The influence of Na2O on the formation and colloidchemical properties of calcium aluminate hydrates

The influence of Na2O on the formation and colloidchemical

properties of calcium aluminate hydrates

Citation for published version (APA):

Spierings, G. A. C. M. (1977). The influence of Na2O on the formation and colloidchemical properties of calcium aluminate hydrates. Technische Hogeschool Eindhoven. https://doi.org/10.6100/IR73086

DOI:

10.6100/IR73086

Document status and date: Published: 01/01/1977

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

THE INFLUENCE OF Na20 ON

THE FORMATION AND COLLOIDCHEMICAL PROPERTIES

OF CALCIUM ALUMINATE HYDRATES

PROEFSCH RI FT

TER VERKRIJGING VAN DE GRAAD VAN DOCTOR IN DE TECHNISCHE WETENSCHAPPEN AAN DE TECHNISCHE HOGESCHOOL EINDHOVEN, OP GEZAG VAN DE RECTOR MAGNIFICUS, PROF.DR. P. VAN DER LEEDEN, VOOR EEN COMMISSIE AANGEWEZEN DOOR HET COLLEGE VAN DEKANEN, IN HET OPENBAAR TE VERDEDIGEN OP

DINSDAG 21 JUNI 1977 TE 16.00 UUR

DOOR

GIJSBERTUS ADRIANUS CORNELUS MARIA SPIERINGS

GEBOREN TE DRUNEN

Dit proefschrift is goedgekeurd door de eerste promotor Dr. H.N. Stein en de tweede promotor Prof.Dr. J.M. Stevels.

VOOR MARIJKE VOOR MIJN OUDERS

CONTENTS

CHAPTER I. INTRODUCTION 7

CHAPTER II. _{THE INFLUENCE OF Na2o ON THE HYDRATION}

OF

c

_{3A. I. PASTE HYDRATION.} 17

Introduction 20

Experimental 20

Hydration of

c

_{3A in Solutions of Alkali Hydroxide}

or Hydroxoaluminate 21

Hydration of xNa 20.(3-x)CaO.Al2o 3 Solid Solutions 24 CHAPTER III. THE INFLUENCE OF Na2o ON THE HYDRATION

OF

c

_{3A. II. SUSPENSION HYDRATION.} 27

Introduction 30

Experimental 30

Hydration of c_{3A in NaOH Solutions in Suspensions} 30 Hydration of 0.25Na 2o.2.75 CaO.Al2o 3 in Suspensions 36 CHAPTER IV. ELECTROKINETIC PROPERTIES OF CALCIUM

ALUMINATE HYDRATES. Introduction

Experimental

Theoretical: Calculation of the surface charge

39

41 41 behind the electrokinetic slipping plane 43

Results and Discussion 46

Conclusion 52

CHAPTER V. THE COAGULATION OF Ca

3Al2(0H)12 IN AQUEOUS ELECTROLYTE SOLUTIONS. Introduction Experimental 5 55 57 57

Results 60

Discussion 63

Conclusion 71

APPENDIX A: ADDITIONAL DATA ON THE

c

₃A-HYDRATION 75

a)

c

₃A hydration in 2M Alkali Hydroxide Solutions 75

b)

c

₃A in mixed 2M NaOH + NaAl(OH)₄ Solutions 76

c) Some Remarks on the Saturated

c

₃A Solution Mechanism 77

APPENDIX B: ADDITIONAL DATA ON MATERIALS AND METHODS 80

a)

c

3A 80

b) Heat Evolution Rate Measurements in Suspension 80

APPENDIX C: INTERACTION BETWEEN PARTICLES IN A SUSPENSION;

DERIVATION OF SOME FORMULAE 85

a) Attraction 85

b) Repulsion 88

c) Total Interaction 93

d) Influence of Surface Potential ~

₀

on the Stability

Ratio during Secondary Coagulation 96

SUMMARY SAMENVATTING DANKBETUIGING KORTE LEVENSBESCHRIJVING 99 102 106 106

CHAPTER I Introduction

Literature references of this chapter are to be found on page 15

INTRODUCTION

Portland cement is manufactured by burning a mixture of calcareous materials (e.g. limestone, chalk etc.) and

siliceous materials (e.g. clays) at temperatures of about 1400°-1500°C. The siliceous minerals usually contain an amount of alkali oxides (mostly Na₂o and K₂o) which partly volatilize at the temperatures existing in the kiln, and which will be withdrawn from the kiln by the vented heated air; further alkali oxides are removed in the dust, in which they are concentrated (1). Consequently the amount of alkali oxides present in the final product is significantly lower than that of the starting materials but i t still amounts to 0.5-1.5% by weight.

clowadays, the alkali oxide content of Portland cement tends toincrease, as a consequence of the use of larger kilns, (for economical reasons), in which less ventilation is possible compared to smaller one; further a larger part of the dust, rich in alkali oxides, is recycled into the kiln, in order to reduce the pollution caused by this dust in the environment. The recent energy shortage (october 1973) and

the sharp increase in fuel costs have even accelerated the

trend towards higher alkali contents of Portland cement (1). Thealkali oxides are incorporated in the phases

existing in Portland cement (C

3A, c3

s,

s~c

2

s and c

4

AF)~ in varying quantities (2) • The alkali oxides also form seperate phases (sulphates) when reacting with the gypsum added, or

xin a part of this thesis the shorthand notation, applied in cement chemistry, is used: C

=

cao,

s

=

Sio₂, A

=

Al

2o3, F

=

Fe₂o₃, H

=

H₂o, N

=

Na₂0, K

=

K₂0; e.g.

C3A

=

3CaO.Al 203

=

Ca 3Al206 etc.

with

so

3 in the burned fuel. The other two phases which take up most of the alkali oxides are

c

₃A and

s-c

₂

s.

_{Na 2}

o

is incorporated into

c

₃A together with some Si0₂ (3%) to form an orthorombic phase with the approximate composition NC 8A3 or an intermediate solid solution (3). K₂

o

forms a solid solution with

s-c

₂

s

of composition

Kc

₂₃

s

₁₂•

The influence of the alkali oxides in Portland ~ement on its hydraulic properties has been the subject of numerous pape:cs (see e.g. ref.4-7). In general the alkali oxides in the clinker minerals and in alkali sulfates accelerate the cement hydration to give a paste with higher early

compressive strength (up to 3-7 days) as compared to a Portland cement paste containing little or no alkali oxides while at longer hydration times the compressive strength becomes lower (4, 5). Further the alkali oxides present in Portland cement used for making concrete, are capable of reacting with certain aggregatesused in it; this reaction mayxesult in local volume expansion, cracking, loss of

strength and in extreme cases even in complete destruction of the concrete (1).

In view of these effects, it is important to un~erstand which role the alkali oxides play in the hydration of

Portland cement and as a first step to elucidate the influence of alkali oxides on the hydration of its separate minerals.

A compound added to Portland cement may influence the hardening process and the ultimate strength of the paste by two effects: a) an influence on the mechanism by which the reaction proceeds leading to a different degree of conversion of the cement after some definite time, or to different

reaction products; b) changing the interaction between the hydrates formed during the hydration process. This thesis describes the influence of Na 2

o

on these two aspects of the hydration of c₃A.

The infleunce of Na2

o

on the reaction kinetics of the hydration of

c

₃A is studied by following the heat evolution rate in pastes and suspensions of

c

₃A in water and NaOH solutions and of Nxc 3_xA (x~0.25} in water. Both cases were

during Portland cement hydration as NaOH in the liquid phase or incorporated in the solid phase as Nxc

₃

_~. NaOH in the liquid phase is formed because Na2so4, present in the clinker, will go into solution almost instantaneously (4), when H_{2o is added; the so4}2- ions will be precipitated as ettringite (c_{3A.3Caso 4 .32H2o) or monosulfate}

(C

3A.Caso4.12H2o), leaving Na+ ions in solution with counter-ion OH- (7). This part of the investigatcounter-ion is described in two papers (8, 9), reprints of which are added to this thesis

(Chapter II en III).

The influence of Na 2

o

on the interaction between the hydrates formed during the c 3A hydration was studied using colloid chemical techniques such as ~-potential measurements on c

3AH6 (= Ca3Al2(0H)12> as formed during the hydration in

suspension of c₃A, and ~-potential measurements and coagula-tion experiments of c_{3AH 6 prepared in advance. This part of} the investigation is described in two additional papers; of these preprints are added (Chapter IV and V).

The ~ascription of these investigations are introduced by a general introduction into the hydration of c 3A.

some details, either of experimental techniques or of theoretical derivations, are included in appendices.

The reaction of

c

_{3A with water}

Among the minerals present in Portland cement c 3A is the most reactive towards water. Although present in

relatively small quantities (varying between 5-15%) (10), it has a large influence on the early hydration and the setting of Portland cement paste.

The hydration behaviour of c 3A is dependent on factors such as temperature (11, 12) water/solid ratio, specific surface of c

3A, the preparation method (13) and the size of the hydrating sample (14). If other compounds are present in the paste of c

3A with water, the reaction

mechanism can be changed significantly and some are therefore used in practice to retard or accelerate the setting of a

Portland cement paste. Examples of these compounds are

formed by the _{simultaneous hydration reaction of c 3s (27-29}.} If sedimentation is prevented in pastes C)A _{+ caso4 + H2o,} the revival is either only of minor importance or totally absent and after caso₄ exhaustion a stronger retardation of the c

3A hydration is found rather than a revival (.30, 31}. Three possible mechanisms have been proposed to explain the retardation of c₃A in early stages~

a) 'rhe solution in the space between c₃A and surrouding hydrates is filled with a solution saturated towards the anhydrous c₃A. The hydration rate is determined by the precipitation of the hydrates; ions such as so 42- lower the solubility of c 3A, thus causing a slow precipitation of ettringite (32, 33}.

b) c

3A reacts only where dislocations are present at the surface; the hydration is accompanied by movement of dislocations; ions adsorbed onto the surface impede this movement thus retarding the hydration.

c) A layer of hydrates around the c₃A grains, e.g. ettringite or hexagonal hydrates c₂AH_{8 and c 3AH6}, forms a prbtective layer impeding the diffusion of H₂o molecules tow~rds the unhydrated c₃A surface.

Whether one of these mechanisms gives an explanation' for the effect of the presence of Na₂o on the hydration of c₃A will bediscussed in Chapter III (34}.

Complete conversion of c₃A during hydration in pastes with water takes a long time; even in the presence of exess water, unhydrates c₃A is known to persist for a long period (14, 22). According to Stein (23, 35) this is due to Al(OH)₃ precipitation on the unhydrated c₃A grain during later stages of the reaction. At temperature >50°C c₃AH₆ is formed

directly at the c₃A surface, thus forming a strongly

retarding layer around the c₃A grains (11, 17, 30). However, complete conversion of c 3A. into c 3AH6 occurs in an autoclave at 17 5°C (36} •

The physico-mechanical properties of a c

3A + H2o paste are changed during the hydration process. At first, when

hexagonal hydrates are predominant, the strength of the paste increases because of the interlocking of these hydrates.

retarders such.as caso4 .2H2

o

(14-18) and ligna sulphonate

(19, 20) and accelerators such as cac1₂ (13; 16). The

influence of these and other compounds on the c

3A and

Portland cement hydration has been studied in an extensive way (cf. ref. 21).

When c₃A is brought into contact with pure water at .

25°C different types of hydrates are formed during the total hydration period. In the first minutes a layer of flakes and foils is formed consisting of poorly crystalline gellike

hydrates which cover the c 3A grains (9, 11, 14). Thereafter,

crystalline hexagonal platey crystals are formed with

+

-compositions C₂AH₈ (Ca₂Al(OH}

6 .Al(OH)4 .3H20) and c4AH13

(or c_4AH19> (ca

2Al4(0H) 6+.0H-.3H20), respectively (8, 9, 11,

14, 22-25). These crystals have a layered structure consisting of ca

2Al(OH)6

+

sheets with Al(OH)4- and OH-,

respectively in the interlayer together with H

2o (26).

Similar crystals with other anions such as Cl- and

so

42- are

formed during the c 3A hydration with cac12 or caso4.2H2

o

present (15, 16); in presence of the latter compound, a hydrate of totally different crystal structure and habit, viz. ettringite (ca

6Al 2 (0H) 12

cso

4)3.26H2

o),

can be formed as

we 11 ( 1 4 -1 7 ) •

This thesis is concerned with the hydration of c 3A in the absence ofother anions.Then the hexagonal platey crystals mentioned are metastable and are converted into the stable.

hydrate c_{3AH 6} (= _{ca3Al2 (0H) 12}

>,

which appears as

icositetrahedra (11, 20). The time, at which this conversion takes place, is strongly dependent on the conditions

mentioned before. The conversion is usually assumed to take place by a through solution mechanism (22, 24) since it has been found to be accelerated by the addition of nuclei of c

3AH6• It is accompanied by an increase in heat evolution

rate: this has been interpreted as an indication for a

retarding action exerted by the hexagonal plate layer present

on the c₃A in the initial stages. A similar revival of the

reaction of c 3A withwater is found in the presence of caso4 , when initially formed ettringite becomes metastable, at least when Ca(OH)2 is present as well (18) or when Ca(OH)2 is

After the conversion to the almost spherical c 3AH6 crystals the strength decreases again (14, 37). However, no detailed model is known at present for those processes determining the "strength" of such a paste.

References

1. S. Diamond, Cem. Concr. Res.~, 329 (1975).

2. H.W.W. Pollit and A.W. Brown, Proc. 5th. Inter. Symp. Chem. Cement, Vol.I, p.322, Tokyo (1968).

3. H. Maki, Cem. Concr. Res.], 295 (1973).

4. W.J. McCoy and O.L. Eshenour, J. Mater.], 684 (1968). 5. L.E. Copeland and D.L. Kontro in "The Chemistry of

Cements" (ed. H.F.W. Taylor), Academic Press, London and New York, Vol.I, p.314 (1964).

6. P.P. Budnikov, R.D. Azelitskaya and A.A. Lokot', Zh. Prikl. Kh •

.!1

1 953 (1968) (J. Appl. Chem. U.S.S.R •

.!1

1

906 (1968).

7.

w.

Lieber, zement Kalk Gips, 26, 75 (1973).

8. G.A.C.M. Spierings and H.N. Stein, Cem. Concr. Res. ~, 265 (1976).

9. G.A.C.M. Spierings and H.N. Stein, Cem. Concr. Res. ~'

487 (1976).

10.H.F.W. Taylor; The Chemistry of Cements, Academic Press, ~ondon and New York, Vol.I, p.3 (1964).

11.E. Breval, Cem. Concr. Res.~, 129 (1976).

12.B.S. Bobrov and A.S. Shlepenkov., Neorg. Mater.~, 1123 (1970).

13.P. Fierens, A. Verhaegen and J.P. Verhaegen, Cem. Concr. Res.

!

1 381 (1974).

14.A. Traetteberg and p .• E. Grattan-Bellew, J. Amer. Ceram. Soc. 58, 221 (1975).

15. A •. Traetteberg and P.J. Serada, Cem. Concr. Res.

!

1 461 (1976).

16. N. Tenoutasse1 Zement Kalk Gips,

12,

459 (1967).

17. R.F. Feldman and V.S. Ramachandran, Mag. Concr. Res. ~~ 185 (1966).

18. H.N. Stein, Rec. Trav. Chim.

JU,

881 {1962). 19. M.B. Milestone, Cem. Concr. Res.

!

1 89 (1976).

20. T.D. Ciach and E.G. swenson, Cem. Concr.,Res.

1,

143 (1971}.

21. K.E. Daugherty and M.Y. Kowalewski jr., Proc. 5th. Inter. Symp. Chem. Cement, Vol.IV1 p.4'21 Tokyo (1968).

22. H .N. Stein 1 J. Appl. Chem. (London) jl, 228 ( 1963) • 23. H.N. Stein, Special Report 90, Highway Research Board,

Washington, p.368 (1968).

24. R.F. Feldman and

v.s.

Ramachandran, J. Amer. Ceram. Soc. 49, 268 (1966).

25. J.F. Young, J. Amer. Ceram. Soc. 53, 65 (1970).

26, S.J. Ahmed and H.F.W. Taylor, Nature 215, 622 (1962). 27. J.G.M. de Jong, H.N. Stein and J.M. Stevels, Proc. 5th.

Inter. Symp. Chem. Cement , Vol.II, p.311, Tokyo (1968). 28. J.G.M. de Jong, Thesis, Eindhoven (1968).

29. W.A. Corstanje, Thesis, ~indhoven (1972).

30. H.N. Stein, J. Appl. Chem. (London)~' 314-325 (1965). 31. C.L.M. Holten and H.N. Stein, Cem. concr. Res.,

submitted for publication.

32. W. Lerch, Proc. Am. Soc. Test. Mat. 46, 1252 (1946). 33. P.K. Mehta, Cem. Concr. Res.

!

1 169 (1976).

34. G.A.C.M. Spierings and H.N. Stein, Klei Keram. 26, 99 (1976).

35. H.N. Stein, Chemisch Weekblad 62, 279 (1966).

36. T. Thorwaldson, W.G. Brown and C.R. Peaker, J. Am. Chem. Soc. 32, 910, 3927 (1930).

37. E.S. Solev'eva and E.E. Segalova, Kolloid. Zhur. ~' 621 (1958).

CHAPTER II

The Influence of Na2

o

on the Hydration of

c

₃A I. Paste Hydration

Reprinted from "Cement and Concrete Research" with permission from Pergamon Press, 1976.

Literature references of this chapter are to be found on page 25

CEMENT and CONCRETE RESEARCH. Vol. 6, pp. 265-272, 1976. Pergamon Press, Inc. Printed in the United States.

THE INFLUENCE OF Na 2o ON THE HYDRATION OF c_3A. I. PASTE HYDRATION

G.A.C.M. Spierings and H.N. Stein Laboratory of General Chemistry

Technological University, Eindhoven, The Netherlands

ABSTRACT

(Communicated by H. F. W. Taylor) (Received Dec. 2, 1975)

The influence of Na 0 on the hydration of c₃A was studied both by following t~e hydration of xNa

2

o.

(3-x)CaO.Al2o 3 (O'x'0.25) in water, and of c A in solutions of NaOH. Low NaOH concentrations preve~t a very early appearance of the second heat evolution peak, indicating a more controlled formation of c₃AH nuclei. Higher NaOH

concentrations advance the s~cond peak; this is ascribed to a decreased stability of the hexagonal hydrates with increasing NaOH concentrations.

Der Einfluss von Na 0 auf die Hydratation des c

3A wurde untersucht sowohl mfttels der Hydratation von

xNa2

o.

(3-x)CaO.Al₂

o

₃ (O'x~0.25) in Wasser und von

c

_{3A in}

NaOH-LOsungen. ·

Niedrige NaOH-Konzentration verhindern ein sehr frfihes Auftreten des zweiten Warmeentwicklungspeaks; dieses deutet auf eine besser beherrschte. Keimbildung des c_{3AH 6} hin. Hohere NaOH-Konzentrationen verfrlihen den zweieen Peak; dieseswird einer abnehrnenden Stabilitat der hexagonalen Hydrate mit steigender NaOH-Konzentration zugeschrieben.

265

266 Vol. 6, No. 2 G. A. C. M. Spierings, H. N. Stein

Introduction

The alkalies in a portland cement clinker have a distinct influence on the strength development of a cement paste prepared from it (1,2}. As a first step in understanding this effect the influence of alkalies on the hydration mechanism of portland cement minerals has to be studied.

The alkalies can be incorporated into a number of phases in

t~e clinker. Part of the Na o is normally taken up by the

c

A. Recent work (3,4} has shown that there exist several series ~f solid solutions of general formula xNa2o. (3-x}CaO.Al2o_{3 , of which} a cubic one with O<tx"0.08, anorthorhombicone with 0.1oc;x<0.20 and a monoclinic one with 0.20~x"0.25 are relevant to the present investigation. When 0.08<x<0.16 a mixture of two phases is found.

Some aspects of the influence of Na

2o on the hydration of

c

3A have been investigated (5-9) but no deta1led study has been

reported. The present investigation deals with the paste hydration of c 3A in solutions of NaOH and of xNa2

o.

(3-x)CaO.Al

2o3 in water. Experimental

Methods

Specific surface was determined by N

2 adsorption in an Areameter ( "Str~hlein") • Free lime was determined by the method of Pressler et al. {10). Scanning electron micrographs (SEM) were made using a Cambridge MK-2A instrument.

x-ray analysis was performed using a Philips diffractometer PW1010 with filtered Cu radiation; the quantities of the compounds present were estimated from the intensities of characteristic peaks, which are given in arbitrary units {vvw<vw<w<mw<m<ms<s<vs). The peaks used were the same as thos~ used by Corstanje, Stein and Stevels (11) together with the 10.7 X peak for c₂AH₈•

Isothermal calorimetry was performed at 25°C as described previously (12). The pastes were prepared and the hydration

reactions arrested as described by de Jong, Stein and Stevels (13). Calcium was determined by a spectrophotometric titration method

(Slanina et al. (15)). Sodium was determined using a sodium ion electrode {Swasey (16}).

Materials

The Na₂o containing

c

_{3A samples were prepared from mixtures} of Caco3 (p.a. Merck), Al2o 3 (U.C.B.; loss on ignition 0.57%) and Na 2co3 lp.a. Merck}. The searting materials were mixed in an agate ball mill, heated tree times in platinum crucibles for two hours at 1325°C with intermediate grinding, and sieved. The fractions with particles smaller than 36 ll m were used.

c

_{3A was}

prepared as described by de Jong, Stein and Stevels {13). Table I contains some data for the materials.

According to Regourd and Guinier(3}, C3A and N0 05

c

_{2 95}A are cubic whil~ _{N0 25}

c

2 75A is ~onoclinic, and N0

15

c

2

.a5A

a muture of the cub1c ana. ort'h.orhomb1c phases. The samp es were

characterized by their X-ray diffractograms which agreed completely with the data given by Regourd and Guinier.

Vol. 6, No. 2 267 C3A PASTES, ALKALI HYDROXIDE, HYDRATION

temperature shortly before use. The alkali hydroxide solutions were prepared from LiOH.H2o (Koch light >99%), NaOH and KOH _{(Titrisol Merck) and RbOH (Koch light >99,8%), The NaAl(OH)} solutions were prepared by dissolving Al ribbon (p.a. Merck)4in aqueous NaOH solution. All preparations were carried out using a glove box wi.th a N₂-atmosphere free from co

2•

TABLE I

Data on Materials Employed

Specific Free Ca/Na

lime su~f~ye cmg % Theoretical Analysis CA N3 C A N0.05C2.95A N0.15C2.85A 0.25 2.75 Results 3210 0.3 2890 <0.1 27.5 3060 0.3 9.5 2670 0.7 5.5 Hydration of c₃A in Solutions of Alkali Hydroxide or Hydroxoaluminate

28.1

1

o.

6 6.0

Fig. 1 gives typical heat evolution curves for pastes (w/s

=

1) made with

c

₃A and water or aqueous NaOH. In Fig. 2, the time

-020 1\

-

...

I I I I .... _{Ill I} I nl I

...

_I

.,.

_I iu 0.15 \ I 10.04 M I

..

I "'

....

I I

..

I 1; 0.10 I Q: I <= g _I .... _I

"

C) \ wo.os +'

...

•

:J:

Heat evolution rates of pastes {w/s

=

1) made using water and aqueous NaOH

""

..

,..

0 "'60 0

....

<: 0

"

..

..

"'

E i= 0 I I I I I I I I 0

:"8

l\

_{: 0}

j '

FIG. 2 Time of second heat evolution peak

8....__

268 Vo I. 6, No. 2

G. A. C. M. Spierings. H. N. Stein

of the second neat evolution peak is plotted against NaOH

concentration. Heat evolution in the first 15 minutes decreases with increas~ng NaOH concentrat~on. With water or NaOH less concentrated than about 0.5M, the t~me of the second peak is not very reproducible. With more concentrated NaOH, the

reproducibil~ty ~s much better, and the occurence of this peak at

very short times ~s prevented. The time of the second peak falls with concentration above 1M. Mor~ et al. (6\ also found this.

X-ray results (Table IIi show that C~AH₆ is formed within 1 C•

minutes and that its amount increases witfi t~me thereafter. At

shorttimes the amount is virtual ly independent of NaOH

concentration. Hexagonal hydrates were only once found (after

20h ~n 2N NaOH) with X-rays, but the SEM (Figs. 3 and 4) showed

the presence after 10 minutes of hydrates other than

c

₃AH_{6 .} These had a platey habit reminiscent of that of hexagonal hydrates. The absence of the characterist~c X-ray peaks indicates that hexagonal

hydrates, if formed, are much disordered.

The effect of varying the alkali cation was studied using 2M solutions. The only observed effect was a small decrease ~n heat

evolut~on rate over the whole period of hydration when the cation radius was increased.

Fig. 5 shows the effect of adding NaAl(OH) ₄ as well as NaOH. The second heat evolution peak is depressed and slightly retarded.

Discussion

Addition of NaOh lowers the solubility of Ca(OH) and

increases that of _{Al(OH) 3 .} Berger, Kotsupalo and Pushnyakova (16)

found that aqueous alkal~ partly decomposes c_{3AH 6} to give Ca(_{OH) 2},

c4AH13 and alurninate ~ons in solution, but Jones (17) found in a study of the C-A-N-H system that c3AH 6 is stable ~n 1% NaOH, even

FIG. 3

SEM _{of C3A} hydrated

for 10 minutes in water

FIG. 4

SEM of c3A hydrated fo1

Vol. 6, No. 2 ₂₆₉ C3A PASTES, ALKALI HYDROXIDE, HYDRATION

Table II x-ray Data on Paste Hydration of e A in water and NaOH solutions• NaOH Hydration

c

₃A C 2AH8 e3AH6 (M) _{time (min)} 0 10 vs 0 _{30 vs} 0 120 s 0.04 10 VS 0.2 1 0 vs 0.2 50 vs 0.2 1200 VS vvw

*Neither ea (OH) nor c_{4AHx was ever detected.}

vw m s vw vw mw m :;.06

_..

....

..

..

<>. "'

-

...

:lL04

..

~ « c 0 :;::; .02 3 0 > UJ ...,

..

"

:r 1 2 Time (h) FIG. 5

Heat evolution rates of pastes (w/s 1) containing both NaOH (2M) and NaAl(OH)₄

of concentrations shown at low aluminate ion concentrations. The result of Berger, Kotsupalo and Pushnyakova can be ascribed to the initial absence in the solution of Al(OH)(. It neve. rtheless seems strange that no solid ea(OH)₂ could be de~ected in the present work.

Retardation of the hydration reaction after the first peak is generally attributed to ~~rmation of a layer of hydrates, which impedes thepassage of ea and aluminate ions into solution. The SEM results (Figs, 3 and 4) confirm that such a layer is formed, However, misfit between e 3A and hexagonal hydrates as regards interatomic distances and habits will make it difficult to obtain a fit on an atomic scale, and some space will exist between the e A and the hydrate crystals. Some retardation mechanisms cdmpatible with the existance of such a space will be discussed in a later paper (18).

The existence of the second heat evolution peak is generally attributed to recrystallization of this layer. The effect of alkali in preventing the very early occurence of this peak can be attributed to retardation of the hydration of the e 3A which could be expected to cause a more dontrolled growth of e 3AH6 nuclei.

The fact that the amount of c 3AH6 formed in the first 10 minutes does not depend on NaOH concentration indicates that the

nucleation of e₃AH₆ is not markedly affected by the type of aluminate ion present in .the solution. The earlier appearance of the second peak at high alkali concentration might be caused by changes in the nucleus growth rate ofc₃AH

6 due to higher aluminate concentrations. However, the second peak ooes not occur sooner when NaAl(OH~- is added initially (Fig. 5)J therefore the earlier occurence of ~he second peak is attributed to changes in the rate at which Al (OH)4- passes into solution from the hexagonal hydrates.

270 Vol. 6. No. 2 G. A. C. M. Spierings. H. N. Stein

...

_...

.... "'

..

:o.oe 1"

..

..

""> -o.os .20.04 +' " 0 _> w .... 0.02

..

..

:c - - c₃A - - - ~.O!h.9s A -<>-'>-N C A 0.15 2.85 -x-ll-ND.25c2. 75 A FIG. 6

Heat evolution rates for pastes of water and Nxc 3_xA preparations

-

.

_....

..

"'

..

... 0.08

"'

2o.os

_..

....

..

n: .20.04 .... ~ 0 > w .... 0.02

..

"'

:c - - N0.05c2.95A + ~.15c2.85A ---- 'Na.asc2.9SA + No.2f2.1SA

---2 3 Time (h) FIG. 7

Heat evolution rates for pastes of water and mixtures of Nxc 3_xA pastes Hydration of xNa

2o. {3-x)CaO.Al2g3 Solid Solutions Figs. 6 and 7 give heat evolution curves for several preparations and mixtures thereof. As with hydration of

c

_{3A in} NaOH solutions, Na2

o

causes the first peak to occur later ~hough

no increase in the effect with Na₂o content was observed beyond

x 0.05 despite the variations in crystal structure. Mori et al.

(6) found a similar effect. The second peak occurred sooner at high Na 2o contents.

x-ray results (Table III) showed that c_{2AH 8 was formed} initially. The amount decreased after the second peak and none was found after 48 h. A SEM (Fig. 8) of a preparation with x

0.25 hydrated for 50 minutes showed particles coated with typical

hexagonal plates. When these Na

2o-containing phases hydrate alkali hydroxide accumulates in the solution. The concentration depends on x, the time or degree of hydration, and the amount of water left. At w/s

=

1 the OH- concentration could easily reach 0.5-1 M1 for example, if in a paste of N0 25c 2 75A that is 33% hydrat~d, and assuming no change in volume•oi e~e liquid phase, the OH

concentration is about 1M. The alkali will have &imilar effects to those found on hydration of pure c

3A in NaOH solutions but the situation is more complex because the alkali hydroxide

concentration in solution varies with time.

The. fact that the mixture of preparations with x

=

0.05 and x = 0.25 behaves in much the same way as the preparation with x

=

Vo l . 6, No. 2 271

C3A PASTES, ALKALI HYDROXIDE, HYDRATION

TABLE III

X-ray Data on Paste ~

Hydration of _{N0 _25c 2_75 A}

Hydration C₃A c_{2AH 8 C3AH6}

time (h)

0.83 vs m m

2.50 s w s

48 s vs

~C AH was not found.

4 X

FIG. 8

SEM of _{N0 _25}c2 _75 A hydrated for 50 min. in water

0.15 indicates that the effects are determined by the alkali hydroxide concentration in solution and not by any particular property of the solid phase; this is consistent with the earlier

statements about the second peak.

Acknowledgement

One of the authors (G.A.C.M. Spierings) gratefully acknowledges

financial support granted by the "ENCI Jubileumfonds".

References

1. L.E. Copeland and D.L. Kantro in H.F'.W. Taylor, The Chemistry of Cements, Vol. I, p. 315, 333. Academic Press, London and

New York (1956).

2. W.J. McCoy and O.L. Eshenour, Proc. 5th. Int. Symp. Chem.

Cement, Tokyo, 1968, Vol. II, p. 347.

3. M. Regourd and A. Guinier, Proc. 6th. Int. Symp. Chem. Cement,

Moscow, 1974 (Principal Paper).

4. I . Maki, Cem. Concr. Res., 2• 295 (1973) .

5. F.T. Vazques, Ion, l!• 372 (1971).

6. H. Mori, G. Sudoh, K. Minigishi and T. Ohta, Rev. 25th. Gen.

Meeting, Cem. Asoc. Japan, 1972, p. 52.

7. K. Murakami, T. Hirobumi and Y. Nakura, Chem. and Ind., 1968,

p. 1769.

8. V.R. Ryazin, Yu.s. Malinin and K.G. Kolenova, Tsement, 1972,

p. 20.

272 Vol. 6, No. 2 G. A. C. M. Spierings, H. N. Stein

9. T.A. Khryzhanovskaya, V.M. Mirak'yan, B.G. Shokotova and A.G. Kholodnyi, Tsement,

ll•

10 (1965).

10. E.E. Pressler,

s.

Brunauer and D.L. Kantro, Anal. Chem.

£!,

894 (1956).

11. W.A. Corstanje, H.N. Stein and J.M. Stevels, cem. Concr. Res., }_, 791 (1973).

12. H.N. Stein, J. Appl. Chem. (London),

..!1,

474 (1961).

13. J.G.M. de Jong, H.N. Stein and J.M. Stevels, J. Appl. Chem.

(London),~ 9 (1968).

14. J. Slanina, P. Vermeer, G. Mook, H.E.R. Reinders and J. Agterdenbos,

z.

Anal. Chem., 260, 354 (1972).

15. C.C. Swasey, Tappi, ~· 1692 (1970).

16. A.S. Berger, N.P. Kotsupalo and V.A. Pushnyakova, Proc. 6th. Int. Symp. Chem. Cement, Moscow, 1974, Section III,

suplementary paper III-4.

17. F.E. Jones, J. J?hys. Chem., 48, 379 (1944).

CHAPTER III

The Influence of Na 2

o

on the Hydration of

c

_3A

Reprinted from "Cement and Concrete Research" with permission from Pergamon Press, 1976.

Literature references of this chapter are to be found on page 38

CEMENT and CONCRETE RESEARCH. Vol. 6, pp. 487-496, 1976. Pergamon Press, Inc. Printed in the United States.

THE INFLUENCE OF Na₂

o

ON THE HYDRATION OF

c

₃A. II. SUSPENSION HYDRATION.

G.A.C.M. Spierings and H.N. Stein Laboratory of General Chemistry

Technological University, Eindhoven, The Netherlands

ABSTRACT

(Communicated by H. F. W. Taylor) (Received March 25, 1976)

The influence of Na2

o

on the hydration of

c

_{3A was}

studied in suspensions from the start of the reaction onwards. The heat evolution rate in very early stages of the hydration, measured at varying NaOH

concentrations, and SEM1indicate that at

NaOH concentrations larger then 0.1 M the reaction mechanism differs from that in water. In these

solutions the hydration is thought to be controlled at first by a more or less amorphous Ca(OH) 2 layer. Der Einfluss von Na₂o auf die Hydration des C3A wurde untersucht vom Anfang der Reaktion an. Die W!rme-entwicklungsgeschwindigkeit in den ersten Hydratations-stadien, gemessen bei verschiedenen NaOH-Konzentrationen, und SEM-Aufnahmen deuten auf einen anderen Reaktions-mechanismus bei NaOH-Konzentrationen grOsser als 0.1 M. -In diesen L6sungen ist die Hydratation im Anfang

beherscht duich eine mehr oder weniger amorfe Ca(OH) 2-Schicht.

487

488 Vol. 6. No. 4 G. A. C. M. Spierings, H. N. Stein

Introduction

In a previous paper (1} the influence of Na₂o on the hydration of C A in pastes was reported. To elucidate some of the aspects of3~e hydration mechanism, especially during the first minutes of hydration, C₃A hydration in suspension (w/s

=

100) was studied from the start of the hydration onwards using isoperibolic calorimetry.

Experimental

Methods for SEM, x-ray analyses and the arresting of hydration were as described previously (1). Conductivity measurements were performed as described by de Jong, Stein and Stevels (2).

The heat evolution rates in the suspensions were measured using a precision calorimeter LKB 8700-1. The principles of these measurements have been described by WadsB (3) • An isoperibolic calorimeter consists in principle of a nearly adiabatic reaction vessel in an environment of constant temperature. The total heat Q(t ) developed after a certain hydration time t₁ can be calculated from the following equation

( 4) : a &T (t) t 13 yaT(t)dt 0 t1J: Q(t1)

=

a&T(t1) + S

1

aT(t)dt, where 0

heat capacity of reaction vessel + contents, temperature difference between environment and reaction vessel after time t (presumably), total heat leak. during t₁ seconds.

&T(t) was calculated by measuring the change in resistance of a thermistor &R with a Wheatstone bridge. The heat leak. from the reaction vessel to the environment was found to be

proportional to &T(t).

The experiments were conducted with an environment

temperature of 25° + o.o1oc and the maximum temperature rise in any experiment was 0.7oc.

The materials used and methods for determining sodium and calcium were those described previously (1). Aluminium was determined as described by Pribil and Vesely (5).

Hydration of c 3A in NaOH Solutions in Suspensions Results

Fig. 1 shows the cumulative heat evolution for c₃A hydrated in suspension in water and NaOH solutions (w/s

=

100). Fig. 2 shows the heat evolution rate obtained by differentiating the curves in Fig. 1. In all cases the heat evolution rate is high during the first seconds and decreases rapidly as hydration proceeds. At higher pH (>0.1M NaOH) a second peak. in the heat evolution rate appears which is accelerated by increasing NaOH concentrations.

Vol. 6, No. 4

600

C3A, HYDRATION, NaOH SOLUTION

-·-~o - • - 0.1 M NaOH - • - 0.4 M NaOH - o - 2 M NaOH - • - 4 M NaOH

.

10 30 60 (min) Time FIG.

Cumulative heat evolution from c 3A hydrated in water and NaOH solutions

Heat evolution rates of CJA hydrated in water and NaOH solutions

489

90

490 Vol. 6, No. 4 G. A. C. M. Spierings, H. N. Stein

I

1 2 3

the NaOH concentration. This "initial" heat evolution rate is the maximum rate of

increase of the thermistor temperature, which is reached a few secondsafter the first contact between the reactants. A time lag arises because of the delay in the response of the thermistor.

At low NaOH concentration {<0.1M) the initial heat evolution rate decreases rapidly with increasing NaOH concentration while higher NaOH concentrations effect only a minor additional decrease. x-ray analyses of the hydration products are shown in Table I.

NaOH Concentration

I.

(M) _{the hydration products on the} Fig. 4 shows SEM's of surfaces of particles after 10 seconds and 10 minutes hydration. In water without NaOH addition the plate-like hydrates are, in many cases, FIG. 3

Initial heat evolution rate versus the NaOH concentration perpendicularly to the

predominantly parallel

x-ray Data for NaOH Hydration time

(M) (min)

0 10

0.04 10

2 10

2 1200

orientated more or less surface of a particle instead of the orientation in paste hydration (1).

TABLE L hydration of C

3A in NaOH solutions C2AH8 C4AH13 c_3AH6 c_3A

w w vs

vw vw vs

w vvw vs

s s

Fig. 5 shows the change in conductivity of C3A suspensions during the hydration in H₂o and dilute NaOH solutions

Discussion

The decrease in conductivity (Fig. 5) after a certain

hydration time (5.7 h for a suspension of

c

_3A in water) indicates the point during the hydration at which the conversion of

hexagonal hydrates into c_{3AH 6 becomes important enough to cause} the solution to leave the metastable point of coexistence of c_{2AH8 ,} c₄AH

13 and aqueous solution (9). Both the conductivity experiments and the behaviour of the second heat evolution peak shows that this conversion is accelerated when the NaOH

Vol. 6, No. 4

C3A, HYDRATION, NaOH SOLUTION 491

a

c d

e f

FIG. 4

SEM's of C₃A hydrated in suspension (w/s - 100) for a 10 sec in H 0 d 1 (J m in in H 0

b 1 0 sec in 0~1M NaOh e 10 m in in 0~1M NaOH

-

_{c 10 sec in 2M NaOH f 1 0 m in in 2M NaOH}

492 Vetl. 6, No. 4

,

~---G. A. C. M. Spierings. H. N. Stein Lnc

.3i

;.:a

__

,

~

lJ

·s:

+:: 0.01 M NaOH u 3 - - - . . .

~

r----

\

I

.

~-....

~

2

···-·-.~~2M.~~o~-·'\

\ ·"'• / 2 V 'I - · - · '-.. I

'j

.'-

-Q. ·-... (fl · · · · -CII 1 ~ \ 0.04M NaOH " ' _ . v ' l ' l v v " 7 v v ' V -f r " - v - v - " ' ~-·-•-•-•-•-•-•-•-•-~ .\ ,/""' 2 4 6 Time (min) '\ 1 / 2 3 4 "-...,./Q.1M NaOH Time (h) 5 6 7 FIG, 5

Increase in specific conductivity with regard to initial conductivity of suspensions of

c

₃A in water and aqueous NaOH concentration increases. In 0.1M NaOH, the conductivity

increases at the time of the second heat evolution peak (compare Figs. 2 and 5). For suspensions without NaOH, a very early appearance of the second peak and the accompanying conductivity decrease that were observed in pastes (1) were not found, presumably because of difficulty in nucleation of C3AH6 • Thus, the influence of NaOH on the second peak in suspensions is always an accelerating one.

In the early hydration stages, the heat evolution rate in pastes is about equal to that in suspensions after the same time. This is remarkable in view of the different orientation of the platey crystals in the hydrate layer in the two cases, and rules out a direct screening action of the hydrates.

The SEM results show that a hydrate layer exists on the

c

_{3A surface.} Because of misfit between the latter and the hydrates, some space will exist between them. Three possible explanations of retardation of

c

₃A hydration will be considered.

1) This space is filled by an aqueous solution saturated towards the anhydrous

c

₃A, the hydration rate of the

c

_{3A then}

being determined either by the diffusion of ions out of the space concerned and of water into it, or by the growth of the hexagonal hydrates in the layer.

2) Feldman and Ramachandran (10) proposed that the hydration of CA is impeded,not by hydrates, but by the blocking of

moy;ffients of surface dislocations by adsorption of ions, such as OH • In this case, the second peak would be caused by the

Vol. 6, No. 4 493 CjA, HYDRATION, NaOH SOLUTION

recrystallization of hexagonal hydrates because the latter is accompanied by some changes in the concentrations of these ions,

3) An aqueous solution occurs in the space between the c A and the hexagonal hydrates, in which conditions differ from th~se in the outside solution: this leads to the formation in this space of an amorphous solid that can match the C A surface on an atomic scale more closely than can a hexagonal h~drate (11). The role of the hexagonal hydrates is then an isolating one, and the rate determining step is the transport between the crystals of the hexagonal hydrates of ions, which tend to attack the retarding agent.

Regarding these three reaction mechanisms the following remarks can be made:

According to the "saturated solution" mechanism the heat evolution rate at t + 0 corresponds to a situation in which the dissolution of

c

₃A is not hindered by hydrates, and follows the equation

2+

-c₃A + 6H₂0 + 3Ca (aq) + 2Al(OH)₄ (aq) + 40H-{aq) However, the dependence of the heat evolution rate at t + 0 on the NaOH concentration is too weak, especially at higher concentrations, to be compatible with the mechanism, at least if equal volumes of liquid are assumed to be saturated per unit of time in all cases. This mechanism could account for the facts only if with increasing NaOH concentration either a larger volume would be accessible to saturation by

c

₃A, or increasing

contributions of hydrate precipitation to the heat evolved at t + 0 are assumed, Both alternatives are improbable; the former because large NaOH concentrations promote precipitation of

,Ca{OH)₂, the latter because this precipitation is an endothermic ·process.

According to the mechanism postulated by Feldman and

Ramachandran (10), the influence of increasing NaOH concentrations at t + 0 is caused by increasing difficulty in movement of

dislocations; the levelling off with higher NaOH_concentrations is due to all surface sites being occupied by OH ions, the decrease of the heat evolution rate is due to gradual exhaustion of surface dislocations, and the second peak can be ascribed to a decrease in OH- concentration, as has indeed been observed in suspensions at that particular stage of reaction by

de Jong, Stein and Stevels (9). However, this effect is small (the pH changes from 12.2 to 12.1), especially when compared with the OH- concentrations p.resent, for instance, in 4M NaOH where the second peak remains quite remarkable (Fig. 2). Thus, the data do not support a surface oislocation movement inhibition by absorbed OH- ions.

The following mechanism appears to be compatible with the results: In NaOH solutions of high concentration the amount of ca2+ going into solution as a result of hydration of the C3A must be small. Aluminate ions may go into solution bu~ ca2+

ions stay behind, their charge being compensated by OH ions. Rearrangement into crystalline Ca(OH)2 is prevented by the adjacent

c

_3A. The resulting primary layer t~us formed prevents contact between

c

_{3A and the solution, and can follow the c 3A}

494 Vol. 6, No. 4

G~ A. C. M. Spierings, H. N. Stein

surface closely because of its in situ formation. Figs. 4a-c show.that after 10 seconds hydration in water a structure consis.ting of plates reminiscent of the hexagonal hydrates is formed, while in 2M NaOH a dense layer is formed which might be capable of preyenting contact between the

c

₃A and the solution. Some H₂0and OH will penetrate through this_layer and react with aluminate ions from the C3A_to form Al(OH)₄ • These aluminate ions replace part of the OH ions in the layer, and at some distance from the

c

₃A rearrangement into hexagonal hydrates may occur, as is shown by X-ray analyses (Table I} and SEM

(Fig. 4d-f).

The decrease in the heat evolution rate with time (Fig. 2) is ascribed to the layer becoming thicker, and the second peak to recrystallization of the hydrates into c₃AH

6, which affects the concentration gradients in the vicinity of the

c

3A

sufficiently to dsstabilize the primary layer. This hypothesis would explain the absence of x-ray reflections due to crystalline Ca(OH) 2 and the presence of only weak lines due to the hexagonal hydrates (Table I}.

It appears from the conductivity curve that,after the formation of the layer seen in Fig. 4b,some hydrate nucleates in the surrounding liquid. This is shown by the conductivity maximum found after about 1 minute (Fig. 5 r the SEM shown in Fig. 4b was taken after 10 sec., which is before the conductivity maximum). This supports the existence of two hydrates:. one at the C₃A surface, a second one somewhat further away from it. The second one is the one seen in Figs. 4e-f. and is thought to be a disordered form of one of the hexagonal hydrates playing an isolating rather than a retarding role.

The mechanism described is not necessarily the mechanism operative in H2o or in Ca(OH)2 solutions. On the contrary, the _{shape of theheat evolution rate at t ~ 0 versus the NaOH} concentration graph (Fig. 3) suggests that a different mechanism becomes operative when the NaOH concentration exceeds 0.1M.

Hydration of 0.25Na

2o.2.75CaO.Al2£3 in Suspensions Results

The total heat developed up to a certain hydration time and the heat evolution rate at the time concerned are both shown in Fig. 6. Thetotal heat liberated after a certain hydration time and the heatllberation rate during the No.2 5

c

2 • 75A hydration are somewhat lower than for

c

3A {Figs. 1 and 2J,

In Fig. 7 the conductivity of a suspension of Nb₀~₅c₂•₇₅A is plotted against time. At certain times designated y

A, B, C and D the hydration was arrested and X-ray analyses of the solid phases and chemical analyses of the liquid phase were performed. Table II shows the results of these analyses. Discussion

When N0 25

c

2 75A is hydrated in suspensions with w/s

=

100 no second peaR in'tne heat evolution rate curve appears (Fig. 7). The total heat developed up to a certain hydration time is only slightly less than the total heat developed during the hydration

Vol. 6. No. 4

FIG. 6 Cumulative heat evolution and heat evolution rate during the hydra-tion in suspension (w/s

=

100) of N0.2Sc2.7SA in H20 FIG. 7 Specific conducti-vity of a suspen-sion (w/s = 100) of N0.25C2.75A in H20

C3A, HYDRATION. NaOH SOLUTION

1 TABLE II 2 3 4 Time 495 2 4 6 Time (min) 5 6 7 (h)

X-ray and chemical analyses data for hydration of N0 •25c₂•₇₅A in H₂o

Tiine concentration in mmol/l

(min)C₂AH_{8 C4AH19 C3AH6 N0.25c2.75A ea} 2+ Al(OH)4 Na+

A 3 vw vs 1.54 1.47 0.58

B 32 w vw s 0.62 0.75 1.22

c

200 s m m w 0.60 0.91 1.86

D 410 s vs vw 0.42 1.96 2.40

of c 3A. Thus the hydration characteristics of No.25c2.7SA in _{suspensions are almost similar to those of C3A, while in pastes} there exists a significant difference (1). The similarity between the hydration in suspension of

c

_{3A and No.2sC2.7SA is to}

496 Vol. 6. No. 4 G. A. C. M. Spierings. H. N. Stein

be expected in view of the small NaOH concentrations eff$ct~d by the former (Fig. 2). Again it appears (see ref. 1) that th~ NaOH concentration in the solution is more important than changes in solid state properties.

References

1. G.A.C.M. Spierings and H.N. Stein, to be published. 2. J. de Jong, H.N. Stein and J.M. Stevels, J, Appl. Chem.

(London),~' 9 (1968).

3. J. WadsO, Science Tools, 13, 33 (1966). (Chem. Abstr. 69,

30771U).

-4. S.R. Domen and P.J. Lamperti, J. Res. Nat. Bur. Stand,, 78A, 595 (1974).

5. R. Pribil and V. Vesely, Chem. Listy, 63, 1217 (1969). (Chem. Abstr. 72, 6243V),

6. H. N. Stein, .Special Report 90, Highway Research Board, Washington D.C., p.368 (1966).

7. Gmelin, Handbuch der Anorganischen Chemie, Calcium Teil B, p.1248, Chemie Verlag, Berlin (1956).

8. F.E. Jones, J. Phys. Chem., ~. 379 (1944).

9. J.G.M. de Jong, H,N. Stein and J.M. Stevels, J. Appl. Cht;m~. (London),~, 25 (1969).

10. R.F. Feldman and V.S. Ramachandran, Mag. Concr. Res., 18 185 (1966).

11. W.A. corstanje, H.N. Stein and J.M. Stevels, cem. Concr. Res., 2_, 791 (1973).

CHAPTER IV

The Electrokinetic Properties of Calcium Aluminate Hydrates

Reprinted from "Colloid and Polymer Science" with permission from Steinkopf Verlag 1976.

Literature references of this chapter are to be foun4 on page 52

ELECTROKINETIC. PROPERTIES OF CALCIUM ALUMINATE HYDRATES.

Introduction.

Calcium aluminate hydrates are formed when ca

3Al2

o

6, a compow.d present in Portland cement, reacts with water (1-3). At 25°C two metastable hydrates ca

2

Al(OH)~.OH-.3H

2

o and

+

-ca2Al(OH) 6.Al(OH) 4 .3H2

o

are formed in the first stage of the reaction; thesehydrates, present as hexagonal plates, have a structure consisting of ca

2

Al(OH)~ sheets with H₂

o

and respectively OH- and Al(OH)~ in the interlayer (4). These hexagonal hydrates recrystallize after some time, depending on the conditions (temperature, water/solid ratio, surface area of reactants) into the cubic ca3Al 2 (0H) 12 , the stable hydrate under _these conditions, which is (in small amounts) already present in the first stage together with the

hexagonal hydrates.

In the past little attention has been paid to the study of the electrokinetic properties of the calcium aluminate hydrates and other hydrates present in hydrating cement (5,6) although these properties will influence the

physico-mech~li9al properties of these systems.

We studied the electrokinetic properties of the calcium alum~nate hydrates and the influence of NaOH on them in the course of an investigation of the influence of Na 2

o

on the react_ion of ca3Al2

o

_{6 with water (7, 8) described in Chapter}

II and III. Experimental.

ca

3Al2

o

6 was prepared as described by de Jong et al (9). The water used was distilled twice and redistilled under reduced pressure shortly prior to use. ca3Al2 (0H) 12 was prepared in an autoclave from ca3Al2

o

6 and water as described by Thorvaldson et al (10). NaOH solutions were prepared from titrisol NaOH (Merck). All preparations were performed in a glove bow with an N2-atmosphere free of C02.

Water and NaOH solutions saturated towards ca

3Al2(0H)12 were prepared by dispersing ca₃Al₂(0H)₁₂ in the solution in

a polythene vessel. This vessel was kept in a co2-free atmosphere and shaken mechanically twice daily for 30 minutes. After 3 weeks the dispersion was filtered, with exclusion of co₂, through a micropore membrane filter with pore size 0.08 pm (Shandon Nuclepore N 008). This filter was used for other filtrations as well.

ca_{3Al 2 (0H) 12 suspensions (0.1 g/1) were prepared by} dispersing the material in solutions previously saturated towards it, using an ultrasonic bath (Megason Ultrasonic, Frequency 80.000 Hz) for 30 minutes.

The reaction of ca₃Al₂

o

₆ (1 g) with water and NaOH solutions (100 ml) took place in a polythene vessel at 25°

+ 0.1°c, with continued stirring. At predetermined times during the reaction a sample of 11 ml was taken and left standing for 30 minutes to allow the much larger unreacted Ca_{3Al 2o6 particles to settle. The} ~-potential of the hydrate crystals present in suspension after the settling was

electrophoretically using a Smith and Lisse cell (11). The Helmholtz-Smulochowski relation was used for the calculation of the ~;-potential from the electrophoretic mobility; this is justified by the fact that the radius of the almost spherical icositetrahedral ca_{3Al 2 (0H) 12 particles as determined by SEM,} ranges from 100 to 1000 nm, while the thickness of the double

-1

layer K ranges from 1 to 2,5 nm (12). In the case of the hexagonal hydrates the thickness of the plates was about. 200 nm. These values also justify the assumption of a flat double layer (see later).

Calcium was determined in the liquid phase after filtration using a spectrophotometric titration method as described by Smit and Stein (13). Aluminium was determined as described by Pribil and Vesely (14.).

x-ray analyses were performed using a Philips diffracto-meter PW 1120 with filtered Ca radiation. Scanning electron micrographs (SEM) were made using a Cambridge MK-2A

Theoretical.

CatouZation of the aurfaae aharge behind the eZeatro-kinetio stipping ptane.

The formula for calculating the surface charge behind

the electrokinetic slipping plane from the ~-potential as

derived from the Gouy-Chapman theory of the diffuse double layer is based on the assumption that the concentration of each ion in the diffuse double layer is determined by the electrical macropotential only. However, using the equality of the electrochemical potential of each type of ions,

throughout the double layer, the number n. of these ions as

J.

function of the distance to the slipping plane is given by:

n·=

tr'n~expL

zielP)

I

fi

I ~

kT

[ 11

where fi is the activity coefficient and the index ~

indicates the value of the quantity in the bulk of the

liquid. The other symbols have their usual meaning. Then,the

surface charge cr~ for flat double layers is given by:

,00

ljJ

1

Or={2et,e:re~zi~ ~~~ exptz~eT

)d$)}

2

[2]

!::, I 0 I

The ration f':'/f. can not be determined experimentally.

J. J. .

However, it may differ significantly from 1 because the activity coefficient is determined primarily by the

atmosphere of ions of opposite charge around any anion. This atmosphere near the interface differs from that in the bulk solution: If the solid has, say, a negative surface charge, anions will be all but absent in its vicinity; the cations will have, near the interface, an activity coefficient higher and the anions an acticity coefficient lower than in the bulk solution. The net effect is a decrease in absolute

value of cr~ as compared with that calculated on the

assumption of constant activity coefficients throughout the diffuse double layer.

Because it is essential in the discussion, whether cr~

increases with increasing NaOH concentration (see later), we estimated this effect, comparing cr~ calculated on the

assumption f~/fi

=

1 throughout the diffuse double layer with cr calculated on the following assumptions. For the

region~O<X~K-

1 _{(x =distance from the slipping plane, K =} the Debye Huckel parameter) fi was taken equal to fi at x

-1 00

07 for x>K , fi was taken to be= fi. Thus:

00 ~ 00

q,

1

Or=

{2Eat:rkT2:

n~

[

-4

exp(- zker

)+(1-~

)exp(- zi:T11r)-1]}

2

[3]

~ j

fi

fi

I

0 -1

where fi

=

f~ at x = 0 and ifiK = ifi at x

=

K ; ifiK was

calculated from the equation for ifi(x) for an electrolyte consisting of uni- and bivalent cations and univalent anions:

zoo

2

2

e n

2,...

where

Y

=

EoErkT

1 this equation can be derived fro, the Poisson-Boltzmann relation.

Activity coefficients at x = 0 were calculated from the Debye-Huckel equation as given by Gimblett and Monk (15); the ionic strength I used in this calculation was taken proportional to the concentration of univalent ions of opposite charge :

wherei+(x

=

0) is the ionic strength used for calculating the activity coefficients of the cations etc.

In both equations

[4]

and

[s] ,

concentrations 'V{ere calculated with f~/fi

=

17 the approximation involved was considered to be a second order effect.

The procedure may be illustrated by Fig.1, showing for

~

=

-17.4 mv at [Na+]

=

10-2M and [ca 2+]

=

2,25x10-3M the course of: a) the potential as calculated from

[4] :

b) the

f+

-

9)> > n E t+

:or

"""":.12

f_

;::;: ... '< I'll .8 0 ... c: 0 _:!~9

_...

Ill 0 ... a. 0 1~ t+

Distance to slipping plane (nm)

FIG.1

The potential and activity coefficients in the liquid phase f o r t = -17,4 mV, [ca2+]

=

2,24 x 10-3M and [Na+]

=

10-2M.

~calculated activity coefficient.

----activity coefficient used for the calculation of at.

FIG.2

The t-po~ential of the calcium aluminate hydrates during the reaction of ca₃Al₂

o

₆ with NaOH solutions versus the NaOH con-centration.

I : reaction stage I (both hexagonal hydrates and Ca₃Al₂(0H)₁₂).

II: reaction stage II (Ca₃Al₂(0H)₁₂). I Vl20 10 0.0 NaOH 45

activity coefficients as calculated from the Debye-HUckel relation using [5] for the ionic strength; c) the activity coefficients as used in the calculation of az;.

Results and discussion.

The ~;-potential of the hydrates in the two stages of the reaction of ca

3Al2

o

6 with water and NaOH solutions is shown in Fig.2. Because of the influence of NaOH on the time at which the reaction stages occur (8), samples were taken after different times according to the NaOH concentration

(Table I) •

TABLE I.

NaOH concentration

Concentration mmol.l -1

(M) stage time ca 2+ _Al(OH)~

0 I 4 h 11 • 6 2.5 0 I I 8.33 h 6.6 4.5 0.01 I 2.33 h 7.7 2.2 0.01 I I 5.50 h 3.4 2.8 0.02 I 2 h 5.3 3.1 0.02 I I 4.83 h 2.5 2.4 0.04 I 1 h 2.2 4.1 0.04 I I 4 h 2.0 2.0 0.1 I 0.83 h 0.9 6.1 0.1 I I 3.50 h 0.9 2.2

The experiments shown in Fig.2. in stage I all refer to the !;-potential of the hexagonal hydrates; in the electro-phoresis cell these hydrates were easily distinguishable as platey crystals from ca₃Al₂(0H)₁₂• No difference, however, was found in the behaviour of the two types of hydrates. In the second stage only ca₃Al₂(0H)₁₂ is present, as evidenced by X-ray analysis.