Tải bản đầy đủ - 0 (trang)
Ch 5. Batteries for Vehicular Propulsion

Ch 5. Batteries for Vehicular Propulsion

Tải bản đầy đủ - 0trang


Halina S. Wroblowa

for the last 70 years. The energy and maintenance costs of electric

vans were shown to be 45-60% of their gasoline and diesel-powered

analogues.3'4 Small fleets of electric buses are in operation in Australia, England, France, Germany, and Japan.4'5 For specialized

purposes (e.g., indoor, underground, and airport transport of people

and cargo; golf carts; forklifts) lead/acid traction batteries are

widely used. Only in the USSR they have not completely displaced

nickel-iron batteries.

Electric passenger cars cannot presently compete with ICEVs,

neither economically nor in terms of user's convenience. Therefore,

only a very limited number of two- and four-seater lead/acidpowered passenger EVs have as yet been constructed.2'4

In the 1960s the interest in electric cars—as a remedy against

urban pollution—was revived and strengthened in the 1970s by the

realization that foreign oil sources can be abruptly shut off, oil

supplies will be exhausted in the not-too-distant future, and indeed

all fossil-fuel supplies are finite. Intensified research in the area of

battery systems, which (at least in the USA) seems to have peaked

in terms of funding around 1980, resulted in considerable improvements of previously available systems and in worldwide efforts to

develop new batteries better suited for EV applications than leadacid. The latter still remains the only battery commercially used

for specialized traction purposes, while other systems are now at

various stages of development. The decrease in the EV battery R&D

effort can be related, on the one hand, to the temporary "oil glut"

and, on the other, to the inherent range limitation of cars powered

by secondary batteries and severe technical difficulties connected

with the necessity of simultaneous optimization of several battery

attributes which are desirable for EV applications. The required

battery attributes are closely linked together and improvement of

one feature usually has an adverse effect on other properties.

2. Battery Classification

Among rechargeable batteries commercially available today (Table

1), only the lead/acid (Pb-A) and Ni-Fe systems can be considered

as potential EV power plants because of the cost and scarcity of

materials used in other batteries. The performance of the Pb-A and

Ni-Fe systems is, however, inherently limited by their low-energy

Batteries for Vehicular Propulsion


Table 1

Commercially Available Rechargeable Batteries




Lead-Acid (SOA)


Nickel-Iron (SOA)








Automotive SLI; traction

(specialized vehicles);

emergency power

Traction and lighting of trains

(predominantly in USSR)

Small-size portable power

Aerospace applications requiring

nonmagnetic components

Military; aerospace

storage capability. Attempts to provide EVs with a range (between

recharges) exceeding that of the SOA (state of art) Pb-A traction

batteries has led to considerable improvements in the technology

of Pb-A ISOA ("Improved" SOA) battery prototypes, "advanced"

Pb-A cells, and Ni-Fe batteries, as well as to the emergence of

several new systems expected to have a more acceptable range on

the basis of their higher theoretical specific energies. With the

exception of a few primary batteries (metal fuel cells), the candidate

systems are electrically rechargeable. They can be classified in a

number of ways, e.g., according to the operating temperature

(ambient or high), type of electrolyte (aqueous, organic liquid,

molten salt, solid; flowing or stationary), or their relative stage of

development. The latter classification, somewhat subjective and

labile, divides the candidate systems into three groups: near term,

advanced, and exploratory. Very roughly, the development of nearterm systems has reached the prototype EV battery stage; of

advanced systems, the battery prototype or module stage; and of

exploratory systems, the small-module, or cell stage. The list and

basic characteristics of near-term and advanced batteries is given

in Table 2, along with data pertaining to some exploratory systems.


Battery characteristics can be described in terms of the desired EV's

performance-, cost-, and safety-related factors, as listed in Table 3.


term and advanced

Pb/H 2 SO 4 (aq)/PbO 2



Zn/ZnCl 2 (aq)/Cl 2 (H 2 O) 6

Zn/ZnBr 2 (aq)/Br 2 complex


Na melt//3"-alumina/S melt


Calcium-Iron disulfide

Ca2Si/chloride melt/FeS2

Lithium-Titanium Disulfide

Li/Li salt in organic solvent/TiS2


Fe/KOH(aq)/O 2


Zn/KOH(aq)/O 2


Al/KOH(aq)/O 2

Conducting polymer (CP)

CP/salt in organic solvent/CP or Li/Li +






Lithium-Iron sulfide



Table 2

Battery Classification













1.92 (aver.)





















480 (aver.)












637 (aver.)




(Whkg- 1 )


Batteries for Vehicular Propulsion


Table 3

Correspondence between the Vehicle and EV Battery Characteristics

EV parameters

1. Performance

Range between recharges

Daily range

Acceleration capability

Climbing speed


Battery parameters

Specific energy (Wh kg l)

Energy density (Wh liter"1)

Self-discharge rate (percent per day)

Charging time (h/C)

Specific peak power (Wkg"1)

Peak-power density (W liter"1)

Specific sustained power (Wkg"1)


Insensitivity to ambient conditions, overcharge,

overdischarge, shock, vibration, etc.

Lack of complexity

2. Cost


Fundamental resource limitations


Geopolitical distribution

Manufacturing, R&D

Cycle life

Overall energy efficiency


3. Safety and environmental


Stationary failure modes

Impact failure modes

Fire and/or explosion hazard

Reactive and/or toxic material release

Environmental impact


Mining and processing effects

1. Performance-Related Requirements

(i) Range between Recharges

EV range depends primarily on the amount of effective energy

stored per unit weight or volume of the battery. The specific energy

(Wh kg"1) of secondary batteries delivered to EV wheels is approximately 2-5% of that of gasoline. Volumetrically, the situation is

not much better, the energy density being within —3-7% of the


Halina S. Wroblowa

gasoline value. This explains the inherent gap in the vehicle range

between "refueling" of battery- and ICE-powered cars, a gap which

cannot be bridged by simply increasing the battery weight. The

battery fraction (battery weight/vehicle curb weight) should not

exceed ~ | to ensure mechanical stability and to avoid the excessive

waste of energy spent on propelling the battery itself.

(a) Specific energy

The energy available for driving purposes is often somewhat

lower than the energy deliverable by the battery. The losses may

be due, e.g., to the operation of battery auxiliaries (pumps in flow

batteries, shunt-current protection, etc.) The translation of the available specific energy into the EV's range must take into account the

vehicle characteristics including its weight, wind and rolling resistance, and the driving profile6 determined by the vehicle's end use

as shown in Fig. 1, based on the average performance of Pb-Apowered vehicles driven in Germany.7 The trend shown can be

rationalized in terms of slower speeds and acceleration (energy

losses are proportional to the square of velocity) of larger German

vehicles driven with various driving profiles, depending on their

mission. Also the drag due to the frontal wind resistance does not

decrease linearly with decreasing vehicle volume. The ordinate

values can vary considerably from those shown in Fig. 1. Thus,


250 i







£ 170

i i










n I




Figure 1. Specific energy of various types of EVs.

Batteries for Vehicular Propulsion


e.g., a 1500-kg compact-type EV with ~ | battery fraction, driven

under urban conditions, as simulated by the U.S. Federal Urban

Driving Schedule,6 requires only —250 Wh mile"1, or ~ 170 Wh t o n 1

mile"1.8 The trend, however, correctly indicates that the most

demanding is a small passenger car. The packaging constraints are

compounded by size effects of the battery itself. Scaling down

battery capacity results in a loss of effective specific energy, which

is relatively small for simple, ambient-temperature systems and

becomes progressively larger with the degree of complexity and

number of auxiliary subsystems in flow- and high-temperature

batteries. As shown in Fig. 2, the specific energy of complex batteries

is severely impaired in the region of 20-30 kWh suitable for small

urban cars. The latter require ~60Whkg~ 1 to attain a 100-mile

range.8 This might indicate that, in spite of some more optimistic

projections, only simpler systems may prove compatible with compact EVs, while more complex batteries would be more suitable

for large family cars, vans, etc.

The effective specific energy of various batteries amounts

presently to some 10-25% of theoretical specific energy (defined

as the reversible work of the cell reaction per unit weight of reactive

electrode materials). The extent of departure from the theoretical

value depends on: (1) the weight of nonactive battery components;


Figure 2. System

specific energy.9











Halina S. Wroblowa

(2) reactant utilization; (3) discharge overvoltage (determined

primarily by ohmic losses); (4) coulombic losses due to parasitic

reactions, self-discharge, and/or shunt currents in series-connected

cells with a common electrolyte; and (4) losses connected with

operation of auxilaries.

Efforts to improve the ratio of effective to theoretical specific

energy involve, in general: (1) replacement, where possible, of inert

grid, container, connector, and current collector materials by lightweight substances; (2) increase of utilization of active materials by

improved cell design and/or use of special additives; (3) careful

modeling and optimization of current collection; and (4) use of

bipolar electrodes.

Although the individual mechanism may differ, the specific

energy of batteries decreases, in general, with increasing rate of

discharge. For example, the specific energy of a certain SOA Pb-A

traction battery varies between 34 and 22 wHkg"1 when discharged

within 5 and 1 h, respectively.1 The capacity C{ available at the

given discharge current / is often represented for porous-plate

simple batteries by the empirical Peukert equation:

Cj =


where K and n are coefficients characteristic for the given battery.

Any comparisons of battery specific energy or ranges have sense

only if the latter are measured for the same value of C/h, where

C denotes the rated capacity of the battery and h—the number of

hours of discharge.

The EV range is affected by the self-discharge which may occur

both during rest and operating periods of the battery. The selfdischarge rate depends on the individual battery characteristics,

ranging from zero for batteries with solid electrolytes (e.g., Na-S)

to considerable values for batteries operated without separators

(e.g., zinc-chlorine hydrate). In batteries with separators, selfdischarge is determined by the rate of ion transport through the

membrane. The most efficient—ion-exchange membranes—are

usually too expensive for practical purposes and microporous

separators are commonly used.

The EV's range can be somewhat extended by the battery's

capability of accepting charge pulses during regenerative braking.

Batteries for Vehicular Propulsion


Depending on the battery and driving profile, such recovery may

increase the range by some 15-25%.10

(b) The volumetric energy density

This defines the packaging requirements for the given EV range.

For simple, ambient-temperature systems, its numerical value is

about double the battery's specific energy. This ratio decreases with

increasing volume of auxiliaries, and, for smaller cars, volumetric

considerations may become range limiting for rather than the

gravimetric effective energy storage capability. This limitation may

pertain particularly to bulky batteries which cannot be spatially

distributed in the car being designed as a single package (e.g.,

high-temperature systems enveloped in a vacuum insulation).

Optimal packaging of batteries will require vehicles specially

designed to accommodate the given battery.

(11) Daily Range

In general, fast recharge decreases the energy efficiency (unless

the self-discharge rate during charge is considerable, as in the case

of the Zn-Br2 battery) and may cause severe cell damage owing to

temperature increases which accompany high charging currents.

Therefore, recharging to full capacity requires 6-10 h for most

systems and the range between recharges usually defines the daily

mileage allowable. Some systems are claimed by their developers

to be capable of a 0.5-h recharge to some fraction (e.g., 50%) of

the rated capacity. If indeed, the high charging rates would not

affect the battery reliability and cycle life, the daily mileage of EVs

could be extended to more convenient distances. The demands on

the infrastructure of electric utilities would be in this case more

severe (although probably not prohibitive) than those anticipated

for overnight recharge.

(MI) Acceleration

Specific peak power (W kg"1) of the battery determines the EV

acceleration capability. Not only the energy storage but also the

power/weight ratio comparisons with ICEs are not favorable for


Halina S. Wroblowa

electrochemical power plants. The obvious reason is the

tridimensional character of hot combustion which occurs in the

reactor's volume, as opposed to the two-dimensional characteristics

of the interfacial electrochemical energy conversion. In spite of the

poor power/weight ratio, acceptable acceleration of EVs is more

easily attainable than the desirable range.

The specific power passes through a maximum at a voltage

corresponding to —50% of the open-circuit value (OCV). The

usable peak power, however, limited in practice to —85% of the

maximum value, is attainable at voltages some 30% lower than

OCV. (Higher-voltage departures would severely affect the cost of

the present dc motors or ac controllers.) Available data usually

refer to the maximum power. Also, published data often refer to

peak power averaged over a 15-s pulse, while actually the value of

peak power during the next 5 s is as important for EV acceleration.

In general, the specific peak power decreases with the depth

of discharge (DOD) (cf. Fig. 3). Therefore, EV battery specification

with respect to range and allowable depth of discharge limit should

pertain to (whichever comes first) DOD values at which EV becomes

incapable of the demanded acceleration, or to which the system


Power kW








Figure 3. Specific peak-power, averaged over the last 10 s of a 20-s

pulse, as a function of the depth of discharge for Lucas-Chloride,

-535 kg, 192-V ISDA tubular Pb-A batteries. EV power-train constraints imposed. (Courtesy of Lucas Chloride EV Systems, 1984.)

Batteries for Vehicular Propulsion


can be safely discharged without impairing its reliability and lifetime (often —80% of the rated capacity is required as a limit).

The EV power requirements depend on the car's end use and

assumed acceleration demands. For example, the specifications for

safe driving in a mixed ICEV and EV urban traffic of a small

compact-type EV have been defined as acceleration from 0 to

50 mph in <20 s.8 The specific peak-power corresponding to these

demands has been calculated as 80-90 wkg"1 for conditions of

optimum use of the AC power train (i.e., for constant-power acceleration following the initial period in which the peak-power is attained

at the maximum tractive effort8). The majority of near-term and

advanced batteries could, at least potentially, meet these demands.

The difference between projected and actual performance of

the present prototypes is often very high. Thus, e.g., the potentially

"best" performance of the LiAl/FeS battery (Argonne National

Laboratory) is compromised at present, primarily by the inadequate

current collection which causes substantial power losses due to the

high internal resistance.

(it;) Specific Energy-Peak power Tradeoffs

The specific energy to peak power ratio is a key battery-design

parameter. Since both cannot be simultaneously increased, the given

system can be designed for maximum range at some loss of peakpower capability, or for high-power operation at some loss of the

achievable range. The sensitivity of the given system to the power

demands is shown in Ragone plots exemplified9 in Fig. 4 for

projected performance of a number of battery systems. (It should

be kept in mind that the data in Fig. 4 differ considerably from

those presently attainable and are of comparative rather than

absolute value). Ragone plots map the specific energy available

from the battery for values of the specific power at which the battery

is discharged. The energy/power relations determined in laboratory

tests have been successfully used to project EV's performance. An

agreement of ±5% has been reported for a number of near-term

batteries10 between the projections based on Ragone plots obtained

for small (6-12 V) modules at the Argonne National Laboratory

and vehicle ranges obtainable under conditions of urban driving

cycle as tested at the Jet Propulsion Laboratory.


Halina S. Wroblowa





I 50







Figure 4. Ragone plots. Characteristics expected for 1990s. Specific power data

for 80% DOD.9

It would seem that battery-battery hybrids might provide better

simultaneous optimization of range and acceleration. However, the

overall specific-energy and efficiency losses, cost, and complexity

of the hybrid systems seem to militate, at present, against this


(v) Reliability

The probability that a battery will function properly under

various kinds of driving and environmental conditions determines

its reliability. The latter is an extremely important parameter which

may adversely affect public acceptance of electric cars to an even

greater extent than their limited range. At present, little is known

about the reliability of full systems other than a number of lead/acid

traction batteries which are highly reliable. (Thus, e.g., batteries

for delivery vans produced by Lucas Chloride EV Systems are

presently sold in England with a two-year unconditional and twoyear conditional warranty.) Data for other batteries are largely

absent since very few have been sufficiently tested even at the

module or cell level and hardly any information exists concerning

effects of actual driving conditions.

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Ch 5. Batteries for Vehicular Propulsion

Tải bản đầy đủ ngay(0 tr)