Corresponding Author

Changbao Zhu
Email:C.Zhu@...

Max Planck Institute for Solid State Research

References

1.
C. Zhu, C.; Mu, X.K.; van Aken, P.A.; Yu, Y; Maier, J.
Single-layered Ultrasmall Nanoplates of MoS2 Embedded in Carbon Nanofibers with Excellent Electrochemical Performance for Lithium and Sodium Storage
2.
Maier, J.
Thermodynamics of Electrochemical Lithium Storage

Department "Physical Chemistry of Solids"

How to get a conversion reaction reversible? Lithium and sodium storage in MoS2 nanodots

Authors

C. Zhu, X. Mu, P.A. van Aken, Y. Yu, and J. Maier

Departments

Physical Chemistry of Solids (Joachim Maier)

It is shown that lithium as well as sodium can be incorporated into MoS2 nanodots with astonishing reversibility. For this purpose ultrasmall, single layers of MoS2 are embedded into carbon fibers via electrospinning. Here the reaction confinement in such a nanodot situation is key for the reversibility. The very same confinement also leads to the substantial lowering of the half-cell voltage which is beneficial in terms of battery function.

Energy storage connected with the possibility of releasing it in the form of electricity is one of the greatest challenges in energy research. Unlike storage exploiting electrical potential differences (capacitors) or gravitational potential differences (hydropower plants), exploiting differences in the chemical potential offers – apart from nuclear reactions – by far the highest specific capacities. Electrochemical storage using synthesis, most promisingly in Li-doped substances, is particularly important as here the transformation from chemical to electrical energy can be highly reversible.

In this context in Li-based batteries, it is sensible to distinguish four storage modes: (i) dissolution into a given phase (single phase storage), examples being LiCoO2 or C, (ii) phase change (two-phase mechanism) transforming a given phase into a Li-richer phase when the solubility limit is exceeded, examples being FePO4/LiFePO4, (iii) decomposition into a multiphase mixture on massive storage, e.g. formation of Li2O:Co or Li2O:Ru on deeply discharging LiCoO2 or RuO2 in a Li-battery, (iv) interfacial storage, which becomes particular relevant if the neighboring phases themselves do not store Li and their grain sizes are very small.

Among these modes, mode (iii), i.e. its decomposition or conversion reaction, is under particularly intense debate as on one hand the theoretical capacity is very high, but on the other hand the kinetic problems associated with it are extraordinarily severe.

Let us consider MoS2 as a relevant example. Reducing it to Mo according to

MoS2+ 4Li → Mo + 2Li2S

involves an introduction of 4 Li per metal, as compared to 1 Li per metal in the case of  FePO4 (two-phase mechanism) and typically much less in a single phase mechanism. This corresponds to a very large theoretical capacity. Unlike a phase transformation where in the ideal case only a single interface is involved moving through a given particle, in the above conversion reaction, the MoS2 phase decomposes into a complex phase mixture of Mo and Li2S.

Severe kinetic problems arise not only on discharging but even more strongly on charging, whereby nucleation and diffusion is necessary to form the electroactive material again. The regeneration will be particularly demanding if the mixture has aged that is if the span between discharging and charging is substantial.

Here we show that problems with conversion reactions can be considerably mitigated – if not removed in the ideal case – by providing the electroactive mass in the form of isolated nanodots. Not only are in such a case the differences between the various storage modes blurred, in the extreme case of lithiating an atomistic cluster rather than a macroscopic phase, nucleation and diffusion problems are getting minimized if not nullified.

Another point that we have to address in this context is the substantial reduction in cell voltage as compared to a macroscopic situation which is very beneficial when referring to negative electrode materials.

Molybdenum disulfide (MoS2) is a layered transition-metal dichalcogenide, which has been widely used as functional material in a large amount of fields such as lubrication, electronic transistors, catalysis, photovoltaics and batteries. Its layered situation is similar to graphite, yet with a different stacking sequence. Recently single-layered MoS2 has attracted great interest due to modified properties when compared to its bulk counterpart. Finding an effective, simple and up-scalable method to prepare single-layered MoS2 or single-layered plates with conducting and easily percolating partners remained a challenge. Moreover, MoS2 is an attractive host for ion intercalation, in particular as negative electrode in Li and Na based batteries. Beyond intercalation, MoS2 can – as mentioned – also be converted into Mo and  by conversion reaction. By constructing different nanostructures, the lithium storage capacity of MoS2 can reach a reversible capacity of 800–900mAh/g at low current rate.

<p class="P1withIndendation"><strong>Fig. 1:</strong> a) Overview TEM-BF micrograph. b) HRTEM image displaying the ultrathin MoS<sub>2</sub> embedded in the carbon nanofiber. c) and d) Corresponding HRTEM images from the marked region in Figure b and c respectively, demonstrating detailed structure of single-layered ultrasmall MoS<sub>2</sub> embedded in the amorphous carbon.</p> Zoom Image

Fig. 1: a) Overview TEM-BF micrograph. b) HRTEM image displaying the ultrathin MoS2 embedded in the carbon nanofiber. c) and d) Corresponding HRTEM images from the marked region in Figure b and c respectively, demonstrating detailed structure of single-layered ultrasmall MoS2 embedded in the amorphous carbon.

[less]

In this work, by pyrolysis of (NH4)2MoS4/PVP nanofibers obtained through an electrospinning process followed by post heat-treatment, for the first time, composites of single-layered MoS2 nanoplates embedded in carbon nanowires were prepared [3]. As shown in Fig. 1(a), the MoS2-carbon fibers are uniform, long, and continuous, with a diameter of around 50nm, connected with each other forming a 3D network. These MoS2 layers are not only ideally thin (0.4nm) but also extremely small (4nm) as far as the lateral dimensions are concerned. These nanodots are randomly embedded in the thin amorphous carbon fibers (Fig. 1(b)–1(d)). The carbon content of this composite is ≈38%wt. This morphology is exciting owing to the extremely short perpendicular distances and the tiny lateral transport length, but also the local electroactive mass will be small and extremely confined.

The electrochemical performance in terms of lithium storage is outstanding as far as rate performance, capacity and cycling behavior is concerned. This is shown in Fig. 2(a). Note in particular that even for as many as 1000 cycles the performance did not worsen. Equally exciting is the fact that these  materials are excellent hosts also for Na as displayed in Fig. 2(b).

<p><strong>Fig. </strong><strong>2:</strong> Left: Excellent electrochemical performance of single-layered MoS<sub>2</sub>-carbon nanofiber composite for lithium batteries. a) Charge and discharge voltage profiles for the first three cycles at 100mA/g. b) Rate performance. c) Cycling performance.</p>
<p>Right:Excellent electrochemical performance of single-layered MoS<sub>2</sub>-carbon nanofiber composite for sodium batteries. a) Charge and discharge voltage profiles for the first three cycles at 100mA/g. b) Rate performance. c) Cycling performance.</p> Zoom Image

Fig. 2: Left: Excellent electrochemical performance of single-layered MoS2-carbon nanofiber composite for lithium batteries. a) Charge and discharge voltage profiles for the first three cycles at 100mA/g. b) Rate performance. c) Cycling performance.

Right:Excellent electrochemical performance of single-layered MoS2-carbon nanofiber composite for sodium batteries. a) Charge and discharge voltage profiles for the first three cycles at 100mA/g. b) Rate performance. c) Cycling performance.

[less]
<strong>Fig. 3:</strong> Carbonfiber containing MoS<sub>2</sub> nanodots, in liquid electrolyte. The differences between the various storage modes are blurred. The confinement of the conversion reaction to the nanoplate renders the conversion reaction reversible and reduces the cell potential of the negative electrode. Zoom Image
Fig. 3: Carbonfiber containing MoS2 nanodots, in liquid electrolyte. The differences between the various storage modes are blurred. The confinement of the conversion reaction to the nanoplate renders the conversion reaction reversible and reduces the cell potential of the negative electrode. [less]

The storage behaviour of such nano-dots is highly exciting from a fundamental point of view. The differences between the usual storage modes – insertion, conversion, interfacial storage – are beneficially blurred for such unique morphology. Moreover, due to confinement in the matrix and the small local electroactive mass, the products of such conversion reaction will not be transported away to form separate metal and sulfide phases and probably even nucleation may not be necessary (Fig. 3).

This will not only lead to higher reversibility but also to lower conversion voltage. In fact the conversion voltage (a realistic equilibrium value is between 0.6 and 1V) is substantially less than the value calculated from bulk thermodynamics (1.5V). Such a shift can be easily understood by the small size of the reaction domains: Let us imagine that unlike the bulk phases, the nano-phases (radius r) Li2S, Mo, MoS2 show higher chemical potentials simply by the capillary term (2γ/r)V where γ is the respective surface tension and V the molar volume. Then the cell voltage will be lowered by [(2γMo/rMo)VMo + 2(2γLi2S/rLi2S)VLi2S - (2γMoS2/rMoS2)VMoS2] / 4F . Assuming as a first estimate a surface tension of ≈1J/m2 and sizes between 0.4nm and 1nm, shifts by several hundreds of millivolts in the observed direction are indeed predicted.

Further assets of the MoS2/C nanostructures shown in Fig. 1 (see also Fig. 3) are fast diffusion of electrons via carbon, fast diffusion of Li+ via electrolyte and carbon, mechanical decoupling in spite of excellent electrochemical coupling thus avoiding cracking and finally the favourable ability of the carbon matrix to function as a binder and particle holder as well as a protector against Ostwald ripening.

All in all, the discussed example highlights the paramount importance of morphology and network structure as far as electrochemical functionality is concerned.


 
loading content