By: Dominik Hawelka

Friction creates a loss of energy and reduces component lifetime. According to the scientific community concerned with friction, wear and wear protection the overall costs related to these issues amount to 4 % of the Gross National Product of industrial countries, a staggering sum. Despite the fact that wear protection coatings are already deployed to improve the life-time of highly stressed components, limiting costs due to tribological loss mechanisms poses a great challenge in many industrial sectors. In particular, within the automobile sector, a high throughput of parts needs to be protected every year. As a consequence there is a strong demand for low-cost production processes for wear protection coatings.

Physical vapor deposition (PVD) is a widely accepted technology to produce high-performance wear protection coatings. It exhibits, however, several distinctive drawbacks such as its high demand for a great amount of technological efforts and the inherent costs as well as the lack of inline-capability. As a low-cost alternative to PVD, wet-chemical processes based on nanoparticulate materials hold great potential as they do not need expensive vacuum technology or any other elaborate equipment. In addition, they can also easily be integrated into an inline process chain.

The FunLas research consortium consisting of Schaeffler KG, Merck KGaA Darmstadt, Biofluidix GmbH, DILAS GmbH and the Fraunhofer ILT has succeeded in producing innovative wear resistant coatings based on such nanoparticulate material. Sponsored by the German Federal Ministry of Education and Research, the jointly run project aims to develop a cost-effective laser-based inline production process.

The overall process can be divided into three steps. At first the sol-gel coating mixture consisting of nanoparticulate zirconia dispersed in a mixture of solvents and additives is applied to hardened steel substrates via dip- or spin-coating. These are energy and resource-saving techniques that are easy to implement. During the coating process evaporation of the main part of solvents takes place. Within the second step the remaining solvents are removed by heating the samples to 150 °C. After drying the deposited layer is about 300 nm (Figure 1).

 

[Figures 1a and 1b, Coated samples with different geometries (a). SEM-micrograph of a coated sample (b)]

Finally the major challenge of this innovative coating process is to implement a thermal post treatment at temperatures > 800 °C, required to achieve functionalisation of the applied films. In this case the functionalisation refers to all processes, such as densification of the coating, that lead to improving the layer hardness and mechanical stability significantly. During this thermal post treatment it is necessary to minimize the thermal load of the steel substrates, which often feature low tempering resistance. Due to its exact local and temporal controllability, the laser is very well suited for this purpose.

In order to fulfill the complex task of generating temperature-time-profiles that meet these two opposing requirements the experimental work is supported by the following modeling approach. In a first step the splitting-up ratio of the laser energy absorbed in the coating and in the substrate is calculated based on the determined optical constants of the coating material. Knowing that 17 % of the overall laser power is absorbed by the coating whereas 29 % is absorbed by the substrate the heat conduction equation is solved for specific sets of process parameters. According to this simulation it is possible to realize a peak temperature of 800 °C at the surface of the coated sample while the heat affected zone is reduced to approximately 20 µm when using pulsed diode laser radiation at a wavelength of 980 nm and an intensity of 4 · 105 W/cm2 (Figure 2b). This is a significant improvement compared to the minimum heat penetration depth of approximately 90 µm realized when continuous diode laser radiation at an intensity of 1.4 · 105 W/cm2 is used to generate a peak temperature of 800 °C.

[Figure 2 a and b, FEM-results of the temperature-time-profile at the surface of  the coated steel substrate demonstrate the correlation between the depth of the heat affected zone and the interaction-time between the laser-beam and the sample (a). According to this simulation the heat penetration depth is reduced to 20 µm when using pulsed diode laser radiation at a pulse intensity of 4 · 105 W/cm2 and a pulse duration of 20 µs to generate a peak temperature of 800 °C (b)]

This significant decrease is due to the reduction of the interaction-time between the laser beam and the coated sample during the laser treatment which is realized with a beam deflection system by guiding the laser beam in meandering loops (Figure 3). The interaction-time, which refers to the period during which energy is transferred to an element of the coated surface is determined by the pulse duration between 2 – 20 µs when using pulsed laser radiation. When continuous diode laser radiation is used, on the other hand, this time is determined by the period required to cross the beam diameter of 340 µm at a given scan velocity. Due to the large size of the mirrors used within the beam deflection system to guide the raw laser beam with a diameter of approximately 30 mm, the scan velocity is limited to 2000 mm/s which results in a minimal interaction time of 170 µs in the latter case.

 

[Figure 3, Schematic view of the laser setup and the laser treatment strategy]

In accordance with the laser treatment strategy shown in Figure 3 there are four laser process parameters for the laser hardening process with pulsed diode laser radiation: The pulse energy EP, the pulse duration tp, the track offset dy and the number of pulses per position n which is the number of pulses deposited during the period required to cross the beam diameter at a given scan velocity. Based on the information obtained by FEM simulations these parameters are adapted systematically to achieve an optimal result. Important information about the mechanical properties of the coatings is obtained by means of nanoindentation hardness measurements carried out with a fisherscope HM500. These measurements, carried out with a Vickers indenter, a test load of 0,5 mN and a load time of 20 s, prove that the laser treatment was successful. By increasing the number of pulses per position the coating hardness is increased to approximately 800 HV (Figure 4). This is a significant improvement compared to an untreated coating (n=0) with a hardness of approximately 160 HV. Therefore the laser treatment led to a substantial reduction of the discrepancy between the hardness of the sol-gel coating and the reference coating produced by PVD. With regard to the low costs of the laser based process this is an incredible success.

[Figure 4, Vickers hardness of dried (n=0) and laser treated coatings (n = 1, n= 4,5) compared to the hardness of a coating produced by PVD, measured with a test load of 0,5 mN and a load time of 20 s (n refers to the number of pulses per position)]

Further investigations on the wear protection performance of the laser treated coatings are carried out by applying an industrially approved FE8-test-procedure. This test is carried out by Schaeffler KG in order to evaluate the protection performance of the laser treated coatings under realistic operating conditions. During this test carried out at loads of 30, 50 and 80 kN, the laser treated coatings show performances similar to currently used ceramic or diamond-like coatings produced by PVD (Figure 5). With these promising results the FunLas consortium has moved a step closer to the aim of developing an inline-capable, cost-effective process as an alternative for conventional PVD-processes.

[Figure 5, Comparison of Fe8-tested sample surfaces (diamond-like coatings produced by PVD (a) and laser-treated ceramic coating (b))]

Credit is due to the German Federal Ministry of Education and Research for funding the research depicted in this article within the framework of the funding measure “Material Processing with Brilliant Laser Sources” (MABRILAS). The author would also like to thank the Schaeffler KG, Merck KGaA, Darmstadt and DILAS GmbH for the excellent cooperation within the project consortium FunLas.