Microstructures and mechanical properties of polymer modified mortars under distinct mechanisms

In this study, two types of acrylic latexes, PA (polyacrylate) and PU/PA (polyurethane modified PA), are investigated in their influences on mechanical properties of mortar under different interaction mechanisms. In light of a previous study, the polymer–cement hydrates interaction mechanisms in PA and PU/PA modified mortars are illustrated respectively, and the microstructures are simulated using a computer model. Through mechanical experiments, it is revealed that the incorporation of polymer tends to reduce the compressive strength and elastic modulus except PA at low P/C ratio, while improve the flexural strength and toughness. As compared with PA, PU/PA is more effective in these influences. All of the influences of PA and PU/PA on mechanical properties can be explained successfully based on the interaction mechanisms and microstructures. In addition, it’s also found that the compressive strength of polymer modified mortar can be roughly estimated based on a modified gel/space ratio, and the incorporation of polymers does not change the relationship between elastic modulus and compressive strength. A hightemperature curing procedure is concluded to be suitable for preparation of high-performance cement composites in short period.

Construction and Building Materials 47 (2013) 579–587 Contents lists available at SciVerse ScienceDirect Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat Microstructures and mechanical properties of polymer modified mortars under distinct mechanisms Hongyan Ma, Zongjin Li ⇑ Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Hong Kong, China h i g h l i g h t s  Microstructures of polymer modified pastes are illustrated by a computer model  Effects of different polymer latexes on mechanical properties of mortar are revealed  Gel/space ratio is modified for polymer modified cement composites a r t i c l e i n f o Article history: Received 25 January 2013 Received in revised form 17 April 2013 Accepted May 2013 Available online June 2013 Keywords: Polymer modified mortar PA PU/PA Interaction mechanism Mechanical properties a b s t r a c t In this study, two types of acrylic latexes, PA (polyacrylate) and PU/PA (polyurethane modified PA), are investigated in their influences on mechanical properties of mortar under different interaction mechanisms In light of a previous study, the polymer–cement hydrates interaction mechanisms in PA and PU/PA modified mortars are illustrated respectively, and the microstructures are simulated using a computer model Through mechanical experiments, it is revealed that the incorporation of polymer tends to reduce the compressive strength and elastic modulus except PA at low P/C ratio, while improve the flexural strength and toughness As compared with PA, PU/PA is more effective in these influences All of the influences of PA and PU/PA on mechanical properties can be explained successfully based on the interaction mechanisms and microstructures In addition, it’s also found that the compressive strength of polymer modified mortar can be roughly estimated based on a modified gel/space ratio, and the incorporation of polymers does not change the relationship between elastic modulus and compressive strength A hightemperature curing procedure is concluded to be suitable for preparation of high-performance cement composites in short period Ó 2013 Elsevier Ltd All rights reserved Introduction In the last half century, polymer modified mortar and concrete have been widely utilized in construction practice, as polymer modification can improve the workability, adhesive strength, waterproofness and many other properties of cement based materials [1–4] Although polymer modified mortar and concrete are mainly used as finishing or repair materials in their history [3], following the development of polymerization and composition techniques, they have been sometimes used massively as the major construction material in some projects, e.g pavement [5] In such applications, polymer modification may be expected to increase the tensile or flexural strength and toughness of the cement-based materials, without inducing severe decrease in compressive strength ⇑ Corresponding author Tel.: +852 2358 8751; fax: +852 2358 1534 E-mail addresses: mhy1103@gmail.com (H Ma), zongjin@ust.hk (Z Li) 0950-0618/$ - see front matter Ó 2013 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.conbuildmat.2013.05.048 In polymers that are commonly used for modification of cement-based materials, in forms of latex, emulsion or re-dispersible powder, SBR (styrene butadiene rubber), EVA (ethylene–vinyl acetate copolymer) and acrylics have been deeply studied and broadly utilized in practice [2,3,6–8] In hardened state, the phenomena of noticeable increase in flexural strength and no improvement or even reduction of compressive strength of polymer modified cement-based composites, as compared with unmodified ones, have been commonly reported [3,4,9] Of course, there are a lot of different and even contrary reports, e.g Pei et al [10] found that the incorporation of polymer latex negatively influence both compressive and flexural strength; Mohammed et al [11] reported that waste latex paint modification with low P/C (polymer to cement mass ratio) might be positive in improving compressive strength of concrete Actually, results from the literature cannot be compared with each other directly, if one does not recognize that polymers are generally added into cement composites in two different ways, say, keeping constant W/C (water to cement ratio) to obtain similar hydration of cement and keeping constant 580 H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 consistency by adjusting the W/C or the inclusion of plasticizer [12,13] The former is a typical laboratory procedure, while the later is a trial-and-error procedure with its results directly applicable in practice Generally, significant improvements of flexural strength were reported in studies using the later method Barluenga and Hernández-Olivares [12] noted that with SBR latex modification, keeping constant W/C, resulted in constant flexural strength and noticeably decreased compressive strength, while keeping constant consistency led to constant compressive strength and significantly increased flexural strength In a recent study, the advantages and disadvantages of different types of polymers, as modifiers of cement composites, were introduced [14], through which it can be seen that acrylics have perfect mechanical properties and durability They are indeed frequently employed in flooring compounds and mortars where the highest level of physical performance (adhesion, abrasion resistance, flexural strength, and impermeability) is required [2,6] Compared with SBR, acrylics as modifier can improve flexural strength more significantly at the same P/C, and may not reduce the compressive strength at low P/C [15] The purpose of the present study is right to modify cement mortar using acrylic latexes, and improve the flexural strength and toughness without reducing compressive strength obviously Mechanisms of the interaction between organic and inorganic phases in acrylics modified mortars have been discussed in the previous study [14] Two types of acrylic latexes were used as modifiers One was polymerized by emulsion polymerization with monomers of MMA (methyl methacrylate), AA (acrylic acid), HEMA (2-hydroxyethyl methacrylate) and cross-linking agent It was labeled as PA (polyacrylate) The other one was PU/PA (polyurethane modified PA) In the present study, the interaction mechanisms are introduced firstly Then, the microstructure evolution of polymer modified cement pastes are simulated using a status-oriented computer model based on the interaction mechanisms Keeping constant W/C, influences of PA and PU/PA latexes on mechanical properties of mortars are investigated experimentally, and explained based on the interaction mechanisms and simulated microstructures Materials and experiments PA and PU/PA latexes were used as polymer modifiers The details of the syntheses and physical properties of them can be found in the previous work [14] Cement that satisfies the requirements of BS EN197-1:2000 for CEM I Portland cement of strength class 52.5N (roughly equivalent to the requirements of ASTM C150 for Type I Portland cement) was used as binder The chemical compositions and physical properties of the cement are listed in Tables and 2, respectively Siliceous sand was used for preparing mortars with and without polymer The fineness modulus of the sand is 1.73, while the average and maximum grain sizes of the sand are 0.33 mm and 2.36 mm respectively Deionized water was used for mixing various mixtures Besides, in the mixing process of polymer modified mixtures, a type of organosilicon defoamer in proper amount was added to suppress the foaming effect of surfactants in the latexes In the preparation of all mortars, with and without polymer, the W/C ratio was kept 0.5, while the sand to cement weight ratio (S/C) was kept Various P/C ratios were selected to investigate its influence on mechanical properties The mix proportions of these mixtures are listed in Table After mixing, the mixtures were cast in steel moulds with different sizes for different test purpose The size of specimen is 50 mm  50 mm  50 mm for compression test, 40 mm  40 mm  160 mm for flexural test, and 25 mm  25 mm  160 mm for fracture energy test After casting, the mixtures were covered with plastic sheets to avoid evaporation After 24 h, all specimens were demoulded and cured according to three different procedures Procedure was a simple wet curing in a moisture room where the temperature and Table Chemical composition (%) of cement SiO2 Al2O3 Fe2O3 CaO MgO SO3 Loss on ignition 21.1 5.6 3.4 65.3 1.6 2.1 0.85 relative humidity were approximately 23 °C and 95% respectively Procedure consisted of a 2-day wet curing and a following dry curing in ambient environment (22 ± °C and 40–70% RH) In procedure 3, a 2-day steam curing at 60 °C was followed by a 4-day oven-drying at 60 °C and subsequently a dry curing in ambient environment Procedure is the most beneficial to cement hydration, and thus to the evolution of mechanical properties of unmodified mortars Procedure is a commonly adopted method for polymer modified mortars as the dry curing is believed to be necessary for polymer film formation [3] Procedure is similar to the one that was used to prepare a high flexural strength cement paste, in which the steam curing and oven drying were used to promote cement hydration and film formation respectively in short stages [16] Mechanical tests were conducted at ages of 3, 7, 28 and 60 days Compressive strengths of mortars were measured according to ASTM C109, on a MTS 815 ROCK Mechanics machine Flexural strengths were tested using three-point bending on a MTS 858 Mini machine, and calculated as ff ¼ 3P p L 2bt ð1Þ where ff represents flexural strength, Pp the peak load, L (140 mm) the span of specimen, b (40 mm) and t (40 mm) the width and height of the cross-section of the specimen respectively Elastic moduli of mortars were estimated from the load– displacement curves of three-point bending according to the following equation, E¼ L3 4bt Á dP dd ð2Þ where dP/dd is the gradient of the load–displacement curve corrected for the small amount of distortion in the three-point loading system As this was not a standard method, the results could not be compared with the data in the literature Thus, only the relative values of elastic moduli Er, i.e the values after being normalized by the elastic modulus of the reference mortar MPC at 60 days, would be shown in the present paper The fracture energy GF of mortars were roughly determined using a simplified method This method imitated the draft recommendation proposed by RILEM Committee on Fracture Mechanics of Concrete-Test Methods [17] based on the fictitious crack model Small-size specimens that could not fulfill the recommendation were used, and the notch length was equal to half depth of the beam It must be noted that GF measured in this way may not be the true values, thus they can only be used to compare with each other, rather than with other data in the literature Latex-hydrates interaction mechanisms and microstructure evolution 3.1 Interaction mechnisms of different types of polymer latices and cement hydrates The interaction mechanisms of the two different types of latexes (PA and PU/PA) and cement hydrates have been studied in a previous work [14] It has been found that after being incorporated into cement paste, PA latex is destabilized and demulsified PA molecules react with cement hydrates chemically to form a compound rather than form high purity film On the other hand, PU/PA latex is just slightly demulsified, and still forms film with high purity Behaviors and properties of PU/PA latex modified paste or mortar can be explained by Ohama’s multi-step model [3] to a large extent, as PU/PA latex is sterically stabilized and PU/PA molecules are relatively passive Behaviors of PA latex modified cement composites have to be explained by the newly developed 4-step model [14] In step 1, immediately after mixing, a large amount of polymer particles adsorb on the surface of cement particles or coalesce because of the demulsification The rest are still dispersed in the aqueous phase Adsorption happens in mixing process and the first several minutes after mixing In step 2, the hydration of cement is successively governed by dissolving and migration of ions Some chemical reactions take place between latex and cement hydrates Polymer particles adsorbed on the surface of cement particles are partially or totally embedded in hydrates This stage lasts for tens of minutes until flocculation occurs In step 3, cement hydration further progresses With the reaction between calcium hydroxide and surfactant, the dispersion is severely destabilized Polymer particles with cement hydrates settle down together, and flocculation happens Due to the porous nature of the flocs, the growth or sedimentation of hydrates and the chemical reactions can continue in it, so that a complex 581 H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 Table Physical properties of cement Specific gravity Blain specific surface area (cm2/g) 3.15 3580 Setting time (min) Initial Final days days 28 days 150 180 24.1 40.7 56.9 Table Mix proportions of mortars in weigh ratio Mark MPC MA05 MA10 MUA05 MUA08 MUA10 MUA15 MUA20 Water 0.5 Cement Sand Polymer solid weight PA PU/PA – 0.05 0.1 – – Compressive strength (MPa) 0.05 0.08 0.1 0.15 0.2 compound of cement hydrates and flocculated PA particles are formed Roughly speaking, this stage ends after wet curing ends At last in step 4, with the drainage of water, in a few spaces, polymer film forms, but rarely and not continuously More general situation is the formation of an organic–inorganic co-matrix 3.2 Simulation of the microstructural evolution of latex modified cement pastes Based on the latex-hydrates interaction mechanisms introduced above, in light of a status-oriented computer model for cement hydration (details can be found in Refs [18,19]), the microstructure evolution of polymer latex modified cement paste can be simulated According to the status-oriented computer model, once W/ C and particle size distribution of cement are fixed, the microstructure of cement paste can be simulated as a function of degree of hydration The simulated microstructure is composed of anhydrous cement grains, inner hydrates layers, outer hydrates layers and large capillary pores The initial microstructure of polymer modified cement paste can be simulated through randomly replacing capillary water voxels by polymer voxels until the polymer volume fraction calculated from P/C is achieved One polymer voxel is used to simulate one polymer particle In Fig 1, the initial states of pure cement paste (W/C = 0.5) and polymer modified cement paste (W/C = 0.5, P/C = 0.1) are compared In the figures, white, dark and magenta represent cement particle, capillary water and polymer particles respectively The term ‘initial state’ means an assumed special state occurs immediately after mixing and before any chemical reaction happens, thus PA and PU/PA modified pastes need not to be distinguished from each other At this state, polymer particles are still uniformly dispersed in aqueous, just like in the polymer latex Keeping the same W/C, in the polymer modified cement paste, the volume fraction of cement is lower than that in pure cement paste due to the incorporation of polymer Note that all two-dimensional images shown below are cross-sectional views cut from the simulated three-dimensional microstructure The evolution of the microstructure of pure cement paste is shown in Fig 2, in which white, blue, green and dark represent anhydrous cement grain, inner hydrates layer, outer hydrates layer and large capillary pore respectively Fig 2a shows the state at the age of days, which is the end of wet curing Fig 2b shows the state at 28 days when the paste is under air-dry curing and has been relatively mature It is assumed that the incorporation of polymer latex in cement paste does not influence the microstructure of inner hydrates layer, but only significantly influence the microstructure of outer hydrates layer To simulate the microstructure evolution of polymer latex modified cement paste, irrespective of the type of latex, the outer hydrates layer is divided into two layers, i.e the compound layer and the composite layer formed in two individual processes The compound layer is formed in the wet curing period under the reaction between cement hydrates and the adsorbed polymer particles The volume fraction of the compound layer can be easily calculated based on the degree of hydration of cement and the adsorption ratio of polymer, and the formation of this layer contacting with the original surfaces of cement particles can be then simulated according to the algorithm as described in Refs [18,19] Polymer voxels in this layer are dispersed as they cannot form any polymer film due to the chemical reactions The composite layer is formed following the withdraw of water and the formation of polymer films It is a composite of cement hydrates and polymer films The volume fraction calculation and Fig Initial states of pure and polymer latex modified cement pastes (100 lm  100 lm): (a) pure cement paste (W/C = 0.5); (b) polymer latex modified cement paste (W/ C = 0.5, P/C = 0.1) 582 H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 Fig Microstructure evolution of pure cement paste (W/C = 0.5, 100 lm  100 lm): (a) days; (b) 28 days overall formation simulation of this layer are similar with that of the compound layer Continuous polymer film in this layer can be simulated by adding polymer voxels adhering to randomly determined seed polymer voxels or existed polymer film voxels until the volume fraction is achieved The microstructure evolution of PA modified cement paste is shown in Fig 3, while that of PU/PA modified cement paste in Fig In these two figures, green and yellow are used to indicate the cement hydrates in the compound layer and composite layer, respectively, while magenta still the polymer phase In the simulations, it is assumed that the degree of hydration of cement in polymer modified pastes are the same as that in pure cement paste at the same age, as it has been proved experimentally that the incorporation of polymer latex only retards early hydration (especially in the first day), but influences little on long-term hydration [20] It can be seen that, regardless of the type of paste, the microstructure consists of several layers Hydrated pure cement paste contains four layers, i.e anhydrous cement grain, inner hydrates layer, outer hydrates layer, and large capillary pore layer Under the effects of polymers, the outer hydrates layer of polymer modified cement paste can be divided into two sub-layers, i.e compound layer and composite layer The compound layer is mainly formed by the organic–inorganic compound in the wet curing period as described above, while the composite layer is a cement hydrates-polymer film interpenetrated structure formed following the withdraw of water in the dry curing period It is obvious that PA modified paste has relatively large volume of compound layer due to the high adsorption ratio of PA, higher than 70%, compared with lower than 20% of PU/PA [14], while PU/PA modified paste has large volume of composite layer at comparable degree of hydra- tion Besides, the much higher volume fraction of polymer in composite layer of PU/PA modified paste guarantees that the formed polymer network is continuous and interpenetrate with cement hydrates The microstructural differences between these two different types of polymer modified pastes are shown more clearly in Fig 5, in which Figs 3b and 4b are locally zoomed in and compared Incorporation of polymer latex can improve the flexural strength of cement composites because of the formation of continuous polymer network which is interpenetrated with cement hydrates, as well as the constrain of polymer on skin and inner micro-cracks [21] Both PA and PU/PA are effective for the later mechanism As a continuous polymer network can form and interpenetrate with cement hydrates in the composite layer of PU/PA modified paste, the PU/PA latex should be more effective in increasing the flexural strength of mortar Generally, incorporation of polymer reduces the compressive strength, as polymer can be considered as pore phase in the cement composite to a certain degree [9], while to what a degree depends on the type of polymer Experimental results 4.1 Compressive strength 4.1.1 Influence of curing method on cement hydration and compressive strength It has been mentioned in Section that three different curing procedures were adopted for different purpose Cement degree of hydration and compressive strength evolution of un-modified mortar cured under different procedures are shown in Fig The Fig Microstructure evolution of PA cement paste (W/C = 0.5, P/C = 0.1, 100 lm  100 lm): (a) days; (b) 28 days H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 583 Fig Microstructure evolution of PU/PA cement paste (W/C = 0.5, P/C = 0.1, 100 lm  100 lm): (a) days; (b) 28 days Fig Microstructural comparison of different types of polymer modified pastes (W/C = 0.5, P/C = 0.1, 50 lm  50 lm): (a) PA modified paste, magnified from Fig 3b; (b) PU/ PA modified paste, magnified from Fig 4b evolutions of hydration degree of the cement (at W/C of 0.5) under 20 °C and 60 °C (isothermal) are from simulations using Hymostruc developed in TU Delft Note that these are rough simulations, as only temperature is considered while the influence of the relative humidity under different curing conditions cannot, due to the limitations of the software However, the accuracy of the simulation result is enough for comparison purpose rather than accurate calculation It can be seen from Fig that the development of com- Fig Influence of curing method on hydration and compressive strength development of un-modified mortar pressive strength is directly related to degree of hydration Due to the high temperature, degree of hydration of cement at days under 60 °C reaches the level of that at 28 days under 20 °C Thus, the compressive strength of mortar cured by procedure reaches a mature level at very early age That is why Bracknbury et al [16] used such a method to get high maturity in a short stage The difference of compressive strength developments following procedure and procedure can be attributed to the availability of water for continuous hydration 4.1.2 Influence of polymer on compressive strength Although the wet curing is beneficial to cement hydration, and the high temperature treating can promote hydration in a short period and get high maturity fast, the wet plus air dry curing following procedure is the most similar with that in practice Thus, more attention focuses on the comparison of unmodified mortar MPC and polymer modified mortars cured by procedure Following curing procedure 2, the influence of polymers on the evolution of compressive strength is shown in Fig Fig 7a shows that the incorporation of PA at the P/C of 0.05 (MA05) increases the compressive strength at each age, but higher incorporation ratio leads to the reduction of compressive strength In Fig 7b, it can be seen that the incorporation of PU/PA tends to reduce the compressive strength of mortar, and the higher the P/C, the larger the reduction, except MUA08 at early ages According to Neville [22], regardless of age and aggregate volume fraction, the compressive strength of concrete is directly 584 H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 proportional to the cube of the gel/space ratio, r, and can be calculated as fc ẳ 234 r 3ị r¼ where fc is the compressive strength in MPa, the 234 MPa is the intrinsic strength of the hydrates gel of cement, and the gel/space ratio, r, is defined as the ratio of the volume of cement hydrates to the sum of the volumes of hydrated cement and of the original capillary pores, i.e r¼ jh v c a v c a ỵ W=C as space to a certain degree Thus, the gel/space ratio in polymer modified cement composite should be modified to ð4Þ where jh is the hydrates volume expansion factor of ordinary Portland cement, which indicates the volume of hydrates generated when unit volume of cement is completely hydrated jh equals to 2.13 according to Sanahuja et al [23] v c is specific volume of cement with the value 0.317 cm3/g a is the degree of hydration of cement Generally, the validity of this model is good in predicting compressive strength of cement composite without polymer, as shown in Fig 8, giving that the cement degree of hydration equals to 0.48, 0.54, 0.58 and 0.60 at the ages of 3, 7, 28 and 60 days, respectively, as determined using the thermogravimetric method described in Ref [19] The poor evolution of degree of hydration at late ages should be attributed to the dry curing condition in the corresponding period Due to the soft nature of polymers compared with cement hydrates, rather than aggregate which is much more rigid and strong, polymer phase should be considered Fig Influence of polymer modifications on the evolution of compressive strength: (a) PA modified mortars vs MPC; (b) PU/PA modified mortars vs MPC jh v c a v c a ỵ W=C þ g Á P=C q ð5Þ P where qP is the density of polymer, and g is the effectiveness coefficient, which indicates to what a degree the polymer can be considered as space Experimentally determined compressive strengths of mortars, as a function of polymer type and P/C, have been plotted in Fig Assuming g = 0, which means treating polymer as aggregate with much higher rigidity, the predicted results according to Eqs (3) and (5) keep constant rather than a function of P/C Assuming g = 1, which means totally treating polymer as space and makes the predicted results much lower than experimental values By fitting experimental results to Eqs (3) and (5), it is found that g = 0.502 for PU/PA and g = 0.257 for PA give the best fitting respectively PU/PA has higher effectiveness coefficient because its rigidity is lower than PA, thus can be treated as space, or pore to a higher degree To sum up the above analysis, when studying compressive strength of polymer modified cement composite, polymer phase can be treated as space to a certain degree, while to what a degree depends on the type, or rigidity of the polymer It must be noted that this theory has a presupposition, i.e only physical interactions occur between the polymer phase and cement hydrates, e.g the effect of SBR latex on compressive strength of cementitious materials [24] According to this theory, the compressive strength decreases with the increasing of P/C undoubtedly However, the mortar MA05 gives an obvious exception, and this may be attributed to the chemical reactions, as described in the previous study [14] In a low dosage of polymer addition, the chemical reactions may make cement hydrates bond to each other to form denser and stronger structure, which enhances the compressive strength of polymer modified mortars While in a high dosage of incorporation, the polymer phase tends to form larger particles which act as defects or partial spaces as described by Eq (5) Thus, there should be a threshold P/C, below which PA enhances compressive strength mainly under the chemical mechanism, while above which the big grouping effect of PA will exceed its positive effect so that the compressive strength can be roughly estimated according to Eqs (3) and (5) By a simple observation on the experimental results, this threshold P/C for PA should be between 0.05 and 0.1 In another research which kept constant W/C, the incorporation of a polyacrylate at low P/C was also reported to improve the compressive strength [25] Fig Experimentally determined compressive strength of MPC cured by procedure and predicted results by Eqs (3) and (5) H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 585 Fig Influence of polymer on compressive strength at 28 days 4.2 Flexural strength The evolutions of flexural strength of mortars listed in Table 3, cured following procedure are plotted in Fig 10 An initial analysis of the data leads to the following remarks regarding the influence of polymer First, at the age as early as days, the flexural strength of almost all polymer modified mortars, irrespective of the type of polymer and P/C ratio, are lower than that of MPC This is because the incorporation of polymer retards the early hydration of cement on the one hand, and the polymer cannot form film in the wet curing environment on the other Second, at late ages, e.g 28 days and 60 days, following the withdraw of water and the gradual formation of polymer film, polymers tend to increase the flexural strength, especially for the cases with P/C from 0.05 to 0.1 PU/PA modifications with P/C > 0.1 seem to be not so effective in improving flexural strength, perhaps because the over-percolation of polymer film is harmful to the continuity of cement hydrates Actually this is also true for PA modified mortars according to limited data involving high polymer content Although no continuous film network can form in PA modified mortars as shown in Fig 3, PA can also improve the flexural strength, as the incorporation of PA latex can effectively limit the formation of micro-cracks and delay the propagation of microcracks when loaded Beside this mechanism, continuous polymer film network can form in PU/PA modified mortars, as shown in Fig 4, which helps to improve the flexural strength more significantly E.g at the age of 60 days, MA05 has a flexural strength of 8.51 MPa, which is 7.59% higher than the 7.91 MPa of MPC, while the value of PU/PA modified mortar (MUA05) at the same P/C and the same age is 9.68 MPa, which is 22.38% higher than MPC The data in Fig 10 clearly highlight an optimum of polymer content, corresponding to P/C between 0.05 and 0.1 irrespective of polymer type, with regard to improving flexural strength A similar conclusion has also been drawn by Bureau et al [13] in their study on mechanical properties of SBR modified mortars As compared with the mortars cured by procedure 2, the hightemperature treated mortars following procedure seem to have higher flexural strength at the age of 28 days, as shown in Fig 11 The high-temperature steam curing not only promotes the early hydration and generates more hydrates fast as proved by Fig 6, but also makes sure the high continuity of hydrates phase at the end of days Under the help of the followed oven curing, polymer film can form with high quality and without disturbing the continuity of the hydrates phase With these mechanisms, PU/PA modifications even with high P/C ratio (0.2) can also improve the flexural strength Fig 10 Influence of polymer modifications on the evolution of flexural strength: (a) PA modified mortars vs MPC; (b) PU/PA modified mortars vs MPC 4.3 Elastic modulus Due to the reasons described in Section 2, only relative elastic moduli of mortars are plotted against age in Fig 12 Simple observation can lead to a conclusion that the incorporation of polymer tends to decrease the elastic modulus of mortar, and the higher the P/C, the larger degree of the decreasing at late ages This is almost a general feature of polymer modified cement composites, especially in the cases of keeping constant W/C [3,20,26] Fig 11 Flexural strength comparison of PU/PA modified mortars cured by different methods at the age of 28 days 586 H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 Compared with others, the MA05 only has limited decreasing in elastic modulus Again, this may be attributed to the chemical interactions between PA latex and cement hydrates According to the ACI building code (ACI 318), the elastic modulus of concrete can be estimated using a binary function of its unit weight and compressive strength Assuming a constant density for normal weight concrete, the elastic modulus will be directly proportional to the square root of compressive strength [27] Other standards may use different equations to link elastic modulus to compressive strength, e.g in Eurocode and CEB-FIP Model Code the elastic modulus of concrete is directly proportional to the cubic root of its compressive strength [28] Although only relative elastic moduli are given in the present study, it should also read Er ẳ k1 fc1=2 6ị or Er ẳ k2 Á fc1=3 Fig 13 The relation between relative elastic modulus and compressive strength ð7Þ where k1 and k2 are constants In Fig 13, the elastic moduli of all modified and unmodified mortars at different ages are plotted against the corresponding compressive strength, and then fitted using Eq (6) It is obvious that the relationship between relative modulus and compressive strength can be fitted by a unique function, irrespective of the type of mortar In other words, the empirical equations used to estimate the elastic modulus of concrete based on compressive strength can also be used for polymer modified concrete Based on the digitized three-dimensional microstructure as described in Section 3, in light of multi-scale micromechanics [29] Fig 12 Influence of polymer modifications on the evolution of elastic modulus: (a) PA modified mortars vs MPC; (b) PU/PA modified mortars vs MPC or finite element methods [30], the elastic properties of polymer modified cement composite may be predicted more accurately 4.4 Toughness The ratio of flexural strength to compressive strength (ff/fc) of a cement composite is an important indicator of its toughness [24,31] Higher ff/fc indicates higher toughness The toughness indicators of all mortars at late ages are shown in Fig 14 It can be seen that the incorporation of polymer latexes of both types can improve the toughness (indicator) The effect of PA is very limited at low incorporation ratio (P/C = 0.05), but considerable somewhat at higher P/C ratio (0.1) PU/PA seems to be more effective in increasing the toughness indicator, due to the formation of continuous polymer film network as shown in Fig The fracture energy GF of PU/PA mortars have been roughly measured, and the results are shown in Fig 15 Cured following procedure 2, the incorporation of PU/PA latex helps to improve the fracture energy When P/C is not larger than 0.1, the degree of improvement increase with the increasing P/C However, when P/C > 0.1, the improvement becomes not so significant, due to the disturbed hydrates continuity induced by the large volume of polymer This trend is consistent with that of toughness indicator, as shown in Fig 14 MUA10 (P/C = 0.1) gives the best performance at 28 days, with an increasing of around 40% in fracture energy, as compared with MPC (P/C = 0) High-temperature treatment following procedure further improves fracture energy, as clearly shown in Fig 15 Under the help of high-temperature steam, large amount of hydrates form in the first days with high continuity, Fig 14 Toughness indicator comparison of mortars cured by procedure H Ma, Z Li / Construction and Building Materials 47 (2013) 579–587 587 (2009CB623200) and from Hong Kong RGC, Systematic studies on magnesium phosphate cement-based concrete (615810) are gratefully acknowledged References Fig 15 Fracture energy of PU/PA modified mortars at 28 days cured by different methods and in the following oven-curing for facilitating polymer film formation, higher P/C results in higher fracture energy Conclusions The following conclusions can be drawn from the present study: (1) PA and PU/PA modified mortars have different microstructure, due to their differences in modification mechanisms This has been clearly illustrated in the present study under the help of a status-oriented computer model for microstructure simulation (2) Incorporation of polymer tends to reduce the compressive strength of mortar The compressive strength of polymer modified mortars can be roughly estimated based on a modified gel/space ratio (3) Polymer modifications reduce the elastic modulus of mortar, but not influence the elastic modulus-compressive strength relationship Irrespective of the type of mortar, its elastic modulus can be estimated based on a unique function describing the relationship between elastic modulus and compressive strength (4) Cured under procedure 2, a curing method similar with the environment in practice or in situ curing, polymer modifications can improve the flextural strength and toughness, and PU/PA performs better than PA due to the formation of highquality polymer film network In PU/PA modified mortars, the P/C range between 0.05 and 0.1 seems to be the most reasonable considering both function and cost (5) Cured under the high-temperature procedure 3, which consists of a cement hydration promotion period and a polymer film formation facilitation period, the flexural strength and fracture energy of PU/PA modified mortars can be further improved Thus, this method has the potential to be used to prepare high-performance cement composites in short period Acknowledgements Financial supports from a China Basic Research Grant, Basic Research on Environmentally Friendly Contemporary Concrete [1] Chandra S, Ohama Y Polymers in concrete Boca Raton, FL.: CRC Press; 1994 [2] Miller M Polymers in cementitious materials Shawbury, UK: Rapra Technology; 2005 [3] Ohama Y Handbook of polymer-modified concrete and mortars Park Ridge, NJ.: Noyes Publications; 1995 [4] Ohama Y Polymer-based admixtures Cem Concr Compos 1998;20:189–212 [5] Yao H, Liang N, Sun L, Meng J The design of the polymer cement concrete pavement and the analysis of test road J Chongqing Jiaotong Univ 2005;24:83–7 [6] ACI-Committee-548 Polymer-modified concrete, ACI 548.3R-03 Farmington Hills, MI.: American Concrete Institute; 2003 [7] Chung DD Use of polymers for cement-based structural materials J Mater Sci 2004;39:2973–8 [8] Ramakrishnan V Latex-modified concretes and mortars: synthesis of highway practice Washington DC: National Cooperative Highway Research Program (NCHRP), Transportation Research Board; 1992 [9] Schulze J, Killermann O Long-term performance of redispersible powders in mortars Cem Concr Res 2001;31:357–62 [10] Pei M, Kim W, Hyung W, Ango AJ, Soh Y Effects of emulsifiers on properties of poly(styrene–butyl acrylate) latex-modified mortars Cem Concr Res 2002;32:837–41 [11] Mohammed A, Nehdi M, Adawi A Recycling waste latex paint in concrete with added value ACI Mater J 2008;105:367–74 [12] Barluenga G, Hernández-Olivares F SBR latex modified mortar rheology and mechanical behaviour Cem Concr Res 2004;34:527–35 [13] Bureau L, Alliche A, Pilvin P, Pascal S Mechanical characterization of a styrene– butadiene modified mortar Mater Sci Eng, A 2001;308:233–40 [14] Ma H, Tian Y, Li Z Interactions between organic and inorganic phases in PAand PU/PA-modified cement-based materials J Mater Civ Eng 2011;23:1412–21 [15] Wu K, Zhang D, Song J Properties of polymer-modified cement mortar using pre-enveloping method Cem Concr Res 2002;32:425–9 [16] Bracknbury WR, Grzeskowiak R, Reid NL, Lynn ME Flexural strengths of polymer modified cement pastes prepared by a novel curing process Cem Concr Res 1988;18:971–9 [17] RILEM committee on fracture mechanics of concrete-test methods Determination of the fracture energy of mortar and concrete by means of three-point bend tests on notched beams Mater Struct 1985;18:285–90 [18] Ma H, Li Z Modeling cement paste at micro-scale In: Choi CK, editor Proceedings of the 2011 world congress on advances in structural engineering and mechanics, Seoul, Korea: Techno-Press; 2011 p 992–1004 [19] Ma H, Li Z Realistic pore structure of Portland cement paste: experimental study and numerical simulation Comput Concr 2013;11:317–36 [20] Odler I, Liang N Properties and development of the microstructure in cement pastes modified by a styrene–butadiene co-polymer Adv Cem Res 2003;15:1–8 [21] Pascal S, Alliche A, Pilvin P Mechanical behaviour of polymer modified mortars Mater Sci Eng, A 2004;308:1–8 [22] Neville AM Properties of concrete 4th and final ed New York: J Wiley; 1996 [23] Sanahuja J, Dormieux L, Chanvillard G Modelling elasticity of a hydrating cement paste Cem Concr Res 2007;37:1427–39 [24] Wang R, Lackner R, Wang PM Effect of styrene–butadiene rubber latex on mechanical properties of cementitious materials highlighted by means of nanoindentation Strain 2011;47:117–26 [25] Medeiros MH, Helene P, Selmo S Influence of EVA and acrylate polymers on some mechanical properties of cementitious repair mortars Constr Build Mater 2009;23:2527–33 [26] Wong WG, Fang P, Pan JK Dynamic properties impact toughness and abrasiveness of polymer-modified pastes by using non-destructive tests Cem Concr Res 2003;33:1371–4 [27] Mindess S, Young JF, Darwin D Concrete 2nd ed Upper Saddle River, NJ.: Prentice Hall; 2003 [28] Noguchi T, Tomosawa F, Nemati KM, Chiaia BM, Fantilli AP A practical equation for elastic modulus of concrete ACI Struct J 2009;106:690–6 [29] Julien S, Luc D, Gilles C Modelling elasticity of a hydrating cement paste Cem Concr Res 2007;37:1427–39 [30] Koenders EA, Dolado JS, van Breugel K, Porro A Nano to microlevel modeling of cement-based materials ACI Special Publ 2009;267:1–10 [31] Wang R, Wang PM, Li XG Physical and mechanical properties of styrene– butadiene rubber emulsion modified cement mortars Cem Concr Res 2005;35:900–6  ... influences of PA and PU/PA latexes on mechanical properties of mortars are investigated experimentally, and explained based on the interaction mechanisms and simulated microstructures Materials and. .. of almost all polymer modified mortars, irrespective of the type of polymer and P/C ratio, are lower than that of MPC This is because the incorporation of polymer retards the early hydration of. .. Tables and 2, respectively Siliceous sand was used for preparing mortars with and without polymer The fineness modulus of the sand is 1.73, while the average and maximum grain sizes of the sand are