The hydrogenation of 1,3-pentadiene over an alumina-supported palladium catalyst: An FTIR study

Article abstract of DOI:10.1039/b413178a

The hydrogenation of a mixture of cis- and trans-1,3-pentadiene over a 1% Pd/Al₂O₃ catalyst at 303 K has been studied using infrared spectroscopy to monitor the changes in the composition of the gas phase over the catalyst as a function of time. The reaction is seen to occur as a consecutive process, with the terminal double bond hydrogenated in advance of the internal double bond. Vibrational assignments have been confirmed through ancillary calculations for a number of C₅ molecules. The reaction profile is consistent with the catalyst presenting two distinct reaction sites: hydrogenation of the terminal double bond occurs at Site α, whilst Site β is responsible for hydrogenation of the internal double bond. Trans-pent-2-ene is identified as the only reaction intermediate. From comparative studies of the hydrogenation of pentenes over the catalyst, the absence of any cis-pent-2-ene in the reaction mixture is tentatively attributed to cis-1,3-pentadiene isomerising at Site α to form trans-1,3-pentadiene. The effect of toluene-d₈ to act as a chemical modifier was also investigated and shown to selectively poison Site β Site α being unperturbed.

Full text of DOI:10.1039/b413178a

View Article Online / Journal Homepage / Table of Contents for this issue
R E S E A R C H P A P E R
The hydrogenation of 1,3-pentadiene over an alumina-supported
palladium catalyst: an FTIR studyw
Elaine Opara,a David T. Lundie,a Timothy Lear,a Iain W. Sutherland,a Stewart F. Parkerb
and David Lennon*a
a Department of Chemistry, Joseph Black Building, University of Glasgow, Glasgow,
UK G12 8QQ. E-mail: d.lennon@chem.gla.ac.uk; Fax: (44)-(0)-141-330-4888;
Tel: (44)-(0)-141-330-4372
b ISIS Facility, Rutherford Appleton Laboratory, Chilton, Didcot, Oxfordshire,
UK OX11 0QX
Received 26th August 2004, Accepted 28th October 2004
First published as an Advance Article on the web 11th November 2004
The hydrogenation of a mixture of cis- and trans-1,3-pentadiene over a 1% Pd/Al2O3 catalyst at 303 K has been
studied using infrared spectroscopy to monitor the changes in the composition of the gas phase over the catalyst
as a function of time. The reaction is seen to occur as a consecutive process, with the terminal double bond
hydrogenated in advance of the internal double bond. Vibrational assignments have been confirmed through
ancillary calculations for a number of C5 molecules. The reaction profile is consistent with the catalyst presenting
two distinct reaction sites: hydrogenation of the terminal double bond occurs at Site a, whilst Site b is responsible
for hydrogenation of the internal double bond. Trans-pent-2-ene is identified as the only reaction intermediate.
From comparative studies of the hydrogenation of pentenes over the catalyst, the absence of any cis-pent-2-ene
in the reaction mixture is tentatively attributed to cis-1,3-pentadiene isomerising at Site a to form trans-1,3-
pentadiene. The effect of toluene-d8 to act as a chemical modifier was also investigated and shown to selectively
poison Site b, Site a being unperturbed.
diene.9 Reactant isomerization was slow over Pd catalysts and,
on the basis of n-pentene distributions, it was concluded that
1. Introduction
Olefin chemistry on the industrial scale is closely linked with
3,4- addition of hydrogen was more important than 1,2-
addition.9 Additionally, pentane yields were shown to be low,
increased refinery capacity and covers a wide range of chemical
processes.1a C2–C4 olefins dominate although higher olefins
(C5–ca. C18) are also industrially significant, with an increasing
with the catalysts exhibiting a high selectivity for the partial
hydrogenation products. Vasquez and Madix examined the
interest in unbranched olefins due to the latter’s linearity linked
reactivity of unsaturated linear C6 hydrocarbons on Pd(111)
to products with favourable advantages such as biodegradabil-
surfaces and showed 1,3-hexadiene to hydrogenate to hexene
ity.1b A crucial descriptor for the performance of unbranched
but, in a similar manner to that reported by Wells and Wilson,9
further hydrogenation to hexane was not observed.10 H–D
higher olefins is whether the double bond is positioned at the
end of the molecule (terminal) or is internal to the molecule.
exchange into the adsorbed alkene was reported, which was
The market for terminal olefins is greater, with a world
manufacturing capacity in 1994 of 2.2 Mtonne.1c Diolefins,
hydrogenated intermediate.10 Subsequent work by Jackson
proposed to occur via reversible C–H bond formation in a half-
or dienes, are also industrially significant, with C4 and C5 1,3-
dienes featuring strongly. Here, the double bonds are conju-
et al. used FTIR spectroscopy to investigate the hydrogenation
of 1,3-hexadiene over a silica-supported Pd catalyst in the
gated and the molecules are more reactive than their monoene
presence and absence of toluene-d8, acting as chemical modi-
counterparts.1d 1,3-Butadiene dominates this market and con-
fier.11 Depending on the order in which the modifier was added
sequently a number of studies have investigated the reaction
to the reaction system, the toluene-d8 could act as either a
characteristics of this molecule. For example, the hydrogena-
promoter or a poison. Furthermore, with respect to the relative
tion properties have been comprehensively studied by Wells
and co-workers2–5 and the general topic has been reviewed by
reactivity of the two olefin groupings, that study revealed
differences in the reaction characteristics of 1,3-hexadiene
Webb.6 More recently, Souza et al. have examined geometric
effects active in butadiene hydrogenation over Pd supported
compared to 1,3-pentadiene. Vasquez and Madix suggest that
the central double bond of 1,3-hexadiene to be hydrogenated
systems7 and Moyes et al. have examined electronic effects in
before the terminal double bond,10 however, the infrared
evidence for the Pd/SiO2 catalyst clearly indicated that the
butadiene hydrogenation catalysed by transition metals.8
Of the higher homologues, Wells and Wilson investigated
terminal double bond was hydrogenated in advance of the
internal double bond.11 Additionally, full hydrogenation of
1,3-hexadiene was reasonably facile over the Pd/SiO2 cata-
the hydrogenation of cis- and trans-1,3-pentadiene catalysed by
Co, Ni, Cu, Pd and Pt catalysts, with the general features of the
lyst,11 whilst this pathway was not accessible for the same
gas phase hydrogenation closely resembling those of 1,3-buta-
molecule on the Pd single crystal.10
w Electronic supplementary information (ESI) available: Comparison
Regarding selective reduction of dienes, usually it is the least
of the inelastic neutron scattering spectrum of pentane (solid line) with
hindered double bond that is preferentially hydrogenated and,
that calculated by DFT (B3LYP/6-31G*(d,p)) (dashed line) (Fig. S1);
when steric hindrance effects are approximately equal, the most
comparison of observed and calculated positions and intensities of cis-
and trans-1,3-pentadiene (Tables S1 and S2). See http://www.rsc.org/
strained bond will be hydrogenated first.12 Selectivity charac-
teristics are known to be metal dependent,12,13 which is thought
suppdata/cp/b4/b413178a/
5588
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
to reflect the relative tendencies of metals to induce conjuga-
tion through double bond migration.12 Additionally, for re-
ductions of dienes to monoenes, it is often practical to stop
hydrogen consumption at the stoichiometric point, as few
reactions will stop spontaneously and will continue on to form
the fully saturated product.12
microscope but, unfortunately, it was not possible to precisely
identify metal particles against the presence of the support
material. This observation is reasonably consistent with the
chemisorption data, in that at 2.4 nm the metal particles are
too small to be detected by this instrument with a support
material that offers poor contrast to metal particles of these
dimensions. Temperature programmed reduction studies con-
firmed that reduction of the palladium occurred below room
temperature.
Against this background, it was decided to investigate the
hydrogenation of 1,3-pentadiene over an alumina-supported
Pd catalyst, to determine the relative sensitivity of the two
olefin functionalities to hydrogenation, and to evaluate the
potential for toluene-d8 to act as a selective modifier for this
reaction. Following on with the approach adopted for the 1,3-
hexadiene study over a supported palladium catalyst,11 FTIR
was selected to monitor the gas phase reaction as distinct
vibrations of the reacting molecules can be used to follow the
course of the reaction sequence. Additionally, by using deut-
erated toluene, any incorporation of hydrogen from the modi-
fier to the product can be readily identified, providing
information on hydrogen exchange processes present at the
catalyst surface. It is noted that the separation of 1,3-penta-
diene from the C5 cracking fraction is very involved,1c,e with
the 1,3-pentadiene existing as a mixture of cis- and trans-
isomers. To mimic the likely industrial feed, this study has
investigated a mixture of cis- and trans-1,3-pentadiene. Hydro-
genation of C5 monoenes was also considered to gauge the
effect of conjugation on the hydrogenation activity. Vibra-
tional assignments have been confirmed through ancillary
calculations for a number of C5 molecules.
2.2 Apparatus and reaction testing
The reactor was based around a modified Graseby-Specac 5661
heated gas cell fitted with KBr windows and a 20140 automatic
temperature controller. Two Brooks 5850E mass flow control-
lers controlled the flow of hydrogen (BOC, 99.995% purity)
and helium (BOC, 99.999% purity) via an in-line purifier
(Messer Grieshiem Oxisorb) into a 50 cm3 stainless steel mixing
vessel packed with glass beads before entering into the reaction
cell. A back pressure regulator (Norgen) and isolation valves
allowed the cell to be isolated and the pressure monitored. The
reactor was housed within a Nicolet Avatar FTIR spectro-
meter and fitted via rubber sleeves to the spectrometer internal
gas purge facility (Peak Scientific) to minimise atmospheric
contributions to the spectra. The reactor was additionally fitted
with an injection septum, through which reagents could be
added to the reaction system.
The catalyst was loaded into the reactor as a pressed disc.
For all the studies presented in this study, the catalyst was
mounted on a glass sample holder so that the catalyst was not
in the path of the infrared beam. In this way, all spectra
presented arise from the composition of the gaseous phase,
with no contribution from the catalyst or the catalyst surface.
This simplifies the analysis considerably. Subsequent studies
have examined the infrared spectrum when the catalyst disc is
mounted in the beam but that work will be presented else-
where. The catalyst mass was selected to yield a full hydro-
genation profile in a time that was sufficiently long (ca. 30 min)
so that infrared spectra of sufficient signal/noise ratio could be
repeatedly recorded during that interval, so as to define a
representative reaction profile. In this manner, 1 part of the
1% Pd/Al2O3 catalyst was diluted with 9 parts of the support
material (g-alumina), ground in a pestle and mortar, then ca.
20 mg of this mixture pressed into a thin disc using a Perkin
Elmer hydraulic press. The catalyst disc was loaded into the
reactor and reduced in the following manner. The cell tem-
perature was maintained at 303 K and a mixture of 6.3% H2–
93.7% He was passed through the reactor at a flow rate of 20
ml minꢀ1 for 25 min, then the hydrogen composition was
increased to yield an equimolar mixture of hydrogen and
helium at the same flow rate. After 5 min the reactor was
isolated at a pressure of 1672 torr (0.223 MPa). Hydrocarbons
were injected into the cell via the septum. 9 ml of 1,3-pentadiene
(piperylene) were used and 5 ml of toluene-d8. Scanning com-
menced as soon as the injection of the hydrocarbon was
complete. This combination of reagents results in a hydrogen:
pentadiene ratio of 49 : 1 and a pentadiene : Pd(s) ratio of
1160 : 1, i.e. hydrogen is in excess of the hydrocarbon and the
hydrocarbon is in excess of palladium surface atoms. Thus, the
infrared cell is acting as a batch reactor under conditions where
reasonable conversions represent multiple turnovers, unhin-
dered by the availability of hydrogen. Infrared spectra were
recorded at a resolution of 4 cmꢀ1, co-adding 8 scans and
requiring an acquisition time of 10 s. The reaction temperature
was maintained at 303 K for all the reactions studied. Experi-
ments on an alumina disc showed no conversion of 1,3-
pentadiene. The toluene-d8 (Aldrich) had a minimum purity
of 99.95% D. The 1,3-pentadiene used was technical grade
(Aldrich, 490% purity) and comprised a mixture of cis- and
trans-isomers. The sample was analysed by gas liquid
This study shows that the reaction proceeds as a consecutive
process, with the terminal double bond hydrogenated in ad-
vance of the internal double bond. The reaction profile is
consistent with the catalyst presenting two distinct reaction
sites: hydrogenation of the terminal double bond is thought to
occur at Site a, whilst Site b is responsible for hydrogenation of
the internal double bond. Whereas previous work reports little
or no isomerisation for 1,3-pentadiene hydrogenation over Pd-
based catalysts,9 cis to trans isomerisation is shown to be facile
over this catalyst and is tentatively associated with Site a. The
effect of toluene-d8 to act as a chemical modifier is noted to be
different to that observed in the 1,3-hexadiene study11 and,
moreover, is seen to selectively poison Site b, leaving Site a
unperturbed.
2. Experimental
2.1 Catalyst preparation and characterisation
The 1% w/w Pd/Al2O3 catalyst was prepared from anhydrous
palladium chloride (Fluka Chemicals) by wet impregnation.
The support material was g-alumina (Degussa Aluminium
Oxid C, BET surface area 102 m2 gꢀ1). Dropwise addition of
HCl was added to a mixed slurry prepared in water to ensure
complete dissolution of the PdCl2. The mixture was transferred
to a rotary evaporator where the water was slowly removed by
maintaining the sample at 353 K under a nitrogen atmosphere
to produce a free-flowing powder. The resulting catalyst was
then dried for 12 h at 373 K.
Atomic absorption measurements were performed using a
Perkin Elmer 1100B spectrophotometer, which revealed the
actual Pd loading of the catalyst to be 0.91%. Carbon mon-
oxide pulse chemisorption was used to evaluate palladium
dispersion. Saturation corresponded to a CO coverage of
19.5 ꢁ 1.4 mmol CO g(cat)ꢀ1, with the error representing ꢁ2
standard deviations in triplicate measurements. This coverage
corresponds to 2.35 ꢂ 1019 surface Pd atoms gꢀ1, assuming a
Pd : CO surface stoichiometry of 2 : 1,14 which represents a
metal dispersion of 45.6%. Assuming the metal particles are
spheres of equal diameter, the chemisorption result equates to
a mean particle size of 2.4 nm. The catalyst was examined by
transmission electron microscopy using a JOEL 1200 FX
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
5589

View Article Online
chromatography and 1H NMR spectroscopy, which showed
the 1,3-pentadiene isomeric distribution to be comprised of
36% of the cis-isomer and 64% of the trans-isomer. Reflecting
the industrial scenario,1c,e a small quantity of cyclopentene was
also present. This comprised o4% of the mixture and there-
fore represents a minor contaminant. Infrared spectroscopy of
the 1,3-pentadiene in the vapour phase confirmed the mixture
of isomers.15 Pent-1-ene (499% purity, Aldrich) and cis-pent-
2-ene (498% purity, Aldrich) were analysed and found to
contain no cyclopentene.
3. Computational studies
Perhaps surprisingly, there have been very few studies of the
vibrational spectra of C5 molecules. Pentane16 and cis- and
trans-1,3-pentadiene17 have been studied but there are no
studies of the isomeric pentenes to our knowledge. Unambig-
uous assignment of the modes of the molecules investigated
here is essential to their correct identification under catalytic
conditions. Density functional theory (DFT) provides frequen-
cies that are accurate to a few per cent for small molecules.18
To provide the necessary vibrational data we have carried out
DFT calculations using the B3LYP functional with the 6-
311*g(d,p) basis set as implemented in GAUSSIAN03.19 Com-
parison with the literature values for cis- and trans-1,3-penta-
diene17 gave generally good agreement, as shown in Tables S1
and S2.w The few disagreements occur for modes that are weak
in both the infrared and Raman spectra, so are less certain. A
stringent test of any calculation of the vibrational spectrum is
comparison with the inelastic neutron scattering (INS) spec-
trum, since this only depends on the atomic displacements in
the modes.20 The INS spectrum of pentane is available from
www.insdatabase.rl.ac.uk and the comparison is shown in Fig.
S1.w The excellent agreement is strong evidence that the DFT
calculations provide a rigorous basis for the assignment of the
spectra.
Fig. 1 The infrared spectrum in the 3150–2770 cmꢀ1 region as a
function of time for Pd/Al2O3 þ H2 þ 1,3-pentadiene: (a) 0, (b) 8, (c)
12, (d) 23 and (e) 27 min. t ¼ 0 represents the time when the injection of
the diene into the reactor is complete.
Collectively, these modes can be used to examine the reaction
coordinate and investigate the principal stages of the hydro-
genation process.
Fig. 1 presents the infrared spectrum in the 3150–2770 cmꢀ1
region as a function of time, with t ¼ 0 representing the time
when the injection of the diene into the reactor is complete. At
this time two bands at 3096 and 3024 cmꢀ1 represent unsatu-
rated functionalities. The anti-symmetric QCH2 stretch at
3096 cmꢀ1 17 is uniquely associated with the terminal olefinic
group and reduces to zero intensity between 8–12 min. The
symmetric –CHQstretch at 3020 cmꢀ1 has residual intensity at
12 min but is absent in the 27 min spectrum. This feature is
associated with the presence of an olefinic functionality and,
therefore, it appears that the terminal double bond is hydro-
genated before the internal double bond. The formation of
saturated species can be represented by the intensity of the anti-
symmetric C–H stretch about 2976 cmꢀ1 21
This is seen to
.
4. Results
increase dramatically within the first 8 min of reaction. The
proportionally greater band intensity of the saturated analogue
is thought to partially reflect the distribution of material within
the reactor. Unsaturated species will favour binding to the
catalyst, reducing their gas phase concentration. Upon forma-
tion of the saturated product, these molecules are released to
the gas phase, where a noticeable increase in intensity is
observed.
Fig. 2 presents the infrared spectrum in the 1880–1730 cmꢀ1
region as a function of time. At t ¼ 0, a sharp band is seen at
1810 cmꢀ1, which is assigned to the overtone of theQCH2 wag
of 1,3-pentadiene.17 On increasing time, the intensity of this
band progressively decreases up to 12 min, when no signal is
detected. This mode is specifically associated with the terminal
The hydrogenation of 1,3-pentadiene was studied under three
reaction environments. Firstly, the diene was added to the
reactor that was pre-charged with an excess of hydrogen gas.
Secondly, the hydrogen charged catalyst was pre-treated with a
potential modifier, toluene-d8, before introduction of the diene.
Thirdly, the order of addition of the reagents was further
shuffled so that the modifier was added prior to the diene
and the hydrogen. Finally, in order to evaluate the relative
rates and performance of the monoene options, the hydroge-
nation of pentene over a hydrogen charged catalyst was also
examined.
4.1. H2 þ 1,3-pentadiene
Post-catalyst reduction, the diene was injected into the reactor
with an overpressure of hydrogen of 836 torr present over the
catalyst and the infrared spectrum of the gas phase continu-
ously monitored. The various reacting entities exhibit a range
of vibrational modes that are characteristic of their molecular
arrangements, which thereby permit the reaction profile for the
hydrogenation reaction to be determined. For 1,3-pentadiene,
the anti-symmetricQCH stretch at 3096 cmꢀ1, the overtone of
the QCH2 wag at 1807 cmꢀ1 and the fundamental of the Q
CH2 wag at 900 cmꢀ1 represent diagnostic regions of the
vibrational spectrum.17 Partial hydrogenation converts the
diene to a monoene with, for example in the case of trans-
pent-2-ene, the symmetric –CHQstretch at 3024 cmꢀ1 and the
trans-CHQCH wag at 969 cmꢀ1 providing distinct vibrational
signatures.21 Full hydrogenation yields pentane. The intensity
of the aliphatic modes reflects conversion but, additionally, the
anti-symmetric methyl deformation exhibits a slightly higher
frequency (1462 cmꢀ1) than its unsaturated counterparts.15,21
Fig. 2 The infrared spectrum in the 1880–1730 cmꢀ1 region as a
function of time for Pd/Al2O3 þ H2 þ 1,3-pentadiene: (a) 0, (b) 5, (c) 8
and (d) 12 min.
5590
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
pour phase infrared measurements showed pent-1-ene to be
characterised by two strong features in this region at 1005
(trans-CH wag) and 920 cmꢀ1 (CH2 wag),21 whereas cis-pent-2-
ene exhibits a band at 940 cmꢀ1. Calculations show that this
mode is the out of phase C1–C2 and C4–C5 stretch. None of
these bands coincide with the 969 cmꢀ1 feature present in Fig.
3. Rather, this band more closely corresponds to that of the
trans-CHQCH wag of trans-pent-2-ene,15,21 to which this
feature is assigned. Thus, this analysis reveals that trans-pent-
2-ene is formed as a reaction intermediate and there is no
evidence for production of other pentenes. The absence of
pent-1-ene confirms the selective hydrogenation of the terminal
double bond in advance of the internal double bond hydro-
genation.
The richness of the vibrational spectrum can then be used to
map out a reaction profile for the series of events occurring
within the reactor upon addition of the diene to the hydrogen-
charged catalyst. This is presented in Fig. 4. The terminal
double bond of 1,3-pentadiene is hydrogenated within a period
of 10 ꢁ 2 min to form uniquely trans-2-pentene. This species is
then hydrogenated further to produce pentane. That process
occurs within 25 ꢁ 2 min. The errors represent the range
observed in duplicate measurements and the data is collated
in Table 1. No other reaction products were identified during
the course of the reaction sequence. The absence of pent-1-ene
in the spectrum eliminates the possibility that the reaction
follows a concurrent pathway and therefore the reaction is
shown to occur via a step-wise (consecutive) process.
The 1,3-pentadiene mixture contains a small quantity of
cyclopentene (o4%). Cyclopentene can be distinguished from
cis- and trans-1,3-pentadiene and the resulting hydrogenation
products by strong bands about 713 and 1067 cmꢀ1. The
former is a cis-C–H wag, whereas the latter is an out of plane
ring deformation. Neither of these features are observable
when the 1,3-pentadiene mixture is injected into the reaction
cell. This could reflect the low concentration of the cyclopen-
tene in the sample or, alternatively, that the cyclopentene is
adsorbed on the catalyst surface. Whichever, this entity is
thought to play no significant part in the observed chemistry.
This matter is considered further in Section 5.1.
Fig. 3 The infrared spectrum in the 1125–790 cmꢀ1 region as a
function of time for Pd/Al2O3 þ H2 þ 1,3-pentadiene: (a) 0, (b) 8,
(c) 12, (d) 23 and (e) 27 min.
double bond of the diene and indicates hydrogenation to be
occurring at this centre. The intensity profile for the 1810 cmꢀ1
band matches that of the anti-symmetricQCH2 stretch at 3096
cmꢀ1 seen in Fig. 1, confirming the initial loss of the terminal
olefinic unit.
The spectrum in the 1125–790 cmꢀ1 region is shown in Fig.
3. The t ¼ 0 spectrum (Fig. 3a) is characterised by the trans-CH
QCH wag (QCH out-of-plane deformation) at 1002 cmꢀ1 17
which represents the diene functionality, and the fundamental
,
of theQCH2 wag at 900 cm
,
ꢀ1 17 which is uniquely associated
with the terminal olefinic group. At 8 min the trans-CHQCH
wag appears to shift to 969 cmꢀ1, then to progressively
decrease in intensity up to 23 min, where no significant
intensity is seen. Although no such shift is apparent with the
QCH2 wag, the intensity of this feature drops dramatically in
the first 8 min but then remains constant up to 27 min. The
residual intensity at 910 cmꢀ1 beyond 8 min is attributed to
saturated product15 and the rapid decrease up to that time is
associated with consumption of the terminal olefinic group.
This analysis is consistent with the rate behaviour of the anti-
symmetricQC–H stretch (3096 cmꢀ1, Fig. 1) and the overtone
of QCH2 wag (1807 cmꢀ1, Fig. 2). The shift of the 1,3-
pentadiene trans-CHQCH wag at 1002 cmꢀ1 to 969 cmꢀ1 is
indicative of the transition from diene to monene, with the
resulting loss of conjugation causing a shift of this mode to
lower frequency.22 Comparing candidate monoenes,15,23,24 va-
4.2. H2 þ toluene-d8 þ 1,3-pentadiene
The second series of diene hydrogenation reactions involved
introducing a potential chemical modifier, toluene-d8, to the
hydrogen charged catalyst in advance of the diene. Whereas
the anti-symmetric QCH2 stretch at 3096 cmꢀ1 was removed
between 9 and 10 min, in a comparable manner to that seen
previously, the symmetric –CHQstretch at 3024 cmꢀ1 was still
present at 77 min. This latter mode represents the presence of
an olefinic grouping and the time taken for the loss of olefinic
Table 1 Time for the complete hydrogenation of the terminal and
internal double bonds within the C5 framework for the five reactions
studied. The errors quoted represent the range from duplicate measure-
ments
Reaction
No.
Reaction
tterminal/min tinternal/min
1
2
H2 þ 1,3-pentadiene.
H2 þ toluene-d8 þ 1,
3-pentadiene.
10 ꢁ 2
10 ꢁ 2
25 ꢁ 2
ca. 100
3
Toluene-d8þ1,3-pentadiene
þ H2.
12 ꢁ 2
43–164
4
5
6
H2 þ cis-pent-2-ene.
H2 þ pent-1-ene.
—
11–14
17–19
25–27
—
27–30
Fig. 4 Reaction profile for the hydrogenation of 1,3-pentadiene over a
1% Pd/Al2O3 catalyst at 303 K. The vibrational modes used to define
the relative concentration/time dependence of the reacting species are
presented in parentheses: 1,3-pentadiene (QCH2 wag at 904 cmꢀ1),
trans-pent-2-ene (trans-CHQCH wag at 969 cmꢀ1) and pentane (anti-
symmetric C–H deformation at 1463 cmꢀ1).
H2 þ 1,3-pentadiene after
3 repeat 1,3-pentadiene
hydrogenation cycles.
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
5591

View Article Online
Fig. 6 The infrared spectrum in the 1100–800 cmꢀ1 region as a
function of time for Pd/Al2O3 þ toluene-d8 þ 1,3-pentadiene þ H2:
(a) 0, (b) 4, (c) 7, (d) 12, (e) 43 and (f) 164 min.
Fig. 5 The infrared spectrum in the 1100–800 cmꢀ1 region as a
function of time for Pd/Al2O3 þ H2 þ toluene-d8 þ 1,3-pentadiene:
(a) 0, (b) 9, (c) 12, (d) 77 and (e) 100 min.
reaction 2, no incorporation of deuterium was observed in the
pentane product.
character is considerably longer than in the absence of a
modifier. This increased reaction time relates specifically to
consumption of the internal double bond.
The 1100–800 cmꢀ1 region is shown in Fig. 5 and confirms
the reduced rate of reaction for the second stage hydrogenation
reaction under these conditions. TheQCH2 wag of 1,3-penta-
diene at 904 cmꢀ1 is lost between 9 and 12 min. The loss of
conjugation again attenuates the 1,3-pentadiene trans-CHQ
CH wag at 1002 cmꢀ1. The resulting presence of trans-pent-2-
ene is signified by the trans-CHQCH wag of this entity at 969
cmꢀ1. This band has minimal intensity at 100 min, indicating
that the trans-pent-2-ene has been fully hydrogenated about
this time.
4.4. Pentene hydrogenation
Previous work from this laboratory on hydrogenation reac-
tions over supported Pd catalysts has shown that certain
catalysts can present a substantial barrier to alkene hydroge-
nation.25 In order to check that no such problems existed with
this catalyst and to ensure that the catalyst was capable of
hydrogenating a terminal and an internal double bond in the
absence of any other olefinic functionality, hydrogenation of
the monoenes pent-1-ene and cis-pent-2-ene was investigated.
Fig. 7 presents the infrared spectrum in the 1050–875 cmꢀ1
region for the hydrogenation of cis-pent-2-ene as a function of
reaction time. The t ¼ 0 plot (Fig. 7a) shows the band at 933
cmꢀ1 (out of phase C1–C2 and C4–C5 stretch of cis-pent-2-ene)
to shift to 969 cmꢀ1, which is characteristic of the trans-CHQ
CH wag.21 The 969 cmꢀ1 feature then decreases in intensity on
increasing reaction time up to 27 min, when no signal is
apparent. This sequence indicates that, in the presence of the
catalyst, the cis-pent-2-ene isomerises to trans-pent-2-ene,
which is then hydrogenated to form pentane. The residual
signal at 915 cmꢀ1 in Fig. 7e, initially evident at 16 min, is
associated with the formation of pentane.15
From duplicate measurements, the complete loss of the
terminal bond requires 10 ꢁ 2 min, which indicates the
modifier to have negligible effect on the first stage hydrogena-
tion process. In contrast, hydrogenation of the internal double
bond requires ca. 100 min, 4 times longer than when no
modifier was present (Table 1). Clearly, the modifier is acting
as a selective poison for the second stage hydrogenation
reaction. It is noted that this modification to reaction rate
has not affected the selectivity profile. Additionally, if the D-
atoms associated with the toluene-d8 were incorporated into
the pentane product, there would be a growth in intensity of
the aliphatic C–D stretching modes around the region of 2100
cmꢀ1 11
No modes were observed around this region at any
.
Pent-1-ene hydrogenation was shown to be facile over this
catalyst and the times for the complete hydrogenation of the
terminal and internal double bonds within the C5 framework
are presented in Table 1. Concentrating on reactions 1 and 5, it
is seen that the time taken for the hydrogenation of pent-1-ene
stage in the hydrogenation process and the modes associated
with the non-deuterated aliphatic were observed to increase in
intensity throughout the reaction. Therefore, deuterium was
not incorporated into the product(s).
4.3. Toluene-d8 þ 1,3-pentadiene þ H2
The final series of diene hydrogenation reactions involved
introducing the potential modifier before the diene and the
hydrogen gas. The anti-symmetricQCH2 stretch at 3096 cmꢀ1
was removed between 12 and 43 min and the symmetric –CHQ
stretch at 3024 cmꢀ1 was still present at 43 min but completely
removed by 164 min. Fig. 6 presents the infrared spectrum in
the 1100–800 cmꢀ1 region as a function of time. The QCH2
wag at 904 cmꢀ1 is lost between 12 and 43 min. The trans-CH
QCH wag shifts to 969 cmꢀ1, indicating formation of trans-
pent-2-ene, and disappears between 43 and 164 min. The rate
of loss of the terminal double bond is noted at 12 ꢁ 2 min,
which is within general agreement for the loss of this entity in
reactions 1 and 2 (Table 1). The loss of the internal double
bond is, however, reported as 43–164 min, which is broadly
consistent with that seen for reaction 2. Once again, as in
Fig. 7 The infrared spectrum in the 1050–875 cmꢀ1 region as a
function of time for Pd/Al2O3 þ H2 þ cis-2-pent-2-ene: (a) 0, (b) 7,
(c) 11, (d) 16 and (e) 27 min.
5592
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
is comparable to that observed for the terminal double bond of
the diene, performed in the absence of the modifier. In a similar
fashion, comparing reactions 1 and 4, the hydrogenation rate
for cis-pent-2-ene is comparable to that observed for the
internal double bond of the diene. In the following section,
these results are considered along with the preceding reactions
to consider the mechanistic implications for the reaction of 1,3-
pentadiene and unsaturated C5 molecules with hydrogen over
this Pd/Al2O3 catalyst.
work on 1,3-pentadiene hydrogenation over a Pd/Al2O3 cata-
lyst clearly identifies the terminal double bond to react first and
is consistent with the early work of Dobson et al.,28 who
identified hydrogenation to occur on multifunctional molecules
in clear cut stages, with hydrogenation over a Pd/C catalyst
occurring first at the terminal double bond then at the internal
double bond. Such an analysis suggests caution is required in
extrapolation of results from metal single crystals to finely
divided supported metal catalysts. This topic has recently been
addressed by Freund and co-workers, who have shown this
matter is not just an issue of a ‘pressure gap’. Their work on
pentene hydrogenation over model Pd/Al2O3 catalysts29,30
shows that full hydrogenation is possible on the Pd/Al2O3
model catalysts but is not accessible on a hydrogen treated
Pd(111) surface. These studies demonstrate the crucial aspect
of hydrogen supply in hydrogenation reactions,31 with metal
particles providing hydrogen that is not available on the
extended structures of the metal single crystal. However,
morphological effects are also expected to contribute to the
overall catalytic performance and it is noted that in the study
of 1,3-pentadiene hydrogenation by Wells and Wilson that
pentane yields over supported Pd catalysts were low.9 In fact,
these authors suggest that alkane formation may require a
distinct surface site which, presumably, is absent or ineffective
on their catalysts. In the presence of a generous hydrogen
supply, the fully saturated product is readily attained in this
work. Furthermore, the earlier study of 1,3-pentadiene on
supported metal catalysts showed the presence of pent-1-ene,
cis- and trans-pent-2-ene in the product distribution,9 which
contrasts markedly with that seen in this study. In fact, their
results are consistent with a concurrent process, rather than the
consecutive series of hydrogenation steps seen in this work.
Such a discrepancy suggests that the active site distributions
and overall performance of the two catalysts are fundamentally
different. Assuming that the initial formation of hydrocarbo-
naceous overlayers is responsible for the catalyst selectivity
characteristics,14,32,33 it is tentatively assumed that the catalyst
conditioning process is different in both cases and leads to
catalysts with different geometric constraints on the active
sites33 and different hydrogen exchange characteristics at these
sites. Linkage of the active site distribution with catalyst
preparative conditions is highly desirable and would constitute
a major advance in heterogeneous catalysis.
5. Discussion
5.1 Mechanistic implications
With reference to Table 1, the similarity between the time taken
for complete hydrogenation of the terminal and internal
double bonds in 1,3-pentadiene and that observed for cis-
pent-2-ene (internal double bond) and pent-1-ene (terminal
double bond) is strongly suggestive of a two-site model. The
data are consistent with two distinct active sites: terminal
double bonds are hydrogenated at Site a, with a reaction time
of ca. 12 min; and internal double bonds are hydrogenated at
Site b, with a reaction time of ca. 26 min under the conditions
studied. It is possible that steric or stereo-chemical effects could
confer two rates of reaction at the same site, however, this
situation would be expected to be sensitive to cooperative
effects, i.e. rates sensitive to the presence of a co-reactant,
and yet the conversion rates for the monoenes match those for
their diene counterparts. Another possibility is that 2 sites
could be operating simultaneously. This appears not to be the
case, as this scenario would lead to the formation of pent-1-ene
from the hydrogenation of the terminal double bond in 1,3-
pentadiene, which is not observed. Furthermore, the absence of
pent-1-ene in the 1,3-pentadiene hydrogenation profile has
already discounted the possibility of concurrent reactions
originating from a single site (see section 4.1). The 2-site model
remains consistent with a consecutive process. Recent work by
Borodzinski invoking steric arguments has proposed two site
regimes as being active for ethyne hydrogenation over sup-
ported Pd catalysts.26,27 In Borodzinski’s model, the active sites
are defined by the metal sites accessible within a hydrocarbo-
naceous overlayer that forms during an initial conditioning
process. These sites are termed Site A and Site E. Site A affords
a smaller Pd footprint and is associated with ethyne hydro-
genation but excludes ethene adsorption and is therefore
inactive for ethene hydrogenation. All reagents adsorb at the
larger Site E, which is responsible for the hydrogenation of
ethene to ethane.27 Recent work from these laboratories has
shown a similar 2-site model to also provide a comprehensive
description for propyne hydrogenation over supported metal
catalysts.14,25 In that work, Type I sites were assigned as being
responsible for full hydrogenation of propyne to propane,
whereas partial hydrogenation of propyne to propene occurs
at Type II sites.14,25 It is possible that similar distributions of
sites are active on this catalyst under the reaction conditions
presented here, with Sites a and b differentiating between
terminal and internal double bonds on steric grounds. Regret-
fully, at this time, it is not possible to interrogate these sites
directly and therefore it is not possible to determine if Sites A
and E27 or Sites I and II14,15 equate to Sites a and b. However,
similar characteristics are indeed observed and the compari-
sons appear to indicate that conventional catalyst preparations
lead to a distribution of reaction sites under steady state
operation, which can confer specific selectivity characteristics
in multi-stage hydrogenation reactions.
It is possible that the small presence of the cyclopentene
impurity in the 1,3-pentadiene sample is perturbing the penta-
diene hydrogenation characteristics. With reference to Table 1,
this possibility is discounted because cis-pent-2-ene and pent-1-
ene give essentially the same reaction times as that identified
for the related units of the 1,3-pentadiene molecule. The
pentene samples are pure and contain no cyclopentene, indi-
cating that reaction times are unaffected by the presence of
cyclopentene in such small quantities. Rather, the absence of
any cyclopentene feature in the gas phase is consistent with this
minority species being rapidly hydrogenated to cyclopentane,
where it takes no further part in the reaction chemistry.
5.2 The effect of the modifier
Table 1 illustrates how the addition of the chemical modifier,
toluene-d8, can affect the relative hydrogenation characteristics
of the reaction system. Interestingly, reactions 2 and 3 show
Site a to be unaffected by toluene-d8 addition, however, the
modifier has a pronounced effect on Site b. Adding toluene-d8
before the diene (reaction 2), leads to a 4-fold increase in the
time taken to hydrogenate the internal double bond (at Site b).
Changing the order in which the modifier is presented to the
catalyst (reaction 3) does not seem to unduly alter this hydro-
genation rate. Thus, it appears that the modifier has no effect
on Site a but it does retard the hydrogenation rate at Site b,
with no apparent sensitivity to the order of the addition
A major difference in the preceding FTIR study of 1,3-
hexadiene hydrogenation over a Pd/SiO2 catalyst11 compared
to the work performed on a Pd single crystal10 was that the
latter study suggested that the internal double bond was
hydrogenated in advance of the terminal double bond. This
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
5593

View Article Online
process. In this way, toluene-d8 is behaving as a selective
poison, in that its presence preferentially retards hydrogena-
tion of the internal double bond, which occurs at Site b.
The modifier is clearly inducing different characteristics in
the Pd/Al2O3/H2/1,3-pentadiene system to that seen for Pd/
SiO2/H2/1,3-hexadiene.11 Crucially, all of the spectra in the
former show no evidence of D incorporation into the products
(spectra not shown). The implication of this is that the toluene-
d8 is molecularly, not dissociatively, adsorbed. This will then
prevent any H/D exchange occurring on the Pd surface and
then, ultimately, transferring into the C5 molecular unit.
Alternatively, if any dissociation of the toluene was occurring,
presumably centred about the methyl group,11,34–36 it occurs to
such a small extent that it is below the detection limits of the
spectral acquisition conditions used here.
Bond.33 Inspection of product selectivities for alkane hydro-
genolysis over supported Pt catalysts38–42 demonstrated that
carbon retention interfered with only a certain class of site on
the catalyst surface, while other sites remained active.33 The
fact that Site a appears to be preferentially attenuated via a
‘traditional’ deactivation route and yet this site remains un-
perturbed by the presence of a chemical modifier, further
indicates the subtleties of selective adsorption that exists in
competitive adsorption on surfaces offering a variety of sites.
5.4 Stereochemical implications
The reactivity of cis-pent-2-ene has been particularly informa-
tive regarding guidance as to how the C5 entities are reacting
on this surface. Fig. 7 shows that the catalyst efficiently
isomerizes the cis-pent-2-ene to trans-pent-2-ene, which then
goes on to react further to form pentane. Additionally, no cis-
pent-2-ene was observed in the 1,3-pentadiene hydrogenation
studies, despite the diene being a mixture of cis- and trans-
stereoisomers. Direct hydrogenation of cis-1,3-pentadiene
would be expected to produce cis-pent-2-ene.9 It is possible
that this stereoisomer is actually selectively retained by the
catalyst and it is only the trans-1,3-pentadiene that goes on to
react. However, the more likely explanation is thought to be
that the catalyst rapidly isomerises the cis-1,3-pentadiene into
trans-1,3-pentadiene, which then goes on to form trans-pent-2-
ene and ultimately pentane. A postulated reaction mechanism
is presented in Fig. 8. The mechanism assumes the well
established Horiti–Polanyi stepwise hydrogen addition pro-
cess, with half-hydrogenated states10,43 and specifically identi-
fies two sites as being responsible for the partial and full
hydrogenation reactions.
5.3 Catalyst deactivation issues
In order to evaluate if catalyst deactivation might dispropor-
tionally affect the hydrogenation kinetics, reaction 6 was
performed where the hydrogenation rates for 1,3-pentadiene
were monitored after a fresh, activated catalyst had undergone
3 complete diene hydrogenation cycles. The results for the
fourth hydrogenation reaction are presented in Table 1, which
shows a modest increase in hydrogenation times for both sites,
consistent with the reacting system exhibiting a degree of
catalyst deactivation, presumably as a result of a small degree
of additional hydrocarbonaceous laydown on increasing reac-
tion time.14,37 The terminal double bond is hydrogenated
within 18 min which represents an increase in reaction time
of ca. 50% compared to a mean reaction time of ca. 12 min
seen for the fresh catalyst. However, a smaller increase of only
10% is seen with the internal double bond, which requires 28.5
min compared to the 26 min seen when a fresh catalyst is used.
Although the increase in reaction times is relatively minor,
these results suggest that catalyst deactivation is preferentially
affecting Site a, yielding a relatively larger attenuation of
hydrogenation rate at this site. A similar link between carbon
deposits and catalytic activity and selectivity has been noted by
Upon examination of Fig. 8 it is intriguing to consider which
of the two active sites is responsible for the isomerisation
process. Reaction 2 provides some insight into this matter.
With prior addition of toluene-d8, Fig. 5 shows trans-pent-2-
ene to be formed at the same rate as observed for the
unmodified catalyst, i.e. Site a is unperturbed. However, the
modifier has selectively poisoned Site b (see Table 1). If Site b
Fig. 8 A postulated reaction scheme for the surface interactions that correspond to the gas phase reaction profile (Fig. 4) for the hydrogenation of
1,3-pentadiene over a Pd/Al2O3 catalyst. The asterisk (*) refers to surface Pd atoms.
5594
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
9
P. B. Wells and G. R. Wilson, J. Chem. Soc. A, 1970, 2442.
were responsible for the isomerisation step, a retardation in the
production of trans-pent-2-ene would be expected. The fact
that this does not occur suggests that Site a is in fact respon-
sible for the isomerisation reactions (cis-1,3-pentadiene -
trans-1,3-pentadiene and cis-pent-2-ene - trans-pent-2-ene),
in addition to hydrogenation of the terminal double bond.
10 N. Vasquez and R. J. Madix, J. Catal., 1998, 178, 234.
11 S. D. Jackson, S. Munro, P. Colman and D. Lennon, Langmuir,
2000, 16, 6519.
12 P. N. Rylander, in Hydrogenation Methods, Academic Press,
Orlando FL, 1985, p. 36.
13 G. C. Bond and J. C. Rank, Proc. 3rd Intl. Cong. Catal., 1965, 2,
1225.
14 D. R. Kennedy, G. Webb, S. D. Jackson and D. Lennon, Appl.
Catal. A: Gen., 2004, 259, 109.
6. Conclusions
15 C. J. Pouchert, in The Aldrich Library of FT-IR Spectra,
Aldrich Chemical Company, Milwaukee, WI, 1st edn., 1985,
vol. 1.
16 N. G. Mirkin and S. Krimm, J. Phys. Chem., 1993, 97, 13887.
17 D. A. C. Compton, W. O. George and W. F. Maddams, J. Chem.
Soc., Perkin 2, 1977, 1311.
Infrared spectroscopy has been used to study the hydrogena-
tion of a mixture of cis- and trans-1,3-pentadiene over a Pd/
Al2O3 catalyst operating under batch conditions. The main
findings can be summarised as follows:
(i) The reaction is seen to occur as a consecutive process,
with the terminal double bond hydrogenated in advance of the
internal double bond.
(ii) The reaction profile is consistent with the catalyst pre-
senting two distinct reaction sites: hydrogenation of the term-
inal double bond occurs at Site a, whilst Site b is responsible
for hydrogenation of the internal double bond.
(iii) trans-Pent-2-ene is identified as the only reaction inter-
mediate. cis-Pent-2-ene is seen to isomerise to trans-pent-2-ene
before going on to be hydrogenated to pentane.
(iv) The effect of toluene-d8 to act as a chemical modifier was
also investigated and shown to selectively poison Site b, Site a
being unperturbed.
(v) On the basis that modifier studies do not inhibit the
formation of trans-pent-2-ene, Site a is tentatively identified as
the site active for the cis - trans-isomerisation process.
(vi) Repeat hydrogenation studies cause a modest drop in
hydrogenation rates, indicating a small progressive hydrocar-
bonaceous laydown on increased reaction time, with Site a
demonstrating a greater degree of sensitivity than Site b.
18 A. P. Scott and L. Radom, J. Phys. Chem., 1996, 100, 16502.
19 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N.
Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J.
B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M.
C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Ragha-
vachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A.
Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M.
W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J.
A. Pople, GAUSSIAN 03W, (Version 6.0), Gaussian, Inc., Pitts-
burgh PA, 2003.
20 P. C. H. Mitchell, S. F. Parker, A. J. Ramirez-Cuesta and J.
Tomkinson, in Vibrational Spectroscopy with Neutrons with Ap-
plications in Chemistry, Biology, Materials Science and Catalysis,
World Scientific, Singapore, 2004.
21 N. B. Colthup, L. H. Daly and S. E. Wiberley, in Introduction to
Infrared and Raman Spectroscopy, Academic Press, New York,
2nd edn., 1975.
22 L. J. Bellamy, in Advances in Infrared Group Frequencies,
Methuen, London, 1968, p. 42.
23 D. Dolphin and A. Wick, Tabulation of Infrared Spectral Data,
Wiley, New York, 1977.
24 N. Sheppard and D. M. Simpson, Q. Rev., 1952, 1.
25 R. Marshall, G. Webb, S. D. Jackson and D. Lennon, J. Mol.
Catal. A: Chem., in press.
Acknowledgements
ICI are thanked for the award of a Lectureship in Hetero-
geneous Catalysis (DL). The EPSRC, ICI Chemicals and
Polymers Ltd and INEOS Chlor Ltd. are thanked for the
awards of Industrial CASE studentships (DTL and IWS).
Synetix and the University of Glasgow are thanked for the
award of a studentship (TL) and assistance with equipment
provision. Rutherford Appleton laboratories are thanked for
the provision of neutron beamtime and for access to computa-
tional facilities.
26 A. Borodzinski and A. Golebiowski, Langmuir, 1997, 13, 883.
27 A. Borodzinski, Catal. Lett., 1999, 63, 35.
28 N. A. Dobson, G. Erlinton, M. Krishnamurti, R. A. Raphael and
R. G. Willis, Tetrahedron, 1961, 16, 16.
29 A. M. Doyle, Sh. K. Shaikhutdinov, S. D. Jackson and H. J.
Freund, Angew. Chem. Int. Ed., 2003, 42, 5240.
References
1
(a) K. Weissermel and H.-J. Arpe, in Industrial Organic Chemistry,
VCH, Weinheim, 3rd edn., 1997, 59; (b) K. Weissermel and H.-J.
Arpe, in Industrial Organic Chemistry, VCH, Weinheim, 3rd edn.,
1997, 74; (c) K. Weissermel and H.-J. Arpe, in Industrial Organic
Chemistry, VCH, Weinheim, 3rd edn., 1997, 75; (d) K. Weissermel
and H.-J. Arpe, in Industrial Organic Chemistry, VCH, Weinheim,
3rd edn., 1997, 105; (e) K. Weissermel and H.-J. Arpe, in Industrial
Organic Chemistry, VCH, Weinheim, 3rd edn., 1997, 116.
B. J. Joice, J. J. Rooney, P. B. Wells and G. R. Wilson, Discuss.
Faraday Soc., 1966, 41, 223.
P. B. Wells and A. J. Bates, J. Chem. Soc. A, 1968, 3064.
J. J. Philipson, P. B. Wells and G. R. Wilson, J. Chem. Soc. A,
1969, 1351.
A. J. Bates, Z. K. Leszeynski, J. J. Philipson and P. B. Wells, J.
Chem. Soc. A, 1970, 2453.
30 A. M. Doyle, Sh. K. Shaikhutdinov and H. J. Freund, J. Catal.,
2004, 223, 444.
31 D. Lennon, R. Marshall, G. Webb and S. D. Jackson, Stud.
Surface Sci. Catal., 2000, 130, 245.
32 G. Webb, Catal. Today, 1990, 7, 139.
33 G. C. Bond, Appl. Catal. A: Gen., 1997, 149, 3.
34 J. M. Orozco and G. Webb, Appl. Catal., 1983, 6, 67.
35 A. M. Coats, E. Cooper and R. Raval, Surf. Sci., 1994, 307–309,
89.
36 X.-C. Guo and R. J. Madix, J. Catal., 1995, 155, 336.
37 D. Lennon, D. R. Kennedy, G. Webb and S. D. Jackson, Stud.
Surf. Sci. Catal., 1999, 126, 341.
38 G. C. Bond and M. R. Gelsthorpe, J. Chem. Soc., Faraday Trans.
1, 1989, 85, 3767.
39 G. C. Bond and L. Hui, J. Catal., 1993, 142, 512.
40 G. C. Bond and M. R. Gelsthorpe, Catal. Lett., 1989, 3, 359.
41 G. C. Bond and J. C. Slaa, J. Mol. Catal. A: Chem., 1995, 101,
243.
2
3
4
5
6
G. Webb, in Catalysis, Specialist Periodical Reports, ed. C. Kem-
ball and D. A. Dowden, The Chemical Society, London, vol. 2,
1978, p. 155.
7
8
P. R. N. Souza, M. M. Pereira, O. A. C. Antunes, D. A. G.
Aranda and J. W. M. Carneiro, Braz. J. Chem. Eng., 2002, 19, 187.
R. B. Moyes, P. B. Wells, J. Grant and N. Y. Salman, Appl. Catal.
A: Gen., 2002, 229, 251.
42 G. C. Bond and R. H. Cunningham, J. Catal., 1996, 163, 328.
43 M. Boudart and G. Djega-Mariadasson, in Kinetics of Hetero-
´
geneous Catalytic Reactions, Princeton University Press, Prince-
ton, 1984, p. 184.
T h i s j o u r n a l i s & T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 5 5 8 8 – 5 5 9 5
5595