Mechanical and Physical Characteristics of Oil Palm Empty Fruit Bunch as Fine Aggregate Replacement in Ordinary Portland Cement Mortar Composites

Abstract

Palm oil empty fruit bunch (OEB) is the largest source of waste in the production of crude palm oil. Utilizing this waste in various applications can help reduce its volume and mitigate adverse environmental effects. In this study, fibers from OEB without any chemical treatment are introduced into Ordinary Portland Cement (OPC)-based mortar to partially replace fine aggregates, aiming to reduce the mortar’s density. The goal of this experimental study is to observe the mechanical and physical performance of the samples according to the effect of the addition of OEB. The composite samples were made by replacing 1%, 2%, and 3% of the weight of quartz sand as the fine aggregate with OEB (fine and coarse). The hardened composites were further tested to determine their compressive strength, and it was found that the replacement of sand with OEB led to a decrease in compressive strength and flowability while alleviating the mortar’s density and affecting the setting time. The decrease in compressive strength was attributed to cavities present in the samples. Flexural tests and 28-day drying shrinkage measurements were carried out on the samples with 1% replacement of sand with OEB. The experiments showed that OEB fibers increased the flexural strength, functioned as a crack barrier, and reduced drying shrinkage.

1. Introduction

Indonesia is the leading country that produces palm oil in the world. In 2022–2023, the total production recorded was 45.5 metric tons. The increasing palm oil production is in line with the production of industrial waste which endangers the surrounding environment. Palm oil processing produces 21% of crude palm oil and 5% of palm kernel oil, and the rest is waste in the form of palm oil empty fruit bunch (OEB), fibers, shells, and other solid waste. OEB is the largest amount of waste, accounting for almost 23% of fresh fruit bunch [1,2].

Currently, the fiber parts of OEB and the shells are used as fuel for boilers, which supply steam for fresh fruit bunch sterilization and turbine rotation. The remaining OEB is returned to the soil due to its high nutrient content [3]. Studies have explored the processing and treatment of OEB into fibers and their applications. OEB is mainly composed of cellulose, hemicellulose, and lignin, with percentages of around 42.7%, 17.1%, and 13.2%, respectively [3,4]. Compared to other natural fibers, OEB has relatively high cellulose and hemicellulose contents (35–45% and 25–40%, respectively) [5]. Through pyrolysis, the cellulose, hemicellulose, and lignin content in OEB can produce biochar, bio-oil, and activated carbon [4,6,7]. OEB can also be used as reinforcement in composites, with treated OEB fibers exhibiting strengths from 52 to 156.3 MPa depending on the treatment, suitable for reinforcement in thermoset or thermoplastic-based composites. Besides its use in polymer matrix composites, OEB also improves the mechanical properties of concrete [2,8,9].

Concrete, a composite of crushed stone and sand (referred to as coarse and fine aggregates), is bound in a cementitious matrix [10]. It is the second most widely used material globally, with an annual consumption of around 30 billion tons, second only to water [11]. Precast concrete is gaining prominence as a superior form of concrete, offering numerous benefits. Its production in a controlled factory setting ensures enhanced quality and consistency with rigorous quality control measures. Precast elements can be manufactured simultaneously with on-site preparations, significantly reducing the overall construction time [12,13,14]. However, due to its density of approximately 2400 kg/m³, precast concrete is heavy and requires the use of heavy machinery [15].

Reducing the density of precast concrete offers several advantages. A lighter structure facilitates easier transportation and installation and diminishes the dead load on buildings, leading to smaller and less costly structural components [16,17]. It often enhances thermal and acoustic insulation, contributing to greater energy efficiency and comfort in buildings [18]. Additionally, lighter precast concrete is generally more resistant to cracking and deformation, which improves the durability and lifespan of structures.

Utilizing OEB as a fiber reinforcement in mortar and concrete can reduce the density of precast concrete, resulting in more lightweight concrete and reducing oil palm waste. Rao and Ramakrishna conducted experiments on the mechanical properties of mortar reinforced with OEB fibers from stalks and spikelets. Their findings included mortar with spikelet fibers exhibiting higher workability compared to stalk fibers; the compressive strength of mortar mixed with OEB being lower than that of mortar without OEB; the flexural strength of mortar with OEB being higher than that of mortar without OEB; and the flexural and split strength of mortar with OEB stalk fibers increasing with the fiber content [2]. Several other studies have shown that the setting time decreases as the amount of OEB increases. The addition of OEB reduces workability, increases split tensile strength, and decreases slump test results in concrete [19,20]. The pre-treatment of OEB before mixing will have different effects on workability, compressive strength, tensile strength, and water absorption [2,20].

According to the standard set by BS EN 206-1, lightweight concrete has an oven-dry specific gravity between 800 kg/m³ and 2000 kg/m³, achieved by partially or completely replacing dense natural aggregates with lightweight aggregates [21]. The lightweight aggregates used in concrete have requirements and specifications that must be met. Generally, lightweight aggregates used in concrete construction have a lighter specific gravity than those often used in general construction. According to BS EN 13055: 2016, an aggregate should have a density of no more than 1200 kg/m³ and consist of two types of aggregates: coarse and fine [22].

In this work, OEB without chemical treatment is chosen as a partial substitute for fine aggregate in Ordinary Portland Cement (OPC) mortar due to its low cost and abundance. To minimize the effect of aggregate variation, no coarse aggregates or crushed stones were incorporated into the samples. Two different lengths of OEB fibers were studied, referred to as fine and coarse OEB, respectively. To study the effect of OEB’s addition into the mortar, the flowability and setting time in the fresh mortar state were measured, while mechanical performance, including compressive and flexural strength, was tested for the cured or hardened samples. Shrinkage was also measured, as the addition of fibers into concrete is known to reduce drying shrinkage.

2. Materials and Methods

2.1. Materials

OEB from the Elaeis guineensis species, received from the plantation PT Perkebunan Nusantara VIII (PTPN VIII) in Indonesia, was sun-dried. At the laboratory, the dried OEB was cut using a chopper and sieved to obtain the desired size. The fine OEB was obtained using mesh number 70, with a particle size of around 212 µm, and the coarse OEB was obtained using mesh number 16, with a particle size of approximately 1.18 mm, in the form of fibers. Images of the fine and coarse OEB, abbreviated as FOEB and COEB, are shown in Figure 1a,b, respectively. Type 1 Ordinary Portland Cement (OPC) was used as the binder for the mortar, with Ottawa quartz sand conforming to ASTM 778 as the fine aggregate or filler, as shown in Figure 1c. Both fibers have a nonuniform length distribution, while the quartz sand has uniform, rounded, translucent particles.

2.2. Methods

2.2.1. Specimen Preparation

The mortar was composed of OPC, water, and quartz sand mixed into a paste. Small amounts of 1%, 2%, and 3% by weight of sand were replaced with OEB, denoted as FOEB1, FOEB2, and FOEB3 for fine OEB and COEB1, COEB2, and COEB3 for coarse OEB. Due to the much lower density and sparsely packed nature of OEB, a higher volume of paste was obtained with an increasing amount of OEB replacement. A cement-to-sand ratio of 1:2.75 by mass was applied according to ASTM Designation C109 [23]: Standard Test Method for Compressive Strength of Hydraulic Cement Mortars using 50 mm cubic specimens. The compositions of the mixtures are provided in

Table 1. Mix design of mortar samples.

Sample	FOEB, g	COEB, g	Quartz Sand, g	OPC, g	Water, g
Control	0.0		1200.0	437.5	219.0
FOEB1	12.0		1188.0	437.5	219.0
FOEB2	24.0		1176.0	437.5	219.0
FOEB3	36.0		1164.0	437.5	219.0
COEB1		12.0	1188.0	437.5	219.0
COEB2		24.0	1176.0	437.5	219.0
COEB3		36.0	1164.0	437.5	219.0

. After mixing, part of the fresh mortar paste was poured into molds for compression and flexural tests, while the remainder was used for flowability and setting time tests.

The compression samples were removed from the molds after 1 day and were left to cure immersed in water for 28 days. After 28 days of water immersion, the cured samples were removed, wiped dry, and left in the air for 1 day before testing. All of the samples wiped dry were weighed before and after the water immersion curing. The same procedure was applied to the flexural test samples. At least 5 samples were tested for compression and flexural strength.

2.2.2. Fresh Mortar Paste Characterization

After mixing, the fresh mortar paste was immediately subjected to flowability measurement according to ASTM C1437 [24]: Standard Test Method for Flow of Hydraulic Cement Mortar, and setting time measurement, referring to ASTM C191 [25]: Standard Test Methods for Time of Setting of Hydraulic Cement by Vicat Needle.

2.2.3. Mechanical and Physical Properties of Mortars

The compressive strength of the 50 mm cube samples was tested according to ASTM C109, and the flexural test was conducted according to ASTM C348 [26]: Standard Test Method for Flexural Strength of Hydraulic-Cement Mortars on samples with dimensions of 40 × 40 × 160 mm. At least 5 samples were tested for each condition. The failure patterns of the compression and flexural samples were observed and analyzed. Drying shrinkage was measured according to ASTM C596 [27]: Standard Test Method for Drying Shrinkage of Mortar Containing Hydraulic Cement on samples with dimensions of 25 × 25 × 285 mm. The length of the samples was measured hourly for the first 24 h after they were released from molds, which occurred 24 h after casting. After the first day, the sample lengths were recorded daily for 28 days. Flexural strength and drying shrinkage tests were only performed on samples with the best compressive strength and the control sample.

Macrostructure observation of the samples was conducted using an optical microscope to inspect the pore and OEB fiber shape distribution on the free surfaces, sidewalls, and fractured surfaces of the samples. The density ($\rho$) of the mortar is calculated by Equation (1), where W is the weight of the sample before curing and V is the volume of the sample.

$$\rho ={\displaystyle \frac{W}{V}}$$

(1)

3. Results and Discussion

3.1. Flowability and Setting Time of Fresh Mortar

The flowability of the mortars containing OEB is illustrated in Figure 2a. The incorporation of OEB into the mortars reduced the flowability of the mortars. The size and surface properties of the fibers significantly affect the flowability of the mortar [22]. The presence of tiny fibers causes a greater decrease in flowability compared to coarse fibers. The current study indicates that OEB can absorb water, and its smaller fibers, which have a larger surface area, absorb more water compared to COEB [28]. The substitution of only 1% by weight of the sand with fine and coarse OEB approximately halved the flowability of the control sample. Substituting greater amounts of OEB decreased the flowability significantly, which indicates a nonlinear correlation between flowability and the surface area of fibers and hence the amount of water absorbed. In addition to being a necessary reactant for hydration, water also serves as the medium for blending and facilitating the movement of fresh paste to fill in the mold.

The Vicat needle test for setting time measures the depth of penetration of a loaded needle in a specified time interval. This measurement method only provides indirect information on the curing or hydration degree [29]. The addition of OEB resulted in a decrease in the setting time of fresh paste, linearly correlated with flowability. Setting time decreased with the amount of OEB, with the fine fibers having a greater impact compared to the coarse fibers, as shown in the graphs in Figure 2b. The mortar without fibers did not set until after about 2 h, giving time for the fresh paste to fill in the mold.

It can be assumed that the decrease in penetration with time into the mortar without fibers was caused by the hardening or curing of the materials, which was slower than the absorption of water as a flowing agent into the mortar with OEB. To test that assumption, the Vicat needle technique was applied to mixtures of cement, sand, and OEB with the same compositions but without water. It was found that full penetration of the Vicat needle was observed in all the mixture compositions regardless of time. It can be concluded that neither OEB nor water absorption obstructed needle penetration and that hydration, although limited, did occur in the mixtures with less water. Similar to flowability, water functioned as a mobility medium. When water was still present, the needle could still move in between the hydration products.

3.2. Compressive Strength

The average compressive strength is tabulated in Figure 3a. The compressive strengths of the control samples, FOEB, and COEB with different weight percentages are 26.9 MPa, 20.4 MPa, 9.1 MPa, 8.6 MPa, 21.8 MPa, 15.3 MPa, and 9.7 MPa, respectively. The compressive strengths are inversely related to the volume of OEB in the mortar samples, which aligns with previous research [2,30]. The coarseness, size, surface area, and grain shape of the fillers affect the compressive strength [31,32,33]. The COEB substitution resulted in a smaller decrease in compressive strength compared to FOEB, with 1 wt.% exhibiting the least decrease in compressive strength, 24% and 19% compared to the control samples, respectively.

The decrease in the strength of FOEB dropped to less than half of the control sample’s strength at 1 wt.% substitution, with no further decrease observed at a higher substitution of 2 wt.%. Conversely, the decrease in strength of COEB was less severe. This may indicate that at the substitution of 2 wt.% FOEB, all the water had been absorbed at the time of mixing, leaving no water for the hydration reaction. It is well understood that hydration products in OPC are the primary contributors to its strength. The lower amount of hydration product and the presence of cavities, which can be observed on the fracture surface of the crushed COEB compression specimen in Figure 3b, were believed to contribute to the strength reduction. Although there is a positive correlation between flowability, setting time, and compressive strength, no direct causal relations could be found. The risk of incomplete filling of the mold caused by low flowability was prevented by vibration during paste pouring, and only perfectly cubic specimens went through testing. Previous studies have suggested that the Vicat needle test is not accurate for determining phase transformation [34] but serves as a practical guide in the application of a material.

Based on these data, COEB1, which had the least decrease in strength, was chosen for further characterization (of shrinkage properties and flexural strength). With a compressive strength of 20 MPa and above, both COEB1 and FOEB1 can be applied for load-bearing structural elements like beams and columns. Less strong but lighter composites can be used for nonstructural applications like walls. The introduction of crushed stone or other types of coarse aggregate will inevitably modify these compositions for OEB-containing concrete.

3.3. Density and Macrostructure

Density was calculated by Equation (1). The replacement of sand with OEB decreased the density, as shown in

Table 2. The averaged weight and density of mortar samples.

Sample	Before Curing, g (a)	Density, g/cm3	After Curing, g (b)	Moisture Take Up, % ((b − a)/a) × 100
Control	264.70	2.12	269.77	1.93
FOEB1	253.17	2.03	257.28	2.03
FOEB2	252.50	2.02	260.65	3.22
FOEB3	242.67	1.94	253.47	4.45
COEB1	257.00	2.06	262.25	2.04
COEB2	255.67	2.05	262.93	2.84
COEB3	247.50	1.98	256.77	4.55

. The fine OEB reduced density more than coarse OEB at the same replacement amount. However, fine OEB caused greater moisture uptake, as shown in the last column of the table, due to its larger surface area [35].

The photographs of the mortar samples in Figure 4 present images of the free surface (the surface on the opposite side of the bottom of the cube mold) of samples containing 1 wt.% of FOEB and COEB. In Figure 4a, the FOEB fibers cover the surface of the samples, making it smooth, whereas the sample surface with COEB fibers is less smooth, with protrusions of sand particles (Figure 4b). A few rounded cavities are seen on the FOEB surface. The bottom surface of the samples reveals irregularly shaped cavities in both mortars with fine and coarse OEB (Figure 4c,d). There are relatively fewer cavities in the FOEB sample. The different shapes of the cavities between the FOEB and COEB can be observed in their sidewall samples in Figure 4e,f. Fine OEB has almost all spherical cavities, while coarse OEB has predominantly irregular cavities with a few spherical ones. The COEB samples have both irregularly shaped and spherical cavities, as seen on the sidewalls in Figure 4f.

The spherical shape of the cavities suggests that they were made by trapped air escaping from the fibers when water was being absorbed [36]. In contrast, the irregularly shaped cavities resulted from the inability of the fresh paste to fill the mold due to hindrance from coarse fibers. Both the FOEB and COEB samples exhibited irregular cavities on the bottom side, suggesting that coarse fibers blocked the fresh mortar flow. The FOEB samples contain predominantly spherical cavities, formed by air trapped when water was absorbed by the fibers, resulting in their lower density compared to the COEB samples, which had a mixture of spherical and irregular cavities. It can be deduced that due to the larger surface area per weight of fine fibers, more cavities, which are mostly spherical, formed in the FOEB samples than in the COEB ones. This phenomenon aligns with the fact that moisture uptake was higher in the fine OEB samples than in the coarse ones, as seen in Table 2.

3.4. Failure Patterns in the Compression Test

The designation of the failure patterns on the specimens after the compression test followed BS EN 12390-3:2009, an illustration of which is given in the Supplementary Materials (Figure S1) [37]. Satisfactory failures are characterized by equally distributed damage on the four sides of a cube, with the other two in contact with the platens undamaged. According to BS EN 12390-3:2009, satisfactory failure patterns are classified into three types and are labeled as a, b, and c in

Table 3. Failure patterns of compressed specimens.

Sample	Pattern Type/Compressive Strength (MPa)
	1	2	3	4	5	6
Control	8	a	4	5	4	a
	28.9	24.2	29.1	23.2	29.9	27.1
FOEB1	b	d	4	1	6	b
	19.7	22.9	21.7	20.0	19.9	18.4
FOEB2	4	9	6	6	a	b
	10.9	8.3	8.4	8.7	9.2	8.6
FOEB3	c	3	3	5	2	c
	8.8	8.7	8.5	8.4	8.6	8.3
COEB1	a	a	a	a	a	8
	19.6	20.7	18.9	24.7	24.3	22.3
COEB2	1	1	8	1	1	1
	15.2	13.3	14.1	15.7	17.5	15.8
COEB3	3	2	a	6	9	6
	9.9	9.3	10.9	10.2	8.1	9.5

, while the unsatisfactory ones are labeled 1 to 9. The failure patterns of each sample tested were photographed (Supplementary Materials, Figures S2–S7), and the pattern types are tabulated in Table 3.

The compressive strength values that resulted from the compression test per ASTM C109 did not represent the material’s exact properties but merely the maximum compressive pressure loading on the specimens before longitudinal tensile cracks or shear failure developed, and it was affected strongly by the friction between the platens and the specimens [38]; thus, it can only be utilized for comparative analysis between the samples. No correlation could be found between the strength values and the failure patterns; however, there was remarkable consistency in the COEB1 samples, with all showing the same satisfactory failure pattern “a”, which can be seen in Figure 5. COEB2 also showed consistency in the failure pattern “1” although not a satisfactory one. This consistency indicated that longer or coarse fibers prevented the disintegration of samples. Further study focusing on the effect of fiber length on the crack propagation or failure modes is therefore necessary.

The compressive strength numbers obtained from the compression test according to ASTM C109 did not accurately reflect the precise qualities of the material.

3.5. Flexural Strength

The flexural test, the results of which are presented in Figure 6a,b, was performed on samples with 1 wt.% COEB, which showed the lowest decrease in strength compared to the controls. The presence of fibers deflected crack propagation from the path where the stress was highest, in the mid-length of the two-point bending flexural test specimens. The introduction of fibers increased the flexural strength by acting as a barrier to crack propagation, as seen in Figure 6c, which was in agreement with findings from other researchers showing that OEB reduces microcracks [39]. The increasing flexural strength values exhibited in Figure 6d demonstrate the potential contribution of utilizing abundant palm oil industry waste in the concrete industry. Statistical analysis by an unpaired t-test shows a significant improvement in flexural strength with a p-value of 0.0009. The deviation was higher in the samples with fibers due to the increased inhomogeneity, as shown in the photograph in Figure 6d.

3.6. Drying Shrinkage

In line with the findings of other researchers, the introduction of OEB reduced the drying shrinkage [39]. The presence of hygroscopic fibers affected drying shrinkage, significantly reducing its rate, as shown in Figure 7. While the shrinkage in the control samples continued to increase at a lower rate until day 28, shrinkage stopped at around day 14 in the samples with OEB. It is hypothesized that the fibers acted as a moisture reservoir, releasing water into the mortar whenever the moisture content became lower than that of the fibers. During the initial hardening stage, moisture release delayed the setting, as mentioned earlier. After the mortar hardened, the moisture-balancing action of the fibers helped minimize drying shrinkage.

4. Conclusions

The small amount of COEB and FOEB used to replace the sand promotes the environmentally friendly material as an additive in mortar samples. The research’s conclusions are as follows:

• The presence of OEB fibers largely reduced flowability, but it did not directly affect setting time.
• The OEB fibers in the OPC-based mortar induced spherical macropores and irregularly shaped cavities that decreased its density and compressive strength. Spherical pores were formed by the air released upon the absorption of water by the fibers. The fibers, especially the longer ones, blocked the flow of the fresh mortar, resulting in irregularly shaped cavities.
• The maximum addition of COEB of 1% still exhibited good compressive strength of more than 20 MPa, reduced from 26.7 MPa of the control samples, and significantly increased flexural strength. The presence of longer fibers in the COEB prevented the disintegration of the samples by stopping crack propagation and at the same time minimized the drying shrinkage to less than 0.04% by day 25.