The shrinkage of hardening cement paste and mortar

(1)

The shrinkage of hardening cement paste and mortar

Citation for published version (APA):

Haas, de, G. D., Kreijger, P. C., Niël, E. M. M. G., Slagter, J. C., Stein, H. N., Theissing, E. M., & Wallendael, van, M. (1975). The shrinkage of hardening cement paste and mortar. Cement and Concrete Research, 5(4), 295-319. https://doi.org/10.1016/0008-8846(75)90087-3

DOI:

10.1016/0008-8846(75)90087-3

Document status and date: Published: 01/01/1975

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at: openaccess@tue.nl

providing details and we will investigate your claim.

(2)

CEMENT and CONCRETE RESEARCH. Vol. 5, pp. 295-320, 1975. Pergamon Press, Inc. Printed in the United States.

THE SHRINKAGE OF HARDENING CEMENT PASTE AND MORTAR

G.D. de Haas, P.C. Kreijger, E.M.M.G. Ni~I, J.C. Stagier, H.N. Stein, E.M. Theissing, M. van Wailendael ~

(Rec'd. Sept. 26, 1973; in final form April 9, 1975) (Refereed)

ABSTRACT

~'his paper i s an a b s t r a c t from the r e p o r t of the commission BIO: "The i n f l u e n c e of the shrinkage o f cement on the s h r i n k - age of c o n c r e t e " , of the N e t h e r l a n d s Committee f o r Concrete Research.

Measurements of puise v e l o c i t y , volume shrinkage and heat of h y d r a t i o n on hardening p o r t l a n d cement support the idea t h a t the f o r m a t i o n of e t t r i n g i t e i s an i m p o r t a n t l i n k i n the mechanism of shrinkage i n the p l a s t i c stage of cement paste and m o r t a r .

3

Mechanical t e s t s on prisms of 4 x 4 x 16 cm gave some i n f o r m a t i o n about the d i f f e r e n c e i n s e n s i t i v i t y t o surface c o r r o s i o n of d i f f e r e n t types of cement.

Das v o r i i e g e n d e B e r i c h t g i b t ein kurzes U e b e r s i c h t yon den A r b e i t des Ausschusz BIO von dem N i e d e r l ~ n d i s c h e n Verein f u r BetonprUfung, h i n s i c h t l i c h d i e E i n f l u s z des Zementschwindens auf das Schwinden des Betons.

Die PrUfung von P u l s g e s c h w i n d i g k e i t , Volumeschwindung und Hydratationsw~rme an e r s t a r r e n d e Paste von P o r t l a n d Zement w~hrend den e r s t e n d r e i Tagen s t ~ r k t e die Auffassung dasz die Bildung von E t t r i n g i t wBhrend der p i a s t i s c h e n Phase von

Zementpaste und M~rtel sehr w i c h t i g i s t zur Erkl~rung des Mechanismus des Schwindprozesses.

Auch mechanische PrUfungen an e r s t a r r t e n Zement- und M B r t e l - prismen wurden d u r c h g e f U h r t zur EzklBrung des Schwindprozesses.

A f f i l i a t i o n s of the a u t h o r s can be o b t a i n e d from CUR, P.O. Box 61, Zoetermeer, The N e t h e r l a n d s .

(3)

Introduction

The purpose of this a r t i c l e is to introduce to a greater public the findings of the commission BlO of the Netherlands Committee for Concrete Research (CUR) which studied the influence of the shrinkage of cement on the shrinkage of concrete.

In this a r t i c l e only the main points of the study w i l l be dis- cussed.

The shrinkage of cement paste during hardening i s caused by chemical r e a c t i o n s of the c l i n k e r minerals w i t h water during the f i r s t hours. At a l a t e r stage i t i s caused by changes in the water content of the hardened paste.

The purpose of the study was to o b t a i n more knowledge about the changes in volume in the f i r s t stage. To do t h i s the f o l l o w i n g p r o p e r t i e s were measured during the hardening process:

a) Heat e v o l u t i o n r a t e ( w i t h a conduction c a l o r i m e t e r ) b) V e l o c i t y of u l t r a s o n i c pulses

c) Volume shrinkage

Next to these measurements also the strength and the shrinkage of the hardened prisms, made w i t h d i f f e r e n t types of cement, were measured.

Experimental

The heat evolution r a t e c a l c u l a t e d per gram cement was measured using Lerch's method (1) by means of an isothermal conduction c a l o r i m e t e r as described by Stein ( 2 ) . The f i r s t peak was not measured because the mixing pzocedure of 2 minutes s t i r r i n g , 3 minutes r e s t , 2 minutes s t i r z i n g was done outside the c a l o r i - meter.

The v e l o c i t y of pulses i n hardening cement pastes and mortars was measured w i t h a Cawkell UCT 2 apparatus at 40 kHz. The specimen was placed in a t h i n rubber c o n t a i n e r between the pizza

(4)

Vol. 5, No. 4 297 SHRINKAGE, HARDENING, CEMENT PASTE, MORTAR

electric measuring crystals of the instrument. The crystals were mounted on a fixed distance; the specimen filled exactly the space between them.

Changes i n volume of a hardening paste or mortar due t o chemical t r a n s f o r m a t i o n s were measured on a specimen of ~ 600 grams placed i n a w a t e r - t i g h t rubber cover t o exclude t h e i n f l u e n c e of d r y i n g or w a t e r - u p t a k e . This sealed specimen was placed i n a vessel com- p l e t e l y f i l l e d w i t h water of 20 ~ 0.1 °C. The v e s s e l was connected w i t h a tube f i l l e d w i t h water to a beaker on the scale of a M e t t l e r balance. When changes of volume occurred the rubber cover f o l l o w e d these changes and water was sucked out or pressed i n t o the beaker. The changes i n weight of the f i l l e d beaker were r e - corded and from these the changes i n volume of the specimen were c a l c u l a t e d . In these experiments the same batch of p o r t l a n d cement c l a s s A was used.

The chemical and c a l c u l a t e d m i n e r a l o g i c a l composition of t h i s p o r t l a n d cement i s given i n t a b l e 1.

Results and Discussion

Heat Evolution Rate

The influence of increasing amounts of the various clinker minerals, of water, of the addition of sand and of the time of aerating the

TABLE I

Analysis of a Homogeneous Batch of Dutch Portland Cement.

Chemical Analysislin % by wei~ht Hineralo~ical Composition in ~ by weight

L.O.I. I000 °C 1.0 L.O.I. 1000 °C 1.0

SiO 2 A1203 F e203 CaO total MgO Na20 K20 503 T__ot al 22.5 5.5 2.6 64.5 1.2 0.1 0.5 2.1 100.0 C2S C3S C3A C4AF CaO free MgO Na20 K20 Ca50 4 Total Specific Surface cm2/g Specific Weiqht 30.8 44.8 10.2 7.9 1.2 0.1 0.5 3.6 100,1 2900 3111

(5)

cement w i t h a i r c o n t a i n i n g carbon dioxide and water vapour on the flow of heat was s t u d i e d .

The i n f l u e n c e of the s p e c i f i c surface and the amount of a l k a l i of the cement, which are thought to be l a r g e , have not been i n v e s t i g a t e d .

In general three periods of increasing and decreasing heat e v o l u t i o n (peaks) can be observed when the hardening of p o r t l a n d cement i s studied w i t h an isothermal conduction c a l o r i m e t e r . On the decreasing branch of the second peak a f t e r ~ 18 hours a small hump or shoulder can sometimes be seen (Figure 1). The cause of t h i s hump i s not clea~; i t i s f r e q u e n t l y a t t r i b u t e d to both P-C2S and C4AF.

The magnitude of the peaks, and the time at which they occur, in the f i r s t place depends on the m i n e r a l o g i c a l composition and fineness of the cement. In t a b l e 2 the changes in time and height of the second and t h i r d peak are summarized. The exact values of time s h i f t and peak height are not given because they depend too much on the m i n e r a l o g i c a l composition of the c l i n k e r used.

The i n f l u e n c e of C3S added (Figure 2) on the second peak i s r a t h e r c l e a r : a l a r g e r surface of C3S i s exposed to water, so more r e a c t i o n heat can be evolved.

2.5 2 4 6 8 10 20 30 40 50 60 2.0 7 . -~ 1.0

~o

time in ~ u r s FIG. 1

I n f l u e n c e of the w/c r a t i o on the heat development of p o r t l a n d cement class A

(6)

TABLE 2

The I n f l u e n c e on Time of Occurrence and Height of the Peaks i n the Heat Flow Curve by varying the Composition

Increasing Second Peak T h i r d Peak Amounts of

Height Time Height Time

C3A - - higher sooner

C3S higher - - -

C a S O 4 - - l o w e r l a t e r water lower l a t e r higher sooner quartz sand - - lower sooner aerating time lower - higher sooner - = no d i s t i n c t i n f l u e n c e

,o I

0 ~-- O ₁₀ ₂₀~ ₃₀ ₄₀ ₅₀ ₆₀ time in hours FIG. 2 I n f l u e n c e o f t h e a d d i t i o n o f C3S on t h e h e a t d e v e l o p m e n t o f p o r t l a n d cement A d d i t i o n o f more C3A ( s e e F i g u r e 3) g i v e s a l a r g e z s u l p h a t e c o n s u m p t i o n d u r i n g t h e f i z s t peak, t h e g r a i n s become c o v e r e d w i t h e t t r i n g i t e c r y s t a l s t h r o u g h w h i c h t h e z e a c t i o n r a t e becomes low ( a p p z . a t 2 h o u r s i n F i g u r e 3 ) . The e t t r i n g i t e l a y e z i s l e s s t h i c k w i t h more C3A because t h e same amount o f gypsum i s a v a i l a b l e - T h e r e f o r e t h e l a y e r i s more p e r m e a b l e f o r w a t e r .

(7)

C 3.0 E U 2~ - E (.J - - (3 U b 7 . . . . cemee~.~,C3A, water I I I ~ l --.-- cte~rt.S~.C~.water

,i

0 10 20 30 ~0 50 gO timeinhou~ FIGo 3

I n f l u e n c e of C3A a d d i t i o n on the heat development of p o r t l a n d cement

This r e s u l t s , e i t h e r according to the " s c a i i n g o f f " mechanism ( 2 ) , (3) or by chemical d e s t r u c t i o n ( 4 ) , ( 5 ) , in a f a s t e r consumption of sulphate u n t i l the stage i s reached at which a l l sulphate i s consumed. This stage w i l l be reached in s h o r t e r time and a new C3A h y d r a t i o n peak (the t h i r d peak) i s formed sooner.

The i n f l u e n c e of the a d d i t i o n of gypsum can be desribed by the same mechanism w i t h the opposite r e s u l t ; the t h i r d peak i s developed l a t e r .

The i n f l u e n c e of the water cement r a t i o on the heat of h y d r a t i o n of p o r t l a n d cement per gram of cement i s given in Figure 1. The second peak i s developed between 2 and 34 hours a f t e r mixing. The heat developed during t h i s peak i s high when the water cement r a t i o i s low.

When i n t e r p r e t i n g these data, i t should be remembered t h a t p a r t of the heat developed by a h y d r a t i n g paste i s used f o r heating the paste (there must e x i s t a temperature d i f f e r e n c e between paste and constant temperature heat sink i n order to make a

(8)

heat flow possible). From c a l i b r a t i o n experiments, the heat flow recorded can be related to the temperature gradient e x i s t i n g over the heat flow meter; the temperature differences are found to be small, e.g. at the top of the second peak in Figure 1: 0.11 °C in the experiment with w/c r a t i o = 0.2, and 0.08 °C in the experiment with w/c r a t i o = 0.6.

Consequently, only a small f r a c t i o n (about 2 %) of the t o t a l heat developed is employed f o r heating the hydration paste; thus, differences in heat capacities of paste + container caused by varying w/c r a t i o s do not noticeably influence the accuracy of the heat evolution rate measurement. Moreover, the heat capacities do not d i f f e r much since they are determined l a r g e l y by that of t h e c o n t a i n e r s .

Therefore, i t must be concluded from the r e s u l t s mentioned in Figure 1, that the hydration reaction connected with the second heat evolution peak (C3S h y d r a t i o n ) i s f a s t e r in pastes of lower w/c r a t i o . This stands in contrast with data reported by Kantro and Copeland (6) but agrees with experiments reported by Locher ( 7 ) .

A slower C3S hydration with increasing w/c r a t i o can be a t t r i b u t e d to the influence of C3A hydration. When a high w/c r a t i o i s used, more C3A reacts with water in the i n i t i a l stages before the C3A i s covered with e t t r i n g i t e , and more material dissolves, so more aluminium i s taken up by the c a l c i u m - s i l i c a t e hydrates formed on the C35grains, shielding them from water. At low w/c r a t i o the C3A w i l l react less because the layer of e t t r i n g i t e is b u i l t up

sooner around these grains since the water i s super-saturated more q u i c k l y .

With t h i s mechanism, a large influence of mixing conditions and cement composition i s comprehensible, which explains the contra- d i c t o r y findings reported concerning the influence of w/c r a t i o on the hydration rate of C3S.

(9)

The t h i r d peak is higher for higher w/c r a t i o and there i s a small s h i f t to shorter times. This difference can be explained by a difference in morphology of the e t t r i n g i t e formed.

At low w/c r a t i o (<0.30) a c l o s e l y packed e t t r i n g i t e layer i s

formed. The layer formed at higher w/c r a t i o consists of e t t r i n g i t e with a more n e e d l e - l i k e s t r u c t u r e , which i s less dense. This layer i s less able to cut o f f the C3A grains from the supply of water. Thus at high w/c r a t i o the process of water transport through the e t t r i n g i t e layer elapses f a s t e r ; sulphate i s consumed sooner and an unhindered C3A reaction can s t a r t .

The r e s u l t i n g t h i r d peak w i l l be higher at higher w/c r a t i o because more water i s available and less h y d r o s i l i c a t e s have been formed

(as evidenced by the lower second peak), and the s i l i c a t e hydrates formed may be less dense, so the water transport to the grains i s less hindered. This t h i r d peak i s not seen at very low w/c r a t i o s because the low amount of water is completely blocked from the C3A by the higher amount of CSH-gel formed during the second peak. The hump on the second peak at these w/c r a t i o s can be a t t r i b u t e d to the reaction of C^A at a stage where the

o

e t t r i n g i t e layer i s repaired a f t e r bursting through c r y s t a l i i z a t i o n pressure (see also shrinkage measurements).

At the s t a r t of the i n v e s t i g a t i o n s i t was found that measurements on the same cement with the same w/c r a t i o in various tests at d i f f e r e n t times gave d i f f e r e n t heights of the peaks and d i f - ferent times ofoccurrence. This was thought to be caused by the influence of humid a i r during storage. To prove t h i s , a batch of portland cement was exposed to a i r with d i f f e r e n t amounts of C02 and water during d i f f e r e n t periods. The moisture from the a i r reacts with C3A and C35 , the product reacts with C02. This was shown by i n f r a - r e d spectroscopy, where the band of the calcium- sulphate decreases with the band of anhydrous calcium-aluminate

(10)

FIG. 4 ~00

The disappearance of anhydrous 300

20o calcium-aluminate by exposure 100 to C02 and humid a i r ,<~cUon m 193~ 0 3 % ( 0 2 - 9 6 % RJ'L 0 2 5 ? 9 12 days oirL,,atir~ at 19.3 pm, i n d i c a t i n g t h a t calcium-sulpho-aluminates and/or calcium-carbo-aluminates are formed (see Figure 4 ) .

The t o t a l heat development of a paste from aerated cement (Figure 5) i s decreasing w i t h i n c r e a s i n g a e r a t i n g time e s p e c i a l l y at the more humid c o n d i t i o n s ~ 4 4 $ r . h . ) . The second peak decreases and d i s - appears at longer a e r a t i n g times. The t h i r d peak i s formed sooner. These peaks are very sharp.

In view of these e f f e c t s , care was taken t h a t a e r a t i n g of the cement used i n the other i n v e s t i g a t i o n s could not take place. The strength of hardened paste or mortar made wLth aerated cement was very low.

The i n f l u e n c e of quartz sand on the flow of heat of p o r t l a n d cement i s given i n the Figures 6 and 7 f o r w/c r a t i o s of 0.50 and 0.30. The a d d i t i o n of sand i n f l u e n c e s predominantly the t h i r d peak. The peak becomes smaller and occurs sooner when the amount of sand i s increased.

sdi!! iiTd ,,12d

~u A " 0 i O * !

- -

2 1 i

! il

:"

/1

' ' I d a y

I [ l l l l l l l ] l

air composition:O.3"l. CO 2 and 96% re( hum

. . . ~ = ~ . .

~P~..--..

10 20 30 40

time in hours FIG. 5

Heat development of par±land cement a f t e r a e r a t i n g f o r a number of days

(11)

T

P ~ n e m hours FIG. 6

Flow of heat of portland cement with d i f f e r e n t amounts of quartz sand (w/c r a t i o O.50)(cement-ag-

gregate r a t i o by weight 1:O;1:1;1:3 and 1:6)

The first minimum comes later, indicating that the first peak lasts longer, so less C3A is available to produce the third peak. The sulphate consumption must be faster too, which results in a

s h i f t to shorter times for t h i s peak.

I t i s not clear how the sand influences the early C3A reaction. The following explanations are possible:

G

0 I~ O _/, ~

8 12 1G 20 2/. 28 32 36 &O ~, ~1 . 5 2 S6 60 (1~ m

~ i m e ~ hours

FIG. 7

Heat development of portland cement with quartz sand (w/c r a t i o 0.30, c/a r a t i o by weight 1:O;

(12)

a) Sand gives nuclei for c r y s t a l l i z a t i o n b) Sand absorbs the sulphate

ad a) The products of the C3A

reaction(C2AH8,C4AH14

and e t t r i n g i t e ) can be formed in the v i c i n i t y of the sand when the c r y s t a l - l i z a t i o n of these products form f a s t e r by nuclei given by the sand. By t h i s the C3A can react longer and f a s t e r , so sulphate i s consumed more q u i c k l y . The outcome of t h i s w i l l be t h a t the t h i r d peak is developed sooner and w i l l be smaller. As C3A grains react longer, the C3S grains w i l l be coated with more hydro-aluminates too, so the second peak will be retarded.

The fact t h a t C3A reacts longer under the influence of sand can be proved by i n t e g r a t i n g the f i r s t peak in the flow of heat curve which was not possible with the method used.

ad b) Sulphate absorption i s not l i k e l y because experiments on mortars with a cement/aggregate weight r a t i o of 1:1 with d i f f e r e n t gradings (ranging from 16-200 cm2/g) of the sand did not show an influence of the surface of the sand.

V e l o c i t y of Ultrasonic Pulses

Up to the time at which the top of the second peak in the flow of heat curve took place, the pulse v e l o c i t y was higher for higher w/c r a t i o s (Figure 8 and 9).

A f t e r t h i s period the pulse v e l o c i t i e s were higher for lower w/c r a t i o s . When the v e l o c i t y i s thought to be correlated to the compressive strength, the early strength of a paste or mortar made with high w/c r a t i o must be higher than one made with a low w/c r a t i o . During the f i r s t hours the v e l o c i t i e s , however, were very low, even lower than in water, so the v e l o c i t y was at t h a t stage not d i r e c t l y related to the strength.

(13)

U U in

=~

>~

T

t

o 1 2 3 & S 1o time in hours FIG% 8

Pulse v e l o c i t y in pastes with d i f f e r e n t w/c r a t i o 1 2 3 & S 10 time in hours FIO. 9 Pulse v e l o c i t y i n mortar ( c / a r a t i o 1 : 1) w i t h d i f f e r e n t w/c r a t i o from Poland, when working i n the Netherlands, by considering the cement paste as a v i s c o - e l a s t i c m a t e z i a l , using a Maxwell model. According to t h i s , the v e l o c i t y i s depending on the frequency of the pulse (see Figure 10). At low frequencies ( t i l l 500 kHz) there i s a large i n f l u e n c e of the viscous p a r t of the model by which the v e l o c i t i e s are very low during the f i r s t 10 hours. For higher

frequencies (2000-4000 kHz) the v e l o c i t y i s much nearer to the v e l o c i t y of sound in water. At these frequencies the e l a s t i c p a r t i s the most i m p o r t a n t .

A f t e r lO hours the d i f f e r e n c e s between the v e l o c i t i e s are smaller but the v e l o c i t i e s are s t i l l higher f o r higher frequencies.

At the same frequency the curves f o r the higher and lower w/c r a t i o cross each o t h e r , but the time at which they cross i s not the same, more d e t a i l s have to be obtained however.

From t h i s model the conclusion i s drawn t h a t the v e l o c i t y i n paste and mortar during the f i z s t hour i s to a high degree dependant on

(14)

E 3OOO

|

',: ILlll

t WlC m~o 0.27 - - m m W l c ~ m 0.36

,/,-~/r/,~

, o o

///.,4'1

J

" / / / ~ - 1 '

///7/.x7

!

[L

l 2000 / _ J / &OOOkHz -- ~ ~ / i

--t'5-H'Z,,tq. kt" ./1

o °1

1 2 3 10 20 30 40 50 ~ t~e in hours FIG. 10

Pulse v e l o c i t y i n paste as a f u n c t i o n of the time f o r d i f f e r e n t pulse frequencies ( e x p e r i m e n t a l )

i n t e r n a l viscous f o r c e s . These forces are formed by the i n t e r - a c t i o n of p a r t i c l e s t o g e t h e r and p a r t i c l e s w i t h f l u i d . The p a r t i c l e s change in number, Form and mass during the hardening process. The f l u i d changes in v i s c o s i t y by the d i s s o l v i n g c l i n k e r m i n e r a l s . Sn the d i f f e r e n c e i n v e l o c i t y between paste or mortar made w i t h e i t h e r low or high w/c r a t i o con probably be a t t r i b u t e d to the d i f f e r e n c e in morphology of e t t r i n g i t e , the degree of h y d r a t i o n of C3A and the degree of s u p e r s a t u r a t i o n of the water. L a t e r t h i s d i f - ference disappears because more p a r t i c l e s ore formed, during the second peak, g i v i n g a more s o l i d gel w i t h more e l a s t i c p r o p e r t i e s ~ ) .

• ) A pulse can be seen as on e x t e r n a l force working on the paste. At low w/c ratio a paste is called dilatant, the opposite of thixotrope, so an external force induces more internal force, that is o more viscous behaviour, resulting in a low velocity. At high w/c ratio the paste is thixotrope so on external force gives on abolition of the viscous forces resulting in higher velocities during the first hours.

(15)

This is particularly distinct in low w/c ratio pastes: here particle to particle interaction would be higher than in high w/c ratio pastes at the same degree of hydration because of the latters' higher porosity; moreover, according to the large heat evolution rate during the second peak more silicate hydrate is formed in law w/c ratio pastes,

Horrors have always higher pulse velocities than pastes at the same w/c ratio because more "solid material" is found per unit of volume.

Shrinkage

In F i g u r e s 11 and 12 t h e s h r i n k a g e curves combined w i t h t h e p u l s e v e l o c i t y and f l o w o f heat are g i v e n f o r p a s t e and m o r t a r w i t h a

cement : a g g r e g a t e w e i g h t r a t i o 1 : 1. D i f f e r e n t w/c r a t i o s were used.

D u r i n g t h e f i r s t hours u n t i l t h e t i m e at which t h e second peak i s f u l l y developed, t h e s h r i n k a g e i s l a r g e r a t l o w e r w/c r a t i o . L a t e r t h e s h r i n k a g e becomes l o w e r f o r t h e l o w e r w/c r a t i o so t h e curves c r o s s each o t h e r d u r i n g t h e second peak. T h i s i s a s c r i b e d t o t h e f o l l o w i n g mechanism: E U w U

%

T

ZSO0 2400 ~00 1600

i ' " " -

i 1

i

lO I I i ) ' I i f4t,. ,II'~ • ., f / i r' • ~ ! / k_.-k~'"

L

5 FIG. II

:~. o.2o> .-~-"

'

"~,,,,,.

_{F.OH. 0.!}

' ! T

I

i f "

at heat(lO-3 J sec: 1 cj "1) . . . . ~s/-~xJe~Yo/(mlsec.) X O.SO(hic~WtC rata) i

! 0.20 ( low WIC ~io)

, ]"

!

15 = time in hours Combined c u r v e s o f f l o w of h e a t , s h r i n k a g e and p u l s e v e l o c i t y f o r p a s t e s 20

(16)

.S •

'I ,~0 L' !

I

~/'

''

!i

i !

! , ,

i : ' i : i

r ~

~..r

~ i

: ~

.shr~-',,:,:,.:~e pulse ve~:ity (mlsec.) :--- !

~..,.~,,..~1 X O.50(hicjhWlC ratio) ~ -

• 020 ( law WIC raUo) ~--

c

ol-

o1~"

I

! I ,

5 10 15 20

. time in hours

FIG. 12

Combined curves of flow of heat, shrinkage and pulse v e l o c i t y f o r mortar (c/a r a t i o 1:1)

D i r e c t l y a f t e r mLxing C3A d i s s o l v e s , g i v : n g a f a s t shrinkage (Figure 13). A f t e r some time the e t t r i n g i t e c r y s t a l s are formed, C3A dissolves more slowly and the v e l o c i t y of shrinkage decreases.

I s

\\\:

,\\ \ I\

N\I \

N \1

1

~'~s &l's Ih

II I l l

WC-m~g.d~ / \

:

illl I J

[i~i

"

'l~Ii II

~.-.k2"J~ I a, ~ , ~ . . ~ - - r ~ s m ~ w g l t l ~ FIG. 1 3.

(17)

The formation of e t t r i n g i t e compensates for the shrinkage because the volume of e t t r i n g i t e i s only a l i t t l e smaller than the t o t a l volume of the s t a r t i n g materials, while a l l other hydrates have a much smaller volume than the s t a r t i n g materials.

At low w/c r a t i o s the e t t r i n g i t e c r y s t a l s are formed close around the C3A and the C3S reaction s t a r t s e a r l y . A f t e r approximately 4 hours the unstable CSH-gel is formed causing an increase of the v e l o c i t y of shrinkage. When the C3S reaction i s at i t s maximum value the v e l o c i t y of shrinkage decreases while stable CSH-gel i s formed.

When the hump on the second peak occurs the v e l o c i t y of shrinkage gets a small decrease that i s followed by an immediate increase a f t e r t h i s shoulder. This e f f e c t can be explained by the formation of shrinkage r e s i s t i n g e t t r i n g i t e during the stage at which the bursted coating i s sealed around the C3A grain.

At high water cement r a t i o s , shrinkage i s less compared with low w/c r a t i o s (Figure 11). This i s supposed to be caused by the d i f - ferent morphology of the e t t r i n g i t e t h a t has been formed. More C3A has reacted but shrinkage is less so the e t t r i n g i t e formed must have more shrinkage r e s i s t i n g properties than the e t t r i n g i t e formed at low w/c r a t i o . When the c r y s t a l s are longer and are more perpendicular to the grain they can keep the grains wider apart. A f t e r 10 hours, shrinkage becomes higher at high w/c r a t i o s than

at low w/c r a t i o s .

This can be a t t r i b u t e d to differences in porosity of the CSH-gel: at the same degree of hydration the i n t e r a c t i o n between the c a l - c i u m - s i l i c a t e hydrate p a r t i c l e s would be lower at higher w/c r a t i o than at lower w/c r a t i o , r e s u l t i n g in a weaker t o t a l system; in the pastes investigated here t h i s is enhanced by a higher degree of C3S hydration in low w/c r a t i o pastes as evidenced by the high heat evolution rate during the second peak.

When the t h i r d peak in the flow of heat curve appears, the v e l o c i t y of shrinkage becomes faster again. This i s caused by the hydration

(18)

of C3A. Probably the shrinkage i s even boosted up by the t r a n s - formation of e t t r i n g i t e in the less voluminous monosulphate and aluminates. The tobermorite gel is not strong enough at t h i s stage to withstand t h i s shzinkage. Moreover, the new components may form in a less stress state or in low stress locales w i t h i n the sample and therefore w i l l not provide the support which the e t t r i n g i t e has previously provided.

The influence of quartz sand on the shrinkage i s as follows: During the f i r s t hours a mortar with c/a weight r a t i o 1:1 shrinks more than a paste with the same w/c r a t i o . A f t e r ~ 8 hours the

shrinkage of a mortar is less (the curve in Figure 12 i s an extension to the r u l e ) . The shrinkage curves f o r low and high w/c r a t i o s cross each other during the formation of the second peak.

The explanation for t h i s i s the same as t h a t given f o r paste. The higher value of the shrinkage of mortar than of paste during the f i r s t hours can be explained i f i t i s admitted (see flow of heat) by the f a c t t h a t the sand has a large influence on the morphology of the e t t r l n g l t e formed. At low w/c r a t i o the compact coating around the C3A i s not formed, C3A reacts longer causing greater shzinkage. The shrinkage r e t a i n i n g influence of the e t t r i n g i t e at higher w/c

r a t i o i s p a r t l y maintained. Shrinkage i s less in r e l a t i o n to mortar with low w/c r a t i o but there i s more shrinkage compared with paste with the same w/c r a t i o , notwithstanding the larger amount of s o l i d material per u n i t of volume.

As a consequence of the observed influence of the sand on shrinkage, a large dependance of the shrinkage on the s p e c i f i c surface of the sand was expected. But the measurements on mortar 1:1 with d i f f e r e n t gradings of the sand (from 16-200 cm2/g) gave no clear r e l a t i o n between these q u a n t i t i e s . So the explanation f o r the influence of sand needs more research as does the explanation for the influence of sand on heat e v o l u t i o n .

(19)

Conclusions

The measurements of the flow of heat, shrinkage and pulse v e l o c i t y show t h a t during the hardening a clear d i f f e r e n c e exists between pastes or mortars made with elgher high or low water/cement r a t i o . During the second peak in the heat evolution, a change takes place in the course of the shrinkage and pulse v e l o c i t y . High s t a r t i n g values become low ones and low s t a r t i n g values become high. For these effects two explanations may be envisaged. One is as follows:

At low water cement r a t i o s the e t t r i n g i t e formed during the f i r s t peak is thought to consist of a great number of small c r y s t a l s t h a t form a compact l a y e r close around the C3A grain hindering the f u r t h e r reaction of the C3A. This gives high shrinkage and low

pulse v e l o c i t y . High shrinkage because a i i p a r t i c l e s can come closer to each other and a low pulse v e l o c i t y because the viscous forces in t h i s compact system are l a r g e r .

At high water cement r a t i o s the e t t r i n g i t e has the p o s s i b l i i t y to form more n e e d l e - l i k e c r y s t a l s t h a t are able to keep the c l i n k e r p a r t i c l e s wider apart thus r e t a i n i n g shrinkage and giving a more " e l a s t i c " network in wich pulses can proceed f a s t e r .

A l t e r n a t i v e l y , the low pulse v e l o c i t y in the i n i t i a l stages in low water cement r a t i o pastes could be ascribed to formation of hydrate layers in the v i c i n i t y of the contact points between the solid particles.

Then the solid particles do not move independantly under the action of an ultrasonic pulse, and damping may occur especially in the transmission of motion from one particle to its neighbour.

The law pulse velocity in the initial stages in low w/c ratio pastes should then be ascribed to a relative large extent of hydration (as evidenced by the heat evolution rate data), which causes a large amount of hydrate crystals (rather flexible in this state) to sur- round the contact points. In the later stages, the hydrate crystals

(20)

obtain so many mutual contact points t h a t they obtain together a certain e l a s t i c s t i f f n e s s ; then the pulse v e l o c i t y i s increased by a higher extent of hydration. Future research i s needed f o r s e t t l i n g t h i s point.

During the second peak in the flow of heat curve, the CSH-gel i s formed. At low w/c r a t i o a gel with good mechanical properties i s formed, that r e s i s t s the shrinkage b e t t e r than the gel formed at high

w/c

r a t i o . This stronger gel i s more e l a s t i c ; the pulse v e l o c i t y becomes higher.

The CSH-gel formed at high w/c r a t i o s i s also less able to r e s i s t the extra shrinkage t h a t occurs during the t h i r d peak, where C3A reacts and e t t r i n g i t e i s transformedinto other products with a smaller volume, such as monosulphate.

The influence of the w/c r a t i o gives the same trends f o r mortars as f o r pastes. The presence of quartz sand, however, influences the formation of e t t z l n g i t e , t h i s r e s u l t s in deviations in the measured values in heat flow, shrinkage and pulse v e l o c i t y compared with paste.

In the heat flow the t h i r d peak comes sooner and has a lower

i n t e n s i t y , because the C3A reacts longer in the f i r s t hours, perhaps through nucleation by the sand. This f i r s t C3A reaction gives higher shrinkage in the e a r l y hours too.

The pulse v e l o c i t y i s then higher, p a r t l y because the morphology of the e t t r i n g i t e i s nat the same as in paste, e s p e c i a l l y at low w/c r a t i o s , or because contact paints between quartz p a r t i c l e s are less effected by hydrate formation than contact points between cement p a r t i c l e s .

Later on, a f t e r the second peak, the sand serves as an i n e r t aggregate and r e s i s t s the shrinkage by the cement and makes possible higher pulse v e l o c i t i e s .

The mechanism by which the sand influences the C3A reaction during the f i r s t hours needs maze detailed study.

(21)

The conclusion from these i n v e s t i g a t i o n s is that good mechanical properties of pastes and mortars can be obtained when during the f i r s t f i v e hours of hardening the shrinkage i s r e l a t i v e l y high, so that a f t e r t h i s p l a s t i c stage one can expect low shrinkage and high pulse v e l o c i t i e s (= high strength). This can be accomplished by promoting the growth of compact e t t r i n g i t e around the C3A grain or by suppressing e t t r i n g i t e formation by diminishing the C3A content. By t h i s C3S can give more tobermorite and so strength w i l l be b u i l t up sooner.

Research on hardened Paste and Mortar

S~multaneously with the preparation of the samples for the afore 3

cited research, prisms of 4 x 4 x 16 cm were made. Prisms which were used for determination of f l e x u r a l and compressive strength were stored during 28 days in water of 20 °C. Prisms f o r shrinkage measurements were s±ored during 91 days at 20 °C and 65 % r e l a t i v e humidity, a f t e r t h i s period they were tested for strength.

Strength and shrinkage of paste and mortar were related to the water cement r a t i o l o g a r i t h m i c a l l y (see Figure 14), and to the amount of gypsum (Figure 15) and of C3A (Figure 16) in portland cement.

At optimum gypsum content the strength i s maximal and the shrinkage

minimal (see Figure 15).

At other amounts of gypsum than the optimum amountt the water cement r a t i o of the sample has a l a r g e r influence on the strength and the shrinkage:

Athigher w/c r a t i o (> 0.40) a high gypsum content gives r e l a t i v e l y low strength. High shrinkage occurs when the amount of gypsum i s low at high water cementratio. When the gypsum content i s nearly constant increasing amounts of C3A give higher strengths but above 10 % C3A shrinkage increases excessively, independant of the w/c r a t i o (Figure 16). Measurements on d i f f e r e n t types of cement showed that aluminous cement shrinks the l e a s t , while normal blastfurnace

(22)

Vol. 5, No. 4- 315 SHRINKAGE, HARDENING, CEMENT PASTE, MORTAR

slag cement shrinks most a t o l l water cement r a t i o s .

The sequence i n decreasing shrinkage of the types of cement becomes for pastes and mortars:

Portland b l a s t - f u r n a c e slag cement class A Portland cement class A

Portland b l a s t - f u r n a c e slag cement class B Portland cement class B

Portland cement class C High-alumina cement

_OE

to.lc

I

1

"0 0 J & =Hi@~-alum~a cement ( H A]

. = ~ cemen, c~ss S • = Blast-furnace slog c ~ class A ~ ~ ~ H C , •=Blast-f . . . lag cemert class A (.riSer)

o =nlast-furnace slag cement class B 8 : ~" = Hi(jh -°lun,/*-~ cemenl

~¢~ ~- ~ shrinko(:Je a f t e r 91 days (~kr)

to

• =Potllancl cernen( class B (RHC I ) O = Portland cement class C ( RHC II)

• =Blasl-fumace slag cement class A ( PBS 2) • =Blas~-f~Jrnace slog cemen( class A (winter') ( PBS 2) o = Blast-furnace slag cemt,.nt class B ( PBS Z)

¢,,, 1 0 0 E Z : [ ,'- 80 ~; 6O .== ~o 8 0 9 = F'o~'tlar~ cem4,n~ cl~.s B ") I = Porttar, d cement class C {

B = Blas~-fumace slag cement ck]ss A I cured a t

I I = Blast .furnace slo~ ceme~ class A (winter) ZO°C-65"/.RH

• =BlaSl-furnace s l ~ cement class B o = High-alum~ c e m e n t • 0

°' °iCr'l

0.2 0.3 0.4 0.50.6070.80.9 =, WlC r a t i o FIG. 14

Strength and shrinkage of cement-stone made from portland cement class B and C, portland

b l a s t - f u r n a c e slag cement class A, A w i n t e r qua- l i t y and B and high-alumina cement

(23)

~ Z Z IE

t

,

°- o,

r

. I I 1 1 120 ~. ~ .~ ,0o ~ I ~ ~ ~ , ~ 0.~, ~ '0.2 k \ ~° , . _ _ L I \ \ 0.3

l

"~°"

2O 28 Oays ~ water o l l l J FN T T ~ T T ' ~ °.3°

...- gyps,urn(*/,.) roiled wilh ck'ker 19.9% C3A)

FIGo 1 5

Strength and shrinkage of cement-stone made from p o r t l a n d cement c l i n k e r and 2, 5 and 8 % gypsum %

T

.c- % 8

T

I I

111 , , L ,

~ I

. o . , o - IX t ! / J . , I \ I ~ , L I 1,,4,,- • " o.zs-~ 8 ~

~

0 . 5 0 " o . w - "

o | I I I I I

I

|00, ~

~ L U ~ _ d

2o 2 e e l s ~ , , ~ r i I I [ _ _ i P I ] i i 1 9 4 . 6 &l & . 6 % ~ %C3A FIG~ 16

Strength and shrinkage of cement-stone made from p o r t l a n d cement w i t h d i f - f e r e n t amounts of C3A

Nhen the aggregate cement r a t i o gets higher both strength and shrinkage become lower, so cement-stone has the highest s t r e n g t h and highest shrinkage.

(24)

16 12 8 Z, 0 i 20=(: - - 6 5 % relative humidity • =1:1 "7 20 A =1:1 ) 'ss~ z .,3 ) .zo - - e = 1 : 6 ) ~ T = 1 : 6 ) ! = U:~[::

i

! I

2 3 z, 5 6 7 8 t l A

I stel

' lli I

I

'

,~1111

!

PortlarKI Blast-tum(xe. sJag/!/~Portland cement

... ;1

B l PorUcnd 110 20 3C Portland Blast-furnace cement 1:1 30165

i i

~ Portlond cement closs BoC 1:1-20165 .t;.~ I 30 ~0 5O ?0 90 9 80 100 =, compressive strength (MN/m 2)

FIG. 17A and B

compressive strength: f l e x u r a l strength ( ~ ) for paste

Ratio and

f o r mortar with d i f f e z e n t aggregate to cement

r a t i o from portland cement and portland blast furnace slag cement cured in water at 20 ~C and 65

relative humidity

(crc) that may be important concezning crack formation in concrete,

. ~ f .

zs gzven as a function of the compressive strength. %

The~-~ zatlo was always higher f o r cement pastes with blastfuznace slag cement than with por±land cement class A, a f t e r hardening in water.

The difference between blast fuznace slag and portland cement

paste hardened under dry conditions (20 °C, 65 ~ r . h . ) was considez- abiy highez ( ~ 2 times) at low compressive strength than at high strength, where the diffezence disappeared gradually (Figuze 17A).

~ c

The r a t i o ~ - ~ at dry conditions was always higher than the zatio for samples sto:ed under water. The d i f f e r e n c e increased at higher compressive strength of the cement-stone.

(25)

For mortars w i t h an aggregate cement r a t i o of 1:1 the type of

=c

cement when hardened under water had no i n f l u e n c e on the r a t i o G--~ (Figures 17A and B). At dry c o n d i t i o n s b l a s t - f u r n a c e slag cement had

GC

approximately 30 % higher values f o r ~ than p o r t l a n d cement.

At the aggregate cement r a t i o of 3 there was no d i f f e r e n c e between the types of cement although the r a t i o again was higher under dry c o n d i t i o n s than under water. At the aggregate cement r a t i o of 6 no d i f f e r e n c e could be found e i t h e r between the types of cement or

between the c o n d i t i o n s . These systematlcs are i n t e r e s t i n g w i t h respect to the outermost l a y e r or skin of concrete where a t r a n s - i t i o n i n aggregate cement r a t i o fzom 6 or 7 to 0 occurs, going from the i n s i d e to the surface. At high f l e x u r a l and compressive strengths of the bulk concrete the f l e x u r a l strength of the skin of concrete can change considerably, e s p e c i a l l y under dry con- d i t i o n s , through a decrease of the aggregate cement r a t i o of the concrete + outermost l a y e r , t a k i n g i n t o account the l a r g e i n - fluence of the type of cement used.

0nly very good curing and a low water cement r a t i o can make i t possible t h a t the f l e x u r a l strength of the outermost l a y e r of concrete made w i t h b l a s t f u r n a c e cement i s not lower than i f made of p o r t l a n d cement ( a t the same compressive s t r e n g t h ) . In case of both, surface c o r r o s i o n and mechanical load (as in f r o s t damage), the use of b l a s t f u r n a c e cement may lead to f a s t e r surface c o r r o s i o n than the use of p o r t l a n d cement w i t h the optimum amount of gypsum at equal compressive s t r e n g t h .

Acknowledgement

The authors g r a t e f u l l y acknowledge the r e c e i p t of a grant from the Netherlands Committee f o r Concrete Research (CUR).

F u l l d e t a i l s of the work described in t h i s paper can be obtained from the s e c r e t a r i a t of the Commission 810, c/o IBBC-TNO, P.O. Box 49, D e l f t , The Netherlands.

(26)