Studies on humidification of PEM fuel cells

Abstract

Proton exchange membrane (PEM) fuel cell is an electrochemical energy conversion device which converts chemical energy of hydrogen into electrical energy via redox reactions. The Pt/C (Platinum on carbon) catalyst layers at anode and cathode facilitate oxidation of hydrogen and reduction of oxygen, respectively. These catalyst layers are separated by an ion-exchange membrane. Besides acting as an electrical insulator thereby preventing shorting of the fuel cell, the main purpose of this membrane is to allow protons to migrate selectively from anode to cathode without allowing the gaseous reactants to diffuse across. Such membranes are typically made of Nafion, which is a sulfonated polytetrafluoroethylene (PTFE) random co-polymer. The co-polymer consists of PTFE backbone and pendant side chains each ending in a sulphonate (SO3-) group. The dissimilar nature of the covalently bonded pendant group and backbone chain results in microphase separation, which is enhanced by solvation (uptake of water or solvent molecules). The microphase separated morphology comprises micelles, which upon water uptake self-assemble into an inter-connected network of nanochannels comprising solvated sulfonic acid groups and water molecules. The sulphonate groups are responsible for water retention and transport in the membrane [1]. This phase-separated morphology gives Nafion its unique ion and solvent transport properties [2]. The hydration of Nafion membranes is achieved partially by water generated at the cathode because of the oxygen reduction reaction. However, it is often observed that the water generated at the cathode is insufficient for optimal hydration of the membrane because of the water lost due to convective evaporation on the cathode, thereby resulting in lower fuel cell performance [3]. Therefore, a common practice is to externally humidify the feed gases supplied to PEM fuel cell using humidification devices such as bubble humidifier, membrane humidifier, evaporators etc. Among these, membrane humidifiers due to their large surface area, compactness and easy-to-use module setup, are the most popular choice for fuel cell humidification. Other methods include passive humidification of cells using novel techniques based on the principle of evaporative cooling such as wicks, direct liquid water injection, porous water transport platesetc. In any case, achieving optimum humidification is critical for water management in PEM fuel cells and its durable performance. In this work, we have focused on membrane humidifiers in the class of active humidification methods and wick based humidification in the class of passive humidification methods. Statement of Problem, Aim and Objectives While there is myriad literature on membrane humidifiers and passive humidification methods for PEM fuel cells, there are several important lacunae which have remained unaddressed. These are: i) comprehensive understanding of mechanisms involved in membrane (both porous and non-porous) based gas humidification ii) rigorous experimental validation of models developed for heat and mass transport in membranes iii) poor pressure tolerance and high cost of dense membrane humidifiers iv) dry-feed operation of PEM fuel cell using passive humidification technique for large active areas. The aim of this thesis is to address these lacunae. Specifically, the work presented in this thesis is aimed at contributing to the field of membrane technology and PEM fuel cells by delineating the following research objectives covering both active and passive methods of humidification. Research objectives and methodology: 1. Understanding gas humidification through dense membranes. The methodology for achieving this objective includes: 1.1. Mathematical modelling of water-to-gas hollow fiber membrane humidifier and 1.2. Validation of model with experimental results on commercial scale Perma Pure humidifier 2. Developing cost effective (up to 5 – 10 folds cost reduction from commercial benchmark humidifier) hollow fiber membranes for gas humidification. The methodology for achieving this objective includes: 2.1. Developing polysulfone based asymmetric hollow fiber membranes for gas humidification. 2.2. Characterizing the hollow fiber membranes for microstructure elucidation. 2.3. Mathematical modelling of water-to-gas asymmetric hollow fiber membrane humidifier 2.4. Validation of model with experimental results on developed humidifiers. 3. Understanding maldistribution of flow inside hollow fiber membrane humidifier. The methodology for achieving this objective includes: 3.1. Flow simulations using COMSOL multi-physics CFD tool. 3.2. Studying the effect of tube symmetry and header geometry on flow distribution. 4. Investigate wick based passive humidification technique for water management in PEM fuel cells under dry-feed operation. The methodology for achieving this objective includes: 4.1. Characterizing carbon cloth for its feasibility to be used as a wicking material. 4.2. Testing single cell with a hydrophilic carbon cloth over the MEA extending out of the cell and dipping in a water reservoir to facilitate wicking action. 4.3. Testing different configurations of fuel cell with wick. 4.4. Electrochemical impedance measurements for comparison of different configurations with control experiments without wick. Key findings of the research work The work presented on membrane humidifiers in this thesis is a combination of analytical theory, numerical simulations and experimental investigations. The summary and key results of each of the major working scheme is provided below. Scheme-1: Mathematical modelling of gas humidification using dense nafion membranes This work focusses on improving the understanding of the mechanisms involved in water-to-gas membrane humidification using dense membranes. It presents a quasi-2D model for water-to-gas hollow fiber membrane humidifier and its validation with experiments performed on commercial Perma Pure FC200 humidifier module. The model takes into account the relevant phase equilibria along with coupled heat and mass transport across dense Nafion hollow fiber membranes. The model is shown to predict the humidifier performance within 8 % deviation from the experimental results. Scheme-2: Experimental and theoretical investigations on polysulfone based membrane humidifiers This work focusses on developing cost effective polysulfone asymmetric membranes which can be used for gas humidification over more expensive dense Nafion membrane-based humidifiers. It presents experimental and theoretical investigations of polysulfone based asymmetric hollow fiber membranes. The experimental section includes development of polysulfone based hollow fiber membranes, their characterization and humidification performance tests. The humidification performance was found to decrease with increasing polymer concentration. Membranes developed from polymer concentration of 27 % were shown to provide high humidification performance without any flooding or entrainment issues. A mathematical model based on resistance-in-series approach is proposed for these membranes and is shown to fit the experimental observations in humidification tests. It is found that in asymmetric membranes, the skin and bulk resistances govern the mass transport whereas heat transport is limited by the interfacial resistance. Membranes made from 27 wt % polymer concentration are found to work comparably with bubble humidifiers for fuel cell operation. Scheme-3: Flow behavior inside humidifiers using CFD simulations This work presents a computational fluid dynamics study on flow behavior inside humidifier module. The effect of symmetry in tube arrangement inside the humidifier module helps provide relevant design guidelines for minimizing flow maldistribution in tubes. Further, shell side flow simulations based on current humidifier design provide suggestions for reducing the dead volume inside humidifier. On tube side, a perfectly symmetric tube arrangement along with a small truncated conical header is observed to show uniform flow distribution. On shell side, a dead volume zone is developed near the inlet/outlet ports because of the current design. It is therefore suggested to have the inlet and outlet ports on the end caps to reduce the dead volume. Scheme-4: Wick based passive humidification of PEM fuel cells. This work focusses on reducing parasitic power loss in PEM fuel cell operation by providing passive means of fuel cell hydration and dry-feed operation. It presents a novel wick based internal humidification technique for low temperature PEM fuel cells. The technique leverages on hydrophilic carbon cloth as a wick placed over the MEA for passive water transport inside the cell without external aid of membrane humidifiers. Configurations with wick placed over anode (Configuration-B) or cathode (Configuration-C) were shown to have performance comparable with a fuel cell operated with external humidifiers on both sides.