UC Berkeley

In the past two decades, lithium-ion batteries have emerged as an increasingly important technology. They are used almost ubiquitously in laptops and cell phones because of their relatively high energy densities when compared to other battery chemistries. More recently, lithium-ion batteries have been employed in the automotive sector in both pure electric vehicles and hybrid electric vehicles. However, one of the major barriers in the widespread adoption of lithium-ion batteries in electric vehicles is the relatively high cost of these batteries. Despite being an inactive component, the battery separator has the highest specific cost of any component in the battery. Battery separators are typically composed of high molecular weight polyolefins (ie. polyethylene, polypropylene) that are made porous in one of two ways. The pores are made by either stretching the films uniaxially to make holes, in the so-called dry method, or by blending with another low molecular weight polyolefin that is extracted later to make holes, in the so-called wet method. Neither of these methods produces holes in an equilibrium fashion. Whereas the cost of raw polypropylene is ~1-2 $ kg^-1, the cost of battery separators is ~120-240 $ kg^-1. This large increase in cost is due primarily to the aforementioned manufacturing and quality control steps. In this thesis, we discuss the use of self-assembling block copolymers to synthesize mesoporous battery separators, which offer the potential to greatly reduce the cost of battery separators.

Our approach for synthesizing mesoporous battery separators utilizes the ability of block copolymers to self-assemble into well-defined morphologies. We blend a polystyrene-block-polyethylene-block-polystyrene (SES) copolymer with homopolymer polystyrene (PS) and then solvent cast nonporous films of these blends. We use a homopolymer selective solvent to remove the homopolymer PS, leaving a mesoporous film. There are several important considerations in this design. The PE block of the copolymer is semi-crystalline, which allows for the retention of a non-collapsing pore phase after homopolymer extraction. Ideally, the homopolymer PS segregates preferentially to the PS phase of the SES copolymer during the blending step. The ability of the homopolymer PS to segregate to this phase is largely dictated by the size of the homopolymer PS (N_PS) relative to the size of the PS chains in the SES copolymer (N_PS,BCP). We define the parameter &alpha as equal to this ratio: &alpha = N_PS/N_PS,BCP . We found that intermediate values of &alpha, between 0.2 and 0.5, allowed the homopolymer PS to reside primarily within the PS domain, thus taking advantage of block copolymer self-assembly and enabling the formation of a continuous porous phase after homopolymer extraction. The morphological characterization of these mesoporous block copolymer films is non-trivial, and we employ a variety of techniques to study them.

This thesis is organized into 5 chapters. Chapter 1 describes the synthesis and preparation of the mesoporous block copolymer membranes, in addition to some of the characterization techniques used in this thesis. In Chapter 2, we discuss the relationship between the conductivity (in a liquid electrolyte) and morphology of the SES separators. We focus on the importance of on the overall porosity of the structure. As traditional block-copolymer characterizations techniques proved inadequate to properly describe the morphology, in Chapter 3, we discuss the use of resonant soft x-ray scattering (RSoXS) to fully characterize the mesoporous block copolymer films. By tuning the incident x-ray energy, we were able to contrast match the three phases of our system (voids, polystyrene, and polyethylene). In Chapter 4, we study the impact of the washing step on the final film properties. Normally, the SES/PS cast films are immersed first in tetrahydrofuran (THF) to remove the homopolymer PS, and then in methanol (MeOH). Both THF and MeOH are nonsolvents for crystalline PE, but THF is a good solvent for both amorphous PE and PS, whereas MeOH is a poor solvent for all phases. Despite the fact that films were exposed to both THF and MeOH, the final solvent treatment (THF or MeOH) has a dramatic effect on both the conductivity and morphology of the film. In Chapter 5, we study the effect of changing the length of the PE block of the SES copolymer while maintaining the length of the PS block. We also benchmark our SES separators using tensile strength experiments and as full cell batteries against a commercial battery separator.