Ethanol conversion into butadiene over Zr-containing molecular sieves doped with silver

Article abstract of DOI:10.1039/c4gc02202e

The synthesis of butadiene-1,3 from ethanol over silver doped (1 wt%) Zr-containing molecular sieves ZrBEA, ZrMCM-41 and ZrO₂ supported on silica has been studied. The catalysts were characterized by X-ray diffraction, nitrogen adsorption-desorption, X-ray photoelectron spectroscopy, ²⁹Si MAS NMR, SEM/EDX, TEM and FTIR of adsorbed CD₃CN. The activity of the catalysts was found to increase in the following range: Ag/ZrO₂/SiO₂ < Ag/ZrMCM-41 < Ag/ZrBEA. This activity range correlated with the content of Lewis acid sites in the catalysts determined by FTIR of the adsorbed CD₃CN. The best catalyst performance in terms of butadiene yield was observed for ZrBEA (Si/Zr = 100) promoted with silver, which showed butadiene selectivity of 56% at ethanol conversion of 48%.

Full text of DOI:10.1039/c4gc02202e

Green Chemistry
View Article Online
View Journal
PAPER
Ethanol conversion into butadiene over
Zr-containing molecular sieves doped with silver†
Cite this: DOI: 10.1039/c4gc02202e
a
a,b
c
Vitaly L. Sushkevich, Irina I. Ivanova* and Esben Taarning
The synthesis of butadiene-1,3 from ethanol over silver doped (1 wt%) Zr-containing molecular sieves
ZrBEA, ZrMCM-41 and ZrO supported on silica has been studied. The catalysts were characterized by
X-ray diffraction, nitrogen adsorption–desorption, X-ray photoelectron spectroscopy, Si MAS NMR,
SEM/EDX, TEM and FTIR of adsorbed CD CN. The activity of the catalysts was found to increase in the fol-
lowing range: Ag/ZrO /SiO2 < Ag/ZrMCM-41 < Ag/ZrBEA. This activity range correlated with the content
2
2
9
3
Received 11th November 2014,
Accepted 12th February 2015
2
3
of Lewis acid sites in the catalysts determined by FTIR of the adsorbed CD CN. The best catalyst perform-
DOI: 10.1039/c4gc02202e
ance in terms of butadiene yield was observed for ZrBEA (Si/Zr = 100) promoted with silver, which
showed butadiene selectivity of 56% at ethanol conversion of 48%.
www.rsc.org/greenchem
unfavorable and were stopped because butadiene was very
efficiently produced from petrochemical sources.
1
. Introduction
Butadiene is one of the most important building blocks for
Nowadays the interest in the technology of on-purpose
the production of a wide variety of synthetic rubbers, elasto- butadiene production from ethanol is growing again. Recent
9
,10
mers and resins. Two routes are mainly pursued for its indus- studies
have demonstrated that the bio-based process could
trial synthesis: isolation from naphtha steam cracker fractions be a promising substitute for the dominant naphtha-based
of paraffinic hydrocarbons or the catalytic and oxidative de- method. However, although significant achievements have
1
hydrogenation of n-butane and n-butene. The first route is the been made towards a fundamental understanding of the
most important, accounting for over 95% of worldwide buta- process and the development of novel efficient catalysts during
2
11–19
11,12
diene production. But the increased production of shale gas, the last few years,
which contains substantial amounts of ethane, has led to a there is still ample room for catalyst improvement.
shift in the feedstock for steam crackers. This circumstance One of the most efficient ways to design and improve cata-
the recent reviews
point out that
will cause a shortage of the longer-chain hydrocarbons such as lysts involves the determination of the key reaction steps
butadiene in the immediate future. Therefore the development leading to the target product, the selection of the optimum
of an alternative, cheap and sustainable process for butadiene catalyst components for each reaction step, the merging of
synthesis from biomass feedstocks is highly desirable.
these components into one multifunctional catalyst and balan-
One of the most promising sources for butadiene pro- cing their functions to achieve the highest activity of, and
duction is ethanol, which can be produced from carbohydrate selectivity to the target product. Our recent kinetic studies on
1
3
biomass. The one-step catalytic process to produce butadiene metal containing oxide catalysts have revealed that among
3
,4
from ethanol was developed by Lebedev and commercialized the five reaction steps which constitute the target reaction
in the USSR in the beginning of the 1920s. Later, Carbide and pathway leading to butadiene (Scheme 1, ref. 11–14,19), steps
Carbon Chemicals Corporation commercialized, in the USA,
5
–7
8
a two-step process
based on Ostromislenskiy reaction.
However after the 1960s, both processes became economically
a
Department of Chemistry, Lomonosov Moscow State University, Leninskye Gory 1,
bld. 3, 119991 Moscow, Russia. E-mail: iiivanova@phys.chem.msu.ru;
Fax: +7(495)939-3570; Tel: +7(495)939-3570
b
A.V. Topchiev Institute of Petrochemical Synthesis, Russian Academy of Science,
Leninskiy prospect, 29, 119991 Moscow, Russia
c
Haldor Topsøe A/S, Nymoellevej 55, 2800 Kgs. Lyngby, Denmark
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4gc02202e
Scheme 1 Main reaction pathway leading to butadiene.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
1
, 2 and 4, namely, ethanol dehydrogenation, acetaldehyde Optima 3000 (Varian Vista). N
2
sorption–desorption isotherms
condensation and Meerwein–Ponndorf–Verley–Oppenauer were measured at 77 K using a Quantachrome Autosorb auto-
MPVO) reduction of crotonaldehyde with ethanol, can be con- matic surface area and a pore size analyzer.
sidered as the key reaction steps. Dehydration of 3-hydroxy- Powder X-ray diffraction (XRD) patterns of the samples were
butanal and crotyl alcohol was shown to proceed very easily, obtained using a Phillips X’Pert Diffractometer with Cu-Kα
(
1
3
even on the weak acid sites of the silica support. The investi- radiation at a wavelength of 1.5456 Å.
2
0–22
gation of the key reaction steps
suggested that Ag-contain-
TEM images were obtained using a JEOL JEM 2010 electron
microscope operating at 200 kV. SEM/EDX analysis was per-
2
0
ing metal sites are very efficient in ethanol dehydrogenation,
whereas Zr-containing Lewis acid sites proved to be very active formed using an Oxford Instruments MIRA3 TESCAN
2
1
and selective in aldol condensation of acetaldehyde and the microscope.
2
2
MPVO reaction of crotonaldehyde with ethanol. The interplay
XPS spectra were recorded on an Axis Ultra DLD (Kratos
between the metal promoter and the acidic sites has been Analysis) spectrometer using an Al Kα X-ray source. Due to the
addressed in ref. 13,20. In this paper, we aimed at elucidating charging of the insulating samples, the binding-energy values
the effect of different Zr-containing Lewis sites, keeping the were referenced to the C 1s line at 284.5 eV (arising from the
same the type and the content of the metal promoter (1 wt% inadvertent carbon contamination).
2
9
Ag). For this purpose, Zr-containing molecular sieves, which
Si solid-state MAS NMR was performed using a Bruker
2
2
proved to be very active in the MPVO reaction have been syn- Avance-400 spectrometer operating at a resonance frequency of
thesized and investigated in the overall reaction. The results 79.46 MHz with a spinning rate of 10 kHz, pulse length of
2
studied in the previous 3 µs, and recycle time of 20 s. The Si chemical shifts are
work. This paper is the first report on the catalytic properties reported relative to TMS.
2
9
are compared with Ag/ZrO
2
/SiO
1
3
of Ag-promoted ZrBEA and ZrMCM-41 molecular sieves in the
ethanol conversion into butadiene.
The acidic properties were studied by FTIR spectroscopy of
adsorbed CD CN. IR spectra were recorded with a Nicolet
Protégé 380 FT-IR spectrometer at a 4 cm optical resolution.
Prior to the measurements, the catalysts were pressed in self-
supporting discs and activated in the IR cell attached to a
3
−
1
2
. Experimental
2
.1 Catalyst preparation
vacuum line at 723 K for 4 h. Adsorption of CD CN was carried
3
ZrBEA zeolites with the Si/Zr ratio within 100–400 were syn-
thesized using the procedure described in ref. 23. ZrMCM-41
was prepared by the modified procedure proposed by Li
out in the room temperature cell. The pressure was measured
using a Barocell gauge. Difference spectra were obtained by
the subtraction of the spectra of the activated catalyst samples
from the spectra of the samples with adsorbate. The subtrac-
tion was carried out using the OMNIC 7.3 package.
2
4
et al., in particular, tin chloride pentahydrate (SnCl
4 2
·5H O)
was replaced by zirconyl chloride octahydrate (ZrOCl ·8H O).
2
2
The molar composition of the gel for the synthesis of
ZrMCM-41 was 1Si : 0.005Zr : 0.35CTABr : 0.21TMA : 0.07Cl :
2.3 Catalyst evaluation
−
Catalytic experiments were carried out in a flow-type fixed-bed
reactor under atmospheric pressure. In a typical experiment,
3
1.5H O. The hydrothermal synthesis was carried out at 413 K
2
for 15 hours. The sample was recovered by filtration, washed
with ample water, dried overnight at 353 K and calcined at
2
g of the catalyst (fraction 0.5–1 mm) was packed into the
quartz tubular reactor and purged with nitrogen at 873 K for
.5 h. Ethanol (95 wt%) was used as a feed. The reaction
8
93 K for 5 hours under a flow of air.
The ZrO /SiO catalyst was prepared by incipient wetness
0
2
2
mixture was fed using a Razel syringe pump. Helium was used
as a carrier gas (molar ratio EtOH/He = 1). The WHSV varied
from 0.12 to 0.64 h− , and the reaction temperature was 600 K.
Gaseous products were analyzed online using a Crystal 2000M
gas chromatograph using the 50 m SE-30 column. Liquid pro-
ducts were separated and analyzed using 50 m SE-30 and 2 m
Porapak Q columns. Methane was used as an external standard
for gaseous products.
impregnation of silica gel (Karpov Chemical Plant) with an
aqueous solution of zirconyl nitrate to attain a ZrO content of
2
1
∼1.0 wt%. After impregnation, the catalyst was dried at 393 K
and calcined at 773 K for 3 h under a flow of air.
All the catalysts were modified with 1 wt% of silver by inci-
3
pient wetness impregnation with AgNO solution followed by
calcination and reduction. Elemental analysis confirmed the
silver content within 1.1–0.9 wt%.
The conversion of ethanol, and selectivity towards the pro-
ducts and yields were calculated as follows:
The molecular sieve samples are assigned as ZrSieve_type
(
x), where x is the Si/Zr ratio. The catalysts promoted with
silver are labeled as Ag/ZrSieve_type(x) or Ag/ZrO /SiO .
mreactedðEtOHÞ
2
2
Ethanol conversion ¼
ꢀ 100%
m
fedðEtOHÞ
2
.2 Catalyst characterization
mj
The elemental analysis was performed using inductively
coupled plasma atomic emission spectroscopy (ICP-AES). Prior
to the analysis, the samples were dissolved in a mixture of HF,
F
j
k
ꢀ 100%
Mrj
Selectivity ¼ m
=46
reactedðEtOHÞ
HCl, HNO
3 3 4
and H PO , followed by neutralization with
Yield ¼ Ethanol conversion ꢀ Selectivity=100;
B(OH) . The solution was analyzed using a Perkin Elmer
3
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
where m
j
– mass of j-product in the reaction mixture, Mr
j
–
Nitrogen adsorption–desorption data point to a very high
F
molecular weight of j-product, kj – number of ethanol mole- surface area of zeolitic and mesoporous catalysts (Table 1).
cules required for product j formation. ZrBEA(200) has a reversible Type-I adsorption/desorption iso-
therm with a steep rise at p/p < 0.01, typical for microporous
0
solids (Fig. 1). Similar isotherms are observed for ZrBEA cata-
lysts with different Si/Zr ratios (not shown in Fig. 1). The
ZrMCM-41 sample shows the Type-IV isotherm with a step at
3
. Results and discussion
3
.1. Catalyst structure and composition
p/p ∼ 0.35, corresponding to the uniformly sized mesopores
0
The chemical and textural characteristics of the catalyst
precursors are presented in Table 1. The amount of ZrO incor-
porated into ZrBEA, ZrMCM-41 and loaded onto silica is close
to the expected values.
of ca. 35 Å in diameter. The silica supported ZrO
2
reveals
2
−1
2
surface areas of 270 m
g
and the average pore sizes of
∼100 Å, typical for amorphous silica gels.
The structure of the catalysts was analyzed by XRD tech-
niques (Fig. 2). The powder XRD patterns of the ZrBEA(200)
sample show typical features of well-crystalline zeolites. For
ZrBEA, the asymmetry of the peak in the range of 2θ = 7°–9°
indicates the presence of two isostructures, polymorphs A and
Table 1 Catalyst characteristics
2
3
Surface Total pore Micropore Average pore B. No peaks due to crystalline ZrO or any other crystalline
2
Si/Zr area
volume
volume
diameter
(Å)
impurity phases are detected. The ZrBEA samples with Si/Zr =
00 and 400 (not presented in Fig. 2) do not show any differ-
Sample
ratio (m g−1) (cm g−1)
2
3
(cm g−1)
3
2
ZrBEA(100)
ZrBEA(200)
ZrBEA(400)
134
263
455
470
470
465
870
270
0.29
0.30
0.30
1.1
0.20
0.20
0.20
—
7
7
7
35
100
ences with respect to ZrBEA(200). The low angle XRD patterns
of ZrMCM-41 show one major (100) reflection and two minor
reflections corresponding to the (110) and (220) planes, point-
ing to a highly ordered hexagonal pore structure. The ZrO2/
ZrMCM-41(200) 184
ZrO /SiO (200) 195
2
2
0.70
—
SiO
2
catalyst exhibits XRD patterns typical for amorphous
materials (Fig. 2).
The local structure of Zr and the incorporation of Zr into
2
9
the silica frameworks were studied using XPS and Si MAS
NMR techniques. The photoelectron peaks for Zr 3d observed
over the ZrBEA, ZrMCM-41 and ZrO /SiO samples are shown
2
2
in Fig. 3. The binding energy of Zr3d5/2 (182.9 eV) observed for
these samples is significantly higher than that of bulk ZrO
2
2
3,25
(
182.2 eV).
This shift towards higher values can be due to
the formation of Si–O–Zr bonds, i.e. the incorporation of Zr
2
6
into the framework of zeolites, MCM-41 or silica. The higher
binding energy of Zr 3d indicates a higher positive charge on
the Zr incorporated into the silica framework due to the lower
electronegativity of Zr than that of Si. These observations are
2
6
in line with the results previously reported for ZrBEA,
2
7
25
2 2
ZrMFI and ZrO /SiO .
2
9
Fig. 4 displays the Si MAS NMR spectra for Zr-containing
materials. ZrBEA(200) shows resonances in the range of −110
4
Fig. 1 Nitrogen adsorption/desorption isotherms obtained at 77 K.
to −120 ppm typical for Q species. These signals can be
Fig. 2 XRD patterns of Zr-containing catalysts.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
assigned to the Si(4Si) sites located in different crystallo-
The spectrum observed for ZrMCM-41 is typical for
graphic positions of zeolite BEA. The spectrum shows a rela- MCM-41 and other amorphous silicas, which do not have
tively complex feature due to the superposition of several short range ordering. The signal at ca. −108.5 ppm is due to
4
resonances for nonequivalent T crystallographic sites in BEA Q (Si(4Si)) species and a shoulder at ca. −100 ppm points to
3
28
29
(
9 sites), which are not resolved. The lack of resolution can be the presence of Q (Si(OH, 3Si)) defects. The Si MAS NMR
accounted for by the presence of Zr, which induces distortion spectrum of ZrO /SiO is very similar to those observed for
2
2
3
of the lattice. No Q -resonance indicative of Si(OH, 3Si) struc- ZrMCM-41, pointing to the same local structure of siliceous
tural defects are observed, which confirms the high quality of species in these samples.
29
ZrBEA material used for this study.
The analysis of the XPS and Si MAS NMR data suggests
that Zr is incorporated in the frameworks of zeolite and
MCM-41 materials. In the case of ZrO /SiO catalyst, the for-
2
2
mation of Si–O–Zr bonds is also evident. However due to the
impregnation procedure used, these species are most probably
located on the surface of silica and the formation of tiny zirco-
nia islands on the silica surface is more probable than the
incorporation of isolated zirconium atoms.
The Ag promoter supported over different catalysts has
been characterized using SEM/EDX and TEM techniques. The
images are shown in Fig. 6 and 7 for the Ag/BEA(100) sample.
The results point to rather uniform distribution of Ag particles
of 2–5 nm size over the catalyst. Such a particle size
was reported to be close to optimal for alcohol
2
0,31
dehydrogenation.
No significant difference was observed in Ag dispersion and
distribution over different supports. In contrast, the acidic pro-
perties showed significant variation.
3
.2. Acidic properties
The nature of the acid sites was investigated by FTIR spec-
troscopy of the adsorbed deuterated acetonitrile. CD CN was
3
selected for this study, since it is a well-established probe
molecule for the characterization of the acidity of zeolitic,
2
9
3
mesoporous and oxide catalysts. The adsorption of CD CN at
room temperature on acidic catalysts leads to the formation of
H-bonds with OH groups and to the coordination of CO to
Lewis acid sites. Due to this interaction, the ν(CuN) vibration
band of the adsorbed acetonitrile shifts to higher wave-
Fig. 3 XPS spectra of Zr-containing catalysts.
numbers with respect to the band of physisorbed CD
(
3
CN
2268 cm− ). This shift is characteristic of the nature (Lewis or
1
OH) and strength of the site. Application of CD CN as a probe
3
molecule allows for easy access to most of the sites due to its
very small size. This brings an additional advantage for the
investigation of the acidity of microporous solids.
In a typical experiment, an equilibrium pressure of 0.5 Torr
3
of acetonitrile-d was introduced in the IR cell and after
adsorption for 1 hour, the spectrum was collected. The results
obtained are shown in Fig. 5. The adsorption of CD
3
CN on all
samples leads to the appearance of the bands at 2303 and
1
2
275 cm− (Fig. 5). These bands can be attributed to the
ν(CuN) vibrations of CD CN adsorbed on the Lewis acid sites
Zr tetrahedral sites) and surface silanol groups, respectively.
3
3
0
(
−
1
For ZrBEA catalysts the band at 2203 cm due to physically
adsorbed CD CN is also observed. Evacuation of the aceto-
3
nitrile at ambient temperature leads to vanishing of the bands
2
9
Fig. 4
Si MAS NMR spectra of Zr-containing catalysts.
corresponding to adsorption of CD CN on silanol groups and
3
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
3
Table 2 Relative intensities of the bands observed after acetonitrile-d
adsorption
Area of the band
centered at 2303 cm−
(Lewis sites)
Area of the band
centered at 2275 cm
(OH groups)
1
−1
Sample
ZrBEA(100)
ZrBEA(200)
ZrBEA(400)
ZrMCM-41(200)
0.97
0.75
0.69
0.66
0.42
1.27
1.07
1.26
2.08
0.92
2 2
ZrO /SiO (200)
components centered at 2303, 2275 and 2268 cm− with fixed
full width at half maximum (FWHM) were used for spectra
fitting. The results are shown in Table 2. The area of the band
1
corresponding to the CD CN adsorption on Lewis sites
3
increases in the following order of the samples: ZrO /SiO <
2
2
ZrMCM-41(200) < ZrBEA(400) < ZrBEA(200) < ZrBEA(100),
whereas the intensity of the band attributed to H-bonded
acetonitrile increases as follows: ZrO /SiO < ZrBEA(100) ≈
2
2
ZrBEA(200) ≈ ZrBEA(400) ≪ ZrMCM-41(200). These results
show that the ZrBEA(100) catalyst possesses the highest Lewis
acidity and the ZrMCM-41(200) sample has a higher concen-
tration of surface Si–OH groups in comparison with other
catalysts.
3
Fig. 5 FTIR spectra of CD CN adsorbed over Zr-containing catalysts.
The lowest content of Zr Lewis sites is observed over ZrO2/
SiO
2
catalysts. This observation can be explained by the
different distribution and localization of Zr atoms in this cata-
lyst. Indeed, in molecular sieve catalysts most of Zr is located
in the form of Zr(OSi)4 isolated species in the T-positions of
the framework. In contrast, in the ZrO /SiO (200) catalyst ZrO
2
2
2
is deposited via impregnation, which usually leads to non-
uniform distribution of Zr on the surface of silica with the for-
mation of ZrO2 aggregates along with the Zr(OSi)4 species.
Therefore, molecular sieves catalysts should contain more iso-
lated Zr sites.
Fig. 6 TEM images of the Ag/ZrBEA(100) catalyst.
3.3. Catalyst evaluation in ethanol conversion into butadiene
3.3.1. Reaction products and the reaction network. The
reaction products observed over the catalysts studied can be
subdivided into 5 main groups as shown in Fig. 8:
(
1) butadiene and small amounts of intermediate products
leading to butadiene (Scheme 1), mainly, acetaldehyde
and crotonaldehyde;
(
(
2) diethyl ether and ethylene obtained via ethanol
dehydration;
3) ethyl acetate formed from acetaldehyde via the
Tischenko reaction and the products of its further trans-
formation – acetone and propylene;
(
4) butanol-1 and butenes formed from crotyl alcohol via
hydrogenation, dehydration and isomerization steps;
5) hexatrienes and other heavy products obtained by cross
condensation of acetaldehyde with crotonaldehyde and
other aldehydes.
Fig. 7 SEM/EDX silver mapping of the Ag/ZrBEA(100) catalyst.
(
physically adsorbed acetonitrile, which points on weak inter-
action with the surface and Si–OH.
For the quantification of the adsorbed CD
3
CN spectra,
The kinetics of these products’ formation over the metal
a deconvolution procedure was performed. Three Gauss–Lorentz promoted oxide catalysts has been studied in detail in our pre-
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
range of catalysts (Table 3) correlates with the relative amount
of Lewis acid sites determined by FTIR of adsorbed CD CN
3
(
Fig. 5 and Table 2).
The analysis of the product distribution and selectivity of
the catalysts was carried out at the same conversion level as
ethanol (Table 4). Similar ethanol conversions within 29–31%
were achieved by variation of WHSV in the range of 0.12–0.64 g
−
1
−1
g
h
. The results indicate that the Ag/ZrBEA(200), Ag/
ZrMCM-41(200) and Ag/ZrO /SiO (200) catalysts show very
similar selectivity in the range of 65–66%. This observation
2
2
4
+
suggests that the Zr sites located in molecular sieve frame-
works and incorporated in amorphous silica exhibit the
same selectivity in the overall reaction. However, some
differences are observed in the by-product distribution
Fig. 8 Main reaction pathways.
(
Table 4):
Ag/ZrMCM-41(200) shows the highest content of the pro-
•
1
3
vious paper. It should be mentioned that the Ag-promoted
Zr-containing molecular sieves show the same set of products
and obey the same mechanism of transformations.
ducts of ethanol dehydration – diethyl ether and ethene,
which is most probably due to the high amount of surface
silanol groups over this sample, as evidenced by the FTIR
3
.3.2 Effect of the type of catalyst. The results on ethanol
spectra of the adsorbed CD
CN (Fig. 5 and Table 2).
3
conversion over the catalysts studied are shown in Table 3. The
comparison of the catalysts with the same composition and
different structures indicate that the molecular sieve catalysts
Ag/ZrBEA(200) and Ag/ZrMCM-41(200) are much more active
• The Ag/ZrO /SiO (200) catalyst shows the highest contri-
2
2
bution of heavy by-products (14%) with respect to Ag/
ZrBEA(200) (7–10%) and Ag/ZrMCM-41(200) (6%). This
result can be related to the difference in the pore struc-
ture of the catalysts: zeolite BEA and MCM-41 have
smaller pores than the silica based catalyst (Table 1 and
Fig. 1) and the formation of large molecules of hexa-
trienes can be inhibited over these catalysts due to geo-
metric factors.
2 2
with respect to Ag/ZrO /SiO (200). Since the content of Ag and
the metal dispersion are similar over all the catalysts, the
reason should be due to the different properties of the sup-
ports, which are governed by the state and location of Zr
atoms.
2
1
Our previous studies on acetaldehyde condensation and
• Ag/ZrBEA(200) provide for the highest amount of butenes
at the expense of ethanol dehydration products and
hexatrienes.
2
2
the MPVO reaction of crotonaldehyde with ethanol pointed
4
+
to the key role of Zr Lewis acid sites in these reaction steps.
This study confirms the above conclusion for the overall reac-
To assess the stability of the catalytic activity of different
tion of ethanol conversion into butadiene. Indeed, the activity catalysts over the time course of the reaction, the yield of buta-
Table 3 Ethanol conversion and product distribution over Zr catalysts (T = 593 K, WHSV = 0.32 g g− h−1, TOS = 3 h)
1
Selectivity (mol%)
Ethanol
Sample
conversion (%)
Butadiene
Ethylene
Propylene
Butenes
Ethyl ether
Ethyl acetate
1-Butanol
6
C +
Ag/ZrBEA(100)
Ag/ZrBEA(200)
Ag/ZrBEA(400)
Ag/ZrMCM-41(200)
47.9
30.6
32.4
29.8
11.9
55.6
66.3
67.2
65.9
66.8
1.5
5.2
4.2
7.4
2.6
3.9
3.1
2.8
3.4
2.8
8.7
6.8
7.0
4.1
3.1
5.6
9.0
8.6
9.9
3.5
0.5
2.1
2.2
2.5
3.9
0.3
0.7
0.7
1.1
2.6
24.0
6.9
7.3
5.7
14.7
2 2
Ag/ZrO /SiO (200)
Table 4 Selectivity of Zr catalysts in the ethanol conversion (T = 593 K, TOS = 3 h)
Selectivity (mol%)
Ethanol
Ethyl
ether
Ethyl
acetate
WHSV
−1
Sample
conversion (%)
Butadiene
Ethylene
Propylene
Butenes
1-Butanol
C
6
+
(g g h−1
)
Ag/ZrBEA(100)
Ag/ZrBEA(200)
Ag/ZrBEA(400)
Ag/ZrMCM-41(200)
30.1
30.6
32.4
29.8
30.5
63.9
66.3
67.2
65.9
64.7
3.7
5.2
4.2
7.4
3.8
3.4
3.1
2.8
3.4
3.4
7.8
6.8
7.0
4.1
3.8
8.6
9.0
8.6
9.9
4.6
1.1
2.1
2.2
2.5
3.9
0.9
0.7
0.7
1.1
2.7
10.6
6.9
7.3
5.7
13.1
0.64
0.32
0.32
0.32
0.12
2 2
Ag/ZrO /SiO (200)
Green Chem.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Green Chemistry
Paper
Fig. 9 Yield of butadiene versus time on stream (T = 593 K, WHSV = 0.32 g g− h−1).
1
diene is plotted versus time on stream in Fig. 9. The results
show that the Ag/ZrBEA(200), Ag/ZrMCM-41(200) and Ag/ZrO2/ molecular sieves shows that the highest butadiene yield of
SiO (200) catalysts demonstrate similar deactivation patterns 27% is achieved so far over Ag/ZrBEA(100). The comparison of
Fig. 9a). It should be noted that the loss of yield is associated this result with the literature data
The comparison of different Ag-promoted Zr-containing
2
1
1–19
(
indicates that the
with the decrease of ethanol conversion, whereas the selecti- activity of molecular sieve catalysts is comparable with oxide
vity to butadiene and other by-products changes only slightly based systems. Further progress in this field requires widening
(Fig. S1†). Therefore the deactivation of the catalysts studied is of the list of molecular sieve structures and optimization of
most probably due to the gradual poisoning of active sites by the reaction conditions.
coke deposits as confirmed by TG analysis (Fig. S2†).
3
.3.3 Effect of ZrBEA composition. To elucidate the influ-
ence of Zr content on catalyst activity, selectivity and de-
activation, Ag/ZrBEA catalysts with a Si/Zr ratio in the range of
4. Conclusions
1
00–400 are compared in Tables 3, 4 and Fig. 9b. The increase Butadiene synthesis from ethanol over a series of Ag-promoted
of Zr content leads to a significant enhancement of the Zr-containing
catalysts including ZrBEA zeolites,
ethanol conversion from 30 to 48% (Table 3) due to the ZrMCM-41 mesoporous material and zirconia supported on
increase of the content of Lewis sites responsible for this reac- silica has been studied. All types of catalysts show the same set
tion (Table 2).
of products and obey the same mechanistic pathways. The
The analysis of the product distribution (Table 4) suggests main side reactions are found to involve: (1) ethanol de-
that all Ag/ZrBEA catalysts have similar selectivity to butadiene. hydration into diethyl ether and ethylene; (2) conversion of
The slight decrease of the selectivity to butadiene and the acetaldehyde into ethyl acetate and further into acetone and
increase of the contribution of heavier products over Ag/ZrBEA propylene; (3) cross-condensation of acetaldehyde with croto-
(100) can be accounted for by the higher density of Zr sites in naldehyde and other aldehydes into heavy by-products; and (4)
Ag/ZrBEA(100). Thereby, the intermediate products such as hydrogenation of crotyl alcohol into butanol-1 followed by its
acetaldehyde or/and crotonaldehyde formed at one site can dehydration and isomerization into butene-1 and butene-2.
undergo further condensation reactions into heavier C + pro-
Whereas all the catalysts demonstrate similar selectivity of
6
ducts at the neighboring sites. The latter can be transformed ca. 65% at the same conversion level of ca. 30%, the activity is
into hexatrienes by the sequence of MPVO and dehydration found to change in the following order of catalysts: Ag/ZrO2/
1
3
reactions.
SiO (200) ≪ Ag/ZrMCM-41(200) < Ag/ZrBEA(200). This activity
2
The deactivation curves for Ag/ZrBEA catalysts are com- range is due to the higher content of isolated Zr sites in mole-
pared in Fig. 9b. The Ag/ZrBEA(200) and Ag/ZrBEA(400) cular sieve catalysts with respect to silica supported catalysts.
samples show similar deactivation curves with a gradual The increase of Zr content in Ag/ZrBEA leads to further activity
decrease of the butadiene yield with time on stream. The Ag/ enhancement. This activity range correlates with the amount
ZrBEA(100) catalyst is more stable during the first 5 hours, but of Lewis acid sites in the catalysts determined by FTIR of
afterwards shows faster deactivation. Such deactivation behav- adsorbed CD CN.
3
ior is typical of the pore blocking mechanism. Further studies
The best catalyst performance in terms of butadiene yield
are necessary for the understanding of different deactivation (27%) is achieved over Ag/ZrBEA(100). On this catalyst, the
behaviors of the Ag/ZrBEA(100) and Ag/ZrBEA(200), Ag/ZrBEA selectivity to butadiene is 56% at an ethanol conversion of
(400) catalysts.
48%.
This journal is © The Royal Society of Chemistry 2015
Green Chem.

View Article Online
Paper
Green Chemistry
1
6 M. Jones, C. Keir, C. Di Iulio, R. Robertson, C. Williams
and D. Apperley, Catal. Sci. Technol., 2011, 1, 267–272.
Acknowledgements
The authors thank the Russian Science Foundation for the 17 M. Lewandowski, G. S. Babu, M. Vezzoli, M. D. Jones,
financial support (grant no. 14-23-00094). Vitaly Sushkevich
R. E. Owen, D. Mattia, P. Plucinski, E. Mikolajska,
A. Ochenduszko and D. C. Apperley, Catal. Commun., 2014,
49, 25–28.
gratefully acknowledges Haldor Tøpsoe A/S for
fellowship.
a PhD
1
1
8 H.-T. Chae, T.-W. Kim, Y.-K. Moon, H.-K. Kim, K.-E. Jeong,
C.-U. Kim and S.-Y. Jeong, Appl. Catal., B, 2014, 150–151,
596–604.
References
9 W. Janssens, E. Makshina, P. Vanelderen, F. De Clippel,
K. Houthoofd, S. Kerkhofs, J. A. Martens, P. A. Jacobs and
B. F. Sels, ChemSusChem, 2014, DOI: 10.1002/
cssc.201402894.
1
2
W. C. White, Chem.-Biol. Interact., 2007, 166, 10–14.
J. Grub and E. Loser, Butadiene in Ullmann’s Encyclopedia of
Industrial Chemistry, Wiley-VCH, Weinheim, 2012.
S. V. Lebedev, British Patent 331482, 1930.
3
4
5
6
20 V. Sushkevich, I. Ivanova and E. Taarning, ChemCatChem,
2013, 5, 2367–2373.
W. J. Toussaint and J. T. Dunn, U. S. Patent 2,357,855, 1944. 21 V. Ordomsky, V. Sushkevich and I. Ivanova, J. Mol. Catal. A:
S. V. Lebedev, J. Gen. Chem., 1933, 3, 698–708.
W. Toussaint, J. Dunn and D. Jackson, Ind. Eng. Chem.,
947, 39, 120–125.
Chem., 2010, 333, 85–93.
1
22 V. Sushkevich, I. Ivanova, S. Tolborg and E. Taarning,
J. Catal., 2014, 316, 121–129.
23 Y. Z. Zhu, G. K. Chuah and S. Jaenicke, J. Catal., 2004, 227,
1–10.
24 L. Li, C. Stroobants, K. Lin, P. A. Jacobs, B. F. Sels and
P. P. Pescarmona, Green Chem., 2011, 13, 1175–1181.
25 B. Zhang, M. Tang, J. Yuan and L. Wu, Chin. J. Catal., 2012,
33, 914–921.
26 S. Damyanova, L. Petrov, M. A. Centeeno and P. Grange,
Appl. Catal., A, 2002, 224, 271–284.
27 B. Rakshe, V. Ramaswamy, S. G. Hegde, R. Vetrivel and
A. V. Ratnasamy, Catal. Lett., 1997, 45, 41–50.
28 X. X. Wang, F. Lefebvre, J. Patarin and J.-M. Basset, Micro-
porous Mesoporous Mater., 2001, 42, 269–276.
29 J. Ryczkowski, Catal. Today, 2001, 68, 263–381.
7
8
9
B. B. Corson, H. E. Jones, C. E. Welling, J. A. Hinckley and
E. E. Stahly, Ind. Eng. Chem., 1950, 42, 359–373.
J. Ostromislenskiy, J. Russ. Phys.-Chem. Soc., 1915, 47,
1
472–1506.
A. D. Patel, K. Meesters, H. den Uil, E. de Jong, K. Bloka
and M. K. Patel, Energy Environ. Sci., 2012, 5, 8430–8444.
1
1
1
1
1
1
0 J. A. Posada, A. D. Patel, A. Roes, K. Blok, A. C. Faaij and
M. K. Patel, Bioresour. Technol., 2013, 135, 490–499.
1 C. Angelici, B. M. Weckhuysen and P. Bruijnincx,
ChemSusChem, 2013, 6, 1595–1614.
2 E. V. Makshina, M. Dusselier, W. Janssens, J. Degrève,
P. Jacobs and B. Sels, Chem. Soc. Rev., 2014, 43, 7917–7953.
3 V. Sushkevich, I. Ivanova, V. Ordomsky and E. Taarning,
ChemSusChem, 2014, 9, 2527–2536.
4 C. Angelici, M. E. Z. Velthoen, B. M. Weckhuysen and 30 M. Boronat, P. Concepcion, A. Corma, M. Renz and
P. C. A. Bruijnincx, ChemSusChem, 2014, 7, 2505–2515. S. Valencia, J. Catal., 2005, 234, 111–118.
5 E. V. Makshina, W. Janssens, B. F. Sels and P. A. Jacobs, 31 K. Shimizu, K. Sugino, K. Sawabe and A. Satsuma, Chem. –
Catal. Today, 2012, 198, 228–344. Eur. J., 2009, 15, 2341–2351.
Green Chem.
This journal is © The Royal Society of Chemistry 2015