Study the Hydration and the Microstructure of Cement Blended with ...

第 32 卷

武

第 17 期

汉

理

工

大

学

学

报

Vo l. 32 N o. 17 Sept. 2010

JOURNAL OF WUHAN UNIVERSITY OF TECHNOLOGY

2010 年 9 月

DOI: 10. 3963/ j. issn. 1671 4431. 2010. 17. 053

Study the Hydration and the Microstructure of Cement Blended with Rice Husk Ash in Comparison with Silica Fume N G UYEN V T 1 , Y E G 1 , V A N B K 1 , B UI D D 2 ( 1. Faculty of Civil Engineer ing and G eosciences, Delft U niversit y of T echnolog y, P . O. Box 5048, 2600 GA, Delft, T he N etherlands; 2. Department of Building M aterials, Hanoi U niversity of Civ il Eng ineering, Hanoi 700803, Vietnam)

Abstract:

R ice husk ask ( RHA) is not a ultra fine material as silica fume ( SF) , but possesses a very hig h specific surface

area because of its porous structure. W ith the similar chemical composition of RHA and SF, t he activ ity of RHA, t herefor e, is different from that of SF . T he objective of this w ork is to study the hy dration and the micr ostructur e of Por tland cement blend ed w ith RHA in compar ison w ith SF . T he test results show that SF r efined the pore structure of cement paste better than RHA . However, t he effect of RHA on cement hydration is more pronouned t han that of SF for the mixture w ith low w ater to binder r atio.

Key words: ash;

CLC number:

1

hydration;

isothermal calo rimetry;

mer cury intrusio n porosimetr y ( M I P) ;

microstructure;

T U 52

Document code:

A

Article ID: 1671 4431( 2010) 17 0252 04

Introduction Silica F ume ( SF) and Rice Husk Ash ( RHA) are classified as highly active pozzolans

SF

rice husk

silica fume

[ 2]

[ 1]

. Sim ilar to t he

, w hen RHA is incorporated in cement, it aff ect s the rat e and t he ext ent of heat generat ion[ 3] . Besides,

RHA has been used successfully replacing SF for production of hig h strengt h/ performance concret e[ 4, 5] . T he reactivity of RHA is att ribut ed t o it s high cont ent of amorphous silica, and to its very larg e surface area g overned by t he microporous structure of individual particles. Meanwhile, the react ivity of SF wit h the similar chemical composit ion comes most ly from t he surface area of t he very fine particles[ 6] . T herefore, RHA might be expect ed to behave dif ferent ly from SF as far as hy drat ion and microst ruct ure of cement are concerned. T he object ive of t his paper is to st udy t he effects of RHA and SF on t he hydration of Portland cement by means of isothermal calorimetry. Besides, the eff ect of RHA and SF on t he microst ruct ure of cement w as also e valuat ed by means of M ercury Inst rusion Porosimet ry ( M IP) .

2 2. 1

Experimental Materials T he materials used in this st udy w ere Port land cement , CEM I 52. 5N; condensed silica fume; rice husk

R eceived date: 2010 05 29. Biogr aphy : N GU YEN V T ( 1977 ) , P h D Candidate. E mail: T . vanNguyen@ tudelft. nl

第 32 卷

第 17 期

NGU Y EN V T , Y E G, VA N B K , et al: Study the Hy dration and the M icrostructur e % %

253

ash; and a polycarbox ilic acid based superplast icizer, Glenium 51, w ith 35% solid content by weig ht . T he SF possesses an amorphous SiO2 cont ent of 97. 2% and its mean part icle size about 0. 1~ 0. 15 m. Rice husk, an a g ricult ural waste material from Viet nam, w as burnt in a drum incinerat or. Det ails of the oven and rice husk combust ion process were described elsew here[ 7] . T he obtained ash w as ground in a vibrat ing ball mill for 90 min. T he ash cont ains 88. 0% amorphous SiO 2 , 3. 8% loss on ignit ion and its mean part icle size is 5. 6 m. 2. 2

Experiment description

T o study t he hydration of cement blended w it h RHA and SF , the isot hermal heat w as measured by T AM Air isothermal calorimet er. Experiment w as done on t he sam ples w ith w at er t o binder ( w / b) ratio of 0. 25, 0. 40 and 0. 60 at 20 ! . T he duration of all t est s w as 120 h. Bef ore mixing, all the m aterials were placed in an oven at least 24 h at the desired testing t em perat ure. T he weig ht of cement paste for the isot hermal measure ment s w as ( 10 ∀ 0. 01) g. A constant dosage of superplast icizer ( 0. 8% of the cement mass) w as employed for past es w ith w / b rat io of 0. 25. Table 1 Mixture proportions and abbreviation of

T he M IP measurements w ere carried out

samples used in the study

in a room w it h the t emperat ure of 16 ! . T he cont act angle of 139# and t he surf ace tension

M ix proportion Cement/ SF / RHA

of 0. 485 N/ m w ere used. T he int rusion pres sure goes from 0. 004 M Pa t o 210 M Pa. T he Washburn equat ion w as used t o calculate t he diameter of pores int ruded at each pressure st ep [ 8] . In t his w ay , t he pores in t he range of 350 m to 0. 007 m could be det ected. T he

Abbreviation w/ b = 0. 25 w/ b = 0. 40 w/ b = 0. 60

100/ 0/ 0

1R EF

2REF

3REF

90/ 10/ 0

1RHA10

2RHA10

3RHA10

90/ 0/ 10

1SF10

2SF 10

3SF 10

80/ 20/ 0

1RHA20

2RHA20

3RHA20

80/ 0/ 20

1SF20

2SF 20

3SF 20

sample preparat ion and measurement procedure of M IP w ere described in detail in reference [ 9] . 2. 3

Mixture proportions T he mixt ure proportions and abbreviation of samples used in t his st udy are present ed in T able 1.

3

Results and Discussion

3. 1

Heat evolution Cement hydration is an ex othermal process. T he isothermal calorimeter records t he heat release or the t her

mal pow er. T he deg ree of hy drat ion at t ime t can be approx imated as t he heat released at t ime t divided by t he max imum heat[ 10] , viz. ( t ) = Q( t ) / Q max

( 1)

w here, Q( t ) is the heat librat ed at t ime t , Q m ax is t he maximum heat w hen all t he cement has hydrated. T he max imum heat of t he binder ( cement + SF or cement + RHA) is calculat ed as follow s Q m ax = P C em ∃ Q Cem + P Poz. Q Poz

( 2)

P Cem, P Poz is t he percentage of cement and pozzolan ( SF or RHA) , respect ively ; Q Cem is the max imum heat of cement hydrat ion calculated according to Bogue

[ 10]

; Q Poz is the maximum heat of SF or RHA calculated from t he

amount of amorphous SiO2 in SF or RHA, w here Q SiO 2 = 780 J/ g [ 11] . 3. 2

The effect of RHA and SF on the hydration process T he result s of heat evolution and t he hy drat ion deg ree for t he sam ples w it h w / b of 0. 25, 0. 40 and 0. 60 at

20 ! are show n in Fig. 1, Fig. 2, Fig. 3, respect ively. Each f igure show s t he heat evolution on the left and t he hydrat ion deg ree on t he right. T he result show s that , com pared t o t he control sample w it hout RHA and SF, a low er and a later accelera t ion peak of t he heat evolution curve of t he RH A sample is observed ( Fig. 1( a) , Fig. 2( a) , Fig. 3( a) ) . H owev er, in case of SF sample, t his peak is earlier and higher. At 20% replacement , t he ext ra peak is clear on t he

254

武

汉

理

工

大

学

学

报

2010 年 9 月

heat evolut ion curve for the RHA sample and the SF sample at about 17 h and 12 h, respect ively ( Fig. 3( a) ) . T his indicat es that t he react ion of RHA and SF w it h cement hydration product att ributes t he ex tent of the heat g enerat ion of cement hydration.

F or cement blended w it h SF and RHA, t he hy drat ion deg ree of t he RHA sample is higher t han t hat of t he SF sam ple w it h w/ b rat io of 0. 25, especially f or 20% replacement percent age ( F ig. 1( b) ) . T his behavior changes gradually w ith an increase of w / b rat io. Wit h w / b ratio of 0. 40 and a given replacement percentage, the hydrat ion deg ree of both RHA and SF samples is similar ( Fig. 2( b) ) . How ever, t he hydrat ion degree of t he RHA sample is low er than t hat of the SF sample w ith w / b rat io of 0. 60 ( F ig. 3( b) ) . F rom t his result , it can be concluded t hat the eff ect of RHA on cem ent hydration is more pronounced than that of SF f or the mixt ures w it h low w / b rat ios. 3. 3

The effect of RHA and SF on the pore structure of cement paste In this part, only t he sam ples w it h w / b rat io of 0. 25 w as evaluated. F ig . 4 show s t he pore structure of t he

cont rol sample and samples incorporat ing RHA or SF at t he age of 28 d. Compared to t he cont rol sample wit hout RHA and SF, t he addit ion of both RHA and SF ref ines significant ly the pore structure of cement paste and reduces t he tot al porosit y in cement paste, in w hich the eff ect of SF is larger than t hat of RHA. T his is underst andable because the mean part icle size of SF is about 0. 1~ 0. 15 m,

第 32 卷

第 17 期

NGU Y EN V T , Y E G, VA N B K , et al: Study the Hy dration and the M icrostructur e % %

255

w hich is much sm aller than t hat of RHA, i. e. 5. 6 m. T heref ore, t he possibilit y of SF to ref ine t he pore struc t ure of the syst em is larger t han that of RHA. T his causes a low er t otal porosity of SF samples compared to RHA samples. 3. 4

Discussion T he higher hy drat ion degree of RHA blended cement w it h t he low w / b rat io can be att ribut ed t o several es

pect s. F irst , because of its porous st ruct ure, RHA has a large specific surface area. During m ix ing, RH A may also absorb a cert ain amount of free w at er into it s pores. T his water is released from t hese pores w hen t he rela t ive humidit y in the past e decreases w it h progress of cem ent hydrat ion process, and theref ore increases the hy drat ion degree of RHA sample at the lat er stage. T his eff ect is larger w hen t he w / b ratio of m ix ture is low er. T his mechanism is similar to that proposed by van Breugel[ 12] for int ernal curing of high strengt h concrete w ith the w ater sat urated light w eight aggreg at es. Second, a pozzolanic react ion corresponds to the second peak on t he exothermic curve. T his peak is clearer w it h an increase of RHA replacement percentages. Moreover, the RHA possesses a large specific surface area, w hich can improve the pozzolanic react ion. T herefore, the more the RHA is added, t he more t he heat w ill be produced. T his can give a higher hydration degree of RHA samples compared to that of SF samples.

4

Conclusions F rom above result s, some main conclusions can be drawn: a) T he RHA is a reactive pozzolan sim ilar to the SF . b) T he eff ect of RHA on cement hydration is more pronounced at t he lat er st age. c) T he eff ect of RHA on cement hydration is larg er than that of SF in pastes w it h a low w/ b rat io. T his

may be att ributed by int ernal curing of RHA and t he pozzolanic reaction of RHA and cement hydration product. d) Bot h RHA and SF not only refine the pore st ruct ure of cement past e but also reduce the t otal porosity, in w hich the ef fect of SF is larger than t hat of RHA. References [ 1]

M ehta P K . Pozzo lanic and Cementitious by products as M ineral A dmix tures for Concrete a Critical Rev iew. ACI Special Publi catio n SP 79, Ed V M M alho tra[ A] , 1983, 1 46.

[ 2]

Cheng yi H, Feldman R F. Hy dration Reactio ns in Portland Cement silica Fume Blends[ J] . Cem Concr Res, 1985, 15( 4) : 585 592.

[ 3]

Feng Q , Yamamichi H, Shoya M , et al. Efficiency of Highly Active Rice Husk Ash on the Hig h str ength Concrete[ A] . T he 11th Inter . Congr ess on the Chemistry of Cement ( ICCC) , Durban, South Afr ica, 2003, 816 822.

[ 4]

Dalhuisen D H, Str oeven P, Bui D D , et al. Replacement of Condensed Silica Fume by Rice Husk Ash for t he Production of High Strength Co ncrete in Developing Countries[ J] . Proc Int Conf U tilization o f Hig h Strength/ High Performance Concrete, Par is, 1996: 235 243.

[ 5]

Chandrasekhar S, Pramada P N , R ag havan P, et al. M icrosilica from R ice Hush Ash As a Possible Substitute for Condensed Silica Fume for High Performance Concr ete[ J] . J of M aterials Science L etters, 2002, 21: 1245 1247.

( 下转第 260 页)

260

武

汉

理

工

大

学

学

报

2010 年 9 月

References [ 1]

Akhtaruzzaman A A , Hasnat A. Properties of Concrete Using Crushed Brick as Aggr eg ate[ J] . Concrete International, 1983, 5 ( 2) : 58 63.

[ 2]

M ohammad A M , Wee T H, Lee S C. Crushed Bricks as Coarse Agg reg ate for Concr ete[ J] . ACI M aterials Jour nal, 1999, 96 ( 4) : 478 484.

[ 3]

Jiang D M , Bai Z G , Wang F C, et al. T est Research on Enviro nment friendly Concrete[ J] . Ar chitecture T echnolog y, 2006, 37 ( 1) : 38 40.

[ 4]

Xing Z X, L iu L J, Zhao Y Q , et al. Research on the M ix ture of Recycle Brick Agg regate Concr ete[ J] . Recycling Research, 2006, 2: 38 40.

[ 5]

China Academy of Building Resear ch concrete. Application of L ight Ag gregate Abroad[ M ] . Beijing: China Architectur e & Building Press, 1982.

( 上接第 255 页) [ 6]

M ehta P K . Rice Husk Ash a U nique Supplementary Cementing M aterial[ A ] . P roc of the Inter Sy m on A dvance in Concrete T echnology, Athens, G reece, 1992, 407 430.

[ 7]

Bui D D. Rice Husk A sh as a M ineral Admix tur e for H igh P erformance Concr ete. P h D T hesis [ D] . Delft U niviversity of T echnology, 2001, Delft, 122.

[ 8]

Washburn E W. N ote on a M ethod of Determining the Distribution of Pore Sizes in a Porous M aterial. Proc of National A cade my of Sciences[ J] . 1921, Washington, ( 7) : 115 116.

[ 9]

Ye G . Ex perimental Study and Numer ical Simulation of the M icrostr ucture and Per meability of Cement itious M ater ials[ D ] . Ph D T hesis, Delft U niviversity of T echnology, 2003, Delft, 186.

[ 10]

Van Breugel K . Simulat ion of Hydration and Formation of Str ucture in Har dening Cement Based M aterials[ D] . Ph D thesis, 1991: 305

[ 11]

Waller V , De Larr ar d F, Roussel P , et al. M odeling the T emperature Rise in M assive HP C Structures[ M ] . Proc. of t he 4th

[ 12]

V an Breugel K , De Vr ies J, T akada K . L ow W ater/ binder Ratio Concretes w ith Low Cracking Risk by U sing L WA [ M ] .

Inter . Sym. on U tilization of High strength/ High performance Concrete, 1996: 415 421. Proc. Int. Conf. Dundee, 1999: 255 2641.