La Metallurgia Italiana - Marzo 2018 by aimnet3

Metallurgia Italiana

International Journal of the Italian Association for Metallurgy

n. 3 Marzo 2018 Organo ufficiale dellâ&#x20AC;&#x2122;Associazione Italiana di Metallurgia. Rivista fondata nel 1909

in evidenza

Giornate di Studio

La fatica termica Aumento della produttività degli stampi attraverso un controllo specifico della fatica termica

9-10 maggio 2018 Bergamo Kilometro Rosso Organizzata da

Ad oggi il Comitato Tecnico Pressocolata compie quasi 20 anni. All’interno del Comitato tutti i suoi componenti hanno sempre lavorato con passione e permettetemi di dire con “Tenacia”, promuovendo numerose Giornate di Studio, che sono state interessanti e proficue per i partecipanti. Nel corso di questi anni i temi affrontati sono stati diversi e diversificati. Oggi con queste giornate sulla Fatica Termica ci siamo posti un obiettivo ambizioso: ne parliamo esaminando il problema da diverse e complementari sfaccettature, ma sempre con lo stesso spirito critico e più completo possibile. Credo che partecipare alle Giornate sia gratificante e culturalmente importante per tutti gli attori della pressocolata. Potrete vedere ed ascoltare, nelle presentazioni, sfumature che forse non avete mai preso in considerazione oppure che per svariati motivi avete sottovalutato o parzialmente ignorato. Benvenuti quindi a tutti, anche stavolta pensiamo non rimaniate delusi ma che usciate dalle giornate con “qualcosa in più” che vi permetterà di avere , speriamo, un approccio ai problemi della pressocolata più completo.

Per informazioni ed iscrizioni:

Fulvio Piana - Coordinatore delle Giornate

AIM · Associazione Italiana di Metallurgia Tel. 02-76021132 / 02-76397770 · E-mail: info@aimnet.it · www.aimnet.it

#aggiornamento #formazione #metallurgia #fatica #faticatermica #produttività #pressocolata

La Metallurgia Italiana

Metallurgia Italiana

International Journal of the Italian Association for Metallurgy

n. 3 Marzo 2018 Organo ufficiale dell’Associazione Italiana di Metallurgia. Rivista fondata nel 1909

International Journal of the Italian Association for Metallurgy Organo ufficiale dell’Associazione Italiana di Metallurgia. House organ of AIM Italian Association for Metallurgy. Rivista fondata nel 1909

Direttore responsabile/Chief editor: Mario Cusolito

n. 3 Marzo 2018

Direttore vicario/Deputy director: Gianangelo Camona Comitato scientifico/Editorial panel: Livio Battezzati, Christian Bernhard, Massimiliano Bestetti, Wolfgang Bleck, Franco Bonollo, Bruno Buchmayr, Enrique Mariano Castrodeza, Emanuela Cerri, Lorella Ceschini, Mario Conserva, Vladislav Deev, Augusto Di Gianfrancesco, Bernd Kleimt, Carlo Mapelli, Jean Denis Mithieux, Marco Ormellese, Massimo Pellizzari, Giorgio Poli, Pedro Dolabella Portella, Barbara Previtali, Evgeny S. Prusov, Emilio Ramous, Roberto Roberti, Dieter Senk, Du Sichen, Karl-Hermann Tacke, Stefano Trasatti Segreteria di redazione/Editorial secretary: Valeria Scarano Comitato di redazione/Editorial committee: Federica Bassani, Gianangelo Camona, Mario Cusolito, Ottavio Lecis, Carlo Mapelli, Valeria Scarano Direzione e redazione/Editorial and executive office: AIM - Via F. Turati 8 - 20121 Milano tel. 02 76 02 11 32 - fax 02 76 02 05 51 met@aimnet.it - www.aimnet.it

siderweb LA COMMUNITY DELL’ACCIAIO

Gestione editoriale e pubblicità Publisher and marketing office: Siderweb spa Via Don Milani, 5 - 25020 Flero (BS) tel. 030 25 400 06 - fax 030 25 400 41 commerciale@siderweb.com - www.siderweb.com La riproduzione degli articoli e delle illustrazioni è permessa solo citando la fonte e previa autorizzazione della Direzione della rivista. Reproduction in whole or in part of articles and images is permitted only upon receipt of required permission and provided that the source is cited. Reg. Trib. Milano n. 499 del 18/9/1948. Sped. in abb. Post. - D.L.353/2003 (conv. L. 27/02/2004 n. 46) art. 1, comma 1, DCB UD Siderweb spa è iscritta al Roc con il num. 26116 Stampa/Printed by: Poligrafiche San Marco sas - Cormòns (GO)

Anno 110 - ISSN 0026-0843 Additive manufacturing Comparison Between Microstructures, Deformation Mechanisms and Micromechanical Properties of 316L Stainless Steel Consolidated by Laser Melting I. Heikkilä , O. Karlsson, D. Lindela, A. Angra, Y. Zhong, J. Olsén 5 Impact of Process Conditions on the Properties of Additively Manufactured Tool Steel H13 processed by LBM L. Wu, T.Klaas, S. Leuders, F. Brenne, T. Niendorf

Lessons Learnt Through the Development of an Application-Specific Methodology for Metal Powder Characterisation for Additive Manufacturing M. Meisnar, S. Baker, C. Fowler, L. Pambaguian, T. Ghidini 20 In-situ Micro-tensile Testing of Additive Manufactured Maraging Steels in the SEM: Influence of Build Orientation, Thickness and Roughness on the Resulting Mechanical Properties K. B. Surreddi, C. Oikonomou, P. Karlsson, M. Olsson, L. Pejryd 27 Surface Oxide State on Metal Powder and its Changes during Additive Manufacturing: an Overview E. Hryha, R. Shvab, H. Gruber, A. Leicht, L. Nyborg

Influence of thermal aging on scc susceptibility of DSS 2304 in the presence of chlorides and thiosulphates F. Zanotto, V. Grassi, A. Balbo, C. Monticelli, F. Zucchi 40 Microstructural characterization and corrosion behaviour of SLM CoCrMo alloy in simulated body fluid M. Seyedi, F. Zanotto, E. Liverani, A. Fortunato, C. Monticelli, A. Balbo 45 Attualità industriale / Industry news Manifestazioni AIM

Properties Of Ti-6Al-4V Components Produced By Digital Metal Binder Jetting Technology edited by: M. Persson, R. Carlström, K. Gustavsson, S. Nilsson, C. Palmqvist, B. Brash - Höganäs, Sweden 52 ®

La Stampa 3D a metallo è arrivata nelle applicazioni oleodinamiche a cura di: AIDRO hydraulics 58 Atti e notizie / Aim news Calendario degli eventi internazionali / International events calendar

Normativa

l’editoriale La Metallurgia Italiana Le tecnologie di fabbricazione additiva per i metalli, probabilmente meglio conosciute con il termine inglese di additive manufacturing, raccolgono in questi recenti anni un interesse crescente, sia nel mondo della ricerca accademica, sia in diversi settori industriali: dall’aerospazio all’automotive, dal biomedicale a quello del gioiello senza dimenticare molti settori tradizionali della meccanica quali quello degli stampi, dei sistemi idraulici, degli scambiatori di calore.

La possibilità di generare forme complesse a volte impossibili da riprodurre mediante i processi tradizionali, per parti su misura o prototipi strutturali in tempi rapidi e con costi ridotti sono caratteristiche molto attraenti che stimolano l’immaginazione

Maurizio Vedani

sia dei ricercatori sia degli imprenditori più intraprendenti. I sistemi commerciali per l’additive manufacturing di parti metalliche sono in rapidissima evoluzione e consentono di intraprendere sperimentazioni su prodotti e materiali con relativa facilità. Vanno tuttavia anche sfatate delle false credenze: non si immaginano ancora i tempi in cui l’additive potrà rimpiazzare i processi di fonderia o di deformazione plastica a noi noti per la produzione di parti metalliche soprattutto in grandi serie. Questa tecnologia rappresenta invece un’ulteriore opportunità che si deve immaginare affiancata - ma non sostitutiva - dei processi più tradizionali, per ampliare le possibilità di produzione ed utilizzo di particolari funzionali e strutturali sempre più prestanti, leggeri, personalizzabili.

Sensibile a queste tendenze, l’Associazione Italiana di Metallurgia attraverso alcuni dei suoi Centri di Studio ha proposto in questi anni diverse manifestazioni sull’additive manufacturing dei metalli ed altre ne ha in preparazione nel prossimo futuro; non ultimo il congresso nazionale di Bologna previsto a settembre che sicuramente sarà platea di numerosi interventi sul tema. Anche La Metallurgia Italiana dedica per la seconda volta un numero focalizzato sulle tecniche ed i materiali per l’additive manufacturing, raccogliendo alcuni dei contributi più rilevanti presentati nel corso di un ampio congresso internazionale organizzato a Milano nell’ottobre 2017.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Comparison Between Microstructures, Deformation Mechanisms and Micromechanical Properties of 316L Stainless Steel Consolidated by Laser Melting I. Heikkilä, O. Karlsson, D. Lindell, A. Angré, Y. Zhong, J. Olsén A powder bed fusion laser technique (PBF-LS) was used to fabricate 316L stainless steel specimens for characterization of microstructures and micromechanical properties under uniaxial loading in-situ in a scanning electron microscope (SEM). Correlations between the microstructure, deformation mechanisms and mechanical properties were investigated. The results show that the morphology of the microstructure is very different when the sample building orientation was altered. In tensile test specimens that were machined from horizontally oriented rectangular beams, smaller grains and a more deformed microstructure were observed. Under uniaxial loading the yield strength and initial work hardening rate was highest in the horizontally built specimens. The uniform and total elongation was better for tensile test samples that were machined from vertically built rectangular specimens. Slip and twinning were the dominant deformation mechanisms with correlation to the observed texture. The observed anisotropic mechanical behavior can be explained by the differences in the distribution of deformed and sub-structured microstructures along the strain path.

KEYWORDS: ADDITIVE MANUFACTURING, STAINLESS STEEL, MICROSTRUCTURE, MECHANICAL PROPERTIES, DEFORMATION INTRODUCTION Powder bed fusion laser technique (PBF-LS) is an additive manufacturing (AM) method that uses layer by layer melting of metal powder to form solid parts of arbitrary geometries. The application of PBF-LS is growing in many areas such as aerospace due to its ability to manufacture complex parts with properties comparable or superior to ones of wrought materials. [1,2] Yet, being a promising component manufacturing method, it is crucial to understand how the process parameters, evolution of microstructure and achieved material properties are related to one another in order to be capable to produce functional parts for specified industrial needs. Through understanding the relation between the thermal history of the material and the microstructure, there is an obvious potential to tailor microstructures through graded deposition and optimized process parameters. Numerous studies have been conducted in an attempt to obtain a set of optimized process parameters to achieve desired properties. Delgado et al. (2011) [3] has shown that the various process parameters such as scanning speed and layer thickness influence the mechanical properties of iron-based products. Zhong et al. (2015) [4] and Saeidi et al. (2015) [5] indicated that the thermal history of a deposition has an important effect on the microstructure and on the final properties of

La Metallurgia Italiana - n. 3 2018

multi-layer stainless steel 316L through intragranular segregation and formation of sub-grain structure. Li et al. (2016) [6,7] showed that the different scanning path strategies influenced the mechanical properties of deposited 316L material. Shifeng et al. (2014) [8] showed that the spacing between molten pool boundaries of adjacent layers in the vertical building direction is much less than the one between two adjacent tracks in the horizontal building direction and correlated this to anisotropic mechanical properties of 316L material. In the present work, the focus is on characterization of the

I. Heikkilä, O. Karlsson, D. Lindell, A. Angré Swerea KIMAB AB, Box 7047, SE-164 07 Kista, Sweden

Y. Zhong, J. Olsén Department of Materials and Environmental Chemistry, Arrhenius Laboratory, Stockholm University, SE-106 91 Stockholm, Sweden © European Powder Metallurgy Association (EPMA). First published in the Euro PM2017 Congress proceedings

Additive manufacturing EXPERIMENTAL DETAILS The raw material of the investigations was gas-atomized 316L powder (Carpenter powder products AB, Torshälla, Sweden), see Table 1.

microstructure features of 316L within the melt pool boundaries by electron backscattered diffraction (EBSD) technique and correlating the as-built microstructure to the deformation mechanisms and micromechanical properties of the material.

Tab. 1 – The chemical composition of the 316L powder (in wt.%).

Element

Powder

0.014

0.70

1.69

0.014

0.004

17.8

12.5

2.38

0.04

0.09

Bal.

165ppm

The fabrication of the specimens was made through PBF-LS technique in N2 atmosphere at ambient pressure using an EOSINT M 270 system (EOS GmbH, Krailling, Germany). The laser source was a continuous Yb- fiber laser (200 W) with 0.1 mm diameter laser spot. In the building process the laser beam was moving at the speed of 850 mm/s using a Meander scanning strategy (45°). The space between adjacent lines was 0.1 mm. The building plate (250×250×20 mm) was lowered 0.02 mm after each layer was completed and a new layer of powder was dispersed evenly upon the melted top. The laser beam was exposed on the new powder surface and 3D specimens

(rectangular blocks) were made by repeating this layer-by-layer process. The longest axis of the rectangular blocks was aligned such that it was either parallel (horizontal building direction) or perpendicular (vertical building direction) to the building plate. The built blocks, with the dimensions 40x14x4 mm, were machined, by milling, into micro tensile samples usable in in-situ tensile testing in SEM, dimensions in fig. 1. The tensile straining direction coincides with the long axis of the block, see fig.1. The total length of the tensile specimens was 36.5 mm. A region of 2 x 1.4 mm was machined in the center of the specimen to assure that the microstructure deforms within the field of view.

c) Fig. 1 – Illustration of the geometry of the in-situ tensile specimens and their position within the as-built blocks they were machined from. a) Horizontal built blocks, b) Vertical built blocks and c) Dimensions of the micro tensile sample. 6

La Metallurgia Italiana - n. 3 2018

Additive manufacturing The surfaces of the produced pins were prepared for metallographic investigations. The final thickness of the polished specimens was 0.85 mm and the width to thickness ratio 1.65. A LEO 1530 scanning electron microscope upgrade to Zeiss Supra 55 equivalent (Carl Zeiss Microscopy GmbH, Muchen, Germany) was used for EBSD investigations with 15 keV and 0.5 µm step size. Tensile testing and pre-straining for microstructural investigation was made using a Gatan 5000 N tensile stage (Gatan Inc, Abingdon, United Kingdom) with a crosshead speed of 0.2 mm/min. Tensile properties were evaluated from two specimens for each block. Texture analysis was made using the MTEX toolbox for Matlab [9]. Crystal plasticity simulations were made using the VPSC code [10]. For the plasticity simulations slip on {111}<110> and twinning on {111}<112> was considered. A Voce hardening law with τ0 = 150 MPa, τ0 = 300 MPa, and θ0 = 300 MPa, = 100 MPa was used to describe the hardening behavior θ0 for both slip and twinning. No attempts were made to scale the corresponding hardening parameters to perfectly match the experimental data since this would have required much

more testing and since the required sample dimensions are not suitable for such a testing rig. Instead, reasonable values for wrought material were used in order to elucidate the effect of texture alone. RESULTS AND DISCUSSION Micro-mechanical properties Stress-displacement curves for the tensile testing are given in Fig.2. The repeatability between the separate measurements was good. The yield strength of the horizontal built specimens was clearly higher than for the verticallly built ones. The yield strength is also some 200 - 300 MPa larger than for typical wrought produced material of the same grade [8]. Further, the ultimate strength was higher for the horizontally built specimens, see Table 2 for the measured values. The two specimens also have very different work hardening characteristics, the horizontally built specimens with higher strength has high initial work hardening rate however it decreases rather fast. The uniform and total elongation is higher for vertical built specimen with a more constant work hardening rate.

Fig. 2 – Tensile properties of the horizontally and vertically built 316L specimens. The elongation is expressed as crosshead displacement, i.e. not strain.

Tab. 1 – Tensile data of as-deposited 316L samples.

Process and pin identification

Yield strength (MPa)

Ultimate tensile strength (MPa)

Horizontal built specimen 1

650

770

Horizontal built specimen 2

700

807

Vertical built specimen 1

635

698

Vertical built specimen 2

590

663

La Metallurgia Italiana - n. 3 2018

Additive manufacturing As-built microstructure Pole figures computed from EBSD data can be seen in Fig.3. The building process produces a sharp [110] type fiber texture along the building direction Z. As a result of this texture [1-11] directions are positioned within the X-Y plane. The microstructure viewed along the X-Y plane can be seen in Fig.4a. From this view the 45o meander scanning path can be seen. The grain size is somewhat smaller than the distance between the scanning lines, 100 µm. A more fine microstructure can be seen along the scanning path within the grains. This microstructure consists of “true grain boundaries” with relative large misorientations (black lines)

and more substructure features with low misorientations (white lines). Within this microstructure relative large intragrain misorientations also exist which is indicative of stored deformation. The microstructure viewed along the Z-X plane can be seen in Fig.4b. The morphology of the grains looks more wedge-like in this view and the finer microstructure along the scanning path cannot be seen. Thus, the effective grain size is larger. Quantitatively speaking the grains size determined by the linear intercept method of >10 o boundaries is ~13 and 23 µm for X and Z, respectively.

Fig. 3 – Pole figures from the as-build microstructure. The build axis Z shows a strong [110] fiber texture. Contours 1x, 2x, etc.

c) Fig. 4 – As-build microstructures as The maps are colored using inverse pole figure (IPF) IPF-X and IPF-Z notation as those are the tensile axes during subsequent horizontal and vertical loading respectively. Grain boundaries are colored using black lines for boundaries with misorientation >10 o and white lines for boundaries with misorientation 3-10 o. a) Microstructure viewed along the X-Y plane b) Microstructure viewed along the Z-X plane c) IPF notation used for coloring, also used in Fig. 6. 8

La Metallurgia Italiana - n. 3 2018

Additive manufacturing The presence of sub-structured, deformed and recrystallized microstructure in the tensile specimens is illustrated with respective colors in Fig. 5. It can be noted that the microstructure of the horizontal built specimen is more deformed along the direction where tensile load is applied, please no-

tice the red color representing deformed microstructure. In the applied deposition process the thermal stresses originating from rapid temperature fluctuations during melting/remelting, solidification and heat dissipation are high enough to induce plastic deformation.

Fig. 5 â&#x20AC;&#x201C; Local mis-orientation maps showing deformed (red), sub-structured (yellow) and recrystallized (blue) microstructure in the a) horizontally and b) vertically built pins in 3D. Deformed microstructure Microstructures after 0.8 mm pre-strain are given in Fig 6. Slip and twinning appears to be the primary deformation mechanisms for both horizontal and vertical loading. <0.1% deformation induced martensite was found and therefore not further considered. Twinning occurs primarily in grains close to <111> parallel to the tensile axis as these are most difficult to deform by slip according to Taylor theory [11]. On the other hand, slip occurs primarily on grains close to <100> parallel to the tensile axis. This is most easily seen in the large red grain on the left hand side of Fig. 5b. The large degree of twinning has increased the grain boundary fraction significantly. Both specimens have a grain size of about 5 Âľm using a similar line intercept methodology as was used for the as-build microstructures. A series of crystal plasticity simulations were made using the VPSC code [9] to further study the relation between loading direction and mechanical response. The texture given in Fig. 3 was used in the simulations along with a synthetic dataset with a random set of crystals, i.e. no texture. Uniaxial strain path was considered along X (horizontal loading) or Z (vertical loading). In a first set of simulations {111}<110>

La Metallurgia Italiana - n. 3 2018

slip alone was considered and in a second set of simulations also {111}<112> twinning. With regard to yield strength, see Fig.6a, the opposite behavior was actually observed, i.e. the yield strength is expected to be slightly larger for vertical loading however Fig. 2. shows the opposite trend. As already mentioned, however, the yields strengths are about twice as high compared to wrought material. Hence, the yield strength is likely governed by the distribution of the deformed microstructure observed in the as-build material, rather than by texture. Besides high yield strength the horizontal loading also results in a very high initial work hardening rate. One explanation for this could be latent hardening effects, the simulations predict deformation on a larger set of systems for horizontal loading along, see Fig. 6b. The simulations also suggest that a larger proportion of the total stain is carried out by twinning when deformed in horizontal, this is in line with the observations in Fig 6. It is however likely that the distribution of the deformed microstructure also plays a role for hardening so this phenomena is not fully understood. The elemental distribution of alloying elements can also be of significance [5]

Additive manufacturing

a) Horizontal load (X).

Stress-strain response. Note that vertical loading is expected to have higher yield strength.

b) Vertical load (Z).

Slip and twinning activity. Horizontal load activates more slip and twinning systems.

Fig. 6 – Crystal plasticity simulations using the VPSC code. The dataset showed in Fig 3 were used to simulate horizontal and vertical load. Random crystals refer to a synthetic dataset for a material without any texture. Fig. 6, microstructures after 0.8 mm prestrain (see Fig. 2). The maps are colored using IPF-X and IPF-Z notation as those are the tensile axes during horizontal and vertical loading respectively. Grain boundaries are colored using black lines for boundaries with misorientation >10 o, white lines for boundaries with misorientation 3-10 o and red lines for twin boundaries. Note the large increase in grain boundary volume compared to as-build microstructures, Fig. 4. CONCLUSIONS The relationship between the microstructure, deformation mechanism and micromechanical properties of 316L for horizontal and vertical loading were investigated. The following observations were made: • The morphology of the microstructure is very different when viewed along the X-Y or Z-X plane, Z being the vertical building direction. For the X-Y plane the structure resembles the scanning strategy with coarse grains between the scan lines and fine grains within the re-melted scanning path. Within the scanning path the microstructure also looks 10

somewhat deformed. • The building produces a strong [110] type fiber texture, consequently the X-Y plane contains [1-11] type directions. • Uniaxial straining along X (horizontal) and Z (vertical) both resulted in yield stress about twice as high compared to wrought produced material. For horizontal loading the yield stress and initial work hardening rate were highest, however the uniform and total elongation was overall better for vertical loading. • Crystal plasticity simulations were made in order to understand the observations, the yield strength differences cannot be governed by texture. Instead the distribution of the deformed microstructure is likely to explain the differences. The work hardening rates can to some extent be explained by the texture however the deformed layer must be further understood in order to fully understand the relation between microstructure and mechanical properties. Further studies e.g. by subsequent heat treatments are motivated as well as different strain paths.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing REFERENCES [1]

A. Ahmadi et al., Effect of manufacturing parameters on mechanical properties of 316L stainless steels parts fabricated by selective laser melting, Mat. Design,112 (2016) 328-338 [2] M. H. Farshidianfar et al., Effect of real-time cooling rate on microstructure in Laser Additive Manufacturing, J. Mat. Proc. Tech. 231 (2016) 468-478 [3] J. Delgado et al., Influence of process parameters on part quality and mechanical properties for DMLS and SLM with iron-based materials, Int. J. Adv. Manuf. Technol. (2012) 60:601-610 [4] Y. Zhong et al., Intragranular cellular segregation network structure strengthening 316L stainless steel prepared by selective laser melting, J. Nuclear Mat. 470 (2016) 170-178 [5] K. Saedi et al., Hardened austenite steel with columnar sub-grain structure formed by laser melting, Mat. Sc. Eng. A 625 (2015) 221-229 [6] K. Saedi et al., Transformation of austenite to duplex austenite-ferrite assembly in annealed stainless steel 316L consolidated by laser melting, J. Alloys and Comp. 633 (2015) 463-469 [7] J. Li et al., Microstructure and performance optimization of stainless steel formed by laser additive manufacturing, Mat. Sci. Tech. 32 (2016) 12: 1223-1230 [8] W. Shifeng et al. Effect of molten pool boundaries on the mechanical properties of selective laser melting parts, J. Mat. Proc. Tech. 214 (2014) 2660-2667 [9] F. Bachmann, R. et al.,Texture Analysis with MTEX - Free and Open Source Software Toolbox, Solid State Phenomena, 160 (2010) 63-68. [10] R. A. Lebensohn et al. A self-consistent anisotropic approach for the simulation of plastic def. and texture development of polycrystals: application to zirc alloys, Acta Metall., 41 (1993) 2611-2624. [11] G. I. Taylor. The mechanism of plastic deformation in crystals. Part I. Theoretical, Proc. Royal Society of London, 145 (1934) 362387.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Impact of Process Conditions on the Properties of Additively Manufactured Tool Steel H13 processed by LBM L. Wu, T.Klaas, S. Leuders, F. Brenne, T. Niendorf H13 tool steel, a versatile chromium hot-work steel, is widely used in demanding industrial applications such as high-pressure die casting. When processed by additive manufacturing (AM), innovative functionalities like conformal cooling can be implemented in these tools in order to optimize the thermal management as well as the linked processing cycle time. However, due to the high carbon content, robust processing of H13 requires high pre-heating temperatures, which cannot be achieved by many commonly available Laser beam melting (LBM) systems. This could be different for the Electron beam melting (EBM) process, where pre-heating temperatures of more than 1000 °C can be achieved. Against this background, the present study provides a first step towards process parameter development for H13 tool steel. For H13 processed by LBM it is shown that application relevant material characteristics, such as defectdensity, surface quality and mechanical properties, are strongly influenced by the processing parameters.

KEYWORDS: HOT WORK STEEL H13, LASER BEAM MELTING, LASER PARAMETER, MELTING POOL, SOLIDIFICATION CRACKS INTRODUCTION During recent years, the field of additive manufacturing (AM) gained steadily growing interest from industry and academia due to its outstanding potentials, i.e. design freedom, lightweight design, realization of complex inner structures, etc.[1]. In recent decades, different AM technologies were established

and improved, while numerous materials were processed. Two well-established methods of AM, Laser beam melting (LBM) and Electron beam melting (EBM) are under consideration in the current work. Although the basic principle is similar, i.e. layer-wise manufacturing from a powder bed, LBM and EBM have some differences, which are listed in Tab.1.

Tab. 1 – List of differences between LBM and EBM

Difference

LBM

EBM

Source

Laser beam

Electron beam

Powder

Fine

Coarse

Vacuum

yes

L. Wu, T.Klaas, S. Leuders H13 is a commercial hot-work steel. The standard composition (according to DIN standard) is listed in Tab.2. Recent investigations show that H13 in general can be processed by AM [2-4]. Studies focused on effect of pre-heating and processing windows on porosity and monotonic properties (cf. Fig.1) however, solid process-microstructure-property relationships allowing transfer of results to other AM facilities are not established so far.

Voestalpine Additive Manufacturing Center GmbH, Hansaallee 321, 40549 Düsseldorf, Germany

F. Brenne, T. Niendorf Institut für Werkstofftechnik (Materials Engineering), Mönchebergstr. 3, 34125 Kassel, Germany © European Powder Metallurgy Association (EPMA). First published in the Euro PM2017 Congress proceedings

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Tab. 2– Nominal chemical composition of H13 steel [4]

Elements

0.35-0.42 0.80-1.20 0.25-0.50

wt.%

Analysis of the melting pool, i.e. its morphology and dimensions, provides a first step towards understanding of the impact of processing parameters on microstructure and defect evolution for both, LBM and EBM. Thus, melting pool analyses could provide

4.80-5.50 1.20-1.50 0.90-1.10

Bal.

the basis for process parameter development in a joint fashion for both techniques. The current study reveals the impact of three parameters for one technique (LBM) only, studies focusing on EBM will be future work.

Fig. 1 – (a) Stress strain curves for samples without, with 200°C and with 400°C preheating [3] (b) Density contour plot as a function of scan speed and power [2]

Experimental details For this study, specimens have been processed under an argon gas atmosphere on a SLM 280HL machine (SLM Solutions GmbH) equipped with a high-temperature (HT) heating stage, employing an yttrium fiber laser with a maximum power of 700 W. The layer

thickness was set to 30 µm and the platform was heated to 200 °C, 300°C and 400°C. As raw material, gas-atomized H13 has been used. Detailed information on the powder is provided in Tab.3.

Tab. 3 – H13 powder used in the current study

Elements

wt.%

0.37

0.97

0.58

5.24

1.74

1.16

Bal.

Size distribution

D10

D50

D90

µm

18.2

28.5

43.5

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 2 – Design of specimens: single laser tracks for melting pool analysis and bulk material for porosity analysis For the investigation of melting pools and porosity using different parameters, specimen were designed as depicted in Fig.2. Using the SLM process, the specimens were built in z-direction and single laser tracks were scanned back and forth in x-direc-

tion. In order to investigate the influence of parameters, 3 x 9 specimens at temperatures of 200 °C, 300°C and 400°C were conducted according to Tab.4.

Tab. 4 – Parameter setting for 9 specimens at temperatures of 200 °C, 300°C and 400°C

Nr.

Laser Power (W)

100

200

300

Scan speed (mm/s)

450

Hatch distance (mm)

0.1

Nr.

Laser Power (W)

100

200

300

Scan speed (mm/s)

900

Hatch distance (mm)

0.1

Nr.

Laser Power (W)

100

200

300

Scan speed (mm/s)

1350

Hatch distance (mm)

0.1

La Metallurgia Italiana - n. 3 2018

Additive manufacturing For characterization, all specimens were cut in the y-z section, mechanically ground and polished, and etched by Nital 3%. For optical microscopy a digital microscopy Keyence VHX-5000 was used. Quantification of porosity and analyses of melting pools was conducted based on optical micrographs. Additionally, microstructure was characterized by scanning electron microscopy (SEM) employing a JEOL JSM-IT 300.

RESULTS AND DISCUSSION As mentioned in the experimental details section, 27 different specimens were built in total. Information on porosity and melting pool dimensions are provided in Tab.5.

Tab. 5 – Porosity and melting pool dimensions

T(°C)

200

300

400

Speed(mm/s) Porosity (%)

No.

Power(W)

100

450

200

Melting pool Width(µm)

Depth(µm)

0.47

116.19

76.50

450

0.02

167.95

156.70

300

450

0.01

191.01

111.23

100

900

5.28

97.47

74.82

200

900

0.03

115.25

72.19

300

900

0.09

158.36

94.03

100

1350

29.39

78.11

64.96

200

1350

3.75

94.86

72.08

300

1350

0.42

105.28

96.23

100

450

0.43

99.34

75.75

200

450

0.01

149.00

116.81

300

450

0.02

188.33

137.60

100

900

9.68

84.96

60.98

200

900

0.04

122.07

127.00

300

900

0.05

116.06

90.41

100

1350

26.91

83.15

62.97

200

1350

2.71

107.15

92.48

300

1350

0.06

116.65

84.81

100

450

0.27

114.32

77.06

200

450

0.23

115.84

75.39

300

450

0.01

200.78

166.90

100

900

8.12

84.92

68.80

200

900

0.01

125.50

113.82

300

900

0.10

134.04

154.33

100

1350

25.39

80.62

57.86

200

1350

2.02

94.02

61.85

300

1350

0.10

120.95

82.97

La Metallurgia Italiana - n. 3 2018

Additive manufacturing In order to analyze the interaction of various parameters on melt pool dimensions and evolution of porosity, Minitab 17 was applied to do the DoE analysis. Fig. 3 (a) and (b) reflect the influence of laser power, scan speed and temperature on the width of the melting pools. The factor temperature does have only minor influence, the

width of the melting pool is stronger influenced by laser power and scan speed. Similarly, the depth of the melting pools is affected. Fig.4 (a) and (b) reveal the same tendencies as Figure 3.

Fig. 3 – Influence of power, speed and temperature on the width of melting pools. (a) Interaction of main factors (b) contour plot for factors power and speed at 300°C

Fig. 4 – Influence of power, speed and temperature on the depth of melting pools. (a) Interaction of main factors (b) contour plot for factors power and speed at 300°C

Fig. 5 – Influence of power, speed and temperature on the porosity. (a) Interaction of main factors (b) contour plot for factors power and speed at 300°C

La Metallurgia Italiana - n. 3 2018

Additive manufacturing The decrease of scan speed and the increase of power results in the reduction of porosity in large parts of the investigated range as highlighted in Fig.5. This phenomenon indicates that porosity is strongly affected by width and depth of the melt pool in the parameter range investigated. The relationship between melting pool dimensions and porosity can be reflected not only from the data presented above, but also from metallographic analyses. Fig. 6 (a), (b) and (c) display optical micrographs from specimen processed with different parameters. With increasing power melting pools enlarge and pores surrounding melting pools gradually shrink and finally di-

sappear. Furthermore, defects can be classified into two types, i.e. inside and surrounding melting pools. Typical defects surrounding melting pools are pores, which are mostly filled with unmolten powder. Fig.7 (a) and (b) display the morphology of pores in the SEM and a corresponding schematic view. Obviously, the junctions of melting pools show high risk for evolution of pores. The schematic depicted in Fig.7 (c) provides a simple description of the underlying mechanism. Obviously, melting pool dimensions have to be set according to hatch distance and layer thickness to avoid this effect.

Fig. 6 â&#x20AC;&#x201C; Melting pools and porosity upon LBM with different parameters (speed 1350mm/s, hatch 0.1mm, temperature 300Â°C are constant) (a) power 100W, (b) power 200W, (c) power 300W

Fig. 7 â&#x20AC;&#x201C; Pores surrounding melting pools (a) morphology of pores as seen in the SEM (b) schematic related to the SEM image (c) schematic description of mechanism leading to porosity

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 8 â&#x20AC;&#x201C; Cracks inside melting pools (a) morphology of cracks in SEM (b) schematic related to the SEM image (c) schematic highlighting the mechanism of solidification cracking [5, 6]

Inside the melting pools, another type of cracks is frequently observed. Fig.8 (a) and (b) show that cellular and elongated structures accompanied by cracks can be seen in individual grains. The type of cracks seen initiates and grows alongside the boundaries of clusters of similar appearance, even across boundaries of melting pools. According to literature [5, 6], these cracks are assumed to be solidification cracks, which form primarily on high-angle grain boundaries (HAGBs) and result from segregation of critical elements. Fig.8 (c) depicts schematically the mechanism responsible for crack formation. Detailed analyses, however, will be done in future work and focus on local grain orientations and character of grain boundaries. Although porosity can be significantly reduced by increase of melting pool dimensions in the region investigated (at constant hatch distance and layer thickness), too high volume energy

density (VED) is detrimental. Fig.9 (a) and (b) display morphology and microstructure of a high VED specimen (power 500W, speed 450mm/s, hatch distance 0.1mm) in the y-z section. Cracks particularly evolve within the large melting pools (red boxes in Figure 9 (a)), simultaneously coarse dendrite structures evolve as can be seen in the SEM micrograph. This kind of solidification structure is strongly influenced by prevailing thermal gradients and growth rate of the solidification front (Fig.9 (c)). For AM, high-power LBM could speed up the manufacturing process. However, the specimens are more susceptible to cracks, non-favorable microstructure, high evaporation etc.

Fig. 9 â&#x20AC;&#x201C; High-power LBM specimens, (a) morphology (b) microstructure obtained by SEM (c) influence of G and R on microstructure [7]

La Metallurgia Italiana - n. 3 2018

Additive manufacturing CONCLUSIONS The current study focused on the influence of the three processing parameters (laser power, scan speed and temperature) in LBM on melting pool dimensions and porosity for hot-work steel H13. The following conclusions can be drawn: 1. In the investigated process parameter region, porosity of specimens decreases when laser power increases and scan speed decreases. This is due to an increase of width and depth of the melting pools. 2. Pores and cracks detected highlight the important role of

melting pool dimensions. Porosity is caused by an unsuitable arrangement of melting pool dimensions, hatch distance and layer thickness. 3. Cracks located inside of melting pools follow grain boundaries. Solidification cracking is expected to be affected by segregation of critical elements. 4. High energy density LBM causes more intense cracking, inappropriate microstructure evolution and evaporation of elements.

REFERENCES [1] [2] [3] [4] [5] [6] [7]

Schmidt, V. and M.R. Belegratis, Laser technology in biomimetics: Basics and applications. 2014: Springer Science & Business Media. Laakso, P., et al., Optimization and Simulation of SLM Process for High Density H13 Tool Steel Parts. Physics Procedia, 2016. 83: p. 26-35. Mertens, R., et al., Influence of Powder Bed Preheating on Microstructure and Mechanical Properties of H13 Tool Steel SLM Parts. Physics Procedia, 2016. 83: p. 882-890. Papageorgiou, D., C. Medrea, and N. Kyriakou, Failure analysis of H13 working die used in plastic injection moulding. Engineering Failure Analysis, 2013. 35: p. 355-359. Lu, L., D. Raabe, and W. Bleck, Characterization of the crack formation mechanism in Ni-based superalloy Inconel 738LC produced by Selective Laser Melting (SLM). 2015, Institut fĂźr EisenhĂźttenkunde, RWTH Aachen Aachen, Germany. Wang, N., et al., Solidification cracking of superalloy single-and bi-crystals. Acta Materialia, 2004. 52(11): p. 3173-3182. Kou, S., Welding metallurgy. New York, 1987.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Lessons Learnt Through the Development of an Application-Specific Methodology for Metal Powder Characterisation for Additive Manufacturing M. Meisnar, S. Baker, C. Fowler, L. Pambaguian, T. Ghidini Additive manufacturing is gaining prominence in high-tech sectors such as aeronautics and space, which recognise the importance of having an in-depth knowledge of the material feedstock used to produce their parts. Recently, the European Space Agency has started to develop its own expertise on powder characterisation within its UK-based Advanced Manufacturing Laboratory, in close collaboration with the European Space Research and Technology Centre. In this paper, a detailed study on the testing methods for determining powder and particle properties will be presented as well as storage and recycling considerations. Currently recognized standard procedures for metal powder characterisation will be critically reviewed. The developed testing methodology will be illustrated on a variety of powder samples. This study is part of a broader initiative that aims to support the establishment of a healthy powder supply chain for additive manufacturing, ensuring reliable and consistent powder feedstock for the space sector.

KEYWORDS: ADDITIVE MANUFACTURING, METAL POWDER, HIGH-TECH, TESTING METHODS

INTRODUCTION The purpose of this paper is to present the work that has been undertaken by the ESA-RAL Advanced Manufacturing Laboratory, based at ECSAT in Harwell (UK), on metallic powders for additive manufacturing (AM). The primary focus was placed on powder characteristics and related test methods, which were obtained through a careful review of the current landscape of international standards on the testing of metal powders. These standard methods were applied and reviewed critically, focusing on their strengths and weaknesses, in order to develop a methodology targeted at additive manufacturing feedstock powders. A literature survey showed that a number of scientific studies have been performed to date demonstrating the influence of the feedstock powder properties on the quality of final AM parts. For instance, it was found that a powder with a wider particle size distribution provides a higher powder bed density and therefore generates higher density parts as well as a smoother surface finish [1]. In contrast, a powder with a narrower range of particle size provides better flowability and results in higher strength and hardness [1,2]. Furthermore, it has been recognised that the selected powder handling and testing methods can have a significant impact as well. For example, especially in cases of broad particle size distribution, it has been found that the technique of powder sampling has a significant impact on the powder characteristics [3]. Finally, storage and recycling of the powder feedstock have been identified as key criteria for AM to be a safe and economically feasible manufacturing process. At

the same time, storage and recycling also have an impact on the final part quality. Powder storage in uncontrolled environments can cause humidity and oxidation which may result in limited flowability or contamination. In contrast, whilst recycling of metal powder is necessary for cost effectiveness, the sieved and reused

M. Meisnar, S. Baker European Space Agency, ESA-RAL Advanced Manufacturing Laboratory, Harwell-Oxford Campus, Fermi Avenue, OX110FD, Didcot, United Kingdom

C. Fowler Science and Technology Facilities Council, Rutherford Appleton Laboratory, Harwell-Oxford Campus, Harwell-Oxford Campus, Fermi Avenue, OX110QX, Didcot, United Kingdom

L. Pambaguian, T. Ghidini European Space Agency, ESTEC, Keplerlaan 1, PO Box 299, 2201 AZ Noordwijk, The Netherlands ÂŠ European Powder Metallurgy Association (EPMA). First published in the Euro PM2017 Congress proceedings

La Metallurgia Italiana - n. 3 2018

Additive manufacturing powder could have an adverse effect in terms of coarsening and higher surface roughness or increased hardness due to oxidation [4]. These studies carried out to date demonstrate that repeatable part characteristics rely on a thorough knowledge of the feedstock powder properties which highlights the need for standardised assessment methods [5]. The methodology developed at the ESA-RAL Advanced Manufacturing Laboratory is based on the existing ASTM standards on testing of metal powders, which were applied to a selected set of powders from different manufacturers. Currently, metallic powder is being produced by many suppliers with different specifications and target qualities. It is therefore not guaranteed that all metal powders are suitable raw material for AM of aerospace parts. Hence, for the purpose of this study, powders with the same nominal chemical composition were chosen from three different types of suppliers: a powder manufacturer (PM), a third party supplier (3P) and an AM machine manufacturer (commonly known as “verified” powder). One objective of this study was to determine the characteristic differences of these three types of powder and discuss the potential impact on final AM parts. The long-term vision is to understand each step from the production to the post-processing of AM feedstock powder and promote a

healthy supply chain of “space-quality” powder. International standardisation and method development Today, the full potential of AM for the aerospace industry is not fully exploited, one of the reasons being that there is a lack of available standards for raw material specifications and qualification of the final parts [5]. The ASTM committee F42 on Additive Manufacturing Technologies was created to bring the industry experts together to identify the specific standard requirements common in the AM industry. As part of this effort, the international standard ASTM F3049 “Standard Guide for Characterizing Properties of Metal Powders Used for Additive Manufacturing Processes” was created. It references many conventional international standards for measuring the characteristics of metal powders. A selection of currently available international ASTM and ISO standards on the testing of metal powders, which were the basis for the devised powder characterisation methodology, is shown in Fig.1. Some of these standards were applied to a number of different powders in order to determine their suitability for testing AM feedstock powders.

Fig. 1 – Overview of international standards used for metal powder characterisation

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Included are, for instance, methods for powder flow and density measurements which utilize the flow of powder through a funnel. However, in powder-bed based AM processes the powder is usually introduced through a hopper and spread out in a single layer on the build plate or on top of consecutive powder layers. This raises the question of the suitability of some of these standard methods with respect to testing powders as raw material for the AM process. In fact, various specialists argue that the applied measurement technique should be as representative of the AM process as possible in order to correctly establish the â&#x20AC;&#x153;process-abilityâ&#x20AC;? of powders in the according stress state [6,7]. An overview of the currently available techniques for ba-

sic powder characterisation and considerations about their strengths and weaknesses are shown in Fig. 2. The developed characterisation methodology is split into two phases: analysis of powder particle properties (Fig. 2, blue) and analysis of collective powder properties (Fig. 2, orange). The grey boxes indicate the most advanced and feasible (in terms of time and cost) techniques currently available. Furthermore, it was found that there is currently no technique for flow measurement that appropriately simulates the condition and stress state of powders as they are used in an actual AM build environment.

Fig. 2 â&#x20AC;&#x201C; Basic methodology with pros (blue) and cons (red)

More complex characteristics such as optical and thermal properties as well as hygroscopicity or inter-particle cohesion that may be of importance with regard to the AM process were not included at this stage. These will be part of further studies. The next sub-sections describe the application of some of these standard methodologies to a selected range of powders and the lessons learnt in terms of their suitability for AM feedstock powder. Furthermore, the characteristics of chemically identical powders from three different suppliers were compared in order to study the variability of powders currently provided on the market. The focus of this study was placed on plasma spheroidized Ti-6Al-4V. Particle Size Distribution A well-established method for measuring the particle size range of a powder is sieving (ASTM B214-16). It is still commonly used in industry due to its relative simplicity and large throughput, but remains a rather crude method if more detailed results are required. In contrast, image analysis via scanning electron microscopy 22

(SEM) and laser light diffraction are much more accurate and were therefore applied in this study. Image analysis via SEM allows direct measurement of the particle dimensions, but the time spent to analyse a sufficient number of particles for reliable statistics is excessive (in the range of days). The added benefit of this method is that it allows concurrent particle morphology analysis. In comparison, laser light diffraction, an indirect method through which the particle size is inferred from the scattering angle of the laser light upon interaction with the powder particles, calculates the particle size distribution (PSD) in a few minutes. The main difference of this technique compared to direct image analysis is that the resulting PSD is given as a volume density as opposed to a number density. This means that measurements via laser light diffraction, due to the nature of the technique, give no information about the number of particles analysed or their shape. Fig. 3 shows SEM images of the three selected Ti-6Al-4V powders (grade 5) acquired in unused condition from three different suppliers and their particle size distributions. It was observed that La Metallurgia Italiana - n. 3 2018

Additive manufacturing the PSDs of these three powders exhibit noticeable differences (Fig. 3d). It was found that the PSD of the powder acquired from a third-party supplier (3P) is noticeably biased towards smaller particles. The largest particles, up to 60 μm and slightly above, were found in powder PM. The standard ASTM F3049-14 mentions that powders with a significant fraction of small particles may have significantly reduced flow rates. This could lead to inconsistent deposition of the powder layers in the AM build chamber and therefore to defects in the final part. A reduced flow rate was indeed observed for powder 3P compared to MM and PM, as shown later.

Nevertheless, these observations on their own do not guarantee detrimental effects on the built part. For instance, a bias toward larger particle sizes, like in powder PM, can lead to better flow properties but reduced packing density and therefore increased porosity of the final part. Interestingly, the PSD of the “verified” powder from the machine manufacturer (MM) is exactly between the other two distributions of powders PM and 3P. Nevertheless, the sensitivity of these slight changes in PSD on built parts remains unknown and will be the focus of a separate study.

Fig. 3 – SEM backscatter electron images and particle size measurement: a) powder from machine manufacturer (MM); b) powder from powder manufacturer (PM); c) powder from third party supplier (3P); d) PSD (linear scale, volume distribution) measured via laser light diffraction (red: PM, green: MM, blue: 3P); Dx(10), Dx(50) and Dx(90) shown in table

Finally, for comparison, the diameters of ~ 500 particles were measured in the SEM and compiled to form a number distribution. Based on the assumption of spherical particles, the number distribution was converted into a volume distribution (as shown in Fig. 4, blue) which matches well with the results from laser light diffraction (red). The data acquisition and analysis for the

La Metallurgia Italiana - n. 3 2018

image analysis method took ~ 1 day, whereas laser light diffraction took only a few minutes. Therefore, laser light diffraction will be chosen for future studies and only fully automated image analysis methods (such as projection methods, see Fig. 2) will be considered as a feasible alternative.

Additive manufacturing

Fig. 4 - Particle size distribution: comparison between laser light diffraction (volume density: red) and image analysis method (particle number: orange, volume density: blue)

Morphology/chemical analysis Both particle morphology and chemical analysis were performed using SEM. Energy-dispersive X-ray Spectroscopy (EDS) confirmed the chemical composition of the powders as Ti-6Al-4V. Taking into consideration the detection limits of the analysis method (quantification of low-Z elements such as Oxygen not reliable with EDS), no signs of contamination or oxidation were found. It has to be noted that appropriate storage in Ar atmosphere was used. The particles from all three powders were observed to be primarily spherically shaped, with powder 3P showing exceptional surface smoothness, regularity and sphericity (Fig. 3c). An enhanced proportion of larger particles can be observed for powder PM (Fig. 3b) with some surface indentations and irregularly shaped particles or satellites. Powder MM consists of regularly shaped, near-spherical particles with some satellites (Fig. 3a). The number of smaller and larger particles in powder MM seems very evenly distributed, as indicated by the PSD in Fig. 3d. Image analysis via SEM was identified as the ideal tool for particle morphology studies, especially high-resolution imaging of a sub-set of particles, and was chosen as one of the preferred methods for future analyses. Another option is the use of fully automated projection methods, which can save time and increase the statistics (as explained in Fig. 2).

Powder Flow and Density The international ASTM standards provide a selection of different methods for the determination of metal powder flow. The angle of repose technique and use of the Hall Flowmeter funnel (ASTM B213-13) are two of the most well established methods. However, their suitability with respect to the AM process is questionable as neither flow through a funnel nor cone-like deposition of powder are characteristic to powder-bed based techniques. Despite this fact, the Hall Flowmeter was used to measure the flow time of an equal (non-compacted) volume of the three powders in this study. Powder PM showed the best flow characteristics, as expected, due to its bias towards larger particles compared to the other powders. In contrast, the longest flow time was measured for the 3P powder with a larger proportion of smaller particles. Throughout the measurements, it was found that the results are very operator-dependent, as the stop watch operation is manual. While a single operator may achieve good repeatability between measurements, the reproducibility of the flow time between different operators is not guaranteed. Another method to determine the flow characteristics of powders that has recently gained prominence is the avalanche angle [6,7], where good flow is represented by smaller angles (Fig. 5). As a result, the fastest flowing powder PM (Fig. 5a) has the smallest avalanche angle (Îą).

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 5 - Powder flow determined via avalanche angle (fâ&#x20AC;Ś flow time)

Both Hall Flowmeter and the avalanche angle method are fairly easy to use and were found to provide qualitatively matching results, each method representing the results differently (time or angle). More advanced equipment designed for dynamic flow and inter-particle force measurements such as rheometers certainly provide very useful information as well. However, an instrument specifically designed for measuring the characteristics of metal powders in actual AM conditions with respect to the deposition method and stress state of the powder under the correct atmospheric conditions is required to accurately simulate the process. Currently, there are ASTM standards for the determination of the apparent powder density via the Hall Flowmeter (ASTM B21213) and the reduced density after compaction of the powder with a tapping apparatus (ASTM B527-15). The ratio between these two characteristics, also known as Hausner ratio, provides information on the packing density and flowability of the powder. According to this test, all three powders in this study were found to meet the ASTM quality criteria (e.g. powder 3P had a Hausner ratio of 1.09, while > 1.25 is considered poor flowability according to ASTM B527-15). This method of establishing the quality of powders with respect to their bulk densities was fairly easy to use, but also involves measurement errors resulting from differences between operators and the manual volume measurement. Other methods for powder density measurements such as gas

La Metallurgia Italiana - n. 3 2018

displacement techniques (pycnometry) or X-ray computed tomography are more accurate and shall be investigated as part of a future study. Storage and Recycling Fig. 6a and b demonstrate that storage (~1 month) in an inappropriate environment (air) can result in agglomeration (caking) and poor flowability which could potentially affect the final part. This highlights the importance of a suitable storage environment and proper handling procedures. The quantities of powder deposited in the build chamber during the AM process are usually very high compared to the powder used for building the actual part. Therefore, the powder remaining in the chamber after the build process is often sieved to remove agglomerated particles and then recycled. Despite the sieving, some powder particles, especially those in close vicinity to the built part, may be altered or attached to other particles. Fig. 6c shows powder particles after an unknown number of cycles in the build chamber. Some recycled particles were found to be irregularly shaped or have a number of satellites attached to them. Changes in powder morphology and PSD after recycling could have detrimental impacts on the final parts and will therefore be studied in the future in more detail.

Additive manufacturing

Fig. 6 - Impact of inappropriate storage (1 month) in air and recycling: a) new powder after storage – caking; b) new powder storage – decreased flowability; c) SEM image of recycled powder particles

Conclusions and perspectives It is well known that the quality and reproducibility of AM parts strongly depends on the feedstock powder characteristics. However, current standard methods for evaluating metal powder properties are not entirely representative of the AM process, highlighting a need for more process-related international standards, especially with respect to flow and density measurements. At the same time, the exact specifications of a “space quality” powder are not yet defined and ESA and its industry partners are working towards developing powder grades for space applications.

A methodology was devised for basic characterisation of metal powders, which will soon be repeated on a number of powders from additional manufacturers and other materials and complemented with more advanced techniques. It was shown that powders available on the market, which are nominally identical chemically, can have characteristics that differ quite drastically from one manufacturer to another. This in turn, may have substantial impacts on the quality of the built parts. A standard powder specification developed for space industry could potentially mitigate this risk.

REFERENCES [1] A.B. Spierings, N. Herres, G. Levy, Influence of the particle size distribution on surface quality and mechanical properties in AM steel parts, Rapid Prototyp. J. 17 (2011) 195–202. doi:10.1108/13552541111124770. [2] B. Liu, R. Wildman, C. Tuck, I. Ashcroft, R. Hague, Investigation the Effect of Particle Size Distribution on Processing Parameters Optimisation in Selective Laser Melting Process, in: Solid Free. Fabr. Symp., Austin, TX, 2011: pp. 227–238. doi:10.1017/ CBO9781107415324.004. [3] G.B. Basim, M. Khalili, Particle size analysis on wide size distribution powders; effect of sampling and characterization technique, Adv. Powder Technol. 26 (2015) 200–207. doi:10.1016/j.apt.2014.09.009. [4] V. Seyda, N. Kaufmann, C. Emmelmann, Investigation of Aging Processes of Ti-6Al-4 V Powder Material in Laser Melting, Phys. Procedia. 39 (2012) 425–431. doi:10.1016/j.phpro.2012.10.057. [5] J.A. Slotwinski, E.J. Garboczi, P.E. Stutzman, C.F. Ferraris, S.S. Watson, M.A. Peltz, Characterization of Metal Powders Used for Additive Manufacturing, J. Res. Natl. Inst. Stand. Technol. 119 (2014) 460–493. doi:10.6028/jres.119.018. [6] A.B. Spierings, M. Voegtlin, T. Bauer, K. Wegener, Powder flowability characterisation methodology for powder-bed-based metal additive manufacturing, Prog. Addit. Manuf. 1 (2016) 9–20. doi:10.1007/s40964-015-0001-4. [7] M. Krantz, H. Zhang, J. Zhu, Characterization of powder flow: Static and dynamic testing, Powder Technol. 194 (2009) 239– 245. doi:10.1016/j.powtec.2009.05.001.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing In-situ Micro-tensile Testing of Additive Manufactured Maraging Steels in the SEM: Influence of Build Orientation, Thickness and Roughness on the Resulting Mechanical Properties K. B. Surreddi, C. Oikonomou, P. Karlsson, M. Olsson, L. Pejryd Selective laser melting (SLM) is frequently used additive manufacturing technique capable of producing various complex parts including thin-wall sections. However the surface roughness is a limiting factor in thin sections produced by SLM process when strength is the main criterion. In this study, the influence of build orientation, thickness and roughness on the resulting mechanical properties of as-built test samples was investigated. Various thin sheets of EN 1.2709 maraging steel built in horizontal and vertical orientations produced by SLM were investigated using in-situ micro-tensile testing in a scanning electron microscope. The mechanical strength and deformation mechanisms were analyzed and explained based on thickness and build orientation. Increased ductility was observed in thicker samples as well as in the horizontal build samples. The results illustrate the potential of the in-situ test technique and aspects important to consider in design guidelines for thin AM structures.

KEYWORDS: SELECTIVE LASER MELTING, IN-SITU MICRO-TENSILE TESTING, BUILD ORIENTATION, DEFORMATION MECHANISM AND MARAGING STEEL INTRODUCTION Selective laser melting (SLM), which is one of the most widely used metal additive manufacturing (AM) techniques is capable of producing various end-use parts with complex shapes and intricate designs which can be difficult to produce with conventional methods. The SLM process uses a laser beam to melt and fuse the metal powder through layer by layer approach to build 3-D parts directly from a CAD design [1]. The SLM method, generally characterized by high cooling rates and high thermal gradients, results in non-equilibrium or fine microstructures [2].This can be a limiting factor in case of brittle materials which cannot resist high internal thermal stresses and lead to quench cracking. The rapid cooling makes SLM process to be used in inert or controlled atmosphere, such as nitrogen or argon, to avoid oxidation and to reduce the contamination from the atmosphere. Usage of inert or controlled atmospheres can also create problems such as entrapment of gas bubbles, formation and entrapment of inclusions which ultimately lead to porosity, unintended impurities and even delamination between layers [3, 4]. The surface roughness is another limiting factor in terms of surface quality especially in the production of thin sections [5]. The poor surface quality is due to surface porosity, sticking of the powder particles to the edge of the thin sections and waviness of solidified built-up layers during SLM. Previous studies on the surface La Metallurgia Italiana - n. 3 2018

quality and roughness show that a careful selection of the size distribution of the powder particles as well as laser parameters can improve the surface quality [5]. The presence of porosity and impurities, build orientation, layer thickness and the interface between the built-up layers as well as the surface quality are main contributing factors when considering the tensile strength and elongation at break for the

Kumar Babu Surreddi, Mikael Olsson Materials Science, Dalarna University, SE-791 88 Falun, Sweden

Christos Oikonomou Uddeholms AB, SE-683 85 Hagfors, Sweden

Patrik Karlsson, Lars Pejryd Department of Mechanical Engineering, Örebro University, SE701 82 Örebro, Sweden © European Powder Metallurgy Association (EPMA). First published in the Euro PM2017 Congress proceedings

Additive manufacturing sections below 1 mm thickness. These factors can be controlled by careful and optimized selection of SLM process parameters, and a good design approach is necessary in order to produce good quality thin sections [6]. It is necessary to study the mechanical properties of thinner sections when strength is the main design parameter. The tensile properties of the sections with the thickness lower than 2 mm have to be carried out by micro-tensile testing. Majority of mechanical properties of maraging steel studies were focused on bulk samples. So the aim of the present work is to study the influence of surface roughness, built orientation and thickness (≤ 1 mm) on the tensile strength

as well as to compare with the bulk tensile properties. Experimental method The material used for this study is maraging steel which is characterized by good tensile strength, high toughness and good weldability. Maraging steels are mainly used in different tooling applications. The material is well suited for additive manufacturing due to its good weldability and resistance to cracking when subjected to rapid cooling and heating [7, 8].

Tab. 1 - The material composition in wt. % of MS1 [9] Fe Bal.

Ni 17-19

Co 8.5-9.5

Mo 4.5-5.2

Ti 0.6-0.8

The maraging steel powder with a commercial name EOS Maraging Steel MS1 (equivalent to 1.2709 alloy composition according to European standard) was used for this study and the nominal composition of MS1 is given in Table 1 [9]. All samples were build using a selective laser melting machine EOS M290 equipped with 400 W Ytterbium fiber laser source working under nitrogen atmosphere. A 40 µm build layer thickness was used for all samples. Bulk tensile test bars were prepared from the selective laser melted build blocks according to DIN EN ISO 68921. The bulk tensile tests and micro-tensile tests were carried out in as-built condition (without any heat treatment) at room temperature. Microtest 5000 N tensile stage from Gatan Inc., with the capability to record stress-strain curves from the Microtest software was used for the micro-tensile testing. This microtest module from Gatan offers in-situ tensile tests of the specimens inside the scanning electron microscope (SEM). The micro-tensile tests were performed on as-received rough samples outside of the SEM and also on polished test samples inside the SEM with the motor speed of 0.1 mm/min. The force and elongation data were collected during the micro-tensile testing and, SEM images were continuously captured from the surface of the polished sample by continuous scanning during the in-situ testing to study the plastic deformation.

Al 0.05-0.15

Cr, Cu each ≤ 0.5

C ≤ 0.03

Mn, Si P, S each ≤ 0.1 each ≤ 0.01

In order to study the mechanical properties of thin plates and to compare them with the bulk material properties, square sections (20 x 20 mm2) in parallel and perpendicular directions to the built orientation, and plates with 0.5 mm, 0.8 mm and 1 mm thickness were built. The microtensile test samples as shown in figure 1 were prepared using water jet cutting in horizontal and vertical directions to the build orientation. The polishing of micro-tensile test samples was carried out by standard grinding and polishing procedures. The surface topography of the as-built and polished samples was characterized using Wyko NT9100 optical profilometer. The 3-D surface roughness parameters were measured at four different areas (3 x 6 mm2) on asbuilt and polished samples. These parameters provide roughness, spatial and hybrid information for 3-D surfaces. The microstructure of the samples was characterized by optical, stereo and scanning electron microscope. SEM-EDS (energy dispersive spectroscopy) was used to analyze the composition of the material and inclusions. Light optical microscopy was carried out using Leica DMRME, stereo microscopy was carried out using Leica MZ16 and SEM was carried out on rough, polished and fractured samples using Zeiss Ultra 55 FEG-SEM equipped with an Oxford Instruments INCA energy dispersive X-ray spectroscopy (EDS) system.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 1 - Schematic diagram showing how horizontal and vertical micro-tensile test samples prepared from the SLM build plate and microscopic images of side, top and front view of the build plate

Results and discussion Figure 1 shows the schematic diagram of the build plate showing the build orientation and the direction of microtensile test samples, and optical and stereo images of side, top and front views in order to understand the build orientation. Table 2 shows the average 3-D surface roughness parameters (S parameters) for as-built samples where Sa is the average surface roughness, Sq is the root mean square roughness and Sz is the Ten Point Height over the complete 3-D surface. The S parameters show a slight increasing trend with increasing thickness. Figure 2 shows the surface

roughness contour plots of as-received and polished samples of 0.5 mm thick micro-tensile samples in horizontal direction to build orientation. The roughness map of asreceived sample clearly reveals the horizontal build orientation and the polished sample shows uniform and smooth surface with 3-D surface roughness, Sa approximately 1 µm. However, it can be seen that there are bright elongated regions along the build orientation representing the porosity and surface defects (figure 1 – front view).

Tab. 2 - Average 3-D surface roughness parameters of as-built plates Thickness 0.5 mm 0.8 mm 1 mm

Sa (µm) 5.2 ± 0.99 5.78 ± 0.72 5.89 ± 0.94

Sq (µm) 6.98 ± 1.45 7.69 ± 0.85 7.84 ± 1.29

Sz (µm) 63.43 ± 18.8 65.35 ± 4.97 77.66 ± 8.55

Fig. 2 - Surface roughness maps of as-built (left) and polished (right) samples of 0.5 mm thick micro-tensile samples in horizontal direction to build orientation. Figures 3(a) and 3(b) show the SEM secondary electron and back scattered images of the unpolished sample respectively. These images show clearly that the top surface of the as-built sample is characterized by built layers (blue arrow), partially molten powder particles adhered to the surface (red arrow) and dark regions in between the built layers (black arrow). The dark regions were characterized with SEM-EDS and was found to be rich in Ti, Al and O, prefera-

La Metallurgia Italiana - n. 3 2018

bly titanium oxide and alumina or combination of both as shown by previous studies [4]. The oxides were deposited generally in between the layers as they float to the top of the molten pool during the SLM process due to lower density [5]. The entrapment of slag or gas within the layer may sometimes lead to weak metallurgical bonding between the built-up layers (blue arrow). Large pores or inclusions on the top of the sample as shown in the figure 1 (front view)

Additive manufacturing can be observed on the polished surface. The roughness of the unpolished samples is mainly due to the surface porosity, partially adhered powder particles and the relatively rough solidified built-up layers. Polishing, removing the top 50-60 Âľm, will remove the adhered particles can obtain 3-D surface roughness Sa ~1 Âľm but reveal the surface porosity. Previous studies on the surface roughness of SLM parts reveal that the surface quality can be increased by selecting smaller layer thickness [5, 10]. Table 3 presents the in-situ SEM micro-tensile test results for the polished samples of 0.5, 0.8 and 1 mm thickness prepared in horizontal and vertical directions to the build orientation. Figure 4(a) shows the force-elongation plots and figure 4(b) shows the engineering stress-strain diagrams for polished 0.5, 0.8 and 1 mm thickness samples. It is difficult to accurately obtain the stress strain data for

the as-built rough samples due to pronounced surface roughness. As can be seen, the ultimate tensile strength (Rm) and the plastic strain at fracture of 0.8 and 1 mm samples are higher as compared to the 0.5 mm thick samples. The horizontal samples exhibit slightly higher elongation at break as compared to vertical samples. Previous studies also revealed the tensile strength of maraging steel in the range of 1100 to 1200 MPa with the ductility of about 8 to 12 % [7, 11, 12]. However micro-tensile test results reveal higher engineering strain values. This is due to the calculation of engineering strain is based on the gauge length of about 5 mm and the elongation during the test was obtained from the distance between the holding pins which are about 26 mm.

Fig. 3 - SEM images ((a) secondary electron mode and (b) backscattered electron mode) of 1 mm thick sample

Tab. 3 - In-situ SEM micro-tensile test results of polished samples Direction thickness 0.5 mm 0.8 mm 1 mm

Horizontal Rm, MPa 1088 1148 1185

Eng. strain, % 20 27 29

Vertical Rm, MPa 1110 1154 1173

Eng. strain, % 18 24 27

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 4 - Force-Elongation curves of 0.5, 0.8 and 1 mm thick polished samples (a) and engineering stress - strain curves of 0.5 and 1 mm thick polished samples (b) obtained during in-situ SEM micro-tensile testing Figure 5 shows the SEM micrographs of 0.5 mm thick polished samples taken during the in-situ micro-tensile testing at various plastic strains after necking. As can be seen, there is a clear necking in both samples resulting in a final fracture at about 60Â° angle to the tensile direction. The bottom images in figure 5 reveal the presence of large pores and cracks at the edges at 18 % of strain just before the fracture. Pores or impurities in between the built-up layers can open up and extend as large pores when the microstructure is exposed to high uniaxial deformation. The lower elongation values for the vertical samples can be connected to the presence of pores and impurities along the build lines which are perpendicular to the tensile direction. Figure 6 shows SEM images of fracture surfaces after micro-tensile testing of 0.5 mm polished samples. The fracture surfaces of horizontal and vertical build samples show

large voids including pores which were resulted from the entrapped gas and the impurities due to slag formation. In addition to the defects, large solid area reveals ductile type fracture. The presence of a high number of pores can be detrimental for the mechanical properties such as fatigue resistance and the tensile strength. Thus the lower tensile strength for 0.5 mm samples as compared to 1 mm samples (figure 4(b)) is due to low amount of solid contact area and the presence of large and high number of pores in the cross section. When the thickness is reduced below 0.8 mm, the amount of porosity and inclusions play major role in strength. The strength and plastic strain at failure sharply decrease with the reduction of thickness below 0.8 mm [13]. Thus, the density and the porosity is main contributing factors in the strength of thin sections produced in SLM process [12].

Fig. 5 - SEM micrographs of 0.5 mm thick polished samples in horizontal (top) and vertical (bottom) directions to the built orientation at various stages of plastic strain during in-situ micro-tensile testing La Metallurgia Italiana - n. 3 2018

Additive manufacturing The main advantage of micro-tensile testing is the possibility to study the deformation mechanisms of the as-build microstructures and the influence of porosity. Figure 7 shows a surface defect before and after micro-tensile testing observed on a horizontal sample in the SEM. It shows

large oxide inclusion on the surface and can be observed that several cracks were initiated and extended after the micro-tensile testing. This type of surface defects can be detrimental for fatigue strength.

Fig. 6 - SEM fracture micrographs of 0.5 mm polished horizontal direction samples (a) and (b), and vertical direction samples (c) and (d)

Fig. 7 - Influence of surface defect and inclusions on the deformation behavior during the micro-tensile test

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Tab. 4 - Bulk tensile test results

Parallel to build (H) Perpendicular to build (V)

Y.S 0.2% (MPa)

Rm (MPa)

Elongation (%)

1105 ± 5 1138 ± 12

1200 ± 5 1197 ± 22

12 ± 0.5 9±3

Bulk tensile test results presented in table 4 reveal the ultimate tensile strength is about 1200 MPa for samples built in parallel and perpendicular direction. However, parallel built samples exhibits higher elongation values as compared to perpendicular built samples similar to the micro-tensile test results. The yield strength (Y.S) is higher in perpendicular built samples as compared to the parallel built samples. The ultimate tensile strength of 1 mm thick samples is closer to the bulk tensile strength. The tensile strength is lower when the thickness is less than 1 mm which can be connected to the amount and size of the porosity and impurities. Conclusions In this study, plates of various thicknesses (0.5, 0.8 and 1 mm) were prepared from EOS maraging steel MS1 using SLM method to study the influence of thickness, built orientation and surface quality on tensile properties. The results were compared with bulk tensile properties obtained from bulk tensile bars. The 3-D surface analysis results show a slight increasing trend with increasing thickness of the samples. The as-built surface is characterized by rough surface due to surface porosity, built-up layers, partial sticking of powder

particles and high quantity of slag in between the built-up layers. Polishing 50 to 60 µm can remove the surface defects and can obtain polished surface with approximately 1 µm average 3-D surface roughness (Sa). The in-situ micro-tensile testing can be used to obtain detailed information of the deformation mechanism of AM samples as well as to study the mechanical properties of micro-tensile test samples (< 2 mm). The micro-tensile test results show that the ultimate tensile strength of thin samples depends on the porosity and the inclusions. The tensile strength values are lower for 0.5 mm thick samples as compared to 1 mm samples due to presence of large and high amount of porosity. Bulk tensile strength are comparable to 1 mm thick samples. Higher elongation was observed in the parallel build samples in both bulk and micro-tensile test samples. Acknowledgements The authors gratefully acknowledge the financial support from the Swedish Knowledge foundation (KK Stiftelsen). The authors would like to thank Mr. Torbjörn Holmstedt and Ms. Karolina Johansson from Lasertech LSH AB for providing the SLM samples.

REFERENCE [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]

Rombouts, Marleen. "Selective Laser Sintering/Melting of Iron-Based Powders (Selectief laser sinteren/smelten van ijzergebaseerde poeders)." (2006) Murr, Lawrence E., et al. "Metal fabrication by additive manufacturing using laser and electron beam melting technologies." Journal of Materials Science & Technology 28.1 (2012): 1-14. Rombouts, Marleen, et al. "Fundamentals of selective laser melting of alloyed steel powders." CIRP Annals-Manufacturing Technology 55.1 (2006): 187-192. Thijs, L., et al. "Investigation on the inclusions in maraging steel produced by selective laser melting." Innovative Developments in Virtual and Physical Prototyping: Proceedings of the 5th International Conference on Advanced Research in Virtual and Rapid Prototyping, Leiria, Portugal, 28 September-1 October, 2011. CRC Press, 2011. Strano, Giovanni, et al. "Surface roughness analysis, modelling and prediction in selective laser melting." Journal of Materials Processing Technology 213.4 (2013): 589-597. Yadroitsev, I., et al. "Strategy of manufacturing components with designed internal structure by selective laser melting of metallic powder." Applied Surface Science 254.4 (2007): 980-983. Kempen, Karolien, et al. "Microstructure and mechanical properties of Selective Laser Melted 18Ni-300 steel." Physics Procedia 12 (2011): 255-263. Hunt, J., F. Derguti, and I. Todd. "Selection of steels suitable for additive layer manufacturing." Ironmaking & Steelmaking 41.4 (2014): 254-256. Material data sheet EOS MaragingSteel MS1 https://cdn3.scrvt.com/eos/c88047245bff2c4b/2f494ef432d0/MS-MS1-M280_ M290_400W_Material_data_sheet_05-14_en.pdf Bacchewar, P. B., S. K. Singhal, and P. M. Pandey. "Statistical modelling and optimization of surface roughness in the selective laser sintering process." Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 221.1 (2007): 35-52. Yasa, Evren, et al. "Microstructure and mechanical properties of maraging steel 300 after selective laser melting." Solid Freeform Fabrication Symposium Proceedings. 2010. Tan, Chaolin, et al. "Microstructure and Mechanical Properties of 18Ni-300 Maraging Steel Fabricated by Selective Laser Melting." (2016). Adam, Guido AO, and Detmar Zimmer. "On design for additive manufacturing: evaluating geometrical limitations." Rapid Prototyping Journal 21.6 (2015): 662-670.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Surface Oxide State on Metal Powder and its Changes during Additive Manufacturing: an Overview E. Hryha, R. Shvab, H. Gruber, A. Leicht, L. Nyborg Quality and usefulness of the powder for additive manufacturing (AM) are strongly determined by the surface composition of the powder. In addition, taking into account harsh conditions during AM process, significant changes in the powder surface chemical composition are taking place, limiting powder recyclability. Hence, knowledge concerning amount of oxides, their composition and spatial distribution on the powder surface determines further powder recycling. This communication summarizes possibilities of qualitative and quantitative analysis of powder surface chemistry by surface-sensitive chemical analyses using XPS and HR SEM coupled with EDX. The effect of alloy composition, AM process applied and powder handling on the surface composition of the powder are addressed. Results indicate significant enrichment in the thermodynamically stable surface oxides in case of high-alloyed powder for both, EBM and LS processes. A generic model for the oxide distribution, depending on the alloy composition and powder surface degradation during AM manufacturing, is proposed.

KEYWORDS: OXIDES, LS PROCESS, SEM, OXIDE DISTRIBUTION

INTRODUCTION The metal powder is the base material for powder bed or blown powder additive manufacturing technologies. At the same time, there is a lack of understanding of the effect of powder properties on the properties of the AM fabricated components, processability of the powder by specific AM process, AM process robustness, etc. Nowadays most powder for AM applications is produced by Electrode Induction melting Gas Atomization (EIGA) or Vacuum Induction Gas Atomization (VIGA), characterized by high-purity but at the same time high cost as productivity is low and manufacturing costs are high. In order to improve process efficiency and taking into account high cost of the powder, powder is typically recycled. At the same time, there is limited understanding of the powder degradation during AM process. Hence, typically applied industrial approach when it comes to the powder degradation are based on empirical approach developed by each user in-house based on the components manufactured, hardware utilized as well as alloys of interest. Metal powder used for AM is characterised by a surface area that is about 10 000 times larger than the surface area of the bulk material of the same mass. This results in the significantly higher surface reactivity of the metal powder in comparison with the bulk metal. Reactivity is increasing with increasing surface area/decreasing particle size. At the same time it is important to emphasize that the pure metallic surface does not exist at ambient conditions – metallic surfaces are covered by surface oxide as well as absorbed species in order to minimize their surface energy. Hence, surface of the metal powder is typically covered by nanometre thick oxide layer that has a complex structure and composition [1-4]. Characteristics of the surface oxide formed 34

are determined by the thermodynamic stability of the oxides formed by the metal/alloy as well as the history of the metal surface exposure to the surrounding conditions during powder manufacturing and handling [1-7]. Temperature and time of the exposure in combination with the oxidation potential of the surrounding atmosphere determines thickness, structure and composition of the oxide formed on the powder surface. Oxide stability itself has an important effect on the extent of the surface oxide formation during manufacturing and handling – better powder passivation in case of reactive powders as e.g. Al, Ti, etc. At the same time, stability of the surface oxide present determines behaviour of the powder during processing as well as type and risk of the defects formation. Therefore, knowledge concerning surface chemistry of the powder and its changes during AM processing is of vital importance in order to establish required powder manufacturing,

Eduard Hryha, Ruslan Shvab, Hans Gruber, Alexander Leicht, Lars Nyborg Department of Industrial and Materials Science, Chalmers University of Technology, Rännvägen 2a, SE-412 96, Gothenburg, Sweden. © European Powder Metallurgy Association (EPMA). First published in the Euro PM2017 Congress proceedings

La Metallurgia Italiana - n. 3 2018

Additive manufacturing handling and recycling routines for a specific alloy and AM technique. The aim of this study is to demonstrate the application of the advanced microscopy and surface analysis methods for analysis of the metal powders for additive manufacturing. Focus is placed on powder bed methods – laser sintering (LS) and electron beam melting (EBM). Surface composition for different kinds of metallic powder with respect to material type and alloy composition are presented. Changes in surface oxide chemistry of the powder during AM processing as well as powder handling are discussed. Experimental Procedure Focus is placed on stainless steel 316L and nickel based superalloy powder Hastelloy X, produced by VIGA, and nickel based superalloy powder Alloy 718, produced by plasma atomization. Powder samples were prepared by lightly pressing the powder onto soft aluminum plates. Chemical analysis of the powder surface was performed by means of X-ray photoelectron spectroscopy (XPS), using PHI 5500 instrument. The analyzed area during XPS analysis was about 0.8 mm in diameter and thus a large number of particles (~100 particles) were analyzed at the same time, giving statistically reliable average result. Composition of the surface compounds was estimated by curve fitting of the characteristic peaks, and their intensities were corrected using standard relative sensitivity factors. Quantification of the results was performed by calibration measurements on pure elemental standards. Determination of the surface oxide layer thickness and compositional profiles was done by alternating ion etching and XPS analysis. Evaluation of the thickness of the surface oxide layer was performed using theoretical model involving the effect of the setup geometry (etching and measurement angles) with respect to the surface geometry (roughness) on the etching profiles and integrated in the software “Powder XPS calculator v.1.2.4”, developed by authors for powder analysis [8]. The etching was performed in argon gas with an accelerating voltage of 4 kV giving an etching rate from 3 to 5 nm·min-1. Surface oxide distribution and morphology was studied by high-resolution scanning electron microscope (HR SEM) LEO Gemini 1550, equipped with INCA Energy/X-Max analyzer. Recycled powders were sampled after AM processing utilizing EBM process (Arcam A2 machine) and LS processing (EOS M290). Hastelloy X powder degradation was studied in case of LS process (EOS M290) and Alloy 718 powder was studied after EBM processing (Arcam A2X).

Fig.1. Powder surface is clean, there are no obvious particulate oxide phases present on the powder surface. Powder is covered by a by thin uniform iron oxide layer, ranging between 2 to 4 nm, depending on the powder manufacturing route. However, significant changes in the appearance of the powder surface are observed after EBM-processing, see Fig.2. Even though powder seems to be unaffected at low magnification, observation of the powder surface at higher magnification clearly shows appearance of the particulate features on the powder surface, sizing up to 200 nm, that are typically characterised by irregular shape. Enrichment in Cr, Mn and Si was also detected in the features based on the EDX analysis [9]. Detailed analysis of the powder surface chemistry by XPS analysis [9] indicate almost no change in the thickness of the iron oxide layer, covering powder surface – 2.9 nm in case of virgin powder and 2.6 nm in case of recycled powder. However, significant changes were observed in the relative fractions of cations for oxide phases on the powder surface, see Fig.3. In case of virgin powder, initial decrease in iron cation content with ion etch depth indicates removal of the iron oxide layer, while maintained level at greater etch depth is connected to the contribution from the un-etched regions on the powder surface. The increased levels of cation content of Mn and Cr until ~10 nm indicate presence of the fine particulates of similar size, also evident from high magnification SEM in Fig.1. In case of recycled powder, similar initial behaviour of the iron oxide content is evident, indicating similar thickness of the homogeneous iron oxide layer, covering most of the powder surface. However, for larger etch depths (>3 nm), situation is very different. Higher content in Cr-cation level is evident, indicating formation of the Cr-rich particulate oxides on the powder surface. The increase in relative Cr-cation content with etch depth, indicate that the thickness (and hence size) of the Crcontaining oxide products is greater than the final etch depth of 20 nm applied. This is in correlation with the SEM observations, see Fig.2, where particulate features sizing up to 200 nm are evident. The EDX analysis of these features confirms enrichment in Cr as well [9]. At the same time, decrease in Mn-cation content compared to that of virgin powder is observed. This is attributed to the Mn-sublimation during EBM processing as powder is kept at high temperature (around 800°C) in vacuum. The same counts for surface contamination of powder by Zn, present in the case of virgin powder but fully absent in case of EBM processed powder.

Results and discussion Stainless steel powder Stainless steel powder as 316L is still the main working horse in case of the laser sintering (LS) characterised by well-established process parameters and robust manufacturing window, supplied by all of the LS hardware manufacturing. This type of alloy has a very well established gas atomization technology for powder fabrication and hence 316L powder of very good quality, produced by variety of the gas atomisation techniques, can be acquired. Typical powder for AM, produced by VIGA, is characterised by very good purity from the surface composition point of view, see La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 1 - SEM micrographs of gas atomized stainless steel powder 316L (Fe-16Cr-10Ni-2Mo-2Mn-0.75Si), showing appearance the powder surface.

Fig. 2 - SEM micrographs of gas atomized stainless steel powder 316L after one EBM cycle (recycled powder), showing appearance of the particulate features on the powder surface.

Fig. 3 - Relative cations concentration in oxide present on the surface of virgin powder vs. etch depth (left) and recycled powder vs. etch depth (right).

Fig. 4 - SEM micrographs of gas atomized stainless steel powder 316L after one LS cycle (recycled powder), showing appearance of the particulate features on the powder surface.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing In case of laser sintering, see Fig.4, effect of the one cycle of processing by laser sintering on the powder surface chemistry is much less evident than in the case of EBM processing, see Fig.2. In general, powder surface seems to be un-affected and only in some places appearance of the spherical particulate sizing up to 200 nm can be observed. In this case, oxide particulates have a dark appearance and perfect spherical shape and are enriched in Mn and Si, based on EDX analysis. Ni-base superalloy powder Results for Hastelloy X powder, manufactured by vacuum induction gas atomisation (VIGA), are presented in Fig.5, indicating high surface purity of powder, characteristic for VIGA. From XPS, it is found that the powder surface is covered by a thin (around 1 nm) uniform nickel oxide/ hydroxide layer with very rare presence of fine particulate features of thermodynamically stable Al-Cr-Ti-Si oxides [7]. However, the surface composition of the powder is significantly changed after laser sintering (LS) process. The SEM observation of the powder surface at high magnification clearly reveal extensive surface coverage by fine particulate features sizing around 20 nm, sporadically forming larger agglomerates on the powder surface,

see Fig.6. The EDX analysis indicate that these features are rich in Cr and Al, see Fig.6. The XPS analysis, see Fig.7, confirms the strong presence of Cr and Al in oxidized form in the surface oxide products after LS-processing. Based on the presented results it can be concluded that powder surface after LS processing is enriched in thermodynamically stable oxides rich in Cr and Al. These oxide phases are very difficult to be removed during the rapid LS processing and hence can be detrimental for the mechanical performance of the final components. However, sporadically much more oxidised metal particles can be found in the powder after LS-processing, see Fig. 8. Such metal particles were not observed in the virgin powder, their appearance already after processing in the laser sintering machine is at first strange. However, by sampling powder at the atmosphere outlet in the build chamber it was detected that almost all the powder there possess this type of oxidation, see Fig. 8. In case of fine particles (~20 Âľm) they are fully oxidised. In case of coarse powder particles with size about 50 Âľm they are oxidised from one side only. This is believed to be the powder from the interface between the volume of material scanned by laser beam and powder bed.

Fig. 5 - SEM micrographs of Hastelloy X powder, produced by VIGA in as-received state.

Fig. 6 - SEM+EDX analysis of the LS-recycled Hastelloy X powder.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 7 - Detailed XPS spectra of Cr (left) and Al (right) on the surface of the Hastelloy X powder in virgin and recycled state, indicating larger oxide fraction in case of recycled powder.

Fig. 8 - SEM+EDX analysis of the “burned” Hastelloy X powder, sampled after one LS-cycle close to the processing atmosphere outlet.

In case of electron beam melting it is typically expected that the issues with the powder recycling in connection to the powder oxidation are less critical as the powder bed is kept under high vacuum during the whole EBM process. However, observation of the powder surface in case of high-alloyed powder indicate presence of strong degradation of powder surface. Typically used in EBM process, like plasma atomised powder, is characterised by high purity, see Fig. 9, with visible dendritic structure and absence of pronounced formation of particulate oxide features on the powder surface. Analysis of the powder from the powder cake after EBM processing indicate though significant changes in the powder surface appearance as well as powder surface chemistry, see Fig. 10. Significant

coverage of the powder surface by the fine particulate features is observed, increasing with the exposure time during the EBM processing. Even analysis of the powder from the powder hopper, not exposed to EBM process, but only exposed in the powder hopper for number of cycles, indicate appearance of the similar powder surface oxidation but to lesser extent. The reason for such powder degradation during EBM process is believed to be connected to the long-term exposure of the powder – dozens and even hundreds of hours – to the high temperature (about 1000 °C) and apparently not good enough vacuum to avoid powder oxidation. This preliminary study clearly indicates importance of the detailed study of the powder degradation during EBM processing.

Fig. 9 - SEM micrographs of Alloy 718 powder, produced by plasma atomisation in as-received state.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 10 - SEM micrographs of Alloy 718 powder after four cycles of EBM processing. Conclusions Combination of the advanced surface and microscopy techniques allows evaluation of the oxygen distribution in the powder as well as chemical composition of the specific oxide features, present on the powder surfaces. Powder for additive manufacturing is typically fabricated using advanced powder fabrication techniques as EIGA, VIGA or plasma atomisation and is characterised by spherical shape and high-purity. The powder is typically covered with homogeneous oxide layer formed by the main element (iron oxide in case of stainless or tool steels and nickel oxide/hydroxide in case of Ni-based superalloys). The thickness of the oxide layer is between 1 and 4 nm, depending on alloy composition, powder manufacturing method and powder handling. Rare presence of the particulate oxide features with size up to 20 nm, rich in oxygen-sensitive elements, can be observed. Improper powder handling as well as processing by additive manufacturing techniques can lead to significant increase in surface coverage of the powder by thermodynamically stable oxides. Stainless steel powder shows rather minor changes in surface oxide state after laser sintering.

More significant coverage of the powder surface by the particulate oxide features is observed in case of 316L powder after EBM processing. Similar tendency is observed in case of Ni-based superalloys where rather extensive degradation of the powder surface was observed in case of EBM processing of Alloy 718 powder. In case of LS-processing, extensive degradation of some fraction of the powder was detected. Results indicate that combination of the advanced surface and microscopy analysis techniques the=e extent of the powder degradation during EBM and LS processing to be clearly determined and hence establish the limits of the powder recyclability. Acknowledgements Support from the Chalmers Areas of Advance in Materials Science and Production as well as funding from the strategic innovation program LIGHTer, provided by Vinnova, are gratefully acknowledged. Further thanks are also extended to Mr Jonas Olson from University West and Mrs Karin Fransson from Swerea IVF for sampling of the powder for analysis.

REFERENCE [1] T. Tunberg, L. Nyborg, , Powder Metall., 1995, vol. 38, no. 2, pp. 120-129. [2] L. Nyborg, I. Olefjord, Powder Metall. Int., 1988, no. 1, pp. 11-16. [3] E. Hryha, C. Gierl, L. Nyborg, H. Danninger, E. Dudrova, Appl. Surf. Sci., 2010, Vol. 256, No.12, pp. 3946-3961. [4] R. Shvab, E. Hryha, L. Nyborg: ” Surface chemistry of the titanium powder studied by XPS using internal standard reference”, Powder Metall., 2017, Vol. 60, No. 1, p. 42-48. [5] K. Zumsande, A. Weddeling, E. Hryha, S. Huth, L. Nyborg, S. Weber, N. Krasokha, W. Theisen, Mater. Char., 2012, vol. 71, pp. 66-76. [6] E. Hryha, R. Shvab, M. Bram, M. Bitzer, L. Nyborg: “Surface chemical state of Ti powders and its alloys: Effect of storage conditions and alloy composition”, Appl. Surf. Sci., in press, doi:10.1016/j.apsusc.2016.01.046. [7] R. Shvab, A. Leicht, E. Hryha, L. Nyborg: ”Characterization of the Virgin and Recycled Nickel Alloy HX Powder Used for Selective Laser Melting”, Proc. World PM2016, ISBN/ISSN: 978-1-899072-47-7. [8] C. Oikonomou, D. Nikas, E. Hryha, L. Nyborg, Surf. Interface Anal., 2014, Vol.46, No.10-11, pp. 1028-1032. [9] A. Leicht, R. Shvab, E. Hryha, L. Nyborg, L. Rännar: “Characterization of virgin and recycled 316L powder used in additive manufacturing”, Proceedings of the SPS2016 conference, Lund, Sweden, 2016.

La Metallurgia Italiana - n. 3 2018

Corrosione Influence of thermal aging on SCC susceptibility of DSS 2304 in the presence of chlorides and thiosulphates F. Zanotto, V. Grassi, A. Balbo, C. Monticelli, F. Zucchi In this paper, the effect of brief aging within the 650–850 °C temperature range on the resistance of DSS 2304 to stress corrosion cracking (SCC) was discussed. Slow strain rate tests (SSRT) were performed on both as-received and heat treated DSS 2304 in the standard NACE TM-0177 solution at pH 2.7 and 25 °C in the presence of 10-3 M S2O32-. High SCC susceptibility was detected with aging at 650 °C for 60 min, at 750 °C for 10 and 60 min and at 850 °C for 10 min, in good agreement with preceding results obtained for pitting resistance and degree of sensitization to intergranular corrosion. The SCC susceptibility is likely connected to chromium concentration levels in the depleted zones, which depend on the adopted aging times and temperatures.

KEYWORDS: DUPLEX STAINLESS STEELS, STRESS CORROSION CRACKING, SLOW STRAIN RATE TESTS, THIOSULPHATE INTRODUCTION Since some years the corrosion behaviour of lean duplex stainless steels (LDSS) is investigated at the “A. Daccò” Corrosion and Metallurgy Study Centre and the effects of brief aging treatments between 650 °C and 850 °C on pitting resistance and on the degree of susceptibility (DOS) to intergranular corrosion (IGC) were investigated on LDSS 2101, DSS 2304 and LDSS 2404 [1-3]. Aging of DSS 2304 between 650 and 850 °C, from 5 to 60 min, negatively influenced its pitting corrosion and IGC resistance. The critical pitting temperature (CPT) obtained in 0.1 M NaCl drastically decreased after aging at 650 °C for 60 min, but heat treatments of 5 and 10 min did not produce any effect. The sample treated at 750 °C for 10 min showed a CPT value 10 °C lower than that of the as-received one. The CPT value further decreased by extending the aging time to 60 min at the same temperature. At 850 °C, CPT decreased with only 5 min aging and did not show significant variations when the treatment time was increased up to 60 min. DOS to IGC measurements performed with Double Loop Electrochemical Potentiokinetic Reactivation (DL-EPR) method in 33% H2SO4 with 0.45% HCl at 20 °C, highlighted that DSS 2304 aged 60 min at 650 °C was the most susceptible to IGC, followed by that aged 10 and 60 min at 650 °C. Conversely, after aging at 850 °C from 5 to 60 min, DSS 2304 evidenced DOS values slightly lower than 1, which indicated a very low susceptibility to IGC [2]. These findings were related to the chromium impoverishment in regions adjacent to chromium carbides, produced after heat treatments at 650 and 750 °C. Instead, the lower influence of heat treatments at 850 °C on the alloy corrosion resistance was ascribed to partial chromium replenishment due to quick chromium diffusion.

The stress corrosion cracking (SCC) susceptibility of LDSS 2101 was also studied, both before and after heat treatments between 650 and 850 °C in standard NACE TM0177 solution in the presence of thiosulphate ions (S2O32−) [4-6]. In fact, as proposed by Tsujikawa et al. [7], the use of S2O32− in replacement of H2S gas can minimize the health hazards in laboratory tests and therefore can reduce the costs to ensure safe working conditions [8]. As-received LDSS 2101 resulted susceptible to SCC in standard NACE TM-0177 solution in the presence of S2O32− already at concentration of 10-4 M, at pH 2.7 and 25 °C [4]. SCC susceptibility significantly increased by increasing S2O32- content. After brief heat treatments between 650 and 850 °C, the resistance to SCC of LDSS 2101 got worse at increasing aging temperature and time [5,6]. In this work, the SCC behaviour of heat treated DSS 2304 was studied in the standard NACE TM-0177 solution at pH 2.7 and 25 °C in the presence of S2O32- at 10-3 M. These tests aim at filling a knowledge gap because very few data are reported in the literature about the SCC behaviour of

F. Zanotto LT Terra&AcquaTech - Università degli Studi di Ferrara, via Saragat 4A, 44122 Ferrara

V. Grassi, A. Balbo, C. Monticelli, F. Zucchi Centro di Studi sulla Corrosione e Metallurgia "A. Daccò"

La Metallurgia Italiana - n. 3 2018

Corrosion lean duplex in environments containing hydrogen sulphide [9].

(supplied by Outokumpu S.p.A. under annealed conditions), having the nominal chemical compositions (wt. %) reported in Tab. 1.

EXPERIMENTAL PART The tests were performed on DSS 2304 stainless steels Tab. 1 – Nominal chemical composition (wt.%) of DSS 2304.

DSS

DSS 2304

0.02

4.8

0.3

0.10

bal.

Tensile samples with an overall length of 23 cm and a gauge portion of 20x5x1.5 mm were machined from a 1.5 mm thick steel sheet. The samples were aged for 10 and 60 min at 650, 750 °C and 850 °C and then air cooled. The resulting microstructures were observed by Zeiss EVO MA15 scanning electron microscope (SEM) in back-scattered electron (BSE), in order to reveal the presence of secondary phases. The susceptibility to SCC was evaluated by SSRT, with a strain rate of 1x10-6 s-1. After thermal aging, the samples were ground parallel to the stress direction down to 800 grit emery papers and screened by a two-component epoxy varnish, so leaving only the gauge portion exposed to the solution. SSRT were performed by inserting the sample in an electrochemical cell, which was filled by deaerated and thermostated 5% NaCl and 0.5% CH3COOH solution (the basic standard solution NACE TM-0177 [10], without saturated H2S gas) in the presence of 10-3 M Na2S2O3, at 25 °C and pH 2.7. During each test, the stress - strain curve was recorded and the corrosion potential values (ECOR, versus Saturated Calomel Electrode (SCE)) were measured. Reference SSRT in air at 25 °C were also carried out. Each test was performed in triplicate. The SCC susceptibility was evaluated by the ratio (R) between the percentage strain to fracture (εf%) in the test solution and that in air at 25 °C. R values equal to or higher

than 0.8 were considered an index of immunity to SCC [11]. At the end of the tests, the gauge length section of the samples were observed with optical stereomicroscope and side surfaces were examined by OM, after polishing and etching with Beraha’s reagent, with the purpose to analyse crack initiation and morphology. RESULTS Fig. 1 shows the microstructures of the short transverse sections of DSS 2304 sheets under both as-received conditions (Fig. 1-a) and aged for 60 min at 850 °C (Fig. 1-b). Elongated austenitic grains (lighter phase) were distinguishable in the ferritic matrix (darker phase). Due to the low molybdenum content in this alloy, no χ and σ secondary phases were detected after heat treatments [12]. Very small black precipitates were observed at the α/γ grain boundaries in the sample aged for 60 min at 650 °C and on all samples aged at 750 °C or 850 °C. At these higher temperatures, the particles grew up with time. The SEM-EDS examination by line-profile analysis on these larger particles indicated they were based on chromium carbides. Other authors reported the precipitation of chromium nitrides and carbides in DSS 2304 aged at 750 and 850 °C for about 45 min [13].

Fig. 1 - BSE-SEM micrographs of DSS 2304 cross sections as-received (a) and aged 60 min at 850 °C (b).

La Metallurgia Italiana - n. 3 2018

Corrosione The stress - strain curves obtained with SSRT in air at 25°C on as-received and heat treated DSS 2304 (here not reported) evidenced that all aging treatments at 650 °C and the short ones at 750 and 850 °C did not modify the mechanical behaviour of the material. Instead, the long treatments at 750 and 850 °C determined a moderate increase in steel ductility (of about 10 %). This slight modification of the me-

chanical behaviour was induced by the subtraction of interstitial carbon atoms from solid solution, after precipitation of relatively large chromium carbide particles at the grain boundaries, after 60 min aging at the two higher investigated temperatures [6].

Fig. 2 - Stress – strain (a) and ECOR - strain (b) curves obtained with SSRT on DSS 2304, both as-received and after aging for 10 min within 650-850 °C range, in NACE TM-0177 in the presence of 10−3 M S2O32−. Fig. 2-a and 2-b collect the stress - strain and E COR – strain curves, obtained during SSRT performed in standard NACE TM-0177 solution in the presence of 10-3 M S2O32-, on both as-received and 10 min aged samples (T= 650, 750 and 850 °C). They show that εf % of as-received sample was about 51 % (Fig. 2-a), almost equivalent to that obtained in air (52 %), while the heat treatments induced much smaller εf %. In fact, decreasing values of about 42, 22 and 21 % were obtained at progressively increasing aging temperature. During SSRT, ECOR values (Fig. 2-b) of the as-received sample remained around -0.2 V SCE, that is in the passive potential range. The ECOR values of the sample aged 10 min at 650 °C had initially the same values recorded for the as-received

sample, then, after a strain of about 15 %, slowly decreased reaching values of about -0.55 VSCE, that are typical of active corrosion conditions, at the end of the test. Instead, the ECOR values of DSS 2304 aged for 10 min at 750 and 850 °C rapidly decreased already during elastic deformation and, after a plastic strain of only 5 %, reached about -0.55 VSCE. At these quite negative ECOR values the S2O32- conversion to reduced S-containing species, such as H2S, is highly favored [14]. Fig. 3-a and 3-b collect the stress - strain and E COR – strain curves obtained during SSRT performed in NACE solution containing 10 -3 M S2O32- on both as-received and 60 min aged samples (T = 650, 750 and 850 °C).

Fig. 3 - Stress – strain (a) and ECOR - strain (b) curves obtained with SSRT on DSS 2304, both as-received and after aging for 60 min within 650-850 °C range, in NACE TM-0177 in the presence of 10−3 M S2O32−.

La Metallurgia Italiana - n. 3 2018

Corrosion Also in this case, the ductility of heat treated samples was much lower than that afforded by the as-received sample, but εf % increased by increasing heat treatment temperature. In fact, values of 18 %, 25 % and 39 % were obtained after long aging treatments at 650, 750 and 850 °C, respectively. Thus, 60 min aging at 650 °C resulted the most SCC susceptible heat treatment in S 2O 32- solution. This result is also supported by ECOR values (Fig. 3-b), which shifted very rapidly to -0.55 VSCE in the case of the sample aged 60 min at 650 °C, while the decay was slower in the case of samples heat treated at 750 and particularly at 850 °C, for the same time.

At the end of SSRT, all samples aged for 10 and 60 min between 650 and 850 °C showed a brittle type fracture and badly corroded surfaces with many secondary cracks as those observable in the macrograph of Fig. 4-a. In contrast, as-received samples evidenced a ductile type fracture with necking both in air and in NACE solution with S2O32-. The micrographs of the longitudinal sections of samples failed by SCC evidenced that in all cases, small pits were present (Fig. 4-b), clearly operating as crack initiators [15]. Then the cracks propagated in the ferrite phase and/or following ferrite/austenite grain boundaries (Figure 4b).

Fig. 4 - Macrograph (a) and micrograph (b) of DSS 2304 aged for 10 min at 850 °C after SSRT in NACE solution containing 10−3 M S2O32− (in b) long transverse section, parallel to load direction). Tab. 2 collects R values obtained from SSRT results, for both as-received and heat treated DSS 2304. The influence of heat treatment on SCC susceptibility in NACE solution containing S2O32− is more or less the same after short aging at 750 and 850 °C (R = 0.42 and 0.41, respectively), but it is

significantly less marked after 10 min at 650 °C (R = 0.83). By extending the aging time, an increase in SCC susceptibility was observed only at 650 °C (R = 0.36), while the susceptibility decreased at both 750 (R = 0.45) and particularly at 850 °C (R = 0.69).

Tab. 2 – R values obtained from SSRT in NACE TM-0177 in the presence of S2O32− at 10-3 M on DSS 2304, both as-received and heat treated for 10 and 60 min between 650 and 850 °C.

Heat treatment conditions

Asreceived

0.98

650 °C

850 °C

10 min

60 min

10 min

60 min

10 min

60 min

0.83

0.36

0.42

0.45

0.41

0.69

DISCUSSION SCC susceptibility results are quite in agreement with previous findings concerning the measurements of CPT values and DOS to IGC, obtained on DSS 2304 samples under the same conditions here investigated [2]. The high SCC susceptibility and the low resistance to pitting and IGC induced by heat treatments with specific time-temperature combinations (60 min at 650 °C, 10 and 60 min at 750 °C and 10 min at 850 °C) are likely related to the formation of dechromized areas in the proximity of chromium carbide particles, La Metallurgia Italiana - n. 3 2018

750 °C

precipitated during heat treatments. In these regions, pits tend to form, acting as crack initiators. Afterwards, cracks propagation occurs preferably in ferrite grains and along ferrite/austenite grain boundaries [16]. The most severe heat treatment consists in 60 min aging at 650 °C. It significantly reduces the SCC resistance of DSS, likely due to the achievement of very low chromium levels in narrow areas. As a consequence, easy surface activation occurs, as proved by the quite negative ECOR values (-0.55 VSCE) reached very quickly during the elastic deformation step 43

Corrosione in the tensile test (Fig. 3-b). At these low potential values, S2O32- reduction to H2S is possible and easily occurs, suggesting that a contribution of hydrogen penetration stimulated by sulphide species to SCC cannot be excluded [7,8]. During the 60 min heat treatment at 850 °C, there is time enough for chromium replenishment in depleted regions, so reducing the alloy SCC susceptibility. Under this aging condition, the sample undergoes a slower activation.

ge between 650 and 850 °C increased SCC susceptibility of DSS 2304 in NACE TM-0177 solution containing 10-3 M S 2O32-. The SCC susceptibility was likely connected to chromium depletion in the vicinity of chromium carbide precipitates, whose extent depended on aging times and temperatures. An aging of 60 min at 650 °C significantly reduced SCC resistance, while a recovery was observed by increasing heat treatment temperature.

CONCLUSIONS Heat treatments of 10 and 60 min in the temperature ranREFERENCE [1] F. Zanotto, V. Grassi, M. Merlin, A. Balbo, F. Zucchi, Corros.Sci., 2014; 94:38-47. [2] F. Zucchi, V. Grassi, M. Merlin, F. Zanotto, Atti del Convegno Giornate nazionali sulla Corrosione e Protezione – Associazione Italiana di Metallurgia, 2013 July 10-12, Naples, Italy, Milan AIM 2013, p. 1-11. [3] F. Zanotto, F. Zucchi, V. Grassi, M. Merlin, A. Balbo, Atti del Convegno Giornate nazionali sulla Corrosione e Protezione – Associazione Italiana di Metallurgia, 2015 June 15-17, Ferrara, Italy, Milan AIM 2015, p. 1-11. [4] F. Zanotto, V. Grassi, A. Balbo, C. Monticelli, F. Zucchi, Corros. Sci., 2014; 80: 205-212. [5] F. Zanotto, V. Grassi, F. Zucchi, M. Merlin, A. Balbo, La Metallurgia Italiana 2016, 12-108:23-33. [6] F. Zanotto, V. Grassi, A. Balbo, C. Monticelli, C. Melandri, F. Zucchi, Corros. Sci., 2018; 130:22-30. [7] S. Tsujikawa, A. Miyasaka, M. Ueda, S. Ando, T. Shibata, T. Haruna, M. Katahira, Y. Yamane, T. Aoki, T. Yamada, Corrosion, 1993; 49:409–419. [8] L. Choudhary, D. D. Macdonald, A. Alfantazi, Corrosion, 2015; 71:1147-1168. [9] E. Johansson, R. Pettersson, Proceeding of European Corrosion Congress 2010 - EUROCORR 2010, EFC, 2010 Sept. 13–17, Moscow, p. 2869–2878. [10] NACE standard TM-0177-90 Standard Test Method Laboratory Testing of Metals for Resistance to Sulfide Stress Cracking in H2S Environments, National Association of Corrosion Engineers (NACE), Houston 1990. [11] M. Barteri, N. De Cristofaro, L. Scoppio, G. Cumino, G. Della Pina, Proc. Corrosion ’95, NACE, Houston 1995, paper 76. [12] B. Deng, Y. Jiang, J. Xu, T. Sun, J. Gao, L. Zhang, W. Zhang, J. Li, Corros. Sci.; 2010, 52:969:977. [13] G. Straffelini, S. Baldo, I. Calliari, and E. Ramous., Metallurg. and Mat. Trans. A Vol. 40A (2009) 2616-2621. [14] P.R. Rhodes, G.A. Welch, L. Abrego, J. Mater. Energy Syst. 1983; 5:3–18. [15] P. Marcus, J. Oudar, Corrosion Mechanism in theory and practice, Marcel Dekker, New York, 1995, p.345. [16] T. Bellezze, G. Giuliani, A. Viceré, G. Roventi, Corros. Sci. 2018; 130:12-21.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing Microstructural characterization and corrosion behaviour of SLM CoCrMo alloy in simulated body fluid M. Seyedi, F. Zanotto, E. Liverani, A. Fortunato, C. Monticelli, A. Balbo CoCrMo base alloys are widely used for the production of orthopaedic implants because of their excellent mechanical properties and high corrosion resistance in biological fluids. However, despite the advances in this sector, untimely implant failures and dissatisfaction are still observed in patients. The fabrication of customized implants by Selective Laser Melting should permit to overcome some common problems, like mismatch between the joint prosthesis and the natural joint conformation, non-physiological load transfer and scarce osteointegration. However, some parameters related to the process optimization in terms of final density and mechanical properties are critical and the evaluation of their influence on corrosion resistance of the components is reputed fundamental. In this work SLM technique was used to produce two types of samples based on ASTM F1527 CoCrMo alloy by using different process parameters. The corrosion behaviour of the two materials was investigated by recording the EIS spectra and the polarization curves during 15 days of immersion in Phosphate-Buffered Saline (PBS) solutions at pH 7.4. The samples showed a fine microstructure, characterized by Mo enrichment at the cell boundaries. Both types of samples showed low corrosion rates in the studied environments because of the formation of a very protective oxide film on the sample surfaces with high resistance to localized corrosion, but the finer microstructure ensured slightly higher corrosion resistance.

KEYWORDS: BIOMATERIALS, SELECTIVE LASER MELTING, CoCrMo ALLOYS, CORROSION, SIMULATED BODY FLUIDS

INTRODUCTION Traditional endoprostheses are widely used to replace joint surfaces damaged by severe trauma or degenerative articular disease. Typical metallic biomaterials used to fabricate implant devices include austenitic stainless steels, Co-Cr and Ti alloys. In particular, the use of high-strength Co-Cr based alloys for improving the durability of such devices has attracted much attention in the past years because of their excellent properties, including high mechanical strength, good biocompatibility, high wear and corrosion resistance [1-3]. Although these materials have been used successfully for over 5 decades, in the case of some specific joint types frequent failures and patients’ dissatisfaction have been reported [4]. The reasons for these failures were related to a number of factors, such as: limited availability of prosthesis size and shape, difference in mechanical properties between bone and implant materials, unsatisfactory osteointegration and foreign body reaction induced by inflammatory cells [4-6]. Selective laser melting (SLM) is a very promising technique that can overcome some of these issues. In fact, this technique allows obtaining completely customizable and scalable prostheses with complex shapes, through the consolidation of metal powders layer by layer. It is, therefore, possible to control the microstructure characteristics of the product and to manufacture dense components, porous graded or fully customized architectures, improving osteLa Metallurgia Italiana - n. 3 2018

ointegration and bone adaptation [7-9]. Biofunctionality and biocompatibility are fundamentals factors for the selection of biomaterial alloys and the corrosion resistance of metal implants in the body fluids is closely related to their biocompatibility [1]. Recently, the potential of using SLM materials was investigated in vitro for dental applications [9,10].The materials fabricated with SLM technology seem to be promising in terms of corrosion resistance, but their behavior in biological environments has not been studied

Mahla Seyedi, Federica Zanotto, Cecilia Monticelli, Andrea Balbo Corrosion and Metallurgy Study Centre “A. Daccò”, University of Ferrara, Department of Engineering, via Saragat 4a, Ferrara, Italy

Erica Liverani, Alessandro Fortunato University of Bologna, Department of Industrial Engineering, viale Risorgimento 2, Bologna, Italy

Additive manufacturing in depth. In a previous work [11], a proper selection of different laser process parameters, bulk and surface fabrication was optimized in terms of mechanical properties. In this work, the same strategy was adopted to fabricate two different types of samples and the corrosion behaviour of the CoCrMo alloy was investigated in solutions simulating biological fluids.

sphate buffered saline (PBS) solution at pH=7.4, whose composition is shown in tab. 3. All samples were exposed to the solutions that simulated body fluids (SBF) for 15 days at the temperature of 37±1°C. Electrochemical impedance spectroscopy (EIS) was used to monitor the evolution of the corrosion behaviour during the immersion period. The spectra were collected at the corrosion potential (Ecor) by imposing a 10 mVrms amplitude excitation voltage in the frequency range 104 –10−3 Hz, and by taking ﬁve measurements per decade. From each spectrum the polarization resistance (Rp) value was estimated as the limit of the real part of the impedance at frequency tending to 0. At the end of tests the polarizations curves (PC) were recorded at a scan rate of 0.1 mV s-1. Preliminary cyclic voltammetry (CV) tests were also conducted, in order to better investigate the electrochemical behaviour of the two different sample types. Each CV test consisted in 5 cycles recorded at a sweep rate of 120 mV s-1 in the potential ranges (-1.5 ÷ +0.7) V SCE and (-0.25 ÷ +0.7) V SCE. Before each test, the samples were polarized at -1.5 VSCE for 80 s. The electrochemical tests were performed with a PAR 2263 potentiostat/FRA/ galvanostat (EIS), and with a PAR 273A potentiostat/galvanostat (PC and CV).

EXPERIMENTAL The studied materials were prepared with SLM technique starting from a commercial powder of CoCrMo alloy (LPW Technology Ltd, Runcorn Cheshire, UK) whose composition meets the standard ASTM F1537. Tab. 1 reports the composition of the raw powder. The samples were fabricated with a SISMA MYSINT100 system (Sisma, Piovene Roccheta, Italy) by using two different sets of parameters, as indicated in tab. 2. The hatch distance (space between two adjacent laser tracks) was maintained constant to the value of 0.06 mm and a chessboard laser scan strategy was adopted. All materials were characterized by a scanning electron microscope (SEM, Zeiss MA15/LaB6) coupled with an energy dispersive spectroscopy (EDS) system (X-Act/INCA, Oxford Instruments) and by X ray diffraction technique (Bruker D8 advances Billerica USA). The biological environment was simulated by using a pho-

Tab. 1 - EDS analysis of CoCrMo powders chemical composition (wt%)

≤ 0.04

0.77±0.05

28.06±0.33

64.23±0.16

0.68±0.11 0.12±0.01

5.92±0.11 0.07± 0.03

Tab. 2 - Process parameters used and obtained relative densities.

Test

Power (W)

Scan speed (mm/s)

Fluence (J/mm3)

Relative density (%)

150

900

138.8

99.8

1200

53.6

Tab. 3 - Composition, electrical conductivity and pH of simulated body fluid (SBF) solution

Na2HPO4 (g/L)

KH2PO4 (g/L)

KCl (g/L)

NaCl (g/L)

Total Cl- (M)

Λ(μS/cm)

1.42

0.245

0.2

8.8

0.15

17.15

7.4

La Metallurgia Italiana - n. 3 2018

Additive manufacturing RESULTS AND DISCUSSION The adopted laser parameters (tab. 2) induced different microstructural characteristics in the fabricated materials. High fluence processes promoted powder melting instead of sintering, so enabling the production of denser materials. In fact, the sample B1, obtained under high fluence conditions, showed a high relative density (99.85 %) and a low level of residual porosity, instead the B2 samples, obtained with low fluence process, were characterized by a lower relative density (95%) with respect to the B1 and by a microstructure with some large pores (up to 200Âľm). The microstructure of the produced samples is shown in fig 1. In both cases, the SEM observation highlighted a fine and elongated cellular microstructure formed within the melt pool. In fact, during the SLM process the alloys undergo a rapid heating, through the laser scan, followed by a very rapid cooling that produces an undercoo-

led melt. These high solidification rates and high level of non-equilibrium conditions induced the development of a microcellular structure and suppressed the formation of micro-sized carbides. Fig. 1 clearly shows a difference in the size of the microcells in the two samples: the sample B1 (fig. 1a) was characterized by coarser microcells with respect to the sample B2 (fig. 1b). This difference is probably due to the higher laser scanning speed and lower laser power (lower fluence) used for B2, which promoted a cell size refinement, due to a shorter persistence at high temperature during the layer build up. An elemental profile analysis carried out by SEM/EDS through these microstructural features on B1 sample, revealed clear Mo enrichment and Co depletion localized at the cell boundaries. Similar effects were observed on B2 sample.

(a) (b) Fig. 1 - SEM image of the microstructures of (a) B1 and (b) B2 samples after electrolytic etching.

Fig. 2 - XRD diffraction pattern of raw powder, B1and B2 specimens.

La Metallurgia Italiana - n. 3 2018

Additive manufacturing

Fig. 3 - A) Rp and Ecor values and B) polarization curves recorded during the exposure to SBF.

Fig. 4 - Cyclic voltammograms recorded on SLM samples in SBF at 120 mVs-1. A) B1; B) B2. Fig. 2 shows the XRD patterns obtained from the starting powder and from B1 and B2 samples, under spinning conditions. The XRD analysis showed that in all cases the metastable fcc phase (γ phase) is mainly retained at room temperature as the martensitic transformation from the metastable fcc to hcp phase (ε phase) is hindered during fast cooling below Tc (martensitic transformation critical temperature), such as during air cooling [12]. However, some amounts of ε phase were detected on both SLM samples (fig. 2, peak at 46.5°) since the transformation γ→ε partially occurred during each new layer build up, due to heating of the sample sublayers at temperatures below Tc. The data indicated a higher amount of this phase in the B2 samples with respect to both B1 and raw powder, suggesting that, during each new layer deposition in B2 samples, the lower laser fluence induced the achievement of lower temperatures in comparison to those present on B1, so permitting a higher degree of martensitic transformation [9]. The corrosion behaviour of the SLM alloys was studied by monitoring the Rp values of both sample types, during 48

15 days of exposure to the selected SBF. The Rp and Ecor values are shown in fig. 3a. After 1 hour immersion in PBS at pH= 7.4, on B1 sample Rp values of 1MΩ cm2 were detected, while B2 showed lower values, close to 0.7 MΩ cm2. For both B1 and B2 samples, Rp values increased rapidly during the first immersion period (1-3 days), with B2 samples always exhibiting higher Rp values, till the end of the immersion period. The corresponding Ecor values increased during the first days of exposure on both samples and then, in the case of B1 sample they settled at about -0.140 VSCE, until the end of exposure, while on B2 they continuously increased up to -0.080 VSCE. These results suggested that the air-formed oxide films on the alloys continuously improved their protectiveness particularly on B2. The polarization curves collected after 1 hour and at the end of exposure period are shown in fig. 3b. At short immersion time, both B1 and B2 showed a passive behaviour and low corrosion currents (icor) around 9x10-8 and 10-7 A/cm2 for B1 and B2 sample respectively. However, after 15 days of immersion, the anodic currents

La Metallurgia Italiana - n. 3 2018

Additive manufacturing significantly decreased, particularly on B2, inducing an ennobling of Ecor and a reduction of icor down to 2x10-8 A/ cm2 for B1 and 9x10-9 A/cm2 for B2 sample. These results confirmed that at the end of the exposure B2 exhibited a higher corrosion resistance in comparison to B1, as highlighted by EIS measurements. At around +0.57 VSCE a rapid increase in the anodic current densities was measured. At these noble potential values both water oxidation and transpassive dissolution may occur [13, 16], but no pitting attack was observed by accurate SEM observations. Fig. 4 collects the first and the fifth (the last) cycle of the CV tests carried out on B1 and B2 samples, at a sweep rate of 120 mV s-1, in the two investigated potential ranges. B1 and B2 samples exhibited similar voltammograms. In the potential range from -1.5 VSCE to 0.7 VSCE, the direct scan of the first cycle (solid blue line) showed initially very high cathodic currents which are likely the sum of multiple reduction reactions (water and oxygen reduction and also surface oxide reduction if oxides are still present after pre-conditioning at -1.5 VSCE). Then, an anodic peak at -0.5 VSCE is detected, related to electroformation of Co(II) hydroxide from Co metal [17] and a second peak is visible at -0.125 VSCE, likely related to further oxidation of Mo oxides [18]. In the range +0.6 ÷ +0.7 V SCE, compatible with both water oxidation and transpassive Cr dissolution, anodic currents increase significantly. During the reverse scan, the peak corresponding to Co(II) hydroxide reduction to Co metal is detected at about -1.1 VSCE [17]. In the subsequent cycles, an increase in anodic currents was observed in the potential range between the two previously quoted anodic peaks (-0.5 and -0.125 VSCE) and in the last cycle (solid red line) only a broad anodic peak centred at -0.25 VSCE was detected. In the CV test performed in the potential range from -0.25 VSCE to +0.7 VSCE, the peak at -0.125 VSCE was again detected in the first anodic cycle (blue dashed line), suggesting that this peak is not due to further oxidation of the species produced by oxidation at -0.5 VSCE. On cycling (last cycle: red dashed line), the anodic current decreased progressively, showing that if the potential range permitting the oxide

La Metallurgia Italiana - n. 3 2018

layer reduction is not reached, a more and more protective surface film forms. This also suggests that in the range +0.6 ÷ +0.7 VSCE negligible transpassivity occurs because cycling including these high anodic potentials does not impair the oxide film protectiveness. The results obtained showed that the process parameters adopted for the production of the samples have important effects on both the microstructural characteristics and the corrosion behavior. The B2 samples fabricated with the low-fluence process displayed some porosity and a refinement of the microstructure due to a shorter persistence at high temperature during the layer build up. As pointed out by both EIS measurements and PC tests, these B2 samples exhibited a better corrosion behaviour with respect to B1 after long immersion times, in neutral PBS solution. It is likely that in the presence of a finer microstructure, lower levels of Mo micro-segregation occur, so inducing a stronger and more uniform passivity. The porosity present on B2 does not affect its corrosion behaviour. Preliminary results obtained with CV tests, clearly stress the high stability of the surface passive film, which is maintained at high anodic potentials if the cathodic potential range allowing for oxide film reduction is avoided. CONCLUSIONS • The selected SLM parameters allow to obtain samples with different microstructural characteristics: B2 sample has a higher porosity than B1 but a finer cellular microstructure. • Samples obtained with SLM technique show low corrosion rates in the tested environment that decrease during the exposure. The porosity present in the B2 does not affect the corrosion behaviour; • The process parameters affect the corrosion behaviour: sample B2 has lower corrosion rates than sample B1 in all tested conditions; • Lower levels of Mo micro-segregation on the cells boundary in the samples B2 allow for a more uniform Mo distribution in the passive film making it more protective.

Additive manufacturing REFERENCE [1] Chen Q, Thouas GA, Mat. Sci. Eng. R. 87 (2015) 1–57. [2] Takayuki N et al., Advance in Metallic Biomaterials, Tissue, Materials and Biological Reaction. 1st ed. Springer-Verlag Berlin Heidelber; 2015. [3] Niinomi M, Sci. Technol. Adv. Mater., 4 (2003) 445-454. [4] Gougoulias N, Khanna A, Maffulli N, Clin. Orthop Relat R, 468 (2010) 199-208. [5] Gilbert JL, Sivan S, Liu Y, Kocagöz SB, Arnholt CM, Kurtz SM, J. Biomed Mat Res.,130A (2015) 211-217. [6] Anderson JM, A. Rodriguez A, Chag DT, Semin. Immunol.; 20, (2008) 86-100. [7] Xin XZ, Chen J, Xiang N, Wei B, Cell Biochem. Biophys., 67 (2013) 983-990. [8] Gu DD, Meiner W, Wissenbach K, Poprawe R. Inter. Mat Rev. a 57 (2012) 133-166. [9] Hedberg YS, Quian B, Schen Z, Virtanen S, Wallinder IO, Dental Materials; 30 (2014) 525-534. [10] Lu Y, Wu S, Gan Y, Li J, Zhao C, Zhuo D, Lin J, Mat. Sci. Eng: C, 49 ( 2015) 517-525. [11] Liverani E, Fortunato A, Leardini A, Belvedere C, Siegler S, Ceschini L, Ascari A, Mat&Des. 106 (2016) 60-68. [12] Lopez HF, Saldivar-Garcia AJ, Metall. Trans. A. 39 (2008) 8-18. [13] Bettini E, Eriksson T, Boström M, Leygraf C, Pan J, Electro. Acta 56 (2011) 9413-9419. [14] Bettini E, Boström M, Leygraf C, Pan J, Nature of Current Increase for a CoCrMo Alloy: “transpassive” Dissolution vs. Water Oxidation Int. J. Electrochem. Sci. 8 (2013) 11791-11804. [15] Vidal VC, Muñoz AL. Corros. Sci. 50 (2008) 1954-1961. [16] Hodgson AWE, Kurz S, Virtanen S, Fervel V, Olsson C-OA, Mischler S. Electro. Acta; 49 (2004) 2167-2178. [17] Metikos-Hukovic MR, Babic R, Corros. Sci 49 (207) 3570-3579. [18] Metikos-Hukovic MR, Pilic Z, R. Babic R, Omanovic D. Acta Biomater. 2 (206) 693-700.

La Metallurgia Italiana - n. 3 2018

Le manifestazioni AIM AIM meetings and events 2018 METALLURGIA DELLE POLVERI Scuola di - Centro MP Imola (BO) c/o SACMI, 19-20 aprile COME GARANTIRE LA CONFORMITÀ DELLE MACCHINE ANCHE A SEGUITO DI MODIFICHE: FORMA E SOSTANZA GdS - Centro AS Brescia, 17 maggio PROFILATI ESTRUSI DI ALLUMINIO: PER UN MERATO GLOBALE CHE VUOLE QUALITA', COMPETENZA E VALORE AGGIUNTO GdS - Centri ML e LPM Milano, 7 giugno AUMENTO DELLA PRODUTTIVITA’ DEGLI STAMPI ATTRAVERSO UN CONTROLLO SPECIFICO DELLA FATICA TERMICA GdS - Centro P Bergamo, 9-10 maggio METALLURGIA DELLE POLVERI Scuola di - Centro MP Maerne di Martellago (VE) c/o POMETON, 10-11 maggio METALLURGIA DI BASE PROPEDEUTICA AI TRATTAMENTI TERMICI Corso - Centro TTM Milano, 16-17-23 maggio ICS 2018 - 7TH INTERNATIONAL CONGRESS ON SCIENCE & TECHNOLOGY IN STEELMAKING Convegno Internazionale e 26° CONVEGNO NAZIONALE TRATTAMENTI TERMICI Convegno Venezia, 13-15 giugno PROBLEMATICHE DEI MATERIALI NEI CICLI COMBINATI TRADIZIONALI ED INNOVATIVI GdS - Centro ME Milano, 28 giugno RADDRIZZATURA E TENSIONI RESIDUE DEL GETTO GdS - Centro P Ceregnano (RO) c/o TMB, giugno

MATERIALI METALLICI E PROCESSI PRODUTTIVI INNOVATIVI PER L'AEROSPAZIO Convegno - Centri ML, MFM e MP Napoli, 19-20 luglio METALLURGY SUMMER SCHOOL - 2a edizione COMET Bertinoro (FC), luglio 37° CONVEGNO NAZIONALE AIM Convegno – SEGR Bologna, 12-14 settembre EOSC 2018 - 8TH EUROPEAN OXYGEN STEELMAKING CONFERENCE Convegno Internazionale Taranto, 10-12 ottobre LA PREVENZIONE E LA GESTIONE DELLE MALATTIE PROFESSIONALI GdS - Centro AS Brescia, 25 ottobre TRATTAMENTI TERMICI DEGLI ACCIAI PER STAMPI A CALDO E A FREDDO PER IL SETTORE AUTOMOTIVE GdS - Centro TTM ottobre GLI ACCIAI INOSSIDABILI - 10a EDIZIONE Corso - SEGR Milano, ottobre/novembre UTENSILI DIAMANTATI GdS - Centro MP Vicenza, 15 novembre LA PRODUZIONE DI GETTI PER APPLICAZIONI STRUTTURALI. ASPETTI METALLURGICI E DI PROCESSO GdS - Centro P Brescia, 16 novembre CLEAN TECH 4 - 4TH EUROPEAN CONFERENCE ON CLEAN TEHNOLOGIES IN THE STEEL INDUSTRY Convegno Internazionale Bergamo, 29-30 novembre RIVESTIMENTI - 1° modulo Rivestimenti PVD e CVD Corso modulare - Centro R Roma, novembre

Per ulteriori informazioni rivolgersi alla Segreteria AIM e-mail: info@aimnet.it oppure visitare il sito internet www.aimnet.it

La Metallurgia Italiana - n. 3 2018

Industry news Properties Of Ti-6Al-4V Components Produced By Digital Metal® Binder Jetting Technology edited by: M. Persson, R. Carlström, K. Gustavsson, S. Nilsson, C. Palmqvist, B. Brash Additive manufacturing is in addition to widespread manufacturing of functional prototypes increasingly deployed for manufacturing of components in Titanium alloys for a wide range of applications e.g. medical, dental, aerospace and industrial. Robust material properties and adhering to material standards are instrumental when moving into mass production of components. Digital Metal® technology, a two-step Additive Manufacturing process, was applied for manufacturing of components in Ti-6Al-4V. Layer wise precision binder jetting on powder bed is followed by a curing to develop component’s green strength. Subsequently non-bonded powder is removed and component sintered to develop density and strength. Process parameters are optimized for the alloy and physical properties examined in relation to material standard ISO 22068. Chemical composition and microstructure has been monitored as a function of process parameters. Mechanical properties investigated in the different directions relative the build direction and dimensional accuracy on components evaluated. KEYWORDS: ADDITIVE MANUFACTURING, BINDER JETTING, TI6AL4V, MIM, EQUIVALENT OXYGEN CONTENT, MECHANICAL PROPERTIES, SURFACE ROUGHNESS

Mats Persson, Ralf Carlström, Karl Gustavsson, Sofia Nilsson, Christer Palmqvist, Benjamin Brash Höganäs AB, S-26383 Höganäs, Sweden © European Powder Metallurgy Association (EPMA). First published in the Euro PM2017 Congress proceedings INTRODUCTION Digital Metal has been involved in additive manufacturing since 2010. The technology used is based on precision ink-jet on powder bed, followed by a separate sintering to obtain final strength of the component. The technique is successfully used to produce components in stainless steels e.g. 316L and 17-4PH. However basically all types of materials, available as powders with suitable morphology and particle size, can be applied to print green bodies. The subsequent sintering process is then optimized for each individual alloy. In this paper the resulting properties from printing Ti-6Al-4V are presented. Additive manufacturing has in recent years gone from being a rapid prototyping technology for plastic and metal components to being a manufacturing process considered for its flexibility, design freedom and weight saving. The principle 52

is that components are built directly, layer by layer, from 3D CAD data. Metal printing technology is dominated by selected melting processes however emerging technologies e.g. binder jetting is gaining interest and is today applied for prototyping as well as serial production of components. In the Digital Metal process components are built layer by layer derived from an original 3D or CAD file. The 3D file is prepared and sliced into 2D files corresponding to 42 µm thick layers of the component. A layer of 42 µm powder is applied in the build box, and the printer passes over the surface and deposits “ink” on relevant spots based on information from the 2D file. The bottom of the build box is lowered, and the printing process is repeated layer by layer until the component is formed in accordance with the original 3D file. The principle of the technique is presented in Figure 1. La Metallurgia Italiana - n. 3 2018

Attualità industriale As printing occurs at room temperature followed by a separate sintering, there is no heat involved during building, thus printing can be performed without protective gas. Since no melting takes place during building, green components can be produced with very high detail levels and tolerances. Furthermore there is no need of support structures during building of the components as surrounding powder bed is providing a stable foundation for the component. Since forming and heat treatment are separated, the process allows for a wide materials selection, where each process step can be optimized for each material. Virtually all materials that can be sintered, can also be utilized in the Digital Metal process, as long as the material is available as powder with adequate properties. Powder properties that affect the quality of the components are e.g. particle size, powder morphology, density, and flowability. To obtain a high resolution the powder particles has to be sufficiently small, to match the building resolution of 42 µm. There is need for sufficient tap density of the powder in order to get high green density. If tap density is not high enough, the green components will have high porosity levels. This leads to high shrinkage during sintering, difficulties to reach high final density, and difficulties to keep tolerances. Spherical particle morphology is preferred, but not completely necessary. Somewhat uneven particles might improve green strength, but might also be a disadvantage regarding uniform density. Powder properties will influence powder layers and

print results, and choice of improper powder might reduce tolerances, but when the process is performed correctly, the technology can offer tolerances down to 50 µm or better, and hole and wall thicknesses down to 200 µm at the present time. Once the green components are cleaned from excess powder they can be sintered. For each material, the heat treatment process has to be optimized for binder removal and densification. If printing and sintering is performed correctly, a component with high density will be possible to obtain. The final result is quite similar to what is obtained from metal injection molding. Ti6Al4V being an alloy with very high affinity to elements oxygen, carbon, nitrogen and hydrogen puts high demand on the thermal processes as it is virtually not possibly to reduce oxygen, and carbon that has been picked up by the component during sintering process Oxygen, Carbon and Nitrogen control is instrumental in order to develop desired mechanical properties. In this study an equivalent oxygen content is applied in order to describe tensile and yield strength. Surface finish may be improved by post processing after sintering, however internal surfaces like channels or holes can be impossible to treat, thus a good as built roughness is important for AM produced components. For stainless steels Digital Metal process yield an average of Ra 6 micron, which easily can be improved by traditional surface treatment processes to Ra 3 micron or better.

Fig. 1 - Overview of the Digital Metal printing process.

La Metallurgia Italiana - n. 3 2018

Industry news Experimental A test program was designed to investigate the repeatability of the process. Six prints were conducted in a Höganäs Digital Metal® P2000 printer. A pre-alloyed plasma atomized, sphe-

rical powder with a particle size below 45micron was used. Powder characteristics are summarized in table 1.

Tab. 1 - Powder characteristics. Lot 1 2

Al(%) 6,5 6,2

V(%) 4,0 3,7

C(%) 0,02 0,02

0(%) 0,13 0,13

Two powder lots were applied for the test program and samples were printed in 6 batches, and divided into four sinterings as outlines in table 2. Test cubes were printed for density, chemistry, resolution and feature tests, see figure 2. Two types of tensile test bars PM style, SS112123 Type N (series 1-6) and MIM style, ISO2740

N(%) 0,01 0,03

H(%) 0,004 0,003

TD(g/cm3) 2,9 3,0

D50(µm) 30 30

(series 7-8) were manufactured for mechanical testing and metallographic evaluation. Tensile test bar with reduced length were made for dimensional control and surface characterization.

Tab. 2 - Experimental set up. Series 1 2

Powder lot 1 1

Print 1 2

Sintering 1 2

3 4 5 6 7 8

1 1 2 2 2 2

3 4 5 6 5 6

3 4 1 3 1 3

All samples were thermally debinded and subsequently sintered under vacuum at 1350°C with a holding time of 240minutes in a VacuaTherm H800 SF. Same heat cycles were applied for sintering 1,2,3 and 4 respectively. Tensile testing was performed in a Zwick Z100, elongation recorded by a clip on extensometer. Surface roughness was measured on samples in the as sintered as well as blasted condition by a MarSurf M 300C. Metallographic evaluation were performed on the TS bars cut cross sectional in the grip section. Porosity analysis images were taken by Light Optical Microscopy (LOM) in a Zeiss Axio Imager A2m system in unetched condition on five images at x200. Samples were subsequen-

tly etched with Kroll’s reagent (100 ml DI-H2O, 3 ml HNO3, 1.5 ml HF) before further LOM investigation. Results Density and chemistry were evaluated on the test cubes and by Archimedes principle. Theoretical full density of the alloy is 4,43g/cm3.

La Metallurgia Italiana - n. 3 2018

Attualità industriale Tab. 2 - Chemical analysis and density measurements on sintered Series 1 2 3 4 5 6 7 8

C (%) 0,075 0,069 0,067 0,051 0,074 0,051 0,074 0,051

N (%) 0,020 0,030 0,016 0,019 0,014 0,019 0,014 0,019

O (%) 0,279 0,291 0,240 0,271 0,280 0,271 0,280 0,271

S (%) 0,002 0,002 0,002 0,002 0,002 0,002 0,002 0,002

Density (g/cm3) 4,16 4,17 4,16 4,17 4,16 4,19 4,16 4,19

Fig. 2 - Test cube for density and chemical analysis. Printed dimensions 11x11x7mm, as sintered 9,5x9,5x6mm Mechanical properties The results from tensile testing are summarised below in table 3. Strength and elongation data is an average of seven

specimens. Series 7 and 8 are ISO 2740 test specimen processed with identical processing as series 5 and 6.

Tab. 3 - Mechanical data from the different production. Note series 7 and 8 are ISO2740 specimens. Series 1 2 3 4 5 6 7 8

Yield strength (MPa) 792 795 789 794 804 791 812 771

Surface roughness Surface roughness was evaluated on a selection of samples and result given in Table 4 below. The top surface of short tensile test bars was measured perpendicular to the long direction. Surface roughness of top surface is Ra 6,5µm in

La Metallurgia Italiana - n. 3 2018

Tensile strength (MPa) 888 888 898 887 896 876 915 877

Elongation (%) 7,9 7,1 7,3 7,3 8,4 8,2 9,0 9,3

the as sintered state and improved to Ra 4µm after blasting operation. An average of all surfaces yields Ra 7,5 and 5,5µm respectively in the as sintered and blasted condition.

Industry news Tab. 4 - Surface roughness measurements As sintered

As blasted

Series

Ra(µm)

Rz(µm)

Ra(µm)

Rz(µm)

1 2 3 4

6,1 5,9 6,7 7,0

37 35 40 42

4,0 4,7 3,6 4,0

22 28 19 22

Metallographic evaluation LOM analysis of unetched samples shows a distributed porosity. Pore fraction measurements resulted in a porosity level of 4,9% which confirms density measurements. Etched samples show an elongated lamellar primary α structure. These lamellas are defined within regions/grains that often are surrounded by coarse α platelets. This is best observed when the

lamellas are aligned in the same plane as the image. When the direction is perpendicular to the image plane the appearance may be more “cloudy-shaped” where the β structure forms a network of half-curves that conceals the alpha structure, as marked in Fig.4b.

Fig. 3 - LOM images of unetched cross section

Fig. 4 - LOM images of etched cross section 56

La Metallurgia Italiana - n. 3 2018

Attualità industriale Discussion All prints show a density level in the range of 94-95%. Impurity levels are well within the normative values in ISO22068:2014 (O<0,4% , C<0,2% and N<0,1%) Performance of titanium alloys is very sensitive to impurity levels. Once O, C, N are dissolve in the titanium matrix it is virtually impossible to remove them during sintering. It is thus of importance to limit the pick-up of these elements in the prior process steps. Printing takes place at ambient temperature thus there is no increase of bonded impurities in this step. The following thermal debinding step is performed as to limit the pick-up. It is however in this step as the final carbon and nitrogen levels are established and the majority of oxygen pick up occurs. The influence of impurities on MIM produced Ti6Al4V material has been investigated [3] and an Oxygen Equivalent

applied to describe the influence of impurities on mechanical properties. Oeq. = [O] + 2[N] + 2/3[C]

[4]

The experimental data from this study plotted against the calculated oxygen equivalent show good agreement compared with the relation found for YS and UTS [3] as seen in figure 5. The equivalent oxygen level is well below 0,45% where the negative impact on elongation is starting to be seen. This is confirmed by the stable elongation values in Table 2. Earlier work [1] has shown that printed components can be successfully densified by HIP. This is an viable route in cases when further improved properties are needed.

Fig. 5 - Strength test data versus Equivalent oxygen content. Dotted line is data on MIM processed materials. [3] Conclusions Ti6Al4V components have successfully been manufactured repeatedly by the Digital Metal binder jetting process. Resulting material properties are in line with properties reported for MIM produced components. ISO22068:2014 standard is attained. Mechanical properties show a dependence on impurity level. A good fit is found when plotting data versus equivalent oxygen content, which also coincide with earlier findings for MIM processed Ti6Al4V. The two test geometries ISO2740 and SS112123 Type N yields comparative mechanical data. Surface roughness is on average Ra 7,5µm and can be easily be improved to Ra 5,5µm by blasting operation.

La Metallurgia Italiana - n. 3 2018

REFERENCES [1]

R Frykholm, B-G Andersson, R. Carlström “Digital Metal® – A technology for production of small precision components at high productivity with different type of metals” [2] R. M. German “Review – Progress in Titanium Metal Powder Injection Molding”, ISSN1996-1944 [3] T Ebel, O Milagres Ferri, W Limberg and F-P Schimansky “Metal Injection Molding of Advances Titanium Alloys” Advances in Powder Metallurgy Particulate Materials 2011 [4] H. Conrad “Effect of interstitial solutes on the strength and ductility of titanium” Progress in Materials Science, 1981, ISSN 00796425 57

Industry news La Stampa 3D a metallo è arrivata nelle applicazioni oleodinamiche a cura di: AIDRO hydraulics Un numero crescente di aziende sceglie la tecnologia di Additive Manufacturing (AM) non solo per la prototipazione rapida ma anche per produzione di oggetti funzionali. Questo è il caso dell’azienda della provincia di Varese, Aidro Hydraulics, che ha introdotto la stampa 3D a metallo affiancando la produzione tradizionale di componenti oleodinamici. Sfruttando la lunga esperienza tecnica, Aidro ha iniziato a produrre soluzioni oleo-idrauliche stampate in 3D a metallo in alternativa ai processi convenzionali (asportazione di truciolo, forgiatura, casting, ecc…) al fine di realizzare oggetti per applicazioni speciali. Tipicamente la produzione di componenti idraulici parte da un pezzo di metallo, in barra o da fusione, e successivamente la parte viene lavorata in CNC per raggiungere la forma desiderata. Invece, la produzione additiva inizia con materiale metallico in polvere che viene fuso in strati ultra sottili e sequenziali, utilizzando un laser ad alta potenza, e strato dopo strato, si costruisce l’oggetto tridimensionale. Aidro impiega la tecnologia additive manufacturing definita “fusione a letto di polvere di metallo” (DMLS o SLM), per creare prodotti idraulici stampati in 3D. Questa nuova tecnologia offre molti vantaggi, come: - maggiore libertà nel design che non deve più sottostare alle limitazioni dei metodi di produzione tradizionali, - l'elevato grado di personalizzazione e soddisfazione delle esigenze del cliente, - la possibilità di realizzare geometrie complesse senza costi aggiuntivi, - forme più compatte e con riduzione di peso, - la possibilità di integrare più funzioni e sensori, - tempi di produzione brevi. Aidro sta sfruttando queste potenzialità per realizzare sistemi oleodinamici innovativi e personalizzati che raggiungono performance migliori o forme più convenienti, che con la produzione idraulica tradizionale non sarebbero realizzabili. Per arrivare a realizzare un oggetto in AM, la prima fase è la progettazione che per il componente oleodinamico viene 58

completamente rivoluzionata: si parte infatti dalla funzione idraulica che si vuole ottenere, poi si definiscono i punti di riferimento, quali attacchi e connessioni del componente idraulico all’interno del sistema completo e poi si tracciano i collegamenti a piacere e dove è più conveniente per l’ingegneria. Una volta definito il nuovo design dell’oggetto, il progetto viene rielaborato con software dedicati al AM per essere poi realizzato con la stampante a metallo, nel giro di qualche ora. La tecnologia additiva ha vari ambiti di impiego, e nel campo dell’oleodinamica, questi sono principalmente la prototipazione veloce e la produzione di serie di prodotti funzionali. Nel caso del rapid prototyping, la stampa 3D consente ai progetti di svilupparsi più rapidamente e un prodotto può passare dalla fase di progettazione al lancio sul mercato in minor tempo. Oltre al tempo, un altro vantaggio legato alla prototipazione del settore idraulico riguarda anche la riduzione dei costi: infatti il costo di un prototipo stampato in 3D può essere molto inferiore rispetto a quello della prototipazione tramite fusione, che richiede stampi dedicati, o delle lavorazioni tradizionali con CNC, che per un numero limitato di pezzi per la prototipazione sono molto costose. Aidro ritiene che, sfruttando questi vantaggi, molti progetti rimasti chiusi in un cassetto a causa dei limiti della prototipazione tradizionale, avranno ora maggiori possibilità di essere realizzati. Per quanto riguarda la produzione vera e propria, con le attuali tecnologie additive si può pensare di produrre vere e proprie serie di oggetti idraulici funzionali. Alberto Tacconelli, Managing Director di Aidro, sostiene infatti che "tramite la produzione additiva si possono progettare e realizzare blocchi oleodinamici con migliori prestazioni, grazie alla capacità di costruire canali interni curvi. Infatti, evitando le lavorazioni meccaniche in CNC che creano intersezioni ad angolo retto, i canali interni del blocco valvole stampato in 3D sono ottimizzati per un maggiore flusso all'interno di uno spazio più piccolo. Inoltre, le potenziali perdite di carico vengono eliminate perché non sono più necessari i tappi esterni per chiudere i fori ausiliari". La Metallurgia Italiana - n. 3 2018

Attualità industriale Aidro ha sviluppato vari progetti di componenti oleodinamici stampati in 3D che vengono tutti validati tramite test funzionali e analisi sulle proprietà meccaniche. I test effettuati da Aidro hanno dimostrato che i prodotti stampati in 3D sono assolutamente paragonabili ai prodotti fabbricati in maniera tradizionale sia nella resistenza alla pressione, sia nella porosità e densità. Inoltre, le proprietà meccaniche sono assimilabili a quelle del metallo da barra. Infatti, il processo "a fusione di letto di polvere" (DMLS) riscalda la polvere di metallo al punto che le particelle si fondono insieme a livello molecolare. La porosità del materiale

La Metallurgia Italiana - n. 3 2018

sinterizzato è controllabile e si può affermare che la densità è del 99,9%. Aidro utilizza principalmente leghe di alluminio (AlSi10Mg) e acciaio inox (AISI316L). Aidro svolge test sulle proprietà meccaniche dei prodotti oleodinamici di produzione additiva in collaborazione con il Dipartimento di Meccanica del Politecnico di Milano. Queste analisi hanno confermato che la porosità è trascurabile in quanto nell’ordine di grandezza di qualche micron e che tutte le altre proprietà meccaniche dei materiali stampati sono di ottimo livello (vedasi la presentazione sottostante).

Industry news Blocco di sinistra è ottenuto con processi convenzionali (cnc). Blocco di destra è stampato in 3D e dimostra come un oggetto tradizionale possa essere virtualmente reinventato usando un approccio alla progettazione diverso ed innovativo: le valvole sono installate dove è necessario e poi vengono collegate con canali dalle forme libere. Inoltre, i canali interni del blocco sono ottimizzati per migliorare il flusso e risparmiare spazio, mentre il rischio di perdite viene eliminato, in quanto non sono più necessarie le perforazioni ausiliarie e i tappi.

Corpo valvole bancabili per la riduzione della pressione. Il corpo di questa valvola è stato ridisegnato per essere stampato in 3D al fine di avere un oggetto più leggero. Infatti, l’additive manufacturing consente un'alta riduzione del materiale utilizzato per la produzione. Nell’esempio di destra, il corpo valvola mantiene gli attacchi esterni come nel prodotto tradizionale, ma è stato ridisegnato e alleggerito per cui il risparmio di peso è del 60%.

Cursori idraulici che Aidro ha riprogettato con fori dalle forme nuove, cioè fori quadrati e fori ovali. I cursori tradizionali sono generalmente circolari perché lavorati con frese rotanti. Le forme quadrate o irregolari non sono realizzabili con i metodi tradizionali, in particolare con le lavorazioni in CNC, mentre la produzione additiva permette di realizzare forme geometricamente complesse. L’idea di Aidro è di sfruttare le nuove forme dei fori per aumentare l’area di passaggio dell’olio all’interno del cursore e quindi ridurre i cali di pressioni. Inoltre, sfruttando la possibilità di creare forme complesse, il cursore di Aidro consolida più componenti in un’unico pezzo stampato 3D. Questo semplifica il processo di assemblaggio e i tempi.

Infine, bisogna tenere conto del fatto che questa nuova tecnologia, essendo ancora nuova ed in fase di sviluppo, ha alcuni limiti legati alle dimensioni degli oggetti stampabili e ai grandi volumi. I produttori di macchine per la stampa 3D

stanno lavorando per sviluppare impianti ad alta produttività per cui si può immaginare che tra qualche anno, gli attuali limiti verranno superati.

La Metallurgia Italiana - n. 3 2018

Attualità industriale

Blocco oleodinamico per il controllo del cilindro ad azione singola, realizzato in acciaio inox (AISI316L). Questo manifold stampato in 3D è stato riprogettato per ridurre l’ingombro e alleggerire il peso.

Blocco oleodinamico doppio stampato in 3d con AlSi10Mg.

Scambiatore di calore acqua-olio stampato in 3d con AlSi10Mg, in un pezzo unico, non richiede assemblaggio di parti come negli scambiatori tradizionali. Aidro sarà presente a EuroPM 2018, per presentare un progetto di blocco oleodinamico stampato in 3D in lega di alluminio (14 - 18 October 2018 - Bilbao, Spain).

La Metallurgia Italiana - n. 3 2018

Aim news Calendario degli eventi internazionali International events calendar

QUOTE SOCIALI AIM 2018 (ANNO SOLARE) Benemeriti (quota minima) 1.750,00 €

April 12-13, Friedrichshafen, Germany European Conference on Heat Treatment - Nitriding and Nitrocarburizing April 30 - May 3, Houston, USA Offshore Technology Conference (OTC) 2018 May 19-26, Perth, Australia ALTA 2018 Nickel-Cobalt-Copper, Uranium-REE-Li and Gold-PM Conference & Exhibition

Sostenitori (quota minima)

750,00 €

Ordinari (solo persona)

70,00 €

Seniores

25,00 €

Juniores

15,00 €

La quota dà diritto di ricevere la rivista dell’Associazione, La

May 23-25, Brno, Czech Republic 27th International Conference on Metallurgy and Materials, METAL 2018

in formato digitale). Ai Soci viene

May 30-31, Haifa, Israel Technological Innovation in Metals Engineering (TIME 2018)

partecipazione alle manifestazioni

Metallurgia Italiana (distribuita riservato un prezzo speciale per la AIM e per l’acquisto delle

June 3-6, Pittsburgh, USA 2018 Superalloy 718 & Derivates: Energy, Aerospace, and Industrial Applications

pubblicazioni edite da AIM.

June 5-7, Erfurt, Germany International Trade Show + Conference for Additive Manufacturing

Per ulteriori informazioni, iscrizioni, rinnovi:

June 5-7, Greenville, USA 4th International Conference on HTSE in Automotive Applications

AIM, Via F. Turati 8

June 10-13, Helsingør, Denmark 4th International Congress on 3D Materials Science (3DMS 2018)

Tel.: 02 76021132/76397770,

June 18-21, Gaithersburg, USA Additive Manufaturing Benchmarks 2018 (AM-Bench 2018)

e-mail: amm.aim@aimnet.it

20121 Milano fax: 02 76020551 www.aimnet.it

July 8-13, Paris, France THERMEC July 15-21, Paris, France International Conference on Composites or Nano Engineering (ICCE-26) Semptember 5-8, Las Vegas, USA MEI2018 (Mining Expo International) September 9-13, Oxford, United Kingdom Eurosuperalloys 2018 September 12-14, Xi’ An, China 25TH IFHTSE September 13-14, Aachen, Germany Metallurgie im Wandel 4.0 October 14-18, Seattle, USA Furnace Tapping 2018 Conference October 16-19, Stockholm, Sweden 3rd Ingot Casting, Rolling and Forging Conference, ICRF 2018 October 30-31 Shangai, China BAC 2018, The 7th Baosteel Biennial Academic Conference November 21-23 Wien, Austria 3rd MAMC2018, Metal Additive Manufacturing Conference 62

La Metallurgia Italiana - n. 3 2018

Aim news AIM – UNSIDER Norme pubblicate e progetti in inchiesta (aggiornamento 28 febbraio 2018) NORME UNSIDER PUBBLICATE DA UNI NEL MESE DI FEBBRAIO 2018 UNI EN ISO 19901-2:2018 Industrie del petrolio e del gas naturale - Requisiti specifici per le strutture in mare - Parte 2: Procedure e criteri di progettazione in zona sismica UNI CEN ISO/TS 17969:2018 Industrie del petrolio e del gas naturale - Linee guida sulla competenza del personale nelle operazioni di gestione dei pozzi NORME UNSIDER RITIRATE DA UNI NEL MESE DI FEBBRAIO 2018 UNI EN ISO 19901-2:2005 Industrie del petrolio e del gas naturale - Requisiti specifici per le strutture in mare - Parte 2: Procedure e criteri di progettazione in zona sismica

ses -- Technical delivery conditions -Part 4: Nickel-alloy steels with specified low temperature properties ISO 9328-5:2018 Steel flat products for pressure purposes -- Technical delivery conditions -- Part 5: Weldable fine grain steels, thermomechanically rolled ISO 9328-6:2018 Steel flat products for pressure purposes -- Technical delivery conditions -- Part 6: Weldable fine grain steels, quenched and tempered ISO 9328-7:2018 Steel flat products for pressure purposes -- Technical delivery conditions -Part 7: Stainless steels

UNI CEN ISO/TS 17969:2015 Industrie del petrolio e del gas naturale - Linee guida sulla competenza del personale

ISO 7500-1:2018 Metallic materials -- Calibration and verification of static uniaxial testing machines -- Part 1: Tension/compression testing machines -- Calibration and verification of the force-measuring system

NORME UNSIDER PUBBLICATE DA CEN E ISO NEL MESE DI FEBBRAIO 2018

ISO 6507-4:2018 Metallic materials -- Vickers hardness test -- Part 4: Tables of hardness values

EN ISO 945-1:2018 Microstructure of cast irons - Part 1: Graphite classification by visual analysis

ISO 15461:2018 Steel forgings -- Testing frequency, sampling conditions and test methods for mechanical tests

EN ISO 4545-4:2018 Metallic materials - Knoop hardness test - Part 4: Table of hardness values ISO 9328-1:2018 Steel flat products for pressure purposes -- Technical delivery conditions -Part 1: General requirements ISO 9328-2:2018 Steel flat products for pressure purposes -- Technical delivery conditions -- Part 2: Non-alloy and alloy steels with specified elevated temperature properties ISO 9328-3:2018 Steel flat products for pressure purposes -- Technical delivery conditions -- Part 3: Weldable fine grain steels, normalized ISO 9328-4:2018 Steel flat products for pressure purpoLa Metallurgia Italiana - n. 3 2018

PROGETTI UNSIDER MESSI ALLO STUDIO DAL CEN (STAGE 10.99) – FEBBRAIO 2018 prCEN ISO/TR 10400 rev Petroleum and natural gas industries Equations and calculations for the properties of casing, tubing, drill pipe and line pipe used as casing or tubing prEN ISO 11960 rev Petroleum and natural gas industries Steel pipes for use as casing or tubing for wells prCEN ISO/TS 27469 Petroleum, petrochemical and natural gas industries - Method of test for fire dampers EN 10247:2017/prAC Micrographic examination of the nonmetallic inclusion content of steels

using standard pictures PROGETTI UNSIDER IN INCHIESTA PREN E ISO/DIS – MARZO 2018 PREN – PROGETTI DI NORMA EUROPEI prEN ISO 29001 Petroleum, petrochemical and natural gas industries - Sector-specific quality management systems - Requirements for product and service supply organizations (ISO/DIS 29001:2018) prEN ISO 19900 Petroleum and natural gas industries - General requirements for offshore structures (ISO/DIS 19900:2018) prEN ISO 10426-3 Petroleum and natural gas industries Cements and materials for well cementing - Part 3: Testing of deepwater well cement formulations (ISO/DIS 104263:2018) prEN ISO 10426-4 Petroleum and natural gas industries Cements and materials for well cementing - Part 4: Preparation and testing of foamed cement slurries at atmospheric pressure (ISO/DIS 10426-4:2018) prEN 448 District heating pipes - Bonded single pipe systems for directly buried hot water networks - Factory made fitting assemblies of steel service pipes, polyurethane thermal insulation and a casing of polyethylene prEN 14419 District heating pipes - Bonded single and twin pipe systems for buried hot water networks - Surveillance systems prEN 15698-1 District heating pipes - Bonded twin pipe systems for directly buried hot water networks - Part 1: Factory made twin pipe assembly of steel service pipe, polyurethane thermal insulation and a casing of polyethylene prEN 15698-2 District heating pipes - Bonded twin pipe systems for directly buried hot water networks - Part 2: Factory made fitting and valve assemblies of steel service pipes, polyurethane thermal in63

Aim news sulation and one casing of polyethylene EN 10139:2016/prA1 Cold rolled uncoated low carbon steel narrow strip for cold forming - Technical delivery conditions prEN ISO 6149-1 Connections for hydraulic fluid power and general use - Ports and stud ends with ISO 261 metric threads and O-ring sealing - Part 1: Ports with truncated housing for O-ring seal (ISO/DIS 61491:2018) prEN 10283 Corrosion resistant steel castings ISO/DIS – PROGETTI DI NORMA INTERNAZIONALI ISO/DIS 1143 Metallic materials -- Rotating bar bending fatigue testing ISO/DIS 20064 Metallic materials -- Steel -- Method of test for the determination of brittle crack arrest toughness, Kca ISO/DIS 19900 Petroleum and natural gas industries -- General requirements for offshore structures

PROGETTI UNSIDER AL VOTO FPREN E ISO/FDIS – MARZO 2018 FPREN – PROGETTI DI NORMA EUROPEI FprEN 13015 Maintenance for lifts and escalators Rules for maintenance instructions FprEN ISO 15138 Petroleum and natural gas industries Offshore production installations - Heating, ventilation and air-conditioning (ISO/FDIS 15138:2018) FprEN ISO 10855-1 Offshore containers and associated lifting sets - Part 1: Design, manufacture and marking of offshore containers (ISO/FDIS 10855-1:2018) FprEN ISO 10855-2 Offshore containers and associated lifting sets - Part 2: Design, manufacture and marking of lifting sets (ISO/FDIS 10855-2:2018)

FprEN 10164 Steel products with improved deformation properties perpendicular to the surface of the product - Technical delivery conditions

ISO/DIS 10426-3 Petroleum and natural gas industries -Cements and materials for well cementing -- Part 3: Testing of deepwater well cement formulations

FprCEN/TR 10366 Steel for the reinforcement of concrete - Weldable reinforcing steels - Reinforcement bars de-coiled by processor

ISO/DIS 945-4 Microstructure of cast irons -- Part 4: Test method for evaluating nodularity in spheroidal graphite cast irons ISO/DIS 23251 Petroleum, petrochemical and natural gas industries -- Pressure-relieving and depressuring systems ISO/DIS 29001 Petroleum, petrochemical and natural gas industries -- Sector-specific quality management systems -- Requirements for product and service supply organizations

ISO/FDIS 1083 Spheroidal graphite cast irons -- Classification ISO/FDIS 17832 Non-parallel steel wire and cords for tyre reinforcement ISO/FDIS 19203 Hot-dip galvanized and zinc-aluminium coated high tensile steel wire for bridge cables -- Specifications ISO/FDIS 15138 Petroleum and natural gas industries -Offshore production installations -- Heating, ventilation and air-conditioning ISO/FDIS 9443 Surface quality classes for hot-rolled bars and wire rod

FprEN ISO 10855-3 Offshore containers and associated lifting sets - Part 3: Periodic inspection, examination and testing (ISO/FDIS 10855-3:2018)

ISO/DIS 20088-3.2 Determination of the resistance to cryogenic spillage of insulation materials -Part 3: Jet release

ISO/DIS 10426-4 Petroleum and natural gas industries -- Cements and materials for well cementing -- Part 4: Preparation and testing of foamed cement slurries at atmospheric pressure

ISO/FDIS 10855-3 Offshore containers and associated lifting sets -- Part 3: Periodic inspection, examination and testing

FprEN ISO 9443 Surface quality classes for hot-rolled bars and wire rod (ISO/FDIS 9443:2018) ISO/FDIS – PROGETTI DI NORMA INTERNAZIONALI ISO/PRF TS 35105 Petroleum and natural gas industries -- Arctic operations -- Material requirements for arctic operations ISO 49:1994/FDAmd 1 Malleable cast iron fittings threaded to ISO 7-1 -- Amendment 1: Chemical composition of the zinc coating -- adjustment to actual requirements regarding hazardous substances ISO/FDIS 10855-2 Offshore containers and associated lifting sets -- Part 2: Design, manufacture and marking of lifting sets ISO/FDIS 10855-1 Offshore containers and associated lifting sets -- Part 1: Design, manufacture and marking of offshore containers La Metallurgia Italiana - n. 3 2018

trentasettesimo convegno nazionale

AIM

Bologna 12-13-14 settembre 2018 DEADLINE INVIO MEMORIE 20 APRILE 2018

SPAZIO AZIENDE E SPONSORIZZAZIONE

Gli interessati a presentare memorie scientifiche (sia per le sessioni orali che per la sessione poster) dovranno inviare entro il 20 aprile 2018, il titolo della memoria, i nomi degli autori e la loro affiliazione ed un sommario di circa 300 parole.

È previsto uno spazio per l’esposizione di apparecchiature, per la presentazione dei servizi e per la distribuzione di materiale promozionale.

Tutte le info sul sito www.aimnet.it/37aim.htm

Le aziende interessate a promuovere la propria attività e sostenere il Convegno potranno richiedere informazioni alla Segreteria AIM (info@aimnet.it . tel. 02 76021132).

Il 37° Convegno Nazionale AIM è sponsorizzato da HEAT TREATMENT

La Metallurgia Italiana - n. 3 2018

GROUP

SEAMLESS TUBE PLANTS

Dal forato al tubo madre – PQF®-Premium Quality Finishing

Dalla billetta al forato – tecnologia perforatore

Finitura – dal tubo madre al tubo finito

SOLUZIONI INTELLIGENTI PER LA FORATURA, L‘ALLUNGAMENTO, LA LAMINAZIONE DI FINITURA Indipendentemente dai metodi utilizzati – tradi-

tolleranze del prodotto e da un vasto campo di

zionali o di avanguardia -computerizzati con acqui-

applicazione.

sizione completa dei dati – grazie a SMS group

E c’è di più: tutto questo fa aumentare in maniera

potrete sempre contare sulla giusta soluzione.

sostanziale

La nostra gamma comprende tutto, dai tubi di

Aggiungiamo valore nell’intera catena del valore,

precisione di alta qualità ai tubi OCTG / di tra-

facciamolo insieme.

vostro

vantaggio

competitivo.

sporto. Potrete così godere dei vantaggi derivanti da un ottimo rapporto costi/efficacia, dalle strette

SMS INNSE S.p.A. Via Milano, 4 20097 San Donato Milanese (MI), Italy Phone: +39 02 2124-1 Fax: +39 02 2124-699 innse@sms-innse.it · www.innse.com

Leading partner in the world of metals