Hydrogen for Fuel Cells from Methanol Reforming

In seeking to overcome the problems associated with hydrogen storage and distribution, numerous approaches have set out to use liquids rich in hydrogen such as gasoline or methanol as a source of hydrogen via on-board reformers. In contrast to pure hydrogen-based systems, they are compact (containing on a volume basis more hydrogen than even liquid hydrogen) and easy to store and handle without pressurization. The possibility of generating hydrogen with more than 80% efficiency by the on-board reforming of gasoline has been demon strated. However, the process is expensive and challenging, because it involves high temperature and needs considerable time to reach a steady operational state. The advantage is that the distribution network for gasoline already exists, though this would not solve the problems of diminishing oil resources and dependence from oil-producing countries. On the other hand, methanol steam reformers operating at much lower temperature (250-350 °C) [95], albeit still expensive, are more adapted for on-board applications. The absence of C-C bonds in methanol, which are difficult to break, greatly facilitates its transformation to high-purity hydrogen with 80-90% efficiency [134]. Methanol, furthermore contains no sulfur, a contaminant for fuel cells. With the reformer operating at low temperature, no nitrogen oxides are formed. The use of an on-board reformer enables the rapid and efficient delivery of hydrogen from a liquid fuel that can be easily distributed and stored on the vehicle. To date, methanol is the only liquid fuel that has been processed and demonstrated on a practical scale in fuel cells for transportation applications. The disadvantages of this system are, however, the added weight, complexity and cost of the overall system, as well as trace emissions that may be produced from the reformer when it burns some of the methanol to provide the necessary heat for hydrogen production [119].

The potential for on-board methanol reformers to power FCVs has been demonstrated by several prototypes constructed and tested by various automobile companies. In 1997, DaimlerChrysler presented the first methanol-fueled FCV the Necar 3, a modified A-Class Mercedes-Benz compact vehicle equipped with a 50-kW PEM fuel cell and a driving range of 400 km. In 2000, an improved version with a 85-kW fuel cell, the NECAR 5 was introduced (Fig. 11.4) [135]. In this vehicle, which was described by the company as being fit for practical use [119], the entire fuel cell and reformer system has been accommodated in the under-body of the car, enabling five passengers and their luggage to be transported at a maximum speed of 150 km h-1 and a driving range approaching 500 km [135]. In 2002, this FCV was the first to complete a coast-to-coast trip across America, from San Francisco to Washington D.C., or a distance of more than 5000 km, with methanol being refueled every 500 km [136]. DaimlerChrysler also presented in 2000, a fuel cell/battery hybrid Jeep Commander SUV powered by methanol. Based on a Ford Focus, Ford constructed THINK FC5 [137], a methanol-fueled FCV with a fuel cell/reformer system beneath the vehicle's floor and characteris-

Daimler Chrysler Necar
Figure 11.4 DaimlerChryslers methanol-fueled NECAR 5 fuel cell vehicle (introduced in 2000, Courtesy of DaimlerChrysler).

tics similar to NECAR 5. Other companies which have developed methanol-pow-ered FCVs include General Motors, Honda, Mazda, Mitsubishi, Nissan, and Toyota. However, most of these companies, including Daimler-Chrysler, have recently concentrated their efforts on vehicles with on-board storage of pure hydrogen.

Georgetown University of Washington DC has been in the forefront in the development of transportation fuel cells for some 20 years. Supported by the U.S. Federal Transit Administration, the University has developed several fuel cell transit buses running on methanol [138]. In 1994 and 1995, Georgetown produced three buses that were the world's first FCVs able to operate on a liquid methanol fuel (Fig. 11.5). These methanol buses, each powered by a 50-kW Phosphoric Acid Fuel Cell (PAFC) combined with a methanol steam reformer, are still operating today. In 1998, an improved second-generation bus using a more powerful 100 kW PAFC provided by UTC Fuel Cells was introduced, followed in 2001 by the first urban transit bus powered by a liquid-fueled 100 kW PEMFC system manufactured by Ballard Power System, a major fuel cell developer. Batteries provide surge power and a means to recover braking energy by regeneration. These two buses, each able to seat 40 passengers, meet all the requirements of the transit industry, are much more quiet than their ICE-powered counterparts, and have a driving range of some 560 km between refueling. The use of methanol allows the refueling to be as quick and easy as with diesel buses. PM and NOx emissions are virtually eliminated, and other emissions are well below even the cleanest CNG buses on the road with the most stringent clean air standards [138]. Capitalizing on previous experience, a third generation fuel cell hybrid bus fueled by methanol is presently under development.

Besides on-board methanol reforming, methanol is also seen as a convenient way to produce hydrogen in fueling stations to refuel hydrogen FCVs. Mitsubishi Gas Chemical has developed a process to produce high-purity hydrogen by steam reforming of methanol using a highly active catalyst which allows operation at relatively low temperature (240-290 °C) and enables rapid start-up and stop, as well as flexible operation. These methanol-to-hydrogen (MTH) units, which range in production capacity from 50 to 4000 m3 H2 per hour, are already used by a variety of customers in the electronic, glass, ceramics, and food processing industries [139]. They provide an excellent reliability, prolonged life service, and minimal

Figure 11.5 Methanol fuel cell buses developed at Georgetown University in front of the U.S. Capitol (2002) (Source: Georgetown University).

maintenance [134]. Based on the technology developed by Mitsubishi Gas Chemical, the first fueling station to supply hydrogen by methanol reforming was constructed in Kawasaki, Japan as part of the Japan Hydrogen & Fuel Cell Demonstration Project (JHFC), which is studying a large array of possible feedstocks for hydrogen generation. According to JHFC, methanol is the safest of all materials available for hydrogen production [140]. Operating at relatively low temperature, the MTH process has a clear advantage over the reforming of natural gas and other hydrocarbons, which needs to be carried out above 600 °C. A smaller amount of energy is necessary to heat methanol to the appropriate reaction temperature. At the Kawasaki station, methanol and water are evaporated and reacted over a catalyst. After purification and separation, the hydrogen produced is compressed and stored to provide FCVs with high-pressure hydrogen [140]. Although methanol reforming is a convenient and attractive way of producing hydrogen, it does not solve the problems associated with the costly and difficult on-board storage of hydrogen.

Significant investigations aimed at further improving methanol reforming to hydrogen, whether focused on on-board or stationary applications, are under way. Hydrogen obtained by methanol reforming in current processes always contains more than 100 ppm CO, a poison for PEM fuel cell catalysts operating below 100 °C. At present, reformed gas has thus to be cleaned to remove CO, lowering the total efficiency of the process. At the Brookhaven National Laboratory, new catalysts have been designed which produce hydrogen with high yield, generating by the same time only marginal amounts of CO. Using a process known as oxidative steam reforming, which combines steam reforming and the partial oxidation of methanol, and different novel catalyst systems, the National Industrial Research Laboratory of Nagoya has also achieved the production of high-purity hydrogen with either zero or only trace amounts of CO, at high methanol conversion and temperatures as low as 230 °C. Oxidative steam reforming of methanol also has the advantage of being - contrary to steam reforming - an exothermic reaction, minimizing energy consumption. The exothermicity of the reaction however, can also be a drawback since the generated heat and consequently the reactors temperature may be difficult to control. In an ideal case, the reaction should therefore only produce enough energy to sustain itself. This is the principle of autothermal reforming. The autothermal reforming of methanol, which combines steam reforming and partial oxidation of methanol in a specific ratio, is an idea first developed in the 1980s by Johnson-Matthey. It is neither exothermic nor endothermic, and thus does not require any external heating once the reaction temperature has been reached. For a fast start-up, the methanol/oxygen feed ratio can be varied, as has been shown for example in Johnson-Matthey's "Hot-Spot" methanol reformer.

Direct Methanol Fuel Cell (DMFC)

In contrast to hydrogen fuel cells, direct methanol fuel cells are not dependent upon hydrogen generation by processes such as electrolysis of water, or natural gas or hydrocarbon reforming. As mentioned earlier, the storage and distribution of hydrogen fuel will require an entirely new infrastructure or a complete overhaul of existing systems, all of which constitutes a major barrier for entry into the commercial market. Methanol, on the other hand, is a clear liquid fuel (b.p. = 64.7 °C, d = 0.791 g mL-1) that does not require special cooling at ambient temperature, and will fit into existing storage and dispensing units with only small modification.

Methanol has a relatively high volumetric theoretical energy density compared to other systems such as conventional batteries and the H2-PEM fuel cell (Fig. 11.6). This is of basic importance for small portable applications, as battery technology may not be able to keep up with the demand for laptops and mobile phones that are lightweight and have extended operating time [141, 142].

There is more hydrogen in 1 L of liquid methanol than in 1 L of pure cryogenic hydrogen (98.8 g of hydrogen in 1 L of methanol at room temperature compared to 70.8 g in liquid hydrogen at -253 °C). Therefore, it transpires that methanol is a safe carrier fuel for hydrogen.

In the past, methanol-based PEM cells have used a separate reformer to release the hydrogen from liquid methanol, after which the pure hydrogen is fed into the fuel cell stack. However, since 1990 researchers at the Jet Propulsion Laboratory, and the authors' group at the University of Southern California [143, 144], have developed a simple DMFC that consists of two electrodes separated by a proton exchange membrane (PEM) and connected via an external circuit that allows the conversion of free energy from the chemical reaction of methanol with air to be directly converted into electrical energy (Fig. 11.7).

The anode is exposed to methanol water mixture fed by flow from an external container, where it is oxidized to produce protons that travel through the PEM by ionic conduction, and electrons that travel through the external circuit by electronic conduction. The cathode containing platinum as catalyst is exposed to oxygen or air, which may be either ambient or pressurized. The PEM is coated on both sides with layers of catalyst (1:1 Pt-Ru catalyst at the anode and Pt catalyst at the cathode), usually supported by a gas diffusion electrically conductive carbon (graphite) electrode at the cathode and a liquid feed-type carbon electrode struc-

Figure 11.6 Theoretical energy density of batteries, H2-PEM fuel cells, and DMFCs.

Figure 11.6 Theoretical energy density of batteries, H2-PEM fuel cells, and DMFCs.

Methanol Reformer

catalyst layer catalyst I

Figure 11.7 The direct methanol fuel cell (DMFC).

catalyst layer catalyst I

Figure 11.7 The direct methanol fuel cell (DMFC).

ture at the anode that facilitates reduction of oxygen and oxidation of methanol, respectively Eqs. (1) and (2).

At room temperature, the overall reaction (Eq. (3)) gives a theoretical open circuit voltage of 1.21 V with a theoretical efficiency close to 97%. Although fundamentally simple, DMFCs presently still perform well below their theoretical Nernstian potential, even under open-circuit conditions because of sluggish redox kinetics and fuel crossover. Advances in both catalysts and membranes have - and are being - made and promise to conquer these issues, in terms of both performance and cost.

PEMs intended for H2-PEM fuel cells are not good candidates for DMFCs due to the issue of methanol crossover from anode to cathode. High crossover rates have a deleterious effect on DMFC performance. As oxygen is reduced to produce a cathodic current, oxidation of crossover methanol simultaneously produces an anodic current that results in a mixed potential and overall reduced cathode potential. Methanol may poison the cathode catalyst (Pt) and block cathode catalyst sites, further reducing its ability to efficiently reduce oxygen, and necessitating an increase in oxygen flow above and beyond stoichiometric requirements. Furthermore, the chemical oxidation of methanol also produces excessive water that hampers cathode performance due to flooding. Instead of meaningful electrical energy, waste heat is generated and fuel utilization efficiency is lowered. Methanol permeates through the PEM by two methods: (i) by simple diffusion due to a concentration gradient; and (ii) by electro-osmotic drag from proton migration when the cell is under an applied current. New membranes based on hydrocarbon/hy-drofluorocarbon materials with reduced cost and cross-over characteristics have been developed that allow room temperature efficiency of 34% [145, 146].

With the advances made in all aspects of DMFC research, many companies are now actively developing low-power DMFCs for portable devices such as cellular phones and laptop computers [147]. The success of these initial devices is pivotal to transition away from rechargeable batteries, which can in theory deliver only 600 W-h kg-1 at best. Currently, rechargeable commercial lithium ion batteries deliver a power density anywhere between 120 and 150 W-h kg-1. Consumers will soon enjoy the benefits of DMFC-powered devices, including longer cellphone talk time, extended usage time on laptop computers, rapid rechargeability, and lighter weight contribution of the power source.

Many issues arise when single fuel cells are assembled into practical stack assemblies such as temperature and pressure control, resistance, and water management. Various materials and designs have been employed to deal with these issues. The Jet Propulsion Laboratory [148] has developed a small, six-cell DMFC stack with total active area of ~48 cm2 (anode and cathode catalyst loading of 4-6 mg cm-2) using ambient air with 1 M methanol at room temperature, having a power density from 6 to 10 mW cm-2. The cell is a flat array, where the cells are externally connected in series sharing a single membrane. The cathode catalyst, when applied to teflonized carbon supports, imparted excellent water-removal properties to the system, although the design also provides increased ohmic resistance. Three two-flat pack arrays are necessary to power a mobile phone, and a 10-h operating time is estimated before methanol replenishment.

Los Alamos National Labs, in conjunction with Motorola [149], have also developed a stack for mobile phones utilizing ceramic fuel plates with microfluidic channels for efficient delivery of methanol and water and removal of CO2. Their four-cell stack with total active area ~60 cm2 gives a power density between 12 and 27 mW cm-2 with catalyst loading of 6-10 mg cm-2 at room temperature.

The Korea Institute of Energy Research (KIER) has developed a 10-W DMFC stack [150]. The cell has a bipolar plate design with six cells each of 52 cm2 active area. Most notably, the stack was operated with 2.5 M methanol, and achieved 6.3 Wat room temperature using ambient oxygen flow. In addition, the Korea Institute of Science and Technology (KIST) has developed and assembled a six-cell monopolar stack [151] with a total active area of 27 cm2, and which produced a power density of 37 mW cm-2 using 4 M methanol and ambient air.

Toshiba has developed a promising DMFC prototype stack for laptop computers. The stack has an average output of 12 W and may be continuously used for 5 h with a 50-mL methanol storage cartridge. To minimize the size of the cartridge, the cell collects output water for recombination with methanol. Sensors are hooked up directly to the PC to tell users when the cartridge needs replacing. NEC has developed similar stacks and, within the next few years, anticipates the stack to have a 40-h operation time.

Both stationary and portable DMFC stacks are now available for consumer purchase. For instance, The Fuel Cell Store offers its SFC A25 Smart Fuel Cell [152], capable of 25 W continuous output, and can cover four days worth of energy demand using only 2 kg of fuel. A larger 50-W model is also available.

In the transportation area, Daimler-Chrysler, working on DMFC for automotive purposes, constructed a one-person go-cart prototype vehicle powered by a 3-kW DMFC (Fig. 11.8) [153]. Recently, a similar vehicle, equipped with a 1.3-kW DMFC known as "JuMOVe", has been developed in Germany by the Julich Research Center [154]. In Japan, Yamaha presented in 2003 its FC06 prototype, the first two-wheeler motor cycle powered by a DMFC with an output of 500 W. Equipped with a 300 W AC outlet, this bike can also serve as an electric power source for outdoor activities or during emergencies [155]. An advanced version of this motor bike, the FC-me, is presently in practical use on a lease basis in Japan (Fig. 11.9). In the United States, Vectrix plans to commercialize in 2006

Figure 11.8 Daimler Chrysler DMFC go-cart (Courtesy of DaimlerChrysler).
Dmfc Stacks
Figure 11.9 (Courtesy © Yamaha Motor Co.).

a hybrid fuel cell scooter powered by a 800-W DMFC attached to a rechargeable battery [156, 157]. The fuel cell continuously recharges the battery, which powers the electric motor. Regenerative braking technology also captures the energy usually dissipated during braking to provide additional battery charging. With a maximum speed in excess of 100 km h-1 and a range of about 250 km at cruising speed, it has characteristics comparable to conventional scooters.

Considerable development efforts are still needed to make larger DMFCs practical, for example to be able to power motor cars, but ongoing progress is impressive. DMFCs offer numerous benefits over other proposed technologies in the transportation sector. By eliminating the need for a methanol steam reformer, the vehicle's weight, cost and the system's complexity can be significantly reduced, thereby improving fuel economy. DMFC systems also come much closer to the simplicity of direct hydrogen-fueled fuel cell, without the cumbersome problem of either on-board hydrogen storage or hydrogen-producing reformers. By emitting only water and CO2, other pollutant emissions (NOx, PM, SO2, etc.) are eliminated. As methanol will eventually be made by recycling atmospheric carbon dioxide, CO2 emissions will not be of any concern and there will be no dependence on fossil fuels.

Methanol as a transportation fuel has a variety of important advantages. In contrast to hydrogen, methanol does not need any energy-intensive procedures for pressurization or liquefaction. Because it is a liquid, it can be easily handled, stored, distributed and carried on board vehicles. Methanol is already used today in ICE vehicles. Through on-board methanol reformers it can act as an ideal hydrogen carrier for FCVs, and be used in the future directly in DMFC vehicles. Employing the same fuel from present ICE to advanced vehicles equipped with DMFCs will allow a smooth transition between existing and new technologies. With the progressive phase-out of MTBE as a gasoline additive, substantial methanol production over-capacity is immediately available to be used as a transportation fuel.

Over the next decade new innovations such as novel proton-conducting materials, membrane-less fuel cells, and cheaper and more efficient catalysts may lead DMFC technology away from the traditional cell structure and design. DMFC is poised to play a critical role in electricity production from methanol in the overall methanol economy structure.

Continue reading here: Fuel Cells Based on Other Fuels and Biofuel Cells

Was this article helpful?

+2 0


  • valerie
    What are problems with methanol fuel cells?
    9 months ago