Selective aerobic oxidation of 5-hydroxymethylfurfural in water over solid ruthenium hydroxide catalysts with magnesium-based supports

Article abstract of DOI:10.1007/s10562-011-0707-y

Solid catalyst systems comprised of ruthenium hydroxide supported on magnesium-based carrier materials (spinel, magnesium oxide and hydrotalcite) were investigated for the selective, aqueous aerobic oxidation of the biomass-derived chemical 5-hydroxymethylfurfural into 2,5-furandicarboxylic acid (FDA), a possible plastics precursor. The novel catalyst systems were characterized by nitrogen physisorption, XRPD, TEM and EDS analysis, and applied for the oxidation with no added base at moderate to high pressures of dioxygen and elevated temperatures. The effects of support, temperature and oxidant pressure were studied and optimized to allow a quantitative yield of FDA to be obtained. Graphical Abstract: [Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-011-0707-y

Catal Lett (2011) 141:1752–1760
DOI 10.1007/s10562-011-0707-y
Selective Aerobic Oxidation of 5-Hydroxymethylfurfural
in Water Over Solid Ruthenium Hydroxide Catalysts
with Magnesium-Based Supports
•
Yury Y. Gorbanev Søren Kegnæs
•
Anders Riisager
Received: 22 June 2011 / Accepted: 16 September 2011 / Published online: 1 October 2011
Ó Springer Science+Business Media, LLC 2011
Abstract Solid catalyst systems comprised of ruthenium
hydroxide supported on magnesium-based carrier materials
(spinel, magnesium oxide and hydrotalcite) were investi-
gated for the selective, aqueous aerobic oxidation of the
biomass-derived chemical 5-hydroxymethylfurfural into
2,5-furandicarboxylic acid (FDA), a possible plastics pre-
cursor. The novel catalyst systems were characterized by
nitrogen physisorption, XRPD, TEM and EDS analysis,
and applied for the oxidation with no added base at mod-
erate to high pressures of dioxygen and elevated tempera-
tures. The effects of support, temperature and oxidant
pressure were studied and optimized to allow a quantitative
yield of FDA to be obtained.
dehydration of hexose carbohydrates obtained from ligno-
cellulosic biomass by, e.g. enzymatic hydrolysis [2, 3].
HMF can be readily oxidized to different potentially
important products, such as maleic anhydride [4], 2,5-dif-
ormylfuran (DFF) [5, 6], 2,5-furandicarboxylic acid (FDA)
(Scheme 1) or its dimethyl ester [7–11]. FDA has been
established by the US Department of Energy (DOE) bio-
mass program as one of the 12 chemicals that in the future
can be used as chemical building block from biomass in
biorefineries [12, 13]. In particular, the two carboxylic
groups present in FDA make it a valuable polymer building
block and hence a possible renewable alternative to tere-
phthalic, isophthalic, adipic and other currently used acids,
produced from fossil-based resources [14].
Keywords 5-Hydroxymethylfurfural Á
2,5-Furandicarboxylic acid Á Aerobic oxidation Á
Ruthenium hydroxide catalysts
Ruthenium-based catalysts are generally known for their
aptitude in aerobic oxidation reactions [15–17] including
applications for oxidation of alcohols to produce aldehydes
or ketones. Hence, homogeneous Ru-complex catalysts
have been found to generate aldehydes or ketones in almost
quantitative yields when employed in organic solvents [18,
19] or ionic liquids [20]. A more preferred way to oxidize
HMF involves heterogeneous catalysis, due to ease of
catalyst separation in possible industrial processes [1].
Accordingly, supported ruthenium hydroxide catalysts
have recently been reported to be efficient catalysts for
aerobic oxidation reactions. Ru(OH)x supported on ceria
has been shown to oxidize alcohols to the corresponding
ketones, aldehydes and acids, and also aldehydes to acids
with high yields at 80–140 °C at ambient air pressure [21],
whereas a Co(OH)2 co-promoted catalyst afforded high
activity even at room temperature [22]. Kozhevnikov et al.
[23] performed oxidation of primary alcohols to alde-
hydes using mixed Ru-Co oxide with 95% yield in toluene
at 110 °C under oxygen atmosphere. Similarly, a ruthe-
nium-functionalized nickel hydroxide composite catalyst
1 Introduction
Biomass is a viable feedstock for production of both
chemicals and novel fuels, which eventually can replace
crude oil and gas (fossil feedstocks) as major raw materials
[1]. 5-Hydroxymethylfurfural (HMF) is a product of the
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-011-0707-y) contains supplementary
material, which is available to authorized users.
Y. Y. Gorbanev Á S. Kegnæs Á A. Riisager (&)
Centre for Catalysis and Sustainable Chemistry, Department
of Chemistry, Technical University of Denmark, DK-2800 Kgs.
Lyngby, Denmark
e-mail: ar@kemi.dtu.dk
123

Catalysts with Magnesium-Based Supports
1753
O
HO
O
O
Ru(OH)x/support
H2O, O2, 140oC
O
Biomass
O
HO
OH
HMF
FDA
O
OH
OH
O
HO
HO
OH
support: MgO, HT, MgAl2O4
OH
OH
HO
HO
Scheme 2 Aerobic oxidation of HMF to FDA with supported
Ru(OH)x catalyst in water
Monosaccharides
OH
Fructose
Glucose
species supported on the porous magnesium-containing
supports: MgO, spinel and HT (Scheme 2). The reactions
were conducted in water with molecular oxygen as the
oxidizing agent as a cheap and abundant resource. The
catalysts were characterized and the effect of reaction time,
pressure and temperature on the catalytic performance
were studied and optimized to obtain near quantitative
yield of FDA.
HO
O
O
HMF
[O]
O
O
O
HO
OH
2 Experimental
FDA
Scheme 1 The schematic route for biomass conversion to FDA
2.1 Materials
(Ru/Ni(OH)2) has been used to oxidize alcohols quantita-
tively to aldehydes or ketones in organic solvents at 90 °C
in the presence of molecular oxygen [24].
HMF ([99%, Sigma-Aldrich), 2-furoic acid (98%, Sigma-
Aldrich), levulinic acid (98%, Sigma-Aldrich), formic acid
(FA) (98%, Sigma-Aldrich), ruthenium(III) chloride (pu-
rum, Sigma-Aldrich), hydrotalcite Mg6Al2(CO3)(OH)16Á
4H2O (HT) and spinel MgAl2O4 (purum, Sigma-Aldrich),
sodium hydroxide ([98%, Sigma-Aldrich), MgO (p.a.,
Additionally, Kaneda and coworkers [25, 26] and Miz-
uno and coworkers [27–31] have reported selective aerobic
oxidations of aromatic and aliphatic alcohols to aldehydes
and ketones and amines to amides with Ru(OH)x supported
by alumina, magnetite and hydroxyapatite. Alcohols and
amines were oxidized to produce aldehydes/ketones and
amides/nitriles, respectively, at 80–150 °C under ambient
pressure of O2 in toluene or PhCF3 with yield above 99%.
Furthermore, alumina-supported ruthenium hydroxide have
been used for oxidation of alcohols in a continuous mul-
tifunctional reactor [32].
¨
Riedel-de Haen AG), DFF (98%, ABCR GmbH & Co.KG),
FDA ([99%, Toronto Research Chemicals Inc.), 5-hydro-
xymethyl-2-furancarboxylic acid (HMFCA) ([99%, Toronto
Research Chemicals Inc.) and dioxygen (99.5%, Air Liquide
Denmark) were all used as received.
2.2 Catalyst Preparation and Characterization
Catalyst supports with basic functionality such as, e.g.
hydrotalcite (HT) and hydroxyapatite have also been
investigated [29, 30]. Synthetic Ru–Co–Al and Ru–Al–Mg
HTs have been reported to catalyze aerobic oxidation of
aliphatic and aromatic alcohols in toluene at 60 °C under
ambient dioxygen pressure, producing aldehydes and
ketones in above 90% yield [33, 34]. Ruthenium- and
ruthenium-cobalt-promoted hydroxyapatite gave yields
higher than 99% [35, 36].
4.88 g of support (i.e. MgO, MgAl2O4 or HT) were added
to 143 mL of 8.3 mM aqueous RuCl3 solution (1.19 mmol
Ru). After stirring for 15 min, 28 mL of 1 M NaOH
solution was added and the mixtures were stirred for 18 h.
Then the catalysts were filtered off, washed thoroughly
with water until the filtrates were neutral (colorless filtrates
suggested absence of ruthenium ions) and dried at 140 °C
for 40 h. Approximately 4.9 g of each catalyst was
obtained containing 2.4 wt% Ru.
Recently, we have screened and obtained promising
results for the oxidation of HMF with ruthenium hydroxide
catalysts on various common supports such as, e.g. mag-
nesium oxide and CeO2 [37]. In this work we have elab-
orated the study and investigated the selective oxidation of
HMF to FDA with solid catalysts containing Ru(OH)x
XRPD patterns were recorded using a Huber G670
˚
powder diffractometer (Cu-Ka radiation, k = 1.54056 A)
in the 2h interval 5–100°.
Surface areas were determined by nitrogen adsorption
and desorption measurements at liquid nitrogen tempera-
ture on a Micrometrics ASAP 2020. The samples were
123

1754
Y. Y. Gorbanev et al.
outgased in vacuum at 100 °C for 4 h prior to measure-
ments. The total surface areas were calculated according to
the BET method.
Table 1 Characteristics of supports and supported Ru(OH)x catalysts
Material
BET surface
area (m2/g)
Ru content
(wt%)a
TEM images were recorded on a FEI Tecnai Transition
Electron Microscope at 200 kV with samples deposited on
a carbon support. EDS analysis was performed with an
Oxford INCA system.
MgO
30
27
63
53
8
–
Ru(OH)x/MgO
MgAl2O4
Ru(OH)x/MgAl2O4
HT
0.75 (1), 2.48 (2)
–
2.41 (1), 2.42 (2)
–
2.3 Oxidation Reactions
Ru(OH)x/HT
6
0.25 (1), 7.55 (2)
a
Oxidations were carried out in stirred Parr mini-reactor
autoclaves equipped with internal thermocontrol (T316
steel, TeflonTM beaker insert, 100 mL). In each reaction the
autoclave was charged with 63 mg of HMF (0.5 mmol) and
10 mL of water. The initial HMF concentration (0.05 M)
was chosen based on experimental data on FDA solubility in
water and extrapolation of this data to 140 °C. Subsequently,
the supported 2.4 wt% Ru(OH)x catalyst was added
(0.105 g, 0.025 mmol Ru). The autoclave was flushed and
pressurized with dioxygen (1–40 bar, ca. 1.6–64 mmol) and
maintained at 140 °C for a given period of time under stirring
(700 rpm). After the reaction, the autoclave was rapidly
cooled with ice to room temperature. The reaction mixture
was made alkaline with 1 mL of 1 M NaOH solution before
filtering off the catalyst, or filtered directly without base,
followed by analysis using HPLC (Agilent Technologies
1200 series, Aminex HPX-87H column from Bio-Rad,
300 mm 9 7.8 mm 9 9 lm, flow 0.6 mL/min, solvent
5 mM H2SO4, temperature 60 °C). In all figures where the
product distribution is shown as a function of time each data
point corresponds to an individual reaction run.
The Ru contents are based on Ru:Al atomic ratios provided by EDS.
The values of (1) and (2) are related to the areas numbered 1 and 2 on
Fig. 1 for the respective support
Notably, only agglomerated crystallites of the respective
supports were observed on the TEM images with no
noticeable ruthenium particles even at higher resolution.
EDS analysis of the catalyst samples (performed on the
parts shown in white circles) revealed an uneven distribu-
tion of ruthenium species on the surfaces of the catalysts
with highest basicity, i.e. MgO and HT. The measured Ru
contents are compiled in Table 1.
3.2 Aerobic Oxidation of HMF
Initially, the catalyzed oxidation of HMF to FDA was
investigated with Ru(OH)x/HT catalyst in water in the
absence of added base at 1 bar dioxygen pressure and a
reaction temperature of 140 °C. In Fig. 2 the formation of
products is shown as a function of reaction time.
As seen in the figure, HMF was fully converted after
26 h of reaction and a quantitative yield of FDA was
obtained after a reaction time of 38 h. Importantly, no
product degradation was observed during the examined
time period. Two intermediate oxidation products were
observed; DFF and HMFCA, thus suggesting a competitive
reaction pathway for HMF oxidation with intermediate
products formation followed by oxidation to FDA
(Scheme 3). This pathway is similar to the route previously
established for the gold-catalyzed conversion of HMF [10,
11]. However, the rates of formation and subsequent oxi-
dation of HMFCA and DFF appeared here to be more
comparable, though with initial faster formation of DFF
under the applied reaction conditions (i.e., 140 °C and
1 bar of O2 pressure).
ICP analysis (Perkin Elmer ELAN 6000 with cross-flow
nebulizer and argon plasma) was performed on diluted
post-reaction mixtures and quantified with ICP standard
solutions.
3 Results and Discussion
3.1 Catalyst Characterization
The BET surface areas of the applied support materials and
the prepared catalysts are listed in Table 1. The surface
areas of the catalysts were very much dependent on the
choice of the metal oxide and, as expected, a small
decrease in the surface areas was observed between the
pure supports and the final catalysts. Moreover, X-ray
powder diffraction (XRPD) patterns of the supported cat-
alysts (not shown) revealed exclusively peaks originating
from the respective supports, since the ruthenium content
on the catalysts was too low to allow detection.
Figure 3 shows the distribution of oxidation products
obtained after oxidation of HMF for 1 h with Ru(OH)x/HT
catalyst at oxygen pressures of 1–40 bar and constant
reaction temperature of 140 °C. As shown in the figure, it
proved possible to get full conversion of HMF within 1 h
by increasing the pressure of oxygen. Moreover, it is evi-
dent from the low pressure results that the oxygen pressure
Representative transmission electron microscopy (TEM)
images of the prepared catalysts are presented in Fig. 1.
123

Catalysts with Magnesium-Based Supports
1755
Fig. 1 TEM images of
Ru(OH)x/MgO (a, b), Ru(OH)x/
MgAl2O4 (c, d), and Ru(OH)x/
HT (e, f) catalysts. White circles
represent the areas analyzed by
EDS
dependence was larger on DFF formation than on HMFCA
formation (i.e. higher reaction order of oxygen in the rate
expression for DFF formation), resulting in a higher rate of
oxidation of the alcohol moiety on HMF compared to the
aldehyde group. When performing the reaction at 2.5 bar
for 1 h it proved therefore possible to form DFF with a
relatively high selectivity of about 75%.
of oxygen at different reaction temperatures (Fig. 4). The
reaction temperature drastically affected the performance of
the catalyst which converted essentially all the HMF within
6 h at 100 °C and above. The major product formed at
100 °C was HMFCA (about 80%, with a selectivity of
*85%) while only low amounts of FDA and DFF were
formed (5–8%). However, at higher temperatures only FDA
was observed giving almost quantitative yield at 140 °C. The
absence of DFF after 6 h of reaction time is most likely a
result of easier oxidation of the aldehyde functionality [38].
In order to elucidate the temperature effect on product
formation, a series of experiments were performed with
Ru(OH)x/HT catalyst with a reaction time of 6 h and 2.5 bar
123

1756
Y. Y. Gorbanev et al.
100
80
60
40
20
0
100
80
60
40
20
0
FDA
HMFCA
DFF
HMF
HMF
HMFCA
DFF
FDA
1
20
2.5
5
10
40
0
5
10
15
20
25
30
35
40
Pressure (bar)
Reaction time (hours)
Fig. 3 Product yields in HMF oxidation with Ru(OH)x/HT catalyst in
water (0.05 M HMF, 1 h, 140 °C, 5 mol% Ru to HMF)
Fig. 2 Product yields in HMF oxidation with Ru(OH)x/HT catalyst in
water (0.05 M HMF, 1 bar O2, 140 °C, 5 mol% Ru to HMF)
100
HO
O
FDA
HMFCA
DFF
O
80
HMF
HMF
O
O
HO
O
60
40
20
0
O
O
OH
DFF
HMFCA
O
O
O
OH
80
120
140
100
130
Temperature (° C)
O
O
O
Fig. 4 Product yields in HMF oxidation with Ru(OH)x/HT catalyst in
water (0.05 M HMF, 6 h, 2.5 bar O2, 5 mol% Ru to HMF)
HO
OH
FDA
Scheme 3 Reaction pathway from HMF to FDA by aerobic oxida-
tion via the competitive formation of the two intermediate products
DFF and HMFCA
which was shown to be optimal reactions conditions for the
HT supported catalyst (see Fig. 4). The obtained product
yields as a function of reaction time are presented in
Fig. 5a–c.
The stability of HMF under the applied reaction con-
ditions was confirmed by conducting an experiment with
pure HT support. Here HMF remained essentially uncon-
verted with only ca. 2% of HMF being oxidized and con-
verted to HMFCA (1.3%) and FDA (0.7%), respectively.
To examine the effect of the support on the catalytic
activity for the ruthenium-catalyzed conversion of HMF to
FDA, catalysts with the alternative magnesium supports,
MgO and MgAl2O4, were also prepared and tested in the
oxidation reaction (characteristics of the supports and cat-
alysts are shown in Table 1). The performance of the cat-
alysts were tested under 2.5 bar of oxygen and 140 °C,
The results in Fig. 5 demonstrate that Ru(OH)x sup-
ported on MgO or HT under the applied reaction conditions
was able to convert almost all of HMF to FDA, as
expected. However, for the Ru(OH)x/MgAl2O4 catalyst
(Fig. 5c) the activity was lower, resulting in a yield of 60%
of FDA after 42 h. Furthermore, a substantial amount
(35%) of FA was formed after 42 h with this catalyst. FA,
in accordance with literature, may originate from degra-
dation of HMF and FDA [39]. Interestingly, no degradation
products were observed when the more alkaline supports
MgO and HT were used. Recently, Corma and co-workers
reported that usage of ceria-supported gold catalyst in
123

Catalysts with Magnesium-Based Supports
1757
Reactions were performed at 140 °C with a reaction time
of 1 h (Fig. 6).
100
(a)
Using the Ru(OH)x/MgAl2O4 catalyst the yield of
HMFCA and FDA increased when the oxygen pressure was
increased, especially up to 5 bar as also found for the
Ru(OH)x/HT catalyst (see Fig. 3). Notably, however, an
increase in dioxygen pressure from 1 to 2.5 bars resulted in
significantly lower formationof FA—from 15 to 0.1%. Based
on this observation, the time dependence experiment with the
spinel-based catalyst was redone at 5 bar (Fig. 7) instead of
2.5 bar (see Fig. 5c) in order to minimize the byproduct.
From Figs. 5c, and 7 it is clear that at both 2.5 and 5 bar
of dioxygen pressure FA formation initiated as the reaction
progressed and high relative concentration of the products
(i.e. HMFCA, DFF and FDA) accumulated. This indicated
that a gradual increase in acidity of the media due to FDA
formation could induce furan cycle decomposition. To
understand these results in more detail, a control experiment
with only Ru(OH)x/MgAl2O4 catalyst and FDA (10 mL
H2O, 0.078 g (0.5 mmol) FDA, 2.5 bar O2, 140 °C, 16 h)
was performed to test the stability of FDA in the presence of
the catalyst. The experiment revealed that 78% of the initial
amount of FDA remained unconverted after the 16 h of
reaction whereas partial degradation led to formation of
18% FA. This clearly established the FA—at least par-
tially—to originate from FDA, and possibly also from HMF
or the intermediate products HMFCA and DFF. Accord-
ingly, it is possible to limit the formation of FA in the
reaction when using Ru(OH)x/MgAl2O4 catalyst by apply-
ing short reaction time and high relative oxygen pressure.
Application of different magnesium-containing supports
resulted in different amounts and distributions of products
(see Fig. 5a–c). To understand this difference post-reaction
solutions from experiments with each of the catalysts were
80
60
40
20
0
HMF
HMFCA
FDA
0
0
0
5
10
15
20
Reaction time (hours)
100
80
60
40
20
0
(b)
HMF
HMFCA
DFF
FDA
5
10
15
20
25
Reaction time (hours)
100
80
60
40
20
0
(c)
HMF
HMFCA
DFF
FDA
FA
FDA
HMFCA
DFF
FA
60
50
HMF
40
30
20
10
0
10
20
30
40
50
Reaction time (hours)
Fig. 5 Product yields in HMF oxidation with Ru(OH)x catalysts in
water supported on a MgO, b HT or c MgAl2O4 (0.05 M HMF,
2.5 bar O2, 140 °C, 5 mol% Ru to HMF)
HMF oxidation in alkaline aqueous media led to formation
of both ring-opening degradation products and 2-furoic
acid [40].
1
40
2.5
5
10
20
Pressure (bar)
To increase the yield of FDA and limit the formation of
FA when using Ru(OH)x/MgAl2O4, the effect of oxygen
pressure on the HMF oxidation was further investigated.
Fig. 6 Product yields in HMF oxidation with Ru(OH)x/MgAl2O4
catalyst in water (0.05 M HMF, 1 h, 140 °C, 5 mol% Ru to HMF)
123

1758
Y. Y. Gorbanev et al.
100
80
60
40
20
0
post-reaction solution. Similarly, the high pH value of 10 in
the post-reaction solution with MgO support can be asso-
ciated to its enhanced dissolution under the reaction con-
ditions (Table 2, entry 2).
HMF
HMFCA
DFF
FDA
FA
As the basicity of the respective support decreases in the
order MgO [ HT [ MgAl2O4 [43], the absence of the
DFF product in the HMF oxidation reaction with MgO
support (Fig. 5a) might be further explained by the highly
basic media (Table 2), possibly facilitating Cannizzaro
reaction of the dialdehyde.
In contrast to the HT and MgO supports, the spinel
support remained significantly more stable under the
reaction conditions permitting only a small amount (0.9%)
of the magnesium to dissolve in the acidic post-reaction
solution. Accordingly, the formation of FA when using
MgAl2O4 support can be rationalized to be related to lower
stability and higher degradation of FDA and HMF in acidic
media. In line with this, no degradation of substrate were
observed in reactions with catalysts based on the HT and
MgO supports, since the solutions here were maintained at
high pH throughout the reactions due to partial dissolution
of the supports. Additionally, the results of the XRPD
analysis did not reveal any change in the spinel structure
before and after its employment in the reaction (see Sup-
plementary information, Fig. S1).
0
5
10
15
20
Reaction time (hours)
Fig. 7 Product yields in HMF oxidation with Ru(OH)x/MgAl2O4
catalyst in water (0.05 M HMF, 5 bar O2, 140 °C, 5 mol% Ru to
HMF)
analyzed by ICP for magnesium and ruthenium content
(Table 2).
The ICP analysis confirmed presence of magnesium ions
in all post-reaction solutions. Especially, for the HT- and
MgO-supported catalysts (Table 2, entries 1 and 2) the
amount of leached magnesium was high (26–38%). Nota-
bly, the concentration of Mg2?-ions leached from the HT-
supported catalyst corresponded to about the amount (i.e.
concentration) of FDA formed, thus indicating that the HT
support acted as a solid base in the reaction and ionized the
FDA to form a Mg-salt which most likely proved more
stable towards degradation. Single crystal XRD analysis of
the isolated Mg-FDA has recently been reported [41].
Interestingly, HT-supported gold nanoparticle catalyst was,
in contrast, recently reported to be reusable and apparently
stable in the oxidation of HMF in water under ambient
oxygen pressure and elevated temperature [42].
In Table 2 the measured amounts of ruthenium in the
post-reaction solutions are also reported. Importantly, only
an extremely small amount (0.01–0.02%) of the ruthenium
metal on the catalysts was dissolved in the examined post-
reaction solutions, thus making especially the Ru(OH)x/
MgAl2O4 catalyst prone for re-use. Hence, an experiment
was performed where this catalyst was recovered by fil-
tration, washed with base and water (to remove any FDA
0,020
For the MgO support a similar tendency was also
observed. The fact that HT dissolved during reaction and
neutralized some of the formed FDA also explains the
otherwise unexpected neutral pH value measured of the
HMF
FDA
0,016
0,012
0,008
0,004
0,000
Table 2 ICP analysis of the post-reaction solutions from the aerobic
HMF oxidation using supported Ru(OH)x catalysts (0.05 M HMF,
2.5 bar O2, 140 °C, 5 mol% Ru to HMF)
Entry Support
[Mg2?
(g/L)
]
Mg
[Run?
dissolved (mg/L) dissolved
]
Ru
pHd
(%)c
(%)c
1a
2a
3b
a
HT
0.980
1.590
26
38
0.9
0.030
0.035
0.046
0.013
0.015
0.020
7
10
2
MgO
1
2
3
4
MgAl2O4 0.157
Reaction run
Measured after 6 h of reaction
Measured after 42 h of reaction
Based on the overall element loading
Fig. 8 Rates of HMF conversion and FDA formation per gram of the
catalyst in the recycling of Ru(OH)x/MgAl2O4 catalyst in water
(0.05 M HMF, 5 bar O2, 140 °C, 5 mol% Ru to HMF, 6 h of reaction
time)
b
c
d
Measured pH values of the post-reaction solutions
123

Catalysts with Magnesium-Based Supports
1759
¨
3. Schafer T, Borchert TW, Nielsen VS, Skagerlind P, Gibson K,
Wenger K, Hatzack F, Nilsson LB, Salmon S, Pedersen S, Heldt-
Hansen HP, Poulsen PB, Lund H, Oxenbøll KM, Wu GF,
Pedersen HH, Xu H (2007) Adv Biochem Eng Biotechnol 105:59
4. Du Z, Ma J, Wang F, Liu J, Xu J (2011) Green Chem 13:554
precipitated on the surface of the catalyst after cooling
down the reaction mixture) and re-used (Fig. 8). As seen
from the results, the spinel-supported ruthenium catalyst
preserved its initial activity clearly making this stable
heterogeneous Ru(OH)x oxidation catalyst interesting for
further investigations.
´
5. Navarro OC, Canos AC, Chornet SI (2009) Top Catal 52:34
6. Ma J, Du Z, Xu J, Chu Q, Pang Y (2011) ChemSusChem 4:51
´
7. Grabowski G, Lewkowski J, Skowronski R (1991) Electrochim
Acta 36:1995
8. Lilga MA, Hallen RT, Hu J, White JF, Gray MJ (2008) US Patent
20080103318, 1 may 2008
9. Casanova O, Iborra S, Corma A (2009) J Catal 265:109
10. Gorbanev YY, Klitgaard SK, Woodley JM, Christensen CH,
Riisager A (2009) ChemSusChem 2:672
11. Taarning E, Nielsen IS, Egeblad K, Madsen R, Christensen CH
(2008) ChemSusChem 1:75
12. Werpy T, Petersen G (2004) Top Value Added Chem Biomass
1:26 http://www.osti.gov/bridge
13. Bozell JJ, Petersen GR (2010) Green Chem 12:539
14. Boisen A, Christiansen TB, Fu W, Gorbanev YY, Hansen TS,
4 Conclusions
Supported catalysts with catalytically active Ru(OH)x
species deposited on the three magnesium-based supports
HT, MgO and spinel (MgAl2O4), have been applied for
aerobic oxidation of HMF to FDA in water without added
base. All three catalysts were found to effectively catalyze
the oxidation of HMF. However, both HT and MgO sup-
ports dissolved partly under the reaction conditions liber-
ating significantly amounts of Mg2? ions, thus making the
mixtures alkaline. This resulted in formation of Mg-FDA
salts stabilized against further degradation. The spinel
support, on the other hand, remained stable under the
reaction conditions which allowed performing the oxida-
tion reaction under base free conditions.
˚
Jensen JS, Klitgaard SK, Pedersen S, Riisager A, Stahlberg T,
Woodley JM (2009) Chem Eng Res Des 87:1318
15. Dijksman A, Arends IWCE, Sheldon RA (2001) Platinum Metals
Rev 45:15
16. Chang S, Lee M, Ko S, Lee PH (2002) Synth Commun 32:1279
17. Vanover E, Huang Y, Xu L, Newcomb M, Zhang R (2010) Org
Lett 12:2246
18. Zhai Y, Liu H, Liu B, Liu Y, Xiao J, Bai W (2007) Trans Met
Chem 32:570
19. Lei ZQ, Kang QX, Bai XZ, Yang ZW, Zhang QH (2005) Chin
Chem Lett 16:846
20. Farmer V, Welton T (2002) Green Chem 4:97
21. Vocanson F, Guo YP, Namy IL, Kagan HB (1998) Synth Com-
mun 28:2577
22. Ji H, Mizugaki T, Ebitani K, Kaneda K (2002) Tetrahedron Lett
43:7179
23. Musawir M, Davey PN, Kelly G, Kozhevnikov IV (2003) Chem
Commun 1414
24. Venkatesan S, Kumar AS, Lee J-F, Chan T-S, Zen J-M (2009)
Chem Commun 1912
25. Yamaguchi K, Mori K, Mizugaki T, Ebitani K, Kaneda K (2000)
J Am Chem Soc 122:7144
26. Mori K, Yamaguchi K, Mizugaki T, Ebitani K, Kaneda K (2001)
Chem Commun 461
27. Yamaguchi K, Mizuno N (2003) Chem Eur J 9:4353
28. Kotani M, Koike T, Yamaguchi K, Mizuno N (2006) Green
Chem 8:735
29. Kim JW, Yamaguchi K, Mizuno N (2008) Angew Chem Int Ed
47:9249
30. Mizuno N, Yamaguchi K (2008) Catal Today 132:18
31. Yamaguchi K, Mizuno N (2010) Synlett 16:2365
32. Bavykin DV, Lapkin AA, Kolaczkowski ST, Plucinski PK (2005)
Appl Catal A 288:175
33. Kaneda K, Yamashita T, Matsushita T, Ebitani K (1998) J Org
Chem 63:1750
34. Matsushita T, Ebitani K, Kaneda K (1999) Chem Commun 265
35. Opre Z, Grunwaldt J-D, Maciejewski M, Ferri D, Mallat T,
Baiker A (2005) J Catal 230:406
The reported data suggests that the reaction pathway for
aerobic oxidation of HMF to FDA with the Ru(OH)x
supported catalysts proceed via relatively slow initial
competitive oxidation to DFF and HMFCA (Scheme 3).
The subsequent oxidations to form the product are fast
since no other intermediates (e.g. 5-formylfuran-2-car-
boxylic acid) were observed.
Importantly, only very low amounts (\0.02%) of the
ruthenium metal inventory was found to dissolve from the
catalysts (irrespectively of the support dissolution) under
the applied reaction conditions. Combined with the
observation that Ru(OH)x/MgAl2O4 preserved its activity
upon reuse, makes this and analogous catalyst systems
based on stable supports attractive alternatives to present
aerobic HMF oxidation catalysts based on metal nanopar-
ticles (e.g. gold catalysts), which often is less active upon
reuse due to particle sintering [10, 44].
Acknowledgments The authors thank Bodil Holten (Centre for
Catalysis and Sustainable Chemistry, DTU Chemistry) for experi-
mental assistance. The work was supported by The Danish National
Advanced Technology Foundation and Novozymes A/S.
36. Opre Z, Grunwaldt J-D, Mallat T, Baiker A (2005) J Mol Catal A
242:224
37. Gorbanev YY, Kegnæs S, Riisager A (2011) Top Catal. doi:
10.1007/s11244-011-9754-2
References
1. Rinaldi R, Schu¨th F (2009) Energy Environ Sci 2:610
2. Park J, Seyama T, Shiroma R, Ike M, Srichuwong S, Nagata K,
Arai Sanoh Y, Kondo M, Tokuyasu K (2009) Biosci Biotechnol
Biochem 73:1072
38. Marsden C, Taarning E, Hansen D, Johansen L, Klitgaard SK,
Egeblad K, Christensen CH (2008) Green Chem 10:168
39. Lewkowski J (2001) ARKIVOC (i) 17–54
40. Casanova O, Iborra S, Corma A (2009) ChemSusChem 2:1138
123

1760
Y. Y. Gorbanev et al.
41. Schau-Magnussen M, Gorbanev YY, Kegnæs S, Riisager A
(2011) Acta Cryst C67:m327
42. Gupta NK, Nishimura S, Takagaki A, Ebitani K (2011) Green
Chem 11:824
43. Di Cosimo JI, Diez VK, Xu M, Iglesia E, Apesteguia CR (1998)
J Catal 178:499
44. Hashmi ASK, Hutchings G (2006) Angew Chem Int Ed 45:7896
123