Applied Catalysis A: General 268 (2004) 43–50

Surface characterisation of Pd–Ag/Al2 O3 catalysts for acetylene

hydrogenation using an improved XPS procedure
Robert N. Lamb a , Bongkot Ngamsom a,b,c,∗ , David L. Trimm c , Bin Gong a ,
Peter L. Silveston b , Piyasan Praserthdam b
a Surface Science and Technology, School of Chemistry, University of New South Wales, Sydney, NSW 2052, Australia
b Department of Chemical Engineering, Centre of Excellence on Catalysis and Catalytic Reaction Engineering,

Chulalongkorn University, Bangkok 10330, Thailand

c Centre of Particles and Catalyst Technologies, School of Chemical Engineering and Industrial Chemistry,

University of New South Wales, Sydney, NSW 2052, Australia

Received in revised form 13 February 2004; accepted 11 March 2004

Available online 7 May 2004

Abstract

Effects of pretreatment on the surface of alumina-supported Pd–Ag catalysts with oxygen or oxygen-containing compounds (NO, N2 O,
CO and CO2 ) have been studied using an improved X-ray photoelectron spectroscopy (XPS). Surface analysis was performed either before or
after the selective hydrogenation of acetylene. Analysis of the surface after reduction shows evidence of a Pd–Ag alloy. The binding energy of
the Pd 3d is not affected by pretreatment, whereas a significant shift of the Ag 3d is revealed after NO and N2 O pretreatment. The surface after
reaction shows no state change of either Pd or Ag compared to those measured prior to reaction, which is in agreement with the reactivity test;
therefore surface modification occurs after pretreatment and is retained even after 8 h on stream. No carbonaceous deposits are formed after 8 h
on stream. Ethylene gain enhancement by NO and N2 O pretreatment is a result of strong adsorption on the surface which may block the sites
responsible for ethylene hydrogenation without facilitating carbonaceous deposits for hydrogen spillover. On the other hand, pretreatment
with O2 , CO or CO2 increases the Pd active sites, which increases C2 H2 hydrogenation activity.
© 2004 Elsevier B.V. All rights reserved.

Keywords: Pd–Ag/Al2 O3 ; Pretreatment; Oxygen-containing compounds; XPS; Acetylene hydrogenation

1. Introduction In our recent communication [15], we reported the promo-

The selective hydrogenation of acetylene over palladium
catalysts is used commercially to remove trace amounts of
acetylene contaminant in ethylene feedstreams for polyethy-
lene production [1–8]. Due to poor selectivity at high acety-
lene conversion and oligomer formation during acetylene
hydrogenation, considerable attention has been focused on
bimetallic systems in which a second metal such as silver,
copper or lead is incorporated into palladium. Substantially
increased catalytic performance as well as reduction in green
oil formation have been reported with supported Pd–Ag cat-
alysts [9–14].
∗ Corresponding author. Tel.: +66-2-7392416-9x144;
fax: +66-2-7392416-9x4.
E-mail address: knbongko@kmitl.ac.th (B. Ngamsom).
tion effect of pretreatment with oxygen or oxygen-containing
compounds (NO, N2 O, CO and CO2 ) on the catalytic perfor-
mance of Pd–Ag/Al2 O3 for the selective hydrogenation of
acetylene. The higher activity was thought to be a result of
increasing the Pd working sites. Pretreatment with NO and
N2 O gave higher ethylene gain, whereas less ethylene gain
was observed with the other pretreatment compounds. Gain
is used in industry as a measure of selectivity in acetylene
hydrogenation. A definition has been given earlier [15].
In this work, we continue our investigation of catalyst
pretreatment by oxygen-containing compounds [15]. In our
previous study, ex situ XPS was used for surface character-
isation. We now employ an improved procedure to ensure
our results were not falsified by exposure to air either dur-
ing storage or sample preparation. This procedural change
has caused a correction to our discussion of the promotion
effect. Using XPS we now observe a shift in the XPS peak

0926-860X/$ – see front matter © 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2004.03.041
44 R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50
position. The source of this has been studied as explained
below. Relevance to catalytic hydrogenation based on our
characterisation is also discussed here.
2. Experimental
2.1. Catalyst preparation and testing
The bimetallic Pd–Ag/Al2 O3 (2.8 wt.% Pd and 2.3 wt.%
Ag) catalysts used in this study were prepared by sequential
impregnation using Pd(NO3 )2 and AgNO3 as the Pd and
Ag sources, respectively. Details have been given previously
[15].
After preparation, the catalyst was pretreated with oxygen
or oxygen-containing gases, i.e., N2 O, NO, CO2 and CO,
prior to use as previously detailed [15].
The selective hydrogenation of acetylene was conducted
at 50 ◦ C for 8 h after reduction and pretreatment of the cat-
alyst. The catalyst used straight away after reduction with-
out further pretreatment was designated as untreated cata-
lyst. Hydrogenation was undertaken at a space velocity of
6000 h−1 . The reactor effluent was analysed by GC using
a carbosieve-S2 column. Again details have been discussed
earlier [15].
The performance of the catalyst was reported in terms
of acetylene conversion and ethylene gain observed from
hydrogen and acetylene concentrations as follows.
Acetylene conversion = the amount of acetylene con-
verted with respect to acetylene in the feed:


dH2
Ethylene gain (%) = 100 × 2 −
dC2 H2
where dH2 and dC2 H2 represent the amounts of total hy-
drogen and hydrogenated acetylene consumed, respectively
(for more details, see [15]).
2.2. Surface analysis
Granular catalyst samples (40/60 mesh) were transferred
from the reactor to an atmospheric bag where the samples
were prepared for analysis by further manual grinding using
a mortar and pestle in an inert (N2 ) atmosphere. Evenly
distributed powders were mounted on sample stubs using a
double-side adhesive tape, then transferred into the analysis
chamber of the XPS machine by a transfer device which
prevented contact with air.
XPS measurements were performed using a Kratos
XSAM800pci surface analysis instrument. A Mg K␣ X-ray
was used as primary excitation and was operated typically at
250 W. The analysis area was approximately 4 mm × 6 mm.
XPS elemental spectra were acquired in the fixed anal-
yser transmission (FAT) mode with 0.1 eV energy steps
at a pass energy of 20 eV. Corrections to binding energy
values, to compensate for sample charging, were made
by assigning the major component of the aluminium (Al)
2p photoelectron peak envelope to 74.6 eV. The spectra
were resolved into Gaussian–Lorentzian components after
background subtraction, using the Shirley fitting routine.
Palladium and silver foils (99.9%) were obtained from
Sigma–Aldrich Pty. Ltd. and used as metallic Pd and Ag
reference spectra after argon ion sputtering. Silver (I) oxide
and palladium oxide (Sigma–Aldrich Pty. Ltd.) were used
without further purification, as standard oxide forms of Ag
and Pd, respectively. Table 1 gives the XPS binding energy
and full width of half maximum (FWHM) values for Ag
3d and Pd 3d signals obtained from reference metallic and
oxide forms of Ag and Pd. The accuracy of the data is
approximately ±0.1 eV. As the values of 3d level splitting
of Ag and Pd are almost unchanged (ca. 6.0 and 5.3 eV,
respectively), only the values of 3d5/2 level are reported here.
3. Results and discussion
3.1. Pd–Ag surface before reaction
The values of binding energy as well as FWHM of Ag
3d5/2 and Pd 3d5/2 lines obtained from untreated and pre-
treated Pd–Ag/Al2 O3 samples are given in Table 2. Figs. 1
and 2 represent the Ag 3d and Pd 3d spectra obtained from
Pd–Ag catalysts before reaction, respectively.
For untreated catalyst, the binding energy of the Pd
3d5/2 is 334.7 eV, which is 0.7 eV lower than the bulk Pd
(335.4 eV). The binding energy of Ag 3d5/2 is at 367.4 eV,
a −0.6 eV shift from bulk silver (368.0 eV). The core level
shifts relative to pure Pd and Ag depend on changes in the
bulk charge around an atomic site. Generally, the core level
binding energy of the central atom increases as the elec-
tronegativity of the attached atoms or groups increases [16].
Since the electronegativity of Pd is higher than Ag, Pd core
levels should shift toward the lower binding energy [17,18].
The shift of Pd to lower energy is consistent with this rule.
Therefore, a solid solution or an alloy between Pd and Ag
is formed upon reduction (untreated sample). Partial charge
transfer from Ag to Pd resulting in lower binding energy
for Ag has also been reported [19–21]. A solid solution
between Pd and Ag was reported by us, based on XRD
measurement [15].
The Pd 3d5/2 binding energies obtained from pretreated
samples lie in the range 334.8–335.2 eV, which indicate that
palladium is in a metallic state. There is no evidence of a
Table 1
Binding energies (eV) of Pd 3d5/2 , Ag 3d5/2 for reference metallic and
oxide forms of Pd and Ag species
Sample Pd 3d5/2 (eV) Ag 3d5/2 (eV) FWHM (eV)
Pd metal 335.4 – –
PdO 337.5 – –
Ag metal – 368.0 0.9
Ag2 O – 368.0 1.3
 R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50 45

Table 2
XPS binding energies of Ag 3d and Pd 3d species of Pd–Ag catalysts before (roman values) and after (italics values) reaction test
Sample Pd 3d5/2 Ag 3d5/2 Pd:Ag atomic ratio

Peak position (eV) Peak position (eV) FWHM (eV)

Reduced 334.7 367.4 1.9 0.66

Pd–Ag/Al2 O3 (untreated) 334.9 367.4 1.8 0.88
O2 -treated 335.0 367.6 2.1 0.68
334.5 367.4 1.8 0.91
CO2 -treated 334.8 367.4 2.1 0.71
334.8 367.4 1.8 1.23
NO-treated 335.2 367.8 1.8 0.73
335.0 367.7 1.7 0.93
N2 O-treated 335.2 367.9 1.5 0.68
335.2 367.8 1.8 0.97
CO-treated 334.9 367.5 1.9 0.75
335.0 367.6 1.8 1.00

(f)

(e)
N(E) (arbitrary units)

(d)

(c)

(b)

(a)

378 376 374 372 370 368 366 364 362

Binding Energy (eV)

Fig. 1. XPS Ag 3d spectra obtained from (a) untreated Pd–Ag/Al2 O3 , (b) O2 -treated, (c) CO2 -treated, (d) NO-treated, (e) N2 O-treated and (f) CO-treated.
All samples before reaction.
46
(f)
(e)
(d)
N(E) (arbitrary units)
(c)
(b)
(a)
344
Fig. 2. XPS Pd 3d spectra obtained from (a) untreated Pd–Ag/Al2 O3 , (b) O2 -treated, (c) CO2 -treated, (d) NO-treated, (e) N2 O-treated and (f) CO-treated.
All samples before reaction.
PdO signal, which would be expected between 335.4 and
337.5 eV (Table 1 shows a +2.1 eV between Pd and PdO).
We speculate, however, that oxygen atoms from pretreatment
compounds are adsorbed on Pd.
After the samples were pretreated with oxygen-containing
compounds, their binding energies measured for Ag 3d5/2
show a positive shift in the range of 0.2–0.5 eV (from 367.4
to 367.6–367.9 eV). From the reference spectra of metallic
and oxide forms of Ag (see Table 1), it can be seen that
there is no shift of Ag 3d5/2 signals between Ag and Ag2 O,
however, the bandwidth broadens by 0.4 eV. Therefore, it is
likely that Ag remains metallic after pretreatment.
The reason for the shift in the XPS signal was inves-
tigated by examining the surface of the untreated and
pretreated catalysts using a transmission cell in a Nico-
let Impact 400 FT-IR. Pretreatment was performed in the
IR cell to avoid sample contamination. Fig. 3 shows the
spectra, only pretreatment with NO and N2 O causes a
significant change in the spectra. Peaks appear at 1650,
1530 and 1457 cm−1 . These indicate a linear surface com-
plex Pd–NO (literature reports a range of 1810–1650 cm−1
R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50
342 340 338 336 334 332
Binding Energy (eV)
[22,23]) as well as nitrate and nitrite species on either silver
or the alumina support (the literature suggests these bands
are at 1550 and 1465 cm−1 [22,23], or between 1600 and
1200 cm−1 [24,25]). No N2 was detected in a separate ex-
periment where the pretreated catalyst was exposed to N2 O
at 90 ◦ C. Thus, there is no oxide formation, confirming our
XPS results. Consequently, the shift in the spectra is prob-
ably explained by strong surface adsorption or perhaps the
formation of surface nitrate and nitrite on Ag.
For O2 , CO2 and CO pretreatment, no significant change
of peak positions or bandwidths of the Ag 3d lines are seen.
Although pretreatment with these compounds improved ac-
tivity, there was no improvement in ethylene gain [15]. For
NO and N2 O pretreatment, significant positive shifts of Ag
3d5/2 signals from the untreated sample were found to-
gether with narrower peak widths (−0.1 and −0.4 eV, re-
spectively). Modification of Ag in Pd–Ag alloy surface by
NOx -treatment, therefore, must have occurred. IR evidence
also suggested a change. There is another explanation of the
observed shift in the Ag 3d binding energy besides the strong
chemisorption or compound formation. A study of charge
 R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50 47

16 50 1530 14 57

NO-treated
absorbance (a.u.)

N2 O-treated

O2 -treated

CO 2 -treated
CO-treated
untreated

1700 1600 1500 1400 1300 1200

-1
wave number (cm )

Fig. 3. FT-IR spectra of Pd–Ag/Al2 O3 surface before and after pretreatment.

redistribution in a series of ion-beam-mixed Pd–Ag alloys
showed that when the Ag composition in a Pd–Ag alloy in-
creased, the Ag 3d core level shifts toward higher binding
energies and approaches the energy of bulk Ag [16]. Thus
the positive shifts of the Ag 3d core levels resulting from
NOx pretreatment may be due to changes in composition of
the Pd–Ag alloy.
3.2. Pd–Ag surface after reaction
Performance of the catalyst for acetylene hydrogenation,
in terms of acetylene conversion and ethylene gain, mea-
sured at 50 ◦ C for 8 h is illustrated in Figs. 4 and 5, re-
spectively. All treated catalysts gave a 3.1–17% increase in
acetylene conversion after 8 h on-stream compared to the

95

90

85
Acetylene conversion (%)

80

75

70

65

60
0 2 4 6 8
time on stream (h)

Fig. 4. Acetylene conversion over Pd–Ag/Al2 O3 vs. time on stream at GHSV = 6000 h−1 and 50 ◦ C: (䉬) untreated Pd–Ag/Al2 O3 , (×) O2 -treated, (䉱)
CO2 -treated, (䊉) NO-treated, ( ) N2 O-treated and (䊏) CO-treated.
48 R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50

95 temperature in the range of 40–90 ◦ C. Important differences

were observed at 50 ◦ C. Thus, testing of the pretreated
catalyst at extended exposure was undertaken at 50 ◦ C.
90

85 Fig. 4 indicates that for the untreated catalyst, conversion

increases over the first 2 h on stream and then becomes con-
80
Ethylene gain (%)

stant. N2 O pretreatment shows a similar behaviour, while

75 CO pretreatment requires about 6 h to reach a stable conver-
sion. On the other hand, NO and CO2 pretreatment slightly
70
changes over 8 h on stream. Curiously, the activity of the
65 O2 pretreated catalyst drops with time. An explanation for
this behaviour was sought out by TPO measurements of
60 the catalyst exposed for 8 h. Carbonaceous deposits were
55 observed neither after 20 min on stream, as reported earlier
[15], nor after 8 h on stream in this study.
50 Ethylene gain results in Fig. 5 are generally similar to the
0 2 4 6 8 conversion data in Fig. 4, except that the gain for O2 -treated
time on stream (h)
catalyst increases slightly with time on stream rather than
Fig. 5. Ethylene gain over Pd–Ag/Al2 O3 at GHSV = 6000 h−1 and decreases.
50 ◦ C: (䉬) untreated Pd–Ag/Al2 O3 , (×) O2 -treated, (䉱) CO2 -treated, The surface of the reacted catalysts was again monitored
(䊉) NO-treated, ( ) N2 O-treated and (䊏) CO-treated.
by XPS. Fig. 6 shows the Ag 3d doublets of samples mea-
sured after reaction. The Pd 3d spectra of the reacted sam-
untreated catalyst. Ethylene gain is enhanced by N2 O and ples are given in Fig. 7. There is no significant change in
NO pretreatment by 17.3 and 10.8%, respectively, whereas the peak positions of the Pd 3d and Ag 3d spectra compared
the other treated catalysts result in a lower ethylene gain to those observed prior to reaction. The fact that there is no
relative to the untreated catalyst. change of electronic state of the pretreated Pd–Ag bimetallic
In our previous investigation using a 20 min exposure, catalysts before and after reaction suggests that modification
we observed that acetylene conversion increased with of the Pd–Ag surface occurs during pretreatment and the

(f)
N(E) (arbitrary units)

(e)

(d)

(c)

(b)

(a)

376 374 372 370 368 366 364 362

Binding energy (eV)

Fig. 6. XPS Ag 3d spectra obtained from (a) untreated Pd–Ag/Al2 O3 , (b) O2 -treated, (c) CO2 -treated, (d) NO-treated, (e) N2 O-treated and (f) CO-treated.
All samples after reaction.

(f)
N(E) (arbitrary units)
(e)
(d)
(c)
(b)
(a)
345
Fig. 7. XPS Pd 3d spectra obtained from (a) untreated Pd–Ag/Al2 O3 , (b) O2 -treated, (c) CO2 -treated, (d) NO-treated, (e) N2 O-treated and (f) CO-treated.
All samples after reaction.
effect of pretreatment is retained even after the catalysts have
been on stream for 8 h. Adsorbed species, such as ethylidyne
(BE = 283.3 eV [26,27]) and others formed in the course
of hydrogenation steps, were also checked by examination
of the C 1s signal. The C 1s signal species should appear
between 285.0 and 285.1 eV, and there was no evidence of
ethylidyne or other carbonaceous species, so the C 1s is not
shown in this paper.
3.3. Correlation between electronic and catalytic properties
Through the results of the electronic characterisation
of the pretreated catalysts before and after reaction, an
explanation for the catalytic behaviour in the selective hy-
drogenation of acetylene can be attempted. All treated cat-
alysts improve the C2 H2 conversion compared to untreated
catalyst.
However, the magnitude of improvement differs consid-
erably. Thus, N2 O pretreatment increases conversion after
8 h on stream from 73 to 88% whereas NO pretreatment has
a very small shift to 76%. Pretreatment with O2 , CO2 have
a similar effect on C2 H2 conversion to that of NO, whereas
the effect of CO pretreatment is similar to that of the N2 O
pretreatment.
R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50 49
343 341 339 337 335 333 331
Binding Energy (eV)
Differences in ethylene gain for the different pretreat-
ments are dramatic. With N2 O, the gain after 8 h increases
from about 76 to 88%. The improvement is slightly less for
NO pretreatment going from 76 to 82%. Gain decreases sig-
nificantly to 70% for CO2 pretreatment and to 63% for O2
pretreatment.
C2 H2 hydrogenation activity enhancement upon pre-
treatment compared to the untreated catalyst is a result
of an increase in Pd active sites, as previously reported
[15].
XPS measurements show little change through pretreat-
ment and subsequent exposure to reactants except for al-
teration in Ag. In our earlier study, no Ag 3d5/2 shift was
seen but the location of this peak in the oxide region of
XPS spectra led us to assume Ag2 O was formed, possi-
bly on the Pd surface. We now think that this interpreta-
tion was incorrect. We believe that the observed shift in
our new XPS results comes from strong chemisorption or
perhaps nitrate/nitrite formation on exposed Ag in Pd–Ag
alloy.
We believe that our reactivity and XPS observation can
be explained on the basis of proposals in the literature that
postulate the existence of three types of active sites on the
palladium surface and one site on the support, as modelled
50 R.N. Lamb et al. / Applied Catalysis A: General 268 (2004) 43–50
H2 adsorption and dissociation
H2 desorption
Ag
Pd
Ag 2 References
1 Pd 3
4
Alumina support
1: Site for oligomer formation
2: Site for direct ethane formation from acetylene
3: Site for ethylene production from acetylene
4: Site for ethane production from ethylene
carbonaceous deposit bridge to support
Fig. 8. Conceptual model showing four main types of surface sites on
alumina-supported palladium catalyst and the role of Ag promoter as
desorption site for transferred H2 .
in Fig. 8 [28–35]. The three sites on the palladium surface
are responsible for selective hydrogenation of acetylene to
ethylene, direct ethane formation from acetylene and
oligomer formation. Ethylene hydrogenation is believed to
take place on the support by means of a hydrogen transfer
mechanism. It is claimed that the carbonaceous deposits
present act as hydrogen bridges for spillover [28]. Ag is re-
ported to prevent hydrogen spillover from Pd to the support
[30–33]. We believe that N2 O and NO pretreatments inter-
fere with the spillover mechanism. This seems the only way
to explain the increase in ethylene gain and the increase in
acetylene activity. However, we observe no carbon on the
catalyst surface after 8 h on stream. Furthermore, gain in-
creases with time on stream for all pretreatments. This would
not be expected if carbonaceous deposits are involved with
ethylene hydrogenation, because those deposits can only in-
crease with time on stream. Thus, if these deposits involved
in the ethylene hydrogenation mechanism, the gain should
decrease.
The decrease in gain for O2 , CO and CO2 pretreatment
is consistent with increased dispersion of Pd on the support,
which increases acetylene hydrogenation activity as well as
the rate of H2 spillover onto the support.
Acknowledgements
The project was financially supported by The Thailand
Research Fund (TRF) and TJTTP-JBIC. B. Ngamsom
would like to thank The Surface Science Technology and
Centre of Particles and Catalyst Technologies, School of
Chemical Engineering and Industrial Chemistry, The Uni-
versity of New South Wales, Australia, for providing ex-
perimental facilities for this work. The technical assistance
of and fruitful discussions with Mr. Wilhelm Holzschuh are
gratefully acknowledged.
[1] Q. Zhang, X. Liu, Q. Zhu, Petrochem. Technol. 27 (1998) 53.
[2] G.C. Bond, P.B. Wells, J. Catal. 5 (1965) 65.
[3] A.N.R. Bos, L. Van De Beld, H.J. Martens, K.R. Westerterp, Chem.
Eng. Commun. 21 (1993) 27.
[4] A.N.R. Bos, L. Van De Beld, H.J. Martens, K.R. Westerterp, Chem.
Eng. Commun. 121 (1993) 55.
[5] C. Gaudiness, A.L. Cabanes, G. Villora, Can. J. Chem. Eng. 74
(1996) 84.
[6] A.H. Weiss, B.S. Gambhir, R.B. La Pierre, W.K. Bell, Ind. Eng.
Chem. Proc. Des. Develop. 16 (1977) 352.
[7] H.R. Aduriz, P. Bodnariuk, M. Dennehy, C.E. Gigola, Appl. Catal.
58 (1990) 227.
[8] S. Leviness, V. Nair, A.H. Weiss, Z. Schay, L. Guczi, J. Mol. Catal.
25 (1984) 131.
[9] B. Cordts, D. Pease, L. Azaroff, Phys. Rev. B 24 (1981) 538.
[10] G. Meitzer, J.H. Sinfelt, Catal. Lett. 30 (1995) 1.
[11] K.H. Chae, et al., Phys. Rev. B 53 (1996) 10328.
[12] I. Coulthard, T.K. Sham, Phys. Rev. Lett. 77 (1996) 4824.
[13] Q. Zhang, J. Li, X. Liu, Q. Zhu, Appl. Catal. A. 197 (2000) 221.
[14] C.E. Gigola, H.R. Aduriz, P. Bodnariuk, Appl. Catal. 27 (1986) 133.
[15] P. Praserthdam, B. Ngamsom, N. Bogdanchikova, S. Phatanasri, M.
Pramotthana, Appl. Catal. A: Gen. 230 (2002) 41.
[16] K.H. Chae, et al., Nucl. Instrum. Meth. Phys. Res. B 106 (1995) 60.
[17] L. Pauling, The Nature of the Chemical Bond, Cornell University
Press, New York, 1960, p. 93.
[18] L.R. David (Ed.), Handbook of Chemistry and Physics, 80th ed., p.
1999.
[19] A.K. Datye, J. Bravo, T.R. Nelson, P. Atanasova, M. Lyubovsky, L.
Pfefferle, Appl. Catal. A: Gen. 198 (2000) 179.
[20] A.M. Venezia, L.F. Liotta, G. Deganello, Z. Schay, L. Guczi, J.
Catal. 182 (1999) 449.
[21] L.F. Liotta, A.M. Venezia, G. Deganello, A. Longo, A. Martorana,
Z. Schay, L. Guczi, Catal. Today 66 (2001) 271.
[22] M. Valden, R.L. Keiski, N. Xiang, J. Pere, J. Aaltonen, M. Pessa, T.
Naunula, A. Savimaki, A. Lahti, M. Harkonen, J. Catal. 161 (1996)
614.
[23] K. Almusaiteer, S.C. Chuang, J. Catal. 180 (1998) 161.
[24] V. Zuzaniuk, F.C. Meunier, J.R.H. Ross, J. Catal. 202 (2001) 340.
[25] F.C. Meunier, V. Zuzaniuk, J.P. Breen, M. Olsson, J.R.H. Ross, Catal.
Today 59 (2000) 287.
[26] Sh.K. Shaikhutdinov, M. Frank, M. Baumer, S.D. Jackson, R.J.
Oldman, J.C. Hemminger, H.-J. Freund, Catal. Lett. 80 (3–4) (2002)
115.
[27] M. Frank, M. Baumer, R. Kuhnemuth, H.-J. Freund, J. Vac. Sci.
Technol. A 19 (2001) 1497.
[28] S. Asplund, J. Catal. 158 (1996) 267.
[29] A.S. Al-Ammar, G. Webb, J. Chem. Soc. Faraday 174 (1978) 195.
[30] A.S. Al-Ammar, G. Webb, J. Chem. Soc. Faraday 174 (1978) 657.
[31] A.S. Al-Ammar, G. Webb, J. Chem. Soc. Faraday 175 (1978) 1900.
[32] J. Margitfalve, L. Guczi, A.H. Weiss, J. Catal. 75 (1984) 417.
[33] J. Margitfalve, L. Guczi, React. Kinet. Catal. 72 (1980) 475.
[34] J.M. Moses, A.H. Weiss, K. Matsusek, L. Guczi, J. Catal. 56 (1984)
417.
[35] A.H. Weiss, S. Leviness, V. Nau, L. Guczi, A. Sarkany, Z. Schay,
Proceedings of the Eighth International Congress on Catalysis, vol.
5, 1984, p. 591.