Volume 4 fuel cells and hydrogen technology 4 09 – molten carbonate fuel cells theory and application

Volume 4 fuel cells and hydrogen technology 4 09 – molten carbonate fuel cells theory and application Volume 4 fuel cells and hydrogen technology 4 09 – molten carbonate fuel cells theory and application Volume 4 fuel cells and hydrogen technology 4 09 – molten carbonate fuel cells theory and application Volume 4 fuel cells and hydrogen technology 4 09 – molten carbonate fuel cells theory and application Volume 4 fuel cells and hydrogen technology 4 09 – molten carbonate fuel cells theory and application

T Leo,FuelCell Energy Inc., Danbury, CT, USA

© 2012 Elsevier Ltd All rights reserved

Glossary

Anaerobic digester A system for disposing of organic

waste materials which uses bacteria to break down the

waste, producing methane and carbon dioxide

byproducts Anaerobic digesters are used to process waste

at municipal wastewater treatment plants and at food

processing plants The gas byproduct can be used as a fuel

in some systems, depending on the tolerance for the

carbon dioxide diluent

Anaerobic digester gas The gas produced as a byproduct of

anaerobic digestion The gas is typically 60 to 70% methane

with the balance being carbon dioxide, plus traces of sulfur

compounds and other impurities

Anode An active fuel cell component functioning as a

negative electrode, where oxidation of fuel occurs,

producing electrons

Cathode An active fuel cell component functioning as a

positive electrode, where reduction of oxidant and

consumption of electrons occurs

Combined heat and power (CHP) An application of fuel

cells (or other types of power plants) where waste heat

produced during power production is used for applications

such as heating water, producing steam, or driving an

absorption chiller This heat utilization avoids the use of

fuel, providing cost and carbon emission savings

Direct FuelCell (DFC) A trade name for molten

carbonate fuel cell, which refers to the fact that

hydrocarbon fuels are sent directly to the fuel cell stack, where they are reformed to hydrogen In some other fuel cell systems the reforming reaction has been done in a pre-processor subsystem

Distributed generation Electric power that is generated where it is needed (distributed throughout the power grid) rather than in a central location Centrally generated power requires extensive transmission networks, while distributed generation does not

Electrolyte A material which allows transport of ions from anode to cathode In the molten carbonate fuel cell the electrolyte is a mix of alkali metal carbonates, and carbonate ions are transported from cathode to anode Fuel cell A device which converts the energy value of a fuel

to electricity electrochemically, without combusting the fuel The functional components of a fuel cell are the anode, the cathode, and the electrolyte matrix

Matrix A thin porous layer between the anode and cathode which contains the fuel cell electrolyte

Molten carbonate fuel cell A fuel cell which uses mixed alkali metal carbonates as the electrolyte The fuel cell operates at a high enough temperature for the carbonates

to become liquid and ionically conductive

Reforming Catalytic conversion of hydrocarbon fuel (such as pipeline natural gas or digester gas) to hydrogen-rich gas The hydrogen-hydrogen-rich gas serves as a fuel for the electrochemical reaction

4.09.1 Introduction

One of the earliest fuel cells to be deployed commercially are the molten carbonate fuel cells (MCFCs), which are high-temperature fuel cells that operate with a variety of fuels with high efficiency and very low emissions Research and development on the system has been conducted since the 1950s, and demonstration and commercialization activities were underway by the late 1990s The current state of the technology is such that it is best used for baseload power generation in stationary power plant systems[1]

Hydrocarbon fuels Natural gas/Digester gas

Steam

CH4 + 2H2O 4H2 + CO

2 + CO3 H2O + CO2 + 2e–

e–

Catalyst Electrolyte

Catalyst

e– 1/2O2 + CO2 + 2e– CO3

Air

Commercial power plants based on MCFCs have been available since 2003, and there are currently more than 60 MW of commercial power plants operating at more than 50 sites around the world

The success of the technology is due to key features related to the high temperature of operation The cells operate at 600–650 °C, higher than phosphoric acid or PEM cells, but lower than SOFC fuel cells At this temperature the reaction kinetics are such that the cells do not need noble metal catalysts to achieve good performance, but the temperature is not so high as to require exotic alloys or ceramics as materials of construction The absence of noble metal catalysts reduces fuel cell stack cost and reduces sensitivity to fuel impurities and carbon monoxide, which allows a high degree of fuel flexibility The operating temperature is also high enough to allow hydrocarbon fuels (e.g., natural gas or anaerobic digester gas (ADG)) to be used without an external reforming system These fuels can be reformed to hydrogen within the fuel cell stack This increases system efficiency and reduces cooling system cost, since the reforming reaction is driven

by waste heat from the fuel cell reaction [2]

The following sections describe the operating principle of internal reforming MCFCs, followed by a discussion of actual power plant configuration and applications

4.09.2 Carbonate Fuel Cell Chemistry and System Configuration

The operating principle is described schematically in Figure 1 When configured in ‘internal-reforming’ systems, the electrochemical cell stack is where hydrocarbon fuels are reformed into hydrogen A mixture of fuel and water vapor is sent to the anode chambers of the cells in the stack Reforming and shift reactions occur, which convert the hydrocarbon fuel to hydrogen:

Reforming: CH4 þ H2O →CO þ 3H2; ΔH650ºC ¼ þ53:68 kcal g mol− 1 Shift: CO þ H2O →CO2 þ H2; ΔH650ºC ¼ −8:5 kcal g mol− 1 Net: CH4 þ 2H2O →CO2 þ 4H2; ΔH650ºC ¼ þ45:18 kcal g mol− 1 The overall conversion is endothermic, which helps provide thermal management to the fuel cell stacks During power generation, more than half of the waste heat produced by the fuel cell reaction is consumed in the reforming reaction Water produced in the fuel cell anode reaction provides additional reforming reactant, and since the hydrogen produced by the reaction is continuously consumed in the fuel cell, the reforming/shift reaction is driven to completion, converting virtually all of the hydrocarbon fuel to hydrogen This complete conversion is obtained despite the fact that the fuel cell operates at significantly lower temperatures, and with less steam input, than would be used in a conventional external reformer system

The electrochemical reactions occurring in carbonate fuel cells are as follows:

At the anode The H2 produced by the reforming reaction reacts with carbonate ion CO3 2 − and produces H

2O and CO2 by the reaction:

2 −

H2 þ CO3 → H2O þ CO2 þ 2e− The CO2 produced in this reaction is sent back to the fuel cell cathodes by the power plant process system (described below)

Power to grid

Mechanical balance of plant

370 °C DFC exhaust

Natural Water

Fuel &

water preheat

Fuel cell anodes

Fuel cell cathodes

Catalytic oxidizer

DC/AC conversion

Fuel treatment

Water treatment

Fuel cell stack module(s)

Electrical balance of plant

Air preheat

600 °C

Air gas or biogas

At the cathode Oxygen from air and carbon dioxide from the anode reaction are reacted to form CO 2 −

3 as follows:

1�

2 þ 2e− → CO3 The CO 2 −

3 ions formed on the cathode electrode are electrochemically transported to the anode, where they are consumed by the anode reaction

These cell reactions require a process system that provides heated fuel and oxidant reactants to the fuel cells at a proper flow rate, and which provides a mechanism to deliver CO2 produced by the anode reaction to the cathodes A number of different system approaches are possible FuelCell Energy (FCE) uses a system that leaves some fuel unutilized to serve as energy input to heat up incoming air The overall power plant process is shown schematically in Figure 2

Water and hydrocarbon fuel (typically, natural gas or methane biogas) are treated to remove sulfur and other impurities, heated to stack temperature, and sent to the fuel cell stacks The reforming/shift reactions described above then convert the hydrocarbon fuel to hydrogen The stack anodes consume about 70% of the hydrogen produced by the reforming reaction The residual 30% hydrogen is used to preheat incoming air in a catalytic oxidizer The heated air is then sent to the cathodes at a temperature of approximately 550 °C Because the air was mixed with the anode exhaust stream in the catalytic oxidizer, the cathode inlet gas contains the CO2 produced by the anodes, which is required for the cathode reaction After the cathode reaction, the cathode exhaust gas exits the fuel cell stacks at approximately 600 °C, and it

is used to provide the preheat to the incoming fuel and water streams, in the power plant heat recovery unit (HRU) The exhaust gases are cooled to 370 °C in the HRU, and can then be used for a variety of waste heat applications, as described below

The theoretical operating temperature of MCFCs at 650 °C is 1.03 V [3] In practice, the cells typically operate at approximately 0.8 V per cell under rated load conditions This voltaic efficiency, combined with the 70% coulombic fuel utilization efficiency, gives

a 54% DC power generation efficiency After accounting for DC to AC power conversion losses and parasitic power requirements for balance-of-plant systems, typical power plant fuel to AC power efficiency is just under 50% on a lower heating value (LHV) basis

4.09.3 Cell Stack and Power Plant Design

Figure 3 is an illustration of the repeating cell components of the fuel cell stack The anode and cathode electrode structures are thin layers of porous nickel alloys The electrodes are laminated on either side of a thin porous ceramic lithium aluminate substrate, which holds the molten carbonate electrolyte within its pores

FUEL OXIDANT

CATHODE BIPOLAR

PLATE

The electrolyte is a mixture of alkali carbonates, which becomes liquid and ionically conductive at 450–510 °C The pore structure of the ceramic matrix, Ni anode, and Ni cathode layers are strictly controlled in order to achieve the desired distribution of electrolyte between the three layers Well-defined liquid–solid–gas interfaces in the cell structure provide the optimum amount of gas reaction sites and ionic mobility Development efforts to extend cell life have focused in large part on maintaining the desired pore size distribution over time Corrugated flow layers, made of 300-series stainless steel, are placed on either side of the anode/matrix/cathode lamination to provide the required gas flow passages for anode fuel gases and cathode oxidant gases Anode and cathode gases are separated by a stainless steel bipolar plate, which also provides the series electrical connection from each cell to the next

In current commercial fuel cell products, a cell stack consists of approximately 400 of these cell package structures When stacked together, the corrugations present gas inlet and exit passages on alternating faces of the stack Figure 4 shows a photograph of a commercial-scale stack, rated at 350 kW net AC output When configured into a stack module, stainless steel manifolds are placed against the appropriate stack faces to direct inlet gases or collect exit gas flows

The cell package and stack structures shown in Figures 3 and 4 are typical of the MCFC technology commercialized by FCE in Danbury,

CT, USA FCE (along with their partner in South Korea, POSCO Power) is currently the only company offering commercial fuel cell power plants based on MCFC technology FCE’s fuel cell system utilizes internal reforming, and it is called the Direct FuelCell (DFC) in reference

to the fact that hydrocarbon fuels are sent directly to the fuel cell stacks, without the need for an external reforming system

DFC300 Single Module Powerplant

300 kW

Single-stack module

DFC1500 One 4-Stack Module 1.4 MW

Cell package and

stack

Four-stack module

DFC3000 Two 4-Stack Modules 2.8 MW

FCE produces three standard power plant configurations based on the DFC design The systems are designated DFC300, DFC1500, and DFC3000 and produce 300, 1400, and 2800 kW, respectively As shown in Figure 5, the standard power plant systems utilize one of two modules: a single-stack module rated at 300 kW or a four-stack module rated at 1.4 MW Both modules use the same basic cell package and stack as described above

All of the systems operate at 47% LHV electrical efficiency with new fuel cell stack modules Current DFC fuel cell stack life is

5 years During this time, the power plant output declines to 90% of the initial rated output value, and the electrical efficiency declines by 10% (4.7 percentage points) to 42.3% At the end of 5 years of operation, the stack modules are replaced with modules with new cells, and initial performance levels are restored

In addition to the stack module, two other subsystems are used in each power plant system The mechanical balance-of-plant section contains the process equipment that prepares air, fuel, and water for use in fuel cell stack modules The electrical balance­ of-plant section converts the DC power into high-quality AC power Like other fuel cell systems, DFC power plants use solid-state static inverter systems, which provide additional benefits to local grid power quality beyond the power conversion function Inverters have the ability to provide or absorb volt–ampere reactive power (VARs), have low frequency distortion, and low fault currents These features make fuel cells more easily integrated into grid systems as distributed resources compared with power generation systems with rotating electric generators

4.09.4 Advantages of MCFC Power Plants

DFC power plant systems compete against a wide variation of power generation technologies, including the default option: the electric grid As a new technology, DFC systems are often more expensive than conventional power generation options, but they have key features which make them a good economic fit for many applications:

• High electrical efficiency, which results in reduced fuel costs and reduced carbon footprint per kilowatt-hour of generated power

• High temperature waste heat, which can be used for a variety of purposes, improving economics and further reducing carbon footprint

• Low emissions and low noise, which facilitate distributed generation installations on-site near power and heat users

The high electrical efficiency is put into perspective in Figure 6, which compares electrical efficiency for a variety of power generation systems over a wide range of system sizes No other commercially available power generation system can match the efficiency of fuel

Low Temp Fuel Cells

On-site distribut generation with

ed Grid support distributed

generation with

Central generation combined heat and

power

combine p

Combined Cycle

heat and ower

d

MCFC

Engines

Gas Turbine

Micro-Coal/Steam Turbine

Average US Fossil Fuel

20

10

0

System size (MW)

2800 kW power

560 kW unused heat Fuel

input

5957 kW

DFC powerplant

1013 kW High-grade use

1360 liters/hour hot water heated from 110 to 120 C

Or

1200 kg/h steam at 520 kPa

Or

1340 kW absorption chilling

1580 kW

exhaust

Low-grade heat use

1415 liters/hour water heated from 15 to 32 C

3157 kW

64% total thermal efficiency

90% total thermal efficiency

38 °C exhaust

370 °C exhaust

cells in distributed generation sizes Combined-cycle power plants start to approach DFC efficiencies at sizes above 10 MW, but for typical on-site power generation applications, DFC systems will be more efficient

Beyond electrical efficiency, DFC systems are almost always deployed in combined heat and power systems, where waste heat in the fuel cell exhaust is used to offset on-site thermal fuel use, further increasing efficiency and reducing carbon footprint The 370 °C exhaust can be used in heat exchangers to produce hot water, steam, or to provide energy to absorption chilling systems Figure 7 illustrates the thermal balance for a number of heat recovery options that could be used with a DFC3000 2.8 MW system High-grade heat uses include hot water, steam, or absorption chilling Enough energy is often left over to provide for low-grade heat recovery to produce additional hot water or to provide other heating functions In the examples shown in Figure 7, the high-and low-grade heat uses result in a total thermal efficiency of 90% [4]

In addition to high electrical efficiency and high-grade waste heat, the third major advantage of DFC fuel cells is very low emissions Table 1 compares the emissions of DFC fuel cell power plants to other distributed generation options and the average US fossil fuel plant The fuel cell systems are orders of magnitude lower in NOx and particulate emissions compared with conventional power generation systems The high electrical efficiency also translates to low CO2 emissions (which become zero when renewable biogas fuel is used)

4.09.5 Applications of MCFC Power Plants

DFC power plant systems are deployed in applications that produce clean power using natural gas fuel or applications that produce clean, renewable power using biogas-derived fuels Applications can also be split into two other categories: on-site power generation

to reduce grid power use and power generation for export to the grid

Table 1 Comparison of DFC power plant emissions to conventional power generation sources

(lb MWh−1) (lb MWh−1) (lb MWh−1) (lb MWh−1) Average US fossil fuel plant 5.06 11.6 0.27 2031

Small gas turbine (250 kW) 1.15 0.008 0.08 1494 DFC fuel cell on natural gas 0.01 0.0001 0.000 02 940 DFC fuel cell on natural gas with CHP 0.006 0.000 06 0.000 01 550 DFC fuel cell on biogas 0.006 0.000 06 0.000 01 0 Source: Emissions estimates for nonfuel cell sources are from ‘Model Regulations for the Output of Specified Air Emissions from Smaller-Scale Electric Generation Resources’, 15 October 2002, The Regulatory Assistance Project, for the National Renewable Energy Laboratory (NREL) [5]

4.09.5.1 Self-Generation Applications

DFC power plants have been used in on-site power applications to meet the baseload power requirements of a wide range of municipal, military, commercial, and industrial customers including wastewater treatment plants, manufacturing facilities, office buildings, hospitals, universities, and military bases Many of the earlier self-generation projects used the smaller DFC300 product, but as the technology has been proven, the demand has been shifting to larger installations, which have a lower installed cost per kilowatt Figure 8 shows an example of a self-generation project at an industrial bakery in Connecticut This early generation DFC1500 system, rated at 1.2 MW, provides more than half of the facility’s power requirement, and excess heat from the power plant is used to support bakery processes, offsetting fuel use in plant boilers After the fuel cell exhaust flows through the steam generator, it flows to another heat exchanger which is used to provide preheat to a facility gas stream This two-level heat recovery gives the project a very high thermal efficiency, enhancing the economics and greenhouse gas reductions achieved by the system The stack module can be seen in the right of the photograph, and the equipment on the left of the picture is the DC–AC power conversion system The power plant exhaust (in the center of the photograph) exits vertically and is then directed down and behind the unit to the steam generation system

Figure 9 shows another on-site combined heat and power (CHP) application, in this case a 1.4 MW DFC1500 at a university in California In this view, the hot water heat recovery system can be seen just behind the power plant equipment

These are two typical examples of combined heat and power installations in on-site self-generation projects This type of project makes up about 40% of the 67 MW of MCFC power plants operating around the world The remaining 60% are applications where power is exported to the utility grid instead of consumed on-site

4.09.5.2 Grid Support Applications

In the grid support market, power is sold into the electric grid, at prices set by feed-in tariffs or by pricing mechanisms under Renewable Portfolio Standards (RPS) programs These projects tend to be larger scale than the self-generation applications The

Figure 9 DFC1500 power plant in combined heat and power application on California State University, East Bay campus

DFC3000 power plant is often being used in multiple-plant configurations for grid support systems The most significant grid support market for fuel cells is in South Korea, where an RPS pricing mechanism provides favorable pricing for fuel cell power Figure 10 shows the largest fuel cell installation in the world, an 11.2 MW site with four DFC3000 power plants in Daegu City, South Korea The system produces power for sale to the local utility under a long-term power purchase agreement, and provides heat

to local neighborhood users South Korea has supported fuel cell grid sales through a feed-in tariff mechanism which has now been transitioned to a renewable energy credit (REC)-based pricing scheme under their national RPS program In South Korea, the high efficiency and very low emissions of fuel cell power plants qualify them for inclusion as renewable power, even when using natural gas fuel The same is true for the state of Connecticut in the United States (and four other states), which has a feed-in tariff type program targeted at establishing 150 MW of renewable, clean generation in the state The state of California is in the process of enacting feed-in tariffs for combined heat and power and renewable power sources, which could support additional grid support fuel cell installations

Driven largely by the scale of the projects, the grid support application is the largest growing market segment for DFC fuel cell power plants Unlike the self-generation application, it is not always possible to identify local users for waste heat from these grid-connected systems Because of this, power plant developers have been evaluating approaches to converting waste heat from the systems into additional power One of the most straightforward of these approaches is to use DFC waste heat to drive an organic Rankine cycle (ORC) system An ORC is a power generation system similar to a steam turbine plant, except that it uses an organic working fluid with lower boiling point than water, to capitalize on lower-temperature heat sources ORCs are used to

Figure 11 320 kW sub-MW DFC/T power plant during field test at Billings Clinic in Billings, MT

recover industrial waste heat or to produce power from geothermal sources When operated from the waste heat of a DFC power plant, the additional power from an ORC bottoming cycle can increase the electrical efficiency of the power plant from 47% to approximately 50%

FCE has also developed a hybrid cycle in which a gas turbine generator is integrated into the fuel cell system at the point of highest temperature (after the anode gas oxidizer) in order to achieve even higher efficiencies The gas turbine is unfired, and runs entirely on waste heat from the fuel cell process The technology has been demonstrated at the sub-MW scale Figure 11 shows a prototype system that was based on the DFC300 product (rated 250 kW at the time) combined with a Capstone microturbine, for a combined rated output of 320 kW The unit was tested at FCE’s Connecticut test facility and then operated

in the field at a hospital site in Billings, MT, USA The maximum LHV electrical efficiency demonstrated was 58% [6] The hybrid system, designated the DFC/T, achieves a higher efficiency than exhaust-based bottoming cycles (like the ORC) because

it is driven by higher temperature heat in the fuel cell system FCE is developing megawatt-scale DFC/T designs for commercialization in the near future

A high-efficiency configuration that is commercially available today involves the use of waste heat from DFC fuel cell systems to support power generation from natural gas pressure letdown energy recovery Natural gas is transmitted over long distances at high pressure, and then reduced in pressure for distribution in local markets This pressure reduction is done at gate stations near major markets, and it is usually done with pressure reducing valves Enbridge Inc., a major natural gas distribution company, and FCE have codeveloped a hybrid system in which the pressure reduction is done in an expansion turbine, which drives an electric generator In order to prevent excess cooling of the expanding natural gas, waste heat from a DFC power plant is used to preheat the high-pressure gas before expansion This system, called the DFC-ERG (for Energy Recovery Generation) has been demonstrated at Enbridge’s headquarter, Toronto, Canada, in a 2.2 MW system consisting of 1.2 MW fuel cell generation and 1 MW expansion turbine generation The maximum efficiency demonstrated by this system has been 70% on an LHV basis [7] The power plant is shown

in Figure 12 The fuel cell system, with four single-stack modules, is shown in the foreground of the photograph, and the turbine generation and heat transfer equipment is behind the fuel cell system

4.09.5.3 Renewable MCFC Power Plant Applications

As described above, MCFC system currently available in the DFC configuration are designed for internal reforming of methane-based fuels, which makes them a good fit for biogas fuels which are predominantly methane, such as ADG ADG differs from pipeline quality natural gas in that it has higher levels of contaminants, such as sulfur compounds and siloxanes, and the methane tends to be diluted with CO2, another by-product of the anaerobic digestion process ADG from municipal wastewater treatment processes consist of about 60% methane and 40% CO2 ADG from brewery or food processing waste digestion has about 70% methane concentration [8]

MCFCs are particularly well suited for operation on these bio-derived fuels, due to its insensitivity to the CO2 diluent As described above, a unique aspect of the carbonate fuel cell chemistry is that CO2 is produced in the anodes and consumed in the cathodes Because of Nernst effects, the presence of CO2 diluent in the fuel will reduce anode performance and improve cathode performance In practice, the cathode gain is roughly equal to the anode penalty – the DFC power plants perform about the same in the presence of the CO2 diluent, as long as methane content is above approximately 50% Figure 13 compares performance from a test stack operating on pure methane versus operation on methane diluted with CO2 The results show that the CO2 diluent had no impact on cell performance This has been born out in many commercial projects at municipal wastewater treatment plants using ADG with 60% methane concentration In fact, a small performance gain (vs natural gas) has been observed at sites with 70% methane/30% CO2 gas compositions, which suggests that at this concentration the cathode gain is slightly more than the anode performance loss

0.500 0.550 0.600 0.650 0.700 0.750 0.800 0.850 0.900 0.950

100% Methane Simulated DG, 60% Methane/40% CO2

Current density (mA/cm2)

As of this writing, there are approximately 9 MW of DFC power plants operating on biogas at 11 sites around the world, mostly in California Most systems are at municipal wastewater treatment sites, but systems are also installed at food processors and breweries Figure 14 shows a 1.4 MW DFC1500 system at a municipal wastewater treatment plant at Turlock Irrigation District in California Figure 15 shows a system at a wastewater treatment facility in Tulare, California, consisting of three sub-MW DFC300 systems After several years of operating these systems, the municipality installed a fourth unit to utilize more ADG and increase the capacity of power generation at the site [9]