Improvement of mechanical properties and mechanism analysis of FMLs by ultrasonic vibration assisted hot press forming

Abstract

The low forming quality of fiber metal laminates (FMLs) limits their application in the aerospace field. As a fundamental research, FMLs high-quality forming process provides key technical support for FMLs engineering applications. In this study, an advanced process of ultrasonic vibration assisted hot press forming (UAHPF) of FMLs is proposed. The specially forming mold of UAHPF was designed and manufactured. Mechanical properties of FMLs were tested. UAHPF increased the tensile strength, bending strength, and shear strength of the FMLs by 13.09%, 11.97%, and 49.51% under the optimal process parameters, respectively. In comparison to traditional forming, UAHPF markedly diminished internal defects in FMLs, augmented the metal-fiber contact area, and narrowed the width of the fiber-resin gap. This novel process is a convenient and effective way to enhance the mechanical properties of FMLs.

Graphical abstract

Improvement of mechanical properties and mechanism analysis of FMLs by ultrasonic vibration assisted hot press forming.

Nomenclature

FMLs

fiber metal laminates

UAHPF

ultrasonic vibration assisted hot press forming

CRITIC

criteria importance through intercrieria correlation

CFRP

carbon fiber reinforced plastic

OOA

out-of-autoclave

σ_m

tensile strength

σ_f

bending strength

τ_m

interlaminar shear strength

T_i

comprehensive performance score

1. Introduction

Lightweight materials such as aluminum alloy and carbon fiber reinforced plastic (CFRP) are extensively utilized in fields such as automobiles and aerospace [[1], [2], [3]]. Given the growing demand for lightweight and structurally integrated equipment, single CFRP and alloy face challenges in meeting high strength and toughness [[4], [5], [6]]. Fiber metal laminates (FMLs) are super hybrid materials obtained by alternating layers of fiber composites and metal plates, providing unique structural properties [7]. Thanks to its high specific strength, specific modulus, and corrosion resistance, FMLs have found applications in Airbus A380 and solid rocket construction [8]. Significant property differences among material components often result in poor forming quality for FMLs [9]. Interfacial delamination frequently occurs during service, posing a constraint on the widespread application of FMLs [10]. Enhancing the forming quality of FMLs will facilitate their utilization across various industrial sectors. The forming quality of FMLs is primarily assessed based on their strength properties, encompassing tensile strength, flexural strength, and interlaminar shear strength. Inadequate shear strength resulting from weak interfacial bonding stands as the primary cause of failure in FMLs components [11]. The fabrication of FMLs mainly includes metal surface pretreatment and FMLs curing. Suitable pretreatment methods and curing processes can effectively improve the mechanical properties of FMLs.

It has been shown that various surface treatments of metals can enhance the mechanical properties of FMLs to some extent. Ye et al. [12] performed nano-scale electrochemical sculpture of titanium plates before the forming of FMLs. The results showed that the interlaminar shear strengths of FMLs were significantly enhanced due to the formation of TiO₂ nanotube arrays on the surface of titanium sheets FMLs. Tsai et al. [13] modified surfaces of aluminum alloy using nitrogen plasma treatment, leading to a 25% increase in the bonding strength between aluminum alloy and GFRTP. Zhang et al. [14] carried out the treatment of magnesium alloy surface with a phosphating solution before manufacturing the fiber metal laminate based on magnesium (MgFML). The phosphating treatment significantly improved the interfacial fracture toughness of MgFML. Molitor et al. [15] discussed various methods of surface treatment for titanium. The results revealed that a short period of alkaline cleaning of aluminum alloy for the fabrication of FMLs could enhance their interfacial bonding properties. The interfacial bonding property and bending resistance of FMLs decreased with the increase of alkali washing time. Aghamohammadi et al. [16] applied four different surface treatment methods on the aluminum layers and investigated their effects on the properties of FMLs. Anodizing, as the most effective surface treatment method, caused significant improvements in flexural strength. Maryan et al. [17] studied the effects of different metal surface treatment processes on the tensile and fatigue properties of FMLs. The anodizing surface modification increased the tensile strength of FMLs by 18.1%. In high fatigue cycles, the anodizing surface modification has the best performance, as compared with other surface modifications. Santos et al. [18] comprehensively compared the mechanical properties of FMLs obtained from anodized and non-anodized aluminum plates. The results show that the compression shear strength of FMLs obtained from anodized aluminum plates increased by 16.7%, the shear strength resistance increased by 12.4%, and the translaminar shear strength increased by 9.2%.

As the most critical link in the fabrication of FMLs, the forming process directly determines the mechanical properties of FMLs. The autoclave forming process has problems such as difficulty expelling bubbles and insufficient resin fluidity, which lead to poor quality control of FMLs components [19,20]. The unique advantages of ultrasound in the fields of chemistry and physics have attracted widespread attention from researchers [21,22]. Arumugham et al. [23] found that ultrasonic assistance can significantly reduce the activation energy of chemical reactions. Shi et al. [24] significantly improved the adsorption capacity of graphitic carbon nitride/Multiwall carbon nanotube (g-C3N4/MWCNT/Bi2WO6) for Ceftriaxone sodium (CS) with ultrasonic assistance. Holtmannspötter et al. [25] revealed the mechanism of acoustic cavitation in reducing resin viscosity and bubble content, which promoted the application of ultrasonic assistance in the field of composite material forming. Sharma and Luzinov [26] found that ultrasonic vibration could improve the curing rate of epoxy resin and remarkably reduce the time of curing. Peng et al. [27] developed an ultrasonic vibration assisted resin transfer molding (UARTM) technique to fabricate CFRP laminates. With the assistance of ultrasonic vibration, the flexural strength, fracture energy, and fiber volume fraction of CFRP were significantly increased. Wang et al. [28] applied ultrasonic vibration to the bonding process of nickel and CFRP sheets after plasma modification of the nickel sheet surface. With the plasma treatment and ultrasonic application, the bonding strength increased by 65%. Yang and Yuan [29] applied ultrasonic vibration assisted processing to the CFRP adhesive bonded joints. The results indicated that ultrasonic vibration can clean the substrates, improve the wettability of substrates, and promote the permeation of adhesive into the substrate microstructure. Wang et al. [30] studied the strengthening mechanism of an ultrasonic vibration-strengthened adhesive bonding method by numerical analyses. The findings indicate that high-frequency vibration caused bubbles in the adhesive layer to stretch, deform, and ultimately, burst at the flow front. The hydraulic pressure difference caused by vibration permeates the adhesive into the microstructure of the adherend surfaces.

The stability of the adhesive bonding joints substrate material makes its mechanical properties unaffected by ultrasonic vibration [31]. Both metal surface anodization and ultrasonic assistance can effectively improve the mechanical properties of the adhesive bonding joints. Before the resin is completely cured, ultrasonic vibration will seriously change the morphology of the prepreg and affect the mechanical properties of FMLs. It is necessary to combine metal anodization and ultrasonic vibration assisted technology to fabricate FMLs to explore the feasibility of applying ultrasonic vibration assisted technology in the forming process of FMLs.

An advanced high-quality forming process for FMLs is proposed based on metal surface anodizing and ultrasonic vibration assisted in this paper, named ultrasonic vibration assisted hot press forming (UAHPF). FML samples were fabricated by different forming processes and then tested for mechanical properties. Process parameter optimization was performed using orthogonal experiments, with the comprehensive performance score of FMLs as the guiding criterion. Surface morphology of prepregs, contact angle, laminate cross-section, and interfaces between materials were characterized to elucidate the strengthening mechanism of UAHPF on mechanical properties of FMLs.

2. Experiment

The fabrication of FMLs by UAHPF requires the selection of appropriate materials, pretreatment methods, and process parameters. Characterizing the mechanical properties of FMLs is a prerequisite for comparing the advantages and disadvantages of different forming processes. Revealing the influence mechanism of UAHPF on the mechanical properties of FMLs requires the use of multiple material characterization methods.

2.1. Materials

The FMLs sample with 2/1 stacking configuration consists of a 1 mm thick CFRP core and 0.8 mm thick 5052 aluminum alloy plates on both sides. CFRP core was molded from 3 K plain woven carbon-fiber/ thermoset epoxy prepreg (produced by Wuxi Shengte Carbon Products Co., Ltd.) with 0.25 mm thickness and 42% resin content. The carbon fiber exhibits a tensile strength, tensile modulus, elongation, density, and diameter of 3530 MPa, 235 GPa, 1.5%, 1.76 g/cm³, and 6 µm. The 5052aluminum alloy (supplied by Jiangsu Taizhou Xingmu Metals Factory), possesses a tensile strength of 215 MPa and a yield strength of 158 MPa.

2.2. Fabrication of FMLs samples

Fig. 1 shows the process of fabricating FMLs using the UAHPF. To enhance the interlayer interfacial bonding strength of the FMLs, the aluminum plates underwent phosphoric acid anodic oxidation. Reagents used for metal surface anodization are provided by China National Pharmaceutical Group Corporation. The aluminum plates were wiped with acetone and then immersed in an alkaline wash solution at 60 ± 5 °C for 1 min, comprising 30 g/L NaOH (96%) and Na₂CO₃ (99.8%). Subsequently, the aluminum plates were immersed in a 400 g/L HNO₃ (68%) solution for 2 min. The aluminum plates underwent ultrasonic cleaning with 60 °C deionized water for 3 min before and after each immersion. The aluminum plate was anodized by immersing them along with stainless steel in an H₃PO₄ (85%) electrolyte solution with a concentration of 120 g/L. The direct current supply voltage, anodizing temperature, and anodizing duration were set at 10 V, 25 ± 5 °C, and 20 min. Following anodizing, the aluminum plates were cleaned and dried using a vacuum drying oven.

Fig. 1.

Flow chart for fabrication of FMLs samples.

During fabricating process of FMLs, the prepregs were neatly laid between the aluminum alloy plates. A release cloth was placed on the outer surface of the metal plate. A caul plate, evenly coated with a release agent, was inserted between the release cloth and the mold. The metal plates and prepregs placed into the mold were restrained by the localization frame. The necessary pressure, temperature, and ultrasonic field conditions for FML forming are attained by adjusting the pressurizing bolts, heating plate, and ultrasonic generator settings. The desired FMLs test samples were obtained after a series of heating, pressing, curing, cooling, and machining

The quality of FMLs is significantly influenced by curing temperature, forming pressure, and ultrasonic parameters. Previous research has demonstrated that ultrasonic vibration assistance at a frequency of 20 kHz notably enhances resin wetting on fibers during composite forming processes [32]. This study utilized an ultrasonic vibration head operating at a frequency of 20 kHz. Five amplitudes (15.8 µm, 20.5 µm, 25.2 µm, 31.1 µm, and 34.3 µm) were uniformly chosen within the adjustable range of amplitude percentages from 50% to 100%. Mold temperatures (25 °C, 50 °C, 80 °C, 100 °C, and 130 °C) during ultrasonication were uniformly chosen within the range from room temperature to the curing temperature recommended by the manufacturer. Following the typical forming process of FMLs, forming pressures were set at 0.05 MPa, 0.25 MPa, 0.50 MPa, 0.70 MPa, and 1.00 MPa [33]. Ultrasonication durations (2 s, 5 s, 8 s, 11 s, and 14 s) were uniformly chosen between 2 s and 14 s.

2.3. Characterization

The surface morphologies of the prepreg were observed using a camera (Canon XA60). To record the impact of ultrasonic vibration on prepregs surface morphologies during the FMLs forming process, a 4 mm thick quartz plate replaced one aluminum plate of FMLs. Resin viscosity-temperature curve was obtained by a rotational rheometer (NETZSCH Kinexus Pro+). A contact angle measuring instrument (JC2000CS, supported by Powerreach Co., Ltd.) is used to evaluate the wettability of resin on the aluminum alloy surface. Cross-sectional morphology of the FMLs samples was observed using the scanning electron microscope (SEM) (JSM-IT800(HL), manufactured by Nippon Electronics Co., Ltd.).

2.4. Mechanical properties

The tensile strength, flexural strength, and interlaminar shear strength of FMLs samples were tested on a universal testing machine (SANS CMT5205, manufactured by MTS SYSTEMS (CHINA) Co., Ltd.). The shape and size of the test sample are shown in Fig. 1. Each test used the five samples with the same process parameters. The average of five test results was taken to ensure the reliability of the experiment. Tensile strength was calculated using Eq. (1).

$${\sigma }_m=\frac{F_m}{b_1{h}_1}$$ (1)

where F_m is the maximum tensile force. b₁ is the sample width. h₁ is the sample thickness.

The long strip sample for bending test was machined according to GB/T 41049–2021 standard. The bending strength δ_f can be calculated by Eq. (2).

$${\delta }_f=\frac{3P_{\mathrm{M}}{L}_1}{2b_2{h_2}^2}\left[1+4{\left(\frac{{\delta }_{\mathrm{M}}}{L_1}\right)}^2\right]$$ (2)

where P_M is the maximum load. L₁ is the sample span. b₂ is the sample width. h₂ is the sample thickness.

The interlaminar shear test was conducted according to the GB/T 35100–2018 with the short strip sample. The interlaminar shear strength τ_m can be obtained by Eq. (3).

$${\tau }_{\mathrm{m}}=\frac{3F_{\mathrm{m}}}{4b_3{h}_3}$$ (3)

where F_m is the maximum load. b₃ is the sample width. h₃ is the sample thickness.

3. Mechanical property evaluation and industrial viability analysis

The practical industrial application of FMLs necessitates comprehensive consideration of tensile, flexural, and interlaminar shear properties. Whether UAHPF is a high-quality forming process needs to be verified through a comprehensive performance evaluation of FMLs. The industrial viability of UAHPF needs to be justified by cost-benefit analysis.

3.1. Optimization of process parameters

The process parameters of UAHPF were optimized using orthogonal experiments, which is an effective approach for studying multi-factor and multi-level experiments. These parameters include ultrasonic amplitude, ultrasonic duration, forming pressure, and mold temperature during ultrasonication. The factor levels for the orthogonal experiments are presented in Table 1. According to Table 1, an orthogonal experimental scheme was designed as shown in Table 2 using the Minitab software. Table 2 indicates that each parameter significantly affects the tensile strength, flexural strength, and interlaminar shear strength of the samples. The results were analyzed using the “Analyze Taguchi Design” function of the Minitab software. From the main effect plot Fig. 2, each process parameter has distinct effects on the tensile strength, flexural strength, and shear strength. Tensile and flexural strengths are most significantly affected by ultrasonic duration. Interlaminar shear strength is most significantly affected by mold temperature during ultrasonication. Tensile strength, flexural strength, and shear strength all increase and then decrease with the increase of each process parameter. Tensile strength, flexural strength, and shear strength have different optimal process parameters. The practical industrial application of FMLs necessitates comprehensive consideration of tensile, flexural, and interlaminar shear properties. A comprehensive performance evaluation method was developed based on the data in Table 2 using the Criteria Importance Through Intercriteria Correlation (CRITIC) method [34].

Table 1.

Factor levels of orthogonal experiments.

Levels	Factors
	Ultrasonic amplitude (µm)	Ultrasonic duration (s)	Mold temperature ( °C)	Forming pressure(MPa)
1	15.8	2	25	0.05
2	20.5	5	50	0.25
3	25.2	8	80	0.50
4	31.1	11	100	0.70
5	34.3	14	130	1.00

Table 2.

Orthogonal experimental scheme and corresponding results.

Case number	Ultrasonic amplitude (µm)	Ultrasonic duration (s)	Mold temperature (°C)	Forming pressure (MPa)	Tensile strength (MPa)	Flexural strength (MPa)	Shear strength (MPa)
1	15.8	2	25	0.05	306.61	412.50	34.93
2	15.8	5	80	0.70	338.45	453.67	42.88
3	15.8	8	130	0.25	364.21	494.83	54.63
4	15.8	11	50	1.00	340.17	462.98	56.92
5	15.8	14	100	0.50	282.74	406.89	38.20
6	20.5	2	130	0.70	330.68	456.41	46.63
7	20.5	5	50	0.25	374.55	484.96	52.98
8	20.5	8	100	1.00	355.28	486.38	54.03
9	20.5	11	25	0.50	294.73	408.74	45.69
10	20.5	14	80	0.05	298.02	414.38	44.29
11	25.2	2	100	0.25	362.52	479.86	50.85
12	25.2	5	25	1.00	363.73	463.02	45.89
13	25.2	8	80	0.50	322.28	450.12	46.22
14	25.2	11	130	0.05	318.85	457.41	52.71
15	25.2	14	50	0.70	302.30	442.96	47.56
16	31.1	2	80	1.00	353.56	470.51	49.72
17	31.1	5	130	0.50	354.94	467.63	46.83
18	31.1	8	50	0.05	346.35	474.25	55.58
19	31.1	11	100	0.70	317.13	440.57	57.45
20	31.1	14	25	0.25	320.56	433.56	49.90
21	34.3	2	50	0.50	335.83	449.92	43.95
22	34.3	5	100	0.05	360.59	448.05	44.63
23	34.3	8	25	0.70	344.65	444.31	46.16
24	34.3	11	80	0.25	365.25	468.25	58.32
25	34.3	14	130	1.00	341.19	459.28	48.37

Fig. 2.

Main effect plot of means.

The original data matrix X is constructed based on the experimental data:

$$X=\left(\begin{array}{ccc}{x}_{11}& \dots & x_{1\mathrm{n}}\\ ⋮& x_{ij}& ⋮\\ x_{\mathrm{m}1}& \cdots & x_{mn}\end{array}\right)$$ (4)

where x_ij denotes the value of the jth evaluation indicator for the ith sample. i is equivalent to the case number in Table 2. j = 1, 2, 3 represent tensile, flexural, and shear strengths.

The shear strength is one order of magnitude lower than the flexural and tensile strengths. Normalization of data was performed to eliminate the impact of the order of magnitude differences on evaluation results.

$$x_{ij}^{\prime }=\frac{x-\min \left(x_j\right)}{\max \left(x_j\right)-\min \left(x_j\right)}$$ (5)

The information-carrying capacity of each indicator was calculated. Firstly, the standard deviation was calculated to indicate the fluctuation of the differences in the values of the indicators.

$$\{\begin{array}{c}\overline{x_j}=\frac{1}{m}{\sum }_{i=1}^m{x}_{ij}\hfill \\ S_j=\sqrt{\frac{{\sum }_{i-1}^m{(x_{ij}-\overline{x_j})}^2}{m-1}}\hfill \end{array}$$ (6)

where S_j is the standard deviation of the data of the jth indicator.

Indicator conflict calculations were then performed:

$$r_{ij}=\frac{{\sum }_{j,k=1}^n\left(x_{ij}-\overline{x_j}\right)\left(x_{ik}-\overline{x_k}\right)}{\sqrt{{\sum }_{j=1}^n{\left(x_{ij}-\overline{x_j}\right)}^2{\sum }_{k=1}^n{\left(x_{ik}-\overline{x_k}\right)}^2}}$$ (7)

$$A_j={\sum }_{i=1}^n\left(1-r_{ij}\right)$$ (8)

where r_ij is the correlation coefficient of the ith indicator with the jth indicator. A_j is the conflict coefficient of the jth indicator.

The informative value C_j for the jth indicator was defined as:

$$C_j=S_j\times A_j$$ (9)

The weighting coefficient W_j of the jth indicator was calculated:

$$W_j=\frac{C_j}{{\sum }_{j=1}^n{C}_j}$$ (10)

Tensile strength weighting coefficient W₁ is 0.3433. The flexural strength weighting coefficient W₂ is 0.2456. The shear strength weighting coefficient W₃ is 0.4111. To mitigate the influence of the range Groups, the 5th value of the three strengths in descending order is taken as the scoring basis. Referring to Table 2, the values for tensile strength, flexural strength, and shear strength are 362.52 MPa, 474.25 MPa, and 54.63 MPa. The comprehensive performance score of the FMLs can be calculated using Eq. (11).

$$T_i=\left(\frac{W_1{\sigma }_{m,i}}{362.52}+\frac{W_2{\delta }_{f,i}}{474.25}+\frac{W_3{\tau }_{m,i}}{54.63}\right)\times 100$$ (11)

3.2. Comprehensive performance evaluation

The comprehensive performance score was then analyzed using the “Analyze Taguchi Design” function of the Minitab software. Mean response is presented in Table 3, while the main effect plot is displayed in Fig. 2. Table 3 reveals that the mold temperature during ultrasonication has the greatest influence on the comprehensive performance score, followed by the ultrasonic duration, forming pressure, and ultrasonic amplitude. Fig. 2 illustrates the variation in the comprehensive performance score with different factor levels. Optimal parameters include an ultrasonic amplitude of 31.1 µm, ultrasonic duration of 8 s, mold temperature during ultrasonication of 80 °C, and forming pressure of 0.70 MPa. To explore the difference between UAHPF and traditional forming process, the comparative experiment shown in Table 4 was designed.

Table 3.

Mean response for comprehensive performance score.

Level	Ultrasonic amplitude	Ultrasonic duration	Mold temperature	forming pressure
1	88.27	89.52	88.67	86.77
2	91.29	93.05	90.34	91.49
3	91.99	95.78	98.45	94.08
4	94.79	94.97	95.85	94.84
5	92.95	85.97	85.98	92.11
Delta	6.53	9.82	12.46	8.07
Rank	4	2	1	3

Table 4.

Fabrication of FMLs by different forming processes.

Group	Forming process	Ultrasonic amplitude (µm)	Ultrasonic duration (s)	Mold temperature ( °C)	Forming pressure (MPa)
A	UAHPF	31.1	8	80	0.7
B	Autoclave forming	0	0	0	0.7
C	Hot press forming	0	0	0	0.7
D	Long time UAHPF	31.1	20	80	0.7

Fig. 3(a) displays the comprehensive performance score of samples from Table 2, Table 4. Samples in Group A obtained the highest score, demonstrating the effectiveness of the comprehensive performance evaluation method. Fig. 3(b) shows the mechanical properties and comprehensive performance scores of FMLs fabricated by four different forming processes. It is evident that samples in Group A exhibit the best mechanical properties and comprehensive performance. Compared to Group B, Group A demonstrates improvements in tensile strength, flexural strength, interlaminar shear strength, and comprehensive performance score by 7.85%, 9.10%, 35.19%, and 18.66%. Compared to Group C, these values increased by 13.09%, 11.97%, 49.51%, and 26.23%. This suggests that UAHPF offers clear advantages over hot press forming and autoclave forming. The tensile and flexural strengths of samples in Group D are lower compared to those in the other groups. This suggests that ultrasonic duration is crucial in determining the forming quality of the samples. Excessive ultrasonic duration can significantly degrade the forming quality of samples.

Fig. 3.

Mechanical properties of FMLs sample: (a) comprehensive performance score of samples fabricated by UAHPF, (b) mechanical properties of sample with different forming processes.

3.3. Cost-benefit analysis

Autoclave forming is robust and well-understood, having benefited from significant research and experience gained from widespread industrial use, and remains a benchmark for competing processes [35]. However, autoclaves involve significant costs for acquisition, operation, and tooling, particularly for large parts. Autoclaves also impose a relatively inflexible manufacturing environment, in which potential part designs are constrained by available vessel sizes, production rates are restricted by scheduling, large autoclaves must sometimes be used inefficiently for small parts, and subcontractor options are limited [36]. The low capital investment, lower consumable costs, and improved energy efficiency of out-of-autoclave (OOA) processing have added to their appeal. However, these methods use very low pressures and typically rely on vacuum bag pressure to consolidate the part, and as a result, the mitigation of voids has become an issue [37]. Compared to autoclave curing of composite laminate, the OOA curing reduces energy consumption by over two orders of magnitude [38]. However, the tensile and shear strengths of the composite laminate were significantly reduced [39]. As a new OOA process, UAHPF not only reduces energy consumption but also significantly reduces the bubble content and improves the mechanical properties of the composites.

4. Strengthening mechanism analysis

In comparison to autoclave forming, UAHPF substantially enhances the mechanical properties of FMLs. The mechanism of UAHPF improving the mechanical properties of FMLs is still unknown. Understanding this mechanism can not only elucidate the impact of each process parameter on the mechanical properties of FMLs but also offer scientific groundwork and theoretical direction for the implementation of this process in industrial production.

4.1. Surface morphology of prepreg

The mechanical properties of FMLs are influenced not only by the load-bearing capacity of the prepreg and metal plate but also by the interfacial bonding strength between them. Due to the stability of the aluminum alloy, only the load-bearing capacity of the prepreg is affected during UAHPF. Fig. 4(a) depicts the changes of prepreg surface morphology with varying ultrasonic duration and the mold temperature under ultrasonication. Without ultrasonic vibration, bubbles are present at the center of the 80 °C prepreg surface. After 2 s of ultrasonication, the bubbles were transferred from the center to the edge. As the ultrasonic duration increases, bubbles at the edge are gradually expelled. At an ultrasonic duration of 8 s, no bubbles are visible in the prepreg.

Fig. 4.

Effects of ultrasonic duration and mold temperature on prepreg properties: (a) surface morphology of prepreg, (b) area of the prepreg, (c) resin viscosity.

As shown in Fig. 4(b), the area of the prepreg initially decreases and then increases as the bubble content decreases. The smallest area of 385.4 mm² appeared at 4 s of ultrasonic duration. When the ultrasonic duration is shorter, the bubble content is reduced and the bubbles are discharged from the center of the prepreg to the edges. At the same time, the resin flows towards the center of the prepreg and replaces the previous position of the bubbles. The prepreg area decreases, and the fiber content increases. After 4 s of ultrasonication, the prepreg boundary starts to expand and its weave structure starts to distort. After 22 s of ultrasonication, due to severe deformation prepreg reaches the maximum area of 684.44 mm².

It is well-known that the mechanical properties of carbon fiber remain nearly unchanged in an air environment up to 200 °C [40]. Comparing with 80 °C, the fibers in the prepregs tend to disperse more at 100 °C. This is attributed to the difference in the restraint strength of the resin to the fiber at different temperatures. From the viscosity curve in Fig. 4(c), the resin viscosity at 80 °C and 100 °C is 11.2 Pa·s and 2.5 Pa·s respectively. Resin in the prepreg has a weaker constraint on the fibers at 100 °C. This is a reason why a lower comprehensive performance score for FMLs fabricated by ultrasonic vibration at 100 °C than at 80 °C. The viscosity of the resin also has a significant impact on its flow and filling on the surface of other materials [41,25]. When the temperature is <80 °C, resin viscosity is too high to effectively flow and fill the surface of other components by ultrasonic vibration. At 80 °C, the resin can not only maintain the original shape of fibers but also sufficiently fill the surface of other components by ultrasonic vibration [42]. At 130 °C, The viscosity and curing degree of resin increase but the fluidity decreases [43]. Ultrasonic vibration scatters the prepreg and hinders the resin crosslinking reaction. FMLs exhibit markedly inferior mechanical properties.

4.2. Contact angle

The wettability of resin on the metal surface is a crucial factor influencing the interface bonding strength of FMLs. Fig. 5 illustrates the contact angle between resin and aluminum alloy surface at various ultrasonic durations. In the absence of ultrasonic vibration, the contact angle measured 115.7° at 80 °C. The contact angle decreased to 88.7°, 72.2°, 50.6°, and 32.3° with increasing ultrasonic duration to 2 s, 4 s, 6 s, and 8 s. This indicates that the wettability was significantly improved with the ultrasonic durations [31]. When the ultrasonic durations are higher than 8 s, the tendency of the contact angle to decrease with the ultrasonic durations slows down until it is nearly stagnant. Ultrasonic vibration on the wettability enhancement effect is close to maximization. Prolongation of ultrasonic durations no longer has a significant effect. A small number of big bubbles can be observed in the resin droplets without ultrasonic vibration. With the ultrasonic duration increased to 4 s, these bubbles transformed into a greater number of smaller bubbles. As the ultrasonic duration continued to increase, the bubbles in the resin droplets suddenly vanish [25]. This is because big bubbles were broken down into many small bubbles under hot pressure. As the ultrasonic duration increases, the ultrasonic cavitation effect causes the small bubbles to collapse and disappear rapidly [44]. Ultrasonic vibration enhances the wettability and comprehensive mechanical properties of the resin, thereby improving the interlayer interfacial bonding strength and overall mechanical properties of the FMLs.

Fig. 5.

Contact angle at different ultrasonic durations.

The effect of ultrasonic duration on contact angle, bubble content, and prepreg structure effectively elucidates why the comprehensive performance score of FMLs first increased significantly and then decreased steeply with increasing ultrasonic duration. When the ultrasonic duration is < 8 s, bubbles in the prepregs are fully expelled as the ultrasonic duration is extended. Simultaneously, the wettability of resin on the metal surface is significantly enhanced [45]. The occurrence of internal defects due to bubbles and weak adhesive interface was notably reduced, leading to a significant enhancement in forming quality. Once the ultrasonic duration exceeds 8 s, the mechanical properties of FMLs decline with further increases in ultrasonic duration. This occurs because as the ultrasonic duration is prolonged, the fiber distribution becomes increasingly scattered, and the enhancement in wettability is not substantial.

4.3. Cross section morphology of samples

The cross-sectional morphology and defect content of the samples from groups A, B, C, and D are shown in Fig. 6. The defect content is the ratio of the defect area to the image area, which is obtained by image processing software. From Fig. 6(c), there are two large and two small glue-deficient defects in the sample of Group C, with a total defect content of 4.05%. The large defects are situated at the interface between aluminum alloy and CFRP, whereas the small glue-deficient defects are present at the interface of various layers of prepreg. One large defect is located inside the CFRP and another is located at the CFRP-Al interface. Internal defects in CFRP are caused by residual bubbles. Anodization introduces too few active functional groups on the metal surface to effective chemical bonding [46]. Resin cannot fully infiltrate anodized metal surfaces under pressure alone. Therefore, the debonding of the CFRP-Al interface produces a large adhesive defect. The cross section of FMLs fabricated by UAHPF is shown in Figure Fig. 6(a). There are no visible defects in the sample from Group A. This suggests that bubbles present in the sample have been effectively expelled. The resin adequately permeates the surface microstructures of both carbon fiber and aluminum alloy, facilitating the formation of well-bonded interfaces between the materials. We have previously studied on interfacial interlocking effect of ultrasonic vibration assisted adhesive bonding based on molecular dynamics simulation [47]. The results showed that the ultrasonic action could improve the permeation and uniformity of adhesive at the interface. The molecular numbers of permeated adhesive increased by 10%, 8% and 7% with the frequency of 20 kHz, 25 kHz and 30 kHz after 10 cycles, and those increased by 6% and 10% with the amplitude of 1 Å and 2.5 Å after 5 cycles. Fig. 6(b) shows that the cross-sectional defect content of the group B sample is 0.21%, with only two minor defects. The size of these defects is comparable to that of the small defects observed in the sample from Group C, both caused by residual small bubbles in the prepreg. Clearly, the severity of defects in the sample from Group C exceeds that in Group B. This disparity arises because the laminate materials underwent vacuuming during the autoclave forming process (Group B), resulting in the expulsion of more bubbles and enhanced resin flow [20]. Group D samples exhibit severely deformed corrugated prepregs and numerous interface debonding. This indicates that excessively prolonged ultrasonic durations will notably diminish the forming quality. Compared with the samples in group C, the defect content of the samples in group D was reduced to 2.29%, which reflects the powerful bubble removal ability of ultrasonic action.

Fig. 6.

SEM images of FMLs samples fabricated by different forming processes: (a-d) FMLs cross sectional morphology of group A, group B, group C, and group D, (e-h) bonding interfaces of the samples in group A and group B, (i-l) Prepreg cross section of group A, group B, group C, and group D.

To delve deeper into the impact of UAHPF on the mechanical properties of FMLs, bonding interfaces of the samples from Groups A and B were comparatively analyzed. Fig. 6(e) and Fig. 6(g) shows the well-bonded metal-resin and fiber-resin interfaces observed in the Group A sample. From Fig. 6(f), it can be observed that a crack, approximately 18 µm in length and with a maximum width of about 1 µm, exists between the resin and the aluminum alloy plate in the Group B sample. Two defects are present at the fiber-resin interface. One defect, situated at the left end of the carbon fiber, measures approximately 1.2 µm in length and has a maximum width of about 0.1 µm. The other defect is located at the upper side of the fiber, with a length of about 2 µm and a maximum width of about 0.3 µm. The top and bottom boundaries of the defects coincide with each other at the metal-resin interface and interface between the resin and fiber end. This indicates that these two defects result from resin cracking during the forming process, reflecting weak interface bonding [19]. The boundary shape of the defect on the upper side of the fiber is irregular. A hollow is present inside the defect. This is caused by uneven and insufficient filling of the resin on the fiber surface. The gap at the fiber-resin interface, as depicted in Fig. 6(g) and Fig. 6(h), also reflects the filling effect of resin on the fiber surface. It can be seen from the figure that the gap of the sample in Group B is 0.065 µm, which is larger than the 0.022 µm gap in Group A. Well-bonded interfaces are the key to the excellent mechanical properties of FMLs fabricated by UAHPF.

The prepreg cross-section of samples from each group was further observed using SEM to precisely examine the fiber distribution. Fig. 6 (i-l) illustrates the SEM images of the prepreg cross-section and their binarized results. The fiber volume fractions (V_f) of prepregs in Groups A, B, C, and D are 68.25%, 65.75%, 59.51%, and 53.63%. The V_f of the samples in Groups A and B is significantly higher than the initial value of the prepreg of 58%. This is attributed to the effectiveness of both UAHPF and autoclave forming in reducing the bubble content of prepreg and enhancing the fiber volume fraction [48]. In comparison, UAHPF demonstrates a superior ability to remove bubbles, thereby effectively reducing the defects observed in Group A samples [43]. Samples in Group A exhibit the highest fiber density and the most superior mechanical properties. The V_f is only enhanced by 1.5% for the sample in Group C without ultrasonic vibration. When the ultrasonic duration is 20 s, the V_f of the sample in Group D drops to 53.63%. This shows that the ultrasonic duration is a key factor affecting the mechanical properties of FMLs fabricated by UAHPF.

According to the analysis above, the impact of the UAHPF on the quality of FMLs can be depicted in Fig. 7. The rise in temperature leads to a decrease in resin viscosity and an increase in fluidity. With the ultrasonic vibration-assisted, the resin tends to permeate into the micro-nano holes present on the surface of the anodized aluminum plate [49]. A limited number of bubbles within the prepreg are expelled as the resin flows under the action of the hot press, augmenting fiber density [27]. The cavitation effect induced by ultrasonic vibration contributes to the rapid reduction and eventual elimination of bubbles within the prepreg, increasing the fiber volume fraction. The application of ultrasonic waves generates dynamic impingement at the interface, thereby enhancing resin wettability on both fiber and aluminum alloy surfaces [47]. UAHPF significantly enhances the quality of FMLs by improving the bonding strength across each interface and enhancing the load-bearing capacity of prepregs.

Fig. 7.

Strengthening mechanism of the UAHPF.

5. Summary and outlook

This study focused on proposing the UAHPF process, a high-quality forming process for FMLs. UAHPF could significantly improve the mechanical properties of FMLs. The optimal process parameters for FMLs forming are ultrasonic amplitude 31.1 µm, ultrasonic duration 8 s, mold temperature 80 °C during ultrasonic application, and forming pressure 0.70 MPa. Compared with autoclave forming, the tensile strength, the tensile strength, flexural strength, interlaminar shear strength, and comprehensive performance score increased by 7.85%, 9.10%, 35.19%, and 18.66%. UAHPF improves the mechanical properties of FMLs by reducing bubble content, increasing fiber volume fraction, and resin wettability. UAHPF reduces the contact angle by 83.4°, reduces the defect content by 0.21%, and increases the fiber volume fraction by 2.5%. In a nutshell, UAHPF not only reduces production costs but also improves the mechanical properties of FMLs and facilitates their utilization across various industrial sectors.

However, UAHPF has some potential limitations of upscaling for real-world manufacturing scenarios. It is difficult to precisely coordinate process parameters such as ultrasonic duration, ultrasonic amplitude, ultrasonic frequency, and mold temperature. During the forming and service process, it is very difficult to non-destructively monitor the mechanical properties and internal defects of the workpiece. In particular, long-term performance monitors of durability and fatigue resistance severely limit upscaling to real-world manufacturing scenarios. We are conducting research on the precise coordination of process parameters and non-destructive monitoring of workpiece forming and long-term performance. We hope that relevant research will achieve good results and be published in high-level journals.

CRediT authorship contribution statement

Yizhe Chen: Conceptualization, Methodology, Investigation, Writing – review & editing. Longyu Dai: Formal analysis, Investigation, Data curation, Writing – original draft. Yusen Yang: Methodology, Validation. Hui Wang: Writing – review & editing, Visualization, Supervision. Shujian Li: Software, Visualization. Lin Hua: Resources, Project administration.

Declaration of competing interest

The authors declare that they have no conflicts of interest in this work.

Biography

Yizhe Chen is a professor at the School of Automotive Engineering, Wuhan University of Technology, Wuhan, China. He received his B.S., M.S., and Ph.D. degree in materials engineering from the Harbin Institute of Technology, Harbin, China. His areas of research interests include metal and hybrid material forming.