Morphology-dependent antibacterial properties of diamond coatings

Abstract Microorganisms promoted corrosion has caused significant loss to marine engineering and the antibacterial coatings have served as a solution that has gained attention. In this study, the chemical vapour deposition technique has been employed to grow three different types of diamond coatings, namely, ultrananocrystalline diamond (UNCD), nanocrystalline diamond (NCD), and microcrystalline diamond (MCD) coatings. The evolution of associated surface morphology and the surface functional groups of the grown coatings have demonstrated antibacterial activity in seawater environments. It is found that different ratio of sp3/sp2 carbon bonds on the diamond coatings influences their surface property (hydrophobic/hydrophilic), which changes the anti-adhesion behaviour of diamond coatings against bacteria. This plays a critical role in determining the antibacterial property of the developed coatings. The results show that the diamond coatings arising from the deposition process kill the bacteria via a combination of the mechanical effects and the functional groups on the surface of UNCD, NCD, and MCD coatings, respectively. These antibacterial coatings are effective to both Gram-negative bacteria (E. coli) and Gram-positive bacteria (B. subtilis) for 1–6 h of incubation time. When the contact duration is prolonged to 6 h or over, the MCD coatings begin to reduce the bacteria colonies drastically and enhance the bacteriostatic rate for both E. coli and B. subtilis.


Introduction
In daily life, human beings are easily exposed to bacteria from different sources. These bacterial microbes have many pathogenic properties, which can cause food poisoning and infectious diseases [1,2]. They can lead to the biofouling of alloys because of their affinity to alloys. Biofouling is long-term corrosion that begins with the formation of a precursor film followed by the adherence of bacteria and algae, and then the attachment of macroalgae [3,4]. Therefore, antibacterial materials and/or coatings are in high demand to tackle these problems. There are a number of antibacterial materials that have attracted much attention, such as Ag and Cu, etc [5]. Recently carbon-based materials have emerged as ideal candidates for preventing bacterial infections [6][7][8][9], due to their excellent physical and chemical properties, such as versatile surface functionality, intrinsic chemical inertness, and physical robustness. Recent interest in industrial applications and medical exploitations of carbon-based materials calls for the need for approved testing protocols for characterizing their interaction with living systems. The biocompatibility, lubricity, stability, and cell adhesion of carbon-based materials such as diamond-like carbon (DLC) and diamond coatings can be improved via modification of their surface to play an effective role in antibacterial application [10][11][12]. Due to their good biocompatibility and excellent mechanical properties, DLC coatings have become excellent implant materials [13]. However, the stability of DLC is not as good as diamond. For antibacterial applications in marine environments, diamond coatings are deemed more suitable than DLC [14]. Due to the diamond's corrosion resistance and long-term stability [15]. Moreover, diamond coatings grown by chemical vapour deposition (CVD) are finding an increasing number of applications due to the superlative properties of diamond coupled with the availability of an eclectic selection of morphologies, and affordability, which can be sourced from a number of commercial suppliers [16][17][18][19][20]. Recent progress has indicated that diamond nanoparticles have an antibacterial ability which can be used in life science, biotechnologies, and medicine [21][22][23]. It is well established that both Ag and Cu can kill a broad spectrum of bacteria. However, the sliver must be in wet conditions to release the ions necessary to disrupt cell membranes.
In comparison with Ag, Cu's two ionic states, make it a more active element regardless of the presence of moisture for an oxidizer [24]. Medina et al. [25] show that nanodiamonds have better bactericidal and bacterial anti-adhesive properties than Ag, but not as good as Cu. However, Cu doesn't have satisfactory biocompatibility. Among the promising antibacterial nanoparticles, the bactericidal effect of nano-Ag is the most widely recognized [26], but its overt toxicity to normal cell is unavoidable. The antibacterial mechanism of Ag mainly lies in changing the permeability of cells [27], damaging DNA [28], inducing cells to produce inhibitors [29], and accordingly resulting in bacterial cell death. Nano-Ag has a strong antibacterial ability, but it will also affect the normal cell. The antibacterial mechanism of Cu is similar to that of Ag. Different from Ag, Cu acts directly on the cell membrane, causing a voltage difference between the inside and outside of the cell membrane, resulting in the rupture of the cell membrane, and then the cell loses nutrition and shrinks, and dies [30]. Both Ag and Cu have varying degrees of antibacterial properties, but still destroy the normal cell as a side effect.
Diamond-based materials are proven non-toxic [31,32]. The antibacterial mechanism of diamond films is mainly through mechanical antibacterial, anti-adhesion, and oxygen-containing surface groups which offer fewer opportunities to damage the normal cells while killing bacteria [33,34]. Hence, there is widespread interest to study the antibacterial properties of diamond-based materials.
May et al. [35] suggest that high-aspect-ratio nanoprotrusions, such as spikes or needles can generate a mechanical bactericidal effect. The supreme antibacterial propensity by embedding silver nanodroplets on nanodiamond coatings is studied in detail [36,37]. The variable antibacterial properties of carbon coatings result from the different growth parameters can be adjusted during the deposition processes [38]. Poor bacterial adhesion provides a good environment for antibacterial materials. By changing the physicochemical properties of the material surface, the interaction between the material and the bacteria can be moderated, so as to prevent the bacteria from adhering to the biomaterial surface, thus preventing the formation of the bacterial biofilm. The difference between hydrophilicity and hydrophobicity is one of the important factors of anti-adhesion [39]. Dunseath et al. discussed the influence of the hydrophobicity and the surface structure of the needle tip on the bio-adhesion which led to improved antimicrobial activity [33].
In this study, diamond coatings with various surface morphology, namely, UNCD, NCD, and MCD were used to study the antibacterial behaviour on Gram-negative bacteria (E. coli) and Gram-positive bacteria (B. subtilis) in simulated marine environments. E. coli is chosen as the model of Gram-negative bacteria in this study. Most of the bacilli of marine origin belong to B. subtilis, according to their phenotypic characteristics, antibiotic susceptibility profiles, and fatty acid patterns. Therefore, B. subtilis is chosen as the model Gram-positive bacteria for the marine and seawater research.

Materials
The UNCD and NCD coatings were prepared by a hot filament chemical vapor deposition system which was reported previously [40,41]. MCD coatings were purchased from Element Six Ltd. Agar 2216E and 2216E liquid medium were obtained from Qingdao Hope Bio-Technology Co., Ltd. Stainless steel sheets 316 were provided by Shanghai Baosteel Group Corporation. All chemicals were of analytical grade and used without further purification. Deionized water was used in all experiments. A Gram-positive bacterium (B. subtilis, CMCC63501) and a Gram-negative bacterium (E. coli, CMCC44102) were provided by Shanghai Luwei Technology Co. Ltd.

Characterization
The crystal quality, phase analysis, and quantitative identification of the UNCD, NCD, and MCD coatings were studied by Raman spectroscopy (Renishaw, UK) with a 514 nm wavelength of the laser, FTIR (Thermo Nicolet 6700, USA) and XPS (XPS, Kratos AXIS ULTRA, UK). The grain orientation and crystallinity were investigated by XRD (Bruker D8 X-ray spectrometer, USA) using Cu-Kα radiation (λ = 0.154 nm) at 40 kV and 40 mA with a grazing incidence angle of 2θ. The range of scanning angle was set from 30° to 100° at a scanning speed of 10°/min and 0.02° step size. The surface morphology of diamond coatings was imaged by SEM (Hitachi S-3600N, Japan) and operated at 20 kV with a working distance of 10 mm. Particle size was obtained from SEM images using the ImageJ software (National Institutes of Health, MA, USA) and Jade software based on the XRD data. The roughness and surface topography of diamond films were examined using an atomic force microscope (AFM, Bruker Dimension Icon calibrated at ScanAsyst Institute). The wettability was evaluated by measuring the static contact angles between a droplet of artificial seawater and the sample surface.

Antibacterial assay
The model microorganisms, E. coli (CMCC63501) and B. subtilis (CMCC44102) were first inoculated onto agar plates. They were then separately transferred to the Luria Broth (LB) medium (E. coli) and 2216E liquid medium (B. subtilis), respectively. The cultures were incubated at a constant temperature of 37°C and shaken for 24 h at 120 rpm to activate the bacteria. Afterward, they were centrifuged for 5 min at 3000 rpm in order to separate the bacteria from the medium. The pellets were washed with a NaCl (0.9%) solution (E. coli) and artificial seawater (B. subtilis) three times to remove the residual medium. Shortly after, the E. coli were diluted in NaCl and B. subtilis diluted in artificial seawater, and the suspensions were analyzed with a microplate reader (MD Spectra-Max 190, USA) at a wavelength of 600 nm to establish a moderate concentration (105-106 CFU/mL). Artificial seawater is used to simulate the breeding environment of B. subtilis in the ocean. The osmotic pressure provided by normal saline can maintain the normal state of E. coli cells and will not lose their normal shape or even die due to water loss or absorption. After that, the UNCD and NCD, as well as MCD coatings, were placed in Erlenmeyer flasks containing 50 mL of bacterial solutions, whereas 50 mL of pure bacterial solutions were used as controls. Five replicates per bacterial species were incubated at 37 °C for 0-6 h (30 min, 1 h, 2 h, 3 h, and 6 h, respectively) with constant shaking at 120 rpm. Then, samples of 100 μl per solution were withdrawn and diluted 10-fold, 100-fold, and 1000-fold. Diluted bacterial solutions (100 μl per sample) were inoculated onto agar plates. Lastly, all agar plates were placed in an incubator with a constant temperature of 37 °C for 24 h to allow bacteria to grow and propagate. After 24 h, the colony numbers were obtained by counting the number of bacterial colonies on the plates and comparing them with controls and the bactericidal rate was calculated by the formula R% = (B-C)/B*100% where B is the bacterial concentration in the control sample, and C is the bacterial concentration in the media from the tested sample. Each experiment was carried out in triplicate, and the values reported in this study represent the means of three replicates. Figure 1 shows the SEM images of the UNCD, NCD, and MCD coatings. The formation mechanism of UNCD and NCD coatings has been extensively studied in Liang's work [42]. It can be seen that the grain size of UNCD and NCD coatings has increased from approximately 2.9 nm (UNCD) to 9 nm (NCD) as calculated by the Image software. For MCD coatings, the surface becomes very rough with irregular clusters.

Materials characterisation
The Raman spectra of diamond samples are shown in Figure 2(a). The diamond composition peak (D peak) appears at around 1332 cm −1 in all samples. The composition peaks (G peak) consisting of graphite and amorphous carbon located at 1580 cm −1 are also visible. It is generally believed that this is related to the disordered sp 3 structure carbon of nanodiamonds. The spectrum of UNCD consists of diamond sp 3 peak and characteristic sp 2 phase. The decrease of diamond grains is usually associated with more disorders, thus the energy range of different phonons involved in the Raman scattering process becomes wider. Raman spectrum of the NCD film was deconvoluted into four peaks at 1331 cm −1 , 1589 cm −1 , 2692 cm −1 , and 2886 cm −1 , as shown in the finely fitted inset in Figure 2(a). The peaks at 1331 cm −1 and 1589 cm −1 can be assigned to diamond sp 3 and graphite, respectively. The peak at 1331 cm −1 further reveals that the film contains mainly diamond and the one at 1589 cm −1 is the G-mode Raman peak which is attributed to the size effect of crystalline graphite [43]. Moreover, there are conspicuous trans-polyacetylene (t-PA) related peaks appearing at 1465 cm −1 and 1190 cm −1 for UNCD which is consistent with the previous report [44,45]. The more widely distributed phonon energy results in the broadening of Raman spectral lines implying that the UNCD and NCD have much smaller grain size as shown in the scanning electron micrographs. However, when the Raman peaks of UNCD and NCD are compared with that of the MCD, it can be seen that the latter has a very sharp sp 3 characteristic peak (FWHM narrowing) indicating that it contains a highly-oriented crystalline diamond. It can be concluded that the MCD coating obtained is of good quality and consists of a large amount of sp 3 carbon and few sp 2 .
The grain orientation and crystallinity were further determined by XRD, as shown in Figure 2(b). It can be seen that the diamond characteristic peaks are dominant in all the samples, with (111) orientation located at ∼ 43° and (110) orientation at ∼76°. The difference in peak intensities is barely distinguishable for NCD and UNCD; these coatings exhibit high crystallinity associated with sharp diffraction peaks and narrower full width at half maximum (FWHM). Moreover, the (111) crystal orientation is dominant in both NCD and UNCD coatings whilst the (110) crystal orientation dominates in MCD coatings.
Furthermore, from the Raman spectrum, the residual stresses can be calculated using the following equation [46]: where σ is the residual stress, and δ is the shift in the Raman spectra.
The calculated residual stress was −3.969 GPa for UNCD, and −5.67 GPa for NCD, respectively, which are compressive in nature owing to the fast growth rate. However, the calculated residual stress for MCD is small enough to be ignored.  . the FtiR spectra of the uncD, ncD, and mcD coatings (left), and amplified regional spectra for uncD (upper right) and mcD (lower right). Figure 3 shows the FTIR spectra of as-grown UNCD, NCD, and MCD coatings. The FTIR spectra for both UNCD and NCD are very similar as shown in Figure 3 (left region). However, different from UNCD, NCD does not have clear surface C-H functional groups and few absorptions attributed to the C-N, C-O-C, and C-C bonds as demonstrated in Figure 3 (right upper region). These groups are possibly from the contamination and abundant graphitic phases owing to the same growth environment for NCD and UNCD. It is worth noting that both NCD and UNCD have large surface-to-volume ratios with high sp 2 proportions. The peak around 2360 cm −1 is related to CO 2 (C = O bond) and unrelated to any surface processes or properties [47,48] For MCD, there is a clear peak around 2850-3000 cm −1 which has three vibration peaks due to C-CH, -CH 3, and -CH 2 bond, and the peaks around 2750 cm −1 to 2500 cm −1 show the presence of -CH 3 and = CH 2 functional groups resulting from the as-grown hydrogen terminations, with the hydroxyl situated around 3400 cm −1 . Meanwhile, the peaks situated at 1558, 1576, 1615, and 1636 cm −1 are attributed to -NH stretching bonds, and their attributed bending bands typical of the carboxyl group and C = O stretching band of ester were located at 1716, 1733, and 1749 cm −1 , respectively. Thus, the spectra of FTIR confirm the existence of -CH 2 , CH 3 , amino, ester, carboxyl, and hydroxyl on the MCD surface.
Raman and FTIR spectra have shown the overt sign of the existence of sp 2 and sp 3 carbon within the above three types of diamond.
X-ray photoelectron spectroscopy (XPS) was employed to verify the chemical compositions of the samples and the results of XPS were calibrated before fitting the curves by the adventitious carbon. Figure 4 shows the XPS survey spectrum of UNCD, NCD, and MCD as well as the high-resolution core level of C1s and O1s. There are only peaks of carbon and oxygen in Figure 4. It is observed that the C1s peaks can be deconvoluted into two main components at 284.8 eV and 285.4 eV, which are attributed to sp 2 and sp 3 carbon, respectively [48,49].
The fine deconvoluted sp 2 and sp 3 peaks from C1s spectra give the ratio of sp 3 /sp 2 shown in Table 1.
It can be seen that the sp 3 /sp 2 ratio increases with the increase of the grain size from UNCD to MCD. The increase of the grain size will decrease the surface to volume ratio of diamonds, resulting in less grain boundaries regions for the sp 2 phases to be trapped and embedded in.
Representative images from the surface contact angle (SCA) test are shown in Figure 5. The SCA is 118.72° for UNCD coatings which exhibit hydrophobic surface properties. The NCD coatings have shown mild hydrophilic properties with an SCA value of 87.11°. It is evident that MCD has the most hydrophilic surface with an SCA value of 42.24°. It can be seen that the hydrophilicity of the samples is in the following order: MCD > NCD > UNCD. As the roughest surface, MCD has the most hydrophilic property. The Blog Wenzel equation proposed by Wenzel describes the relationship between roughness and wettability, the equation is shown below [50]: where θ* is the measured contact angle on a real surface with an inherent roughness, θ is the contact angle on an ideal surface (also called Young's contact angle), and r is the roughness ratio between the actual and projected solid surface area (r = 1 for a smooth surface and r > 1 for a rough surface) [51]. The equation suggests that hydrophilic surfaces will become more hydrophilic with the increase of roughness. Compared with UNCD and NCD, the MCD surface contains more polar molecules such as -CH3, -CH2, -COOH, -OH, and -NH2, which will promote a more hydrophilic surface. As molecules with polar groups have a greater affinity for water, they can Figure 4. the XPS spectra of the diamond coatings. the inset is the fine deconvoluted c1s peak for fitted spectra of sp 3 (red) and sp 2 (green) components.  Figure 5. the contact angle of uncn, ncD, and mcD coatings.
attract water molecules more easily. Therefore, the increase in roughness makes the hydrophilic surface of MCD more hydrophilic. Before the biofilm was known, the main strategy to achieve the antibacterial effect was focused on anti-adhesion between bacteria and the surfaces of their surrounding medium. The mechanism of biological adhesion is rather complex [52], which can be divided into two parts: 1) Protein deposition, and 2) A layer of the extracellular matrix after the attachment of cells. In particular, the initial adhesion process in the bio-membrane is related to the protein composition of the extracellular polymer. It has been shown that hydrophobic materials with special treatment have certain anti-adhesion properties, but they are unable to keep their effectiveness in seawater for a long time [53]. In addition, protein adsorption is closely related to bacterial adhesion in seawater. Thus, the surface which can interfere with the protein adsorption will offer good anti-adhesion properties in seawater [54]. Figure 6 shows the typical two-dimensional and three-dimensional morphology of UNCD, NCD, and MCD coating measured by AFM on a 5 μm x 5 μm area. The surface roughness of UNCD, NCD, and MCD are shown to be around 29.6 nm, 38.4 nm, and 61.7 nm, respectively. It's known from the mechanical point of view that a smoother surface tends to offer a less adhesive base to bacteria, but some studies suggest that the nano rough surface has better anti-adhesion ability than the nano smooth surface [55,56]. Thus, the surface roughness alone doesn't affect the adhesion of bacteria exclusively in our studies.

The results of antibacterial assays
The antibacterial activity of the coating against E. coli and B. subtilis was evaluated and the antibacterial rate was used to quantitatively analyse the antibacterial activity of the coatings by the viable cell counting technique as described in Section 2.3. The number of colonies and bacteriostatic effect of diamond coatings for E. coli and B. subtilis after six hours of contact is shown in Figure  7. As a control, bare glass slides and silicon wafers were used for the comparison study. Both surfaces have a high number of colonies after exposure to both types of bacteria. This implies that their intrinsic antibacterial effect from glass slides and silicon wafers can be ignored. Therefore, when used as diamond substrates the silicon influence is negligible.
Among the three types of diamond coatings used, it appears that MCD is the most effective coating in killing both E. coli and B. subtilis, whilst UNCD is the least effective antibacterial coating.
The number of colony-forming units dropped and the bacteriostasis rate of MCD reached 56.84% for E. coli and 94.74% for B. subtilis as shown in Figure 7(c) and (d), respectively.

Discussion
This study has shown three types of diamond coatings have antibacterial properties against E. coli and B. subtilis. MCD is the most effective coating. There are a number of factors to determine the antibacterial effects, such as surface roughness, surface textures, surface wettability, and surface functional groups, etc. A detailed discussion is given below.
First of all, surface roughness is one of the key factors affecting cell activity. It is known that the UNCD surface offers the largest surface-to-volume ratio than NCD and MCD, thus it provides the biggest contact area between the diamond and bacteria during the contact-killing stage. Therefore, it is expected that UNCD should have the most effective antibacterial properties. However, our experiments show that both E. coli and B. subtilis experienced the most significant physical and mechanical damage on the roughest MCD surfaces instead of UNCD, which is controversial to what was expected. The possible mechanism responsible for this observed phenomenon is due to the fact that the rough surface acts as the nanotribology platform to destroy the bacterial cells [57].
Secondly, the surface texture is another important parameter affecting cell activity. Surface textures of various geometry have been artificially designed and fabricated on diamond surfaces. Dunseath's group has fabricated a sharp protuberance structured diamond film which can kill E. coli bacteria via the penetration of the needle-like surface texture [33]. Ye's group has designed the mushroom texture with super-hydrophobic surfaces which can prohibit the growth of the bacteria effectively [58]. Although UNCD is not the most effective antibacterial coating through this study, it has the potential to be used as the anti-fouling coating due to its low friction and low surface energies. UNCD offers a smooth surface, which can prevent the attachment of larger microorganisms. The surface texture is critically important as it may influence the surface wettability of the materials.
Surface wettability is characterized by the equilibrium contact angle (CA) of a sessile droplet on the surface in air. The hydrophilic (SHI) surface in the air (CA < 90°) becomes hydrophobic underwater and vice versa for the hydrophobic (SHO) surface (CA > 90°) [59]. Our data indicate UNCD coatings show the CA results are in the following order: UNCD > NCD > MCD, which suggests UNCD has a hydrophobic surface and MCD has a hydrophilic surface as shown in Figure 5. In general, a hydrophobic surface (UNCD in this study) is expected more effective to inhibit the adhesion and reproduction of bacteria and thus achieves a superior antibacterial performance [60]. But our results indicate the hydrophilic MCD coating is more effective than UNCD for both types of bacteria. This implies that wettability is not a decisive factor determining the antibacterial behaviour of diamond coatings. Although hydrophilicity is not the direct cause of the inactivation of bacteria, it can improve the antibacterial rate of bactericidal surfaces. It is proposed that the possible reason for the best antibacterial performance of the most hydrophilic surface is associated with bacterial anti-adhesion due to the fact that the hydration layer formed on the hydrophilic surface has a repulsive force and a steric hindrance effect. This is in agreement with the report by Gui et al. [61] It is worth noting that the chemical properties of diamond surfaces play an important role in the antibacterial behaviour of the coating. Surface functionalization is believed to be an effective approach to tuning the physical and chemical properties of diamond. Understanding and controlling the functionalization process of diamond surfaces thus holds considerable promise for antibacterial coatings development. It is known that hydrogen-terminated diamond surface shows hydrophobic wetting characteristics, besides chemical inertness, whilst oxygen-terminated diamond surface shows hydrophilic wetting features. Akhavan et al. [62] reported that oxygen-containing functional groups can help diamond coatings kill/inactivate the bacteria. From the FTIR results as shown in Figure 3, MCD contains -COOH, -OH, -NH 2, and easter which may enhance antibacterial properties. For the carboxyl group, it can inhibit bacterial adhesion through the electrostatic effect, while hydroxyl groups (epoxide) will generate oxidation pressure with the production of reactive oxygen species (ROS) to reduce cell activity [63]. Meanwhile, hydroxyl groups can be produced through the ring-opening reaction of the epoxy group catalysed by acids and bases. This indicates that the higher content of the hydroxyl will cause a higher probability of ROS production. These oxygen-containing functional groups might be the decisive factor responsible for the antibacterial behaviour of diamond coatings in ocean environments. Compared with UNCD and NCN, MCD possesses higher carbon sp 3 content. The difference in sp 3 /sp 2 carbon ratio is reported to relate to the variation of bacterial adhesion to diamond-like carbon coatings [64]. The higher sp 2 content may promote the hydrophilic characteristic which may affect the adhesion of bacteria. This may influence the bacterial colonies but has not demonstrated the killing effect to bacteria. Figure 8 shows the schematic antibacterial effect of different diamond surfaces. With the increase of the surface roughness from UNCD to MCD, there are holes and vacancies remaining on the diamond surfaces to interact with bacteria of different size. E. coli is a typical Gram-negative rod bacterium with 1.0-2.0 micrometres long whilst B. subtilis is a Gram-positive rod bacterium with 2-6 µm long. Smaller E. coli can be easily trapped within those holes and vacancies on the diamond surface, and bigger B. subtilis have a higher probability to contact with irregular surface textures on diamond, resulting in 94.74% bactericidal rate for B. subtilis on MCD surfaces. This phenomenon suggests that diamond coatings are more susceptible to B. subtili, which is in agreement with Beranova's result that the susceptibility was different based on the basic morphological difference between the cell wall structures of Gram-negative and Gram-positive bacteria [65]. Researchers have shown that when the microstructural pattern size on the surface is close to the size of the bacteria, the contact points provided for the bacteria will be reduced. The reduction of contact points makes the attachment of bacteria on the surface difficult, thus inhibiting the formation of biofilms effectively [66].
In summary, the antibacterial mechanisms of diamond coatings have been discussed. There are a number of factors to determine the antibacterial effect, such as surface roughness, surface textures, surface wettability, and surface functional groups. The interaction between bacteria and diamond surface is rather complex. The antibacterial performance of diamond coatings results from the combination of the above factors discussed.

Conclusion
UNCD, NCD, and MCD coatings have been characterised systematically by SEM, Raman spectroscopy, XRD, FTIR, surface contact angle, and AFM. It is found that MCD coatings show the strongest antibacterial properties against E. coli and B. subtilis in seawater environments. Many surface characteristics, such as surface roughness, surface textures, surface wettability, and surface functional groups may have an impact on the antibacterial properties of diamonds. However, none of these physical parameters has a decisive influence. The antibacterial process of diamond coatings against E. coli and B. subtilis is very complicated. The possible physical mechanism responsible for the antibacterial behaviour of diamond has been explained. The results demonstrated in this study have provided some new insight which will benefit to the development of the anti-corrosion and anti-fouling coatings for Oceaneering applications. The development of surface functionalised diamonds in a controlled and reproducible way can nowadays be achieved in several different manners and will be ready for exploitation for industrial application. The antibacterial performance may be further enhanced by multiple killing mechanisms such as doping diamonds with antifungal metals such as nano-silver or nano-copper to form a hybrid material, and by chemically attaching antibacterial drugs to the diamond surface. Further experiments on these prospects are currently on the way.