1. Introduction
Carbon fiber-reinforced plastics (CFRPs), in which carbon fibers [
1,
2] are used as reinforcement materials, are lightweight and have excellent strength. CFRPs are used in a wide range of products, from fishing rods to structural parts of airplanes and vehicles [
3,
4,
5]. The prepreg structure, or the fabric woven from a continuous fiber reinforced with thermosetting resin, has been the mainstream in CFRP. A long period is required for the curing reaction, resulting in long manufacturing times and other problems, making them unsuitable for mass production [
6]. However, the emergence of carbon fiber-reinforced thermoplastics (CFRTPs) reinforced with short fibers with a length of several hundred micrometers to several millimeters and the application of injection molding techniques to CFRTPs have brought about a higher degree of freedom in shape, enabling the mass manufacturing of products with complex three-dimensional shapes. The problem with short fibers is their reinforcing effect, which is inferior to that of continuous fibers, which, in turn, results in the low mechanical strength of products [
7]. In recent years, sheets of continuous fibers reinforced with thermoplastics have been developed, and techniques utilizing the features of both continuous and short fibers have been proposed to produce complex three-dimensional molded products with high specific strength in a short molding cycle time. Specifically, hybrid injection molding, a type of overmolding technique, has been introduced in practical use, in which a CFRTP sheet with a woven structure is first compression molded while being softened by radiation heating, and then a short carbon fiber-reinforced thermoplastic is injected over the primary molded product to produce molded products with complex shapes, such as ribs for reinforcement effects [
8,
9,
10,
11,
12]. With this molding technique, however, the CFRTP sheet comes into contact with the low-temperature surface of the mold cavity during compression molding and solidifies. As a result, the transferability of the shape of the cavity surface to the molded products decreases. It has, therefore, been difficult to use a hybrid injection molding technique for the manufacturing of defect-free molded products. To address this issue, heat-and-cool hybrid injection molding using the variotherm technique has been applied. This hybrid injection molding technique involves heating the mold cavity surface to a temperature higher than the melting point of the base polymer of the CFRTP sheet [
13,
14].
Regarding the hybrid injection molding technique without heating and cooling, a number of studies have been conducted in a wide range of areas, from sheet materials to injection-molded polymers [
15]. Among them, hybrid injection-molded products are widely studied regarding the molded parts’ strength, mainly because they are used as lightweight structural parts that do not require assembly. One such study focused on the effect of the injection molding process conditions on the adhesive mechanism and adhesive quality between fiber-reinforced thermoplastic sheets and long fiber-reinforced thermoplastics [
16]. The relationship between the preheating time of the thermosetting resin sheet and injection molding process conditions, such as the temperature and injection rate of the injection-molded polymer, and the adhesive strength has been studied using measurements of the melt pressure inside the cavity and mold temperature [
17,
18]. It has been clarified that the preheating time has a significant effect on the adhesive strength, with higher cavity pressure and mold temperature conditions resulting in higher adhesive strength. Similar studies to the above have been conducted for CFRTP sheets using measurements of the melt pressure and temperature in the cavity [
19]. It has been clarified that as the temperature of the injection-molded polymer increases and the injection rate increases, the interface temperature increases, resulting in an increase in adhesive strength. Moreover, an increase in preheating time causes roughness on the sheet surface, resulting in an increase in adhesive strength due to the anchor effect. Studies are also being conducted on pretreatment methods for sheet surfaces to increase adhesive strength. For example, the improvement in the adhesive strength in the rib part has been attempted by cutting the continuous fibers near the surface of the CFRTP sheet shorter and entangling these fibers with the injection-molded polymer at the base of the rib [
20]. Furthermore, the effects of the linear expansion coefficient and the bending modulus of the CFRTP sheet and injection-molded polymer on the warpage of molded products have been examined [
21]. It has been clarified that greater warpage is generated in the molded products in injection-molded polymers with higher linear expansion coefficients and in injection-molded polymers with a lower bending modulus. In addition, a study has been conducted on the influence of the environmental temperature on the molded product’s strength [
22], as well as a study on the development of FEM simulation methods to assist in the design of molded products [
23].
On the other hand, few studies have been conducted on heat-and-cool hybrid injection molding. Overmolding molds that can be heated by applying rapid heat cycle molding (RHCM), one of the variotherm techniques, have been fabricated to study the effect of mold heating on the surface quality of molded products [
24]. It has been clarified that heating to 150 °C lowers the surface roughness Ra of CFRTP sheets from 0.43 μm to 0.06 μm, compared with conventional molding at a constant temperature of 85 °C. However, the effects of the mold heating temperature and injection molding process conditions on the surface properties and surface defects of CFRTP sheets in heat-and-cool hybrid injection molding have not been studied over a wide range of conditions. For this reason, it has been impossible to take extensive measures when the dimensional accuracy of the molded products is low, the surface properties are poor, or molding defects such as gas burning are generated.
There are several variotherm techniques for the heating and cooling of a mold, such as circulating a heating medium and a cooling medium alternately through the mold [
25,
26], using an electric heater [
27,
28], or using an electromagnetic induction-heating system [
29,
30,
31,
32,
33,
34,
35], which have been applied in practice. A method of heating the mold surface using infrared radiation has also been proposed [
36]. However, infrared radiation heating is rarely used in the manufacturing process. This is because the heating rate is lower and the molding cycle time is longer than that of the variotherm techniques described above because of the large energy losses due to reflection on metal surfaces with relatively low surface roughness, such as mold surfaces. However, compared with the above variotherm techniques, infrared radiation is used in research applications where the molding cycle time is not an issue because it is easier to heat mold surfaces where nano- and micro-shapes are processed or where surface defects are generated from outside the mold. Some of the authors of this paper have proposed a heating and cooling injection mold that can be easily heated using a far-infrared radiation heater and a three-dimensional cavity plate. The effects of heating the mold cavity surface on the appearance of molded products composed of fiber-reinforced thermoplastics have been examined using this mold [
37,
38]. Since an infrared radiation heater is often used to heat the sheets in overmolding, we speculated as to whether it would be possible to heat both the sheet and the mold surface simultaneously using only one far-infrared radiation heater.
In this study, the above molds [
38] are used as the basis, and a mold is designed and fabricated to continuously perform the following three processes using only an injection molding machine: (1) the radiation heating of both the CFRTP sheet and the mold cavity surface using a far-infrared radiation heater, (2) the compression molding of the CFRTP sheet, and (3) the injection molding of the melt. The effects of the heating temperature conditions of the mold, the injection molding process conditions, the number of CFRTP sheet layers, and the compression ratio of the sheet on the appearance characteristics of the molded products are examined using this mold and a far-infrared radiation heater.
3. Results
3.1. Carbon Fiber Tow Orientation
Figure 10 shows the appearance of the hybrid injection-molded product obtained using the CFRTP sheet with the fiber tow orientation of 0°/90°. The molded product is obtained under the following conditions: the number of sheet layers is one,
α is 12.5%, the heating temperature is 50 °C → 280 °C → 50 °C, and Δ
t is 10 s. By using the developed mold and a far-infrared radiation heater, it is confirmed that heat-and-cool hybrid injection-molded products can be obtained using only an injection molding machine. Using this molding technique, the effects of the sheet conditions and injection molding process conditions on molded products are examined, as described below.
Figure 11 shows the molded products obtained by changing the fiber tow orientation. The number of CFRTP sheet layers is one,
α is 12.5%, the heating condition is 50 °C → 240 °C → 50 °C, and Δ
t is 15 s. To examine the distortion of the woven pattern, the points where the fiber tows in two directions intersect are connected by yellow dotted lines. The position of the gate hole is displaced laterally from the center and the shape of the hole is deformed into an ellipse in both molded products. This is assumed to be because the CFRTP sheet moves slightly during compression molding, owing to the imbalanced clamping force of the sheet-holding plates. It has been observed that, if the clamping force is too large, the CFRTP sheet is not drawn into the cavity during compression molding, resulting in sheet fracture. These results indicate that caution is needed in adjusting the clamping force applied to the CFRTP sheet. Distortion occurs in the molded product with a +45°/−45° fiber orientation, whereas no distortion occurs in the molded product with a 0°/90° fiber orientation. This is assumed to be because in the 0°/90°-sheet-molded product, the orientation of the fiber tow is aligned with the direction in which the CFRTP sheet is pulled during compression molding. The subsequent experiments are conducted under the fiber tow orientation condition of 0°/90°, under which no distortion occurs.
3.2. Effects of Heating Temperature of Three-Dimensional Cavity Plate and Δt
Figure 12 shows the digital camera images of observation area
C (denoted in
Figure 3b) for the molded products, obtained by setting Δ
t to 0 s and changing the heating temperature of the three-dimensional cavity plate. The number of CFRTP sheet layers is one and
α is 12.5%. The positions of the ribs molded on the back surface of the molded products are indicated by dotted lines. Dent surface defects are generated in the compression area of the molded products under all heating conditions. The surface defects do not disappear, even when the heating condition is 50 °C → 280 °C → 50 °C, although the surfaces of the molded products gradually become smooth with the increasing heating temperature. To observe the surface defects in detail, a shape analysis laser microscope is used to magnify the compression area
D (denoted in
Figure 3b), and the results are shown in
Figure 13. The results at 50 °C → 180 → 50 °C and 50 °C → 280 → 50 °C are shown. Dent surface defects are generated in the compression area of the molded product under heating conditions of 50 °C → 180 °C → 50 °C. The fiber weave pattern is exposed at the bottom of the dents, which hereafter is referred to as the “exposed fiber weave pattern”. The surface defects do not completely disappear although the area where the surface defects are generated decreases with the increasing heating temperature.
Figure 14 shows the observation results of molded products formed by changing Δ
t under the heating temperature conditions of 50 °C → 280 °C → 50 °C where surface defects do not completely disappear, as shown in
Figure 13. The area where the exposed fiber weave pattern is generated gradually decreases with increasing Δ
t.
Figure 15 shows the images taken with a shape analysis laser microscope and the surface shapes of the molded products obtained by setting Δ
t to 0 s and 15 s. The surface becomes smooth and the dents and the exposed fiber weave pattern completely disappear when Δ
t is 15 s. A similar tendency is also observed under other heating conditions.
As described above, surface defects are generated when the heating temperature of the three-dimensional cavity plate is low. These defects do not completely disappear although the area where the surface defects are generated gradually decreases with the increasing heating temperature. Δt strongly affects the appearance of molded products. Surface defects are generated in the compression area of the molded product when the melt is injected immediately after the completion of the compression molding of the CFRTP sheet. On the other hand, it is found that as Δt increases, a flat surface profile without surface defects is generated on the molded product’s surface.
3.3. Effects of Number of CFRTP Sheet Layers and Compression Ratio
Figure 16 shows the molded products obtained by changing the number of CFRTP sheet layers and
α.
Figure 17 shows an enlarged view of the compression area of the molded product taken with a shape analysis laser microscope. The heating condition is 50 °C → 280 °C → 50 °C and Δ
t is 0 s. When the number of CFRTP sheet layers is the same, the exposed fiber weave pattern is generated at a low
α of 12.5%. The surface defects are not generated at a high
α of 37.5%. When
α is the same, the generation of the surface defects is suppressed more with two CFRTP sheet layers than with one CFRTP sheet layer, resulting in a glossy molded product surface. Molded product surfaces with excellent glossiness are generated when the number of CFRTP sheet layers is two and
α is 37.5%. However, the fiber tow width in the compression area of the molded product is different from that in the rib part, resulting in an irregular weave pattern. In particular, the weave pattern in the rib part is distorted and appears to be compressed from the top and bottom, as shown in
Figure 16d.
As described above, as the number of CFRTP sheet layers and the sheet compression ratio α increase, smooth molded product surfaces without defects are generated. However, the fiber weave pattern becomes irregular when the number of CFRTP sheet layers is two and the sheet compression ratio α is high.
3.4. Observation of Cross Sections of Molded Products
Figure 18 shows the cross section
E of the molded product (as denoted in
Figure 3b). Both the front and back of the CFRTP sheet are embedded in the injection-molded polymer in the frame part of the molded product. This is assumed to be because the CFRTP sheet in the frame part is not sandwiched between the three-dimensional cavity plate and the core insert and is held aloft in the air during compression molding.
Figure 19 shows the cross section of the fiber tow in the compression area of the molded product obtained at different compression ratio
α values when the numbers of CFRTP sheet layers are one and two. The heating condition is 50 °C → 280 °C → 50 °C and Δ
t is 0 s. The width of the fiber tow,
W, does not change with
α in the molded products with one CFRTP sheet layer, as shown in
Figure 19a,b. The width of the fiber tow in layer
II on the core insert side,
Wb, changes little with
α in the molded products with two CFRTP sheet layers, as shown in
Figure 19c,d. However, the width of the fiber tow in layer
I on the three-dimensional cavity plate side,
Wf, increases with
α because the fiber tow is deformed in the thickness direction of the molded product and flattened. As a result,
Wf becomes larger than
Wb.
3.5. Results of Measurement of Sheet Temperature
Figure 20 shows the temperature measurement results at position
A in the CFRTP sheet when Δ
t is 0 s, and the heating condition of the three-dimensional cavity plate is changed. The parameter
t1 indicates the time to start lifting the far-infrared radiation heater from the three-dimensional cavity plate surface, and
t2 indicates the time to complete compression molding. As the heating temperature of the three-dimensional cavity plate increases, the temperature of the CFRTP sheet increases during the time period from the start of the lifting of the far-infrared radiation heater (
t1) to the completion of compression molding (
t2).
At the time of t1, when the temperature of the three-dimensional cavity plate reaches its maximum, the surface temperature of the sheet is 210 °C, which is higher than the temperature of the plate under the condition of 50 °C → 180 °C → 50 °C. At the same time, for the condition of 50 °C → 280 °C → 50 °C, the surface temperature of the sheet is 240 °C, which is lower than the temperature of the plate. The causes of this are discussed below. The CFRTP sheet is hidden behind the three-dimensional cavity plate; therefore, it is difficult to heat it by radiation. At the start of heating, the surface temperature of the sheet, which has a smaller heat capacity than the plate, rises more rapidly. On the other hand, when it reaches a high temperature, the temperature of the plate reaches 280 °C by radiation and heat transfer from the heater, but that of the sheet does not reach 280 °C because heating by radiation is difficult.
The CFRTP sheet temperatures at
t1 and
t2 increase as the heating temperature of the plate increases. It is assumed that the viscosity of the base polymer of the CFRTP sheet decreases as the sheet temperature increases with the heating temperature [
39], and the CFRTP sheet is more susceptible to molecular diffusion under the condition of a high temperature at the time of compression completion
t2, resulting in the squeeze flow of the low-viscosity base polymer over the surface of the CFRTP sheet during compression molding. As a result, it is assumed that the surface properties of the three-dimensional cavity plate are easily transferred to the surface of the CFRTP sheet.
Figure 21 shows the temperature measurement results at position B in the rib part of the CFRTP sheet and the three-dimensional cavity plate during heat-and-cool hybrid injection molding, as measured while changing Δ
t. When Δ
t is 0 s, the sheet temperature decreases until the completion of compression molding (
t2), then increases again, and decreases rapidly. On the other hand, when Δ
t is 15 s, the sheet temperature decreases rapidly even after the completion of compression molding (
t2), decreasing to 150 °C after 15 s (
t3). The results suggest that the sheet temperature at position
A in the compression area of the molded product, where both sides of the CFRTP sheet are in contact with the cavity surface, is lower than that at the completion of compression molding
t2 in
Figure 20. The sheet temperature again increases when the melt is injected 15 s later. This is a result of the two-stage molding process: (1) the compression molding stage—heat is transferred from the CFRTP sheet to the cavity surface when they come into contact, causing a decrease in the sheet temperature; (2) injection molding stage—heat is transferred from the injection-molded polymer to the CFRTP sheet, causing an increase in the sheet temperature. However, the sheet temperature when the melt is injected 15 s later is lower than that at the completion of compression molding (
t2). Under such temperature condition, it is difficult for molecules to diffuse at the interface between the base polymer of the CFRTP sheet and the injection-molded polymer, and the adhesive strength between the sheet and the polymer is presumably lower than in the case of Δ
t = 0 s.
As described above, it is found that the heating temperature of the three-dimensional cavity plate and Δt affect the temperature of each part of the CFRTP sheet during injection molding.
4. Discussion
Surface defects with the exposed fiber weave pattern are generated in the compression area of the molded product when Δt is 0 s, namely, when injection molding is started immediately after the compression molding of the CFRTP sheet. On the other hand, molded products having a smoother surface without defects are obtained when Δt is increased. The sheet temperature increases with the heating temperature of the three-dimensional cavity plate. The viscosity of PMMA, the base polymer of the CFRTP sheet, decreases in the compression area of the molded product during compression molding. As a result, the base polymer squeezes out onto the surface of the CFRTP sheet and adheres to the three-dimensional cavity plate. This is assumed to be one possible reason that the surface properties of the three-dimensional cavity plate are fully transferred to the molded product’s surface, resulting in mold products with excellent surface properties. However, molded products with the optimal surface properties are not obtained when only the heating temperature is increased. It is found that the surface defects completely disappear only when Δt is increased. Moreover, surface defects are less likely to be generated when the number of CFRTP sheet layers and the sheet compression ratio α are high. These results are discussed below.
Figure 22 shows a diagram of the predicted behavior of the melt flowing from the rib part to the frame part of the mold. Only the runner and the melt in the frame part are drawn while the CFRTP sheet is omitted in this figure. The melt passes through the rib part, as shown in
Figure 22a, and flows into the frame part, as shown in
Figure 22b. The frame part becomes fully filled with the melt when the melt fronts meet at the corners of the frame part, as shown in
Figure 22c.
Figure 23 shows a diagram of the estimated mechanism of the generation of surface defects with the exposed fiber weave pattern at the
G–G cross section in
Figure 22c. Immediately after compression molding, as shown in
Figure 23a, the sheet remains softened inside the frame, without being cooled, because the sheet is not sandwiched between the three-dimensional cavity plate and the core insert. Then, the melt flows through the space on the core insert side of the sheet in the frame part when injection starts, as shown in
Figure 23b. The melt pressure increases after the space on the core insert side of the sheet is filled with the melt. As a result, the melt flows through the gaps between the fiber tow of the sheet into the unfilled space on the three-dimensional cavity plate in the frame part, as shown in
Figure 23c. In this study, the pressure distributed on the CFRTP sheet and the melt pressure are not measured by the pressure sensor. Therefore, the following discussion is only speculative. Assuming that the holding pressure of 60 MPa is directly applied to the melt inside the frame part if ignoring the pressure loss, it can be considered that the pressure distributed in the sheet of the compression area of the molded product is lower than the pressure of the melt inside the frame area under the low
α condition. As a result, the residual air in the unfilled space is pushed by the melt toward the contact area between the CFRTP sheet and the three-dimensional cavity plate, and it moves into the compression area of the molded product, as shown in
Figure 23d. When Δ
t is short, the viscosity of PMMA, the base polymer of the CFRTP sheet, is low because PMMA is maintained at a high temperature. It is assumed that the air moving into the compression area of the molded product pushes the PMMA inward in the compression area of the molded product, resulting in the generation of dent surface defects with the exposed fiber weave pattern. When Δ
t is long, the residual air is also pushed to the compression area of the molded product. However, as shown in
Figure 23e, the air diffuses throughout the contact area between the CFRTP and the three-dimensional cavity plate because the sheet in the compression area of the molded product has already cooled and solidified as the heat from the sheet is transferred to the cavity wall. As a result, it is assumed that the above surface defects are not generated. When
α is high, it is assumed that little air penetrates the contact area between the CFRTP sheet and the three-dimensional cavity plate because the pressure inside PMMA in the compression area of the molded product is higher than the melt inside the frame part due to stronger compression. The surface defects are, therefore, not generated.
Figure 24 demonstrates why surface defects are not generated when the number of CFRTP sheet layers is two, even when
α is the same. As seen in the cross section in the compression area of the molded product shown in
Figure 19d, only the fiber tow of layer
I on the three-dimensional cavity plate side is greatly deformed when the number of CFRTP sheet layers is two and
α is 37.5%. When the CFRTP sheet has two layers, layer
II of the CFRTP sheet is on the core insert side, as shown in
Figure 24a. When mold clamping starts, layer
II first comes into contact with the core insert, which is kept at a low temperature and begins to solidify from this moment, as shown in
Figure 24b. The thickness of layer
II, whose viscosity has increased since solidification began, changes little even when it is further compressed during mold clamping. On the other hand, the thickness of layer
I, which is still softened, largely decreases when it is compressed. Layer
I is significantly deformed, resulting in increased pressure inside the PMMA. Therefore, little air penetrates the contact area between the CFRTP sheet and the three-dimensional cavity plate. This is assumed to be the reason that surface defects are not generated when the CFRTP sheet has two layers. It is also assumed that as
α increases, layer
I is greatly deformed in the compression area of the molded product, protruding out of the compression area of the molded product and into the rib part, compressing the CFRTP sheet in the rib part. As a result, the weave pattern becomes irregular in the rib part, as shown in
Figure 16d.