INSTITUTE
OF FUNDAIVENTAL
TECHNOLOGICAL
RESEARCH
P O L I S H
A C A D E M Y
O F S C I E N C E S
MATRIX
GOMPOSITES
10
E d i t e d
b y
A.M. BRANDT'
J. OLEK
M. A. GLINICKI
ano
C. K. Y. LEUNG
Previous Proceedings of the International Symposia
on Brittle Matrix Composites in Poland
Brittle matrix composites -1
JABI-ONNA,12-15 September 1985, 590 pages.
Published by E&FN SPON London. ISBN l-851 66-041-0
Brittle matrix composites - 2
Cedzrna, 20-22 September 1988,674 pages.
Published by E&FN SPON London. ISBN l-851 66 360-6
Brittle matrix composites - 3
Warsaw, 17-19 September 1991,606 pages.
Published by E&FN SPON London. ISBN 1-851 66-687-7 Brittle matrix composites - 4
Warcaw, 13-15 September 1994, 702 pages.
Published by Woodhead Publ. Ltd, Cambridge, UK and IFTR in Warsaw.
tsBN l-85573-1 83-X / 83-9021146-0-1
Brittle matrix composites - 5
W a r s a w , l 3 - 1 5 O c t o b e r l q q 7 . 6 l 8 p a g e s .
Published by Woodhead Publ. Ltd, Cambridge, UK and IFTR in Warsaw
ISBN 1-85573-358-7 / 83-90662',1 -3-6 Brittle matrix composites - 6
Warsaw, 9-l I October 2000, 579 pages.
Published by Woodhead Publ. Ltd, Cambridge, UK and IFTR in warsaw
rsBN 1-85573-551-2 I 83-91038'7 -4-2 Brittle matrix composites - 7
Warsaw Il- l5 Oclober 2001. 564 pages.
Published by Woodhead Publ. Ltd, Cambridge, UK and IFTR in Warsaw
tsBN l-85573-769-8 / 83-91'7926-6-8
Brittle matrix composites - 8
warsaw. 23-25 Oclober 2006.603 pages.
Published by Woodhead Publ. Ltd, Cambridge, UK and IFTR in Warsaw
rsBN 1-84569-03 1 - l / 83-89687-09-7
Brittle matrix composites - 9
Warsaw, 25-28 October 2009, 451 pages.
Published by Woodhead Publ. Ltd, Cambridge, UK and IFTR in Warsaw
rsBN 978-l-84569-7'7 5-4 I 9',1 8-83-8968',7 -48-',7
ISBN 978-0-85709-988-4 (print)
ISBN 978-0-85709-989-1 (ontide)
Sale cement compositcs SRCC the rope effect in HPFRC concrete D o m i n i L L O C O N . P o 1 a n d . . . .
In0uencc of mixturc composition on shrinkage cracking oflightweighl
scl f-consol idat ing concrctc
Maria KASZYNSKA ard Adam ZIELINSKI, Poland ... KEYNOTE PAPER
Shape simulation ofgranular particles in concrete and applications in DEM l\tan HE, Chind, Piet STROEVEN, The Netherlands,
E r i c P I R A R D a n d L u c C O U R A R D , B e l s i u m . . . 2 7 5 2 5 1
Durable concretc structures with cracks which hcal thcmselves
Kim VAN TITTELBOOM, Manuel DIERICK and Nclc DF. BEl,lE. Belgian ... 285
Ncw cco-mechanical index for concrete structures
Alessandro P FANTILLI, llemardino CHtAtA
and Samanta CORR ADO, ltal), .. . . ... 295
lnfluence ofaggregate typc on thc durability ofconcrete made ofblendcd cements
with calcerous fly ash
Mariusz DABROWSKI and Michal A. GLlNlCKl, Poiazrl
Porosity ofgrccn concrcte based on a gap-gradcd blcnd
Nghi L.B. LE and Piet STROEVEN, The Netherlands
Characteristics ofsccondary flcxurcs in uniaxial tension test ofconcrete
Hiroshi AKITA, Hideo KOIDB and Hirozo MIHASHI,.|apan 13ehavior ofhybrid fiber ECC panels subjected to low and high vclocity lmpact a rcvlew
M. MAALEJ, S.T. QUEK, J. ZHANG and VW.J. LIN, U,4t
. l 1 5
325
3 3 5
Expcrimcntai study ofshear failure mechanism in concrcte beams
M a r t a S L O W l K , P o l a n d . . . . . . 3 4 5
Modelling offracture process zone and rclatcd cncrgy dissipation during
quasi briltlc fracture
V a c I a r V L S L L V , P e I r I - R A N I i K , T o l n i s P A I L
and Zbyndk KERSNER, Czech ltepublic ... 355
lnfluence ofcalcerous fly ash on concrete rcsistancc to migration ofchlorides
Prcc. Int. Symp. ,,Brittle Malri\ Composil..\ 10" A.M.Bran h, J.Olek, M.A.GIinicki, C.K.YLetng, eds.
Warsaw Otlober 15 17.2012
IFTR and Woodhead Puhl., Warsaw 2012
KEYNOTE PAPER
SHAPE SIMULATION OF CRANULAR PARTICLES IN CONCRETE
AND APPLTCATIONS IN DEM
Huan HEI, Piet STROEVEN,, Eric PIRARD3, Luc COURARD3
lCollege of Civil Engineering and Architecture, Beijing University ofTcchnology. Pingleyuan 100, 100124, Bcijing, China
'zFaculty ofCivil Engineering and Geosciences, Dclft University ofTcchnology
Stcvinweg l, 2628 CN Dchl, the Netherlands
lGcMMe, Minerals Engineering-Materials-Environment, Ulrivcrsity of Liegc.
Chcmin des Chevreuils 1, 4000 Lidge, Bclgium
e-mail: 2 p.strocvcn@tudelft.nl
ABSTRACT
Aggregate occupics at least three-quarters of the volumo of concrete. so its impacl on concreie s propcrhcs ls large. The sicvc curve iraditionally dcfines the aggregalc size range. Another csscntial properlv is grain shape. Bolh, size and shapc influence workabilily and ihe mechanical and durabilily properlios ofconcrete On thc olher hand. the shape of cemenl parlicles plays also an impodant role in ihe hydration proccss due 10 sur{ilce dissolution in ihe hardcning process. Addilionally. grain dispersion. shape and size govem the pore pefsolation Drocess that is olcrncial inporlance for concreie durabilily
Discrctc element modeling (DEM) is commonly employed ibr sinulation of concrctc st|ucture To bc
abl€ doing so, lhe assessed grain shape should bc implemented. Thc approaches for aglregale and ccmc.L
structure sinuiation by a concurrcnt algorilhm-based DEM sysiem are discusscd in this papcr. Both nggrcgatc and ccmcnl were expcrimcntally analyzed by X ray tonography mcthod recenlly Thc resLrlts pfo\rdc a r.al cxperimenial dalabase. e.g. surlace area versus volumc distribution, ibr sinulalion of padiclcs In concf.lc tcchnology. Optimun solutions are obtaincd by different simplificd shapes proposcd lbr aggregat€ rnd cemenl respectivcly. ln rhis way, rcliable concepls Ior aggregate srructure and licsh cement paslc can be sinrulatcd b) a D E M s y s t e m .
Ke)svords
Aggregatc, concrete, DEM, shape, packing.
INTRODUCTION
A particle is dctined as the smallcst discrete unit of a powder mass that can not be easily subdivided l1j. Thc shapes ofreal particlcs are irregular, either from coarsc aggregatc or from
mineral admixture in concrcte. Aggegaic occupies at least three-quarters of the volumc of
concrete, so its impact on concrete's properties is largc. The sieve curvc traditionally dcfines
thc aggregate size range. Anothcr csscntial properly is grain shapc. Both, sizc and shapc
influencc workability and the mechanical and durability properties of concrete Thc dcfinition
of the actual grain shape as wcll as the represcntation in a discrctc clement modeling (DEM)
systcm is complicated, howevcr. Furthcr, the real shape of aggrcgatc or binder particles can
vary widcly. The spherc is therefore gcncrally adopted in conventional simulation systcms'
cce-]rns Jo ezrs rxnlllrxeul eql pu€ aTts o^ets a./ ,sreleurDled o^\l oql ^luo ocuts
lJol pJqsrua -Jo (oltelnurs cql ser"llldurs f pccri srq.1 .l0l] suotte8rtscnur pleu s.onrJ iurmo11o.1 ,)l.o.r
peqsn-r:r luese:der o1 pcdolcnep ue serr3rns pJltc.p,] g-t qt11\ sJdeqs Jo dnojF .rlrJccls V 'cpqred aq1;o uolleouc8 Jql ioJ [ercnlc sr qslu e Sursn 'doJcJcrlJ elcrged ,{je,le:o3 ,(resscolu ere sapou uorlelnJl€J q8noua .Ir'{gc ur uo[eln.l]:rc Jo ,{cuctcglc ro1 surel8 {Jol peqsuJ Jo uorlelnurs roJ pesn se,tr uorpcq,{1od eqt 'fuBJtuoc eqt uO sclsnred 1e,rer3:c,rtr eqt tucs.ld.t ot pet.eles se,r prosdrllc cqJ I 3ll ur u,{oqs se',{lc,trlJodsel'sclcrped pedeqs-.re1nFue pul? prpunoJ iutlu:rso:dcr,)Jol pcqsmc pup ul8rlo lel^nu Jo le^e.t8 :sdnolS o,{t olut pspr^tp ccl uEJ etclJuo. ur ete8er88e ,,{lluteueC
sNMl) sJ,vcsdccv do Notr\r'.lnt\its sdvHs
'lIIats,{s
tAtrEC pacue^pe ue Butsn ,{q pJllnpuoc lcilunJ cq uec elsed lueuo. rlsc{ -+o clnlJru}solJrLu
pu€ qeBerSSE lo unlcnrls pclced ,{lcsuep oql Jo uorletcuc8 .rlJ uotlulnurs JlJruEd tu3tuJ.
pue e1e8aj33e o1 soqceordde pe:_re;ud eql se pcsodord ele sor8JlrJls uoqelnurs tu oJJrp
'srs,{loup
srql uo peseg lg'/l secue.rcJ.r se pe^.res stlnsar Jld pun Il eqr Kqe:cq,r .suru5 lucLlroc se IIc,r se e1e3er,33e:og suortnlos oduqs reldrurs .uros qlt,\\ pcl.npuoJ oslu se,r,{pn}s srs,{1eue edeqs V urJls^^s tr1gg paseq-uqlroSle ltlcllniuoJ 1ucrs,{qd r ur p.telodrotllr eq
11r,n qceordde uoqeptuts srqf-patucs.rd sl fSolouq.ol etelcuoc ur perldde sur€J8 luculec puB
sle3e:83e pedeqs-,fiurtrqle Jo uouetu.selder cq1 ro1 ser8elells uorlelnurs .:cded srql ul 'creSer8Be
to uortulnturs :o1 poldopr aq us. spoqtcu uorlelnurs luc:c.grq ,{e,r leorlcerd uoru r
sr scloqred lueragrp tucserdet ot scdeqs lere,les :o cuo iursooq. ,.loJcFqJ .16l czrs puE edcqs lenlrE slr q .r qcDo 'sJl.tt.rEd Jo leqrunu eiJ€l r? lJpout ot leoqcud 1ou st tl ,spjo^\ leqlo ul Surunsuoc-eur1 (:clndluoJ) pue J^lsucdxe ool sr I IAC ur scdeqs clcqred lerj Futsn .ro^.,{\oH
't\gg ur uollelnurrs :og s:oleruered e1qen1e,r splc leql lueulec pu€ at€Aol88e Jo s.srqqep
lelo.rurredxe luu opr,rord sllnsor iqJ '[S'/] (lJ) ,(qdedoruo] fer-X ro (JJd) ,{qdEl8oruot
-olcrur ^q p.zfleue e:e,tr c1e8er83e pu€ luourec ec$elc1.l euos .,{11uc:roX .qlnser uorlclnLrtrs peserq ,{lqrssod Jo lsoc cql tE 'Je,^c,'r\oq :uor}elnulo-+ uultuoSle ur uottEcr}tldrurs tuerellut
Jo Jsne.oq'sL!rets,{s qrns ut l€cucqds porunsse ,{11ejeuei sr ad€qs olctued .lJpou peseq
-egeurl lelrBrp eqt ulo:g lredy ssofund srql roj pedo1e,,'cp
ueaq e,req [9,g.7] slcpour rxnnul]uoJ se lle,'tr se f!1 leporu pcseq-eiEurr I€U8lp V etarluoJ 3o sergcdord pue lueurdole,rep
olnlculsoJcrru Suqe8qsc,lur go (lqrqrssod eql slcpu.: qleordde 1,196 E ul uoqelp,{q tuolue:)
'[7] ,{]r|qernp ctJj.uoc toJ clueilodur let.nlc.+o sr leqt ssJtold uoltelorjcd ojod oql ua^o8
ezrs pue.deqs 'uorsledsrp utr?ri ',{1€uollrppv ssecord Suruep:eq cq} ur uorlnLossLp arq+rns ol
rnp sseco:d uo4upfq eql ut Jlol tueuodrur ue osle s,{eid sclrrged lueuec Jo Jdeqs eqJ
Shape sinulation oJ grunulat pa icles in oncrete and appttuations in DEM
2 7 7
,&
I Gruver1ffi
Crushed rock..,,@w
'"'*&&W
E l l i p s o i d s AggregateFig. 1. Simulation stralcgy ofarbitrary shapcd aggregate; (top) differently slrapecl cllipsoids represent river gravel; (bottom) diftbrcntly shaped polyhedrons represcnt crushed rock
ery,e.@&,
lelrchedron Penlaicdfotl He{ahcdrol
Ww,e,s"
Heplahedron
P o l l h c d r a
0ctahedron
Fig. 2. Nine regular polyhedra with facet number 4-g
clement, arc required. Fig. 2 illustrales the meshcd particles with differcnt number of facets.
The quantity proportions of each polyhedron in the whole group of crushed rock proposed by
Guo [0].
_ Recently, in a study ofErdogan er d/. t8l, the shapcs offbur different Lypes or aggregarc!
GR, LS, IN and AZ were investigaled by the CT method and reconstructed by the sphcrical
harmonic technique [8]. The most inter€sting information fiom their shape analysii is the surface rrca {5) vs. volume (t, rnformation ofeach particle of different types of aggregatc. A function S=alr' was used to regress the corrclation of .S and Iz in eactr tyie of aggiegite ;S1.
The cocfficients a and b were obtained for each t)?e of aggregate as shown in faUie f [tj.
The variances ofthe regrcssions were all higher than 0.98.
As an ideal value of D in standard shapes is 2/3, thc regression function S=a,l/?/J was
employed to fit differenl lypcs of aggregate. For effective expcrimental conditions as to
volume (<4000 mm') and size range, a, is determined for cach tlpe of aggregatc with a variation lower than 57o, as shown in Tablc L The impoflant shape index, sphcricity, can be defincd as the surface area ratio of thc equivalent sphcrc (with -qual volume) and the real
particle. An approximate valuc for sphericily in each typc of aggrcgate can as a lonscquence
bc calculated bv
. 4 o t 3 v t "
S p h c r i . t n - ' i
,{q pJurelqo oq ueo S' pue y' p.^r.t.p .q uei sdrqsuorlelJl I S tucrcJJtp qlr\\ sprosdrllc tuereJJrp 'uorterIe^ roloLuErBd ,{fl prosdrlls
ue Jo edeqs .q] f,unu-rclep (srxe lsauolls , pue ezts untpsur
q JsaSuol slueseJder,) sexD ledrouud ccJql pequJsJp .q uEc Suolqo ol Jtulqo rro4 8ul8u€l
'sodeqs jo ,{teue,\ E sreteurel€d
t ,{luo Lllt^\ esnllJeq Sunseretllt ,{llDIn.ru€d .le sptosdtlll
'seruedold godsuetl elt seJucnUur
leqt uotuodotd ZJ_l Jql ro.+ puE ol.lcuoJ .Jo scqJcdoJd lerrueqceu oql roJ lueuodrut st e.le..p.]rns se I/lgC qlr,{\ pelueserdel ,{llJdold cq u€r etaJruo. ur crntJn4s lJol paqsu. Jrlt 'Lrolt€lnurs p.grldurrs slql ,{g .s.d,{t S.I pue dD.Jo stlnsej l€lucruuodxa eqt qlr,,r\ ,{ldruo. h ecl uEJ elpJq,{lod elrsodluoc Jo uoqelu.Iolrn
7-g 'suorgodord seseql ,{B f 3rg ur se 11c,n se I elqeJ ur u,{oqs posodotd sr ,{lo8cle.
qJec Jo uoluodord rJlloq e 'eroJeleqJ -eiuer clrrlred e3re1 eql ur uorttr^ep euros sr Jl.ql ttlg '11e,r
elrnb S-I pue UD se1e3er35e;o €lep rqt sll] lesodo:d s.ong.Surleeds fllercucg .g .3rg
ur u^\oqs'l8l S.l pue dCJo sllnscr lelucunjodxs qtt.n poleduoc eq uec [01] ono,{q p.sodord
poqlcur apeLg,{1od olrsoduoc Jrll 'lycN € elqD.l. ut pclsrl se ',{ro8o1ec cdeqs qoec 30 uotuodojd e lJor pJr{snj. ro-+ uorle8ltso^ut lelu.ruu.dxe srq Luo{ pcsodoJd osle [01] onD r:rpcq,{1od
rulnB .uru cscqlJo secrpur edeqsjo eiuer eqt urqlt,4\ cje S-l pue dC saldups ltoi peqsu.
jo sccrlur .dpqs eqt leql setejlsuourcp I 'ecueloJaj e se ereqds eqt qlr,^{ Z clqel ur palstl
r:e c:pcq,{1od reln8e: 1uare;3rp;o secrpur adeqg -[g] l, l, truSoplg Jo sqnso: lelueuuadxc
qll^\ pcz^leue uoql erE lJor paqsu. io uortelnuts roJ [01] onD ,{q sedeqs pcsodord aq1
uorlelnuns Jelleq eqt ro; uorpodord pesr,ror e pue [01]
ong fq p.sodord uorpodo:d cdeqs qcec pue upeq,{1od :e1n3$ oq} Jo serrput edeqg 7 c1qe1
.,-,,/,r=S uoucunJ {q pa1elno1ece: sr {lcueqds pue slucrrq-lcoo lrlt8aSSejo sed,{1 :no1 .ro-1 [8]-i, ,r, u€3oprE ,{q punoj sluercggaoc uortelrrro.l,s I elqpJ_ 'elnleu pepunol oloru Jlcql .+o lln$l e se ,{lrcucqds rc8rel f1e,rrle1er e J^eq ZV pue p1 sed,{1 1c,tuB :c,ru lernleu .qJ sJqr.u.qds rellrttrs eltnb e,^.eq ot 91 pue 19 sad,{1 Irol p.rlsnl. oll} s.{\orls U I elqeJ, ur pc1sr1 ore sorlroueqdsJo qlnscj cqJ ,/.1 pue S Jo drqsuorlelcl ur lucrcgJeoc eqt sr , D q.rq,\\ ur
clu!1looJ onl puE C{vutd Jtra Nit,\:lod-Ls rcrd'EUrq) rtH unnH
8 L Z
t 8 t
0 0 t
cqds % L I%0.s b z 9
8 / 0
8
II uorpeqelco ,t,Ll % 0 9z L s
s 8
0
8 I uorpaqElro% s
s I
,/00.01 9 0 90 8 0
L ll uornrqttdeH % 9 9 |ot,o.0
t
r 8',9
t 8 0
L l uorpJqetdeHvool
% 9 ' L I 6 t 9 9 L 0 9 Il uorpeqexel l ,1,01 % 9 L l00.9
1 8 0 9lactnSt I uorlraqP.xrH
oA9 % 9 7 , 1 I L 9z L 0
I ll uorpaqelurd % 9 % 9 Z I 9 6 90 t . 0
s
I uolporlElued % 9 %0.01 I Z ' L 1 9 0v
uorp0rlErle-L uotuoootoP0sr^ed onD Iq uoruodord ,ftnrnqdg requlnu lc.ec sodrq5
I 8 0 9 6 S 8 6 0 1 9 0 I 6
0 8 0
9 0 96 6 0
2 9 0
9 8 NI6 1 0
al ()
166 0
n 9 0
s L
S I6 1 0 9 6 6 0
t9'0
I 8UD
(lrrL.reqdg /lr tuJrJuJJo.)
Shdpc sinulatio ol !:rdnlllar pqrticles in Lon.rete d d dpplit.ltion.t in DEM 279
0 1000 2000 3000 4000
v o t u m e ( m m 3 )
Fig. 3. S-/ curves ofcrushed rock types (GR and LS), ofcornpositc polyhedra
and of thc sphere
cquations from ! lj: 4
3
(2)
s :26.,0f
in which, E=(S'?66s'?4r a2sin2 d;sin2 1/ is a cocfficient of thc flrst fundamental form and 1l is a polar angle. Flatness and elongation are defincd as r/b and 6/a, respectivcly. So. the distribution contours of sphcricity ar.d S/Ir" (or r.r' ) with different clongation and flalncss values can bc constructed, as depicted in Fig. 4.
Figure 4 (a) rcvcals that if elongation and llatncss are both larger than 0.5, sphcricity of an ellipsoid will be close to 1. ,S//"'-contours with elongation and flatncss are plotted in Fig. 4(b). With experimental values of' IN and AZ typcs in Table l, thc hatched rcgions in Fig. ,1 can bc sclected as thc optimum solution. The pref'erred elongation and flatness of an cllipsoid
should cxceed 0.3 and 0.4, respectivcly. This conclusion is comparablc with the experimcntal
rcsults on gravcl in Fig.4. Thc selected ellipsoids are thus ncither very elongated nor flat.
SHAPE SIMULATION OF CEMENT GRAINS
Computer X-ray micro-tomography offers a potenlial solution for shapc assessment of
ccments, as shown by Garboczi and Bullard [7] and by Bullard and Garboczi ll2.l. Thc microstructures ofhydrated cements based on actual grain shape and on spherical shapc wcre lbund significantly diffcrent in this study. The rcsults provide an cxperimental dalabasc ofthis
cement that yiclds some valuable paramctcrs for DEM simulation ofccmcnt hydration.
Using grains in numerical simulation similar to the real ones would be too expcnsivc. So, it is crucial finding simpler shapes that are sufficicntly representativc. Similar as in thc casc o1'the aggrcgate simulation. a shape analysis study was thcrefore conducted with some
simpler shapes. Based on this analysis, a simulation strategy is proposed for cement. Morc
realistic but slill cost-cffective panicle shapcs should thereforc be implemented in DEM svstcms. n- r 5oo E
g
H 1000"n lao
(3)
srxelsc8uol qlr,{\ solclued 0001.+o sJsEa l] slue\ol aJnBlC stqJ e^tnJ eaualcjJl JeluarutJcdxe ue qlrM erpJqeloo pelereuoS ^Iuopuel Jo suottnqulslp / s eql s,roqs (e)g :Jn3rc
-Jdpqs atrtuedJo,{ttsjc^rp Jrll.lo-+
SuttunocJe pclldde cq uea rollllue^ (uopuej palrlurl etuos .lrrJnraa caucta.lcr eql Jo leql ol Jsolc sr uorpJqet5o ueJo e^lnJ /-S cqt Jcurs [g1I ur se.rsod:nd slqt roj pe,{oldruc rq uec sexE eerql 'uoneue^ loleLuejEd ,{q scdoqs reln5er:r olur uotleuloJsurxl lersuc ue 8ul,{{olle Jo esneacq ,{pn1s uorlelnuns eqt roJ petcolcs sr uolpcqetJo Jq-L suorlnlos Surstuold ts^ou .qt J.Jlo / uojpaqelro .qt pue / uorpeq€tdcq Jql tellt pepnlcuoc sr lt !Z elqeJ pu€ E Srjl urou
'eg
0 se petelnlle. Jq uc.,{lr.uoqds'cseJ slqt ul enle^ srqt.leurxoldde croleteql plnoqs scdDqs pJsodotd yoE ueql la,'!\ol uorleue^ Jo tuotcq]Joc E Lilt,4\ .srs,{leue uolssaliJ.l {q punol se,t 8 9-,, tuercgJeoc '[tJ.rut 6n,r'Ut t,OO0,0l lo ounto^ letueuucdx. .^rt.qlc uc urqtrlA
r,-r,1,r:S,,{q p.ss5rdxc osle eq,{l1cr.ruli uer drqsuortnlt.t T 5 J[tnlo^ le^o.Jc1lns eql'9 31J ul pcllold rje sccrput edeqs esJql.+o son[€^'[/] ctucj.I.r leluotuueclxe aqt ruo.q pcurptqo uorleuuo;ul lueuodur lsoltr Jrll su,n drqsuo4ulc.r (4 5) olunlo^-ol-€Jle Jtp.!ns iq1 sV 'srs{1eue edeqs srqt roJ pclooles
oslt ele Z 8rC ur erpeq{1od eurg .uorlereprslroJ stql uo p.seq sureri lueruoJ Jo uorltlnlllrs loJ alqeuos€al arollr Luoes e:pcq,{1o.1 .sreloruorcttu Jo Jlecs r uo sad€qs relnBuE ol pepuel ole sureJF luaLuoa IDurJ Jo cdrqs .JtuoH sur€13 lucurJJ IIetUs otLll tunsd,{3 3:o tunolue lleuls e qlt,'rr lcl1ul rql ur punori eJe slalutll lueurer ^\DU .stclaullluJa Jo ole.s e uo selJlued pJpunor,{lqinol pu€ .FJel 'prpq .le s:a{urlJ tu.urec ,lre1 suorteutcleJ -+o sscJord e ieue 'Jle ',{elc 'cuolsJrurl sE qins sleueleLu ,rer ,{q epeut sr lua[raJ
Erpcqflod .telnFcJ cruos pue lueuroo oluJJcJeJ.lo seAlnc ctunlo,l snsrc,r eere e*3:ng .E .3tg
00009t 000001 0000s 0 t lorpaqeloq.t --' r.'.'. €qro . uorp.lexqr -u -uorp6qet-u6d tuorp€qeluod -oorp.Lte4.f .. .. ]aE €9t ft.lJ 000q o oooor H : 3 0 0 0 9 ! r ( q ) u o a e 6 ! o :
u)tvunoJ rn't pue cdvutd .!rit NtASouJs rrrd EurqJ tH ulnFJ
(e) d ! r L ,l , r l i
#e
ts
0 8 2
prosorllJ ue Jo ss.uleLl pue uoriEfluolr tu.rc-l-llp gl^{r-.,1/,S (q) puE,{}r.ucqds (e)Jo srnoluor uortnqutsrC i 3r:J
,':t;t
' i ; l E
Shape rinulation ol grunuhr p.trticlcs in conLrete and dppticatio s i,n DE M
in l0-50 pm sizc range. The ,t- I/ relationships comply well with experimcnts. Fig. 6(b) shows
an cxample ofa loosc packed sfucturc ofarbitrary octahedrons supposedly reprcsenting fresh
ccmcnt.
Morphological comparisons can additionally be madc between scctions of simulated
particle struclures and ofreal ones. As an example, Fig. 7(a) shows a random scction of the
simulated structure shown in Fig. 6(b) and a 2D scction ofa rcal cement structurc obtained by
X-ray micro-tomography. Cement particles in Fig. 7 (b), displayed in light grey, obviously
revcal the angular shape that is similarly revcalcd by the scction of thc polyhedral cement particlcs in Fig. 7(a). This adds to thc aforementioncd quantitative matching of S- I/ features of
real and simulated cement structures, as shown by Fig. 6(a). Therefore, it can be concluded
that thc prcposed simulation method providcs a significantly improvcd option lor modelling ofccmcnt particles as compared to the conventional sphere system. This will also have impacl on the hydration process, as wcll.
[ . " J , " "
2 8 1
5000 .(-4000 d 3 0 0 0 3 2ooo .D 1000 0 0 5000 10000 15000 20000 25000 votume (Fm3) (a) (b)Fig. 6(a) Computcr simulation of 1000 particle in l0-50 pm size range and cxpcrimental regression results [7] and ft) visualized structure ofcompacted grains
Fig. 7(a) section ofthe simulated structure (1000 octahcdron grains in 10-50 pm sizc range); (b) 2D section ofa rcal fresh cement structure (Cement- 133) obtained by X-ray
'
662-162,0102,(.n)Zt .se}soduoJ etercuoJ lueueJ .rruouola pue
fllllqerlrr ',{llltqrsee.} .}o slcedsv :,{.rteLutsolod etelcuoJ !.V.C €^elo) '.[,nH ..d ,ue^eo4s .Z
'6961 uole.uud
'uolt€repeg soulsnpul J.p,rod lelel^l -sorlsuclceleqJ Bur]ced clotu€d ,.r,{.U .uelulog .l
saJNsusdsu
crsed
rueuac
Llseg
ro emrrnrrs
er€F
aae
c4srreer
"ro*
n
".,n
"u"8
un"ii?:tfiiiit:j:#il:
uollelnurs .seq'] Jo uorlElueueldurl .selcqed 1eu qlr,r erpeqflod pesodord 1o sernlcn_r1s
pelced cqt;o ,(1ue1rurrs eql s,\\oqs uosueduoc lecr8oyoqd:ol,\I [;1] ,uo4erp,{q
}uour.J .roJ lelcn]c sl qcrq^\ 'sereqds se sdrqsuoqelu 7 5 cllsrl?u elou rslJo ,{eqJ .uot}nlos runur.ltio ue pcJeprsuoc cq upc elpcqelcoJo uorlerJel pelrurl p l€qt u,t oqs se,tr ll .sprosdrllc pue elpoq,{lod
pessedruocue ,{pnls slql euo lecuoqds pefoldure ,{luoruuro:r eq1 se sedeqs crlsrls:rr erolu
,{1ecrs,{qd;o lucrussess€ .ql Surnslnd pelcnpuo. sem ,{pn1s e .uorleFulrs uoqelpfq toC '[91] odeqs reln8er e uo uoq.unj eurs
e Sursoduuedns ,{q peDptsuoc ,{ldurrs eq osle uEc uoqeFrrrrs eJnlycl eceJJns lcjouuequnj
[3] suopeS4sa,tur ls]ueuucdxe ut alqelle^€ uo4euuoJut l-S eqt qll\\ setlduroc sprosd4le
lp,U ^lo^lelcr Jo pur:1 y ur8uo Fr^nU Jo lo^rl3 Jo uoqeFlurs ro; pesn eq uec sprosdrlg .[g]
s8urpug lelueuucdxe Luo{ pcl.elxe suo4eleJlocl-S cqt q}r,'rl,{ldruoc ro11cq urc poqlar.u srql 'clercuoo
ur pesn lco.r pcqsuc eql lueseJdeJ fl.rcdold uEr [61] sedeqs eldurs pesodold eutu
qlr,,r\ €Jpeq,{lod .lrsoduoJ posodold st surelS elegel58e podeqs-f.ru:1tqre 3o sur8tro lucjejgrp
uo pesEq ,{SetEl]s uo4Elnlurs y .[9.1] secuucaer se pesn ere,r (fJl ro) ro 13 Xq lueurc:r
puu e1e8ei58e lecr Jo uoqeuuoJur edeqs .qJ .luerucc Jo s€ ll.,r se ele8a$Be Jo uoqelnuls
lol sedeqs rcldruts euros qlr.r pet.npuor se^\,{pnls srsfleue edeqs e ,etoJol.ql
'LZll sllnsrr uouelnurs peserq flqrssodjo tsoc cql l€ sr
su-lcls^s uottElnturs l€uotluo,\uoc ur pounsse ^lleuuou adDqs.l.rued l€.ucqds e.ssrleque^eN
'c,t rlelueserder flluercgJns ere 1eq1 sedeqs :eldurs Surpug IercruJ sr lr ,oS .[6] eultl lrlndluor pue lueurdrnbe ol s€ c^tsuedxa oot eq plno,,lr seuo le eqt se :eltuls ert ]Bql uortelnurrs I€.rrelunu ur sure,r8 go lunoue e8rel e 3ursl-1 .g .iU ul poltold ele seldruExg .[9l.gl] ue:8ord qsaur C€ e qll,\{ selgord leluozuoq pue lecrua^ 3q} Aururqruoc ,{q palon:isuocer eq ,{lqFnor uec edeqs elcrlred reln8e:rr ue ,slsxleuc c8erur Ct ltuollpe4 r{1r.,lr ue^g [rt,/] HS pue (JJrl ro) 13 3o luerudole,rep qlr.^{ .{llErceds? ,etqrssod ,{llEcruq.ot sr edeqs rEln8oJ]r ue Jo uorlcn{suo.eX xelduoc ,{lqelrleur eJoJeleql sr edeqs 3:o uoq€lnturs Jo uortrur1cp
.qJ sedeqs:e1n8eut e,req 'sle.reuru lo tueurc .cteier88e Jeqtte ,etol.uoc ur pesn sclJruEd
',{lc,rqcedso: 'qseul
Ct esJ€oc (lg8u) pue qscu Ct eug (gal) qll,| ure:B pelcrulsuo.er eJo sn^er^ om.1 .g.Brg
sNolsn.rJNoJ oNV NOtSSnJStO
Shape sinulation d gtunulff pdrticles in e crete un.l applications in DLM
283
3. Bentz, D.P., Three-dimcnsional computer simulation of portland cement hydration and
microstructurc developmcnt. J. ofAmerican. Ceramic Society,80(l), 199j,3,21.
.1. van Breugel, K., Simulation of hydration and fomation ofstructure in hardening cemellt-based matc als. Delf't University Prcss, Delfl 1997.
5. Navi, P., Pignat, C., Thrcc-dimensional characterizatioll of thc pore struclurc of a simulaled comcnt paste. Ccment and Concrcte Research, 29(4), j qqq, 507-514.
6. Stroeven, M., Discretc Numerical Model for thc Stluctural Asscssment of Composite
Materials. PhD fhesis, Mcine|ra, Dclf't, 1999.
7. Garbocz| E.J., Bullard, J.W., Shape analysis ofa reftrence cemcnt. Cement and Concrete
Research. 34. 2004. 1933-1937.
ll. Erdogan S.T., Quiroga P.N., Fowlcr D.W., Saleh H.A., Livingston R.A., Garboczi E.J,
Kctcham P.M., Hagedom J.G., Satterficld S.G., Three-dimensional shape analysis of
coa$e aggregates: New techniques for and prcliminary rcsults on several diffcrent coarsc
aggrcgates and rcfbrcnce rocks. Cement and Concrele Rcsearch, 36(9), 2006, l6l9-162j.
9. Thomas, P.A. and Bray, J.D., Capluring non-spherical shape of granular media with disk
clusters. J. of Gcotechnical and Ceoenvironment Engincering, 125(2-3), 1999, 169- I 78.
l0.Guo, W., Somc material parametem on numerical statistical continuum mcchanics of
concrete, TU Delft Repor125-88-38, Dclft, 1988.
I l. Weisslcin, E.\N ., Ellipsoid. From MathWorld--A Wolfram Wcb Resourcc,
http://mathworld.wol{iarr.com/Ellipsoid.html
12. Bullard, J.W., Carboczi, E.J., A model investigation of the influence of pafliolc shape on
Portland cement hydration, Cemcnt and Concrcte Research. 36, 2006. 1007-1015.
13. Hc, H., Guo, 2., Stroeven, P. and Stroevcn, M., Discrcte element approach to packing of
arbitrary shaped particlcs in concretc, In2ynieria Matcrialowa, 29(4), 2008. 403-401.
14. Carboczi, E.J., Three-dimensional mathcmatical analysis of particlc shape using X-ray
tomography and spherical harmonics: Application to ag$cgatcs used in concretc. Cemenl
and Concrete Research. 32( I 0), 2002, 162 1 - 1 638.
15. Hc, H., Guo,2., Stroeven, P., Stroeven, M., Sluys, L.J., Characterization ofthe packing of
aggregate in concrete by a discrete element approach. Materials Characterization,60(10).
2009, 1082-1087.
16.Hc, ll., Computational Modclling of Pafiiclc Packing in Concrctc. phD lhcsis. Delli
