Effects of Blended-Cement Paste Chemical Composition Changes on Some Strength Gains of Blended-Mortars

Effects of chemical compositions changes of blended-cement pastes (BCPCCC) on some strength gains of blended cement mortars (BCMSG) were monitored in order to gain a better understanding for developments of hydration and strength of blended cements. Blended cements (BC) were prepared by blending of 5% gypsum and 6%, 20%, 21%, and 35% marble powder (MP) or 6%, 20%, 21%, and 35% brick powder (BP) for CEMI42.5N cement clinker and grinding these portions in ball mill at 30 (min). Pastes and mortars, containing the MP-BC and the BP-BC and the reference cement (RC) and tap water and standard mortar sand, were also mixed and they were cured within water until testing. Experiments included chemical compositions of pastes and compressive strengths (CS) and flexural strengths (FS) of mortars were determined at 7th-day, 28th-day, and 90th-day according to TS EN 196-2 and TS EN 196-1 present standards. Experimental results indicated that ups and downs of silica oxide (SiO2), sodium oxide (Na2O), and alkali at MP-BCPCC and continuously rising movement of silica oxide (SiO2) at BP-BCPCC positively influenced CS and FS of blended cement mortars (BCM) in comparison with reference mortars (RM) at whole cure days as MP up to 6% or BP up to 35% was blended for cement.


Introduction
Increasing of marble and brick manufacturing and lack of suitable landfills for wastes and ineffective purification systems for marble waste (MW) and brick waste (BW) have resulted in a move to raise both volume of these wastes and loss of the useful minerals. MW and BW occur, respectively, at 2,592,000 (tons) and 3,800,000 (tons) in Turkey every year as like whole world [1].
Activation of cement, such byproducts as fly ash (FA), silica fume (SF), MP, and BP, has been appealed by researchers for more than ten decades. Several researches have presented benefits of byproducts replaced by sand in mortar and concrete such properties as strength and workability [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17]. Compatibility of byproducts for cement and mortar is assessed according to standard methods and technological studies (XRD, SEM, EDS, and TEM) and equations suggested by researchers to compute C 3 S, C 2 S, C 3 A, and C 4 AF compounds in cement [18][19][20][21][22][23][24][25]. Suggested standards, technological studies, and equations are inadequate to explain effects of blended-cement paste chemical composition changes on some strength gains of blended-mortars. Compounds linked strength gain at hydrated cement pastes are C 3 S, C 2 S, C 3 A, and C 4 AF are composed of CaO, SiO 2 , Al 2 O 3 , and Fe 2 O 3 known major oxides. Basically, these major oxides underlie the main reason for which cements show strength gain quickly (false strength) or strength gain slowly. However, CaO and SiO 2 of OPC pastes at 28 d were, respectively, over 9% greater and over 6% less than that of OPC at 7 d. Due to these changes of CaO and SiO 2 , OPC mortars at 28th-d showed over 25% greater quick strength gain in comparison with OPC mortars at 7th-d (Tables 3, 4, 5, and 6). This strength gain tendency of cement was considered to be able to change chemical compositions in which cement paste is hydrated by C 3 S, C 2 S, C 3 A and C 4 AF.
Therefore, this research was planned to monitor the effects of blended-cement paste chemical composition changes on some strength gains of blended mortars. This 2 The Scientific World Journal MW or BW can also be partially added for CEMI42.5 cement clinker and gypsum in order to positively improve chemical compositions of cement pastes and strength gains of blended mortars according to physical and activation properties of MP and BP (Table 1) [1].
Mineralogical properties of MP measured by XRD are given in Figure 1. MP mainly contains middle pure levelcalcite (CaCO 3 ). There are found some other minor minerals,  for instance, quartz, dolomite, hematite, and magnetite in XRD pattern of MP ( Figure 1) [1]. Figure 2 showed that most of the MP particles consisted of near-spherical and spheres were quite smooth. It is also apparent that MP is composed of various sized particles ranging from several to dozens of ultrafine micrometers. Additionally, microstructure of MP has slightly little spherical and amorphous hollow ( Figure 2). There are mainly Ca, C, and O element peaks in EDS of MP similar to its chemical composition ( Figure 3). As measured through absorbance spectra, results show that MP turned out big and little annular and sphere particle [1]. Figure 4 shows X-ray diffraction of BP. Characteristic mineralogical properties of BP consist of silica oxide (SiO 2 ), aluminum oxide (Al 2 O 3 ), magnesium orthosilicate (2[Mg 0,96 Fe 0,04 ]O⋅SiO 2 ), and magnesium silicate (Mg 2 SiC 4 ) ( Figure 4) [1].
SEM images of BP and EDS element counts of BP are shown in Figures 5 and 6, respectively. SEM observations of BP shown in Figure 5 indicated amorphous structure of BP. As measured through absorbance spectra, results show that BP turned out big and little annular and flat particle. Grains of BP exhibit irregular shapes with flat surfaces covered by small debris which are similar cement grains in paste and mortar. BP samples also contain immature needle-like products. It is apparent that populations of these products in the BP are significantly higher than that of the MP and these products are more compacted, as well ( Figure 5) [1]. On the contrary to MP observations, BP includes plenty of rich silicate, aluminate, calcite, and hematite minerals which

Sample Preparation.
Difference of this research than others is that sample preparation consists of three stages. First stage is preparation of blended cements. Blended cements were prepared by blending of MW or BW for CEMI42.5N cement clinker at the percents of 6, 20, 21, and 35 and adding gypsum up to 5%, and these mixture proportions were grinded in ball mill together at 30 min. Blending ratios of MW, BW, and gypsum for cement clinker were chosen according to TS EN 196-1 in order to be appropriate for this up-to-date standard. Codes of cement samples and mixture proportions and grinding time are given in Table 2 [1]. Second stage is preparation of cement pastes containing MP-blended cements or BP-blended cements or reference cement in order to measure chemical composition changes at 7th-d, 28th-d, and 90th-d. A medium planetary mixer was used for every paste mix using this following procedure: (1) add water and cements in bowl and mix them for 90 s at low speed; (2) stop mixer for 15 s to scrape bowl; mix for another 90 s at high speed; (3) cast pastes in cubic mold 50 mm for 4 min totally. Pastes were mixed with water : cement ratio of 1 : 2 [1,26,28].
Last stage deals with preparation of mortars containing MP-blended cements or BP-blended cements or reference  7) cast each mortar mixes in prism mould 40 × 40 × 160 mm as three layers; (8) collapse each layer 60 times [1,28]. Mortars were mixed with water : cement : sand ratio of 1 : 2 : 6. Codes of blended cement mortars, reference cement mortars, and mixture proportions of mortars for one standard 3-gang mould are presented in

Chemical Compositions Experiment (with Wet Analysis).
Blended-cement pastes and reference cement pastes were dried in laboratory and grinded at the end of curing days of the 7-day, 28-day, and 90-day ages. TS EN 196-2 standard was known wet chemical analysis was followed to analyze the chemical compositions of the pastes [26]. After 24 h casting in plastic mould in automatic controlled curing cabinet at 98% relative humidity, these pastes were subjected to water curing in the same cure cabinet until testing at 21 ∘ C and 98% relative humidity. For each mix at each age, three samples were analyzed and the average value of 162 samples was taken to be the representative chemical properties.

Compressive Strength (CS) Experiment.
CSs of mortars were determined cubic samples which were used portions of prisms broken in flexural strength experiment at 7 d, 28 d, and 90 d in accordance to standard method [28]. For each mix at each age, twelve samples were tested with an ELE hydraulic testing machine and the average value of 324 samples was taken to be the representative strength.

Flexural Strength (FS) Experiment.
Standard method was followed to measure flexural strength of prism mortar samples 40 × 40 × 160 mm with one-point loading at 7 d, 28 d, and 90 d. After 1 day of casting, samples were demolded and cured in water in automatic controlled curing cabinet at 21 ∘ C and 98% relative humidity until testing. Curing water was refreshed fortnightly [28]. For each mix at each age, six samples were tested with an ELE hydraulic testing machine and the average value of 162 samples was taken to be the descriptive strength.

Influences of the BCP-SEM Image Changes to CS and FS of Blended Cement Mortars (BCM).
Results of SEM observations and EDS graphs of hardened pastes of CEMI42.5MP6P and CEMI42.5NBP20P at 7th-d and 28th-d and 90th-d are presented in Figures 7 and 8. It was observed that in hydration of CEMI42.5MP6P at 7 d, clinker grains were surrounded by radiating fibers of calcium silicate hydrate (C-S-H) resembling pattern of C-S-H of CEMI42.5 cement. Randomly oriented portlandite (CH) crystals and prismatic ettringite crystals were widely dispersed through paste (Figure 7, upper left). However, in CEMI42.5MP6P at the ages of 7 days it was found that marble grains were covered with an amorphous (with respect to C-S-H) layered CH hydration products. Matrix phase is mainly composed of short radicular outgrowths of C-S-Hs around clinker grains and needleshaped ettringite crystals (Figure 7, upper left). Microstructure of hydrated paste of CEMI42.5MP6P at the ages of 28 days was presented by amorphous gel filling spaces between hydrated particles. In pastes of CEMI42.5MP6P, layered accumulations of CH crystals of about 10 m in width are intermingled through paste (Figure 7, middle left). There is a visible densification around marble grains due to partial hydration of marble grains, leading to formation of additional CH. At the ages of 90 days marble grains were well located in matrix and were sunk in a layered CH. Observation of paste based on CEMI42.5MP6P demonstrated that marble grains were turned out amorphous reaction prisms. In CH phase, matrix of CEMI42.5MP6P is found to be richer in calcium than that of reference cement paste. At 90th-day, microstructure of CEMI42.5MP6P was further densified with respect to reference cement paste (Figure 7, lower left). However, compressive strength of CEMI42.5MP6M samples at whole cure periods is notably lower than that of brick powder-blended cement mortar (BP-BCM) and reference cement mortar (RCM). This can be explained by the difference at degree of hydration that occurs in each of two compounds such as CH and C-S-H. A lower degree of hydration occurs in cement paste with 6% MP, where a number of hydration compounds are spotted during SEM imaging, shown in Figure 7. EDS analysis also supported that they were CH, since they exhibited Ca/Si ratios of between 1.76 and 2.8 at 7 d, 28 d, and 90 d. In comparison, hydrates exhibited none traces of carbon at whole periods. Therefore, this low compressive strength exhibited by CEMI42.5MP6M samples may be attributed to increasing of Ca/Si ratio although MP is aiding in densifying microstructure in cement systems by accelerating hydration (Figure 7). For CEMI42.5BP20P at the ages of 7 days it was found that brick grains were covered with a numerous hydration products of C-S-H gel, layered CH, prism ettringite, and little hollow. Matrix phase of CEMI42.5BP20P is mainly composed of short fibrous C-S-H forming around clinker grains and needle-shaped ettringite crystals, and randomly oriented portlandite (CH) crystals are widely dispersed through paste (Figure 8, upper left) CEMI42.5BP20P at the ages of 28 days was presented by amorphous gel filling hollow between hydrated particles; this gave certain stability to paste structure. In CEMI42.5BP20P pastes, layered accumulations of C-S-H crystals of about 3 m in width are intermingled through paste (Figure 8, middle left). There is a visible densification around brick grains due to partial hydration leading to formation of additional C-S-H. At the ages of 90 days, brick grains were well dispersed through matrix. And they penetrated in a layered CH in order to convert CH to fibrous C-S-H forming gel. Observation of paste based on CEMI42.5BP20P demonstrated that brick grains were turned out fibrous reaction prisms. In C-S-H phase, matrix of CEMI42.5BP20P was found to be richer in silicate with respect to reference cement at 90th-day ( Figure 8, lower left). However, compressive strengths of CEMI42.5BP20M samples at whole cure periods were the greatest in these mortars. This can be explained by difference at degree of 8 The Scientific World Journal hydration that occurs in each of one compound as C-S-H. A high degree of hydration occurs in cement paste with 20% BP, where a number of hydration compounds are spotted during SEM imaging, shown in Figure 8. EDS analysis also supported that they were silicate, since they exhibited Si/Ca ratios of about 0.53, 0.63, and 0.96 at 7 d, 28 d, and 90 d, respectively. In comparison, hydrates exhibited none traces of carbon at whole period. Therefore, the highest compressive strength exhibited by CEMI42.5BP20M samples may be attributed to increase of Si/Ca ratio at 90th-day although BP was aiding in densifying microstructure in cement systems by accelerating strength gain slowly (Figure 8).

Influences of the CCC-BCP to CS and FS of Blended
Cement Mortars. 7th-day, 28th-day, and 90th-day LOI and chemical properties of MP-blended cement pastes and BPblended cement pastes and CEMI42.5N cement pastes are presented in Tables 4, 5, and 6, respectively. Table 7 shows 7th-day, 28th-day, and 90th-day CS and FS of MP-blended mortars and BP-blended mortars and CEMI42.5N cement mortars [1]. According to the literature, results showed that mortars containing cement with high LOI were slower strength gains than that of mortars containing cement with low LOI. An increase in LOI of 1% can be anticipated to reduce mortar strength by nearly 4 MPa [20-22, 24, 29-33]. A parallel result was clearly observed in this research due to MP increased LOI in cement and cement pastes, too. As MP was increased up to 35% in cement paste by mass, averages of CaO, MgO, K 2 O, and LOI of MP-blended cement pastes (MP-BCP) at 7 d were, respectively, over 11%, 1%, 9%, and 15% higher with respect to reference paste (RP); averages of SiO 2 , Al 2 O 3 , Fe 2 O 3 , SO 3 , Na 2 O, and alkali of MP-BCP at 7 d were over 20%, 26%, 34%, 29%, 61%, and 46% less than that of RP. Changes of chemical compositions of MP-blended cement pastes (MP-BCPCC) at 7 d caused the decrease of both CS and FS of MP-blended mortars (MP-BM). CS and FS of MP-BM were, respectively, over 23%, 7.5 MPa and 12%, 0.6 MPa less than that of reference mortars (RM) at 7 d, except FS of CEMI42.5MP6M. It was surprisingly detected 0.03 MPa greater FS of CEMI42.5MP6M in comparison with RM although CEMI42.5MP6P oxides changed such as other MPblended cement pastes at 7 d (Tables 3 and 6). Since MP was increased from 6% to 35% (over 5.5 times) in cement pastes, increasing of MP created 30% CS and 24% FS reduction associated with increasing of 9% CaO, 46% SO 3 , 37% K 2 O, and 33% LOI at MP-BM at 7 d. In view of these changes of MP-BCPCC, CS and FS of MP-BM at 7 d were located in relatively wide ranges of 21.8-31.7 MPa and 4.1-5.4 MPa. These values are the lowest at MP-BMs. Increasing of MP in cement pastes caused a delay at strength of MP-BM at later ages, nevertheless, MP showed the most pronounced enhancing effect on strength gain at early ages: for 20% MP mix at 28 d, CS and FS could be improved by as much as 40% and 12% with respect to MP-BM at 7 d, respectively, (Tables 4 and 7).
Similar strength reduction happened in MP-BM at 7 d was monitored at BP-BM. Since BP was augmented up to 35% in cement paste by mass, BP decreased CaO, MgO, Na 2 O, LOI, and alkali over 3%, 37%, 56%, 8%, and 26% lesser and also increased SiO 2 , Al 2 O 3 , Fe 2 O 3 , SO 3 , and K 2 O over 11%, 30%, 21%, 14%, and 88% higher in comparison with RP at 7 d. Changes of chemical compositions of BP-blended cement paste (BP-BCPCC) resulted reducing both CS and FS of BP-blended mortars (BP-BM). CS and FS of BP-BM were, respectively, over 2%, 0.8 MPa and 4%, 0.2 MPa lower than that of RM at 7 d, except CS of CEMI42.5BP6M and both CS and FS of CEMI42.5BP20M (Tables 3 and 6) [20-22, 24, 32, 33]. It was monitored that CEMI42.5BP20M had the greatest CS and FS in these mortars at 7 d. Result infers that BP has got over 9% greater acceleration effect on strength gain with respect to RM at 7 d. Nevertheless, this effect is not as high as blended mortars containing 20% MP at 28 d. As BP was increased from 6% to 35% (over 5.5 times) in cement pastes, the increasing of BP originated 25% CS and 10% FS reduction accompanying with increasing of 69% Al 2 O 3 , 52% Fe 2 O 3 , 94% MgO, 75% SO 3 , 220% K 2 O, and 25% Na 2 O at BP-BM at 7 d. In view of these changes of BP-BCPCC, CS and FS of BP-BM at 7 d were located in relatively wide ranges of 27.7-37.5 MPa and 4.6-6 MPa. These values are the lowest in BP-BM. However, BP showed the least pronounced enhancing effect on strength gain at early ages: for 20% BP mix at 28 d, CS and FS were, respectively, improved by as much as 23% and 15% in comparison with 7 d (Tables 4 and 7).
It was shown in the literature that an increase in alkali of pastes reduced strength gain by nearly 1.5 MPa at 28 d [20-22, 24, 29-33]. Strength reduction was not monitored in