Table of contents

N-oxyl compounds, in particular 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO), have attracted considerable attention in alcohol oxidations due to their remarkably low-cost, high selectivity, and metal-free nature. TEMPO-mediated oxidations are often performed in high pH electrolytes as high OH - alleviates both kinetic and thermodynamic limitations. However, the degradation of TEMPO is also favorable in base, resulting in significant loss of catalytic activity and challenging long-term electrosynthesis. By contrast, ionic Liquids (ILs) possess high alkaline chemical stability. Herein, we designed electrode composites in which ILs are used as on part of a biphasic system (electrode phase and buffer phase) preserving TEMPO catalytic stability. Polycaprolactone (PCL), a binder that also possesses low melting point, was applied as a comparison. The results indicated that the IL-TEMPO composite showed no loss of current for glycerol oxidation at pH 10.0 after 2.0 h, while 67.1% current was lost using PCL-TEMPO. This stability enhancement was further evaluated in an 800 μL electrochemical cell using bulk chemical analysis and successive cycles of glycerol oxidation. The strategy demonstrated here not only offers an opportunity to prepare catalytic systems with enhanced stability, but also converts what are typically homogeneous catalysts to heterogenous systems. In subsequent studies, we aim to exploit poly(ionic liquids) (PILs) to further regulate TEMPO catalytic activity at various pH conditions.

Figure 1

The carbon dioxide capture and storage (CCS) plays an important role in reducing greenhouse gases, but its high energy cost is a critical issue. The CO₂ separation is recognized as a key process for reducing cost, and a membrane separation method is one of the promising candidates with a low energy cost. Ion gels comprising polymer networks and ionic liquids (IL), which exhibit unique properties such as non-volatility and CO₂ absorption selectivity, have attracted attention as environmentally friendly separation membranes. Previous work in our group has shown that sulfonated polyimide (SPI) with an imidazolium cation exhibits good compatibility with an IL. SPI can hold high content of IL (~75 wt%) and form a uniform, flexible, tough (~10 MPa of Young's modulus) and thin composite membrane.^[1-3] This SPI was synthesized by random copolymerization of a sulfonated ionic monomer and a non-ionic monomer. The composite membranes separated into bicontinuous two phases of ionic and non-ionic domains, which contribute to the formation of gas diffusion path and mechanically tough network, respectively. By using multiblock copolymers of SPI, it is expected that the size and connectivity of each domain are improved, which could lead to the increase of the CO₂ permeability and mechanical toughness of the membrane. In this study, we fabricated an 1-buty-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ([C₄mim][NTf₂])/SPI composite membranes composed of multiblock-type SPI (mb-SPI) and compared their property with those of random-type SPI (r-SPI).

75 wt%[C₄mim][NTf₂]/mb-SPI composite membranes exhibited higher CO₂ permeability (372 barrer) than the corresponding r-SPI (235 barrer) while the gas selectivity (CO₂/N₂) is 26, which is of the same magnitude as that of r-SPI (CO₂/N₂ is 28). The mb-SPI can hold higher IL content up to 80 wt% than r-SPI, resulting in higher CO₂ permeability (480 barrer; CO₂/N₂ is 27). The higher ionic conductivity also suggested the higher connectivity of ionic domains in mb-SPI system. On the other hand, from tensile tests, it was found that the mb-SPI membrane was less ductile than the r-SPI membrane. The result of dynamic mechanical analysis (DMA) indicates that the difference of internal structure between the membranes composed of mb-SPI and r-SPI is one of the reasons of the poor mechanical property.

To fabricate more strong membrane, we used SPI with more rigid polymer backbone. Rigid mb-SPI exhibits high CO₂ permeability (450 barrer) and gas selectivity (CO₂/N₂ is 24), comparable with the conventional one, and the mechanical property was superior to the conventional one. From these results, we concluded that the mb-SPI membrane having high CO₂ permeability and toughness was successfully fabricated.

References

[1] A. Ito et al., Sulfonated polyimide/ionic liquid composite membranes for carbon dioxide separation, Polymer J., 49, 671-676 (2017).

[2] A. Ito et al., Sulfonated Polyimide/Ionic Liquid Composite Membranes for CO₂ Separation: Transport Properties in Relation to Their Nanostructures, Macromolecules, 51, 7112-7120 (2018).

[3] E. Hayashi et al., Application of Protic Ionic Liquids to CO₂ Separation in a Sulfonated Polyimide-Derived Ion Gel Membrane, ACS Applied Polym. Mater., 1, 1579-1589 (2019).

With the continued rise in atmospheric CO₂ levels and the potential impact on climate, carbon capture and storage has gained prominence as a potential technological solution to reduce CO₂ emissions, resulting in increased interest in advanced materials to facilitate the removal of CO₂ from energy-related emissions. While a number of approaches have been explored to separate CO₂ from mixed streams, ionic liquids are promising materials due to their negligible vapor pressure, high thermal stability, and the ability to tune physicochemical properties such as CO₂ solubility. The incorporation of free volume into ionic liquids through appropriate development of bulky cation/anion pairs or through the creation of porous ionic liquids can impact both gas transport and solubility. Macrocyclic ionic liquids where the bulky cation consists of a nitrogen heterocycle connected by ether linkages can selectively remove CO₂ from mixed gas streams and offer a potential avenue for CO₂ capture. Porous ionic liquids are emerging materials that combine the attributes of porous materials within a nonvolatile liquid matrix. Permanent porosity is incorporated into the ionic liquid by adding a framework material (zeolitic imidazolate framework) where the cations and anions are too large to diffuse into the host. We investigated the selective absorption of CO₂ in both macrocyclic and porous ionic liquids and examined the effect of porosity on CO₂ sorption capacity and gas transport. Relevant physical and chemical properties of these unique materials such as solubility, density, viscosity and structure were measured. In addition, analysis of the vibrational modes of both the ionic liquid and the CO₂ through in situ attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) measurements was used to better understand the mechanisms of CO₂ interaction with the ionic liquids and the impact on solubility and transport. Moreover, we find that certain of these ionic liquids exhibit a unique binding and release of CO₂ which could open new avenues for CO₂ capture and storage.

In the past few years, more and more attention has been focused on processes occurring in small volumes such as microdroplets and thin films due to the increases in reaction rates and other phenomena recently reported in the literature. Our group has primarily investigated the thermodynamics of physical and chemical changes taking place in acoustically-levitated droplets ranging in volume from five to 10 microliters in size. Levitating a droplet using a standing acoustic wave allows for convenient study of reactions occurring within the droplet or any process reliant on the droplet's surface using visualization techniques such as microphotography and infrared imaging. We have used this technique in the past to study the evaporation process of levitated solvent droplets, including those of ionic liquids, as well as the neutralization thermodynamics of strong acids and bases within a levitated droplet.

In this work we present the initial characterization of droplet "shells" formed on the surface of levitated solvent droplets which contain a noble metal salt and a small piece of transition metal wire. In these experiments silver nitrate or a gold (III) salt was dissolved in a levitated solvent droplet (either in deionized water, silver bis(trifluoromethanesulfonyl)imide ionic liquid, or a DMF/water solution) after which a small (~0.5cm) section of zinc or copper wire was dropped into the levitated drop. The resulting reduction of the noble metal salt to the noble metal solid was observed for each solvent/noble/transition metal combination using a DSLR camera equipped with a close-up lens. In many cases a dark precipitate formed on the surface of the wire, indicating the formation of metal solid much in the same fashion as the silver "tree" general chemistry experiment. However, in the case of the silver nitrate/DI water/copper and gold(III)/DI water/zinc combinations, a metallic solid ( shiny, silver-colored for silver nitrate and black for gold(III)) formed a shell at the droplet/air interface. More study of the shell structures is planned including examination using SEM. While specific applications of these structures is not immediately clear, we envision they may be of interest as high surface area catalysts.

Figure 1

Deep eutectic solvents (DESs), compromised of hydrogen bond donors and acceptors, are concentrated electrolytes with extremely low volatility and tunable physical properties, and therefore they are promising for large scale energy storage. The electrode-electrolyte interfaces of concentrated and complex electrolytes such as DESs have not been rigorously studied. Owing to the similarity of DESs to ionic liquids (ILs) where a camel shape potential-dependent capacitances have been reported with significant deviations from dilute electrolytes, it will be critical to understanding the DES behavior near polarized surfaces to inform electrochemical kinetics. In this work, the excess charge accumulation of mixtures of choline chloride and ethylene glycol with 1:2 (Ethaline DES composition), 1:4 and 1:6 molar ratios are probed on silver and glassy carbon electrodes via electrochemical impedance spectroscopy. To study the surface species and changes with applied potential, in-situ surface-enhanced Raman spectroscopy was employed on an electrochemically roughened silver substrate as the electrode. Results from these studies suggest Ethaline DES is different than most ILs and present u-shaped differential capacitance and more in line with regular electrolytes. We will present the results from our ongoing studies of the electrode-DES interfaces outlining the impact of composition and electrode material.

Recent advancements in electrochemical polishing have minimized surface roughness and enhanced conductivity properties of high-purity (>99.9% composition) aluminum surfaces in ways that machine polishing and simple chemical polishing cannot compare. Aluminum metals have a high propensity to deliver powerful electric charges with little resistance, thus making them effective for a wide range of electroconductivity experiments. The effects of an acid-free ionic liquid electrolyte solution were tested to determine potential energy thresholds during electropolishing treatments based upon temperature, experiment duration, current, and voltage.

High purity aluminum metal specimens were used in electropolishing treatments with an acid-free ionic liquid electrolyte prepared from quaternary ammonium salts as an environmentally friendly alternative to acid-based solutions. An ionic liquid solution was formed from ethylene glycol and choline chloride combined in a 2:1 ratio. Similar ionic liquid solutions have been successfully employed to electropolish metal alloys, but little research has been conducted with this approach to polish rare earth metals like aluminum (1-5). Linear sweep voltammetry and chronoamperometry tests were initially used to determine ideal conditions for electrochemical polishing, while Atomic Force Microscopy revealed nanoscale effectiveness of the ionic liquid relative to an industry standard acid polishing treatment of 1 M Phosphoric Acid. Surface characterization via root mean squared roughness before and after electrochemical polishing treatments in 10x10 µm sample regions were used to estimate polishing efficiency throughout a 15-minute treatment period at 70 deg C and an average voltage of 2V.

Results from electropolishing treatments in the ionic liquid electrolyte revealed a significant change in root mean squared roughness from 159.31 nm to 26.649 nm and resulted in an overall mass loss of 0.0394 g. A change of 132.661 nm during the 15-minute treatment resulted in an electropolishing efficiency of 83.272% and a shiny mirror finish 6x smoother than the same aluminum surface prior to treatment.

Comparatively, the phosphoric acid electropolishing treatment yielded only a nominal improvement in root mean squared roughness of 97.786 nm, yet resulted in a greater mass loss of 0.0458 g for a lower smoothing efficiency rating of 38.546%. The electropolishing rate via the Phosphoric Acid solution is greater, yet the efficiency is less favorable due to significant pitting at low current densities (Figure 1). Hydrogen contamination was observed to adversely impact the overall root mean squared roughness of all sampled areas, yet this may be remedied via costly degassing treatments at temperatures of greater than 800 deg C to achieve more comparable results to the ionic liquid solution by mitigating some of the pitting and bubbling caused by hydrogen attack. In this study, a green polishing solution was prepared from simple inorganic salts, alkyls, and other organic solvents. This eco-friendly solution produced polished surfaces superior to those surfaces treated with industry standard acid electrochemistry treatments of 1M H₃PO₄.

References:

1. Loftis, Jon Derek, and Tarek M. Abdel-Fattah. "Nanoscale electropolishing of high-purity nickel with an ionic liquid." International Journal of Minerals, Metallurgy, and Materials26, no. 5 (2019): 649-656.

2. Loftis, Jon Derek, and Tarek M. Abdel-Fattah. "Nanoscale electropolishing of high-purity silver with a deep eutectic solvent." Colloids and Surfaces A: Physicochemical and Engineering Aspects511 (2016): 113-119.

3. Tarek M. Abdel-Fattah, Derek Loftis and Anil Mahapatro, "Nanoscale Electrochemical Polishing and Preconditioning of Biometallic Nickel-Titanium Alloys" Nanoscience and Nanotechnology, 5(2), (2015): 36-44

4. Abdel-Fattah, Tarek M., Derek Loftis, and Anil Mahapatro. "Nanosized controlled surface pretreatment of biometallic alloy 316L stainless steel." Journal of biomedical nanotechnology7, no. 6 (2011): 794-800.

5. Tarek M.Abdel-Fattah, Jon Derek Loftis, and Anil Mahapatro. "Ionic liquid electropolishing of metal alloys for biomedical applications." ECS Transactions25, no. 19 (2010): 57-60.

Figure 1

Hard chromium coatings are often used to improve the wear and corrosion resistance of different tools and components. Often electrolytes based on hexavalent chromium compounds, which are highly oxidative and carcinogenic, are still used to plate hard chromium layers. A key challenge is the development of an alternative baths free of Cr(VI). Such an alternative could be chromium coatings from a Cr(III)-based electrolyte.

Over the last years alternative electrolytes besides water were studied for electroplating. One of these alternatives are the deep eutectic solvents, DES, of which ethaline is a prominent candidate. It is 1:2 molar ratio mixture of choline chloride and ethylene glycol. This DES show good solubility for various metal salts. The electrochemical potential window of most DES is typically in the range of 3 V. Ethaline has excellent physical and chemical properties, such as: it is stable under ambient conditions, easy to prepare and handle, cheap (manufactured on the M tons scale globally) and biodegradable (used as a B-vitamin in animal food), requires no additional registration for process applications, it has good conductivity and low viscosity.

This paper will discuss the electroplating of Cr layers from ethaline on steel substrates. Different concentrations of CrCl₃ × 6 H₂O in the DES were investigated. The influence of LiCl, NaCl, KCl, formic acid and oxalic acid in the electrolyte as well as the influence of temperature were also investigated. Furthermore, the role of deposition technique (direct current or pulsed current) and of the hydrodynamic conditions on the quality of the chromium layers were evaluated.

The electrodeposition of some µm thick and adherent chromium films was proved to be possible from ethaline. The hardness of the deposited Cr layers was up to 2500 N/mm², depending on the process parameters.

The electrodeposition of iron alloys is a significant research topic, owing to the potential applications such as corrosion/wear-resistant coatings¹, magnetic thin films², and catalyst electrodes³. Although aqueous electrolytes have been mostly used for electrodeposition, vigorous hydrogen evolution on the cathode can lead to hydrogen embrittlement and precipitation of metal hydroxides². In addition, less noble elements such as aluminum and silicon cannot be co-deposited from aqueous electrolytes due to the limited electrochemical window.

To overcome the limitations of aqueous solutions, non-aqueous systems such as molten salts and ionic liquids have gained significant research attention. The electrochemical behaviour of iron complexes in imidazolium- and pyrrolidinium-based ionic liquids, and electrodeposition of metallic Fe was demonstrated in some liquids^4,5. Many ionic liquids are, however, not viable on a practical scale.

In this study, we explored an alternative electrolyte to ionic liquids for electrodeposition of iron. A mixture of ferric chloride and acetamide is shown to give a liquid at ambient temperature. This eutectic mixture has a higher conductivity than comparable eutectics formed from either aluminium or zinc chloride. Ferric chloride disproportionates to form both anionic and cationic iron species identified using spectroscopic methods. The electrodeposition of dense metallic iron film was demonstrated.

References

1. N. Tsyntsaru et al., Surf. Eng. Appl. Electrochem., 48, 491–520 (2013)

2. G. Panzeri et al., Electrochim. Acta, 271, 576–581 (2018)

3. N.-C. Lo et al., J. Electrochem. Soc., 163, D9–D16 (2016).

4. P. Giridhar, B. Weidenfeller, S. Z. El Abedin, and F. Endres, ChemPhysChem, 15, 3515–3522 (2014).

5. Y. Zhu, Y. Katayama, and T. Miura, J. Electrochem. Soc., 159, D699–D704 (2012).

A range of techniques including physical property measurements, neutron scattering experiments, ab initio molecular dynamics and classical molecular dynamics simulations are used to probe the structural, thermodynamic, and transport properties of a deep eutectic solvent comprised of a 1:2 molar ratio of choline chloride and ethylene glycol. This mixture, known as Ethaline, has many desirable properties for use in a range of applications and therefore understanding its liquid structure and transport properties is of interest. Simulation results are able to capture experimental densities, diffusivities, viscosities, and structure factors extremely well. The solvation environment is dynamic and dominated by different hydrogen bonding interactions. Dynamic heterogenities resulting from hydrogen bonding interactions are quantified. Rotational dynamics of molecular dipole moments of choline and ethylene glycol are computed and found to exhibit a fast and slow mode.

1. Introduction

Pyroprocessing is a promising technology for a fast reactor cycle with a highly concentrated minor actinides (MA: Np, Am and Cm)-bearing metallic fuel. Separation of actinides (An) from fission products such as lanthanide elements (Ln) is one of the difficult challenges to establish practical pyroprocessing. Liquid Cd cathode (LCC) with LiCl-KCl bath has been generally used for An recovery in our process [1], however, An/Ln separation performance using LCC was not high enough to fabricate a highly concentrated MA-bearing fuel from materials having a high ratio of Ln/An. Our recent studies showed that a combination of liquid Ga cathode with LiCl-KCl bath had potential to give a higher An/Ln separation performance, and currently we are developing an innovative pyroprocessing using the liquid Ga electrode. Solubility of U and Pu in liquid Ga is smaller than in liquid Cd. Therefore, the solid precipitates (An-Ga alloy) formation could have a significant influence on designing the process compared to the case of using LCC. However, the influence of the solid precipitates formation on An/Ln separation performance is unclear.

In this study, galvanostatic electrolysis test on the recovery of U, Pu and Am in liquid Ga was carried out to investigate the influence. Precipitation of Pu-Ga alloy was observed by SEM/EDX on the cross section of Ga electrode after the electrolysis, and separation factors (SFs) for each metallic elements (M) based on Ce, which is defined as (M/Ce concentration ratio in liquid Ga)/(M/Ce concentration ratio in the salt phase), were calculated to evaluate An/Ln separation performance.

2. Experiment

All experiments were performed in glove boxes with high-purity Ar atmosphere. Cd-0.1wt%Li alloy anode, liquid Ga cathode and Ag/AgCl reference electrode were used in LiCl-KCl eutectic melt at 773K. Two runs of galvanostatic electrolysis experiments were carried out applying the constant current of 5~10 mA (cathodic current densities: 3.76~7.53 mA/cm²). In the first experiment, An were recovered in the liquid Ga electrode from LiCl-KCl melt containing An chlorides, where concentrations of recovered An in the liquid Ga were almost the solubility. In the second experiment, two times larger amounts of An than the solubility were recovered in liquid Ga from LiCl-KCl melt containing both An and Ln chlorides. After the electrolysis, the Ga and the molten salt were sampled and dissolved into nitric acid. Concentrations of elements in the samples were analyzed by ICP-AES. γ spectrometry was also used to analyze Am concentration in samples. Cross section of solidified Ga electrode was observed by SEM/EDX.

3. Results and discussion

It was confirmed that SFs of An seen in Table 1 were around ten times higher values than those obtained in the case of LCC. Furthermore, the SFs of An in the case of the amount of An recovered in liquid Ga exceeding solubility (second experiment) were comparable to those in the case of the An concentration below solubility. The SEM/EDX observation showed that grain shaped precipitates with 1～10 μm was formed homogeneously inside the Ga and the size of the precipitates increased with the amount of recovered An in liquid Ga. Chemical form of the precipitates was assigned to be Pu-Ga alloy by the EDX. It was also found that the precipitates did not accumulate on the surface of liquid Ga. In the present experimental conditions, the amount of An in the liquid Ga had an influence on the precipitation formation, while the precipitation had no influence on An/Ln separation performance.

Acknowledgement

This work is supported by the Innovative Nuclear Research and Development Program from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

Reference

[1] S. Kitawaki, et al. Nucl. Technol., 162 (2008) 118-123.

[2] T. Murakami et al., J. Nucl. Radiochem. Sci., 16 (2016) 5-10.

Figure 1

Electrolytic uranium oxide reduction has the potential to be a key process for recycling spent fuel from commercial light water reactors to advanced nuclear reactors—including molten salt reactors and metal fueled reactors. However, several problems with the process have been identified that need to be addressed to support efficient, cost-effective commercial implementation. Process optimization requires attention to metal corrosion, anode stability, cell efficiency, and cathode product purity. Generation of oxygen bubbles at the anode combined with high temperature (650^oC) and molten chloride salt (LiCl + 1 wt% Li₂O) create a highly oxidizing environment for metals needed for salt containment and shrouding of the anode. The current standard choice for the inert anode is platinum metal, which has fairly good stability in the system. But given the high cost of platinum, alternative anode materials need to be discovered. Efficient removal of oxygen bubbles requires use of a shroud around the anode. The easy fabricability of metals makes them a reasonable choice for the shroud material, but corrosion due to interaction with the salt and oxygen needs to be minimized. Cell efficiency is affected by hydroxide impurity in the salt. Methods for removing water and hydroxides from the salt will be reported that improve cell efficiency. Cell efficiency can also be affected by controlling the reduction mechanism. Uranium oxide can either be reduced chemically by reaction with electrochemically generated lithium atoms or electrolytically with direct reduction of UO₂. The reduction mechanism also affects the buildup of entrained lithium oxide in the cathode product. Entrained lithium oxide is problematic as it gets carried on to subsequent processing steps such as electrorefining and molten salt oxidation.

The U.S. Department of Energy is currently supporting the design and planned construction of the Versatile Test Reactor (VTR) capable of performing experiments in a fast neutron field. This will allow for study of the effect of fast neutron irradiation on nuclear fuels, materials, and functional components. The molten salt irradiation experiment will utilize the VTR to study the chemical changes and speciation in molten chloride and/or fluoride salts.

In preparation for future experiments involving irradiation of candidate molten salt fuels, the redox potential measurement and control methods in molten NaCl-CaCl₂-UCl₃ at 500^oC were studied. The goal was to identify a metal/metal chloride redox buffer that can be used to control the redox potential, thereby helping to mitigate the corrosion risks posed by irradiation of the uranium salt fuel. An ideal candidate would both adequately reduce the redox potential and possess a low neutron interaction cross section.

The first metal/metal chloride redox buffer investigated was Zr/ZrCl₄. ZrCl₄ has a standard state reduction potential of -2.02 V vs. Cl^-/Cl₂ at 500° C. The standard state reduction potentials for UCl₃ and CrCl₂ are -2.41 and -1.55 V, respectively. Zr metal should, thus, oxidize to ZrCl₄ and control the redox potential at a level that simultaneously prevents Cr corrosion while leaving UCl₃ unreduced from the salt.

In order to test Zr metal for redox control, a 6.2-mm diameter Zr rod (99%) was submerged in molten NaCl-CaCl₂-UCl₃ salt at 600°C. The open circuit potential (OCP) of the salt was measured at a tungsten working electrode vs a Ag/AgCl reference electrode encased in a mullite tube. NiCl₂ was added to the salt to simulate the effect of fission on redox potential. After submerging the Zr rod, the measured open circuit potential between the W WE and the RE dropped by -0.63 V over a period of 100 minutes.

There is concern as to whether the rate of reaction of insoluble Zr metal with the salt can keep up with the rate of potential change due to the irradiation of the salt. Other redox buffer candidates being considered include uranium, calcium, and sodium. The latter two choices are attractive because of their predicted solubility in the molten salt, supporting homogenous redox control reactions. The problem with the latter three choices is that amounts contacted with the salt will need to be precisely controlled to avoid reduction of UCl₃. OCP measurements will also be reported for molten NaCl-CaCl₂-UCl₃ using U, Na, and Ca as the redox buffer. Conclusions will be drawn regarding the most effective metal to use in the future VTR experiments in order to control the redox potential of the irradiated salt.

The authors have been investigating a new RE (RE=Dy and Nd) recycling process from Nd-Fe-B permanent magnets, which is characterized by the use of RE-IG (IG=Fe, Ni, Co) alloy diaphragms for separation of REs in molten salts (Fig. 1) [1, 2]. In this process, the RE-IG alloy diaphragm performs as a bipolar electrode and RE ions permeate via three steps: (1) reduction of RE ions to form alloy on the anolyte side of the diaphragm, (2) diffusion of RE atoms in the diaphragm, and (3) oxidation of RE atoms to dissolve into the catholyte as RE ions on the other side of the diaphragm. Because the reaction and diffusion rates depend on the kinds of RE elements and the electrolytic conditions, selective permeation of Dy and/or Nd is possible under an appropriate condition. Using this selective permeation phenomenon through the alloy diaphragms, Dy and Nd can be separately recovered from Nd-Fe-B permanent magnets only by molten salt electrolysis, as shown in Fig. 1.

Recently, the authors reported selective Dy permeation through a RE-Ni alloy diaphragm in molten LiCl-KCl eutectic melts containing NdCl₃ and DyCl₃ at 450 ℃ [3]. Whereas, in the present study, selective permeation of Nd was examined using a similar experimental setup at the different electrolytic conditions. The obtained molar ratio of permeated Nd/Dy was around 6. Interestingly, EDX analysis revealed that the molar ratio of Nd/Dy in the alloy diaphragm was less than 0.05, which was two orders of magnitude lower than that of the permeated amount. These results suggested that the diffusion rate of Nd atoms in the alloy diaphragm was extremely high compared to that of Dy.

Acknowledgement

This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO) of Japan.

Reference

[1] T. Oishi, H. Konishi, T. Nohira, M. Tanaka and T Usui, Kagaku Kogaku Ronbunshu, 36, 299–303 (2010).

[2] T. Oishi, Molten Salts (Yoyuen Oyobi Koon Kagaku), 63, 78–83 (2020).

[3] T. Oishi, M. Yaguchi, Y. Katasho and T. Nohira, Rare metal technology 2020, 151–156 (2020).

Figure 1

Pyroprocessing of spent nuclear fuels which uses chemical and electrochemical reactions in molten salts has been developed as one of the promising technologies for advanced nuclear fuel cycle. A main step of the pyroprocessing is electrorefining process where a recovery of actinides from spent nuclear fuels is accomplished in an electrolyte, LiCl-KCl melt. After the electrorefining process, fission products (rare earths (REs), alkalis and alkaline earths) chlorides accumulate in the melt, the used LiCl-KCl melt.

The accumulated fission products must be removed from the used melt to be stabilized in a waste form suitable for a geological disposal. For this purpose, a zeolite is utilized in the conventional used salt treatment process: The zeolite is contacted with the used melt to absorb and occlude the fission products in its structure and then the zeolite is converted to a stable chlorine-containing waste form, a glass-bonded sodalite. However, the achievable concentration of fission products in the glass-bonded sodalite is limited to lower values (1.6 ~ 4.3 wt%) compared with that in the conventional oxide-base glass waste form (6.4 ~ 15.2 wt%).

In order to pursue a reduction of an environmental burden, this paper proposes a novel used salt treatment process utilizing the glass to decrease the volume of the waste form. REs, dominant among fission products, are electrochemically recovered from the used melt in a form of their silicides (REⁿ⁺ + xSi + ne^- → RESi_x). Then, the silicides are oxidized to be dissolved in the glass matrix. Si is selected as the alloying material, meaning that REs are not required to be separated from the Si before stabilizing in the glass waste form which is composed mainly of Si. After the electrochemical recovery of REs, the remaining fission products, alkalis and alkaline earths, are removed from the used melt in a similar manner to the conventional one; absorption and occlusion in the zeolite to be converted to the glass-bonded sodalite.

To investigate a basic behavior of a series of an individual RE silicide (La-Si, Ce-Si, Pr-Si, Nd-Si, Sm-Si, Eu-Si and Gd-Si) formation, cyclic voltammetry and potentiostatic electrolysis were performed using a Si wafer (n-type, <110>, 0.02 Ωcm) as the working electrode in LiCl-KCl melt containing the corresponding RE chloride at 723 K. It was confirmed that all of the above REs were recovered from the melt as their silicides. Silicides with various morphologies were found to be formed on the Si electrode; a pillar shaped silicide growing perpendicularly to the Si substrate for Sm, a dense and thin silicide layer covering the Si substrate for Eu and a porous silicide layer between a dense silicide layer and the remaining Si substrate for the rest of the REs. The obtained results confirmed a feasibility of the key step composing of the proposed used salt treatment process; REs recovery as silicides from the used LiCl-KCl melt. In addition, a preliminary test of the proposed used salt treatment process, Ce recovery as silicide from LiCl-KCl-CeCl₃ melt and the oxidation of the Ce silicide to be dissolved in a glass matrix, was demonstrated and an optimization of the processing condition is under way.

Acknowledgement Parts of the presented results were obtained in the project "Development of highly flexible technology for recovery and transmutation of minor actinide" entrusted to CRIEPI by the Ministry of Education, Culture, Sports, Science and Technology (MEXT).

There has been a renewed interest over the past decade in the use of molten salts as coolants and fuels in next-generation nuclear reactors that are used for electricity generation and to supply high-temperature, low-pressure heat. For solid-fuel salt-cooled reactors, molten salts are used as the coolant for solid-fueled reactors that operate at high temperatures (>500 °C) while for liquid salt fueled reactors, fissile, fertile, and fission products are dissolved in a homogeneous molten salt which serves as both the fuel matrix and the primary coolant. Two key challenges with a liquid salt fueled reactor is to manage the solubility and oxidation state of the materials dissolved in the molten salt and to manage the interfacial interactions of the molten salt with the reactor materials to limit corrosion. To understand how the structure and dynamics of molten salts impact their physical and chemical properties, such as viscosity, solubility, and thermal conductivity, as well as chemical reactivity, it is necessary to determine the structure and speciation of the molten salt at the atomic/molecular scale. The Molten Salts in Extreme Environments (MSEE) Energy Frontier Research Center is addressing these challenges through a coordinated experimental and theoretical effort to elucidate the atomic and molecular basis of molten salt behavior, including interactions with solutes and interfaces and under coupled extremes of temperature and radiation. The structure of bulk salt mixtures, such as MgCl₂/KCl, and solute speciation in molten salts has been studied to better understand behavior in these complex environments using combined X-ray scattering and spectroscopy, neutron scattering, optical spectroscopy and computational modeling. The chemical and morphological evolution of metal/molten salt interfaces has been examined using X-ray tomography and electron microscopy to better understand corrosion processes in molten salt systems. This presentation will discuss the challenges in molten salt chemistry for nuclear energy applications and highlight recent results in understanding the structure, properties and reactivity of high temperature molten salts.

Molten Salt Reactors (MSRs) are a potential game-changing technology for next-generation nuclear power. Although the MSR concept was demonstrated at Oak Ridge National Laboratory (ORNL) in the 1960's, it had not gathered attention from commercial vendors until recently. A fundamental knowledge of salt chemistry, including the speciation and solubility of corrosion and fission products is central to the safe and reliable operation of MSRs. The structural properties of fused salt solutions, especially the coordination geometry around the metal ions in the melt, are intricately dependent on the melt composition and temperature. In this work, a combination of in-situ electrochemistry and optical absorption spectroscopy techniques are utilized to understand the speciation and structure of lanthanides in molten chloride salts as a function of temperature, concentration and melt composition. The spectroelectrochemistry of metal species in molten salt media can be used as a potential process monitoring technique enabling the quantitative measurement of lanthanide and actinide metals within molten salt reactors or nuclear fuel pyro-processing applications.

This work was supported as part of the Molten Salts in Extreme Environments Energy Frontier Research Center, funded by the U.S. Department of Energy (US-DOE), Office of Science, Basic Energy Sciences, at BNL, INL and ORNL under contracts DE-SC0012704, DE-AC07-05ID14517 and DE-FC02-04ER15533, respectively.

In the last decade, there has been a sustained resurgence of the Molten-Salt Nuclear Reactor (MSR) concept. As the MSRs operate, chemical compositions and physical properties of salts change because of the creation of fission products, and the effects of corrosion and radiolysis. Therefore, computer simulations of the microscopic structure of multi-component molten salts are necessary to predict changing thermophysical properties. However, first-principles simulations inevitably involve simplifying approximations and therefore require careful experimental validation. Here we present detailed experimental and simulation studies of the structure of molten NaCl-CrCl₃ and NaCl-CrCl₂ near-eutectic mixtures. These compositions have been chosen to elucidate the behavior of Cr ions in the melts since Cr is the most important corrosion product. Although the corrosion will result in relatively small quantities of Cr in the melt, we used much larger Cr concentration to measure the structure around Cr ions with high resolution. The results of X-ray and neutron diffraction with isotopic substitution, X-ray absorption spectroscopy, and ab initio molecular-dynamic simulations are in agreement. In NaCl-CrCl₃ we found CrCl₆³⁺ octahedra and their networks, as well as intermediate-range order, which manifests itself in the prepeak at ~ 1 Å^-1 with non-monotonic temperature behavior.

Figure 1 shows one of the main results of this work, namely the comparison of measured and calculated total and partial PDFs of the molten NaCl-CrCl₃ near-eutectic mixture. The consistency between the measured and calculated PDFs is remarkable. Neutron diffraction with isotopic substitution allows for a high-fidelity determination of the distances between Cr ions and other species in the melt. By utilizing a combination of neutron diffraction with isotopic substitution of ⁵³Cr, X-ray diffraction, and ab-initio molecular dynamic simulations, we discovered the formation of chains of CrCl₆^3- corner-, edge-, and face-sharing octahedra, with Cr-Cl distances of 2.4 Å, and Cl-Cl distance of 3.3 Å. We also found evidence for intermediate-range order that is likely related to inter-chain correlations. In this regard, these ternary salt shows a structure that is reminiscent of that of KCl-MgCl₂, LiF-BeF₂, and NaCl-UCl_n.

Neutron and X-ray diffraction augment each other to determine partial PDFs Neutrons are particularly useful for studies involving, for example, Cr and Ni, where isotopes with a large difference in neutron cross-section are available. But when isotopic substitution is not achievable, the combination of X-rays and neutrons can extract the information about partial PDFs not otherwise available. The combination of neutrons and X-rays is also useful to overcome experimental constraints, such as compatibility between the salts and crucibles, potential radiation damage to crucibles, and sometimes large background from the sample environment. For example, quartz crucibles are suitable for both neutrons and X-rays and can be used with chloride salts but not with fluoride salts. The latter can be contained in vanadium crucibles for neutrons experiments. Our measurements demonstrate the relative strengths and weaknesses of both techniques for understanding the structure of multi-component salts. In conclusion, the understanding of the structure around Cr ions will be used to calculate properties of molten salt components (e.g., diffusivity) in the presence of Cr impurities. These properties will affect macroscopic thermo-physical properties such as thermal conductivity and viscosity of melts containing that, in turn, are necessary for designing safe and efficient MSRs. The extension of these methods to study melts containing up to 10 components will be discussed.

We acknowledge useful discussions with Richard Mayes, Stephen Raiman, Jake Mcmurray (ORNL), and Raluca Scarlat (UC Berkeley). This material is based upon work supported by the Department of Energy under Award Number DE-NE0008751. This research used resources at the Spallation Neutron Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated by the Oak Ridge National Laboratory, and the beamline 28-ID-1 (PDF) of the National Synchrotron Light Source II, a DOE Office of Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-SC0012704.

Figure 1

A variety of molten salts are proposed as liquid fuels or coolants in a new fleet of molten salt nuclear reactors that would have operational and safety advantages over present reactor systems. The molten salt will be exposed to high radiation levels under these conditions, therefore understanding the chemical effects of radiolysis on the molten salt fuel or coolant is essential to reliable, efficient and sustainable reactor operation. This understanding begins with identifying the primary salt radiolysis products and characterizing their reactivities, in pure salt and also under conditions that represent loading with fuel (actinides), fission products and corrosion products. These reactions form subsequent products that determine the chemical evolution of the molten salt fuel over the duration of its lifetime in the reactor. To obtain predictive insight into this radiation-driven chemical evolution, we investigate these reactions by performing pulse radiolysis transient absorption spectroscopy at the BNL Laser-Electron Accelerator Facility using a high-temperature sample configuration. Examples will be given of how the composition of the molten salt determines the identities and reactivities of the primary radiolysis products, and we will follow their reactions with metal ions that are representative of the reactor fuel environment. This work was supported as part of the Molten Salts in Extreme Environments Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science.

Figure 1

The bimetallic structure concept consisting of a corrosion resistant surface layer joined on top of an ASME code-approved substrate becomes a promising solution to address the corrosion issue of ASME substrate materials in molten salt environment. Specifically, refractory materials such as molybdenum, tungsten, and alloys containing them are found to be most resistant in molten salt environment and hence they have great potential to be used as the protective salt-facing surface layer. Due to the well-recognized difficulties of welding/3D-printing refractory materials (i.e., oxidation, brittle phase formation, cracking, etc.), cold spray, a deposition technology that minimizes material heating, avoids undesired thermal effects and maintains dimension stability, is an ideal approach for fabricating the bimetallic components. A corrosion resistant cladding material that is also amenable to cold spray process is desired and has been developed. Promising cladding structure has been designed, and tested by both molten salt corrosion and cold spray trials for down-selection. Prototyping for clad adhesion test will also be conducted as proof-of-concept validation. This program is sponsored by DOE Office of Nuclear Energy.

A thorough understanding of radiation effects on molten salt media is necessary to support the design, development, and deployment of molten salt reactor, using either fuel salt or solid fuel with salt as the coolant. Fundamental issues include the determination of the initial yields and reactivity of the primary radiolytic species, i.e., the solvated electron (e_solv^–), chlorine atom (Cl^•), and dichloride radical anion (Cl₂^•–) as their reactivity varies depending on the environment such as the salt cation and temperature. Here we report on the reaction kinetics for e_solv^– and Cl₂^•– in molten LiCl-KCl eutectic doped with Zn²⁺ ions. Electron pulse radiolysis absorption spectroscopy was used to observe the transient behavior of e_solv^– and Cl₂^•– on the nanosecond to microsecond time scale. Experiments were performed using the Brookhaven National Laboratory Laser-Electron Accelerator Facility (LEAF)¹ and utilizing a recently developed high-temperature sample holder.² Decay kinetics of of e_solv^– and Cl₂^•– in 9.41 mM ZnCl₂ in LiCl-KCl are shown in the figure. e_solv^– and Cl₂^•– decays in different time scale due to different reaction kinetics and decay speed increase as temperature.

This work was supported as part of the Molten Salts in Extreme Environments Energy Frontier Research Center, funded by the U.S. Department of Energy (US-DOE), Office of Science, Basic Energy Sciences, at BNL, INL and ORNL under contracts DE-SC0012704, DE-AC07-05ID14517 and DE-FC02-04ER15533, respectively. The Laser Electron Accelerator Facility of the BNL Accelerator Center for Energy Research is supported by the US-DOE Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences under contract DE-SC0012704.

1. Wishart, J. F.; Cook, A. R.; Miller, J. R.; Rev. Sci. Instrum., 2004, 75 (11), 4359.

2. Phillips, W. C.; Layne, B.; Gakhar, R.; Horne, G. P.; Ramos-Ballesteros, A.; Iwamatsu, K.; LaVerne, J. A.; Pimblott, S. M.; Wishart, J. F.; Rev. Sci. Instrum., 2019, in peer review.

Figure 1

Molten fluoride salts have attracted interest as fuel and/or coolant for several different molten salt reactor (MSR) designs. Fluoride salts have many advantageous properties for these applications, such as high ionic conductivity, good heat transfer capacity, and high solubility for actinides. However, the impurities of the fluoride salt corrode the metals of MSR such as H₂O, HF, and metal impurities. Therefore, it is important to monitor the oxides in molten fluoride salt. In this study, LiF-NaF-KF (FLiNaK) salt was used as a less toxic surrogate for FLiBe because of the toxicity of beryllium. In this work, we focused on measuring the concentration of oxide by adding Li₂O in the FLiNaK salt using two electrochemical methods: cyclic voltammetry (CV) and square waver voltammetry (SWV). Moreover, after the electrochemical tests, the salts were analyzed for concentration of oxygen by LECO in order to compare with electrochemical results. We observed an oxidation peak corresponding to oxide using CV with a tungsten working electrode. Then SWV was used to measure the oxidation peak for oxide at various concentrations (0.1 to 2.0 wt.% of Li₂O). SWV can measure the low concentration, because SWV is more sensitive to the oxidation current than CV through eliminating the non-faradaic current.

Molten chloride and fluoride salts possess various properties such as good thermal conductivity and stability, large specific heat and low viscosity which make them important as fuels and/or coolants for molten salt reactors (MSR). Molten chloride salts are also selected as heat transfer and storage fluids to carry thermal energy from a solar concentrator to a steam generator in concentrating solar power (CSP) technologies. In both the applications the nickel-based alloys are used as structural materials. However high operating temperatures and extremely corrosive nature of molten salts promote faster corrosion of the structural materials. Hence, the corrosion behavior of the structural materials needs to be understood and monitored with time when exposed to the corrosive molten salt environment. Electrochemical techniques are useful for understanding the corrosion mechanism of metals and alloys in molten salts. Among them, electrochemical noise (ECN) is a simple in-situ technique to monitor the corrosion in real time without applying any external signals to the metal/alloy in corrosive environments. A unique advantage of ECN technique is the possibility to detect and analyze the early stages of the localized corrosion process during corrosion initiation as well as a type of corrosion.

The corrosion of metals and alloys in molten chloride is strongly affected by impurities (O₂, H₂O) present in the atmosphere, temperature changes, and concentrations of various chemical impurities (metal oxides and metal chlorides). These impurities have been considered as strong oxidizing agents for metals and alloys in molten chloride salts. Chromium is a major alloying element in nickel-based alloys. Cr can be oxidized into metal chlorides CrCl_x (x=2,3) due to their favorable Gibbs free energies of the formation as compared to NiCl₂ at 623 K. Cr³⁺ and Cr²⁺ coexist in the molten salt during corrosion of nickel-based alloys, while Cr³⁺ is the primary valence state. It is important to study the effect of these multi-valent ions on corrosion of metals/alloys in molten salts. To the best of our knowledge, no study on the effect of Cr³⁺ on the metals/alloys in molten chloride medium has been reported.

The effect of CrCl₃ on the corrosion behaviour of Ni and Ni-20 wt %Cr (NiCr) alloy in purified molten ZnCl₂ at 623 K was investigated for the first time using ECN measurements in an argon-filled glovebox. ECN measurements were carried out in a three-electrode setup using two working electrodes and platinum as a reference electrode in molten CrCl₃-ZnCl₂ at 623 K. Ni and NiCr corrosion behavior was examined in two different combinations : 1) identically coupled electrodes (Ni-Ni and NiCr-NiCr vs Pt) and 2) galvanically coupled electrodes (Ni-NiCr vs Pt). Different types of current and potential noise fluctuations have been identified for pitting and other forms of corrosion. Analysis of fluctuations in current and potential noise in Ni and NiCr in all the three combinations concluded that Ni and NiCr exhibited localized corrosion. The addition of CrCl₃ to ZnCl₂ accelerated the dissolution of Cr and Ni at the metal-salt interface. After ECN studies, the working electrodes were examined by scanning transmission electron microscopy (STEM) coupled with energy dispersive X-ray spectroscopy (EDS) and electron energy loss spectroscopy (EELS). STEM/EDS/EELS observations confirmed pit formation, Ni and Cr dissolution, and Cr depletion at the salt-metal interfaces. Ni-rich and Cr-depleted precipitates were observed in the salt with a galvanically coupled NiCr electrode after ECN studies. The in-depth ECN results will be discussed in conjunction with micro- and nano-scale imaging of dissolution of Ni and Cr elements at the metal-salt interface, and plausible corrosion mechanisms will be proposed. The present ECN results on Ni and NiCr are compared with ECN studies in pure ZnCl₂ salt.

This work was supported as part of the Molten Salts in Extreme Environments Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science.

1. Introduction

The consumption and demand of Nd-Fe-B magnets, so-called neodymium magnets, is increasing year by year. Since the rare earth elements such as Nd can be obtained from sparse mineral resources, the technology for the recycling system of Ne-Fe-B magnets are highly required. The authors have been developing a technique to separately recover Nd and Dy from used Ne-Fe-B magnets using alloy diaphragm and molten salt [1].

Although the electrochemical behavior of Nd and Dy was intensively investigated, the Ne-Fe-B magnet contains B as well as Fe, and the behavior of B in molten salt has not been clarified yet. Therefore, in this study, we investigated the behavior of B in LiCl–KCl eutectic melts at 723 K. The LiCl–KCl eutectic melt was chosen because a high separation ratio of Nd and Dy was obtained in our past experiments [2].

2. Experimental

The experiments were conducted in LiCl–KCl (44:56 wt%) eutectic melt in a dry Ar atmosphere at 723 K inside a glove box with an electronic furnace. 300 g of LiCl–KCl was vacuum dried at 473 K for more than 24 hours before elevating the temperature. As working electrodes, Ni, and Ag wire (φ1 mm) electrodes were used for cyclic voltammetry. Ni and Ag plate electrodes (5 mm×15 mm×0.1 mm(t)) were used for electrodeposition. A glassy carbon rod was used as a counter electrode. A reference electrode was an Ag⁺/Ag electrode, which was prepared by immersing an Ag wire in a LiCl−KCl melt containing 1 mol% of AgCl set in a Pyrex glass tube. The potential of the reference electrode was calibrated by the Li⁺/Li potential obtained by open circuit potential measurement on a Mo wire (φ1 mm) electrode just after electrodepositing Li metal. All the potential hereafter is described in reference to this Li⁺/Li potential.

Cyclic voltammetry was conducted at KBF₄ concentrations of 0, 0.5 and 2.0 mol%. Potentiostatic and galvanostatic electrolysis was conducted at KBF₄ concentrations of 2.0 mol%. After the electrolysis, the plate electrodes were rinsed by distilled water, and then characterized by SEM and XRD.

3. Results and Discussion

3-1. Cyclic Voltammetry

Cyclic voltammograms (CV) obtained by Ni and Ag electrodes are shown in Fig.1(a) and (b), respectively. In the case of Ni electrode, cathodic current was observed at around 1.2 V shown "a" in the figure. The couple of peaks "b" and "c" at around 0.0 V correspond to Li metal deposition and dissolution, respectively. In addition to current peak "a", a small current peak "e" appeared at 1.8 V, which seems to correspond to a formation of Ni-B compounds. Except for the peak "e", essentially the same results were obtained in the case of Ag electrode. In both cases, the peak current at around 1.2 V increased with the concentration of KBF₄ (shown as "a" in the figures). This peak can correspond to the reduction of B(III) to B(0), although further investigation is needed to confirm it.

3-2. Characterization of Products Prepared by Potentiostatic and Galvanostatic Electrolysis

Fig. 2 shows a microscope image and a surface SEM image of the deposit on the Ni plate electrode obtained by potentiostatic electrolysis at 1.2 V, Q= −50 C. The granules with a diameter of c.a. 1-2 μm were observed. From XRD measurement, the diffraction peaks coincided to Ni, Ni₂B, and NiCl₂ was confirmed. Weak peaks that might correspond to B were partially observed.

Next, the galvanostatic electrolysis was conducted at −9.8 mA cm⁻², Q= −15 C for a Ni and an Ag plate electrode, respectively. In the case of Ni electrode, Ni, Ni₂B, NiCl₂ and KCl were detected by XRD analysis. This is almost the same result as the potentiostatic electrolysis at 1.2 V. Ag was selected as the substrate because no alloys of Ag and B exist in phase diagram. From the XRD results, only clear diffraction peaks coincided to Ag was confirmed. Other peaks were also detected, but they did not correspond to elemental B or other expected compounds. From the observation by sight, some deposits were actually formed. Therefore, the deposit can be amorphous B. In the future, further analysis and experiments are needed to clarify the behavior of B on Ag electrode.

Acknowledgement

This work was partly supported by the New Energy and Industrial Technology Development Organization (NEDO) of Japan.

References

[1] T. Oishi, H. Konishi, T. Nohira, M. Tanaka and T Usui, Kagaku Kogaku Ronbunshu, 36, 299–303 (2010).

[2] T. Oishi, M. Yaguchi, Y. Katasho and T. Nohira, Rare metal technology 2020, 151–156 (2020).

Figure 1

To further the development of sustainable energy, it is necessary to understand material behavior under extreme environments. The need to transport high-temperature heat in next-generation nuclear or large-scale solar thermal power plants drives increased interest in molten halide salts as a heat transfer fluid. Specifically, molten KCl-MgCl₂ and related mixtures are attractive, as they have some of the highest volumetric heat capacities of viable systems at 700 ^oC and lowest cost. However, the prevalent challenge for molten KCl-MgCl₂ is corrosion with the encasing structural Ni-based alloys. Literature has shown that this corrosion generally proceeds through the selective dissolution of Cr along with intergranular corrosion. Our prior work showed that in addition to intergranular corrosion, a percolation dealloying phenomenon can contribute to the morphological changes of metal corrosion in molten salt; this dealloying process leads to a porous structure formation instead of corrosion along the grain boundaries. However, the underlying conditions and kinetics that determine the dominating phenomenon remain unclear.

In our present study, binary Ni-20Cr (wt. %) alloy was studied in molten KCl-MgCl₂ (50 – 50 mol. %) at elevated temperatures (500, 600, 700 and 800 ^oC) by multimodal X-ray and electron microscopy techniques. In situ synchrotron X-ray nano-tomography was used to observe the real-time morphological evolution. 3D visualizations present apparently different pathways for the material's morphological evolution at different temperatures. With quantification of morphological characteristics, including relative volume change, porosity, feature size distribution and corrosion distance propagation as a functions of temperature and time, possible rate-determining factors are discussed, including long-range diffusion-control (such as dissolved Cr ions diffusing outward or salt ions diffusing inward) vs. interface-control (such as surface diffusion or reaction at the interfaces). Furthermore, by X-ray Absorption Near-Edge Structure (XANES) spectroscopic imaging, the oxidation state of Ni in the remaining sample was analyzed. In addition, high-resolution Transmission Electron Microscopy with Energy-Dispersive X-ray Spectroscopy was applied to observe the corrosion products at the surface of the alloy and salt penetration at the nanoscale. Overall, this study reveals the temperature-dependent corrosion behavior of Ni-20Cr alloy in molten KCl-MgCl₂ salts, which could be applied to improve the materials used in large-scale energy facilities such as nuclear and solar power plants, highlighting the importance of considering temperature-dependent kinetics routes. On the broader impact, the knowledge could be applied to a wider community, such as when utilizing the molten salt dealloying phenomena to fabricate nano-/mesoporous metals for functional applications.

This work was supported as part of the Molten Salts in Extreme Environments (MSEE) Energy Frontier Research Center (EFRC), funded by the U.S. Department of Energy Office of Science.

１．Introduction

Rare earth elements (REs) were used in many industrial materials in the past few decades. Especially, the demand for Nd–Fe–B magnets has rapidly increased due to the popularization of electric vehicles (EVs) and hybrid electric vehicles (HEVs) in recent years. Since the magnets in motors need to work under high temperature environment, the addition of Dy to magnets is necessary to maintain their strong magnetic properties. However, the supply risk of Dy is high due to its insufficient reserves and uneven distribution. Based on this background, an efficient recycling process of Dy from Nd magnet scraps is a social demand.

We proposed a separation / recovery method using the differences of the formation potential and formation rate of RE–Ni (RE=Nd, Dy) alloys [1]. In our previous reports, we have already studied on the formation of RE–Ni alloys and separation ratio of Dy/Nd in molten LiCl–KCl at 723 K [2], NaCl–KCl at 973 K [3] and LiF–CaF₂ at 1123 K [4]. Although the higher separation ratio was achieved in the chloride systems at 723–923 K, alloy formation rate was higher in the fluoride system at 1123 K.

In the present study, we focused on a low vapor pressure chloride salt CaCl₂, to achieve both high separation ratio of Dy/Nd and high alloy formation rate. Electrochemical formation of RE–Ni alloys was investigated in CaCl₂–RECl₃ at 1123 K.

２．Experimental

The experiments were conducted in a dry Ar atmosphere at 1123 K. Reagent-grade CaCl₂ with 1 mol% NdCl₃ or DyCl₃ was loaded in a graphite crucible. The crucible was placed at the bottom of a stainless-steel vessel in an air-tight Kanthal container and dried under vacuum at 773 K for 24 h. Mo and Ni wires were used as the working electrodes. A carbon rod and a Ni²⁺/Ni electrode were used as the counter and reference electrodes, respectively. The potential was calibrated by the deposition potential of Ca metal on a Mo wire. Electrochemical formation behaviors of RE–Ni alloys were investigated by cyclic voltammetry. Samples prepared by potentiostatic electrolysis were analyzed by XRD and SEM/EDX.

３．Result and discussion

Fig. 1 shows cyclic voltammograms at a Ni electrode measured before and after the addition of 1.0 mol% RECl₃. The cathodic currents increase from around 1.0 V (vs. Ca²⁺/Ca) in both the NdCl₃-added system and the DyCl₃-added system, suggesting the formation of RE–Ni alloys. For the NdCl₃ system, the cathodic current largely increases from 0.48 V, which is likely attributed to the formation of Nd–Ni intermetallic compound with a high Nd concentration. Several anodic current peaks probably correspond to the dissolutions of Nd from different Nd–Ni alloy phases. For the DyCl₃ system, a steep increase of cathodic current is observed from 0.51 V, indicating the formation of Dy–Ni intermetallic compound with a high Dy concentration.

To confirm the formation of RE–Ni alloys, potentiostaic electrolysis was conducted at a Ni plate at 0.50 V for 1 h in both the NdCl₃ and DyCl₃ systems. Fig. 2-a shows a cross-sectional SEM image of the sample obtained in NdCl₃ system. An alloy layer with the thickness of approximately 10 μm was observed. From XRD and EDX results, the alloy layer was confirmed to be NdNi₅. Fig. 2-b shows a cross-sectional SEM image of the sample obtained in DyCl₃ system. Ni plate was completely alloyed with Dy and the thickness of plate increased from 100 μm to 200 μm due to the alloying. The formation of DyNi₂ was confirmed by XRD and EDX. Based on these results, selective formation of Dy–Ni alloy can be expected at 0.50 V in molten CaCl₂–NdCl₃–DyCl₃.

Acknowledgement

This work was supported by Grant-in-Aid for JSPS Fellows 19J20301.

The present address of Kouji Yasuda is Graduate School of Engineering, Kyoto University.

References

[1] T. Oishi, H. Konishi, T. Nohira, M. Tanaka and T. Usui, Kagaku Kogaku Ronbunshu, 36, 299 (2010).

[2] H. Konishi, H. Ono, T. Nohira and T. Oishi, ECS Trans., 50, 463 (2012).

[3] K. Yasuda, K. Kondo, S. Kobayashi, T. Nohira and R. Hagiwara, J. Electrochem. Soc., 163, D140 (2016).

[4] T. Nohira, S. Kobayashi, K. Kondo, K. Yasuda, R. Hagiwara, T. Oishi and H. Konishi, ECS Trans., 50, 473 (2012).

Figure 1

High-temperature molten salts are widely used in production and refining of non-ferrous and rare metals, in nuclear technology and solar storage systems. However, application of such technologies is limited by the problem of finding suitable corrosion resistant materials capable of withstanding a prolonged contact with molten salts.Current study is focused on the corrosion of metallic materials in 3LiCl-2KCl-based melts.

Metallic tantalum and molybdenum, nickel-chromium-molybdenum alloys (VDM Alloy C-4, Hastelloy G-35 and specially designed KhN62M alloy) were chosen as the objects of the investigation. Corrosion tests were performed in 3LiCl-2KCl and 3LiCl-2KCl-UCl₃ (5 wt. % U) melts. The corrosion tests run for 100 h, and the temperature was varied from 450 to 750 °C. The corrosion rates were determined by both gravimetric technique and chemical analysis of quenched melts. The last technique is very useful in case of alloys formation on the surface of tested samples.

It was found that all the above metallic materials can be used in 3LiCl-2KCl melts at 450 and 550 °C. Increasing working temperature led to intensification of corrosion processes. This effect was more pronounced for nickel-based alloys where increasing temperature to 650 °C also resulted in changing character of the corrosion. The undesirable intergranuluar corrosion was noticed for the nickel-based alloys at 750 °C and this was induced by the formation of secondary phases at the grain boundaries.

Metallic tantalum and molybdenum were subjected to only gradual etching even at 750 °C. Their corrosion rates did not exceed 0.01 mm/year and these materials can be recommended for the application in contact with molten chlorides at 750 °C.

Addition of uranium(III) chloride to the melt led to increasing corrosion rate for all the materials studied. The corrosion mechanism, however, did not change much with temperature. The presence of oxidant (U³⁺) resulted in acceleration of corrosion etching. The rate of the exchange reaction between uranium chlorides and components of the materials can be accelerated by the formation of U-Ni alloys. Formation of metallic uranium-containing layers on the surface of nickel-based samples deserves a separate attention as an element of possible currentless protection coatings.

The corrosion resistance of tantalum and molybdenum is also sufficient at 750 °C in 3LiCl-2KCl-UCl₃ (5 wt. % U). However, it should be remembered that their mechanical properties, weldability, and high-temperature resistance in air are noticeably worse compare to the nickel-based alloys.

In the recent years, much attention has been focused on the synthesis of various carbon nanomaterials (CNMs) because of their wide range of potential applications. These materials have been fabricated by different methods such as the laser and arc evaporation of graphite, catalytic pyrolysis of hydrocarbons, disproportionation of CO on metal-catalysts etc. Promising method for synthesis of CNMs is high-temperature electrochemical synthesis (HTES) in molten salts. Although much research has been devoted to this subject, little is known about the mechanism of HTES and which fundamental reactions taking place during this process.

In this work an attempt to confirm the mechanism of cation-anion interaction between the CO₃^2- anion and the strongly polarizing cations (Li⁺, Ca²⁺, Mg²⁺) is made. The mechanism of these interactions has a complex nature. Cation-anion interaction can result either in the formation of cationized anions (metal complexes) or in anion dissociation both under the direct influence of cations and through the intermediate stage of formation of "short-lived" metal complexes.

To test this hypothesis, the quantum chemistry modeling and the cyclic voltammetry experiments were done. It has been shown by the quantum chemistry method that by changing the cationic composition of the electrolyte, one can transform anionic carbonate complexes into a new active state – cationized carbonate complexes. Experimental confirmation of cation-anion interaction was studied in the chloride melts Na,K,Cs|Cl, Na,K,Rb|Cl, Na,K|Cl by the cyclic voltammetry method at temperature ranges of 550–580 °C and of 700–800 °C, which are much lower than the thermal decomposition temperature. The electrolyses was carried out in electrolytes under potentiostatic conditions at a potential of -1.1 V at the concentrations of lithium, calcium and magnesium chlorides, corresponding to the clear observation of cathodic waves. According to the data of a chemical and an X-ray phase analysis a black powder of the cathodic product was carbon. The morphology of electrolytical carbon is presented in the figure below.

Based on the obtained results the following conclusions can be done:

1. In the Na,K,Rb|Cl melt containing weakly polarizing Na⁺, K⁺, Rb⁺ cations, the CO₃^2- anion shows no electrochemical activity in the temperature range of 570–700 °C. This is accounted for by the large values of the activation barriers to the two- and four-electron reduction of CO₃^2-.

2. The addition of strongly polarizing cations (cations with high specific charge) results in the activation of the carbonate ion, which is caused by a large excess of Li⁺ and Ca²⁺ cations. In the case of Mg²⁺ cations, the carbonate ion shows electroactivity at much lower concentrations.

3. Independent of the polarizing power of the cation (melt acidity), the CO₃^2- reduction process in molten chlorides occurs in the same potential range of -0.7 to -0.9 V versus a silver reference electrode.

4. The CO₃^2- electroreduction process under the action of strongly polarizing cations occurs at temperatures much lower than the thermal decomposition temperature of the corresponding carbonate. This suggests that the formation of CO₂ is not the result of thermal decomposition, but is a consequence of the polarizing action of cations.

Figure 1

Molten salt nuclear fast nuclear reactor (MSNFR) is one of the prospective design in frame of Generation IV concept. This technology is also required for reprocessing of spent nuclear fuel (SNF) in molten salts to extract valuable components. Recycling SNF consists of extraction uranium and plutonium with the disposal of minor actinides and fission products. These stages need the materials that retain their corrosion and mechanical properties for long time under the influence of high temperatures, radiation fields and contact with molten salts. Various steels and alloys were widely studied to assess their use under such extreme conditions. Alternative materials include ceramics, composite and carbon materials, which have high corrosion resistance in various media and can be used up to 1000 °C.

In the present work the corrosion and mechanical properties of nitride ceramics (Si₃N₄ and BN) and carbon-based materials (carbon-carbon composite material (C/C) and high-density carbon) were studied in a molten mixture of lithium, sodium, and potassium fluorides (FLiNaK) in the temperature range of 550–750 °C under inert atmosphere. Corrosion tests were performed under static conditions, and the duration of each test was 100 h to enable the comparison of the experimental data.

It was found that carbon-containing materials (C/C and high-density carbon) showed high corrosion resistance in the melt. C/C samples had lower corrosion rates (less than 0,1 mm/year in the entire temperature range), while the high-density carbon showed better mechanical properties. The main disadvantage of these materials is their impregnation by molten salt due to relative high porosity. However, C/C composite and high-density carbon are promising structural materials for salt media based on FLiNaK in the selected temperature range.

Tested nitride ceramics (BN, Si₄N₃) demonstrated relatively low corrosion resistance in fluoride systems compare to carbon-containing and metallic materials. The corrosion rates of these materials in FLiNaK exceeded the value of 1 mm/year. The plastic properties of nitride ceramics were also poor. Possible application of such materials for MSNFR and SNF reprocessing is under further investigation.

Method of currentless transfer in molten salts was used for synthesis of tantalum and niobium carbides coatings and crystals of molybdenum carbide on carbon fibers. Temperature of process was varied from 800 up to 900 °C, and time of process from 0.5 up to 24 hours.

Figure 1. Micrographs of carbon fiber with TaC and NbC coatings and Mo₂C crystals (from left to right), obtained in molten salts melts by currentless transfer.

The splicing of fibers with each other was not observed; coatings were uniform in a cross section as well as along the fiber. Thickness of TaC and NbC coatings on carbon fiber was 50 – 250 nm. Size of crystals of Mo₂C was approximately several micrometers.

Investigation of refractory metal (Ta, Nb, Mo) carbides on carbon fiber of the Carbopon-B-22 brand was carried out by impedance spectroscopy method. The values of the standard rate constants of charge transfer and capacities of the double electric layer were determined for synthesized materials in water solution with hydrogen peroxide.

Coatings and crystals of refractory metal carbides on carbon fibers can be used as catalysts and electrocatalysts. The kinetics of the electrocatalytic decomposition of hydrogen peroxide on the surface of TaC/C, NbC/C and Mo₂C/C composite materials was studied. This reaction was investigated on measuring the volume of released oxygen. The initial carbon fiber of the Carbopon-B-22 brand was used as the cathode. The composite materials made of carbon fiber coated with tantalum carbide TaC (NbC) or modified by molybdenum carbide Mo₂C served as the anode. The reaction order was established, the rate constants and activation energies of the hydrogen peroxide decomposition were calculated. The composition with the highest electrocatalytic activity was determined.

Figure 1

Mg and Li are smelted by electrolysis in molten chloride, and carbon-based material is used for the anode. Carbon-based anode reacts electrochemically with oxide ions dissolved in the bath, causing oxidative consumption. Therefore, the development of the so-called inert anode is desired. The authors proposed an inert anode by forming a stable passivation film on it and investigated the anodic behavior and the conditions to form a film on MoSi₂ in molten LiCl-KCl containing Li₂O. In this study, the behavior of MoSi₂ anode and film formation on it have been investigated in molten MgCl₂-NaCl-CaCl₂.

MgCl₂-NaCl-CaCl₂ electrolytic bath with about 1 mass% MgO was used at about 973K in an Ar atmosphere. A MoSi₂ rod was used as working electrode. A Cu plate and Ag/AgCl were used as counter and reference electrode, respectively. Anodic behavior of MoSi₂ was investigated by cyclic voltammetry and potentio-static electrolysis. The change in the MoSi₂ surface with electrolysis was analyzed by XRD and SEM-EDX.

Fig. 1 shows a cyclic voltammogram in the bath with MgO. Anodic current density increased at about 1.6 V (vs. Mg / Mg²⁺) and an anodic current peak appeared at about 1.8 V (vs. Mg / Mg²⁺). After the peak, the anodic current density sharply decreased. Since this anodic current peak was not seen without addition of MgO, it was indicated that the passivation phenomenon occurred in the bath with MgO.

Fig. 2 shows the SEM image and element distributions of the cross-section of MoSi₂ after potentio-static electrolysis for 4h at 3.1V (vs. Mg / Mg²⁺) in the bath with MgO. A passivating film of about 3 µm-thick was seen on the anode surface, and Si was found on it. Mg and O were concentrated in the passivation film, and SiO₂ and MgSiO₃ were detected by XRD. From the quantity of electricity during the potentio-static electrolysis, the thickness of the passivation film and the weight change, the current contributions for the passivation film formation, and for Mo dissolution were about 2% each, while about 96% of the electricity was shown to be spent by other reactions. Considering the electrolysis condition, 3.1V (vs. Mg/Mg²⁺), most of the current seemed to be used for the gas generation. It is concluded that MoSi₂ is potential for the inert anode in molten chloride-containing oxide ions.

Figure 1

Modern aerospace engineering relies on Al-based alloys with enhanced high-temperature strength properties and oxidation resistance. These requirements can be fulfilled by modifying alloys' structure employing rare earths metal doping. Direct addition of metallic rare earths to aluminium is complicated due to high chemical activity of molten rare-earth metals and significant vapor pressure of aluminium at elevated temperatures. To overcome these drawbacks, lanthanide-containing master alloys with relatively low REE content can be employed instead of pure scandium.

Currently there are three methods for preparing aluminium-based alloys containing rare earths, i.e. direct fusing, molten salt electrolysis and exchange or metallothermic reduction reactions. In the last case, Al–Gd and Al–Sc master alloy can be prepared by aluminothermic reduction of GdF₃ or ScF₃ in a molten binary NaF–KCl salt mixture at 800–1050 °С.

In the present work we suggest the modification of the above mentioned method for aluminium–cerium and aluminium–neodymium master alloy production with a lanthanide concentration of 1 to 7 wt. %. Based on thermodynamic calculations we proposed that the high-temperature exchange reaction between metallic aluminum and corresponding fluoride containing salt electrolyte should be carried out at temperatures from 740 to 780 ° C to provide the higher recovery of REE. The influence of the time of the exchange reaction and «aluminium–salt» ratio on the process to obtain ligatures of various compositions is analyzed. The yield higher than 85 % is achieved.

The microstructure of the alloys represents a metal matrix of aluminum in which intermetallic compounds of Ln₃Al₁₁ compositions are distributed In the case of an exchange reaction between aluminum and cerium fluoride, the formation of a new phase, presumably related to the Al₄Ce intermetallic compound, is found.

An innovative production process of Al metal is desired to reduce CO₂ and perfluorocarbon gas generation. Our previous study showed that calcium titanate and silicate dissolved in molten CaCl₂. Ti and Si metals could be electrodeposited, and it was indicated that Ti and Si reductions depended on the molar ratio of CaO to TiO₂ or SiO₂. In this study, direct production of liquid Al metal was attempted in molten CaCl₂ with various calcium aluminates.Molten CaCl₂ at 1373 ~ 1573K in an Ar atmosphere was used as a bath. Calcium aluminate (molar ratio of CaO to Al₂O₃, R_CaO/Al2O3 =1.0, 12/7 or 3.0) was prepared by sintering a molded mixture of CaO and Al₂O₃, and it was added in CaCl₂ beforehand. Mo and graphite rods were used a working and counter electrode, respectively. Mo wire was used as a quasi-reference electrode. The cathodic reaction was investigated by cyclic voltammetry, and potentio-static electrolysis was carried out. The electrodeposit was analyzed by XRD and SEM-EDX.Although no sharp current peak was seen, cathode current increased overall with calcium aluminate addition. The shape of CVs slightly changed with R_CaO/Al2O3, but R_CaO/Al2O3 seemed not to affect Al reduction significantly.Al and Al-Mo alloy were detected in the electrode after the electrolysis at -1.8V at 1573K regardless of R_CaO/Al2O3, but the Al amount was very small. So, a Mo cathode with BN receiver in the bottom shown Fig.1 was used to prevent deposited Al metal from running down from the cathode. Al metal covered the surface by electrolysis, and a thick Al-Mo alloy layer was found in its cross section. It was shown that calcium aluminate dissolved in molten CaCl₂ above 1573K, and that Al metal could be deposited electrochemically. The molar ratio of CaO to Al₂O₃ slightly affected the Al deposition though its influence have not clarified well.

The results of the electrochemical reduction of zirconium dioxide in molten electrolytes based on calcium or magnesium chloride on a liquid gallium cathode, bypassing the stage of granulation and sintering with carbohydrates, are presented. The liquid gallium cathode provides not only reliable contact with zirconium dioxide, but also favourable conditions for its reduction. The contact area of the gallium cathode with fine oxide powder is much larger than the contact area of the granulated and sintered zirconium dioxide with a solid conductor. This ensures a more uniform cathode polarization. Due to the lower specific mass, zirconium dioxide is located on the surface of gallium cathode, the convective movement of which provides more intense mass transfer at the interface of the phases and removal of recovery products from the zone of electrode reaction. Products of electrolysis under such conditions do not block neither the surface of zirconium dioxide nor the surface of the cathode. Zirconium, which is formed during the renewal, due to a larger specific mass, precipitates to the bottom of the electrolyser, and the layer of gallium protects it from interaction with the components of the molten electrolyte. In addition, due to the formation of alloys, the reduction of metal cations on liquid cathodes proceeds at more positive potentials than on solid indifferent cathodes, which reduces the specific energy consumption by electrolysis. The results of voltammetric studies confirm this conclusion.

The reduction product is fine-grained zirconium powder with an average particle size of 1—3 microns, and purity of 99.9 %. As the density of the current increases, the value of the specific surface of the powder, the specific volume of the micropore and their average radius decrease. The degree of extraction depends on the composition of the electrolyte mixture and naturally decreases when replacing cations in the melts both on the basis of calcium chloride and on the basis of magnesium chloride in the following sequence Na⁺ > K⁺ > Li⁺. The melt based on compounds of calcium and sodium chloride provides the best performance. The removal degree of zirconium from such melt reaches 77 %.

The electrodeposition of Al has been studied in various ionic liquids represented by the 1-ethyl-3-methylimidazolium chloride-AlCl₃ system. However, AlCl₃-based (chloroaluminate) ionic liquids are generally handled in an inert atmosphere of nitrogen or argon because they are highly hygroscopic and susceptible to hydrolysis through the reaction with moisture in the air. To maintain an inert atmosphere, a closed system that typically uses a glove box is required, but operation in a closed system leads to low productivity and high cost. Therefore, electrodeposition in an inert atmosphere is unsuitable for mass production on an industrial scale. To alleviate this problem, we have previously studied the feasibility of Al electrodeposition in dry air using chloroaluminate ionic liquids. Because the electrodeposition process in dry air does not require a glove box, it is considered to be more productive than the electrodeposition process under an inert atmosphere.

In our previous study, four different chloroaluminate ionic liquids consisting of different organic cations ([cation]Cl-AlCl₃) with a molar ratio of [cation]Cl : AlCl₃ = 1 : 2 were used as Al electrodeposition baths to study Al electrodeposition behavior in dry air. The cations used were 1-ethyl-3-methylimidazolium (EMI⁺), 1-butylpyridinium (BP⁺), 1-butyl-1-methylpyrrolidinium (BMP⁺), and trimethylphenylammonium (TMPA⁺). We found that electrodeposition in dry air using the bath composed of EMI⁺ or BP⁺ could not produce an Al film covering the entire surface of the substrate, whereas electrodeposition using the bath composed of BMP⁺ or TMPA⁺ could produce an Al film covering the entire surface of the substrate in dry air as well as in an inert atmosphere.

The reason for the difference in the results of Al electrodeposition in dry air according to cationic species has not yet been ascertained, but it is speculated that dissolved oxygen has some influence on the Al electrodeposition. In the present study, we measured the concentration of oxygen dissolved in four types of ionic liquids in dry air.

For measuring the dissolved oxygen concentration, ionic liquids with a composition of [cation] Cl: AlCl₃ = 1: 0.9 were used. This composition was chosen because only AlCl₄⁻, which is an ionic species that is difficult to be reduced to Al metal, is present in the bath when the AlCl₃ mole fraction is less than 50 mol% (Lewis base)in chloroaluminate ionic liquids. Consequently, Al electrodeposition does not occur, and low extent of electrochemical reaction other than Al electrodeposition at a potential of 0 V vs. Al / Al (III) or less can be investigated. The concentration of dissolved oxygen was measured by chronoamperometry using two types of electrodes, a macroelectrode and a microelectrode, both of which have a circular active area with a diameter in the order of millimeter and micrometer, respectively. In the case of chronoamperometric measurement with a macroelectrode, the reduction current of oxygen follows the equation in the upper row in Table 1, assuming that the reaction rate is controlled by the planar diffusion of oxygen from the bulk electrolyte to the surface of the electrode. Upon measuring the current–time response, two unknowns remain, namely, the concentration and the diffusion coefficient of dissolved oxygen; and it is not possible to determine the dissolved oxygen concentration through the chronoamperometric measurement using the macroelectrode alone. In the case of chronoamperometric measurement with a microelectrode, the geometry of the diffusion of the reactant to the electrode surface changes to semi-spherical, and the reduction current of oxygen thus follows the equation in the lower row in Table 1. The concentration and diffusion coefficient of dissolved oxygen can thus be determined through the two chronoamperometric measurements.

The chronoamperometric measurements revealed that there was no significant difference in the oxygen concentrations of the ionic liquids with different cationic species. The results that we have obtained so far are as follows: 1) In an inert atmosphere, an Al film covering the entire surface of the substrate can be electrodeposited from a bath containing any cation species. 2) In dry air, an Al film covering the entire surface of the substrate can only be electrodeposited from some baths composed of specific cation species. 3) There is no significant difference in the dissolved oxygen concentrations of the baths in dry air, even though the cation species in the baths are different. In consideration of these results, the reason for the different results of Al electrodeposition in dry air according to the cationic species is neither dissolved oxygen alone nor cationic species alone but the combined action of both dissolved oxygen and the cation.

Figure 1

Molten salts have physical properties that include thermal stability over a large temperature range, low vapor pressure at high temperature, and are nearly immune to radiological effects. Due to these favorable properties, molten salts are being considered as a promising candidate for next generation heat transfer fluids for use in generation IV nuclear reactors. One drawback of molten chloride salts thermo-physical properties is their low specific heat capacity (around 1.0–1.5 (J/(g×K)). It has been shown that addition of nanoparticles in molten salts leads to a large increasing in the specific heat capacity of these salts, once melted with the particles. The aim of this study is to investigate the effect that nanoparticles have on specific heat capacity of zinc potassium chloride solutions. The hollow carbon nanospheres (hCNS, outer diameter is 20–30 nm; wall thickness 7 nm) and zeolite nanoparticles (ZSM-5, size 10–30 nm) were chosen for this study. The colloidal solutions of ZnCl₂–KCl (46 mol% KCl) eutectic melt (T_m.p. = 220 ^oC) with different amounts, 0.3 to 2 wt% of nanoparticles, were prepared by mechanical mixing of required amount of salt with nanoparticles at 350^oC for at least 24 hours until stable colloidal solutions were formed. Specific heat capacity of the samples was measured using a differential scanning calorimeter (HDSC; PT1000, Linseis Inc.) under the argon atmosphere according to established literature procedures. Measurements were carried out between 250–350^oC temperature region with 10^oC/min temperature rate using platinum crucibles with platinum lids for containment. The obtained experimental results (fig. 1) demonstrate that specific heat capacity of ZnCl₂–KCl eutectic melt increases with increasing amounts of both types of nanoparticles used to form the colloidal solution.

Figure 1

A direct synthesis route for high purity anhydrous binary and ternary salt mixtures has been developed. This atom efficient, solvent free process is easily scalable with potential to produce salt eutectics with purity standards required to meet the needs of industrial heat transfer and nuclear applications. The methodology involves mechanochemical mixing of alkali and alkaline earth chloride salts followed by subsequent drying under dry air laminar flow. Each salt mixture was then analyzed for waters of hydration, purity, and oxide content through subsequent thermogravimetric analysis, X-ray diffraction, and strong acid titration. This process is proposed to provide a cost effective, single pot synthesis for both fuel and coolant chloride salt compositions for use in generation IV molten salt reactor systems. This work was supported as part of the Molten Salts in Extreme Environments Energy Frontier Research Center, funded by the U.S. Department of Energy Office of Science.

1. Introduction

Lithium secondary batteries have been widely used as power sources for small electronic devices. However, there are still many challenges for the widespread use as large storage batteries, such as the volatility and flammability of the organic-solvent-based electrolytes, and the uneven distribution and scarcity of lithium and cobalt resources.

Therefore, we have been developing sodium and potassium-ion secondary batteries using FSA-based ionic liquid (IL) electrolytes (FSA = bis(fluorosulfonyl)amide) because the FSA-based ILs have excellent electrochemical stability and ionic conductivities, and sodium and potassium resources are abundant in the Earth's crust [1,2].

In this study, we focus on FTA-based ILs (FTA = (fluorosulfonyl)(trifluoromethylsulfonyl)amide) as promising electrolytes for alkali metal-ion batteries. The FTA-based ILs have lower melting points and higher thermal stability than the FSA-based ones, which is advantageous for broadening the operating temperature range of batteries [3,4]. The physicochemical properties of M[FTA]–[C₄C₁pyrr][FTA] ILs (molar fraction x(M[FTA]) = 0.20; M = Li, Na, K, Rb, Cs; C₄C₁pyrr = N-butyl-N-methylpyrrolidinium) were investigated to discuss the effect of alkali metal-ion species.

2. Experimental

All reagents were handled under argon atmosphere. The M[FTA]–[C₄C₁pyrr][FTA] (x(M[FTA]) = 0.20; M = Li, Na, K, Rb, Cs) ionic liquids were prepared, and their ionic conductivities, viscosities, densities, and electrochemical windows were measured. The electrochemical windows were determined by cyclic voltammetry (CV) measurements using a three-electrode cell. The components of the three-electrode cell were as follows; a copper disk electrode and a glassy carbon disk electrode were used in the negative and positive potential region as the working electrodes, respectively, a platinum mesh electrode was used as the counter electrode, and an Ag⁺/Ag electrode was used as the reference electrode.

3. Result and discussion

Fig. 1 shows the Arrhenius plots of the ionic conductivities for M[FTA]–[C₄C₁pyrr][FTA] ILs. The order of the ionic conductivities (mS cm⁻¹) is Na(1.7) < Li(2.0) < K(2.2) < Rb(2.3) < Cs(2.6) at 298 K. In terms of ion size, the smallest Li⁺ is considered to be the most favorable for the ionic conduction. However, its ionic conductivity is lower than those of K⁺, Rb⁺, and Cs⁺-based systems. One possible explanation is that the higher charge densities of Li⁺ and Na⁺ lead to the strong ion interaction with FTA⁻, resulting in the lower ion mobilities of alkali-metal ion complexes. We also measured their viscosities and densities of the FTA-based ILs at 273–368 K and constructed a Walden plot, which suggested that all the FTA-based ILs do not have a special ion conduction mechanism such as the Grotthuss mechanism.

Fig. 2 summarizes the results of the CV measurements for M[FTA]–[C₄C₁pyrr][FTA] ILs. The order of their electrochemical windows (V) is Na(5.33) < Li(5.45) < K(5.58) < Rb(5.64) < Cs(5.74) at 298 K. The redox peaks were observed in the negative potential regions for all the electrolytes, which corresponds to the deposition and dissolution of alkali metals. On the other hand, irreversible oxidation currents were observed in the positive potential region for all the electrolytes. Since there was little difference in the anode limits for all the electrolytes, these oxidation currents are considered to be the decomposition of FTA⁻ anions [4]. The trends of the physicochemical properties obtained in this study are very similar to those for FSA-based ionic liquids reported by our group [1].

References

[1] T. Yamamoto, K. Matsumoto, R. Hagiwara, T. Nohira, J. Phys. Chem. C, 121, 18450 (2017).

[2] R. Hagiwara, K. Matsumoto, J. Hwang, T. Nohira, Chem. Rec., 18, 1 (2018).

[3] K. Kubota, T. Nohira, R. Hagiwara, H. Matsumoto, Chem. Lett., 39, 1303 (2010).

[4] K. Kubota, H. Matsumoto, J. Phys. Chem. C, 117, 18829 (2013).

Figure 1

Lanthanide elements, which are present in spent fuel from fast nuclear reactors can be converted into molten salts by anodic dissolution. The lanthanide elements are the most difficult fission products to separate from actinides due to their similar chemical properties. Thus a knowledge of the electrochemistry and thermodynamics of lanthanide compounds in molten salt systems is very useful for the understanding the recycling of spent nuclear fuel.

The electroreduction of SmCl₃ in alkali chloride melts (NaCl-KCl, KCl, CsCl) was studied in the temperature range 973-1173 K by different electrochemical methods. The diffusion coefficients (D) for Sm(III) and Sm(II) were determined by linear sweep voltammetry, chronopotentiometry and chronoamperometry methods showing that D decreases with increase of samarium oxidation state, while the activation energy for diffusion increases. Decreasing values of D were also obtained when the cation in the second coordination sphere changed from Na to Cs. It was shown that such changes are due to the decreasing of counter polarizing effect of cations with increasing cationic radius.

The standard rate constants of charge transfer (k_s) for the Sm(III)/Sm(II) redox couple were calculated on the basis of cyclic voltammetry data using Nicholson's equation. The following row of the standard rate constants of charge transfer has been experimentally determined: k_s (KCl) < k_s (CsCl) < k_s (NaCl-KCl). The nature of working electrode on the rate of charge transfer for the Sm(III)/Sm(II) redox couple was studied. It was found that the values of k_s determined at a molybdenum electrode were higher than those at a glassy carbon electrode.

The formal redox potentials E^*_{Sm(III)/Sm(II)} were obtained in alkali chlorides melts from the linear sweep voltammetry data. From the values of the formal redox potentials were calculated the Gibbs energies and equilibrium constants for the reaction:

SmCl_2(sol.) +1/2 Cl_2(g.) « SmCl_3(sol.) (1)

It was determined the increasing of the formal entropy for the reaction (1) in transition from NaCl-KCl to CsCl melt that associated with a greater degree of ordering of the reaction products due to the complex formation.

Alkali chloride melts are considered as prospective working media for non-aqueous pyrochemical reprocessing of spent nuclear fuels (SNFs). Separation of fissile materials from fission products in pyrochemical reprocessing can be achieved electrochemically and amongst all the fission products zirconium has the closest electrochemical properties to uranium. Uranium and plutonium fission produces several zirconium isotopes (from Zr-91 to Zr-97) and depending on the reactor neutron spectrum, nuclear fuel type, burnup and cooling time SNF arriving for reprocessing can contain ca. 5–13 kg of zirconium (as the fission product excluding cladding) per ton. Electrochemical separation of uranium and zirconium in fused salts is a challenging task.

The present work was devoted to studying the electrochemical behavior of zirconium and uranium in 3LiCl–2KCl based melts using cyclic voltammetry and cathodic polarization. The experiments were performed in LiCl–KCl–ZrCl₄, LiCl–KCl–UCl₄, LiCl–KCl–UCl₃ and LiCl–KCl–ZrCl₄–UCl₄ melts.

On a solid tungsten electrode zirconium(IV) ions were reduced to Zr(0) in two stages and the metal deposition potentials were between –2.17 and –2.07 V (at 532–637 ^oC) vs. Cl^–/Cl₂ couple. Changing the solid electrode to liquid zinc, gallium or gallium–zinc eutectic alloy (3.64 wt. % Zn) resulted in significant shift of zirconium deposition potential in the positive direction. Examples of the polarization curves are shown in the Fig. Polarization measurements performed in LiCl–KCl–ZrCl₄–UCl₄ melt on the Ga–Zn electrode showed that reduction of U(IV) to U(III) and deposition of zirconium occurred at very similar potentials and deposition potential of uranium was significantly more negative. Separation factor for uranium/zirconium couple was also determined.

Fig. Polarization of Ga–Zn eutectic alloy cathode in LiCl–KCl–UCl₄ (524 ^oC) and LiCl–KCl–ZrCl₄ (550 ^oC) melts.

Figure 1

Alkali chloride melts have numerous potential applications in nuclear fuel cycle including pyrochemical reprocessing of spent nuclear fuels, uranium electrowinning and electrorefining. Oxygen is a common technological impurity that can affect uranium speciation and behavior in fused salts. The present work was devoted to studying the reactions of oxygen with solutions of uranium tetrachloride in molten alkali chlorides. The experiments were performed in LiCl–KCl, NaCl–KCl–CsCl and NaCl–CsCl eutectic based melts at 450–750 ^oC. Pure oxygen and argon–oxygen mixtures (containing ca. 1 and 10 % O₂) were used. Amount of oxygen passed through the melt varied from less than one to over 100 moles per mole of uranium present. Effect of moisture (0.4–2.5 % H₂O) presence in oxygen or Ar–O₂ mixtures was also investigated. The course of the reaction was followed by in situ electronic absorption spectroscopy measurements with the spectra recorded at the certain time intervals.

Depending on temperature, cationic melt composition and oxygen-to-uranium molar ration the reaction resulted in oxidation of uranium(IV) to soluble uranyl chloride and/or precipitation of uranium dioxide. Analysis of the spectra provided the information on kinetics of U(IV) concentration change. Increasing temperature, O₂ : U(IV) molar ratio or decreasing mean radius of alkali cations of the solvent melt resulted in faster decrease of U(IV) concentration in the melt.

Under certain conditions U(IV) can be oxidized to UO₂Cl₄^2– without precipitation of UO₂. Therefore sparging the melt with oxygen can be used as a way of separating uranium from certain fission products, for example rare earth elements. Rare earth chlorides react with oxygen yielding oxychlorides or oxides insoluble in alkali chloride melts. Interaction of oxygen with melts containing a mixture of uranium and rare earth chlorides was therefore also investigated and an example of the spectra recorded in LiCl–KCl–UCl₄–NdCl₃ melt is shown in Fig.

Fig. Spectra recorded in the course of reaction of O₂ with LiCl–KCl–UCl₄–NdCl₃ melt at 550 ^oC. Arrows show the direction spectra changed.

Figure 1

Alkali chloride based melts can be used as working electrolytes in prospective technologies of pyrochemical reprocessing of spent nuclear fuels. After extracting fissile materials the melts contain ions fission product elements having more negative electrode potentials than uranium and plutonium. These include rare earth, alkaline earth and alkali metals. One of possible methods for removing rare earths from alkali chloride based melts is phosphate precipitation. Alkaline earth metals also form sparingly phosphates and therefore phosphate precipitation can be potentially used for removing strontium and barium from molten electrolytes.

The present work was aimed at studying the reaction of solutions of strontium or barium chloride in alkali chloride based melts with sodium orthophosphate. The experiments were conducted in LiCl–KCl and NaCl–KCl–CsCl eutectic based melts at 550 ^oC, and in NaCl–KCl equimolar mixture at 750 ^oC.

To determine the conditions required for complete removal of strontium and barium, the initial phosphate-to-barium molar ratio in the melt was set at 0.5; 1; 2; 4; 6 and 8. Residual alkaline earth content in the melt was determined by chemical analysis. Precipitated strontium and barium phosphates were subjected to X-ray powder diffraction analysis.

Selecting a possible method of separating solid precipitate from molten salt requires information on size of particles forming the solid phase. Particle size was determined by laser diffraction and examples of particle size distribution curves for precipitates formed in LiCl–KCl–BaCl₂ melts are shown in Fig. The precipitates consisted of particles ranging from 0.5 to 100 microns. Increasing the initial PO₄^3– : Ba²⁺ molar ratio resulted in increasing particle size but even at the molar ratio of six over 50 % particles were less than 10 μm.

Fig. Particle size distribution curves for barium phosphate precipitated from LiCl–KCl eutectic based melts at 550 ^oC. Initial PO₄^3– : Ba²⁺ mole ratio was 0.5 (line 1); 1.0 (2); 2.0 (3); 4.0 (4) and 6.0 (5).

Figure 1

Today, existing methods of recycling spent nuclear fuel do not solve the problem of separation of transuranium elements and lanthanides. Electrolysis of molten salts is promising for separation of lanthanides and actinides. To solve this problem, it is necessary to have experimental data on the electrochemical behavior of various lanthanides in molten salts.

The aim of this study is the electrochemical investigation by cyclic voltammetry of the Sm(III)/Sm(II) redox couple in NaCl-KCl, KCl and CsCl melts containing SmF₃.

Electrochemical studies were carried out in a temperature range of 973-1173 K by cyclic voltammetry using an AUTOLAB PGSTAT 20 potentiostat with a package of application programs GPES (version 4.4.). The sweep rate (v) varied between 0.1 up to 3.0 V s^-1. The melt container was a glassy carbon crucible of SU-2000 brand, which also served as an auxiliary electrode. Tungsten wire was used as a working electrode and platinum wire was utilized as a quasi-reference electrode.

It was found in NaCl-KCl-SmF₃, KCl-SmF₃ and CsCl-SmF₃ melts that the recharge process was reversible up to the scan rate 1.0 V s^-1. In this range of sweep rate the diffusion coefficients (D) of Sm(III) were calculated using the Randles–Shevchik equation. The diffusion coefficients decrease with a change of the composition of the second coordination sphere from sodium to cesium. Similar dependencies for D values are well known. It is associated with a decrease of the counter-polarizing effect during the transition from Na to Cs, which in turn causes a decrease of the metal – ligand bond length and increase the strength of samarium complexes.

A transition from reversible to quasi-reversible process was found at v > 1.0 V s^-1 in NaCl-KCl-SmF₃, KCl-SmF₃ and CsCl-SmF₃ melts.

The standard rate constants of charge transfer (k_s) of the redox couple Sm(III)/Sm(II) were determined by cyclic voltammetry in all studied melts by using the Nicholson's equation, which is valid for quasi-reversible processes.

The following series of the standard charge transfer constants was found k_s(CsCl)<k_s(KCl)<k_s(NaCl–KCl). According to the theory of elementary charge transfer, the smaller and stronger bond complexes require a higher rearrangement energy, and in consequence, the charge transfer proceeds at a slower rate. Therefore, a decrease of the standard rate constants for the redox reaction would have been expected, which is in an agreement with obtained experimental results.

The electrodepositions of tungsten and its alloys has been of considerable interest because of usual properties of the metal and alloys. The electrodepositions of W and Co from water solutions and Na₂WO₄-WO₃-CoO at 1023 K were investigated in more works. The possibility of electrodeposition of W-Co alloys from carbamide melts at 408 K has been examined. When studying the electrochemical behaviour of cobalt and tungsten oxides and its compounds (WO₃, Li₂WO₄, Na₂WO₄, K₂WO₄) in molten carbamide, it can be concluded that maximum limiting currents are typical of the system (NH₂)2CO-Li₂WO₄. Cyclic voltammograms showed the potentials of deposition of tungsten and cobalt. Layered Co-W coatings on nickel cathodes have been obtained by the electrolysis of the molten system (NH₂)₂CO-CoO- Li₂WO₄ at current densities of 5-10 mA/cm². The cathodic products were analyzed by XRD and JEM-2100F SEM.

Fig. SEM of the surface morphology of W-Co coating deposited by electrolysis from (NH₂)₂CO-CoO- Li₂WO₄ melt.

Figure 1

The goal of the present investigation was to study of Mg²⁺, Ca²⁺, Sr²⁺ and Ва²⁺ strongly polarizing cations influence on the standard rate constants of charge transfer (k_s) for the redox couple Ti(IV)/Ti(III) in the KCl-KF(10 wt.%)-K₂TiF₆ melt.

Electrochemical studies were carried out in the temperature range of 1073-1173 K by cyclic voltammetry. Voltammetric curves were recorded at a glassy carbon electrode vs. a glassy carbon quasi-reference electrode. The sweep rate (v) varied from 0.1 up to 2.0 V s^-1. The electrochemical redox process Ti(IV)+e^-↔Ti(III) was classified as quasi-reversible at a sweep rate 1.0 V s^-1≤ν≤2.0 V s^-1.

The values of k_s were calculated by using the Nicholson's method [1]. It was shown that k_s increase with increasing the temperature. It is due to the number of particles capable of overcoming the potential barrier increases [2].

Addition of alkaline earth metal cations resulted in increasing of k_s up to the certain ratio of Me²⁺/Ti(IV) for all alkaline earth metal cations. It was determined the linear dependence of k_s on ionic potential of alkaline earth metal cations.

The activation energies of the charge transfer in the case of strongly polarizing cations addition were calculated. It was shown that values of the activation energy for molten systems with strongly polarizing cations are considerably less than activation energy of the initial melt and decrease with increasing of ionic potential.

It was determined that the k_s for the redox couple Ti(IV)/Ti(III) in the KCl-KF(10 wt.%)-K₂TiF₆ melt (for all systems) are less than k_s in the NaCl-KCl based melt [3].

References:

[1] R.S. Nicholson, J. Anal. Chem., 1965. Vol. 37 (1965), 1351.

[2] B.B. Damaskin, O.A. Petriy, Introduction of electrochemical kinetics, Vysshaya Shkola Publ., Moscow, 1975 (in Russian).

[3] D.A.Vetrova, S.A. Kuznetsov, Proceedings of the Kola Science Center RAS, 2015. P. 214.

Mixtures of ionic liquids (ILs) containing single-walled carbon nanotubes (SWCNTs) were prepared and characterized to obtain electrolytes with optimized transport properties for the use in energy storage devices such as supercapacitors. Imidazolium ILs bearing cations with side chains of different functionality coupled with bis(trifluoromethylsulfonyl)amide (NTf₂^-) or bis(fluorosulfonyl)amide (FSA^-) anions were used in mixtures containing up to 5 wt % SWCNTs. It was determined that the ionic conductivities of all ILs studied can be significantly increased with the addition of single-walled carbon nanotubes when the loading of the nanotubes is above 2 wt %. Moreover, higher conductivities were observed for the ILs containing the NTf₂^- anion and 5 wt % SWCNTs. These results can significantly contribute to the development of improved energy storage devices. The work at Brookhaven National Laboratory was supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences under contract DE-SC0012704. The work at Brooklyn College was supported by the National Science Foundation, Solid State and Materials Chemistry Program, Division of Materials Research, EAGER award #1841398.

Among Li⁺ transport properties of electrolytes, ionic conductivity and Li⁺ transference number (t_Li) are key factors contributing to enhancement of the battery performance. High t_Li enables rapid charging even if the ionic conductivity is lower than that of conventional organic electrolytes and would also suppress the lithium dendrite formation despite the relatively low mechanical properties of the electrolytes ^{[1], [2]}. Our group focuses on molten Li salt solvate electrolytes including glyme (G4)-based solvate ionic liquids (SILs) owing to their attractive thermal and electrochemical properties. Recently, we have reported that high t_Li (= 0.77) resulting from unique Li ion hoppling/exchange conduction mechanism in sulfolane (SL)-based electrolytes allows for the improved charge-discharge rate performance of Li-ion and Li-S batteries ^{[3], [4]}. Few studies, however, have clarified the relationship between t_Li, which is affected by the correlations of ionic motions and molecular structure for highly concentrated electrolyte systems like molten Li salt solvates. In this study, we attempted to understand the cross-correlations of ionic motions and clarify their effects on the transport properties in molten Li salt solvate electrolytes with different solvent and anion structures.

Table 1 Two transference numbers of the electrolytes.

Table 1 illustrates two different t_Li values of the electrolytes; one is estimated from self-diffusion coefficients measured by pulse field gradient (PFG)-NMR (t_{Li ^NMR}) and the other is estimated from the electrochemical method (t_{Li ^PP}) ^{[5], [6]}. In the case of [Li(G4)][TFSA], t_{Li ^PP} is extremely low compared with t_{Li ^NMR}, which implies the concentration polarisation in this electrolyte is more pronounced under the electric field. By contrast, for [Li(SL)₃][TFSA], t_{Li ^PP} is higher than t_{Li ^NMR}. This suggests that the interionic correlations can enhance t_{Li ^PP} in [Li(SL)₃][TFSA]. We further discuss the interionic cross-correlations and their relationship with Li ion transport properties of other electrolytes in detail.

References

[1] K. M. Diederichsen; E. J. McShane; B. D. McCloskey, ACS Energy Lett., 2017, 2, 2563-2575.

[2] P. Barai; K. Higa; V. Srinivasan, J. Electrochem. Soc., 2018, 165, 11, A2654-A2666.

[3] K. Dokko; D. Watanabe et al., J. Phys. Chem. B, 2018, 122, 10736-10745.

[4] A. Nakanishi; K. Ueno et al., J. Phys. Chem. C, 2019, 123, 14229-14238.

[5] P. G. Bruce; J. Evans; C. A. Vincent, Solid State Ionics, 1988, 28-30, 918-922.

[6] M. Watanabe; S. Nagano et al., Solid State Ionics, 1988, 28-30, 911-917.

Considering the well-known, global challenges the energy economy and climate policy face, batteries and fuel cells are widely discussed for potential applications in the future due to their many benefits. There are some difficulties connected with these technologies though. For PEM-fuel cells one of the main problems is linked to their membranes. These are based on Nafion^®, a sulfonated fluorocopolymer and therefore have a limited application temperature of 80 °C at ambient pressure due to dehydration and corresponding loss of conductivity at higher temperatures. [1] Due to this limit in operating temperature there is a significant need for alternative membrane materials, preferably with precise and defined size and geometry that can exhibit high ion mobilities and ionic conductivities above 100 °C.

In recent years research on ionic liquids (ILs) has experienced a revival. These salts with melting points below 100 °C are promising components in energy devices such as batteries, solar or fuel cells, owing to their high thermal and electrochemical stability, non-flammability and high ionic conductivities. However, to prevent leaking and realise proper function of these devices immobilizing the IL in a matrix is necessary. [2] The resulting ionogels (IGs) then combine the characteristics of the respective IL with the useful properties of the polymer, i.e. its mechanical stability. This immobilization can be realized in three different ways: doping of polymers with the IL, polymerization of vinyl monomers in the IL, and polymerization of polymerizable ILs. [3]

To obtain membrane materials with precisely controlled sizes, shapes and geometries along with the necessary performance stereolithography and 3D-printing of suitable materials are a promising method. [4]

The aim of this work therefore is the synthesis and characterization of ILs for ion- and especially proton-conduction. The ionic conductivities of these compounds range between 10^-2 – 10^-4 S/cm at elevated temperatures. Moreover, with wide electrochemical and thermal stability windows (e. g. ΔE up to 3 V, T_g around -90 °C and T_d over 200 °C (for some of them)) these ILs are promising for ion transport in fuel cell membranes above 100 °C and their properties are in addition studied under aspects of ion mobility. The immobilization of these ILs is furthermore realized via different methods, as mentioned above. The corresponding transparent and flexible IGs, in part containing high wt% of IL, display promising thermal and mechanical stability and reach ionic conductivities of up to 10^-3 S/cm at elevated temperatures. This study also demonstrates successful 3D-printing and structuring of IGs, which clearly enables the design of materials with different requirements by simply adapting the size and shape.

[1] A. Martinelli, A. Matic, P. Jacobsson, L. Börjesson, A. Fernicola, S. Panero, B. Scrosati, H. Ohno, J. Phys. Chem. B, 2007, 111, 12462.

[2] Y.-S. Ye, J. Rick, B.-J. Hwang, J. Mater. Chem. A, 2013, 1, 2719.

[3] J. Lu, F. Yan, J. Texter, Progress in Polymer Science, 2009, 34, 431.

[4] K. Zehbe, A. Lange, A. Taubert, Energy Fuels, 2019, 33, 12885.

In order to resolve energy issues, novel technologies for more effective utilization of carbon resources such as biomass and organic waste are needed. Direct carbon fuel cells (DCFCs) are fuel cells utilizing solid carbon as fuel. The total reaction can be expressed as follows.

C(s) + O₂(g) = CO₂(g) (1)

DCFCs as well as typical fuel cells are expected to achieve relatively high efficiency even on a small scale. Due to utilizing solid fuel, storage and transportation of fuel are easy. However, conventional DCFCs had two big problems such as the way to continuously supply solid fuel and low cell performance. We have proposed novel tubular molten-carbonate type direct carbon fuel cells (TMC-DCFCs) [1] to resolve those problems. This study investigated basic performance of TMC-DCFCs.

We have already developed tubular molten carbonate-type fuel cells (TMCFCs) with comparatively high robustness and good durability against impurities [2,3], and the manufacturing method of them were applied for TMC-DCFCs. Each cell component of TMC-DCFCs was almost same with typical MCFCs. Cathode was NiO-3%MgO, electrolyte matrix was LiAlO₂, anode was Ni-2%AlCr and they were formed by slurry coating methods. Electrolyte was 60%Li₂CO₃-40%Na₂CO₃ molten carbonate. Activated carbon powders (MD: 43 µm) or carbonized wood powders (MD: 31 µm) was used as solid carbon fuel. A tubular cell was inserted into the mixture of solid carbon fuel and molten carbonate same with electrolyte in the weight ratio of 80/20. In this paper, we refer to the mixing ratio between solid carbon and molten carbonate as "the carbon/carbonate ratio". Carbonized wood powders were sawmill residues from conifer carbonized at 623 K and included 50.3% of volatile matter, 46.3% of fixed carbon and 3.4% of ash. Therefore, the net fixed carbon/carbonate ratio was 65/35 when using carbonized wood powders.

Single cell tests of continuous power generation were conducted for 2 hours and about 0.72 V of cell voltages with 230 mA cm^-2 of current densities, which corresponded to over 160 mW cm^-2 of power densities, were achieved at 1073 K in both cases using activated carbon and carbonized wood (Fig. 1). TMC-DCFCs had high cell performance and could sufficiently utilize carbonized wood powders as fuel. These results suggested that the change of the carbon/carbonate ratio was permissible to some extent and a small amount of ash hardly affected the cell performance. On the other hand, it is presumed that the contact between the anode and solid carbon might be poor when the carbon/carbonate ratio is too large or the flooding might occur when the carbon/carbonate ratio is too small. Hence, effects of the mixing ratio and ash contents should be investigated in more detail.

Furthermore, the stack test using two cells connected in series was conducted at 973 K. The container filled with the fuel mixture was separated into two sections by the alumina board to be electrically insulated. Results of the stack test will be reported at our presentation.

References

[1] A. Ido, M. Kawase, J. Power Sources, 449, 227483 (2020).

[2] M. Kawase, J. Power Sources, 285, 260-265 (2015).

[3] M. Kawase, J. Power Sources, 371, 106-111 (2017).

Figure 1

Currently, the problem of global warming and exhaustion of fossil fuel has become a big problem on the global scale, therefore a highly efficient electric power generating device that does not emit greenhouse gases such as carbon dioxide is required. Especially, since high temperature fuel cell such as solid oxide fuel cell (SOFC) or molten carbonate fuel cell (MCFC) operate at high temperature, these have the advantages that efficiency is high and expensive noble metal catalysts are unnecessary for the electrode. However, the operation at high temperature leads to degeneration of the cell component, which makes it unsuitable for long lifetime. Thus, attempts to lower the operating temperature have been made.¹

Ceria-based oxide doped with trivalent Sm and Gd (SDC or GDC) was noticed, it achieves sufficient oxygen ion conductivity even in the middle temperature range of 773-973 K. However, with only ceria-based oxide, it indicates electron conductivity, which makes open circuit voltage lower. Adding carbonate makes composite indicated good electron insulator, ceria-based oxide/molten carbonate composite electrolyte is one of the candidate of middle temperature fuel cells. In the composite material, the molten carbonate is impregnated in the oxide and the interfacial layer is formed between solid oxide and carbonate due to solid-liquid interaction.

In this study, we applied nano-ordered SDC powder prepared by the complex polymerization method to composite of SDC/molten carbonate eutectics.² The ionic conductivity using the ac impedance measurement, the thermal properties and spectral properties of carbonate was measured in order to discuss the influence of the solid phase to the ionic conduction near the solid phase.

Ceria and SDC were prepared by the Pechini method using citric acid and ethylene glycol mixed with predetermined amounts of Ce(NO₃)₃ and Sm (NO₃)₃ aqueous solution. Obtained Sm³⁺ doped CeO₂ samples indicate Sm ratios of 10 and 20 mol% (SDC10 and SDC20). As the liquid (melt) phase, alkali carbonates were dehydrated in CO₂ at 473 K for 48 hours, and the prepared eutectics; (Li_0.52Na_0.48)₂CO₃ (LN) and (Li_0.43Na_0.32K_0.25)₂CO₃ (LNK) were used. Oxide and eutectics were mixed so as to obtain predetermined liquid phase volume fraction to obtain CeO₂-eutectics or SDCs- eutectics. The mixed sample was pressed at 60 MPa for 30 minutes and fired at eutectic temperature or higher to prepare pellets for AC impedance measurement in the temperature range of 623-773K in 25 Hz -1 MHz. The eutectic point of the bulk LN is seen at 771 K, and the eutectic point is shifted to lower temperature by mixing with oxide. When the liquid phase volume fraction is around 60-90 vol%, no remarkable change in the eutectic point of LN was confirmed. However, the significant decrease in the eutectic point was confirmed from the region of 60 vol% or less in any oxides. The endothermic peak associated with melting decreased as the amount of liquid phase decreased, and finally disappeared below 30 vol% or less. On the other hand, when compared with past results using CeO₂ (1.5 m² g^-1) having a small specific surface area as solid phase, the melting enthalpy was confirmed even when the liquid phase was 5 vol%.

The temperature dependences of electrical conductivity in ceria-based oxide/LN and LNK coexisting system are shown in Fig.1. For using LN eutectics, the transition point due to melting of the eutectic molten salt was observed at around 720 K at the liquid phase of 25 vol% or more. The conductivity at the liquid phase volume fraction of 45 vol% is smaller than the value of 15-35 vol% in the region below the transition point. As the liquid phase amount increases, the bulk molten salt crystallizes, from which, it is considered that the crystallized molten salt inhibited the electrical conduction pathway. In addition, at 15 vol%, the conductivity did not change remarkably even before and after the transition point, and there was no melting enthalpy change at 25 vol% or less, so from these results, the molten salt existing at the solid interface is in molten state even below the transition point. For LNK system, the effect of solid phase is more effective than that for LN systems. The activation energy of conductivity increased with decreasing apparent average thickness. This phenomenon was remarkable where an abrupt increase was observed below the apparent average thickness of 1 nm.

1. B. Singh et al., J. Power Sources, 339, 103(2017).

2. P. Ramos-Alvarez, et al., J. Mater. Sci., 52, 519 (2017).

Figure 1

Aluminum-anion rechargeable battery (AARB), in which chloroaluminate anions control the cathodic and anodic reactions, has been getting a lot of attention recently, because Al metal has several advantages, such as very high theoretical capacity (2980 mAh g^-1, 8046 mAh cm^-3), less expensive and mass consumable material, and moderate reactivity in air. It is known that Al metal deposition/stripping reversibly proceeds in Lewis acidic chloroaluminate melts.¹

4[Al₂Cl₇]^- + 3e^- ⇌ Al + 7[AlCl₄]^-

In general, AARB uses layered sp² carbon materials as cathode active materials that can cause the following electrochemical reaction.¹

nC + [Al anion]^- ⇌ C_n[Al anion] + e^- (Al anion: [AlCl₄]^- and [Al₂Cl₇]^-)

These materials show a favorable coulomb efficiency over 99 % and are available for rapid charge-discharge. But, the charge-discharge capacities are only 60 ~ 90 mAh g^-1, if the electrolyte is Lewis acidic organic chloroaluminate ionic liquids (ILs). Although, in the inorganic ones, e.g., AlCl₃–NaCl–KCl, the capacities are over ca. 130 mAh g^-1,² it is still disportionate to high capacity of Al metal anode. Therefore, novel high capacity cathode active materials are required to design appealing AARBs. In this study, we have applied the sulfur-carbon composite material (SPEG) synthesized from the mixture of sulfur and polyethylene glycol³ to the cathode active material for AARB. Electrochemical behavior of the SPEG electrodes was examined in an inorganic 61.0-26.0-13.0 mol% AlCl₃–NaCl–KCl IL (eutectic point: 366 K).

The procedure used for the preparation of SPEG was identical with that described in previous articles.² SPEG composite electrodes were prepared by pressing the mixtures of x wt% SPEG, 100-x-5 wt% conductive additive (ketjen black (KB) or multi-walled carbon nanotube (MWCNT), and 5 wt% polytetrafluoroethylene (PTFE) onto molybdenum (Mo) plate current collectors. Al metal plates were employed as anodes. After purification of AlCl₃, NaCl, and KCl, the inorganic AlCl₃–NaCl–KCl IL was obtained by heating the mixture consisting of 61.0 mol% AlCl₃, 26.0 mol% NaCl, and 13.0 mol% KCl at 403 K. The final product was clear and colorless. Electrochemical experiments were carried out using a two-electrode type sealed cell set up in an electric furnace with a temperature-control device. The experimental temperature was 393 K. All the procedures are conducted in an Ar gas-filled glove box with O₂ and H₂O < 1 ppm.

Figure 1 shows cyclic voltammograms recorded at Mo and SPEG composite electrodes in the two-electrode type sealed cell with a 61.0-26.0-13.0 mol% AlCl₃–NaCl–KCl electrolyte at 393 K. There is no change in current density in the voltammograms at the Mo electrode, implying that Mo electrode is usable for the current collector in AARB. As to the SPEG composite electrodes, a pair of reduction and oxidation waves appeared at ca. 1.05 V and 1.25 V, respectively. However, the shapes of voltammograms depended on the conductive additive species and their weight ratio. In the case of the KB conductive additive, redox waves with the influence of electric double layer capacitance were observed, when the weight ratio was over 45 wt%. It was difficult to see the waves below 45 wt%. Meanwhile, the use of MWCNT was very effective to improve the cathode performance, and clear redox waves appeared at the additive amount of 10 wt%. Given the fact that electrochemical behavior of the sulfur in an organic AlCl₃–[C₂mim]Cl (1-ethyl-3-methylimidazolium chloride) IL,⁴ the redox waves observed in Fig. 1 would originate in the following reaction.

8[Al₂Cl₇]^- + 6e^- + 3S ⇌ Al₂S₃ + 14[AlCl₄]^-

These results suggest that the SPEG composite electrodes can work as the cathode in AARB. Charge-discharge tests were conducted under various conditions. Favorable rate capability comparable to the layered sp² carbon materials was recognized.¹ Besides, the SPEG electrodes had very high cathode capacities. The SPEG cathode with KB (45 wt%) showed 323 mAh (g-SPEG)^-1 at 5000 mA (g-SPEG)^-1. If the MWCNT was employed as a conductive additive, the capacity was decreased to 266 mAh (g-SPEG)^-1 at 5000 mA (g-SPEG)^-1, but the cyclability was stable up to 600 cycles. From these findings, we concluded that SPEG composite electrodes prepared in this research are promising cathodes for AARB.

This research was supported by the MIRAI program (grant number JPMJMI17E9), JST.

References

1. T. Tsuda, G. R. Stafford, and C. L. Hussey, J. Electrochem. Soc., 164, H5007 (2017) and references therein.

2. C.-Y. Chen, T. Tsuda, S. Kuwabata, and C. L. Hussey, Chem. Commun., 54, 4164 (2018).

3. T. Kojima, H. Ando, N. Takeichi, and H. Senoh, ECS Trans., 75, 201 (2017).

4. T. Gao, X. Li, X. Wang, A. J. Pearse, K. J. Gaskell, et al., Angew. Chem. Int. Ed., 55, 9898 (2016).

Figure 1

Electrochemical energy storage (EES) technologies, such as rechargeable batteries, supercapacitors and supercapatteries, are featured by their capability of direct storage of electricity, fast charging-discharging, and modular flexibility and convenience for scaling [1,2]. However, they also suffer from several obvious shortcomings, for example, high manufacturing cost, low specific energy (or energy density), and short storage time in comparison with technologies based on combustion of fossil fuels. In fact, the specific energy values resulting from either thermochemical or electrochemical oxidation are very much comparable, considering the thermodynamics of the following common reactions that convert chemical energy in the fuels to work, i.e. Gibbs energy change. Therefore, the key to increase the specific energy of EES technologies depends largely on how to achieve the conversion.

The values given in brackets are calculated without considering the mass of oxygen gas.

In these oxidation reactions, the reactants are the fuels that can all be reproduced or regenerated from the oxide products via appropriate electrochemical reduction driven by renewable energy (light, heat and electricity) [3]. Therefore, these fuels can all be called regenerative fuels [4]. In this presentation, the author will analyze and discuss the feasibility of, and some preliminary findings from using molten salt electrochemistry to enable the regenerative fuels technology in the context of affordable, durable and portable high density energy storage.

References:

(1) Chen GZ, Supercapacitor and supercapattery as emerging electrochemical energy stores, Int. Mater. Rev., 62 (2017) 173–202.

(2) Xia L, Yu LP, Hu D, Chen GZ, Electrolytes for electrochemical energy storage, Mater. Chem. Front., 1 (2017) 584–618.

(3) Ijije HV, Lawrence RC, Siambun NJ, Jeong SM, Jewell DA, Hu D, Chen GZ, Electro-deposition and re-oxidation of carbon in carbonate containing molten salts, Faraday Discuss., 172 (2014) 105–116.

(4) Xia L, Chen GZ, High density electrochemical energy storage via regenerative fuels, Chin. J. Catal.40 (s1) (2019) 111-119.

Acknowledgement: The author thanks the financial supports from the EPSRC (EP/J000582/1 EP/F026412/1), Royal Society (Braine Mercer Feasibility Award, 2006), and Ningbo Municipal People's Governments (3315 Plan and 2014A35001-1).

Figure 1

Rechargeable potassium-ion batteries are promising candidates for low-cost largescale energy storage systems due to their cost-effectiveness, sustainability and comparable energy density. With an increasing number of publications related to electrolytes for potassium-ion batteries, ⁽¹⁾ room temperature ionic liquids (RTILs) have also been suggested as feasible electrolytes owing to their superior properties, such as non-flammability and wide electrochemical window.⁽²⁾In our research group, we reported bis(trifluoromethanesulfonyl) amide (TFSA)-based RTILs as suitable for the operation of potassium-ion battery when coupled to high voltage cathode materials.⁽³⁾ In this study, we evaluate the physicochemical properties of TFSA-based RTILs with KTFSA, in addition to assessing their application for high-voltage potassium-ion battery with honeycomb layered cathode oxides. ⁽⁴⁾

N-methyl-N-propylpyrrolidinium (Pyr₁₃) TFSA and 1-ethyl-3-methylimidazolium (EMI) TFSA were selected as RTILs in this study. A comparison of their physicochemical properties with same concentration of KTFSA, LiTFSA, and NaTFSA reveal RTILs containing KTFSA to show relatively high ionic conductivity and wide electrochemical window. When a symmetrical cell of potassium was constructed using Pyr₁₃TFSA with 0.5 M of KTFSA, deposition and dissolution reaction could be performed repeatedly without increase in overvoltage. Finally, we evaluated the charge/discharge performance of K₂Ni_2-xCo_xTeO₆ (where x=0.25, 0.5 and 0.75) high-voltage cathode materials using Pyr₁₃TFSA with 0.5 M of KTFSA as an electrolyte. Results show that each positive electrode has good cycle performance, and the average voltage changed depending on the Co content in the cathode. We conclude that TFSA-based RTILs are potential candidates for high-voltage potassium-ion battery.

References

1. T. Hosaka, K. Kubota, A. S. Hameed and S. Komaba, Chem. Rev. (2020).

2. T. Yamamoto, K. Matsumoto, R. Hagiwara and T. Nohira, J. Phys. Chem. C, 121, 18450 (2017).

3. T. Masese, K. Yoshii, Y. Yamaguchi, T. Okumura, Z.-D. Huang, M. Kato, K. Kubota, J. Furutani, Y. Orikasa, H. Senoh, H. Sakaebe and M. Shikano, Nat. Commun., 9, 3823 (2018).

4. K. Yoshii, T. Masese, M. Kato, K. Kubota, H. Senoh and M. Shikano, ChemElectroChem, 6, 3901 (2019).

Acknowledgement

This work was conducted in part under the auspices of the Japan Society for the Promotion of Science (JSPS KAKENHI Grant Numbers 19K15686).

Today, ionic liquids are more and more present in many fields and more particularly in electrochemistry. Indeed, their physical and chemical properties are appealing and attractive. They are conductive solvents in which organic and inorganic salts can be dissolved depending on the nature of the anion and cation that make up the ionic liquid. However, only very few studies have reported their use in membraneless redox flow batteries (RFBs) for the storage of renewable energy ^{1, 2}. The concept of membraneless redox-flow batteries was first reported by Ferrigno et al.³ in 2002, with the development of a millimeter-scale redox fuel cell based on the vanadium aqueous electrolyte solutions.

In this work, we have developed an ionic liquid membraneless RFB by using 1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide (C₂mimTFSI) as supporting electrolyte and Quinone (Q) and iron chloride (FeCl₂) as electroactive species in a microfluidic system. Polarization curve and cyclic voltammetry were used to characterize the electrochemical properties as well as the performance of the microbattery. The proof-of-concept of the system has been shown with an open circuit potential of 0.6 V, obtained with both polarization curve and cyclic voltammetry, and with a current density ranging from 0.3 to 0.65 mA cm^-2 for total flow rates of 10 to 20 µL min^-1. As shown on fig. 1(b), a maximum of power of 40 µW cm^-2 has been obtained. Such a technology is promising and performances can be enhanced by using 3D electrodes and optimizing the choice of the redox mediators (concentration, potential, etc.)

1. Navalpotro, P.; Palma, J.; Anderson, M.; Marcilla, R., A Membrane-Free Redox Flow Battery with Two Immiscible Redox Electrolytes. Angew. Chem. Int. Ed. Engl. 2017,56 (41), 12460-12465.

2. Chen, R.; Bresser, D.; Saraf, M.; Gerlach, P.; Balducci, A.; Kunz, S.; Schroder, D.; Passerini, S.; Chen, J., A Comparative Review of Electrolytes for Organic-Material-Based Energy-Storage Devices Employing Solid Electrodes and Redox Fluids. ChemSusChem 2020,13 (9), 2205-2219.

3. Ferrigno, R.; Stroock, A. D.; Clark, T. D.; Mayer, M.; Whitesides, G. M., Membraneless vanadium redox fuel cell using laminar flow. Journal of the American Chemical Society 2002,124 (44), 12930-12931.

Figure 1

Thermal batteries are important power sources in applications requiring a long shelf life, fast activation, and reliable power delivery. They operate at high temperatures and utilize molten salt electrolytes, typically eutectic mixtures, which are inert solids at ambient temperatures and provide fast ionic conductivity above the melting point of the salt. The size, energy density, and mechanical properties of thermal batteries are limited by the current manufacturing processes, which are slow and labor-intensive. The development of printable, coated-film electrodes and components will lead to batteries with increased energy density, a more efficient production process, and form factor flexibility.

The electrolyte salt (LiCl-KCl eutectic) and MgO separator materials are combined in a single 'electrolyte binder' mixture (EB). Slurries were prepared with the EB using DMSO as the carrier solvent and additional dissolved LiCl as a binder for the printed film. The slurries show a large change in viscosity as a portion of the electrolyte dissolves during the mixing process. The initial coarse slurry mixture quickly increases in apparent viscosity before smoothing out into highly shear-thinning suspension with a small overall particle size (<25 µm). The slurries can be easily printed directly on coated cathode substrates, using a slot-die or tape casting coating method. After drying, the film provides a stable separator layer, which can be cut or punched into the desired form and assembled into battery cells. Micro-CT imaging shows the resulting multi-layer film with a distinct cathode layer, and separator layers with a final dry film thickness on the order of 220 µm.

Thermal analysis data indicates that neither the modified salt content nor the dissolution and re-crystallization process significantly alter the melting behavior of the electrolyte salts and separator coatings when compared to the standard thermal battery separator mixtures. Thermogravimetric analysis (TGA) indicates that the solvent is sufficiently removed from the printed films during the drying process, and differential scanning calorimetry (DSC) measurements show that the melting point of the electrolyte in the coated separator film shifts by less than 5ºC, in comparison to the standard LiCl-KCl eutectic mixture. Single-cell electrochemical tests using the combined multi-layer separator/cathode films yielded voltages of >1.9 V delivering a capacity >100 mAh/g and with a lower internal resistance than standard thermal battery cells.

Figure 1. Cross-sectional micro-CT image of printed multi-layer electrode, consisting of carbon fiber support with cathode film and separator/electrolyte overcoat.

Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy's National Nuclear Security Administration under contract DE-NA0003525.This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government. SAND No. SAND2020-5349 A

Figure 1

Ionic liquids are being extensively explored as an alternative to carbonate-based electrolytes in reversible batteries, as they can exhibit lower flammability and volatility, and better overall safety. The polarity and acidity/basicity of ionic liquids can be tuned to control solubility, and can influence the electrochemical window and interfacial properties when used in batteries. In this study, 1-ethyl-3-methyl-imidazolium bis(trifluoromethanesulfonyl) imide (EMIM.TFSI) and 1-methyl-1-propypyrrolidinium bis(trifluoromethanesulfonyl)imide (MPPY.TFSI) were selected because they are well-known, relatively stable ionic liquids with large electrochemical windows which show strikingly different behavior with carbonaceous anodes due to differences in interface passivation. High surface area mesoporous hard carbon anodes were employed to provide a large signal from the interfacial chemistry of these electrolytes during reversible cycling. A combination of operando small-angle neutron scattering (SANS) and ex-situ electrochemical studies were used to understand the differences in solid electrolyte interphase (SEI) for these ionic liquids, and the relationship of SEI formation to hard carbon microstructural changes. Reversible hard carbon expansion is observed in the first cycle for the electrolyte lithium bis(trifluoro-methanesulfonyl)imide (LiTFSI)/EMIM.TFSI related to EMIM+ intercalation and deintercalation before a stable SEI is formed, while a largely irreversible framework expansion of 15% is observed for the LiTFSI/MPPY.TFSI electrolyte. There is only relatively minor expansion and contraction in subsequent cycles after a suitable solid electrolyte interphase (SEI) has formed. Irreversible framework expansion in conjunction with SEI formation is found to be essential for the stable cycling of hard carbon electrodes.[1]

Acknowledgement: Research was sponsored by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division. Research at Spallation Neutron Source user facility was sponsored by the Division of Scientific User Facilities, Office of Basic Energy Sciences, US Department of Energy, under contract DE-AC05-00OR22725 with UT-Battelle, LLC.

[1] Bridges, C.A.; Sun, X.G.; Guo, B.K.; Heller, W.T.; He, L.L.; Paranthaman, M.P.; Dai, S. "Observing Framework Expansion of Ordered Mesoporous Hard Carbon Anodes with Ionic Liquid Electrolytes via in Situ Small-Angle Neutron Scattering." ACS Energy Lett.2017, 2, (7), 1698-1704.

Graphite is considered by the US Department of Energy to be a critical element. Due to its widespread use as anodes in batteries, the demand for high purity graphite for battery applications is expected to increase for the foreseeable future. Synthetic graphite, due to its high purity, is generally the preferred choice for battery applications. The current process for the synthesis of graphite is very energy intensive and requires heat treatment of carbon precursors to temperature up to 3300 K for successful graphitization. Herein, an electrochemical method for graphitization is reported, which requires significantly lower temperatures. The cathodic electrochemical polarization of amorphous carbons in molten salts at around 1100 K can transform amorphous carbons to highly graphitic carbon structures. This process is facilitated by molten salts and has the great potential to save significant energy compared to the current existing process.

sp²-hybridized carbon materials, such as carbon nanotube and graphene nanoplatelet (GNP), have attracted a great deal of attention for their unique physicochemical properties. The carbon materials emerged as the promising candidates to replace carbon blacks as polymer membrane electrolyte fuel cell (PEMFC) catalyst supports, because those can prevent carbon oxidation and Pt nanoparticle catalyst agglomeration during start and stop cycling of the PEMFC.¹ However, the problem is how to anchor Pt nanoparticles over the smooth chemical inert surface of these carbon materials. It has a big risk leading to a decrease in the desired stability of GNP by the commonly used method of grafting functional groups onto the destroyed honeycomb structure.² Recently, based on the ionic liquid (IL)-based two-step method,³ we established a simple and mass-production available one-pot pyrolysis method with ionic liquid (IL one-pot process) for preparing Pt metal and PtNi alloy nanoparticle-supported multi-walled carbon nanotube composite electrocatalysts for oxygen reduction reaction (ORR).⁴ In the present study, this method was applied to two kinds of GNPs with different thicknesses, 20~30 layers (GNP-20) and 3 layers (GNP-3). The impacts of IL and GNP species on the catalytic performance to ORR were examined in detail.

Two types of ILs, N,N,N-trimethyl-N-propylammonium bis(trifluoromethanesulfonyl)amide ([N_1,1,1,3][Tf₂N]) and 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ([C₄mim][Tf₂N]) were used as reaction media. These ILs were purified by an appropriate pretreatment process prior to use. 2.5 g L^-1 of platinum(II) acetylacetonate was mixed with 1.25 g L^-1 GNP-20 in the ILs. The Pt nanoparticle supported GNP-20 composite materials (Pt/GNP-20s) were prepared by agitating the IL mixtures at 573 K for 4 hrs under N₂ atmosphere. Similarly, Pt/GNP-3s were prepared using GNP-3 instead of GNP-20. The resultant composites were washed with acetonitrile and water several times and dried in vacuum overnight. The final products were characterized by TEM, ICP, and XRD. Their electrocatalytic performances were evaluated by electrochemical measurements.^3,4

Figure 1 shows TEM images of Pt nanoparticle supported GNPs prepared by the IL one-pot process method. Compared to the specimens prepared in [N_1,1,1,3][Tf₂N], those in [C₄mim][Tf₂N] have smaller Pt nanoparticle size and show better Pt nanoparticle dispersion, if carbon supports are the same. These results seem to be related to the molecular structures of organic cations in reaction media. [C₄mim][Tf₂N] contains a larger cation with an aromatic ring, which can form a π-π stacking structure on the GNPs,⁵ would confine the nucleation points for Pt nanoparticle deposition at the anchor points. Difference in the GNP species affected Pt nanoparticle size on the Pt/GNP-20s and Pt/GNP-3s. For example, when [N_1,1,1,3] [Tf₂N] was used, mean particle sizes for the Pt nanoparticle on GNP-20 and GNP-3 were 3.8 nm and 3.3 nm, respectively. This result is attributed to a difference in specific surface area, i.e., Pt nanoparticle growth is suppressed on the GNP-3 having a larger surface area and more nucleation points. As given in Table I, the similar tendency was also recognized in [C₄mim][Tf₂N].

Electrocatalytic performances for the obtained Pt/GNP-20s and Pt/GNP-3s were evaluated by commonly-used electrochemical approaches in N₂ and O₂ saturated HClO₄ solutions along with a commercial catalyst (TEC10V30E). All the electrochemical data are summarized in Table I. The Pt/GNP-3 prepared in [C₄mim][Tf₂N] shows the highest mass activity (439.34 A g^-1) among all Pt nanoparticle supported GNPs because of the smallest mean particle size. Considering oxygen accessibility, Pt nanoparticle density per unit area of the basal plane on GNPs may also affect the electrocatalytic performances. Interestingly, after 15000 cycle durability test, all the Pt/GNPs showed higher mass activity retention rates than TEC10V30E. This should be due to the chemically stable surface structure at the basal plane on the GNP.

References

1. X.Zhou, J. Qiao, L. Yang, and J. Zhang, Adv. Energy Mater., 4, 1301523 (2014).

2. L.Xin, F. Yang, S. Rasouli, J. Xie, et al., ACS Catal., 6, 2642 (2016).

3. K.Yoshii, T. Tsuda, T. Torimoto, S. Kuwabata, et al., J. Mater. Chem. A, 4, 12152 (2016).

4. Y.Yao, R. Izumi, T. Tsuda, S. Kuwabata, et al., ACS Appl. Energy Mater., 2, 4865 (2019).

5. T. Fukushima, A. Kosaka, Y. Ishimura, T. Yamamoto, T. Aida, et al., Science, 300, 2072 (2003).

The search for synthesis methods which are simple, green, reliable, scalable and generalizable for functional metal oxide nanoparticles has been an area of interest for many researchers in recent years. Because of its meritorious features including environmental friendliness, low cost, simple to operate, easy to scale-up, etc., molten-salt synthesis (MSS) method becomes an excellent bottom-up synthesis technique of nanomaterials with various chemical compositions and morphologies. In this talk, I will start with the status, potential and challenges of MSS for the synthesis of nanomaterials to give a concise flavor on the importance of synthesis on the properties and application of nanomaterials. I will discuss different aspects of MSS such as the role of used molten salt, the choice of desirable molten salt, the effect of various synthesis parameters, typical oxosalts and their electrochemical aspects. More importantly, I will cover the recent progress of MSS for inorganic metal oxide nanoparticles. Other than the nanomaterials of a few binary oxides synthesized by the MSS, I will highlight how the MSS method has been successful in synthesizing complex metal oxide nanoparticles, such as AMO₂ delafossites, AMO₃ perovskites, AM₂O₄ spinels, and A₂M₂O₇ pyrochlores. Specifically, in the past few years, we have focused on the studies of lanthanide and actinide doped pyrochlore A^III₂M^IV₂O₇ nanoparticles (where A = trivalent rare earth ions, and M = Zr⁴⁺, Hf⁴⁺, etc.) useful for solid-state lighting, X-ray scintillators, thermometry, bioimaging, and radioactive waste containment. We have achieved substantial tunability of their particle size, crystal phase, and more importantly, luminescence properties. We have gained a clear understanding of the influences of synthesis conditions, particle morphology and composition on their photoluminescence and radioluminescence. Therefore, MSS opens a new avenue for making size and shape tunable nanomaterials for various catalytic, optoelectronic, magnetic and electrical applications.

Introduction

Incident lights interact with noble metal nanoparticles (NPs) to form collective oscillations of free electrons, creating a strong electromagnetic field on their surface. This phenomenon is called localized surface plasmon resonance (LSPR). Recently metal oxide NPs with appreciable concentration of free electrons have been also reported to exhibit LSPR peaks, the peak wavelengths of which were controlled from the visible to near-IR regions.[1] When a plasmonic NP was combined with a semiconductor, the excitation of LSPR could cause the injections of electrons and/or holes from plasmonic NPs to the semiconductor, that is, the plasmon-induced charge separation (PICS).[2] Since the PICS was reported using plasmonic metal oxide NPs,[3] the intense research has been devoted to studying plasmonic metal and metal oxide NPs for the application to photovoltaics, photocatalyts, and biosensors.

Recently we reported the preparation of noble metal NPs by metal sputtering deposition onto room-temperature ionic liquids (RTIL) under a reduced pressure (RTIL/metal sputtering), where NPs of several nanometers in size were stably and uniformly dispersed without additional stabilizing agents.[4] This technique enabled the clean preparation of plasmonic metal NPs, such as Au[4], Ag[5], and AgAu alloy[6]. In this study, we apply the RTIL/metal sputtering to the preparation of molybdenum oxide (MoO_x) NPs. Thus-obtained NPs exhibited the LSPR peak, the excitation of which induced the PICS.

Experimental

The sputter deposition of molybdenum were carried out on the surface of 1-hydroxyethyl-3-methylimidazolium tetrafluoroborate (HyEMI-BF₄) and 1-ethyl-3-methylimidazolium tetrafluoroborate (EMI-BF₄) used as RTILs for 1 h with a discharge current of 30 mA under an argon pressure of 3.0 Pa. The deposited NPs were partially oxidized by the heat treatment at 473 K for 30 min in air, resulting in the formation of MoO_x NPs. The photoelectrochemical properties were measured in a 0.5 M Na₂SO₄ aqueous solution with the ITO electrode immobilized with MoO_x NPs as working electrodes, a Pt counter electrode and a Ag/AgCl reference electrode. The photocurrents were detected with monochromatic light irradiation under the potential application at +0.5 V vs Ag/AgCl.

Results and Discussion

TEM measurements revealed that Mo NPs sputter-deposited in HyEMI-BF₄ had an average diameter of 6.8 nm and then the heat treatment at 473 K increased the particle size to ca. 65 nm. XPS spectra of thus-obtained particles indicated that the thus-obtained NPs were composed of MoO_x containing Mo(V) and Mo(VI) species. The MoO_x NPs prepared in HyEMI-BF₄ showed a LSPR peak at around 840 nm, the peak wavelength of which was close to that of LSPR reported for chemically synthesized MoO_3-x NPs.[1] On the other hand, MoO_x NPs prepared in EMI-BF₄ showed no LSPR peak in the visible and near-IR regions. Anodic photocurrents were observed by the irradiation to MoO_x NP-immobilized ITO electrodes with wavelength shorter than ca. 1000 nm. Regardless of the kinds of MoO_x NPs, the action spectra of photocurrents showed the increase of IPCE at the wavelength below 600 nm due to the photoexcitation of interband transition of MoO₃, being in agreement with their extinction spectra. Furthermore, the MoO_x NPs prepared in HyEMI-BF₄ showed a peak at 840 nm, the peak wavelength of which agreed with that of their LSPR peak. In contrast, MoO_x NPs formed in EMI-BF₄ exhibited no photocurrent by the irradiation of near-IR lights.

Conclusively, we successfully prepared the MoO_x NPs with RTIL/metal sputtering technique followed by the heating treatment. The NPs formed in HyEMI-BF₄ showed the LSPR peak in near-IR region, the photoexcitation of which induced the PICS from the MoO_x NPs to the ITO electrodes.

References

(1) A. Agrawal, S. H. Cho, O. Zandi, S. Ghosh, R. W. Johns, and D. J. Milliron, Chem. Rev., 118, 3121 (2018).

(2) Y. Tian and T. Tatsuma, J. Am. Chem. Soc., 127, 7632 (2005).

(3) S. H. Lee, H. Nishi, and T. Tatsuma, Nanoscale, 10, 2841 (2018).

(4) T. Torimoto, K. Okazaki, T. Kiyama, K. Hirahara, N. Tanaka, and S. Kuwabata, Appl. Phys. Lett., 89, 243117 (2006).

(5) T. Suzuki, K. Okazaki, T. Kiyama, S. Kuwabata, and T. Torimoto, Electrochemistry, 77, 636 (2009).

(6) K. Okazaki, T. Kiyama, K. Hirahara, N. Tanaka, S. Kuwabata, and T. Torimoto, Chem. Commun., 6, 691 (2008).

Ionic liquids (ILs) are now being recognized as the third group of solvents, following water and organic solvents. They are easily available and possess unique properties such as nonvolatility, high thermal stability, and designability, which make it possible to use them on demand and under harsh conditions. Our study had been focused on understanding of the unique properties of ILs and on their utilization as neoteric solvents for innovative polymeric materials and devices^1,2 that can help realize a sustainable society. ILs exhibit unique solubility toward polymers;¹ this opens up a new field of intelligent materials chemistry. By utilizing the unique solubility, we have proposed soft materials containing ionic liquids, which we named "ion gels".^1,2 Ion gels are a novel platform for many applications such as electrolyte membranes for batteries³ and fuel cells,⁴ actuators,⁵ gas-separation membranes,⁶ and electric double layer transistors.⁷

Especially, due to the recent surge in flexible and wearable devices, highly durable ion gels have attracted much attention. In this lecture, I will address the recent advances in the development of ion gels that have healing functions against mechanical damages. As stimuli-responsive healing strategy, light- and thermally-induced healing of ion gels are discussed mainly based on block copolymer self-assenbly changes in ILs.^8-10 Then, self-healable ion gels based on supramolecular¹¹ and dynamic bond¹² chemistry are addressed. By judicious designing of polymer nanostructures in ILs and interactions between polymer chains and IL cations and anions, tough, highly stretchable, and self-healable ion gels are recently demonstrated.

References

1. T. Ueki, T. M. Watanabe, Macromolecules, 41, 3739 (2008).

2. Y. Kitazawa, K. Ueno, M. Watanabe, Chem. Record, 18, 391 (2018).

3. Y. Kitazawa, K. Iwata, R. Kido, S. Imaizumi, S. Tsuzuki, W. Shinoda, K. Ueno, T. Mandai, H. Kokubo, K. Dokko, M. Watanabe, Chem. Mater.,30, 252 (2018).

4. S.-Y. Lee, A. Ogawa, M. Kanno, H. Nakamoto, T. Yasuda, M. Watanabe, J. Am. Chem. Soc. 132, 9764 (2010).

5. S. Imaizumi, H. Kokubo, M. Watanabe, Macromolecules, 45, 401 (2012).

6. A. Ito, T. Yasuda, T. Yoshioka, A. Yoshida, X. Li, K. Hashimoto, K. Nagai, M. Shibayama, M. Watanabe, Macromolecules, 51, 7112 (2018).

7. M. Matsumoto, S. Shimizu, R. Sotoike, M. Watanabe, Y. Iwasa, T. Aida, J. Am. Chem. Soc., 139, 16072 (2017).

8. R. Tamate, K. Hashimoto, T. Ueki, M. Watanabe, Phys. Chem. Chem. Phys., 20, 25123 (2018).

9. T. Ueki, R. Usui, Y. Kitazawa, T. Lodge, M. Watanabe, Macromolecules, 48, 5928 (2015).

10. C. Wang, K. Hashimoto, R. Tamate, H. Kokubo, M. Watanabe, Angew. Chem. Int. Ed., 57, 227 (2018).

11. A. Saruwatari, R. Tamate, H. Kokubo, M. Watanabe, Chem. Commun., 54, 13371 (2018).

12. R. Tamate, K. Hashimoto, T. Horii, X. Li, M. Shibayama, M. Watanabe, Adv. Mater., 30, 1802792 (2018).

The application of additive manufacturing in various areas of chemistry has increased dramatically in recent years including in the field of ionic liquids (ILs). It has been shown that ILs can be incorporated into the resins used in stereolithography (SLA) 3-D printing either as part of the polymer backbone or as an additive. Results have shown a change in the properties of the resulting printed polymer based on the structure of the ILs. The incorporation of the ILs into 3-D printed polymers provides the ability to customize the chemical and physical properties of a printed part. In particular, ionic liquids have been shown to act as plasticizers when incorporated into 3-D printed PMMA.

In this presentation, we will explore the effects that selected ILs have on the polymerization of the various resins used in SLA printing including the IL's effects on the resin's heat of polymerization. These experiments have incorporated a new technique to measure the temperature change during the polymerization process using a "TinyLev" acoustic levitator. This device provided a contact-free, container-less sample holder for a 10 µL droplet of the resin/IL mixture suspended on a standing acoustic wave. A FLIR camera was used to record the temperature change as the polymerization reaction progressed. Since the polymerization of SLA resins is typically initiated by UV light, these experiments used a 405 nm laser to initiate the polymerization reaction. A variety of ILs with different cation and anion functionalities were examined with the collected data compared to pure resin. Thermal decomposition data for these samples collected using DSC and TGA was corroborated with GPC data in order to determine the effect of the ILs on the molecular weight. This data, combined with the FLIR thermography results, provides an understanding of the role ILs have in the polymerization reaction of the resin—either as additives or as a co-polymer.

Figure 1

Biomaterials such as cotton and silk are abundant natural resources that have demonstrated material properties equal or superior to many synthetic polymers. Using ionic liquids (ILs) and the Natural Fiber Welding (NFW) process, these biomaterials can be chemically and physically enhanced. Here we present our progress on using imidazolium-based polymerizable ionic liquids (Poly-ILs) in the NFW process to prepare polyionic biocomposites. A variety of Poly-ILs with different cation and anion structures were synthesized and subsequently characterized using ¹H and ¹³C NMR. Each Poly-IL was evaluated for its ability to (i) polymerize and (ii) solubilize a biopolymer (cellulose) matrix, revealing anion structural motifs that either limit or enhance a Poly-IL's ability to polymerize and weld. Optimized parameters yielded polyionic biocomposite materials whose physicochemical properties were evaluated using ATR, SEM and EIS.

Figure 1

Ionic liquids (ILs), which are salts with low melting points, show unique properties including non-volatility and high ionic conductivity. By combining an ionic liquid (IL) with an amphiphilic ABA-type triblock copolymer, polystyrene (PSt)-b-poly(methyl methacrylate) (PMMA)-b-PSt (SMS), insoluble A blocks aggregate and form cross-linking points, whereas soluble B blocks form a gel network swollen with the IL.^[1,2] As a result, a self-standing and highly ion-conductive ion gels can be obtained, and they have been applied to electrochemical devices such as polymer actuators. In these devices, the mechanical toughness and ionic conductivity are important parameters. Thus, we have investigated the correlation between microstructures formed by the block copolymer and the properties of ion gels such as ionic conductivity and mechanical property. In particular, we focused on a gyroid structure in which both insoluble and soluble segments form continuous phases, because this unique microstructure is expected to exhibit high mechanical strength and favorable ionic conductivity, simultaneously. In this study, we directly observed microstructures using AFM to investigate how microstructures affect properties of the ion gel. Herein, we used poly(ethylene oxide) (PEO), which has a lower glass transition temperature (T_g) than PMMA, as a middle block (PSt-PEO-PSt, SOS). Consequently, SOS showed various microstructures including the gyroid depending on block ratio and content of IL. It was found that ionic conductivity and mechanical property strongly depended on the connectivity of microstructures. Namely, the gyroid structure showed the enhanced mechanical property and good ionic conductivity.

In addition, we observed in-situ change in the microstructure for a stimuli responsive ion gel containing temperature-responsive poly(N-isopropylacrylamide) (PNIPAm) blocks: PNIPAm is known to show lower critical solution temperature (LCST)-type phase behavior in water while showing upper critical solution temperature (UCST)-type phase behavior in ILs.^[3] PNIPAm-b-PEO-b-PNIPAm (NON) exhibits gel-to-sol transition in IL with increasing temperature due to the UCST-type phase transition of PNIPAm blocks. We have also reported photo-responsive gel-to-sol transition of ion gels by introducing a photo-responsive azobenzene moiety to either IL or polymer.^[2] These transition phenomena have been studied in terms of a macroscopic property such as rheological property. However, little is known about their microstructure during the photo-responsive change. Thus, we observed the photo-responsive change in the microstructure using AFM and investigated how the microstructure change is synchronized with the rheological response. We found that UV irradiation induced the change in the microstructures from a microphase-separated gel to a sol with a vague structure, and it was correlated with the photo-responsive change in the mechanical property and ionic conductivity.

Acknowledgements

This work was financially supported by the Grant-in-Aid for Scientific Research for Basic Research S (15H05758) from Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.

References

[1] Imaizumi S, Kokubo H, Watanabe M. Macromolecules, 2012, 45, 401–409.

[2] Tamate R, Hashimoto K, Ueki T, Watanabe M. Phys. Chem. Chem. Phys., 2018,20, 25123–25139.

[3] Ueki T, Watanabe M. Chem. Lett., 2006, 35, 964–965.

Introduction

Eutectic gallium-indium (Ga-In) has excellent properties, such as low melting point of 15.3 °C, high thermal and electronic conductivity, and metallic luster. Ga-In has promise as a liquid electron-conducting material owing to its low viscosity, negligible vapor pressure and low toxicity.^{1, 2)} On the other hand, ionic liquids are ambient temperature molten salts that have attracted considerable attention because of unique properties such as high ionic conductivity, non-volatility and thermal stability. We proposed that ion gels, composed of macromolecular networks swollen with ionic liquids, exhibit self-standing film-forming ability in addition to the unique properties of ionic liquids. In this study, we prepared composite gel materials containing ionic liquid and Ga-In. This composite gel (metal gel) might have high electronic conductivity based on Ga-In and high ionic conductivity originated from the ionic liquid, as well as good mechanical properties based on the polymer, such as flexibility and strechability. These new materials are applicable to flexible or stretchable devices in wearable and flexible electronics applications.

Experimental

We chose hydrogen bonding copolymers of N,N-dimethylacrylamide (DMAAm) and acrylic acid (AAc) (P(DMAAm-r-AAc)) as the matrix polymers. This copolymer was combined with a hydrophobic ionic liquid, 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)amide ([C₂mim][NTf₂]) to form an ion gel.³⁾ In order to improve dispersibility of Ga-In in the composite gel, bulk Ga-In was ultra-sonicated in ethanol and the suspension of Ga-In microdroplets was mixed with P(DMAAm-r-AAc) and [C₂mim][NTf₂]. The composite gels were prepared by solution casting method either in the air where thin oxide layer is formed on the Ga-In particles⁴⁾ or under inert atmosphere to examine the effects of preparation conditions on their properties.

Results and Discussion

In tensile tests, Young's modulus increased with increasing volume fraction of Ga-In in the composite gels. In rheological measurements, storage modulus was higher than loss modulus, confirming soft solid-like behavior of the composite gels. In both measurements, modulus of composite gels was higher than that of ion gels. We found difference in the temperature dependent rheological properties between the composite gels prepared in air and under inert atmosphere. The presence/absence of the surface oxide layer on the Ga-In particles was likely responsible for the difference in the rheological responses. Electronic conductivity was improved by a factor of 10⁶ for the composite gels prepared under inert atmosphere compared to that of the composite gels prepared in the air. It was found that the oxide layers on the Ga-In particles had a significant impact on the rheological and electronic properties. However, electronic conductivity of the composite gels prepared under inert atmosphere was still low compared to that of bulk Ga-In. To achieve high electronic conductivity comparable to the bulk value, volume fraction of Ga-In microdroplets needs to be increased in the composite gels. In order to improve dispersibility of high-loading Ga-In in the composite gel, Ga-In microdroplets were prepared with dispersants. The results suggested that there is a trade-off between dispersibility of the Ga-In microdroplets and the electronic conductivity: better dispersibility of Ga-In microdroplets resulted in lower electronic conductivity.

Acknowledgement

This study was supported in part by Core Research for Evolutionary Science and Technology (CREST) of the Japan Science and Technology Agency (JST).

References

1) Kazem, N. et al, Adv. Mater.,2017, 29, 1-14.

2) Anderson, T. J. et al, Phase Equilibria, 1991, 12, 64-72.

3) Tamate, R. et al, Adv. Mater, 2018, 30, 1802792

4) Ren, L. et al, Adv. Funct. Mater.,2016, 26, 8111-8118.

Figure 1

Olefin/paraffin separation by cryogenic distillation is one of the most energy-consuming processes not only among the entire chemical processes but also among all the energy-consuming activities by human beings. The purification of propylene and ethylene accounts for 0.3% of global energy use (1). Membrane separation is an attractive alternative, owing to its high energy-efficiency compared to conventional distillation processes. Liquid membranes that utilize gas transport through the liquid phase confined in the pore structures have shown promising performance due to the faster transport compared to solid membrane materials (2).

Since the liquid phase is where the transport of olefin/paraffin molecules occurs, the liquid membrane performance in terms of permeability, selectivity, and stability mainly relies on the properties of the liquid component. Ionic liquids are a suitable material since the liquid component in the membranes provide stable separation performance owing to their extremely low vapor pressure and good thermal/chemical stability. Moreover, the limitless turnability of ionic liquids can be manipulated to enhance the molecular interaction with certain gas molecules to result in better gas solubility. For olefin/paraffin separation, in addition to physical solubility of the olefin in the ionic liquid, silver ions are commonly added to further improve olefin solubility. The silver ions can complex with olefin molecules via interaction between the double bonds of olefins and the orbitals of silver ions.

Here we have investigated the effect of the anion, concentrations of the silver salt, and temperatures on the physicochemical properties of the silver-containing ionic liquid mixtures. The macroscopic physicochemical properties, such as viscosity and density, affect the filling of the membrane pores and stability to transmembrane pressure. However, we show that they are also indicators of ion aggregation in the mixtures, which can be correlated with relevant membrane properties, including solubility, diffusivity and silver ion reactivity with contaminant gases. The physiochemical viewpoint on the silver-containing ionic liquid mixture will provide appropriate insight for understanding the olefin/paraffin gas transport in the ionic liquid mixture, which leads to a design idea for better separation performance.

(1) D. S. Sholl, R. P. Lively, Seven chemical separations: to change the world: purifying mixtures without using heat would lower global energy use, emissions and pollution--and open up new routes to resources. Nature532, 435 (2016).

(2) C. M. Sanchez, T. Song, J. F. Brennecke, B. D. Freeman, Hydrogen Stable Supported Ionic Liquid Membranes with Silver Carriers: Propylene and Propane Permeability and Solubility. Industrial & Engineering Chemistry Research59, 5362-5370 (2020).

Molten Salt Reactors (MSRs) are the leading candidates out of the six-generation IV advanced nuclear power reactor designs chosen for deployment for nuclear energy in US. MSR technology is particularly attractive due to better passive safety, operation at atmospheric pressure, high thermal efficiency, lower spent fuel per unit energy and increased solubility of fission products in molten salts.

To facilitate robust and economical design of such systems, accurate knowledge and fundamental understanding of the structure and speciation of salts and metals in and near molten salt environments is necessary. Speciation of solutes is important because it determines how chemical behavior in solution and complex speciation from fission and corrosion products impact kinetic, thermodynamic, and transport properties of the molten salt systems. Investigating effects of radiation on metallic species and radiation-induced nucleation and growth of metallic nanoparticles under ionizing radiation is crucial for predicting the stability and reactivity of molten salts.

Molten Salts in Extreme Environments (MSEE) EFRC aims to investigate speciation of metals and radiation-induced reactions in molten salt systems by utilizing synchrotron based XAS methods. In our work, Extended X-ray Absorption Fine Structure (EXAFS) and X-Ray Absorption Near Edge structure (XANES) are used to investigate local coordination environment and chemical structure of metal species such as Nickel in Zinc Chloride based molten salt systems. In addition, the effect of metal concentration and temperature on changes in local and chemical structure of metal is studied. XAS studies are complemented by optical ultra-violet spectroscopy studies of molten salts, enabling a direct correspondence between UV-Vis peak shape and coordination geometry and number, determined by EXAFS. Such knowledge of speciation of metals and radiation-induced nanoparticles in molten salt environments will provide critical understanding needed to predict and control the physical and chemical properties of molten salts and corrosion mechanisms in molten salt systems. This work was supported as part of the Molten Salts in Extreme Environments, Energy Frontier Research Center, funded by the U.S. Department of Energy Office of Science.

1. Introduction

From the standpoint of energy conservation, the development of a recovery process for rare earth metals with reduced energy consumption is desired. In previous investigations, we demonstrated the recovery of Nd metal using low-temperature molten salts (LTMSs) [1,2], because an LTMS has many useful physicochemical properties [3] such as a wide electrochemical window, low liquid-phase temperature, and high ionic conductivity. In our findings, the reduction process of Dy(III) proceeded in two steps by way of Dy(II), [Dy(III)+e^-→Dy(II), Dy(II)+2e^-→Dy(0)] in [NTf₂^-] based ionic liquids (ILs), because [Dy^(III)(NTf₂)₅]^2- and [Dy^(II)(NTf₂)₄]^2- in ILs were confirmed by Raman spectroscopy and DFT calculations [4]. However, there were no information about the detailed solvation conformations of Dy(II) and Dy(III) in K[NTf₂] melts. Therefore, we have investigated the electrochemical behavior of Dy(II) and Dy(III) in K[NTf₂] melts.

2. Experimental

The K[NTf₂] melts including Dy(III) (molar fraction: x_Dy=0.1) were applied as the electrolytic for the electrochemical measurement and this solution was dried at 373 K in a vacuum chamber (<0.1 MPa) for 24 h. Cyclic voltammetry (CV) at 493±1.0 K was carried out using a cylindrical cell constituted from three-electrode system under Ar flow. The Pt electrode was used as a working electrode that was mirror polished using alumina paste. Platinum wires (φ=0.7 mm) were employed as a counter electrode and a quasi-reference electrode (Q.R.E), because the potential obtained using a Pt Q.R.E was stable and exhibited a good reproducibility at medium temperature. All the potentials were compensated for the redox reaction of K/K⁺ couple. As for the electrodeposition process, a prismatic Dy rod and a Cu substrate were employed as an anode and a cathode, respectively. The anode was surrounded by a soda lime tube with a Vycor glass filter at the bottom to prevent the diffusion of dissolution components from the anode into electrolyte. The potentiostatic electrolysis was carried out while stirring the electrolyte at 500 rpm in order to increase the current density of the electrodeposition process. An in-depth analysis of the electrodeposits was conducted with Al-Kα radiation by XPS. The sputtering rate was 27.2 nm min^-1 estimated from the sputtering rate of the Si standard.

3. Results and discussion

From the voltammetric analysis of Dy(III) in K[NTf₂] melts, there were no apparent anodic peaks corresponding to an oxidation of Dy(0) in voltammogram. This result indicated that the reduction of Dy(III) was an irreversible process. It was also consistent with the reference [5], which was reported that no oxidation peak of Dy(0) in DMF and DMPT. There were two reduction peaks around +1.0 V and +2.3 V vs. K/K⁺ in voltammogram. The reduction peak around +2.3 V would be based on Dy(III)/Dy(II) charge transfer reaction. The reduction peak around +1.0 V would be corresponded to Dy(II)/Dy(0) deposition process. Regarding the charge transfer reaction, it was confirmed that the plot of the cathodic peak of the current density vs. the square root of the scan rate showed a good linear relation. It indicated that the charge transfer reaction was controlled by the diffusion process.

In the electrodeposition process, the blackish-brown electrodeposits had a relatively strong adhesion on the Cu substrate. The Dy3d_5/2 peak of the electrodeposits was shown in Fig. 1. The Dy3d_5/2 peaks for Dy metal and Dy₂O₃ were theoretically positioned at 1295.8 eV and 1289.0 eV, respectively [6]. The Dy3d_5/2 spectra of the top surface and the middle layer showed a relatively good agreement with the theoretical value. This study enabled us to conclude that Dy metal was able to electrodeposit in a metallic state from the K[NTf₂] melts.

Acknowledgement

This research was partially supported by the Grant-in-Aid for Scientific Research (No. 18H03404) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

References

[1] H. Ota, M. Matsumiya, T. Yamada, T. Fujita, S. Kawakami, Sep. Purif. Technol., 170 (2016) 417-426.

[2] M. Matsumiya, H. Ota, K. Kuribara, K. Tsunashima, J. Electrochem. Soc., 164(8) (2017) H5230-H5235.

[3] K. Kubota, T. Tamaki, T. Nohira, T. Goto, R. Hagiwara, Electrochim. Acta, 55 (2010) 1113-1119.

[4] M. Matsumiya, R. Kazama, K. Tsunashima, J. Mol. Liq., 215 (2016) 308-315.

[5] J. Londermeyer, M. Multerer, M. Zistler, S. Jordan, H.J. Gores, W. Kipferl, E. Diaconu, M. Sperl, G. Bayreuther, J. Electrochem. Soc., 153(4) (2006) C242-C248.

[6] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray photo-electron spectroscopy, Perkin-Elmer Corp., (1992).

Figure 1

Molten halides can be effectively employed for electrorefining of refractory and radioactive metals with relatively negative reduction potentials and high melting temperatures. In the present study electrorefining of uranium was studied for developing pyrochemical reprocessing technology of metallized spent nuclear fuels. 3LiCl-2KCl molten salt was chosen as the working electrolyte. Uranium containing electrolytes were prepared by dissolving UCl₃ in the solvent salt melt in an argon filled glove box to obtain various desired concentrations of uranium in the electrolyte. The working temperature in the refining experiments varied between 450 and 750 °C.

The experiments on uranium electrorefining were carried out in a stainless steel semi-industrial water-cooled electrolyser. Crude uranium metal was loaded in a specially designed molybdenum anodic basket. Glassy-carbon crucible was used to hold the melt. Molybdenum rod was employed as a cathode.

The main goal of the electrolytic refining process is obtaining high-purity uranium metal. Specific energy consumption is also very important. The results of the electrolytic refining depend on various parameters. In the present study current density, temperature, uranium concentration in the electrolyte, and specific quantity of electricity passed through the cell were selected as variables, and the current efficiently was chosen as a response factor.

Dense and compact uranium deposits were obtained under low current densities, while application of relatively high current densities led to the formation of uranium dendrites. Intermediate values of current density allowed producing more coherent deposits that could be easily scraped off the cathode. Increasing temperature led to a sharp change of the cathodic deposits morphology. The reason for this is the formation of tetragonal beta-uranium instead of the low-temperature orthorhombic alpha modification.

The values of cathodic current efficiency in the most of experiments confirmed the three-electron scheme of uranium reduction. Cathodic current density under certain conditions exceeded 95 %. The amount of salt retained in the cathodic deposit varied from 7 to 25 %, values typical for dendrite electrolytic metals.

Molten fluoride salts can be used as the fuel and coolant for molten salt reactors (MSR) and electrolytes for spent nuclear fuel (SNF) reprocessing. 46.5%LiF-11.5%NaF-42%KF (FLiNaK) melt is the prospective media for these purposes due to their desirable thermophysical and nuclear properties. Finding construction materials with sufficient corrosion and mechanical resistance is the most challenging task for practical implementation of MSR concept.

In the present study the corrosion and mechanical properties of different types of construction materials were investigated. The materials included various low carbon Ni-Cr-Fe-Mo, Ni-Cr-Mo, and Ni-Mo alloys, and metals with relatively positive electrode potentials.

The corrosion experiments were performed in FLiNaK melt at different temperatures (from 550 to 750 °C) in the specially designed stainless steel cells under high-purity argon atmosphere. Corrosion properties of studied materials were investigated under static conditions, and the duration of each test was 100 h to enable the comparison of the experimental data. In a special series of experiments fluorides of typical fissile nuclides and fission products were added to the salt electrolyte to estimate the influence of the red-ox potential on the corrosion resistance of the materials.

Corrosion rates were determined from the weight loss measurements and chemical analysis of quenched melts. Surface and microstructure of corroded samples was examined by various microscopic techniques. Mechanical properties of investigated materials were also studied at the ambient temperature and at 600 °C.

Advantages and limitations of different types of construction materials were evaluated on the basis of data obtained. The effect of temperature on corrosion and mechanical properties of the studied materials were determined. Possible mechanisms of corrosion of various materials in fluoride melts were proposed.

The alloys based on Ni-Cr-Mo and Ni-Mo systems, and molybdenum and its alloys were selected for further long-time tests under dynamic conditions to determine the resource of materials in contact with molten fluorides for MSR and SNF recycling technologies.

A liquid metal electrode has interesting property such as phase change from a liquid to a solid depend on alloys composition. There are many reports of solid metal deposition on a liquid electrode, there are few papers of liquid metal deposition on a liquid electrode. In this study, Ga was used as the liquid electrode and Na electrodeposition was carried out in ionic liquid electrolyte at 160 ^oC. Since the melting point of Ga and Na are 29.8 and the 97.8 ^oC, at the moment when Na is electrodeposited, the liquid Na presents on the liquid Ga. In this experiment, we focused on the surface change of the Ga electrode before and after the electrodeposition of liquid Na.

The experiment was conducted in a glove box under Ar atmosphere. NaTf₂N (Sodium bis (trifluoromethanesulfonyl) imide) -TEATf₂N (Tetraethylammonium bis (trifluoromethanesulfonyl) imide) mixed ionic liquid with molar ratio of 1 : 4 was used as an electrolyte at 160 ^oC. Liquid Ga or solid Ag plate was used as a working electrode and liquid Na was used as counter electrode and reference electrode. The voltammogram measurement was performed in the range from 2.5 V to -0.5 V (vs. Na/Na⁺) with sweep rate of 10 mV s^-1. Constant current electrolysis was carried out at 5 mAcm^-2 for 3 minutes. The liquid Ga electrode surface was taken by a digital camera. The time change of the open circuit potential of the Ga electrode after constant current electrolysis was measured by potentiostat.

In voltammogram measurement, cathodic current on Ag working electrode flowed at lower than 0.0 V (vs. Na / Na ⁺). In the liquid Ga electrode, cathodic current started at 0.6 V, then the cathodic current increased again at lower than about 0 V. The color of the liquid Ga surface changed slightly from bright metallic to black and gray. When the potential was reversed to anodic direction, the color of the Ga surface changed to gray, black and bright metallic. In liquid Ga electrode, start of cathodic current was more positive potential compared with solid Ag electrode. It is considered the cathodic current at range from 0.0 to 0.6 V may relate to formation of Na-Ga intermetallic compound. In constant current electrolysis, the potential change showed in two steps with similar the voltammogram measurement, color of the Ga surface changed to black in the first step and gray in the second step.

In the open-circuit potential measurement, 3 potential plateaus of 1.9, 0.58, and 0.38 V (vs. Na / Na ⁺) were observed. The potential plateau at 1.9 V corresponds to pure Ga, the potential plateaus 0.58 and 0.38 V may correspond to Na-Ga intermetallic compounds with different composition. It is known from the Na-Ga binary phase diagram that there are two intermetallic compounds, Ga₄Na and Ga₃₉Na₂₂. In the surface color change of Ga electrode to black, it is considered that formation of Ga₄Na on the surface because small amount of Na liquid was electrodeposited on the Ga electrode. Further, by depositing Na on the surface, Ga₃₉Na₂₂ may be formed on the surface, and even if more Na is deposited on Ga₃₉Na₂₂, the intermetallic compound does not change any more, and metal Na is deposited on the Ga₃₉Na₂₂. Therefore, the color of the electrode surface may have changed from black to gray.

Nitrous oxide (N₂O), of which global warming potential is 310 times of CO₂, is a powerful greenhouse gas [1]. The concentration of N₂O has increased steadily from various agricultural and industrial activities such as cultivating crops in low-pH soil and manufacturing semiconductor devices [2,3]. Therefore, the importance of studying N₂O reduction is emphasized. Thermal decomposition methods focusing on thermal catalyst have been reported, but there are limits such as process temperature and low efficiency [4,5].

Electrochemical reduction of N₂O can be an attractive method since it is conducted at low temperatures and has low energy consumption. However, in aqueous solutions, the electrochemical reduction of N₂O inevitably accompanies hydrogen evolution reaction (HER), which is the main side reaction. There were many trials to reduce HER through metal and metal oxide catalysts, but they also had limits in terms of efficiency and high cost of noble metal materials [6-9].

Ionic liquids are promising electrolytes in gas reduction field replacing water because they can avoid hydrogen evolution and enhance the solubility of gaseous reactants. Some imidazolium-based ionic liquids were adopted to electrochemical reduction of CO₂ because of their favorable properties such as wide electrochemical potential window and high CO₂ dissolution capacity [10,11]. Cations of ionic liquids are known to accept oxygen from CO₂, and then return to initial state at counter electrode emitting oxygen [12]. According to the same principle, ionic liquids may also be suitable for the electrochemical reduction of N₂O.

In this study, an ionic liquid/organic solvent mixture was adopted as an electrolyte for the electrochemical reduction of N₂O. The ionic liquid and organic solvent used for electrolyte were 1-butyl-3-methylimidazolium tetrafluoroborate ([BMIM][BF₄]) and propylene carbonate (PC), respectively. The use of [BMIM][BF₄] in place of aqueous solutions could improve the solubility of N₂O and also avoid the hydrogen evolution reaction, which is the main side reaction in an aqueous solution.

Herein we observed the difference of LSV curves in aqueous solution and [BMIM][BF₄]. Hydrogen evolution reaction was excluded, but the low current density was obtained in [BMIM][BF₄]. The low electrical conductivity and high viscosity of [BMIM][BF₄], the main disadvantages in the context of electrochemical reactions, were compensated on introducing PC. We measured the viscosity and electrical conductivity in various [BMIM][BF₄]/PC composition, and the composition of the [BMIM][BF₄]/PC was optimized considering both the electrical conductivity and N₂O stability in the solution. The current and faradaic efficiency could be evaluated during chronoamperometry in aqueous solution and [BMIM][BF₄]/PC. For the optimized [BMIM][BF₄]/PC, high current efficiency and faradaic efficiency (>90%) were achieved for N₂O reduction to N₂.

Figure captions

Figure 1. LSV curves for Cu electrode in (a) 0.3 M K₂SO₄ aqueous solution and (b) [BMIM][BF₄]. The scan rate was 50 mV s^-1.

Figure 2. (a) Viscosity (black line) and conductivity (blue line) changes for [BMIM][BF₄]/PC according to [BMIM][BF₄] content (b) Current efficiency–time curves recorded at –1.3 V vs. SCE in 0.3 M K₂SO₄ aqueous solution (blue line) and at –2.27 V vs. Fc/Fc⁺ in 75 vol% [BMIM][BF₄] (red line).

References

[1] C. Kroeze, Sci. Total Environ., 143 (1994) 193-209.

[2] P.J. Crutzen, Q. J. Royal Meteorol. Soc., 96 (1970) 320-325.

[3] R. N Van den Heuvel, S. E Bakker, M. Jetten, M. Hefting, Geobiology, 9 (2011) 294-300.

[4] M. Konsolakis, ACS Catal., 5 (2015) 6397-6421.

[5] Z. Liu, F. He, L. Ma, S. Peng, Catal. Surv. Asia, 20 (2016) 121-132.

[6] S. Baek, K.H. Kim, M.J. Kim, J.J. Kim, Appl. Catal. B, 217 (2017) 313-321.

[7] S. Baek, K.H. Kim, I. Choi, O.J. Kwon, J.J. Kim, Chem. Eng. Trans., 335 (2018) 915-920.

[8] A. Kudo, A. Mine, J. Electroanal. Chem., 426 (1997) 1-3.

[9] K.H. Kim, T. Lim, M.J. Kim, S. Choe, S. Baek, J.J. Kim, Electrochem. Commun., 62 (2016) 13-16.

[10] D. Vasilyev, E. Shirzadi, A.V. Rudnev, P. Broekmann, P.J. Dyson, ACS Appl. Energy Mater., 1 (2018) 5124-5128.

[11] L. Sun, G.K. Ramesha, P.V. Kamat, J.F. Brennecke, Langmuir, 30 (2014) 6302-6308.

[12] L. Gu, Y. Zhang, Unexpected CO₂ splitting reactions to form CO with N-heterocyclic carbenes as organocatalysts and aromatic aldehydes as oxygen acceptors, J. Am. Chem. Soc., 132 (2010) 914-915.

Figure 1

1. Introduction

Titanium is known as a metal which has excellent properties such as high specific strength and high corrosion resistance. However, there are two problems that prevent widespread use of titanium. One is the high smelting cost and the other is the poor workability. As a method to solve these problems, titanium plating is an attractive technique because superior surface properties of titanium can be utilized. For some applications, physical vapor deposition (PVD) is practically used as a titanium plating method. However, the PVD method has high plating cost and can be used for only simple-shaped substrates. On the other hand, electrodeposition of titanium in high-temperature molten salt is a promising plating method which is expected to have lower cost and can be used for complex shaped substrates. Accordingly, there have been many reports on the titanium electrodeposition in high-temperature molten salts [1–3].

We have already reported the electrochemical behaviors of Ti(III) ions and the electrodeposition of titanium in KF–KCl and LiF–LiCl molten salts [4–7]. Furthermore, we found that LiF–LiCl has an advantage for electrodepositing smooth titanium films due to its lower melting point (774 K at the eutectic composition [8]); smoother titanium films were obtained at lower temperature by suppressing grain growth of titanium [9]. Thus, in the present study, we investigated the effect of electrolysis conditions such as current density and concentration of Ti(III) ions on smoothness of titanium films in LiF–LiCl eutectic molten salt.

2. Experimental

The experiments were conducted in LiF–LiCl eutectic molten salt in an Ar glove box at 823 K. Li₂TiF₆(0.50–5.0 mol%) and Ti sponge (0.33–3.3 mol%) were added to the bath and Ti(IV) ions were converted to Ti(III) ions by comproportionation reaction. Ni plate, Mo flag, and Au flag electrodes were used as the working electrode. The counter and reference electrodes were Ti rods. The potential of the reference electrode was calibrated by Cl₂/Cl⁻ potential measured at a glass-like carbon rod electrode. Samples were prepared by galvanostatic electrolysis of Ni plate substrates. The samples were analyzed by SEM/EDX after washing with distilled water and 1M Al(NO₃)₃ aqueous solution to remove adhered salts.

3. Result and Discussion

The electrodeposition was carried out by changing the added amounts of Li₂TiF₆from 0.50 to 5.0 mol% and Ti sponge from 0.33 to 3.3 mol%. The cathodic current density was also varied from 25 to 1000 mA cm⁻². The surface roughness of the obtained samples was measured by SEM. As a result, smooth Ti films were obtained at high Ti(III) concentrations and low current densities. As a typical example of the smooth film, Fig. 1 shows an appearance of the sample prepared at 50 mA cm⁻² and 60 C cm⁻² after addition of 3.0 mol% Li₂TiF₆and 2.0 mol% Ti sponge. The electrodeposited titanium film had metallic luster. Fig. 2 shows a cross-sectional SEM image of the film, indicating the smooth and compact deposit.

In the poster session, the solubility of Ti(III) ions will be presented. Also, the effects of Ti(III) ion concentration and cathodic current density on the morphology will be discussed.

Acknowledgement

The present address of Kouji Yasuda is Graduate School of Engineering, Kyoto University.

References

[1] H. Takamura, I. Ohno, and H. Numata, J. Jpn. Inst. Metals, 60, 388 (1996). [in Japanese]

[2] A. Robin and R.B. Ribeiro, J. Appl. Electrochem.,30, 239 (2000).

[3] V. V. Malyshev and D. B. Shakhnin, Mater. Sci., 50, 80 (2014).

[4] Y. Norikawa, K. Yasuda, and T. Nohira, Mater. Trans., 58, 390 (2017).

[5] Y. Norikawa, K. Yasuda, and T. Nohira, Electrochemistry,86, 99 (2018).

[6] Y. Norikawa, K. Yasuda, and T. Nohira, J. Electrochem. Soc., 166, D755 (2019).

[7] Y. Norikawa, K. Yasuda, and T. Nohira, J. Electrochem. Soc., 167, 082502 (2020).

[8] J. Sangster and A. D. Pelton, J. Phys. Chem. Ref. Data, 16, 509 (1987).

[9] Y. Norikawa, K. Yasuda, K. Numata, M. Ogawa, M. Majima, and T. Nohira, Abstract of The 50th Symposium on Molten Salt Chemistry, p. 76, Tokyo, Japan (2018).

Figure 1

The production of high strength carbon fibers is an energy intensive process where a significant cost involves the wet or dry spinning of polyacrylonitrile (PAN) fiber precursors. Melt spinning PAN fibers would allow for significant reduction in production cost and production hazards. Ionic liquids (ILs) are an attractive fiber processing medium due to their negligible vapor pressure and low toxicity. In addition, they are carbon forming precursors; upon carbonization residual ILs can enhance carbon yield although primarily useful for plasticized melt spinning of PAN precursor fibers. In this presentation, the influence of the molecular structure of the ILs and the control of the plasticizing interactions with PAN during melt spinning will be discussed. The structure, thermal and mechanical properties of the melt spun PAN fibers are investigated by a combination of various characterization methods, such as differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), X-Ray Diffraction (XRD) and mechanical testing. Our results demonstrate that ionic liquid structure and counter-anions influence the PAN fiber formation. More specifically, ILs containing bromide counter-anions produced PAN precursor fibers with increased mechanical properties compared to ILs containing chloride anions. We believe that our research can provide foundation to understand the influence of ILs on melt spinning of PAN fibers and gives the guidelines for the more cost/energy efficient production of PAN-based carbon fibers.

1. Introduction

Recently, the demand for solar cells has been increasing to realize low-carbon society. Among various types of solar cells, crystalline Si solar cells have dominated the photovoltaic market due to the high efficiency, excellent stability, and abundant natural resources. However, the low productivity of the Siemens process and the considerable kerf loss in the Si-slicing step are the main drawbacks in the conventional production method of the Si substrates. Since the demand for crystalline Si solar cells will continue to increase, the development of an alternative production method of Si substrate is strongly required.

One of the expected methods is a direct formation of Si films on an inexpensive substrate [1–3]. We have proposed the electrodeposition of Si films from KF–KCl molten salt, in which SiCl₄ gas is used as a Si source [4,5]. We have already investigated the optimum conditions for obtaining adherent, compact, and smooth Si films in molten KF–KCl–K₂SiF₆ at 923 K [6], and the effect of temperature and current density on the Si electrodeposition in molten KF–KCl–K₂SiF₆ at 923–1073 K [7]. However, the detailed analysis including the evaluation of semiconductor characteristic has not been performed for the Si films.

In the present study, we prepared crystalline Si films on graphite substrates under several conditions in molten KF–KCl–K₂SiF₆ at 1023 K. The obtained Si films were analyzed by XRD and SEM/EDX. The semiconductor characteristic was also evaluated by a photoelectrochemical measurement which had been used in the report by Bard et al. [2,3].

2. Experimental

2.1 Electrodeposition of Si

The electrodeposition was conducted in molten KF–KCl–K₂SiF₆ in Ar atmosphere at 1023 K in a glove box. As a Si source, 3.5 mol% of K₂SiF₆ was added to the bath. A graphite plate was used as the working electrode. The counter and the reference electrodes were Si rods. Si films prepared by galvanostatic electrolysis were analyzed by XRD, SEM/EDX, and a photoelectrochemical measurement.

2.2 Photoelectrochemical measurement

Photoelectrochemical measurement was conducted by linear sweep voltammetry. The electrolyte was prepared by mixing TBAClO₄ (0.3 M) and EVBr₂ (0.05 M) in CH₃CN at room temperature.

2TBAClO₄ + EVBr₂ → EV(ClO₄)₂ + 2TBABr↓

After the reaction was completed, the supernatant liquid in which EV(ClO₄)₂ was dissolved was used as the electrolyte. A Pt plate and an Ag⁺/Ag electrode were used as the counter and reference electrode, respectively. A Xe lamp (100 mW cm⁻²) was used as a light source. The light was chopped at a frequency of 1 Hz during the linear sweep voltammetry.

3.Result and discussion

3.1 Electrodeposition of Si

Galvanostatic electrolysis was conducted under the conditions of the added amount of K₂SiF₆ of 3.5 mol% and the cathodic current density of 100 mA cm⁻² for 15 min. As shown in Fig. 1(a), the deposit on a graphite substrate has dark gray color. From XRD analysis, the deposit was confirmed to be crystalline Si. Fig. 1(b) shows a surface SEM image of the Si film, indicating compact and smooth surface with crystal grains of around 10 μm. The current efficiency was calculated to be 90.1 % from the amount of electricity and the weight increase.

3.2 Photoelectrochemical measurement

The Si film on a graphite substrate was covered with an insulating coating to expose only the Si part and to prescribe the electrode area. Fig. 2 shows a photoresponse of the Si film during the linear sweep voltammetry, where the light is chopped at a frequency of 1 Hz. Cathodic currents change depending on the light chopping, indicating that the obtained Si film is p-type semiconductor.

Acknowledgement

The present address of Kouji Yasuda is Graduate School of Engineering, Kyoto University.

References

[1] E. Juzeliunas and D. J. Fray, Chem. Rev., 120, 1690 (2020).

[2] J. Zhao, H. Yin, T. Lim, H. Xie, H. Hsu, F. Forouzan, A. J. Bard, J. Electrochem. Soc., 163, D506 (2016).

[3] X. Zou, L. Ji, J. Ge, D. R. Sadoway, E. T. Yu, and A. J. Bard, Nat. Commun., 10, 5772 (2019).

[4] K. Maeda, K. Yasuda, T. Nohira, R. Hagiwara, and T. Homma, J. Electrochem. Soc., 162, D444 (2015).

[5] K. Yasuda, K. Maeda, R. Hagiwara, T. Homma, and T. Nohira, J. Electrochem. Soc., 164, D67 (2017).

[6] K. Yasuda, K. Maeda, T. Nohira, R. Hagiwara, and T. Homma, J. Electrochem. Soc., 163, D95 (2016).

[7] K. Yasuda, K. Saeki, T. Kato, R. Hagiwara, and T. Nohira, J. Electrochem. Soc., 165, D825 (2018).

Figure 1

The need for safe, low cost, high energy and density storage devices is ubiquitous the world over. Satisfying this need requires various energy storage and conversion applications, one of which is the aluminum ion batteries. One problem with expansion in their development and commercialization is the availability of optimized electrolytes. Recently, deep eutectic solvents (DES) have gained attention as electrolytes for energy storage devices. These solvents are similar to ionic liquids, but are generally cheaper and easier to prepare. Additionally, they are known to dissolve metal oxides which plague AIBs operation, and their solubility is dependent on the hydrogen bond donor. Because of their inherent tunability, we studied the effect of AlCl₃ concentration (1:1 – 1.7:1 molar ratio), amide type (acetamide (AA), butyramide (BA), propionamide (PA)), viscosity reducing additive type and concentration, on the aluminum ion species and transport in AlCl₃:amide DES electrolytes using multi-Nuclear (¹H, ²⁷Al) Magnetic Resonance (NMR) and AC Impedance Spectroscopy techniques, as a function of temperature. Below are a few noteworthy observations.

Figure 1. VFT plots of the ionic conductivity for AlCl₃:AA (left) and AlCl₃:PA (right) DES electrolytes.

Thirdly, the variable temperature ¹H and ²⁷Al spectra and T₁ data for the pure and additives included DES electrolytes were very dependent on concentration, additive type and concentration, and species type. For example, in the 1:1 molar ratio electrolyte, the assigned AlCl₂(amide)⁺ species had the longest T₁, while the AlCl₂(amide)₂⁺ had the shortest. Additionally, the inclusion of additives such as fluoroethylene carbonate (FEC) and propylene carbonate (PC) caused species specific increasing local ion dynamics, possibly through the reduction of bulk viscosity effects and Coulombic interactions. These results and more will be expounded upon in our presentation.

Firstly, for both AlCl₃:PA and AlCl₃:BA DES electrolytes, a maximum in ionic conductivity was observed at the 1.3:1 molar ratio. For AlCl₃:AA DES the maximum occurred at the 1.5:1 molar ratio. Secondly, for the three amides conductivity values fell in the range of ~1 – 11 mS/cm over the temperature range of 288 – 363K. Additionally, the temperature dependence displayed curve-like behavior and were fitted using the VFT equation to reveal high levels of fragility and dynamic behaviors similarities to some pure ILs where the effective inter-conversion between the trans and cis conformations of the anion facilitated faster ion dynamics.

Figure 1

The Ni-Ti shape memory alloys are currently a topic of notable interest in medicine. They provide a unique opportunity to make novel surgical implants and instruments for vascular and orthopedic surgery. However, a high nickel content in these alloys could cause the problem of biocompatibility because of nickel toxic effect. In turn, a tantalum is successfully used in medicine as a wire or sheet. It does not irritate the living tissue and does not harm the functioning of the organism, but it has a high specific weight. Therefore, using of tantalum coatings on various materials is more appropriate than utilization of a bulk metal. Thus, it is necessary to improve a corrosion resistance of nickel-titanium alloys (nitinol) that can be achieved with applying of protective porousless tantalum coating.

For obtaining of tantalum coatings the NaCl-KCl-NaF(10 wt%)-K₂TaF₇(10 wt%) melt was used at a temperature 1023 K. Direct (DC), unsteady state (USC) and pulsed current (PC) were supplied for electrodeposition. A cathodic current density was changed from 5 to 200 mA/cm² at the galvanostatic mode. The same cathodic current densities and the time of electrolysis (t_el = 0.5-5.0 s) with pauses (t_p = 1-5 s) were used at unsteady state and pulsed electrolysis. It was found that a pulsed electrolysis allow to produce coatings with lower roughness than galvanostatic electrolysis.

Electroreduction of tantalum complexes in chloride-fluoride melt at a nitinol electrode was investigated. A voltammetric curve registered at a nitinol electrode except of the electroreduction peak of tantalum fluoride complexes to tantalum has several peaks corresponding to the formation of intermetallic compounds of nickel and tantalum.

In the case of formation intermetallic compounds between a substrate and coating is an important aspect of the coating adhesion to substrate. The adhesion was measured by cross-sections method using a tester of adhesion Elcometer 107. The measurements showed that coatings obtained by DC and USC could be classified as adhesion to international standards ISO (1) and ASTM (4B). At the same time, coatings deposited on nitinol by PC have a maximum class of adhesion ISO (0) and ASTM (5B). These results were obtained for intermetallic compounds with a thickness around 1-2 μm.

Additional experiments were carried out for obtaining of thick layers of intermetallics and study their composition by SEM and EDX.

The porosity of tantalum coatings was determined by Erhard's technique based on measuring a current dissolution at a certain potential, in which dissolves a substrate material and coatings remain passive.

Corrosion resistance of the composition tantalum coating-nitinol was studied in different corrosion media.

High purity rare earth metals are essential for research and use as high-performance materials. Even a minute amount of impurities can have significant influence on physical and chemical properties of metals, so some special properties of rare earth materials are exhibited only when they are in the state of high purity. Molten salts are promising reaction media for extractive metallurgy. In particular, molten chlorides are good electrolytes for selective dissolution or deposition of pure reagents; using molten salts provides a promising route for treatment of raw materials. In addition, molten salts proved to be suitable media for metal electrowinning and electrorefining. Available experience in high-temperature electrochemistry allows producing metals in the solid form. One of the advantages of molten salts is their variety. So, it is possible to find a solvent, which chemical and electrochemical characteristics are suitable to carry out a given process.

The goal of this work was studying the mechanism of electrochemical reduction of dysprosium ions in molten LiCl–KCl eutectic for production of high purity dysprosium metal.

The cathodic reduction of Dy(III) ions was studied on inert molybdenum electrode in the temperature range of 723–843 K under inert atmosphere. Cyclic voltammograms contained one cathodic peak at–3.19±0.11 V and the corresponding anodic peak at –2.95±0.11 V vs. chlorine reference electrode, Fig. The cathodic peak potential was not constant and shifted to the negative values with increasing scan rate. The cathodic peak current was directly proportional to the square root of the polarization rate. It was found that increasing scan rate led to an expected increase of irreversibility of the cathode process. The number of electrons (n) of the electrode reaction for reduction of Dy(III) ions was determined by square-wave voltammetry and the calculated value was equal to 2.93±0.05. According to the theory of linear sweep voltammetry the redox system Dy(III)/Dy(0) is irreversible and controlled by the rate of the charge transfer.

So, cathodic reduction of dysprosium ions proceeded in one three-electron electrochemical step at the potential of –3.19 V versus the Cl^–/Cl₂ reference electrode:

DyCl₆^3– + 3 e^– = Dy + 6 Cl^–.

Temperature dependence of Dy(III)/Dy couple apparent standard potential was determined by chronopotentiometry at the zero current. The experimental values are described by the linear equation:

E*_Dy/Dy(III) = –(3.401±0.009) + (6.2±0.1)10^–4T.

The apparent standard Gibbs energy change, enthalpy, and entropy of dysprosium trichloride formation from the elements in fused LiCl–KCl eutectic and activity coefficient of DyCl₃ were calculated.

The influence of different parameters on the composition of the cathode product was investigated. It was found that for fine purification of dysprosium from its impurities, the electrolysis should be carried out in two stages, in which the first stage is the purification electrolysis, and the second one is the base electrolysis.

The reported study was funded by Russian Foundation for Basic Research according to the research project No. 20-03-00743.

Figure. Typical cyclic voltammograms for the reduction of dysprosium trichloride on molybdenum electrode (S = 0.14 cm²) in fused LiCl–KCl eutectic at 723 K. m(DyCl₃) = 4.110^–2 mol/kg. Scan rates, V s⁻¹: 1 – 0.075; 2 – 0.2; 3 – 0.5.

Figure 1

Melts based on mixtures of alkali and/or alkaline earth halides are considered as prospective media for electrowinning rare earth metals as well as for pyrochemical reprocessing spent nuclear fuels. Fluoride or mixed fluoride-chloride baths can be operated at high temperatures yielding molten rare earth metals (REMs) thus simplifying separation of the metal and salt. One of the problems in using fluoride melts is possible formation of fluorine at the anode. This can be avoided by adding a rare earth oxide to the melt both as the source of REM and oxide ions. The latter will be oxidized to oxygen producing carbon mono- or dioxide at the anode. The limiting factor for feeding the electrolysis bath with REM oxides is their solubility in the halide melt.

The aim of the present study was determining the effect of temperature and melt composition on solubility of REM oxides in fused halides. The experiments were performed in CaCl₂–CaF₂ mixtures containing 20 or 75 mol. % CaF₂; BaCl₂–BaF₂ mixtures containing 15 or 73 mol. % BaF₂; equimolar CaF₂–BaF₂ mixture and NaCl–NaF eutectic mixture (34 mol. % NaF). Solubility of REM oxides was determined by the method of isothermal saturation. Time required for reaching the equilibrium between solid REM oxides and fused salts was determined in a preliminary set of experiments. To compare the behavior of 4f- and 5f-elements, solubility of uranium dioxide was also measured.

The measurements were performed at the temperatures up to 1400 ^oC under argon atmosphere. The lower limit of the temperature range varied from 700 to 1100 ^oC depending on the melting temperatures of the salt mixtures used. Oxides of yttrium, lanthanum, cerium, praseodymium, neodymium and samarium were selected for the study. To assess a possible mutual influence of rare earth elements on solubility of their oxides in fused salts the solubility of a mixture of REM oxides was determined in a separate series of experiments and concentrations of individual REMs in the melt was determined.

Solubility of REM oxides increased with increasing temperature and an example of effect of temperature on solubility of neodymium oxide in various melts is shown in Fig.

Fig. Concentration of neodymium in alkali and alkaline halide based melts saturated with Nd₂O₃. Melt: CaCl₂–CaF₂ 20 mol. % (1); CaCl₂–CaF₂ 75 mol. % (2); BaCl₂–BaF₂ 73 mol. % (3); BaCl₂–BaF₂ 15 mol. % (4); CaF₂–BaF₂ 50 mol. % (5; and NaCl–NaF 34 mol. % (6).

Figure 1

Current research has shown that Li₂CO₃-Na₂CO₃ electrolyte has better ionic conductivity, a lower rate of cathode dissolution and a lower electrolyte loss than Li/K carbonate. Lower oxygen gas solubility has been one of the disadvantages of the Li/Na electrolyte. Modification of the electrolyte such as the addition of alkali metals and alkaline earth metals have led to an improvement in the oxygen reduction reaction (ORR). In this experiment, the effects of lanthanum, strontium, cesium, and calcium were analyzed using several electrochemical half-cells and 7 cm² coin-type MCFCs operated at 973 K. A eutectic mixture of Li₂CO₃ and Na₂CO₃ (52:48) was used as the electrolyte. For the half-cell, the reference electrode compartment was supplied with a 33% Air/67% CO₂ gas mixture while the electrolyte compartment was supplied with different CO₂/Air gas compositions. The anode of the coin cells was supplied with a mixture of H₂, CO₂ and H₂O in the ratio 72:14:14 mol% and the cathode was supplied with air and CO₂ in the ratio 70:30 mol%. 0.5 mol% La₂O₃, 3 mol% Cs₂CO₃, 3 mol% CaCO₃, and 3 mol% SrCO₃ were added separately and also as a combination of two (i.e. La₂O₃/SrCO₃, CaCO₃/La₂O_3, and Cs₂CO₃/La₂O₃) to the Li/Na electrolyte. The effects of these additives were observed and analyzed using Electrochemical analysis techniques such as electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV) were used for the half-cell analysis. Steady-state polarization (SSP), internal resistance (IR) using the Nyquist plot, and step chronopotentiometry were employed in the analyses of the coin cell. The mass transfer parameter (C√D) values obtained from the cyclic voltammetry measurements increased as the oxygen partial pressure increases in all electrolytes. From the results obtained, the combined addition of Ca/La proved to have the most significant effect on the overall performance of the cell.

Today, quantum-chemical studies of the mechanism of electron transfer in molten salts based on a direct calculation of the transition state are faced with practically insurmountable computational difficulties. As our experience shows, model systems assigned to investigate the mechanism of charge transfer should include, in addition to the electroactive complex, two more of its coordination spheres. For electrolytes based on alkaline earth metal halides, this can be, for example, MTiF₆+12MX₂ type systems (M – Mg, Ca, Sr, Ba; X – F, Cl). The search for a transition state by standard methods will require enormous computer time and is almost unrealistic. In this paper, we proposed another approach, which is based on the analysis of frontier molecular orbitals (FMO) under various deformations of the initial structure.

The aim of this work was the using of the method of frontier molecular orbitals to study the mechanism of the electron transfer in the CaTiF₆+12CaCl₂ model system.

The geometry optimization of structures was performed with the Firefly program package, partially based on the source code of the GAMESS(US) program, by the density functional theory DFT/UHF method with the use of the B3LYP hybrid functional. For the F and Cl atoms quasi-relativistic basis set Stuttgart RLC ECP was used; for Ti and Ca – Stuttgart RSC 1997 ECP basis set.

In this work, we were interested in the state of the complex TiF₆^2- near the cathode surface. For this reason, on one side of the model system, a flat boundary layer consisting of 12-15 chlorine and calcium ions was formed. The titanium complex is in contact with one of the calcium cations belonging to the boundary layer, because our previous studies indicate the bridge nature of electron transfer to the TiF₆^2- complex. Note that adding of carbon cluster to the main system CaTiF₆+12CaCl₂ would require a drastic increase of the computation time. However, for qualitative conclusions, it is enough to analyze the CaTiF₆+12CaCl₂ system.

Based on the experimental data and the quantum-chemical analysis of FMO in the CaTiF₆ + 12CaCl₂ system, a mechanism for the charge transfer was proposed. The geometric structure of the transition state was shown to be not intermediate between the initial and final states of the system. As a result, it was found that the structure of this state is much less ordered than the initial structures of the boundary layer. The high efficiency of FMO method allows recommending this method as the instrument for testing hypotheses on the mechanism of electron transfer in molten salts.

Ammonium (NH₄⁺) ion has some unique properties as a charge carrier, i.e., natural abundance and lightweight properties. In addition, compared with metal cations (such as Li⁺, Na⁺, and K⁺), NH₄⁺ ion shows the highest intercalation potential and comparable cycling performance for Prussian blue-type positive electrode materials, KM[Fe(CN)₆] (M = Ni and Cu).[1] More recently, the first rocking-chair-type NH₄⁺ ion battery has been proposed.[2] The electrolytes for NH₄⁺ ion batteries, however, have rarely been investigated except for aqueous solutions and solid state electrolytes [2,3]. Therefore, electrolytes which enable fast NH₄⁺ ion transport at medium temperatures, i.e. 100–200 °C, have been of great interest.

Molten salts, especially ionic liquids (ILs), have received considerable attention as novel electrolyte materials for future electrochemical devices owing to their various characteristics such as negligible volatility, low flammability, high thermal and electronic stability, and high ionic conductivity. We have lately reported a series of studies on a hydronium (H₃O⁺) solvate IL, which is described as [H₃O⁺·18C6]Tf₂N (18C6 = 18-crown-6-ether; Tf₂N = bis(trifluoromethylsulfonyl)-amide, Tf = SO₂CF₃).[4–6] In [H₃O⁺·18C6]Tf₂N, protic H₃O⁺ ion (i.e., solute) is solvated by 18C6 ligand (i.e., solvent) to form a [H₃O⁺·18C6] complex cation (i.e., solvate), and the counter anion is the Tf₂N^– anion.

It is well known that 18C6 features a six-oxygen cavity and matches well a C_3v coordination with NH₄⁺ as well as H₃O⁺, while the degree of off-center of [NH₄⁺·18C6] is larger than that of [H₃O⁺·18C6]. As far as we know, however, NH₄⁺-based molten salts or ILs for electrolytes have never been reported, while alkyl ammonium-based ones have already been studied.[7]

Herein, we report synthesis and physicochemical properties of [NH₄⁺·18C6]Tf₂N (Figure 1). We synthesized NH₄Tf₂N by neutralization of HTf₂N with NH₃ before adding equimolar amount of NH₄Tf₂N to 18C6 to obtain [NH₄⁺·18C6]Tf₂N. A differential scanning calorimetry revealed that [NH₄⁺·18C6]Tf₂N melts at around 100 °C (Figure 2). The ether-coordinated ammonium amide has potential applications as electrolyte materials toward a new class of ammonium ion batteries.

This work was supported financially by Grants-in-Aid for Scientific Research (B) (No.19H02490: A. K.) from the Japan Society for the Promotion of Science. A. K. also thanks the Joint Usage/Research Program on Zero-Emission Energy Research, Institute of Advanced Energy, Kyoto University (ZE2020A-09).

[1] C. D. Wessells et al., J. Electrochem. Soc., 159, A98 (2012).

[2] X. Wu et al., Angew. Chem. Int. Ed., 56, 13026 (2017).

[3] R. C. T. Slade et al., J. Solid State Chem., 32, 87 (1980).

[4] A. Kitada et al., J. Electrochem. Soc., 165, H121 (2018).

[5] A. Kitada et al., J. Electrochem. Soc., 165, H496 (2018).

[6] K. Kawata et al., J. Electrochem. Soc., 167, 046508 (2020).

[7] J. Qu et al., Tribol. Lett., 22, 207 (2006).

Figure 1

In light of the ongoing transition from non-renewable to renewable energy sources, organic and hybrid solar cells have been studied due to their cost-efficiency and the smaller environmental impact. The current work is aiming for the establishment of ionic liquid precursors (ILPs) for the synthesis of complex metal chalcogenides. In organic photovoltaic systems these can then be used, e.g., as the active layers due to them acting as additional p-type semiconductors. The major difference to more conventional approaches is that the ILP not only acts as a metal source but also as a morphology directing template and as a stabilizer for the resulting nanoparticles.^1,2 The general approach is using metal-containing ionic liquids, where the metal is an integral part of the IL anion as the metal source and to generate mixed metal chalcogenides directly from these complex precursors.^3,4

In the current proof-of-concept study, thirteen N-butylpyridinium salts, including three monometallic compounds [C₄Py]₂[MCl₄], nine bimetallic compounds [C₄Py]₂[M_1-x^aM_x^bCl₄] and one trimetallic compound [C₄Py]₂[M_1-y-z^aM_y^bM_z^cCl₄] (M = Co, Cu, Mn; x = 0.25, 0.50 or 0.75 and y = z = 0.33), were synthesized and their structure, thermal, optical and electrochemical properties were analysed. All compounds are ILs with melting points between 69 and 93 °C. The successful synthesis of mixed metal ILs with distinct compositions was confirmed using ICP OES, with single crystal X-ray diffraction showing consistent asymmetric units. Furthermore, X-ray diffraction shows that all ILs are isostructural, thus confirming that both the anion and the cation are stable and reliable building blocks for MILs. A possible application regarding the direct use of these ILs in electrochemical devices was further analysed. The electronic conductivity at room temperature is between 10^-2 and 10^-8 S cm^-1. This correlates with the optical band-gap measurements indicating rather poor semiconductors with large band gaps. However, at elevated temperature approaching the melting points, the conductivities reach up to 1.47 10^-1 at 70 ºC. As a result, future electrochemical applications are possible, especially in a moisture-sensitive environment at room temperature. Moreover, corrosive behaviour and the correlation between conductivity and composition were investigated for the first time, e.g. showing a positive influence with the addition of manganese into the compound, possibly acting as an electron scavenger. Additionally, cyclic voltammetry shows promising electrochemical stability windows between 2.5 and 3.0 V.⁵ Furthermore, a big advantage of these ILPs is the fact that their properties, such as band gaps, can directly be adjusted by proper choice of the metals in the ILs.

Through a reaction of these metal-containing ILs with a sulphur source the respective metal chalcogenide (MC) nanoparticles will form.¹ The p-type semiconductor nanoparticles will then be used as a hole transport material in the bulk heterojunction to improve the charge transport as well as providing a broad optical absorption window.

1 Y. Kim, B. Heyne, A. Abouserie, C. Pries, C. Ippen, C. Günter, A. Taubert and A. Wedel, J. Chem. Phys., , DOI:10.1063/1.4991622.

2 A. Abouserie, G. El-Nagar, B. Heyne, R. Sarhan, Y. Kim, C. Pries, E. Ribacki and C. Günter, Hierarchically structured Copper Sulfide Microflowers with Excellent Amperometric Hydrogen Peroxide Detection Performance, 2019.

3 K. Thiel, T. Klamroth, P. Strauch and A. Taubert, Phys. Chem. Chem. Phys., 2011, 13, 13537.

4 A. Abouserie, K. Zehbe, P. Metzner, A. Kelling, C. Günter, U. Schilde, P. Strauch, T. Körzdörfer and A. Taubert, Eur. J. Inorg. Chem., 2017, 2017, 5640–5649.

5 C. Balischewski, K. Behrens, K. Zehbe, C. Günter and A. Taubert, In preparation.

Introduction

Tungsten metal is known for its robustness, especially high melting and boiling points and hardness. However, its hardness and brittleness make it difficult to work. Thus, electrodeposition of flat and dense tungsten films is worth investigating. There are a number of reports on the electrodepositon of tungsten from high temperature molten salts such as fluorides [1], chlorides [2], oxides [3], etc. It is known that dense and coherent tungsten deposits are more easily obtained from molten fluorides compared to molten chlorides. One drawback of typical fluoride melts like LiF–NaF–KF is difficulty in removing the adhered molten salts by water washing because LiF and NaF have limited solubility to water [4]. Although good tungsten deposits can be electrodeposited from molten oxides [3], the operation temperature tends to be very high, typically over 1123 K.

From this background, we proposed new electrolyte baths, molten KF–KCl [5,6] and CsF–CsCl. Since all the components salts of KF, KCl, CsF and CsCl have large solubility to water [4], solidified salts on deposited films can be easily removed by water washing.

Results & Discussion

First, electrodeposition of tungsten was investigated in eutectic KF–KCl melt to which WO₃ was added at 923 K. Cyclic voltammograms at copper and gold electrodes indicated the deposition of tungsten metal at more negative potential than 1.55 V vs. K⁺/K. The effects of electrolysis conditions on the morphology of deposits were studied by changing the cathodic current density from 12.5 to100 mA cm⁻². As shown in Fig.1(a), a dense tungsten film with a thickness of approximately 20 µm was obtained on a copper substrate electrode at 12.5 mA cm⁻² for 120 min. From the Raman spectra of molten KF–KCl and molten KF–KCl–WO₃ at 923 K, the coordination structure of W(VI) complex ion was indicated to be fac-[WO₃F₃]³⁻.

Secondly, eutectic CsF–CsCl melt was used to lower the operation temperature. After adding 2.0 mol% of WO₃, the electrodeposition was investigated using copper and gold electrodes at 823–923K. As shown in Fig.1(b), a flat and dense tungsten film was obtained by galvanostatic electrolysis at 12.5 mA cm⁻² for 120 min at 873 K.

Acknowledgement

This study is supported by Sumitomo Electric Industries, Ltd. The present address of Kouji Yasuda is Graduate School of Engineering, Kyoto University.

References

[1] S. Senderoff, G. Mellors, Science, 153, 1475 (1966).

[2] M. Masuda, H. Takenishi, A. Katagiri, J. Electrochem. Soc., 148, C59 (2001).

[3] K. Koyama, M. Morishita, T. Umezu, Electrochemistry, 67, 667 (1999).

[4] D. R. Lide (Ed.), CRC Handbook of Chemistry and Physics, 88th Edition, Chap.4, p.43-101, CRC Press, Boca Raton (2007).

[5] K. Maeda, K. Yasuda, T. Nohira, R. Hagiwara, T. Homma, J. Electrochem. Soc., 162, D444 (2015).

[6] T. Nohira, Molten Salts, 62, 3 (2019).

Figure 1

Aluminum is one of the first metals that has been successfully deposited from ionic liquids (ILs) and has been in the focus of intense research. However, the electrochemical kinetics of the aluminum deposition from ILs is still not fully understood.

This paper will discuss recent results from the authors' lab on the electrochemical kinetics of the deposition of aluminium from 1-ethyl-3-methyl-imidazolium based ILs. Overpotential measurements and electrochemical impedance spectroscopy (EIS) experiments were employed to extract kinetic data from potential-time transients (see Fig. 1). The transients were interpreted in terms of different contributions to the total overpotential. This includes a detailed discussion of the Ohmic drop, charge transfer, concentration and nucleation overpotential. The charge transfer mechanism of aluminum was investigated in terms of the anodic and cathodic rate-determining step (RDS) and oxidation and reduction mechanisms are suggested. Density Functional Theory (DFT) calculations were performed to evaluate if the suggested mechanism is thermodynamically reasonable.

Fig. 1 Potential-time transient for a constant-current step (j = 1.25 mA cm^-2) at an aluminum electrode in [EMIm]Al₂Cl₇ at 300 K. The inset shows the first 500 ms of the potential-time transient.

Figure 1

1. Introduction

Titanium metal has superior properties such as high specific strength, high corrosion resistance, biocompatibility, etc. In addition to these properties, crustal abundance of titanium is more than 40 times larger than those of commonly used copper and nickel. However, titanium is not used widely and commonly due to two problems: the high cost of present smelting method and the poor workability of itself. Therefore, both a new smelting method and a processing method are required.

Plating titanium metal on an inexpensive substrate is one of the methods to resolve these problems. Electrodeposition of titanium is a promising plating method from the viewpoints of the cost and the flexibility in shape of substrate. Thus, electrodeposition of titanium metal has been investigated for a long time using high-temperature molten salts [1–4].

We have already reported that compact, smooth, and adherent titanium films are electrodeposited in molten KF–KCl and LiF–LiCl containing Ti(III) ions at 923 K [5–8]. However, crystal grains of titanium become large with the increase of a film thickness, which results in a rough surface. Concerning the crystal grain size, Wei et al. electrodeposited titanium in molten LiCl–NaCl–KCl–K₂TiF₆ (+ Ti metal) at 723–973 K and reported that the grain size became small as the temperature decreased [1]. This report suggests that titanium films are expected to have a smoother surface at a lower electrodeposition temperature. Therefore, in the present study, we investigated the effect of temperature on morphology and smoothness of electrodeposited titanium films in LiF–LiCl eutectic melt at 823–973 K. The effect of temperature on the electrochemical behavior of Ti(III) ions was also studied.

2. Experimental

The experiments were conducted in LiF–LiCl eutectic melt (LiF:LiCl = 30:70 mol%, m.p.: 774 K [9]) in an Ar glovebox. The experimental temperature was varied from 823 K to 973 K. Li₂TiF₆(2.00 mol%) and Ti sponge (1.33 mol%) were added to the bath and Ti(III) ions were prepared by comproportionation reaction. Ni plate, Mo flag, and Au flag electrodes were used as the working electrodes. The counter and reference electrodes were Ti rods. The potential of the reference electrode was calibrated by Li⁺/Li potential measured at a Mo flag electrode. Samples prepared by galvanostatic electrolysis of Ni plate substrates were analyzed by SEM/EDX after washing with distilled water to remove adhered salts.

3. Results and Discussion

Galvanostatic electrolysis was conducted at 823, 873, 923, and 973 K. Fig. 1 shows the optical and surface SEM images of the samples. The cathodic current density and electrolysis time were 100 mA cm⁻² and 10 min, respectively. The surface roughness (Sa) was measured by SEM, which is also shown in Fig. 1. All the samples have a metallic luster. The sample at 823 K has the highest brightness with metallic luster. The surface SEM observations indicate that the crystal grain size increases as the temperature rises. This tendency is reasonably explained by a previous work on the temperature dependent grain growth of Ti [10]. The value of Sa also increases as the temperature increases. The smoothest Ti film with Sa = 2.05 ± 0.22 μm was obtained at 823 K. These results conclude that Ti film with smoother surface can be electrodeposited at a lower temperature by suppressing the grain growth of Ti.

In the presentation, more detailed analysis results and the effect of temperature on the electrochemical behavior of Ti(III) ions will be discussed.

Acknowledgement

A part of this work was supported by JSPS Fellows grant number 19J15015. The present address of Kouji Yasuda is Graduate School of Engineering, Kyoto Univ.

References

[1] D. Wei, M. Okido, and T. Oki, J. Appl. Electrochem., 24, 923 (1994).

[2] H.Takamura, I. Ohno, and H. Numata, J. Jpn. Inst. Metals, 60, 388 (1996). [in Japanese]

[3] A. Robin and R. B. Ribeiro, J. Appl. Electrochem.,30, 239 (2000).

[4] V. V. Malyshev and D. B. Shakhnin, Mater. Sci., 50, 80 (2014)

[5] Y. Norikawa, K. Yasuda, and T. Nohira, Mater. Trans., 58, 390 (2017).

[6] Y. Norikawa, K. Yasuda, and T. Nohira, Electrochemistry, 86, 99 (2018).

[7] Y. Norikawa, K. Yasuda, and T. Nohira, J. Electrochem. Soc., 166, D755 (2019).

[8] Y. Norikawa, K. Yasuda, and T. Nohira, J. Electrochem. Soc., 167, 082502 (2020).

[9] J. Sangster and A. D. Pelton, J. Phys. Chem. Ref. Data, 16, 509 (1987).

[10] K. Okazaki and H. Conrad, Metall. Trans., 3, 2411 (1972).

Figure 1

Electrodeposition of Al-Ti alloys from EmImCl-AlC₃ ionic liquid containing TiCl₄

using cathodic-anodic pulse technique

Takahiro Kizawa¹, Taisuke Nozawa², Takehiko Kumagai³, Hisayoshi Matsushima³,

and Mikito Ueda³

(Graduate School of Engineering, Hokkaido University¹, NIPPON STEEL CORP.²

Faculty of Engineering, Hokkaido University³)

To improve the pitting corrosion resistance of aluminum in chloride environment, alloying with transition metals is conducted. Titanium is one of candidate of the transition metals for improvement of the corrosion resistance of aluminum, and Al-Ti alloys with high titanium concentration has better corrosion resistance¹⁾. The electrodeposition of Al-Ti alloys from EmImCl(1-ethyl-3-methyl-imidazolium chloride)-AlCl₃ containing TiCl₄ or TiCl₂ has been widely investigated for enhancement of titanium concentration in the electrodeposits^{2, 3)}. However, the electrodeposition of Al-Ti alloys by pulse electrolysis has not been investigated. Cathodic-Anodic Pulse deposition (CAP deposition), which combines Al-Ti deposition and Al dissolution, is considered to be one of the effective techniques to increase the Ti concentration in the electrodeposits. In this study, we investigated the electrodeposition of Al-Ti alloys by the CAP deposition in EmImCl-AlCl_3-containing TiCl₄ at 338 K.

EmImCl-AlCl₃ ionic liquid (molar ratio of 1 : 2) with 50 mM TiCl₄ was prepared in the glove box with Ar atmosphere. Voltammogram measurement was recorded at a potential range from -0.3 V to 1.5 V (vs. Al / Al (III)) and scan rate was 20 mV s^-1. Electrodeposition experiments of constant current, current pulse and CAP electrolysis were carried out. In the constant current and current pulse electrolysis, current density was from 3 to 10 mA cm^-2 and duty ratio was 0.83 (t_on = 1.0 s, t_off = 0.2 s). In the CAP electrolysis, the cathodic current density was 5 mA cm^-2, the anodic current density was from 0.50 to 2.0 mA cm^-2. The charge density was 20 C cm^-2 in the all experiments. The electrodeposits were washed with distilled water and ethanol, then dried, and weighed. After observing the surface of the electrodeposits by SEM, the electrodeposits were dissolved in an HF solution and the Ti concentration were measured by ICP-AES.

In the voltammogram measurement, three cathodic current waves at 0.95 V (vs, Al / Al (III)), 0.35 V, and 0.06 V were observed. The waves are considered to correspond to reduction reaction of Ti (IV) / Ti (III), Ti (III) / TI(II) and Ti(II) / Ti, respectively. At a potential lower than 0 V, the cathodic current increased due to electrodeposition of the Al-Ti alloys. During the positive potential sweep, oxidation waves corresponding to the three reduction waves were observed at 0.25 V, 0.55 V, and 1.1 V, respectively. In the voltammogram measurement, dissolution potential of Al-Ti alloy shifted to the noble side compared to the it of pure aluminum. It was found that the potential shift is an effect of titanium.

In the constant current electrolysis, the maximum Ti concentration in the electrodeposit was 11 at% at 5 mA cm^-2. Since the potential of electrodeposition of Al and Ti is close and the concentration of Al ions in the ionic liquid is much higher than the concentration of Ti ions, the electrodeposition of Al occurs preferentially during the electrolysis. In the current pulse electrolysis, the maximum Ti concentration in the electrodeposits was 20 at% at 5 mA cm^-2. It is considered that the Ti ions decrease at the electrode surface during the electrolysis, then recovery of the Ti ions occurs at off-time of the pulse, therefore Ti concentration in the electrodeposit increases compare to it of the constant current electrolysis. In the CAP electrolysis, the maximum Ti concentration was 21 at% at the anodic current density of 1.5 mA cm^-2. Comparing this result with the current pulse electrolysis, the increase in Ti concentration is small. It is indicating the dissolution reaction of only Al from Al-Ti alloys is difficult to occur.

1) J. R. Davis, Corrosion of Aluminum and Aluminum Alloys, ASM International,Materials Park,OH (1999).

2) R. T. Carlin, R. A. Osteryoung, J.S. wiles, and J. Rovng, Inorganic Chemistry, 29, 3003 (1990)

3) T. Tsuda, C. L. Hussey, G. R. Stafford, and J. E. Bonevich, J. Electrochem. Soc., 150, C234 (2003)

Al is widely used in daily necessities, automobiles, building materials and plating materials. In addition, since Al formed into a mirror surface has a reflectance of more than 90% in the visible light region, it has been used as a reflection plate of an optical device such as a microscope and an LED light in the optical field. Our group aims to deposit bright aluminum electroplating on a resin that is even lighter than aluminum. This technology can contribute to further weight reduction of mobile devices. However, there is a few reports of formation of bright aluminum electroplating film with a thickness of several 10 μm. Therefore, research was conducted focusing on the bright condition of aluminum electroplating and the uniformity of the thicker plating film. In this study, 1-ethyl-3-methylimidazolium chloride (EmImCl) and AlCl₃ a mixture ionic liquid and 1,10-Phenanthroline were selected as the electrolyte for Al plating and additive, respectively.

EmImCl-AlCl₃ ionic liquid (molar ratio of 1 : 2) with 20 mM 1,10-Phenanthroline was prepared in the glove box with Ar atmosphere. A Cu plate or a Cu-plated ABS resin was used for the working electrode and an Al plate or Al mesh was used for the counter electrode. Electrodeposition experiments were carried out by current pulse electrolysis with 20 mA cm^-2 and duty ratio of 0.83 (t_on = 1.0 s, t_off = 0.2 s). Charge density was 59.7 Ccm^-2 (Theoretical film thickness of 20μm). For the electroplated surface, surface observation and cross-sectional observation were carried out by SEM. The surface roughness and reflectance were measured by AFM and optical reflection probe(reference: Al mirror of 90% reflectance)

In the experiment using Cu plate working and Al plate counter, bright Al formed at only edge of the Cu plate with non-stirring condition of electrolyte. When the stirring rate of the electrolyte was increased, the area of the bright part increased, and the stirring rate of 400 rpm was appropriate rate for formation of bright surface. This may because the additive reaches the whole surface of the Cu surface by stirring. The roughness and the reflectance of the Al electroplated surface formed under the condition of the 400 rpm was 2-3 nm from the AFM measurement and 75%(for the reflectance of the Al mirror) from the optical reflection measurement. However, since the flow of the electrolyte for the Cu substrate was not uniform, bright part was not also uniform, then electroplating in a non-stirring condition was investigated. As a result, it was found that the same effect as stirring of the electrolyte was obtained by using an Al mesh as the counter electrode.

When the cross section of the electroplated specimen was observed by SEM, thickness was thick at the edges and thin at the center of the specimen. Next, in order to reduce a current concentration for the edge part of Cu substrate, support electrodes were placed around the Cu substrate. It was found that the thickness of Al electroplating was almost uniform for the substrate by the supporting electrode. Even if the underlayer was replaced with a Cu-plated ABS resin from the Cu plate, a bright Al-plated film was formed with a uniform thickness. Since the heat resistant temperature of ABS resin is about 100 ^oC, Al electroplating on ABS resin cannot be achieved in molten salt electrolysis (150 ^oC). These results were achieved by using an EmImCl-AlCl₃ ionic liquid.

The electrodeposition of aluminum (Al) has been extensively investigated for a range of emerging applications in electrical, electronic, and automotive industries due to its excellent corrosion and wear tolerance. One highly prospective process is the Al electrodeposition from an electrolytic bath comprised of an air and moisture-stable ionic liquid, such as 1-ethyl-3-methylimidazolium chloride ([EMIm]Cl), and an aluminum precursor, which is normally aluminum chloride (AlCl₃). Recent spectroscopic studies have shown that Al deposition results from electrochemical reduction of chloroaluminate complexes. In general, the chloroacidity determines reactivity and electrochemistry in the ionic liquid electrolyte. However, the Al deposition chemistry has been complicated by the chemical nature of different chloroaluminate complexes species, their chemical equilibria, and interconversion [20, 21]. The [AlCl₄]^-, [Al₂Cl₇]^-, and Cl^- are the main anions in the solution of alkylimidazolium chloride and AlCl₃. More complicated anions such as [Al₃Cl₁₀]^-, [Al₄Cl₁₃]^-, and [Al₂OCl₅]^-, have also been detected when the AlCl₃ concentration is high.

The distribution of the chloroaluminate complexes in the AlCl₃/[EMIm]Cl system is heavily dependent on AlCl₃ concentration and reaction temperatures. When the molar ratio is less than 50%, [AlCl₄]^- is the only chloroaluminate complex present. In contrast, when is greater than 50%, the [AlCl₄]^- concentration decreases and [Al₂Cl₇]^- is formed. When is further increased to 75%, the [Al₃Cl₁₀]^- concentration reaches 48%, accompanied by [Al₂Cl₇]^-, [Al₄Cl₁₃]^- and Al₂Cl₆. Our thermodynamic calculations indicate that increasing temperature mainly affects the distribution of [AlCl₄]^-, [Al₂Cl₇]^- and [Al₄Cl₁₃]^- when is higher than 50%.

As of now, standardized and reproducible procedures have not yet been established due to challenges with the sensitivity of the process chemistries to environments and operation conditions. The properties of Al deposits were affected by many factors such as: the composition of the mixture (AlCl₃-to-IL ratio), substrate-pretreatment, stirring, additives, and reaction temperatures. In order to elucidate how the temperature affects the quality of Al deposits, we have focusedly investigated microstructure changes of Al thin films deposited from an acidic imidazolium-based tetrachloroaluminate bath as a function reaction temperature using several materials characterization techniques. Our presentation will introduce the interesting structural changes of Al films with increasing temperatures, and their correlation with the process chemistry and the kinetics of the electrodeposition.

References:

[1] S.M. Al Farisi, S. Hertel, M. Wiemer, T. Otto, Aluminum Patterned Electroplating from AlCl₃–[EMIm]Cl Ionic Liquid towards Microsystems Application, Micromachines, 9 (2018) 589 (1-14).

[2] G. Franzen, B.P. Gilbert, G. Pelzer, E. DePauw, The anionic structure of room-temperature organic chloroaluminate melts from secondary ion mass spectrometry, Org. Mass Spectrom. 21 (1986) 443–444.

Aluminium is produced by the Hall-Heroult process by electrowinning in a molten fluoride electrolyte with dissolved alumina at ~950 ^oC. Several metallic impurities typically end up in the produced metal due to the fact that most metals are more noble than aluminium. Such co-deposition of metals may be utilized to form aluminium alloys. Another advantage is that many metal oxides are readily soluble in molten cryolite based electrolytes. The more noble metals will deposit at the limiting current density. The alloy composition is essentially determined by the bath concentration of the dissolved cations of the alloying element. Therefore it is possible to produce alloys of certain compositions by controlled addition of metal oxides. Interesting candidate alloying elements are silicon, iron, titanium and manganese but also exotic elements such as scandium are possible. Electrochemical studies were carried out in laboratory experiments and the behaviour of dissolved metal oxides was studied in industrial cells. High current efficiencies for the co-deposition of the alloying elements Si, Fe, Ti and Mn were confirmed. Significant savings for producing alloys by this method are expected.

High purity (>99.9% composition) copper metal specimens were used in electropolishing treatments with an acid-free ionic liquid electrolyte prepared from quaternary ammonium salts as a green polishing solution. A prominent ionic liquid composed of ethylene glycol (HOCH₂CH₂OH) and choline chloride (HOC₂H₄N(CH₃)₃⁺Cl⁻), sometimes abbreviated as 2EG:1VB₄, has been used in other studies to successfully electropolish various metal alloys, but has only been used in limited capacity to polish pure rare earth metals (1-5), like copper, the focus of this study. Voltammetry and chronoamperometry tests were conducted to determine the optimum conditions for electrochemical polishing, while Atomic Force Microscopy (AFM) revealed nanoscale effectiveness of the ionic liquid relative to an industry standard acid polishing treatment of 1 Mol Phosphoric Acid (H₃PO₄). Surface morphology comparisons summarized electrochemical polishing efficiency of these relative treatments by providing root mean square roughness prior to and post-electrochemical polishing.

Electropolishing using the ionic liquid revealed a mirror finish >10 times smoother than the same copper metal surface prior to electropolishing. This transition manifested in a marked change in root mean squared roughness from 167.15 nm to 14.744 nm and resulted in a mass loss of 0.0535 g for an electropolishing rate of 59.444 µg/s when subjected to a 900 second treatment with an average voltage of 1.5 V. By comparison, the phosphoric acid electropolishing treatment yielded an improvement in root mean squared roughness of 85.019 nm, resulted in a mass loss of 0.067 g for a higher electropolishing rate of 74.444 µg/s when subjected to the same treatment conditions. However, the it can be noted that while the elctropolishing rate via the acid solutionis greater, there are extreme peaks and troughs etched on the smoothed surface due to pitting at low current densities (Figure 1). Hydrogen contamination can be observed to affect the overall root mean square average of the region, which may be remedied via a relatively costly degassing treatment at temperatures up to 800 deg C to achieve more comparable results to the ionic liquid solution.

After these electropolishing pretreatments, the resulting copper surfaces were subsequently electroplated during a 1 hour treatment at 2 V with an 2EG:1VB₄ ionic liquid solution that had previously been used to electropolish high-purity niobium metal. A thin film of niobium metal coated the polished copper during electrodeposition resulting in an average mass gain of 0.002 g, and a root mean squared roughness of increase of -2.101 nm for the ionic liquid pretreated sample, compared to the -2.417 nm resulting roughness from the phosphoric acid pretreatment.

As a result, pretreatments revealed a cheaper, recyclable, and environmentally-friendly approach to electopolishing pure copper was demonstrated to be more effective than an acidic industry standard electrolyte. Additionally, electrodeposition of niobium metals on each pretreatment sample provides a reasonably cost-effective method to combine the electroconductivity benefits of copper and ferroelasticity attributes of niobium metals.

References

1. Derek Lofis and Tarek M Abdel-Fattah, International Journal of Minerals, Metallurgy and Materials, 26(5), 649–656 (2019)

2. Derek Lofts and Tarek M. Abdel-Fattah, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 511C, 113-119 (2016)

3. Tarek M. Abdel-Fattah, Derek Loftis and Anil Mahapatro, Nanoscience and Nanotechnology, 5(2), 36-44 (2015)

4. Tarek M. Abdel-Fattah, Derek Loftis and Anil Mahapatro, Journal of Surfaces and Interfaces of Materials, 3, 67-74 (2015)

5. Tarek M. Abdel-Fattah, Derek Loftis and Anil Mahapatro, Journal of Biomedical Nanotechnology, 7(6) 794-800 (2011)

Figure 1

In this study, aluminum is electrodeposited from ionic liquids comprising a melt of 1-butyl-3-methylimidazolium chloride (BMIC) and aluminum chloride (AlCl₃) at the AlCl₃ mole fraction of 0.667 (molar ratio of 1:2). The electrochemical behavior of chloroaluminate species in the ionic liquid is investigated by the chronoamperometry (CA) technique at different temperatures. The diffusion coefficients (D) of Al₂Cl₇^- species in such ionic liquid are calculated at various temperatures. The SEM was used to measure the grain size of the deposits. The calculated average radius of the Al deposit was 6.3 × 10^-6, 5.2 × 10^-6, and 4.3 × 10^-6 m at 363 K, with an applied potential of 1.7, 1.75, and 1.8 V respectively. The calculated grain radius is comparable to the experimental data of electrodeposits. In addition, the rate of nucleation during the electrodeposition is also investigated.

Silver halides, AgX (X = Cl^–, Br^–, and I^–), are known to be typical poorly soluble salts in aqueous solution. AgX is also almost insoluble in an amide-type ionic liquid, 1-butyl-1-methylpyrrolidinium bis(trifluoromethylsulfonly)amide (BMPTFSA). However, AgX has been known to dissolve in the media in the presence of X^– by forming halogenocomplexes, [AgX_n]^(n–1)–. Dissolution of AgCl has been reported in basic chloroaluminate ionic liquids[1] and deep eutectic solvents containing Cl^–. In these reports, however, electrochemical behavior of [AgX_n]^(n–1)– has not been studied in detail. We have already reported that some metal chlorides are soluble in BMPTFSA in the presence of Cl^– and that electrodeposition of the metals is possible[2,3]. In the present study, dissolution of AgX has been investigated in BMPTFSA with addition of BMPX. Furthermore, electrodeposition of Ag and electrochemical preparation of Ag nanoparticles have been examined in the Lewis basic ionic liquids.

AgX was found to dissolve in BMPTFSA containing 0.5 M BMPX. The dissolved species was determined to be [AgX₃]^2– by potentiometric technique. The deposition and dissolution of Ag were observed in cyclic voltammetry. The redox potential of [AgX₃]^2–/Ag was more negative than that of [Ag(TFSA)₃]^–/Ag and dependent on the kind of halide. The diffusion coefficient of [AgX₃]^2– was determined by chlonoamperometry. The initial stage of electrodeposition of Ag on a glassy carbon electrode was regarded as progressive. Potentiostatic cathodic reduction of [AgX₃]^2– resulted in deposition of Ag. The morphology of Ag deposits depended on the reduction potential. Formation of Ag nanoparticles dispersed in the ionic liquid was confirmed after potentiostatic cathodic reduction at a negative potential using transmission electron microscopy[2].

References

[1] T. M. Laher and C. L. Hussey, Inorg. Chem., 22, 1279 (1983).

[2] Y. Katayama, Y. Oshino, N. Ichihashi, N. Tachikawa, K. Yoshii, and K. Toshima, Electrochim. Acta, 183, 37 (2015).

[3] M. Sano, N. Tachikawa, K. Yoshii, N. Serizawa, and Y. Katayama, Electrochemistry, 86, 260 (2018).

1. Introduction

The global production of photovoltaic (PV) cells has increased by a factor of approximately 360, i.e., 0.285 GW year⁻¹ to 102.4 GW year⁻¹, from the year 2000 to 2018 [1,2]. Among the many types of solar cells, crystalline Si solar cell accounted for 96.9% of the worldwide production in 2018 [3]. Therefore, crystalline Si would be most likely to remain as the main stream of the PV industry in the long term.

Due to the high energy cost of the current production process of solar-grade silicon (SOG-Si), the next-generation production process for SOG-Si with low energy cost is required. For the past two decades, we have been studying the direct electrolytic reduction of solid SiO₂ to Si in molten CaCl₂ as a new production process of SOG-Si [4–6]. Recently, we have focused on electrochemical reduction of dissolved SiO₂ in CaO-added molten salts [7].

In this study, we selected eutectic molten NaCl–CaCl₂ as the electrolyte, because its melting point is comparatively lower than that of single CaCl₂ melt, which enables the investigation in a wide temperature range. The structure and electrochemical reduction of dissolved SiO₂ were investigated in molten NaCl–CaCl₂–CaO.

2. Experimental

All experiments were conducted in a dry Ar atmosphere. The molten salts were prepared as follows: Firstly, NaCl and CaCl₂ powders were mixed in a eutectic composition (NaCl:CaCl₂ = 47.9:52.1 mol%), and then certain amounts of CaO (0–2.0 mol%) and SiO₂ (0 or 1.0 mol%) powders were added to the molten eutectic mixture. For Raman spectroscopy, a platinum pan loaded with the mixed salt was placed in an air-tight high-temperature stage, and then heated to 1023 K.Then the structure of dissolved SiO₂ was investigated by Raman spectroscopy. For electrochemical measurement, the mixed salts were loaded into a graphite crucible. A graphite plate was used as a working electrode. The counter and quasi-reference electrodes were a graphite rod and a Si rod, respectively. The potential was calibrated by the deposition potential of Na metal on a Mo wire. Samples obtained by the galvanostatic electrolysis were analyzed by XRD and SEM/EDX.

3. Result and discussion

Fig. 1 shows Raman spectra of molten (a) eutectic NaCl–CaCl₂, (b) NaCl–CaCl₂–2.0 mol% CaO, and (c) NaCl–CaCl₂–2.0 mol% CaO–1.0 mol% SiO₂ at 1023 K. Within the wave number range of 700–1200 cm⁻¹, a band at 845 cm⁻¹ is observed only for molten salt (c), whereas no apparent bands for molten salts (a) and (b). The structure of dissolved SiO₂ in molten salt (c) is considered as SiO₄⁴⁻ ion, considering that the main Raman bands for the stretch vibration of SiO₄⁴⁻, SiO₃²⁻, and Si₂O₅²⁻ in silicate melts have been reported at around 850, 950, and 1100 cm⁻¹, respectively [8]. Since the SiO₄⁴⁻ ion is regarded as the dissolved species of Ca₂SiO₄, this ion is consistent well with the ratio of added CaO/SiO₂ (2.0 mol% and 1.0 mol%).

The galvanostatic electrolyses were also carried out at graphite plates at 50–200 mA cm⁻² in molten NaCl–CaCl₂–1.9 mol% CaO–0.97 mol% SiO₂. From XRD and SEM/EDX analysis of the samples, deposition of Si was confirmed, which suggested the reduction of SiO₄⁴⁻ ion as follows:

SiO₄⁴⁻ + 4 e⁻ → Si(s) + 4 O²⁻ (1)

In the presentation, the effect of CaO concentration on the dissolution behavior of SiO₂ and electrochemical reduction of dissolved SiO₂ will be discussed.

Acknowledgement

The present address of Kouji Yasuda is Graduate School of Engineering, Kyoto University.

References

[1] Photovoltaic Market 2017, RTS Corp., Tokyo, Japan (2017). [in Japanese]

[2] Industrial Rare Metal 2019, Arumu Publ. Co., Tokyo, Japan (2019). [in Japanese]

[3] Photovoltaic Market 2019, RTS Corp., Tokyo, Japan (2019). [in Japanese]

[4] T. Nohira, K. Yasuda, and Y. Ito, Nat. Mater., 2, 397 (2003).

[5] K. Yasuda, T. Nohira, R. Hagiwara, and Y. H. Ogata, Electrochim. Acta, 53, 106 (2007).

[6] T. Toba, K. Yasuda, T. Nohira, X. Yang, R. Hagiwara, K. Ichitsubo, K. Masuda, and T. Homma, Electrochemistry, 81, 559 (2013).

[7] Y. Ma, K. Yasuda, and T. Nohira, Abstracts of The 51st Symposium on Molten Salt Chemistry, p. 24-25, Sapporo, Japan (2019).

[8] B. O. Mysen, Structure and Properties of Silicate Melts, Elsevier, Amsterdam, Netherlands (1988).

Figure 1

The direct electrochemical reduction of titanium oxide to titanium metal in molten CaCl₂ salt has been proven by the FFC Cambridge process. [1] Subsequently, the process has been applied to a number of refractory metals. However, there are limitations to the FFC Cambridge process. For example, the current efficiency of the process is quite low, 10-40%, to achieve sufficiently low oxygen content, 0.3%, in the final titanium product. [2] This could be due to a number of reasons, such as the increasing concentration of O^2- ions in the melt as the reduction process proceeds. The metal oxide electrode has an inherent pore structure and sponge-type substrate electrodes can be used with a range of pore sizes, which has the advantage of high surface area and access of the melt to the oxide. However, when the metal oxide is reduced, oxide ions accumulation in the pores can significantly change the potential needed for the reduction, as shown in the Littlewood [3] predominance diagram. It can also result in the formation of other unwanted metal phases, such as those that include the salt's metal, as it reacts with the oxide ions. Oxide ion build up close to the electrode surface can ultimately bring the reduction process to a halt, leaving the inner parts of the metal oxide unreduced.

1. Chen, G.Z., D.J. Fray, and T.W. Farthing, Direct

electrochemical reduction of titanium dioxide to

titanium in molten calcium chloride. Nature,

2000. 407(6802): p. 361-364.

2. Schwandt, C., D.T.L. Alexander, and D.J. Fray,

The electro-deoxidation of porous titanium

dioxide precursors in molten calcium chloride

under cathodic potential control. Electrochimica

Acta, 2009. 54(14): p. 3819-3829.

3. Littlewood, R., Diagrammatic Representation of

the Thermodynamics of Metal‐Fused Chloride

Systems. Journal of the Electrochemical Society,

1962. 109(6): p. 525.

Our group has focused on the Al electrolysis using room-temperature ionic liquids (RTILs). Al can be deposited from the chloroaluminate ILs, but a practical technology for depositing Al from even the ILs has not been established. There are several problems such as the low limiting current density and the deposition in a dendritic form. In particular, the surface smoothness of electrolytic Al foil is required for practical application. Many articles on the Al electrodeposition using the ILs have ever been reported. However, few articles systematically investigate the correlation between the Al electrodeposits and the operating conditions (parameters). Ueda and co-workers reported that the surface smoothness was improved by adding 1,10-phenanthroline anhydrate (OP) to a Lewis acidic AlCl₃-EMIC (1-ethyl-3-methylimidazolium chloride) melt [1]. In order to effectively scale up from the laboratory level to the practical level, it is desirable to make the above correlation clear sufficiently. In this study, we investigated on the parameters influencing to obtain a smooth electrolytic Al foil.

The chloroaluminate ionic liquids consisting of anhydrous AlCl₃ and 1-ethyl-3-methylimidazolium chloride (EMIC) for 2:1 molar ratio were prepared as an electrolyte in an Ar-filled glove box. 20 mmol dm⁻³ OP (1,10-phenanthroline anhydrate) was added to the electrolyte as an additive. Galvanostatic electrolysis method was carried out in a conventional three-electrode cell with stirring at room temperature (RT) and 50 ℃. A Ti plate was employed as a cathode. The electricity was controlled to 30 C cm⁻². Surface morphology was observed using a field-emission scanning electron microscope (FE-SEM). Arithmetic mean roughness (Sa) was observed using an atomic force microscope (AFM).

The current efficiency of the resulting Al foil at the operating temperature of RT and 50 ^oC was 99.6% in the latter as compared to 84.8% in the former at a current density of 52.6 mA cm⁻². The current efficiencies at the operating temperature of 50 ^oC were higher than those of RT also at other current densities. The Al foil was obtained without the reductive decomposition of EMI⁺ cation (-2.2 V vs. Al / Al (Ⅲ)) even at the high current density (63.2 mA cm⁻²) by increasing the operating temperature to 50 ^oC. Regardless of the addition of OP, the current efficiencies were more than 90% even at the current density of 50 mA cm^-2.

Fig. 1 shows the FE-SEM images of the Al foil obtained at the various operating conditions. The crystal grain shape was the same, like a texture, regardless of the operating conditions. In contrast, the crystal grain size became larger with increasing the operating temperature. Moreover, the addition of OP to the AlCl₃-EMIC melt at 50 ^oC also affected the crystal grain size to be small while the addition of OP did not affect the overpotential for the electrolysis. The AEM images showed that the Sa value became larger with increasing the operating temperature. By adding OP to the melt, the Sa value became extremely smaller, indicating that the addition of OP strongly suppresses the roughness.

These results indicate that the smooth electrolytic Al foil would be obtained even with high operating temperature and high current density by adding OP to the AlCl₃-EMIC melt.

Acknowledgements

This work is based on results obtained from a project commissioned by the New Energy and Industrial Technology Development Organization (NEDO).

References

[1] H. Takahashi, C. Namekata, T. Kikuchi, H. Matsushima, and M. Ueda, J. Surf. Finish. Soc. Jpn., 68, 208 (2017).

Figure 1

1. Introduction

From the perspective of process simplification and secondary waste reduction, the solvent extraction-electrodeposition method is effective in which the ionic liquid (IL) phase containing the extracted complexes is used as an electrodeposition bath and the extracted complexes are electrodeposition to recover the metals efficiently. We have applied the solvent extraction-electrodeposition method to Ru [1], In [2] and Nd [3] in the IL system based on bis(trifluoromethyl)sulfonyl amide (NTf₂^-). In addition, the complexation states between [NTf₂^-] and the metal have been clarified from a computational chemical point of view [4] based on Raman spectroscopy and density functional theory. The electrochemical quartz crystal microbalance method (EQCM) was used to evaluate the electrodeposition behavior of the extracted platinum complex in the IL phase after the extraction by solvent extraction. The purpose of this study is to analyze the electrodeposition behavior of extracted platinum complex, the weight change of the electrode interface, and the alternation of viscosity or density of the IL at the electrode interface.

2. Experimental

The resonance frequency and resistance of an AT-cut Pt-coated quartz oscillator [9 MHz, φ=5.0 mm, Seiko EG&G, QA-A9M-PT(P)] were determined using an electrochemical quartz crystal microbalance (EQCM) system (Seiko EG&G, QCA922) with a well-type cell. The EQCM system was heated with a heating mantle controlled by a thermostat with a PID controller. The applicability of the EQCM technique at elevated temperature was demonstrated in our previous study [5]. Voltammetric measurements were carried out using an electrochemical analyzer (ALS-440A, BAS Inc.) with a Pt-coated quartz oscillator in the EQCM system as a working electrode. The surface area of the QCM crystal was 1.9635×10^-5 m². Two Pt wires with inner diameters of 0.5 mm were used as the counter and quasi-reference electrodes (QRE). The counter electrode was surrounded by a Vycor glass filter at the bottom to prevent diffusion of the decomposition components from the anode into the electrolyte. The Pt QRE showed high stability and good reproducibility of the potential at elevated temperature. The potential was compensated against a Fc/Fc⁺ redox couple. The electrochemical behavior of the extracted platinum complex was investigated at 323 K with a sweep rate of 2.0 mV s^-1.

3. Results and discussion

The result of the CV/EQCM analysis of [R₃NH]₂[PtCl₆]_IL in Alamine336/[P₂₂₂₅][NTf₂] was shown in Fig. 1. Preliminary, in the case of Alamine336/[P₂₂₂₅][NTf₂] system, which did not contain the Pt(IV) extraction complex and consisted of the extractant and the diluent, there were no main peaks other than the solvent decomposition during this potential region. It indicated that Alamine336 extractant did not undergo the electrochemical decomposition. On the other hand, a slight reduction peak was observed at -0.53 V for the extracted Pt(IV) complex, which showed little change with respect to Δm, suggesting that it was corresponded to the charge transfer reaction of Pt(IV)/Pt(II). The main reduction peak was observed at -1.65 V, and simultaneously an increase in Δm and a decrease in Δηρ were also observed in this study. The molecular weight of the electrodeposited species: M_app was corresponded to be 193.7 calculated from Δm, suggesting that the main reduction reaction was the electrodeposition of Pt(II)/Pt(0); the decrease in Δηρ represented a localized viscous change in the IL phase near the electrode interface, indicating the consumption of Pt(II) by the electrodeposition process. The change in Δηρ during the electrodeposition process was consistent with the EQCM analysis of Ag(I)/Ag(0) [6].

References

[1] Y. Song, Y. Tsuchida, M. Matsumiya, K. Tsunashima, Hydrometallurgy, 181 (2018) 164-168.

[2] M. Matsumiya, M. Sumi, Y. Uchino, I. Yanagi, Sep. Purif. Technol., 201 (2018) 25-29.

[3] M. Matsumiya, Y. Kikuchi, T. Yamada, S. Kawakami, Sep. Purif. Technol., 130 (2014) 91-101.

[4] Y. Tsuchida, M. Matsumiya, K. Tsunashima, J. Mol. Liq., 269 (2018) 8-13.

[5] N. Sasaya, M. Matsumiya, S. Murakami, K. Nishihata, K. Tsunashima, Electrochim. Acta, 194 (2016) 304-309.

[6] N. Serizawa, Y. Katayama, T. Miura, J. Electrochem. Soc., 156 (2009) D503-D507.

Figure 1

Understanding the electrochemical deposition of titanium and its alloy from the ionic liquid electrolyte is technologically very essential and necessitates a wide range of attention to gain more insights into their electrochemistry. In this work, the electrochemical behavior of titanium-chloroaluminate anion species in the Lewis acidic alkyl imidazolium-chloroaluminate ionic liquid electrolyte is investigated by electroanalytical techniques such as cyclic voltammetry (CV), chronoamperometry (CA), and chronopotentiometry (CP). The effect of scan rate on peak potential and peak current, constant current density, constant applied potential, and the electrolyte temperature are discussed. It is shown that the redox reaction of the Ti(II)/Ti(0) deposition from titanium hepta-chloroaluminate ([Ti(Al₂Cl₇)₄]²⁻) anions is a quasi-reversible process with a 2-electron transfer reaction while that of Al(III)/Al(0) deposition from hepta-chloroaluminate ([Al₂Cl₇]⁻) anions proceeds through 3-electron transfer. The cyclic voltammograms indicated that the reduction of [Al₂Cl₇]⁻ and [Ti(Al₂Cl₇)₄]²⁻ species to corresponding aluminum and titanium forms is a diffusion-controlled phenomenon. The diffusion coefficients of the titanium hepta-chloroaluminate anion are determined under the studied experimental conditions at different temperatures using CV and CP techniques.

Imidazolium based chloroaluminates were the first widely studied room-temperature ionic liquids. However, work in these systems has waned over the past 25 years in favor of ionic liquids with less reactive anions. Researchers are beginning to rediscover chloroaluminates because they possess some very advantageous properties that hold significant promise for applications such as energy storage, electrodeposition, and chemical synthesis. Chloroaluminates are not simple mixtures of a single cation with a single anion. It is more appropriate to think of these systems as binary mixtures of two ionic liquids which share the same cation. The binary nature of the chloroaluimates leads to highly tunable physical and chemical properties including liquidus range, electrochemical stability window, dc ionic conductivity, zero-shear viscosity, and static dielectric permittivity. However, the binary nature also leads to significant experimental challenges with respect to characterization of the structure and anion speciation in the chloroaluminates. The majority of work characterizing the imidazolium based chloroaluminate was conducted more than 25 years ago and focused on physical property measurements across limited compositions and/or narrow temperature/frequency ranges. The data from these early studies revealed important aspects of chloroaluminate physical properties; however, due to their limited scope they failed to make significant progress in understanding underlying structure-property relationships which are critical to realizing the full potential of chloroaluminate ionic liquids as electrolyte systems. Our current work seeks to reexamine imidazolium chloroaluminate ionic liquids leveraging the improved experimental tools and capabilities available today. In this presentation we will discuss our initial studies of imidazolium chloroaluminate ionic liquids using broadband dielectric spectroscopy, oscillatory shear rheology, and Raman spectroscopy. We will present preliminary work on the impact of changes in cation structure and ionic liquid composition on anion speciation, transport properties, and dynamics.

In this work, the conductivity (k) of Lewis acidic ionic liquid (IL) comprising a melt of organic chloride and aluminum chloride (AlCl₃) with different molar ratios of organic chlorides with AlCl₃ in the temperature range of 70-110 °C are reported. The ILs prepared with three imidazolium-based organic chlorides such as EMIC, BMIC, and HMIC are studied. The molar ratios of 1:1.5, 1:1.8, and 1:2.0 are selected for the study to maintain the acidity of the ILs for possible metal electrodeposition applications. The electrochemical behavior of anion species in the three ILs is studied using electrochemical impedance spectroscopy (EIS) technique and obtained the precise and fast measurement of conductivity values. The measured k values of all the ILs followed the Arrhenius law. The activation energies of conduction determined by fitting the experimental conductivity data are in good agreement with the literature. The conductivity of ILs with lower alkyl group organic chlorides decreased with the higher mole fraction of AlCl₃. In contrast, the conductivity of IL having a higher alkyl group organic chloride (HMIC) increased with the increasing mole fraction of AlCl₃. Based on the thermodynamic calculations, the differences in conductivities are attributed to the anion concentration, the molecular structure, cation-anion interactions, and hydrogen bonds in the ILs.

The research and development of magnesium secondary batteries using magnesium anode have been attracted much attention as one of the prospective post-lithium ion batteries due to the high theoretical capacity (3830 Ah/dm³) and without no resource constraints. However, it has been quite difficult to construct the magnesium batteries because practical electrolytes for the battery has not yet been developed.

We previously reported that TFSA-based ILs were not suitable for a Mg anode even in alkali metal binary molten salt such as MgTFSA₂-KTFSA, which indicated that the existence of the TFSA anion disturbed a magnesium redox on Mg electrodes. [1] On the other hand, relatively smooth magnesium redox could be observed in 1-butyl-3-methylimidazolium chloride melts containing Mg[TFSA]₂. [2] However, the electrochemical window of the imidazolium melt was limited by the reduction decomposition of the imidazolium.

In this study, we reported that ambient temperature molten chlorides could be obtained by mixing a quaternary ammonium chloride and MgCl₂, which did not contain the TFSA anion and also imidazolium cations. The thermal properties and also physicochemical properties were investigated. The electrochemical response of a magnesium metal in the melt were also shown to indicate the possibility as an electrolyte for a magnesium anode battery.

[1] K. Kubota, H. Matsumoto, J. Electrochem. Soc., 167 (2020) 020541.

[2] H. Matsumoto, K. Kubota, ECS Transactions, 86(14) (2018) 21.

Whether used as electrolytes or solvents for synthesis both the bulk viscosity and the microscopic dynamics of ions in ionic liquids often determine the applicability of these. Hence, it is important to understand the connection between the structures of ionic liquid ions and these key properties. Features that contribute to these behaviours include ion mass, size, interactions, and flexibility and the ability of the ions to give rise to nanoscale segregation of charged and nonpolar domains.

In this work, we combine experimental and theoretical studies of a wide range of ionic liquids composed of structurally similar ions in an attempt to separate the relative contributions of these features to the transport properties of ionic liquids. We target a number of design parameters, such as the ions' central atoms (e.g., ammonium or phosphonium cations), ion substitution (e.g., alkyl chain length and degree of substitution of cations and fluorination of the anion).

In addition to the direct effects of structural changes to an individual ion, we show that the significance of these changes is often moderated by the nature of the counter ion of the ionic liquid. We also demonstrate that targeted design of functional groups based on structure-property relations can yield ionic liquids of exceptionally high fluidity.

The flexibility to manipulate various interactions in ionic liquids by changing the cation, anion, or the various functional groups on the ions has drawn researchers to design tailor-made ionic liquids for a given application. Another approach that is gaining importance in tuning the properties of ionic liquids is to form binary, ternary, and reciprocal ionic liquid mixtures. Results from molecular simulations conducted in our research group have demonstrated that the binary ionic liquid mixtures in which two anions are combined in different molar ratios for a given cation can lead to either native or non-native structures in the bulk depending on the difference between the hydrogen bonding ability of the participating anions. The non-native structures differ markedly from those for the pure ionic liquid analogues and impact the physical dissolution mechanism for gases such as CO₂.

Given the importance of ionic liquids as electrolytes in the next-generation of batteries, it is likely that such ionic liquid mixtures will increasingly be investigated for their ability to transport ions to and from charged surfaces. However, very little is currently known regarding how the presence of a surface or an interface affects the local and long-range structures of binary ionic liquids. In this talk, we will present our findings from molecular simulation studies by examining the structural response of binary ionic liquid mixtures comprised of a single cation in combination with two anions when presented with a graphite surface. We will compare and contrast these structures to those obtained when binary ionic liquid mixtures are formed with a single anion and two cations and a graphite surface is introduced. Additionally, we will elucidate the influence of graphite on the transport and electrical properties of binary ionic liquids.

Mixtures of amines with carboxylic acids give not only limited ionicity, but also fascinating profiles of conductivity vs mixing ratio. These acids, when mixed with pyridine or tertiary amines, produce conductivity maxima at 5:1 (not 1:1) mole:mole mixing ratios. Other historical cases of amine/acid mixtures have produced double maxima vs mixing ratio. A good theory to quantitatively explain these is, after over 100 years, still lacking. Here we report on our recent efforts to extend our 2018 pyridine/acetic acid theory to apply to trialkylamine/acid cases. Unlike the pyridine case, where the maxima in volume contraction, viscosity, ion concentration, and conductivity are all at the 5:1 ratio, the trialkylamine cases show more structural complexity; for instance, triethylamine with propionic acid produces these four maxima at four unique ratios (2:1, 3:1, 4:1, and 5:1 respectively). Our extension incorporates some excellent advances put forward by Huyskens in 1980 for these trialkylamine cases.

With applications that include thermal energy storage in Gen3 concentrated solar power (CSP), molten salt batteries in grid scale energy storage, fuel and coolant in molten salt reactors (MSR), and oxide reduction/electrorefining of spent nuclear fuel; molten salts are potentially the most important all around medium for advanced energy systems. In particular, molten halide salts are known to be extremely hygroscopic, typically resisting complete thermal dehydration. Incomplete removal of water leads to hydrolysis; promoting formation of volatile HCl and insoluble oxide/hydroxide impurities in the molten salt. Depending on the application, this can accelerate corrosion, lower product yield through formation of insoluble oxides and oxychlorides, and/or reduce cell efficiency in electrolytic processes. Consequently, the generation, speciation, and electrochemical response of oxide and hydroxide impurities in various molten chloride salts have been widely studied. In this presentation, previously published work with LiCl-Li₂O, MgCl₂-KCl-NaCl, CaCl₂, and CaCl₂-CaO-CeCl₃ molten chloride salts will be reviewed and discussed to reveal underlying commonality and lessons learned. In LiCl-Li₂O and MgCl₂-KCl-NaCl, LiOH and NaOH, respectively, were found in the salt after melting water-contaminated starting material. Electrode response signals were ascribed to both LiOH and NaOH contamination, and a nondestructive, in-situ method for determining their concentrations using cyclic voltammetry (CV) was developed. In CaCl₂, Ca(OH)₂ was found to be unstable, reverting to CaO at operating temperatures. Electrode signals were ascribed to the CaO, and an in situ measurement technique using CV to measure the CaO concentration was developed. The effect of CaO contamination on an electrorefining process was further studied in CaCl₂-CaO-CeCl₃, where CaO presence was shown to promote the formation of both insoluble Ce₂O₃ and CeClO. Electrochemical methods in all cases were further substantiated through support analyses including titration, thermogravimetric analysis, quadrupole mass spectrometry, and inductively coupled plasma mass spectrometry. These quantitative electroanalytical methods are shown to be both effective and versatile, showing promise as the development of real-time, in-line sensors for many molten salt processes continues to be of great consequence. Moreover, their versatility in different molten chloride salt systems is encouraging, as they can become an integral step in the evaluation of new use-cases for other molten halide salts and/or their suitability for new applications.

A fundamental understanding of oxide concentration is critical for operation of advanced nuclear reactors which use molten salts. Molten fluoride salts have been proposed as fuel solvents, coolants, and tritium breeding blankets in fission and fusion nuclear reactors due to their high temperature operation and favorable neutron-interaction properties. An electrochemical sensor which can measure the real-time concentration of oxide anions in the salt melt could be used to identify air ingress to the system. Additionally, similar electrochemical analyses used in the development of a sensor could be used to study the fundamental properties and behavior of oxides in molten fluoride salts. These properties include the activity coefficient, diffusion coefficient, and solubility of oxides. Understanding of these properties could, in turn, inform sensor development.

A further understanding salt chemistry can also help predict and control the effects of oxide concentration on salt corrosion. Material integrity under exposure to high-temperature molten salts is an impactful technical challenge for the design of nuclear reactors that employ molten salts. Oxide anion concentration and complexation of cations and anions in the melt can influence corrosion. Therefore, electrochemical oxide sensors for molten fluoride salts would be instrumental in corrosion studies. Fundamental understanding of oxide chemistry in fluorides would also advance understanding of corrosion mechanisms in fluoride salts.

Cyclic voltammetry, square wave voltammetry, and chronopotentiometry have been previously used to study the electrochemical behavior of oxides in fluoride salt melts. Square wave voltammetry (SWV) is particularly useful in this application because it limits the noise caused by oxygen bubbling from the oxidation of O^2- anions, as found by Massot et al.. Titrations of known additions of oxide to the melt have previously been used to show that the intensity of the SWV oxide oxidation peak is linear with oxide concentration.

Oxides have yet to be studied electrochemically in 2LiF-BeF₂ (FLiBe) or LiF-NaF-KF (FLiNaK) molten salts, which are of interest for nuclear reactors. FLiBe is a likely coolant and solvent candidate for molten salt nuclear reactors. FLiNaK has been studied as a surrogate salt with similar thermo-physical properties to FLiBe. It is easier to handle experimentally since it presents no beryllium health hazard, but its chemical and neutronic behavior are different. FLiNaK is often used in salt loops for reactor design, safety analysis, component testing, and corrosion testing. Therefore, availability of oxide electrochemical sensors for FLiBe and FLiNaK would be instrumental in reactor operation as well as reactor development.

This talk will summarize the application of SWV to quantification of oxides in FLiBe and FLiNaK melts. It will also discuss the available chemical and thermodynamic information for oxides in FLiBe and FLiNaK. Ultimately, an understanding of oxide behavior and a quantification of oxide concentration in molten fluoride salts will aide in the development of an electrochemical sensor for detecting leaks in molten salt reactors.

Molten salts are one of the promising alternative classes of new materials for electrochemical processing. They have high chemical stability and offer much wider electrochemical window than conventional aqueous media. This enables various electrochemical reactions to be conducted without decomposing background media. In addition, the high operation temperature of molten salt boosts the mass transfer of reactant of electrochemical reaction, so that the processing time is shortened. Owing to such advantages, the molten salt media have intensively dealt with the electrowinning of metal and electrochemical separation processes. However, knowing the chemical behavior of electrodes during such processes is crucial to understanding the mechanism of electrochemical reactions that occur during the process. One of the most powerful methods of examining the system is to monitor the visual status of electrodes in real time while the electrochemical process is running. However, due to some practical limitations, it has rarely been applied to the molten salt electrochemical system.

In the present work, the real-time monitoring system for molten LiCl-KCl eutectic salt medium is set up. Instead of an optically opaque ceramic crucible, a cylindrical quartz tube is chosen for the electrochemical cell of our experimental system. With four optically accessible quartz windows mounted on the electrical furnace, a commercially available CMOS camera can be installed externally to monitor the molten salt system without any special modifications.

Real-time monitoring was carried out to investigate the electrode behavior in various electrochemical systems. First, the deposition behavior of cobalt in molten CoCl₂-LiCl-KCl system (1wt% CoCl₂) is compared on a tungsten working electrode. As shown in Figure 1, there is a clear morphological structural difference in the cobalt deposit between the galvanostatic and the potentiostatic electrodeposition. While the metallic cobalt was deposited evenly on the surface of the working electrode by galvanostatic electrodeposition, rapid growth of the dendritic structure was observed using the potentiostatic electrodeposition. This suggests that the structural properties of deposits can vary significantly due to the nature of the electrodeposition methods.

In electrochemical experiments in molten salt system that do not produce solid phase products, it is difficult to elaborately understand the behavior of chemical species during the experiment. However, real-time monitoring enables noteworthy observation from these experiments. During the galvanostatic lithium reduction in the molten LiCl-KCl system, the liquefied metallic lithium formed the oval bulge on the surface of the tungsten working electrode. At the same time, the unknown indigo-colored flow was observed on the surface of the deposited metallic lithium. The clear mechanism remains to be explored, but it appears to be related to the re-oxidization of metallic lithium to Li(I) ions. In the molten ZrCl₄-LiCl-KCl system (1wt% ZrCl₄) with tungsten working electrode, the reduction of Zr(IV) ions to Zr(III) ions during the potentiostatic electroreduction was visualized and identified by the formation of brown-colored flow near the surface of the working electrode. This confirms the result of our previous work and suggests that Zr(IV) is reduced to Zr(III) with a brown color in the first step of CV [1].

Furthermore, with the video signal from the CMOS camera to the PC that controls the potentiostat, the visual changes can be synced with the measured current or potential during the experiment. This enables us to compare the visualized morphological properties on the surface of the working electrode under different applied potential conditions during the CV experiment (Figure 2).

[1] H. L. Cha, J.-I. Yun, Electrochem. Commun. 84, 86 (2017)

Figure 1

Natural cellulosic biopolymer matrices are readily dissolved using ionic liquid (IL) solvents. The dissolved biopolymer can be reconstituted from the ionic liquid using a solvent exchange. Variations in the solvent exchange process have been shown to strongly influence the final morphology in reconstituted cellulose from IL-cellulose solutions, including mesopore formation. We will report on the use of variable solvent exchange to form mesoporous all-cellulose composites. We have characterized these materials using AFM (height and nanomechanical mapping), SEM, XRD, and gas physisorption techniques. We will discuss these results, including the effects of variable IL treatment and solvent exchange parameters on the BET surface area, pore size, and material properties (e.g. modulus and adhesion).

Due to the large volume fraction of nanoscale-reinforcement in cellulose-rich CNC-polymer nanocomposites and the extensive surface area, nearly all the soft/polymeric phase is interfacial (interphase polymer) and therefore, under nanoconfinement. It is well-established that the dynamic properties of any interphase polymer are determined by the structure of the polymer and strength of the interactions with the hard phase, and can deviate significantly from that of the neat polymer. With optimal interfacial interactions, this leads to significantly enhanced performance in many natural composites, such as those found in crustacean and insects shells, and bone. Yet, the role of interfacial interactions, polymer structure, and nano-confinement in helicoidal cellulose nano crystal nanocomposites remains incompletely understood. Here, we use fluorescently labelled helicoidal cellulose nanocomposites and fluorescent lifetime imaging microscopy (FLIM) to study the nature of the effect of polymer structure (cross linked, blend) on the confinement effect in alkyl ammonium modified sulfate-CNC. nanocomposites.

Figure 1

Ionic liquids (ILs) are promising solvents for the treatment and processing of cellulosic biopolymer materials. Despite the discovery of numerous ILs capable of dissolving varying amounts of cellulose, a detailed understanding of the dissolution process in ILs is still lacking. In this study, a new approach, Raman spectral mapping, is employed to investigate the kinetics of cellulose decrystallization and dissolution in a prototypical solvent IL, 1-ethyl-3-methylimidazolium acetate. The cellulose utilized is in the form of mercerized cotton yarns. By monitoring Raman bands characteristic of the crystalline domains, we investigate the degree and spatial extent of crystallinity in IL treated yarns. To elucidate the interplay of kinetic (viscosity) and thermodynamic (solvation) factors in the decrystallization process, the treatment time and temperature were systematically varied. A detailed analysis of the experimental results reveals three distinct temperature-dependent decrystallization regimes including (i) homogeneous decrystallization and cotton yarn collapse, (ii) fast-heterogeneous decrystallization of the collapsed yarn, and (iii) slow heterogeneous decrystallization. These results are discussed within the framework of the current understanding of cellulose and polymer dissolution.

Cellulosic biopolymer materials, such as cotton and linen, have been shown to function well as a support matrix for the synthesis of metallic nanoparticles. Nanoparticles grown within the biopolymer matrix routinely exhibit uniform size (~5 nm diameter) and even distribution for a variety of metallic systems, including Fe-Pd. Our recent work reports the synthesis and characterization of lignocellulose-stabilized superparamagnetic Fe-Pd biocomposites well suited to a variety of applications, including catalysis, micro-robotics, and electromagnetic interference (EMI) shielding. In continuation of this research, we have examined the effects of the type of cellulosic material on the magnetic behavior using vibrating sample magnetometer (VSM) measurements. Additionally, we report the use of ionic liquid treatment to form layered biocomposites, including magnetic behavior characterization.

Molten salts and ionic liquids are a unique class of solvents that offer many advantages, such as negligible vapor pressures, wide liquidus ranges, good thermal stability, and tunable solubilities of both organic and inorganic species. This presentation will provide an overview of my research in the following five areas related to molten salts and ionic liquids: (1) speciation and thermodynamics of actinides and fission products in molten salts, (2) metal-ion extraction based on ionic liquids, (3) gas separation through functionalized ionic liquids, (3) synthesis of functional materials (e.g, carbon, porous materials) using molten salts and ionic liquids as reaction media, (4) electrolytes based on task-specific ionic liquids, and (5) emerging porous ionic liquids. The overall goal of my research is to investigate fundamental principles underpinning ionic media through tailored interactions for energy-related applications.