Hybrid Reinforcement of Sisal and Polypropylene Fibers in ...€¦ · Fiber Preparation and Characterization Commercial unbleached sisal kraft pulp from Brazil was used in the experiments

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

Hybrid Reinforcement of Sisal and Polypropylene Fibersin Cement-Based Composites

G. H. D. Tonoli1; H. Savastano Jr.2; S. F. Santos3; C. M. R. Dias4; V. M. John5; and F. A. R. Lahr6

Abstract: Several studies using vegetable fibers as the exclusive reinforcement in fiber-cement composites have shown acceptablemechanical performance at the first ages. However, after the exposure to accelerated aging tests, these composites have shown significantreduction in the toughness or increase in embrittlement. This was mainly attributed to the improved fiber-matrix adhesion and fibermineralization after aging process. The objective of the present research was to evaluate composites produced by the slurry dewateringtechnique followed by pressing and air curing, reinforced with combinations of polypropylene fibers and sisal kraft pulp at different pulpfreeness. The physical properties, mechanical performance, and microstructural characteristics of the composites were evaluated beforeand after accelerated and natural aging. Results showed the great contribution of pulp refinement on the improvement of the mechanicalstrength in the composites. Higher intensities of refinement resulted in higher modulus of rupture for the composites with hybridreinforcement after accelerated and natural aging. The more compact microstructure was due to the improved packing of the mineralparticles with refined sisal pulp. The toughness of the composites after aging was maintained in relation to the composites at 28 days ofcure.

DOI: 10.1061/�ASCE�MT.1943-5533.0000152

CE Database subject headings: Aging; Fibers; Composite materials; Durability; Portland cement; Hybrid methods.

Author keywords: Accelerated aging; Cellulosic fibers; Composites; Durability; Polypropylene fibers; Portland cement; Refinement,sisal fibers; Vegetable fibers.

Introduction

Fiber-cement products had been widely used in the world due totheir versatility as corrugated and flat roofing materials, claddingpanels, and water containers presented in large number of build-ing and agriculture applications �Ikai et al. 2010�. The main rea-son for incorporating fibers into the cement matrix is to improve

1Postdoctorate Student, Dept. of Structural Engineering, Escola deEngenharia de São Carlos, Univ. of São Paulo, Avenida Trabalhador SãoCarlense, 400, 13566-590, São Carlos/SP, Brazil. E-mail:[email protected]

2Full Professor, Dept. of Food Engineering, Faculdade de Zootecnia eEngenharia de Alimentos, Univ. of São Paulo, Avenida Duque de CaxiasNorte, 225, 13635-900 Pirassununga/SP, Brazil �corresponding author�.E-mail: [email protected]

3Postdoctorate Student, Dept. of Food Engineering, Faculdade deZootecnia e Engenharia de Alimentos, Univ. of São Paulo, AvenidaDuque de Caxias Norte, 225, 13635-900 Pirassununga/SP, Brazil. E-mail:[email protected]

4Graduate Student, Dept. of Construction Engineering, Escola Politéc-nica, Univ. of São Paulo, 05508-900, São Paulo/SP, Brazil. E-mail:[email protected]

5Associate Professor, Dept. of Construction Engineering, EscolaPolitécnica, Univ. of São Paulo, 05508-900, São Paulo/SP, Brazil. E-mail:[email protected]

6Full Professor, Dept. of Structural Engineering, Escola de Engen-haria de São Carlos, Univ. of São Paulo, Avenida Trabalhador São Car-lense, 400, 13566-590, São Carlos/SP, Brazil. E-mail: [email protected]

Note. This manuscript was submitted on December 10, 2008; ap-proved on July 16, 2010; published online on July 19, 2010. Discussionperiod open until July 1, 2011; separate discussions must be submitted forindividual papers. This paper is part of the Journal of Materials in CivilEngineering, Vol. 23, No. 2, February 1, 2011. ©ASCE, ISSN 0899-
1561/2011/2-177–187/$25.00.
JOURNAL OF MA

J. Mater. Civ. Eng. 201

the toughness, tensile strength, and the cracking deformationcharacteristics of the resultant composite. It is well known thatpolymer synthetic fibers, like polyvinylalcohol �PVA� and poly-propylene �PP� fibers, lead to the improvement of the postpeakductility performance, performance under fatigue, impactstrength, and also help to reduce the shrinkage cracking �Balaguru1994; Coutts 1987; Hannant 1978; Mindess 1993�.

The success of the cellulose pulp in a broad field of applica-tions can be influenced by the choice of the production methodand by the ability to capture the optimum mechanical perfor-mance of these fibers �Coutts and Warden 1992�. It is also depen-dent on overcoming major concerns related to fiber degradation incementitious media �Agopyan et al. 2005�. One of the possibletreatments to enhance mechanical performance of composites re-inforced with cellulosic pulp is the refinement process, which iscarried out in the presence of water, usually by passing the sus-pension of pulp fibers through a disk refiner composed by a rela-tively narrow gap between the rotor and the stator �Britt 1970;Clark 1987�. Cellulosic fibers are intrinsically strong, and the re-finement greatly improves their processability, which is necessaryif the composite is manufactured using the Hatschek productionmethod �Coutts 1984�. The main effect of refinement in cellulosicfiber structure as a result of mechanical action is the fibrillation ofthe fibers surface �Coutts 1987�. These fibrillated and shorter fi-bers are responsible for the formation of a net inside the compos-ite mixture with the consequent retention of the cement matrixparticles during the dewatering stage of manufacturing process.

A previous study �Tonoli et al. 2007� showed that refinementcauses intense fibrillation of the sisal pulp fibers, favoring theirplasticization and mechanical anchorage in the cement-based ma-trix. Such a behavior led to significant improvement of the modu-
lus of rupture �MOR� of cement-based composites. However, in
TERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2011 / 177

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

general, the better adhesion of the fibers and the fiber mineraliza-tion reduced the incidence of fiber pull-out during the compositefracture with the consequent damage to the toughness of the ma-terial. Fiber adhesion with the matrix is improved after aging dueto the reprecipitation of cement hydration products �rich in cal-cium� around the fibers and fiber mineralization is the process ofreprecipitation of cement hydration products into the lumen of thefibers �Mohr et al. 2005; Tonoli et al. 2009, 2010�. The use of asmall content of PP fibers could represent an alternative to main-tain the toughness of the composites in the long run, while thecellulose pulp fibers are responsible to capturing the cement par-ticles during the manufacturing process. Therefore, the use of twotypes of fibers in a suitable combination can potentially result inperformance synergy, whose hybrid performance exceeds the sumof individual fiber performances �Banthia and Sappakittipakorn2007�. Attempts have been made at identifying fiber combinationsthat produce the maximum synergy �Ding et al. 2009; Ding et al.2010; Qian and Stroeven 2000�. Banthia and Gupta �2004� de-scribed three different categories for hybrids reinforcement: hy-brids based on fiber constitutive response �one type of fiber isstiffer and the second is flexible�, hybrids based on fiber dimen-sions �one type is smaller and the second is larger�, and hybridsbased on fiber function �one type improve the production processand the second improve mechanical properties, for example�.

PP fibers have been studied as alternative low-cost fiber forreinforcement in fiber-cement roofing products �Kalbskopf et al.2002�. These fibers present elevated ductility and they are resis-tant to alkaline attack �Hannant 1998� when compared to cellu-lose fibers. However, they present low elastic modulus �3.5–4.8GPa� in relation to other reinforcing fibers �Zollo 1997� and someother disadvantages as weak chemical bonding with cement dueto their hydrophobic nature and the consequent poor mechanicalanchoring to the cementitious matrix.

This paper investigates the effect of cellulose pulp refinementon the properties of fiber-reinforced cement composites afteraging. The objective of this present study is to evaluate the per-formance of fiber-cement material using conventional kraft sisal�Agave sisalana� pulp with three different pulp freeness levelsand PP fiber as durable toughening reinforcement. The referredcomposites are also compared with composites exclusively rein-forced with sisal pulp fibers in different pulp freeness levels anddifferent aging conditions.

Experimental

Fiber Preparation and Characterization

Commercial unbleached sisal kraft pulp from Brazil was used inthe experiments. Part of the unrefined pulp was submitted to astirring process in water, only to provide fiber dispersion, main-taining the original freeness level given by the Canadian StandardFreeness �CSF� of 680 mL. Another part of the pulp with fiberconcentration of 5.3 g/L in water was postrefined in a Bruno diskrefiner model 2RA-12, as reported in Tonoli et al. �2007�. Disksare 300-mm diameter with 3-mm width bar, 3-mm width groove,and 7.5° angle bar configuration �Fig. 1�. A specific edge load ofapproximately 0.3 W·s /m was used for the refinement, operatingat current intensity of 50 A. Pulp was passed 10 and 15 timesthrough the refiner, for the achievement of CSF degrees of 220and 20 mL, respectively. The CSF test is a widely recognizedstandard measure of the drainage properties of pulp suspensions
�Coutts and Ridikas 1982� and it relates well to the initial drain-
178 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRU


age rate of the wet pulp pad during the dewatering process. Lowfreeness values �less than 300 mL� are indicative of high degreeof external fibrillation and/or shortage of the fibers, leading tolong drainage periods during the test. The CSF value of each pulpwas determined in accordance with the correspondent BrazilianStandards NBR 14344 �Brazilian Technical Standards Association�ABNT� 2003�.

The main physical attributes of the pulp were previously char-acterized �Tonoli et al. 2007� by two procedures: particle sizeanalyzer �Galai CIS-100 equipment, NIWA, Auckland, NewZealand� and, only for entire fibers �fibers that appear to be notteared by the refining process�, optical microscopy. Microscopicmorphological characterization was performed in at least 100 en-tire fibers for each pulp by image acquisition with the equipmentOlympus PV10-CB. Besides the average length and width of en-tire fibers, this analysis allows the evaluation of the lumen diam-eter �Table 1�. The analysis with Galai CIS-100 consists in theevaluation of the attributes of the whole fibrous material presentin the pulp �Table 2�. Average length and width, number of fibersper gram, and fines content were analyzed and stored with the aidof the Wshape v.1.0 software �Ferreira et al. 2006�.

Fig. 1. View of the bars configuration of the plate in a opened diskrefiner �3-mm width bar, 3-mm width groove, and 7.5° angle bar�

Table 1. Microscopic Morphological Characterization of Entire Fibers�Data from Tonoli et al. 2007�

Morphological properties

Sisal pulps

Unrefined,CSF 680 mL

Refined

CSF 220 mL CSF 20 mL

Average length �mm� 2.59�0.66 2.55�0.56 2.30�0.55

Average width ��m� 21.33�4.93 21.02�3.99 19.72�4.07

Lumen diameter ��m� 9.45�4.90 9.10�4.04 8.96�3.34

Table 2. Pulp and Fiber Physical Properties �Data from Tonoli et al.2007�

Properties

Sisal pulps

PPfiber

Unrefined,CSF 680 mL

Refined

CSF 220 mL CSF 20 mL

Average length �mm� 1.66�0.02 1.13�0.05 0.79�0.01 �6.00

Average width ��m� 22.2�0.5 18.7�0.2 20.0�0.3 �28.0

Aspect ratio 75 60 40 215

Fibrous material�106 fibers /g�

4.69�0.02 8.88�0.93 15.64�0.17 —

Fines content �%� 27.2�1.2 40.6�1.3 42.0�0.7 —

ARY 2011

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

The reduction of the thickness of cell wall was observed forrefined pulps �Tonoli et al. 2007�, and can be associated with theexternal fibrillation that causes delamination of the surface layers�Fardim and Durán 2003� and consequently peeling off of theouter layers to the liquid media �Somboon et al. 2007�. The fibersalso acquire a ribbonlike shape, which helps in explaining thetendency of diminution of lumen diameter in the refined pulps�Table 1�. The microscopy analysis did not reflect the character-istics of the whole pulp, which includes the cut fibers and thegeneration of fines. The characterization of the fibrous material�Table 2�, i.e., intact plus damaged fibers performed in Galaiequipment gives a more precise idea of the actual morphology ofthe pulp. Table 2 depicts the drastic changes caused by the refine-ment to the universe of fibers in the pulp. Average length was themost affected property, with the decrease of 52%. Average widthremained practically unchanged and fibrous material and finescontent increased 233 and 54%, respectively. The tensile strengthof the individual pulp fibers is expected to be between 500 and900 MPa, and the elastic modulus of 25 to 40 GPa �Campbell andCoutts 1980�.

PP fibers �monofilaments�, with approximately 6-mm lengthand 28-�m width �circular section�, with 300-MPa tensilestrength, 3.3-GPa elastic modulus, and 25% strain at failure �datasupplied by Fitesa S.A., Brazil�, were used for mix design withhybrid reinforcement.

Cement and Carbonate Characterization

Ordinary Portland cement �OPC-CPIIE�, according to BrazilianStandards NBR 11578 �Brazilian Technical Standards Association�ABNT� 1991� �with additions of blast furnace slag=6–34% andcarbonate filler=0–10%�, was used in the composites. Oxidecomposition of the OPC is �wt %�: 25.3% SiO2, 58.6% CaO,5.83% Al2O3, 2.88% Fe2O3, 3.02% MgO, 0.19% P2O5, 0.84%K2O, 0.19% SrO, 0.08% MnO, and 0.03% ZrO2. Further carbon-ate filler was used for partial substitution of OPC in order toreduce costs concerning the production of fiber cement. Oxidecomposition of the carbonate filler �wt %�: 2.47% SiO2, 44.6%CaO, 0.45% Al2O3, 0.37% Fe2O3, 8.56% MgO, 0.08% P2O5,0.17% K2O, 0.04% TiO2, 0.05% SrO, and 0.06% MnO. Particlesize distribution was evaluated by Mastersizer S long bed 2.19 v.in a Malvern equipment as depicted in Fig. 2.

The carbonate filler was chosen in order to present the sameparticle distribution of the OPC. According to Fig. 2, 50% of theparticles are smaller than 13.6 and 14.5 �m for OPC and carbon-ate filler, respectively. Most of particles �90%� are smaller than

Fig. 2. Discrete particle size distribution �DPSD� and cumulativeparticle size distribution �CPSD� of OPC �CPIIE� and carbonate ma-terial

41.8 and 50.1 �m for OPC and carbonate filler, respectively.

JOURNAL OF MA


Formulation and Composite Preparation

Cement composite pads measuring 200 mm�200 mm and rein-forced by sisal pulp and PP fibers were prepared in the laboratoryusing a slurry vacuum dewatering and pressing technique de-scribed in details by Savastano et al. �2000�. Formulations �Table3� were established based on studies published elsewhere�Bezerra et al. 2006; Tonoli et al. 2007�. Pulp and PP fibers wereboth previously dispersed in water by mechanical stirring at 1,700revolutions per minute �rpm� for 1.5 h and 30 min, respectively.Then, cement and carbonate filler were added and the mixtureformed with approximately 20% of solids was stirred at 1,700rpm for additional 20 min. The slurry was transferred to theevacuable casting box and vacuum was applied �approximately80-kPa gauge� until a solid surface formed. Three pads of eachformulation were pressed simultaneously at 3.2 MPa for 5 min,then sealed wet in a plastic bag to cure at room temperature for 2days and immersed in water for 5 days. Pads were cut wet intofour 165 mm�40 mm flexural test specimens using a watercooled diamond saw. Specimen thickness was approximately 5mm. Samples were allowed to air cure in an internal environmentof 27�2°C and 65�5% of relative humidity for a period of 20days, prior to mechanical and physical testing. On completion ofthe air curing, specimens were tested at 28 days after production.Specimens were soaked in water for 24 h prior to mechanical testunder saturated condition.

Composites produced in the present work were compared withcomposites reinforced only with sisal pulps in the same pulp free-ness as described detailed in Tonoli et al. �2007�. The formulationof the composites with only sisal pulp was 78.8% cement �CP-IIE�, 16.5% carbonate filler, and 4.7% of sisal pulp �% by mass�,following the same cure procedures.

Physical and Mechanical Characterization

Water absorption �WA�, bulk density �BD�, and apparent voidvolume �AVV� values were obtained from the average of 10specimens for each mix design, following procedures specified byASTM C948 �ASTM 1981�. Mercury intrusion porosimetry�MIP� was performed using Micromeritics Poresizer 9320 withpressure of up to 200 MPa. The assumptions made were0.495 g /cm2 mercury surface tension and 13,534 kg /m3 mer-cury density. Equilibration time in both low and high pressurewas 10 s. The advancing/receding contact angle was assumed to

Table 3. Mix Design of Fiber-Cement Composites Reinforced with SisalPulp and Polypropylene Fibers

Raw materialSP-680

�% by mass�SP-220

�% by mass�SP-20

�% by mass�

Sisal pulp �CSF 680 mL� 3.0a — —

Sisal pulp �CSF 220 mL� — 3.0 —

Sisal pulp �CSF 20 mL� — — 3.0

PP fibers 1.7b 1.7b 1.7b

Cement �CP-IIE�c 78.8 78.8 78.8

Carbonate fillerd 16.5 16.5 16.5aEquivalent volume fraction= �2.8%.bEquivalent volume fraction= �2.6%.cNBR 11578 �Brazilian Technical Standards Association �ABNT� 1991�;additions of blast furnace slag=6–34%, and carbonate material=0–10%; density=3,080 kg /m3.dDensity=2,850 kg /m3.

be 130°. The amount of mercury intruded at each pressure inter-


1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

val was recorded. Specimens were cut with a cubic side length ofapproximately 5 mm, dried at 70°C for 24 h and stored in anair-tight recipient prior to evaluation. Sample mass was approxi-mately 2 g. The technique was adopted in order to evaluate thepore size distribution as usually applied in the characterization ofcement-based materials �Kuder and Shah 2003; Mehta and Mon-teiro 2006; Nita et al. 2004�.

Mechanical tests were performed using the universal testingmachine Emic DL-30,000 equipped with 1-kN load cell. Four-point bending configuration was employed to evaluate the limit ofproportionality �LOP�, the MOR, the modulus of elasticity�MOE�, and the toughness �T� of the specimens. A 135-mm spanand a deflection rate of 1.5 mm/min were adopted in the bendingtests, in order to determine the LOP, MOR, and MOE, followingcalculations specified in Eqs. �1�–�3�:

LOP =Plop · Lv

b · h2 �1�

MOR =Pmax · Lv

b · h2 �2�

MOE =276 · Lv

3

1,296 · b · h3 · �m� �3�

where Plop= load at the upper point of the linear portion of theload versus deflection curve; Pmax=maximum load; Lv=majorspan; b and h=specimen width and depth, respectively; and m=tangent of the slope angle of the load versus deflection curveduring elastic deformation.

Toughness �T� was defined as the energy absorbed during theflexural test divided by the cross-sectional area of the specimenunder investigation �Eq. �4��, as described previously �Tonoliet al. 2007; Tonoli et al. 2009� and adapted from RILEM �1984�.The absorbed energy was obtained by integration of the areabelow the load versus deflection curve at the point correspondingto a reduction in load carrying capacity to 30% of the maximumreached. The deflection during the bending test was collected bythe deflectometer positioned in the middle span, on the down sideof the specimen. The values of stress �� were calculated usingthe Eq. �1� for each load value, P. The values of deflection weredivided by a span of 135 mm and called strain �� in the presentwork. Each test was finalized when � decreased to 0.3�MOR.Fig. 3 presents a typical stress versus strain curve, which definesLOP, MOR, and toughness �T�. MOE was not defined in Fig. 3because it was determined by Eq. �3� using load versus deflection

Fig. 3. Definition of the mechanical parameters in a typical stressversus strain curve

curves



Toughness =absorbed energy

b � h�4�

Scanning electron microscopy �SEM, Zeiss LEO 440 and Hitachi-S2500 microscopes� was applied for the characterization of fiber-matrix interface on fractured surface of specimens that hadundergone mechanical tests similarly to the procedures used inSavastano et al. �2005�. Samples were previously gold coated in aBal-Tec MED 020 sputtering system.

Natural and Accelerated Aging Tests

The accelerated aging test aims to simulate natural aging withexposure to cycles of soaking and drying. Specimens were suc-cessively immersed in water at 20�5°C for 18 h and exposed tothe temperature of 60�5°C for 6 h in a ventilated oven as de-scribed in the Standard EN 494 �European Committee for Stan-dardization 1994�. Each soak and dry procedure represents onecycle and it was performed for 50 and 100 cycles, respectively, tosimulate weathering exposition.

Some specimens were submitted to natural weathering for 1year in rural environment. This test was carried out in Pirassu-nunga, State of São Paulo, Brazil �latitude 21°59�S�. During theperiod of exposition of the specimens, the rainfall was approxi-mately 1,550 mm, the maximum and minimum temperatures were37.0 and 4.9°C, respectively, the highest radiation peak was ap-proximately 1 ,210 W /m2. The maximum and minimum relativehumidities were 93.3 and 41.3%, respectively. The compositeswere positioned in a metallic structure facing north with 30° slopein relation to the horizontal plane.

Statistical Analyses

The results of physical and mechanical tests were subjected totwo-way ANOVA �three levels of the cellulose pulp freeness andfour different aging conditions� at the 95% confidence level ��=0.05�, similarly as applied in Tonoli et al. �2010�.

Results and Discussion

Physical Properties

The results of the physical properties are depicted in Table 4.Composites with hybrid reinforcement seems to present higherapparent porosity �AVV� than composites only reinforced withcellulose pulp fibers as a probable consequence of the poorerpacking of the matrix due to the addition of long and hydrophobicsynthetic fibers �Bentur and Mindess 1990�. The use of PP fibersincreased the WA of the composite by around 10% when com-pared with the composites without PP fibers from the work con-ducted by Tonoli et al. �2007�. The poor dispersion of thehydrophobic PP fibers led to an inefficient packing of the com-posite and thus contributing to the higher porosity of the matrixphase. Similar values of porosity were found by Bezerra et al.�2006� for hybrid composites in analogous conditions of cure andaging, however using PVA fibers, cellulose pulps, and other min-eral additions.

The addition of PP fibers diminished the BD of the compositesin relation to that found in the previous study �Tonoli et al. 2007�with composites only reinforced with sisal pulp. This fact is di-
rectly related with the higher porosity of the matrix and it is also
ARY 2011

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

due to the substitution of the matrix ��2.6 g /cm3� and thecellulose fibers ��1.5 g /cm3� by the lighter phase representedby the PP fibers ��=0.9 g /cm3�.

The difficulty in the composite compacting �by tamping� withunrefined pulps �CSF 680 mL� can be held responsible for thelarge porosity in some cases. As described by Coutts �2005�, thefibrilation generated by the refinement operation contribute to thehigher malleability of the cellulose fibers and improves the pack-ing of the phases in the composites. Furthermore, the unrefinedfibers have not been collapsed and present higher lumen diam-eters �Table 1�, which contribute to the large porosity.

Accelerated aging �100 cycles� and natural aging caused repre-cipitation of cement hydration products �rich in calcium� aroundand inside the cellulosic fibers and around PP fibers. In the case ofcomposites after 1 year of natural aging, carbonation of the ce-ment matrix also took place �Tonoli et al. 2010�. Therefore, theconsequences of aging were the significant increase of the BDand the decrease of WA and AVV of the composites with the threedifferent levels of pulp freeness �Table 4�.

MIP

Fig. 4 presents the effect of PP addition in the pore size distribu-tion related to composites only reinforced with refined sisal pulp�Tonoli et al. 2007�. It was outlined in Fig. 4�a� for SP-220 theincidence of larger pores in the region between 1,000 and 3,000nm. This profile of the pore size distribution of composites withsisal pulp and PP fibers confirms that the incidence of large cap-illary pores is in agreement to the increase of the WA as previ-ously discussed.

Fig. 4�b� shows the effect of 100 aging cycles on the compos-ites with PP fibers and refined pulp �220 CSF mL�. Differences inthe curve of the pore size distribution were evident mainly for thepores with dimention below 100 nm. During the 100 soak and drycycles, the continuous hydration and reprecipitation of the port-land cement took place. This led to the reduction of the matrixvoids, as indicated by the AVV values in Table 4. Larger poresattributed to fibers �in the interval from 1,000 to 5,000 nm� werenot significantly affected by the soak and dry cycles. Reprecipi-tation of cement hydration products after accelerated aging was

Table 4. Average Values and Standard Deviation of Ten Specimens forWater WA, AVV, and BD of the Hybrid Composites Reinforced with PPFibers and Sisal Pulp at Three Pulp Freeness �i�� and at Different AgingConditions �j��

Composites�i�� Condition �j��

WA�%�

AVV�%�

BD�g /cm3�

SP-680 28 days �unaged� 21.1�1.3 32.8�1.1 1.56�0.04

50 cycles �aging� 19.2�3.0 31.5�2.0 1.61�0.09

100 cycles �aging� 13.6�2.4 22.9�3.2 1.69�0.07

Natural aging �1 year� 15.0�0.9 26.1�1.4 1.73�0.02

SP-220 28 days �unaged� 20.9�0.7 33.5�0.6 1.61�0.03

50 cycles �aging� 18.5�2.0 29.9�2.4 1.62�0.05

100 cycles �aging� 15.1�1.7 25.4�2.1 1.68�0.05


SP-20 28 days �unaged� 20.4�0.8 33.5�0.7 1.64�0.03

50 cycles �aging� 16.7�0.7 27.9�0.8 1.67�0.04

100 cycles �aging� 15.5�1.0 26.3�1.4 1.71�0.01


not enough to fill larger pores �1,000 to 5,000 nm�. It is expected

JOURNAL OF MA


that the natural carbonation of the cement matrix after 100 agingcycles was lower �almost negligible� than after natural aging�Tonoli et al. 2010�.

Table 5 depicts the summary of the intrusion data achievedwith MIP analysis for unaged and aged composites reinforcedwith refined sisal+PP fibers �SP-220� or with only refined sisal�SS-220� from previous work �Tonoli et al. 2007�. The BD of theSP-220 determined by MIP was in accordance with that shown inTable 4.

After 100 aging cycles, a reduction of approximately 40% inthe total volume of mercury intruded for composites with PPfibers, SP-220, and around 30% for composites only reinforcedwith refined sisal pulp, SS-220, was observed. Table 5 also showsthe significant reduction of the total pore area after aging for bothcomposites categories.

Fig. 4. �a� Pore size distribution; �b� cumulative mercury intruded incomposites reinforced with refined sisal+PP fibers �SP-220� and ex-clusively with refined sisal �SS-220� at 28 days and after 100 agingcycles. Data from Tonoli et al. �2007�.

Table 5. Intrusion Data Summary for the Composites with Hybrid Rein-forcement �Refined Sisal Pulp+PP Fibers� and Exclusively Reinforcedwith Sisal Pulp

28 days �unaged� 100 aging cycles

SP-220 SS-220a SP-220 SS-220a

Total intrusion volume �mL/g� 0.068 0.048 0.040 0.033

Total pore area �m2 /g� 6.11 7.76 3.82 3.05

Median pore diameter-volume�nm�

93.3 64.4 147.5 145.8

Median pore diameter-area �nm� 14.2 10.2 13.5 15.4

Average pore diameterb �nm� 44.2 24.9 41.7 43.1

BD �g /cm3� 1.59 1.85 1.73 1.81aData from Tonoli et al. �2007�.b
4 V /A �V=volume of mercury intruded and A=total area of pores�.

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

Even though the total intrusion volume was higher for thecomposites with refined sisal pulp+PP fibers �SP-220� at 28 daysin comparison with the composites with only refined sisal pulp�SS-220� �Table 5�, the total pore area of the composites with PPfibers was lower. This can be explained by the greater amount ofpores with lower sizes �below 100 nm� in composites only rein-forced with sisal pulp at 28 days, as illustrated in Fig. 4�a�. Thepores of lower size can be attributed to the lumen of the cellulosefibers. In the case of composites reinforced with only sisal pulp,the continuous hydration and reprecipitation of cement pores�below 100 nm� with the consequent increase of hydration prod-ucts after aging filled the smaller the value of the average porediameter after 100 aging cycles.

Mechanical Properties and Microstructure

Composites with refined pulp �SP-220 and SP-20� presented thelarger values of LOP, at 28 days �Fig. 5�a��. Refinement softensthe fiber, increasing its flexibility, which allows it to wrap aroundthe cement and other minerals and makes intimate contact withparticles, binding them tightly together �Coutts and Kightly 1982�and providing a more packed composite.

Fig. 6 shows SEM micrographs of fibers prior to the inclusionin the composite. Fig. 6�a� depicts the smooth surface of the un-refined fibers and in the Figs. 6�b and c� the external layers werepartially pulled out from the sisal fibers after refining. These ex-ternal layers improve the anchorage of the fibers into the cemen-titious matrix. In the Figs. 7�a and b�, the external layers of therefined fibers can be seen largely bonded to the cementitious ma-trix. In this case, the primary layer remained adhered to the matrixand the leftover fiber was pulled out. The refined fiber-cementpaste bond seems to be stronger than that of the unrefined fiber-cement paste bond �Mohr et al. 2005; Tonoli et al. 2007�. It can

Fig. 5. Average values and standard deviation of: �a� LOP; �b� MOR;and �c� toughness �T� in different aging conditions of compositesreinforced with sisal pulp+PP, in function of three levels of pulpfreeness

be seen the poor adhesion of the unrefined fibers in the composite,



SP-680, after aging as presented in the Fig. 8. Fig. 8�a� depicts thevoid of up to 3 �m at the interface between fiber and matrix after100 aging cycles.

Fig. 8�b� presents the sheathlike reprecipitation of cement hy-dration products into this interface around the unrefined fibersafter 1 year of natural aging. This reprecipitation improved theanchorage of the fibers and increased the values of LOP, MOR,and T of the composites �SP-680� in relation to those at 28 daysof cure �Fig. 5�.

Fig. 5�b� shows that the composites reinforced with PP fibersand refined sisal pulp �SP-220 and SP-20� presented average val-ues of MOR greater than those composites containing unrefinedpulp �SP-680� in all aging conditions. Besides, the lower pulpfreeness �CSF 20 mL� led to 22 and 10% of increase in the aver-age value of MOR of the composite when compared to the inter-mediate pulp freeness after 100 aging cycles and 1 year of naturalaging, respectively. Therefore, the large superficial contact per-formed by refined celulosic pulp enhanced the MOR and im-proved the load transferrance from the matrix to the fibers.

Fig. 9 shows the comparisons between the typical stress versusstrain curves of the composites with the same content of PP fiber

Fig. 6. SEM micrographs of the sisal pulp fibers: �a� unrefined �CSF680 mL�; �b� refined �CSF 220 mL�, pull-out of the primary cell wallof the fiber; and �c� refined �CSF 20 mL�, intensive fibrillation of thefibers

but with sisal pulp with different freeness levels �Table 3�. Typical

ARY 2011

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

stress versus strain curves means that their profiles represent theaverage mechanical behavior of composites submitted to fourpoint bending test. The curves were chosen based on the averagevalues of MOR and toughness �T�.

The PP fibers guaranteed a pseudoplastic behavior of all com-posites presented in this work. However, the mechanical behavior

Fig. 7. SEM micrographs of fracture surfaces in hybrid compositeswith PP fibers and refined pulp �CSF 20 mL�: ��a� and �b�� naturalaging �arrows indicate primary layer adhered to the matrix�

Fig. 8. SEM micrographs of fractured surfaces of composites withunrefined pulp �CSF 680 mL�: �a� voids at the fiber-matrix interfaceafter 100 aging cycles; �b� after 1 year of natural aging �arrow indi-cate the sheathlike reprecipitation of cement hydration products�

JOURNAL OF MA


of the composites after accelerated and natural aged �Fig. 9�shows the effect of adhesion of the primary layer of the sisal pulpfiber to the matrix caused by the reprecipitation of the cementhydration products in the fiber/matrix interface �Mohr et al. 2005;Mohr et al. 2006; Tonoli et al. 2009�. This fact indicates that therewas sufficient bond between the sisal pulp fibers and the cementmatrix to induce a good mechanical anchorage and interlockingeffects. It was observed considerable number of sisal pulp fibersthat were broken instead of the frictional slip provided by pull-out. There were also a combined effect between refine degrees ofthe sisal pulp and aging, what promoted the sawtooth curves.Besides, the curves of the composites SP-20 indicates that itsfracture process were more instable than others composites be-cause presented major drop in the ramps.

The LOP of the composites was maintained after aging forSP-220 and SP-20, and increased for SP-680 �Figs. 5�a and b��.This increase in the LOP is due to the reduction of porosity and/orWA �Fig. 10� with the aging by the reprecipitation of the cementhydration products in the fiber surroundings. Soroushian et al.�1994� using both refined virgin and recycled Pine kraft pulp re-ported this as the main mechanism of the composite embrittle-ment and responsible for the increase of the bending resistanceafter repeated soak and dry cycles. The increase of the LOP withaging was also found by Bezerra et al. �2006� and Hannant andHughes �1986�.

The ductile behavior �great postcracking regions in the Fig. 9�of the composites even after aging was attributed to the longer PPfibers �6 mm�. The hybrid fiber systems may render the fiber-reinforced composites more efficient compared to only sisal pulpfibers at all freeness levels �Tonoli et al. 2007�. This suggests thatduring the application of the load upon the composite, the sisalfibers fractured first, then the pulling out of PP fibers started.Rupture of the sisal pulp fibers and pull out of the PP fibers arepresented in the Fig. 11. These two events induce the better per-formance of the composite due to the distinct behavior of eachfiber during the mechanical test, indicating a synergy effect in theaction of these two fiber types. Additionally the PP fibers presentmodulus of elasticity about 10 times lower than the sisal pulp �asreported above� what is also an important factor to explain theimprovement of the toughness of the composite containing theplastic fiber.

Figs. 9�c and d� show successive peaks in the stress versusstrain curves which indicate that the multicracking is a character-istic of the aged composites in this work. This behavior wasmainly a consequence of the improvement in adhesion for bothcellulosic pulp and PP fiber by the densification of the interfacialtransition zone with the reprecipitation of cement hydration prod-ucts �Mohr et al. 2006� and also due to the continued hydration ofthe cementitious matrix after aging �Bezerra et al. 2006�. In thecase of cellulose fibers, the anchoring is higher than their strengthand therefore they rupture. Furthermore, the mineralization �re-precipitation of cement hydration products into the fiber� of somecellulose fibers can contribute to the additional peaks in the post-cracking region. In general, aging led to the increase of LOP,MOR, and MOE of the composites for all levels of pulp freeness,however, it is observed that for the composites with refined pulps�SP-220 and SP-20� the higher peaks �MOR� occurred in the post-craking region after 100 aging cycles and 1 year of natural aging�Fig. 9�.

In Fig. 12, the synergy among the PP fibers and sisal pulp atdifferent freeness levels was evaluated as a percentage compari-son �positive or negative� of the mechanical properties of the
hybrid fiber-reinforced composites to that of the reference �rein-

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

forced with only sisal pulp�, as reported by Ding et al. �2010�.Generally, MOR and MOE of the composites decreased �negativepercentages� with the exchange of the sisal pulp by PP fibers, inrelation to those composites with only sisal pulp �Figs. 12�a andb��, implying that the hybrid has poorer MOR and MOE perfor-mances than the composites without PP fibers. This mechanicalbehavior was observed before and after the aging cycles and natu-ral aging. The MOE results were similar when the different pulpfreeness in composites with hybrid reinforcement �sisal+PP fi-bers� were compared, independent of aging.

Toughness is often correlated to the length of reinforcing fi-bers. As the stress is transferred from the matrix to the fiber,debonding can take place at the interface and the fiber may bepulled out from the matrix, generating considerable frictional en-ergy losses, which contribute to the toughness of the composite�Savastano et al. 2003�. Unrefined cellulosic fibers are longer and

Fig. 9. Typical stress versus strain curves for composites reinforcedand PP fibers: �a� unaged; �b� after 50 aging cycles; �c� after 100 agi

Fig. 10. Limit of proportionality �LOP� in function of the waterabsorption �WA� of composites exclusively reinforced with sisal pulpat different levels of pulp freeness and different aging condition andcomposites reinforced with sisal+PP fibers. Dashed line is the linearregression of all series together.



without fibrillation, which leads to poor anchoring into the matrixand considerable incidence of pulled out fibers. Coutts �1984�reported similar results of greater toughness associated to com-posites reinforced with unbeaten wood fibers when compared tothose with beaten fibers.

isal pulp at three levels of pulp freeness �CSF 20, 220, and 680 mL�les; and �d� after 1 year of natural aging

Fig. 11. SEM micrographs of the fracture surface of the hybrid com-posites with PP fibers and sisal pulp: �a� SP-220 after 50 aging cycles,rupture of sisal pulp fibers �arrow 1� and pull out of PP �arrow 2�fibers; �b� SP-20 after 100 cycles, disruption of the PP fiber �arrow�

with sng cyc

ARY 2011

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

Furthermore, for the unaged �28 days� composites with inter-mediate �SP-220� and lower �SP-20� freeness values the additionof PP fibers resulted in an increase of the toughness over com-posites reinforced with only sisal pulp �Fig. 12�c��, as a clearsignal of positive synergy. In this case, cellulosic fibers are fibril-lated and the anchorage is improved, the fracture prevails over thepull out and the toughness decreases for samples with only sisalfibers.

The use of refined pulp in composites with PP fibers �SP-220and SP-20� has caused an increase of at least 60% in the tough-ness of the composites at 28 days when compared to that withunrefined pulp �Fig. 5�c��. The positive synergy observed by en-hancement of the toughness for composites with refined pulp �SP-220 and SP-20� was partially provided by the improved retentionof cement particles �Coutts and Kightly 1982� and also avoidingthe stratification of the long and hydrophobic PP filaments alongthe thickness of the composite. These features turned the cemen-titious matrix more compact with the consequent improvement inthe contact of PP fibers, leading to more efficient reinforcementby the synthetic fibers. The fiber to matrix bonding may also bechanged during the aging of the composites. In cementitious com-posites, the microstructure around the fiber �interfacial transitionzone� is normally different from the bulk microstructure. Thepresence of this zone results in a gradient of microstructurecaused by the wall effect. The wall effect happens as preferen-tially smaller particles pack next to the fiber surface. Therefore,this special microstructure of the transition zone leads to a com-plex crack pattern. There is a reduction of toughness between 28days and 50 cycles for SP-20 and SP-220 caused by the weaken-ing of the cellulosic fibers in the matrix, but with time the tough-ness increases due to the improved adhesion of PP fibers in thematrix and consequently higher friction energy during their pull-ing out. Mohr et al. �2005� also reported the reduction and fol-lowing increasing of mechanical properties in composites with

Fig. 12. Variation of the mechanical properties of the compositesafter the addition of PP fibers in relation to composites with only sisalpulp fibers in the different pulp freeness: �a� variation in MOR; �b�variation in MOE; and �c� increment in toughness

unrefined and refined fibers with the evolution of wet and dry

JOURNAL OF MA


cycles. The lower pulp freeness �SP-20� did not provide signifi-cant increase in the toughness when compared with the interme-diate pulp freeness �SP-220� after both 100 aging cycles and 1year of natural weathering.

The extremities of the PP fibers, that were pulled out from thefractured cement matrix have expanded and exfoliated creating ananchoring effect �Fig. 13�. These structures were also found byBeaudoin �1990�. It was observed by the writer from X-ray analy-sis that the exfoliated material is organic and being identified asPP.

The maintenance of the toughness �T� after the aging wasaccomplished with the addition of PP fibers. When compared withcomposites exclusively reinforced with 4.7% by dry mass of sisalpulp after 1 year of natural aging, the PP fibers provided an in-crease of at least 200% in the toughness �T�. The PP fibers pro-vide higher absorption of energy with the attrition during thepulling out, and are responsible for the enhancement of the com-posite toughness. Higher values of toughness were found byBezerra et al. �2006� and Dias et al. �2010� for hybrid compositesreinforced with PVA fibers and cellulose pulps. Those improvedtoughness can be atributed to the higher contents of PVA fibersand other mineral additions used by the writers.

Conclusions

This present experimental study reported the use of sisal pulp in

Fig. 13. SEM micrographs of fracture surfaces of unaged �28 days�hybrid composites SP-20: �a� PP fibers pulled out of the matrix; �b�detail of the free ends expanded and exfoliated

conjunction with PP fibers, as an attempt for synergy of suitable


1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

hybrids reinforcements in low-cost fiber-cement application. Thesynergy effect of the hybrids reinforcements were simultaneouslyachieved by both different lengths of the reinforcing elements andalso functional aspects of the fibers. The addition of PP fibers hasimproved the toughness of the composites after 28 days of cure,especially those containing refined cellulose pulps. The decreaseof WA and capillary porosity in the composites with refined sisalpulp was a clear indication of the improved packing of the com-posites in relation to those with unrefined sisal pulp. This morecompact microstructure optimized the mechanical performance.In addition, refinement is responsible for the fibrillation of thefibers, which favored their anchorage into the cementitious ma-trix. It was inferred that high intensities of pulp refinement im-proved the mechanical performance �both MOR and toughness�of the composites with PP fibers, at the initial age �28 days� andafter aging as well. Cement based composites with hybrid rein-forcement of PP fibers and sisal refined pulp maintain their me-chanical properties even after aging with interesting performancefor possible future application for fiber-cement products.

Acknowledgments

Financial support for this research project was provided by Pro-grama Habitare-Financiadora de Estudos e Projetos �Finep�, Con-selho Nacional de Desenvolvimento Científico e Tecnológico�CNPq�, Coordenação de Aperfeiçoamento de Pessoal de NívelSuperior �Capes�, and Fundação de Amparo à Pesquisa do Estadode São Paulo �Fapesp�, Brazil. The writers were supported bygrants offered by CNPq, Capes, and Fapesp, Brazil. Sisal kraftpulp was provided by Lwarcel Celulose e Papel Ltda. and PPfibers and cementitious raw materials were kindly furnished byInfibra Ltda. The writers would also like to thank Dr. Marie-AngeArsène from Université des Antilles et de la Guyane for helpingwith SEM studies.

References

Agopyan, V., Savastano, H., Jr., John, V. M., and Cincotto, M. A. �2005�.“Developments on vegetable fibre-cement based materials in SãoPaulo, Brazil: An overview.” Cem. Concr. Compos., 27�5�, 527–536.

ASTM. �1981�. “Test method for dry and wet bulk density, water absorp-tion, and apparent porosity of thin sections of glass-fiber reinforcedconcrete.” ASTM C948, West Conshohocken, Pa.

Balaguru, P. �1994�. Thin reinforced concrete products and systems, ACI,Detroit, 25–42.

Banthia, N., and Gupta, R. �2004�. “Hybrid fiber reinforced concrete�HyFRC�: Fiber synergy in high strength matrices.” Mater. Struct.,37�10�, 707–716.

Banthia, N., and Sappakittipakorn, M. �2007�. “Toughness enhancementin steel fiber reinforced concrete through fiber hybridization.” Cem.Concr. Res., 37�9�, 1366–1372.

Beaudoin, J. J. �1990�. Handbook of fiber-reinforced concrete: Principlesproperties, developments and applications, Noyes Publications, ParkRidge, N.J.

Bentur, A., and Mindess, S. �1990�. Fibre reinforced cementitious com-posites, Elsevier Applied Science, London.

Bezerra, E. M., Joaquim, A. P., Savastano, H., Jr., John, V. M., andAgopyan, V. �2006�. “The effect of different mineral additions andsynthetic fiber contents on properties of cement based composites.”Cem. Concr. Compos., 28�6�, 555–563.

Brazilian Technical Standards Association �ABNT�. �1991�. “Compound



Portland cement—Test methodologies and aggregates.” NBR 11578,Rio de Janeiro, Brazil �in Portuguese�.

Brazilian Technical Standards Association �ABNT�. �2003�. “Cellulosicpulp—Determination of drenability—Canadian standard freeness.”NBR 14344, Rio de Janeiro, Brazil �in Portuguese�.

Britt, K. W. �1970�. Handbook of pulp and paper technology, 2nd Ed.,Chaps. 1–3, 5, Van Nostrand Reinhold, New York.

Campbell, M. D., and Coutts, R. S. P. �1980�. “Wood fibre-reinforcedcement composites.” J. Mater. Sci., 15�8�, 1962–1970.

Clark, J. A. �1987�. Pulp technology and treatment for paper, Chaps.6–8, 12–14, Miller Freeman, San Francisco.

Coutts, R. S. P. �1984�. “Autoclaved beaten wood fibre-reinforced cementcomposites.” Composites, 15�2�, 139–143.

Coutts, R. S. P. �1987�. “Fibre matrix interface in air-cured wood-pulpfibre cement composites.” J. Mater. Sci. Lett., 6�2�, 140–142.

Coutts, R. S. P. �2005�. “A review of Australian research into natural fibrecement composites.” Cem. Concr. Compos., 27�5�, 518–526.

Coutts, R. S. P., and Kightly, P. �1982�. “Microstructure of autoclavedrefined wood-fibre cement mortars.” J. Mater. Sci., 17�6�, 1801–1806.

Coutts, R. S. P., and Ridikas, V. �1982�. “Refined wood fibre-cementproducts.” Appita J., 35�5�, 395–400.

Coutts, R. S. P., and Warden, P. G. �1992�. “Sisal pulp reinforced cementmortar.” Cem. Concr. Compos., 14�1�, 17–21.

Dias, C. M. R., Savastano, H., Jr., and John, V. M. �2010�. “Exploring thepotential of functionally graded materials concept for the developmentof fiber cement.” Constr. Build. Mater., 24�2�, 140–146.

Ding, Y., You, Z., and Jalali, S. �2010�. “Hybrid fiber influence onstrength and toughness of RC beams.” Compos. Struct., 92�9�, 2083–2089.

Ding, Y., Zhang, Y., and Thomas, A. �2009�. “The investigation onstrength and flexural toughness of fibre cocktail reinforced self-compacting high performance concrete.” Constr. Build. Mater., 23�1�,448–452.

European Committee for Standardization. �1994�. “Fibre-cement profiledsheets and fittings for roofing—Products specification and test meth-ods.” EN 494, Brussels, Belgium.

Fardim, P., and Durán, N. �2003�. “Modification of fibre surfaces duringpulping and refining as analysed by SEM, XPS and ToF-SIMS.” Col-loids Surf., A, 223�1–3�, 263–276.

Ferreira, C. R., Fantini, M., Jr., Colodette, J. L., Oliveira, R. C., Gomide,J. L., and Carvalho, A. M. M. L. �2006�. “Technological assessmentof Eucalyptus wood clones: Part 2—Bleached pulp quality for print-ing and writing papers.” Scientia Forestalis, 71, 9–18.

Hannant, D. J. �1978�. Fibre cements and fibre concretes, Wiley, NewYork.

Hannant, D. J. �1998�. “Durability of polypropylene fibers in portlandcement-based composites: Eighteen years of data.” Cem. Concr. Res.,28�12�, 1809–1817.

Hannant, D. J., and Hughes, D. C. �1986�. “Durability of cement sheetsreinforced with layers of continuous networks of fibrillated polypro-pylene films.” Proc., RILEM Symp. on Developments in Fibre Rein-forced Cement and Concrete, R. N. Swamy, R. L. Wagstaffe, and D.R. Oakley, eds., University of Sheffield, Sheffield, U.K.

Ikai, S., Reichert, J. R., Rodrigues, A. V., and Zampieri, V. A. �2010�.“Asbestos-free technology with new high toughness polypropylene�PP� fibers in air-cured Hatschek process.” Constr. Build. Mater.,24�2�, 171–180.

Kalbskopf, R., De Lhoneux, B., Van Der Heyden, L., and Alderweireldt,L. �2002�. “Durability of fiber-cement roofing products.” Proc.,Inorganic-Bonded Wood and Fiber Composite Materials Conf., Uni-versity of Idaho, Moscow, Id., 347–353.

Kuder, K. G., and Shah, S. P. �2003�. “Effects of pressure on resistence tofreezing and thawing of fiber-reinforced cement board.” ACI Mater.J., 100�3�, 463–468.

Mehta, P. K., and Monteiro, P. J. M. �2006�. Concrete: Microstructure,properties, and materials, 3rd Ed., McGraw-Hill, New York.

Mindess, S. �1993�. “Fibre reinforced concrete challenges and prospects.”
Fibre reinforced concrete, N. Banthia and S. Mindess, eds., Univ. of
ARY 2011

1.23:177-187.

Dow

nloa

ded

from

asc

elib

rary

.org

by

ESC

OL

A P

OL

ITE

CN

ICA

- U

SP o

n 05

/08/

13. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

British Columbia, Vancouver, Canada, 1–11.Mohr, B. J., Biernacki, J. J., and Kurtis, K. E. �2006�. “Microstructural

and chemical effects of wet/dry cycling on pulp fiber-cement compos-ites.” Cem. Concr. Res., 36�7�, 1240–1251.

Mohr, B. J., Nanko, H., and Kurtis, K. E. �2005�. “Durability of kraftpulp fiber-cement composites to wet/dry cycling.” Cem. Concr. Com-pos., 27�4�, 435–448.

Nita, C., John, V. M., Dias, C. M. R., Savastano, H., Jr., and Takeashi, M.S. �2004�. “Effect of metakaolin on the performance of PVA and cel-lulose fibers reinforced cement.” Proc., 17th ASCE Engineering Me-chanics Conf., University of Delaware, Newark, Del.

Qian, C., and Stroeven, P. �2000�. “Fracture properties of concrete rein-forced with steel-polypropylene hybrid fibers.” Cem. Concr. Compos.,22�5�, 343–351.

RILEM.�1984�. “Testing methods for fibre reinforced cement-based com-posites.” Matériaux et Constructions, 17�102�, 441–456.

Savastano, H., Jr., Warden, P. G., and Coutts, R. S. P. �2000�. “Brazilianwaste fibres as reinforcement for cement-based composites.” Cem.Concr. Compos., 22�5�, 379–384.

Savastano, H., Jr., Warden, P. G., and Coutts, R. S. P. �2003�. “Potentialof alternative fibre cements as building materials for developingareas.” Cem. Concr. Compos., 25�6�, 585–592.

Savastano, H., Jr., Warden, P. G., and Coutts, R. S. P. �2005�. “Micro-

JOURNAL OF MA


structure and mechanical properties of waste fibre-cement compos-ites.” Cem. Concr. Compos., 27�5�, 583–592.

Somboon, P., Kang, T., and Paulapuro, H. �2007�. “Disrupting the wallstructure of high-freeness TMP pulp fibres and its effect on the energyrequired in the subsequent refining.” Pulp Pap. Mag. Can., 108�10�,30–34.

Soroushian, P., Marikunte, S., and Won, J. �1994�. “Wood fiber reinforcedcement composites under wetting-drying and freezing-thawingcycles.” J. Mater. Civ. Eng., 6�4�, 595–611.

Tonoli, G. H. D., Joaquim, A. P., Arsène, M.-A., Bilba, K., and Savas-tano, H., Jr. �2007�. “Performance and durability of cement basedcomposites reinforced with refined sisal pulp.” Mater. Manuf. Pro-cesses, 22�2�, 149–156.

Tonoli, G. H. D., Rodrigues Filho, U. P., Savastano, H., Jr., Bras, J.,Belgacem, M. N., and Rocco Lahr, F. A. �2009�. “Cellulose modifiedfibres in cement based composites.” Composites, Part A, 40�12�,2046–2053.

Tonoli, G. H. D., Santos, S. F., Joaquim, A. P., and Savastano, H., Jr.�2010�. “Effect of accelerated carbonation on cementitious roofingtiles reinforced with lignocellulosic fibre.” Constr. Build. Mater.,24�2�, 193–201.

Zollo, R. F. �1997�. “Fiber-reinforced concrete, an overview after 30 yearof development.” Cem. Concr. Compos., 19�2�, 107–122.


1.23:177-187.

Documents

Hybrid Reinforcement of Sisal and Polypropylene Fibers in ...€¦ · Fiber Preparation and Characterization Commercial unbleached sisal kraft pulp from Brazil was used in the experiments