Current State of Knowledge: Direct Carbon Fuel Cells (DCFCs) Technology With Alkaline (Hydroxide) Electrolyte

SUMMARY

Among numerous modern high‐performance energy technologies that allow for conversion of chemical energy into electricity and heat, the most interesting are direct carbon fuel cells (DCFCs). They are the only ones among all types of fuel cells that allow for a direct conversion of chemical energy stored in the solid fuel (carbon) into electricity. Furthermore, they are characterized by high performance, which is not limited due to the Carnot's rule, low emissions of such substances as SO2, NOx, fly ashes and others, and a relatively simple design since there are no moving components. Nowadays, DCFCs have been developed all over the world. These cells differ first and foremost in the electrolyte they use. The type of electrolyte determines both configuration of the device and operating temperature. The paper discusses current state of knowledge concerning DCFC technology with alkaline (hydroxide) electrolyte and presents the results of research and development studies concerning such cells all over the world. Furthermore, main factors and parameters that impact on the operation of individual cells and potential challenges that have to be overcome in order to develop these technologies were characterized and discussed.

1. INTRODUCTION

Noticeable tendencies in the development of energy technologies are being observed in the European Union and all over the world, with their major task being not only to increase generation capacity but also to improve the efficiency of conversion of chemical energy of coal combined with activities commonly termed “clean coal technologies.” Among numerous modern high‐performance and low‐emissions energy technologies that allow for transformation of coal into electricity and heat, the most interesting are direct carbon fuel cells (DCFCs). The technology is promising and has been extensively explored recently in research and development studies. The aim of this paper is to review the current state of knowledge concerning DCFC technology with alkaline (hydroxide) electrolyte and presents the results of research and development studies conducted by the key technology players related to those fuel cells all over the world. Furthermore, main factors and parameters that impact on the operation of individual cells, key areas for further improvements and potential challenges that have to be overcome in order to develop this technology have also been characterized and discussed.

1.1 Direct carbon fuel cells—general overview

Direct carbon fuel cells are the only ones among other types of the fuel cells that are capable of direct conversion of chemical energy of solid carbonaceous fuels into electricity. The substrates supplied to DCFCs include elemental carbon (contained in, eg, hard coal, brown coal, carbonized biomass, graphite, carbon black, coke etc.) and oxygen (pure or contained in the atmospheric air), whereas products include electricity, pure stream of carbon dioxide, and mineral residue. Carbon fuel is supplied to anode chamber of the cell and, with electrochemical reactions that occur in elevated temperature, is oxidized to CO2, generating electricity (see Figure 1).

Figure 1

General scheme and basic operating principle of a direct carbon fuel cell [Colour figure can be viewed at wileyonlinelibrary.com]

Carbon fuel cells offer a range of benefits, among which are the use of solid carbonaceous fuels, eg, fossil coals1 or carbonized biomass of various origins2, 3 (even organic wastes like coffee grounds4 or wastepaper5), very high theoretical efficiency (100%), and practical efficiency (50‐80%),6-8 which translates into minimum 50% reduction of CO2 emissions per unit of generated electricity (compared with the conventional heat power plant) and low emissions of pollutants (eg, SO2, NOx, fly ashes). Furthermore, DCFCs can represent the element of distributed energy systems that generate electricity and heat from biomass with power from several to several hundred kWe. With the above benefits, these fuel cells can become a key technology for clean electricity and heat production in the future, especially in the small and medium size power systems.

First information about DCFC was published in the mid twentieth century. In 1855, Becquerel,9 and then, in 1877, Jablochkoff,10 construed electrochemical devices that used carbon rod as an anode, platinum, or iron melting pot (used simultaneously as an electrolyte container and cells' cathode) and electrolyte in the form of molten KNO3 or NaNO3. These devices generated electricity but were unstable as a result of electrolyte degradation.

1.2 Types of direct carbon fuel cells

Nowadays, DCFCs have been developed all over the world. These cells differ first and foremost in the electrolyte they use. The type of electrolyte determines both configuration of the device and operating temperature. There are four basic types of electrolytes: molten carbonates,11-14 solid oxide ceramics (mostly Yttria‐Stabilized Zirconia),15-17 water solutions of hydroxides,18, 19 and molten hydroxides.2, 20-22 Since recently, cells which combine two electrolytes (molten carbonate and solid oxide), so‐called hybrid DCFCs, have also been developed.23-26 The overview diagrams of functioning of DCFCs that used individual electrolytes are presented in Figure 2.

Figure 2

A general diagram of functioning of direct carbon fuel cells with electrolyte: A, molten carbonates; B, solid oxide ceramic; C, hydroxides; D, hybrid [Colour figure can be viewed at wileyonlinelibrary.com]

With different types of electrolytes, electrochemical processes that occur in DCFC cells on electrodes differ from each other. The effect of these processes is difference of potentials between electrodes, while their product is carbon dioxide generated in general reaction (1).

      (1)

2. HYDROXIDE ELECTROLYTE DIRECT CARBON FUEL CELL

In DCFCs with alkaline electrolyte, water solutions of hydroxides or molten hydroxides are used as electrolyte. The following reactions occur at individual electrodes of such cells:

1. anode.

       (2)

2. cathode.

       (3)

       (4)

In 1891, Thomas A. Edison patented a carbon fuel cell design27 (Figure 3A), used in 1896 by an American electrical engineer William W. Jacques to develop, demonstrate, and patent28, 29 his own DCFC (Figure 3B), which he named “carbon electric generator”. The cell was made of an iron melting pot (being a cathode) and a carbon rod that represented both fuel and receiver of the generated electrical charge (anode current collector). Air, pumped by means of a pump, was distributed on the container's bottom by means of a special perforated aerator. Cell's operating temperature was 673 to 773 K, while electrolyte was replaced by molten sodium and potassium hydroxides. The presented fuel cell was characterized by voltage slightly higher than 1 V and current density of 116 mA cm−2. Jacques also demonstrated a stack of 100 individual cells of his own design with total power of 1440 W (16 A, 90 V).

Figure 3

Sketches of carbon fuel cells: A, by Edison; B, by Jacques. Figures reproduced from References 27, 29

Based on the results of his research work, Jacques found that in order to produce 1336 Wh of electrical power in a cell, one needs 1 lb (≈0.454 kg) of carbon, of which 0.4 lb (≈0.182 kg) is used in the cell and the remaining 0.6 lb (≈0.272 kg) is burnt on the furnace's grate in order to maintain the desired cell's operating temperature. Finally, Jacques evaluated efficiency of his cell at the level of 32% compared with the gross calorific value of carbon.28

In 1904, Haber and Bruner31 found that reaction of electrochemical oxidation of carbon in the cell designed by Jacques did not result from direct electrochemical carbon oxidation but occurred from indirect reactions. They also argued that caustic soda used as electrolyte is consumed through irreversible reactions which generate carbonates and hydrogen. The last statement, relating to generation of carbonates of alkaline metals, became the basic problem of DCFC with hydroxide electrolyte for many years.

In the early 1990s, the interest in the technology of DCFCs with hydroxide electrolyte was revived. At present, the major contemporary research centers that are developing the technology include:

• Hawaii University, Hawaii Natural Energy Institute, USA,
• Scientific Applications and Research Associates, USA,
• West Virginia University, USA,
• Brown University, USA,
• Czestochowa University of Technology, Energy Engineering Department, Poland.

3. AQUEOUS‐ALKALINE DIRECT CARBON FUEL CELL

Early research on DCFCs with electrolyte in the form of water solutions of hydroxides were performed in autoclaves at temperatures of 473 K and pressure of 3 MPa. The autoclave container represented the anode, whereas cathode was provided by an iron bar. The fuel was brown coal dispersed in the electrolyte. The experiments yielded the open‐circuit voltage of ca. 0.55 V, which, however, dropped to zero very quickly.32

Later, the examinations of such cells began in Hawaii Natural Energy Institute—HNEI (Hawaii University, USA). The design of the cell schematically shown in Figure 4 was composed of a porous ceramic pipe with fuel (charcoal) which was poured inside the pipe. At the bottom of the pipe, there was a nickel plug, while a nickel piston was used in the upper part to press the fuel in order to obtain the highest possible electrical conductivity. Electrical wires were connected to both the plug and the piston in order to transfer the electrons. Around the porous ceramic pipe there was a metallic foil with the wound silver screen, which, with the sparger that was made of nickel tubing that supplied the air, represented the cathode assembly. Both anode and cathode assemblies were placed in an alumina tube, a bottom part of which was closed by a nickel base. Two electrical band heaters were placed around the tube. The electrolyte was water solutions of potassium, sodium, lithium, cesium, and magnesium hydroxides.18, 19, 33

Figure 4

Schematic diagram of the DCFC with aqueous‐alkaline electrolyte examined in HNEI. Reprinted with permission from Reference 18. 2007 American Chemical Society

The cell operated at temperatures of 353 to 518 K using pressed charcoal obtained through carbonization of corncob, macadamia nutshell (macshell), and fructose at temperatures of 1173, 1223, and 1323 K. If the temperature of fuel cell operation exceeded 373 K, the entire fuel cell was closed in a pressure vessel that allowed for regulation of pressure over the range of 2.8 to 3.6 MPa, which prevented electrolyte from boiling and evaporation of electrolyte.

During the tests, the researchers examined, eg, the effect of the types of material and various designs of cathode (see Figure 5), chemical composition of electrolyte and its temperature, and the type of charcoal and its amount on the cell performance characteristics.18

Figure 5

Different cathode designs tested in a DCFC developed by HNEI: 1—perforated nickel pipe that supplies air, 2—Porous sparger made of high‐nickel alloy, 3—Silver foil, 4—Silver screen, 5—Teflon pipe. Adapted with permission from Reference 18. 2007 American Chemical Society

Among various solutions tested for the cathode assembly, the best results were obtained for the variants presented in Figure 5A,F. Furthermore, various catalytic materials were examined for cathode (platinum, silver, gold, and palladium), and finally, it was found that the best solution is the arrangement presented in Figure 5F, where, instead of silver, a platinum foil was used, for which stabilization of electrical potential occurred the fastest.18

The highest values of basic electrical parameters were achieved for the cell fuelled with carbonized corncobs (with the amount of 0.5 g), with electrolyte in the form of a mixture of water solution of potassium and lithium hydroxides (6 M KOH/1 M LiOH) at temperature of 518 K and pressure of 3.58 MPa. The open‐circuit voltage between the anode and the cathode was 0.574 V, and the measured values of maximum current and power densities were 43.6 mA cm−2 and 6.5 mW cm−2, respectively.18

According to the authors, cell's electrodes should operate in different temperatures: cathode below 500 K and anode above 510 K. Two‐temperature design of the cell is aimed to eliminate certain problems encountered during the research, eg, evaporation of water from the cathode chamber. Furthermore, it was found that carbonate ions (CO32−) are unstable in the electrolyte at temperature of 573 K and are decomposed with generation of CO2 and reconstruction of hydroxyl ions (OH−).18 Moreover, cell's voltage remained nearly the same regardless whether or not hydroxyl or carbonate ions were involved in reactions on the anode.19

Challenges that the HNEI team has to face include in particular the improved voltage and electrical parameters obtained from the cell, development of the method of continuous supply of fuel, and evaluation of opportunities for scaling‐up the prototype. An undeniable benefit of this type of DCFC is a low operating temperature, which translates into the possibility of using cheap structural materials and, therefore, insignificant total cost of the DCFC system.

4. MOLTEN HYDROXIDE DIRECT CARBON FUEL CELL (MH‐DCFC)

Molten hydroxide DCFCs have many well‐known advantages,34 eg, high ionic conductivity (conductivity of molten hydroxides at 723 K is about 50% higher than in the case of molten carbonates at 923 K), higher electrochemical activity of carbon (higher anodic oxidation rate and lower overpotential), and lower operating temperatures which allow for the usage of less expensive materials for cell fabrication and helps to avoid generation of CO from the undesired Boudouard reaction. Furthermore, the use of molten hydroxides allows for placing both anode and cathode in one chamber filled with electrolyte, which is impossible in the case of cells with molten carbonate electrolyte since kinetics of oxygen reduction reaction in molten carbonates is substantially lower than in electrochemically aggressive molten alkalines.35

Major drawback of this technology is the limited stability of the electrolyte. Unwanted carbonates that cause electrolyte degradation are formed during operation of a MH‐DCFC. The formation of carbonates in the reactions of carbonaceous fuel and molten hydroxides has already been well studied. In Jacques early experiments, the formation of carbonates has already been observed, but the important work was done later by Goret and Tremillo.36, 37 They led the way to understand the mechanism behind the formation of carbonates and pointed out that the carbonates were attributed to the reactions between generated CO2 and molten hydroxide either through a chemical (5) or an electrochemical (6) process occurring on the anode. Moreover, electrochemical reaction (6) is found to be composed of two stages: chemical (7), that occurs very fast, and electrochemical (8), with substantially lower kinetics.34, 36-38

       (5)

       (6)

       (7)

       (8)

As can be observed, the rate of formation of carbonates depends on concentration of O2− ions (8), and, therefore, water in the electrolyte (7). Through the increase in the amount of water in the hydroxide electrolyte (eg, through moistening air supplied to the cell and/or maintaining a high pressure of water vapor over the electrolyte surface), the direction of the reactions (5)–(7) is shifted to the left (Le Chatelier's principle), and, consequently, the formation of CO32− ions is reduced. Furthermore, ionic conductivity of electrolyte is increased while structural material corrosion is reduced, resulting from limiting the formation of oxide ions generated during reactions (9) and (10).20, 34

       (9)

       (10)

Formation of carbonates can be limited or effectively inhibited by addition of such oxides as MgO, As2O3, Sb2O3, SiO2, or diphosphates (pyrophosphates) and persulfates from the group of oxyanions to the electrolyte.34 A positive effect of MgO addition to the hydroxide electrolyte was already demonstrated by Jacques.29

Zecevic and co‐workers39 found that neither water content nor oxide additives exerted substantial reduction of carbonate formation at temperatures below 923 K, so it is necessary to ensure operation of the cell at temperatures over 923 K. Moreover, the examinations performed by Kling40 in the West Virginia University revealed a rate of sodium carbonate formation in the MH‐DCFC of 0.67 mole of Na2CO3 per 100 kJ of electricity generated in the cell that operates at temperature of 873 K.

5. SCIENTIFIC APPLICATIONS & RESEARCH ASSOCIATES (SARA)

Based on the design and the principle of operation of the Jacques DCFC,28, 29 an American company Scientific Applications & Research Associates (SARA), supported by the Electric Power Research Institute designed and patented their own MH‐DCFC. Over many years of research, the team developed two designs of the cell:

• With one electrolyte chamber,20, 34, 35, 38, 39, 41, 42
• With two electrolyte chambers.35, 43-45

5.1 Cell with one electrolyte chamber

In the first design, as displayed in Figure 6A, the cylindrical graphite rod (used as both fuel and anode current collector) submerged in molten sodium hydroxide contained in a cylindrical or perpendicular container which was simultaneously used as a cathode current collector. The moistened air was supplied to the bottom of the container with electrolyte and distributed on its walls by means of a special perforated aeration system made of low‐carbon steel. The desired cell operation temperature (in the range of 673‐923 K) was obtained by electrical heating belts.34

Figure 6

SARA's MH‐DCFC prototypes with one electrolyte chamber: A, schematic; B, photographs of individual models. The figure (A) was redrawn and photo (B) was reprinted from Reference 34. Copyright 2004, with permission from Elsevier [Colour figure can be viewed at wileyonlinelibrary.com]

The main disadvantage of this system was that the air supplied to the cell had a direct contact with the surface of both cathode and anode—direct contact between anode and oxygen led to anode degradation and low electrochemical activity of the fuel.

Several prototypes were obtained during the research, performed with the cell with one electrolyte chamber, characterized by a simple design and using cheap structural materials. Individual generations of the prototypes presented in Figure 6B differed mainly with the materials used for the construction of cathode. In the case of the 0th generation cell, the cathode was made of a standard steel (C‐1018). In the next prototypes, the cathode was made from nickel foam coated with silver deposited on the C‐1018 steel (first generation), C‐1018 steel coated with nickel foam (second generation), and mild steel with addition of 2 wt% titanium (Fe2Ti; 98% Fe‐2% Ti)—third and fourth generation.20, 34, 39

The preliminary study showed that:

• C‐1018 steel oxidizes very fast and corrodes in the presence of oxygen contained in the electrolyte. Cathode surface becomes nonconductive after oxidation due to the formation of Fe2O3,20
• Silver coating covering the nickel foam is unstable in the molten hydroxide and is dissolved in electrolyte, leading to uncovering the nickel layer, followed by oxidation to low‐conducting NiO,20
• The prototype denoted as Mark II‐D (second generation, see magnified picture in Figure 6B) had an active surface of anode (graphite rod with diameter of 2 cm) of 26 cm2 and cathode (electrolyte container) made of C‐1080 steel coated with nickel foam from the side of electrolyte. The distance between electrodes was 1.3 cm. The cell generated the voltage (under non‐current conditions) with value of 0.85 V and maximum current of 7 to 8 A (≈270 mA cm−2) and maximum power density of 57 mW cm−2. The cell model operated for ≈100 hours, and it was found that nickel used for the cathode oxidized to non‐conducting nickel oxide, which had a negative effect on the obtained values of generated current and power.20, 34, 38, 42
• The cathode, made of mild steel with addition of titanium, allowed for the cell operation for over 540 hours. With the use of mild steel doped with 2 wt% of titanium (Fe2Ti) causes formation of a layer of oxide on the material surface, which represents the degenerated semi‐conductor characterized by a stable electrical conductivity and good corrosion resistance. This material has also very good chemical stability and promising catalytic properties for the oxygen reduction reaction. Due to the positive results of examinations, Fe2Ti steel was continued to be used for construction of consecutive generations of MH‐DCFCs (third and fourth generations).20

A prototype denoted as Mark III‐A (third generation, see magnified picture in Figure 6B) had an anode with surface area of 300 cm2, whereas cathode (which also represented the electrolyte container) was made of Fe2Ti. The distance between electrodes in this prototype was 3 cm. The open circuit voltage was 0.75 V, with the current intensity exceeding 40 A (≈130‐150 mA cm−2), and the average value of generated power density was 40 mW cm−2 (the maximum peak power output was 180 mW cm−2 sustained for 5 to 10 seconds). The obtained electrical parameters were recorded during the test which took 540 hours. The MH‐DCFC operation was under a mixed activation‐ohmic‐mass transfer control. The activation polarization was mainly due to slow anodic oxidation of graphite, the ohmic polarization was probably due to a large electrode spacing whereas the mass transfer polarization was due to slow oxygen diffusion to the cathode surface. Efficiency of the non‐optimized Mark III‐A cell that operates at constant current density (50 mA cm−2) was evaluated at 60%. Furthermore, the estimations revealed that for the power plant with the examined cell, efficiency of 70% to 75% can be achieved. It was also found that the operation of the examined cell depends on the cathode material, intensity of aeration, operating temperature, and size of the device.20, 34, 38, 42

A prototype of the fourth generation (MARK IV‐A) was also construed, with a cathode that was not the container for the electrolyte, but it represented another component of the cell. It allowed for increasing the cathode active surface area and reduction of the distance between the electrodes. Furthermore, a gold wire immersed in the melt was used in this prototype as a reference electrode used for independent measurements of anode and cathode potentials. During the tests of the MARK IV‐A prototype, the anode was provided by a graphite rod with diameter of 7.62 cm and active surface of 450 cm2. The cathode was a perforated cylinder made of Fe2Ti steel surrounding graphite rod. The examinations were performed for various temperatures (788‐923 K), content of water in the electrolyte, and cathode air flow rate. The effect of carbonate content in the electrolyte on cell's efficiency was also examined.39, 42

The results of the examinations showed (similarly to MARK III‐A) that higher electrolyte temperature and higher air flow rate resulted in improved cell operation. Based on the performance characteristics, the highest open‐circuit voltage (≈0.7 V), current density (≈110 mA cm−2), and power density (≈30 mW cm−2) were determined for temperature of 929 K.39

It was also found that the electrolyte with carbonate (Na2CO3) with the amount of 12 mol% led to lower electric parameters obtained from the cell, mainly due to a lower cathode performance. The authors also found that the rate of degradation of cell performance was similar to that recorded for pure NaOH electrolyte. Considering that the evaluated content of carbonate at the end of the experiment should be ca. 35 mol%, it was concluded that carbonate content does not affect the cell degradation rate within this carbonate content. Other tests demonstrated that the decline of cell performance over time was mainly caused by the phenomena that occurred on the cathode side, most likely connected with generation of corrosion products on cathode surface. Therefore, it was concluded that the steel with addition of titanium used for building the cathode is not, as initially presumed, the best material resistant to the conditions in the cell (temperature and chemically aggressive electrolyte). Therefore, authors conducted the examinations for alternative materials which demonstrated that the highest corrosion resistance is observed for the nickel doped with 2 wt% titanium (Ni2Ti)—over five‐times higher compared with Fe2Ti steel. Furthermore, it was demonstrated that resistance of polarized materials (which participated in electrochemical reactions) is substantially higher than for non‐polarized materials (that did not participate in the reactions, such as electrolyte container).39

In order to obtain higher electrical parameters, the authors indicate the necessity of optimization of MH‐DCFC design, which should include determination of, eg, optimal value of air flow rate, dimension and amount of air bubbles (including installation of additional turbulence promoters), surface roughness of cathode current collector and its active surface area, temperature and composition of the electrolyte (including use of additions that improved oxygen reduction reaction on the cathode), and a use coal instead of graphite.20, 38

5.2 Cell with two electrolyte chambers

A schematic of the MH‐DCFC system with two electrolyte chambers is illustrated in Figure 7, which contains a special porous separator installed between the anode (graphite rod) and cathode (electrolyte container). This solution was aimed to eliminate the problem of degradation of electrolyte by the carbonates generated during cell operation. With this cell design, the composition of electrolyte near anode (anolyte) and cathode (catholyte) is different and these electrolytes should not be mixed together. Anolyte is a mixture of molten carbonates and hydroxides, whereas catholyte contains mainly molten hydroxides. The separator that separates electrode chambers can be made of porous ceramic materials or porous metals resistant to corrosion, while its thickness should be small enough to limit the resistance of the flow of ions and ensure a minimal mechanical strength.35, 43, 45

Figure 7

The schematic diagram of a SARA's MH‐DCFC prototype with two electrode chambers. The figure was redrawn based on Reference 35 [Colour figure can be viewed at wileyonlinelibrary.com]

Working principle of the MH‐DCFC prototype consists in that hydroxyl ions restored on the cathode (electrolyte container inner surface) are transported by porous separator to the anode, where proper electrochemical reaction of carbon oxidation occurs. Transport of ions through separator occurs mainly in the electrical field generated between the anode and cathode (due to negligible dimensions of pores, the effect of diffusion and convection on ion movement was assumed to be insignificant). The negatively charged ions (OH− and CO32−) are moved to the anode and form anolyte, while additional ions (Na+, Li+, K+) are transported towards the cathode side, forming catholyte. Carbonate ions are collected near the anode surface, leading to the oxygen reduction on the cathode occurring exclusively in the hydroxide electrolyte. Carbonate ion concentration in the anode chamber increases with time, and, when it reaches a certain level, these ions start taking part in the anodic oxidation of carbon producing electrons and CO2 through reaction (11). 43, 45

       (11)

The increase in concentration of CO32− ions is stopped by the reaction according to Equation 11, and subsequently, gaseous carbon dioxide is released. Thus, anodic carbon electrochemical oxidation takes place in a mixed hydroxide/carbonate electrolyte and therefore the performance of the anode is better than in pure carbonate melts.

Apart from fuel cell configuration presented in Figure 7, SARA also proposed other design variants, as shown in Figure 8A,B. They differ in potential applications of particulate carbonaceous fuels through placing them in a porous perforated metal basket used as a anode current collector. Furthermore, in the design presented in Figure 8B, cell cathode does not represent the electrolyte container, but it is a separate component.43

Figure 8

Various design variants for SARA's MH‐DCFC with two electrode chambers. The figures were redrawn based on Reference 43 [Colour figure can be viewed at wileyonlinelibrary.com]

Therefore, the MH‐DCFC concept allowed for the use of particulate carbon fuel and stabilization of the electrolyte composition, which allowed for longer cell operation without degradation of electric parameters.

During the tests of the new design showed in Figure 7, the cell operated at temperature of 843 to 923 K with a separator made of porous sintered nickel and porous zirconium oxide. The test made in the first of the tested materials (sintered nickel) demonstrated that after 160 hours, cell operation stopped (causes were not determined).45 The cell with ceramic separator operated for 120 hours without degradation of electric parameters (it was initially assumed that 30 hours is needed in order for all the anolyte to react with CO2 and turn into carbonate). Composition analysis of the electrolyte in the cathode chamber after the test revealed the lack of carbonates and suggested the correctness of the adopted concept.35

No effect of carbonate ions on cathode was found after 500 hours of cell's operation (as presumed, these ions did not get through a porous separator to the cathode side). Furthermore, the time needed for conversion of hydroxide electrolyte near anode (anolyte) to the mixture of hydroxides and carbonates depends on the value of currents generated by the cell.44

At the present stage of the development of the DCFC technology, SARA finds it necessary to conduct a greater number of experiments which would allow to primarily determine the requirements for carbon fuel cleanliness (mainly hard coal) for the purposes of fuelling the cell and, consequently, the improvement in parameters of its operation. The current system has neither a capture nor management system for volatile matter contained in the coal, released during cell's operation. Coal devolatilization would be one of the stages of preparation of coal necessary for the improvement in the effectiveness of the entire system with MH‐DCFC. Moreover, SARA started cooperation with the West Virginia University in order to determine opportunities for manufacturing adequate coal rods used as anodes in a new cell's configuration.45

SARA expected application of its cell in stationary energy systems. The Joint Industry Program was developed for this purpose with the company's industrial partner, American Electric Power. Further development of the cell within Joint Industry Program was aimed to build a power plant with capacity of 100 MW. 46

6. WEST VIRGINIA UNIVERSITY (WVU)

After 2004, the SARA's research team started collaboration with the West Virginia University—(USA) to develop the methods to manufacture solid cylindrical coal rods that represent the replacement fuel for previously used graphite electrodes.35

Solid coal‐derived electrodes were made of various amounts of solvent extracted carbon ore (SECO) and petroleum coke (PetCoke). A standard coal tar binder pitch was used for binding the SECO and PetCoke together. SECO is a low‐ash extract material that is produced from bituminous coal at WVU by solvent extraction. Furthermore, Hackett and his colleagues from the WVU conducted numerous tests to determine cell performance characteristics (the design and principle of operation of the WVU's prototype presented in Figure 9 was similar to the cell developed by SARA illustrated in Figure 6A) depending on the properties of the fuel. The cell operated at temperatures of 873 to 973 K, with electrolyte in the form of molten NaOH. Graphite rods were also used during the test as a reference fuel.47

Figure 9

Schematic diagram of MH‐DCFC developed in the West Virginia University.Figure redrawn based on Reference 47 [Colour figure can be viewed at wileyonlinelibrary.com]

Using graphite, maximum current density was 230 mA cm−2, whereas open circuit voltage was 0.788 V. Coal rods yielded higher voltage (1.044 V), but the obtained maximum current density was only 35 mA cm−2. Maximum power density generated in the cell fuelled with graphite did not exceed 84 mW cm−2, whereas in the case of coal, the values were not higher than 33 mW cm−2. According to authors, differences in the obtained electrical parameters were connected with the larger resistance of coal than graphite.47

Contrary to the cell fuelled with graphite rods, the cell that used coal‐derived rods was characterized by unsteady operation. During some of the tests, coal rods started to crack and break. According to the researchers, it was due to the use of binder pitch as a bond, which was preferentially attacked in the electrochemical oxidation reaction. Therefore, the coal‐derived rods disintegrated over time in the electrolyte.47

7. BROWN UNIVERSITY

Recently, the research on MH‐DCFC has also begun in the Brown University (USA).21, 48 The experiments were performed for several design variants of anode and cathode.

The anode was made in two configurations:

A1) Basket made of nickel mesh mounted on the frame made of chromium‐nickel wire (also named “tea‐bag” anode, TBA),

A2) Nickel container (Ni alloy 200), perpendicularly shaped (25 mm × 25 mm × 5 mm), with holes bored in one of the sides, covered with the nickel mesh (also named rectangular box anode, RBA).

The cathode was also made in two various design variants:

C1) Nickel pipe with diameter of 6.35 mm placed inside the bigger pipe with diameter of 12.7 mm. A nickel mesh was wound outside the smaller pipe, representing the surface on which electrochemical oxygen reduction occurred (also named annular tube cathode, ATC),

C2) In this solution, the cathode was made analogously as anode A2, but in this case an air sparger that produces smaller oxygen bubbles was incorporated inside the nickel container (also named rectangular box cathode, RBC).

The scheme of the cell's design with A1 anode and C1 cathode is presented in Figure 10. The electrolyte used during the examinations was molten NaOH or eutectic mixture NaOH‐KOH (54‐46 mol%) placed in the melting pot made of ceramic heated by an electric heater. Activated carbon C‐3014 with particle size of 0.355 to 0.5 mm was used as a fuel, whereas air supplied to cathode was initially moistened. A silver wire with 1‐mm diameter was used as a reference electrode.21

Figure 10

Scheme of MH‐DCFC developed in the Brown University.Figure redrawn based on Reference 21 [Colour figure can be viewed at wileyonlinelibrary.com]

The open‐circuit voltage obtained between electrodes of the cell with the setup as shown in Figure 10 operating at temperature of 823 K was 1.055 V. Maximum current density was 110 mA cm−2, while maximum power density was 34 mW cm−2. No significant changes in the value of recorded electrical parameters generated from the cell and no significant electrolyte degradation was found during 100‐hour test (most likely due to supplying of moistened air to the cell; higher water content in electrolyte limits reactions of carbonate formation (see Subchapter 2.2). The test performed for various cell's operating temperatures in the range of 643 to 823 K showed that the increase in temperature leads to the increased conductivity of both electrolyte and oxide layers formed on the surfaces of electrodes while cell's power also increases. Furthermore, the temperature increase leads to limitation of overpotential losses.21

The obtained cell's performance characteristics with two different electrolytes at temperature of 773 K revealed that the open‐circuit voltage of tested MH‐DCFC operating with pure NaOH electrolyte was higher than in the case of NaOH‐KOH mixture. On the other hand, maximum current density values were higher for the NaOH‐KOH electrolyte. Furthermore, the use of the binary mixture of NaOH and KOH hydroxides allowed for the operation of the cell at temperatures lower than in the case of NaOH only, which allows for limitation of the rate of corrosion of materials used for construction of the cell.21

Changes in the anode design did not lead to significant changes in cell performance, whereas the use of C2 cathode led to the increase in power density by ca. 50%.21

The use of particulate hard coal (as‐received Pittsburgh #8 bituminous coal) as a fuel yielded open circuit voltage of 1.26 V. Maximum current and power densities were 122.2 mA cm−2 and 50.6 mW cm−2, respectively. The obtained values were higher for the raw coal compared with pyrolyzed coal. This was probably attributed to relatively high hydrocarbon content and more disordered structure of raw coal. On the other hand, pyrolysis process improved fuel stability in the contact with the molten electrolyte at 723 K. Moreover, it was found that the susceptibility of coals to electrochemical oxidation was correlated with their thermal heating values.48

Obtained results will be incorporated into a hydrodynamic fuel cell design that allows for continuous fueling and ash removal. This is the next step in the development of a practical operating MH‐DCFC capable of utilizing various types of carbonaceous fuels.21

8. CZESTOCHOWA UNIVERSITY OF TECHNOLOGY

The only research centre in Europe that examines carbon fuel cells with hydroxide electrolyte is the Energy Engineering Department (EED) in the Czestochowa University of Technology (Poland).

During the preliminary experiments, the research team in EED designed and made three prototypes of MH‐DCFCs shown in Figure 11, which allowed to select and choose proper materials for individual cell's components.49

Figure 11

The view of MH‐DCFC prototypes developed in EED: A, prototype I, made of carbon steel; B, prototype II, made of stainless steels 300 series; and C, prototype III, made of nickel and high‐nickel alloys [Colour figure can be viewed at wileyonlinelibrary.com]

Eventually, the preliminary studies and modifications led to obtaining the model made of nickel (prototype III presented in Figure 11C), characterized by stable operation and replicable results of the measurements of electrical parameters (not burdened with the effect of corrosion processes) in the same conditions during individual time intervals.2, 49

Figure 12 presents a detailed design of prototype III. The electrolyte container was made of nickel (Ni alloy 201). Cathode was represented by a pipe made of Inconel® 600 alloy with the external diameter of 42 mm (3‐mm wall thickness) and length of 135 mm, welded to the cell's cover from the bottom side. The air was supplied to the bottom part of the cathode by means of the pipe with external diameter of 6 mm and wall thickness of 1 mm, installed centrally inside the main cathode pipe. The air was distributed on the internal part of the pipe with 1‐mm holes, bored on its perimeter with the distance of 10 mm from the pipe's base (see the enlarged view in bottom right of Figure 12B). Anode was made of a pipe (Nickel alloy 201) with external diameter of 19.1 mm and wall thickness of 1.65 mm. From the bottom, the pipe was plugged with a special plug. In the side of the anode, 74 holes with diameter of 6 mm were bored to allow electrolyte to flow into its interior. A rolled‐up nickel mesh was placed inside the pipe (see the enlarged view in upper right of Figure 12B). Its task was to prevent fuel from getting to the electrolyte and receive electrons generated from electrochemical carbon oxidation. After a portion of fuel was placed inside the anode chamber, a special perforated plug was introduced, with its task being to immobilize fuel particles and allow for discharge of the generated gaseous CO2.2, 49

Figure 12

The prototype III MH‐DCFC developed in EED: A, design; B, 3D rendering visualization [Colour figure can be viewed at wileyonlinelibrary.com]

Besides the stable and replicable location of electrodes with respect to each other (the electrodes were installed in the cell's cover), the prototype III had separate anode and cathode chambers, which enabled elimination of the likelihood of mixing air that left the cell on the cathode side with CO2 generated on the anode side.49

During previous studies, the research team focused on, eg, verification of application potential for various forms and shapes of the fuel supplying the cell. Firstly, the positively completed tests using solid graphite and coal electrodes made authors to attempt to use particulate hard coal and biochar being the product of carbonization of various types of biomass as a fuel.

The results of the study demonstrated that in the case of the solid graphite electrode which is nearly pure elemental carbon, the cell generated electricity with several times lower power density than for biochars and hard coals. The discrepancies obtained could be explained by differences in crystalline structure of fuels: disordered structure of hard coals and biochars resulted in higher reactivity and susceptibility to electrochemical oxidation compared with ordered graphite structure.2

Furthermore, the results of the examinations pointed to a correlation between oxygen content in individual fuels and maximum power densities obtained from the cell as shown in Figure 13. The higher oxygen content in individual fuels (which was correlated with relative content of oxygen‐containing functional groups on the fuel grains surface), the higher maximum power density. However, unequivocal confirmation of this observation requires further research, which the authors has already planned.2

Figure 13

The relationship between maximum power density and oxygen content in fuel [Colour figure can be viewed at wileyonlinelibrary.com]

Another stage of the study was to determine the effect of various operating conditions,50 chemical composition of electrolyte,22 and thermo‐chemical processing of the fuel51, 52 on MH‐DCFC performance. During the study, the researchers examined the effect of, eg, fuel particles size, amount of air supplied to the cathode, electrode surface area, chemical composition, and electrolyte temperature on such electrical parameters such as current and power densities, electromotive force, internal resistance, etc.

Temperature has a significant effect on cell's operation; it determines, eg, electrolyte conductivity and electrochemical carbon activity. The results of the study unequivocally showed that the elevated temperature caused the increase in electrical parameters of the cell (see Figure 14A). The power and current densities achieved at temperature of 773 K were almost twice as high as those obtained for 673 K. The increase in temperature had an effect not only on the rate of electrochemical reactions (limitation of activation polarization) but also on ion mobility (reduction of diffusion polarization) and electrolyte conductivity (reduced ohmic polarization).50

Figure 14

The influence of various operating conditions including electrolyte temperatures (A), cathode inlet air flow rates (B), electrolyte compositions (C), and fuel particles size (D) on the current and power densities of MH‐DCFC developed in EED

Amount of air supplied to the cell should be adjusted to stoichiometry of electrochemical reactions and the potential hydroxyl ions creation on the cathode surface. This was confirmed by the results of the measurements that indicated that both too small and too high air flow rate reduces the values of the power density. At low air flow rate (0.03 dmn3 min−1), high concentration polarization was observed, which could have been directly related to insufficient amount of substrate (O2) for reduction reactions on the cathode side. Furthermore, too high value of air flow rate (0.8 dmn3 min−1) led to formation of big gas bubbles directly near the cathode surface, limiting the zone of reaction and decline in the value of current and voltage. The results of the study showed that the cell operated the best for the air flow rate of 0.5 dmn3 min−1 (see Figure 14B).50

Biochar particle size has a direct effect on the magnitude of active surface of the anode, that is, the surface of the contact between particles and current collector, which determines the amount of carbon molecules that are able to release electrons per time unit. The change in particle size of biochar that fuelled the cell had a direct effect on the value of voltage and maximum current density: with the increase in grain size, the values of listed electrical parameters decreased significantly. The highest values of electrical parameters were obtained for the fuel grain size of 0.180 to 0.250 mm and 0.425 to 0.500 mm, whereas the lowest values were obtained for the particle size of 1.0 to 1.4 mm. Since the rate of electrochemical oxidation of carbon is related to the contact surface area between electrolyte, fuel particles, and current collector, the best results were obtained for the particle size of 0.425 to 0.500 mm (see Figure 14D), which also ensured adequate contact area between fuel and current collector, and allowed for a free flow of electrolyte between fuel particles and permeating into the pore interior.50

The examinations also evaluated the effect of electrolyte composition on MH‐DCFC operation characteristics. Three various eutectic mixtures of hydroxides were chosen for the tests: NaOH‐LiOH (90‐10 mol%), NaOH‐LiOH (70‐30 mol%), and NaOH‐KOH (50‐50 mol%). Two fuels were used to supply the cell: solid graphite electrode (reference fuel) and particulate biochar obtained from carbonization of apple tree chips. The results indicated that the best electrolytes due to the obtained electrical parameters achieved by the cell were mixtures of NaOH and KOH with mole ratio of 1:1 and NaOH and LiOH with mole ratio of 9:1 (see Figure 14C). Considering the lowest temperature of the fuel cell and the highest obtained power densities, the eutectic mixture of NaOH‐KOH turned out to be the best electrolyte.22

The influence of fuel thermo‐chemical pre‐treatment of carbon fuels on performance of MH‐DCFC was also examined.51 Two carbon‐rich substances (hard coal and biochar) pre‐treated with HNO3 (at two different temperatures: 294 and 353 K) were used as fuels. The results confirm that HNO3 treatment of fuels can increase current and power densities and decrease internal resistance of tested MH‐DCFC. FTIR spectrums showed significant differences in the surface functional groups, made during the acidic treatment. Diversity and content of the oxygen‐containing functional groups on the surface of fuel particles were increased after nitric acid treatment. Moreover, formation of this oxygen groups tightens the mezo‐ and macroopores; thus, it reduces, eg, pore volume. Besides, the ash content of tested fuels has been decreased to more than half after a chemical treatment. The experimental results demonstrated a correlation between oxygen content in fuel (that is associated with the presence of oxygen‐containing functional groups in the fuel matrix) and maximum measured power density. The amounts of surface oxygen functional groups directly affect the electrochemical discharge rate of fuels because of the effect of the occurrence of a large number of free reactive sites. It can therefore be concluded that the amount and type of surface oxygen groups have a crucial effect on the reaction rate in the tested MH‐DCFC and were more important than, eg, elemental carbon content.51

The effects of thermal treatment of three raw coal samples on the performance characteristics of the MH‐DCFC were also analyzed, and it was concluded that the electrochemical oxidation of coals at MH‐DCFC anode is strongly dependent on elemental carbon, ash, and oxygen content in the fuel. Obtained results indicated that the maximum power and current densities were higher for raw coals rather than the pyrolyzed ones. Considering the performance characteristics of examined fuel cell, it becomes clear that amount and diversity of oxygen‐containing groups on the surface of the raw coal particles are desirable and largely responsible for fuel electrochemical reactivity.52

Depending on the fuel type and process parameters, power density ranged from 18 to 42 mW cm−2 was obtained. The highest values were observed for the cell fuelled with commercial charcoal (with particles size of 0.18‐0.25 mm), electrolyte (NaOH‐KOH) temperature of 673 K, and air flow rate of 0.5 dmn3 min−1.

The energy efficiency evaluated using the method similar to other heat engines was 41% (relative to the biochar lower heating value), which means a very good and promising result compared with other technologies of biomass chemical energy conversion into electricity. Furthermore, the determined electrochemical efficiency was 59% (at the voltage of 0.65 V) and was barely by 5% lower than theoretically achievable in the conditions studied. Fuel utilization factor was 95%, which means that most of the fuel was converted into electricity during fuel cell operation.6

9. NUMERICAL MODELING AND SIMULATION OF MH‐DCFC

Mathematical models coupled with computational and numerical simulations are important tools for design and optimization of fuel cells, stacks, and fuel cell power systems. Nowadays, numerical modeling and computer simulation are used for improving and analyzing common fuel cells, such as proton exchange membrane fuel cell, molten carbonates fuel cell, and solid oxide fuel cell. Recently, Xing et al53 attempted to model the effects of operational parameters (eg, temperature, pressures in the cathodic and anodic compartments, oxygen flow rate or amount and type of fuel), and construction parameters (eg, height of the anodic compartment) on the performance characteristics of MH‐DCFC, which has not been studied in literature so far. The plotted cell polarization curves are obtained for different values of the input parameters mentioned above, and it is found that the various polarizations affecting the cell performance are in the following order: ohmic > anode activation > cathode concentration > cathode activation. Furthermore, obtained results from computational simulations were compared with the available experimental data taken from the literature, although a direct comparison seems difficult because of the differences in the modeling approaches and assumptions, as well as cell geometric dimensions, operating parameters, and other properties of tested MH‐DCFC prototypes. Regardless of the difficulties resulting from such a comparison, it was concluded that the developed model is correct and allows for a simulation of the cell's operation. Nonetheless, carbonization phenomena inside the electrolyte have been simplified or neglected, and so far, the influences of, eg, molten sodium hydroxide conductivity (which is temperature dependent parameter), conductivity of the interconnection between the fuel and the electrolyte, and gas diffusion away from the electrochemical reaction sites on MH‐DCFC performance have been ignored.

10. CONCLUSIONS

Direct carbon fuel cells represent the technology that enables a direct conversion of chemical energy through electrochemical reactions into electricity at high efficiency. This technology is relatively simple compared with other fuel cell technologies and requires no expensive preparation of any gaseous fuel, as well as accepts all carbonaceous substances, as potential fuels. Previous results of the research studies concerning DCFCs with hydroxide electrolyte confirmed the opportunities for using many types of fuels, from solid graphite and carbon electrodes, particulate raw and modified hard coals through to carbonized biomass (charcoal, biochar) of various origins. The experiments performed in laboratories for various carbon fuels allow for recognition and assessment of the phenomena occurring on the electrodes during cell's operation, including the determination of the effect of individual physical and chemical properties of fuels, such as specific area, particle size, content of surface oxygen functional groups, crystalline structure, thermo‐chemical pre‐treatment, and content of contaminants on cell performance. To complement this overview, a summary of the main findings, achievements, and test results obtained in the individual research centers dealing with hydroxide electrolyte DCFC technology is presented in Table 1. Moreover, in Figure 15, the performance comparison of all studied DCFCs is presented.

Table 1. Comparison of the main finding and test results obtained in the individual R&D centers dealing with hydroxide electrolyte DCFCs

Figure 15

Comparison of maximum power densities in individual DCFCs with hydroxide electrolyte [Colour figure can be viewed at wileyonlinelibrary.com]

The data presented in Table 1 and Figure 15 indicated that the best electrochemical activity (highest power and current densities) was achieved for graphite rod immersed in the molten NaOH electrolyte in the MH‐DCFC developed in the West Virginia University while the worst results were determined for DCFC with aqueous‐alkaline electrolyte examined in Hawaii University where the fuel was in the form of granulated charcoal. Graphite is very conductive and performs well, probably due to the enhanced charge transport, even though it is relatively less reactive chemically compared with other fuels such as biochar, charcoal, and coal. However, particular attention should be put on investigating the possibility to supply the fuel cell with an “unpressed” and “unprepared” loose granular carbon fuel, ie, the type of fuel that is more easily achievable in a bulk and may be directly used for electricity generation without any preprocessing.

Although, the formation of carbonates in the reaction of carbonaceous fuel and molten hydroxides has been well studied, it is still necessary to improve the stability of the hydroxide electrolyte (or find a way for its effective regeneration with the recovery of hydroxides) for the practical application. It seems that this is a main challenge that technology is currently facing.

Development of mathematical models and conducting numerical simulations are useful for evaluation, optimization, and prediction of DCFCs performance. Until now, only one numerical model of MH‐DCFC has been developed, but soon more scientists and engineers are expected to be involved in this field. Therefore, it can be expected that the developed models will be more accurate and will enable a faster progress of this technology.

The international effort towards developing hydroxide electrolyte DCFC technology is relatively small in comparison to molten carbonates fuel cell, solid oxide fuel cell, and proton exchange membrane fuel cell technologies with only a few major key players. Most of the research and development work is performed at universities and research organizations. Apart from SARA, there is no other major well‐established company working in this area. All groups are testing small cells over short time periods (few hundred hours at the most). Moreover, typical power densities achieved from those DCFCs do not exceed 100 mW cm−2 (see Figure 15).

Despite the several advantages, this technology is still at an early stage of development and requires further research focused on investigation of the reaction kinetics, fuel delivery, materials degradation, and optimal operation parameters, before it will come to the phase of commercialization. Long‐term research is needed to assess the opportunities for practical implications and lifetime of the DCFCs discussed in this paper. They should provide a number of useful data which will allow for optimization of the cell's design, determination of the degree of degradation and corrosion of structural materials, and assessment of the possibilities of constant feeding with carbonaceous fuels and, in many situations, moving to the stage of prototyping of semi‐industrial models.

Source: Andrzej Kacprzak - Wiley

The 10 largest coal producers and exporters in Indonesia:

1. Indo Tambangraya Megah (ITMG)
2. Bukit Asam (PTBA)
3. Baramulti Sukses Sarana (BSSR)
4. Harum Energy (HRUM)
5. Mitrabara Adiperdana (MBAP)
6. Adaro Energy (ADRO)
7. Bumi Resources (BUMI)
8. Samindo Resources (MYOH)
9. United Tractors (UNTR)
10. Berau Coal

Click Here for Top World Clean Coal Technologies Contractors