Patent Application
20050260117
1. Field of the Invention
This invention relates to a method of producing carbon nanostructures for use as catalyst supports in fuel cells
2. Discussion of the Prior Art
During the last few years, carbon nanostructures such as nanofibers and single or multi-walled nanotubes have received a lot of attention in the field of fuel cell research and development. The surface area, porosity and electromagnetic properties of such structures make them more suitable as catalyst supports in fuel cell electrode assemblies than carbon black powders.
Proton exchange membrane fuel cells (PEMFCs) have the highest power density when compared to other types of fuel cells. In order to improve fuel cell performance, reactant and catalytic material utilization must be optimized. PEMFCs rely on carbon-supported catalysts to produce the electrochemical reaction that generates such high power densities. In order to improve the electrochemical reaction rate, the catalyst should be as highly dispersed and as small as possible to expose the greatest surface area to the reactants. The catalyst layer usually consists of platinum crystallites with a mean diameter of a few nanometers highly dispersed on the surface of an amorphous carbon black powder.
One of the best catalyst performances in recent past was with a Pt loading of 0.1 to 0.2 mg/cm2 for 20 wt. % Pt/C [see Z. Qi et al, J. Power Sources, 113, pp. 37-43 (2003)]. Recently, it was shown by T. Yoshitake et al [see Physica B, 323, pp. 124-126 (2002)] that using single walled carbon nanohorns instead of carbon black increases the power density of PEMFCs by improving catalyst dispersion, and that iron can be used instead of or with Pt as a catalyst at the cathode of PEMFCs [see H. Wang et al, J. Phys. Chem. B, 103 pp 2042-2049 (1999) and T. Todat et al, J. Electrochem. Chem., 460, pp 258-262 (1999)]. In general, methods used to produce carbon nanostructures are somewhat complicated. Accordingly, there is a need for a simple method of producing carbon nanostructures for use as fuel cell catalyst supports.
The object of the present invention is to meet the above mentioned need by providing a relatively simple method of producing a novel carbon nanostructure which is an effective catalyst support.
Accordingly, the present invention relates to a method of producing carbon nanostructures comprising the steps of:
More specifically, in accordance with a first step in the method set out above, carbon and transition metal such as iron, nickel, cobalt, titanium and/or vanadium are mixed with carbon, and the resulting mixture is subjected to ball milling in an iron crucible under a hydrogen pressure of about 1 MPa. Preferably, before mixing the carbon and transition metal(s), the carbon is treated under vacuum or with a flowing inert gas to remove contaminants. In one embodiment of the invention, between the mixing and ball milling steps, the mixture was placed in a reaction vessel, the vessel was evacuated, and pure H2 was introduced into the vessel at a pressure of 0.1 to 21 MPa. The ball milling can be effected using any type of milling machine. Ball milling breaks the crystal structures of the elements to form nanocrystalline and amorphous grains, the hydrogenation of carbon, and the mechanical alloying of carbide and/or transition alloys. The hydrogenation of carbon and eventually the production of methane is catalyzed by the presence of transition metal. The hydrogen concentration in the metal reaches up to about 4.5 wt percent after a milling time of 48 hours. Pure graphite ball milled in an inert atmosphere for 150 hours and then heated at 1400° C. for 6 hours results in nucleation and growth of carbon nanotubes. Some Fe in the sample comes from the crucible walls. In one embodiment, the addition of Co to the mixture alloys with Fe to produce FeCo and Fe3CO7 crystallites, thereby preventing later formation of iron carbides. Ball milling is effected under a hydrogen atmosphere which creates C—H covalent bonds.
In a second step of the process, the sample is heated in a quartz tube at temperatures ranging from 500 to 1000° C. for one hour under an inert atmosphere, in this case helium or argon. During heating, pure metal or alloy crystallizes and catalyzes methane production from previously hydrogenated carbon. The subsequent catalytic decomposition of methane on nano-crystalline carbide or transition metal synthesizes porous carbon nanostructures. The addition of transition metals meets several goals. The metals catalyze the hydrogenation of carbon and the subsequent production of methane, and catalyze the decomposition of methane on nanocrystals to produce the carbon nanostructures. The use of transition metals substantially reduces the milling and heat treating time as well as the heat treating temperature.
After each step of the process, X-ray diffraction analyses are performed to characterize sample structures and determine the average Fe crystallite diameter, Fe being the main metal constituent. At the end of the process, atomic absorption spectra are obtained to evaluate the amount of C, Fe and Co in the catalyst composition. TEM images are taken to determine whether carbon nanostructures are present in the sample and whether the metal crystallites are well distributed. BET analysis are performed to obtain the specific surface area and porosity of the product. Specific electrical conductivity of the samples is measured with a low frequency LCR impedance meter with the samples compressed in a hollow acrylic cylinder and sandwiched between two metal plungers.
Several samples were prepared containing various amounts of carbon, iron and cobalt, and the mean crystallite diameters were determined by X-ray diffraction analysis. The exact composition of each sample was determined by atomic mass spectroscopy. The specific conductivity of the samples heated at different temperatures (except for sample number 4, which was purified, i.e. the transition metals were omitted) was also determined. The metal composition of eleven samples and the above listed properties thereof are set out in the following table.
In Table 1, T is the temperature to which sample was heated, d is the diameter of particles in angstroms and a is the specific electrical conductivity of the samples in siemens (S) per square centimeter.
From the results set out in the above table, it has been established that a temperature of 600° C. is required to initiate the production of carbon nanostructures. The average metal cluster diameter varies from 294 to 401 Å for the electrically conducting samples.
In the accompanying drawings:
Referring to the drawings,
The BET analysis of sample number 3 gives a specific surface of 49 m2/g. The specific surface of sample number 4 is 246 m2/g. This increase in specific surface comes mainly from the external carbon surface and from the meso porous structure. The meso porous volume up to a pore dimension of 30 nm is evaluated at 0.51 cm3/g. The analysis of the N isotherms suggests that samples prepared in accordance with the method of the present invention have good macro porous structures. A large pore volume means that mass transfer losses in PEMFCs should be reduced thus resulting in higher power densities.
Referring to
The metal content of sample number 3 was reduced to zero in order to measure the electrical conductivity of the carbon itself the small difference in conductivity between samples 3 and 4 proves that the metal content of the samples increases the conductivity by only a small amount, even with an Fe loading of 63.8 wt %.
Thus, there has been described a simple method for producing carbon powders containing amorphous and structured carbons, which can be used as catalyst supports for PEMFCs. The TEM images (
