A MICROCHANNELED SOLID ELECTROLYTE AND RELATED ELECTROLYZER FOR ENHANCED ELECTROCHEMICAL REDUCTION OF CO2

Abstract

The present techniques relate to a microchanneled solid electrolyte (MSE), an electrolyzer and a method including or using the MSE for in-situ regeneration and collection of CO2 during a CO2 electroreduction operation. The MSE includes an anion conducting layer configured to conduct (bi) carbonate anions from a surface of an adjacent cathode; a cation conducting layer configured to conduct protons from a surface of an adjacent anode; and an integrated channel layer comprising multiple microchannels formed between the anion conducting layer and the cation conducting layer. The microchannels define a hollow path extending across the integrated channel layer for receiving the (bi) carbonate anions from the anion conducting layer and the protons from the cation conducting layer, thereby locally regenerating and collecting CO2 along and within the microchannels.

Description

TECHNICAL FIELD

The present techniques relate to electrochemical reduction of carbon dioxide to carbon products, and more particularly to a microchanneled solid electrolyte facilitating regeneration and collection of carbon dioxide.

BACKGROUND

The electrochemical carbon dioxide reduction reaction (CO₂RR) allows for utilization of intermittent renewable electricity and the mitigation of CO₂ emissions. Copper-based CO₂RR catalysts generate valuable multi-carbon chemicals (C₂⁺) at industrially relevant reaction rates and selectivities. Performing CO₂RR at industrially relevant reaction rates generates alkaline conditions which are favourable for CO₂RR but promote the rapid chemical conversion of CO₂ to carbonates and bicarbonates. In conventional C₂⁺ electrolyzers, between 70 and 95% of the supplied CO₂ is converted to carbonates and bicarbonates, and is thus lost to the electrolyte, anode tail gas, and/or cathode salt precipitation, as exemplified in FIG. 1. A small fraction of the remaining CO₂ is converted to products in conventional systems.

Recovering lost CO₂ reactant can incur an economic penalty 0.7 to 5 times the ethylene (C₂H₄) market price (Sisler et al. (2021). Ethylene Electrosynthesis: A Comparative Techno-economic Analysis of Alkaline vs Membrane Electrode Assembly vs CO₂—CO—C₂H₄ Tandems. ACS Energy Lett. 6, 997-1002). In addition to trapping CO₂, the accumulation of salt within the system limits CO₂RR performance and stability. Known solutions to reduce the loss of CO₂ include employing bipolar membranes, solid-state electrolyte packed beds, and surrounding the locally alkaline cathode with acidic electrolyte that regenerates gaseous CO₂.

However, there are still challenges to be overcome to offer a CO₂RR system minimizing CO₂ losses, while achieving satisfactory yield in carbon products, stability and energy efficiency.

SUMMARY

There is provided herein a microchanneled solid electrolyte that can be part of a CO₂RR system to favour in-situ regeneration of CO₂ and recycling of the regenerated CO₂ to a cathode side of the system.

In one aspect, there is provided a microchanneled solid electrolyte (MSE) for in-situ regeneration and collection of carbon dioxide (CO₂) during a CO₂ electroreduction operation. The MSE includes an anion conducting layer configured to conduct (bi) carbonate anions from a surface of an adjacent cathode; a cation conducting layer configured to conduct protons from a surface of an adjacent anode; and an integrated channel layer comprising multiple microchannels formed between the anion conducting layer and the cation conducting layer. The microchannels define a hollow path extending across the integrated channel layer for receiving the (bi) carbonate anions from the anion conducting layer and the protons from the cation conducting layer, thereby locally regenerating and collecting CO₂ along and within the microchannels.

In some implementations, the microchannels can be defined on a surface of the cation conducting layer to produce the integrated channel layer forming a one-piece structure with the cation conducting layer. Alternatively, the microchannels can be defined on a surface of the anion conducting layer to produce the integrated channel layer forming a one-piece structure with the anion conducting layer. Further alternatively, the integrated channel layer can be a separate microporous structure positioned between the anion conducting layer and the cation conducting layer.

In some implementations, the microchannels are in fluid communication with one another to define a network that directs the regenerated CO₂ from a central region of the integrated channel layer to an edge region of the integrated channel layer.

In some implementations, the network of microchannels defines an interconnected diamond pattern. Optionally, the network of microchannels defines an interconnected square pattern. Further optionally, the network of microchannels defines an interconnected circular pattern.

In some implementations, the microchannels are sized and shaped to maintain a maximum pressure below 100 kPa, 90 kPa, 80 kPa, 70 kPa, 60 kPa, 50 kPa, 40 kPa, 30 kPa, 20 kPa or 10 kPa.

In some implementations, the microchannels are sized and shaped to maintain a voltage drop below 500 mV, 400 mV, 300 mV, 200 mV, or 100 mV.

In some implementations, the microchannels have a channel depth of at most 125 μm, between 10 and 100 μm or between 20 and 80 μm.

In some implementations, the microchannels have a pore path width between 25 and 150 μm or between 25 and 75 μm.

In some implementations, the integrated channel layer has a porosity between 5% and 95%, between 10% and 80%, between 15% and 70%, between 20% and 60% or between 25% and 50%.

In some implementations, the microchannels are uniform across the integrated channel layer.

In some implementations, the microchannels have a non-circular cross-section including cross-sections of elliptical, circular, rectangular, or square shape.

In some implementations, the anion conducting layer comprises fixed cations derived from piperidinium, imidazolium, or benzimidazolium.

In another aspect, there is provided a system for electroreduction of CO₂ into carbon products. The system includes:

• • an electrolyzer comprising:
    • a cathode flow field having an inlet to receive a CO₂ gas stream; a cathode in fluid communication with the cathode flow field to operate electroreduction of the CO₂ gas stream;
    • an anode flow field having an inlet to receive an anolyte stream; an anode in fluid communication with the anode flow field; and
    • a microchanneled solid electrolyte (MSE) positioned between the cathode and the anode in a forward-biased configuration, the MSE being as defined herein and releasing a concentrated CO₂ stream comprising CO₂ and water to the cathode flow field when the CO₂ gas stream is electroreduced; and
  • a recycle loop in fluid communication with the cathode flow field for recovering the concentrated CO₂ stream from an outlet of the cathode flow field and redirect the regenerated and collected CO₂ from the concentrated CO₂ stream back to the cathode flow field for serving as at least a part of the CO₂ gas stream.

In some implementations, the recycle loop comprises: a first tubing for recovering the concentrated CO₂ stream comprising water and regenerated CO₂ from the microchannels; a liquid-gas separator in fluid communication with the first tubing to receive the concentrated CO₂ stream and separate the concentrated CO₂ stream into a regenerated CO₂ gas stream and a water stream; and a second tubing interconnecting the liquid-gas separator to the cathode flow field for recycling the regenerated CO₂ gas stream to the cathode flow field.

In some implementations, the second tubing is in fluid communication with the inlet of the cathode flow field to provide the regenerated CO₂ gas stream along with the CO₂ gas stream to the cathode flow field.

In some implementations, the concentrated CO₂ stream comprises at least 80%, 85%, 90% or 95% of CO₂, and an anode tail gas recovered from the anode flow field comprises at most 1%, 2%, 3%, 4% or 5% of CO₂.

In some implementations, the anolyte is free of mobile alkali metal cations. Optionally, the anolyte is water or a solution of H₂SO₄, HClO₄, or a combination thereof.

In some implementations, the CO₂ gas stream is supplied at a CO₂ feed rate between 0.25 sccm·cm⁻² and 2 sccm·cm⁻², 0.5 sccm·cm⁻² and 1.5 sccm·cm⁻², or 0.8 sccm·cm⁻² and 1 sccm·cm⁻²

In some implementations, the electrolyzer is operated at a current density between 40 mA·cm⁻² and 240 mA·cm⁻², 50 mA·cm⁻² and 200 mA·cm⁻², 60 mA·cm⁻² and 160 mA·cm⁻², 80 mA·cm⁻² and 120 mA·cm⁻², or 90 mA·cm⁻² and 100 mA·cm⁻².

In another aspect, there is provided a method for reducing CO₂ losses during electroreduction of CO₂ in a CO₂RR electrolyzer comprising a cathodic compartment and an anodic compartment. The method includes:

• • supplying a CO₂ gas stream to the cathodic compartment operating CO₂ reduction reactions producing (bi) carbonate and hydroxide anions;
  • supplying an anolyte stream to the anodic compartment operating anodic reactions producing protons;
  • allowing generation and collecting of a concentrated CO₂ stream in microchannels positioned between the cathodic compartment and the anodic compartment by
    • directing (bi) carbonate and hydroxide anions from the cathode to the microchannels via an anion conducting layer, and
    • directing protons from the anode to the microchannels via a cation conducting layer; recovering the concentrated CO₂ stream from the CO₂RR electrolyzer;
  • separating water and a regenerated CO₂ gas stream from the concentrated CO₂ stream; and
  • recycling at least a portion of the regenerated CO₂ gas stream to the cathodic compartment of the CO₂RR electrolyzer.

In some implementations, supplying the CO₂ gas stream is performed at a CO₂ feed rate between 0.25 sccm·cm⁻² and 2 sccm·cm⁻², 0.5 sccm·cm⁻² and 1.5 sccm·cm⁻², or 0.8 sccm·cm⁻² and 1 sccm·cm⁻².

In some implementations, the method includes operating the CO₂RR electrolyzer at a current density between 40 mA·cm⁻² and 240 mA·cm⁻², 50 mA·cm⁻² and 200 mA·cm⁻², 60 mA·cm⁻² and 160 mA·cm⁻², 80 mA·cm⁻² and 120 mA·cm⁻², or 90 mA·cm⁻² and 100 mA·cm⁻².

In some implementations, the method includes controlling at least one of: a maximum pressure within the microchannels below 100 kPa, 90 kPa, 80 kPa, 70 kPa, 60 kPa, 50 kPa, 40 kPa, 30 kPa, 20 kPa or 10 kPa; and a voltage drop below 500 mV, 400 mV, 300 mV, 200 mV, or 100 mV.

In some implementations, the method includes providing the microchannels with at least one of: a channel depth of at most 125 μm, between 10 and 100 μm or between 20 and 80 μm; and a pore path width between 25 and 125 μm or between 25 and 75 μm.

In some implementations, the method includes providing the microchannels in an integrated channel layer between the anion conducting layer and the cation conduction layer, the integrated channel layer having a porosity between 5% and 95%, between 10% and 80%, between 15% and 70%, between 20% and 60% or between 25% and 50%.

In some implementations, the method includes providing the anion conducting layer with fixed cations derived from piperidinium, imidazolium, or benzimidazolium.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic representation of a conventional membrane electrode assembly sustaining CO₂RR and including an anion-conducting layer, showing that at least 70% of the CO₂ is lost to the anode side.

FIG. 2 is a schematic representation of a cross-sectional view of the CO₂RR membrane electrode assembly including a microchanneled solid electrolyte sandwiched between a cathode and an anode, and showing that at most 3% of the CO₂ is lost to the anode side.

FIG. 3 is a schematic representation of a partially exploded view of the microchanneled solid electrolyte showing, for example, in-situ regeneration of CO₂ and water via reaction of (bi) carbonates and protons within the microchannels, and further transfer of the regenerated CO₂ through the microchannels.

FIGS. 4A to 4F include schematic representations of the successive steps (for the fabrication of a cation-conducting layer with integrated microchannels in images (4B, 4C, 4E and 4F). (4A) the fabrication of a master by a standard photolithography process. (4D) hot embossing a cation-conducting layer with the master. (4B) A silicon wafer with a spin-coated photoresist layer. (4C) Photomask. (4E) A developed rigid master on the silicon wafer. (4F) Optical microscopy images of the cation-conducting layer with integrated microchannels.

FIG. 5 is a schematic of a CO₂ electrolyzer assembled with a microchanneled solid electrolyte (MSE).

FIGS. 6A and 6B are two photographs of a CO₂ electrolyzer assembled with an MSE: (6A) Cathode side view of electrolyzer; and (6B) CO₂ recycling and feedback with liquid and gas separator.

FIG. 7 is a schematic of poly(aryl piperidinium) functional group from the anion-conducting layer promoting CO₂RR.

FIGS. 8A to 8C includes three graphs of a computational fluid dynamics model for different channel pattern designs, all designs share the same land to channel area ratio with 75 μm pore path size diamond pattern design (1 cm² electrode area). (8A) Pressure distribution map for diamond pattern design. (8B) Pressure distribution map for square pattern design. (8C) Pressure distribution map for circular pattern design.

FIGS. 9A to 9D includes three graphs of a computational fluid dynamics model for microchannels having different pore path sizes showing (9A) Channel pressure distribution for a 25 μm pore path size (1 cm² electrode area); (9B) Channel pressure distribution for a 75 μm pore path size (1 cm² electrode area); (9C) Channel pressure distribution for a 150 μm pore path size (1 cm² electrode area), and (9D) being a zoomed portion of FIG. 9A.

FIGS. 10A and 10B include a graph (10A) showing a computational fluid dynamics model of the maximum channel pressure induced by the regeneration and flow of internally captured CO₂ in microchannels having a pore path size of 75 μm (100 cm², 100 mA cm⁻²) and a zoomed view thereof (10B).

FIG. 11 is a graph of full cell voltage (V) vs time (min), showing CO₂RR performance using MSEs with different microchannel thicknesses: 50 μm and 25 μm (75 μm pore path size, diamond pattern design, 0.01 M H₂SO₄ anolyte, 100 mA cm⁻², and 20 sccm CO₂ flow rate).

FIGS. 12A and 12C are respectively two photographs (12A and 12B) of 50 cm² CO₂ electrolyzer with MSE; and a graph (12C) showing composition of the gas collected from the 50 cm² MSE-electrolyzer anode vs time(s).

FIGS. 13A to 13C includes six optical microscope images of the microfluidic channels on the cation-conducting layer: (13A, 13C, 13E) three pre-electrolysis images, and (13B, 13, 13F) three post-electrolysis images.

FIG. 14 is a graph of a computational fluid dynamics model of the dissolved CO₂ concentration and pH distribution along a thickness of a conventional MEA electrolyzer including an anion-conducting layer (ACL) (and no ICL and CCL). (0.1 M KHCO₃ anolyte, 100 mA cm⁻²).

FIGS. 15A and 15B are two graphs of a computational fluid dynamics model of (15A) HCO₃⁻ and (15B) CO₃²⁻ concentration along a thickness of a conventional MEA CO₂ electrolyzer equipped with an anion-conducting layer (and no ICL and CCL) (0.1 M KHCO₃ anolyte, 100 mA cm⁻²).

FIG. 16 is a schematic of 1D microchanneled solid electrolyte (MSE) that is used for a computational fluid dynamics model as shown in FIGS. 17 and 18.

FIG. 17 is a graph of the computational fluid dynamics model of the dissolved CO₂ concentration and pH distribution along a thickness of a CO₂ electrolyzer equipped with a MSE (as shown in FIG. 16) for internal CO₂ capture (0.01 M H₂SO₄ anolyte, 100 mA cm⁻²).

FIGS. 18A and 18B are two graphs of the computational fluid dynamics model of (18A) HCO₃⁻ and (18B) CO₃²⁻ concentration along a thickness of the CO₂ electrolyzer equipped with a MSE (as shown in FIG. 16) for internal CO₂ capture (0.01 M H₂SO₄ anolyte, 100 mA cm⁻²).

FIG. 19 is a graph showing gas composition at an integrated channel layer (ICL) outlet of the CO₂RR MSE electrolyzer at different applied current densities (mA·cm⁻²).

FIG. 20 is a graph showing gas composition at an anode outlet of the CO₂RR MSE electrolyzer at different applied current densities (mA·cm⁻²).

FIG. 21 is a graph showing a CO₂ consumption rate as anodic CO₂, internally capture CO₂, and electroreduced CO₂ for the CO₂RR MSE electrolyzer at different applied current densities (mA·cm⁻²).

FIGS. 22A and 22B are two graphs for comparing (22A) Current-voltage behaviour and (22B) CO₂ consumption distribution ignoring unreacted CO₂ (120 mA cm⁻²) for the MSE and anion-conducting solid electrolyte (ACL) with various anolytes.

FIG. 23 is a graph showing experimental cell voltages of CO₂RR MSE electrolyzer (0.01 H₂SO₄ anolyte) and calculated voltage penalty caused by adding integrated channels at different applied current densities (mA·cm⁻²).

FIGS. 24A to 24D includes two photographs 24A and 24B and two graphs 24C and 24D showing CO₂RR with forward-bias bipolar membrane without internal CO₂ capture domain: photograph (24A) is a schematic of (bi) carbonate, CO₂ and water formation; photograph 24(B) is a post-electrolysis image; graph (24C) shows a current density (mA·com-2) vs time(s) demonstrating unstable CO₂RR performance when using 0.01 M H₂SO₄ anolyte, −3.8 V cell voltage, and 20 sccm cm⁻² CO₂ feed rate into the electrolyzer; and graph (24D) comparing a total cell voltage (V) at different current densities (mA·cm⁻²) between the MSE and the forward-bias bipolar membrane configuration (the current density plotted is from the short initial period of stable operation shown in the red box).

FIGS. 25A and 25B show CO₂RR in a CO₂ electrolyzer with solid-state electrolyte packed beds: schematic (25A) shows (bi) carbonate, CO₂ and water formation; and graph (25B) provides comparative total cell voltage (V) at different current densities (mA·cm⁻²) for a CO₂ electrolyzer using the MSE or solid-state electrolyte beads.

FIGS. 26A and 26B relate to an experiment using an MSE cell with 0.01M sulphuric acid anolyte, and 0.1M potassium sulphate that was added at 9600s (at 100 mA cm⁻² and CO₂ feed rate of 20 sccm cm⁻²): graph (26A) shows the cell voltage (V) and Faradaic Efficiency (FE) (%) versus time(s) relating to CO₂RR performance; and photograph (26B) shows salt precipitates on the cathode after a 1.5-hour operation of adding potassium sulphate (in a central region thereof).

FIG. 27 is a graph of FE (%) of major CO₂RR products at different current densities (mA·cm⁻²) with a CO₂ feed rate of 20 sccm·cm⁻² using a CO₂RR MSE electrolyzer.

FIGS. 28A and 28B relate to experiments with a CO₂RR MSE electrolyzer with 0.01M sulfuric acid anolyte (H₂SO₄), and pure DI water anolyte, at 120 mA cm⁻² and 20 sccm CO₂ flow rate: graph (28A) showing comparative current-voltage behaviour; and graph (28B) showing comparative FE of H₂ and other major CO₂RR products.

FIGS. 29A to 29F includes six graphs for comparison of MSE voltage and selectivity with three different anion-conducting solid electrolyte: graph (29A) shows total cell voltage (V) at different current densities (mA·cm⁻²) using PiperION; graph (29B) shows FE (%) of major CO₂RR products at different current densities (mA·cm⁻²) using PiperION; graph (29C) shows total cell voltage (V) at different current densities (mA·cm⁻²) using Sustainion; graph (29D) shows FE (%) of major CO₂RR products at different current densities (mA·cm⁻²) using Sustainion; graph (29E) shows total cell voltage (V) at different current densities (mA·cm⁻²) using Aemion; and graph (29F) shows FE (%) of major CO₂RR products at different current densities (mA·cm⁻²) using Aemion.

FIGS. 30A to 30C show CO₂RR performance comparison of feed-in reactant gas with recycled CO₂ stream and without recycled (with 0.01M sulphuric acid anolyte and a CO₂ feed rate of 20 sccm cm⁻²), including graph (30A) showing experimental cell voltages of different current densities with/without recycled CO₂ stream; graph (30B) showing FE at different current densities demonstrating CO₂RR product distribution with recycled CO₂ stream; and graph (30C) showing FE at different current densities demonstrating CO₂RR product distribution without recycled CO₂ stream.

FIG. 31 is a graph of the computational fluid dynamics model (based on FIG. 16) of the pH distribution at different dissolved CO₂ content (100% is the ambient CO₂ solubility limit) along a thickness of the MSE electrolyzer.

FIGS. 32A to 32C include three graphs of a computational fluid dynamics model (based on FIG. 16) of (32A) HCO₃⁻, (32B) CO₃²⁻, and (32C) CO₂ concentration in a CO₂ electrolyzer installed with the MSE for internal CO₂ capture (0.01 M H₂SO₄ anolyte, 100 mA cm⁻², and 10% of the CO₂ saturation concentration on cathode boundary condition) along a thickness of the electrolyzer.

FIGS. 33A to 33C includes three graphs of the computational fluid dynamics model (based on FIG. 16) of (33A) HCO₃⁻, (33B) CO₃²⁻, and (33C) CO₂ concentration in a microfluidic bipolar membrane CO₂ electrolyzer installed with the MSE for internal CO₂ capture (0.01 M H₂SO₄ anolyte, 100 mA cm⁻², and 50% of the CO₂ saturation concentration on cathode boundary condition) along a thickness of the electrolyzer.

FIG. 34 is a graph showing FE (%) of major CO₂RR products and a ratio of Cd₂₊ products to carbon monoxide at different CO₂ feed rates (sccm·cm⁻²) and at a current density of 100 mA·cm⁻².

FIG. 35 is a graph showing Energetic Efficiency (EE) (%) of C₁ and C₂⁺ products and a single-pass CO₂ utilization (%) at different CO₂ feed rates (sccm·cm⁻²) and at a current density of 100 mA cm⁻².

FIG. 36 is a graph showing total cell voltage (V) and FE (%) versus time (h) for CO₂RR products, C₂⁺ products and ethylene in particular demonstrating CO₂RR stability using the MSE (0.01 M H₂SO₄ anolyte, 100 mA cm⁻², and 1 sccm cm⁻² CO₂ feed rate). The CO₂ consumption rate for these conditions is quantified in FIG. 37.

FIG. 37 is a graph showing CO₂ consumption rate for the same conditions as per FIG. 36 thereby quantifying CO₂ loss detected from different sources using the MSE (100 mA cm⁻², 1 sccm cm⁻² CO₂ feed rate, 0.01 M H₂SO₄ anolyte).

While the invention will be described in conjunction with example embodiments, it will be understood that it is not intended to limit the scope of the invention to such embodiments. On the contrary, it is intended to cover all alternatives, modifications and equivalents as may be included as defined by the present description. The objects, advantages and other features of the present invention will become more apparent and be better understood upon reading of the following non-restrictive description of the invention, given with reference to the accompanying drawings.

DETAILED DESCRIPTION

The electrochemical reduction of carbon dioxide (CO₂) is a promising route to convert carbon emissions into valuable chemicals and fuels. In conventional electrolyzers producing multi-carbon products, 70-95% of the supplied CO₂ is converted to (bi) carbonates, limiting the carbon efficiency of electrochemical CO₂ conversion. These (bi) carbonate anions can be lost to the aqueous electrolyte, converted back to gaseous CO₂ and diluted in the anode tail gas, and/or combined with alkali metal cations from the electrolyte to form solid salt precipitates.

More particularly, CO₂ can engage in a series of chemical equilibria with HCO₃⁻, CO₃²⁻, H⁺, and OH in aqueous acidic (E1-E2) and alkaline (E3-E4) solutions:

$\begin{array}{cc}{\mathrm{CO}}_2+H_{2}O\leftrightarrow H^{+}+{\mathrm{HCO}}_3^{-}& \left(\mathrm{E1}\right)\end{array}$ $\begin{array}{cc}{\mathrm{HCO}}_3\leftrightarrow H^{+}+{\mathrm{CO}}_3^{2-}& \left(\mathrm{E2}\right)\end{array}$ $\begin{array}{cc}{\mathrm{CO}}_2+{\mathrm{OH}}^{-}\leftrightarrow {\mathrm{HCO}}_3^{-}& \left(\mathrm{E3}\right)\end{array}$ $\begin{array}{cc}{\mathrm{HCO}}_3^{-}+{\mathrm{OH}}^{-}\leftrightarrow {\mathrm{CO}}_3^{2-}+H_{2}O& (\mathrm{E4})\end{array}$

The (bi) carbonates refer herein to at least one of carbonate ions and bicarbonate ions that are generated from the CO₂ according to one or more of the chemical reactions E1 to E4, depending on a local pH.

The hydroxides consumed in the reactions can be supplied from the catholyte (if one is present) or from CO₂ electrolysis. These (bi) carbonates ultimately end up in one of three primary locations: the electrolyte, anode tail gas, or salt precipitates. As an example, we will consider a conventional zero-gap electrolyzer with an anion-conducting layer and an aqueous electrolyte containing alkali metal cations. To maintain electroneutrality during electrolysis, negative charges must be expelled from the cathode to the anode (or vice versa with positive charges). If there is no CO₂ loss to (bi) carbonates, then hydroxide ions must be the sole charge carrier through the anion conducting layer. If alkali metal cations from the anode are the charge carrier, then one CO₂ can be lost to salt precipitation for every one to two cations transferred, depending on whether bicarbonate or carbonate precipitates are being formed, respectively. If (bi) carbonates are the charge carriers, then they migrate from the cathode to the anode, embodying one CO₂ per anion. These (bi) carbonate ions will first adjust the anolyte pH (e.g., if the initial anolyte contained hydroxides, then the pH will decrease as hydroxides are gradually replaced with these (bi) carbonates), entombing CO₂ in the anolyte. Once the anolyte pH is stationary, any additional (bi) carbonates cannot be retained by the anolyte so they will be reverted to CO₂ gas and released in the anode tail gas.

The cathodic CO₂ reduction reaction (CO₂RR) is often paired with the anodic Oxygen Evolution Reaction (OER). OER has well-developed catalysts and consumes an inexpensive reactant feedstock. Operation of a conventional CO₂RR electrolyzer with an anion-conducting layer produces an anode tail gas with a typical composition including 60-70% v/v CO₂ and 30-40% v/v O₂. Recovering CO₂ from the anode tail gas may require costly separation processes.

To reduce or prevent losses of CO₂ (i.e., to the electrolyte, to the anode tail gas or to salt precipitation), there is provided a microchanneled solid electrolyte (MSE) comprising a network of microchannels where regeneration and collection of CO₂ are facilitated. The microchannels are distributed across the MSE and within a plane of the MSE. The microchannels should be understood herein as defining elongated openings that are in fluid communication with one another to offer a hollow path within the MSE, with this hollow path serving to conduct the regenerated CO₂ and water out of the MSE. The network of microchannels can define various patterns in accordance with the curvature and direction of each microchannel. For example, a microchannel can define a curved path. For example, a microchannel can define a linear path. It should further be noted that the microchannels are said to be distributed as a network because the microchannels can be connected to one another such that the defined hollow path does not include any blockage for the fluid to flow therethrough.

The MSE can be positioned in a zero gap configuration between a cathode and an anode of a CO₂RR system such that the microchannels are distributed between a cathode side and an anode side of the system. The use of the MSE allows for recovering CO₂ in a concentrated CO₂ stream including at least 80%, 85%, 90% or 95% of CO₂, while producing an anode tail gas that is substantially exempt of CO₂ (e.g., including at most 1%, 2%, 3%, 4%, 5%, 6% or 7% of anodic CO₂).

More particularly, referring to FIG. 2, the MSE (2) includes an anion-conducting layer (4) for direct contact with the cathode (6), a cation-conducting layer (8) for direct contact with the anode (10), and an integrated channel layer (ICL) (12) comprising the network of microchannels (14) being distributed at an interface between the anion conducting layer (4) and the cation conduction layer (8). The ICL can be referred to as an intermediate microporous channel layer. The ICL (12) can be formed from a surface of the cation-conducting layer (8) such that the ICL (12) is integral with the cation-conduction layer (8), both the ICL (12) and cation-conducting layer (8) forming a microchanneled membrane (13).

The cation-conducting layer of the MSE can conduct protons from the anode side of the system to the microchannels, and the anion-conducting layer of the MSE can conduct (bi) carbonates ions from the cathode side of the system to the microchannels for in-situ regeneration thereof to CO₂ upon encountering the locally acidic conditions. Advantageously, the (bi) carbonates ions are converted to CO₂ before reaching the anode side of the system, thereby preventing the regenerated CO₂ from being mixed with the anode tail gas. The microchannels can thus be referred to as an internal CO₂ capture domain.

It should be noted the anion-conducting layer of the MSE can further conduct, in addition to the bicarbonate ions but in a much lower amount, hydroxide ions to the microchannels. Water can thus be produced, along with CO₂ and be recovered as part of the concentrated CO₂ stream that flows out the MSE from the microchannels. The gas phase of the concentrated CO₂ stream can further include traces of H₂ and O₂, following diffusion from the cathode and anode, respectively.

In-situ regeneration of CO₂ refers to an «internal» regeneration and capture/collection of gaseous CO₂ within the microchannels of the MSE, such that the regenerated CO₂ can be transferred in gas phase out of the MSE along the microchannels and further recovered without being lost to the anode side. The CO₂ is thus regenerated en route to the anode—by desorbing CO₂ in the absence of O₂ that is produced at the anode side—which facilitates recycling of the CO₂ while eliminating two primary sources of CO₂ loss: the anode tail gas and electrolyte.

The microchannels facilitate mass and ion transfer; thereby allowing for water transport, ion transport, and gaseous CO₂ evolution and collection along and within the microchannels. The network of microchannels of the ICL can be characterized by a porosity and a channel hydraulic diameter. Such parameters can remain the same or vary across the ICL. Selection/adjustment of the porosity and the channel hydraulic diameter can be tailored to a desired range/threshold for an internal pressure build-up and an ionic conductivity of the ICL.

With regard to pressure, one can understand that the generation of CO₂ and H₂O within the microchannels can yield an internal pressure build-up. For example, in accordance with the characteristics of the material that is selected for use as the ICL, the porosity and channel hydraulic diameter can be selected/adjusted for a maximum pressure to remain below 100 kPa, 90 kPa, 80 kPa, 70 kPa, 60 kPa, 50 kPa, 40 kPa, 30 kPa, 20 kPa or 10 kPa.

With regard to ionic conductivity, the presence of the microchannels between the ACL and the CCL lead to a voltage drop (voltage losses). For example, for a given applied current density, the porosity and channel hydraulic diameter can be selected/adjusted for the voltage drop between the ACL and the CCL to remain below 500 mV, 400 mV, 300 mV, 200 mV, or 100 mV.

The porosity is understood herein as an area defined by the microchannels and that is available for generation of CO₂ (along with H₂O) and transport of the resulting concentrated CO₂ stream, being further divided by a total area of the ICL. In other words, the porosity can be characterized by a land to channel area ratio. For example, the porosity of the ICL can be between 5% and 95%, between 10% and 80%, between 15% and 70%, between 20% and 60% or between 25% and 50%.

The dimensions of the microchannels can include a pore path width, a channel depth and a hydraulic diameter. The pore path width can be understood as a width of the microchannel taken in a transverse direction of the microchannel and in the plane of the ICL (or in a plane parallel to the ICL plane). The channel depth can be understood as a depth taken in a transverse direction of the microchannel and in a transverse plane that is perpendicular to the plane of the ICL. The hydraulic diameter is as known in the field of uniform and non-uniform cross-section channels, i.e. being generally defined by the following formula: (4× channel cross-sectional area)/(channel perimeter). For example, the channel depth of at most 125 μm, between 10 and 100 μm or between 20 and 80 μm. For example, the pore path width can be between 25 and 150 μm, or between 25 and 75 μm. In some implementations, the cross-section of the microchannels can be circular such that the pore path width is substantially equal to the channel depth. In some implementations, the cross-section of the microchannels can be non-circular including cross-sections of elliptical, circular, rectangular, or square shape. In some implementations, the cross-section of the microchannels can be non-uniform, leading to variations in the pore path width and channel depth along the microchannels.

The network of microchannels can have various geometric patterns, including an interconnected diamond pattern, an interconnected square pattern, or an interconnected circular pattern as exemplified in FIGS. 8A to 8C. The pattern refers herein to the shape defined by microchannels when they are extending across the ICL. The term “interconnected” refers to the ability of the microchannels to be in fluid communication with one another via at least a portion of the pattern unit (diamond, square, circle, etc.) such that the generated concentrated CO₂ stream can flow within the ICL through the microchannels, e.g., from a central region of the ICL to an edge region of the ICL. Experimental results tend to demonstrate that the diamond pattern would allow for a lower internal pressure build-up when compared to other patterns of similar porosity.

Although implementations illustrated in the Figures show formation of the microchannels on the surface of the CCL to form the ICL and the CCL as a one-piece structure, it should be noted that the ICL including the microchannels can alternatively be provided as a separate microporous structure or the microchannels can be formed at a surface of the ACL to produce the MSE having microchannels allowing in-situ regeneration and collection of CO₂.

In some implementations, referring to FIG. 3, the ICL (12) can be formed on the surface of the CCL (8). The expression «on the surface» can be understood as within a surface layer of the CCL and across the depth of this surface layer, such that the ICL and CCL are a one-piece structure. The pattern of the network of the microchannels (14) can be defined by the shape conferred to pillars (16) of the CCL (8) at a surface thereof. For example, multiple spaced-apart diamond-shaped pillars (16) can be formed from the surface of the CCL (8) of the MSE (2), and have defined dimensions (for example, d_p=400 μm and d_q=250 μm) to provide a given porosity to the resulting ICL (12). For example, the network of microchannels (14) and corresponding pillars (16) can be formed on the surface of the CCL (8) via techniques including soft lithography, hot embossing, laser ablation, micro-milling or any combinations thereof. An example implementation of a method to manufacture the MSE is further detailed herein. Such microchannels result from open channels defined on the surface of the CCL or ACL and being further closed to define the hollow path upon contact with the opposed ACL or CCL, i.e. when the MSE is assembled in a zero-gap configuration.

The characteristics of the ICL (including porosity and channel hydraulic diameter, for example) can be tailored to specific operating parameters, including the current density, temperature, a cathode chamber pressure and an anode chamber pressure. For example, for a higher current density operation, more CO₂ and water have to be removed via the integrated microchannels, and thus the ICL can be designed with larger pores.

Although the implementations illustrated in the Figures show microchannels being uniform across the plane of the ICL, it should be noted that the microchannels could be non-uniform across the plane of the ICL, e.g., in terms of channel hydraulic diameter, porosity, and/or geometric pattern of the network. For example, as the fluid flow can increase from the center to the edges/outlet of the ICL, the porosity and/or channel hydraulic diameter can increase from the center to edge regions of the ICL in order to reduce the pressure drop there between.

Referring to FIG. 5, there is provided an MSE electrolyzer comprising the MSE as above described with respect to FIGS. 2 and 3 (an anion conducting layer ACL and a cation conducting layer CCL being modified to form microchannels of an integrated channel layer ICL) in a forward-biased fashion (i.e., with the ACL facing the cathode and the CCL facing the anode). The MSE electrolyzer further includes a cathode (gas diffusion electrode) and an anode for sandwiching the MSE in a zero gap configuration. It is shown that the cathode and anode can be maintained and aligned with adjacent components via a surrounding gasket. The MSE electrolyzer further includes an anode flow field comprising an inlet for being supplied with a liquid anolyte and an outlet for releasing used anolyte. The MSE electrolyzer further includes a cathode flow field comprises a CO₂ inlet for being supplied with a CO₂ gas stream and an outlet for releasing a product mixture including CO₂RR products and CO₂. The cathode flow field further includes secondary outlets for releasing the concentrated CO₂ stream, being a mixture of H₂O and regenerated CO₂.

Although implementations illustrated in the Figures show two secondary outlets on the cathode flow field for the ICL, it should be noted that any number of secondary outlets could be implemented to release the concentrated CO₂ stream from the microchannels. For example, for a cell area above 100 cm², multiple secondary outlets can be defined in the cathode flow field to reduce the pressure drop that may result from the transport of generated CO₂ from one part of the ICL towards a single remote secondary outlet.

Referring to FIG. 5, it is typical to align the MSE such that transport within the ICL is in-plane with gravity to facilitate bubble escape via buoyancy.

As better seen in FIGS. 6A to 6B, the MSE electrolyzer further includes a recycle loop that collects the concentrated CO₂ stream including the regenerated CO₂ from the secondary outlet of the cathode flow field (originating from the integrated channel layer) and recirculates the CO₂ to the primary inlet of the cathode flow field (directed towards the cathode compartment). To do so, the recycle loop includes a liquid-gas separator to separate the concentrated CO₂ stream into a CO₂ gas stream and a liquid water stream, such that only the separated CO₂ gas stream is recycled to the cathode flow field to maximize CO₂ utilization.

Although the implementations illustrated in the Figures show no secondary inlets on the cathode flow field, it should be noted that any number of inlets could be implemented to supply, for example, a purge fluid into the ICL. The purge fluid, gas or liquid, could flow through the ICL via the microchannels to prevent any build-up of CO₂ and/or water bubbles. A secondary inlet could also be provided on the cathode flow field to circulate the captured CO₂ back to the cathode flow field but as a separate stream from the CO₂ feedstream that is fed via the main inlet of the cathode flow field.

Salt Precipitation Prevention

Alkali metal cations can be viewed as essential for CO₂RR because they modulate the cathode catalyst interface and stabilize negatively-charged reaction intermediates (e.g., CO₂ (_ads)″), thereby promoting CO₂RR. In zero-gap systems, alkali metal cations in the anolyte can migrate through the anion conducting layer. However, the formation of salt precipitates is inevitable from the reaction of alkali metal cations and (bi) carbonate. It is known that the gradual accumulation of salt blocks transport, degrades performance and is a barrier to stable CO₂RR operation.

Contrary to conventional systems relying on mobile alkali metal cations in aqueous electrolytes, the present MSE can further include fixed cations that are part of the anion conducting layer to prevent CO₂ loss by salt formation from free alkali metal cations and (bi) carbonates.

Referring to FIG. 7, the fixed cations from the ACL (quaternary ammonium cation containing N⁺, labelled stable cation) can modulate the surface of the adjacent cathode (including the Cu substrate catalytic layer). Various fixed cations can be provided by changing the nature of the anion conducting layer. For example, the fixed cations can be piperidinium, a quaternary ammonium cation contained within a poly(aryl piperidinium) ACL, commercially known as PiperION. The quaternary ammonium piperidinium cation was shown to have excellent stability in alkaline conditions and a high ionic conductivity of hydroxides and (bi) carbonates. These cations are known in conventional anion-conducting solid electrolyte layers for fuel cell applications which, due to their hydrophobic backbones, achieve relatively low water uptake and mitigate catalyst flooding. Water management can be crucial for efficient CO₂RR as a flooded catalyst layer can introduce CO₂ mass transport losses and harm the energy efficiency.

Other commercially available anion-exchange membranes including Sustainion, Aemion/Aemion+, Fumasep, Pention, and Selemion can be used as the ACL in the present MSE. For example, the fixed cations can thus be piperidinium, imidazolium or benzimidazolium.

Example Implementation of the MSE Fabrication

The MSE includes a «three-layer» structure made of the anion conducting layer (ACL), the integral channel layer (ICL), and the cation conducting layer (CCL). The ICL is not limited to be a defined layer that is separate from the ACL or CCL, and can be integrated, i.e., forms a one-piece structure, with the CCL, for example. Various ways can be employed to define the microchannels in the surface of the CCL and thus form the ICL, such that the ICL is integral with the CCL.

It should be noted that the «cation conducting layer» may be interchangeably referred to as a «cation exchange membrane», and the «anion conduction layer» may be interchangeably referred to as an «anion exchange membrane».

Referring to an example illustrated in FIGS. 4A to 4F, fabrication of the MSE can include formation of internal microfluidic channels at the surface of a cation conducting layer 8 to form the ICL 12. The ICL 12 is formed integral with the cation conducting layer 8, thereby defining the microchanneled membrane 13 (also referred to as the cation exchange membrane with micro flow channels 13 in FIG. 5). A master 7 for the microfluidic channels was prepared by spin-coating positive photoresist SU-8 2050 (Microchem, USA) with a height of 50±5 μm on a silicon wafer. The master was patterned with designed high-resolution transparency masks (CAD/Art Services, Inc., USA) following standard photolithography procedures. The cation conducting layer 9, Nafion™ 117 (Fuel Cell Store, USA), was hot embossed with the prepared master molds under a temperature of 220° F. and pressure of 1.25 MPa for 5 minutes to form the ICL at the surface of the CCL. The modified Nafion™ 117 layer was cleaned by immersing in a 3% wt H₂O₂ solution at 80° C. for 60 min and rinsed in DI water at 80° C. for 60 min. Then, the modified layer was soaked in 1 M H₂SO₄ at 80° C. for 60 min, and finally rinsed in DI water at 80° C. for 60 min to remove H₂O₂ and H₂SO₄ residues. Anion conducting layers (PiperION TP-85, 40 μm, Versogen, USA), Sustainion (Sustainion® X37-50 Grade RT, Dioxide Materials, USA) and Aemion (AF1-HNN8-50-X, lonomr Innovations Inc., Canada) were used. To finalize the MSE, the anion conducting layers were soaked in 1 M KOH at for 5 hours at room temperature for activation, and then were rinsed in DI water to remove KOH residues 3 times. The formed layers (ACL and the ICL-modified CCL) were then assembled in a zero gap configuration with the ICL interface being sandwiched between the ACL and the CCL.

Example Implementation of Electrode Preparation

Various anodes and cathodes can be used in combination with the MSE to form the MSE electrolyzer. The anode includes an anodic catalyst that can be iridium, ruthenium, rhodium, platinum, gold, or combinations thereof. The cathode includes a cathodic catalyst that can be copper, silver, gold, zinc, bismuth, tin, iron, nickel, aluminum, palladium, or combinations thereof.

For example, the cathode and anode that were used for the below reported experimentation were produced as follows. A polytetrafluoroethylene (PTFE) based copper cathode was prepared by plasma sputtering. Approximately 300 nm of copper catalyst was sputtered onto the PTFE substrate using an AJA International ATC Orion 5 Sputter Deposition System (Toronto Nanofabrication Centre, Canada). The anode electrode was prepared by spray-coating iridium chloride on platinized titanium felt (Fuel Cell Store, USA) followed by thermal decomposition. An 8×8 cm² area of titanium felt was etched in boiling 0.5 M oxalic acid for 30 minutes. The etched titanium felt was then spray coated on a hot plate held at 80° C. with a solution consisting of 200 mg IrCl₃XH₂O (99.8%, Alfa Aesar, USA) dissolved in 13 mL of ethanol. The titanium felt coated with IrCl₃ was then calcined at 500° C. for 10 minutes, and cut to size to obtain the final anode electrode.

Operation of an MSE Electrolyzer

Several operation parameters can be controlled to enhance the operation of the MSE electrolyzer and minimize CO₂ losses.

For example, it was demonstrated that lowering the CO₂ feed rate/concentration that is fed to the cathode inlet led to an increase in selectivity towards ethylene. In some implementations, the CO₂ feed rate can be adjusted/controlled between 0.25 sccm·cm⁻² and 2 sccm·cm⁻², 0.5 sccm·cm⁻² and 1.5 sccm·cm⁻², or 0.8 sccm·cm⁻² and 1 sccm·cm⁻²

In another example, it was demonstrated that operating at a controlled current density allows for maintaining stability of the MSE electrolyzer over an extended period with reduced CO₂ losses. In some implementations, the current density can be adjusted/controlled within an industrially relevant range that is between 40 mA·cm⁻² and 240 mA·cm⁻², 50 mA·cm⁻² and 200 mA·cm⁻², 60 mA·cm⁻² and 160 mA·cm⁻², 80 mA·cm⁻² and 120 mA·cm⁻², or 90 mA·cm⁻² and 100 mA·cm⁻².

In some implementations, operating the MSE electrolyzer can include continuously or periodically purging the microchannels by circulating a purging fluid through the microchannels of the ICL to prevent any build-up of CO₂ and/or water bubbles. For example, the purging fluid can be any gas or liquid including water and/or CO₂.

When using the present MSE including the integrated microchannels and fixed poly(aryl piperidinium) cations on the ACL, the CO₂ losses can be controlled to at most 5%, 4%, or 3% of the CO₂ converted in the reactor, with a CO₂RR faradaic efficiency (FE) of at least 60%, 70%, 80%, 90% or 95%, a C₂H₄FE of at least 10%, 20%, 30%, 40%, 50% or 60%, and a C₂₊ energetic efficiency (EE) that can be of at least 5%, 10%, 15%, 20% or 25%, during at least 100, 110, 120, 130, 140, 150, 160, 170, 180, 190 or 200 hours of stable operation.

Although implementations of the MSE are described in relation with operation of a CO₂RR system based on oxygen evolution reaction, it should be noted that the MSE as described herein can be part of an electrolyzer system operating other anodic reactions which also create gas phase products (or use gas phase reactants), such as the hydrogen oxidation reaction or the chlorine evolution reaction.

Experimental Results and Discussion

Geometry of the Network of the Microchannels of the MSE

As seen on graphs of FIGS. 8A to 8C, simulations were performed according to a computational fluid dynamics model to evaluate the pressure build-up of different channel pattern designs including interconnected diamond, square, and circular patterns. All pattern designs formed a network of microchannels providing a same porosity to the ICL (i.e., having a same land to channel area ratio), with a pore path width of 75 μm for a 1 cm² electrode area. More particularly, graphs of FIGS. 8A to 8C map the pressure distribution over the surface of the ICL according to the pattern design. It was observed that the diamond pattern exhibited a lower pressure build-up than the other microstructures of similar porosity.

For example, the interconnected channel pattern (diamond pillars, d_p=350 μm, d_q=250 μm) can be formed on the surface of the cation-conducting layer CCL via soft lithography and hot embossing, to produce the ICL-modified CCL. Referring to graphs of FIGS. 9A to 9D and corresponding Table S8, the mass transfer resistance of the diamond-patterned microchannels having a pore path width between 25 μm and 150 μm was analyzed using computational fluid dynamics modelling. Pore path widths over 75 μm resulted in acceptably low two-phase flow resistance and internal pressure build up: 0.227 kPa for a 1 cm² area (also referred to a maximum pressure drop from center to edges of the ICL), and 44.1 kPa for an industrial scale 100 cm² area. Simulations of internal CO₂ regeneration in the limiting case of a large-area liquid-filled layer resulted in maximum pressures well below the yield strength of commercial solid anion-conducting layers. It was experimentally shown that smaller microchannels led to repeated cycles of pressure build-up and release. As apparent from FIG. 11, these pressure oscillations are visible on the voltage response and can hamper long-term electrolyzer operation, particularly for the ICL including microchannels having a channel depth of about 25 μm. As pressure builds in the ICL, the pressure increase can push the ACL apart from the ICL-modified CCL, which results in a loss of interfacial contact between the layers and in an increase in ionic resistance. The full cell voltage signal (e.g., between cathode and anode) thus oscillates as the voltage drop at the ICL-ACL interface increases.

TABLE S8
Computational fluid dynamic modelling of the mass transfer resistance
in the variable pore path size integrated channels.
	Reactor Size		Pore Path Size	Maximum Pressure Drop
1	cm2	25	μm	1.66	kPa
1	cm2	75	μm	0.227	kPa
1	cm2	150	μm	0.072	kPa
100	cm2	75	μm	44.1	kPa

Referring to FIGS. 12A to 12C, a pilot-scale 50 cm² system, shown FIGS. 12A and 12B, was also tested and demonstrated that stable operation and low anodic CO₂ loss were achieved as seen in FIG. 12C. The pilot-scale 50 cm² electrolyzer was tested integrating an MSE with 0.01 M H₂SO₄ anolyte, 100 mA cm⁻², and a 3.6 sccm cm⁻² CO₂ feed rate. The cathode consisted of 200 nm of sputtered Cu deposited on carbon paper (AvCarb MGL190). The full cell voltage of the 50 cm² cell fluctuated between −4.6 to −4.8 V. The voltage was higher than expected in this preliminary experiment since gas build-up caused delamination of the layers and increased the ionic resistance of the MSE. Improved microchannel flow will be realized via future improvements in cell design, cell assembly, and MSE manufacture. In addition, as seen from the optical microscope images of the microchannels of the ICL shown in FIGS. 13A to 13C, there was no detectable change in the ICL, before and after CO₂ electrolysis.

Comparative Experimental Results: Conventional CO₂RR System Vs. CO₂RR MSE System

Numerical multiphysics simulations were performed to better understand CO₂ crossover within a conventional system including a ACL as seen in FIG. 1, FIG. 14, or in FIGS. 15A and 15B). More particularly, the CO₂ reduction reaction (CO₂RR) and the oxygen evolution reaction (OER) on the cathode and anode were modelled, respectively, in one dimension (1D) using COMSOL Multiphysics version 5.5. COMSOL's Transport of Diluted Species and Secondary Current Distribution physics modules were applied to model the interactions between aqueous CO₂, HCO₃⁻, CO₃²⁻, H⁺, OH⁻ and SO₄²⁻ in a time-dependent study. Building upon recent modelling work (McCallum, C., Gabardo, C. M., O'Brien, C. P., Edwards, J. P., Wicks, J., Xu, Y., Sargent, E. H., and Sinton, D. (2021). Reducing the crossover of carbonate and liquid products during carbon dioxide electroreduction. Cell Reports Physical Science, 100522), the model included an acidic environment (0.01 M H₂SO₄ anolyte) and the physical mechanism of gaseous CO₂ bubbling out when it is oversaturated. The aqueous CO₂ concentration in the electrolyte was limited to the solubility limit.

CO₂ Solubility

The CO₂ solubility in pure water was governed by Henry's Law (E7-E8). The CO₂ was assumed to be an ideal gas, and the solubility determined by pressure, temperature, and salinity effects. A series of Sechenov equations were used to determine the solubility as a function of the solution salinity (E9-E11). As the salt concentration increased, the resultant CO₂ solubility decreased in the electrolyte.

$\begin{array}{cc}{\left[{\mathrm{CO}}_2\right]}_{\mathrm{aq},0}={K_0\left[{\mathrm{CO}}_2\right]}_g& \left(\mathrm{E7}\right)\end{array}$ $\begin{array}{cc}\ln K_0=93.4517\left(\frac{100}{T}\right)-60.2409+23\text{.3585}\ln \left(\frac{T}{100}\right)& \left(\mathrm{E8}\right)\end{array}$ $\begin{array}{cc}\log \left(\frac{{\left[{\mathrm{CO}}_2\right]}_{\mathrm{aq},0}}{{\left[{\mathrm{CO}}_2\right]}_{\mathrm{aq}}}\right)=K_s{C}_s& \left(\mathrm{E9}\right)\end{array}$ $\begin{array}{cc}{K}_s=\sum \left(h_{\mathrm{ion}}+h_G\right)& \left(\mathrm{E10}\right)\end{array}$ $\begin{array}{cc}{h}_G=h_{G,0}+h_T\left(T-298.15\right)& (\mathrm{E11})\end{array}$

The Sechenov constant (K_s) and molar concentration (C_s) of an electrolyte solution were calculated. The aqueous CO₂ solubility was calculated depending on the concentration of OH⁻, CO₃₂⁺, and HCO₃⁻; ions. The specific parameters are shown in Table S1 below.

TABLE S1
Corresponding Sechenov constants13
Ion           hion
OH−           0.0839
CO22−         0.1423
HCO3−         0.0967
hG, 0 for CO2 −0.0172
hT for CO2    −0.000338

Carbonate Equilibrium and Electrolyte Equilibrium

The chemical reactions predicted a steady-state equilibrium of aqueous CO₂, HCO₃⁻, CO₃²⁻, H⁺, and OH (E3-E6). The water dissociation (E12) and the dissociation of H₂SO₄ electrolyte (E13) were also considered in this system. The dependence of the rate constants on temperature and salinity was implemented as found in literature¹⁴. The corresponding equations are listed below:

$\begin{array}{cc}{H}_{2}O\leftrightarrow H^{+}+{\mathrm{OH}}^{-}& \left(\mathrm{E12}\right)\end{array}$ $\begin{array}{cc}{H}_2{\mathrm{SO}}_4\leftrightarrow 2H^{+}+{\mathrm{SO}}_4^{2-}& \left(\mathrm{E13}\right)\end{array}$

Ohm's Law and Poisson Equation

Ohm's Law is applied to the CO₂ electrolyzer (E14). The conductivities of different electrodes and electrolytes are considered, and the corresponding values are listed in Table S2

$\begin{array}{cc}i=-\sigma \frac{\partial \varphi }{\partial x}& \left(\mathrm{E14}\right)\end{array}$

Where σ is the electrical/ionic conductivity of different media.

TABLE S2
Electrical/ionic conductivity of different domains
Domain                           Electrical/ionic conductivity [S/m]
Anion-conducting Layer           8.0
Integrated Channel Layer         16.20
Cation-conducting Layer          24.92
Solid-phase of IrOx Anode Catalyst 1.4e7
Liquid-phase of IrOx Anode Catalyst 0.0657

The Poisson equation (E15) determined the combination of electroneutrality and the induced space charge for ion-exchange membranes.

$\begin{array}{cc}{ϵ}_0{ϵ}_r\frac{\partial \psi }{\partial x}=F\sum z_i{c}_i+{\rho }_{\mathrm{aem}}& \left(\mathrm{E15}\right)\end{array}$

The ∈_o and ∈_r represent the permittivity of vacuum and relative permittivity of water, respectively. The y is the combination of electroneutrality and induced space charge for the ion-exchange membrane. The P_aem represents the space charge for the membrane, and it exists exclusively in the membrane domain. The corresponding values are listed in Table S3.

TABLE S3
Parameters for AEM
Parameters                     Value (unit)
Permittivity of vacuum         8.8542e−12 (F/m)
Relative permittivity of water 80 (1)
Membrane space charge          1 (M)

The amalgamation of electrolyte potential, electroneutrality, and the induced space charge for membranes induced the electromigration effect of the charged species (CO₃²⁻, HCO₃⁻, H⁺, OH⁻ and SO₄²⁻) (E16).

$\begin{array}{cc}V={\varphi }_l+\psi & \left(\mathrm{E16}\right)\end{array}$

Catalyst Electrochemical Reactions:

The cathode catalyst layer reaction produces CO, H₂, C₂H₄, ethanol, formic acid, and acetic acid (E17-E22). Oxygen is evolved at the anode catalyst layer (E23). The Faradaic efficiency and corresponding partial current density for each reaction are listed in Table S4.

$\begin{array}{cc}\mathrm{HER}& \\ 2H_{2}O+2e^{-}\to H_2+2{\mathrm{OH}}^{-}& \left(\mathrm{E17}\right)\end{array}$ $\begin{array}{cc}{\mathrm{CO}}_2\mathrm{RR}& \\ 2H_{2}O+2e^{-}\to H_2+2{\mathrm{OH}}^{-}& \end{array}$ $\begin{array}{cc}{\mathrm{CO}}_2+H_{2}O+2e^{-}\to \mathrm{CO}+2{\mathrm{OH}}^{-}& \left(\mathrm{E18}\right)\end{array}$ $\begin{array}{cc}2{\mathrm{CO}}_2+8H_{2}O+12e^{-}\to C_2{H}_4+12{\mathrm{OH}}^{-}& \left(\mathrm{E19}\right)\end{array}$ $\begin{array}{cc}2{\mathrm{CO}}_2+9H_{2}O+12e^{-}\to C_2{H}_{5}OH+12{\mathrm{OH}}^{-}& \left(\mathrm{E20}\right)\end{array}$ $\begin{array}{cc}{\mathrm{CO}}_2+2H_{2}O+2e^{-}\to CHOOH+2{\mathrm{OH}}^{-}& \left(\mathrm{E21}\right)\end{array}$ $\begin{array}{cc}2{\mathrm{CO}}_2+6H_{2}O+8e^{-}\to C{H}_3\mathrm{COOH}+8{\mathrm{OH}}^{-}& \left(\mathrm{E22}\right)\end{array}$ $\begin{array}{cc}\mathrm{OER}& \\ 2H_{2}O\to O_2+4H^{+}+2e^{-}& \left(\mathrm{E23}\right)\end{array}$

TABLE S4
Faradaic efficiency and partial current density
	Reaction	FE (%)	Partial Current Density (mA/cm2)
	COER	60	60
	C2H4ER	20	20
	HER	5	5
	EtOHER	9	9
	HCOOHER	3	3
	CH3COOHER	3	3

Species Transport

Diffusion and electromigration were considered for all species according to the Nernst-Planck set of equations in all domains (E24, E25). The effective diffusivity of species was governed by the Millington and Quirk set of equations. (E26, E27) All layers except the electrolyte diffusion boundary were considered as porous domains.

$\begin{array}{cc}\frac{\partial c_i}{\partial t}=\sum R_i-\frac{\partial j_i}{\partial x}& \left(\mathrm{E24}\right)\end{array}$ $\begin{array}{cc}{J}_i=-\frac{D_i\partial c_i}{\partial x}-\frac{z_i{D}_i}{\mathrm{RT}}{\mathrm{Fc}}_i\frac{\partial V}{\partial x}& \left(\mathrm{E25}\right)\end{array}$ $\begin{array}{cc}{D}_i=\frac{{ϵ}_p}{{\tau }_{F,i}}{D}_{F,i}& \left(\mathrm{E26}\right)\end{array}$ $\begin{array}{cc}{\tau }_{F,i}={ϵ}_p^{-1/3}& \left(\mathrm{E27}\right)\end{array}$

Where C_i, D_i, ∈_p, τ_F,i and z_i are the species concentration, diffusion coefficient, charge number, porosity coefficient, and tortuosity coefficient, respectively. The corresponding diffusion coefficient and charge number are listed in Table S5. The corresponding porosity coefficients of different domains are listed in Table S6.

TABLE S5
Diffusion coefficients and charge in the system17-19
	Species	Di(m2s−1)	zi (−)
	CO2	1.91e−9	0
	CO32−	0.923e−9	−2
	HCO3−	1.185e−9	−1
	H+	9.31e−9	+1
	OH−	5.26e−9	−1
	SO42−	1.07e−9	−2
	H2SO4	2.31e−9	0
	H2O	2.57e−9	0

TABLE S6
Porosity coefficient of different domains
Domain                   Porosity Coefficient
Anion-conducting Layer   0.1
Integrated Channel Layer 0.35
Cation-conducting Layer  0.1
IrOx Anode Catalyst      0.95

Referring to FIG. 14 and FIGS. 15A and 15B, the anodic oxygen (O₂) evolution reaction makes the anode locally acidic and shifts the (bi) carbonate equilibria towards dissolved gaseous CO₂. The dissolved CO₂ then exceeds the solubility limit in the liquid anolyte and gaseous CO₂ comes out of the solution, mixing with the generated O₂. It is noted that this mixing of the CO₂ with anode-produced O₂ results in costly downstream separation.

Based on the same model hypothesis and software, numerical multiphysics simulations were performed for a one dimension (1D) microchanneled solid electrolyte (MSE) with internal CO₂ capture domain, according to an implementation of the present invention. Referring to FIG. 16, the geometry consisted of a cathode catalyst layer (Cathode), an anion-conducting layer (ACL), an integrated channel layer (ICL), a cation-conducting layer (CCL), a porous iridium oxide anode catalyst layer (Anode), and an anolyte layer (Anolyte). The microchannel pattern was fabricated on one side of the CCL. A constant CO₂ concentration was applied to the cathode boundary point, whereas equilibrium concentrations of H⁺ and SO₄²⁻ were specified at the anolyte boundary point. A constant current density was applied to the cathode catalyst layer with an electrical ground assigned to the right-hand boundary of the anode layer.

Based on the hypothesis that the anion-conducting layer would achieve high pH conditions at the cathode favourable for CO₂RR and the cation-conducting layer would provide proton transport and an acidic pH for internal CO₂ regeneration (<4 pH), the transport of (bi) carbonates and protons to the internal CO₂ regeneration domain was modelled at the interface of the anion- and cation-conducting layers.

Referring to FIG. 17 and FIGS. 18A and 18B, the simulations indicate that under operating conditions, the ICL would be sufficiently acidic to transform incoming (bi) carbonates to CO₂, exceeding the solubility limit and causing gas-phase CO₂ evolution. For example, the cation-conducting layer can provide proton transport to achieve a pH of at most 4, and optionally between 2.5 and 4.

CO₂RR MSE Electrolyzer Operation

To test internal CO₂ regeneration experimentally, a pH-swing CO₂ capture domain at the interface of anion- and cation-conducting layers with integrated channels for CO₂ extraction was built. Structuring the channels in the cation-conductive layer was preferable due to the mechanical and thermal stability of the material, and the conductivity is higher than that of the anion-conductive layer. The assembly was integrated into a zero-gap CO₂ electrolyzer making use of a copper catalyst on the cathode and iridium oxide catalyst on the anode.

	TABLE S7(A)
	Integrated Channel Layer	Anode
Current Density				CO2RR				CO2RR
(mA cm−2)	CO2	Hydrogen	Oxygen	Products	CO2	Hydrogen	Oxygen	Products
−40	97.52% ±	0.74% ±	1.72% ±	0.02% ±	6.72% ±	0.14% ±	93.14% ±	0.01% ±
	0.59%	0.13%	0.15%	0.01%	0.52%	0.04%	0.56%	0.00%
−80	97.59% ±	0.74% ±	1.62% ±	0.06% ±	6.94% ±	0.10% ±	92.95% ±	0.01% ±
	0.94%	0.11%	0.13%	0.02%	0.41%	0.03%	0.43%	0.00%
−120	97.74% ±	0.59% ±	1.58% ±	0.09% ±	7.00% ±	0.08% ±	92.91% ±	0.01% ±
	0.82%	0.09%	0.10%	0.02%	0.40%	0.04%	0.43%	0.00%
−160	97.98% ±	0.65% ±	1.24% ±	0.13% ±	7.46% ±	0.06% ±	92.47% ±	0.01% ±
	0.90%	0.14%	0.20%	0.02%	1.37%	0.05%	1.40%	0.00%
−200	97.15% ±	0.74% ±	0.95% ±	0.15% ±	6.56% ±	0.04% ±	93.39% ±	0.00% ±
	1.07%	0.12%	0.13%	0.01%	0.57%	0.02%	0.55%	0.00%
−240	97.26% ±	1.24% ±	1.30% ±	0.21% ±	6.90% ±	0.03% ±	93.05% ±	0.002% ±
	1.42%	0.16%	0.20%	0.04%	0.72%	0.02%	0.74%	0.01%

TABLE S7(B)
Current Density	CO2 Consumption Rate (sccm cm2)
(mA cm−2)	Anodic	Internally Captured	Electroreduced
−40	0.011 ± 0.002	0.082 ± 0.014	0.230 ± 0.046
−80	0.022 ± 0.005	0.238 ± 0.030	0.426 ± 0.055
−120	0.035 ± 0.024	0.401 ± 0.083	0.562 ± 0.040
−160	0.049 ± 0.009	0.544 ± 0.110	0.674 ± 0.195
−200	0.051 ± 0.015	0.654 ± 0.070	0.686 ± 0.058
−240	0.062 ± 0.019	0.781 ± 0.096	0.727 ± 0.145

Tables S7 (A) and S7 (B) provide results with respect to gas leaving from the CO₂RR MSE electrolyzer with a CO₂ feed rate of 20 sccm cm⁻² into the electrolyzer and 0.01 M H₂SO₄ anolyte. Table S7 (A) provides the composition of the gas from integrated channel layer and anode; and Table S7 (B) provides the CO₂ consumption rate from different sources (error bars represent the standard deviation of three independent measurements).

Referring to FIG. 19, when the system was operated at current densities ranging from 40 to 240 mA cm⁻², the composition of the gas leaving the microchannels was constant at 98% v/v CO₂. Also present were traces of H₂ and O₂, owing to diffusion from the cathode and anode, respectively. Referring to FIG. 20, analysis of the anodic tail gas indicated a CO₂ concentration of ˜7% v/v, independent of current density. This fraction of CO₂ diffuses to the anode in dissolved form, due to concentration gradients between the interface and the anode. Referring to FIG. 21, based on the measured stream compositions, all CO₂ pathways (electroreduced, internally regenerated/recirculated, and released at the anode) were quantified over the range of current densities. In all cases, the fraction of CO₂ converted in the reactor which is lost to the anode tail gas is between 3% and 4%, as compared to 70-95% in conventional electrolyzers, indicating that internal capture can reduce CO₂ crossover to the anode by more than 20 times.

Theoretical Voltage Penalty Calculation from Implementing MSE

Referring to FIG. 23, a CO₂RR MSE electrolyzer was operated at cell voltages from −3.6 V to −4.3 V to reach current densities from 40 cm⁻² to 240 mA cm⁻², respectively. Referring to FIGS. 22A and 22B, it was observed that the MSE had comparable cell voltages (e.g., within 100 mV at current densities >120 mA cm⁻²) to conventional anion-conducting solid electrolytes (anion conducting layer, ACL) with near-neutral anolytes, but with much lower anodic CO₂ loss.

The theoretical voltage penalty from implementing MSE was obtained by comparison with a conventional forward-bias bipolar system (i.e., with the ACL facing the cathode and the CCL facing the anode) that does not have internal microchannels. The voltage penalty can be explained by ionic conductivity changes resulting from adding the microfluidic channels. Comparing with the same thickness non-porous proton exchange layer (conductivity provided in Table S2), the microchannel domain was implemented with microchannels covering 35% of the total area (Table S6), which caused a decrement of the ionic conductivity by 35%. The voltage penalty of the conductivity changes at the different electrolyzer operation current densities but can be quantified by the following equation:

$\begin{array}{cc}\Delta V_i=\left(\frac{1}{{\sigma }_{\mathrm{nonporous}}}-\frac{1}{{\sigma }_{\mathrm{microchannel}}}\right)\frac{\mathrm{iL}}{A}& \left(\mathrm{E28}\right)\end{array}$

where ΔV_i is the voltage penalty at the electrolyzer operation current density i. σ_nonporous and σ_microchannel are the ionic conductivities of the non-porous proton exchange layer and microchannel domain, respectively. L is the thickness of the microfluidic channels domain, and A is the unit conduction surface area (1 cm²) of the microchannel domain.

Referring to FIG. 23, the voltage penalty due to narrowing the ionic transport path through the integrated channel layer was minimal, due to the thin channel geometry and the relatively high ionic conductivity of the cation-conducting layer. The voltage penalty of integrated microchannels to the cation-conducting layer was calculated to be <1% of the full cell voltage.

To verify this experimentally, an electrolyzer using the same materials but without integrated microchannels was operated, and the cell voltages were found to be indistinguishable within measurement error, further confirming that the CO₂ capture microchannels do not present a significant voltage penalty as seen in FIG. 24D. However, in absence of the integrated microchannels, the CO₂ formed at the junction of the cation and anion-conducting layers could not escape, as represented in FIG. 24A. The trapped CO₂ caused the solid electrolyte to delaminate, resulting in less than 1 hour of stable operation, as shown in FIGS. 24B and 24C, which is consistent with other literature reports using conventional forward-bias bipolar configurations.

For further comparison, another conventional electrolyzer including solid-state electrolyte packed beds between the ion-conducting solid electrolyte layers, as represented in FIG. 25A, can allow CO₂ to escape. However, referring to FIG. 25B, higher cell voltages (e.g., ˜700 mV, or ˜20%, more at 200 mA cm⁻²)—than in the case of an electrolyzer including an anion conducting layer-were required due to higher electrolyte and interfacial losses in the cell. The integrated channel layer (ICL) is thus shown to facilitate effective removal of CO₂ without stability or cell voltage penalty.

Salt Precipitation

The CO₂RR MSE electrolyzer was tested with a small amount of potassium (0.1 M potassium sulphate, K₂SO₄) added to the electrolyte (0.01 M H₂SO₄) part way through the experiment. The results, observed in FIG. 26A, demonstrate that the CO₂RR performance degraded rapidly upon adding the potassium cations, with H₂ evolution dominating within 1 hour. Upon disassembly of the electrolyzer, salt precipitates were visible on the cathode as seen in FIG. 26B, signifying that alkali metal cations migrated through both solid electrolyte layers. Similar to other CO₂RR systems, free alkali metal cations in the electrolyte of a MSE electrolyzer limit stability.

The CO₂RR MSE-electrolyzer including fixed poly(aryl piperidinium) as seen in FIG. 7 was shown to achieve CO₂RR in the absence of free alkali metal cations in the anolyte, thereby avoiding salt precipitation and presenting a route to long-term stability. More particularly, when operating without free alkali metal cations and using the MSE in the electrolyzer assembly, the selectivity towards CO₂RR products was maintained above 80% in the range of 40-240 mA cm⁻², reaching a maximum of 95%, as seen in FIG. 27 and corresponding Table S9.

TABLE S9
CO2RR product distribution at different current densities (0.01M
sulphuric acid anolyte with a CO2 feed rate of 20 sccm cm−2).
Current
Density	Faradaic Efficiency (%)
(mA cm−2)	H2	CO	CH4	C2H4	Formate	Acetate	Ethanol	N-propanol
−40	5.0% ±	74.6% ±	0.4% ±	12.0% ±	3.21% ±	0.69% ±	2.94% ±	0.29% ±
	0.60%	1.47%	0.12%	0.65%	1.06%	0.35%	1.52%	0.06%
−80	3.55% ±	64.31% ±	0.38% ±	19.01% ±	4.23% ±	1.29% ±	6.26% ±	0.46% ±
	0.63%	3.47%	0.12%	2.28%	0.84%	0.60%	1.31%	0.07%
−120	4.74% ±	35.40% ±	0.70% ±	28.70% ±	7.20% ±	4.80% ±	16.30% ±	1.50% ±
	1.72%	2.85%	0.12%	3.61%	2.44%	0.88%	3.86%	0.25%
−160	4.79% ±	25.00% ±	1.11% ±	35.88% ±	6.90% ±	5.40% ±	18.40% ±	1.30% ±
	1.51%	2.26%	0.38%	2.96%	1.12%	1.28%	2.03%	0.29%
−200	9.68% ±	15.45% ±	2.46% ±	36.54% ±	5.70% ±	7.50% ±	21.20% ±	1.40% ±
	1.27%	1.71%	0.40%	2.87%	1.23%	3.60%	2.48%	0.46%
−240	21.76% ±	12.24% ±	3.63% ±	30.68% ±	4.00% ±	7.30% ±	19.30% ±	1.00% ±
	2.43%	1.14%	0.97%	5.85%	0.66%	1.65%	3.19%	0.42%

The low H₂ FE demonstrates that the cathodic environment, modulated by the fixed piperidinium cation of the anion conducting layer, was sufficiently alkaline to suppress H₂ production. Referring to FIG. 27, at low currents, carbon monoxide was the dominant product (above 60% at 40 mA cm⁻²). Selectivities towards C₂H₄ and ethanol increased with current and at 200 mA cm⁻² reached 35% and 21%, respectively.

To assess the validity of an alkali-metal-free anolyte scenario, a control test with pure water anolyte was performed. Referring to FIGS. 28A and 28B, the selectivity is shown to be comparable to the case of a 0.01 M sulphuric acid (H₂SO₄) anolyte. Analysis of the MSE outlet and anolyte outlet streams was further performed and results are shown in Table S10 below, which further confirmed a lack of alkali metal cations present (<2 ppm).

TABLE S10
The concentration of common alkali metal cations
in the cell determined by ICP-OES.
	Li	Na	K	Cs
Cathode and	0.002 ppm	0.804 ppm	0.635 ppm	<0.001 ppm
MSE outlet
Anolyte	0.001 ppm	1.468 ppm	0.336 ppm	<0.001 ppm

Similar tests were run with 0.01 M H₂SO₄ anolyte and 20 sccm CO₂ flow rate for alternative fixed cations provided by other anion-conducting layers. Referring to FIGS. 29A to 29F, the tests resulted in successful CO₂RR without free alkali metal cations in the anolyte.

In some implementations, the anolyte can thus be free of mobile alkali metal cations, and the anolyte can for example be water or a solution of H₂SO₄, HClO₄, or a combination thereof. When using an acid solution, the solution can have a concentration of at most 0.5 M, for example from 0.005 to 0.05 M.

During the tests performed on the CO₂RR MSE electrolyzer with 0.01 M H₂SO₄ anolyte and 20 sccm CO₂ flow rate, the CO₂ captured from the ICL was redirected back to the cathode inlet. Due to the high purity of the captured CO₂, upon comparison to a system fed exclusively with a fresh CO₂ stream, no observable performance difference was observed when the system was co-fed with the internally regenerated CO₂ stream, as seen in FIGS. 30A to 30C (FIG. 30B showing performance with captured CO₂ recycle, and FIG. 30C showing performance without recycle).

Further tests were performed to provide highly alkaline conditions and achieve the rate-determining step for C₂⁺ production, C—C coupling, at lower overpotentials. Lowering the CO₂ concentration was performed to lessen (bi) carbonates formation and increase the local pH. Referring to graphs of FIGS. 31, 32A to 32C and 33A to 33C, simulations indicated that lowering the CO₂ concentration from 100% to 10% would increase the cathode pH by 1.3 to ˜14.7—the pH equivalent of 5 M potassium hydroxide. It should be noted that PCL in the graphs refers to a proton conducting layer acting as the CCL.

Beyond pH effects, the hypothesis that lower partial pressures of CO₂ (or CO) reactant yield favourable reaction kinetics was verified. The CO₂ supply to the reactor was reduced, enabling a greater fraction of CO₂ to be consumed and greater dilution of CO₂ with gas products, all while recirculating internally regenerated CO₂. Referring to FIG. 34 and corresponding Table S11, decreasing the flow rate resulted in a substantial increase in C₂H₄ production at the expense of carbon monoxide, from a C₂H₄ selectivity of 12% at 20 sccm cm 2 to 51% at 0.8 sccm cm⁻². Decreasing the flow rate further increased H₂ selectivity due to CO₂ mass transport limitations. Referring to FIG. 35, the single-pass CO₂ utilization (black squares), calculated using the fraction of the incoming CO₂ supply that is converted to CO₂RR products, increased monotonically with decreasing flow rate, up to 83% at 0.25 sccm cm⁻² and 100 mA cm⁻². The highest C₂⁺ energetic efficiencies (EEs) of ˜25% were obtained at flowrates 0.8 and 1.0 sccm cm⁻², but operation at a flowrate of 1.0 sccm cm⁻² yielded less H₂ and a higher CO₂RR EE.

TABLE S11
CO2RR Product distribution for different CO2 supply rates. (0.01M H2SO4 anolyte, 100 mA cm−2)
CO2 Inlet Faradaic Efficiency (%)
Flow-rate	Faradaic Efficiency (%)
(sccm cm−2)	H2	CO	CH4	C2H4	Formate	Acetate	Ethanol	N-propanol
20	4.60% ±	6.46% ±	1.39% ±	11.66% ±	7.20% ±	2.31% ±	9.84% ±	0.98% ±
	0.96%	62.19%	0.25%	1.93%	2.03%	1.39%	1.64%	0.19%
2	7.14% ±	36.20% ±	1.33% ±	27.46% ±	5.20% ±	4.77% ±	16.28% ±	1.47% ±
	1.52%	1.76%	0.27%	1.60%	3.58%	0.84%	3.21%	0.25%
1	7.95% ±	12.90% ±	1.37% ±	43.16% ±	3.90% ±	8.20% ±	21.00% ±	1.17% ±
	2.58%	1.40%	0.87%	3.58%	0.61%	1.00%	6.74%	0.41%
0.8	10.72% ±	8.21% ±	2.14% ±	51.33% ±	1.74% ±	5.46% ±	19.15% ±	1.25% ±
	2.22%	1.23%	1.30%	2.16%	0.78%	0.95%	5.04%	0.21%
0.5	18.82% ±	7.41% ±	5.76% ±	44.12% ±	1.52% ±	5.30% ±	15.25% ±	0.96% ±
	2.47%	1.60%	1.99%	4.75%	0.25%	1.35%	5.23%	0.31%
0.25	32.35% ±	7.43% ±	9.64% ±	28.48% ±	1.27% ±	4.30% ±	12.76% ±	0.96% ±
	5.93%	1.52%	1.87%	2.76%	0.21%	1.46%	6.08%	0.35%

MSE Stability

To demonstrate the stability of the electrolyzer with the MSE, electrolysis was performed galvanostatically at 100 mA cm⁻² at a CO₂ feed rate of 1 sccm·cm⁻², as seen in FIG. 36. Referring to FIG. 37, operating at this current density and flow rate is shown to limit CO₂ loss to ˜3%. After an initial break-in period of 10 hours, the FE towards C₂H₄ and all CO₂RR products remained constant for another 190 hours at ˜43% and >91%, respectively, with no indications of reduced performance at the end of the test period. Referring to Table S12, while achieving minimal CO₂ loss, the MSE electrolyzer was shown to be the longest operating stable CO₂-to-C₂⁺ electrolyzer at industrially relevant current densities.

TABLE S12
The longest demonstration in current literature operated
at industrially relevant current densities.
Metallic Cation	CO2	Operation	C2+
in Electrolyte	Loss	Time	FE	Reference
None	3%	200	h	77%	This work
0.1M KHCO3	70%*	55	h	80%	Nat. Commun. 12, 2808 (2021)
0.1M KHCO3	70%*	157	h	81%	ACS Energy Lett., 6, 809-815 (2021)
0.1M KHCO3	70%*	100	h	70%	ACS Energy Lett. 5, 2811-2818 (2020)
0.1M KHCO3	70%*	60	h	~60%	Science 367, 661-666 (2020)
1M KOH	90%*	110	h	~80%	Nature 581, 178-183 (2020)
0.1M KHCO3	70%*	180	h	~70%	Nature 577, 509-513 (2020)
0.1M KHCO3	70%*	100	h	80%	Joule 3, 2777-2791 (2019)
1M KOH	90%*	150	h	83%	Science 360, 783-787 (2018)

*CO₂ loss is typically not reported, so estimates were applied based on measurements with similar electrolytes from other reports for flow cells and zero-gap electrolyzers.

It should be noted that all CO₂RR experiments were performed using an electrolyzer with an active area of 1 cm² at room temperature (˜ 20° C.) according to a set up as detailed in FIG. 5 and photographed in FIGS. 6A to 6B. During a CO₂RR experiment, the aqueous 0.01 M H₂SO₄ anolyte was circulated through the anode flow channel at a flow rate of 25 mL min⁻¹ using a peristaltic pump. The CO₂ gas flow was bubbled through water at room temperature for humidification prior to entering the electrolyzer. All voltages reported are full cell voltages without iR compensation.

Product Analysis

The CO₂RR gas products were analyzed in 1 mL volumes using a gas chromatograph (PerkinElmer Clarus 590) equipped with a thermal conductivity detector (TCD) and a flame ionization detector (FID). The liquid products were quantified using 1H nuclear magnetic resonance spectroscopy (NMR) on an Agilent DD2 500 spectrometer with dimethyl sulfoxide (DMSO) as an internal standard. For the stability tests, the electrolyzer was run uninterrupted with a fixed current density. The gas and liquid products were sampled periodically. DI water was periodically added to the electrolyte to compensate for water dragged through the cation-conducting layer. The products were manually collected periodically for analysis. The FE of CO₂RR gas products was calculated using the following equation:

${\mathrm{FE}}_{\mathrm{gas}}=x_i\times v\times \frac{z_i{\mathrm{FP}}_o}{\mathrm{RT}}\times \frac{1}{I_{\mathrm{total}}}\times 100\%$

where xi is the volume fraction of gas product i, v is the outlet gas flow rate in sccm, z_i is the number of electrons required to produce one molecule of product i, F is the Faraday Constant, P_o is atmosphere pressure, R is the ideal gas constant, T is the temperature, and j_total is the total current.

The FE of CO₂RR liquid products was calculated using the following equation:

${\mathrm{FE}}_{\mathrm{liquid}}=n_i\times \frac{z_{i}F}{Q}\times 100\%$

where n_i is the number of moles of liquid product i, and Q is the cumulative charge as the liquid products were collected.

The EE of CO₂RR products was calculated using the following equation:

${\mathrm{EE}}_i=\frac{E_i^{◦}}{E_{\mathrm{cell}}}\times {\mathrm{FE}}_i$

where E^o_i is the thermodynamic potential for each product,⁵⁰ and E_cell is the applied cell voltage without iR compensation.

It should be noted that the same numerical references refer to similar elements. Furthermore, for the sake of simplicity and clarity, namely so as to not unduly burden the figures with several references numbers, not all figures contain references to all the components and features, and references to some components and features may be found in only one figure, and components and features of the present disclosure which are illustrated in other figures can be easily inferred therefrom. The embodiments, geometrical configurations, materials mentioned and/or dimensions shown in the figures are optional, and are given for exemplification purposes only. Therefore, the descriptions, examples, methods and materials presented in the claims and the specification are not to be construed as limiting but rather as illustrative only.

In the above description, the term “about” means within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, i.e., the limitations of the measurement system. It is commonly accepted that a 10% precision measure is acceptable and encompasses the term “about”.

In the above description, an embodiment is an example or implementation of the inventions. The various appearances of “one embodiment,” “an embodiment”, “some embodiments” or “some implementations” do not necessarily all refer to the same embodiments/implementations. Although various features of the invention may be described in the context of a single embodiment/implementation, the features may also be provided separately or in any suitable combination. Conversely, although the invention may be described herein in the context of separate embodiments/implementations for clarity, the invention may also be implemented in a single embodiment/implementation.

All references detailed herein above and below are incorporated by reference.

• Gabardo, C. M., O'Brien, C. P., Edwards, J. P., McCallum, C., Xu, Y., Dinh, C.-T., Li, J., Sargent, E. H., and Sinton, D. (2019). Continuous Carbon Dioxide Electroreduction to Concentrated Multi-carbon Products Using a Membrane Electrode Assembly. Joule 3, 2777-2791.
• Larrazábal, G. O., Strøm-Hansen, P., Heli, J. P., Zeiter, K., Therkildsen, K. T., Chorkendorff, I., and Seger, B. (2019). Analysis of Mass Flows and Membrane Crossover in CO 2 Reduction at High Current Densities in an MEA-Type Electrolyzer. ACS Applied Materials & Interfaces 11, 41281-41288.
• Endródi, B., Kecsenovity, E., Samu, A., Halmágyi, T., Rojas-Carbonell, S., Wang, L., Yan, Y., and Janáky, C. (2020). High carbonate ion conductance of a robust PiperION membrane allows industrial current density and conversion in a zero-gap carbon dioxide electrolyzer cell. Energy & Environmental Science 13, 4098-4105.
• O'Brien, C. P., Miao, R. K., Liu, S., Xu, Y., Lee, G., Robb, A., Huang, J. E., Xie, K., Bertens, K., Gabardo, C. M., et al. (2021). Single Pass CO 2 Conversion Exceeding 85% in the Electrosynthesis of Multicarbon Products via Local CO 2 Regeneration. ACS Energy Lett., 2952-2959.

Claims

1. A microchanneled solid electrolyte (MSE) for in-situ regeneration and collection of carbon dioxide (CO2) during a CO2 electroreduction operation, the MSE comprising: an anion conducting layer configured to conduct (bi) carbonate anions from a surface of an adjacent cathode;a cation conducting layer configured to conduct protons from a surface of an adjacent anode; andan integrated channel layer comprising multiple microchannels formed between the anion conducting layer and the cation conducting layer;wherein the microchannels define a hollow path extending across the integrated channel layer for receiving the (bi) carbonate anions from the anion conducting layer and the protons from the cation conducting layer, thereby locally regenerating and collecting CO2 along and within the microchannels.

2. The MSE of claim 1, wherein the microchannels are defined on a surface of the cation conducting layer to produce the integrated channel layer forming a one-piece structure with the cation conducting layer.

3. The MSE of claim 1, wherein the microchannels are defined on a surface of the anion conducting layer to produce the integrated channel layer forming a one-piece structure with the anion conducting layer.

4. The MSE of claim 1, wherein the integrated channel layer is a separate microporous structure positioned between the anion conducting layer and the cation conducting layer.

5. The MSE of any one of claims 1 to 4, wherein the microchannels are in fluid communication with one another to define a network that directs the regenerated CO2 from a central region of the integrated channel layer to an edge region of the integrated channel layer.

6. The MSE of claim 5, wherein the network of microchannels defines an interconnected diamond pattern.

7. The MSE of claim 5, wherein the network of microchannels defines an interconnected square pattern.

8. The MSE of claim 5, wherein the network of microchannels defines an interconnected circular pattern.

9. The MSE of any one of claims 1 to 8, wherein the microchannels are sized and shaped to maintain a maximum pressure below 100 kPa, 90 kPa, 80 kPa, 70 kPa, 60 kPa, 50 kPa, 40 kPa, 30 kPa, 20 kPa or 10 kPa.

10. The MSE of any one of claims 1 to 9, wherein the microchannels are sized and shaped to maintain a voltage drop below 500 mV, 400 mV, 300 mV, 200 mV, or 100 mV.

11. The MSE of any one of claims 1 to 10, wherein the microchannels have a channel depth of at most 125 μm, between 10 and 100 μm or between 20 and 80 μm.

12. The MSE of any one of claims 1 to 11, wherein the microchannels have a pore path width between 25 and 150 μm or between 25 and 75 μm.

13. The MSE of any one of claims 1 to 12, wherein the integrated channel layer has a porosity between 5% and 95%, between 10% and 80%, between 15% and 70%, between 20% and 60% or between 25% and 50%.

14. The MSE of any one of claims 1 to 13, wherein the microchannels are uniform across the integrated channel layer.

15. The MSE of any one of claims 1 to 14, wherein the microchannels have a non-circular cross-section including cross-sections of elliptical, circular, rectangular, or square shape.

16. The MSE of any one of claims 1 to 15, wherein the anion conducting layer comprises fixed cations derived from piperidinium, imidazolium, or benzimidazolium.

17. A system for electroreduction of CO2 into carbon products, the system comprising: an electrolyzer comprising: a cathode flow field having an inlet to receive a CO2 gas stream;a cathode in fluid communication with the cathode flow field to operate electroreduction of the CO2 gas stream;an anode flow field having an inlet to receive an anolyte stream;an anode in fluid communication with the anode flow field; anda microchanneled solid electrolyte (MSE) positioned between the cathode and the anode in a forward-biased configuration, the MSE as defined in any one of claims 1 to 16 and releasing a concentrated CO2 stream comprising CO2 and water to the cathode flow field when the CO2 gas stream is electroreduced; anda recycle loop in fluid communication with the cathode flow field for recovering the concentrated CO2 stream from an outlet of the cathode flow field and redirect the regenerated and collected CO2 from the concentrated CO2 stream back to the cathode flow field for serving as at least a part of the CO2 gas stream.

18. The electrolyzer of claim 17, wherein the recycle loop comprises: a first tubing for recovering the concentrated CO2 stream comprising water and regenerated CO2 from the microchannels;a liquid-gas separator in fluid communication with the first tubing to receive the concentrated CO2 stream and separate the concentrated CO2 stream into a regenerated CO2 gas stream and a water stream; anda second tubing interconnected the liquid-gas separator to the cathode flow field for recycling the regenerated CO2 gas stream to the cathode flow field.

19. The system of claim 18, wherein the second tubing is in fluid communication with the inlet of the cathode flow field to provide the regenerated CO2 gas stream along with the CO2 gas stream to the cathode flow field.

20. The system of any one of claims 17 to 19, where the concentrated CO2 stream comprises at least 80%, 85%, 90% or 95% of CO2, and an anode tail gas recovered from the anode flow field comprises at most 1%, 2%, 3%, 4% or 5% of CO2.

21. The system of any one of claims 17 to 20, wherein the anolyte is free of mobile alkali metal cations.

22. The system of any one of claims 15 to 20, wherein the anolyte is water or a solution of H2SO4, HClO4, or a combination thereof.

23. The system of any one of claims 17 to 22, wherein the CO2 gas stream is supplied at a CO2 feed rate between 0.25 sccm·cm−2 and 2 sccm·cm−2, 0.5 sccm·cm−2 and 1.5 sccm·cm−2, or 0.8 sccm·cm−2 and 1 sccm·cm−2.

24. The system of any one of claims 17 to 23, wherein the electrolyzer is operated at a current density between 40 mA·cm−2 and 240 mA·cm−2, 50 mA·cm−2 and 200 mA·cm−2, 60 mA·cm−2 and 160 mA·cm−2, 80 mA·cm−2 and 120 mA·cm−2, or 90 mA·cm−2 and 100 mA·cm−2.

25. A method for reducing CO2 losses during electroreduction of CO2 in a CO2RR electrolyzer comprising a cathodic compartment and an anodic compartment, the method comprising: supplying a CO2 gas stream to the cathodic compartment operating CO2 reduction reactions producing (bi) carbonate and hydroxide anions;supplying an anolyte stream to the anodic compartment operating anodic reactions producing protons;allowing generation and collecting of a concentrated CO2 stream in microchannels positioned between the cathodic compartment and the anodic compartment by: directing (bi) carbonate and hydroxide anions from the cathode to the microchannels via an anion conducting layer, anddirecting protons from the anode to the microchannels via a cation conducting layer;recovering the concentrated CO2 stream from the CO2RR electrolyzer;separating water and a regenerated CO2 gas stream from the concentrated CO2 stream; andrecycling at least a portion of the regenerated CO2 gas stream to the cathodic compartment of the CO2RR electrolyzer.

26. The method of claim 25, wherein supplying the CO2 gas stream is performed at a CO2 feed rate between 0.25 sccm·cm−2 and 2 sccm·cm−2, 0.5 sccm·cm−2 and 1.5 sccm·cm−2, or 0.8 sccm·cm−2 and 1 sccm·cm−2.

27. The method of claim 25 or 26, comprising operating the CO2RR electrolyzer at a current density between 40 mA·cm−2 and 240 mA·cm−2, 50 mA·cm−2 and 200 mA·cm−2, 60 mA·cm−2 and 160 mA·cm−2, 80 mA·cm−2 and 120 mA·cm−2, or 90 mA·cm−2 and 100 mA·cm−2.

28. The method of any one of claims 25 to 27, comprising controlling at least one of: a maximum pressure within the microchannels below 100 kPa, 90 kPa, 80 kPa, 70 kPa, 60 kPa, 50 kPa, 40 kPa, 30 kPa, 20 kPa or 10 kPa; anda voltage drop below 500 mV, 400 mV, 300 mV, 200 mV, or 100 mV.

29. The method of any one of claims 25 to 28, comprising providing the microchannels with at least one of: a channel depth of at most 125 μm, between 10 and 100 μm or between 20 and 80 μm; anda pore path width between 25 and 125 μm or between 25 and 75 μm.

30. The method of any one of claims 25 to 29, comprising providing the microchannels in an integrated channel layer between the anion conducting layer and the cation conduction layer, the integrated channel layer having a porosity between 5% and 95%, between 10% and 80%, between 15% and 70%, between 20% and 60% or between 25% and 50%.

31. The method of any one of claims 25 to 30, comprising providing the anion conducting layer with fixed cations derived from piperidinium, imidazolium, or benzimidazolium.