Analysis of Laser Deep-Alloyed Samples
The cross-sectional areas of the samples produced with different laser powers are obtained in Fig. 5. They illustrate the increase in melt volume with increasing laser power. The samples are mainly free of pores. Cracks were found underneath the surface for laser powers of 3 kW and sporadically for laser powers of 4 kW. For higher laser powers of 5 kW, no cracks occurred, but some asperities were observed in the center regions of the former melt pool. With increasing laser power from 3 to 5 kW, the melt pool area increases from 8.3 to 19.9 mm2 (Fig. 6). Between 3 and 4 kW, the increase in melt pool area is 85%. With a further increase in laser power to 5 kW, the melt pool area increases further by 29%. The diagram shows that the hardness of the samples is affected by the increasing melt pool area. The micro-hardness decreases from 440 HV0.1 to 423 HV0.1. The deviation of the measured values is lowest for samples generated with a laser power of 4 kW, whereas it is highest for samples generated with 3 kW.
Figure 7 shows the average chromium and nickel content over the cross-sectional area of the sample as a function of the laser power. The measurement was taken close to the surface at a depth of 0.2 mm. Both the chromium content and the nickel content decrease with increasing melt volume, indicated by the melt pool area because of the increasing laser power. The standard deviation of the values shows only a small spread. Between 3 kW and 4 kW, the decrease in chromium content is 19%. With a further increase in laser power to 5 kW, the chromium content decreases further by 10%.
Figure 8 shows the deviation of hardness and chromium content within a sample generated with a laser power of 4 kW. The chromium content was measured at 0.2 mm underneath the surface. The micro-hardness was measured both, below and directly on the surface. The chromium content is between 4.8 and 5.7 wt%. On the surface, the average micro-hardness of 441 HV0.1 is slightly higher than below the surface with an average of 431 HV0.1. However, higher deviations are observed in micro-hardness at the surface in a range between 406 HV0.1 and 488 HV0.1 compared to those below the surface. Below the surface, the hardness varies between 410 HV0.1 and 454 HV0.1.
Figure 9 reveals the differences in the microstructure. The base material C15 consists of different grain sizes. Ferrite and perlite are found in the base material. The deep-alloyed material is etched with Kalling etchant, and depending on the used laser power, the martensitic structure is formed differently. 3 kW shows a rather inhomogeneous microstructure compared to 4 kW and 5 kW laser powers.
Analysis of Deep Rolled Samples
By analyzing the process forces in deep rolling, first information is gained during the processing. Changes in, for example, workpiece hardness or topography are visible in a slight increase or decrease in the rolling force Fr. The force curve is shown in Fig. 10 as an example for a laser deep-alloyed sample produced with 3 kW laser power. Regarding the force curve, different areas along the track can be found.
The base material has the lowest hardness and the highest roughness due to the pre-milling of the surface. All measured forces show oscillating behavior in this area. In contrast, the alloyed area has a polished surface due to the sample preparation so that fluctuations in force cannot be traced back to the topography. The sudden increase in the feed force clearly shows the entering of the deep rolling tool into the laser deep-alloyed area. It can also be deduced from the sudden drop of this force where the tool has left the laser deep-alloyed area again. In the example shown, the deep rolling force has a constant level between investigated positions 1–5. Only in the middle region of the laser deep-alloyed sample (position 3), a different behavior can be observed. Further information can be determined by investigating the plastic deformation induced by the deep rolling process. As track depth and width are influenced by the material flow behavior and material properties, e.g., the hardness and yield strength, changes within the area of a laser deep-alloyed sample could be observed for varied material composition. Several tactile measurements are taken at intervals of one millimeter along the track (cf. Fig. 10: position 1–5). The highest values are obtained in the middle of the samples, independent of the laser power (cf. Fig. 11). For a laser power of 5 kW, a crater was found in the middle of the sample, which made it impossible to measure the track geometry. The gradient of values determined within a sample decreases with increasing laser power. The depths and widths are set to constant levels, and the standard deviation of the results decreases with increasing laser power.
As the hardness was determined to decrease with increasing laser power, it was expected that both the track widths and the track depths in deep rolling would increase with lower material hardness. Instead, the track width increases slightly with higher material hardness. The different behavior of track depth and track width could be traced back to elastic behavior of the indenter. An elastic deformation of the tool ball during the deep rolling process could lead to the observed deviations.
Analysis of LiSE-Hardness Indentations
The measured laser-induced indentation geometries are shown in Fig. 12 for the deep-alloyed materials and the base material. The results demonstrate that the created indentation depth and indentation diameter increase when higher laser powers are used during laser deep alloying. The highest indentation depths and diameters are measured on the untreated base material. The calculated depth and diameter deviation are the smallest for the 4 kW laser power deep-alloyed samples. For lower laser powers of 3 kW and for higher laser powers of 5 kW, the measured depth and diameter deviation increase. It can also be obtained that the standard deviations are high for the measured indentation diameter and depth. This is especially the case for the untreated samples. The minimum and maximum deviations are indicated by the horizontal broken lines for the measurements on the base material. Furthermore, the coefficient of variation is determined for the created diameter, the indentation depth and from the hardness measurements on the samples. The coefficient of variation describes the ratio of the standard deviation to its mean value. Larger measured values may lead to larger absolute deviations. This stochastic parameter describes the repeatability of a measurement and makes it comparable to other measurement results. Thus, the coefficient of variation for the indentation diameter and the indentation depth is compared to the hardness coefficient of variation determined by conventional Vickers hardness measurements.
It is shown in Fig. 13 that a larger coefficient of variation also leads to a larger coefficient of variation for the indentation depth and indentation diameter. The largest coefficients of variation are calculated for the base material, whereas the second largest coefficients of variation are obtained for the 3 kW laser deep-alloyed sample. The lowest coefficients of variation are observed for the 4 kW sample. It can also be observed that the coefficient of variation for the diameter is on the average 2 times higher than the hardness coefficient of variation. The coefficient of variation for the indentation depth is on the average 6.5 times higher compared to the hardness coefficient of variation.
Comparison Between Methods
The descriptors determined utilizing the two methods presented above are plotted as plastic deformation over the average Vickers hardness of the laser deep-alloyed samples in Fig. 14. To compare the two methods, average values of the descriptors for each method are shown in this graphic. The position dependency is not considered, since it cannot be guaranteed that the values were determined at comparable positions using both methods. The comparison between the measurements shows that the indentation depth, the indentation diameter and track depth correlate with the hardness. A larger hardness value results in a lower geometrical value. This behavior cannot be clearly confirmed for the track width, since the value determined at the highest hardness differs from the mentioned trend.