AVS2011 Session EN+NS-ThM: Nanostructures for Energy Storage and Fuel Cells I

One of a real-world application of Electrochemical Double Layer Capacitors (EDLCs) is their use in prototype electric vehicles. An EDLC simply consists of two non-reactive, porous electrodes and an electrolyte separated with a membrane. Activated charcoals are one of a commonly used electrode systems which has drawbacks such as limited energy storage due to their large ion size. Therefore, a replacement electrode system with a higher surface area is necessary for a superior EDLC performance. Modified graphene is one of a promising candidate with its high theoretical surface area (2630 m²/g). EDLC working principle also depends on how effectively it can polarize the electrolyte solution. In principle, ionic liquids are promising electrolyte systems with their high non‑volatility, non-flammability, thermal stability and good solvating ability.

This study focuses on the intercalation studies of ionic liquids such as N-methyl-N,N,N-tris(2-hydroxyethyl)ammonium iodide, N-methyl-N,N,N-tris(2-hydroxyethyl)ammoniummethyl sulfate, 1-butyl-3-methylimidazoliumemethylsulfate and N-octyl-N-methylpiperidinium methylsulfate. To achieve a high surface area material with modified graphene, it is essential to understand the interfacial interactions upon intercalation of ionic liquids.

We study powder X-Ray Diffraction Analysis to characterize the interlayer distance of modified graphene. For instance, its d-spacing increases by about 4-12 Å after intercalating as-synthesized graphite oxide (GO, ~9Å) with N-methyl-N,N,N-tris(2-hydroxyethyl)ammonium methylsulfate (IL) at room temperature. Annealing at 300-500°C, (002) orientation of GO disappears as a result of thermal exfoliation. In addition, in-situ Infrared Absorption Spectroscopy (IRAS) measurements were performed to explore the intercalation chemistry of modified graphene in the presence of ionic liquids. Therefore, we perform thermal reduction studies in vacuum coupled with IRAS measurements to characterize the chemical interactions during thermal exfoliation of reduced GO. After annealing GO intercalated with this IL, the loss of C-N and C-O containing species is identified at ~1000-1500 cm^-1 and 800-1200 cm^-1 with contribution from C-OH groups at 3000-3700 cm^-1. Complete removal of these species is observed after a ~500°C anneal resulting in a weak infrared absorbance intensity of sp²-hybridized C=C species at ~1580 cm^-1. The presence of new formation of sheet-to-sheet linking or bonding motifs was also studied with X-ray Photoelectron Spectroscopy (XPS).

*Research supported by the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering under Award DE-SC001951.

In its goal of developing more fuel efficient vehicles, the US Department of Energy in collaboration with the US auto industries are focusing on high-power and high energy lithium-ion batteries to meet the energy storage requirements for HEV and PHEV applications. Under these auspices, Argonne National Laboratory is investigating several different lithium ion chemistries in order to address the calendar life, cost and safety of high power and high energy lithium ion batteries for transportation applications

Silicon nanowires have recently garnered significant attention as a potential candidate to replace graphite as the negative electrode in a lithium-ion battery. Silicon’s earth abundance, extensive knowledge base, and its theoretical capacity of 4200 mAh/g make it an attractive material for this purpose. While the high incorporation of lithium presents a problem as silicon undergoes a significant specific volume expansion upon intercalation (up to 400%), nanowires permit facile radial strain relaxation and allow lithiation without pulverization. In this work, we report on the first systematic study of nanowire length and doping on the cycling performance of NW electrodes. Silicon nanowires were grown on stainless steel substrates via the vapor-liquid-solid technique in a cold-wall low pressure chemical vapor deposition reactor. Growth times ranging from 5-60 minutes and doping concentrations between 10¹⁸ – 10²⁰ atoms cm^-3 were studied. The half cells were cycled against lithium metal between (1) 0.01 and 2.0 V and (2) 0.07-0.7 V at a C/20 rate for 20 cycles. Cycling performance at different depths of discharge was also studied. Nanowires grown at short and long times both exhibit lower capacities than those grown at intermediate times. We attribute this effect to the loss of nanowire contact at the interface between the stainless steel and nanowire array for long nanowires, while the percolation network formed by short nanowires is not sufficiently robust to prevent loss of electrical connectivity upon wire breakage further from this interface. Nanowires at higher doping concentrations were not found to improve cycling performance drastically, which we attribute to the dopant effects on the growth mechanism of the nanowires and placement of dopant atoms in the silicon matrix that may have inhibited lithium atom insertion. We will present novel electrode fabrication routes that overcome these two challenges.

We created the first nano-battery inside a transmission electron microscope (TEM), allowing for real time atomic scale observations of battery charging and discharging processes. Two types of nano battery cells [1], one ionic liquid based, and the other all solid based, were created. The former consists of a single nanowire anode, an ionic liquid (IL) electrolyte and a bulk LiCoO₂ cathode; the latter uses Li₂O as a solid electrolyte and metal Li as anode. Four case studies will be presented: 1) Upon charging of SnO₂ nanowires in an IL cell, a reaction front propagates progressively along the nanowire, causing the nanowire to swell, elongate, and spiral. The reaction front contains high density of dislocations, which are continuously nucleated at the moving front and absorbed from behind. This dislocation cloud indicates large in-plane misfit stresses and is a structural precursor to electrochemically-driven solid-state amorphization. 2) In charging Si nanowires, the nanowires swell rather than elongate. We found the highly anisotropic volume expansion in lithiated Si nanowires, resulting in a dumbbell-shaped cross-section which developed due to plastic flow and necking instability. Driven by progressive charging, the stress concentration at the neck region led to cracking, eventually splitting the single nanowire into sub-wires. 3) Carbon coating not only increases rate performance but also alters the lithiation induced strain of SnO₂ nanowires. The SnO₂ nanowires coated with carbon were charged 10 times faster than the non-coated ones. Intriguingly, the radial expansion of the coated nanowires was completely suppressed, resulting in reduced tensile stress at the reaction front, as evidenced by the lack of formation of dislocations. 4) The lithiation process of individual Si nanoparticles was observed in real time in a TEM. A strong size dependent fracture behavior was discovered, i.e., there exists a critical size with a diameter of ~150 nm, below which the particles neither cracked nor fractured upon lithiation, above which the particles first formed cracks and then fractured due to lithiation induced huge volume expansion. For very large particles with size over 900 nm, electrochemical lithiation induced explosion of Si particles was observed. This strong size-dependent fracture behavior is attributed to the competition between the elastic energy and the surface energy of the nanoparticles. These results highlight the importance of in-situ studies in understanding the fundamental sciences of lithium ion batteries.

1. J.Y. Huang et al., Science 330, 1515-1520 (2010); Nano Lett. (revised); ACS Nano (in press).