A Comprehensive Review on Attempts to Improve Rubber-Cement Matrix Interface
Introduction
Incorporating rubber aggregates into cementitious mixtures typically leads to reduced mechanical properties, primarily due to bond defects between the rubber aggregates and the cement matrix This issue is compounded by the low stiffness of rubber aggregates Various studies have explored methods to enhance the bond between these aggregates and the cement matrix, employing techniques such as coatings and treatments for the rubber aggregates.
- Washing or soaking rubber aggregates with water;
- Pre-treating rubber aggregates (sodium hydroxide NaOH);
- Adding chemical admixtures (styrene-butadiene rubber latex, silane coupling agent…) to cementitious mixtures;
- Coating rubber aggregates with cementitious materials (cement particles, cement paste, mortar…) before mixing with other concrete/mortar ingredients;
- Adding silica fume or nano silica as additives into mixtures
Each method for enhancing the mechanical properties of rubberized cement-based composites has its own strengths and weaknesses Despite numerous attempts documented in the literature, none have successfully met the goal of significantly improving these properties.
Figure 1.1 Primary attempts to improve mechanical properties of rubberized cement-based composites
Typical previous attempts to improve rubber-cement matrix interfacial bond
1.2.1 Rubber aggregates pre-treated with sodium hydroxide NaOH
Sodium hydroxide NaOH is a popular chemical agent to pretreat rubber aggregates before introducing them into cementitious mixtures According to Segre and Joekes (2000) and
Li et al (2004), this NaOH treatment enhanced the hydrophilic properties of rubber aggregates Normally, rubber particles have smooth surface (Thomas et al., 2014, Colom et al.,
When sodium hydroxide (NaOH) interacts with rubber surfaces, it roughens the texture by creating voids and promoting an alkali-silica reaction with the silica present in the rubber (Mohammadi et al., 2015) Additionally, NaOH effectively removes zinc stearate from the rubber surface, which improves adhesion properties and enhances surface homogeneity (Segre et al., 2002).
The procedure of rubber pre-treatment process using NaOH involves the following steps illustrated in Figure 1.2 (Youssf et al., 2014):
(i) wash rubber aggregates by water to remove any impurities and dust;
(ii) immerge rubber aggregates in NaOH solution at a certain pH level for a given period in a container;
(iii) wash rubber aggregates again by stirring in water until the pH reaches 7;
(iv) finally leave rubber aggregates for air dry
Figure 1.2 Procedure of rubber pre-treatment using sodium hydroxide (Youssf et al., 2014)
Regularly checking the pH of NaOH solution every 5 minutes with a pH meter is essential to maintain a pH of 14, with a total process time of approximately 30 minutes (Youssf et al., 2014; Sgobba et al., 2015) Mohammadi et al (2015) determined that the optimal immersion time for rubber aggregates in a 10% NaOH solution at pH 14 is 24 hours, while Najim and Hall (2013) recommend a pre-treatment duration of only 20 minutes Treatment of rubber aggregates with sodium hydroxide at varying intervals (20 minutes, 2 hours, 24 hours, 48 hours, and 7 days) results in a rougher surface texture, as illustrated in Figure 1.3 (Mohammadi et al., 2015).
Figure 1.3 SEM images of rubber surface (a) no treatment (b) after 20 minutes (c) after 2 hours (d) after 24 hours (e) after 48 hours (f) after 7 days (Mohammadi et al., 2015)
Segre and Joekes (2000) observed that a cement paste specimen containing 10% as-received rubber by mass exhibited a fractured surface, revealing that rubber particles were easily extracted The presence of empty spaces at the rubber-matrix interface indicated poor adhesion between untreated rubber aggregates and the cement paste In contrast, an adhesive interface was established between NaOH-treated rubber particles and the matrix, demonstrating improved bonding.
Figure 1.4 Microstructure of cement paste samples with: untreated tire rubber (a), 10 % by mass of NaOH-treated tire rubber (b) (Segre and Joekes, 2000)
Research by Mohammadi et al (2014) indicates that rubberized concrete made from water-soaked aggregates shows a 22% increase in compressive strength compared to untreated rubber aggregates (size 0.075-4.75 mm), while flexural strength only rises by 8% Additionally, Youssf et al (2018) found that pre-treating rubber aggregates by soaking them in water for 24 hours and allowing them to air dry can enhance compressive strength by 15%, specifically for rubber sizes of 9.5-16.0 mm with a 20% partial replacement of normal weight concrete stone Mohammadi et al (2014) also highlighted the advantages of washing or water-soaking rubber aggregates.
(i) Diminish the repelling-water characteristic of rubber particles due to a gradual release of air bubbles entrapped on the surface of rubber aggregates (Figure 1.5);
(ii) Reduce the hydrophobic behavior of rubber aggregates
Figure 1.5 Effect of water-soaking method: full of air bubbles on rubber surface (left) and a release of the air bubbles after 24-hour soaking (Mohammadi et al., 2014)
Research indicates that pre-treating rubber aggregates with NaOH enhances the mechanical properties of rubberized cement-based composites Youssf et al (2014) found that replacing 20% of sand volume with NaOH-treated rubber aggregates (1.18 - 2.36 mm) improved compressive strength by 6% and 15% at 7 and 28 days, respectively, and increased elastic modulus by 12%, though it also resulted in a 12% decrease in splitting tensile strength compared to untreated aggregates Similarly, Mohammadi et al (2015) reported a 25% increase in compressive strength and a 5% increase in flexural strength when using as-received rubber aggregates (0.075-4.75 mm).
In 2017, research demonstrated that modifying rubber aggregates through NaOH surface treatment and various coating techniques, including normal cement and blended cements, significantly enhanced the compressive strength of concrete mixtures Specifically, mixtures incorporating NaOH-treated or cement-Na2SiO3-coated rubber aggregates outperformed other rubberized options Due to its simplicity and effectiveness in improving mechanical performance and long-term durability for rigid pavement applications, the NaOH treatment method was recommended for further use (Guo et al., 2017) However, contrary findings by Albano et al (2005) indicated minimal improvements in compressive and tensile strengths when using NaOH pre-treated rubber as a fine aggregate replacement Additionally, Li et al (2004) noted that NaOH treatment was only effective for small-sized rubber tire powders, as larger aggregates did not benefit from the treatment due to their reduced surface area for chemical reactions.
The impact of NaOH pre-treated rubber aggregates (L20-P) versus untreated rubber aggregates (L20-N) on the properties of rubberized concrete has been analyzed, revealing significant differences in key performance metrics The study, conducted by Youssf et al (2014), highlights variations in slump, compressive strength, modulus of elasticity, and splitting tensile strength, demonstrating the potential benefits of using pre-treated rubber aggregates in enhancing the mechanical properties of rubberized concrete.
Segre and Joekes (2000) found that cement paste with 10% untreated rubber aggregates had lower water capillary absorption compared to the control Additionally, specimens with NaOH-treated rubber aggregates absorbed even less water than those with untreated rubber The study concluded that better adhesion in the rubber-cement matrix resulted from using NaOH-treated aggregates, which also led to reduced mass loss during abrasive testing.
Figure 1.7 Effect of NaOH-treated rubber aggregates on water capillary absorption (Segre and Joekes, 2000)
Figure 1.8 Effect of NaOH-treated rubber aggregates on abrasion resistance (Segre and Joekes,
Research indicates that the durability of rubberized concrete subjected to freeze-thaw cycles is significantly enhanced when utilizing NaOH-treated rubber particles compared to untreated rubber particles (Si et al., 2017) In this study, rubber aggregates sized between 1.44 mm and 2.83 mm were pre-modified by soaking in a 4% NaOH solution for 40 minutes at room temperature The concrete specimens were formulated with 15% as-received rubber aggregates alongside varying amounts of NaOH-treated rubber aggregates.
In a study by Si et al (2017), rubberized concrete containing varying volumes of 0-4 mm-sized sand (15%, 25%, 35%, and 50%) was subjected to approximately 500 freeze-thaw cycles The findings revealed that NaOH-treated rubber particles exhibited a strong bond with the cement paste, which significantly limited mass loss and reduced the decrease in relative dynamic modulus of the concrete specimens Specifically, the untreated rubberized concrete (C-AS-15) experienced a 7% change in dynamic modulus, while the treated version (C-OH-15) only saw a 2% change from the 246th to the 508th cycle.
Figure 1.9 Effect of NaOH-treated crumb rubber aggregates on mass of rubberized concrete under freeze-thaw cycles (Si et al., 2017)
Figure 1.10 Effect of NaOH-treated crumb rubber aggregates on dynamic modulus of rubberized concrete under freeze-thaw cycles (Si et al., 2017)
1.2.2 Rubber aggregates glued with sand
Meddah et al (2014) utilized the MEDAPLAST resin from Granitex Factory in Algeria to adhere fine sand to the surface of rubber aggregates These sand-glued rubber aggregates partially replaced coarse aggregate in the mix, contributing to the production of rubberized concrete specifically designed for roller compacted concrete pavement (RCCP).
Figure 1.11 Rubber treated with gluing-sand (Meddah et al., 2014)
Gluing sand particles to untreated rubber aggregates (3-8 mm, 25% volume substitution of coarse aggregates) significantly enhances the compressive and flexural strengths of roller compacted concrete, achieving increases of 29% and 15%, respectively In contrast, rubber aggregates treated with NaOH solution for 24 hours, followed by washing and drying, only demonstrate a 11% increase in compressive strength and a 24% increase in flexural strength compared to untreated mixtures However, the compressive strength of the NaOH-treated rubber aggregates remains lower than that achieved through the sand-gluing method.
Figure 1.12 Effect of sand-glued rubber aggregates on compressive and flexural strengths (Meddah et al., 2014)
1.2.3 Rubber aggregates pre-coated with cement paste or mortar
Najim and Hall (2013) examined the mechanical-dynamic properties, porosity, and fracture energy of self-compacting rubberized concrete, incorporating rubber aggregates (2-6 mm, 12% by volume) pre-treated through various methods such as water washing and NaOH immersion Their findings revealed that rubber aggregates pre-coated with mortar significantly enhanced fracture toughness and energy absorption compared to other treatments and untreated aggregates Specifically, the compressive and splitting tensile strengths of the concrete with mortar pre-coated aggregates (SCRC3) increased by 37% and 19%, respectively, compared to those using untreated aggregates (SCRC1) This improvement is attributed to a stronger interfacial bond between the rubber and cement matrix.
Figure 1.13 Rubber aggregates pre-coated with mortar (left) and with cement paste (right) (Khalid Battal Najim and Hall, 2013)
Figure 1.14 Rubber-cement matrix interfacial bond with mortar-precoated rubber aggregates (bottom) and untreated ones (top) (Khalid Battal Najim and Hall, 2013)
In a study by Li et al (1998), cement paste and methocel cellulose ethers were used to pre-coat rubber particles in rubberized concrete, where 33% of the sand volume was replaced with 2.5 mm rubber aggregates The findings revealed that cement paste-coated rubber particles (RBC-2) outperformed methocel cellulose-coated particles (RBC-3) in terms of practicality and performance The use of cement paste-coated rubber aggregates led to increases in compressive strength, modulus of elasticity, and maximum flexural load by 31%, 15%, and 3%, respectively, compared to untreated rubber (RBC-1) However, the compressive strength of the rubberized concrete remained significantly lower than that of the control concrete (RBC-0) Additionally, the flexural test highlighted the importance of post-peak residual bearing capacity in RBC-2, demonstrating the effectiveness of cement paste pre-coating in enhancing the bond between the rubber aggregates and the cement matrix.
Figure 1.15 Effects of cement paste or methocel cellulose ethers-coated rubber aggregates on compressive strength and elastic modulus (a), and flexural strength (b) (Li et al., 1998)
The literature review indicates that there is little significant difference between precoating rubber aggregates with mortar or cement paste and using untreated rubber aggregates in mortar or concrete mixtures The primary issue lies in the bond defect between the rubber aggregates and the cement matrix Coating the rubber with mortar or cement paste has shown improvements in certain mechanical properties, as noted by Li et al (1998) and Khalid Battal, Najim, and Hall (2013) However, the question arises as to why rubber aggregates should be coated when they are to be included in cementitious mixtures The observed increase in mechanical properties may not be directly linked to the enhancement of the rubber-cement matrix interface, but rather could result from reduced air entrapment during mixing and casting, or from the formation of a stronger shell around the rubber aggregates.
Summary
From the above review of the literature in this chapter, it can be summarized as following:
The bond defect between rubber aggregates and the cement matrix in rubberized cement-based composites significantly impacts their properties, highlighting the need to enhance the rubber-cement matrix interface Several studies indicate that the hydrophobic nature of rubber aggregates leads to air entrapment during mixing and casting, resulting in numerous air bubbles on their surface when immersed in water This phenomenon contributes to a compromised Interfacial Transition Zone, characterized by increased porosity and high permeability, ultimately diminishing the quality of the composite.
Numerous studies have sought to enhance the adhesion between rubber aggregates and the cement matrix, achieving varying levels of success in mitigating mechanical strength loss However, research indicates that the strength improvements from rubber treatment are minimal and remain significantly lower than that of conventional concrete or mortar without rubber aggregates (Rashad, 2015) Additionally, only a limited number of studies have explored the impact of rubber treatments on the transfer properties of rubberized cement-based composites, primarily focusing on aspects such as porosity and water absorption.
Various treatment methods for rubber aggregates have been explored, including simple techniques like washing, soaking in water, and pre-treating with sodium hydroxide (NaOH) Other methods involve coating rubber with limestone, cement paste, or mortar, as well as gluing sand onto the rubber surface or using polymeric agents as standard admixtures in cementitious mixtures Research by Meddah et al (2014), Mohammadi et al (2015), Khalid Battal Najim and Hall (2013), Segre and Joekes (2000), and Youssf et al (2014) highlights the effectiveness of NaOH as a chemical agent for treating rubber aggregates, as it enhances the surface properties of the material.
Increasing the surface roughness of rubber aggregates leads to greater entrapment of air bubbles (Richardson et al., 2016) To enhance interlocking bonding, methods such as gluing sand onto rubber aggregates (Meddah et al., 2014), pre-coating with cement paste or mortar (Khalid Battal Najim and Hall, 2013), or using limestone (Onuaguluchi and Panesar, 2014) have been proposed Coating rubber aggregates with resin and sand is expected to create a more hydrophilic surface, which is beneficial for improving composite properties However, the difference in performance between using coated rubber aggregates and mixing crude rubber directly with mortar or concrete ingredients may be minimal Any observed increase in mechanical properties from the slurry coating method could be due to reduced air entrapment during mixing or the formation of a stronger shell around the rubber aggregates compared to the standard matrix.
Alternative methods for treating rubber aggregates, such as oxidation and the use of silane coupling agents alongside carboxylated styrene butadiene rubber latex, present complexities that hinder their large-scale application Specifically, these treatments necessitate high-temperature hydrolysis at 70 °C, as noted by Li et al (2016), or require drying at 110 °C, making them impractical for widespread use.
Burning rubber aggregates at 250 °C in a controlled oxygen/nitrogen environment can modify waste rubber properties, making them hydrophilic (Chou et al., 2010) However, this high-temperature process can release harmful emissions and increase rubber stiffness, which may negatively impact the mechanical advantages of incorporating rubber aggregates into cementitious mixtures These aggregates are primarily used to enhance strain capacity and reduce cracking in cement-based composites due to restrained shrinkage (Ho et al., 2009; Turatsinze and Garros, 2008; Turatsinze et al., 2005) Therefore, methods that stiffen rubber aggregates can counteract their beneficial effects.
Chemical agents such as latex, acrylic, vinylic, and styrene-butadiene polymeric admixtures are commonly incorporated into concrete mixtures to enhance the bond between rubber aggregates and the cement matrix Research indicates that these admixtures can be added directly to the concrete or mixed with rubber particles shortly before integration For example, styrene-butadiene-rubber (SBR) latex is recommended to be combined with rubber aggregates and then added to the concrete components promptly.
2005) Another approach is to premix rubber aggregates with SBR latex and to wait for 20
The study by Bowland et al (2012) indicates that mixing should occur within 43 minutes to ensure optimal integration of materials The research found that the addition of Styrene-Butadiene Rubber (SBR) at levels of 20% and 150% of the rubber aggregate by mass significantly enhances the mechanical properties of concrete This improvement is attributed to the SBR not only coating the rubber aggregate surface but also being evenly distributed throughout the concrete matrix.
Recent studies indicate that incorporating silica fume as a partial replacement for cement (approximately 15% by mass) in rubberized concrete mixtures significantly enhances their strength compared to conventional mortar and concrete This strength improvement is attributed to the pozzolanic reaction and the filling effect of silica fume, rather than solely to the enhancement of the rubber-cement matrix interface.
Most studies indicate that rubber aggregate treatments have a minimal impact on the mechanical properties of rubberized cement-based materials, primarily due to the low stiffness of rubber aggregates However, if the bond between the rubber and cement matrix is enhanced, the transfer properties of these materials could potentially match those of traditional mortar without rubber aggregates.
This research aims to establish a strong bond between rubber aggregates and the cement matrix, and to examine how this bonding influences the properties of the composite, particularly focusing on transfer properties that significantly impact the sustainability of rubberized cement-based applications.
Improvement of Rubber-Cement Matrix Interface and Characterization of
Mortar constituents
Rubber aggregates, derived from mechanically grinding used tyres into sizes of 0 to 4 mm, exhibit a specific gravity of 1.2 and minimal water absorption Their modulus of elasticity ranges from 0.01 to 0.1 GPa, with compact and expanded rubber aggregates measuring 68 MPa and 12 MPa, respectively Notably, rubber aggregates are hydrophobic, repelling water and trapping air bubbles on their surface, which are released into the cement matrix upon immersion Energy Dispersive X-ray Spectrometry (EDS) analysis reveals that carbon is the primary component of rubber aggregates, with zinc stearate present as an additive to enhance oxidation resistance and maintain hydrophobicity Additionally, the thermal expansion coefficient of rubber aggregates significantly differs from that of the cement matrix.
Figure 2.1 (a) Rubber aggregates, sand, and (b) cement type used
In this study, 30% of rubber aggregates were used as a partial replacement of sand by an equivalent volume in order to increase strain capacity of mortar Rounded siliceous sand (0-
4 mm, specific gravity of 2.62 and water absorption of 1.9%) (Figure 2.1 a) and cement CEM I
52.5R (Figure 2.1 b) whose main compositions is given in Table 2.1 were prepared The slight difference in size gradation between rubber aggregates and sand is illustrated in Figure 2.4
Figure 2.2 Air bubble-entrapment on rubber surface immersed in water
Figure 2.3 Chemical composition of studied rubber aggregates
To enhance the workability of rubberized mortar and prevent segregation, a polycarboxylate-type superplasticizer and a high molecular weight synthetic polymer-based admixture were utilized, owing to the low density and hydrophobic characteristics of rubber aggregates.
Table 2.1 Main compositions of Cement CEM I 52.5R
SO3 (%) CaO/SiO2 MgO (%) C3S (%) C2S (%) C3A(%) C4AF(%)
Figure 2.4 Grading curves of sand and rubber aggregates
Approaches to improve rubber-cement matrix interface
The bond defect between rubber aggregates and the cement matrix in rubberized cement-based materials significantly impacts the mechanical and transfer properties of these composites Most treatments for rubber aggregates have shown minimal effect on improving these properties, leading to a general consensus that the mechanical properties of rubberized cement-based composites are reduced due to the low stiffness of the rubber aggregates However, if the bond between the rubber aggregates and the cement matrix is enhanced, the transfer properties of these composites could potentially rival those of traditional mortar without rubber aggregates.
This study explores the use of a coating agent to enhance the bond between rubber aggregates and the cement matrix For optimal results, the coating must adhere well to the rubber aggregates and be compatible with the cement paste Additionally, the research suggests incorporating an air detraining admixture to minimize entrapped air during mixing and casting, aiming to achieve a rubberized mixture with air content comparable to that of control mortar This admixture is recommended to reduce the interface gap between the rubber aggregates and the cement matrix.
48 cement matrix At best, if the cement paste and the rubber aggregates will come to contact with each other, a good interface can be induced
Rubber aggregates possess hydrophobic properties, necessitating the use of a styrene-butadiene-type copolymer bonding resin to enhance their interaction with water This coating facilitates the hydration process at the interface between the cement matrix and the copolymer-coated rubber aggregates Notably, the copolymer not only bonds effectively to the rubber aggregates but also improves both cohesion and adhesion to the cement matrix, as detailed in the product information.
The copolymer bonding agent enhances the tensile strength of mortar while reducing Young's modulus of elasticity, which are crucial properties for increasing the strain capacity of cement-based composites.
The addition of air-detraining admixture, particularly those made from modified polysiloxanes, significantly reduces the formation of pinholes, bubbles, and air pockets in cementitious mixtures during mixing and application This de-aerating effect enhances the density of concrete, resulting in increased compressive strength, reduced porosity and permeability, and improved durability By minimizing excess air bubbles, the overall appearance of exposed concrete is enhanced, leading to a more compact surface in poured concrete.
Rubber aggregates possess hydrophobic properties that lead to significant air entrapment during the mixing process This entrapment causes a partial release of air bubbles, ultimately increasing the air content in rubberized cement-based mixtures, as noted by Wong et al.
In 2009, it was found that total porosity and the presence of micro-cracks significantly impair the transfer properties of cement composites, more so than the porosity of the Interfacial Transition Zone (ITZ) Therefore, minimizing air entrapment is essential for reducing the density of air pores within the matrix and improving the bond at the rubber-cement matrix interface.
Compositions of studied mortars
This study utilized two primary compositions: control mortar and untreated-rubberized mortar, based on prior research from the Laboratoire Matériaux et Durabilité des Constructions de Toulouse (Nguyen, 2010) In the rubberized mortar, 30% of the sand volume was substituted with rubber aggregates ranging from 0-4 mm Additionally, two types of admixtures were incorporated to enhance the interface between the rubber and the mortar.
Four distinct mortar mix proportions were prepared, as detailed in Table 2.2, incorporating aggregates and a cement matrix The acronyms UR and CR denote mixtures containing Uncoated Rubber and Coated Recycled Aggregate (RA), respectively Additionally, the letters A and P refer to the Air-detraining admixture and Polymer bonding resin, which serve as treatment materials to enhance the rubber-cement matrix interface.
In the coating procedure, rubber aggregates were mixed with a styrene-butadiene copolymer (2% of RA by mass) and then allowed to stabilize at room temperature (20 °C) and 50% relative humidity for one hour This essential step prevents the polymer layer on the rubber surface from being removed during subsequent mixing with the pre-mixed mixture The copolymer coating, combined with the pre-mixing of sand, cement, water, and other admixtures, effectively limits air void formation in the cementitious matrix (Kim and Robertson, 1997) The premixed mixture was prepared similarly to the control mix (0R) and uncoated rubberized mixtures (30UR, 30UR-A), where sand and cement—either with or without rubber aggregates—were premixed for three minutes before adding water and other admixtures to complete the mixing in two minutes.
Table 2.2 Mortar mix proportions (values in kg/m 3 )
The study determined the optimal amount of air-detraining admixture made from modified polysiloxanes by balancing the air content between a control mix and a rubberized cement-based mixture As shown in Figure 2.5, the addition of rubber aggregates resulted in high air content and low unit weight, attributed to the air-entrapment phenomenon on the rubber surface and the low specific gravity of the rubber aggregates (Fedroff et al., 1996) The purpose of using air-detraining admixture is to reduce air content and gradually increase unit weight With the control mixture exhibiting an air content of approximately 3%, the adjusted quantity of air-detraining admixture was set at 2.56.
The air content and fresh density of the mixtures were determined following NF EN 12350-7 (2012) and NF EN 12350-6 (2012) standards Fresh mortar was compacted into a watertight container of known volume, and its mass was measured The unit weight of the fresh mortar was then calculated by dividing the mass by the container's volume, with results expressed to the nearest 10 kg/m³.
Figure 2.5 Effect of air-detraining admixture on air content and unit weight of rubberized cement-based mixture
Improvement of rubber-cement matrix interface
The Scanning Electron Microscope (SEM) has been utilized since the 1960s to examine the microstructural characteristics of cementitious composites SEM can analyze various forms, including fragments, sectioned and polished surfaces, or powders, producing detailed images through secondary or backscattered electrons that reveal surface characteristics For qualitative analysis, backscattered electron images combined with Energy Dispersive X-ray Spectrometry (EDS) are effective in identifying and assessing the chemical constituents of these composites The ASTM C1723-16 standard (2016) outlines the application of SEM-EDS techniques for concrete analysis For quantitative X-ray microanalysis, such as point, line, or area scans, flat and polished specimens must be prepared through sectioning, drying, resin impregnation, and grinding.
51 polishing, (v) cleaning (with alcohol), (vi) coating, (vii) observations Note that micro-cracks can be induced during drying process and unsuitable polishing progress may affect microstructural details to be observed
X-ray diffraction (XRD) is a powerful tool used to both qualitatively-quantitatively identify and qualify the crystalline structure in cement-based composites (Zürz & Odler, 1987; Scrivener, Füllmann, Gallucci, Walenta, & Bermejo, 2004; Snellings, Salze, & Scrivener, 2014) When an X-ray beam strikes a powdered/flat sample, the X-rays are scattered by random atoms and the diffracted X-ray obtained produces a diffraction pattern containing information about the atomic arrangement within the crystal It is necessary to recall that only crystalline materials, not amorphous materials produce a diffraction pattern From the given diffractometer, information of chemical compositions in cementitious composites such as portlandite, ettringite, carbonate, quartz and so on will be obtained Note that each crystalline material has a unique diffraction pattern according to Bragg’s law The proportion of phases in Portland cement and Portland-cement clinker using x-ray powder diffraction analysis will be based on ASTM C1365-06(2011) standard (2011)
SEM studies on rubberized cement-based composites reveal voids at the rubber-cement matrix interface (N Segre & Joekes, 2000; Turki et al., 2009; Corinaldesi et al., 2011) Various treatments of rubber aggregates have shown improved adhesion, with NaOH-treated particles exhibiting a strong bond with the cement matrix (Segre and Joekes, 2000), and silica fume enhancing adhesion on rubber surfaces (Pelisser et al., 2011) Additionally, rubber aggregates coated with mortar or polymeric admixtures also demonstrated effective bonding (Khalid Battal Najim and Hall, 2013; Albuquerque et al., 2005) EDS analysis by Pelisser et al (2011) indicated high carbon content on rubber surfaces, while EDX results showed a predominance of silicon over calcium in mixtures with 50% rubber substitution, alongside significant calcium carbonate presence (Turki et al., 2009) Furthermore, Sgobba et al (2015) utilized X-ray diffraction to analyze mineral phases and found no significant differences in hydrated products between control and rubberized concrete specimens.
The literature review indicates that many scanning electron microscopy (SEM) images focus on the rubber-cement matrix interface to address issues related to rubber incorporation in cementitious mixtures and the impact of various treatments on rubber aggregates Additionally, energy-dispersive spectroscopy (EDS) and X-ray diffraction (XRD) techniques have been employed to analyze these interactions further.
This study identifies the global chemical compositions of rubberized cement-based composites, focusing on the rubber-cement matrix interface rather than the chemical nature of the epitaxial reaction The analysis indicates insufficient information regarding the effects of treatment methods on this interface To address this, Scanning Electron Microscopy (SEM) was employed to examine the rubber-cement matrix interface (ITZ) and the cement matrix where rubber particles were extracted Observations included the surfaces of untreated and coated rubber aggregates, which were compared to their original morphology Additionally, the chemical nature of the epitaxial reaction at the interface was analyzed using X-ray Diffraction (XRD) on model samples.
Figure 2.6 SEM observations on rubber-cement matrix interface of fractured sample (a) and on footprint formed by rubber extraction (b), and crystalline phases identification on model samples by XRD (c)
Microstructural analyses of uncoated and coated rubber aggregates, along with the Interfacial Transition Zone (ITZ) between the rubber and cement matrix, were conducted using Scanning Electron Microscopy (SEM) Each mortar sample, measuring approximately 10 mm x 10 mm x 10 mm, was fractured, and three rubber aggregates were meticulously extracted from the fractured surfaces, leaving imprints in the cementitious matrix Prior to observation, all samples were cleaned with alcohol and coated with a thin layer of carbon The study focused on examining the cement matrix within the holes and the interface between the rubber aggregates and the cement matrix.
The morphological texture of untreated and coated rubber aggregates was visualized and compared before their incorporation into the mixture and within the composites to identify any changes.
Figure 2.7 Fractured sample preparation for SEM observations
To assess the bonding mechanism in civil materials and structures, the pull-out or pull-off test is commonly utilized He et al (2016) performed pull-out tests and found minimal differences in bonding strengths between untreated and sulphonated rubber aggregates with a cement matrix, specifically 0.17 MPa and 0.239 MPa, respectively The low stiffness of rubber aggregates leads to deformation during these tests, making it challenging to accurately record the bonding strength with the available apparatus Therefore, X-ray diffraction (XRD) analysis using Cu-Kα radiation is an effective method for identifying the crystalline phases present on the surface of rubber aggregates For this analysis, model samples consisting of 10 mm rubber aggregates, partially immersed in a cementitious paste with a water-cement ratio of 0.47, were prepared After curing these samples for 28 days in a controlled environment (20 °C and 95% relative humidity), the rubber aggregates were extracted to prepare testing samples for crystalline phase identification on both the rubber aggregate and cement matrix sites This phase identification is crucial for confirming the bonding phenomenon at the interface.
Figure 2.8 Rubber aggregates of 10 mm (a), specimen preparation (b), and model samples for XRD analysis (c)
2.4.2.1 Morphology of rubber-cement matrix interface
Morphological studies of as-received and coated rubber aggregates were conducted using SEM, revealing that the surface of the rubber aggregates is smooth, consistent with findings by Thomas et al (2014) Coating the rubber particles created a thin layer of copolymer on their surface, which aids in understanding the changes at the rubber-cement matrix interface in rubberized composites.
Figure 2.9 Surface morphology of as-received and coated RA (SEM images x500)
The mortar produced with untreated rubber aggregates shows a notable gap at the interface between the rubber particles and the cementitious matrix, attributed to the water-repellent nature and air entrapment of the rubber aggregates, a phenomenon known as "bond defect" (Turatsinze et al., 2006) This indicates a lack of bonding between rubber aggregates (R) and the cement matrix (C) when compared to conventional aggregates Upon extracting the rubber aggregates, the cement matrix reveals a porous texture In contrast, the rubberized mortar with air-detraining admixture still exhibits bond defects, yet the interface transition zone (ITZ) appears reduced, allowing for better contact between the rubber aggregates and the cement paste Cracks are observed at the interface in both scenarios, and the surface of uncoated rubber aggregates remains smooth, indicating no bonding occurred at the interface with uncoated rubber aggregates.
Figure 2.10 Morphology of rubber-cement matrix interface (SEM images x500)
Rubber coating with styrene butadiene copolymer bonding resin, followed by mixing with a pre-mixed cementitious mixture, significantly enhances the interface between rubber aggregates and the cement matrix At a magnification of x500, no bond defects are visible, and the cement-matrix site appears compact after rubber extraction SEM images at x1000 show that rubber particles adhere tightly to the cement matrix and embed within the mortar, indicating a stronger interfacial bond and a denser microstructure due to the copolymer's bonding effects The crack path is localized in the cement matrix rather than at the rubber-cement interface, which is typically observed with uncoated rubber aggregates A thin layer of bonding resin on the copolymer-coated rubber aggregates is notably altered after hydration, suggesting possible chemical interactions at the interfacial zone Localization of the interface between rubber aggregates and the cement matrix is challenging EDS analysis identified high carbon content on the rubber side, confirming the presence of rubber aggregates, while another point analysis identified the cement matrix at the interface.
57 Figure 2.11 Different imprints of rubber aggregates in cement matrix (SEM images x1000)
Figure 2.12 Effect of rubber coating on the interface (SEM image x1000)
58 Figure 2.13 Typical EDS point analysis at the interfacial zone: (a) cement matrix and (b) rubber particle
2.4.2.2 Chemical nature of epitaxial reaction at rubber-cement matrix interface
XRD analysis, as shown in Figures 2.14 and 2.15, reveals diffraction peaks primarily associated with portlandite in the interfacial zone between the rubber and cement matrix The cement matrix depicted in Figure 2.14 exhibited similar hydrated products Notably, the untreated rubberized mix displayed the highest intensity of portlandite at a peak of 18°, while the control mix and copolymer-coated rubberized mix showed a slight increase at a peak of 34.1° compared to other matrixes However, the diffractogram details are insufficient to determine if the cement matrix effectively bonds with the rubber aggregates.
Figure 2.14 XRD diffractogram of the ITZ on cement matrix side
Polymer-coated rubber aggregates exhibit a significantly higher intensity of portlandite (CH) at peaks of 18° and 34.1° compared to uncoated rubber surfaces, with or without air-detraining admixture This similarity in the development of hydrated products between the coated rubber surface and control cement paste demonstrates a strong bond between the cement matrix and the polymer-coated rubber aggregates The co-polymer bonding resin facilitates water contact with the newly created surface, enhancing the hydration process at the interfacial zone Although increased portlandite is present at the interface, it may be obstructed by the polymer film during crystallization, resulting in a denser and well-bonded interface.
The diffractogram reveals that the peak of portlandite at 34.1° is significantly higher than at 18°, although the peak at 18° appears more pronounced This discrepancy can be attributed to the orientation of portlandite, which is often found in a plate-like formation localized within the interfacial transition zone (ITZ) of samples (Mehta and Monteiro, 1988) Additionally, the water-repellent effect surrounding uncoated rubber aggregates may restrict the hydration process, leading to the formation of a porous and weak interface.
Figure 2.15 XRD diffractogram of the ITZ on rubber aggregate side
Compressive strength and modulus of elasticity in compression
The 28-day compressive strength was determined according to NF EN 12390-3 standard (2012) After curing cylindrical specimens (112 mm in diameter and 220 mm in height) under controlled atmosphere (20 °C temperature and 95% relative humidity) for 27 days, top and bottom faces of the specimen were ground to impose high parallelism and good contact with loading plate The compressive test was carried out using a hydraulic testing machine with a load capacity of 4000 kN After the application of the initial preload (30 kN), controlled load applied to the specimen increases continuously at the loading rate of 0.5 MPa/s until the difference of two consecutive loads recorded less than 20%
The 28-day modulus of elasticity was determined according to RILEM CPC8 (1975) recommendations A special ring extensometer illustrated in Figure 2.16 consists of three
The study utilized 61 displacement sensors positioned at 120-degree angles on the lateral sides of test samples to measure longitudinal displacement Four loading cycles were conducted at a rate of ± 0.5 MPa/s, with the stress amplitude varying from an initial 0.5 MPa to 30% of the specimen's maximum bearing capacity The modulus of elasticity was subsequently calculated using the normal stress and the average longitudinal deformation recorded by the sensors during the final loading cycle.
Where σa is 30% of the average strength, σb is the basic strength (0.5 MPa), εa,n and εb,n are the measured longitudinal strains to stresses σa and σb for the cycle n (n = 4), respectively
Figure 2.16 Ring extensometer (left) and testing set-up for modulus of elasticity test (right)
Photographs from compressive tests on cylindrical specimens reveal a distinct difference in failure modes between control mortar and rubberized mortar specimens The rubberized specimens maintain their original shape despite experiencing significant vertical displacement and horizontal deformation, indicating some recoverable behavior upon unloading Eldin and Senouci (1993) note that high horizontal internal tensile stresses develop at the top and bottom of rubber aggregates during compression, leading to numerous tensile cracks in the rubberized specimens However, rubber aggregates exhibit superior energy absorption and tensile deformation capabilities compared to conventional aggregates, serving two key functions in rubberized cement-based composites: they act as springs that delay crack opening and prevent the propagation of micro-cracks at the rubber-cement matrix interface, thereby reducing sharpness and promoting stress relaxation.
Research indicates that the propagation of micro-cracks in rubberized mortar can be effectively delayed, preventing the coalescence of these micro-cracks into larger macro-cracks (Ho et al., 2012b; Turatsinze and Garros, 2008) Consequently, rubberized mortar specimens demonstrated resilience under compression, showing no complete disintegration.
Figure 2.17 Failure due to compression of the control mortar (left) and the rubberized mortar (right)
The 28-day compressive strength (fc) and modulus of elasticity (E) for mortars are shown in Figure 2.18 and Table 2.3 One should note that each value is the average from 3 tests for compressive strength and of 2 tests as determining Young modulus These results are in agreement with the ones reported by several previous authors, namely the presence of rubber aggregates induces a reduction in mechanical properties compared to control mortar (Turatsinze et al 2005, Nguyen et al 2010) As observed, the use of an air-detraining admixture or the coating of rubber aggregates by a co-polymer bonding resin can slightly reduce the loss of compressive strength and of the modulus of elasticity For some authors (Fedroff et al., 1996; Khatib and Bayomy, 1999), the decrease in compressive strength was attributed to the increase in air content of cementitious mixtures as rubber aggregates were incorporated However, as demonstrated above, when air-detraing admixture was used to obtain the rubberized mixture having the same air content as the control mix, the decrease in compressive strength was still observed Hence, the decrease in the compressive strength and modulus of elasticity can be attributed to the low stiffness of rubber aggregates (Turatsinze et al., 2005) or to the void content in the rubberized composite as low stiff rubber aggregates may be viewed as voids (Ho et al., 2012b; Khatib and Bayomy, 1999; Turatsinze and Garros,
Figure 2.18 Compressive strength and modulus of elasticity of different mortars
Table 2.3 Compressive strength and modulus of elasticity
Compositions Compressive strength (MPa) Modulus of Elasticity (GPa) fc Standard deviation E Standard deviation
Concluding remarks
Rubber aggregates are hydrophobic, trapping air and repelling water upon contact with mixing water, which increases the air content in rubberized cementitious mixtures Energy Dispersive Spectroscopy (EDS) analysis reveals that carbon is the primary component of these rubber aggregates Additionally, previous studies have reported that the low stiffness of rubberized aggregates influences the properties of rubberized composites.
Rubberized mortar exhibited higher air content and lower unit weight compared to control mortar, primarily due to air entrapment and the low density of rubber aggregates during mixing To address this, the use of air-detraining admixture proved effective in reducing air content and increasing the unit weight of the rubberized mortar By adjusting the dosage of the air-detraining admixture, it is possible to achieve a rubberized mixture with air content comparable to that of the control mixture.
The Interfacial Transition Zone (ITZ) between untreated rubber aggregates and the cement matrix exhibits a porous texture due to air entrapment on the rubber surface and the water-repellent nature of the aggregates These characteristics adversely affect the hydration process at the rubber-cement interface To mitigate air entrapment, the use of air-detraining admixtures can enhance the contact between the two phases, improving overall performance.
The application of a rubber coating agent enhances the density of the interfacial transition zone (ITZ), resulting in a crack-free rubber-cement matrix interface at the same magnification A cement matrix layer is visible on the surface of polymer-coated rubber aggregates Therefore, coating rubber aggregates and allowing one hour for the copolymer to fully densify before lightly mixing with a pre-mixed cementitious mixture is an effective method to improve the rubber-cement matrix interface.
XRD analysis proves to be a more effective method than traditional pull-out tests for assessing the bonding strength of cement paste on rubber aggregates within cementitious mixtures The analysis revealed significantly higher intensities of portlandite (CH) at peaks of 18° and 34.1° on copolymer-coated rubber aggregates compared to uncoated ones, indicating that the copolymer coating enhances the hydration process at the interfacial transition zone (ITZ) between the coated rubber aggregates and the cement matrix This results in a denser and better-bonded rubber-cement matrix interface Conversely, when untreated rubber aggregates are incorporated, XRD analysis shows no cement paste presence, leading to a porous rubber-cement matrix interface, suggesting that hydration is hindered at the interface with untreated rubber aggregates.
Incorporating rubber aggregates into cementitious mixtures leads to reduced compressive strength and modulus of elasticity, primarily due to the low stiffness of rubber aggregates and the high porosity at the rubber-cement matrix interface.
The incorporation of an air-detraining admixture or the application of a co-polymer bonding resin to rubber aggregates can effectively minimize the decrease in compressive strength and elasticity modulus in a 65 matrix.