Metallization of Carbon Fiber-Reinforced Plastics (CFRP): Influence of Plasma Pretreatment on Mechanical Properties and Splat Formation of Atmospheric Plasma-Sprayed Aluminum Coatings

Abstract

Carbon fiber-reinforced plastics (CFRPs) have broad applications as lightweight structural materials due to their remarkable strength-to-weight ratio. Aluminum is often used as a bond coating to ensure adhesion between CFRPs and further coatings with a higher melting temperature. However, challenges persist in optimizing their surface properties and adhesion attributes for diverse applications. This investigation explores the impact of sandblasting and plasma pretreatment on CFRP surfaces and their influence on plasma-sprayed aluminum coatings. Two distinct CFRP substrates, distinguished by their cyanate ester and epoxy resin matrices, and two different aluminum powder feedstocks were employed. Plasma pretreatment induced micro-surface roughening in the range of 0.5 µm and significantly reduced the contact angles on polished specimens. Notably, on sandblasted specimens, plasma-activated surfaces displayed improved wetting behavior, which is attributed to the removal of polymeric fragments and augmented fiber exposure. Aluminum splats show a better interaction with carbon fibers compared to a polymeric matrix material. The impact of plasma activation on the coating adhesion proved relatively limited. All samples with plasma activation had deposition efficiencies that increased by 12.5% to 34.4%. These findings were supported by SEM single-splat analysis and contribute to a deeper comprehension of surface modification strategies tailored to CFRPs.

1. Introduction

Carbon fiber-reinforced plastics (CFRPs) are currently one of the most significant lightweight structural materials used in various sectors including aerospace, automotive, marine, wind energy, sports equipment, and consumer goods, owing to their remarkable combination of high rigidity and strength coupled with low density. Nevertheless, CFRPs are not devoid of shortcomings, notably their limited resistance to low temperatures and surface wear. To surmount these challenges, the application of protective or functional coatings has gained attention. Among the methodologies used, thermal spraying stands out as a widely adopted technique for depositing such coatings. Notably, the successful application of thermally sprayed coatings on CFRPs extends to industries like printing, wherein wear-resistant and corrosion-protective coatings, as well as functional coatings, are employed on sizable CFRP rolls, employing thermal spraying technology [1]. Moreover, the utility of thermal-sprayed coatings for aerospace applications is evident in enhancing wear and corrosion resistance, along with the augmentation of thermal insulation properties. Furthermore, metallic coatings find relevance in safeguarding aircrafts through lightning protection measures.

In the context of bonding CFRPs to polymeric materials or metals by using adhesive agents, extensive research has been undertaken. Investigating atmospheric plasma treatment has revealed enhanced wetting behavior and increased surface energy. This treatment triggers an augmentation in the surface oxygen content, generating functional groups like alkoxy, carbonyl, and carboxyl groups. Particularly, carboxyl groups are correlated with heightened adhesion and shear strength [2,3]. Plasma pretreatment has also demonstrated its efficacy in enhancing the metallization of CFRPs [4,5].

Thermal spraying is a beneficial technique for improving the surface properties of CFRPs and has been used industrially for decades. Nevertheless, it is not possible to apply ceramic materials and metals with high melting points directly to the CFRPs due to substrate decomposition and a lack of coating build-up. This means that low-melting metals are essential as bond coatings for materials with elevated melting points, such as ceramics or hard metals [6]. Metals with low melting points, such as aluminum or zinc, create promising coatings and good results in terms of adhesion strength. Coating CFRPs presents a challenge due to their low thermal conductivity and resistance. The application of thermally sprayed metals onto CFRPs has been explored via methods like cold gas spray [7,8,9,10,11], wire arc spray [12,13,14], flame spray [6], HVOF [15], and APS [16,17,18].

In this publication, APS was used to apply a thin aluminum coating which acts as bond coating for additional coatings. Ceramic or hard metal coatings are able to further improve electrical or mechanical properties to enhance the properties of the composite and extend the area of application. Plasma spraying enables the application of both the bond coating and the additional coating with the same torch. A thin and homogeneous bond coating is enough to ensure the adhesion of an upcoming coating and enables the transfer of heat to the polymer. Compared to HVOF with powder or suspension feedstock, the heat transfer to the substrate can be reduced with the use of plasma spraying by around four when using data and equations from Müller and Floristán Zubieta [19,20].

Successful coatings for CFRPs necessitate low melting bond coatings that adhere effectively to both the fibers and the polymeric matrix. Various strategies aim to enhance coating adhesion. Diverse mechanical, thermal, or chemical surface treatments such as grinding, sandblasting, laser roughening, and various acids have been investigated as potential pretreatment methods [21,22,23,24,25,26]. Approaches incorporating particles (metal or sand) or fabrics into CFRP surfaces during their production have demonstrated promising outcomes, including heightened adhesion strength and improved resistance to contact corrosion [12,27]. Ganesan et al. conducted a comprehensive investigation into various pretreatment methods and their impact on the adhesion strength of copper coatings on CFRPs. Notably, their research revealed an intriguing relationship between the coating thickness and adhesion strength. A lower coating thickness was found to correlate with a higher adhesion strength due to reduced residual stresses. The study encompassed chemical, mechanical, and thermal pretreatment techniques. Specifically, chemical and thermal pretreatments were explored, resulting in a noticeable reduction in contact angle when compared to untreated or sandblasted specimens. Among these methods, chemical treatment exhibited the highest adhesion strength, closely followed by the sandblasted specimens [28]. Another approach involved laser pretreatment, which aimed to create diverse surface structures on CFRPs. This encompassed laser roughening of CFRP surfaces, selective matrix removal, and laser microstructuring. Subsequently, copper was applied via wire arc spraying, and shear adhesion strength measurements were conducted [25]. Another approach aimed at the combined utilization of laser pretreatment and wire arc-sprayed NiCr led to a substantial adhesive strength of 20.3 ± 2.7 MPa. However, it is noteworthy that not all laser-structured specimens were entirely covered by the NiCr coating, so the cohesive strength of the glue may influence the adhesion strength values, and fracture images were not given. A technique involving matrix removal, which increased the exposure of carbon fibers at the surface and facilitated complete coating coverage, yielded the highest recorded adhesion strength [26].

In the broader context, pretreatment through sandblasting emerges as a strategy that, while promoting a beneficial surface roughness for mechanical anchoring, paradoxically leads to unfavorable wetting behavior. Moreover, the reduction in matrix material appears advantageous for promoting the adhesion of thermally sprayed metals onto CFRPs. However, a comprehensive understanding of the adhesion mechanisms governing the interaction between metallic splats and CFRPs, as well as their associated influences, remains incomplete. Notably, the combination of multiple pretreatment methods has yet to be thoroughly explored. This paper aims to address these gaps through a focused analysis of individual aluminum splats adhering to polished CFRPs with varying polymeric matrices. Our investigation also delves into the effects of atmospheric plasma pretreatment on both sandblasted and polished CFRP surfaces. Furthermore, we examine the wetting behavior of pretreated CFRPs and assess the adhesion of metallic splats to CFRPs. By undertaking these inquiries, we aim to contribute to the elucidation of adhesion mechanisms and enhance the current understanding of this complex interface.

2. Materials and Methods

Two distinct types of CFRP substrates were utilized in this study. The first was composed of cyanate ester resin matrix CE662, featuring Toray C-fibers and a fabric composition [0° I 90°]3, produced by Structural Engineering GmbH & Co.KG (Pulheim, Germany). The second CFRP substrate comprised an epoxy resin matrix (Epoxid E323), sourced from Carbon-Werke GmbH (Wallerstein, Germany), with a fabric composition [0° I 90°]3. The fiber volume content for the cyanate ester resin matrix substrate was determined to be 44.0%, while the epoxy resin matrix substrate exhibited a fiber volume content of 56.2%. Aluminum was employed as the coating material, with two distinct powder feedstocks featuring different size distributions. Aluminum −45 + 20 µm (GTV Verschleißschutz GmbH, Luckenbach, Germany) and aluminum −90 + 45 µm (Metco 54NS, Oerlikon Metco, Wohlen, Switzerland) were utilized. The morphology of the powder feedstock is illustrated in Figure 1, while the powder size distribution is presented in Figure 2. The sample preparation involved the following steps: sandblasting of 150 mm × 30 mm samples using corundum F60 (pressure: 0.235 MPa, distance: 100 mm from substrate), subsequent acetone cleaning after sandblasting, plasma pretreatment (applied to half of the samples), coating application, sample cutting, and testing. The coating and plasma pretreatment employed an F6 plasma torch (GTV Verschleißschutz GmbH, Luckenbach, Germany), ensuring uniform powder feeding rates for both powder types.

The surface topography analysis comprised two methods: tactile measurements using a Mahr Perhometer (Göttingen, Germany) at five randomly selected locations (dimensions: length 5.6 mm, width 1.75 mm, five measurements), and white light interferometry (WLI) using a Bruker Contour GT device (Billerica, MA, USA). The splat examination involved preparing specimens with a surface roughness below R_a 0.08 through automated grinding and polishing. The contact angle measurements utilized an OCA20 contact angle measurement instrument (dataphysics instruments, Filderstadt, Germany) with deionized water (surface tension: 72.3; 51.0 polar; 21.8 disperse) and n-hexane (surface tension: 18.4; 0.0 polar; 18.4 disperse) as testing fluids [29,30]. The Young–Laplace Fitting method was applied, utilizing a 0.5 mm diameter nozzle. The particle size distribution (PSD) measurements were conducted using a Mastersizer 3000 (Malvern Panalytical, Malvern, UK). The scanning electron microscopy (SEM) analysis employed a NeoScope JCM-5000 (JEOL, Tokyo, Japan) for powder analysis and a JSM-6490LV (JEOL, Tokyo, Japan) for splat examination. The stereomicroscopy was performed using Nikon SMZ-10A (Tokyo, Japan). Deposition efficiency (DE) was assessed according to DIN EN ISO 17836 [31]. Adhesion strength was determined in accordance with DIN EN 14916 [32]. For adhesion strength testing, a Precision Adhesion Tester (P.A.T.) (DFD Instruments, Kristiansand, Norway) was employed, utilizing a 14.2 mm diameter test stamp and DELO AD 897 adhesive (DELO Industrie Klebstoffe GmbH & Co. KGaA, Windach, Germany). The test setup is visualized in Figure S18 (Supplementary Materials). The coated 150 mm × 30 mm specimens were cut to square 28 mm × 30 mm specimens. After hardening of the glue, the coating around the test stamp was removed using a diamante hollow drill. The actual adhesion strength of the coating was measured when the failure occurred within the coating or the coating/CFRP interface. If the failure occurred within the CFRP (cohesion failure), the adhesion strength was higher than the measured value and greater than the strength of the CFRP substrates. Five individual measurements were conducted on each sample, and the actual fractured area was calculated and corrected through image processing. Fracture analysis was performed with ImageJ grayscale analysis of the stamp surfaces. In the following sections, the materials will be referred to as follows:

• Al −45 = aluminum GTV Verschleißschutz − 45 + 20 µm;

• Al −90 = aluminum Oerlikon Metco, NS54 − 90 + 45 µm;

• CFRP-CE = carbon fiber-reinforced plastic–cyanate ester resin matrix system;

• CFRP-EP = carbon fiber-reinforced plastic–epoxy resin matrix system;

• PA = plasma-activated;

• wP = without plasma activation;

• po = polished sample;

• sb = sandblasted sample.

		PA	Al −45	Al −90
			(−45 + 20 µm)	(−90 + 45 µm)
Amperage	[A]	470	470	470
Ar/H2	[SLPM]	42/5	42/5	42/5
Travel speed	[mm/s]	500	600	600
(Spraying) distance	[mm]	80	140	140
Torch passes	[-]	1	1	1
Cooling pressurized air	[MPa]	0.6	0.6	0.6
Feeding rate	[g/min]	-	12.1	11.8
Ar flowrate feeder	[SLPM]	-	5	5.5

3. Results

3.1. Polished Specimens

The CFRP specimens underwent polishing processes, resulting in roughness values measured by tactile measurement as documented in Table 1. Little alterations are observed in the WLI measurements. Comparing the po-wP and po-PA samples, a discernible elevation in micro-surface roughness is apparent, as showcased in Figure 3 and Figure S1 (Supplementary Materials). Table 2 enumerates the changes before and after plasma pretreatment, demonstrating more pronounced differences in distance in areas between fibers and areas with resin in samples with CE matrices. The decomposition of the matrix material contributes to heightened micro-roughness, and fibers were not decomposed by the hot plasma plume. On the CFRP-EP-po samples, this increase in distance can also be seen, but not as pronounced as on the CFRP-CE-po samples.

In order to conduct a splat analysis, we introduced a mask featuring three holes with a diameter of 1 mm. The coating was applied on the mask, with single splats reaching the CFRP trough the holes. The specimens were initially examined using stereomicroscopy after coating, followed by immersion in an ultrasonic bath before stereomicroscopy and subsequent SEM analysis. In a comparison between samples with only polishing (po-wP) and polishing and plasma activation (po-PA), it is evident that po-wP samples exhibit fewer splats than PA samples sharing the same parameters. The application of ultrasonic excitation led to the near-total dislodgment of splats on the po-wP samples. This phenomenon is aptly demonstrated in Figure 4 and Figures S2–S4 (Supplementary Materials). Notably, images before and after ultrasonic examination were captured at identical positions, and this effect was consistent across all samples. Intriguingly, splats appear to adhere more firmly to CFRP-po-PA surfaces. On PA samples, plasma activation instigates micro-roughening, yielding a modest elevation in roughness (0.2 µm for EP, 0.8 µm for CE). Given that the dominant adhesion mechanism revolves around mechanical anchoring, this marginal increase in micro-roughness is projected to enhance the adhesion of individual splats on polished surfaces. Inspecting the SEM analysis illustrated in Figure 5 and Figures S5–S7 (Supplementary Materials), it becomes apparent that splats display different adhesion and wetting characteristics on carbon fibers when compared with the polymeric matrix. Aluminum splats tend to bridge adjacent carbon fibers, while the polymeric surface remains incompletely covered by the splats. Aluminum splats favorably wet carbon fibers, orient in the fibers’ direction, and solidify. Areas with resin are less wet on average, and splats are not adhering well on resin-dominant areas, as visualized in Figure 6. The wP-samples exhibit a more pronounced tendency toward splat splashing, as small splat residues are visible. An increased carbon fiber content on the surface of PA samples could potentially heighten thermal conductivity, thereby facilitating enhanced splat solidification and mitigating splashing tendencies. Furthermore, the micro-roughening effect could contribute to augmented splat anchoring. The stereomicroscopic images corroborate these observations, with splats on samples that lack plasma pretreatment (wP) experiencing adhesion loss post-ultrasonic excitation.

3.2. Roughened Specimens

3.2.1. Surface Topography

This investigation focused on sandblasted (sb-wP) and sandblasted and additionally plasma-activated specimens (sb-PA). Upon comparing the surface structures using SEM (Figure 7), both surfaces exhibit fractured carbon fibers. The plasma-activated (PA) specimens display more defined edges with fewer fragmented polymeric particles (indicated in Figure 7, left). On a macroscopic level, the PA surface appears more reflective, featuring an increased prominence of carbon fibers compared to the polymeric matrix. Plasma pretreatment seems to remove or decompose polymeric fine dust particles and less-adhering fibers from the surface. This effect is more noticeable in CE specimens, whereas EP specimens exhibit a milder effect due to a higher matrix presence on the surface. The WLI measurements (Figures S8 and S9, Supplementary Materials) and surface topography measurements (Figure S10, Supplementary Materials) are taken from the same specimen spot and reveal no discernible change in the macroscopic surface topography as a result of plasma activation.

3.2.2. Contact Angle

Comparing the contact angle measurements with water as a dispersant, as visualized in Figure 8 (left), the PA samples exhibit a reduced contact angle in comparison to the wP samples. The change in contact angle for CFRP-CE is more pronounced than on the CFRP-EP samples. However, there is no observable change when a non-polar liquid is used, as shown in Figure 8 (right).

3.2.3. Microstructure

The cross-section in Figure 9 shows that homogeneous and thin aluminum coatings were successfully applied via APS using both types of powder feedstocks. The coatings appear dense, with some open pores and almost no oxidation. A higher magnification of the coating’s cross-section is shown in Figure S19 (Supplementary Materials). Oxidation at the contact area between the C-fibers and aluminum occurs due to the use of a water-based diamond suspension during preparation. The coating thickness was measured as 28.0 ± 11.8 µm for the CFRP-CE sample with fine powder feedstock (−45) and 23.2 ± 9.8 µm for the coarse powder feedstock. The SEM surface analysis shown in Figure 10 reveals molten and solidified splats on the surface. With a coarse powder feedstock (−90), the splats appear better molten, while the surface from finer powder shows a rougher texture with less pronounced splats, suggesting a lower temperature or speed during particle impact.

3.2.4. Deposition Efficiency (DE)

The deposition efficiency is visualized in Figure 11, ranging from 30.0% to 48.3%. For CFRP-CE and CFRP-EP, the DE values are similar when comparing specimens produced with the same powder feedstocks. Coatings formed from finer powder feedstocks generally exhibit higher DEs compared to those from coarser feedstocks. Plasma activation leads to an increase in DE ranging from 12.5% to 34.4% for the initial aluminum pass. This effect is more pronounced when using a coarser powder feedstock (−90 +45 µm). Overall, the wP-samples display noticeably lower DEs than the PA samples do. Subsequent passes of aluminum on CFRP result in increased DEs due to the superior adhesion of aluminum to previously solidified aluminum rather than to CFRP. Representative samples include CFRP-CE-sb-PA Al −45 (first pass (45.4%), second pass (52.4%), third pass (50.8%)) and CFRP-CE-sb-PA Al −90 (first pass (44.4%), second pass (62.2%), third pass (67.0%)), as depicted in Figure 12.

3.2.5. Adhesion Strength

The adhesion strength of plasma-sprayed aluminum on CFRP is depicted in Figure 13 and ranges between 3.5 and 5.4 MPa. Comparatively higher values are attainable when coating CFRP-EP substrates, as opposed to CFRP-CE. Coatings using a coarser powder feedstock exhibit slightly better adhesion compared to finer powder fractions. Contrary to expectations, plasma pretreatment does not notably enhance adhesion strength.

Fracture analyses on the tested specimens reveal more cohesive fiber damage in samples with plasma pretreatment (PA samples), as shown in Figure 14. For CE samples, cohesion failure predominantly occurs within the CFRP, with no discernible differences between samples with or without plasma pretreatment. In Al −90 samples, more cohesive failure is observed, while Al −45 samples show more adhesive failure. In comparison, the EP samples display a higher occurrence of adhesion failure between the aluminum and CFRP. The EP samples without plasma pretreatment exhibit slightly more cohesion failure within the composite compared to samples with plasma activation. Fractures are presented in Tables S1 and S2 (Supplementary Materials). The strength of CFRP alone in the tensile direction was measured to be as low as 4.7 ± 0.4 MPa for CFRP-CE and 5.7 ± 0.8 MPa for CFRP-EP. These values were obtained through adhesion strength tests without removing the fibers around the test stamp. Overall, the measurements indicate that the adhesion strength of the aluminum coatings on CFRPs exceeds the strength of the substrate. However, the true adhesion strength cannot be measured due to the substrate’s limited strength, as cohesion failure occurs within the CFRP. Samples with higher strengths (CFRP-EP) exhibit less cohesive failure within the CFRP than CFRP-CE, which has a lower strength. Figures S11 and S12 (Supplementary Materials) show a stereomicroscopic analysis of single-splat experiments, revealing no significant loss of splats even after ultrasonic excitation. There is no observed difference in splat loss between wP and PA samples, as seen in the polished samples, suggesting higher adhesion of splats through mechanical clamping on the roughened surface. A single-splat examination is shown in Figure 15 and Figures S13–S16 (Supplementary Materials). All splats mainly anchor in the valleys of the surface. No differences in splat geometry or splashing behavior are observed when comparing CFRP-EP and CFRP-CE samples. On wP-samples, a higher content of polymeric material and fragments are present at the surface compared to PA samples. More voids can be observed within the splats of wP-samples, especially in Figure S17 (Supplementary Materials), indicating a lower wetting at specific areas of the CFRP surface. Overall, the difference between the splats of wP and PA samples is less pronounced than on polished samples.

4. Discussion

4.1. Plasma Pretreatment Effects

Plasma pretreatment had discernible effects on the surface of CFRP specimens. This treatment effectively removes or decomposes small polymeric fragments and loosely adhering broken fibers resulting from sandblasting. Consequently, an increased fiber presence was observed on the surface. Additionally, the contact angle decreased in comparison to specimens that were solely sandblasted. Especially for the plasma-pretreated polished specimens, three notable outcomes emerged: First, there was an increase in fiber presence at the surface. Second, micro-roughening occurred trough polymeric matrix decomposition by the hot atmospheric plasma in the range of 0.5 µm, visualized in Figure 3. After plasma treatment, the surface roughness was slightly higher, represented by the increase in distance between the matrix and fibers. The C-fibers have a high thermal conductivity and are thus conducting the heat from the hot plasma in the fiber direction, leading to a temperature regulation. The polymeric matrix, on the other hand, has both a lower thermal conductivity and lower decomposition temperature. Thus, the increase in roughness must come from evaporated polymeric matrices by the hot plasma plume. A micro-roughening of the surface, as presented here, has not yet been observed in the literature. Third, we observed an improvement in wetting behavior. The improvement in wetting behavior following different plasma pretreatments is also shown in various publications and linked to an increase in COO groups on the surface. Ganesan et al. proved an increase in COO groups by XPS [28]. In the field of electroless plating, a plasma treatment leads to increased surface energy and thus increased adhesion [5]. In joint bonds, a decrease in the surface angle by plasma activation is recognized, also leading to an increased adhesion strength of the bonds [2].

4.2. Single-Splat Deposition

The examination of single splats on polished specimens highlighted improved adhesion on plasma-activated samples. The number of splats on the plasma-activated polished samples (po-PA) remained higher than on the polished samples without plasma activation (po-wP) after ultrasonic excitation, so the adhesion of these splats seems to be higher. Remarkably, aluminum exhibited superior solidification on carbon fibers, leading to finger-like structures that formed bridges connecting adjacent carbon fibers. This distinctive behavior suggests enhanced wetting of aluminum on carbon compared to its wetting on the polymeric matrix. Notably, due to the low thermal conductivity of the polymeric matrix, the prolonged solidification of splats can lead to splashing and decreased or lost adhesion. Splashed splats which have already lost the adhesion to CFRP are shown in Figure 5a,b. Overall, increased wetting behavior, a higher carbon fiber content, and an increase in micro-roughness on the surface are changes that follow plasma activation and may lead to increased adhesion of single splats on polished samples. On sandblasted but not plasma-activated specimens (sb-wP), this effect was less pronounced than on the polished samples (po-wP). On the sb-wP-samples, polymeric residues and areas where the polymeric matrix is present at the surface were clearly visible. Plasma activation reduces the amount of polymeric material on the surface, promoting faster solidification due to the heightened thermal conductivity.

4.3. Adhesion Mechanisms

For polished specimens, plasma activation resulted in micro-roughening, wetting improvement, and residue removal, leading to better adhesion of metallic particles on CFRPs. For sandblasted specimens, plasma activation effectively removed residues and improved wetting. Micro-roughening could not be confirmed by the SEM images. Overall, plasma activation did not enhance the coating’s adhesion properties on the sandblasted specimens. Interestingly, coatings formulated with larger-particle-size powder feedstocks exhibited slightly improved adhesion. Larger splats likely interacted more with carbon fibers than with the polymeric matrix, forming extensive contact areas that are conducive to robust adhesion. This preference for aluminum–C-fiber interaction over aluminum–polymeric matrix interaction is evident in Figure 6. Notably, the predominant adhesion mechanism remains mechanical anchoring, which is provided by using sandblasting as the pretreatment method. Figures S11 and S12 (Supplementary Materials) show single splats before and after ultrasonic excitation with nearly the same splats present, as plasma pretreatment does not lead to any change in the adhesion of splats on the sandblasted specimens. On both samples, splats do adhere.

The adhesion strength measurements of the plasma-sprayed aluminum samples reached a maximum of 5.4 MPa. This represents the highest value in the tests. The measurements indicate that the adhesion strength of the coatings exceed the strength of the CFRP substrate itself when a high number of cohesion failures appears. Therefore, the true adhesion strength cannot be measured, as the CFRP substrate fails before the coating does. In comparison to the literature, cold-sprayed aluminum coatings reach an adhesion strength of 2.26 MPa, while flame-sprayed aluminum achieves 6.0 MPa on Polyimide CFRPs [6,33]. No comparable data for plasma-sprayed aluminum are available. Plasma-sprayed cooper coatings achieve an adhesion strength of ~3.0 MPa [27].

4.4. Deposition Efficiency Observations

A significant increase in DE was observed on all sandblasted and plasma-activated specimens (sb-PA) compared to non-plasma-activated specimens (sb-wP). Overall, the DE can be increased by 12.5% to 34.4% due to plasma activation. The increase in DE is likely due to the faster solidification of splats, attributed to the higher amount of carbon fibers on the surface compared to a sandblasted surface, where a polymeric matrix and polymeric fragments are still present. Better wetting behavior and higher thermal conductivity on the surface prevent single splats from splashing and losing their adhesion on PA samples. This results in a higher DE on all samples with PA. Notably, finer aluminum powders exhibited higher DE values, which we posit might be due to faster solidification of smaller particles, thereby reducing splat splashing and leading to an increasing DE. This potentially indicates the presence of partially solidified small particles adhering to the surface, albeit not strongly. This phenomenon seems to be linked to a lack of change in adhesion but a higher DE. Compared to other thermal spray techniques, the DE of aluminum on CFRPs is quite high, although comparable data for thermally sprayed coatings are rare. For instance, cold-sprayed copper samples achieve a DE of only ~8% or 12% [10,34], while cold-sprayed Sn reached a DE of ~20% [35]. Only warm-sprayed Titanium coatings achieve a higher DE at ~60% [36]. DE data for thermally sprayed aluminum from are not available from the literature [37].

5. Conclusions

In summary, this study has explored the adhesion characteristics of aluminum splats on different surfaces, showing the effects of plasma pretreatment and the powder feedstock size of aluminum coatings on two different CFRPs. Several key findings emerge from our investigation:

• -. Aluminum splats exhibit superior adhesion to carbon fibers compared to polymeric matrices, indicating a potential avenue for enhanced material bonding.

• -. Plasma pretreatment effectively reduces polymeric residue on surfaces, resulting in improved wettability and contact angle reduction. The process also leads to a strong increase in deposition efficiency (up to 48.8%), making it a promising technique for surface modification prior to coating.

• -. Micro-roughening induced by plasma pretreatment contributes to enhanced adhesion of single metallic splats on polished samples. This finding holds significance for specialized applications demanding a specific surface topography and minimal roughness, such as galvanic and paint applications.

• -. Plasma activation shows limited influence on the adhesion of thermally sprayed metallic particles on sandblasted CFRPs. Mechanical anchoring on rough surfaces, particularly through sandblasting, plays a predominant role in achieving adhesion.

• -. Although a coarse powder feedstock tends to slightly improve adhesion compared to a finer powder feedstock, coatings produced from finer powder feedstocks exhibit higher DEs.

• -. Due to significant cohesion failure within the CFRP during adhesion strength measurements, the true adhesion strength of the coatings must be higher than 5.4 MPa.

• -. Plasma activation enhances the DEs of coatings on sandblasted specimens. Furthermore, the DE of the first layer of aluminum on CFRP is lower compared to upcoming passes, where the particles are solidifying on already solid aluminum particles.

Author Contributions

Data curation, C.S.; Formal analysis, C.S.; Funding acquisition, A.K.; Investigation, C.S. and W.S.; Methodology, C.S.; Supervision, A.K.; Validation, C.S.; Visualization, C.S.; Writing—original draft, C.S.; Writing—review and editing, A.K. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Research data will be made available on request.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. (a) SEM image of aluminum powder Al −45; (b) SEM image of aluminum powder Al −90.

Figure 2. (a) PSD of aluminum powder Al −45; (b) PSD of aluminum powder Al −90.

Figure 3. (a) Surface topography of CFRP-CE-po-wP; (b) surface topography of CFRP-CE-po-PA.

Figure 4. Top surface of polished CFRP-CE with splats: (a) Al −90 CFRP-CE-po-wP; (b) Al −90 CFRP-CE-po-wP—after ultrasonic excitation; (c) Al −90 CFRP-CE-po-PA; (d) Al −90 CFRP-CE-po-PA—after ultrasonic excitation.

Figure 5. Top surface of polished CFRP with splats: (a) Al −45 of CFRP-CE-po-wP; (b) Al −45 of CFRP-CE-po-PA; (c) Al −90 of CFRP-CE-po-wP; (d) Al −90 of CFRP-CE-po-PA.

Figure 6. (a) SEM BSE image of aluminum splats preferentially adhering to C-fibers. (b) Stereomicroscopic image of splat investigations.

Figure 7. Surface topography of CFRP-CE-sb-wP (a); surface topography of CFRP-CE-sb-PA (b).

Figure 8. Contact angle measurements—dispersant water and n-hexane.

Figure 9. SEM cross-section analysis: (a) CFRP-CE-sb-PA-Al −45; (b) CFRP-CE-sb-PA-Al −90.

Figure 10. SEM surface analysis: (a) CFRP-sb-PA-Al −45; (b) CFRP-sb-PA-Al −90.

Figure 11. Deposition efficiency of aluminum coatings on CFRP.

Figure 12. DEs of further passes of aluminum applied onto CFRP-CE-sb-PA with fine (−45) and coarse (−90) powder feedstock.

Figure 13. Adhesion strength of CFRP substrates and coatings on sandblasted (sb-wP) and sandblasted and plasma-activated (sb-PA) CFRPs.

Figure 14. Results of fracture analysis after adhesion strength test in accordance with DIN EN ISO 14916.

Figure 15. (a) Top surface of CFRP with splat, Al −90 of CFRP-CE-sb-wP; (b) top surface of CFRP with splat, Al −90 of CFRP-CE-sb-PA.

Table 1

Surface roughness values Ra and Rz of polished specimens.

Specimen	po-wP
	Ra [µm]	Rz [µm]
CFRP-CE	0.06 ± 0.00	0.29 ± 0.02
CFRP-EP	0.08 ± 0.01	0.42 ± 0.07

Table 2

Change in surface of polished specimens following plasma activation.

Specimen	po-wP(Polished)	po-PA(Polished + Plasma Activation)	Difference [µm]
	Distance Fiber–Resin [µm]	Distance Fiber–Resin [µm]
CFRP-CE	0.046	0.825	0.779
CFRP-EP	0.278	0.478	0.200

Supplementary Materials

The supporting information can be downloaded at https://www.mdpi.com/article/10.3390/coatings14091169/s1.

References

1. Menningen, M.; Weiss, H.; Fischer, U. Metallic wear resistant coatings for carbon fibre epoxy composite rolls. Surf. Coat. Technol.; 1995; 71, pp. 208-214. [DOI: https://dx.doi.org/10.1016/0257-8972(94)01024-D]

2. Zaldivar, R.J.; Nokes, J.; Steckel, G.L.; Kim, H.I.; Morgan, B.A. The Effect of Atmospheric Plasma Treatment on the Chemistry, Morphology and Resultant Bonding Behavior of a Pan-Based Carbon Fiber-Reinforced Epoxy Composite. J. Compos. Mater.; 2010; 44, 137. [DOI: https://dx.doi.org/10.1177/0021998309345343]

3. Zaldivar, R.J.; Kim, H.I.; Steckel, G.L.; Nokes, J. Effect of Processing Parameter Changes on the Adhesion of Plasma-treated Carbon Fiber Reinforced Epoxy Composites. J. Compos. Mater.; 2010; 44, pp. 1435-1453. [DOI: https://dx.doi.org/10.1177/0021998309355846]

4. Prysiazhnyi, V.; Stupavská, M.; Ráheľ, J.; Kleber, C.; Černák, M.; Rafailović, L.D. A comparison of chemical and atmospheric plasma assisted copper plating on carbon fiber reinforced epoxy polymer surfaces. Surf. Coat. Technol.; 2014; 258, pp. 1082-1089. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2014.07.026]

5. Rafailovi, L.D.; Stupavska, M.; Prysiazhnyi, V.; Polt, G.; Rohr, O.; Nixon, S.; Rahel, J. Impact of atmospheric pressure plasma treatment on surface metallization of CFRP composites. Surf. Coat. Technol.; 2021; 412, 127046. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2021.127046]

6. Abedi, H.R.; Salehi, M.; Shafyei, A. Microstructure, Tensile Adhesion Strength and Thermal Shock Resistance of TBCs with Different Flame-Sprayed Bond Coat Materials onto BMI Polyimide Matrix Composite. J. Therm. Spray Technol.; 2017; 26, pp. 1669-1684. [DOI: https://dx.doi.org/10.1007/s11666-017-0607-9]

7. Ganesan, A.; Affi, J.; Yamada, M.; Fukumoto, M. Bonding behavior studies of cold sprayed cooper coatings on the PVC polymer substrate. Surf. Coat. Technol.; 2012; 207, pp. 262-269. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2012.06.086]

8. Chen, C.; Xie, X.; Xie, Y.; Yan, X.; Huang, C.; Deng, S.; Liao, H. Metallization of polyether ether ketone (PEEK) by copper coating via cold spray. Surf. Coat. Technol.; 2018; 342, pp. 209-219. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2018.02.087]

9. Lupoi, R.; O’Neill, W. Deposition of metallic coatings on polymer surfaces using cold spray. Surf. Coat. Technol.; 2010; 205, pp. 2167-2173. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2010.08.128]

10. Che, H.; Vo, P.; Yue, S. Metallization of carbon fibre reinforced polymers by cold spray. Surf. Coat. Technol.; 2017; 313, pp. 236-247. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2017.01.083]

11. Gabor, T.; Akin, S.; Jun, M.B.-G. Numerical studies on cold spray gas dynamics and powder flow in circular and rectangular nozzles. J. Manuf. Process.; 2024; 114, pp. 232-246. [DOI: https://dx.doi.org/10.1016/j.jmapro.2024.02.005]

12. Rezzoug, A.; Abdi, S.; Kaci, A.; Yandouzi, M. Thermal spray metallisation of carbon fibre reinforced polymer composites: Effect of top surface modification on coating adhesion and mechanical properties. Surf. Coat. Technol.; 2018; 333, pp. 13-23. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2017.10.066]

13. Devaraj, S.; Anand, B.; Gibbons, M.; McDonald, A.; Chandra, S. Thermal spray deposition of aluminum and zinc coatings on thermoplastics. Surf. Coat. Technol.; 2020; 399, 126114. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2020.126114]

14. Wang, R.; Song, D.; Liu, W.; He, X. Effect of arc spraying power on the microstructure and mechanical properties of Zn–Al coating deposited onto carbon fiber reinforced epoxy composites. Appl. Surf. Sci.; 2010; 257, pp. 203-209. [DOI: https://dx.doi.org/10.1016/j.apsusc.2010.06.065]

15. Ivosevic, M.; Knight, R.; Kalidindi, S.R.; Palmese, G.R.; Sutter, J.K. Adhesive/cohesive properties of thermally sprayed functionally graded coatings for polymer matrix composites. J. Therm. Spray Technol.; 2005; 14, pp. 45-51. [DOI: https://dx.doi.org/10.1361/10599630522765]

16. Sun, G.; He, X.; Jiang, J.; Sun, Y.; Zhong, Y. A Study on the Deposition of Al₂O₃ Coatings on Polymer Substrates by a Plasma Spray/Micro-Arc Oxidation Two-Step Method. J. Therm. Spray Technol.; 2013; 22, pp. 27-35. [DOI: https://dx.doi.org/10.1007/s11666-012-9865-8]

17. Sun, G.; He, X.; Jiang, J.; Sun, Y. Parametric study of Al and Al₂O₃ ceramic coatings deposited by air plasma spray onto polymer substrate. Appl. Surf. Sci.; 2011; 257, pp. 7864-7870. [DOI: https://dx.doi.org/10.1016/j.apsusc.2011.04.057]

18. Melentiev, R.; Yudhanto, A.; Tao, R.; Vuchkov, T.; Lubineau, G. Metallization of polymers and composites: State-of-the-art approaches. Mater. Des.; 2022; 221, 110958. [DOI: https://dx.doi.org/10.1016/j.matdes.2022.110958]

19. Müller, P. Prozessdiagnostik und Beschichtungstechnik beim Plasma- und Hochgeschwindigkeitsflammspritzen von Keramischen Suspensionen; Shaker Verlag GmbH: Dylan, Germany, 2016; ISBN 978-3-8440-4585-7

20. Floristán Zubieta, M. Off-Line Generation of Robot Trajectories for Thermal Spray Processes with Respect to the Heat and Mass Transfer: Off-Line Generierung von Keramischer Bauteile (IFKB); Shaker Verlag GmbH: Dylan, Germany, 2013; ISBN 978-3-8440-1881-3

21. Kodokian, G.K.A.; Kinloch, A.J. Surface pretreatment and adhesion of thermoplastic fibre-composites. J. Mater. Sci. Lett.; 1988; 7, pp. 625-627. [DOI: https://dx.doi.org/10.1007/BF01730315]

22. Davies, P.; Courty, C.; Xanathopoulos, N.; Mathieu, H.-J. Surface treatment for adhesive bonding of carbon fibre-poly(etherether ketone) composites. J. Mater. Sci. Lett.; 1991; 10, pp. 335-338. [DOI: https://dx.doi.org/10.1007/BF00719701]

23. Fischer, F.; Kreling, S.; Jäschke, P.; Frauenhofer, M.; Kracht, D.; Dilger, K. Laser Surface Pre-Treatment of CFRP for Adhesive Bonding in Consideration of the Absorption Behaviour. J. Adhes.; 2012; 88, pp. 350-363. [DOI: https://dx.doi.org/10.1080/00218464.2012.660042]

24. Gebauer, J.; Burkhardt, M.; Franke, V.; Lasagni, A.-F. On the Ablation Behavior of Car-bon Fiber-Reinforced Plastics during Laser Surface Treatment Using Pulsed Lasers. Materials; 2020; 13, 5682. [DOI: https://dx.doi.org/10.3390/ma13245682] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33322785]

25. Gebauer, J.; Franke, V.; Gustke, K.; Lampke, T.; Lasagni, A.F. Improving the bond strength of metal-FRP-hybrids with thermal sprayed copper using pulsed laser-based processing approaches. Proceedings of the Laser in Manufacturing Conference; Virtual, 21–24 June 2021.

26. Gebauer, J.; Klotzbach, U.; Lasagni, A.-F. Functionalization of fiber-reinforced plastic based on laser micro Structuring. Proceedings of the Laser-Based Micro- and Nanoprocessing XIII; San Francisco, CA, USA, 5–7 February 2019; Volume 10906, [DOI: https://dx.doi.org/10.1117/12.2505114]

27. Lopera-Valle, A.; McDonald, A. Application of flame-sprayed coatings as heating elements for polymer-based composite structures. J. Therm. Spray Technol.; 2015; 24, pp. 1289-1301. [DOI: https://dx.doi.org/10.1007/s11666-015-0302-7]

28. Ganesan, A.; Yamada, M.; Fukumoto, M. The Effect of CFRP Surface Treatment on the Splat Morphology and Coating Adhesion Strength. J. Therm. Spray Technol.; 2014; 23, pp. 236-244. [DOI: https://dx.doi.org/10.1007/s11666-013-0003-z]

29. Rulison, C. Two-Component Surface Energy Characterization as a Predictor of Wettability and Dispersibility; Application Note #213 Krüss Gmbh: Hamburg, Germany, 2000.

30. Kaelble, D.H. Dispersion-Polar Surface Tension Properties of Organic Solids. J. Adhes.; 1970; 2, pp. 66-81. [DOI: https://dx.doi.org/10.1080/0021846708544582]

31. DIN EN ISO 17836:2018-03 Thermisches Spritzen—Bestimmung der Auftragrate beim thermischen Spritzen. ISO: Geneva, Switzerland, 2018; [DOI: https://dx.doi.org/10.31030/2723759]

32. DIN EN ISO 14916:2017-08 Thermisches Spritzen—Ermittlung der Haftzugfestigkeit. ISO: Geneva, Switzerland, 2017; [DOI: https://dx.doi.org/10.31030/2588875]

33. Zhou, X.L.; Chen, A.F.; Liu, J.C.; Wu, X.K.; Zhang, J.S. Preparation of metallic coatings on polymer matrix composites by cold spray. Surf. Coat. Technol.; 2011; 206, pp. 132-136. [DOI: https://dx.doi.org/10.1016/j.surfcoat.2011.07.005]

34. Akin, S.; Nath, C.; Jun, M.B.-G. Selective Surface Metallization of 3D-Printed Polymers by Cold-Spray-Assisted Electroless Deposition. ACS Appl. Electron. Mater.; 2023; 5, pp. 5164-5175. [DOI: https://dx.doi.org/10.1021/acsaelm.3c00893]

35. Fukumoto, M.; Ganesan, A.; Yamada, M.; Tsunekawa, Y. Thick Metallic/Composite Coating on C-FRP Substrate by Plasma Spraying. Proceedings of the 21st International Symposium on Plasma Chemistry (ISPC 21); Cairns, QLD, Australia, 4–9 August 2013.

36. Ganesan, A.; Takuma, O.; Yamada, M.; Fukumoto, M. Microstructure and Mechanical Properties of Warm-Sprayed Titanium Coating on Carbon Fiber-Reinforced Plastic. J. Therm. Spray Technol.; 2016; 25, pp. 788-796. [DOI: https://dx.doi.org/10.1007/s11666-016-0392-x]

37. Affi, J.; Okazaki, H.; Yamada, M.; Fukumoto, M. Fabrication of Aluminum Coating onto CFRP Substrate by Cold Spray. Mater. Trans.; 2011; 52, pp. 1759-1763. [DOI: https://dx.doi.org/10.2320/matertrans.T-M2011807]