Effect of Atmospheric Pressure Plasma Treatment on the Lap-Shear Strength of Adhesive-Bonded Sheet Molding Compound Joints

Atmospheric pressure plasma treatment (APPT) technology was used herein to treat sheet molding compound (SMC) substrates to increase the lap-shear strength of adhesive-bonded SMC joints. Further, the mechanisms behind the lap-shear strength improvements in APPT-treated adhesive-bonded SMC joints were explored. A maximum lap-shear strength about three times that of the as-received SMC joints was achieved when the APPT distance was set to 20 mm. The surface roughness, which exhibited little benefit to the lap-shear strength, was determined to not be the primary reason for the increase in lap-shear strength. Specifically, X-ray photoelectron spectra revealed that an increased amount of O-containing groups (i.e., C–O–H, C–O–C, H–O–C=O or R–O–C=O) following APPT contributed to the improved lap-shear strength. In addition, the surface free energy increased significantly after APPT, which improved the lap-shear strength of the adhesive-bonded SMC joints. Compared to the change of surface morphology, the changes in both the surface chemical property and surface free energy played larger roles in increasing the lap-shear strength of APPT-treated SMC joints.


Introduction
Sheet molding compounds (SMCs) have gained extensive use in the area of new-energy vehicles to realize the light-weight requirements thereof [1,2]. However, a method whereby to joint the SMC substrates to ensure adequate automotive safety becomes important as the number of joining demands increases [3]. Adhesive bonding has the advantages of excellent mechanical properties, uniform stress distribution and low damage to the SMC substrates compared with the conventional joining techniques of welding and bolted connections [4]. Therefore, adhesive bonding has emerged as one B Junying Min Junying.Min@tongji.edu.cn B Jianping Lin jplin58@tongji.edu.cn 1 of the most effective joining methods for SMC. However, the SMC joints are not widely used as-received because of their poor lap-shear strength caused by the mold release agent used during demolding [5]. To achieve a greater and more durable lap-shear strength, the surface of the SMC substrates must therefore be cleaned and activated before joint fabrication [6]. Previous studies have reported adhesion improvements after wet chemical cleaning or mechanical abrasion [7]. Recently, the atmosphere pressure plasma treatment (APPT) technology has become more important in surface modifications of composites owing to its advantages of environmental protection and high efficiency [8,9]. Furthermore, as a functional technique, it affects only the substrate surface and changes the surface properties.
The APPT is widely applied as a polymer pre-treatment in practical situations [10,11]. It is worth noting that the APPT has drawn the attention of both industry and academy in recent years because of its ease of operation in industrial production and its selective treatment of substrates [12]. In a previous study, Noeske et al. [10] have found that the joints of polyethylene, polypropylene, polyvinylidenefluoride, polymers polyethyleneterephthalate and polyamide-6 following plasma treatment exhibited the fracture modes of substrate failure and cohesive failure after tensile test-ing, while the as-received joints exhibited adhesive failure between the polymer and adhesive. Gonzalez and Hicks [13] have investigated the surface modifications of polymers after remote APPT, while Zaldivar et al. [14] have used APPT to improve the strength of composites. Additionally, Yun et al. [15] have studied how the APPT affects polyimide/novolac epoxy resin joints using various gas types and aging temperatures. However, few studies exist on the mechanism of activating SMC substrates by APPT. The surface of substrates can be damaged or melted by high gas temperatures, induced by the heat sensitivity of the fiber-reinforced polymer. The optimal parameters for the application of this relatively new plasma treatment need to be studied in detail. Therefore, it is essential to explore the effects of APPT on the surface properties and adhesion performances of SMC substrates.
In this study, APPT was used to clean the surface and activate the SMC substrates. Quasi-static lap-shear testing was used to estimate the effect that APPT had on the lap-shear strength. Scanning electron microscopy (SEM) and roughness measurements were used to estimate the morphology modifications following APPT, while X-ray photoelectron spectroscopy (XPS) was used to characterize the chemical properties of the plasma-treated and as-received SMC substrates. Finally, the effect of APPT on the wettability of the SMC substrates was evaluated in the respect of contact angle and surface free energy (SFE).

Materials
The SMC used herein was fabricated from carbon fiber and vinyl ester resin and was strengthened by a fairly high proportion of short carbon fibers. The versatile epoxy backbone and the modifiers included in the resin system increased the toughness of the SMC. A mold release agent was used in the molding process. The SMC plates 3 mm thick were cut into substrates 100 × 25 mm 2 in area. A two-component roomtemperature-cured epoxy adhesive (Scotch-Weld DP460, 3 M) was used to joint the APPT-treated and as-received SMC substrates. The mechanical properties of the SMC used herein and the fully cured structural adhesive at room temperature are listed in Tables 1 and 2, respectively.

Plasma Treatment
The APPT apparatus illustrated in Fig. 1a was employed to treat the SMC substrates. In the plasma treatment process, a plasma jet flowing from a circular nozzle was sprayed over the SMC substrate, whose path is illustrated in Fig. 1b. The width of the plasma jet was set at 4 mm, and the treated area of SMC substrates (14 × 25 mm 2 ) was chosen to be slightly wider than the bonding area (12.5 × 25 mm 2 ). The gas flow pressure was set to 0.25 MPa for all substrates, and the atmospheric speed close to the exit of the circular nozzle was 7.2 m/s. Further, an operating discharge voltage of 8 kV and a discharge frequency of 12 kHz were used. To achieve a continuous treatment across the substrate area, the SMC substrates were placed on an automated moving platform that had a lateral speed of 10 mm/s. According to previous studies [17][18][19], the lap-shear strength of the adhesive-bonded joints is related to several factors of the APPT process such as the distance between the substrate and APPT apparatus nozzle, the working gas, and the duration of APPT exposure. In this work, the distance between the APPT apparatus nozzle and the substrate was varied to explore its effect on the lap-shear strength of the adhesive-bonded SMC joints. Vertical distances between the nozzle and the SMC substrates of 25, 20, 15, and 10 mm were used, and a photograph of the plasma jet operating with a treatment distance of 10 mm is given in Fig. 2.

Quasi-Static Lap-Shear Tensile Test
Using the ASTM D1002-2001 test method standard [20], quasi-static lap-shear tensile tests of the adhesive-bonded SMC joints were carried out to determine the lap-shear strength in a universal testing machine (MTS Systems Corp.).
To mitigate bending during the lap-shear tensile test of the adhesive-bonded SMC joints, two plates 3.0 mm thick were applied, which ensured that the tensile direction was roughly parallel to the SMC joint. To obtain the peak loads of the lap-shear joints, the crosshead speed was set at 2 mm/min until fracture. The lap-shear strength was acquired as an average value of five measurements obtained with the same conditions. Finally, analysis of the fracture surface was performed to estimate the fracture modes of the adhesive-bonded joints.

Characterization of Surface Morphologies
The influence that APPT has on the SMC substrate surface morphology was studied with SEM, while the surface roughness of the SMC substrates was measured by a 3D optical profiler. In this way, accurate measurements were obtained of the modifications induced on the substrate surface morphology by APPT.

Surface Chemistry Analysis
The surface chemistry state of the APPT-treated and asreceived SMC substrates was characterized by XPS (Escalab 250Xi, Thermo Scientific) whose X-ray source had a spot size of 500 μm. In the XPS spectra, the peak at 284.6 eV was referenced as C 1 s. In addition, a complete analysis of the different peaks was conducted using a Lorentzian-Gaussian peak shape.

Contact Angle Measurement and SFE
Surface wettability, which is an essential factor for improving adhesion, is commonly evaluated using contact angle values and the SFE [21]. Here, the contact angles on the SMC substrate surfaces were measured by the contact angle measurement system (OCA-20, Dataphysics) previously described in Ref. [22]. Table 3 lists the room-temperature surface tension data of the three contact angle testing liquids (i.e., distilled water, diiodomethane, and ethylene glycol), where the test liquid drop volume was 2 μL. The contact angle was calculated by sessile drop technology, where each average contact angle was calculated based on five identical measurements.
The SFE of the SMC substrates was calculated using Young's equation (refer to Eq. (1)), which represents the relation between the contact angle θ and the SFE [23]. The interfacial energy between the solid and the liquid γ SL can be expressed as where γ S and γ L are the SFE values of the solid and liquid, respectively [24]. Owens and Wendt [25] have proposed a geometrical way of connecting the dispersion and polar interactions in accordance with the ideas of Fowkes [26]. Following their proposal, the estimation for γ SL was expressed as where γ d S and γ d L are the dispersion components of SFE values of the solid and liquid, respectively, where γ To acquire the γ d S and γ p S values of the solid, we need to measure the contact angles of more than two liquid types whose surface tension components can be acquired. Herein, three typical measuring liquid types were involved to determine the γ d S and γ p S of the SMC substrates. The fracture mode of the adhesive-bonded joint is another indicator that can be used to evaluate the adhesive bonding performance. The fracture modes of various SMC joints are shown in Fig. 5. The fracture mode of the as-received SMC joint in Fig. 5a was adhesive failure, which indicated that the interface strength between the adhesive and the as-received SMC substrates was less than the shear strength of the 3 M DP460 adhesive. With APPT at a distance of 25 mm, the fracture mode was also adhesive failure, as suggested by Fig. 5b. With APPT at a distance of 20 mm, the fracture mode of the joints was adherend failure (Fig. 5c), which presented with a neat fracture at the bonding edge. In addition, the lap-shear strength reached 23.1 MPa. We note the lap-shear strength herein was always lower than 25.9 MPa. This is owing to the fact that the SMC substrates will fracture if the tensile force reaches 8.1 kN (108 MPa × 25 mm × 3 mm 8.1 kN) according to the tensile strength of SMC in Table 1, and the fact that for adhesive-bonded joints the lap-shear strength is 25.9 MPa (8.1 kN/(12.5 mm × 25 mm) 25.9 MPa) when the tension reaches to 8.1 kN. In addition, a bending moment exists at the junction of the joints. Therefore, the lap-shear strength value was always lower than 25.9 MPa. At the same time, the interface strength between the SMC substrates and adhesive was higher than 23.1 MPa as evidenced by the lack of adhesive failure. When the APPT distance was either 15 or 10 mm, the fracture mode was a thin delamination fracture, as shown in Fig. 5d. This is owing to the fact that at APPT distances of 15 and 10 mm the substrate temperature will increase to levels that can damage the SMC substrates and thus induce a delamination fracture. Furthermore, this indicates that the interface strengths produced at APPT distances of 15 and 10 mm are also greater than the strength between the carbon fiber and vinyl ester resin. Because the fracture modes in samples with treatment distances of 20, 15 and 10 mm were either adherend or delamination failure, the specific interface strength values between the adhesive and APPT-treated SMC substrates could not be measured. Nevertheless, according to Figs. 4 and 5, the interface strength between the adhesive and SMC substrates was improved after APPT. Additionally, the interface strengths at APPT distances of 20, 15 and 10 mm were higher than that with an APPT distance of 25 mm.
This relationship between the fracture mode and interface strength indicates that the interface strength between the SMC substrates and adhesive was enhanced by the APPT. To understand the reasons underlying the interface strength increase following APPT, further analyses were performed that focused on as-received SMC substrates and APPT-treated SMC substrates at treatment distances of 25 and 20 mm. The effect of APPT on the surface morphologies, chemical properties and SFE of SMC substrates will be explored in detail. Figure 6 presents the surface morphologies of APPT-treated and as-received SMC substrates. In Fig. 6a, we observe that the as-received SMC substrate surface morphologies are relatively out-of-flatness and that some material exists above the SMC substrates, which is most likely the release agent applied for the molding process. Much more uniform surface morphologies after APPT are observed in Fig. 6b, c, which indicates that the plasma jet removed SMC substrate surface material (e.g., the release agent and possibly somewhat of the resin). Roughness testing was used to estimate the surface morphology modification of SMC substrates induced by APPT. Figure 7 exhibits the surface roughness (R a ) values of APPTtreated and as-received SMC substrates, where it can be found the APPT treatment at the distances of 25 and 20 mm did not have an obvious effect on the surface roughness. This result is consistent with the SEM micrographs in Fig. 6.

Chemical Properties of Plasma-Treated SMC Substrates
Typical XPS survey spectra for the APPT-treated (20 and 25 mm distances) and as-received SMC substrates are given in Fig. 8, wherein obvious and intense peaks emerging in the spectra of the three samples demonstrate the existence of C  Table 4, where the amount of silicon is observed to decrease after APPT while the amount of nitrogen increased gradually with decreasing distance between the APPT nozzle and SMC substrate. Mold release agents are widely applied in SMC substrate fabrication where silicone elastomer-polydimethylsiloxane, comprising silicon-oxygen-silicon as chain and side-   chained carbon atoms, is the main component of the mold release agent. In the results herein, the amount of silicon decreased following APPT, which suggested that the mold release agent was gradually removed by the APPT. The increasing nitrogen amount was owing to the fact that nitrogen compounds present in the air entered the surface of resin when plasma treatment was applied [27]. Further, the amount of oxygen increased and the amount of carbon decreased significantly after plasma treatment. To further probe the detailed chemical states of carbon present in the samples, narrow-scan spectra were obtained and peak differentiation analyses were conducted. The XPS C 1 s binding energy spectra are shown in Fig. 9, and the functional group distributions on the surface of the SMC substrates are given in Table 5. The carbon signal is commonly used to calibrate the energy of the other existing signals in the XPS spectrum, so it is a trustworthy signal from which we can derive relative amounts. Here, the C 1 s peak was deconvoluted and found to divide into three distinct peaks. The peak at 284.6 eV corresponds to C-C or C-H bonds [28,29], while the higher oxidation states of carbon located at 286.7 eV correspond to C-O-H or C-O-C functional groups [28,29]. Finally, the peak at 288.5 eV is attributed to H-O-C=O or R-O-C=O groups [30,31]. The low oxidation states of carbon (i.e., peak at 284.6 eV) experienced a decreased signal intensity after APPT. The amounts of the three types of chemical bonds and the O/C ratio can be obtained from the deconvoluted C 1 s peak, and Table 5 clearly shows that the APPT increased the O/C ratio. When the APPT distance was 20 mm the O/C ratio was at a maximum, which is consistent with the O 1 s spectra results shown in Table 4. In addition, it can be seen that the amount of groups containing the C element had a dramatic decline and the amount of O-containing groups (C-O-H, C-O-C, H-O-C=O, R-O-C=O) strongly increased as the treatment distance decreased to 20 mm.

SFE of APPT-Treated SMC Substrates
The contact angles (illustrated in Fig. 10) of three kinds of liquids on the SMC substrates are listed in Table 6. After APPT, the contact angles of distilled water on SMC substrates decreased obviously, and thus APPT contributes to the wettability of SMC substrates. The SFE of the SMC sub- strates was estimated based on the method introduced by Owens and Wendt [25] and by Fowkes [26]. It was found that the SFE, which consisted of polar and dispersion components, increased significantly after APPT, as presented in Fig. 11. The SFE of the as-received SMC substrates mainly possessed a dispersion component of about 27.18 mJ/mm 2 , while the polar component was only 3.26 mJ/mm 2 . The dispersion component exhibited almost no change after plasma treatment while, interestingly, the polar component increased remarkably to 12.04 and 36.20 mJ/mm 2 as the treatment distances reached to 25 and 20 mm, respectively.

Discussion
Based on the above results, the lap-shear strength of adhesivebonded SMC joints was found to be remarkably enhanced by APPT. When the treatment distance was 20 mm, the maximum lap-shear strength was obtained, which was about three times more than the lap-shear strength of the as-  Fig. 8 and Table 5. Ni et al. [32] have found that, following APPT, the amount of ester groups (R-O-C=O) on the substrate surface decreased, whereupon more carboxyl groups (H-O-C=O) and hydroxyl groups (C-O-H) appeared. This offers an explanation for the increasing amount of carboxyl groups (H-O-C=O) and hydroxyl groups (C-O-H) following APPT. Blank et al. [33] have confirmed that chemical reactions can exist between epoxide groups and carboxyl or hydroxyl groups, which can operate as a crosslinking mechanism, as shown in Fig. 12. The possible reactions that may appear between the residual epoxide groups of the adhesive and the carboxyl groups (H-O-C=O) and hydroxyl groups (C-O-H) are exhibited in Fig. 13. These reactions suggest that APPT can cause the chemical groups of vinyl ester resin to become more active, which facilitates the combination of  [34], an increased amount of carboxyl groups can improve the lap-shear strength.
The SFE was found to increase significantly after APPT, and two reasons were posited to explain this phenomenon. First, the APPT increased the amount of oxygen, which facilitated the existence of polar molecules. Zaldivar et al. [34] have also concluded that wetting characteristics are influenced positively by the amount of oxygen. Second, the APPT removed the inert mold release agent present on the SMC substrates, where the former possesses a low surface tension.
Additionally, to clearly compare the relationship between the lap-shear strength and the surface roughness, the lapshear strength and roughness of SMC substrates treated with APPT at various distances are shown in Fig. 14. Similar surface roughnesses are observed on the APPT-treated and as-received SMC substrates, but the lap-shear strengths varied remarkably. It may be concluded that the surface morphology modifications were not responsible for the variation of the lap-shear strength. Specifically, the surface roughness contributed little benefit to the lap-shear strength and was not CA contact angle, SD standard deviation Fig. 12 The chemical reactions of a an epoxide group and a carboxyl group and b an epoxide group and a hydroxyl group [31] Fig. 13 Possible chemical combination between the residual epoxide groups of the adhesive and the carboxyl or hydroxyl groups  [35] and Schafer et al. [36]. In particular, the hydroxyl and carboxyl groups increased significantly after APPT and may be the primary reason for the improvement in the lap-shear strengths of APPT-treated SMC joints.