Advances in alternative cementitious binders
M.C.G.Juenger a ,⁎,F.Winnefeld b ,J.L.Provis c ,J.H.Ideker d
a
University of Texas at Austin,Department of Civil,Architectural and Environmental Engineering,1University Station C 1748,Austin,Texas 78712,USA
b
Empa,Swiss Federal Laboratories for Materials Science and Technology,Laboratory for Concrete and Construction Chemistry,Überlandstras 129,CH-8600Dübendorf,Switzerland c倒排计划
University of Melbourne,Department of Chemical and Biomolecular Engineering,Parkville,Victoria 3010,Australia d
Oregon State University,School of Civil &Construction Engineering,220Owen Hall,Corvallis,Oregon 97331,USA
a b s t r a c t
a r t i c l e i n f o Article history:
元宵节的古诗有哪些
Received 8April 2010
Accepted 17November 2010Keywords:A.Hydration
D.Alkali-activated cement D.Calcium aluminate cement D.Sulfoaluminate
D.Granulated blast furnace slag
There is a burgeoning interest in the development,characterization,and implementation of alternatives to Portland cement as a binder in concrete.The construction materials industry is under increasing pressure to reduce the energy ud in production of Portland cement clinker and the associated greenhou gas emissions.Further,Portland cement is not the ideal binder for all construction applications,as it suffers from durability problems in particularly aggressive environments.Several alternative binders have been available for almost as long as Portland cement,yet have not been extensively ud,and new ones are being developed.In this paper,four promising binders available as alternatives to Portland cement are discusd,namely calcium aluminate cement,calcium sulfoaluminate cement,alkali-activated binders,and supersulfated cement
s.The history of the binders,their compositions and reaction mechanisms,bene fits and drawbacks,unanswered questions,and primary challenges are described.
©2010Elvier Ltd.All rights rerved.
Contents 1.Motivation .............................................................12322.Specifying alternative binders ....................................................12333.
Alternative binders .........................................................12333.1.Calcium aluminate cements ...........
.......................................12333.1.1.Hydration and property development ..
.......................................12343.2.Calcium sulfoaluminate cements .........
.......................................12353.2.1.Raw materials and binder composition ........................................12353.2.2.Hydration ......................................................12363.2.3.Properties ...............
.......................................12363.3.Alkali-activated binders .............
.
......................................12373.3.1.Reaction mechanisms and binder structure ......................................12373.3.2.Reaction kinetics ...................................................12383.3.3.Primary challenges ...........
.......................................12383.4.Supersulfated cements .............
.......................................12393.4.1.Raw materials and binder composition ........................................12393.4.2.Hydration ......................................................12393.4.3.Properties ...............
.......................................12404.Conclusions .............................................................1240Acknowledgements ............................................................1241References .........................
.......................................
1241
1.Motivation
Since the development of Portland cement over 175years ago,it has become the dominant binder ud in concrete for construction.Annual worldwide Portland cement production is approaching 3Gt
filedCement and Concrete Rearch 41(2011)1232–1243
⁎Corresponding author.Tel.:+15122323593.
E-mail address:mjuenger@mail.utexas.edu (M.C.G.
Juenger).0008-8846/$–e front matter ©2010Elvier Ltd.All rights rerved.doi:
10.res.2010.11.012
Contents lists available at ScienceDirect
Cement and Concrete Rearch
j o u r n a l h om e p a g e :h t t p://e e s.e l s evi e r.c o m /C E M C O N /d e fa u l t.a s p
[1].Praid for its versatility,durability,and economic value,Portland cement concrete is receiving increasing recognition for its relatively low embodied energy compared to other building materials,as shown in Fig.1[2],and for its u of local materials,thereby reducing energy and pollution costs associated with material transport.
However,Portland cement is not without problems.Becau such vast quantities are produced,manufacturing of Portland cement consumes 10–11EJ of energy annually,approximately 2–3%of global primary energy u.Furthermore,Portland cement production results in approximately 0.87t of carbon dioxide for every tonne of cement produced [3];this accounts for 5%of manmade CO 2emissions [3].The cement industry is under pressure to reduce both energy u and greenhou gas emissions and is actively eking alternatives to this familiar and reliable material.
Coupled with the interest in eking low-energy,low-CO 2binders is an interest in finding re-u for waste materials from other industries.Portland cement concrete already accommodates a wide variety of waste materials ud as supplementary cementing materials,including fly ash from coal combustion,ground granulated blast furnace slag from iron production,and silica fume from ferrosilicon production.However,the are ud to replace only a portion of the cement in concrete,typically on the order of 10–50%(although sometimes ud in greater quantities).There is interest in pushing this envelope further,to create binders made entirely or almost entirely from waste materials.
Additional motivation for exploring alternatives to Portland cement can be derived from its shortcomings in certain applications and environments.For instance,rapid-repair applications demand a faster strength gain than Portland cement concrete can provide.Similarly,environmental conditions with high acidity or high sulfate concentra-tions can cau substantial degradation of Portland cement concrete.For such cas,there is a demand for Portland cement alternatives.
New binders are being developed for concrete that promi to reduce the environmental impact of construction,u a greater proportion of waste materials,and/or improve concrete performance.The materials reprent a substantial departure from the traditional chemistry of Portl
and cement,and therefore do not bene fit directly from the many years of rearch into its reaction mechanisms,property development and durability.Furthermore,new materials have dif ficulty finding acceptance in the construction industry,making implementation challenging.In this paper we discuss some of the alternative binders which are attracting increasing attention in rearch and practice,summarizing the current understanding,gaps in knowledge and challenges.This paper does not address all of the possible Portland cement alternatives that are being developed,studied,and ud.Rather,we have chon four materials that show particular promi as Portland cement alternatives:calcium alumi-nate cement,calcium sulfoaluminate cement,alkali-activated binders,and supersulfated cement.
The challenges facing new concrete binders are twofold.Primarily,there are many fundamental questions to be addresd with respect to
processing,chemical and physical behavior,and performance.Secondly,after a strong basic understanding of the material is in place,it is important to establish standard composition and/or performance parameters for the materials and to incorporate them into building codes and speci fications.In this paper the challenges of standardization and speci fications are addresd jointly in the next ction,and we then proceed to discuss the scienti fic challenges unique to each binder p
arately.
2.Specifying alternative binders
Speci fications for building materials can generally be classi fied as either prescriptive or performance-bad.Certainly,a prescriptive speci fication for Portland cement would preclude the u of an alternative binder.A performance-bad speci fication,however,may provide suf ficient flexibility to allow the u of a non-Portland cement binder.There are differing degrees of prescription in cement standards and speci fications in place worldwide.In the United States,ASTM has parallel prescriptive (ASTM C150[4])and performance-bad (ASTM C1157[5])standards for cement,but the acceptance of ASTM C1157is not yet widespread among state regulatory author-ities.ASTM C 1600[6]has recently been adopted and covers the broader category of rapid-hardening hydraulic cements in a perfor-mance-bad approach.In the European Union,EN 197-1[7]is a predominantly prescriptive cement standard which is referenced by the concrete standard EN 206-1[8],and this would appear to place some restrictions on the u of non-Portland cements in that region,unless product-speci fic Technical Approvals can be obtained.Each EU nation also has its own t of national appendices which sit beneath the EU Standards,and some of the are more permissive than others in terms of the scope for introducing alternative binder chemistries.
Other nations including Canada and Australia have good scope for acceptance of materials on a performance basis within existing legislative frameworks.There also exists in Ukraine a highly developed framework of prescriptive standards governing speci fic class and formulations of non-Portland cements,which have been generated through 50years of development of alkali activation technology.
International developments in standards for non-Portland cements are being driven and monitored by RILEM Technical Committee 224-AAM.This committee has a speci fic focus on alkali-activated binders,but the availability of performance-bad standards is motivated by the desire to u performance rather than chemistry as the primary criterion for acceptance of a binder type,since composition-bad criteria are necessarily binder-speci fic.The focus of the RILEM Committee does not speci fically limit the applicability of its outcomes to alkali-activated materials.
音量键Probably the most daunting challenge facing developers of perfor-mance-bad standards is exactly how a testing regime may be designed which is suf ficiently inclusive to enable its u to test and validate a wide range of binder systems,but which is also restrictive enough to ensure good performance of materials when they are mixed and placed under less-controlled real-world conditions.The lection of curing conditions (for example,whether lime –water curing is uful for no
n-Portland cement systems),whether the most critical tests are conducted on precursors,pastes,mortars or concretes,and the need to transfer as much as possible of knowledge from Portland cement and concrete technology to the new binder systems,are all esntial areas which require input from both the commercial and rearch ctors if satisfactory outcomes are to be achieved.3.Alternative binders 3.1.Calcium aluminate cements
Calcium aluminate cements (CACs)are a specialty class of cements containing primarily monocalcium aluminate (CA)and
sometimes
Fig.1.Embodied energy of common building materials [2].
1233
M.C.G.Juenger et al./Cement and Concrete Rearch 41(2011)1232–1243
C12A7and/or CA2.Silica may be prent in small amounts in the form of C2S and/or C2AS(gehlenite).Small amounts of ferrite may also be prent in the form of a C4AF solid solution with variable A/F ratio[9]. Originally developed in the early1900s,near La Teil,France by Jules Bied of the J.&A.Pavin de Lafarge company,CACs were invented to resist sulfate attack[9].Calcium aluminate cement concrete has veral distinct advantages over traditional Portland cement concrete, including rapid strength gain upon tting and enhanced resistance to abrasion,sulfate attack and alkali–silica reaction.Furthermore, production of CAC results in lower CO2emissions than Portland cement production.Currently,CACs are mainly ud in refractory and building chemistry applications,such asfloor screeds and rapid-hardening mortars[9].However,CACs are gaining renewed interest in the construction industry due to their rapid hardening and enhanced durability properties compared to other cementitious binders.
Despite the fact that CAC was developed over100years ago and has many advantageous characteristics,it is not ud as extensively as
Portland cement.Widespread u of CAC is limited by two primary challenges.First,a process called“conversion”occurs in hydrated CAC over time,whereby metastable hydrates convert to stable hydrates, leading to an increa in porosity and subquent decrea in strength.Several building collaps in the1970s were initially attributed to CAC conversion,and many structural codes subquently banned u of the material.Later investigations revealed that in one of the collaps,improper structural detailing was to blame,and other failures were a result of improper material usage despite manufac-turer recommendations.Since this time,intensive rearch has provided a greater understanding of CAC chemistry and behavior.A report by the Concrete Society in1997provided improved guidance for predicting long-term properties,and,as a result,the technical concrete market has en resurgence in interest and u of this alternative cementitious binder[10].Secondly,CAC is expensive compared to Portland cement,with the cost related directly to the limited supply of bauxite,the main source of alumina in CAC production.Both of the challenges need resolution if this material is to gain acceptance as a viable Portland cement alternative.
3.1.1.Hydration and property development
In Portland cement the temperature history affects primarily the rate of reaction,whereas in CAC the temperature during hydration also impacts the phas that form and the rate of transition from metas
table to stable he conversion process).At low curing temperatures,metastable hydrates CAH10and C2AH8form.It is generally accepted that the predominant metastable hydrate formed at temperatures below~15°C is CAH10[11].As the curing temper-ature increas to30°C,C2AH8is also formed;however,the formation of CAH10is not thermodynamically favored and formation of C2AH8is slow as temperatures approach30°C[12,13].The conversion of the hydrates to the stable C3AH6pha is accompanied by the formation of AH3gel and the relea of water.This is a thermodynamically inevitable process.As a result of the conversion process,the paste increas in porosity and subquently the strength of the material decreas.At higher curing temperatures(N70°C)the stable hydrate C3AH6is predominantly formed[9].Fig.2shows a schematic of the conversion process and approximate temperature ranges for the formation of metastable and stable hydrates.
While the conversion process may take years in thefield,it can be simulated in the laboratory by curing at38°C immediately after casting.This leads to an accelerated formation of the stable hydrates, and the subquent minimum in strength is typically realized within 5days[10].Fig.3shows scanning electron microscope images of CAC microconcretes cured isothermally at20°C and38°C,promoting the formation of metastable(higher strength)and stable(lower strength) hydration
products,respectively.In Fig.3a,despite the large amount of unreacted CA prent,the microstructure is relatively den,filled with hydration product(CAH10)and discrete porosity similar to a traditional high-performance Portland cement microstructure.In sharp contrast,in Fig.3b there is little unreacted CA.Additionally, the porosity that exists is evident throughout the entire microstruc-ture,resulting in a lower strength material for38°C isothermal curing than for20°C isothermal curing.
It is recognized that conversion is an inevitable process and rearch has shown that the best approach when using CAC concrete is to design for the lower,converted strength rather than for the maximum strength,thereby eliminating long-term problems post-conversion[10].Recent rearch in CAC systems has focud less
on Fig.2.Schematic of conversion implies densification of hydrates which leads to incread porosity and strength
reduction.
Fig.3.SEM images of CAC microconcrete:a.unconverted,high plete conversion,lower strength,incread porosity(both scale bars=20μm).C=unreacted CA,A=AH3gel,C1=CAH10,C2=C2AH8,C3=C3AH6,R=partially reacted CA grain,P= pore.
1234M.C.G.Juenger et al./Cement and Concrete Rearch41(2011)1232–1243
strength development,but instead on characterizing and predicting dimensional stability in order to reliably predictfield performance,as well as blending the cement with supplementary cementing materials in order to reduce the cost of concrete made from CAC.
Recent rearch efforts investigating calcium aluminate cements were highlighted in the2008Centenary Conference[14].The proceedings include information specific to hydration,incread understanding of mechanical and volumetric properties,durability, refractory and building chemistry applications,and the u offillers and reactive powders in conjunction with CACs.
Early-age microstructural development was investigated by Pöllmann et al.[15]using cryo-scanning electron microscopy(SEM) coupled with heat-flow calorimetry and in situ X-ray diffraction. Combining
the techniques they were able to obrve the transfor-mation of AH3gel to crystalline AH3phas at early ages(up to8h after mixing).From8to15h,cryo-SEM showed an increasing density of CAH10and the crystallization of layered,hexagonal crystals of C2AH8.
Ideker et al.[16,17]demonstrated the profound effect of curing temperature on early-age volume changes.Specifically,under isothermal curing,the formation of metastable hydrates(primarily CAH10)was linked to shrinkage.Converly,formation of stable phas(especially C3AH6)was linked to expansion of the material. The mechanisms governing the volume changes are not obvious. Simple volume change calculations of the hydrates for conversion from CAH10to C2AH8or CAH10to C3AH6show volume increas of4.4 and2.4%,respectively.This includes the relea of water from CAH10 (if this is not done,an incorrect net shrinkage is calculated).The rearchers have conjectured that the ability of the system to accommodate water relea and subquent water movement within the pore structure could be linked to expansion as a result of incread hydraulic pressure[16,17].
Lamberet et al.[18]and Alexander et al.[19]highlighted the improved performance of CAC in wer tunnel linings compared to OPC.The prence of higher concentrations of aluminum ions in CAC systems combined with low pH levels prevalent in wastewater conveyance were shown to inhibit bac
terial growth,thereby reducing damage in CAC-bad mortar linings.
The reaction kinetics and long-term properties of Portland cement-dominated and calcium aluminate cement-dominated lf-levelingflooring mortars were investigated by Kighelman et al.[20]. They found that CAC-dominated systems were more stable volumet-rically due to early-age strength gain and a denr microstructural formation.They also found that the CAC-dominated systems showed improved abrasion resistance compared to OPC-dominated systems.
Goslin[21]found that the incorporation of SCMs,in particular ground granulated blast furnace slag(GGBFS)and silica fume,created more space and available water for the hydration of CA.At early ages this resulted in further hydration of CA in the systems compared to pure CAC systems.Reactivity of SCMs was not obrved until later ages(N1day),similar to Portland cement systems,and as a result the incorporation of SCMs did not significantly enhance early-age strength gain.Infield applications of CACs,an accelerator(typically Li2SO4-bad)is commonly ud to regulate tting time.In pure CAC systems Goslin found that an accelerator actually reduced hydration of CA at early ages and decread strength gain due to an increa in the formation of denr stable hydrates(C3AH6and AH3)at early ages. Thisfinding has implications on the u of CAC infield applications,as it ems that the problem of strength loss associated with conversion ma
y be partially mitigated through the u of SCMs and/or chemical accelerators.
Little work has been done on modeling the reaction kinetics of CAC,as rearch has generally focud on developing predictable mechanical behavior and gaining a better understanding of the formation of metastable and stable hydrates.More detailed investiga-tions of hydration kinetics will be welcomed in the future and will further enhance our understanding of calcium aluminate cement systems and how to utilize them through avoidance of or proper characterization of conversion.
3.2.Calcium sulfoaluminate cements
Calcium sulfoaluminate(CSA)cements contain ye'elimite(C4A3S
P
) as a major constituent(30–70%).Ye'elimite was introduced as a cementitious pha in the1960s,when it was patented by Alexander Klein as an expansive or shrinkage compensating addition to cementitious binders(“Klein's compound”)[22].While CSA cements are not widely ud in Europe and the U.S.,they have been produced, ud and standardized in China for about30years[23–29],where they are known as the“third cement ries.”Two types of clinkers are defined,sulfoaluminate belite clinker(containing mainly(C4A3S
P
)and C2S)and ferrialuminate clinker(containing mainly(C4A3S美国竞选
P
),C4AF and C2S).The clinkers are interground with different levels of calcium sulfate in order to obtain rapid-hardening,high strength,expansive, or lf-stressing cements.CSA cements have been ud in China as a binder for concrete in bridges,leakage and epage prevention projects,concrete pipes,precast beams and columns), prestresd concrete elements,waterproof layers,glassfiber rein-forced cement products,low temperature construction and shotcrete [23,24,26,27,30].In addition,due to their low pH,their low porosity and the ability of ettringite and AFm phas to bind heavy metals, calcium sulfoaluminate cements and their blends with Portland cement are of interest in thefield of hazardous waste encapsulation [31–35].
CSA cements are receiving increasing attention becau they promi to provide a low-CO2alternative to Portland cement[36]. Compared to alite,which releas0.578g CO2per g of the cementing pha when made from calcite and silica,calcium sulfoaluminate clinker releas only0.216g CO2per g of cementing pha when made from limestone,alumina and anhydrite.Thefirin
g temperature ud to produce CSA clinker is typically1250°C,about200°C lower than that ud for Portland cement clinker.In addition,this type of clinker is easier to grind than Portland cement clinker[23].
3.2.1.Raw materials and binder composition
CSA clinker can be produced from limestone,bauxite(iron-rich bauxite in the ca of ferrialuminate clinker)and calcium sulfate (anhydrite or gypsum)[23–29,37,38].The high cost of bauxite prents an economic challenge for CSA cements,just as for CAC. Therefore,a significant amount of effort has been put into exploring industrial by-products or waste materials such asfly ash,blast furnace slag,phosphogypsum,baghou dust or scrubber sludge for the manufacture of calcium sulfoaluminate-bad clinkers[39–42]. Generally,the same production process as for Portland cement clinker,either in shaft or in rotary kilns,can be applied[26],using a clinkering temperature between1250and1350°C.
Depending on the raw meal composition,CSA clinkers can contain various other hydraulic phas such as belite,calcium aluminoferrite, excess anhydrite or free lime,calcium aluminates,perovskite or gehlenite[24,38,43].The latter two phas can be regarded as hydraulically inactive.To increa the r
eactivity of the belite pha, which is responsible for late strength development of sulfoaluminate belite cements,minor ingredients can be added to the raw meal[44].
Usually about15–25wt.%of gypsum is interground with the clinker for optimum tting time,strength development and volume stability.Depending on the level of calcium sulfate addition,CSA cements with different properties can be obtained[26,28,29].CSA cements can also be ud in blends with other cementitious materials including Portland cement,burnt oil shale or limestone[28,32,45–49] to improve their strength development or to formulate rapid tting/ hardening binder systems.
1235
M.C.G.Juenger et al./Cement and Concrete Rearch41(2011)1232–1243
3.2.2.Hydration
关于荷叶的古诗The kinetics of pure ye'elimite hydration and product development are in fluenced by the addition of calcium sulfate or calcium hydroxide
[50–54].With water alone,(C 4A 3S P
取个网名
)
reacts with water according to Eq.(1)to form monosulfate and aluminum hydroxide,with the latter being usually X-ray amorphous.The kinetics of this reaction are quite slow,exhibiting a dormant period of veral hours.The addition of gypsum or anhydrite accelerates the hydration kinetics,and ettringite is formed according to Eq.(2),together with aluminum hydroxide,instead of monosulfate.When the calcium sulfate is consumed,monosulfate is formed according to Eq.(1).The ratio between ye'elimite and calcium sulfate determines the ratio between ettringite and monosulfate in the final product.Above a calcium sulfate to ye'elimite molar ratio of 2,only the reaction according to Eq.(2)occurs.With the addition of calcium hydroxide,ye'elimite reacts very rapidly to form C 4AH x ,whereas the combined addition of calcium hydroxide and calcium sulfate leads to the rapid formation of ettringite according to Eq.(3).
In CSA cements,which generally contain veral hydraulic phas,similar reactions take place [23,24,27,30,44,49,55–59].Usually the
(C 4A 3S P
)is more reactive than the other accessory phas like C 2S,C 4AF or CA [29,44,58–60].Depending on clinker composition,additional hydration products such as strätlingite (C 2ASH 8)(Eq.(4)),calcium
silicate hydrates or CAH 10may form.Fig.4shows the pha development of a calcium sulfoaluminate cement containing belite with ongoing hydration,derived by thermodynamic modeling [58].The water-to-cement ratio needed for complete hydration is determined by the amount of calcium sulfate added,and is at a maximum around an addition of 30%calcium sulfate [23,54].This is higher compared to Portland 0.78for pure ye'elimite reacting with 2molar equivalents of anhydrite [61],or around 0.60for technical cements [23].In comparison to Portland cement,cements bad on calcium sulfoaluminate react faster,and most of the hydration heat evolution occurs between 2and 24h of hydration [59].Typical values for heat of hydration are clo to 400J/g cement after 72h by conduction calorimetry [58,62].
毕业答辩C 4A 3S +18H →C 3A ⋅CS ⋅12H +2AH 3monosulfate formation ðÞ
ð1Þ
C 4A 3+2H 2+34H →C 3A ⋅⋅32H +2AH 3ettringite formation ðÞ
ð2Þ
C 4A 3S +CS H 2+6CH +74H →3C 3A ⋅3CS ⋅32H ettringite formation ðÞ
ð3Þ
C 2S +AH 3+5H →C 2ASH 8str a tlingite formation ðÞ
ð4Þ
The hydration of calcium sulfoaluminate cements depends mainly on the amount and reactivity of the added calcium sulfate [23,29,54,57,58,63,64].Therefore,a formula for the calculation of optimum sulfate level (Eq.(5))to obtain the different types of CSA cements was developed in China,bad on stoichiometric calculations [29]:
C T =0:13⋅M S
ð5Þ
where C T =ratio gypsum/clinker,A =mass %of ye'elimite in the clinker,S _
=mass %of SO 3in the gypsum,M =molar ratio gypsum/ye'elimite,and the value 0.13is a stoichiometric factor containing all the conversions between mass and molar units.The value M (and also whether gypsum or anhydrite is ud)is related to the type of CSA cement.M =0–1.5with a low-c
alcium sulfate content yields a rapid-hardening or high strength cement,which is also con firmed by the experimental data given in [57].Higher sulfate levels are applied to formulate expansive (M =1.5–2.5)and lf-stressing cements (M =2.5–6).Shrinkage compensating,expansive and lf-stressing cements generate expansive (compressive)stress in the final,dried paste,mortar or concrete,which are low (b 1MPa)in the ca of shrinkage compensated and high (up to 8MPa)in the ca of lf-stresd systems [65].The latter have to be restrained by a suitable reinforcement in order to avoid excessive expansion and crack formation.Thus,according to Eq.(5),the properties of the CSA cements are directly related to the formation kinetics and to the total amount of the voluminous ettringite pha in the hardened system.
There are not many data available on pore solution chemistry in this system [27,44,58,66].The liquid pha is dominated at an early age by Na,K,Ca,Al and sulfate,until the added calcium sulfate has been consumed.The pH value in this period is between 10and 11.After consumption of the calcium sulfate,a strong decrea of calcium and sulfate concentrations and an increa of pH to about 12.5occur.During the first hours of hydration,silicate concentrations in the pore solutions are lower than for OPC,wheras after veral days they are comparable in both systems.
Microstructural investigations [24,27,57–59,67],Fig.5,have revealed mainly the prence of space-filli
ng ettringite needles,together with monosulfate,aluminum hydroxide,and calcium silicate hydrates and/or strätlingite,leading to a very den,low porosity microstructure.
3.2.3.Properties
The tting times of CSA cements depends on their ye'elimite content,the kind and content of minor phas,and the amount and reactivity of the added calcium sulfate.Typical values are between 30min and 4h [23,27–29,44].Compared to Portland cement,CSA cements in general reach higher early and late strengths [23,26–30,56,57].
CSA cements exhibit a chemical shrinkage,which is related to the fact that the apparent density of the water bound in the hydrated phas,such as ettringite,is higher than the density of free water.It can be calculated through thermodynamic modeling that CSA cements should have a theoretical chemical shrinkage of about 11cm 3/g cement after 28days [58],whereas a Portland cement reaches about 4–5cm 3/g.Chemical shrinkage of CSA cement was experimentally found to be of the same order of magnitude as the predicted value [62].It should also be noted that expansion may occur if ettringite forms in reasonable amounts after tting,which can be triggered by the amount of added calcium sulfate [29].Calcium oxide and calcium hydroxide accelerate ettringite formation,and so can also lead to expansion [27,57].Further,due to the high water/cement ratio needed for complete hydration,which is typically around 0.60,CSA cements tend to undergo lf-desiccation,as a low water-to-cement ratio of 0.30–0.45is typically ud
[23].
Fig.4.Pha development of a CSA cement (water/cement =0.80)as a function of hydration time calculated by thermodynamic modeling,taken from [58].
1236M.C.G.Juenger et al./Cement and Concrete Rearch 41(2011)1232–1243
