Astm stp 705 1980

ENGINE COOLANT TESTING: STATE OF THE ART A symposium sponsored by ASTM Committee D-15 on Engine Coolants AMERICAN SOCIETY FOR TESTING AND MATERIALS Atlanta, Ga., 9-11 April 1979 ASTM

ENGINE COOLANT

TESTING: STATE OF

THE ART

A symposium sponsored by ASTM Committee D-15 on Engine Coolants AMERICAN SOCIETY FOR TESTING AND MATERIALS Atlanta, Ga., 9-11 April 1979

ASTM SPECIAL TECHNICAL PUBLICATION 705

W H Ailor Reynolds Metals Company editor

List price $32.50 04-705000-12

•

AMERICAN SOCIETY FOR TESTING AND MATERIALS

1916 Race Street, Philadelphia, Pa 19103

NOTE The Society is not responsible, as a body, for the statements and opinions advanced in this publication

Printed in Baltimore, Md

May 1980

This publication on Engine Coolant Testing: State of the Art contains

papers presented at a symposium held 9-11 April 1979 at Atlanta, Georgia

The symposium was sponsored by the American Society for Testing and

Materials through its Committee D-15 on Engine Coolants W H Ailor,

Reynolds Metals Company, served as symposium chairman and editor of

this publication

Single Cylinder Engine Tests for Evaluating the Performance of Crankcase

Lubricants, Part I: Caterpillar IG2 Test Method, STP 509A (Part I),

1979, bound, $9.75, 04-509010-12; looseleaf, $12.75, 04-509011-12

Single Cylinder Engine Tests for Evaluating the Performance of Crankcase

Lubricants, Part II; Caterpillar IH2 Test Method, STP 509A (Part

II), 1979, bound, $9.75,04-509020-12; looseleaf, $12.75,04-509021-12

Single Cylinder Engine Tests for Evaluating the Performance of Crankcase

Lubricants, Part III: Caterpillar ID2 Test Method, STP 509A

(Part III), 1979, bound, $9.75, 04-509030-12; looseleaf, $12.75,

04-509031-12

LP-Gas Engine Fuels, STP 525 (1973), $4.75, 04-525000-12

Low-Temperature Pumpability Characteristics of Engine Oils in Full-Scale

Engines, DS 57 (1975), $16.00, 05-057000-12

to Reviewers

This publication is made possible by the authors and, also, the unheralded

efforts of the reviewers This body of technical experts whose dedication,

sacrifice of time and effort, and collective wisdom in reviewing the papers

must be acknowledged The quality level of ASTM publications is a direct

function of their respected opinions On behalf of ASTM we acknowledge

with appreciation their contribution

ASTM Committee on Publications

Jane B Wheeler, Managing Editor Helen M Hoersch, Associate Editor Ellen J McGlinchey, Senior Assistant Editor Helen Mahy, Assistant Editor

Introduction 1

Automotive Engine Coolants: A Review of Tlieir Requirements and

Methods of Evaluation—L C ROWE 3

Experience of the British Standards Institution in the Field of Engine

Laboratory Research in the Development and Testing of Inhibited

Coolants in Boiling Heat-Transfer Conditions—A. D MERCER 53

Discussion 78

Simulated Service Tests for Evaluation of Engine Coolants—

ROBERT SCHULMEISTER AND HELMUT SPECKHARDT 81

Discussion 100

Research and Development Efforts in Military Antifreeze

Formula-tions—J H CONLEY AND R G JAMISON 102

Discussion 108

Corrosion Testing of Furnace and Vacuum Brazed Aluminum

Radiators—KAZUHIDE NARUKI AND YOSHIHARU HASEGAWA 109

Discussion 131

Use of Electrochemical Techniques for Corrosion Testing of

Anti-freezes—E F. O'BRIEN, S T HIROZAWA, AND J C WILSON 133

Discussion 145

Chemical Properties as a Tool for Maintaining High-Quality Engine

Antifreeze Coolants in the Marketplace—T. P YATES AND

MARYLOU SIANO 146

Discussion 154

How Good is the ASTM Simulated Service Corrosion Testing of

Engine Coolants?—j v CHOINSKI AND J F MAXWELL 156

Discussion 165

Discussion 187

Static Vehicle Corrosion Test Method and Its Significance in

Engine Coolant Evaluations for Aluminum Heat Exchangers—

Statistical Treatment of Laboratory Data for ASTM D 1384-70

Discussion 231

Refinement of the Vibratory Cavitation Erosion Test for the

Screening of Diesel Cooling System Corrosion Inhibitors—

R D HUDGENS, D P CARVER, R D HERCAMP, AND

J LAUTERBACK 2 3 3

Discussion 266

Electrochemical Corrosion of an Aluminum Alloy in Cavitating

Discussion 281

Cavitation Corrosion—B D OAKES 284

Discussion 292

Evaluation of a Novel Engine Coolant Based on Ethanediol

Developed to Replace AL-3 (NATO S735) as the Automotive

Antifreeze Used by the British Army—E. W BEALE,

BRIAN BEDFORD, AND M J SIMS 295

Discussion 307

Cooling System Corrosion in Relation to Design and Materials—

E BEYNON, N R COOPER, AND H J HANNIGAN 3 1 0

Introduction

A critical component for any internal combustion engine is its coolant

system The combination of dissimilar metal components, including cast

iron, brass, zinc, aluminum, solders, etc., operating in a liquid system at

increasingly higher temperatures creates potentially severe corrosion and

heat transfer problems

The use of alcohol as an antifreeze for engine coolants has given way to

inhibited ethylene glycol solutions in available local supply waters for

year-round operation The diversity of inhibitors available for corrosion and

erosion protection has further complicated the coolant picture Higher

flow rates have introduced cavitation and erosion problems as new

con-cerns

During 9-11 April 1979, ASTM Committee D15 on Engine Coolants

sponsored an International Symposium on the State of the Art in Engine

Coolant Testing The sessions were held at the Sheraton-Biltmore Hotel in

Atlanta, Ga The 21 papers presented included both invited papers and

offered papers from knowledgeable persons in the automotive and coolant

manufacturing fields Authors came from England, West Germany, Japan,

Switzerland and, of course, the United States

The symposium was designed to present the current thinking of those

involved with engine coolant testing and to indicate areas for work to meet

new problems The sessions were of special value to newcomers in the field

and served as educational lectures At the same time, the continuing

efforts towards standardization of test methods were reported by members

of ASTM Committee D15 on Engine Coolants, based on more than 30

years of committee efforts

The papers and discussion resulting from this symposium make up this

Special Technical Publication The book should be very useful to engineers,

chemists and others concerned with engine and solar heat exchangers and

designers, stylists and others whose work involves heat transfer equipment

All Committee D15's test methods may be found in the current Annual

Book of ASTM Standards {Part 30) In the 1978 edition there were 21

methods

ASTM STP 120B on Selection and Use of Engine Coolants and Cooling

System Chemicals (1974) is an updated revision of earlier helpful

dis-cussions on engine cooling systems, antifreeze-coolants, installation and

service, and cooling system chemicals

The contributions and efforts of all the members of ASTM Committee

D15 on Engine Coolants is appreciated Many members served in planning

the program, chairing the sessions, reviewing the papers, supervising the

social functions Special thanks are due our overseas authors and session

chairmen who not only prepared excellent oral and written presentations

but helpfully have written answers for many questions raised at the sessions

The support and interest of their organizations and for all participating

companies is gratefully acknowledged

Members of DlS's Organizing and Planning Committee included:

Norman R Cooper, Union Carbide Corp.; Donald L Cramer, Houston

Chemical Co.; Joseph C Gould, E I duPont de Nemours; Vincent R

Graytok, Gulf Research and Development Co.; Donald L Wood, Shell

Development Co.; and Charles W MacKenzie, Radiator Reporter

W H Ailor

Metallurgical Research Div., Reynolds Metals Co., Richmond, Va 23261; editor

Automotive Engine Coolants:

A Review of Their Requirements

and Methods of Evaluation

REFERENCE: Rowe, L C , "Automotive Engine Coolants: A Review of Their

Requirements and Methods of Evaluation," Engine Coolant Testing: State of the Art,

ASTMSTP 705, W H Ailor, Ed., American Society for Testing and Materials, 1980,

p p 3 - 2 3

ABSTRACT: A brief review of early automobiles shows the development of the engine

cooling system, and some of the associated problems with these early cars are discussed

A liquid is commonly used to transfer heat from an operating automobile engine to

a radiator where the heat can be dissipated to the air In order that the liquid perform

effectively, it must have the appropriate chemical and physical properties Of foremost

consideration is the capability of the fluid to transfer heat over a wide range of

operating conditions In addhion, the fluid must be stable, must not freeze when not in

use or boil during or after engine operation, and must not cause or allow excessive

corrosion of the parts it contacts

To determine how well the cooling system is capable of performing its function, it is

necessary to perform a variety of tests to evaluate the operational characteristics of

component parts, the properties of the coolant fluid and its long-range stability, and

the capability of the fluid to minimize corrosion of all materials Tests range from the

shorttime laboratory test to the longer and more comprehensive field test Operating

conditions are often difficult to simulate in the laboratory, and the test tends to be

restrictive Field tests are usually more definitive but can be difficult to control

However, the end result of an effective development program over a number of years

has been a cooling system that has provided good durable service

KEY WORDS: engine coolants, engine cooling system, coolant properties, coolant

testing, antifreeze, heat transfer, corrosion

A brief review of our early automobiles provides some insight into the

reasons for the need and the development of an effective engine cooling

system There has been continual improvement over the past 75 years in

the design of the cooling system and in the quality of the antifreeze

cool-ants However, many of the same problems that were found with the earlier

'Departmental research scientist, Physical Chemistry Dept., General Motors Research

Laboratories, Warren, Mich 48090

cooling systems still exist today because the motorist is not sufficiently

informed or concerned to provide the required maintenance or to use

proper engine coolants

The history of the self-propelled land vehicle is only a little over 200

years, but much has happened during that short period of time The

earliest vehicles could hardly be classed as automobiles, as we know them

today, because many of them were merely some form of "land carriage"

with a means of propulsion added These early vehicles were primarily

steam operated, and one of the first patents for such a vehicle was granted

to Oliver Evans in 1787 in Maryland for a steam wagon that was only able

to operate for a short distance at a very slow speed before breaking down

[1].^ A commercial practical gasoline engine was produced by Etienne

Lenoir in France in 1860 [1], Carl Benz from Germany is given the credit

in 1885 for the first road vehicle propelled by an internal combustion

engine [2] It was not until 1892 that the Duryea brothers produced the

first gasoline engine in the United States [1.3] Following this introduction,

the interest in vehicular transportation grew at an increasing rate

The internal combustion engine required some means for removing the

excess heat from the engine because all of it was not transformed into

mechanical energy If the heat was not removed, the engine overheated and

soon malfunctioned Both air and water were used to cool engines in the

early part of the 20th century Air cooling required that the outer surface

of the combustion chamber be in direct contact with the surrounding air to

remove the heat The amount of heat that could be removed was limited by

the metal surface area that came in contact with the air, so it was not

unusual to add a few metal ribs to the surface to increase the contact area

[4] Air cooling was more feasible in these early days than in later years

because the early low-horsepower engines were small and produced little

excess heat Air cooling was often preferred by the manufacturer of the

small, light car because it added less weight than a water-cooling system

and was a simple design In addition, it was not affected by freezing

temperatures

In water-cooled engines, water passed through a jacket surrounding the

combustion chamber, absorbed the heat, and transmitted it to a radiator

where it could be dissipated to the air One of the distinct advantages to

the water-cooled system is that the surface area of the radiator can be

made many times greater than that of the engine block, permitting a more

rapid transfer of heat to the air Two systems were used to circulate the

water; namely, natural circulation and forced circulation with a pump In

natural circulation, a water tank was placed above the engine The water

passed by gravity from the bottom of the tank through a radiator and into

the bottom of the engine water jacket where it was heated [5] The hot

2 The italic numbers in brackets refer to the list of references appended to this paper

water rose to the top of the jacket and then back into the tank again This

system was fraught with difficulties and had many limitations Other

means were needed to improve cooling and to facilitate circulation of the

water, and the best and most frequently used method was forced

cir-culation with a pump, usually a rotary, centrifugal pump

There were early concerns about the location of the engine, whether it

should be placed in the front, rear, or under the middle of the vehicle

The cooling system was destined to play a significant role in this selection,

as indicated by one author's comments after a reliability race in 1902 from

New York to Boston The author stated that "troubles with the cooling

system are shown to have exceeded those due to faulty ignition and fuel

feed as causes for delay" [6] The author went on to speculate that some of

the troubles could be traced to the long piping necessary to take the cooling

water from the engine in the rear of the car to the radiator at the front It

was suggested that a compact cooling system with the radiator and engine

both at the front reduced the opportunity for leaks and the formation of

deposits to clog the system As the cars grew heavier, the distribution of

weight became of concern, and there was less objection to a better

dis-tribution of weight by placing the engine over the front axle The

im-portance of a reliable, more efficient cooling system continued to take on

greater significance, and the trend to water-cooled systems increased

The automobile was such a completely new experience to people in these

early days that the owners could hardly be expected to be concerned with

the cooling system when they tended to neglect other basic procedures that

were necessary for dependable vehicle operation They were chided because

they forgot to recharge the acetylene generator that supplied gas lamps or

neglected to tighten the brake bands [7] Even running out of gas was

attributed to carelessness because the owner forgot to remove the filling

cap "to sound with a lead pencil, bit of string, wire, or clean stick to

determine the quantity remaining in the tank" [7] The cooling system

received little owner concern because loss of coolant would be noted by

boiling liquid before the cylinders became overheated However, the same

writer suggested drawing off some old water and filling the system with

fresh water before a trip to avoid the necessity of "bothering some roadside

resident for water and the loan of a bucket" [7]

This simplistic approach was not endured for long The automobile had

provided the people with a new degree of freedom They could now travel

longer routes and explore lesser traveled areas, and they demanded more

reliability As the popularity of the automobile grew, it was no longer

regarded as a warm weather vehicle but one that could be used any time of

the year It became necessary to use a substance that would not freeze by

itself or that would lower the freezing point of water when mixed with it

Many people were satisfied to add any substance to water as long as it

depressed the freezing point, but there were those that warned against the

use of certain materials Salts, such as calcium or sodium chloride, were

known to be destructive to metal and were not recommended Glycerin

that was not chemically pure was said to attack both metal and rubber

components, and it rapidly degraded and had to be replaced, adding to the

expense of its use

The obvious concern for freezing of the coolant is indicated in a letter to

the editor of The Automobile in 1904 A substitute for water is suggested

in this extraction from the letter "I have just made a test of an antifreeze

article It is no less than an inexpensive lubricating oil I put this

clear (no water) in a 4-cylinder Toledo car and it cooled every bit as

much as water and it will not freeze" [8] Regardless of this individual's

experience, there never has been a trend to the use of lubricating oils for

cooling

The motorist turned next to the use of wood (methyl) alcohol for freezing

protection because it was cheaper than grain (ethyl) alcohol which was

taxed at $2.10 per proof gallon [9] The tax was removed from ethyl alcohol

in 1907 because it was being used increasingly for industrial purposes

Ethyl alcohol proponents claimed the following advantages over methyl

alcohol: (a) lower freezing point (only true for pure alcohol—a 50 percent

solution has a higher freezing point), (b) higher boiling point, (c) cheaper

because less of it was needed, and (cf) more uniform because it contained

no solids and required no filtering, and (e) less destructive to parts of the

cooling system [9] The growing need for alcohol is indicated in this

state-ment regarding availability: "If the plans of the United States Departstate-ment

of Agriculture are consummated, denatured alcohol will, within the next

few years, be manufactured by every farmer in the country from his waste

material" [9] Although this claim was never fulfilled, it is interesting that

similar claims are being suggested today in regard to the use of alcohol as

a gasoline substitute

There has been continued improvement in cooling system design and in

the quality of coolant materials Much of the credit for these improvements

belongs to organizations such as ASTM, the Society of Automotive

En-gineers (SAE), The Chemical Specialties Manufacturing Association

(CSMA), and similar organizations in other countries Information

bul-letins, standards, and specifications have been written to give guidance in

the selection and use of coolant materials ASTM Committee D15 on

Engine Coolants deserves much of the credit for these standards This

committee was formed in 1947 as the Engine Antifreeze Committee with

the following scope:

The study of engine antifreezes, including terminology, identification and classification,

methods of sampling and testing of engine antifreeze and cooling system corrosion

inhibitors; interpretation and significance of tests; and the preparation of specifications

This was quite an assignment, but over 20 methods, practices, or

speci-fications have been written and continually revised over the intervening

years, and these standards are used throughout the world Committee

members have contributed additional information and knowledge on the

subject through published papers, reports, and seminars Even the name

of the committee was changed to "Engine Coolants" in 1972 in recognition

of a greater concern for the entire coolant rather than just its antifreeze

aspect This continued activity has resulted in fairly well defined

pa-rameters for an engine coolant

Requirements of an Engine Coolant

The requirements of the coolant must be directed to fulfilling the

ob-jective of transferring heat from the engine to the radiator for dissipation

Many of the desired characteristics and requirements of an engine coolant

are listed in Table 1

No single material can satisfy all these requirements for an engine

coolant, so some concessions must be made The first requirement, high

specific heat and good thermal conductivity, is basic to the function of the

coolant to transfer heat from one site to another Water with a specific

heat of one is the best material available for this purpose It meets many of

the other desired requirements, but it fails to meet three very essential

categories; specifically, freezing, boiling, and corrosivity These

short-comings of water were mentioned in the brief review of the early

auto-mobiles, and it is interesting that even today there are still automobile

owners who disregard the requirements of a good coolant and use either

water alone as a coolant or highly diluted ethylene glycol antifreeze

To overcome the deficiencies of water, it was necessary to find a

sub-stitute for it or to mix it with other materials to improve its characteristics

The choice of materials has usually been divided among (a) salts, (b)

alcohols, (c) petroleum products, and (d) polyhydroxy alcohols (glycerol

and glycols) Unfortunately, these materials may satisfy those requirements

that water cannot, but they may be less acceptable in other areas

Salts have never been satisfactory because it requires a high concentration

TABLE 1 —Characteristics and requirements of an engine coolant

High specific lieat and good thermal conductivity Fluidity within the temperature range of use Low freezing point

High boiling point Noncorrosive to metals; minimum degradation of nonmetals Chemical stability over the temperature range and conditions of use Nonfoaming

Low flammability; high flash point Reasonable compatibility with other coolants or oil Low toxicity; no unpleasant odor

Reasonable cost; available in large quantities

of salt for freezing protection These solutions are highly corrosive, and it

is difficult to inhibit corrosion through the addition of small quantities pf

chemicals In addition, radiator-tube blockage may be enhanced through

salt recrystallization in tubes If a salt solution should leak into the engine

compartment, its high conductivity may contribute to a shortcircuit in

the electrical system

Alcohols have many satisfactory characteristics, and alcohol-water

mixtures were used successfully as engine coolants for many years They

fail to meet present-day requirements for a high boiling coolant that will

allow higher operating temperatures The flash point and flammability of

alcohols are not as satisfactory as that of other materials The corrosion

inhibitor systems in alcohol antifreeze coolants were not completely

satis-factory in the past, but it is presumed that they could have been improved

with further research if the justification had developed

Petroleum products, such as oil or kerosene, have had limited use as a

substitute for water mixtures The heat transfer characteristics are not

satisfactory, and engines tend to run hotter than when water mixtures are

used Petroleum-base coolants, if used, would be restricted to mild

tem-peratures and moderate driving conditions where little stress is placed on

the engine There is a tendency for these materials to be more flammable

than desired, and they are subject to ignition under certain conditions

Some products may attack polymeric materials in the cooling system

Finally, the engine may be damaged from overheating because the high

boiling point of these products does not provide the warning by boilover

that water-base materials do

A water solution of glycerol should be a satisfactory engine coolant

under most conditions if properly inhibited Corrosion was the principal

problem in its early use because inhibitors were not used However, glycerol

is not as good a freezing point depressant as the other polyhydroxy alcohols

The most satisfactory product in this category has been ethylene glycol,

which often has small percentages of diethylene glycol or propylene glycol

mixed with it Although ethylene glycol does not meet all the requirements

of a coolant, it provides a fairly good overall balance It has a specific heat

of about 60 percent that of water, but a 50:50 mixture of ethylene glycol

and water raises this to about 80 percent In addition, a 50 percent solution

has a 7°C (15°F) higher boiling point than water alone, and this provides a

definite advantage It has been demonstrated that a car at idle at an

ambient temperature of 38°C (100°F) that contains a 50 percent mixture

of ethylene glycol and water can be operated 40 percent longer than when

water alone is used before boiling occurs [10] Thus, some of the

dis-advantage of the lower specific heat is offset by the higher boiling point To

meet other engine coolant requirements, small quantities of chemicals are

added to the ethylene glycol concentrate: (a) inhibitors to prevent metal

corrosion, (b) alkaline substances to provide a buffering action against

acids, (c) an antifoam agent to reduce foaming tendencies, (d) a dye to

provide identity, and (e) a small amount of water to dissolve certain

chem-icals and to provide stability It is quite evident that the ethylene glycol

coolant concentrate is a carefully formulated product A basic outline of

the composition of such a product is shown in Table 2

ASTM Reference Coolant (D 3585-77) is a typical example of a

formu-lated product Certain physical and chemical properties of a product must

be checked to ensure that the product meets the specific requirements of

the cooling system Standard test methods have been developed by ASTM

Committee D15 for this purpose A list of these methods is shown in

Table 3 with a reference to the specific function of each method There is

no ASTM standard for the complete chemical analysis of an antifreeze

coolant, but published procedures are available from other sources [11,12]

Many of these standard methods are used for quality control to ensure

that a product falls within certain specified limits or to determine whether

a product meets engineering specifications that may include composition

or physical characteristics Some methods may be used in test programs to

TABLE 2—Basic composition of ethylene glycol coolant concentrate

Composition Concentration, % Ethylene glycol 80 min Other glycols 15 max Multi-inhibitor system

Buffer or neutralizer Foam suppressor Dye

Water

TABLE 3—Standard ASTM test methods to determine physical and chemical properties of

an engine coolant concentrate

Numerical

Designation Title

D 1119-65 Ash Content of Engine Antifreezes, Antirusts, and Coolants

D 1120-72 Boiling Point of Engine Coolants

D 1121-72 Reserve Alkalinity of Engine Antifreeze, Antirusts, and Coolants

D 1122-58 Specific Gravity of Engine Antifreezes by the Hydrometer

D 1123-73 Water in Engine Coolant Concentrate by the Iodine Reagent Method

D 1177-65 Freezing Point of Aqueous Engine Coolant Solution

D 1287-78 pH of Engine Antifreezes, Antirusts, and Coolants

D 1881-73 Foaming Tendencies of Engine Coolants in Glassware

D 1882-66 Effect of Cooling System Chemical Solutions on Organic Finishes for

Automotive Vehicles

D 3634-77 Trace Chloride Ion in Engine Antifreezes/Coolants in the Presence

of Mercaptobenzothiazole

determine the changes to the coolant after use in a car The absolute

composition of the coolant concentrate can only be obtained by chemical

analysis, but many of these standard tests will provide an indirect measure

of composition For example, specific gravity will differentiate between

methyl alcohol and ethylene glycol, but it is not absolute because some

combination of materials can be mixed to give a specific gravity that may

be close to that of a single substance Combinations of tests can be more

definitive If boiling point and freezing point tests are used in addition to

that of specific gravity, the identification is narrowed to that substance that

can satisfy the specific results from all three tests The advantage to the

use of these methods over that of a complete analysis is one of time

be-cause they can usually be completed rather fast Thus, these physical test

methods are well suited to quality control

Reserve alkalinity and pH are not directly related to the performance of

a product, but because they indicate whether a product is alkaline and to

what degree, they are a measure of the concentration of alkaline inhibitors

that may be used to prevent corrosion Ash is also a qualitative measure of

inorganic inhibitors, but because it includes all noncombustibles, it is not

usually recommended for this purpose It can be used to show the presence

of harmful substances, such as chloride salts, that might have been used as

a freezing point depressant; the total ash would be much higher than that

usually found for glycol antifreeze coolant

There are a number of properties of the coolant concentrate that concern

its storage, handling, and vehicle use that are of interest The product

should not react with protective coatings on the vehicle It should have

reasonable chemical stability over the temperature range and conditions of

use to ensure that it continues to satisfy the requirements of the system

and to avoid the formation of degradation products that might make the

product more corrosive Stability during storage is also an important

re-quirement of the product Both short- and long-time storage tests should

be run to ensure that the concentrate is stable at normal storage

tempera-tures and that it is not corrosive to the storage container; the corrosivity of

the concentrate can be quite different from that of a dilute solution

There are a number of concerns related to the satisfactory use of a

cool-ant in service For example, the coolcool-ant will foam if air or exhaust gas

leaks into the cooling system The resultant increase in coolant volume can

cause loss of coolant which, in turn, can cause overheating Foaming should

be minimized through the use of an antifoam agent Foaming tests can be

conducted in the laboratory to determine whether an antifoam is effective,

but this assumes the presence of the antifoam in the cooling system Many

antifoam agents are not miscible with the ethylene-glycol-water mixture

and float on the top Thus, when coolant is withdrawn from a large bulk

storage tank for addition to the automobile cooling system during

manu-facture, the antifoam can be excluded unless the coolant mixture is

agi-tated The same situation applies to quart or gallon containers of antifreeze

bought by the car owner for addition to the cooling system; the antifoam

may be left on the walls of the container if it is not shaken adequately

be-fore addition The life of the antifoam material can be short during use in

the cooling system, either because the antifoam coats out on metal surfaces

or is degraded through a reaction with other chemicals Thus, laboratory

test results cannot be expected to predict the life of the product, but the

tests should help to exclude inferior products from use

Compatibility of a coolant with other fluids is a desirable requirement

but not as essential as other requirements For example, if coolant should

leak into the oil system, reasonable compatibility is desired to avoid

sludg-ing of the oil, but excessive leakage will cause other problems so it is best

to avoid the leak through proper maintenance Compatibility with other

coolants is desirable because of the occasional need to add coolant to make

up for a loss The use of incompatible materials may result in the formation

of a precipitate through the interaction of inhibitors or the reaction of an

inhibitor with soluble substances in the water The precipitate affects heat

transfer through accumulation on metal walls and in radiator tubes, and the

loss of inhibitors affects corrosion protection The problem can be avoided

in part by the use of the same antifreeze coolant for initial fill and makeup

and by using water with low dissolved solids content

Safety requirements should also be given consideration Low flammability

and high flash point are required to avoid the hazards of fire in case of a

spill or leak onto a hot engine component Products made from ethylene

glycol and water meet this requirement well Minimal effect to the skin is

desirable because the product is handled both on the automobile production

line and by the customer in changing his coolant The avoidance of an

un-pleasant odor is understandably desirable

The material that satisfies most requirements still fails if it cannot be

provided at a reasonable cost and in large quantities It has been shown in

a survey conducted by the CSMA that about 200 million gallons of

anti-freeze on an average have been marketed annually over the past five years

[13] This use requires a ready source of material

Corrosion Test for Engine Coolants

A good engine coolant must minimize the degradation of nonmetals and

the corrosion of metals Excessive deterioration of any material can lead to

a loss of coolant through a hole or break in the wall of the material or to

reduction in the strength of a material and subsequent malfunction of the

part The products of the deterioration can deposit on metal surfaces and

reduce heat transfer or in tubes and restrict coolant flow In practically all

cases the ultimate effect is an overheated engine that causes further loss of

coolant and subsequent engine malfunction plus a notable reduction in

heat to the passenger compartment where it is needed for warmth in the

winter So it is important to determine the effects of the coolant on

ma-terials

There have been few specific tests to evaluate the effects of coolants on

nonmetallic materials, such as radiator or heater hoses Standard test

pro-cedures to measure the hardness of elastomeric materials have been used

to determine the change in material hardness after exposure to coolant

solutions, but the interpretation of results is difficult because the

relation-ship to service life is not clearly understood Much of the evaluation is done

using component parts, such as hoses, either in service simulation tests,

engine or dynamometer tests, or in actual car operation Failure of the

material to perform satisfactorily is usually determined by changes in

physi-cal characteristics, such as loss of compressibility around clamps,

perma-nent set, breakdown of fibers or the basic material, etc Of course, loss of

coolant from a cracked or burst hose is an immediate indication of failure

Polymeric materials may degrade, and the degradation products that are

leached into the coolant may increase the rate of metal corrosion

There-fore, it is advisable to include nonmetals in the system when testing for

metal corrosion

The corrosion of metals in the cooling system has received more attention

than the degradation of nonmetals, and perhaps rightfully so because

metals constitute the largest percentage of material surface Tests may be

run to evaluate the effects of a coolant or of a particular type of corrosion

on a specific metal, but it is more usual to evaluate complete metal systems

Metal specimens or complete parts, such as a pump or radiator, may be

used in the evaluation The important consideration in any event is that

one must have a reasonable understanding of corrosion and how it is

pre-vented and in the interpretation and use of test results

The objective in coolant testing is to simulate, in part or in whole, the

physical and chemical reactions that occur in the engine cooling system, at

least those that are thought to affect corrosion This is a very difficult task

because operational conditions are seldom constant within a single vehicle

and will vary widely among different vehicles The best that can be done in

most cases is to select average conditions that can be controlled and that

are assumed to affect corrosion It is evident that no single test can provide

complete corrosion information about all driving conditions, and it is

usually necessary to run a series of different tests to obtain a comprehensive

evaluation of a product or to study the effects of operational parameters

[14] The time required to run a large number of tests and the space

essary to accommodate the tests are important Therefore, it may be

nec-essary to accelerate some test parameters to reduce the time period, but

there may be some sacrifice in the reliability of results once this is done

This does not mean that such tests are poor and the results meaningless

Much can be learned from well-designed, accelerated tests as long as the

limits of the test are understood and appreciated In most cases the results

cannot be applied directly to service, but they can be used for guidance to

indicate the direction for further testing or they may indicate potential

problem areas

Coolant tests can be divided into the three main categories shown in

Table 4 The list of tests is not to be considered all inclusive but rather to

indicate the scope of tests Laboratory tests are usually simpler and less

time consuming than the other tests, and they are generally more

economi-cal to run

Laboratory tests have been one of the primary means for evaluating

engine coolants Examples of different kinds of laboratory test equipment

are shown in Fig 1 One of the first tests developed by ASTM Committee

D15 was the Corrosion Test for Engine Coolants in Glassware (D 1384-70)

(Fig le) Metal specimens are immersed in a solution of antifreeze coolant

and corrosive water The test is run for two weeks at a constant temperature

of 88°C (190°F), constant solution volume of 750 ml, and with continuous

aeration Metal surface areas of specimens of six different alloys commonly

found in the cooling system are about equal

Although this immersion test is typical of many simple corrosion tests, it

is not too difficult to list a number of variations from service conditions

that might have an effect on the credibility of results from this test [15]

These variations cannot be considered unique because they could apply

equally to other tests Some examples are as follows:

• The coolant temperature in service varies from ambient to hot

(per-haps as high as 121°C (250°F)) and then cools again to ambient, and these

cyclic temperature changes can affect the stability of protective films at

metal surfaces and increase corrosion The higher temperatures in service

may increase corrosion rates and can contribute to the breakdown of glycols

TABLE 4—Engine coolant corrosion tests

Laboratory Tests Glassware tests Electrochemical tests Simulated service tests Special tests

Pump cavitation High temperature-high pressure Impingement

Crevice Aluminum transport Engine Tests

Engine dynamometer test No-load engine test Field Service Tests Proving ground tests Fleet tests

FIG 1—Laboratory tests: (a) Controlled-potential test equipment, (b) cell for

electrochemi-cal tests, (c) heat-transfer test, (d) metal conductance test, and (e) simple immersion test

and chemical innibitors The higher metal surface temperatures at heat

rejection areas can affect the corrosion rate of the metals involved

• Aeration in service may be greater or less than the constant purge

used in this test, affecting the oxidation phases of the corrosion reaction or

the breakdown of glycol and inhibitors Entrained air can increase corrosion

through erosion or impingement

• Metals in service are not found with a surface area ratio of 1:1, and

the volume of coolant to metal surface area is different in service from that

in the laboratory test These variations in ratios of surface areas or solution

volume to surface area can affect corrosion rates and galvanic effects

be-tween metals

• The surface condition of metals prepared for test is different from

that of metals in the engine cooling system

• The corrosive water used in the test represents an average degree of

corrosivity, but environmental conditions in service can be quite different

• The depletion rate of inhibitors cannot be taken into account in a

short two-week test

It should not be assumed from the above comments that the test is

use-less In fact, it can serve a very useful purpose as long as its limitations are

recognized, and the results are interpreted in the proper light When the

ASTM D-1384 test method was adopted as a standard, its limitations were

indeed recognized by ASTM Committee D15 because it is recommended

for use as a screening test This implies that the test can be used to make

a division between presumably good inhibitor systems and poor ones

Furthermore, it has the advantage of permitting a large number of

candi-date inhibitors or products to be evaluated in a short time and with the use

of little space The assumption is made in this case that results are

indica-tive of performance in service Although this may be true, there can be

errors because of the differences discussed previously or the fact that

inhibi-tors may be more effective under liquid flow rather than static conditions

or they may provide good initial protection but fail rapidly with longer use

in service

A special phase of testing came into prominence in recent years involving

the use of electrochemical techniques [16] These tests can be quite useful in

studying mechanisms of reactions, and they can often supply information

that is difficult to obtain by other means The polarization resistance method

(Fig la,b) has been used for many studies because corrosion rates can be

measured rapidly, and the rate is that which is measured at any moment of

time as opposed to average rates that are usually determined by weight loss

measurements This capability offers the advantage of being able to

mea-sure a rate in situ, thus allowing one to determine the effects of operating

parameters on corrosion [17] Walker studied the effect of coolant flow on

the corrosion of aluminum through the use of an actual pump and timing

chain cover [18] The results suggested continuous passivity at low flow

rates, and a breakdown and repair of the passive oxide film at higher flow

rates The technique was also used to measure corrosion rates in an

auto-motive cooling system under operating conditions A specially designed cell

(Fig 2a) was installed in the hose allowing the coolant to flow through the

cell The results showed that temperature was the dominant parameter that

determined the rate of corrosion in a normal operating vehicle

Other electrochemical methods that have been used include the

measure-ment of the change in metal conductance to show the effects of corrosion

(Fig Id) [19], the direct measurement of galvanic corrosion current to

show the effect of bimetallic couples [20], and the determination of anodic

polarization curves to study solution effects and metal variations (Fig la, b)

[16], These methods can provide useful information within the constraints

of the technique, and they should be regarded as tools that can provide

?'

FIG 2—Specimen holders: (a) For electrochemical measurements in recirculating coolant

tests or engine or car tests, and (b to d) variations in holders for weight-loss measurements in

engine or field service tests

another piece of information that can assist with development work or the

resolution of problems The information can never be absolute because

(a) the tests are too short and can only indicate initial and short-time

effects, (Z>) solution volumes in laboratory tests are usually small and may

not reflect the proper metal-surface-to-solution-volume ratio, and (c) the

chemistry of reactions may be different because many service parameters

are not included in the test However, when the tests are used and the

re-sults are interpreted properly, the information can provide much guidance

in design and development work and in the selection of important

param-eters for more extensive testing by other techniques

Other tests have been developed to simulate service conditions more

closely, with the goal of trying to improve the reliability of results A

num-ber of these laboratory tests are shown in Fig 3 The ASTM Method for

Simulated Service Corrosion Testing of Engine Coolants (D 2570-73)

(Fig 3a) that was developed by ASTM Committee D15 is an example of

such a test The metal-surface-to-coolant-volume ratio was improved

through the use of automotive parts and a cast iron or aluminum pot with

a large surface area to simulate the engine block; the use of these larger

parts increased the coolant volume until it approached that in the engine

cooling system The coolant is circulated at the flow rate that occurs in an

automobile at 96 km/h (60 mph) Test time was increased to about seven

weeks Although the temperature is still not as high as that in service, an

effort was made to cycle it to some degree by cooling the system to ambient

temperature twice a week over an 8-h period Although a tremendous effort

has been made to improve simulation of service conditions, one is never

sure that the exact conditions that affect corrosion have been selected

Interlaboratory test results have shown poor reproducibility for solder

and aluminum alloys Repeatability tends to be somewhat better but not as

good as desired Although the cause of these variations has not been

de-termined at this time, it would appear that there is some variation in

sur-face condition that affects the formation or breakdown of the passive film

on these two materials The test has had much value as a research tool and

as a qualification test, as long as its limitations were appreciated and

sufficient tests were run to establish the limits of repeatability and

repro-ducibility

Corrosion may occur under actual driving conditions but not in a general

corrosion test, such as the simulated service test recommended by ASTM

(D 2570-73), because some condition that occurs while the car is operating

is not simulated in the laboratory test Special tests have been developed

to respond to these needs As discussed earlier, corrosion rates at a metal

surface can vary, depending on whether the metal is immersed in a solution

that is heated by an outside source or whether the solution is heated by

transmitting heat through the metal of interest In the latter case, metal

surface temperatures are higher than the bulk solution temperature, and

^tt^

FIG 3—Recirculating-coolant laboratory tests: (a) ASTM Simulated Service Test, (b)

elec-trochemical cell installed in hose, (c) heat being transferred through metal specimens, (d)

ASTM cavitation test, and (e) thermostat housing crevice test

surface boiling can take place, which is what often happens when the

engine is being operated Various approaches have been tried to simulate

this effect, and some examples are shown in Figs Ic and 3c In all cases,

the corrosion has been more severe when the heat is transferred through

the metal than when the metal is simply immersed in a hot solution It is

more difficult to inhibit metal corrosion under these conditions, and tests

have shown that high concentrations of phosphate inhibitor can increase

the corrosion of aluminum substantially at heat-rejecting surfaces [21]

Tests have been developed to emphasize other specific car-service

con-ditions Some of these tests involve the impingement or erosion effect that

occurs from high velocity fluids striking a metal surface [22] High

tem-peratures and high pressures have been used to create different stresses on

the metal surface and to increase the rate of corrosion [23] Specific types

of corrosion have been emphasized in tests, such as the crevice corrosion

that can occur between the neck of the thermostat housing and the rubber

hose (Fig 3e) [14] It is very possible that the solubility of the products of

corrosion may be greater at the point of formation because of a higher

solution temperature than at some other part in the system, such as the

radiator, where the solution is being cooled [24,25] Precipitation that

occurs at these cool areas can reduce heat transfer or coolant flow This

mechanism is thought to be applicable to aluminum alloys, and efforts are

being made currently to develop a test procedure that will simulate this

condition

ASTM Test for Cavitation Erosion-Corrosion Characteristics of

Alumi-num Pumps with Engine Coolants (D 2809-77) was developed in

Com-mittee D15 to evaluate the special effects of cavitation-erosion on metal

corrosion (Fig 3rf) Although this type of corrosion can occur on the

cool-ant side of cylinder walls under certain conditions, it is more often

associ-ated with an aluminum pump or timing chain cover—the parts used in the

development of the test The advantages to the test are that it can show

the types and concentrations of inhibitors that affect cavitation and the

characteristics of pump design that contribute to the effect

The continued improvement in corrosion test procedures is usually

ob-tained by a better simulation of the operating process in service In the

case of the cooling system, this is done most appropriately by the use of an

actual engine The engine is operated either under controlled laboratory

conditions or in an operating vehicle (Fig 4), and specimens may be

in-stalled in the cooling system using holders, such as those shown in Fig 2

Two types of laboratory engine tests have been used predominately although

there can be many modifications [14] The ASTM Test for Engine Coolants

by Engine Dynamometer (D 2758-78) (Fig Aa,b) is the more demanding

of the two tests because the conditions can be varied to simulate

heavy-duty driving conditions The second engine test requires little attention to

run and is more economical than the dynamometer test; it is commonly

FIG 4—Engine tests: (a,b) Dynamometer tests, (c to e) test specimen holders installed for

field service tests, and measurement being made electrochetnically in (d)

referred to as the no-load engine test In this test procedure the engine is

operated continuously on a test stand without an applied load, although

the engine can be turned on or off to cycle the test Both tests provide the

same physical conditions found in car operation, that is, the same design,

metal ratios, volume of solution, and transfer of heat through the metal It

should be expected that results will be more reliable than those from other

tests, and they are, even though engine tests are still laboratory tests in the

sense that operating conditions are controlled and may even be accelerated

Actual parts can be examined, but results can also be based on the

ex-posure of corrosion specimens in the system Repeatability and

reproduci-bility of results from specimen exposure are normally better than those

from simpler tests

Engine tests also have their shortcomings Elapsed exposure time is less

than that in service even though operating time may be comparable, so the

results may not be indicative of long-time coolant use in a vehicle Although

test conditions can be varied in an engine test, the variability found among

the multitude of driving conditions in service would be difficult to

dupli-cate The no-load engine test is less costly than the dynamometer test, but

one of the drawbacks to any engine test is the cost Not only are the engine

parts costly, but only a limited number of tests can be run at one time

be-cause of space requirements and the needed instrumentation These tests

are not used extensively for these reasons

The final approach to all corrosion testing is the evaluation of materials

under actual use conditions, such as ASTM Test for Engine Coolants in

Vehicle Service (D 2847-78) (Fig 4c,d,e) Specimens of metals may be

in-corporated in the cooling system to determine corrosion rates or parts from

the system may be evaluated after the test In most cases the coolant is

examined and analyzed before and after test to determine changes

Variations in these tests are usually associated with variations in vehicle

service, which can be divided into about three subcategories

Proving Ground Service Tests

The evaluation is done in vehicles operated under specific, controlled

conditions either at a proving ground site or outside the site but following

a specific route or pattern Vehicles may be operated continuously to

accelerate the effects of corrosion

Fleet Service Tests

A specific group of vehicles is selected that will operate under some

repetitive set of conditions, such as those associated with police cars, taxis,

or municipally driven vehicles Some factor or type of corrosion is often

stressed under these conditions; for example, metals are exposed to high

temperatures in taxis for long periods because of almost continuous

opera-tion and lots of time at idle Cavitaopera-tion condiopera-tions can be stressed in police

cars because of faster accelerations and high-speed driving

Group Service Tests

These tests are quite similar to fleet tests except that driving conditions

are less controlled or perhaps even unknown Tests in this category include

employee-owned cars, sales personnel cars, and perhaps rental cars

If a sufficient number of test vehicles are run, the results can provide a

good indication of what to expect in service operation Results should not

be accepted without proper study because the type of service, the

main-tenance of the vehicles, and the conditions and rate at which mileage was

accumulated may have varied If specimens were used to determine

corro-sion rates, the results will not be indicative of the corrocorro-sion effect that

occurs at heat-rejecting surfaces However, corrosion of this type may be

indicated by the accumulation of insoluble corrosion products on metal

surfaces or in the coolant

The cost of running vehicle tests is high, but as long as a good statistical

sample is used, they represent the best method to obtain reliable

informa-tion about the performance of materials in the engine cooling system and

about potential problem areas

Summary

Heat must be removed continuously from its source in the internal

com-bustion engine to avoid overheating and subsequent engine malfunction

This objective is accomplished through the use of a liquid coolant that is

pumped continuously through the engine system to a radiator where some

of the heat is dissipated to the air To ensure adequate coolant performance

and continued reliability, the coolant must have a high specific heat, a

low freezing point, and a high boiling point It must be noncorrosive,

chemically stable, and nonfoaming Other desirable requirements of a

coolant that must be considered are its flammability, toxicity, and odor

Above all, it must be available in quantity and at a reasonable cost

In order to determine whether a coolant does have the desired

require-ments, it must be evaluated and tested There are tests that can be used

to determine certain chemical and physical characteristics of the coolant,

and if it is considered necessary, the coolant can be chemically analyzed

by standard procedures It is more difficult and time-consuming to

deter-mine whether the coolant has an appropriate inhibitor system to prevent

corrosion under all conditions of service The means of testing fall into

three general categories: (1) laboratory tests, (2) engine tests, and (3) field

service tests The tests become progressively more complex, but the results

can be expected to be more reliable as the simulation of field conditions

improves When tests are properly used and the results properly

interpre-ted, one can expect to provide a good-operating engine cooling system

[4] Petard, R M., The Automobile, Vol 10, 6 Feb 1904, p 172

[5] River, A D., The Automobile, Vol 10, 25 June 1904, p 684

[6] "The Cooling System," The Automobile, Vol 8, 17 Jan 1903, p 72

[7] McKilwin, "Keeping the Land Yacht Shipshape," Harpers Weekly, Harper & Brothers,

New York, N.Y., 2 Jan 1909

[8] Wilson, A T., The Automobile, Vol 9, 16 Jan 1904, p 63

[9] "Comparative Value of Non-Freezing Solutions," Harpers Weekly, Harper & Brothers,

New York, N.Y., 1909

[10] Beynon, E., Cooper, N., and Hannigan, H., Soap and Chemical Specialties, Vol 47,

No 2, 1971, p 44

[//] Beynon, E in Encyclopedia of Industrial Chemical Analysis, F Snell and C Hilton, Ed.,

Interscience Publishers, New York, N.Y., Vol 6, 1968, p 9

[12] Loranger, R B., "Methods for the Chemical Analysis of a Formulated Antifreeze,"

Research Publication GMR-833, Research Laboratories, General Motors Corp., Warren,

Mich., Jan 1969

[13] "Comparison of the Antifreeze Market for Years 1973 Through 1977," Chemical

Special-ties Manufacturing Association, Annual Survey, 22 Feb 1978, private communication

[14] Rowe, L C in Handbook on Corrosion Testing and Evaluation W Ailor, Ed., Wiley,

New York, 1971, p 625

[15] Rowe, L C in Corrosion Inhibitors, C Nathan, Eds., National Association of Corrosion

Engineers, Houston, Tex., 1973, p 173

[16] Walker, M S and Rowe, L C in Electrochemical Techniques for Corrosion, R

Ba-boian, Ed., National Association of Corrosion Engineers, Houston, Tex., 1977, p 79

[17] Walker, N S and France, W D., "In Situ Electrochemical Evaluation of Corrosion

Inhibition in an Automotive Cooling System," Research Publication GMR-830, Research

Laboratories, General Motors Corp., Warren, Mich., March 1969

[/*] Walker, M S., Materials Performance, Vol 13, No 7, July 1974, p 37

[19] Rowe, L C and Walker, M S., Corrosion, Vol 17, No 7, July 1961, p 353t

[20] Walker, M S., "The Galvanic Corrosion Behavior of Dissimilar Metal Couples Used in

Automotive Applications," Research Publication GMR-2538, Research Laboratories,

General Motors Corp., Warren, Mich., Oct 1977

[21] Walker, M S., Materials Performance, Vol 12, July 1973, p 29

[22] "Corrosion Test for Engine Coolants," LP-461H-94, Engineering Office, Chrysler Corp.,

Detroit, Mich., 1966

[23] "High-Temperature-Pressure Corrosion Test for Engine Coolants," LP-461H-111,

Engineering Office, Chrysler Corp., Detroit, Mich., 1967

[24] Kawamoto, S and Suzuki, H., "The Problems of Using Antifreezing Corrosion Inhibitor

in Aluminum Engines," Technical Review, Mitsubishi Heavy Industries, June 1974,

p 102

[25] "Omni-Horizon Troub\e," Radiator Reporter, Vol 6, March 1978, p 25

Experience of the British Standards

Institution in the Field of Engine

Coolants

REFERENCE: Mercer, A D., "Experience of the British Standards Institution 'm the

Field of Engine Coolants," Engine Coolant Testing: State of the Art, ASTM STP 705,

W H Ailor, Ed., American Society for Testing and Materials, 1980, pp 24-41

ABSTRACT! The development in the United Kingdom of specifications and standards

for inhibited coolants for internal combustion engines is briefly reviewed Three types of

British Standards that are currently in existence are described These are concerned

respectively with compositional specifications, recommendations for the care and

maintenance of cooling systems, and, more recently, performance specifications The

motivation for the preparation of these standards and their technical contents are

re-viewed and some indications of the direction of future work discussed

KEY WORDS; standards, specifications, antifreezes, coolants, corrosion, corrosion

tests, corrosion inhibitors

The need for standards, whether national, military, or company

state-ments of recommended or mandatory procedures, in a particular field of

ac-tivity usually arises as that acac-tivity begins to affect increasing numbers of

manufacturers and users This has certainly been the case with engine

coolants in that the demand for, and use of, standards has grown with the

in-crease in the number of vehicles powered by internal combustion engines

Despite advances in engine technology and the advent of, for example,

elec-tric propulsion, the need to remove heat from the site of the explosive

reac-tion responsible for providing the motive power has remained an unchanging

feature of the vast majority of engines Also, for well over sixty years the basic

arrangement of the cooling circuit of an engine cooling circuit has remained

essentially unchanged

Cooling systems inevitably contain a number of different metals and alloys

and corrosion of some or all of these by the coolant will be unavoidable even

'National Physical Laboratory, Teddington, Middlesex, United Kingdom

when this is pure water The problem is intensified when various types of

an-tifreeze agent are incorporated into the coolant Ethylene glycol (ethanediol)

is frequently used for this purpose but produces acidic oxidation products,

particularly in the presence of hot metal surfaces, vigorous aeration, and

metal ions each of which can catalyze the oxidation process The consequent

corrosive action of such coolants is now almost invariably countered by the

additional presence of corrosion inhibitors

However, although systematic studies leading to the development of

stan-dard methods of preventing corrosion by incorporation of corrosion

in-hibitors into the coolant have taken place only within the last thirty years,

some research was in progress prior to that time and inhibited coolants were

available Indeed, a review made by the National Research Council of

Canada (NRCC) in 1942 of antifreeze literature and patents in Chemical

Abstracts and the Industrial Arts Index for the period 1907-1941 quotes 138

references to the subject [1]? The importance of the problem was more

ex-plicitly recognized in the United States By 1952 the National Bureau of

Standards (NBS) had already been issuing leaflets on automotive antifreezes

for over 25 years "as an expedient means of answering the many letters of

inquiry on this subject [2]."

Apart from the impetus given by the increase in the number of vehicles on

the roads, there were two other factors that contributed to the awareness of

the need for maintaining cooling systems in a clean uncorroded condition

and in turn to the eventual introduction of British standards One was the

importance, before the jet-engine era, of the problem of preventing corrosion

in aircraft engines since this could have more serious repercussions than in

land vehicles Sidery and Willstrop in a paper in 1939 to the Royal

Aeronautical Society [3] pointed out the corrosive nature of ethylene glycol

solutions and described the use of the phosphate salt of triethanolamine as a

corrosion inhibitor; the organic salt being preferred to the alkali metal salts

since the latter were less soluble in glycol The second factor, and one clearly

linked with the first, was the effect of the Second World War Shortages of

glycol meant that many chemicals had to be evaluated, often in a very short

period, as alternatives and freedom from corrosivity was an obvious

require-ment Much of this work in the United Kingdom was conducted in the

Chemical Research Laboratory which later became the National Chemical

Laboratory and is now part of the National Physical Laboratory The result

of these activities was a better understanding of the properties and

effec-tiveness of various antifreeze agents and corrosion inhibitors

This work, conducted in the middle and late 1940s, had shown that there

was no realistic substitute for ethylene glycol as the antifreeze material

Com-positions based on calcium and magnesium chlorides, chemicals that were

more readily available, were extremely corrosive and could not be

satisfac-2 The italic numbers in brackets refer to the list of references appended to this paper

torily inhibited, a fact that was also recognized by the NBS [2] Nevertheless,

the work revealed the extremely good properties of sodium benzoate as a

cor-rosion inhibitor for steel in aqueous solutions This property was not,

however, maintained with cast iron and the additional presence of sodium

nitrite was necessary to protect this material [4,5] The results of this work

were published in the scientific press and the benzoate-nitrite inhibitor for

ethylene glycol antifreeze was rapidly adopted by industry in the United

Kingdom and subsequently, in the rest of Europe

Work in the laboratories of Rolls-Royce during and after the war years had

led to the introduction of an antifreeze consisting of ethylene glycol

contain-ing triethanolammonium phosphate (TEP) and sodium

mercaptobenzo-thiazole (NaMBT) as corrosion inhibitors [6] This coolant was developed

specifically for the Merlin aircraft engine which was predominantly of light

alloy construction The presence of NaMBT was necessary to counteract the

corrosive action of triethanolamine on copper and the consequent corrosion

of aluminum alloys

Experience in service soon showed that, although TEP-NaMBT and

ben-zoate-nitrite formulations had been primarily intended for aluminum alloy

and cast iron systems respectively, each gave satisfactory performance with

the other material and could thus be used generally in commercial vehicles

Nevertheless, in practice the distinction was recognized between the original

purposes of the two types of formulation, so that instructions for use often

recommended the type of engine construction for which each was more

suitable

The commercial success of these formulations, and also of an ethanediol

antifreeze containing sodium tetraborate as a corrosion inhibitor that had

been introduced at about this time, thus provided a suitable background for

moves to produce British Standards for inhibited antifreezes

British Standards

Compositional Types

Following a request to the British Standards Institution (BSI) by the Local

Authorities' Advisory Committee, a technical committee was established in

1956 under the authority of the Chemical Industry Standards Committee, to

prepare specifications for inhibited antifreezes Preliminary discussions led

to the conclusion that there were three main types of inhibitor formulation

then in use in the United Kingdom Work went ahead to prepare three

separate standards based on these formulations for use with ethylene glycol,

now known as ethanediol The composition of the latter was required to

com-ply with BS 2537 for specific gravity, distillation range, ash, acidity, and

chloride and sulphate contents [7]

These standards, namely, BS 3150, BS 3151 and BS 3152 were issued in

1959 and specified the following requirements^ for the undiluted ethanediol:

• BS 3150 (also known as Type A) [8] will contain TEP and NaMBT;

that is, ethanediol containing 0.9 to 1.0 percent (by weight) of phosphoric

acid (H3PO4) with addition of triethanolamine to give a pH value of a 50

per-cent (volume/volume) aqueous solution of 6.9 to 7.3, together with 0.2

to 0.3 percent (by weight) sodium mercaptobenzothiazole calculated as

C7H4NS2Na

• Bs 3151 (Type B) [9] will contain sodium benzoate and sodium nitrite;

that is, ethanediol containing 5.0 to 7.5 percent (by weight) of sodium

ben-zoate calculated as C7H502Na and 0.45 to 0.55 percent (by weight) of sodium

nitrite calculated as NaN02; the pH value of a 25 percent (volume/volume)

aqueous solution to be 7.0 to 8.5

• BS 3152 (Type C) [10] will contain sodium tetraborate, that is

ethanediol containing 2.4 to 3.0 percent (by weight) of sodium tetraborate

decahydrate calculated as Na2B407 • IOH2O; the pH value of a 25 percent

(volume/volume) aqueous solution to be 7.8 to 8.2

The recommended dilution for each type is 1:3 with water to give 25

per-cent ethanediol solutions which provide adequate freezing point depression

in the United Kingdom and the necessary concentration of inhibitors

An appendix to each standard briefly describes the purpose and use of

an-tifreeze solutions with comments on the preparation of cooling systems,

maintenance of the antifreeze concentration, the useful life of an antifreeze

solution and a note on the inadvisability of mixing the different types

Standards for the Care and Maintenance of Cooling Systems

It was soon realized that the information given in the appendixes to these

standards could usefully be extended to cover the general treatment of

cool-ing systems to provide optimum freedom from corrosion These

considera-tions led to the publication in 1965 of BS 3926, "Recommendaconsidera-tions for the

Use and Maintenance of Engine Cooling Systems [ / / ] " The practice of

us-ing inhibitors in coolus-ing waters, as opposed to antifreeze solutions, was

in-troduced in this standard The intention was to overcome the problems that

could arise when the user drained the inhibited antifreeze from the engine at

the end of the winter and replaced it by uninhibited water on the assumption

that this was an acceptable procedure, since corrosion was widely believed to

be associated only with the presence of antifreeze The alternative procedure

of leaving the inhibited antifreeze in the cooling system for twelve months

was also recommended This was an arbitrary period of time based on

•'This has been summarized from the standards that also describe specific gravity and freezing

point requirements, instructions for pH determination, marking and analytical methods, etc

general experience with the three types of inhibitors described in BS

3150-3152

The approach in BS 3926 was extended further in BS 4959,

"Recommen-dations for Corrosion and Scale Prevention in Engine Cooling Water

Systems," issued in 1974 [12], which presented a more detailed discussion of

the development of scale (water hardness) and corrosion and of the

preven-tion of these phenomena British Standard 4959 includes definipreven-tions of types

of corrosion, for example, general wasting, pitting, crevice corrosion, etc.;

features of protective action against scaling and corrosion; cleaning and

in-hibition of cooling systems; factors affecting the choice of treatment; and a

summary of the properties of various types of corrosion inhibitors The

pro-tection of the cooling systems of engines that remain filled with coolant

dur-ing idle periods or which are drained durdur-ing longer periods, for example,

during transport and storage, is also considered The recommended

pro-cedures apply to aqueous coolants generally and are not restricted to

an-tifreeze solutions

Despite the general applicability of the three standards BS 3150-3152, the

view was expressed in the United Kingdom at the time of their publication,

and subsequently, that it would be more relevant to base standards on

per-formance in specified tests rather than on clearly defined compositions A

performance specification would have the merit of allowing new formulations

to be introduced on the basis of quantitative tests without necessarily

divulg-ing the inhibitor composition.'' Also, an opportunity would be provided for

an objective assessment of the effectiveness of the inhibitors rather than the

more qualitative assessment of general satisfactory experience attributed to

the compositional specifications of BS 3150-3152 Furthermore, a

perfor-mance specification would be less restrictive than formulation-based

specifications and would allow for the introduction of "general purpose"

an-tifreezes

Performance Standards

At the time of preparing BS 3150-3152 the testing of coolants had not

reached the stage where a performance test could be recommended with any

degree of confidence A subsidiary panel was therefore formed in the early

1960s to consider the possibility of devising a performance standard

The problems in producing such a standard were immediately obvious,

particularly in relation to methods for assessing corrosion resistance; for

ex-ample, the form that the test should take and the method to be used for a

quarilitative assessment of the effectiveness of inhibition of the formulation

''However, in order to substantiate that formulations subsequently supplied are equivalent to

formulations originally approved by performance tests, some tests to establish chemical identity

will be required in order to eliminate the need for repeating expensive performance tests

Of these the latter has presented the greater difficulty and has not as yet been

entirely resolved The problem is to devise test methods and criteria that will

adequately represent service conditions and equally ensure that good

in-hibitor formulations are not rejected as a result of shortcomings of the test

By 1973 the work of the committee had reached a stage where it was

con-sidered desirable to publish a document describing in some detail the

methods that should be recommended for performance testing This

stan-dard, BS 5117 [13], was issued in 1974 Part of the foreword to this document

reads as follows: "It is emphasized that this standard is only concerned with

methods of testing antifreeze solutions No attempt has been made to set

criteria of satisfactory performance for any test method described and no

such criteria should be construed from the wording of the test methods or any

explanatory paragraphs." Thus, the document was restricted to test methods

only and was issued in an attempt to get wider experience in their use and

in-terpretation so that eventually a more definitive document could be

pro-duced

Four ways of assessing the corrosion inhibiting efficiency of antifreeze

solutions are decribed in BS 5117 including (1) glassware tests with (a) hot

and (fe) cold immersion conditions, (2) a recirculating rig test, (3) a static

engine test, and (4) a field test

The glassware tests provide a simple procedure using readily available

ap-paratus which is intended as a rapid preliminary sorting test to determine

whether a solution merits further testing by the more elaborate (and probably

more expensive) methods The standard emphasizes that the results obtained

should not be regarded in themselves as an adequate indication of how a

solution will behave in an engine In the hot immersion test, specimens of

metals and alloys typical of those found in the cooling system of an engine,

are fully immersed in the test solution at 90°C (194°F) with aeration for 14

days Corrosion is evaluated from the mass losses at the end of the test The

test vessel consists of a 1000-ml glass beaker with the assembly of specimens,

mounted on a central insulated rod, resting on the base The specimens

in-clude steel, copper, brass, solder, cast aluminum and cast iron and

corres-pond to the test bundle specified in ASTM Corrosion Test for Engine

Coolants in Glassware (D 1384-70) The same equipment is also used for the

cold immersion test in which the system is left at room temperature for 14

days after an initial test period at 90°C (194°F); that is, to reproduce the

con-ditions in which an engine remains idle after an initial filling and circulation

at the working temperature of the coolant The apparatus is represented

diagrammatically in Fig 1

The recirculating rig test provides a procedure for the laboratory

evalua-tion of the potential corrosivity of an antifreeze soluevalua-tion in condievalua-tions more

closely simulating those in the cooling system of an engine An obvious

prob-lem exists in the design of such a rig because of the wide variations in engine

construction and operating conditions and the need to produce only one type

Gas

distribution tube

Key to test specimens

Brass legs

Steel Brass washers wasKiers

FIG 1—Diagram of the glassware test apparatus

of rig The test was therefore designed to reproduce the service conditions

thought to be the most demanding of an antifreeze while using areas of metal

surfaces exposed to the coolant that were realistic with respect to the ratios of

those found in practice

The principle of the method is to circulate a hot aerated solution of

an-tifreeze in water in a rig constructed from materials used in cooling systems

(but not actual components of an engine) and which contains assemblies of

weighed specimens The general arrangement of the apparatus is shown in

Fig 2 It consists essentially of a reservoir, a head tank, a pump and

connect-ing pipework The solution is heated by two electric cartridge heaters

in-serted into sheaths and supported in the reservoir These sheaths are made of

an aluminium alloy typical of cylinder head materials and thus also act in

assessing the behavior of the coolant in heat transfer conditions In the latest

design of the rig these sheaths have longitudinal holes drilled to

accom-modate thermocouples so that the temperature of the metal may be

monitored during the test Cooling of the circulating solution is by a cold

finger, usually of brass, inserted into the head tank The reservoir and head

tank are made from cast iron and brass respectively; the pipework is of

heat-Pressure release valve

Brass reduction and bend

Air flow

- meter

Air inlet pipe Brass lid

Copper cold finger

Brass t>ead tank

Hose connection

Thermostat control gauge

Hose specimen holder Plug for heater Brass lid & exit pipe Cartridge heater location Pump Aluminium sheaths Cost iron reservoir Rug for pump Hose connection Hose elbow Visual flow indicator Drain cock

Time clocks

FIG 2—Recirculating rig test—diagram showing test rig layout

and pressure-resistant rubber or plastic and the circuit includes a flowmeter

and provision for aeration of the solution Two assemblies of specimens,

prepared as for the glassware test, are mounted in the reservoir and four

assemblies in the section of pipework above the reservoir (Fig 3) The system

volume is 2750 + 50 ml

In all tests covered by BS 5117 and DD 53 (discussed later in this section),

the antifreeze is diluted 1:3 with a standard test water containing 0.148-g

sodium sulfate (Na2S04), 0.165-g sodium chloride (NaCl), and 0.138-g

sodium hydrogen carbonate (NaHC03) per litre of distilled water

Tests are conducted with aeration at 10 ± 5 ml air/min and a system gage

pressure of 70 ± 7 kPa The coolant temperature is set at 95 ± 2°C (203 ±

3.6°F) with a flow of cooling water through the cold finger adjusted to give an

outlet temperature of 35°C (95°F) The total test period is six weeks

in-cluding planned shut-down periods when the system returns to room

temperature conditions A schedule for these is specified and is typically 9 to

10 h at 95°C (203°F) followed by 5 h at room temperature, the system being