Selective production of methoxyphenols from dihydroxybenzenes on alkali metal ion-loaded MgO

Article abstract of DOI:10.1016/j.jcat.2006.08.009

Selective O-methylation of dihydroxybenzenes (DHBs; catechol, resorcinol, and hydroquinone) to methoxyphenols (MPs) was carried out with dimethylcarbonate on MgO and alkali metal ion (Li, K, and Cs)-loaded MgO between 523 and 603 K. Catalytic activity and product selectivity varied with respect to DHB substrates. Selectivity for O-methylated products increased with increasing basicity of alkali ions; however, K-MgO showed high and stable activity toward MPs. Selectivity for MPs obtained from three substrates increased in the following order: catechol < resorcinol < hydroquinone. The mode of interaction of substrates on the catalysts surface influenced reactivity and product selectivity. It is likely that the low reaction temperatures used (< 603 K) kinetically control and favor high MP selectivity from DHBs. Calcined and spent catalysts were characterized by XRD, surface area, SEM, thermal analysis, NMR, and XPS. XRD analysis revealed the formation of alkali oxide phases on alkali-loaded MgO. Crystallite size and surface area of the catalysts decreased after methylation reactions, except on K-MgO. TGA showed 40-60 wt% coke deposition on spent catalysts. TGA in N₂ followed by air and ¹³C CP-MAS NMR measurements indicated the nature of deposited carbon to be molecular species, graphite, MgCO₃ and polyaromatics. XPS revealed the nature and availability of active sites on the spent catalysts, as well as the same changes with reaction conditions and correlated with catalytic activity.

Full text of DOI:10.1016/j.jcat.2006.08.009

Journal of Catalysis 243 (2006) 376–388
www.elsevier.com/locate/jcat
Selective production of methoxyphenols from dihydroxybenzenes
on alkali metal ion-loaded MgO
∗
Munusamy Vijayaraj, Chinnakonda S. Gopinath
Catalysis Division, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411 008, India
Received 2 August 2006; accepted 15 August 2006
Abstract
Selective O-methylation of dihydroxybenzenes (DHBs; catechol, resorcinol, and hydroquinone) to methoxyphenols (MPs) was carried out with
dimethylcarbonate on MgO and alkali metal ion (Li, K, and Cs)-loaded MgO between 523 and 603 K. Catalytic activity and product selectivity
varied with respect to DHB substrates. Selectivity for O-methylated products increased with increasing basicity of alkali ions; however, K-MgO
showed high and stable activity toward MPs. Selectivity for MPs obtained from three substrates increased in the following order: catechol <
resorcinol < hydroquinone. The mode of interaction of substrates on the catalysts surface influenced reactivity and product selectivity. It is likely
that the low reaction temperatures used (<603 K) kinetically control and favor high MP selectivity from DHBs. Calcined and spent catalysts
were characterized by XRD, surface area, SEM, thermal analysis, NMR, and XPS. XRD analysis revealed the formation of alkali oxide phases on
alkali-loaded MgO. Crystallite size and surface area of the catalysts decreased after methylation reactions, except on K-MgO. TGA showed 40–
13
6
0 wt% coke deposition on spent catalysts. TGA in N followed by air and C CP-MAS NMR measurements indicated the nature of deposited
2
carbon to be molecular species, graphite, MgCO and polyaromatics. XPS revealed the nature and availability of active sites on the spent catalysts,
3
as well as the same changes with reaction conditions and correlated with catalytic activity.
©
2006 Elsevier Inc. All rights reserved.
Keywords: MgO; K-MgO; Alkali-loaded MgO; Dihydroxybenzene; Methoxyphenol; XPS; O-methylation; Thermal analysis
1
. Introduction
duction of methoxyphenols (MPs) from DHBs is an interesting
and challenging problem from both process and kinetic control
standpoints. However, few reports are available in the literature
Innovations in solid catalysts for the production of fine
[
31–46].
Ono et al. [31–34] reported the selective O-methylation of
chemicals and bulk chemicals have lead researchers to at-
tempt vapor-phase methylation (or alkylation) of phenols with
dimethylcarbonate (DMC), methanol (or alcohols) over metal
oxides [1–23], sulfates [24,25], phosphates [26], and zeolites
catechol over alkali hydroxide/nitrate-loaded alumina. DMC
was found to be a more efficient methylating agent than
methanol over alumina. The main product obtained was 2-MP,
[
27–29]. O-methylated dihydroxybenzenes (DHBs) and phe-
with 70% selectivity at 583 K. Over K/Al O , a high 1,2-
2
3
nols are widely used in the production of variety of fine chem-
icals and valuable synthetic intermediates [30]. For instance,
2
chol, is an important precursor in the preparation of vanillin.
Similarly, 3-methoxyphenol (3-MP) and 4-methoxyphenol
dimethoxybenzene (1,2-DMB) yield of 97% was obtained at
83 K with DMC:catechol = 4. Li/Al2O3 was found to be a
selective catalyst for the 2-MP formation (84%) with 100%
catechol conversion at 583 K. On Cs/Al O , catechol methyla-
5
-methoxyphenol (2-MP), derived from O-methylation of cate-
2
3
tion yielded 2-MP and catechol carbonate (PCC), which further
reacted with DMC, resulting in 1,2-DMB formation.
(
4-MP) are used as UV inhibitors, antioxidants for oil and
grease, and polymerization inhibitors. Selective catalytic pro-
Cavani et al. [35–37] found boron phosphate (BPO ) to be an
4
efficient catalyst for the monoetherification of catechol, aimed
*
at the production of 2-MP. BPO catalyst with B/P = 1 dis-
Corresponding author. Fax: +91 20 2590 2633.
E-mail address: cs.gopinath@ncl.res.in (C.S. Gopinath).
4
played best results in terms of catechol conversion, 2-MP se-
0021-9517/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2006.08.009

M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
377
lectivity, and stability with TOS at 548 K. The optimal surface
acidity and low temperature made the undesired reactions of tar
formation and ring methylation much slower. However, this cat-
alyst on alumina support had a lower activity than BPO4. Vish-
wanathan et al. [38] reported the vapor-phase methylation of
catechol over solid base catalysts, like Cs-loaded TiO2, Al2O3,
and SiO2 at 623 K. Good selectivity to 2-MP was achieved,
and TPD studies indicated the presence of weak basic sites and
their participation in the O-methylation process. Rao et al. [39,
the MgO by an impregnation method using an aqueous solution
of LiOH·H O, KNO , and CsNO at 353 K. The as-synthesized
2
3
3
catalysts were dried at 383 K for 12 h, pelletized, and sieved
to a mesh size of 25–30. An identical but blank wet impreg-
nation was carried out with MgO, and the resulting Mg(OH)2
was used after activation as pure MgO catalyst. The catalytic
measurements were carried out in vertical downflow glass reac-
tor (18 mm i.d.) with 0.5 g of sieved catalyst packed between
inert porcelain beads and placed inside a vertical furnace (Geo-
mechanique, France) [49,50]. The catalyst was activated in situ
with a stream of air at 673 K for 2 h for nitrate and hydroxide
decomposition, then flushed with stream of N2 (10 mL/min)
while bringing down to the desired reaction temperature. Note
that the above in situ activation decreases the weight of the
loaded catalyst due to the decomposition of nitrate/hydroxide,
and the exact weight was obtained by simulating the same in
thermogravimetric analysis (TGA) and used for space velocity
calculations. The feed mixture (DMC + DHB) was passed into
the reactor using a syringe pump in a N2 stream (10 mL/min).
Products were identified by gas chromatography–mass spec-
trometry (GC–MS), using a Schimadzu GC-17A equipped with
a QP 5000 mass spectrometer, and analyzed by GC, using an
Agilent 6890 J-413 with an HP-5.5% phenyl methyl silox-
ane capillary column equipped with a flame ionization detec-
tor.
4
0] found that using Mg–Al hydrotalcite as a catalyst for the
synthesis of 2-MP gave an 80% yield at 573 K. A concerted
mechanism was proposed involving acid–base pair sites with
nucleophilic attack at the methyl carbon atom of DMC by the
phenolic oxygen or phenolate anion.
Renken et al. [41,42] studied regulation to control the se-
lectivity in vapor-phase methylation of catechol to 2-MP over
modified alumina at 598 K. A 20-fold change in the O/C-
methylation ratio was achieved by varying the catalyst acid–
base properties. Catalytic activity and selectivity toward 2-
MP formation was found to increase with surface acidity.
Sivasanker et al. [43,44] reported O-methylation of DHB on
alkali metal ion-loaded silica at relatively high temperature
(
673 K). Catalyst activity and selectivity for DMB increased
with metal loading and basicity of the metal ions. The order of
reactivity of the three DHBs increased in the following order:
catechol < hydroquinone < resorcinol. Assuming the methy-
lation process as pseudo-first-order kinetics, the ratio of rate
constants of second methylation to first methylation (K2/K1)
was found to predict the ratio of DMB to total methoxy prod-
ucts [44]. Recently, Zhu et al. [45] reported 2-MP production
from Ti containing AlPO4 at 553 K. In another report, Fu et al.
2.2. Catalyst characterization
XRD patterns of the powder samples were recorded with
a Rigaku Geigerflex instrument using CuKα radiation
[
46] explored ZnCl2-modified Al2O3 for O-methylation of cat-
(
1.5405 Å) with a Ni filter to identify the different phases. The
echol with MeOH at 553 K. A weak coordination of catechol
crystallite size of the catalysts was calculated using Scherrer’s
equation [51]. The BET surface area and the pore volume (Vp)
of the catalysts were determined by N2 adsorption–desorption
method at 77 K using a Quanta chrome NOVA-1200 adsorption
unit. Activated catalysts were subjected to energy-dispersive
X-ray (EDX) analysis for bulk composition; the results were
within the expected range. Thermal analysis was measured by
a Perkin–Elmer Diamond TG/DTA with Al2O3 as the inter-
nal standard. About 10 mg of the spent K-MgO catalyst was
taken after methylation testing with any DHB for 1 h and was
subjected to temperature ramping up to 1500 K in the flow
of air or N2 (100 mL/min) at a heating rate of 10 K/min.
with Zn2 sites led to 82% 2-MP selectivity. ZnCl2 leaching
and surface coking during the reaction run were the main causes
of catalyst deactivation. It is clear from the literature reports that
the alkali metal ions enhance the basicity of the catalyst [47,48],
which in turn improves the O-methylation activity.
+
Although O-methylation for catechol has been evaluated,
that of resorcinol and hydroquinone has not been measured in
detail [43,44]. We have attempted O-methylation of all three
DHBs with DMC on MgO and alkali ion (Li, K, and Cs)-
loaded MgO to selectively produce MPs. The high and sta-
ble activity observed on K-MgO indicates an optimum basic-
ity required for O-methylation. X-ray diffraction (XRD), sur-
face area, NMR, photoemission and thermal analysis on cal-
cined and spent catalysts reveal the mechanisms behind cata-
lyst deactivation. A plausible methylation mechanism is sug-
gested from the reaction and spent catalyst characterization re-
sults.
13
C cross-polarization magic-angle spinning (CP-MAS) NMR
measurements were carried out with an MSL Bruker spec-
trometer operating at 75.15 MHz on solid spent catalysts from
catechol and resorcinol methylation after 1 h TOS. X-ray pho-
toelectron spectra (XPS) were acquired from a VG Microtech
Multilab ESCALAB 3000 spectrometer using a nonmono-
chromatized MgKα X-ray source (1253.6 eV) [49,50]. Base
pressure in the analysis chamber was maintained in the range
2
. Experimental
−
10
2
.1. Catalyst preparation and methylation
of 3–6 × 10
Torr. Energy resolution of the spectrometer
was set at 0.8 eV with MgKα radiation at a pass energy of
20 eV. The error in the reported binding energy (BE) values is
± 0.1 eV.
Magnesium oxide light (Merck, India) was used as a cata-
lyst/support. Alkali metal ions (Li, K, and Cs) were loaded onto

378
M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
Fig. 1. Selective O-methylation of catechol to 2-methylphenol at 583 K on MgO and 5 mmol alkali ion-loaded MgO, with DMC:catechol = 2:1 feed mixture. Alkali
ion content loaded on MgO in mmol is given in parentheses. Catechol methylation on 3 mmol K-loaded MgO at 583 K also shown to show the dependence on alkali
loading.
studies were carried out exclusively on 5 mmol-loaded systems.
In fact, the stable activity on K-MgO clearly indicates that the
genesis of activity was due mostly to potassium and the asso-
ciated overall intermediate basicity on K-MgO. The decreased
catechol conversion on MgO and Li/Cs-MgO suggests that a
low or high basicity did not assist the reaction. Indeed, a simi-
lar trend was observed with other substrates, like resorcinol and
hydroquinone, and the same explanation holds.
On MgO (Fig. 1a), 2-MP selectivity was 80% and C-
methylated products formed dominantly over 1,2-DMB and
catechol carbonate (PCC). However, with Li (K) loading on
MgO, 2-MP selectivity decreased marginally to 75% (70%) and
Scheme 1. Catechol methylation with DMC (DMC:catechol = 2) on MgO and
C-methylated product selectivity dropped to 5% on K-MgO
alkali ion-loaded MgO. Highly selective (97%) production of 1,2-DMB with
(Figs. 1b and 1c). 1,2-DMB (PCC) formation decreased (in-
DMC:catechol = 4 hints a direct methylation.
creased) with TOS on MgO as well as on Li-MgO. In contrast,
the sizeable amount of 1,2-DMB formed (24%) on K-MgO
indicates a higher O-methylation capacity compared with the
other systems. Suppression of PCC formation at the cost of 1,2-
DMB on K-MgO was noted and compared with that on Li- or
Cs-MgO catalysts.
3
. Results
3
.1. Catechol methylation
Fig. 1 shows catechol conversion and 2-MP selectivity
Feed composition studies on K-MgO at 573 K with DMC:
catechol = 2 and 4 are shown in Fig. 2. Needless to say that
on pure MgO and Li- and K-loaded MgO with DMC at
83 K. The amount of alkali metal nitrate/hydroxide loaded
was 5 mmol per gram of MgO. Reaction results from 3 mmol
5
1
00% catechol conversion was obtained in both cases; how-
ever, product selectivity changed dramatically with the amount
of DMC. An abrupt change occurred in 1,2-DMB selectivity
KNO -loaded MgO are also shown in Fig. 1. Li- and Cs-
3
loaded MgO showed very similar results, and hence the lat-
ter results are not shown. 2-MP formed as a major product
along with PCC, 1,2-DMB, and C-methylated products (3-
methylcatechol, 3-methylguaiacol, and 4-methylguaiacol), as
(
from 24 to 97%) for a change in the DMC:catechol ratio from
2 to 4. Only trace amounts of 2-MP and C-methylated products
and no PCC were formed at a feed ratio of 4. Fig. 2 also shows
DMC conversion with respect to TOS, indicating that 60–65%
DMC reacted for the complete conversion of catechol to methy-
lated products, and hence effective utilization occurred.
The effect of reaction temperature (523–583 K) was eval-
uated on K-MgO with DMC:catechol = 2 (Fig. 3). Catechol
conversion increased from 25% at 523 K to nearly 100% at
+
indicated in Scheme 1. The 5 mmol K -loaded MgO (K-MgO)
showed stable and 100% catechol conversion; however, on
all other systems, the conversion decreased continuously. The
+
3
mmol (and 1 mmol) K -loaded MgO showed lower con-
version for all three DHBs, with the extent of O-methylation
+
decreasing with a decreasing amount of K ions. All further

M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
379
Fig. 2. Feed composition dependence of O-methylation catalytic activity and
−
1
products selectivity on K(5)-MgO at 583 K at WHSV = 9.7 h
.
5
53 K and then reached 100% at 583 K. 2-MP selectivity was
marginally affected at 523 K and remained at around 70% be-
tween 553 and 583 K. Note that PCC formation increased with
TOS (18–36%) at 523 K. 1,2-DMB (PCC) selectivity increased
(
decreased) with increasing temperature from 523 to 583 K.
This indicates that 1,2-DMB formed at the expense of PCC at
high reaction temperature and that this formation was not likely
through subsequent methylation of 2-MP.
Fig. 4. Space velocity dependence of (a) catechol conversion and (b) the ratio
between 2-methylphenol to 1,2-dimethoxybenzene (mono to di methylation)
at 583 K on K(5)-MgO with DMC:catechol = 2:1 feed mixture. It is to be
noted that at high WHSV, 2-MP was produced exclusively and hence 2-MP
to 1,2-DMB ratio is not given in panel (b).
Weight hour space velocity (WHSV)-dependent studies
were carried out on K-MgO with a DMC:catechol = 2 feed at
5
83 K (Fig. 4). The results clearly indicate that catechol conver-
−
1
Fig. 3. Reaction temperature dependence of catechol conversion and products selectivity on K(5)-MgO at a WHSV = 9.7 h
with DMC:catechol = 2:1 feed
mixture.

380
M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
sion was significantly affected by changes in WHSV. Although
catechol conversion decreased to 1/3 by increasing the space
on Li- and K-MgO (MgO and Cs-MgO) catalysts. Some high
boiling biphenolic molecular fragments were also identified
from reaction products by GC-MS. 3-MP selectivity remained
high (>70%) on all of the catalyst systems independent of TOS.
Cs-MgO showed faster deactivation than MgO even though
the highest 3-MP selectivity (>90%) was observed. Significant
aromatic ring methylation was observed on all systems except
Cs-MgO.
Fig. 5 shows a systematic increase in resorcinol conversion
with increasing temperature from 543 K on K-MgO; however,
deactivation was also observed with increasing TOS at all tem-
peratures studied. At lower temperatures (543 and 563 K),
3-MP formed selectively (90–97%) with traces of C-methylated
resorcinol and 1,3-DMB. Secondary products formed appre-
ciably at 583 K and above at the expense of 3-MP. 1,3-DMB
formed in the initial few hours of reaction at 583 K and above
at the expense of 3-MP, indicating the stepwise methylation of
−
1
velocity from 9.7 to 29 h , the initial conversion level for any
WHSV value remained at the same level for up to 12 h, indicat-
ing stable catalytic activity (Fig. 4a). Fig. 4b expresses 2-MP
selectivity in terms of the 2-MP to 1,2-DMB (mono to di) ratio,
showing a ratio of >1, indicating the selective monomethyla-
tion over dimethylation. The above methylation ratio increased
exponentially at higher space velocity, and 2-MP selectivity
increased to 100% with no C-methylated products at a space
−1
velocity >14.5 h
.
3
.2. Resorcinol methylation
Table 1 shows resorcinol methylation activity and product
selectivity on MgO and alkali metal ion-loaded MgO. 3-MP
formed as a major product with 1,3-DMB (Scheme 2) and
other C-methylated products (4-methylresorcinol, 6-methyl-3-
guaiacol, and 4-methyl-3-guaiacol) as side products on all of
the catalysts. A high initial resorcinol conversion (>65%) at
TOS = 1 h was observed in the TOS-dependent studies, indicat-
ing the possibly high potential of all catalysts. K-MgO showed
3-MP to 1,3-DMB.
Fig. 6 shows the effect of the DMC:resorcinol molar ra-
tio on reactant conversion and product selectivity on K-MgO.
It is very interesting to note that the catalyst deactivation rate
decreased with increasing DMC:resorcinol ratio from 2 to 4.
A stable resorcinol conversion was observed for at least up to
8
0–90% conversion up to TOS = 3 h. Nonetheless, the conver-
sion decreased from the initial levels to about ꢀ40% (ꢀ15%)
Table 1
Comparison of O-methylation activity of resorcinol with DMC:resorcinol =
2:1 feed mixture at 583 K and TOS = 3 h
Catalysts
WHSV in h
MgO (8.6)
Li-MgO (8.8)
K-MgO (9.7)
Cs-MgO (8.6)
Conversion
(wt%)
Products selectivity (%)
−
1
(
)
3-MP
1,3-DMB
Others
28
79.7
69
83
3.5
4.5
11
16.8
26.5
6
43.6
81.6
21.2
Scheme 2. Resorcinol methylation with DMC on MgO and alkali ion-loaded
MgO. Ratio of 3-MP/1,3-DMB varies with the ratio of DMC:resorcinol as well
as time on stream.
94
3
3
−
1
Fig. 5. Reaction temperature dependence of resorcinol conversion and products selectivity on K(5)-MgO at WHSV = 9.7 h with DMC:resorcinol = 2:1 feed
mixture.

M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
381
−
1
Fig. 6. Feed composition (DMC:resorcinol) dependence of O-methylation catalytic activity and products selectivity on K(5)-MgO at 583 K and at WHSV = 9.7 h
.
1
1 h for a DMC:resorcinol ratio of 4 (Fig. 6c). 3-MP (1,3-DMB)
selectivity increased (decreased) after the transient state at a
TOS of ca. 1 h. The above results clearly demonstrate that ex-
cess DMC hindered the catalyst deactivation. Recyclability of
K-MgO was tested after the reactions shown in Figs. 6a and 6b.
There was a considerable decline of 5% conversion on every re-
cycling after complete carbon burning; however, the selectivity
remained the same.
Fig. 7 shows the effect of space velocity on resorcinol con-
version and 3-MP/1,3-DMB (mono to di) selectivity ratio at
various TOS on K-MgO with DMC:resorcinol = 4 at 583 K.
A stable resorcinol conversion was observed for all space veloc-
ities used, although conversion decreased with increasing space
velocity (Fig. 7a). O-methylation activity plotted in terms of
−
1
mono to di ratio was >1 at WHSV > 14.5 h . This indicates
a high 3-MP selectivity is possible by reducing the substrate–
catalyst contact time. Also note that with increasing TOS, the
mono/di ratio increased substantially at all WHSVs.
3
.3. Hydroquinone methylation
Hydroquinone methylation on MgO- and alkali ion-loaded
MgO with a DMC:hydroquinone = 2 feed ratio at 583 K was
carried out; the results are given in Table 2. Methanol was added
in the feed as solvent for all hydroquinone methylation studies,
because it is sparingly soluble in DMC. As such, methanol had
no methylation activity on these catalysts. This was confirmed
by experiments with methanol and hydroquinone that showed
no methylation. 4-MP formed selectively along with small
amounts of 1,4-DMB and 2-methylhydroquinone (Scheme 3).
MgO, Li, and Cs-MgO showed ca. 30% initial conversion (from
TOS-dependent studies; not shown) with >85% 4-MP selec-
tivity. Nevertheless, K-MgO showed stable hydroquinone con-
version (80%) and selectively produced 4-MP (95%). Gener-
Fig. 7. Space velocity dependence of (a) resorcinol conversion and (b) the ratio
between 3-methoxyphenol to 1,3-dimethoxybenzene (mono to di methylation)
at 583 K on K(5)-MgO with DMC:resorcinol = 4:1 feed mixture.

382
M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
Table 2
Comparison of O-methylation activity of hydroquinone with DMC:hydroqui-
none = 2:1 feed mixture at 583 K and TOS = 3 h
Catalysts
WHSV in h
MgO (8.6)
Li-MgO (8.8)
K-MgO (9.7)
Cs-MgO (8.6)
Conversion
(wt%)
Product selectivity (%)
−
1
a
(
)
4-MP
1,4-DMB
Others
24
28
83
10
85
90
94
98
1
1
4
2
14
9
2
0
a
Mainly consisting 2-methylhydroquinone.
Scheme 3. Hydroquinone methylation with DMC on MgO and alkali ion-loaded
MgO catalysts.
Fig. 9. Space velocity dependence of hydroquinone conversion and 4-methoxy
phenol selectivity at 583 K on K(5)-MgO with DMC:hydroquinone = 2:1 feed
mixture.
Space velocity influenced hydroquinone conversion and 4-
MP selectivity, as shown in Fig. 9. Reaction was measured on
K-MgO with DMC:hydroquinone = 2. Hydroquinone conver-
sion decreased rapidly from 95 to 3% with increased space
−
1
velocity from 3.2 to 34 h . 4-MP was produced with 90–100%
selectivity at all space velocities used.
Fig. 8. Feed composition (DMC:hydroquinone) dependence of hydroquinone
and DMC conversion and products selectivity on K(5)-MgO at 583 K and at
−
1
3.4. Textural properties of activated and spent catalysts
WHSV = 9.7 h
.
ally, selectivity for 4-MP increased considerably on alkali ion-
loaded MgO.
The powder XRD patterns from calcined catalysts are shown
in Fig. 10a. It is evident that the single phase of MgO peri-
clase (ASTM 4-0829) remained unaffected by alkali metal ion
loading. However, the new phases observed can be attributed to
alkali metal oxides (Li2O, K2O, and Cs2O) (JCPDS 9-355, 23-
493, 26-1327, and 10-248); indeed, Cs-MgO was dominated by
Cs2O phase. Crystallite sizes calculated using Scherrer’s equa-
tion [51] for all of the catalysts are given in Table 3. Nanocrys-
talline MgO is evident from the crystallite size of about 12 nm
for all of the calcined catalysts except Cs-MgO, which had
a size of 19 nm. Scanning electron microscopy results (not
shown) clearly hinted that the particle size remained the same
on MgO and alkali-loaded MgO. Surface area and pore volume
Fig. 8 shows the feed composition dependence with DMC:
hydroquinone molar ratios of 2 and 4 on K-MgO at 583 K. At a
DMC:hydroquinone ratio of 2, 80% hydroquinone conversion
with 95% selectivity for 4-MP was found (Fig. 8a). At a molar
ratio of 4, 100% hydroquinone conversion was observed; how-
ever, selectivity for 4-MP dropped to 70% with an increase in
1
,4-DMB formation (30%). Higher DMC content in the feed
led to O-methylation of both OH groups. The DMC conversion
was 50% at a DMC:hydroquinone ratio of 2 and 30% at a ratio
of 4, indicating a significant loss of DMC in side reactions like
gasification, dissociation, and others.

M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
383
Table 3
Physical and textural properties of activated and spent MgO and alkali metal
ion-loaded MgO catalysts
Catalysts
Crystallite
size (nm);
spent
BET surface
area (m /g);
spent
Pore volume
activated
(10 cc/g)
2
−
2
(
activated)
(activated)
MgO-Res
6.5 (11.8)
8.4 (12.6)
10.6 (11.8)
10.1 (19.3)
11 (11.8)
21 (60)
24 (49)
18 (28)
6 (16)
24 (28)
21 (28)
13
11
3
Li-MgO-Res
K-MgO-Res
Cs-MgO-Res
K-MgO-Cat
K-MgO-Hq
1
–
–
11.5 (11.8)
3
.5. Thermal analysis
Figs. 11a–11c show TG and derivate TG (DTG) profiles
from spent K-MgO in air atmosphere after catechol, resorci-
nol, and hydroquinone methylation for 1 h at 583 K. The initial
weight loss (<10%) observed at around 400 K on all of the
spent catalysts is attributed to the loss of water. A large amount
of carbon deposition is evident from the weight loss observed
between 420 and 750 K in the TGA profiles. The total percent
weight loss observed at 1000 K was about 40% with catechol,
6
0% with resorcinol, and 50% with hydroquinone spent cata-
lysts. For catechol and resorcinol, the DTG showed two-stage
weight losses at 575–645 K and 670–740 K, respectively; with
hydroquinone, one major weight loss at 640 K and a minor loss
at 710 K were observed. The weight loss at 740 and 710 K
shown in Figs. 11b and 11c, respectively, may be due to the
presence of high boiling biphenolic compounds. The presence
of these compounds was indicated in the GC–MS analysis of
resorcinol methylation products.
Fig. 11d shows the differential thermal analysis (DTA) traces
in which a positive microvolt (µV) heat flow at respective
temperatures was observed due to carbon burning in air at-
mosphere. This demonstrates that the observed TG weight
losses are exothermic in nature, as expected. Further, the peak
at 640–670 K in all three cases is interesting, hinting at the sim-
ilar nature of the carbon deposits, which is likely polyaromatics
Fig. 10. XRD pattern from MgO and alkali metal ion-loaded MgO. (a) Ac-
tivated catalysts (calcined at 673 K for 2 h) and (b) spent catalysts (after
O-methylation of dihydroxybenzene carried out for 4 h TOS at 583 K).
(
Vp) measured from N2 sorption isotherms using BET method
on alkali ion-loaded MgO decreased gradually with increasing
alkali metal ion content and ionic radius, as shown in Table 3.
The very low Vp on Cs-MgO suggests effective pore filling
and hence rapidly declining catalytic activity for all DHB sub-
strates.
[
45,46].
Adsorbed molecular components, such as reactants and
XRD of representative spent catalysts after reaction for 4 h
on stream (Fig. 10b) showed no new reflections, whereas the
intensity of reflections due to alkali metal oxides (Li2O, K2O,
and Cs2O) decreased or disappeared after the reaction. Further,
the broad features observed for spent catalysts compared with
the sharp and intense features on calcined (MgO, Li-MgO, and
Cs-MgO) catalysts indicate a decrease in crystallinity and crys-
tallite size; this was marginal (severe) in the case of K-MgO
products, will become oxidized in air; thus, it is difficult to
account for the weight losses by carrying out thermal analysis
only in air atmosphere. However, thermal analysis in N at-
2
mosphere might reveal (to a limited extent) the nature of the
desorbing components. Figs. 11e, 11f, and 11g (thick traces)
show the TG results carried out in N2 atmosphere for K-MgO
spent catalysts with all three DHB reactants. Note that the
weight loss for catechol, resorcinol, and hydroquinone is ap-
proximately 40% at 1473 K, in contrast to the higher weight
loss due to complete burning at 1000 K in air atmosphere for
resorcinol and hydroquinone. Two broad weight losses were
noted for all of the spent catalysts in the DTG profiles at
500–800 K and 1100–1300 K. DTA traces (Fig. 11h) show an
endothermic weight loss processes, unlike that occurring in air
atmosphere. In the case of catechol, these two weight losses
were not very distinct, whereas resorcinol and hydroquinone
(
Cs-MgO and MgO) (Table 3). Surface area also decreased
for spent catalysts, indicating the presence of significant car-
bonaceous products on the catalyst surface. The percentage
drop in surface area after reaction was lower for K-MgO (14–
3
6%) than for any other catalyst. In fact, catechol and hydro-
quinone methylation on K-MgO hardly deactivated, and the
aforementioned decrease in surface area hardly affected the per-
formance.

384
M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
−
1
Fig. 11. Thermal analysis of K(5)-MgO spent catalyst after methylation of all three dihydroxybenzenes at 583 K, WHSV = 9.7 h for 1 h with DMC:DHB = 2:1
feed mixture. TGA and derivative of TGA (DTG) weight loss under air (a–c) and N (e–g) atmosphere with corresponding DTA in (d) and (h), respectively. TGA
2
and DTG carried out subsequently in air atmosphere, after N atmosphere analysis of catechol (e) and resorcinol (f) and corresponding DTA in (h).
2
showed sharp and distinct weight losses. The spent catalysts
subjected to thermal analysis in N2 atmosphere were subse-
quently subjected to thermal analysis in air atmosphere without
being removed from the thermal analyzer; the results for cat-
echol and resorcinol catalysts are shown in Figs. 11e and 11f
(
thin traces), respectively. Exothermic DTA profiles for these
catalysts showed a single sharp peak (Fig. 11h) at around 830 K,
again indicating a similar nature of carbon, which is likely to
be polyaromatic, because it cannot be oxidized/desorbed in ni-
trogen.
3
.6. 13C CP-MAS NMR
Fig. 12 shows 13C CP-MAS NMR results from spent K-
MgO catalysts after catechol and resorcinol methylation reac-
tion at 583 K for 1 h. Peaks related to adsorbed reactants and/or
products (112 ppm), carbons without protons but attached to
heteroatom (152 ppm), alkyl groups/chains (20 ppm), MgCO3
13
Fig. 12. C CP-MAS NMR spectra from K(5)-MgO spent catalyst after cate-
−
1
chol/resorcinol O-methylation at 583 K, WHSV = 9.7 h for TOS = 1 h with
DMC:dihydroxybenzene = 2:1 feed mixture.
(
170 ppm), and polyaromatics (125 ppm) were identified from
the different broad features. Clearly, the spectral feature was
strongly dominated by aromatic species [52]. The strong peak
at 125 ppm can be attributed to polyaromatic species due to
polymerization of reactant/product molecule in different envi-
ronment. This was the main species blocking the active sites. In
the case of catechol, this strong peak appeared very sharp and
symmetrical; however, in resorcinol it was broad with shoulders
from different carbon species mentioned above.
Alkyl carbon species was also observed (20 ppm) on both
catalysts, indicating the presence of methylated products or
alkyl chains/groups. The alkyl feature was relatively broad and
strong for resorcinol catalyst. It is worth correlating the lower-
intensity alkyl feature and more effective DMC utilization of
catechol case (Fig. 2) compared with resorcinol (Fig. 6). Car-
bonate carbon species, likely from MgCO3 [53], was also ob-

M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
385
Table 4
Surface atomic ratio of K-MgO fresh and spent catalysts
a
Atomic
ratio
Fresh
catalyst
Spent catalyst
Catechol Resorcinol 4:1, DMC:Res Hydroquinone
K/Mg
K/C
Mg/C
0.024
2.565
0.478
0.186
0.665
2.153
0.055
0.025
0.080
0.017
0.078
4.690
4.769
0.147
0.086
0.582
0.667
–
–
–
(
K + Mg)/C
a
K/MgO catalyst was subjected with all reactants at optimum reaction con-
−
1
ditions (space velocity = 6 h ; reaction temperature = 583 K; TOS = 1 h).
2:1 ratio of DMC:DHB was employed for all cases and 4:1 ratio was employed
additionally with resorcinol.
spent catalysts. Very poor or no Mg 2p intensity on resorcinol
2+
(
DMC:resorcinol = 2:1) can be attributed to poor surface Mg
concentration.
Figs. 13c and 13d show K 2p and K-L3M45M45 Auger elec-
tron emission, respectively, from calcined and spent K-MgO
catalysts. The calcined sample exhibited two peaks at 293.3
and 296 eV, attributed to K 2p3/2 and 2p1/2, respectively. The
spent samples showed similar features at lower BEs, at 292.8
and 295.5 eV. The higher K 2p intensity on all of the spent cat-
alysts compared with the calcined catalyst is interesting, hinting
at significant surface modification and altered surface composi-
tion due to reaction. This difference between calcined and spent
catalysts in K 2p level was also well reflected in the Auger spec-
tral results. The K-L M M level appeared at 249 eV for the
3
45 45
calcined catalyst, compared with 250 eV for the spent catalysts.
3
.7.2. Surface composition
Table 4 gives the surface atomic ratios of calcined and spent
K-MgO catalysts. It is evident from the Mg/C and K/Mg ratios
that surface Mg content decreased during reaction. Very low
Mg/C was observed after reaction with 2:1, DMC:resorcinol.
However, reaction with 4:1, DMC:resorcinol feed had a greater
amount of surface Mg, as evident from the very low K/Mg
(0.017) and very high Mg/C (4.769) ratios. Surface K concen-
tration was high for the spent catalysts compared with the cal-
cined catalyst, except for 4:1, DMC:resorcinol reaction, as ev-
ident from the K/Mg ratio. Comparable K/C ratios were found
for all of the spent catalysts, except for a very high K/C af-
ter the catechol reaction, indicating the surface was enriched
with K. (K + Mg)/C showed very high (low) carbon content
after 2:1 (4:1), DMC:resorcinol reaction, indicating a drastic
change in surface composition, depending on the DMC content
in the reaction. Catechol and hydroquinone catalysts showed
comparable (K + Mg)/C ratios.
Fig. 13. X-ray photoemission spectra from K(5)-MgO spent catalysts after
−
1
O-methylation of dihydroxybenzenes at 583 K and at WHSV = 9.7 h
for
1
h on TOS with DMC:DHB = 2:1 feed mixture. XPS results from fresh
K(5)-MgO is also given for comparison. Dotted lines are guide to the eye.
served at 168 ppm in the resorcinol catalyst. Indeed, more
MgCO3 was evident on resorcinol from the relatively high
weight loss at around 1000 K than in catechol, which showed
only marginal weight loss (Figs. 11e–11f).
3
3
.7. XPS analysis
.7.1. O 1s, Mg 2p, C 1s, and K 2p core-level analysis
Fig. 13a shows O 1s spectra from K-MgO spent and calcined
catalysts. The O 1s spectrum was very broad for the calcined
catalyst, with a main peak centered at 530.2 eV and a shoulder
at 532.3 eV attributed to the hydroxyl groups from adsorbed
moisture on the catalysts [54]. Spent K-MgO catalysts showed
a sharp peak at 531 eV after a 1-h methylation test. A shift
in BE of around 1.8 eV (532 eV) was observed only on re-
sorcinol (DMC:resorcinol = 2:1 feed). This indicates an onset
in change in surface characteristics toward deactivation in re-
sorcinol (Fig. 5). Fig. 13b shows Mg 2p spectra from calcined
and spent K-MgO catalysts. A sharp peak at 50.3 eV was ob-
4. Discussion
4.1. O-methylation of DHBs on alkali ion-loaded MgO
Selective 2-MP production from catechol on MgO and
5 mmol alkali ion (Li, K, or Cs)-loaded MgO was carried out
with DMC; only K-MgO showed stable activity. Secondary
products like PCC, 1,2-DMB, and C-methylated catechol for-
mation competed with each other depending on the reaction
served for the calcined sample, as is typical for Mg2 . However,
+
the Mg 2p intensity observed at 49.4 eV was low on all of the

386
M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
Fig. 14. Pictorial representation of (a) catechol carbonate and (b) 1,2-dimethoxybenzene formation from catechol. O-methylation and intermolecular interactions on
c) resorcinol and (d) hydroquinone.
(
conditions. PCC formation increased with TOS on all of the
systems except K-MgO, in which a sizeable amount of 1,2-
DMB formed. In general, 2-MP selectivity (70–80%) was not
significantly affected by side products, catalysts, and reac-
tion conditions, with the exception of a higher DMC:catechol
catalyst surface. Note that C-methylated product formation was
significant on all of the catalyst systems except Cs-MgO, indi-
cating that moderate-strength basic sites may be responsible for
aromatic ring methylation. Indeed, the same trend was observed
with catechol and hydroquinone. The maximum and stable O-
methylation activity for all three DHB substrates observed on
K-MgO compared with other systems indicates that an opti-
mum basicity is required, not high (Cs-MgO) or low basicity
(MgO and Li-MgO).
3-MP selectivity in resorcinol methylation was mainly af-
fected by sequential methylation to 1,3-DMB. Note that when-
ever the resorcinol conversion increased, a simultaneous in-
crease in 1,3-DMB was also observed. Even with high DMC
content and stable resorcinol conversion on K-MgO at 583 K,
a large 1,3-DMB selectivity (75%) was observed during ini-
tial TOS (Fig. 6c). Thus, it is clear that in the transient state at
(
4) ratio. However, 1,2-DMB formation through PCC methy-
lation was confirmed by temperature-dependent and WHSV-
dependent studies. Thus, a sequential methylation step of 2-MP
to 1,2-DMB may either be ruled out or may have occurred
at a minimum level. The mono to di O-methylation ratio of
>
1 and the stable 2-MP selectivity of 70–75% at space veloc-
−1
ity <14.5 h indicated no sequential methylation of 2-MP to
,2-DMB. A linear increase in O-methylation capacity with K
concentration and very low levels of side products at high space
1
−1
velocity (ꢁ14.5 h ) clearly hinted that optimizing the process
conditions with selective removal of 2-MP could lead to the
selective production of 2-MP.
ꢁ
563 K, sequential methylation of 3-MP to 1,3-DMB occurred.
Similarity between the hydroquinone and catechol cases is
evident by the selective formation of methoxy phenols on K-
MgO with very low levels of side products at high WHSV
3
-MP selectivity increased to a greater extent at higher TOS as
−1
well as at higher space velocity (ꢁ18 h ), as evident from the
mono to di O-methylation ratio shown in Fig. 7b. The foregoing
findings indicate that a change in the catalyst surface occurred
in the first few hours on stream and reached a state at which
stable activity was observed, meriting further investigation.
−1
(
ꢁ14.5 h ). In contrast, ꢁ90% 4-MP selectivity at all WHSV
values indicated a perfect perpendicular orientation of the hy-
droquinone aromatic ring to the catalyst surface. Significant
formation of 1,4-DMB occurred at DMC:hydroquinone = 4 at
the cost of 4-MP, indicating sequential methylation; however,
isomerization of methylated hydroquinone to 1,4-DMB cannot
be ruled out. Indeed, ꢁ95% 1,2-DMB selectivity observed at
DMC:catechol = 4 supports the parallel orientation of catechol
on the catalyst surface (Fig. 14b).
4
.2. Reactivity of DHBs
Reaction results suggest that although the three DHB com-
pounds are just isomers, they have significantly different reac-
tivities. The variation in reactivity trends is due mainly to dif-
ferences in acidity. The acidities of these three compounds are
in the following order: catechol < hydroquinone < resorcinol.
Because MgO and alkali metal ion-loaded MgO are basic cat-
alysts, most acidic resorcinol chemisorbs strongly and deacti-
vates the active basic sites, which does not occur with other two
isomers. This finding is supported by the highest carbon content
observed in thermal analysis on spent catalyst from resorcinol.
Unlike catechol or hydroquinone methylation, resorcinol
methylation activity declined rapidly on all of the catalyst sys-
tems, due to resorcinol condensation producing polyphenolic
chains that block the active sites, as shown in NMR and thermal
analysis. However, K-MgO showed relatively better activity
among the systems studied. Further, stable resorcinol conver-
sion was observed with increasing amounts of DMC, due to
prevention of the intermolecular interaction of resorcinol on the

M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
387
The probability of the hydroxyl groups reacting toward cat-
alyst surfaces drives the product selectivity. For instance, cate-
chol has a high likelihood of both hydroxyl groups interacting
on the catalyst surfaces, as shown in Figs. 14a and 14b. Thus,
the formation of PCC and 1,2-DMB would be expected, as
was indeed observed. In general, 1,2-DMB formation is likely
through any of three possible routes: direct O,O-dimethylation
of catechol, PCC ring opening followed by dimethylation, and
methylation of 2-MP. 1,2-DMB formation through sequen-
tial methylation of 2-MP was already ruled out. Temperature-
dependent studies of product selectivity have shown that 1,2-
DMB forms only at the expense of PCC. An exception is direct
O,O-dimethylation of catechol to 1,2-DMB, which likely is
possible when excess DMC is available in the feed (Fig. 14b).
In resorcinol and hydroquinone, only one of the hydroxyl
groups interacted with the catalysts surfaces. Thus, the inter-
acting hydroxyl group was becoming methylated, and the se-
lectivity for MPs was quite high. However, the free hydroxyl
group likely facilitated an intermolecular interaction between
resorcinol molecules, leading to polyaromatic chain formation
Partial decomposition of polyaromatics to hydrogen-defi-
cient polyaromatics or graphitic carbon during thermal analysis
in N2 atmosphere is evident from the successive weight loss at
850 K during air atmosphere analysis of the same sample. It is
evident from the weight loss [resorcinol (63%) > hydroquinone
(50%) > catechol (40%)] that the most acidic resorcinol made
the highest molecular mass polyaromatic chains, followed by
hydroquinone and catechol. Condensation or polymerization of
the acidic reactants was boosted by the basic sites of the cata-
lysts.
XPS reflects the changes on the catalysts due to reaction
through changes in BE and surface atomic composition. The
2−
O
ions on the surface of spent catalysts have an electronic
interaction with DHB protons and hence a shift in the BE of
O 1s core level was observed compared with calcined cata-
lyst. Resorcinol, having the most acidic protons, interacted very
strongly, producing a large shift in O 1s BE. The increasing O
1s BE from 530.2 eV indicates that the basicity of the surface
decreased during methylation, to a large extent for resorcinol.
Thus, the descending basicity led to the deactivation with TOS
during resorcinol methylation.
(Fig. 14c). The spatial arrangement of OH groups in resorcinol
and its acidity enhanced the condensation process. In hydro-
quinone (Fig. 14d), the free OH group spatial arrangement was
such that it could condense with another gas-phase molecule;
however, the stable activity suggests that this condensation was
unlikely. Thus, no extensive polyaromatic chain formation with
hydroquinone was observed, as evident from thermal analysis
and catalytic activity.
Mg 2p core level appeared at a lower BE (by 1 eV) on spent
catalysts compared with fresh catalyst. It is likely that disso-
ciative adsorption of DHB as hydroxy phenolate anions and
protons occurred on the catalyst surface; the former made the
Mg sites electron-rich. The decreasing Mg 2p spectral inten-
sity on spent catalyst indicates that the surface Mg species
was buried under the deposited carbon species. The absence of
Mg 2p core-level emissions in resorcinol (DMC:resorcinol = 2)
reveals a severe coke deposition and hence a burial of Mg into
the bulk. Indeed, Cs-MgO-Res showed no Cs2O phase on XRD,
supporting the significant coke deposition. The very low surface
Mg content on K-MgO-Res (2:1) catalyst changed dramatically
to very high when using excess DMC (K-MgO-Res—4:1), be-
cause excess DMC in the feed hindered effective intermolecular
interaction, as suggested earlier. This also confirms the need for
basic MgO support for the reaction.
The K 2p core level observed at lower BE than that of fresh
catalyst is also reflected in the K-L3M45M45 Auger transition
appearing at higher kinetic energy. This shift in the binding and
kinetic energies are due to the interaction of hydroxy phenolate
anions with the K site. The high K content on spent catalyst sur-
faces compared with that on fresh K-MgO indicates significant
surface segregation of K due to reaction at >583 K.
2+
4
.3. Physicochemical and spectroscopic analysis
XRD of spent catalysts revealed that reaction conditions led
to a significant breakup of crystallites on MgO, Cs-MgO, and
especially Li-MgO. Thus, heavy carbon deposition on the cata-
lyst surfaces buried the active centers completely and rapidly,
leading to faster deactivation on the above catalyst systems.
Surface area analysis of spent catalysts also supports the above
conclusions. In contrast, marginal changes in the crystallite size
and surface area of K-MgO show its robustness to reaction con-
ditions and hence stable activity.
Thermal analysis in N2 atmosphere revealed the presence of
adsorbed reactants and/or products along with polyaromatics,
as evident from the low-temperature (below 1000 K) weight
loss peak. The minor weight loss observed at 580 K in air at-
mosphere and the broad weight loss observed at 500–850 K
in N2 atmosphere in catechol and its optimum reaction tem-
perature at 583 K supports that the above carbon species are
from adsorbed molecular components. Similar molecular com-
ponent desorption in N2 atmosphere was seen in hydroquinone
5. Conclusion
A comprehensive study on O-methylation of DHBs with
DMC was carried out on MgO and alkali metal ion (Li, K,
and Cs)-loaded MgO under vapor-phase conditions. Pure MgO
had O-methylation activity that declined rapidly with time on
stream. Alkali metal ion-loaded MgO showed better activity
than pure MgO with all substrates. Among the catalyst sys-
tems screened, K-MgO showed the best mono O-methylation
activity with relatively high stability for all substrates, hinting
at the optimum basicity requirement. Reactivity and product
selectivity were influenced by substrate acidity and mode of in-
(
at 570–850 K) and resorcinol (at 573–1000 K). The high tem-
perature (740 K) of the second weight loss peak in resorcinol
13
suggests the strong chemisorption of resorcinol. C CP-MAS
NMR results obtained on the spent catalysts exhibiting chemi-
cal shifts characteristic of different carbon species were in good
agreement with thermal analyses. Chemical shifts relating to
carbonate carbon on MgCO3 indicated the interaction between
MgO and CO2 or DMC directly.

388
M. Vijayaraj, C.S. Gopinath / Journal of Catalysis 243 (2006) 376–388
teraction on the catalyst surfaces. Catechol interacting through
,2 hydroxyl groups on the catalyst surfaces gave more sec-
[18] T.M. Jyothi, T. Raja, M.B. Talawar, B.S. Rao, Appl. Catal. A Gen. 211
2001) 41.
19] T. Mathew, N.R. Shiju, K. Sreekumar, B.S. Rao, C.S. Gopinath, J. Catal.
10 (2002) 405;
T. Mathew, N.R. Shiju, V.V. Bokade, B.S. Rao, C.S. Gopinath, Catal.
Lett. 94 (2004) 223.
[20] T. Mathew, S. Sylesh, B.M. Devassy, M. Vijayaraj, C.V.V. Sathya-
narayana, B.S. Rao, C.S. Gopinath, Appl. Catal. A Gen. 273 (2004) 35.
(
1
[
ondary products, including PCC and veratrole; however, 2-MP
selectivity possibly can be further maximized by varying the
reaction conditions. Resorcinol and hydroquinone interactions
left one hydroxyl group free in the space, and this free spa-
tial hydroxyl group aided polyaromatic chain formation and
blocked the active sites leading to catalyst deactivation with
TOS. Low temperature (<610 K) led to high selectivity for
MPs compared with the dimethoxybenzene selectivity reported
in the literature. Large amounts of carbon deposits, as detected
by thermal analysis, shed light on the mechanisms behind this
catalyst deactivation. 13C NMR revealed the presence of heav-
ier polyaromatics and surface carbonates on spent catalysts. The
surface basicity decreased during DHB interaction, as shown by
XPS studies, and the significant basicity loss led to faster deac-
tivation in the case of resorcinol methylation.
2
[
21] T. Mathew, M. Vijayaraj, S. Pai, B.B. Tope, B.S. Rao, S.G. Hegde, C.S.
Gopinath, J. Catal. 227 (2004) 175.
[
[
22] R. Bal, S. Sivasanker, Appl. Catal. A Gen. 246 (2003) 373.
23] R. Bal, S. Sivasanker, Green Chem. 2 (2000) 106.
[24] R. Pierantozzi, A.F. Nordquist, Appl. Catal. 21 (1986) 263.
[
[
[
25] M.C. Samolada, E. Grigoriadou, Z. Kiparissides, I.A. Vasalos, J. Catal.
152 (1995) 52.
26] G. Sarala Devi, D. Giridhar, B.M. Reddy, J. Mol. Catal. A Chem. 181
(2002) 173.
27] S. Balsama, P. Beltrame, P.L. Beltrame, P. Carniti, L. Forni, G. Zurelti,
Appl. Catal. 13 (1984) 161.
[28] E. Santacesaria, M. Discrio, P. Ciambelli, D. Gelosa, S. Carra, Appl.
Catal. 64 (1990) 101.
[
[
29] V. Durgakumari, S. Narayanan, L. Guczi, Catal. Lett. 5 (1990) 377.
30] Phenol Derivatives, in: Ullmann’s Encyclopedia of Industrial Chemistry,
vol. A19, VCH, Weinheim, 1991, p. 354.
Acknowledgments
The authors thank the reviewers for many helpful sugges-
tions for improving the manuscript. They also thank Drs. P.R.
Rajmohanan, T.G. Ajithkumar, P.A. Joy, and S. Mayadevi for
helpful discussions on the NMR and thermal analysis results.
M.V. thanks CSIR, New Delhi for a research fellowship.
[31] Y. Fu, T. Baba, Y. Ono, Appl. Catal. A Gen. 166 (1998) 419.
[32] Y. Fu, T. Baba, Y. Ono, Appl. Catal. A Gen. 166 (1998) 425.
[
[
[
[
[
33] Y. Fu, T. Baba, Y. Ono, Appl. Catal. A Gen. 176 (1998) 201.
34] Y. Fu, T. Baba, Y. Ono, Appl. Catal. A Gen. 178 (1999) 219.
35] L. Calzolari, F. Cavani, T. Monti, Solid State Chem. Catal. 3 (2000) 533.
36] F. Cavani, T. Monti, Catal. Org. Reactions 82 (2001) 123.
37] F. Cavani, T. Monti, D. Paoli, Stud. Surf. Sci. Catal. 130 (2000) 2633.
References
[38] V. Vishwanathan, S. Ndou, L. Sikhwivhilu, N. Plint, K. Vijayaraghavan,
N.J. Coville, Chem. Commun. (2001) 893.
[
39] M.B. Talawar, T.M. Jyothi, P.D. Sawant, T. Raja, B.S. Rao, Green Chem. 2
2000) 266.
[
1] T. Kotanigawa, M. Yamamoto, K. Shimokawa, Y. Yoshida, Bull. Chem.
Soc. Jpn. 44 (1971) 1961.
(
[40] T.M. Jyothi, T. Raja, M.B. Talawar, B.S. Rao, Appl. Catal. A Gen. 211
(2001) 41.
[2] T. Kotanigawa, K. Shimokawa, Bull. Chem. Soc. Jpn. 47 (1974) 1535.
3] K. Tanabe, H. Hattori, T. Sumiyoshi, K. Tamaru, T. Kondo, J. Catal. 53
[
(
1978) 1.
[41] L. Kiwi-Minsker, S. Porchet, R. Doepper, A. Renken, Stud. Surf. Sci.
Catal. 108 (1999) 149.
[42] L. Kiwi-Minsker, G. Jenzer, L. Pliasova, A. Renken, Stud. Surf. Sci.
Catal. 121 (1999) 159.
[
[
4] H. Grabowska, W. Kaczmarczyk, J. Wrzyszcz, Appl. Catal. 49 (1989) 351.
5] H. Grabowska, W. Mista, L. Syper, J. Wrzyszcz, M. Zawadski, J. Catal.
1
60 (1996) 134.
[
[
[
[
6] H. Grabowska, W. Mista, L. Syper, J. Wrzyszcz, M. Zawadski, Appl.
Catal. 144 (1996) L1.
7] H. Grabowska, W. Mista, L. Syper, J. Wrzyszcz, M. Zawadski, Angew.
Chem. Int. Ed. Engl. 35 (1996) 1562.
8] H. Grabowska, W. Mista, J. Trawczynski, J. Wrzyszcz, M. Zawadski,
Appl. Catal. A Gen. 220 (2001) 207.
[43] R. Bal, B.B. Tope, S. Sivasanker, J. Mol. Catal. A Chem. 181 (2002) 161.
[44] R. Bal, S. Mayadevi, S. Sivasanker, Org. Process Res. Dev. 7 (2003) 17.
[45] X. Zhu, X. Li, M. Jia, G. Liu, W. Zhang, D. Jiang, Appl. Catal. A Gen. 282
(2005) 155.
[46] Z. Fu, Y.Y.D. Yin, Y. Xu, H. Liu, H. Liao, Q. Xu, F. Tan, J. Wang, J. Mol.
Catal. A Chem. 232 (2005) 69.
9] H. Grabowska, R. Klimkiewicz, W. Tylus, P.J. Godowski, Appl. Catal. A
Gen. 240 (2003) 111.
[47] S.B. Waghmode, R. Vetrivel, S.G. Hegde, C.S. Gopinath, S. Sivasanker,
J. Phys. Chem. B 107 (2003) 8517.
[
[
10] E. Santacesaria, D. Grasso, S. Carra, Appl. Catal. 64 (1990) 83.
11] V. Durgakumari, S. Narayanan, J. Mol. Catal. 65 (1991) 385;
V. Durgakumari, G. Sreekanth, S. Narayanan, Res. Chem. Intermed. 14
[48] V.K. Diez, C.R. Apesteguia, J.I. Di Cosimo, J. Catal. 240 (2006) 235.
[49] T. Mathew, B.S. Rao, C.S. Gopinath, J. Catal. 222 (2004) 107.
[50] M. Vijayaraj, C.S. Gopinath, J. Catal. 241 (2006) 83.
[51] N.F.M. Henry, J. Lipsons, W.A. Wooster, The Interpretation of X-Ray Dif-
fraction Photographs, Macmillan, London, 1951.
[52] The Aldrich Library of C and H FT NMR Spectra, vol. 2, first ed.,
Aldrich Chemical Co., Inc., Milwaukee, 1993.
[53] H.W. Papenguth, R.J. Kirkpatrick, B. Montez, P.A. Sandberg, Am. Min-
eral. 74 (1989) 1152.
(
1990) 223.
12] A. Vinu, M. Karthik, M. Miyahara, M. Murugesan, K. Ariga, J. Mol. Catal.
30 (2005) 155.
[
13
1
2
[13] S. Velu, C.S. Swamy, Appl. Catal. A Gen. 119 (1994) 241.
[14] S. Velu, C.S. Swamy, Appl. Catal. A Gen. 145 (1996) 141.
[15] S. Velu, C.S. Swamy, Appl. Catal. A Gen. 162 (1997) 81.
[16] W.C. Choi, J.S. Kim, T.H. Lee, S.I. Woo, Catal. Today 63 (2000) 229.
[17] T. Tsai, F. Wang, Catal. Lett. 73 (2001) 167.
[54] A. Miyakoshi, A. Ueno, M. Ichikawa, Appl. Catal. A Gen. 219 (2001)
249.