Consortium Research Program Status

Program Status

A key objective of the research program is to develop, understand and assess the performance of fuel cell structures featuring nanostructured catalytic electrodes and a proton conducting dense ceramic membrane that is impermeable to methanol or water. At the end of the project, it is expected that these fuel cell structures will be able to provide about 1 W of power and reach power densities in excess of 100 mW/cm2.

Motivation for our program are several fold, namely, to address the need for reliable, high energy power sources for portable electronics, to understand and predict how new behavior and properties evolve as dimensionality moves into the nanoscale regime, and to develop appropriate synthesis tools for controlled fabrication of nanostructures.

This program addresses most of these issues in a single platform that involves an all solid-state ultra thin fuel cell architecture fabricated by using well-established semiconductor micro-fabrication techniques for miniaturization.

A key feature of our DMFC structure is the dense, anhydrous proton conducting solid oxide electrolyte, such as yttria doped barium zirconate (BYZ) deposited as a thin layer of only 50-100 nm. This reduces the area specific resistance (ASR) of the proton conducting membrane down to about 0.5 ohm.cm2 at 60oC, which compares well with the ASR value for Nafion-based membranes currently employed for PEMFCs and DMFCs. The proton conducting solid electrolyte also serves as an impervious diffusion barrier for methanol and water molecules, completely eliminating chemical shorting and the need for water management.

Thin film anodes are made of nanoporous catalytic metals such as Pd, Ru, Ni, Pt, and their alloys. We plan to explore and study other nanoporous catalytic anodes such as WO3 and RuO2 as well as doped and/or decorated zeolite structures which are known to be active catalysts for methanol oxidation. In some cases, the anode current collector will also serve as a chemical barrier layer for the proton conducting membrane. At the cathode, porous catalytic films of Pt and/or mixed conducting perovskites are employed for oxygen reduction. Overall thickness of the fuel cell structure is expected to be several hundred nanometers.

We are currently constructing additional ALD reactors and developing the window of ALD deposition parameters and appropriate precursors to fabricate the quaternary BYZ composition. ALD process is ideally suited not only to grow high-quality pin-hole free BYZ films at the nanoscale but also to grow them with uniform conformality in three dimensions, since it is essentially a surface reaction limited chemical vapor deposition process where one reaction cycle produces only one monolayer. This allows precise control of film composition, microstructure and thickness.

Also, we employ computational tools for molecular simulation and modeling both to predict and to understand the energetics and kinetics of rate processes governing cell performance.

In summary, this program is aiming at fabricating a 1 W DMFC based on a 100 mW/cm2 MEA, which is composed of nanoporous catalytic elecrodes and an 50 - 100 nm thin solid state proton conducting membrane that operates around 60 oC.

Direct Methanol Fuel Cells (DMFCs)


Why DMFC?

Rapid proliferation and growth of portable electronics such as cell phones and laptops into the consumer marketplace created an urgent need for high energy content, small and reliable energy generation and storage systems.

Currently, Li-ion or Ni-metal hydride batteries are widely employed for primary power sources for portable electronic devices. They have several disadvantages that limit performance, reliability and capacity. They provide a limited amount of energy dictated by the mass of their electrodes, and require frequent charging and, hence, access to a wall outlet. They have lifetimes limited at best to 1000 charge/discharge cycles. And their volumetric and gravimetric energy densities require significant weight and volume in order to provide sufficient energy, as compared to methanol.

In contrast, DMFC power sources provide high energy content for reliability, long lasting device operation, and instant refueling via a portable methanol cartridge. The superior energy content of methanol is evident in Table I when compared to energy densities of batteries commonly used as power sources for portable devices.

[Table 1]


How do DMFCs work?

Fuel cells are electrochemical devices that convert chemical energy stored in the bonds of the fuel into electrical energy. In its basic form, a fuel cell is composed of an ion selective membrane in between a cathode and anode layers, where the oxidant is reduced and the fuel is oxidized, respectively. During this process, electrons released at the anode are shuttled through the external circuit back to the cathode, generating electricity, [...???]

DMFCs employ a methanol-water solution as the fuel, and air to provide the oxidant. The role of water is merely to promote an internal reforming reaction to generate protons, which are then transported across a proton selective membrane, and react with oxygen in the air in contact with the cathode. The electrode reactions and the schematic design of a typical DMFC is provided below. (INSERT DMFC SCHEMATIC from Gur ppt)


Problems and Challenges with Conventional DMFCs

Current DMFC technologies employ a proton conducting ionomeric polymer membrane (typically NafionTM) as the electrolyte whose conductivity depends strongly on its degree of hydration. If the polymer membrane dries for some reason, the fuel cell ceases to operate. This necessitates the implementation of an active and dynamic humidification or water management system, which naturally reduces the system efficiency and adds to the size and weight of the fuel cell. Additionally, methanol cross-over presents a major problem in DMFCs where unreacted methanol crosses over from the anode compartment through the polymer membrane into the cathode compartment where it is oxidized to produce CO2. This chemical short-circuiting gives rise to polarization losses at the cathode, decreases the open circuit voltage of the fuel cell, and reduces the fuel utilization efficiency.

Ideally, DMFCs should operate with stochiometric (i.e., 1:1 molar ratio) methanol and water on the fuel side. Unfortunately, methanol cross-over and its associated problems get worse as the methanol concentration increases. To mitigate methanol crossover, only dilute methanol solutions much below the stochiometric amount, generally about 1M (~4 wt % methanol in water), can be employed as fuel in conventional DMFCs. But this reduces both the capacity and the overall efficiency of the system. Currently, DFMCs have conversion efficiencies in the range 20% to 30%.

It is clear that incremental advances in polymeric membranes and noble metal catalytic electrodes will only result in marginal improvements in these fuel cells. A radically different approach is required in order to improve fuel utilization and conversion efficiencies in a dramatic way.


Our Radical Solution for Next Generation DMFCs

Our novel DMFC concept employs dense, ultra thin film proton-conducting ceramic membranes with nanostructured catalytic electrodes. The cell architecture is fabricated on patterned Si substrates that further extend the prospects of this design for on-chip power applications.

This all-solid-state MEA architecture deviates dramatically from the conventional DMFCs described above. By replacing the water-saturated polymeric electrolyte with a dense ceramic proton conducting membrane, methanol cross-over and water management problems of DMFCs are completely eliminated. This approach promises major improvements in device performance by increasing the open circuit voltage to its theoretical value of 1.2V, minimizing cathodic polarization, and increasing both fuel utilization and conversion efficiency.

In addition, this configuration allows us to use the proper stoichiometric proportions of water and methanol, since the crossover problem is not an issue anymore with this architecture. Normally DMFc based on Nafion or other polymeric membranes cannot employ methanol concentrations much higher that 3-4% due to severe crossover losses and membrane degradation due to reaction with methanol. Table II compares the energy available to Nafion based DMFCs that employ about 4% (i.e.,1 molar) methanol solution with that of the Stanford design that employs the stoichiometric amount of 69% (17 molar). It is clear that the Stanford design provides a significant enhancement in the amount of available energy.

[Table 2]

An important advantage of the Stanford architecture is its compatibility with existing chip fabrication processes. We fabricate our ultra thin film membrane electrode assemblies (MEA) on single crystal Si substrates using MEMS processes in conjunction with advanced deposition techniques such as pulsed laser deposition (PLD) and atomic layer deposition (ALD), with total MEA thicknesses of several hundred nanometers. These micro- and nano-fabrication methodologies and our ultra thin film MEA architecture are ideally suited for on-chip manufacturing and can readily be adopted by the semiconductor industry in their chip production lines.


Possibility of On-Chip Power?

Our solid state DMFC cell architecture that is fabricated on patterned Si substrates further extends the prospects of this design for on-chip power applications. Compatibility of our fabrication methods and materials with those of chip manufacturing steps poses an attractive opportunity for the development of on-chip power modules.

Moreover, our DMFC offers added benefits since it is significantly lighter and smaller than batteries (see Table I), provides high energy content, minimizes ohmic loses in chip circuitry that otherwise would give rise to increased Joule heating, serves as a heat sink for the microprocessor, and provides efficient, reliable and more importantly, longer lasting power in proportionately smaller volume.

We are actively seeking industrial partners to participate in this ground-breaking effort and work with us to help design and integrate our DMFC architecture into chip manufacturing processes to provide on-chip power.