Angewandte
Research Articles
Chemie
(Figures 3c and d). At higher magnification, the ridges of the
analyzed by AFM before and after electrocatalysis in the
cubes can be observed and they appear to be parallel to the
bulk of the electrode. This suggests that all the nanocubes are
growing along the h110i direction. In addition, as Cu has
a face-centered cubic (FCC) packing structure, the exposed
surface of such nanostructures has to be dominated by the
{100} facet. This could not unambiguously be confirmed by
analyzing the post-catalysis electrode by bulk XRD as the
change in the surface morphology is too small compared to
the resolution of the instrument. Nevertheless, by comparing
the peak heights of XRD spectra of a freshly electropolished
Cu electrode with post catalysis Cu electrodes with and
without 1-Br2, an increase of the Cu(100) peak is observed in
the post-catalysis Cu electrode with 1-Br2 (Figure S15).
In term of size, the cubic nanostructures display average
widths of 275 nm for the biggest ones and 100 nm for the
smallest. According to atomic force microscopic (AFM)
images, the nanocubes present an average high of 80–90 nm
(Figure S14c). SEM energy dispersive X-ray (EDX) analysis
confirms that the nanostructures are mainly comprised of
copper (Figure S14d).
presence of 10 mm of 1-Br2 (Figure 4). After 5 min of contact
between the electrode and the electrolyte, AFM images
These structural observations contrast with previous
reports on nanostructured copper electrodes.[16–18] This is the
first example of stable copper nanocubes under electro-
catalytic conditions over prolonged reaction times (> 40 h,
Figure S16). Also, the observed nanocubes after electro-
catalysis are well defined and oriented in the same direction,
contrasting with the typical morphologies of copper electro-
des after catalysis displaying agglomerated nanofeatures.
Inorganic halide salts are well known for inducing nano-
structuring process in copper surfaces.[36] Thus, we decided to
study the influence of the bromide counteranion in 1-Br2 on
the formation of the observed nanocubes at the electrode
surface. When a polycrystalline copper electrode is left in
contact with an aqueous solution of KBr (20 mm) for 5 min,
and then subjected to 3 measurements of potentiostatic
electrochemical impedance spectroscopy (PEIS), AFM im-
ages show the presence of nanocubes similar in shape to those
formed with additive 1-Br2, with a height of ca. 80–90 nm
(Figure S17a). This observation suggests that the bromide
anion can promote the formation of nanostructures, by
a corrosive process,[37] under our catalytic conditions. It is
worth noting that the PEIS measurements were performed at
open circuit voltage, resulting in a nanostructuring process
under mild conditions. This approach contrasts with the
previous reports where high oxidative potentials have been
used during the nanostructuring of Cu electrodes.[37,38]
When KBr (20 mm) was tested as an additive for CO2RR
at ꢁ1.07 VRHE, the product profile (Table S5) was similar to
that of bare Cu, with hydrogen and methane being the
dominant products (FEH = 56%, FECH = 20%). Ex-situ
Figure 4. Ex-situ AFM images of Cu electrodes a) after electropolish-
ing; b) after 5 min in contact with a 10 mm solution of 1-Br2 in 0.1m
KHCO3; c) after 5 min in contact with a 10 mm solution of 1-Br2 in
0.1m KHCO3 and 3 cycles of PEIS; d) after 65 min of electrocatalysis
at ꢁ1.07 V in 0.1m KHCO3 with 1-Br2.
already show some degree of well-defined nanostructuring
(Figure 4b), indicating that the bromide anion of the 1-Br2
additive is effective in corroding the electrode surface. After
three cycles of PEIS, AFM images clearly show the presence
of well-defined nanostructures (Figure 4c). These observa-
tions confirm that the nanostructuring process occurs via
corrosion of the copper surface, even before electrocatalysis.
As expected, the AFM images taken after electrocatalysis
(Figure 4d) show well-defined cubic nanostructures, consis-
tent with the previously discussed SEM images.
To study the effect of the counteranion in the molecular
additive, the corresponding phenanthrolinium ditriflate, di-
chloride and diiodide molecules were synthesized and tested
in CO2RR ([1-X2] = 10 mm, X = OTfꢁ, Clꢁ and Iꢁ, Table 1
and S6). The 1-(OTf)2 and 1-Cl2 derivatives show lower values
of total current densities compared with that of 1-Br2. Also,
the FE for ethylene production decreases 10–15%. Ex-situ
post-catalysis AFM images of the electrodes from the experi-
ments with 1-(OTf)2 and 1-Cl2 do not show well-defined
nanostructures (Figures S18a and S18b). On the other hand,
the selectivity of 1-I2 with respect to ethylene generation is
very similar to that with 1-Br2 (FEC H = 43%, jC H
=
2
4
2
4
2
4
post-catalysis AFM images of the electrode indicate a lower
level of nanostructuring compared to the case of 1-Br2 as
additive (Figure S17b). These observations allow us to infer
that although the organic film is not required for the
nanostructuring step, it is critical to preserve the nanocubes
during electrocatalysis.
ꢁ1.07 mAcmꢁ2). The overall current density is, however,
lower and comparable to that obtained with 1-(OTf)2 and 1-
Cl2. Compared to the organic film derived from 1-Br2,
a thicker film is observable on the electrode when 1-I2 is
used, accounting for the decrease in total current density
(Figure S18c). Well-defined nanostructures are observed by
AFM after dissolving the film with DMSO (Figure S16d).
These observations support previous reports on the incapacity
To further explore the role of bromide anion in the
nanostructuring process, the surface of a copper electrode was
Angew. Chem. Int. Ed. 2019, 58, 2 – 9
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&&&&
These are not the final page numbers!