Improved Plating Adhesion of Polyarylate Fibers via Supercritical CO₂ and Impregnation of Slightly Soluble Oil on the Fiber Surface

Vectran®, a polyarylate fiber (1670 DTEX, 1500DEN) composed of 300 fiber strands, was obtained from Kuraray Co. Ltd., Tokyo, Japan. CO₂ (+ 99.99%) was purchased from Uno Oxygen Co., Inc., Tokyo, Japan and used as received. Hexadecylpyridinium chloride, octadecylamine, oleic acid, octadecanol, and sulfuric acid were purchased from FUJIFILM WAKO Pure Chemical Co., Ltd., Tokyo, Japan. Palladium(II) acetate complex, which was employed as a catalyst for electroless plating pretreatment, was purchased from Aldrich Chemical Co., Ltd. Chemicals, including OPC-80 catalyst ML and OPC-SAL-M, which were used to impregnate the catalyst into the fiber in a water system as a control, were purchased from Okuno Chemical Industry Co., Ltd., Tokyo, Japan. ATS-ADDCOPPER IW-A (solution A), ATS-ADDCOPPER IW-M (solution M), and ATS-ADDCOPPER IW-C (solution C) were purchased from Okuno Chemical Industry Co., Ltd., Tokyo, Japan. Thermosetting epoxy resin was supplied by Huntsman Japan Co., Ltd. and was used to prepare epoxy balls. The epoxy resin was cured by mixing bisphenol epoxy resin and butanediol glycidyl ether in a 1:1 ratio (Table 1).

Cu plating solutions were prepared by adding 10 mL of solution A, 16 mL of solution B, and 2 mL of solution C to 172 mL of distilled water. The plating solution was heated to a temperature of 42 ± 2 °C using a hot plate while stirring with a magnetic stirrer. The fiber samples were plated by dipping them into the solution for 15 min.

The four-point probe method was employed to measure the line resistivity of the plated fibers using a DC Milliohm meter (Agilent 34401A, Agient Technology Co., Ltd).

Thermosetting epoxy resin was applied to the surface of the plated Vectran® fibers to form epoxy resin balls. The samples were heated to 70 °C in air for 2 h to cure the epoxy balls. A jig was prepared by forming a slit on a metal plate to hook the epoxy balls. The jig was set in an Instron mention, and the plating film adhered to the epoxy resin balls from the Vectran® fiber. The interfacial share strength (IFSS) between the plating film and Vectran® fiber was calculated using Eq. (1). The surfaces of the epoxy resin balls and fibers after peeling were examined using a scanning electron microscope (SEM, Hitachi S-3400N) and a Keyence VHX-2000 microscope.where F is the maximum detection load (N), L the epoxy resin diameter (mm), and d the fiber diameter (mm).

After weighing the plated Vectran® fiber in a quartz decomposition vessel, nitric, hydrochloric, and sulfuric acids were added. The fiber was decomposed using an automatic microwave digester (DISCOVERSP-D80, CEM). Thereafter, it was scalped to 100 mL with ultrapure water and filtered (0.45 μm) after filtration. Inductively coupled plasma luminescence spectrometry (ICP-LS, Thermo Fisher Scientific Co., Ltd) was then employed to measure the Pd content of the fiber.

The amount of Pd at several nanometers from the surface of the fiber sample subjected to supercritical or reduction treatment was measured by X-ray photoelectron spectroscopy (XPS). All samples were analyzed using JAEOL 9010 MCY-XPS with monochromatic Mg Ka radiation (hv = 1253.6 eV) operated at 10 kV and 10 mA. The pressure in the cell was approximately 10⁻⁷ Pa. To measure the amount of Pd on the fiber surface, the binding energy scale was set to 285.0 eV for C 1s.

The distribution of Pd on the fiber surface was evaluated using time-of-flight secondary ion mass spectrometry (TOF-SIMS; ION-TOF TOF-SIMS 5). The ion source was Bi₃⁺, and the spectrometer was operated at 10 keV, 0.2 pA, and 10 kHz.

Electron probe X-ray microanalysis (JXA-8500F, JEOL Co., Ltd.) was performed to observe the interface between the plating film and the fiber. Using an ion source, the voltage and probe current were 0 kV and 50 nA, respectively.

Table 2 shows the samples after ScCO₂ and reduction treatments. The Vectran® fibers were yellow. After Pd impregnation, the fibers changed to orange or gray, and no further changes were observed during annealing. The orange color indicates that the metal complex was not reduced and the oil did not metalize the fiber surface. In particular, hexadecylpyridinium chloride was less reduced than oleic acid and octadecanol. When Pd acetate was impregnated into Vectran® fibers through supercritical treatment after adding oil to the surfaces, the fibers became darker, indicating that a large amount of Pd was incorporated into them. Therefore, hexadecylpyridinium chloride was less reduced than octadecanol and existed as a complex soluble in supercritical CO₂.

We evaluated the adhesion and bending fatigue resistance of the sample impregnated with Pd after applying oil and plating. Table 3 lists the bending fatigue test results. Figure 4 shows the effect of bending times on the electrical resistance of the samples. The addition of a surfactant or oil increased the bending resistance of the fibers and improved the adhesion between the fibers and the plated metal. When the rate of increase in the resistance after the bending test increased, the metal film peeled off. Furthermore, when Pd acetate was impregnated into the fibers through supercritical treatment, higher adhesion was obtained than when Pd was impregnated onto the fiber surface in an aqueous system and then plated (Fig. 5).

The specific resistance of the samples with oil was smaller than that of the samples with no oil applied. After supercritical treatment, the samples with oil were darker than those without oil. This shows that a large amount of Pd was impregnated into the fibers when oil was added, resulting in many anchors between the fiber and copper, thereby preventing peeling. To further understand the reasons for the obtained results, we measured the adhesion strength of the plating film using microdroplets (MDs).

Table 4 shows the effect of oil on the MD and bending fatigue test results, and Fig. 6 shows the effect on the IFSS of the samples. Figures 7 and 8 show the microscope and SEM images of the MD samples, respectively. To measure the IFSS between the plating film and the fibers, epoxy spheres were formed on the surface of the plating Vectran® and torn off. Samples subjected to supercritical treatment without oil exhibited higher adhesive strength than those impregnated with Pd in a water system. Furthermore, samples with higher IFSS exhibited higher bending resistance (Table 4). The IFSS and bending resistance of the samples were significantly improved when hexadecylpyridinium chloride was added to the fibers (Fig. 6). As shown in the microscopic images, the brown Cu plating coating on all of the fibers turned white during the MD test (Fig. 7), indicating that the interfacial adhesion between the Cu coating and the fibers could be evaluated using MDs. There was no significant difference between the surfaces of the samples treated in SCF and water (Fig. 8).

It was thought that the greater the amount of palladium on the surface of the fiber, the better the plating adhesion. Therefore, we analyzed the amount of Pd on the fiber surface using XPS (Fig. 9). Similar to the general case, when electroless plating was applied to the resin, such as Vectran®, the fiber surface was impregnated with Pd by placing the fiber in an aqueous solution in which the Pd colloid was dispersed. Metal coatings were formed on the fiber surface by plating in an aqueous solution. The fibers impregnated with Pd using an aqueous solution were used to prepare a control sample for comparison. Figure 10 shows the XPS spectra of the samples, and Table 5 lists the XPS data. The amount of Pd in the “ScCO₂ Without Oil” sample was lower than that in the water system sample. XPS analysis revealed that the amount of Pd on the surface of Vectran® before and after reduction impregnation with oil agents, such as hexadecylpyridinium chloride and cationic oil, was higher than that without oil agents (Table 5, Fig. 10). Electroless plating was conducted in an aqueous solution, and the adhesion of the plating was expected to improve because the water absorption of Vectran® is low and the amount of Pd present on the fiber surface is high.

The abovementioned high adhesion between the metal plating and the fibers was obtained because ScCO₂ has a hydrophobicity comparable to that of hexane and can swell Vectran® fibers, which is highly hydrophobic. The amount of Pd on the surface of Vectran® fibers after supercritical treatment was mapped using TOF–SIMS. Since this device detects secondary ions, the amount detected in the “shadow” of the fibers is small. For this reason, we also included “Total ion”, which includes secondary ions originating from fibers, and compared it with “Pd” to discuss uniformity. When each oil was applied on the samples, Pd was distributed throughout the Vectran® fibers (Fig. 9 a-1, b-1, c-1, and d-1). Furthermore, when analyzed under high magnification, the sample with oleic acid showed a mixture of light- and dark-colored areas, indicating high and low Pd concentrations, respectively, whereas the sample with hexadecylpyridinium chloride showed a more uniform Pd distribution (Fig. 9 a-2, b-2, c-2, and d-2). Figure 11 a-1 and b-1 show secondary electron images of the plated Vectran® fibers. There was no significant difference between the samples subjected to supercritical and water treatments. However, the ratio of Cu to fiber at the interface of the samples treated was lower than that of the samples treated in SCF, as revealed by SEM analysis. Bubbles and similar things are trapped in the sample treated in SCF (Figs. 10, 11). This is because owing to the low polarity of the Vectran® fiber, only a small amount of Pd adhered to its surface, and the samples were not uniformly plated in the initial stage of the plating process, resulting in a low density.

We considered not only the interface but also the inside of the fibers. ICP was employed to analyze the Pd distribution in the fiber samples. Table 6 lists the ICP analysis results for the Vectran® fiber samples with Pd. The samples subjected to supercritical treatment with oil applied on the surface showed a small ratio of Pd on the fiber surface (measured by XPS) to that in the entire fiber (measured by ICP), indicating that Pd was impregnated in the fiber direction from the center of the interface between the plating film and the fiber. Thus, the anchor effect was exhibited and adhesion improved.

We investigated the reason for the increase in the Pd impregnation rate on the surface when oil was applied to the fiber surfaces. Figure 12 shows the changes in the weight of the samples after adding various oils to the surface and treating them in ScCO₂ at 80 °C and 25 MPa for 30 min. Oleic acid and octadecanol have a carboxyl group and a hydroxyl group. They reduced Pd acetate and metalized it on the fiber surfaces. Thus, Pd could not penetrate deep into the fiber [17]. Because most of the oils dissolve in ScCO₂, the fibers could not be impregnated with a large amount of Pd. In contrast, hexadecylpyridinium chloride remained on the fiber surface without dissolving in ScCO₂ because it is a surfactant salt. Vectran®, a liquid crystalline resin, has a low degree of swelling in ScCO₂, and the amount of Pd that can adhere to the fiber surface per unit of time is limited. In contrast, by arranging the salt of a cationic surfactant with a high affinity for Pd acetate on the fiber surface, Pd acetate was concentrated on the surface of the Vectran® fiber.

ScCO₂ is generally less soluble than liquids because of its lower density. In addition, if the affinity of a functional agent to ScCO₂ is too high, when impregnating the functional agent on the fiber, the functional agent remains in the ScCO₂, and the amount of agent impregnated inside the fiber decreases. The fibers were impregnated with functional agents in ScCO₂ using the following steps:

When using ScCO₂ as the medium, adding oil to the fiber increases the Pd impregnation rate owing to the following two reasons:

Vectran® is a liquid crystalline polymer with a complex crystal structure to microfibrils. Pd was introduced into the interior of the fiber by swelling with ScCO₂, not in the interior of the crystal but in the region where the interaction between the fibrils was relatively small compared with that of the crystal.