Direct Catalytic Conversion of Ethanol to C5+ Ketones: Role of Pd–Zn Alloy on Catalytic Activity and Stability

Article abstract of DOI:10.1002/anie.202005256

Ethanol can be used as a platform molecule for synthesizing valuable chemicals and fuel precursors. Direct synthesis of C₅₊ ketones, building blocks for lubricants and hydrocarbon fuels, from ethanol was achieved over a stable Pd-promoted ZnO-ZrO₂ catalyst. The sequence of reaction steps involved in the C₅₊ ketone formation from ethanol was determined. The key reaction steps were found to be the in situ generation of the acetone intermediate and the cross-aldol condensation between the reaction intermediates acetaldehyde and acetone. The formation of a Pd–Zn alloy in situ was identified to be the critical factor in maintaining high yield to the C₅₊ ketones and the stability of the catalyst. A yield of >70 percent to C₅₊ ketones was achieved over a 0.1 percent Pd-ZnO-ZrO₂ mixed oxide catalyst, and the catalyst was demonstrated to be stable beyond 2000 hours on stream without any catalyst deactivation.

Full text of DOI:10.1002/anie.202005256

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Direct Catalytic Conversion of Ethanol to C5+ Ketones: Role of
PdZn Alloy on Catalytic Activity and Stability
Authors: Senthil Subramaniam, Mond Guo, Tanmayi Bathena,
Michel Gray, Xiao Zhang, Abraham Martinez, Libor Kovarik,
Konstantinos A. Goulas, and Karthikeyan Ramasamy
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202005256
Link to VoR: https://doi.org/10.1002/anie.202005256

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
Direct Catalytic Conversion of Ethanol to C5+ Ketones: Role of Pd-
Zn Alloy on Catalytic Activity and Stability
Senthil Subramaniam,[a,b] Mond F. Guo,[a,b] Tanmayi Bathena,[c] Michel Gray,[a] Xiao Zhang,[a,b] Abraham
Martinez,[a] Libor Kovarik,[a] Konstantinos A. Goulas,[c] and Karthikeyan K. Ramasamy*[a]
Abstract: Ethanol can be used as
a
platform molecule for
and diesel), lubricants and fuel blendstocks. C5+ ketones
themselves also have direct applications for specialty cleaning
solvents.[1]
synthesizing valuable chemicals and fuel precursors. Direct synthesis
of C5+ ketones, building blocks for lubricants and hydrocarbon fuels,
from ethanol was achieved over a stable Pd-promoted ZnO-ZrO2
catalyst. The sequence of reaction steps involved in the C5+ ketone
formation from ethanol was determined. The key reaction steps were
found to be the in-situ generation of the acetone intermediate and the
cross-aldol condensation between the reaction intermediates
acetaldehyde and acetone. The formation of a Pd-Zn alloy in-situ was
identified to be the critical factor in maintaining high yield to the C5+
ketones and the stability of the catalyst. A yield of >70% to C5+ ketones
was achieved over a 0.1% Pd-ZnO-ZrO2 mixed oxide catalyst, and the
catalyst was demonstrated to be stable beyond 2000 hours on stream
without any catalyst deactivation.
Catalytic approaches to generate C5+ ketones have focused
primarily on processes starting with acetone, n-butanol, and
ethanol (ABE) mixtures generated from fermentation. Anbarasan
et al. developed an efficient Pd/C-K3PO4 catalyst system that
utilizes the nucleophilic carbonyl carbon atoms of acetone and the
electrophilic α-carbon of the aldehyde intermediate, generated
from the in-situ dehydrogenation of the feed alcohol, to selectively
initiate the cross-aldol condensation.[1c] Several groups have
followed this pathway; most recently, Wang et al. have reported
ceria (Ce) -based mixed-oxide catalysts to selectively convert an
ABE mixture to C7 ketones (4-heptanone).[2] However, little work
has been done on establishing an equivalent process starting
from pure ethanol, which requires a modified catalytic system as
it lacks an initial ketone group to dictate the reaction. Despite this,
the overlap with the ABE chemistry makes ethanol a natural
candidate to support the renewable production of C5+ ketones,
particularly given its flexibility as a feedstock and ease of
synthesis. In addition to accessing an established commercial
infrastructure, with a US production volume near 15 billion
gallons,[3] using ethanol would help diversify the potential carbon
sources to include resources such as municipal solid waste, flue
gas, and shale gas.[4]
Here we report a novel approach to convert ethanol in a
one-pot reaction to produce long-chain ketones (C5-C11), based
on the in-situ generation of both the aldehydes and acetone
required for the suitable cross-aldol condensation to form longer
carbon chains. This is made challenging by the nature of ethanol
carbon-carbon (C-C) coupling chemistries, as they typically
involve complex cascade-type reactions with multiple
intermediate steps across multi-functional sites (e.g., acidic, basic,
and redox).[5] The reactive nature and low thermal stability of the
intermediate compounds means that parallel chemistries such as
dehydrogenation, dehydration, esterification, Tishchenko
reactions may occur, producing a range of products such as
aldehydes, esters, alkenes, and ketones.[6] As the promoted
mixed oxides catalysts typically used for these chemistries are
similar in nature,[7] selective C-C coupling relies on the delicate
balancing of catalyst characteristics and operating conditions to
avoid activation of unwanted side reactions. As products, C5+
ketones are typically only observed as minor fractions in ethanol
coupling reactions. However, ketones are regularly generated
during the formation of i-butene over ZnO-ZrO2 catalyst, which
proceeds through an acetone intermediate formed from
acetaldehyde coupling. The high reaction temperature leads to
Introduction
Growing awareness of the need for energy independence and
sustainability has driven increased interest in developing energy-
efficient technologies to produce renewable fuels and chemicals.
As sustainable sources of small oxygenate compounds become
more widely available, developing facile coupling strategies that
can extend the carbon chain length would enable renewable
catalytic pathways towards replacing a wider class of larger
petroleum products. One promising approach is to first produce
higher molecular weight ketones (C5+) as intermediate precursor
molecules. C5+ ketones possess both electrophilic and
nucleophilic functionalities, as well as sufficient carbon chain
length, that make them suitable building blocks, analogous to
alkenes and aromatics in a petroleum refining complex, for
producing infrastructure compatible hydrocarbon fuels (e.g. jet
[a]
Dr. S. Subramaniam, Dr. M. F. Guo, M. Gray, Dr. X. Zhang, Dr. L.
Kovarik, A. Martinez, Dr. K. K. Ramasamy
Chemical and Biological Processing Group
Pacific Northwest National Laboratory
Richland, WA 99354, USA
E-mail: karthi@pnnl.gov
[b]
[b]
Dr. S. Subramaniam, Dr. M. F. Guo, Dr. X. Zhang
The Gene and Linda Voiland School of Chemical Engineering and
Bioengineering
Washington State University
Pullman, WA 99164, USA
T. Bathena, Dr. K. A. Goulas
Chemical, Biological, and Environmental Engineering
Oregon State University
Corvallis, OR 97331 USA
Supporting information for this article is given via a link at the end of
the document.
rapid self-condensation of acetone into diacetone alcohol that
8]
produces i-butene by decomposition,[7a,
but appropriate
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
modification of the catalyst and conditions to instead favor
acetone coupling with acetaldehyde can offer a pathway towards
forming longer ketones. Literature on ABE condensation
reactions has shown that promotion of Pd is particularly suitable
to selectively enable the hydrogenation of α, β-unsaturated
ketones to cease the undesired reactions and improve the C5+
ketones products.[1c, 9]
Based on these previous works, a multifunctional Pd
promoted ZnO-ZrO2 catalyst was developed to upgrade the
ethanol to C5+ ketones in a one-pot catalytic process. (Figure 1)
Evaluation of the operating conditions and the compositional
changes in the catalyst with respect to the reaction mechanism
showed this formulation to be highly active toward this chemistry,
with over 70% selectivity to ketones at near complete conversion.
Characterization by X-ray Absorption (XAS) and High-Angle
Annular Dark Field - Transmission Electron Microscopy (HAADF-
TEM) revealed the in-situ formation of a dispersed Pd-Zn alloy
formation during reaction, which significantly modified the Pd
functionality and was correlated with high ketone yields. Further
long-term evaluation assessing the commercial viability of this
catalytic system demonstrated stable catalyst performance for
over 2000 hours on stream without deactivation.
formation of carbon monoxide (CO) and methane from ethanol
decomposition, presumably via sequential dehydrogenation and
metal-catalyzed decarbonylation reactions.[10] Rapid deactivation
of the Pd, likely due to CO poisoning as commonly reported,[11]
was marked by the disappearance of these decomposition
products, trending towards unpromoted ZrO2.
Ethanol reaction over ZnO primarily produced ethyl acetate
(44.5%) followed by acetone with an ethanol conversion of around
64%. Ethyl acetate can be formed via either the Tishchenko
reaction from the condensation of two acetaldehydes or through
the dehydrogenation of a hemiacetal formed through the ethanol-
acetaldehyde interaction. The redox functionality of ZnO along
with its minimal acidic sites favor the dehydrogenation of ethanol
followed by either ester formation or aldol condensation over
basic sites, rather than dehydration.[12] Conversion of ethanol to
acetaldehyde and further condensation to acetone has been
explored on different structures of ZnO, and these experimental
results were consistent with those found in literature.[13] The
addition of 0.1% Pd to ZnO improved the ethanol conversion to
~98% and largely eliminated ethanol dehydration products,
reflecting significant improvement in dehydrogenation. Pd also
contributed to greater ketone formation and condensation activity,
resulting in a much higher selectivity to ketones (~58%) over
esters, with a small increase to coupled alcohol products as well.
However, the selectivity to acetone was quite high (~19.3%) with
2-pentanone being other major ketone product, resulting in a low
average ketone carbon chain length (~4). Interestingly, very little
ethanol decomposition products were detected, unlike in the
reaction over the 0.1%Pd-ZrO2 catalyst, with the increase in CO2
directly corresponding to increased ketone formation.
Furthermore, the catalyst was found be very stable in contrast to
the rapid Pd deactivation observed on ZrO2. These results
indicate a significant change in the catalytic and physical nature
of the Pd functionality in the presence of ZnO.
Figure 1. Overview of the production of higher ketones (C5+) from biomass via
a hybrid biological/chemical pathway.
Results and Discussion
Over the mixed oxide ZnO-ZrO2 catalyst, the conversion of
ethanol was ~99% with a 25.9% selectivity toward ketones.
Though the high conversion and reduced ethanol dehydration (%
ethylene) signify higher dehydrogenation activity compared to the
ZnO and ZrO2, the presence of unsaturated enones such as 3-
penten-2-one indicate insufficient C=C hydrogenation activity.
These cross-condensed intermediates likely lead to the
Catalyst Performance and Reaction Mechanism
To develop an active catalyst and understand the reaction
mechanism for the direct conversion of ethanol to C5+ ketones,
various combinations of ZnO and ZrO2 catalysts were synthesized
with and without the Pd metal promoter. These catalyst
formulations were evaluated at 370 °C temperature, 300 psig
pressure, and 0.15 h-1 weight hourly space velocity (WHSV). The
detailed results of these experiments are presented in Table 1,
see supplementary section for the catalyst preparation, catalyst
evaluation and product analysis methods.
Reactions over ZrO2 produced primarily ethanol
dehydration products, namely diethyl ether and ethylene with a
combined selectivity close to 50%. Additional dehydration
products were detected in the form of 1,3-butadiene (11.3%) as
well as propene and pentene (27%), corresponding to aldol
condensed crotyl alcohol and acetone/2-pentanone, respectively.
These products can be attributed to the acid sites present in ZrO2
that lead to unimolecular and bimolecular dehydration (Table S4).
considerable cyclic formation, specifically
7 and 9-carbon
cyclohexenones and phenols through Michael addition/ring
annulation and may have contributed to an observed deactivation
trend, presumably as a result of coking. Self-condensation of
acetone was also observed as isobutene formation following
unselective decomposition on balanced Lewis acid-base pair
sites, as recognized in ZnO-ZrO2 mixed oxides.[14] Significant
dehydration rates also led to propene and pentene side products
from C3/C5 ketones and alcohols. Accounting for these side
reactions, it is evident that though condensation rates on ZnO-
ZrO2 are considerably higher compared to ZnO and ZrO2, ketone
Combined with the lack of redox sites for ethanol dehydrogenation, selectivity remains limited due to insufficient redox activity.
this results in the preferential generation of dehydration products
over ZrO2. Next, ZrO2 was promoted with 0.1% Pd to provide the
necessary redox sites, with the results showing a minor increase
in ketone formation and mild decrease in dehydration products
such as ether, as well as hydrogenation of most alkenes to
alkanes. At the same time, the introduction of Pd led to additional
To better assess the introduction of Pd, given the previously
observed interaction of Pd with ZnO, an initial test was performed
using a physical mixture of ZnO-ZrO2 and Pd-promoted SiO2 as a
control, with the same total Pd loading. Here, as in the Pd-
promoted ZrO2 catalyst, a significant quantity of methane and CO
was produced as decomposition products, with an even lower
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
[a]
Head 1[a]
ZrO2
0.1%Pd-ZrO2
ZnO
0.1%Pd-ZnO
ZnO-ZrO2
ZnO-ZrO2+Pd-SiO2
0.1%Pd-ZnO-ZrO2
Conversion
C5+ Ketones
Acetone
93.0
0.6
99.3
5.48
3.38
23.4
37.4
12.8
14.4
2.11
0.49
0.56
64.1
4.2
97.7
38.5
19.3
2.8
99.7
12.4
13.5
13.7
24.5
1.10
18.1
14.0
1.51
1.29
100.0
7.06
4.23
13.8
36.4
11.2
19.5
6.33
0.38
1.10
100.0
65.4
5.8
0.7
14.8
15.4
2.1
DEE/Ethane/Ethylene
C3+ Alkane/Alkene
CO/Methane
CO2
49.7
36.1
0.4
2.5
1.6
9.1
0.0
1.0
0.0
5.8
6.2
8.8
14.7
0.7
Cyclics[b]
0.0
2.7
0.1
Alcohol/Aldehyde
Ester
5.2
9.5
16.0
11.6
1.7
1.0
45.1
0.0
Table 1. Ethanol conversion and product selectivity over various catalyst combinations between ZnO, ZrO2 and Pd. Reaction conditions: Temperature 370 °C;
pressure 300 psig; WHSV 0.15 h-1. The analysis results were provided for the sample collected at time on stream of 24 hour. 100% pure ethanol was used as feed.
[a] Physical mix of ZnO-ZrO2 and Pd-SiO2 with Pd content equivalent to 0.1wt%. [b] Cyclic compounds represent compounds such as cyclohexenones,phenols etc.
selectivity toward ketones (11.3%). Though small alkene products
are hydrogenated, the continued formation of cyclic compounds
and as well as isobutene indicates the ineffective hydrogenation
of unsaturated ketone intermediates by the physically mixed Pd,
despite the presence of H2. As with Pd-ZrO2, rapid deactivation
as measured by the decrease in CO and methane products over
ZnO-ZrO2-Pd-SiO2 resulted in a deactivated product profile largely
identical to unpromoted ZnO-ZrO2. The Pd in a physical mixture
temperatures, until 370 °C and above when dehydration becomes
relevant. This results in a large window of temperatures where the
ketone yield is considerable, with the major change being the
distribution of ketone products as the average chain length grows
from about 5 to 6. Comparison with the results from unpromoted
ZnO-ZrO2, in Figure S8b show that without Pd, ketone yield never
becomes significant at lower temperatures due to the lower
conversion and high ester selectivity, and remains limited by
cyclic and dehydration side product formation at 370 °C. It is clear
that Pd significantly improves the dehydrogenation/hydrogenation
activity, and at the same time also appears to contribute to the
overall selectivity of ketones over esters. To confirm this effect,
ZnO-ZrO2 catalysts with 0.05 wt% and 0.5 wt% Pd were also
prepared and tested at 340 °C. Ketone formation was found to be
highly dependent on the Pd concentration, as overall yield nearly
doubled at 0.5 wt% loading compared to 0.05 wt%, corresponding
to an improvement in ketone selectivity from 43.95 to 74.6% in
conjunction with a small 8% increase in conversion. In addition,
no esters were observed during reaction over 0.5 wt%, resulting
in a product profile similar to that observed at 370 °C for 0.1 wt%
Pd. However, the average ketone chain length remained relatively
invariant to Pd loading, showing a slight increase from 5.0 to 5.1,
suggesting that the effect of Pd is related the formation of the
ketone moiety rather than ketone chain growth. Similar results
were obtained by studying Pd concentration on ZnO catalyst,
showing that added Pd could increase the total ketone yield to >
60% even at 340 °C, but had no effect on the ketone distribution
with a static chain length of 4.0 (See Supplementary Section
Figure S9). To better understand the reaction mechanism,
reaction data was gathered for the 0.1wt%-ZnO-ZrO2 catalyst for
lower conversions by varying space velocity. Figure 2b and S9
shows the corresponding changes in the product distribution,
where increasing WHSV corresponds to lower conversions and a
large shift in selectivity away from ketones towards esters and
alcohols/aldehydes, as well as a reduction in product chain
lengths. The same trends were observed with decreasing reaction
thus
does
not
significantly
contribute
to
the
dehydrogenation/hydrogenation activity or ketone formation, and
the Pd is instead mostly active toward the unselective
decomposition of ethanol.
In contrast, directly impregnating 0.1%Pd onto a ZnO-ZrO2
catalyst resulted in the highly selective production of ketones
(~71%), with a higher average carbon number (~6) and minimal
acetone selectivity (<10%). A significant proportion were C7+
ketones, resulting from the sequential cross condensation of
ketones with aldehydes. The detailed product distribution from
this experiment is provided in the inset of Figure 2a. The stable
catalyst performance and lack of ethanol decomposition products,
compared to the physical mixture, again point to a significant
change in the nature of Pd upon direct contact with the support,
as observed with Pd-ZnO. The catalyst now proved much more
effective in C=C hydrogenation, largely eliminating the formation
of cyclic and isobutene products resulting from unsaturated
condensed ketones intermediates.
Further study of the 0.1%Pd-ZnO-ZrO2 catalyst was
conducted across a range of temperatures from 300 to 385 °C,
with the changes to carbon selectivity of the major products in
presented in Figure S8a. Beginning at 300 °C, conversion was
74% with esters being the predominant products along with some
alcohol condensation products. Raising the temperature to 340 °C
results in near complete conversion and ketone selectivities
above 60%, with a minor fraction of side products consisting
primarily of C6+ aldehydes and esters. These largely disappear as
ketone formation becomes more favorable with increasing
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
temperatures, indicating the in-situ formation of acetone was a
slower secondary process. The rates of formation of key
compounds as a function of contact time is shown in Figure 2b.
Lower contact times (i.e. < 20 min) were not tested due to the
cyclic compound formation from unsaturated condensations.
Importantly, we note that the total ketone production is
approximately one-to-one by moles to CO2. This is consistent with
a decarboxylation pathway for the production of intermediates, or
the reaction pathway, as shown in Figure 2c, they were found to
be largely mitigated upon the introduction of Pd, specifically those
promoted on zinc containing catalysts. Elevated dehydrogenation
activity suppressed dehydration reactions while effective
hydrogenation of α,β-unsaturated ketone intermediates
prevented potential subsequent decomposition or cyclic products.
Pd was found to directly increase ketone formation over esters
and alcohols.
and exhibited a wide operating reaction
through water gas shift from CO generated from
decarbonylation pathway.[15]
a
temperature regime without catalyzing unselective ethanol
decomposition. This surprising effectiveness of Pd towards
achieving high ketone selectivity over ZnO-ZrO2 catalysts
warranted further investigation. Supplementary section includes
the detailed discussion on the reaction steps involved in the
formation of acetone (Figure S6) and experimental details on the
intermediate feed experiments to demonstrate the evidence for
the cross-aldol condensation between acetaldehyde and acetone
to form 2-pentanone (Figure S7 and Table S4).
These experimental results demonstrate that a sequential
C-C coupling pathway can be used to efficiently build long chain
ketones up to C13 starting from ethanol in a single catalyst bed.
The regular distribution of ketone products at odd carbon number
intervals is consistent with the preferential cross-condensation of
aldehydes with acetone formed in-situ, which limits excess CO2
byproduct formation (see SI for details). Though other major side
products can form from the numerous intermediate steps along
a) 100
b)
Acetaldehyde
C4+ Condensation
CO2
CO2
16.37%
Acetone
11.04%
0%Pd
Ketones Products*
25
20
15
10
5
Ethylene
Ethyl Acetate
C6+ Esters
0.05%Pd
0.1%Pd
0.5%Pd
Alcohols
2.74%
80
Acetone
C5+ Ketones
Alkenes
6.83%
2-Pentanone
33.54%
C11+ Ketones
6.28%
60
40
20
0
C9 Ketones
6.57%
2-Heptanone
6.87%
4-Heptanone
9.75%
0
40
80
120
160
200
Aldol Condensation
Esters
Ketones
Contact Time (Min)
Figure 2. (a) Carbon yield distribution of major product groups with varying Pd loading on ZnO-ZrO2 at 340 °C, 0.21 MPa EtOH, 1.86 MPa N2 and 0.15 h-1, inset
showing detailed product carbon selectivity from the ethanol conversion over 0.1%Pd-ZnO-ZrO2 at 370 °C, 0.21 MPa EtOH, 1.86 MPa N2 and 0.15 h-1; (b) Molar
yield of key products as a function of contact time conducted by varying catalyst loading for ethanol conversion over 0.1%Pd-ZnO-ZrO2 at 370 °C, 0.21 MPa EtOH,
1.86 MPa N2 and 0.15 h-1. *Denotes the sum of products including all ketones such as acetone, 2-pentanone and 3-penten-2-one, their respective alcohols such as
isopropanol and 2-pentanol, their respective dehydration products such as propene and trans-2-pentene, as well as isobutene from acetone self-condensation (c)
Simplified reaction mechanism for the ethanol conversion to C5+ ketones and the potential side reactions (compounds shown in gray are not found in product stream.
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
Catalyst Characterization
The catalyst evaluation results demonstrate that intimate contact
is required between Pd and ZnO, as well as between ZrO2 and
predominately as PdO. Upon heating to 370 °C in the presence
of ethanol, the Pd in the catalyst reduced to the metallic state
(PdO→Pd). (Figure S3a) The ambient-condition Extended X-Ray
ZnO. To understand the intrinsic reasons for the site requirements, Absorption Fine Structure (EXAFS) spectrum fit is consistent with
we undertook an extensive catalyst characterization studies using
X-ray Absorption Spectroscopy (XAS), Temperature
a four-coordinated Pd-O species. (Figure S3b) In the R-space of
the operando scan at 370 °C in the presence of ethanol (Figure
3a), we observe scattering at distances shorter than the expected
Pd-Pd scattering. This suggests the presence of a scattering atom
other than Pd as nearest neighbor to Pd. Based on the relative
atomic sizes, we fit the EXAFS spectrum using Pd-Pd scattering
at 2.733Å and Pd-Zn scattering at 2.565Å. To achieve a
meaningful fit without high correlations, we used the Debye-
Waller factors from our previous work on Pd-Cu alloys.[17] The
small coordination numbers obtained (Table S2) suggest small
nanoparticles of Pd-Zn alloy.
Programmed Reduction (TPR), and Transmission Electron
Microscopy (TEM), experimental details listed in the
supplementary section. Wang et al. researched the interaction of
ZnO and ZrO2. In their study, which focused on the aldol
condensation reaction to convert ethanol to isobutene, they
reported that the incorporation of ZnO into the ZrO2 promotes the
formation of new Lewis acid-base active structures (i.e., Zr-O-Zn)
on the catalyst surface, which play a critical role in the activity of
the cascade aldol-condensation chemistry.[16] The TPD and Py-
IR results from our studies on the ZnO-ZrO2 catalyst show a
similar trend in the acid–base characteristics (Figure S2, Table
S3). The addition of 0.1% Pd to the ZnO-ZrO2 results in a further
increase in the number of Lewis acid and base sites. However,
this did not significantly impact the cross-condensation chemistry,
as experiments reacting acetone and butanal over Pd-promoted
and unpromoted ZnO-ZrO2 resulted in largely identical product
profiles of primarily 4-heptanone. (Figure S8) Further
investigation of the incorporation of Pd over the ZnO-ZrO2 matrix
and the resulting effect on its redox characteristics was thus
necessary to explain the higher catalyst activity of Pd-ZnO-ZrO2
compared to the other catalyst compositions tested.
This picture is also consistent with the results of TPR
experiments (Figure S4). Two reduction peaks may be observed
o
for the 0.1%Pd-ZnO-ZrO2 catalyst: a small peak at 80 C and a
larger one at 340 oC. We tentatively attribute these to the
reduction of PdO to Pd and to the formation of a Pd-Zn alloy as a
result of Zn reduction, respectively. We observe no peaks under
500 oC for the temperature-programmed reduction of ZnO,
leading us to conclude that hydrogen spillover from Pd to ZnO is
responsible for the formation of the Pd-Zn alloy. The as-prepared
and spent 0.1% Pd-ZnO-ZrO2 catalysts were further analyzed
using STEM imaging and EDS mapping. Consistent with the
requirement for intimate contact between ZnO and ZrO2, the
catalyst support material was confirmed to be predominately a
mixture of the two oxides.
The Pd K-edge X-ray Absorption Near Edge Structure
(XANES) spectra of the 0.1%Pd-ZnO-ZrO2 catalyst suggest that
under ambient conditions the Pd in the catalyst exists
1.5
a)
b)
3
Magnitude
Fit
Pd-Zn
2.565 Å
1.0
2
1
0
Pd-Pd
2.733 Å
0.5
PdZrZn under CO/He
PdZrZn under He
Pd/SiO2 under CO/He
Pd/SiO2 under He
3180 3185
0.0
1
2
3
4
3165
3170
3175
R (Å)
Energy (eV)
Pd = 51.5 at.%
Zn = 48.5 at.%
O(K)
O(K)
Zr(K)
Zr(K)
c)
d)
Zn(K)
Pd(K)
Zn(K)
Pd(K)
Figure 3. (a) Pd K-edge EXAFS in operando under ethanol flow and 370 °C and atmospheric pressure; (b) Normalized Pd LIII-edge XANES spectra of reduced
0.1% Pd on ZnO-ZrO2 catalyst compared to reduced Pd on SiO2 control, and under CO flow; (c) HAADF-STEM and corresponding Zr, Zn, and Pd EDS mapping
on fresh 0.1% Pd loaded on ZnO-ZrO2 catalyst; and (d) HAADF-STEM and corresponding Zr, Zn, and Pd EDS mapping on spent 0.1% Pd loaded on ZnO-ZrO2
catalyst after >100 hours on stream.
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
This can be seen in the results from energy dispersive
spectroscopy (EDS) mapping (Figure 3c, 3d), which shows both
Zn and Zr uniformly distributed throughout the oxide phase. In
the as-synthesized sample, individual Pd nanoparticles of ~2.7
nm in diameter were identified in the EDS maps (Figure 3c); the
HAADF imaging provided only minimal contrast. In the spent
sample (collected after 100 hours on stream at 370 °C and 300
psig), shown in Figure 3d, we find that the Pd nanoparticles are
enriched in Zn with an average Pd/Zn ratio of approximately 1 with
the measured average particle size of 10.2 nm. This TEM analysis
supports the results of the EXAFS analysis indicating the
formation of the Pd-Zn alloy phase in the presence of ethanol at
the reaction temperature. This is consistent with the work of
Waele et al., who have noted the activity of Pd-Zn alloys in ethanol
dehydrogenation.[18] These results point to the possible role of Zn
as a modulator for the Pd reactivity, much like the role of Cu in
PdCu alloys in alcohol dehydrogenation reactions.[11b]
Pd LIII-edge XANES was also performed to clarify the
associated changes in Pd after alloy formation. Prior to reduction,
the catalyst is oxidized. The oxidation is evidenced by the height
of the white line. Consistent with the TPR results, there is a
reduction of the strength of the white line between the 80 and 130
oC scans, suggesting a reduction of the Pd. (Figure S5a)
Furthermore, at higher temperatures, the edge shifts to higher
energies, suggesting that there is an electronic modification of the
Pd. This could be ascribed to alloying with Zn. (Figure S5b)
Further evidence for that can be found by comparison of the
reduced 0.1%Pd-ZnO-ZrO2 catalyst with a Pd-SiO2 standard; the
edge is shifted by about 0.7 eV, consistent with alloying.[19]
Moreover, there is a feature approximately 15 eV above the edge
position similar to that reported by Tew, et al. for the bonding of
The formation of Pd-Zn alloy modulates the role of Pd in the
catalyst matrix in several ways. First, the alloy increases the rate
[18]
of ethanol dehydrogenation to form acetaldehyde selectively
.
Second, the alloy selectively hydrogenates α,β-unsaturated
ketone intermediates such as 3-penten-2-one to avoid the
formation of undesired cyclic compounds. Finally, the Pd-Zn alloy
creates a balanced environment between the redox, acid and
base active sites to selectively form the C5+ ketones from ethanol
via the complex cascade reaction mechanism.
Catalyst Stability
To assess the stability of the 0.1%Pd-ZnO-ZrO2 catalyst, a long-
term experiment was conducted for a period over 2000 hours in a
plug flow reactor arrangement. Figure 4 represents the detailed
product selectivities from the samples collected at periodic
intervals over the 2000-hour experiment. The ethanol conversion
was consistently ≥ 99% for the duration of the experiment. During
this study, extreme fluctuations in the operating conditions
occurred periodically due to incidents such as power outages.
These fluctuations included temperature, pressure, ethanol feed
flow rate, and carrier gas flow rate. The catalytic activity quickly
returned to its normal level once the operating conditions were
restored every time these fluctuations occurred. Throughout the
duration of the experiment, the catalyst did not show any
deactivation towards the ethanol conversion and the product
selectivity. The formation of Pd-Zn alloy can be seen to play a key
role in the high level of stability of the Pd-ZnO-ZrO2 catalyst over
time and its robustness across wide operating windows.
Typical deactivation mechanisms for ZnO-ZrO2 and other
catalysts, such as TiO2, hydroxyapatite and basic metal oxides
during gas-phase aldol condensation chemistry were reported to
Pd and H;[20] this suggests bonding between Pd and a heteroatom, be due to the strong adsorption of high molecular weight
compounds over the active sites, and this was demonstrated
in this case Zn. with the shift corresponding to charge transfer of
the PdZn alloy. Furthermore, the XANES spectra showed no
changes under the presence of CO, indicating weak CO binding
characteristics that differ from bulk Pd.[21] These results offer a
direct explanation for the absence of the decarbonylation
products observed during EtOH conversion over Pd-Zn catalysts.
It has been reported elsewhere that intermetallic alloying
between Pd and Zn offers an improved dispersion of Pd in the
catalyst matrix in addition to the unique and homogeneous
collaborative active sites, providing improvement in a variety of
reactions such as alkyne hydrogenation[22] and steam reforming
[23]. In this work, for the first time we report the formation of a Pd-
Zn alloy on a ZnO-ZrO2 catalyst and the role of the alloy in the
tandem dehydrogenation-aldol condensation chemistry. The alloy
formation makes this catalyst both highly active, stable, and
selective to producing C5+ ketones from ethanol. In general, Pd in
metallic form is known to decompose ethanol and produce
methane, CO, and hydrogen[23] which is clearly evident in our
experiments with both the Pd impregnated ZrO2 catalyst and the
physical mix of Pd-SiO2 plus ZnO-ZrO2 catalysts. However, in the
case of Pd impregnated on ZnO and the Pd impregnated on ZnO-
ZrO2 catalyst, the selectivity to the methane and CO is negligible.
We attribute that to the in-situ formation of Pd-Zn alloy by
hydrogen spillover from Pd to the ZnO on the 0.1%Pd
impregnated on ZnO-ZrO2 catalyst matrix, consistent with a
PdO→Pd→Pd-Zn reductive process.
through facile regeneration by calcination in air.[14, 24] Pd metal can
also be easily deactivated from particle growth for various reasons
such as coke deposition and coke transformation, modification of
the Pd surface and the variations of physical properties, as well
as chemical poisoning and leaching.[25] The interaction of Pd with
Zn prevents the Pd from sintering and mitigates coke deposition
and chemical poisoning, stabilizing the catalyst for more than
2000 hours of time on stream, even at temperature regimes where
Pd diffusion would otherwise be high and sintering likely
Figure 4. Product selectivity snapshot at various points time on stream during
the 2000 hour catalyst stability experiment. Experiment was conducted at
370 °C temperature, 300 psig pressure, 0.15 hr-1 WHSV with nitrogen carrier
gas and ethanol feed solution.
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
[4]
[5]
a) C. Manochio, B. R. Andrade, R. P. Rodriguez, B. S.
Moraes, Renew Sust Energ Rev 2017, 80, 743-755; b) A.
D. Li, M. Khraisheh, Int J Chem React Eng 2010, 8.
E. C. D. Tan, L. J. Snowden-Swan, M. Talmadge, A.
Dutta, S. Jones, K. K. Ramasamy, M. Gray, R. Dagle, A.
Padmaperuma, M. Gerber, A. H. Sahir, L. Tao, Y. Zhang,
Biofuels, Bioproducts and Biorefining 2017, 11, 41-66.
R. A. Dagle, A. D. Winkelman, K. K. Ramasamy, V.
Lebarbier Dagle, R. S. Weber, Industrial & Engineering
Chemistry Research 2020, 59, 4843-4853.
a) J. M. Sun, K. K. Zhu, F. Gao, C. M. Wang, J. Liu, C. H.
F. Peden, Y. Wang, J Am Chem Soc 2011, 133, 11096-
11099; b) C. Smith, V. L. Dagle, M. Flake, K. K.
Ramasamy, L. Kovarik, M. Bowden, T. Onfroy, R. A.
Dagle, Catal. Sci. Technol. 2016, 6, 2325-2336.
J. M. Sun, R. A. L. Baylon, C. J. Liu, D. H. Mei, K. J.
Martin, P. Venkitasubramanian, Y. Wang, J Am Chem Soc
2016, 138, 507-517.
K. A. Goulas, S. Sreekumar, Y. Song, P. Kharidehal, G.
Gunbas, P. J. Dietrich, G. R. Johnson, Y. C. Wang, A. M.
Grippo, L. C. Grabow, A. A. Gokhale, F. D. Toste, Journal
of the American Chemical Society 2016, 138, 6805-6812.
L. Di, S. Yao, M. Li, G. Wu, W. Dai, G. Wang, L. Li, N.
Guan, ACS Catalysis 2015, 5, 7199-7207.
a) E. I. Gürbüz, D. D. Hibbitts, E. Iglesia, Journal of the
American Chemical Society 2015, 137, 11984-11995; b)
K. A. Goulas, Y. Song, G. R. Johnson, J. P. Chen, A. A.
Gokhale, L. C. Grabow, F. D. Toste, Catalysis Science &
Technology 2018, 8, 314-327; c) J. L. Davis, M. A.
Barteau, Journal of the American Chemical Society 1989,
111, 1782-1792.
L. Wu, A. Gokhale, K. A. Goulas, J. E. Myers, F. Dean
Toste, C. D. Scown, ACS Sustainable Chemistry &
Engineering 2018, 6, 14523-14532.
H. Zhang, J. M. Sun, C. J. Liu, Y. Wang, J Catal 2015,
331, 57-62.
C. Liu, J. Sun, C. Smith, Y. Wang, Applied Catalysis A:
General 2013, 467, 91-97.
R. A. Dagle, A. Platon, D. R. Palo, A. K. Datye, J. M.
Vohs, Y. Wang, Applied Catalysis A: General 2008, 342,
63-68.
R. A. L. Baylon, J. Sun, L. Kovarik, M. Engelhard, H. Li, A.
D. Winkelman, Y. Wang, Applied Catalysis B:
Environmental 2018, 234, 337-346
K. A. Goulas, S. Sreekumar, Y. Song, P. Kharidehal, G.
Gunbas, P. J. Dietrich, G. R. Johnson, Y. C. Wang, A. M.
Grippo, L. C. Grabow, A. A. Gokhale, F. D. Toste, Journal
of the American Chemical Society 2016, 138, 6805-6812.
J. De Waele, V. V. Galvita, H. Poelman, C. Detavernier, J.
W. Thybaut, Applied Catalysis A: General 2018, 555, 12-
19.
a) C. Yang, Z. Wu, G. Zhang, H. Sheng, J. Tian, Z. Duan,
H. Sohn, A. J. Kropf, T. Wu, T. R. Krause, J. T. Miller,
Catalysis Today 2019, 323, 123-128; b) S. Gatla, J.
Radnik, N. Madaan, M.-M. Pohl, O. Mathon, A. Rogalev,
V. Narayana Kalevaru, A. Martin, S. Pascarelli, A.
Brückner, Chemistry – A European Journal 2015, 21,
15280-15289.
M. W. Tew, J. T. Miller, J. A. van Bokhoven, The Journal
of Physical Chemistry C 2009, 113, 15140-15147.
T. B. Bolin, T. Wu, N. Schweitzer, R. Lobo-Lapidus, A. J.
Kropf, H. Wang, Y. Hu, J. T. Miller, S. M. Heald, Catalysis
Today 2013, 205, 141-147.
Conclusion
In summary, we have developed a new Pd-promoted ZnO-ZrO2
catalyst to convert ethanol to C5+ ketones in a one-pot reaction
with ketones selectivity >70% at 100% ethanol conversion, with
catalyst stability >2000 hours of time on stream at 370 °C. We
also showed that (1) the reaction mechanism includes the in-situ
generation of acetone followed by the cross condensation
between acetaldehyde and acetone to generate C5+ ketones from
ethanol and (2) the formation of an intermetallic Pd-Zn alloy in-
situ during the reaction is essential for the activity, selectivity, and
stability of the catalyst.
[6]
[7]
[8]
[9]
Acknowledgements
Pacific Northwest National Laboratory (PNNL) is operated by the
Battelle Memorial Institute for the U.S. Department of Energy
under contract no. DE-AC05-76RL01830. This research was
conducted as part of the Co-Optimization of Fuels & Engines (Co-
Optima) project sponsored by the U.S. Department of Energy
(DOE) Office of Energy Efficiency and Renewable Energy (EERE),
Bioenergy Technologies and Vehicle Technologies Offices. The
authors also thank the Laboratory Directed Research and
Development program at PNNL for funding the project. The views
and opinions of the authors expressed herein do not necessarily
state or reflect those of the U.S. Government or any agency
thereof. Neither the U.S. Government nor any agency thereof, nor
any of their employees, makes any warranty, expressed or
implied, or assumes any legal liability or responsibility for the
accuracy, completeness, or usefulness of any information,
apparatus, product, or process disclosed, or represents that its
use would not infringe privately owned rights. Portions of this work
were performed at the DuPont-Northwestern-Dow Collaborative
Access Team (DND-CAT) located at Sector 5 of the Advanced
Photon Source (APS). DND-CAT is supported by Northwestern
University, E.I. DuPont de Nemours & Co., and The Dow
Chemical Company. Portions of this work were performed at
Sector 9 of the Advanced Photon Source (APS). This research
used resources of the APS, a DOE Office of Science (DOE-SC)
user facility operated for DOE-SC by Argonne National
Laboratory under Contract No. DE-AC02-06CH11357.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
Keywords: Mixed oxide catalyst, Ethanol upgrading, Aldol
condensation
[1]
a) Y. Wang, M. Peng, J. Zhang, Z. Zhang, J. An, S. Du, H.
An, F. Fan, X. Liu, P. Zhai, D. Ma, F. Wang, Nature
communications 2018, 9, 5183; b) Madhesan
[20]
[21]
Balakrishnana, E. R. Sacia, G. G. Sanil Sreekumara, A. A.
Gokhale, C. D. Scown, F. D. Toste, A. T. Bell, Proc Natl
Acad Sci 2015, 112, 7645–7649; c) P. Anbarasan, Z. C.
Baer, S. Sreekumar, E. Gross, J. B. Binder, H. W. Blanch,
D. S. Clark, F. D. Toste, Nature 2012, 491, 235-239.
Y. H. Wang, M. Peng, J. Zhang, Z. X. Zhang, J. H. An, S.
Y. Du, H. Y. An, F. T. Fan, X. Liu, P. Zhai, D. Ma, F.
Wang, Nat Commun 2018, 9.
[22]
[23]
H. Zhou, X. Yang, L. Li, X. Liu, Y. Huang, X. Pan, A.
Wang, J. Li, T. Zhang, ACS Catalysis 2016, 6, 1054-1061.
A. Casanovas, J. Llorca, N. Homs, J. L. G. Fierro, P.
Ramírez de la Piscina, Journal of Molecular Catalysis A:
Chemical 2006, 250, 44-49.
a) K. A. Goulas, S. Dery, P. Dietrich, G. R. Johnson, A.
Grippo, Y. C. Wang, E. Gross, Catalysis Today 2019, 336,
186-192; b) Z. D. Young, S. Hanspal, R. J. Davis, ACS
Catalysis 2016, 6, 3193-3202.
[2]
[3]
[24]
U.S.Energy Information Administration 2018.
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
[25]
P. Albers, J. Pietsch, S. F. Parker, J Mol Catal a-Chem
2001, 173, 275-286.
This article is protected by copyright. All rights reserved.

10.1002/anie.202005256
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents (Please choose one layout)
Layout 2:
RESEARCH ARTICLE
Senthil Subramaniam, Mond F. Guo,
Tanmayi Bathena, Michel Gray, Xiao
Zhang, Abraham Martinez, Libor
Kovarik, Konstantinos A. Goulas, and
Karthikeyan K. Ramasamy*
*Page No. – Page No.
Direct Catalytic Conversion of
Ethanol to C5+ Ketones: Role of Pd-Zn
Alloy on Catalytic Activity and
Stability
The formation of Pd-Zn alloy on a Pd-ZnO-ZrO2 results in the modification of the
Pd electronic structure and enables the highly selective formation of C5+ ketones
from renewable ethanol (>70%yield) for extended catalysts lifetimes above 2000
hours
This article is protected by copyright. All rights reserved.