Tertiary butylation of phenol on Cu1-xCoxFe 2O4: Catalysis and structure-activity correlation

Article abstract of DOI:10.1016/j.jcat.2003.12.001

A systematic study of catalytic tertiary butylation of phenol was carried out with isobutene as a function of temperature, feed composition, time on stream, space velocity, and catalyst composition on Cu_1-xCo _xFe₂O₄ (x=0 to 1) system. Tertiary butylation of phenol gives three products, namely, 2-tert-butyl phenol, 4-tert-butyl phenol, and 2,4-di-tert-butyl phenol. The phenol conversion and selectivity of these products depend on the reaction parameters. A good correlation was found between the activity, in terms of phenol conversion and various product selectivities for this reaction, and the acid-base properties of the catalysts. High activity is achieved with x=0.5 composition, illustrating the importance of a 1:1 combination of Cu and Co and the necessity for optimum concentrations of acid-base centers for this reaction. A reaction mechanism involving the interaction of phenoxide from phenol and the tert-butyl cation from isobutene on Cu_1-xCo_xFe₂O₄ is proposed. X-ray photoelectron spectroscopy and X-ray induced Auger electron spectroscopic analysis of fresh and spent catalysts revealed a partial reduction of metal ions due to reaction. Valence band studies clearly revealed an increase in the overlap of metal ion 3d bands from fresh to spent catalysts as reflected from a large decrease in the energy gap between them. The better catalytic results observed with x=0.5 are attributed to an optimum distribution of Cu species with heteroatom neighbors, maximum overlap between the Cu and Co 3d bands, and intermediate acid-base character.

Full text of DOI:10.1016/j.jcat.2003.12.001

Journal of Catalysis 222 (2004) 107–116
www.elsevier.com/locate/jcat
Tertiary butylation of phenol on Cu1−xCoxFe2O4:
catalysis and structure–activity correlation
Thomas Mathew, Bollapragada S. Rao, and Chinnakonda S. Gopinath ∗
Catalysis Division, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411 008, India
Received 7 July 2003; revised 24 November 2003; accepted 1 December 2003
Abstract
A systematic study of catalytic tertiary butylation of phenol was carried out with isobutene as a function of temperature, feed composi-
tion, time on stream, space velocity, and catalyst composition on Cu
Co Fe O (x = 0 to 1) system. Tertiary butylation of phenol gives
x
1−x
2 4
three products, namely, 2-tert-butyl phenol, 4-tert-butyl phenol, and 2,4-di-tert-butyl phenol. The phenol conversion and selectivity of these
products depend on the reaction parameters. A good correlation was found between the activity, in terms of phenol conversion and various
product selectivities for this reaction, and the acid–base properties of the catalysts. High activity is achieved with x = 0.5 composition, illus-
trating the importance of a 1:1 combination of Cu and Co and the necessity for optimum concentrations of acid–base centers for this reaction.
A reaction mechanism involving the interaction of phenoxide from phenol and the tert-butyl cation from isobutene on Cu
Co Fe O is
x
2 4
1−x
proposed. X-ray photoelectron spectroscopy and X-ray induced Auger electron spectroscopic analysis of fresh and spent catalysts revealed a
partial reduction of metal ions due to reaction. Valence band studies clearly revealed an increase in the overlap of metal ion 3d bands from
fresh to spent catalysts as reflected from a large decrease in the energy gap between them. The better catalytic results observed with x = 0.5
are attributed to an optimum distribution of Cu species with heteroatom neighbors, maximum overlap between the Cu and Co 3d bands, and
intermediate acid–base character.
2003 Elsevier Inc. All rights reserved.
Keywords: Tertiary butylation; Phenol; Cu
overlap; Metal ion distribution
Co Fe O ; Ferrospinel; X-ray photoelectron spectroscopy; Acidity; Basicity; Acid–base pair; Valence band
x
2 4
1−x
1. Introduction
depending on the catalysts and on the reaction conditions.
Most of the above-mentioned catalysts are focused on the
production of 4-tertiary butylphenol (4-tBP), 2,4-dtBP, and
2,6-dtBP. 2-tBP is produced as the main product of phenol
butylation with isobutene in the presence of an acidic ion-
exchange resin catalyst [1]. Though cation exchange resin
catalysts have an advantage due to environmental friendli-
ness, they have disadvantages due to tedious workup, less
activity, and less stability at high temperatures.
Ferrites of spinel-type structure are found to be highly
active toward many aromatic alkylation reactions such as
phenol alkylation, aniline methylation, and pyridine methy-
lation [15–19]. The catalytic effectiveness of these systems
is due to the ability of the metallic ions to migrate between
the sublattices without altering the structure, which makes
the catalyst efficient for many organic transformation reac-
tions. The Cu1−xCoxFe2O4 system was found to be highly
active for phenol alkylation with different alkylating agents
and has been characterized in detail [19–23]. The prime at-
traction of this catalyst system is its high ortho products
Tertiary butylation of phenol has been studied exten-
sively owing to industrial interest in the production of a
variety of materials [1]. For example, 2-tertiary butylphe-
nol (2-tBP) is a starting material for the synthesis of an-
tioxidants and agrochemicals. Triphosphate and benzotria-
zole derivatives of 2,4-ditertiary butylphenol (2,4-dtBP) are
employed as costabilizers for polyvinyl chloride and UV
absorbers in polyolefins, respectively. Various catalysts re-
ported for tertiary butylation of phenol include Cr2O3, ion-
exchange resins, ZrO2, sulfated zirconia, aluminum hydro-
talcites, molecular sieves, and zeolites including mordenite,
SAPO-11, HY, Hβ, AlMCM-41, and FeMCM-41 [2–14].
Unlike other phenol alkylation, tertiary butylation of phe-
nol gives numerous products including 3-tert-butylphenol
*
Corresponding author.
E-mail address: gopi@cata.ncl.res.in (C.S. Gopinath).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2003.12.001
2003 Elsevier Inc. All rights reserved.

108
T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
selectivity with high phenol conversion. Cu–Co combination
in terms of synergism and 3d energy band overlap were sug-
gested to be responsible for this efficient process [21]. As
part of continuing studies on alkylation in our laboratory, we
carried out tertiary butylation of phenol with isobutene (IB).
Effects of various parameters such as reaction temperature,
time on stream (TOS), space velocity, mole ratio of phenol to
IB, and catalyst composition were studied. Effects of these
parameters on phenol conversion and selectivity to various
products were also studied. A good correlation was found
between activity, in terms of phenol conversion and selectiv-
ity of various products for this reaction, and the acid–base
properties of the catalysts. Special attention was paid to the
associated changes in terms of surface composition and the
overall electronic structure of catalysts due to reaction.
the reactor in such a way that the catalyst was sandwiched
between the layers of inert porcelain beads. The upper por-
tion of the reactor was also filled with inert beads and served
as a vaporizer-cum-preheater; it was maintained at 200 ◦C
or the reaction temperature, whichever was higher, to ensure
that phenol was fully vaporized before it reached the cata-
lyst bed. The feed mixture containing phenol and IB gas in
the appropriate ratio was passed simultaneously. IB flow was
measured with a mass flow controller. Phenol was heated at
55 ◦C and pumped through a syringe pump (ISCO, Model
500D). The feed line from the pump to the reactor was also
heated at 55 ◦C. The products were collected in a condenser
cooled with ice-cold water and the analysis was performed
with GC, GC-MS, and GC-IR methods. The accuracy of the
catalytic activity values reported in this article was deter-
mined by repeated measurements and is reproducible within
5%.
2. Experimental
Cu1−xCoxFe2O4 (x = 0, 0.25, 0.50, 0.75, and 1) samples
were prepared by coprecipitation technique and character-
ized by various physicochemical techniques as reported in
earlier articles [20–23]. Briefly, stoichiometric amounts of
premixed metal nitrate solutions were added to NaOH solu-
tion with continuous stirring. The final pH of the resulting
solution was adjusted to between 9.5 and 10. The precipitate
3. Results
3.1. Characterization of Cu1−xCoxFe2O4
Catalysts were characterized with respect to chemical
composition structural and textural properties have already
been reported [20–23]. In Fig. 1 shows the XRD patterns
of calcined (a) and spent (b) catalysts. Fresh Cu-rich cata-
lyst (x = 0) exhibits a diffraction pattern attributed to cu-
bic spinel phase and a considerable amount of CuO and
Fe2O3. Substitution of Cu with Co increases the overall crys-
tallinity of the spinel phase and all the peaks are indexed
(ASTM Card Nos. 1-1121 and 3-0864). The XRD pattern of
spent catalysts displays Cu0 at x = 0 along with CuO and
Fe2O3; however, only a trace amount of Cu0 is detected at
0.25 ꢀ x ꢀ 0.75 and there is no major change on CoFe2O4
after reaction. Table 1 lists the unit cell values (a) for all
catalysts, which decrease with increasing x. Deviation at
x = 0 composition is due mainly to the significant amount of
impurities present. Surface area and pore volume (Table 1)
have relatively high values at x = 0.5; however, other com-
positions, except x = 0.0, are associated with comparable
values, hinting that there might be no major influence of tex-
tural properties on catalytic properties. Relative acidity and
basicity values were measured from the pyridine and CO2
adsorbed on the catalyst surfaces at 100 ◦C and room tem-
perature, respectively [23]. A systematic increase (decrease)
in relative acidity (basicity) is observed with increasing x
value.
was washed with demineralized water until Na+ and NO3
−
ions disappeared. The precipitates were filtered, dried at
80 ◦C, and calcined at 500 ◦C. The chemical compositions of
calcined (fresh) catalysts were determined by X-ray fluores-
cence (XRF) spectroscopy. X-ray diffraction (XRD) patterns
of the powder catalysts were recorded with Cu-Kα radiation
(λ = 1.5405 Å) with a Ni filter to obtain unit cell parame-
ter (a) and crystallite size. The BET surface area and pore
volume of the catalysts were determined by a N2 adsorption–
desorption method at 77 K using a Quantachrome NOVA-
1200 adsorption unit. For X-ray photoelectron spectroscopy
(XPS) measurements, a VG Microtech Multilab ESCA 3000
spectrometer with a non-monochromatized Mg-Kα X-ray
source (hν = 1253.6 eV) was employed with in situ scraped
fresh catalyst pellets and powder samples of spent catalysts.
Energy resolution of the spectrometer was set at 0.8 eV with
Mg-Kα radiation at a pass energy of 20 eV. Binding en-
ergy calibration was performed with the Au 4f7/2 core level
at 83.9 eV. Spent catalysts analyzed by XRD and XPS, af-
ter butylation at 200 ◦C for 10 h, had a 1:3 composition of
PhOH:IB, unless otherwise stated.
The reactions were conducted in a vapor-phase setup at
atmospheric pressure in a vertical, downflow, fixed-bed glass
reactor 17–18 mm in internal diameter. This glass reac-
tor was placed inside a split-type furnace (Geomechanique,
France), which can maintain different temperatures in the
upper and lower halves [22]. Three grams of fresh cata-
lyst with a particle size around 20 mesh (different particle
sizes did not have any significant effect on the kinetic re-
sults reported here) was charged each time in the center of
3.2. Catalytic activity studies
3.2.1. Effect of isobutene:phenol mole ratio
Tertiary butylation of phenol was carried out at 200 ◦C
with various IB:phenol mole ratios (IB:PhOH) between 1
and 7 on Cu0.5Co0.5Fe2O4 to determine the optimum feed
ratio (Fig. 2a). The main products of tertiary butylation of

T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
109
Fig. 2. (a) Isobutene:phenol feed composition dependence of phenol conver-
sion and selectivity of all tertiary butylation products on Cu Co Fe O ,
0.5 0.5
2 4
◦
at 200 C and TOS of 3 h. (b) Catalyst composition dependence of phe-
nol conversion and selectivity of tertiary butylation products at 200 C and
◦
IB:PhOH of 3 are shown for Cu
lectivity of 4-tert-butylated phenol with increase in Co content and increase
Co Fe O . Note an increase in the se-
x
2 4
1−x
in isobutene in the feed.
Table 2
Influence of catalyst composition on phenol conversion and product distri-
a
bution over Cu
Co Fe O (x = 0, 0.25, 0.50, 0.75 and 1)
x
1−x
2 4
Product distribution (mol%)
Catalyst composition, x
0
0.25 0.50 0.75
1
2-tert-butyl phenol
4-tert-butyl phenol
2,4-Di-tert-butyl phenol
2,6-Di-tert-butyl phenol
Others (including o-cresol
and 2,6-xylenol)
6.49 8.25 12.76 3.97 1.50
4.38 5.83 8.19 4.46 4.37
2.08 1.41 7.49 1.59 0.34
Fig. 1. X-ray diffractograms of (a) fresh and (b) spent Cu
Co Fe O
x
2 4
1−x
0.17 0.13 0.48 0.12
0
catalysts. X-ray diffractogram of spent catalysts were obtained after ter-
◦
0.29 0.30 0.70 0.20 0.16
tiary butylation of phenol at 200 C for 10 h with a 1:3 composition of
phenol:isobutene. Note that a significant amount of metallic Cu is seen for
Phenol conversion
2-tert-butyl phenol selectivity
4-tert-butyl phenol selectivity
13.40 15.92 29.62 10.34 6.37
48.43 51.82 43.08 38.39 23.55
32.69 36.62 27.65 43.13 68.60
x = 0.0 spent catalyst.
2,4-Di-tert-butyl phenol selectivity 15.52 8.86 25.29 15.38 5.34
Table 1
Total ortho selectivity
49.70 52.64 44.70 39.45 23.55
Textural and acid–base characteristics of Cu
Co Fe O
x
2 4
1−x
a
Phenol butylation is carried out at a phenol:isobutene mole ratio of 1:3
Composition,
a
S
Pore volume
Relative acidity
(basicity)
BET
2
◦
−1
.
−2
a
at 200 C and a WHSV of 1 h
x
(Å)
(m /g)
(cc/g) (10
)
0.0
8.3898
8.4051
8.4012
8.3982
8.3997
28.8
34.0
43.8
36.6
36.8
5.1
6.7
10.9
6.4
300 (580)
381 (510)
479 (420)
623 (275)
782 (210)
0.25
0.50
0.75
1.0
in the amount of aliphatics formed (results not shown) due to
oligomerization as well as cracking of IB. 4-tBP selectivity
increases as IB:PhOH increases up to 5 and thereafter mar-
ginally decreases. Though the selectivity of 4-tBP increased
with a large amount of IB in the feed, there is a perceptible
drop in the selectivity of both 2-tBP and 2,4-dtBP.
5.9
a
Obtained from FT-IR studies of pyridine and CO adsorbed cata-
2
lysts [23].
phenol are 2-tBP, 4-tBP, and 2,4-dtBP. A trace amount of
2,6-dtBP is also formed. It can be seen that phenol conver-
sion increases as the IB:PhOH increases to 3 and thereafter
marginally increases to 5 and then decreases. However, com-
paratively high yields of 2-tBP and 2,4-dtBP were observed
at a feed ratio of 3:1. Further, too much of either of the re-
actants in the feed deactivates the catalyst rapidly, hence an
optimum feed ratio of 3:1 was maintained for further stud-
ies. It is to be noted that the amount of unreacted IB increases
with increasing feed ratio. Concurrently, there is an increase
3.2.2. Effect of catalyst composition
Fig. 2b illustrates the catalyst composition dependence of
phenol conversion and selectivity of tertiary butylated phe-
nols on Cu1−xCoxFe2O4 at 200 ◦C with IB:PhOH mole ratio
of 3:1 and WHSV of 1 h−1. Conversion and product se-
lectivity of phenol tertiary butylation on Cu1−xCoxFe2O4
under optimum conditions are also summarized in Table 2.
The important observations are: (1) Phenol conversion in-
creases from x = 0 to 0.5 and then decreases with further
increase in x. (2) 2-tBP selectivity decreases as Cu con-

110
T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
tivity of 4-tBP and 2,4-dtBP also remains the same, around
25–30 mol%, up to 8 h; however, the selectivity changes af-
ter 8 h. It is to be noted here also that phenol conversion and
2,4-dtBP selectivity vary hand-in-hand, as in the temperature
dependence studies.
3.2.5. Effect of weight hourly space velocity (WHSV)
The effect of WHSV on catalytic performance was stud-
ied in the range 0.5–2 h−1 using the x = 0.5 catalyst at
200 ◦C (Fig. 3c). It can be seen from the data that both high
WHSV and low WHSV are not helpful for high catalytic
performance. Phenol conversion is maximum at an optimum
WHSV of 1 h−1. High phenol conversion with a good yield
of 2-tBP is achieved at a WHSV of 1–1.5 h−1. It was also
found that 2,4-dtBP selectivity decreases with increasing
WHSV. Addition of 2-tBP and 2,4-dtBP selectivity gives a
constant value of 67 mol% independent of WHSV values,
strongly indicating that tertiary butylation is sequential and
tertiary butylation at the ortho position takes place first fol-
lowed by a second tertiary butylation at the para position.
This also suggests that there is no isomerization between 2-
tBP and 4-tBP. 4-tBP selectivity does not show significant
WHSV dependence.
Fig. 3. Tertiary butylation of phenol is shown on Cu Co Fe O , with
IB:PhOH = 3, as a function of (a) reaction temperature, (b) time on stream,
and (c) weight hourly space velocity.
0.5 0.5
2 4
tent decreases. (3) 4-tBP selectivity increases at x > 0.5,
however, with a concurrent decrease in phenol conversion.
(4) x = 0.5 is associated with maximum phenol conversion
and high selectivity towards 2,4-dtBP formation. A sudden
increase in 2,4-dtBP and a parallel decrease in 2-tBP selec-
tivity at x = 0.5 hints that tertiary butylation is sequential.
3.3. XPS analysis
3.2.3. Effect of reaction temperature
XPS has been used to examine the changes in oxidation
state and surface composition of catalysts. Spent catalyst of
x = 0.5 composition obtained after reaction for 10 h with a
1:3 ratio of IB:PhOH at 300 ◦C (hereafter, x = 0.5@300 ◦C)
was also subjected to XPS analysis.
Fig. 3a shows the effect of reaction temperature on the
conversion of phenol and selectivity of various products on
x = 0.5 at a space velocity of 1 h−1 and TOS of 3 h. Phe-
nol conversion increases from 175 to 200 ◦C and afterward
falls at high temperatures. Maximum phenol conversion is
observed at 200 ◦C. Phenol conversion increases from 3 to
30 mol% as the temperature changes from 175 to 200 ◦C.
The one-to-one change observed between phenol conversion
and 2,4-dtBP selectivity indicates the first-order dependence
of phenol conversion on 2,4-dtBP yield. An increase in tem-
perature does not change 2-tBP selectivity significantly. 4-
tBP selectivity also remains the same from 175 to 350 ◦C,
except at 200 ◦C. A significant increase in 2,4-dtBP selectiv-
ity at the cost of 4-tBP and 2-tBP selectivity indicates again
that at high phenol conversion, the reaction pattern shifts sig-
nificantly to ditertiary butylation; this point is worth investi-
gating further. It is found that as the temperature increases,
the amount of unreacted IB decreases and the amounts of
aliphatics increase.
3.3.1. Cu 2p core level and Cu–L3M45M45
Photoemission spectra from the Cu 2p3/2 core level for
both fresh and spent Cu1−xCoxFe2O4 catalysts are shown
in Figs. 4a and b, respectively. It is clear that all fresh cat-
alysts exhibit the Cu 2p3/2 main peak at a binding energy
(BE) of 934.2 0.2 eV with good satellite intensity (IS)
around 942 eV, confirming the existence of Cu2+. The in-
tensity of satellite to mainline (IS/IM) is between 0.5 and
0.6 in all cases (Fig. 4d), which is very close to that of CuO
(0.55) [24]. The above facts indicate the charge density on
Cu is the same at all catalyst compositions.
Two different features are observed for all spent catalysts
with low IS. The first feature at 934.2 0.2 eV corresponds
to Cu2+ as in the calcined catalysts, and the second feature
at 932.6 0.2 eV corresponds to Cu+ and/or Cu0 species.
Cu2+ has a satellite feature due to its open-shell 3d9 config-
uration, while Cu0/Cu+ does not show a satellite due to its
closed-shell configuration (3d10). The intensity of main line
also decreases from fresh to spent catalysts. A careful con-
volution (not shown) reveals that the intensity of the low-BE
component increases with increase in x. x = 0.5@300◦C
displays a high-intensity feature at 932.6 eV and a shoulder
at 934 eV with a very weak satellite. This indicates that the
3.2.4. Effect of time on stream
The effect of TOS is expected to shed light on deactiva-
tion of any catalyst and its influence on product selectivity.
Since x = 0.50 was found to be a better catalyst for the phe-
nol tertiary butylation reaction, the TOS dependence of this
catalyst at 200 ◦C was investigated and is shown in Fig. 3b.
It was found that the conversion of phenol is high up to
5 h and then decreases significantly. The selectivity of 2-tBP
remains constant at 45 mol% irrespective of TOS. The selec-

T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
111
Fig. 5. Co 2p
core-level photoemission spectra of (a) fresh and (b) spent
3/2
Cu
spent Cu
Co Fe O catalysts. (c) XAES from Co L M
M
of fresh and
XPS of spent
3/2
x
1−x
2
4
3
45 45
Co Fe O catalysts for x = 0.5 and 1. Co 2p
x
1−x
2
4
◦
catalyst at a reaction temperature of 300 C is also displayed in (b) and (c)
to show the effect of higher temperature in increasing surface Co concen-
tration. The changes observed in intensity at x = 0.5 from fresh to spent
catalysts are indicated by two arrows.
Table 3
XPS parameters from Co 2p core level in Cu
Co Fe O catalysts
x
2 4
1−x
Composition, x
BE of Co 2p
(mainline–satellite energy gap, eV)
mainline
2p –2p
3/2 1/2
energy gap (eV)
Fig. 4. Cu 2p
photoemission spectra of (a) fresh and (b) spent
3/2
3/2
Cu
Co Fe O catalysts. Broadening on the lower-binding-energy side
x
2 4
1−x
2+
F a
demonstrates a partial reduction of Cu
on spent catalysts. Cu 2p
3/2
0.5
780.5 (4.7)
780.2 (3.4 and 6.1)
780.9 (4.6 and 6.1)
780.5 (6.2)
780.2 (5.7)
780.4 (6.4)
780.5 (6.7)
780.1 0.3
779.6
15.5
15.6
15.5
15.9
16.0
15.8
15.9
15.5
–
◦
F
XPS of spent catalyst at a reaction temperature of 300 C is also displayed
in (b) to show the extent of the larger reduction due to higher temperature.
0.75
F
1.0
0.5
(c) Cu L M
M
45 45
Auger electron spectra from spent catalysts. (d) Satel-
S
3
S (300 ◦C)
lite-to-main line intensity ratio (I /I ) for fresh and spent catalysts and
M
S
0.5
+
2+
the ratio of Cu to Cu on spent catalyst. The reducibility of Cu increases
with Co content.
S
0.75
S
1.0
b
CoO
high reaction temperature (300 ◦C) leads to a large reduction
b
Co O
2
3
4
in Cu2+
.
b
Co O
3
780.5
15.0
a
Lower-valence Cu species can be distinguished by exam-
ining the Cu–L3M45M45 Auger peaks shown in Fig. 4c. All
the compositions exhibit a peak at a kinetic energy (KE)
around 917.8 eV, characteristic of Cu2+ species. Addition-
ally, a second component is also observed at about 916.2 eV
and it is very intense on x = 0.5@300 ◦C, which is charac-
teristic of Cu+ species. It should be noted here that the KEs
of the Cu–L3M45M45 Auger peak values for Cu, CuO, and
Cu2O are 918.4, 917.6, and 916.5 eV, respectively [25,26].
Present KE values of 916.2 and 917.8 eV suggest contri-
butions of Cu+ and Cu2+ species, respectively, on spent
catalysts and no Cu0.
Fig. 4d shows the IS/IM ratios of fresh and spent catalysts
and the reducibility of Cu. An important observation to be
noted here is that the IS/IM ratio of spent catalysts decreases
linearly from x = 0.0 to 0.5 and then increases marginally
at x = 0.75; however, fresh catalysts display a value close
to 0.55 and no significant change with x. It is also to be
noted that the IS/IM ratio changes considerably from fresh
to spent for x = 0 and 0.5 compositions; however, it remains
Superscripts F and S indicate fresh and spent catalysts.
Refs. [27–29].
b
the same for x = 0.25 and 0.75 compositions. Reducibility
of Cu, calculated as the ratio of Cu+ to Cu2+, increases lin-
early with increasing Co content. This is totally contrary to a
decrease in Cu reducibility for phenol methylation [21], and
ethylation and isopropylation [22] reactions.
3.3.2. Co 2p core level and Co–L3M45M45
Co 2p3/2 photoemission spectra for fresh (a) and spent (b)
catalysts are shown in Fig. 5 and the results are given in
Table 3. The important features are: (1) the intensity of
shakeup satellites decreases with increasing x; (2) the en-
ergy gap between the 2p3/2 peak and its satellite increases
with increasing x on fresh and spent catalysts; and (3) an
increase in the energy gap between Co 2p spin orbit dou-
blets is observed from fresh to spent. Spent catalyst of
x = 0.5@300◦C displays significantly high intensity fea-
tures, indicating a higher Co concentration on the surface.

112
T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
Table 4
Surface atomic ratio of fresh and spent Cu
tert-butylation reaction at 200 C for 10 h
Co Fe O catalysts after
x
1−x
2 4
◦
x
Fresh
Spent
Cu/Fe Co/Fe (Cu + Co)/Fe Cu/Fe Co/Fe (Cu + Co)/Fe
0.00 1.17
0.25 1.04
0.50 0.79
–
0.20
0.32
1.17
1.24
1.11
0.68
0.74
0.68
(0.44)
0.44
–
–
0.16
0.40
(0.47)
0.45
0.66
0.68
0.90
1.07
(0.91)
0.89
0.66
a
a
a
0.75 0.50
0.50
0.75
1.00
0.75
1.00
–
a
Results obtained from the sample subjected to phenol tert-butylation
◦
reaction at 300 C for 10 h.
This is further corroborated by surface composition val-
ues (see above) (Table 4). High temperature leads to large
segregation of Co2+ on the surface. Comparison of the BE
values of catalysts with those of standard Co compounds
(Table 3) clearly indicates the catalyst surfaces are in gen-
eral composed of both Co2+ and Co3+. A high IS for x = 0.5
suggests contributions by Co2+ and high-spin Co3+, since
low-spin Co3+ shows poor Is [27–29].
Co–L3M45M45 Auger spectra of fresh and spent catalysts
of x = 0.5 and 1 are shown in Fig. 5c exhibit broad features
between 760 and 780 eV. Two features at around 767 and
772 eV are virtually the same for fresh and spent catalysts
at x = 1. However, at x = 0.5, a shift in energy is observed
for the lower-KE component on spent catalyst surfaces com-
pared with fresh, and the shift is substantial for the catalyst
exposed to 300 ◦C. Further, the high intensity indicates a
large segregation of Co on the surface at 300 ◦C. x = 0.75
also shows characteristics similar to those x = 0.5.
Fig. 6. Fe 2p core-level XPS of (a) fresh and (b) spent Cu
Co Fe O
x
2 4
1−x
catalysts. A clear shift is observed in satellite energy from fresh to spent
catalysts (thick dotted arrows), and a weak shoulder seen below 710 eV on
spent catalysts (broken line) for intermediate compositions (0 < x < 1) in-
dicate partial reduction of iron. (c) XAES from Fe–L M
M
45 45
of fresh and
3
spent Cu
Co Fe O catalysts for x = 0.5 and 1 compositions. Broad-
x
1−x
2 4
ening due to an additional feature is denoted by a solid arrow and fresh
catalyst feature is denoted by the thick dotted arrow.
3.3.3. Fe 2p core level and Fe–L3M45M45
Fig. 6 shows the Fe 2p3/2 core level photoemission from
(a) fresh and (b) spent Cu1−xCoxFe2O4 catalysts. There
are two main differences between fresh and spent catalysts.
(1) The main peak at 711 eV with a clear satellite at 718.5 eV
due to Fe3+ is observed on fresh catalysts; IS is relatively
weak on spent catalysts and shifts to 719.5 eV. (2) Spent
catalysts also indicate a shoulder around 709.6 eV, character-
istic of Fe2+ species in tetrahedral coordination [30,31]; and
the overall broadening observed for the satellites of all spent
catalysts (except x = 1) supports contributions by both Fe3+
and Fe2+ [31]. Fig. 6c displays the Fe–L3M45M45 spectra
for x = 0.5 and 1.0 catalysts. Fresh catalysts are very similar
and no difference is observed. There is a clear line broaden-
ing on the spent catalysts and a new low-KE feature appears
as a shoulder on spent catalysts; however, it is not consider-
able at x = 1. A shoulder seen at 697.5 eV (broken line) on
all spent catalysts is characteristic of Fe2+ species [21], and
the lower intensity of this feature at x = 1 indicates the ex-
tent of Fe reduction might be low. x = 0.5@300 ◦C shows
comparatively large intensity of Fe2+, suggesting that high
temperature induces large reduction. Fe 2p and Fe–LMM
result indicate the partial reduction of Fe3+ to Fe2+ due to
the phenol tertiary butylation reaction.
Fig. 7. Valence band XPS of (a) fresh and (b) spent Cu
Co Fe O cat-
x
2 4
1−x
alysts. The significant broadening of the valence band observed on spent
catalysts is due to partial reduction of metal ions and consequent larger
overlap of 3d energy bands of metal ions.
3.3.4. Valence band photoemission
Changes observed on spent catalysts from core level are
reflected strongly in the valence band (VB) spectra too,
shown in Fig. 7 with details. The main VBs observed have
contributions predominantly from metal 3d bands below
10 eV and the assignments of various bands are from the
photoionization cross-section values [32] at hν = 1253.6 eV
and as explained earlier [21,22].
A few important observations are worth highlighting
from the data: (1) The Cu 3d bands observed at 4.2 eV on

T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
113
fresh catalysts (solid line) are found to be broadened on the
lower-BE side toward the Co 3d region (broken line) on
spent catalysts, indicating a reasonable overlap of energy
bands of Cu and Co 3d. This is marked by a considerable
dip in intensity between Cu and Co 3d bands on fresh cat-
alysts and no such dip on spent catalysts. Additionally, on
x = 0.50@300◦C, the 3d bands overlap is stronger in such
a way that the Cu 3d band is shifted to 3 eV and the energy
gap is just 1 eV between the Cu and Co 3d bands (dotted
arrow, small). (2) The Cu 3d8 satellite appears on all the
compositions containing Cu for fresh and spent catalysts;
however, the satellite is more broadened with less inten-
sity for intermediate compositions. (3) The shift in Fe 3d
bands to lower BE is discernible on spent catalysts, except
at x = 1. The shift in the trailing edge of VBs to lower BE
to 7–8 eV, compared with 9 eV on fresh catalyst, testifies to
the above. (4) There is a weak but finite photoemission in-
tensity at EF on spent catalyst (except x = 1), revealing the
conducting nature of the catalyst. (5) Co 3d bands remain
close to the same BE around 2.3 eV on fresh and spent cata-
lysts. The preceding points illustrate an appreciable increase
in the overlap of 3d energy bands on spent catalysts com-
pared with fresh catalysts due to a partial reduction of metal
ions.
Fig. 8. Comparison of atomic ratio of Cu + Co/Fe for all compositions
of fresh and spent catalysts (left panel) and phenol conversion and product
◦
yields on Cu
Co Fe O catalysts at 200 C, TOS = 3 h (right panel).
x
1−x
2 4
Note the correlation between high heterogeneity and high catalytic activity
for x = 0.5.
Cu and high Co content (Table 4). This is supported by the
poor activity of Co-rich x = 1; however, x = 1 is selective to
4-tBP. Despite the Fe-rich surface for x = 0 and x = 1, the
different reactivities are due to the Cu and Co species, re-
spectively. Even through there are a few differences, x = 0
and x = 0.25 have comparable activity, indicating the im-
portance of Cu to this reaction.
3.3.5. Surface composition
Table 4 summarizes surface composition for both fresh
and spent catalysts from XPS data. These data are extremely
helpful in understanding the distribution and heterogeneity
of metal ions on the surface, as it directly influences the cat-
alytic activity. The important points to be noted from Table 4
are: (1) Generally, Cu/Fe and Co/Fe ratios change linearly
with x on fresh as well as spent catalysts. (2) Cu + Co/Fe
values indicates the spent catalysts are generally enriched
with Fe compared with the respective fresh catalysts, except
at x = 0.5, indicating the heterogeneity of the metal ions
is maintained at a very high level on the surface. (3) Cu-
rich compositions show a large decline in Cu/Fe on spent
catalysts, whereas the Co/Fe ratio changes marginally from
fresh to spent for all x values. (4) x = 0.5@300◦C and x =
0.75 show almost the same atomic ratio; this is further cor-
roborated by the similar catalytic activity in Figs. 2b and 3a.
This last point hints at the synergetic behavior between reac-
tion temperature and catalyst composition. Coke depositions
of 10–20%, depending on the reaction conditions and cata-
lyst composition, were observed from the C 1s core level
and thermogravimetric analysis (data not shown) [22].
Fig. 8 shows the interdependence of surface composi-
tion ((Cu + Co)/Fe) on the left panel and PhOH conversion
with product yield on the right panel. It is clear from Fig. 8
that the heterogeneity dictates the catalytic activity to a large
extent and the same is exemplified for the x = 0.5 composi-
tion which demonstrates a high phenol conversion and better
product yield. This also hints at the importance of a 1:1 ra-
tio of Cu to Co for the reaction. Albeit good heterogeneity
for x = 0.75, the not so good activity is likely due to low
4. Discussion
4.1. Catalytic performance
From the activity data, it is evident that Cu-rich samples
are active for tertiary butylation and Co-dominant composi-
tions show poor activity, emphasizing the important role of
Cu in this reaction. However, the equal bulk contents of Cu
and Co at x = 0.5 show maximum activity and follow sim-
ilar trends as in other phenol alkylations [21,22]. Except at
x = 0.5, phenol conversion is between 5 and 16 mol% for all
other compositions, suggested that tertiary butylation activ-
ity is due mainly to the Cu + Co combination, of which Cu is
highly active. An important aspect of the butylation reaction
is related to product selectivity too. At an optimum temper-
ature of 200 ◦C, IB attaches to the phenolic ring to give tert-
butyl phenols. Unlike lower alkylations, tertiary butylation
of phenol gives three major products, namely, 2-tBP, 4-tBP,
and 2,4-dtBP, and the selectivity of each product depends on
the various reaction parameters. The total ortho selectivity
observed with any composition is ꢀ 50%, whereas phenol
methylation and ethylation are noted for high ortho selec-
tivity of > 90% [21,22]. Two reasons are the bigger size
and higher stability of t-butyl species compared with other
lower alkyl species, and hence it is worth analysis by in situ
spectroscopy methods. Further, the linear change observed

114
T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
between phenol conversion and 2,4-dtBP selectivity under
most conditions hints that 2,4-dtBP production is controlled
by first-order dependent phenol conversion and sequential
butylation of 2-tBP. High 4-tBP selectivity on x = 0.5 at
high IB:PhOH ratio (Fig. 2a) indicates the surface could be
dominated by Co, as in x = 0.5@300 ◦C.
and mesoporous materials [7,9]. However the requirement
for strong acidic sites for 2,4-dtBP on the above-mentioned
material is contrary to the present results, suggesting the p-
tert-butylation mechanism is different on ferrites. It is to be
noted that the catalyst having intermediate acid–base char-
acter (x = 0.5) has maximum yields of 2,4-dtBP and 2-tBP.
Alkylation of phenols, just like other aromatic com-
pounds, follows the Friedel–Kraft mechanism. From elec-
trophilic substitution it is clear that the tert-butyl cation at-
tacks the phenol ring preferentially in positions ortho and/or
para to the –OH group. The ratio of isomers formed ini-
tially is kinetically controlled and is determined by phenolic
OH groups and incoming tert-butyl cations in the present
case. Though the mono-tert-butylated phenols are more ac-
tive than phenol, subsequent tertiary butylation occurs only
when there is less steric hindrance.
Phenol is adsorbed as phenolate by dissociative adsorp-
tion on an acid–base site. Simultaneously, protonation of IB
occurs due to H+ from the above process. The protonated
IB rearranges to give an appreciably stable tert-butyl cation,
which in turn attacks phenoxide either from the adsorbed or
gaseous state. The relatively high selectivity of 2-tBP than
4-tBP indicates that the carbocation attacks preferentially
from the adsorbed state. However, high Co content reverses
the above trend and favors para-tert-butylation probably
from gas phase. FT-IR investigation of phenol methylation
with methanol on Cu1−xCoxFe2O4 shows exclusive ortho-
alkylation, due to the close proximity of the ortho positions
of phenoxide to the catalyst surface, compared with the para
positions [33]. Once formed, 2-tBP acts as a precursor to the
formation of 2,4-dtBP. Nonetheless, steric hindrance of the
tert-butyl group prevents the further attack by IB at the ortho
position of already formed 2-tBP, illustrating negligible for-
mation of 2,6-dtBP. In addition, once the 2-tBP forms, the
phenyl ring cannot be coplanar to the catalyst surface due
to the bulky size of the t-butyl group, indicating the para
position of the phenyl ring is far removed from the surface
and supported by negligible second ortho tertiary butylation.
However, the tert-butyl cation can attack, from the gaseous
state, phenol or 2-tBP, resulting in the formation of 4-tBP
and 2,4-dtBP, respectively. This is in contrast to the reac-
tion between phenol and any alcohol where the formation
of para alkylated product is < 1% [22]. Tertiary butylation
of phenol has been carried out with tert-butyl alcohol and
the conversion was found to be ꢀ 5% under any condition
and hardly any 4-tBP was detected [22]. This clearly indi-
cates that both IB and tert-butyl alcohol in the adsorbed state
might not produce any p-alkylated product over the present
catalyst system. The above results and discussion hint that
both Hinshelwood and Eley–Rideal mechanisms are possi-
ble for tertiary butylation on Cu1−xCoxFe2O4.
From WHSV studies it is clear that the selectivities of
2-tBP and 2,4-dtBP show reverse trends with a changing
contact time. Due to long contact time sequential alkylation
is increasingly feasible at low WHSV and entirely supports
the high selectivity of 2,4-dtBP at low WHSV. The foregoing
facts indicate the reaction is kinetically controlled and prod-
uct distribution can be optimized. Further, they also demon-
strate an increase in the selectivity of 2,4-dtBP at the expense
of 2-tBP. An increase in 2-tBP selectivity with increasing
WHSV indicates that 2-tBP is preferred over other products
at short contact time, probably due to phenol orientation on
the catalyst surface, which is easily ortho-butylated. Possi-
ble dealkylation reactions at higher temperatures may also
account for a decrease in the selectivity of 2,4-dtBP. Lower
t-butylation activity at higher temperatures is due to the side
reactions of IB with to butylenes and aliphatics.
4.2. Reaction mechanism
A detailed investigation of the acid–base properties of
the Cu1−xCoxFe2O4 catalyst system was carried out sepa-
rately [23]. Important points from FT-IR studies help in gain-
ing an understanding of tertiary butylation activity: (1) The
Cu1−xCoxFe2O4 surface is dominated by Lewis acid char-
acter. (2) A relatively weak acidic character at x = 0 in-
creases with Co content to highly acidic character at x = 1.0
with an increase in number and strength of acidic sites (Ta-
ble 1). (3) The basic character of the catalyst decreases with
increasing x. (4) The acid–base character on the catalyst
surface is due predominantly to octahedral cations, and the
Lewis acidity changes significantly depending on the eas-
ily reducible (Cu2+) or nonreducible (Co2+) ions in the
neighborhood. Generally, robust Co2+ in the neighborhood
increases the acidity. A simple comparison of phenol con-
version and acid–base character demonstrates that neither
strong acidic nor strong basic character helps the tertiary
butylation reaction. It is also evident that the intermediate
acid–base character observed for the x = 0.5 composition
(Tables 1 and 2) with high catalytic activity indicates an
optimum level of acid–base sites are required for better cat-
alytic performance in terms of phenol conversion and over-
all product yield. Strong acidity promotes oligomerization
and cracking of IB; thus the conversion of phenol is ex-
pected to be lower, which is the case with Co-rich composi-
tions (Fig. 2b). In addition, butyl phenols may also undergo
dealkylation in the presence of strong acid sites, account-
ing for the low conversion of phenol at x ꢁ 0.75. However,
the highly acidic property at x = 1 improves 4-tBP selectiv-
ity. It is in agreement that strong acid sites favor 4-tBP and
weak to moderate acid sites favor 2-tBP as in zeolites [11]
4.3. Electronic structure aspects
That (phenol + IB) creates a reducing atmosphere is con-
firmed by the observation of H2, CO, and cracked products

T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
115
of IB in gaseous products. The XRD pattern of spent cata-
lysts obtained at 200 ◦C shows Cu0 on x = 0; however, only
a trace amount is detected on other catalyst compositions.
An optimum temperature of 200 ◦C is not capable of accel-
erating the reduction process is very much reflected in the
XRD and XPS results. Fe carbides, Fe0, and Cu0 detected
in the methylation and ethylation spent catalysts [21,22]
are not detected in the XRD pattern of tertiary butylation
spent catalysts. Nevertheless, the unreacted phases of CuO
and α-Fe2O3 present on the fresh catalyst at x = 0, which
remain after the reaction to some extent, could be the rea-
son for sizable amounts of Cu0 and Fe2O3. This contrasts
with the observation of magnetite on spent catalysts (x = 0)
after methylation and ethylation [21,22] between 300 and
400 ◦C.
The changes observed on main peaks, satellites, and
Auger features of Fe, Co, and Cu are a clear-cut indication
that partial reduction of these species occurs during the re-
action. It is useful to highlight the following observations
from XPS: (1) Copper reduction is seen for x ꢀ 0.75 and the
extent of reduction increases linearly with X as well as reac-
tion temperature. (2) There is partial reduction of all metal
ions and a consequent increase in the overlap of 3d energy
bands in the spent catalysts, especially at intermediate com-
positions. (3) The (Cu + Co)/Fe ratio at x = 0.5 remains
around 1.1 before and after reaction. Thus, a high phenol
conversion and a Cu + Co/Fe ratio close to 1 at x = 0.5
ensure the optimum distribution of active species to have
a highly heterogeneous surface. The high activity observed
at x = 0.5 highlights the importance of a 1:1 bulk ratio of
Cu:Co. The above also suggests that the surface atomic ratio
remains almost the same throughout the reaction period on
x = 0.5, but the low activity at a TOS of 10 h is due mainly
to coke deposition. Apparently the coke deposition merely
decreases the percentage of metal ion on the surface, but not
the metal ion ratio significantly and hence the above obser-
vation. This also suggests that coke formation does not affect
other results significantly for x = 0.5. Simple calcination in
air restores the original activity. A change from low activity
observed at x = 1 to better activity at x ꢀ 0.75 indicates the
importance of Cu2+ for better performance.
XPS studies on spent catalysts of Cu1−xCoxFe2O4 re-
vealed that Cu and Co 3d energy bands overlap to a consid-
erable extent and play an important role in catalytic perfor-
mance. The relatively strong overlap of 3d energy bands of
metal ions in methylation and ethylation reactions compared
with tertiary butylation is due mainly to the high tempera-
ture (350–375◦C) employed in the former cases compared
with 200 ◦C in the latter.
5. Conclusions
This article reports the results from the catalytic study of
tertiary butylation of phenol with isobutene on
Cu1−xCoxFe2O4 as a function of various reaction parame-
ters and catalyst compositions. Fresh and spent catalysts
were subjected to detailed analysis by XRD, XPS, and
XAES. Phenol tertiary butylation gives three major products,
namely, 2-tBP, 4-tBP, and 2,4-dtBP, and the phenol conver-
sion and selectivity of these products depend on the reaction
conditions. It is found that x ꢀ 0.5 compositions are compar-
atively more active, indicating the importance of Cu for this
reaction. x = 0.5 shows better activity, emphasizing the im-
portance of the 1:1 bulk combination of Cu + Co for overall
performance.
It is found that the acid–base properties of the system
have considerable influence on product selectivity. A large
number of strong acid sites on x = 1 does help in the forma-
tion of 4-tBP. On the other hand, the selectivities of 2-tBP
and 2,4-dtBP are lowest on catalysts with strong acidity;
however, catalysts having intermediate acid–base property
have maximum products yield. The high activity achieved
with x = 0.5 demonstrates that an optimum concentration of
acid–base centers is needed for phenol adsorption and sub-
sequent polarization of both phenol and the alkylating agent.
In phenol alkylation with isobutene, the tertiary butyl carbo-
cation can attack, from the adsorbed as well as the gaseous
state, the phenol, resulting in the formation of para tertiary
butylated products such as 4-tBP and 2,4-dtBP. Steric hin-
drance of the tert-butyl group prevents the further attack of
IB at the ortho position of already formed 2-tBP, illustrating
negligible formation of 2,6-dtBP.
XPS analysis of both fresh and spent Cu1−xCoxFe2O4
catalysts indicates that the oxidation states of the metal ions
and their distribution on the surface have considerable influ-
ence on activity and selectivity toward phenol tertiary buty-
lation. XRD and XPS studies reveal that redistribution as
well as partial reduction of metal ions occurs to various lev-
els due to reaction. The high activity at x = 0.5 demonstrates
that uniform distribution of reactive Cu species with hetero-
geneous neighbors of Co and Fe is essential. Valence band
studies show a change from marginal overlap between 3d
bands of metal ions on fresh catalysts to a significant overlap
on spent catalysts. It is clear that the x = 0.5 catalyst brings
out synergetic behavior due to 3d energy band overlap and
an intermediate acid–base character and enhances catalytic
The extent of Cu reducibility increases with increasing
x on spent catalysts. This is in excellent agreement with
the results obtained from temperature programmed reduc-
tion (TPR) of Cu1−xCoxFe2O4 in H2 atmosphere [20]. Two
important reduction processes involved are Cu2+ → Cu+
and Cu+ → Cu0, at increasing temperature between 180 and
280 ◦C. Cu2+ → Cu+ occurs around 200 ◦C and the amount
of H2 consumed increases with x. Additionally, a gradual de-
crease in the reduction temperature is observed for Cu+
→
Cu0 as x increases. Since the tertiary butylation of phenol is
carried out at 200 ◦C, which is at the Cu2+ → Cu+ reduction
temperature, the decrease in Cu+ → Cu0 reduction tem-
perature also favors the above. The preceding TPR results
entirely support increasing Cu reducibility with increasing x
at 200 ◦C as observed in XPS.

116
T. Mathew et al. / Journal of Catalysis 222 (2004) 107–116
activity. It is likely that a combination of all the factors—
optimum acid–base character, relatively high surface area,
uniform distribution of active species with heterogeneous
neighbors, and 3d bands overlap—contributes to overall re-
activity and further quantitative experiments are necessary
to single out the factor that contributes the most to activity.
[12] A.H. Padmasri, V.D. Kumari, P.K. Rao, Stud. Surf. Sci. Catal. 113
(1998) 563.
[13] J.W. Yoo, C.W. Lee, B. Wang, S.E. Park, Res. Chem. Intermed. 27
(2001) 561.
[14] M. Nagai, T. Yoda, M. Kodomari, J. Catal. 201 (2001) 105.
[15] R. Rajgopal, R. Vetrivel, B.S. Rao, Catal. Lett. 65 (2000) 99.
[16] K. Sreekumar, T. Mathew, S.P. Mirajkar, S. Sugunan, B.S. Rao, Appl.
Catal. A 201 (2000) L1.
[17] K. Sreekumar, T. Mathew, B.M. Devassy, R. Rajagopal, R. Vetrivel,
B.S. Rao, Appl. Catal. A 205 (2001) 11.
[18] K. Sreekumar, T.M. Jyothi, T. Mathew, M.B. Talawar, S. Sugunan,
B.S. Rao, J. Mol. Catal. A 159 (2000) 327.
Acknowledgment
T.M. thanks CSIR, New Delhi, for a senior research fel-
lowship.
[19] B.S. Rao, K. Sreekumar, T.M. Jyothi, Indian Patent 2707/98 (1998).
[20] K. Lázár, T. Mathew, Z. Koppány, J. Megyeri, V. Samuel, S.P. Mi-
rajkar, B.S. Rao, L. Guczi, Phys. Chem. Chem. Phys. 4 (2002) 3530.
[21] T. Mathew, N.R. Shiju, K. Sreekumar, B.S. Rao, C.S. Gopinath, J. Ca-
tal. 210 (2002) 405.
[22] T. Mathew, Synthesis and Characterization of Mixed Oxides Contain-
ing Cobalt, Copper and Iron and Study of Their Catalytic Activity,
PhD thesis, University of Pune, 2002.
[23] T. Mathew, N.R. Shiju, B.B. Tope, S.G. Hegde, B.S. Rao, C.S.
Gopinath, Phys. Chem. Chem. Phys. 4 (2002) 4260.
[24] G. Moretti, G. Fierro, M.L. Jacono, P. Porta, Surf. Interface Anal. 14
(1989) 325.
[25] S. Poulston, P.M. Parlett, P. Stone, M. Bowker, Surf. Interface Anal. 24
(1996) 811.
References
[1] H.-W. Voges, in: B. Elvers, S. Hawkins, G. Schulz (Eds.), Ullmann’s
Encyclopedia of Industrial Chemistry, vol. A19, VHC, Weinheim, Ger-
many, 1991, p. 328.
[2] M. Inoue, S. Emoto, Chem. Pharm. Bull. 20 (1972) 232.
[3] S. Velu, C.S. Swamy, Res. Chem. Intermed. 26 (2000) 295.
[4] A. Sakthivel, N. Saritha, P. Selvam, Catal. Lett. 72 (2001) 225, refer-
ences therein.
[26] B.R. Strohmeier, B.E. Leyden, R.S. Field, D.M. Hercules, J. Catal. 94
(1985) 514.
[27] M.S. Stranick, M. Houalla, D.M. Hercules, J. Catal. 106 (1987) 362.
[5] K.G. Chandra, M.M. Sharma, Catal. Lett. 19 (1993) 309.
[6] R.A. Rajadhyakasha, D.D. Chaudhari, Ind. Eng. Chem. Res. 26 (1987)
1276.
[28] N.S. McIntyre, M.G. Cook, Anal. Chem. 47 (1975) 2210.
[29] Z. Zsoldos, L. Guczi, J. Phys. Chem. 96 (1992) 9393.
[30] P. Mills, J.L. Sullivan, J. Phys. D 16 (1983) 723.
[31] T. Fujii, F.M.F. de Groot, G.A. Sawatzky, F.C. Voogt, T. Hibma, K.
Okada, Phys. Rev. B 59 (1999) 3195.
[32] J.J. Yeh, I. Lindau, Atom. Data Nucl. Data Tables 32 (1985) 1.
[33] T. Mathew, S. Pai, B.B. Tope, S.G. Hegde, B.S. Rao, C.S. Gopinath,
J. Phys. Chem. B, submitted.
[7] S.K. Badamali, S. Sakthivel, P. Selvam, Catal. Lett. 65 (2000) 153.
[8] K. Zhang, H. Zhang, G. Xu, S. Xiang, D. Xu, S. Liu, H. Li, Appl.
Catal. A 207 (2001) 183.
[9] S.K. Badamali, S. Sakthivel, P. Selvam, Catal. Today 63 (2000) 291.
[10] S. Subramanian, A. Mitra, C.V.V. Satyanarayana, D.K. Chakrabarty,
Appl. Catal. A 159 (1997) 229.
[11] K. Zhang, C. Huang, S. Xiang, S. Liu, D. Xu, H. Li, Appl. Catal. A 166
(1998) 89.