Characterization and reactivity of Pd/MgO and Pd/?3-Al2O3 catalysts in the selective hydrogenolysis of CCl2F2

Article abstract of DOI:10.1021/jp012357a

Al₂O₃ and MgO supported Pd catalysts with 6 wt % loading are prepared by the wet impregnation method. The catalysts are made into two parts; one of them is dried at 110 ?°C, and the other one is calcined at 500 ?°C. Conversion of CCl

Full text of DOI:10.1021/jp012357a

1024
J. Phys. Chem. B 2002, 106, 1024-1031
Characterization and Reactivity of Pd/MgO and Pd/γ-Al2O3 Catalysts in the Selective
†
Hydrogenolysis of CCl2F2
Hari Padmasri Aytam, Venugopal Akula, Krishnamurthy Janmanchi,
Seetha Rama Rao Kamaraju, and Kanta Rao Panja*
Catalysis and Physical Chemistry DiVision, Indian Institute of Chemical Technology, Hyderabad-500 007, India
Kishan Gurram and J. W. Niemantsverdriet
Schuit Institute of Catalysis, Technical UniVesity of EindhoVen, 5600 MB EindhoVen, The Netherlands
ReceiVed: June 20, 2001; In Final Form: October 18, 2001
Al2O3 and MgO supported Pd catalysts with 6 wt % loading are prepared by the wet impregnation method.
The catalysts are made into two parts; one of them is dried at 110 °C, and the other one is calcined at 500
°C. Conversion of CCl2F2 in hydrogen is carried out under identical reaction conditions on both dried and
calcined catalysts after the catalysts are prereduced in H2 at 400 °C for 3h. The fresh and the used catalysts
are characterized by BET-surface area, X-ray diffraction (XRD), temperature programmed reduction (TPR),
temperature programmed desorption (TPD) of NH3, and X-ray photoelectron spectroscopy (XPS). XPS data
shows that surface Pd species are more in MgO supported catalyst than in Al2O3 supported one. In used
catalysts, surface F- concentration is more on MgO than on Al2O3 supported Pd catalyst. The MgO supported
Pd catalyst (dried) showed higher reactivity and CH2F2 selectivity compared to other catalysts. MgO support
is found to be superior to Al2O3 support for Pd for the reaction.
1. Introduction
the presence of CH2 species in the used catalysts. Formation of
propene is observed as an addition product with the addition of
C2H4 to the surface CH2 species.15 The nature of the support
has a great influence on the characteristics and activity of
supported Pd catalysts for the hydrogenolysis of CCl2F2 and
selectivity to CH2F2. The present work is aimed at the study of
the conversion of CCl2F2 in hydrogen to selectively form CH2F2
over Pd supported on MgO and comparing the activity of this
catalyst with that of Pd/γ-Al2O3 under identical reaction
conditions. In this regard, Pd/MgO and Pd/γ-Al2O3 have been
prepared and characterized both before and after the reaction
by XRD, XPS, TPR, and TPD of NH3 and BET-surface area to
understand the nature and the role of the two different supports
toward the selective hydrodechlorination of CCl2F2. The physical
characteristics of the catalysts are correlated with the reactivity
of the catalysts, wherever possible.
Selective hydrogenolysis of chlorofluorocarbons (CFCs) is
an important reaction, as it is one of the best routes not only
for the conversion of CFCs into ozone friendly substances, but
also for the synthesis of hydrofluorocarbons (HFCs) and
hydrochlorofluorocarbons (HCFCs), the elusive replacements
of the ozone depleting CFCs.1-20 Hydrodechlorination of CFCs
involves the hydrogenolysis of C-Cl bond. Supported palladium
catalysts are reported to be very selective and preferred catalysts
for this reaction in the formation of hydrofluorocarbons, the
desired products. According to Coq et al.,10 metal fluoride
supported Pd catalysts are more selective for CH2F2 formation
in the hydrodechlorination of CCl2F2. Juszczyk et al.,11 showed
that the extent of hydrodechlorination of CCl2F2 greatly depends
on Pd dispersion over Pd//γ-Al2O3 catalysts. Larger Pd particles
are reported to be more easily transformed into Pd carbide and
are more active and selective toward CH2F2. Makkee et al.12,13
have studied the hydrodechlorination of CCl2F2 over Pd on
activated carbon and Pd black catalysts and reported that PdCx
species facilitates high CH2F2 selectivity. Kim et al.21 reported
a long catalyst life and high selectivity to CHCl3 in the
hydrodechlorination of CCl4 over Pt supported on MgO. Basicity
of MgO seems to be responsible for retarding the coke formation
and suppression of C2 oligomers, which are observed to have
formed in good amounts over other supported catalysts.22 MgF2-
supported Pd is reported by Malinowski et al.,14 to be a very
stable and selective catalyst for the synthesis of CH2F2 (HFC-
32) by the hydrodechlorination of CCl2F2 (CFC-12). Deshmukh
and d′Itri15 have done the transient kinetic experiments using
C2H4 as a trapping agent for the surface carbenes and found
2. Experimental Section
2.1. Preparation of Catalysts. MgO and γ-Al2O3 supported
6 wt % Pd catalysts are prepared by the wet impregnation
method using PdCl2 as a precursor. Alumina, (Harshaw) with a
BET- surface area of 220 m2g-1 is calcined at 500 °C for 5 h
in air before use. MgO is prepared by the precipitation of Mg-
(OH)2 from aqueous Mg(NO3)2 precursor using dil. aqueous
NaOH solution followed by washing with distilled water, drying
overnight at 110 °C in an oven and calcination in air at 500 °C
for 18 h. The BET-surface area of the MgO support thus
prepared is found to be 110 m2g-1. Requisite amount of acidified
(with dil. HCl) aqueous PdCl2 solution is used to impregnate
the support particles (18/20 BSS mesh) and the excess solution
is evaporated on a water bath and then dried overnight in oven
at 110 °C. The dried catalysts are made into two parts and one
part of alumina supported sample is calcined at 500 °C in air
for 5 h and that of MgO supported one is calcined at 500 °C in
* To whom correspondence should be addressed. Fax: +91-40-
7160921 & 7163757. E-mail: pkr@iict.ap.nic.in.
† IICT Communication Number: 4827.
10.1021/jp012357a CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/09/2002

Pd/MgO and Pd/γ-Al2O3 Catalysts
J. Phys. Chem. B, Vol. 106, No. 5, 2002 1025
TABLE 1: BET-surface Areas and Ammonia Uptakes of the
Fresh and Used Catalysts of 6 wt %Pd Supported on
γ-Alumina and Magnesia
air for 18 h. The dried and calcined catalysts samples are reduced
in hydrogen at 400 °C for 3 h before the reaction is carried out.
2.2. Characterization of Catalysts. XRD patterns are
obtained on a Siemens D5000 X-ray diffractometer using Ni
filtered Cu KR radiation. Surface area measurements are made
by N2 adsorption at liquid nitrogen temperature using BET-
method at 10-6 Torr on all-glass high vacuum system.
TPR studies are made on an on-line reactor system connected
to a GC equipped with a thermal conductivity detector (TCD)
and a data station for recording the TPR profiles. About 200
mg of catalyst sample is placed in the reactor and heated at a
ramp of 5 °C/min from ambient temperature to 700 °C by
passing the reducing gas mixture (6%H2 in argon) over the
catalyst. Prior to the start of the temperature program, the
reducing gas (20 mL/min.) is passed over the catalyst for 30
min. The gas mixture from the outlet of the reactor is passed
through a dilute alkali (KOH) trap to remove HCl/HF, the
byproducts of the reaction, and then through a molecular sieve
trap to remove the moisture. The mixture is then let into the
GC column. There is a provision to analyze reduction products
of the catalysts by using a flame ionization detector (FID) of
the GC connected in series to TCD and for recording the
profiles.
The XPS spectra are recorded on a VG Escalab 200 MK
spectrometer equipped with an Al-KR source and a hemi-
spherical analyzer connected to a five-channel detector. During
measurements the base pressure of the system is maintained
around 5 × 10-10 mbar. The spectra are recorded with constant
pass energy of 20 eV. Binding energies are determined by
computer fitting of the measured spectra. Samples are pressed
in indium foil for the study. Binding energy correction is
performed by using the C 1s peak at 284.6 eV as a reference.
The relative intensities of the surface composition of different
elements are measured by correcting them with their corre-
sponding atomic sensitivity factors.23
TPD experiments are performed using a reactor system
connected to a GC equipped with a TCD, which in turn is
connected to a data station. Ammonia is used as a probe
molecule for evaluating the acid site distribution of the catalysts
in the present study. In a typical experiment ∼250 mg of catalyst
is taken in the reactor and pretreated in a flowing He at 250 °C
for 2 h. After pretreatment, the temperature is brought down to
80 °C (under He flow) where the catalysts are saturated with
flowing NH3 gas for 2h. Then the catalysts are flushed with He
at 100 °C for 3h in order to remove the physisorbed NH3 on
the catalyst surface. Temperature programmed desorption has
been performed at a rate of heating of 5 °C/min from 100 to
500 °C. Acidity measurements are standardized by taking pulse
injections of known amounts of NH3 at the same experimental
conditions.
2.3. Activity Study. Reaction of CCl2F2 in hydrogen over 6
wt %Pd/MgO and 6 wt %Pd/γ-Al2O3 (dried and calcined)
catalysts is carried out in a fixed bed microreactor connected
on-line to a gas chromatograph using a FID. An alkali (KOH)
trap is connected to the outlet of the reactor to remove HCl and
HF that form as byproducts before the product mixture is
injected into the GC using a six-way sampling valve. The
reactivity of the catalysts is carried out at a reaction temperature
of 250 °C, a molar ratio of H2/CFC-12 of 8, and a GHSV (gas
hourly space velocity) of 4,000 h-1. During a preliminary
study,24 these reaction conditions have been found to be
optimum for obtaining good activity and selectivity. The
products are analyzed by GC under steady-state conditions (at
4-5 h of the reaction) using a Porapak-Q (3 m long and 1/8 in.
dia made of ss) column and confirmed by GC-MS analysis.
BET-surface
area (m2g-1
)
fresh
used
catalyst fresh used weaka moderateb strongc weaka moderateb strongc
Pd/MgO
dried
Pd/MgO
calcined
Pd/Al2O3
dried
Pd/Al2O3
calcined
32
41
27
33
62
92
36
42
22
36
42
22
34
65
18
16
29
46
100
49
120
65
78
31
42
38
115
152
16
51
27
117
a NH3 uptake in µmolg-1 in the temperature range from 200 to 300
°C. b NH3 uptake in µmolg-1 in the temperature range from 300 to
400 °C. c NH3 uptake in µmolg-1 in the temperature range from 400
to 500 °C.
3. Results & Discussion
3.1. BET-Surface Areas. The surface areas of the supports
MgO and γ-Al2O3 measured by the BET-method reduced
drastically after the impregnation of Pd and further the surface
areas (Table 1) of the fresh catalysts are found to decrease after
the reaction indicating transformation of the catalyst surface into
metal fluoride or metal hydroxo fluorides that occur during the
course of reaction.
3.2. XRD Results of Fresh and Used Catalysts. XRD
patterns of the fresh and used Pd/Al2O3 and Pd/MgO catalysts
are shown in Figures 1 and 2, respectively. In Al2O3 and MgO
supported Pd catalysts in both calcined and dried forms, signals
due to the support i.e., γ-Al2O3 (2θ ) 66° with “d” value of
1.4, 2θ ) 45° with “d” value of 1.99 and 2θ ) 36.1° with “d”
value of 2.46) and MgO (2θ ) 42.3° with “d” value of 2.11,
2θ ) 61.5° with “d” value of 1.49 and 2θ ) 77.3° with “d”
value of 1.22) & Mg(OH)2 (2θ ) 18.4° with “d” value of 4.77,
2θ ) 37.5° with “d” value of 2.37 and 2θ ) 50.4° with “d”
value of 1.79) can be seen. No signals due to PdCl2 are observed
from these patterns. In used Pd/MgO and Pd/Al2O3 catalysts,
both calcined and dried, PdCx phases (2θ ) 39.5° with “d” value
of 2.303, 2θ ) 45.5° with “d” value of 1.991 and 2θ ) 66.5°
with “d” value of 1.405) and R-Pd phase (2θ ) 40° with “d”
value of 2.25, 2θ ) 46.5° with “d” value of 1.95 and 2θ )
67.9° with “d” value of 1.38) are observed. Juszczyk et al.11
have reported the presence of PdCx and AlFx phases in used
catalysts after hydrodechlorination of CCl2F2 over Pd/Al2O3.
Phases due to MgO and MgF2 are observed in Pd/MgO used
catalysts. The nonstoichiometric phase due to AlF1.65(OH)1.35
xH2O and γ-Al2O3 phases are found in Pd/Al2O3 used catalysts.
Kanta Rao and co-workers25 reported the transformation of
catalyst material into metal hydroxo fluorides during the
dismutation of CCl2F2 over alumina based catalysts and Kemnitz
and co-workers26 reported the formation of large amount of
AlF2(OH) pyrochlore structure during the activation of γ-Al2O3
with CFC molecule in gas phase. Coq et al.10 reported the same
in the hydrogenolysis reaction of CCl2F2 over Pd/alumina.
3.3. TPR Analysis. TPR of Fresh Catalysts. Figure 3 shows
the TPR patterns of fresh Pd/γ-Al2O3 and Pd/MgO catalysts
both dried and calcined varieties. The fresh catalysts of Pd/γ-
Al2O3 showed only a single negative peak around 90 °C, which
may be attributed to the decomposition of â-PdHx.20,27 In
hydrogen atmosphere, PdCl2 or PdO gets reduced easily at
ambient temperature to Pd metal28 and further interacts with
hydrogen forming â-PdHx which appears to have occurred
during the passage of reducing gas prior to the start of the

1026 J. Phys. Chem. B, Vol. 106, No. 5, 2002
Aytam et al.
Figure 1. XRD patterns of fresh and used 6 wt %Pd/γ-Al2O3 catalysts (dried and calcined). (a) Pd/γ-Al2O3 (fresh; dried) (b) Pd/γ-Al2O3 (fresh;
calcined) (c) Pd/γ-Al2O3 (used; dried) (d) Pd/γ-Al2O3 (used; calcined).
temperature program. The intensity of this peak is more on
calcined catalyst, which may be attributed to lower Pd dispersion
in this catalyst. According to Vannice et al.29,30 in Pd supported
on active carbon suppression of â-PdHx formation is due to
carbon contamination of the surface and bulk Pd (reduction of
Pd particle size).
Chen et al.33 reported to have observed two desorption peaks
one at 300 °C and the other at 400 °C in the TPD of hydrogen
over Pd/MgO calcined catalysts. The high-temperature positive
peak above 500 °C shows further consumption of hydrogen
which may be attributed to the spill over of the hydrogen
activated on metallic Pd on to the MgO support.34
The TPR patterns of Pd/MgO fresh catalysts are found to be
complex unlike the patterns of Pd/γ-Al2O3 fresh catalysts. No
negative peak corresponding to â-PdHx decomposition is
observed over Pd/MgO fresh catalysts. One probable reason may
be the high dispersion of Pd over MgO as evident from the
XPS data. Increase in dispersion of Pd in Pd/SiO2 system or
decrease in Pd crystallite size is reported to result in the decrease
in the intensity of â-PdHx decomposition peak.27
A strong positive peak with a Tmax at ∼160 °C is observed
in both dried and calcined fresh catalysts of Pd/MgO with a
shoulder ∼270 °C in calcined one. This peak probably corre-
sponds to PdO reduction and/or to hydrogen consumption of
metallic Pd (XPS data reveals the stabilization of some part of
palladium in metallic form over MgO support) that is formed
and/or hydrogen adsorption directly on MgO.31,32 The intensity
of the peak with Tmax at ∼160 °C being higher with the calcined
catalyst compared to that over dried one gives credence to PdO
reduction.
TPR of Used Catalysts. The TPR patterns of used Pd/Al2O3
and Pd/MgO catalysts (Figure 4) show a negative peak centered
at ∼100 °C both in calcined and dried forms. This peak is
attributed to the decomposition of â-PdHx.27 Positive signals at
∼425 °C in used Pd/Al2O3 calcined and dried catalysts can be
attributed to the consumption of hydrogen by the carbonaceous
species present on the catalyst surface. Evolution of CH4 is
confirmed from the FID pattern of the TPR reactor outlet stream
(Figure 4c). The negative signal at ∼575 °C may be attributed
to the reaction of CHx species (x g 2) with low amount of H2
and evolution of hydrocarbons. Negative signals centered at Tmax
∼450 °C are found in both calcined and dried used Pd/MgO
catalysts. Signal centered at a Tmax of 680 °C is present in
calcined used catalyst only. These peaks may be attributed to
the evolution of methane confirmed from FID pattern of the
TPR reactor outlet stream and shown in Figure 4f particularly
by the reduction of coke moieties (CHx, x g 2) present on the
catalyst surface. Deshmukh and d′Itri have reported the presence
of surface CH2 species in the used Pd/AlF3 catalysts.15
A negative peak centered at ∼400 °C is seen in both the fresh
catalysts, which may be attributed to the partial desorption of
hydrogen that is chemisorbed on both palladium and magnesia.
3.4. XPS Results. Figure 5 shows the X-ray photoelectron
spectra of 6 wt %Pd/γ-Al2O3 fresh and used catalysts in two

Pd/MgO and Pd/γ-Al2O3 Catalysts
J. Phys. Chem. B, Vol. 106, No. 5, 2002 1027
Figure 2. XRD patterns of fresh and used 6 wt %Pd/MgO catalysts (dried and calcined). (a) Pd/MgO (fresh; dried) (b) Pd/MgO (fresh; calcined)
(c) Pd/MgO (used; dried) (d) Pd/MgO (used; calcined).
Figure 3. TPR patterns of fresh 6 wt %Pd/γ-Al2O3 and 6 wt %Pd/
Figure 4. TPR patterns of used 6 wt % Pd/γ-Al2O3 and 6 wt %Pd/
MgO catalysts. (a) Pd/γ-Al2O3 (used; dried) (b) Pd/γ-Al2O3 (used;
calcined) (c) FID pattern of Pd/γ-Al2O3 (used; calcined) (d) Pd/MgO
(used; dried) (e) Pd/MgO (used; calcined) (f) FID pattern of Pd/MgO
(used; calcined).
which is close to that of PdCl2 (338 eV).35 On the other hand,
the binding energy value of Pd/Al2O3 fresh calcined catalyst
MgO catalysts. (a) Pd/γ-Al2O3 (fresh; dried) (b) Pd/γ-Al2O3 (fresh;
calcined) (c) Pd/MgO (fresh; dried) (d) Pd/MgO (fresh; calcined).
different forms (dried and calcined) each. Pd/Al2O3 fresh
catalysts (dried and calcined) showed one doublet indicating
the Pd in oxide form. The Pd/Al2O3 dried catalyst (Figure 5a)
showed a binding energy (BE) value of 337.6 eV for Pd 3d 5/2
,

1028 J. Phys. Chem. B, Vol. 106, No. 5, 2002
Aytam et al.
Figure 6. XPS spectra of 6 wt %Pd/MgO. (a) Pd/MgO (fresh; dried)
(b) Pd/MgO (fresh; calcined) (c) Pd/MgO (used; dried) (d) Pd/MgO
(used; calcined).
Figure 5. XPS spectra of 6 wt %Pd/ γ-Al2O3. (a) Pd/γ-Al2O3 (fresh;
dried) (b) Pd/γ-Al2O3 (fresh; calcined) (c) Pd/γ-Al2O3 (used; dried)
(d) Pd/γ-Al2O3 (used; calcined).
between the B.E. values of Al 2p in oxides (74.2-74.6 eV)
and AlF3 (76.1-77.0 eV),23 indicating the transformation of
alumina to perhaps aluminum hydroxo fluoride.
(Figure 5b) is at 337.2 eV indicating that Pd is in oxide form,
which is logical for the calcined catalyst. The used catalysts-
(Figure 5c and 5d, for dried and calcined respectively) showed
the Pd 3d5/2 B. E. values of around 334.9 eV, which indicates
that Pd is reduced to Pd metal during the hydrogenolysis
reaction.
The F 1s B.E. values have been reported by Kemnitz et al.38
to be in the range of 687.0 to 687.8 eV for AlF2.3(OH)0.7‚H2O
and â-AlF3 catalysts from the XPS analysis of these samples.
The F 1s B.E values over Pd/Al2O3 used catalysts are found to
be 686.4 eV and 685.5 eV (not shown in the figure) in the dried
and calcined ones, respectively. These values also suggest the
possible formation of aluminum hydroxo fluoride from alumina
during the course of the reaction (with the release of HF). The
F 1s spectra of Pd/MgO used catalysts (not shown in the figure)
showed the photo peaks at B.E values of 685.5 and 685.4 eV
for dried and calcined catalysts, respectively. These values
confirm the transformation of MgO to MgF2.
Figure 6 shows the XPS spectra of 6 wt % Pd/MgO fresh
and used catalysts of dried and calcined forms. In Pd/MgO dried
catalyst (Figure 6a) the B.E of Pd 3d5/2 is 336.0 eV and in
calcined catalyst, this value is 336.9 eV with a shoulder at 335.6
eV (Figure 6b). Although the B.E. value of the dried catalyst
indicates that Pd is present, mainly in metallic form, that of the
calcined catalyst shows that it is predominantly present as PdO
as well as metallic Pd. In dried catalyst, it might also contain
some in the form of Pd(H2O)2+ 35
.
The B.E. values of Pd 3d5/2
The surface compositions of Pd/Al, Pd/Mg, F/Mg, F/Al, and
Pd/C ratios are presented in Table 2. The ratios have been
calculated from the relative intensities obtained from corrected
XPS of Pd 3d, F 1s, Mg 2s, Al 2p, and C 1s (by their
corresponding atomic sensitivity factors as cited in the litera-
ture).23 Even though Pd loading of 6 wt % is maintained in all
the catalysts, it is evident from Table 2 that Pd is in more
dispersed state on MgO surface than on alumina in both fresh
and used catalysts. The F/Al ratio (F/Al < 3) indicates a partial
transformation of alumina to hydroxo fluorides. On the other-
hand, the F/Mg ratio, which is found to be >2, suggests high
enrichment of fluoride on Pd/MgO catalyst surface during the
course of hydrogenolysis reaction. The apparent surface enrich-
ment of fluoride in used Pd/MgO catalysts beyond a F/Mg ratio
of 2 may be attributed to the deposition of heavy halogenated
hydrocarbons on the surface of the catalyst during the hydro-
in the Pd/MgO used catalysts are 334.6 and 335 eV for dried
and calcined varieties, respectively. These values clearly indicate
the existence of Pd in metallic form (B.E. value of Pd 3d5/2 in
metallic form is 335.2 eV). The B.E. values of Pd 3d5/2 show
that Pd is stabilized to some extent in the form of metal over
MgO supported fresh catalysts and on the other hand as Pd2+
over alumina supported fresh catalysts.36,37
On comparison of Al 2p and Mg 2p B.E. values of fresh and
used catalysts, the extent of fluorination appears to be higher
over MgO than on alumina. The B.E. of Mg 2p of fresh Pd/
MgO is shifted from 49.8 eV to 50.8 eV23 in used catalysts
(not shown in figure) indicating that MgO is transformed
partially to MgF2. Similarly, the B.E. values of Al 2p in fresh
Pd/alumina catalysts (74.4 eV) are changed to 74.8-75.2 eV
in used catalysts (not shown in figure), which are intermediate

Pd/MgO and Pd/γ-Al2O3 Catalysts
J. Phys. Chem. B, Vol. 106, No. 5, 2002 1029
TABLE 2: Surface Compositions of the Pd, Al, Mg, C, and
F Calculated from the XPS Analysis of the Fresh and Used
Al2O3 and MgO Supported 6 wt %Pd Catalysts
TABLE 3: Hydrodechlorination of CCl2F2 over 6 wt % Pd
Supported on MgO and γ-Al2O3, Calcined and Dried
Catalysts at a Reaction Temperature of 250 ×bcC, GHSV of
4,000h-1 and H2/CCl2F2 of 8
catalyst
Pd/Al2O3
dried; fresh
Pd/Al2O3
calcined; fresh
Pd/Al2O3
dried; used
Pd/Al2O3
calcined; used
Pd/MgO
dried; fresh
Pd/MgO
calcined; fresh
Pd/MgO
dried; used
Pd/MgO
calcined; used
Pd:Al
Pd:Mg
F:Al
F:Mg
Pd:C
% selectivities
% conversion
0.0308
catalyst
of CCl2F2
CH2F2 CH4 CH3F CHClF2 others
0.0157
0.0050
0.0139
Pd/MgO
dried
Pd/MgO
calcined
Pd/Al2O3
dried
70
64.3 32.6
-
-
3.1
4.1
3.6
2.6
-
1.1016
1.1671
-
0.0461
0.0905
58
27
50
63.6 29.2 3.1
-
-
-
45.6 49
52.4 37
-
1.8
5.0
0.0438
0.0459
0.0246
0.0243
Pd/Al2O3
calcined
3.0
acidity in Pd/Al2O3 used catalysts compared to that in Pd/MgO
used catalysts may be due to the carbonaceous deposits on the
former in more quantities than on the later. It is reported that
carbon coverage occurred during the aromatization reaction on
alumina drastically reduces its acidity.44
3.6510 0.1121
3.5740 0.1326
genolysis of CCl2F2. Wiersma et al.13 reported the presence of
heavy halogenated hydrocarbons in used Pd/C catalyst after the
hydrodechlorination of CCl2F2. It appears that the extent of
heavy halogenated hydrocarbons present in used Pd/MgO is
higher than that present in used Pd/Al2O3.
Kemnitz et al.26 reported MgF2 to have shown lower amounts
of NH3 desorption as compared to AlF3 and other Mg doped
AlF3 compounds. They have also shown MgF2 to possess a
sufficient amount of Lewis acid sites, which are different in
strength from the sites observed over AlF3. Differences in the
acidity of Pd/alumina and Pd/magnesia used catalysts observed
in this study vis-a`-vis to that reported by Kemnitz et al. may
arise out of the differences in the nature of catalyst, nature of
reaction conditions as well as reaction duration.
3.5. Acidity Results. During the hydrogenolysis of CCl2F2
on supported Pd catalysts, change is expected not only in the
structure of Pd (due to the formation of PdCx species), but also
in the support structure (due to the transformation of oxide
supports to halides or oxyhalides by the release of HF/HCl
during the reaction). Thus, there is every possibility of change
in the acid site distribution of the catalyst after the reaction.
This may affect the product distribution during time on stream
analysis of the reaction until the catalyst is fully stabilized in
the corrosive atmosphere of hydrogen halides. Knowledge of
the variation in the acid site population of catalysts before and
after reaction may throw some light in assessing the product
composition at steady-state conditions as well as at the start of
the reaction. Temperature programmed desorption (TPD) of NH3
is a well-known technique in finding out the acid site population
present in the catalyst.39 The acid site population of fresh and
used alumina and magnesia supported Pd catalysts are shown
in Table 1. All the catalysts show an increase in the number of
acid sites of all the strengths when hydrogenolysis reaction of
CCl2F2 is carried over them.
The extent of increase in the number of acid sites is more in
used Pd/MgO (dried) catalyst than in calcined one unlike in
used Pd/Al2O3 catalysts, where there is only a marginal increase
after the reaction. In Pd/MgO, the support is perhaps more
fluorinated compared to that of Pd/Al2O3, which might have
been partially fluorinated. This fact is further confirmed from
XRD and XPS results of the fresh and used catalysts. In dried
catalyst, the easier exchange of hydroxyl groups on magnesia
by F- is expected as magnesia gets transformed to brucite in
aqueous medium during the preparation step. Transformation
of periclase to brucite in aqueous medium is easy40 and also
the exchange of O2- and OH- on magnesia by Cl- is
documented.41 Thus, easier transformation of Mg(OH)2 to MgF2
in the presence of HF can be expected over Pd/MgO dried
catalysts getting highly fluorinated. The fluoride ligands sur-
rounding the metal cations are known to exhibit higher Lewis
acidity than oxides or hydroxides,42,43 thus hydroxo fluorides
can be expected to show lower Lewis strength than fluorides.
It is worth noting that strong acid sites on Pd/alumina fresh
and used are more than in the fresh and used Pd/MgO catalysts,
with the exception of Pd/MgO used dried catalyst. The lower
3.6. Activity Results. The product distributions on different
catalysts recorded at steady-state conditions are presented in
Table 3. The initial activity measurements are also discussed
in order to explain the formation of various other products during
the course of the reaction. The time on stream analysis of the
products is given in Figure 7. The steady-state is attained at
4-5 h of the reaction in the hydrogenolysis of CCl2F2 over all
the catalysts studied. From Figure 7, it can be observed that
the selectivity to CH4 decreased from ∼45% to ∼30% with time
over Pd/MgO catalysts. However, on Pd/Al2O3 catalysts, there
is an initial decrease in the first hours of reaction, and thereafter,
the selectivity to CH4 is stabilized at ∼40-50%. Formation of
CH4 occurs through defluorination of CCl2F2 to CCl2F (inter-
mediate species) and subsequently by dechlorination to yield
CH4. Although the selectivity to CH2F2 is about the same
(∼64%) on both dried and calcined Pd/MgO catalysts, the
conversion of CCl2F2 is higher (70%) in dried catalyst than that
(58%) on calcined one. However, it is different on γ-Al2O3
supported Pd catalysts. Although the conversion is higher on
calcined catalyst, it is much lower than those on MgO supported
Pd catalysts. The selectivity to CH2F2 is also lower (45-53%)
on alumina supported catalysts. Under the corrosive reaction
conditions where HF/HCl are formed during the course of
reaction, support transformation into respective metal oxy/
hydroxo fluorides or fluorides would be expected. Formation
of the main desired product CH2F2 is dependent on the
desorption rates of intermediate species :CF2 from catalyst
surface. It is also reported that support surfaces with Lewis acid
sites induce electron deficient/vacant sites on Pd surface.10 The
formation of byproduct CHClF2 (R-22) is observed only on
Al2O3 supported catalysts. CH3F (R-41) is formed over both
Pd/MgO and Pd/Al2O3 calcined catalysts. Although CHClF2
forms via the adsorbed [CClF2]* intermediate species, formation
of CH3F may occur via hydrodefluorination of [:CF2]* inter-
mediate to adsorbed [:CHF]* species and subsequent hydroge-
nation to yield CH3F.10 No halogen exchange products namely
CClF3 and CCl3F are observed in this study.

1030 J. Phys. Chem. B, Vol. 106, No. 5, 2002
Aytam et al.
Figure 7. Time-on-stream activity of CCl2F2 conversion in hydrogen over dried and calcined 6 wt %Pd/MgO and 6 wt %Pd/γ-Al2O3 catalysts. (a)
Pd/γ-Al2O3 (dried) (b) Pd/γ-Al2O3 (calcined) (c) Pd/MgO (dried) (d) Pd/MgO (calcined).
The surface compositions of the catalysts with respect to Pd/
Al and Pd/Mg ratios (Table 2) are obtained from the XPS
analysis of the fresh and used catalysts. It can be seen that Pd/
Mg ratios are more or less same in calcined and dried catalysts.
However, in Pd/Al2O3 catalysts the Pd/Al ratio is 3 times more
in the calcined catalyst than that in dried catalyst. The
transformation of catalyst surface occurs during the initial
activity through reaction of the catalysts with by-products of
reaction, HCl and HF. The XPS data of the used Pd/MgO dried
catalyst show that F/Mg ratio is >3, which is nearly three times
higher than that on Pd/Al2O3 used catalyst (F/Al ∼ 1.1). TPR
analysis of the used catalysts revealed that the amount of coke
deposition is more over Pd/MgO calcined than that on the dried
used catalyst. Pd/MgO dried catalyst which shows highest
increase in acidity upon reaction, has also exhibited highest
activity and good CH2F2 selectivity among the catalysts studied
probably on account of more electron deficient character of Pd
in this catalyst.
hydrogenolysis reaction) compared to that on Al2O3 supported
Pd catalysts. The nature of carbonaceous deposits occurred
during the reaction appears to be dissimilar on both the
supported catalysts. XRD analysis of the used catalysts indicated
the presence of AlF1.65(OH)1.35 xH2O phase over Pd/γ-Al2O3
used catalysts and MgF2 phase on Pd/MgO used catalysts due
to surface transformation during the course of reaction. XPS
data shows that Pd is present in different environments on Al2O3
and MgO supports, respectively. Surface Pd concentration is
more on MgO than on Al2O3. During the course of hydro-
genolysis reaction, there is more surface enrichment of fluoride
on MgO than on Al2O3 support. The relative enhancement in
the acidity upon hydrogenolysis reaction is highest on dried Pd/
MgO catalyst. Another probable reason for higher acidity on
Pd/MgO used catalyst may be that the extent of coking is higher
on Pd/Al2O3 used catalyst resulting lower acidity.
Among the catalysts studied, Pd/MgO, dried variety has
shown highest hydrodechlorination activity with high CH2F2
selectivity. Pd//γ-Al2O3 catalysts, though have shown different
activities, the selectivities to CH2F2 and CH4 are around 40-
50%.
4. Conclusions
MgO supported Pd catalysts have shown higher hydrogenoly-
sis activity for CCl2F2 and selectivity to CH2F2 (R-32) compared
to those on Al2O3 supported Pd catalysts. This behavior can be
attributed to higher dispersion of Pd on MgO surface than on
Al2O3 surface and also to the presence of higher amounts of
F- in MgO supported Pd catalysts (occurred during the
Acknowledgment. AHP thanks University Grants Commis-
sion, New Delhi, and AVG & JKM thank Council of Scientific
and Industrial Research-New Delhi for the award of fellow-
ships. The authors thank Dr. K. V. Raghavan, Director, I.I.C.T,
Hyderabad, for his keen interest in this work. Our grateful thanks

Pd/MgO and Pd/γ-Al2O3 Catalysts
J. Phys. Chem. B, Vol. 106, No. 5, 2002 1031
(22) Kulkarni, P. P.; Deshmukh, S. S.; Kovalchuk, V. I.; d′Itri, J. L.
Catal. Lett. 1999, 61, 161.
(23) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Hand-
Book of XPS; Perkin-Elmer: Eden Prairie, MN, 1992.
(24) Padmasri, A. H. Hydrodechlorination of CCl2F2 oVer Calcined
Hydrotalcite Supported Palladium Based Catalysts, Ph.D. Thesis, 2000,
Osmania University.
(25) Venugopal, A.; Rama Rao, K. S.; Sai Prasad, P. S.; Kanta Rao, P.
J. Chem. Soc. Chem. Commun. 1995, 2377.
(26) Kemnitz, E.; Hess, A.; Rother, G.; Troyanov, S. J. Catal. 1996,
159, 332.
(27) Fagherazzi, G.; Beneditti, A.; Polizzi, S.; Di Mario, A.; Pinna, F.;
Signoretto, M.; Pernicone, N. Catal. Lett. 1995, 32, 293.
(28) Stobbe-Kreemers, A. W.; Soede, M.; Makkee, M.; Scholten, J. J.
F. Appl. Catal. A: General 1995, 131, 367.
(29) Krishnan Kutty, N.; Vannice, M. A. J. Catal. 1995, 155, 312.
(30) Krishnan Kutty, N.; Vannice, M. A. Appl. Catal. A: General 1998,
173,
(31) Claus, P.; Berndt, H.; Mohr, C.; Radmik, J.; Shin, E.-J.; Keane,
are due to the reviewer for his constructive criticism and
suggestions for improving the quality of the manuscript.
References and Notes
(1) Ohnishi, R.; Wang, W.-L.; Ichikawa, M. Chem. Letts. 1991, 841.
(2) Coq, B.; Cognion, J. M.; Figueras, F.; Tournigant, D. J. Catal. 1993,
141, 21.
(3) Coq, B.; Hub, S.; Figueras, F.; Tournigant, D. Appl. Catal. A 1993,
101, 41.
(4) van de Sandt, E. J. A. X.; Wiersma, A.; Makkee, M.; van Bekkum,
H.; Moulijn, J. A. Recl. TraV. Chim. Pays-Bas 1996, 115, 505.
(5) Karpinski, Z.; Early, K.; d′Itri, J. L. J. Catal. 1996, 164, 378.
(6) Wiersma, A.; van de Sandt, E. J. A. X.; Makkee, M.; van Bekkum,
H.; Moulijn, J. A. Stud. Surf. Sci. Catal. 1996, 101, 369.
(7) Ueda, W.; Tomioka, S.; Morikawa, Y.; Sudo, M.; Ikawa, T. Chem.
Lett. 1990, 879.
(8) Ahn, B. S.; Lee, S. C.; Moon, D. J.; Lee, B. G. J. Mol. Catal. A
1996, 106, 83.
(9) Scott, S. P.; Sweetman, M.; Thomson, J.; Fitzgerald, A. G.;
Sturrock, E. J. J. Catal. 1997, 168, 501.
M. A. J. Catal. 2000, 192, 88.
(10) Coq, B.; Figueras, F.; Hub, S.; Tournigant, D. J. Phys. Chem. 1995,
99, 11 159.
(32) D′Ercole, A.; Giamello, E.; Pisani, C.; Ojamae, L. J. Phys. Chem.
B, 1999, 103, 3872.
(11) Juszczyk, W.; Malinowski, A.; Karpinski, Z. Appl. Catal. A:
General 1998, 166, 311.
(33) Chen, Y. Z.; Hwang, C. M.; Liaw, C. W. Appl. Catal. A: General
1998, 169, 207.
(12) van de Sandt, E. J. A. X.; Wiersma, A.; Makkee, M.; van Bekkum,
H.; Moulijn, J. A. Appl. Catal. A: General 1997, 155, 59.
(13) Wiersma, A.; van de Sandt, E. J. A. X.; den Hollander, M. A.; van
Bekkum, H.; Makkee, M.; Moulijn, J. A. J. Catal. 1998, 177, 29.
(14) Malinowski, A.; Juszczyk, W.; Pielaszek, J.; Bonarowska, M.;
Wojciechowska, M.; Karpinski, Z. J. Chem. Soc. Chem. Commun. 1999,
8, 685.
(15) Deshmukh, S.; d′Itri, J. L. Catal. Today, 1998, 40, 377.
(16) Ribeiro, F. H.; Gerken, C. A.; Somorjai, G. A.; Kellner, C. S.;
Coulston, G. W.; Manzer, L. E.; Abrams, L. J. Catal. 1998, 176, 352.
(17) van de Sandt, E. J. A. X.; Wiersma, A.; Makkee, M.; van Bekkum,
H.; Moulijn, J. A. Appl. Catal. A: General 1998, 173, 352.
(18) Ohnishi, R.; Wang, W.-L.; Ichikawa, M. Appl. Catal. A: General
1994, 113, 29.
(34) Itoh, T.; Kuramoto, M.; Yoshida, M.; Tokuda, T. J. Phys. Chem.
1983, 87, 4411.
(35) L′ Argentiere, P. C.; Figoli, N. S.; Arcoya, A.; Seoane, X. L. React.
Kinet. Catal. Lett. 1991, 43(2), 413.
(36) Okumura, K.; Tanaka, H.; Niwa, M. Catal. Lett. 1999, 58, 43.
(37) Shelef, M.; Haack, L. P.; Soltis, R. E.; de Vries, J. E.; Logothetis,
E. M. J. Catal. 1992, 137, 114.
(38) Hess, A.; Kemnitz, E.; Lippitz, A.; Unger, W. E. S.; Menz, D.-H.
J. Catal. 1994, 148, 270.
(39) Falconer, J. L.; Schwarz, J. A. Catal. ReV. Sci-Eng. 1983, 25(2),
141.
(40) Nageshwara Rao, V.; Sai Prasad, P. S.; Prasad, K. B. S.; Kanta
Rao, P. J. Chem. Soc. Chem. Commun. 1990, 278.
(41) Sales, E. A.; Bugli, G.; Ensuque, A.; Mendes, M. de J.; Bozon-
Verduraz, F. Phys. Chem. Chem. Phys. 1999, 1, 491.
(42) Hess, A.; Kemnitz, E. J. Catal. 1994, 149, 449.
(43) Peri, J. B. J. Phys. Chem. 1968, 72, 2917.
(19) Bonarowska, M.; Malinowski, A.; Juszczyk, W.; Karpinski, Z. Appl.
Catal. B: EnViron. 2001, 30, 187.
(20) Sai Prasad, P. S.; Lingaiah, N.; Chandrashekar, S.; Rama Rao, K.
S.; Kanta Rao, P.; Raghavan, K. V.; Berry, F. J.; Smart, L. E. Catal. Lett.
2000, 66, 201.
(21) Kim, S. Y.; Choi, H. C.; Yang, O. B.; Lee, K. H.; Lee, J. S.; Kim,
Y. G. J. Chem. Soc. Chem. Commun. 1995, 2169.
(44) Sai Prasad, P. S.; David Raju, B.; Rama Rao, K. S.; Salvapathy,
G. S.; Kanta Rao, P. Appl. Catal. A: General 1992, 83, 141.