Lubricant savings in sheet metal forming through thermally oxidized wear protection layers

Increasing demands on quality, variety of shapes and complexity of deep-drawn components with simultaneous price stability make the development of innovative solutions in the field of deep-drawing indispensable. For ecological and economic reasons, there is a great interest in reducing the amount of lubricant required in forming processes to a minimum. In this work, tool coatings for forming tools are to be investigated and further developed in order to use their friction-reducing properties on the tribo-system and thus enable a reduction in the use of lubricants. For this purpose, in this research strip drawing tests are carried out to investigate the friction-influencing properties of different coatings and their potential for lubricant saving. Therefore, cylindrical forming heads are provided with different coating systems. Strips of aluminum sheet are drawn around the forming heads at an angle of 90°. The amount of lubricant applied is varied. Friction values are recorded and analyzed for each system. Wear mechanisms are identified, examined and evaluated macroscopically and microscopically. For this purpose, energy dispersive X-ray spectroscope (EDX) and scanning electron microscope (SEM) examinations are carried out. A statement on the potential of the investigated coatings for lubricant saving is derived from the results.


Introduction
The tribological conditions in sheet metal forming influence process parameters such as temperature, process forces, tool wear and the robustness of the production process [1].The quality of the manufactured component is primarily determined by the type and quantity of lubricant used, the topography of the sheet metal and tool surface and by process-dependent factors such as the blankholder force [2], [3].Due to the increased demands on sheet materials, modern deep-drawing tools must be able to exhibit high strength, toughness and wear resistance.Multi-stage deep-drawing tools for the stepby-step forming of sheet metal are mainly used when the complexity of the component or a high overall drawing ratio no longer allow forming in a single stage [4].An optimization of the tribological boundary conditions for multi-stage tools with regard to the reduction of friction lowers the number of necessary drawing steps and extends the process limits, so that the increased complexity is kept as low as possible compared to deep drawing in one step [5], [6].One possibility for optimization often provides for additional lubrication of the sheet metal beyond the initial factory lubrication in order to improve the tribological system [7].This additional lubrication is particularly relevant in the subsequent stages, since the oil film on the sheet is already greatly reduced as a result of the first stage, which sometimes leads to the occurrence of solid or mixed friction in the process [8].The disadvantage of additional lubrication, however, is increased lubricant consumption, which should be kept as low as possible for ecological and economic reasons [9].Optimizations with regard to the surface microstructure show another possibility for improving the tribological system.They lead to a reduction of friction and wear values [10].A friction reduction through the application of suitable tool coatings shows advantages in terms of adhesion reduction on the one hand, on the other hand the application of surface structures to a forming tool also shows an increase in tool life as well as improved lubrication conditions [11].Within the framework of the DFG priority program "SPP 1676 -Dry Forming in Forming Technology", dry sheet metal forming by means of selective oxidation of tool surfaces was fundamentally investigated.For this purpose, a process was developed for the thermal generation of an oxide layer in the submicrometric range under inert gas.Wear investigations were able to show that iron oxide systems have a frictionreducing effect and are also wear-resistant [12].Further investigations also suggested that the combination of an oxide layer and minimal oiling of the workpiece can significantly improve the tribology in the deep drawing process [9].The sequential deposition of amorphous silicon dioxide also stabilizes the coating system and brings good dry lubrication properties with high hardness into the tribological system.The layer thicknesses of the oxide systems achieved so far amounted to approx.300 nm and must be adapted to conventionally applied tool coatings (~ 2 µm) in the future.Since the deposition of the oxide layer is slower the thicker it becomes, the production of a comparable layer thickness is a challenge that requires the development of alternative methods with faster growth rates and higher activation energies.In the context of this research, various coating systems are applied to cylindrical forming heads in order to investigate its effects on reducing the lubricant requirement in a forming process.Strip drawing tests with deflection on aluminum strips are carried out.The amount of lubricant is varied.The influence of the coatings on the tribological system, the friction values and the tool wear is analyzed and evaluated.Wear mechanisms are detected and evaluated by means of EDX and SEM measurements.

Materials and Methods
As testing material strips of Aluminum (EN AW-5657) in a sheet thickness of 0.8 mm are used.This is an alloy with small amounts of magnesium to increase the strength.EN AW-5657 is suitable for applications that require shiny surfaces.The alloy is weldable and hot and cold formable.Typical applications for this material are cladding for household appliances, car parts, power and window shutters as well as parts for the cosmetics industry.The mechanical properties of EN AW-5657 are shown in Table 1 [13]: Table 1.Mechanical properties of EN AW-5657.

Young's modulus [N/mm²] Density [g/cm²] Thermal conductivity [W/mK]
210,000 7.7 20 Two different heat treatment methods were used to modify the surfaces of the test specimens.Firstly, an oxidizing heat treatment under strongly reduced oxygen partial pressure was carried out in a continuous inert gas furnace at 500°C, producing thin surface layers of Fe2O3 (Oven 1 without SiH4).Using nitrogen as inert gas with approximately 0,1 ppm oxygen a further reduction of the O2-level can be achieved by adding small amounts of SiH4 (monosilane) (Oven 2 with SiH4) [12].
On the other hand, an ultra-short time laser treatment (LB1) of the test specimens in air was tested, which was carried out with a Nd:YAG laser under variation of the laser parameters like pulse repetition frequency, power and trave speed of beam, (see Fig. 1).This type of surface treatment resulted in a characteristic geometric microstructure of sparse depressions due to microscopic melting processes.With this microstructure, an increased lubricant retention capacity and thus more effective lubrication is aimed for.
Generally in all cases a periodical structure was created on the surface in accordance to the applied distances of 100 µm between discrete lasered lines and the laser spot diameter of about 50 µm, cf.light microscopic images in figure 1.
However, scanning electron microscopic analysis of the lasered sample, Fig. 2, reveal different topographic structures on the microscopic scale strongly depending on the laser process parameters.A pulse repetition rate of 7 kHz yield remelted surfaces with a smoothly undulating surface, while a treatment with 4 kHz applying the same beam travel speed (75 mm/s) result already in a fissured microstructure.The structural depth appears to be very shallow in all cases, amounting to only a few micrometers.
From a tribological point of view, a slightly undulating surface structure should be expected to exhibit lower sliding friction in the tool application.

Strip drawing test
In the strip drawing test, strips of sheet metal are drawn around cylindrical forming heads at an angle of 90°.The forming heads are initially lubricated with a defined amount of lubricant using a spraying system manufactured by the company Raziol GmbH.Renoform DYO 5006 from the company FUCHS LUBRICANTS GmbH with a viscosity of 126 mm²/s at 40 °C is used as lubricant.A film of defined lubricant mass is sprayed onto the sheet metal strips via a nozzle (see Fig. 3).The amount of lubricant is sprayed onto the sheet over a preset pulse duration.The quantity applied is checked beforehand using a precision scale to ensure that the lubricant application is reproducible.The strip drawing test stand replicates the load conditions experienced at the edge of a deep drawing die during drawing.Therefore, a sheet metal strip is drawn around a cylindrical test specimen.The counterholder force FC is maintained at constant level, while the tensile force FD is adjusted based on factors like the sheet material and the friction between the test specimen and the sheet strip.The test parameters are set constant to a drawing speed of vDraw = 10 mm/s and a constant counterholder force of FC = 2 kN.The test specimen is tempered to a constant temperature of 40 °C via a heating cartridge with integrated thermocouple.The friction values can be calculated from the force ratios of counterholder to drawing force by a formulaic relationship according to [8].The experimental setup is shown in Fig. 4: Three lubrication conditions are investigated.The strips are tested with oil amounts of 12 g/m 2 , 1 g/m 2 and without lubrication.The strips have a thickness of 0.8 mm, a width of 46 mm and a length of 1000 mm.The forming heads have a diameter of 16 mm and are hardened to a value of 57 HRC.

Test specimen
Fig. 5 visualizes the optical changes of the surface modified specimens.Both the laser treatment (LB1, corresponds to surface 2 in Fig. 1) and the heat treatment under nitrogen without silane (Oven 1 without SiH4), lead to a brown discoloration of the surface resulting from the formation of a thin alpha iron oxide coating (thickness 0,5-1 µm).The reduction of the oxygen partial pressure by silane addition leads to a significantly lower oxidation of the surface, which do not yet have closed layers and microscopically consist of discrete αFe2O3 crystallites.Optically, these layers appear transparent.The dot diagrams show the measured friction values for each test carried out in the order of the test sequence.Here it can be seen that the friction values increase after each stroke for all examined surfaces.This indicates that the surfaces are subjected to wear, which worsens the contact conditions and increases the friction.The forming heads "Oven 2 with SiH4" show the lowest averaged friction value of µ = 0.17 and therefore lays with about 19 % below the friction values of the uncoated reference with an average friction value of µ = 0.21.The highest averaged friction value of µ = 0.25 is achieved with the forming heads "Oven 1 without SiH4".It should be noted that the friction values of the layer "Oven 1 without SiH4" clearly deteriorate over the testing procedure from µ = 0.2 in stroke 1 to µ = 0.31 in stroke 3.This suggests that the wear process has already been initiated to a greater extent.Since the forming heads "LB1" show no significant difference to the uncoated reference, these will not be considered in the following.

Figure 6.
Friction values for 12 g/m² DYO5006 Fig. 7 shows the evaluation of the friction values for tests with an oil amount of 1 g/m 2 .Therefore, the influence of a reduced lubricant quantity is analyzed.In this case, also, the surface "Oven 2 with SiH4" with an averaged friction coefficient of µ = 0.195 turns out to be the most promising one and is close to the averaged friction coefficient of the uncoated reference surface with µ = 0.205.The increase of the friction coefficients per stroke is also not significant for "Oven 2 with SiH4" and reaches from µ = 0.19 to µ = 0.21.For the uncoated reference the friction coefficient reaches from µ = 0.195 at the first stroke to µ = 0.22 at the third stroke.The surface "Oven 1 without SiH4", on the other hand, shows a significantly worse performance with an average friction coefficient of µ = 0.28 and, with a significant increase in the friction values from µ = 0.22 in stroke1 to µ = 0.35 in stroke 3, also strongly advancing wear.
Furthermore, experiments are carried out without lubricant to investigate the dry friction properties of the coating systems.For this purpose, "Oven 2 with SiH4", the most promising coating so far, is compared with the uncoated surface of the reference forming head.The results show, that the sheet metal strips crack in the second stroke for both surfaces.The failure pattern can be seen in Fig. 8.This is caused by a massive increase in wear during the dry tests.Due to the decreasing of surface quality, the frictional resistance between the forming head and the aluminum sheets increases until the drawing force becomes so great that the sheet tears.

Wear mechanism
As part of the work, SEM (with acceleration voltage 20kV) and EDX studies were also carried out.These were used to investigate the dominant wear mechanisms for the various coating systems.The surface of forming head with coating LB1, "Oven 2 with SiH4", "Oven 1 without SiH4" and Uncoated forming head was studied.Using the BSE (back scattered electrons) and corresponding images, the difference in composition between the components of the sample was visualized in different shades of grey (Fig. 9).Since a substance with larger atoms (for example, iron Z=26) scatters electrons much more strongly compared to a substance consisting of light atoms (for example, aluminium Z=13), they create a stronger signal and are lighter in the SEM image (Fig. 9).The SEM results show, that for all coatings macroscopic stripes appear in the tensile direction of the forming zone.These stripes were examined by EDX to determine their elemental composition and the presence of aluminium, iron, chromium and other elements.The various deposits of the elements are visualised in Fig. 9 in the coloured illustrations, which all show the same enlarged area presented in der SEM picture on the left.The EDX measurements show that these stripes consist primarily of aluminium, which indicates that sheet material is deposited on the forming heads.The dominant wear mechanism is therefore an adhesive material transfer.This leads to a deterioration of the surface roughness and thus to a progressive increase in the frictional force.Qualitative differences in the material transfer from the aluminium band to the surface of the tool steel cannot initially be derived from the SEM examinations.

8
In the case of the surface-modified tool samples, a measurable change in the high adhesion tendency of the aluminium without lubricant cannot be demonstrated yet on a microscopic scale.

Conclusion and Outlook
Within the scope of this research, strip drawing tests were carried out with differently coated specimens in order to investigate the influence of tool coatings on the tribological conditions of a deep drawing process.It could be shown that the coating "Oven 2 with SiH4" causes a reduction of the friction factor compared to an uncoated tool surface as reference under saturated tool lubrication conditions.For experiments with reduced lubricant quantity, a slight improvement of the tribological conditions could be achieved with this layer compared to a conventional tool surface.However, a complete elimination of the need for lubricants could not be achieved.Adhesion is the primary wear mechanism identified for all surfaces.In future research, other layer systems such as further modified surfaces via laser treatment will be investigated, in order to focus on lubricant retaining structures for tool steel surfaces.The most promising variants will be applied on a deep-drawing die in a forming tool to investigate and evaluate their performance under real process conditions.

Fig. 6
Fig. 6 shows the evaluation of the friction values with an oil amount of 12 g/m 2 for all forming heads.The bars show the mean friction values over three tests with the corresponding standard deviation.