Journal of CO₂ Utilization 20 (2017) 208–217

Contents lists available at ScienceDirect

Journal of CO2 Utilization

journal homepage: www.elsevier.com/locate/jcou

Electrochemical conversion of CO2 to formic acid utilizing Sustainion™ MARK

membranes
⁎
Hongzhou Yang, Jerry J. Kaczur , Syed Dawar Sajjad, Richard I. Masel
Dioxide Materials Inc., 3998 FAU Blvd, Suite 300, Boca Raton, FL 33431, United States

A R T I C L E I N F O A B S T R A C T

Keywords: Formic acid generated from CO2 has been proposed both as a key intermediate renewable chemical feedstock as
Electrochemical well as a potential chemical-based energy storage media for hydrogen. In this paper, we describe a novel three-
Formic acid compartment electrochemical cell configuration with the capability of directly producing a pure formic acid
Anion exchange membrane product in the concentration range of 5–20 wt% at high current densities and Faradaic yields. The
CO2 reduction
electrochemical cell employs a Dioxide Materials Sustainion™ anion exchange membrane and a nanoparticle
CO2 utilization
Sn GDE cathode containing an imidazole ionomer, allowing for improved CO2 electrochemical reduction
performance. Stable electrochemical cell performance for more than 500 h was experimentally demonstrated at
a current density of 140 mA cm−2 at a cell voltage of only 3.5 V. Future work will include cell scale-up and
increasing cell Faradaic performance using selected electrocatalysts and membranes.

1. Introduction cathodic hydrogen generation reaction can dominate at low pH,

significantly lowering the cell formic acid/formate Faradaic efficiency.
The concept of CO2 utilization as an alternate means of reducing Because of this limitation, no one has operated an electrochemical
carbon emissions and atmospheric CO2 in place of carbon capture and system generating only formic acid. These systems have always
sequestration (CCS) has been attracting increasing interest in recent produced alkali metal formate salts because of the cell electrolyte
years [1–9]. CO2 conversion into viable feedstocks for chemicals and requirement. A formate salt product from these cells would have to go
fuels offers both a carbon neutral manufacturing process cycle and the through a second step acidification process in order to produce formic
potential for achieving an economically significant reduction in atmo- acid.
spheric CO2 when applied on a global scale using renewable energy Another formic acid/formate operating issue is that better gas
resources, such as solar, wind, nuclear, and tidal technologies. The diffusion electrodes need to be developed. GDEs have the advantage in
focus of this paper is on the experimental work in developing a new that they allow direct contact between gaseous CO2 and the electro-
method of electrochemically converting CO2 directly into a pure chemical catalysts, improving mass transfer and allowing the reduction
aqueous formic acid product, which can be directly utilized as a reactions to occur at significantly higher concentrations of CO2. But the
renewable chemical feedstock. GDEs that have been reported so far have issues with liquid flooding,
The reduction of CO2 to formate/formic acid has been investigated even after hours of operation, so they need improvement.
by many researchers over the past 30 years and is summarized in several In this paper, we discuss the development of two key improvements
excellent review papers [10–13]. The research can be divided into that can lower the barriers to commercialization. First, we have
studies involving metal electrodes and metal catalysts in the form of flat modified the GDE structure that is commonly used, by adding an anion
plates, high surface area electrode materials, and metal organic catalysts exchange membrane, allowing the operation of a GDE cathode without
[14–42], and in the implementation of gas diffusion electrodes (GDEs) liquid flooding. Second, we modified the cell design so that the cathode
[43–46] operating in both aqueous and non-aqueous electrolytes. is held at a pH of 7–11 so that hydrogen formation is suppressed, and
Presently, though, several limitations need to be overcome before employed an acidic center compartment where the resultant product
the process can be commercialized. It has generally been found that the formate stream is acidic, so we can directly produce formic acid rather
Faradaic efficiency of CO2 conversion to formate/formic acid is than a formate salt product. Only deionized water water is used in the
significantly lower at lower aqueous pH conditions (pH 3–5) than at cell design, and not requiring any salt additions as electrolytes.
higher pH conditions (pH 7–11). The reason is that the competing

⁎
Corresponding author.
E-mail address: [email protected] (J.J. Kaczur).

http://dx.doi.org/10.1016/j.jcou.2017.04.011
Received 3 February 2017; Received in revised form 25 April 2017; Accepted 26 April 2017
Available online 06 June 2017
2212-9820/ © 2017 Elsevier Ltd. All rights reserved.
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

Fig. 1. Formic acid 3-compartment cell configuration. CO2 reduction is facilitated by the presence of an imidazolium ionomer present in the nanoparticle Sn catalyst at the GDE cathode.

2. Experimental methods supporting salt electrolytes, only deionized water is used for both the
anode compartment and center flow compartment.
Key to the work was the development of a novel 3-compartment
formic acid cell design. The anode compartment operates on a DI water 2.2. Three-compartment formic acid cell test system and isometric cell
feed, the GDE cathode operates in a pH range of 7–11, and the formic drawing
acid stream in the center compartment operates at a pH of 1–5 with an
input stream of DI water. The center compartment is bounded by a Fig. 2 shows the formic acid cell experimental test system set-up.
cation membrane on the anode compartment side and by an anion Fig. 3 shows an isometric drawing of the formic acid cell assembly.
membrane on the cathode compartment side. These two membranes
adjoin the cell center compartment without adding a significant voltage 2.3. Equipment and chemicals
resistance. Details on the cell construction and configurations are
detailed below. 2.3.1. Equipment
BK Precision 9121A power supply (B & K Precision Corporation);
2.1. Three-compartment formic acid cell configuration PHD 2000 Infusion syringe pump (Harvard Apparatus); NE-9000
peristaltic pump (New Era Pump Systems, Inc.); Dakota Data
The 3-compartment formic acid cell configuration detail is shown in Instruments GC Series CO2 mass flow controller, 0–50 mL/min;
Fig. 1. Table 1 provides a summary detail of the various anode, cathode, Oakton pH 700 pH meter (Oakton Instruments); Oakton Cond 6+
and center flow compartment configurations that were evaluated. The Handheld Conductivity meter (Oakton Instruments); GC: HP 6890 Plus
formic acid cell consisted of three compartments: a cathode compart- + GC (Agilent); modified fuel cell hardware (Fuel Cell Store, 5 cm2)
ment where the cathode operates at a pH of 7–11 using a Sustainion™ was used as the electrochemical cell. A GDE cathode was prepared by
anionic ionomer, a center flow compartment that is held at a pH spraying a Sn catalyst layer (99.9%, 60–80 nm, US Research
between 1 and 5 using strongly acidic ion-exchange bead electrolyte Nanomaterials, Inc.) containing 5% carbon nanotubes (Sigma-Aldrich,
media, and an anode compartment that is even more acidic using a zero multi-walled carbon nanotubes, > 90% carbon, Item 659258) sus-
gap spacing with the anode where a Nafion membrane ionomer serves pended in an isopropanol/water mixture onto a 50% PTFE proofing
as the electrolyte. CO2 gas was introduced into the cathode compart- Toray paper (TGP-H-120 50%, Fuel Cell Earth) containing 2 wt%
ment and flowing into the GDE cathode catalyst structure, where CO2 Sustainion™ ionomer. The anode was either a GDE electrode prepared
reacts with water coming from the center compartment through the by spraying a nanoparticle IrO2 catalyst (99.99%, Alfa Aesar) onto 5%
anion membrane to the GDE catalyst/membrane interface to form PTFE proofing Toray paper (TGP-H-120 5%, Fuel Cell Earth) or a
formate ions. The generated formate ions ionically pass through the titanium anode expanded metal screen having an IrO2-based catalyst
anion exchange membrane into the center flow compartment. The coating supplied by Water Star, Inc. (Parkman, OH).
reaction in the anode compartment is the oxidation of water into
protons (H+) and oxygen, with the protons passing through a PFSA 2.3.2. Chemicals and materials
(perfluorinated sulfonic acid) cation exchange membrane into the Deionized water (DI water) (18.2 MΩ cm−1, from Milli-Q
center flow compartment. The protons and formate ions combine in water system); formic acid (reagent grade > 95% Sigma–Aldrich);
the center flow compartment to produce formic acid. Deionized water is sodium hydroxide (standardized 0.1 N, LabChem Inc. for titrations);
passed into the center flow compartment to remove the formic acid 1% phenolphthalein solution (LabChem); Amberlite® IR120 (hydrogen
product from the cell. The formic acid cell design does not require any form, Sigma–Aldrich); Dowex® 50WX2 (hydrogen form,

209
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

Table 1
Formic acid cell assembly configuration summary.

Cathode configurations
Cathode POCO Graphite block with serpentine flow path
GDE cathode Toray paper, 50% PTFE proofing
GDE catalyst Sn Nanoparticles (99.9%, 60–80 nm, US Research Nanomaterials, Inc.), catalyst layer air atomized onto GDE cathode with varying amounts of PTFE
suspension and 5% imidazolium-based ionomer, and 5% carbon nanotubes.
CO2 flowrate 20 mL/min, not humidified for ambient temperature operation
Anode configurations
Anode 1 POCO Graphite block with serpentine flow path. The anode was a GDE electrode with a 2 mg cm−2 IrO2 catalyst (99.99%, Alfa Aesar) onto 5% PTFE
proofing Toray paper
Anode 2 Titanium Grade 2 anode block with serpentine flow path. Anode screen having IrO2-based metal oxide coating on expanded titanium (Water Star)
was spot welded onto the titanium anode block
Center flow compartment
Ion exchange media Amberlite® IR120 strong acid ion exchange resin fill, 620–830 μm beads, and other selected ion exchange resins
Center compartment 1 mm thickness, 2.25 cm × 2.25 cm, 0.50 mL (empty), estimated 40% void volume, i.e., 0.2 mL free volume with ion exchange resin
Center compartment frame Polycarbonate, machined flow path for solution flow in/out
DI water feed Selected rate of 0.03–1 mL/min into center flow compartment

Anolyte recirculation rate: DI water, 8–12 mL/min

Anion membrane
Dioxide Materials Sustainion™ X37 imidazolium-based anion exchange membrane:
60–80 micron thickness (wet), ion exchange capacity of 1.05 mEq/g (950 EW), membrane conductivity in water, through plane, of about 60–70 mS cm−1
Cation membranes
DuPont Nafion® Membranes Tested: 212, 115, 324
Cell active membrane geometric area: 5 cm2
Cell operating temperature: Ambient, 20–25 °C

Sigma–Aldrich); Dow Amberlite® IRN-77 (ion exchange resin, nuclear flow compartment is bounded by a cation exchange on the anode side
grade, VWR Scientific); Duolite® Resin C433 (Alfa Aesar); Dupont and an anion exchange membrane on the cathode side. Fig. 4 shows the
Nafion® 115, 212, and 324 membranes (Ion Power Inc.). Liquid CO2 proposed main and secondary reactions as well as ion transport that
(NuCO2, Stuart, FL) was used as the gas supply and vaporized for use in may occur during cell operation.
the lab.

2.4.1. Main reaction

2.3.3. Analytical The electrochemical reduction of CO2 starts when CO2 reacts with
Formic acid concentrations were determined by titration using a water on the cathode forming formate (HCOO−) and hydroxide (OH−)
0.1 N standardized NaOH reagent, phenolphthalein indicator, and a ions:
Class A 50 mL buret. Formic acid sample size weights/volumes for the
titration were sized large enough to obtain titration volumes of about CO2 + H2O + 2e− → HCOO− + OH− (1)
20 mL or more, so as to minimize the error in the formic acid Simultaneously, the oxidation of water occurs at the anode, forming
concentrations to less than 1%. oxygen gas and protons (H+):

2H2O → 4H+ + 4e− + O2 (2)

2.4. Formic acid cell compartment reactions and ion transport
Both the formate ions and hydroxide ions migrate through the anion
The formic acid cell has a complex set of reactions that occur at the exchange membrane into the center flow compartment where they
anode, the GDE cathode, and the center flow compartment. The center meet the protons produced from the anode compartment which pass

Fig. 2. Formic acid cell experimental test system set-up.

210
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

Fig. 3. Formic acid cell exploded isometric assembly drawing.

Fig. 4. Formic acid cell configuration showing proposed electrochemical reactions and ion transport.

through the cation membrane to yield formic acid and water: Formic acid losses by oxidation at the anode:

H+ + OH− → H2O (3) 2HCOOH + O2 → 2CO2 + 2H2O (9)

H+ + HCOO− → HCOOH (4)

2.5. Center flow compartment conductive media selection

2.4.2. Side reactions The 3-compartment formic acid cell design requires a thin center
Side reactions that can occur on the cathode: flow compartment spacing and a high-conductivity electrolyte to
minimize the cell IR drop. The conductivity of aqueous formic acid
CO2 + H2O + 2e− → CO + 2OH− (5) alone, without any added acid or anion salt carrier such as sulfate or
bicarbonate, is extremely low, only being between 5 and 10 mS cm−1
2H2O + 2e− → H2 + 2OH− (6) for formic acid concentrations ranging between 5 and 20 wt% (see Fig.
S4). A higher-conductivity electrolyte was required to minimize the IR
CO2 + OH− → HCO3− (7) resistance across the thin center flow compartment. We decided on
using an ion exchange resin media fill to improve the electrolyte/
The bicarbonate ion (HCO3−)
can move through the anion exchange solution conductivity.
membrane where it reacts with protons produced on the anode to The use of an ion exchange media to improve cell voltage and
produce CO2: performance in electrochemical cells has been successfully employed in
electrodeionization [47–52], electrodialysis [53,54], and other electro-
H+ + HCO3− → CO2 + H2O (8)
chemical systems [55,56].

211
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

80 2.7. Sustainion™ anion exchange membrane

70
The Sustainion™ membrane used in the formic acid cell is based on
60 the development of a series of new proprietary anion exchange
membranes. These membranes have a stable polystyrene backbone that
50
has been functionalized with an imidazolium group, resulting in a
40 Amberlite IR120 unique membrane that is both very conductive and alkaline stable.
Amberlite IRN77 Details of the synthesis and membrane are extensively discussed in
30 Dowex 50WX2 recently issued patents and publications [55–62]. The electrochemical
Duolite C433
20 CO2 reduction reaction, occurring with the tin catalyst on the cathode,
Aqueous Formic Acid
is theorized to be facilitated by the addition of an imidazole-based
10 ionomer incorporated into the cathode catalyst layer. The formation of
a CO2-imidazolium complex is proposed to lower the free energy
0
0 5 10 15 20 25 required for the reduction reaction [63]. In addition, the Sustainion™
membrane at the membrane-cathode catalyst interface may also be
involved in the reduction reactions. The Sustainion™ membranes are
Fig. 5. Graphic plot of the conductivity of hydrogen form ion exchange resin slurries also being used in other electrochemical cell applications, such as the
saturated with water and various formic acid concentration solutions. reduction of CO2 to CO and in water electrolysis cells, producing H2 and
O2, where they allow electrochemical operation at very high current
densities at low voltages [62–65].

In order to determine suitable ion exchange resin media for the cell,
3. Results and discussion
the ionic conductivity of four types of ion exchange resin media were
experimentally measured in water and in 5, 10, 15 and 20 wt% formic
3.1. Single-pass operation cell performance
acid solutions. The hydrogen-form resins (as received) were saturated
with the various solutions and then carefully packed into the space
The formic acid cell was operated in two different modes, namely, a
between the conductivity probe and the sensor probe plastic shield. The
single-pass mode and a center compartment recirculation mode. In
Oakton Conductivity instrument has a plastic shield that has to be in
single-pass operation mode, DI water was directly metered into the
place to correctly measure conductivity (see Fig.S6 in the Supporting
center compartment using a syringe pump operating at selected flow
Information to see the conductivity sensor and its plastic cover shield).
rates. In a standard formic acid cell configuration using a graphite
The various ion exchange resin/saturated formic acid solution media
anode flow field and Nafion® 115 membrane, the cell performance
mixture conductivities were then individually measured. The measured
results are shown in Figs. 6 and 7 . The calculated residence time in the
ionic conductivities of the aqueous resin media mixtures are detailed in
center compartment ranged from about 4 min at the 0.05 mL flowrate
Table S2 (Supporting Information) and are summarized graphically in
and about 2 min at the 0.10 mL DI water flowrate using a ion exchange
Fig. 5.
void volume value of about 40%.
The use of cation ion exchange resin media provided a substantial
In Fig. 6, we can see that the cell voltage increased with increasing
conductivity increase over aqueous formic acid alone. The ion exchange
current density, operating at a voltage of 3.15 V at 100 mA cm−2 and
resin/formic acid matrix mixtures showed an increase in ionic con-
3.75 V at a current density of 200 mA cm−2. The high current density
ductivity over formic acid solutions alone by a factor of about 6–10. The
obtained at such relatively low cell voltages was largely attributed to
hydrogen-form sulfonic acid resins, such as the Amberlite® IR120 and
utilizing a GDE cathode and Sustainion™ anion membrane in the novel
the Dowex® 50WX2, provided the highest ionic conductivities.
cell configuration described previously.
Amberlite® IR120 was chosen as the primary center flow compartment
It was observed that the flow rate of DI water into the central
ion exchange material for testing in the formic acid cell center flow
compartment significantly affected the formic acid cell performance in
compartment in our experimental runs. Dowex® 50WX2, which has a
terms of formic acid product concentration and Faradaic efficiency.
very fine median particle size range of 100–200 mesh (75–150 μm)
Fig. 7 shows the formic acid concentration and Faradaic efficiency as a
demonstrated a very high pressure drop in the center flow compartment
function of the center compartment flow rate in a typical single-pass
when evaluated, and was not further utilized in the experiments.
operation mode having a graphite anode flow field, IrO2-coated GDE

2.6. Cation ion exchange membrane selection

The 3-compartment formic acid cell design required the selection of

a suitable cation ion exchange membrane. The fundamental require-
ments for the cation exchange membrane are that it have a high ionic
conductivity, resistance to oxidation reactions at the anode, and ability
to block or limit the transport of formic acid/formate ions into the
anode compartment. Typical low-cost hydrocarbon based cation mem-
branes, such as those employed in electrodialysis systems, such as those
produced by Tokuyama Soda under the Neosepta® trade name as well as
other manufacturers, are not resistant to oxidation. The only type of
membranes that are both highly conductive and have a high oxidation
resistance are the PFSA based membranes, such as those sold by DuPont
under the Nafion® trademark as well as other brands sold by other
various manufacturers worldwide. PFSA membranes are available in Fig. 6. Formic acid cell current density versus voltage profile for a typical cell
different thicknesses, equivalent weights (EW), and with/without an configuration operated at different current densities in a single-pass operation mode into
internal fabric/mesh reinforcement. the center compartment at a DI water flow rate of 0.1 mL/min.

212
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

1
100 Table 3
14 First formic acid extended run electrolyzer configuration.
90
80 12 Anode

Formic Acid Conc. / wt %

Faradaic ĸciency / %

GDE, IrO2 loading: 5 mg cm−2 on Toray paper, POCO graphite block with
70 serpentine flow path
10
60 Anolyte: Deionized water
8 Cathode
50 Faradaic ĸcciency GDE, nanoparticle Sn, Sn loading: 5 mg/cm2 with 5% PTFE, 5% carbon
nanotubes, and 5% imidazolium-based ionomer
40 Formic Acid Concentra
C n 6
CO2 gas flow rate: 20 mL/min
30 Center flow compartment
4 Resin fill: Amberlite® IR120
20 DI water single-pass mode input flow rate: 0.03 mL/min
2 Anion membrane
10
Dioxide Materials Sustainion™ X37 imidazolium-based anion membrane
0 0 Cation membrane
0 0.05 0.1 0.15 0.2 0.25
0 0.3 0.35 0.4 0.45 0.5 0.55 DuPont Nafion® 212
Cell active membrane projected area: 5 cm2
Flow Rate / mL min-11 Cell operating temperature: Ambient, 20–25 °C
Cell current density: 140 mA cm−2
Fig. 7. Faradaic efficiency and formic acid concentration as a function of center
compartment flow rate in single-pass operation mode with graphite anode flow field,
IrO2 coated GDE anode, and Nafion® 115 membrane. Error bars show the range of results
at the various flow rates. acid concentrations used in the starting solutions. For example, an 85%
Faradaic efficiency was obtained by recirculating an initial volume of
DI water solution through the center compartment. The solution
anode, Sn GDE cathode and Nafion® 115 membrane. At low flowrates, increased to a 2.3 wt% formic acid concentration after 2 h. The
e.g. 0.05 mL/min, the cell produced about 9.7 wt% formic acid at a Faradaic efficiency dramatically dropped to 47% when a starting
47% Faradaic efficiency. The Faradaic efficiency improved with 9.9 wt% formic acid solution was recirculated through the center
increasing flowrate, increasing to 88% at 0.5 mL/min. However, the compartment. The cell performance using a recirculation mode opera-
formic acid product concentration was only about 2 wt% at this tion through the center compartment was found to be significantly
flowrate as shown in Fig. 7. A number of the experimental runs, more lower as compared to single-pass operation mode in obtaining the same
than 20, were conducted at the 0.05 mL/min and 0.10 mL/min flow formic acid product concentration.
rates to make comparisons with the other factors that we investigated in
potentially improving the performance the formic acid cell. The runs 3.3. Formic acid cell extended time performance runs
were conducted for varying time periods ranging from 8 h to 120 h and
longer. These factors included anode and cathode modifications, A first extended formic acid cell run was conducted to determine
selection of ion exchange media employed in the center compartment, cell voltage and performance stability over time. The detailed cell
and the evaluation of various Sustainion™ anion exchange membrane configuration is given in Table 3. The cell was continuously operated in
formulations. a single-pass operation mode using deionized water for a total of about
Operationally, as the formic acid concentration in the cell center 550 h. The test results are shown in Fig. 8.
compartment increased, an increasingly larger fraction of the formic The formic acid cell voltage was found to be surprisingly stable,
acid product might cross over into the cell anolyte compartment operating in a range of 3.3–3.4 V over 550 h of operation. The formic
through the cation membrane where formic acid may be lost via acid concentration gradually decreased with time, starting at an 18 wt
reaction (8). Faradaic formic acid losses due to formic acid crossover % concentration and decreasing to 15 wt% after 550 h. The formic acid
and decomposition factors are further discussed in Section 3.5. cell average Faradaic efficiency was determined to be only about 30%
during the run. The cell performance agreed with the trend in Fig. 8,
3.2. Recirculated center flow compartment mode operation cell performance where the cell operates at lower Faradaic efficiencies when producing a
very high formic acid product concentration at specified flow rates in
Cell performance using a recirculated mode operation, where single-pass operation mode. The formic acid cell, on disassembly, was
various concentrations of formic acid were circulated through the
center flow compartment for a period of time, was evaluated.
Typically, formic acid concentration in the recirculated solution
increased with time since there was no DI water added to the
recirculation stream. However, the concentration increase slowed after
several hours of operation, indicating a lower Faradaic efficiency. To
further study the relationship between formic acid concentration and
Faradaic efficiency, several formic acid solutions with different con-
centrations were used as starting solutions for the cell. The cell was then
operated for a period of 2 h, and the final formic acid concentrations
were determined by titration in calculating the Faradaic efficiency. As
shown in Table 2, the Faradaic efficiency decreased with higher formic

Table 2
Effect of formic acid concentration on Faradaic efficiency in the recirculated center flow
compartment mode operation.

Starting formic acid concentration (wt%) 0 9.9 19.5

Final formic acid concentration (wt%) 2.3 11.2 20.4 Fig. 8. First extended run formic acid cell performance using graphite flow field,
Faradaic efficiency (%) 85 47 32 nanoparticle Sn cathode GDE, Nafion® 212 cation membrane, and IrO2 coated GDE

213
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

Table 4
Second formic acid cell extended time run electrolyzer configuration.

Anode
Titanium Grade 2 anode block with serpentine flow path. Anode screen having
IrO2-based metal oxide coating on expanded titanium (Water Star) was spot
welded onto the titanium anode block
Anolyte: Deionized water
Cathode
GDE, nanoparticle Sn, Sn loading: 5 mg cm−2
5 wt% PTFE, 5 wt% carbon nanotubes,, 5 wt% imidazolium-based ionomer
CO2 gas flow rate: 20 mL/min
Anion membrane
Dioxide Materials Sustainion™ X37 imidazolium-based anion membrane
Cation membrane
DuPont Nafion® 324
Center flow compartment
Resin fill: Amberlite® IR120
DI water single-pass mode input flow rate: 0.10 mL/min
Cell operating temperature: Ambient, 20–25 °C
Cell active membrane projected area: 5 cm2
Cell current density: 140 mA cm−2 Fig. 9. Second extended run cell performance using a titanium flow field, IrO2 coated
titanium mesh anode, and Nafion® 324 membrane.

found to have significant oxidation erosion in several locations of both

the graphite anode graphite flow field and GDE. showed only small traces of formic acid, but this would not be a
A big potential Faradaic efficiency loss factor in the formic acid cell definitive determination of formic acid transport into the anode
is the significant transport of formate ions into the anode compartment compartment since all of the transported formate could be oxidized
through the cation membrane. The effect of cation membrane selection immediately at the anode. A better method would be to determine the
is discussed in Section 3.4. amount of CO2 produced by the oxidation of formate ions that would be
A second extended time cell run was conducted, but instead using a present in the anolyte oxygen product gas stream. Table 6 shows the
Ti metal anode flow field and an IrO2-based coating on expanded data on the area ratio of CO2 and O2 found in the anolyte gas by GC
titanium which was spot welded onto the Ti anode flow field. This with the cells operating at formic acid product concentrations and using
anode configuration would negate any anodic oxidation issues related various cation membranes selected in the tests.
to the use of a graphite flow field/carbon paper based GDE as anode From the data, the thickness of the Nafion PFSA type cation
substrates when generating oxygen. A Nafion® 324 membrane was also membranes was found to be a critically important factor in minimizing
evaluated in the testing in looking at reducing formic acid crossover as the transport of formate ions through the membrane into the cell
discussed in Section 3.4. anolyte compartment. The thinner Nafion® 212 was initially chosen in
The detailed results of the revised formic acid cell configuration and our experimental runs (the 500 h extended run) to obtain lower cell
cell performance are presented in Tables 4 and 5 and in Fig. 9. After voltages, but this data showed that the transport of formic acid through
142 h of operation, the cell voltage increased from 3.3 V to 3.7 V. a thicker membrane was significantly reduced, and thus an important
Surprisingly, the concentration of the formic acid product increased factor in achieving improved formic acid cell Faradaic performance.
with time, starting at a concentration of 8.1 wt% concentration and Nafion® 324, which is a bilayer PFSA membrane having an 1100 EW
increasing to 9.4 wt% after 142 h of operation. and a 1500 EW layer, showed the lowest transport of formate ion into
The improvement in cell performance is also reflected by the the anolyte, and would be the preferred cation membrane for use in the
improved Faradaic efficiency as shown in Table 5. The cell Faradaic formic acid cell.
efficiency started at 80.5% and increased to 94.0% at the end of the cell The permeation rate of formic acid based on Nafion® membrane
run. On disassembly, the cell anode indicated no anode or cathode thickness has been previously reported [66]. Nafion® 112 showed a 2–3
deterioration. The significant improvement in the cell formic acid times higher formic acid flux rate than Nafion® 117 (178 μm thickness)
Faradaic performance may be the combined outcome of using both in a two compartment permeation cell having formic acid on one side of
the Nafion® 324 membrane and the IrO2-based titanium anode on the Ti the membrane and deionized water in the other compartment. The
flow field. permeation cell was not used in an operating electrochemical cell. The
amount of the swelling of the Nafion® ionomer at high concentrations of
3.4. Formate ion transport into anode compartment formic acid was cited as a factor in the formic acid permeation or flux
rate.
As mentioned above in Sections 3.1 and 3.3, the transport of
formate ions into the anode compartment through the cation exchange Table 6
Summary of GC analysis results of the formic acid cell anolyte gases using different cation
membrane may be a significant potential source of Faradaic efficiency
membranes when operating the center flow compartment at various formic acid
loss in the formic acid cell. concentrations.
In order to determine the amount of formate ion transport through
the cation exchange membrane into the anolyte compartment, with Membrane Membrane thickness Formic acid concentration ACO2 /AO2
subsequent oxidation at the anode, two tests were conducted. Analysis (μm) (wt%)

of the recirculating anolyte solution during a number of the runs DuPont Nafion® 50.8 15 2.3
212
Table 5 DuPont Nafion® 127 5 0.29
Faradaic efficiency of 142 h extended cell run with titanium flow field and IrO2 coated 115 8.7 0.31
titanium mesh anode and Nafion® 324 membrane. 15.6 0.56
16.8 0.70
Cell operation in hours 3.4 25.5 47.2 119.6 142 DuPont Nafion® 150 8.5 0.03
Faradaic efficiency 80.5% 81.0% 82.0% 91.8% 94.0% 324

214
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

We propose that the dramatically reduced formic acid transport that occurs during acid decomposition of formic acid, for example when
we found using the Nafion® 324 was due to the ionomer properties of concentrated sulfuric acid is added to formic acid as a well-known
the Nafion® 1500 EW membrane layer. This membrane polymer has a laboratory method of generating CO.
much lower water content than the other Nafion® membranes that we Other potential sources of CO and H2, may be from the cathode
evaluated and would have a reduced number of sulfonic acid functional compartment, as dissolved aqueous side-product species. The very low
groups per gram of ionomer. This ionomer environment should slow solubilities of CO (0.03 g/kg of water) and H2 (0.0016 g/kg water) [69]
down the transport of the negatively charged formate anions. In make this unlikely, however. Gas permeation through the anion
addition, as is well known, formic acid at higher concentrations forms membrane from the catholyte reaction was also unlikely so long as
formate–formate dimers, which can form neutral charged molecules. the anion membrane was intact, having with no physical gas permea-
Neutral molecules can easily pass through membranes. The formic acid tion.
conductivity versus concentration curve, as shown in Fig. S4 in the The experimental results indicate that there may be some electro-
supporting information, shows a conductivity maxima at about 30 wt% chemical formic acid decomposition reactions that may be occurring,
formic acid. Formic acid concentrations above 30 wt% should show most probably at the cation membrane interface with the center
increasing concentrations amounts of the neutral formic acid dimer compartment ion exchange resin, which is the strongest acid location.
species. A large flux of protons enters the center flow compartment at this
interface. Potential formic acid losses may be about 10% or less based
3.5. Gas formation in center flow compartment on the GC data in Table 7.
The formic acid product from the cell, ranging from 5 to 20 wt% on
Gas bubbles were observed in the formic acid solution product concentration, was found to be stable, showing no signs of gas evolution
stream exiting the cell center flow compartment. One potentially or solution decomposition during storage.
formed gas is CO2, which may be produced by the transport of
carbonate ions (CO3−) or carbonic acid from the cathode compartment
through the anion membrane into the center flow compartment. These 3.6. Formic acid cell compartment voltage potential measurements
carbonate and bicarbonate species should decompose to CO2 in the low
pH environment of the formic acid/strong acid ion exchange media in The operating voltage potentials of the formic acid cell anode,
the center flow compartment. Other gases that may be generated are center compartment, and cathode were measured to determine the
CO and H2 from the potential decomposition of formic acid in the voltage potentials in the cell and the voltage polarization plot for the
center flow compartment. Strong acidic conditions exist in the hydro- cell. A voltage polarization plot for the cell was also conducted and
gen form cation exchange resin at the cation exchange membrane shown in Fig. 10.
interface, where a large flux of protons is entering the center flow The formic acid cell measured potentials were as follows:Cell Open
compartment. Circuit Voltage: 1.80 volts
Table 7 shows the GC analysis of the center flow compartment
Measurements made operating at 140 mA cm−2 current density:
product gas. The data show that CO2 was the predominant gas product,
with small amounts of CO and H2. The total volume of CO2 gas Anode Potential: 1.40 − 1.45 volts
generated was not measured, so the total amount of formic acid
decomposition into H2 or CO could not be quantified.
No gases were noted in or exiting the center flow compartment Cathode Potential: 1.40 − 1.50 volts Center Compartment:
when the current to the cell was shut off when the solution flow was 0.40 − 0.60 volts (voltage drop including both membranes)
continued through the compartment. In addition, no gases were
evolved from any of the formic acid/strong acid ion exchange resin
mixtures that were prepared in the ionic conductivity experiments in Total cell volts: 3.40
Section 2.5.
The decomposition of formic acid, whether by acid or catalytic Total cell resistance by impedance measurements: 800 mΩ.
decomposition routes, may occur by two different pathways [67,68] as The voltage polarization plot for the formic acid cell is shown
in reactions (10) and (11). Fig. 10. The voltage OCP (open circuit potential) was found to be about
1.80 V. There appears to be a mixed cathodic cell reaction occurring at
HCOOH → CO2 + H2
(10) ΔG = −48.4 kJ mol−1 Decarboxylation cell voltage potentials between 1.8 and 3.0 V, at current densities of
0–40 mA cm−2.
HCOOH → CO + H2O ΔG = −28.5 kJ mol−1 Dehydration (11)
4
The small amounts of CO and H2 that were found in the center flow
3.5
compartment off-gas were found in approximately equimolar amounts.
It is well known that various catalysts, such as Pt and Pd, will 3
Cell Voltage / V

decompose formic acid solutions according to reaction (10), producing

2.5
CO2 and H2. The coproduction of some CO indicates that the decom-
position reaction may have an equal selectivity where both products are 2
formed. Decomposition reaction (11), forming CO and water, typically
1.5
Table 7 1
Summary of GC analysis results of gas formation in the center flow compartment.
0.5
Gas products from Peak area from GC analysis AH2 /ACO2 ACO /ACO2
center flow 0
compartment
0 20 40 60 80 100 120 140 160
AH2 ACO ACO2
Current Density / mA cm-2
H2 CO CO2 14.7 19.2 1465.3 0.010 0.013
Fig. 10. Cell voltage polarization plot of the formic acid cell.

215
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

3.7. Formic acid cell operation using a hydrogen-consuming anode Appendix A. Supplementary data

The formic acid cell was configured with a hydrogen-consuming Supplementary data associated with this article can be found, in the
anode to determine the lowest possible cell voltage that could be online version, at http://dx.doi.org/10.1016/j.jcou.2017.04.011.
obtained, in addition to getting further confirmation of the cell
operating cathode potentials. For this purpose, a Pt/C catalyst instead References
of IrO2 catalyst was deposited onto a Nafion® 115 membrane as well as
the GDE anode. Other components were the same as those used in the [1] M.M. Halmann, Chemical Fixation of Carbon Dioxide: Methods for Recycling CO2
formic acid cell assemblies previously described. The cell was operated into Useful Products, CRC Press, Boca Raton, FL, 1993, pp. 67–127.
[2] M. Aresta, Carbon Dioxide as Chemical Feedstock, Wiley-VCH, Weinheim, 2010, pp.
at a voltage of 1.83–1.88 V at a current density of 160 mA cm−2 1–13 ISBN-13 9783527324750.
(800 mA current) continuously for a period of 68 h. The measured [3] M. Aresta, A. Dibenedetto, Utilisation of CO2 as a chemical feedstock: opportunities
cathode potential, when operating the cell at 140 mA cm−2, was and challenges, Dalton Trans. 28 (2007) 2975–2992.
[4] D.T. Whipple, P.J.A. Kenis, Prospects of CO2 utilization via direct heterogeneous
measured to be in the range of −1.38 V to −1.48 V, which was electrochemical reduction, J. Phys. Chem. Lett. 1 (2010) 3451–3458.
consistent with the cell cathode potential measurements detailed in [5] E.A. Quadrelli, G. Centi, J.-L. Duplan, S. Perathoner, Carbon dioxide recycling:
Section 3.6. The formic acid cell, operating in a single-pass operation emerging large-scale technologies with industrial potential, ChemSusChem 4
(2011) 1194–1215.
mode with a water flowrate of 0.10 mL/min, produced a 9.2 wt%
[6] S. Sankaranarayanan, K. Srinivasan, Carbon dioxide – a potential raw material for
formic acid product at a Faradaic efficiency of 81%. the production of fuel, fuel additives and bio-derived chemicals, Indian J. Chem.
51A (September–October) (2012) 1252–1262.
[7] R.I. Masel, R. Ni, Z. Liu, Q. Chen, R. Kutz, L. Nereng, D. Lutz, K. Lewinski, Unlocking
the potential of CO2 conversion to fuels and chemicals as an economically viable
4. Summary route to CCR, Energy Procedia 63 (2014) 7959–7962.
[8] R. Masel, Z. Liu, D. Zhao, Q. Chen, D. Lutz, L. Nereng, CO2 conversion to chemicals
A novel 3-compartment electrochemical cell design with the cap- with emphasis on using renewable energy/resources to drive the conversion, in:
S. Snyder (Ed.), RSC Green Chemistry, No. 43, Commercializing Biobased Products:
ability of generating a pure formic acid product from the reduction of Opportunities, Challenges, Benefits, and Risks, 2016, pp. 215–257, , http://dx.doi.
CO2 is detailed. The cell configuration consisted of an anode compart- org/10.1039/9781782622444-00215 eISBN: 978-1-78262-244-4 (Chapter 10).
ment, a center flow compartment containing a cation exchange resin [9] M. Aresta, A. Dibenedetto, E. Quaranta, State of the art and perspectives in catalytic
process for CO2 conversion into chemicals and fuels: the distinctive contribution of
electrolyte bounded by a PFSA cation exchange membrane on the chemical catalysis and biotechnology, J. Catal. 343 (2016) 2–45.
anode side and a proprietary Sustanion™ anion exchange membrane on [10] M. Scibioh, B. Viswanathan, Electrochemical reduction of carbon dioxide: a status
the cathode side, and an imidazole-nanoparticle Sn catalyst-based GDE report, Proc. Indian Natl. Sci. Acad. 70 (2004) 407–462.
[11] B. Hu, C. Guild, S.L. Suib, Thermal, electrochemical, and photochemical conversion
in the cathode compartment. The Sustainion™ anion membrane used in of CO2 to fuels and value-added products, J. CO2 Util. 1 (2013) 18–27.
the formic acid cell has a stable polystyrene backbone that is functio- [12] X. Lu, D.Y.C. Leung, H. Wang, M.K.H. Leung, J. Xuan, Electrochemical reduction of
nalized with an imidazolium functional group, resulting in a membrane carbon dioxide to formic acid, ChemElectroChem 1 (2014) 836–849.
[13] D. Pletcher, The cathodic reduction of carbon dioxide – what can it realistically
that is both very conductive and alkaline stable. The addition of an
achieve? A mini review, Electrochem. Commun. 61 (2015) 97–101.
imidazole-based ionomer into the nanoparticle Sn based cathode- [14] Y.B. Vassiliev, V.S. Bagotzky, N.V. Osetrova, N.A. Mayorova, Electroduction of
catalyst enabled CO2 reduction reaction at the cathode GDE-anion carbon dioxide. Part 1. The mechanism and kinetics of the electroreduction of CO2
in aqueous solutions on metals with high and moderate hydrogen overvoltages, J.
membrane interface. The electrochemical cell operated at a current
Electroanal. Chem. 189 (1985) 271–294.
density of 140 mA cm−2 range at a cell voltage of only 3.5 V with [15] H. Noda, S. Ikeda, Y. Oda, K. Imai, M. Maeda, K. Ito, Electrochemical reduction of
formic acid Faradaic efficiencies of up to 94%. High formic acid carbon dioxide at various metal electrodes in aqueous potassium hydrogen
Faradaic efficiencies were found to be critically dependent on the carbonate solution, Bull. Chem. Soc. Jpn. 63 (1990) 2459–2462.
[16] M. Todoroki, K. Hara, A. Kudo, T. Sakata, Electrochemical reduction of high
selection of the cation membrane utilized on the anolyte compartment pressure CO2 at Pb, Hg and In electrodes in an aqueous KHCO3 solution, J.
side of the cell. Thicker Nafion® membranes were effective in minimiz- Electroanal. Chem. 394 (1995) 199–203.
ing crossover formate ion losses into the anode compartment. Nafion® [17] M. Azuma, Electrochemical reduction of carbon dioxide on various metal electrodes
in low-temperature aqueous KHCO3 media, J. Electrochem. Soc. 137 (1990)
324 was found to provide the lowest formate/formic acid crossover into 1772–1776.
the anolyte compartment. Two modes of cell operation, a center [18] Y.Y. Hori, H. Wakebe, T. Tsukamoto, O. Koga, Electrocatalytic process of CO
compartment single-pass versus recirculation modes were investigated, selectivity in electrochemical reduction of CO2 at metal electrodes in aqueous
media, Electrochim. Acta 39 (1994) 1833–1839.
with the single pass mode being preferred. The novel formic acid cell [19] M. Jitaru, D.A. Lowy, M. Toma, B.C. Toma, L. Oniciu, Electrochemical reduction of
design indicates a potential route for future commercial use of formic carbon dioxide on flat metallic cathodes, J. Appl. Electrochem. 27 (1997) 875–889.
acid as a sustainable chemical feedstock for generating downstream [20] M. Jitaru, Electrochemical carbon dioxide reduction – fundamental and applied
topics (review), J. Univ. Chem. Tech. Metall. 42 (4) (2007) 333–344.
chemicals, as well being a viable chemical-based energy storage [21] Y. Hori, Electrochemical CO2 reduction on metal electrodes, Mod. Asp.
medium in hydrogen storage/generation. Electrochem. 42 (2008) 89–189.
[22] H. Li, C. Oloman, Development of a continuous reactor for the electro-reduction of
carbon dioxide to formate. Part 1. Process variables, J. Appl. Electrochem. 36
(2006) 1105–1115.
Acknowledgements [23] C. Oloman, H. Li, Electrochemical processing of carbon dioxide, ChemSusChem 1
(2008) 385–391.
Parts of this work were supported by DOE under contract DE- [24] H. Li, C. Oloman, Development of a continuous reactor for the electro-reduction of
carbon dioxide to formate. Part 2. Scaleup, J. Appl. Electrochem. 37 (2007)
SC0004453 and by 3M Company. The opinions here are those of the 1107–1117.
authors and may not reflect the opinions of DOE or 3M. Assistance from [25] S.R. Narayanan, B. Haines, T.I. Valdez, Electrochemical conversion of carbon
colleagues, collaborators and friends from 3M are gratefully acknowl- dioxide to formate in alkaline polymer electrolyte membrane cells, J. Electrochem.
Soc. 158 (2) (2011) A167–A173.
edged. [26] M. Alvarez-Guerra, S. Quintanilla, A. Irabien, Conversion of carbon dioxide into
The authors filed a U.S. patent application on the electrochemical formate using a continuous electrochemical reduction process in a lead cathode,
process and designs discussed here, which was recently published [70]. Chem. Eng. J. 207 (2012) 278–284.
[27] M. Alvarez-Guerra, A. Del Castillo, A. Irabien, Continuous electrochemical reduc-
The authors all have a financial stake in the outcome on the sale of the tion of carbon dioxide into formate using a tin cathode: comparison with lead
Sustainion™ membranes discussed. cathode, Chem. Eng. Res. Des. 92 (4) (2014) 692–701.
Special thanks to Dioxide Materials personnel Brandon Robinson, [28] B. Innocent, D. Liaigre, D. Pasquier, F. Ropital, J. Léger, K.B. Kokoh, Electro-
reduction of carbon dioxide to formate on lead electrode in aqueous medium, J.
Daniel Barba, and Broc Dunlap in the preparation and fabrication of cell Appl. Electrochem. 39 (2009) 227–232.
materials for the test cells. [29] B. Innocent, D. Pasquier, F. Ropital, F. Hahn, J.M. Léger, K.B. Kokoh, FTIR
spectroscopic study of the reduction of carbon dioxide on lead electrode in aqueous

216
H. Yang et al. Journal of CO₂ Utilization 20 (2017) 208–217

medium, Appl. Catal. B: Environ. 94 (2010) 219–224. Electrodeionization with ion-exchange textile for the production of high resistivity
[30] D.T. Whipple, E.C. Finke, P.J.A. Kenis, Microfluidic reactor for the electrochemical water: influence of the nature of the textile, Desalination 114 (1997) 165–173.
reduction of carbon dioxide: the effect of pH, Electrochem. Solid-State Lett. 13 (9) [49] E. DeJean, J. Sandeaux, R. Sandeaux, C. Gavach, Water demineralization by
(2010) B109–B111. electrodeionization with ion exchange textiles. Comparison with conventional
[31] Z.M. Detweiler, J.L. White, S.L. Bernasek, A.B. Bocarsly, Anodized indium metal electrodialysis, Sep. Sci. Technol. 33 (6) (1998) 801–818.
electrodes for enhanced carbon dioxide reduction in aqueous electrolyte, Langmuir [50] A. Mahmoud, L. Muhr, G. Grévillot, G. Valetin, F. Lapique, Ohmic drops in the ion-
30 (2014) 7593–7600. exchange bed of cationic electrodeionisation cells, J. Appl. Electrochem. 36 (2006)
[32] E. Irtem, T. Andreu, A. Parra, M.D. Hernández-Alonzo, S. Garcia-Rodriguez, 277–285.
J.M. Riesco-Garcia, G. Penelas-Pérez, J.R. Morante, Low-energy formate production [51] D.V. Golubenko, J. Kŕivčik, A.B. Yaroslavtsev, Evaluating the effectiveness of ion
from CO2 electroreduction using electrodeposited tin on GDE, J. Mater. Chem. A 4 exchangers for the electrodeionization process, Pet. Chem. 55 (10) (2015) 769–775.
(2016) 13582–13588. [52] E. Korngold, L. Aronov, O. Kedem, Novel ion-exchange spacer for improving
[33] A.S. Agarwal, Y. Zhai, D. Hill, N. Sridhar, The electrochemical reduction of carbon electrodialysis I. Reacted spacer, J. Membr. Sci. 138 (1998) 165–170.
dioxide to formate/formic acid: engineering and economic feasibility, [53] K. Zhang, M. Wang, D. Wang, C. Gao, The energy-saving production of tartaric acid
ChemSusChem 4 (2011) 1301–1310. using ion exchange resin-filling bipolar membrane electrodialysis, J. Membr. Sci.
[34] J. Wu, F.G. Risalvato, F.-S. Ke, P.J. Pellechia, X.-D. Zhou, Electrochemical reduction 341 (2009) 246–251.
of carbon dioxide I. Effects of the electrolyte on the selectivity and activity with Sn [54] B. Zhang, H. Gao, Y. Chen, Enhanced ionic conductivity and power generation using
electrode, J. Electrochem. Soc. 159 (7) (2012) F353–F359. ion-exchange resin beads in a reverse-electrodialysis stack, Environ. Sci. Technol.
[35] J. Wu, F.G. Risalvato, F.-S. Ke, P.J. Pellechia, X.-D. Zhou, Electrochemical reduction 49 (2015) 14717–14724.
of carbon dioxide II. Design, assembly, and performance of low temperature full [55] J.J. Kaczur, D.W. Cawlfield. Electrochemical process for producing chlorine dioxide
electrochemical cells, J. Electrochem. Soc. 160 (9) (2012) F953–F957. solutions from chlorites. U.S. Patent 5,092,970 (1992).
[36] L.R. Zhang, P. Gao, L. Lei, Studies on the Faradaic efficiency for electrochemical [56] D.W. Cawlfield, J.J. Kaczur, H. Arabghani. Process for removing catalyst impurities
reduction of carbon dioxide to formate on tin electrode, J. Power Sources 253 from polyols. U.S. Patent 5,254,227 (1993).
(2014) 276–281. [57] R.I. Masel, B.A. Rosen. Catalyst Mixtures. U.S. Patent 8,956,990 (2014).
[37] S. Zhang, P. Kang, T.J. Meyer, Nanostructured tin catalysts for selective electro- [58] R.I. Masel, Q. Chen. Carbon Dioxide Conversion to Fuels and Chemicals. U.S. Patent
chemical reduction of carbon dioxide to formate, J. Am. Chem. Soc. 136 (2014) 9,193,593 (2015).
1734–1737. [59] R.I. Masel, Q. Chen, Z. Liu, R. Kutz. Ion-Conducting Membranes, U.S. Patent 9,370,
[38] J. Wu, P.P. Sharma, B.H. Harris, X.X. Zhou, Electrochemical reduction of carbon 773 (2016).
dioxide: IV dependence of the Faradaic efficiency and current density on the [60] R.I. Masel, Q. Chen, Z. Liu, R. Kutz. Electrochemical Device for Converting Carbon
microstructure and thickness of tin electrode, J. Power Sources 258 (2014) Dioxide to a Reaction Product. U.S. Patent 9,481,939 (2016).
189–194. [61] R.I. Masel, B.A. Rosen, W. Zhu. Devices and Processes for Carbon Dioxide
[39] J.L. White, J.T. Herb, J.J. Kaczur, P.W. Majsztrik, A.B. Bocarsly, Photons to formate: Conversion into Useful Fuels and Chemicals. U.S. Patent 9,181,625 (2014).
efficient electrochemical solar energy conversion via reduction of carbon dioxide, J. [62] R.I. Masel, A. Salahi-Khogin, B.A. Rosen. Catalysts for Electrochemical Conversion
CO2 Util. 7 (2014) 1–5. of Carbon Dioxide. U.S. Patent 9,012,345 (2015).
[40] J. Qiao, Y. Liu, F. Hong, J. Zhang, A review of catalysts for the electroreduction of [63] B.A. Rosen, A. Salehi-Kohjin, M.R. Thorsen, W. Zhu, D.T. Whipple, P.J.A. Kenis,
carbon dioxide to produce low-carbon fuels, Chem. Soc. Rev. 43 (2014) 631–675. R.I. Masel, Ionic liquid-mediated selective conversion of CO2 to CO at low
[41] P. Bumroongsakulsawat, G.H. Kelsall, Tinned graphite felt cathodes for scale-up of overpotentials, Science 334 (2011) 643–644.
electrochemical reduction of aqueous CO2, Electrochim. Acta 159 (2015) 242–251. [64] Z. Liu, R.I. Masel, Q. Chen, R. Kutz, H. Yang, K. Lewinski, M. Kaplun, S. Luopa,
[42] D. Du, R. Lan, J. Humphreys, S. Sengodan, K. Xie, H. Wang, S. Tao, Achieving both D.R. Lutz, Electrochemical generation of syngas from water and carbon dioxide at
high selectivity and current density for CO2 reduction to formate on nanoporous tin industrially important rates, J. CO2 Util. 15 (2015) 50–56.
foam electrocatalysts, ChemistrySelect 1 (2016) 1711–1715. [65] R. Kutz, Q. Chen, H. Yang, S.D. Sajjad, Z. Liu, R.I. Masel, Sustainion™ imidazolium
[43] M.N. Mahmood, D. Masheder, C.J. Harty, Use of gas-diffusion electrodes for high- functionalized polymers for CO2 electrolysis, Energy Technol. (2017), http://dx.
rate electrochemical reduction of carbon dioxide. I. Reduction at lead, indium- and doi.org/10.1002/ente.201600636.
tin-impregnated electrodes, J. Appl. Electrochem. 17 (1987) 1159–1170. [66] Y.-W. Rhee, S.Y. Ha, R.I. Masel, Crossover of formic acid through Nafion®
[44] D. Kopljar, A. Inan, P. Vindayer, N. Wagner, E. Klemm, Electrochemical reduction membranes, J. Power Sources 117 (2003) 35–38.
of CO2 to formate at high current density using gas diffusion electrodes, J. Appl. [67] N. Akiya, P.E. Savage, Role of water in formic acid decomposition, AIChE J. 44 (2)
Electrochem. 44 (2014) 1107–1116. (1998) 405–415.
[45] S. Sen, B. Skinn, T. Hall, M. Inman, E.J. Taylor, F.R. Brushett, Pulsed electro- [68] M. Tsuji, M. Hattori, D. Shimamoto, Hydrogen production from formic acid
deposition of tin electrocatalysts onto gas diffusion layers for carbon dioxide decomposition at room temperature, Chim. Oggi – Chem. Today 34 (3) (2016)
reduction to formate, MRS Adv. (2017) 1–8, http://dx.doi.org/10.1557/adv.2016. 55–58.
652. [69] G.W.C. Kaye, T.H. Laby, Tables of Physical and Chemical Constants and Some
[46] N. Furuya, T. Yamazaki, M. Shibata, High performance Ru–Pd catalysts for CO2 Mathematical Functions, 15th ed., Longmans, Green, Co., NY, USA, 1986 ISBN-13:
reduction at gas-diffusion electrodes, J. Electroanal. Chem. 431 (1997) 39–41. 978-0582463547, 219.
[47] E. Korngold, Electrodialysis processes using ion exchange resins between mem- [70] J.J. Kaczur, H. Yang, S.D. Sajjad, R.I. Masel. Method and System For
branes, Desalination 16 (1975) 225–233. Electrochemical Production of Formic Acid from Carbon Dioxide. U.S. Patent
[48] E. DeJean, E. Laktionov, J. Sandeaux, R. Sandeaux, G. Pourcelly, C. Gavach, Application 2017/0037522 (2017).

217