Electrochemical twin reactor

Abstract

The underlying invention is related to a device and a process for the direct production of an energy storing fluid from a reactant and an electrolyte in an electrochemical twin reactor, comprising a first reactor part which has at least one anode and an electrolyte for dissolving the reactant and a second reactor part which is comprising at least one further anode and a further electrolyte as a hydrogen source.

Description

[0001] Electrochemical twin reactor

[0002] The underlying invention is related to a device and a process for the direct production of an energy storing fluid from a reactant and an electrolyte in an electrochemical twin reactor, comprising a first reactor part which has at least one anode and an electrolyte for dissolving the reactant and a second reactor part which is comprising at least one further anode and a further electrolyte as a hydrogen source.

[0003] Implementing an energy transition these days is a huge challenge for all sectors of industry that use large amounts of fossil fuels, such as transport, logistics and the production industry. Currently, the energy requirements of these sectors are mainly met by fossil fuels, which need to be replaced with regenerative electricity or regenerative fuels.

[0004] An alternative to fossil fuels are special fuels (X-fuels), which are synthesized from electrolysis-based hydrogen. The most important X fuel is currently hydrogen itself. However, before hydrogen can be utilized as an energy source on a widespread basis, there are still some considerable hurdles that need to be overcome in terms of its transportation and storage. These include either high space requirements for its storage or other energetically unfavorable conditions. Ammonia (NH3) could become an alternative solution to allow the required hydrogen to be stored and transported with ease.

[0005] It should be mentioned first that ammonia has primarily been known as a source material for agricultural fertilizers. However, it is also a high-quality energy carrier, particularly as a storage medium for hydrogen, because ammonia can be liquefied at a moderate temperature of -33 °C, it has a volumetric hydrogen content which is significantly higher than that of compressed hydrogen at 700 at. In comparison to hydrogen, liquefied ammonia makes it easier to transport large volumes to wherever it is needed. Because hydrogen produced from ammonia does not contain any carbon oxides or methane, it is also free from greenhouse gases.

[0006] The conversion of hydrogen into ammonia is a viable option for using it as an energy source both on-site for industrial processes on a large scale as well as on a decentralized basis in the logistics sector. However, ammonia is not very suitable for generating energy through combustion as it is virtually incombustible in the air. But it has to be stated that in a cracking reactor ammonia can be split into nitrogen andhydrogen when suitable catalysts are applied. A mixture of ammonia, hydrogen and nitrogen is suited for homogeneous combustion and can be used as an energy source. Lithium-mediated N2 reduction reaction (LiNRR) is in the spotlight as an alternative to Haber Bosch process for ammonia synthesis. The Haber-Bosch synthesis of ammonia from N2 and H2 is a root cause of the population growth from about two billion in early twentieth century to more than 8 billion today, making its invention as one of the keystones in defining Anthropocene epoch. While its importance cannot be overlooked and benefits are undeniable, disruption of the global nitrogen cycle, along with consuming a considerable portion of the global energy production and its share in greenhouse gas emissions due to H2 production from steam reforming, made the search for an alternative ammonia synthesis process a priority. From thermodynamic point of view, the exothermic nature of ammonia synthesis from gaseous nitrogen and hydrogen (N2 + 3H2 2NH3 + 46 kJ mol-1), demands high pressure and low temperature to favor the equilibrium toward ammonia. On the other hand, slow reaction kinetics at low temperatures forces Haber-Bosch plants to operate at elevated temperatures, making it an energy-intensive process.

[0007] There are various reaction setups proposed to replace the Haber-Bosch process, among which electrochemical synthesis seems particularly promising. In ammonia electrosynthesis, the reaction is enforced by renewable electricity, circumventing the harsh operation conditions required for thermochemical process, opening a path for ambient temperature and pressure synthesis. Among NH3 electrosynthesis systems, so far only lithium-mediated nitrogen reduction reaction (LiNRR) has been proven to be able to break extremely stable N2 triple bond under ambient conditions to form ammonia.

[0008] The first mention of LiNRR dates to 1930 by Fichter et al. in which they showed ammonia is produced by electrolyzing halogenic lithium solution containing alcohols with N2 gas bubbling on platinum cathode. Atypical LiNRR system consists of N2 gas, an organic solvent, a lithium salt with high solubility and stability in the solvent, a proton source, and a proton shuttle to deliver the H+to the cathode. Current understanding of the LiNRR is that the ammonia electrosynthesis initiates with electrodeposition of Li+on the cathode, followed by decomposition of non-aqueous electrolyte in contact with metallic lithium and formation of a surface layer called solid-electrolyte interphase (SEI). The SEI is made of a mixture of organic and inorganic compounds, within whichthe metallic lithium reacts with N2 to form LisN , then further reacts with a proton source to result the deprotonated form of the proton donor and NH3. The SEI plays a defining role in ammonia production performance by providing a protective passivation layer for the cathode, and at the same time regulating the reaction by modulating the diffusion of N2, Li+, and H+, which considered to be the bottlenecks of a LiNRR system.

[0009] Unlike traditional electrocatalytic mechanisms where reaction cycle occurs on a two-dimensional surface in a short timeframe, LiNRR takes place in three-dimensional SEI at more extended intervals, adding more complexity to the system. Different strategies have been tested to achieve Faradaic efficiencies (FE) near 100% and ammonia production rates of ~2.5 pmol crrr2s’1, like employing non-aqueous gas diffusion electrodes (GDE), using high N2 gas pressure, and variety of proton donors. So far, H+is provided either by sacrificing the proton shuttles and solvent, or by coupling LiNRR with H2 oxidation reaction (HOR), which poses additional challenges to practical applications. One might think of water as a source of H+but adding H2O to the electrolyte solution will result in undesired side reactions and competition between LiNRR and hydrogen evolution reaction (HER), lowering the current efficiency.

[0010] In this aspect several investigations have been conducted to achieve improved results. Therefor the basis for electrochemical synthesis of ammonia with Li-rich SEI electrodes can be found in a paper e.g. - Electrosynthesis of ammonia with high selectivity and high rates via engineering of the solid-electrolyte interphase - ScienceDirect, Volume 6, Issue 9, 21 September 2022, Pages 2083-2101.

[0011] An ammonia-producing reactor with a palladium electrode and two chambers, one of these is filled with an aqueous electrolyte and one with a non-aqueous electrolyte, is known from application US 2012048742 A1.

[0012] Further the US 2024150903 A1 provides an electrochemical method for direct ammonia production from water and nitrogen using palladium and lithium catalysts. Additionally, the DE 102013019341 A1 discloses an electrolysis cell with a first part comprising an anode and an alkaline electrolyte, a second part comprising a further anode and a further electrolyte with water or hydrogen as a hydrogen source, wherein the two parts are interconnected by a cathode, whereby hydrogen and oxygen can be formed from water as energy carriers.Finally, WO 2022 / 031256 A1 discloses a structure with a cathode enclosed by two anodes, wherein an aqueous electrolyte is introduced via an opening and an antiseptic solution consisting of silver atoms and water is drained off via a drain, wherein the silver atoms being produced by the dissolution of the anode.

[0013] However, in the state of the art it is mentioned that it is problematic extracting protons from an aqueous source and feed them into a non-aqueous solution, without contaminating the target reaction environment with water.

[0014] The problem is solved by the underlying independent claims 1 and 7.

[0015] The concept of electrochemical production of NH3 as an energy storing liquid can be achieved by an electrochemical twin reactor, especially a palladium membrane reactor (PMR), which paves the way to separate the proton source for hydrogenation as one part of the electrochemical twin reactor from the other part for solving the reactant in another reactor part. A central electrode or a central cathode as a membrane-shaped metal layer or a metallized membrane or metallized membranous ceramic device can be used to separate or vice versa connect both reactor parts in an electrochemical cell as an electrochemical twin reactor. The membrane-shaped metal layer or the metallized membrane or the metallized membranous ceramic device encompasses a thickness in the range from 25 pm to 250 pm.

[0016] It is possible to use a metal for the metal layer or metallized membrane or the metallized membranous ceramic device like Pa, Ni, Fe, Ti, V, Nb, Ta, Y, Pd, V, Cr or alloys of these elements or a further metal with a face-centered cubic lattice to selectively absorb and accommodate individual hydrogen atoms within its structure. By integrating the metal like PMR with an electrochemical system it is possible to extract hydrogens from an aqueous source, in which the metal serves as both the cathode for the aqueous chamber and the hydrogen source for the non-aqueous side. As a material for the anode(s) can be used the material selected from a group of elements comprising platinum, stainless steel, or graphite or alloys these elements. As in conventional electrochemical reactors like PMR the electrical charge is only used for extracting H+from an aqueous solution and accommodating them into the metal structure through a Faradaic process. In the underlying invention a lithium-mediated ammonia electrosynthesis has been developed, by coupling it with a hydrogenation e.g. a PMR-LiNRR. This is achieved by running two electrochemical reactions in series,in which a metal e.g. Pd acts simultaneously as the cathode for both electrochemical systems, physically separating aqueous and organic compartments, while selectively permeating H+from an aqueous solution to the LiNRR chamber. The cathode acts therein like a diffusion barrier for Protons to cross into the LiNRR chamber, but unlike conventional PMR, both chambers of the twin reactor undergo Faradaic reactions. In a typical LiNRR system, small quantity of alcohol improves the production of ammonia, which is considered to play two major roles in the system. First, the delivery of protons to SEI increases by adding an alcohol or an alcoholic solution to an organic solvent like THF, hence the alcohols can react with lithium nitride to form the desired product. The organic solvent like an alcohol or the alcohol containing solution can modulate the mass transfer ratio between H+and N2, significantly affecting the performance. On the other hand, an excess H+results in parasitic side reactions and wasting electrical charge, while H+deficiency put a cap on the NH3 production. In addition to being a proton donor (or proton shuttle), the alcohol or alcohol solution also modulates the SEI structure in a way that improves continuous reactivity between H+, N2, and Li.

[0017] Metallic lithium in contact with an organic solvent like THF, alcohol or alcohol solutions e.g. ethanol or methanol or other alcohols undergoes a spontaneous reaction, which results in lithium ethoxide and hydrogen gas. High concentration of the alcohols e.g. methanol, ethanol etc. in LiNRR quickly dissolves the electroplated lithium, hampering the N2 activation step. On the other hand, absence of ethanol or methanol or other alcohols causes the electrodeposited species to pile up on the cathode, forming a thick and dense SEI layer. The compact SEI formed in the absence of methanol, ethanol or other alcohols complicates the reactivity between N2 and electrodeposited lithium, while the slight presence of methanol, ethanol or other alcohols facilitates the transformation of SEI to a more dynamic layer, allowing efficient diffusion and reaction of H+, N2, and Li. In the underlying application the alcohol or alcoholic solution should not exceed a concentration of a volume 1.0% by volume.

[0018] The organic solvent for the electrolyte, normally a lithium salt source e.g. tetrafluoroborate (LiBF4) and small quantities of an alcohol for generating an optimized SEI is put in one part of the twin reactor to dissolve the reactant. The Lithiumsalt can be selected from a group comprising LiBF4, LiCICU, LiCFsSOs. The reactant can be a liquid or a gas like air, especially pressured air or pure N2 pressured or unpressured orCO2 pressured or unpressured. A liquid to be used can comprise formic acid, paraformaldehyde in solution or carbonic acid. In the other part of the twin reactor as an electrolyte and as hydrogen source an acid is selected from a group comprising HCI, HNO3or H2SO4.

[0019] By applying the respective potential at the first and second reactor part, the atomic hydrogen is formed at one end of the center cathode in a first reactor part and at the other end in a second reactor part the solid electrolyte interphase (SEI) is generated, which accelerates the reaction of the reactant to the energy storing fluid.

[0020] To the electrochemical twin reactor and its electrodes different currents in a range of at least 0.1 mA cnr2and 3 mA cnr2are applied to the anodes and to the center cathode as membranous metal layer or the metallized membrane or membranous metallized ceramic unit.

[0021] It should be stated that the electrochemical twin reactor can be used in the production of an energy storing fluid throughout the chemical industry.

[0022] Embodiments

[0023] An electrochemical twin reactor for the direct production of an energy storing fluid is investigated, choosing a first reactor part LiNRR reaction subunit as one chamber, wherein a Li-salt is solved in tetrahydro furan (THF) as an organic solvent, especially molar lithium tetrafluoroborate (UBF4) as a lithium salt and small quantities of ethanol as proton donor for SEI building are given as starting materials. For the second reactor part, the PMR reaction subunit as another chamber, 1M H2SO4 is used as electrolyte and hydrogen source for all the experiments.

[0024] In Figure 1 is shown a schematic drawing of a twin reactor. It can be seen that the twin reactor is comprising two reactor parts 1 , 2 with an organic chamber 1 and an aqueous chamber 2. Within these reactor parts 1 , 2 are located an anode 4, 5 comprising Pt on each side which are stuck into the respective chambers 1, 2 filled with an electrolyte. In case of the reactor part 1 the electrolyte is LiBF4 in THF and in case of the reactor part 2 the electrolyte is aqueous H2SO4. Combined are these two reactor parts with a center cathode 3, which reacts like a diffusion barrier for the protons of the chamber 2 and as a transformation lattice to form atomic hydrogen during the diffusion step of H+. When applying voltage or current to the electrodes on one side of the center cathode in chamber 1 a SEI 6 is formed which modulates the reaction of N2 as a reactant withthe Li+compounds and atomic hydrogen to generate NH3. It could be found that a slight presence of an alcohol in the reactor part 1 or organic chamber 1 leads to an increase of the ammonia production.

[0025] Materials

[0026] Information about chemicals used in this study is presented in Table S1.

[0027] Table S1 - Chemicals used within this application

[0028] >

[0029] >

[0030] >

[0031]

[0032] Twin -Reactor

[0033] A gas-tight H-cell with 30 mL capacity on each side was used as the twin reactor cell. Metallized palladium foil was placed between two gaskets and fixed between the two chambers as a first and second reaction part by a spherical joint clamp. The exposed surface area of palladium to the reaction solutions was ~0.785 cm2on each side. Two platinum electrodes were used as anodes for each chamber. The aqueous side was filled with 15 mL of 1M H2SO4, denoted as PMR chamber or aqueous chamber. The non-aqueous side was filled with 15 mL of 1M LiBF4 dissolved in THF with varying amounts of alcohols, denoted as LiNRR chamber or organic chamber. A PTFE cap fitted with a rubber O-ring and a screw cap were used to seal the reaction chamber and isolate it from the atmosphere. A glass tube with fritted tip was used to inject gas into LiNRR chamber. The gas flow rate was set to 10 standard cubic centimeters per minute (seem) by an ALICAT digital mass flow meter. The outlet gas was connected to a gas trap filled with vacuum pump oil. The LiNRR solution was under constant stirring by a magnetic stirrer. All cell parts, separators, and electrodes were sonicated and washed with DI water and acetone and dried in an oven at 110 °C prior to use.

[0034] Experiments

[0035] Figure 2 proofs the dependence of ammonia production to the ethanol concentration in the reaction solution. The results disclose that ammonia should be produced in the presence of ethanol. Significant amount of ammonia could be detected when ethanol concentration reaches 0.25% v / v for both LiNRR and PMR-LiNRR systems. As expected, there is an optimum alcohol concentration for LiNRR, which is 1% v / v in the underlying setup.

[0036] As can be seen in Figure 2 the coupling of LiNRR with PMR hydrogenation shifts the ammonia production behavior in relation with alcohol concentration, bringing the optimum ethanol concentration to 0.5% v / v. The enhancement in production rate of ~3 times is recorded for 0.25% and 0.5% ethanol after merging PMR and LiNRR, but it shows no improvement for 1%, which then gets worse by increasing ethanol content compared to conventional LiNRR.

[0037] Based on the atomistic kinetic model developed by Andersen et al., when the incoming H+to the reaction surface is more than, LiNRR is governed by N2 limited regime, andwhen there is more N2 compared to H+the system is governed by H+limited regime. Thus, for LiNRR to work effectively, there should be a balance between the mass transport of H+and N2.

[0038] Figure 3 presents the Faradaic efficiency as a function of ethanol concentration, which shows roughly a similar trend with ammonia production rate. The sum of charge passed from both PMR and LiNRR systems was used for calculating the FE. The merged PMR-LiNRR shows higher FE compared to LiNRR on 0.25% and 0.5% ethanol due to three-fold enhancement of ammonia production rate. Beyond that, even though the production rate of both systems is similar at 1 % ethanol, the excess charge injected to PMR chamber made FE to drop for the merged system. The results point out to the necessity of considering both electric input and product output in identifying the optimum reaction condition of successive Faradaic processes. The 0.5% v / v ethanol concentration was selected for the rest of the experiments.

[0039] Proton donor concentration is not the only parameter affecting the performance of the system. The reaction rate in conventional LiNRR is controlled by manipulating the electrical current passing the system. In the merged PMR-LiNRR system, the electrical current passing through each system can affect the performance, therefore variation of both PMR current density (JPMR) and LiNRR current density (JLiNRR) were investigated.

[0040] As reflected in Figure 4, increasing JPMR has beneficial effect on the ammonia production rate in general, but it reaches a plateau at higher values. JPMR can directly be attributed to the amount of H+injected to the LiNRR system. By providing excess H+, the mass transfer balance between N2 and H+is disturbed, which pushes the system into N2 limited regime. The unreacted H+recombines as H2 on the organic side, reducing the efficiency of the system. Formation of small bubbles was visible on higher PMR current densities on the non-aqueous side of the palladium membrane, which can be attributed the H2 evolution.

[0041] The diminish of Faradaic efficiency of NH3 electrosynthesis and H2 production is shown in Figure 5. The initial boost of FE fades away when JPMR increases, showing that there is an optimum window for hydrogenation current density to have a meaningful enhanced ammonia production.

[0042] Figure 6 shows how JUNRR affects ammonia FE and production rate.As the next step, the effect of LiNRR current density on ammonia production was investigated, by using the same solution as the previous section, while PMR current density was fixed at 3mA crrr2and LiNRR current density was varied. As expected, higher LiNRR current density resulted in higher production rate and FE (Figure 7). Further the proof of concept for the underlying twin reactor system was investigated in tetrahydrofuran (THF) as a solvent for 1 M lithium tetrafluoroborate (LiBF4) and small quantities of an alcohol in the organic chamber being the so called first reactor part. The presence of alcohol was found to enable the generation of a stable SEI to transport N2 and Li+to the cathode. The so called second reactor part encompasses as an electrolyte and hydrogen source an aqueous solution of 1 M H2SO4.

[0043] Pt was exclusively used as the anodes in both reactor parts and the center cathode was made of Pd. With 0.5% vol. alcohol in the organic compartment and applying -3 mA cm’2current to either the Pd in both aqueous and organic sides (PMR-LiNRR) or only to the organic side (LiNRR), the Li production rate could be quantified. It was found that using ethanol resulted in the highest NH3 yield, as quantified via NMR.

[0044] Keeping with ethanol as the alcohol source, it can be stated that increasing the applied current on the PMR side (while keeping the LiNRR current at -3 mA cm’2) boosted the NH3 production rate, providing strong evidence that H2O was acting directly as the Flsource (Fig.8a).

[0045] In addition to this, it was found that having a continuous current on the LiNRR side was also essential. When keeping the PMR current constant (-3 mA cm’2), increasing the LiNRR current also resulted in an increase in NH3 production. This illustrates the need to continuously reduce Li+to a metallic Li layer which can react with N2 and the use of the dual-electrochemical system to simultaneously drive the PMR and LiNRR process. We note that we kept the ethanol concentration deliberately low. As the concentration increased beyond 0.5% vol. the ethanol began to increasingly serve as the H-donor and the system behaved more like the standard LiNRR setup, as evidenced by a diminishing rate-enhancement conferred by the PMR part of the reactor.

[0046] While a typical reaction run lasted 1 h, sufficient to result in a NH3 concentration that can be reliably measured, it was attempted to probe the longevity of the initial iteration of the PMR LiNRR setup. When (-3 mA cm’2) was applied to both sides of the Pd foil (PMR-LiNRR), the concentration of NH3 continuously increased within a 24 h periodand plateaued afterwards (Fig. 8b), potentially due to a breakdown in the Pd-Li-SEI interface.

[0047] Finally, it could be proved that the NH3 produced resulted from N2 reduction as opposed to contamination or other sources. Therefore,15N2 isotopes were used, scrubbed for impurities, as the reagent and produced15NHs in comparable quantities (Fig. 8c). While Faradaic efficiency, quantified here using the total charge passed through both electrochemical reactors, reached only modest levels of up to 5%, this might be improved by the use of higher pressures of air or N2 and advances in reactor engineering.List of reference symbols

[0048] 1 first reactor part / Organic chamber

[0049] 2 second reactor part / Aqueous chamber 3 Center electrode / center cathode

[0050] 4 Organic chamber anode

[0051] 5 Aqueous chamber anode

[0052] 6 Solid electrolyte interphase (SEI)

Claims

Claims1. Electrochemical twin reactor for the direct production of an energy storing fluid from a reactant and an electrolyte, comprising a first reactor part (1) which has at least one anode (4) and an electrolyte for dissolving the reactant and a second reactor part (2) which is comprising at least one further anode (5) and a further electrolyte as a hydrogen source,wherein the first reactor part (1 ) and the second reactor part (2) are connected to one another via a center cathode (3) and in the second reactor part (2) produced hydrogen diffuses as an atomic or radical source through the center cathode (3) which reacts with the dissolved reactant in the reactor part (1) forming the energy storing liquid.

2. Twin reactor according to claim 1 ,wherein the center cathode (3) comprises a membrane-shaped metal layer or a metallized membrane or metallized membranous ceramic unit.

3. Twin reactor according to claim 1 or 2,wherein the center cathode (3) is used as a diffusion barrier to lead atomic or radical Hydrogen from the second reactor part (2) through the center cathode (3) to the first reactor part (1 ) solely.

4. Twin reactor according to one of the preceding claims,wherein the metal for the center cathode (3) is comprising elements selected from a group comprising Pa, Ni, Fe, Ti, V, Nb, Ta, Y, Pd, V, Cr or alloys of these elements.

5. Twin reactor according to one of the preceding claims,wherein the center cathode (3) is comprising a membranous metal layer or metallized membrane which encompasses a thickness in the range from 25 pm to 250 pm.

6. Twin reactor according to one of the preceding claims,wherein the material of the anodes (4,5) is selected from a group of elements comprising platinum, stainless steel, or graphite or alloys of these elements.

7. Process for producing an energy storing fluid in an electrochemical twin reactor, comprising at least two reactor parts with a first reactor part (1) which comprises at least one anode (4) and an electrolyte for dissolving a reactant and a second reactor part (2) which has at least one further anode (5) and a further electrolyte as hydrogen source,wherein the first reactor part (1 ) and the second reactor part (2) are connected to one another via a center cathode (3) which is formed as a shaped metal layer membrane or as a metallized membrane or as a metallized membranous ceramic unit, wherein by applying the respective potentials on the first and second reactor part (1 ,2), in the second reactor part (2) hydrogen is produced which diffuses as an atomic or radical source through the center cathode (3) and reacts after the diffusion ending with the dissolved reactant in the reactor part (1) forming the energy storing liquid.

8. A process according to claim 7,wherein a gas or liquid is used as the reactant.

9. A process according to claim 8,wherein the gas is nitrogen or carbon dioxide.

10. A process according to one of the preceding claims,wherein ammonia or formaldehyde or formic acid is produced as the energy storing fluid.

11. A process according to one of the preceding claims,wherein Lithium salt, preferably LiBF4, LiCICU, LiCFsSOs, is used as an electrolyte in the first reactor part (1) and an acid preferably HCI, HNOs or H2SO4 is used as hydrogen source in the second reactor part (2).

12. A process according to one of the preceding claims,wherein the electrolyte in the first reactor part (1) is present in an alcoholic solution which does not exceed a concentration of 1.0% by volume.

13. A process according to one of the preceding claims,wherein different currents in a range of at least 0.1 mA.crrr2and 3 mA. cm’2are applied to the anodes (4,5) and to the center cathode (3) as a membranous metal layer or a metallized membrane or a membranous metallized ceramic unit.

14. A process according to one of the preceding claims,wherein by applying the respective potential at the first and second reactor part (1 ,2), the atomic hydrogen is formed at one end of the center cathode (3) in reactor part (2) and at the other end of the center cathode (3) a solid electrolyte interphase (SEI) (6) is generated in reactor part (1), which accelerates the reaction of the reactant to the energy storing fluid.

15. Use of the electrochemical twin reactor according to any one of the preceding claims 1 to 6 for use in the chemical industry.

Images