Effect of Methyl cellulose/poly(acrylic acid) Blends on Physico-Mechanical Properties of Portland Cement Pastes

The paper deals with the influence of blend polymers on the physico-mechanical properties of Portland cement pastes including, W/C ratio, setting time, workability, density, porosity and compressive strength. Blend polymers were used based on methyl cellulose (MC) of different molecular weight (4x104 and 8.3x104 g/mol) and poly(acrylic acid) (PAA). The result showed that, the physico-mechanical properties of the cement pastes are dependent on molecular weight of MC. However, the addition of MC/PAA blend into cement, the W/C-ratio, setting time and density decrease while the compressive strength and flow increase. Also blend polymer containing MC (8.3x104 g/ mol) is more effective than that containing MC (4x104 g/mol) keywords: Blend polymer; cement; sustainable, workability; compressive strength.

reinforcements 8 . Polymers can be solutions such as water-soluble polyacrylate, methylcellulose, polyvinyl alcohol and polyacrylamide [9][10][11][12] . In our laboratory, it is of interest to prepare polyacrylate based on acrylic acid and 2-hydroxy ethyl acrylate (2-HEA) with another acrylic monomer with different chemical composition as chemical admixtures for cement pastes, mortar and concrete [13][14][15][16] . Our previous work reported that the addition of polyacrylate to cement pastes, improved the specific characteristics of Portland cements due to to the formation of a long side chain, which results in a polymer film that improves the strength. For example, the effect of copolymer latexes based on 2-HEA with each of styrene (2-HEA/St), 2-hydroxy ethyl methacrylate (2-HEA/2-HEMA) and vinyl ether of ethylene glycol (2-HEA/VEEG) on properties of cement pastes were studied [17][18][19] . The results showed that, the strength of cement pastes increases with increasing the ratio of 2-HEMA, VEEG content in the copolymer latexes and decreasing with increasing St in the latexes. On the other hand, methylcellulose has found to decrease the mechanical properties of concrete because of the increased air entrainment. For example, the addition of methylcellulose (0.2-0.8% by weight of cement) to cement pastes was found to increase the tensile strength by up to 72%, decrease the tensile modulus by up to 57%, decrease the compressive strength by up to 30%. However, the physco-mechanical of cement pastes decreased with increasing the concentration of methyl cellulose 20,21 . Improving the properties of polymers can be achieved by chemical modifications or by a new novel architecture. Another possible approach to improve polymer properties, which has received much attention, is polymer blend since blending is a simple process. Polymer blend is an effective approach to develop new materials exhibiting combinations of properties that cannot be obtained by individual polymers. Negim et al 22 prepared blend polymers based on on methyl cellulose (MC) of different molecular weight (4x10 4 and 8.3x10 4 g/mol) and poly(acrylic acid) (PAA). The viscosity measurements showed that all blends have non-Newtonian shear thinning (pseudoplastic) behavior. These blends have a single glass transition indicating that these blends can form a miscible phase due to the formation of hydrogen bonds between the hydroxyl group of MC and the carboxyl group of PAA. The MC/PAA blends exhibit good mechanical properties, thermal stability, characteristics of a MC-PAA polymer network. The work was further extended to include the application of the blend polymers to modify the properties of cement pastes.

Materials
In our work, we used two samples of methylcellulose (MC): one of them, of molecular weight M 1 = 4x10 4 g/mol, (viscosity of 2% aqueous solution at g=1,7 is 400 cps), the second one of M 2 = 8.3x10 4 g/mol, (viscosity of 2% aqueous solution at g=1,7-1.9 is 4000 cps) purchased from Aldrich (USA). PAA (number average molecular weight 7.5 x 10 5 g mol -1 ) was purchased from Aldrich (USA) and used without purification. The mixtures of MC and PAA were prepared by direct mixing of the initial aqueous polymer solutions of the same concentrations; thus, the total polymer concentration in the mixture was kept constant during each experiment at different pH The raw materials used in the present study are Portland cement clinker (PCC) and raw gypsum (G). Each of those raw materials was separately ground in a steel ball mill until the surface area of respectively 3650 and 2800 cm 2 /g was achieved. The chemical composition of the raw materials is shown in Table 1. The mineralogical composition of the PCC sample is C 3 S, 58.79 %; â-C 2 S, 17.68 %; C 3 A, 8.08 %; C 4 AF, 9.72 %. The Portland Cement (PC) was prepared by mixing 96 % PCC and 4 % G (by weight) in a porcelain ball mill for one hour using 3 balls to ensure complete homogeneity of the cement. The Blaine surface area of the cement sample was 3350 cm 2 /g 18 .

Mixing and testing
At time of sample preparation, water mixed with the chemical admixtures was added gradually to about 200 g of the dry cement powder to determine the water of consistency and setting time using Vicat apparatus 23,24 . The control mix (M0) consists of Portland cement (PC) and water. In mixes M1 and M2, blend polymers were added respectively, whereas in mixes M3 and M4, methylcellulose (MC) 4 X 10 4 and 8.3 x 10 4 g/mol were added respectively were added. The addition rate was 0.5% by weight of cement. Further details about the mixes are given in Table 2. Workability test using the flow table was conducted as per BS 1881, Part 105: 1984.
To about 500 gm of the dry cement powder, water mixed with the hydrophilic copolymer was added using the predetermined water of consistency. The cement pastes were directly moulded into one-inch cube stainless steel moulds. After moulding, the mould was manually vibrated for 3-4 minutes and then on a vibrator for other 3 minutes to assure the complete removal of air bubbles and voids to produce suitable pastes. The moulds were kept in a humidity chamber at 100% R. H and constant room temperature for 24 hours, demoulded and then cured in tap water till the time of testing (3, 7, and 28 days) for compressive strength 25 . Three samples were taken for each hydration age and the mean value was considered. The curing water was renewed every week. The broken specimens from the determination of compressive strength were immersed in 1: 1 methanol-acetone mixture to stop the hydration. The mechanism of hydration was also studied by measuring the chemically combined water cement, which was carried out based on ignition loss 26 .

Characterization of the prepared blend polymers
The miscibility behavior and physicomechanical properties between methyl cellulose (MC) of different molecular weights (4 × 10 4 and 8.3 × 10 4 g/mol) and poly(acrylic acid) (PAA) have been previously reported by authors [22]. The rheological features for the obtained blends are strongly dependent on the molecular weight of the MC. The viscosity measurements showed that all blends have non-Newtonian shear thinning (pseudoplastic) behavior. These blends have a single glass transition indicating that these blends are able to form a miscible phase due to the formation of hydrogen bonds between the hydroxyl group of MC and the carboxyl group of PAA. The MC/PAA blends exhibit good mechanical properties, thermal stability, characteristics of a MC-PAA polymer network.    Application of chemical admixture for cement W/C ratio Water/Cement ratio is the ratio of the weight of water to the weight of cement used in a concrete mix and has an important influence on the quality of concrete produced, because most failures of cement structures are essentially due to either too much or too little water being added during the mixing process. The results of water of consistency of the various Portland cement pastes mixed with different chemical admixtures, are shown in Figure 1. The W/C ratio of cement pastes decreases from 0.28 to 0.25 and 0.2 with the addition of chemical admixture M1 and M2 to cement pastes mixes respectively, and increases to 0.31, 0.33 and 0.35 with the addition of admixtures M5, M4 and M3 to cement pastes mixes respectively. The same behavior was reported by Fu and Chung 27 when they studied effect of different latexes and methylcellulose on the properties of cement pastes. This ratio tends to change irregularly per the type of chemical admixtures added to Portland cement during the mixing process. This may be due to a relationship between the W/C ratio of Portland cement and the particle size as well as viscosity of the chemical admixtures used, which cause variable water requirements to produce suitable pastes [14][15][16][17][18][19] .

Setting time (initial and final)
The setting time (initial and final) of cement pastes premixed with different chemical admixtures is given in Figure 2. The results show that initial and  M4) and PAA act as retarder as in the study reported in Ref. 28,29 .

Flow
Workability is often defined as the amount of mechanical work required for full compaction of the concrete without segregation. The final strength of the concrete is largely influenced by the degree of compaction. Table 3 shows a sharp increase in flow of cement pastes premixed with admixtures (M1 & M2) and a decrease in flow of cement pastes premixed with admixtures (M3, M4 and M5). From Table 3 32 , density of concrete decreases with addition of poly vinyl alcohol and styrene-butadiene rubber to concrete mixes. Density values of concrete depend on many factors such as type of polymers, ratio of polymer and molecular weight 25,28,32 . Figure 3 shows that density value of cement pastes decreases with increasing molecular weight of MC.

Compressive strength
The results of compressive strength of cement pastes premixed with blend polymers, MC and PAA are represented as a function of curing time in Figure 4. The results show that, the compressive strength values of the pastes premixed with the blend polymers increases gradually with curing time up to 28 days, but compressive strength of pastes premixed with MC and PAA are lower during the early ages of hydration up to 28 days. The 28 days compressive strength of cement pastes premixed with blend polymers tend to be higher than those premixed with MC and PAA. This is principally due to the gradual increase in the process of polymerization or crystallization resulting from an increase in the branching of the used blend polymers. However, the type of admixture used plays role in determining cement pastes strength 33 .

CONCLUSION
The physico-mechanical proper ties of cement pastes containing different chemical admixture including blend polymer (MC/PAA), MC and PAA were investigated. Polymer blends (MC/ PAA) are new materials exhibiting combinations of properties that cannot be obtained by individual MC and PAA. However, mixing of cement pastes with blend polymers evidently improves most of the specific characteristics of the cement pastes. The reduction in W/C ratio of cement pastes premixed with polymer blends is more than that of MC and PAA respectively, is associated with longer initial and final setting times. So polymer blends act as water reducing agent when mixed with cement pastes. Flow of cement pastes mixed with polymer blends is higher than that of cement pastes containing MC and PAA respectively. Flow of cement pastes premixed with blend polymers is higher than that of pastes containing MC and PAA, due to the ball bearing action of blend polymer particles. The density of cement pastes mixed with blend polymer enhances and lower than that of reference and pastes premixed with MC and PAA respectively due to the lower weight of blend polymers than that of MC and PAA. The compressive strength of cement pastes containing blend polymers increases sharply compared to those of reference, pastes mixed with MC and PAA respectively, due to the high content of voids in presence of MC and PAA. The molecular weight of MC plays important role in the physicomechanical properties of cement pastes. However, blend polymers including MC with high molecular weight is more effect than those of MC with low molecular weight.