Home Eco-friendly utilisation of agricultural waste: Assessing mixture performance and physical properties of asphalt modified with peanut husk ash using response surface methodology
Article Open Access

Eco-friendly utilisation of agricultural waste: Assessing mixture performance and physical properties of asphalt modified with peanut husk ash using response surface methodology

  • Alaaeldin A. A. Abdelmagid EMAIL logo , Aichen Yue , Guocan Chen , Hai Guo and Aboubaker I. B. Idriss EMAIL logo
Published/Copyright: May 3, 2025
Download Article (PDF)
Become an author with De Gruyter Brill
Submit Manuscript Author Information
REVIEWS ON ADVANCED MATERIALS SCIENCE
From the journal REVIEWS ON ADVANCED MATERIALS SCIENCE Volume 64 Issue 1

Abstract

In recent years, researchers have been increasingly motivated to incorporate agricultural waste into asphalt pavements due to its potential to mitigate disposal challenges, conserve natural resources, and improve pavement performance. This study explores the incorporation of various dosages of peanut husk ash (PA) into asphalt to produce modified asphalt, with a focus on analysing the resultant physical properties. Response Surface Methodology was employed to examine the influence of asphalt content and PA dosage on the strength and volumetric characteristics of asphalt mixtures. The mixtures were subjected to tests for indirect tensile strength, Marshall stability, and flow. The findings indicate that asphalt binders can be modified with up to 8% PA without encountering phase separation issues during elevated temperature storage. Additionally, the study highlights the significant impact of preparation factors on the properties of the asphalt mixtures. The optimal values for achieving the highest desired response variables were determined to be 7.5% PA and 5.4% asphalt content.

Keywords: agricultural waste; peanut husk ash; asphalt mixture; sustainable construction materials

List of abbreviations

ANOVA

analysis of variance

CCD

central composite design method

dT

softening point difference

ITS

indirect tensile strength

MQ

Marshall quotient

PA

peanut husk ash

PAR

penetration aging ratio

PI

penetration index

RSM

Response Surface Methodology

RTFO

rolling thin film oven

SEM

scanning electron microscope

SPI

softening point increment

T 1.2

equivalent breaking point

T 800

equivalent softening point

VAI

viscosity aging index

VMA

voids in the mineral aggregate

VTM

voids in the total mix

ΔT

flexible temperature range

1 Introduction

For the last many years, the development of highway engineering construction in the world has always maintained a high growth rate. Due to the combined effects of vehicular loading and climatic conditions, asphalt undergoes aging and consequent performance degradation, leading to distress such as rutting and cracking in asphalt pavements. The aforementioned types of damages may exert a considerable influence on both the safety of drivers and the longevity of pavement [1,2,3]. Therefore, enhancing the serviceability of asphalt pavement has always been a key topic of current scholars. The utilisation of techniques for modifying asphalt has consistently been regarded as a crucial method for enhancing the functional capabilities of asphalt pavement. For example, modifiers for asphalt such as SBS, SBR, PE, tire pyrolysis carbon, and rubber have achieved certain application results in practical projects [4,5,6,7]. The proposition of a national strategy for sustainable resource management has generated growing interest in the utilisation of waste materials as modifiers for asphalt in recent times [8,9,10,11].

The global production of abundant peanut husks, an agricultural byproduct with low economic value that can amount to 30 million tons per year [12], poses a significant environmental challenge for their proper disposal. One method that has been attempted to address this issue is the process of mass burning. However, this method leads to environmental pollution and resource waste. Hence, the utilisation and advancement of biomass materials as an inexpensive form of construction materials has emerged as a popular trend among numerous researchers.

In recent years, some scholars have begun to study biomass ash-modified asphalt. Some of these conclusions show that the vast majority of biomass ash contains many components beneficial to asphalt [13,14,15,16,17]. These waste materials, such as rice husk ash and fly ash, have shown positive effects on enhancing mechanical strength and high-temperature stability, emphasising the potential of such waste materials to enhance the sustainability and performance of asphalt pavements [18]. Besides, some scholars studied the feasibility of using biomass ash in HMA mixture as a filler. They reported that there was a considerable enhancement in bonding among asphalt and particles of the aggregate [19,20,21,22]. Due to the metastable surface structure, the particle size of solid waste such as biomass ash, can be ground very fine and become more active [23,24], so that it can be employed as a modifier to enhance the properties of the mixture [25]. Peanut husk ash (PA) falls under the category of biomass ash [26]. The use of PA in the construction field to minimise the cost, enhance the strength, and modify brick, concrete, and cement, besides stabilising soil, has been considered in earlier research [27,28]. The potential of PA as an appropriate substitution to traditional cement in concrete applications has been explored in numerous studies [29,30]. These investigations report that PA possesses the tendency to enhance the compressive strength of the material. Recent studies have investigated the incorporation of PA into asphalt binders and mixtures, focusing on varying dosages and performance metrics. These studies primarily addressed isolated variables or fixed dosages, leaving a gap in understanding the synergistic effects of PA and asphalt content. A summary of key findings from prior works is provided in Table 1.

Table 1

Previous studies on PA-modified asphalt

Study Dosages of PA (%) Performance evaluated Main results
Abdelmagid et al. [31] 2, 4, 6, 8, 10 Physical, rheological, and storage stability properties Improved high-temperature performance but reduced low-temperature cracking resistance. Optimal dosage: 8%
Wang et al. [32] 1, 2, 3, 4, 5, 6 Self-healing, fatigue, rheological properties Improved self-healing (4% optimal) but reduced low-temperature ductility
Abdelmagid et al. [33] 2, 4, 6, 8, 10 Marshall stability, fatigue resistance, rutting resistance Enhanced adhesion and stiffness; 8% PA optimal for fatigue and thermal resistance

Besides preventing environmental contamination brought on by inappropriate handling of peanut husk, PA application in the production of modified asphalt may also lower production costs and enhance asphalt performance.

The utilisation of statistical modelling and optimisation techniques offers significant advantages due to their inherent ability to effectively address multiple objectives and constraints simultaneously. Moreover, these methodologies provide valuable insights into the complex interplay between variables and their influence on specific responses [34]. Response surface methodology (RSM) is a statistical approach that finds extensive utility in the analysis, modelling, and optimisation of a process [35,36,37]. This is accomplished by assessing individual influences and their interplay with a reduction in experimental run numbers [38,39,40]. RSM was first applied in the chemical industry to optimise the operation process. Subsequently, this method has been widely used in biology, medicine, aerospace, mechanical engineering, and other fields. Recently, in light of on-going advancements and progress in this theoretical approach, the response surface method has garnered significant interest among pavement engineering researchers. In pavement engineering, the implementation of RSM involves assessment of the attributes of asphalt mixtures. Bala et al. applied RSM to optimise the content of asphalt and nano silica according to the performance characteristics of the mixtures [41]. Baghaee Moghaddam et al. employed RSM to scrutinise the HMA factors of polyethylene terephthalate stone-matrix asphalt [42]. Khodaii et al. developed predictive models using RSM for the evaluation of the stripping characteristics of the HMA. Their model incorporated two preparation factors, namely grading and lime concentration [43]. Generally, the asphalt content, grading, temperature, modifier content, etc., are often the independent factors for RSM when testing asphalt mixes, while response variables are those that reflect the performance of the mixture [44,45,46].

2 Statement of the problem and aims

The annual generation of peanut husk, exceeding one hundred million tons, constitutes a substantial agricultural waste stream. Current disposal practices, such as burning and landfilling, pose environmental concerns. Utilising this waste as a replacement raw material in asphalt pavement offers a promising solution to mitigate environmental pollution, and achieve cost reductions in construction. This study is crucial in addressing the challenges posed by the disposal of agricultural waste and promoting sustainable development. Besides, the study of the use of PA as a modifier for asphalt and mixture represents a novel and sustainable solution to the challenges posed by agricultural waste and should be given significant attention and consideration. In this context, we expand on our previous work [31], which investigated the rheological properties of PA as a binder modifier, by examining its effects on asphalt mixtures. Therefore, it is important to conduct a feasibility study to evaluate the efficacy of using PA. Notably, a comprehensive review of available literature indicates the absence of prior studies that have employed RSM to investigate the combined influence of asphalt and PA on engineering characteristics.

In light of the above, the current work has studied the characteristics of asphalt binders modified with various dosages of PA and their mixtures. An evaluation of the physical characteristics of neat and modified asphalt was conducted through softening point, penetration, and viscosity before and after rolling thin film oven (RTFO) test, in addition to ductility and storage stability. Additionally, the scanning electron microscope (SEM) was used to observe the microstructure changes and the dispersion of PA into the asphalt. PA and asphalt content were examined for their effects on the volumetric strength of the HMA mixture utilising the RSM. The strength features of the HMA mixtures, namely Marshall stability and Indirect Tensile Strength (ITS), were evaluated. The RSM was then utilised to optimise the preparation process and conduct a statistical analysis of the influential variables.

3 Materials and test program

3.1 Asphalt

The 60/80 asphalt was utilised as the base asphalt for test analysis. Its physical characteristics are softening point, 48°C; ductility, 25°C, greater than 100 cm; penetration, 25°C, 67 dmm; and viscosity 0.35 Pa·s, 135°C.

3.2 PA

The peanut husk used in this study was first cleaned to remove any foreign materials, such as dirt, dust, or other organic matter, ensuring that the material was suitable for further processing. The peanut husk was burned at 600°C for 120 min, with careful control of the combustion process to prevent the formation of char or carbonised material, ensuring the complete conversion of organic matter into ash while minimising unburned carbon [47]. Then the obtained PA was crushed and dried through a 0.075 mm sieve to be used as PA for asphalt modification. The peanut husk and obtained PA are demonstrated in Figure 1.

Figure 1 (a) Peanut husk and (b) PA (0.075 mm).
Figure 1

(a) Peanut husk and (b) PA (0.075 mm).

3.3 Aggregate

The proportion of aggregate in the asphalt mixture is more than 90%. As an important component of the mixture, the mechanical properties and gradation of aggregate will have a great impact on the performance of the mixture. The coarse and fine aggregates that were utilised are crushed basalt minerals. Table 2 reveals aggregate properties based on JTG E42-2005 [48], and the aggregate gradation is presented in Figure 2.

Table 2

Aggregates properties

Property of aggregate Values Requirement
Sand equivalent (%) 93.7 ≥60
Water absorption (%) 0.39 ≤2.0
Crushing value (coarse aggregate) 13.1 ≤28
Apparent specific gravity (g·cm−3) 2.78 ≥2.60
Apparent specific gravity (fine aggregate) 2.69 ≥2.50
Los Angeles abrasion value (%) 15.1 ≤30
Figure 2 Gradation chart for aggregates.
Figure 2

Gradation chart for aggregates.

3.4 PA-modified asphalt

The asphalt was heated to the melting state at 155°C. Then, different dosages of PA modifier (2, 4, 6, 8, and 10%) were incorporated and mixed at 155°C using a slow speed mixer for 15 min to ensure the proper dispersion of PA into the asphalt. After that, the high-speed shear apparatus was applied at 155°C, 2,500 rpm for 30 min (Figure 3) [25]. The prepared PA-modified asphalt underwent a slow speed stirring process in an oven at 155°C 15 min to remove the remaining air introduced by the high-speed shear mixer.

Figure 3 High-speed mixer.
Figure 3

High-speed mixer.

3.5 Performance of asphalt binder

3.5.1 Physical properties

The physical properties of PA-modified asphalt were assessed before and after short-term aging by applying viscosity, ductility, penetration, and softening point tests [49]. Additionally, the penetration was done at 30, 15°C, as well as 25°C to utilise the JTG E20-2011 approach to evaluate asphalt temperature sensitivity. Tests were conducted on a minimum of three samples, with each test being repeated three times. The average value of the test outcomes was taken as the final result.

3.5.2 Temperature indexes

The penetration index (PI), flexible temperature range (ΔT), equivalent breaking point (T 1.2), and equivalent softening point (T 800) were assessed as indicators for the temperature performance of the modified asphalt. T 1.2 reflects the asphalt’s low-temperature stability, with lower values signifying better resistance to cracking and T 800 evaluates high-temperature stability, with higher values indicating improved resistance to softening and rutting at elevated temperatures. The ΔT represents a range of temperature from T 800 to T 1.2, where a wider ΔT suggests adaptability to a broader temperature range. The PI gives an indication of the sensitivity of the asphalt to changes in temperature, with a higher PI indicating less sensitivity to temperature changes, ensuring consistent performance. It was employed for the purpose of assessing the asphalt temperature susceptibility. These indexes were calculated based on the penetration values at 15, 25, and 30°C using Eqs. (1)–(5). To demonstrate the linear correlation between penetration (expressed as lg Pen) and temperature (T), Eq. (1) was utilised.

(1) lg Pen = ε + γ T ,

(2) T 1.2 = lg 1.2 − ε γ ,

(3) T 800 = lg 800 − ε γ ,

(4) ∆ T = T 800 − T 1.2 ,

(5) PI = 20 − 500 γ 1 + 50 γ ,

where γ and ε are coefficients of regression.

3.5.3 High-temperature storage stability

One of the main concerns that researchers have been focusing on is the stability of modified asphalt during storage at elevated temperatures. In the storage stability test, asphalt was introduced into an aluminium tube oriented vertically, with dimensions of 0.025 m in diameter and 0.14 m in height. The tube was then sealed and subjected to an elevated temperature of 163°C for a duration of 48 h in an oven. Afterward, it was frozen for 4 h at −4°C. After that, the measurement of the variance in softening points between the top and bottom samples of the aluminium tube was employed to assess its storage stability.

3.6 Mixture preparation

According to JTG F40–2004 [50], all mixture samples were prepared by the Marshall mix design method. The Marshall method consists of a number of steps. In order to achieve a temperature of 160°C during the mixing process, 1,200 g of aggregate and asphalt were put in an oven. Then, they were put in a mixer for 180 s, and the obtained mixture was placed in a mould with a diameter of 0.1016 m and a height of 0.0635 m. Afterward, the mould was put inside the Marshall automated compactor as shown in Figure 4 (and compacted 75 times on upper and lower surfaces, respectively). After 24 h at room temperature, the compacted mixture was removed out of the mould (Figure 5).

Figure 4 Marshall compactor.
Figure 4

Marshall compactor.

Figure 5 Marshall specimens.
Figure 5

Marshall specimens.

3.7 Performance of mixtures

3.7.1 Strength performance

The characteristics of strength in terms of Marshall stability and flow were measured based on JTG E20 2011 [49]. Before starting the test, all compacted samples were kept for 30 min at 60°C in a water bath. The stability determines the asphalt material’s capacity to withstand its maximum load at a 50.8 mm·min−1 loading rate, while the vertical deformation takes place when the maximum load is attained and describes the flow. Dividing the stability of every sample by its flow allowed one to find the Marshall Quotient (MQ).

The tensile strength of compacted asphalt mixtures was gauged utilising the ITS test. The values of ITS were obtained using Eq. (6).

(6) ITS = 2,000 P π tD ,

where D is the diameter of specimen, mm; t is the height of specimen, mm; P is the maximum load, N.

3.7.2 Volumetric characteristics

This research took into account the volumetric properties, including voids in the total mix (VTM) and voids in the mineral aggregate (VMA), which are important for pavement performance and durability. VMA and VTM were obtained through Eqs. (7) and (8), respectively [49].

(7) VMA = ( 1 − γ f × P s γ sb ) × 100 ,

(8) VTM = 1 − γ f γ t × 100 ,

where γ t is the theoretical maximum density; γ f is the bulk specific gravity of the mixture; γ sb is the bulk specific gravity of the aggregate; P s is the aggregate, percent by total weight of the mix.

3.8 RSM

Within this investigation, the central composite design method (CCD) is adopted and carries out a three-level two-factor test design. CCD is often used because of its appropriateness for consecutive experiments and high-quality design space forecasts [40,43]. First of all, the asphalt content (recorded as V1) and PA content (recorded as V2: PA content) were determined as two independent variables. From low to high, the asphalt content was determined to be 4.5, 5, 5.5, 6, and 6.5%. The PA content is 2, 4, 6, and 8% by weight of asphalt binder. Eight percent of PA was selected to be the highest content based on the asphalt performance tests (Figures 3–7). In addition, the following response values were chosen: (VTM, VMA, flow, Marshall stability, MQ, ITS). 13 test groups were developed utilising Design-Expert 13.0 programme. The response surface design consisted of 5 groups designated as the “centre point.” Each group had the same amount of asphalt percentage and PA, which were used to correct errors in the test process. The other 8 groups were utilised to explore the association among responses and independent factors. The relationship between the anticipated response and the variable value conforms to the quadratic equation, as shown in Eq. (9) [40], from which the best parameter value can be obtained.

(9) Y = β 0 + ∑ i = 1 n β i x i + ∑ i = 1 n β ii x i 2 + ∑ i < 1 n β ij x i x j + ϵ ,

where Y is the predicted response; n is the number of factors; ϵ is the statistical error; x i and x j is the independent variables; β ij , β ii , β i , and β 0 is the interaction coefficient, quadratic coefficient, linear coefficient, and intercept coefficient, respectively.

Analysis of Variance (ANOVA) is a widely used statistical method that assesses how different identifiable sources of variation contribute to the overall variability observed in measured responses [51,52] and was carried out to evaluate the accuracy of the model fitting and the significance level of asphalt content and PA concentration and their interaction with VTM, VMA, MQ, Marshall stability, ITS, and, flow.

4 Results and discussion

4.1 Asphalt binder performance

4.1.1 Classical asphalt binder tests

The penetration values at various temperatures (15, 25, and 30°C) of neat asphalt and asphalt binder modified with different content of PA are illustrated in Figure 6. Most noticeably of all, with the climb in concentration mass of PA, the penetration values decreased at different temperatures. There was a slight reduction in penetration at 15°C, while it decreased considerably at (30, 25°C). With the addition of 8% of PA, the neat asphalt’s penetration dropped by (30.4, 35.8, and 7.4%) at 30, 25, and 15°C, respectively. In addition, the neat asphalt revealed the highest penetration values. In contrast, the highest decrease in penetration was achieved by using 10% of PA, demonstrating that PA could boost the consistency of the asphalt and enhance its high-temperature performance. This finding arises due to the presence of activated silica and calcium and the porosity of the PA particle, which allows the high-molecular-weight components to be absorbed. The presence of a thin coating of asphaltene on the particle of PA leads to form new bonds, therefore reducing the degree of penetration.

Figure 6 Effect of PA content on penetration at various temperatures.
Figure 6

Effect of PA content on penetration at various temperatures.

Figure 7 displays the effects of PA on the ductility and softening point. From Figure 7 the softening point values increased steadily with growth PA content, suggesting that adding PA boosted the elevated temperature performance. The results of the penetration test were well corroborated by this result. The softening point of neat asphalt rose from 48°C to 53.8°C and 57°C when 4% PA and 8% PA were added, respectively. The lower the content of PA is, the less enhancement impact is. At elevated temperatures, asphalt binder with a great softening point has lower susceptibilities to rutting. Therefore, the increase in the values of softening point suggests that employing this type of asphalt to prepare the mixture might result in superior rutting performance. Also worth noting is that the ductility fell dramatically as the concentration mass of PA rose. Based on Figure 7, the PA increment led to grow the asphalt binder hardness and a significant drop in its ductility. This may be explained by the cohesion between PA particles and asphalt binder, as well as their capacity to absorb the binder resulting in the reduction.

Figure 7 Effect of PA content on ductility and softening point.
Figure 7

Effect of PA content on ductility and softening point.

Figure 8 illustrates the relationship between the asphalt viscosity at a temperature of 135°C and the different concentration mass of PA. The results demonstrate that a positive relationship exists between viscosity and PA, wherein an elevation in the latter leads to a corresponding increase in the former. The improvement in viscosity after adding PA could be due to the non-spherical and porosity form, robust absorption tendencies, as well as the existence of activated silica and calcium. The asphalt modified with PA exhibited a high value of viscosity compared to neat asphalt, as revealed in Figure 8. The viscosity of neat asphalt increased by 20, 37.1, 48.6, 65.7, and 82.9% with 2, 4, 6, 8, and 10% PA, respectively. However, it should be noted that all samples satisfied the Superpave standard’s requirement that at 135°C the viscosity ought not to surpass 3 Pa·s. This criterion is critical because excessively high viscosity can cause problems related to the workability of the asphalt. Thus, at elevated temperatures, using up to 10% of PA will not cause problems in compacting, mixing, or pumping.

Figure 8 Effect of PA content on viscosity.
Figure 8

Effect of PA content on viscosity.

The reduction in penetration and ductility, coupled with increased viscosity and softening point, suggests interfacial interactions between PA particles and the asphalt binder. These interactions are further supported by FTIR analysis from our previous work [33], which revealed the disappearance of an absorption peak at 2,186 cm⁻¹ in the neat asphalt after modification with PA. This change, along with the emergence of new peaks, indicates that chemical reactions occurred between PA and the asphalt binder, likely involving the formation of new bonds or complexes. Such chemical interactions, combined with the physical adsorption of asphaltenes onto PA’s porous surface, contribute to the enhanced stiffness and performance of the modified binder.

Table 2 displays the findings of the examination of the properties of the aged asphalt binders after (RTFO) aging through number of indexes including softening point increment (SPI), penetration aging ratio (PAR), and viscosity aging index (VAI). These indexes were determined from Eqs. (10)–(12) [53].

(10) SPI = Softening point aged − Softening point unaged ,

(11) PAR = Penetration aged Penetration unaged × 100 ,

(12) VAI = Viscosity aged − Viscosity unaged Viscosity unaged × 100 .

The findings derived from the analysis of Table 3 indicate that the utilisation of PA in the asphalt binder produced a rise in PAR, which was found to be directly associated with the rise in PA concentration. Conversely, a drop was observed in VAI and SPI values, which implies that PA had a favourable influence on enhancing the asphalt binder’s ability to resist aging [53].

Table 3

Aging indexes

Index PA content
0% 2% 4% 8% 6% 10%
SPI (%) 8.7 8.2 7.9 7.6 7.2 6.35
PAR (%) 64.5 68 70 73 75 78
VAI (%) 68.5 67.2 65 64.5 62 58.6

4.1.2 Indices for measuring the performance of temperature

Regression analysis was utilised to find the linear relationships between the logarithm of penetration (Y) and temperature (X), as listed in Table 4. In order to scrutinise the impact of the addition of PA contents, the number of temperature indexes (T 800, T 1.2, ΔT, and PI) were determined through Eqs. (1)–(5).

Table 4

Indices that measure the performance of temperature

PA (%) Linear regression equation ΔT (°C) T 1.2 (°C) T 800 (°C) PI
0 lg Pen = 0.72 + 0.0442 T 63.89 −14.5 49.3 −0.65
2 lg Pen = 0.74 + 0.0422 T 66.9 −15.6 51.4 −0.35
4 lg Pen = 0.77 + 0.0391 T 72.2 −17.6 54.6 0.15
6 lg Pen = 0.79 + 0.0373 T 75.7 −18.9 56.7 0.47
8 lg Pen = 0.81 + 0.0349 T 80.9 −21.0 59.9 0.93
10 lg Pen = 0.84 + 0.0324 T 87.2 −23.5 63.7 1.45

Generally, the aim of using T 800 is to scrutinise the elevated temperature performance of asphalt binders. Table 4 illustrates the tendency of the T 800 to fluctuate is substantially identical to the softening point. Compared to PA-modified asphalt, neat asphalt has the lowest value of T 800, which proves again the addition of PA could boost the performance of asphalt binder at elevated temperatures. The T 1.2 was utilised to evaluate the PA-modified asphalt in terms of resistance to low-temperature cracking. The better behaviour of the binder at low temperatures could be seen when the PA concentration increased, as exhibited in Figure 9.

Figure 9 Impact of PA content on temperature performance indexes.
Figure 9

Impact of PA content on temperature performance indexes.

The greatest values of T 1.2 and T 800 were found in the 10% PA, with a 29 and 61% rise, respectively. Better low-temperature performance is indicated by a lower value of T 1.2, while better high-temperature performance is indicated by a higher T 800 value. Additionally, as demonstrated in Table 4, there was a rise in the ΔT with an increase in the PA concentrations. Based on these results, asphalt binder modified with PA would have improved performance properties at both low and high temperatures.

Table 4 also displays the computed PI values. Table 4 shows that the PI of neat asphalt is lower than that of PA-modified asphalt binder. Lower values of PI indicate higher temperature susceptibility, which may lead to less rutting resistance. Table 4 shows that the values of PI rose as the PA concentration climbed. The use of PA at concentrations of 2, 4, 6, 8, and 10% resulted in PI increases of 45.9, 123.3, 172, 241, and 321%, respectively. This indicates that asphalt binders mixed with PA exhibited reduced sensitivity to temperature variations.

4.1.3 High-temperature storage stability

Huge differences in the softening point that occurs during storage under certain conditions indicate a significant amount of modifier segregation. The outcome of the storage stability of PA is exhibited in Figure 10. Figure 10 demonstrates that the PA softening point difference (dT) tends to rise with an increase in PA. The (dT) after 48 h of storage should not exceed 2.2°C. As for the data in Figure 10, it demonstrates that at a PA concentration of 10%, dT values more than 2.2°C were attained, whereas the remaining PA-modified asphalt binders were within the allowed range. However, this may be explained by the fact that at elevated temperatures during storage, modifier segregation would occur as a result of the different density and solubility parameters between PA particles and asphalt, besides the gravity influence on particles. The impact became intense as the concentration mass of PA increased, resulting in a more remarkable change in the softening point of the bottom and top sides, consequently leading to segregation once the PA level in the sample exceeded 8%.

Figure 10 Effect of PA on the storage stability.
Figure 10

Effect of PA on the storage stability.

4.2 RSM

The parameters of the independent variables and the outcome of the response values are listed in Table 5. With these results, RSM was utilised to discover interactions between the parameters of the independent variables and responses.

Table 5

CCD experiment and its outcomes

Run Factor Response
Asphalt PA VMA VTM Stability Flow MQ ITS
(%) (%) (%) (%) (kg) (mm) (kg mm−1) (kPa)
1 4.5 4 18.9 8.1 1422.5 3.21 443.148 910
2 5.5 4 17.2 3.69 1419.44 3.53 402.109 1,020
3 6.5 8 19.2 2.1 1184.91 3.34 354.763 796
4 5.5 4 17.4 3.58 1423.52 3.43 415.021 1,026
5 6.5 4 18.2 2.4 1174.71 3.65 321.839 771
6 5.5 0 16.4 3.37 1389.87 3.56 390.414 945
7 5.5 4 18 3.52 1425.56 3.63 392.717 1,029
8 4.5 8 19.5 7.2 1452.08 2.92 497.286 960
9 6.5 0 17.7 2.4 1143.1 3.99 286.492 780
10 5.5 4 17 3.76 1428.62 3.49 409.347 1,022
11 4.5 0 18.7 7.9 1309.32 3.49 375.162 820
12 5.5 4 17.3 3.71 1442.9 3.4 424.382 995
13 5.5 8 18.1 3.02 1489.81 3.31 450.092 1,066

The statistical analysis details and how preparation parameters affect the features of asphalt mixture are covered in the following sections.

4.2.1 Predictive models analysis

One of the crucial components of the data analysis is evaluating the adequacy of the model [54]. Employing the CCD technique, the RSM software identified a varying number of influential factors across 13 randomised test sets. Table 5 systematically presents the observed responses according to these factors. After doing a regression analysis, to forecast response variables the RSM recommended a linear model and a quadratic equation model. ANOVA was utilised to assess the statistical significance values of the predictive models, as shown in Table 6.

Table 6

Quadratic regression (ANOVA)

Responses Source SS df MS F-value p-value
Stability Model 166500 5 33291.87 140.06 <0.0001
V 1 77332.16 1 77332.16 325.35 <0.0001
V 2:PA content 13490.11 1 13490.11 56.75 0.0001
V 1 V 2 2547.82 1 2547.82 10.72 0.0136
V 1 60323.98 1 60323.98 253.79 <0.0001
V 2:PA content 118.78 1 118.78 0.4997 0.5025
Lack of fit 1342.32 3 447.44 5.57 0.0653
Flow Model 0.6684 2 0.3342 37.29 <0.0001
V 1 0.3083 1 0.3083 34.40 0.0002
V 2:PA content 0.3602 1 0.3602 40.19 <0.0001
Lack of fit 0.0569 6 0.0095 1.16 0.4638
MQ Model 35257.78 5 7051.56 41.67 <0.0001
V 1 20709.55 1 20709.55 122.38 <0.0001
V 2:PA content 10422.87 1 10422.87 61.59 0.0001
V 1 V 2 725.02 1 725.02 4.28 0.0772
V 1 2 3147.01 1 3147.01 18.60 0.0035
V 2 2 44.28 1 44.28 0.2617 0.6247
Lack of fit 599.37 3 199.79 1.37 0.3734
ITS Model 135600 5 27115.24 100.14 <0.0001
V 1 19608.17 1 19608.17 72.41 <0.0001
V 2:PA content 12788.17 1 12788.17 47.23 0.0002
V 1 V 2 3844.00 1 3844.00 14.20 0.0070
V 1 2 82085.39 1 82085.39 303.14 <0.0001
V 2 2 151.10 1 151.10 0.5580 0.4794
Lack of Fit 1162.29 3 387.43 2.11 0.2412
VMA Model 9.41 5 1.88 17.41 0.0008
V 1 0.6667 1 0.6667 6.17 0.0420
V 2:PA content 2.67 1 2.67 24.67 0.0016
V 1 V 2 0.1225 1 0.1225 1.13 0.3224
V 1 2 4.97 1 4.97 45.98 0.0003
V 2 2 0.0047 1 0.0047 0.0438 0.8403
Lack of fit 0.1886 3 0.0629 0.4427 0.7354
VTM Model 52.30 5 10.46 679.19 <0.0001
V 1 44.28 1 44.28 2875.19 <0.0001
V 2:PA content 0.3038 1 0.3038 19.72 0.0030
V 1 V 2 0.0400 1 0.0400 2.60 0.1511
V 1 2 7.52 1 7.52 488.02 <0.0001
V 2 2 0.4538 1 0.4538 29.46 0.0010
Lack of fit 0.0687 3 0.0229 2.34 0.2142

df is the degree of freedom; MS is the mean square; SS is the sum of squares.

The significance of the model may be determined by using the F-value and 95% confidence level. For stability, flow, MQ, VTM, VMA, and ITS, the model F-value of 140.06, 37.29, 41.67, 679.19, 17.41, and 100.14, respectively, shows that the model is significant, and Table 6 reveals that for all models, the likelihood of a model F-value of this size arising due to noise is 0.01%. All suggested models and their terms, except for a few irrelevant ones with a p-value beyond 0.05, were statistically significant with a 95% confidence level, since their p-values were less than 0.05. The backward elimination technique was employed to remove these terms with the purpose of boosting the models. Eqs. (13)–(18) illustrate the outcomes of the constructed models.

(13) Stability = − 2676.603 + 1564.863 V 1 + 46.556 V 2 − 6.31 V 1 V 2 − 150.287 V 1 2 ,

(14) Flow = 2.456 + 0.227 V 1 − 0.061 V 2 ,

(15) MQ = − 355.549 + 309.244 V 1 + 28.932 V 2 − 3.366 V 1 V 2 − 32.230 V 1 2 ,

(16) ITS = − 4187.768 + 1901.190 V 1 + 54.167 V 2 − 7.75 V 1 V 2 − 175.214 V 1 2 ,

(17) VTM = 68.301 − 20.863 V 1 + 0.146 V 2 + 1.64 V 1 2 − 0.025 V 2 2 ,

(18) VMA = 59.563 − 15.262 V 1 + 0.167 V 2 + 1.357 V 1 2 .

The residuals analysis was carried out in terms of lack of fit to scrutinise the accuracy of the model. Lack of fit should be non-significant with p beyond 0.05 which suggests a good model, and there is a significant effect on output responses. Conversely, if the lack of fit is significant with p less than 0.05, it means the experimental data cannot be fit by the proposed model, and the responses are not affected significantly by the independent variables. Table 6 shows that the p-value of lack of fit for all models is not significant, implying that the accuracy of all models is high, and that models are adequate for the purpose of data explanation in the field of experimentation.

Additionally, the suitability of the proposed models was evaluated graphically. The correspondence between actual and predicted ones is shown in Figure 11. The prediction accuracy increases with decreasing distances between points and the diagonal, while the error increases with increasing top-to-bottom deviations. As can be observed, the whole distribution does not include any outliers, which implies that the ones that were suggested may correctly predict the properties of a mixture that contains PA.

Figure 11 Correlation between the actual values obtained from experiments and the values predicted by a model.
Figure 11

Correlation between the actual values obtained from experiments and the values predicted by a model.

Table 7 illustrates the results from the ANOVA analysis, including the (adequacy precision, R², R adj 2 , R pre 2 ). Fit quality is assessed using R 2, where R 2 over 0.8 indicates a well-fitted model [55]. All models had values over 0.8, as shown in Table 7. The high values of R 2 and R adj 2 demonstrate that the models are statistically sound and that the actual and anticipated responses matched [42]. AP is utilised to quantify signal-to-noise ratio, and a value over 4 indicates a good model [56]. All models in Table 7 exhibit AP values exceeding the acceptable range, confirming their suitability for navigating the design space.

Table 7

Indices that assess the appropriateness of the model

Responses R 2 R adj 2 R pre 2 AP
Stability 0.990 0.983 0.936 31.59
Flow 0.882 0.858 0.800 20.74
MQ 0.968 0.944 0.825 22.73
ITS 0.986 0.976 0.908 26.02
VMA 0.926 0.872 0.758 12.68
VTM 0.998 0.997 0.986 70.73

pre = predicted; adj = adjusted.

4.2.2 Strength performance analysis

The resistance of an HMA mixture to permanent deformation and shoving when subjected to traffic loads is known as stability. Therefore, it needs to be sufficient to manage traffic effectively, and at the same time, not greater than what the traffic circumstances necessitate. Both independent factors exert a considerable effect on the stability, as shown in Figure 12. The values of Marshall stability first climbed with increasing asphalt content, peaked at 5–6% asphalt content, and then began to decrease. The film thickness increased as a result of the asphalt concentration exceeding the peak value, preventing the load from being transmitted effectively. Compared to the unmodified asphalt mixture, adding PA led to strengthening the asphalt mixture. As the content of PA increased, the values of stability showed an increasing tendency, demonstrating the enhanced performance of the PA-modified asphalt mixture. The highest value of stability was achieved when using 8% of PA with asphalt content of 5.5%, while the lowest value was obtained when using asphalt content of 6.5% with 0% of PA (control specimen).

Figure 12 Impact of V 1 and V 2 on stability.
Figure 12

Impact of V 1 and V 2 on stability.

Flow is the capability of mixtures to adapt to slow moves and settlement with no crack, and it could be used to evaluate the mixture performance when subjected to traffic loading. The impact of the asphalt binder percentage and PA concentration on the flow is shown in Figure 13. The values of flow rose with increasing asphalt content while decreasing as the PA was added. As the amount of asphalt increased, thick films of asphalt formed, prohibiting the aggregate from interlocking strongly and leading flow to increase. However, the flow reduction that occurred with the PA addition could be associated with the hardness enhancement of the asphalt mixture modified with PA compared to the unmodified sample, which enhanced resistance to deformation.

Figure 13 Impact of V 1 and V 2 on flow.
Figure 13

Impact of V 1 and V 2 on flow.

The evaluation of MQ is a critical factor in gauging the ability of the asphalt mixture to withstand permanent deformation. Figure 14 demonstrates the influence of the asphalt binder and PA on MQ. As the quantity of asphalt was increased from 4.5 to 5.5%, it demonstrates that the MQ grew, and subsequently decreased when a larger percentage was applied, this suggests that the optimum asphalt concentration is between 5 and 6%. What’s more, the MQ increased when the PA was added to the asphalt mixture at different concentration levels, offering greater resistance to rutting. According to the findings, both variables significantly impacted the MQ. A greater MQ was observed in the asphalt mixture made with 8% PA and an asphalt content of 4.5%.

Figure 14 Impact of V 1 and V 2 on MQ.
Figure 14

Impact of V 1 and V 2 on MQ.

The impact of the asphalt binder and PA concentration on the tensile strength is illustrated in Figure 15. It is a useful characteristic that may affect the HMA crack resistance. It can be seen that the difference in asphalt concentration between the lower and intermediate ranges demonstrates a substantial impact on the values of ITS. Furthermore, the use of PA at certain levels of asphalt content significantly enhanced the tensile strength. The observed improvement in tensile strength with the incorporation of PA into the asphalt mixture is primarily attributed to the uniform dispersion of PA particles within the mixture. This uniform dispersion enhances the cohesion between the asphalt binder and aggregates, leading to improved mechanical properties and increased resistance to cracking. Additionally, the filler effect of PA particles likely contributes to better load distribution throughout the mixture, further strengthening the bond between the components and enhancing the mixture’s overall durability. The positive correlation between asphalt binder percentage and PA concentration also suggests that PA acts as a reinforcement agent, improving the tensile strength of the modified asphalt mixture. This improvement is particularly significant compared to unmodified asphalt mixtures, where the lack of PA results in lower tensile strength and, consequently, less resistance to cracking. On the whole, asphalt binder percentage and PA concentration affected the tensile strength positively.

Figure 15 Impact of V 1 and V 2 on ITS.
Figure 15

Impact of V 1 and V 2 on ITS.

4.2.3 Volumetric characteristics analysis

The particles of aggregate in the mixture need to be sufficiently spaced apart for the pavement to get sufficient film thickness, besides allowing enough air void and effective asphalt (not absorbed by aggregate particles). This void of volume space between the particles is known as VMA, which is a vital volumetric in ensuring the durability of the mixture. Figure 16 represents the influence of the asphalt binder and PA on VMA. It is noteworthy to highlight the first finding that the VMA values declined considerably as the asphalt concentration grew from 4.5 to 5.5% and then climbed marginally by adding 6.5% asphalt. Also worth noting is that at a given concentration of asphalt binder, the value of VMA rose with increasing PA concentration, which suggests that the PA affected the VMA significantly.

Figure 16 Impact of V 1 and V 2 on VMA.
Figure 16

Impact of V 1 and V 2 on VMA.

VTM is the volume between the particles of aggregates coated with asphalt in the mixture, which is an important characteristic impacting pavement durability. The impact of the asphalt concentration and PA on the VTM is shown in Figure 17. It is worth remarking here that VTM was considerably affected when the percentage of asphalt climbed from 4.5 to 6.5%, there was a significant decrease in the value of VTM. What’s more, the value of VTM rose with the addition of 4% PA before falling at the 8% PA concentration. Moreover, the values of VTM with a standard range (3–5%) were obtained when using a concentration of 4–8% PA and 5.5% of the binder, this promotes the use of HMA-containing PA to improve pavement durability.

Figure 17 Impact of V 1 and V 2 on VTM.
Figure 17

Impact of V 1 and V 2 on VTM.

4.2.4 Optimisation of mixture parameters

The multiple response optimisation was carried out with the Design-Expert software to determine an optimal ratio of asphalt and PA content which could be used to get optimised responses in terms of maximising the behaviour of asphalt mixtures, besides meeting the requirements of the Marshall design. The target values for different responses selected as (VTM 3–5%, VMA ≥ 17%, ITS Maximum kPa, MQ Maximum kg·mm−1, Stability Maximum kg, Flow 2–4 mm) based on JTG F40–2004 [50]. To achieve the mixture's optimal performance in terms of tensile strength and durability, the ITS and stability were adjusted to their maximum values.

The RSM program chose the best options, which are shown in Figure 18, and had a 0.83 as the maximum combined desirability. The findings indicated that 5.4 and 7.5% were the optimum amounts of asphalt binder and PA, respectively.

Figure 18 RSM plot; desirability for the optimum solutions.
Figure 18

RSM plot; desirability for the optimum solutions.

In order to confirm the optimisation findings and the precision of the response models, more experimental work was conducted by using the optimum contents. Eq. (19) was used to determine the error between the actual data and the model, and all findings were tabulated in Table 8. A very good agreement was observed with a proportional error under four percent when comparing experimental results with the predicted ones, as shown in Table 8.

(19) Error ( % ) = E xperimental value − predicted value E xperimental value × 100 .

Table 8

Comparison of obtained and predicted results

Response Actual predicted Error (%)
ITS (kPa) 1061.3 1063.79 −0.24
VMA (%) 17.7 18.01 −1.69
VTM (%) 3.55 3.58 −0.85
MQ (kg·mm−1) 450.62 458.31 −1.71
Flow (mm) 3.15 3.21 −2.00
Stability (kg) 1441.60 1489.81 −3.34

4.3 SEM

The SEM images of PA, neat asphalt, and asphalt binders modified with some selected PA percent were observed using a JSM-7500F SEM, as shown in Figure 19. According to the image in Figure 19(a), these particles have irregular, non-uniform, and porous structures. Besides, the surface of the particles has smaller pores, which may be the reason for the high PA absorption. It seems that a surface that is not level and the angularity of particles with an irregular form produces an increase in friction between the particles and the asphalt binder, thereby generating more stress than spherical particles, particularly when the asphalt binder stiffens at low temperatures. Figure 19(b) represents the neat asphalt, while Figure 19(c)–(e) reveal asphalt binder modified with 2, 6, and 10% PA, respectively. As for the images in Figure 19(c)–(e), bright dots depict the particles, whilst the black backdrop illustrates the binder. Most noticeably of all, it can be seen that in spite of the fact that the PA particles ranged in size from nanometres to micrometres, PA particles homogeneously dispersed through the binder. Also worth noting is that none of the asphalt binders modified with 2, 6, and 10% PA had any observable boundaries between PA particles and asphalt, nor considerable agglomeration.

Figure 19 SEM images of (a) PA, (b) neat asphalt, (c) 2% PA, (d) 6% PA, and (e) 10% PA modified asphalt.
Figure 19

SEM images of (a) PA, (b) neat asphalt, (c) 2% PA, (d) 6% PA, and (e) 10% PA modified asphalt.

5 Conclusion

The primary objective of this study was to evaluate the efficacy of utilising PA as a modifier for both asphalt binder and mixture. To achieve this, a range of experimental tests and statistical analyses using RSM were employed to scrutinise the physical characteristics of asphalt binders that were mixed with varying PA content, and the mechanical behaviour of the resultant mixtures. Based on the findings of this study, the following conclusions can be drawn:

  • Enhanced binder properties: The addition of PA to the asphalt binder increased viscosity and softening point, while decreasing penetration and ductility. This indicates that PA effectively stiffens the binder and enhances its resistance to rutting.

  • Improved aging resistance: Incorporating PA into the asphalt binder improved PAR values and reduced VAI and SPI values, demonstrating a beneficial impact on the aging resistance of the binder.

  • Temperature resistance and susceptibility: Enhanced T 1.2, T 800, and PI values with PA incorporation suggest improved resistance of the binder to both high and low temperatures, reducing its temperature susceptibility.

  • Optimal PA concentration: A maximum concentration of 8% PA by mass is recommended for the asphalt binder to prevent phase separation issues during transportation or storage at elevated temperatures.

  • Mixture performance and optimisation: ANOVA results indicated that PA and asphalt content significantly affect the HMA mixture’s volumetric and strength properties. Adding PA increased MQ, VMA, tensile strength, and Marshall stability while reducing VTM and flow. The optimal blend of 5.4% asphalt binder and 7.5% PA was identified to achieve the highest performance and meet Marshall mix design criteria.

As agriculture waste continues to increase, there is a pressing need for sustainable solutions that can reduce its environmental impact. A viable way to address this issue is to employ PA as a modifier for asphalt mixture and binder. The integration of PA as a modifier has the potential to boost the properties and performance of asphalt binder and mixture, resulting in more durable and long-lasting infrastructure. While the study demonstrates the efficacy of PA as a sustainable modifier for asphalt binders and mixtures, there are several limitations that need to be acknowledged. First, all tests and analyses were conducted under controlled laboratory conditions. Although the findings indicate improvements in mechanical properties and performance, these results may not fully represent the behaviour of PA-modified asphalt under real-world environmental conditions. Moreover, while the study achieved successful results for PA content up to 8% without phase separation, further research is required to fully understand the long-term storage stability and field applicability at higher percentages, especially when subjected to continuous high-temperature exposure.

Overall, the utilise of PA as a modifier for asphalt and mixture is a crucial step towards promoting sustainable development in the pavement industry, and it highlights the importance of finding innovative ways to utilise agricultural waste.

  1. Funding information: This research was funded by the National Foreign Expert Project in China (Project Number: Y20240076).

  2. Author contributions: All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

  3. Conflict of interest: The authors state no conflict of interest.

  4. Data availability statement: All data generated or analysed during this study are included in this published article.

References

[1] Li, Y., C. Liu, G. Yue, Q. Gao, and Y. Du. Deep learning-based pavement subsurface distress detection via ground penetrating radar data. Automation in Construction, Vol. 142, 2022, id. 104516.10.1016/j.autcon.2022.104516Search in Google Scholar

[2] Shyaa, A. S. and I. D. R. Abd Rahma. Asphalt pavement rutting distress and affects on traffics safety. Journal of Traffic and Transportation Engineering, Vol. 10, 2022, pp. 35–39.10.17265/2328-2142/2022.01.004Search in Google Scholar

[3] Wu, H., A. Shen, Y. Guo, Z. Lyu, and L. Fan. Mechanical response of basalt fiber asphalt pavement with low temperature-load coupling in heavy frozen areas. CICTP 2020, 2020, pp. 1883–1895.10.1061/9780784483053.158Search in Google Scholar

[4] Leng, Z., Z. Tan, H. Yu, and J. Guo. Improvement of storage stability of SBS-modified asphalt with nanoclay using a new mixing method. Road Materials and Pavement Design, Vol. 20, No. 7, 2019, pp. 1601–1614.10.1080/14680629.2018.1465842Search in Google Scholar

[5] Neto, S. A. D., M. M. Farias, J. C. Pais, P. A. A. Pereira, and J. B. Sousa. Influence of crumb rubber and digestion time on the asphalt rubber binders. Road Materials and Pavement Design, Vol. 7, No. 2, 2006, pp. 131–148.10.1080/14680629.2006.9690030Search in Google Scholar

[6] Zhang, J., Z. Yao, T. Yu, S. Liu, and H. Jiang. Experimental evaluation of crumb rubber and polyethylene integrated modified asphalt mixture upon related properties. Road Materials and Pavement Design, Vol. 20, No. 6, 2019, pp. 1413–1428.10.1080/14680629.2018.1447505Search in Google Scholar

[7] Li, C., S. Yi, Q. Deng, Y. Gan, and Z. Hu. Optimizing the performance of TPCB/SCA composite-modified asphalt using improved response surface methodology. Reviews on Advanced Materials Science, Vol. 63, No. 1, 2024, id. 20230182.10.1515/rams-2023-0182Search in Google Scholar

[8] Abo-Shanab, Z. L., A. A. Ragab, and H. M. Naguib. Improved dynamic mechanical properties of sustainable bio-modified asphalt using agriculture waste. International Journal of Pavement Engineering, Vol. 22, No. 7, 2021, pp. 905–911.10.1080/10298436.2019.1652301Search in Google Scholar

[9] Debbarma, S., G. Ransinchung, S. Singh, and S. K. Sahdeo. Utilization of industrial and agricultural wastes for productions of sustainable roller compacted concrete pavement mixes containing reclaimed asphalt pavement aggregates. Resources, Conservation and Recycling, Vol. 152, 2020, id. 104504.10.1016/j.resconrec.2019.104504Search in Google Scholar

[10] Hung, V. Q., A. Jayarathne, C. Gallage, L. Dawes, P. Egodawatta, and S. Jayakody. Leaching characteristics of metals from recycled concrete aggregates (RCA) and reclaimed asphalt pavements (RAP). Heliyon, Vol. 10, No. 9, 2024, id. e30407.10.1016/j.heliyon.2024.e30407Search in Google Scholar PubMed PubMed Central

[11] Rispoli, O. and O. O. Ajibade. Comparative life cycle assessment of a novel sustainable road pavement system adopting recycled plastic from PET bottles and carbonated aggregate. Heliyon, Vol. 10, No. 2, 2024, id. e24354.10.1016/j.heliyon.2024.e24354Search in Google Scholar PubMed PubMed Central

[12] Tejedor, J., V. Cóndor, C. E. Almeida-Naranjo, V. H. Guerrero, and C. A. Villamar. Performance of wood chips/peanut shells biofilters used to remove organic matter from domestic wastewater. Science of The Total Environment, Vol. 738, 2020, id. 139589.10.1016/j.scitotenv.2020.139589Search in Google Scholar PubMed

[13] Abdelmagid, A. A. A. and C. P. Feng. Evaluating the effect of rice-husk ash and crumb-rubber powder on the high-temperature performance of asphalt binder. Journal of Materials in Civil Engineering, Vol. 31, No. 12, 2019, id. 04019296.10.1061/(ASCE)MT.1943-5533.0002962Search in Google Scholar

[14] Abdelmagid, A. A. A. and C. P. Feng. Laboratory evaluation of the effects of short-term aging on high temperature performance of asphalt binder modified with crumb rubber and rice husk ash. Petroleum Science and Technology, Vol. 37, No. 13, 2019, pp. 1557–1565.10.1080/10916466.2019.1590409Search in Google Scholar

[15] Bhattacharya, S. C., P. A. Salam, H. L. Pham, and N. H. Ravindranath. Sustainable biomass production for energy in selected Asian countries. Biomass & Bioenergy, Vol. 25, No. 5, 2003, pp. 471–482.10.1016/S0961-9534(03)00085-0Search in Google Scholar

[16] Fareed, A., S. B. A. Zaidi, N. Ahmad, I. Hafeez, A. Ali, and M. F. Ahmad. Use of agricultural waste ashes in asphalt binder and mixture: A sustainable solution to waste management. Construction and Building Materials, Vol. 259, 2020, id. 120575.10.1016/j.conbuildmat.2020.120575Search in Google Scholar

[17] Han, B., Y. Xing, C. Li, J. Yan, X. Zhang, and L. Zhang. Effect of waste crab shell powder on matrix asphalt. Reviews on Advanced Materials Science, Vol. 63, No. 1, 2024, id. 20240059.10.1515/rams-2024-0059Search in Google Scholar

[18] Si, H., D. Shen, M. N. Amin, S. U. Arifeen, M. T. Qadir, and K. Khan. Producing sustainable binding materials using marble waste blended with fly ash and rice husk ash for building materials. Reviews on Advanced Materials Science, Vol. 63, No. 1, 2024, id. 20240049.10.1515/rams-2024-0049Search in Google Scholar

[19] Foroutan Mirhosseini, S. A., M. M. Khabiri, A. Kavussi, and M. H. Jalal Kamali. Applying surface free energy method for evaluation of moisture damage in asphalt mixtures containing date seed ash. Construction and Building Materials, Vol. 125, 2016, pp. 408–416.10.1016/j.conbuildmat.2016.08.056Search in Google Scholar

[20] Jeffry, S. N. A., R. P. Jaya, N. Abdul Hassan, H. Yaacob, and M. K. I. M. Satar. Mechanical performance of asphalt mixture containing nano-charcoal coconut shell ash. Construction and Building Materials, Vol. 173, 2018, pp. 40–48.10.1016/j.conbuildmat.2018.04.024Search in Google Scholar

[21] Mistry, R., S. Karmakar, and T. Kumar Roy. Experimental evaluation of rice husk ash and fly ash as alternative fillers in hot-mix asphalt. Road Materials and Pavement Design, Vol. 20, No. 4, 2019, pp. 979–990.10.1080/14680629.2017.1422791Search in Google Scholar

[22] Nisar, M. A., S. B. A. Zaidi, A. Fareed, J. S. Carvajal-Munoz, and I. Ahmed. Performance evaluation of Nano wood ashes in asphalt binder and mixture. International Journal of Pavement Engineering, Vol. 23, No. 10, 2022, pp. 3318–3332.10.1080/10298436.2021.1892107Search in Google Scholar

[23] Nair, D. G., A. Fraaij, A. A. K. Klaassen, and A. P. M. Kentgens. A structural investigation relating to the pozzolanic activity of rice husk ashes. Cement and Concrete Research, Vol. 38, No. 6, 2008, pp. 861–869.10.1016/j.cemconres.2007.10.004Search in Google Scholar

[24] Sinsiri, T., W. Kroehong, C. Jaturapitakkul, and P. Chindaprasirt. Assessing the effect of biomass ashes with different finenesses on the compressive strength of blended cement paste. Materials & Design, Vol. 42, 2012, pp. 424–433.10.1016/j.matdes.2012.06.030Search in Google Scholar

[25] Xue, Y., S. Wu, J. Cai, M. Zhou, and J. Zha. Effects of two biomass ashes on asphalt binder: Dynamic shear rheological characteristic analysis. Construction and Building Materials, Vol. 56, 2014, pp. 7–15.10.1016/j.conbuildmat.2014.01.075Search in Google Scholar

[26] Idriss, A. I., J. Li, Y. Guo, Y. Wang, E. A. Elfaki, and E. A. Ahmed. Improved sintering quality and mechanical properties of peanut husk powder/polyether sulfone composite for selective laser sintering. 3D Printing and Additive Manufacturing, Vol. 10, 2021, pp. 111–123.10.1089/3dp.2021.0036Search in Google Scholar PubMed PubMed Central

[27] Edeh, J. E., M. Joel, and J. M. Aburabul. Groundnut shell ash stabilized reclaimed asphalt pavement, as pavement material. Advances in Materials Research, Vol. 824, 2013, pp. 3–11.10.4028/www.scientific.net/AMR.824.3Search in Google Scholar

[28] Trivedi, N. S., S. A. Mandavgane, S. Mehetre, and B. D. Kulkarni. Characterization and valorization of biomass ashes. Environmental Science and Pollution Research, Vol. 23, No. 20, 2016, pp. 20243–20256.10.1007/s11356-016-7227-7Search in Google Scholar PubMed

[29] Alabadan, B., C. Njoku, and M. Yusuf. The potentials of groundnut shell ash as concrete admixture. Agricultural Engineering International: CIGR Journal, Vol. 8, 2006, pp. 1–8.Search in Google Scholar

[30] Ikumapayi, C. Properties of groundnut shell (Arachis hypogaea) ash blended portland cement. Journal of Applied Sciences and Environmental Management, Vol. 22, 2018, id. 1553.10.4314/jasem.v22i10.03Search in Google Scholar

[31] Abdelmagid, A. A. A., Y. Qiu, and E. Yang. Laboratory investigation of the impact of peanut husk ash as an asphalt binder modifier on physical and rheological properties. Construction and Building Materials, Vol. 401, 2023, id. 132920.10.1016/j.conbuildmat.2023.132920Search in Google Scholar

[32] Wang, B., J. Shen, S. Li, and W. Wang. Peanut shell powder as a sustainable modifier and its influence on self-healing properties of asphalt. Materials, Vol. 16, No. 20, 2023, id. 6618.10.3390/ma16206618Search in Google Scholar PubMed PubMed Central

[33] Abdelmagid, A. A. A. and Y. Qiu. Repurposing agricultural waste: The effectiveness of peanut husk ash in improving asphalt mixture properties. Construction and Building Materials, Vol. 411, 2024, id. 134476.10.1016/j.conbuildmat.2023.134476Search in Google Scholar

[34] Kikuchi, S., N. Kronprasert, and S. M. Easa. Aggregate blending using fuzzy optimization. Journal of Construction Engineering and Management, Vol. 138, No. 12, 2012, pp. 1411–1420.10.1061/(ASCE)CO.1943-7862.0000557Search in Google Scholar

[35] Iweka, S. C., K. C. Owuama, J. L. Chukwuneke, and O. A. Falowo. Optimization of biogas yield from anaerobic co-digestion of corn-chaff and cow dung digestate: RSM and python approach. Heliyon, Vol. 7, No. 11, 2021, id. e08255.10.1016/j.heliyon.2021.e08255Search in Google Scholar PubMed PubMed Central

[36] Borode, A. and P. Olubambi. Modelling the effects of mixing ratio and temperature on the thermal conductivity of GNP-Alumina hybrid nanofluids: A comparison of ANN, RSM, and linear regression methods. Heliyon, Vol. 9, No. 8, 2023, id. e19228.10.1016/j.heliyon.2023.e19228Search in Google Scholar PubMed PubMed Central

[37] Alateyah, A. I., M. El-Shenawy, A. Nassef, M. El-Hadek, M. M. Z. Ahmed, H. Kouta, et al. Optimizing the ECAP processing parameters of pure Cu through experimental, finite element, and response surface approaches. Reviews on Advanced Materials Science, Vol. 62, No. 1, 2023, id. 20220297.10.1515/rams-2022-0297Search in Google Scholar

[38] Azargohar, R. and A. Dalai. Production of activated carbon from Luscar char: experimental and modeling studies. Microporous and Mesoporous Materials, Vol. 85, No. 3, 2005, pp. 219–225.10.1016/j.micromeso.2005.06.018Search in Google Scholar

[39] Baghaee Moghaddam, T., M. Soltani, and M. R. Karim. Stiffness modulus of Polyethylene Terephthalate modified asphalt mixture: A statistical analysis of the laboratory testing results. Materials & Design, Vol. 68, 2015, pp. 88–96.10.1016/j.matdes.2014.11.044Search in Google Scholar

[40] Montgomery, D. Design and analysis of experiments, Jon Wiley & Sons. Inc, Hoboken, NJ, 2012.Search in Google Scholar

[41] Bala, N., M. Napiah, and I. Kamaruddin. Nanosilica composite asphalt mixtures performance-based design and optimisation using response surface methodology. International Journal of Pavement Engineering, Vol. 21, No. 1, 2020, pp. 29–40.10.1080/10298436.2018.1435881Search in Google Scholar

[42] Baghaee Moghaddam, T., M. Soltani, M. R. Karim, and H. Baaj. Optimization of asphalt and modifier contents for polyethylene terephthalate modified asphalt mixtures using response surface methodology. Measurement, Vol. 74, 2015, pp. 159–169.10.1016/j.measurement.2015.07.012Search in Google Scholar

[43] Khodaii, A., H. F. Haghshenas, and H. Kazemi Tehrani. Effect of grading and lime content on HMA stripping using statistical methodology. Construction and Building Materials, Vol. 34, 2012, pp. 131–135.10.1016/j.conbuildmat.2012.02.025Search in Google Scholar

[44] Haghshenas, H. F., A. Khodaii, M. Khedmati, and S. Tapkin. A mathematical model for predicting stripping potential of Hot Mix Asphalt. Construction and Building Materials, Vol. 75, 2015, pp. 488–495.10.1016/j.conbuildmat.2014.11.041Search in Google Scholar

[45] Lv, S., J. Yuan, X. Peng, M. Borges Cabrera, S. Guo, X. Luo, et al. Performance and optimization of bio-oil/Buton rock asphalt composite modified asphalt. Construction and Building Materials, Vol. 264, 2020, id. 120235.10.1016/j.conbuildmat.2020.120235Search in Google Scholar

[46] Yıldırım, Z. B. and M. Karacasu. Modelling of waste rubber and glass fibber with response surface method in hot mix asphalt. Construction and Building Materials, Vol. 227, 2019, id. 117070.10.1016/j.conbuildmat.2019.117070Search in Google Scholar

[47] Xu, W., T. Y. Lo, and S. A. Memon. Microstructure and reactivity of rich husk ash. Construction and Building Materials, Vol. 29, 2012, pp. 541–547.10.1016/j.conbuildmat.2011.11.005Search in Google Scholar

[48] JTG E42-2005. Test Methods of Aggregate for Highway Engineering, Ministry of Transportation Highway Research Institute, China: Beijing, 2005.Search in Google Scholar

[49] JTG. JTG E20–2011 ‘Standard test methods of bitumen and bituminous mixtures for highway engineering’, China Ministry of Transport, Beijing, 2011.Search in Google Scholar

[50] JTG. JTG F40–2004 ‘Technical specifications for construction of highway asphalt pavements, China Ministry of Transport, Beijing, 2004.Search in Google Scholar

[51] Wang, Y., Q. Zhang and M. Liu. Analysis of variance. in Textbook of Medical Statistics: For Medical Students, Guo, X. and F. Xue, eds, Springer Nature Singapore, Singapore, 2024, pp. 99–124.10.1007/978-981-99-7390-3_8Search in Google Scholar

[52] Saisana, M. Analysis of Variance. in Encyclopedia of quality of life and well-being research, F. Maggino, ed., Springer International Publishing, Cham, 2023, pp. 183–187.10.1007/978-3-031-17299-1_83Search in Google Scholar

[53] Xu, S., J. Yu, C. Hu, L. Jia, and L. Xue. Evaluation of aging performance of bitumen containing layered double hydroxides intercalated by UV absorbents. International Journal of Pavement Engineering, Vol. 20, No. 4, 2019, pp. 499–505.10.1080/10298436.2017.1309198Search in Google Scholar

[54] Körbahti, B. K. and M. A. Rauf. Application of response surface analysis to the photolytic degradation of Basic Red 2 dye. Chemical Engineering Journal, Vol. 138, No. 1, 2008, pp. 166–171.10.1016/j.cej.2007.06.016Search in Google Scholar

[55] Hamdane, H., Y. Tamraoui, S. Mansouri, M. Oumam, A. Bouih, T. El Ghailassi, et al. Statistical modeling of geopolymers from dual-alkali activation of un-calcined phosphate sludge and their potential applications as sustainable coating materials. Journal of Cleaner Production, Vol. 283, 2021, id. 125421.10.1016/j.jclepro.2020.125421Search in Google Scholar

[56] Hafeez, A., S. A. Ammar Taqvi, T. Fazal, F. Javed, Z. Khan, U. S. Amjad, et al. Optimization on cleaner intensification of ozone production using artificial neural network and response surface methodology: Parametric and comparative study. Journal of Cleaner Production, Vol. 252, 2020, id. 119833.10.1016/j.jclepro.2019.119833Search in Google Scholar
Received: 2024-11-28
Revised: 2025-02-27
Accepted: 2025-04-07
Published Online: 2025-05-03

© 2025 the author(s), published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

  1. Review Articles
  2. Utilization of steel slag in concrete: A review on durability and microstructure analysis
  3. Technical development of modified emulsion asphalt: A review on the preparation, performance, and applications
  4. Recent developments in ultrasonic welding of similar and dissimilar joints of carbon fiber reinforcement thermoplastics with and without interlayer: A state-of-the-art review
  5. Unveiling the crucial factors and coating mitigation of solid particle erosion in steam turbine blade failures: A review
  6. From magnesium oxide, magnesium oxide concrete to magnesium oxide concrete dams
  7. Properties and potential applications of polymer composites containing secondary fillers
  8. A scientometric review on the utilization of copper slag as a substitute constituent of ordinary Portland cement concrete
  9. Advancement of additive manufacturing technology in the development of personalized in vivo and in vitro prosthetic implants
  10. Recent advance of MOFs in Fenton-like reaction
  11. A review of defect formation, detection, and effect on mechanical properties of three-dimensional braided composites
  12. Non-conventional approaches to producing biochars for environmental and energy applications
  13. Review of the development and application of aluminum alloys in the nuclear industry
  14. Advances in the development and characterization of combustible cartridge cases and propellants: Preparation, performance, and future prospects
  15. Recent trends in rubberized and non-rubberized ultra-high performance geopolymer concrete for sustainable construction: A review
  16. Research Articles
  17. Investigation of the corrosion performance of HVOF-sprayed WC-CoCr coatings applied on offshore hydraulic equipment
  18. A systematic review of metakaolin-based alkali-activated and geopolymer concrete: A step toward green concrete
  19. Evaluation of color matching of three single-shade composites employing simulated 3D printed cavities with different thicknesses using CIELAB and CIEDE2000 color difference formulae
  20. Novel approaches in prediction of tensile strain capacity of engineered cementitious composites using interpretable approaches
  21. Effect of TiB2 particles on the compressive, hardness, and water absorption responses of Kulkual fiber-reinforced epoxy composites
  22. Analyzing the compressive strength, eco-strength, and cost–strength ratio of agro-waste-derived concrete using advanced machine learning methods
  23. Tensile behavior evaluation of two-stage concrete using an innovative model optimization approach
  24. Tailoring the mechanical and degradation properties of 3DP PLA/PCL scaffolds for biomedical applications
  25. Optimizing compressive strength prediction in glass powder-modified concrete: A comprehensive study on silicon dioxide and calcium oxide influence across varied sample dimensions and strength ranges
  26. Experimental study on solid particle erosion of protective aircraft coatings at different impact angles
  27. Compatibility between polyurea resin modifier and asphalt binder based on segregation and rheological parameters
  28. Fe-containing nominal wollastonite (CaSiO3)–Na2O glass-ceramic: Characterization and biocompatibility
  29. Relevance of pore network connectivity in tannin-derived carbons for rapid detection of BTEX traces in indoor air
  30. A life cycle and environmental impact analysis of sustainable concrete incorporating date palm ash and eggshell powder as supplementary cementitious materials
  31. Eco-friendly utilisation of agricultural waste: Assessing mixture performance and physical properties of asphalt modified with peanut husk ash using response surface methodology
  32. Benefits and limitations of N2 addition with Ar as shielding gas on microstructure change and their effect on hardness and corrosion resistance of duplex stainless steel weldments
  33. Effect of selective laser sintering processing parameters on the mechanical properties of peanut shell powder/polyether sulfone composite
  34. Impact and mechanism of improving the UV aging resistance performance of modified asphalt binder
  35. AI-based prediction for the strength, cost, and sustainability of eggshell and date palm ash-blended concrete
  36. Investigating the sulfonated ZnO–PVA membrane for improved MFC performance
  37. Strontium coupling with sulphur in mouse bone apatites
  38. Transforming waste into value: Advancing sustainable construction materials with treated plastic waste and foundry sand in lightweight foamed concrete for a greener future
  39. Evaluating the use of recycled sawdust in porous foam mortar for improved performance
  40. Improvement and predictive modeling of the mechanical performance of waste fire clay blended concrete
  41. Polyvinyl alcohol/alginate/gelatin hydrogel-based CaSiO3 designed for accelerating wound healing
  42. Research on assembly stress and deformation of thin-walled composite material power cabin fairings
  43. Effect of volcanic pumice powder on the properties of fiber-reinforced cement mortars in aggressive environments
  44. Analyzing the compressive performance of lightweight foamcrete and parameter interdependencies using machine intelligence strategies
  45. Selected materials techniques for evaluation of attributes of sourdough bread with Kombucha SCOBY
  46. Establishing strength prediction models for low-carbon rubberized cementitious mortar using advanced AI tools
  47. Investigating the strength performance of 3D printed fiber-reinforced concrete using applicable predictive models
  48. An eco-friendly synthesis of ZnO nanoparticles with jamun seed extract and their multi-applications
  49. The application of convolutional neural networks, LF-NMR, and texture for microparticle analysis in assessing the quality of fruit powders: Case study – blackcurrant powders
  50. Study of feasibility of using copper mining tailings in mortar production
  51. Shear and flexural performance of reinforced concrete beams with recycled concrete aggregates
  52. Advancing GGBS geopolymer concrete with nano-alumina: A study on strength and durability in aggressive environments
  53. Special Issue on Recent Advancement in Low-carbon Cement-based Materials - Part II
  54. Investigating the effect of locally available volcanic ash on mechanical and microstructure properties of concrete
  55. Flexural performance evaluation using computational tools for plastic-derived mortar modified with blends of industrial waste powders
  56. Foamed geopolymers as low carbon materials for fire-resistant and lightweight applications in construction: A review
  57. Autogenous shrinkage of cementitious composites incorporating red mud
  58. Special Issue on AI-Driven Advances for Nano-Enhanced Sustainable Construction Materials
  59. Advanced explainable models for strength evaluation of self-compacting concrete modified with supplementary glass and marble powders
https://doi.org/10.1515/rams-2025-0107
Keywords for this article
agricultural waste; peanut husk ash; asphalt mixture; sustainable construction materials
Creative Commons
BY 4.0

Articles in the same Issue

  1. Review Articles
  2. Utilization of steel slag in concrete: A review on durability and microstructure analysis
  3. Technical development of modified emulsion asphalt: A review on the preparation, performance, and applications
  4. Recent developments in ultrasonic welding of similar and dissimilar joints of carbon fiber reinforcement thermoplastics with and without interlayer: A state-of-the-art review
  5. Unveiling the crucial factors and coating mitigation of solid particle erosion in steam turbine blade failures: A review
  6. From magnesium oxide, magnesium oxide concrete to magnesium oxide concrete dams
  7. Properties and potential applications of polymer composites containing secondary fillers
  8. A scientometric review on the utilization of copper slag as a substitute constituent of ordinary Portland cement concrete
  9. Advancement of additive manufacturing technology in the development of personalized in vivo and in vitro prosthetic implants
  10. Recent advance of MOFs in Fenton-like reaction
  11. A review of defect formation, detection, and effect on mechanical properties of three-dimensional braided composites
  12. Non-conventional approaches to producing biochars for environmental and energy applications
  13. Review of the development and application of aluminum alloys in the nuclear industry
  14. Advances in the development and characterization of combustible cartridge cases and propellants: Preparation, performance, and future prospects
  15. Recent trends in rubberized and non-rubberized ultra-high performance geopolymer concrete for sustainable construction: A review
  16. Research Articles
  17. Investigation of the corrosion performance of HVOF-sprayed WC-CoCr coatings applied on offshore hydraulic equipment
  18. A systematic review of metakaolin-based alkali-activated and geopolymer concrete: A step toward green concrete
  19. Evaluation of color matching of three single-shade composites employing simulated 3D printed cavities with different thicknesses using CIELAB and CIEDE2000 color difference formulae
  20. Novel approaches in prediction of tensile strain capacity of engineered cementitious composites using interpretable approaches
  21. Effect of TiB2 particles on the compressive, hardness, and water absorption responses of Kulkual fiber-reinforced epoxy composites
  22. Analyzing the compressive strength, eco-strength, and cost–strength ratio of agro-waste-derived concrete using advanced machine learning methods
  23. Tensile behavior evaluation of two-stage concrete using an innovative model optimization approach
  24. Tailoring the mechanical and degradation properties of 3DP PLA/PCL scaffolds for biomedical applications
  25. Optimizing compressive strength prediction in glass powder-modified concrete: A comprehensive study on silicon dioxide and calcium oxide influence across varied sample dimensions and strength ranges
  26. Experimental study on solid particle erosion of protective aircraft coatings at different impact angles
  27. Compatibility between polyurea resin modifier and asphalt binder based on segregation and rheological parameters
  28. Fe-containing nominal wollastonite (CaSiO3)–Na2O glass-ceramic: Characterization and biocompatibility
  29. Relevance of pore network connectivity in tannin-derived carbons for rapid detection of BTEX traces in indoor air
  30. A life cycle and environmental impact analysis of sustainable concrete incorporating date palm ash and eggshell powder as supplementary cementitious materials
  31. Eco-friendly utilisation of agricultural waste: Assessing mixture performance and physical properties of asphalt modified with peanut husk ash using response surface methodology
  32. Benefits and limitations of N2 addition with Ar as shielding gas on microstructure change and their effect on hardness and corrosion resistance of duplex stainless steel weldments
  33. Effect of selective laser sintering processing parameters on the mechanical properties of peanut shell powder/polyether sulfone composite
  34. Impact and mechanism of improving the UV aging resistance performance of modified asphalt binder
  35. AI-based prediction for the strength, cost, and sustainability of eggshell and date palm ash-blended concrete
  36. Investigating the sulfonated ZnO–PVA membrane for improved MFC performance
  37. Strontium coupling with sulphur in mouse bone apatites
  38. Transforming waste into value: Advancing sustainable construction materials with treated plastic waste and foundry sand in lightweight foamed concrete for a greener future
  39. Evaluating the use of recycled sawdust in porous foam mortar for improved performance
  40. Improvement and predictive modeling of the mechanical performance of waste fire clay blended concrete
  41. Polyvinyl alcohol/alginate/gelatin hydrogel-based CaSiO3 designed for accelerating wound healing
  42. Research on assembly stress and deformation of thin-walled composite material power cabin fairings
  43. Effect of volcanic pumice powder on the properties of fiber-reinforced cement mortars in aggressive environments
  44. Analyzing the compressive performance of lightweight foamcrete and parameter interdependencies using machine intelligence strategies
  45. Selected materials techniques for evaluation of attributes of sourdough bread with Kombucha SCOBY
  46. Establishing strength prediction models for low-carbon rubberized cementitious mortar using advanced AI tools
  47. Investigating the strength performance of 3D printed fiber-reinforced concrete using applicable predictive models
  48. An eco-friendly synthesis of ZnO nanoparticles with jamun seed extract and their multi-applications
  49. The application of convolutional neural networks, LF-NMR, and texture for microparticle analysis in assessing the quality of fruit powders: Case study – blackcurrant powders
  50. Study of feasibility of using copper mining tailings in mortar production
  51. Shear and flexural performance of reinforced concrete beams with recycled concrete aggregates
  52. Advancing GGBS geopolymer concrete with nano-alumina: A study on strength and durability in aggressive environments
  53. Special Issue on Recent Advancement in Low-carbon Cement-based Materials - Part II
  54. Investigating the effect of locally available volcanic ash on mechanical and microstructure properties of concrete
  55. Flexural performance evaluation using computational tools for plastic-derived mortar modified with blends of industrial waste powders
  56. Foamed geopolymers as low carbon materials for fire-resistant and lightweight applications in construction: A review
  57. Autogenous shrinkage of cementitious composites incorporating red mud
  58. Special Issue on AI-Driven Advances for Nano-Enhanced Sustainable Construction Materials
  59. Advanced explainable models for strength evaluation of self-compacting concrete modified with supplementary glass and marble powders
Sign up now to receive a 20% welcome discount
Institutional Access
How does access work?
Have an idea on how to improve our website?
Please write us.
© 2025 De Gruyter Brill
Downloaded on 14.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/rams-2025-0107/html
Scroll to top button