Construction and Building Materials 43 (2013) 348–361

Contents lists available at SciVerse ScienceDi rect

Construc tion and Buildi ng Materia ls

journal homepage: www.elsevier .com/locate /conbui ldmat

Strength and elastic properties of mortars with various percentages of environmentally sustainable mineral binder

0950-0618/$ - see front matter � 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.conbuildmat.2013.02.061

⇑ Corresponding author. Tel.: +40 232 278683x1455; fax: +40 232 239213. E-mail addresses: iotoma@ce.tuiasi.ro (I.-O. Toma), covadan@ce.tuiasi.ro

(D. Covatariu), anamtoma@ce.tuiasi.ro (A.-M. Toma), gtaranu@ce.tuiasi.com(G. Taranu), mbudescu@tuiasi.ro (M. Budescu).

Ionut-Ovidiu Toma ⇑, Daniel Covatariu, Ana-Maria Toma, George Taranu, Mihai Budescu ‘‘Gheorghe Asachi’’ Technical University of Iasi, Faculty of Civil Engineering and Building Services, No. 43rd, Prof. D. Mangeron Blvd., Ia s�i 700050, Romania

h i g h l i g h t s

� High percentages of Kerysten has benefits in terms of density. � High percentages of Kerysten has a beneficial effect on the flexural strength of mortar. � High surface area which contributes to hardened properties of mineral matrices. � All specimens with Kerysten exhibited lower values for the compressive strength. � Some equations accurately estimate the modulus of elasticity of mortar with Kerysten.

a r t i c l e i n f o

Article history: Received 20 August 2012 Received in revised form 1 February 2013 Accepted 26 February 2013 Available online 30 March 2013

Keywords:Sustainable mineral binder Cement replacement Strength characteristics Modulus of elasticity

a b s t r a c t

The production of cement alone has increased dramatically over the past 80 years due to a continuous increase in demand for concrete. Due to increased public and scientific community awareness on the environmen tal impact of the construction industry extensive efforts have been put to reduce the CO 2emissions of the Portland cement plants. A major step in the direction of sustainable development has been done with the development of alternative cementitious binders. An important source of such mate- rials is the industry. The present pap er presents the results of a research work focused on the suitability of using a new mineral binder as partial replacement of ordinary Portland cement in concrete. At this stage of the research the strength and elastic properties of mortars incorporating various percentages of the new binder were investigated. The binder is obtained from industrial by-products, most of them being disposed in landfills. The experimental results lead to the conclusion that lighter structural or non-struc- tural elements could be made with the incorporation of the new binder. In terms of bending tensile strength the obtained values were much higher than the reference mix proportion. On the other hand, the compressive strength did not match that of the reference mix. The volume of Portland cement sub- stitution as well as the type of cement plays an important role on the strength and elastic properties of the mortars.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Mortars have been used for a variety of applicati ons since ancient times [1]. The mortars were made using mainly three types of binders: mud, probably the oldest type of binder used for pre- paring the mortar, gypsum and lime. The binder contributes to the workability of the mortar, whereas the aggregat e influencesthe mechanical propertie s and helps in controlling shrinkage related problems [2–4]. With the first introduction of the Portland cement in 1824 [5] the strength of cementitious construction materials, e.g. mortars, concrete, became higher and higher [6],

following the trend of constantly increasing expectations related to their performanc e [7–9].

The production of cement alone has increased dramatically over the past 80 years due to a continuous increase in demand for con- crete. According to a U.S. Geological Survey statistic report, the ce- ment production increased, worldwide, from 62.4 million metric tons, in 1926, to 3.06 billion metric tons, in 2009. Taking into ac- count that the cement content in normal strength concrete ranges from 10% to 15%, by weight, [10] and that the concrete industry is the largest consumer of natural resources in the world, roughly 11.5 billion tons a year [11], one could only imagine the environ- mental burden the constructi on industry creates [12–14].

Due to increased public and scientific community awarene ss on the environmental impact of the constructi on industry [15,16]extensive efforts have been put to reduce the CO 2 emissions of the Portland cement plants [17], particulate air emissions, noise

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 349

pollution, water pollution [18], etc. The innovation in the construc- tion industry is continuously driven not only by the need of reduc- ing the environmental footprint [19] but also by the increasing demand of reducing the costs in compliance with the evolution of the global market [20]. This can only be achieved by a constant and sustained innovation process which should also anticipate the future environm ental constraints [21–23].

A major step in the direction of sustainable development has been done with the development of alternativ e cementitious bind- ers [24]. An important source of such materials is the industry. Fly ash, a by-produ ct of the coal-fired power plants and municipal incinerators [25], has been successfu lly used in constructi on indus- try to improve the properties of concrete , such as strength and durability [26,27]. Ground granulated blast furnace slag has been used since 1950 either as partial replacemen t of ordinary Portland cement or as fine aggregat e in concrete mix design [19] with highly improved resistance to aggressive environments [28]. Silica fume, avery reactive pozzolan due to its chemical and physical properties, has a relatively short period of time since its application in con- crete industry. Since its first use in 1970, silica fume became one of the world’s most valuable and versatile admixture for concrete. Significant improvements in the early age strength propertie s of mortars have been reported when silica fume was used [29,30].In addition, significant increases in the tensile and compressive strength of concrete have been observed [31] along with being avery advantag eous stabilizin g agent for self compacting concrete ,thus preventing segregati on [32]. Recent studies show the influ-ence of the silica fume on the dynamic properties of concrete, such as the dynamic elastic modulus and the damping ratio [33].

In view of the new stricter regulations in terms of CO 2 emissions[15] as well as continuous developmen ts in both electric power and steel production one can only expect a significant decrease in the quantities of resulted fly ash and blast furnace slag [21].There are, however, other sources of production for the supple- mentary cementitious materials (SCMs), some of them used for centuries [34,35] but still requiring mining the natural deposits and some of them more recent, obtained from the treatment of industrial by-products such as phosphogypsum (PG) [36,37] whichhas been stored in landfills scarring the scenery and creating many environmental problems.

The present paper presents the results of a research work focused on the suitability of using a new mineral binder as partial replace- ment of ordinary Portland cement in concrete. At this stage of the research the strength and elastic properties of mortars incorporat- ing various percentages of the new binder were investiga ted before moving to concrete and structural concrete elements . As the finalgoal is to evaluate the suitabilit y of using the mineral binder in rein- forced concrete elements, the next stage consists in determining the potential corrosion upon traditional steel reinforcement.

The binder is obtained from industrial by-products , most of them being disposed in landfills, such as phospho gypsum, lacto- gypsum, flue gas desulphurisation gypsum and can be entirely recycled after its expiration date [38]. Taking into account that gypsum results from the production process of many organic and inorganic fertilizer s, pigments, metals etc. it has become a signifi-cant ecological problem [39]. It is the belief of the authors that the use of a new mineral binder based on the anhydrous calcium sul- phate in the b-anhydrite III’ form as SCM has both environmental and economical advantages and justification [40,22,23].

Recent research works show the possibility of using the green binder in dwelling structures [41]. The newly proposed building system is based on reinforcing the mineral matrix, obtained by replacing part of the ordinary Portland cement with the new bin- der, by means of glass fibres. However, at this stage of the research the use of the green mineral binder together with steel reinforce- ment is still unclear from the point of view of corrosion effects.

2. Materials

2.1. Ordinary Portland cement

Two types of cement were used as this stage of the research, both of them being produced according to the SR EN 197-1:2002 standard specifications [42]. The firsttype was a CEM I 42.5R cement with high early strength. It is a general purpose cement being suitable for all uses in works requiring high strength values at early ages. The other type of cement considered was a CEM II B-M (S-LL) 32.5R cement also with high early strength. It is considered to be a composite cement with 65–75% Portland cement and 25–35% ground granulated blast furnace slag and lime. This second type of cement was considered in view of the fact that almost two thirds of the European cement market corresponds to CEM II cement [43]. Both cement types are readily available on the market and widely used, this being the main reason of their selection for this research.

2.2. Sand

As with the cement, two types of sand were considered, with different maxi- mum grain sizes. It is widely recognized now that the aggregate size plays an important role on the strength and elastic characteristics of mortars [44,45] andconcrete [46,47]. The maximum grain sizes of the two types of sand were 0.3 mm and 1 mm. The larger grain size sand is also readily available on the market. The sand was used in its dry state for each mix proportion considered at this stage of the research.

2.3. Recycled mineral binder – anhydrous calcium sulphate (Kerysten)

The anhydrous calcium sulphate (ACS) was used as a replacement for the ce- ment. The percentages used in this research were: 15%, 20%, 25%, 30%, 35% and 40% of the total volume. Together with the cement, the recycled mineral binder is used to obtain a more eco-friendly binder for mineral matrices. The anhydrous cal- cium sulphate, in the form of b anhydrite III 0 (CaSO4�eH2O) [38], was obtained from industrial wastes, most of them unrecyclable such as: phosphogypsum (industrialwaste from the production of phosphoric acid, a key ingredient for fertilizers and detergents), flue gas desulphurization gypsum, FGD (industrial waste from coal firepower plants), lactogypsum (industrial waste from the production of lactic acid, used mainly in food preservatives), etc. The opportunity for using such industrial unrecyclable wastes in construction industry, especially the phosphogypsum, has recently been recognized by researchers as having net benefits for the environment [48,49].

According to the invention patent [38], the production of ACS, from now re- ferred to as Kerysten, involves low temperatures, less than 750 �C, and no CO 2 emis-sion at all. Moreover, it is entirely recyclable after its expiration date. The grain size ranges from 5 lm to 100 lm and the BET N2 surface is larger than 10 m2/g.

The absolute density of the powdery material depends on the gypsum source and can range from 2360 kg/m 3, for natural gypsum and phosphogypsum, to 2740 kg/m 3, for flue gas desulphurization gypsum.

3. Experimen tal procedure

3.1. Sample preparati on and curing conditions

Table 1 presents the mix proportions considered for the mortars in this research. The mineral binder was assumed to occupy 50% of the total volume whereas the aggregates to occupy the remaining 50%. The mineral binder consisted either only of Portland cement or as a mixture between the Portland cement and Kerysten. The water to mineral binder ratio of 0.4 was kept constant throughout the experimental works. Only tap water was used without any super-plas ticizers or setting retarders.

Each specimen was named according to the type of cement used, CEM I or CEM II, followed by the letter K for the mix propor- tions where Kerysten was used to replace a part of the cement as mineral binder. The percentage of the total volume occupied by the Kerysten follows the letter K. Finally, the designation s ‘‘0.3’’ or ‘‘1’’ were used to specify the maximum grain size of the sand. Hence, C1K20-0.3 defines the mix proportio n for which CEM Iwas used in the mineral binder together with 20% Kerysten and the maximum grain size of the sand was 0.3 mm.

The mortar paste was cast in 160 � 40 � 40 mm (Length �Thicknes s �Width) prisms and in 50 � 100 mm cylinders. The moulds were covered by wet cloth and kept for 24 h at constant room temperature of 23 �C. After demouldi ng the specimens

Table 1Mix proportions and specimen designations.

Specimen designation Binder Sand Water/binder ratio

Cement type Kerysten Max. grain size

CEM I CEM II 0.3 mm 1.0 mm

(%) (%) (%) (%) (%) –

C1K0-0.3 50 – – 50 – 0.4 C1K15-0.3 35 15 C1K20-0.3 30 20 C1K25-0.3 25 25 C1K30-0.3 20 30 C1K35-0.3 15 35 C1K40-0.3 10 40 C1K0-1.0 50 – – – 50 C1K15-1.0 35 15 C1K20-1.0 30 20 C1K25-1.0 25 25 C1K30-1.0 20 30 C1K35-1.0 15 35 C1K40-1.0 10 40 C2K0-1.0 – 50 – – 50 C2K15-1.0 35 15 C2K20-1.0 30 20 C2K25-1.0 25 25 C2K30-1.0 20 30 C2K35-1.0 15 35 C2K40-1.0 10 40

350 I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361

containing only Portland cement as mineral binder were placed in water for curing until the day of testing in order to prevent shrink- age cracking. The other specimens, containing both Portland ce- ment and Kerysten as mineral binder, were cured in air, at room temperature , until the day of testing. The choice of air curing for the specimens containing Kerysten was based on the fact that being a derivative of gypsum, high expansion could be induced by the presence of water [40,50]. However, a recent study showed improved mechanical propertie s of mortars made with Kerysten and cured in water for various periods of time [51].

Fig. 1. Three-point loading test of mortar prisms.

3.2. Testing methodology

The prisms were tested at the ages of 1, 2, 3, 7, 14, 21 and 28 days. Six specimens were cast for each considered curing time interval, resulting in a total number of 3528 prisms. Before loading, the prisms were measured and weighed. This was done both to check for changes in dimension and to determine the density of the hardened mortar.

The prisms were subjected to a three point bending test, as shown in Fig. 1, with a loading rate of 50 N/s accordin g to the stan- dard specifications [52]. The method was used to determine the flexural strength of mortar prisms. The resulting half prisms were examined for signs of visible cracks and then subjected to uniaxial compression test, Fig. 2, with a loading rate of 2400 N/s [52].

The cylinders were used to determine the static modulus of elasticity in compression according to ASTM C469 specifications[53] at the ages of 21 and 28 days. The loading rate was 0.23 N/ mm2/s. Before being tested, both ends of the cylinders were ground to ensure a smooth and plane surface. The axial strain was mea- sured with the help of a 50 mm gage length extensom eter, Fig. 3,whereas the transverse strain was measured by means of a circum- ferential extensom eter, also shown in Fig. 3. Six cylinders were cast for each age and mix proportion mentioned in Table 1, resulting in a total number of 252 cylinders. Out of the six cylinders, only fivewere used to determine the static modulus of elasticity and Pois- son’s ratio. The first cylinder was loaded up to failure in order to determine the loading range specified in ASTM C469 provision s[53]. For each of the remaining five cylinders, 5 loading cycled were

conducte d and out of which only the last three were used to deter- mine the modulus of elasticity and Poisson’s ratio.

4. Results and discussion s

The obtained results are presente d in terms of the density, strength characterist ics and elastic properties. The strength char- acteristics refer to the bending tensile strength and uniaxial com- pressive strength. The elastic properties focused on in the present paper were Young’s modulus and Poisson’s ratio of the mortar.

4.1. Density

The density is an important parameter to be taken into account especiall y when a material is used either for repair [54] or for

Fig. 2. Uniaxial compression test on the resulting half prisms.

Fig. 3. Loading set-up for the determination of elastic properties of mortar mixes.

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 351

strengthening techniques [55] as well as for new constructions. Being a newly developed mineral binder, the assessme nt of the influence of Kerysten on the density of the mortar specimens was one of the objectives of this research.

The dimensio ns of the mortar prisms were measured by means of a digital calliper with the accuracy of 0.01 mm at the day of test- ing. Four measurements were taken for the length of the prisms and three for width and thickness, respectively . After the measure- ment, the specimens were weighed with the precision of 0.01 g.

It should be mentioned that there were no visible cracks on the surface of the prisms no matter what the testing age was.

Table 2 presents the obtained results in terms of the geometri- cal dimensions and the density for all mix proportio ns considered in the present paper. All values were obtained at the age of 28 days and they represent the average values measure d for six specimens for each mix proportion.

The computed values for the density, at the age of 28 days, cor- responding to each mix proportion do not change significantly for the mixes made with the same percentage of gypsum based min- eral binder. The differences between similar mix proportions were between ±0.2%. It can be therefore concluded that the type of cement and the type of sand used in this research do not have sig- nificant influences on the density of the mortars. The statement is

not generally valid [56] and was based only on the types of sand used in this research. The highest value for the density was ob- tained for the mortar made with only Portland cement and sand in equal proportions by volume. The obtained values are very close to those reported in the literature for sprayed concrete [54,57].

The mortars containing Kerysten as mineral binder replacing the Portland cement exhibited lower values for the density. The variation was between 10% and 19% depending on the percentage of the new mineral binder used in the mix proportion.

Fig. 4 presents the variation of the density for the first batch of mixes based on CEM I cement and sand with a maximum grain size of 0.3 mm. It can be observed that the values change significantlyduring the first 7 days and then steadily increase or decrease, depending on the mix, until the 28th day. The reference mix, C1K0-0.3, shows an increase in the density whereas all other mixes exhibit lower values for the density compared to the value deter- mined at 24 h after casting. Significantly lower values of the den- sity were obtained for the mortars with more than 25% gypsum based mineral binder in their mix.

Fig. 5 presents the change in the values of the density for the second batch of mixes based on CEM I cement and sand with amaximum grain size of 1.0 mm. The trend observed in Fig. 4 canalso be reported in this case – the higher the Kerysten percentage, the lower the density of mortars. However, the way the values de- crease is somehow different than in the previous case. There is alarger scattering of the results for the mixes containing 25%, 30% and 35% of Kerysten especially during the first 14 days of curing. This could be attributed to different rates of losing water as all the specimens made with Kerysten were cured in air, at room tem- perature , until the day of testing. The larger maximum grain size could allow for more or less water being trapped under it, depend- ing on the position of the sand grain [46].

Fig. 6 shows the variation of the density for the third batch of mixes based on CEM II B-M (S-LL) cement and sand with a maxi- mum grain size of 1.0 mm. The trend is very similar to the one ob- served in Fig. 5.

Based on the results presented so far it can be concluded that the higher the percentage of Kerysten in the mix proportion, the lower the values of the density for the correspondi ng mortar. On the other hand, it can be observed that the density at the age of 28 days for the mixes made with 25%, 30% and 35% gypsum based binder is almost the same and it does not depend on the type of cement used.

Fig. 7 presents the change in dimensions , expresse d in percent- age, of the mortar prisms at the age of 28 days for the mortar mixes made with CEM I cement and sand with a maximum grain size of 0.3 mm. It should be mentioned that none of the specimens exhib- ited a reduction in the dimensions despite the fact that all mixes were cured in air at room temperature , with the exception of the C1K0-0.3 case which was cured in water. No cracks, due to shrink- age [58] were observed either. It can be observed that the change in length was less significant, less than 1%, than the change in the width and thickness of the prisms. At the same time, the most stable mix, from the point of view of geometrical dimensions, was the C1K35-0.3 case. On the other hand, the C1K25-0.3 exhibited the largest variations of the dimensions.

Similar observations can be made for the results shown in Fig. 8for the mortars made with CEM I cement but with 1.0 mm maxi- mum grain size of sand. As with the previous case, the length change is less significant than the variation of width or thickness. The mix proportions made with more than 25% Kerysten showed the smallest change in dimensions whereas the other mix propor- tions exhibited larger values. Based on the obtained results it can be concluded that the aggregate size influences the geometri cal stability of the specimens. This observati on is supported by earlier results published in the scientific literature [40,47,58].

Table 2Geometrical dimens ions and densities at 28 days (average values of six specimens).

Specimen Width (mm) COV (%) Thickness (mm) COV (%) Length (mm) COV (%) Mass (kg) Density (kg/m3)

C1K0-0.3 40.15 0.10 40.14 0.10 160.81 0.11 564.25 2141.87 C1K15-0.3 40.38 0.11 40.38 0.11 160.88 0.1 515.96 1971 C1K20-0.3 40.35 0.11 40.34 0.1 161.21 0.32 495.19 1776.4 C1K25-0.3 40.73 0.12 40.73 0.13 161.3 0.33 496.01 1817.15 C1K30-0.3 40.11 0.10 40.09 0.1 160.55 0.22 472.06 1828.51 C1K35-0.3 40.02 0.10 40.02 0.1 160.28 0.15 466.47 1887.11 C1K40-0.3 40.25 0.09 40.23 0.11 160.5 0.17 461.67 1853.63 C1K0-1.0 40.45 0.10 40.43 0.1 160.88 0.15 556.31 2114.43 C1K15-1.0 40.60 0.10 40.55 0.11 161.02 0.29 505.04 1905.15 C1K20-1.0 40.85 0.11 40.72 0.12 161.1 0.22 501.48 1762.34 C1K25-1.0 40.11 0.09 40.1 0.1 160.19 0.19 467.55 1814.64 C1K30-1.0 40.14 0.11 40.15 0.1 160.25 0.1 467.03 1808.35 C1K35-1.0 40.12 0.12 40.12 0.12 160.55 0.11 466.44 1804.95 C1K40-1.0 40.09 0.10 40.12 0.11 160.19 0.09 454.07 1871.35 C2K0-1.0 40.51 0.10 40.49 0.1 160.55 0.1 565.22 2146.35 C2K15-1.0 40.60 0.15 40.55 0.09 160.98 0.15 502.29 1895.23 C2K20-1.0 40.86 0.11 40.87 0.11 160.55 0.18 501.80 1871.63 C2K25-1.0 40.56 0.13 40.55 0.12 161.02 0.2 482.49 1821.88 C2K30-1.0 40.25 0.09 40.22 0.1 160.68 0.1 472.58 1816.78 C2K35-1.0 40.15 0.10 40.11 0.11 160.4 0.11 466.86 1807.35 C2K40-1.0 40.18 0.10 40.2 0.1 160.22 0.17 463.91 1792.58

17001750180018501900195020002050210021502200

0 7 14 21 28

Den

sity

[kg/

m3]

Curing time [days]

C1K0-0.3

C1K15-0.3

C1K20-0.3

C1K25-0.3

C1K30-0.3

C1K35-0.3

C1K40-0.3

Fig. 4. Density of specimens made with CEM I cement and 0.3 mm sand.

17001750180018501900195020002050210021502200

0 7 14 21 28

Den

sity

[kg/

m3]

Curing time [days]

C1K0-1.0

C1K15-1.0

C1K20-1.0

C1K25-1.0

C1K30-1.0

C1K35-1.0

C1K40-1.0

Fig. 5. Density of specimens made with CEM I cement and 1.0 mm sand.

352 I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361

Fig. 9 presents the results for the mortar mixes made with CEM II B-M (S-LL) cement and sand with a maximum grain size of 1.0 mm. The length change is again less significant than the trans- versal dimensio ns. Very small variation s of the prisms dimensions were obtained for the mixes with high percentages of calcium sul- phate based mineral binder, that is for 30–40% of the total volume.

The graphica l results coupled with the data shown in Table 2 leadto the conclusion that using high percentages of Kerysten in a mix proportion has benefits both in terms of density as well as in an

increased stability of the geometrical dimensio ns of the specimens .Still, the numerical values for the dimensio ns of the specimens are not that far off the reference dimensio ns (40 mm � 40 mm �160 mm), both in absolute values or expresse d as percentages. The type of cement has only a slight influence on the results, with larger values measured for the width and thickness of the prisms made with CEM II cement, especiall y for the reference mix and for the mix proportio ns made with 15%, 20% and 25% Kerysten. Similar observati ons were made in a recently published studies by

Mor

tar e

xpan

sion

[%]

1.11

%

1.06

%

0.55

%

1.48

%

1.36

%

0.63

%

2.08

%

1.77

%

0.68

%

0.27

%

0.25

%

0.12

%

0.35

%

0.37

%

0.16

%0.30

%

0.30

%

0.34

%

0.22

%

0.30

%

0.12

%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

Width Thickness Length

C1K0-1.0

C1K15-1.0

C1K20-1.0

C1K25-1.0

C1K30-1.0

C1K35-1.0

C1K40-1.0

Fig. 8. Dimension changes for the mortar batch made with CEM I cement and 1.0 mm maximum sand grain size.

Mor

tar e

xpan

sion

[%]

0.37

%

0.35

% 0.50

%

0.94

%

0.94

%

0.55

%

0.87

%

0.84

%

0.75

%

1.79

%

1.79

%

0.81

%

0.27

%

0.22

%

0.34

%

0.05

%

0.05

%

0.17

%

0.62

%

0.57

%

0.31

%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

Width Thickness Length

C1K0-0.3

C1K15-0.3

C1K20-0.3

C1K25-0.3

C1K30-0.3

C1K35-0.3

C1K40-0.3

Fig. 7. Dimension changes (expansion) for the mortar batch made with CEM I cement and 0.3 mm maximum sand grain size.

17001750180018501900195020002050210021502200

0 7 14 21 28

Den

sity

[kg/

m3]

Curing time [days]

C2K0-1.0

C2K15-1.0

C2K20-1.0

C2K25-1.0

C2K30-1.0

C2K35-1.0

C2K40-1.0

Fig. 6. Density of specimens made with CEM II cement and 1.0 mm sand.

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 353

Mor

tar e

xpan

sion

[%]

1.26

%

1.21

%

0.34

%

1.48

%

1.36

%

0.61

%

2.10

%

2.13

%

0.34

%

1.38

%

1.36

%

0.63

%

0.62

%

0.55

%

0.42

%

0.37

%

0.27

%

0.25

%0.45

%

0.50

%

0.14

%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

Width Thickness Length

C2K0-1.0

C2K15-1.0

C2K20-1.0

C2K25-1.0

C2K30-1.0

C2K35-1.0

C2K40-1.0

Fig. 9. Dimension changes for the mortar batch made with CEM II B-M (S-LL) cement and 1.0 mm maximum sand grain size.

354 I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361

Rodriguez et al. [59] and Garg and Pundir [60]. On the other hand, the type of sand does seem to have a more significant influence on the dimensional stability, especiall y for the mixes with higher per- centages of calcium sulphate based mineral binder 25–40%. The lar- ger the grain size diameter was, the lower the change in the width, thickness and length especially for the mixes made with 25% and 40% Kerysten. The authors believe that further research should be conducted at the material level in order to fully understand the underlying mechanism s related to the hydration of the cement combined with the new environmental ly friendly mineral binder.

4.2. Strength characteristics

4.2.1. Flexural tensile strength Portland cement based mortars and concrete are the most

widely used constructi on materials. Their advantag es reside in their low cost, high compress ive strength and ease of fabrication, to name just a few. However, there are some limitations amongst which low tensile strength, brittleness and, to some extent, long term durabilit y. Extensive research effort has been put into improving these properties. Using steel as reinforce ment for con- crete partially solves the problem of the insufficient tensile strength of the mineral matrix.

The pending question is how would this new supplementary cementitious material would influence the flexural tensile strength of mortars?

The values of the flexural tensile strength determined accordin gto SR EN 196-1 [52] for the mortar batch made with CEM I 42.5R cement and sand with maximum grain size of 0.3 mm are shown in Fig. 10 as average values of 6 determination s. It can be observed that the reference mix exhibits the largest value for the tensile strength for the first 72 h from casting but quickly falls behind the mix proportions with high percentages of Kerysten starting from the age of 7 days. The C1K20-0 .3 and C1K25-0.3 cases exhibit a slightly lower gain in strength during the first 7 days but at the age of 28 days the values are 20% and 24% higher than the refer- ence mix, respectively . On the other hand, the C1K15-0.3 mix shows similar values for the flexural tensile strength as the refer- ence mix. This is encouraging since the same strength can be ob- tained using less Portland cement.

The percentage of calcium-s ulphate based binder has a net ben- eficial effect on the flexural tensile strength of mortar with 42% improvement over the reference mix for the C1K40-0.3 obtained at the age of 28 days. Taking into account that the specimens were

cured in air at room temperature and that no extra moisture was provided , it clearly recommends this new binder for use in areas where supplying enough moisture for the hydration of cement might be an issue. On the other hand, for a given percentage of Kerysten , the evolution of the tensile strength seems to be largely influenced by the cement. This statement is based on the fact that from the age of 14 days the gain in strength is only minor with re- spect to the evolution until that day. Since a high early age strength cement was used, this phenomeno n could be attributed to the type of cement and not to the presence of the new cementitious binder.

The results for the second batch of mortars made with CEM I42.5R but with coarser sand with a maximum grain diameter of 1.0 mm are shown in Fig. 11 . Similar trends could be observed when coarser sand was used as the ones showed in Fig. 10 .Although the use of finer grain sand generally produces better re- sults in terms of flexural behaviour of concrete elements [61],based on the obtained results it can be concluded that increasing the maximum grain size of sand does not seem to have significantinfluence on the flexural tensile strength of mortar mixes studied in the present research.

Again, the percentage of calcium-sulpha te based binder has anet beneficial effect on the flexural tensile strength of mortar with 43% improvement over the reference mix for the C1K40-1. 0 ob- tained at the age of 28 days, only slightly better than the result ob- tained for the C1K40-0.3 at the same age.

Fig. 12 presents the results obtained for the batch made with CEM II cement and the coarser sand. The trend is very similar to the previous two batches of mortars for the first 14 days of curing. At the age of 28 days however , even the mix proportion made with 15% Kerysten shows higher flexural strength than the reference specimen by as much as 21%. This could be attributes to the chem- ical reactions between the Kerysten and the constituents of the CEM II B-M (S-LL) 32.5R cement. This statement is further sup- ported by the fact that the values of the flexural tensile strength are almost the same with the ones obtained for the batches made with CEM I cement.

The high values for the flexural tensile strength obtained for all mix proportions made with 30%, 35% and 40% Kerysten, irrespec- tive of the type of Portland cement used or maximum grain size of sand, were directly influenced by the presence of the calcium sulphate based binder with its high surface area [38] which con- tributes to the rheological and hardened properties of mineral matrices [62]. A recently published study [59] demonstrated that the addition of gypsum causes a noticeab le increase of ettringite

4.99 5.13 5.3 5.48 6.

17 6.45 6.

73

3.44 4.

03 4.69 5.

38

6.49 6.

79 6.85

2.99 3.

35 3.77

4.58

7.97 8.

34 8.5

2.92

2.97 3.12

4.55

8.19 8.

56 8.89

2.92 3.

24 3.5

7.46

9.52 9.

89 10.0

1

2.8 3.02 3.

6

7.27

9.3 9.

87

9.91

3.09 3.21 3.

89

7.35

11.1

2

11.3

5

11.7

8

0

2

4

6

8

10

12

14

1 2 3 7 14 21 28

Flex

ural

tens

ile s

tren

gth

[MPa

]

Curing time [days]

C1K0-0.3C1K15-0.3C1K20-0.3C1K25-0.3C1K30-0.3C1K35-0.3C1K40-0.3

Fig. 10. Flexural tensile strength for the mortar batch made with CEM I cement and 0.3 mm maximum sand grain size.

5.21

5.24 5.38 5.56

6.41 6.

76 6.89

3.14 3.34 3.58

4.38

6.32 6.42 6.

84

3.03 3.

63

3.67

6.37

9.29 9.38 9.53

3.07 3.

41

3.42

8

9.53

10.3

9

10.9

3

2.84 3.

41 3.51

7.55

9.17 9.

88 10.1

2

3.09 3.

74

5.05

7.66

9.52

11.2

8

11.6

3

3.37

4.26 4.

58

9.88

11.3

5

11.6

7 12.2

8

0

2

4

6

8

10

12

14

1 2 3 7 14 21 28

Flex

ural

tens

ile s

tren

gth

[MPa

]

Curing time [days]

C1K0-1.0C1K15-1.0C1K20-1.0C1K25-1.0C1K30-1.0C1K35-1.0C1K40-1.0

Fig. 11. Flexural tensile strength for the mortar batch made with CEM I cement and 1.0 mm maximum sand grain size.

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 355

formation due to the reaction between the gypsum and the cal- cium aluminates present in the cement. The shrinkage of the ce- ment paste, which is defined as the volume reduction of cementitious materials by hydration, is, in this case, countera cted by the expansion of gypsum.

It is generally accepted that shrinkage cracking has significantinfluence on the values for Young’s modulus and flexural tensile strength of mineral matrices [63] and is highly depende nt of the paste volume and binder compositi on. Moreove r, the use of expan- sive additives may lead to substantial reduction of autogeneous shrinkage, up to 50%, or even complete elimination of this phe- nomenon for concretes with water/bind er ratio of 0.3 [64].

The Kerysten used to replace ordinary Portland cement in the mortar mixes also acts as an expansive additive, countera cting the shrinkage. Coupled with smaller grain size, it leads to higher values of the flexural tensile strength, compare d to the reference specimen, from as early as 7 days for the mixes with 30%, 35% and 40% Kerysten.

Figs. 10–12 reveal the fact that the gain in flexural tensile strength, of specimens with 30%, 35% and 40% green binder, shows

a steep increase from the age of 3 days until 7 days and almost con- stant values from the age of 14 days.

The type of cement used in the mix proportio n seems to have aslight influence in this behaviour. The mixes with 25–40% Kerysten and 1.0 mm maximum grain size of sand exhibit almost similar values of the flexural tensile strength from the age of 14 days until the end of the considered time interval. On the other hand, the type of sand considered for the mix proportions leads to a monotonic in- crease in the values of the flexural tensile strength, at the ages of 14, 21 and 28 days, with the increase in the percentage of Kerysten.

4.2.2. Compressive strength The compressive strength was determined on the resulted half

prisms broken in flexure. The resulted parts were visually exam- ined for the presence of cracks before being subjected to the uniax- ial compression test according to code specifications [52]. For each mix proportio n, the presented results were obtained by averaging 12 values determined from the 6 prisms cast and broken in flexure.After the averaging, the each of the determined 12 values was checked to fit within ± one standard deviation from the average va-

4.11 4.

86 5.26 5.

68 6.24 6.33 6.41

3.63 4.

03 4.24 4.

92

6.11

7.65 8.

15

3.04

3.92 4.

27

5.9

8.76

9.85 10

.07

2.97 3.09 3.

58

7.12

9.08 9.

36 9.52

2.68 3.

38 3.59

7.25

8.52

9.45 9.

77

2.88

3.9 4.12

8.63 9.

27 9.89 10

.51

3.22 3.

72

5.17

8.27

9.64 10

.01 11

.17

0

2

4

6

8

10

12

14

1 2 3 7 14 21 28

Flex

ural

tens

ile s

tren

gth

[MPa

]

Curing time [days]

C2K0-1.0C2K15-1.0C2K20-1.0C2K25-1.0C2K30-1.0C2K35-1.0C2K40-1.0

Fig. 12. Flexural tensile strength for the mortar batch made with CEM II cement and 1.0 mm maximum sand grain size.

27.5

9

37.9

4

47.0

9

55.5

6 63.6

6

67.2

5

76.0

6

31.1

5

32.7

2

36.3

6

37.1

6

40.3

6

41.3

5

41.9

15.1

3 19.4

25.3

9 32.0

3

35.4

4

36.0

5

37.3

3

11.6

2 17.5

8 23.9

2 30.0

7 35.5

9

35.6

3

37.2

5

10.2

8

20.5

7

22.5

2

25.8

1

35.6

1 40.3

4

42.1

5

10.1

2 16.1

2 22.1

9 28.7

8 33.9

8 41.4

6

43.2

2

9.91 14

.5 18.2

2 26.1

8

35.2

2 40.8

8

43.0

1

0

10

20

30

40

50

60

70

80

90

1 2 3 7 14 21 28

Com

pres

sive

str

engt

h [M

Pa]

Curing time [days]

C1K0-0.3C1K15-0.3C1K20-0.3C1K25-0.3C1K30-0.3C1K35-0.3C1K40-0.3

Fig. 13. Compressive strength for the mortar batch made with CEM I cement and 0.3 mm maximum sand grain size.

356 I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361

lue. If one measurement would not fit it would have been automat- ically discarded. However, all the determined values for each mix proportion shown in Table 1 were close to one another without any scattering of the results.

The values of the compress ive strength for the different mix proportions made with CEM I cement and finer sand are shown in Fig. 13 . It can be observed that all specimens made with Kery- sten exhibited much lower values for the compressive strength than the reference mix. Replacing the Portland cement with the gypsum based mineral binder seems to have a significant influenceon the strength of mortar. Similar results have been reported in the scientific literature [65]. It can be observed that even though dur- ing the first 7 days there is a general tendency of decreasing the compressive strength with the increase in the percentage of Kerysten, from the 14th day onward the results do not seem to be so significantly influenced by the percentage of the new mineral binder in the mix proportion. The obtained results, with the excep- tion of the reference mix, showed variations between 13% and 15%. This is an encouraging phenomeno n as the possibility of using

large percentages of calcium sulphate based mineral binder could have net benefits for the environment [19,21,24].

The results obtained for the second batch, made with CEM I ce- ment but with coarser sand, are shown in Fig. 14 . The trend is very much similar with the results obtained for the mix proportions made with finer sand. With the exception of the C1K15-1.0 case for which the obtained values were between 9% and 15% than C1K15-0. 3 case, for all other mix proportions the differences ran- ged from �2% to 7%. This observation is valid for curing periods of 14, 21 and 28 days. As with previous results, the compressive strength of specimens made with Kerysten does not seem to be sig- nificantly influences by the percentage used in the mix proportio n. The variation interval is 8–10%, better than before. This could be due to the coarser sand which gives better stability and less scat- tering of the results [2].

The third batch of mortars was made with CEM II cement and coarser sand. The values of the compress ive strength are shown in Fig. 15 . Since CEM II cement contains less Portland cement, the values for the compressive strength are lower than in the other

25.5

9

38.1

4

38.8

2 47.0

6 54.8

1 63.1

3

77.6

6

13.9 17

.51 25

.12 31

.96

34.9

9

35.5 38

.43

16.3

7

17.1

5

20.7

4

32.8

8

36.5

1

37.9

1

39.6

2

15

20.0

1 25.6

6 33.0

4

36.5

1

38.8

5

39.5

8

10.4

7

13.5

1

22.9

2

34.1

4

35.8

8 39.8

9

41.1

5

9.91 13

.19

23.1

2 29.4

1 36.8

9 41.9

5

42.5

8

11.9

8

14.5

21.1

7

35.1

5

38.2

3

41.0

2

42.2

7

0

10

20

30

40

50

60

70

80

90

1 2 3 7 14 21 28

Com

pres

sive

str

engt

h [M

Pa]

Curing time [days]

C1K0-1.0C1K15-1.0C1K20-1.0C1K25-1.0C1K30-1.0C1K35-1.0C1K40-1.0

Fig. 14. Compressive strength for the mortar batch made with CEM I cement and 1.0 mm maximum sand grain size.

19.9

33.0

5

44.5

2

47.8 51

.43 59

.62 66

.19

13.8

19.6

6 27.2

9

37.3

6

38.8

39.3

4

41.2

6

13.8

1 17.9

1 25.3

6

35.0

2

35.7

2

35.8

35.8

1

10.4

3 16.8

6

26.0

4

26.4

1

35.8

4

36.2

2

36.7

7

9.81

15.2

3 23.5

5

25.1

7 31.1

3 36.1

6

37.0

1

9.78 13

.49

14.2

5

24.1

2 29.3

9 33.9

2

37.2

7

9.73 13

.58

15.0

1 22.2

8 28.2

7 33.8

5

370

10

20

30

40

50

60

70

80

90

1 2 3 7 14 21 28

Com

pres

sive

str

engt

h [M

Pa]

Curing time [days]

C2K0-1.0C2K15-1.0C2K20-1.0C2K25-1.0C2K30-1.0C2K35-1.0C2K40-1.0

Fig. 15. Compressive strength for the mortar batch made with CEM II cement and 1.0 mm maximum sand grain size.

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 357

two cases. A similar trend related to the evolution of the compres- sive strength was observed in this case, as well. However , in this case, the results obtained for the mix proportions with Kerysten began to stabilize at a later age, 21 days, than in previous cases, 14 days. The strength gain seems to be related, in this case, to the amount of CEM II cement – the higher the percentage, the fas- ter the specimens exhibit higher strength.

Still, when it comes to the very early stages, that is 24 and 48 h, the compressive strength seems to be governed more by the pres- ence of Kerysten than by the type of cement. The values are slightly lower for CEM II mortars but not by much, 1.3–9%. The type of sand also influences the values of the compressive strength as finer sand leads to higher densities of the prisms, Table 2, and consequently to slightly higher values for the compressive strength [57,58,61].

4.3. Elastic properties

Nowadays the emphasis is more and more on the utilization of cementitious materials to produce sustainable cements [66,67].Their efficient utilization [68–70] has captured the interest of the

scientific community due to their net benefits for the environment. However , there is still little information on their influence upon the elastic properties of the mineral matrix.

4.3.1. Modulus of elasticity The modulus of elasticity, together with strength characteris-

tics, is a key material property in design practice. Its assessment, especiall y for cement based materials, is quite difficult to be pre- dicted because it is influenced by the properties and quantities of the constituent parts in the mix proportion [71,72].

The experimental ly obtained values are presente d in Table 3 forall mix proportions considered in the research. As expected , the modulus of elasticity increases from the age of 21 days until the age of 28 days for all mixtures by 5–20%. The most significant in- crease rates, >13%, were observed for the control specimens made with CEM I cement and for all mix proportions containing more than 30% Kerysten.

Substituti ng the Portland cement by the new calcium sulphate based mineral binder results in a drop in the values of the modulus of elasticity by an average value of 10% at the age of 21 days. For

Table 3Elastic properties of mortars (average values determin ed on five cylinders).

Specimen Modulus of elasticity experimental values (GPa)

Modulus of elasticity ACI363R-92 Eq. a(GPa)

Modulus of elasticity BS8110-Part2 Eq. b (GPa)

Modulus of elasticity ACI318M-05 Eq. c (GPa)

Poisson’s coefficientexperimental values

21 Days 28 Days 21 Days 28 Days 21 Days 28 Days 21 Days 28 Days 21 Days 28 Days

C1K0-0.3 29.33 34.79 34.13 35.85 32.76 35.21 38.54 40.99 0.202 0.199 C1K15-0.3 25.11 27.48 28.25 28.39 28.16 28.38 30.22 30.42 0.202 0.198 C1K20-0.3 25.78 27.21 26.83 27.18 26.90 27.47 28.22 28.72 0.203 0.199 C1K25-0.3 26.23 28.02 26.72 27.16 26.73 27.45 28.05 28.69 0.202 0.199 C1K30-0.3 26.45 30.98 27.99 28.45 27.70 28.43 29.85 30.51 0.201 0.201 C1K35-0.3 26.52 31.15 28.28 28.73 27.94 28.64 30.26 30.90 0.203 0.202 C1K40-0.3 26.68 31.55 28.13 28.67 27.75 28.60 30.05 30.82 0.205 0.201 C1K0-1.0 28.15 32.55 33.28 36.16 31.54 35.53 37.34 41.42 0.201 0.199 C1K15-1.0 24.41 25.89 26.68 27.48 26.42 27.69 28.00 29.14 0.201 0.199 C1K20-1.0 25.46 25.74 27.34 27.80 27.20 27.92 28.94 29.58 0.202 0.2 C1K25-1.0 25.47 29.68 27.59 27.79 27.61 27.92 29.29 29.57 0.204 0.201 C1K30-1.0 25.11 31.24 27.87 28.20 27.71 28.23 29.68 30.15 0.204 0.202 C1K35-1.0 25.34 31.12 28.40 28.56 28.26 28.52 30.44 30.67 0.206 0.201 C1K40-1.0 25.98 31.27 28.16 28.49 27.95 28.45 30.10 30.56 0.206 0.201 C2K0-1.0 25.93 28.28 32.54 33.91 31.26 33.24 36.29 38.24 0.201 0.198 C2K15-1.0 23.99 26.12 27.72 28.23 27.46 28.25 29.48 30.19 0.2 0.198 C2K20-1.0 23.52 25.72 26.76 26.77 27.16 27.16 28.12 28.13 0.202 0.199 C2K25-1.0 22.91 26.35 26.88 27.03 27.11 27.35 28.29 28.50 0.203 0.201 C2K30-1.0 23.68 27.69 26.86 27.10 27.02 27.40 28.26 28.59 0.205 0.2 C2K35-1.0 23.54 28.01 26.24 27.17 25.97 27.45 27.37 28.69 0.206 0.2 C2K40-1.0 23.72 27.89 26.22 27.09 26.00 27.40 27.34 28.59 0.206 0.203

a Ec ¼ 3:32ffiffiffiffif 0c

pþ 6:9 where f 0c is the compressive strength in MPa; valid for 21 MPa 6 f 0c 6 83 MPa.

b Ec;t ¼ Ec;28ð0:4þ 0:6 fcu;t

fcu;28Þ; for t P 3 days and Ec;28 ¼ K0 þ 0:2f cu;28 where K0 is a constant related to the modulus of elasticity of the aggregate, taken as 20 GPa for normal-

weight concrete; fcu;28 is the characteristic cube strength at 28 days, in MPa; fcu;t is the characteristic cube strength at any given age, provided the sample is at least 3 days old. c Ec ¼ 4:7

ffiffiffiffif 0c

pwhere f 0c is the compressive strength in MPa.

358 I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361

the specimens made with CEM I cement, the decrease in the values of modulus of elasticity becomes less than 10% for high volumes of Kerysten of 35% and 40%, respectively . On the other hand, the mix- tures made with CEM II cement showed less scattering of the re- sults for the modulus of elasticity , the average difference being 9%. The decisive influencing factor, given the mix proportions shown in Table 1, seems to be the composition of the cementitious paste, as previously reported in the literature [73–75].

The differenc e between the values of the modulus of elasticity of the control specimen and the other mortars increases at the age of 28 days, reaching a value of 15.5% for the batch made with CEM I ce- ment and finer sand. However, it can be observed that increasing the volume of Kerysten in the mix proportion the values become closer to that of the reference mix. With coarser sand, however , the aver- age difference stays almost the same, namely 10.42% compare d to 10.14% for early ages. Again, for high volumes of Kerysten, e.g. 30–40%, the difference drops at a mere 4% between the modulus of elas- ticity of the reference mix and the rest of the considered cases.

The mix proportions made with CEM II cement seemed to be less sensitive to the variation of cement percentage in terms of the values for the modulus of elasticity. The average difference be- tween the reference mix and the other mixes was 4.6% slightly higher for the specimens with smaller percentages of Kerysten .When the volume of calcium sulphate mineral binder increased beyond 30% the obtained moduli of elasticity were 0.95–2% lower than the reference mix. The difference in behaviour compared to the other two batches made with CEM I could be attributed also to the presence of other mineral admixtures in the CEM II cement. Even though further research should be conducted in this direc- tion, the obtained results are encourag ing and lead to the conclu- sion that the new environm entally sustainable mineral binder could be successfully used as replacemen t for the Portland cement.

Since there are no specially derived equations for the predictio nof the modulus of elasticity of mortars, a few of the currently avail- able equation s for concrete have been used in order to check whether they are suitable to use in this particular case. The values

are also presented in Table 3 for the equations proposed in ACI 363R-92 [76] report and BS 8110-85 Part 2 [77] and ACI 318M- 05 [78] codes. Previous research works have shown good agree- ment between the equation proposed in ACI 363R-92 [76] andthe experimentally obtained results for sprayed concrete on simi- lar sized specimens with the present work [54]. The other two equation s, although derived for concrete , lead also to good approx- imations of the experime ntal results.

However , equation s used in Table 3 should not be applied for any given case as they were not derived for mortars but for con- crete. They can be used to obtain a quick estimation of the results but should not substitute the latter. Further research work should be conducted in this filed in order to propose a similar equation that can be used to estimate the modulus of elasticity of mortars.

Fig. 16 presents the relationship between the compress ive strength of mortars, at the age of 21 days, and the corresponding modulus of elasticity . The three lines represent the governing equation s mentioned above. Even though the equation proposed in the British Standard allows for the evaluation of the modulus of elasticity at different ages [77] it overestimat es the results in this particular case. However, it can be observed that [77] givesthe best estimation of the results even though it was develope dfor concrete and not mortar.

On the other hand, at the age of 28 days, Fig. 17 , both [76] and[77] estimate the values of the modulus of elasticity quite accu- rately. However, a clear trend is difficult to establish given the nar- row range of compressive strengths presented here.

4.3.2. Poisson’s ratio The obtained values for the Poisson’s ratio are presented in

Table 3. It can be observed that there are no significant variations in the values of the ratio among the mix proportions . At the age of 28 days the values are slightly lower than those at 21 days. This could be attributed to the increased stiffness of the cylinders due to the higher values for the modulus of elasticity. The obtained values of the Poisson’s ratio at the age of 28 days are in line with

20

22

24

26

28

30

32

34

36

38

40

30 35 40 45 50 55 60 65 70

Mod

ulus

of E

last

icity

[GPa

]

Compressive Strength [MPa]

ACI 363R-92BS 8110-85 -Part 2ACI 318M-05C1-0.3C1-1.0C2-1.0

Fig. 16. Relationship between the compressive strength and modulus of elasticity at the age of 21 days.

20

22

24

26

28

30

32

34

36

38

40

42

35 40 45 50 55 60 65 70 75 80

Mod

ulus

of E

last

icity

[GPa

]

Compressive Strength [MPa]

ACI 363R-92BS 8110-85 -Part 2ACI 318M-05C1-0.3C1-1.0C2-1.0

Fig. 17. Relationship between the compressive strength and modulus of elasticity at the age of 28 days.

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 359

the predicted values for high-strength concrete given in [76]. Her- ve et al. [72] obtained similar values for Portland cement based mortar made with different types of round siliceous sand having fine, medium and coarse grain sizes.

5. Conclusions

The paper presents the findings of a research work focused on the evaluation of the strength and elastic properties of mortars made with different percentages of a new environmental ly sus- tainable cementiti ous material. The new gypsum-bas ed mineral binder is obtained from industria l by-products , most of them being disposed in landfills. The construction industry has to look for alternatives to Portland cement if the concept of sustainability was to be fully implemented [67].

Based on the obtained results, the following conclusions can be drawn:

(1) The higher the percentage of Kerysten in the mix proportio n, the lower the values of the density for the correspondi ng mortar. The density at the age of 28 days for the mixes made

with 25%, 30% and 35% gypsum based binder is almost the same and it does not depend on the type of cement used. Using high percentages of Kerysten in a mix proportion has benefits both in terms of density as well as in an increased stability of the geometri cal dimensions of the specimens. The type of cement has only a slight influence on the results, with larger values measured for the width and thickness of the prisms made with CEM II cement. On the other hand, the type of sand does seem to have a more significant influ-ence on the dimensional stability, especially for the mixes with higher percentages of calcium sulphate based mineral binder 25–40%. The larger the grain size diameter was, the lower the change in the width, thickness and length espe- cially for the mixes made with 25% and 40% Kerysten.

(2) The high values for the flexural tensile strength obtained for all mix proportions made with 30%, 35% and 40% Kerysten, irrespective of the type of Portland cement used or maxi- mum grain size of sand, are directly influenced by the pres- ence of the calcium sulphate based binder with its high surface area [38] which contributes to the hardened proper- ties of mineral matrices [62].

360 I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361

(3) All specimens made with Kerysten exhibited much lower values for the compressive strength than the reference mix. Replacing the Portland cement with the gypsum based mineral binder seems to have a significant influence on the strength of mortar. Still, when it comes to the very early stages, that is 24 and 48 h, the compressive strength seems to be governed more by the presence of Kerysten than by the type of cement. The values are slightly lower for CEM II mortars but not by much, 1.3–9% compared to CEM I mor- tars. The type of sand also influences the values of the com- pressive strength.

(4) Substituti ng the Portland cement by the new calcium sul- phate based mineral binder results in a drop in the values of the modulus of elasticity by an average value of 10% at the age of 21 days. For the specimens made with CEM Icement, the decrease in the values of modulus of elasticity becomes less than 10% for high volumes of Kerysten of 35% and 40%, respectively. On the other hand, the mixtures made with CEM II cement showed less scattering of the results for the modulus of elasticity, the average difference being 9%. The differenc e between the values of the modulus of elasticity of the control specimen and the other mortars increases at the age of 28 days, reaching a value of 15.5% for the batch made with CEM I cement and finer sand. How- ever, it can be observed that increasing the volume of Kery- sten in the mix proportion the values become closer to that of the reference mix. The mix proportions made with CEM II cement seemed to be less sensitive to the variation of cement percentage in terms of the values for the modulus of elasticity .

(5) Some of the existing equations for predictin g the values of the elasticity modulus for concrete succeed in quite accu- rately estimating the modulus of elasticity of mortar, as well. However , a clear trend is difficult to establish given the nar- row range of compress ive strengths presented here.

There are still questions that have not been answered in this study but they are currently investiga ted, namely the long-term durability of such mortars as well as their geometric stability at ages longer than 28 days. Although some informat ion is already available in the literature [59,60] on similar gypsum based mortars there still the need for more experimental results before the new mineral binder can be used in structural elements made of con- crete. This goal is envisaged by the authors after supplementar yinvestigatio ns have been conducted at the material level in terms of fresh properties, strength and elastic characterist ics, durability issues of the mineral matrix itself.

Acknowled gements

This paper was supported by the project ‘‘Develop and support multidiscipli nary postdoctora l programs in primordial technical areas of national strategy of the research – development – innova- tion’’ 4D-POSTDOC, contract nr. POSDRU/89/1.5 /S/52603, project co-funded from European Social Fund through Sectorial Opera- tional Program Human Resources 2007–2013.

The authors would also like to thank Carpatcement Holding S.A. for supplying the CEM I type of cement used in this research. Their contribution is greatly appreciated.

References

[1] Elsen J. Microscopy of historic mortars – a review. Cement Concr Res 2006;36:1416–24.

[2] Haach VG, Vasconcelos G, Lourenco PB. Influence of aggregates grading and water/cement ratio in workability and hardened properties of mortars. Constr Build Mater 2011;25:2980–7.

[3] Rao GA. Long-term drying shrinkage of mortar – influence of silica fume and size of fine aggregate. Cement Concr Res 2001;31:171–5.

[4] Atis CD, Kilic A, Sevim UK. Strength and shrinkage properties of mortar containing a nonstandard high-calcium fly ash. Cement Concr Res 2004;34:99–102.

[5] Aspin J. An improvement in the modes of producing an artificial stone. In: Patent ed., 5022:1824, UK, 1824.

[6] Chen JJ, Kwan AKH. Superfine cement for improving packing density, rheology and strength of cement paste. Cement Concr Comp 2012;34:1–10.

[7] Chang PK. An approach to optimizing mix design for properties of high- performance concrete. Cement Concr Res 2004;34:623–9.

[8] Long GC, Wang XY, Xie YJ. Very-high-performance concrete with ultrafinepowders. Cement Concr Res 2002;32:601–5.

[9] Bentur A, Mitchell D. Material performance lessons. Cement Concr Res 2008;38:259–72.

[10] Environmental Council of Concrete Organizations, Cement, Concrete and the Environment, Fact Sheet. In: Environmental Council of Concrete Organizations; 2012.

[11] Mehta PJMMPK. Concrete: microstructure, properties and materials. 3rd ed. Mc-Graw-Hill; 2005.

[12] Arikan M. Feasibility analysis of manufacturing high performance – ecological – cement in Turkey. Build Environ 2004;39:1125–30.

[13] Koroneos C, Dompros A. Environmental assessment of brick production in Greece. Build Environ 2007;42:2114–23.

[14] Blom I, Itard L, Meijer A. Environmental impact of dwellings in use: maintenance of facade components. Build Environ 2010;45:2526–38.

[15] Rehan R, Nehdi M. Carbon dioxide emissions and climate change: policy implications for the cement industry. Environ Sci Policy 2005;8:105–14.

[16] Spaargaren G, Mol APJ. Greening global consumption: redefining politics and authority. Global Environ Change 2008;18:350–9.

[17] Ali MB, Saidur R, Hossain MS. A review on emission analysis in cement industries. Renew Sust Energy Rev 2011;15:2252–61.

[18] Habert G, Billard C, Rossi P, Chen C, Roussel N. Cement production technology improvement compared to factor 4 objectives. Cement Concr Res 2010;40:820–6.

[19] Meyer C. The greening of the concrete industry. Cement Concr Comp 2009;31:601–5.

[20] Sorrentino F. Chemistry and engineering of the production process: state of the art. Cement Concr Res 2011;41:616–23.

[21] Scrivener KL, Kirkpatrick RJ. Innovation in use and research on cementitious material. Cement Concr Res 2008;38:128–36.

[22] Schneider M, Romer M, Tschudin M, Bolio H. Sustainable cement production- present and future. Cement Concr Res 2011;41:642–50.

[23] Shi CJ, Jimenez AF, Palomo A. New cements for the 21st century: the pursuit of an alternative to Portland cement. Cement Concr Res 2011;41:750–63.

[24] Juenger MCG, Winnefeld F, Provis JL, Ideker JH. Advances in alternative cementitious binders. Cement Concr Res 2011;41:1232–43.

[25] Lee TC. Recycling of municipal incinerator fly-ash slag and semiconductor waste sludge as admixtures in cement mortar. Constr Build Mater 2009;23:3305–11.

[26] Chindaprasirt P, Homwuttiwong S, Sirivivatnanon V. Influence of fly ash fineness on strength, drying shrinkage and sulfate resistance of blended cement mortar. Cement Concr Res 2004;34:1087–92.

[27] Karahan O, Atis CD. The durability properties of polypropylene fiber reinforced fly ash concrete. Mater Des 2011;32:1044–9.

[28] O’Connell M, McNally C, Richardson MG. Performance of concrete incorporating GGBS in aggressive wastewater environments. Constr Build Mater 2012;27:368–74.

[29] Rao GA. Development of strength with age of mortars containing silica fume. Cement Concr Res 2001;31:1141–6.

[30] Sezer GI. Compressive strength and sulfate resistance of limestone and/or silica fume mortars. Constr Build Mater 2012;26:613–8.

[31] Bhanja S, Sengupta B. Influence of silica fume on the tensile strength of concrete. Cement Concr Res 2005;35:743–7.

[32] Vikan H, Justnes H. Rheology of cementitious paste with silica fume or limestone. Cement Concr Res 2007;37:1512–7.

[33] Giner VT, Ivorra S, Baeza FJ, Zornoza E, Ferrer B. Silica fume admixture effect on the dynamic properties of concrete. Constr Build Mater 2011;25:3272–7.

[34] Yang KH, Hwang HZ, Kim SY, Song JK. Development of a cementless mortar using hwangtoh binder. Build Environ 2007;42:3717–25.

[35] Oh D, Song D, Go S. Strength properties of activated hwangtoh mortars. Adv Mater Res–Switz 2011;168–170:709–15.

[36] Smadi MM, Haddad RH, Akour AM. Potential use of phosphogypsum in concrete. Cement Concr Res 1999;29:1419–25.

[37] Singh M. Treating waste phosphogypsum for cement and plaster manufacture. Cement Concr Res 2002;32:1033–8.

[38] KANDCO. Process for the industrial manufacture of compositions based on anhydrous calcium sulphate in the B-Anhydrite III’ form and corresponding compositions and binders, Organisation Mondiale de la Propriete Intellectuelle. In: O.M.d.l.P. Intellectuelle editor, WO 2010/003827 A1, France; 2010.

[39] Singh M, Garg M. Making of anhydrite cement from waste gypsum. Cement Concr Res 2000;30:571–7.

[40] O’Rourke B, McNally C, Richardson MG. Development of calcium sulfate-ggbs- Portland cement binders. Constr Build Mater 2009;23:340–6.

[41] Taranu G, Budescu M, Plesu R, TIO. New building system made of glass fiberreinforced mineral matrix composites. Adv Mater Res 2013;649:25–8.

I.-O. Toma et al. / Construction and Building Materials 43 (2013) 348–361 361

[42] SR EN 197-1: Cement. Part I: Composition, specifications and conformity criteria for normal use cements. Romanian Standard Association; 2002 [in Romanian].

[43] CEMBUREAU. World statistic review – 1996–2006. In: E.C. Association; 2008. [44] Toma I-O, Taranu G, Toma A-M, Budescu M. Influence of cement and sand type

on the strength characteristics of mortars with various contents of green binder. Procedia Eng 2011;21:196–203.

[45] Stefanidou M, Papayianni I. The role of aggregates on the structure and properties of lime mortars. Cement Concr Comp 2005;27:914–9.

[46] Diamond S, Kjellsen KO. Scanning electron microscopic investigations of fresh mortars: well-defined water-filled layers adjacent to sand grains. Cement Concr Res 2008;38:530–7.

[47] Cortes DD, Kim HK, Palomino AM, Santamarina JC. Rheological and mechanical properties of mortars prepared with natural and manufactured sands. Cement Concr Res 2008;38:1142–7.

[48] Garg M, Minocha AK, Jain N. Environment hazard mitigation of waste gypsum and chalk: use in construction materials. Constr Build Mater 2011;25:944–9.

[49] Taher MA. Influence of thermally treated phosphogypsum on the properties of Portland slag cement. Resour Conserv Recy 2007;52:28–38.

[50] Martinez-Aguilar OA, Castro-Borges P, Escalante-Garcia JI. Hydraulic binders of fluorgypsum–portland cement and blast furnace slag, stability and mechanical properties. Constr Build Mater 2010;24:631–9.

[51] Toma IO, Covatariu D, Toma AM, Taranu G, Budescu M. Greening of concrete industry by incorporating gypsum-based industrial wastes as supplementary cementitious materials. Adv Mater Res 2013;649:246–9.

[52] SR EN 196-1. Methods for testing cement – Part 1: determination of strength. Romanian Standard Association; 2006 [in Romanian].

[53] ASTM C469/C469M. Standard test method for static modulus of elasticity and poisson’s ratio of concrete in compression. American Society for Testing and Materials; 2010.

[54] Goodier CI, Austin SA, Robins PJ. Low-volume wet-process sprayed concrete: hardened properties. Mater Struct 2008;41:99–111.

[55] Papanicolaou CG, Triantafillou TC, Karlos K, Papathanasiou M. Textile- reinforced mortar (TRM) versus FRP as strengthening material of URM walls: in-plane cyclic loading. Mater Struct 2007;40:1081–97.

[56] Kovler K, Roussel N. Properties of fresh and hardened concrete. Cement Concr Res 2011;41:775–92.

[57] Le TT, Austin SA, Lim S, Buswell RA, Law R, Gibb AGF, et al. Hardened properties of high-performance printing concrete. Cement Concr Res 2012;42:558–66.

[58] Grassl P, Wong HS, Buenfeld NR. Influence of aggregate size and volume fraction on shrinkage induced micro-cracking of concrete and mortar. Cement Concr Res 2010;40:85–93.

[59] Tovar-Rodriguez B, Barra M, Pialarissi S, Aponte D, Vasquez E. Expansion of mortars with gypsum contaminated fine recycled aggregates. Constr Build Mater 2013;38:1211–20.

[60] Mridul G, Pundir A. Comprehensive study of fly ash binder developed with flyash – alpha gypsum plaster – Portland cement. Constr Build Mater 2012;37: 758–65.

[61] Kang SH, Ahn TH, Kim DJ. Effect of grain size on the mechanical properties and crack formation of HPFRCC containing deformed steel fibers. Cement Concr Res 2012;42:710–20.

[62] Bentz DP, Ferraris CF, Galler MA, Hansen AS, Guynn JM. Influence of particle size distributions on yield stress and viscosity of cement–fly ash pastes. Cement Concr Res 2012;42:404–9.

[63] Leemann A, Lura P, Loser R. Shrinkage and creep of SCC – the influence of paste volume and binder composition. Constr Build Mater 2011;25:2283–9.

[64] Meddah MS, Suzuki M, Sato R. Influence of a combination of expansive and shrinkage-reducing admixture on autogenous deformation and self-stress of silica fume high-performance concrete. Constr Build Mater 2011;25:239–50.

[65] Yao Y, Yang DW, Zeng JL, Sun LX, Li WJ. Influences of fly ash and fluorgypsumon the hydration heat and compressive strength of cement. J Therm Anal Calorim 2011;106:869–74.

[66] Zhang TS, Yu QJ, Wei JX, Zhang PP. Efficient utilization of cementitious materials to produce sustainable blended cement. Cement Concr Comp 2012;34:692–9.

[67] Gartner EM, Macphee DE. A physico-chemical basis for novel cementitious binders. Cement Concr Res 2011;41:736–49.

[68] Tosun K, Felekoglu B, Baradan B, Altun IA. Effects of limestone replacement ratio on the sulfate resistance of Portland limestone cement mortars exposed to extraordinary high sulfate concentrations. Constr Build Mater 2009;23:2534–44.

[69] Felekoglu B, Turkel S, Kalyoncu H. Optimization of fineness to maximize the strength activity of high-calcium ground fly ash–Portland cement composites. Constr Build Mater 2009;23:2053–61.

[70] Bellmann F, Stark J. Activation of blast furnace slag by a new method. Cement Concr Res 2009;39:644–50.

[71] Goncalves JP, Tavares LM, Toledo RD, Fairbain EMR, Cunha ER. Comparison of natural and manufactured fine aggregates in cement mortars. Cement Concr Res 2007;37:924–32.

[72] Herve E, Care S, Seguin JP. Influence of the porosity gradient in cement paste matrix on the mechanical behavior of mortar. Cement Concr Res 2010;40:1060–71.

[73] Maia L, Azenha M, Faria R, Figueiras J. Influence of the cementitious paste composition on the E-modulus and heat of hydration evolutions. Cement Concr Res 2011;41:799–807.

[74] Azenha M, Faria R, Magalhaes F, Ramos L, Cunha A. Measurement of the E- modulus of cement pastes and mortars since casting, using a vibration based technique. Mater Struct 2012;45:81–92.

[75] Maia L, Azenha M, Geiker M, Figueiras J. E-modulus evolution and its relation to solids formation of pastes from commercial cements. Cement Concr Res 2012;42:928–36.

[76] ACI 363R-92. State-of-the-art report on high strength concrete. American Concrete Institute Committee 363; 1992 [reapproved in 1997].

[77] BS 8110-85. Structural use of concrete. Part 2: Code of practice for special circumstances. British Standard; 1985.

[78] ACI 318M-05. Building code requirements for structural concrete and commentary. American Concrete Institute Committe 318; 2005.