Novel anhydrous organic proton conductors for fuel cells

In recent years, fuel cells have attracted much attention because they directly, cleanly, and efficienctly convert chemical into electrical energy. An essential component of a fuel cell is the electrolyte, which conducts ions. Polymer-electrolyte membranes have been widely used in fuel cells.¹ Liquidlike regions of water between polymer chains can act as proton carriers. Protons can jump between the water molecules but mainly ride on them as H₃O⁺ oxonium ions. However, the high water content required for such a system limits the fuel cells' maximum operating temperature to 100^°C. Maintaining a high water content also requires a heavy humidifier and water tank. If higher operating temperatures could be achieved, the system would provide higher energy efficiency and enable simplified heat management and the use of the heat generated (as well as the electricity). A fuel cell that required less water could also be a smaller, lighter source of mobile power. Much recent research has focused on developing a solid electrolyte that operates in the intermediate temperature range (100 – 200^°C). Organic proton conductors lacking crystallized water (anhydrous) are strong candidates.

Among inorganic compounds, a few solid acidic salts — such as potassium and cesium hydrogen sulfate (KHSO₄ and CsHSO₄, respectively) — are known to be good proton conductors.^2,3 They usually consist of oxyanions, such as sulfate (SO₄²⁻), linked by hydrogen bonds. CsHSO₄exhibits a rather drastic phase transition at about 140^°C, during which the proton conductivity increases by almost four orders of magnitude.⁴ At this higher temperature, the structure of CsHSO₄ loosens to allow SO₄²⁻ ion rotation so that a proton on one of the oxygen ions can approach an oxygen ion of a neighboring sulfate ion, thus facilitating proton transfer along hydrogen-bonded SO₄²⁻ chains. This suggests that a key requirement for effective proton conduction is a 1D hydrogen-bonded structure.⁵

Figure 1. Schematic representation of columnar arrangement in the LC solid and fluid phases of A and B. (a) Solid phase of A and B and low-temperature fluid phase of B, (b) high-temperature fluid phase of B. The arrows aand b represent the fundamental unit vectors of a lattice unit cell (i.e., the lattice constants).

It is often difficult to control the manner and degree of intermolecular hydrogen bonding of conventional organic compounds because they are usually crystalline and amorphous substances. In contrast, liquid-crystalline (LC) materials assemble themselves to form various phases. The assembled structures are directed by the choice of phase so that structurally controlled materials could be prepared by quenching from the appropriate fluid phase to the solid state.^6,7 LCs are among the most promising compounds for applications in proton-conducting materials. To develop novel fuel-cell electrolytes showing protonic conductivity even under dry conditions, we have designed a new family of thermotropic LCs made of simple aliphatic compounds exhibiting a columnar phase.

We synthesized 1,3-dioctadecylurea [CO(NHOct)₂] (compound A) and the corresponding rare-earth metal complex [CO(NHOct)₂]_0.57[TbCl₃]_0.43 (compound B). Although A — which lacks metal ions — showed only a melting point, the terbium complex (B) exhibited an LC phase across the wide temperature range of 109 – 179^°C. Compound B also exhibits enantiotropic polymorphism in the LC fluid phases, i.e., two different phase structures in the LC temperature range.

Temperature-dependent x-ray-diffraction studies reveal a more detailed phase structure (see Figure 1). The solid phase of the parent compound (A) is columnar with a 2D rectangular lattice with lattice constants a=8.92 and b=6.29nm. This implies that compound A shows an LC rather than a crystalline solid phase. We obtained similar results for compound B, the terbium complex. Both the solid and low-temperature LC fluid phases are rectangular columnar. In the high-temperature LC fluid phase we observe a hexagonal columnar structure with lattice parameter a=3.33nm. Additionally, the regular slightly distorted hexatic array of uniform channels was observed directly by transmission-electron microscopy. For the solid sample the intercolumnar distance is roughly 5nm. Optical rotatory measurements revealed that each column is made up of helical molecular chains based on the formation of 1D hydrogen bonds with the urea moieties, i.e., the low-mass LC molecules form a noncovalently linked polymer (a pseudopolyurea) with a helical structure.

We used electrochemical-impedance spectroscopy to perform a preliminary evaluation of the ionic conductivity of these molecular chains. The Cole-Cole plots obtained for the urea derivatives showed semicircles in the high-frequency region, which relate the movements of ions in the bulk electrolyte. The ionic conductivity increased with temperature and reached of order 10⁻⁵S/cm in the high-temperature region. The terbium complex exhibited ionic conductivity of ~10⁻⁶S/cm even in the dry solid state, retaining a columnar structure. We assume that this is because protonic diffusion proceeds in two steps: transfer in a hydrogen bond followed by a jump to another hydrogen bond along the 1D molecular-chain axis. The proton-conducting mechanism of the urea-type LC materials may depend on proton transport through a conductive pathway based on amide-imide acid tautomerism (see Figure 2).

Figure 2. Schematic illustration of the conductive pathway of protons based on 1D hydrogen-bonded molecular chains.

We have designed and synthesized 1D nanomaterials on the basis of urea-based columnar LCs. We hope that this novel class of anhydrous organic proton conductors will permit development of high-operating-temperature fuel cells. Applications of 1D nanomaterials are currently leading to the development of small and lightweight fuel cells — which need neither humidifier nor water tank to maintain a high humidity – as a power source for vehicles and mobile devices.

Grateful acknowledgement is made of the Toyota Motor Company's support of this work, which was funded in 2006 by the Industrial Technology Research Grant Program of the New Energy and Industrial Technology Development Organization of Japan.