1
Application of Mössbauer Spectroscopy to Energy Materials

Pierre-Emmanuel Lippens Jean-Claude Jumas and Josette Olivier-Fourcade

Institut Charles Gerhardt de Montpellier, UMR 5253 CNRS, Université de Montpellier, ENSCM, Pôle Chimie Balard Recherche 1919 route de Mende, Montpellier Cedex 5, France

1.1 Introduction

The increasing demand for energy and the environmental problems caused by fossil fuels make necessary the fast development of renewable and clean energy sources such as solar or wind. However, their intermittent nature requires highly efficient energy storage systems. Furthermore, clean transportation, such as electric or hybrid vehicles, needs power sources of high energy density. In this regard, lithium-ion batteries can be considered as one of the most promising technologies. Although lithium-ion batteries are used in electronic devices, power tools, electric bikes, and more, there are still many challenges to overcome for new applications that require higher energy and power densities, improved safety, and lower cost. This explains the current intensive research on electrode materials [1, 2].

The performance of currently used carbon and layered lithium transition metal oxides as negative and positive electrode materials, respectively, has almost reached the upper theoretical limits. New negative electrode materials such as silicon and tin have high theoretical capacities, but they suffer from strong volume variations during cycling, which leads to capacity fading and strongly reduces the cycle life [3, 4]. Among the different approaches proposed to improve the performance of such materials, downsizing the particle size or dispersing particles within an electrochemically inactive matrix have been proposed [5, 6]. For the positive electrodes, new Fe-containing materials such as LiFePO4, LiFe1-xMnxPO4, or LiFePO4F appear to be very promising for environmental and economic aspects [7, 8]. Finally, the potential problems of cobalt and lithium supply in a near future have recently led to the development of alternative technologies such as Na-ion batteries [9, 10].

Improving the performance of electrode materials requires a good knowledge of the electrochemical reaction mechanisms in batteries. Mössbauer spectroscopy is often used for the characterization of materials, but is also a unique tool to follow such reactions at the atomic scale from the analysis of the Mössbauer parameters: isomer shift (given in this chapter relative to a-Fe and BaSnO3 for 57Fe and 119Sn isotopes, respectively), quadrupole splitting and hyperfine magnetic field [11, 12]. Different examples are considered to illustrate the application of Mössbauer spectroscopy to electrode materials for Li-ion and Na-ion batteries. This includes tin-based negative electrode materials: ß-Sn, tin oxides, tin borophosphates, tin intermetallics, and tin-silicon composites, as well as iron-based positive electrode materials: LiFePO4, LiFe1-xMnxPO4, Fe1.19PO4(OH)0.57(H2O)0.43, and Na1.5Fe0.5Ti1.5(PO4)3.

Reducing emission and pollution of existing transportation based on fossil fuels is also a great challenge. Reforming catalysis is a major petroleum refining process for the production of hydrogen or high-octane gasoline. In particular, catalytic reforming is a chemical process used to convert naphtha, produced during petroleum refining, into high-octane number gasoline. If Pt/Al2O3 was the first naphtha-reforming catalyst, a great progress has been achieved with supported bimetallic-reforming catalysts, in which Pt is promoted by another metal such as Sn, that offer high selectivity at low pressure [13]. Improved selectivity can be obtained by the addition of a second promoter as In [14]. Mössbauer spectroscopy has been widely used in catalysis [15, 16]. Some examples are presented here to show how complex redox processes in Sn-Pt based catalysts involving many different tin species can be elucidated.

1.2 Mössbauer Spectroscopy for Li-ion and Na-ion Batteries

1.2.1 Characterization of Electrode Materials and Electrochemical Reactions

The main specifications of electrode materials for Li-ion or Na-ion batteries are specific and volumetric capacities, nominal potential, cycle life, calendar life, rate capability, safety, environmental impact, and recycling. It is of course impossible to optimize all these features at the same time. For example, layered lithium transition metal oxides, as commonly used positive electrode materials for Li-ion batteries, have high capacity and high potential vs. Li+/Li. Lithium iron phosphate (LiFePO4) has lower capacity but provides better safety and higher electric power. Thus, new electrode materials are needed to improve the performance of Li-ion and Na-ion batteries. This requires a better knowledge of the electrochemical reaction mechanisms that take place during the charge-discharge cycles.

Different techniques have been used for the characterization of electrode materials and to follow the reactions within the batteries. This includes ex situ experiments, i.e. the electrode material is extracted from the battery, in situ experiments, i.e. the electrode material is in the battery, and operando experiments as electrochemical reactions proceed [17]. X-ray diffraction (XRD) and Mössbauer spectroscopy have often been combined to obtain complementary information about long range and local properties, respectively [18]. It is thus possible to determine the formed species even if they are amorphous, to follow changes in local structure or oxidation state in order to elucidate the reaction mechanisms.

In situ and operando Mössbauer and XRD measurements require specific electrochemical cells that allow transmission of ?-rays and reflection of X-rays, respectively, while voltage or electric current is imposed. Such a cell is based on the usual positive electrode/separator/electrolyte/negative electrode configuration [19, 20]. To investigate the electrochemical reactions for a given electrode material, a half-cell is commonly used, where the negative electrode is metallic lithium (or sodium) and the positive electrode is the formulated electrode material under investigation. The formulation consists of mixing the electrochemically active material (nanoparticles or microparticles) with an electronic conductive additive, like carbon black, and a binder to form a slurry that is casted onto the metal current collector [21]. In the case of an electrochemical half-cell, the discharge corresponds to the lithiation (or sodiation) of the electrode material under investigation and the charge to delithiation (or desodiation).

1.2.2 Tin-Based Negative Electrode Materials for Li-ion Batteries

1.2.2.1 Electrochemical Reactions of Lithium with Tin

Metallic and semi-metallic elements can be used as negative electrode materials for high energy Li-ion batteries. For example, each Sn atom in ß-Sn can react with a maximum of 4.4 Li, which corresponds to specific and volumetric capacities of 992 mAh·g-1 and 2111 mAh·cm-3, respectively [3]. This is more than 2.6 times the capacity of currently used graphite (372 mAh·g-1 and 719 mAh·cm-3). In addition, the average potential of ß-Sn in a Li half-cell is higher than that of carbon, which reduces lithium plating and improves safety. Unfortunately, the formation of Lix < 4.4Sn phases during lithiation induces a strong increase of the particle volume (>300%) that causes cracks, while delithiation leads to the pulverization of the particles. These two effects are responsible for mechanical and electrical instabilities of the negative electrode film, which drastically reduces the cycle life of the battery. XRD and 119Sn Mössbauer spectroscopy have been combined to obtain deeper insights into Li-Sn alloying reactions.

The experimental voltage curves of ß-Sn in a Li half-cell, obtained during the first cycle in galvanostatic regime at low current density, show different well-defined plateaus that can be attributed to two-phase reactions (Figure 1.1) [22].

The voltage plateaus are observed for both lithiation and delithiation processes, showing the reversibility of the mechanism. The following alloying reactions have been proposed by considering the different crystalline phases of the commonly accepted Li-Sn phase diagram [24]:

First-principles calculations of the cell voltage were performed with different methods based on density functional theory (DFT) for reactions (1.1)-(1.6) [22, 23]. The comparison between the experimental and theoretical voltage profiles suggests that the two first plateaus at average experimental voltages of 0.75 V and 0.65 V reflect the formation of Li2Sn5 (reaction 1.1)...