Co cation effects on activity and stability of isolated Pd(II) cations in zeolite matrices for selective catalytic reduction of nitric oxide with methane

Article abstract of DOI:10.1016/S0021-9517(02)93722-X

Zeolite-supported Pd catalysts perform well for NO reduction by using methane as a reducing agent. The catalytic properties of palladium ion-exchanged zeolite catalysts toward selective reduction of NO with methane (CH₄-SCR) were studied in the presence of water vapor at 500°C. Pd/HZSM-5 with a Si/Al ratio of 19.8:1 deactivated under the reaction conditions. The durability was slightly enhanced when the ratio decreased, but the NO conversion gradually decreased. Mordenite-supported palladium demonstrated stability in NO conversion during the durability test reaction. Co addition on Pd/HZSM-5 enhanced the catalytic activity and no deactivation was found on the sample loaded with 3.3 wt % cobalt. An isolated Pd²⁺ on HZSM-5, which was the active and selective site for CH₄-SCR, was not eluted by NaCl titration after adding 3.3 wt % cobalt. NH₄NO₃ titration indicated that the isolated Pd²⁺ existed on the Pd-Co(3)/HZSM-5. Titration also showed that some portion of cobalt was located as an isolated Co²⁺ in the sample. The effect of the isolated Co²⁺ on the stability of the isolated Pd²⁺ was detected not only on ZSM-5 with a different Si/Al ratio but also on zeolites with different size cavities, such as MOR.

Full text of DOI:10.1016/S0021-9517(02)93722-X

Journal of Catalysis 211, 75–84 (2002)
doi:10.1006/jcat.2002.3722
Co Cation Effects on Activity and Stability of Isolated Pd(II)
Cations in Zeolite Matrices for Selective Catalytic
Reduction of Nitric Oxide with Methane
Masaru Ogura,1,2 Susumu Kage, Tomoko Shimojo, Junko Oba, Masayoshi Hayashi,
Masahiko Matsukata, and Eiichi Kikuchi
Department of Applied Chemistry, Waseda University, 3-4-1 Okubo, Shinjuku, Tokyo 169-8555, Japan
Received December 18, 2001; revised April 17, 2002; accepted June 12, 2002
mal stability in the 1960s and of Zn(II) and Ga(III) for aro-
Palladium ion-exchanged zeolites (ZSM-5, mordenite, ferrierite, matization in the 1970s showed additional ways to use zeo-
Y, beta) were tested to evaluate catalytic activity for selective re- lite catalysts in the petrochemical industry. Many reactions
duction of nitric oxide with methane (CH4-SCR) and stability in
have achieved high yields by the use of zeolite-supported
metal cation catalysts. A recent development by Panov and
co-workers regarding the direct synthesis of phenol from
benzene using N2O, catalyzed by Fe cations on zeolites (1),
the presence of 10% water vapor at a relatively high temperature
◦
of 500 C. Deactivation of the catalytic activity was inevitable on
Pd/HZSM-5 with a Si/Al ratio of 19.8 under the reaction condi-
tions. When the Si/Al ratio decreased, the durability was slightly
clarified the need for more knowledge about the catalytic
enhanced. Among the zeolites tested, palladium supported on mor-
denite showed stable NO conversions. The addition of cobalt effec-
tively stabilized the activity of Pd–H-zeolites, particularly with the
properties of metal cations occluded in zeolite cavities. An-
other well-known reaction catalyzed by metal cations on
zeolites is selective catalytic reduction of nitric oxide by the
addition of 3.3 wt% cobalt onto Pd/HZSM-5, of which no deacti-
2
+
vation was noticeable for at least 50 h. Isolated Pd cations on use of hydrocarbons (HC-SCR). To date, Cu (2), Co (3),
HZSM-5 were the active and selective sites for CH4-SCR. The iso- Ga (4), In (4), and Fe (5) incorporated in zeolite matrices
2
+
+
lated Pd cations could not be exchanged by Na cations in NaCl as cations have been found to promote this reaction.
titration test after the addition of 3.3 wt% cobalt, while NH4NO3
titration revealed that the isolated Pd existed in the sample. This
Among numerous metal cations on zeolites or metal ox-
2
+
ides that catalyze HC-SCR, zeolite-supported palladium
catalysts have garnered much attention (6–14) since they
show a high performance for NO reduction by the use of
methane as a reducing agent. Palladium on zeolites was first
reported by Nishizaka and Misono to show high catalytic
activity for CH4-SCR (6). Loughran and Resasco (7) noted
2
+
means that Co changes the ion-exchange property of the iso-
2
+
lated Pd , leading to a much higher stability than Pd/HZSM-5
2
+
on the stability
with regard to durability. The effect of Co
of the isolated Pd2 cations and their sitings in zeolite cavities
+
is discussed. ꢀc 2002 Elsevier Science (USA)
Key Words: CH4-SCR; Pd/HZSM-5; stability; Pd–Co/HZSM-5,
2
+
ion-exchange property; isolated Pd ; NaCl titration; NH4NO3 that palladium sites and acid sites are preferably intimate
titration.
with each other, and that the acidity is needed to transform
palladium into a highly dispersed state such as an isolated
2
+
Pd while the palladium on a nonacidic support tends to
form a PdO cluster (8). With respect to the redispersion of
the palladium species, Adelman and Sachtler reported that
the exposure of a reactant mixture of NO2–CH4–O2 pro-
INTRODUCTION
Zeolites have prevailed in various catalytic systems since
the discovery of a method to produce them synthetically,
and the significant amount of subsequent evolution has at-
tracted interest in their structural and catalytic properties.
Most zeolite utilization is in their protonic forms as acid
catalysts and occurs in the field of petroleum chemistry.
Also, ion exchanges of rare-earth elements for hydrother-
2
+
vides the reconstruction of the isolated Pd species (9).
They clearly showed the effect of the acid sites in zeolite
on the creation of the active palladium species as follows:
+
2+
PdO + 2H → Pd + H2O. The stoichiometry for the for-
mation of active sites on Pd-zeolites was also indicated by
Bell and co-workers (10), who deduced from the studies of
infrared spectroscopy that the palladium active species are
stabilized at the pair sites of the next nearest neighboring Al
1
Current address: Department of Chemical System Engineering, Grad-
uate School of Engineering, The University of Tokyo, 7-3-1 Hongo,
Bunkyo-ku, Tokyo 113-8656, Japan.
−
+
+
−
of zeolite, denoted by Z H (PdO)H Z . Later, they con-
firmed in their computational studies how the next near-
est neighboring Al sites facilitate the stabilization of the
2
To whom correspondence should be addressed. Fax: +81-(0)3-
5
8417771. E-mail: ogura@chemsys.t.u-tokyo.ac.jp.
75
0
021-9517/02 $35.00
c 2002 Elsevier Science (USA)
All rights reserved.
ꢀ

7
6
OGURA ET AL.
palladium species (15). Previously, we established that the
palladium species on HZSM-5 are located in an isolated
EXPERIMENTAL
Catalysts Preparation
The zeolites used in this work are summarized in
state, that is, the isolated Pd2 , as determined by “NaCl
+
titration” (12).
More than one decade has passed since the first report on
HC-SCR catalyzed by Cu cations on ZSM-5 was published. Table 1. ZSM-5 with Si/Al ratios 11.9, 19.8, and 35 were
Unfortunately, however, this reaction has not yet been re- supplied by Tosoh Co. Mordenite (MOR) with Si/Al ratio
alized in practice (16). Zeolite-based catalysts, which are 7.5 and ferrierite (FER) with Si/Al ratio 8.5 were manu-
knownbytheirhighactivitiesandselectivities, cannotmain- factured by Catalysis & Chemicals, Inc., and Tosoh Co.,
tain their activities under practical conditions and a sta- respectively. Faujasite Y with Si/Al ratio 2.8 was supplied
ble catalytic activity for a more prolonged period than the by Catalysis & Chemicals, Inc. Zeolite beta (BEA) with
conventional experimental conditions is indispensable for Si/Al ratio 15 was synthesized by a steam-assisted crystal-
practical application. The zeolite-based palladium catalyst lization method, which is described in detail elsewhere (19).
is well known to show high activity for CH4-SCR, even in Palladium catalysts supported on those zeolites were pre-
the presence of water vapor (17), while it shows low dura- pared according to the description in our previous work
+
bility against H2O (11, 18). One of the most serious causes (12). First of all, an NH -form zeolite was prepared by ion
4
for the low durability is the agglomeration of palladium exchange with 1 M NH4NO3 aqueous solution. Then, typi-
cations into bulky PdO particles outside of the zeolite pore cally, NH4-ZSM-5 was ion exchanged by stirring in a 0.1 M
structure, not only by hydrothermal treatment (11) but also Pd(NH3)4Cl2 solution at room temperature. The chemical
even in use under the reaction conditions (18).
composition of catalysts was determined by means of induc-
In our recent work, we observed a pronounced effect tively coupled plasma (ICP) atomic emission spectroscopy.
of cobalt species on the durability of Pd–Co/HZSM-5 for The amount of palladium on zeolite was typically fixed at
CH4-SCR in the presence of H2O (18). We evaluated the 0.4 wt%. When the palladium loading was changed, the
effect of second components on the catalytic performance abbreviation as Pd(content: wt%)/H-zeolite was used. A
of Pd/HZSM-5 zeolite with a durability test of NO reduc- palladium catalyst, in which 0.4 wt% palladium was sup-
◦
tion with CH4 at a relatively high temperature of 500 C in ported on silica, was also prepared by impregnation of
◦
the presence of water vapor for a prolonged period, and SiO2 in a 0.1 M Pd(NO3)3 solution at 80 C. All the cata-
among the second components tested, Co, Rh, Ag, Ce, and lysts were dried at 110 C overnight, followed by calcina-
Fe introduced individually to the Pd/HZSM-5 enhanced the tion at 500 C in 20% oxygen-containing flow prior to reac-
◦
◦
durability. In particular the addition of 3.3 wt% cobalt led to tion. After the calcination, the catalyst was cooled to the
astable NOconversionfor more than 40 h in the presence of temperature where the catalytic activity test would be car-
H2O. Here we discuss the roles of the cobalt species on the ried out. Cobalt onto Pd/HZSM-5 were introduced indi-
stable performance of Pd–Co/HZSM-5 while making ref- vidually by ion exchanging in a 1 M cobalt acetate solu-
◦
erence to the cations sited in some types of zeolite cavities. tion at 80 C. The loadings of cobalt were at 1, 2.2, 3.3, and
TABLE 1
Zeolite Catalysts Used
Si/Al
Pd/Al
Co/Al
Catalyst
Pd/HZSM-5
Zeolite type molar ratio Pd (wt%) molar ratio Co (wt%) molar ratio
MFI
MFI
MFI
MFI
MFI
MFI
MFI
MFI
MFI
MFI
MFI
mordenite
mordenite
ferrierite
faujasite Y
beta
19.8
19.8
19.8
19.8
19.8
19.8
19.8
19.8
11.9
11.9
35.0
7.5
0.40
1.0
2.0
0.05
0.12
0.24
—
0
0
0
—
—
—
0.72
0.21
0.48
0.72
1.12
—
Pd(1)/HZSM-5
Pd(2)/HZSM-5
Co(3)/HZSM-5
Pd–Co(1)/HZSM-5
Pd–Co(2)/HZSM-5
Pd–Co(3)/HZSM-5
Pd–Co(5)/HZSM-5
Pd/HZSM-5(11.9)
Pd–Co(3)/HZSM-5(11.9)
Pd/HZSM-5(35)
Pd/HMOR
0
3.3
1.0
2.2
3.3
5.2
0
3.3
0
0
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.40
0.05
0.05
0.05
0.05
0.03
0.03
0.08
0.02
0.02
0.02
0.01
0.04
0.45
—
—
Pd–Co(3)/HMOR
Pd/HFER
Pd/HY
7.5
8.5
3.3
0
0.30
—
2.8
0
—
Pd/HBEA
15.0
0
—

STABILITY OF ISOLATED Pd(II) CATIONS
77
5
.2 wt% on H-zeolites, followed by ion exchange of palla- was used for the measurements. The sample was pressed
dium. The content of cobalt loaded on zeolites is denoted in into a thin disk, set into the XPS sample holder, and out-
−
8
this paper as Pd–Co(content: wt%)/H-zeolite, for example, gassed below 1.0 × 10 Pa before each measurement. The
Pd–Co(3)/HZSM-5 (see Table 1).
2p level of silicon, included in every sample, was taken as
the internal reference.
Catalytic Activity Test
27Al-MAS NMR Studies
Reduction of NO was carried out in a fixed-bed flow
reactor, by feeding a mixture of 100 ppm NO, 2000 ppm
Al coordination states on the zeolite samples were con-
2
7
CH4, 10% O2, and 0 or 10% H2O in He balance at a rate firmed by Al-MAS NMR, recorded on a JEOL GSX-400
3
−1
of 100 cm min to 0.1-g catalyst at the temperature of at the frequency of 104 MHz, pulse width of 4.2 µs, recycle
◦
4
00 or 500 C. Effluent gases were analyzed by means of time of 5 s, and scanning times of 200. Chemical shifts were
3+
gas chromatography and chemiluminescence NOx analy- referenced to an external standard of [Al(H2O)6]
.
sis. Catalytic activities were evaluated by the level of NO
conversion to N2. N2O was not detected during the activity
tests under the conditions used in this study.
RESULTS
Catalytic Performance of Pd/HZSM-5 and the Effect
of Cobalt on the Activity and Stability
NaCl and NH4NO3 Titration Methods
Aqueous solutions of NaCl and NH4NO3 were used as
Figure 1 shows the catalytic performance of Pd/HZSM-5
◦
the media for the ion exchange of palladium species on for CH -SCR at 400 and 500 C. Under a dry condition
4
2+
◦
zeolites. It was found that the isolated Pd could be quan- at 500 C, NO conversion on this catalyst was almost un-
+
titatively exchanged by Na cations in the NaCl solution changed over 20 h. It showed high catalytic activity and
◦
(
12). After the ion exchange, the amounts of cobalt and/or stability at 400 C even in the presence of H O. On the con-
2
palladium eluted into the filtrate were measured with ICP. trary, the catalytic activity decreased with time onstream at
◦
The amount of palladium cations from Pd(0.4)/HZSM-5 500 C under wet conditions, resulting in a very low level of
thus was determined 0.037 mmol/g-zeolite, which coincided NO conversion after 7 h. The decreased catalytic activity
well with thewholeamountofpalladium inthis catalyst. Be- could not be recovered at all by removal of H O from the
2
cause the exchangeable amount of palladium cations from reactant stream, whereas the level of CH conversion was
4
PdO(0.4)/SiO2 was negligibly small, it is obvious that the recovered to the level attained in the dry condition, mean-
nature of palladium species on HZSM-5 is quite different ing that the selectivity of CH toward NO reduction was
4
from that in PdO/SiO2 in terms of ion-exchange ability. This lowered by the deactivation.
technique is, therefore, useful to determine the amount of
the isolated Pd in Pd/HZSM-5 catalysts, and we will here- tive effects of cobalt on the catalytic activity of Pd/HZSM-5
We described in our previous reports (17, 18) the promo-
2+
after refer to it as “NaCl titration.”
2
and on its durability in the presence of H O. As shown
The amount of the isolated Pd2 was determined by in Fig. 2, Co-loaded Pd/HZSM-5 showed high stability in
+
the method of NaCl or NH4NO3 titration using a 0.02 M
NaCl or a 1 M NH4NO3 aqueous solution. After calcina-
◦
tion at 500 C or after the durability test reaction, 30 mg of
a palladium-containing sample was stirred in the solution
◦
at 80 C for 1 h, filtered, and the concentration of palla-
dium in the filtrate was determined by ICP. No additional
2
+
Pd was observed to be eluted even with a longer period
of ion exchange, except for Pd/HMOR. The mass balance
of palladium on/out of the catalyst was always confirmed,
as reported previously (12). The palladium eluted out of
the catalyst was sited as isolated Pd2 species in the ion-
exchange sites of the zeolite. The residual palladium on the
catalyst was precipitated species, that is, PdO.
+
X-Ray Photoelectron Spectroscopic Studies
The surface states of palladium species were studied by
means of X-ray photoelectron spectroscopy (XPS). Spec-
tra of Pd 3d were recorded with EscaLab 220-I (VG Instru-
FIG. 1. EffectofH2O onthecatalyticactivityofPd/HZSM-5forCH4-
SCR: NO, 100 ppm; CH4, 2000 ppm; O2, 10%; total flow rate, 100 cm /min
3
in the absence (ꢀ) or in the presence of 10% H2O (ꢁ, ꢂ); catalyst weight,
◦
◦
ments). Non-monochromated Mg Kα radiation (1254 eV) 0.1 g; reaction temperature, 400 C (ꢂ), 500 C (ꢀ, ꢁ).

7
8
OGURA ET AL.
2
+
the order in the ion exchanges of Co up to 3.3 wt% and
2
+
Pd never changed the catalytic activity or stability, that is,
Co(3)–Pd/HZSM-5andPd–Co(3)/HZSM-5showedsimilar
NO conversion over the prolonged period of reaction.
Qualitative and Quantitative Analyses of Pd(II) Species
on Pd/HZSM-5 and Pd–Co/HZSM-5
NaCl titration, that is, an ion exchange of an isolated
Pd2 in Pd/HZSM-5 by use of Na in the aqueous solution
+
+
of NaCl, was found to effectively determine the amount of
2
+
2+
isolated Pd cations (12). The amounts of the isolated Pd
thus determined before utilization in the durability test are
summarized in Table 2. As shown previously (12), palla-
dium species in Pd/HZSM-5 were located entirely in the
FIG. 2. Durability of (ꢁ) Pd/HZSM-5, (ꢃ) Co(3)/HZSM-5, and (ꢀ)
2
+
cavities of HZSM-5 zeolite as isolated Pd cations when
the palladium content was below 1 wt% and the Si/Al ra-
tio of HZSM-5 was 19.8. Pd/HZSM-5 used in this study
Pd–Co(3)/HZSM-5 for CH4-SCR under wet conditions: NO, 100 ppm;
3
CH4, 2000 ppm; O2, 10%; total flow rate, 100 cm /min in the presence of
◦
1
0% H2O; catalyst weight, 0.1 g; reaction temperature, 500 C.
also contained all of the palladium species as isolated Pd2
+
cations. The addition of cobalt up to 1 wt% did not affect
2
+
the durability test, while cobalt that was ion-exchanged on the number of isolated Pd cations. The number of isolated
2
+
HZSM-5 hardly showed NO conversion under the same Pd cations, however, decreased with an increase in cobalt
experimental conditions.
content for Pd–Co(2)/HZSM-5 and Pd–Co(3)/HZSM-5,
2
+
The stability effect strongly depended on the loading and almost no isolated Pd cations were exchanged by
+
amount of cobalt, a typical example of which is illus- Na in the NaCl titration from Pd–Co(3)/HZSM-5, even
◦
trated in Fig. 3. The initial activity for NO reduction in- after the prolonged period of titration for 24 h at 80 C.
creased by loading Co(1) and then decreased with increas-
Figure 4 shows the XPS spectra of Pd 3d for Pd/HZSM-5
ing cobalt content. We found that almost no palladium was andPd–Co/HZSM-5andforPdOonSiO2. Thevaluesofthe
loaded on the Pd–Co(5)/HZSM-5, suggesting that no ion- binding energy corresponding to the Pd 3d5/2 on each sam-
exchangeable sites remained in Co(5)/HZSM-5, of which ple are summarized in Table 3. The absence of a spectrum
the Co/Al ratio corresponded to 1.12. Deactivation of these in the Pd(0.4)/HZSM-5 sample indicates that palladium
catalysts was suppressed as the amount of cobalt increased, species on the sample are well dispersed and never located
and the activity of Pd–Co(3)/HZSM-5 changed little dur- as an aggregated state on the zeolite surface. A moderate
◦
ing the durability test over 50 h. It is interesting to note that spectrum was detected after reaction at 400 C in the pres-
ence of water vapor. Pd(1)/HZSM-5 also showed a spec-
trum with an intensity similar to that of Pd(0.4)/HZSM-5
after use in the reaction. The dispersed state of palla-
dium species at 0.4 wt% was slightly influenced by the
addition of 1 wt% cobalt on HZSM-5, and a spectrum could
TABLE 2
NaCl Titration of Pd/HZSM-5 and Pd–Co/HZSM-5
Pd2
+
Co
2+
Pd /Pd
2+
Co /Co
2+
a
molar ratiob
Catalyst
(mmol/g) (mmol/g) molar ratio
Pd(0.4)/HZSM-5
Pd(1)/HZSM-5
Pd(2)/HZSM-5
Pd–Co(1)/HZSM-5
Pd–Co(2)/HZSM-5
Pd–Co(3)/HZSM-5
0.037
0.082
0.082
0.032
0.004
0.002
—
—
—
0.011
0.042
0.083
0.98
0.87
0.44
0.85
0.11
0.04
—
—
—
0.06
0.11
0.15
FIG. 3. Effect of cobalt on the catalytic performance of Pd/HZSM-5
for CH4-SCR under durability test conditions: (ꢁ) Pd/HZSM-5, (ꢄ) Pd–
Co(1)/HZSM-5, (ꢂ) Pd–Co(2)/HZSM-5, (ꢀ) Pd–Co(3)/HZSM-5, (ꢅ)
Pd–Co(5)/HZSM-5; NO, 100 ppm; CH4, 2000 ppm; O2, 10%; total flow
a
2+
The value is the molar ratio of Pd eluted into NaCl solution per
total Pd in the catalyst.
3
b
2+
rate, 100 cm /min in the presence of 10% H2O; catalyst weight, 0.1 g;
The value is the molar ratio of Co eluted into NaCl solution per
total Co in the catalyst.
◦
reaction temperature, 500 C.

STABILITY OF ISOLATED Pd(II) CATIONS
79
TABLE 3
X-Ray Photoelectron Spectra of Pd 3d5/2 on Pd/HZSM-5,
Pd–Co/HZSM-5, and PdO/SiO2
Catalyst
Binding energy of Pd 3d5/2 (eV)
Pd(0.4)/HZSM-5
After reactiona
—
337.3
Pd(2)/HZSM-5
337.1
336.4
Pd–Co(1)/HZSM-5
Pd–Co(3)/HZSM-5
After reactionb
338.3
338.5
338.1
PdO(Pd : 0.4)/SiO2
336.4
336.3
After reactiona
334.4
a
◦
NO–CH4–O2 reaction in the absence of H2O at 400 C.
b
◦
NO–CH4–O2 reaction in the presence of H2O at 500 C.
Palladium has covalent bonds with oxygen atoms in the ox-
ide, so that a higher level of binding energy corresponds to
a more cationic state of palladium; therefore, the divalent
palladium species on the zeolite maintains a cationic state
2
+
like that of the isolated Pd even on Pd–Co(3)/HZSM-5. It
should be noted that the intense peaks were obtained only
after calcination and that the divalent palladium species on
Pd–Co(3)/NH4ZSM-5, before calcination, were similar to
those on Pd/HZSM-5.
The palladium species in Pd–Co(3)/HZSM-5 were also
2
+
found to be isolated Pd by means of the ion exchange with
+
NH called “NH4NO3 titration” in this work. The titration
4
results are summarized in Table 4. First, the exchangeabili-
+
+
ties of palladium species with Na and NH were compared
4
by the use of aqueous NaCl and NH4NO3 solutions, respec-
tively. Pd(2)/HZSM-5 was used as a reference, since palla-
dium species on this sample were confirmed by NaCl titra-
2
+
tiontoexistinbothstatesasisolatedPd andaprecipitated
2
+
PdO, as shown in Table 2. The amounts of the isolated Pd
+
+
exchanged by Na and NH cations are comparable, and,
4
2+
therefore, it is confirmed that only the isolated Pd cations
on ion-exchangeable sites in zeolite are eluted by means of
these titration methods. As obviously shown in Table 4,
almost all the palladium species in Pd(0.4)–Co/HZSM-5
+
catalyst could be exchanged by NH4 cations, irrespective
of cobalt content. This means that Pd(0.4)–Co/HZSM-5
FIG. 4. XP spectra of Pd 3d on Pd/HZSM-5, Pd–Co/HZSM-5, and
PdO/SiO2.
TABLE 4
be detected. In contrast, intense peaks were detected on
Pd(2)/HZSM-5 and Pd–Co(3)/HZSM-5, indicating that di-
valent palladium species on the sample were concentrated
on the surface or near the pore mouth of ZSM-5 zeolite. We
NH NO Titration of Pd/HZSM-5 and Pd–Co/HZSM-5
4
3
Pd2
+
Co
2+
Pd /Pd
2+
2+
Co /Co
Catalyst
(mmol/g) (mmol/g) molar ratio molar ratio
deduced previously (17) that excess loading of cobalt leads Pd(0.4)/HZSM-5
0.033
0.068
0.032
0.034
0.037
—
—
0.011
0.159
0.356
0.88
0.36
0.85
0.91
0.99
—
—
0.06
0.43
0.64
to the agglomeration of the divalent palladium species on Pd(2)/HZSM-5
Pd–Co(1)/HZSM-5
the outer surface of zeolite. Also, PdO/SiO2 had surface di-
Pd–Co(2)/HZSM-5
valent palladium species, although the binding energy of the
Pd–Co(3)/HZSM-5
divalent palladium species differed from that of the zeolites.

8
0
OGURA ET AL.
contains every palladium as the isolated Pd2 cation in
+
the matrix of Co/HZSM-5. Comparing Tables 2 and 4 also
+
shows that cobalt species were rarely exchanged with Na ,
+
2+
while NH was able to exchange with isolated Co cations
in zeolite. The amount of Co revealed by NH4NO3 titra-
tion was 0.64 Co /Co in Pd–Co(3)/HZSM-5. This means
4
2+
2+
2
+
Co /Al = 0.46, indicating that the cobalt species exists as
2
+
isolated Co cations in the ion-exchange sites of zeolite.
The residual cobalt species were considered to be precipi-
tated cobalt oxide.
To understand the role of cobalt species on the stabiliza-
2+
tion of the isolated Pd active for CH4-SCR, the NH4NO3
titration was carried out on the samples after the durabil-
2
+
ity test. Table 5 compares the amounts of isolated Pd and
2
+
Co sustainableafterthereaction, asshowninFig. 3. Deac-
FIG. 5. Effect of zeolite type on the catalytic activity and durabil-
ity of isolated Pd for CH4-SCR: (ꢀ) Pd/HZSM-5, (ꢄ) Pd/HMOR,
tivated catalysts such as Pd/HZSM-5, Pd–Co(1)/HZSM-5,
2+
and Pd–Co(2)/HZSM-5 held fewer isolated Pd2 cations (ꢅ) Pd/HFER, (ꢂ) Pd/HBEA, (ꢁ) Pd/HY, (ꢃ) Pd–Co(3)/HMOR; NO,
+
3
1
00 ppm; CH4, 2000 ppm; O2, 10%; total flow rate, 100 cm /min in the
than the fresh ones. Interestingly, the amount of isolated
◦
2
+
presence of 10% H2O; catalyst weight, 0.1 g; reaction temperature, 500 C.
Co cations also decreased in those samples after the dura-
bility test. In contrast, Pd–Co(3)/HZSM-5 possessed the
2+
2+
isolated Pd as well as Co in amounts similar to that of
the fresh catalyst.
activity for CH4-SCR, even after 10 min of reaction, while
only Pd/HMOR catalyzed this reaction. Interestingly, the
catalytic activity of Pd/HMOR increased during the initial
period of the activity test, and after 10 h a stable NO conver-
sion was given. Ohtsuka and Tabata reported that Pd/MOR
Palladium (0.4 wt%) was loaded onto Co/HZSM-5 which
had been prepared after ion exchange of Co(3)/HZSM-5
with NH4NO3, and the subsequently prepared Pd–Co/
HZSM-5 was tested in the durability test. The activity of
this catalyst was as high as that of Pd–Co(3)/HZSM-5 in
the initial period of reaction, while the activity decreased
much faster than that of the Pd–Co(3)/HZSM-5. These
◦
showed a higher stability at 450 C than Pd/ZSM-5, but the
conversion gradually decreased in their catalytic test in the
◦
presence of water vapor (14). In our observation at 500 C,
however, no decline in the conversion was seen. It should
be noted that a longer time of 3 h was taken to titrate the
2+
results suggest that isolated Co cations removed from
+
2+
Co(3)/HZSM-5 by NH4 before loading of Pd play an
2+
isolated Pd completely in Pd/HMOR (see Table 6); on
2
+
important role in the stabilization of the isolated Pd on
ZSM-5.
◦
the other samples 1 h at 80 C was enough to achieve ion-
exchange equilibrium of Na in zeolites.
+
The catalytic performance of Co-loaded Pd/HMOR was
also included in Fig. 5. Interestingly, a higher NO conver-
sion than that for Pd/HMOR was kept stable from the be-
Effect of Zeolite Type on the Activity and Stability
of Pd(II) Species
Figure 5 compares the catalytic activities of palladium ginning of the durability test up to 20 h. As in the case
supported on HMOR, HFER, HY, and HBEA zeolites, and of Pd–Co(3)/HZSM-5, the amount of the isolated Pd2 on
+
Table 6 summarizes the amounts of isolated Pd2 cations HMOR, determined by NaCl titration, decreased with the
+
in these Pd-zeolites, determined by the NaCl titration. Un- addition of cobalt (see Table 6) and the NH NO titra-
4
3
der the conditions used in the durability test, palladium tion revealed that the palladium species was entirely iso-
2
+
supported on HFER, HY, and HBEA showed no catalytic lated Pd . It seems that the stabilization effect of the
TABLE 5
Pd2 /Pd Determined by NaCl and NH4NO3 Titrations for Pd/HZSM-5 and Pd–Co/HZSM-5:
+
Effect of Reaction
NaCl titration
NH4NO3 titration
Catalyst
Before reaction
After reaction
Before reaction
After reaction
Pd/HZSM-5
0.98
0.85
0.11
0.04
0.24
0.52
0.54
0.01
0.88
0.85
0.91
0.99
0.68
0.76
0.52
1
Pd–Co(1)/HZSM-5
Pd–Co(2)/HZSM-5
Pd–Co(3)/HZSM-5

STABILITY OF ISOLATED Pd(II) CATIONS
81
TABLE 6
Pd2 /Pd Determined by NaCl Titration
+
for Pd/H-zeolites and Pd–Co/HMOR
Pd2 /Pd molar ratio
+
Catalyst
Pd/HMOR
Pd/HFER
Pd/HY
0.59 (1)a
0.50
0.81
Pd/Hbeta
Pd/Co(3)/HMOR
0.40
0.12 (1)b
a
2+
2+
Pd /Pd became 1 after titration for 3 h.
b
Pd /Pd became 1 by NH4NO3 titration for 1 h.
isolated Pd2 by isolated Co can work in different zeolite
cavities.
+
2+
FIG. 6. Effect of Si/Al ratio in ZSM-5 for the catalytic activity and
durability of isolated Pd2 for CH4-SCR: Si/Al ratio: (ꢀ) 11.9, (ꢄ) 19.8,
(
1
+
ꢂ) 35; (ꢁ) Pd–Co(3)/HZSM-5 (11.9); NO, 100 ppm; CH4, 2000 ppm; O2,
Pd/HFER and Pd/HBEA were also inactive for CH4
combustion. All of the catalysts used in this study contained
3
0%; total flow rate, 100 cm /min in the presence of 10% H2O; catalyst
◦
weight, 0.1 g; reaction temperature, 500 C.
2+
isolated Pd , as confirmed by NaCl titration, and seen in
Table 6, along with precipitated PdO. Palladium supported
on zeolite Y has been reported to be inactive for CH4-SCR
by many researchers. Also, in our experiments, Pd/HY was
inactive for the reaction even though it contained isolated
was not found in either of the fresh catalysts. Although
not so much of the extra-framework Al could be ob-
served on Pd/HZSM-5 after the test, none was observed on
Co(3)/Pd/HZSM-5. This indicates that Si/Al ratio is hardly
changed during the reaction, and that the irreversible loss
2
+
Pd , but on the other hand, it was active for CH4 com-
bustion. From these results, it appears that the activities of
the isolated Pd2 and precipitated PdO for CH4-SCR and
CH4 combustion are affected by their location surrounded
by zeolitic aluminosilicate framework.
+
Effect of Si/Al Ratio in ZSM-5 Framework
on the Stability of Pd(II) Species
The catalytic activities of palladium supported on
HZSM-5 with different Si/Al ratios in the framework were
evaluated by the durability test of CH4-SCR, as illustrated
in Fig. 6. The activity and selectivity against combustion
(
not shown) increased with the decrease in Si/Al ratio in
2
+
the ZSM-5 framework. Also, the stability of isolated Pd
increased with decreasing Si/Al ratio. The catalytic perfor-
mance of Co-loaded Pd/HZSM-5 with a Si/Al ratio of 11.9
was also included in this figure. A higher and more stable
NO conversion than Pd/HZSM-5(11.9) up to 20 h was given
by the addition of Co(3). In this catalyst, also, the amount
2+
of the isolated Pd decreased in response to cobalt addi-
+
tion, and NH titration revealed that the palladium species
4
2+
was entirely the isolated Pd (results are not shown). From
these results, we concluded that not only the amount of the
isolated Pd2 cations but also their stability correlates with
+
the Si/Al ratio of ZSM-5 framework. In contrast, the stabi-
2+
lization effect of isolated Co on the catalytic activity of
the isolated Pd2 was also observed in ZSM-5 cavities, even
though the Si/Al ratio of the framework was changed.
Figure 7 shows MAS NMR spectra of Al on Pd/HZSM-5
and Co(3)–Pd/HZSM-5 with Si/Al ratio of 19.8 before
+
27
FIG. 7.
Al-MAS NMR spectra for Pd/HZSM-5 (a, b) and Pd–
◦
Co(3)/HZSM-5 (c, d) after reaction at 500 C in the absence (a, c) or in the
and after durability tests. Al in octahedral coordination presence (b, d) of water vapor. Asterisk represents the spinning sideband.

8
2
OGURA ET AL.
of catalytic activity of Pd/HZSM-5 is mainly due to the for- ion-exchange sites of the zeolite. Binding energy of metal,
mation of extra-framework PdO species.
which is measured by XPS, is sometimes shifted to a lower
level when the size of the metal is changed. In our case, how-
ever, the palladium species on zeolites is the same (isolated
DISCUSSION
2
+
Pd ) and never influenced in size. Therefore, the XPS re-
sults obtained in this study indicate that divalent palladium
species on Co(3)/HZSM-5 are in a more cationic state than
not only covalently bonding PdO but also the divalent pal-
ladium species on HZSM-5. The divalent palladium species
on Pd–Co(3)/HZSM-5 were stabilized and could maintain
their electronic state after the durability test, while some
portion of those on HZSM-5 were lowered in the binding
Roles of Cobalt on the Stability of Isolated Pd(II)
in Zeolite Cavities
Considering the results obtained in this work, we specu-
late that the cobalt included in Pd/H-zeolites will (1) en-
hance zeolites’ acidity to stabilize the isolated Pd2+; (2)
minimize the residual ion-exchangeable sites so the zeo-
lite framework will not collapse, as discussed in the case of
Fe/ZSM-5 (5); and (3) suppress the dealumination from the
zeolites’ framework to maintain ion-exchange sites for the
◦
energy even after the reaction at 400 C. This corresponds
2+
to the characteristic mobility of an isolated Pd in zeolite
cages as a hydrated cation, leading to the formation of PdO
observed after hydrothermal treatments or observed on de-
activated samples (11). The changes in selectivity for CH4-
2
+
isolated Pd
.
For acidity, one can expect this degree of enhancement
based on the promotive effect brought about by HMOR
having a higher acid strength than HZSM-5 and other
zeolites used in this study. However, NH3-TPD revealed
2+
SCR strongly support the fact that the isolated Pd sites
active for SCR are agglomerated into PdO, which shows
CH4 combustion activity. Because CH4 conversion on the
Pd–Co(3)/HZSM-5 was unchanged, and then the selectivity
was also unchanged during the test, we note that this sta-
ble performance is not due to a change in the nature of the
that isolated Co2 cations ion-exchanged in zeolites never
changed the acid strength of zeolites, while the acid amount
decreased.
+
Armor and Farris suggested in their report (20) that
cobalt-ion-exchanged ZSM-5 following treatment under
hydrothermal conditions maintains the same high surface
area as it had before the treatment. A not-so-significant
decrease in the acid amount of zeolites determined by the
NH3-TPD studies could be observed after the durability
tests for over 20 h even on Pd/HZSM-5. Additionally, Pd–
Co(3)/HMOR and Pd–Co(3)/HZSM-5 with Si/Al ratios of
2+
2+ 2+
active site, such as forming a pair site of [Pd –O–Co
] ,
2+
but rather to the high stability of the isolated Pd active
sites.
The enhancement in the stability of the isolated Pd
2+
2+
by isolated Co cations can be judged by the change
in ion-exchange property. Thermodynamic affinities of
monovalent cations for the ion exchange into ZSM-5
were determined and reported in the literature (23) hav-
7
.5 and 11.9, respectively, had many more acid sites but
+
+
+
+
ing the following sequence: Cs > Rb > NH > H3O >
4
showed stable activities during 20 h. Therefore, the residual
acid sites did not so significantly influence the durability of
zeolite. Certainly, the addition of cobalt led to the decrease
of octahedral aluminum species after the durability test, as
+
+
+
K > Na > Li . A physical retardation of ion exchange
cannot be taken into account when one considers the sizes
of ZSM-5 cavity and cations and that the titration period
◦
used (1 h at 80 C) was long enough to reach ion-exchange
27
qualitativelyconfirmedby Al-MASNMR, showninFig. 7.
A more plausible deduction from the NMR study was that
cobalt affected the magnetic field of zeolites and Co-loaded
zeolite samples gave disordered spectra. As revealed in the
durability tests and the NH4NO3 titration studies, isolated
equilibrium. Until now, the results have not yet led us to
conclude that a “site-selective” extraction of the isolated
2+
Pd from zeolites occurs during titration. Therefore, the
2+
change in the ion-exchange property of the isolated Pd
by the addition of isolated Co cations is the change in
2+
2
+
2+
Co is the effective species for stabilizing the isolated Pd
thermodynamic affinity onto ZSM-5: the difficulty of ion
2+
in the zeolite cavities. The isolated Co has unpaired elec-
trons and thus shows a paramagnetic character, resulting in
+
+
exchange with Na was diminished by using NH , which
4
exhibits stronger affinity to exchange in ZSM-5.
a highly active species for ESR (21). The isolated Pd2 is
+
2+
It can be concluded that the isolated Co species im-
also affected by co-cations from the viewpoint of the stabil-
ity against redox treatments (22). Therefore, isolated Co
cations might change an electronic density within a local
range of zeolite cavities.
2+
prove the stability of the isolated Pd cations in the ion-
2
+
exchange sites of zeolite by enhancing thermodynamic
2+
affinity of the isolated Pd to the zeolite framework, re-
sulting in higher catalytic activity and stability for CH4-SCR
under wet and thermal conditions.
This phenomenon reminds us of the results obtained by
the XPS studies in which the Pd 3d spectrum on 3.3 wt%
Co-loaded Pd/HZSM-5 was shifted to a higher level of
binding energy. The NH4NO3 titration indicates that all
of the palladium species on Pd/HZSM-5 and even on Pd–
Pd(II) and Co(II) Sitings in Zeolite Cavities
2+
Mortier suggested that divalent cations such as Ca and
Co(3)/HZSM-5 were located as the isolated Pd2 in the Ba are likely to coordinate to six- or eight-membered
+
2+

STABILITY OF ISOLATED Pd(II) CATIONS
83
rings (MR), not to pentasil ones. (24). Wichterlova and in the α site to retard stabilization of the isolated Pd2 in
co-workers suggested in their recent reports (25–30) that the same site. The results observed by the XPS that divalent
ZSM-5 has three kinds of ion-exchangeable sites for iso- palladium spectrum on Pd–Co(3)/HZSM-5 appeared after
+
2
+
lated Co , the α, β, and γ sites in 6MR, which are lo- calcination can be explained by the ion exchange, which
cated at the straight channel, the channels’ intersection, and occurs during calcination to compete for α sites. Therefore,
2
+
the sinusoidal channel of ZSM-5, respectively, and that the the isolated Pd in Pd–Co(3)/HZSM-5 must be located
2
+
site selectivity changes by changing the loading amount of next to the α-siting Co , as illustrated in Fig. 8. By be-
2
+
2+
cobalt. In this study, the isolated Co , exchangeable with ing in the closest location to the isolated Pd , the isolated
+
2+
2+
2+
NH , is effective for stabilizing the isolated Pd ; there- Co can act as a stabilizer of the isolated Pd in zeolite
fore, isolated Pd and Co sitings in ZSM-5 cavities were cavities.
discussed according to the results already reported in the
literature.
4
2+
2+
A similar interaction between isolated Pd2 and Co
was observed in the larger cavities of MOR than those of
+
2+
The Co/Al ratio in Pd–Co(3)/HZSM-5 is 0.72, and the ZSM-5. However, according to the literature (25, 28, 29),
Co2 /Co ratio determined by NH4NO3 titration was ca. 0.5; the sequence of Co siting tendency in MOR differs from
+
2+
2
+
2+
therefore, Co /Al = 0.36. Co cation is primarily located that in ZSM-5: the primary site is an α site, followed by
at β sites in ZSM-5 when a small amount of cobalt is loaded, a β site. In contrast, Descorme et al. suggested that the
and then α sites with gradual increase of Co . Considering isolated Pd in MOR is supposed to be located in a side
2+
2+
2
+
the Co /Al ratio of 0.36 in Pd–Co(3)/HZSM-5, the isolated pocket of MOR corresponding to β sites, and a gradual
2
+
Co is thought to be located on both α and β sites, accord- change in siting is observed in the sites facing the main,
ing to the literature (28, 29). When loading lesser amounts straight channel of MOR, that is, the α sites, resulting in a
of cobalt, the isolated Co2 is located only on β sites.
+
gradual increase in NO conversion (32). We conclude that
According to Mortier’s (24) and Takaishi’s rule (31), di- some correlation exists between the ion-exchange property
2
+
valent cations are to be located at the sites in 6MR. The α of the isolated Pd in HMOR and the gradual change and
site is the most released 6MR site among the three kinds of subsequent stability of the Pd/HMOR in NO conversion.
sites in ZSM-5, so when Pd(NH3)2 is ion-exchanged, such The ion-exchange of the isolated Pd in the Pd/HMOR,
+
2+
4
a bulky amine complex will be first located in the α site. evaluated by NaCl titration, was found to take a longer time
2
+
In comparison to the ion-exchange properties of isolated to elute all of the isolated Pd cations, indicating that the
2
+
2+
2+
2+
Co and Pd in ZSM-5, Co has much more affinity isolatedPd isprimarilystabilizedinthelessactive β sitein
2+
for the ion exchange sites than Pd . When enough iso- side pockets of MOR. Further, the higher NO conversion of
lated Co cations coexist in ZSM-5 along with Pd(NH3)
to be located in α sites, the isolated Co must be placed an increase in the activity suggests that the isolated Pd in
2
+
2+
Pd–Co(3)/HMOR from the beginning of a reaction without
4
2+
2+
FIG. 8. Illustration of isolated Pd2 and Co sitings in ZSM-5 cavities.
+
2+

8
4
OGURA ET AL.
the catalyst might exist in the more active α sites, and it may
REFERENCES
interact with the α-siting Co2 in the same straight channel
+
1
2
. Panov, G. I., Sobolev, V. I., and Kharitonov, A. S., J. Mol. Catal. 61, 85
of MOR.
(
1990).
In summary, the characteristic properties of isolated Pd2+
cations are clarified by locating them in the ion-exchange
. Iwamoto, M., Yahiro, H., Yuu, Y., Shundo, S., and Mizuno, N.,
Shokubai 32, 430 (1990).
3. Li, Y., and Armor, J. N., Appl. Catal. B1, L31 (1992).
. Kikuchi, E., Ogura, M., Terasaki, I., and Goto, Y., J. Catal. 161, 465
2
+
sites of zeolites. Also, the promising effect of isolated Co
cations on stabilizing the mobile, isolated Pd2 is devel-
oped by intimate siting in the zeolite cavities, as has been
proposed using Ir/In-zeolites (33). Therefore, for the high
performance of metal cations, zeolite must act as a lig-
and (34) by which cationic species are stabilized, or as a
solid solvent in which the cationic species act freely, result-
ing in pronounced catalytic activity and selectivity, which
would never be attained as metal, metal oxide, or on oxide
supports.
+
4
(
1996).
5
6
7
. Feng, X., and Hall, W. K., Catal. Lett. 41, 45 (1996).
. Nishizaka, Y., and Misono, M., Chem. Lett. 1295 (1993).
. Loughran, C. J., and Resasco, D. E., Appl. Catal. B7, 113 (1995).
8. Ali, A., Alvarez, W., Loughran, C. J., and Resasco, D. E., Appl. Catal.
B14, 13 (1997).
. Adelman, B. J., and Sachtler, W. M. H., Appl. Catal. B14, 1 (1997).
0. Aylor, A. W., Lobree, L. J., Reimer, J. A., and Bell, A. T., J. Catal. 172,
53 (1997).
1. Descorme, C., Gelin, P., Lecuyer, C., and Primet, M., Appl. Catal. B13,
85 (1997).
2. Ogura, M., Hayashi, M., Kage, S., Matsukata, M., and Kikuchi, E.,
Appl. Catal. B23, 247 (1999).
3. Koyano, G., Yokoyama, S., and Misono, M., Appl. Catal. A188, 301
9
1
1
1
1
4
1
CONCLUSIONS
Palladium ion-exchanged zeolite catalysts were tested to
(
1999).
evaluate their catalytic properties toward CH4-SCR in the 14. Ohtsuka, H., and Tabata, T., Appl. Catal. B21, 133 (1999).
1
5. Rice, M. J., Chakraborty, A. K., and Bell, A. T., J. Catal. 186, 222
(1999).
6. Traa, Y., Burger, B., andWeitkamp, J., MicroporousMater. 30, 3(1999).
7. Ogura, M., Sugiura, Y., Hayashi, M., and Kikuchi, E., Catal. Lett. 42,
presence of water vapor at a relatively high temperature
of 500 C. Pd/HZSM-5 with a Si/Al ratio of 19.8 deacti-
vated under the reaction conditions. When the Si/Al ratio
decreased, the durability was slightly enhanced but the
◦
1
1
185 (1996).
NO conversion gradually decreased. Among the zeolites 18. Ogura, M., Kage, S., Hayashi, M., Matsukata, M., and Kikuchi, E.,
Appl. Catal. B27, L213 (2000).
9. Matsukata, M., Ogura, M., Osaki, T., Rao, P. R. H. P., Nomura, M.,
and Kikuchi, E., Top. Catal. 9, 77 (1999).
0. Armor, J. N., and Farris, T. S., Appl. Catal. B4, L11 (1994).
1. El-Malki, E.-M., Werst, D., Doan, P. E., and Sachtler, W. M. H.,
J. Phys. Chem. B104, 5924 (2000).
2. Choo, H., Prakash, A. M., Zhu, Z., and Kevan, L., J. Phys. Chem.
B104, 3608 (2000).
tested, mordenite-supported palladium showed stability in
1
NO conversion during the durability test reaction.
The addition of cobalt on Pd/HZSM-5 improved the cata-
lytic stability, and no deactivation was observed on the
2
2
sample loaded 3.3 wt% cobalt. An isolated Pd2 on HZSM-
+
2
5
, which was the active and selective site for CH4-SCR, was
not eluted by means of NaCl titration after the addition of
.3 wt% cobalt. In contrast, NH4NO3 titration revealed that
2
2
3. Szostak, R., “Molecular Sieves,” second ed., Blackie, London, 1998.
4. Mortier, W. J., “Compilation of Extra Framework Sites in Zeolites.”
Butterworth, Stoneham, MA, 1982.
3
2
+
the isolated Pd existed on the Pd–Co(3)/HZSM-5. We
also found by the titration that some portion of cobalt was 25. Dedecek, J., and Wichterlov a´ , B., J. Phys. Chem. B103, 1462 (1999).
2
+
2
6. Kaucky, D., Dedecek, J., and Wichterlov a´ , B., Microporous Mater. 31,
75 (1999).
7. Wichterlov a´ , B., Dedecek, J., and Sobal ´ı k, Z., Proceedings of 12th
International Zeolite Conference, Mater. Res. Soc. Symp. Proc. 941
located as an isolated Co in the sample. XPS results sug-
gested that the isolated Co2+ cations change an electronic
density within a local range of zeolite cavities, and then
2
2
+
change the ion-exchange property of the isolated Pd . The
(
1999).
thermodynamic affinity of the isolated Pd2 to ZSM-5 was 28. Kaucky, D., Vondrov a´ , A., Dedecek, J., and Wichterlov a´ , B., J. Catal.
+
enhanced by the addition of the isolated Co2 , leading to
+
1
94, 318 (2000).
9. Dedecek, J., Kaucky, D., and Wichterlov a´ , B., Microporous Mater.
5–36, 483 (2000).
2
3
a higher stability than Pd/HZSM-5 under the durability
3
2+
conditions. The effect of the isolated Co on the stabil-
0. Sobal ´ı k, Z., Dedecek, J., Kaucky, D., Wichterlov a´ , B., Drozdova, L.,
and Prins, R., J. Catal. 194, 330 (2000).
ity of the isolated Pd2 was observed not only on ZSM-5
+
with a different Si/Al ratio but also on zeolites having dif- 31. Takaishi, T., Kato, M., and Itabashi, K., Zeolites 15, 21 (1995).
2
+
3
2. Descorme, C., Gelin, P., Lecuyer, C., and Primet, M., J. Catal. 177, 352
(1998).
ferent size cavities, such as MOR. The isolated Pd and
2
+
Co should be sited intimately in zeolite cavities to inter-
act with each other. An isolated metal cation such as Pd
is stabilized by a ligature with a ligand-like complex such as
3
3. Kikuchi, E., and Ogura, M., Cat. Surv. Jpn. 1, 227 (1997).
4. Armor, J. N., “Science and Technology in Catalysis 1994, Studies in
Surface Science and Catalysis” (Y. Izumi, H. Arai, and M. Iwamoto,
Eds.), Vol. 92, p. 51. 1995.
2
+
3
2
+
Co zeolites.