Efficient green methanol synthesis from glycerol

Article abstract of DOI:10.1038/nchem.2345

The production of biodiesel from the transesterification of plant-derived triglycerides with methanol has been commercialized extensively. Impure glycerol is obtained as a by-product at roughly one-tenth the mass of the biodiesel. Utilization of this crude glycerol is important in improving the viability of the overall process. Here we show that crude glycerol can be reacted with water over very simple basic or redox oxide catalysts to produce methanol in high yields, together with other useful chemicals, in a one-step low-pressure process. Our discovery opens up the possibility of recycling the crude glycerol produced during biodiesel manufacture. Furthermore, we show that molecules containing at least two hydroxyl groups can be converted into methanol, which demonstrates some aspects of the generality of this new chemistry.

Full text of DOI:10.1038/nchem.2345

ARTICLES
PUBLISHED ONLINE: 14 SEPTEMBER 2015 | DOI: 10.1038/NCHEM.2345
Efficient green methanol synthesis from glycerol
1
†
1
1
1
2
Muhammad H. Haider , Nicholas F. Dummer , David W. Knight , Robert L. Jenkins , Mark Howard ,
*
1
1
1
Jacob Moulijn , Stuart H. Taylor and Graham J. Hutchings
*
The production of biodiesel from the transesterification of plant-derived triglycerides with methanol has been
commercialized extensively. Impure glycerol is obtained as a by-product at roughly one-tenth the mass of the biodiesel.
Utilization of this crude glycerol is important in improving the viability of the overall process. Here we show that crude
glycerol can be reacted with water over very simple basic or redox oxide catalysts to produce methanol in high yields,
together with other useful chemicals, in a one-step low-pressure process. Our discovery opens up the possibility of
recycling the crude glycerol produced during biodiesel manufacture. Furthermore, we show that molecules containing at
least two hydroxyl groups can be converted into methanol, which demonstrates some aspects of the generality of this
new chemistry.
here is presently a drive towards identifying new sustainable acrolein21,25. We considered that the dehydration reaction could
routes to important platform chemicals and fuels that can also be base catalysed. On this basis we reacted aqueous glycerol
–5
T
interface bioderived feedstocks1 with current petrochemical (Supplementary Figs 1–3) over MgO, a well-known basic oxide,
6
–8
and chemical industries that are based primarily on fossil fuels
Much emphasis has been placed on biorefinery processes . At acrolein formation catalysed by strong acids (that is, 500–600 K,
.
under reaction conditions similar to those we had used for
9–11
2
1,25
present a number of processes have been developed and commercia- 10% glycerol in water) . In these initial exploratory experiments
lized, including bioethanol and biobutanol as well as the production we used pure glycerol in line with previous experimental studies.
of biodiesel from the transesterification of plant-derived triglycerides We observed (Supplementary Table 1) that acrolein was still
with methanol1 . This production of biodiesel produces impure formed, but as a minor product. Surprisingly, we identified metha-
glycerol as a by-product at roughly one-tenth the mass of the biodie- nol as the major product (Fig. 1). Indeed, we had found methanol as
sel and consumes methanol derived from fossil fuels
of this crude glycerol can present a problem for this technology and catalysed chemistry . At a relatively low temperature (523 K) and
effectively is providing a brake on further development . Pure refined with MgO, acrolein was observed with a selectivity of about 10% and
glycerol is a high-value material with uses in pharmaceuticals and methanol at about 30%. However, as the reaction temperature was
foodstuffs; however, at present crude glycerol from biodiesel pro- increased the formation of acrolein diminished as the conversion
duction contains high levels of impurities that prevent its use in continued and methanol became the dominant product
2–14
15
16–18
. Utilization a very minor product (about 1%) in the previously published acid-
2
1,25
19
2
0
this form. Glycerol conversion by oxidation into glyceric acid , dehy- (Supplementary Fig. 4 and Supplementary Table 1). With CaO, a
dration into acrolein2
1–25
and hydrogenation into methanol
26,27
has stronger base, the formation of methanol was enhanced, although
been demonstrated, but to date only by using refined glycerol, the overall conversion was decreased (Fig. 1). We then made a
28
which is an expensive and valuable material . In this work we inves- number of mixed magnesium/calcium oxides (Fig. 1 and
tigated a new reaction of glycerol with water using very simple basic Supplementary Table 1) by mixing freshly prepared MgO and
or redox oxide catalysts to produce methanol in high yields, together CaO, and observed that these mixed oxides retained the high con-
with other useful chemicals, in a one-step low-pressure process. Our version levels associated with MgO, but exhibited a much higher
discovery opens up the possibility of recycling the crude glycerol pro- selectivity than the separate oxides, which indicated the presence
duced during biodiesel manufacture and so provide a means to of a synergistic effect. SrO and mixed oxides of strontium and mag-
replace methanol derived from fossil fuel.
nesium were not as effective (Fig. 1 and Supplementary Table 1). It
The conversion of glycerol has been the focus of extensive is clear that a range of products was formed in addition to methanol.
research as it is a highly functionalized molecule readily derived These included acetol and ethanal, and at low temperatures acrolein
from biomass. One desirable target is to convert glycerol into and ethylene glycol. Other products formed in low selectivities
methanol, which is a major chemical intermediate of immense (<5%) were ethanol, propanal, 1-propanol, 2-propanol, allyl
utility. However, the central problem for the conversion of glycerol alcohol, 2,3-butanedione, 2-hexanone, acetone and CO . Acetol
2
into methanol is that hydrogen has to be introduced, as demon- and acrolein are the products of dehydration. Further reduction of
27
strated by Wu et al. , who hydrogenated glycerol with H . We acrolein gives allyl alcohol and propanal; reduction of acetol gives
2
wanted to explore the reactivity of glycerol using water as a potential acetone. However, the formation of methanol requires carbon–
hydrogen source specifically under conditions in which synthesis carbon bond scission and a source of hydrogen. Detailed isotopic
gas (CO + H ) is not required as a key intermediate.
labelling experiments were carried out to explore this new chemistry
2
further. The presence of methanol was confirmed by using
1
3
13
Results
1,2,3- C-tris-glycerol (Aldrich 99%), which resulted in C-metha-
Our initial experiments focused on extending our earlier studies nol identified by the presence of a doublet in the proton NMR spec-
concerning the acid-catalysed dehydration of glycerol into trum centred at 3.3 ppm with a coupling constant of 142 Hz.
1
2
Cardiff Catalysis Institute, School of Chemistry. Cardiff University, Cardiff CF10 3AT, UK. Hull Research & Technology Centre, Saltend, Hull HU12 8DS, UK.
†
Present address: Sabic Technology & Innovation Center (STC), PO Box 42503, Riyadh 11551, Kingdom of Saudi Arabia. *e-mail: mhhaider@yahoo.com;
Hutch@cardiff.ac.uk
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
1
©
2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2345
ARTICLES
1
1
1
1
1
8
6
4
2
0
8
6
4
2
0
a
a = MgO
100
80
MeOH
Conversion
STYMeOH
100
80
60
40
20
0
b = Mg CaO
3
x
x
c = MgCaOx
d = MgCa
e = CaO
O
3
6
4
2
0
0
0
0
f = MgSr O
3
x
g = SrO
5
60
580
600
620
640
660
680
Temperature (K)
b
100
MeOH
Conversion
80
5
53
573
593
613
Temperature (K)
60
40
Figure 1 | Catalytic activity of the metal-oxide and mixed metal-oxide
materials. a–g, Space-time yield (STY) is defined as the grams of methanol
produced per kilogram of catalyst per hour, and is presented as a function of
the reaction temperature. The activity of the catalysts generally increases
20
2
−1
0
with increasing temperature. a, Over MgO, BET surface area 144 m g .
2
−1
2
−1
2
−1
0.5
1.0
Catalyst (g)
1.5
2.0
b, Mg
3
CaO
x
, 25 m g . c, MgCaO
x
2
, 17 m g . d, MgCa
3
−1
O
x
, 11 m g .
2
−1
−1
2
e, CaO, 13 m g . f, MgSr
3 x
O , 3 m g . g, SrO, 3 m g . The experiments
were carried out in a stainless-steel fixed-bed flow reactor housed in a furnace
for temperature control. Experiments were performed under the following
Figure 2 | Catalytic conversion of glycerol over cerium oxide and
selectivity to methanol. a, Effect of temperature on the conversion (mol%)
of glycerol (10 wt%) and methanol selectivity (mol%), which indicates that
the space–time yield (STY) of methanol reaches a plateau with increasing
−1
2
conditions: catalyst (0.5 g), feed flow (1 ml h , 10 wt% glycerol/H O), inert
−1
carrier (100 ml min ), three hours. Full reaction data concerning the
conversion and selectivity are given in Supplementary Table 1.
−1
temperature. Experimental conditions: 0.5 g catalyst, 100 ml min inert
1
8
18
−1
Reactions with H2 O did not lead to the formation of O-metha- carrier, 1 ml h feed flow, products collected for three hours. b, Influence of
nol (Supplementary Figs 5 and 6). The use of D O led to a 50% contact time on glycerol conversion (mol%) and methanol selectivity
2
decrease in the glycerol conversion, which indicates the presence (mol%) at 613 K suggests that the MeOH selectivity can improve with
−1
of a significant isotope effect. On the basis of these results we con- increased contact time. Experimental conditions: 100 ml min inert carrier,
−1
cluded that water was acting as a source of the hydrogen required for 1 ml h feed flow, products collected for three hours. Experimental error is
methanol formation in this reductive process.
±5% as represented by the error bars.
We investigated the method of MgO preparation to determine
whether better catalysts could be obtained. We made a series of four
We investigated the reaction of other oxygenates over MgO as a
magnesium oxides using different heat treatments (Supplementary catalyst. Methanol was formed from ethylene glycol and 1,3-propa-
Fig. 4) and observed that for each of these the product selectivities nediol (Fig. 3), but not from 1- or 2-propanol. It is apparent that
were almost identical, but the activities were related directly to the molecules require more than one hydroxyl group for this reaction
surface area of the MgO (Supplementary Fig. 4), which indicates to be observed. In this initial stage of the study we have not explored
that an important aspect of catalyst design is to maximize the fully the potential range of substrates from which methanol can
surface area. With the most active of these MgO samples (denoted be formed.
MgO (A), see Supplementary Fig. 4), we then investigated higher con-
To deduce the possible reaction pathways by which methanol is
centrations of glycerol. We found we could achieve similarconversions formed, both over MgO, a non-reducible basic oxide, and over
with higher concentrations of glycerol (up to 30%) by increasing the CeO , a reducible oxide, we reacted separately the observed
2
catalyst mass, and that the conversion could typically be maintained products over these catalysts. Methanol, ethanol, acetone, 1- and
at about 25% with 40% methanol selectivity (Supplementary Fig. 7). 2-propanol and acrolein proved unreactive, which indicates these
Extending the reaction time to 35 hours showed no loss of activity to be terminal products. We consider that both thermal dehy-
or selectivity and the catalyst performance was stable over this dration and radical fragmentation in a reductive atmosphere,
2
9
period (Supplementary Fig. 7).
In the next set of experiments we used lanthanide-based oxides inate this degradation of glycerol 1 (Fig. 4). Double dehydration
Supplementary Table 2) and were pleased to find that CeO
, a redu- under these basic conditions generates a relatively small amount
which would be present in steam at this temperature , would dom-
(
2
cible oxide, was very effective. This suggests that a wide range of oxide of acrolein (2), which becomes lower at higher temperatures, in
catalysts may be effective catalysts for this new chemistry. With MgO contrast to the related acid-catalysed reaction. The major
(Fig. 1 and Supplementary Table 1) we found that conversions were pathway appears to feature monodehydration with the loss of a
typically about 25% and we could not significantly improve on this. terminal hydroxyl and the formation of enol 3, tautomeric with
However, with CeO and by either increasing the temperature or acetol (4), followed by radical fragmentation related to a Norrish
2
increasing the catalyst mass, we could achieve complete glycerol con- type-1 process to give the methanol precursor 5 and the acetyl
version and increase the methanol selectivity to 60% (Fig. 2).
radical 6; subsequent reduction leads to methanol and ethanal
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
©
2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2345
ARTICLES
100
80
60
40
20
0
0
.5 wt%
3 wt%
Methanol
10 wt%
Methanol
Methanol
Ethanol
Ethanal
Others
Ethanol
Ethanal
Others
Ethanol
Ethanal
Others
Conversion
Conversion
Conversion
480
500
520
540
560
580
600
620
640
Temperature (K)
Figure 3 | The influence of reaction temperature on the conversion and product selectivities (mol%) over MgO (A) with different feed concentrations of
,3-propanediol. The formation of methanol requires a reactant with at least two hydroxyl groups as no products were detected with 1- or 2-propanol.
1
−1
−1
Reaction conditions: 1 ml h feed flow, 100 ml min inert carrier, 0.25 g catalyst (0.5 g for 10 wt% feed) for three hours. ‘Others’ represents a combination
of acrolein, propionaldehyde, allyl alcohol and 1-propanol in mol%. Experimental error is ±5% as represented by the error bars.
(
7), respectively. Further reduction of the latter could account for Table 3) show that we obtained very similar outcomes using this
the formation of ethanol (9); arguably, the unlikely formation of crude material as those using refined glycerol (Fig. 5).
,3-butanedione (8) provides strong support for the intermediacy
of the acetyl radical 6.
2
Discussion
We regard a second pathway, initiated by a C–C bond cleavage, We consider that the initial potential exploitation of the new chem-
as minor when using MgO. Such a reaction would generate the same istry we have identified is in the fatty acid methyl ester (FAME)-
methanol precursor 5, together with the ethylene glycol radical 10, based biodiesel industry, which is the source of the crude glycerol.
which could lose a hydrogen radical to give enediol (11) and thence Although the availability of glycerol as a by-product is rather
hydroxyethanal (12), fragmentation of which, again by a Norrish stable for the time being, we consider that there remains an oppor-
type-1 process, would give more of the methanol radical 5 and for- tunity to optimize the overall production of biodiesel by incorporat-
maldehyde precursor 13, which could also be reduced to methanol. ing biomethanol based on our new chemistry into the FAME
When carried out over ceria there is a distinct increase in the for- product, and thereby make better use of the vegetable oil feedstock.
mation of products derived from the latter pathway, mostly metha- In FAME, the methanol only accounts for a small percentage of the
nol from at least two reactions, at the expense of those (7 and 8) product molecule (about 11%), and is limited to about 7% in
derived from the major route. It is unclear which factor determines the diesel blend. There is, therefore, significant value in increasing
this change—eithergreaterinitial C–C bond cleavage (to give 5 and 10) the efficiency with which we use crop-based feedstocks, and increas-
or a slowing of monodehydration that leads to acetol (4).
ing the renewable content of the biofuels derived from them. Using
Until now we had been using refined glycerol, which is a renewable methanol to make FAME enables around a 10% increase
premium product and does not represent a viable economic starting in its renewable content, which could be very helpful to the industry
point for methanol synthesis. Glycerol is formed as a by-product of at this time. It is certainly true that there are alternative uses for gly-
biodiesel production, in which fatty triglycerides, derived from veg- cerol in the production of higher-value chemicals, and it is also
etable oils, tallow and even waste from the food industry, are trans- probable that methanol derived from glycerol will not be cost com-
esterified using methanol. Crude glycerol contains many impurities, petitive with methanol derived from natural gas. However, we do
including traces of NaOH (the catalyst used in its manufacture), not anticipate that the new chemistry will find its initial application
unreacted or partly reacted triglycerides, nitrogenous compounds outside the biodiesel arena and we anticipate that the new process
of plant origin and long-chain acids and long-chain alkanes. In will permit 100% renewable FAME to be produced rather than
our final set of experiments we used crude glycerol from a biodiesel 90% renewable FAME. As to the likelihood of its development,
plant (Biodiesel Amsterdam BV). The crude glycerol contained two our process has several benefits. First, the process design is very
phases, namely, aqueous glycerol in one phase and a minor com- simple and the conditions are mild. Second, the methanol produced
ponent of unreacted triglycerides and other organic material can be used directly in the transalkylation process for the pro-
present in a separate phase (Supplementary Figs 8 and 9). duction of FAME. It is always a very favourable situation when stoi-
We separated the aqueous glycerol layer from the organic layer chiometric amounts of chemicals needed in a process are prepared
and then treated the aqueous layer with activated carbon on-site.
(Supplementary Fig. 10). After this simple treatment, the crude gly-
It is also important to consider competing uses for crude gly-
cerol was reacted with CeO
2
. The results (Fig. 5 and Supplementary cerol. Recently, the integration of glycerol conversion into syngas
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
3
©
2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2345
ARTICLES
H
OH
H
H
–
H2O
OH
Taut
O
– H2O
O
HO
OH
HO
HO
H
1
2
O
H
H
7
OH
OH
OH
O
H
–
H2O
Taut
O
+
EtOH
HO
HO
OH
HO
OH
HO
HO
9
x 2
H
1
1
3
4
5
6
O
H
O
8
MeOH
OH
OH
OH
OH
H
–
H
H
Taut
+
HO
OH
OH
OH
OH
O
OH
O
1
5
10
11
12
5
13
Figure 4 | Proposed mechanism for the formation of methanol from glycerol (1). Over base catalysts glycerol can undergo dehydration to form reactive
species, which results in the production of methanol as the major product and of other secondary products, such as acrolein (2), 2,3-butanedione (8) and
ethanol (9). Taut, enol–keto tautomerism.
conversion step is needed and the conditions are mild at 523–680 K
and atmospheric pressure. The reaction is based on heterogeneous
catalysis in the gas phase. A simple process design (single-phase
fixed-bed reactor, and easy product separation by distillation) is
possible. Indeed, the separation of methanol from crude glycerol
by distillation is common practice in existing biodiesel plants. Of
course, catalyst development work has to be done to optimize the
catalyst, and this has yet to be carried out. In particular, catalyst
stability, often negatively influenced by real feedstocks, is crucial
for a satisfactory practical process. Our exploratory study shows
promising results: stable catalyst performance of over more than a
day and impurities in the crude glycerol do not cause large problems
with catalyst stability.
1
00
80
Pure glycerol
9
0
0
0
0
0
0
0
0
0
0
70
Crude glycerol
8
7
6
5
4
3
2
1
60
50
40
30
20
10
0
Therefore, we consider that our results pave the way for a new
catalytic route from aqueous glycerol to methanol, to be used to
recycle crude glycerol as methanol in a biodiesel production unit.
We have not attempted to optimize the catalyst design and there
is no doubt immense scope to generate catalysts with enhanced
activity and selectivity. However, we have shown that methanol
can be produced in a new catalytic reaction that requires neither
high pressure nor hydrogen.
1
0
15
20
25
30
35
40
45
50
55
Feed concentration (wt%)
2
Figure 5 | Catalytic activity of CeO increases the glycerol feed
concentration for both pure and crude glycerol. The pure glycerol solutions
were prepared by diluting glycerol (99.9%) with water, whereas the crude
glycerol solutions were prepared by diluting crude glycerol (about 85 wt%
in water). The catalyst is tolerant of impurities in the feed stream in the
case of the reactions with crude glycerol; however, over three hours the
conversion is lower than that with the corresponding pure solutions. Glycerol
conversion is represented by open symbols and methanol selectivity by
Methods
The MgO (A) catalyst was prepared by first calcining high-purity hydroxide (99%
Sigma Aldrich) at 723 K for 24 hours. The resulting solid was then sieved between 250
−
1
−1
half-filled symbols. Reaction conditions: 1.0 g ceria, 1 ml h feed flow,
and 425 µm followed by refluxing in water (15 ml g for three hours). The resulting
−1
slurry was dried at 383 K for 24 hours and then heated at 875 K under flowing N
2
1
00 ml min inert carrier, three hours duration at 613 K. Experimental
−1
(
100 ml min ) for three hours. A range of MgO catalysts (denoted (B)–(D))
was also prepared by varying the thermal treatment of the hydroxide precursor
Supplementary Information). The oxide catalysts of Ca, Sr, La and Ce were also
error is ±5% as represented by error bars.
(
prepared by the same procedure (without sieving) from their respective high-purity
hydroxides (99%+ Aldrich). Mg/Ca mixed oxides were prepared by physically
grinding different proportions of MgO (A) and CaO before pelleting and sieving
coupled with the production of the resultant methanol has been
studied. This study, referred to as the Supermethanol concept, is a
good benchmark with which to compare our new chemistry. In
(
250–425 µm). Mixed metal-oxide catalysts of Mg/Sr were prepared by mixing the
−3
the Supermethanol process several chemical conversions are corresponding nitrate solutions (total molarity, 1 mol dm ) in an appropriate ratio.
needed that involve harsh conditions: the reforming reaction The solutions were heated to 343 K and aqueous ammonia was added to form a
precipitate (pH = 9, 10), which was collected by evaporating to dryness, and the
(
(
24–27 MPa, 950–1,000 K) and the methanol synthesis reaction
24–27 MPa, 470–520 K) and, in addition, reactions and separations
2
catalysts formed by heating at 1,073 K under flowing N for three hours. Surface areas
were determined according to the Brunauer–Emmett–Teller (BET) method.
Catalytic reactions were evaluated using a gas-phase plug flow microreactor
are needed to tune the CO/CO ratio to the right value. Compared to
2
this process, our process is remarkably simple. Only one chemical (Supplementary Figs 10 and 11). The aqueous glycerol feed was introduced into a
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
©
2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2345
ARTICLES
preheater and vaporizer (573 K) using an HPLC pump with a precisely controlled
11. Budarin, V. L. et al. Use of green chemical technologies in an integrated
−
1
flow rate (0.017 ml min ). The vaporized feed was then swept through the reactor
biorefinery. Energ. Env. Sci. 4, 471–479 (2011).
−
1
system in a flow of nitrogen carrier gas (100 ml min ). All of the catalysts were
pressed and sieved to a uniform particle size (250–425 µm) before use, and were
packed into an 8 mm inner diameter (i.d.) stainless-steel tube between plugs of silica
12. Atabani, A. E. et al. A comprehensive review on biodiesel as an alternative
energy resource and its characteristics. Renew. Sust. Energ. Rev. 16,
2020–2093 (2012).
13. He, L. et al. Biodiesel synthesis from the esterification of free fatty acids and
alcohol catalyzed by long-chain Brønsted acid ionic liquid. Catal. Sci. Technol.
3, 1102–1107 (2013).
14. Schwab, A. W., Bagby, M. O. & Freedman, B. Preparation and properties of
diesel fuels from vegetable oils. Fuel 66, 1372–1378 (1987).
15. Otera, J. Tranesterification. Chem. Rev. 93, 1449–1470 (1993).
16. Lee, S. in Handbook of Alternative Fuel Technology (eds Lee, S., Speight, J. G. &
Loyalka, S.) 297–322 (CRC Press, 2007).
3
wool. The catalysts were packed to a uniform volume of 0.25 to 5 cm , which
−
1
permitted typical gas hourly space velocities of 2,000–24,000 h . The catalyst bed
was heated using an electric furnace placed around the reactor tube and the
temperature of the catalyst was maintained using a proportional integral derivative
temperature controller linked to a thermocouple placed in the catalyst bed. After
exiting the catalyst bed the lines were trace heated to prevent any condensation
taking place. Unreacted glycerol and the reaction products were collected for analysis
in a series of cold traps. Three traps were used, as this was found to be the most
efficient method and ensured that any carry-over from the first trap was collected
subsequently. Crude glycerol was supplied by Biodiesel Amsterdam BV and treated
by decantation of the aqueous phase followed by simple filtration through charcoal
to remove coloured impurities.
17. Twigg, M. V. & Spencer, M. S. Deactivation of supported copper metal catalysts
for hydrogenation reactions. Appl. Catal. A 212, 161–174 (2001).
18. Wainwright, M. S. Catalytic processes for methanol synthesis—established and
future. Stud. Surf. Sci. Catal. 36, 95–108 (1988).
Reaction products, collected in the cold traps, were combined for analysis,
which was performed offline using a Varian CP 3800 gas chromatograph equipped
with a capillary column (ZBWAX plus, i.d. 0.53 × 30 m). Gas samples were also
collected and analysed off-line by means of a Varian CP 3800 GC with a Porapack
Q 1/8″ × 2 m column. Product selectivities (mol%) were calculated from the moles of
product recovered divided by the total moles of all products.
19. Thompson, J. C. & He, B. B. Characterisation of crude glycerol from biodiesel
production from multiple feedstocks. Appl. Eng. Agric. 22, 261–265 (2006).
20. Carrettin, S., McMorn, P., Johnston, P., Griffin, K. & Hutchings, G. J. Selective
oxidation of glycerol to glyceric acid using a gold catalyst. Chem. Commun.
696–697 (2002).
21. Haider, M. H. et al. Rubidium and cesium doped silicotungstic acid catalysts
supported on alumina for the catalytic dehydration of glycerol to acrolein.
J. Catal. 286, 206–213 (2012).
1
The liquid-phase products were analysed by H NMR spectroscopy. NMR
spectra were recorded at room temperature on a Bruker DPX 500 MHz Ultra-Shield
NMR spectrometer (1H 500.13 MHz), and quantified with a 1% Me
internal standard contained in a sealed glass ampoule, which was calibrated against a
known concentration of methanol. Typically, 0.7 ml of sample and 0.1 ml of D
were placed in an NMR tube along with the internal standard. A solvent-suppression
program was run to minimize the signal that arises from the water. Chemical shifts
4
Si/CDCl
3
22. Atia, H., Armbruster, U. & Martin, A. Dehydration of glycerol in gas phase using
heteropoly acid catalysts as active compounds. J. Catal. 258, 71–82 (2008).
23. Chai, S.-H., Wang, H.-P., Liang, Y. & Xu, B.-Q. Sustainable production of
acrolein: gas-phase of glycerol over 12-tungstophosphoric acid supported on
2
O
ZrO
2 2
and SiO . Green Chem. 10, 1087–1093 (2008).
were reported in parts per million relative to Me
4
Si. Formaldehyde was determined
24. Tsukuda, E., Sato, S., Takahashi, R. & Sodesawa, T. Production of acrolein
from glycerol over silica supported heteropoly acids. Catal. Commun. 8,
1349–1353 (2007).
25. Haider, M. H. et al. Effect of grafting zirconia and ceria onto alumina as a
support for silicotungstic acid for the catalytic dehydration of glycerol to
acrolein. Chem. Eur. J. 20, 1743–1752 (2014).
26. Tsang, S. C. E., Oduru, W. O. & Redman, D. J. Methanol production process.
Patent application WO 2009130452 A1 (2009).
27. Wu, C. T. et al. A non-syn gas catalytic route to methanol production. Nature
Commun. 3, 1050 (2012).
using HPLC. The liquid sample was drawn through a silica gel packed cartridge
coated with 2,4-dinitrophenylhydrazine (DNPH). Any formaldehyde within the
reaction solution readily forms a stable derivative with the DNPH reagent. The
derivative was eluted from the column with acetonitrile and analysed by reverse-
phase chromatography using a photodiode array detector set at 360 nm. The
presence of formaldehyde was confirmed via the comparison of retention times with
those of standard DNPH derivatives of this compound. Quantification of the
formaldehyde DNPH derivative was achieved against a range of formaldehyde
DNPH solutions of known concentration. Formaldehyde was detected only in trace
quantities (parts per million).
28. Yang, F., Hanna, M. A. & Sun, R. Value-added uses for crude glycerol—
a byproduct of biodiesel production. Biotech. Biofuels 5, 13 (2012).
29. Burghardt, H. et al. in Proc. 18th World Hydrogen Energy Conf. (eds Stolten, D. &
Grube, T.) 343–348 (Jülich, 2010).
Received 5 February 2015; accepted 10 August 2015;
published online 14 September 2015
References
Acknowledgements
1
.
Huber, G. W., Cortright, R. D. & Dumesic, J. A. Renewable alkanes by aqueous
The authors acknowledge the financial support of Cardiff University, the Engineering and
Physical Sciences Research Council (Grant No. EP/J013420/1), the European Research
Council (ERC-2011-ADG, Grant Agreement No. 291319: ‘After The Gold Rush’) and the
Technology Strategy Board. This work initially formed part of the Glycerol Challenge and
was cofunded by the Technology Strategy Board’s Collaborative Research and Development
programme after an open competition. The Technology Strategy Board is an executive body
established by the UK Government to drive innovation. The authors also thank Biodiesel
Amsterdam BV for supplying crude glycerol.
phase reforming of biomass derived oxygenates. Angew. Chem. Int. Ed. 43,
1
549–1551 (2004).
2
.
.
Klass, D. L. Biomass for Renewable Energy, Fuels and Chemicals (Academic
Press, 1998).
Huber, G. W., Iborra, S. & Corma, A. Synthesis of transportation fuels
from biomass: chemistry, catalysts, and engineering. Chem. Rev. 106,
3
4
044–4098 (2006).
4.
5.
6.
Cortright, R. D., Davda, R. R. & Dumesic, J. A. Hydrogen from catalytic
reforming of biomass-derived hydrocarbons in liquid water. Nature 418,
9
64–967 (2002).
Author contributions
Braden, D. J., Henao, C. A., Heltzel, J., Maravelias, C. C. & Dumesic, J. A.
Production of liquid hydrocarbon fuels by catalytic conversion of biomass-
derived levulinic acid. Green Chem. 13, 1755–1765 (2011).
Han, J., Sen, S. M., Alonso, D. M., Dumesic, J. A. & Maravelias, C. T. A strategy
for the simultaneous catalytic conversion of hemicellulose and cellulose from
lignocellulosic biomass to liquid transportation fuels. Green Chem. 24,
M.H.H. prepared and tested the catalysts and designed the initial experiments, N.F.D
designed the reactor and provided assistance with experimental design and D.W.K.
provided mechanistic insights into the chemistry. Detailed analysis was provided by R.L.J.
S.H.T provided expertise on catalyst preparation. G.J.H. directed the overall research and all
the authors contributed to the analysis of the data and the writing of the manuscript.
6
53–661 (2014).
Additional information
7
8
9
1
.
.
.
Ertl, G., Knözinger, H., Schûth, F. & Weitkamp, J. Handbook of Heterogeneous
Catalysis Vol. 6 (Wiley-VCH, 2008).
Lloyd, L. L., Ridler, D. E. & Twigg, M. V. Catalyst Handbook 2nd edn (Wolfe
Publishing, 1989).
Farmer, T. J. & Clark, J. H. Green chemistry for 21st century industry. Catal. Rev.
2
All the data created during this research are openly available from the Cardiff University
data archive at http://dx.doi.org/10.17035/d.2015.100110. Supplementary information is
available in the online version of the paper. Reprints and permissions information is
available online at www.nature.com/reprints. Correspondenceand requests for materials should
be addressed to M.H.H. and G.J.H.
5, 7–14 (2012).
0. Lin, C. S. K., Luque, R., Clark, J. H., Webb, C. & Du, C. A seawater-based
biorefining strategy for fermentative production and chemical transformations
of succinic acid. High Energ. Dens. Phys. 4, 1471–1479 (2011).
Competing financial interests
The authors declare no competing financial interests.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
5
©
2015 Macmillan Publishers Limited. All rights reserved