Influence of coal ash on the concrete properties and its performance under sulphate and chloride conditions

This study investigated the influence of coal bottom ash (CBA) on the concrete properties and evaluate the effects of combined exposure of sulphate and chloride conditions on the concrete containing CBA. During concrete mixing, cement was replaced with CBA by 10% of cement weight. Initially, concrete samples were kept in normal water for 28 days. Next, the specimens were moved to a combined solution of 5% sodium sulphate (Na2SO4) and 5% sodium chloride (NaCl) solution for a further 28 to 180 days. The experimental findings demonstrated that the concrete containing 10% CBA (M2) gives 12% higher compressive strength than the water cured normal concrete (M1). However, when it was exposed to a solution of 5% Na2SO4 and 5% NaCl, gives 0.2% greater compressive strength with reference to M1. The presence of 10% CBA decreases the chloride penetration and drying shrinkage around 33.6% and 29.2% respectively at 180 days. Hence, this study declared 10% CBA as optimum that can be used for future research.


Introduction
Coal is a source of energy to produce electricity. The burning of coal in a furnace of a power plant results in the generation of noncombustible ashes like coal bottom ash (CBA) and fly ash (FA). Typically 1 MW electricity is produced through 15-18.75 tons burning of coal and left behind around 4.3 tons of CBA and 11 tons of FA (Asokan et al. 2005). Approximately annual coal ash generation is about 600-800 million tons worldwide (Hui et al. 2009). Coal ash that is removed from flue gases in an electro-precipitator is named FA which is around 70-80% of total ash and the remaining 20-30% of heavyweight ash falls into the bottom of the furnace, called CBA (Singh et al. 2015). In America, the coalfired power plants produce around 22.6 million tons of FA and 3.8 million of CBA (ACAA 2017), particularly Malaysian power plants contribute around 6.8 million tons of FA and about 1.7 million tons of CBA annually (Rafieizonooz et al. 2016) and in Indian around 155 power plants are operated on coal and produces approximately 169.25 million tons of FA and 34 million tons of CBA (CEA 2017). It was reported by Central Electricity Authority, New Delhi, that the fly ash is almost 63.28% utilized for different purposes i.e., concrete, cement, earth filling, masonry, road embankments, agriculture and others (CEA 2017). However, a high volume of fly ash is already utilized in the concrete (Hooton et al. 1994) (Jayaranjan et al. 2014) and it was recognized as a cement constituent and standardized vide BS EN 197-1 (2011). It is also adopted in cement manufacture, to conquer the problem of carbon dioxide (CO 2 ) emissions in the environment. However, the use of CBA is still unexplored, and it is being directly disposed-off into the open ponds which occupies a huge land area and contaminated the soil and underground water resources (IDEM 2017). According to the Physicians for Social Responsibility (PSR) (United States Affiliate of International Physicians for the Prevention of Nuclear War 1985) which declared that coal ash has a dangerous and toxic material in storage and disposal under wet condition, they suggested that dry storage should be for the extreme command to avoid leaching, moving or leakage of toxicants. The use of coal for electricity generation requires efforts to manage the safe storage, disposal and reuse of huge amount of coal ash; therefore, more research is required to be carried out on the possible use of CBA in concrete construction to minimize environmental pollution (Deonarine et al. 2015). CBA is porous in nature and dark grey in colour and after grinding, it poses fine particles as presented in Fig. 1. The chemical characteristics of CBA and fly ash are almost similar but coarser than fly ash. The CBA particles mostly fall in the range of 4.75 mm to 90 μm (Singh et al. 2015). The particles size of CBA are coarser and almost comparable to that of fine aggregate due to that CBA is formerly considered sand replacement material in concrete, which causes the reduction in compressive strength of concrete (Singh and Siddique 2016). However, after the grinding process, it offers a good opportunity to be applied as cement, which could deliver better compressive strength (Mangi et al. 2018a) (Khan and Ganesh 2016). But, its performance may be affected in the aggressive environment that represents marine environment such as the combined condition of sulphate and chloride.
Presently, the performance of the concrete structure under aggressive conditions is a challenging task for the engineers and solutions for that problem have become more popular. Limited studies have been started on the concrete with supplementary cementitious material (SCM) under the combined effects of sulphate and chloride. A review of literature on the concrete comprising different SCM under sulphate and chloride attacks is summarized in Table 1. Maes and De Belie (2014) investigated the sulphatechloride combined effect on the concrete containing blastfurnace slag as cement. They declared that chloride penetration increased as sulphate content increased but sulphate attack is mitigated by the presence of chloride. Furthermore, Mangi et al. (2018b) investigated individual effects of sodium sulphate and sodium chloride on concrete containing CBA. They were acknowledged that CBA has substantial potential to reduce sulphate and chloride effects in individual solution. Besides that, Snelson and Kinuthia (2010) considered pulverized fuel ash (PFA) as cement replacement in the mortar. It was soaking in sodium sulphate solution for 504 days. They found that mortar with PFA gives good sulphate resistance.
Considering the performance of geopolymer mortars containing lignite bottom ash exposed to 3% sulphuric acid and 5% sodium sulphate solutions, it was investigated by Sata et al. (2012) that bottom ash mortars deliver good performance and it is less susceptible to sodium sulphate and sulphuric acid solutions. Okoye et al. (2017) evaluated geopolymer concrete containing fly ash and silica fume, exposed to 2% sulphuric acid and 5% sodium chloride solutions. Strength was declined around 36% and 8% in control mix and concrete with fly ash and silica fume when exposed to 2% H 2 SO 4 at 90 days Compressive strengths were declined around 18% and 0% in control mix and concrete containing fly ash and silica fume when exposed to 5% NaCl at 90 days. Hence, concrete including silica fume in sulphuric acid and chloride solution gives satisfactory and higher performance than the control mix. Stroh et al. (2016) considered concrete with fly ash and slag exposed to the combined attack of sulphate and chloride at laboratory conditions with NaCl and Na 2 SO 4 solution different concentrations. They declared that rapid access and binding of chloride ions by alumina, causing Friedel's salt formation. However, more sulphate ions encourage ettringite formation and sulphate react partially with portlandite creating gypsum and pH depletion. Low pH leads to depletion in Friedel's a b salt. Moreover, Kazi Tani et al. (2018) evaluated the influence polyethylene terephthalate (PET) on the concrete performance when exposed to 5% sodium sulphate (Na 2 SO 4 ). They declared that PET blend cement reduces the effects of Na 2 SO 4 . Modified concrete/mortars need to be introduced and investigated for a better environment and sustainable development.
The literature review indicated that the earlier studies were conducted on single-solution exposure of sulphate or chloride and rare studies were found on combined effects of sulphate and chloride solution. However, the actual conditions are different from the individual solutions. Most of the real structures are exposed to collective solutions, especially in the marine environment. Therefore, the novel appraisal of this study is to investigate the performance of concrete incorporated ground CBA exposed to the combined solution of sulphate and chloride, in which solutions represent the marine environment.

Scope of study
This study evaluated the performance of concrete incorporated CBA under sulphate-chloride salt conditions. Three groups of specimens were prepared as shown in Table 2. Group-I specimens are initially kept in normal water for 28 days as recommended by ACI Committee 318 (1985) to achieve a targeted compressive strength of 35 MPa. Afterward, specimens were transferred into 5% Na 2 SO 4 + 5% NaCl solution and a compressive strength test was performed at 28, 56, 90 and 180 days. Next, group-II specimens were cured in normal water and RCPT testing was performed at 28 days as an initial test and 180 days as matured concrete. Next, the results were compared with initial RCPT test results with RCPT test at 180 days. Group-III specimens were initially cured in normal water for 7 days as per ASTM C596 (ASTM C596 2010); afterward, specimens were kept at the room ambient temperature and drying shrinkage readings were taken at 1, 3, 7, 14, 28, 56, 90 and 180 days. Figure 2 shows the experimental setup for compressive strength, RCPT instrument and drying shrinkage instrument.

Materials
Ordinary Portland cement (OPC) in accordance with BS EN 197-1 (2011) having consistency as 30%, initial setting time 90 min, final setting time 270 min and specific gravity 3.10 was used. The CBA obtained from a thermal power station, Selangor, Malaysia, was used as OPC replacement. Original CBA was grinding for 20 h in a ball mill grinder as shown in Fig. 3. The chemical characteristics of OPC and CBA were assessed through an X-ray fluorescence (XRF) test, refer to Table 3. However, it was found that the CBA is rich in SiO 2 ,  -05 2005). In addition to that, the particle size of OPC and CBA was also evaluated through a particle size analyser (PSA) and results are provided in Fig. 4. It was observed through the PSA that OPC is finer than ground CBA. The OPC particles fall in the range of 3.8 to 21.2 μm, whereas CBA particles were in the range of 3.7 to 50.5 μm. However, the specific gravity of OPC and CBA were recorded as 3.10 and 2.41, respectively. Additionally, a scanning electron microscope (SEM) of CBA is also provided in Fig. 5, which shows CBA particle shapes, indicated as irregular, sharp and spherical particles.

Concrete mix
The scope of this study is to involve 10% CBA as SCM; it was also formerly recommended as an optimum proportion in concrete production (Mangi et al. 2018a) (Mangi et al. 2019). Therefore, two concrete mixes were prepared with fixed water to binder ratio (w/b) of 0.50. The first mix contains 100% OPC and the second mix incorporated 10% CBA as SCM. Fine aggregates passing from 5 mm sieve, coarse aggregates nominal maximum size of 10 mm was used in this study. The ACI method was used to prepare concrete mix proportions as shown in Table 4.

Results and discussion
Compressive strength performance To evaluate the pozzolanic reaction of CBA in concrete, the specimens were prepared with 10% CBA as a replacement for OPC. Moreover, specimens were cured in normal water and a combined solution of 5% sodium sulphate (Na 2 SO 4 ) and 5% sodium chloride (NaCl). Experimental results are summarized in Table 5, and Fig. 6 demonstrated the compressive strength relationship with respect to exposure period concrete, while Fig. 7 indicated the percentage variation in compressive strength under normal water and combined solution of 5% Na 2 SO 4 + 5% NaCl. It was observed that the performance of M2 under normal water was found to be superior to M1. In the early days, the compressive strength of M2 was lower than M1, because the pozzolanic reaction is not yet initiated; then, at 56 days and onwards, the compressive strength of M2 was noticed greater . It was noticed that the hydration process is affected due to CBA in concrete due to a reduction in the amount of Ca(OH) 2 (Jaturapitakkul and Cheerarot 2003). However, M2 concrete which contained 10% ground CBA cured in normal water gives the compressive strength 60.0 MPa at 180 days, which is almost 12% greater than the control mix (M1) concrete. While the same concrete gives 0.2% higher compressive strength when exposed to a combined solution of 5% Na 2 SO 4 + 5% NaCl at 180 days, it was observed that M1 concrete when exposed to combined 5% Na 2 SO 4 + 5% NaCl solution reduces its strength from 44 to 40.6 MPa at the age of 28 days and 55.3 to 51.05 MPa at 90 days; this is due to reduction in binding ability in presence of the sulphatechloride condition. Previously, it was explained that the binding of incoming sulphate ions consequences in the formation of ettringite and gypsum (De Weerdt and Justnes 2015). Formation of ettringite in the first phase until bonding capacity It was also observed in this study that control mix (M1) concrete, when exposed to a combined solution of 5% Na 2 SO 4 + 5% NaCl, losses its binding ability, whereas M2 concrete which contains 10% CBA, enhances the silica and alumina content in the cement paste and resulting in the development of compressive strength.

Micrograph and X-ray diffraction (XRD) analysis
Scanning electron microscopy (SEM) and X-ray diffraction (XRD) technique were adopted to analyse the problems associated with the concrete when it was exposed to the combined  Fig. 8 SEM image of the concrete without CBA exposed to a combined solution of 5% Na 2 SO 4 + 5% NaCl Fig. 9 SEM image of the concrete with CBA exposed to a combined solution of 5% Na 2 SO 4 + 5% NaCl Fig. 10 XRD diffraction pattern of CM concrete exposed to 5% Na 2 SO 4 + 5% NaCl Fig. 11 XRD diffraction pattern of concrete with CBA exposed to 5% Na 2 SO 4 + 5% NaCl sulphate-chloride at 180 days. Figure 8 shows the condition of control mix (M1) concrete exposed to a combined solution of 5% Na 2 SO 4 and 5% NaCl, whereas Fig. 9 shows the condition of concrete with 10% CBA (M2) exposed to a combined solution of 5% Na 2 SO 4 and 5% NaCl. The micrograph of M1 and M2 concrete indicated two major features. One is the holding salts and the second is the formation of C-S-H gel.
The reduction in concrete strength is mainly caused by less formation of C-S-H gel (Aggarwal and Siddique 2014). The formation of C-S-H gel is retarded due to salt of sulphate and chloride. It was observed that concrete with CBA (M2) has well formation of C-S-H gel as compared to normal concrete (M1). Furthermore, it was observed that the amount of calcium silicate was found higher in the control mix (CM) as compared to concrete containing CBA. The presence of CBA in concrete produces more C-S-H gel since CBA is high in aluminium and OPC paste is high in calcium carbonate, which makes reaction among these two compounds forms tricalcium aluminate (C 3 A) and gives adequate strength development. However, the development of ettringites (hydrous calcium aluminium sulphate) formation gives good strength as the volume of voids was occupied by the ettringites (Aggarwal and Siddique 2014). Moreover, the CM exposed to a combined solution of 5% Na 2 SO 4 + 5% NaCl was found to be more severe because the presence of NaCl salts produces chloroaluminate in chloride solutions and deterioration takes place by de-calcifications that was noticed in all specimens. It was noticed that the hydration of cement formed different products such as quartz (SiO 2 ), portlandite (calcium hydroxide), belite (C 2 S), alite (C 3 S), calcium silicate hydrate (CSH), C 3 A (tricalcium aluminate), and ettringite (C 3 A (CS)H 32 ). The pozzolanic reaction was detected in the concrete containing CBA, reaction among Ca (OH) 2 (calcium hydroxide). It shows that CBA could create well shape CSH gel which delivered higher compressive strength performances (Figs. 10 and 11).

Chloride penetration performance
The durability of concrete can be assessed through a rapid chloride penetration test (RCPT). It is necessary to check chloride penetrability when a concrete structure is built within the chloride environment. This study involves CBA as supplementary cementitious material in concrete. However, CBA contains 17.65% aluminium oxide (Al 2 O 3 ) and OPC contains 3.95%, which means that concrete containing CBA (M2) holds more amount of Al 2 O 3 . Therefore, chlorides physically and chemically bound due to adsorption on the surface of hydrates C-S-H and C-A-H. The addition of CBA increases the amount of Al 2 O 3 , which gives the better chloride binding ability (Argiz et al. 2018). Besides that, a fine particle of CBA also fills the free pore (Argiz et al. 2017). These phenomenal changes give the lower penetration of chloride ions through the concrete incorporated 10% CBA. It was observed that concrete with 10% CBA (M2) gives the lower chloride penetration as the age of concrete increases. It was observed from Fig. 12

Drying shrinkage performance
This study evaluated the effect of CBA on the drying shrinkage (DS) of concrete was also evaluated. However, it was previously noticed that the shrinkage occurred in the first 2 months and for a later age, very less variation in shrinkage of concrete (Saha 2018). It was experimentally known that the DS in the control mix (M1) is higher than the concrete containing 10% CBA (M2) even up to 180 days as shown in Fig. 13. The DS is significantly depending on the fineness of supplementary cementing materials (SCM); it was formerly noticed that concrete containing fly ash nano-particle gives the higher DS (Gao et al. 2017) and DS decreases when the coarser fly ash was used (Saha 2018). The DS in concrete containing 10% CBA (M2) were noticed lower than the control mix (M1), which validated that ground CBA particle is coarser than the ordinary Portland cement (OPC). However, CBA addition as SCM reduces the CaO contained in the cement paste which leading towards a reduction in the hydration process. Thus, lower DS was recorded in M2 as compared to M1.

Conclusions
This study declared CBA as supplementary cementitious material (SCM) due to the presence of a substantial proportion of three main oxidize, i.e., silica, alumina and ferrous. The following conclusions can be drawn.
i. Concrete incorporated 10% CBA gives 12% higher compressive strength than normal concrete cured in normal water. Once exposed to a combined solution of sulphatechloride, it gives around 0.2% higher compressive strength as compared to normal concrete. ii. It was detected through micrographs that concrete containing CBA has well formation of C-S-H gel as compared to the control mix concrete. However, the development in the strength was noticed. iii. Incorporation of 10% CBA in concrete reduces the chloride penetration around 33.6% at the age of 180 days. It indicated that CBA could be utilized in concrete to enhance the durability performance of concrete. iv. Concrete containing 10% CBA exhibits around 29.2% lower drying shrinkage as compared to the normal concrete at the age of 180 days.
It was observed that the replacement of cement with CBA offers technical and environmental advantages, which are considered important aspects in sustainable concrete construction.
However, this study considered samples under submerged conditions and future studies could be considered wetting and drying conditions with different proportions of CBA.
Acknowledgements The writers also gratefully acknowledge the support of Mehran University of Engineering and Technology, Pakistan.
Materials availability Locally available materials were used in this study.
Author contributions • Sajjad Ali Mangi: Preparation of original draft, experimental performance, analysis.

Declarartions
Ethics approval All ethical standards have been followed during this research.
Consent to participate Not applicable.
Consent to publish Not applicable.

Conflict of interest
The authors declare no competing interests.