Integrated conversion of 1-butanol to 1,3-butadiene

Article abstract of DOI:10.1039/c8ra02977f

Renewed interest in production of 1,3-butadiene from non-petroleum sources has motivated research into novel production routes. In this study, we investigated an integrated process comprising 1-butanol dehydration over a γ-Al₂O₃ catalyst to produce a mixture of linear butenes, coupled with a downstream K-doped Cr₂O₃/Al₂O₃ catalyst to convert the butenes into butadiene. Linear butene yields greater than 90% are achievable at 360 °C in the dehydration step, and single-pass 1,3-butadiene yields greater than 40% are achieved from 1-butene in a N₂ atmosphere in the dehydrogenation step. In the integrated process, 1,3-butadiene yields are 10-15%. In all cases, linear C4 selectivity is greater than 90%, suggesting that 1,3-butadiene yields could be significantly improved in a recycle reactor. Doping the Cr₂O₃ catalyst with different metals to promote H₂ consumption in a CO₂ atmosphere did not have a large effect on catalyst performance compared to an undoped Cr₂O₃ catalyst, although doping with K in an N₂-diluted atmosphere and with Ni in a CO₂-enriched atmosphere showed slight improvement. In contrast, doping with K and Ca in a CO₂-enriched atmosphere showed slightly decreased performance. Similarly, employing a CO₂-enriched atmosphere in general did not improve 1,3-butadiene yield or selectivity compared to reactions performed in N₂. Overall, this study suggests that an integrated dehydration/dehydrogenation process to convert 1-butanol into 1,3-butadiene could be feasible with further catalyst and process development.

Full text of DOI:10.1039/c8ra02977f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Integrated conversion of 1-butanol to 1,3-
butadiene†
Cite this: RSC Adv., 2018, 8, 24068
*
Jacob S. Kruger,
Tao Dong, Gregg T. Beckham
and Mary J. Biddy
Renewed interest in production of 1,3-butadiene from non-petroleum sources has motivated research into
novel production routes. In this study, we investigated an integrated process comprising 1-butanol
dehydration over a g-Al2O3 catalyst to produce a mixture of linear butenes, coupled with a downstream
K-doped Cr2O3/Al2O3 catalyst to convert the butenes into butadiene. Linear butene yields greater than
90% are achievable at 360 ꢀC in the dehydration step, and single-pass 1,3-butadiene yields greater than
40% are achieved from 1-butene in a N2 atmosphere in the dehydrogenation step. In the integrated
process, 1,3-butadiene yields are 10–15%. In all cases, linear C4 selectivity is greater than 90%,
suggesting that 1,3-butadiene yields could be significantly improved in a recycle reactor. Doping the
Cr2O3 catalyst with different metals to promote H2 consumption in a CO2 atmosphere did not have
a large effect on catalyst performance compared to an undoped Cr2O3 catalyst, although doping with K
in an N2-diluted atmosphere and with Ni in a CO2-enriched atmosphere showed slight improvement. In
contrast, doping with K and Ca in a CO2-enriched atmosphere showed slightly decreased performance.
Similarly, employing a CO2-enriched atmosphere in general did not improve 1,3-butadiene yield or
selectivity compared to reactions performed in N2. Overall, this study suggests that an integrated
dehydration/dehydrogenation process to convert 1-butanol into 1,3-butadiene could be feasible with
further catalyst and process development.
Received 6th April 2018
Accepted 19th June 2018
DOI: 10.1039/c8ra02977f
rsc.li/rsc-advances
Several pathways to convert biomass to a renewable C4
stream exist commercially, incorporating syngas, methanol,
Introduction
and ethanol intermediates en route to 1-butanol, isobutanol,
succinic acid, or butanediol.2,6 The pathways to convert these
C4s into 1,3-butadiene are less technically developed, though
the embodied chemistries are relatively mature. Specically, 1-
butanol could be dehydrated to 1- and 2-butenes, and these
butenes could be dehydrogenated to 1,3-butadiene (Scheme 1).
A single-stage integrated process would be challenging due
to the different conditions required for alcohol dehydration and
olen dehydrogenation. In particular, olen dehydrogenation is
typically carried out at temperatures greater than 550 ꢀC and
1,3-Butadiene is a large-market commodity chemical used in
the manufacture of synthetic polymers and other materials.1–3
Historically, it has been produced as a byproduct of non-
catalytic steam cracking of the naphtha fraction of petroleum
for ethylene production, or by catalytic dehydrogenation (direct
or oxidative) of n-butane and n-butenes. The advent of inex-
pensive light hydrocarbons, such as shale gas, has led to non-
naphtha ethylene routes, concomitantly leading to supply
shortages and price uncertainty of butadiene.2,4,5 These uncer-
tainties, as well as environmental concerns surrounding fossil
feedstocks, have resulted in a renewed interest in “on-purpose”
catalytic dehydrogenation of C4s and in developing alternative
renewable pathways to 1,3-butadiene.1,2,4 Several of these path-
ways have been examined, starting in the early 1900s. For
example, 1,3-butadiene can be produced in yields up to 60%
from ethanol over Co or Cr/MgO catalysts, and butanediols can
be dehydrated to butadiene over several transition metal oxides,
though butadiene yields have generally been ꢁ25% or less.1
weight-hourly space velocities (WHSVs) greater than 150 hꢂ1
.
High temperature is required to activate the olens on the
dehydrogenation catalyst, and the high WHSV is required to
minimize unfavourable side reactions that lead to C1–C3
cracking products and coke. Additionally, dehydrogenation
catalysts (Cr2O3 or Fe2O3 supported on Al2O3) are typically
doped with K2CO3 or K2O, which neutralize catalyst acidity that
would be necessary for alcohol dehydration. Similarly, alcohols
tend to react by dehydrogenation at such temperatures to
produce an undesired carbonyl, rather than by dehydration to
produce a desirable olen.7 On the other hand, 1-butanol can be
dehydrated to linear butenes in greater than 95% yield over a g-
Al2O3 catalyst at temperatures of 350–410 ꢀC and WHSVs of 1–10
hꢂ1.8 This difference in reaction conditions may be one reason
National Renewable Energy Laboratory, National Bioenergy Center, 15013 Denver
West Parkway, Golden, CO 80401, USA. E-mail: Jacob.Kruger@nrel.gov
† Electronic supplementary information (ESI) available: Reactor temperature
prole, control reaction product distributions, additional doped catalyst results,
effect of H2O-rich. See DOI: 10.1039/c8ra02977f
24068 | RSC Adv., 2018, 8, 24068–24074
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Scheme 1 Dehydration of butanol to linear butenes, and dehydrogenation of butenes to 1,3-butadiene.
that it has been suggested,1 but never demonstrated, that an that lead to unfavourable cracking reactions, which in turn lead
integrated dehydration–dehydrogenation process to convert 1- to coke.9
butanol to 1,3-butadiene could be feasible.
Thus, in addition to demonstrating an integrated butanol-
We hypothesized that olen dehydrogenation could also be to-butadiene process, we were motivated to explore the
operated at lower temperature by decreasing the space velocity, potential of new equilibrium-shiing catalysts for butene
and that the lower temperature would also be advantageous in dehydrogenation. Herein we report process development for
mitigating side reactions. While cracking equilibria still 1-butene dehydrogenation over a series of Cr2O3/Al2O3 cata-
strongly favour C1–C3 products at temperatures as low as lysts doped with K, Ca, Ni, Mo, Fe, and Cu/CeO2, and inte-
450 ꢀC, the kinetics of these reactions are much slower.9 Thus, gration with 1-butanol dehydration.
we were motivated to explore the integrated conversion of 1-
butanol to 1,3-butadiene in a combined dehydration–dehydro-
genation reactor.
Experimental
Additionally, we were interested in improving yields in the
Materials
butene dehydrogenation step. Linear butenes can typically be
directly dehydrogenated over CrOx/Al2O3 or Fe2O3-containing
catalysts in single-pass yields of 40–50% or oxidatively dehy-
drogenated over ZnFe2O4, Bi2MoO6, or Sn/Sb-containing cata-
lysts in single-pass yields of 60–70%.1,9–12 The oxidative
dehydrogenation shis the dehydrogenation equilibrium
toward 1,3-butadiene by converting the produced H2 to H2O.
The presence of O2 and H2O concomitantly mitigate coke
formation on the catalyst surface.
In dehydrogenation without O2, alternative concepts for
shiing the dehydrogenation equilibrium may also be feasible.
In particular, (1) coke gasication or methanation, (2) CO2
methanation, and (3) reverse water-gas shi (RWGS) are three
reactions that could consume produced H2.
1-Butene (5.4 mol% in N2) and a GC calibration standard of 1-
butene, cis-2-butene, trans-2-butene, 1,3-butadiene, and n-
butane (1 mol% each in N2) were purchased from Matheson
Trigas. A GC calibration standard for CH4, C2H4, C2H6, C3H6,
C3H8, and 1-butene was purchased from Praxair. N2, CO2, and
zero-grade air were purchased from Airgas. 1-Butanol (99%)
and hydrated nitrate salts of Cr, K, Ca, Ni, Fe, Cu, and Ce were
purchased from Sigma-Aldrich, as was the oxalate salt of Mo.
The g-Al2O3 catalyst support was from Sasol (Puralox Nwa-
155). Inert quartz chips (nominally 30–50 mesh) were from
´
Dupre Minerals.
Catalyst synthesis
C + 2H2 / CH4
(1)
(2)
(3)
Catalysts were synthesized by depositing aqueous solutions of
the nitrate salts on a Sasol Puralox Nwa 155 Al2O3 support by
incipient wetness. The baseline catalyst was 8.7 wt% Cr2O3. The
doped catalysts had the same loading of Cr2O3 plus 1 wt% of the
dopant metal, unless otherwise specied, mixed as the nitrate
or oxalate salt in the same solution as the Cr. In the case of Cu/
CeO2, the dopant solution was prepared separately and applied
by incipient wetness to the baseline Cr2O3/Al2O3 catalyst. The
metal loading for that catalyst was 0.1 wt% Cu and 0.9 wt% Ce.
Aer depositing the salt solution, the catalysts were dried
under vacuum at 40 ꢀC overnight, then transferred to a ceramic
CO2 + 4H2 / CH4 + 2H2O
H2 + CO2 / H2O + CO
Of these three, reaction (3) has proven promising in dehy-
drogenation of ethylbenzene to styrene and diethylbenzene to
divinyl benzene13–15 and to some extent in butene
dehydrogenation.16–19
To this end, certain catalysts are known to promote these
desired reactions. In particular, coke gasication can be
promoted by K, Ca, and Ni;20 Ni, Fe, and Mo can hydrogenate
CO2 to CH4;21,22 and RWGS can be promoted by Fe2O3 and Cu/
CeO2.13,23 K and Ca can also help to poison catalyst acid sites
ꢀ
dish and calcined by the following program: ramp to 95 C at
25 ꢀC minꢂ1, hold for 1 h, ramp to 550 ꢀC at 5 ꢀC minꢂ1, hold for
10 h.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 24068–24074 | 24069

View Article Online
RSC Advances
Paper
heated to the reaction temperature using the control thermo-
couple in the thermowell, the wall temperature of the furnace
was the same as when the control thermocouple was embedded
in the catalyst bed, conrming that the temperature prole was
the same during measurement and reaction.
During reactor operation, a reaction-purge–regeneration-
purge cycle using four mass ow controllers was imple-
mented, with the reactor control soware automatically
switching between steps. The temperature, time, and ow
parameters used for each cycle unless otherwise indicated are
shown in Table 1.
For dehydration experiments, a 0.2 g bed of the same g-Al2O3
material used to support the Cr2O3 catalysts was positioned 5
inches (12.7 cm) above the centre of the reactor tube. At this
point, the reactor temperature was 360 ꢀC when the axial centre
of the reactor tube was at 450 ꢀC. The butanol ow rate was 0.02
mL minꢂ1, selected to approximate the conditions of the
5.4 mol% 1-butene mixture used in the dehydrogenation cata-
lyst screening experiments. The N2 ow rate was 177 sccm. For
integrated dehydration–dehydrogenation experiments, the
conguration for both sets of experiments was combined. A
0.2 g bed of g-Al2O3 was positioned 5 inches (12.7 cm) above the
reactor centre and a 0.6 g bed of K–Cr2O3/Al2O3 was positioned
at the reactor centre, the 1-butanol ow rate was 0.02 mL minꢂ1
and the N2 ow rate was 177 sccm.
Catalyst characterization
Catalysts were characterized by N2 physisorption to determine
surface area, by NH3 temperature programmed desorption
(TPD) to quantify acid sites, and by X-ray diffraction (XRD) to
evaluate crystal structure. N2 physisorption was carried out on
a Quantachrome Quantisorb SI four-station instrument. Prior to
analysis, catalysts were outgassed under vacuum at 350 ꢀC
overnight. Eight adsorption points for each catalyst were
recorded to establish an isotherm from which the surface area
was calculated, using a 30 s equilibration time for each point.
NH3 pulse chemisorption was carried out on an Altamira
Instruments AMI-390 unit. Samples were treated for 2 h at
450 ꢀC under 50 mL minꢂ1 Ar, then cooled to 120 ꢀC under
owing He before dosing the sample with 25 ꢃ 5 mL pulses of
10% NH3 in He. The average of the post-saturation pulses
(typically the last 15–20 pulses) was used as the reference peak
area. The adsorbed peak area was calculated as the sum of the
difference between the observed peak area and the reference
peak area for the unsaturated pulses (typically rst 5–10 pulses).
The NH3 areas were quantied on a TCD and converted to
a quantity of adsorbed molecules assuming ideal gas behaviour
of the pulse gas.
XRD was carried out on a Rigaku Ultima IV X-ray diffrac-
tometer using Cu K-a radiation. The operating voltage and
current were 40 kV and 44 mA, respectively. A scan range of 5–
80ꢀ 2q, a scan speed of 5ꢀ minꢂ1 and a point spacing of 0.05ꢀ 2q
were used.
Reactor effluent passed through a condenser and then to
local exhaust ventilation, with a slip stream sampled by online
GCMS. The system consisted of an Agilent 6890 Plus GC
equipped with a TCD, FID, and a 5973 MS, which analysed
samples in parallel. The GC column was a 30 m ꢃ 0.32 mm ID
GS-GASPRO column, operating in ramped ow mode with the
following program: 2.3 mL minꢂ1 for 3 min, ramp at 1 mL
minꢂ2 to 2.7 mL minꢂ1, hold at 2.7 mL minꢂ1. The corre-
sponding oven program was 50 ꢀC for 3 min, ramꢂp1 at
Reactor system
The reactor system was a xed-bed downow stainless steel
tubular reactor, 0.500 ID and 1400 length. Unless otherwise spec-
ied, the catalyst bed (0.6 g) was positioned in the middle of the
reactor tube, with 30–50 mesh SiO2 packed upstream and
downstream to facilitate mixing and heat transfer. The particles
were held in place by a plug of quartz wool on either end of the
reactor, and a thermocouple was placed in the middle of the
catalyst bed. In the case of the sequential dehydration–dehy-
drogenation, the thermocouple tip was in the middle of the
dehydrogenation catalyst bed. The reactor was brought up to
temperature under owing N2, and typically allowed to equili-
brate overnight before starting a reaction. To measure the
temperature prole of the reactor, the top of the reactor was
tted with a thermowell that accompanied two K-type thermo-
couples. A control thermocouple remained in the centre of the
tube while the other moved from the bottom of the tube to the
top of the tube in 0.500 increments, and waiting 30–60 s at each
point for the temperature to stabilize. When the reactor was
ꢂ1
ꢀ
ꢀ
ꢀ
15 C min to 75 C, hold for 2 min, ramp at 50 C min to
250 ꢀC, hold for 2.83 min. The inlet conditions were 250 ꢀC,
initial pressure of 10.64 psi, and a split ratio of 20 : 1. The
carrier gas for the system was He. Hydrocarbons were quanti-
ed on the FID, using N2 (detected on the TCD) as an internal
standard, and gas mixes of authentic standards for C1–C4
compounds to develop response factors. For C5 and heavier
compounds, FID response factors were calculated based on the
method of Scanlon,24 though these compounds typically
comprised less than 0.5% carbon yield.
Yields and selectivities are reported on a carbon molar basis.
Selectivities are based on the amount of butene fed rather than
as a fraction of products detected.
Table 1 Reactor parameters for butene dehydrogenation
Parameter
Reaction
450 ꢀ
Reaction purge
450 ꢀ
Regeneration
Regeneration purge
450 ꢀ
Temperature
Duration
C
C
450 ꢀC
74 min
C
72 min
6 min
6 min
Gas feed
100 sccm N2 or CO2 +
100 sccm 1-butene mix
200 sccm N2 or CO2
100 sccm Ar + 100 sccm zero air
200 sccm Ar
24070 | RSC Adv., 2018, 8, 24068–24074
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Screening of butene dehydrogenation reaction conditions
Results
To establish reaction conditions for butene dehydrogenation
that would maximize butadiene selectivity, maximize single-
pass yield, and minimize cracking reactions, we explored
different temperatures and space velocities with the baseline
Cr2O3/Al2O3 catalyst. The results of these experiments are
shown in Fig. 2. Selectivity and yield are optimal at 450 ꢀC
and 0.76 hꢂ1 weight hourly space velocity (WHSV), while both
lower and higher temperatures show lower selectivity. At
400 ꢀC, the lower selectivity is likely due to coke formation as
no other compounds were observed in the effluent stream,
Butanol dehydration
We initially explored butanol dehydration under conditions
similar to those reported by Pines and Haag.8 Using a tempera-
ture of 360 ꢀC and WHSV of 4.86 hꢂ1 (0.2 g g-Al2O3 catalyst bed
and 0.02 mL minꢂ1 butanol ow rate), we were able to obtain
>95% carbon balance and 91% yield to linear butenes over
ꢁ100 min time-on-stream, as shown in Fig. 1. These were the
conditions we employed in the integrated dehydration–dehy-
drogenation experiments.
We also explored butanol reactivity at 450 ꢀC, consistent with
the butene dehydrogenation conditions reported below.
Butanol conversion in a SiO2-packed tube (without catalyst) was
100%, but selectivity to butenes was low, and a signicant
amount of black solid was deposited in the reactor. Because
conversion of butenes under these conditions was negligible
(see below), the high conversion of butanol suggests thermal
reactions through non-butene routes predominated, such as
cracking, coke formation, dehydrogenation to butyraldehyde,
and decarbonylation of butyraldehyde to produce propene and
CO.
ꢀ
while at 500 C, both coke formation and cracking reactions
play a role, as C1–C3 hydrocarbons were detectable, along
with small amounts of C5 hydrocarbons, benzene, and
toluene. These results show that high butadiene yields and
selectivities can be obtained at temperatures signicantly
lower than those typically used for direct butene dehydro-
genation. Notably, following the analysis of Tyuryaev,9 the 1-
butene dehydrogenation equilibrium at a 1-butene partial
pressure of 0.027 atm and temperatures of 400 ꢀC and 450 ꢀC
would give 18–22% and 35–39% yields to butadiene, respec-
tively. At 500 ꢀC, 57–60% butadiene yields would be pre-
dicted. Thus, it appears that the butadiene yield at 400–
450 ꢀC is limited by equilibrium, while at 500 ꢀC, it is limited
by side reactions.
We attempted to mitigate these nonselective reactions by
inserting a g-Al2O3 catalyst bed upstream of the axial tube
centre to convert the butanol to butenes at 450 C. However,
ꢀ
when the g-Al2O3 bed was positioned within the isothermal
zone of the reactor, a signicant amount of isobutene was
produced through skeletal isomerization. Thus, we measured
the temperature prole within the reactor tube and placed the
g-Al2O3 bed 5 inches (12.7 cm) above the axial centre of the
reactor, at which point the temperature was 360 ꢀC (Fig. S1†),
consistent with Pine and Haags.8 This was the location of the
g-Al2O3 bed in the integrated experiments.
Thus, we established 450 ꢀC and a space velocity of 0.76 hꢂ1
WHSV as the operating conditions for our remaining experi-
ments. Notably, the butadiene selectivity and yields are
comparable to common industrial catalysts reported for direct
dehydrogenation.9 Under these same conditions, a control
reaction (reactor tube packed with only SiO2) showed <5%
conversion of 1-butene (Fig. S2†).
Fig. 1 Dehydration of 1-butanol to 1- and 2-butenes over a g-Al2O3 Fig. 2 Yields and selectivity for butene dehydrogenation over a Cr2O3/
catalyst. Reaction conditions: 0.02 mL minꢂ1 butanol, 177 sccm N2, Al2O3 catalyst. Results calculated at 50 min time on stream with
0.2 g catalyst, 350–400 ꢀC.
a 2.7 mol% concentration of 1-butene in N2.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 24068–24074 | 24071

View Article Online
RSC Advances
Paper
Effect of catalyst dopant and reaction atmosphere
Integration of butanol dehydration and butene
dehydrogenation
In the interest of improving the butene dehydrogenation
portion of the reactor by subsequently reacting produced H2, With these results, we were motivated to integrate the dehy-
we tested a number of 1 wt% doped catalysts in N2 and CO2- dration and dehydrogenation reactions in a single reactor
rich atmospheres. The feed composition of these atmo- tube. The reactor conguration consisted of a 0.6 g bed of the
spheres were 97.3% N2 and 47.3%N2/50.0% CO2, respectively, 1 wt% K-doped Cr2O3/Al2O3 catalyst at the centre of the reactor
with 2.7% 1-butene in both cases. For two reaction cycles in tube (at 450 ꢀC), a 0.2 g bed of g-Al2O3 positioned 5 inches (12.7
N2, two in CO2, and one in N2, the doped catalysts were cm) above the centre of the tube (at 360 ꢀC), a butanol ow of
generally comparable to the undoped Cr2O3/Al2O3 catalyst, 0.02 mL minꢂ1, and an N2 ow of 177 sccm. Under these
except that the K-doped catalyst was slightly better in N2, the conditions, we were able to obtain ꢁ13% single-pass yields
Ni-doped catalyst was slightly better in CO2, and the K- and over 10 h TOS, as shown in Fig. 4. The gap between 200 and
Ca-doped catalysts were slightly worse in CO2. These results 300 min TOS represents a regeneration step, though the
are shown in Fig. 3. We also tested Ni- and Mo-doped cata- catalyst performance appeared to be decreasing only slightly
lysts at 0.25 wt% and 5 wt% metal dopant loadings, but the with TOS. In dehydrogenation of 1-butene, catalyst perfor-
results were similar to the 1 wt% loading (Fig. S3†). The mance decreased more rapidly (though maintaining higher
corresponding selectivities are also quite high, with several overall yields). Thus, the integrated process with butanol feed
catalysts giving greater than 80% selectivity at 30 min time- appears to be more stable than just the dehydrogenation. The
on-stream (TOS, Fig. S4†).
reason for this difference is not clear at this point, but it may
The CO2-enriched atmosphere did not lead to higher be that the oxygen in the butanol feed is able to mitigate coke
butadiene yields as hypothesized, despite the more favourable formation. Because this oxygen would likely only see the
equilibria for H2 conversion. (For example, correlations for the dehydrogenation catalyst as H2O, it is likely that any coke-
water-gas shi equilibrium25 at 1 atm and 450 ꢀC predict mitigating function is due to H2O rather than 1-butanol
greater than 98% conversion of produced H2 in a 50% CO2 specically.
atmosphere). In these reactions, there was not a signicantly
The yield of butadiene in the integrated process is lower than
higher amount of CH4 produced (as would be expected from expected from the previous runs, which produced >40% yields
methanation of C or CO2), nor was any condensate collected of butadiene from 1-butene at 30 min TOS. We tentatively
from the reactor knockout pot, (as may be expected from H2O ascribe the lower yields to the water sensitivity of Cr2O3/Al2O3
production via enhanced RWGS activity). Thus, it seems that catalysts,9 as H2O is produced in the dehydration step. Indeed,
the catalysts synthesized here are either not capable of acti- dehydrogenation activity was much lower when co-feeding 0.02
vating CO2 or not capable of catalysing the desired reactions at mL minꢂ1 of H2O with the 1-butene/N2 mixture over these
450 ꢀC. Further catalyst development will be the focus of future catalysts (Fig. S5†)
work.
Notably, while the single-pass butadiene yield is only 13%,
the average selectivity to linear C4 compounds is >95%
(Fig. S6†). The remaining mass is likely due to coke formation,
Fig. 3 Conversion and selectivity in 1-butene dehydrogenation over
1 wt% metal-doped Cr2O3 catalysts, interpolated to 30 min time-on-
stream. Left-most set: reaction cycles 1–2 in N2. Centre set: reaction
cycles 3–4 in N2/CO2. Right-most set: reaction cycle 5 in N2.
Fig. 4 Integrated butanol dehydration and butene dehydrogenation.
24072 | RSC Adv., 2018, 8, 24068–24074
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Table 2 Surface area and acidity of selected catalysts synthesized in
this work
Conflicts of interest
There are no conicts to declare.
Catalyst
BET surface area (m2 gꢂ1
)
NH3 uptake (mmol gꢂ1
)
Al2O3
Cr2O3
1K-Cr2O3
153
139
137
242
424
334
Acknowledgements
This work was authored by the National Renewable Energy
Laboratory (NREL), operated by Alliance for Sustainable Energy,
LLC, for the U.S. Department of Energy (DOE) under Contract
No. DE-AC36-08GO28308. This work was supported by the
Laboratory Directed Research and Development (LDRD)
Program at NREL. The views expressed in the article do not
necessarily represent the views of the DOE or the U.S. Govern-
ment. The U.S. Government retains and the publisher, by
accepting the article for publication, acknowledges that the U.S.
as some CO2 could be detected during the regeneration cycle.
Thus, with a recycle reactor, the overall achievable yield of
butadiene from butanol would be nearly quantitative using the
present conditions.
Catalyst characterization
Government retains
a nonexclusive, paid-up, irrevocable,
We characterized the fresh catalysts for surface area, acidity,
and phase composition. Catalyst surface area and acidity are
shown in Table 2, while phase composition is shown in Fig. S7.†
The g-Al2O3 support had a surface area of 153 m2 gꢂ1, which
was reduced slightly aer deposition of Cr2O3 and calcining to
decompose the precursor salt. The presence of K in the Cr
precursor solution does not signicantly decrease the surface
area further. Because high-valent Cr catalysts are known to
decompose NH3 at temperatures above about 500 ꢀC,26 we
quantied acid sites by NH3 pulse chemisorption instead of
NH3 TPD. The NH3 uptake on the g-Al2O3 support suggests an
acid site content of 242 mmol sites/g. Addition of Cr2O3
increases the acid site content to 424 mmol sites/g, while the
presence of K in the Cr precursor solution mitigates the
increase in acid sites to 334 mmol sites/g. Doping Cr oxide
catalysts with K has been hypothesized to poison acid sites,9
though recent work has suggested that K also changes the
surface morphology of the supported Cr oxide phase.27 Any
structural changes were not apparent by XRD, as only peaks for
the g-Al2O3 support could be detected in the prepared catalysts
(Fig. S7†).
worldwide license to publish or reproduce the published form
of this work, or allow others to do so, for U.S. Government
purposes.
Notes and references
1 E. V. Makshina, M. Dusselier, W. Janssens, J. Degreve,
P. A. Jacobs and B. F. Sels, Chem. Soc. Rev., 2014, 43, 7917–
7953.
2 M. J. Biddy, C. Scarlata and C. Kinchin, Chemicals from
Biomass: A Market Assessment of Bioproducts with Near-Term
Potential. Technical Report NREL/TP-5100-65509, National
Renewable Energy Laboratory, 2016.
3 Z. Xi and W.-X. Zhang, Synlett, 2008, 2008, 2557–2570.
4 M. D. Jones, Chem. Cent. J., 2014, 8, 53.
5 P. C. A. Bruijnincx and B. M. Weckhuysen, Angew. Chem., Int.
Ed., 2013, 52, 11980–11987.
6 Nexant, Final Report: Renewable Chemicals & Materials
Opportunity Assessment, Major Job Creation and Agricultural
Sector Engine, Prepared for US Department of Agriculture, 2014.
7 J. S. Kruger, R. Chakrabarti, R. J. Hermann and
L. D. Schmidt, Appl. Catal., A, 2012, 411, 87–94.
8 H. Pines and W. O. Haag, J. Am. Chem. Soc., 1961, 83, 2847–
2852.
Conclusions
9 I. Y. Tyuryaev, Russ. Chem. Rev., 1966, 35, 59.
¨
We have integrated 1-butanol dehydration with butene dehy- 10 J. Grub and E. Loser, in Ullmann's Encyclopedia of Industrial
drogenation to produce 1,3-butadiene in a one-tube, two-stage
process. Dehydration of 1-butanol gives linear butenes in
Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, 2000,
DOI: 10.1002/14356007.a04_431.
>90% yield over a g-Al2O3 catalyst at 350–410 ꢀC. Dehydroge- 11 R. Julius, F. Robert, A. Claudia and D. Olaf, AIChE J., 2017,
nation of 1-butene gives single-pass butadiene yields greater
63, 43–50.
than 40% at 450 ꢀC for a 2.7 vol% butene feed in N2, with 12 H. Lee, J. C. Jung, H. Kim, Y.-M. Chung, T. J. Kim, S. J. Lee,
selectivity greater than 80% at 30 min TOS. The integrated
process demonstrated here gives an average single-pass buta-
S.-H. Oh, Y. S. Kim and I. K. Song, Catal. Commun., 2008, 9,
1137–1142.
diene yield of 13%, with overall C4 selectivity greater than 95%. 13 S. Chen, A. Sun, Z. Qin and J. Wang, Catal. Commun., 2003, 4,
With a recycle reactor conguration, overall yields of 1-butanol 441–447.
to butadiene would be much higher. Additionally, these results 14 L. Zhang, Z. Wu, N. C. Nelson, A. D. Sadow, I. I. Slowing and
are obtained at temperatures signicantly lower than those S. H. Overbury, ACS Catal., 2015, 5, 6426–6435.
traditionally used for direct butene dehydrogenation, repre- 15 Y. Yue, L. Zhang, J. Chen, D. K. Hensley, S. Dai and
senting a potential route for decreasing process severity in S. H. Overbury, RSC Adv., 2016, 6, 32989–32993.
butadiene production. Thus, the process developed here is an 16 W. Yan, Q. Y. Kouk, J. Luo, Y. Liu and A. Borgna, Catal.
important step in the production of renewable butadiene.
Commun., 2014, 46, 208–212.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 24068–24074 | 24073

View Article Online
RSC Advances
Paper
17 W. Yan, J. Luo, Q.-Y. Kouk, J. E. Zheng, Z. Zhong, Y. Liu and 23 X. Qi and M. Flytzani-Stephanopoulos, Ind. Eng. Chem. Res.,
A. Borgna, Appl. Catal., A, 2015, 508, 61–67. 2004, 43, 3055–3062.
18 W. Yan, Q.-Y. Kouk, S. X. Tan, J. Luo and Y. Liu, J. CO2 Util., 24 J. T. Scanlon and D. E. Willis, J. Chromatogr. Sci., 1985, 23,
2016, 15, 154–159. 333–340.
19 Y. Gao, B. Wang, B. Yan, J. Li, F. Alam, Z. Xiao and T. Jiang, 25 C. Callaghan, PhD thesis, Worcester Polytechnic Institute,
React. Kinet., Mech. Catal., 2017, 122, 451–462. 2006.
20 A. A. Lizzio and L. R. Radovic, Ind. Eng. Chem. Res., 1991, 30, 26 L. Li, Z. H. Zhu, S. B. Wang, X. D. Yao and Z. F. Yan, J. Mol.
1735–1744. Catal. A: Chem., 2009, 304, 71–76.
21 T. A. Le, M. S. Kim, S. H. Lee, T. W. Kim and E. D. Park, Catal. 27 V. S. Sullivan, S. D. Jackson and P. C. Stair, J. Phys. Chem. B,
Today, 2017, 293, 89–96.
2005, 109, 352–356.
¨
¨
22 S. Ronsch, J. Schneider, S. Matthischke, M. Schluter,
¨
M. Gotz, J. Lefebvre, P. Prabhakaran and S. Bajohr, Fuel,
2016, 166, 276–296.
24074 | RSC Adv., 2018, 8, 24068–24074
This journal is © The Royal Society of Chemistry 2018