Graphite-epoxy composites for fuel-cell bipolar plates: Wet vs dry mixing and role of the design of experiment in the optimization of molding parameters

Highlights

• Bipolar plates with conductivity above US DOE were obtained from low cost materials.

• When an epoxy binder is used, processed graphite is preferred over natural graphite.

• Preparation by wet mixing is preferred over dry mixing the composite components.

• Design of Experiment gave useful hints for future improvements.

Abstract

Bipolar plates (BPs) are key components of Proton Exchange Membrane Fuel Cells mainly employed in hydrogen-powered electric vehicles. Here, a reliable and detailed experimental method to prepare graphite-epoxy composites suitable for manufacturing BPs is reported. Dry and wet mixing procedures were compared and a simple composition was optimized, with regard to electrical conductivity. The adoption of wet mixing of the components and the choice of the conductive filler were the main factors that contributed to the achievement of good electrical and mechanical properties. The addition of a small percentage of carbon black as a secondary filler was also advantageous. The effects of molding parameters (pressure, temperature and time) on a fixed-composition graphite-epoxy composite were modeled using a Design Of Experiments approach, which provided valuable information for future improvements. Conductivity values well above the US DOE requirements were obtained.

Introduction

The need for independence from fossil fuels and for limitation in CO₂ emission [[1]] pushes the international community towards the search and evaluation of alternative energy production systems [[2], [3], [4]]. Among them, the Fuel cell (FC) technology is attracting increasing interest for its multipurpose applications and high thermodynamic yield [[5], [6], [7]]. FCs permit to get electricity from hydrogen and air in a controlled way, forming water as a harmless byproduct. Key components of Proton Exchange Membrane (PEM) FCs are bipolar plates (BPs) [8,9]. They are responsible for fundamental functions in FCs: electrical connections, supply of gaseous reactants, and removal of heat and byproducts. BPs represent also a significant part of the FC cost, around 30–50% [10,11], and weight, around 80%. Any improvement in costs and weight can impact the economy of this sector.

Graphite is a suitable material for BPs manufacturing, thanks to a combination of thermal and electrical stability, but its inherent brittleness represents a weak point for its employment when compared to metal BPs. The latter, on the contrary, suffer from low corrosion resistance [12,13]. Graphite-polymer composites can be used to improve the mechanical performance of bulk graphite and to maintain manufacturing temperatures well below those employed in sintering conditions; the concurrent reduction in electrical conductivity can be controlled through the development of optimized formulations and production methods. In order to work as BP, a composite must have high mechanical strength, thermal and electrical conductivities, corrosion resistance, low gas permeability, and density, accordingly to the technical targets defined by the US DOE [14] (Table 1).

Composite graphite-polymer BPs can be realized by mixing various conductive carbonaceous fillers with a polymeric binder, often a thermosetting one. The resulting powder or paste is then pressed and cured to obtain the final BP [15]. The conductive filler is the largest ingredient by mass and volume, often in excess of 80 wt%, in order to meet the US DOE required conductivities. Natural graphite, expanded graphite and other processed graphite types are typically employed, frequently mixed with lower amounts of secondary fillers like carbon blacks (CBs), carbon nanotubes, nanofibers, or others.

Addition polymers, like epoxy or some kind of phenolic resins, are among the most commonly employed binders. They undergo curing mechanisms free from small molecular weight by-products thus preventing the formation of voids inside the material. Aside from the presence of reinforcing fibers, the mechanical integrity of the composite is highly dependent from the ability of the binder to glue together the components. In this context, the adhesion between binder and filler is an essential issue [15].

Despite the good availability of information around the formulation of graphite-polymer composites, the experimental activity involved in their manufacturing is often not well documented in the literature, with few exceptions [16]. However, the final properties of composites are critically dependent from subtle preparation details. This is the reason that pushed us to thoroughly describe, here, our experimental activity aimed at the realization of a graphite-epoxy composite having adequate electrical and mechanical features to work as PEMFC BPs and based on cost effective and commercially available materials. In particular, we found that wet mixing is by far superior to dry mixing and that another key issue is the choice of the type of graphite filler. Once a good manufacturing method and an optimized formulation of the composite were devised, the effects of molding parameters (pressure, temperature, and time, measured using a custom-made pressing machine) were also studied by means of a two-level full factorial Design Of Experiments (DOE) approach, that led to a further improvement of the performances of the obtained composites. Appropriate mixing and pressing conditions, together with polymer and filler choice, proved to be essential parameters in order to obtain compliant and reliable results.