Plasma-Based Amorphous Carbon Coatings on Polymeric Substrates for Biomedical Applications: A Critical Review Focused on Adhesion

Abstract

Material surfaces are of primary importance in biomaterial development, significantly influencing implant lifespan and clinical success. Consequently, coating technologies are frequently employed to modify surface properties and functionality. Plasma-based amorphous carbon coatings have been widely applied to all classes of substrates to improve their tribology, corrosion resistance, hardness, and even biological properties. Plasma technology is widely recognized to be effective, not only for the deposition of amorphous carbon coatings but also for substrate pre-treatment, in which it may play a key role in activating surfaces and enhancing interfacial adhesion. Amorphous carbon coatings can be classified into two major categories: diamond-like carbon (DLC) and polymer-like carbon (PLC), according to their mechanical properties. Regardless of their nature, the adhesion of both types of amorphous carbon coatings to the substrate has always represented a major challenge. Several strategies have been reported to enhance the adhesion of DLC coatings to silicon wafers, metals, and glass substrates. However, few studies report strategies aimed at controlling the adhesion of (both types of) amorphous carbon coatings to polymeric substrates, polymeric implants, and polymeric devices. Therefore, this work aims to provide a state-of-the-art review on the adhesion of amorphous carbon coatings to polymeric substrates for biomedical applications. Furthermore, this review presents the main techniques used to assess adhesion and the strategies available to improve adhesion between coatings and polymeric substrates.

1. Introduction

Surfaces play a pivotal role in ensuring the performance of biomaterials as they mediate the interactions between the material and the biological environment. The physicochemical properties of surfaces, including roughness, wettability, chemical composition, and surface energy, are known to directly influence cellular adhesion, protein adsorption, and overall biocompatibility [1,2,3].

Coatings have emerged as a versatile solution to enhance biomaterial surface properties, offering tailored functionalities without compromising the bulk characteristics of the substrate [1,4]. Amorphous carbon coatings have been extensively studied for biomedical applications, primarily due to their outstanding mechanical properties [5]. Specifically, hard amorphous carbon coatings, commonly known as diamond-like carbon (DLC), continue to attract increasing interest due to their unique properties, such as hardness and wear resistance, which can significantly enhance the performance and lifespan of implants [6,7,8,9,10,11].

Recent reports show that DLC coatings have been explored as platforms for releasing antibacterial agents [12,13,14], providing new possibilities for the development of antimicrobial and antiviral surfaces. This is of particular pertinence for medical devices, implants, and environments in which reducing bacterial/viral colonization and biofilm formation is a critical requirement. Despite these advancements, the adhesion of DLC coatings has always been reported as a major challenge in ensuring the performance and durability of coated medical devices. Several studies have extensive reports on strategies to enhance the adhesion, wear resistance, and durability of DLC coatings deposited on silicon wafers [15,16,17] and metallic substrates, including stainless steel [18,19,20,21,22,23] and titanium alloys [19,24,25,26].

Despite their widespread use as biomaterials, the adhesion of DLC to polymeric substrates has not been extensively studied. Therefore, the adhesion performance and stability over time of DLC coatings deposited on polymers remain largely unexplored and underreported.

Using the keywords ‘DLC’, ‘adhesion’, and ‘polymeric substrate’ as inclusion criteria yielded only 90 articles, in contrast to the 350 articles retrieved for ‘DLC’, ‘adhesion’, and ‘metallic substrate’ (Web of Science database, 2025). As shown in Figure 1, polymers (either soft or hard) present a different set of challenges for the deposition of DLC, primarily due to their lower surface energy, higher thermal expansion coefficients, and lower mechanical strength compared to metals. These differences lead to poor adhesion, the cracking and delamination of the coating under stress or deformation, compromising the integrity and performance of the coated polymer substrates.

To address this gap, soft amorphous carbon coatings, known as polymer-like carbon (PLC), may be considered high-potential candidates for application on polymeric substrates. PLC shares some of the favorable properties of DLC, such as chemical resistance and biocompatibility [10,27], but its structure can be tailored to match (at the interfacial level) the mechanical and thermal properties of the polymeric substrate onto which it is deposited. This compatibility should lead to improved adhesion and stability on polymeric surfaces by mitigating stresses arising from mismatches in thermal expansion and mechanical flexibility [28]. Despite this, there are a limited number of studies reported in the literature that address this topic. In this scenario, the adhesion performance of amorphous carbon coatings on polymers, which is essential for guaranteeing efficacy and extending implant life, needs to be better understood.

This review focuses on the critical issue of assessing the adhesion of amorphous carbon coatings on polymers used in medical devices and implants. It addresses the assessment of adhesion performance, the merging and matching of properties between the coating and polymer substrate, and the main strategies available to enhance adhesion, detailing their use and the rationale behind their selection. Among these strategies, plasma technology stands out as a particularly versatile approach, not only for the deposition of various types of amorphous carbon coatings but also for substrate pre-treatment. In this context, plasma processes can play a decisive role in modifying surface chemistry and topography, introducing functional groups, and removing contaminants, all of which are crucial for activating polymer surfaces and promoting strong interfacial bonding with the coating. Its versatility, cleanliness, and ability to operate at relatively low temperatures make it particularly suitable for treating thermally sensitive polymeric substrates used in biomedical applications [29,30].

This review aims to provide an overview of the deposition and classification of amorphous carbon materials. Then, there is a section dedicated to state-of-the-art research on the adhesion of amorphous carbon coatings on polymeric medical devices, detailing the primary techniques used to evaluate adhesion and summarizing key findings. Finally, the critical factors influencing the adhesion of these materials are thoroughly reported and discussed.

2. Amorphous Carbon Coatings

2.1. Deposition Methods

Several amorphous carbon coatings have been deposited using different techniques based on physical vapor deposition (PVD) and chemical vapor deposition (CVD) processes [7]. Physical vapor deposition encompasses a variety of methods to deposit thin films by the condensation of a vaporized form of the material, involving purely physical processes. Some examples include ion beam deposition, pulsed laser deposition, filtered cathodic vacuum arc (FCVA), high-power impulse magnetron sputtering (HiPIMS), and laser ablation [7,11]. When a volatile compound of the material is vaporized and then breaks down or reacts on the substrate surface to form a solid coating, the process is called chemical vapor deposition (CVD). A relatively high temperature is required for CVD processes, in some cases, near 1000 °C [7,11].

Plasma-enhanced or plasma-assisted chemical vapor deposition (PECVD or PACVD) is a modified version of conventional CVD processes and one of the most common and efficient methods for depositing amorphous carbon coatings on polymers. This technique allows the deposition of coatings at much lower temperatures, which is a crucial factor in preventing the thermal degradation of polymeric substrates [7,29,31]. In the deposition of DLC by PECVD, a precursor gas (typically hydrocarbons like methane, acetylene, or a mix of methane/hydrogen) is introduced into a vacuum chamber. In general, the precursor gas is activated into a plasma state, during which it dissociates into reactive ions, radicals, and excited species. These fragments diffuse toward the substrate and chemically bond with surface sites generated during the pre-treatment. Subsequent incoming species further react with the growing film, progressively forming the amorphous carbon network that constitutes the DLC coating [32,33]. The ion energy required for DLC film growth is typically provided by applying a bias to the substrate, with the highest reported sp³ fraction obtained at around 100 eV [33]. Using methane (CH₄) as a precursor often favors higher sp³/sp² ratios compared to other gases under similar conditions, since the carbon atom is already in sp³ hybridization. Another common approach is to use a CH₄/H₂ mixture, which promotes an increase in the hydrogen present in DLC coatings [34,35]. When the ion energy is lower than 100 eV, and the sp³ comes from the chemical saturation of the C=C bonds by hydrogen, the resulting film is called PLC. These films are usually deposited without substrate bias and display distinct properties compared to DLC, including a lower density and reduced hardness. However, their more flexible nature makes them particularly suitable for polymeric substrates, as they generally adhere better and present a lower risk of fracture or delamination when subjected to bending, stretching, or other deformations [33,36].

Overall, PECVD enables the properties of amorphous carbon coatings to be tailored, including improving stability and adhesion, through careful control of deposition conditions and plasma parameters. Their specific influence on coating performance will be detailed in Section 4.2.

2.2. Classification

The ternary phase diagram for amorphous carbon materials, from which their classification is derived, was first reported by Robertson et al. [37]. Since then, several studies have deposited amorphous carbon coatings using different PVD and CVD techniques, which in some cases deviate from the DLC region of the ternary diagram but are still referred to as DLC coatings [7,28]. In their recent work, Ohtake et al. [38], redefined the DLC region on the ternary diagram considering 74 different amorphous carbon coatings. Among them, 35 were deposited by PVD methods, including plasma-based ion implantation and deposition (PBII&D), arc ion plating (AIP), ionized evaporation (IE), and magnetron sputtering (MS), and 39 were produced by PECVD. Taking into consideration this study and the latest reports in the literature, a classification of amorphous carbon coatings based on C-C sp³ and hydrogen content, as well as hardness, is proposed and summarized in

Table 1. Classification of amorphous carbon materials based on C-C sp³, hydrogen, and hardness. The reported ranges were compiled by the authors based on the minimum and maximum values found in the literature.

Type	C-C sp3 (%)	Hydrogen (%)	Hardness (GPa) *
Pure amorphous carbon (a-C)	10–50 [6,38]	<5 [6,38]	9–25 [38]
Tetrahedral amorphous carbon (ta-C)	50–90 [6,38,39]	<5 [6,38]	25–90 [38,39,40]
Hydrogenated amorphous carbon (a-C:H)	10–60 [38,39]	10–50 [38,39,41,42]	9–45 [38,39]
Tetrahedral hydrogenated amorphous carbon (ta-C:H)	50–90 [38,41]	5–50 [38]	9–50 [38,40]
Polymer-like carbon (PLC)	Up to 70% sp3 ** but commonly presents a higher proportion of sp2 carbon bonds [36,38,41]	40–70 [36,38,41]	0.5–9 [38]
Graphite-like carbon (GLC and GLCH) ***	Predominantly consist of sp2 carbon bonds [41,43]	≤20 [41]	5–20 [43]

* The hardness values reported in this table can vary depending on the technique used and the coating thickness. ** Most of the sp³ bonds are H-terminated, and the material is soft and has low density. *** GLC and GLCH are part of the group of amorphous carbon materials; however, they were not included in this review due to the lack of studies regarding GLC and/or GLCH coatings on polymers addressing adhesion.

.

This table highlights the fundamental role of hydrogen in determining the mechanical properties of amorphous carbon coatings. Coatings (a-C, ta-C, a-C:H, and ta-C:H) with a percentage of hydrogen less than 50% are commonly classified as DLC and show higher hardness values [6,38,39,40,41,42]. In contrast, when the percentage of hydrogen ranges between 40% and 70%, the coatings are classified as PLC [36,38,41]. It is important to note that the hardness values indicated in this table vary depending on the technique used to measure hardness, the thickness of the coating, and the substrate [38,39,40,43].

3. Adhesion of Amorphous Carbon Coatings on Polymers for Biomedical Applications

This section provides an overview of the most common techniques used to qualitatively and quantitatively assess the adhesion of amorphous carbon coatings on polymers. Since both the choice of polymeric substrate and the type of amorphous carbon coating are closely linked to their intended biomedical application, this aspect is also briefly discussed. In addition, 16 of the most relevant studies assessing adhesion for biomedical applications found in the literature were further critically evaluated. Particular attention was paid to the strategies used to improve adhesion and to the type of coating deposited according to its application. However, before explaining in detail these techniques and the studies, the fundamental concept of adhesion needs to be summarized.

3.1. Fundamentals and Adhesion Assessment Techniques

The term adhesion refers to the phenomenon whereby two surfaces are held together by interfacial forces that can include mechanical, chemical, and physical interactions [44]. Adhesion between identical materials is termed homo-hesion, whereas adhesion between dissimilar materials is termed hetero-hesion. In the context of thin coatings and experimental practice, the terms bond strength and adhesion strength are commonly used and can be defined as follows:

• The maximum force per unit area exerted when two materials are separated.
• The energy (work) applied to separate or detach two materials from each other.

When two materials are detached from each other, the corresponding failure is defined as an adhesive failure if the break occurs at the interface (the boundary layer where the film or coating meets the substrate), whereas, if the failure occurs between the coating itself, it is considered a cohesive failure [44].

Table 2. The most common techniques used to evaluate the adhesion of amorphous carbon coatings on polymeric substrates. Each method is presented with a brief description, a schematic, and its limitations, as well as the recommended system (coating/substrate) for its use. Data are from [44,45,46].

Technique (Adhesion Test)	Description	Applied on: Type of Coating/ Type of Substrate				Limitations
		Hard/ Hard	Hard/ Soft	Soft/ Soft	Soft/ Hard
Rockwell indentation (formation of crater and analysis with optical microscopy) ASTM D785 [50]	A conventional Rockwell hardness test is carried out, followed by a qualitative analysis of the network of coating cracks and flaking around the edges.					The interpretation of the cracks can be subjective.
Cross-cut tape test ASTM D3359 [51]	Involves creating a cross-hatch pattern of cuts on the coated surface, applying adhesive tape, and then peeling it off. This qualitative test provides a quick way to gauge adhesion strength, and results are often interpreted using standardized rating scales.					The cross-cut pattern can potentially damage both the coating and the underlying polymeric substrate, leading to imprecise results.
Ball-on-disc test ASTM G99 [52]	Qualitative/semi-quantitative technique. A ball (steel, Al2O3, Si3N4) is pressed against the substrate mounted on a disc with a predefined force, establishing a contact area between them. Then, the disc is rotated at a constant speed while the ball remains stationary or is moved in a specific pattern across the disc’s surface. The wear track on the disc can reveal information about how well the coating adheres to the substrate.					An indirect measurement of adhesion. Misinterpreted data.
Pull-off test ASTM D4541 [53]	Quantitative technique, in which a force is applied perpendicular to the substrate surface, causing the coating to detach or “pull off” from it. The force required to remove the coating is measured and used to assess the adhesion strength.					Difficult to distinguish between cohesive failure, adhesive failure at the interface, and failure within the substrate.
Pull-out test ASTM not specified	Quantitative analysis. A force is applied parallel to the substrate surface, causing the coating to slide or be pulled out from the substrate.					Might not provide the most reliable or meaningful data, and it might be difficult to identify the mode of failure.
T-peel test ASTM D1876 [54]	This quantitative test measures the force required to peel apart two flexible materials that are bonded together. It has a T-shaped geometry, and the peeling force is applied at the free ends of the T. The recorded force vs. displacement data is analyzed to determine the peel strength.					Flexibility is required. Hard coatings are more prone to cracking than peeling, making it difficult to accurately measure adhesion strength using this method.
Scratch test ASTM D7027 [55]	This method typically uses a spherical-tipped diamond indenter to scratch a coating with progressively increasing force, either continuously or in stages. The critical load necessary to detach the coating from the substrate is a qualitative indicator of the adhesion strength between the coating and substrate.					Difficulty in identifying failure modes. Complex data interpretation to determine the exact point of failure.
Tensile stress relaxation test and analysis of cracks and/or delamination ASTM D2990 [56]	Dog-bone-shaped specimens are subjected to tensile deformation, in which one end is connected to a load cell while the other is fixed. The coating delamination or cracking is qualitatively analyzed in the center area after the release of the strain.					It can be used only with flexible substrates. Complex fracture behavior and process for the identification of crack type (adhesive or cohesive).

The ✓ symbol denotes that the technique is suitable for evaluating adhesion on the specified coating/substrate combination.

details the most frequently reported tests in the literature. These include the cross-cut tape test, pull-off test, pull-out test, T-peel test, and scratch test. Additionally, less common methods like Rockwell indentation, ball-on-disc, and tensile tests have been adapted to obtain indirect information on adhesion and are often supplemented by microscopic analyses [44,45,46]. This table highlights the most suitable techniques to measure the adhesion as a function of the polymeric substrate and coating type (hard or soft). The classification of the type of polymeric substrate was based on the average hardness values (i.e., the surface resistance of a material to deformation, particularly permanent deformation) determined by Shore A, Shore D, or Rockwell R scales for polymers [47,48,49] as already presented in Figure 1. As far as coatings are concerned, given the indentation test results found in the literature, this review indicates that PLC is considered the softest coating, while a-C or ta-C are classified as the hardest coatings, and a-C:H coatings are considered intermediate in terms of hydrogen and sp³ carbon content (see Table 1).

3.2. Biomedical Applications of Polymers and Amorphous Carbon Coatings

Polymer selection for a specific medical device, particularly implants, depends on several factors, including the following: functional requirements, biocompatibility, mechanical properties, and regulatory compliance. Manufacturers carefully consider these features when selecting the most suitable polymer for each application.

For orthopedic implants (e.g., total joint, knee, and hip prostheses, as well as intervertebral disc implants), frequently used polymers include ultra-high-molecular-weight polyethylene (UHMWPE), medical-grade high-density polyethylene (HDPE), and polyether ether ketone (PEEK) [8,9,57,58,59,60]. These polymers are selected for such applications due to their biocompatibility and mechanical properties, including wear resistance and low friction coefficients. In dentistry, PEEK has been used for dental implants as an alternative to titanium, as it has an elastic modulus closely matching that of human bone [61]. In the case of polymethyl methacrylate (PMMA), this polymer has been used to design temporary crowns or bridges that are placed on the dental implant, as dental fillings, and for the fabrication of dental implants. Interestingly, PMMA can also be used for intraocular and contact lenses due to its optical transparency and recognized biocompatibility [60,62]. In the same field of orthopedic implants, there are recent advancements in the use of 3D-printed artificial scaffolds made of polycarbonate (PC) to replace damaged joints in the hip or knee by tailoring the porosity and mechanical strength of this polymer [63]. This polymer has also been used for optical applications such as ophthalmic lenses due to its optical transparency [64].

For tubular medical devices such as urinary and cardiovascular catheters, soft polymers, including polydimethylsiloxane (PDMS or silicone), polyurethane (PU), and polytetrafluoroethylene (PTFE), are commonly used. These materials are particularly valued for their flexibility, elasticity, and non-toxicity [65,66,67,68,69]. Additionally, PTFE has been used to produce stent delivery systems, vascular grafts, and heart valves [70]. Polyethylene terephthalate (PET) and polypropylene (PP) are also widely employed in these applications due to their mechanical strength, chemical resistance, and non-toxicity [71,72].

Despite the valuable properties of these polymers, researchers have extensively investigated their surface modification, particularly through the application of amorphous carbon coatings, to enhance their mechanical and biological performance. As illustrated in Figure 2, such coatings have been applied across a wide spectrum of biomedical devices, including orthopedic, cardiovascular, dental, optical, and tubular medical systems, each benefiting from tailored surface properties.

DLC coatings, in particular, are well known for their hardness and excellent wear resistance; these coatings also exhibit outstanding resistance to chemical and physiological environments, reducing the risk of material degradation caused by oxidative, corrosive, or enzymatic processes and minimizing potential cytotoxic or inflammatory responses [8,11,42]. In vitro cell viability studies showed no cytotoxic response toward different human and mouse-derived cell lines, including human osteoblast-like [73], MG-63 [8], hBMSC [61], hBMNL [72], mouse fibroblast [74], MC3T3 [75], and NIH-3T3 cells [67]. Regarding in vivo studies, different substrates have been coated with a variety of amorphous carbon materials, primarily DLC. These substrates were implanted between 3 and 12 months in transcortical sites of sheep, intramuscular locations of rats [73], skeletal muscles of rabbits [76] and guinea pigs [77]. The results showed no toxicological effects, inflammation, or excess scar tissue in any of the animals tested. Furthermore, diverse studies assessed the hemocompatibility of DLC coatings, demonstrating a reduced tendency for platelet adhesion, effectively lowering the risk of thrombus formation [78,79,80,81,82,83,84]. This biological property makes DLC coatings particularly interesting for blood-contacting medical devices, such as vascular grafts and stents, for which thromboresistance is critical [40].

Beyond biological performance, ensuring the strong adhesion of amorphous carbon coatings to polymeric substrates remains a critical challenge, particularly under operational physiological conditions. Therefore, complementary testing is often necessary to better replicate real-life conditions for each specific application.

For instance, in orthopedic implants, the wear resistance and the ability to withstand mechanical stresses are crucial properties [8,9,85]. These materials are subjected to constant movement and friction. Therefore, techniques such as ball-on-disc tests and fatigue tests are essential to evaluate their performance under realistic conditions. For dental implants, they not only withstand the cyclic load of chewing, biting, and grinding but are also subjected to a complex environment that includes constant exposure to saliva and bacteria [86]. To ensure their durability, tensile and compression tests, as well as fatigue tests, are essential for evaluating the mechanical resistance and adhesion of these materials. Flexible implants, such as cardiovascular catheters and stents, must balance flexibility with shape retention while supporting blood flow [87]. To achieve this, tensile tests are used to determine the material’s ability to stretch, while bending tests are used to evaluate its flexibility without breaking in static and dynamic conditions. Subsequent microscopic analyses are carried out to observe how the coating reacts to these deformations.

3.3. Reported Adhesion Studies on Polymeric Substrates

As part of this review, 16 relevant studies were critically evaluated for their assessment of the adhesion of amorphous carbon coatings on polymeric substrates intended for biomedical applications. The studies, summarized in

Table 3. Summary of studies evaluating the adhesion of amorphous carbon coatings on polymeric substrates for biomedical applications. The classification of the amorphous carbon type is based on the terminology used by each original study, as no standardized criteria were applied, and definitions vary among authors. Hardness values do not correspond to those in Table 1 due to the conditions of the technique applied and other factors, like coating thickness. Low values for the intensity ratio of the disorder (D) peak to the graphite (G) peak (ID/IG) obtained by Raman spectroscopy represent higher sp³ carbon, while higher values of the ID/IG ratio represent higher sp² carbon.

Substrate	Pre- Treatment	Type of Amorphous Carbon	Analysis to Determine the Type of Amorphous Carbon	Adhesion Test	Adhesion Results	Application /Biological Performance	Ref.
UHMWPE	Etching by Ar+	a-C:H (1.5 µm)	Raman ID/IG: 0.2–0.3	Rockwell D penetration and modified scratch tests with a ball indenter	Maximum normal load of 10.6 N. The adhesion of the coatings to the UHMWPE substrates can be considered sufficient for use in total knee replacements.	Total knee replacements/ Not cytotoxic to MG-63 cells 1 in vitro	[8]
UHMWPE	Etching by Ar+	a-C	Raman ID/IG: 0.67 Indentation Hardness: 2.16 GPa	Rockwell and scratch tests	The sample deposited at −60 V bias exhibited greater adhesion than those at −40 V and −80 V. However, the sample produced with a gradient bias voltage exhibited superior adhesion compared to the one deposited with a balanced bias voltage. Maximum normal load of 3 N.	Total joint replacements /Not assessed	[57]
UHMWPE	O2 plasma	DLC	ID/IG is not presented in this paper, but the authors affirmed that the DLC coating spectra obtained by Raman were similar to a typical DLC coating with a mixture of sp3- and sp2-bonded carbons.	Ball-on-disc test and scratch test	After the ball-on-disc test, the DLC coating without pre-treatment exhibited cracks, while the DLC coating pre-treated with oxygen plasma showed no cracks, indicating improved adhesion strength. In the scratch test, the critical loads for DLC-coated UHMWPE could not be reached to indicate the adhesive strength to the substrate.	Total hip and total knee joint prostheses/ Not assessed	[60]
Medical-grade HDPE	Ar plasma by reactive ion etching (RIE) Ar and O2 plasma by PECVD	DLC and Si-DLC	Values of sp3/sp2 ratio, hydrogen content, and ID/IG ratio are not mentioned.	Scratch test	DLC coatings with 21 at.% of Si-doped deposited by PECVD showed greater adhesion compared to pristine DLC thin coatings deposited by pulsed laser deposition (PLD) on HDPE substrates that were modified with RIE. Maximum load is not reported.	Hip implants/ Not assessed	[58]
PEEK	O2 plasma before PLC deposition O2 plasma and titanium interlayer before a-C:H deposition	PLC and a-C:H	XPS: PLC, sp2 C=C 64.5%, sp3 C-C 20.4%, C-O 15.1%. a-C:H, sp2 C=C 65.3%, sp3 C-C 25.5%, C-O 9.3%. Nanoindentation hardness: PLC, 3.2–3.6 GPa; a-C:H, 5.1–5.2 GPa	Pull-off and scratch test	PLC coatings showed higher pull-off force (1400–1500 N) with cohesive fracture compared to a-C:H coatings (980–1390 N). PLC coatings reached rupture at 5 N, but partially adhered to PEEK in the scratch test at a normal load of 10 N. The a-C:H showed delamination at 10 N.	Intervertebral disc implants and dental implants/ Not assessed	[59]
PEEK	Cleaning with Ar plasma as pre-treatment NH3 plasma after DLC deposition	DLC and NH2-DLC	XPS showed an increase in carbon content after the deposition and confirmed the amination of PEEK. The authors did not mention the sp3 or sp2 content. Nanoindentation tests showed an increased value from 0.38 GPa to 2.99 GPa after DLC and NH2-DLC deposition.	Scratch test	In the results of scratch tests, neither cracks nor spallation can be found around the scratch tracks, and no failures of the deposited DLC coatings were visible after a loading force of 100 N, indicating excellent adhesion between the coatings and PEEK substrates.	Orthopedic implants /NH2-DLC favorable to osteogenic performances of hBMSCs 2 in vitro and can facilitate the peri-implant bone regeneration in vivo	[61]
PC	O2 plasma	SiOx-DLC, a-C:H, and Si-a-C:H (ICP-CVD) C2H2 + 10%TMS 3, O2 + C2H2	250 nm thickness Hardness: DLC: 13.1 GPa, Si-DLC: 11.8 GPa, SiOx-DLC: 10.79–11.9 GPa	Scratch test	While the undoped DLC coatings have relatively poor adhesion strength, Si doping with 10 vol.% TMS in DLC coatings enhances the adhesion strength from 21.5 mN to 24.8 mN. Incorporation of a-SiOx network in DLC coatings by supplying 5 vol.% O2 in plasma leads to further increase in adhesion strength up to 37.8 mN.	Orthopedic implants and optical applications/ Not assessed	[88]
PC	Not mentioned. The doping could be considered as pre-treatment	a-C, SiO2 (ECR-type IBS 4), a-C:H, a-C:H:Si bipolar-type PBII 5	XPS and ERDA 6: a-C 100% C (140 nm), a-C:H 81% C, 19% H (40 nm), a-C:H:Si 63% C, 20% H, 17% Si (40 nm), SiO2 (265 nm)	Scratch test	a-C:H:Si showed the highest adhesion strength (maximum critical load of around 32 mN).	Orthopedic implants and optical applications/ Not assessed	[64]
PMMA	Ar+ ion sputtering	DLC	The authors confirmed the presence of D- and G-bands, characteristic of the DLC structure, by Raman spectroscopy. FTIR peaks: 1450 cm−1 correspond to sp3-bonded CH2, peaks at 1245 cm−1, and between 1515 cm−1 and 1640 cm−1 are related to sp2/sp3 bonded C–C.	Pull-off test	DLC coatings are significantly adhered to the PMMA surfaces. All the coated samples show adhesion strength between 2.96 and 3.25 MPa.	Artificial dentures, bones, and ophthalmic intraocular lenses/ Not assessed	[62]
PMMA	O2 plasma	DLC	Raman ID/IG is not presented in this paper, but the authors affirm that the DLC coating spectra are similar to a typical DLC coating with a mixture of sp3- and sp2-bonded carbons.	Ball-on-disc test and scratch test	Cracks were observed in the DLC coating without pre-treatment. DLC-coated PMMA showed a smoother surface than uncoated PMMA. The adhesion strength obtained by the scratch test for DLC-coated PMMA was 101.5 mN with plasma pre-treatment, compared to 42.5 mN without pre-treatment.	Contact lenses and denture teeth/ Not assessed	[60]
PDMS	Not mentioned	DLC	DLC coatings deposited at different powers were analyzed by Raman, and the ID/IG ratios obtained were as follows: (a) 300 W: 1.30 (b) 500 W: 1.02 (c) 900 W: 2.11	Scratch test	The highest adhesion value (58.94 N) was registered for samples produced at the highest power (900 W). Applying the lowest power (300 W) resulted in the thickest carbon coating and the poorest mechanical properties (18.70 N).	Not specified/ Not assessed	[65]
PDMS	O2 plasma	a-C:H or DLC	Raman ID/IG: 0.2–0.6 The G-band shift to a higher frequency indicates that the DLC on silicone is more graphitic in nature.	Pull-off test	DLC coatings with a refractive index <1.85 showed greater adhesion (~20 MPa), while for those >1.85, the adhesion was approximately 12 MPa.	Not specified/ Not assessed	[66]
PU	O2 plasma	DLC	All spectra had similar band shapes of typical DLC Raman spectra at any position exhibiting a shouldered peak (D-band at 1350 cm−1) and a broad peak (G-band at 1570 cm−1).	Tape test and nano-scratch test	On untreated substrates, DLC coatings were easily removed after the tape test. Coatings pre-treated with O2 plasma adhered strongly to substrates. Nano-scratch tests revealed that O2 plasma improved the adhesion strength of the coating by 10% compared to the untreated coating. Maximum load of 100 mN.	Medical devices and implants/ Non-cytotoxic NIH-3T3 cells 7 in vitro	[67]
PTFE	N2 plasma	DLC doped with fluorine	All coatings exhibited two broad peaks in the Raman spectrum (G- and D- bands), indicating that DLC was successfully coated onto PTFE despite fluorine doping.	Pull-out test	The adhesion strengths of all samples with different CF4 ratios (0% to 80%) were improved with N2 plasma pre-treatment. The sample with the highest adhesion strength was the 0% CF4 ratio with ~4 MPa compared to ~1.8 MPa for the untreated sample.	Not specified/ Not assessed	[68]
PTFE	Ar or O2 plasma and a-C:H:F coating interlayers	a-C:H 60.1 at.% of C, and 39.9 at.% of H. a-C, H-free, 100 at.% of C.	Two peaks in the Raman spectra: the D-peak (1350 cm−1) and the G-peak (1530 cm−1), which is representative of the DLC coating. HR-ERDA 8 was used to study the H content in the coatings.	T-peel test	The peel strengths of Ar and O2 plasma pre-treated samples were improved in comparison with those without pre-treatment. Ar plasma showed the maximum peel strength of around 2.7 N/cm on H-free a-C coatings. The authors did not mention the results for a-C:H coating.	Stents and prosthetic vascular grafts/H-free a-C inhibited thrombus formation in vitro	[70]
PET	Doping with N and Si	N-DLC and Si-DLC	Raman ID/IG ratio is 0.82 for pure DLC, 1.12–1.42 for N-DLC, showing that nitrogen doping enhances sp2 bonding formation. The ID/IG ratio for Si-DLC samples is 0.54–0.76, confirming that Si doping facilitates the formation of sp3 bonding.	Scratch test	No delamination was observed for N-DLC samples, while coating detachment was observed around the scratch for Si-DLC, indicating better adhesion of N-DLC compared to Si-DLC. Increase in load from 0.1 to 0.5 mN.	Artificial heart valves and blood vessels/Not assessed	[71]
PP	Ar and O2 plasma	a-C:H doped with SiOx	The Raman bands found are not the typical ones for DLC coatings. However, after the deconvolution, the authors identified five peaks corresponding to the different carbon forms.	Pull-off test	Ar treatment results in 4.4 times higher pull-off force (~15 kg/cm2) than that of the initial substrate. With O2, the maximum force (~24 kg/cm2) was 7 times higher compared to the substrate without pre- treatment.	Blood-contacting devices, such as heart valves and catheters/Not cytotoxic to hBMNLs 9 in vitro	[72]

¹ MG-63 cells: Human osteosarcoma cell line derived from bone cancer. ² hBMSCs: Human bone marrow-derived mesenchymal stem cells. ³ TMS: Tetramethyl silane. ⁴ ECR-type IBS: Electron cyclotron resonance-type ion beam sputtering system. ⁵ Bipolar-type PBII: Bipolar-type plasma-based ion implantation. ⁶ ERDA: Elastic recoil detection analysis. ⁷ NIH-3T3 cells: Mouse embryonic fibroblast cell line established from Swiss mouse embryos. ⁸ HR-ERDA: High-resolution elastic recoil detection analysis. ⁹ hBMNLs: Human blood mononuclear leukocytes.

, are organized according to substrate type, coating type, adhesion testing method, reported results, intended application, and biological performance. This overview provides a basis for comparing different strategies used to enhance coating adhesion across various polymeric materials and medical device applications.

Table 3 demonstrates the wide range of amorphous carbon coatings on polymeric substrates (both soft and hard). All polymer surfaces were pre-treated or modified before the coating deposition to enhance the adhesion, with plasma surface modification being the most commonly used technique. The type of amorphous carbon coating deposited also plays a significant role in adhesion. However, many of these studies do not report the chemical and structural characterization of the coatings, simply using the generic term “DLC”. Additionally, the use of different adhesion testing techniques complicates the direct comparison of results. Among these methods, the scratch test stands out as the most frequently used method. However, variations in test conditions and maximum loads applied (ranging from a few mN to 100 N) make direct comparisons difficult. Although no clear trend emerges, the available data suggest that flexible coatings, such as PLC and a-C:H, generally exhibit stronger adhesion to polymeric substrates compared to a-C coatings, regardless of whether the substrate is hard or soft.

At the same time, it is important to emphasize that only 6 out of the 16 studies summarized in Table 3 included biological evaluations (such as cytotoxicity or hemocompatibility), and just one reported in vivo results. This highlights a recurring gap in the field, in which studies tend to focus either on adhesion or on biological performance, rarely addressing both aspects within the same work. As a result, while the available data suggest promising compatibility, they remain insufficient to fully capture the long-term behavior of these coatings under physiological conditions. Moving forward, integrating adhesion assessments and biological testing within the same study would be highly valuable, as it would provide a more realistic evaluation of coating performance in clinically relevant environments.

Taken together, these findings underline that both the choice of pre-treatment and the type of amorphous carbon coating are decisive factors for adhesion, yet their biological validation is still fragmented. This will be explored in more detail in the following section.

4. Perspectives on the Main Factors That Influence the Adhesion of Amorphous Carbon Coatings

The factors that influence the adhesion of amorphous carbon coatings to polymeric substrates include the following: pre-treatment, the type of amorphous carbon coating, and, in some cases, the use of dopants. These components are interconnected and dependent on each other, as shown in Figure 3. This section will focus on breaking down each component in order to better understand their effects on adhesion and to inform and guide the current knowledge base on the adhesion of amorphous carbon coatings on polymeric implants.

4.1. Effect of Pre-Treatment

Based on the studies presented in Table 3, pre-treatment is essential to ensure the stable adhesion of DLC coatings on polymeric substrates, with plasma surface modification being the most common and effective method. Figure 4 illustrates the three principal plasma-based pre-treatment processes, classified into cleaning and etching, surface activation, and coating/interlayer deposition.

The most critical step in the pre-treatment process is thorough surface cleaning. Any contaminants, such as oils, greases, or oxides present on the substrate surface, can significantly weaken adhesion [29]. Proper cleaning ensures that the amorphous carbon coating adheres well to the substrate. Most studies report two common cleaning approaches: ultrasonic baths with solvents and plasma cleaning. Plasma not only avoids solvent use, offering environmental benefits, but also enables surface activation and thin-film deposition in a continuous process. Some researchers used plasma pre-treatment with argon, oxygen, or a mix of both prior to deposition to clean and activate the polymeric surface [29,89,90]. For example, polymers like PDMS are inert, meaning that their chemical interaction with other materials is weak [91,92,93]. In these situations, inert gases like argon are used to clean and etch the surface, whereas reactive gases like oxygen or nitrogen activate the substrate surface, making it more chemically reactive [29,91]. This activation promotes adhesion by enhancing the bonding between the amorphous carbon coating and the substrate. However, this varies depending on the polymeric substrate and the possible chemical bonds that can be formed. As seen in Table 3, oxygen plasma is the most common pre-treatment used for a wide range of polymers [58,59,60,66,67,70,72,88], while nitrogen plasma was reported only for PTFE [68].

Another pre-treatment strategy is the use of an intermediate layer between the substrate and the amorphous carbon coating. This interlayer acts as a buffer, reducing the stress mismatch between the coating and the substrate, and eventually enhancing the adhesion. Yamato et al. [70] employed a fluorine-incorporated a-C:H interlayer deposited onto PTFE to enhance the adhesion between the PTFE and a-C coating. PTFE is a fluorocarbon polymer composed primarily of carbon and fluorine atoms. When fluorine was introduced into a-C:H, the surface chemistry of the interlayer became more closely matched to that of PTFE. This similarity in chemical structure improved the interfacial compatibility between the layers, promoting better bonding. In addition, fluorine doping can promote F-F interactions across the interface with the fluorine atoms in PTFE. In instances in which a substrate is soft, but the properties of a hard coating are required, flexible interlayers can be used, for example, an a-C:H interlayer with a high hydrogen content or a PLC interlayer. The flexibility of the buffer layer ensures that it can withstand stresses similar to those of the substrate without failing, thus maintaining the integrity of the overall coating system. It helps bridge the gap between the mechanical properties of a hard outer layer and those of a soft substrate.

4.2. Effect of Coating Type

The choice of amorphous carbon coating will undoubtedly influence its adhesion, given the wide range of structural, chemical, and mechanical properties. DLC coatings exhibit various structures, which can be broadly categorized as H-free a-C, H-free ta-C, a-C:H (an H content from 10 to 40%), and PLC (an H content between 40 and 60%). The main characteristic of H-free coatings is their high hardness, reaching high values up to 90 GPa. However, as hydrogen increases, the hardness of the coating decreases. When the amount of hydrogen exceeds 40%, the amorphous carbon coatings become soft, like PLC materials [6,33,38]. The hydrogen content determines the hardness of amorphous carbon coatings and, consequently, their adhesion to polymeric substrates. This is attributed to hydrogen’s ability to enhance surface interactions and reduce internal stresses [94,95,96]. a-C:H and PLC coatings have shown lower internal stresses than a-C or ta-C coatings. These coatings normally exhibit better adhesion because their low internal stress prevents delamination from polymeric substrates.

The type of amorphous carbon coating produced depends on the plasma parameters used during the deposition process [6,33]. Amongst these parameters, the greatest determinants are pressure, power, and bias voltage [28,97,98,99]. Specifically, pressure influences plasma density, reactivity, growth kinetics, and ultimately, the mechanical properties of amorphous carbon coatings. Higher pressure increases dissociation and ionization rates, facilitating hydrogenation and promoting PLC characteristics. Conversely, DLC coatings are typically produced at lower pressures (0.01 to 1.0 Torr), depending on the deposition method, system configuration, and desired coating properties [36].

Power is linked to the ionization rate and dissociation of particles in the plasma. Generally, increasing power during coating deposition raises the ID/IG ratio. Since a higher value of the ID/IG ratio corresponds to a lower fraction of C-C sp³ hybridized carbon, coatings obtained at a higher power exhibit a reduced sp³ content, decreased density, and increased flexibility, all characteristic of PLC coatings [65].

Another crucial parameter is the bias voltage applied. This parameter dictates the energy of the incoming ions to the substrate. Dufils et al. [59] deposited PLC coatings on PEEK by applying no bias to the substrate. The resulting coatings exhibited high adhesion, which was related to the low hardness (between 3.2 and 3.6 GPa) and low residual internal stresses of the PLC coating. When the substrate was biased, the hardness (5.2 GPa) and the internal stresses of the resulting coating increased, denominated as a-C:H. Thus, the adhesion of this coating decreased, leading to delamination.

There is an evident relationship between the plasma parameters (i.e., pressure, power, and bias voltage) and the final properties of the coating. It is important to consider the nature of the substrate and the final application when determining the key parameters of the resultant coating, as a clear compromise between the desired coating properties (e.g., hardness, wear resistance, antibacterial, and cell viability) and the adhesion performance is often required.

4.3. Effect of Dopants

An important factor influencing adhesion performance is the use of dopants. These are generally incorporated into amorphous carbon coatings to enhance the mechanical, tribological, and biological properties (e.g., cell viability and hemocompatibility) of polymeric substrates. Their influence depends on different factors, including dopant type, polymer nature, and the intended application. As shown in Table 3, common examples of dopants include nitrogen, fluorine, silicon, and their derivatives. For instance, when DLC was doped with silicon (Si-DLC), particularly at concentrations of about 17–21 at.%, a noticeable improvement in adhesion was observed for polymers such as HDPE [58] and PC [64,88]. For PET substrates, Ashtijoo et al. [71] demonstrated that nitrogen-doped DLC (N-DLC, 4–7 at.% N) outperformed Si-DLC (1.9–4.1 at.% Si) in terms of adhesion. They attributed this to nitrogen doping, which promoted sp² configurations through the formation of C=N bonds, reducing surface roughness and internal stresses. By contrast, Si doping formed sp³ Si–C, increasing the hardness, Young’s modulus, and roughness. In the case of fluorine-doped DLC (F-DLC) coatings, enhanced adhesion to PTFE was expected due to the inherent chemical compatibility between fluorine-containing species and the fluoropolymer surface. However, Ozeki et al. [68] reported a counterintuitive result: the DLC coating without CF_x doping (0%) achieved the best adhesion performance. This finding suggests that, in this case, the surface pre-treatment prior to deposition played a more decisive role in promoting adhesion than the chemical modification of the coating through fluorine incorporation.

Other dopants, such as iron, copper, and silver, have been used mainly for their antibacterial activity, which is a crucial and often overlooked feature for medical devices and implants [13,14,100,101]. However, these studies were excluded from the scope of this review because they did not report effects on adhesion. A search in the Web of Science database (2025) revealed that only 2 of 25 relevant articles evaluated both antibacterial and adhesion properties on polymers, while the rest focused solely on end-use applications, leaving adhesion unaddressed or reported elsewhere.

This gap underscores the relevance of the present work. Regardless of the biological or antibacterial efficacy of a coating, poor adhesion can compromise implant performance. Once implanted, medical devices are subjected to complex physiological environments in which coating–substrate bonding dictates long-term stability and function. Thus, an adhesion evaluation is essential to ensure both durability and clinical success. In designing antibacterial coatings, dopant type, size, and concentration must also be carefully optimized to balance antibacterial performance with mechanical integrity.

5. Conclusions

This review aimed to highlight and discuss the importance of the adhesion of diverse amorphous carbon coatings on polymeric substrates, particularly in the context of biomedical applications. To the best of the authors’ knowledge, this is the first review in this field. The impact of this work is twofold: (1) it serves as a comprehensive resource summarizing the state of the art in amorphous carbon coatings on polymeric substrates, and (2) it acts as a tool to help overcome the challenges associated with the use of these substrates. Furthermore, it aims to encourage future researchers to conduct extensive adhesion analyses. Based on the conclusions of this research, a SWOT analysis was developed, which is presented in Figure 5.

The classification of amorphous carbon materials has been summarized according to the latest research. However, most studies still lack essential structural (e.g., sp²/sp³ ratio and hydrogen content) and mechanical (e.g., hardness) characterization, which are needed to better understand the relationship between coating type, substrate, and adhesion behavior. This review also addressed the adhesion techniques currently used in the literature, which have been classified according to the most suitable coating/substrate system for obtaining the most reliable results. This is relevant because different polymeric substrates and their respective biomedical applications require specific adhesion techniques to predict performance under real-life conditions.

The commercialization of amorphous carbon coatings has been constrained by their weak adhesion to substrates, mainly due to high internal stresses and mismatched chemical bonding. In this context, evidence in the literature has demonstrated the critical role of both pre-treatment and coating selection in determining adhesion properties. Selecting these parameters appropriately can reduce the mismatch in mechanical properties between amorphous carbon coatings and polymers. Despite this, most studies do not report adhesion or coating stability, raising concerns about the effectiveness of these materials, especially their long-term reliability under physiological conditions. In this sense, assessing adhesion is imperative to guarantee the stability and longevity of medical implants and devices. In addition, research on PLC and a-C coatings remains limited. Given the unique set of challenges posed by polymers, whether soft or hard, further exploration of PLC coatings, a-C:H coatings with a high hydrogen content, and the use of flexible interlayers could shed light on how to address these issues.

Therefore, future research should focus on improving adhesion and evaluating long-term mechanical stability and biological performance, with more comprehensive in vitro and in vivo evaluations.