Back to Table of Contents

46864-G7
Templating Ion-Conducting Membranes Using Block Copolymers

Thomas Epps, University of Delaware

The need for clean and sustainable energy sources is growing rapidly, and electrochemical devices such as batteries and fuel cells are considered promising solutions.  Polymer electrolyte membranes for use in lithium batteries and fuel cells require high ionic conductivities to decrease the internal potential losses. Besides ionic conductivities, adequate mechanical strength is necessary to prevent short-circuiting between electrodes and to reduce dendrite formation in lithium batteries.  Additionally, thermal stability, chemical resistance, and controlled permeabilities are extremely important due to the high temperatures and potentially harsh chemical environments inside electrochemical cells.

To achieve the above requirements for lithium ion batteries, block copolymers containing conducting blocks like poly(ethylene oxide), and complementary blocks, have been employed to vary the morphology and properties of conducting materials. The self-assembly of these copolymers permits the design of flexible and sturdy membranes that contain conducting channels for ion transport.  Most copolymer/salt doping has focused on the low salt concentration regime, analogous to ether oxygen to lithium cation ([EO]:[Li]) ratios ranging from 50:1 to 12:1; however, recent literature has shown the promise in using higher salt concentrations.

For our studies to date, we have employed cylinder-forming poly(styrene-b-ethylene oxide) [PS-PEO] block copolymers, where the polymer electrolyte cylinders were formed via a self-assembly process inside a polystyrene matrix.  The three lithium salts used were, LiClO₄, LiCF₃SO₃, and LiAsF₆.  The block copolymers were doped with salts at [EO]:[Li] ratios ranging from 48:1 to 3:1.  We have shown previously that the salt is located solely within the PEO domains of the block copolymers. The nanoscale behavior of these materials was examined through small-angle x-ray scattering (SAXS), transmission electron microscopy (TEM), and differential scanning calorimetry (DSC).

We show that the domain spacing of our polymers increased as the degree of salt-doping within the PEO domains increased.  This trend is shown succinctly in the SAXS profiles in Figure 1, where the primary scattering peak shifts to smaller q with increased salt.  Additionally, the neat (non-doped) material, along with the polymers at salt doping ratios ranging from 48:1 to 6:1, exhibited a hexagonally packed cylindrical structure, while the highest doped (3:1) sample displayed a lamellar microstructure.  From these data and equation 1, we calculated the domain spacing of each sample. The existence of the hexagonally packed cylinder and lamellar nanostructures was corroborated by TEM.

d*=2p/q*                   (equation 1)

The complete temperature dependent behavior of the PS-PEO:LiClO₄ block copolymer complexes is shown in Figure 2.  One particular point of interest is the discontinuity in domain spacing for the 3:1 sample as the temperature is cooled from 147 ^oC to 138 ^oC.  This discontinuity was linked to the crystallization of the 3:1 EO:LiClO₄ complex in a recently published work.  We note that the crystallization occurred above the glass transition temperature of the PS matrix; however, the increased segregation strength upon doping likely promoted the confined crystallization of the PEO domains, leading to the larger domain spacing of the lamellae below the discontinuity.  Similar results were obtained for the other two PS-PEO-salt complexes studied.  

Differences in the domain spacing between copolymers doped lithium salts at similar ratios, but with the various counterions, also were found in this study.  The LiAsF₆-doped polymer exhibited the largest characteristic size, while the LiCF₃SO₃-doped polymer showed the smallest domain spacing.  These differences in domain spacing are likely related to alkali metal – counterion interactions in the presence of the PEO.  In the near future, we will relate these differences to the relative degree of association between the lithium metal and the counterion in the presence of the polymer electrolyte.  Additionally, we found that the addition of salt to the PS-PEO block copolymers led to crystallization of the PEO/salt complexes at temperatures above the glass transition temperature of the PS matrix.  In each case the crystallization remained confined within the nanoscale motif that was established prior to crystallization.  Thus, we can locate and predict order-order transitions that are directly related to crystallization of the PEO-lithium complexes. 

            We will continue to evaluate the influence of the lithium counterions on the thermodynamics of block copolymer self-assembly, to understand and stabilize conducting networks and explore the conductivity of these self-assembled materials.

Figure 1.In-situ SAXS profiles of LiClO₄-doped PS-b-PEO as a function of salt concentration. Samples were held in the SAXS under vacuum and at 166 ^oC for 1 h prior to data acquisition. The diffraction peaks are identified by q/q*, where q* is the primary peak. Curves are shifted vertically for clarity.

Figure 2.(color online) Domain spacing vs temperature calculated from in-situ SAXS data for LiClO₄-doped PS-PEO at [EO]:[Li] ratios from 48:1 to 3:1. Note the discontinuity in domain spacing in the 3:1 sample as the polymer-salt complex passes through an order-order phase transition.

Back to top