Establishment of Operating Current Density

Understanding the relationship between current density and voltage is essential to setting the design current density. There are several factors that influence this choice. One strategy would be to operate at the highest current density possible, which would be at the limiting-current density. As seen from Equation 14.2 for the space–time yield, higher current densities increase Y and result in a smaller reactor volume for a fixed rate of production. A smaller reactor (smaller electrodes and less separator area) corresponds to lower initial costs. However, operation near the limiting current has its drawbacks.

1. As the limiting current is approached, V_cell increases rapidly. Both the voltage efficiency and the energy efficiency decrease with increasing current density, see Equation 14.5.
2. Operation at the limiting current may lower the current efficiency, especially for multistep reactions. Thus, the faradaic efficiency, η_f, is reduced. A lower faradaic efficiency reduces both the energy efficiency and the space–time yield. This situation is therefore particularly bad—a larger reactor that is less efficient.
3. The limiting-current density may exceed the capability of the available electrode materials or separator membranes and may therefore unacceptably shorten the lifetime of these materials. For example, high current operation may lead to excessive cell temperatures that damage the physical components of the cell.
4. Constant current operation at the limiting current may not be robust from an operational standpoint, since a change in the inlet conditions or in the operating conditions may reduce the limiting current and lead to undesirable side reactions if operation at the same value of the current is continued.

In the end, the current density will be selected so that the profitability of the electrolytic process is maximized. The choice of current density is both important and complex. Data on reaction rates, mass-transport rates, ohmic losses, current efficiencies, and heat removal are needed for design purposes. For the problems that we will consider in this chapter, you will either be given the current density, instructed to operate at the mass-transfer limit, or be given sufficient information to establish the relationship between the cell voltage and current in order to determine operation below the limiting current. In all cases, we will assume a uniform current density for simplification purposes.

While simplification is necessary for our initial treatment of the topic, let's not forget that we, the electrochemical engineers, have many tools at our disposal to alter the polarization curve. The flow rate of reactants can be increased. Higher flows lead to increased rates of mass transfer and greater limiting-current densities. The gap between electrodes can be reduced, resulting in a lower ohmic drop and better energy efficiency. We may be able to change the concentration of reactants, which directly affects the limiting-current density. Catalysts can be added to reduce kinetic polarizations and improve faradaic efficiency. Porous electrodes may be used to increase the specific surface area of the electrodes, and the temperature of the process may be changed.

Electrical Configurations

Once the current density is fixed, the electrode area follows directly from Faraday's law and the production rate. From our experience with electrochemical systems, we know that the potential of individual cells will be on the order of a few volts, and certainly less than 10 volts. The scale of industrial processes requires enormous amounts of electrical power. Most often the electrical power to drive the electrolytic process comes from high voltage alternating current (AC) that is then rectified. It's not practical to supply vast quantities of direct current (DC) at a potential of just a few volts. The solution is to place cells electrically in series to build voltage. The same approach is used for batteries, double layer capacitors, and fuel cells for high-power applications. The number of series connections is established from the system voltage and potential of an individual cell.

(14.17)

Determining the electrode area and the number of cells connected in series is not the end of the story. When a system contains more than a single anode and cathode pair in series, there are two general methods for making electrical connections: monopolar and bipolar. These terms have the same meanings as they did for batteries and fuel cells. In the monopolar configuration (Figure 14.7a), a separate electrical connection is made to each electrode. The current through the cell is divided among the electrodes electrically connected in parallel, . Many electrode pairs can be combined in a single cell. In the individual cell, all of the anodes in a cell are at the same potential, as are all of the cathodes. The voltage between each anode and cathode pair is the same and equal to the cell voltage. Both surfaces of each electrode are active. These cells can then be connected in series as needed to add voltage.

The second means of connecting multiple electrode pairs is the bipolar stack. Here, assemblies of electrode pairs separated with a solid conductive plate are stacked like a deck of cards. Current in a bipolar stack flows straight through the stack and eliminates the need for connections to each internal electrode; the current distribution tends to be more uniform in the bipolar arrangement. The use of narrow-gap cells is considerably easier in bipolar stacks. There are, however, a couple of important disadvantages. Because the current flows from cell to cell through a bipolar stack, failure of one cell results in failure of the entire stack. In contrast, electrodes in a monopolar arrangement function independently. Also, because the difference in potential from one end of the bipolar stack to the other is large, it is possible for a portion of the current to skip one or more cells and flow directly to another cell downstream (see Figure 14.7b). This phenomenon is referred to as a bypass or shunt current. Bypass currents reduce the faradaic efficiency. Electron-transfer reactions are still needed for the bypass currents, and these reactions may be undesired or destructive to the cell, as is the case for corrosion reactions. Often this damage is more important than the small loss of efficiency. Bypass currents can be reduced by eliminating bypass pathways, but this can be difficult to do in an industrial cell where, for example, electrolyte from different cells flows into a common manifold. Because of these characteristics, bipolar stacks are standard in fuel cells and redox-flow batteries, but used less frequently in industrial electrolysis.

Another key reactor characteristic that needs to be determined is whether or not a divided cell should be used. A divided cell uses a separator to create distinct anolyte and catholyte solutions. If possible, we prefer not to have a separator since it represents an extra resistance to current flow between the electrodes. However, as we have noted previously, the anode is at a higher potential than the cathode in an electrolytic cell. Therefore, a product or by-product produced at the anode can be reduced spontaneously at the cathode. Similarly, a product or by-product produced through reduction at the cathode can be oxidized at the anode. In addition, soluble products may react with each other in solution. Thus, a principal purpose of the separator is to prevent loss of faradaic efficiency by minimizing or eliminating transport of reaction products in order to prevent undesirable reactions. For example, the diaphragm in a chlor-alkali cell helps to keep Cl₂ that is dissolved in the electrolyte from reaching the cathode where it would react. Consequently, the faradaic efficiency is increased by preventing Cl₂ reduction to Cl⁻ at the cathode. A separator can also maintain purity of the anolyte and catholyte solutions. For example, the diaphragm in a chlor-alkali cell helps to reduce the amount of Cl⁻ in the catholyte, which increases the value of the liquid NaOH product. Finally, separators can prevent the formation of explosive mixtures such as H₂/Cl₂. The following questions may be useful in considering whether or not to use a separator:

1. To what extent is the desired product likely to react at the opposite electrode?
2. Are there undesirable solution phase reactions that may be prevented through the use of a separator?
3. Are there safety issues that can be addressed through the use of a separator?
4. Will use of a separator to create distinct anolyte and catholyte solutions enhance the value of product streams or avoid an expensive downstream separation process?

Flow Configurations

Industrial electrochemical reactors are usually flow reactors. Streams of reactants into and out of the reactor are an essential aspect of both continuous and semicontinuous operation. Reactors can also incorporate internal convection to improve rates of transport, as well as to improve the concentration and temperature distributions. Flow can be important for the removal of gases evolved in the reactor in order to minimize the resistance as discussed in Section 14.3. For assemblies of parallel plate electrodes, the flow is principally coplanar across the electrode surface; whereas with plug-flow reactors, the flow is in void spaces of the porous electrode.

Flow patterns are often unique to the application and too numerous to categorize succinctly. Nonetheless, we will identify the basic flow arrangements for an assembly of cells. For multiple electrode pairs that are housed together in one assembly (a cell for monopolar or a stack for bipolar), there are two principal flow arrangements: parallel flow and series flow. Figure 14.8 illustrates the difference between series and parallel flow for a monopolar design. It is clear that parallel flow will have a lower pressure drop. Series flow enables a greater fraction of the reactants to be converted in a single pass through the assembly, at the expense of a larger pressure drop. Problems associated with gas evolution can be exacerbated with series flow as bubbles accumulate along the flow path. Hybrids of parallel and series are also possible. Industrial practice favors parallel flow where conversion can be increased by placing cell stacks in series with respect to flow, or by separation and recycle of reactants.

Reactor Volume

The volume of the reactor can be estimated from a knowledge of the electrode area required to meet the desired production rate and the specific area of the reactor, . Referring back to Figure 14.7, we see that for the parallel plate construction, we can estimate the specific area from the cell dimensions, specifically the thicknesses of the electrodes and the width of the gap between electrodes. The calculation is straightforward, and the result can be summarized by a quantity called cell pitch (see Section 10.4). This parameter is simply the number of electrode pairs per unit length when the repeating units are stacked together.

(14.18a)

(14.18b)

where the factor of 2 in Equation 14.18b accounts for the two active faces of the electrode. With porous electrodes, there is a lot of internal surface area inside the electrode, represented by a, the specific interfacial area of the electrode. However, even for porous electrodes, we often speak of the superficial current density rather than the true current density based on the internal area. When using the superficial current density with a porous electrode, it is not necessary to include the internal surface area in our analysis.

Scale-Up

The process of designing a system for industrial electrolysis is sequential but with some iteration as noted previously. This characteristic is best illustrated through an envisioned process to scale-up a reactor. Referring to Figure 14.9, we might start with fundamental electrode studies as described in Chapter 6. Basic kinetic data are obtained and side reactions identified. The effects of temperature and reactant concentration are often examined at this stage. The second stage is a complete system of anode, cathode, and electrolyte, but at a subscale—a small single cell where the reactants are supplied in large excess to each electrode. The electrode area of this subscale cell might be a factor of 10 or more less than A_c. Of course, at this point, the area for an individual electrode and the total area are only estimates. These estimates will be refined at each stage. For this subscale cell, uniform current density is assumed and often there are no mass-transfer limitations.

The next step is to increase the area of the cell to its full size; that is, A_c. At this stage, the flow configuration is set and the design includes the effect of finite utilization (conversion) of reactants, u:

(14.19)

Utilization can be defined for an electrode, a cell, or a cell stack. Finally, these individual electrode pairs are connected together to form a system. As noted previously, electrodes are almost invariably connected together electrically (series–parallel combination) to build voltage. The cell may also be combined to form one or more mechanical assemblies. The flow rates of reactants and products are critical elements of the final cell design.

Finally, we note that there is off-the-shelf hardware available for initial evaluation of a process and prototype development. One example is the so-called plate-and-frame assembly shown in Figure 14.10. An off-the-shelf reactor such as this facilitates the development of new electrochemical processes.

Synthesis of Inorganic Chemicals

Electrolytic production of chlorine and sodium hydroxide, introduced earlier in this chapter, represents the largest electrolytic industry. The process produces chlorine, sodium hydroxide, and hydrogen from a salt solution. Production takes place at 60–95 °C and 0.1–1 MPa. Two types of cells dominated this industry for many years: the diaphragm cell described at the beginning of this chapter and a mercury cell. The mercury cell permitted operation at higher current densities and resulted in products of higher purity relative to the diaphragm cell, with similar energy requirements when the energy required to concentrate the dilute NaOH from the diaphragm cell is included. A combination of new technological developments and environmental concerns has led to the development of a third type of cell, a membrane cell, that takes advantage of a cation-exchange membrane, improved electrodes with reduced overpotentials, and corrosion-resistant polymers for cell construction to produce higher purity products than the diaphragm cell at lower energy consumption rates than the mercury cell. Therefore, most new chlor-alkali cells are of the membrane type.

Although at a much lower production scale, diaphragm and membrane cells are also used industrially to produce chlorine and hydrogen from hydrochloric acid. Hydrochloric acid is a by-product of several important nonelectrochemical industrial processes. Serious environmental concerns associated with handling and disposing of HCl can be avoided by converting it to useful products through electrolysis.

A number of other inorganic chemicals are produced by electrolysis at a smaller scale than chlorine production. For example, fluorine gas and other strong oxidizing agents such as KMnO₄, H₂O₂, and Na₂S₂O₈ (sodium persulfate) can be produced electrochemically. Recently, reagents such as hydrogen peroxide (H₂O₂) have been produced in situ by electrolysis at the quantity needed for optimal use. Another example of inorganic chemical production by electrolysis is high quality MnO₂ for battery applications. In addition to these and other contemporary products, there are many other products that can be made electrochemically, but are not currently manufactured that way because of cost. For example, water electrolysis can be used to produce hydrogen and oxygen gas at high purity; however, except for some specialized applications, other methods of producing hydrogen and oxygen are currently more economical.

Electrowinning of Metals

Electrowinning is the production of metals from ores by electrodeposition from a melt or solution. The most important industrial electrowinning process is the production of aluminum using the Hall–Héroult process. In fact, more electrical power is consumed in aluminum production than in any other electrolytic process. The key innovation, made simultaneously and independently in 1886 by Hall in the United States and Héroult in France, was the discovery that alumina (Al₂O₃) is soluble in cryolite (sodium hexafluoroaluminate) at about 1000 °C, resulting in a conductive solution. The overall reaction for the production of aluminum is

The precise details of the chemistry are not fully understood; consequently, it is difficult to write a complete set of electrochemical reactions. However, the cathodic reaction is clearly the reduction of aluminum. Molten aluminum is denser than the cryolite solution and falls to the bottom of the crucible where it forms the cell cathode; it is periodically siphoned off as the desired product as shown in Figure 14.11. The carbon anode is consumed in the reaction and is lowered gradually into the cell at a rate of about 2 cm per day to maintain the desired cell gap. The other principal reactant, alumina, is added periodically to the melt through a hopper. Typical faradaic efficiency is near 90%, but energy efficiency is low, on the order of 25%.

Other reactive metals produced by electrowinning from a molten salt include lithium, magnesium, and sodium, where chloride-based salts are typically used.

Copper and zinc are the principal metals recovered by electrowinning from aqueous solutions. The hydrometallurgical process used to do this includes acid leaching followed by extraction and then electrowinning. Historically, most copper has been made by smelting, a competing process. Electrowinning is performed in lined concrete tanks into which alternate rows of anodes and cathodes are placed. The spacing between electrodes is about 5 cm. The operating current density for copper ranges from 150 to 1500 A·m⁻², although maximum values of 350–400 A·m⁻² are more common. The cathodic reaction is the reduction of the metal, which is plated onto the cathode. The anodic reaction is oxygen evolution on, for example, Pb electrodes. Electrolyte temperatures range from 40 to 60 °C, cell voltages from 1.9 to 2.5 V, and current efficiencies from 80 to 95%. Air sparging, electrolyte circulation, or ultrasonic agitation can be used to increase mass transport and, consequently, the maximum current density. The purity of the copper produced by electrowinning can be quite high (99.999%) and is typically ready for market. In contrast to copper, most zinc is now produced by electrowinning, where the process used is similar to that used for copper. The cell voltage for zinc electrowinning, however, is somewhat higher at 3.3 V.

Electrorefining

In contrast to electrowinning, the purpose of electrorefining is to purify rather than to recover the metal. Aspects of copper electrorefining were used to illustrate several concepts in Chapter 4—you may want to review those parts. Metallic copper, often from a smelting process (approximately 99.5% Cu), is used as the anode. During the refining process, the copper anode is dissolved and copper is plated at the cathode. Any impurities that are more noble than copper stay with the anode and are not dissolved. Impurities that are more active than copper dissolve with the copper into the electrolyte. These active impurities, however, remain in the electrolyte and do not plate out with the copper at the cathode; they are later precipitated out or otherwise removed or recovered from the electrolyte. The net result is the electrodeposition of high-purity copper at the cathode (e.g., 99.999%).

A variety of metals can be purified by electrorefining. For example, nickel, cobalt, lead, and tin can all be refined electrochemically in aqueous solution. Active metals such as aluminum can also be purified in this manner with use of a molten salt electrolyte. The cell voltage for electrorefining tends to be lower than that used for electrowinning as the equilibrium potential is essentially zero for the electrodes of nearly the same composition. Current densities are also modest in order to maintain high purity product and avoid anode passivity (where applicable). As a result, the operating cell voltage for copper electrorefining is only about 0.25 V.

Electrosynthesis of Organic Compounds

A large number and variety of organic reactions can be carried out electrochemically. In fact, organic electrochemistry is considered to be a mature branch of organic synthesis, and most organic reactions that involve electron transfer can be performed by electrochemistry. Types of reactions include oxidation and reduction of functional groups, cleavage, substitutions (e.g., halogenation), additions (e.g., hydrogenation), coupling (e.g., dimerization), and rearrangement. See Further Reading at the end of this chapter for examples of specific reactions.

Organic electrosynthesis reactions may be performed directly or indirectly. Direct synthesis reactions are heterogeneous reactions that take place directly on the surface of the electrode. In most cases, the electrochemical reaction forms a reactive intermediate or radical that undergoes further reaction in close proximity to the electrode surface to produce the desired product. Indirect electrosynthesis reactions take place via a mediator, which in turn reacts homogeneously in solution with the organic reactant to produce the desired product. The mediator is regenerated electrochemically once it has reacted to affect the desired synthesis. Therefore, there are no waste or disposal concerns since the mediator is recycled and not consumed. Most mediators (catalysts) for indirect synthesis are inorganic redox couples such as the following:

1. Reductions: Sn^4+/Sn²⁺, Cr³⁺/Cr²⁺, Ti⁴⁺/Ti³⁺, Zn²⁺/Zn, Na⁺/NaHg
2. Oxidations: Ce³⁺/Ce⁴⁺, Cr³⁺/Cr⁶⁺, Mn²⁺/Mn³⁺, Mn²⁺/Mn⁴⁺, Ni (OH)₂/NiOOH, I⁻/I₂,Br⁻/Br₂,Cl⁻/ClO⁻

Indirect reactions may be advantageous when they can be used to replace organic reactions that have a high overpotential and sluggish kinetics or that tend to passivate the electrodes. Indirect reactions are also favored when the redox catalysts can provide enhanced selectivity. The catalyst regeneration and the chemical reaction steps can take place in the same reactor (in-cell) or in different reactors (ex-cell). An ex-cell strategy, made possible through the use of indirect reactions, permits separate optimization of the catalyst and organic reactions. Multiphase reactions are also possible with the catalyst regeneration in the aqueous phase and the organic reaction in a separate organic phase. Use of multiple phases can facilitate product separation and enhance the commercial viability of a process. As always, however, there are trade-offs between the simplicity of a direct process and the enhanced flexibility of an indirect process that must be considered carefully in the design process.

In spite of the many possibilities that exist, relatively few organic electrosynthesis reactions have been successful industrially. Even some of the early successes are no longer performed commercially. Steckhan (2012) estimated that approximately 200 reactions have been performed at the pilot scale with more than 100 commercially available. It is difficult to get a precise number because the details of many of industrial processes are often kept confidential.

The most significant industrial process is the production of adiponitrile, an intermediate in the production of Nylon^®. It is the only organic electrosynthesis process where the volume of production is consistent with that of a commodity chemical (300,000 metric tons·yr⁻¹). The reactions are shown below. The cathode reaction is the electro-hydro-dimerization of acrylonitrile, and oxygen is evolved at the anode.

(14.20)

The overall reaction is

(14.21)

These reactions occur in an undivided bipolar stack using aqueous sulfuric acid as the electrolyte. Another process, currently under development, is the electrochemical synthesis of ethylene glycol, which has the potential to become another high-volume process.

There are many potential advantages to the electrochemical synthesis of organic compounds. The inherent advantage is that electrons serve as the oxidizing and reducing agents. These electrons are, in general, inexpensive and clean relative to chemical agents. The rate of reaction is activated with potential rather than temperature. Thus, the mild conditions characteristic of electrochemical synthesis are well suited for chemicals that are heat sensitive. Also, since the current is directly proportional to the reaction rate, these reactions are inherently easier to control. Closely connected to the ability to control the process is the potential for high selectivity from electrochemical processes. Selectivity is particularly important for high-value specialty products. In spite of these advantages, the number of commercial processes is small, as are the volumes produced, as mentioned previously.

Given the advantages of organic electrosynthesis, why are there not more successful industrial processes? What are the key factors that contribute to a successful process? It turns out that energy costs and initial capital costs associated with the electrochemical cells are not typically the problem. At the risk of overgeneralizing, the factors that are often most important are the availability and cost of the reactants, the reaction yield that can be obtained, the ability to inexpensively separate the product(s) from reactant(s), the availability of a suitable, stable electrolyte, and the ability to achieve an acceptable production rate. Because of the low conductivity of organic solvents, it is necessary to add a supporting electrolyte, which must then be separated from the product downstream. Many commercially successful processes involve water soluble reactants and products and utilize sulfuric acid as the electrolyte. Alcohols and acetic acid are also used industrially with some frequency. Cosolvents can be used to enhance solubility if needed. Separation can be facilitated by phase separation where feasible.

Another important factor that may easily be overlooked is the need to consider organic electrosynthesis as a design alternative early in process development. It is difficult and expensive to consider such options at an advanced stage of design. Consequently, it is important for commercial success that a company has the expertise needed to consider electrochemical options as part of their normal design process. This factor is likely to become more important in the future as society shifts to solar power as the primary energy source from which electricity can be generated directly rather than from fuels as is currently the case.

Electrolytic Fuel Generation

As we look to the future, it seems clear that the sun will be our primary source of energy. In addition to solar thermal methods, solar energy can be captured in the form of energetic electrons and holes, inherently an electrochemical process. Also, since the availability of solar energy is cyclic, electrochemical processes can be used for energy storage in order to provide the energy needed in off cycles.

Perhaps more important, the transition to solar energy as the primary source of energy for society will likely be accompanied by a shift from fuel-based energy use to direct use of electricity. Electrochemical devices will undoubtedly play a critical role in this shift. Electricity has traditionally been the “high-end” form of energy since it has most frequently been generated using fuels. Consequently, it has historically been more efficient to use fuels directly, where possible, rather than to use fuels to generate electricity, which is subsequently used for the intended application. This will change as electricity is generated directly from renewable sources; hence, electrochemical processes that use electricity directly will have an added economic advantage. There will also likely be a shift from centralized generation of electricity to a distributed solar-based system, which will again impact the use and scale of electrochemical devices. The same type of shift is already occurring in the transportation industry, driven by environmental concerns, where portability will continue to depend on electrochemical devices.

What role, if any, will electrolytic processes play with respect to fuels in a solar-based system? In applications where direct use of electricity is not viable, electric power can be used to generate solar fuels such as hydrogen. This process is a reversal of the previous paradigm where fuels were used to generate electricity. Electrolysis of water to produce hydrogen and oxygen is perhaps the electrolytic process first considered for fuel generation.

Water Electrolysis

The electrolysis of water is essentially a fuel cell in reverse, where electricity is used to create hydrogen and oxygen from water. Therefore, water electrolyzers reflect the types of technologies that we considered for fuel cells in Chapter 9. The three principal types of electrolyzers are alkaline, PEM, and solid oxide.

Most commercial water electrolysis is performed with alkaline electrolyzers. The electrode reactions are as follows:

and

The net reaction is, of course, just the splitting of a water molecule to produce hydrogen and oxygen. Figure 14.13 shows the equilibrium potential and thermoneutral potential (−ΔH_Rx/nF) for water electrolysis as a function of temperature. These data assume the water is a vapor. Near room temperature, where water is a liquid, the two values are close to one another. At high temperatures, there is large difference—which is important for high-temperature electrolysis. The equilibrium potential, which is proportional to the free energy change for the reaction and, therefore, the minimum potential needed to split the water, decreases with increasing temperature. In contrast, the thermoneutral potential does not change significantly with temperature. Operation at a cell voltage greater than the equilibrium potential but less than the thermoneutral potential would lead to net cooling due to the reversible heat term (TΔS) for water electrolysis, and would require the addition of heat to the reactor. Above the thermoneutral potential, the irreversible losses in the cell are sufficient to generate the heat needed for the reaction. Note that, while convenient, analysis with the thermoneutral potential is approximate for an open system such as an electrolyzer since it does not account for the enthalpy of the inlet and outlet streams. Therefore, where possible, use of the full energy balance, Equation 14.22, is preferred.

Voltage losses as a function of current density in a typical alkaline electrolyzer are shown in Figure 14.14. The overpotentials for both the anodic and cathodic reactions are significant. As expected, ohmic losses become more important at the higher current densities. Since the energy efficiency is a direct function of the operating voltage, a trade-off exists between the absolute amount of hydrogen that can be produced in a given electrolyzer and the energy efficiency at which it can be produced. Commercial electrolyzers typically operate at cell potentials below 2 V and current densities between 1000 and 3000 A·m⁻². Note that at 25 °C, U^θ = 1.229 V and the thermoneutral potential is 1.481 V (assuming liquid water). Given the relatively sluggish kinetics at these temperatures, operation below the thermoneutral potential is not practical.

Both monopolar and bipolar alkaline electrolyzers are available, although the bipolar configuration is more common. The electrolyte consists of 25–30 wt% KOH, which has a relatively high electrical conductivity. The decision to operate at high rather than low pH is based on material's cost and stability. Specifically, corrosion problems are less severe at high pH than under acidic conditions. Cells traditionally operate at temperatures ranging from 65 to 100 °C. Input or makeup water must be relatively pure (κ < 500 μS·m⁻¹) in order to avoid the buildup of impurities in the cell. Alkaline electrolyzer technology is considered to be mature with a life expectancy of up to 15 years. High-temperature alkaline cells have also been developed; these cells operate at temperatures up to 150 °C in order to improve conductivity and reaction kinetics, although water management is an issue at the high temperatures.

Water electrolyzers based on proton exchange membranes (PEM) are also manufactured. These electrolyzers first appeared during the Space Race in the 1960s, and use technology similar to that of PEM fuel cells. The electrode reactions are as follows:

Hydrogen production rates in PEM electrolyzers are low relative to alkaline cells, in spite of the fact that the current densities are higher. The low production rate is due to the smaller electrode surface area (superficial) in the cell. Use of an ion-exchange membrane provides enhanced safety and increased purity due to the low gas permeability of the membrane. PEM electrolyzers operate well at partial load and have good dynamic performance. The major disadvantages of these electrolyzers are higher initial cost due to the membrane and catalysts, shorter lifetimes, and lower hydrogen production capacity.

A promising technology for water electrolysis that is still at the research stage is that of solid oxide electrolyzers (SOEs). Analogous to solid oxide fuel cells, these devices operate at high temperature and utilize a solid, ceramic O²⁻ conducting electrolyte. The reactions are as follows:

Operation at temperatures up to 1000 °C provides some important advantages. Reaction rates are much faster at high temperature, which circumvents the need for expensive catalysts. Also, the equilibrium voltage decreases with increasing temperature—at 1000 °C U = 0.922 V. That means that less electrical power is needed to drive the reaction in the desired direction to produce hydrogen and oxygen. As seen in Figure 14.13, at 1000 °C, the thermoneutral potential is 1.291 V. It is easy to conceive of a condition where the cell potential is below the thermoneutral potential. This means that, if available, high-temperature heat rather than electricity could be used to provide some of the energy needed to carry out the reaction. Consequently, these SOEs are viewed as a candidate for coupling with high-temperature gas-cooled nuclear reactors to provide both the electrical energy and high-temperature heat for optimal operation. SOEs also provide the advantage of fuel flexibility since, for example, they are capable of directly reducing CO₂ to CO or a combination of H₂O and CO₂ to syngas (H₂ and CO) if desired. The primary concern with SOEs is the lack of suitable materials to provide the lifetime needed for industrial applications.

In spite of recent and continuing developments, water hydrolysis is still more expensive, in general, than hydrogen generation from hydrocarbon sources. Consequently, only about 4% of hydrogen is currently produced from water. The largest electrolytic hydrogen plants worldwide are located in proximity to hydroelectric generation facilities in order to take advantage of inexpensive off-peak power.

Other Electrolysis Processes

There are several other electrolytic processes that may offer advantages over water electrolysis. Two key advantages include: (1) use of a waste stream to produce the desired fuel and perhaps other valuable products while simultaneously cleaning up the stream and (2) reduction of the voltage, and hence the energy, required for hydrogen production. It may also be possible to design or modify processes that have traditionally treated hydrogen as an undesirable by-product to produce hydrogen as one of the intended products. Two examples of alternative processes for hydrogen generation include the electrolysis of HCl and the electrolysis of NH₃.

Waste streams rich in ammonia can be treated by electrolysis to produce hydrogen gas while simultaneously cleaning up the stream. The equilibrium potential for the electrolysis cell is 0.06 V, much lower than that for water electrolysis (1.229 V). Thus, hydrogen production from ammonia waste streams would require much less energy than that required for water electrolysis. In a similar fashion, urea-contaminated wastewater can also be treated by electrolysis (U = 0.37 V). In addition, both of these processes contribute in a positive way to the reduction of the amount of fixed nitrogen, which has increased significantly as a result of human activity. Finally, another electrochemical process for hydrogen production involves the use of a photoelectrochemical cell. This process will be treated separately in Chapter 15.

Wastewater Treatment

Several of the processes just mentioned involve the use of electrochemical reactors in environment friendly ways to produce useful products while simultaneously cleaning up waste streams. Another use of electrochemical technologies is for the cleanup of industrial effluent streams containing dilute concentrations of toxic materials. To illustrate, electrochemical methods can be used to remove toxic metal ions and have been the subject of renewed interest as regulations have been tightened. For many metals, it is no longer feasible to meet the specified limits for effluent discharge with use of conventional hydroxide precipitation methods. Also, the cost of disposing of the precipitation sludge has increased dramatically as such disposal must ensure that ground contamination by leaching out of the metal ions does not occur. In addition, remediation of water pollution caused by low concentrations of pharmaceutical residues is of significant recent interest and can be done electrochemically. In the treatment below we assume the following:

• The concentration of the contaminant is dilute. Therefore, removal of the contaminant does not change the liquid flow rate to any appreciable extent.
• The removal or cleanup of the contaminant is mass-transfer limited.
• The inlet flow rate and concentration are known.
• The outlet target concentration is known (typically set by regulation).
• Operation is continuous and steady state.
• The reactor cross section is constant and variations only occur in one dimension along the length of the reactor.
• Axial diffusion is not significant.
• Excess supporting electrolyte is used to enhance the conductivity and reduce the potential drop in solution.

The situation that we consider here involves a three-dimensional porous electrode. Our objective is to estimate the size of reactor needed to clean up the stream to the desired level. The flow-through and flow-by configurations, as well as the required material and charge balances, were presented in Chapter 5. Key results are repeated here for convenience. The concentration distribution under mass-transfer control is a function of x only:

(5.57)

where v_x is the actual velocity of the fluid in the pores in the direction of interest (εv_x would be the superficial velocity). Even though the local rate of reaction is controlled by mass transfer, current must still flow in solution between the upstream counter electrode and the three-dimensional electrode of interest. Again, as we saw in Chapter 5, the current in solution is

(5.60)

Finally, there is a potential drop in solution associated with the current flow. We consider the most common situation where σ ≫ κ. Under such conditions, the change in overpotential is equal to the change in the potential in solution across the thickness of the electrode.

(5.62)

where

You may be wondering why we care about the potential drop and change in overpotential if the system is at limiting current and controlled by mass transfer. To illustrate, consider Figure 14.14, which shows the current density as a function of overpotential. At limiting current, the current does not change with changing overpotential. However, if the overpotential is increased too much, the current will again increase due to side reaction(s). Side reactions should be avoided since they consume power and may have significant additional negative impacts on the process. For the process under consideration, side reactions are avoided by limiting the potential drop in solution, which limits the range of overpotentials in the system. The maximum depends on the specific chemical system, but is usually between 100–300 mV. Equation 5.62 can be rewritten as

(14.23)

where the superficial velocity, v_s = /A_c. Equation 5.57 can be used to relate the concentration at the outlet to the thickness L of the electrode as follows:

(14.24)

In writing the equation in this form, we recognize that k_cm is a function of the velocity through the bed, v_x, which will vary with both and A_c. The following correlation from Wilson and Geankoplis for packed beds at low Re can be used:

(14.25)

This correlation applies to a packed bed of spherical particles: 0.0016 < Re < 55, 168 < Sc < 70,600, and 0.35 < ε < 0.75. Mass-transfer coefficients in electrochemical reactors typically vary between 10⁻⁶ and 10⁻⁴ m·s⁻¹.

Our goal is to perform a preliminary calculation of the size of the flow-through reactor. In doing so, we note that there are multiple combinations of cross-sectional area and length that will provide the desired outlet concentration. Also, the simple model used here seems to indicate that the length of the porous bed should be as short as possible since a short length would minimize the potential drop and the pressure drop through the bed (Figure 14.15). However, practical issues such as the nonuniformity of current across a large area and the difficulty of fabricating and obtaining uniform flow over a thin bed with a large cross section become important. Where possible, we recommend that a cross-sectional area consistent with a commercially available electrolyzer be chosen in order to avoid the design and fabrication of a custom system. In practice, the final design decisions will be based on an optimal return on investment.

The preliminary design calculation can be performed as follows:

1. Choose an initial cross-sectional area for the reactor (best if aligns with that of a commercial available reactor).
2. From the total flow rate, cross-sectional area, and void fraction, calculate v_s and v_x.
3. Use v_x to estimate k_c for the bed with use of a correlation or from experimental data.
4. Calculate the length L of the bed with use of Equation 14.24.
5. Repeat the above steps for several possible values of the cross-sectional area until a value for L that is comparable to that of the height and width of the unit is obtained.
6. Use Equation 14.23 to check if the maximum potential drop has been exceeded for the chosen value of L.
7. If the absolute value of the potential drop is lower than the specified maximum, the initial sizing of the electrolyzer is complete. If the potential drop is too high, increase the cross-sectional area until the potential drop no longer exceeds the specified maximum.

Illustration 14.14 demonstrates this general procedure.