Full Papers
can be observed in Figure 3 that standard deviations from the
average value are similar for the different techniques and Mo
loadings. Moreover, it is clear that effective interaction be-
tween Pt and Mo is only achieved at Mo/Pt ratios lower than
0.3. The advantage of selective deposition of organometallic
precursor of oxophilic metal onto the pre-reduced noble metal
in the parent catalyst that the CSR method offers is reduced
when there is weak interaction between the precursor and re-
duced metal surface. Although the differences in composition
distributions for CSR and IWI catalysts were significant for
RhMo/C system,[17] which is characterized by a strong interac-
tion between the two metals, they appear to be less significant
for the PtMo/C system with a weaker interactions.
that do not contribute to improving the catalytic reactivity.
This result demonstrates the superiority of CSR catalysts as
compared to IWI catalysts. The CSR catalysts are more suitable
as model bimetallic materials for the study of the effect of
component ratio and in the elucidation of nature of active
sites in conjunction with theoretical studies. Additionally, CSR
catalysts are beneficial from an economical point of view,
where targeted deposition results in saving of the expensive
oxophilic promoters such as rhenium.
Specific reactivity plots for PtMo/C catalysts are presented in
Figure 4b. These plots of specific rate versus Mo level also
demonstrate a bell-shape curve for the CSR catalysts, indicat-
ing the creation of bi-functional active sites as Mo is progres-
sively deposited, followed by the blocking of the Pt surface at
higher Mo levels. Thus, it can be concluded that even with
a lower affinity of Mo for Pt, the deposition of Mo onto the Pt
nanoparticle surface can be achieved to create a maximum in
the number of active sites for this system. Although the CSR
catalysts show a bell-shaped curve in the plot of specific rate
versus composition, the specific rate curves for the IWI cata-
lysts are rather broad, and a clear maximum in the rate is not
observed. This behavior suggests inefficient alloying in the IWI
catalysts, resulting in a wide range of composition distribution
such that the promotional effect of increasing Mo content is
less responsive to composition change. Therefore,
even for a system with a weaker affinity between the
bimetallic components, CSR catalysts demonstrate
more efficient alloying and result in a more uniform
nanoparticle composition as compared to IWI cata-
lysts.
To further study the differences between CSR and IWI cata-
lysts, the catalytic activities of the different catalysts for the
RhMo/C and PtMo/C systems were evaluated in the selective
CO hydrogenolysis of 2-(hydroxymethyl)tetrahydropyran
(HMTHP) to 1,6-hexanediol (Scheme 1). In our previous work, it
was concluded that the most active catalysts for this reaction
consist of nanoparticles with bi-functional active sites com-
posed of small ensembles of the highly reducible metal (Rh or
Pt for hydrogenation) and small ensembles of the oxophilic el-
ement (Mo or Re for the generation of acidity) in close proxim-
ity.[17]
The differences between catalyst synthesis meth-
ods are further compared based on turnover frequen-
cies (rates normalized by exposed noble metal) (Fig-
ure 5a and b). For the RhMo system, a continuous
upward trend in TOF for the catalysts prepared by
Scheme 1. Reaction Scheme of 2-(hydroxymethyl)tetrahydropyran (HMTHP) hydrogenoly-
sis.
CSR can be attributed to the fact that the specific
rate increases with Mo loading, while the number of
exposed Rh surface atoms decreases. This behavior is
Figure 4a plots the specific rate per gram of catalyst for the
RhMo/C system with respect to the Mo content (ICP value).
The plot of specific rate versus Mo level shows a bell-shaped
curve. It can be seen that the addition of around 10 at% Mo
(ICP value) using the CSR approach was sufficient to achieve
the maximum promotion, beyond which there was an accumu-
lation of Mo at the surface that did not contribute to the for-
mation of active site but rather reduced the number of exist-
ing active sites by covering the Rh surface. An important ob-
servation is that the specific rate maxima for IWI1 and IWI2 are
displaced to a higher Mo loading (~25 at% Mo, ICP value). This
displacement is a strong indication of efficient alloying and
uniformity in the CSR catalysts, where targeted deposition of
oxophilic component resulted in a smaller amount of Mo
being sufficient to achieve similar conversion rates. Composi-
tional non-uniformity for the IWI catalysts required a higher
loading of oxophilic promoter to create the same number of
bi-functional active sites, because inherently nanoparticles rich
in either component or monometallic particles were formed
an indication of effective alloying by means of selective depo-
sition of Mo onto the metal nanoparticles. However, different
trends were obtained for those catalysts prepared by incipient
wetness impregnation (IWI1 and IWI2). In this case, the initial
increase in specific rate is identical to the CSR system, where
the rate increases and the number of exposed Rh surface
atoms decreases, resulting in an increase in TOF. However, on
further increasing the Mo content, the TOF decreased. This be-
havior is a further evidence of broader composition distribu-
tions for the IWI catalysts, suggesting the presence of Rh rich
particles along with surface Mo rich domains, the former con-
tributing to a smaller reduction in the exposed Rh but not con-
tributing to an increase in activity. A bell shaped curve can be
also observed for the CSR-PtMo/C catalysts, as a result of the
broader composition distribution in comparison with CSR-
RhMo/C catalysts.
Selected spent catalysts were also characterized after reac-
tion. Samples after reaction were filtered, washed with milli-Q
water and acetone, and dried before CO chemisorption analy-
ChemCatChem 2015, 7, 3881 – 3886
3884
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim