Use of Cellulose Fiber From Hemp Core

Industrial Crops and Products 39 (2012) 89 96

Contents lists available at SciVerse ScienceDirect

Industrial Crops and Products

journa l h o me page: www.elsev ier .com

Use of cellulose bers from hemp core in ber-ceoccul ro

R. Jarabo utja Chemical Eng id 280b Faculdade de xias Nc Grup LEPAMA

a r t i c l

Article history:Received 21 NReceived in reAccepted 14 F

Keywords:Fiber-cementPulping conditHemp coreAgricultural wFlocculation

feasihroul and

of thired fetentiertieof theighe

1. Introduction

The use of agricultural wastes as reinforcement of compositematerials is one of the most important targets in todays materialsresearch (Aet al., 2009)wastes fromcellulose b

Industriathose of temintensive orequires crois biodegradThis, togethexplains thecomposite mThe most cocially the cothe most vabe separatesidered a wet al., 2011)

CorresponComplutense Tel.: +34 91 39

E-mail add

The use of cellulose bers as reinforcing agents in compositebuilding materials offers many advantages over glass bers, suchas the possibility to manufacture products with low density andgood biodegradability (Joshi et al., 2004; Li et al., 2006; Mwaikambo

0926-6690/$ doi:10.1016/j.hankari et al., 2011; Ashori and Nourbakhsh, 2010; Rabi. The present research is concerned with the use of the

the industrial hemp, Cannabis Sativa L., to obtain usefulers to manufacture ber-cement products for roong.l hemp is one of the crops having a highest yield amongperate zones, being at the same time one of and less

nes. It is highly self-compatible what means that itp rotation. Similar to other lignocellulosic bers, hempable and environmentally friendly (Mutj et al., 2007).er with its high strength and durability and low density,

increase in the use of hemp bers in the manufacture ofaterials (Carver, 1941; Hayo and van der Werf, 2004).

mmon material obtained from hemp is its bers, espe-arse ones which are extremely strong and durable. Asluable bers are located in the phloem, they must oftend from the xylem material (hemp core), which is con-aste in the hemp industry (Sedan et al., 2008; Trodec.

ding author at: Chemical Engineering Department, Chemistry Faculty,University of Madrid, Avda. Complutense s/n, Madrid 28040, Spain.4 42 45.ress: [email protected] (M.C. Monte).

and Ansell, 2003). However, cellulose bers also have some disad-vantages, for instance, they present low modulus of elasticity, highmoisture absorption, they decompose in alkaline environmentsand are susceptible to biological attack, and they have variablemechanical and physical properties (Swamy, 1990). Some of thesedisadvantages could be overcome using bers from hemp core witha higher durability than the traditional cellulose bers from pineand with a specic tensile strength of 0.750 GPa, and specic mod-ulus of elasticity of 60 GPa. These properties make hemp bers acandidate material with potential as reinforcement ber (Li et al.,2004).

Many authors have studied the use of hemp bers as reinforce-ment for building materials based on cement (Dalmay et al., 2010;Placet, 2009; Sedan et al., 2008; Trodec et al., 2011). However, theuse of hemp core bers for this purpose has not been yet reported.This research aims to contribute to ll this lack of knowledge.

In general, cell wall polymers and their matrices determine themechanical and physical properties of natural bers. In view of this,a number of researchers have reported the chemical compositionof natural cellulose bers and some of them have recommendedimproving the natural cellulose ber properties by modifying thecell wall polymers (Golbabaie, 2006; Wielage et al., 2003; Young,1997). Furthermore, the incorporation of cellulose bers in a poly-mer or mineral matrix can involve an interface incompatibility

see front matter 2012 Elsevier B.V. All rights reserved.indcrop.2012.02.017ation, retention, drainage and product pa, E. Fuentea, M.C. Montea,, H. Savastano Jr. b, P. Mineering Department, Complutense University of Madrid, Avda. Complutense s/n, MadrZootecnia e Engenharia de Alimentos, Universidade de So Paulo, Avenida Duque de CaP, Universitat de Girona, Avd. Llus Santal S/N, 17071 Girona, Spain

e i n f o

ovember 2011vised form 9 February 2012ebruary 2012

ions

astes

a b s t r a c t

The aim of this paper is to study thewaste, hemp core (Cannabis Sativa L.), tber-cement production. The physicadepend on the nature and morphologymust be kept between the limits requhow its use affects the occulation, rand the mechanical and physical prop

The use of pulp obtained by means 180 C during 90 min resulted in the hthe studied hemp core pulps./ locate / indcrop

ment production. Effect onperties

c, C. Negroa

40, Spainorte, 225, 13635-900 Pirassununga, SP, Brazil

bility of using cellulose bers obtained from an agriculturalgh different new environmental friendly cooking processes for

mechanical properties of the ber reinforced concrete, whiche bers, matrix properties and the interactions between them,or its application. Therefore, the morphology of the bers andon and drainage processes in the ber-cement manufacture,s of the ber-cement product have been studied.

hemp core cooking in ethanolamine at 60% concentration atst solids retention and the best mechanical properties among

2012 Elsevier B.V. All rights reserved.

90 R. Jarabo et al. / Industrial Crops and Products 39 (2012) 89 96

Table 1Cooking conditions of hemp core using semi-chemical process.

Pulps T (C) Time (min) NaOH (%) Yield (%)

P1 P2

Table 2Cooking condi

Pulps

P3P4

between bchemical prthe chemic(e.g., electroeffect of thtions betweefciency wber (Negro

The prosources detfore, it affecof this stagea semi-chemboth at diffthe cellulos

First, a cocarried out on the berthe study ocement suseffect of usiproperties o

2. Materia

2.1. Materi

The rawfrom the cuLEPAMAP g

The hemlulose and to obtain hseparated, ocore bers. 10 mm (Mu

Tables 1pulp by a secess emploand the yie

The ethacial produc

Organosrotator digetrol. The puliquor to soexperimentand pulping

The pulpwith a solualyzer of thin a hydrap

Table 3Composition of the ber-cement suspensions studied.

Raw materials Dry weight (%)

se 5% II cemsilica

withd, cru

conus stcore ned

sincmanuortlar thee sizo 50rosilf micsphemou

ocsionc polek p06 g/ed intrati

com sumed, bM1 ws preulp wed asouglly uironm

eredeted to

ethod

Dens beomet

radi (10

ore tC forapplyred:180 90 25 49.1140 30 15 68.8

tions of hemp core using organosolv (chemical) process.

T (C) Time (min) Ethanolamine (%) Yield (%)

180 90 60 28.2155 30 40 53.6

ers and matrix, which may be overcome through theetreatment of bers, with the aim of modifying eitheral nature of the ber surface or the surface propertiesstatic charge, conformation of adsorbed polymers). The

is treatment is probably a modication of the interac-en the mineral matrix and bers, which increases theith which stress is transferred from the matrix to the

et al., 2005; Ouajai and Shanks, 2005).cess through which bers are obtained from naturalermines the properties of the bers surface and, there-ts the interface matrixber surface. To study the effect

on the nal product two different treatments, namelyical pulping process and a chemical pulping process,

erent cooking conditions have been applied to obtaine bers from the hemp core.mprehensive study of the morphology of the bers was

(Jarabo et al., 2012); then, the effect of using these bers-cement manufacture process was determined throughf the occulation, retention and drainage of the ber-pensions prepared with the different bers. Finally, theng the obtained bers on the physical and mechanicalf ber-cement probes was determined.

ls and methods

als

material used for this study was the hemp core, wastelture of C. Sativa L. grown in Spain and supplied by theroup.p ber is constituted by 43% cellulose, 16% hemicel-

8% lignin. Hemp straw is a sub-product in the processemp strands from stalk of the plant. This stalk can ber decorticated, into two main components: bers and

The initial length of these core bers lies between 5 andtj et al., 2007; Barber et al., 2011).

and 2 show the cooking conditions used to obtain hempmi-chemical process and a chemical (organosolv) pro-ying NaOH and ethanolamine as solvents, respectively,ld obtained in each case.nolamine PA-ACS solution was supplied as a commer-t by Panreac.

CelluloASTMMicro

screenpresse

Thepreviohemp

Reerencein the

A Pused foparticl2 m t

Mictype ophous small a

ThesuspenanioniHatsch7.4 1dissolvconcen

Theries ispreparthem: M2 waKraft pwas us

Althnormato envbers win the ascribe

2.2. M

2.2.1. The

tipycnatomicing 1 A

Befat 60

times measuolv pulping were performed in a 25-L stainless steelster with a heat exchanger system and pressure con-lping conditions were maintained constant as follows:lid ratio: 6:1 and digester pressure 57 bars. Selectedal conditions were: solvent concentration, temperature

time.ing with NaOH was carried out in the same digester buttion of NaOH having 1.4 g/L of anthraquinone as a cat-e chemical reaction. The cooked pulps were deberedulper and screened through a Somerville vibratory at

Vp (cm3) =

Density (g

where Vp iholder (16.the pressur(PSI), P2 is the W is theent 91%4%

0.15 mm slot size; the screened pulp was washed,mbled and stored at 4 C.

ditions chosen in these tables are the result best from audy carried out to optimize these cooking processes forbers (Barber et al., 2011).unbleached pine Kraft pulps (35 SR) were used as ref-e this pulp is commonly used to provide cellulose bersfacture of ber-cement through the Hatschek process.

nd cement (type II/AV 42.5) containing 12% y ash was probes. It is a ne powder with a wide distribution of

es, being the 80% of the particles in the interval from m.ica was also used to manufacture of the test probes. Therosilica employed was composed of ultra-thin amor-res of SiO2 with a particle size around 0.5 m containingnts of crystalline quartz (less than 0.5%) as impurities.culant employed to study the behavior of ber-cements and to prepare the ber-cement probes was anyacrylamide (APAM) commonly used in the industrialrocess (Negro et al., 2006) with a molecular weight ofmol and a charge density of 13.4%. The occulant was

distilled water to prepare solutions of APAM with aon of 1.5 g/L.position of the prepared ber-reinforced cement slur-marized in Table 3. Five ber-cement slurries wereeing the source of cellulose the only difference betweenas prepared using pulp P1 as a source of cellulose bers,pared with P2, M3 with P3, M4 with P4 and the pineas used to prepare the ber-cement slurry MR, which

a reference.h in air curing ber-cements synthetic bers (PVA) aresed due to their capacity to withstand degradation dueental conditions, in the present research, no synthetic

used to prepare the probes so as to avoid interferencesrmination of the ber-cement properties that can be

the use of cellulose bers.

s

ity of the bersr density was measured using a gas pycnometer (Mul-er). Helium was the gas employed because its smallus enables it to enter into crevices and pores approach-10 m).he measurement, the samples were dried in an oven

24 h. The volume of each sample was determined veing Eqs. (1) and (2) to the pressure and volume values

Vc Vr((

P1P2

) 1

)(1)

/cm3) = WVp

(2)

s the sample volume, Vc is the volume of the sample

163 cm3), Vr is the reference volume (8.068 cm3), P1 ise measurement after pressurizing the reference volumethe pressure measurement after including Vc (PSI) and

dry weight of sample.

R. Jarabo et al. / Industrial Crops and Products 39 (2012) 89 96 91

2.2.2. Morphological characterization of bersThe morphological characterization of the bers was carried

out using a ber and pulp morphological analyzer, Mor, V7.9.13.E(Techpap, France).

This chaconsisting ocamera. Imbers, whicysis, whichwere analyparametersweighted incoarseness,

Fines arepulps weretheir dimenlength valu75 m, neet al., 2012;

The samby adding 1the suspensterization w

2.2.3. FloccA comm

(FBRM) M5employed tthe propertlength distrcentrationsa focal poinwindow, whcircular patthe focusinthe reecteoptical bethe intercepduration ofmovement.In this waypulses rececase). Each ticles in sureects thecal parametevolution, emeasured pand Schell,

In the pimmersing pared withAfter 10 minAPAM was lution of thstirring inteformed (dereduced aga(Jarabo et a

2.2.4. RetenThe equ

was a vacuua barrier: thup to the ada mesh in th

and it is connected to a vacuum pump and to a probe in whichthe ltrate is stored and weighted in real time. The nal volume ofltrate is measured. In a typical trial, 400 mL of ber-cement sus-pension, prepared with water saturated in Ca(OH)2, were stirred

rpmcrea

Afterringn waacedh thened o obtd cais of

Prepar rerougals wsolidble 2

wasas fowatestic y fore ba

emoecimrepa

mecinat

Mechchane Emg con

rupity (Mtionspan

thely recsing site f

(MPa

Pa)

(kPa)

/m2)

LMAXf the

majotivelyeec

absoint coum s

mece coracterization is based on an image analysis system,f a diode that emits unpolarized light and a micro-

aging is performed until the equipment counts 5000h is the optimum value for subsequent statistical anal-

is performed using a computer software. The imageszed using a specic program to determine different

of the bers and pulps: bers, arithmetic length, length length, average length weighted in area, average width,

number of microbrills and nes length. generated mainly during the beating stage when the

debered. There are differentiated from the bers bysions. The equipment was programed to consider beres between 100 and 10,000 m, width between 5 ands length below 100 m and width below 5 m (Jarabo

Moral et al., 2010).ples for morphological characterization were prepared

g of dry bers to 600 mL of water and homogenizingion in a laboratory disintegrator ENJO-692. The charac-as done in duplicates.

ulation of ber-cement suspensionercial focused beam reectance measurement probe00L, manufactured by Mettler Toledo, USA, waso monitor the occulation process and to determineies of the ocs. The FBRM device measures the chordibution on real time over a wide interval of solid con-. A laser beam is generated by a diode and focused ont in the plane next to the external surface of the probeich is inside the suspension. The focal point describes a

h at a constant speed of 2000 m/s due to the rotation ofg lens. When a particle intercepts the focal point path,d light reaches the detector through the probe and ther. The detector receives light pulses. A chord length ofting particle is determined as the product of the time

the light pulse by the linear speed of the focal point Thousands of particles can be measured each second., a chord length distribution is obtained from the lightived during the measurement time (set in 5 s in thisdistribution is a function of the size and shape of par-spension. The evolution of the chord size distribution

aggregation or dispersion of particles. Many statisti-ers can be calculated from the distribution to follow its.g., the mean chord size and the total number of countser second (Blanco et al., 2002; Hubbe, 2007; Kerekes1992; Negro et al., 2007).resent study, the occulation trials were carried outthe probe in a 400 mL ber-cement suspension, pre-

water saturated in Ca(OH)2, and stirred at 800 rpm., stirring intensity was reduced to 400 rpm. 100 ppm of

added 5 min later, to induce occulation, and the evo-e ocs was studied at 400 rpm for 4 min. After that, thensity was increased to 800 rpm to break down the ocsocculation) for 2 min and, nally, stirring intensity wasin to 400 rpm to induce the reocculation of the systeml., 2010).

tion and drainage of ber-cement suspensionipment used for measuring the retention and drainagem drainage tester (VDT). It has two jars separated bye upper jar keeps the ber-cement suspensions stirreddition of the occulant dosage. The second jar containse bottom to carry out the dewatering of the suspension

at 600was deadded.the stipensiowas plthrougof draiorder ttion ananalys

2.2.5. Fibe

tory thmateriwith a evacuagauge)cake wof the in a plathis wafrom thwere rtest spwere ption ofdeterm

2.2.6. Me

machinbendinulus ofelasticcalcula

A sused indigitalprocescompo

MOR

LOP (M

MOE

SE (kJ

wherepoint ois the respecis the d

Thethe pomaxim

Theafter th during 6 min in the upper jar. Then, stirring intensitysed to 300 rpm and, 5 min later 100 ppm of APAM werer 15 s of contact time between occulant and mixture,

was stopped, the barrier was removed and the sus-s drained to the second jar in which an 18 mesh wire. The suspension was drained under vacuum (0.2 atm)

lter and a computerized balance recorded the masswater over time. The drainage curve was analyzed inain the drainage rate for the different occulants. Reten-ke humidity were determined gravimetrically after thethe formed cake (Fuente et al., 2010).

ration of ber-cement probesinforced cement probes were prepared in the labora-h a slurry vacuum de-watering technique. The matrixere added and dispersed in Ca(OH)2 saturated waters concentration of 20%. The slurry was transferred to an00 mm 200 mm casting box and a vacuum (60 kPa

exerted until excess water was removed and a solidrmed. The pad was pressed at 3.2 MPa, so that the restr was removed. After pressing, the plates were sealedbag at room temperature. Three pads were prepared in

each formulation. After 2 days; the pads were removedgs and placed in water. Twenty-six days later, the padsved from the water and four 200 mm 50 mm exuralens were diamond wet-sawn from each pad. Eight padsred to provide sufcient specimens for the determina-hanical properties and four pads were prepared for theion of physicals properties.

anical properties of ber-cement probesical tests were performed in the universal testingic DL-30,000 equipped with 1 kN load cell. A four pointguration was employed in the determination of mod-

ture (MOR), limit of proportionality (LOP), modulus ofOE) and specic energy (SE) of the specimens following

specied in Eqs. (3)(6).of 100 mm and a deection rate of 0.5 mm/min were

bending test (Savastano et al., 2000). Test data wereorded and reduced using automatic data collection andfacilities. Eight exural specimens were tested for eachormulation.

) =(

LMAXb h2

) (Sdown Sup) (3)

=(

LLOPb h2

) (Sdown Sup) (4)

= tg(

LMAX

) (Sdown Sup)

3

b h3 (5)

= Absorbed energyb h (6)

is the maximum stress, LLOP is the stress at the upper linear portion of the stressstrain curve, (Sdown Sup)r span, b and h are the specimen width and thickness,, tg is the initial slope of the stressstrain curve and tion of the composite.rbed energy is the area under the stressstrain curve torresponding to a reduction in carrying capacity to thetress (Tonoli et al., 2010).hanical properties of the probes were measured 28 daysnstruction of the sheet.


Table 4Morphological characterization of pulps.

Process Semi-chemical Organosolv Kraft

Pulps P1 P2 P3 P4 PR

FibersFibers (106/g 45.2 28.5 10.8Arithmetic l 345 352 456Length weig 459 497 1129Average leng 472 526 1344Average wid 24.0 27.3 25.5Coarseness ( 0.06 0.10 0.18Microbrills 1.68 1.48 1.62

FinesFines numbe 8947 15,474 49,192Fines length 19.5 25.1 40.3

2.2.7. PhysiWater a

were obtainprocedures

Four spe

3. Results

3.1. Morph

The resubers and pprocess arefor compari

It is notifrom the heTherefore, tthan that omatrix.

P1 and Plength weigwhich lowecan be due tpulps P1 antion during

Regardinsolv), the oand with lo

NaOH usamorphousthe surfacethe coarsencoarseness Organosolvcelluloses abe used to oorganosolv surface is h

The percPR and higthe interactimprove th

It has bequick and 2011).

3.2. Floccul

Fig. 1 shand ocs in

r 600 s of stirring at 400 rpm, the value of the mean chordas constant. 900 s after starting the trial, the addition of oc-

to the suspension caused a fast increase in the mean chorde to the aggregation of particles to form larger ocs. A max-value of this parameter was reached at 10 or 15 s and thened decreasing with different rate depending on the stabilityength of the ocs in such hydrodynamic conditions (evolu-

ocs). When the stirring was increased to 800 rpm, part ofaining ocs was broken, thus decreasing the mean chord

eoc by tcreaan beachent frion ocs fover, hord

suspand er theose

resuare mg read theded

4) 32.0 61.2 ength (m) 356 389hted in length (m) 539 574 th weighted in area (m2) 583 617 th (m) 19.2 29.0 mg/m) 0.08 0.40

(%) 1.38 1.66

r 13,149 30,470 (%) 25.4 32.7

cal properties of ber-cement suspensionbsorption, bulk density and porosity values at 28 daysed from the tested exural specimens following the

specied in ASTM C 948-81 (ASTM C 948-81).cimens were used to determine these properties.

ological characterization of pulps

lts of the morphological characterization of hemp coreulps obtained by both semi-chemical and organosolv

shown in Table 4. Kraft pine pulp was also characterizedson.ceable that the length of the bers and nes obtainedmp core is much shorter than that of the pine bers.he number the bers per gram on hemp pulps are highern pine pulp. This could improve their dispersion in the

3, presented lower values of bers arithmetic length,hted in length, average width and coarseness alongr nes length than those for the pulps P2 and P4. Thiso the more aggressive cooking conditions used to obtaind P3, which increase the probability of bers degrada-

the process.g the effect of the process (semi-chemical or organo-

rganosolv pulps (P3 and P4) had slightly shorter berswer coarseness than the semi-chemical pulps.ed in semi-chemical process is well-known to remove

materials, such as hemicelluloses and pectins, from of hemp bers (Trodec et al., 2009); this reducesess of the bers, which is lower for P1. However, theof the organosolv bers is even lower than that of P1.

process consists of dissolving lignin, extractives, hemi-nd pectin in ethanolamine but in such way that they canbtain valuable chemical products. The efciency of theprocess in removing these compounds from the berigh even when soft cooking conditions is applied.

Aftesize wculantsize duimum it startand strtion ofthe remsize (dshownwas de

It cwas rediffereevolutthe oMoreomean cthe MRin M1 but aftthan th

Theerties cookinfavorewas ad

d s

ize (

m

)

50

60

70

80entage of microbrills is similar in pulps P2, P3 andher than in the two other pulps. This could increaseion among the bers and the matrix and, consequently,e properties of the ber-cement.en demonstrated that a ber analyzer (MorFi) can gainreliable results (Bartl and Pico, 2009; Pico and Bartl,

ation of ber-cement mixtures

ows the evolution of the mean chord size of particles cement suspensions.

800

Mean

ch

or

10

20

30

40

Fig. 1. Evolutireocculationculation). The reocculation ability of the system ishe increase in the mean chord size when the stirringsed to 400 rpm again.

observed that the largest maximum mean chord sized in the MR suspension. This value, however, is not toom those reached in suspensions M2 and M4 and thef the mean chord size in these two cases indicates thatrmed were more stable than those formed from MR.after the deocculation and reocculation stages, the

size in M2 and M4 remained larger than in the case ofension. The maximum mean chord size values reached

M3 were lower than those reached in the other cases, evolution stage the values obtained were slightly lower

in MR.lts indicate that the occulation process and oc prop-ore affected by the cooking conditions than by the

gent used, and that the use of soft cooking conditions formation of larger and more stable ocs when APAMto the ber-cement pulp.

MR

M1

M2

M3

M4

00 rpm 40 0 rpm800 rpm

locc

ulat

ion

lutio

n of

floc

s

tion

cula

tionTime (s)

1300120011001000900

100 ppm

APAM

F

Evo

Def

locc

ula

Reflo

c

on of the mean chord size during the occulation, deocculation and of the ber-cement suspensions.


Table 5Retention and humidity of ber-cement suspensions.

Fiber-cementsuspension

Source of bers Retention (%) Humidity (%)

M1 P1 66.5 1.87 46.8 2.05M2 P2 29.1 4.90 63.7 0.80M3 P3 64.7 1.57 50.5 0.90M4 P4 18.5 5.70 64.7 1.33MR Unbleached Kraft pine 57.3 2.30 57.9 1.20

3.3. Retention and drainage of ber-cement suspensions

Fig. 2 shows the drainage curves of the ber-cement suspensionsconsidered in this research.

Drainage took place in two steps: rst, the suspension was l-trated and a cake was formed with a fast water removal, whichcorresponds to the rst part of the drainage curves (linear partwith a high slope); secondly, the cake was compressed and thick-ened and water removal rate decreases toward zero. During therst stage, only the water between ocs was removed, while partof the water inside the ocs was removed during the compressionstage. Most of the solids loss with the ltrate takes place duringthe rst stage while the second stage determines the nal humid-ity and the formation properties of the cake. Drainage time can beobtained asto zero.

There arthe ber-ceone for MRintensity ofest values fcompressioweight of rthe case of

Table 5 scake was foimum valuetogether wisolids, werestage in theremaining iscopicity extwo cases.

These dition of partiand from th

0

Filtr

ate

(g

)

0

100

200

300

400

500

F

0

0.5

1

1.5

2

2.5

3

3.5

4003002001000

%

Chord size (m)

MR

M1

M2

M3

M4

Fig. 3. Chord lengths distribution of the ber-cement suspensions after the additionof the APAM.

Fig. 3 shpensions Mwere muchspaces amofast. This e

d MRose form

amoge ra

choumeThe sutesse ineten

the an w

echa

4 shorve otainum ee of Mhaviothe clid paan M

M1 M2 M3 M4 MR the time required to reduce the drainage curve slope

e notable differences between the drainage curves ofment suspensions containing hemp core bers and the. The initial drainage rate was notably affected by the

cooking conditions of the pulp and it took the low-or M1 and M3 suspensions. There was no appreciablen stage in the drainage curves of M2 and M4. The nalecovered ltrate was also higher in these cases and inM1.hows that the solids retention was very low when thermed from suspensions M2 and M4, taking its max-

in the case of suspensions M1 and M3. These values,th the drainage curves, indicate that most of the mineral

lost with the water drained during the rst drainage case of M2 and M4 with only some of the mineral solidsn the cake and the majority of bers, whose high hygro-plains the high humidity of the cakes formed in these

fferences can be explained from the chord size distribu-cles in suspension after the occulation process (Fig. 3)e oc properties.

MR

M1

M2

M3

M4 anthan thken, despacesdrainasolids.

Thewere nsions. contribdecreawhen rtent ofless th

3.4. M

Fig.The cuthat obmaximthe casical beunder and sorigid th

4

6

8

tre

ss

(M

Pa

)Time (s)

14012010080604020

M4

ig. 2. Drainage curves of the ber-cement suspensions.

0

2

0.00

S

Fig. 4. ows that the chord size distributions of occulated sus-2, M4 and MR were similar, and that the ocs formed

larger than in the case of M1 and M3. Therefore, theng the ocs are also larger and water can drain veryxplains the high initial drainage rate observed by M2,. However, the ocs formed from MR are much weakerformed from M2 and M4 and, therefore, they were bro-ed and compressed by the vacuum forces causing theng the ocs to diminish, which implies a decrease in thete (compression stage) and a higher retention of mineral

rd size distributions of M1 and M3 show that thererous small ocs and particles present in these suspen-paces among the ocs were, therefore, smaller, what

to an increase in the retention of solids, but causes a the drainage rate. The humidity of the cake decreasestion increases, and this is due to the higher mineral con-cake (water absorbed by the mineral particles is muchater absorbed by the bers).

nical and physical properties of the probes

ws the stressstrain curves of the specimens at 28 days.btained with M3 resulted to be the most similar one toed with MR is, although it is worth noticing that thelongation before the failure of the probe was higher in3, while the maximum stress was lower. This mechan-

r, high elongation and stress reached, and the large areaurves can be associated with a good array of the bersrticles into the composite. The other probes were more3 and MR, as shown by their low strain values.0.020.01

Strain (mm/mm)

Stressstrain curves of the composite reinforced with hemp.


0.00

2.00

4.00

6.00

8.00 a b

c

MO

R (

MP

a)

0

000

000

12000

16000

a)

0.00

2.00

4.00

6.00

8.00

LO

P (M

Pa)

.00

.15

.30

.45

Fig. 5. lus of

Thereforthe best mepared with

Fig. 5 shof rupture, energy of thtwo applied

These gbers, M3 ylimit of prowere lowerworth poinwere reneand degradbers and mthan those onicant effenot contribas ller andsuspension

Furthermprepared frobtained fotg.

A largera better disbearing capber was ususe of pine uct and it mthe beatingimproved

Despite ties, the phporosity of considered.bers from are probablmatrix.

of tbilitdroxh hy

of t watas sleasoGolbs wases.

resuits o

n Sta341-

d

MRM4M3M2M1

4

8

MO

E (M

P

MRM4M3M2M1

0

0

0

0

Sp

ecif

ic E

nerg

y (

KJ/m

2)

Effect of the hemp on the mechanical properties: (a) modulus of rupture, (b) modu

e, the probes prepared from M3 may be the ones withchanical properties among the ber-cement probes pre-hemp core bers.ows the results of the mechanical properties: moduluslimit of proportionality, elastic modulus and specice probes of ber-cement made of hemp pulps with the

processes.ures show that, among the mixtures with hemp coreielded the probes with the highest modulus of rupture,portionality and specic energy, although these values

than the values obtained with the reference MR. It isting out that the bers used in the preparation of MRd bers. Rening increases the presence of microbrillse ber surface, which enhances the interaction amongatrix. Pine bers and nes lengths are notably greaterf hemp core bers and these parameters also have a sig-

Oneis its athe hythrougchange

TheM4 wmain rloses (procesproces

Thethe limChileaNo. 21ct on the mechanical properties of the probes. Fines doute signicantly to ber-cement strength but act more, in most cases, inuence negatively the drainage of the

(Tonoli et al., 2009).ore, Fig. 5 shows that the elastic modulus of the probes

om M4 were the highest, even higher than the valuesr the reference probes. This is due to the high value of

number of bers per gram could be associated withtribution of bers in the matrix and thus fewer loadsability. However, this was not the case when hemp coreed. Despite on the lower number of bers per gram, theimproved the mechanical properties of the nal prod-ay be attributed to both the superior ber quality and

process, which produces brillation and consequentlyber matrix bonding (Savastano et al., 2000).the large differences observed in mechanical proper-ysical properties (water absorption, bulk density andthe probes) are similar (Fig. 6) between the composites

Therefore, they do not explain the effect of the use ofhemp core on the mechanical properties. These effectsy related to the interaction between the bers and the

4. Discussi

The effefor ber-cebers morpselves depeto the berat hard coosmall ocs ((Fig. 2) andlower stren(Table 5), pand the othThis indicaM1 and M3suspension

Albeit aM3, the mewere betterthe bers mrelated to tpercentagefore, it affeMRM4M3M2M1

MRM4M3M2M1

elasticity, (c) limit of proportionality and (d) specic energy.

he drawbacks of the use of cellulose in ber-cementy to absorb moisture from the environment. Due toyl and oxygen-containing groups, moisture is retaineddrogen bonding. This causes a swelling and dimensionalhe bers affecting the composites.er absorption of the probes prepared from M3 andightly lower than those from M1, M2 and MR. Then for moisture absorption is attributed to hemicellu-abaie, 2006), whose removal through the organosolvs more efcient than those for semi-chemical and Kraft

lts obtained with hemp core and pine bers fall withinf physical and mechanical properties according to thendard (Norma chilena NCh186/1-2003) and the DecretoMEIC.on

ct of using hemp core bers in the Hatschek processment suspensions and probes is closely related to thehology and to the occulation process, which them-nd on the cooking conditions. The addition of APAM-cement suspensions prepared with the pulps obtainedking conditions (M1 and M3) induced the formation ofFigs. 1 and 3), which slowed down the drainage process

an enhancement of the compression stage due to thegth of the ocs. This increased the retention of solidsart of them by means of their interaction with the berer part through blocking the small voids among ocs.tes a better performance in mechanical properties of

ber-cement probes comparing with the ber-cements M2 and M4; however, the drainage is affected.

high retention solid was obtained for both M1 andchanical properties of the probes obtained from M3

than those for M1. This can be explained consideringorphology. Although the main effect on the process ishe cooking conditions, the pulping process affects the

of microbrills, the length, and the coarseness; there-cts the interaction between bers and matrix and the


0.00

5.00

10.00

15.00

20.00

a

b

c

Wa

ter

Ab

so

rpti

on

(%

by

ma

ss

)

0.000

0.400

0.800

1.200

1.600

Den

sit

y (

g/c

m3)

0.00

10.00

20.00

30.00

40.00

Perm

eab

le v

oid

vo

lum

e

(% b

y m

as

s)

Fig. 6. Effect odensity and (c

dispersion microbrillincrease thebetween thneous surfabetween thbe better fo

In additimay explainby the incre

At the levbers coueffects astension, c

It is also pcellulose occur in t

5. Conclus

The novmaterials, pfriendly and

of ber-cement through new environmentally friendly cookingprocesses.

The cooking process and cooking conditions used to preparethe pulp from the hemp core determined the feasibility of the use

pulp. Coo

90 m for nicaler, anditemen

wled

auttion 07-6

to active we Unill as

Diasity ledg

of GMRM 4M3M2M1

MRM 4M3M2M1

Bulk density Fibers density

of the factureduringstudiedmechaHowevpulp cober-c

Ackno

TheInnovaCTM20Jarabolaboraand thedge aZaqueuUniveracknowversityMRM 4M3M2M1

f the hemp on the physical properties: (a) water absorption, (b) bulk) permeable void volume.

of bers in the matrix. The increase in the number ofs, and the slight decrease in the length and coarseness

surface area, which is available for chemical bondinge bers and the matrix. This leads to a more homoge-ce made of cellulose, which will enhance the adhesione bers and the matrix making this ber-cement probermed.on, there are other mechanisms and phenomena that

the higher modulus of rupture of M3 probes inducedase in the microbrills:

el of the microbrils, the longitudinal tension in theld involve not only tensional stresses, but also torsion

well. This torsion stress, which initially impedes theould subside or relax as a result of the repeated stress.robable that rearrangements and re-orientations of themicrobrils and/or changes in the crystalline fractionhe bers (Placet, 2009).

ions

elty of this paper lies in the use of agricultural wastearticularly, industrial hemp straw (C. Sativa L.), as eco-

renewable source of cellulose bers in the manufacture

References

Ahankari, S.S.,reinforcedposites: coTechnol. 7

Ashori, A., Noresidues. W

ASTM (Americdard TestApparent Conshoho

Barber, L., Pfor paperm865872.

Bartl, A., Pico741746.

Blanco, A., Fuebeam ree734740.

Carver, G.W., Pop. Mech

Dalmay, P., SmPropertieson the pro

Decreto No. 21NCR 193:1

Fuente, E., Jarsepiolite oConstr. Bu

Golbabaie, M.,of Plant Ag

Hayo, M.G., vathe effect

Hubbe, M.A., 2review of (2), 2963

Jarabo, R., Fuesepiolite oConcr. Res

Jarabo, R., Moncultural reCrops Prod

Joshi, S.V., Drzenvironme35, 3713

Kerekes, R.J., Scrowding

Li, Z., Wang, Lreinforceds obtained as source of bers for ber-cement manu-king the hemp core in ethanolamine at 60% at 180 Cin was the optimal pulping process among the optionshemp core pulping. These conditions provided the best

properties for hemp core ber ber-cement probes. more thorough optimization of the pulping process andioning (e.g. rening) based on the effect of bers on thet mechanical properties would be required.

gments

hors are grateful to the Ministry of Science andof Spain for the nancial support to the Project6793-C03-03 and for funding the scholarship of Rococomplish her Ph.D. Thesis. This research is part of a col-ork between the Universidad Complutense de Madrid

versity of Sao Paulo; the authors would like to acknowl-sistance offered by Mr. Gustavo H.D. Tonoli and Mr.s de Freitas at the Laboratory of Rural Construction ofof Sao Paulo, Brazil. Finally, the authors would like toe the contribution of the LEPAMAP group of the Uni-irona in supplying hemp pulps.

Mohanty, A.K., Misra, M., 2011. Mechanical behaviour of agro-residue poly(3-hydroxybutyrate-co-3-hydroxyvalerate), (PHBV) green com-mparison with traditional polypropylene composites. Compos. Sci.1, 653657.urbakhsh, A., 2010. Bio-based composites from waste agriculturalaste Manage. 30, 680684.

an Society for Testing and Materials), 2000. ASTM C948-81, Stan- Method for Dry and Wet Bulk Density, Water Absorption, andPorosity of Thin Sections of GlassFiber Reinforced Concrete, Westcken, USA.lach, M.A., Prez, I., Puig, J., Mutj, P., 2011. Upgrading of hemp coreaking purposes by means of organosolv process. Ind. Crops Prod. 34,

, D., 2009. Characterization of short bers. Chem. Eng. Trans. 17,

nte, E., Negro, C., Tijero, J., 2002. Flocculation monitoring: focusedctance measurement as a measurement tool. Can. J. Chem. Eng. 80,

1941. Auto body made of plastics resists denting under hard blows.. Mag. 76 (6).ith, A., Chotard, T., Sahay-Turner, P., Gloaguen, V., Krausz, P., 2010.

of cellulosic bre reinforced plaster: inuence of hemp or ax bresperties of set gypsum. J. Mater. Sci. 45, 793803.341-MEIC, 1992. Lminas planas de brocemento. Especicaciones.992. La Gaceta No. 127.abo, R., Moral, A., Blanco, A., Izquierdo, L., Negro, C., 2010. Effect ofn retention and drainage of suspensions of berreinforced cement.ild. Mater. 24, 21172123.

2006. Applications of Biocomposites in Building Industry. Departmentriculture, University of Guelph.n der Werf, 2004. Life cycle analysis of eld production of bre hemp,of production practices on environmental impacts. Euphytica, 1323.007. Flocculation and redispersion of cellulosic ber suspensions: aeffects of hydrodynamic shear and polyelectrolytes. BioResources 231.nte, E., Moral, A., Blanco, A., Izquierdo, L., Negro, C., 2010. Effect ofn the occulation of suspensions of bre-reinforced cement. Cem.. 40, 15241530.te, M.C., Blanco, A., Negro, C., Tijero, J., 2012. Characterisation of agri-sidues used as a source of bres for bre-cement production. Ind.. 36, 1421.al, L.T., Mohanty, A.K., Arora, S., 2004. Are natural bres compositesntally superior to glass bres reinforced composites? Compos. Part A

76.chell, C.J., 1992. Characterization of ber occulation regimes by afactor. J. Pulp Paper Sci. 18 (1), 3238.., Wang, X., 2004. Compressive and exural properties of hemp ber

concrete. Fibers Polym. 5 (3), 187197.


Li, Z., Wang, X., Wang, L., 2006. Properties of hemp reinforced concrete composites.Compos. Part A 37, 497505.

Moral, A., Monte, M.C., Cabeza, E., Blanco, A., 2010. Morphological characterizationof pulps to control paper properties. Cell. Chem. Technol. 44 (10), 473480.

Mwaikambo, L.Y., Ansell, M.P., 2003. Hemp bres reinforced cashew nut shell liquidcomposites. Compos. Sci. Technol. 63, 12971305.

Mutj, P., Lpez, A., Vallejos, M.E., Lpez, J.P., Vilaseca, F., 2007. Full exploitationof Cannabis sativa as reinforcement/ller of thermoplastic composite materials.Compos. Part A 38, 369377.

Negro, C., Snchez, L.M., Fuente, E., Blanco, A., 2005. Effects of occulants and sizingagents on bending strength of ber cement composites. Cem. Concr. Res. 35,21042109.

Negro, C., Blanco, A., San Po, I., Tijero, J., 2006. Methodology for occulant selectionin ber cement manufacture. Cem. Concr. Res. 28, 9096.

Negro, C., Blanco, A., San Po, I., Tijero, J., 2007. In-line occulation monitoring ina Hatschek machine for ber-cement manufacture. Compos. Part A: Appl. Sci.Manuf. 38, 2633.

Norma Chilena, 2006. Fibrocemento Planchas Parte 1: Planchas planas Requi-sitos. NCh186/1-2003.

Ouajai, S., Shanks, R.A., 2005. Composition, structure and thermal degradation ofhemp cellulose after chemical treatments. Polym. Degrad. Stab. 89, 327335.

Pico, D., Bartl, A., 2011. Image analysis for simultaneous determination of sphericaland brous particles. Chem. Eng. Trans. 24, 619624.

Placet, V., 2009. Characterization of the thermo-mechanical behaviour of hempbres intended for the manufacturing of high performance composites. Compos.Part A 40, 11111118.

Rabi, J.A., Santos, S.F., Tonoli, G.H.D., Savastano, H.J., 2009. Agricultural wastes asbuilding materials: properties, performance and applications. In: Cornejo, D.N.,Haro, J.L. (Eds.), Building Materials: Properties, Performance and Applications.Nova Publishers, Inc., New York, pp. 299342.

Savastano, J.H., Warden, P.G., Coutts, R.S.P., 2000. Brazilian waste Fibres asreinforcement for cement-based composites. Cem. Concr. Compos. 22,379384.

Sedan, D., Pagnoux, C., Smith, A., Chotard, T., 2008. Mechanical properties of hempbre reinforced cement: Inuence of the bre/matrix interaction. J. Eur. Ceram.Soc. 28, 183192.

Swamy, R.N., 1990. Vegetable bre reinforced cement composites a false dreamor a potential reality? In: Sobral, H.S. (Ed.), Proceedings, 2nd InternationalSymposium on Vegetable Plants and their Fibres as Building Materials. RilemProceedings 7. Chapman and Hall, London, pp. 38.

Tonoli, G.H.D., Fuente, E., Monte, C., Savastano Jr., H., Rocco Lahr, F.A., Blanco, A.,2009. Effect of bre morphology on occulation of brecement suspensions.Cem. Concr. Res. 39, 10171022.

Tonoli, G.H.D., Santos, S.F., Joaquim, A.P., Savastano, J.H., 2010. Effect of acceleratedcarbonation on cementitious roong tiles reinforced with lignocellulosic bre.Constr. Build. Mater. 24, 193201.

Trodec, M., Peyratout, C.S., Smith, A., Chotard, T., 2009. Inuence of various chemicaltreatments on the interactions between hemp bres and a lime matrix. J. Eur.Ceram. Soc. 29, 18611868.

Trodec, M., Rachini, A., Peyratout, C., Rossignol, S., Max, E., Kaftan, O., Fery, A., Smith,A., 2011. Inuence of chemical treatments on adhesion properties of hemp bres.J. Colloid Interface Sci. 356, 303310.

Young, R.A., 1997. Utilization of natural bers: characterization, modica-tion and applications. In: Leao, A.L., Carvalho, F.X., Frollini, E. (Eds.),LignocellulosicPlastics Composites. USP, Brazil.

Wielage, B., Lampke, T., Utschick, H., Soergel, F., 2003. Processing of natural brereinforced polymers and the resulting dynamic-mechanical properties. J. Mater.Process. Technol. 139, 140146.

Documents

Use of Cellulose Fiber From Hemp Core