Enhanced Interfacial Adhesion of Nylon 66 to Epoxy Resin EPON 825 by Non-thermal Atmospheric Pressure Dielectric Barrier Discharge Plasmas

Abstract

Poly(hexamethylene adipamide), nylon 66, is a popular plastic that requires high surface wettability and strong adhesive bonds for many applications. However, pristine nylon is difficult to bond due to its hydrophobic nature and poor surface wettability. The objective of this work was to modify the physio-chemical surface properties of nylon 66 via a novel atmospheric plasma surface treatment approach using oxygen (O₂) or water vapor (H₂O) plasma glow. The surface hydrophilicity of the plasma-treated nylon surface was substantially enhanced immediately after either helium (He)/H₂O or He/O₂ plasma surface treatment. The average water contact angle was reduced from 65 degrees to ~30 degrees after He/H₂O plasma and ~40 degrees after He/O₂ plasma treatments. The improved hydrophilicity was also evidenced by the increased intensities of the surface oxygen and hydroxyl bonds in the X-ray photoelectron spectra. The interfacial adhesion strength of nylon surfaces before and after plasma treatment was further evaluated by uniaxial tensile tests of nylon single-joint lap shears bonded with three adhesives, i.e., thermoset epoxy resins EPON 825/ JEFFAMINE D-230 and EPON825/JEFFAMINE D-2000, and the thermoelastic polyurethane adhesive Sikaflex 252. The most significant improvements in bond strengths due to plasma treatment were found for lap shears bonded with the EPON 825/JEFFAMINE D-230 epoxy resin; their shear strengths with maximum loads were more than doubled—from 299–451 to 693–1594 N—after plasma treatment and were further enhanced by a factor of four to 895–1857 N after a subsequent silane treatment. In contrast, the bond strength of lap shears bonded with EPON 825/JEFFAMINE D-2000 and Sikaflex was not significantly improved because of the different a, re-affirming the importance of adhesive bulk properties This work presents the preliminary success of effective surface functionalization leading to enhanced interfacial adhesive bonds for nylon 66 via the development of scalable atmospheric plasma surface treatments.

1. Introduction

Nylon 66, a thermoplastic synthesized by polycondensation of hexamethylenediamine (hexane-1,6-diamine) and adipic acid [1,2], is one of the most used plastics. It has many unique properties, including its strong, tough, low-friction, good electrical and thermal-resistant surface features. For these reasons, it is traditionally used to make a variety of plastic products, as described in [3]. Many potential final applications of nylon 66 require satisfactory plastic-to-plastic binding strengths leading to strongly reinforced composites, which is difficult to achieve with pristine nylon 66 due to its hydrophobic surface nature and poor surface wettability [4,5].

On the other hand, several thermoset epoxy resins, such as EPON 825, is a high purity bisphenol A epichlorohydrin liquid epoxy which is widely employed in fabricating various polymeric composites [6,7] including nylon. It is mainly because of its easy formulation and curing when using suitable application-dependent agents. To further enhance the resistance of a nylon composite against high mechanical impacts, it is important to seek feasible, scalable, and easy-to-implement methodologies to improve the surface binding strength of nylon to popular thermoset adhesives such as EPON 825, with tailored surface wettability and chemical functionality [8]. Different approaches have been applied to improve the morphological, mechanical, and chemical properties of nylon-based surfaces via a variety of techniques, including UV and electron beam irradiations [9,10], chemical treatments [11,12], nanoparticle coating [13], and plasma treatments [14,15,16,17,18,19,20,21,22,23,24,25]. These previous studies attempted to characterize nylon surface wettability and chemistry, as well as seeking feasible plasma-enhanced surface alterations. These efforts focused on surface science studies as a function of experimental conditions, providing few insights on the effect of impact on the resultant nylon composites bonded with popular adhesives.

In a previous work [26], we reported a preliminary study in which we altered the surface hydrophobicity of nylon 66 via a non-thermal atmospheric dielectric barrier discharge (DBD) plasma glow utilizing a custom-made L-shape plasma reactor with either 5% oxygen (O₂) or water vapor (H₂O) plasma in helium (He). In that study, we demonstrated enhanced binding capability with EPON 825 [26]. The present work describes our continuing efforts to advance the technical understanding of the complicated correlations of key plasma-treatment conditions, including gas type, composition, and duration, on nylon surface chemistry and wettability. We discuss how the plasma treatment affects the resultant effectiveness/ineffectiveness of nylon’s binding strength to the popular thermoset epoxy resin EPON 825 using two curing agents, i.e., JEFFAMINE D-230 and JEFFAMINE D-2000, which have significantly different epoxy crosslinking characteristics. The different properties of the thermoset adhesives and their resulting bond strengths in nylon composites are also compared to that of a thermoelastic, one-component polyurethane adhesive, using Sikaflex 252 as an example. The purpose is to compare untreated and plasma-treated nylon surfaces to study the possible effects of plasma beyond interfaces in terms of the different thermal responses of adhesives (thermoset versus thermoelastic) and different levels of curing (JEFFAMINE D-230 versus JEFFAMINE D-2000).

2. Materials and Methods

2.1. Nylon 66 Sample Preparation

Two types of poly(hexamethylene adipamide), i.e., nylon 66 samples (Goodfellow Corporation, Pittsburgh, PA, USA), were used. First, 0.5-mm-thick films were cut into 2.5 mm × 2.5 mm squares for plasma treatments under various conditions. The subsequent characterization included X-ray photoemission spectroscopy (XPS) and water contact angle (WCA) measurements. Next, 50-µm-thick sheets were cut into 5 mm × 30 mm strips for plasma treatment under the best-identified conditions, i.e., 0.5-mm-thick films; these were then made into lap shears for Instron mechanical tests.

2.2. Plasma Setup and Surface Treatments

All atmospheric plasma surface treatments were performed using a custom-made L-shaped dielectric barrier discharge (DBD) reactor. Photos of the reactor have been shown previously elsewhere [26]. Briefly, the reactor (~20 cm × 15 cm × 8 cm) consisted of a custom-built acrylic gas manifold between two symmetric mica-bonded copper (Cu) high-voltage (HV) electrodes (~8 in × 3 in). Mica was chosen because of its good electrical properties, high temperature resistance, and ease of fabrication [27]. The gas manifold had a total of eight splits and 16 parallel channels across a total distance of 6 inches to allow the feed gas to enter the reactor from the top, splitting four times and exiting the reactor evenly through a narrow slit (1.5 mm wide) to generate a uniform plasma glow under the bottom plate of the reactor. He (ultrahigh purity, Earlbeck Gases & Technologies, Baltimore, MD, USA) at a volumetric flow rate of 6 standard liters per minute (SLPM) was used as the diluting gas to introduce either H₂O vapor through a glass bubbler or O₂ (Air Products, 99.999%) of different compositions (1 to 5 vol.%, i.e., 0.06 to 0.3 SLPM) as the feed gas. Figure 1 illustrates the entire experimental apparatus setup, including gas flow meters (Alicat Scientific Precision Gas Mass Flow Controllers, MCR-series, Tucson, AZ, USA ) for independent flow rate adjustments of the precursor gas (H₂O or O₂) and He, an AC, HV pulsed power source at ~33 kHz frequency and ~12 kV load condition, an oscilloscope (Teledyne LeCroy WaveRunner 6 Zi, Chestnut Ridge, NY, USA), a HV probe (North Star PVM-5; 1000×, Bainbridge, WA, USA), and a current monitor (Pearson Model 8181, Peterson Electronics, Inc., Palo Alto, CA, USA), in addition to the DBD reactor. The cut, as-received nylon samples were cleaned with ethanol and methanol, air dried, and then left to rest on a motorized stage that moved back and forth horizontally at a rate of ~25 mm/s. This was placed underneath the bottom HV electrode plates, where gas exited at 2 mm above the surface of nylon samples. The entire stage also served as the ground electrode, so that a uniform, light purple plasma glow appeared when the power source was turned on.

2.3. Characterization of Surface Functionality before and after Plasma Treatments

The hydrophilicity/hydrophobicity of the nylon surface before and after plasma treatment was examined by measuring the static WCA using an automated goniometer/tension meter (ramé-hart Model 290, Succasunna, NJ, USA, using ramé-hart DROPimage Advanced v2.6). First, ~2.5 µL of high-performance liquid chromatography (HPLC)-grade distilled water was dispensed on the film to create a real-time image of the two-dimensional drop profile through a charged-coupled device (CCD). The tangent angles were then measured on the left and right sides with the vertex at the three-phase lines which formed between the film and the drop profile. At least five measurements were taken at different spots on the nylon surface to obtain an average WCA value for each sample. An aging study was also performed by repeating the WCA measurements for the He/O₂ or He/H₂O plasma-treated films after one and six days, respectively.

Near-surface compositional measurements were performed by XPS using a Kratos Axis Ultra system (Kratos Analytical Ltd., Manchester, UK) equipped with a hemispherical analyzer. The sample was irradiated with a 150-W monochromatic Al Ka (1486.7 eV) beam, and both magnetic and electrostatic lenses were used to select photoelectrons from a 1 × 2 mm area of the nylon surface with a take-off angle of 90°. The pressure in the XPS chamber was held at 10⁻⁹−10⁻⁸ torr. High-resolution spectra for carbon (C) 1s, nitrogen (N) 1s, and oxygen (O) 1s were acquired using 20.0-eV pass energy and calibrated relative to 285.0 eV (C-C binding state). All spectra were processed, deconvoluted, and their surface compositions calculated using the Kratos v2 software. Areas with high-resolution peaks were used in quantifications to calculate the elemental compositions and relative ratios among important functional groups in different chemical environments after deconvolution.

2.4. Silane Treatment

The plasma-treated nylon sample with the best surface functionality was subsequently dipped into a 1 wt% 3-aminopropyltriethoxysilane (APS) (Sigma-Aldrich, St. Louis, MO, USA) solution in 90/10 ethanol/water for 1 min, followed by drying in nitrogen (N₂) and oven heating at 100 °C for 1 h.

2.5. Adhesive Preparation

Three adhesives were used to make the lap shears. A low-viscosity, high-purity bisphenol A epichlorohydrin epoxy resin EPON 825 (Momentive Inc., Columbus, OH, USA) was mixed with a difunctional polyetheramine curing agent, either JEFFAMINE D-230 or JEFFAMINE D-2000 (Huntsman Inc. USA, Woodlands, TX, USA) [6], in a mixing container until the viscosity noticeably increased. Both the D-230 and D-2000 agents had oxy-propylene repeating units in their backbone with different average molecular weights (Figure 2) [7], thus requiring different mixing ratios and curing times to achieve a suitable viscosity for bonding. A ready-to-use, single-component, polyurethane-based, non-sag elastomeric sealant, i.e., Sikaflex 252, was also explored when making the lap shears, as a comparison to the two EPON 825 thermoset epoxy resins cured with JEFFAMINE D-230 and D-2000.

Figure 3 shows the flow charts of preparation procedures for the three different adhesives. The experimental conditions were based on both the recommendation from the manufacturer for JESSAMINE D-230 and D-2000 [7] and our previous experiences.

2.6. Lap Shear Assembly Procedure

Single-lap-joint specimens were assembled by bonding two 100 mm × 25 mm × 5 mm pieces of nylon using the three adhesives prepared as described previously. The lap-shear joints were bonded using a custom-made 80/20 aluminum fixture to ensure optimal alignment of the 12.5 mm × 25 mm overlap and a uniform 2.5-mm bond thickness with pressure applied by weights on top of the fixture. Figure 4 shows the setup and representative snapshots of the lap-shear fabrication process using a custom-made, built-in house, bonding tool fixture that ensured proper alignment of the top overlap and a uniform bond thickness of 2.5 mm. Pressure was applied to the adhesive bond by placing a lead weight on top of the fixure (see the last two pictures at the bottom of Figure 4).

2.7. Uniaxial Tensile Test with an Instron

Tensile test experiments were performed using the lap shears fabricated with the three different adhesives (Section 2.3) to evaluate the improved adhesive bonding strength of nylon surfaces after plasma treatment only (using the identified best plasma conditions) and after silane treatment with the APS ethanol solution for the lap shears fabricated using the EPON825/JEFFAMINE D-230 adhesive. The lap-shear joints were placed under tension according to ASTM D1002 at a rate of 1.27 mm/min with an Instron (Model 1123 with a 5500 series controller and Bluehill Universal Materials Tesing software from Instron) until the specimen failed. The lap-shear strength was calculated as the maximum load divided by the bonded area, as measured from the failed specimens. The mode of failure (adhesive, cohesive, or mixed) was determined by visual inspection. Figure 5 exhibits a single-joint lap shear undergoing tensile force with the Instron equipment. Figure 5a is a snapshot of the entire setup. Figure 5b provides a closer look at the adhesive bond region of the lap shear joint, as indicated by the red box in Figure 5a.

3. Results and Discussion

In this section, we first report the results from the WCA measurements to discuss the surface wettability of the plasma-treated nylon films and the way in which we determined the best conditions for the He/O₂ and He/H₂O plasmas. Surface chemistry studies of representative plasma-treated nylon films are then presented via high-resolution XPS spectra with a special focus on changes in the key functional groups which are most relevant to important binding states of C1s in the nylon structure before and after plasma treatment. The Instron test results are discussed last to evaluate the improved bonding strength due to the enhanced surface wettability of the adhesives after plasma treatments using the best-determined conditions and to determine whether the bonding capability was further improved after silane treatment.

3.1. Water Contact Angle (WCA) Measurements

Figure 6 shows the average WCA measurements for the surface of the nylon films after He/H₂O (Figure 6a) and He/5% O₂ (Figure 6b) plasma treatments for different exposure times, as compared with that of an untreated pristine nylon surface (65.3 degrees). Our measured WCA for the original (ethanol/methanol), cleaned, untreated nylon surface was relatively smaller than the values reported in the literature [12,13]. The nylon surface wettability significantly increased after the He/H₂O plasma treatment for 180 s, as indicated by the reduction in its average WCA from 65.3 to 33.2 degrees. It appeared that a 180-s exposure time was sufficient, because the WCA remained about the same after further increasing the plasma treatment time to 8 min (480 s). On the other hand, the improvement of the surface wettability was less with the He/O₂ plasma, with the average WCA decreasing from 65.3 to ~46 degrees, even though it appeared that a shorter treatment exposure time (only 30 s) was needed. In comparison with the He/H₂O plasma treatment, our results suggested that functionalization of a nylon surface via atmospheric He/O₂ plasma may require a more high-throughput plasma source to increase the interfacial chemical reactions between nylon and the active species in the plasma glow during treatment to facilitate further post-plasma surface hydrolysis by air exposure, since moisture was lacking during plasma exposure. Gao et al. modified nylon films with 1% and 2% O₂ plasma in He, respectively, using a plasma jet with a 13.56-MHz radio-frequency power source. Those authors reported results as low as nearly 30 degrees of WCA after plasma treatment, corresponding to a significantly rough surface, as revealed by atomic force microscopy [8]. In contrast, we applied a 5% O₂ in He plasma in a DBD configuration with a kilohertz power source. Similar to the He/H₂O plasma treatment, our nylon surface wettability remained about the same, even when the plasma treatment time was doubled to 60 s or extended to 180 s, suggesting that 30 s is sufficient when using He/O₂ plasma.

Figure 7 shows the WCAs measured for 1- and 6-day-old plasma-treated nylon surfaces. The purpose was to quantify how quickly the modified surface functionality (i.e., increased hydroxyl groups) was lost (i.e., the so-called “hydrophobicity recovery”) [18,19] and how the changes correlated to the duration of the He/H₂O and He/O₂ plasma treatments, respectively. As exhibited in Figure 7, while improved wettability was sustained for all the plasma-treated nylon surfaces, the wettability decreased after one and six days from the significant improvement right after treatment (Figure 5 and Figure 6). The overall results suggest that the longer the plasma exposure, the worse the plasma-improved surface wettability over time. Such “surface aging”, or the so-called “surface recovery” phenomenon, has been reported before for nylon surfaces that were modified by other plasma conditions [17,19,20]. We also found that the increase in average WCA was most apparent for the 180-s He/H₂O plasma-treated surface. The average WCA of the resultant nylon surface increased from 36 to 46 degrees (a 28% increase) after one day and further increased by another 9% to 51 degrees six days later. Except for the 30-s He/H₂O plasma-treated surface, the wettability of all other treated surfaces worsened significantly after six days, with the 60-s He/O₂ plasma treatment showing the most pronounced effect. Taking into account the observed aging phenomenon, we decided to perform subsequent tests: suspension in an APS ethanol solution for silane treatment, as well as bonding with a thermoset epoxy (EPON825/JEFFAMINE D-230 or EPON825/JEFFAMINE D-2000, respectively) or Sikaflex 252 resin to fabricate single-joint lap shears for mechanical strength tests. The results are described in Section 3.3.

3.2. Surface Functionality via X-ray Photoemission Spectroscopy (XPS)

In general, we found that shoulders at higher energy states than the main peaks appeared for the high-resolution O1s, N1s, and C1s scans of the plasma-treated nylon surfaces. Figure 8 displays the spectra of the nylon surfaces before (Figure 8a) and after He/H₂O plasma treatment for 300 s (Figure 8b). The O1s and N1s peaks became asymmetric, with small peaks hidden at the higher binding states after the plasma treatment. The changes in the O1s and N1s spectra corresponded well to the increased intensities of the higher energy states relative to the main C peak at 285 eV (C-C) for the C1s spectrum. The estimated composition of the untreated nylon film was 79% C, 11% O, and 10% N, which is consistent with the 1-to-1 molar ratio for N to O for nylon 66—formula (C₁₂H₂₂N₂O₂)_n.

The surface composition was calculated by quantifying the XPS spectra acquired from the nylon surfaces after the He/O₂ and He/H₂O plasma treatments under different conditions. Figure 9 shows the C, O, and N composition of the nylon surface after He/O₂ plasma treatment for 180 s with 1%, 2%, and 5% O₂ content, respectively, in the plasma glow. The results are quantitatively listed in

Table 1. Composition of as-received and plasma-treated nylon surfaces, as represented by the atomic percentages of carbon (C), oxygen (O) and nitrogen (N); and their relative ratios including C/O, C/N and O/N.

Sample	C (%)	O (%)	N (%)	C/O	C/N	O/N
As-received	78.50	11.10	10.60	7.07	7.41	1.05
He/1% O2 plasma 180 s	75.90	15.60	8.50	4.87	8.93	1.84
He/2% O2 plasma 180 s	77.70	14.70	7.60	5.29	10.22	1.93
He/5% O2 plasma 180 s	75.70	15.10	9.30	5.01	8.15	1.63
He/H2O plasma 30 s	75.48	12.94	11.59	5.83	6.51	2.34
He/H2O plasma 60 s	74.26	14.15	11.60	5.25	6.40	2.92
He/H2O plasma 180 s	70.80	16.78	12.42	4.22	5.70	1.35

along with those for the He/H₂O plasma treatment. Figure 10 shows the respective C, O, and N composition for the He/5% O₂ (Figure 10a) and He/H₂O (Figure 10b) plasmas for durations of 30, 60, and 180 s. When compared with the untreated nylon surface, the decrease in the C content with consistently increased O content was comparable among all He/O₂ and He/H₂O plasma-treated samples. These changes in the elemental composition also led to comparable C/O and C/N ratios. For the O₂ plasma treatment (Figure 10a), the O/N ratios for the surfaces treated by He/5% O₂ plasma were much lower (1.63) than those treated by He/1% O₂ plasma (1.84) and He/2% H₂O plasma (1.93), suggesting either a higher level of C-O bonds than C-N bonds or a better-preserved amine-containing nylon backbone when adding the surface hydroxyl groups. For the He/H₂O plasma treatment (Figure 10b), the continuous decreases in both the C/O and C/N ratios with the increase of treatment duration mainly resulted from the gradually decreased C and increased O contents, while the N composition remained about the same. Compared to the nylon surfaces treated by He/5% O₂ plasma (Figure 10a), those treated by He/H₂O plasma resulted in a significant increase in O content and a slight increase in N content with a corresponding gradual decrease in C content when the plasma exposure time increased. These findings suggest that He/H₂O plasma was more effective than He/5% O₂ plasma (the highest concentration explored in this work) at changing the C1s chemical environment, with an increased total O content favoring increased surface wettability (Figure 6) and likely a less damaged nylon backbone containing its signature amine functional groups, leading to a much lower O/N ratio of 1.35.

To analyze the different binding states of surface carbons associated with O and N in a nylon 66 structure before and after plasma treatment, it is important to determine the C chemical environment. Figure 11 shows nylon’s chemical structure with sp³ carbons in the backbone (-C-C*-C-) at 285.0 eV, amide carbons near the C-O double bond [-N-C*-(C=O)] at ~285.6 eV, amine hydroxyl carbons (-N-C*OOH) at ~286.5 eV, and carbonyl carbons (-C=O) at ~288.1 eV. These C* carbons at different binding states are also numbered in Figure 11 to differentiate them from each other and their respective peaks in the deconvoluted high-resolution C1s spectra for the untreated sample. The He/H₂O plasma-treated nylon surfaces are exhibited in Figure 12. The two intermediate binding energies at 285.6 and 286 eV resulted from carbons covalently bonded with nitrogen (N-C*) or oxygen (O-C*), respectively, and their relative intensities with respect to the C major peak at 285 eV (-C-C*-C-) did not appear to change significantly. However, the peak responsible for the C double bond with O at 288.1 eV increased after He/H₂O plasma treatment. Considering the increase of total O content with the corresponding decrease of C content (Figure 10b), it is possible that the polymer chains were damaged by the plasma and the fresh active C atoms chemically bonded with the active O and hydroxyl species to form surface carboxylic functional groups (-COOH), as illustrated and annotated by a number (4) in Figure 11a. The bond breaking and -COOH formation were responsible for the increase of the -C=O bonding state at ~288.1 eV observed for the plasma-treated nylon surface. The chemical environment for N atoms changed only slightly before and after plasma treatment. The major peak was still the C-N peak. However, hydroxylation and oxidation in the plasma resulted in small peaks that started to appear at higher energy states. Because of the more significant changes in the C and O contents, those in the N content became less pronounced. It is difficult to distinguish the C*-N peak from the C*-O peak in a high-resolution C1s spectrum because they appear at about the same binding energy.

The different C bonding states of nylon 66 surfaces were previously reported by different groups. Louette et al. [28] deconvoluted the C1s spectrum into four different binding states including the backbone carbons at 285 eV, similar to our analyses. Pappas et al. [17] reported a similar comparison of high-resolution C1s spectra for a nylon sample before and after plasma treatment. In contrast to the present research, those authors obtained a significant increase in the peak intensity at approximately 286.5 eV and assigned this to the -C-NH-C=O functional group. It should be noted that their plasma treatment was much more robust in comparison to ours, with a plasma power of 850 W and a much higher operating frequency of 90 kHz [17].

3.3. Bond Strengths of Nylon Single-Joint Lap Shears via Uniaxial Tensile Tests

As noted in Section 2 and the discussion of plasma effectiveness illustrated previously, a 180-s He/H₂O plasma treatment was chosen when preparing the nylon surface for fabricating the nylon single-joint lap shears for Instron mechanical tests because that condition led to the best surface wettability among all the plasma-treated nylon surface samples. Figure 13 illustrates the binding strength to EPON 825 adhesive cured with JEFFAMINE D-230 for as-received, plasma-treated only, and plasma/silane-treated nylon surfaces. Multiple specimens were tested for each nylon surface condition. As anticipated for EPON 825 epoxy resin (Figure 14), the shear strength reached a maximum prior to the distinctly complete separation between the two pieces of nylon strips. The range of maximum loads increased more than double from 299–451 to 693–1594 N with corresponding increased joint extensions from 0.58–1.19 to 1.20–3.41 mm after plasma treatment. After a further silane treatment, the maximum loads were further increased by four times to 895–1857 N as compared with specimens made from the untreated nylon surfaces. Compared with the plasma-treated specimens, the more than 10% enhancement in the binding strength after silane treatment can be attributed to the additional silano bonds on the nylon surface, which provided increased interfacial interactions between the nylon surface and the epoxy adhesive (EPON 825/JEFFAMINE D-230). The modulus of EPON 825 epoxy was believed to have increased because one end of the silanols interacted with the plasma-induced hydroxyl functional groups on the nylon surface while the other end chemically reacted with the bulk epoxy structure [1]. In addition, the 1.22–3.02 mm joint extensions were comparable to the range seen without silane treatment, i.e., plasma treatment only (1.20–3.41 mm). The bond was also cohesive, as indicated by the post-testing separated joints for the lap shears (Figure 13), because the bonds were hardened and tough after cooling.

Figure 15 presents the binding strengths of lap-shear joint extensions made with (a) EPON 825/JEFFAMINE D-2000 epoxy resin and (b) Sikaflex, respectively, using as-received and plasma-treated nylon surfaces. Different from EPON 825/JEFFAMINE D-230 epoxy resin, which showed a significant benefit of the plasma and silane treatment on the enhancement of binding strength, the plasma treatment did not significantly improve the binding strength of EPON 825/JEFFAMINE D-2000 adhesive. Using JEFFAMINE D-2000 instead of D-230 to cure the same EPON 825 epoxy, the maximum loads tested were recorded as only 123–366 N, with a corresponding joint extension of 0.36–0.84 mm (pink lines in Figure 15a), which were 20% and 30% lower, respectively, than those for JEFFAMINE D-230-cured EPON 825 epoxy resin, with untreated nylon surfaces. The maximum loads improved moderately to 196–299 N with a corresponding joint extension of 0.46–0.90 mm after plasma treatment (green lines in Figure 15a). Similar to specimens made with EPON 825 epoxy cured by D-230, the comparable joint extensions before and after plasma treatment affirmed the cohesive bonding nature of epoxy resin (EPON 825) despite of the lower density of crosslinking in D-2000 due to its higher molecular weight [6,22]. Nevertheless, both JEFFAMINE D-230 and D-2000 are still popular in the adhesive community because their amine nitrogen (H₂N-) is sterically hindered in nucleophilic reactions by the pendant methyl group -CH₃ for both curing agents leading to favorable moderate interactions with the epoxy resin EPON 825. Our findings seemed to suggest that plasma and silane treatments for enhancement of surface wettability are more beneficial to adhesives with higher crosslinking [29,30].

Compared with the thermosetting epoxy adhesive EPON 825, the binding strength of untreated and plasma-treated nylon surfaces to Sikaflex, which is a ready-to-use, single-component, polyurethane-based, non-sag elastomeric sealant, showed a dramatically different behavior. Because of the high elasticity in the Sikaflex bond, the two nylon strips did not completely separate from each other, as with the EPON 825-bonded lap shears, with 1.85–2.59 mm extensions after reaching a much lower range of highest maximum loads (130–211 N) (pink lines in Figure 15b). The binding strength of these nylon lap shears was not even half of that of the EPON 825/JEFFAMINE D-230 bonds, even though the two pieces of nylon strips remained connected with more than twice the extensions. Also shown in Figure 15b is the increased maximum load of 256 N with a corresponding extension greater than 6.5 mm after plasma treatment, which was several times larger than any other joint extension value for separating the two nylon pieces in this work. This result suggested that even the best plasma treatment condition tested in this work was not able to change the generic nature of a bond adhesive for nylon 66. The correlations between the maximum load and the maximum joint extension before the failure of a lap-shear joint shown in Figure 14 and Figure 15 also underscore the fact that the improved interfacial bond strengths between the plasma-treated nylon surface and the adhesive do not influence the failure mode.

4. Conclusions

This paper discussed the results of our investigations into alterations of the surface wettability for commercial nylon 66 plastics via atmospheric pressure oxygen (He/O₂) and water vapor (He/H₂O) plasmas with a L-shape DBD custom-made reactor. The WCA results showed that the 180-s He/H₂O plasma treatment resulted in the best surface wettability, as indicated by the corresponding lowest average WCA (33 degrees) from its original 65 degrees before plasma treatment. Improved wettability was sustained after one and six days of exposure to ambient conditions with a continued lower WCA when compared with the untreated surface, although aging studies did reveal, to a certain extent, hydrophobicity recovery, during which the WCA increased slightly.

The increased oxygen content and hydroxyl surface functionality for plasma-treated nylon surfaces were confirmed by the high resolution C1s scans in the XPS spectra, which showed relatively increased peak intensities toward the higher binding state at ~288 eV. The overlapping of C*-N and C*-OH peak positions at ~287 eV in the C1s spectra makes it difficult to discern the increase in the hydroxyl peak. The relative peak intensity corresponding to the carbonyl group, C-N-(C*=O), at ~288 eV increased gradually with the increase of treatment duration for the He/H₂O plasma.

The enhanced bond strength to adhesives due to the improved surface wettability was demonstrated by uniaxial tensile tests of single-joint lap-shear tests made of nylon substrates that were treated by He/H₂O plasma for 180 s with and without further silane treatment. The results revealed the different extents of enhancement in the bond strength of nylon surfaces with a strong dependence on the molecular weight of curing agent (JEFFAMINE D-230 versus JEFFAMINE D-2000) for the same thermoset epoxy adhesive (EPON 825) and thermoelasticity (EPON 825 versus Sikaflex 252). The He/H₂O plasma-treated nylon surface was found to be the most effective for improved binding strength, more than doubling its tolerable maximum loads from 299–451 to 693–1594 N. With subsequent silane treatment, the maximum loads were found to further increase by another 20% to 895–1857 N, which was more than four times that of the untreated nylon surface. This was attributed to the silano groups that were covalently bonded with the nylon surface hydroxyl groups, allowing more effective bonds to epoxy adhesives to occur. On the other hand, with the less-ideal curing agent, JEFFAMINE D-2000, having a higher molecular weight, the same He/H₂O plasma surface treatment environment (as well as the same exposure time of 180 s) did not significantly improve the binding strength. In comparison with EPON 825, the binding strengths of the untreated nylon surfaces to the thermoelastic one-component polyurethane adhesive Sikaflex were low, i.e., 130–211 N. This value was less than half the bond strength of those with the thermoset EPON 825 adhesive. Although the He/H₂O plasma treatment led to a higher binding strength with an increased maximum load (256 N), the lap shear still failed to break cohesively, with a nearly 7-mm joint extension due to the high elasticity of the adhesive. The resultant correlation between the maximum load and joint extension before the failure of a lap-shear joint underscored the fact that the improved strength of adhesive bonds following effective plasma surface treatments does not alter the failure mode of an adhesive, which is an intrinsic bulk property.

Overall, this work provides valuable insights into feasible, moderate atmospheric plasma surface treatment methodologies and how they correlate to nylon surfaces, adhesive bond materials, and their resultant binding strengths.