Liquid Phase Oxidation of p-cymene by (VO)2P2O 7 and VO(PO3)2 Catalysts

Article abstract of DOI:10.1007/s10562-010-0278-3

Vanadium phosphate oxide (VPO) catalysts derived from VOHPO ₄?0.5H₂O and VO(H₂PO₄) ₂ precursors showed improved rates in the catalysed liquid phase oxidation of p-cymene. Both catalysts showed high performance with conversions of up to 30% achieved within 3 h with a selectivity towards the tertiary cymene hydroperoxide (TCHP) of 75-80% when compared to lower conversions and poorer TCHP selectivity in the non-catalysed industrial oxidation processes that takes about 8-12 h reaction time. The temperature dependent structural changes during activation within the catalysts were studied by in situ XRD and TGA-MS. The in situ XRD showed that the VO(H₂PO₄)₂ form mainly an amorphous phase at temperature up to 600 °C while above 650-750 °C a stable VO(PO₃)₂ phase was obtained. The in situ XRD studies of the (VO)₂P₂O₇ phase derived from VOHPO₄?0.5H₂O showed to became more crystalline with increasing temperature from 400 to 750 °C. No formation of VOPO₄ phases were observed in the activated (VO)₂P₂O₇ catalyst even at high temperature under the in situ XRD conditions. Graphical Abstract: VPO catalysts derived from VOHPO₄?0.5H₂O and VO(H₂PO₄)₂ precursors improve oxidation rates and improved tert-cymene hydroperoxide selectivity during the liquid-phase oxidation of p-cymene compared to non-catalyzed oxidations. [Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-010-0278-3

Catal Lett (2010) 135:105–113
DOI 10.1007/s10562-010-0278-3
Liquid Phase Oxidation of p-cymene by (VO)2P2O7 and VO(PO3)2
Catalysts
•
Peter R. Makgwane Ernst E. Ferg
•
•
David G. Billing Ben Zeelie
Received: 31 August 2009 / Accepted: 16 January 2010 / Published online: 5 February 2010
Ó Springer Science+Business Media, LLC 2010
Abstract Vanadium phosphate oxide (VPO) catalysts
derived from VOHPO4ꢀ0.5H2O and VO(H2PO4)2 precur-
sors showed improved rates in the catalysed liquid phase
oxidation of p-cymene. Both catalysts showed high per-
formance with conversions of up to 30% achieved within
3 h with a selectivity towards the tertiary cymene hydro-
peroxide (TCHP) of 75–80% when compared to lower
conversions and poorer TCHP selectivity in the non-cata-
lysed industrial oxidation processes that takes about 8–12 h
reaction time. The temperature dependent structural chan-
ges during activation within the catalysts were studied by
in situ XRD and TGA-MS. The in situ XRD showed that
the VO(H2PO4)2 form mainly an amorphous phase at
temperature up to 600 °C while above 650–750 °C a stable
VO(PO3)2 phase was obtained. The in situ XRD studies of
the (VO)2P2O7 phase derived from VOHPO4ꢀ0.5H2O
showed to became more crystalline with increasing tem-
perature from 400 to 750 °C. No formation of VOPO4
phases were observed in the activated (VO)2P2O7 catalyst
even at high temperature under the in situ XRD conditions.
1 Introduction
The selective oxidation of p-cymene is an important
industrial chemical process because of its product tertiary
cymene hydroperoxide (TCHP) which is an intermediate
material for synthesis of p-cresol [1–3]. The current
industrial liquid phase oxidation process of p-cymene is a
non-catalysed reaction process is run as two phases,
aqueous and organic. This process is characterised by poor
space time yield due to slow rates of 8–12 h to achieve 15–
20% conversion with a TCHP selectivity of 65–70% [1].
The use of molecular oxygen and heterogeneous catalysts
has attracted much attention because of the improved
reaction rates and product selectivity achieved in many
liquid phase oxidation reactions [4, 5]. Vanadium phos-
phate oxide (VPO) is one of the important industrial het-
erogeneous catalysts for gas phase oxidations of C2–C5
hydrocarbons [6–9]. This catalyst had been reviewed in
number of publications about which phase or phases are
actually catalytically active [10–14]. The vanadyl pyro-
phosphate phase [(VO)2P2O7] is generally assumed to be
the contributing factor to make this phase of VPO catalyst
more active.
Keywords In situ XRD ꢀ Liquid phase oxidation ꢀ
p-cymene ꢀ Vanadium phosphate ꢀ
Tertiary cymene hydroperoxide
While the (VO)2P2O7 phase is assumed to be the active
catalytic phase, some studies have suggested that the par-
ticipation of the V?4 and V?5 oxidation states in the cat-
alyst material provides the basis for this catalyst to operate
via the Mars–Krevelen mechanism in the reaction [15–17].
Thus, the observed gain in n-butane conversion and maleic
anhydride selectivity had been attributed to the good
reducibility and oxidizibility of the (VO)2P2O7 catalyst [15,
16, 18]. Apart from the above matters pertaining to the
performance of VPO catalysts, a number of factors can
influence its performance such as its preparative route,
starting material type for preparations, detailed phase
P. R. Makgwane ꢀ E. E. Ferg (&) ꢀ B. Zeelie
InnoVenton: NMMU Institute of Chemical Technology Institute,
Nelson Mandela Metropolitan University, P. O. Box 77000,
Port Elizabeth 6031, South Africa
e-mail: ernst.ferg@nmmu.ac.za
D. G. Billing
School of Chemistry and DST-NRF Centre of Excellence
in Strong Materials, University of the Witwatersrand,
P.O. Wits, Johannesburg 2050, South Africa
123

106
P. R. Makgwane et al.
composition and reaction medium [19–22]. The (VO)2P2O7
phase is obtained from VOHPO4ꢀ0.5H2O precursor that is
transformed under reaction conditions (n-butane/O2 flow,
400 °C) or calcination in N2 at 550 °C [19]. The purity of
the VOHPO4ꢀ0.5H2O phase is also one of the critical fac-
tors affecting the catalytic properties and performance of
the (VO)2P2O7 catalyst. The presence of the VO(H2PO4)2
phase in the VOHPO4ꢀ0.5H2O precursor had been associ-
ated with poor catalytic performance and low surface area
of the activated (VO)2P2O7 catalyst [23, 24]. The
VO(H2PO4)2 phase is soluble in water and can be easily
eliminated to produce a pure VOHPO4ꢀ0.5H2O phase. The
pure VO(H2PO4)2 phase had been previously prepared and
its activated catalyst showed mainly vanadium polyphos-
phate, VO(PO3)2 [24].
N2 flow conditions in a tube furnace while VO(H2PO4)2 to
VO(PO3)2 at 650 for 4 h under N2 flow.
2.2 Characterisation of Catalysts
Ambient powder XRD measurements were obtained by
using a Bruker D8 with Cu radiation in standard Bragg–
Brentano geometry using a Ni filter at the detector. The
scan range was 5–60° 2h at 0.02 steps.
In situ XRD was done on a Bruker D8 Advance fitted
with an Anton Paar XRK900 reactor, and operated under
inert atmosphere conditions by passing N2 gas over the
sample. Measurements were made in parallel beam
geometry, with the diffractometer equipped with a primary
¨
beam Gobel mirror, a radial Soller slit, a Vantec-1 detector
While vanadium phosphate catalyst has been used
extensively in many gas phase oxidations due to its unique
catalytic properties very little is known of its application in
liquid phase oxidation. The present study investigated the
changes in phase composition behaviour of vanadium
phosphate catalyst prepared from certain precursors,
VOHPO4ꢀ0.5H2O and VO(H2PO4)2 to (VO)2P2O7 and
VO(PO3)2, respectively during in situ XRD activation and
their catalytic activities in p-cymene liquid phase oxidation
are reported. Of particular interest was the activity of the
VO(PO3)2 catalyst phase prepared from the VO(H2PO4)2
precursor in the liquid phase oxidation of p-cymene since it
has been reported previously that its presence in the
VOHPO4ꢀ0.5H2O precursor influences the catalytic perfor-
mance of the (VO)2P2O7 catalyst in gas phase oxidations.
and using Cu-Ka radiation. The in situ XRD measurements
were recorded at 50 °C intervals with heating rate of
10 °C min-1 over the temperature range between 350 and
750 °C.
The resulting diffraction patterns were identified by
using the PDF database [25] and their structures refined by
Rietveld refinement using Topas V3 [26] with structural
information obtained from the ICSD [27].
The surface area of the catalysts samples were deter-
mined by BET nitrogen absorption using a Gemini
Micromeritics surface area analyser. Samples we first
degassed at 170 °C N2 flow 2 h prior analysis to remove
any moisture present.
Particle size analysis was done by laser diffraction on a
Malvern Mastersizer particle size analyzer.
TGA analysis was done on a TA Q600 heating the
sample from room temperature to 800 °C at 10 °C min-1
under N2 gas. The simultaneous MS analysis of the
decomposition gases was done using a Pfeifer ThermostarÒ.
2 Experimental
2.1 Catalyst Preparation
2.3 Catalytic Testing Procedure
Both VO(H2PO4)2 and VOHPO4ꢀ0.5H2O precursors were
prepared by using VOPO4ꢀ2H2O as starting material [19, 23].
The VOPO4ꢀ2H2O was prepared by reacting V2O5 (11.8 g)
with H3PO4 (85%, 115.5 g) in distilled water (24 ml g-1
solid). The mixture was stirred under reflux at 120 °C for 8 h.
The yellow solid was then recovered by vacuum filtration,
washed with 100 ml distilled water followed by 100 ml
acetone and then air dried at 110 °C for 16 h. In the second
step, 4 g of VOPO4ꢀ2H2O was refluxed for 21 h in 80 ml
of2-butanone [for VO(H2PO4)2] and isobutanol (for
VOHPO4ꢀ0.5H2O), respectively. The resulting precipitates
were filtered out from the solvent, washed with acetone, and
dried for 16 h at 110 °C. The VOHPO4ꢀ0.5H2O precursor
was further refluxed in water (9 ml g-1 of solid) for 3 h to
eliminate all water dissolving impurities including the pos-
sible presence of VO(H2PO4)2. The VOHPO4ꢀ0.5H2O pre-
cursor was activated to (VO)2P2O7 at 550 °C for 4 h under
Catalytic testing was performed in a jacketed 500 ml glass
batch reactor fitted with two baffles and impeller-type
stirrer blade. To a reactor was added p-cymene (35 g),
oxidized cymene oil as initiator (2 g) to initiate the reaction
(containing 58% content of TCHP) and the respective VPO
catalysts (0.0175 g). Only activated catalysts of the two
precursors, (VO)2P2O7 and VO(PO3)2 were tested as the
actual catalyst for p-cymene liquid phase oxidations. The
VOHPO4ꢀ0.5H2O precursor showed previously to give
5–7% low activity and TCHP selective in the p-cymene
oxidation reaction [28]. The initiator containing 58%
content of TCHP was obtained by concentrating the oxi-
dised cymene oil under vacuum distillation by removing
some of the unreacted p-cymene starting material. The
desired reaction temperature was controlled at 100 °C by
circulating the hot oil through the reactor jacket. A reflux
123

Liquid Phase Oxidation of p-cymene
107
condenser was attached to the reactor. Undiluted pure O2
gas was introduced through a gas inlet tube at the top of the
reactor at a flow rate of 40 ml min-1. The reactor top head-
space was small in order to ensure the minimal build-up of
the oxygen to avoid any possible explosion. However, it
must be noted the reactor was operated as an open system,
thus, allowing the escaping of the gaseous reactants
through the reflux condenser. All catalytic testing experi-
ments were performed at ambient pressure. The oxidation
products samples containing the hydroperoxides were
collected at regular intervals and treated with triphenyl-
phosphine in methanol for GC-FID analysis using a
Supelco SPB-20 column (30 m 9 0.25 mm i.d. 9 0.25 lm).
5 days [29]. This reported experimental value and the
one obtained in the present study, 10.7% for the
VOHPO4ꢀ0.5H2O precursor are slightly different to
the expected theoretical value of 10.5% calculated accord-
ing to the decomposition equation below:
2VOHPO4 ꢀ 0:5H2O ! ðVOÞ2P2O7 þ 2H2O
The differences of the experimental and expected the-
oretical values seems to suggest that although the XRD
structure confirmed the formation of only VOHPO4ꢀ
0.5H2O (Fig. 1b) as the solely phase there is a possibility of
trace amount of other phases present. These could account
to about 0.2% of the total 10.7% weight loss which were
not detectable by the XRD due to its limited sensitivity in
detection. Hannour et al. [24] obtained a TGA decompo-
sition weight loss of 13% for VO(H2PO4)2 precursor phase
at temperature of 405 °C which is the same to our value
obtained in this study. The decomposition pattern of the
VO(H2PO4)2 to VO(PO3)2 catalyst phase can be illustrated
by the simple equation below that shows the VO(H2PO4)2
precursor to posses water probably absorbed on its surface
structure:
3 Results and Discussion
3.1 Characterization of the Catalyst
The XRD patterns of the VOHPO4ꢀ0.5H2O and
VO(H2PO4)2 precursors showed only the presence of the
respective pure phases with no other impurities that are
typically observed when using the described preparative
methods (Fig. 1a, b). The structural changes, of the two
precursors, that occur with increasing temperature were
studied by in situ XRD and TGA-MS. The TGA-MS curves
of the two precursors, VOHPO4ꢀ0.5H2O and VO(H2PO4)2
illustrated in Fig. 2a and b, respectively showed differ-
ences in their thermal decomposition weight loss curves
with increasing temperature. The TGA-MS decomposition
curve of the VOHPO4ꢀ0.5H2O precursor showed a total
weight loss of 10.7% while the VO(H2PO4)2 precursor
displayed a loss of 13%. The results obtained agree closely
with the previous reported data by Hannour et al. and
Johnson et al. and, respectively [24, 29]. The TGA values
of between 11 and 12% total weight loss for the
VOHPO4ꢀ0.5H2O precursor were obtained at different
temperatures and exposure times ranging between 3 and
VOðH2PO4Þ2! VOðPO3Þ2þH2O þ 1=2O2
The simultaneous MS scan of the decomposition gases
of the samples from the TGA confirmed the presence of
water only. This is indicated by the m/z of 18 in Fig. 2a and
b, respectively.
The in situ XRD of the VOHPO4ꢀ0.5H2O precursor
recorded at 50 °C intervals with heating rate of
10 °C min-1 over the temperature range between 350 and
750 °C is illustrated in Fig. 3. The TGA curve in Fig. 2a
showed that the maximum loss of the intercalated water
from the VOHPO4ꢀ0.5H2O structure occurred between the
temperature ranges of 350–400 °C. This is consistent with
the in situ XRD results (Fig. 3). At 400 °C, the in situ XRD
patterns showed that the VOHPO4ꢀ0.5H2O structure com-
pletely collapsed with visible peaks that corresponded to
the (VO)2P2O7 phase (broad peaks at around 23° and 29°
2h). There was also a small visible peak on the shoulder of
the broad peak at 23° 2h XRD reflection. The peak became
clearly resolved at about 22° 2h with the increase of the in
situ XRD activation temperature up to 650 °C. An addi-
tional peak at around 18° 2h started appearing at about
500 °C and became clearly visible at 600 °C. Above
650 °C, these two peaks disappeared and a well crystalline
(VO)2P2O7 catalyst phase was formed. These peaks prob-
ably correspond to some intermediate phase which could
not be identified due to the limited amount of diffraction
information available and could not be clearly associated
with the formation of intermediate phases belonging to the
VOPO4 groups of materials. It was demonstrated that
the transformation of VOHPO4ꢀ0.5H2O to (VO)2P2O7
Fig. 1 XRD of a VOHPO4ꢀ0.5H2O and b VO(H2PO4)2 precursors
123

108
P. R. Makgwane et al.
Fig. 2 TGA-MS curve of a
VOHPO4ꢀ0.5H2O and b
VO(H2PO4)2
proceeded through complex processes which include
dehydration, oxidation and reduction in the reaction
condition [15, 30–32]. The (VO)2P2O7 catalyst activated
under reaction conditions usually compose of trace
amount of VOPO4 phases in addition to the main phase
(VO)2P2O7. On the other hand, it was demonstrated that
the direct topotactic transformation of VOHPO4ꢀ0.5H2O
to pure (VO)2P2O7 catalyst phase can be achieved under
controlled inert atmospheric condition such as N2 or He
flow [29, 33, 34]. The absence of the XRD detectable
VOPO4 phase formation in our study (Fig. 3) confirms
these earlier observations by Johnson et al. [29] and
Bordes et al. [32].
The in situ XRD of the (VO)2P2O7 sample recorded at
room temperature after heating to 750 °C showed that only
the (VO)2P2O7 phase was present, showing that it is stable
once cooled to room temperature under the in situ activa-
tion conditions. These structural changes observed during
the in situ XRD indicate the complexity involved during
the transformation of the VOHPO4ꢀ0.5H2O precursor to a
more crystalline (VO)2P2O7 catalyst phase [34]. Much of
the variations in the peaks shape and intensities probably
123

Liquid Phase Oxidation of p-cymene
109
For preparation of the final activated catalysts, the
VOHPO4ꢀ0.5H2O precursor was activated at 550 °C for 4 h
under N2 flow conditions in a tube furnace to (VO)2P2O7
catalyst phase (Fig. 5a) while the VO(H2PO4)2 at 650 for
4 h under N2 flow to VO(PO3)2 catalyst (Fig. 5b). The final
XRD pattern of the VO(PO3)2 catalyst material showed
the presence of two phases, a mixture of b-VO(PO3)2
(tetragonal) and VO(PO3)2 (monoclinic) phases. Using the
structural information available for these phases from
ICSD [27] and doing quantitative Rietveld refinement of
the whole XRD VO(PO3)2 diffraction pattern [26], the
sample contained about 84.96% b-VO(PO3)2 and 15.04%
VO(PO3)2 phases. The ex-situ XRD of the (VO)2P2O7
catalyst obtained in Fig. 5a showed to give more crystalline
phase of (VO)2P2O7 catalyst when compared to the one
obtained under the in situ XRD activation conditions (at
550 °C in Fig. 3). This difference in the crystallinity can be
related to the activation temperature exposure period which
has been shown to have an impact on the catalyst crystal-
linity and surface area [36]. It should be noted that the in
situ XRD diffractogram in Fig. 3 (550 °C) was obtained
after the catalyst spent about 55 min at 550 °C under N2
flow while the ex-situ XRD diffraction patterns illustrated
in Fig. 5a was recorded at room temperature after the
catalyst was calcined under N2 conditions at 550 °C for 4 h
in a tube furnace.
Fig. 3 In-situ XRD structure transformation of VOHPO4ꢀ0.5H2O to
(VO)2P2O7
have an impact on the physical properties of the catalyst,
thus its catalytic performance [35].
Figure 4 illustrate the in situ XRD pattern of the
VO(H2PO4)2 precursor transformation with temperature
studied under the same condition as the in situ XRD for
VOHPO4ꢀ0.5H2O precursor. The results showed different
VO(H2PO4)2 precursor decomposition pattern when com-
pared to the VOHPO4ꢀ0.5H2O precursor. At a temperature
range of 500–600 °C, the VO(H2PO4)2 structure com-
pletely collapsed to form an amorphous phase (Fig. 4).
Different to the previous VOHPO4ꢀ0.5H2O sample, the
VO(H2PO4)2 material remained partially amorphous up to
650 °C, where the formation of the VO(PO3)2 phase was
observed. The formation of vanadium trimetaphosphate,
VO(PO3)3 usually observed in the VO(PO3)2 catalyst phase
material was not present in the final activated VO(PO3)2
catalyst. Again, the in situ XRD pattern recorded at 750 °C
was very similar to the one done subsequently at room
temperature, indicating that the VO(PO3)2 catalyst material
remained stable once cooled to room temperature.
The BET surface area and particle size of the precursors
[VOHPO4ꢀ0.5H2O and VO(H2PO4)2] and their respective
activated catalysts [(VO)2P2O7 and VO(PO3)2] are sum-
marised in Table 1. The VO(H2PO4)2 phase and its acti-
vated catalyst displayed a lower surface area when
compared to the VOHPO4ꢀ0.5H2O precursor and its active
(VO)2P2O7 catalyst. The higher surface area exhibited by
the VOHPO4ꢀ0.5H2O and (VO)2P2O7 phases could be
related to their respective smaller particle size which
occurred from the different preparative methods used for
the catalyst types. The obtained low BET surface area of
the VO(H2PO4)2 precursor and its activated catalyst
VO(PO3)2 was typical for this type of VPO catalyst phase
[13].
3.2 Catalytic Activity Testing
The catalytic behaviour of (VO)2P2O7 and VO(PO3)2 cat-
alysts were evaluated in p-cymene oxidation. The non-
catalytic p-cymene oxidation in the presence of an initiator
was performed in order to evaluate the oxidation rates of
the non-catalytic reaction and the catalytic effect contri-
bution from the reactor system components such as the
internal glass reactor surface and the stainless steel
impeller-type blade that may occur. Only the conversion of
less than 1% after 5 h was obtained as illustrated in Fig. 6
for non-catalysed reaction when compared to more 30% in
Fig. 4 In-situ XRD structure transformation of VO(H2PO4)2 to
VO(PO3)2
123

110
P. R. Makgwane et al.
Fig. 5 XRD rietveld refined structures of: a (VO)2P2O7, and b VO(PO3)2 catalyst
oxidation reaction time was prolonged for non-catalytic
reaction a conversion of about 16% was achieved after a
reaction period of 8 h. The long induction period observed
in the non-catalysed p-cymene oxidation indicate the dif-
ficulty of initiating the autoxidation reactions by activation
of the C–H bond energies. However, when the p-cymene
oxidation was done in the presence of the two catalysts
[(VO)2P2O7 and VO(PO3)2], respectively, the improved
oxidation rates were observed with initiation time reduced
to about less than 1 h (Fig. 6).
Table 1 Surface parameters of the VPO catalysts
Catalyst
BET SA (m2 g-1
)
Particle size
*D [4.3] (lm)
VO(H2PO4)2
VOHPO4ꢀ0.5H2O
VO(PO3)2
0.53
25.0
27.34
17.24
7.83
0.74
(VO)2P2O7
36.34
3.95
* D [4.3] Volume mean diameter
Table 2 shows comparison rates of (VO)2P2O7 and
VO(PO3)2 catalysts in p-cymene oxidation. Both catalysts
showed improved oxidation rates resulting in conversions
of more than 30% when compared to non-catalytic oxida-
tion reaction. These conversions are significantly better
the catalysed reaction by the two catalysts, (VO)2P2O7 and
VO(PO3)2. The low level of conversions suggested that
none or very little catalysis results from the internal reactor
components and any catalytic effects may therefore safely
be interpreted in terms of the added catalyst. When the
123

Liquid Phase Oxidation of p-cymene
111
phase in the (VO)2P2O7 catalyst will not influence the
catalytic performance of the (VO)2P2O7 catalyst for liquid
phase oxidation of p-cymene to the same extent as reported
for other gas phase oxidation reactions [23]. More inter-
estingly was that the results showed that both VO(PO3)2
and (VO)2P2O7 catalysts phase seemed to expose some
minimal catalytic site that are important for p-cymene
liquid phase oxidation.
In the liquid phase oxidation of p-cymene, the inter-
mediate TCHP is produced as the main product along with
several side products such as primary cymene hydroper-
oxide, p-methylacetophenone, p-isopropylbenzoic acid,
p-isopropylbenzaldehyde and p-isopropylbenzyl alcohol.
Table 3 shows the typical concentration profiles of
p-cymene liquid phase oxidation products distribution
catalysed by (VO)2P2O7 catalyst. The results showed that
the main product was the desired TCHP with about 25%
yield at about 35% substrate conversion. In addition, the
effect of the prolonged reaction time and high conversion
was clearly indicated by the high amount formation of
by-products (Table 3) and the drop of TCHP selectivity
(Table 2 for (VO)2P2O7 catalyst) as the conversion and
reaction time increased. As results, it is a common practice
in liquid phase oxidation as in many complicated reaction
systems to stop the conversion where the selectivity
towards the desired product and space time yield are sat-
isfactory. In the present study it appears as if the conver-
sions up to 30% gives a superior process performance of
both selectivity and reaction rates.
Fig. 6 Comparison of catalysed and non-catalysed p-cymene oxida-
tion rates. Symbols (filled square) (VO)2P2O7; (open triangle)
VO(PO3)2 and (filled circle) no catalyst. Other reaction conditions:
p-cymene (35 g), initiator (2 g), O2 flow rate 40 ml min-1, temper-
ature 100 °C and agitation rate (2,000 rpm)
than the typical 15–20% conversions for industrial-scale
p-cymene oxidations and in significantly shorter reaction
times (3 h as compared to 8–12 h). Surprisingly, the
VO(H2PO4)2 derived catalyst, VO(PO3)2 which is associ-
ated with poor catalytic performance [23, 24], also showed
substantial high catalytic activity performance during the
liquid phase oxidation of p-cymene relatively comparable
to the (VO)2P2O7 catalyst. However, its selectivity
(Table 2) was somewhat poorer when compared to
(VO)2P2O7 catalyst phase. The (VO)2P2O7 catalyst also
showed slightly better oxidation rates when compared to
VO(PO3)2 catalyst phase. The observation that the
(VO)2P2O7 catalyst showed both higher activity perfor-
mance and TCHP selectivity may possibly originate from
the higher surface area (36.95 m2 g-1) when compared to
VO(PO3)2 catalyst (0.74 m2 g-1) and the unique surface
layer catalytic sites possessed by the (VO)2P2O7 phase. The
results seem to suggest that the presence of the
VO(H2PO4)2 precursor phase in the VOHPO4ꢀ0.5H2O
precursor or its activated catalyst in a form of VO(PO3)2
The observed results in the present study are signifi-
cantly better when compared to the previous reported by
Senseman et al. [31] and Serif et al. [2] in the liquid phase
oxidation of p-cymene. In particular, Serif et al. [2]
obtained the hydroperoxide content of about 19% (after
197 h) and 40% (after 35.5 h) at 60 and 80 °C, respec-
tively. More interestingly were the product distributions
obtained by Senseman et al. [31] where the main reaction
products were found to be p-toluic acid (or p-methyl
Table 2 Comparison of (VO)2P2O7 and VO(PO3)2 activity in p-cymene oxidation
Time (minute)
Conversion mol%
(VO)2P2O7
TCHP selectivity (mol %)
TCHP yield (mol %)
(VO)2P2O7
VO(PO3)2
(VO)2P2O7
VO(PO3)2
VO(PO3)2
60
2.5
6.8
0.7
2.2
100
100
93.3
2.5
6.3
0.7
2.1
90
94.8
85.2
79.8
77.4
69.3
120
150
180
210
15.6
25.7
33.2
38.1
9.5
83.7
77.2
71.5
64.1
13.4
20.5
24.7
26.4
7.9
16.9
21.4
27.7
13
15.3
17.8
35 g p-cymene, 2 g initiator (oxidised cymene oil with 58% content of cymene hydroperoxide), 40 ml min-1 O2 flow rate, 2,000 rpm stirring
rate and 100 °C
TCHP Tertiary cymene hydroperoxide
123

112
P. R. Makgwane et al.
and oxidation rates than the VO(PO3)2 catalyst. The in situ
XRD results showed that the crystallinity of the (VO)2P2O7
catalyst seemed to be increasing with increase in activation
temperature and under the in situ conditions no formation
of other phases such as VOPO4 were observed. The
amorphous phase was obtained at in situ XRD temperatures
below 600 °C for VO(H2PO4)2 whereas above 650 °C, a
well crystalline phase of VO(PO3)2 was observed.
Table 3 Product distribution of p-cymene oxidation using
(VO)2P2O7 catalyst
Time (minute) (mol %)
p-Cymene TCHP PMAP PIBAL PCHP Cumic
0
100.0
97.5
93.2
84.4
74.3
66.8
61.9
0.0
2.5
0.0
0.0
0.0
0.5
1.6
4.2
5.8
0.0
0.0
0.0
0.4
0.9
1.1
1.3
0.0
0.0
0.4
0.8
1.4
1.4
1.5
0.0
0.0
0.0
0.5
1.3
1.8
3.1
60
90
6.4
120
150
180
210
13.4
20.5
24.7
26.4
Acknowledgments We thank NRF, Tshumisano and Sasol for
financial support. R.P. Forbes and M.K. Rayner are thanked for their
assistance in the recording of the in situ XRD data.
Reaction conditions: 35 g p-cymene, 2 g initiator (oxidised cymene
oil with 58% content of cymene hydroperoxide), 40 ml min-1 O2
flow rate, 2,000 rpm stirring rate and 100 °C
References
1. Fiege H (1991) Ullmann’s encyclopedia of industrial chemistry,
cresols and xylenols, Wiley-VCH Verlagsgesellschaft, A8:25
2. Serif GS, Hunt CF, Bourns AN (1953) Can J Chem 31:1229
3. Clonts KE, McKetta RA (1978) in: Kirk-Othmer (3rd Ed.),
Encyclopaedia of chemical technology: ‘‘cresols and cresylic
acids’’, Interscience, New York, 3:212
TCHP Tertiary cymene hydroperoxide; PMAP p-Methyl acetophe-
none; PIBAL p-Isopropyl benzaldehyde; PCHP Primary cymene
hydroperoxide; Cumic p-Isopropyl benzoic acid
benzoic acid), formaldehyde, formic acid, terephthalic
acid, CO2 and water which was different to the products
achieved in this work and by others [1, 2]. Based on the
results achieved in the present study of using the
(VO)2P2O7 as catalyst, a higher catalytic performance for
the liquid phase oxidation of p-cymene to TCHP in terms
of the process space time yield and product selectivity can
be achieved by improving the kinetic rate determining step
that are required for the C–H bond activation. Based on the
catalytic performance results of the two VPO catalysts, the
studies on (VO)2P2O7 catalyst long-term performance and
catalytic properties (e.g. structure stability, leaching,
deactivation and regeneration) in liquid phase oxidation of
p-cymene are reported elsewhere [37].
4. Sheldon RA, Kochi JK (1981) Metal-catalyzed oxidation of
organic compounds. Academic Press, New York
5. Suresh AK, Sharm MM, Sridhar T (2000) Ind Chem Eng Res
39:3958
6. Di Serio M, Cozzolino M, Tesser R, Patrono P, Pinzari F, Bonelli
B, Santacesaria E (2007) Appl Catal A Gen 320:1
7. Sookraj SH, Engelbrecht D (1999) Catal Today 49:161
8. Argyle MD, Chen K, Bell AT, Iglesia E (2002) J Catal 208:139
9. Pillai UR, Sahle-Demessie E (2003) N J Chem 27:525
10. Hutchings GJ (2004) J Mater Chem 14:3385
11. Centi G (1993) Catal Today 16:147
12. Bordes E (1988) Catal Today 3:163
13. Guliants VV, Benziger JB, Sundaresan S, Wachs IE, Jehng JM
(1996) Catal Today 28:275
¨
14. Bluhm H, Havecker M, Kleimenov E, Knop-Gericke A,
¨
Liskowski A, Shlogl R, Su DS (2003) Topics Catal 23:99
15. Hutching GJ, Chomel AD, Olier R, Volta JC (1994) Nature
368:41
16. Centi G, Frifiru¨ F (1988) Chem Rev 88:55
17. Mars P, van Krevelen DW (1954) Chem Eng Sci 3:41
18. Xue ZY, Schrader GL (1999) J Phys Chem B 103:9459
19. Hutchings GJ, Sananes MT, Sajip S, Kiely CJ, Burrows A,
Ellison IJ, Volta JC (1997) Catal Today 33:161
20. Taufig-Yap YH, Saw CS (2008) Catal Today 131:285
21. Kamiya Y, Hiyosni N, Ryumon N, Okuhara T (2004) J Mol Catal
A Chem 220:103
22. Koyano G, Okuhara T, Misono M (1998) J Am Chem Soc
120:767
23. Bartley JK, Wells RPK, Hutchings GJ (2000) J Catal 195:423
24. Hannour FK, Martin A, Kubias B, Lu¨cke B, Bordes E, Courtine P
(1998) Catal Today 40:263
25. Powder diffraction file-2, International Centre for Diffraction
Data, Pennsylvania, USA, Release 2004
26. Topas V3: General profile and structure analysis software for
powder diffraction data; Bruker AXS, Karlsruhe, Germany, 2005
27. Inorganic Crystal Structure Database V 1.4.1, Fachinformat
Ionszentrum Karlsruhe, Germany, 2006
28. Harmse NI (2005) DTech thesis, Nelson Mandela Metropolitan
University
29. Johnson JW, Johnston DC, Jacobson AJ, Brody JF (1984) J Am
Chem Soc 106:8123
4 Conclusion
Both (VO)2P2O7 and VO(PO3)2 catalysts showed improved
p-cymene oxidation rates when compared to the non-cat-
alysed industrial-scale oxidation process with substrate
conversions of above 30% achieved in 3 h. Suprisingly,
even the VO(PO3)2 associated with poor catalytic perfor-
mance in gas phase oxidations showed a high catalytic
performance for p-cymene oxidation closely comparable to
the (VO)2P2O7 catalyst. This indicated that the presence of
VO(H2PO4)2 in VOHPO4ꢀ0.5H2O or its activated catalyst
in a form of VO(PO3)2 in VOHPO4ꢀ0.5H2O precursor
activated catalyst, (VO)2P2O7 will not influence the cata-
lytic performance of the (VO)2P2O7 catalyst for liquid
phase oxidation of p-cymene to the same extent as reported
for other gas phase oxidation reactions. However, the
(VO)2P2O7 showed to give slight better TCHP selectivity
123

Liquid Phase Oxidation of p-cymene
113
30. Centi G (1993) Catal Today 16:5
35. Kiely CJ, Burrowa A, Sajip S, Hutchings GJ, Sananes A, Tuel A,
Volta JC (1996) J Catal 162:31
36. Bordes E, Courtine P (1984) J Solid State Chem 55:270
37. Makgwane PR, Ferg EE, Zeelie B (2010) Appl Catal A Gen
373:132
31. Senseman CE, Stubbs JJ (1932) Ind Eng Chem 24:1184
32. Abon M, Bere KE, Tuel A, Delichere P (1995) J Catal 156:28
33. Kiely CJ, Burrowa A, Hutchings GJ, Bere KE, Volta JC, Tuel A,
Abon M (1996) Faraday Discuss 105:103
34. Nguyen PT, Sleight AW, Roberts N, Warren WW (1996) J Solid
State Chem 122:259
123