Experimental Analysis on Shrinkage and Swelling in Ordinary Concrete

Resea ch A icle Expe imen al Analysis on Sh inkage and Swelling in O dina y Conc e e Ba ba a Kucha czyko á, Pe Danjk, Dalibo Kocáb, and Pe Misák Facul y o Ci il Enginee ing, B no Uni e si y o Technology, Ve eˇ ´ ı331/95,60200B no,CzechRepublic Co espondence should be add essed o Ba ba a Kucha czyko ´ a; ba ba a.kucha czyko a@ u b .cz Recei ed 2 Ma ch 2017; Accep ed 23 Ap il 2017; Published 22 May 2017 Academic Edi o : Ka suyuki Kida Copy igh © 2017 Ba ba a Kucha czyko ´ a e al. This is an open access a icle dis ibu ed unde he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. The pape deals wi h he expe imen al de e mina ion o sh inkage de elopmen du ing conc e e ageing. Th ee conc e e mix u es we e made. They di e ed in he amoun o cemen in he esh mix u e, 300, 350, and 400 kg/m3. In o de o de e mine he in luence o plas icise on he p og ess o olume changes, ano he h ee conc e e mix u es we e p epa ed wi h plas icise in he amoun o 0.25% by cemen mass. Measu emen s we e pe o med wi h he goal o obse ing he in luence o cemen and plas icise con en on he o e all de elopmen o olume changes in he conc e e. Changes in leng h and mass losses o he conc e e du ing ageing we e measu ed simul aneously. The con inuous measu emen o conc e e mass losses caused by d ying o he specimen’s su ace p o ed use ul du ing he in e p e a ion o esul s ob ained om he conc e e sh inkage measu emen . Du ing he i s 24 hou s o ageing, all he conc e e mix u es exhibi ed swelling. I s magni ude and p og ess we e in luenced by cemen , wa e , and plas icise con en . Howe e , a loss o mass caused by wa e e apo a ion om he su ace o he specimens was also eco ded in his s age. The measu ed p og ess o sh inkage co esponded well o he p og ess o mass loss. 1. In oduc ion Conc e e sh inkage can be de ined as a olume change ha akes place du ing wo consecu i e s ages o conc e e ageing, he ea ly s age o se ing and he long- e m s age o ha dening [1]. These changes a e in bo h cases connec ed wi h mois u e lea ing he esh o ha dened conc e e. Gene ally speaking, h ee main ca ego ies o sh inkage can be ecognised – plas ic sh inkage, chemical sh inkage (which includes au ogenous sh inkage), and sh inkage due o d ying. Some sou ces men ion conc e e swelling [2–7]. F om he long- e m poin o iew also he ca bona ion sh inkage mus be aken in o accoun . The de e mina ion o he eal p og ess o olume changes o conc e e has ecei ed inc eased a en ion om ci il enginee s and conc e e p oduce s in ecen yea s. This is connec ed mainly wi h he de elopmen o new conc e es as a as composi ion and p oduc ion echnology is conce ned [3, 8–10]. Scien i ic sou ces desc ibe a numbe o app oaches o de e mining he alue o conc e e sh inkage [5, 11, 12]. Howe e , hese a e mos ly me hods o sepa a ely de e - mining he indi idual componen s o conc e e sh inkage in he ea ly age, such as me hods o de e mining plas ic o au ogenous sh inkage, desc ibed, o example, in [2, 3, 6, 7, 12, 13], and me hods o de e mining conc e e sh inkage due o d ying, which a e de ined mos ly in he na ional s anda ds o a ious coun ies. Thus, con empo a y expe imen al mea- su emen s a e no su icien o cap u ing he o al change in conc e e olume caused by chemical eac ions, plas ic se lemen , and d ying. These in luences ha e ei he la ge o smalle signi icance in e e y s age o conc e e ageing. While he o al sum o hese in luences can be e y signi ican in he ea ly s age o se ing and ha dening, he mos signi ican in luence om a long- e m pe spec i e appea s o be conc e e d ying [4]. Con empo a y app oaches o conc e e sh inkage mea- su emen a e based mainly on he de e mina ion o ela i e leng h change. In mos cases, measu emen begins a e specimens ha e been emo ed om hei moulds, which is ypicallynosoone han heageo 24hou s.Inimpo an o complica ed conc e e s uc u es, sh inkage is measu ed di ec ly on a conc e e elemen using a special ype o wi e s ain gauge designed o be embedded in he conc e e. Hindawi Ad ances in Ma e ials Science and Enginee ing Volume 2017, A icle ID 3027301, 11 pages h ps://doi.o g/10.1155/2017/3027301 2 Ad ances in Ma e ials Science and Enginee ing Suchgaugesa e ypically ied o he ein o cemen cageo he measu ed elemen by means o eba ex ensions [14– 18]. Howe e , his measu emen is e y expensi e and i is usually no possible o d aw gene alisa ions om he esul s. In p ac ical e ms, i is mo e ad an ageous o de e mine he p og ess o he olume changes in conc e e by es ing specimens in a labo a o y. The signi icance o each ype o sh inkage is s ongly dependen on he composi ion o he composi e, he cu ing me hod,and hesizeo he es specimen[19].Expe ience gained om measu emen s pe o med in ecen yea s indi- ca es he necessi y o assessing he magni ude o sh inkage in wo basic s ages, in he ea ly s age o se ing and ha dening ( oughly up o 48 hou s a e mixing cemen wi h wa e ) and a “la e ages” which can be coun ed om 48 h o conc e e ageing. A e cemen has been mixed wi h wa e , hey eac oge he andchemicalsh inkageoccu s.Along wi h he chemical sh inkage, au ogenous sh inkage also appea s. Au ogenous sh inkage occu s o as long as he e is wa e p esen in he po e s uc u e. Fo his eason, his phenomenon is o en called sel -desicca ion sh inkage [1, 4]. Ve y soon a e esh conc e e has been made, plas ic sh inkage also akes place. This is in a e y ea ly s age when he conc e e is s ill esh and a solid s uc u e which would “hold he body in place” has no ye o med. The magni ude o plas ic sh inkage depends on he e apo a ion o wa e om esh conc e e o i s abso p ion in o a d y conc e e body o soil unde nea h [4, 20]. I s magni ude is also in luenced by esh mix u e bleeding. A no less impo an , ye o en neglec ed, ype o olume change is de o ma ion caused by empe a u e change. Such olume changes can be obse ed a he ea ly s age o se ing as well as h oughou ha dening and long- e m ageing. In hei pape om 1997, Tazawa and Miyazawa discuss he in luence o empe a u e on au ogenous sh inkage; as empe a u e ises, he magni ude o au ogenous sh inkage a an ea ly age inc eases, while a la e ages his in luence appea s negligible. In he ex ensi e scien i ic wo k om 2001 [2], Hol discusses he connec ion be ween empe a u e inc ease in conc e e and i s he mal expansion du ing he ea ly s age o se ing and ha dening. Amin e al. ha e also deal wi h he in luence o empe a u e on sh inkage o conc e e du ing his pe iod, which hey discuss in de ail in a pape published in 2010 [21]. In ecen yea s he e has been g ea ly inc eased in e es in de e mining olume changes in conc e e a an ea ly age. Also, he e m “conc e e swelling” can be encoun e ed mo e and mo e equen ly in connec ion wi h he phenomenon obse able in he i s 24 hou s a e he mixing o cemen wi h wa e . The au ho s o his pape deal wi h his phe- nomenon mo e ex ensi ely in 2006 du ing he expe imen al e i ica ion o a me hod o measu ing olume changes in conc e e cas in sh inkage d ains p oduced by Schleibinge [12]. Ve y in e es ing esul s ha e been ob ained du ing measu emen s pe o med on conc e es wi h di e en po ous agg ega e con en s, o du ing he obse a ion o he in luence o he wa e sa u a ion o po ous agg ega e on olume changes in ligh weigh conc e e. The e a e se e al ac o s which can be a he o igin o his phenomenon. In mos cases conc e e swelling is connec ed wi h excess ee wa e in he Table 1: Iden i ica ion o es conc e es. Assumed cemen dosage [kg] No plas icise Plas icise 0.25% o cemen mass 300 0/1 1/1 350 0/2 1/2 400 0/3 1/3 conc e e mix u e which causes he conc e e o bleed. This excess wa e is d awn back in o he ha dened cemen pas e by weak capilla y suc ion a e plas ic se ling has inished. This eabso p ion is he cause o swelling [2, 22, 23]. Swelling can also be s eng hened by he o ma ion o some expansi e hyd a ion p oduc s, abso p ion o wa e in o he C-S-H gel [4, 7], o he gene a ion o hyd a ion hea [2]. Swelling also occu s in conc e es which a e en i ely subme ged in wa e [4]. I appea s ha in mos adi ional cemen composi es he mos subs an ial pa o he o al amoun o sh inkage is ac ually de o ma ion caused by d ying. This de o ma ion occu s h oughou se ing and ha dening and can be educed ia sui able composi ion design and especially by choosing a sui able cu ing me hod o he inal conc e e elemen . Sh inkage due o d ying occu s in all ypes o cemen composi es placed in an en i onmen wi h ela i e humidi y <95% [20]. 2. Expe imen al Pa 2.1. Tes ed Ma e ial. The expe imen s d aw on esul s pub- lished wi hin he p ojec GAˇ CR 13-18870S [24–27]. The pu pose o he measu emen s was o de e mine olume changes in o dina y conc e e and especially o eco d he p og ess o ea ly and la e -age sh inkage. The basic o mula o he conc e e consis ed o dense agg ega e wi h a maximum pa icle size o 16 mm, CEM I 42.5 R Po land cemen and wa e . Th ee conc e e mix u es we e made which di e ed in cemen con en . They di e ed in he amoun o cemen in he esh mix u e, 300, 350, and 400 kg/m3. In o de o de e mine he in luence o plas icise on he p og ess o olume changes, ano he h ee conc e e mix u es we e p epa ed wi h plas icise in he amoun o 0.25% by cemen mass (polyca boxyla e e he based echnology). The amoun o cemen in he eshmix u ewas hesamelikein he e e ence mix u es wi hou plas icise . In o al six di e en mix u es, h ee wi h plas icise and h ee wi hou , we e p epa ed o he expe imen (see Table 1). Wa e con en was adjus ed o achie e app oxima ely he same wo kabili y in all he conc e es. Measu emen s we e hen pe o med wi h he aim o examining he in luence o cemen and plas icise con en on he o e all p og ess o olume changes in he conc e e.Ala genumbe o specimenswe emade omeach ypeo conc e e,eachbeingapp ox.0.4m 3in olume. In o de o ensu e he homogenei y o he amoun o conc e e necessa y o making all he specimens, all conc e e mix u es we e p oduced in a conc e e mixing owe . Fo mo e de ailed in o ma ion on he composi ions o he conc e e mix u es, see Table 2. Also, Table 3 lis s he mix u es’ basic p ope ies in he esh s a e. These p ope ies a e di ec ly connec ed wi h Ad ances in Ma e ials Science and Enginee ing 3 Table 2: Composi ion o esh conc e es. Componen s pe 1 m3o esh conc e e Uni s Conc e e ID 0/1 0/2 0/3 1/1 1/2 1/3 CEM I 42.5 R (Mok ´ a) [kg] 300 350 400 300 350 400 Sand (B a ˇ cice) 0–4 [kg] 925 875 825 925 875 825 Agg ega e (Olb amo ice) 4–8 [kg] 185 185 185 185 185 185 Agg ega e (Olb amo ice) 8–16 [kg] 695 695 695 695 695 695 Wa e Mixing [kg] 190 190 190 165 165 165 In agg ega e [kg] 14 14 14 14 14 14 To al [kg] 204 204 204 179 179 179 Sika ViscoC e e 4035 [kg] 0 0 0 0.75 0.88 1.00 w/c a io (wmixing/cemen ) [—] 0.63 0.54 0.48 0.55 0.47 0.41 Table 3: Basic p ope ies o esh conc e es. Conc e e ID 0/1 0/2 0/3 1/1 1/2 1/3 Bulk densi y o esh conc e e [kg/m3] 2,320 2,320 2,290 2,280 2,300 2,300 Flow able [mm] 410 390 420 360 350 370 Ai con en [%] 2.7 2.5 2.5 2.6 2.8 3.2 conc e e beha iou especially du ing he ea ly s age o i s se ing. The p ope ies o he esh conc e e we e de e mined in acco dance wi h ˇ CSN EN 12 350 [28]. Two pa ame e s we e obse ed du ing he measu emen , namely, ela i e leng h de o ma ion (sh inkage/swelling) and mass loss (e apo a ion o wa e om he su ace o he conc e e). Th ee es specimens, sh inkage d ains illed wi h conc e e, we e manu ac u ed om each mix u e. 2.2. Tes Equipmen and P ocedu e. The measu emen o sh inkage was pe o med using a es ing de ice made by he company Schleibinge Ge ¨ a e Teube u. G eim GmbH [12]. Sh inkage d ains o 1000 mm in leng h and 60 × 100mm in c oss-sec ion we e used o eco ding he leng h changes measu edalong hecen alaxiso hespecimensusingan induc i i y senso leaning agains he mo able head o he d ain. These d ains a e p ima ily designed o sh inkage mea- su emen in he ea ly s age o cemen composi e se ing and ha dening. Special ma ke s we e embedded in o he uppe su ace o he conc e e placed in he sh inkage d ains in o de o acili a e subsequen long- e m measu emen o ela i e de o ma ion. In his way, wo gauging bases we e c ea ed o u he measu emen (see Figu e 1). This a angemen enables he con inuous cap u e o he o al ela i e leng h changes o he conc e e since he momen he conc e e is placed in o he d ain un il i s long- e m ageing a e he specimen has been emo ed om he sh inkage d ain. De ails abou he ma ke ypes, d awing and hei a angemen can be ound in [29]. Figu e 1 shows he sh inkage d ains illed wi h conc e e and eady o measu emen . The sh inkage d ains we e illed wi h conc e e and placed on o a special weighing able which allowed con inuous eco ding o mass losses caused by ee d ying o he spec- imen su aces (see Figu e 2). In his way, changes in he leng h and mass losses o he conc e e in he sh inkage d ains 1 3 2 Figu e 1: Sh inkage d ains eady o measu emen (1: embedded ma ke s; 2: induc i i y senso ; 3: mo able head). we e measu ed simul aneously (see Figu e 2). The design o he weighing able as well as he de ails o he measu emen p ocedu e a e p o ec ed by indus ial p ope y igh s held by B no Uni e si y o Technology egis e ed in he na ional pa en da abase unde applica ion numbe 2013-961 [30]. Fo de ails abou measu emen echniques see [31]. Due o he consis ency o he esh conc e e, measu e- men was s a ed app oxima ely one hou a e he conc e e was pou ed in o he d ains. Sh inkage was measu ed in he d ains placed on he weighing able in a labo a o y a a empe a u e o 21±2∘Cand ela i ehumidi yo 83±3%un il heconc e ewas3daysold.The opsu aceo heconc e e was no p o ec ed om d ying. The specimens we e hen ex ac ed om he d ains and s o ed in a labo a o y a a s able empe a u e o 21 ± 2∘Cand ela i ehumidi yo 55 ± 3%. Fu he measu emen s we e pe o med using a Hollan s ain gauge(seeFigu e3)whichwas ixedon o hesu aceo hespecimens.Theposi ioningo hegaugingpoin swas p ede ined by he ma ke s embedded a spacing o 200 mm. The specimens we e hen le o d y eely o he en i e ime o measu emen and we e weighed a egula in e als. 4 Ad ances in Ma e ials Science and Enginee ing Table 4: Basic cha ac e is ics o ha dened conc e es. 0/1 0/2 0/3 1/1 1/2 1/3 Bulk densi y [kg/m3]2310 (11.0) 2320 (8.62) 2330 (18.5) 2320 (8.08) 2330 (4.04) 2360 (1.73) Comp essi e s eng h [N/mm2]33.3 (0.19) 44.5 (0.49) 55.8 (2.23) 42.8 (0.84) 50.8 (0.22) 56.4 (2.21) Dynamic modulus o elas ici y 𝐸c L [N/mm2]29000 (503) 34600 (661) 35000 (281) 30600 (563) 31200 (670) 32900 (671) No e. All ma e ials’ cha ac e is ics a e ep esen ed by mean alues and (s anda d de ia ion) bo h in oduced in he same uni s. Values o dynamic moduluso elas ici y 𝐸c L we e de e mined using he esonance me hod acco ding o [32]. (a) 1 2 3 (b) Figu e 2: Weighing able o con inuous measu emen o mass loss (a); sh inkage d ains placed on he weighing able (1: sh inkage d ain wi h he induc i i y senso ; 2: weighing able; 3: embedded ma ke s). 1 Figu e 3: A Hollan s ain gauge (1: embedded ma ke s). 3. Resul s and Discussion The measu emen esul s a e p esen ed he e mainly in isual o m. To in oduce he es ed conc e es hei basic ma e ials’ cha ac e is ics in ha dened s a e a e summa ized in Table 4. Fo cla i y, measu emen a an ea ly age, ha is, conc e e in sh inkage d ains, is shown sepa a ely om he p og ess o o al sh inkage, ha is, measu emen in he sh inkage d ains as well as measu emen a e he specimens we e emo ed. Gene ally speaking, he obse ed in luences, ha is, cemen and plas icise con en , a e no e y signi ican in e ms o he alues o o al de o ma ion. Conce ning cemen dosage (Figu e 4(a), on he uppe pa ), he g ea es o al sh inkage de e mined a he age o 300 days was eco ded in conc e e 0/1.Howe e , he alueo sh inkageincompa isonwi h o he conc e es (0/2 and 0/3) is di e en by no mo e han 150 𝜇m/m. The p og ess and alues o mass losses eco ded o e he whole ime o measu emen we e he same o all he conc e es being examined. Ra he in e es ing esul s we e obse ed in he measu emen s eco ded in he ea ly s age o conc e e se ing and ha dening. Figu e 4(b) shows he p og ess o he sh inkage o conc e es wi h di e en cemen con en . The esul s indica e a di e en beha iou o each conc e e in he i s 24 hou s o ageing. The diag am shows ha sho ly a e commencing he measu emen , all he conc e es swell despi e wa e e apo a ion (as seen om he mass losses, Figu e 4(b), bo om pa ). I can also be obse ed ha he swelling ime mo es along wi h an inc ease in cemen con en . While he s age o swelling in he case o conc e e 0/1 (300 kg o cemen pe 1 m3o esh conc e e) ends app ox. 18 hou s a e he beginning o measu emen , in he case o conc e e 0/2 (350 kg/m3as abo e) he du a ion o swelling is 24 hou s and o conc e e 0/3 (400 kg/m3as abo e) i is 30 hou s. The magni ude o swelling is dependen on he amoun o cemen and wa e con ained in he conc e e. Gene ally speaking, he highe he cemen dosage was, he g ea e he alue o olume inc emen eco ded was. The abo e-desc ibed p og ess o conc e e olume changes can be explained in he ollowing way. Fi s ly, his phenomenon is s ongly dependen on he chemical composi ion, spe- ci ic su ace, and g ading o he cemen used. Secondly, he p og ess o swelling is also connec ed wi h mois u e dis ibu ion in he po e s uc u e o he conc e e. In suppo o he conclusions d awn by Hol [2], his phenomenon is Ad ances in Ma e ials Science and Enginee ing 5 0 50 100 150 200 250 300 0 0 100 0/1 0/2 0/3 Sh inkage (m/m) Mass loss (g/g·10−4) Time (days) Time (days) 0 50 100 150 200 250 300 −100 −200 −300 −400 −500 −200 −400 −600 −800 −1000 0/1 0/2 0/3 (a) 0 12243648 0 12 24 36 48 0 25 50 75 100 0 25 Sh inkage (m/m) Mass loss (g/g·10−4) Time (hou s) Time (hou s) −25 −50 −75 −100 −125 −150 −25 −50 0/1 0/2 0/3 0/1 0/2 0/3 (b) Figu e 4: P og ess o sh inkage/swelling and mass loss o conc e e 0/1, 0/2, and 0/3 ((a): measu emen o e he whole ime o se ing and ha dening; (b): ea ly age measu emen ). linked wi h he con en o mixing wa e in conc e e ha ises o he su ace o he composi e (a e agg ega e and cemen dis ibu ion in he esh conc e e dispe sion sys em has s abilised) and causes conc e e bleeding. A e plas ic se lemen has inished, his excess wa e on he specimen uppe su ace is d awn back in o he se ing and la e ha d- ening cemen pas e, e illing he po es c ea ed du ing cemen hyd a ion. This wa e eabso p ion causes he swelling o conc e e. Simila explana ion can be ound in Repo o RILEM Technical Commi ee [22], whe e his phenomenon is asc ibed o he ea ly au ogenous de o ma ion called as au ogenous expansion. Thi dly, he swelling can also be suppo ed by he o ma ion o some hyd a ion p oduc s ha ha e a g ea e mola olume han he o iginal eac an s. And las ly, i is necessa y o sugges ha he hea gene a ed du ing cemen hyd a ion has a subs an ial e ec on he olume inc ease o he conc e e as well because i causes i s he mal expansion. The phenomenon o swelling is clea ly isible om he measu emen s in he i s 24 hou s o ageing when he hyd a ion hea is gene a ed a he highes a e (app ox. 200 J/g pe 24 h) [31, 33]. The pa icula con ibu ion o all he abo e-men ioned phenomena is he objec o ongoing expe imen al measu emen . When he es specimens a e emo ed om he d ains and hei su ace is exposed o ai , he wa e con en o he conc e e begins o dec ease apidly due o ee desicca ion which esul s in sh inkage o a much g ea e magni ude han swelling. 6 Ad ances in Ma e ials Science and Enginee ing 0/1 1/1 0 −200 −400 −600 −800 −1000 Sh inkage (m/m) 0 50 100 150 200 250 300 Time (days) 0 100 Mass loss (g/g·10−4) Time (days) 0 50 100 150 200 250 300 −100 −200 −300 −400 −500 0/1 1/1 (a) 0 12243648 0 25 50 75 100 Sh inkage (m/m) Time (hou s) −25 −50 0 12 24 36 48 0 25 Mass loss (g/g·10−4) Time (hou s) −25 −50 −75 −100 −125 −150 0/1 1/1 0/1 1/1 (b) Figu e 5: P og ess o sh inkage/swelling and mass loss o conc e es 0/1, and 1/1 ((a): measu emen o e he whole ime o se ing and ha dening; (b): ea ly age measu emen ). Conce ning he compa ison o he o al sh inkage o conc e es wi h he addi ion o plas icise , i can again be s a ed ha no signi ican di e ences we e eco ded in he absolu e alues o o al de o ma ion de e mined o he indi- idual conc e es in he g oups o conc e e specimens being compa ed (see Figu es 5–7). Howe e , some di e ences we e obse ed a he ime when de o ma ion s abilised. Conc e es wi hou plas icise saw a ela i ely sha p inc ease in sh inkage be o e he age o 90 days, a e which hey began o s abilise. The addi ion o plas icise slowed he sh inkage down al eady a he age o app oxima ely 30 days. The highes di e ence in sh inkage alues was eco ded be ween conc e es 0/1 and 1/1. Ne e heless, Figu e 5(a) (on he uppe pa ) shows ha his di e encein o alsh inkageisnohighe han150𝜇m/m. The alues and p og ess o mass losses we e he same o bo h obse ed conc e es (see Figu e 5(a) bo om pa ). Simila esul s o sh inkage p og ess we e eco ded in he case o conc e es 0/2 and 1/2 and 0/3 and 1/3, espec i ely. These esul s co espond well wi h he measu ed p og ess o mass losses associa ed wi h cemen and wa e dosage (see Table 2). Figu es 5(b), 6(b), and 7(b) (on he bo om pa s) show ha , in he ea ly s age o conc e e se ing and ha dening, he mass o he conc e e specimens wi h plas icise dec eases mo e apidly (in compa ison wi h he conc e e wi hou plas icise ). I can be obse ed ha his end is changing app oxima ely a e 7 days o conc e e ageing (see Figu es 6(a) and 7(a) bo - om pa s). Ra he in e es ing esul s in e ms o he p og ess o conc e e olume changes we e again ob ained om he ea ly age measu emen (Figu es 5(b), 6(b), and 7(b)). All he pe o med measu emen s exhibi a simila end, conc e es Ad ances in Ma e ials Science and Enginee ing 7 0 −200 −400 −600 −800 −1000 Sh inkage (m/m) 0 50 100 150 200 250 300 Time (days) 0 100 Mass loss (g/g·10−4) Time (days) 0 50 100 150 200 250 300 −100 −200 −300 −400 −500 0/2 1/2 0/2 1/2 (a) 0 12243648 0 25 50 75 100 Sh inkage (m/m) Time (hou s) −25 −50 0 12 24 36 48 0 25 Mass loss (g/g·10−4) Time (hou s) −25 −50 −75 −100 −125 −150 0/2 1/2 0/2 1/2 (b) Figu e 6: P og ess o sh inkage/swelling and mass loss o conc e es 0/2, and 1/2 ((a): measu emen o e he whole ime o se ing and ha dening; (b): ea ly age measu emen ). wi h plas icise see a educ ion in he alue o swelling. Also, he cu es o ini ial olume changes o he conc e e wi h plas icise eco ded du ing i s 48 hou s o measu emen di e om he cu es o he conc e e wi hou plas icise . This phenomenon is closely linked wi h he adjus men o wa e con en . Once he plas icise is added, he wa e con en is educed which in luences he p og ess o he se ing o he esh conc e e. This is also connec ed wi h hea low andwi h he eleaseo hyd a ionhea in he i s 24hou s o ageing.I isal eadyknownandhasbeenp o enin p e ious expe imen al measu emen s [33] ha he addi ion o plas icise has a subs an ial in luence on cemen hyd a ion. The addi ion o plas icise e a ds cemen hyd a ion and has a c i ical in luence on i s o e all p og ess. The ma gin o hyd a ion e a da ion is s ongly dependen on he amoun and p ope ies o he cemen and plas icise used in he conc e e mix u e. Based on he measu emen esul s i can be s a ed ha he highe he amoun o cemen is, he g ea e he di e ences a e in he beha iou o he conc e e specimens in he g oups being compa ed (see Figu es 5(b), 6(b), and 7(b) uppe pa s). On he o he hand, all conc e es wi h plas icise (1/1, 1/2 and 1/3) exhibi e y simila beha iou o e he whole ime o ageing, independen ly o he amoun o cemen dosage (see Figu e 8). The p og ess o conc e e olume changes a he ea ly age co esponds well wi h he p og ess o hea low measu ed o pa icula mix u es [33]. The smalles e ec o he addi ion o plas icise was ound in conc e es 0/1 and 1/1 (Figu e 5) which, compa ed wi h o he conc e es, con ained a low amoun o cemen and had a high w/c a io (see Tables 1 and 2). 8 Ad ances in Ma e ials Science and Enginee ing 0 −200 −400 −600 −800 −1000 Sh inkage (m/m) 0 50 100 150 200 250 300 Time (days) 0 100 Mass loss (g/g·10−4) Time (days) 0 50 100 150 200 250 300 −100 −200 −300 −400 −500 0/3 1/3 0/3 1/3 (a) 0 12243648 0 25 50 75 100 Sh inkage (m/m) Time (hou s) −25 −50 012243648 0 25 Mass loss (g/g·10−4) Time (hou s) −25 −50 −75 −100 −125 −150 0/3 1/3 0/3 1/3 (b) Figu e 7: P og ess o sh inkage/swelling and mass loss o conc e es 0/3, 1/3, and ((a): measu emen o e he whole ime o se ing and ha dening; (b): ea ly age measu emen ). Qui e in e es ing esul s we e obse ed in ela ionship be ween sh inkage and mass loss de elopmen (Figu e 9). No di e ences a e obse ed be ween mix u es 0/1, 0/2 and 0/3, and 1/1. Howe e , sligh di e ences a e obse ed a he end o cu es o mix u es 1/1, 1/2, and 1/3 which shows he e ec o di e en imes when he mass o es specimens has been s abilised. 4. Conclusion Recen expe ience wi h measu emen s indica es he necessi y o assessing he magni ude o sh inkage in wo basic s ages, in he ea ly s age o se ing and ha dening ( oughly up o 48 hou s a e cemen is mixed wi h wa e ) and a he “la e ages” which begin a 48 h and mo e. Pe o med ea ly age measu emen s showed di e ences in he de elopmen o conc e es’ sh inkage which could no be de ec ed by measu emen s pe o med on demoulded specimens ( ypi- cally 24 hou s a e cas ing). Measu emen s conduc ed by di e en esea ch eams showed ha gene alisa ions canno be d awn om he absolu e alues o he es esul s bu a he om he ends isible in he eco ded da a p og ess. The pe o med expe imen s con i med his. Despi e he ac ha he measu ed ea ly age de o ma ions we e no e y signi ican in e ms o absolu e alues, he esul s we e use ul o unde s anding he gene al beha iou o he conc e es being examined. Con inuous measu emen o he mass losses caused by ee d ying o he es specimens’ su ace p o ided use ul da a o he in e p e a ion o he esul s o conc e e sh inkage measu emen . Conce ning he cemen Ad ances in Ma e ials Science and Enginee ing 9 1/1 1/2 1/3 0 −200 −400 −600 −800 −1000 Sh inkage (m/m) 0 50 100 150 200 250 300 Time (days) 0 100 Mass loss (g/g·10−4) Time (days) 0 50 100 150 200 250 300 −100 −200 −300 −400 −500 1/1 1/2 1/3 (a) 1/1 1/2 1/3 0 12243648 0 25 50 75 100 Sh inkage (m/m) Time (hou s) −25 −50 0 12243648 0 25 Mass loss (g/g·10−4) Time (hou s) −25 −50 −75 −100 −125 −150 1/1 1/2 1/3 (b) Figu e 8: P og ess o sh inkage/swelling and mass loss o conc e es 1/1, 1/2, and 1/3 ((a): measu emen o e he whole ime o se ing and ha dening; (b): ea ly age measu emen ). dosage, du ing he i s 24 hou s o ageing, all mix u es exhibi ed swelling he magni ude and p og ess o which we e in luenced by he con en o cemen and wa e . Gene ally speaking, he highe he cemen dosage was, he g ea e he alue o olume inc emen eco ded was. In absolu e alues, Figu e 4(b) on he uppe pa anges app oxima ely om 35 o 85 𝜇m/m. Howe e , a simul aneous loss in mass was egis e ed in his s age as well. Conc e es wi h plas icise see a educ ion in he alue o swelling. Also, he cu es o ini ial olume changes eco ded du ing he i s 48 hou s o mea- su emen di e signi ican ly in compa ison wi h conc e es wi hou plas icise . In e ms o he absolu e alues o o al de o ma ion, he obse ed in luences (cemen and plas icise con en )a eno e ysigni ican .Themeasu edp og esso sh inkage well co esponds wi h he p og ess o mass losses. Conce ning he absolu e alue o sh inkage de e mined a 300 days o age, all he examined conc e es ange om 650 o 850 𝜇m/m. Howe e , di e ences we e eco ded in he ime a which de o ma ion s abilised. Conc e es wi hou plas icise saw a ela i ely sha p inc emen o sh inkage be o e he age o 90 days a e which hey began o s abilise. A e he plas icise had been added, he p og ess o sh inkage slowed down in app oxima ely 30 days. In he ea ly s age o conc e e se ing and ha dening, he mass o conc e e specimens wi h plas icise , compa ed wi h he conc e e wi hou plas icise , dec eases mo e apidly. I can be obse ed ha his end is changing app oxima ely a e 7 days o conc e e ageing. Conc e e swelling esul s usually om he ac ion o se e al