Perforated membranes for the efficient conversion of carbon monoxide to organic compounds

This application claims priority from U.S. Provisional Patent Application No. 63/386,711, filed on Dec. 9, 2022, is a continuation-in-part of U.S. patent application Ser. No. 18/333,391, filed on Jun. 12, 2023, and is a continuation of U.S. patent application Ser. No. 18/460,186 filed Sep. 1, 2023, the contents of each of which are incorporated herein by reference in their entirety.

This invention was made with government support under Contract No. DE-AC36-08GO28308 awarded by the Department of Energy. The government has certain rights in the invention.

The electrocatalytic conversion of carbon monoxide into useful products such as formic acid is an important step in the reduction of atmospheric carbon monoxide levels and achieving a carbon neutral energy economy. Unfortunately, current methods of CO and COconversion suffer from issues regarding low energy efficiency, hurdles to scalability and the requirement of energy intensive separations of mixed product streams.

Described herein are devices and methods for the facile, efficient electrocatalytic conversion of carbon dioxide to high purity formic acid and distilled water. The described devices utilize perforated membranes to reduce energy requirements and extend electrochemical cell life by reducing damage caused by fluids being formed between internal ion exchange membranes.

In an aspect, provided is a device comprising: a) an anode; b) a cathode; and c) a cationic exchange membrane and an anionic exchange membrane positioned between the anode and cathode; wherein at least one of the cationic exchange membrane and the anionic exchange membrane is a perforated ion exchange membrane; wherein the device is an electrochemical cell for the conversion of carbon dioxide to organic compounds via electrolysis.

In another aspect, provided is a device comprising: a) an anode comprising a porous carbon electrode and an anode catalyst; b) a cathode comprising a porous carbon electrode and a cathode catalyst; c) a perforated cationic exchange membrane, positioned between the anode and the cathode; and d) an anionic exchange membrane positioned between the anode and the cathode; wherein the device is an electrochemical cell for the conversion of carbon dioxide to organic compounds via electrolysis.

The perforated ion exchange membrane may have various physical configurations and modifications to allow generated products to easily flow towards the anode without becoming trapped between the various other components of the electrochemical cell, for example, between the anion exchange membrane and the cation exchange membrane (which in this case, would be perforated). Examples of configurations included alternating lines (i.e., alternating parallel segments of membrane and void space, either vertically or horizontally), holes of void space, polka-dots (circles of membrane), crisscrossing lines or other geometric configurations. The perforated ion exchange membrane may have void space greater than or equal to 5%, 10%, 15%, 20%, 25%, 30%, 40% or 50% of the surface area of the membrane or the facing surface of the membrane. The perforated ion exchange membrane may have void space selected from the range of 0.1% to 50% of the surface area of the membrane or the facing surface of the membrane. The membrane may also be a catalyst coated membrane, for example, a membrane coated with Pt/HSC or Pt/V. Examples of useful membranes include Nafion or Nafion-based membranes and other polymers or fluorinated polymers.

The cathode may be a gas diffusion electrode and may comprise additional components including a cathode catalyst, a porous carbon electrode and additional membranes including anion exchange membranes. An example of an anion exchange membrane is a functionalized poly (aryl piperidinium) polymer membrane.

The anode may also comprise additional components including a porous carbon electrode and one or more additional catalyst layers.

In an aspect, provided is a method comprising: a) providing an electrochemical cell comprising: i) an anode; ii) a cathode; iii) a cationic exchange membrane and an anionic exchange membrane positioned between the anode and cathode; wherein at least one of the cationic exchange membrane and the anionic exchange membrane is a perforated ion exchange membrane; b) generating an electrical current between the anode and the cathode; c) flowing carbon dioxide gas to the cathode, thereby reacting the carbon dioxide gas and generating organic compounds.

The provided carbon dioxide gas may contain additional molecules or reactants, for example, potassium hydroxide and potassium bicarbonate.

The devices and methods provided herein may also be used for the facile conversion of carbon monoxide into organic compounds. Accordingly, the various references to COmay be substituted with CO or a combination of both COand CO and provide the same or similar efficiencies and advantages. For clarity, any reference in this application to COcan also be applied to CO.

The embodiments described herein should not necessarily be construed as limited to addressing any of the particular problems or deficiencies discussed herein. References in the specification to “one embodiment”, “an embodiment”, “an example embodiment”, “some embodiments”, etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

As used herein the term “substantially” is used to indicate that exact values are not necessarily attainable. By way of example, one of ordinary skill in the art will understand that in some chemical reactions 100% conversion of a reactant is possible, yet unlikely. Most of a reactant may be converted to a product and conversion of the reactant may asymptotically approach 100% conversion. So, although from a practical perspective 100% of the reactant is converted, from a technical perspective, a small and sometimes difficult to define amount remains. For this example of a chemical reactant, that amount may be relatively easily defined by the detection limits of the instrument used to test for it. However, in many cases, this amount may not be easily defined, hence the use of the term “substantially”. In some embodiments of the present invention, the term “substantially” is defined as approaching a specific numeric value or target to within 20%, 15%, 10%, 5%, or within 1% of the value or target. In further embodiments of the present invention, the term “substantially” is defined as approaching a specific numeric value or target to within 1%, 0.9%, 0.8%, 0.7%, 0.6%, 0.5%, 0.4%, 0.3%, 0.2%, or 0.1% of the value or target.

As used herein, the term “about” is used to indicate that exact values are not necessarily attainable. Therefore, the term “about” is used to indicate this uncertainty limit. In some embodiments of the present invention, the term “about” is used to indicate an uncertainty limit of less than or equal to ±20%, ±15%, ±10%, ±5%, or ±1% of a specific numeric value or target. In some embodiments of the present invention, the term “about” is used to indicate an uncertainty limit of less than or equal to ±1%, ±0.9%, ±0.8%, ±0.7%, ±0.6%, ±0.5%, ±0.4%, ±0.3%, ±0.2%, or ±0.1% of a specific numeric value or target.

As used herein, the term “organic compounds” refers to reaction products of CO and COelectrochemical reactions, including formic acid. The addition of additional reactants such as KOH and KHCOand recycling or re-reacting product molecules allows for the production of larger organic compounds, for example, C-Ccarboxylic acids, ethanol and ethylene, as examples. Organic compounds may also comprise CO and CO, regardless of whether a person having skill in the art would consider CO as “organic.” For example CO may be converted in to COor COmay be converted in to CO.

As used herein, the term “perforated ion exchange membrane” refers to a non-continuous polymer layer capable of exchanging ions, either cations or anions. While embodiments use a traditional membrane other forms of polymer thin film deposition (such as electrospinning) are also considered.

provides a cross-sectional schematic of an electrochemical cellutilizing a perforated ion exchange membranepositioned in between the anodeand the cathode. In some embodiments, the perforated ion exchange membraneis cationic and positioned proximate to the anode; however, the perforation may occur on the cathode side of the cell and utilize an anionic perforated membrane.

The anodemay include several features including an electrodeand a catalystor a catalytic layer. In some embodiments, a catalyst may be dispersed on the perforated (or unperforated) cationic membrane. The anion electrodemay be porous, including porous carbon.

The cathodemay also include several features including an electrodeand a catalystor a catalytic layer.

Additionally, in embodiments in which the perforated ionic membraneis cationic and proximate to the anode side of the electrochemical cell, an additional perforated or unperforated anionic exchange membranecan be included proximate to the cathode.

The electrochemical cellalso includes a gas flow chamberfor providing reactants to the cell. Example reactants include high purity carbon dioxide gas (e.g., greater than 90%), potassium hydroxide and potassium bicarbonate. The reactants are typically in the gas phase but may include liquids including dispersed liquid droplets. After conversion via electrolysis, products flow through the product flow chamber. Example products include formic acid and water in the liquid phase.

The benefit of utilizing two ionic exchange membranes (one anionic and one cationic) is illustrated by, which provides Faraday efficiency for a formic acid generating cell with a single membrane. Without the inclusion of the second, cationic exchange membrane a large amount of the formic acid product is oxidized, greatly reducing yields. However, the use of two full membranes causes products to build in between the two membranes causing ballooning or delamination at the interface between the membranes, as shown in. Evidence of this problem is shown in the images provided in, where the used cell with a perforated cationic exchange membrane () shows a smooth interface between the two membranes the unperforated membrane cell () shows clear evidence of delamination and damage. By perforating one of the membranes, in this case the cationic exchange membrane on the anode side, cell life is increased by about two hundred times () and initial Faraday efficiency is increased by more than five times ().

provides an example configuration of a perforated cationic exchange membrane, as described herein. The anode electrode is Pt/HSC 0.25 mg/cmon 211 patterned membrane with Toray paper (no microporous layer (MPL)). The anode is being provided with 0.8 SLPM Hand 10 mL/min of deionized (DI) water. The anionic exchange membrane is 80 μm thick Versogen. The cathode is 39BB gas diffusion layer (GDL) with 2.5 mg/cmBiOand an ionomer, I: catalyst 0.02. The cathode is provided with 2 SLPM COat 100% RH with premixed 2 mL/min 1M KOH. The cell temperature is 30° C. Experimental results forare provided in.

provides another example configuration of a perforated cationic exchange membrane, as described herein. The anode electrode is Pt/HSC 0.25 mg/cmon 117 patterned membrane with Toray paper (no MPL). The anode is being provided with 0.8 SLPM Hand 10 mL/min of DI water. The anionic exchange membrane is 80 μm thick Versogen. The cathode is 39BB GDL with 2.5 mg/cmBiOand an ionomer, I: catalyst 0.02. The cathode is provided with 2 SLPM COat 100% RH with premixed 5 mL/min 1M KOH. The cell temperature is 30° C. Experimental results forare provided in.

provides another example configuration of a perforated cationic exchange membrane, as described herein. The anode electrode is Pt/HSC 0.25 mg/cmon NC700. The anode is being provided with 0.8 SLPM Hand 10 mL/min of DI water with a 10 g/hr DI water humidifier. The anionic exchange membrane is 80 μm thick Versogen. The cathode is 39BB GDL with 4 mg/cmBiOand an ionomer, I: catalyst 0.02. The cathode is provided with 2 SLPM COat 100% RH with premixed 2 mL/min 1M KOH. The cell temperature is 60° C. Experimental results forare provided in.

In this example, a zero-gap membrane electrode assembly (MEA) architecture was developed for the direct electrochemical synthesis of formic acid from carbon dioxide. The MEA contained a composite forward bias bipolar membrane with a perforated cation exchange membrane (PCEM) which allowed formic acid generated at the membrane interface to exit through the anode flow field at concentrations up to 0.25 M. As a result, this system is poised to leverage currently available materials and stack designs ubiquitous in fuel cell and Helectrolysis spaces, enabling a more rapid transition to scale and commercialization. The PCEM configuration achieved >80% Faradaic efficiency (FE) to formic acid at <2 V and 300 mA/cmin a 25 cmcell. More critically, a 55-hour stability test at 200 mA/cmshowed stable Faradaic efficiency and cell voltage.

The electrochemical reduction of carbon dioxide to formic acid using renewable electricity has been shown to reduce production cost compared to traditional fossil-based approaches by as much as 75%. Formic acid has a wide range of applications, from an effective and economical hydrogen-storage and transportation medium, to a feedstock for the chemical or biomass industry. Formic acid has even been identified as an input for subsequent conversion to sustainable jet fuel intermediates utilizing metabolic engineering. With momentum building for a formic acid economy, multiple research efforts have focused on the optimization of catalyst selectivity. However, the majority of efforts remain on small-scale H-cell or liquid flow cells operating at low current densities (<50 mA/cm). To reduce cost, enable commercialization, and increase subsequent market penetration, electrochemical carbon dioxide reduction (CO2R) must be performed at high current densities (>200 mA/cm) and Faradaic efficiency (FE) while maximizing the utilization of materials and stack components from fuel cell and water electrolysis technology that enable CO2R devices to leverage economies of scale. Furthermore, to enhance the production utility and avoid additional downstream processing, formic acid should be targeted as an end product instead of formate salt.

Along these lines, recent efforts have been made to develop industrially relevant gas diffusion electrode (GDE) based CO2R to formate/formic acid devices. These device configurations can be categorized into three main groups: 1. Flowing catholyte, 2. Single membrane (either cation exchange membrane (CEM) or anion exchange membrane (AEM)), and 3. Interlayer configurations. Simplified cross sections for these configurations are shown in. For catholyte configurations, an electrolyte chamber is created between the membrane and the cathode GDE. The flowing catholyte serves to provide ionic access to the cathode catalyst layer, though its necessity to control formate selectivity has been debated. Nevertheless, recently, researchers achieved up to 90% FE to formate at 500 mA/cmusing a carbon supported SnOcathode with 1.27 mm thick catholyte layer. The thick catholyte layer, coupled with a reverse-bias bipolar membrane (BPM) to limit ion crossover, resulted in an operating voltage of 6 V and a 15% energy efficiency. To improve energy efficiency, a single CEM configuration was implemented, achieving 93.3% FE at a partial current density of 51.7 mA/cm. Both approaches yielded formate as opposed to the preferred product, formic acid. Additional processing requirements aside, in a CEM configurations, formate salt (e.g., KCOOH) frequently accumulates in the GDE and flow field resulting in transport limitations and eventual cell failure.

To combat the formation of formate salt, the use of a solid electrolyte interlayer between the CEM and AEM to promote the formation of formic acid, which achieved 91.3% FE at 200 mA/cmin a 5 cmcell, yielding a 6.35 wt. % formic acid solution. In a separate experiment, a similar configuration achieved 83% carbon dioxide (CO) to formic acid FE at 200 mA/cm, examining the durability of the system over 100 hrs. While small scale results are promising, the added cost of the porous ion-exchange resin, additional complexity in stack design, and the pressure required to push formic acid through the porous interlayer make interlayer configurations challenging to scale to larger systems (e.g., 1000 cm).

To help visualize the net impact of the various designs, we tabulated formate/formic acid production per kWh for all the previously referenced systems and plotted them in. Here, it is evident that any system containing a catholyte or interlayer reaches a maximum in performance at low current densities with performance decreasing at higher current densities. Additionally, while energy efficient CEM configurations yield the highest production of mol/kWh, salt accumulation results in a rapid decrease in performance at high current densities.

To mitigate the previously discussed failure modes, we developed a membrane electrode assembly (MEA) containing a composite forward bias BPM with a perforated cation exchange membrane (PCEM). This architecture is detailed in. The anode is supplied with hydrogen (H), and protons are generated by hydrogen oxidation reaction (HOR). The introduction of a PCEM layer in a BPM system enables formate ions generated at the cathode to traverse the AEM, combine with protons to form formic acid at the BPM interface and interstitial CEM pores, subsequently departing through the anode GDE and flow field. Using this configuration, we achieve >80% FE to formic acid at <2 V and 300 mA/cmfor a 25 cmcell. Most critically, this design leverages commercially available components and device architectures used for fuel cell and water electrolyzer stacks, enabling a more rapid transition to scale.

To suppress Hevolution, a plethora of COreduction efforts have utilized MEA configurations and AEM membranes, coupled with high molarity electrolytes (e.g., 1-10 M KOH) to create alkaline conditions on the cathode (as shown in). In these configurations, formate ions generated at the cathode traverse the membrane as negatively charged species where they then form KCOOH and are ushered out of the system through the anode KOH stream. While formate FE and cell voltage are initially favorable,, stability tests resulted in an approximately 30% FE reduction over just 10 hrs (). It should be noted that the use of 1 M KOH anolyte is critical to both minimize anode overpotential in basic oxygen evolution reaction (OER) systems and enable ionic accessibility within the cathode catalyst layer. High KOH concentrations can, however, introduce carbonate/bi-carbonate salt formation at the cathode and result in potassium formate production as opposed to the preferred product, formic acid. When anolyte concentration is reduced to 0.1 M KOH, both cell voltage and formate oxidation (formate loss) increase (), illustrating the zero-sum tradeoff.

To target formic acid generation, Hwas supplied at the anode to a carbon supported Pt (Pt/C) catalyst. While this AEM configuration produced lower cell voltage and the FE for CO and Hremained low (<20% total FE from 100-300 mA/cm), there was significant oxidation of formic acid at the anode. In fact, from 100-400 mA/cm>60% of the total product FE was unaccounted for due to formic acid oxidation. Along similar lines, forward bias BPMs that generate protons at the anode have been previously examined to enable formic acid production (). It is worth revisiting the performance and failure modes for BPM configurations for such systems absent more costly and complex interlayer configurations that would prove difficult to scale. As shown on, the BPM configuration cell failed after 40 mins at 200 mA/cm, accompanied by a voltage surge to over 5 V. Significant delamination at the CEM/AEM interface was observed after the test. In addition to formate, anions such as carbonate, bicarbonate, and hydroxide can also transport through the AEM membrane, reacting with protons at the CEM/AEM interface, forming COgas and liquid water, which can lead to BPM delamination () and ultimately cell failure.

Based on the performance of the above-mentioned configurations, a new MEA architecture was targeted that: 1. produced formic acid, 2. limited formic acid oxidation, 3. leveraged currently commercially available materials, 4. leveraged fuel cell and electrolyzer components and designs, and 5. provided adequate electrode ionic conductivity with minimal electrolyte in an energy efficient system. A schematic for this architecture is shown inand detailed further in. Here, the PCEM layer provides pathways for formic acid and anions to migrate from the CEM/AEM interface, reducing species accumulation. Simultaneously, the PCEM pathways direct formic acid into the diffusion media and flow field, avoiding concentration build up near the Pt/C anode catalyst, reducing the likelihood for formic acid oxidation. Polarization results using 80, 40 and 25 μm thick AEMs are shown in. It is important to mention that while most of the formic acid produced at the CEM/AEM junction can pass through the perforated channels on the CEM, there will inevitably be a localized area of high formic acid concentration at the interface. The high formic acid concentration gradient can lead to a small amount of formic acid directly diffusing through the CEM, with subsequent oxidation of formic acid at the anode. Additionally, formic acid will also diffuse back through the AEM to the cathode. Thus, while the total cell voltage increases with AEM the thickness, as expected, use of thicker AEMs prevents formic acid back diffusion, increasing cathode pH and reducing Hproduction ().

shows the formic acid concentration and pH distribution throughout the thickness direction of the MEA using finite element Poisson-Nernst-Plank simulation. It is not surprising that the CEM/AEM interface exhibits the greatest concentration of formic acid, 0.23 mol/L, as formic acid is produced at this interface. As the AEM thickness increases, there is a faster drop-off of formic acid concentration through the AEM, which suggests a greater mass transport resistance and a smaller flux of formic acid due to back diffusion.show the resulting pH and formic acid within the cathode catalyst layer caused by back diffusion, and 2D formic acid concentration distribution, respectively. With thinner AEM membranes, the formic acid concentration near the cathode is higher and cathode pH becomes acidic. Therefore, while thicker AEM membranes lead to higher ohmic losses, they are critical to prevent formic acid back diffusion to the cathode and maximizing a high net system formic acid FE. Ultimately, increasing AEM thickness to 80 μm enabled >80% FE to formic acid at <2 V and 300 mA/cmfor a 25 cmcell.

As described herein, beyond FE and energy efficiency, understanding system stability (degradation at constant conditions) and/or durability (degradation under dynamic conditions) is critical to mitigating losses and enabling scalable processes. To test the stability of the PECM based architecture, the cell was held at 200 mA/cmfor 55 hrs. The overall results are displayed inwith results from the first 3 hrs highlighted in. When a Pt/C anode catalyst was used, the cell voltage increased dramatically, from 1.3 to 1.8 V, within the first 30 minutes (). At longer durations, the cell voltage remained nearly constant, yielding a degradation rate of 0.6 mV/hr () At the beginning of test, the FE for formic acid collected at the anode is 76.5%, and the cathode FE for hydrogen is 19.2%. After the first hour of testing, the FE for hydrogen dropped to 13.8%, suggesting an improved selectivity for formate. However, the system FE for formic acid dropped to 62.7% at 1 hour, and the anode formic acid oxidation rate increased from near zero at the beginning of the test to 17.0%. Subsequently, the FE for H, CO, formic acid, and the anode formic acid oxidation rate remained stable for the duration of the experiments. The increase of formic acid oxidation over the first hour is likely related to formic acid accumulation at the PCEM/AEM interface. As the concentration of formic acid builds up, it will not only exit through membrane perforations but also via diffusion through the CEM itself, entering the Pt/C anode layer. Since formic acid is a liquid at 60° C., its accumulation can cause mass transport issues and lead to preferential oxidation over hydrogen gas.

To elucidate degradation mechanisms, the morphology for both beginning-of-test (BOT), as prepared, and end-of-test (EOT), post 55-hr stability tested samples were characterized using nano-Xray computed tomography (nano-CT), as shown in. Based on the 3D structure, the EOT sample has a larger catalyst particle size compared to the BOT sample. The average catalyst cluster grows from 930 nm to 1207 nm in diameter, likely also shifting from physically interacting aggregates to agglomerated structures. The High-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image and energy-dispersive X-ray spectroscopy (EDS) results are shown in. While the BOT catalyst layer contains a large portion of smaller catalyst particles, as well as some larger agglomerates, at the EOT, the catalyst layer can be divided into two separate regions: one with significantly larger solid particles and a more porous region containing a significant number of smaller particles. The EDS mapping shows that the large solid particles are Bi rich, likely metallic Bi, while the porous region is oxygen rich. When the cell is operated at 200 mA/cm, the negative potential at the cathode will lead to reduction of BiO, as evidenced by in situ X-ray absorption spectroscopy results discussed below. The HAADF-STEM and EDS mapping results indicate that certain particles of BiOundergo a reduction process, causing them to lose oxygen and coalesce into larger metallic particles. Simultaneously, Bi from smaller particles may reattach to the larger metallic cores, resulting in an increase in the size of those particles. X-ray diffraction (XRD) patterns of the BOT and EOT cathodes support the interpretation of the EDS data (): with only crystalline BiOdetected in the BOT cathode and crystalline Bi metal detected at the EOT. To understand the effect of cathode potential on the oxidation state of the BiOcathode catalyst, in-situ X-ray absorption spectra (XAS) were acquired at the Bi Labsorption edge in 0.1 M KOH from the open circuit potential (+0.3 V vs RHE) to −1.5 V vs RHE. The onset of the reduction of the BiOphase at −0.85 V vs RHE was observed, as indicated by a decrease in the white line intensity in the near-edge region of the spectrum, with 90% reduction to Bi metal at −1.1 V vs RHE (). Regardless of the mechanism and despite the pronounced change in cathode morphology, catalyst oxidation state and crystallite structure, overall formate selectivity in the cathode, inferred from Hand CO FE along with formic acid production, remains largely unaffected.

Cell operating conditions can play a role in the transport, accumulation, and kinetics for formic acid oxidation. With a similar boiling point to water (100.8 vs 100° C. at 101.3 kPa), the transport and accumulation of formic acid will be impacted by cell temperature.displays partial current densities as a function of temperature. As operating temperature increases, formic acid FE decreases, which is accompanied by the increased partial current density for hydrogen evolution reaction (HER). The rate for formic acid oxidation is also higher at higher temperatures, while in constrast, CO production is less sensitive to the temperature change.

To quantify exactly how much opportunity exists to improve the energy efficiency as formic acid oxidation is suppressed, a Hreference electrode was used to discern voltage loss contributions. At current densities below 500 mA/cm, the cathode potential remained lower than −1.25 V. The anode potential is divided into two main parts, the theoretical overptential of HOR predicted by the HOR exchange current density and Bulter-Volmer equation from previous measurements, and the remainder attributed to formic acid oxidation. A small rate of formic acid oxidation reaction at the anode can lead to a significant increase in the anode potential due to its much slower reaction kinetics compared to HOR. The result indicates that nearly 500 mV of overpotential can be eliminated by completely supressing anode formic acid oxidation.

To test this assessment, the deionized water (DI) flow rate was varied on the anode inlet to reduce formic acid effluent concentration.show the resulting FE, formic acid concentrations and cell voltages at 200 mA/cmas a function of anode DI flow rate. As the DI flow rate increased from 3.3 mL/min to 25 mL/min, formic acid anode concentration decreases from 0.27 to 0.08 mol/L. The reduced concentration improved overall formic acid FE, decreasing HFE, as cathode pH becomes more basic due to reduced back diffusion of formic acid. The reduced formic acid concentration at the highest DI flow rate also nearly eliminates formic acid oxidation, pushing the overall cell voltage to just below 1.7 V at 200 mA/cm. When viewed as a complete system, technoeconomic analysis will need to be utilized to inform the best combination of stack operating conditions relative to subsequent distillation to the 85 wt. % formic acid commercial standard. Nevertheless, whether through the use of anode catalysts with improved Hto formic acid selectivity, or device operation, PCEM based architectures achieve vastly improved energy efficiency when formic acid oxidation is suppressed.

In summary, we investigated several zero-gap MEA configurations for COto formic acid reduction and proposed a novel structure that contains a composite forward bias bipolar membrane including a perforated cation exchange membrane (PECM) to facilitate the mass transport of formic acid generated at the membrane interface. This configuration generated >96% formic acid up to 0.25M (using an anode DI flow rate of 3.3 mL/min). At higher DI flow rates (25 mL/min), the configuration yielded >80% FE 200 mA/cmcurrent density at 1.7 V using a 25 cmcell. At moderated anode DI flow rates (10 mL/min) the PECM configuration maintained a stable voltage and high FE for formic acid over a 55 hr test at 200 mA/cm. The high stability and selectivity achieved with commercially available catalysts and polymer membrane materials can be further amplified by combining it with optimized electrocatalysts. The approach for COfor formic acid presented here, eliminates the need for anolyte and catholyte chambers, interlayer components, and specialized materials thereby improving cell energy efficiency and reducing system complexity to facilitate system scale up.

Cathode Gas Diffusion Electrode Fabrication—All materials and reagent grade solvents were used as received unless otherwise noted. Bismuth oxide catalyst (BiO, 80 nm) was purchased from US Research Nanomaterials, Inc. The polymer (AP1-CNN8-00-X) powder was supplied by IONOMR. Omnisolv® grade n-propanol (nPA) and ultra-pure water (18.2 Ω, Milli-Q® Advantage A10 Water Purification System) were obtained from Millipore Sigma. ACS Certified methanol and acetone were purchased by VWR Chemicals BDH® and Fisher Chemical, respectively. The polymer powder was combined in a 1:1 weight mixture of acetone and methanol to render a 6.5 wt. % polymer dispersion. The catalyst ink was prepared by combining 20 g of BiO, ultra-pure water, nPA, and ionomer dispersion into a 30 mL jar. This formulated recipe contained 30 wt. % catalyst, an ionomer-to-catalyst weight ratio of 0.02, and an alcohol-to-water weight ratio of 2:3 (40 wt. % nPA). 70 g of Glen Mills 5 mm zirconium oxide grinding media were added to this mixture prior to mixing. Samples were placed on a Fisherbrand™ Digital Bottle Roller at 80 rpm for 26 hrs. Inks were allowed to settle for 20 min before coating. BiOinks were coated at 22° C. utilizing a ½″×16″ wire wound lab rod (RD Specialties—60 mil diameter) on a Qualtech automatic film applicator (QPI-AFA6800). 5 mL of catalyst ink deposited on 7.5″×8″ Sigracet 39 BB carbon gas diffusion media (Fuel Cell Store) were rod coated at a fixed average speed of 55 mm/sec. These coated electrodes were transferred to an oven and dried at 80° C.

Membrane Electrode Assembly-For the composite membrane configuration that contains anion exchange membrane (AEM) and perforated CEM. Nafion NC700 (Chemours, USA) with a nominal thickness of 15 μm is used as the CEM layer. The anode catalyst was directly spray coated on the CEM, with an ionomer to carbon ratio of 0.83, and a coating area of 25 cm. High surface area supported platinum (50 wt. % Pt/C, TEC 10E50E, TANAKA precious metals) with a loading of 0.25 mg/cmwas used as the anode catalyst. Nafion D2020 (Ion Power, USA) was used as the anode catalyst layer ionomer. The CEM perforation was conducted by cutting parallel lines on the CEM membrane with a spacing of 3 mm. During cell assembly, the perforated catalyst coated CEM membrane was place on a 25 cmToray paper (5 wt. % PTFE treated, Fuel Cell Store, USA). An AEM membrane (PiperION, Versogen, USA) was placed on top of the CEM and the cathode GDE on top of the AEM. A PTFE gasket was used for both anode and cathode with thicknesses to achieve an optimum GDE compression of 18%.

Electrochemical Testing: During the test, the assembled cell was held at 60° C. (for temperature dependent study, 30, 60 and 80° C.), with the anode supplied with 0.8 slpm hydrogen and cathode supplied with 2 slpm COgas. Both anode and cathode gas stream were humidified with a relative humidity of 100%, and a cathode back pressure of 37.6 kPa. During the operation, the cathode gas stream was mixed with 2 mL/min of 1 molar KOH, to facilitate cathode catalyst layer utilization and ion conduction. The anode gas stream was mixed with 10 mL/min DI water, to help remove formic acid at the anode. The electrochemical testing was conducted using a Garmy Potentiostat (Reference 30K, Gamry, USA). The cell was conditioned using liner scanning voltammetry from 0 to 250 mA/cmfor 4 times with a scan rate of 2.5 mA/cmprior to polarization curve measurement. The polarization curves were obtained using galvanostatic mode, with the cell held at certain current densities for 4 mins before collecting cathode gas and anode liquid sample.

Reference Electrode and Voltage Breakdown: We used a hydrogen reference electrode in the MEA to separate the cathode and anode potentials. The structure of the reference electrode is shown in figure SA strip of Nafion membrane (Nafion 211, IonPower, USA) is used as the ion bridge to connect the MEA membrane with the reference electrode. One end of the Nafion strip was connected to a 1 cmgas diffusion electrode (GDE) that has a loading of 0.25 mg Pt/cm(50 wt. % Pt/C, TEC10E50E, TANAKA precious metals) spray coated on 29BC carbon paper (Fuel Cell Store, USA). Custom-made Polyether ether ketone (PEEK) hardware was used for gas sealing and to guarantee good contact between the GDE and the Nafion strip, as well as to connect the reference electrode to the fuel cell hardware. The other end of the Nafion strip was linked to an overhanging edge of the cell's CEM.

Product Quantification: The gas samples were collected from the cathode, after the effluent gas passed through the condenser and gas-liquid separator. The collected gas was analyzed using a 4900 Micro GC (10m, molecular sieve, Agilent) at least three times. Samples were collected in Supel™ Inert Multi-Layer Foil Gas Sampling Bags (Sigma-Aldrich) for a specific duration (30 seconds) and manually inserted into the Micro GC within two hours of collection. The injection temperature was set to 110° C. Carbon monoxide (CO) and hydrogen (H2) were separated within a heated (105° C.) and pressurized (28 psi) 10 m MS5A column using argon (Matheson Gas-Matheson Purity) as the carrier gas. The compounds were detected on an integrated thermal conductivity detector (TCD). The GC chromatogram and calibration curves of CO and Hare shown in figure S. Liquid formic acid sample were collected from the anode with a certain period (120 seconds) and filtered with PTFE 0.22 μm syringe filters into 2 mL vials. The liquid products in the vials were analyzed using an Agilent 1260 Infinity II Bio-inert High-Performance Liquid Chromatography (HPLC) system, where a 20 μL sample volume was injected via autosampler (G5668A) with a mobile phase of 4 mM sulfuric acid (HSO) flowing at 0.6 mL/min (G5654A quaternary pump). The products were separated on a heated (35° C., G7116A column thermostat) Aminex HPX-87H 300×7.8 mm Column (Bio-Rad) with a preceding Micro-Guard Cation H guard column. Formic acid was detected on a Diode Array Detector (DAD) at 210 nm with a bandwidth of 4 nm.

Liquid product (Formic acid) FE is calculated using equation 1,