Mench M.M. Fuel Cell Engines

Подождите немного. Документ загружается.

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

410 Other Fuel Cells

The major disadvantages of the PAFC are mainly in comparison to the PEFC, whose

development has been greatly accelerated at the expense of the PAFC:

1. The system power density is not appropriate for automotive applications. The

PAFC is a bulky, heavy system compared to the PEFC. Area speciﬁc power (0.2–

0.3 W/cm

) is less than half that of a PEFC [49].

2. The required use of platinum catalyst with nearly the same loading as PEFCs despite

the increasal operating temperature.

3. The relatively long warmup time until the electrolyte is conductive at ∼160

◦

(although this is much less than of a problem the MCFC or SOFC).

4. Although the PAFC has been successfully demonstrated in buses, much higher

power densities can be achieved with PEFC systems.

5. The ultimate cost of PAFC units is $3000–4000 per kWe, which is three to four

times higher than needed for ubiquitous stationary power market application [45].

Although the PAFC has many advantages compared to the PEFC or the SOFC and

a demonstrated history of reliable, high-efﬁciency performance, development of these

systems has slowed tremendously as a result of the failure to achieve a competitive market

price. Most of this research has been supplanted by SOFC and PEFC development. Some

attempts at achieving a high-temperature PEFC system might actually be considered a new

generation of PAFCs, however, combining phosphoric acid–based membranes with a PEFC

system for high-temperature operation. In these systems, beginning with Wainright et al.

[50], a polybenzimidazole (PBI) membrane structure is impregnated with phosphoric acid

and can be operated at 200

◦

C. There have been many such papers on this approach [e.g.,

51–55]. Performance is relatively high, but durability of these systems remains a major

question.

Ultimately, the main disadvantages of the PAFC are low system power density and

cost. The PAFC is too bulky to be used for mobile applications and too expensive to be used

in many commercial stationary power applications. Until these issues can be overcome, the

PAFC will remain primarily an option for premium stationary power applications only.

7.4 ALKALINE F UEL CELLS

The AFC has been developed for space, naval, and transportation applications. The AFC

global electrochemical reaction is the same as for other hydrogen air fuel cells, which is

the same as a balanced combustion reaction

→ H

O (7.10)

The global anode HOR is slightly different than the PEFC and PAFC:

+ 2OH

−

→ 2H

O + 2e

−

(7.11)

Electrochemical redox reactions have been called “cold combustion” by some, since the overall reaction is

similar, yet it occurs at lower temperature.

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

7.4 Alkaline Fuel Cells 411

DM - Teflonized

Electrolyte Matrix

+ 2OH

O + 2e

+ 2H

+ 4e

4OH

Cathode

Anode

Current collector/

flowfield land

Figure 7.23 Charge transfer processes at electrodes in AFC.

The global ORR on the cathode is also slightly different than the PEFC and PAFC:

+ 2H

O + 4e

−

→ 4OH

−

(7.12)

The AFC and PAFC are both liquid electrolyte solutions. The AFC, however, is based on

an alkaline electrolyte solution of potassium hydroxide (KOH) in water. Other alkaline

solutions can be used, notably sodium hydroxide (NaOH), but KOH is an inexpensive,

easily handled solution that is used in other areas such as agriculture, so a distribution

network already exists. The KOH solution molarity is typically between 30 and 80%,

depending on the operating temperature. A higher molarity reduces the vapor pressure of

the solution, and thus high-temperature systems require a high electrolyte concentration.

As shown in Figure 7.23, hydrogen oxidation occurs at the anode via reaction with the

hydroxide ion (OH

−

) in the electrolyte to generate water (in the PEFC and PAFC, water

is generated at the cathode). At the cathode, a balanced rate of hydroxide is formed via

reaction with water, electrons, and oxygen. This leads to a water management challenge

on the AFC, discussed in detail later in this section. The electrolyte solution conductivity

and vapor pressure are strong functions of water content. At high current density, the water

consumption at the cathode tends to dry the cathode and can even cause solidiﬁcation of

the electrolyte. In low-temperature AFCs, a ﬂooding problem exists on the anode where

water is generated. This ends up being less of a concern than for PEFCs where electrode

ﬂooding of the cathode is common, since the HOR is a facile reaction and hydrogen is

highly diffusive so that mass transport is not generally limited.

7.4.1 Operation and Conﬁgurations

Operating Conditions Unlike most other fuel cell systems that operate within a limited

range governed by material or kinetic parameters, the AFC has been developed and operated

with a variety of catalysts over a very broad range of temperature, pressure, and electrolyte

solution concentration. Table 7.1 shows some of the operational parameters for a selection

of AFC systems built and tested throughout the years. Although the electrolyte solution

must have sufﬁcient water for conductivity, like the PEFC electrolyte, the water generated

by the reaction can be used for this, and external humidiﬁcation is not needed.

The original AFCs developed by Francis Bacon had very admirable performance

compared to any modern fuel cell (0.8 V at 1 A/cm

). The anode and cathode pressure and

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

412 Other Fuel Cells

Table 7.1 Operational Parameters of Some AFC Systems

Catalyst

Temperature Pressure Electrolyte Anode–

System and Year (

◦

C) (MPa) Conﬁguration Cathode Performance

Bacon cell/1950s 200 4.50 Recirculating Ni–NiO 0.8 V at 1 A/cm

Apollo, 1960s 230 0.34 Static Ni–NiO 1.5 kWe/109 kg

Space Shuttle orbiter,

1980s

93 0.41 Static Pt/Pd–Au/Pt 12 kWe/93 kg

Siemens, 1986 80 0.22 Recirculating Ni–Ag 0.8 V at 0.4 A/cm

Russian photon

system, 1993

100 1.20 static Pt–Pt Efﬁciency 65–75%

Source: Data from [56].

temperature were 4.5 MPa and 200

◦

C, respectively, and the high temperature of Bacon’s

AFC enabled the use of non–noble metal nickel catalysts. The Gemini space program in

the United States utilized a PEFC, but the Apollo program (Figure 7.24) and early Space

Shuttles used AFCs to generate auxiliary power and potable water [56]. Future Space Shuttle

systems will apparently utilize PEFCs for auxiliary power [57]. In the Apollo system, the

sintered nickel electrodes were augmented with up to 40 mg/cm

platinum (cost was not an

issue for the Apollo program) to increase performance, despite reduced operation pressure.

The AFCs have also been developed for military applications, including the Swedish and

German navies, Russian aerospace applications, and power generation by Siemens [58].

Figure 7.24 Cluster of three 1.5-kWe (maximum 2.3-kWe) AFC units used on NASA Apollo

program. (Reproduced with permission from Ref. [58].)

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

7.4 Alkaline Fuel Cells 413

For automotive use, an array of thirty-two 5-kWe AFCs were constructed by Union Carbide

and used for propulsion in a General Motors six-passenger van in 1967 [56]. Additionally, a

6-kWe hydrogen/air AFC was installed by K. Kordesch in a lead–acid dual-cell hybrid, and

actually ran on public roads for three years with a six-bottle array of compressed hydrogen

on the roof [56], using air that was run through a soda lime bed to remove CO

. Modern

AFCs now operate at nearly the same conditions as PEFCs, at slightly elevated pressures

(∼ 2 atm) and around 100

◦

One of the major limitations of the AFC system is the inability to tolerate CO

, normally

found in air in levels of around 380 ppm (0.0383%) [59]. In industrialized areas or locations

with combustion activity, the local CO

concentration can be even greater. The CO

in the

air or in impure hydrogen reacts with the potassium hydroxide in the electrolyte to form

solid carbonate crystals through the following reaction [60]:

+ 2KOH

(

)

→ K

3(s)

+ H

O (7.13)

The carbonate particle formation reduces ionic conductivity and can block pores of the

electrode, leading to reduced performance. Carbon dioxide intolerance can be managed

in three ways: (1) operation on pure oxygen and hydrogen, (2) use of CO

scrubbers to

eliminate CO

from the input stream, or (3) use of a recirculating electrolyte (discussed

below) to remove the carbonate from the electrolyte and avoid buildup. Operation on pure

reactants is really only feasible for aerospace applications where pure oxygen and hydrogen

are available as fuel for liquid rocket systems. However, the AFC can be coupled with an

electrolyzer system to create a reversible fuel cell, discussed in Chapter 1. Use of CO

scrubbers to eliminate the CO

from the air stream is not 100% efﬁcient, and if using

hydrogen from reformer-based systems, scrubbing is also needed for the hydrogen feed

stream. The resulting equipment is expensive and adds to the bulk of the total system. A

recirculating electrolyte can be used to increase the CO

tolerance and lifetime of AFC

systems somewhat and is discussed later in this section.

Electrolyte System The AFC can be categorized into two main conﬁgurations, static

electrolyte and mobile electrolyte systems. A schematic of the mobile electrolyte system

is shown in Figure 7.25. In this system, the electrolyte is pumped from the stack into

an electrolyte reservoir. The mobile electrolyte is constrained within the porous electrode

structure either by asbestos or other porous separation layer between the electrode and

the mobile electrolyte or by careful control of the differential pressure in the anode and

cathode and the surface tension in the porous electrode structure as in the MCFC and

PAFC liquid electrolyte systems. The use of a mobile electrolyte offers the following major

advantages:

1. Heat Management The electrolyte can be used as a coolant to remove heat from

the stack, eliminating the need for a separate coolant system. The electrolyte is

passed through a radiator upon leaving the stack.

2. Water Management The product water from the anode can be removed from the

system through an evaporator, reducing the parasitic ﬂow requirements on the anode

and maintaining a proper water balance.

3. Electrolyte Poison Management The electrolyte is easily poisoned by small

amounts of CO

. Recirculation of the electrolyte can be used to ﬁlter and

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

414 Other Fuel Cells

Figure 7.25 Mobile recirculating electrolyte system in AFCs. Heat transfer, water removal, and

impurity removal occur through the ﬂushed electrolyte. (Reproduced with permission from Ref. [58].)

remove carbonate particles that form as a result of CO

impurity. Alternatively,

the electrolyte can be completely ﬂushed and replaced on a regular basis, given the

inexpensive nature of the KOH solution.

4. Electrolyte Gradient Management During operation, the electrolyte solution be-

comes concentrated (and possibly solidiﬁed) at the cathode and diluted at the anode

since water is generated at the anode and consumed at the cathode. This affects the

conductivity and vapor pressure of the electrolyte solution. Some of this gradient is

offset by water diffusion from the anode to the cathode. With a ﬂowing electrolyte,

convective ﬂow of the electrolyte minimizes these gradients.

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

7.4 Alkaline Fuel Cells 415

The major disadvantages of the recirculating electrolyte include the following:

1. The additional hardware and high-temperature pumps that are required to handle

the elevated temperature ﬂuid, transfer heat, clean impurities, and remove vapor add

to the bulk and reduce the reliability of the system.

2. Since all the electrolyte between individual plates is connected in the AFC sys-

tem, there is a potential for internal short circuits that rob power. This is managed

by maximizing the electrolyte path length between individual cells, by combin-

ing cells in a mixed series–parallel arrangement to reduce the maximum voltage

potential.

3. The recirculating electrolyte is affected by changes in orientation and gravity unlike

a capillary-pressure based static system. In zero-gravity situations, such as the Space

Shuttle, the choice of static electrolytes is partially due to the capillary pressure

control of these systems.

A schematic of the static electrolyte system is shown in Figure 7.26. In this conﬁg-

uration, an asbestos or other porous medium is soaked with the electrolyte solution and

maintained between the sintered metal or metal porous mesh electrodes. The advantages

of this approach are basically the opposite of the recirculating electrolyte system disadvan-

tages: The static system is more compact and simpliﬁed and does not suffer from internal

shorting of the electrolyte. However, the static system is almost certainly unsuitable for

operation on anything besides completely pure reactants, since CO

poisoning cannot be

remediated and damage will quickly accrue. Two other disadvantages of the static elec-

trolyte are heat and water management. A separate coolant ﬂuid must be used in the static

system to remove water heat, and the water generated at the anode must be removed using

excess ﬂow stiochiometry at the anode, which is parasitic.

A majority of terrestrial AFC applications use circulating electrolyte systems be-

cause operation on pure oxygen and hydrogen is not feasible, while space applications

demanding less bulk, more reliability, and zero-gravity operation use static electrolyte

systems.

Stack Conﬁguration The individual cells in bipolar plate stacks such as the PEFC are

typically connected in series, with current collection across the entire electrode surface

along the interface between the bipolar plate landings and the DM. The ﬂow ﬁelds in

AFCs are similar to those used in other fuel cells, and various parallel and serpentine

conﬁgurations are used to optimize mass, heat, and reactant/product transport.

A subset of stacks are designed using monopolar plates. Monopolar plates are used for

many AFC applications [61]. In this design, there is a PTFE sheet between the electrode and

the ﬂow ﬁeld to prevent the liquid electrolyte from passing through the electrode into the

channel, which can be by static forces or by weeping, which is caused by electro-osmotic

pressure-induced motion resulting from current ﬂow. The PTFE coating also prevents

electrical conduction between the land–DM interface, and the current cannot pass from

the anode of one cell to the cathode of the adjacent cell through the ﬂow ﬁeld plate.

Therefore, the individual cells do not need to be connected completely in series, as in the

bipolar plate design stack. Instead, they are connected in a series–parallel arrangement to

optimize power, compactness, and durability. A monopolar arrangement allows the unique

advantage of isolating single cells in the event of replacement or damage as also realized

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

416 Other Fuel Cells

Figure 7.26 Static electrolyte system in AFCs. Heat transfer, water removal, and impurity removal

occur through the anode and cathode ﬂows. (Reproduced with permission from Ref. [58].)

in SOFC bundle arrangements. In a series arrangement, if one cell is damaged, the entire

stack fails. Current collection from the electrode in the monopolar design occurs from a

metal current collector that frames and contacts the outside of the electrochemical active

area. Obviously, excellent current conduction along the electrode surface is needed to avoid

maldistribution of current and dead zones in the middle of the active area. As a result, this

approach is typically used only with metal mesh electrodes and small-area (<400-cm

)

AFCs [62].

Electrode Materials The AFCs developed by F. Bacon utilized nonnoble sintered nickel

metal catalysts. The high electrical conductivity of these porous electrodes permits use of

current collection from monopolar stack plates [62]. The sintered metal is often applied

in two separate layers, with large pores on the channel side, so that control of the liquid

electrolyte–gas phase interface can be achieved by capillary and gas-phase pressure forces.

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

7.4 Alkaline Fuel Cells 417

The use of Raney metals, which are physical mixtures of an active metal such as nickel

or silver with an inactive one such as aluminum, is another approach [26]. The aluminum

and active metal powder in Raney metals are not alloyed, and subsequent treatment with a

strong alkali solution dissolves the aluminum, leaving a porous active metal structure with

high electrical conductivity. The lower temperature modern AFCs mostly use noble metals

rather than nickel, although other spinels or perscovites can be used at the cathode with the

alkaline electrolyte [63].

Modern electrodes are similar to PEFC electrode structures with a porous carbon cloth

DM, consisting of a carbon support material with ﬁne metal catalysts, interdispersed with

PTFE for hydrophobicity and pressed onto a nickel mesh to improve in-plane conductivity.

As discussed, to prevent electrolyte weeping loss into the channels, a thin layer of PTFE

is used to coat the electrode, and a monopolar design must be used. This is not a problem,

however, considering the nickel mesh used on the electrode surface permits the high

in-plane conductivity required to enable efﬁcient current collection around the electrode

current-conducting frame.

7.4.2 Summary of Advantages and Disadvantages

Alkaline fuel cells are an older fuel cell technology, ﬁrst seriously developed and ap-

plied in aerospace and naval applications, where cost was not much of a concern and

pure fuel and oxidizer were available. For a variety of reasons, development of this tech-

nology has almost ceased, but it may be revived in the future if a solution to the disad-

vantages of the system are found. The basic advantages of the AFC system include the

following:

1. The highest demonstrated operating efﬁciencies of any fuel cell system due to the

reduced polarization during the ORR compared to acid-based electrolytes.

2. Use of a low-cost electrolyte, potassium hydroxide (KOH).

3. Flexibility in the choice of operating conditions and use of nickel-based electrode

structures in high-temperature AFCs.

4. Inexpensive raw materials of the stack.

The main limitations which have hindered development of the AFC to date include the

following:

1. An intolerance to CO

, which forces the use of CO

removal equipment, pure

oxygen and hydrogen, or a recirculating electrolyte system that severely reduces

system energy density.

2. The need to manage water and electrolyte removal from the anode and electrolyte

solution, which complicates system design and control.

Development of AFC systems was undertaken by several major companies and research

organizations for over 40 years, with interest waning since the middle 1990s. Alkaline

fuel cells represent a low-cost, high-efﬁciency system that does not necessarily need noble

metal catalysts. Siemens developed AFC technology for over 20 years and made substantial

progress, only to ultimately abandon research efforts in favor of alternative technology [64].

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

418 Other Fuel Cells

Ultimately, because of the high efﬁciency on pure oxygen and hydrogen and the need

to run on pure reactants, the AFC is best suited for niche use in reversible fuel cell systems

or aerospace applications. However, the intrinsic advantage of reduced ORR losses with

an alkaline electrolyte will always remain attractive. A breakthrough in CO

tolerance,

aCO

-tolerant alkaline electrolyte with similarly good ORR kinetics, or cost-effective

ﬁltration would likely spur renewed interest. Some CO

tolerance is now possible

[65], so there may be a future for this technology. Development of alkaline-based polymer

electrolyte fuel cells is also a promising future technology.

7.5 BIOLOGICAL AND OTHER FUEL CELLS

Just when it seems all of the possible combinations of fuel cells have been examined, there

are more. In this section, a very brief description of some of the fuel cell concepts with

future potential is given as a brief introduction. Biological fuel cells, or biofuel cells, use

biological processes to convert organic fuel directly into electrons. A microbial biofuel

cell uses living microorganisms for the electrochemical reaction. Since the organisms are

living, the system is very sensitive to environmental changes which can quickly kill off

the microorganisms. Enzymatic biofuel cells utilize redox enzymes as biocatalysts for the

redox reactions. Bioprocessing of water products can also be used to directly generate

hydrogen for conventional fuel cells. The energy density of these bio-based systems is

several orders of magnitude lower than conventional fuel cell systems, however. Excellent

reviews and summaries of biofuel cells are given in [66–68].

Metal air batteries are sometimes referred to as fuel cells. Really, they are hybrid

systems that operate on ﬂowing air like fuel cells but oxidation of stored metal like a

battery. Usually, an alkaline electrolyte is used in these system because of the favorable ORR

kinetics. Because the fuel is stored as a solid and the oxygen is not stored onboard, instead

absorbing from the ambient air during discharge, the system can achieve a higher speciﬁc

charge and energy density compared to full-battery systems. Essentially, this behaves as a

battery with the oxidizer stored in the ambient air. It is an energy storage system that does

not obtain a true equilibrium because the fuel is gradually depleting during discharge, and

therefore these are not really fuel cells. Reviews of zinc–air and seawater aluminum–air

systems can be found in [69, 70].

7.6 SUMMARY

The world of fuel cells comes in many varieties. Although PEFCs are featured in Chapter 6,

they are by no means the only option. Several other fuel cell systems can potentially reach

commercial production following current development trajectories. In a generic sense,

the advantages and disadvantages of each fuel cell fall along the lines of temperature

and electrolyte classiﬁcation. High-temperature fuel cells suffer from reduced maximum

thermodynamic efﬁciency and slow startup times but can use non–noble metal catalysts and

reform a wide variety of carbon-based fuel streams, while low-temperature fuel cells can

have rapid startup but have higher raw material costs and a need for pure hydrogen streams.

All liquid electrolyte systems suffer from vaporization losses and enhanced corrosion.

c07 JWPR067-Mench December 19, 2007 17:46 Char Count=

Application Study: System Design 419

Individually, the ﬁrst real development thrust was focused on the AFC for space

applications. The AFC has a broad temperature and pressure operation range and the

highest operating efﬁciency of the various fuel cell combinations due to the reduced ORR

polarization. However, its low power density and inability to handle CO

in air without a

scrubber or recirculating electrolyte conﬁguration have limited its usefulness.

The PAFC was the next system to be seriously developed following the AFC and is

the ﬁrst mass-produced fuel cell system to reach the consumer market in premium power

applications. Through several generations of design, stationary 200-kWe PAFC systems

showed reliable (>95%) service and high combined efﬁciency (>80%). Ultimately, the

total system cost has limited development of this technology.

Molten carbonate and solid oxide fuel cells are the high-temperature fuel cell technolo-

gies available. Both systems have the potential for direct internal reformation of carbon-

based fuel sources and used as part of a CHP or cogeneration system for achieving high

overall conversion efﬁciency. Issues with the liquid electrolyte loss and accelerated cata-

lyst and separator plate corrosion in the MCFC must be resolved, and SOFC systems will

continue to seek less expensive and more reliable manufacturing methods that increase

the stack component robustness. It is a desire of SOFC design to reach lower operating

temperatures approaching that of the MCFC (650

◦

C), so that metallic parts can be used.

This is being approached with anode-supported designs, novel electrolyte materials, and

ultrathin electrolyte structures.

There are also many other fuel cell technologies that are not yet mainstream but may

become mainstream in the future. Metal–air battery systems are really a hybrid between a

battery with stored fuel and a fuel cell with ﬂowing oxidizer. Biological fuel cells represent

laboratory-based nascent technologies with very low power density. However, the potential

advantages of these systems for power neutral waste remediation is worthy of future

development.

It is a common misconception that no fuel cell system has been marketed as a product.

Many have, and more are coming. The high-temperature systems even eliminate the need for

a hydrogen infrastructure, so that implementation of stationary power applications can occur

right away. Given the high efﬁciency, low emission, and relatively silent operation of PAFC,

SOFC, and MCFC systems, it seems highly likely that at least one of these technologies

will be come mainstream for stationary power applications in the coming decades.

APPLICATION STUDY: SYSTEM DESIGN

The ancillary components in many fuel cell systems consume 50–75% of the cost and bulk

of the system, are responsible for many of the degradation issues, and can consume 20–30%

of the gross power output of the system. In this assignment, choose a fuel cell system (e.g.,

a molten carbonate fuel cell for stationary power applications) and draw a process ﬂow

diagram of the system. Try to include all of the components necessary for operation,

including pumps, blowers, preheaters, and power conditioning. An example of a process

ﬂow diagram for a 3-MWe molten carbonate fuel cell operating on landﬁll gas is given

below. On a separate page, list the function of each component you show. Make sure you

have not left anything out. Will the ﬂow be able to come into and leave the stack? What will

happen to the unused fuel? Will the power be properly conditioned to suit the application?