Oxidation of cyclohexane, cyclohexanone, and cyclohexanol to adipic acid by a non-HNO3 route over Co/Mn cluster complexes

Article abstract of DOI:10.1006/jcat.2002.3756

μ₃-Oxo-bridged Co/Mn cluster complexes, CoMn₂(O), exhibit high catalytic activity and selectivity for the aerial oxidation, in the homogeneous liquid phase, of cyclohexane, cyclohexanol, and cyclohexanone to adipic acid by a non-HNO₃ route. The cluster complexes are superior to the individual Co and Mn acetates. The yields of adipic acid are comparable to those in the current processes using HNO₃. Electron paramagnetic resonance and electronic spectra indicate that the cluster complex is probably decomposed during the reaction process. Attempts to use these cluster complexes encapsulated in zeolite-Y as solid catalysts for the oxidation reactions were unsuccessful due to significant leaching out of the complexes during the reaction.

Full text of DOI:10.1006/jcat.2002.3756

Journal of Catalysis 212, 39–45 (2002)
doi:10.1006/jcat.2002.3756
Oxidation of Cyclohexane, Cyclohexanone, and Cyclohexanol to Adipic
Acid by a Non-HNO3 Route over Co/Mn Cluster Complexes
S. A. Chavan, D. Srinivas, and P. Ratnasamy1
National Chemical Laboratory, Pune 411 008, India
Received February 4, 2002; revised July 18, 2002; accepted July 25, 2002
ing technology. Asahi Chemical Industry, BASF, Bayer, and
DuPont employ catalytic or thermal processes to destroy
the N2O (6). Solutia and Rhodia subsidiary Alsachimie
(Chalampe, France) recycle the N2O back into their pro-
duction processes (phenol and HNO3, respectively). An al-
ternative non-HNO3 route to AA using cleaner oxidants
such as O2 or H2O2 is highly desirable.
µ -Oxo-bridged Co/Mn cluster complexes, CoMn2(O), exhibit
3
high catalytic activity and selectivity for the aerial oxidation, in
the homogeneous liquid phase, of cyclohexane, cyclohexanol, and
cyclohexanone to adipic acid by a non-HNO3 route. The cluster
complexes are superior to the individual Co and Mn acetates. The
yields of adipic acid are comparable to those in the current processes
using HNO3. Electron paramagnetic resonance and electronic spec-
tra indicate that the cluster complex is probably decomposed during
the reaction process. Attempts to use these cluster complexes en-
capsulated in zeolite–Y as solid catalysts for the oxidation reactions
were unsuccessful due to significant leaching out of the complexes
c
Although many variations of the initial cyclohexane ox-
idation (to KA oil) step have been developed, technol-
ogy for the second stage (conversion of KA oil to AA)
is fundamentally the same as originally developed by Du
Pont in the early 1940s (1). A number of alternative pro-
cesses for producing AA have been investigated but none
of them has been commercialized so far (7–9). Tanaka
(7, 8) has reported a single-step, liquid-phase, aerial ox-
idation of cyclohexane to AA (yields of AA = 70–75% at
cyclohexane conversions of 50–75% and 343–373 K). BASF
has developed a process based on hydrocarboxylation or
carboalkoxylation of butadiene (9). Jacobs and cowork-
ers (10–12), Balkus and coworkers (13, 14), Herron et al.
(15) and Raja and Ratnasamy (16, 17) studied this oxida-
tion with tert-butylhydroperoxide or iodosylbenzene over
metalphthalocyaninecomplexesencapsulatedinzeolite–Y.
Sato et al. (18) oxidized cyclohexene with 30% H2O2 over
Na2WO4 · 2H2O. Recently, Thomas and coworkers (19–21)
and Luna et al. (22) oxidized cyclohexane and n-hexane
with air to AA over metal-containing aluminium phosphate
molecular sieves. However, none of these studies using solid
catalysts has the potential to replace the current commer-
cial process due to their low yields of AA and/or use of
expensive oxidants such as H2O2, tert-butylydroperoxide,
and iodosylbenzene.
during the reaction.
ꢀ 2002 Elsevier Science (USA)
Key Words: aerial oxidation; oxidation of cyclohexane, cyclohex-
anol, and cyclohexanone; oxidations over µ -oxo-bridged Co/Mn
3
cluster complexes; adipic acid via a non-HNO3 route.
INTRODUCTION
Adipic acid (AA), a valuable raw material used in the
production of nylon-6,6, fibers, plasticizers, and food addi-
tives, is manufactured by the oxidation of cyclohexanone,
obtained either by hydrogenation of phenol or, more com-
monly, by the oxidation of cyclohexane (1–3). In the first
stage, cyclohexaneisoxidizedwithoxygen(423–473 K;115–
175 psig), using a soluble cobalt catalyst (usually cobalt
naphthenate, cobalt octonate, or cobalt acetate) to a mix-
ture of cyclohexanol–cyclohexanone (referred to as KA
oil; conversion <10%; selectivity ≈70–90%). In the second
stage, this mixture is further oxidized, at 343–364 K, to AA
(yield of AA = 70–95%) by 40–60% HNO3 in the presence
of copper and vanadium catalysts (2). During the second
stage, significant amounts of environmentally harmful NOx
effluents (NO, NO2, and N2O) are produced (4, 5). While
NO and NO2 are recycled to the HNO3 plant, N2O is usu-
ally vented and constitutes a major source of this highly
polluting species. According to the global warming treaty
signedinKyotoalltheAAmanufacturershadtoinstallN2O
removal systems by the end of 2000, adding cost to the exist-
In our studies on the oxidation of para-xylene to tereph-
thalic acid we had reported (23) that µ3-oxo-bridged
cluster complexes (Fig. 1) are formed from Co and Mn ac-
etates in acetic acid (HOAc) medium during the oxidation
process. These complexes (especially the heteronuclear
CoMn2(O)), in both neat state and when encapsulated in
zeolite–Y, exhibited higher activity and selectivity in the
para-xylene oxidation than the individual metal acetates
(24, 25). We now report the aerial oxidation of cyclohex-
anone, cyclohexanol, and cyclohexane to AA using these
1 To whom correspondence should be addressed. Fax: +91-20-5893355.
E-mail: prs@ems.ncl.res.in.
39
0021-9517/02 $35.00
c
ꢀ 2002 Elsevier Science (USA)
All rights reserved.

40
CHAVAN, SRINIVAS, AND RATNASAMY
was cooled to 298 K and the conversion of reactant(s) and
liquid products formed were determined by gas chromato-
graphic analysis. In the case of cyclohexane oxidation the
catalytic runs were conducted for 8 h. The main reaction
products were adipic, valeric, glutaric, and succinic acids.
These acids were analyzed by GC and identified by GC–MS
after esterification with BF3 (23). Mass balances were ver-
ified. Addition of water controls the oxidation of HOAc to
CO2 and CO in the Co/Mn catalyzed reactions (26). Blank,
controlled experiments were conducted with no catalyst.
The reactions did not proceed in the absence of any cata-
lyst, indicating that they are catalytic in nature.
Caution! The oxidation reactions are potentially explo-
sive and necessary care and safety precautions should be
taken (1, 3).
RESULTS AND DISCUSSION
Characterization of µ3-Oxo-Bridged Co/Mn Cluster
Complexes (24, 25)
FIG. 1. Structure of µ3-oxo-bridged Co/Mn cluster complex.
The cluster complexes showed IR peaks at 1605 and
1404 cm−1 for Mn3(O), 1609 and 1402 cm−1 for Co3(O),
and 1616 and 1604 cm−1 for CoMn2(O) due to the asym-
metric and symmetric stretching vibrational modes of the
COO− group of acetate ligands (Fig. 2). The separation
between νas(COO−) and νs(COO−) indicates bidentate co-
ordination of acetate molecules, in syn–syn mode, to the
metal ions. The characteristic peak of the µ3-oxo-bridge
(sensitive to the coordinated metal ion) appeared at 665,
692, and 685 cm−1 for Mn3(O), Co3(O), and CoMn2(O),
respectively (Fig. 2).
µ3-oxo-bridged cluster complexes, CoMn2(µ3-O)(OAc)6
(py)3, Mn3(µ3-O)(OAc)6(py)3, and Co3(µ3-O)(OAc)6
(py)3 (hereafter referred to as CoMn2(O), Mn3(O), and
Co3(O), respectively; py = pyridine). To our knowledge this
is the first non-HNO3, aerial oxidation route for the prepa-
ration of AA from cyclohexanone wherein the yields of AA
are comparable to those obtained in the current, commer-
cial practice.
EXPERIMENTAL
The “neat” and zeolite–Y-encapsulated µ3-oxo-bridged
cluster complexes (CoMn2(O), Mn3(O), and Co3(O)) were
prepared and characterized as reported earlier (25).
(c) CoMn2(O)
(b) Co3(O)
Procedures
FT-IR spectra of the complexes were recorded on a
Shimadzu FT-COM 1 spectrophotometer as Nujol mulls.
UV–visible spectra were measured using a UV-101PC spec-
trophotometer. EPR spectra were recorded using an X-
band Bruker EMX spectrometer (23).
(a) M n3(O)
Catalytic Activity Studies
The oxidation reactions were conducted in a titanium-
linedParr4843reactor. Inatypicalreaction4.21mlofcyclo-
hexanone, 4.3 ml of cyclohexane, or a mixture of cyclohex-
anol and cyclohexanone was taken in 38 ml of glacial HOAc
and 1.9 ml of distilled water; 0.114 ml of methylethylketone
(MEK; oxidation initiator) and a known amount of the cat-
alyst were then added. The reactor was pressurized with
2000
1750
1500
1250
1000
750
500
Wavenumber (cm -1
)
FIG. 2. FT-IR spectra (Nujol mull) of µ3-oxo-bridged cluster com-
air to 700 psig and heated to 373 K. After 4 h, the reactor plexes.

OXIDATION TO ADIPIC ACID BY A NON-HNO3 ROUTE
41
TABLE 1
254
(c) CoMn2(O)
Selective Oxidation of Cyclohexanone over µ -Oxo-Bridged
3
Co/Mn Cluster Complexesa
345
Product distribution (wt%)
Run
no.
Temp. Conv.
(K)
254
Catalyst
(wt%) TOFb SA
GA
VA
AA
(b) Mn3(O)
1
2
3
4
Co3(O)
Mn3(O)
CoMn2(O)
CoMn2(O)
363
363
353
363
363
373
423
11.9
93.7
82.8
92.2
90.1
97.6
99.9
136 10.6
15.1
13.9
18.7
14.7
14.4
11.5
21.5
0
0
1.1
0
3.3
0
74.3
83.1
75.0
78.4
80.0
86.6
67.2
320
1104
3.0
5.2
6.9
2.3
1.9
6.6
941
1048
1024
1104
1130
250
4a CoMn2(O)
5
6
(a) Co3(O)
400
CoMn2(O)
CoMn2(O)
355
4.7
a Reaction conditions: catalyst weight = 7.5 mg (8.6–8.9 µmol);
200
300
Wavelength (nm)
500
cyclohexanone = 4.21 ml (40.5 mmol); oxidant = air (550 psig when hot);
oxidation initiator = MEK (0.114 ml); medium = HOAc (38 ml) + water
(1.9 ml); reaction time = 4 h. Run 4a was without MEK.
b Turnover frequency (TOF) = number of moles of substrate converted
per mole of the catalyst per hour.
FIG. 3. UV–visible spectra of µ3-oxo-bridged cluster complexes in
HOAc.
The acetic acid solutions of the cluster complexes showed
a partially resolved UV band, characteristic of a µ3-oxo
bridge, ataround320nmforMn3(O)and355and345nmfor
Co3(O) and CoMn2(O), respectively (Fig. 3). The position
of this band, attributed to oxygen-to-metal charge transfer,
is sensitive to the coordinated metal ion.
200 to 900 psig) and reaction time (Table 2). Under our re-
action conditions, complete conversion of cyclohexanone
was achieved at 6 h (Table 2, Run 12). Product selectiv-
ity, however, was less sensitive to changes in pressure and
reaction time.
Detailed EPR spectroscopic studies of the cluster com-
plexes have already been reported by us earlier (25). Spec-
tral variations in the range 77–298 K were consistent with
intermolecular antiferromagnetic interactions (25). Poly-
crystals of Mn3(O) complex showed a broad, isotropic sig-
nal at g = 2.01 with a peak-to-peak linewidth of 500 G.
Co3(O) cluster complexes also showed an isotropic signal
(giso = 2.259), but with a linewidth of 1075 G. CoMn2(O)
clusters were EPR silent.
Oxidation of a Mixture of Cyclohexanone
and Cyclohexanol
The oxo-bridged clusters are highly efficient in the oxi-
dation of cyclohexanone–cyclohexanol (-one/-ol) mixtures.
While the ketone (-one) is oxidized directly to the acid
products, the alcohol (-ol) is oxidized initially to -one and
then, subsequently, to the acid products (1–3). The catalytic
TABLE 2
Oxidation of Cyclohexanone
Effects of Pressure and Reaction Time on Product Distribu-
tion in the Oxidation of Cyclohexanone: Catalyst, CoMn2(O);
temperature, 363 Ka
Effect of cluster composition and reaction temperature.
Adipic acid (AA) was the major product while succinic
(SA), glutaric (GA), and valeric acids (VA) were the mi-
nor products in the oxidation of cyclohexanone. The com-
position of the cluster complex has a marked effect on the
reactivity (conversion and TOF) and product selectivity.
Table 1 lists the catalytic activity data of the cluster com-
plexes. Though Mn3(O) and CoMn2(O) clusters exhibit
similar activity, the yield of AA was more with the former
(Table 1, Runs 2 and 4). Among the three clusters, Co3(O)
exhibited least activity and AA selectivity (Table 1, com-
pare runs 1, 2, and 4). On increasing the temperature from
353 to 423 K, while the conversion and TOF of cyclohex-
anone increased, the yield of AA passed through a maxi-
mum around 373 K (Table 1, Runs 3–6).
Product distribution (wt%)
Run Pressure Reaction Conv.
no.
(psig)
time (h) (wt%) TOF SA
GA
VA
AA
1
2
3
4
5
6
7
8
9
200
350
550
700
900
550
550
550
550
550
550
550
0.75
0.75
0.75
0.75
0.75
0.25
0.50
1.0
1.5
2
51.6
71.0
89.8
94.9
98.2
3112
0
14.7
13.1
15.7
14.9
13.2
14.9
13.0
15.1
13.4
15.3
13.8
17.1
5.8
5.7
1.7
3.0
3.8
1.9
2.1
2.1
2.1
3.3
2.2
3.0
79.5
80.3
79.7
79.0
80.0
80.2
82.1
79.5
82.4
78.4
81.7
76.3
4282 1.0
5415 2.9
5723 3.2
5922 3.0
69.5 12573 3.0
84.7
88.0
89.1
90.6
90.0
99.9
7662 2.7
3980 3.3
2687 2.1
2049 3.0
1357 2.3
755 3.6
10
11
12
3
6
Effect of pressure and reaction time. The conversion of
cyclohexanone and AA yield increased with pressure (from
a Other reaction conditions are the same as in Table 1.

42
CHAVAN, SRINIVAS, AND RATNASAMY
TABLE 3
Selective Oxidation of a Mixture of Cyclohexanol (-ol) and Cyclohexanone (-one) over µ -Oxo-Bridged Co/Mn
3
Cluster Complexes at 363 Ka
Individual conv. (wt%)
Product distribution (wt%)
Run
no.
Substrate
(-ol : -one)
Overall conv.
(wt%)
Catalyst
-ol
-one
SA
GA
VA
AA
1
2
3
4
5
6
7
8
CoMn2(O)
CoMn2(O)
CoMn2(O)
CoMn2(O)
Mn3(O)
Mn3(O)
Mn3(O)
Mn3(O)
0 : 100
10 : 90
20 : 80
50 : 50
0 : 100
10 : 90
20 : 80
50 : 50
92.2
85.9
84.5
79.1
93.7
87.1
80.5
84.7
—
100
100
91.3
—
100
96.1
94.7
92.2
84.5
80.9
65.5
93.7
86.1
75.6
74.1
6.9
3.3
3.4
3.3
3.0
2.9
2.2
2.3
14.7
17.5
15.2
12.5
13.9
16.7
13.1
13.3
0
78.4
77.9
79.7
82.5
83.1
79.6
84.1
83.7
1.4
1.7
1.7
0
0.8
0.6
0.8
a Run time = 4 h. Other reaction conditions are the same as in Table 1.
activity data are presented in Table 3. The conversion of -ol Oxidations over Mn3(O), Co3(O), and CoMn2(O)
increased (from 91 to 100 wt% over CoMn2(O) and from
94.7 to 100 over Mn3(O)) with increasing concentration of
-one in the reaction mixture (Table 3). In the absence of
-one the conversion of -ol was very low (45 wt%). In con-
trast, theindividualconversionof-onedecreased(from84.5
to 65.5 wt% over CoMn2(O) and from 86.1 to 74.1 wt%
over Mn3(O)) with increasing concentrations of -ol in the
reaction mixture (Table 3). In the absence of -ol the con-
version of -one was higher (92.2 wt% over CoMn2(O) and
93.7 wt% over Mn3(O)). The overall conversion of the reac-
tion mixture (-one + -ol) decreased from 92 to 79 wt% over
CoMn2(O) and from 93.7 to 84.8 wt% over Mn3(O) with
increasing -ol concentration. AA selectivities, however, are
not significantly changed (80 5%) (Table 3). It appears
that -ol is a poison in cyclohexanone oxidation.
Complexes Encapsulated in Zeolite–Y
The encapsulated cluster complexes were prepared as re-
ported earlier (18). The encapsulated complexes exhibited
activity (in cyclohexanone oxidation with conversion = 3.7,
97.1, and 98.4 wt% and AA selectivity = 91, 82.5, and
79.3 wt% for Co3(O)–Y, Mn3(O)–Y and CoMn2(O)–Y, re-
spectively) similar to that of the neat clusters. However,
significant leaching out of the metal ions during the reac-
tion was observed. This was in contrast to the situation in
the oxidation of para-xylene at 473 K and 700 psig wherein
such leaching was negligible (25). The reasons for these dif-
ferences are under investigation.
In Situ Spectroscopic Studies
In situ electronic and EPR spectroscopic studies had en-
abled us, earlier, to identify the oxo-bridged clusters as the
active species in para-xylene oxidation (23). Similar studies
for cyclohexanone oxidation were performed on the fol-
lowing solutions:
Oxidation of Cyclohexane
The catalytic activity of the cluster complexes in cy-
clohexane oxidation decreased in the order Co3(O) >
CoMn2(O) > Mn3(O) (Table 4). Succinic and glutaric acids
were the major products. It may recalled that adipic acid
is the major product in the oxidation of cyclohexanone–
cyclohexanol (Tables 1–3).
1. Mn(OAc)2 in HOAc + pyridine + water (298 K;
363 K for 45 min; 700 psig air, 363 K, 45 min).
2. Solution 1 + MEK (700 psig air, 363 K, 45 min).
3. Solution 2 + cyclohexanone (700 psig air, 363 K,
45 min).
TABLE 4
4. A mixture of Co(OAc)2 and Mn(OAc)2 (3 : 1 mmol
ratio)inHOAc + pyridine + water(298 K;363 Kfor45min;
700 psig air, 363 K, 45 min).
Aerial Oxidation of Cyclohexane over µ -Oxo-Bridged Co/Mn
3
Cluster Complexesa
5. Solution 4 + MEK (700 psig, 363 K, 45 min).
Product distribution (wt%)
Run
no.
Conv.
6. Solution 5 + cyclohexanone (700 psig, 363 K, 45 min).
Catalyst
(wt%) TOF -ol -one SA GA VA AA
Electronic Spectroscopy
1
2
3
Co3(O)
Mn3(O)
CoMn2(O)
85.6
41.8
51.6
489 5.0 1.5 16.2 55.0 0.4 22
248
—
1.0 59.6
1.1
—
—
38.3
36.2
Solution 1, containing Mn(OAc)2, did not exhibit any ab-
sorption in the visible region. However, on heating at 363 K
for45minandpressurizingwithair(700psig)anarrowband
at 490 nm, a broad band at 445 nm, and a shoulder around
293 3.7 1.3 31.6 27.2
a Cyclohexane = 4.4ml(40.7mmol);reactiontime = 8h. Otherreaction
conditions are the same as in Table 1.

OXIDATION TO ADIPIC ACID BY A NON-HNO3 ROUTE
43
In the presence of cyclohexanone (solution 6) a reduction
in the intensity of these bands (Fig. 5, curve h) was ob-
served in addition to the appearance of a weak band at
524 nm corresponding to Co(II) species. The electronic
spectroscopic studies of Mn(OAc)2 and a mixture of
Co(OAc)2 and Mn(OAc)2 solutions reveal that during the
oxidation reaction a small number of M(III) ions in the
cluster complex are reduced to M(II) species.
(c)
1
(d)
330
445
490
500
EPR Spectroscopy
(b)
(a)
Solution 4, containing a mixture of Co(OAc)2 and
Mn(OAc)2, showed a resolved sextet-line hyperfine pat-
tern (Aiso = 94.5 G) centered at giso = 2.0012 due to the
manganese(II) acetate complex (Fig. 6, Curve a). Although
Co(II) acetate is paramagnetic (S = 3/2), its EPR signals
cannot be detected at ambient temperatures due to the
short spin lattice relaxation times. When the solution was
pressurized with air (700 psig at 363 K for 45 min) the over-
all spectral intensity decreased (Fig. 6, curve b), the signals
broadened, and the spectral width increased from 524 to
538 G. The center of the hyperfine pattern shifted from
giso = 2.012 to 1.998. These features indicate the forma-
tion of magnetically interacting Co/Mn cluster complexes
(CoMn2(O)) at high pressure (700 psig) and temperatures
(363 K) from individual cobalt and manganese acetate salts.
This result is in agreement with the electronic spectroscopic
studies, wherein these cluster complexes were observed and
characterized by bands at 345 and 455 nm (Fig. 5). The for-
mal oxidation state of Mn ions also changes from +2 to
+3 on the formation of the cluster complexes. When MEK
was added and the system pressurized with air (700 psig,
365 K, 45 min; solution 5) a significant decrease in EPR
signal intensity was observed, indicating the formation
0
400
600
700
Wavelength (nm)
FIG. 4. Electronic spectra of manganese complexes: (a) Mn(OAc)2
in HOAc + pyridine + H2O at 363 K, 45 min; (b) solution in (a) at 700 psig
air, 363 K, 45 min; (c) Mn(OAc)2 in HOAc + pyridine + H2O + MEK
at 700 psig air, 363 K, 45 min; (d) Mn(OAc)2 in HOAc + pyridine + H2O +
MEK + cyclohexanone at 700 psig air, 363 K, 45 min.
330 nm (Fig. 4, curves a and b) were observed. The former
(490 nm) corresponds to d–d transition of a monomeric
Mn(III) species. The latter two bands are attributed to d–
d transition and oxygen ⇒ Mn(III) charge transfer transi-
tion of µ3-oxo-bridged Mn cluster complex (Mn3(O)); the
assignment of the 330-nm band was also confirmed from
Mn3(O) cluster complexes prepared independently (vide
supra)). As anticipated, the intensities of Mn3(O) bands
increased with the concentration of O2 (higher pressure
of air). When the experiment was conducted with MEK,
the color of the solution changed from pale yellow to deep
brown. Solution 2 showed intense bands at 445 and 330 nm
due to Mn3(O) cluster complex (Fig. 4, curve c; dilution
factor = 20). MEK promoted the oxidation of Mn from the
+2 to +3 state and formation of the more active Mn3(O)
cluster complex. In the presence of cyclohexanone (solu-
tion 3), some reduction in the intensity of Mn3(O) bands
was observed (Fig. 4, curve d; dilution factor = 20), suggest-
ing that part of Mn3(O) is reconverted to Mn(II) species in
the presence of cyclohexanone. The Mn3(O) cluster and
Mn(II) species are, perhaps, in a quasiequilibrium.
Solution 4, containing the mixture of Co(OAc)2 and
Mn(OAc)2 (3 : 1 mmol), exhibited d–d transitions in the
visible region, at 524 and 490 nm, corresponding to Co(II)
acetate complexes. On heating at 363 K for 45 min and
pressurizing with air (700 psig) the intensity of these bands
increased (Fig. 5, curves e and f). With MEK, solution 5
(700 psig, 363 K, 45 min) showed marked changes in the po-
sition and intensity of the UV–visible bands (Fig. 5, curve g;
dilution = 20). New bands appeared, around 570, 455, and
345 nm. The broad band at 570 nm is due to a Co(III)
species and those at 455 and 345 nm are characteristic of µ3-
3.5
(d)
3.0
2.5
345
2.0
(c)
1.5
490
500
524
455
570
1.0
0.5
0.0
(b)
(a)
300
400
600
700
Wavelength (nm)
FIG. 5. Electronic spectra of cobalt and manganese complexes: (a)
Co(OAc)2 + Mn(OAc)2 (3 : 1 mmol) in HOAc + pyridine + H2O at 363 K,
45 min; (b) solution in (a) at 700 psig air, 363 K, 45 min; (c) Co(OAc)2+
Mn(OAc)2 in HOAc + pyridine + H2O + MEK at 700 psig, 363 K,
45min;(d)Co(OAc)2 + Mn(OAc)2 inHOAc + pyridine + H2O + MEK +
oxo-bridged Co/Mn cluster complexes (CoMn2(O)) (27). cyclohexanone at 700 psig, 363 K, 45 min.

44
CHAVAN, SRINIVAS, AND RATNASAMY
impurities (1). The acid impurities (such as GA, VA, and
Co(OAc)2+Mn(OAc)2 (3:1 mmol)
SA) can form also by oxidative degradation of AA. A plau-
sible mechanism for the degradation of AA is as follows
(28):
MIII(OAc)3 + HOOC(CH2)4COOH
(a)
↔ MIII(OAc)2(OOC(CH2)4COOH) + HOAc,
[1]
[2]
MIII(OAc)2(OOC(CH2)4COOH)
→ MII(OAc)2 + CO2 + •CH2(CH2)3COOH,
(b)
(c)
•CH2(CH2)3COOH + O2 → HOOC(CH2)3COOH, [3]
MIII(OAc)3 + HOOC(CH2)3COOH
(d)
↔ MIII(OAc)2(OOC(CH2)3COOH) + HOAc,
[4]
MIII(OAc)2(OOC(CH2)3COOH)
2000
2500
3000
3500
4000
4500
5000
→ MII(OAc)2 + CO2 + •CH2(CH2)2COOH,
[5]
Magnetic Field (G)
•CH2(CH2)2COOH + O2 → HOOC(CH2)2COOH. [6]
FIG. 6. EPR spectra of reaction solutions at 298 K: (a) Co(OAc)2 +
Mn(OAc)2 (3 : 1 mmol) in HOAc + pyridine + H2O at 363 K, 45 min;
(b) solution in (a) at 700 psig air, 363 K, 45 min; (c) Co(OAc)2 +
Mn(OAc)2 in HOAc + pyridine + H2O + MEK at 700 psig, 363 K, 45 min;
(d) Co(OAc)2 + Mn(OAc)2 in HOAc + pyridine + H2O + MEK + cyclo-
hexanone at 700 psig, 363 K, 45 min.
Two features may be noted: (i) the decomposition of adipic
to glutaric and succinic acids, and (ii) the partial reduction
of the metal ions (from the tri- to the divalent state). Such a
change in the oxidation state of metal ions in cyclohexanone
oxidation was indeed observed by us by in situ electronic
and EPR spectroscopic studies (vide supra). During the ox-
idation reaction, the metal ions are reoxidized by O2.
of the EPR-silent µ3-oxo-bridged Co/Mn clusters (Fig. 6,
curve c). When cyclohexanone was added (solution 6),
the overall intensity increased (solution 6; Fig. 6, curve d)
due to partial decomposition of the cluster complexes to
monomeric Co(II) and Mn(II) species. Similar conclusions
were drawn also from the electronic spectroscopic studies
(see above).
CONCLUSIONS
A novel, non-HNO3 route for the preparation of adipic
acid by aerial oxidation of cyclohexanone or a mixture
of cyclohexanone and cyclohexanol or cyclohexane using
µ3-oxo-bridged Co/Mn cluster complexes is reported. The
yields of adipic acid are comparable to those in current
processes.
Speculated Mechanism of Cyclohexanone Oxidation
In the oxidation of cyclohexanone, 1,2-cyclohexanedione
is an intermediate (Fig. 7) (1). While the keto-form yields
the desired product AA, the enol-form leads to other acid
REFERENCES
1. Davis, D. D., and Kemp, D. R., in “Kirk–Othmer Encyclopedia of
Chemical Technology” (J. I. Kroschwitz and M. Howe-Grant, Eds.),
Vol. 1, p. 466. Wiley, New York, 1991.
2. Castellan, A., Bart, J. C. J., and Cavallaro, S., Catal. Today 9, 237 (1991).
3. Dumas, T., andBulani, W., in“OxidationofPetrochemicals:Chemistry
and Technology.” Applied Science, London, 1974.
4. Thiemens, M. H., and Trogler, W. C., Science 251, 932 (1991).
5. Schuchardt, U., Cardoso, D., Sercheli, R., Pereira, R., da Cruz, R. S.,
Guerreiro, M. C., Mandelli, D., Spinace´, E. V., and Pires, E. L., Appl.
Catal. A 211, 1 (2001).
6. Scott, A., Chem. Week., Feb. 18, p. 37 (1998).
7. Tanaka, K., CHEMTECH 555 (1974).
8. Tanaka, K., Hydrocarbon Process. 114 (1974).
9. Chauvel, A., Delmon, B., and Holderich, W. F., Appl. Catal. A 115,
173 (1994).
10. Parton, R. F., Vankelecom, I. F. J., Casselman, M. J. A., Bezoukhanova,
C. P., Uytterhoeven, J. B., and Jacobs, P. A., Nature 370, 541 (1994).
11. Thisbault-Starzyk, F., Parton, R. F., and Jacobs, P. A., Stud. Surf. Sci.
Catal. 84, 1419 (1994).
12. Parton, R. F., Peere, G. J., Neys, P. E., Jacobs, P. A., Claessens, R., and
Baron, G. V., J. Mol. Catal. A 113, 445 (1996).
FIG. 7. Possible mechanism of cyclohexanone oxidation.

OXIDATION TO ADIPIC ACID BY A NON-HNO3 ROUTE
45
13. Balkus, K. J., Jr., Eissa, M., and Lavado, R., J. Am. Chem. Soc. 117,
10753 (1995).
21. Thomas, J. M., Raja, R., Sankar, G., and Bell, R. G., Nature 398, 227
(1999).
14. Balkus, K. J., Jr., Eissa, M., and Lavado, R., Stud. Surf. Sci. Catal. 94,
713 (1995).
22. Luna, F. J., Ukawa, S. E., Wallau, M., and Schuchardt, U., J. Mol. Catal.
A 117, 405 (1997).
15. Herron, N., Stucky, G. D., and Tolman, C. A., J. Chem. Soc. Chem.
Commun. 1521 (1986).
23. Chavan, S. A., Halligudi, S. B., Srinivas, D., and Ratnasamy, P., J. Mol.
Catal. A 161, 49 (2000).
16. Raja, R., and Ratnasamy, P., Catal. Lett. 48, 1 (1997).
17. Raja, R., and Ratnasamy, P., U.S. Patent 5,767,320 (1998).
18. Sato, K., Aoki, M., and Noyori, R., Science 281, 1646 (1998).
19. Dugal, M., Sankar, G., Raja, R., and Thomas, J. M., Angew. Chem. Int.
Ed. 39, 2310 (2000).
24. Chavan, S. A., Srinivas, D., and Ratnasamy, P., Chem. Commun. 1124
(2001).
25. Chavan, S. A., Srinivas, D., and Ratnasamy, P., J. Catal. 204, 409 (2001).
26. Partenheimer, W., Catal. Today 23, 69 (1995).
27. Vincent, J. B., Chang, H.-R., Folting, K., Huffman, J. C., Christou, G.,
and Hendrickson, D. N., J. Am. Chem. Soc. 109, 5703 (1987).
28. Schulz, J. G. D., and Onopchenko, A., J. Org. Chem. 45, 3716 (1980).
20. Raja, R., Sankar, G., and Thomas, J. M., Angew. Chem. Int. Ed. 39,
2313 (2000).