Tuning Filler Shape, Surface Chemistry and Ion Content in Nano-filled Polymer Electrolytes.

Open Access
- Author:
- Ganapatibhotla, Lalitha
- Graduate Program:
- Chemical Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 16, 2014
- Committee Members:
- Janna Kay Maranas, Dissertation Advisor/Co-Advisor
Janna Kay Maranas, Committee Chair/Co-Chair
Scott Thomas Milner, Committee Member
Jeffrey M Catchmark, Committee Member
Enrique Daniel Gomez, Committee Member
Michael John Janik, Committee Member - Keywords:
- polymer electrolytes
lithium ion batteries
nanoparticles
nanowhiskers
polymer dynamics
QENS
reflectometry - Abstract:
- We investigate how nanofiller surface chemistry and aspect ratio affect the performance of nanofilled solid polymer electrolytes. Polymer-based electrolytes are an attractive alternative to the flammable, volatile, and toxic organic electrolytes currently used in lithium ion batteries. Despite their advantages, polymer electrolytes suffer from low room temperature ionic conductivities (10-8 to 10-6 S/cm, target for device applications ~ 10-3 -10-2 S/cm) and low modulus (1-10 MPa, target to prevent dendrite formation ~ 6 GPa). Spherical ceramic fillers are known to improve conductivity and mechanical properties of polymer electrolytes. However the mechanism by which nanoparticles enhance conductivity is not clearly understood. If we can understand this mechanism, we can tailor the filler-electrolyte design to obtain highly conducting matrices. To understand this mechanism, we characterize acidic nanoparticle filled electrolytes and compare them to neutral particle-filled electrolytes previously measured in our lab. Dielectric spectroscopy measurements indicate that the highest increase in conductivity occurs at the eutectic composition (EO/Li=10) and is independent of filler surface chemistry. Because lithium ion conduction depends on polymer segmental dynamics, we measure PEO dynamics using quasi-elastic neutron scattering and do not observe any change in polymer dynamics with particle surface chemistry. When we examine the elastic incoherent structure factor associated with the rotational process, fillers are found to restrict the rotation of the highly conducting PEO6:LiClO4 tunnels. At the eutectic composition, these tunnels are stabilized at the filler surface even above PEO melting temperature. Marginal stability theory predicts formation of alternating layers of coexisting phases at the eutectic composition. To explain these trends, we propose a new mechanism, via stabilization of alternating layers of PEO and highly conducting PEO6:LiClO4 tunnels at the filler surface. When compared to spherical particles, more such structures would be stabilized at a filler surface with high aspect ratio. Consistent with this hypothesis, neutral gamma-Al2O3 nanowhiskers (2-4 nm in diameter and 200-400 nm in length) intensify the effect of neutral gamma-Al2O3 nanoparticles. The diameters of the two fillers are similar, but the change in aspect ratio (1 to 100) improves conductivity by a factor of 5. Interestingly, this enhancement occurs at battery operation temperatures! Although the change in aspect ratio does not affect thermal transitions and segmental dynamics at optimal whisker loading, the rotation of PEO6 remnants is distinct at the eutectic composition. Rotation geometry is described by a combination of two EISF models- one that describes more restricted layers immediately next to the whisker surface and the second pertains to the subsequent less-restricted layers. Thus multiple layers are stabilized at the longer whisker surface and provide larger increase in conductivity than the particles. Because the mechanism by which nanofillers enhance conductivity is related to stabilization of conducting structures at the filler-electrolyte interface, we determine the interface morphology using neutron reflectometry. For this, we spin-coat the unfilled electrolytes EO/Li = 8, 10 on sapphire substrate, which has the same surface chemistry as -Al2O3. When freshly-spin coated on sapphire substrate, the non-eutectic sample does not exhibit any segregation of layers. Interestingly, the freshly spin-coated eutectic sample forms layers with alternating high and low salt concentrations, very similar to the eutectic lamellae predicted by the marginal stability theory for eutectic solidification. Surprisingly, such lamellae do not develop further when the sample is annealed at eutectic temperature and the salt concentration in the polymer decreases gradually away from the surface of sapphire. To take fullest advantage of the surface mechanism and obtain larger increases in conductivity we tailor the aspect ratio of high aspect ratio fillers. Commercial availability of alumina nanowhiskers is limited to neutral surface chemistry and aspect ratio of 100, cellulose nanowhiskers provide a model system where a wide range of surface chemistries may be accessed with variable aspect ratio. We synthesized cellulose whiskers of two different aspect ratios [cotton whiskers: aspect ratio ̴ 10, acetobacter whiskers: aspect ratio ̴ 200] and tested their influence on conductivity and morphology of polymer electrolytes. Similar to all fillers studied in this work, both types of cellulose whiskers provide highest increase in conductivity at the eutectic composition, with the longer acetobacter whiskers providing a marginally higher increase than the cotton whiskers. Although both cellulose whiskers do not alter the crystallinity or glass transition temperature at the optimal 1 wt% loading, they amplify the faint cold crystallization behavior observed in the unfilled eutectic electrolyte without changing the overall crystallinity. At the non-eutectic compositions, cellulose whiskers behave similar to the acidic nanoparticles. To determine the function of nanofillers in entire composition range of the phase diagram, we extend the range of measurements on the nanofilled PEO+LiClO4 electrolyte to EO/Li = 4 to 100. Because PEO+LiAsF6 electrolytes have similar phase diagram as the PEO+LiClO4 electrolytes, we augment the study of nanofilled PEO+LiAsF6 complexes to the PEO+LiClO4 electrolytes. At compositions near the high and low ends of the phase diagram, the effect of nanofillers on conductivity is governed by reduction in crystallinity of PEO and PEO-salt complexes. In the absence of PEO6, fillers interact directly with PEO and suppress crystallization. This is consistent with our findings from reflectometry experiment where sapphire surface prefers to interact with the salt-rich layers. Around the eutectic composition fillers restrict the highly conducting PEO6 complex at their surface and any increase in conductivity is due to stabilization of these conducting tunnels. For room temperature applications, lithium hexafluoroarsenate seems to be the better salt than lithium perchlorate. At temperatures higher than the eutectic temperature (50 ℃), conductivity levels off at the value set by the eutectic composition. To sum up, we suggest a mechanism by which nanoparticles increase conductivity of polymer electrolytes in a range of compositions, and extend this mechanism to obtain high room temperature conductivities via high aspect ratio fillers.