Communications
these values did not change across electrocatalyst testing. A Bio-
Logic VMP3 potentiostat was used in galvanostatic or potentiostat-
ic mode, depending on the experiment. For catalyst activation,
Experimental Section
Ni(NO ) ·6(H O) (Sigma–Aldrich) was used as nickel precursor,
3
2
2
À1
Cu(NO ) ·3(H O) (Sigma–Aldrich) and RuCl (Sigma–Aldrich) were
3
2
2
3
a cyclic voltammogram was taken at a scan rate of 20 mVs from
used as copper and ruthenium precursors, respectively. 10% PtRu/
VC (from http://www.fuelcellstore.com; accessed June 2016) and
À1.5 V to 0 V vs. Ag/AgCl and a subsequent 20 min chronoamper-
ometry at À1.5 V vs. Ag/AgCl were run before electrolysis. Catalyst
testing was conducted at a constant potential of À1.5 V (vs. Ag/
1
0% Pt/VC (Sigma–Aldrich) were used as-received.
+
The catalysts were prepared by incipient wetness impregnation.
First, the VULCAN XC72R carbon black (CABOT) support was acid-
treated to decrease hydrophobicity and to remove any impurities,
by stirring in 6m HCl for 16 h. After filtrating to pH 5, the retentate
was dried at 608C overnight. Then, 500 mg were ground using
a mortar and pestle for 30 min under dropwise addition of
AgCl) for a total reaction time of 5 h. To compensate for H -con-
sumption in the catholyte solution and to maintain a constant pH
throughout the reaction, 25 vol% H SO was continuously added
2
4
by a syringe pump (Harvard System) to the catholyte, using a 5 mL
syringe (BD). To prevent local pH decrease, a small-diameter hose
was placed within the solution, leading to near immediate and uni-
form proton distribution. The rate of addition varied by experi-
ment. Changes in the electrolyte conductivity and pH were moni-
tored and found to be negligible. Sample volumes taken for high
performance liquid chromatography (HPLC, Agilent 1200) were
251.35 mL for the starting concentration, 100 mL for all subsequent
samples. Volume loss due to gas evolution was compensated each
time before taking the sample by adding DI water. During chro-
2
.975 mL metal salt solution. The impregnated catalysts were dried
in the oven at 608C overnight and reduced at 4008C in a pure hy-
À1
drogen stream at a flow rate of 0.2 mLmin . Reduction time was
À1
1
h at a heat ramp of 28Cmin .
2
To improve conductivity and remove impurities, a 26 cm piece of
Nafion 117 (N117, Ion Power) was treated using the following pro-
cedure: the membrane was first boiled for 1 h in 3% H O and sub-
2
2
sequently boiled in DI water 3 times, for 30 min each time. The
membrane was rinsed with DI water after each boil. The mem-
brane was then boiled in 0.25m H SO for 1 h, and again boiled in
DI water 3 times for 30 min each time and rinsed with DI water in-
between.
Catalyst powder (10 mg) was dispersed in a mixture of 400 mL of
DI water, 400 mL of isopropanol (Sigma–Aldrich), and 6.9 mL of 5%
Nafion solution (Liquion, Ion Power). The slurry was sonicated for
noamperometry, N gas was purged into the catholyte headspace
2
in order to dilute the hydrogen stream and all gas phase products.
The exit streams of both electrolyte compartments were fed
through an ice bath in order to trap any gas-phase products that
formed during the reaction and would otherwise be missed as
products from the HPLC analysis. The total reaction time was 5 h
with 0.1 mL liquid samples being taken after 5, 15, 45, 100, 160,
230 and 300 min. The solution temperature was monitored before
and after the reaction by a thermometer and remained constant at
room temperature. All data were collected at least in triplicate.
Concentrations were determined by HPLC for liquid-phase reac-
tants, and by gas chromatography (GC, Agilent) for gaseous prod-
ucts.
2
4
1
0 min and then sprayed onto a gas diffusion layer (GDL, 10 BC,
2
Ion Power), obtaining a coated area of 3.5 cm . The air brush
Speedaire) was sonicated and rinsed before spraying each catalyst.
(
Intermediate weighing of the GDL was performed to ensure an
À2
overall catalyst mass of 3.5 mg, for a total loading of 1 mgcm .
2
With an overall GDL size of 15ꢂ45 mm , the coated GDL was con-
Additional characterization details can be found in the Supporting
Information.
tacted at the left space using a copper wire of 1 mm diameter.
Silver paint (SPI) provided adhesion and small contact resistance.
The GDL was covered with Kapton tape (DuPont) on edges, front-,
and backside, leaving only the spray-coated areas uncovered. Only
the sprayed surface area was exposed to the electrolyte to elimi-
nate contamination of the silver paint and copper wire.
The electrochemical experiments were carried out in an H-cell (see
Supporting Information, Figure S1) containing 14 mL of electrolyte
in each compartment and separated by the protonated N117
membrane. In all experiments, a gas diffusion electrode (described
before) was used as a working electrode, with a platinum coil
counter electrode and a Ag/AgCl reference electrode (both BASi).
The working and reference electrodes were housed in the same
compartment while the counter electrode was placed in the other
compartment. Each electrolyte compartment was stirred using
a stir-plate at 1600 rpm. To minimize evaporation, both compart-
ments were capped with rubber stoppers. While a needle provided
pressure compensation for the anolyte, the catholyte exhaust was
led into 14 mL of water to reabsorb any species that left the catho-
lyte. For catalyst screening, prior to experiments, the electrolyte so-
Acknowledgements
The authors acknowledge the contribution of H. Luo, Y. Wang, G.
Veith, M. Orella, and O. Hinrichsen. T.B. acknowledges the Ernest-
Solvay-Foundation and the Elite Network of Bavaria for financial
support. S.H. acknowledges financial support by the National Sci-
ence Foundation Graduate Student Fellowship and CAREER
Award (1122374).
Keywords: electrocatalysis
·
heterogeneous catalysis
·
hydrogenation · nickel · renewable resources
[
[
lution was purged with N for 15 min to remove dissolved oxygen.
[4] Encyclopedia of Applied Electrochemistry (Eds.: G. Kreysa, K. Ota, R. F. Sa-
2
A 0.5m solution of Na SO4 (>99% purity, anhydrous, Sigma Al-
2
drich) at pH 2 was used as the catholyte while 0.5m Na SO4 at
2
[
pH 0.6 was used as the anolyte. Solution pH was adjusted using
[
sulfuric acid (H SO , 95–98% purity, Sigma–Aldrich). 51.35 mL of hy-
2
4
droxyacetone (95% purity, pH~3.5, Alfa Aesar) was added to
4.2 mL of catholyte to obtain a 50 mm solution of hydroxyace-
1
tone. When not in use, the hydroxyacetone was stored in a lab re-
frigerator (1–78C, Marvel 6FRF). Minimal amounts of hydroxyace-
tone dimer, formic, and acetic acid were found in the stock and
&
ChemSusChem 2016, 9, 1 – 8
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Carroll, Kyler J.
