.
8

LA-UR- 87:2393

LOS Alamos h’al,onal Laboralor) IS operated by ~ae un,ver$,ty O! Cal,torma tor the unltect Slale$ Oeparlmenl 01 Energy “ntj~, CWI,KI w.7405. Er.JG.3fj

TITLE FUEL CELLS FOR EXTRATERRESTRIALAND TERRESTRIAL APPLICATIONS

AUTHOR:S) SUPRAMANIAMSRINIVASAN, (MEE-11)

LA-uR--87-2393

DE87 013160

SUBi\41TlED TO Symposium on “’El~ctrode Materials and Processes for Energy
Conversion and Storage,” 171st Meeting of the Electrochemical
society.
PhilcidlIlphia, PA
Mav 1O-15, 1987

DMCLAIMER

I“hls rcp,:l WIIS prcpnrcd usun uccount of work spmrnwcd hy ur. ugcncyof the United Stutcs

(h)vcrntl,cnt. Nciil?r(h [lrlltc[l St/ltcs (iovcrntncnl rim unytigcncy lhercof, nnrunydthcir

ctnployccs, mukm ilny wurrllnly, cxprcsf or implied, or uwurncs urty lcgd Iiuhility nr rcsponni.

t~i!ily f,w the uccdrncy, conlplclcrww, (w u~cfulncw of uny informutim, uppHrulu*, prducl,or

prt.cw diwlmcd, {Jr rcprc+cnts 1}1111its usc WIWIII trot infringe privutcly owned righ(t, Refcr-

cncc hcrcln II) uny spccii’ic ctmmcrciul pr,nlud, prmcw, or scrvicc hy trudc numc, Irndcrmrrh,

munufllcturcr, or otherwise (km nut ncccwwrily constitute or imply its cndmxcrncnt, rcmrl-

ttwnlltlt(m, or (Iworing hy Ihc [Jnilcd SIUICN (hwctnmcnt or uny ugcnq (hereof I’hc vicwx

,Ind ,Iplrlt(ms I)f Ilu!hors cxprcwcd herein do not ncccwwrily \tmv (w Icflct’t IhIIXC of !hn

II IIIIC(I Stlltcq(;t)vcrnltlctll or, lnyllgcncythcrcttf

~va(, P1,ln~,CPt, fth, \a~l, <l? , P ,,,,,, . v ,wL~,I,,wbti!,aI I,WU a LNUVWI,,t*,Wtttitw,~,tINMtou,ovaL,ti.,vv tuy*, tv, tewo,&@ttgw! uuuu,, bt, otfepfoti”cq

ltlpptof!l,tt cd 1(,*n, O! lb!< tn.,~t,,jl,n?, nf lo nllow olh~ft to do 10 fo~ () S (iOvafnmenl Duf Doi?!

T!,r I(IN AIn@~II-. NdtI(IfIfi I iat),,a,i),v tw,@BI% lhnl II!P c)ut,I,qhe. ,dmnt, tv Ih, n ntl,cl?aswo?h Wvlofmea undarlho ausp,coaolthous @@n?lm@t\tnlln@tg)

. . .—

MASTER
Losmk?ilmosLos Alamos National Laboratory

About This Report
This official electronic version was created by scanning the best available paper or microfiche copy of the original report at a 300 dpi resolution.  Original color illustrations appear as black and white images.

For additional information or comments, contact: 

Library Without Walls Project 
Los Alamos National Laboratory Research Library
Los Alamos, NM 87544 
Phone: (505)667-4448 
E-mail: lwwp@lanl.gov



FUEL CELLS FOR EXTRATERRESTRIALAND
TERRESTRIAL APPLICATIONS

S. Srinivasan
Los Alamos National Laboratory

?!echanical & Electronic Engineering Division, ?ISD429
Los Alarnos, New Hexico 87545

ABSTRACT

The fuel cell is a nineteenth century invention and a twentieth century

technology development. me to the high power and energy density, high

efficiency, reliability, and production of pure water, hydrogen-oxygen fuel

cell systems have no competition as auxiliary power sources for space

vehicles. The alkaline fuel cell system is a well developed and proven

technology for this application. The solid polyner electrolyte eyutem may be

its future competitor. The energy crisis of 1973 stimulated research,

development ●nd demonstration of the phosphoric acid, molten carbonate, solid

oxide and solid polymer electrolyte fuel cell systems using natural gas,

petroleum or coal derived hydrogen (and carbon monoxide for the high

temperature uyetems) for terrestrial applications. The direct methanol-air

fuel cell is still an electrochemist’s dream, Though considerable

technological advances have been made, tha prenent price of cmde oil, and the

high c~pital cost-s and limited lifetime of fuel cell systems impede their

terrestri~l application in the developed countries, Conversely, tha

potential for lower capital COSLS Of labor intensive manufacturing processes

and the r~latively higher fossil fuel prices make these eyu~ems more

atti”activo for such applications in the developing countries,

~HE FUEL CELLS - A NINETEENTH CENTURYINVENTION AND ~ENTImH CENTURY

~~CHNOLWY Dl!VELOPHEtJT——— _ —.

Though Sir william Grove discovorod fuel cells as early as 1839, it was

only in tho 1960’s that fuel cell oystem. found their flrot major

application-auxiliery power sources for the Gemini opsce flights. However,

a number of concepts relating to tho ●dvantages of convort!ng chemical energy

to electrical energy, typas of fu~ls, and tcmperaturo rmngeo for oporation

6978K



were alluded to in the research work during the intervening 120 years.
1-3

Though the historical development of fuel cells has been covered adequately in

several books and review articles, it is worth summarizing the highlights

the progress made in this period.

● Sir William R. Grove demonstrated the first fuel cell in 1839 by

of

supplying hydrogen to one platinum electrode and oxygen to another with

both electrodes contained in sulfuric acid. He connected six of these

fuel cells in series and used the electricity generated to decompose

water. Platinum is still the best elactrocatalyst for aqueous

electrolyte fuel cells!

● Becquerel devised a fuel cell in 1855 consuming carbon by immersing it in

molten nitre (potassium nitrate) contained in a platinum vessel.

Platinum also served as the counter electrode. The faradaic efficiency

was low because of the direct chemical oxidation of carbo~ by potassium

nitrate.

8 ?lond and Langer (1889) developed a fuel cell battery in which platinum

sheets with many small holes and covered with platinum black ware used as

electrodes, Dilute sulfuric acid, retained in a porous plaster-of-Paris

matrix, served as the electrolyte.

* Ostwu!d (1894) recognized the potentially high efficiency of fuel cells,

i.e., the conversion of the free energy chanve of a chemical reuc;tion

directly to electricity.

● Jacqu@s (1897) designed and built a 1,5 Icw fuel cell b~ttery, Each cell.

consisted of an iron pot containing fused sodium hydroxide into which was

placed a carbon rod as the consumable anode,

● The pioneer of the fuel CO1l nra in the twentieth century is Francis T.

Bacon, He started the development of ● hydrogen/oxyBen fuel cell battery

in 1932 A 5 kw fuel cell power plant was built and tested in 1952 In

2 69:8K



orcler to stabalize the gas and electrolyte within the electrodes, these

were of a dual porosity construction (Teflon was not discovered at that

time) . The single cell consisted of a

lithiated nickel oxide cathode and 85%

and pressures were 200-240”C and 39 to

sintered nickel anode, sintered

KOH. The operating temperatures

40 atm.

● The Bacon fuel cell system served as thl~ foundation for the fuel cell

power plants (auxiliary power sources) for the Apollo Space Vehicles,

developed by Pratt and Uhitney Aircraft., Division of United Technologies

Corporation.

The energy crisis of 1973 was the impetus for a large number of research

and development programs on fuel cell technology for terrestrial

applications. The status of fuel cell technologies for extraterrestrial and

terrestrial applications is summarized in the next two sections, while the

final section deals with the prognosis of economics and applications Of fuel

cell systems.

Before dealing with these topics, it if worthwhile recollecting Some fuel

cell principles in respect to systems engineering aspects:

● The fuel determines type of fuel cell and fuel processor.

● The operating temperature determines method of product heat and water

removal.

● The application determines power rating, need for bottoming cycle,

pow?r conditioning and scale-up considerations.

● Start-up time and peaking capabilities are of primary concern in

respect to developing fuel cells for tranapf)rtation ●pplications.

?t~
The heart of the fuel cell system is the electrochemical cell stack.

‘ ‘,S

.b

Its basic components are the porous gas diffusion electrodes,

electrolyte which may or may not be held in a matrix and the bipolar
;+ plnte , Their choice design and fabrication procedure is ●gain
-J’

) determined by the primary fuel ●nd by the operating temperature and

pramuure of the fuel cell power plant.

3 6978E



THE SPACE AGE - THE FIRST H.AJORAPPLICATION OF FUEL CELL SYSTEMS

The main advantages of fuel cell systems for space applications are high

power and energy density with respect to weight and volume, high efficiency,

few moving parts, minimum noise and vibration and reliability. For this

application, the logical fuel and oxidant are cryogenic hydrogen and oxygen.

The Gemini space flights used a 1 kw General Electric solid polymer
4

electrolyte fuel cell. The distinctive feature of this cell is that a

solid electrolyte was used in the form of an ion exchange membrane

(polystyrene sulfonate). The two electrodes which consist of the

electrocatalyst (finely divided platin~) and a plastic material for

wet-proofing the electrode are incorporated on fine metallic (platinum or

titanium) wire screens. These electrodes are bonded on both sides of the

solid polymer electrolyte. Metallic current collectors are ribbed on to each

electrode. The heat renoval was by circulating a coolant through tubes in the

bipolar current collector to a space radiator. Wicks were used on the cathode

side of the current colloctor to absorb the ]~roduct water and carry it over by

capillary action. The water (1 pint/kwh) was used for drinking purposes by

the astronauts.

The Gemini fuel cell ~;ystem was used for missions of duratioa about 1-2

weeks. Though it satisfied the perfomnance requirenlentr for such flights,

there were some problems with this technology - the power der~sities attained

were not high enough; the polystyrene sulfonate ion exchange membrane

electrolyte ws not stable u~cler the electrochemical environments in the cell;

and, the platinum loading was quite high. Thus, the alkaline fuel cell

technology was chosen for the A,Q01105 and space shuttle6 flights.

The Pratt and Whitney Division of United Technologies Co~oration

developed alka]ine fuel cell systems for the Apollo flights - these fuel cells

used the B&con fuel cell technology with dual porosity electrodes (nickel for

the anode and lithiated nickel oxide for the cathode). The dual porosity

electrodes served the purpose of stabilizing the electrolyte meniscus within

the elect~-ode. Unlike the Bacon fuel cells, these cells used a high

concentration potassium hydroxide (80-85% whi~h is practically the molten

salt), at a higher temperature (250” rather than

atmospheric pressure; r~markable improvements in

200SC) and close to

performance have been

69786



reported in futuristic versions of the alkali fuel cell power plants for the

space shuttle flights and these are illustrated in Figure 1. The electrodes

are Teflon bounded and contain unsupported platinum (anode) and gold-platinum

alloy electrocatalysts. The main difference in the electrochemical cell is

the use of an extremely thin (5o vm) electrolyte matrix made with potassium

titanate. Figure 1 shows that extremely high current densities are achieved

in the purely activation controlled region (about 0.5 A/ctr12 at 0.9 V), while

the slope of the linear region is amazingly low (0.05 ohm cm2). This low

vaiue of the slope is essential for attaining high current densities (up to 10

A/cm*) at moderate cell potentials (0.5 V). The Apollo fuel cell systems

(rated power about 2 khl) performed quite satisfactorily for missions which

lasted 2-4 weeks. For the space shuttle flights of the 1970’s and 1980’s,

United Technologies Corporation/International Fuel Cells Corporation,

developed a high performance alkaline fuel cell systeu (12 W), with platinum

anodes (10 mg/cm2 of Pt), and gold (90%) - platinum (10%) cathodes (20

mg/cm2 of Au-Pt alloy). The electrolyte is 35% KOH and is immobilized in a

reconstituted asbestos matrix. An electrolyte reservoir plate is on the anode

side to supply electrolyte when needed. The cells operate at about 60-70”C

under a few atm pressure. Current densities of about 1 A/cu12 at cell

potentials of 0,80 V are obtained in these cells. The objectives of BJASA are

to further improve the performance of these cells by increasing the operating

temperature to about 120°C, For this purpose, the gold-platinum

electrocatalyst will have to be modified to i~rove the performance and

increase the stability, perhaps by incorporating a third component in the

alloy,

A strong competitor for the alkaline fuel cell system is possibly the

advanced solid polymer electrolyte fuel cell. Recent results from Siemens,

Gemnany; Ballard Technologies Corporation, Canada; and Los Alsunos National

Laboratory (LANL), IJSA, strongly indicate the 6ttainment of high power

densities with stablc~ membranes (for example, Mefion and Dow Hembranes) and

low noble metallic loadings (0.5 mg/cm2). Current densities of 5 emp/cm2

at a cell potential of 0.5 V, 80”c and 8 atm pressure have been reported by

Ballard Technologies corporation and Siemens in s~lid polymer electrolyte (Dow

membrane) fuel cells with TefJon bonded electrodes containing 4 mg/cm2

unsupported platinum. The reactants are pure H2 and 02. The Los Alamos

6978E



National Laboratory investigators are focussing on achieving high power

densities in such fuel cells using low noble platinum loading
6-8

electrodes. The manner in which this is achieved is by impregnation of a

proton conductor (e.g., tJafion) into a *’conventional*” porous 6as diffusion

electrode used with liquid electrolytes. The rationale for this approach is

that the reaction zone for a fuel cell reaction at a solid electrodelsolid

electrolyte interface is <lose to a *’three phase line.” Impregnation of the

proton conductor into the electrode structure considerably extends the three

phase zone and approaches that in the case of a liquid electrolyte. Current

densities of 1 A/cm2 have been attained at cell potentials of 0.5 V. The

slopes of the linear region in the cell potential–current density plots are

0.1 and 0,2 ohm cm2 for the Ballard Technologies Corporation and LANL single

cells, respectively, The main reason for the lower slope, which is central

for attainment of higher current densities appears to be the use of the Dow

membrane in the former and the Nafion membrane in the latter case. The Dow

membrane used by Ballard Technologies Corporation has a smaller thickness (125

vs 1751Jsn), higher acidity and better water retention characteristics than

the I#afion membrane used in the LANL single cells. The Ballard fUel Cells

were also operated at a higher pressure (8 vs 5 atm) than the WL ones.

THE ENERGYCRISES OF 1973 - A RENAISSANCEOF FUEL CELL TECHNOLOGYFOR

TERRESTRIAL APPLICATIONS

The NASAR&D programs on fuel cell technologies initiated in the 19S0’S,

stimulated similar programs in the USA and several European countries during

this and succeeding decade for terrestrial applications. Considerable efforts

were made to use fuels such as natural gas, the higher hydrocarbons, methanol,
1

ethanol and other organic fuels, directly in fuel cells. In the latter

part of the 1960’s, the interest faded away because of the low efficiency and

power density achieved in fuel cells which use organic fuels dir6ctly as the

anodic reactant. United Technologies Corporation (UTC), with their strong

‘“forte” in alkaline fuel cell technology for extraterrestrial applications,

also provided leadership in the development of fuel cells for terrestrial

applications. Even prior to the energy crisis of 1973, UTC embarked on a

progrm with some support from electric and gas utilities to develop fuel cell

6



power plants utilizing natural gas. Such types of fuel cell power plants

being considered as substitutes for gas turbines whic~ were and still are

being used as peaking devices in electric utility power plants.

were

The energy crisis of 1973 was the ●’rationale*’ for renaissance of fuel
3

cell technology for terrestrial applications. The Energy Research and

Development Administration (ERDA which subsequently became the U. S.

Department of Energy) and the Electric Power Research Institute (EPRI)

stressed the need for conservation of the petroleum fuels and for the

increasing utilization of coal, natural gas, and renewable energy resources.

The potential benefits of fuel cell power plants for peak shaving, load

following and intermediate and base load power generation in electric utility

applications are:

● The chemical energy of a fuel is directly transfomned into electricity.

Thus, the theoretical efficiency is not limited by the Carnot cycle as is

the case for conventional themal power plants.

9 Fuel cell power plant efficiencies are independent of the rated power

above say about 100 W. This is not so with oil or coal burning power

plants, gas turbines, and diesel generators, where the efficiency is

reasonably constant only at

● Fuel cells can be made in a

permits locating such power

the megawatt power level.

variety of sizes and they are modular. This

plants at several grid systems to reduce

transmission line costs,

con~ested urban centers,

facilities are expensive

● The siting advantages of

This advantage is particularly important in

where the needed transmission and distribution

to install;

fuel cell power plants can be extended because

of their capabilities for waske heat utilization (as for example, space

heating, hot water, and absorption cooling). The combined efficiency for

electricity and heat production can be as high as 90%, This is quite

attractive for gas utilities having extensive gas distribution systems

~%,

,* \&

\l

i’ 7
t

~. ‘ .,)

\u
u“

\

6978E



● Fuel cell power plants make a lot of sense for installation

industries which need both electricity and high grade heat.

in large

In the late 1970’s, programs were initiated at Brookhaven National

Laboratory and LOS Alamos National Laboratory for the investigation of fuel
9

cell power plants for vehicular power plants. Some of the above advantages

of fuel cell power plants over thermal ones for electric and gas utility power

generation apply equally well for vehicular power plants. The significant one

is that the projected efficiencies for fuel cell power plants are at least

twice as high as internal combustion and diesel engines.

In the 1970’s, it was realized that because activation overpotentials for

anoclic oxidation of organic fuels are too high and poisoning problems by their

intermediates are encountered, it is necessary to process these fuels to

hydrogen prior to entry into the electrochemical cell stack. In the high

temperature fuel cell systems, carbon monoxide is also an efficient anodic

reactant, Natural g~.,s and coal are the primary fuels of most interest for

terrestrial applications of fuel cells. Methanol, a liquid fuel which is mass

produced from natural gas and in the future possibly from coal, is another

attractive fuel, particularly for transportation applications. This fuel is

the most electroactive organic fuel but its activity is about three orders

less than that of hydrogen, Thus , at the present time, even methanol will

have to be ste~-refo~ed to hydrogen before entry into the anode gas

chamber. The processing of organic fuels restricts electrolytes to carbon

dioxide rejecting ones - phosphoric acid, molten carbonate, solid oxide, and

solid polymer electrolyte (the last is with processed methanol as fuel). The

following subsections briefly describe the status of each of these

technologies. For more details, the reader is referred to recent
3,10

publications,

Phosphoric Acid Fuel Cell Systems: Apart from the alkaline fuel cell systems

for space applications, this technology is the most advanced. Systems ranging

from 1 khl to 5 w have been designed, built and tested. The most advanced

version ~perates at 205”C, 8 atm pressure, at a current density of 325

mA/cn12 and 0.7 V/cell. The noble metal loading of the anode is 0.25 mg/cm2

8 6978E



and of the cathode is 0.5 mg/cm2. Some platinum alloy electrocatalysts

(with Cr, V) SI.OWimproved performance but lifetime is a problem. These

sygtems are being considered for load following in electric utilities and on

site integration energy systems. The chemical + electrical energy

conversion efficiency at the rated power level is about 40% and chemical +

(electrical & heat) energy conversion ~~fficiency is over 8~. Lifetimes Gf

over 5 years have been attained on demonstration systems.

The materials used for the fabrication of components in a phosphoric acid

fuel cell are shown in Fig. 2. The anode and cathode consist of porous gas

diffusion electrodes. These electrodes earl be prepared by one of three

methods. The spraying ❑ethod (Fig- 3), many steps of which can be automated,

is the most conunonly used. Filtration and rolling methods are also used. The

fomer is good for preparing experimental electrodes for research purposes.

The rolling method is somewhat tedious, but result in electrodes which are

reproducible from the point of view of electrochemical perfownce. The

silicon carbide matrix (about 50-100 pm thick) iS generally deposited on top

of an electrode (say, cathode) by a wet process which involves preparation of

a silicon carbide - Teflon emulsion and Bpra:*ing or continuous feeding to the

electrode . The purpose of the small quantity of Teflon is to serve as a

binder, The electrode-silicon carbide assembly is then heated to dry the

matrix and sinter the Teflon in it. The porous carbon serves three pu~oses -

it is grooved to permit gas distribution; in the United Technologies

Corporation design it serves as an electrolyte reservoir; and it is a current

collector.

The active layer in the electrode consists of platinum supported on

carbon and Teflon. The Teflon content in the layer is about 40%. Platinum

crystallite are deyosited on the high surface area carbon generally by m

collcidal method followed by heat pretreatment. The aim is to have platinum

crystallite with a diameter of about 50 A“. An electron microscope

photograph of the supported electrocatalynts iB shown in Fig. 4. One of the

perfomed degradation mechanism~ in fuel ceils electrodes is the agglomeration

or sintering of platinum cry8tallites, causing loss of electrochernicelly

active surface area.

phenomenon is due to

e
!-- ‘

15 is generally qccepted at the prysent timo that khis

surface migration of platinum

9



crystallite, iollowed by their coalescence. The low

significantly influenced by the electrolyte. Another

temperature sintering is

cause for performance

degradation is poisoning of mainly the anode. Poisons which can be

encountered are CO and H2S from reformed fuels, organic impurities in cell

component materials and metallic impurities, from say cooling pipes.

Plans are underway to build 11 and 7 1/2 UWphosphoric acid fuel cell

power plants by International Fuel Cells Corporation (a division cf United

Technologies Corporation) and Westinghouse Corporation. Figure 5, a schematic

of the fotmer one, demonstrates that the electrochemical cell stack is a small

but vital component of the entire system. These power plants are being

considered for ‘*load followers” by the electric utilities. Such types of

power plants are being strongly identified in Japan for the stated and

dispersed power generation applications. Toshiba is in International Fuel

Cells Corporation for the development and installation of these

plants.

Molten Carbonate Fuel Cell Systems: This is en ideal system to

types of power

be used with

reformed natural gas and gasj.fied coal. Progress has recently been made at

Energy Research Corporation for the internal reforming of natural gas (i.e.,

with catalyst system ahead of the anode). This eliminates a major component -

the fuel processor. Carbon monoxide produced from natural gas or coal is a

reactant and not a poison for the molten carbonate fu 1 cell system unlike in

the phosphoric acid fuel cell system. The components of the single cell in a

molten carbonate fuel cell are represented in Fig. 6, The electrolyte

(lithium and potassium carbonate) is held in a tile (lithium aluminate). The

anode is made of porous nickel and the cathode of porous nickel oxide. The

reactions in a molten carbonate fuel cell are depicted in Fig. 7. The cathode

material is transformed into lithiated nickel oxide by its interaction with

the electrolyte. Lithiated nickel oxide is a better electronic con4uctor and

also more corrosion resistant than nickel oxide. However, corrosion of the

cathode material is one of the most serious problems in this fuel cell. Use

of Wi-Cr or Ni.-Cu alloys inhibits the sintering of the Ui and the presence of

Cr reduces the sulfur poisoning. Attempts are being made to find spinels and

perovskites as substitutes for the lithiated nickel oxide cathodes which tend

6978B



to corrode nlowly in the electrolyte. The bipolar plate conoints of two

layers - nickel on the hydrogen side and stainless steel on the oxygen side.

The morphology ~d porosity of each of the cell components is sho~ in Fig.

8. The pores in the electrolyte tile are relatively nmall compared to those

in the electrodes. Activation overpotential losses are minimum in thin cell.

The cell potential - current density relation is linear (Fig. 9) and the cells

are designed to operate at 0.7 V and 150 mA/cm2. The slope of this line is

about four times as high as in the linear region of the same plot for the

phosphoric acid fuel cell system. High grade heat, liberated by this system,

can be used for electr],city generation in a gas turbine or used as such by

chemical industry. Efficiencies for electricity generation as high as 60% are

projected.

Molten carbonate fuel cell programs are ongoing in the U,S.A., Japan, and

Italy. Fuel cell powerplants with an electrode area of 1/2 to 2 m2 are

plnnned to be built. The multicell stack will consist of 75 to 700 cells,

Each of these will deliver a power output of 5 KU to 1 ?fU. International Fuel

Cells Corporation is the fore~nner for th~ development of molten carbonate

fuel cells, There are two lines of approach. One is to build a stack with

electrode area of 0.1 m2 and the other with that of 0.8 m2, The arean of

active research ace (i) i~roved electrolyte management, (ii) stabilization Of

cathode, (iii) reduction of anode compaction and low cell to end plate

resistance; and (iv) examintition of effects of H2 S on molten carbonqte fuel

cell perfomnance,

Solid Electrolyte Fuel Cell Systems: The rationale for developing solid

electrolyte fuel cell systems is its high projected efficiency for electricity

&aneration, by using gasified coal, ag compared with those of the phosphoric

acid End molten carbonate fuel cells, A ochematic of a oolid electrolyte fuel

cell power plant is shawn in Fig. 10. In recent years, attention has been

focuaaed on natural gag, iasteud of coal ●B the primary fuel. Thio is beta’~se

of th) vast proven resemes of this fuel in many countries including the

U.S.A. It i- ●lso simplsr to process natural gas than coal to tha fuel call

fuels (H2 ●nd CO). The proceseod fuel enters the electrochemical cell stack

where tho chemical energy of the fuel is tran~formed to electricity. This

P,
o

~ql [’& Ji
rfq ;’\

11 6978E

T;\du



fuel cell with the yttria stabilized zirconia (12% yttria in zirconia)

electrolyte operates at 1000”C. The high quality waste heat enters the steam

generator which in turn s’ipplies steam to the steam turbine to produce more

electricity. The total efficiency for the production of electrical energy

from the chemical energy of coal or natural gas using tl~e fuel cell system and

the gas turbine iS projected to exceed 60%.

Rapid advances in this technology have been made at Westinghouse

Corpmation during the last S years. l%is is in respect to a tubular ciesign

(closed at one end) with connections of cells in series and parallel. Layers

of the cathode, electrolyte and anode material are deposited on the closed end

porous zirconia tube (see Fig. 11) by chemical or electrochemical vapor

deposition techniques. The cathode of one cell is connected to the anode of

the next cell by a strip of interconnection and nickel felt. TWO cells are

connected in parallel with a nickel felt. The series - parallel connection in

four cells is represented in Fig. 12. The nickel felt is in a reducing

environment and is stable. Nickel is the anode electrocatalyst and strontium

doped lanthanum ~nganite the cathode electrocatalyst. Magnesium doped

lanthanum chromi!,e is the material used for the interconnection. The

operating temperature is 1000”C. The cell potential-current density relation

is linear. The slope is relatively high - about 0.60 O?Un/CII!2. Thus, the

current density at about 0.65 V is about 150 mA/cm2. The improvement in

performance of oxygen over air, as seen in Fig, 13, is due to thermodynamic,

electrode kinetic and “barrier layers” effects. The H2/air fuel cell

performance can be improved by minimizing the “barrier layer” effect, i.e.,

reducing the thicknesses of the support tubes, and of the cathode layer as

well as increasing the porosity.

The power versus current or current density plot is also shown in Fig.

13. Each tube has a designed power output of 18 watts, The cell performance

is stable over several hundred hours, This led to the design, engineering and

construction of multicell stacks, A fuel cell bundle consists of twelve

tubular cells, arranged in a three-parallel by four series matrix. A series

connection af two bundles is made at one side and power terminals are

connected to each of the bundles on the other side. Thus, the twenty-four

cells exhibit ●ight times

n

voltage and three times the current of a single

cell. ?-~

PI \b1, ~’

[

rf’ “,tl 12 69788

;U@”



The generator which consists of the twenty-four cell stacks also has a

thermal insulation package. The latter includes the cell-support etructure

snd gas distribution systems. The temperature distribution within the cell

bundle is maintained only by waste heat generation from the fuel cells and

combustion of the spent gases in the exhaust chamber (15% of supplied fuel).

Heat exchange in the spent fuel combustor preheats the inlet air supply and

limits the exhaust gas temperature.

The successful design, construction, and performance evaluation of the

24-cell stack led to the construction of a 5 KU fuel cell generator. This

multicell stack consists of 324 single cells. The sub-cell stacks, consisting

of the 24 cells, are connected in series to form the multi-cell stack.

Performance evaluation of the 5 W fuel cell system is now in progress. Like

in the molten carbonate fuel cell system, hydrogen and carbon monoxide may be

ueed as processed fuels. ~ain, because of the high grade heat, it is

designed to senerate more electricity in a bottoming cycle or be used as such

for cogeneration applications. Efficiencies for electricity generation will

be higher than that for the molten carbonate system.

~lid Polymer Ulectroiyte Fuel Cell Systems: Thir system is being considered

for transportation application. Since the system operates at less than

100”C, methanol is the most appropriate liquid fuel to be reformed and used in

this fuel cell system, As stated in the preceding section, progress had been

made Lt Los Alamos National Laboratory to reduce the noble metal loading to

one tenth that in the state-of-the-art iystem developed by General Electric

Company/Hamilton Standards, This is done by incorporating a proton conductor

into the electrode structure to extend the three dimensional reactor tone,

The water management in the system can be ●chieved by optimum humidification

of tim system, Prospects are good for attaining high power densities - over

500 mW/cm2.

Pi-t Hethanol ?UO1 Cell: A fuel Cell Researcher’s Dream: A fuel cell

researcher’s dream in to oxidise organic fuels (preferably methanol which is

●asily produced from natural gee or coal) rather than process these to

hydrogen, which is in tu oxidiaaid to hydro~en in the electrochemical call.

~!!ifl “

J“ \
\\

T 13 69?8E
P!,’ ‘\\.

[

w’
,0 ‘b)



Methanol is the most active electroorganic fuel but its activity in three

orders of magnitude less than that of hydrogen. Thus, under these conditions

current-densities of about 50 mA/CK12●i-e obtained at cell potentials of 0.4

v. Even at these cur~-ent densities, there is a performance degradation with

time. The main cause for i~ is that the intermediates formed during methanol

oxidation poison the platinum electrocatalyst. The poisoning effects are less

with some alloy elgctrocatalysts (e.g, Pt-Ru, Pt-Sn) and platinum

electrocatalysts with ad atoms (Bi, Pb, Sn, Gel. The probable mechanisms for

inhibition by such electrocatalysts is that the second elements is either in

an oxidised state which catalyzes the oxidation of the organic intermediates

or an inhibitor for hydrogen adsorption on platinum, thereby preventing the

formation of a hydrogenated organic inhibitor.

Attempts have been made to oxidise methanol in a phosphoric acid fuel

cell at elevated temperatures (say 200°C). In this case, methanol was fed as

a vapor into the cell. The main problem was

from the anode to tho cathode where it acted

reduction reaction.

that unreacted methanol migrated

as a depolarizer for the oxygen

PROGNOSIS OF ECONOMICSND APPLICATIONS OF FUEL CELL SYST~S.—

Thta applications of fuel cell systems have been well demonstrated in

space vehicles, Capital costs are not such an Overriding factor for space

applications as for torresirial applications. The alkaline and posuibly the

solid polymer {~lectrolyte fuel cell systems will continuo to ba the ●uxiliary

power sources for space vehicles. Regenerative fu~l cells will have to be

developed for long space flights (more than a few months) and space

laboratories,

The present price of crude oil makes it more difficult. for the ●ntry of

fuel cell systems into the terrestrial ● rena particularly in the dweloped

countries, Fuel cells may find nearer term replications in tho developing

countries, particularly for dispersed power generation uning biomess fuels

(biogas, alcohols). Since the manufacturing processe~ for fuel cells ● re

labor intensive, the capital costs of these systems could be coi~siderably

lower in the developing countries, Further, the cost of fuels in these

countrios ● re realistically and relatively (~n respect to per capita income)

higher in the developing countries, Thus, the need for fuel cell technology

f

697flu

. ,.



is higher in the developing countries. In the developed countries, fuel cell

technologies can make an impact on-site integrated energy and cogeneration

applications, Environmental constraints can also accelerate the entry of fuel

cell technologies.

The bottom line with respect to economics of fuel cells for terrestrial

applications, will be the cost of electricity generated, which in turn will

depend on capital costs, efficiency and lifetime.Taking into consideration the

performances of the four types of fuel cells for terrestrial applications, the

major factors to be overcome are: (i) develop automated techniques to reduce

the capital costs of the system to about $1000-$2000/kU (high capital cost at

the I@ level and lower at the HW level) for stationary applications and

$200/kW for transportation applications; (ii) achieve reliability of

performance ~Ter the required lifetimes; (iii) demonstrate economic, technical

and environmental advantages of dispersed, on-site integrated energy and

cogeneration power plants; and (iv) find niche for electric vehicles - in

plant, military, buses, trucks, and automobiles?

The factors which impede tha commercialization of fuel C0118 are:

● Capital cost, operating cost, lifetime, reliability, weight, and

volume.

● Capital costs and powerlweight and powerlvolume which &re primarily

dependent on power density.

● Operating costs which are detemined by amortization of capital costs

(hence lifetime) and efficiency,

● Material costs of fuel cell components which are relatively low

except for the solid polymer electrolyte fuel cell system,

● Capital coste which are relativ~ly high becausa manufacturing coots

are labor intwwive,

,, “
15 6978B



● For fuel cell systems to have an efficiency greater than 50%, single

cell potential must be greater than 0.75 V (i.e., half the

thermoneutral potential for the reaction Ha + 1/2 0. + HOO
4

which iS 1.48 V.

● Open circuit potential for all fuel cells is

attain high current densities with this open

approximately 1.0 V. To

circuit pot~ntlal and

cell potentiel of 0.75 V, the slope (dE/di) in the linear region of

the cell potential versus current density plot has to be small, say

about 0.1 ohm cm2. The lowest values of (dE/di) reported for the

five types of fuel cell systems are found in Table I. It can be

readily seen from this table that the only fuel cell systems which

can attain high current densities arc the ones with alkaline and

solid polymer electrolyte. The phosphoric acid fuel cell system is

considerably closer to attaining higher current densities than the

molten carbonate or sulid electrolyte fuel cell systems bectiuse of

tha respective (dEldi) values.

This work was carried out under the auspices of the U. S Department of

Energy. The author wises to thank Dr. Jack T, Brown of Westinghouse

corporation for lendi,~g the slide material on solid oxide elclctrolyte fuel

cells, which wore quite valuable for the preparation and presentation of this

paper,

REFERENCES

1, J.O’M. Bockris, and S, Srinivauan, Fuel Calls: Their Electrochemistry,

McGraw Hill Publishing Company,

2* K. V. Kordosch, J, Igloctrochem,

New York (1969).

80C., 125, ??C (1978).

3. S, S, Pomer, lM,, Assessment of Research Nods for Advanced Fuel Calls,

69781!

r
\



4.

5.

6,

7.

8.

9,

10,

11.

*r

,\

‘“\\’#*

H. T. R. Haget, Chapter 11 in C. Berger, Ed. , Handbook of Fuel Cell

Technology, Prentice Hall, Inc., Englewood Cliffs, NJ (1968).

C. C. 140rrill, Proc. Ann. Power Sources Conf., 19, 38 (1965).

1. D. Raistrick, proceedings of the Symposium on Diaphragms, Separators

and Ion Exchange ?fembranes, 86-13, 172 Electrochemical Society, Inc.,

Princeton, WJ (1986).

S. Srinivasan, E. A. Ticianelli, C. R. Derouin, and A. Redondo, J. Power

Sources. In Prass.

E. A. Ticianelli, W. Paik, C. R. Derouin, I. D. Raistrick, A. Redondo,

and S. Srinivasan, Extol, Abs,, 171st Meeting of the Electrochemical

Society, Honolulu, Hawaii, October 18-23, 1987. In Press.

P. Stonehart, B. D. McNicol, E. J. Taylor, mid S. Srinivasan, Chapter 8

in B. D. tlcblicol and D, A. J. Rand, Eds., Power Sources for Electric

Vehicles, Elsvier, Amsterdam (1984).

Fuel Cells: Technology Status Report, !!organtown Energy Technology

Center, U. S, Department of Energy, Report #DOE/HETC-86-0241 (DE

86006619), Plorgantown, West Virginia (1985).

6978if



TA8LE I

Lowest Values of S1OPCS of Linear Reftions of Cell Potential-
Current Density Plots@E/di) Reported for the Five Types

~f Puel Cells

Type of Fuel Cells ~dE/di), ohm cm~

Alkaline 0-05
Solid Polymer o-1o
Phosphoric Acid 0-20
Molten Carbonate 0-80
Solid Electrolyte O-65

18 6978E



l-.
-y .

.

m

nEwcEDwEmH7

0.s U#

“EEr



I Porousanduchve gra~ite l~er !I

I
ANOOE pofous P~suppxted WI

cafbon m!h a Teflon binder I
I

ELICTROIYT[ W(I. m an inarl

sititxtn-based mattix I
!

I POrOUSCOIW!UCt*graphitelayer I



FUEL CELL ELECTRODE FABRICATION
SPRAYING METHOD

r 4 i -—
Pt ON C CATALYST

TEFLON 30
FOTO FLOW

WATER
I

MIXING

r-+HOMOGENEOUSCATALYST
MIXTURE I

DRYING

PRE$kMJG

SINT;RING
)
I BACKING I

I DIFFUSIONLAYER I
1 I I

I
4

SPRAYING I

PRESSING

SINT;RING

.=





..-”.

E&
‘..

c.- -
t&l~,-..- [!Jt!!Jtlr!!JA W-—--”” T II—— ... .9.
~;~ g -“ -“1 ~ =

-...”..
--G-

I ----—



I #i side of bilayer cell separator plate I
t

Fine-pore Ni
ANODE: Porous Ni

bubble-barrier

I ELECTROLYTE: Li/KC03 supported onineti liA!02 I
CATHODE: Porous NiO PUI keted stainless steel

current Cokxtor

Stainless steel side of bilqer cell separetor pltie
4

MOLTEN

A 31527

CARBONATE FUEL CELL CONSTRUCTION MATERIALS



““ r7TsFw%02 + C02+ h-~cof
OXIDANT
AIR + C02 CATMODE

EXHAUST

OVERALL nwnoa——



.

,, .\. ..:::..:::..:....=

\



.

●✎

“t o

I 1 1 I 1 1
w la 1$9 m Z*O



s HOT

FUEL~
CO:::::IONi

STEAM
SOFC 1 w GENERATOR

AIR ~ !

D-C.
STEAM

TURBINE

!NVERTER

A.C.
GENERATOR

\ 1

\

-

/
COMBINED A.C. OUTPUT



INTERCONNECTION

AIR

ELECTROLYTE



●

\ir Fk-rtti I-In: \fujird bnttunum Mangantic Slurw Coal/Sinlrr
(C-*)

%kl Eklrc+w o l-a-)- YIIru-5&lizd Zirconia EVD”

Inlmrmnrcfwl O(B?4 Mdird LaIwIunuwI Chrunilc E\’D”

Fur! FM d nfh-)1 NKkAz+CaliaCk-wKlw SJurw Ccd/Stnlrr
( An(&) CA’dl-zlnxnw Gllwl

CELL-WELL CONNECllON IN THE M)LID OXIDI! FUEL CELL



-———- ----- —. —.. - -—-

.

.

‘1*, ,<