Mario Wachtler

Magnesium batteries are currently attracting a lot of interest as a next generation battery technology. One critical issue is to find a suitable electrolyte and herein we explore an electrolyte based on magnesocene (MgCp₂) in tetrahydrofuran (THF), aiming for low‐voltage Mg batteries, with respect to: Mg plating characteristics, electrochemical stability windows, electrolyte speciation, and electrolyte decomposition reactions; both experimentally and computationally. Overall, the electrolyte does not seem to decompose on a Mg metal anode and most likely reduced solvation of Mg²⁺ by the Cp⁻ anion is important and species such as MgCp₂THF₂ may play an important role for Mg plating with small overpotential. The oxidation limit is largely determined by the Cp⁻ anion and density functional theory predicted oxidation reactions point to polymerized end‐products to be possible. Furthermore, in silico substitution studies enable us to establish the prospects of some Cp⁻ anion derivatives to further improve the oxidative stability, but still the Mg²⁺ solvation must be monitored for ease of reduction and Mg plating.

As part of a systematic study on the formation and composition of the solid electrolyte interphase (SEI) in lithium‐ion batteries (LIBs), going stepwise from highly idealized electrodes such as highly oriented pyrolytic graphite and conditions such as ultrahigh vacuum conditions to more realistic materials and reaction conditions, we investigated the decomposition of simplified electrolytes (ethylene carbonate (EC)+1 M LiPF₆ and dimethyl carbonate (DMC)+1 M LiPF₆) at binder‐free graphite powder model electrodes. The results obtained from cyclic voltammetry and ex situ X‐ray photoelectron spectroscopy half‐cell measurements – in particular on the effect of cycling rate, solvent and electrode – are explained in terms of a mechanistic model where electrolyte decomposition occurs at the SEI | electrode interface and where transport of solvent and salt species through the growing SEI plays an important role for explaining the observed change from preferential salt decomposition to solvent decomposition with increasing cycling rate.

Metal-oxygen batteries employing non-aqueous electrolytes are a promising alternative to conventional lithium-ion batteries due to their high specific energies. One of the most appealing metals in such an application is magnesium, considering its price and its volumetric charge density. While previous reports already indicated the formation of a blocking layer on the oxygen-electrode surface during oxygen reduction, associated with a low reversibility, the reasons for this behavior are not yet understood. This work focuses on elucidating the composition as well as the electrochemical behavior of this blocking layer employing X-ray photoelectron spectroscopy, infrared spectroscopy and measurements with a rotating ring-disc electrode setup. The electrochemical measurements indicate that the blocking layer exhibits an effect similar to the valve effect, which is known from the oxidation of aluminum for example. Combining potentiodynamic and –static measurements, it could be shown that there is a current flow at cathodic potentials due to migration of ions within the layer. A further outcome of the electrochemical measurements is that the anodization of this surface layer proceeds in a self-accelerating manner. The spectroscopic analysis of the surface layer suggests the formation of magnesium peroxide on the surface as well as the presence of some sulfur and chloride containing compounds which must arise from the decomposition of the electrolyte.

Unlike ferrocene, bis(η⁵‐cyclopentadienyl)magnesium (magnesocene, MgCp₂) is slightly dissociated in solvents, such as ethers, resulting in electrolyte solutions with low conductivity. MgCp₂/tetrahydrofuran solutions make possible reversible magnesium plating and stripping with low over‐potentials for many cycles. The Mg deposits appear with a cauliflower‐like morphology. IR and NMR spectroscopy confirm that the electrolyte is stable and not decomposed during prolonged cycling. The anodic stability limit is in the range of 1.5 V (at platinum) and 1.8 V versus Mg/Mg²⁺ (at stainless steel), which may be sufficient for low‐voltage cathode materials. MgCp₂ is a first example of a completely new class of halide‐free electrolytes, which may open up a new research direction for future magnesium metal and magnesium‐ion batteries.

Unlike ferrocene, bis(η⁵‐cyclopentadienyl)magnesium (magnesocene, MgCp₂) is slightly dissociated in solvents, such as ethers, resulting in electrolyte solutions with low conductivity. MgCp₂/tetrahydrofuran solutions make possible reversible magnesium plating and stripping with low over‐potentials for many cycles. The Mg deposits appear with a cauliflower‐like morphology. IR and NMR spectroscopy confirm that the electrolyte is stable and not decomposed during prolonged cycling. The anodic stability limit is in the range of 1.5 V (at platinum) and 1.8 V versus Mg/Mg²⁺ (at stainless steel), which may be sufficient for low‐voltage cathode materials. MgCp₂ is a first example of a completely new class of halide‐free electrolytes, which may open up a new research direction for future magnesium metal and magnesium‐ion batteries.

The flammability of electrolytes is an important aspect of the thermal safety behavior of Li-ion batteries. Flash points (FPs) and self-extinguishing times (SETs) of 25 solvents (including carbonates, ethers, esters, lactones, dinitriles, a sulfone, and others), 3 solvent mixtures, and 15 electrolytes are presented. The FPs have been measured according to the Abel and Pensky-Martens closed-cup methods using 12 mL of sample and electric ignition. The SETs have been determined with the pure liquids, without any carrier substrates. A correlation of the FPs with the SETs, as well as with vapor pressures and boiling points is attempted. Furthermore, the effect of the addition of two non-flammable solvents [1-butyl-1-methylpyrrolidinium bis(trifluoromethyl-sulfonyl)imide and methyl nonafluorobutyl ether] and a flame-retardant additive (trimethyl phosphate) on both the SET and the FP of a carbonate-based electrolyte is discussed. As an alternative to their experimental determination, the FPs of the pure solvents have been calculated from other physical and chemical properties of the substance, and the FPs of the solvent mixtures and electrolytes from the FPs of their flammable constituents. Some of the models predict the FPs with an accuracy of ± 10°C, which may be sufficient to estimate the flammability hazards for many applications.