alkalis in concrete

8/19/2019 Alkalis in Concrete

http://slidepdf.com/reader/full/alkalis-in-concrete 1/16

I

Ser

TIS

National Research Conseil national

N2 d \t Council Canada

no 236

de recherches Canada

c.

BLDG

EVALU ATION OF TEST METHODS FOR ALKALI-AGGREGATE

REACTIV ITY

y P.E. Grattan-Bellew

Reprinted from

Danish Concrete Association

6th. International Conference

Alkalis in Concrete: Research and Practice

Technical University of Denmark

Copenhagen, 22-25 June 1 98 3

Proceedings, p. 303

314

N R - I S T I 1

BLDG.

RES.

L I B R A R Y

84

11- O

6

B I B L ~ O T H Z Q U E

Reth.

Bttim

N R I I S T

DBR Paper No. 1236

Division of Building Research

Price $1

OO

OTTAWA NRCC 23749



L es msthodes d e s s a i d e l a r h c t i o n a l c a l i - g r a n u l a t comprennent

c e l l e s q u i p e rm e t t en t d e d e te rm in er l a r e a ct iv it e p o t e n t i e l l e

d u n g r an u l a r ( l e s s a i chimiqu e e t l a m6t hod e p et r o g r aph iq u e

propos ee pour d e terminer l a n g l e d ex t i nc t i on on du la to i r e du

q u a rt z ) e t c e l l e s q u i v i s e n t m esurer l e x p a n s i m d un

gr an u l a t dans l e b e ton (msthodes du p r i sme de b e ton, du bar re au

d e m o r ti e r e t d e l g p r o u v e t t e c y l in d r iq u e ). P l u s i e u r s

mo di f i ca t i on s on t e t5 apport ees l a m ethode chimique,

ASTM

C289 p ou r p r e nd r e en c o n si d e r a ti m l e s v a r i a t i o n s

r e g i on a le s d e s g r a n u l a t s . l

r i n r i n l A U e u e s

d e

mesu re d e l ex p a n s i f en

oeuvre. Pour t e n t e r Ye

s

d 1 accS l e r a t i o n o n t

s

d a n s d e s s o l u t i o r

1 au toc lavage . re

in terp r e t ees avec ,

_

6lev ees,

l

s e p r ~

n e s e p r o d u i r a n

atmosphgrique

.



DANISH

CONCRETE ASSOCIATION

DBF

LK LIS

N CQMCRlETlE

Research

and

Practice

Paper reprinted from the

PROCEEDINGS

TECHNICAL UNIVERSITY

of Denmark

COPENHAGEN

22. 25.

J U N E 983

Editors: G M ldorn Steen Rostam



EVALUATION OF TEST METHODS FOR ALKALI-AGGREGATE REACTIVITY

P.E. Grattan-Bellew

Di vis ion of Bui ldin g Research, Nati ona l Research Council Canada

Ottawa, Canada K1A OR6

ABSTRACT

Tes t methods f o r a l ka li - agg r egat e r e ac t i v i t y can be d i v ided i n t o t hose t ha t

de termine the po te nt i a l r e ac t iv i ty of an aggregate the chemica l t e s t and th e

proposed pe t rogr aphic method f o r de termining th e undula tory ex t in c t io n angle

of qua r tz) and those t ha t purp or t t o measure the expansi9i ty of an aggrega te

i n conc re t e conc r e t e p ri sm ,

mortar bar , and rock cy lin de r methods) .

Severa l

mo di fi ca tio ns have been developed f o r th e chemic al method, ASTM C289, t o tak e

i n t o accoun t r eg i on a l va r i a t i on s i n aggr ega te s . The main d i sadvan tage of t h e

methods measuring expansion

i s

t h a t they t ake a long t i m e

To t r y t o overcome

t h i s problem, va rio us methods of a cc el er at io n have been proposed, vary ing from

immersion of t h e s am ples i n s a l t s o l u t i o n a t a n e le v a t e d te mp er at ur e t o

auto clav ing. Autoclaving methods must be rega rded w it h some ca uti on , however,

because a t e lev ate d temperatures and pressures hydrothermal rea ct i on s occur

t h a t do no t t ake p l ace a t a tmospher ic pr e ssu re .

KEY WORDS:

Evalua t ion , Tes t s , Alka l i , Aggregate .

1.

INTRODUCTION

The i de a l method of de te r mi ni ng t h e po t e n t i a l a l k a l i expans i v i t y of an

aggregate i n concre te should be appl i ca ble t o a l l rock types , should be r ap id ,

t ak ing no more tha n a few days t o complete, and must g ive r e l i a b le r e su l t s .

The quest f o r t h i s t e s t has kept a number of rese arch ers busy fo r many yea rs ,

bu t l i k e t h a t of t h e a l chemi s t s be f o re them i t w i l l not succeed.

M v e r s i t y i n

th e composi tion, g ra in s iz e , and poros i ty, and the presence o r absence of

secondary cementing minerals i n rocks pre di ca te s aga ins t t h e development of a

univer s a l ly app l i cable , r ap id t e s t method. Some t e s t s , f o r example th e

conc re t e p ri sm method, a r e ap p l i cab l e t o a l l r ock t ypes bu t may t a ke s o l ong

t h a t t h e s t r u c t u r e w i l l be b u i l t b e fo r e t h e t e s t i s complete.

Other

t e s t s

l i k e t h e mor ta r ba r and t he chemi ca l methods, a r e s u i t a b l e f o r agg rega t es

e x h i b i t i n g c l a s s i c a l a l k a l i - s i l i c a r e a c t i v i t y b u t n ot f o r r e a c t i v e c ar b on a te

aggregates. A s a re s u l t , a number of d i f fe re n t t e s t methods have been, and

continue t o be, developed. This review

w i l l

examine t he e f fe ct iv en es s and

l im i ta t io ns of ex i s t in g and proposed t e s t methods.

2. CHEMICAL TEST

The ch emic al method, ASTM

C 89 111, was developed a s a ra pi d means of

d i f f e re n t i a t in g between pot ent i a l l y r eac t ive and non-react ive s i l i ca-be ar ing

aggr ega t es . The c r i t e r i a f o r eva l ua t ion wer e devel oped from t h e r e su l t s of

p a ra l l e l s e t s of mor tar bar and chemical t e s t s on many aggregates f rom th e

Uni ted St at es 121. The rocks te st ed included some con tain ing opal , ch er t o r

chalcedony a s the re ac ti ve component and othe rs i n which vo lc an ic gla ss was

the reactive component.

As

t h e two main re ac ti ve components, op al and

vo l can i c g l a s s , cons i s t of c r yp t o c r ys t a l l i ne S i 02 and an a l umi nos i l i ca t e

g l a s s , r e spec t i ve l y ,

i t i s

s u r p r i s i n g t h a t one s e t of c r i t e r i a were fo und t o

be appropr ia t e f o r eva lua t ion of both types .

The problem of obta in ing meaningfu l va lues fo r th e d i sso lved s i l i c a Sc) and

t he r educ t ion i n a l ka l i n i t y Rc) a r e compli ca ted i f ca r bona t e s o r su l pha t e s ,

e .g., gypsum 131, a r e prese nt i n th e aggregate because they r ea ct wi th NaOH t o



reduce

i t s

a l k a l i n i t y .

The values of Sc and Rc ar e a ls o very susc ept i b le t o

operator inf luence.

For example, i f th e sample i s not p roper ly s iz ed and a

significant amount of -150

u

m a te r i a l

i s

i nc lu de d i n t h e

t e s t

the value of

Sc w i l l be markedly increased. Grinding quar tz t o s iz e s

less

than about

100 u g r e a t l y i n c r e a s e s i t s s o lu b i l i t y F ig u re 1 ).

The phenolphthale in ind ic ato r does not g iv e a sharp end poin t and th e value

of Rc determined is thu s, t o some ex te nt , dependent on th e judgement of th e

operator. Soaie of th e problem inh eren t i n th e chemical

tes t

were discussed

by Dent Glasser and Kataoka 141.

Their main conclusion i s th a t the va lue of

Rc, as currently determined,

i s

ra th e r meaning less s inc e

i t

inc ludes

r ed uc ti on s i n bo th OH- and t he c o nc e nt ra t io n of ~ a +n t h e s o lu ti o n .

They

suggested th a t th e method should be a l te re d s o t ha t OH can be determined by a

pH meter and ~ a +y a sp ec tr os co pi c method.

The q u e st i o n a r i s e s a s t o t h e need t o d e t e r d n e t h e r ed uc ed a lk a l i n i t y , Rc;

i s de termina t ion of Sc a lone suf f ic ie n t f o r p red ic t i ng the po ten t ia l

re ac t i v i ty of an aggrega te? To inv es t iga te th i s ,

t e s t

r e s u l t s f o r a s e r i e s of

quar tz sands containing varying amounts of opal 121 were pl ot te d i n Figure 2;

i t i s e v ide n t t h a t e x pa ns io n of m o rt ar b a r s c o r r e l a t e s b e t t e r w i th t h e f a c to r

Rc/Sc than with Sc a lone, ind icat ing th at f or th ese aggregates

i t

i s b e t te r t o

determine both Rc and Sc.

I t

h a s been e s t a b l i s h e d t h a t t h e c r i t e r i a give n i n C289 a r e n o t a d e qu a te f o r

the evaluat ion of Icel and ic sands conta ining volcanic gl as s or of Danish f l i n t

g rave l s , f o r which l o w values of th e amount of d iss olved s i l i c a

are

i n d i c a t i v e

of po ten t ia l r eac t iv i ty . New c r i t e r i a were developed fo r th e i r eva lua t ion a s

fol low s pe rs on al comnunication, Hakon Olaf sson)

Icelandic sands:

non -re act ive Sc <I 00 nsnol/L

po te nt ia ll y re ac ti ve Sc 100 t o 200 mmol/L

r ea c t i v e Sc s200 mmol/L

Danish gravels:

non r e a c t i v e Sc <50 mmol/L

pot en t ia l ly reac t iv e Sc 50 t o 200 mmol /~

r e a c t i v e Sc >200 mmol/L

I n c o n t ra s t ,

i n Br i t a i n t he Thames Valley f l i n t g rav e l s y ie lded va lues of Sc

i n exce ss of 100 mmol/L y e t caused no problems i n con cr et e 131. I t appears

t h a t s e p a ra t e s e t s of

c r i t e r i a need t o be developed on a r eg io nal ba si s, even

f o r evalua tion of what ar e nominally t he same rock types.

I n t he late-ex pansive aggr egat es of th e Malmesbury Formation i n th e

Cape Peninsula of South Africa, i n which s tr ai ne d quartz i s t h e r e a c t i v e

component,

t h e c r i t e r i a s p e c i f i ed i n C289 a r e e f f e c t i v e f o r d i f f e r e n t i a t i n g

rea cti ve and non-reactive aggregates;

but i n th e Tygerberg Formation from th e

same reg ion lower val ues of Sc were obtaine d from re ac ti ve agg regat es.

The

resu l t s o f the chemical te s t on thes e aggregates, reported by Brandt,

Oberhols ter and Westra 151, a r e r ep lo t te d i n Figure 3.

It

i s e v id e nt t h a t t h e

dividing l in e between r eac t ive and non-reactive aggregates must l i e a t a value

of Sc between 35 and 50 nnnol/L.

I n Se r b ia t h e che mica l method, ASTM G289, used by Mitr ov iE and

~ u r i E 6 1 i n d i c a t e d t h a t t h e a g gr eg a te s a r e p o t e n t i a l l y r e a c t i v e a s d id t h e

mortar bar tes t ) , a l though no problems have been observed i n co ncret e.

There

a p p e a r s t o be some doubt concerning th e va li d it y of both methods f o r

eva lua t i ng these pa r t i c u l a r aggregates .

I n A u s t r a l ia , t h e c r i t e r i a s p e c i f i e d i n C289 a r e e v id e n t ly s a t i s f a c t o r y ;

they have been adopted s an Austra l ian s tandard, S 1141 Section 39, 1954.



2 1 Discussion

T h e c h e d c a l

t e s t i s

the only rapid ,

currently

av ai lab le method, th at has

been thoroughly evaluated. With some m d i f i r a t i o n of th e c r i t e r i a used t o

def ine re ac t i ve aggrega tes , and i n co-n juac tion wi t h a pe t rograph ic

examination,

it

can

e

u se d t o e v a lu a t e t h e p o t e n t i a l r e a c t i v i t y o f a w id e

range of s i l ic eo us aggregates . Improvements might be made i n th e an al yt ic a l

p rocedures so tha t the reduc t ions i n OH- and i n ~ a re determined separate ly .

There

is

some

e v id e nc e t h a t d e te rm ina ti on o f d i s s o lv e d s i l i c a a lo n e i s

sufficient t o d i f fe re n t i a te between re ac t iv e and non-react ive aggregates .

Owing t o t h e w id e d iv e r s i t y i n the conposit lo-n gr a i n s i z e and perm eabi l i ty of

rocks,

it i s

most u n l i k e ly t h a t u n iv e r s a l l y a p p l i c a b l e c r i t e r i a c an be

deve loped t o d i f f e r en t i a t e between rea c t iv e and non- reac tive aggrega tes ;

ins tead , c r i t e r i a app l icab le t o spe c i f i ed rock types should be deve loped.

2.2 Modified ch em ical

test

A

d i f i e d c h e d c a l t e s t was d ev elop ed f o r e v a lu a t i n g t h e p o t e n t i a l

r e a c t i v i t y o f a g g re g a te s i n sh Sch les wig -lb lst ein reg ion of Germany. The

aggregate

i s

d iv id e d i n to two s i z e f r a c t i o n s : 1-2 mm and 2-4 mm These a r e

dige sted f n NaOH s ol ut io n f o r 60 d n a t BO C.

The

samples are

washed, dried,

and weigh t lo s s de te rmined ; t h i s represe n ts t he amount of s i l i c a d i s so lved ou t

by N H and gives,

i n p r i n c i p l e , a measure of t h e p o t e n t i a l r e a c t i v i t y of t h e

aggregate.

The l im i t a t i o n s of t h i s t e s t have been disc uss ed by Jensen,

Chatterji, Christensen, Thaulow and Gudmndsson 171. t would be applicable

o n ly t o a g g r eg a t es c o n t ain ing r e a c t i v e

s i l ica ,

i.e., opal,

cher t or chalcedapy

and possi bly a ls o t o aggregate composed of volc anic g lass .

3. MORTAR B R METHOD

The mo rt ar b a r t e s t , ASTM C227-8

/ 8 /

i s

t h e most widely used method of

eva lua t ing the po t en t ia l a lk a l i - rea c t iv i ty of aggrega tes .

A

number of

v a r i a t i o n s i n t h e m ethod o f e v a l ua t i ng t h e t e s t r e s u l t s a r e i n u s e by

di ff er en t organizat ion s . Despi te the a lmost univ ersa l acceptance of th e

mortar bar method,

i t

has several shortcomings: It u s u al l y t a k e s from s i x

months t o a ye ar t o complete,

except wi th some opal-bearing aggregates th at

can be evaluated i n about th re e months.

Such long lead

times

f requent ly do

not ex i s t be fore cons t ruc t ion

i s

s t a r te d , even i n major s t ru c t u r es fo r which

th e d u r a b i l i t y of t h e c o n c r et e i s of c r i t i c a l importance.

The evaluat ion of th e re ac t i v i ty of concrete aggregate by t h e mortar ba r

method assumes t ha t expansion of concrete i n the f i e l d cor re l a t es wi th

expansion of t e s t mortar bars . Although t h i s holds t r ue fo r some aggregates ,

th ere ar e exceptions. The carbonate aggregate from Kingston, Ontario , f o r

example,

shows much g re at er expansion i n co ncre te prisms than i n mortar bars ,

a s do late-expansive quartz-bearing aggre gates such a s Malmesbury aggregat e

from th e Cape Penin sula i n South Af ri ca and a greywacke from Nova S co ti a,

Canada 191 ( s ee Figu re 4, 73-50).

3.1 In te rp re ta ti on of observed expansion

Recommended c r i t e r i a fo r eva lua t ion of th e re su l t s of th e mor ta r ba r t e s t

a r e g iv en i n ASTM

C33-82,

Appendix XI.1.3 / lo /

I t

must be s t r ess ed t ha t

th es e a r e recommendations, not a mandatory pa r t of t h e stand ard. Although

t he y a r e f r e q ue n t ly i n t e r p r e t e d a s

limits,

t h i s

i s

i n c o r r e c t .

According t o C33, expansions a re t o be considered excess ive i f they exceed

0.05% a t thr ee months o r 0.10% a t s i x months; the three-month da ta should only

be used, however, when a six-month t e s t i s not possib le . The a l k a l i con ten t

of th e cement should be 0.6%, pr pref era bly 0.8%. With t h e hig h a l k a l i

cements now commonly i n use, an a l k a l i l e v e l of 1.0% o r g r e a t e r would be more



ap pr o pr ia te . The problem wi th t he se recommended maximum pe rm is si b le

expansions i s

th a t they assume a fa s t r a t e of expans ion wi th a sh or t

i n i t i a t i o n p eri od .

slow ly expand ing agg re ga te may show mo rtar ba r

expans ions we l l below t he va lues cons ide red t o be de l e te r io us , accord ing t o

C33, but th e cont inuin g slow expansion

w i l l

e v e n t u a l l y l e a d t o e x c e s si v e

s t r e s s and cracking.

t y p i c a l example of exp ans ion of a morta r b ar made w it h Malmesbury

aggre gate and a cement con tai nin g 0.82 a l k a l i i s shown i n Figure 5 (pers onal

comnunication, R.E. Ob erh ols ter ) . Th is agg reg ate i s known t o ca use

de le t e r i ou s expans ion i n conc re te made wi th a h igh a l k a l i cement.

The problem

of eva l ua t i ng the re su l t s of expansion of morta r ba r s made wi th l a te ex pa ns iv e

agg reg ates h as a l s o been disc uss ed by Kennerly e t a l . 1111 and by Sims 131.

3.2 Corps of En gine ers

The C orps of Eng ine er s c r i t e r i a f o r e va lua t i on of t hz r e s u l t s of t he mor t ar

bar t e s t a re po ssi bly more ap pro pri ate than tho se recommended i n C33.

EM 1110-2-2000 1121 s t a t e s th a t expansion of 0.10 o r more a t any age shou ld

be considered de l e te r io us . An expansion of 0.05 or more a t s i x months may

a l so be cons ide red de le te r io us . When expans ions c los e t o the se

l i m i t s

a r e

recorded, th e tre nd of th e time-expansion curve should be taken in t o account.

T h i s

i s

i n l i n e wi th t h e concept of de te rmining the r a t e of expansion proposed

by Grattan-B ellew 191. Cracking of mortar b ar s i s generally observed when

expansion exceeds about 0.04 ; i f c racki ng i s accepted a s ev idence of

de le t e r i ou s reac t ion , expansions of l e s s than 0.10 must in d i ca t e po te n t ia l

a lka l i -aggrega te re ac t i v i ty . The problem i s t h a t a t p r e s e n t no t e nough

informat ion on the co rr e l a t io n between expansion of mortar bar s and f i e l d

c o n c r e t e e x i s t s

t o s e t r e a l i s t i c s a f e l i m i t s o f ex pa nsio n f o r m or ta r ba rs .

3.3 Canadian Sta nd ard s Ass oci atio n

The Canadia n st an dar d (CAN3.A23.1-M77, Appendix B3.4 113 1) s t a t e s t h a t

expans ions i n excess of 0 .04 a t any age a r e to be cons ide red de le te r io us .

t

a l so note s t ha t t he shape of the expans ion curve i s of ten of importance i n

making a judgement.

The 0.04 l i m i t for expans ion i s i n agreement w ith

f requent o bse rva t ions of c racking when th i s l e v e l of expansion i s exceeded;

the l i m i t may, however, be too re s t r i c t i v e f o r some aggre gates f o r which

expans ion may ju s t exceed t h i s va lue and then le v e l of f .

3.4 Accelera ted Danish mortar bar te s t

Th i s t e s t , de s c r ibe d by J ens en e t a l . 171,

i s

a n a c c e l e r a t e d ve r s ion of t he

mortar bar method C227 and i s use d f o r D an is h f l i n t g r a ve l s.

The mortar bars

a r e made i n th e usu al way and cured i n a fo g room a t 23OC f o r 27 days. They

a r e

then placed i n s a tu ra t ed NaCl s ol ut io n a t 50°C and expansion i s monitored

w i t h time. No

l i m i t s

have been proposed. Th is method

i s

r e po r t e d t o be

s a t i s f a c t o r y f o r t h e a gg r ega te s t e s t e d i n Denmark, bu t t he r e a r e no r e po r t s of

i t s

use elsewhere.

3.5 Chinese autoc laved mortar bar te s t

Th i s

i s

a no the r a c c e l e r a t e d mor ta r ba r t e s t . Mort ar ba r s 1

1

4 cm a r e

prepared and cured f o r one day i n a fog room. Following t h i s , they a r e steam

cured a t 100°C f o r 4 h and t hen immersed i n 10

KOH

s o l u t i o n an d au to c la v ed a t

150°C f o r 6 h. Expansions a r e recorded af t e r each phase of the cu r ing cycle .

l a r ge number o f a l ka l i - s i l i c a r e a c t i ve a gg r e ga t es w ere t e s t e d i n t h i s way by

Ming-shu, Su-fen and Shi-hua 1141, who concluded th a t t h e method i s capab le of

d i f f e r en t i a t in g be tween rea c t i ve and non-reac tive aggregate s .

More tes t ing

s hou ld be c a r r i e d ou t t o e va lua t e

i t s

e f f e c t i ve ne s s f o r a wide r va r i e t y o f

rock types. Hydrothermal t e s t s must, however, be regard ed wi th some ca ut io n



because of the possibility that the observed expansion might be distorted by

hydrothermally induced reactions that would not

occur in concrete under normal

conditions.

3.6 Discussion

No firm limit of expansion can be specified that is applicable to all rock

types. Cracking of mortar bars is frequently observed when expansion exceeds

about 0.04 , and this value might be considered indicative of potentially

deleterious expansion. In some cases, however, this could result in

classification of some aggregates that perform satisfactorily in concrete as

deleterious. It would be desirable to have a time limit for the mortar bar

test; owing, however, to wide divergences in initiation periods and in rates

of expansion it is not possible to establish a universally applicable time

limit. It is possible that regionally applicable time-expansion limits such

as those proposed in C33 may be established for some suites of aggregates.

Accelerated tests have so far not been evaluated for a wide enough range of

aggregates to permit conclusions to be drawn as to their applicability.

4. CONCRETE PRISM TEST

The concrete prism test (CSA A23.2-14A 1151) was developed in Canada for

evaluating the expansivity of reactive dolomitic limestones that do not cause

significant expansion in mortar bars.

The test has also been used success-

fully for late-expansive siliceous aggregates and limestones containing

reactive silica. Similar tests have been used in New Zealand and

South Africa.

The criteria for evaluation of the test results are spe ified

in CSA A23.1.M77, Appendix B3.5. Linear expansions of more than about 0.03

indicate potentially deleterious expansion. Appendix 3.5 concludes that some

aggregates expand beyond the three-month period and that, where possible, the

test should be continued until expansion has virtually ceased. This is sound

advice, but it may often not be practical since with some late-expansive

aggregates it may take up to two years for expansion to taper off. The use

of the rate method of evaluating expansivity of aggregates 191 may help in

evaluating the expansivity of concrete prisms before expansion tapers off.

It is very desirable to carry out a petrographic examination in conjunction

with the concrete prism test so that the type of reactivity to be expected can

be determined; this helps in the correct evaluation of early test results if

it is not possible to wait until the expansion has tapered off.

The concrete prism method may be the most reliable test for all types of

aggregate. In it, the aggregate is tested in the same size range and with the

same cement:aggregate ratio that is used in field concrete.

For some l a t e

expansive siliceous aggregates and for carbonate reactive aggregates such as

those from Kingston, Ontario, the concrete prism test is the only satisfactory

method since these aggregates do not cause significant expansion in mortar

bars.

Expansion can be significantly accelerated by storage of the prisms at

38°C and 100

R

instead of at 23°C as specified. A few preliminary

experiments to try and accelerate expansion by

immersion of the prisms in NaCl

solution, after the method of the Danish accelerated mortar bar test, were not

very successful, but more research is needed before any firm conclusions can

be drawn.

5.

CONCRETE CUBE TEST

This test was developed in Germany 1161,

and cracking is the criterion used

to determine whether a cement-aggregate combination is reactive. There are

two versions of it. In the first, 30-cm cubes are made and expansion is



accelera ted by s tor age i n a f og room a t 40 .

Other exposure condi t ions a r e

al s o used, f o r example, s to rag e on th e roof of the resear ch laboratory .

I n the second version of the t e s t

10-cm cubes a r e st or ed a t 65 RH wi th the

bottom 1 cm immersed i n w ater a t 20°C.

I n some samples con taining re ac t iv e

agg rega te marked cra ckin g was observed ju s t above t he wate rl ine .

I n B ri ta i n , a t th e Cement and Concrete Associat ion, a modified ver sio n of

the 10-cm cube t e s t

i s

being inves t iga ted (D.W. Hobbs, per so na l communica-

ti on ). Cubes a r e immersed i n a water bath a t 23'C and crack formation i s

observed.

A

temperature of 23'C i s reported t o be the optimum f o r so lu bi l i ty

o f s i l i c a .

In South Africa ,

30-cm cubes a r e used a t outdoor exposure s i t e s where they

may be s ubmitted t o var ious moisture and weather exposure cond it ions.

Te sts involving the obse rvat ion of cracking have a major advantage over

tests req uirin g ac ccur ate measurement of leng th change i n th a t no ela bo ra te

appara tus i s required.

This means t h a t th e concr ete cube t e s t could rea di ly

be ca rr ied out by pi t , quarry or ready-mix operators .

The disadvantage i s

t h a t un less rap id ly expanding aggrega tes a r e be ing eva lua ted

i t

would take an

unacceptably long time f or vi s i bl e cracks t o develop. From laboratory

observ at ions by t he autho r

it

is known th a t a con si de ra bl e amount of ex pans ion

can frequent ly be observed b efore cracks become v is ib le t o th e unaided eye.

6. ROCK CYLINDER METHOD

The rock c y li nde r method (ASTM C586-69 1171) was desi gne d t o eva lu a te t h e

pot en t i a l re ac t iv i ty of a lk a l i reac t iv e carbona te aggrega tes , bu t it h a s a l s o

been used wi th vary ing degrees of success t o eva lua te th e r ea c t iv i ty of

a lk al i- si l i ca re ac t i ve aggregates /5, 10, 18, 191. Paragraph 9.3 of C586

s t a t e s th at expansive behaviour of aggregate (carbonate) i n concre te

i s

qua l i t a t ive ly pred ic t ed by the res u l t s of the rock cy l inder t es t .

Grattan-Bellew 1191 found moderate co rr e la ti on between expansion of co nc re te

made wit h s i x samples of Kingston aggr eg ate and expansion of mi nia tur e rock

prisms made wit h th e same aggregates. Much poorer co r re la ti o n was observed

f o r expansion of co nc re te prisms and th at of minia ture rock prisms made with

late-expanding s i l ic e ou s aggregates. Kennerley e t a l . 1111 found t h a t some

prism s of New Zealand greywacke expanded ex ce ss iv el y i n

NaOH

but t h a t t he

same rock caused no expansion i n concrete.

I t

i s concluded t ha t t he rock cy l i nde r t e s t ,

o r va r i a t i on s o f i t can only

be

used to eva lua te the re ac t iv i ty of ce r t a i n rock types.

I t s optimum use

would probably be f o r evalua t ing th e pot en t ia l expansivi ty of indi vid ual hori -

zons i n a quarry containing a rock type f o r which co rr el at io n of expansion of

rock c yl ind er s and th a t of con cre te prisms has previou sly been demonstrated.

6.1 Acce lera ted rock prism test

An acce le r a t ed rock pr ism t e s t f o r eva lua t ion of the po te n t i a l re ac t iv i ty o f

carb onate agg reg ate s has been developed by Kazimir 1201 and adopted a s

Czechoslov akian st an da rd CSM 7 1160. Rock pr ism s 1

1 3

cm a r e prepa red

and au tocl av ed i n NaOH so lu ti on a t 215OC and 2.1 MPa f o r 6 h. Th is t e s t i s

reported t o be su cce ssf ul i n di f fe re nt ia t i ng between expansive and non-

expansive carbonate aggregates.

Some re se ar ch was conducted by t h e au thor on

t he app l i c ab i l i t y of t h i s t e s t t o l a te - expanding s i l i c e ou s aggrega te s , bu t i t

was discontinued a f t e r a felds par- l ike phase was found t o have formed i n the

prisms. Expansions occurred, but as re act i ons were taking place ( th es e do not

occur a t one atmosphere)

i t

seemed most unlikely that the observed expansions

would c o r re la te with expansions of the same prisms under atmospheric

condit ions.

More res ea rc h should probably be done a t vari ous temperatures and

p re s su re s t o see whether an acc eler ated t e s t could be developed f o r la te-

expansive s i l iceous aggregates .

3 8



7. GEL PAT TEST

This test was developed in England as a rapid method of identifying

potentially deleterious minerals such as opal or chert in mortar samples.

It does not give any indication of the expansivity of the aggregate.

As reported by Sims 131, a smoothed, sawn surface of the mortar is prepared

and immersed face down in alkali for three days. Gel forms on the reactive

particles, which can thus be readily identified and counted to yield a measure

of the percentage of reactive component in the aggregate.

A somewhat similar test, CSN 72-1162 1211, is used in Czechoslovakia to

accelerate rim development around prisms of siliceous aggregates embedded in

mortar. It is useful for preliminary screening of potentially alkali-silica

reactive aggregates, but additional tests would be necessary to obtain a

measure of the expansivity of the aggregate in concrete.

8. PETROGRAPHIC EXAMINATION

Petrographic examination, ANSIIASTM C595 1221, has been changed from a

technique for determining the mineralogical composition of a rock to a

semi-quantitative method of determining the potential reactivity of certain

rock types by Dolar-Mantuani s development of a technique for determining the

undulatory extinction angle, UE, and relating it to the expansivity of

aggregate 1231. Following on earlier work 124-261, describing the development

of the technique for measuring strain in the quartz lattice by means of the

undulatory extinction angle, Dolar-Mantuani suggested that aggregates with

undulatory extinction angles of less than 15 deg may be characteristic*

nowreactive aggregates.

Recent unpublished results obtained by the present author show that some

South African aggregates (a quartzite from the Witwatersrand, a greywacke, and

a spotted slate from the Malmesbury Formation in the Cape Peninsula,

all of

which cause excessive expansion in concrete) have UE angles of 36, 36 and

43 deg,

respectively, indicating that there is some correlation between high

UE angle and the reactivity of the aggregates. More research into this method

should be carried out.

It has great potential as a rapid method of evaluating

the potential reactivity of quartz-bearing aggregates, particularly the late-

expansive types common in Precambrian terrains.

It would be useful to conduct

a series of interlaboratory measurements of UE angles on reference samples so

that participants could adjust their measuring techniques to ensure results

consistent with those of other operators. In a limited test carried out in

Ottawa,

a variation of

1

deg was found among measurements made by three

participants on the same quartz grains in a thin section. Such a wide

variation may have resulted because one or more of the participants measured

the extinction range, ER, rather than the undulatory extinction angle, UE, as

defined by DolarMantuani.

8.1

Use of scanning electron microscope

Use of the scanning electron microscope (SEM)

to extend the scope of the

petrographic method by permitting characterization of texture and composition

of fine argillaceous rocks may, in due course, make it possible to distinguish

between expansive and non-expansive horizons in a particular deposit.

Figure 6 shows micrographs of two rocks from the Pittsbury quarry near

Kingston, Ontario. Figure 6a shows the highly expansive dolomitic limestone

characterized by isolated dolomite rhombs in a matrix of calcite; Figure 6b

shows a dolostone, which is non-expansive. This sample is composed almost

entirely of a more coarse grained dolomite. Electron probe analysis showed

that the Ca/Mg count rate ratio was different in the two samples

(6a: Ca/Mg 3.6; 6b: Ca/Mg 3.01 indicating a higher percentage of Ca



i n t h e r e a c t i v e d ol om it e;

t

i s the ref ore , a meta stable protodolomite 1271.

The SEM has one gr ea t advantage over ot he r techniq ues, e .g., pet rog rap hic

microscopy), that i s th e speed wi t h which obse rvat ion and an al ys is can be

done; i n about half an hour a chi p of rock ca n be coated , examined, and

analys ed by an expqrienced ope rato r .

9. CONCLUSION

There i s a d e f i n i t e t rend towards modif ic a t ion of th e s ta nda rd ASTM t e s t s

f o r u se i n c e r t a i n r e g i o n s and f o r s p e c i f i e d t y p es o f a g g r eg a te .

Th i s has

come abou t a s a r e s u l t of inadequac ies in th e s t andard t e s t me thods, e i t h e r

b ec au se t h ey f a i l t o p r e d i c t c o r r e c t l y t h e r e a c t i v i t y of a n a g gr e ga t e o r

because the t e s t tak es to o long t o complete. The t e s t methods have been

modified i n two ways:

expansion has been accce le r a t e d , a s i n the Ch inese

a ut oc la ve d m or ta r ba r t e s t ; t h e c r i t e r i a f o r t h e c he mic al t e s t , a s

recommended, have been modif ied to pre dic t th e re ac t i v i ty of lo ca l aggrega tes

more accu ra te l y , a s i n I ce land and Denmark where s p ec i f i c l i m i t s on th e

d i sso lved s i l i c a have been p roposed.

M od if ie d c r i t e r i a f o r e x i s t i n g t e s t methods may b e l i m i t e d t o s p e c i f i e d

aggrega tes in ce r t a i n r eg ions . For example , t h e p roposed Dan ish c r i t e r i a fo r

e v a l u a t in g t h e r e s u l t s o f t h e c he mi ca l t e s t of f l i n t ag g re g a te s , t h a t

Sc values between 50 and 200 mmol/L a r e ind ic at iv e of po te nt ia l re ac t i v i ty ,

would be qu i t e unsu i t ed f o r th e eva lua t i on of Thames Va l l ey f l i n t s , which a r e

non-expansive i n conc rete de sp i t e having Sc value s i n excess of

100 mmol/L.

There i s s t i l l a gr ea t need fo r bet t er , more rap id methods of d iagnosing

p o t e n t i a l l y r e a c t i v e a gg r eg a te s . F or r e aso n s a l r e a d y o u t l i n e d t h e new t e s t s

w l l p ro ba bl y b e d ev el op ed f o r t h e e v a l u a t i o n of c e r t a i n a g g r e g a te s i n

sp e c i f i c r e g i o n s .

The most promising f o r f ut ur e development a r e th e chemical

method, i n conjunct ion wi th improvements i n techniques f o r pet rograp hic

evalu at ion , and poss ib ly some type of a utocla ve ac ce le ra t i on f o r mor tar or

concre te expans ion t e s t s .

Acknowledgement

The author i s i ndeb ted t o P.J. Lefebvre f o r h i s en thus iasm fo r th e work and

the p rec i s ion wi th which he ca r r i ed ou t th e experimen t s r epor t ed i n t h i s

paper.

Th i s i s a con t r ibu t i on f rom the Div i s ion of Bu i ld ing Research,

Na tion al Research Co uncil Canada, and i s pub l ished wi th th e approva l of th e

Di rec to r of the Divi s ion.

References

/1/

ASTM C289-81, P o t e n t i a l R ea ct iv it y of Aggre gate s Chemic al Method),

Annual Book of ASTM St an da rd s, P a rt 14, C on cre te and Mi ne ral Ag greg ates .

American Soc ie t y f o r Te st i ng and Ma ter ia l s , 1916 Race St . , Phi lad elphia ,

PA 19103, U.S.A., 198 2, 198-205.

/2/

MIELENZ R.C., GREENE,

K.T

nd BENTON,

E J Chemical Test for

Re ac ti vi ty of Aggregates with Cement Alk ali s; Chemical Proces ses i n

Cement-Aggregate Reaction.

Jour na l o f th e American Concre te In s t i tu t e ,

19 3) , 1947, 193-221.

3

EMS

.

The App l i ca t ion and Re l i ab i l i t y o f S tandard Tes t ing P rocedures

fo r Po ten t i a l A lka l i Reac t iv i ty . P roc . , 5 th In t e rn a t i on a l Confe rence on

Alk ali-A ggre gate Re ac tio n, Cape Town, S252/ 13, 1981, 12 p.

/4/ DENT GLASSER, L.S. and KATAOKA,

N .

Some Obs erva tion s on th e Rapid

Chemical Te st f o r Pot en t i a l ly R eact iv e Aggregate .

Cement and Concrete

Research, 11 1981, 191-196.



/5/ BRANDT, M.P., OBERHOLSTER, R.E. and WESTRA, W.B., A Contribution to the

Determination of the Potential Alkali Reactivity of Tygerberg Formation

Aggregates.

Proc., 5th International Conference on Alkali-Aggregate

Reaction, Cape Town, S,252/11, 1981, 6 p.

I61

MITROVIC, N. and DUCIC, V., Prilog, Zbornik, 1 MK-a, 10, 1981, 195-204.

/7/

JENSEN, A.D., CHATTERJI, S., CHRISTENSEN, P., THAULOW, N. and

GUDMUNDSSON, H., Studies of Alkali-Silica Reaction

Part 1, A Comparison

of Two Accelerated Methods. Cement and Concrete Research,

12

1982,

641-647.

I81

ASTM C227-81, Potential Alkali Reactivity of Cement-Aggregate

Combinations (Mortar Bar Method).

Annual Book of ASTM Standards,

Part 14.

American Society for Testing and Materials, 1916 Race Street,

Philadelphia, PA 19103, U.S.A., 153-158.

/9/

GRATTAN-BELLEW, P.E., A Review of Test Methods for Alkali Expansivity of

Concrete Aggregates. Proc., 5th International Conference on

Alkali-Aggregate Reaction, Cape Town, S25219, 1981, 12 p.

1101 ASTM C33-82, Appendix XI.

Methods for Evaluating Potential Reactivity of

an Aggregate (Para. 1.2, Method C289). Annual Book of ASTM Standards,

Part 14.

American Society for Testing and Materials, 1916 Race Street,

Philadelphia, PA 19103, U.S.A., 13-22.

1111 KENNERLEY, R.A., ST. JOHN, D.A. and SMITH, L.M., A Review of Thirty Years

of Investigation of the Alkali-Aggregate Reaction in New Zealand.

Proc., 5th International Conference on Alkali-Aggregate Reaction,

Cape Town, S252112, 1981, 9 p.

1121 ENGINEER MANUAL EM 1110-2-2000, Engineering and Design Standard Practice

for Concrete, (Appendix B, 3C).

Department of the Army, Office of the

Chief of Engineers, Washington, D.C. 20314, 1971.

1131 CSA CAN3-A23.1-M77, Appendix B3, Tests for Alkali-Aggregate Reactivity

(3.4, ASTM Standard C227, Potential Alkali Reactivity of Cement-Aggregate

Combinations (mortar bar method). 3.5, CSA Test Method A23.2-14A,

Alkali-Aggregate Reaction, National Standard of Canada)

CAN3 AZ3.2-7 7,

Concrete Materials and Methods of Concrete Construction, Methods of Test

for Concrete.

Canadian Standards Association, Rexdale Boulevard,

Rexdale, Ontario, Canada, 1977, 122-123.

1141 MING-SHU, Tang, SU-FEN, Han, and SHI-HUA, Zhen, A Rapid Method for

Identification of Alkali Reactivity of Aggregate. Cement and Concrete

Research,

2

983, 417-422.

1151 CSA A23.2-14A, Alkali-Aggregate Reaction (Concrete Prism Test).

National

Standards of Canada, CAN3 A23.1-M77, CAN3 A23.2-M77, Concrete Materials

and Methods of Concrete Construction. Methods of Test for Concrete.

Canadian Standards Association, Rexdale Boulevard, Rexdale, Ontario,

Canada, 1977, p. 183-185.

1161 BONZEL, Von J. and DAHMS,

J.

Alkalireaktion im Beton, Herstellung

Verwendung 23, 11, S4951500 und H.12, S5971554, 1973.

1171 ASTM C586-69, Standard Test Method for Potential Alkali Reactivity of

Carbonate Rocks for Concrete Aggregates (Rock Cylinder Method). Annual

Book of ASTM Standards, Part 14, Concrete and Mineral Aggregates.

American Society for Testing and Materials,

1916 Race Street,

Philadelphia, PA 19103, U.S.A., 1982, 357-360.

1181 DUNCAN, M.A.G., SWENSON, E.G., GILLOTT, J.E. and FORAN, M.R., Alkali-

Aggregate Reaction in Nova Scotia, Part I, Summary of a 5-Year Study.

Cement and Concrete Research,

2 1973, 55-70.

I191 GRATTAN-BELLEW, P.E., Evaluation of Miniature Rock Prism Test for

Determining the Potential Alkali-Expansivity of Aggregates. Cement and

Concrete Research,11 981, 699-711.

1201 KAZIMIR,

J.,

MoZnosE Skratenia Sk613ky Rozpinania Dolomitovy ch ~ornl n

Alkaliami, Staveb. Cas. 25, c s. Veda Bratislava, 1977, 413-432.



1211 CSN 72 1162, Stanovenie Odolnostri Prirodnc ho ~tavebnkho 9me6a Proti

~ilicifikich.

MDT 691.2.620.1 679.682.1, Ceskoslovenskti Stgtna Norma

Schvhle~: 4.10. 1981.

1221 ANSIIASTM 295-79, Standard Reconmended Practice for Petrographic

Examination of Aggregates for Concrete. Annual Book of ASTM Standards,

Part 14, Concrete and Mineral Aggregates. American Society for Testing

and Materials, 1916 Race Street, Philadelphia, PA 19103, U.S.A., 1982

218-229.

1231 DOLAR-MANTUANI, L.M.M., Undulatory Extinction in Quartz Used for

Identifying Potentially Alkali-Reactive Rocks. Proc., 5th International

Conference on Alkali-Aggregate Reaction, Cape Town, S252136, 1981, 6 p.

1241 DEHILLS, N.S. and CORVALAN, J., Undulatory Extinction in Quartz Grains of

Some Chilean Granitic Rocks of Different Ages.

Geological Society of

America, Bulletin 75, 1964, 363-366.

1251 DOLAR-MANTUANI, L.M.M.. Petrographic Aspects on Siliceous Alkali-Reactive

Aggregates. P;OC., ~ ~ ~ ~ o s i u m o n ~ ~ l k a l i ~ ~ ~ ~ r e ~ a t eeaction, Reykjavik,

1975. p 87-100.

1261 M A ~ R ;

.,

Presentation at Symposium on Alkali-Aggregate Reactivity.

Annual Meeting of Transportation Research Board, Washington, D.C., 1976.

/27/ SWENSON, E .G. and GILLOTT, J. E Alkali-Carbonate Rock React ion. Highway

Research Record, No. 45, 1964, 21-40.



Fi gur e

1

E f f e c t of g r a i n s i z e o f q u a r t z o th a o u a t af d i s s o l v e d s i l i c a

det erm ine d by t h e chem ica l met hod , ASTM C289

1 0 0 0

5 0 0

300

-

aoa

0

4 8 12 16 20

24

O P A L

I N

Q U A R TZ S A N D

l l . l L , l ,

b

1 -

\

\

-

Figure 2

P lo t showing va ri a t i o n i n Sc and Rc/Sc determined by t h e chemical

t e s t ,

STM

C289, and leng th change of mor tar ba rs of qu ar tz sand

con tain ing var io us amounts of op al

6 0

\

-

3 0

1 5 -

-

-

5 0 m " 1 ' t " "0 0 4 0 0 6 0 0 8 0 0 1 0 0 0

MEAN G R l N

S 1 Z E ,LL m

2 5 0 1 1 1 1 1 1 1 1 1

2 0 0

-

1 5 0

-

INNOCUOUS

1 0 0

-

t a

DELETERIOUS

D I S S O L V E D S I L I C A

S c

m m o l l l

Figure

Resul t s of chemica l t e s t , STM C289, f o r s o w re ac t iv e and non-

re ac t i ve South Afr ican aggregates . Non-react ive: No no r i te ) and

WS quar tz sand)

3 3



T h i s p ap er w h i le b ein g d i s t r i b u t e d i n

re p r i n t fo rm by t h e D i v i s i on of B u i l d i ng

Resea rch remai ns t h e copyr i gh t of t h e

o r i g i n a l p u b l i s h e r . t should no t be

r ep ro du ce d i n whole o r i n p a r t w i th o u t t h e

pe rmi ss ion of t h e pub l i she r .

l i s t of a l l p u b l i c a t i o n s a v a i l a b l e from

t h e D i v i s i on may be ob t a i ned by wr i t i ng t o

t h e P u b l i c a t i o n s S e c t i o n D i v i s i o n of

B u i l d i n g R e s e a r c h N a t i o n a l R e s e a r c h

C o u n c i l o f C a na d a O t ta w a O n t a r i o

KI 0R6.

alkalis in concrete

Documents