10.1002/cctc.201601224
ChemCatChem
FULL PAPER
outgassed at 350 ºC for 3 h. Total specific surface area calculations were
made using the Brunauer-Emmett-Teller (BET) equation.
Although not showing great influence in the rate of hydrogenation
of glucose, the presence of oxygen surface groups on the catalytic
support had an improving effect on the selectivity to sorbitol from
glucose, which increased more than 10% by using Ru/CNT2
comparatively to Ru/CNT0.
Despite the results obtained performing both individual steps of
conversion of cellulose, the original multi-walled carbon
nanotubes stand as the best support to Ru nanoparticles in order
to selectively catalyze the direct conversion of cellulose to sorbitol.
The qualitative and quantitative determination of oxygenated surface
functional groups was performed by temperature programmed desorption
– mass spectrometry (TPD-MS). CO and CO2 TPD profiles were obtained
in a fully automated AMI-300 equipment (Altamira Instruments); the
sample was submitted to a 5 ºC·min-1 heating to 1100 ºC under helium
flow (25 cm3·min-1). CO and CO2 signals were monitored by a quadrupole
mass spectrometer (Dymaxion 200, Ametek).
Transmission electron microscopy (TEM) analysis were obtained using a
JEOL2010F instrument.
A measurement of about 250 ruthenium
nanoparticles was carried out from TEM micrographs, in order to obtain
information about the metal particle size distributions. The average
diameter was calculated by ꢀꢁ ꢂ ∑ ꢄ, where ꢆꢇ is the number of particles
Experimental Section
ꢃ ꢅ
ꢄ
∑
ꢅ
ꢄ
with diameter ꢀꢇ . Metal dispersion was obtained using the following
Chemicals and Materials.
∑
ꢉꢊꢄ
equation: ꢈ ꢂ ∑ , where
ꢌ and ꢌꢎꢇ correspond to the number of atoms
ꢍꢇ
Nanocyl-3100 multi-walled carbon nanotubes with an average outside
diameter of 9.5 nm, average length of 1.5 µm and carbon purity higher than
95% were provided by Nanocyl. The metal precursor ruthenium (III)
chloride (RuCl3 99.9%, Ru 38% min) and microcrystalline cellulose were
supplied by Alfa Aesar. Glucose (99.5%) was obtained from Sigma.
Sulphuric (> 95%) and nitric (≥ 65%) acids were purchased from Fisher
Chemical and Sigma-Aldrich, respectively. All solutions were prepared in
ultrapure water obtained in a Milli-Q Millipore System with a conductivity
of 18.2 µS·cm-1.
ꢉꢋꢄ
on the surface and the total amount of atoms of each particle, respectively.
Further details can be found elsewhere [8]
.
Catalyst Evaluation.
The catalytic experiments were performed in a 1000 mL stainless steel
reactor (Parr Instruments, USA Mod. 5120). In a typical hydrolytic
hydrogenation experiment the reactor was loaded with 750 mg of cellulose
(previously ball-milled for 4 h at 1200 rpm), 300 mg of catalyst and 300 mL
of water. After that, the reactor was flushed three times with N2 to remove
air. Then, the mixture was heated under nitrogen atmosphere to 205 ºC
and stirred at 150 rpm. When the desired temperature was achieved, the
reaction was initiated by pressurizing the system with hydrogen (50 bar;
this pressure was kept constant during the entire experiment). The reaction
was stopped after 5 h.
In a typical hydrogenation experiment, glucose was used as substrate
instead of ball-milled cellulose and the reaction was stopped after 1 h. In
a typical hydrolysis experiment, the supports (without the impregnation of
Ru) were used as catalysts and the system was not pressurised with
hydrogen but kept under nitrogen atmosphere. In this case, time t = 0 was
considered when the reaction mixture achieved the desired temperature
(205 ºC); the reaction was carried out for 5 h.
The filtered water-soluble products were analysed by high performance
liquid chromatography (HPLC) using an Alltech OA-1000 ion exclusion
column and a refractive index (RI) detector. A 0.005 mol·L-1 H2SO4 mobile
phase was used at a 0.5 mL·min-1 flow rate and an injection volume of 30
µL was selected. C2-C6 polyols, including ethylene glycol (EG), propylene
glycol (PG), glycerol, sorbitol, erythritol, mannitol and glucose were
detected. The yield (Y) of each product was defined as the ratio of the
number of moles of product formed to the total number of moles of
substrate (cellulose or glucose) initially present, taking the stoichiometry
into account; the selectivity (S) was defined as the ratio of the yield of that
product to the conversion of substrate. Total organic carbon (TOC) data
Preparation Procedures.
Commercial Nanocyl-3100 multi-walled carbon nanotubes (sample CNT0)
were submitted to different chemical and thermal treatments in order to
obtain materials with different surface chemical properties, while
maintaining the original textural properties as far as possible:
- Oxidation treatment: An adequate amount of sample CNT0 was
introduced in a Soxhlet extraction apparatus connected to a boiling flask,
with a HNO3 6 mol·L-1 solution, and to a condenser. The acid solution was
heated to boiling temperature, and the reflux was stopped after 3 h. The
oxidized material was subsequently washed with distilled water until
neutral pH of the rising water was attained, and then dried at 110 ºC for 24
h (sample CNT1). In order to produce carbon nanotubes with a larger
amount of surface groups giving a stronger acid character, the original
sample (CNT0) was also oxidized directly in a Pyrex round-bottom flask
containing the acid solution (sample CNT2), instead of using the Soxhlet
extraction apparatus.
- Thermal treatment: Sample CNT2 was used for the thermal
treatments since it is important that the starting material presents a large
amount of surface groups [11, 18]. Support CNT2 was treated under nitrogen
flow (100 cm3·min-1) for 1 h at different temperatures (400, 700 and 900
ºC) and the following samples were prepared: CNT2_400, CNT2_700 and
CNT2_900. These thermal treatments selectively remove oxygen-containing
surface groups (previously introduced in the oxidation treatment) by
thermal degradation.
was obtained with a Shimadzu TOC 5000-A and the conversion (ꢏꢐꢑꢒꢒꢓꢒꢔꢕꢑ
)
determined using the equation:
ꢘꢙꢚꢛꢜ ꢙꢝ ꢞꢙꢞꢟꢚ ꢙꢠꢡꢟꢢꢣꢤ ꢤꢟꢠꢥꢙꢢ ꢣꢢ ꢞꢦꢛ ꢠꢛꢜꢧꢚꢞꢟꢢꢞ ꢚꢣꢨꢧꢣꢩ
ꢏꢐꢑꢒꢒꢓꢒꢔꢕꢑꢖ%ꢗ ꢂ ꢘꢙꢚꢛꢜ ꢙꢝ ꢤꢟꢠꢥꢙꢢ ꢣꢢ ꢤꢛꢚꢚꢧꢚꢙꢜꢛ ꢤꢦꢟꢠꢡꢛꢩ ꢣꢢꢞꢙ ꢞꢦꢛ ꢠꢛꢟꢤꢞꢙꢠ ꢪ 100
(1)
(2)
Preparation of Catalysts.
A ruthenium catalyst, with a metal content of 0.4 wt%, was prepared by
classical incipient wetness impregnation of commercial multi-walled
carbon nanotubes (CNT0) with an aqueous solution of the metallic
precursor (RuCl3). After impregnation, the resulting material was dried
overnight at 110 ºC. Next, the catalyst was submitted to thermal treatment
at 250 ºC under a nitrogen flow of 50 cm3·min-1 for 3 h and reduction at
250 ºC under a hydrogen flow of 50 cm3·min-1 for 3 h. The sample was
denoted as Ru/CNT0.
The conversion of glucose (ꢏꢫꢒꢓꢐꢔꢕꢑ) was calculated using the equation:
ꢗꢮꢘꢙꢚꢛꢜꢖꢬꢚꢧꢤꢙꢜꢛ
ꢏꢫꢒꢓꢐꢔꢕꢑꢖ%ꢗ ꢂ ꢘꢙꢚꢛꢜ ꢖꢬꢚꢧꢤꢙꢜꢛ
ꢗ ꢪ 100
ꢭ
ꢯ
ꢘꢙꢚꢛꢜ ꢖꢬꢚꢧꢤꢙꢜꢛ
ꢗ
ꢭ
Further details can be found elsewhere [10a]
.
Acknowledgements
Ruthenium was also supported on the modified carbon nanotube samples
following the same procedure, and the corresponding catalysts were
This work was financially supported by: project POCI-01-0145-
FEDER-006984 – Associate Laboratory LSRE-LCM funded by
denominated as Ru/CNT1, Ru/CNT2, Ru/CNT2_400, Ru/CNT2_700 and
Ru/CNT2_900
.
FEDER through COMPETE2020
Competitividade e Internacionalização (POCI) – and by national
funds through FCT - Fundacao para a Ciencia e a Tecnologia. J.J.
- Programa Operacional
Characterization of Materials.
̂
̧
̃
Surface area measurements of the supports and the different catalysts
were performed using a Quantachrome NOVA Surface Area and Pore Size
analyser with N2 at -196 ºC as sorbate. Prior to analysis, the samples were
Delgado is grateful to Ramon y Cajal program and the ATOM
8
This article is protected by copyright. All rights reserved.