Anode fuel utilisation

Figure 5.44 Fuel processor efficiency versus anode fuel utilisation for different fuel processor configurations steam reforming with 100, 90 and 80% efficiency (SR 1.0, 0.9, 0.8) and autothermal reforming with 80% efficiency the arrow compares steam reforming and autothermal reforming at 80% efficiency for a hydrogen utilisation of 73% [16].

The reaction product H20 is mixed with the anode gas and its concentration increases with higher fuel utilisations Uf as shown in Figure 2.2. The fuel utilisation Uf is defined as the ratio of the utilised fuel and the maximum available fuel,... 

Concerning practical applications, high fuel utilisations result in low of hydrogen to water pressure ratios at the outlet of the anode, whereas high hydrogen pressures occur at the entry of the anode. Thereby, the non-linear dependency of the Nemst voltage from the fuel utilisation has to be taken into account with respect to the integral solution of Equation (2.59). 

In Case B, the hydrogen flow rate is chosen with 2.72-10-6 mobs to approximate a linear change in the molar hydrogen fraction from 0.7 at the entry to 0.299 at the outlet of the anode along the cell area. In this range of the molar fraction, the Nernst voltage changes approximately inversely proportional with the fuel utilisation. 

The MEAs of the MCFC are sandwiched between stainless steel or nickel bipolar plates, which collect current from, and distribute fuel/steam and oxidant air plus recirculated anode gas via the porous electrode structures to, the chemical reactions at the electrolyte/gas interfaces. Figure 5.1. In the long-anode, narrow-cathode, rectangular MTU fuel cell, fuel utilisation is about 80%, oxygen utilisation about 50%. 

The loss of methanol is often transposed into a crossover current - the current equivalent to that which would be produced by the methanol, had it reacted properly on the fuel anode. This current, ic, can be used with the useful output current i to give an important figure of merit for a DMFC, which is the fuel utilisation coefficient r)f. This gives the ratio of the fuel that is usefully and properly reacted on the anode to the total fuel supplied, the difference being accounted for by some fuel crossing-over and being lost at the cathode. 

Table 10,1 Contributions to ASR for a Rise-type anode-supported cell (Nl-YSZ/YSZ/tSM-YSZ) at 8S0°C tested in a plug flow-type configuration at S and 8S% fuel utilisation (FU). Rehji is calculated using a specific conductivity of YSZ of 0.045 S/cm, Sconnect is an estimation, Rp.eichem is the sum of typical anode and cathode polarisation resistances measured in separate electrode experiments, Rp.aiff is calculated using a diffusion coefficient of 10 cm /s, 30% porosity, a tortuosity factor of 3 and a thickness of 0.1cm, and fip,conver is Calculated using Eq. (10) with i = 0,5 A/cm ...

The assumption of a position-independent local area specific resistance is an approximation, which is not always justifiable. Part of the anode polarisation resistance is dependent on fuel composition. However, often this part is small. If the cell is not isothermal, the local resistance will vary with position due to its temperature dependence. Also the actual flow pattern may be much more complex than just co-flow. Even so, if the fuel utilisation is large. ASRcor derived from Eqs. (4) and (5) will always be a better characteristic of a cell than a value derived neglecting the fuel utilisation (Eq. (1)). More precise evaluation of ASRcor requires a rigorous 3-D modelling of the cell test. 

After correcting for the effect of non-negligible fuel utilisation, the cell resistance is still significantly smaller when measured with 20% water in the feed than with 5%. This reflects a gas composition dependence of some of the loss terms in Eq. (2). In reference [45], it is argued that the observed composition dependence is primarily due to the composition dependence of the diffusive losses on the anode side (diffusion overvoltage), and it is shown how one may utilise characteristics obtained with different water vapour/hydrogen ratios to assess the magnitude of the diffusion loss [45]. 

Figure 10.12 shows a number of ASR values obtained at different temperatures for an anode-supported cell together with values modelled from the available knowledge of the cell components [39]. The measured ASRs have been corrected for fuel utilisation. The electrolyte resistance and the electrode polarisations only approximately follow Arrhenius expressions in reality. The assumed values of activation energies and pre-exponentials are given in the figure caption. For the diffusion resistance, which is the only non-temperature activated term of the considered losses, a conservative estimate is used. 

Figure 10.13 Area-specific cell resistances corrected for fuel utilisation (ASRcor) measured at various temperatures ofan anode-supported cell. Thefuel was hydrogen withca. 5% water vapour at aflowrate of 301/ h and the airflow was 140 l/h. Cell area 16 cnr. ...

Due to the cell voltage drop as a result of the decreasing fuel concentration towards the exhaust side of the stack, only a certain percentage of the available fuel can be electrochemically converted to electricity and heat. An overall utilisation of 85-90% is considered a practical maximum. At higher fuel utilisations, nickel may oxidise locally. A catalytic burner is used to burn the remaining fuel from the anode side with the surplus air from the cathode side. 

It is important to monitor the concentration of methane using a sensor as the fuel crossover may occur. The problem of fuel crossover is severe for the PEM direct methanol fuel cell. Methanol dissolves quickly in water and reaches the cathode. The presence of platinum catalyst on the cathode will oxidise the fuel, though not as effectively as the anode catalysts. Hence, fuel wastage will occur at the cathode reducing the cell voltage. This phenomenon yields mixed potential and the fuel utilisation coefficient //f is defined as... 

Hence, at the anode, the concentration of product H2O increases with the increasing fuel utilisation factor. As discussed in Chap. 3,... 

Fig. 14 Evolution of a the anode and b cathode probability of failure during IV characterisation, operation at the nominal point with creep of the GDL and MIC and thermal cycle to room temperature for the co- (CO) and counter-flow (COU) configuration [89]. Anode-supported cell and compressive gaskets and SRU geometry depicted in Fig. 5. PR methane conversion fraction in the reformer, j current density, FU fuel utilisation, Pf probability of failure. Reproduced here with kind permission from Elsevier 2012...

Fig. 18 Effect of fuel undersupply of a SRU in a stack due to creep in the anode GDL (lines no creep, dots with creep), during operation at a constant current density, corresponding to initial specific system power of 0.2 W cm (black) and 0.3 W cm (grey). Counter-flow eonfigu-ration, 25 % of pre-reformed methane and maximum allowable temperature of 1150 K. j eurrent density, FU fuel utilisation [90]. Reproduced here with kind permission from Elsevier 2012...

Another option is to use redox stable anodes to raise the actual fuel utilisation, without making the system more complex. This is perhaps one of the best reasons for the use of redox stable anodes as they can tolerate these high steam conditions. 

Anode gas recycle can also be used with these systems to further enhance the fuel utilisation or to supply the steam for any reforming processes. 

At the outlet of the anode the gas (RG) consists of the non-utilised fuel and the reaction products CO2 and H2O. This mass flow is equal to the mass flow of the utilised fuel and of the transferred oxygen by the ion conduction through the electrolyte. The stoichiometric demand of oxygen related to the inlet fuel mass flow is given by the figure //.o2o- Finally, we get for the enthalpy flow at the anode outlet... 

Most fuel cells involve electrochemical oxidation of hydrogen on the anode. There are only few examples of direct conversion of other fuels on the anode. In high-temperature fuel cells, it is possible to convert the fuel to hydrogen inside the cell utilising the heat from the electrochemical reaction [411], but otherwise it is necessary to convert the primary fuel outside the stack into a hydrogen-rich gas which is fed... 

An application of pre-reforming is the feed pre-conditioning for high temperature fuel cells, such as solid oxide fuel cells. When steam reforming is applied, the steam generation of the fuel cell anode may balance the steam consumption by steam reforming [38]. To utilise the anode off-gas, a certain amount needs to be split and refed to the pre-reformer [38]. 

---