The interaction of tricalcium silicate and tricalcium aluminate during their hydration

The interaction of tricalcium silicate and tricalcium aluminate

during their hydration

Citation for published version (APA):

de Jong, J. G. M. (1968). The interaction of tricalcium silicate and tricalcium aluminate during their hydration.

Technische Hogeschool Eindhoven. https://doi.org/10.6100/IR75565

DOI:

10.6100/IR75565

Document status and date:

Published: 01/01/1968

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.

THE INTERACTION OF TRICALCIUM

SILICA TE AND TRICALCIUM ALUMINA TE

DURING THEIR HYDRA TION

THE INTERACTION OF TRICALCIUM SILICATE AND TRICALCIUM ALUMINATE DURING THEIR HYDRATION

THE INTERACTION OF TRICALCIUM

SILICA TE AND TRICALCIUM ALUMINA TE

DURING THEIR HYDRA TION

PROEFSCHRIFT

TER VERKRIJGrnG VAN DE GRAAD VAN DOCTOR rn DE TECHNISCHE WETENSCHAPPEN AAN DE TECHNISCHE HOGESCHOOL TE ErnDHOVEN, OP GEZAG VAN DE RECTOR MAGNIFICUS, DR. K. POSTHUMUS, HOOGLERAAR IN DE AFDELING DER SCHEIKUNDIGE TECHNOLOGIE, VOOR EEN COMMISSIE UIT DE SENAAT TE VERDEDIGEN OP "DINSDAG 25 JUNI 1968 DES NAMIDDAGS TE 4 UUR

DOOR

JAN GODFRIED MARIE DE JONG

GEBOREN TE GELEEN

DIT PROEFSCHRIFT IS GOEDGEKEURD DOOR DE PROMOTOR PROF. DR. J.M. STEVELS.

Aan mijn ouders Aan Ans en Marc

CONTENTS

Chapter I. I. 1. I.1 .1 I. 1. 2. I.1.2.1. I.1.2.2. I.1.2.3. I.1.3. I. 2. I.2.1. I. 2. 2. I. 2. 2. 1 I. 2. 2. 1. 1. I.2.2.1.2. Introduction

The hydration of tricalcium s i l i -cate The

c

₃

s

hydration Introduction Experimental results Experimental results tion of pure

c

₃

s

Acceleration of the by amorphous silica of the

hydra-c

₃

s

hydration

Retardation of the

c

₃

s

hydration by calcium oxide and by calcium hy-droxide 1 3 16 16 16 21 21 28 29

Discussion of the mechanism of the 31 hydration of

c3s

The influence of aluminium hydrox- 33 ide and calcium oxide on the hy-dration of tricalcium silicate Introduction

Experimental results

Paste experiments by isothermal calorimetry

33 34 34

The influence of crystalline alu- 34

minium hydroxide on the hydration of

c3s

The influence of amorphous alumin- 35 ium hydroxide on the hydration of

8 I.2.2.1.3. I.2.2.2. I.2.2.2.I, I.2.2.2.2, I.2.2.2.3. I.2.2.3. I.2.2.3.1 I.2.2.3,2. I. 3. Chapter II. II .1. II .2. II.2.1. II.2.2. II. 3. II. 3. 1 • II.3.2. II.3.2.1. II.3.2.2. II.4. Chapter III

the influence of a mixture of amor-38 phous aluminium hydroxide and cal-cium oxide on the hydration of

c

₃

s

Paste experiments by other methods 41

Qualitative investigation 41

Quantitative investigation SI

Discussion of the results of the 53 paste experiments

Suspension experiments

Results of the suspension experi-ments

Discussion of the results of the suspension experiments

59 59

62

References 63

The hydration of tricalcium alumi- 67 nate

Introduction 67

The influence of amorphous alumin- 69 ium hydroxide on the hydration of

c

_3A

Suspension experiments Paste experiments

The influence of amorphous silica on the hydration of

c

₃

A

Introduction

Experimental results and discus-sion 69 78 84 84 86 Paste experiments 86 Suspension experiments 87 References 92

Interaction of tricalcium silicate 95 and tricalcium aluminate during

I II. 1. III.2. III.2.1 III.2.2. ! I I . 3. !II .3 .1. III.3.2. III.4. Appendix.A. A.1. A, 1. 1. A. 1. 2. A. 1. 3. A. 1 • 3. 1 A. 1. 3. 2. A. 1. 3 .3. A.1.3.4. A .2. A. 2. 1. A. 2. 2. A. 2. 3. A .2 .3. 1. A.2.3.2. A. 2 .3. 3. A. 2. 4. A. 2. 4. 1. A. 2. 4. 2. Introduction Experimental results Paste experiments Suspension experiments 95 97 97 101

Discussion of the experimental re- 107 sults

c3s hydration c 3A hydration References

Preparation and characterisation of the compounds investigated

107 109

11 2

114

Mëthods of characterisation of the 114 raw materials

Microscopie examination 1 14 Chemical determination of the free 114 calcium oxide content

Determination of the particle size 114 Air permeability method 1 14 Determination of surface area with 1 IS a gas adsorption areameter

Sieving

The Coulter counter Materials Tricalcium silicate Tricalcium aluminate Aluminium hydroxide Gibbsite Bayerite

Amorphous aluminium hydroxide

Silica Quartz Amorphous silica 1 1 5 1 1 5 116 1 16 120 1 21 1 21 122 122 123 1 23 123 9

10 A.2.5. A. 2 .6. A. 2. 7. A.2.8. A. 3. Appendix B. B. 1 B. l .1 B. 1. 2. B. 2. B. 2. 1. B. 2. 2. B. 2. 3. B. 2. 4. B.2.4.1. B. 2 .4. 2. B. 2. 4. 3. B. 2. 4 .4. B. 3. B. 3. 1. B. 3. 2. B. 4. B .4. 1 B. 4. 2. B. 5. B .5. 1 Calcium oxide Calcium hydroxide Water Nitrogen References Experimental methods Paste experiments Preparation of pastes Isothermal calorimeter Suspension experiments 124 1 24 124 1 24 125 126 1 26 126 126 129 Preparation of suspensions 129 D e t e r mi na t i on o f t h e e 1 e c t r i c a 1 c o n - 1 2 9 ductivity Determination of the pH

Chemical analysis of the water phase

Separation of liquid and solid

phases

Determination of the calcium

con-centration

130 130

130

13 l

Determination of the aluminate con-131 centration

Determination of the silicate con- 131

centration X-ray analysis

Qualitative X-ray analysis

Quantitative X-ray analysis

Infrared spectroscopy

Qualitative infrared analysis Quantitative infrared analysis Electron-optical examination Electron microscopy 131 1 31 135 l 3 5 136 137 137

B .5. 2.

B. 6.

B. 7.

B. 8.

Appendix C.

Selected area electron diffraction 137

Differential thermal analysis

Chemical determination of free calcium hydroxide

References

Cement chemistry nomenclature

Summary Samenvatting Dankbetuiging Levensbeschrijving 138 138 140 142 145 148 151 152 11

INTRODUCTION

Portland cement is by f ar the most important cement in terms of the quantity produced. I t is produced by the high-temperature reaction (1300 - 1S00°C) between lime on the one hand (obtained by decarbonating calcareous materials such as limestone) and silica, alumina and iron oxide on the other (obtained by heating argi l l a -ceous materials such as clay). The product which is known as clinker, is afterwards ground and mixed with a few per cent. of gypsum.

Portland cement compounds:

consists principally of the following

(1) tricalcium silicate

(c

₃

s*),

contaminated with small amounts of _{A1 2}

o

_{3 and MgO} ("alite")

(2) S - dicalcium silicate _{(SC 2 S);} (3) tric~lcium _{aluminate (C 3 A);}

(4) tetracalcium aluminoferrite, a phase with varying composition;

Portland cement is a hydraulic cement, i.e. when made into a paste with water, i t sets and hardens as a re-sult of chemical reactions between the water and the compounds present in the cement.

hydrates are formed.

During this hardening

A. In the hydration reactions at room temperature of

*

For cement chemistry nomenclature reference to Appendix C.

is made

which constitute about 75% of a Port-land cement by weight, the following calcium hydro-silicates can be formed: the crinkly foil-type C-S-H

( I) , the cigar-shaped C-S-H (II), and afwillite the latter is only formed during the ball mill hydration of

c

₃

s.

Tobermorite "gel"(badly

crys-tallised C-S-H(II)) is the main cementing material in hardened pastes of Portlandcement; i t gradually builds up a threedimensional network that gives con-crete its hardness.

B. The

( 1 )

( 2)

(3)

following hydroaluminates may be distinguished:

. + .

cub1c

c

_{3 AH 6 . Four H} ions can be replaced by one Si 4+ ion (hydrogarnet) and part of the Al 3+ ions

3+

by Fe ions;

hexagonal

c

_{3 A.CaX 2 .Hm(X can be e.g.} OH ;

2- . 3+

!

so

4 ) ; aga1n, part of the Al ions

b 3+ .

replaced y Fe ions;

hexagonal

c

_{3 A.3Cax 2 .Hm} (X can be e.g.

1co

2-2 3 can be

known as ettringite). Again, 3+ ions can be replaced by Fe

part of the

so

2

-4 Al 3+

ions.

The hydration reaction of Portland cement has a very complicated mechanism by which a large number of hydra-tes can be formed as is evident from the above informa-tion. Moreover, there is the complication that the com-pounds in Portland cement undergo not only their own hydration reaction, but also influence the reactions on each other. It is convenient, therefore, to study first the reactions of the various compounds with water and the nature and properties of their hydration products. After that,the setting and hardening processes of Port-land cement may be considered.

In this investigation i t is tried to get a better in-sight into the reaction mechanisms of the two important cement compounds, namely

c

_{3 A and}

c

₃

s.

For this purpose 14 not only the reactions of the pure compounds with water

are considered but also the influence which additives have on these hydration reactions. It is then tried to get an insight into the mutual interaction of

c

_{3 A} and

c

₃

s

during their hydration. In this way an insight may perhaps be obtained into the complete hardening mech-anism of Portland cement.

References

(1). H.F.W. Taylor, "The Chemistry of Cements", (1964) (London and New York: Academie Press)

(2). F.M. Lea, C.H. Desch, "The Chemistry of Cement and Concrete", rev. edn. (1956) (Edward Arnold Ltd., London)

CHAPTER I

THE HYDRATION OF TRICALCIUM SJLICATE

r.

1. _{The c 3}

s

hydration

I.1.1. Introduction

The mechanism of the hydration of tricalcium silicate has not yet been fully explained in spite of many re-search efforts (J*). One feature is the induction peri-od, during which practically no reaction takes place. This is evident

(a) from data obtained by isothermal calorirnetry (2,3); (b) from results obtained by rneans of X-ray

quantita-tive analysis (4-6);

(c) from shrinkage rneasurements (7).

Thi.s "dorrnant period" is of great importance in connec-tion with the rheological properties of cement pastes.

To elucidate the rnechanisrn _{of the hydration of c 3}

s

de-tai led investigations have been carried out on the equilibria in the systern cao

-

_{Si0 2}

-

_{H2 0.} This has brought to light that calcium silicate hydrates can be formed the composition of which depends on the cao per-cent age of the solution ( 8' 9) .

A number of hypotheses for this hydration have been de-scribed. These hypotheses can be divided into two parts ( 1 0) :

(a) the "through solution" theory.

The calcium and silicate ions from

c3s

dissolve, hydrate and hydrolyse during the solution reac-tions. When the concentrations of these ions in so-lution are high enough, crystallisation of hydrated calcium silicates occurs either throughout the so-lution or on the remaining

c

₃

s

particles (11-13). This theory was proposed by,among others, Greenberg and Chang (14). Their experimental results,however, could also be interpreted by Stein and Stevels in a different way, viz. by way of a direct mechanism

( l 5). Segalova et al. (16) suppose that the solu-bility limit of anhydrous

c

₃

s

plays the determining part for the reaction velocity. It appears diffi-cult, however, to bring this hypothesis in line with experimental results (17). Andreeva et al.(18)

interpreted as indicating

their strength data in

c

₃

s

suspensions a transition from a metastable to a stable hydrosilicate;

(b) the "direct mechanism" theory.

Water penetrates into the crystal structure of

c

₃

s,

which changes into a hydrate without first dissolv-ing ( 19). Water or OH ions diffuse through the hy-drated material and enter the anhydrous structure. For this diffusion process zur Strassen (20) arriv-ed at an equation, which was elaborated later by Turriziani et al. (21). The equation of Turriziani et al. (21) seems to be in better agreement with experirnental results, except for the first few hours. For this reaction time a 3-step mechanisrn was suggested. Volkov et al. (22) also arrived at an equation for the diffusion process.

A number of authors exFlain the autocatalytic character of the hydration of

c

₃

s

by the

drate into another. According

transition of to van Bemst

one hy-( 23) the crinkly foil-type hydrate C-S-H(I) changes into the 17

cigar-shaped C-S-H(II) under the influence of increas-ing calcium concentration in the water phase. This hy-po thesis seems to be supported by observations made with the electron microscope (24-27), The transition of a first hydrate into C-S-H(II) has been also proposed by Kantro et al. (28) and by Budnikov et al. (29). Ac-cording to these investigators, however, the first hy-drate is richer in Cao than C-S-H(I).

According to Brunauer and Greenberg (30), who started from an idea of Rebinder (31), the active

c

₃

s

surface increases owing to shattering of the particles caused by hydrate growth in surface cracks.Powers (32) asserts that a hydrate film around the

c

₃

s

grains is disrupted by the difference in osmotic pressure between the bulk solution and the solution near the

c

₃

s

surface.

Tsumura (6) followe,d the

c3s

hydration by determining the quantities of CH and

c

₃

s

by X-ray quantitative ana-lysis. He found practically no reaction during the first few hours of the hydration. He distinguished two reac-tion stages: the surface reaction during which the wa-ter penetrates the grain with constant velocity, and the diffusion reaction during which the diffusion of the water through the reaction layer formed

locity determining step.

is the

ve-The only hypothesis that accounts for all the facts seems to be that of Stein and Stevels (3). They give the following explanation, based partly on results of Kantro et al. (1,28): when

c

3

s

comes into contact with water, a hydrate (F.H. first hydrate) is formed close to the

c

3

s

surface. This first hydrate is formed by the water penetrating Even with an electron microscope the hydrate cannot be distinguished from the

c

3

s.

This hydrate has a very high Ca0/Sio 2 ratio (about 18 3:1 molar) and retards further hydration strongly. It

is converted to a second one (S.H.

=

second hydrate ) which retards the hydration reaction in a les5er degree and has a lower Ca0/Sio 2 ratio.Once nuclei of this sec-ond hydrate have been formed,the conversion takes place with a higher velocity (autocatalytic effect). The

sec-ond hydrate changes into a third(T.H. third hydrate),

the badly crystallised C-S-H(II), known as "tobermorite

gel". Schematically the reaction course of the

hyd~a-tion of c 3 s may be described as follows:

3CaO.Si0 2 + _H20

-1- directly

3CaO.Si0 2 .mH 2o (F.H.)

partly directly, accelerated

-1- by nuclei of S.H.

0.8 - _{1.5CaO.Si0 2 .nH 20 (S.H. ;C-S-H(I)?)} + _{Ca(OH) 2}

+ _{Ca(OH) 2}

-1- after some hours

1.5 - _{2CaO.Sio 2 .pH 20 (T.H.; C-S-H(II)?)}

The identification of S.H. with C-S-H(I) and T.H. with

C-S-H(II) is based only on the study of electron

mi-croscope photographs.

The transition of F.H. into S.H. is accelerated by a

decrease of the concentrations of Ca 2+ ions and OH-ions

in the water phase. This is taken by Stein and Stevels

to be the reason why amorphous silica accelerates the

hydration of c3s. Their conductivity and concentration

data on suspension hydration of c 3 s are consistent with the autocatalytic type of reaction,

is increased by aerosil.

whose acceleration

The mechanism proposed _{for the c 3 s hydration by Kawada}

and Nemoto (33) resembles the one suggested by Stein

Ca0/Si0₂ ratio ascribed to the hydrates formed (F.H.: CaO/Si0 2 "'- 1.4; S.H.: CaO/Si0 2 "' 2.4; T.H.: CaO/Si0 2 "'

1 .8).

I t should be not~d. however, that Kawada and Nemoto's conclusion was based on the free CH content found after extraction of hydrating pastes with water (as opposed to organic solvents) .Further, the following additional objections are to be raised against their interpietation of the results:

(a) the results do not agree with the electron micro-scopic habitus of the hydrates (3;41);

(b) the influence of amorphous silica cannot be plained in this manner, because i t is not to be ex-pected that this compound accelerates the transi-tion of a _{hydra te with a Ca0/Sio 2 ratio}

=

1 .4 in to a hydrate with a higher CaO/Si0 2 ratio (42);

(c) the calcium balance is not correct. For, if after hour of reaction

S -

77.

c

₃

s

has been hydrated in a paste consisting of, say, 1 g

c

₃

s

and 0.5 ml wa-ter, and if the hydrates formed dur ing that time have a mean CaO/Si0 2 molar ratio = 1 .6 (as fellows from Figs. IE _{and 1D 1 •} respectivel_y, of their pa-per), 17. 1 mg "CaO" or 22.6 mg CH must have been formed during that time. Where have these gone? They cannot have been precipitated as free CH,since Kawada and Nemoto do not find any. Nor does "CaO" stay bebind in the water phase, since they mention that af ter that time saturation of the water phase has just been reached; the latter, therefore, con-tains only 0.6 mg "CaO" in 0.5 ml.

Also according to Kawada and Nemoto the

c3s

hydration can get really started only if the liquid phase has reached such a supersaturation that crystal nuclei of the tobermorite-like hydrate are separated by which a chain reaction is formed. lts reaction velocity and 20 heat evolution grow less with decreasing

supersatura-tion. Reaction velocity is then determined only by the velocity at which the water can diffuse through the tobermorite-like hydrate layer which covers the unhy-drated c 3

s.

According to Malinin (13) the hydration of c3s gets really started too,if the liquid phase becomes saturated with CH.

It appeared desirable to collect more data by which to test the hypothesis of Stein and Stevels. From this hy-pothesis i t fellows

(a) that the velocity of CH precipitation is slow first, increases later to reach a maximum, and then decreases;

(b) that the addition of extra Cao retards the reac-tion, since conversion of the first hydrate to the second is expected to be retarded by the presence of additional ca 2+ and OH- ions in solution;

(c) that the c 3 s hydration is retarded by addition of solid CH and that the duration of the "dormant pe-riod" is not determined by the time required to reach supersaturation of the liquid phase.

These items were tested in the present investigation.

I.1.2. Experimental results

I.1.2.1. Experimental results of the hydration of pure

c

₃

s

In Fig. I.l(A) the heat evolution rate during hydration is plotted versus time. The heat evolution per unit of time increased during the first few hours of the hydration as described earlier in the literature ( 2 '3) . With the sample _{of c 3 s used it shows a maximum} after 5 hours qnd then i t decreases gradually. 21

15

20

25

30

-Ttme(hr)

Fig. l . i . Influence of CaO on heat evolution rate in pastes of c 3

s

A. 2.00 g c 3

s

+ 2.00 ml water

B.

_{2. 00 g c 3}

s

+ Ö.025 g cao + 2.00 ml water c. 2.00 g c 3

s

+ 0.05 g cao + 2.00 ml water

D. _{2.00 g c 3}

s

+ 0.095 g cao + 2,00 ml water

At the beginning of the hydration (after 1 hour) the electron micrographs (Fig. I.2A) show a crinkly foil-type hydrate, resembling the hydrate which Grudemo (34) supposes to be C-S-H(I).~he same foils were s t i l l visi-ble after four hours though at that time already ci gar-shaped agglomerates were clearly visible resembling agglomerates observed with tobermorite gel, C-S-H(II)

Fig. I.2. Electron micrographs of

c

₃

s,

hydrating in pastes (2.00 g

c

₃

s

+ 2.00 ml water) .Time af-ter hydration:

(A) l

h;

(B) 4

h;

(C)

200 h; (D) 200 h; (E) electron diffractogram of (D), viz. of CH

Fig. I.2 A

1

Fig. I.2 C

Fig. I.2 D

all C-S-H(I) had been converted into tobermorite gel.

Later, these cigar-shaped fibres increased in size and

number. After 24 h, and s t i l l more clearly after 200 h,

practically all visible particles showed to be the

fi-brous hydra te (Fig. I. 2C).

Calcium silicate _{hydrates with a Ca0/Sio 2} molar ratio

exceeding l .4 are reported (35) to acquire a more and

more needle-like character. Indeed, Kalousek (36) found a "convers ion point" _{round the CaO/Si0 2 ratio of} l .33

for the transition of the calcium silicate hydrates,

viz. between C-S-H(I) and C-S-H(II). However, not all

authors (37t agree with the composition of the hydrate

at this "conversion point".

In contrast to Grudemo's results (34),electron

diffrac-tion patterns of these hydrates could not be obtained;

we only obtained electron diffraction micrographs of CH (Figs. I.2D and I.2E). Neither were electron

diffracto-grams obtained when a sample, prepared by shaking

c

3

s

for six months in an excess of water so that the f i -brous hydrate only was present, was examined (Fig.1.3).

Indefinite rings were visible, but these could not be

but attributed to the carbon film, because they were

also visible when no hydrate was present on the carbon

film in the electron beam. In this case X-ray analysis

showed the presence of C-S-H(II), particularly

evi-denced by a peak at d

=

9.8

R.

A

peak at d

=

12.5

R,

characteristic of C-S-H(l), was absent. In the paste

experiments the tobermorite gel was not observed by

X-ray diffraction analysis. Apart from the peaks

men-tioned i t is difficult to distinguish between the

hy-drates. Additional peaks corresponding to d values of

3.07, 2.8 and 1.8

R,

coincide practically with peaks of

c

3

s.

26

Fig. I.3. Electron micrograph of

c

₃

s,

hydrating in suspension (3.00 g

c

3

s

+ 30.00 ml water) af-ter 6 months'hydration

4000 3000

2000

1500 1000 500

. - - - Wave number (cm-

1)

Fig. I.4. Infrared patterns of A. dry

c

₃

s

B.

c

₃

s,

after 6 months' paste hydration (2.00 g

c

₃

s

+ 2.00 ml water)

of Lehmann and Dutz (38), Runt (39) and Diamond et al. 943 cm-I and 893 cm- 1 , as-( 40) • The maxima at about

disappeared in the course of the reac-tion and were replaced by one braad maximum which had been shifted to lower wavelengths. The deformation

-1 -1

bands at 525 cm and 455 cm , also ascribed to

c

₃

s,

tended to disappear ultimately (Fig. I.4).

CH was also found. It was measured quantitatively by chemica! analysis and by infrared spectroscopy in the course of the reaction. The results of the chemical

0 . f O . - - - .

"

2

5

10

15 20 25

30

_____.. Tiine (hr)

Fig. I.5. A. Free CH determined by the extraction method, and B. Loss on ignition as a func-tion of the time during the reacfunc-tion

analysis of CH are reproduced in Fig. I.SA, together with the results of the measurements of the loss on ig-nition after removal of free water (Fig. I.SB). A simi-lar result was obtained by infrared spectrophotometry. Unfortunately, concentrations lower than 1% CH could not be measured, which was evident from a calibration line. The infrared determinations of

c

₃

s

and CH have been represented in Fig. I.6.

Eca(OH)₂

1

1400 ....

1 ' ,

\

\ \

1200

\.

₈₀₀

'b-.7,

.Q,.

.

. 600 - - - - 0

l

-~

---/

--~---=

)( .

600

200

Fig. I.6. Extinction of the

c

₃

s

peak (o---o) and of the CH peak (x~~x) per 10 mg sample as a function of time during the reaction

2.00 g

c

₃

s

+ 2.00 ml water

I. 1.2.2. Acceleration of the

c

₃

s

hydration by amorphous silica

Stein and Stevels (3) have investigated the influence exerted by amorphous SHx on the hydration of

c

₃

s.

Mea-surements by isothermal calorimetry show that aerosil

heat evolution rate is distinctly greater case without amorphous silica.

than in the

I. 1.2.3. Retardation _{of the C 3 S} hydration by calcium oxide and by calcium hydroxide

In Fig. I. 1. and Fig. 1. 8. i t is shown how the addition of Cao and CH respectively affects the heat evolution rate during the hydration of

c

₃

s.

It is seen that the hydration is retarded by the addition of a very small

5

10

15

20

_____..Time (hr)

Fig.

1.7.

Heat evolution rate versus time for

c

₃

s

past es + - - - + 2.00 g

c3s

+ 2.00 ml water 25 2.00 g

c3s

+ 0.40 g quartz + 2.00 ml water o~~-o 2.00 g

c3s

+ 0.40 g quartz + 0.050 g aerosil + 2.00 ml water

(from H.N. Stein, J.M. Stevels, J, appl.

8 6

4

'""

i'°' 2

\:

"'

8

0

....

~

6 Cl>

è

₄

c:: ~ .ö;!

~

2

~

0

î

10

8

A

6 4-2 0 0

2

4-

6

8

10 12 14- 16 18

_... Time (hr)

Fig. I.8. Influence of CH on heat evolution rate l.n

pastes of

c

₃

s

A. 2.00 g

c

3

s

+ 2.00 ml water

B.

2. 00 g

c

₃

s

+ 0.025 g CH + 2.00 ml water

c.

2.00 g

c

₃

s

+ 0.05 g CH + 2.00 ml water

quantity of CaO or CH;this retardation is less distinct

with higher quantities. In the presence of CaO or CH

the maximum heat evolution rate is lower and the whole

heat peak has been spread more than in the reaction of

I .1 .3. Discussion of the mechanism of the hydration of

c

₃

s

It would seem that the quantitative results of e.g. Angstadt and Hurley (4,5), Greenberg and Chang (14,15), Tsumura (6), Kawada and Nemoto (33),and those mentioned above can be explained by the hypothesis of Stein and Stevels (3,15,17,41,42) with regard to the hydration of c 3 s, as explained in the introduction (Chapter I.1.1.). This hypothesis

(l ,28), too.

rests on the results of Kantro et al.

It follows from the hypothesis of Stein and Stevels that an increase of the concentrations of Ca 2+ ions and OH ions in the water phase retards the transition of F.H. into S.H., and therefore slows down the c 3 s

hydra-tion. It is true that the quantity of ca 2+ and OH- ions in the solution would be independent of the quantity of free CaO if the equilibrium concentrations adjust them-selves directly.However, the formation of supersaturat-ed CH solutions is under these conditions well known

( l ) ; the addition of _{Cao or CH to c 3 s} is expected to increase the supersaturation of the solution towards

CH~ especially during early reaction stages, and, con-sequently, _{to retard c 3 s hydration.} This retardation can, therefore, be considered as a confirmation of <he hypothesis of Stein and Stevels. According to the hypo-thesis of Kawada and Nemoto (33) and also according to Malinin (13) the addition of solid CH should accelerate the hydration of c3s' because in this case the super-saturation of the liquid phase is reached more quickly. The transition of a hydrate with low _{CaO/Si0 2 ratio} ( l • 4)

(33)

into a hydrate with _{higher CaO/Si0 2 ratio (2.4)} should also be accelerated by the adding of CaO or CH. However, from the above mentioned results it is

The less pronounced character of this retardation which is observed when very large quantities of CaO or CH are added,

First,

can be explained by either of two alternatives. heat evolution at the beginning of the reaction may result in renewed acceleration of the formation of because in this case only a few more nuclei of S.H.

Ca 2+ and OH- ions are found in the solution than after

the addition of small quantities of free

cao

or CH.

Otherwise i t may be supposed that a direct conversion of F.H.

of T.H.

to T.H. takes place in which nucleus formation

is accelerated by the presence of excess of

Ca2+ _{and OH} ions in the solution. A choice between these alternatives is not yet possible.

From the curve of quantity of free CH versus time (Fig.

I . 5 . ) ' the maximum rate of the CH evolution is seen to

coincide with the maximum heat evolution rate. In

ac-cordance with the above mentioned hypothesis,

extrapo-lation of the rising part produces a line that does not pass through the origin but cuts the time axis at posi-tive values.

It should be remarked, as one hydrate changes into an-other, that i t makes no sense to describe the hydration

process of by means of an equation. This is the

reason why the suggested quadratic law of Turriziani et al. (21) does not hold during the first hours of the hydration of

c3s.

It is concluded that, although the hypothesis put for-ward by Stein and Stevels cannot yet be considered as proved, the results available at present are in

I . 2. The influence of aluminium hydroxide and calcium oxide on the hydration of tricalcium silicate

I.2.1 . Introduction

In order to consider if and how

c

_{3A influences the}

c

₃

s

hydration (cf. Chapter III), the influence of Al 3+ ions needs to be investigated.

Kalousek (43) has shown that Al 3+ ions may be incorpo-rated in at least one of the calcium hydrosilicates, viz. the well-crystallised tobermorite. He suggested

. . 13+ d + .4+ .

that subst1tut1on of A an H for S1 m1ght also occur during the hydration of cement at normal tempera-ture. Indications for this effect has been found with the electron microscope by Copeland and Schulz (25). The observations of Kalousek on the effects of Al sub-stitution in tobermorite were largely confirmed by Dia-mond et al. (44). Also according to Copeland et al. (45) tobermorite gel, formed by the hydration of pastes of will react chemically with aluminates bath during and after the formation of the gel. Changes in external form of the gel were observed by electron microscope examination of the gel

of pastes.

particles and surface replicas

The system CaO - _{Al 2}

o

_{3 - Si0 2 - H20 was investigated in} detail by Strätling (46), Strätling and zur Strassen

(4 7)' and Dörr (48). Arkosi investigated this system with the electron microscope (49).

In this investigation attention is given to the influ-ence of _{AH 3 on the hydration of} Since the heat evolution per unit of time on pastes of

c

₃

s

may be in- 33

fluenced by the various AH 3 modifications, viz. gibb-site, _{bayerite and amorphous AH 3 ,} the reaction mecha-nism was investigated qualitatively by isothermal calo-rimetry, and by X-ray diffraction, D.T.A., infrared and electron microscope methods.The course of CH concentra-tion was followed quantitatively during the reacconcentra-tion by chemical methods; the course of

c

₃

s

concentration by quantitative X-ray analysis.

r.

2. 2.

r.

2. 2. 1.

I.2.2.1.1.

Experimental results

Paste experiments by isothermal calorimetry

The influence of crystalline aluminium hy-droxide on the hydration of

c

₃

s

In Fig.

r.

9. the results of the experiments with the isothermal calorimeter are reproduced in which the in-fluence of the addition of crystalline _{AH 3 on the} hy-dration of

c

3

s

was investigated. By adding gibbsite the maximum of the heat evolution rate falls slightly la-ter. The same occurs s t i l l later after the addition of bayerite, and the difference with the original situa-tion is then s t i l l clearer. Same indication of a "pre-peak" is found, whereas the

c3s

peak, in addition to shifting, is lower and broader. Since bayerite is less stable than gibbsite (62), i t is likely that the alumi-nium ions will go more quickly in solution from this preparation than from the stable gibbsite.With prepara-tions from which the aluminium ions go even more quick-ly in solution, _{e.g. amorphous AH 3 , these effects were} 34 expected to be even more pronounced.

Fig.

r.

9.

I.2.2.1.2.

20 25 30

- - - Tlme (hr)

Influence of crystalline AH

₃

on heat evolu-tion rate in pastes of

c3s

A. 2.00 g

c3 s

+ 2.00 ml water

B. 2.00 g

c3s

+ 0.20 g gibbsite + 2.00 ml water

c.

2.00 g

c3s

+ 0.20 g bayerite + 2.00 ml

water.

The inf luence of amorphous aluminium hydrox-ide on the hydration of

c

₃

s

The results of the experiments with the isothermal cal-orimeter on the influence of the addition of amorphous AH 3 on the hydration of

c

3

s

are reproduced in Fig.I.10. Four maxima in heat evolution rate are always found. The addition of _{amorphous AH 3} causes the main peak (peak IV) to be shifted towards later hydration times. In the case of small quantities (0.01 g) the shifting 35

----. Heat

evolutlon

rek

(1<14

cal

sec-1

per g

c;

S)

~

i:

~

~

"'

~

ti:

..::?.. ~ ~

_lb.

~

In

'""

Fig. I . 1 0. Influence of amorphous AH

3 on heat evolution rate in

pastes of

c

3

s

A. 2. 00 g

c

3

s

+ 2.00 ml water B. 2.00 g

c

3

s

+ 2.00 ml- water + 0. 01 g amorphous AH3

c.

2.00 g

c

3

s

+ 2.00 ml water + 0.04 g amorphous AH3 D. 2.00 g

c

3

s

+ 2.00 ml water + 0.07 g amorphous AH3 E. 2.00 g

c

3

s

+ 2.00 ml water + 0 .10 g amorphous AH3 F • 2.00 g

c

3

s

+ 2.00 ml water + 0.25 g amorphous AH3 G • 2.00 g

c

3

s

+ 2.00 ml water + 0.40 g amorphous AH3

extends further than in the case of larger quantities (up to 0.04 g), but with s t i l l larger quantities the peak is again shifted to later times. The peak always becomes flatter for

AH 3 , in other words

increasing quantities of amorphous the maximum heat evolution rate at this stage is always smaller.

E

2

c;;-u"'O

'O.

q.

c...

_Q

~2

Î~

- 0

'te

_{,g_}

*

_Q.

.e

2

e

:§

0 ~

~

q.

15

~2

î

3

A

145 165 Fig. I.11.

Influence of mixture of amorphous _{AH 3 and cao on heat evolution} rate in past es of c 3

s

A. 2.00 g c 3

s

+ 2.00 ml water + _{0. 01 g amorphous AH 3}

B. 2.00 g c 3

s

+ 2.00 ml water + _{0.01 g amorphous AH 3} + 0.o1 g cao c. 2.00 g c 3

s

+ 2.00 ml water + _{0. 01 g amorphous AH 3} + 0.02 g cao D. 2.00 g c 3

s

+ 2.00 ml water + 0.o1 g amorphous _{AH 3} + 0.04 g cao

0

0

5

25 45 65 85 105 125 145 165

--.T;me(hr)

Fig. I . 1 2.

Inf luence of mixture of amorphous AH 3 and cao on heat evolution ra te in pastes A. 2.00 g c 3

s

+ B • 2.00 g c 3

s

+

c.

2.00 g c 3

s

+ D • 2.00 g c 3

s

+ E • 2.00 g c 3

s

+ I.2.2.1.3. of

c

₃

s

2. 00 ml water + _{o. 1 0 g amorphous AH 3}

2. 00 ml water + _{0.10 g amorphous AH 3} + 0.01 g cao 2.00 ml water + _{0. 10 g amorphous AH 3} + 0 .02 g cao 2.00 ml water + 0. 10 g _{amorphous AH 3} + 0. 04 g cao 2. 00 ml water + _{0. 10 g amorphous AH 3} + 0,08 g Cao

The influence of a mixture of amorphous aluminium hydroxide and calcium oxide on the hydration of

c3s

The course of the hydration reaction is similar i f , in 38 addition to amorphous _{AH 3 ,}

cao

is added to the

c3s

(Figs. I.11" I.12., I.13.). The only difference is

th~t the third and the fourth heat evolution peaks

ap-pear more quickly and are more pronounced with increas-ing quantity of cao. In some cases the second and the

40

8

5

30

55

Fig. I. 13.

Inf luence of mixture of amorphous AH 3 and cao on heat evolution rate in past es of c 3

s

A. 2. 00 g c 3

s

+ 2.00 ml water + _{0.40 g amorphous AH 3}

B. 2._{00 g c 3}

s

+ 2.00 ml water + _{0 .40 g amorphous AH 3} + 0.01 g cao c. 2._{00 g c 3}

s

+ 2.00 ml water + _{0.40 g amorphous AH 3} + 0.02 g cao

D • 2._{00 g c 3}

s

+ 2.00 ml water + _{0.40 g amorphous AH 3} + 0.04 g cao

third peaks merged into each other. If amorphous _{AH 3 is}

not present, cao retards the hydration of c3s, as has

been demonstrated in Chapter I. 1 .2.3.

Table I.1.

Results of the reactions:

(a) 2.00 g

c

3

s

+ 0.40 g amorphous AH 3 + 2.00 ml water

(b) 2.00 g c 3

s

+ 0.40 g amorphous _{AH 3} + 0.08 g Cao +

2.00 ml wat er

Rel.et ion Dl (fefenti·al thermal 1nfrared Elect ron

stopped an.ilys is; endother.m.i l X-ray analysis invest 1g•t ion. inicroscope

!lfter peak

" different N•w peaks

" invest igat Lon

tempe~•tures, oc

(•) tos0 (m); 225° (m1.1) No n<w products :H 70 cm-1 Uncovered CJS gra.ins 1 h HQO (•) wert found 1410 with nearby amorphous

1470 "f takes" Heat 7 h 105° (raw); 225° ( • ) peak 250° (>) 25 h 1oso (v1.1): 150° (•w) 3410

,.

•1 225° (m); no0 (•)

:

.

:

~

6)

<h• pedi. at 1410 CDl- l predomi.nates (b) 1 so0 (11); 225° (vs) 1650 cm-1 Uncovered c,s grains

'

h H0° ( s ) ; 470° (m) Th• rest of <h• spec- wi th h11re •nd the re trum analogous •o (•) ( ibrts

" <h• surf a.ce " 2S h Heat (•) 120° (vs); 180° (v!.'} ~~~:~t,~7~! 3670 -1 3470 '" -1 llydrates on

".

C)S Cl'!I ;

peak IIÎ so h 22so (111); 250° (>) hy- 3(.40 )410 grain;

.

kind of

)QQO (vv); 4000 (v1.1) drate") 1410 1160 "sponge" 1l4D 711

"

h 1100 (vs); 1so 0 (w) C2ASHg (o) 600

'"

1800 (v); 2250 (mv) Th• ~ i 1 ic at e band

·

·-2soo (111); 100° (w) tveen 800 and 1000 cn:i·I )SQO (1111.1); 4000 (w) changes Î t t (orm

Tr.ans i tion 119 h 120° (vs); 1so0 (V'l5) C2ASH8 (vs) Th• .sa me hexagonal between 1800 (w); 22So (•) Ct.AHI) (vw) pl111tti

heet peek 2so0 (111); 300° (w) "C4AH 11"(vw)

1 f l •nd IV 3500 (u); t.000 (v1.1) Ct.AHl g (vv) C3ACaC03H12 (u) (b) 1200 (vs); 1 soo (vvs) C2ASH:5 (vs) 3670 '" _, 3620 '" -1

"

h 2 no (111w); 250° (1111.>) C1AC<lC01M12 ( >) 3S20

3000 (w); 350° (iaw) 3400-3SOO broad bend;

'"' relt of

,"

spe

c-truia simi lar <o (o)

Heat pe<1\t (.) 120° (vs); 1500 (Via) c 2ASH5 (vvs) )670 c~-1; 3650 cm-1 Hydr.ates

~7n~h~/~!u'\

IV l ' I h 1800 (vvw);2250 (m) C1.,AHu (•) 3St.0 (?) 3520 grain; a

250° (:Il) ; 300° (w) "Ct.AH13" (•) }1.10 )t.t.0 h1U&ROnal pllltl'lS

3S0° (nw) C1.AH19 (vw) Tho rest of •h•

spec-C]'.\CaC03 H1 z (<) tfU!"L analoi.',S •o (•)

" 119 h; only

"

.

nax imo. Of ,,, sillcate

190 h 1200 (vs); J so 0 (vvs) CH (v11s) bands betwi?e-n 800 ond

240 h 2250 (vw); 2500 (vvw) C2ASH8 (vvs) 1000 cm-1 disa'ppedr

3000 (111w); 350° (111w) C 4 AH:13 (s)

470° (vvs) "C1,,AH13'"(vs)

~~~~!~o~~7~cj>

28 days •h• sa11e pattern _" CH (vvs} )670

'" -1 ; 36SO <• -1

'

l'IOnths (,) 240 h 875° C2 ASlls (vs) JSOO

(vague band): only C4Ml13 (~) )t.00-341.0 broad band; tht

...

peak

..

4100 "C 4 AH' 3'' (111) rest of

".

sp..:ctr1.1111

sim-KfOVS o•• C4 A~ l 'l (vw} i l ar •o (>) " 2•0 h C 3AC.:ic03 H1 2 (<."5) (b) 120° (vs): 150° (vs) CH (vvs) 14 diiy& 2B 0 (111v); 250° (mw) C2ASHg (11$) 3000 (1.1); )500 (w) c 4 AHJ) {w) i.700 ( yy ~) "C4AH11''(111w) ~~~~!lo

3

~7: c~l

1111& • very very strong; vs • ·vf!:ry 1trQn~; s • stronR: m • nitdium; m1.1 • onediu"'. wi'ak; w • weak.

I.2.2.2. Paste experiments by other methods

I.2.2.2.1. Qualitative investigation

The reaction products of the following reactions were

investigated:

(a) 2.00 g

c

₃

s

+ _{0.40 g amorphous AH 3} + 2.00 ml water

(b) 2.00 g

c

₃

s

+ _{0.40 g amorphous AH 3} + 0.08 g Cao +

2.00 ml water.

The products were investigated by X-ray diffraction,

D.T.A" infrared analysis and electron microscopy. The

results are summarised in Table I . J.

X-ray analysis

The results of the X-ray analysis have been reproduced

in Figs. I.14. and I.15. After heat peak I and during

and after heat peak II no new products can be observed

by X-ray analysis. During heat peak III the X-ray peak

of the gehlenite hydrate _{(C 2ASH 8 ) appears which always}

becomes stronger as a function of time. At the end of

this heat peak the peaks of the c 4AHn products also ap-pear with and without incorporated co 2 (C 4 AH 19 , c 4AH 13 ,

11_{C4 AH 13}11 _{and c 3A.Caco 3 .H 12 ). During heat peak IV all}

these X-ray peaks become stronger and the peak of CH

also appears.

Differential thermal analysis

The D.T.A. results have been reproduced in Figs. I.16.

and I.17. The D.T.A. graphs show a peak at about 60°c,

probably originating from free water. There is always a

braad band at about 660°C, toa, probably caused by

Caco 3 . c 3

s

shows peaks at 6S0°c, 920°c and 960°c. The

latter two are found continuously up to a reaction time

of 240 h. For the ascription of the D.T.A. peaks to the

different calcium aluminate hydrates is referred to

Chapter II.2. J. Moreover, weak endothermal effects

Fig. 1.14. 42

R

z

Q

m

v

R

_R+W+S

Il

1

1 1'+Z

s

3+,R

~'

1

W.

1

p P+W 1 1 1 Il

w

1 1

il

1 1 1

y

1

R+W .1 ₁ 1 1 _" R

_N"

1

l

1

1

Pl' 1

1

1

;r 1 IJ[

1

1

II

1

1

I

5

9 9

15 17 17 i!) 250 -211'

X-ray analysis of the paste reaction:

2.00 g c3s + 2.00 ml water + _{0.40 amorphous AH 3}

The reactions were stopped after:

I. 0 h II. h; 7 h; 25 h 11 I. 50 h IV. 75 h v. 1 19 h v l . 145 h VII. 190 h. VIII. 28 d; 24 0 h 6 m For the marking of The diffractograms widths of s l i t from 2e 5 - 90) 8 - 18°) ( 1 7 - 50°) the we re wit h with with

products see Appendix c.

made with different

!o !o 10

co J

R

s

Q

R

p

15°

N

R+W+S

III

R

p

ril.

I

25°

Fig. I.15, X-ray analysis of the paste reaction:

2.00 g c 3

s

+ 2.00 ml water + 0.40 g amorphous AH 3 +

o.os

g cao

The reactions were stopped after: I. II. III. IV. 0 h 5 h 24 h 14 d

For the marking of the products see Appendix C

to

c

_{4 AHn products,} fractograms.

too,

The D.T.A. pattern of the second heat peak the reaction of 5 ml 0.50 g amorphous AH 3 ter pattern remains

is of and the

as is evident from X-ray

dif-the experiment with cao af ter similar to that obtained after supersaturated CH solution with 0.20 g amorphous SH The

lat-x

same from about 1 hour until 3 months reaction time. The X-ray pattern of the product of this reaction shows a number of braad peaks. These peaks can be partly attributed to c_{3 A.Caco 3 .H12} hut a number of broad bands could not be identified(d ~ 4.3~;

1h 7h 2Sh 50 100 200 .n? Ij()() -Temp.("C)

m

l/00 -Temp.(°C) 500 500 200

Fig. I.16. D.T.A. of the paste reaction:

2.00 g

c

₃

s

+ 2.00 ml water + _{0,40 g amorphous AH 3}

The reactions were stopped after:

A.

h

B.

7 h

c.

25 h D. 50 h

E.

75 h F. l l 9 h G • 145 h H. 1 90 h·

'

240 h

r.

4400 h ( 6 m)

D.T.A. pattern is attributed to

c

_{3 A.Caco 3 .H 12} (see Chapter II.2.1.), the peak at 250°c must be ascribed to an amorphous calcium aluminate hydrate, in which s i l i -cate ions are probably present (C-A-(S-)H),because this D.T.A. peak is not found after the reaction of 5 ml of supersaturated CH solution with 0.50 g amorphous AH 3 in 44 the absence of amorphous SHx.

Sh

24h

350h

100

200

300

400

500

---.. Temp.(°C)

Fig. I.17. D.T.A. of the paste reaction:

2.00 g c₃s + 2.00 ml water + 0.40 g amorphous

AH 3 +

o.os cao

The reactions were stopped after:

A.

S h

B. 24 h

c.

350 h

The endothermal D.T.A. peak at 120°c must be attributed to gehlenite hydrate, since i t is found in all pastes showing gehlenite hydra te X-ray diffraction lines (50). In the literature (SI ,52) a different D.T.A. peak is 45

given for the gehlenite hydrate, but probably the water of that sample is bound otherwise than in our compound. The D.T.A. peak at 470°C must be attributed to CH.

Infrared analysis

The results of the infrared analysis have been

summa-rised in Table l . i . The many l i t t l e peaks on the

water-band in the infrared spectrum must be attributed to

c 2ASH8 and to the calcium aluminate hydrates, partly on

X-ray diffraction evidence, partly on evidence

literature (see Chapter II.2.1 .).The peak at 3470

in the

-1

cm

which is clearly visible in the beginning of the

reac-tion,must be ascribed to the amorphous C-A-(S-)H, as is

evident from the D.T.A. experiments. CH has a peak in

-1

the infrared spectrum at 3650 cm • This

product,con-taminated by traces of carbonate, shows streng infrared

absorption maxima at about 1410 and 1470 cm-I .According

to Runt (53) these two maxima originate from the v 3

vi-bration of the carbonate at 1430 cm-! The carbonate

-1

bands of _{c 3 A.Caco 3 .H 12 are at 1400 and} 1360 cm ( 5 4'

55). Th e e f f eet in our · experiments at · 136 0 cm-! must

therefore be attributed to this complex. The effect at

1140 cm-! should probably be attributed to gehlenite

hydrate. The effects at 715 and 600 cm-I are carbonate

-1

bands (38);the new peak at 425 cm is probably

attrib-utable to CH. If _{c 3}

s

is hydrated, the silicate bands

-1

at 940 and 890 cm disappear and are replaced by a

broad maximum that has been shifted to a lower

wave-length (38-40). The deformation bands at 520 and 450

-1

cm _{attributed to c 3}

s,

disappear for the most part,

too. This can already be observed during the third heat peak.

Electron microscopy

Jµ

..

.

1

Fig. I.23 A

Fig. I.23 B ~·

Fig. I.23, Electron micrographs taken during the reaction 2.00 g

c

₃

s

+ 2.00 ml water + 0.40 g

amor-phous AH 3

The reaction was stopped:

during heat peak II: A. af ter 4 h du ring heat peak III: B • af ter 43 h

c.

af ter 119 h during heat peak IV: D. af ter 200 h E • af ter 200 h

F. af ter 200 h

G. electron diffractogram of (F) , viz. of

1µ

'·

'

.

Fig. I.23 C

f

µ

Fig. I.23 48 Fig. I.23 E

Fig. I.23 F

Fig. I.23 G

and I.24. Electron diffraction patterns could be ob-tained neither from the "flakes" during heat peak II (Fig. I. 23A), nor from the "sponges" during heat peak III (Fig. I.23B) and nor from the "mats" during heat peak IV (Fig. I.23E). These are probably aluminium-con-taining hydrates, which could not be identified because electron diffraction patterns could not be obtained. Electron diffractograms could be made of the hexagonal plates (calcium aluminate hydrates and gehlenite

so

Fig. I.24 A

·

1µ

Fig. I.24

B

l_..:,..~__;___t::::~:::::::JL..

~

.... "".

Fig. I.24.

Electron micrographs taken during the

reaction

2.00 g

c

₃

s

+ 2.00 ml water + 0,40 g amor-phous AH 3 + 0,08 g Cao

The reaction was stopped:

during heat peak

II:

A. after 5 h during heat peak

II!: B.

after

25

h

I.2.2.2.2. Quantitative investigation

The variation in the amount of

CH

liberated during the reaction of 2,00 g

c

₃

s

with 2.00 ml water and 0.01 g amorphous

AH

₃ is given in Fig.

I.lBA.

The free

CH

was determined chemically, The results of the ignition loss measurements on dried pastes are shown in Fig. I.18B. For the reaction of 2.00 g

c

₃

s

with 0.40 g amorphous

~

0.10

~

tl ~

8

0-08

t

:;::

}.o.~

t

0.04-J

0.02

0

7

--%Cao

--·

A

t

6

5

4

J 2 1:,..-_ _ _ ; 0

_o

_{20 40 60 80}

₁₀₀

_{120 140}

- . .

Ti'me

(hr)

160

Fig. I.18. Reaction of 2.00 g

c

₃

s

with 0.01 g amorphous

AH

_{3 and 2,00 ml water}

A. course of formation of

CH

(determined by extraction) as a function of time

Fig. I.19. 0.22~---.

Q.20

0·18 ~ 0.16 ..!?

&

0.11/-·c:

-2' 0.12

t

0-10 u-061----'---'---'---"---'"--'--~---, 10

%Cao

• 8

1

6 2 O'--~-'-~~'--~--'-~~'--~_._~~..__~__.

0

50 100 150 200 250 300

___....Time (hr)

Reaction of 2.00 g c 3

s

with 0.40 g amorphous AH 3 and 2 .00 ml water

A. course of formation of CH (determined by extraction) as a function of time

B. ignition loss as a function of time

AH 3 and 2.00 ml water similar results are given in

Figs. I.19A. and I.19B.

During the reaction of 2.00 g c 3

s

with 2.00 ml water,

and 2. 0._{0 g c 3}

s

with _{0.40 g amorphous AH3 ,} 0.08 g cao and 2.00 ml water the amounts of unreacted c₁

s

were

de-termined at various times by means of quantitative

X-ray analysis. The results are given in Tables I.2.and 52

r.

3.

Table 1.2. Quantity of

c

₃

s

during the reaction 2.00 g

c

₃

s

+ 2.00 ml water

experiment stopped _{%C 3}

s

calculated

after _{from RI}

*

1 96 Table I.3. min 98 h 70

Quantity of

c

₃

s

during the reaction 2.00 g

c

₃

s

+ _{0.40 g amorphous AH 3} +

0.08 g CaO + 2.00 ml water

from

-

--experiment stopped _{%C 3}

s

calculated

af ter _{from R 1} from

1 min 100 105 5 h 99 104 24 h 80 85 96 h _{40 40} R2 R2

The conclusion is that the quantity of

c

₃

s

consumed af-ter 96 h during the reaction with amorphous AH 3 and Cao is greater than that without these two components.

I.2.2.2.3. Discussion of the results of the paste ex-periments

In the presence of amorphous _{AH 3 the reaction of}

c

₃

s

with water shows a very complicated course. The experi-ments in the isothermal calorimeter always give a pat-tern similar to that given in Fig. I.10.

*

For the meaning of R 1 and R2 ref erence is made to

The following mechanism is in accordance with all these

observations. According to the hypothesis of Stein and

Stevels

(3,15,42)

with regard to the hydration of

c

₃

s

(as explained in Chapter I. l. 3.),

(F.H.) is formed, when

c

₃

s

comes

the "first hydrate" into contact with wa-ter.This is a close-fitting hydrate, invisible by

elec-tron microscope methods, which retards the hydration

reaction strongly. This hydrate is thought to be formed

on the

c

₃

s

surface during the first heat peak. When CaO

had been added to the

c

₃

s

the former was converted into

CH to a great extent during this peak I .

is accompanied by several reactions during

Peak II which, l i t t l e ,

as is evident f~om quantitative X-ray analysis

if any,

c

₃

s

is converted, The heat evolution

during this peak II, calculated by means of integration

of the results obtained with the isothermal

calorim-eter, is

with

0.40

(30

min

-22 cal/g

c

₃

s

for the reaction of 2.00 g

c

₃

s

g amorphous AH 3 , 0.08 g CaO and 2.00 ml water

5 h). If i t is assumed that during heat peak

II 5% of the

c

₃

s

.is hydrated at most (which is thought

to be the analysis)

least

440

.of 2.00 g a value of

inaccuracy range of the quantitative X-ray

and no other reactions have taken place at

cal/g

c

₃

s

would be evolved. For the reaction

c

3

s

with 2.00 ml water in the absence of AH 3

120

cal/g was found, and this agrees well

with results in the literature (56-59), although an

exact comparison is difficult since no corrections were applied for the hydrate surface energy and for the heat

of adsorption of water on the surface. The dif ference

between the 120 cal/g value for the heat of hydration

of

c

₃

s

_{in the absence of AH3 on the one hand,} and the

440

cal/g value for the heat of hydration of

c

3

s

in the

presence of AH 3 during the second heat evolution peak

on the other, is such that i t is very unlikely that i t

therefore, indicates that in addition to the hydration of els a number of other reactions take place during heat peak II. It should be noted that the additional free cao intensifies the heat peak, but does not in-crease the els consumption considerabl~. For these rea-sons this heat peak is attributed to a reaction of the amorphous AHl with CH, the latter being formed not only by the hydration of Cao but also by the conversion of F.H. + S.H. + CH (3). The free Cao appears to react, not directly but only after 1 hour, with AHl; this in-dicates chat either the amorphous AHl must undergo a certain activation, or the hydration of Cao is re-tarded in the presence of AH 3 at the start of the reac-tion. Experiments in the isothermal calorimeter and D.T.A. data about the reaction between CaO, _H2o and amorphous AHl confirmed this hypöthesis. The results of these experiments are given in Fig. I.22., which shows that in pastes of Cao + _{H2o a second heat peak is} ob-served in the presence of AHl.

In pastes of els + H

₂

o + AHl the conversion of the F.H. after the first peak is slowed down immediately, pre-sumably because aluminium ions are incorporated into the calcium silicate hydrate formed. For, it is evident from quantitative X-ray analysis (Table I.l.),that only l i t t l e , if any, els is hydrated. Kalousek et al. (43-45) have shown thac'aluminium ions may be incorporated into tobermorite. Again, a close-fitting hydrate, in-visible by electron microscope methods, on the c 3 s sur-face is formed and this stops further hydration. This aluminium-containing calcium hydrosilicate is thought to be more stable with respect to recrystallisation than alumini~m-free calcium hydrosilicate, in analogy to Kalousek's results on Al-con~aining tobermorite(4l). As mentioned above, the amorphous AHl reacts with the CH and fotms a calcium aluminate hydrate in which sili- 55