[0001] This invention relates to a method for electrodeposition of chromium containing coatings from trivalent chromium based electrolytes.

[0002] Chromium coatings are used in a wide variety of industries. Traditionally, chromium coatings are applied by means of electroplating from electrolytes based on hexavalent chromium (Cr(VI)) and catalyst ions such as sulphate or fluoride. Three fundamentally different classes of chromium coatings can be distinguished : functional coatings (hard), decorative coatings and coatings on steel for packaging purposes, also known as Electrolytic Chromium Coated Steel (ECCS) or Tin Free Steel (TFS).

[0003] Functional chromium coatings consist of a thick layer of chromium (typically from 0.5 to over 1000 prn) to provide a surface with functional properties such as hardness, corrosion resistance, wear resistance, and low friction. Applications of functional chromium coatings include strut and shock absorber rods, hydraulic cylinders, crankshafts and industrial rolls. Carbon steel, cast iron, stainless steel, copper, aluminium, and zinc are substrates commonly used with functional chromium.

[0004] Decorative chromium coatings consist of a thin layer of chromium (typically from 0.1 to 1.0 prn) most often applied over a polished surface or bright nickel to provide a bright surface with wear and tarnish resistance. Decorative chromium coatings are for example found on automotive trims and bumpers, bath fixtures, shower heads and small appliances.

[0005] ECCS is an acronym for Electrolytic Chromium Coated Steel . ECCS is also often called Tin Free Steel (TFS), because this material was originally developed as a lower cost alternative for tinplate due to the high tin prices at the time. This material consists of a thin gauge (0.13 - 0.49 mm) low-carbon steel substrate with a very thin coating comprising a base layer of chromium metal (50 - 150 mg/m ² ) and a top layer of chromium oxide (7 - 35 mg/m ² ). This material is particularly suitable for use in the packaging industry. ECCS is typically used in the production of DRD two-piece cans and components that do not have to be welded, such as ends, lids, crown corks, twist-off caps and aerosol bottoms and tops. ECCS excels in adhesion to organic coatings. There are also many examples of 'non-packaging' applications of ECCS, like automotive components (oil filters, cylinder head gaskets), building trade (space bars for double glazing, light reflectors) and house ware (cake tins, gas canister components).

[0006] For the production of ECCS generally three types of chromium plating processes are in use throughout the world. The three processes are "one step vertical process" (V-l), "two step vertical process" (V-2), and the "one step horizontal high current density process" (HCD) and based on Cr(VI) electrolytes. The specifications of ECCS are standardized under Euronorm EN 10202 : 2001. The two-step vertical process uses a sulphuric acid free Cr(VI) electrolyte for applying the chrome oxide layer in the second step. Sulphuric acid is needed for a good efficiency in applying chrome metal and is therefore always used for the chrome metal plating step in these processes. The "one step vertical" and the "one step horizontal high current density (HCD) process" always have sulphate in the oxide layer because the chromium metal and chromium oxide are produced simultaneously in the same electrolyte (Boelen, thesis TU Delft 2009, page 8-9, ISBN 978-90-805661-5-6). In all cases the ECCS consists of a chromium oxide layer on top of the chromium metal.

[0007] As said traditional chromium electrolytes contain hexavalent chromium which is usually added to the electrolyte as chromium trioxide (Cr0 ₃ ). Hexavalent chromium is nowadays considered a hazardous substance that is potentially harmful to the environment and constitutes a risk in terms of worker safety. To comply with occupational safety and health regulations, the toxicity of hexavalent chromium requires an expensive exhaust system to capture any aerosols being released during electrolysis and also a complex waste water treatment of the effluents. The harmfulness of hexavalent chromium (Cr(VI), Cr ⁶⁺ ) is attributed to its high oxidising potential and its easy permeation of biological membranes. In contrast, trivalent chromium (Cr(III), Cr ³⁺ ) is not known to be harmful to body tissue. In fact, it is an important component of a balanced human and animal diet and a deficiency is detrimental to the glucose and lipid metabolism in mammals. Therefore, trivalent chromium plating is considered a benign technology to replace hexavalent chromium plating.

[0008] Commercial trivalent chromium plating processes for applying decorative coatings have already been in use since the mid-1970s. However, trivalent chromium plating processes for applying functional coatings or producing ECCS appear to be more cumbersome due to issues with process stability and inferior coating quality.

[0009] One major concern addressed in many scientific papers and patents on trivalent chromium plating is the possible oxidation of trivalent chromium to hexavalent chromium at the anode. Besides water also some Cr(III) might be oxidised unintentionally to Cr(VI) at the anode, because the electrode potentials for the oxidation of water to oxygen and the oxidation of Cr(III) to Cr(VI) are very close.

[0010] Typical anode materials used for chromium deposition processes from trivalent chromium electrolytes are carbon or platinised titanium anodes as described by Ward and Christie in GB 1333714 A 29-12-1970 . The use of graphite anodes is advised in the Technical Data Sheet for the commercial TriChrome ^® Plus process developed by Atotech.

[0011] Also, metals coated with a pure metal oxide, e.g. iridium or ruthenium oxide, or a mixed metal oxide (MMO), e.g . iridium/ruthenium or iridium/tantalum oxide are used as mentioned in US 2010108532 A 30- 10-2008 . The metal substrate can be any metal that does not dissolve in the electrolyte, such as titanium, tantalum, niobium, zirconium, molybdenum or tungsten, but preferably titanium is used. These anodes suppress the oxidation of Cr(III) as a result of their electrocatalytic properties for promoting the oxidation of water, but some oxidation of Cr(III) might still occur simultaneously especially over a period of operation and at higher anodic potentials.

[0012] In addition, as disclosed in US 4461680 30-12-1983 , a so-called depolariser might be added to the electrolyte, such as a bromide containing salt. The presence of bromide can assist with suppressing the oxidation of trivalent chromium, presumably through interacting with other chemical species present at the anode surface. The exact reaction mechanisms involving bromide seem complex and remain largely unresolved at this time. It should be noted that bromide can be converted to bromine at sufficiently high anode potentials. Bromine vapour is hazardous when inhaled . The MAC (Maximum Allowable Concentration) value for bromine is 0.7 mg/m ³ .

[0013] Oxidation of Cr(III) to Cr(VI) can also be avoided by using a shielded anode as described in for example GB 1602404 A 6-4-1978 . A lead anode in a different compartment comprising a sulphuric acid anolyte was used that was separated from the trivaient chromium electrolyte with an ion selective membrane. In this case, the plating current is carried by hydronium cations, which can freely move through the membrane. Cr(III) ions, typically being present in the form of an octahedral complex with for example a formate or acetate ligand are blocked by the membrane. So, the membrane effectively prevents any physical contact of Cr(III) with the anode, thus preventing oxidation of Cr(III) to Cr(VI). However, this type of arrangement is expensive and difficult to maintain. The membrane also introduces an additional resistance in the electrical circuit, which significantly increases the overall cell resistance. Moreover, this arrangement is only suited for sulphate electrolytes and not for chloride electrolytes.

[0014] It is the object to prevent the oxidation of trivaient chromium to hexavalent chromium when plating from a trivaient chromium plating bath.

[0015] According to a first aspect of the invention the object is reached with a method for electrodeposition of a chromium containing coating from a trivaient chromium based electrolyte comprising a trivaient chromium compound, a chelating agent, an optional conductivity enhancing salt such as a chloride, an optional depolariser, an optional surfactant and an optional acid or base for adjusting the pH of the electrolyte, on an electrically conductive steel strip, which may already be coated with one or more coating layers, in a continuous electrodeposition line, wherein the steel strip acts as the cathode, and wherein at least one hydrogen gas diffusion anode is used at which hydrogen gas is oxidised thereby preventing the oxidation of Cr ³⁺ to Cr ⁶⁺ and, if a chloride is used as a conductivity enhancing salt, the oxidation of chloride to chlorine gas, by using an anode potential which is less anodic than the potential at which Cr(III) is oxidised to Cr(VI). H ⁺ (protons) in an aqueous solution bind to one or more water molecules, e.g . as hydronium ions (H ₃ 0 ⁺ ). The oxidation of H ₂ (g) to H ⁺ (aq) prevents the occurrence of undesirable oxidation reactions which occur at a higher anodic overpotential when using an anode at which water (H ₂ 0) is oxidised to oxygen (0 ₂ (g)).

[0016] The reaction H ₂ (g)→ 2H ⁺ (aq) + 2e ^" occurs at an anode potential of 0.00

V (SHE). The reaction 2H ₂ 0→ 4H ⁺ (aq) + 0 ₂ (g) + 4e ^" occurs at an anode potential of 1.23 V (SHE). When an anode at which water is oxidised to oxygen is used, then reactions are possible which would not have been possible when using an anode at which hydrogen gas is oxidised .

[0017] One of such undesirable oxidation reactions is the oxidation of Cr(III) to Cr(VI) and this oxidation reaction can be completely excluded by using a hydrogen gas diffusion anode (GDA) at which H ₂ (g) is oxidised to H ⁺ .

[0018] In an embodiment of the method H ₂ (g) is oxidised at the gas diffusion anode to H ⁺ (aq) with a current efficiency of at least 99%, preferably of 100%. The higher the current efficiency, the smaller the likelihood of undesirable side reactions. It is therefore preferable that the current efficiency is at least 99%, and preferably 100%. Based on thermodynamic and kinetic considerations it can be argued that using a hydrogen gas diffusion anode completely eliminates the risk of Cr(III) oxidation as the anode operating potential is much too low for Cr(III) oxidation to occur.

[0019] Thermodynamically, under standard conditions (i.e. a temperature of 25 °C and a pressure of 1 atm) an electrode potential of > 0 V is already sufficient for oxidising H ₂ (g) to H ⁺ (aq), whereas an electrode potential of > 1.23 V is required for oxidising H ₂ 0 to 0 ₂ (g). Cr(III) can only be oxidised to Cr(VI) when the electrode potential is > 1.35 V.

[0020] The electrode potential is measured against the standard hydrogen electrode. The standard hydrogen electrode (abbreviated SHE), is a redox electrode which forms the basis of the thermodynamic scale of oxidation- reduction potentials. Its absolute electrode potential is estimated to be

4.44 ± 0.02 V at 25 °C, but to form a basis for comparison with all other electrode reactions, hydrogen's standard electrode potential (E°) is declared to be zero at all temperatures. Potentials of any other electrodes are compared with that of the standard hydrogen electrode at the same temperature.

[0021] The prevailing equilibrium (zero current) potential can be calculated from the Nernst equation by filling in the appropriate temperature, pressure and activities of the electro-active species. The anode operating (nonzero current) potential needed to generate a specific anodic current is determined by the activation overpotential (i.e. the potential difference required for driving the electrode reaction) and the concentration overpotential (i.e. the potential difference required to compensate for concentration gradients of electro-active species at the electrode).

[0022] Due to the low anode overpotential required for the oxidation of H ₂ (g) to H ⁺ (aq), the anode operating potential will always stay far below the value at which Cr(III) oxidation can take place (see Fig. 1 where the current is plotted against the anode potential in SHE). Firstly this results in a lower energy consumption of the electrodeposition process. Secondly, at an anode potential below about 1.35 V oxidation of Cr(III) to Cr(VI) is not possible (indicated with the crossed through arrow).

[0023] In an embodiment the chromium containing metal coating comprises a base layer of chromium metal and a top layer of chromium oxide deposited from a trivalent chromium based electrolyte.

[0024] In an embodiment the chromium containing metal coating comprises a chromium metal-chromium oxide (Cr-CrOx) layer deposited in a single plating step from a trivalent chromium based electrolyte. This Cr-CrOx coating layer consists of a mixture of Cr-oxide and Cr-metal. The Cr- oxide is not present as a distinct layer on the outermost surface, but is mixed through the whole layer. With the phrase single plating step is meant here that a coating layer comprising chromium metal and chromium oxide is deposited simultaneously. Of course there may be more than one of these single process steps one after the other if, for instance, a thicker coating layer comprising chromium metal and chromium oxide layer is to be deposited.

[0025] In an embodiment the steel strip is a steel substrate for packaging applications selected from :

• a conventional non-passivated electrolytic, optionally flowmelted, tinplate, or • a recrystallisation annealed single or double reduced packaging steel substrate, or

• a cold-rolled and recovery annealed blackplate, or

• a cold-rolled and recovery annealed electrolytic, optionally flowmelted, tinplate,

wherein one or both sides of the steel substrate is coated with a chromium metal - chromium oxide (Cr-CrOx) coating layer produced in a single plating step from the trivalent chromium based electrolyte.

[0026] In an embodiment more than one chromium metal - chromium oxide (Cr-CrOx) coating layer is deposited on one or both sides of the substrate. Each deposited Cr-CrOx coating layers consist of a mixture of Cr-oxide and Cr-metal.

[0027] In an embodiment no depolariser is added to the electrolyte. When a hydrogen gas diffusion anode is used then the addition of a depolariser to the electrolyte is no longer needed.

[0028] The use of a hydrogen gas diffusion anode has the added advantage that the use of a chloride containing electrolyte becomes possible without the risk of chlorine formation. This chlorine gas is potentially harmful to the environment and to the workers and is therefore undesirable. This means that in the case of a Cr(III) electrolyte the electrolyte could be partly or entirely based on chlorides. The advantage of using a chloride based electrolyte is that the conductivity of the electrolyte is much higher compared to a sulphate only based electrolyte, which leads to a lower cell voltage that is required to run the electrodeposition, which results in a lower energy consumption.

[0029] The oxidation reaction of dissolved hydrogen on an active electrocatalyst surface is a very fast process. As the solubility of hydrogen in a liquid electrolyte is often low, this oxidation reaction can easily become controlled by mass transfer limitations. Porous electrodes have been specifically designed to overcome mass transfer limitations. A hydrogen gas diffusion anode is a porous anode containing a three-phase interface of hydrogen gas, the electrolyte fluid and a solid electrocatalyst (e.g. platinum) that has been applied to the electrically conducting porous matrix (e.g. porous carbon or a porous metal foam). The main advantage of using such a porous electrode is that it provides a very large internal surface area for reaction contained in a small volume combined with a greatly reduced diffusion path length from the gas-liquid interface to the reactive sites. Through this design the mass transfer rate of hydrogen is greatly enhanced, while the true local current density is reduced at a given overall electrode current density, resulting in a lower electrode potential.

[0030] A gas diffusion anode assembly to be used in the proposed electrodeposition method, typically comprises the use of the following functional components (see Fig . 2) : a gas feeding chamber 1, a current collector 2 and a gas diffusion anode, which consists of an hydrophobic porous gas diffusion transport layer 3 combined with an hydrophilic reaction layer 4 (see Fig. 2). The latter is made up of a network of micropores that are (partly) drowned with liquid electrolyte. Optionally, the reaction layer is provided with a proton exchange membrane on the outside 5, like a Nafion ^® membrane, to prevent the diffusion of chemical species (like anions or large neutral molecules) present in the bulk liquid electrolyte inside the gas diffusion anode, as these compounds can potentially poison the electrocatalyst sites, causing degradation in electrocatalytic activity.

[0031] The main function of the gas feeding chamber is to supply hydrogen gas evenly to the hydrophobic backside of the hydrogen gas diffusion anode. The gas feeding chamber needs two connections: one to feed hydrogen gas and one to enable purging of a small amount of hydrogen gas to prevent the build-up of gas phase contaminations potentially present in trace amounts in the hydrogen gas supplied. The gas feeding chamber often contains a channel type structure to ensure that hydrogen gas is distributed evenly over the hydrophobic backside.

[0032] The electrical current collector 2 is (usually) attached to the hydrophobic backside 3 of the hydrogen gas diffusion anode to enable the transport of the electrical current generated inside the anode to a rectifier (not shown in Fig. 2). This current collector plate must be designed in such a way to enable the hydrogen gas to contact the backside of the hydrogen gas diffusion anode so it can be transported to the reactive side inside the gas diffusion anode. Usually this is accomplished by using an electrically conductive plate with a large number of holes, a mesh or an expanded metal sheet made from e.g. titanium.

[0033] The functionality of gas feeding channels and electrical current collector can also be combined into a single component, which is then pressed against the hydrophobic back-side of the gas diffusion anode.

[0034] Once the hydrogen gas diffuses through the hydrophobic backside of the hydrogen gas diffusion anode it comes into contact with the electrolyte, which is present in the hydrophilic part of the anode, i.e. the reaction layer (see Fig. 2, right hand side). At the gas-liquid interface (between 3 and 4) the hydrogen gas dissolves into the electrolyte and is transported by diffusion to the electrocatalytic active sites of the hydrogen gas diffusion anode. Usually platinum is used as electrocatalyst, but also other materials like platinum-ruthenium or platinum-molybdenum alloys can be used. At the electrocatalytic sites the dissolved hydrogen is oxidised : the electrons that are generated are transported through the conductive matrix of the gas diffusion anode (usually a carbon matrix) to the current collector 2, while the hydronium ions (H ⁺ ) diffuse through the proton exchange membrane into the electrolyte.

[0035] The method according to the invention can be executed with trivalent chromium containing electrolytes based on the use of chloride and/or on sulphate containing chemicals.

[0036] In an embodiment the electrodeposition of a chromium containing coating is achieved using an electrolyte comprising a trivalent chromium compound, a chelating agent, an optional conductivity enhancing salt, an optional depolarizer, an optional surfactant and to which an acid or base can be added to adjust the pH .

[0037] In an embodiment the electrodeposition of a chromium containing coating is achieved using an electrolyte in which the chelating agent comprises a formic acid anion, the conductivity enhancing salt contains an alkali metal cation and the depolarizer comprises a bromide containing salt.

[0038] In an embodiment the cationic species in the chelating agent, the conductivity enhancing salt and the depolarizer is potassium. The benefit of using potassium is that its presence in the electrolyte greatly enhances the electrical conductivity of the solution, more than any other alkali metal cation, thus delivering a maximum contribution to lowering of the cell voltage required to drive the electrodeposition process.

[0039] In an embodiment the chromium containing metal coating is deposited from the trivalent chromium based electrolyte at a temperature of between 40 and 70°C, preferably of at least 45°C and/or at most 60°C. This temperature range provides shiny metallic coating layers.

[0040] The steel strip is usually provided in the form of a strip of low carbon (LC), extra low carbon (ELC) or ultra low carbon (ULC) steel with a carbon content, expressed as weight percent, of between 0.05 and 0.15 (LC), between 0.02 and 0.05 (ELC) or below 0.02 (ULC) respectively. Alloying elements like manganese, aluminium, nitrogen, but sometimes also elements like boron, are added to improve the mechanical properties (see also e.g . EN 10 202, 10 205 and 10 239). In an embodiment of the invention the substrate consists of an interstitial-free low, extra-low or ultra-low carbon steel, such as a titanium stabilised, niobium stabilised or titanium-niobium stabilised interstitial-free steel.

[0041] In an embodiment the coated substrate is further provided with an organic coating, consisting of either a thermoset organic coating, or a thermoplastic single layer polymer coating, or a thermoplastic multi-layer polymer coating. The Cr-CrOx layer provides excellent adhesion to the organic coating similar to that achieved by using conventional ECCS.

[0042] In a preferred embodiment the thermoplastic polymer coating is a polymer coating system comprising one or more layers comprising the use of thermoplastic resins such as polyesters or polyolefins, but can also include acrylic resins, polyamides, polyvinyl chloride, fluorocarbon resins, polycarbonates, styrene type resins, ABS resins, chlorinated polyethers, ionomers, urethane resins and functionalised polymers.

[0043] In an embodiment the Cr-CrOx coating layer applied onto non-passivated tinplate contains at least 20 mg Cr/m ² , to create a tin oxide passivating effect. This thickness is adequate for many purposes.

[0044] In an embodiment the Cr-CrOx coating layer applied onto non-passivated tinplate contains at least 40 mg Cr/m ² , preferably at least 60 Cr/m ² , to create a tin oxide passivating effect and to prevent or eliminate sulphur staining. To prevent or eliminate sulphur staining, a layer of 20 mg Cr/m2 was found to be too thin. Starting at thicknesses of about 40 mg Cr/m ² the sulphur staining is already much reduced, whereas at a layer thickness of of at least about 60 mg Cr/m ² sulphur staining is practically eliminated .

[0045] A suitable maximum thickness was found to be 140 mg Cr/m ² . Preferably the Cr-CrOx coating layer applied onto non-passivated tinplate contains at least 20 to 140 mg Cr/m ² , more preferably at least 40 and/or at most 90 mg Cr/m ² , and most preferably at least 60 and/or at most 80 mg Cr/m ² .

[0046] These embodiments aim to replace hexavalent chromium passivated tinplate. The major advantage besides the elimination of hexavalent chromium from manufacturing is the potential to create a product with superior sulphur staining resistance and improved corrosion resistance.

[0047] In an embodiment the Cr-CrOx coating layer applied onto blackplate is at least 20 mg Cr/m ² , to create a material that approaches the functionality of ECCS (e.g. excellent adhesion to organic coatings in combination with a moderate corrosion resistance). Preferably the Cr-CrOx coating layer applied onto blackplate is at least 40 and more preferably at least 60 mg Cr/m ² . A suitable maximum thickness was found to be 140 mg Cr/m ² . Preferably the Cr-CrOx coating layer applied onto blackplate contains at least 20 to 140 mg Cr/m ² , more preferably at least 40 mg Cr/m2, and most preferably at least 60 mg Cr/m ² . In an embodiment a suitable maximum is 110 mg Cr/m ² .

[0048] The Cr-CrOx coated blackplate aims to replace ECCS. The major advantage besides the elimination of hexavalent chromium from manufacturing is the potential to create a product for applications for which the superior corrosion resistance properties of tinplate are not required . From a process point of view, the fact that the Cr-CrOx coating layer is applied in a single plating step means that two process steps are combined, which is beneficial in terms of process economy and in terms of environmental impact.

[0049] The Cr-CrOx coating can also be applied to a cold-rolled and recovery annealed blackplate, or to a cold-rolled and recovery annealed electrolytic, and optionally flowmelted, tinplate. These substrates have a recovery annealed substrate, rather than the recystallised single reduced ETP or blackplate or the double reduced blackplate. The difference in microstructure of the substrate was not found to materially affect the Cr- CrOx coating . From a process point of view, the fact that the Cr-CrOx coating layer is applied in a single plating step means that two process steps are combined, which is beneficial in terms of process economy and in terms of environmental impact.

[0050] It was found that the material according to the invention can be used in combination with thermoplastic coatings, but also for applications where traditionally ECCS is used in combination with lacquers (i.e. for bakeware such as baking tins, or products with moderate corrosion resistance requirements) or as a substitute for conventional tinplate for applications where requirements in terms of corrosion resistance are moderate.

[0051] In an embodiment the composition of the electrolyte was (Table 1) : 120 g/l basic chromium sulphate, 250 g/l potassium chloride, 15 g/l potassium bromide and 51.2 g/l potassium formate. The pH was adjusted to values between 2.3 and 2.8 measured at 25 °C by the addition of sulphuric acid. The bath was kept at 50°C.

[0052]

Table 1 - Composition of Cr(III) electrolyte with KCI

[0053] In an embodiment the composition of the electrolyte was (Table 2) : 120 g/l basic chromium sulphate, 80 g/l potassium sulphate, 15 g/l potassium bromide and 51.2 g/l potassium formate. The pH was adjusted to values between 2.8 and 3.4 measured at 25 °C by the addition of sulphuric acid . The bath was kept at 50°C.

[0054]

Table 2 - Composition of Cr(III) electrolyte with K ₂ S0 ₄ molar mass c c compound CAS No.

[g/mol] [g/i] [M] basic chromium sulphate

307.11 [10101-53- CrO H S0 ₄ x N a ₂ S0 ₄ x n H ₂ 0 120 0.385

(n = 0) 8]

16.7 wt-% Cr

potassium sulphate

174.26 [7778-80-5] 80 0.459 (K ₂ S0 ₄ )

potassium bromide (KBr) 119.00 [7758-02-3] 15 0.126 potassium formate

84.12 [590-29-4] 51.2 0.609 (CHK0 ₂ )

[0055] The invention is now further explained by means of the following, non- limitative example.

[0056] A double-walled glass vessel connected with a thermostat bath was filled with a trivalent chromium electrolyte. The temperature of the electrolyte was kept constant at 50 ± 1 °C by circulation of hot water through the double-walled glass vessel.

[0057] The composition of the electrolyte was: 120 g/l basic chromium sulphate, 250 g/l potassium chloride, 15 g/l potassium bromide and 51 g/l potassium formate. The pH was adjusted to 2.3 measured at 25 °C by adding sulphuric acid.

[0058] The experiments were conducted using a three electrode system (i.e. a working electrode, a counter electrode and an auxiliary electrode) connected to a galvanostat (an Autolab PGSTAT 20 potentiostat/galvanostat from Metrohm). A galvanostat maintains a controlled constant current as defined by the user between the working electrode and the counter electrode, while the potential of the working electrode is monitored as a function of time vs. the potential of the reference electrode.

[0059] In the comparative examples the working electrode (anode) was a pure platinum electrode with an electro-active surface area of 5 cm ² , the counter electrode (cathode) was a strip of low-carbon steel sheet and the reference electrode was an Ag/AgCI electrode filled with a saturated KCI solution. At the platinum anode H ₂ 0 is oxidised to 0 ₂ (g) during the experiments, which is representative for the current state of art. [0060] In the inventive examples, the Pt anode was replaced by a hydrogen gas diffusion anode.

[0061] A current was applied and the anode potential was recorded vs. time. The current was increased stepwise and for each setting of the current the anode potential was monitored. A very stable anode potential was already attained within a few seconds after the onset of the current. In Fig. 3, the results of the potential measurements are plotted for both cases. These results are corrected for the ohmic drop residing between the anode and the reference electrode, which was placed at a distance of 4 mm from the anode using a value of 37.8 S/m for the conductivity of the electrolyte. Clearly, for all settings of the current the anode potential of the gas diffusion anode is substantially (about 0.95 V) lower than the potential of the Pt anode. At these much lower anode potentials oxidation of Cr(III) to Cr(VI) is impossible.

[0062] The anodes can be implemented in an industrial line without difficulty.

Their dimensions and operational requirements are comparable to the anodes normally used in such a line. The presence of hydrogen may require some additional safety measures.