Influence of Modified Stalk Fibers on the Fatigue Performance of Asphalt Binder

Abstract

The type and content of modified stalk fibers significantly influence the fatigue properties of asphalt binder. In this study, different concentrations of NaOH solution were used to modify stalk fibers, and scanning electron microscopy (SEM) was used to observe the effect of the modified concentration on the fiber morphology. A dynamic shear rheology (DSR) test and a linear amplitude sweep (LAS) test were conducted to analyze the effects of the fiber type and content on various factors such as the complex shear modulus G*, phase angle δ, and fatigue parameters (A₃₅ and B). Consequently, the fatigue life N_f of the fiber asphalt binder was calculated using a viscoelastic continuum damage model. The results show that stalk fibers modified using a 5% alkali solution exhibited the best oil absorption and heat resistance, the asphalt binder with a 1.5%–2% fiber content exhibited the best resistance to fatigue, and the fatigue performance of the asphalt binder with different types of fibers was superior when fiber doping was at 1.5%. Additionally, the fatigue parameter A₃₅ of the modified cotton and corn stover fibers increased by 40.5% and 57.6%, respectively, and the fatigue parameter B decreased by 5.8% and 4.8%, respectively, compared with that of the unmodified stover fibers. Finally, the modified corn stalk fiber asphalt binder with a 1.5% fiber content demonstrated the best fatigue resistance.

1. Introduction

Fatigue cracking is a major problem in asphalt pavements. Modern asphalt binder theories indicate that asphalt binders are important components of bituminous mixtures and that their properties have a significant impact on the fatigue performance of bituminous mixtures [1,2]. Existing research has demonstrated that fibers can be used to stabilize, reinforce, and increase the viscosity of asphalt binders; therefore, they are often added to asphalt binders as modifiers to enhance their fatigue performance [3,4]. Fibers are renewable resources with the advantages of environmental friendliness, wide availability, and a low cost. They are typically used to replace traditional fiber modifiers such as basalt, glass, and polyester fibers. However, existing agricultural stalk fibers, such as cotton and corn stalk fibers, have disadvantages, such as a poor heat resistance and small specific surface area. Therefore, there is an urgent need to modify stalk fibers by removing pectin and waxes from the fiber surfaces, improving their thermal stability, and increasing their surface roughness to enhance their bonding with the binder. Consequently, the fatigue performance of the fiber asphalt binder is improved.

Numerous studies on the fatigue performance of conventional fiber-modified asphalt binders have been conducted in China and abroad. Zhu et al. conducted dynamic shear rheology (DSR) tests and concluded that conventional fibers can significantly improve the fatigue performance of asphalt binders and that changes in the fiber content have a significant effect on the fatigue performance of the binder [5,6]. In recent years, new crop of stalk fibers have gained increasing attention owing to their economic and environmentally friendly features. Liu et al. used crop stalk fibers to modify asphalt and found that crop stalk fibers can be used to replace traditional fibers and exhibit a more evident improvement effect on asphalt oil absorption and anti-fatigue performance [7,8]. However, numerous studies have reported that the surfaces of stalk fibers contain more silicon and wax, and the fat and pectin in the fibers degrade into small molecular compounds at high temperatures. This reduces the adsorption of substrate materials by the crop stalk fibers, thereby causing difficulty in the formation of an ideal bonding interface on the modified asphalt [9,10]. To address this problem, various pretreatment modification methods for fibers, generally classified as biological, physical, or chemical, have been proposed. Biological modification methods primarily include microbial methods and biological enzymes. Physical modification methods primarily include steam blasting and mechanical crushing, and chemical modification methods primarily include acid treatment and alkali treatment methods [11]. Alkali treatment is the most valuable and promising method for the chemical modification of crop stalk fibers. Kuity reported that the physical properties of the filler, particularly the specific surface area characteristics, had a more significant effect on the fatigue performance of an asphalt binder than the chemical properties [12]. A high-concentration NaOH solution is an ideal modifier for crop stalk fibers because it effectively removes pectin and waxes from the fibers in advance via saponification and by turning the fiber bundles into smaller single bundles, thereby increasing the specific surface area and surface roughness of the fibers [13,14]. Most of the existing research on the use of crop stalk fibers in asphalt focuses on the mixture preparation process and road performance, whereas research on the modification of asphalt binders is scarce. Essentially, the existing research on crop stalk fiber is incomprehensive. In addition, most of these studies, particularly those researching the modification of fibers, only focused on a single fiber and did not compare a variety of fibers under the same experimental conditions. In this study, lignin, cotton stalk, and corn stalk fibers were all used in the study of asphalt binder fatigue performance, and NaOH solution was used to chemically modify the stalk fibers. Experimental designs and research were conducted on fatigue performance, which has practical research significance and broad application prospects.

In this study, based on the requirements of lignin fiber in asphalt pavement, the fatigue performance of a crop-stalk-fiber-modified asphalt binder was investigated in depth based on previous research. 3wt%, 5wt%, and 7wt% NaOH solutions were used as the fiber modifier, and three types of fibers were used for the asphalt matrix experiments: lignin fiber, cotton stalk fiber, and corn stalk fiber. An asphalt binder without any added fibers was set up as a control, and five experimental groups were set up with added lignin fibers, modified cotton stalk fibers, cotton stalk fibers, modified corn stalk fibers, and corn stalk fibers. A scanning electron microscope (SEM) was used to observe the surface morphology characteristics of cotton and corn stalk fibers modified using different concentrations of NaOH solution and to analyze the optimal alkali solution concentration for stalk fiber surface modification. The effects of the fiber type and dosage on the fatigue performance of the asphalt binder were analyzed via a dynamic shear rheology (DSR) test and a linear amplitude sweep (LAS) test [15,16]. The fatigue life of the fiber asphalt binder was calculated using a viscoelastic continuum damage (VECD) model to provide an experimental basis and theoretical support for the fiber type selection and reasonable dosage determination. A flowchart of this research is shown in Figure 1.

2. Materials and Methods

2.1. Materials

2.1.1. Asphalt

Asphalt of the type 70# road petroleum provided by the Qilu Branch of Sinopec was used in this study. The main asphalt technical indicators are listed in

Table 1. Major asphalt technical indicators.

Tested Property	Units	Value	Specification	Test Method
Penetration (25 °C)	0.1 mm	68	60–80	JTG E20-2011 [17]
Softening point	°C	47.4	≥46
Ductility (10 °C)	cm	30	≥15
Ductility (15 °C)	cm	≥150	≥100
Density (15 °C)	g/cm3	1.036	-
Flash point	°C	268	≥260
Solubility	%	99.7	≥99.5

.

2.1.2. Filler

Ground limestone powder was selected as the filler. The main filler technical specifications are listed in

Table 2. Major mineral powder technical indicators.

Index	Apparent Density (t/m3)	Hydrophilic Coefficient (%)	Moisture Content (%)	The Content of <0.075 mm (%)	Test Method
Test Results	2.796	0.41	0.32	85.7	JTG E42-2005 [18]

.

2.2. Stalk Fiber Modification and Performance Analysis

2.2.1. Preparation of the Modified Stalk Fibers

In this study, different concentrations of NaOH solution were used to modify the stalk fiber. First, 3wt%, 5wt%, and 7wt% NaOH solutions were prepared with deionized water. To do this, add 15 g of stalk fiber to 500 mL of alkali solution, put it in a thermostatic magnetic mixer, and mix for 30 min at a constant speed (as shown in Figure 2). After the chemical treatment, the residual alkali solution was washed off the surface of the fibers using deionized water, and the fibers were heated at a low temperature to a constant weight in an electric blast dryer at 45 °C. A comparison of the straw fibers before and after modification is shown in Figure 3.

2.2.2. Modified Stalk Fiber Conventional Technical Performance

The physical properties of the different fibers are listed in

Table 3. Physical properties of the fibers.

Item	Fiber Type					Test Method
	Lignin Fiber	Modified Cotton Stalk Fiber	Cotton Stalk Fiber	Modified Corn Stalk Fiber	Corn Stalk Fiber
Fiber length (mm)	<5	≤4	≤4	≤4	≤4	JT/T 533-2004 [20]
PH	8.4	7.97	6.5	8.35	7.06
Oil Absorption rate	8.13	7.81	6.3	7.70	5.71
Thermal loss (%)	5.6	5.5	9	5.6	7.3

. When taking 5wt% modified concentration as an example, the two stalk fibers exhibit weak alkalinity upon NaOH modification. Additionally, the modified stalk fiber can be effectively combined with asphalt [19] to attain better reinforcement, stabilization, and viscosity, which can increase the oil absorption times of stalk fibers and improve the bond between the fiber and asphalt. Simultaneously, the heat resistance of the modified fiber is enhanced, which can avoid the heat loss of the fiber during the mixing of the asphalt binder and mixture and increase the durability of the fiber.

2.2.3. Microscopic Morphological Analysis of Modified Stalk Fibers

The adsorption of asphalt by the fibers is related to the diameter of the fibers, surface roughness, and other physical properties. In this study, the fibers were modified using NaOH to enhance their surface roughness via alkali erosion. The bumps and pits generated on the fiber surface are conducive to the adsorption of asphalt and increase the interlocking at the interface with asphalt. In this way, a three-dimensional spatial network structure is formed inside the asphalt. The structure plays the role of a bridge, making the fiber and asphalt more closely combined, so that the asphalt has better stability at high temperatures and has improved resistance to crack growth.

We performed a scanning observation of the corn and cotton stalk fiber samples before and after modification with different concentration of alkali solution by SEM. The microscopic morphological characteristics of the fiber surfaces were observed using a 1000× field of view, as shown in Figure 4 and Figure 5.

It can be seen in Figure 4a and Figure 5a that the two kinds of unmodified stalk fibers are thick and straight under a 1000× field of vision: the surface of the corn stalk fiber is smooth, and the surface of the cotton stalk fiber is rough. After modification with different concentrations of alkali solution, the lignin components in corn stalk fiber were eroded by the alkali solution, resulting in many narrow and long micro-gaps on the fiber surface; a small number of lumen structures collapsed, and the inner area of the lumen was exposed. With an increase in modifier concentration, the number of narrow and long micro-gaps increases, and the collapse of the lumen structure deepens. As can be seen in Figure 4d, when the modifier concentration is 7wt%, the corn stalk fiber is seriously eroded by the alkali liquor, resulting in a decline in the overall mechanical strength of the fiber. Impurity molecules such as pectin and hemicellulose on the surface of cotton stalk fiber are dissolved by the NaOH solution, which lacks the binding effect of pectin. The fiber bundle is more likely to become loose and separate, resulting in some monofilament fibers on the surface of stalk fiber, thus increasing the roughness of the fiber surface. When the concentration of modifier increased from 5wt% to 7wt%, the erosion of stalk fiber by alkali solution deepened, the number of monofilament fibers on the fiber surface increased sharply, and the original structure of fiber was destroyed by the alkali solution.

To sum up, NaOH-solution-modified stalk fiber can increase the surface roughness of the fiber, essentially increasing the specific surface area of the stalk fiber, so that the effective contact area between the modified stalk fiber and the asphalt increases under the same fiber content, thus increasing the adsorption capacity of the fiber to the asphalt. However, too high a concentration of alkali liquor will destroy the original structure of the fiber and reduce the overall stability of the fiber, so for this study, we decided to use 5wt% NaOH solution as the stalk fiber modifier.

2.3. Analysis of the Chemical Composition of Stalk Fibers

To study the changes in the composition of corn and cotton stalk fibers before and after modification, data were collected and analyzed using a Fourier transform infrared spectrometer. The results are shown in Figure 5.

As shown in Figure 6, the shapes of the stalk fiber spectra before and after modification are similar, and the main difference lies in the change in the intensity of the absorption peaks. The main reason for the weakening of the peak intensities of the -OH group of the modified stalk fibers is the chemical reaction between the -OH and NaOH. The absorption peaks located near 1730 cm⁻¹ in the corn and cotton stalk fibers, which were formed owing to the stretching vibration of C=O bonds in the hemicellulose, are significantly reduced after modification, indicating that the C=O bonds were destroyed, and part of the hemicellulose was dissolved. The continuous vibration absorption peaks located near 1515 cm⁻¹ and 1650 cm⁻¹ in the modified corn and cotton stalk fibers are also significantly weakened, indicating that the lignin structure of the corn and cotton stalk fibers was partially decomposed after modification. This is confirmed by the collapse of the cavity structure in the microscopic morphology diagram. Additionally, the absorption peak at 1247 cm⁻¹ in the modified stalk fiber almost disappears, indicating that a saponification reaction caused by the action of NaOH during modification occurred, resulting in the breakage of the acetyl ester bond between the molecular structures of cellulose, hemicellulose, and lignin in the modified maize and cotton stalk fibers. This led to the partial dissolution of cellulose, hemicellulose, and lignin.

2.4. Preparation of the Fiber Asphalt Binder

The fiber asphalt binder samples were prepared using a high-speed shear mechanism, and the specific process is outlined below. First, the fiber and mineral powders were dried in an oven at 130 °C until they were completely dehydrated. The fiber and mineral powders were evenly mixed using the dry mixing method, in which the fiber contents (the proportion of fiber to matrix asphalt) were 0, 1.5, 2, and 2.5%, and the powder/asphalt ratio (by weight) was 0.8. The matrix asphalt was heated in an oven at 135 °C and placed in a constant-temperature heating sleeve at 145 ± 5 °C before the evenly mixed fiber and mineral powder was added. A high-speed shear machine was used for shearing at a low speed of 2000 rpm for 10 min, followed by a high speed of 5000 rpm for 40 min. Therefore, the mineral powder and fibers in the asphalt matrix were evenly dispersed. Finally, shearing was conducted at a low speed of 1500 rpm for 10 min to eliminate excess bubbles in the fiber asphalt binder and make it stable. In order to prevent segregation, the prepared sample should be tested in the next step as soon as possible.

2.5. Experimental Methods

2.5.1. Dynamic Shear Rheology (DSR) Test

In project 9–10 of the National Cooperative Highway Research Program (NCHRP), a DSR test was proposed to characterize the fatigue performance of asphalt using the degree of decay of the complex shear modulus (G*) and phase angle (δ) based on the representation method and dissipative energy method [21]. In this study, a DSR was used to obtain the appropriate dosage range of the fiber fatigue performance [22].

The stalk fiber asphalt binder was tested before and after modification, and lignin fiber asphalt binders were used as the control group for comparative analysis. The tests were conducted at 25 °C, with a loading frequency of 10 rad/s and a strain level of 1.5%. Based on the storage modulus and loss modulus obtained in the experiment, the influence of the content and type of modified stalk fiber on the fatigue performance of the binder was evaluated.

But the loss modulus (G*sin δ) being used as a fatigue factor to indicate an asphalt binder’s long-term performance has caused a lot of controversies. Firstly, the loss modulus does not consider the nonlinear viscoelastic behavior of the asphalt binder and cannot reflect the actual mechanical state and process of pavement fatigue cracking. Secondly, the loss modulus draws a conclusion about fatigue performance by taking base asphalt as the research object and does not apply to modified asphalt [23].

Therefore, a few studies have been conducted to develop an accurate, effective, and time-saving binder fatigue test, which led to the introduction of the LAS test [24,25,26].

2.5.2. Linear Amplitude Sweep (LAS) Test

Based on the DSR test, Johnson et al. proposed the LAS test to study the damage resistance ability of asphalt under cyclic loading and predicted the fatigue life of asphalt based on the viscoelastic continuum damage (VECD) model [27,28]. The LAS test was divided into two parts: frequency and amplitude scanning. In frequency scanning, the load was applied to the sample at a strain level of 1.25%, at a test temperature of 25 °C, and with a frequency range of 0.2–30 Hz to determine the asphalt rheological properties and obtain the damage distribution coefficient.

During the linear amplitude sweep test, the temperature was 25 °C, loading frequency was 10 Hz under the strain control mode, loading time was 300 s, and loading amplitude increased linearly from 0.1% to 30% to accelerate fatigue damage.

The parameters used to characterize the rheological properties of the asphalt were determined using data obtained from the frequency scanning test α (Equation (1)):

$$\mathsf{\alpha }=1+\frac{1}{\mathrm{m}}$$

(1)

where the gradient “m” is determined via linear fitting of the logarithm of the storage modulus “${\mathrm{G}}^{\prime }$($\mathsf{\omega }$)” and angular velocity “ω” (Equation (2)):

$$\mathrm{l}\mathrm{o}\mathrm{g}{\mathrm{G}}^{\prime }\left(\mathsf{\omega }\right)=\mathrm{m}\times \mathrm{l}\mathrm{o}\mathrm{g}\left(\mathsf{\omega }\right)+\mathrm{b}$$

(2)

where ${\mathrm{G}}^{\prime }$($\mathsf{\omega }$) = |G*| cosδ($\mathsf{\omega }$) (MPa); herein, ω represents the angle frequency (Hz).

Based on the amplitude scanning results, the cumulative damage D(t) of the asphalt binder was calculated as follows (Equation (3)):

$$\mathrm{D}(\mathrm{t})\cong {\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{N}}{[\mathsf{\pi } \mathrm{I}}_{\mathrm{D}}{\mathsf{\gamma }}_0^2(\left|{\mathrm{G}}^{*}\right|\sin }{\mathsf{\delta }}_{\mathrm{i}-1}-\left|{\mathrm{G}}^{*}\right|{\sin \mathsf{\delta }}_{\mathrm{i}}){]}^{\frac{\mathsf{\alpha }}{1+\mathsf{\alpha }}}({\mathrm{t}}_{\mathrm{i}}-{\mathrm{t}}_{\mathrm{i}-1}{)}^{\frac{1}{1+\mathsf{\alpha }}}$$

(3)

where:

I_D—initial damaged value of |G*| from the 1.0% applied strain interval, (MPa);

γ₀—Strain value applied to a given data point, dimensionless;

α—Undamaged asphalt binder parameters calculated from (1);

t—Testing time (s).

Assuming that D(0) = 0, the relationship between D(t) and |G*| sinδ can be obtained as follows (Equation (4)):

$$\left|{\mathrm{G}}^{*}\right|\sin \mathsf{\delta }={\mathrm{C}}_0-{\mathrm{C}}_1(\mathrm{D}{)}^{{\mathrm{C}}_2}$$

(4)

where:

C₀—constant, whose value is 1;

C₁—anti-logarithm of the intercept, calculated according to Equation (5);

C₂—gradient.

$$\log ({\mathrm{C}}_0-\left|{\mathrm{G}}^{*}\right|\sin \mathsf{\delta })={\mathrm{C}}_2\log (\mathrm{D})+\log ({\mathrm{C}}_1)$$

(5)

When |G*| sinδ drops to 35% of the initial value, the corresponding D(t) value is used as the evaluation standard for fatigue failure, D_f (Equation (6)):

$${\mathrm{D}}_{\mathrm{f}}=0.35(\frac{{\mathrm{C}}_0}{{\mathrm{C}}_1}{)}^{\frac{1}{{\mathrm{C}}_2}}$$

(6)

Based on the above known parameters, the performance parameters A₃₅ and B of the asphalt binder in the VECD fatigue model can be obtained using Equations (7) and (8):

$${\mathrm{A}}_{35}=\frac{\mathrm{f}{({\mathrm{D}}_{\mathrm{f}})}^{\mathrm{k}}}{\mathrm{k}({\mathsf{\pi }\mathrm{I}}_{\mathrm{D}}{\mathrm{C}}_1{\mathrm{C}}_2{)}^{\mathsf{\alpha }}}$$

(7)

$$\mathrm{B}=2\mathsf{\alpha }$$

(8)

where:

f—loading frequency, which is 10 Hz;

k—1 + (1 − C₂)α.

The fatigue life N_f of the asphalt binder is calculated based on Equations (5)–(9)

$${\mathrm{N}}_{\mathrm{f}}=\mathrm{A}{({\mathsf{\gamma }}_{\max })}^{-\mathrm{B}}$$

(9)

where γ_max is the maximum expected strain of the asphalt binder in a given pavement structure (%).

3. Results

3.1. Analysis of the DSR Test Results

Dynamic shear rheology tests were conducted on the stalk fiber asphalt binder and lignin fiber asphalt binder using mixing proportions of 0, 1.5, 2, and 2.5%. The test results are listed in

Table 4. DSR test results for the different asphalt binders.

Asphalt Binder Type	Storage Modulus (MPa)	Loss Modulus (MPa)	Tan (δ)	Phase Angle (Degree)
1.5% Cotton stalk fiber	1.28	2.07	1.61	58.17
2.0% Cotton stalk fiber	0.78	1.40	1.79	60.81
2.5% Cotton stalk fiber	1.06	1.76	1.65	58.86
1.5% Modified cotton stalk fiber	1.32	1.94	1.47	55.72
2.0% Modified cotton stalk fiber	0.89	1.48	1.65	58.85
2.5% Modified cotton stalk fiber	0.92	1.53	1.66	58.96
1.5% Corn stalk fiber	1.24	1.67	1.35	53.41
2.0% Corn stalk fiber	1.21	1.87	1.54	57.04
2.5% Corn stalk fiber	1.47	2.14	1.46	55.64
1.5% Modified corn stalk fiber	1.05	1.53	1.46	55.62
2.0% Modified corn stalk fiber	0.95	1.48	1.56	57.34
2.5% Modified corn stalk fiber	0.97	1.72	1.78	60.62
1.5% Lignin fiber	1.06	1.83	1.73	59.94
2.0% Lignin fiber	0.83	1.31	1.59	57.76
2.5% Lignin fiber	1.11	1.47	1.32	52.80
No fiber	0.65	1.35	2.08	64.33

.

The storage modulus (G*cosδ) gives information about the amount of structure present in a material. The loss modulus (G*sinδ) represents the viscous part, or the amount of energy dissipated in the sample. It also called the fatigue parameter, and the larger the value, the faster the energy loss [29]. The lost energy is directly related to the fatigue damage of the specimen during the loading process. The smaller the value, the slower the fatigue damage development, and the better the fatigue performance.

From Figure 7, it can be seen that with an increase in fiber content, the storage modulus and loss modulus increased at first and then decreased for all the fiber asphalt binders, except lignin fiber. An increase in storage modulus means an increase in resistance to deformation. When fiber content is about 1.5%, it can supply the highest deformation resistance ability. A decrease in the loss modulus means an increase in resistance to fatigue. When the fiber content is about 2%, it can supply the highest resistance to fatigue.

Therefore, from the perspective of the permanent deformation and fatigue resistance of the asphalt binder, when the fiber content is in the range of 1.5%–2%, all types of fibers can improve the fatigue performance of the asphalt binder.

3.2. Analysis of the LAS Test Result

3.2.1. Stress–Strain Curve Analysis of Fiber Asphalt Binder

(1)

Influence of fiber content

The LAS experimental results at 25 °C were analyzed, and the stress–strain curves of the five fiber asphalt binders with different fiber contents were obtained, as shown in Figure 8.

It can be seen in Figure 8 that the stress–strain curves of the five kinds of fiber asphalt binders are consistent with an increase in fiber content. Before reaching the peak stress, the stress increases linearly with an increase in strain. After reaching the peak stress, the stress decreases with an increase in strain and then yields. The yield stress of a fiber asphalt binder decreases with an increase in fiber content, indicating that the bearing capacity of an asphalt binder decreases with an increase in fiber content. The five fiber asphalt binders had the best bearing capacity when the fiber content was 1.5%.

The yield strain values of the five kinds of fiber asphalt binder at fiber contents of 1.5%, 2%, and 2.5% decreased with increases in fiber content, indicating that the dependence of asphalt binder on applied strain increased with increases in fiber content under repeated load. The width of the stress–strain peak zone of the five kinds of fiber asphalt binder with 1.5% fiber content is larger than that of the other two contents of fiber asphalt binder, indicating that the stress sensitivity of a fiber asphalt binder with a content of 1.5% is smaller in this content range, and its fatigue resistance is better. Therefore, the incorporation of too much fiber causes the light components in the asphalt to be adsorbed. An increase in the consistency of the asphalt binder increases its brittleness. The excessive incorporation of the fiber into the asphalt binder increases the stress sensitivity, and the asphalt binder is more likely to be destroyed in advance. The above conclusions show that the five kinds of fiber asphalt binder have a wide peak area and low stress sensitivity when the fiber content is 1.5%.

(2)

Influence of fiber type

The LAS experimental results of the five kinds of fiber asphalt binder with a fiber content of 1.5% at 25 °C were compared with those of a non-fiber asphalt binder. A diagram depicting the stress–strain relationships the six kinds of asphalt binder is shown in Figure 9.

From Figure 9, it can be seen that the stress–strain curves of the six asphalt binders have the same change trend, but their yield stresses and yield strains are different. The yield stresses of the five kinds of fiber asphalt binder are greater than that of the non-fiber asphalt binder. The yield stresses of the modified corn straw fiber and the modified cotton straw fiber asphalt binders are greater than that of the corn straw fiber and the cotton straw fiber asphalt binders, indicating that the incorporation of straw fiber can enhance the load bearing capacity of asphalt binder, and the incorporation of modified straw fiber can further improve its bearing capacity. The yield stress of the two kinds of modified straw fiber asphalt binder is higher than that of the lignin fiber asphalt binder, which shows that the effect of modified straw fiber on the bearing capacity of asphalt binder is better than that of lignin fiber. Because the incorporation of fibers reduces the stress sensitivity of asphalt binder, it can still maintain a certain stress state when it reaches the peak stress.

3.2.2. Fatigue Performance Analysis of Fiber Asphalt Binder

(1)

Influence of fiber content

To study the effect of the fiber content on the fatigue performance of the asphalt binder, the LAS experimental data at 25 °C were fitted to obtain the fatigue parameters A₃₅ and B for the five fiber asphalt binders at fiber contents of 0%, 1.5%, 2%, and 2.5%, as shown in Figure 10.

The fatigue parameter A₃₅ indicates the ability of the asphalt binder to resist fatigue damage, which represents the strength of the internal structural integrity under the asphalt binder damage condition, while the fatigue parameter B indicates the strain sensitivity of its fatigue performance. The smaller the absolute value of parameter B, the smaller the decay rate of the fatigue life of an asphalt binder. Figure 10 reveals that with an increase in the fiber content, the fatigue parameter A₃₅ exhibited a decreasing trend for the five types of fiber asphalt binders, while the fatigue parameter B exhibited an increasing trend. When the fiber content is 1.5%, the fatigue properties of asphalt binders with different kinds of fibers are better. This is because the fibers exist as insoluble matter in the asphalt binder and act as stabilizers, reinforcements, and crack arrestors for the asphalt binder. A large number of micro-pores on the fiber surface can absorb the light oil in the asphalt, which is equivalent to increasing the content of asphaltene, increasing the proportion of structural asphalt, and improving the fatigue performance of the asphalt binder. However, with an increase in fiber content, it is easy to cause agglomeration in the binder, which reduces the storage modulus of the binder, and the structural integrity of asphalt binder thus worsens under repeated load. The fiber content has a great influence on the strain sensitivity of asphalt binder, and the fatigue life decay rate of an asphalt binder increases with an increase in fiber content.

(2)

Influence of fiber type

To study the influence of the fiber type on the fatigue performance of the asphalt binder, the LAS test data at 25 °C were processed to obtain the fatigue parameters A₃₅ and B for the five types of fiber asphalt binder at a content of 1.5%. The parameters are listed in

Table 5. Fatigue parameters of the fiber asphalt binders at a content of 1.5%.

	Types	No Fiber	Lignin Fiber	Modified Cotton Stalk Fiber	Cotton Stalk Fiber	Modified Corn Stalk Fiber	Corn Stalk Fiber
Parameter
A35		148,598	442,905	416,728	296,652	452,125	286,902
B		4.72	4.42	4.38	4.65	4.36	4.58

.

As shown in Table 5, fiber type has a significant effect on the fatigue parameters of asphalt binder at a fiber content of 1.5%. Compared to unmodified stalk fibers, the fatigue parameter A₃₅ of modified cotton and corn stalk fibers increased by 40.5% and 57.6%, respectively, and the fatigue parameter B decreased by 5.8% and 4.8%, respectively. This indicates that the incorporation of modified stalk fibers improved the fatigue resistance of the asphalt binder. Moreover, the modified stalk fiber asphalt binder presents better durability, and the incorporation of the modified stalk fiber results in the same effect as that of lignin fiber, which is applied in complex loading environments.

(3)

Fatigue life analysis of the fiber asphalt binder

To study the influence of the stalk fibers on the fatigue life of the asphalt binder, the fatigue life N_f of the asphalt binder was calculated and the results are presented in Figure 11.

As shown in Figure 11, the fatigue life of each fiber asphalt binder decreases with an increase in the fiber doping content under the same strain level. The fatigue life of the five types of fiber asphalt binder also decreases linearly with increasing strain, and decreases more significantly in the 2.5%–5% strain level with increasing fiber dosage. Additionally, the fatigue life of the asphalt binder with modified corn stover fibers, modified cotton stover fibers, and lignin fiber is decreased at a dosage of 1.5%.

At the same strain level, fiber incorporation improves the fatigue life of asphalt binder. At the 2.5% strain level, the fatigue life of the modified corn stover fiber and modified cotton stalk fiber asphalt binders is much longer than that of the fiber-free asphalt binder, which proves that modified stalk fibers have a positive effect on the fatigue resistance of asphalt binder. This is because an appropriate number of fibers were evenly distributed in the asphalt matrix, and the three-dimensional space network that formed hindered the expansion of the asphalt binder microcracks, thereby prolonging the fatigue life of the asphalt binder [30]. After NaOH modification, the surface roughness of the stalk fibers increases. The specific surface areas of the modified corn stalk fibers are also larger than those of the cotton stalk fibers. Under high-speed shearing, better adsorption and winding effects can be achieved in asphalt macromolecules, which makes the fiber–asphalt interlocking structure firmer. Under external loads, the curled monofilament fibers and surface pores provide greater friction resistance; thus, the binder is not easily deformed. With an increase in the strain level, the effect of the fiber on the fatigue life of the binder decreases, indicating that the effect of the strain load on the fatigue life of the binder is higher than that of the fiber. The fatigue life ranking is as follows: modified corn stover fiber gum paste ≥ modified cotton stalk fiber gum paste ≥ lignin fiber gum paste > corn stover fiber gum paste > cotton stalk fiber gum paste.

In conclusion, the modified corn stalk fiber with a fiber content of 1.5% exhibits the best fatigue life and is recommended for the improvement of the fatigue properties of asphalt binders.

4. Discussion

In this study, NaOH solution, which can improve the surface roughness, specific surface area, and oil absorption ability of the fiber and enhance the adhesion between the fiber and asphalt, was used to modify stalk fibers through alkali erosion. Existing studies have shown that physical property indicators such as the fiber aspect ratio, surface roughness, and specific surface area affect the rheological properties of asphalt binders [31,32]. Chen et al. modified a corn stover using electron microscopy and found that the surface of the modified fibers was rough, the compatibility of the fibers with asphalt was improved, and the overall mechanical properties of the fiber asphalt binder were improved according to tests that assessed low-temperature creep and viscosity [8]. This is similar to the method of alkali erosion employed in this study, in which a NaOH solution was used to improve the surface morphological characteristics of corn stalk fiber, thereby enhancing the mechanical biting force between the fiber and asphalt to improve the interface characteristics of both.

Jia et al. reported that the length-to-diameter ratio of a fiber significantly influences the effect of fiber-reinforced asphalt, and fibers with a large length-to-diameter ratio can effectively overlap to form a network system [33]. In this study, a large length-to-diameter ratio was observed in the corn stalk fibers, and the fibers overlapped with each other to form a tight three-dimensional space network in the asphalt, which enhanced its structural stability and inhibited the generation and development of microcracks.

In this study, lignin, cotton stalk, and corn stalk fibers were modified using a 5wt% NaOH solution. Using a detailed test setup and result analysis, the effects of the fiber type, fiber content, and modification on the fatigue performance of the asphalt binder were determined. Moreover, the influence trends and mechanisms of the three variables on the fatigue performance of the asphalt binder were analyzed using DSR and LAS tests, and the optimal conditions under these three variables were determined. The results indicate that the fatigue resistance of the modified stalk fiber asphalt binder is better than that of the unmodified stalk fiber asphalt binder, and the number of fibers has a significant impact on the strain sensitivity of the asphalt binder.

Chen et al. studied the rheological properties of three and five types of fiber asphalt binders and found that different fiber types and dosages have different degrees of influence on the performance of the binder [34]. Each fiber was given the best dosage, and the performance of the fiber asphalt binder was the highest under the optimal dosage. Existing studies have reported that the fatigue properties of asphalt binders are primarily related to the material and the interactions between the materials [35]. The fibers and bitumen liquid phase are mainly bonded via mechanical connection and wetting adsorption, and the adsorption capacity and bonding capacity of the fibers and bitumen are enhanced after alkali treatment. However, fibers and bitumen are incompatible, which may result in the formation of voids or air bubbles between the fibers and bitumen. With the addition of an excessive fiber content, the voids gradually increase, and fiber agglomeration occurs. This is one of the reasons for the reduced fatigue performance of the asphalt binders.

Chen et al. [35]. reported that an increase in corn stalk fiber content first causes a decrease in the stiffness of the fiber asphalt binder followed by an increase, indicating that there is an optimal binder fiber content. At the optimal fiber content, the corn stalk fibers were evenly dispersed in the binder [8], similar to the results in this study. Furthermore, the experimental results revealed that the fatigue properties of the binder vary with the fiber content. When the fiber content was in the range of 1.5%–2%, the five fibers significantly improved the fatigue performance of the asphalt binder, and the modified corn stalk fiber with a 1.5% fiber content had the best fatigue life.

In this study, only the modification of the 70# matrix asphalt by adding fibers was investigated, and a comparative study with the other types and grades of asphalt was not conducted. In addition, only the fatigue properties of the fiber–asphalt binders were studied. Thus, the rheological properties of fiber–asphalt binders should be studied in the future. These results on fiber asphalt binders can serve as a reference for future research on the road performance of fiber–asphalt mixtures.

5. Conclusions

In this study, a method of preparing road stalk fiber by chemical modification was presented, and a series of experiments on fiber asphalt binder were carried out. SEM was employed to investigate the micromorphology of stalk fibers modified using NaOH solution, and DSR tests and LAS tests were conducted to investigate the influence of the fiber type and content on the fatigue performance of fiber asphalt binder. The main conclusions are summarized as follows:

(1)

Modifications using NaOH solution changed the fiber microstructure, changed the original fiber bundle into monofilament fibers, increased the specific surface area and surface roughness of the stalk fiber, and increased the fiber bonding with the asphalt.

(2)

Based on the variation in storage modulus and loss modulus, when the modified stalk fiber content is in the range of 1.5%–2%, all types of fibers can improve the fatigue performance of asphalt binder.

(3)

For the same fiber type, the fatigue resistance performance is the best when the fiber content is 1.5%, and at this content level, the fatigue life decay rate is the smallest based on fatigue life N_f. Therefore, the best modified stalk fiber content for asphalt binder is recommended to be 1.5%.

(4)

The fatigue life N_f of modified corn stalk fiber asphalt binder is the longest under different strain levels, so the modified corn stalk fiber is recommended as the best fiber to improve the fatigue performance of asphalt binder.