Ethanol Conversion to Butadiene over Isolated Zinc and Yttrium Sites Grafted onto Dealuminated Beta Zeolite

Article abstract of DOI:10.1021/jacs.0c06906

Zinc and Yttrium single sites were introduced into the silanol nests of dealuminated BEA zeolite to produce Zn-DeAlBEA and Y-DeAlBEA. These materials were then investigated for the conversion of ethanol to 1,3-butadiene. Zn-DeAlBEA was found to be highly active for ethanol dehydrogenation to acetaldehyde and exhibited low activity for 1,3-butadiene generation. By contrast, Y-DeAlBEA was highly active for 1,3-butadiene formation but exhibited no activity for ethanol dehydrogenation. The formation of 1,3-butadine over Y-DeAlBEA and Zn-DeAlBEA does not occur via aldol condensation of acetaldehyde but, rather, by concerted reaction of coadsorbed acetaldehyde and ethanol. The active centers for this process are Si-O-Y(OH)-O-Si or Si-O-Zn-O-Si-O groups closely associated with adjacent silanol groups. The active sites in Y-DeAlBEA are 70 times more active than the Y sites supported on silica, for which the Y site is similar to that in Y-SiO2 but which lacks adjacent hydroxyl groups, and are 7 times more active than the active sites in Zn-DeAlBEA. We propose that C-C bond coupling in Y-DeAlBEA proceeds via the reaction of coadsorbed acetaldehyde and ethanol to form crotyl alcohol and water. The dehydration of crotyl alcohol to 1,3-butadiene is facile and occurs over the mildly Br?nsted acidic Si-OH groups present in the silanol nest of DeAlBEA. The catalysts reported here are notably more active than those previously reported for both the direct conversion of ethanol to 1,3-butadiene or the formation of this product by the reaction of ethanol and acetaldehyde.

Full text of DOI:10.1021/jacs.0c06906

Subscriber access provided by The University of Melbourne Libraries
Article
Ethanol Conversion to Butadiene over Isolated Zinc and
Yttrium Sites Grafted onto Dealuminated Beta Zeolite
Liang Qi, Yanfei Zhang, Matthew A. Conrad, Christopher Russell, Jeffrey T. Miller, and Alexis T. Bell
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c06906 • Publication Date (Web): 29 Jul 2020
Downloaded from pubs.acs.org on July 31, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Ethanol Conversion to Butadiene over Isolated Zinc and Yttrium
Sites Grafted onto Dealuminated Beta Zeolite
1
,2
2
3
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Liang Qi , Yanfei Zhang , Matthew A. Conrad , Christopher K. Russell ,
3
1,2
Jeffrey Miller , and Alexis T. Bell*
1Chemical Sciences Division
Lawrence Berkeley National Laboratory
Berkeley, CA 94720
2Department of Chemical and Biomolecular Engineering
University of California
Berkeley, CA 94720
3Davidson School of Chemical Engineering
Purdue University
West Lafayette, IN 47907
Submitted to
Journal of the American Chemical Society
June 21, 2020
To whom correspondence should be addressed: alexbell@berkeley.edu
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
Zinc and Yttrium single sites were introduced into the silanol nests of dealuminated BEA
zeolite to produce Zn-DeAlBEA and Y-DeAlBEA. These materials were then investigated for
the conversion of ethanol to 1,3-butadiene. Zn-DeAlBEA was found to be highly active for
ethanol dehydrogenation to acetaldehyde and exhibited low activity for 1,3-butadiene
generation. By contrast, Y-DeAlBEA was highly active for 1,3-butadiene formation but
exhibited no activity for ethanol dehydrogenation. The formation of 1,3-butadine over Y-
DeAlBEA and Zn-DeAlBEA does not occur via aldol condensation of acetaldehyde but, rather,
by concerted reaction of co-adsorbed acetaldehyde and ethanol. The active centers for this
process are ≡Si-O-Y(OH)-O-Si≡ or ≡Si-O-Zn-O-Si-O≡ groups closely associated with adjacent
silanol groups. The active sites in Y-DeAlBEA are 70 times more active than the Y sites
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
supported on silica, for which the Y site is similar to that in Y-SiO but which lacks adjacent
2
hydroxyl groups, and are 7 times more active than the active sites in Zn-DeAlBEA. We propose
that C-C bond coupling in Y-DeAlBEA proceeds via the reaction of co-adsorbed acetaldehyde
and ethanol to form crotyl alcohol and water. The dehydration of crotyl alcohol to 1,3-butadiene
is facile and occurs over the mildly Brønsted acidic ≡Si-OH groups present in the silanol nest
of DeAlBEA. The catalysts reported here are notably more active than those previously
reported for both the direct conversion of ethanol to 1,3-butadiene or the formation of this
product by the reaction of ethanol and acetaldehyde.
2
ACS Paragon Plus Environment

Page 3 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Introduction
Public concern with the environment and growing recognition of the need to reduce carbon
emissions have motivated the search for processes that can produce basic chemicals from
renewable sources of carbon.1-3 An attractive option is the use of biomass-derived ethanol
(EtOH) because it can be readily converted to basic chemicals such as ethylene, acetaldehyde
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
AcH), isobutene, 1,3-butadiene (1,3-BD) and n-butanol.4-7 Of these products, 1,3-BD is
particularly interesting because it is a monomer for the production of styrene-butadiene rubber
SBR) and polybutadiene.8-9 Moreover, techno-economic analysis and life-cycle assessments
(
have shown that bioethanol-based processes for producing 1,3-BD are attractive.6, 9-11
Two processes have been proposed and utilized for the EtOH to 1,3-BD reaction. In the
one-step process, first developed by Lebedev,12-13 EtOH is converted directly over a single
catalyst having active centers for both EtOH dehydrogenation and C-C bond coupling. In the
two-step process, developed by Ostromisslensky, 1,3-BD is generated through conversion of
EtOH and AcH, using a catalyst that is highly active for C-C bond coupling.6-7, 9, 11 Considerable
effort has been devoted to improving the catalyst performance for both the one-step and two-
step processes.1-2, 6-9 Mixed binary oxide compositions, e.g. SiO with MgO, ZrO , HfO , Y O
2
2
2
2
3
or Ta O , have been found to be effective for C-C bond formation, and these materials doped
2
5
with d-metal (Zn, Cu, Cr, Au, Ag) are effective for EtOH to 1,3-BD.2, 6, 8-11
The investigation of single-site catalysts for the conversion of EtOH to 1,3-BD has been
motivated by the discovery that open Sn sites (e.g., ≡Sn-OH) in the framework of SnBEA,14
promote the isomerization of glucose to fructose, as well as many other processes involving
biomass-derived compounds (e.g., carbohydrates, 3-dihydroxyacetone, and acetone).15-19 These
findings have stimulated the evaluation of Y, Zr, Ti, and Ta incorporated into the framework
of zeolites for the EtOH-to-1,3-BD process.11, 20-24 Of particular note, Ivanova et al. have found
that hydrothermally synthesized Zr-BEA is more active than ZrO /SiO or ZrO /MCM-41
x
2
x
2
2
because of a higher concentration of framework Zr sites. In studies, these authors found that
Ag/Zr-BEA catalysts prepared by a post-synthetic method exhibited higher aldol condensation
2
1
activity compared to a hydrothermally synthesized one. In a related approach, Dzwigaj et al.
have introduced tantalum into dealuminated BEA and dispersed nanoparticles of Ag, Cu, and
Zn on the resulting material. CuTaSi-BEA proved to be the most efficient catalyst for the one-
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
step Lebedev process.11 Zhang and coworkers have introduced Hf and Zn onto Silicalite-1
nanosheets and added lithium to modify the acidity of the catalyst. The LiZnHf-MFI zeolite
nanosheets exhibits 73% 1,3-BD selectivity and good stability for the EtOH-to-1,3-BD
24
process. Zr, Hf, Ta, and Y have all been introduced into dealuminated BEA together with Zn.
An exceptionally high butadiene productivity of 2.33 g1,3-BD gcat−1 h−1 and a selectivity of ∼63%
were achieved at 673 K with 2wt%Zn-8wt%Y/BEA20, 25-26
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Prior research has indicated that both the identity of the Lewis acid center (Zr, Hf, Sn, Ta,
Y) associated with the siliceous support and the environment immediately surrounding this
center affect catalyst activity.14-16, 20, 27-28 The significance of the active site environment was
first proposed by Corma et al. for SnBEA. The authors noted the existence of both “closed” and
“
open” Sn sites corresponding to a fully coordinated Sn site (Sn-(OSi) ) and a partially
4
14
hydrolyzed Sn site ((HO)-Sn-(OSi) ) respectively. Generally, the open site was more active
3
for a number of Lewis acid catalyzed reactions, such as the Baeyer−Villiger oxidation of cyclic
ketones and the isomerization of glucose to fructose.14, 17, 29 Ivanova et al. also proposed the
coexistence of closed and open Zr sites in Zr-BEA,21, 23 and found that the Zr open sites were
more active for AcH aldol condensation. Li and co-workers found that ZnO and YO clusters
x
x
supported in the silanol nest of dealuminated BEA were much more effective than similar
2
0
clusters supported on SiO , and this difference was ascribed to differences in the structural
2
confinement by each support.
The mechanism of C-C bond formation involved in the conversion of EtOH to 1,3-BD has
been related to the composition of the active site and its environment. Several authors have
proposed that C-C bond formation during the EtOH-to-1,3-BD process involves aldol
condensation of two AcH molecules.2, 6, 8-9, 30 On the other hand, Mella et al. have proposed that
over MgO, a carbanion is formed by abstraction of a proton from the methyl group of EtOH
and that this critical intermediate then reacts with AcH to form crotyl alcohol which, in turn,
3
1
dehydrates to 1,3-BD. Roman-Leshkov found the aldol condensations of aromatic aldehydes
with acetone over hydrothermally synthesized Hf-, Sn-, and Zr-BEA zeolites was promoted by
an enolization pathway.32 Ivanova et al. have proposed that the identity of the metal center
2
7
defines the critical step in the C-C bond coupling mechanism. Over Sn-BEA, they suggested
4
ACS Paragon Plus Environment

Page 5 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
that C-C coupling occurs after the initial enolization step; whereas, over Zr-BEA and Ti-BEA,
they concluded that enolization is less likely due to the very high transition-state barrier, and
that instead C-C coupling occurs via a concerted single-step mechanism involving co-
adsorption of two AcH molecules. The preceding discussion suggests that the formation of C-
C bonds during the condensation of EtOH to form 1,3-BD may involve different elementary
steps and depend on the identity of the isolated Lewis center and its immediate environment.33
In this study, we introduced isolated Zn and Y sites into the silanol nest of dealuminated
BEA (DeAlBEA). The structure of these sites was characterized by FTIR, XANES, and EXAFS.
Zn-DeAlBEA and Y-DeAlBEA were then investigated for the reaction of EtOH to 1,3-BD. Zn-
DeAlBEA was found to be highly active for EtOH dehydrogenation but exhibited little activity
for the formation of C-C bonds and 1,3-BD. By contrast, Y-DeAlBEA was highly active for C-
C bond formation via the reaction of AcH with EtOH but was relatively inactive for the
dehydrogenation of EtOH to AcH. ≡Si-O-Y(OH)-O-Si≡ with adjacent silanol groups was
identified as the active center for crotyl alcohol formation from EtOH and AcH. We also show
that by introducing the maximum number of metal atoms in the form of isolated sites, we can
produce exceptionally active catalysts for both the one-step and two-step EtOH-to-1,3-BD
processes.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Experimental methods
Catalyst Synthesis.
Zn-DeAlBEA and Y-DeAlBEA zeolites were prepared in a two-step procedure, consisting
of dealumination of H-Beta and subsequent introduction of Zn and Y species via aqueous
impregnation. In a typical process, commercial H-Beta with a Si/Al ratio of 15.3 (determined
by ICP) (Tianjin Shenneng Co.) was suspended in 13 M nitric acid at 373 K for 20 h to obtain
DeAlBEA.20, 34-35 The resulting powder was filtered, washed with MQ deionized water until to
a pH of 6-7, and then dried at 373 K under vacuum overnight. Afterward, 0.5 g of the DeAlBEA
was impregnated with metal nitrate aqueous solutions. For Zn-DeAlBEA and Y-DeAlBEA
synthesis, the molar concentrations of Zn(NO ) and Y(NO ) solutions were about 0.0218,
3
2
3 3
0
.0436, 0.109, 0.218 and 0.436 mol/L for M/Al ratios (mole of Zn or Y introduced per mole of
Al in H-Beta prior to dealumination) of 0.075, 0.15, 0.375, 0.75 and 1.5, respectively. Generally,
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
.4 mL solution was needed for 0.5 g DeAlBEA. The resulting slurry was dried at room
temperature for 5 h and then further dried at 323 K under vacuum for 2 h. The powdered mixture
was then ground for 20 min. Finally, the catalyst was calcined at 823 K for 6 h in static air. The
zeolite samples after calcination were used directly as the catalysts without reduction. The final
product is labeled as xM-DeAlBEA, where x indicates the M/Al ratios (0.15Zn-DeAlBEA
means the Zn/Al ratio is 0.15). As discussed below, all Al in H-BEA is removed by
dealumination to produce silanol nests containing four ≡Si-OH groups. Therefore, M/Al
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
represents the average number of Zn or Y atoms per silanol nest. Y-SiO was prepared using a
2
modification method of the procedure reported before.36
Characterization Techniques.
Powder X-ray diffraction (PXRD) patterns were acquired with a Bruker D8 GADDS
diffractometer equipped with a Cu−Kα source (40 kV, 40 mA). BET surface areas were
calculated from nitrogen adsorption isotherms obtained using a Micromeritics Gemini VII 2390
surface area analyzer. Fourier transform infrared (FTIR) spectra were acquired using a Thermo
Scientific Nicolet 6700 spectrometer equipped with a liquid nitrogen-cooled MCT detector.
Samples for IR spectroscopy were prepared by pressing 0.036 g of catalyst into a 20 mm-
diameter pellet and placing it into a custom-built transmission cell fitted with CaF windows.
2
All pellets for FTIR were pretreated in dry air at 773 K for 1 h; spectra were acquired at 393 K.
For pyridine adsorption FTIR, 2 L of pyridine was injected for each experiment and then
purged with He for 25 min to remove physically adsorbed pyridine. The Zn concentration of
Zn-DeAlBEA was determined by ICP-OES using an Optima 7000 DV instrument. The Y
loading was analyzed by Galbraith Laboratories, Inc. Solid-state 27Al NMR spectra were
acquired using a Bruker Avance I spectrometer equipped with a 11.4 T magnet operating at
2
7
1
30.3 MHz for Al. Samples for NMR spectroscopy were packed into a zirconia rotor with an
outer diameter of 4 mm and the rotor was sealed with a Kel-F drive cap. Nitrogen gas was used
for magic angle spinning (MAS) and the data were acquired at rotor speed of 10 KHz. The flip
2
7
angle of Al excitation pulse was set to 15 degree with a radio frequency strength of 53.2 kHz,
27
and the Al NMR chemical shift was referenced to 1 M Al(NO ) aqueous solution at 0 ppm.
3
3
A recycle delay of 0.5 s was used and 8192 scans were collected for both spectra.
In situ X-ray absorption spectroscopy (XAS) data were acquired at beamline 10-BM-B of
6
ACS Paragon Plus Environment

Page 7 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
the Advanced Photon Source (APS), located at Argonne National Laboratory (ANL). The
sample was packed into a self-supporting wafer in a six-shooter and loaded into a quartz tube
sealed by Ultra-Torr fittings with Kapton windows to allow photon transmission axially through
the cell. The sample was treated in flowing He for 20 min at 623 K and then cooled and scanned
at room temperature in He. For Zn-DeAlBEA, data were acquired at the Zn K edge (9659 eV)
in transmission mode, with an energy resolution of 0.3 eV and edge energy precision greater
than 0.1 eV. The sample was scanned in fast scan mode, and a third ion chamber measured a
Zn foil for energy calibration. For Y-DeAlBEA, data were acquired at the Y K edge (17038.4
eV) with a 0.3 eV energy resolution and edge energy precision greater than 0.1 eV. All Zn and
Y data were analyzed using WinXAS 3.1 software using standard data reduction procedures to
obtain the X-ray absorption near edge structure (XANES) and extended X-ray absorption fine
structure (EXAFS). The coordination number and bond distances between Zn orY and its
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
nearest neighbors were determined using a least-squared fit performed on isolated k -weighted
-1
Fourier transform data (k = 2.7 – 10.0 Å ).
Measurements of Catalytic Activity.
Reaction rate data were obtained using a quartz, packed-bed reactor (4 mm inner diameter).
Quartz wool was placed below the catalyst bed to hold it in place. The reactor temperature was
maintained using a tube furnace equipped with a Watlow temperature controller and a K-type
thermocouple. Prior to reaction, the catalyst was heated in air (Praxair, zero grade) flowing at
3
0 mL min−1 to 823 K for 40 min before it was cooled to the reaction temperature.
In a typical experiment, EtOH (Sigma-Aldrich, 99.5%) or EtOH with a certain amount of
AcH (Sigma-Aldrich, 99.5%) was introduced into a He (Praxair, 99.999%) stream using a
syringe pump (KD Scientific, Legato 100 series). All experiments were carried out at a pressure
of 1 atm. Product streams were analyzed by gas chromatography using an Agilent 6890A GC
fitted with a Plot-Q capillary column (30 m × 0.32 mm × 0.25 μm) and a flame ionization
detector.
The evolution of reactant conversion is not presented at very low conversion level due to
the small fluctuations (∼5%) of the syringe pump. To investigate the influence of the reactant
partial pressure on both dehydrogenation and C-C coupling reactions, we mainly focus on the
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
formation rate of acetaldehyde and 1,3-butadiene. The error for the rate measurement is around
5
% due to fluctuation of the He flow rate.
Results and discussion
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catalyst Characterization
As shown in Figure S1, X-ray diffraction patterns of parent H-BEA, dealuminated H-BEA
(DeAlBEA), Zn-DeAlBEA, and Y-DeAlBEA show that the zeolite structure is retained after
dealumination and introduction of Zn or Y. Only characteristic peaks of BEA were detected for
all Zn-DeAlBEA and Y-DeAlBEA samples, indicating the absence of bulk ZnO or Y O . BET
2
3
experiments showed that all of the catalysts exhibited a microporous surface area around 350-
2
−1
3
70 m g (Table S1).
(a)
(b)
Figure 1. FTIR spectra of HBEA and DeAlBEA (a) and following pyridine adsorption (b).
The IR spectrum of the H-BEA presented in Figure 1a has characteristic bands at 3740 cm-1
-1
37-39
(
isolated silanol) and 3610 cm (Brønsted acid O−H groups).
After dealumination, the
bands for isolated silanol groups increase significantly and shift to a lower wavenumber, 3735
-1
-1
cm , and a broad, intense band for silanol nests appears at 3500 cm . These changes indicate
that upon dealumination, internal silanols and silanol nests are created within the crystal
34
framework. Comparison of the IR spectra after pyridine adsorption on H-BEA and DeAlBEA,
presented in Figure 1b, shows that almost all of the bands for Brønsted acid (1545 and 1637
-1
-1
cm ), Lewis acid (1620 and 1455 cm ), and the combination of Brønsted and Lewis acid site
(1490 cm ) are removed by dealumination. This conclusion is supported by 27Al NMR
-1
spectroscopy (Figure S2), which shows that virtually all Al is removed by dealumination.
8
ACS Paragon Plus Environment

Page 9 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
-
1
Interestingly, two new IR bands appear at 1592 and 1445 cm in the spectrum of pyridine
adsorbed on DeAlBEA. These two bands are ascribed to the interaction of pyridine with H
atoms of the Si-OH groups in the silanol nests.40 To obtain further evidence for this
interpretation, we acquired IR spectra of pyridine adsorbed silica, which has isolated silanol
groups but almost no silanol nests (Figure S3a). Consistent with our proposed interpretation of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
-1
the band at 1592 cm and 1445 cm in Figure 1b, bands attributed to hydrogen-bonded pyridine
were barely visible for pyridine interacting with silica (Figure S3b).
FTIR spectra of 0.375Zn-DeAlBEA and 0.375Y-DeAlBEA indicate that after the
-1
introduction of Zn and Y, the band for silanol nests at 3500 cm decreased in intensity (Figure
-1
2
1
a). Correspondingly, the intensity of the band for hydrogen-bonded pyridine at 1592 cm and
-1
-1
445 cm also decreased (Figure 2b). Instead, two new signals appeared at 1610 cm and 1450
-1
cm for 0.375Zn-DeAlBEA, which are attributable to pyridine interacting with Zn Lewis acid
sites. The position of these bands are similar to those observed for pyridine adsorbed on ZnO.41
-
1
Pyridine adsorption on 0.375Y-DeAlBEA also produced a new signal at 1604 cm . A second
-1
band at around 1445 cm was not observed in this case because of the overlap with the original
-1
-1
hydrogen-bonded pyridine signal (also at 1445 cm ). Pyridine adsorption bands at 1605 cm
-1
42
and 1444 cm were also observed over Y O , indicating the presence of Y Lewis acid sites.
2
3
Therefore, the IR spectra of adsorbed pyridine provide clear evidence for Zn and Y Lewis acid
sites introduced into the silanol nests of dealuminated BEA, most probably through the
formation of M-O-Si bonds.22-23
(
b)
(a)
Figure 2. FTIR spectra of DeAlBEA, 0.375Zn-DeAlBEA and 0.375Y-DeAlBEA (a) and of
pyridine adsorbed on these materials (b).
FTIR spectra of adsorbed pyridine were also obtained for Zn-DeAlBEA and Y-DeAlBEA
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with M/Al ratios from 0.075 to 1.5. Because of the overlap of Lewis acid signals and hydrogen
-1
bonded pyridine at 1445 cm for Y-DeAlBEA and DeAlBEA catalysts, we focused only on the
-1
-1
bands at 1610 cm and 1604 cm for Zn and Y sites, respectively. Figure 3a shows that, as the
-1
Zn/Al ratio increases from 0.075 to 0.375, the intensity of the band at 1610 cm increases
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
continuously, which also occurs for Zn/Al ratios of 0.375 to 1.5, albeit at a lower rate.
-1
Comparing 1610 cm peak area and Zn/Al ratio shows a linear relationship between 0.075 and
0
.375, but for higher Zn/Al ratios, the slope of the line decreases significantly (Figure 3c). The
-1
linear increase in the intensity of the band at 1610 cm below Zn/Al = 0.375 suggests that for
such Zn/Al ratios, Zn is present as isolated Zn Lewis acid centers. However, above Zn/Al =
0
.375, ZnO nanoclusters may be formed within the zeolite channels. A similar pattern is
observed for Y-DeAlBEA. As seen in Figure 3d, the integrated area is almost the same for
M/Al = 0.075, 0.15 and 0.225 for Zn- and Y-DeAlBEA. The integrated area for 0.375Y-
DeAlBEA is a little lower than expected when compared with that of 0.375Zn-DeAlBEA. That
is possibly due to the slower diffusion of the solvated yttrium species in the porous system of
the zeolite compared with the solvated Zn species. Therefore, the IR spectra of adsorbed
pyridine suggest that below M/Al ≈ 0.375, Zn and Y Lewis acid sites are introduced
n+
successfully into DeAlBEA through reaction of the aqueous metal cations (M (OH) (H O) )
x
2
6-x
with the Si-OH groups in silanol nests.
1
0
ACS Paragon Plus Environment

Page 11 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(
a)
(b)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(c)
(d)
Figure 3. Pyridine adsorbed FTIR spectra of Zn-DeAlBEA (a) and Y-DeAlBEA (b) samples
with different M/Al ratios, and the plots of integrated area of the Lewis acid bands at 1610
-1
−1
cm (c) and 1604 cm (d) as a function of the M/Al ratio.
(a)
(b)
Figure 4. X-ray absorption spectroscopic analysis of 0.15Zn-DeAlBEA (blue) and 0.3Zn-
DeAlBEA (red) after dehydration at 623 K in He and scanned at room temperature in He.
XANES (a); Fourier transforms of EXAFS data (b).
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. EXAFS fits for Zn-DeAlBEA and ZnO
XANES Coordination
Bond
distance (Å)
1.98
Δσ2
(x10-3)
Sample
E0 (eV)
(
keV)
number
4
ZnO (ref.)
9.6618
0.0
4.5
4.5
2.5
2.5
2.5
0
.15Zn-DeAlBEA 9.6626
3.7 ± 0.2
4 ± 0.2
1.97 ± 0.01
1.97 ± 0.01
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0.3Zn-DeAlBEA 9.6626
The oxidation state and coordination geometry of both the Zn and Y sites was further
investigated using in situ X-ray absorption spectroscopy. The XANES spectra and the
calculated Zn K-edge energies for Zn-DeAlBEA are presented in Figure 4a and Table 1,
respectively. The Zn K-edge energies for Zn-DeAlBEA are similar to that of ZnO (Figure 4a),
2
+
consistent with Zn . These sites are considered to be isolated based on the absence of Zn-O-
Zn backscattering for both 0.15Zn- and 0.3Zn-DeAlBEA (Figure 4b). The results of fitting of
the EXAFS data for the first shell Zn-O coordination are given in Table 1. For the Zn-DeAlBEA,
there are four Zn-O bonds at 1.97 Å, a bond distance very close to that for bulk ZnO. The XAS
results presented in Figure 4 and Table 1 are very similar to those for Zn grafted onto SiO . In
2
2+
this case, Zn is reported to be tetrahedrally coordinated but with slightly shorter Zn-O bonds
4
3
of 1.94 Å. The small difference between this Zn-O bond distance and that reported here may
be due to the difference in the environment of the Zn site. Considering the XAS information
and Zn site surrounding environment, two Zn-O-Si covalent bonds were proposed to be created
and the Zn center is also datively coordinated with two adjacent silanol groups.
Upon treatment in H at 823 K (Figure S4 and Table S3), the XANES energy decreases
2
slightly (0.4 eV), which supports the observation of Zn-H formation.43 Likewise, the 0.3Zn-
DeAlBEA sample decreases in coordination number from 4.0 to 3.4, which is consistent with
the loss of a Zn-O bond and formation of a Zn-H bond.
1
2
ACS Paragon Plus Environment

Page 13 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(
a)
(b)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. (a) The XANES region for Y O (black), 0.15Y-DeAlBEA (blue) and 0.3Y-
2
3
DeAlBEA (red) after dehydration at 623 K in He and scanned at room temperature in He. (b)
The real (solid) and imaginary (dashed) parts of the Fourier transform of the EXAFS for Y O
2
3
(black) and the 0.15Y-DeAlBEA sample (blue). The 0.3Y-DeAlBEA sample is identical,
within error, to 0.15Y-DeAlBEA, and is omitted.
Table 2. EXAFS fits for Y-DeAlBEA and Y O
2
3
XANES Coordination
Bond
distance (Å)
2.27
Δσ2
Sample
E0 (eV)
(x10-3)
(
keV)
number
6
Y O (ref.)
17.0504
17.0499
17.05
0
8
7
0.2
3
2
3
0
.15Y-DeAlBEA
3.7 ± 0.2
3.9 ± 0.2
2.27 ± 0.01
2.27 ± 0.01
0.3Y-DeAlBEA
3.5
XANES spectra for Y O , 0.15Y-DeAlBEA, 0.3Y-DeAlBEA are presented in Figure 5a.
2
3
The white line of Y O spectrum has a double peak with maxima at 17050.4 and 17062 eV,
2
3
4
4
suggesting the presence of Y in two different sites. By contrast, the XANES spectra of the
two Y-DeAlBEA samples contain only a single edge, indicating the presence of only one Y
site. Based on the edge energy, the inflection point of the absorption peak, both 0.15Y-
DeAlBEA and 0.3Y-DeAlBEA samples are consistent with Y(III). Fitting the Fourier transform
of the EXAFS data yields the coordination number and bond distance given in Table 2. Based
on the very small signals for the second shell Y-O-Y backscattering at around 3.3 Å, Y appears
3+
to be present as a single-site Y . The coordination number is close to four and the average Y-
O bond distance is 2.27 Å. This Y-O bond length is consistent with other Y containing
compounds; for example, the Y-OSi and the Y-O(THF) bonds in Y(OSiPh ) (THF) lie in
3
3
3
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
5
ranges of 2.12-2.14 Å and 2.37-2.46 Å, respectively, while the average Y-O bond distance in
the aqueous Y(H O) complex is around 2.37 Å.46-47 An average Y-O bond distance of 2.27
3+
2
8
Å for Y-DeAlBEA can be ascribed to a yttrium coordination environment comprising covalent
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Y-OSi bonds with shorter distance and dative Y-O(H)Si bonds with longer distance.36 The
structure of the observed Y single site is somewhat different from that reported for Y single
sites grafted onto SiO2,36 which were found to have a Y-O bond distance of 2.31 Å but a
coordination number of 4.7. In that case, Y was proposed to be present as Si-O-Y(OH)-O-Si
with the Y also datively coordinated with one Si-OH and one H O. Considering the similar Y-
2
O bond distance observed in our work, the structure of Y for Y-DeAlBEA is also proposed to
be Si-O-Y(OH)-O-Si, but the Y is datively coordinated with only one adjacent Si-OH based.
The reason why Y does not datively coordinate with two adjacent Si-OH to achieve a
coordination number of 5 may be due to steric hindrance from the local active site environment
(silanol nest inside channels of the zeolite). Together with the FTIR characterization result, the
proposed structures for Zn and Y single site in the silanol nest of BEA are depicted in Scheme
1. For all four samples tested by in situ X-ray absorption spectroscopy, the data at lower loading
has more noise (Figure S5) but the XANES and EXAFS were consistent with that of the higher
loaded samples, suggesting that the majority of species are the same.
Scheme 1. Generation of Zn and Y single sites in silanol nests.
Catalytic activity of Zn-DeAlBEA
Previous studies have suggested that Zn cations promote EtOH dehydrogenation, whereas
Y, Zr, and Ta cations promote C-C bond formation in mixed metal Zn-M/BEA (M stands for
Y, Zr or Ta ).11, 20-21, 24-25, 28, 48 However, investigations of these reactions over Zn and Y cations
1
4
ACS Paragon Plus Environment

Page 15 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
grafted into silanol nests is rarely reported. To obtain a comprehensive understanding of the
role of single Zn and Y sites in the EtOH-to-1,3-BD process, we investigated the chemistry
occurring over Zn-DeAlBEA and Y-DeAlBEA separately. The aim of this effort was to
establish the roles of isolated Zn and Y cations on the dehydrogenation of EtOH and the
formation of C-C bonds, an essential step in the formation of 1,3-BD. However, prior to doing
so, we first investigated the activity of silanol groups in the silanol nests of DeAlBEA, since
such groups are known to be more acidic than other forms of silanol groups, the acid strength
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
49
of silanol groups increasing in the order terminal < geminal < vicinal < nest. Evidence for the
acidity of nested silanol groups also comes from the IR spectra of adsorbed pyridine on
DeAlBEA shown in Figure 1b and Figure S3.
The reaction of EtOH over DeAlBEA at 593 K produced ethylene with 97% selectivity
14
(
Figure S6), consistent with earlier reported results. However, at lower temperatures, 573 K
and 553 K, the EtOH conversion decreased, and the major product changed from ethylene to
diethyl ether (DEE), as would be expected from the thermodynamics of the two competing
reactions.50-51
Figure 6a shows the products formed during the reaction of EtOH over 0.075Zn-DeAlBEA
as a function of TOS. During the initial phase of the reaction, ethylene and DEE are the major
products observed. As the TOS increases, the formation of ethylene decreases monotonically,
whereas the formation of DEE increases, passes through a maximum, and then decreases with
TOS. The formation of AcH steadily increases and finally reaches a plateau after about 300
min. The TOS for attainment of a plateau in AcH formation decreases as the Zn/Al ratio
increases (Figure 6b and Figure 6c). We attribute the EtOH dehydration and etherification
activities of Zn-DeAlBEA to silanol nests or isolated silanol groups, whereas the formation of
AcH via EtOH dehydrogenation occurs on isolated Zn sites.20, 48 We propose that EtOH
dehydration and etherification activity decreases as silanol sites are gradually poisoned by AcH.
Evidence supporting this hypothesis is presented below. We note that 0.075Zn-DeAlBEA
requires a longer TOS to attain a plateau in AcH formation. This is because this catalyst has the
lowest rate of acetaldehyde formation and the highest concentration of residual silanol nests
compared to 0.15Zn-DeAlBEA and 0.3Zn-DeABEA. Small amounts of 1,3-BD are also
generated over Zn-DeAlBEA, indicating that the Zn active site can do both dehydrogenation
1
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and C-C coupling reaction. However, as seen in Figure 6d the rates of 1,3-BD formation are
nearly an order of magnitude lower than the rates of AcH formation. We note further that the
rates of AcH and 1.3-BD formation are the same over 50 mg of 0.15Zn-DeAlBEA and 25 mg
of 0.3Zn-DeAlBEA, consistent with what would be expected for Zn being present as isolated
sites. To confirm further that isolated sites of Zn are responsible for the formation of AcH,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
experiments were conducted at 573 K, in order to eliminate the formation of C products. As
4
shown in Figure S7, the formation of AcH is linearly proportional to the loading of Zn in Zn-
DeAlBEA.
(a)
(b)
(c)
(d)
Figure 6. Evolution of EtOH conversion and product distribution with TOS over 0.075Zn- (a),
.15Zn- (b), 0.3Zn-DeAlBEA (c), and the rates of AcH and 1,3-BD formation over 0.15Zn-
DeAlBEA and 0.3Zn-DeAlBEA (d). Catalyst mass = 50 mg for 0.075Zn-DeAlBEA and
0
0
5
.15Zn-DeAlBEA, 25mg for 0.3Zn-DeAlBEA, T = 593 K, He flow rate = 25 mL/min, PEtOH
.3 kPa.
=
1
6
ACS Paragon Plus Environment

Page 17 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
The effect of EtOH partial pressure on the rate of AcH formation is illustrated in Figure 7a
for three different temperatures. The EtOH conversion was kept to around 5% for high EtOH
partial pressure (5.3 kPa) and below 20% for very low partial pressure (0.2 kPa). All the data
were collected in the stable AcH formation stage. The apparent order in EtOH partial pressure
decreases from about 1.0 towards zero order, as the EtOH partial pressure increases from 0.2
to 5.3 kPa at all temperatures examined.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A possible mechanism for dehydrogenation of EtOH on isolated Zn sites in Zn-DeAlBEA
is presented in Scheme 2. The process starts with dissociative adsorption of EtOH on a Zn−O
atom pair to form a surface ethoxide and hydroxyl species (Steps 1 and 2). This step is followed
by abstraction of an α-H atom in what is thought to be the rate-limiting step for dehydrogenation
(Step 3). The α-H atom then reacts with the H atom of the hydroxyl species to form molecular
hydrogen and adsorbed AcH (Step IV). The reaction cycle is completed with the desorption of
H and AcH (Steps V and VI).
2
Scheme 2. Reaction mechanism for EtOH dehydrogenation to AcH over Zn-DeAlBEA.
A rate expression for the dehydrogenation of EtOH over Zn-DeAlBEA can be derived from
Scheme 2 and is shown as Eq. 1 (see Supporting Information for a derivation). In Eq. 1, k is
1
the rate coefficient associated with the rate-limiting step, while K is the equilibrium constant
E
1
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
for EtOH adsorption onto Si-O-Zn-O-Si active sites. As shown in Figure 7a, Eq. 1 gives a very
good description of the rate of AcH formation for the three temperatures examined. The
temperature dependence of all rate and adsorption equilibrium constants were obtained by
fitting Eq. 1 to the rate data obtained and presented in Figure 7b. The activation energy for
EtOH dehydrogenation determined from an Arrhenius plot is 140.5 ± 1.2 kJ/mol, which is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
similar to the value reported for HAP (Hydroxyapatite) (142 kJ/mol), but higher than that
5
2
reported for Cu/C/SiO (67.6 kJ/mol) . The enthalpies of adsorption for EtOH over Zn single
2
site is -71±9.1 kJ/mol, lower than that over HAP, -104.6 kJ/mol, very close to the values
measured over ZnO for methanol and propanol, ca. 80 kJ/mol.53
푘1퐾퐸푃퐸푡푂퐻
푟퐴푐퐻 = 1 + 퐾퐸푃퐸푡푂퐻
(1)
(a)
(b)
Figure 7. Rates of AcH formation as a function of EtOH partial pressure at different
temperatures over 0.075Zn-DeAlBEA. The solid line for AcH formation represents the fitting
of Eq. 1 to the data (a). Arrhenius plots for ethanol dehydrogenation and adsorption (b). Catalyst
mass = 5 mg, total flow rate at STP = 42.5 mL/min.
Catalytic activity of Y-DeAlBEA
Figure 8a illustrate the products formed upon reaction of EtOH over 0.075Y-DeAlBEA at
5
93 K. The principal product produced is ethylene, with a selectivity of about 95%, and almost
no AcH or 1,3-BD. We note that in contrast to Zn-DeAlBEA, the formation of ethylene does
not decrease with TOS. This finding is similar to that reported for the reaction of EtOH over
20
DeAlBEA, YO /DeAlBEA and ZrO /DeAlBEA. When 1.4 kPa AcH is fed together with
x
x
EtOH (Figure 8b), the rate of EtOH dehydration decreases significantly; the generation of DEE
is enhanced initially but then rapidly decreases as well. This effect of AcH suggests that it
1
8
ACS Paragon Plus Environment

Page 19 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
adsorbs competitively on the sites that promote EtOH dehydration, and thereby inhibit this
process. This interpretation also explains the observed decrease in the dehydration activity of
Zn-DeAlBEA seen in Figure 6 once AcH is formed as a product.
Cofeeding AcH and EtOH (1.4 kPa and 5.8 kPa) results in the formation of 1,3-BD with a
selectivity of 75% based on the total conversion of EtOH (20%) and AcH (60%) shown in
Figure 8b. This means that, although the addition of AcH decreases the conversion of EtOH,
the efficiency of C-C bond formation increased significantly. We note that 0.075Y-DeAlBEA
contains a high concentration of silanol nests; therefore to determine whether silanol nests are
active for C-C coupling, the reaction of EtOH and AcH was also carried out over DeAlBEA
under identical reaction conditions. Figure S8 shows that in this case, the 1,3-BD formation is
negligible compared with that for 0.075Y-DeAlBEA. Therefore, we conclude that the
formation of 1,3-BD from EtOH and AcH is catalyzed by isolated Y sites in Y-DeAlBEA.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(a)
(b)
Figure 8. EtOH (a) conversion and EtOH and AcH (b) conversion and major product
distribution with TOS over 0.075Y-DeAlBEA. Catalyst mass = 50 mg, T = 593 K, total flow
rate at STP = 24 mL/min, PEtOH = 5.8 kPa, PAcH = 1.4 kPa.
We next investigated the effects of Y loading on the rate of 1,3-BD formation. For these
studies the conversion of AcH and EtOH were kept below 25% and 5%, respectively, at 573 K.
Figure 9 shows that the rate of 1,3-BD formation correlates linearly with the Y/Al ratio from
0
1
.0375 to 0.15, and extrapolates to zero in the absence of any Y. Consequently, if the rate of C4
,3-BD formation is normalized by the Y molar amount, the TOF is almost constant.
1
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 9. 1,3-BD formation rate over Y-DeAlBEA as a function of Y/Al ratios. Catalyst mass
=
5 mg, T = 573 K, PEtOH = 2.7 kPa, PAcH = 0.3 kPa, total flow rate at STP = 41 mL/min. The
rate of 1,3-BD formation is normalized by catalyst mass and moles of Y in the catalyst,
respectively.
C-C bond formation in the conversion of EtOH to 1,3-BD and 1-butanol is generally
thought to proceed via aldol condensation involving the reaction of two adsorbed AcH
molecules via aldol condensation form acetaldol.2-3, 6, 20-21, 27-28 This product readily dehydrates
to form crotonaldehyde, the presumed precursor to both 1,3-BD and 1-butanol.
We, therefore, investigated the aldol condensation of AcH over Y-DeAlBEA. At 573 K,
AcH polymerized for the initial 5 min and after 25 min, the polymerization ceased and only
very a small amount of crotonaldehyde was detected. The polymerization of AcH is probably
caused by the weak BrØnsted acidity of silanol nest, which are easily poisoned by the
polymerized AcH.54-55 The polymerization of AcH could also be observed over DeAlBEA but
this process could be almost totally inhibited by cofeeding water or EtOH.
We then turn to the generation of 1,3-BD by cofeeding EtOH and AcH. The influence of
EtOH partial pressure on the rate of 1,3-BD formation was investigated at 573 K over 0.075Y-
DeAlBEA, using a constant AcH partial pressure of 0.3 kPa. As seen in Figure 10a, the rate of
1
,3-BD formation increases with increasing EtOH partial pressure before reaching a plateau.
The influence of AcH partial pressure on the rate of 1,3-BD formation was also investigated
for a constant EtOH partial pressure of 2.7 kPa. As seen in Figure 10b, the rate of 1,3-BD
formation increases with increasing the AcH partial pressure up to a partial pressure of about
0.2 kPa, after which it decreases. The results presented in Figure 10 clearly demonstrate the
2
0
ACS Paragon Plus Environment

Page 21 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
involvement of both EtOH and AcH in the formation of 1,3-BD.
The influence of EtOH partial pressure on the rate of 1,3-BD formation was also
investigated on 0.3Zn-DeAlBEA; however, the influence of AcH partial pressure was not
performed over this catalyst because Zn itself can catalyze EtOH dehydrogenation which means
that the AcH partial pressure in the system could not be maintained constant. As seen in Figure
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0a, the rate of C 1,3-BD formation over Zn-DEAlBEA is about 7 times lower than that over
4
Y. Nevertheless, the influence of EtOH partial pressure on the rate of 1,3-BD formation is quite
similar, suggesting that the mechanisms of C-C bond formation is similar over Y and Zn sites.
(a)
(b)
Figure 10. 1,3-BD TOF over 0.075Y-DeAlBEA and 0.3Zn-DeAlBEA as a function of EtOH
partial pressure with PAcH = 0.3 kPa (a), and 1,3-BD TOF as a function of AcH partial pressure
over 0.075Y-DeAlBEA with PEtOH = 2.7 kPa (b). Catalyst mass = 5 mg for 0.075Y-DeAlBEA
and 10 mg for 0.3Zn-DeAlBEA, T = 573 K, total flow rate at STP = 41 mL/min. Solid curves
are drawn as a guide for the eye.
Table 3. 1,3-BD formation rate and TOF over 1wt%Y-SiO and 0.15Y-DeAlBEA
2
1,3-BD TOF × 10-4
Catalyst
-7
1
,3-BD formation rate × 10 mol/min
mol/mol Y/s
1
wt%Y-SiO2 a
.15Y-DeAlBEA b
1.09
4
0
10.66
287
a
Catalyst mass = 40 mg
Catalyst mass = 5 mg
b
T = 573 K, TOS = 5 min, total flow rate at STP = 41 mL/min, PEtOH = 2.7 kPa, PAcH = 0.3
kPa.
Work by Getsoian et al. has shown that single Y sites can be grafted onto surface of silica
2
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with the assistance of citric acid.36 EXAFS characterization of such sites indicates that Y is
present as Si-O-Y(OH)-O-Si, a structure similar to that found in the present work for Y grafted
into the silanol nests of DeAlBEA. To assess whether Y sites on Y-SiO are active for C-C
2
bond formation via the reaction of EtOH and AcH, we synthesized 1wt%-Y-SiO following the
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
recipe reported by Getsoian et al. and compared its performance with that of 0.15Y-DeAlBEA.
The rates of 1,3-BD formation for both catalysts are presented in Table 3. Although 40 mg of
1
wt% Y-SiO was used for these experiments, the rate of 1,3-BD formation was far less than
2
that measured using 5 mg 0.15Y-DeAlBEA (the weight percent for 0.15Y-DeAlBEA is 1.1%).
If normalized by the total Y molar amount, the turnover frequency for 1,3-BD formation over
Y-DeAlBEA is 70 times higher than that over Y-SiO . We attribute the large difference in the
2
activity of Y site for the two catalysts to differences in the local environment of the site. While
the structure of the Y site is similar for both catalysts, for Y-DeAlBEA the Y site is adjacent to
two silanol groups, whereas that in Y-SiO interacts with only a single silanol group.
2
Evidence for the differences in Y-DeAlBEA and Y-SiO were evident in the IR specra of
2
adsorbed pyridine (Figure S9). Despite that much higher loading of Y on Y-SiO (5.2 wt%)
2
than on 0.375Y-DeAlBEA (2.75 wt%), the peak area of the band characterizing pyridine
associated with Lewis acid signals was much larger for 0.375Y-DeAlBEA. More importantly,
-1
the IR band intensities for pyridine associated with Brønsted acid groups at 1637 cm and 1545
-1
cm could be observed over 0.375Y-DeAlBEA, but were absent for Y-SiO and DeAlBEA.
2
We note as well that bands for pyridine interacting with Brønsted acid sites on 0.375Zn-
DeAlBEA (Figure 2b) could barely be discerned. Therefore, the mild Brønsted acid of Y-
DeAlBEA is ascribed to the interaction of the Y-OH group with the adjacent ≡Si-OH in the
nest.
We propose that the formation of 1,3-BD over isolated Zn and Y sites occurs via a
concerted reaction involving co-adsorbed AcH and EtOH, as shown in Scheme 3. This scheme
is based on the structure of Y and Zn supported on DeAlBEA and on kinetics of 1.3-BD
formation observed in the present study. As shown in Figure 10, AcH adsorption is much
stronger than that of EtOH. Since the carbonyl oxygen atom of AcH is a strong Lewis base, it
3+
27
interacts with the Y cation, which is a strong Lewis acid, and the α-H of AcH interact with
basic oxygen atom of Y-OH. We envision that the adsorption of EtOH (Step 2) also occurs
2
2
ACS Paragon Plus Environment

Page 23 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
through the interactions of the O atom of EtOH with the Y Lewis acid center. The co-adsorption
of AcH is thought to inhibit EtOH dehydration (Figures 6 and 8), whereas the co-adsorption of
EtOH suppresses the polymerization of AcH, as noted earlier. We propose that the Y-OH group
in the hydroxyl nest is able to protonate the carbonyl oxygen atom of AcH, and then the α-H of
AcH regenerates the Y-OH, resulting in the formation of an enolate intermediate. Because of
these steps, the α-C of AcH will be negatively charged and the α-C of EtOH will be positively
charged. The next step (Step 3/4), proposed to be rate limiting, is formation of a C-C bond via
reaction of the α-C of EtOH with the α-C of AcH. The formation of the C-C bond at the end of
Step IV facilitates the cleavage of the C-H bond on the -C of EtOH and its transfer to the -
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
C of what was the enolate to form a CH group. The resulting product is 1,3-butanediol, which
2
31
rapidly dehydrates to crotyl alcohol (Step VI). Dehydration of crotyl alcohol to form 1,3-BD
(Step VII) is very facile, and, in fact, passage of crotyl alcohol over DeAlBEA readily produces
1
,3-BD. In summary, the total catalytic cycle can be regarded as a combination of C-C coupling
and hydrogen transfer catalyzed by a combination of Brønsted acidic and Lewis acidic sites.
2
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 3. Reaction mechanism for EtOH and AcH to 1,3-BD over Y-DeAlBEA
While a Zn single site located in a silanol nest can catalyze C-C coupling in a manner
similar to Y single sites, the Zn sites are much less active for the following reasons. First, the
polarizability of the grafted metal atom and the Brønsted basicity of the oxygen atom bound to
this metal atom are reported to influence the C-C coupling catalytic activity.17 Since the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
6
3+
polarizability of a metal atom is proportional to its volume, the polarizability of Y should
2+
3+
be stronger than that of Zn . Moreover, the Lewis acidity of Y is also stronger than that of
2+ 57
Zn . Second, although there are two silanol groups adjacent to Zn and Y in both Zn-
DeAlBEA and Y-DeAlBEA, respectively, Zn does not have an associated OH group (i.e.,
ZnOH) that can interact with the adjacent silanols in Zn-DeAlBEA. We believe that absence of
a basic M-OH group in Zn-DeAlBEA makes it harder to accomplish the first hydrogen transfer
(Step 3 in Scheme 3) than for Y-DeAlBEA. The Lewis basicity of the O atom in Y-OH is also
likely enhanced by the interaction of this hydroxyl group through hydrogen bonding with the
adjacent Si-OH groups in the silanol nest. This interpretation is supported by the observation
of much weaker IR bands for pyridine interacting with Brønsted acid sites on 0.375Zn-
DeAlBEA compared to those on 0.375Y-DeAlBEA (Figure 2b). We hypothesize that these two
factors significantly influence the activation of AcH and consequently C-C bond formation.
According to the mechanism presented in Scheme 3, the ratio of EtOH and AcH consumed
to form 1,3-BD should be one. Using the data from Figure 8b and taking into account the
consumption of EtOH for dehydration and etherification, we find that the observed ratio, in
fact, is very close to one, as shown in Figure S10. We note that the generation of butenes was
taken into consideration in these calculations, since the generation of this product occurs via
the dehydration of butanol, which is formed through Meerwein–Ponndorf–Verley (MPV)
reaction between crotyl alcohol and EtOH. When one molecule of butene is generated, one
more molecule of EtOH is consumed and one more AcH molecule is generated.
퐶 퐻 푂퐻 + 퐶퐻 퐶퐻푂→퐶퐻 = 퐶퐻퐶퐻 = 퐶퐻 + 2퐻 푂
(2)
2
5
3
2
2
2
We have also considered two other possible mechanisms to explain our results. The first
2
4
ACS Paragon Plus Environment

Page 25 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
is the aldol condensation mechanism, proposed previously for the formation of 1,3-BD by
3
0
Toussaint and co-workers. In this case, two AcH molecules condense to form crotonaldehyde,
which then undergoes MPV hydrogenation by EtOH to form crotyl alcohol, and the latter
product dehydrates to form 1,3-BD. According to this mechanism, the rate of condensation
should increase with AcH partial pressure and then reach a plateau.5, 58 If this mechanism were
operative, the rate of 1,3-BD formation should be second order in AcH initially and then reach
zero order at high AcH partial pressures. This pattern differs from that seen in Figure 10, which
exhibits first order dependences on both AcH and EtOH at low partial pressures of each reactant.
This suggests that the role of EtOH on 1,3-BD formation is not due hydrogen transfer to
crotonaldehyde but, rather, to involvement in the C-C bond formation process itself.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The Prins mechanism would be another alternative to the one proposed in Scheme 3. In
this case, EtOH first dehydrates to ethylene and this product then reacts with AcH to form a
5
9
C-C bond. No evidence to support this mechanism was found. In fact, we observed that the
addition of AcH suppresses the rate of EtOH dehydration to ethylene for Y-DeAlBEA (see
Figure 7b). We also observed that the addition of ethylene to AcH and water fed over 0.15Y-
DeAlBEA at 593K did not produce 1,3-BD.
An expression for the rate of C-C formation via the reaction EtOH and AcH can be derived
from the mechanism presented in Scheme 3 (see Supporting Information for details). The
resulting expression is given by Eq. 3. Here, PEtOH and PAcH are the partial pressures of EtOH
and AcH, respectively; k is the rate constants for the C-C coupling of EtOH and AcH to crotyl
Y
alcohol and water; K and K are the equilibrium constants for the adsorption of EtOH and
E
A
AcH onto Y sites (Y single site with adjacent silanols), respectively; K is the equilibrium
Y
constant for co-adsorption of EtOH and AcH; and KAAY is the equilibrium constant for the co-
adsorption of two AcH molecules. Eq. 3 shows that at low reactant partial pressures, the rate of
C-C coupling will increase with increasing partial pressures of EtOH and AcH. The inhibition
of the rate of C-C bond formation observed in Figure 10b for AcH partial pressures greater than
0
.2 kPa indicates that the adsorption of AcH is stronger than that of EtOH; therefore too high
AcH partial pressure inhibits EtOH adsorption and decreases the rate of C-C coupling. Eq. 3
also indicates that for sufficiently high EtOH partial pressures, the rate of C-C coupling
approaches zero order in EtOH, consistent with what is shown in Figure 10a.
2
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
푘푌퐾푌푃퐸푡푂퐻푃퐴푐퐻
푟푐 ― 푐 = 1
(3)
2
+ 퐾퐸푃퐸푡푂퐻 +퐾퐴푃퐴푐퐻 +퐾퐴퐴푌퐾퐴푌[푃퐴푐퐻] + 퐾푌푃퐸푡푂퐻푃퐴푐퐻
Comparison of catalyst activities
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The 1,3-BD activity of the catalysts developed in this study can be compared with those
reported in the literature. Up till now, the best 1,3-BD productivity reported for one-step EtOH-
2
8
to-1,3-BD catalyst is that for ZnTa-TUD-1 (TUD-1 is a sponge-like mesoporous silica). At
−
1
6
2
73 K with EtOH partial pressure of 4.56 kPa, a WHSV of 8 h , the 1,3-BD productivity was
.45 g/g /h with a selectivity of 66%. The best 1,3-BD productivity reported for a two-step
cat
process is that for Zr/MCF (MCF is a sponge-like mesoporous silica material with ultra-large
6
0
−1
nanopores). At 673 K with EtOH/AcH ratios (1.17−1.64), WHSV of 3.7 h , this catalyst has
a 1,3-BD productivity of 1.4 g/g /h and a 1,3-BD selectivity of 70%. For comparison, we
cat
consider the performance of 0.15Zn-0.225Y-DeAlBEA and 0.375Y-DeAlBEA for the one-step
and two-step EtOH-to-1,3-BD process, respectively, evaluated at 673 K (Table 4).
Table. 4 Catalyst performances for EtOH and EtOH/AcH conversion to 1,3-BD
0
.15Zn-0.225Y-
0.375Y-
DeAlBEA b
673
ZnTa-
TUD-1c
673
66.0
--
Zr/MCFd
Catalyst
DeAlBEA a
673
52.0
9.6
Temperature (K)
673
70.0
--
1
,3-BD Sel. (%)
66.4
Butenes Sel. (%)
6.2
C olefins Sel. (%)
61.6
3.6
72.6
--
--
4
1
,3-BD Productivity (g/g /h)
6.6
2.5
1.4
6~8
--
cat
Ethylene Sel. (%)
DEE Sel. (%)
11.2
2
9.4
--
3
--
AcH Conv. (%)
AcH Sel. (%)
--
51.3
--
--
7.1
51.3
~30
~80
7.0
93-95
EtOH Conv. (%)
94.0
53.7
a
Catalyst mass = 20 mg. He flow rate = 40 mL/min, EtOH flow rate = 0.32 mL/h, TOS = 5
-1
min, PEtOH = 5.3 kPa, WHSV = 12.24 h .
2
6
ACS Paragon Plus Environment

Page 27 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
b
Catalyst mass = 10 mg. He = 40 mL/min, TOS = 5 min, P = 2.5 kPa and PEtOH = 4.1 kPa,
AcH
-1
WHSV = 31.5 h .
c
-1
Catalyst mass = 30 mg, PEtOH = 4.56 kPa, WHSV = 8 h , TOS = 3 h.
d
-1
Catalyst mass = 60 mg, EtOH/AcH = 1.17−1.64, WHSV = 3.7 h , TOS = 1 h.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The reaction of 5.3 kPa EtOH fed at a WHSV of 12.24 g/g /h over 0.15Zn-0.225Y-
cat
DeAlBEA occurs with an EtOH conversion of 94% at 673 K after 5 min TOS. The selectivity
to all C olefins is 61.6% and the corresponding 1,3-BD productivity is 3.6 g/g /h, whereas the
4
cat
selectivity to 1,3-BD selectivity is 52%. The reaction of 4.1 kPa EtOH and 2.5 kPa AcH over
0
.375Y-DeAlBEA at 673 K produced 1,3-BD at the rate of 6.6 g/g /h with a selectivity of
cat
66.4%. To the best of our knowledge, the 1,3-BD productivity observed here over 0.15Zn-
.225Y-DeAlBEA and 0.375Y-DeAlBEA for the one-step and two-step EtOH-to-1,3-BD
0
process are the highest values reported. We note however, that despite the high 1,3-BD
productivity of 0.375Y-DeAlBEA, the catalyst rapidly deactivates; the 1,3-BD productivity
decreases from 6.6 g/g /h to 4.6 g/g /h during 40 min TOS (Figure S11a). 1,3-BD productivity
cat
cat
via the one-step process over 0.15Zn-0.225Y-DeABEA is more stable under similar conditions
Figure S11b), which is attributed to the existence of hydrogen from EtOH dehydrogenation,
(
which inhibits the generation of coke precusors.61
The rapid deactivation of 0.375Y-DeAlBEA is likely caused by the generation of coke
which gradually cover the active sites during the reaction.26 Both AcH and 1,3-BD could be
active coke precursors,26, 62 and the acidic hydroxyl groups involved in C-C coupling (see
Scheme 3) likely also contribute to the deactivation of these sites. Coking might be mitigated
by lowering the reaction temperature or cofeeding water or hydrogen with
the reactants, and by increasing the EtOH/AcH ratio. Another alternative would be to carry out
the reaction in a fluidized bed reactor-regenerator configuration similar to that used for the
MTO reaction over SAPO-34 catalyst, which also rapidly deactivates due to coke deposition.63
Conclusion
Isolated Zn and Y sites supported on dealuminated BEA were investigated for the
conversion of EtOH to butadiene. This reaction involves a sequence of three steps –
2
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 34
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dehydrogenation of EtOH to AcH followed by reaction of EtOH with AcH to form crotyl
alcohol, which then rapidly dehydrates to form butadiene. Zn-DeAlBEA exhibits high activity
for EtOH dehydrogenation and low activity for C-C bonds formation via the coupling of AcH
and EtOH. By contrast, Y-DeAlBEA exhibits no activity for EtOH dehydrogenation and high
activity for C-C bonds formation. The dehydration of crotyl alcohol is facile and does not
require either Zn or Y sites, and is catalyzed by the mildly Brønsted acidic Si-OH groups present
in silanol nests of DeAlBEA. Investigations of the mechanism and kinetics of C-C bond
formation over Zn-DeAlBEA and Y-DeAlBEA indicate that this process does not proceed via
aldol condensation of AcH but, instead, involves the concerted reaction of adsorbed AcH and
EtOH. It is also notable that the rate of C-C bond formation over single Y sites in Y-DeAlBEA
is 70 time higher than that over isolated silica-supported Y sites. Based on these findings, we
propose that C-C bond formation over both Zn and Y site present in DeAlBEA requires the
participation of hydroxyl groups adjacent to the metal site. Thus, ≡Si-O-Y(OH)-O-Si≡ and ≡Si-
O-Zn-O-Si≡ groups together with the adjacent hydroxyl group constitute the active center for
crotyl alcohol formation from EtOH and AcH. Very good EtOH-to-1,3-BD performance is
obtained over both the 0.15Zn-0.225Y-DeAlBEA and 0.375Y-DeAlBEA catalysts for one-step
and two-step processes, respectively. When the reaction is performed over 0.15Zn-0.225Y-
DeAlBEA at 673 K, the 1,3-BD productivity is 3.6 g/g /h with a C olefin selectivity of 61.6%
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
cat
4
(1,3-BD selectivity of 52% and butenes selectivity of 9.6 %). Over 0.375Y-DeAlBEA catalyst
at 673 K, it is also encouraging to find that with AcH partial pressure of 2.5 kPa and EtOH
partial pressure of 4.1 kPa, the 1,3-BD productivity could reach 6.6 g/g /h with a C olefins
4
cat
selectivity of 72.6% (1,3-BD selectivity of 66.4% and a butene selectivity of 6.2%).
Supporting Information
Supplemental catalyst characterization (BET, XRD, 27Al NMR, ICP, FTIR-Py, EXAFS,
XANES) (Figures S1-S5, S9, Table S1-S3); supplemental reaction results (Figures S6-S8, S10-
S11), and derivation of rate expressions (PDF).
Acknowledgments
This work was supported by the Office of Science, Office of Basic Energy Sciences of the U.S.
Department of Energy under Contract No. DE-AC02-05CH11231. L. Qi also acknowledges
2
8
ACS Paragon Plus Environment

Page 29 of 34
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
support from the Dalian Institute of Chemical Physics, Chinese Academy of Sciences, People’s
Republic of China. We thank Christopher R. Ho for helpful discussion. This research used
resources of the Advanced Photon Source at Argonne National Laboratory. MAC, CKR and
JTM (Purdue University) were supported by the National Science Foundation under
Cooperative Agreement no. EEC1647722. The Advanced Photon Source was supported by the
U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract
No. DE-AC02-06CH11357. MRCAT operations and the beamline 10-BM were supported by
the Department of Energy and the MRCAT member institutions. We also thank Dr. Wei-Chih
Liao (UC Berkeley) for assistance with NMR measurements.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
9
ACS Paragon Plus Environment