CHEMSUSCHEM
FULL PAPERS
explained as a result of the increase in the acid strength of
PTA/MIL-100(Cr) on the one hand and the decrease in the sur-
face area of PTA/MIL-100(Cr) on the other hand as PTA load-
ings increase. Due to the fact that glucose hydrogenation to
sorbitol occurs through a surface-type catalysis,[28] decreasing
the surface area of the catalyst would presumably result in de-
creased sorbitol yields. Additionally, with Ru-PTA/MIL-100(Cr)
(24.2 wt% PTA) as the catalyst, a further dehydration of sorbitol
to sorbitan and isosorbide was also observed, suggesting
a strong acidity of Ru-PTA/MIL-100(Cr) with high PTA loadings,
which may also result in decreased sorbitol yields.
cellulose is generally controlled by the catalyst acid strength,
and therefore, Ru-PTA/MIL-100(Cr) shows the highest catalytic
activity in the Ru-POM/MIL-100(Cr) series. In the case of
Ru-PMA/MIL-100(Cr), because of the relatively lower thermal
stability, higher oxidation potential, and lower hydrolytic stabil-
ity of PMA relative to tungsten heteropolyacids,[50,51] molybde-
num heteropolyacids are frequently deactivated due to their
reduction by the organic reaction medium; it is not uncom-
mon for them to show lower activities than those expected
from their acid strengths.[50,51] Accordingly, on the basis of
a combination of acid strength and stability of catalyst, catalyst
efficiency in the hydrolytic hydrogenation of cellulose decreas-
es in the order: Ru-PTA/MIL-100(Cr)>Ru-STA/MIL-100(Cr)>
Ru-PMA/MIL-100(Cr) (Table 5, runs 1–5).
Apart from PTA/MIL-100(Cr), both STA/MIL-100(Cr) and
PMA/MIL-100(Cr) were also evaluated as supports to investi-
gate the effect of POMs on cellulose conversion. Both
Ru-STA/MIL-100(Cr) and Ru-PMA/MIL-100(Cr) resulted in rela-
tively low sorbitol yields of 27.6% and 11.2%, respectively,
under the same reaction conditions (Table 5, runs 4 and 5).
Therefore, Ru-PTA/MIL-100(Cr) (3.2 wt% Ru, 16.7 wt% PTA) was
considerably more active and selective (in terms of sorbitol for-
mation) than Ru-STA/MIL-100(Cr) and Ru-PMA/MIL-100(Cr)
(Table 5, runs 1–5). Zhang et al. discussed the effect of W in
supported tungsten carbide and bimetallic catalysts:[29–31] high
conversions and a shift in selectivity to EG in the presence of
W during cellulose conversion into EG were observed. Further-
more, Palkovits et al. revealed a cooperative effect between W
and Ru for the dual functional catalyst system Ru/C and PTA
during the conversion of cellulose into sugar alcohols.[27]
Considering selectivities of up to 63.2% for C6 sugar alcohols
(sorbitol and mannitol) over Ru-PTA/MIL-100(Cr) (Table 2,
run 6), in principle our results seemed to agree with the above
proposition: the presence of W enhances the selectivity for
certain products, a fact which may be attributed to coopera-
tive effects between tungsten, ruthenium, and the substrate.
Thus, the differences in product selectivity between STA, PMA,
and PTA could originate from differences in their interaction
with the substrate and ruthenium. Further investigations will
be carried out to elucidate underlying principles. Therefore,
with a constant Ru loading, the catalytic performance of Ru-
POM/
Different amounts of Ru were loaded on the
PTA/MIL-100(Cr) (16.7 wt% PTA) support, and the effect of Ru
loadings on sorbitol yields was also studied (Table 5, runs 2,
6–10). The use of Ru-PTA/MIL-100(Cr) with a Ru loading of
1.2 wt% produced sorbitol in 23.0% yield with a cellulose con-
version of 78.2%. Furthermore, a combined analysis of data
from different loading amounts suggested that Ru-PTA/
MIL-100(Cr) with a Ru and PTA loading of 3.2 and 16.7 wt%, re-
spectively, showed the highest sorbitol yield and the highest
selectivity for sorbitol. These results were in agreement with
the Ru loading effect on the hydrolytic hydrogenation of
cellobiose.
Furthermore, data in Table 1 (runs 2, 11–13) and Table 5
(runs 2, 14–16) revealed that ruthenium significantly promoted
cellobiose and cellulose conversion. However, the acid site
density of Ru/MIL-100(Cr) (1.62 mmolgꢀ1) and Ru-PTA/
MIL-100(Cr) (3.53 mmolgꢀ1) only increased slightly relative to
their corresponding supports (Table 4, runs 1 and 3). The in-
crease in cellobiose and cellulose conversion thus can presum-
ably be related to hydrolytic hydrogenation and hydrogenoly-
sis processes of cellobiose and cellulose catalyzed by the
ruthenium catalyst. As a result, both sorbitol, formed by a hy-
drolytic hydrogenation process, and 3-b-gluco-pyranosyl-d-glu-
citol,[10,14,24] generated through a hydrogenolysis process, were
observed in the metal-catalyzed reaction.
MIL-100(Cr) decreased in the order: Ru-PTA/MIL-100(Cr)
(16.7 wt% PTA)>Ru-PTA/MIL-100(Cr) (24.2 wt% PTA)>Ru-PTA/
MIL-100(Cr) (8.3 wt% PTA)>Ru-STA/MIL-100(Cr) (23.3 wt%
STA)>Ru-PMA/MIL-100(Cr) (17.1 wt% PMA). The above results
are, however, quite different from the acid-strength sequence
of the support POM/MIL-100(Cr) shown in Table 4, which fur-
ther indicates the delicate balance between acid strength, sur-
face area, and metal of catalyst.
Fukuoka et al. reported that both supported Pt and Ru cata-
lysts were effective for the conversion of cellulose into sugar
alcohols.[11,15] In our case, relative to Pt, Rh, and Pd, supported
Ru was more active and selective for the formation of sorbitol,
and Ru-PTA/MIL-100(Cr) resulted in the highest yield in sorbitol
among the examined catalysts (Table 5, runs 2, 11–13). Addi-
tionally, both supports MIL-100(Cr) and PTA/MIL-100(Cr) were
used as catalysts in blank experiments for cellulose conversion
into sorbitol. No sugar alcohols were observed and only trace
amounts of glucose were detected by hydrolysis of cellulose
with MIL-100(Cr) and PTA/MIL-100(Cr) as a Lewis acid and
a Brønsted acid, respectively, under the above conditions
(Table 5, runs 14 and 15). Using Ru/MIL-100(Cr) as a catalyst,
a low sorbitol yield of 13.5% was observed, which was in
sharp contrast with Ru-PTA/MIL-100(Cr) producing sorbitol
yields of 49.0% (Table 5, runs 16 and 2, respectively). These re-
sults demonstrated that a good balance between the two cata-
Notably, PMA/MIL-100(Cr) showed a relatively high acid
strength (Table 4), whereas Ru-PMA/MIL-100(Cr) had the lowest
yield in sorbitol (Table 5, runs 1–5). Previous studies indicate
that the rate of cellulose hydrolysis is strongly dependent on
acid concentration.[52] The catalytic activity of heteropolyacids,
both in homogeneous and heterogeneous systems, usually
parallels their acid strength (PTA>STA>PMA).[50,51] Being
a stronger acid and, therefore, a more efficient proton donor,
PTA usually exhibits higher catalytic activities than other
heteropolyacid catalysts. Furthermore, the hydrolysis rate of
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 12
&
7
&
ÞÞ
These are not the final page numbers!