Static and fatigue behavior of basalt fiber-reinforced thermoplastic epoxy composites

Abstract

The static, fatigue properties and their damage mechanism of basalt fiber-reinforced thermoplastic epoxy polymer composites are investigated. The stress–life curves and stiffness degradation under long-term cyclic loading were tested for basalt fiber-reinforced thermoplastic epoxy polymer. An advanced fatigue loading equipment combined with in situ scanning electron microscopy was used in the tests to track the damage propagations and analysis the fracture surfaces of all specimens. Results were also compared with those of thermosetting epoxy-based basalt fiber-reinforced polymer composites. The results show that the basalt fiber-reinforced thermoplastic epoxy polymer has good interface properties between the fiber and new thermoplastic epoxy, which results in high tensile strength and ductility. Different degradation rates of low- and high-cycle fatigue loads are observed for the basalt fiber-reinforced thermoplastic epoxy polymer composites. Under high fatigue stress levels, a high degradation rate of the fatigue life is found because the dominating damage pattern showed fiber fractures. At low and medium fatigue stress levels, the damage pattern is dominated by matrix cracking and interface debonding, which results in a low degradation rate of the fatigue life. A bilinear phenomenological fatigue model has a higher accuracy for fitting the stress–life data than linear fatigue models. In addition, 80–90% stiffness degradations are observed before failure for all stress levels at a stress ratio of 0.8. Furthermore, compared with thermosetting epoxy-based basalt fiber-reinforced thermoplastic epoxy polymer, the basalt fiber-reinforced thermoplastic epoxy polymer has similar static strength and similar fatigue life at high-stress levels. However, at low-stress levels, the fatigue life of the this polymer is much higher than that of thermosetting one.

Keywords

Basalt fiber-reinforced thermoplastic epoxy polymer composites thermoplastic epoxy resin static behavior fatigue performance fatigue damage

Introduction

As a competitive alternative to traditional materials used in various types of infrastructure, fiber-reinforced polymer (FRP) composites offer many promising characteristics for civil applications, including superior mechanical and chemical behaviors, low weight, and reduced lifecycle costs.^1–3 Continuous basalt fiber-reinforced polymer (BFRP) composites are a relatively new FRP material with superior mechanical properties^2,3 and corrosion resistance^4–6 compared with traditional glass FRP (GFRP), and are more affordable than traditional carbon FRP (CFRP). Furthermore, BFRP composites possess superior creep behavior with a creep rupture stress of 54% of their tensile strength,⁷ which allows them to be used more sufficiently in prestressing and cable applications in comparison to GFRP. Intensive studies on the technological, mechanical, and rational applications of thermosetting resin-based BFRP materials have been conducted.^8–13

Interest in thermoplastic composites is growing from such advantages as a good secondary processing performance through reheating, recyclability, higher toughness, and superior impact properties over thermosets.¹⁴ This solves the problem in which FRP tendons cannot be flexed in the field (such as in a stirrup or end bend hook). Polypropylene (PP) and polyethylene (PE) have been mainly investigated as a matrix for thermoplastic FRP composites. Such thermoplastic FRP composites have a higher notched impact strength, reduced creep tendency, and better stability at elevated temperatures in humid conditions than thermosets.¹⁵ Goel et al.¹⁶ studied the fatigue behavior of PP/20 vol.% E-glass FRP and found that fiber-pullout/fiber fracture/matrix fracture were the energy-absorption mechanisms during tensile and fatigue testing. The static and fatigue strengths of thermoplastic PP/PE FRP composites^16,17 are much lower than those of their thermosetting counterparts owing to their complexity of manufacturing and poor interface properties. Furthermore, no investigations have been carried out on new basalt fiber-reinforced thermoplastic epoxy polymer (BFRTP) composites.

In this study, a new kind of thermoplastic resin, a two-component thermoplastic epoxy resin, was used to investigate the fatigue life and microdamage mechanism of BFRTP composites. The thermoplastic epoxy resin possesses the advantages of both thermoplastic polymers, such as a good secondary processing performance, and epoxy resins, such as good interface properties with fibers. To investigate their suitability of application as prestressing tendons in fatigue-sensitive structures, basalt fiber-reinforced thermoplastic epoxy composites were tested for both their static and fatigue properties using advanced fatigue test equipment and an in situ scanning electron microscope (SEM). All results were also compared with those of the thermosetting epoxy BFRP composites described in Zhao et al.¹⁸

Experiment procedure

Materials and specimens

The thermoplastic resin

The thermoplastic resin used in this study is a two-component liquid epoxy resin (Nagase ChemteX Co., Wuxi, China). The mechanism of the curing reaction of the thermoplastic resin differs from the traditional thermosetting epoxy resin, which is shown below.

where R and R′ could be the following structures for example.

The traditional thermosetting epoxy resin reacts with the curing agent during the heating process to produce a reticulate molecular structure, whereas the curing reaction of the thermoplastic epoxy resin during the heating process makes the molecular chain of the resin extend in a linear manner. The system is suitable for manufacturing protruded BFRP bars and stirrups. It solves the problem in which FRP bars cannot be flexed in the field (such as stirrups and end crooked hooks) without changing its performance based on the thermal processing, which is more suitable for civil construction requirements.

The properties of the thermoplastic epoxy resin used in this experiment are shown in Table 1. It can be seen that the viscosity of the thermoplastic epoxy resin is very high at room temperature, which directly affects the stirring reaction with the curing agent, and then affects the impregnation of the fiber. Therefore, the resin needs to be heated for a period of time from 100℃ to 200℃ to be transparent before it can be used to mix with the curing agent, otherwise, it will affect the uniform mixing of the resin and the curing agent. According to the exothermic peak of thermoplastic epoxy resin, the curing temperature was determined to be 180℃, and the curing time was determined to be 90 min. It is worth noting that the Tg of the thermoplastic resin is very close to the peak temperature seen in civil construction due to solar warming (∼80℃). The usage of this thermoplastic epoxy-based BFRP in the structural members which are under the blazing sun for quite a long time should be cautious.

Table 1.

Thermomechanical properties of the epoxy resin materials.

Type	The ratio of resin to curing agent	Curing exothermic peak	Tg	Viscosity at room temperature
Thermoplastic resin	50 : 1	167℃	83℃	124,000 cp
Thermosetting resin	100 : 8	148℃	125℃	150 cp

To evaluate tensile properties of the thermoplastic epoxy resin, tensile tests for properties of resin castings were carried out according to the standard GB/T-2567-2008. The specimen is prepared by casting body mold as shown in Figure 1. The size of the specimen is shown in Figure 2. Considering the common release agents such as Vaseline and silicone grease, it is found that silicone grease as the release agent of thermoplastic epoxy resin has a complete release effect and will not affect the integrity of the specimen. Five valid samples in the same batch were guaranteed. The data in the resin tests were collected via DH 3716N provided by Donghua Testing Technology Co. Ltd.

The BFRTP

A system of continuous longitudinal basalt fibers in a reinforced thermoplastic epoxy composite was tested in this study, of which the manufacture and testing method is similar to that of thermosetting epoxy-based BFRP composites in Zhao et al.¹⁸ The basalt fibers were CBF13-1200 (1200 tex, Jiangsu GVM, Co.) with silane sizing. Small size dumbbell-shaped specimens (shown in Figure 3) were used during the tests for the BFRTP. The specimens were produced using a filament winding method. The pretension of the fiber filament was controlled using a weight of 150 g. An end tab consisting of two layers of basalt fiber sheets was anchored to the wound pre-impregnated continuous basalt fiber bundle. To smooth the change of stress between the bundle and the tabs, the out layers of the tabs were 10 mm shorter than the inside layers. The specimens were post-cured for 1.5 h at 180℃ and cooled slowly in the air before use. The tested continuous BFRP composite is a unidirectional composite whose fiber orientation is along the loading direction.

Test setup and procedure

The specimens were then tested for both the quasi-static and fatigue tests using the advanced fatigue loading equipment combined with an in situ SEM (Shimadzu SEM Servopulser), as shown in Figure 4(a). The system can conduct fatigue loading tests and SEM observations simultaneously. A pair of steel wedges was fixed to the fatigue servo system, and used to grip the coupon transferring the fatigue load to the specimens, as shown in Figure 4(b). Aluminum shims were placed between the wedges and specimens to smooth the change in stiffness between them and to avoid grip failure during the fatigue testing.

In the quasi-static tests, a stroke rate of 2 mm/min was adopted. The fatigue tests were conducted under a constant stress amplitude cyclic load with a stress ratio of 0.8 at room temperature, as shown in Table 1. A stress level indicates the ratio of maximum applied fatigue stress to the ultimate tensile stress. Four stress levels were applied to the specimens, and to compare the stress–life (S–N) curves, the results covered a lifetime of between 10³ and 10⁷ cycles. It was found that 10⁷ cycles will be more suitable than 10⁶ cycles to an analysis of the long-term fatigue life as the stiffness and fatigue damage of FRPs become stable between 5 × 10⁶ and 1 × 10⁷ cycles.^18,19 A load frequency of 10 Hz was selected. The stiffness of the specimens, which was defined as (maximum load − minimum load)/(maximum deformation − minimum deformation) in one cycle, was recorded every few cycles.

Results and discussions

Static tensile properties

The thermoplastic resin

The static tensile properties of the two resin castings are shown in Table 2. The thermoplastic epoxy resin has the best tensile properties, and its strength is about 20% higher than that of the thermosetting epoxy resin, and the elastic modulus has little difference. The fracture elongation of thermoplastic epoxy resin is also larger than that of thermosetting epoxy resin. This can be a result of larger plasticity of thermoplastic epoxy. With higher strength and fracture elongation of thermoplastic epoxy resin, its BFRTP can be expected to have better performance than thermosetting epoxy resin based BFRPs.

The basalt fiber-reinforced thermoplastic epoxy polymer

Table 2.

Tensile properties of the epoxy resin materials.

Resin type	Tensile strength (MPa)	Average tensile strength (MPa)	Average modulus of elasticity (GPa)	Fracture elongation (%)
Thermoplastic epoxy resin	71.4	65.4	3.0	3.0
	59.6
	74.2
	60.8
	60.9
Thermosetting epoxy resin	48.7	54.7	3.0	2.1
	46.9
	55.6
	60.7
	61.5

Five coupons of the BFRTP were conducted in the quasi-static test, whose results are presented in Table 3. The mean tensile strength was 2765 MPa, which is larger than that of the thermosetting epoxy BFRP (2615 MPa).¹⁸ The coefﬁcient of variation (CV) was relatively low (≈5%), which demonstrates that the mechanical properties of the specimens are relatively stable. The CV is higher than that of the thermosetting epoxy BFRP (2.77%)¹⁸ because it is harder to control the quality during the manufacturing of the BFRTP specimens than of the thermosetting epoxy BFRP.

Table 3.

Summary of the static tests of basalt fiber-reinforced thermoplastic epoxy polymer composites (BFRTP).

Specimen	Loading (kN)	Tensile strength (MPa)
1	1.17	2543
2	1.28	2783
3	1.27	2761
4	1.38	3000
5	1.26	2739
Means	1.27	2765
Standard deviation	0.07	162
Coefﬁcient of variation (%)	5.87	5.87

The load–displacement curves of specimens are shown in Figure 5. All curves showed an almost linear behavior prior to reaching about 90% of the peak load. When the specimens were loaded at the stress before failure, the specimens exhibited nonlinear stress–displacement or were still able to carry a certain load after a drop in the peak load. This behavior is a result of nonuniform fiber breaking.

The macroscopic and microscopic failure modes of the specimens were observed during the static tensile test, as shown in Figure 6. The macroscopic failure mode of the thermoplastic epoxy resin-based specimens was the tearing between the fiber bundles, which differs from a blownout failure mode that is often seen in thermosetting resin-based BFRP.^{12,13,18,20,21} From an SEM image taken after a static failure, shown in Figure 6(b), interface debonding between the broken fibers and the unbroken fibers was observed. However, it can also be seen that the parts far from the debonding parts were relatively intact and that the resin can still hold the fibers, which may be the result of good properties of the resin and good adhesion between the basalt fiber and thermoplastic epoxy resin.

Stress–life curves

The S–N data of the specimens are presented in Table 4 and Figure 7. A linear model was applied to fit the data using the least-squares method according to equation (1).¹⁸

r_{max} = \frac{σ_{max}}{σ_{ult}} = A log (N) + B

(1)

where r_max refers to stress level (the ratio of the maximum stresses, σ_max, to the ultimate strength, σ _ult ), N is the corresponding number of cycles to failure. Parameters A and B are the model parameters determined after application of the process to the available fatigue data, where parameter A characterizes the slope of the S–N fatigue curve which reveals the degradation rate of fatigue life. The parameter B indicates that the extrapolation of the S–N curve to one cycle, which should be around 1.0 in this case.

Table 4.

Stress–life (S–N) data of basalt fiber-reinforced thermoplastic epoxy polymer (BFRP) specimens.

Specimens	Stress level (%)	Failure cycles	Avg. num. of cycles	CV (%)
BFRP-90-1	90	50,436	57,791	39.7
BFRP-90-2		39,426
BFRP-90-3		83,511
BFRP-87-1	87	219,675	159,918	34.8
BFRP-87-2		109,644
BFRP-87-3		150,436
BFRP-85-1	85	14,665,10	26,113,51	42.0
BFRP-85-2		27,134,21
BFRP-85-3		36,541,24
BFRP-83-1	83	55,017,76	43,068,73	33.0
BFRP-83-2		46,841,62
BFRP-83-3		27,345,83

CV: coefficient of variation.

R²= 0.9146 is obtained for the fitting of a linear model. However, it can clearly be seen that the trend of data above a stress level of 0.87 (less than 2 million cycles) is different from those below. Therefore, a bilinear phenomenological fatigue model^18,22 is adopted to describe different degradation rates, as shown in Figure 7. The accuracy of the fitting increased into R²= 0.9546, which indicated that the bilinear phenomenological fatigue model has a higher accuracy for fitting the S–N data than linear fatigue models. The bilinear phenomenological fatigue behavior also indicates different failure mechanics under different stress levels and life cycles. This phenomenon is similar to the different degradation rates of low- and high-cycle fatigue loads observed in unidirectional GFRP composites²² and thermosetting epoxy BFRP composites.¹⁸

Figure 8 compares the stress and fatigue life data of thermosetting¹⁸ and BFRTP composites. At high-stress levels, similar fatigue life of thermosetting epoxy BFRP was observed compared to that of BFRTP. At a low-stress level, the fatigue life of the BFRTP is a little higher than that of thermosetting epoxy BFRP. A conservative fatigue life prediction of BFRTP composites is 80.51%, achieving a fatigue life of 1 × 10⁷ cycles, which is slightly higher than 71.04% for BFRTP in Zhao et al.¹⁸

Stiffness degradation

As has been well documented,^23–26 a change in elastic modulus is a strong indicator of damage to material. The typical degradation curves of the tensile modulus using data collected at various stress levels are shown in Figure 9, where the tensile stiffness and fatigue cycles are presented at a normalized scale.

Typical critical limits of approximately 80–90% of the initial stiffness were found for all stress levels. However, the degradation patterns for different stress levels were varied. For stress levels of more than 87%, a decrease in stiffness can be found throughout the fatigue cycles until a sudden drop before the failure of the specimen. For stress levels of less than 85%, a three-stage stiffness degradation pattern, an initial rapid drop then a plateau and finally a drop before failure, can be observed. The patterns of stiffness degradation were also found in the test on thermosetting BFRP specimens,¹⁸ but the amount of degradation for thermoplastic and thermosetting BFRP specimens differs, as shown in Figure 10. The former has a higher critical stiffness limit and a lower stiffness degradation rate than the latter.

Fatigue damage and its mechanism

Figure 11 shows the typical failure surface at a stress level of 90%. Short matrix cracks and interface debonding from the fiber breaking points occurred in the specimens. A massive amount of matrix debris remained on the surface of the broken fibers in the failure pattern shown in Figure 11.

The specimens at a stress level of 83% showed a typical matrix cracking mode as shown in Figure 12. The matrix cracks were long at the beginning, then propagate toward a longer length with an increase in the number of cycles, as shown in Figure 12(a) to (c). It can also be seen in Figure 12(a) to (c) that the matrix cracks can propagate transversely across many fibers without interface debonding, and the interface debonding only occurred on the fracture surface shown in Figure 12(d).

The SEM images show that the dominant failure is always controlled by fiber breaking at a high-stress level, whereas matrix cracking and interface debonding are the main damage patterns at low and medium fatigue stress levels for BFRP, which is the reason for the knee point of the stress level-Log(N) (Figure 7). These damage patterns were also found during the test of the thermosetting BFRP specimens.¹⁸ The accumulation and development of matrix cracking, interface debonding and wear damage of BFRTP were slower than that of the thermosetting epoxy resin BFRP, and thus the BFRTP had a higher fatigue life at the same stress level.

The fatigue failure mechanism of two types of BFRPs was studied by microscopic observation, whose schematic illustration is shown in Figure 13. Under fatigue load, both types of BFRP have matrix crack phenomenon at the initial loading because the tensile stress of the resin matrix is far lower than that of the fiber. The stress concentration occurs at the crack point of the matrix and the maximum stress concentration occurs at the crack tip, which leads to the gradual propagation of cracks with the increase of the fatigue cycle. Due to the different chemical structures of the two resin matrices, the initial crack development process is also quite different. The molecular chain of thermosetting epoxy resin is a three-dimensional reticulated solid structure that has lower strength, plasticity and fracture elongation. The matrix crack generated under fatigue loading for thermosetting epoxy resin-based BFRP has a fast crack growth rate under fatigue loading. The molecular chain of thermoplastic epoxy resin is a linear structure, which has higher strength, plasticity, and fracture elongation. Under fatigue load, the stiffness decreases slowly, and the initial matrix crack has resilience plasticity or toughness with the increase of the cycle period. Although new matrix cracks continue to appear, the crack propagation speed is slow.

Because the strength and fracture elongation of thermosetting epoxy resin are lower than that of thermoplastic epoxy resin, its performance is relatively weak, which results in matrix crack and interface debonding mixed failure mode. The strength and elongation of thermoplastic epoxy resin are high and the propagation speed of matrix crack is slow, which prevents the interface from debonding from developing. As a result, the cooperative working of fibers can be well maintained and the fracture of fibers is not easy to occur. A better fatigue performance can be maintained for the BFRTP.

Conclusions

In this paper, the static and fatigue behaviors of BFRTP composites were investigated in both the microscale and macroscale. The fatigue failure phenomenon and fatigue–life curves were studied through BFRTP tests and the damage process was monitored using in-situ SEM. The major conclusions are as follows.

1. Compared with thermosetting epoxy BFRP, the static strength and fatigue life at high-stress levels of BFRTP are close to those of thermosetting epoxy BFRP. However, at low-stress levels, the fatigue life of the former is much higher than that of the latter.

2. Under high levels of fatigue stress, a higher degradation rate of fatigue life was revealed than that at low and medium levels. In addition, the bilinear phenomenological fatigue model achieves greater accuracy for fitting the S–N data than linear fatigue models. A conservative fatigue life prediction of BFRTP composites is 80.51%, achieving a fatigue life of 1 × 10⁷ cycles. The critical limit of fatigue stiffness degradation prior to failure was approximately 80–90% of the initial stiffness for all stress levels for a stress ratio of 0.8.

3. SEM images confirm that the BFRTP has two types of fatigue damage patterns: fiber fracture under high levels of fatigue stress, and matrix cracking and interface debonding at low and medium levels of fatigue stress.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: National Key Research and Development Program of China (2017YFC0703000) and the National Science Foundation of China (51678139).

ORCID iD

Xin Wang

References

Meier

. Composite materials in bridge repair. Appl Compos Mater 2000; 7: 75–94.

Wang

, et al. Enhancement of basalt FRP by hybridization for long-span cable-stayed bridge. Compos Part B Eng 2013; 44: 184–192.

Wang

Iwashita

, et al. Tensile fatigue behaviour of FRP and hybrid FRP sheets. Compos Part B Eng 2010; 41: 396–402.

Dong

, et al. Bond durability of BFRP bars embedded in concrete under seawater conditions and the long-term bond strength prediction. Mater Design 2016; 92: 552–562.

Wang

, et al. Durability of basalt fibers and composites in corrosive environments. J Compos Mater 2015; 49: 873–887.

Shi

Zhu

, et al. Durability of BFRP and hybrid FRP sheets under freeze-thaw cycling, Advanced Materials Research. Trans Tech Publ 2011; 163: 3297–3300.

Wang

Shi

Liu

, et al. Creep behavior of basalt fiber reinforced polymer tendons for prestressing application. Mater Design 2014; 59: 558–564.

Wang

. Evaluation of FRP and hybrid FRP cables for super long-span cable-stayed bridges. Compos Struct 2010; 92: 2582–2590.

Wu Z, Wang X and Wu G. Basalt FRP composite as reinforcements in infrastructure. In: Proc., 17th Annual International Conference on Composites/Nano Engineering (ICCE-17), International Conference on Composites/Nano Engineering (ICCE), New Orleans, 2009, pp. 21–24.

10.

Wang

Zhao

, et al. State-of-the-art review of FRP composites for major construction with high performance and longevity. Int J Sustain Mater Struct Syst 2014; 1: 201–231.

11.

Colombo

Vergani

Burman

. Static and fatigue characterisation of new basalt fibre reinforced composites. Compos Struct 2012; 94: 1165–1174.

12.

Wang

. Integrated high-performance thousand-metre scale cable-stayed bridge with hybrid FRP cables. Compos Part B Eng 2010; 41: 166–175.

13.

Wang

Iwashita

, et al. Tensile fatigue behaviour of FRP and hybrid FRP sheets. Compos Part B Eng 2010; 41: 396–402.

14.

Baets

Devaux

Verpoest

. Toughening of basalt fiber-reinforced composites with a cyclic butylene terephthalate matrix by a nonisothermal production method. Adv Polym Technol 2010; 29: 70–79.

15.

Hartness

Husman

Koenig

, et al. The characterization of low cost fiber reinforced thermoplastic composites produced by the DRIFT™ process. Compos Part A Appl Sci Manufac 2001; 32: 1155–1160.

16.

Goel

Chawla

Vaidya

, et al. Characterization of fatigue behavior of long fiber reinforced thermoplastic (LFT) composites. Mater Characterizat 2009; 60: 537–544.

17.

Tarpani

Canto

Saracura

RGM

, et al. Compression after impact and fatigue of reconsolidated fiber-reinforced thermoplastic matrix Sslid composite laminate. Procedia Mater Sci 2014; 3: 485–492.

18.

Zhao

Wang

, et al. Fatigue behavior and failure mechanism of basalt FRP composites under long-term cyclic loads. Int J Fatigue 2016; 88: 58–67.

19.

Schütz

Gerharz

. Fatigue strength of a fibre-reinforced material. Composites 1977; 8: 245–250.

20.

Wang

, et al. Enhancement of basalt FRP by hybridization for long-span cable-stayed bridge. Compos Part B Eng 2013; 44: 184–192.

21.

Zhao

Wang

, et al. Effect of stress ratios on tension–tension fatigue behavior and micro-damage evolution of basalt fiber-reinforced epoxy polymer composites. J Mater Sci 2018; 53(13): 9545–9556.

22.

Harik

Klinger

Bogetti

. Low-cycle fatigue of unidirectional composites: bi-linear S–N curves. Int J Fatig 2002; 24: 455–462.

23.

Varvanifarahani

Shirazi

. Prediction of stiffness degradation and damage of unidirectional GRP composites under fatigue cycles: science and engineering of composite materials. Sci Eng Compos Mater 2007; 14: 197–204.

24.

Kashtalyan

Soutis

. Stiffness degradation in cross-ply laminates damaged by transverse cracking and splitting. Compos Part A Appl Sci Manufact 2000; 31: 335–351.

25.

Zhang

Herrmann

. Stiffness degradation induced by multilayer intralaminar cracking in composite laminates. Compos Part A Appl Sci Manufact 1999; 30: 683–706.

26.

Whitworth

. A stiffness degradation model for composite laminates under fatigue loading. Compos Struct 1997; 40: 95–101.