Chemical Engineering Study Project 4 Ammonia-based fuel cells Name: James Nash Matriculation number: 1323742  Abstract 
 Ammonia appears to be a promising chemical for use as a fuel. It is easier to store and transport then hydrogen, and it is already one of the highest produced bulk chemicals in the world. This paper sought to review the main types of fuels cells that can utilise ammonia in their design, which included alkaline fuel cells AFCs, solid oxide fuel cells SOFCs, proton exchange membrane fuel cells PEMFCs, and alkaline membrane fuel cells AMFCs. Each was reviewed to get a better understanding of its potential, and to identify the bottlenecks remaining in their development. The principal drawbacks of each fuel cell are; For AFCs, they are not operable with carbon dioxide in the air, which seriously limits the scope of their use. SOFCs are operated at very high temperatures, 673K – 1273K, leading to issues with material durability. PEMFCs cannot be operated with ammonia, so require equipment to crack the ammonia into hydrogen before it can be used. AMFCs are like AFCs, but the membrane allows it to be operated with carbon dioxide present in the cell. The biggest challenge faced by both PEMFCs and AMFCs is water flooding, where water produced in the cell stays near the electrodes, inhibits their performance, leading to cell failure. Optimisation and control could be used in areas such as the water management in AMFCs and PEMFCs, and with the conditions in SOFCs to make sure the ammonia is being cracked into nitrogen and hydrogen completely. For the near future, the most likely candidate to reach commercialisation would appear to be a hydrogen-conducting solid oxide fuel cell.   Table of contents 1. Introduction 5 2. Theory 7 3. Ammonia-based fuel cells 11 4. Solid-oxide fuel cells 12 5. Proton-exchange membrane fuel cells (PEMFCs) 17 6. Alkaline-membrane fuel cells 20 7. Future for ammonia-fuel cells 25 8. Conclusion 27 9. References 31  Introduction In this review, the main fuel cells which can use ammonia as its fuel were identified and discussed. A brief look at ammonia itself as a fuel was undertaken, with comparisons between itself and hydrogen discussed. The main hurdles to be overcome by these technologies were investigated, and areas where control and optimisation could play a role were considered. This aim of this report was to provide an overall view of each major fuel cell, and attempt to find both the most realistic application of ammonia in fuel cells, and the most innovative possibility for its use. With the growing global demand to reduce carbon dioxide emissions, hydrogen would appear to be an ideal chemical for the energy industry. However due to its very small size/density, hydrogen itself is not a suitable replacement for fossil fuels. The main issue holding back hydrogen is the requirement that it must be stored as either a compressed gas under conditions around 690 bar and 15 or as a liquid at temperatures below -240 and atmospheric pressures [1]. Research has shown that transporting the fuel as pure hydrogen results in much larger efficiency losses compared to transport in the form of ammonia [2]. Fossil fuels are becoming increasingly more difficult to get out of the ground, and one day they will be used up. It is important that viable replacements for these fuels be discovered early, and have time to be researched thoroughly, and for innovations to be made. The development of our current energy infrastructure was a process that took time and money to get to where we are today. When considering ammonia’s viability as a fuel, the fact that it is still in its early stages must be considered. In terms of alternative fuels to hydrocarbons, ammonia is one of the leading substances to be used. It is 17.76% hydrogen by weight, and as it does not contain a carbon atom it will not release when utilised in a fuel cell. It is also one of the largest produced chemicals in the world due to its massive demand in the agriculture sector for use as fertilizer [3]. The main types of fuel cell discussed within this report are; alkaline fuel cells (AFC), solid-oxide fuel cells (SOFC), proton exchange membrane fuel cells (PEMFC) and alkaline membrane fuel cells (AMFC). Direct ammonia fuel cells were first investigated in the late 1960’s [4], they were based on alkaline fuel cells, operated at around 50-200 , using an electrolyte solution of KOH. These experiments were key in showing that ammonia could be used directly to power a fuel cell, and that this fuel cell could be operated at suitable pressures and temperatures. There current applications seem quite limited due to the issues when CO2 is present. Solid-oxide fuel cells also operate using ammonia as the feed, but the ammonia is cracked inside the fuel cell to produce hydrogen. To do this, the fuel cells are operated at high temperatures, typically in the range of 773 – 1273K. This is to ensure that as much of the ammonia is converted to hydrogen as possible before it contacts the electrodes. SOFCs aim to produce only N2 in terms of nitrogen containing compounds, so that no dangerous NOx are present. [3] These fuel cells are either used with an oxygen-conducting electrolyte or with a proton conducting electrolyte. A positive for SOFCs are that they can be operated with ammonia, but they could also be adapted to use other hydrogen containing compounds, such as hydrogen, biogas, bio-ethanol and bio-methanol. Flexibility with fuels is always useful when using products which could be subject to price fluctuations [3]. This variety of possible fuels for use should also be beneficial in speeding up research on areas beneficial to all SOFCs in general, like fabrication. As solid-oxide fuel cells require high temperatures, they are suited to being used in place of traditional generators in the process industry, where there is excess heat available. For use in the home, the high temperatures would prove a safety concern, as would any potential leakage of the ammonia, so much care would be required. Proton-exchange membrane fuel cells are designed to be operated at room temperature, which makes them an option for use in the automobile industry. These fuel cells cannot utilize ammonia directly, as it poisons the catalyst by depositing nitrogen compounds over time. Therefore, it must be coupled with an ammonia electrolytic cell which can provide the hydrogen stream for the PEMFC. As stated above, these fuel cells have applications in motor vehicle transportation. Alkaline-membrane fuel cells are operated in a very similar fashion to traditional AFCs, which is by the transfer of OH- ion through the electrolyte. These fuel cells are favourable because they do not allow the hydroxide ions to react with carbon dioxide, which would precipitate K2CO3 and NA2CO3 and decrease the efficiency of the fuel cell. Having fuel cells that are compatible with CO2 makes the system much more manageable, and here it allows for running using just air and ammonia [5]. This type of fuel cell was investigated in Ref. [6] using a nickel anode and a MnO2 cathode. Their ability to be operated using normal air means that all they require to start is a supply of ammonia, and the modest operating conditions, compared to SOFC, make them a realistic option for use in many home appliances and in motor vehicles. The goal with ammonia fuel cell technologies should be to eventually replace fossil fuels use in many industries. The main by-product from the cells is water, and research is being done into sustainable methods of ammonia production, which would allow the industry to be potentially carbon neutral. Theory All fuel cells consist of the same basic ingredients; they have two electrodes, a positively charged anode and a negatively charged cathode, and an electrolyte which carries the charged particles between electrodes. This can differ depending on the type of fuel cell, but the moving electrolyte is typically a proton (H+) or a hydroxide ion (OH-). Some cells use membranes to selectively allow certain compounds between the electrodes whilst preventing others from traveling through, to enhance the efficiency. The fuel cells discussed here, primarily Alkaline fuel cells (AFCs), proton exchange membrane fuel cells (PEMFCs) and solid-oxide fuel cells (SOFCs) are all operated under quite different conditions. The traditional ammonia based fuel cell does not rely on hydrogen coming into the cell. The ammonia reacts directly at the anode, where it reacts with the hydroxide ion and loses 3 electrons. Meanwhile oxygen is being added to the cathode where it reacts with water in the cell and the 3 electrons to form the hydroxide ion. In these fuel cells, it is the hydroxide ion which is transferred through the cell. Ammonia is one of the world’s largest produced basic chemicals, and in 2010, 159 million tonnes of ammonia were produced worldwide. As shown in figure 1 [3] [7], almost a third of this was produced in China, India had nearly 9%, Russia nearly 8% and the United States almost 6.5%. The basic infrastructure for mass-producing ammonia is already in place in many countries.  Figure 1: Worldwide ammonia production [3] Therefore, these countries will not have to worry about developing completely new infrastructure for ammonia transportation, and plants for ammonia synthesis, they can expand on their current systems, and make them more efficient to increase their supply of ammonia. The most well established method of producing ammonia is to recover hydrogen from hydrocarbon plant gases (i.e. Natural gas, LNG etc.) using steam reforming, then combine this hydrogen stream with a nitrogen stream producing ammonia using the Haber-Bosch process. (1). This method has been around for over 100 years, meaning it is well practised and optimised, the trouble with relying on this method for increasing the supply of ammonia is that the initial step relies on using hydrocarbons with a global-warming potential. This method can be suitable for now while ammonia is still in the research stage for being used commercially in fuel cells, but if we are aiming to become carbon neutral other avenues must be addressed. To do this it is necessary to source hydrogen from sources other than carbon containing compounds. Using renewable energy such as wind/solar and the electrochemical synthesis of water to create hydrogen would be an ideal way of producing hydrogen without the emission of CO2. There are a few other methods that have been suggested. One of which utilises biological systems, where molecules of nitrogen can be converted directly to ammonia without the need for extreme reaction conditions [8-10]. This process is natural, has been developed over billions of years, and utilizes proteins called metalloenzymes. The enzyme in question here, nitrogenize, has molybdenum and iron which help catalyse the reaction between the atmospheric nitrogen, the protons and the electrons in the air to form ammonia [9,10] [3]. The Haber-Bosh process suffers from several drawbacks that make electrochemical synthesis of ammonia look more favourable as a way forward. Firstly, it must take place under extreme reaction conditions, at pressures of 150-300 bar, and temperatures around 500 . The reaction is also exothermic, so increasing the reaction temperature will result in a decreased yield of ammonia, while decreasing temperature will negatively affect the rate of the reaction. The conversion of ammonia, which is around 10-15%, is thermodynamically limited [11], the pollution from the process is significant, and it requires a large amount of energy [12,13]. Another benefit of electrochemical synthesis is that it has been illustrated that water in the form of steam can be used instead of hydrogen, which would simplify the process [14]. Large quantities of ammonia are also allowed to simply be emitted into our atmosphere with little effort to try and capture it. Nearly 54 million metric tons of gaseous ammonia escape into the atmosphere every year. The major sources of this are domestic animal excreta (40.2%), synthetic fertilizers (16.7%), burning of biomass (10.9%) etc. It is surely feasible to be able to capture a significant quantity of this ammonia before it is released into the atmosphere [3]. Hydrogen appears to be the ideal fuel for a green economy. Its only by-product in either a fuel cell or in combustion is water. But the drawbacks of using pure hydrogen as a fuel have seriously hindered its development as the next energy revolution. Storing a sufficient amount hydrogen to be used for example in automobiles is proving to be a big challenge. It is just not possible using todays best technologies to store the required amount of hydrogen in a car. Due to its tremendously low density, hydrogen has 25% less energy content than gasoline based on volume. For this reason, there is no technology today that can store a quantity of hydrogen that allows for the 482km range that the average internal combustion engine of today manages [15]. In terms of raw energy density, Liquid ammonia has 11.5MJ/L, liquid hydrogen has 8.491 MJ/L (which only exists under extreme cryogenic conditions, temperatures below -250 for around atmospheric pressures.), and compressed hydrogen which has a value of 4.5 MJ/L at 690 bar and 15 . For the near future, it is not possible to store a sufficient quantity of hydrogen on board a car and still maintain the ranges/freedom of modern cars. In terms of the transportation and storage of the fuels, research into this area showed that ammonia suffers far less energy losses than both gaseous and cryogenic H2 [2]. To transform the hydrogen into ammonia, and transport it through 1 mile of pipeline resulted in 84.9% efficiency of energy transport. Transportation as a gas through a hydrogen pipeline for 1 mile, then compression to necessary 690bar resulted in 72.5% efficiency. In comparing the energy losses experienced when each fuel is liquefied and stored, to synthesize ammonia, liquefy it and then store it also results in 84.9% efficiency of energy transport, while to liquefy hydrogen and then store it results in only 53.9% efficiency [2]. Therefore, in terms of transportation and storage, ammonia is shown to be favourable when compared to pure hydrogen. Another positive for ammonia’s use as a fuel is that it already has extensive transportation infrastructure in place. In the united states alone, there are over 4800km of ammonia pipeline, which could potentially allow for distribution costs on par with liquid petroleum gasoline (LPG) [15]. There are safety concerns surrounding ammonia’s widespread use that must be addressed. It is toxic to humans, if inhaled directly can be fatal, and it is a corrosive substance. Direct contact can lead to severe burns [3]. OSHA (Occupational Safety and Health Administration) has set an 8-hour exposure limit of 25ppm and a short term (15-minute) exposure limit of 35ppm for ammonia in the workplace. NIOSH (The National Institute for Occupational Safety and Health) recommends that the level in the workroom air be limited to 50ppm for 5 minutes of exposure [16]. Ammonia has a very narrow range of flammability, 16-25% by volume in air, making it less likely to explode than traditional fuels like gasoline, with flammability occurring around 1-7% by volume in air [17]. With ammonia, it is detectable at quite low concentrations due to its pungent odour, with the average odour threshold around 5ppm [3]. This means that ammonia would be easily detected, whether it was in a car, or coming out of a fuel generator in the home, giving someone time to clear the area before any damage could be done. There has also been research into the application of metal amines with low ammonia partial pressure in fuel tanks to try and combat the risk of large-scale ammonia release [18]. Anhydrous ammonia is also lighter than air, so depending on the weather conditions at the time, one would expect it to disperse into the atmosphere, not pool on the surface. From Ref. [2], ammonia would be just as safe to use as a transportation fuel as gasoline is today. However, one drawback about ammonia’s mass production as a fuel would be that, in at least the United States, it is still classified as a substance that is toxic by inhalation. This means that it requires a hazardous safety permit to transport quantities over 13,248 L [3] Ammonia-based fuel cells As mentioned in the introduction, the first type of fuel cell introduced to handle ammonia as a fuel was a traditional alkaline fuel cell. In an ammonia fuel cell, the anodic oxidation reaction occurs per equation (2) [4], In a complete ammonia-oxygen fuel cell, this reaction is coupled with a reduction in oxygen at the cathode, equation (3) [4], These cells work based on the idea that the ammonia can be oxidised directly into nitrogen and water using the hydroxide ion given by the electrolyte solution, and the oxygen stream makes up the hydroxide again at the cathode. Ammonia fuel cells can be potentially run at room temperature; however, they do suffer from low power density. This can be related to low catalytic activity of the electrodes, so finding catalytic materials which exhibit higher activities at lower temperatures is an important step moving forward [4]. Work done by Simons, Cairns and Surd [4] was centred around both ammonia-oxygen and ammonia-air cells in the late 1960’s. They operated fuel cells between 25 and 140 , with an aqueous potassium hydroxide solution as the electrolyte. In their paper, they found that for the ammonia-oxygen cells, they exceeded reports for all other fuels besides hydrogen and hydrazine. When run with air instead of pure oxygen, the fuel cell suffered from failure due to the presence of carbon dioxide in the system. The durability of these fuel cells is a big issue, due to reactions that take place between the KOH electrolyte and CO2, which causes precipitates to form on the cathode and eventually leads to fuel cell failure. These flaws were discovered over 168 hour runs of the fuel cell, which corresponds to 1 weeks running. This flaw in the fuel cell can only be dealt with by the removal of CO2 from the oxygen stream. This requires access to a pure oxygen stream, or the scrubbing of carbon dioxide from the air before it enters the fuel cell. Tests done using caesium hydroxide as the electrolyte also suffered from a similar problem with the creation of caesium carbonate, however in this case the carbonate did not precipitate onto the cathode. It did however lower the conductivity of the system, which also eventually caused fuel cell failure. The performance was markedly lower in both cases as time went on. Another possible avenue to overcome this issue would be more research into potential electrolyte solutions which would not form carbonate ions in the fuel cell, however little evidence of this research has been discovered. Another issue found in the experiment was that continued operation of the fuel cells appeared to alter the wettability of the entire electrode due to the exposure to the hot alkali solution. This eventually lead to a marked decay in the performance of the electrode. There does not appear to be any simple ammonia alkaline fuel cell that has been developed to overcome the problem of CO2 deteriorating the performance over time. There does seem to be potential here for direct oxygen supply being used, or air that has had its carbon dioxide removed. There are issues still associated with this however, that it is obviously more expensive to have to treat the incoming air instead of using it directly, which diminishes the economic potential of the fuel cell. There appears to be a shift in the research in the direction of AMFCs, PEMFCs and SOFCs. Alkaline membrane fuel cells are essentially an advancement on this idea, which aims to cut out the problem of CO2 degrading the fuel cell whilst still utilising ammonia directly as a fuel. Solid-oxide fuel cells Solid-oxide fuel cells can utilise ammonia directly as a fuel, as they are operated at high enough temperatures for the thermal decomposition of ammonia into hydrogen and nitrogen to occur. This means that SOFCs typically must operate at temperatures above 500C, which is the point where the reaction (4): starts to become significant [19]. SOFCs can be run both with an oxygen-ion conducting electrolyte (SOFC-O) or with a proton-conducting electrolyte (SOFC-H). It was found that for direct ammonia fuel cells, the power density is only slightly lower than that of a fuel cell that utilises hydrogen directly [20]. In SOFC-O cells, the electrochemical reactions that take place if ammonia reaches the anode are; (5) (6) (5) will be the rate-limiting step due to the slow rate of diffusion of oxygen through the cell. Due to this there will be a small amount of NO produced at the anode [3]. Figure 2: A schematic description of the SOFC-O ammonia fed fuel cell [3] As illustrated in figure 2, in this configuration the ammonia enters the anode side, where it is decomposed into hydrogen and nitrogen due to the high temperatures of the fuel cell. Any ammonia which is not decomposed before it reaches the anode will produce the reaction (5) and (6). The hydrogen atoms will react on the anode, where they lose an electron, and react with the oxygen ions which have taken electrons from the cathode, to form water. One problem with this configuration is that the inert nitrogen gas produced at the anode will dilute the fuel concentration, which lowers the reversible cell potential. To negate these effects requires sufficient understanding of the thermodynamics equilibrium of the decomposition reaction at the set operating conditions [3]. When operated at higher temperatures, the ammonia decomposes very quickly in the cell into hydrogen and nitrogen, so one would expect very high concentrations of hydrogen relative to ammonia, so any formation of NO can be expected to be negligible. SOFC-H fuel cells have a higher theoretical efficiency maximum efficiency compared to SOFC-O’s. The schematic differences between the two can be seen by comparing figure 1 and figure 3. Figure 3: schematic of SOFC-H fuel cell. Again, ammonia is decomposed to form hydrogen and nitrogen, and while on the anode side nitrogen acts as an inert diluent, the hydrogen atoms transfer to the electrolyte interface, they are oxidised and travel across the electrolyte as protons. Near the cathode interface they form water with the oxygen that enters the cathode by electrochemical reaction. Here we can see the main advantage of this configuration of fuel cell, that while water is still produced in the reaction it does not act as another diluent for the fuel, as it is formed on the cathode side, and can simply leave with the oxygen stream. The cells can be made more efficient by increasing the transfer of ions through the electrolyte. The conductivity of electrolytes is exponential with respect to increasing temperature, and is inversely proportional to the thickness [21]. One of the main aims of research into SOFCs is the reduction of the operating temperatures while maintaining power density. The development of a solid electrolyte with higher conductivity at lower temperatures will therefore be dependent on that material also having good mechanical stability to work effectively. Another issue faced by ammonia solid oxide fuel cells is the electronic conductivity of the ammonia synthesis catalysts. No suitable catalysts have been developed, which presents an obstacle that must be overcome before ammonia SOFCs can be commercialized. If the ammonia can be decomposed quickly enough, the performance of ammonia SOFCs approaches the performance of direct hydrogen fed SOFCs. As the decomposition of ammonia is mildly endothermic, increasing the temperature will yield a faster rate of reaction, and result in a greater conversion into hydrogen. The reaction rate increases remarkably when the temperature moves from 673K to 1273K, from approximately 1 to 3.6 x respectively [22]. This means that by operating at high temperatures, one can ensure ammonia decomposition will occur before it has time to reach the anode-electrolyte interface. The study provided in [3] gave an indication of the level of ammonia decomposition at 673K, 773K and 873K. It was found that at 673K the decomposition was insufficient, less than 10% at the anode. This was taken to be too low a temperature for operation of a SOFC-H fuel cell, unless there was a very catalytically active material for ammonia decomposition discovered. However, at 773K there was a marked improvement in the decomposition of ammonia, although it was still not quite at 100%. Even without 100% decomposition, this means that a proton conducting solid oxide fuel cell could operate under these conditions with acceptable cell performance. At the higher temperature of 873K, 100% conversion of ammonia in the cell to hydrogen and nitrogen was observed [23]. Due to the high temperature requirements of these fuel cells, and the long start-up time associated with this, they are unlikely to be of use in personal automotive transportation. Though in other modes of transport, for instance in planes, or ships, this start up time wouldn’t be as big an issue and the high temperatures would not pose the same problem. The main area where SOFCs could be utilised appears to be in combined heat and power plants, where excess heat given off by other equipment can be used to power the SOFC. They can also be coupled with heat engine recovery devices. There are still many advancements that must be made before a commercialized SOFC can be realised. Probably the most important one is the durability of the materials used in construction. Due to the problems with operating at high temperatures, much of the research into SOFCs in the last 10 years has been focused on reductions in the operating temperatures [24]. The problem is in designing fuel cell anodes with materials which can be operated at temperatures at least around 800K, that are thermally stable as the fuel cell is taken from room temperature up to operating temperatures, which do not degrade upon successive reduction and oxidation cycles (redox stability), and do not form non-conductive phases at electrode-electrolyte or electrode-interconnect interfaces (chemical stability). These issues present significant areas where research must be done to develop more suitable materials. In order of importance however, the one which probably needs the most attention would be redox stability, as many compounds have failed when put through repeated oxidation and reduction cycles, with post-mortem analysis showing sign of degradation, and changes in their structure, volume and in their TEC (thermal expansion coefficient) upon reduction [24]. There is yet to be a material developed which meets all three of these criteria and could be used in a commercial SOFC. On a more positive note, one of the issues associated with SOFC operation using typical hydrocarbons like ethane, methane, propane, methanol etc. is the coking of the anode in the fuel cell. This is where upon oxidation of the fuel, a layer of carbon is deposited onto the anode, leading to a serious decline in performance. With the use of ammonia in these systems however, this problem gets negated as there are no carbon atoms present. This is a major positive in the further development of SOFCs with ammonia as their fuel, as research attempts into replacements of the traditional Ni anode to try and stop coking from occurring have not been very successful [25]. The large number of possible fuels that can be used in SOFCs can only be a positive for ammonia SOFC, as it will lead to more research into the general construction/fabrication of these fuel cells, which in the long run will benefit ammonia fuel cells.  Proton-exchange membrane fuel cells (PEMFCs) PEMFCs have been developed as a low temperature fuel cell. This type of fuel cell is operated with a thin, permeable electrolyte sheet, an example of which can be seen below. Figure 4: Schematic of a PEMFC [26] Their applications are primarily being investigated for use in motor vehicle transportation. These fuel cells cannot utilise ammonia directly, as it would poison the Pt/C anode catalyst and react with the acidic Nafion membrane. However, ammonia can be utilised with these fuel cells indirectly by coupling the PEMFC with an ammonia electrolytic cell. The idea is that once the PEMFC is running, part of the power generated from the cell will be used to run the AEC, making it a self-sustaining system. The AEC can provide the pure hydrogen stream to the PEMFC that it requires to run. Unlike more traditional battery powered vehicles, hydrogen fuel cell vehicles HFCVs can operate in cold conditions, making them more wide reaching. They also have high efficiencies, and using ammonia the car will not emit any harmful pollutants such as NOx, VOCs, SO2, and CO2 which cause serious harm to human beings when not regulated properly. As well as being better for the environment, HFCVs emit little noise pollution, and they convert the energy in the hydrogen to power with a much higher efficiency (50-60%) compared to the efficiencies of today’s typical hydrocarbon run vehicles (20-30%) [15]. Theoretically, electrolysis of ammonia should require 95% less energy than the electrolysis of water, specifically 1.55 Wh g-1 H2 compared to 33 Wh g-1H2 [15]. In a study done by Vitse et al [27], the cost per kilogram of hydrogen for each electrolysis was calculated and they found that for ammonia it was $0.89/kg H2 compared to $7.1/kg H2 for water. This illustrates that the electrolysis of ammonia appears favourable when compared to the electrolysis of water. For the overall electrochemical reaction to take place (7) , only 0.06V are required, which is significantly lower than the required 1.23V to produce the same quantity of hydrogen from water. Per the thermodynamics, for 1g of hydrogen ammonia electrolysis will use up 1.55Wh. From this 1g of hydrogen in a PEMFC, which is utilising the reverse water electrolysis reaction, it can theoretically generate 33Wh. This means that once the 1.55Wh are sent back to the AEC, the maximum net energy potential of the system would become 31.45Wh. This energy can then be used to meet any demands one has; in transportation, this would include both driving the motor and recharging the start-up battery in the car. The only problem with these numbers is that current PEMFCs run at between 50-60% efficiency, and that even though ammonia is transformed to hydrogen with 100% Faradaic efficiency, there are problems associated with large ammonia oxidation potentials occurring which decrease the overall efficiency [15]. One way forward for this technology; improving current density of the electrodes, which has been shown to significantly decrease the storage costs. Figure 5 [15]: how storage costs vary with electrode current density. As can be seen from this image, small increases in electrode current density can have a dramatic effect on the storage costs, measured in $/kWh. In the paper, they could achieve around 130 mAcm-2, However if 2200 mAcm-2 was achieved, the Department of Energy’s (DOE) storage cost requirements for 2010 could be met. While this seems like a very significant increase, the exponential nature of this curve means even an increase from 130 to ~260 would result in a very high decrease in costs. It would go from about 130 $kWh-1 to approximately 40 $kWh-1. There are still several areas of concern that must be dealt with before PEMFCs are ready for commercialization and there use in motor vehicles becomes economically viable compared to hydrocarbon-combustion vehicles. These fuel cells are typically used with an acidic membrane using a compound such as Nafion. This substance is not compatible with ammonia, and it can suffer from a decrease in cell efficiency from concentrations as low as 1ppm, as the H+ ions are replaced by NH4+ ions both on the catalyst and in the bulk membrane, which leads to a reduction in conductivity [6] [28]. This rules out the use of a cracking reactor in motor vehicles, because to ensure such a complete reaction of the ammonia would require incredibly high energy inputs and significant storage space. This method would therefore seem impossible to incorporate into modern transportation, especially individual modes of transport like cars/motorcycles. That the PEMFCs discussed here must be coupled with an AEC adds to the complexity of the overall process, as it requires further research into both areas. The ammonia would also poison the Pt/C anode catalyst in PEMFCs, so it does seem to be unlikely that developments in materials could lead to a PEMFC that could run on ammonia directly, which will mean that its efficiency is lower due to the need to send some of its own electricity back into an electrolytic ammonia cell. The use of a Pt/C catalyst occurs in almost all PEMFCs, which would be a high initial cost when looking to mass produce fuel cells. Coupling a PEMFC to a cracking reactor for ammonia, plus potentially purification equipment for ammonia scrubbing is an option. This poses significant energy problems however, and it would be questionable whether it was worth using a PEMFC for this job where a SOFC would seem better suited. The large temperature difference between the cracking section and the PEMFC would undoubtedly lead to further energy losses. One of the main issues in a PEMFC is the water management. The water produced at the cathode side is very difficult to move out of the cell, which compromises the transfer of the oxygen to reaction sites on the cathode [29]. In a PEMFC, the rate limiting steps are the reaction of oxygen at the cathode, or the oxygen reduction rate (ORR), and the transfer of oxygen through the water in the cell [29]. These two are combined to be the limiting step, as the reaction is delayed by both the slow diffusion of the oxygen to the cathode, and the slow adsorption onto the catalytic surface. Compared to the rate of oxidation of the hydrogen at the anode side, the ORR is 4-6 orders of magnitude smaller, despite research made into improving catalysts for this reaction. The ionic conductivity of the membrane depends strongly on the level of water inside the system, with higher conductivities seen as the humidity of the system increases. So too low a water content in the system, and the conductivity of the electrolyte decreases, it can result in the dehydration of the membrane, which would degrade performance [29]. Having too high a concentration of water inside the system results in a decreased rate of both the oxygen reduction reaction, and the transportation of oxygen throughout the cell, while if one removes too much water then it risks dehydrating the membrane which results in fuel cell degradation. This then becomes an issue of maintaining water levels to make sure that it can be run properly, which poses a serious challenge. In Figure 6 [29]: graph of cell voltage against current density with varying levels of flooding. It has been shown that by allowing the level of flooding in the cell to increase, it decreases the total voltage the cell can produce at higher current densities. No data was provided on what the actual level of flooding was for each curve, but operating as close to ‘no flooding’ as is possible without damaging the membrane will maximise the cell voltage obtained at a given current density. Alkaline-membrane fuel cells These fuel cells are an advancement on traditional alkaline fuel cells, where they aim to deal with the primary concern of carbon dioxide from the air poisoning the catalyst. These fuel cells have some advantages on proton exchange membrane fuel cells, in that they don’t require precious metals to be operated (Like platinum in traditional PEMFCs). AMFCs can be operated with more inexpensive materials as the catalysts, like nickel, silver and manganese oxide. Common alkaline membranes used today are based on organic quaternary ammonium hydroxides linked to polymers [30]. In using these membranes, it has been reported that carbon dioxide in air can no longer poison the anode, and that its introduction at the cathode can even lead to an improvement in the fuel cell performance [31]. One study into AMFCs available is investigating room-temperature direct ammonia fuel cells with alkaline electrolyte membrane. They used Ni and MnO2 as the catalysts [6]. Figure 7: General configuration of an ammonia based alkaline membrane fuel cell Ammonia is introduced at the anode, where it is converted into hydrogen and nitrogen. The hydrogen reacts with the OH- ions and forms water, losing electrons while the water and unreacted nitrogen leave the cell. Water and air come in the cathode side, where they form hydroxide ions which can travel through the membrane and react on the anode, while the membrane stops the carbon dioxide from travelling across the cell and depositing onto the anode. Figure 8: Cathode, anode and overall electrochemical reactions in AMFC when utilising ammonia as fuel [6] (8) (9) (10). This illustrates that the overall theoretical open-circuit voltage (OCV) is 1.17 V, which is only just below the OCV for a fuel cell that was using hydrogen directly as fuel (1.23V) [32]. Like the problem discussed in the PEMFC section, water flooding is one of the central issues that they must deal with before the fuel cell can be operated for an acceptable amount of time. In alkaline membrane fuel cells, this problem is slightly different as the water isn’t being produced at the cathode side, but at the anode side. So, while the cathode doesn’t suffer as much from flooding, there is dilution of fuel at the anode, and now the anode is in danger of being flooded. It was recommended more research should be done into this issue [6]. Lan and Tao demonstrated that using inexpensive membrane electrolytes and non-noble catalysts, it was possible to run an alkaline-membrane fuel cell with ammonia that was on par with a hydrogen fuel cell. They also found that using nano sized nickel catalysts, they could exhibit higher catalytic activity than was possible using more expensive catalysts containing platinum. Even with an increase in the loading of the catalyst required, the result was an overall reduction in the cost of catalyst despite improvement in activity [6]. They concluded that with further optimization, this type of fuel cell has the potential to power electric vehicles in the future, once a higher power density cell is developed, and once the safe control of ammonia in the fuel cell is possible. Alkaline membrane fuel cells appear to be the most innovative technology in terms of direct ammonia fuel cells. They are operable at room temperatures and pressures; they require no extra purification equipment and they don’t suffer from the same drawbacks as alkaline fuel cells in that they can process CO2. This is due to the absence of free cations such as K+, therefore not allowing carbonate salts to occur on the electrodes which degraded the fuel cell [28]. There are still many issues faced by this type of fuel cell before any commercial application can be realised. Suzuki et al [28] have found that ammonia permeation through the membrane had a terminal effect on the electrodes. This is due to the presence of nitrogen adsorbing species (Nad), which deposited onto the electrodes and lead to a decline in performance with time. The theory behind this mechanism is that both NH2ad and NHad get produced at the Pt electrode through sequential dehydrogenation of the ammonia. The next step is that they are combined to form the species N2Hxad (x = 2 - 4). Lastly, by further dehydrogenation of these species, N2 is produced at the cathode, where it adsorbs and blocks active sites on the catalyst [28]. In this experiment, they were also investigating how different anode catalysts affected the performance of the fuel cell using both ammonia and hydrogen as fuels. They discovered a marked difference between the hydrogen fuelled cells and the ammonia ones. The reason put forward for why this result had not been discovered before was that other experiments were not running at high enough voltages for this effect to be noticeable. When running at low voltages, reductions in performance are much more difficult to determine, so other experiments did not note this effect. Figure 9: open circuit voltage for AMFC fuel cells against time, for hydrogen and ammonia fuels [28]. This graph shows that when ammonia was introduced as the fuel at 10 minutes into the test, there was an immediate drop in performance. The level of the drop depended on the type of catalyst used on the anode, and the order of OCV for the anode materials was Pt-Ru/C > Pt/C > Ru/C. This difference in OCV was not observed during the initial stage of the test with hydrogen, so the difference is down to the catalytic activity of the anodes for ammonia oxidation. The reduction in the OCV will also be dependent on the permeability of ammonia through the membrane, so tests were run on the exhaust gas from the cathode side using mass spectroscopy. The mass intensity of both N2 and NO were measured in the exhaust gas from the cathode, again by running the fuel cell with hydrogen for 10 minutes then switching to an ammonia fuel.  Figure 10: NO and N2 mass intensity against time from cathode exhaust gases [28]. m/e = 28 represents nitrogen and m/e = 30 represents nitrogen monoxide. What this graph illustrates that the introduction of ammonia as a fuel at the anode leads to the formation of nitrogen-containing compounds on the cathode side of the fuel cell. Therefore, it is proof that the current alkaline electrolyte membrane is permeable to ammonia, and that the ammonia is oxidised on the cathode side to create both nitrogen and nitrogen monoxide. This permeability coefficient was calculated using (11), where P is the permeability coefficient, l is the membrane thickness, and p is the partial pressure of the target gas on the supply side (negligible partial pressure on other side of the membrane). The target gases here were ammonia and hydrogen. J is the gas flux through the membrane, and in this experiment, it was calculated using (12), where n is the obtained molar quantity of permeated gas, t is the sampling time, and A is the contact area of the membrane with the supplied gas. Using these two equations, it was found that the permeability of ammonia under wet conditions was 1.26 * 10-6 [cm3 (STP) cm cm-2 s-1cmHg-1], which was approximately 2000 times higher than the value calculated for hydrogen. The equations above are dependent on the gas permeation obeying the solubility-diffusion mechanism, where the permeated gas flux varies linearly with the partial pressure gradient over the membrane p. This condition was met under both wet and dry conditions, therefore confirming that the permeation of ammonia across the membrane follows the solubility-diffusion mechanism [28]. The molar flux of ammonia was found to increase linearly with its partial pressure under wet conditions, while under dry conditions it remained relatively constant. Under dry conditions, the amount of ammonia which permeated through the membrane was around 60 times smaller than under the maximum humidity condition [28]. Perhaps by altering the humidity of the ammonia fuel, it would be possible to maximise the OCV in the cell without allowing too high a flux of ammonia through the membrane. The performance of the fuel cells was found to degrade with repeated running, which was observed to be caused by the accumulation of adsorbed nitrogen over the anode catalyst [28]. However, it was found that by running the fuel cell using hydrogen fuel, one could completely recover the performance from before ammonia was added. Therefore, the Nad species can be reductively stripped from the anode by running the fuel cell with hydrogen. They conclude that the accumulation of Nad over the anode did reduce the cells performance, but that reductive desorption of Nad into the hydrogen fuel could recover the performance of the fuel cell [28]. This would suggest that it could be possible to maintain an AMFC if a mixed supply of both ammonia fuel and hydrogen were used, however this was not mentioned in the work above. The further development of anode catalysts remains a priority for alkaline membrane fuel cells, and research into electrolyte membrane materials should be done to try and limit the permeability of the ammonia fuel between the anode and cathode. Future for ammonia-fuel cells Alkaline membrane fuel cells – The possibility of using a mixture of hydrogen and ammonia as combined fuel should be researched more heavily. Work should be carried out on finding a ratio between the two fuels which could be used to optimise the fuel cell while retaining the activity of the anode. Here control could come in, measuring the performance of the anode and altering the composition of the incoming fuel to maximise performance. There is potential for coupling with partial ammonia electrolysis, or with an incomplete cracking of ammonia, which would require less intensive conditions than the cracking required in SOFCs. The conditions in these two reactors would be altered to deliver the optimum mix of fuel to the AMFC. Most experiments have been run at around atmospheric temperature, which is part the cause of the relatively low power output of the cells. This suggests that research into catalysts with better activity at lower temperatures should be done. Running these fuel cells at slightly higher temperatures would possibly overcome some of these issues, preferably without making the cost of equipment much more expensive. I think research into operation just above 100 could be viable, to attempt to minimise the effect of water flooding in the cell, by converting at least some of this water into steam which should decrease the effect it has on the electrodes. Membranes must be developed for these fuel cells which limit the penetration of ammonia between the anode and the cathode. Proton exchange membrane fuel cells – These cells suffer primarily from water flooding which lead to declines in performance. Here optimisation and control could play a role in finding the ideal water content in the cell to keep the ionic conduction of the membrane high without resulting in water deposition onto the anodes and accumulation of water in the cell limiting the rate of the electrochemical reaction. Research would be required into the specific quantity of water that is allowable in the cell. Work must be done into better anode/cathode materials, with possible emphasis on making them more hydrophobic, and ensuring that they do not change their wettability towards water with time. Current set ups of PEMFCs connected to an ammonia electrolytic cell require ~60% of the output from the fuel cell to be redirected to the AEC. As research is being done into both areas, advancements in this technology can be expected to occur faster than usual, as any improvement in either area results in the combined set-up being more efficient. Based on the theoretical output from the system as discussed in the PEMFC chapter of the report, the theoretical output of the PEMFC AEC system would be 31.45 Wh g-1 H2. Accounting for the fact that PEMFCs usually cannot operate at higher than 60% efficiency, the output from the system could still be as high as 18.87 Wh g-1 H2. Solid oxide fuel cells – These fuel cells can operate using ammonia directly as the fuel, and because of the high temperatures of the operation, it is possible to completely transform the ammonia into hydrogen and nitrogen before it reaches the anode, meaning that it combines ammonia cracking with a hydrogen fuel cell. This is important as it avoids many of the issues faced by the other fuel cells, which is that ammonia appears to degrade fuel cell electrodes over time [28]. The problem with these fuel cells, as briefly stated in chapter 4, is that they are not currently durable enough for commercial use. Advancements in material technology are required before they could be operated for long periods confidently, and fabrication costs would also need to drop. The SOFC-H fuel cell does seem to be the more promising of the two fuel cells, due to the lack of fuel dilution at the cathode by water. If a redox stable anode could be developed, and improvements be made for the ceramic components inside this fuel cell, then their full commercial use could be realised [3]. Alkaline fuel cells – These cells suffer from many issues as discussed in the AFC section of the report, along with their inability to be run with carbon dioxide [4]. In terms of research into ammonia fuel cells, it appears that alkaline membrane fuel cells are the natural progression of this technology. Both PEMFCs and AMFCs require more research before they can realistically be used in fuel cell vehicles, but if the issues faced by these fuel cells could be overcome, they both provide an opportunity to replace traditional hydrocarbon vehicle. SOFCs require too high a temperature to realistically be used in the home/car. Their applications should be in industry, where they could prove a viable replacement for traditional generators, and can use any waste heat from the processes. Conclusion This report has discussed the main technologies currently in use for the utilisation of ammonia as a fuel. The need to diversify from fossil fuels has been briefly discussed, and ammonia’s potential uses as a fuel due to its high concentration of hydrogen have been examined. Hydrogen is, on the surface, the ideal fuel for a green economy. However, as mentioned in the introduction, due to its incredibly low energy density and the issues associated with its efficient transportation, it does not seem suitable given our current energy requirements. Ammonia, therefore, is put forward as a more efficient way of transporting hydrogen as a fuel, as it is useable under much less extreme conditions of temperature and pressure [2]. The fuel cells investigated were; Alkaline fuel cells, solid-oxide fuel cells, proton exchange membrane fuel cells and alkaline membrane fuel cells. The main fuel cells of note from research were SOFCs, PEMFC and AMFCs. They all offer promise for the future, but all suffer from drawbacks that are currently holding back their commercial potential. The one which would appear closest to actual operation would have to be the proton conducting solid oxide fuel cell (SOFC-H). This fuel cell has been shown to be operable at higher efficiency than its oxide conducting counterpart (SOFC-O), due mainly to the lack of water diluting the fuel on the anode side [3]. This fuel cell acts as an ammonia cracker because it is operated at high enough temperatures to facilitate the reaction, This has both positives and negatives aspects. Because the ammonia is cracked before it has time to reach the anode, there is no chance for it to react to form NO, which would be a pollutant, and no deposition of nitrogen containing compounds from ammonia on the anode is expected, which is a problem that AMFCs suffer from. The drawbacks of high temperature operation are one of the fundamental problems with these fuel cells. There are no materials currently existing that can handle the high temperatures of these reactors, and exhibit all the necessary attributes to be a sufficient catalyst. These include problems with chemical stability, and especially the issue of redox stability; many materials that have been put forward suffer terminal degradation during successive reduction and oxidation cycles [3]. However, in comparison with the other fuel cells, the SOFC-H would appear to be closest to a commercial application. The PEMFC is not operable with ammonia directly, as discussed in its chapter in the report. Due to this, for motor vehicle transport the idea is to couple a PEMFC with an ammonia electrolytic cell (AEC) which can take in the ammonia, and will produce a pure hydrogen stream that is sent to the fuel cell to produce the energy for the car. The drawback of this arrangement is that the energy efficiency of the fuel must obviously go down as a part of the output must be sent back into the AEC to make it self-sustaining. Having a self-sustaining fuel cell running a vehicle would be ideal, but for the moment they still suffer drawbacks such as water flooding, where the excess water in the system slowed the reaction rate by affecting both the rate of oxygen diffusion to the cathode, and the actual rate of the reaction at the cathode [29]. For wide-spread use, the fuel cells would also need to be made much more durable, being able to take the strain of operation in a moving vehicle [3]. Alkaline-membrane fuel cells: These fuel cells are quite like PEMFCs, however with a few details differing. Catalysts for AMFCs have been developed using cheaper metals than found in PEMFCs. This is a positive when thinking of attempting to mass-produce the fuel cell, as it can drastically lower the initial capital investment. There will be dilution of fuel on the anode side, and now the anode is in danger of suffering from water-flooding [6]. The authors recommended more research should be done into this issue. Permeation of ammonia through the membrane was also found to have a terminal effect on the cathode. Some of the side effects of this problem were illustrated in figures 8 and 9, which showed how the introduction of ammonia as a fuel caused a marked reduction in the voltage of the cell, and lead to the production of nitrogen containing compounds (NOx and N2) at the cathode side respectively. Moving forward, more research into ways of mitigating the effect of water flooding is crucial, to both AMFCs and PEMFCs futures [29]. If this problem could be dealt with then both fuel cells would be a significant step closer to commercial viability. Ammonia penetration through the membrane is another problem to be addressed. This requires more research into the membranes used for these fuel cells. You want to maximise the ionic conductivity to increase the speed of the reaction, but as Ref. [28] showed, one must try and minimise the permeation of ammonia into the cathode side of the fuel cell. It leads to reductions in efficiencies and creates dangerous NOx. Once on the cathode side of the fuel cell, the ammonia deposits nitrogen compounds onto the catalyst, leading to reductions in efficiency with time. It was found that if you switched to hydrogen fuel for a few minutes, the fuel cell would recover its efficiency completely. This indicates that the hydrogen is causing these adsorbed compounds to desorb back into the fuel cell, and is in effect regenerating the fuel cell. Here I believe that this effect should be studied further, to determine if a mixture of ammonia/hydrogen could be used in the cell without any declines in performance. There are potential areas where optimisation and control could play an important role in the running of ammonia fuel cells, some of which were discussed in the future of ammonia fuel cells chapter. However, the research doesn’t seem to be focused on this area, as there are obstacles that must be overcome with all of them before this area becomes more significant. Once these fuel cells become more developed this area will be of increased interest in maximising the output from the cells. Research into fuel cells is incredibly important as we attempt to shift from our current, fossil fuel dominated energy systems, to environmentally friendly alternatives. As carbon fuels become more and more scarce, and cleaning up emissions becomes more expensive, other areas for energy generation must be explored. Ammonia fuel cells have the potential to produce carbon neutral energy, provided there is the necessary work done to ensure they can be mass-produced, and that renewable energies like wind/solar etc. play a key role in ammonia synthesis. The fuel cells discussed in this paper are all still at the experimental phase, but all of them are steadily getting more efficient, and the problems that hold them back are being investigated and mitigated. Hydrogen is an ideal fuel for a responsible and clean energy economy, and ammonia is one the best ways of utilising this fact in our current transportation infrastructure. It seems unlikely that ammonia fuel cells will be revolutionising our energy/transportation sectors in the next 5 years, but in the long term they could form a very important step in the development of new and innovative energy production. 9. References [1] NIST Webbook, isothermal properties of hydrogen. Accessed 28-01-17 [2] Olson, N., and Holbrook, J. (2007). Available at: http://www.powershow.com/view/5b55a-MWZjZ/NH3_The_Other_Hydrogen_TM_powerpoint_ppt_presentation. Accessed 25-01-17 [3] A. Afif, N. Radenahmad, Q. Cheok, S. Shams, J. H. Kim, A. K. Azad, “Ammonia-fed fuel cells: a comprehensive review”, Renewable and Sustainable Energy Reviews, 60, 2016, pp 822-835 [4] E.L. Simons, E. J. Cairns, D. J. Surd, “The Performance of Direct Ammonia Fuel Cells”, Electrochemical Science, 116(5), 1969, pp 556-561 [5] R. Lang, S. Tao, “Ammonia as a suitable fuel for fuel cells”, Front. Energy Res. 2(35), 2014, pp 1-4 [6] Lan, R., and Tao, S. W. (2010). “Direct ammonia alkaline anion-exchange membrane fuel cells.” Electrochem. Solid State Lett. 13, B83–B86. doi:10.1149/1.3428469 [7] Ammonia production by country [Online]. Available: 〈http://www.index mundi.com/minerals/?product 1⁄4 ammonia〉. [8] Rod TH, Logadottir A, Nørskov JK. “Ammonia synthesis at low temperatures.” J Chem Phys 2000;112:5343–7.
 [9] Schrock R. “Nitrogen fix.” Presented at the MIT Technology Review; 2006. [10] Skulason E, Bligaard T, Gudmundsdóttir S, Studt F, Rossmeisl J, Abild- Pedersen F, et al. “A theoretical evaluation of possible transition metal electro-catalysts for N2 reduction.” Phys Chem Chem Phys 2012;14:1235–45. [11] Marnellos G, Stoukides M (1998) “Ammonia synthesis at atmospheric pressure” Science 282:98–100
 [12] Liu R, Xu G (2010) Chin J Chem 28:139–142 [13] Xu G, Liu R (2009) Chin J Chem 27:677–680
 [14] Amar, I. A., Lan, R., Petit, C. T., and Tao, S. W. (2011). “Solid-state electrochemical synthesis of ammonia: a review.” J. Solid State Electrochem. 15, 1845–1860. doi:10.1007/s10008-011-1376-x
 [15] Boggs, B. K., and Botte, G. G. (2009). “On-board hydrogen storage and production: an application of ammonia electrolysis”. J. Power Sources 192, 573–581. doi:10.1016/j.jpowsour.2009.03.018 [16] A. f. T. S. a. D. Registry, AMMONIA; 2004.
 [17] https://www.ccohs.ca/oshanswers/chemicals/flammable/flam.html accessed: 18-01-17 [18] Klerke, A., Christensen, C. H., Nørskov, J. K., and Vegge, T. (2008). “Ammonia for hydrogen storage: challenges and opportunities.” J. Mater. Chem. 18, 2304–2310. doi:10.1039/B720020J [19] Perman, E., and Atkinson, G. (1904). “The decomposition of ammonia by heat.” Proc. R. Soc. Lond. 74, 110–117. doi:10.1098/rspl.1904.0091 [20] Meng G, Jiang C, Ma J, Ma Q, Liu X. “Comparative study on the performance of a SDC-based SOFC fueled by ammonia and hydrogen.” J Power Sources 2007;173:189–93. [21] Gellings PJ, Bouwmeester H. “Handbook of solid state electrochemistry.” USA: CRC press Inc; 1997. [22] Yin S, Xu B, Zhou X, Au C. “A mini-review on ammonia decomposition catalysts for on-site generation of hydrogen for fuel cell applications.” Appl Catal A: Gen 2004;277:1–9. [23] Ma Q, Peng R, Tian L, Meng G. “Direct utilization of ammonia in intermediate- temperature solid oxide fuel cells.” Electrochem Commun 2006;8:1791–5. 
 [24] Cowin, P. I., Petit, C. T., Lan, R., Irvine, J. T., and Tao, S. W. (2011). “Recent progress in the development of anode materials for solid oxide fuel cells.” Adv. Energy Mater. 1, 314–332. doi:10.1002/aenm.201100 108 [25] Ma Q, Peng R, Lin Y, Gao J, Meng G. “A high-performance ammonia-fueled solid oxide fuel cell.” Journal of Power Sources 161 (2006) 95–98 [26] http://americanhistory.si.edu/fuelcells/basics.htm Accessed 27-01-17 [27] F. Vitse, M. Cooper, G.G. Botte, “On the use of ammonia electrolysis for hydrogen production.” Power Sources 142 (2005) 18–26.
 [28] Suzuki, S., Muroyama, H., Matsui, T., and Eguchi, K. (2012). “Fundamental studies on direct ammonia fuel cell employing anion exchange membrane.” J. Power Sources 208, 257–262. doi:10.1016/j.jpowsour.2012.02.043 [29] Li et al. “A review of water flooding issues in the proton exchange membrane fuel cell,” journal of power sources 178 (2209) 103-117 [30] J. R. Varcoe, R. C. T. Slade, E. L. Yee, S. D. Poynton, D. J. Driscoll, and D. C. Apperley. “An ETFE-derived radiation –grafted anion-exchange membrane with properties spefically tailored for application in metal-cation-free alkaline polymer electrolyte fuel cells” Chem. Mater., 19, 2686 2007.
 [31] M. Unlu, J. F. Zhou, and P. A. Kohl, “Study of alkaline electrodes for hybrid polymer electrolyte fuel cells” Electrochem. Solid-State Lett., 12, B27 (2009) [32] A.V. Saurabh, R. Datta, “The effect of hydrogen crossover on open-circuit voltage in polymer electrolyte membrane fuel cells,” journal of power sciences 195, 8, 2010 pg. 2241-224  34