Doped DLC for bipolar plates (BPP)

Abstract

The present invention relates to a metallic bipolar plate comprising a metal substrate and at least one layer of non-hydrogenated transition metal-doped diamond-like carbon (DLC) disposed on the metal substrate. The non-hydrogenated DLC comprises at least one transition metal selected from groups 4d, 5d, and 6d of the periodic table of elements, with a portion of the at least one transition metal present in the non-hydrogenated DLC in the form of a carbide of the at least one transition metal as a matrix. The non-hydrogenated transition metal-doped DLC has an indentation hardness of 35 GPa or more, preferably 40 GPa or more. Metal-doped DLC used as a coating for metallic bipolar plates in fuel cells and electrolyzers exhibits a longer life (>10,000 hours) and high corrosion resistance and low interfacial contact resistance even under severe operating conditions. Therefore, the present invention also relates to such uses, and to fuel cells, particularly proton exchange membrane fuel cells, and electrolyzers comprising such coated metallic bipolar plates. The present invention further provides a cathodic arc discharge deposition method for depositing a coating of non-hydrogenated DLC, preferably non-hydrogenated transition metal doped DLC, comprising at least one of the transition metals listed above.

Description

[Technical Field]

[0001] The present invention relates to a layer system comprising non-hydrogenated transition metal doped diamond-like carbon (DLC) and a coating thereof provided on a substrate. Furthermore, the present invention relates to fuel cells, in particular proton exchange membrane fuel cells (PEMFCs), and electrolyzers comprising metallic bipolar plates with a coating of non-hydrogenated transition metal doped DLC according to the invention, and corresponding uses of the layer system. Finally, the present invention relates to a method for depositing a coating of non-hydrogenated DLC, preferably a coating of non-hydrogenated transition metal doped DLC according to the invention, which is a cathodic arc discharge deposition method. [Background technology]

[0002] Diamond-like carbon (DLC) is a metastable form of amorphous carbon. DLC films have found widespread applications in science and technology. J. Robertson provides a comprehensive overview of the material and its applications in Materials Science and Engineering R 37, 2002, 129-281. As reported by Robertson, it is necessary to distinguish between hydrogenated and non-hydrogenated forms of DLC. Another criterion for the classification of DLC is the sp 3 For non-hydrogenated materials, sp 3 When the bonding ratio reaches a high degree, the material is commonly referred to as tetrahedral amorphous carbon, abbreviated as ta-C. S. Xu et al. reported in Philosophical Magazine Part B, 76:3, 351-361 that ta-C films were deposited by filtered cathodic vacuum arc technique on silicon at room temperature. 3 Bond fractions (approximately 80% or higher) were obtained. Compressive stresses ranging from 7.5 to 12 GPa and hardnesses ranging from 20 to 55 GPa were observed. The maximum hardness was observed at the highest sp 3 It was found to be consistent with the fraction. Doped carbon coatings have also been developed. A. Abou Gharam et al. studied the high-temperature tribological behavior of W-DLC against aluminum in Surface and Coatings Technology 206 (2011), 1905-1912. W-DLC coatings were deposited using a physical vapor deposition (PVD) system. The addition of W to hydrogenated DLC reduced the friction coefficient in the temperature range of 400°C to 500°C. The problem is that doping of hydrogenated DLC with W, for example, occurs at the atomic level by sputtering. Non-hydrogenated DLC (ta-C) has better wettability with many lubricants and better temperature stability than hydrogenated DLC. Z. Wang et al., International Journal of Hydrogen Energy 42 (2016), 5783-5792, deposited W-doped carbon on a stainless steel substrate. They used a near-field unbalanced magnetron sputtering ion plating (CFUBMSIP) system with a bias voltage of -60 V. The CFUBMSIP system was equipped with one tungsten (W) target, two graphite targets, and one chromium (Cr) target for the deposition of a thin Cr seed layer and a thin intermediate MCx layer (where M is Cr and W). K. Hou et al. deposited niobium (Nb)-doped amorphous carbon (aC) films on stainless steel substrates. Similar to Z. Wang et al. in the aforementioned scientific paper, they used a CFUBMSIP system. The bias voltage was -100 V. The CFUBMSIP system was equipped with one Nb target, two graphite targets, and one titanium target. Niobium carbide is claimed to be embedded in the aC matrix, and the presence of pure niobium has been reported. A simulated first-principles calculation assuming one Nb atom in the supercell revealed approximately 58% sp 3 The simulation results suggest that the sp fraction is up to 54%. 3 However, the maximum sp fraction of 54% 3The fraction was obtained for films in which the cathode current of the Nb cathode was kept at 0, so no Nb was sputtered from the target. However, the Raman results for Nb-doped aC films showed a lower sp 3 In particular, considering the teachings of A.C. Ferrari, Diamond and Related Materials 11 (2002), 1053-1061, an I of about 2.5 is suggested. D / I G The ratio suggests this. No hardness data were reported in the paper by K. Hou et al. D. Zhang et al. (Carbon 145 (2019), 333-344) studied amorphous carbon films doped with silver (Ag) or co-doped with Ag and chromium using a CFUBMSIP system. They found that the higher the dopant percentage, the lower the hardness and graphitization. They reported simulated hardness values ​​that were significantly higher. For pure C, their simulations showed a simulated hardness of 56 GPa. However, their measurements showed compressive stresses of all doped C coatings ranging from 2.40 GPa to 3.37 GPa. M. Andersson et al. (Vacuum 86 (2012), 1408-1416) deposited and characterized magnetron-sputtered amorphous Cr-C films. They used non-reactive DC magnetron sputtering from an elemental target. The films were found to be X-ray amorphous, with no crystallites present. They reported a hardness of 6.9 GPa for a Cr dopant level of about 15 at.% (i.e., atomic %), increasing to 10.6 GPa at about 75 at.% Cr. Y. Lin and S. Zhang (J. Nanosci. Nanotechnol. 16 (2016), 12720-12725) investigated the effect of Cr addition on the properties of graphitic carbon (GLC) films. The films were deposited by unbalanced magnetron sputtering. They reported a hardness of 10.4 GPa for pure C and up to 17.4 GPa for Cr-doped GLC films. Although dopant levels were not explicitly stated, significantly lower dopant levels can be expected for Cr target powers increasing from 0.1 kW to 0.3 kW and carbon from 0 to 5 kW. A. Amanov et al. (Tribology International 62 (2013), 49-57) deposited Cr-doped and undoped DLC films using unbalanced magnetron sputtering (UBMS). They reported a hardness of 22.47 GPa for Cr-doped DLC and 10.75 GPa for undoped DLC. A. Ya. Kolpakov et al. (Nanotechnologies in Russia, 5 (2010), 160-164) used a pulsed vacuum arc method to deposit ta-C coatings doped with nitrogen, tungsten, or aluminum. The level of dopants such as tungsten is not mentioned, nor is the amount of W in the composite graphite-based cathode. The pulse repetition frequency in the pulsed vacuum arc method was 2.5 Hz. The doped carbon coatings are described as being amorphous in structure and free of crystalline impurities. For W-doped films, microhardnesses HV up to 20 GPa were found. No evidence of pure W droplets was found. The use of arc discharge to produce ta-C has also been described by R. Horsfall, Proc. Soc. Vacuum Coaters (1998), 60-85, who used a DC arc discharge. Pulsed discharge to produce ta-C is also described in V. N. Inkin et al., Diamond and Related Materials 13(2004), 1474-1479. In summary, with unbalanced magnetron sputtering, excluding filters to filter out uncharged particles, the hardness of pure C is rather low, previously reported in the range of 20 GPa, but the addition of substantial levels of dopants, typically in the range of 10 at.% or more, shows an increase in hardness caused by a higher proportion of carbides up to a hardness of about 22.5 GPa, as in the above-mentioned scientific paper by A. Amanov et al. As mentioned above, DLC films have found a variety of applications in science and technology. Amorphous carbon films have also been investigated by several researchers as protective coatings for bipolar plates (BBPs) in polymer electrolyte membrane (PEM) or proton exchange membrane (PEMFC) fuel cells. BPPs have important functions in PEMFCs and PEMFC stacks. For example, they separate individual cells in a PEMFC stack, distribute fuel gas and separate them, act as current collectors, facilitate heat and water removal, provide mechanical support for other components, and serve as the backbone or core component of the FC stack. For example, in the aforementioned scientific paper, Z. Wang et al. deposited W-doped carbon on austenitic stainless steel substrates by CFUBMSIP and studied the interfacial contact resistance (ICR) and corrosion resistance. They observed that W dopant levels between 2.54 at.% and 24.41 at.% increased corrosion resistance. They obtained the best results for 2.54 and 8.56 at.%. They attributed the corrosion protection to the formation of a tungsten oxide layer. In the above-mentioned scientific paper, D. Zhang et al. found that amorphous carbon (aC) films doped with Ag and Cr simultaneously achieved low ICR. Furthermore, their tests showed improved lifetime with less corrosion and less out-diffusion when used as a coating on metallic BPP with dopant levels of 4.89 at.% Ag and 12.37 at.% Cr. P. Yi et al. provide a review of carbon-based coatings for metal BPPs used in PEMFCs in International Journal of Hydrogen Energy 44 (2019), 6813-6843. Transition metal carbide (TMC) coatings are also mentioned in this review article. By way of background, fuel cells and electrolyzers utilizing polymer electrolyte membranes (PEMs) have a bipolar plate (BPP), consisting of two halves welded or glued together at the edges of the plate. Attached between the bipolar plate and the PEM membrane is a gas diffusion layer (GDL). The GDL is typically a woven graphite fiber sheet. On the anode side, the gas diffusion layer should transport electrons to the BPP, and on the cathode side, it should transport electrons from the BPP to the gas. In both cases, the first requirement is a low ICR between the BPP and GDL, providing low electrical resistance to current flow from one side of the BPP to the other. The second requirement is that the bipolar plates do not corrode, leading to an increase in ICR. The third requirement is that the BPP does not release ions that could poison the catalyst near the PEM membrane. These two requirements are related because, in some cases, the released ions could be the product of a corrosion process. Different applications require different life spans: high volume passenger vehicles target operating lives in the range of 5,000 to 8,000 hours. Trucks should reach 20,000-30,000 hours. Preferably, trucks should even reach 50,000 hours of operating life. In applications such as trains and electrolyzers, even longer life is required. The protection that the coating provides depends on the fuel cell voltage. Higher voltages result in more rapid oxidation and out-diffusion. In fuel cell automotive applications, high voltages can also occur due to peak voltages during start-up and shutdown conditions. Therefore, coatings that provide good protection at high fuel cell voltages and varying, unsteady fuel cell voltages are desirable for such applications. Key target data have been established by the USA Department of Energy (https: / / www.energy.gov / eere / fuelcells / doe-technical-targets-polymer-electrolyte-membrane-fuel-cell-components, accessed November 19, 2019) and are as follows: ICR<10mΩ.cm 2 Cathodic corrosion current <0.1μA / cm 2 Anodic corrosion current <1μA / cm 2 To address the three conditions above, coatings have been applied to metallic BBPs, and corrosion simulations indicate that they can be expected to last well in excess of 10,000 hours. Numerous examples of coatings can be found in the literature, ranging from pure gold coatings to metal nitride and carbon coatings. Gold coatings offer low resistivity and good protection, but are expensive. Nitride coatings have limited conductivity. Here, carbon coatings come into consideration. Pure carbon coatings are being applied in small quantities by various suppliers, and the required lifespan appears to be suitable for high-volume automotive applications. Doped carbon coatings are also being developed, as seen above in the scientific papers by Z. Wang et al., D. Zhang et al., and K. Hou et al., and the review paper by P. Yi et al. One drawback of soft carbon coatings on BPP in fuel cells (especially PEMFC) and electrolyzers found by the inventors is that the weakly bonded carbon dissolves under longer lifetimes (>10,000 hours) and harsh operating conditions (high operating voltages), such as those found in fuel cells and electrolyzers for trucks and trains. More specifically, the weakly bonded carbon dissolves over time, ultimately resulting in a loss of corrosion protection and electrical conductivity. Therefore, there was a need for an alternative and further improved coating that avoids the above problems in the coatings for BPP known from the literature and in particular meets all three of the above requirements for BPP, i.e., low ICR, no corrosion, and no ion release, even when used in electrolyzers or fuel cells under the longer lifespan of fuel cells typical of trucks and trains. In view of the prior art, there was a particular need for a coating for metallic BPP that further reduces ICR and further improves corrosion resistance, even when used in electrolyzers or fuel cells under the harsh operating conditions summarized above. DLC coatings have also found applications in industry to reduce friction and improve wear resistance. Hydrogenated diamond-like carbon (aC:H) has typically been applied for this purpose. The limitations of hydrogenated diamond-like carbon are its hardness, which ranges from 20 to 35 GPa, and its application temperature, which is limited to approximately 300°C. A further limitation is that low friction depends on the presence of water. To improve temperature stability, 4d, 5d, or 6d transition metal elements, such as W, are added to hydrogenated DLC. An additional benefit of adding easily oxidizable transition elements, such as W, Ta, or V, to hydrogenated DLC is that the metal reacts with oxygen, adding lubricity at higher temperatures. The aforementioned scientific paper by A. Abou Gharam et al. showed that the addition of W to hydrogenated DLC resulted in a reduction in the friction coefficient in the temperature range of 400 to 500°C. The problem is that the addition of dopants, such as W, to hydrogenated DLC is only achieved at the atomic level by sputtering. Physical and chemical vapor deposition coatings (CVD, including PVD and plasma-assisted chemical vapor deposition, PACVD) and derivatives are used to enhance the performance of substrates. Performance improvements can be directed, for example, toward improved wear resistance or reduced friction. To tailor coating properties, coatings can be doped with additional elements, altering their composition, texture, and internal stress levels. In general, doped coatings have been widely applied over the past 20 years. Many experiments have been conducted on the performance of transition metal-doped hydrogenated DLC (aC:H:Me); for an overview, see S. Yazawa et al., Lubricants 2 (2014), 90-112. A beneficial effect of transition metal additions is that elements such as W and Mo generally form sulfides with good lubricating properties. This may play a role for both fuel-contacting elements and lubricant-contacting elements, as these may contain sulfur. Furthermore, W has been shown to have a beneficial effect on the friction coefficient of contacts lubricated with molybdenum dithiocarbamate (MODTC). A reference to the beneficial effect of tungsten incorporated into DLC layers on different lubricants is given in B. Vengudusamy et al., Tribology International 54 (2012), 68-76. International Publication No. WO 2014 / 000994 describes the use of organometallic precursors to form hydrogenated diamond-like carbon coatings of W. The main reason for this application was to increase the deposition rate during the process. In all of the above methods, the transition metal elements are in the plasma phase at the atomic level, and the atoms reach the surface primarily as single atoms. Because a parallel flow of carbon is present in the above experiments, the transition metal is atomically incorporated into the aC:H coating, thereby bonding the transition metal to carbon and forming carbides. For example, a detailed description of Ti incorporation into hydrogenated DLC (aC:H) is given in WJ Meng et al., J. of Appl. Phys. 88, (2000), 2415-2422. In summary, there is great interest in the industry to provide coating materials with superior properties as coatings for BPPs in fuel cells, especially PEMFCs, that offer the above-mentioned advantages for longer life and harsh conditions during operation as well as electrolyzers. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2014 / 000994 [Non-patent literature]

[0004] [Non-Patent Document 1] J.Robertson.Materials Science and Engineering R 37,2002,129-281 [Non-patent document 2] S.Xu et al.Philosophical Magazine Part B,76:3,351-361 [Non-patent document 3] A.Abou Gharam et al.Surface and Coatings Technology 206(2011),1905-1912 [Non-patent document 4] Z. Wang et al., International Journal of Hydrogen Energy 42(2016), 5783-5792 [Non-Patent Document 5] K. Hou et al. [Non-patent document 6] ACFerrari.Diamond and Related Materials 11(2002),1053-1061 [Non-Patent Document 7] D. Zhang et al.(Carbon 145(2019),333-344) [Non-patent document 8] M.Andersson et al.(Vacuum 86(2012),1408-1416)

Non-licensed literature 9

Non-licensed literature 10

Non-licensed Document 11

Non-licensed Document 12

Non-licensed Document 13

Non-licensed Document 14

Non-licensed Document 15

Non-licensed Document 16

Non-licensed Document 17

Non-licensed Document 18

[0005] Non-hydrogenated transition metal doped DLC The present application fulfills these needs by providing a non-hydrogenated transition metal-doped diamond-like carbon (DLC) as defined in claim 1. Accordingly, the non-hydrogenated DLC comprises at least one transition metal selected from groups 4d, 5d, and 6d of the periodic table, with a portion of the at least one transition metal being present in the form of a carbide of the at least one transition metal in the non-hydrogenated DLC matrix. The non-hydrogenated transition metal-doped DLC is characterized by having a hardness of 35 GPa or more, preferably 40 GPa or more, measured on a film of the non-hydrogenated transition metal-doped DLC deposited on a polished substrate at an indentation depth of less than 10% of the film thickness. In this application, the non-hydrogenated transition metal doped DLC according to claim 1 for use in the present invention may be referred to for brevity as "doped DLC according to the present invention." The transition metal selected from groups 4d, 5d and 6d present in the doped DLC according to the present invention may be referred to herein for brevity as "transition metal" or abbreviated as "TM." The preamble of claim 1 is formulated taking into consideration the scientific papers by Z. Wang et al., in which amorphous WC was observed, and by K. Hou et al., in which NbC phase was observed. However, the W-doped carbon films of Z. Wang et al. and the Nb-doped carbon films of K. Hou et al. have hardnesses of less than 35 GPa. If the hardness is less than 35 GPa, the coating is too soft and will not tolerate the weakly bonded, i.e. sp, properties when used as a coating for BPP in fuel cells, especially for longer lifespans and harsh operating conditions such as high voltages during operation. 2The bonded carbon atoms dissolve, leach, and diffuse out of the coating, with an associated undesirable increase in ICR. For example, the hardness of the doped DLC according to the present invention can be in the range of 40-60 GPa. The hardness of the doped DLC according to the present invention is more preferably ≧45 GPa, which is the hardness of the leached sp 2 It is even more beneficial to avoid increased ICR due to catalyst poisoning by bound carbon and out-diffusion. The hardness of the doped DLC according to the present invention is reported in GPa in this application. The hardness is measured by nanoindentation of a Vickers square pyramid indenter on a film of non-hydrogenated transition metal-doped DLC deposited on a flat, polished, hardened substrate with an indentation depth of less than 10% of the film thickness. The flat, polished, hardened substrate has a surface roughness Ra of 0.01 μm and Rz of 0.25 μm. The hardness of the flat, polished, hardened substrate used was 83.6 HRa (Rockwell hardness A, HRA), 62.1 HRC (Rockwell hardness C, HRC), and 747 HV10 (Vickers hardness at a load of 10 kgf). Details of the hardness measurement are described in the "Description of the Invention" section of this application. The doped DLC according to the present invention has a much higher hardness than the BPP coatings known from the literature such as Z. Wang et al. and K. Hou et al., which is due to the higher proportion of sp, typically above 60%. 3 In non-hydrogenated DLC, the hardness is 3 Fractions, i.e., sp, sp 2 and sp 3 Hybridization state, i.e., sp, sp 2 and sp 3 sp in terms of the total carbon atoms in the bonded carbon material 3 The doped DLC according to the present invention typically has an sp ≥ 60%, preferably ≥ 70%, more preferably ≥ 80%, and most preferably ≥ 85%. 3 It has a fraction. Its high sp 3Due to the fraction, the doped DLC according to the present invention can be represented as ta-C, i.e., tetrahedral amorphous carbon. 3 Due to the higher proportion of bonded carbon, the weakly bonded carbon will not dissolve when the doped DLC according to the present invention is used as a coating on a BPP, resulting in a longer lifespan, e.g., greater than 10,000 hours, and harsh operating conditions such as high voltages, e.g., those encountered in fuel cells for trucks or trains, and electrolyzers. The doped DLC according to the present invention is excellent as a coating to reduce friction and / or improve the wear resistance of the surface to which it is applied because of its high hardness and because it is a non-hydrogenated DLC. Raman spectroscopy was used to identify the sp 3 The fraction and the presence of transition metal(s) carbides, i.e., the presence of TM-C bonds in the doped DLC according to the present invention, can be evaluated. For example, when TM is W, the presence of TM-C bonds can be evaluated by measuring the peak intensity of 80-150 cm for an excitation wavelength of 532 nm. -1 It can be detected in the Raman spectrum by peaks in the range of sp 3 Regarding the fraction, the position of the G peak in Raman spectroscopy and the intensity of the D peak (I D ) to the intensity of the G peak (IG), i.e., I D / I G From sp 3 This is described by A.C. Ferrari in the aforementioned scientific paper and is shown in Figure 2 of that paper. The Raman excitation wavelength used here was 532 nm. For the doped DLC samples according to the present invention, the position of the G peak is typically at 1,605 cm -1 At this excitation wavelength, more than 60% of the carbon atoms are sp 3 Refers to a percentage. In the present invention, the compositional analysis of the doped DLC and material coatings according to the present invention is preferably carried out by electron probe microanalysis (EPMA). In particular, the content of at least one transition metal in the doped DLC according to the present invention can be determined by EPMA. When a doped DLC according to the present invention is designated as "non-hydrogenated," this means that hydrogen is not intentionally added during deposition, particularly that no significant amount of hydrogen is added during deposition of the material. However, some hydrogen coming from water vapor in the system may be present and incorporated into the doped DLC according to the present invention. Thus, with such a hydrogen source, the "non-hydrogenated" doped DLC according to the present invention may have a hydrogen content of less than 1 at.% resulting from water vapor in the system, particularly when, for example, the material is deposited with high productivity and short pumping and heating times of the deposition chamber. For the same reasons as described above for hydrogen, small amounts of oxygen from water vapor and trace air in the system, and small amounts of argon from trace air and inert gas atmospheres in the system, may be present and incorporated into the doped DLC of the present invention. Therefore, with such oxygen and argon sources, the non-hydrogenated doped DLC of the present invention may have an oxygen content of less than 1 at.% and an argon content of less than 1 at.% from water vapor, trace air, and inert gas atmospheres in the system, particularly, for example, when the material is deposited with high productivity and short pumping and heating times of the deposition chamber. Preferably, the oxygen content is less than 0.5 at.% and the argon content is less than 0.5 at.%. Most preferably, the oxygen content is less than 0.1 at.% and the argon content is less than 0.1 at.%. The doped DLC of the present invention is doped with at least one transition metal. The transition metal is selected from groups 4d, 5d, and 6d of the periodic table of elements. Therefore, the at least one transition metal is selected from the group consisting of titanium (Ti), zirconium (Zr), hafnium (Hf), vanadium (V), niobium (Nb), tantalum (Ta), chromium (Cr), molybdenum (Mo), and tungsten (W). Preferably, the at least one transition metal is selected from the group consisting of chromium, molybdenum, and tungsten. Most preferably, it is tungsten. All of the above transition metals used in the present invention can form carbides. Therefore, they can all be present in the form of carbides in the non-hydrogenated DLC matrix, as required by the present invention. The content of the at least one transition metal is not particularly limited, but according to a preferred embodiment, it is in the range of 0.1 to 5 at.% for non-hydrogenated transition metal-doped DLC. According to a preferred embodiment, the content of the at least one transition metal in the doped DLC according to the present invention is 0.2 to 2.5 at.%, more preferably 0.3 to 2.0 at.%, most preferably 0.5 to 1.5 at.% or 1 to 5 at.%, more preferably 2 to 4 at.%. When the doped DLC according to the invention is for use as a coating on a bipolar plate of a fuel cell or electrolyzer, the content of the at least one transition metal (also referred to herein as dopant level) is preferably 0.1 to 2.5 at.%, more preferably 0.3 to 2.0 at.%, most preferably 0.5 to 1.5 at.%, for the non-hydrogenated transition metal doped DLC, i.e. in the coating or relevant layer within the coating, which may be a multilayer coating. The non-hydrogenated transition metal-doped DLC of the present invention is preferably a homogeneous material. This is because the cathodic arc discharge deposition method of the present invention, which can produce doped DLC of the present invention, allows for the deposition of homogeneous coatings. This is due to the fact that a carbon target doped with at least one transition metal is used as the target in the cathodic arc discharge deposition in the present invention. Literature examples using CFUBMSIP (e.g., Z. Wang and K. Hou in the aforementioned scientific paper) use cathodes with pure carbon and pure metal, which are either facing each other in the deposition chamber (in the case of a two-cathode system) or at 90 degrees (in the case of a four-cathode system). Due to the rotation of the substrate table, the concentration of the transition metal exhibits layer-by-layer modulation of the transition metal content, with sizes on the order of 2-30 nm, depending on the rotation speed, deposition rate, and equipment configuration. This is because, for example, when a TM target and two graphite targets are used in the CFUBMSIP method (e.g., as used by Z. Wang and K. Hou in the aforementioned scientific paper), the TM content of the coating is higher facing the TM target during rotation and slightly lower facing the graphite target. In the doped DLC coating according to the present invention, the above-mentioned nanoscale layer-by-layer modulation of the transition metal content across the coating thickness, especially on the order of 2-30 nm, can be avoided. That is, in the doped DLC according to the present invention, the at least one transition metal can be distributed uniformly throughout the coating, especially across the coating thickness (even if present as carbides or metal droplets). In the present invention, at least one transition metal in the form of a carbide is present in the non-hydrogenated DLC matrix, i.e., preferably uniformly distributed. The size of the TM carbide phase or domain is not particularly limited. The carbides may exist as atomically distributed WC units at the lower end and as islands, particularly those with sizes in the nm range, i.e., "nano-sized islands," at the upper end. As used herein, the nanometer range, i.e., nano-sized, is defined to include sizes between 0.1 nm and 100 nm. According to a preferred embodiment, the TM carbides are present atomically distributed down to nano-sized islands, more preferably nano-sized islands with a maximum size of 2 nm. Even more preferably, the carbide of at least one transition metal is present as nano-sized islands with a size of about 0.5 nm to 2 nm. This has proven beneficial for meeting the stringent requirements of BPP coatings, particularly to ensure that the doped DLC according to the present invention contains covalently bonded transition metals, resulting in the desired low ICR. At the same time, the desired high hardness is achieved by the presence of at least a portion of the carbides of at least one transition metal as nano-sized islands with a size of up to 2 nm. In other words, the presence of at least one carbide of a transition metal as islands with a size of up to 2 nm in a non-hydrogenated DLC matrix makes it possible to achieve a DLC coating that combines low ICR with high hardness. Typically, very hard ta-C coatings have a high ICR. This can be explained as follows: On the one hand, a higher proportion of sp 3 Bound carbon usually leads to improved hardness of the DLC coating. On the other hand, higher sp 3 A higher fraction usually leads to a lower conductivity of the DLC coating and therefore a higher ICR. In the present invention, the size of the transition metal carbide islands can be determined by transmission electron microscopy (TEM), in particular bright-field TEM (BFTEM) and high-angle annular dark field with spot for scanning TEM (HAADF-STEM). According to a preferred embodiment, another portion of the at least one transition metal, preferably the other portion, is present in the form of metal droplets, which preferably have a diameter of less than 1 μm, preferably between 0.1 and 100 nm, preferably between 0.5 and 40 nm. Such small TM droplets distributed in the matrix of the doped DLC according to the invention and thus embedded in the coating of the material can form a particularly effective source of free transition metal and therefore further contribute to a low ICR when the material is used as a coating on the BPP of a fuel cell or electrolyzer. In order to distinguish between transition metals present in the form of carbides and metals present in the form of droplets in the non-hydrogenated DLC matrix, transmission electron microscopy (TEM) studies using bright-field TEM (BFTEM) and high-angle annular dark-field scanning TEM (HAADF-STEM) were carried out in the present invention. The combination of BFTEM and HAADF-STEM in conjunction with Raman spectroscopy allows for differentiation between the two forms of transition metals present in the non-hydrogenated DLC matrix in the material according to the present invention. Furthermore, the size of the metal droplets could be determined by TEM, particularly the BFTEM and HAADF-STEM of the present invention. Furthermore, for the specific embodiment below, it is shown that W, which has a melting point above 3,400° C., can form metal droplets in a cathodic arc discharge deposition method according to the present invention in a non-hydrogenated DLC matrix according to a preferred embodiment of the present invention. This is therefore also possible for the other transition metals mentioned above for use in the present invention, which have lower melting points. According to a preferred embodiment, the portion of the at least one transition metal present in the form of a carbide in the doped DLC according to the present invention can be present in an amount of 60 at.% or less, more preferably 50 at.% or less, even more preferably 40 at.% or less, even more preferably 30 at.% or less, and most preferably 20 at.% or less, relative to the total content of transition metals in the material. As found by the inventors, the doped DLC according to the present invention contains almost no "free" transition metal (or isolated transition metal), in the sense that it is not in the form of a carbide (e.g., in the form of carbide islands) or in the form of metal droplets. That is, according to a preferred embodiment of the doped DLC according to the present invention, a total of ≥ 85 at.%, preferably ≥ 90 at.%, more preferably ≥ 95 at.% of the at least one transition metal is present in the matrix of the non-hydrogenated DLC in the form of a carbide (preferably as carbide islands) and / or in the form of metal droplets. It has therefore been found that the remaining part of the at least one transition metal, i.e. the above-mentioned percentage difference of the amount of transition metal present in the form of carbides up to 100%, is preferably present in the form of metal droplets of the transition metal. Thus, in a preferred embodiment of the present invention, the above-mentioned proportion of transition metal in the form of carbides and the proportion of transition metal in the metal droplets add up to 100% of the transition metal present in the doped DLC according to the present invention. Layer System The doped DLC of the present invention can form a layer system comprising at least one layer provided on a substrate. The substrate is preferably a metal substrate, particularly a stainless steel substrate or a titanium substrate, especially an ultra-thin sheet stainless steel, which has higher electrical conductivity, better manufacturability, and lower cost compared to conventional BPP graphite materials. The thickness of one or more layers of the doped DLC of the present invention present in the coating and layer system of the doped DLC of the present invention can be measured by SEM in the present invention. BPP coating applications in fuel cells and electrolyzers The doped DLC according to the present invention is useful as a coating on a bipolar plate of a fuel cell or electrolyzer due to its excellent combination of low ICR, high corrosion resistance, and no ion release even under harsh operating conditions. Therefore, the present invention also relates to the use of the doped DLC according to the present invention as a coating on a bipolar plate in a fuel cell or electrolyzer. Furthermore, the present invention relates to a proton exchange membrane fuel cell (PEMFC) or electrolyzer comprising a metal BPP having a coating of the doped DLC according to the present invention. The present invention also relates to a fuel cell comprising a proton-conducting polymer electrolyte membrane having a first surface and a second surface opposite the first surface, an anode formed on the first surface, and a cathode formed on the second surface. The anode side is provided with, in order from the first surface of the proton-conducting polymer electrolyte membrane, a gas diffusion layer, a metal bipolar plate, and a device for supplying hydrogen. The cathode side is provided with, in order from the second surface of the proton-conducting polymer electrolyte membrane, a gas diffusion layer, a metal bipolar plate, an oxygen-containing gas supply device, and a device for removing water formed during fuel cell operation. The anode and cathode are electrically connected via their corresponding metal BPPs, i.e., via the metal BPP on the anode side and via the metal BPP on the cathode side. At least one BPP, preferably both BPPs, is coated with the doped DLC according to the present invention. Thus, each metal BPP-containing coating constitutes a layer system according to the claims, comprising at least one layer of the doped DLC of the present invention provided on a substrate. The substrate is preferably a steel, titanium, or aluminum substrate, more preferably a stainless steel or Ti substrate. The thickness of the substrate, particularly the stainless steel or Ti substrate, may be as low as 0.05-0.1 mm. The substrate may be cleaned by ion bombardment to remove native oxides from the substrate surface. Ion bombardment of the substrate before depositing further layers thereon promotes adhesion of the further layers and reduces ICR. For example, the substrate can be cleaned by Ar etching, i.e., argon ion bombardment. In the layer system, an adhesion layer can be provided directly on the substrate, on which at least one layer of the doped DLC of the present invention can be formed. The adhesion layer can be, for example, a layer of metal Cr or metal Ti. The layer system can also include a multilayer, each of which includes at least one layer of the doped DLC of the present invention. Preferably, the layer system is a multilayer of at least one layer of non-hydrogenated transition metal-doped DLC, in which the content of at least one transition metal is X at.% per layer, and at least one layer of non-hydrogenated transition metal-doped DLC, in which the content of at least one transition metal is 0-0.8 times X at.% per layer and / or at least one layer of ta-C. Thus, according to a preferred embodiment, the multilayer has at least two layers, one with a higher transition metal content and one with a lower transition metal content (or no transition metal at all, as is the case with ta-C layers). The multilayer may contain more than two alternating layers of high transition metal and low / no transition metal. The layer system can also include a transition layer between an adhesion layer and a single layer of the inventive doped DLC or a multilayer of or including the inventive doped DLC described above, by ramping down the adhesion metal layer and ramping up the inventive doped DLC or ta-C layer. When the layer system according to the invention, in particular for use as a BPP in fuel cells and electrolyzers, comprises a single layer of the doped DLC according to the invention, the layer thickness is preferably in the range of 50 nm to 3 μm, preferably in the range of 80 nm to 1 μm. This also applies when the individual layers of the doped DLC according to the invention are present as a multilayer. The multilayer preferably has a thickness in the range of 0.1 to 30 μm, more preferably 0.2 μm to 10 μm. This excludes the thickness of any adhesive and transition layers present on the substrate, as well as, of course, the thickness of the substrate itself. Deposition method It can be denoted as "ta-C" and typically has over 60% sp 3 The non-hydrogenated DLC having a high hardness of ≥ 35 GPa, preferably ≥ 40 GPa, according to claim 1, can only be produced when a high plasma density, i.e., a high degree of ionization, is present. Therefore, the doped DLC according to the present invention cannot be obtained by standard unbalanced magnetron sputtering, such as that employed by Z. Wang et al. and K. Hou et al. Unlike high-power impulse magnetron sputtering (HIPIMS), the degree of ionization is too low for standard unbalanced magnetron sputtering. The present invention further provides a method for depositing a coating of non-hydrogenated DLC containing at least one transition metal selected from groups 4d, 5d, and 6d of the periodic table of elements. The method is a cathodic arc discharge deposition method. In the cathodic arc discharge, a pulsed current is superimposed on a direct current (DC). The pulsed current has a pulse frequency in the range of 10 kHz to 100 kHz. A carbon target doped with at least one transition metal is used as the target in the cathodic arc discharge. The target is directly connected to the cathode. Each pulse of the pulsed current induces a voltage rise measured at the cathode at a rate exceeding 5 V / μs. Each pulse of the pulsed current has an active pulse width of less than 30 μs. In this method, the ionization degree of the evaporated target material approaches 100%. The dopant level of the target can be 0.5 at.% to 10.0 at.%, preferably 1.0 at.% to 8 at.%, more preferably 1.0 at.% to 6 at.%, and more preferably 0.5 at.% to 2.5 at.% of at least one transition metal selected from Groups 4d, 5d, and 6d of the Periodic Table of Elements. The at least one transition metal can be a transition metal mentioned above, such as tungsten (W). Using electron probe microanalysis (EPMA), the inventors have found that approximately 60% of the transition metal present in the target as a dopant is found in the doped DLC of the present invention. For example, 8 at.% W dopant in the target provides a 5 at.% W dopant in the coating of the doped DLC of the present invention, and 2 at.% W in the target provides a 1.2 at.% W dopant in the coating. Cathodic arc discharge deposition is a physical vapor deposition (PVD) technique in which an electric arc is generated between an anode and a target that acts as or is connected to the cathode. The electric arc vaporizes material on the surface of the target in the area where the arc is present. The vaporized target material is deposited on a substrate to form a coating of the target material on the substrate. A. Anders, in the textbook "Cathodic ARCs", Springer, 2008, ISBN 978-0-387-79107-4, provides a detailed introduction to cathodic arc discharge deposition. The difference between cathodic arc and standard unbalanced magnetron sputtering is that the ionization of the deposited atoms is much higher in arc discharge than in unbalanced magnetron discharge. The ionization of carbon (C) in DC arcs is described by A. Anders in "Cathodic ARCs", Springer (2008), ISBN 978-0-387-79107-4, paragraph 4.3 on pages 194-195, and is already 100% singly ionized for arc currents of 200 A, and partially doubly ionized at higher currents. The ionization of carbon in magnetron sputtering is much lower, with typical values ​​for C sputtering in the 5% range.3 The higher fraction of ta-C is attributed to a sub-implantation process by Lifschitz et al. (Physical Review B, Vol. 41, No. 15, pp. 10468-10480). Carbon atoms reach high enough energy to penetrate the surface of the growing film to a depth of typically three atomic layers. At this depth, the implanted atoms experience high pressure, which causes sp 3 In a standard (unbalanced) magnetron (non-HIPIMS), the average energy is too low due to the low degree of ionization to have sub-implantation for most of the C atoms. In this cathodic arc discharge deposition method, a cathodic arc discharge is generated or powered by supplying a DC current superimposed with a pulsed current. This superimposition of current significantly reduces the generation of macroparticles of target material, i.e., carbon doped with at least one transition metal, during the cathodic arc discharge process. Furthermore, the formation of relatively deep craters with sharp edges on the target surface can be minimized. In particular, the pulses superimposed on the DC current split the electric arc into multiple arcs, causing the first arc to erupt and round the edges through evaporation in the process, thus avoiding the formation of craters with sharp edges, especially deep craters. Such sharp edges can be ejected from the target as macroparticles of target material, thereby affecting the quality of the resulting coating. In this method, the pulses of the pulsed current have a high rise rate. Therefore, the split into multiple arcs occurs within a particularly short time. Therefore, high quality coatings can be achieved, and since the formation of such macroparticles can be suppressed, the deposition method does not need to use macroparticle filters, which greatly simplifies the process. The target is directly connected to the cathode, ie, there are no intermediate layers or structures between the target and the cathode. The pulse current has a pulse frequency in the range of 10 kHz to 100 kHz, preferably in the range of 20 kHz to 90 kHz, more preferably in the range of 30 kHz to 80 kHz, and even more preferably in the range of 40 kHz to 70 kHz. By selecting a pulse frequency in the range of 10 kHz to 100 kHz, the formation of macroparticles of the target material can be particularly efficiently and reliably suppressed. In manufacturing applications, cathodic arc discharge deposition processes can be performed in deposition chambers, especially vacuum chambers, with multiple cathodes. For example, if there is a bank of cathodes depositing simultaneously, the arc pulses can be synchronized with delays set between the different arc sources so that only one pulse occurs at a time, avoiding overloading the bias voltage power supply. In "Production of Highly Ionized Species in High-Current Pulsed Cathodic Arcs" (R. Sangines, A.M. Israel, I.S. Falconer, D.R. McKenzie, and M.M. Bielek, Applied Physics Letters 96, 221501

[2010] ), Figure 2 shows that in pulsed arc deposition of Al with a fairly long pulse of 600 μs and a peak current of 800 A, doubly ionized Al is formed but reaches its maximum value after 40 μs. It then rapidly decreases, leaving only singly ionized Al. The arcs are continuously splitting in approximately 60 A increments. The arcs repel each other. The fact that doubly ionized Al is significantly reduced after 100 μs means that after that period, the plasma resembles a DC arc. To stay in the regime where higher ionization occurs and does not recombine with the neutral and return to standard DC arc conditions, the method uses a high pulse voltage rise rate so that all arcs are still close together and the active pulse width of each pulse of pulsed current is kept relatively short: the maximum time that the arc voltage still rises is less than 30 μs. Each pulse of pulsed current superimposed on the DC current induces a voltage rise at the cathode, i.e., the arc discharge voltage, at a rate exceeding 5 V / μs. The voltage can be measured between the cathode and the anode, which can be at ground potential. The use of pulses that induce such a high voltage rise rate at the cathode, to which the target is directly connected, allows the generation of plasma with particularly high plasma intensity, plasma density, and plasma temperature. Generating such a high-density plasma results in the formation of molten transition metal droplets (droplets) on the surface of the target. Furthermore, the high-density plasma ensures that larger transition metal droplets ejected from the target surface are broken down into smaller droplets during their transfer from the target to the substrate to be coated, thereby reducing the droplet size. This further improves the quality of the coating. In particular, incorporating such smaller droplets into the coating can significantly increase the coating's conductivity, which is particularly beneficial for coatings on bipolar plates in fuel cells or electrolyzers, for example. Furthermore, the droplets can improve the friction-reducing properties of the coating. The droplet size reduction achieved by the high-density plasma during droplet transfer from the target to the substrate can prevent the incorporation of large droplets into the coating, thereby resulting in a coating with a smooth and uniform surface structure. Preferably, the rate of rise of the voltage measured at the cathode induced by the pulse is greater than 8 V / μs, more preferably greater than 10 V / μs, even more preferably greater than 12 V / μs, even more preferably greater than 14 V / μs, even more preferably greater than 16 V / μs. Each pulse of the pulsed current superimposed on the DC current has an active pulse width of less than 30 μs. The active pulse width of a pulse is defined as the time during which the arc current induced by the pulse has not yet decayed. The arc current induced by the pulse is measured at the cathode. The arc current induced by the pulse may begin to decay when the pulsed power supply supplying the pulse is turned off. By shutting off the pulse after such a relatively short time (less than 30 μs), it is possible to reliably prevent the arc from splitting into a regime where the distance between the arcs is greater and the plasma density falls to approximately that of a DC arc plasma. The pulse of the pulse current may have an active pulse width of 1 μs or more and less than 30 μs, preferably in the range of 2 μs to 20 μs, more preferably in the range of 4 μs to 10 μs, and particularly preferably in the range of 5 μs. The DC current may be in the range of 50 A to 1000 A, preferably in the range of 100 A to 800 A, more preferably in the range of 200 A to 600 A, and even more preferably in the range of 250 A to 500 A. Selecting a DC current in the range of 50 A to 1000 A can further promote the formation of molten transition metal droplets on the target surface. The peak current of the DC current superimposed with the pulsed current may be greater than 200 A. At the cathode, the DC current superimposed with the pulsed current is measured. By selecting a peak current greater than 200 A, the droplet size can be reduced by the high-density plasma during droplet transfer from the target to the substrate in a particularly efficient manner. The pulses of the pulsed current may have a pulse separation in the range of 20 μs to 200 μs, preferably in the range of 40 μs to 150 μs, more preferably in the range of 60 μs to 120 μs, and most preferably in the range of 80 μs. The method of the present invention can be carried out at a deposition temperature in the range of 50°C to 180°C, preferably in the range of 70°C to 150°C. The deposition temperature is measured at the substrate to be coated. Particularly preferably, the deposition temperature is maintained between 70°C and 150°C. By controlling the deposition temperature so that it does not exceed 150°C, it is particularly reliably ensured that the Young's modulus and hardness of the coating are not significantly reduced. A bias may be applied to the substrate in the cathodic arc discharge deposition process. The bias may be in the range of 10 V to 100 V, preferably in the range of 20 V to 80 V, and more preferably in the range of 40 V to 60 V. Particularly preferably, the bias may be 50 V. The background pressure in the cathodic arc discharge deposition process is 1x10 -5 mbar~5x10 -4 mbar range, preferably 2x10 -5 mbar~1x10 -4 mbar range, more preferably 4x10 -5 mbar~6x10 -5 Particularly preferably, the background pressure in the cathodic arc discharge deposition process is in the range of 5×10 -5 It can be mbar. The cathodic arc discharge deposition process may be carried out in an atmosphere containing an inert gas. More specifically, the cathodic arc discharge deposition process may be carried out in an atmosphere containing argon (Ar) or nitrogen (N). In particular, the cathodic arc discharge deposition process may be carried out in a deposition chamber into which such gases have been introduced. The background pressure of the inert gas, particularly Ar, is preferably 1×10 -4 mbar~9x10 -4 mbar range, preferably 2x10 -4 mbar~8x10 -4 mbar range, more preferably 4x10 -4 mbar~6x10 -4 Particularly preferably, the background pressure of the inert gas, in particular Ar, may be in the range of 5x10 -4By carrying out a cathodic arc discharge deposition process in such an atmosphere, ignition of the electric arc can be facilitated. Prior to depositing the coating, the substrate to be coated may be cleaned, for example by ion etching, using Ar or metal ions. The coating may be deposited directly onto the surface of the substrate, i.e., there may be no intermediate layer between the surface and the coating. Alternatively, an initial adhesion layer may be provided on the surface of the substrate to be coated before the coating is deposited thereon. The initial adhesion layer may be, for example, a metallic chromium (Cr) layer or a metallic titanium (Ti) layer. The method for applying the initial adhesion layer is not limited, and any CVD or PVD method, including, for example, sputtering, may be used. Deposition parameters such as direct and pulsed current characteristics, deposition temperature, substrate bias, gas atmosphere and pressure may be kept at least substantially constant during the cathodic arc discharge deposition process. In some embodiments, the pulsed current can have a pulse frequency in the range of 20 kHz to 90 kHz. The dopant level in the target can be in the range of 0.5 at.% to 10.0 at.%. Each pulse of the pulsed current can induce a voltage rise at a rate of greater than 8 V / μs, measured at the cathode. Each pulse of the pulsed current can have an active pulse width in the range of 2 μs to 20 μs. The peak current can be greater than 200 A. The pulses of the pulsed current can have a pulse separation in the range of 20 μs to 200 μs. The method can be carried out at a deposition temperature in the range of 50° C. to 180° C. In the cathodic arc discharge deposition process, a bias in the range of 10 V to 100 V can be applied to the substrate. The background pressure in the cathodic arc discharge deposition process can be greater than 1×10 -5 mbar~5x10 -4 It may be in the mbar range. The coating deposited by the method of the present invention may be a coating of doped DLC according to the present invention. The coating deposited by the method of the present invention may have the characteristics, properties and features described above. The inventors have found that the higher the content of transition metals, such as tungsten, in the target, the more likely they are to appear as metal droplets in the doped DLC coating of the present invention. This is similar to the target surface, where large particles of molten transition metal separate along the arc track under the influence of the arc. When the arc strikes such transition metal particles, the high-density plasma caused by the high voltage rise rate of over 10 V / μs causes the droplets to be ejected and break up into smaller droplets, while the droplets migrate from the target to the substrate being coated, reducing their size. The proportion of transition metals, such as W, that are isolated on the target surface as metal droplets is more than linearly proportional to the target dopant level. In this way, the relative proportions of transition metals present in the form of carbides and as metal droplets in the doped DLC of the present invention can be adjusted. Figure 11 shows an example of a target exposed to an 8 at.% W arc. Small craters can be seen on the target surface. The white droplets are W. The present invention further provides a non-hydrogenated transition metal doped DLC coating obtainable by the method of the present invention. Embodiment Specific embodiments of the present invention are summarized below. (1) Non-hydrogenated transition metal-doped diamond-like carbon (DLC), wherein the non-hydrogenated transition metal-doped DLC comprises at least one transition metal selected from Groups 4d, 5d, and 6d of the periodic table of elements, a portion of the at least one transition metal being present in the form of a carbide of the at least one transition metal in the non-hydrogenated DLC matrix, and the non-hydrogenated transition metal-doped DLC has a hardness of 35 GPa or more, preferably 40 GPa or more. The hardness can be measured on a film of the non-hydrogenated transition metal-doped DLC deposited on a polished substrate at an indentation depth of less than 10% of the thickness of the film. (2) The non-hydrogenated transition metal-doped DLC according to item (1), wherein at least a portion of the carbide of at least one transition metal is present as islands in the non-hydrogenated DLC as a matrix. (3) The non-hydrogenated transition metal-doped DLC according to item (2), wherein the islands have a size of 2 nm or less. (4) The non-hydrogenated transition metal-doped DLC according to any one of items (1) to (3), wherein the non-hydrogenated DLC is tetrahedral amorphous carbon, i.e., ta-C. (5) The non-hydrogenated transition metal-doped DLC according to any one of items (1) to (4), wherein another portion of the at least one transition metal is present in metallic form as droplets of the transition metal. (6) The non-hydrogenated transition metal doped DLC according to item (5), wherein the droplets of the transition metal have a diameter of less than 1 μm, preferably 0.1 to 100 nm, preferably 0.5 to 40 nm. (7) A non-hydrogenated transition metal-doped DLC according to item (5) or (6), wherein a total of at least 85 at.%, preferably at least 90 at.%, of the at least one transition metal is present in the matrix of the non-hydrogenated DLC in the form of carbides, preferably as carbide islands and / or in the form of metal droplets. (8) sp carbon atom 3 The non-hydrogenated transition metal-doped DLC according to any one of items (1) to (7), wherein the fraction is 60% or more, preferably 70% or more, more preferably 80% or more, and most preferably 85% or more. (9) The non-hydrogenated transition metal-doped DLC according to any one of items (1) to (8), wherein the transition metal is selected from the group consisting of chromium, molybdenum, and tungsten, and preferably tungsten. (10) The non-hydrogenated transition metal-doped DLC according to any one of items (1) to (9), wherein the content of at least one transition metal is in the range of 0.1 to 5 at.%, preferably in the range of 0.2 to 2.5 at.%, more preferably in the range of 0.3 to 2.0 at.%, and most preferably in the range of 0.5 to 1.5 at.% for the non-hydrogenated transition metal-doped DLC. (11) The non-hydrogenated transition metal-doped DLC according to any one of (1) to (10), which has a hardness in the range of 40 GPa to 60 GPa. (12) A layer system comprising at least one layer of the non-hydrogenated transition metal-doped DLC according to any one of items (1) to (11) provided on a substrate. (13) The layer system according to item (12), wherein the layer is a homogeneous layer. (14) The layer system according to item (12) or (13), wherein the layer thickness is in the range of 50 nm to 3 μm. (15) At least one layer of non-hydrogenated transition metal doped DLC, the content of at least one transition metal being X at.% for said layer; At least one layer of non-hydrogenated transition metal doped DLC, the content of at least one transition metal being 0 to 0.8 times X at.% for said layer and / or at least one layer of ta-C; 15. The layer system according to any one of items (12) to (14), which is a multilayer having (16) The layer system according to item (15), wherein the multilayer has a thickness in the range of 0.1 μm to 30 μm, preferably in the range of 0.2 μm to 10 μm. (17) The layer system according to any one of items (12) to (16), wherein an adhesive layer is provided directly on the substrate, and at least one layer of non-hydrogenated transition metal-doped DLC is formed thereon. (18) The layer system according to any one of items (12) to (17), wherein the substrate is a metal substrate. (19) The layer system according to any one of items (12) to (18), wherein the metal substrate is a stainless steel substrate, a titanium substrate or an aluminum substrate. (20) The layer system according to item (19), wherein the metal substrate is a stainless steel substrate. (21) The layer system according to item (19), wherein the metal substrate is a titanium substrate. (22) Use of the non-hydrogenated transition metal doped DLC according to any one of items (1) to (11) as a coating for bipolar plates in a fuel cell or electrolyzer. (23) The use according to item (22), wherein the non-hydrogenated transition metal-doped DLC has a content of at least one transition metal in the range of 0.3 at.% to 2.5 at.%. (24) The use according to item (22) or (23), wherein the coating has a thickness in the range of 50 nm to 3 μm, preferably 0.1 μm to 1 μm. (25) A proton exchange membrane fuel cell or electrolyzer comprising a metal bipolar plate having a coating of the non-hydrogenated transition metal-doped DLC according to any one of items (1) to (11). (26) a proton-conducting polymer electrolyte membrane having a first surface and a second surface opposite the first surface; A fuel cell comprising an anode formed on a first surface and a cathode formed on a second surface, a gas diffusion layer, a metal bipolar plate, and a hydrogen supply device on the anode side, in this order from the first surface; The cathode side is provided with, in order from the second surface side, a gas diffusion layer, a metal bipolar plate, a device for supplying oxygen-containing gas, and a device for removing water formed during operation of the fuel cell; the anode and cathode are electrically connected via their corresponding metallic bipolar plates; A fuel cell, wherein at least one of the bipolar plates has a coating of the non-hydrogenated transition metal-doped DLC according to any one of items (1) to (11). (27) The fuel cell or electrolyzer according to item (25) or (26), wherein the coating of non-hydrogenated transition metal-doped DLC has a thickness in the range of 50 nm to 3 μm, preferably 0.1 μm to 1 μm. (28) A method for depositing a coating of non-hydrogenated DLC containing at least one transition metal selected from Groups 4d, 5d, and 6d of the Periodic Table of the Elements, the method being a cathodic arc discharge deposition method, in which a direct current is superimposed with a pulsed current in the cathodic arc discharge, the pulsed current having a pulse frequency in the range of 10 kHz to 100 kHz, a carbon target doped with at least one transition metal is used as a target in the cathodic arc discharge, the target is directly connected to the cathode, each pulse of the pulsed current induces a voltage rise at a rate of more than 5 V / μs measured at the cathode, and each pulse of the pulsed current has an active pulse width of less than 30 μs. (29) The method according to item (28), wherein the peak current is greater than 200 A. (30) The method according to item (28) or (29), wherein the coating is a coating of the non-hydrogenated transition metal-doped DLC according to any one of items (1) to (11). (31) A coating of non-hydrogenated transition metal-doped DLC according to any one of items (1) to (11), obtainable by the method according to item (28) or (29). [Brief explanation of the drawings]

[0006] [Figure 1] 1 is a cross-sectional schematic diagram illustrating the structure of a proton exchange membrane fuel cell (PEMFC) including a bipolar plate (BPP) coated with doped DLC according to one embodiment of the present invention. [Figure 2] 1 is a cross-sectional view of a cathodic arc discharge deposition apparatus for carrying out a deposition method according to an embodiment of the present invention. [Figure 3] 1 shows Raman spectra of specific examples of doped DLC coatings according to the present invention. The samples had different dopant levels, i.e., transition metal, W content, as follows: B1: 0 at.% (reference); B2: 0.3 at.%; B3: 0.6 at.%; B4: 1.4 at.%; B5: 1.4 at.%. The Raman excitation wavelength was 532 nm with a 50 cm edge filter. [Figure 4] 1 is a scanning electron microscope (SEM) image of 80 nm of W-doped ta-C with a W concentration of 1.4 at.% according to an embodiment of doped DLC according to the present invention. The wafer window with the Si3N4 film is visible, onto which the doped ta-C is deposited. The dark circles are W. [Figure 5] 1 is a high-resolution transmission electron microscope (HRTEM) image in HAADF-STEM mode of ta-C doped with 1.4 at.% W, an example of doped DLC according to the present invention. By focusing, only 20 nm slabs are observed as protrusions. [Figure 6] FIG. 10 shows ICR test results for ta-C with different W dopant percentages, according to certain examples of the present invention. [Figure 7] FIG. 1 shows the corrosion current (Icorr) values ​​of corroded samples coated at 90° C. before and after potentiostatic treatment with an embodiment of doped DLC according to the present invention. [Figure 8] HR-TEM bright field of 80 nm of 5 at. % W-doped ta-C on 20 nm of Si3N4, according to an embodiment of doped DLC according to the present invention. The dark circles are W droplets. [Figure 9] TEM micrograph of 80 nm of 5 at.% W-doped ta-C in dark field mode, according to an embodiment of the doped DLC according to the present invention. The bright spots are pure W droplets embedded in the ta-C matrix. [Figure 10] 1 is a HAADF-STEM micrograph of 5 at.% W-doped ta-C according to an embodiment of the doped DLC according to the present invention. The W appears white when the micrograph is taken in dark field mode. [Figure 11] 1 is an SEM image of a graphite target containing 8 at.% W after use in a cathodic arc discharge deposition process according to the present invention. The white droplets are W. DETAILED DESCRIPTION OF THE INVENTION

[0007] fuel cell A typical structure of a fuel cell in which the doped DLC according to the present invention can be used as a coating will now be described with reference to FIG. The fuel cell 20 shown in Figure 1 provides a schematic representation of a fuel cell, particularly a PEMFC, or a fuel cell stack, particularly an individual fuel cell within a PEMFC stack. The fuel cell shown in Figure 1 comprises a proton-conducting polymer electrolyte membrane (PEM) 22. The membrane 22 may also be referred to as a proton exchange membrane (PEM). As indicated by the arrows, the PEM membrane 22 transfers protons (H + ) can conduct electricity. Typically, the anode and cathode are provided as coatings on the first and second surfaces of the PEM, respectively, and these coatings contain an electrocatalyst. The fuel cell shown in FIG. 1 further comprises gas diffusion layers 30, i.e., a GDL on the anode 30a side and a GDL on the cathode 30b side. The fuel cell further comprises bipolar plates 28, preferably made of metal, in particular stainless steel, titanium, or aluminum, in particular stainless steel or Ti. A BPP is located on the anode 28a side, and a BPP is located on the cathode 28b side. Hydrogen (H2) and oxygen (O2) (preferably supplied in the form of air) are used as fuel gas, as indicated by the corresponding arrows in the figure. The fuel gas flows through the space where the GDL is located, and with the help of the BPP, which has a profile with valleys from top to bottom, it allows the gas to reach the entire area where the GDL is attached. In the BPP on the cathode side 28b, channels also allow water (usually in the form of water vapor) formed during fuel cell operation to be removed. As shown in Figure 1, the anode 24 and the cathode 26 are electrically connected through the metal BPP via a wire 32. As shown in Figure 1, the current (e - ) flows from the anode 24 via the wire 32 through its corresponding BPP 28a, through the BPP on the cathode side 28b to the cathode 26. The wire 32 is provided with a consumer 34. According to the present invention, at least one of the metal BPPs 28a, 28b has a coating of doped DLC (not shown) according to the present invention. The fuel cell shown in Figure 1 generates an electric current (e - First, at the anode 24, hydrogen is oxidized by the catalytic material to generate protons and electrons. + ) passes through the PEM 22 and reaches the cathode 26. + ) and the generated electron (e - ) passes through the anode 24, GDL 30a, BPP 28a, and conductive carriers in the external wire 32 to reach the cathode 26. At the cathode 26, the protons and electrons react with oxygen (O) contained in the oxidant gas, catalyzed by an electrocatalyst, to produce water (HO). Deposition method In the following, a method for depositing a coating of non-hydrogenated DLC comprising at least one transition metal selected from groups 4d, 5d and 6d of the periodic table of the elements according to an embodiment of the present invention will be described with reference to FIG. In this embodiment, the method is carried out using a cathodic arc discharge deposition apparatus 2 as shown in FIG. 2. The apparatus 2 includes a deposition chamber 11, an anode 4, a target 6, a cathode 7, and a current source 8 electrically connected to the anode 4 and the cathode 7. The cathode 7 may be formed of a metal such as copper (Cu). For example, the cathode 7 may be a metal plate, such as a Cu plate. The cathode 7 may be cooled, for example, water-cooled. The anode 4 is integral with the wall of the deposition chamber 11. The anode 4 may form part of the wall of the deposition chamber 11. The anode 4 may be made of the same material as the wall of the deposition chamber 11, for example, a metal such as steel. The anode 4 has a substantially annular shape, as shown in cross section in FIG. 2. For example, the anode 4 may be a substantially annular metal plate, for example, a substantially annular steel plate. In particular, the anode 4 may be a substantially annular flat plate, such as a substantially annular flat metal plate. The anode 4 and the cathode 7 are arranged in a plane substantially perpendicular to the direction from the cathode 7 to the target 6 (see FIG. 2). The cathode 7 is arranged in a central opening of the substantially annular anode 4. The target 6 is directly connected to the cathode 7, i.e., there is no intermediate layer or structure between the target 6 and the cathode 7. In particular, a surface, e.g., a back surface, of the target 6 may be in contact, e.g., complete contact, with the cathode 7, e.g., the body of the cathode 7. The target 6 may be attached to the cathode 7, e.g., by a bolt connection (not shown). The target 6 is directly electrically connected to the cathode 7, e.g., by direct contact between the back surface of the target 6 and the cathode body. The current source 8 is configured to supply a direct current superimposed with a pulsed current to generate an electric arc between the target 6 and the anode 4. The electric arc vaporizes material on the surface 10 of the target 6 in the area where the arc exists. The vaporized target material 12 (see FIG. 2) transfers from the target 6 to a substrate 14 and is deposited on the substrate 14 to form a coating of the target material thereon. The anode 4, target 6, cathode 7, and substrate 14 are arranged in a space formed within a deposition chamber 11. The deposition chamber 11 may be a vacuum chamber, for example, an ultra-high vacuum (UHV) chamber. The anode 4 is integral with the wall of the deposition chamber 11 and is connected, i.e., electrically connected, to the wall of the deposition chamber. The current source 8 is electrically connected to the anode 4 through the wall of the deposition chamber 11 (see FIG. 2 ). In particular, the wall of the deposition chamber 11 may be made of a conductive material, such as a metal, thus establishing an electrical connection between the current source 8 and the anode 4. In other embodiments, multiple cathodes 7 may be arranged within the deposition chamber 11. The background pressure in the deposition chamber 11 is 5x10 -5 The cathodic arc discharge deposition process of this embodiment can be carried out in an atmosphere containing an inert gas, particularly Ar, in the deposition chamber 11. For example, the pressure of Ar in the deposition chamber 11 can be 5×10 -4 It can be mbar. In this embodiment, the target 6 is a carbon target doped with tungsten (W). Thus, a coating of non-hydrogenated DLC containing W is formed on the substrate 14 by the deposition process. The dopant level of the target 6 is in the range of 0.5 at.% to 8.0 at.% W. The superimposed current supplied by the current source 8 has the following characteristics: a pulse current having a pulse frequency in the range of 10 kHz to 100 kHz is superimposed on a DC current of 50 A. The pulses of the pulse current have an active pulse width of 5 μs and a pulse interval of 80 μs. The peak current of the DC current superimposed with the pulse current is greater than 200 A. The DC current and the pulse current are measured at the cathode 7. Each pulse of pulsed current induces a rise in voltage measured at cathode 7, i.e., the arc discharge voltage, at a rate of more than 5 V / μs. The voltage is measured between cathode 7 and anode 4, which is at ground potential. By employing such current superposition, the generation of macroparticles of the target material during the cathodic arc discharge process can be significantly reduced. Furthermore, the generation of craters on the target surface can be minimized. Furthermore, it is possible to ensure that small molten W droplets are incorporated into the coating, thereby enhancing the coating's electrical conductivity and friction-reducing properties. Therefore, a particularly high-quality coating can be provided on the substrate 14. The pulse superimposed on the DC current causes the electric arc to split into multiple arcs, as detailed above. In this embodiment, the current of each single arc resulting from this splitting, i.e., the arc current, may be, for example, approximately 60 A. At peak currents, the total number of arcs may be, for example, five or six. When the arc current is maintained at a high level, the arcs resulting from the pulse-induced arc splitting repel each other, resulting in plasma characteristics similar to those of a DC arc. Therefore, the active pulse width is kept short, i.e., less than 30 μs. In this embodiment, the active pulse width is 5 μs, as detailed above. In this embodiment, the cathodic arc discharge deposition process is carried out at a deposition temperature in the range of 70° C. to 150° C. The deposition temperature is measured at the substrate 14. Furthermore, a bias of 50 V is applied to the substrate 14. Deposition parameters such as DC and pulsed current characteristics, deposition temperature, substrate bias, gas atmosphere and pressure are kept substantially constant during the cathodic arc discharge deposition process of this embodiment. Prior to depositing the coating, the substrate 14 may be cleaned, for example, by Ar etching. The coating may be deposited directly on the surface of the substrate 14. Alternatively, an initial adhesion layer may be provided on the surface of the substrate 14 before the coating is deposited thereon. The initial adhesion layer may be, for example, a metallic chromium (Cr) layer or a metallic titanium (Ti) layer. From the initial adhesion layer to the doped carbon layer, there can be an abrupt transition or ramp down of the metal deposition rate and a ramp up of the doped carbon deposition rate. [Example]

[0008] The present invention will now be further illustrated by way of examples which, of course, should not be construed in a limiting sense. The coatings were deposited by cathodic arc discharge on a W-doped carbon target. The arc discharge was a DC arc with superimposed pulses. Samples were prepared with a DC arc current of 50 A, 5 μs pulse width, 80 μs pulse interval, and an arc peak current greater than 200 A. W dopant levels of 0.5 to 5 at.% were applied. The deposition temperature was maintained between 70 °C and 150 °C. The bias energy voltage applied to the substrate was 50 V. The background pressure was typically 5 x 10 -5 A small amount of Ar was allowed in the chamber to ensure proper arc ignition, typically 5x10 -4 This results in an Ar pressure of 1000 mbar. The product to be coated, i.e., the substrate, is cleaned by Ar etching, i.e., argon ion bombardment, before deposition. The addition of an initial adhesion layer, such as metallic Cr or metallic Ti, has been applied to many samples, but coatings without an adhesion layer have also been produced. The ta-C coatings were deposited without changing any of the parameters during the deposition process, which means that conditions such as pressure, gas atmosphere, bias voltage, and substrate temperature were kept constant during the deposition process. analysis Composition analysis Compositional analysis of the coatings was performed by electron probe microanalysis (EPMA) using an accelerating voltage of 5 keV, a current of 200 nA, and 10 test probes per sample. The compositions investigated in detail are shown in Table 1. JPEG0007747744000001.jpg79166Hardness Hardness is indentation hardness H ITThe hardness was measured by nanoindentation on a flat, polished, hardened substrate using a diamond indenter and a microhardness tester (Fischerscope H100) in accordance with ISO 14577 (i.e., the English versions of ISO 14577-1:2015, July 15, 2015, ISO 14577-2:2015, July 15, 2015, and ISO 14577-4:2016, November 1, 2016). The diamond indenter used has a Vickers geometry. That is, the indenter is a diamond with a square pyramid base with a plane at a 22° angle to the horizontal, or in other words, it is shaped as an orthogonal pyramid with a square base and an angle of 68° between the axis of the diamond pyramid and one of the faces (Vickers pyramid). The hardness is expressed in GPa. The roughness Ra of the substrate was less than 0.06 μm. Specifically, the surface roughness Ra of the substrate (test plate) was 0.01 μm and Rz was 0.25 μm. The hardness of the flat-ground hardened substrate used was 83.6 HRa (Rockwell hardness A, HRA), 62.1 HRC (Rockwell hardness C, HRC), and 747 HV10 (Vickers hardness at a load of 10 kgf). The size of the test plate, i.e., the flat-ground hardened substrate, was 15 x 6 mm. The load of the indenter and the thickness of the coating deposited on the test plate were selected so that the indentation depth was less than 10% of the coating thickness. For the measurement, 10 points (uniformly distributed on the coating surface to be measured) were taken per sample, and the 10 hardness indentation values ​​H were measured. IT The average value of the indentation hardness in GPa is used. IT The relationship between the hardness and Vickers hardness (Hv) is as follows: Hv=94, 53H IT is. Raman testing Raman studies were performed on samples before and after the etching process and with different dopant contents. The Raman excitation wavelength was 532 nm. The G peak position was at 1605 cm -1 At this excitation wavelength, more than 60% of sp 3 Indicates the fraction. 80~150cm -1At 0.6 at.% W and 1.4 at.% W, a peak representing WC bonds becomes visible. Pure ta-C samples and those with 0.3 at.% W do not show this peak. It is visible at 0.6 at.% W and 1.4 at.% W, and is more intense at higher W contents. Raman does not provide information on whether the WC bonds indicate WC crystallites or bonds between individual W atoms and C. Examination of the coatings after the etching process showed no substantial difference in the Raman spectra before and after oxidation. TEM examination of the samples TEM studies were performed with bright-field TEM (BFTEM) and high-angle annular dark field with spot for scanning TEM (HAADF-STEM) at a beam voltage of 200 kV. The sample consisted of a silicon (Si) wafer with a 20 nm thick Si3N4 layer on top, measuring 1 × 1 mm. 2 The Si was etched away by etching. Thus, a window was created. 80 nm of ta-C, doped with 0.3, 0.6, and 1.4 at.% W, was deposited on top of the Si3N4 foil. Most of the Si3N4 foil broke under the compressive stress of the doped ta-C film, but undisturbed foil could be found at the corner of the window. Observation was performed from the side of the doped DLC coating of the present invention deposited on the Si3N4 foil. The combination of BFTEM and HAADF-STEM imaging allowed for the differentiation of W droplets and carbon clusters. W droplets were identified in the 80 nm film. Figure 4 shows a coating doped with 1.4 at.% W observed by BFTEM. The small black dots are W droplets, as confirmed by EDX analysis. In Figure 4, the W droplet sizes range from 2 to 40 nm. In other samples, droplets up to 100 nm were observed. Figure 5 shows a HAADF-STEM image of the same sample doped with 1.4 at.% W. Sub-nm clusters of W were visible (W appears white in HAADF-STEM). It should be noted that due to the focus of the HRTEM, only slots with a thickness of approximately 20 nm were observed, rather than a total film thickness of 80 nm. While precise measurement of such particles in thick coatings is not possible, crude measurements yielded cluster sizes of ~0.5 nm. W has a bcc structure with a lattice spacing of 0.34 nm. WC typically has a hexagonal distribution with lattice constants of 0.29 and 0.28 nm. Therefore, clusters consist of a fairly limited number of W atoms, ranging up to 10 atoms. Assuming that the islands and droplets are spherical, a semi-quantitative analysis of the amount of W visible in the metal droplets and the amount visible in the WC islands was performed. In a coating of doped DLC according to the present invention doped with 0.6 at.% W, 50% of the W was found to be attributable to the WC islands and 50% to the metal droplets. In a film doped with 1.4 at.% W, approximately 30% of the W can be attributed to the WC islands and 70% to the metallic W droplets. This is consistent with the expectation that the amount of droplets increases more than linearly with the percentage of W in the target. The very clear presence of WC in the Raman spectrum is also due to only 1.4 at.% W, the smaller crystallites being WC. Quantitative analysis showed that there was almost no "free" W. The excellent properties of the above materials as coatings for BPP (stainless steel metal BPP) in PEMFCs and electrolysers were confirmed by testing their ICR and corrosion performance. Interfacial contact resistance (ICR) Within a PEM fuel cell or PEM electrolyzer, there are two critical contact points where electron flow is hindered: the catalyst-gas diffusion layer (GDL) and the GDL-BPP contact. In an FC stack, the BPP is located between two GDLs, so the electrical resistance of the BPP contact is R GDL-BPP-GDL GDL is also expressed as R GDLA typical GDL has two different resistances: the first is the flow of electricity through the GDL layer, cm 2 The first is the "through-plane" resistance, which is the electrical resistance perpendicular to the area. The second is the "in-plane" resistance, which is the electrical resistance per cm within the layer of the GDL. R GDL(面内) is not used for the purpose of ICR measurements on BPP because electrons are transported perpendicularly rather than parallel through the GDL layer. The final electrical resistance at the contact point between the GDL and the catalyst is R Cat.-GDL The BPP also has electrical resistance, but the electrical resistance of these BPPs is R GDL(面貫通) This can be ignored because it is 1 / 10,000 of the electrical resistance of R Cat.-GDL is measured for one GDL, then R GDL-BPP-GDL and the desired result is the ICR from the sample only, so the ICR is calculated using Equation 1.

number

number

Claims

1. A metallic bipolar plate comprising a metal substrate and at least one layer of non-hydrogenated transition metal doped diamond-like carbon (DLC) provided on the metal substrate, wherein the non-hydrogenated transition metal DLC comprises at least one transition metal selected from Groups 4d, 5d, and 6d of the Periodic Table of Elements, and a portion of the at least one transition metal is present in the non-hydrogenated DLC as a matrix in the form of a carbide of the at least one transition metal; the non-hydrogenated transition metal doped DLC has a hardness of 35 GPa or greater, the hardness being measured on a film of the non-hydrogenated transition metal doped DLC deposited on a polished hardened substrate, with an indentation depth of less than 10% of the thickness of the film; A metallic bipolar plate, wherein some of said carbides of said at least one transition metal are present as islands in said non-hydrogenated DLC as a matrix.

2. 10. The metallic bipolar plate of claim 1, wherein the non-hydrogenated transition metal doped DLC has a hardness of 40 GPa or greater.

3. The metallic bipolar plate of claim 1 , wherein the islands have a size of up to 2 nm.

4. A metallic bipolar plate according to any one of claims 1 to 3, wherein another portion of said at least one transition metal is present in said non-hydrogenated DLC in said metallic form as droplets of said transition metal.

5. 5. The metallic bipolar plate of claim 4, wherein the metal droplets have a diameter of less than 1 μm.

6. 6. A metallic bipolar plate according to claim 4 or 5, wherein the metal droplets have a diameter of 0.1 to 100 nm.

7. A metallic bipolar plate according to any one of claims 4 to 6, wherein the metal droplets have a diameter of 0.5 to 40 nm.

8. 8. The metallic bipolar plate according to claim 1, wherein the non-hydrogenated transition metal doped DLC has a content of the at least one transition metal in the range of 0.1 to 5 at. %.

9. The metallic bipolar plate according to any one of claims 1 to 8, wherein the hardness of the non-hydrogenated transition metal doped DLC is in the range of 40 GPa to 60 GPa.

10. The metallic bipolar plate according to any one of claims 1 to 8, wherein the non-hydrogenated transition metal doped DLC has an sp3 fraction of carbon atoms of 60% or more.

11. 11. The metallic bipolar plate of claim 10, wherein the hardness of the non-hydrogenated transition metal doped DLC is greater than 60 GPa.

12. A metallic bipolar plate according to any one of the preceding claims, wherein the at least one layer of non-hydrogenated transition metal doped DLC has a thickness in the range of 50 nm to 3 μm.

13. Use of a metallic bipolar plate according to any one of claims 1 to 12 in a fuel cell or electrolyzer.

14. 14. The use according to claim 13, wherein the non-hydrogenated transition metal doped DLC has a content of the at least one transition metal in the range of 0.3 at. % to 2.5 at. %.

15. A proton exchange membrane fuel cell or electrolyzer comprising a metallic bipolar plate according to any one of claims 1 to 12.

16. 16. The proton exchange membrane fuel cell or electrolyzer of claim 15, wherein the layer of non-hydrogenated transition metal doped DLC has a thickness of 50 nm to 1 μm.

17. 1. A method for depositing a coating of non-hydrogenated DLC comprising at least one transition metal selected from groups 4d, 5d, and 6d of the periodic table of the elements, the method being a cathodic arc discharge deposition method, in which a direct current is superimposed with a pulsed current in the cathodic arc discharge, the pulsed current having a pulse frequency in the range of 10 kHz to 100 kHz, a carbon target doped with the at least one transition metal is used as a target in the cathodic arc discharge, the target is directly connected to a cathode, each pulse of the pulsed current induces a voltage rise measured at the cathode at a rate of more than 5 V / μs, and each pulse of the pulsed current has an active pulse width of less than 30 μs.

18. The method of claim 17, wherein the peak current is greater than 200 A.

19. The method according to claim 17 or 18, wherein the coating is a coating of non-hydrogenated transition metal doped DLC according to any one of claims 1 to 11.

20. The metallic bipolar plate according to any one of claims 1 to 12, wherein the metallic substrate is a stainless steel substrate or a titanium substrate.

Images