Production of targeted aromatics by using Diels-Alder classes of reactions with furans and olefins over ZSM-5

Article abstract of DOI:10.1039/c2gc35767d

In this paper we study the co-feeding of olefins (ethylene and propylene) with a series of furanic compounds (furan, 2-methylfuran, furfural, and furfuryl alcohol) over ZSM-5 catalyst in a continuous flow fixed-bed reactor at a temperature range of 450-600 °C. We show the importance of Diels-Alder cycloaddition reactions when olefins are co-fed with furanics over ZSM-5. Co-feeding propylene with furan (C₄ diene) increased the toluene aromatic selectivity from 22% to 59%. Similarly, co-feeding propylene with 2-methylfuran (C₅ diene) increased the xylenes aromatic selectivity from 9% to 27%. Co-feeding of ethylene with furans did not change the aromatics selectivity. The increase of toluene selectivity was also observed in the co-feeding of propylene with furfural and fufuryl alcohol. Furfural and furfuryl alcohol both underwent decarbonylation reactions to produce CO and furan. The reaction conditions that maximize Diels-Alder products were a propylene to 2MF molar ratio of 1 at 450 °C. Increasing the temperature to 600 °C increased the selectivity of benzene, toluene and CO. Decreasing the temperature to 300 °C increased the coke yield and decreased the xylene yield. The xylene selectivity went through a maximum at a propylene to 2-methylfuran molar ratio of 1:1.

Full text of DOI:10.1039/c2gc35767d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2012, 14, 3114
www.rsc.org/greenchem
PAPER
Production of targeted aromatics by using Diels–Alder classes of reactions
with furans and olefins over ZSM-5†
Yu-Ting Cheng and George W. Huber*
Received 18th May 2012, Accepted 21st August 2012
DOI: 10.1039/c2gc35767d
In this paper we study the co-feeding of olefins (ethylene and propylene) with a series of furanic
compounds (furan, 2-methylfuran, furfural, and furfuryl alcohol) over ZSM-5 catalyst in a continuous
flow fixed-bed reactor at a temperature range of 450–600 °C. We show the importance of Diels–Alder
cycloaddition reactions when olefins are co-fed with furanics over ZSM-5. Co-feeding propylene with
furan (C4 diene) increased the toluene aromatic selectivity from 22% to 59%. Similarly, co-feeding
propylene with 2-methylfuran (C5 diene) increased the xylenes aromatic selectivity from 9% to 27%.
Co-feeding of ethylene with furans did not change the aromatics selectivity. The increase of toluene
selectivity was also observed in the co-feeding of propylene with furfural and fufuryl alcohol. Furfural
and furfuryl alcohol both underwent decarbonylation reactions to produce CO and furan. The reaction
conditions that maximize Diels–Alder products were a propylene to 2MF molar ratio of 1 at 450 °C.
Increasing the temperature to 600 °C increased the selectivity of benzene, toluene and CO. Decreasing the
temperature to 300 °C increased the coke yield and decreased the xylene yield. The xylene selectivity
went through a maximum at a propylene to 2-methylfuran molar ratio of 1 : 1.
inexpensive silica-alumina catalysts are used; (3) fungible liquid
1
Introduction
aromatics (BTX) and light olefins (C2–C6 olefins) are produced
that fit into the existing petrochemical infrastructures; (4) water
is not required for this process; (5) CFP can be used to convert a
range of different lignocellulosic feedstocks; (6) the pretreatment
of biomass is simple (drying and grinding); and (7) the process
uses fluidized-bed reactors that are already commercially used
today in the petroleum refinery.
We have recently shown that aromatics and olefins can be
directly produced from wood via CFP in a bubbling fluidized-
bed reactor using zeolite catalyst.12 The lignin fraction of the
biomass primarily forms coke and carbon dioxide through free
radical chemistry.26–30 Some lignin can also form monomeric
compounds.26 The cellulose and hemicellulose in the biomass
form anhydrosugars.12,14,31 These anhydrosugars undergo de-
hydration reactions and form furanic compounds, such as furan,
2-methylfuran, fufural, and furfuryl alcohol.32 The furanic com-
pounds enter zeolite pores and undergo a series of acid-catalyzed
oligomerization, decarbonylation, decarboxylation, and de-
hydration reactions producing aromatics, olefins, CO, CO2,
water, and coke.33,34 The aromatics + olefins yield is a function
of hydrogen to carbon effective ratio (H/Ceff) of the feedstocks,
which is defined as eqn (1), where C, H, and O are the moles of
carbon, hydrogen, and oxygen, respectively.35,36
Lignocellulosic biomass has attracted significant attention world-
wide as a feedstock for renewable liquid fuels and chemical pro-
duction due to its low cost and availability.1–9 However, it is not
currently used as a feedstock to make liquid fuels due to techno-
logical and economic challenges.1,4,9 Several processes for
obtaining biofuels from biomass are currently under develop-
ment.9,10 The ideal process to produce biofuels and biochemicals
from lignocellulosic biomass would be a single step reactor
where solid biomass is directly converted into a liquid fuel at
short residence times. Catalytic fast pyrolysis (CFP) is such a
process for the production of renewable aromatics including
benzene, toluene, and xylenes (BTX).11–13 In this single step
process biomass, including wood, agricultural wastes, or fast-
growing energy crops is fed into a fluidized-bed reactor with zeolite
catalysts. During CFP biomass is rapidly heated (>500 °C s−1
)
to intermediate temperatures (400–600 °C) where the biomass
decomposes into oxygenated pyrolysis vapors.14–16 The pyrol-
ysis vapors then enter the zeolite pores where they are converted
into aromatics and light olefins, as well as CO, CO2, water, and
dehydrated oxygenates.11,17–25 The advantages of CFP are: (1)
all the reaction chemistry occurs in a single reactor; (2)
Department of Chemical Engineering, University of Massachusetts
Amherst, 159 Goessmann Lab, 686 N. Pleasant St., Amherst, MA, USA.
E-mail: huber@ecs.umass.edu; Fax: +1 (413) 545 1647;
Tel: +1 (413) 545 0276
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2gc35767d
H ꢀ 2O
H=Ceff
¼
ð1Þ
C
It was found that the yield of desired products (aromatics +
olefins) increased with the increase of H/Ceff ratios but undesired
3114 | Green Chem., 2012, 14, 3114–3125
This journal is © The Royal Society of Chemistry 2012

View Article Online
products (coke) decreased. One method to increase H/Ceff ratios
in the feedstock is to co-feed hydrogen-rich hydrocarbons such
as olefins. In our previous study we fed ethylene and propylene
with biomass feedstocks into a bubbling fluidized-bed reactor
and obtained improved aromatics yield.12
in one step and produces compounds with 6-membered ring
structure that is favorable in the ZSM-5 pores. Brandvold
obtained 30% theoretical yield of p-xylene by using acid-washed
activated carbon or zeolite catalysts.54 Dauenhauer et al. per-
formed the same reactions (DMF + ethylene) over synthesized
HY and ZSM-5 catalysts at 200–300 °C.56 They found Brønsted
acids catalyzed dehydration reaction, which is the rate-determin-
ing step, and thus promoted the Diels–Alder cycloaddition. In
these studies the reactions were performed at low temperatures
(100–300 °C), high pressure (10–1500 psig), and long residence
time (1–48 h) that are far from CFP conditions.
These results suggest that Diels–Alder classes of reactions
could be used to make tune the aromatic distribution in CFP. The
required dienes (furanics, such as 2-methylfuran) can be
obtained from hydrodeoxygenation of furfural in high
yields.57–59 Furfural is now produced primarily in China with a
production of 150–220 kg per metric ton of lignocellulosic
biomass. Recently, improved furfural yield and process cost has
been reported.60,61 Therefore, it is highly desirable to produce
aromatics from those furanics. The objective of this paper is to
demonstrate how CFP could be tuned to more selectively make
targeted aromatics by adjusting the relative concentration of
different reactants within the CFP process. Alternatively targeted
furanic compounds could be produced from biomass and co-feed
over zeolites in the gas phase to make selective aromatics. A
range of furanic compounds can be produced from biomass by
aqueous phase dehydration reactions.5,62 In this paper we will
report the product selectivity for conversion of different furanic
compounds over ZSM-5 with the goal of showing how the aro-
matic selectivity could be adjusted in CFP by changing the
product concentration or reaction conditions. Fig. 2 shows the
different reactions we will study in this paper, where different
furans (diene) react with olefins (dienephile) forming aromatics
and water.
We have used furan (C4H4O) as a probe molecule to identify
some of the key classes of reactions that occur during biomass
CFP.12,33,37 In this model reaction a wide distribution of aro-
matics were obtained including BTX, and smaller amount of
undesired styrene, indene, and naphthalene. The ratio of benzene
to toluene to xylenes (BTX ratio) is typically 6 to 5.5 to 1.33
It would be highly desirable to be able to tune this ratio for
several reasons: (1) the ratio is far from the market demands in
petrochemical industry (55 to 11 to 34);38 (2) xylenes are the
most valuable aromatic and the xylene selectivity is low; and
(3) each of the aromatics has different uses in the petrochemical
industry.38–42 For example, benzene is alkylated with olefins to
make alkylbenzenes for production of polystyrenes.40,41 Benzene
is also used as a feedstock to make polycarbonates and poly-
urethanes.40,41 Toluene is a molecule with high octane number
that can be blended into gasoline (>20 wt%).43,44 Toluene can
also be converted into benzene and xylenes by disproportiona-
tion reactions, or used to make TDI (toluene diisocyanate) for
production of polyurethanes.39–42 p-Xylene is the precursor to
make terephthalic acid which is then used to make PET (poly-
ethylene terephthalate).39,40 Xylenes can also be blended into
gasoline and are a good octane enhancer.43,44 Due to the diver-
sity of these three basic aromatics (BTX), their market demands
are increasing, especially in p-xylene.38,45
Several researchers have used zeolite catalysts to produce aro-
matics and olefins from biomass, beginning first in the early
1980s with the work of Chen et al.25,35,46–53 For example,
woods were converted by using a two-step reactor where woods
were first pyrolyzed into pyrolysis vapors in a non-catalytic
reactor followed by a second fixed-bed reactor with zeolite to
upgrade the pyrolysis vapors.48–50 Olazar and coworkers
reported 12% aromatic yields in a single stage reactor using
ZSM-5 catalyst.47 Corma and coworkers have converted sorbitol
and glycerol to aromatics over a series of zeolite catalysts where
ZSM-5 was found to be the most selective zeolite to make
aromatics.46
In addition to producing aromatics from these biomass with
zeolite catalysts, recently another route to form aromatics from
furanic compounds in the liquid phase has been suggested, as
described in Fig. 1,54 where 2,5-dimethylfuran (DMF) reacts
with an ethylene to form p-xylene and water. This is a typical
Diels–Alder reaction followed by dehydration. The Diels–Alder
reaction is a cycloaddition reaction where a conjugated diene
adds to an alkene (dienephile) to yield cyclohexene derivatives.55
The Diels–Alder reaction is desired for CFP since it takes place
Fig. 1 Cycloaddition of 2,5-dimethylfuran and ethylene over ZSM-5
catalyst forms p-xylene.
Fig. 2 Examples of cycloaddition (Diels–Alder condensation and
dehydration) of furan-ring with olefins.
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 3114–3125 | 3115

View Article Online
reaction, the spent catalyst was regenerated at 600 °C in a
60 mL min−1 flowing air. The CO formed during regeneration
was converted to CO2 by a CO converter (copper oxide, CuO,
Sigma-Aldrich). CO2 was trapped by a CO2 trap (Ascarite,
Sigma-Aldrich). The coke yield was obtained by measuring the
weight change of the CO2 trap.
Furfural and FA were fed into the reactor by using a bubbler
surrounded by a water bath at 45 °C. The helium stream
(408 mL min−1) flowed through the bubbler and carried furfural
or FA vapor into the reactor. The other operations were the same
as furan or 2MF conversion.
For olefins co-feeding experiments, a 2% ethylene or a 2%
propylene balanced by helium (Airgas) was co-fed into the
reactor with different furans feedstocks. Different olefins/furans
ratios were obtained by diluting the olefins by helium.
Products are identified by GC/MS and quantified by GC/FID/
TCD. In the entire study less than 0.05% of the carbon was
obtained in the condenser. The majority of the products were in
either the gas phase or coke deposited on the catalyst.
2
Experimental and materials
2.1 Materials
Furans, including furan (Sigma-Aldrich, liquid, ≥99%), 2-methyl-
furan (2MF, Sigma-Aldrich, liquid, 99%), furfural (Sigma-
Aldrich, liquid, 99%), and furfuryl alcohol (FA, Sigma-Aldrich,
liquid, 98%) were used as feedstocks without any pretreatment.
A ZSM-5 catalyst from Zeolyst (CBV 3024E, SiO2/Al2O3 = 30)
was used as the catalyst for this study. Before being used, the
catalyst was calcined at 600 °C in a flowing air (Airgas, breathing-
grade air, dehumidified by a drierite tube, the flow rate is
60 mL min−1) for 5 hours in the flow reactor described below.
The following calculations were used in this paper.
moles of carbon in a product
Overall selectivity ¼
ꢁ 100%
moles of carbon in all products
ð2Þ
moles of carbon in an aromatic product
moles of carbon in all aromatic products
Aromatic selectivity ¼
ꢁ 100%
3
Results
ð3Þ
3.1 Furan conversion
moles of carbon in an olefinic product
moles of carbon in all olefinic products
Table 1 summarizes the results for conversion of furan, olefins,
and furan + olefins over ZSM-5. In the first two runs we adjusted
furan conversion by changing the space velocity. The aromatic
distribution (or aromatic selectivity) was not a strong function of
furan conversion when only furan was used as a feed. Only
naphthalene shows an increase in aromatic selectivity from 7 to
15% when increasing furan conversion. Co-feeding of 2% pro-
pylene with the furan caused a dramatic change in the distri-
bution of aromatics. For example, at 600 °C the toluene
selectivity increased from 25 to 59% by co-feeding 2% pro-
pylene with furan. The overall aromatic selectivity increased
from 29.2 to 53.8% when the propylene was co-fed with furan at
600 °C. The carbon monoxide selectivity also decreased with
the propylene co-feed. This means more oxygen was removed
by dehydration reactions rather than decarbonylation reactions.
In the Diels–Alder reactions oxygen is removed from furan
through dehydration reactions producing water as opposed to
decarbonylation reactions which produce CO (see the reaction
(3) in Fig. 2). These results suggest that Diels–Alder reactions
are important when olefins are added to the reactor feed.
Olefin selectivity ¼
ꢁ 100%
ð4Þ
Weight hourly space velocity ðWHSVÞ
flow rate of a feedstock ðg hꢀ1
weight of catalyst ðgÞ
Þ
¼
ð5Þ
moles of a xylene isomer
moles of all xylene isomers
Xylenes distribution ¼
ꢁ 100% ð6Þ
Deactivation rate ¼
yield of 2MF ðtÞ ꢀ yield of 2MFðt ¼ 0 ꢀ 1:5 minÞ
yield of 2MFðt ¼ 0 ꢀ 1:5 minÞ ꢁ t
ꢁ 100%
ð7Þ
2.2 Catalytic conversion of furans in the continuous flow fixed-
Fig. 3 and Table 1 show the overall product carbon selectivity
as a function of temperature with furan and propylene as the
feeds. The CO and CO2 selectivity increased with increasing
temperature, suggesting the rate of decarbonylation and decar-
boxylation increase with temperature compared to the other
reactions. This result was also seen with pure furan where the
selectivity of CO and CO2 was increased from 7 to 18% and 0 to
1% respectively when the temperature increased from 450 to
650 °C.33 However, the CO selectivity decreased when pro-
pylene was added to the feed, suggesting that Diels–Alder reac-
tions occurred at all of these temperatures. The change of CO2
selectivity was smaller than the change in CO. The coke selecti-
vity decreased with increasing temperature. The aromatic selecti-
vity goes through a maximum at 550 °C. In contrast, the olefins
selectivity goes through a minimum at 550 °C. The change of
bed reactor
Details of reactions and analyses are given in the support infor-
mation. The catalytic reactions were carried out in a flow fixed-
bed reactor. Prior to reactions, the catalyst was calcined as
described above. Helium was used as the carrier gas. Furan and
2MF were introduced into the helium stream by a syringe pump
(Fisher KDS-100) and was set to bypass the reactor for 30 min
before switching the helium stream to go through the reactor. An
air bath condenser was used to trap the heavy products. Gas
phase products were collected by air bags. All runs were done at
atmospheric pressure. No pressure drop was detected across the
catalyst bed.
After reaction, the reactor was purged by helium at a flow rate
of 408 mL min−1 for 45 s at the reaction temperature. After
3116 | Green Chem., 2012, 14, 3114–3125
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 Products distribution obtained from conversions of furan, olefins, and furan + olefins mixtures over ZSM-5
Feedstock
Furan
He
Furan
He
Furan
1.92%
Ethylene
Furan
1.94%
Propylene
Furan
2%
Propylene
Furan
2%
Propylene
Furan
2%
Propylene
Non
1.92%
Ethylene
Non
1.94%
Propylene
Non
2%
propylene
Carrier gas
Temperature (°C)
600
10.4
—
6.0
—
600
5.9
—
6.0
—
600
10.4
11.1
6.0
2.6
50
600
10.4
15.6
6.0
2.4
65
550
10.4
16.0
6.0
2.5
59
450
10.4
16.0
6.0
2.5
54
300
10.4
16.0
6.0
2.5
14
600
10.4
11.1
0.0
—
—
4
600
10.4
15.6
0.0
—
—
14
450
9.1
9.1
0.0
—
—
35
Furan WHSV (h−1 a
)
Olefins WHSV (h−1 a
Pfuran (torr)
Olefins/furan mol. ratio
Furan conversion (%)
Olefin conversion (%)
)
48
—
64
—
1
16
25
22
2
Overall selectivity (%)b
CO
CO2
13.9
1.1
14.1
1.8
15.0
1.7
7.5
1.5
4.4
0.8
2.8
0.2
0.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Olefins
Aromatics
Coke
19.1
29.2
33.8
2.9
16.9
42.6
23.0
1.6
17.0
40.2
23.9
2.1
29.0
53.8
7.4
22.7
65.6
5.5
29.4
60.2
5.4
31.0
34.5
19.0
15.3c
20.2
21.2
58.6
0.0
69.3
22.9
7.7
86.9
8.9
4.1
Oxygenates
0.7
1.1
2.0
0.0
0.0
Aromatic selectivity (%)
Benzene
Toluene
27.5
25.1
4.6
24.3
21.8
3.9
28.4
30.1
7.9
9.6
15.8
6.1
15.3
58.6
13.6
5.9
3.9
0.8
8.4
53.7
18.3
5.1
8.4
2.1
5.1
38.9
35.1
1.1
7.9
1.8
1.9
20.9
47.1
0.0
8.4
1.9
31.8
13.4
3.5
35.9
32.7
14.8
2.2
9.8
3.7
9.6
35.1
43.5
0.0
0.0
0.0
Xylenes
Styrenesd
9.8
5.9
5.8
Indenesd
20.6
11.2
1.3
24.4
19.1
0.5
31.8
12.9
0.7
Naphthalenesd
Alkylbenzenesd
2.1
1.8
4.2
10.3
19.7
0.9
11.8
Olefin selectivity (%)
Ethylene
Propylene
C4 olefins
Allene
C5 olefins
C6 olefins
C7 olefins
38.7
35.1
4.3
4.6
14.2
3.1
41.7
35.6
4.6
4.3
11.3
2.5
—
56.4
—
47.1
—
29.5
—
48.2
0.0
7.1
11.6
3.6
10.3
—
58.9
0.0
17.4
12.0
1.5
—
53.8
—
29.2
—
59.1
10.4
4.5
19.2
6.8
24.4
71.9
0.6
3.0
0.0
0.0
32.3
0.7
7.1
3.6
0.0
39.6
0.1
5.2
6.1
1.9
44.8
0.7
0.7
0.1
0.0
55.9
0.0
8.9
4.8
1.2
0.0
0.0
0.0
Xylenes distribution (%)e
p-Xylene
m-Xylene
NA
NA
NA
36
47
17
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
37
47
16
o-Xylene
a Weight hourly space velocity (WHSV) is defined as flow rate of furan, ethylene, or propylene (g h−1) divided by weight of the catalyst (g). b For
conversion of mixtures, ethylene or propylene are excluded. c Oxygenates includes 86% C7 oxygenates formed directly by addition of furan and
propylene. In the other runs most oxygenates are benzofuran (>50%). d Styrenes include styrene and methylstyrenes; indenes include indene, indan
and methylindenes; naphthalenes include naphthalene, dihydronaphthalene, and methylnaphthalenes. Alkylbenzenes includes ethylbenzene, 1,2,4-
trimethylbenzene, 4-ethyltoluene, and propylbenzene. e NA: not analyzed.
Fig. 3 (a) Overall products selectivity and (b) aromatics distribution as a function of reaction temperature obtained from conversion of furan + 2%
propylene over ZSM-5 catalyst; reaction conditions: temperature 300–600 °C, WHSV 10.4 h−1, and furan partial pressure 6 torr.
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 3114–3125 | 3117

View Article Online
coke and aromatics selectivity with temperature was similar to
pure furan.33 However, for pure furan conversion the selectivity
of benzene was comparable with toluene at 450–600 °C. The
relatively high toluene selectivity in Fig. 3(a) was due to Diels–
Alder reactions where toluene should be the major product, as
shown in reaction (3) in Fig. 2.
co-fed with the 2MF. The xylenes was a mixture of all three
isomers: p-xylene, m-xylene, and o-xylene. Table 2 also shows
the distribution of the xylene isomers. At 600 °C m-xylene is the
most favourable isomer with or without propylene co-feeding.
According to the reaction (5) in Fig. 2, m-xylene should domi-
nate xylene isomers for 2MF + propylene reaction. Isomerization
between xylenes inside ZSM-5 pores may also be occurring in
this reaction.67,68
The ratio of the xylene distribution is a function of reaction
temperature as shown in Table 2. Lower temperatures led to an
increase in p-xylene and decreases in m-xylene and o-xylene.
This might be attributed to the formation of coke inside the
ZSM-5 catalyst that can slightly block the pore and modify the
diffusivity of the different xylene isomers. Researchers have
found that the diffusion rate of p-xylene inside ZSM-5 is at least
1000 times faster than o- and m-xylenes.38,69 The increasing dif-
fusion resistance by pre-coking zeolites will create more diffu-
sion difference between p-xylene and the other xylene isomers.38
The coke yield was similar at 600 and 450 °C, but m-xylene and
o-xylene have lower diffusivity at 450 °C due to lower temp-
erature. Therefore, p-xylene increased its selectivity among xylene
isomers. This suggests that the coke induces space confinement
that causes the p-xylene selectivity to further increase.
Fig. 4(a) shows the overall products carbon selectivity as a
function of temperature for conversion of 2MF + propylene.
Details are shown in Table 2. Similar to the conversion of
furan + propylene the CO selectivity increases with temperature
indicating that decarbonylation reactions increase with temp-
erature. Aromatics went through a maximum at 450 °C. The
olefins and CO2 selectivity did not change much with temp-
erature. The distribution of aromatics formed was a strong func-
tion of temperature as shown in Fig. 5(b). The xylenes and
higher alkylbenzenes decreased with increasing temperature, that
was similar to furan + propylene reactions. However, the increase
of temperature from 300 to 450 °C did not change xylenes selec-
tivity. The 2MF conversion was low (13%) at 300 °C with the
propylene conversion being 20% at this same temperature. This
suggests that more of the aromatics were formed from propylene
aromatization instead of 2MF at this temperature. Table 1 also
shows that propylene has a high aromatic selectivity towards
xylenes and alkylbenzenes at 450 °C. These results suggest that
450–600 °C is the best temperature for Diels–Alder reaction of
furanics with olefins over ZSM-5.
As shown in Fig. 3(b), the benzene, toluene and styrene selec-
tivity increased with increasing temperature. The xylenes and
other alkylbenzenes selectivity decreased with increasing temp-
erature. These results suggest that at low temperatures, hydro-
carbon pool products are more favourable than Diels–Alder
reaction type products (toluene). It has been found that at high
temperatures dealkylation of alkylbenzenes occurs, implying that
alkylation reactions are suppressed at high temperatures.63,64
Another interesting result is that in our previous study, at 450 °C
furan conversion produced mostly benzofuran and coke with
18 and 47% overall selectivity, respectively.33 However, with the
existence of propylene in the feedstocks, benzofuran (oxyge-
nates) shows less than 2% overall selectivity, and coke shows
only 5% overall selectivity. This is consistent with our previous
study that co-feeding of molecules that have a high H/Ceff ratio
with furan could reduce coke and make more aromatics.36,65 As
shown in Table 1, when ethylene was fed into the reactor, only
4% conversion was observed. Similarly, co-feeding of furan with
ethylene, only 1% ethylene conversion was observed. The furan
conversion only showed a slightly increase from 48 to 50%
when ethylene was co-fed with furan. The carbon monoxide
selectivity did not decrease and the aromatics distribution did not
change significantly with ethylene co-feeding. It is suggested
that the reaction (1) in Fig. 2 did not occur since benzene did not
increase its selectivity by ethylene co-feeding. These results
suggest that ethylene is not a good dienephile due to its low reac-
tivity. Other types of catalysts should be introduced for ethylene
conversion.66
3.2 2-Methylfuran (2MF) conversion
Table 2 summarizes the results for the conversion of 2MF and
2MF + propylene over ZSM-5 catalyst. 2MF was more reactive
than furan. At 600 °C, 2MF was completely converted at a
WHSV with only 64% of furan converted at this same condition.
About 50% of the oxygen was removed as CO (theoretical CO
selectivity is 20%) with the remaining oxygen being removed as
water. More olefins and aromatics, and less coke were obtained
from 2MF than with furan. As has been previously discussed the
hydrogen to carbon effective ratio (H/Ceff) of a biomass derived
feedstock can be used to predict the amount of olefins and aro-
matics that can be produced.36,65 The aromatic + olefin yield
increases with increasing H/Ceff ratio. 2MF and furan have
H/Ceff ratios of 0.8 and 0.5 respectively. 2MF also has a higher
aromatic + olefin yield than furan. The selectivity of different
aromatics is similar for both 2MF and furan. 2MF had a higher
xylenes selectivity and lower naphthalene selectivity than furan.
Co-feeding of propylene with 2MF at 600 °C dramatically
increases the xylenes selectivity as shown in Table 2. This
increase in xylene selectivity is expected due to an increase in
Diels–Alder reactions as shown in Fig. 1 reaction (5). The CO
selectivity decreased from 11% to 6% when propylene was
3.3 2MF conversion: olefins/2MF ratios
We also carried out the 2MF + propylene reactions at 450 °C
with different olefins to 2MF molar ratios (Ole/2MF). As shown
in Table 2, the 2MF conversion increased with an increase in the
Ole to 2MF ratio. This suggests that propylene helps convert the
2MF into aromatics. Fig. 5(a) shows the overall products carbon
selectivity as a function of Ole/2MF ratio. The CO selectivity
decreased with an increase in propylene concentration. The coke
and CO2 selectivity decreased with increasing Ole to 2MF ratio.
These suggest that the existence of hydrogen-rich reactants like
propylene reduced coke formation during furans CFP, and
shifted the method of oxygen removal to a dehydration route
instead of decarbonylation or decarboxylation. The aromatic
3118 | Green Chem., 2012, 14, 3114–3125
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 2 Products distribution obtained from conversion of 2MF (2-methylfuran) and 2MF + propylene mixtures over ZSM-5
Feedstock
Carrier gas
2MF
He
2MF
2% Propylene
2MF
2% Propylene
2MF
2% Propylene
2MF
2% Propylene
2MF
2% Propylene
2MF
2% Propylene
Temperature (°C)
600
5.7
—
4.9
—
98
—
600
5.7
9.1
4.9
3.09
99
450
5.7
9.1
4.9
3.09
92
300
5.7
9.1
4.9
3.09
13
200
5.7
9.1
4.9
3.09
0
450
5.7
2.9
4.9
1.00
86
450
5.7
1.0
4.9
0.35
79
2MF WHSV (h−1 a
)
Propylene WHSV (h−1 a
P2MF (torr)
Olefins/2MF molar ratio
2MF conversion (%)
Propylene conversion (%)
)
31
42
20
14
43
35
Overall selectivity (%)b
CO
CO2
11.0
0.5
5.9
0.1
1.2
0.2
0.0
0.0
0.0
0.0
3.0
0.8
4.8
1.5
Olefins
Aromatics
Coke
23.3
47.3
16.6
1.4
27.8
59.6
6.2
27.5
66.1
5.0
23.5
51.7
24.8
0.0
14.4
20.6
65.0
0.0
11.7
71.1
12.1
1.3
8.9
60.4
21.6
2.8
Oxygenatesc
0.4
0.0
Aromatic selectivity (%)
Benzene
Toluene
23.8
24.5
9.2
24.4
28.6
26.9
2.6
10.3
3.1
6.8
17.5
48.9
0.8
6.8
1.0
0.9
9.2
51.3
0.0
3.4
1.0
1.8
5.1
35.7
0.5
10.9
11.7
34.4
6.9
17.1
46.5
2.2
9.4
1.6
7.4
17.3
35.7
4.1
18.7
1.7
Xylenes
Styrenesd
4.8
Indenese
25.4
10.1
2.2
Naphthalenese
Alkylbenzenese
4.0
18.2
34.2
16.4
15.1
Olefin selectivity (%)
Ethylene
Propylene
C4 olefins
Allene
C5 olefins
C6 olefins
C7 olefins
28.7
35.0
15.6
0.3
17.4
3.0
61.6
—
30.4
—
9.1
—
61.2
0.0
20.2
9.5
0.0
0.0
—
61.9
0.0
18.3
19.9
0.0
47.1
—
56.7
—
28.0
0.1
7.3
3.0
0.0
49.3
0.0
8.0
8.6
3.7
40.3
0.0
5.7
5.2
1.7
30.1
0.0
5.2
6.7
1.3
0.0
Xylenes distribution (%)e
p-Xylene
m-Xylene
36
46
17
32
49
19
43
42
14
54
37
9
NA
NA
NA
41
45
14
43
43
13
o-Xylene
a Weight hourly space velocity (WHSV) is defined as flow rate of 2MF or propylene (g h−1) divided by weight of the catalyst (g). b For conversion of
mixtures, ethylene or propylene are excluded. c Oxygenates contain furan and furylethylene. d Styrenes include styrene and methylstyrenes; indenes
include indene, indan and methylindenes; naphthalenes include naphthalene, dihydronaphthalene, and methylnaphthalenes. Alkylbenzenes includes
ethylbenzene, 1,2,4-trimethylbenzene, 4-ethyltoluene, and propylbenzene. e NA: not analyzed.
Fig. 4 (a) Overall products selectivity and (b) aromatics distribution as a function of reaction temperature obtained from conversion of 2MF
(2-methylfuran) + 2% propylene over ZSM-5 catalyst; reaction conditions: temperature 300–600 °C, WHSV 5.7 h−1, and 2MF partial pressure 4.9 torr.
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 3114–3125 | 3119

View Article Online
Fig. 5 (a) Overall products selectivity and (b) aromatics distribution as a function of olefins/2MF (2-methylfuran) ratios obtained from conversion of
2MF + propylene over ZSM-5 catalyst; reaction conditions: temperature 450 °C, WHSV 5.7 h−1, and 2MF partial pressure 4.9 torr.
Fig. 6 (a) Selectivity of p-, m-, and o-xylenes, and (b) total xylenes yield as a function of reaction time on stream, obtained from 2MF (2-methyl-
furan) + 2% propylene conversion. Reaction conditions: temperature 600 (#) and 450 (*) °C, 2MF WHSV 5.7 h−1, and 2MF partial pressure 4.9 torr.
selectivity goes through a maximum with the Ole to 2MF ratio.
When pure propylene is used the aromatics selectivity is 8.9%.
The aromatics had a maximum selectivity at an Ole to 2MF ratio
of 1, suggesting that bimolecular reactions were favored for aro-
matics production. This is consistent with Diels–Alder reaction
as shown in Fig. 2. Fig. 5(b) shows the aromatic distribution as a
function of Ole to 2MF ratio. When 2MF was in the feed, the
xylene selectivity increased when the Ole/2MF ratio increased
from 0.35 to 1.0, and did not change at higher Ole/2MF ratio.
Pure propylene conversion shows more toluene but less xylenes.
on stream. The p-xylene selectivity then decreased to a steady
state value of 40% and 50% at 450 and 600 °C, respectively.
The overall p-xylene yield rapidly decreased within 50 min time
on stream as shown in Fig. 6(b). This deactivation is due to
coking on the catalyst surface. The rates of catalyst deactivation
at 600 and 450 °C initially are 334 and 218 h−1 respectively, and
are 10 and 0 h−1 respectively at 50 min time on stream. The
deactivation rates for 2MF + propylene were faster than pure
furan conversion over ZSM-5 where its deactivation rate was
126 h−1 in the beginning and was 5.5 h−1 at 1 hour time on
stream.33
3.4 2MF conversion as a function of time on stream
3.5 Furfural and furfuryl alcohol (FA) conversion
Fig. 6(a) shows the xylene isomers distribution as a function of
reaction time on stream obtained for conversion of 2MF + pro-
pylene at 450 and 600 °C. p-Xylene selectivity went through a
maximum of 60% and 55% at around 20 min time on stream for
reactions at 450 and 600 °C, respectively. The m-xylene went
through a minimum of 30% for both temperatures at 20 min time
Fig. 7 shows the overall products carbon selectivity and aro-
matics distribution obtained from different furanic feedstocks
(including furan, furfural, furfuryl alcohol (FA) and 2MF) with
and without co-feeding of propylene at 600 °C. Furfural had a
higher CO selectivity than the other furans. However, the
3120 | Green Chem., 2012, 14, 3114–3125
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 7 (a) and (b) Overall selectivity of products; (c) and (d) their aromatics distribution obtained from conversion of different feedstocks; reaction
conditions: temperature 600 °C; WHSV 10.4 h−1 for furan, 5.7 h−1 for 2MF (2-methylfuran), 9.0 h−1 for furfural, and 3.3 h−1 for FA (furfuryl
alcohol); partial pressure 6 torr for furan, 7 torr for furfural, 2.5 torr for FA, and 4.9 torr for 2MF.
aromatic distribution (BTXN selectivity) for furfural and furan
was similar. The furfural conversion also showed high selectivity
of furan in the products (15%). These suggest that the first step
for furfural conversion is decarbonylation to form furan and CO.
The produced furan then enters the furan conversion network as
shown in Fig. 8. When co-feeding propylene, oxygenates (furan)
showed low selectivity. This suggests that the propylene helps
convert the produced furan by Diels–Alder reactions. 2MF
formed more aromatics than furan and furfural. Furan and
furfural have lower H/Ceff ratios than 2MF. FA has H/Ceff ratio
0.4 that is lower than furan (0.5) but higher than furfural (0).
Although FA showed higher aromatics selectivity than furan, it
produced more undesired polycyclic aromatics (Fig. 7(c)) such
as indenes and naphthalenes, than the other feeds. 2MF showed
a similar aromatics distribution as furan and furfural, suggesting
that 2MF may be involved in similar reaction mechanisms.
The aromatic and olefin selectivity increased and the coke and
CO selectivity decreased when propylene was added to the
feeds. The toluene selectivity increased by 34 and 29% when
propylene was co-fed with furan and furfural, respectively. The
xylene selectivity increased by 18% when propylene was co-fed
with 2MF. The toluene and benzene selectivity did not change
much when propylene was co-fed with 2MF (≤4%). The
benzene, toluene and xylenes selectivity increased when propy-
lene was co-fed with FA. The Diels–Alder reaction effect was
not significant for FA + propylene but we still can see that the
toluene increased its selectivity more than benzene and xylenes.
This suggested that FA was firstly converted into furan and then
furan reacted with propylene that formed toluene.
4
Discussion
4.1 Reaction network
We have described the overall reaction network for furan conver-
sion at 600 °C in our previous work.33,66 In this paper we have
shown the importance of Diels–Alder reactions between different
furanic species and olefins within this network. Fig. 8 summa-
rizes this chemistry. Furan (C4H4O) first undergoes one of three
reactions: (1) decarbonylation to form allene (C3H4) and CO,
(2) Diels–Alder reaction with another furan followed by de-
hydration forming benzofuran and water or (3) Diels–Alder reac-
tion with propylene (or another olefin) to produce toluene (or
another aromatic) and water. The allene undergoes oligomerization
and cracking reactions to produce a series of olefins. Benzene
can be produced either from benzofuran decarbonylation or
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 3114–3125 | 3121

View Article Online
Fig. 8 Overall reaction network for conversion of furan and furan + propylene over ZSM-5 catalyst at elevated temperatures.
allene dimerization/dehydrogenation. There are two possible
routes to form the other aromatics. The first route is alkylation
and cracking of an aromatic with allene or another olefin
forming ethylene and another aromatic. The second route is a
Diels–Alder reaction between olefins and furan. For example,
propylene reacts with furan producing toluene and water. Poly-
cyclic aromatics can also be formed via Diels–Alder reaction
between aromatics and furan. For example, the benzene can react
with furan and form naphthalene and water. By adjusting the
concentration of olefin to furan in the reactor the rate of this
second pathway can be increased causing the selectivity to
toluene or other targeted aromatics to increase.
The addition of propylene to the reactor increases the rate of
furan conversion. This suggests that the Diels–Alder reactions
are faster than the decarbonylation reactions. For example, at
450 °C without propylene co-feeding, the furan conversion was
only 22% and the aromatics selectivity was 20%.33 Benzofuran
(18% carbon selectivity) and coke (47% carbon selectivity) were
the products produced in highest yield. However, as shown in
Table 1, at 450 °C and with the existence of propylene, 54%
furan was converted and only 5% converted carbon formed coke
at the same furan partial pressure and space velocity.
compound for xylenes production. The reaction of 2MF with
propylene at 600 °C can produce xylenes with selectivity com-
parable to benzene and toluene (the BTX ratio is 0.9 : 1 : 1.1).
The xylene selectivity is also a strong function of temperature.
Fig. 3(b) and 4(b) show that the xylenes and alkylbenzene selec-
tivity decreased with increasing temperature. This suggests that
at higher temperatures dealkylation reactions begin to dominate.63,64
The overall aromatic selectivity however goes through a maximum
with temperature at 450 °C for the reaction of 2MF and propylene.
The p-xylene selectivity (this selectivity only includes
p-xylene, m-xylene, and o-xylene) was a function of temperature
(Table 2). The p-xylene selectivity increased with decreasing
temperature (Table 2). A similar temperature-dependence of
xylene isomers was also seen by S. Al-Khattaf for the reaction of
m-xylene transformation using pre-coked ZSM-5 catalyst.67 In
Al-Khattaf’s study the ratio of p-xylene to o-xylene was
increased by decreasing the temperature from 500 to 400 °C in
both fresh and pre-coked ZSM-5 catalysts. p-Xylene has a
higher diffusivity than the other two isomers and thus it’s selec-
tivity increases with decreasing temperatures.38,67 At high temp-
eratures the effect of the diffusivity is reduced so that all isomers
can easily diffuse out from the ZSM-5 pores. At lower tempera-
tures the space confinement imposed by ZSM-5 pores is more
important. It has been suggested that coking of the zeolite pores
will decrease the pore volume and cause a change in diffusivity
of the different xylene isomers.67 In the reactions studied in
this paper coke builds up on the catalyst surface with time on
stream and causes an increase of p-xylene selectivity, as shown
in Fig. 6(a). This suggests that small amounts of coke can help
maximize the p-xylene selectivity. However, too much coke
4.2 Maximizing p-xylene selectivity
p-Xylene is a large volume commodity chemicals that is used as
a starting material to produce PET. Here we would like to
suggest reaction conditions for maximizing p-xylene selectivity.
Xylenes are produced during biomass CFP.12,33,36,66 According
to Tables 1–3 and Fig. 2, 2MF is the best starting furanic
3122 | Green Chem., 2012, 14, 3114–3125
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 3 Products distribution obtained from conversion of furfural, furfural + propylene mixtures, FA (furfuryl alcohol), and FA + propylene
mixtures over ZSM-5
Feedstock
Carrier gas
Furfural
He
Furfural
2% Propylene
Furfural
2% Propylene
FA
He
FA
2% Propylene
Temperature (°C)
600
9.0
—
7.0
100
—
600
9.0
9.1
7.0
100
53
450
9.0
9.1
7.0
100
64
600
3.3
9.1
2.5
100
600
3.3
9.1
2.5
100
29
Furans WHSV (h−1 a
)
Propylene WHSV (h−1 a
)
Pfurfural or PFA (torr)b
Furfural and FA conversion (%)
Propylene conversion (%)
—
Overall selectivity (%)c
CO
CO2
29.3
3.3
11.3
1.9
7.3
0.1
13.0
1.5
2.0
0.0
Olefins
Aromatics
Coke
18.8
16.7
16.6
15.2
38.7
42.7
3.4
46.6
38.6
2.9
14.7
42.4
25.8
2.6
58.5
34.4
4.8
Oxygenatesd
2.0
4.5
0.3
Aromatic selectivity (%)
Benzene
Toluene
35.5
28.6
6.9
21.0
57.9
14.2
2.1
2.3
1.6
6.1
38.5
38.5
0.6
5.3
0.9
9.1
13.1
13.3
9.6
33.4
17.1
4.4
23.4
38.2
21.7
3.5
6.2
3.7
Xylene
Styrenese
6.9
Indenese
11.3
10.3
0.6
Naphthalenese
Alkylbenzenese
1.0
10.2
3.3
Olefin selectivity (%)
Ethylene
Propylene
C4 olefins
Allene
C5 olefins
C6 olefins
C7 olefins
38.1
42.8
8.2
4.5
5.8
0.62
0.0
58.3
—
28.6
—
55.4
0.0
7.9
6.0
43.5
36.1
12.5
3.2
4.7
0.0
55.8
—
35.6
0.0
5.1
1.0
0.0
40.9
0.0
2.4
0.9
0.0
2.07
0.0
Xylenes distribution (%)
p-Xylene
m-Xylene
37
46
17
32
51
18
40
46
14
40
43
17
31
49
20
o-Xylene
a Weight hourly space velocity (WHSV) is defined as flow rate of furfural or propylene (g h−1) divided by weight of the catalyst (g). The amount of
furfural was estimated by Antoine Equation at 45 °C. b The vapor pressure of furfural was estimated by Antoine equation at 45 °C. c For conversion of
mixtures, ethylene or propylene are excluded. d Oxygenates contain 97–99% furan and 1–3% 2MF. e Styrenes include styrene and methylstyrenes;
indenes include indene, indan and methylindenes; naphthalenes include naphthalene, dihydronaphthalene, and methylnaphthalenes. Alkylbenzenes
includes ethylbenzene, 1,2,4-trimethylbenzene, 4-ethyltoluene, and propylbenzene.
deposition will deactivate the catalyst and decrease the p-xylene
yield as shown in Fig. 7(b).
biomass is directly fed into a fluidized bed reactor with a zeolite
catalysts to produce aromatics and olefins by CFP.12 Ethylene
and propylene are produced in this CFP reaction and can be
recycled back into the reactor to produce more aromatics.12,33
The second route is where biomass is selectively dehydrated to
form furanic compounds. These furanic compounds then are
converted into aromatics via Diels–Alder reactions. As discussed
in the Introduction, furanic compounds are produced from fur-
fural that is produced from biomass with low cost. Therefore,
our results show these other routes to convert the furanics into
targeted aromatics may become commercially viable.
The third parameter which affects the xylene selectivity is the
ratio of propylene to 2MF. The xylene selectivity increases with
increasing propylene to 2MF molar feed ratio. However the
xylene selectivity decreased when pure propylene was fed into
the reactor. The ratio of p-xylene, m-xylene, and o-xylene was
not a function of the olefin to 2MF feed ratio as shown in
Table 2. This indicated that the xylene isomers distribution more
likely was controlled by temperature and the zeolite structure
(e.g. the space confinement imposed by the zeolites) rather than
Ole/2MF ratios. The Ole/2MF ratios simply changed the overall
selectivity of xylenes.
5
Conclusions
4.3 Aromatics are produced from renewable biomass
Furan, furfural, and 2MF produced aromatics, olefins, CO, CO2,
coke and water over ZSM-5. The addition of propylene to the
feed increased the aromatics selectivity and the selectivity of
This paper thus shows two different routes to make aromatics
from biomass using zeolite catalysts. The first route is where
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 3114–3125 | 3123

View Article Online
11 T. R. Carlson, T. R. Vispute and G. W. Huber, ChemSusChem, 2008, 1,
397–400.
12 T. R. Carlson, Y.-T. Cheng, J. Jae and G. W. Huber, Energy Environ. Sci.,
2011, 4, 145–161.
13 J. P. Diebold and J. W. Scahill, Abstr. Pap. Am. Chem. Soc., 1987, 193,
70.
14 Y. C. Lin, J. Cho, G. A. Tompsett, P. R. Westmoreland and G. W. Huber,
J. Phys. Chem. C, 2009, 113, 20097–20107.
15 M. S. Mettler, S. H. Mushrif, A. D. Paulsen, A. D. Javadekar,
D. G. Vlachos and P. J. Dauenhauer, Energy Environ. Sci., 2012, 5, 5414.
16 P. J. Dauenhauer, J. L. Colby, C. M. Balonek, W. J. Suszynski and
L. D. Schmidt, Green Chem., 2009, 11, 1555–1561.
17 H. B. Goyal, D. Seal and R. C. Saxena, Renew. Sust. Energy Rev., 2008,
12, 504–517.
18 A. Demirbas, Energy Sources Part A-Recovery Util. Environ. Eff., 2007,
29, 753–760.
19 T. R. Carlson, G. A. Tompsett, W. C. Conner and G. W. Huber, Top.
Catal., 2009, 52, 241–252.
20 S. Czernik, J. Scahill and J. Diebold, J. Sol. Energy Eng., Trans. ASME,
1995, 117, 2–6.
21 J. Adam, E. Antonakou, A. Lappas, M. Stocker, M. H. Nilsen,
A. Bouzga, J. E. Hustad and G. Oye, Microporous Mesoporous Mater.,
2006, 96, 93–101.
22 M. Olazar, R. Aguado, M. J. San Jose and J. Bilbao, J. Chem. Technol.
Biotechnol., 2001, 76, 469–476.
23 D. Fabbri, F. Fabbri, G. Falini, V. Baravelli, A. Magnani, C. Torri,
H. Maskrot and Y. Leconte, J. Anal. Appl. Pyrol., 2008, 82, 248–254.
24 D. Fabbri, C. Torri and V. Baravelli, J. Anal. Appl. Pyrol., 2007, 80,
24–29.
25 A. Aho, N. Kumar, K. Eranen, T. Salmi, M. Hupa and D. Y. Murzin,
Fuel, 2008, 87, 2493–2501.
26 P. R. Patwardhan, R. C. Brown and B. H. Shanks, ChemSusChem, 2011,
4, 1629–1636.
27 J. H. Jae, G. A. Tompsett, Y. C. Lin, T. R. Carlson, J. C. Shen,
T. Y. Zhang, B. Yang, C. E. Wyman, W. C. Conner and G. W. Huber,
Energy Environ. Sci., 2010, 3, 358–365.
28 J. Cho, S. Chu, P. J. Dauenhauer and G. W. Huber, Green Chem., 2012,
14, 428–439.
toluene and xylenes due to the increase in the Diels–Alder reac-
tions. The toluene selectivity increased by the reaction of furan
(a C4 diene) with propylene (a C3 dienephile). Xylenes selec-
tivity increased by reaction of 2-methylfuran (2MF, a C5 diene)
with propylene (a C3 dienephile). The reaction of propylene with
furfural and furfuryl alcohol produces more toluene than xylene
because furfural and furfuryl alcohol are converted into furan
and CO inside the catalyst before they undergo conversion into
aromatics.
When furan, furfural and 2MF were fed by themselves the aro-
matic distribution was similar (benzene, toluene, xylene and
naphthalene distribution) for all three feeds, suggesting they all
produce common intermediates that are converted into aromatics
in this process. We have previously shown that these furanic
compounds are good model compounds for conversion of solid
lignocellulosic biomass into aromatics by catalytic fast pyrolysis
(CFP). The amount of aromatics produced did depend on the
feed. 2MF produced more aromatics and olefins than furan and
furfural. Furfuryl alcohol produced more aromatics than furan.
However, furfural alcohol also produced more polycyclic aro-
matics than furan. Co-feeding of the propylene with furan, 2MF,
furfural and furfuryl alcohol increased the rate of toluene or
xylene production depending on the diene that was fed. This
illustrates that Diels–Alder reactions are important reactions in
the conversion of lignocellulosic biomass into aromatics by CFP.
The reaction conditions that maximize xylene production were
a propylene to 2MF molar ratio of 1 at 450 °C. The xylene distri-
bution is a function of temperature and zeolite structure. Lower
temperature (e.g. 450 °C) and more space confinement (e.g. pre-
coked zeolites) caused the increase of p-xylene. In general, the
Diels–Alder reactions provide an alternative route to produce tar-
geted aromatics from renewable biomass.
29 M. T. Klein and P. S. Virk, Ind. Eng. Chem. Fundam., 1983, 22, 35–45.
30 F. P. Petrocelli and M. T. Klein, Macromolecules, 1984, 17, 161–169.
31 T. R. Carlson, J. Jae and G. W. Huber, ChemCatChem, 2009, 1, 107–110.
32 M. S. Mettler, S. H. Mushrif, A. D. Paulsen, A. D. Javadekar,
D. G. Vlachos and P. J. Dauenhauer, Energy Environ. Sci., 2012, 5,
5414–5424.
33 Y.-T. Cheng and G. W. Huber, ACS Catal., 2011, 1, 611–628.
34 T. R. Carlson, J. Jae and G. W. Huber, ChemCatChem, 2009, 1,
107–110.
35 N. Y. Chen, T. F. Degnan and L. R. Koenig, Chemtech, 1986, 16,
506–511.
36 H. Y. Zhang, Y. T. Cheng, T. P. Vispute, R. Xiao and G. W. Huber,
Energy Environ. Sci., 2011, 4, 2297–2307.
Acknowledgements
This work is supported as part of the Catalysis Center for
Energy Innovation, an Energy Frontier Research Center funded
by the U.S. Department of Energy, Office of Science, Office of
Basic
Energy
Sciences
under
Award
Number
DE-SC0001004. G. W. H. is a shareholder and chairman of the
Scientific Advisory Board of Anellotech.
37 A. J. Foster, J. Jae, Y.-T. Cheng, G. W. Huber and R. F. Lobo, Appl.
Catal. A: Gen., 2011, under reviewed.
38 T. C. Tsai, S. B. Liu and I. K. Wang, Appl. Catal., A-Gen., 1999, 181,
355–398.
39 L. F. Hatch and S. Mater, Hydrocarbon Process., 1979, 58, 189–200.
40 J. G. Speight, Chemical Process and Design Handbook, McGraw-Hill,
New York, 2002.
41 M. Bejblova, N. Zilkova and J. Cejka, Res. Chem. Intermed., 2008, 34,
439–454.
42 J. Cejka and B. Wichterlova, Catal. Rev. Sci. Eng., 2002, 44, 375–421.
43 S. S. Vasu, D. E. Davidson and R. K. Hanson, Combust. Flame, 2008,
152, 125–143.
44 A. Violi, S. Yan, E. G. Eddings, F. Sarofim, S. Granata, T. Faravelli and
E. Ranzi, Combust. Sci. Technol., 2002, 174, 399–417.
45 R. G. Harvan, Oil Gas J., 1998, 96, 56–61.
46 A. Corma, G. W. Huber, L. Sauvanaud and P. O’Connor, J. Catal., 2007,
247, 307–327.
47 M. Olazar, R. Aguado, J. Bilbao and A. Barona, AIChE J., 2000, 46,
1025–1033.
48 J. Diebold and J. Scahill, Prepr. Pap. – Am. Chem. Soc., Div. Fuel
Chem., 1987, 32, 297.
49 P. A. Horne, N. Nugranad and P. T. Williams, J. Anal. Appl. Pyrol., 1995,
34, 87–108.
Notes and references
1 L. R. Lynd, C. E. Wyman and T. U. Gerngross, Biotechnol. Prog., 1999,
15, 777–793.
2 C. E. Wyman, B. E. Dale, R. T. Elander, M. Holtzapple, M. R. Ladisch
and Y. Y. Lee, Bioresour. Technol., 2005, 96, 2026–2032.
3 C. E. Wyman, Annu. Rev. Energy Environ., 1999, 24, 189–226.
4 N. R. Singh, W. N. Delgass, F. H. Ribeiro and R. Agrawal, Environ. Sci.
Technol., 2010, 44, 5298–5305.
5 Y. Roman-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic, Nature,
2007, 447, 982–U985.
6 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem., Int. Ed.,
2007, 46, 7164–7183.
7 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic, Science,
2005, 308, 1446–1450.
8 R. R. Davda, J. W. Shabaker, G. W. Huber, R. D. Cortright and
J. A. Dumesic, Appl. Catal. B: Environ., 2005, 56, 171–186.
9 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098.
50 P. A. Horne and P. T. Williams, Fuel, 1996, 75, 1043–1050.
10 Y. C. Lin and G. W. Huber, Energy Environ. Sci., 2009, 2, 68–80.
3124 | Green Chem., 2012, 14, 3114–3125
This journal is © The Royal Society of Chemistry 2012

View Article Online
51 A. A. Lappas, M. C. Samolada, D. K. Iatridis, S. S. Voutetakis and
I. A. Vasalos, Fuel, 2002, 81, 2087–2095.
52 T. Q. Hoang, X. L. Zhu, T. Sooknoi, D. E. Resasco and R. G. Mallinson,
J. Catal., 2010, 271, 201–208.
53 X. L. Zhu, L. L. Lobban, R. G. Mallinson and D. E. Resasco, J. Catal.,
2010, 271, 88–98.
54 T. A. Brandvold, United States, 2010.
60 A. S. Mamman, J. M. Lee, Y. C. Kim, I. T. Hwang, N. J. Park,
Y. K. Hwang, J. S. Chang and J. S. Hwang, Biofuels Bioprod. Bioref.,
2008, 2, 438–454.
61 R. Xing, W. Qi and G. W. Huber, Energy Environ. Sci., 2011, 4,
2193–2205.
62 R. Weingarten, G. A. Tompsett, W. C. Conner and G. W. Huber, J. Catal.,
2011, 279, 174–182.
55 K. P. C. Vollhardt and N. E. Schore, Organic Chemistry: Structure and
Function, W. H. Freeman, 2005.
56 C. L. Williams, C.-C. Chang, P. Do, N. Nikbin, S. Caratzoulas,
D. G. Vlachos, R. F. Lobo, W. Fan and P. J. Dauenhauer, ACS Catal.,
2012, 2, 935–939.
57 R. S. Rao, R. T. K. Baker and M. A. Vannice, Catal. Lett., 1999, 60,
51–57.
58 D. G. Manly and A. P. Dunlop, J. Org. Chem., 1958, 23, 1093–1095.
59 H. Y. Zheng, Y. L. Zhu, B. T. Teng, Z. Q. Bai, C. H. Zhang, H. W. Xiang
and Y. W. Li, J. Mol. Catal. A: Chem., 2006, 246, 18–23.
63 S. Al-Khattaf, Energy Fuels, 2006, 20, 946–954.
64 J. Cejka, J. Kotrla and A. Krejci, Appl. Catal., A-Gen., 2004, 277, 191–199.
65 T. P. Vispute, H. Y. Zhang, A. Sanna, R. Xiao and G. W. Huber, Science,
2010, 330, 1222–1227.
66 Y.-T. Cheng, J. Jae, J. Shi, W. Fan and G. W. Huber, Angew. Chem., Int.
Ed., 2012, 51, 1387–1390.
67 S. Al-Khattaf, Chem. Eng. Process., 2007, 46, 964–974.
68 M. Guisnet, N. S. Gnep and S. Morin, Microporous Mesoporous Mater.,
2000, 35–6, 47–59.
69 J. Wei, J. Catal., 1982, 76, 433–439.
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 3114–3125 | 3125