Magnesium fluoride as a catalytic support in hydrodechlorination of CCl2F2 (CFC-12)

Article abstract of DOI:10.1039/a900731h

Magnesium fluoride-supported palladium and ruthenium appear to be useful catalysts in the reaction of CCl₂F₂ (CFC-12) hydrodechlorination; after doping Pd/MgF₂ with gold the selectivity for CH₂F₂ increased from ~ 70 to almost 90%.

Full text of DOI:10.1039/a900731h

Magnesium fluoride as a catalytic support in hydrodechlorination of CCl2F2
CFC-12)
(
a
a
a
a
b
a
A. Malinowski, W. Juszczyk, J. Pielaszek, M. Bonarowska, M. Wojciechowska and Z. Karpi n´ ski*
a
Institute of Physical Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, PL-01224 Warszawa, Poland.
E-mail: zk@ichf.edu.pl
Faculty of Chemistry, A. Mickiewicz University, Grunwaldzka 6, PL-60780 Pozna n´ , Poland
b
Received (in Liverpool, UK) 26th January 1999, Accepted 9th March 1999
2
21
Magnesium fluoride-supported palladium and ruthenium
appear to be useful catalysts in the reaction of CCl (CFC-
2) hydrodechlorination; after doping Pd/MgF with gold
the selectivity for CH increased from ~ 70 to almost
0%.
of MgF
average pore size 20 nm, micropore volume 0.22 cm g
showed a rutile structure. The reaction of CF Cl with
2
(BET surface area 43 m g , Barrett–Joyner–Halenda
3
21
2
F
2
)
1
2
2
2
2
F
2
dihydrogen was conducted in a glass flow system. The
experimental procedure and reaction conditions have been
reported elsewhere.11 Turnover frequencies (TOFs) were calcu-
lated on the basis of metal loadings and dispersions (from
chemisorption measurements) and are shown in Table 1.
9
The need to replace environmentally detrimental chloro-
fluorocarbons (CFCs) by benign hydrochlorofluorocarbons
(
HCFCs) and hydrofluorocarbons (HFCs) has resulted in
Our results indicate that MgF
catalysts exhibit interesting properties in CCl
dechlorination. The first positive point is that CHF
produced by these catalysts, as often happens on catalysts
capable of Cl/F exchange. Secondly, the time-on-stream
behaviour (Fig. 1) shows an initial increase in activity, which
appears to be stable during the next 17 hours. Thirdly, the
2
-supported palladium-based
1
considerable interest in catalytic hydrodechlorination of CFCs.
It is obvious from the patent literature that there is much activity
in the development of new catalysts capable of very selective
chlorine removal from a CFC molecule, i.e. avoiding undesired
hydrodefluorination. Although this can be achieved with the use
of heterogeneous catalysts containing such transition metals as
iridium, ruthenium and, especially, palladium, nevertheless the
formation of over-dehalogenated products still presents a major
concern. The role of a support (active carbon and AlF
reported most often in the patent literature) appears important;
however, for this class of reaction one must also consider
modifying the catalyst by doping the active phase (preferably
palladium) with other elements.5
F
2 2
hydro-
is not
3
2
selectivity towards CH
2
F
2
production is reasonably high
-supported
( > 70%), resembling the catalytic behaviour of AlF
3
7,8
3
4
7
3
are
palladium catalysts. It can be speculated as others have done,
that the Lewis acidity of MgF , as in the case of AlF , decreases
electron density in neighbouring Pd sites. In effect, electron-
deficient Pd species bind :CF carbene radicals (commonly
2
3
2
,6
2,7,8,12
accepted intermediate species
) less strongly, so a hydro-
This report presents preliminary results obtained for MgF
supported Pd and Ru catalysts. Several metal fluorides have
exhibited interesting properties as carriers for palladium (AlF
2
-
genative desorption of these radicals distinctly prevails over
3
,
4,7,8
4
ZrF , TiF
3
2
), but MgF has not been tested yet. In addition,
9
MgO, as a support in hydrodechlorination, is not resistant to
the highly corrosive conditions of this reaction which results
from considerable evolution of HCl, which can transform the
support into MgCl
behaviour of alumina-supported catalysts, due to inevitable
albeit undesired) formation of HF during CFCs hydro-
2
. Similarly, by analogy to the catalytic
7,8
(
dehalogenation, MgO is (at least partly) transformed into
magnesium fluoride, which when located at a metal/support
interface might favorably modify the electronic state of the
metal particles.7 All these facts and speculations have
encouraged us to investigate this area.
,8
The MgF
were prepared by impregnation of MgF
reaction of MgCO with aqueous solution of HF ) with
appropriate metal salts. X-Ray diffraction (XRD) examination
2
-supported Pd, Pd–Au, Ru and Ru–Au catalysts
2
(obtained in the
2
Fig. 1 Time-on-stream behaviour of 2 wt% Pd/MgF (0.121 g, circles) and
10
3
Pd–Au/MgF (0.201 g, squares) catalysts in CCl F hydrodechlorination.
2
2 2
4 2 2
(a) Overall conversion, (b) selectivities to CH and CH F .
Table 1 Hydrodechlorination of CCl
F
2 2
on Pd/MgF
2
, Pd–Au/MgF
2
, Ru/MgF
2
and Ru–Au/MgF
2
catalysts: steady-state turnover frequencies (TOFs), product
selectivities and apparent activation energies (E
a
2 2 2
s). Reaction temperature 453 K, H –CCl F ratio 10:1.
a
Selectivity (%)
1
E
a
/kJ mol21
Catalyst
Pd/MgF
CH
4
CH
2
F
2
CClF
3
CHClF
2
C
2
H
6
C
2
H F
4 2
TOF/s2
b
2
19.0
9.3
67.5
56.4
b
72.0
86.0
7.0
—
0.4
—
8.5
3.8
19.2
21.7
0.2
0.1
3.8
3.2
—
—
2.5
1.9
0.0765
0.0820
0.0053
0.0160
61.0 ± 0.4
61.8 ± 1.0
60.3 ± 0.3
54.1 ± 2.5
b
Pd–Au/MgF
2
b
Ru/MgF
Ru–Au/MgF
Minor products (e.g. CH
2
b
2
12.9
3.4
a
3
Cl, S% < 0.5%). 2 wt% Pd/MgF
2
(metal dispersion, H/Pd = 0.02
3
); 3.5 wt% Pd0.7Au0.3/MgF
). Subscripts in chemical formulae denote atomic fractions.
2
(H/Pd = 0.02); 2.4 wt% Ru/MgF
2
(
CO/Ru = 0.20); 2 wt% Ru0.75Au0.25/MgF
2
(CO/Ru = 0.03
2
Chem. Commun., 1999, 685–686
685

hydrodefluorination. Interestingly enough, the selectivity for
CH increases with Au doping: from ~ 72% for the Pd/MgF
to 86% for the Pd–Au/MgF (Fig. 1 and Table 1).
As expected, preliminary XRD data manifest a substantial
degree of Pd–Au mixing in the MgF -supported catalyst after
selectivity variations. Studies are in progress to prepare new
Pd–Au and Ru–Au catalysts (with different gold contents),
supported on MgF and active carbon.
2
F
2 2
2
2
2
reduction. Two Pd–Au solid solutions are present: Pd0.81Au0.19
and Pd0.4Au0.6 (indexes denote atomic fractions). This sig-
nificant alloying concerns both reduced as well as spent Pd–Au/
Footnotes and references
1 Z. Ainbinder, L. E. Manzer and M. J. Nappa, in Handbook of
Heterogeneous Catalysis, ed. G. Ertl, H. Knözinger and J. Weitkamp,
VCH, Weinheim, 1997, vol. 4, p. 1677.
2
MgF catalyst. XRD profiles obtained with used Pd–Au catalyst
are somewhat complicated because large amounts of carbon
2
3
4
5
A. Wiersma, E. J. A. X. van de Sandt, M. A. Hollander, H. van Bekkum,
M. Makkee and J. A. Moulijn, J. Catal., 1998, 177, 29.
For example, J. I. Darragh, UK Patent 1980, 1578933; V. N. M. Rao, US
Patent 1992, 5136113.
(
originating from CCl
2
F
2
) are incorporated into the Pd-based
11,12
phase.
Since a Pd–C solution has a lattice parameter
(a
PdC0.13 = 0.399 nm1 ) larger than Pd (aPd = 0.3890 nm), it
3,14
is difficult to assess whether an increase in the lattice parameter
of Pd is due to Au (aAu = 0.4078 nm) or C incorporation. In
order to estimate to what extent Pd diffraction lines are affected
by carbon dissolution and alloying with Au further studies are
underway in our laboratory. This issue seems important,
because, as is expected, a close contact between Pd and Au is
For example, S. C. Kellner and V. N. M. Rao, US Patent 1989,
4
873381.
R. Onishi, I. Suzuki and M. Ichikawa, Chem. Lett., 1991, 841; R. Onishi,
W.-L. Wang and M. Ichikawa, Appl. Catal. A: Gen., 1994, 113, 29.
6 V. N. M. Rao, US Patent 1995, 5447896; S. Morikawa, S. Samejima, M.
Yositake and S. Tatematsu, Eur. Patent 1989, 0 347830 A2.
7
8
9
B. Coq, J.-M. Cognion, F. Figuéras and D. Tournigant, J. Catal., 1993,
41, 21.
B. Coq, F. Figuéras, S. Hub and D. Tournigant, J. Phys. Chem., 1995,
9, 11159.
essential for obtaining higher selectivity towards CH
Pd–Ag/graphite catalyst tested in CFC-12 hydrodechlorination
by Coq et al.15 showed selectivity for CH
to be similar to that
2 2
F . The
1
2 2
F
9
of the Pd/graphite catalyst. Such a result can be interpreted by
an apparent absence of Pd–Ag interaction in the catalyst and,
indeed, their XRD study of Pd–Ag/graphite showed no
formation of Pd–Ag solid solution.
H. C. Choi, S. H. Choi, O. B. Yang, J. S. Lee, K. H. Lee and Y. G. Kim,
J. Catal., 1996, 161, 790.
1
0 J. Haber and M. Wojciechowska, J. Catal., 1988, 110, 23.
11 W. Juszczyk, A. Malinowski and Z. Karpi n´ ski, Appl. Catal. A: Gen.,
1998, 166, 311.
12 E. J. A. X. van de Sandt, A. Wiersma, M. Makkee, H. van Bekkum and
J. A. Moulijn, Catal. Today, 1997, 35, 163.
The Ru/MgF
CH (Table 1), but the selectivity to CHClF
in the case of Pd/MgF , in a qualitative agreement with
Wiersma et al.2 Again, doping with gold increased the
selectivity towards partial hydrodechlorination (CH
CHClF ) at the cost of CH formation.
In conclusion, we have shown fair to respectable selectivities
towards partial hydrodehalogenation of CCl over Pd, Pd–
Au, Ru and Ru–Au catalysts supported on MgF . Our results
indicate that the presence of gold in the catalyst is beneficial for
2
catalyst showed rather low selectivity to
2
F
2
2
was higher than
2
13 J. Stachurski and A. Fr a˜ ckiewicz, J. Less-Common Metals, 1985, 108,
2
49.
2 2
F +
1
1
4 S. B. Ziemecki, G. A. Jones, D. G. Swartzfager, R. L. Harlow and J.
Faber, Jr., J. Am. Chem. Soc., 1985, 107, 4547.
5 B. Coq, S. Hub, F. Figuéras and D. Tournigant, Appl. Catal. A: Gen.,
2
4
2 2
F
1
993, 101, 41.
2
Communication 9/00731H
686
Chem. Commun., 1999, 685–686