Surface Reconstruction of Ultrathin Palladium Nanosheets during Electrocatalytic CO2 Reduction

Article abstract of DOI:10.1002/anie.202009616

A surface reconstructing phenomenon is discovered on a defect-rich ultrathin Pd nanosheet catalyst for aqueous CO₂ electroreduction. The pristine nanosheets with dominant (111) facet sites are transformed into crumpled sheet-like structures prevalent in electrocatalytically active (100) sites. The reconstruction increases the density of active sites and reduces the CO binding strength on Pd surfaces, remarkably promoting the CO₂ reduction to CO. A high CO Faradaic efficiency of 93 % is achieved with a site-specific activity of 6.6 mA cm^?2 at a moderate overpotential of 590 mV on the reconstructed 50 nm Pd nanosheets. Experimental and theoretical studies suggest the CO intermediate as a key factor driving the structural transformation during CO₂ reduction. This study highlights the dynamic nature of defective metal nanosheets under reaction conditions and suggests new opportunities in surface engineering of 2D metal nanostructures to tune their electrocatalytic performance.

Full text of DOI:10.1002/anie.202009616

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Surface Reconstruction of Ultrathin Palladium Nanosheets during
Electrocatalytic CO2 Reduction
Authors: Yong Zhao, Xin Tan, Wanfeng Yang, Chen Jia, Xianjue Chen,
Wenhao Ren, Sean C. Smith, and Chuan Zhao
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.202009616
Link to VoR: https://doi.org/10.1002/anie.202009616

10.1002/anie.202009616
Angewandte Chemie International Edition
COMMUNICATION
Surface Reconstruction of Ultrathin Palladium Nanosheets during
Electrocatalytic CO2 Reduction
Yong Zhao,+ Xin Tan,+ Wanfeng Yang, Chen Jia, Xianjue Chen, Wenhao Ren, Sean C. Smith, Chuan Zhao*
Dr. Y. Zhao, W. Yang, C. Jia, Dr. X. Chen, Dr. W. Ren, Prof. C. Zhao
School of Chemistry, University of New South Wales, Sydney, New South Wales 2052, Australia
Email: chuan.zhao@unsw.edu.au
Dr. X. Tan, Prof. S. C. Smith
Integrated Materials Design Laboratory, Department of Applied Mathematics, Research School of Physics, The Australian National University, Canberra,
ACT 2601, Australia
+ These authors contributed equally.
Supporting information for this article is given via a link at the end of the document.
Abstract: A surface reconstructing phenomenon is discovered on a
defect-rich ultrathin Pd nanosheets catalyst for aqueous CO2
electroreduction. The pristine nanosheets with dominant (111) facet
sites are transformed into crumpled sheet-like structures prevalent in
electrocatalytically active (100) sites. The reconstruction increases
the density of active sites and reduces the CO binding strength on Pd
surfaces, remarkably promoting the CO2 reduction to CO. A high CO
faradaic efficiency of 93% is achieved with a site-specific activity of
6.6 mA cm-2 at a moderate overpotential of 590 mV on reconstructed
50 nm Pd nanosheets. Experimental and theoretical studies suggest
other transition-metal nanocrystals may be susceptible to the
dynamic changes (e.g., local pH,[12] electric field,[13] intermediate’s
adsorption[14]) and undergo irreversible structural evolutions (e.g.,
phase transformation,[11] roughening or reconstructing,[14b,15]
nanoclustering[4a,16]
)
which in turn significantly alter the
electrocatalytic performance. To date, the structural
transformation of Pd catalysts during CO2 reduction and its impact
on catalytic properties have been rarely reported, particularly for
the thermodynamically unstable defect-rich Pd nanocrystals.
Here we report a dramatic surface reconstructing phenomenon
on a series of defect-rich ultrathin Pd NSs (20 nm, 50 nm and 120
nm in diagonal length) during aqueous CO2ER. Under reaction
conditions, the ultrathin hexagonal Pd NSs with dominant (111)
facet sites can transform into irregularly crumpled structures with
an increased exposure of (100) facets within an hour. This
structural transformation remarkably increases the surface
density of active sites and reduces the CO binding strength on Pd
surfaces, thus promoting the CO2-to-CO conversion. A high CO
selectivity of 93% can be achieved on the reconstructed 50 nm
Pd NSs at -0.7 V vs. reversible hydrogen electrode (RHE).
The ultrathin hexagonal Pd NSs with three different sizes were
prepared by adapting reported synthetic methods with
modifications.[17] Transmission electron microscopy (TEM)
analysis revealed the uniformity of hexagonal Pd NSs (Figure 1a,
Supporting Information Figure S1). Their average diagonal size
was 20.5 nm, 50.8 nm, and 120 nm, respectively. High angle
annular dark-field scanning TEM (HAADF-STEM) analysis
showed two lattice spacings of 0.223 nm and 0.191 nm on a
typical 50 nm Pd NS (Figure 1b, Figure S2), corresponding to the
(111) and (100) facets of face-centered cubic (fcc) Pd,
respectively.[17,18] The basal plane was dominantly hosted by
(111) facets. Numerous step edges, stacking defaults, and grain
boundaries were observed on the basal plane with partially
amorphized edges as well. These results revealed the defect-rich
nature of the as-obtained Pd NSs. In addition, these Pd NSs can
be well dispersed and attached onto commercial carbon blacks
without obvious aggregation (Figure 1c, Figure S3). High angle
the CO intermediate as
a key factor driving the structural
transformation during CO2 reduction. This study highlights the
dynamic nature of defective metal nanosheets under reaction
conditions and suggests new opportunities in surface engineering of
2D metal nanostructures to tune their electrocatalytic performance.
Electrocatalytic reduction of carbon dioxide (CO2ER) into value-
added chemicals and fuels is a promising approach to store
renewable energy and reduce the anthropogenic CO2 emission.[1]
The success of this venture relies on the development of efficient
and robust catalysts to control product selectivity at practical
potentials or currents.[2] State-of-the-art catalysts mainly include
the nanostructured metals, functionalized carbons and supported
metal complexes.[3] Elucidating the relationships between
electrocatalytic performances (i.e., selectivity and activity) and
operating structures (i.e., active sites) is crucial for designing and
exploring high-performance catalysts.[4]
Palladium (Pd) metal is well known for its excellent capability
of converting CO2 into CO or formate with high selectivity up to
90%,[4b,5] where both products are among the most desired ones
due to their market scale and economic viability.[6] The product
selectivity of Pd catalysts is highly dependent on their surface
binding abilities for key reaction intermediates such as *CO and
*COOH.[5a] By altering the surface structures (e.g., particle size,[4b]
lattice strain,[7] facets[8]) or compositions,[2a,9] the intermediates’
binding strength can be tuned to optimize the selectivity and
activity of Pd catalysts. For example, the Pd(100) facet exhibited
a higher catalytic activity than (111) facet due to its relatively low
binding strength to CO.[8] Pd nanoparticles (NPs) and nanosheets
(NSs) presented a highly size-dependent selectivity for CO2-to-
CO conversion, being manipulated by the surface ratio of
intrinsically active corner and edge sites.[4b,10] Such structure-
property relationships have been summarized recently,[14] and
most have assumed that the active sites of Pd catalysts remain
unchanged during CO2ER.[11] However, Pd nanostructures like
annular bright-field (HAABF) STEM analysis suggested
a
thickness of 1.0 ~1.5 nm with a clear view of various defective
sites mentioned above (Figure 1d). Infrared spectroscopy
analysis (Figure S4) revealed that the defective 2D Pd NSs were
stabilized by residual surfactants, which did not significantly block
the surface Pd sites for electrochemical measurements (Figure
S5). All above data indicate that the synthesized Pd NSs have
abundant structural defects on the basal surfaces and edges.
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202009616
Angewandte Chemie International Edition
COMMUNICATION
potentials between -0.5 V and -1.0 V vs. RHE in CO2-saturated
0.1 M KHCO3. Carbon monoxide (CO), formate, and hydrogen
(H2) products were detected (Figure S6). CO production was
dominant and H2 evolution was depressed in a wide potential
range from -0.6 V to -1.0 V, whereas formate production was
preferable at -0.5 V. All the three electrodes exhibited a typical
volcano-type distribution of CO faradaic efficiency (FECO) in the
whole potential range from -0.5 V to -1.0 V (Figure 2a). Pd-50 and
Pd-120 showed a peak FECO of 93% and 86% respectively at -0.7
V, whereas Pd-20 exhibited a peak of 91% at -0.8 V. Their mass-
specific catalytic activity, i.e., CO partial current density, exhibited
a size-dependent increasing trend (Figure 2b). Specifically, a
activity of 18 A g-1 was achieved on Pd-50 at -0.8 V, which is much
higher than that of 53 nm Pd NSs (~3 A g-1)[10] and even higher
than that of 4.5 nm Pd NPs (~15 A g-1)[4b] and 20 nm Pd
icosahedra (~5
A
g-1)[7]. Moreover, the three electrodes
demonstrated high FECO values (>75%) at -0.7, -0.8 and -0.9 V
with small difference (<10%) at each potential. This is surprisingly
different from the reported size-dependent CO selectivity for
structurally stable ultrathin Pd NSs, where the large Pd NSs (23
nm and 53 nm) exhibited poor selectivity (<50%) due to the low
density of active edge sites.[10] The high CO selectivity and mass
activity imply the formation of new active sites on our Pd NSs
under reaction conditions. Besides, the three electrodes revealed
a near steady-state i-t profiles during the electrolysis at -0.7 V
where the dramatic decrease of i at the beginning may be caused
by the CO poisoning on some Pd sites (Figure 2c).[5b] The FECO
of Pd-50 and Pd-120 remarkably increased from 66% to 90% and
from 50% to 83% after 2700 s, respectively, and that of Pd-20
increased from 77% to 84%. It indicates that the structure or
density of active sites on Pd surfaces varies during the electrolysis,
that is, a surface reconstruction occurs.
Figure 1. Structure analysis of 50 nm Pd NSs. (a) TEM image of Pd NSs laid
flat on the carbon film of Cu grid, inset is the diagonal length distribution; (b)
HAADF-STEM image of an individual Pd NS, inset shows its Fourier transform
pattern; (c) TEM image of Pd NSs loaded on carbon black; (d) HAABF-STEM
image of two NSs vertically attached on carbon black.
To evaluate their CO2ER performances, the ultrathin Pd NSs
dispersed on carbon supports were uniformly loaded on the gas-
diffusion-layer covered carbon papers. The resultant electrodes
were denoted as Pd-20, Pd-50, and Pd-120, respectively.
Controlled potential electrolysis was then performed at various
(c)
0
100
(b)
(a)
0
-20
-40
-60
-80
-10
80
60
40
20
0
-20
Pd-20
-30
Pd-50
Pd-120
@-0.7 V vs. RHE
100
75
Pd-20
Pd-50
Pd-120
Pd-20
Pd-50
Pd-120
50
0
1000
2000
3000
-1.0
-0.9
-0.8
-0.7
-0.6
-0.5
-1.0
-0.9
-0.8
-0.7
-0.6
-0.5
Time (s)
Potential (V vs. RHE)
Potential (V vs. RHE)
(d) 1.2
(e)
(f)
0
Pd-20
2 mA cm-2
@ -0.8 V
After 3000 s
0.6
0.0
0.6
1.2
0.6
0.0
-10
-20
-30
-40
@ -0.7 V
@ -0.6 V
@ -0.5 V
Pd-50
Pd-120
Pd-20
Pd-50
Pd-120
Pristine Pd-50
-1.0
-0.9
-0.8
-0.7
-0.6
-0.5
0.4
0.6
0.8
1.0
1.2
OCP
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4
Potential (V vs. RHE)
Potential (V vs. RHE)
Potential (V vs. RHE)
Figure 2. Potential-dependent CO2ER performances and electrochemical surface properties of Pd-20, Pd-50 and Pd-120. (a) CO faradaic efficiency; (b) CO mass
activity; (c) i-t curves at -0.7 V with the calculated FECO at 300 s, 1500 s, and 2700 s; (d) ECSAs of Pd-20, Pd-50 and Pd-120 after electrolysis at various potentials
for 3000 s, OCP represents open-circuit potential; (e) CO site-specific activity; (f) CO stripping voltammetry profiles of pristine Pd-50 and reconstructed Pd-50 at
various potentials.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202009616
Angewandte Chemie International Edition
COMMUNICATION
The structural evolution was verified by the variation of
electrochemically active surface areas (ECSAs) (Figure 2d) after
CO2ER measured by a Cu underpotential deposition method.[19]
Applied cathodic potential affected the ECSAs significantly, and
different ECSA evolution trends were observed for the three
electrodes. The Pd-120 showed an obvious increase in ECSA
after electrolysis at a low overpotential of -0.5 V, which can be
attributed to its restacking on carbon blacks (Figure S3 and S8).
In contrast, the increase occurred at a more negative potential of
-0.7 V and -0.9 V for Pd-20 and Pd-50, respectively, revealing the
ECSA evolution of these isolated Pd nanosheets under CO2ER
conditions. Particularly, the ECSA of Pd-50 only showed a slight
increase in the potential range of -0.6 V ~ -0.8 V; however, its
FECO increased remarkably during the electrolysis (Figure 2c),
indicating that the reconstructed Pd-50 surfaces are intrinsically
more active for CO production than the pristine surfaces.
Figure 3. Structural analysis of Pd NSs after CO2ER. (a-f) Typical grayscale
and false-color HAADF-STEM images of 50 nm Pd NSs after electrolysis at -
0.7 V for 300 s (a,d), 1500 s (b,e), and 3000 s (c,f), respectively; false color
highlights the surface roughness of Pd NSs.
Moreover, Pd-50 exhibited relatively high site-specific activity (iCO
)
compared with Pd-20 and Pd-120, while Pd-20 NSs with a
relatively high density of edges did not show obvious advantages
in iCO as displayed in mass activity. At -0.7 V, the iCO of Pd-50
reached 6.6 mA cm-2, nearly three times higher than that (2.3 mA
cm-2) of Pd-20. This implies that the structural transformation
mainly occurs on the inert (111) basal planes, inducing the
formation of CO2ER-favorable Pd sites and more intensely
promoting the CO formation on larger Pd NSs such as Pd-50.
Additionally, the three electrodes exhibited similar Tafel slopes
close to 120 mV dec-1 (Figure S9), suggesting that they have a
To visualize the structural transformation, HAADF-STEM
analysis was performed on the Pd-50 NSs after CO2ER (Figure
3, Figure S12). With an electrolysis of 300 s, the ultrathin sheet
morphology almost maintained with some nanopores penetrated
the basal planes and some edges slightly curled or thickened
(Figure 3a). The lattice spacings of (111) planes was contracted
to ~0.215 nm (Figure 3d, Figure S12). After a 1500 s electrolysis,
the flat basal surfaces were partially roughened and amorphized
(Figure 3b,e), and then were transformed into crumpled sheet-like
structures with numerous Pd(100) facets exposed around the
ridges, edges or pores after 3000 s (Figure 3c,f). The (111) lattice
spacing increased from ~0.220 nm to ~0.234 nm concurrently
(Figure S12). The enhanced exposure of (100) facets and
amorphization of (111) basal surfaces are consistent with the CO
stripping analysis in Figure 2f. The compressive lattice strain at
the early stage (<1500s) of CO2ER is triggered by the applied
cathodic potential, which can be relaxed eventually by forming
crumpled structures.[15a] Similar reconstructions with an increased
exposure of (100) facets were also detected on Pd-20 and Pd-
120 after reactions (Figure S13). This reconstruction created
numerous CO2ER-favorable (100) sites on the basal surfaces of
Pd NSs and weakened the dependence of CO production on the
edges, thus leading to the nearly size-independent high CO
selectivity for the three Pd NSs (Figure 2a).
To investigate the role of CO2ER in the reconstruction, Pd-50
electrodes were applied for electrolysis in phosphate buffer
solution (PBS) with various gas environments, and their CO
stripping spectra were recorded after electrolysis (Figure 4a).
Based on CO stripping analysis, the electrode applied in CO2-
saturated PBS exhibited an increased exposure of Pd(100)
surface after electrolysis, while its counterpart in Ar environment
did not. CO was the major electrolysis product as well in CO2-
saturated PBS, but only H2 was detected in Ar-saturated PBS
(Figure S14). It suggests the CO2ER reaction is the major driving
force for the structural transformation from (111) to (100).
Moreover, increased Pd(100) sites was also detected on the
electrode after electrolysis in CO environment under the same
potential, demonstrating the unique capability of CO molecule in
reconstructing Pd NSs. This also implies that the CO
similar rate-determining step towards CO production, i.e., one
•- [20]
electron transfer to CO2 forming CO2
.
The CO desorption is another key step for the CO2-to-CO
conversion. Suitable CO binding ability facilitates the CO
formation on Pd surfaces, which can be analyzed by an
electrochemical CO stripping voltammetry method.[16] Pd-50 was
took as a case study (Figure 2f, Figure S10). The stripping curve
of pristine Pd-50 consisted of two major peaks located at 1.07 V
and 0.87 V, corresponding to the oxidation of adsorbed CO at
Pd(111) and Pd(100) facets, respectively.[21] The relatively high
(111) peak intensity revealed the dominant exposure of (111)
facets on pristine Pd-50, and the broadening of both peaks were
attributed to the variety of defective sites. After electrolysis at
various potentials, the CO oxidation peak on (111) surfaces was
weakened while the peak related to (100) planes was significantly
intensified, indicating a major surface reconstruction from (111)
facets to (100) facets. The negative shift of (111) peak was
probably caused by the H adsorption on Pd.[22] X-ray diffraction
analysis confirmed the structural evolution of fcc Pd crystals
without forming other phases (Figure S11). Moreover, the surface
reconstruction was highly sensitive to the applied potential, as
revealed by the shift of the intensified (100)-related CO oxidation
peaks. Particularly, this peak shifted positively to 0.93 V after an
electrolysis at -0.8 V, approximate to the CO stripping potential on
Pd(110) facets.[21a] It demonstrates that the pristine (111) planes
can transform into a more open (110) planes at a higher
overpotential of -0.8 V vs. RHE. Basically, a more negative
oxidation potential suggests a weaker CO adsorption, and a weak
CO binding strength facilitates CO formation on Pd sites.[22] The
prevalence of reconstructed (100) or (110) surfaces with relatively
low CO binding strength can well explain the higher
electrocatalytic selectivity and activity observed for CO formation
in this study, compared with the previously reported structurally
stable 53 nm Pd NSs.[10]
intermediates play
a key role in inducing the surface
reconstruction of Pd NSs during CO2ER. In addition, the
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202009616
Angewandte Chemie International Edition
COMMUNICATION
(a)
(b)
2 mA cm-2
0.1 M PBS @ -0.7 VRHE
during electrolysis in the presence of CO, which can be supplied
either by CO2ER reaction or by external CO sources (Figure 4a,b).
Surface reconstruction is usually driven by the thermodynamic
demand for the lowest possible surface free energy.[23] This rule
may govern the roughing or crumpling of Pd NSs under CO2ER
conditions due to the intrinsically high surface free energy of the
defect-rich 2D metallic nanosheets.[24] However, the surface
reconstruction from Pd(111) to Pd(100) is unexpected, since the
(100) facets have a higher surface free energy than the more
compact (111) facets. Based on the above results, we conclude
that the surface reconstruction is driven by the highly negative
potential of CO2ER and assisted by the CO intermediates
produced in the reaction. Once the cathodic potential is applied,
CO2 can be adsorbed on Pd nanosheets and reduced to CO
intermediate that can be strongly bound to Pd(111) surface. The
highly negative potential offers sufficient repulsive electrostatic
forces to trigger the lattice contraction of the CO intermediates-
covered (111) basal surfaces and drive the movement of Pd
atoms from the defective sites such as step edges. The detached
Pd atoms eventually assemble into more open (100) regions or
aggregate into amorphous bumps in a local electric field. In this
process, the CO intermediates act as ligands to lower the energy
barriers for the detachment or as surfactants to direct the re-
assembly of Pd atoms.[25] Figure 4e schematically illustrates this
surface reconstruction process. Further understanding of the
roles of CO intermediates in the reconstruction may be achieved
by in situ spectroscopic methods such as Infrared and Raman,[14]
and will be explored in our future work.
In summary, a surface reconstruction phenomenon was
demonstrated on defect-rich ultrathin Pd NSs during CO2
electroreduction. The ultrathin Pd NSs with a dominant exposure
of (111) facets were transformed into crumpled sheet-like
structures prevalent in (100) regions within an hour electrolysis.
This reconstruction not only increased the density of active sites
but reduced the binding strength to CO intermediates, thus
facilitating the CO2-to-CO conversion with a high CO selectivity of
up to 93% at a modest overpotential of 590 mV. This work
highlights the instability of defective 2D metal nanostructures and
emphasizes the importance of correlating catalytic performances
to the operating surface structures under reaction conditions.
CO2 saturation @-0.7 VRHE
2 mA cm-2
pH = 6.8 (KHCO3)
CO2 (pH=6.6)
CO (pH=6.8)
Ar (pH=6.8)
pH = 6.6 (PBS)
pH = 5.6 (PBS)
pH = 3.1 (PBS)
Pristine Pd-50
0.6
0.4
0.6
0.8
1.0
1.2
0.4
0.8
1.0
1.2
Potential (V vs. RHE)
Potential (V vs. RHE)
1
(d)
(c)
Pd (111)
Bias: 0 V
0
-1
-2
Bias: -1.1 V
-3
-4
CO2 (g)
CO (g)
*CO
Reaction coordinate
*COOH
(e)
Figure 4. Experimental and theoretical investigation of the driving force for the
surface reconstruction on Pd NSs. (a-b) CO stripping voltammetry profiles of
Pd-50 after being electrolyzed in 0.1 M PBS with various gas environment (a)
and in CO2-saturated 0.1 M PBS with different pH (b); (c) Calculated Gibbs free
energy diagrams for CO2-to-CO reduction on Pd(111) surface at two different
bias; (d) Gibbs free energy changes for *CO adsorption on Pd(111) and the
transformation from Pd(111) to Pd(100) with different CO coverage at a -1.1 V
vs. SHE; (e) A schematic illustration of the surface reconstruction of ultrathin Pd
NSs during CO2ER.
reconstruction was not significantly affected by pH under CO2ER
conditions (Figure 4b). In acidic PBS (pH=3.1), the structural
transformation occurred while FECO dropped to 38% (Figure S14).
The structural transformation was also observed for the
electroded used in Ar-saturated bicarbonate solution with
relatively low FECO (Figure S15). These results indicate that a
moderate CO production during CO2ER may be sufficient to drive
the structural transformation of defect-rich Pd basal surfaces.
DFT calculations were further carried out to provide atomistic
insights into the structural transformation on Pd surface. As
shown in Figure 4c, CO2 can be converted to CO via two key
intermediates, i.e., *COOH and *CO, on Pd(111) at a bias of -1.1
V vs. SHE (-0.7 V vs. RHE). Due to the strong binding ability of
Pd(111), the reaction rate of CO production is limited by the *CO
desorption step (Figure 4c). A high *CO coverage can be
achieved on Pd(111) surface, as revealed by the calculated
negative Gibbs free energy changes (∆G) for CO adsorption on
Pd(111) with an increasing *CO coverage (Figure 4d, Figure S16).
Then the free energy changes for the transformation from (111)
to (100) surface were calculated in the presence of *CO (Figure
4d, Figure S16-17). For bare Pd(111), the transformation is
thermodynamically prohibited (∆G = 1.70 eV). With the increase
of *CO coverage, the required energy for the transformation is
significantly lowered. Particularly, this structural transformation
becomes spontaneous (∆G = -0.03 eV) when the Pd(111) sites
are fully covered by *CO, suggesting the crucial role of high local
*CO coverage in reconstructing Pd(111) to Pd(100). This explains
our experimental observations that the reconstruction only occurs
Acknowledgements
The study was supported by the Australian Research Council
(FT170100224). We thank the UNSW Mark Wainwright Analytical
Centre for providing access to their TEM, FTIR, XRD, NMR and
facilities. This research was undertaken with the assistance of
resources provided by the National Computational Infrastructure
(NCI) facility at the Australian National University; allocated
through both the National Computational Merit Allocation Scheme
supported by the Australian Government and the Australian
Research Council grant (LE190100021).
Keywords: CO2 reduction • electrocatalysis • Pd nanosheet •
CO • surface reconstruction
[1]
a) S. Chu, Y. Cui, N. Liu, Nat. Mater. 2017, 16, 16-22. b) C. Costentin,
M. Robert, J.-M. Savéant, Chem. Soc. Rev. 2013, 42, 2423-2436. c) W.
Ren, X. Tan, W. Yang, C. Jia, S. Xu, K. Wang, S. C. Smith, C. Zhao,
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202009616
Angewandte Chemie International Edition
COMMUNICATION
Angew. Chem. Int. Ed. 2019, 58, 6972-6976. Angew. Chem. 2019, 131,
748-752. d) D. Kim, C. S. Kley, Y. Li, P. Yang, Proc. Natl. Acad. Sci.
U.S.A 2017, 114, 10560-10565.
7046-7050.
[2]
a) J. H. Lee, S. Kattel, Z. Jiang, Z. Xie, S. Yao, B. M. Tackett, W. Xu, N.
S. Marinkovic, J. G. Chen, Nat. Commum. 2019, 10, 1-8. b) J. Huang, R.
Buonsanti, Chem. Mater. 2018, 31, 13-25. c) M. B. Ross, P. D. Luna, Y.
Li, C.T. Dinh, D. Kim, P. Yang, E. H. Sargent, Nat. Catal. 2019, 2, 648-
658.
[16] Y. Li, F. Cui, M. B. Ross, D. Kim, Y. Sun, P. Yang, Nano Letts. 2017, 17,
1312-1317.
[17] a) X. Huang, S. Tang, X. Mu, Y. Dai, G. Chen, Z. Zhou, F. Ruan, Z. Yang,
N. Zheng, Nat. Nanotech. 2011, 6, 28-32. b) Y. Li, Y. Yan, Y. Li, H. Zhang,
D. Li, D. Yang, CrystEngComm 2015, 17, 1833-1838.
[3]
a) J. Gu, C.S. Hsu, L. Bai, H. M. Chen, X. Hu, Science 2019, 364, 1091-
1094. b) F. Li, A. Thevenon, A. Rosas-Hernández, Z. Wang, Y. Li, C. M.
Gabardo, A. Ozden, C. T. Dinh, J. Li, Y. Wang, J. P. Edwards, Y. Xu, C.
McCallum, L. Tao, Z.-Q. Liang, M. Luo, X. Wang, H. Li, C. P. O’Brien, C.-
S. Tan, D.H. Nam, R. Quintero-Bermudez, T.T. Zhuang, Y. C. Li, Z. Han,
R. D. Britt, D. Sinton, T. Agapie, J. C. Peters, E. H. Sargent, Nature 2020,
577, 509-513. c) F. P. G. de Arquer, C.T. Dinh, A. Ozden, J. Wicks, C.
McCallum, A. R. Kirmani, D.H. Nam, C. Gabardo, A. Seifitokaldani, X.
Wang, Y. C. Li, F. Li, J. Edwards, L. J. Richter, S. J. Thorpe, D. Sinton,
E. H. Sargent, Science 2020, 367, 661-666. d) Y. Wu, Z. Jiang, X. Lu, Y.
Liang, H. Wang, Nature 2019, 575, 639-642. e) H. B. Yang, S.-F. Hung,
S. Liu, K. Yuan, S. Miao, L. Zhang, X. Huang, H.Y. Wang, W. Cai, R.
Chen, Nat. Energy 2018, 3, 140. f) C Jia, K. Dastafkan, W. Ren, W. Yang,
C. Zhao, Sustain. Energy Fuels 2020, 3, 2890-2906.
[18] D. Xu, X. Liu, H. Lv, Y. Liu, S. Zhao, M. Han, J. Bao, J. He, B. Liu, Chem.
Sci. 2018, 9, 4451-4455.
[19] a) S. Szabo, I. Bakos, F. Nagy, T. Mallát, J. Electroanal. Chem. Interf.
Electrochem. 1989, 263, 137-146. b) T. Takashima, T. Suzuki, H. Irie,
Electrochim. Acta 2017, 229, 415-421.
[20] a) Y. Zhao, J. Liang, C. Wang, J. Ma, G. G. Wallace, Adv. Energy Mater.
2018, 8, 1702524. b) Y. Zhao, C. Wang, G. G. Wallace, J. Mater. Chem.
A 2016, 4, 10710-10718. c) Y. Zhao, C. Wang, Y. Liu, G. G. Wallace,
Adv. Energy Mater. 2018, 8, 1801400
[21] a) M. Hara, U. Linke, T. Wandlowski, Electrochim. Acta 2007, 52, 5733-
5748. b) A. Januszewska, R. Jurczakowski, P. J. Kulesza, Langmuir
2014, 30, 14314-14321.
[22] C. W. Lee, N. H. Cho, K. T. Nam, Y. J. Hwang, B. K. Min, Nat. Commun.
2019, 10, 1-8.
[4]
a) J. Huang, N. Hörmann, E. Oveisi, A. Loiudice, G. L. De Gregorio, O.
Andreussi, N. Marzari, R. Buonsanti, Nat. Commum. 2018, 9, 1-9. b) D.
Gao, H. Zhou, J. Wang, S. Miao, F. Yang, G. Wang, J. Wang, X. Bao, J.
Am. Chem. Soc. 2015, 137, 4288-4291.
[23] G. A. Somorjai, Y. Li, Introduction to Surface Chemistry and Catalysis,
John Wiley & Sons: New York, 1994.
[24] a) Y. Chen, Z. Fan, Z. Zhang, W. Niu, C. Li, N. Yang, B. Chen, H. Zhang,
Chem. Rev. 2018, 118, 6409-6455. b) T. Ling, J. J. Wang, H. Zhang, S.
T. Song, Y. Z. Zhou, J. Zhao, X. W. Du, Adv. Mater. 2015, 27, 5396-5402.
[25] Y. Q. Su, J.X. Liu, I. A. Filot, E. J. Hensen, Chem. Mater. 2017, 29, 9456-
9462.
[5]
[6]
[7]
[8]
a) D. Gao, H. Zhou, F. Cai, J. Wang, G. Wang, X. Bao, ACS Catal. 2018,
8, 1510-1519. b) X. Min, M. W. Kanan, J. Am. Chem. Soc. 2015, 137,
4701-4708.
a) T. Zheng, K. Jiang, N. Ta, Y. Hu, J. Zeng, J. Liu, H. Wang, Joule 2019,
3, 265-278. b) W. Yang, S. Chen, W. Ren, Y. Zhao, X. Chen, C. Jia, J.
Liu, C. Zhao, J. Mater. Chem. A 2019, 7, 15907-15912.
H. Huang, H. Jia, Z. Liu, P. Gao, J. Zhao, Z. Luo, J. Yang, J. Zeng, Angew.
Chem. Int. Ed. 2017, 56, 3594-3598. Angew. Chem. 2017, 129, 3648-
3652.
a) H. Dong, L. Zhang, P. Yang, X. Chang, W. Zhu, X. Ren, Z.J. Zhao, J.
Gong, Chem. Eng. Sci. 2019, 194, 29-35. b) A. Klinkova, P. De Luna, C.-
T. Dinh, O. Voznyy, E. M. Larin, E. Kumacheva, E. H. Sargent, ACS Catal.
2016, 6, 8115-8120.
[9]
a) Y. Wang, L. Cao, N. J. Libretto, X. Li, C. Li, Y. Wan, C. He, J. Lee, J.
Gregg, H. Zong, J. Am. Chem. Soc. 2019, 141, 16635-16642. b) B. Jiang,
X.G. Zhang, K. Jiang, D.Y. Wu, W.B. Cai, J. Am. Chem. Soc. 2018, 140,
2880-2889. c) S. Zhu, Q. Wang, X. Qin, M. Gu, R. Tao, B. P. Lee, L.
Zhang, Y. Yao, T. Li, M. Shao, Adv. Energy Mater. 2018, 8, 1802238.
[10] W. Zhu, L. Zhang, P. Yang, C. Hu, Z. Luo, X. Chang, Z. J. Zhao, J. Gong,
Angew. Chem. Int. Ed. 2018, 130, 11718-11722. Angew. Chem. 2018,
130, 11718-11722.
[11] D. Gao, H. Zhou, F. Cai, D. Wang, Y. Hu, B. Jiang, W.B. Cai, X. Chen,
R. Si, F. Yang, Nano Res. 2017, 10, 2181-2191.
[12] M. Ma, K. Djanashvili, W. A. Smith, Angew. Chem. Int. Ed. 2016, 55,
6680-6684. Angew.Chem. 2016, 128, 6792-6796.
[13] L. D. Chen, M. Urushihara, K. Chan, J. K. Nørskov, ACS Catal. 2016, 6,
7133-7139.
[14] a) S. Zou, R. Gómez, M. J. Weaver, Surf. Sci. 1998, 399, 270-283. b) C.
M. Gunathunge, X. Li, J. Li, R. P. Hicks, V. J. Ovalle, M. M. Waegele, J.
Phys. Chem. C 2017, 121, 12337-12344.
[15] a) Y.G. Kim, J. H. Baricuatro, A. Javier, J. M. Gregoire, M. P. Soriaga,
Langmuir 2014, 30, 15053-15056. b) W. T. Osowiecki, J. J. Nussbaum,
G. A. Kamat, G. Katsoukis, M. Ledendecker, H. Frei, A. T. Bell, A. P.
Alivisatos, ACS Appl. Energy Mater. 2019, 2, 7744-7749. c) Y. Fu, M. R.
Ehrenburg, P. Broekmann, A. V. Rudnev, ChemElectroChem 2018, 5,
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202009616
Angewandte Chemie International Edition
COMMUNICATION
Table of Contents
A surface reconstruction is demonstrated on defect-rich ultrathin Pd nanosheets during CO2 electroreduction. The pristine inactive
basal surfaces are transformed into crumpled Pd(100)-dominant active surfaces with reduced CO binding strength, promoting the CO2
conversion to CO with high selectivity and productivity.
6
This article is protected by copyright. All rights reserved.