Processability and cracking behaviour of novel high-alloyed tool steels processed by Laser Powder Bed Fusion

Abstract

Concerning tooling applications, Laser Powder Bed Fusion (LPBF) enables new features such as internal cooling channels that can be implemented in cutting or shaping tools. Thus, higher cutting speeds are feasible thanks to the more efficient cooling that could not be obtained by channels fabricated with conventional methods. However, the alloys exploited for the cutting tools production usually contain high levels of carbon, which makes their LPBF processability challenging due to their high crack-susceptibility. In this work, an approach based on the use of basic physical/empirical indicators has been employed to map the processability of six novel high-alloyed tool steel grades. A large experimental campaign with variable energy densities, single and double passes, as well as different focal points was designed. The results exhibit highly dense but cracked parts. In particular, the LPBF processability deteriorates with increasing carbon content, suggesting that mostly chemistry, rather than process parameters, plays a key role in the determination of the LPBF feasibility. The cooling rate, cooling time between 800 °C and 500 °C, equivalent carbon content, solidification interval, martensite start temperature and volumetric energy density were employed as indicators to provide a rapid classification of processability. The work demonstrates that the combined use of the indicators can better explain the cracking behaviour of carbon-containing tool steels. At a screening level, this approach based on complementar use of physical/empirical tools, may significantly shorten the experimental effort during the design of new compositions, especially when dealing with crack susceptible alloys like carbon-containing tool steels.

Introduction

The layer-by-layer deposition nature of the laser powder bed fusion (LPBF) process enables new complex geometries for cutting and shaping tools, which cannot be realized with conventional manufacturing processes. In recent years, the fabrication of tools with LPBF processes has become more and more attractive since the process allows the realization of near-net-shaped tools with internal cooling channels to enhance the thermal control and productivity through higher cutting speeds, or with a lightweight design, to improve vibrations control during cutting operation. Fayazfar et al. (2018) reviewed the LPBF processability of several ferrous alloys. Nevertheless, in the scientific literature, very few works focus on LBPF processability of tool steels (see Table 1). According to Sander et al. (2016) tool steels with high strength and low toughness are susceptible to cracking, which makes the process very challenging. The short interaction times and high cooling rates typical of the LPBF process cause large thermal gradients resulting in fine microstructures with high strength and residual stresses. Moreover, Saewe et al. (2019) stated that the temperature gradients developed during the process induce a preferential grain growth along the building direction, which can cause brittleness and trigger cracking. In addition to this, the combination of high carbon content and high cooling rates lead to the formation of a very fine and brittle martensitic microstructure.

The majority of the scientific publications are focused on the understanding of the typical defects encountered, the analysis and evolution of the microstructure, the application of preheating, as well as other non-conventional strategies to reduce defects and achieve fully dense parts. Amongst the studied materials M2 HSS, AISI H13 HSS, AISI M50, HS 6−5-8−3, FeCrMoVC and FeCrMoVWC stand out. Table 1 gives a summary of the tool steels found in the literature with the main outcome regarding their processability.

It can be seen that the LPBF processability studies of High Speed Steels (HSS) are limited. Buls and Humbeeck (2014) showed that (Liu et al., 2011; Saewe et al., 2020, 2019, 2018) extensive cracking, delamination and distortion are present in the built parts. With the aim of investigating M2 HSS LPBF feasibility, (Buls and Humbeeck, 2014) and Liu et al. (2011) consider the aforementioned defects the result of the high thermal stresses induced during the process along with the high carbon content of the alloy. Similarly, Saewe et al. (2020) addressed the cause of cracking to the combination of the high carbon content and the rapid solidification of steels, investigating the LPBF feasibility of AISI M50 and H 6−5-8−3. To counterbalance these unwanted defects, especially cracking, preheating of the baseplate was performed. Buls and Humbeeck (2014) and Liu et al. (2011) stated that a baseplate preheating up to 473 K was sufficient to obtain crack – free parts of M2 HSS. Also Saewe et al. (2020) obtained crack – free specimens exploiting baseplate preheating up to 773 K for LPBF HS6−5-8−3 and AISI M50. This confirms that the beneficial effect of baseplate preheating comes with the reduction of temperature gradients inside the parts.

On the other hand, several works deal with the LPBF feasibility of AISI H13, a versatile Cr-Mo tool steel widely used in hot work tooling applications. Krell et al. (2018) and Mertens et al. (2016) stated that Cr-Mo tool steel exhibits a complex processing behaviour due to the combination of carbon content, rapid solidification and additional stresses caused by martensite formation, often resulting in cracks and distortion. Narvan et al. (2019) reported a significant reduction of cracks inside the AISI H13 parts whenever baseplate preheating is exploited, usually up to 573 K. Yan et al. (2017) observed a mitigation of residual stresses left in the parts whenever baseplate preheating is applied. Beal et al. (2008) investigated the effect of the scanning strategy on the processability of AISI H13 as an alternative solution for the reduction of thermal stresses, porosity and shrinkage.

Other publications focused on the LPBF processability of Fe85Cr4Mo8V2C1, Fe85Cr4Mo8V1C1 and for FeCr4Mo1V1W8C1 by Sander et al. (2017a,b) demonstrated that highly dense and crack-free parts can be obtained using baseplate preheating up to 773 K. Due to the high carbon contents, the microstructure usually comprises of martensite, retained austenite and carbides. Moreover, the high cooling rates of LPBF induce an extraordinarily fine microstructure, usually with elongated grains in the direction of the heat flow (building direction). This is consistent with all the other works, even though the compositions of materials are different.

Recently Platl et al. (2020a) studied LPBF of a cold work tool steel with high carbon content and investigated the evolution of defect structure. Extensive cracking along with porosity were observed at different energetic inputs. According to the work of Cunningham et al. (2017), at low energetic conditions, porosity is often generated due to insufficient melting of the powder (lack of fusion mechanism). Oppositely, as observed by Martin et al. (2019), at high energetic conditions, porosity may be formed due to gas entrapment in the melt pool (keyhole mechanism) mostly associated to melt pool instabilities. Concerning the predominant cracking mechanism, a clear correlation between solidification structure and potential stress accumulations, which result from the complex thermal cycle during LPBF, was found.

The major outcomes of the literature point out that LPBF of tool steels remains highly challenging due to the cracking phenomenon, which may be induced by different contributing factors. A large fraction of the published works is based on iterations of different chemical compositions starting from conventional tool steels. Simple analytical and empirical tools would be of great aid to the compositional development of the next generation tool steels. The chemistry of the materials employed in LPBF is usually adopted from alloys processed by the conventional manufacturing processes such as casting. Instead, during the LPBF processes, the rapid cooling and solidification may cause the defects related to the formation of fragile phases as well as high internal stresses. At a material development level, one solution can be to test each new alloy in an iterative way with large experimental plans. This approach, based on intensive experimental tests, should be aided with analytical tools that reduce the process iterations as well as provide insights to the further steps of the alloy design. The processability of tool steels is indeed one of the current topics in AM that can be tackled by combining computational and experimental approaches as discussed by Smith et al. (2016). For instance (Yan et al., 2018) demonstrated that the use of quick calculation analytical tools can be exploited for establishing the process, material, property relationship especially when large experimental runs are concerned. As discussed by Kouraytem et al. (2020), combining physics and data driven approaches, the results can also be better interpreted along different alloying grades and LPBF machine types providing. To the authors’ best knowledge, no previous works have attempted to assess the feasibility of using known rapid analytical and empirical calculations as LPBF processability indicators of new alloys, in particular tool steels.

In this work, the LPBF feasibility of six high-alloyed tool steels grades novel to the LPBF process were studied. In particular, the cracking behaviour was investigated. An experimental campaign with variable energy density, number of passes and focal position was applied to all the material compositions using an industrial LPBF system able to work with small quantities of powder. Among the proposed indicators, cooling rate (K/s), cooling time (s), martensite start temperature (K), solidification intervals (K) and volumetric energy density (J/mm³) were assessed. Crack density was associated to the process parameters as well as the analytical and empirical indicators. The results were also used to interpret the defect formation mechanisms.