Effect of Graphene Oxide on the Mechanical Properties and Durability of High-Strength Lightweight Concrete Containing Shale Ceramsite

Abstract

An effective pathway to achieve the sustainable development of resources and environmental protection is to utilize shale ceramsite (SC), which is processed from shale spoil to produce high-strength lightweight concrete (HSLWC). Furthermore, the urgent demand for better performance of HSLWC has stimulated active research on graphene oxide (GO) in strengthening mechanical properties and durability. This study was an effort to investigate the effect of different contents of GO on HSLWC manufactured from SC. For this purpose, six mixtures containing GO in the range of 0–0.08% (by weight of cement) were systematically designed to test the mechanical properties (compressive strength, flexural strength, and splitting tensile strength), durability (chloride penetration resistance, freezing–thawing resistance, and sulfate attack resistance), and microstructure. The experimental results showed that the optimum amount of 0.05% GO can maximize the compressive strength, flexural strength, and splitting tensile strength by 20.1%, 34.3%, and 24.2%, respectively, and exhibited excellent chloride penetration resistance, freezing–thawing resistance, and sulfate attack resistance. Note that when the addition of GO was relatively high, the performance improvement in HSLWC as attenuated instead. Therefore, based on the comprehensive analysis of microstructure, the optimal addition level of GO to achieve the best mechanical properties and durability of HSLWC is considered to be 0.05%. These findings can provide a new method for the use of SC in engineering.

1. Introduction

Lightweight concrete (LWC), compared with normal-weight concrete (NWC), has been widely and prosperously developed in the past decades due to its lower density and higher thermal insulation performance [1]. LWC can reduce the weight of cement by up to 20% without affecting the required strength, thus contributing to saving raw materials and transportation costs [2]. In addition, it was reported that LWC could reduce thermal energy consumption by approximately 15% in order to achieve thermal comfort in European buildings [3]. At present, many researchers have further explored the potential of mechanical properties and durability of LWC in order to develop high-strength lightweight concrete (HSLWC) with excellent performance [4,5]. Aslam et al. found that using oil-palm-boiler clinker (OPBC) to replace different proportions of coarse aggregate produced lightweight concrete with a 28-day compressive strength of 47 MPa [6]. Kılıç et al. showed that using basalt pumice produced structural HSLWC with a 90-day compressive strength of 43.8 MPa [7]. Rossignolo et al. reported that using local lightweight aggregates in Brazil as coarse aggregates could manufacture HSLWC with a 28-day compressive strength of 53.6 MPa [8]. As the scope of application expands, HSLWC has gradually evolved from nonstructural materials to structural materials. Typically, HSLWC can be used as an efficient structural material to increase the number of stories in high-rise buildings, extend the span of bridges, and strengthen the corrosion resistance of offshore platforms [9,10]. Furthermore, many countries are investigating the production of HSLWC from a variety of construction waste and recycled aggregate, such as ceramicite, fly ash, OPBC, pumice stone, and geopolymers, which has been facilitated in terms of environmental protection, the economy, and sustainable development [11,12,13].

Natural resources such as river sand and stone have been excessively consumed, while the accumulation of large quantities of shale spoil has also produced ecological pollution in China. Therefore, the government requires considerable human and material resources for the disposal and recycling of shale spoil every year. Under the incentive of the increasing demand of the light aggregate industry and the updating of ceramsite production technology, shale spoil can be mass produced into shale ceramsite (SC) and shale pottery sand (SPS) with different particle size distribution sunder high-temperature calcination processing [14]. Based on the characteristics of an increased number of pores, lighter weight, and higher strength, SC is considered as a good lightweight aggregate for producing HSLWC [15]. On the one hand, the bond strength of the interface between cement slurry and SC observed by SEM is an important contribution to strengthening the mechanical properties and durability of HSLWC. On the other hand, the porous structure of SC is not conducive to the compactness of concrete, thus limiting the physical reinforcement ability. With the same ratio of sand, the higher the content of SC, the lower the ultimate compressive strength of LWAC [16]. Although adversely affected by the high water absorption of SC, the slump of HSLWC is satisfactory under the condition of a low water–binder ratio [17].

Increasing the compressive strength to 55 MPa and improving durability are challenging problems for HSLWC, which is fundamentally limited by the lightweight and porous characteristics of aggregate [18]. The use of steel fiber, carbon fiber, and resin fiber as additives to strengthen and toughen the aggregate is a common way to overcome these problems [19,20]. Adding an appropriate amount of nanosilica and fly ash resulted in a positive synergistic effect on the mechanical properties and durability of SC concrete [21]. Nevertheless, nanoscale pores and microcracks still nucleate and grow in the hydration reaction of cement mortar during long-term loading [22]. Graphene oxide (GO) is a unique two-dimensional nanosheet structure that has attracted great attention in the research to improve the performance of cement mortar and concrete [23,24]. At present, the main industrial GO preparation process involves extraction from graphene with the modified Hummer’s method, and then cooling and drying in a vacuum [25]. With the advances in and development of preparation technology, GO is bound to be mass manufactured at a lower fabrication cost. Additionally, previous studies have demonstrated that the amount of GO added to cement-based materials has been optimized to a lower dosage [26,27]. Therefore, it is economically and technically feasible for the addition of GO to strengthen the performance of concrete. In general, the mechanism through which GO enhances concrete performance is currently in an intense exploration stage, and two strongly supportive theories are emphasized as follows: (1) the formation of flower-like crystals to regulate hydration reaction [28]; (2) the provision of nanoscale filling to enhance compactness [29].

The application of GO in cement-based materials mainly focuses on workability and mechanical properties. The better adhesion between GO and cement mortar leads to the lower slump and shorter setting time for fresh concrete [30]. GO mixed with in different proportions in cement mortar results in different degrees of enhancement in the mechanical properties. The addition of 0.03% GO as the optimal dose greatly reduced the total porosity of cement mortar and correspondingly increased the compressive strength by more than 40% [31]. Simultaneously, using the same proportion of GO was the most effective in promoting the growth and regulation of flower-like crystals [32]. When the addition amount exceeded 0.04%, the enhancement effect was inhibited due to the agglomeration of GO [33]. Similar conclusions on mechanical properties have been confirmed by other researchers for the application of pervious concrete and ultra-high-performance concrete (UHPC) [34,35]. However, limited information has been presented on the influence of GO on durability. Yu et al. reported that the incorporation of GO had a favorable impact on the durability of UHPC prepared from recycled sand [36]. Zeng et al. experimentally found that GO extended the life of cement mortars by more than two times when the specimens were exposed to sulfate attack [37]. In general, the many research achievements regarding GO have mainly concentrated on cement mortar and UHPC [38,39], but few studies have been conducted to investigate the influence of GO on the mechanical properties and durability of HSLWC. Hence, it is imperative to further analyze the potential of GO to stimulate HSLWC with SC as aggregate, so as to provide practical reference for construction.

The aims of this study were to use SC as a coarse aggregate to produce grade 60 HSLWC and to design six groups of GO mixtures with different contents to investigate the reinforcement effect and determine the optimal GO content. The following three indicators were used for comprehensive verification: (1) mechanical properties (compressive strength, flexural strength, and splitting tensile strength) tests in accordance with the GB/T 50081-2002 procedure; (2) durability (chloride penetration resistance, freezing–thawing resistance, and sulfate attack resistance) tests in accordance with the GB/T 50082-2009 procedure; (3) microstructure characterization.

2. Materials and Methods

2.1. Materials

Ordinary Portland cement (P.0 42.5 R), as the binder in all mixtures, was produced from ** et al. found that concrete with a migration coefficient greater than 18 × 10⁻¹² had poor resistance to marine environments, and concrete with a migration coefficient less than 8 × 10⁻¹² had good resistance to natural environments [52]. GO can provide better corrosion resistance for HSLWC to adapt to marine environments. Hence, it could be inferred that the migration coefficients of all mixtures in the study were within the satisfactory range.

3.2.2. Freezing–Thawing Resistance

Figure 11 presents the freezing and thawing test results of specimens with different GO incorporation contents up to 250 days. As shown in Figure 11a, the mass loss rate of all mixtures gradually increased with the increase in cycles. Meanwhile, the mass losses of the specimens incorporating GO were lower than those of the specimens without GO, which indicated that GO could effectively slow down the mass losses of the during in the freezing and thawing cycles. After 250 cycles, the mass loss rates of GO-0, GO-2, GO-4, GO-5, GO-6, and GO-8 were 2.91%, 2.23%, 1.65%, 1.10%, 1.35%, and 1.92%, respectively. At this time, the surface of the specimens only grew some small holes but did not deteriorate to form cracks or peeling. The mass loss rate curve of GO-5 was higher than the mass loss rate curve of other mixes, indicating that 0.05% GO could reduce the mass loss rate to the greatest extent. As shown in Figure 11b, the relative dynamic elastic modulus had the same variation characteristics as the mass loss rate. After 250 cycles, the relative dynamic elastic moduli of GO-0, GO-2, GO-4, GO-5, GO-6, and GO-8 were 95.4%, 96.2%, 97.3%, 98.4%, 97.9%, and 96.7%, respectively. The dynamic elastic modulus curve of GO-5 was lower than the dynamic elastic modulus curve of other mixes, indicating that 0.05% GO could prevent the attenuation of the dynamic elastic modulus to the greatest extent.

The above results showed that 0.05% GO, as the optimal dosage, could help HSLWC obtain the best freezing and thawing resistance in this study. Freezing and thawing damage mainly depends on the swelling of capillary water in micro pores of concrete, whereas the addition of GO can refine pores and reduce porosity, thus impeding the free flow of capillary water. In addition, the mass loss rate and relative elastic modulus of all mixtures were in the range of 1.10–2.91% and 95.4–98.4%, respectively, which met the requirements of GB/T 50082-2009. With the contribution of GO, UHPC can obtain a relatively dynamic modulus in the range of 95.93–98.51% after 300 cycles [36]. It also should be highlighted that all the mixes in this study had excellent freezing and thawing resistance.

3.2.3. Sulfate Attack Resistance

Figure 12 shows the sulfate attack resistance results of HSLWC with different GO incorporation contents up to 150 cycling times in a sulfate solution. As shown in Figure 12a, the mass of all mixtures slightly increased in the first 60 cycles and gradually decreased in the remaining 90 cycles, indicating that the mass loss ratio showed a trend of first negative growth and then positive growth. The specimens without GO had a more significant vibration amplitude in their mass change than those with GO, which was mainly due to the fact that GO prevented deterioration in the wet and dry cycles. After 150 cycles, the mass loss rates of GO-0, GO-2, GO-4, GO-5, GO-6, and GO-8 were 2.95%, 2.21%, 1.65%, 1.54%, 1.87%, and 1.99%, respectively. As shown in Figure 12b, the corrosion resistance coefficient had the same trend as the mass loss rate of first increasing and then decreasing. The corrosion resistance coefficient of the specimens without GO decreased after 60 cycles, while those of the specimens with GO decreased after 90 cycles. After 150 cycles, the corrosion resistance coefficients of GO-0, GO-2, GO-4, GO-5, GO-6, and GO-8 were 86.3%, 89.3%, 93.7%, 97.4%, 94.6%, and 91.3%, respectively. Accordingly, the nonlinear enhancement effect was probably attributable to the uneven dispersion or supersaturation of GO, suggesting that the optimal content for sulfate resistance was 0.05%.

Corrosion resistance was reported to be related to ion transport and pore structure [53]. At the early stage of corrosion, ettringite crystals were continuously formed and accumulated, filling capillary pores, thus temporarily increasing the compressive strength and weight. In the later stage of erosion, sulfate can gradually consume and destroy the skeleton of hydration products, thus reducing the compressive strength and weight. Considering the nanofold morphology, GO could block or cut off ion transport, thus mitigating corrosion damage [54]. Hence, all mixtures in this study had excellent corrosion resistance according to the requirements of GB/T 50082-2009.

3.3. Microstructure

Figure 13 shows the SEM images of samples randomly investigated from different mix proportions at 28 days. As shown in Figure 13a, some typical crystals were observed in the specimens without GO. These typical crystals were similar to those produced by hydration reaction in cement mortar, such as layered crystals, rod-like crystals, and sheet-like crystals, which were assembled from the composite formed by AFt, AFm, and CH. This process was also accompanied by the formation of nanoscale pores and microcracks.

The formation and growth of flower-like crystals could be observed from the samples containing GO, as shown in Figure 13b–f. In particular, in contrast with GO-0, many clusters of flower-like crystals formed and grew in an orderly manner in the interfacial transition zones and pores of the GO-2 mixture (Figure 13b). When the GO content increased from 0.04% to 0.06%, the petals of the flower-like crystals grew stronger and gradually matured (Figure 13c–e). When the content of GO reached 0.08%, the shape of the flower crystals were almost unchanged, and the number of flower crystals decreased (Figure 13f). Lv et al. confirmed that GO can participate in the hydration reaction to produce a unique and dense flower-like crystal [24]. Chuah et al. also reported that these flower-shaped crystals were beneficial to improving the mechanical properties and durability of concrete [55].