Assessing the strength and durability behaviour of concrete enhanced with kaolin clay and Shea nut shell particles

Abstract

The aim of this study was to investigate the individual and combined effect of kaolin clay (KC) and shea nut shell particles (SSP) on the physical, strength, and durability properties of concrete. The cement content was replaced with up to 20% KC, and the fine aggregate content was replaced with up to 40% SSP. The concrete samples were cured for 7, 14, 28, and 90 days, and tested. The total S_iO₂+Al₂O₃+Fe₂O₃ content of the KC studied was 88.23%; hence, it possesses pozzolanic properties. Initial and final setting times of the KC pastes steadily increased to 154 min and 325 min at the 20% KC representing 37.5% and 14.4% increase over the control. The control mix obtained a workability value of 102 mm, and decreased steadily to 68 mm at the 20% KC and 40% SSP concrete mix, representing 50% reduction. Maximum compressive and split tensile strength values of 25.97 N/mm² and 4.14 N/mm² were obtained at the 15% KC and 30% SSP concrete, representing 9% and 5% increase over the control concrete respectively in 90 days curing. Furthermore, concrete with 15% KC and 30% SSP recorded 61.5% and 22.1% decrease in water absorption and sulfate attack, over the control concrete respectively in 90 days curing. It is concluded that the addition of KC and SSP positively influenced the physical, strength and durability of concrete, and therefore recommend 15% KC and 30% SSP replacements of cement and fine aggregate respectively for concrete production.

Keywords:

• Kaolin clay
• Shea nut shell
• compressive strength
• split tensile strength
• water absorption

Reviewing Editor:

• Ian Phillip Jones
• University of Birmingham
• United Kingdom of Great Britain and Northern Ireland

Subjects:

• Building and construction
• Architecture
• Sustainable Development

1. Introduction

The United Nations has reported that the world population will witness a high growth of up to 9.7 billion in 2050 and 11.20 billion in 2099 (United Nations Report, 2020). The report further indicates that 55 percent of the population lives in urban communities and that 90 percent of urban population growth is projected to occur in Asia and Africa, particularly in India, China and Nigeria. According to Bebr et al., (2021), there is a housing deficit of 268 million housing units affecting 1.26 billion people globally. They further indicated that an estimated 40 million housing units will have to be added to provide adequate housing for the growing population by 2050.

Figures released by the Ghana Statistical Survey Department (Ghana Statistical Service 2022) indicate that Ghana’s population stood at a little over 31 million in 2020, and an average population growth of 2.2% a year, with 56.7% of the population living in urban centres thereby widening the gap in housing development. According to the Report, Ghana has a housing deficit of 1.8 million units and will require an estimated amount of 2.5 billion US Dollars to provide a minimum of 170,000 housing units annually to close the gap.

Globally, approximately 17.5 billion metric tons of concrete are used annually, in which the cement and aggregate components consist of approximately 2.6 billion metric tons and 13 billion metric tons respectively (Wu et al., 2018). This massive use of concrete largely affects the overall cost of housing provision and leads to environment problems as a result of cement production and aggregate extraction (Amankwah, Bediako & Kankam, 2015; Botchway & Masoperh, 2019). Cement production is predicted to generate approximately 10 - 15% of global carbon dioxide emissions if the production trend continues (Schneider et al., 2011). Currently, the primary aim of the building industry is to minimize the construction costs and environmental pollution resulting from cement production and aggregate extraction. For instance, it has been established that the use of blended cement and nut shell aggregate can reduce the cost of concrete production by up to 7% (Danso-Boateng, 2021) and between 12% and 16% carbon emissions and environmental degradation (Bediako, 2015).

As a result extensive research on cement and aggregate alternatives has been conducted over the past thirty years. For example, numerous research efforts have been made in the use of clay pozzolana (Bediako et al., 2017; Sarfo-Ansah et al., 2014) and agricultural/industrial based wastes such as gum arabic powder (Zakka et al., 2015), shea nut shell ash (Tsado et al., 2014; Zievie et al., 2016), locust bean pod ash (Yalley, 2019), cereal floors (Alfa & Adeleke, 2021) and nano-plastics (Yalley, 2021) to partially replace cement in concrete production. Although the findings indicate that the use of clay and agro-based waste ash pozzolanas as partial replacements for cement improved concrete properties, these studies concentrated on short-term strength properties.

In other studies, agro-based shells and industrial waste particles such as oyster shells (Yang et al., 2005), groundnut shells (Sada et al., 2013), sheet glass (Sankh et al., 2014), coconut shells (Ramadhansyah et al., 2016), walnut shells (Kamal et al., 2017; Husain et al., 2017), corncob ash particles (Memon et al., 2019), rubber seed shells (Wang et al., 2021) were used to partially replace fine aggregates in concrete production. In all these studies, the strength and durability properties improved marginally with further addition and significantly with an extended curing duration of 28 days. However, durability and long-term (90 days) effects have not been studied. Construction waste materials have also been studied as fine aggregate alternatives. For instance, Keerthinarayana and Scrinivasan (2010) used crushed waste bricks from foundry bed and walls as partial replacement for fine aggregate in concrete production. They found that compressive strength increased up to the 25% replacement compared to the control concrete at the 7, 14 and 28 days curing. Mageswari and Vidivelli, (2010) partially replaced natural sand with sheet glass powder and reported an increase of compressive and split tensile strengths at the 10% replacement at the 28 and 45 days curing.

Presently, building material researchers are studying the coupling effect of pozzolanic materials and nut shell particles as partial replacements for cement and aggregate, respectively, to produce more economical concrete with enhanced multiple properties (He et al., 2020). Concrete produced with a combination of these materials not only has reduced cost but also shows good thermal properties and acceptable strength and durability performance. For example, studies have found that partially replacing cement with rice hush ash and fine aggregate with rubber seed shell fine particles improved the lightweight and durability properties of concrete (Momtazi & Zanoosh, 2011). Also replacing 10% cement with fly ash and 10% fine aggregate with coconut shell particles not only resulted in a lightweight concrete but also improved the strength and durability properties (Udaykumar et al., 2014; Rao et al., 2015). Other studies have also reported that replacing 10% cement with groundnut shell ash and 50% fine aggregate with quarry dust produced lightweight concrete with improved strength and durability characteristics (Kumar & Lemessa, 2017). Copper slag ash and ceramic waste fine particles (Chitra & Mohan, 2017) and cow dung ash, and cow dung fine particles (Khan et al., 2022) as partial replacements for cement and fine aggregate in concrete. In all of these studies, the concentration was based on the use of industrial and agro-based waste powder or ash as the pozzolanic material.

The durability of concrete materials and elements is related to their ability to resist both external (temperature and humidity variations, sulfate or chloride attack, etc.) and internal damage. Recently, aspects of concrete strength and durability performance have become a major subject of discussion, especially when concrete is subjected to a severe environment. The corrosion of steel bars is the main factor influencing both concrete strength and durability. The corrosion products of the steel reinforcement expand, developing high pressures within the concrete, which causes cracking and spalling of the concrete cover and exposes the rebar to further corrosion activity. It is possible to influence concrete characteristics to control the project strength and durability by including supplementary cementitious and aggregate materials. These materials originate from several sources and have different effects on the properties of fresh and hardened concrete. The addition of pozzolanic materials to Portland cement and nut shells to fine aggregates is an effective way to reduce concrete costs and prevent environmental pollution (Agopyan et al., 2005). Cement based polymers and nut shell aggregates tend to dilute the alkaline content in the cement, thereby improving the durability of the concrete natural material composite (Pimentel, 2006). There has been an increase in research on the usage of cement-based polymers and agricultural nut shells, which are economically and environmentally friendly.

Although, clay materials have been extensively studied in Ghana as potential cementitious materials in concrete production, the scope has been on the red-loamy clay in southern Ghana. There is little information or literature on the potential use of brown-loamy kaolin clay in northern Ghana as an enhancer of concrete properties. Studies have shown that shea nut shell ash possesses pozzolanic reactive properties and can be used as a cementitious material for concrete production (Tsado et al., 2014; Zievie et al., 2016) and as clay stabilizer for the production of fired clay bricks (Adazabra et al., 2017). However, their use as fine aggregates in concrete production has not been investigated. Again, the authors have not yet read any information on the coupled use of clay and shea nut shell fine-grained particles for concrete production. This central research gap justifies the need for this study. For concrete cost reduction and environmental sustainability, this study evaluates the individual and coupled effects of brown-loamy kaolin clay and shea nut shell fine-grained particles as partial replacements for cement and fine aggregate, respectively, on the strength and durability of concrete.

According to Ghana Geological Survey Department, (2009) the land area of Ghana is 238,535 km². Greater part of this land area, according to Bediako, (2015), and Amankwah and Suglo, (2020) possessed clay deposits estimated at about 1500 billion metric tons. Again, Naangmenyele et al. (2020) reported that Ghana is the fourth leading producer of shea nut worldwide, at about 200,000 metric tons annually, and processes 15,000 tons of the nuts into butter for the international market while 70,000 metric tons are processed for the local market. Shea trees occupy about 77,670 km² of land area in the northern and middle belt of Ghana, and produce over 300,000 metric tons of the nut shells as wastes (SNV Netherlands Development Organisation, 2011; Adazabra et al., 2017). With the abundance of clay deposits and high annual processing of shea nut into butter and nut shell waste generation, it is obvious that clay and shea nut shells will be in abundance, and as such, cost savings in their use as cement and aggregate in concrete production, will be achieved. Thus, the motivation for using the koalin clay and shea nut shell particles is based on their availability, low cost and sustainability.

2. Materials and Methods

2.1. Materials

The constituent materials used in the production of the concrete samples in this study were cement, water, kaolin clay, shea nut shell particles, fine aggregates and granite stones.

2.1.1. Cement

The cement used was ordinary Portland cement grade 32 N conforming to BS EN 197-1 (2011) specifications produced locally by Ghana Cement Company Limited (GHACEM).

2.1.2. Water

Potable tap water free from impurities and chemicals supplied to the laboratory by the Ghana Water Company Limited, which conformed to BS EN 1008 (2002), was used to mix the constituent materials.

2.1.3. Kaolin clay and shea nut shells

Kaolin clay was obtained from a large clay deposit in Charia and the shea nut shells sourced from a local shea butter extraction waste dumping site in Dokpong, all in the Wa Municipality, Ghana. The clay and shell particles conformed to BS EN 8615-2 (2019) and BS EN 12620 (2019) requirements respectively.

2.1.4. Aggregates

Good quality river sand and crushed granite stones with an average size of 10 mm, conforming to BS EN 12620 (2019), were used.

2.2. Testing Methods and Procedures

2.2.1. Kaolin clay (KC) Preparation

The kaolin clay lumps shown in Figure 1 were air-dried to a constant weight in a laboratory environment before grinding to break up the agglomerates using a Thomas grinding machine. The kaolin clay was further pulverized into a very fine particle size (45-150 microns) using a cone crusher and a vibratory mill. The clay was passed through a sieve with an aperture size of 63 µm. The powdered kaolin clay sample shown in Figure 2 was sent to the Building and Road Research Institute (BRRI) of the Council for Scientific and Industrial Research (CSIR-Ghana) Laboratory, Kumasi for the strength activity index (SAI) and, chemical and physical property analysis.

Figure 1. Kaolin lay lumps.

Figure 2. Milled kaolin clay (KC) powder.

2.2.2. Shea nut shell particles (SSP) Preparation

Shea nut shells (Figure 3) were preconditioned by washing to remove all dirt and loose materials. They were then air-dried in the laboratory for three days before they were broken into pieces and manually ground into small fine-grained pieces using a hammer and a local grinding stone, and then sieved through a 4.750 mm BS sieve. The fine-grained particles that passed through the 4.750 mm sieve (Figure 4) were then tested for specific gravity, water absorption, and fineness.

Figure 3. Shea nut shells.

Figure 4. Shea nut shell particles (SSP).

2.2.3. Aggregates quality assessment

The aggregates were tested in accordance with the BS EN 12620 (2019) specifications for the following properties: organic impurities, specific gravity, water absorption, fineness modulus, grain size distribution, abrasion strength, flakiness and elongation indices.

2.2.4. Mix design

The cement was replaced with kaolin clay based on the weight of the cement. The samples were labelled as A₀ for the control sample, and K_x denotes samples with x% kaolin clay. The fine aggregate was also replaced with shea nut shell particles by weight of fine aggregate and labelled as A₀ for the control sample and S_y for samples with y% of shea nut shell particles. To study the combined effect of the two additives, the cement and fine aggregate were replaced with kaolin clay and shea nut shell particles using A₀ for the control sample and K_xS_y for samples with x% kaolin clay and y% shea nut shell particles. Based on trial mixes and with a targeted cube strength of 25 N/mm² a binder-to-aggregate ratio of 1:2:4/0.55 (binder: sand: stones/water/binder ratio) was used to prepare the samples. Concrete materials were mixed using a concrete mixer. The fine aggregate, cement, and kaolin clay were first mixed in a dry state to form a uniform mixture before the granite stones were added. Clean water was added in two phases and mixed to a uniform color and consistency. Twelve cubes of size 150 mm for compression strength, water absorption and sulphate attack tests, and twelve cylinders of diameter 150 mm and height of 300 mm for split tensile test were cast for each replacement level. In all a total of 260 number cube and cylinder specimens were produced.

2.2.5. Test Procedures

2.2.5.1. Physical properties

To determine the consistency and setting times of the mix, the Vicat Method, using the Vicat Apparatus was employed in line with the BS EN 196.3 (2000) specifications. Test blocks were prepared and tested for initial and final setting times using the initial and final set needles, respectively. The workability was determined by measuring the slump using a conical mold in accordance with BS EN 12350-2 (2019).

2.2.5.2. Strength properties

The prepared samples were cured in water for 7, 14, 28, and 90 days and tested. The compressive strength test was performed by gradual application of load on the cubes until failure using a compression testing machine in line with BS EN 12390-3 (2019) specifications. The test setup and destructive modes are shown in Figures 5 and 6. Again, the split tensile strength test was performed by applying a steady load using a universal compression testing machine in conformity with BS EN 12390-6 (2019) stipulations as showcased in Figures 7 and 8.

Figure 5. Cube being crushed.

Figure 6. Destructive mode of cubes.

Figure 7. Cylinder being weighed.

Figure 8. Cylinder being crushed.

2.2.5.3. Durability test

For the water absorption test, the cured cubes were dried to constant weight (M₁) and then immersed in water for 24 hours. The cubes were removed and, cleaned, and their weights were measured again (M₂). The differences in weight before and after immersion were determined, and the percentage of water absorption was calculated.

The sulfate attack test was also performed after the cubes were cured in clean water for 28 days and dried to constant weights (M₁). The cubes were cured in a water tank containing 5% magnesium sulfate (M_gSO₄) solution. The cubes were weighed again after removal from the immersed solution (M₂), and the degree of sulfate attack was measured by the percentage loss of weight at 7, 14, 28, and 90 days of immersion.

The data obtained from various experimental studies were analyzed based on descriptive statistics using the Statistical Package for Social Sciences Version 16.0. Tables and graphs displaying the values and means for the additives and curing durations are used to explain the results of the analysis.

3. Results and discussions

3.1. Materials properties

The chemical properties of the kaolin clay (KC) studied are listed in Table 1. The total S_iO₂ + Al₂O₃ + Fe₂O₃ (88.23%) was higher than the minimum of 70% specified for clays that produce pozzolanas (BS EN 8615-2, 2019). The S_iO₂ content of 53.39% exceeded the minimum limit of 49%, whereas the SO₃ and LOI contents were below the minimum and maximum percentage values of 0.1% and 10%, respectively, indicating that the clay is chemically stable as a pozzolana. Again, the S_iO₂ percentage values recorded fall within the range of 55 – 70% specified for clays that yield high strength (Wu et al., 2017). The total S_iO₂ + Al₂O₃ + Fe₂O₃ content of the brown-loamy kaolin clay used fall within the range of the total S_iO₂ + Al₂O₃ + Fe₂O₃ content of the red-loamy clays studied in southern Ghana. The total S_iO₂ + Al₂O₃ + Fe₂O₃ content of Tanoso clay was 84.68% (Amankwah et al., 2015), Nyamebekyere clay 90.45% (Bediako et al., 2017), Afari clay 80.11%, Mfensi clay 89.85% (Endene et al., 2020) and, clay from an unknown deposit 93.16% (Boakye & Khorami, 2023).

Table 1. Chemical properties of kaolin clay used (%).

SiO2	Al2O3	Fe2O3	CaO	MnO	MgO	N a2O	K2O	P2O5	SO3	LOI
53.39	25.91	8.93	4.56	0.07	3.50	1.75	1.60	0.26	0.01	0.02

The physical properties of the kaolin clay, shea nut shell particles, and aggregates are presented in Table 2. The kaolin clay has a high percentage of fines, 61.2% of its particle size is less than 90 µm and according to Atterberg’s limits it is classified as clay with intermediate plasticity. The kaolin clay properties conformed to the BS EN 8615-2 (2019) specifications. Again, the grading of the shea nut shell particles is good for concrete production, as 35.4% of its particles pass through the 0.600 mm sieve. The aggregate properties were within the established BS EN 12620 (2019) suitability limits.

Table 2. Summary of materials physical properties.

Property	Clay	Nut shells	Sand	Stones	BS EN Ref
Bulk specific gravity	__	2.47	2.60	2.56	2.38 – 2.75
Apparent specific gravity	__	__	__	2.65	2.38 – 2.75
Specific gravity	3.03	__	__	__	__
Liquid limit	43	__	__	__	__
Plastic limit	25	__	__	__	__
Plasticity index	18%	__	__	__	__
Water absorption	__	1.41	14%	2.3%	≤ 20%
Fineness modulus	__	3.0	3.2	6.60	2.0 – 8.0
Abrasion strength	__	__	__	26%	≤ 40%
Flakiness index	__	__	__	12.7%	≥ 15%
Elongation index	__	__	__	9.7%	≥ 10%

3.2. Physical properties of KC and SSP concrete

3.2.1. Setting times

The control mortar paste (A₀) exhibited the lowest consistency value of 35.1%, as shown in Table 3. Beyond this, the value increased consistently with increasing kaolin clay content to 37.5% for the mortar paste with 20% kaolin clay content (K₂₀). A similar occurrence was reported by Amankwah et al., (2015), where the consistency of paste increased as the calcined clay content increased. This shows that the kaolin clay needs a higher water content to form a workable mix. The initial and final setting times of the paste also increased steadily with further addition of the kaolin clay to 154 min and 325 min at the 20% KC compared to 112 min and 284 min obtained by the normal pastes, representing 37.5% and 14.4% increased respectively. This development was expected because the kaolin clay had high silica (53.39%) and fines (61.25%) contents, which might have influenced the steady increase in both the initial and final setting times of the paste.

Table 3. Effect of KC on consistency and setting times.

Sample	Cement (%)	Kaolin clay (%)	w/b ratio	Consistency (%)	Setting times (mm)
					Initial	Final
A0	100	0	0.55	35.1	112	284
K5	95	5	0.55	35.8	123	297
K10	90	10	0.55	36.2	132	308
K15	85	15	0.55	36.9	145	317
K20	80	20	0.55	37.5	154	325

3.2.2. Workability

From the results shown in Table 4, the concrete mixes with kaolin clay (KC) content experienced a steady decline in slump values as the kaolin clay content increased compared to the control mix. Similar observations of slump decline were made in the works of Amankwah et al., (2015), and, Boakye and Khorami, (2023), who studied the effect of clay pozzolana on the slump behavior of concrete mix. In contrast, concrete mixes with shea nut shell particle (SSP) content exhibited a steady increase in slump up to 147 mm at the 40% (S₄₀) replacement level, more than the 114 mm recorded by the normal mix. This outcome is consistent with the work of Oyebisi et al., (2019), who also found that the slump values of a cashew nut shell particles concrete mix increased from 30 mm to 70 mm as the substitution of fine aggregate increased from 0% to 20%.

Table 4. Effect of KC and SSP on concrete slump.

Mix	Cement (%)	Kaolin clay (%)	w/b ratio	Slump height(mm)	Collapsed height (mm)	Degree of workability
A0	100	0	0.55	114	186	High
K5	95	5	0.55	103	197	High
K10	90	10	0.55	94	206	Medium
K15	85	15	0.55	87	213	Medium
K20	80	20	0.55	76	224	Medium
S10	100	0	0.55	119	181	High
S20	100	0	0.55	123	177	High
S30	100	0	0.55	131	169	High
S40	100	0	0.55	147	153	High
K5S10	95	5	0.55	102	192	High
K10S20	90	10	0.55	95	205	Medium
K15S30	85	15	0.55	84	216	Medium
K20S40	80	20	0.55	68	232	Medium

3.3. Strength properties of concrete

3.3.1. Compressive strength of concrete Enhanced with KC

From the results shown in Figure 9, it is observed that, the concrete cubes exhibited a steady appreciation in compressive strength as the kaolin clay (KC) content inclusion was increased to 15% cement replacement and from 7 to 90 days of curing. After seven days of curing, concrete cubes with 15% kaolin clay content obtained the highest compressive strength value of 18.47 N/mm² and which increased to 33.99 N/mm² after 90 days of curing, as compared to the control and other experimental concrete cubes. For all curing durations, the compressive strength values increased consistently up to a clay content of 15% and declined with further addition. The increase in compressive strength from 7 to 90 days of curing age was significantly high in all replacement levels. This result was anticipated because the kaolin clay worked as a filler material owing to its high percentage content of fines (61.2%), as well as pozzolana due to its high silica content (53.39%). The fine particles created more filling capacity, which promoted early strength development, whereas the high silica content led to late strength development owing to the pozzolana reactivity of the kaolin clay polymer. Amankwah et al., (2015) observed similar characteristics of the influence of clay pozzolana on the strength behavior of concrete samples.

Figure 9. Variation of compressive strength with KC content and curing duration.

3.3.2. Split tensile strength of concrete Enhanced with KC

The split tensile strength behavior exhibited a trend similar to that of the compressive strength. For instance, in Figure 10 the maximum split tensile strength value obtained in 7 days curing duration was 2.94 N/mm² at the 15% kaolin clay (KC) content. In 90 days curing duration, a maximum split tensile strength of 5.41 N/mm² was again obtained at the 15% kaolin clay content. The split tensile strength values progressively increased in all replacement levels up to the 90 days curing duration but declined beyond the 15% kaolin clay content. This result was once again anticipated because the kaolin clay worked as a filler material owing to its high percentage content of fines (61.2%), as well as pozzolana due to its high silica content (53.39%). The fine particles created more filling capacity, which promoted early strength development, whereas the high silica content led to late strength development owing to the pozzolana reactivity of kaolin clay polymer. Amankwah et al., (2015) observed similar characteristics of the effects of type one and two clay pozzolanas on the split tensile strength of concrete. They reported that the split tensile strength significantly increased with curing duration for all grades of cylinders.

Figure 10. Variation of split tensile strength with KC content and curing duration.

3.3.3. Compressive strength of concrete Enhanced with SSP

The compressive strength decreased steadily with further inclusion of the shea nut shell particles (SSP) across all curing durations for the experimental cubes, but it showed a consistent increase as the curing duration was prolonged (Figure 11). For instance, in 7 days the control cubes recorded a compressive strength value of 16.38 N/mm² and this dropped to 14.17 N/mm² at the 40% shea nut shell particles content. In 90 days the control cubes compressive strength increased to 32.62 N/mm² whiles the 40% shea nut shell particles content cubes compressive strength also increased to 27.50 N/mm². This loss of strength with increased shea nut shell content and appreciation with prolonged curing duration can be linked to the smaller unit weight of the nutshell particles owing to the lower specific gravity (2.47) compared to that of the fine aggregate (2.60) coupled with the development of more pores owing to the particle shape. Again, more particles (35.4%) passed through the 0.600 mm sieve, which is responsible for the substantial increase in strength as the curing duration is prolonged. This result is in agreement with those of earlier studies. Rao et al., (2015) incorporated up to 20% of coconut shell particles as a partial substitution for fine aggregates in concrete production. They observed that the control sample obtained the maximum strength, followed by a gradual reduction to the 20% substitution level. Notwithstanding, the compressive strength values recorded at all replacement levels at 28 and 90 days of curing satisfied the 28 curing days minimum compressive strength of 25 N/mm² stipulated in the BS EN 12390-3 (2019) for normal concrete.

Figure 11. Variation of compressive strength with SSP content and curing duration.

3.3.4. Split tensile strength of concrete Enhanced with SSP

The split tensile strength also witnessed a steady decline from the control to the experimental concrete cylinders as the shea nut shell particle (SSP) content increased, and a consistent appreciation as the curing duration was prolonged (Figure 12). The control concrete cylinders recorded a split tensile strength of 2.59 N/mm² in 7 days and increase to 5.19 N/mm² after 90 days curing. In a similar trend, the 40% shea nut shell particle content cylinders recorded a split tensile strength value of 2.26 N/mm² in 7 days and increased to 4.40 N/mm² in 90 days curing duration. This loss of strength with increased nutshell content and increase in strength with longer curing durations once again, is attributable to the smaller unit weight of the nutshell particles due to the lower specific gravity (2.47) compared to that of the fine aggregate (2.60) coupled with the development of more pores due to the particle shape. Again, more particles (35.4%) passed through the 0.600 mm sieve, which is responsible for the substantial increase in strength as the curing duration is prolonged. This result is in agreement with those of earlier studies. Modani and Vyawahare, (2013) employed sugarcane bagasse waste fines as a partial replacement for fine aggregates in their study. They observed that the split tensile strength of concrete cylinders decreased with the addition of sugarcane bagasse waste fines as compared to the control concrete cylinders and increased as the curing duration was prolonged, which fits perfectly well with the present result obtained.

Figure 12. Variation of split tensile strength with SSP content and curing duration.

3.3.5. Compressive strength of concrete Enhanced with KC and SSP

The concrete test samples attained the maximum compressive strength values at the 15% kaolin clay (KC)-cement replacement and 30% shea nut shell particles (SSP)-fine aggregate replacement level (K₁₅S₃₀) over the control test cubes in all curing durations (Figure 13). Beyond the (K₂₀S₃₀) replacement level, the compressive strength values decreased. The marginal strength appreciation with an increase in the kaolin clay polymer and shea nut shell particle content and subsequent drop beyond the K₁₅S₃₀ test samples, as well as the substantial strength appreciation with prolonged curing duration, was expected. The nutshell particle content reduced the rate of strength development with an increase in the replacement levels, whereas the kaolin clay inclusion accelerated the strength gain as the curing duration increased owing to the high silica content. In a previous study using fly ash and coconut shell particles as cement and fine aggregate replacements, Rao et al., (2015) reported that 10% fly ash and cement and 20% coconut shell/fine aggregate replacement levels produced the maximum strength after 28 days of curing.

Figure 13. Variation of compressive strength with KC/SSP and curing duration.

3.3.6. Split tensile strength of concrete Enhanced with KC and SSP

The test cylinders obtained the highest split tensile strengths at the 15% kaolin clay (KC)-cement substitution and 30% shea nut shell particles (SSP)-fine aggregate substitution (K₁₅S₃₀) levels compared to the control concrete test samples (Figure 14). The minimal split tensile strength increased with further addition of the kaolin clay polymer and nutshell content, and a subsequent drop beyond the K₁₅S₃₀ concrete test samples, as well as the substantial strength appreciation with long-term curing duration, was anticipated. The shea nut shell particles reduced the rate of strength development with an increase in the substitution levels, whereas the kaolin clay polymer accelerated the strength gained as the curing duration was increased owing to the high silica content in the kaolin clay. This study was inspired by the work of Kumar and Lemessa, (2017), who studied the coupled effect of groundnut shell ash as a partial replacement of cement and quarry dust as a partial replacement of fine aggregate on the split tensile strength behavior. They found that concrete cylinders containing 10% groundnut shell ash and 50% quarry dust obtained the highest split tensile strength after 28 days of curing.

Figure 14. Variation of split tensile strength with KC/SSP content and curing duration.

The analysis of variance (ANOVA) test results for the compressive strength data generated are listed in Table 5. The results showed a significant influence or contribution of the two factors on the compressive strength development behavior. P-values of 0.000 were recorded for both factors, that is, curing duration, F(1,40) = 2311.783, p < 0.05, and kaolin clay/shea nut shell content, F(1,40) = 19.512, p < 0.05. It can be observed that the curing duration contributed more to the strength behavior than the clay/shell content. From earlier discussions on descriptive statistics, this was expected. It is also found that the coefficient of correlation, R² = 0.994 is greater than 0.95. Hence, it is presumed that the model is a statistical fit to predict the compressive strength data because 99.4% of the variability can be explained by the model. This implies that the equation was valid for up to 15% clay and 30% shell content after 90 days of curing. To determine the contribution levels of curing duration and kaolin clay/shea nut shell contents on the split tensile strength behavior of the concrete cylinders, an analysis of variance (ANOVA) test was conducted at a significance level of 0.05. The statistical test results presented in Table 6 recorded P-values of 0.000 for both curing durations, F(1,40) = 1308.125, p < 0.05, and clay/shell content, F(1,40) = 9.366, p < 0.05. Because the variance ratio (F) value for curing duration is much higher (1308.125) than that of the clay/shell content (9.366), this implies that the curing duration contributed far more to the split tensile strength data than the clay/shell content. Notwithstanding, the R-square statistic generated by the model indicates that 99.0% (R² = 0.990) of the variation in the split tensile strength behavior can be linked to the curing duration and clay/shell content. With this high explanatory power of the model, the reliability of the data could be described as excellent and valid up to the 15% kaolin clay and 30% shea nut shell (K₁₅S₃₀) replacement level after 90 days of curing.

Table 5. ANOVA analysis of compressive strength of KC and SSP cubes data (N/mm²).

Source	Sum of squares	df	Mean square	F-value	P-value
Model	2323.790^a	19	122.305	369.653	0.000
Intercept	37914.640	1	37914.640	114593.054	0.000
Duration	2294.652	3	764.884	2311.783	0.000
KC/SSP	25.824	4	6.456	19.512	0.000
Dura. & KC/SSP	3.314	12	0.276	0.835	0.615
Error	13.235	40	0.331
Total	40251.664	60
Corrected total	2337.024	59

a. R² = 0.994 (Adjusted R² = 0.992).

Table 6. ANOVA analysis of split tensile strength of KC and SSP cubes data (N/mm²).

Source	Sum of squares	df	Mean square	F-value	P-value
Model	57.639^a	19	3.034	208.929	0.000
Intercept	963.283	1	963.283	66341.791	0.000
Duration	56.982	3	18.994	1308.125	0.000
KC/SSP	0.544	4	0.136	9.366	0.000
Dura. & KC/SSP	0.113	12	0.009	0.651	0.786
Error	0.581	40	0.015
Total	1021.503	60
Corrected total	58.220	59

a. R² = 0.990 (Adjusted R² = 0.985).

A correlation analysis between the split tensile strength and compressive strength of concrete cylinders and cubes produced with kaolin clay and shea nut shell particle contents was performed, and the results are shown in Figure 15. It appears from the results that there is a very good relationship between the split tensile strength and compressive strength of the concrete samples, with a coefficient of determination (R²) of 0.997. The two strength properties (split tensile strength and compressive strength) are positively influenced by the inclusion of kaolin clay/shea nut shell contents and curing duration.

Figure 15. Relationship between compressive strength and split tensile strength.

3.4. Durability properties of concrete

3.4.1. Water absorption of concrete Enhanced with KC

From the results shown in Figure 16, the concrete cubes without kaolin clay (KC) content absorbed more water than the experimental cubes. It was also noticed that the water absorption percentage values declined steadily from the control concrete cubes to the 20% kaolin clay content concrete cubes for all curing durations. This gradual reduction in water absorption as the kaolin clay polymer content increased can be attributed to the high fine percentage, which promoted densification, thus making the cubes less permeable. This trend of decreasing water absorption with increasing pozzolanic material content and prolonged curing durations has been observed by other researchers. For instance, Ferraro and Nanni, (2012) conducted a similar study and found that the rice husk ash pozzolana content in concrete steadily decreased water absorption up to the 15% substitution level across all curing durations. They attributed this development to the high fines content of rice husk ash compared to that of cement.

Figure 16. Variation of water absorption with KC content and curing duration.

3.4.2. Sulfate attack of concrete Enhanced with KC

The sulfate attack of the kaolin clay (KC) cubes showed a similar trend to the water absorption behavior. From the results plotted in Figure 17, the weight loss of the concrete cubes as a result of sulfate attack also decreased with further addition of the kaolin clay and increased with prolonged immersion duration. This steady decrease in weight loss as the kaolin clay content increased can again be attributed to the high fine percentage, which promoted densification, thus making the cubes less permeable to the magnesium sulfate solution. Sarfo-Ansah et al., (2014) observed similar weight-loss features. They found that the inclusion of clay pozzolana as a partial substitution of cement caused a drop in sulfate attack in the concrete cubes up to the 25% cement replacement level, and an increase in sulfate attack as the immersion duration was prolonged.

Figure 17. Variation of sulfate attack with KC content and curing duration.

3.4.3. Water absorption of concrete Enhanced with SSP

The results indicate a steady decline in the water absorption percentage of concrete cubes with shea nut shell particles (SSP) content over that of the control cubes as the shea nut shell particles content and curing duration increased (Figure 18). This developmental pattern can be attributed to the grinding and fine nature of the shea nut shell particles, as well as their low water absorption behavior. The high filling capacity and low permeability of the nut shell particles decrease the water absorption rate of the concrete cubes. This result confirms the outcome of previous studies that used ago-based nut shell particles as a fine aggregate replacement. In a recent study using groundnut shell particles as a fine aggregate replacement, Buari et al., (2019) reported a decrease in water absorption percentage from 2.67% to 2.24% as the replacement level increases to 40% addition of the groundnut shell particles.

Figure 18. Variation of water absorption with SSP content and curing duration.

3.4.4. Sulfate attack of concrete Enhanced with SSP

The sulfate attack, as a result of the weight loss percentage of the shea nut shell particles (SSP) concrete cubes, increased marginally with increasing shea nut shell content and prolonged curing duration, as shown in Figure 19. It was observed that the concrete cubes recorded lower weight losses in 7 days immersion periods. Weight loss then increased steadily with increasing shea nut shell particle content and immersion duration up to the 40% replacement level and 90 days. This behavior can again, be linked to the grinding and fine nature of the shea nut shell particle as well as their low water absorption behavior. The high filling capacity and impermeable nature of the nutshell particles influenced the marginal increase in the rate of sulfate attack on the cubes. This trend has also been observed by other researchers who used agricultural waste nutshells as fine aggregates. In a previous study, Yang et al., (2010) found that oyster shell particles gradually increased weight loss in concrete cubes as the fine aggregate substitution and cube immersion duration in the solution increased.

Figure 19. Variation of sulfate attack with SSP content and curing duration.

3.4.5. Water absorption of concrete Enhanced with KC and SSP

The results plotted in Figure 20 indicate a consistent decrease in the water absorption percentage values from the control concrete cube (A₀) to the K₂₀S₄₀ experimental concrete cubes. The results clearly show that water absorption decreases with further addition of the kaolin clay and shea nut shell particles, as well as with curing duration. This behavior could be linked to the lower free lime content in the kaolin clay and the greater release of free lime in the cement owing to prolonged immersion in the magnesium sulfate solution. In an earlier study using metakaolin and rice husk as cement and fine aggregate partial replacement, Chatveera and Lertwattanaruk, (2009) noticed a substantial decline in water absorption of experimental concrete cubes compared to control concrete cubes as the percentage addition and curing duration increased.

Figure 20. Variation of water absorption with KC/SSP content and curing duration.

3.4.6. Sulfate attack of concrete Enhanced with KC and SSP

The sulfate attack as a result of the weight loss percentages of the concrete cubes is presented in Figure 21. Sulfate attack decreased with an increase in the kaolin clay (KC) and shea nut shell particle (SSP) content and increased with prolonged immersion duration. For example, in 7 days immersion period, the weight loss of the control cubes was 0.17%, which decreased to 0.13% at the K₂₀S₄₀ replacement level. After 90 days of immersion, the control cubes exhibited 2.46% weight loss and 1.81% weight loss at the K₂₀S₄₀ replacement level. In a review of clay work, Mousavi et al., (2021) highlighted that clays calcined at the right temperatures had a positive impact on the resistance to sulfate attack by concrete. This behavior could be linked to the lower free lime content in the kaolin clay and the greater release of free lime in the cement owing to prolonged immersion in the magnesium sulfate solution. Again, the sulfate attack behavior might have also been greatly influenced by the high acid content of the shea nut shell fine particles. A similar occurrence was observed by Sarfo-Ansah et al., (2015), where the combined effect of calcined clay, steel slag, and granite dust decreased concrete chemical attack more than calcined clay or steel slag alone.

Figure 21. Variation of sulfate attack with KC/SSP content and curing duration.

The analysis of variance (ANOVA) test results shown in Table 7 were generated and analyzed to determine the contributions of the additives and curing duration on the water absorption data. From the results, the curing duration, F(1,40) = 74.119, p < 0.05, and clay/nutshell content, F(1,40) = 125.958, p < 0.05, contributed significantly to the water absorption data. It is also evidenced that the kaolin clay and shea nut shell particles addition contributed to the water absorption data more than curing duration. Again, the R-square value of (R² = 0.950) shows that 95.0% of the variation in the water absorption percentage data can be explained by the model. This suggests that the incorporation of the kaolin clay and shea nut shell particles as partial replacements of cement and fine aggregate resulted in a significant reduction in the water absorption percentage at all replacement levels across all curing durations.

Table 7. ANOVA analysis of water absorption of KC and SSP cubes data (%0).

Source	Sum of squares	df	Mean square	F-value	P-value
Model	213.202	19	11.221	40.087	0.000
Intercept	3278.056	1	3278.056	11710.551	0.000
Duration	62.243	3	20.748	74.119	0.000
KC/SSP	141.034	4	35.259	125.958	0.000
Dura. & KC/SSP	9.924	12	0.827	2.955	0.005
Error	11.197	40	0.280
Total	3502.455	60
Corrected total	224.399	59

a. R² = 0.950 (Adjusted R² = 0.926).

4. Conclusion and recommendations

The effect of the kaolin clay, shea nut shell fine-grained particle, and both kaolin clay and shea nut shell fine-grained particle combined, on the strength and durability behavior of concrete has been studied. From the experimental study, the following conclusions can be made:

• The study found that the kaolin clay used possessed the right and optimum quantities of chemical compounds that produce pozzolana cement. The clay becomes pozzolanic reactive at an optimum calcination temperature of 800 °C. The inclusion of the kaolin clay in the concrete mix progressively increased setting times and decreased workability. This makes it suitable for concrete production in hot and windy climates as well as situations where high to medium workability is prescribed.

• Shea nut shell could generate high fine-grained particles when properly grounded. It is light in weight, low water absorbing and acidic. The inclusion of the SSP in the concrete mix progressively increased the workability, and because the values are less than 150 mm maximum prescribed by the standards, their use can be beneficial in situations where high workability is required.

• Concrete cubes and cylinders with kaolin clay achieved maximum strength at the 15% cement replacement level representing 7.38% for the compressive strength and 6.5% for the split tensile strength over the control cubes and cylinders. Again, in terms of concrete durability, KC reduced concrete water absorption and sulphate attack at the 15% replacement level by 31.3% and 25% respectively compared to the control concrete at 90 days curing. It is therefore, recommended that 15% kaolin clay should be adopted by concrete producers as optimum replacement for cement in concrete production in areas where delays in setting times are required.

• Though concrete cubes and cylinders with shea nut shell particles content marginally reduced strength properties from the control concrete to 40% replacement level, the progressive cumulated percentage drop is minimal at the 30% replacement compared to the benchmark concrete cubes and cylinders. The reduction of strengths at the 30% shea nut shell particle contents is 8.9% for the compressive strength and 9.1% for the split tensile strength compared to the control samples at 90 days curing. For the durability behavior, SSP reduced water absorption by 26.4% and increased sulphate attack by 21% at the 30% replacement compared to the control concrete at the 90 days curing. For economic reasons, the study recommends 30% shea nut shell particle content for concrete production to concrete manufacturers.

• Concrete cubes and cylinders that have both kaolin clay (KC) and shea nut shell particle (SSP) content combined also achieved better strength properties at the 15% kaolin clay content and 30% shea nut shell particle content. The percentage increase for the compressive strength and split tensile strength is 9% and 5% respectively over the control samples in 90 days curing. Again, durability properties were improved with the combined use of kaolin clay and shea nut shell particles as cement and fine aggregate replacement respectively. Water absorption was reduced by 61.5% and sulphate attack reduced by 22.1% at the 15% kaolin clay and 30% shea nut shell particles contents compared to the control samples.

The ANOVA results revealed that the experimental factors KC, SSP and curing duration contributed significantly on the result outcomes with curing duration performing better in the strength behavior whiles KC, SSP, and combined KC/SSP also performed better in the durability behavior compared to curing duration. Furthermore, with high explanatory power of the model developed, the reliability of the data could be described as excellent and valid up to the 15% KC and 30% SSP replacement at 90 days curing. Hence, it is concluded that the addition of the KC and SSP positively influenced the physical, strength and durability behavior of concrete, and therefore recommend 15% KC and 30% SSP replacements of cement and fine aggregate respectively for concrete production to concrete manufacturers. The separate and combine use of KC and SSP can contribute substantially to construction cost reduction as well as provide some level of relief to environmental pollution arising from cement production processes and shea nut shell waste disposal.

Authors contributions

Zievie Patrick wrote the entire research Thesis from which this paper was produced. Peter Paa-Kofi Yalley, Humphrey Danso and Kwaku Antwi supervised the thesis from conceptualization, literature review, methodology, data analysis, and manuscript writing.

Acknowledgement

This work is part of a PhD Thesis in Construction Technology in the School of Graduate Studies, Akenten Appiah-Menka University of Skills Training and Entrepreneurial Development. We therefore acknowledge the contributions of Staff and Laboratory Technicians of the Building Technology and Estate Management Department of the Dr. Hilla Limann Technical University, Wa and the Ghana Highway Authority Materials Division, Wa.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Raw experimental data were generated at the Building Technology and Estate Management Department laboratory at the Dr. Hilla Limann Technical University, Wa, Ghana. Processed data supporting the findings of this study are available from the corresponding author (Z.P) on request.

Additional information

Funding

The full cost of this research work was borne by the authors.

Notes on contributors

Patrick Zievie

Zievie Patrick is an academic staff with the Building Technology and Estate Management Department of the Dr. Hilla Limann Technical University, Wa, Ghana. He is currently a final year student pursuing PhD programme in Construction Technology at the Akenten Appiah-Menka University of Skills Training and Entrepreneurial Development, Kumasi, Ghana. He is being supervised and mentored by two Professors of the University, Prof. Peter Paa-Kofi Yalley and Prof. Humphrey Danso. Zievie Patrick is a Professional member of the Institution of Engineering and Technology-Ghana (IET-Gh). His research interest areas include production of alternative designs and materials for cost saving rural housing and indigenous construction practices.