Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
ENCYCLOPÉDIE RECHERCHE
SUR L’ALUMINIUM AU QUÉBEC
Journée des étudiants du REGAL
ENCYCLOPEDIA RESEARCH
ON ALUMINIUM IN QUEBEC
REGAL Students’ Day
16e édition
16th edition 2 0 1 9
Les textes et illustrations sont une reproduction fidèle et respectent en tous points les travaux des étudiants.The texts and illustrations here in are accurate, faithful reproductions of the students’ work.
Table des matièresTable of contents
Introduction / Introduction 6
Préface / Preface 7
Formations / Formations 8
Conférences / Lectures 11
Affiches / Posters* 15
AXE 1 : Production de l’aluminium / Aluminium Production 15
AXE 2 : Transformation et applications / Transformation and applications 44
Remerciements / Acknowledgements 81
*La liste complète des affiches est disponible au début de chaque section.
ENCYCLOPÉDIE DE LA RECHERCHE SUR L’ALUMINIUM AU QUÉBEC 2019THE ENCYCLOPEDIA OF RESEARCH ON ALUMINIUM IN QUEBEC 2019
Centre de recherche sur l’aluminium - REGAL Université Laval 1065, avenue de la Médecine, local 1752A Québec (Québec) G1V 0A6 Canada Téléphone : 418 656-2362 www.regal-aluminium.ca
Montage graphique / Graphic assembly Service de reprographie de l’Université Laval
Coordination / Coordination Valérie Goulet-Beaulieu Centre de recherche sur l’aluminium - REGAL
Crédit photo de la 16e Journée des étudiants du REGAL / 16th REGAL students’ day photo credits Christian Desjardins et Valérie Goulet-Beaulieu
Tous droits réservés. Sauf à des fins de citations, toute reproduction d’un extrait quelconque de ce livre, par quelque procédé que ce soit, est strictement interdite sans la permission écrite de l’éditeur.
All rights reserved. Other than for purpose of citation, all reproduction of any part of this book, by any process is strictly forbidden without the permission of the publisher.
© Centre de recherche sur l’aluminium - REGAL, 2019 Dépôt légal - Bibliothèque nationale du Québec, 2019 / Copyright - Quebec National Library Dépôt légal - Bibliothèque nationale du Canada, 2019 / Copyright - Canadian National Library ISBN : 978-2-9815930-4-7
16E JOURNÉE DES ÉTUDIANTS DU REGAL
16TH REGAL STUDENTS’ DAY
6 Comité éditorialEditorial committee
Équipe d’organisation
Support à la JER et à l’Encylopédie
Houshang Alamdari Directeur du REGAL
Guillaume Gauvin
Adéline Paris
Valérie Goulet-Beaulieu Coordonnatrice du REGAL
Donald Picard
Geoffroy Rouget Olivier Lacroix
Carl Duchesne Président de la JER 2019
Justin Plante
7PréfacePreface
Au nom du Partenariat Canado-Norvégien sur la recherche et l’enseignement dans la production primaire de l’aluminium (CaNAl), c’est avec un grand honneur que je signe la préface de l’édition 2019 de l’Encyclopédie de la recherche sur l’aluminium au Québec. Le Centre de recherche sur l’aluminium - REGAL, qui fait la promotion de la recherche sur l’aluminium dans le triangle de l’enseignement supérieur, de la recherche et de l’industrie au Québec, est un exemple à suivre dans la promotion de la collaboration internationale. Les partenaires norvégiens de CaNAl, NTNU et SINTEF ont participé à la Journée des étudiants du REGAL 2019. CaNAl vise à consolider et à officialiser la collaboration entre la Norvège et le Canada dans le domaine de la recherche liée à la production d’aluminium. Ce partenariat vise aussi à renforcer la position des deux partenaires en tant que contributeurs majeurs à la R-D de l’industrie mondiale de l’aluminium.
Un avenir durable exige une réduction de la consommation d’énergie et des émissions de gaz à effet de serre, tel que le CO2. Ces deux aspects
sont très pertinents pour la production d’aluminium. Le Canada et la Norvège sont les principaux pays acheteurs d’aluminium dans le monde occidental, grâce à l’énergie hydroélectrique propre comme principale source d’énergie dans les deux pays. La recherche et le transfert de connaissances, facilités par le REGAL, sont essentiels pour rendre l’aluminium encore plus important pour un avenir durable.
On behalf of CaNAl – the Norwegian-Canadian Partnership in Research and Education on Primary Production of Aluminium – it is with great honor I signed the preface of the 2019 Edition of the Encyclopedia of Aluminium Research in Quebec. The Aluminium Research Centre – REGAL, promoting research on aluminium in the knowledge triangle between higher education, research and industry in Quebec, is an example to follow also in the promotion of international collaboration. The Norwegian partners in CaNAl, NTNU and SINTEF, participated in the Regal Student Day in 2019. CaNAl aims to strengthen and formalize the collaboration between the two main research environments in Norway and Canada in the field of aluminum production and aims to reinforce the position of the two partners as major R&D contributors to the global aluminum industry.
A sustainable future demands reduction in the energy consumption and emission of greenhouse gasses such as CO2. These two aspects are
highly relevant to aluminum production. Canada and Norway are major aluminum procuring countries in the western world, based on clean hydroelectric energy as the main energy source in both countries. Research and knowledge transfer, as facilitated by REGAL are key to make aluminium even more important for a sustainable future.
Tor GrandeProfesseur / Professor Norwegian University of Science and Technology (NTNU)
FORMATIONSFORMATIONS
9FormationsFormations
FORMATIONSFORMATIONS
Une quantité importante de données sont recueillies chaque jour à partir des procédés industriels. Ces données sont stockées et archivées dans des bases de données et récupérées pour l’analyse des procédés et l’amélioration continue. Il n’en est pas autrement pour l’industrie de l’aluminium primaire. Alors qu’elle commence à adopter le concept d’Industrie 4.0, il devient de plus en plus important pour les ingénieurs de procédés et le personnel de R-D d’être en mesure d’analyser efficacement les mégadonnées et d’en extraire l’information afin de prendre de meilleures décisions au moment opportun. Ce cours a pour but de fournir une brève introduction aux méthodes multivariées à variables latentes, qui sont utilisées depuis les années 1990 pour l’analyse, à la surveillance, le contrôle et l’optimisation des procédés. La nature des données de procédés sera tout d’abord présentée. Deux des méthodes à variables latentes les plus populaires, à savoir l’analyse en composantes principales (ACP) et la régression des moindres carrés partiels (PLS), seront ensuite décrites. Des exemples provenant de l’industrie de l’aluminium primaire seront finalement traités afin d’illustrer les méthodes et leur application dans le cadre de procédés industriels typiques.
Trois forces majeures sont à l’origine de bouleversements dans la manière dont nous concevons, produisons et utilisons nos produits. Il y a tout d’abord le pivot des consommateurs, du gouvernement et de l’industrie vers des objectifs de durabilité comme changement nécessaire pour l’avenir de notre civilisation. Vient ensuite la disponibilité à des coûts toujours plus faibles de la technologie, notamment les capteurs, la puissance de calcul et les algorithmes ainsi que la connectivité. Enfin, la personnalisation des produits définit de nouvelles pour gérer un nombre croissant de variations dans des délais limités. Ces forces perturbent la façon dont nous concevons, testons, produisons et exploitons nos produits et systèmes. En parvenant à comprendre et à utiliser les changements majeurs susmentionnés dans un objectif de durabilité, de numérisation et de personnalisations, nous pourrons jeter les bases des produits et des services concurrentiels de demain. Ce cours expliquera comment l’aluminium peut jouer un rôle dans ce contexte et présentera des idées ainsi que des exemples passés, présents et futurs d’application et d’innovations utilisant l’aluminium tout en tenant compte des réglementations ainsi que des besoins et des désirs du client. Une attention particulière sur la valeur potentielle des collaborations et des échanges entre les différents acteurs et disciplines sera portée tout au long de la présentation, un aspect essentiel lorsqu’il est question de l’aluminium. Les propriétés et les avantages compétitifs de l’alliage dépendent fortement de la compréhension de l’ensemble du processus, de la fonderie jusqu’à l’assemblage. Les étudiants qui assisteront à cette présentation devraient être en mesure de mieux comprendre le processus d’innovation ainsi que les différentes modalités et habitudes de travail pertinentes pour le développement de produits en aluminium.
Analyse par variables latentes de mégadonnées (Big Data) dans l’industrie de l’aluminium primaire
Pylônes électriques
Carl Duchesne Professeur Université Laval
Geir Ringen NAPIC, NTNU, Norvège
Analysis by Latent Variables of Big Data for the Primairy Aluminium Industry
Electricity Transmission Towers
Everyday, massive amounts of data are collected from industrial processes, archived in data historians, and retrieved for process analysis and continuous improvement. The primary aluminum industry is no exception. As it begins to adopt the concepts of Industry 4.0, it becomes increasingly important for process engineers and R&D staff to be able to analyze Big Data efficiently, and extract valuable information for making timely decisions. This short course aims at providing a brief introduction to multivariate latent variable methods, which are successfully used since the mid 1990’s for process analysis, monitoring, control and optimization. The course will cover the following topics. The nature of process data will be overviewed first. Then, two of the most popular latent variables methods will be described, namely Principal Component Analysis (PCA) and Projection to Latent Structure (PLS). A number of examples taken from the primary aluminum industry will finally be used to illustrate the methods in the context of typical process applications.
Three major forces are causing upheaval in the way we are engineering, producing and using products: firstly, the pivot of consumers, government and eventually industry towards sustainability goals as mandatory shift for the future of our civilizations, secondly, the unprecedent availability of digital technology at ever lower costs, notably sensors, computing power, connectivity and algorithms, and third, personalization and customization of products define new standards for how systems can handle increasing number of variations under strict time constraints. These forces introduce a disruption into how we design, test, produce and run products and systems. If we manage to understand, use and utilize the aforementioned major shifts of sustainability goals, digitalization and customization we can lay the foundation of global competitive future products and services. This lecture will outline how aluminium can play a role within this context, and give ideas and examples of past, present and future aluminium innovations under different modes of innovations. Modes, such as; regulations, combinations, system chocks, accidents, collaborations, research borders, customer needs/wants etc – all influence our design and material choices. Throughout the lecture the potential value of cross-collaborations and cross-fertilization among value chain actors and disciplines will be emphasized. This is of particular importance when it comes to make rational decisions within the long and diverse aluminium value chain, where the final product features and competitiveness depend highly upon understanding the whole picture from casthouse to assembly. Students attending this lecture should be able to better understand innovation processes and innovation work modes relevant for developing future aluminium products.
10 FormationsFormations
Raynald Gauvin Professeur
Université McGill
Laszlo Kiss Professeur, Université
du Québec à Chicoutimi
Nick Parson Rio Tinto
Ce cours présentera la caractérisation des matériaux à l’aide de microscopes électroniques à balayage (MEB) modernes pour deux modes d’opérations permettant d’améliorer la résolution spatiale; l’analyse à basse tension (faible voltage) des matériaux en vrac et l’analyse d’échantillons minces (transparents aux électrons) à 30 keV afin de réaliser la microscopie électronique en transmission (MET) dans le MEB. La microanalyse quantitative par rayons X et l’analyse par diffraction d’électrons rétrodiffusés (EBSD) seront abordées. Enfin, la microscopie électronique en transmission à balayage (METB) à 30 keV sera présentée pour le microscope électronique SU-9000 de Hitachi en effectuant l’analyse aux rayons X du lithium, la spectroscopie de pertes d’énergie (EELS) ainsi que la diffraction d’électrons par faisceau convergent (CBED) à une résolution spatiale de l’ordre du sous-nanomètre.
Les conditions optimales de fonctionnement d’une cuve d’électrolyse d’aluminium exigent une distribution homogène spatiale et temporelle des paramètres intensifs tels que la température, la concentration d’alumine ainsi que le voltage. Les fluctuations temporelles et spatiales sont corrélées et ils sont responsables des instabilités opérationnelles. La méthode d’alimentation des cuves en alumine exerce une influence déterminante sur la génération de ces inégalités.
L’aluminium domine l’industrie de l’extrusion des métaux, car il peut être transformé à chaud en profilés minces, complexes et avec de faibles tolérances à un coût relativement faible. Ce cours débutera par l’examen des différents types d’alliages d’aluminium utilisés dans l’industrie de l’extrusion ainsi que leurs applications. Les alliages de la série 6XXX représentent environ 90 % des tonnages extrudés, car ils offrent une bonne résistance mécanique, ils possèdent une bonne extrudabilité et d’excellentes caractéristiques de finition. La métallurgie du procédé d’extrusion sera présentée en détail, de même que les aspects de l’écoulement du métal dans le conteneur et la filière, qui contribuent tous à la qualité du produit final.
Microscopie électronique à balayage pour la caractérisation des matériaux
Dissolution de l’alumine dans les cuves d’électrolyse d’aluminium
Extrusion de l’aluminium
Scanning Electron Microscopy for Material Characterization
Scanning Electron Microscopy for Material Characterization
Aluminium Extrusion
This course will present state of the art materials characterization with modern scanning electron microscopes (SEM) in two modes of operation for improving the spatial resolution; low voltage analysis of bulk materials and analysis of thin specimens (electron transparent) at 30 keV to perform transmission electron microscopy (TEM) in the SEM. Quantitative X-Ray microanalysis and electron backscattering diffraction (EBSD) analysis will be covered. Finally, analytical scanning transmission electron microscopy (STEM) at 30 keV will be covered with a dedicated electron microscope, the SU-9000 from Hitachi that perform x-ray analysis of Lithium, electron energy loss spectroscopy (EELS) and convergent beam electron diffraction (CBED) with sub-nanometer spatial resolution.
The optimal operating conditions of an aluminium reduction cell require a homogenous distribution through space and time of different intensive parameters such as temperature, alumina concentration and voltage. Temporal and spatial fluctuations are correlated and are responsible for operational instabilities. The alumina method used in the reduction cells has a decisive influence on the generation of these inequalities and instabilities.
Aluminium dominates the metal extrusion industry as it can be hot worked into thin complex profiles with good tolerances at relatively low cost. This session will start by examining the various alloy types used in extruded form and their end uses. The 6XXX alloy family represents ~ 90 % of extruded tonnages as it can provide useful strength, good extrudability and excellent finishing characteristics with the added advantage that the extrusion step can be used as an in-situ solution treatment. The associated process metallurgy will be described in detail along with aspects of metal flow in the container and die which all contribute to the final quality of the end product.
CONFÉRENCESLECTURES
12 ConférencesLectures
Revêtements nanocomposites à base d’argent sur une surface en aluminium anodisé dotés d’excellentes propriétés antibactériennes
L’aluminium (Al) est un matériau attrayant pour les techniques de constructions et pour les applications générales en milieu hospitalier, telles que les surfaces fréquemment touchées (poignées de portes, barrières de lits et tables de lits). Cependant, ces surfaces constituent un réservoir bactérien idéal pour la propagation des infections nosocomiales. Chaque année, plus de 200 000 patients contractent des infections nosocomiales au Canada, dont 8 000 entraînent le décès. Par conséquent, il est idéal de concevoir des surfaces antibactériennes en aluminium avec des revêtements antibactériens à base d’argent. Dans cette étude, une approche en deux étapes a été utilisée pour fabriquer des surfaces en aluminium anodisées antibactériennes. L’aluminium anodisé revêtu de nanocomposites à base d’argent (aluminium anodisé revêtu d’un composé d’ammonium quaternaire (Ag/QUATs/AAO), l’aluminium anodisé revêtu d’Ag Polyméthylhydrosiloxane (Ag/PMHS/AAO) et l’aluminium anodisé revêtu d’Ag (Ag/AAO) ont été caractérisés par MEB, EDS, DRX et FTIR. La durabilité de ces revêtements a été étudiée à l’aide d’un test à l’abrasion. La propriété antibactérienne de Ag/QUATs/AAO a montré un excellent taux de réduction bactérienne de 99,3 %, 99,0 % et 100 % pour le modèle bactérien Staphylococcus aureus (Gram +), Pseudomonas aeruginosa (gram -) et Escherichia coli, respectivement. La propriété antibactérienne de Ag/PMHS/AAO a entraîné un taux de réduction bactérienne de 99,3 %, 99,0 % et 99,5 %, tandis que celui de Ag/PMHS/QUATs/AAO ont entraîné des taux de réduction bactérienne de 99,999 %, 99,999 % et 100 %, mais les surfaces en aluminium sans nanocomposite d’argent (aluminium anodisé et Al tel que reçu) présentaient un taux de réduction de 10,0 %, 9,0 % et 8 % et 0 %, 0 % et 0 %, respectivement, pour les bactéries modèles ci-dessus. Les surfaces en aluminium antibactériennes revêtues de nanocomposites à base de fibres sont idéales pour une utilisation en tant que surfaces tactiles antimicrobiennes en milieu hospitalier afin de réduire la prévalence des infections nosocomiales.
Silver-based nanocomposite coatings on anodised aluminium surface with excellent antibacterial properties
Aluminium (Al) is an attractive material for engineering constructions and for general hospital settings application such as frequently touched surfaces (doorknobs, bed rails and over-bed table). However, these surfaces provide a perfect bacterial reservoir for the spread of nosocomial infections. More than 200,000 patients contract nosocomial infections in Canada, of which 8,000 result in death, per year. Therefore, it is ideal to engineer aluminium antibacterial surfaces using silver-based antibacterial coatings. In this study, a two-step approach has been deployed to fabricate antibacterial anodised aluminium surfaces. The silver-based nanocomposite coated anodised aluminium (Ag-Quaternary ammonium compound coated anodised aluminium (Ag/QUATs/AAO); Ag-Polymethylhydrosiloxane coated anodised aluminium (Ag/PMHS/AAO); Ag coated anodised aluminium (Ag/AAO) has been characterised by SEM, EDS, XRD and FTIR. The durability of these coatings has been studied using abrasion test. Antibacterial property of Ag/QUATs/AAO showed excellent bacterial reduction rate of 99.3%, 99.0 % and 100 % for model bacterial Staphylococcus aureus (gram +), Pseudomonas aeruginosa (gram -) and Escherichia coli respectively. Antibacterial property of Ag/PMHS/AAO resulted in bacterial reduction rate of 99.3%, 99.0% and 99.5%, whilst that of Ag/PMHS/QUATs/AAO resulted in bacterial reduction rates of 99.999 %, 99.999% and 100%. However, aluminium surfaces without silver nanocomposite (anodised aluminum and as-received Al) showed 10.0%, 9.0% and 8.0%; and 0 %, 0 % and 0 %, respectively for the above model bacteria. Those silver-based nanocomposite coated antibacterial aluminium surfaces are ideal for use as antimicrobial touch surfaces in hospital environment for reducing the prevalence of nosocomial infections.
Henry Agbe Université du Québec
à Chicoutimi
Développement des paramètres du procédé de fabrication additive par fusion arc des alliages d’aluminium
Le procédé de fabrication additive par fusion arc (WAAM) regroupe un ensemble de techniques de fabrication additive utilisant un arc électrique comme source de chaleur et un fil métallique comme matière première pour la fabrication de composants métalliques 3D dans un processus de fabrication en multicouches. La production de composants d’alliages d’aluminium par le procédé WAAM est plus difficile qu’avec d’autres alliages en raison du nombre accru de pores de gaz et de la structure dendritique grossière au cours de WAAM. Dans cette étude, un système constitué d’une source GMAW pulsée synergique et d’un bras robotisé a été utilisé pour réaliser le processus WAAM. L’alliage étudié est un alliage Al-Si issu du dépôt de métal d’apport ER4143 utilisé pour la fabrication de murs multicouches en tant qu’échantillons réalisés par WAAM. Les paramètres de soudage ont été étudiés à l’aide de plans d’expériences (DOE), puis pour construire les échantillons, les résultats du DOE ont été utilisés pour ajuster les paramètres de soudage. Les échantillons ont été étudiés par des essais de métallographie, de microdureté et de résistance à la traction. Finalement, l’influence des paramètres de soudage et de l’apport de chaleur sur les propriétés métallurgiques et mécaniques des parois en aluminium a été discutée.
Development of parameters of the wire and arc additive manufacturing (WAAM) process of aluminum alloys
Wire and arc additive manufacturing (WAAM) process is a group of additive manufacturing (AM) techniques that use an electric arc as a heat source and a metal wire as a feedstock for the fabrication of 3D metallic components in a layer-based manufacturing process. The production of aluminum alloys components through the WAAM process is more challenging than other alloys due to the more gas pores and coarse dendritic structure during the WAAM process. In this study, a system consisting of a synergic pulsed GMAW source and a robotic arm were employed to conduct the GMAW-based WAAM process. The studied alloy is an Al-Si alloy originating from the ER4143 filler metal deposition for the manufacture of layered walls as component samples made by WAAM. The welding parameters were studied by using Design of Experiments (DOE), and then to build the specimens, the DOE results were used to adjust the welding parameters. The samples were investigated by metallography, microhardness, and tensile strength tests. Finally, the influence of welding parameters and the heat input on the metallurgical and mechanical properties of the aluminum walls were discussed.
Mehdi Gharagozloo École de Technologie
Supérieure
13ConférencesLectures
Le bio-pitch comme liant pour l’anode de carbone dans la production de l’aluminium
Le brai de goudron de houille (pitch) est le seul liant efficace pour la production des anodes. Cependant, ce matériau issu de sources fossiles est non-renouvelable et contient des hydrocarbures aromatiques polycycliques (HAP) qui sont des composants cancérogènes. Les contraintes environnementales nous incitent donc à produire un liant renouvelable pour remplacer le brai de goudron de houille dans la production des anodes. La bio-huile, l’un des produits de pyrolyse de la biomasse, peut être soumise à un traitement thermique afin de produire un matériau semblable au brai de goudron de houille, nommé le bio-pitch. Ce matériau pourrait être intéressant en raison de ses faibles teneurs en soufre et en HAP. Dans ce travail, le bio-pitch a été mélangé aux fines particules de coke pour déterminer sa capacité à produire une matrice liante uniforme et déformable. Comparativement à la matrice liante à base de brai de goudron de houille, la matrice liante à base de bio-pitch a présenté une densité et une résistivité électrique similaires. Les résultats de ce travail ont mis en évidence la possibilité d’utiliser le bio-pitch pour remplacer le brai de goudron de houille sans nuire aux performances de l’anode.
Bio-pitch as a binder for carbon anodes in aluminum production
Coal tar pitch is the only recognized binder for anode production. However, this material is a fossil-based, non-renewable binder, contains polycyclic aromatic hydrocarbons (PAH) which are considered as carcinogenic components. This situation created a need to produce a renewable and environmentally friendly binder to replace coal tar pitch for anode production. Bio-oil, one of the bio-mass pyrolysis products, can be heat-treated to produce a pitch-like material, named as bio-pitch. This material could be an attractive binder due its low sulfur and PAH contents. In this work, bio-pitch was mixed with the coke fines to determine its ability to produce a uniform and deformable binder matrix. Comparing to the coal tar pitch-based binder matrix, bio-pitch binder matrix showed similar density and specific electrical resistivity. The results of this work pointed out the possibility of using the bio-pitch to replace the coal tar pitch without impairing the anode performance.
Laser Powder Bed Fusion (LPBF) de AlSi10Mg avec étroite Particle Size Distribution (PSD)
La famille des matériaux AlSi10Mg a attiré l’attention des industries aérospatiales et automobiles avec leurs bonne résistance et densité à faible coût. Ici, nous discuterons la possibilité d’utiliser une poudre avec une distribution granulométrique des particules (Particle Size Distribution (PSD)) plus étroite dans le procédé de fusion laser sur lit de poudre (Laser Powder Bed Fusion (LPBF)). Nous avons analysé la microstructure de matériaux imprimés en relation avec les propriétés de la poudre, les paramètres d’impression et les propriétés mécaniques. Finalement, les effets des traitements thermiques T5 et T6 sur les propriétés mécaniques et microstructurales des matériaux ont été examinés.
Laser powder bed fusion of AlSi10Mg powders with narrow particle size distribution
The material family AlSi10Mg has inspired the attention of the aerospace and automotive industries because of the primary properties of strength and low density. Here we will discuss the possibility of using powder with a narrow Particle Size Distribution (PSD) in the process of Laser Powder Bed Fusion. We have analyzed the microstructure of manufactured materials in relation to the properties of the powder, the printing parameters, and the mechanical properties. Finally, the effects of T5 and T6 heat treatment on the mechanical and microstructural properties of the material was examined.
Asem HusseinUniversité Laval
Tiffany TurnerMcGill University
14 ConférencesLectures
Impact du brossage d’oxydes de surface sur les propriétés mécaniques de joints à recouvrement en aluminium 2024 soudés par friction-malaxage
Les propriétés mécaniques des joints soudés par friction-malaxage sont affectées par l’état des surfaces avant le soudage. L’effet des oxydes n’a pas été couvert de manière exhaustive dans la littérature. Ce projet étudie l’impact de la préparation de surface sur les propriétés mécaniques de joints à recouvrement lors de sollicitations monotone et cyclique. Des plaques d’AA2024 T3 AlClad ont été soudées et deux types de préparation de surface ont été effectués : non brossés et brossés manuellement. Des essais de traction et de fatigue à amplitude constante ont été faits. La résistance mécanique est semblable pour tous les échantillons. La défaillance se produit en cisaillement dans la zone soudée. Le chemin de fissure coïncide avec des fragments de AlClad dispersés dans la zone de soudure. Ces fragments offriraient donc un chemin préférentiel lors de la défaillance en cisaillement des échantillons de traction. Les échantillons de fatigue ont tous subi une défaillance en mode I, la fissure se propageant dans la plaque du dessus. La vie en fatigue est similaire pour les échantillons non brossés et brossés. Ces résultats suggèrent que le brossage manuel pour retirer les oxydes n’affecte pas les propriétés mécaniques du joint, mais qu’un brossage permettant le retrait partiel du AlClad pourrait avoir un effet.
The impact of oxide brushing on the mechanical properties of aluminum 2024 friction-stir-welded lap joints
The mechanical properties of friction-stir-welded joints are affected by the state of the mating surfaces. The effects of oxides have not been extensively covered in the literature for friction-stir-welded lap joints. This project focuses on the preparation of the faying surfaces of aluminum overlap joints. The impact of this step in the welding process on the mechanical resistance and the fatigue life of the joints is assessed. AA2024-T3 AlClad plates were friction-stir-welded with two types of surface preparation: as-clad (unbrushed) and manually brushed. Tensile and constant-amplitude fatigue tests were conducted. Mechanical strength is similar for both surface preparation (brushed and as-clad), with failure in shear of all tested specimens occurring through the welded zone. The failure path coincides with the cladding fragments in the welded zone and pure aluminum was detected on the fracture surfaces. This suggests that the cladding material fragments offered a preferred crack path for failure in shear. Fatigue specimens failed on the retreating side of the weld and fatigue lives for both preparations are similar. These results suggest that manual brushing to remove the oxide scale does not affect the mechanical properties of the joint, but that extensive brushing to remove partially the cladding material could.
Bénédicte Robitaille École Polytechnique
de Montréal
Étude expérimentale du soudage au laser de joints simple recouvrement en alliage d’aluminium 5052-H36
La demande croissante de l’allègement des structures dans l’industrie automobile fait des alliages d’aluminium (AA) des candidats potentiels pour les applications structurelles. Avec l’introduction de différentes séries AA, les nouvelles techniques d’assemblage basées sur la technologie de soudage au laser devraient être plus investiguées et mieux comprises. Ce travail présente une étude globale du soudage au laser pour un joint à simple recouvrement des plaques AA5052-H36. Ce travail est divisé en deux parties principales. Dans la première partie, nous avons été capables d’optimiser les paramètres de la procédure (mode, puissance, vitesse, distance focale et fréquence) nous permettant d’obtenir des joints soudés ne contenant aucun type de défaut, sans aucune préparation de la surface et sans l’utilisation d’aucun matériau de rapport ou gaz de protection. En ce qui concerne la deuxième partie, l’influence des paramètres manufacturiers, tels que la forme de la soudure, l’écart entre les plaques et la seconde passe de soudage, sur le comportement mécanique des joints soudés est présentée. Tous les échantillons préparés ont été testés en traction jusqu’à la rupture. La comparaison du comportement en traction des joints soudés a permis de faire ressortir l’influence de chacun des paramètres. En effet, a) la deuxième passe de soudage fournit des joints avec une résistance en cisaillement plus élevée. Cela est dû à la diminution de la concavité au niveau de la surface du cordon de soudure ; b) la forme du cordon de soudure influence la reproductibilité des résultats. Les résistances en cisaillement les plus élevées sont obtenues lorsque la direction du cordon de soudure coïncide avec la direction du chargement ; c) les plaques soudées avec un écart montrent une résistance en cisaillement plus élevée que celle obtenue sur des plaques soudées sans écart. Cette augmentation de la résistance est due à l’augmentation des interfaces de soudage obtenues dans le cas des joints soudés des plaques avec écart. Finalement, ces résultats sont d’une grande utilité pour l’intégration de la technologie de soudage au laser dans des milieux industriels.
Experimental investigation of laser welding of single-lap joints of 5052-H36 aluminum alloys
The increasing demand for lightweight components in the automotive industry makes aluminum alloys (AA) attractive candidates for structural applications, among others. With the introduction of different AA series, emerging joining techniques based on laser welding technology should be better understood. This paper presents an overall study of laser welding for a single lap joint on AA5052-H36 sheets. The influence of second welding pass, stitch-weld shape, and part-to-part gap, on the mechanical behavior of the joints is studied. The findings from the first part of the study provide guidance for laser welding of 5000 series aluminum alloy. The welding parameters for the realization of a first and a second welding pass have been optimized to produce welded joints, without blowholes, defects, or porosities. As for the second part of this study, all the prepared welded specimens were tested in tensile mode until rupture. By comparing the tensile curves, the following results are obtained for each parameter under investigation: 1- The second welding pass provides more aesthetic joints with higher shear resistance due to a decrease in the upper concavity of the welding seam; 2- The stitch-weld shape influences the repeatability of the results and the highest shear resisting stresses are obtained when the welding seam direction corresponds to the loading direction; 3- Finally, the part-to-part gap leads to higher shear resistance owing to two main reasons: the obtained welding surfaces are larger in the case of part-to-part gap and the solidified material in the fusion zone shows higher hardness. These findings are of great value for incorporating laser welding technology in the product assembly line.
Mohamad Idriss Université de Sherbrooke
PRODUCTION DE L'ALUMINIUMALUMINIUM PRODUCTION
AXE 1
Prix CQRDA remis à Redouane Farid
Prix CRITM remis à Ali Elashery
Prix Hatch remis à Nicolas Wawrzyniak
Prix RIO TINTO remis à Goril Jahrsengene
Prix ALOUETTE remis à Justin Plante
Prix ICSOBA remis à Asem Hussein, Bénédicte Robitaille et Eileen Espiritu
Prix AluQuébec-CeAl remis à An Fu
Prix ELYSIS remis à Mohammad Kavand
Prix conférence remis à Asem Hussein Prix de participation remis à Thierno Barry Saidou
Prix Metra aluminium remis à Satish Kumar Tumulu
Prix ALCOA remis à Daniel Rodrigues
Axe | Axis 1Répertoire des affiches | Posters directory
Les étudiants dont le nom est suivi d’un astérisque (*) sont récipiendaires d’un prix d’excellence pour leur affiche. Students whose name is followed by an asterisk (*) are recipients of an award of excellence for their poster.
Alarie, Jonathan Cause et ampleur des erreurs associées à la mesure du taux de dissolution de l’alumine dans la cryolithe par la méthode gravimétrique 17
Amara, Belkacem Effet du taux de chauffage sur les propriétés du biocharbon 18
Baiteche, Anwar Caractérisation d’une pâte à brasquer écologique utilisée dans l’industrie de production d’aluminium primaire 19
Barry, Thierno Saidou Évaluation de l’effet de la configuration des fours à anneau circulaires à sommet ouvert sur la résistivité électrique des anodes en carbone 20
Branvik, Trond Suivi in situ de l’atmosphère d’un four pendant la cuisson d’anodes pour l’électrolyse de l’aluminium 21
Bureau, Julie Étude sur la réaction entre la chaux et le SO2 émis au cours du procédé d'électrolyse dans la production de l'aluminium primaire 22
Cazenave, Lucie Contributions à la caractérisation du comportement thermomécanique des matériaux thermo-réactifs dans les cellules d'électrolyse 23
Chen, Bowen Évolution des propriétés des anodes pendant la cuisson 24
Hussein, Asem* Bio-pitch comme liant pour l’anode de carbone dans la production de l’aluminium 25
Ishak, Elias Estimation des propriétés physiques du coke à l’aide de techniques d’émission acoustique 26
Jahrsengene, Gøril* Spéciation d’impuretés dans le coke de pétrole par spectrométrie d’absorption des rayons X (XANEX et EXAFS) 27
Kansoun, Zahraa Mise en forme des anodes - Quantification de l’effet de la vibro-compaction 28
Kavand, Mohammad* Modèle multi-échelle FEM – DEM pour les réactions de gazéification des anodes de carbone 29
Khangholi, Siamak Nikzad Optimisation de la conductivité électrique (CE) et des propriétés mécaniques des alliages conducteurs de la série 6xxx avec divers rapports Mg / Si 30
Lacroix, Olivier Optimisation de la mise en forme des anodes des cuves Hall-Héroult 31
Lu, Ying Synthèse et caractérisation du brai biologique produit à partir du bio-huile 32
Luneng, Raymond Analyse thermogravimétrique d’isolants thermiques exposés à la vapeur de sodium 33
Manolescu, Petre L’estimation net de consommation de carbone dans la cuve d’électrolyse d’aluminium avec des méthodes multivariées 34
Mollaabbasi, Roozbeh Propriétés rhéologiques de la matrice liante 35
Morais, Brigitte Étude des facteurs influençant les résultats de la flexion des carottes d’anode 36
Rakotondramanana, Lovatiana Transport de masse par les vagues à l’interface bain-métal dans une cuve d’électrolyse 37
Rastegari, Armita Amélioration de la qualité de l’anode par la modification du brai 38
Rodrigues, Daniel* Établissement de la région de spécification de qualité multivarié pour des anodes cuites de charbon 39
Roger, Thomas Analyse thermique de l’injection de poudre refroidit dans un liquide 40
Sadeghi-Chahardeh, Alireza Étude des effets de la distribution granulométrique sur le comportement mécanique des agrégats de coke via la méthode des éléments discrets 41
Senanu, Samuel Pitting Wear on Cathodes in Aluminium Electrolysis Cells 42
Tran, Thi Hang Étude de la microstructure et des propriétés mécaniques des pâtes de ciment avec LCLL 43
17
Journée des étudiants – REGAL8 et 9 octobre 2019
Jonathan Alarie 1, L. I. Kiss 1, S. Poncsák 1, T. Roger 1, R . Santerre 2
1 Université du Québec à Chicoutimi2 Conseiller technique sur la réduction de l’aluminium, retraité de Rio Tinto
Cause and magnitude of errors in the gravimetric methods to
determine alumina dissolution rate in cryolithe.
Cause et ampleur des erreurs associées à la mesure du taux de dissolution de l’alumine dans la
cryolithe par la méthode gravimétrique.
Frequency (w/wn)
Frequency (w/wn)
x104
Cause et ampleur des erreurs associées à la mesure du taux de dissolution de l’alumine dans la cryolithe par la méthode gravimétrique
Cause and magnitude of errors in the gravimetric methods to determine alumina dissolution rate in cryolithe
To properly increase Hall-Héroult process overall efficiency, alumina dissolution in cryolite is now one of the main concerns. In the past years, the gravimetric method has been developed to follow the evolution of the apparent weight of alumina samples. Gravimetric method seems to be one of the best methods to study alumina dissolution. However, to ensure a good measurement, it is useful to identify the potential errors and their import-ance. Some of these errors can be mitigated while others need to be quantified. This paper presents some of the most influent errors affecting gravimetric measurements and their residual effect on recordings. Therefore, errors brought by conception, like vibration, as well as environmental interference, notably liquid convection, are exposed and estimated. These errors combined with those of the instruments used lead to an uncertainty in the recordings that is then determined. Finally, the margin of error around typical data is exposed to calculate the relative error.
L’optimisation du procédé Hall-Héroult est maintenant tributaire du taux de dissolution de l’alumine dans la cryolite. Les dernières années ont vu apparaître des techniques de gravimétrie permettant de suivre l’évolution du poids apparent d’un échantillon pendant sa dissolution. Cette méthode semble être l’une des meilleures tech-niques à disposition pour l’étude de la dissolution de l’alumine. Cependant, il est nécessaire de bien identifier les causes d’erreurs ainsi que leur importance afin d’obtenir des résultats fiables. Certaines de ces erreurs pourront ensuite être atténuées, tandis que d’autres seront toujours présentes. Cet article présente donc les erreurs sur la mesure les plus importantes ainsi que leur amplitude et importance. Conséquemment, les interférences appor-tées par la conception de l’appareil, notamment les vibrations, ainsi que celles provenant de l’environnement, tel que la convection naturelle, sont discutées et estimées. En combinant ces erreurs avec celles des instruments, il est possible de borner l’incertitude des mesures autour des enregistrements. Finalement, la comparaison de la marge d’erreur aux données typiques recueillies permet d’obtenir l’erreur relative des mesures.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Jonathan AlarieUniversité du Québec à Chicoutimi
L. I. Kiss - UQAC S. Poncsák - UQAC T. Roger - UQACR. Santerre - Rio Tinto
1818Effet du taux de chauffage sur les propriétés du biocharbonEffect of heating rate on biocoke properties
L’industrie de l'aluminium fait face à de nombreux défis ; parmi eux, la diminution des gaz à effet de serre (GES) tout en conservant la qualité des anodes de carbone. Cette qualité est affectée par les différentes matières premières entrant dans sa composition. Le remplacement partiel du coke de pétrole par le biocharbon permettra la réduction des GES et rendra l’anode plus écologique et respectueuse de l’environnement. Ce projet consiste à étudier l'effet du taux de chauffage sur la qualité du biocharbon. Pour cela, nous avons fabriqué quatre types de biocharbon à quatre taux de chauffage différents. Plusieurs tests ont étés réalisés afin de déterminer l'effet du taux de chauffage sur les propriétés du biocharbon. Les résultats sont présentés dans cette affiche.
The aluminum industry faces many challenges, among them the reduction of greenhouse gases (GHG) while maintaining the quality of carbon anodes. This quality is affected by the different raw materials that are used in anode production. Partial replacement of the petroleum coke by biocoke will lead to the reduction of GHG and make the anode more ecological and environment-friendly. This project involves studying the effect of the heating rate on the quality of biocoke. For this, four types of biocokes were produced at four different heating rates. Several tests were carried out in order to determine the effect of the heating rate on the biocoke properties. The results are presented in this poster.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL 8 et 9 octobre 2019
Effet du taux de chauffage sur les propriétés du biocharbon
Effect of heating rate on biocoke
properties Belkacem Amara1, Duygu Kocaefe1, Yasar Kocaefe1, Dipankar Bhattacharyay1, Jules Côté2,
André Girard3 1Université du Québec à Chicoutimi, Département des sciences appliquées, 555, boul. de l’Université,
Chicoutimi, Québec G7H 2B1; 2Aluminerie Alouette Inc., 400, Chemin de la Pointe-Noire, Sept-Îles, Québec G4R 5M9; 3Boisaco inc. 648, Chemin du Moulin, C.P. 250, Sacré-Cœur, Québec G0T 1Y0
2
Introduction et problématique Objectifs Caractériser le coke et les biocharbons produits
sous différentes conditions,
Étudier l’impact du taux de chauffage sur les propriété du biocharbons,
Déterminer l’effet de présence du biocharbon dans les anodes sur leurs propriétés,
Déterminer le meilleur biocharbon en comparant l’impact de différents biocharbons sur la qualité des anodes fabriquées avec ces derniers.
Objectifs Caractériser des quatre biocharbons produits sous
différentes conditions,
Étudier l’impact du taux de chauffage sur les propriété du biocharbons,
Déterminer l’effet de présence du biocharbon dans les anodes sur leurs propriétés,
Déterminer le meilleur biocharbon en comparant l’impact de différents biocharbons sur la qualité des anodes fabriquées avec ces derniers.
Mouillabilité FT-IR Densité en vrac MEB BET
Biocharbon 2 Biocharbon 1 Biocharbon 3
Caractérisations
Détermination des groupes fonctionnels
à la surface d’un échantillon
Étude de la mouillabilité du
biocharbon par le brai
Détermination de la densité
tassée
Caractérisation des matériaux par la
détermination de la structure
Biomasse Calcination
Méthodologie
Dans ces dernières décennies, les alumineries ont rencontré plusieurs défies incluant la dégradation de la qualité du coke de pétrole. À cause de la croissance de la demande, l’industrie est en recherche des matières premières alternatives. Le biocharbon est un choix intéressant. Le coke de pétrole, une des matières premières majeures pour la fabrication des anodes, peut être remplacé partiellement par ce produit à l’origine du bois. Ce choix rendrait l’anode plus écologique et respectueuse de l’environnement en diminuant l’émission de GES.
Biocharbon 4
Détermination de la porosité du biocharbon
Anode Anode
Cathode
Al2O3 e−
e− e−
e−
Al Al Al Al Bain Bain
Réfractaire
Matières premières
Brai de houille
Rejets d’anodes
Coke
2 Al2O3 + 3C + énergie électrique → 4 Al +3 CO2
Cal
cina
tion
Bois
Biocharbon
Conclusions Suite à une recherche bibliographique approfondie, une méthodologie a été établie pour déterminer l’impact des biocharbons produits sous
différentes conditions sur les propriétés des anodes à fabriquer avec ces derniers. Le test de mouillabilité a donné des angles de contact semblable pour les quatre biocharbons malgré les taux de chauffage différents. La densité tassée des quatre biocharbon est presque la même malgré la différence des taux de chauffage. L’analyse microscopique montre que les quatre biocharbons fabriqués présente la même structure.
Remerciements Nous remercions: • le Fonds de Recherche du Québec – Nature
et technologies (FRQNT) pour le soutien financier, et
• Boisaco et Aluminerie Alouette pour les matériaux et le soutien financier.
Quatre taux de chauffage
Résultats et discussion
Échantillon Élément
Carbone Oxygène % massique % massique
Biocharbon 1 95,73 4,27
Biocharbon 2 97,35 2,65
Biocharbon 3 92,26 7,74
Biocharbon 4 94,93 5,07
0
20
40
60
80
100
120
140
0 200 400 600 800
Ang
le d
e co
ntac
t (°)
Temps (s)
biocharbon 1
biocharbon 2
biocharbon 3
biocharbon 4
Densité en vrac adimensionnelle des biocharbons Évolution de l’angle de contact des biocharbons Analyse Microscopique des biocharbons
B-1
B-4 B-3
B-2
B-1
B-3 B-4
B-2
0,0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1,0
Biocharbon 1 Biocharbon 2 Biocharbon 3 Biocharbon 4
Den
sité
en
vrac
Amara BelkacemUniversité du Québec
à Chicoutimi
D. Kocaefe - UQACY. Kocaefe - UQAC
D. Bhattacharyay - UQAC J. Côté - Aluminerie
Alouette Inc.A. Girard - Boisaco inc.
19Caractérisation d’une pâte à brasquer écologique utilisée dans l’industrie
de production d’aluminium primaireCharacterization of an ecofriendly ramming paste used in the
primary aluminium production industry
Ramming past is a carbonaceous material that is used in the construction of electrolysis cells for the production of aluminium. The quality of the compaction can influence the pot’s operation since its purpose is to prevents any liquid aluminium or electrolytic bath leakage between the cathode blocks or the lining. A paste that does not meet the technical specifications, especially in terms of mechanical properties, can shorten the life of the cell. Properties of ramming paste based on coal tar pitch are generally well known, which is not necessarily the case with the new generation of ramming pastes, so-called ecofriendly or clean ramming pastes. In this work, we will study one of this clean ramming past currently used in industry. The density after compaction is one of the parameters that can greatly influence the thermomechanical and electrical behavior of the ramming paste. In order to reveal this relationship between density and physical properties, samples will be made at different levels of compaction and then baked at 1000°C. The samples will be characterized before and after baking (electrical and mechanical characterization). This poster focuses on the proposed methodology for studying the density effect on post-compaction properties.
La pâte à brasquer est un matériau carboné utilisé dans la construction de la cuve d’électrolyse pour la produc-tion de l’aluminium. La qualité de sa mise en place peut influencer le fonctionnement des cuves d’électrolyse, car elle assure l’étanchéité de cette dernière vis-à-vis de l’aluminium liquide et du bain électrolytique. Une pâte qui ne rencontre pas les spécifications techniques, notamment au niveau des propriétés mécaniques, peut abréger la durée de vie de la cuve. Les pâtes à base de brai de goudron sont bien connues, ce qui n’est pas nécessairement le cas avec les pâtes de nouvelle génération dites écologiques. Le but de ce projet est d’étu-dier le comportement d’une de ces pâtes de nouvelle génération actuellement utilisée en industrie. La densité après compaction est l’un des paramètres pouvant influencer grandement le comportement thermomécanique et électrique de la pâte. Afin de révéler cette relation entre densité et propriétés physiques, des échantillons seront fabriqués à différents niveaux de compaction puis cuits à 1000 °C. Les échantillons seront caractérisés avant et après la cuisson (caractérisation électrique et mécanique). La présente affiche met l’accent sur la méthodologie proposée pour révéler l’effet de densité sur les propriétés après compaction.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Objectives
Methodology
Fig. 5 Non-compacted ramming paste[1]
Fig. 6 Samples compaction[1]
Fig. 7 Samples baking [1]
Fig. 8 Example of samples [2] Fig. 9 Tensile strength [1]
Fig. 10 Compressive strength [1]
Fig. 11 Electrical risitivity [1]
Fig. 12 Thermogravimetry test [2]
Recent studies have shown that the clean ramming paste has weak points [1]such as:• The junction force between the layers;• The volumetric change which manifests itself as shrinkage during the
baking process.
Fig. 2 Joining force between the layers [1]
Fig. 4 The delamination [1]
Fig. 3 Volume change [1]
Thermo-chemo-mechanical characterization of multilayers clean ramming paste joint.
Caractérisation du comportement thermo-chimio-mécanique des joints multicouches
d’une pâte à brasquer propre.Anwar Baiteche1, Houshang Alamdari1, Mario Fafard2, Donald Picard 2, Julien Lauzon-
Gauthier3, Jayson Tessier31Department of Miniral, Metallurgical and Materials Engineering, Université Laval
2 Department of Civil and water Engineering, Université Laval 3Alcoa Corporation, Aluminum Center of Excellence, Aluminerie de Deschambault
The two problems mentioned above can cause a delamination between thelayers. This can cause the leaking of liquid metal and electrolytic bath into thecell lining which in turn can reduce the life expectancy of the electrolysis cell.
The role of ramming paste in theelectrolysis cell is :
• To prevent the infiltration of theliquid metal and the bath into thecell lining;
• To absorb the thermal expansion ofcathodes blocks and avoid crackingduring the cathodes life.
Fig. 1 Electrolysis cell section to show the ramming past position [1]
To accomplish this work and achieve the stated objectives, samples with different densities willbe manufactured and characterized. The results will reveal the influence of density on thebehavior of the ramming paste.The tests that will be carried out are:• Physical proprieties of ramming past:
o Compressive strength;o Electrical resistivity;o Thermogravemitry and dilatometry analysis;o Density.
• Surface proprieties of the layers:o Tensile strength between layers.
References[1] H. Maali «Electromechanical characterization of the ramming paste and the aging effect on its performance» master thesis2018.[2] S. Zaglafi «Thermomechanical characterization of the carbon anode during baking» master thesis 2018.
Main objective• To study the effect of density on the thermomechanical properties of an
clean ramming paste.Secondary objectives• Increase the joining force between the layers in order to avoid
delamination;• Study the thermo-chemical behavior of the ramming paste.
Context Problematic
Anwar BaitecheUniversité Laval
H. Alamdari - ULM. Fafard - UL D. Picard - ULJ. Lauzon-Gauthier - AlcoaJ. Tessier - Alcoa
2020Évaluation de l’effet de la configuration des fours à anneau circulaires à sommet ouvert sur la résistivité électrique des anodes en carboneEvaluating the effect of open top ring furnaces configuration on the electrical resistivity of carbon anodes
La résistivité électrique de l'anode en carbone est fortement influencée par les paramètres du processus de cuisson et la configuration du four. Dans la plupart des usines de production de carbone, les anodes sont cuites dans un four à anneau supérieur ouvert dans lequel elles sont empilées dans plusieurs fosses. L'empilement entraîne toutefois une augmentation des contraintes mécaniques sur les anodes inférieures. Par conséquent, la microstructure anodique et la résistivité électrique pourraient dépendre de la position de cuisson à l'intérieur du four. Les résultats présentés dans cette étude visent donc à mettre en évidence cet effet, à partir d'anodes industrielles cuites dans la même fosse.
Carbon anode electrical resistivity is greatly influenced by the baking process parameters and furnace config-uration. In most carbon plants, anodes are baked in open-top ring furnace in which they are stacked in several pits. Stacking, however, leads to an increase in mechanical stress on the bottom anodes. Consequently, anode microstructure and thus electrical resistivity could depend on the baking position inside the furnace. The results presented in this study thus aim to highlight this effect based on industrial baked anodes in the same pit.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Thierno Saidou Barry Université Laval
D. Picard - ULG. Gauvin - UL
D. De Araujo Costa Rodrigues - UL
D. Ziegler - AlcoaH. Alamdari - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
One way ANOVA test table – Position of anode
Source of Variation SS df MS F P-value F crit
Between Groups 43.890 2 21.945 3.735 0.031 3.179
Within Groups 299.637 51 5.875
Total 343.527 53
Effect of additional pressures on the bottom anodes Thermal dilatation Pitch softening
Anode internal structure rearrangement Anode electrical resistivity variation according to their position in the furnace
In many prebaked carbon anode plants for aluminium production open top ringfurnaces are used for the baking process. However, the configuration of thesefurnaces due to the anodes position therein has effect on their internal microstructureand electrical resistivity
Thierno Saidou Barry1, Donald Picard2, Guillaume Gauvin2, Daniel De Araujo Costa Rodrigues3, Donald Ziegler3 , Houshang Alamdari1
1 Department of Mining, Metallurgical and Materials Engineering, Université Laval 2 NSERC/Alcoa Industrial Research Chair MACE3 and Aluminium Research Centre – REGAL, Université Laval
3 Department of Chemical Engineering, Université Laval4 Alcoa Primary Metals, Alcoa Technical Centre, 859 White Cloud Road, New Kensington, PA, 15068, USA
Evaluating the effect of open top ring furnaces configuration on the carbon anodes electrical resistivity
Evaluation de l'effet de la configuration des fours à anneau supérieurs ouverts sur la résistivité électrique des anodes en
carbone
One way ANOVA test tableSource of Variation SS df MS F P-value F crit
Between Groups 0.043E-05 2 0.022E-05 4.036 0.018 3.017
Within Groups 2.3E-05 429 5.36E-08
Total 2.343E-05 431
4.2. Van der Pauw method data analysis – Electrical resistivity
1. Introduction
2. ObjectiveA way to improve the Hall-Héroult process energy efficiency by understanding the
effect of open top ring furnces on the electrical resistivity of anodes
3. Methodology
𝐏𝐏𝟐𝟐
𝐏𝐏𝟏𝟏
3.1 Four point probes method
4. Results and Discussions
Top location
Bottom location
Place Number of voltage measured onone anode
Face 1 4 4 4Face 2 4 4 4Total 8 8 8
V
A
I=5 A
10 cm
3.2. Van der Pauw method
The mean of BA is statisticallydifferent to the mean of TA
The mean of MA is notstatiscally different to the meanof TA and BA
The voltage of 18anodes weremeasured
For all anodes 432voltagemeasurementswere done
Observation
4.3. Electrical resistivity of anodes according to the cored sample location
6. Futur works Laboratory anodes baking under load Electrical resistivity analysis according to the weight of
the applied load
7. Acknowledgement We acknowledge our research partner NatureSciences and Engineering Research Council ofCanada (NSERC), Aluminium Research Center –REGAL and ALCOA
14 cm
Samples coring
16 cm
2 cm
BA (Bottom anodes)
MA (Middle anodes)
TA (Top anodes)
CS 1
CS 2
CS 3
TA MA BA TA MA BA
Anode from pit 1 Anode from pit 2
Two way ANOVA test table
Source of Variation SS df MS F P-value F crit
Sample 65.535 2 32.768 6.946 0.002 3.204Columns 43.890 2 21.945 4.652 0.014 3.204
Interaction 21.816 4 5.454 1.1561 0.343 2.579Within 212.285 45 4.717Total 343.527 53
5. ConclusionsWe have highlighted that, the electrical resistivity of industrial carbon anode was influenced by their
position in the furnace during the baking The ANOVA results shown that, statistically the average of the electrical resistivity of BA was Lower
than the electrical resistivity of TA
𝐏𝐏𝟐𝟐
𝐏𝐏𝟏𝟏
𝑃𝑃1 > 𝑃𝑃2
CS : Cored Samples from anode
Non destructive method – Full scale anode
Destructive method – Cored
samples
Resistance analysis
Four point probes method
Electrical resistivity analysis
Van der Pauw method
Step 1
Step 2
0,00150,001520,001540,001560,00158
0,0016
TA MA BA
Res
ista
nce
(Ω)
Position of anodes in the furnace
One way ANOVA analysis
4.1. Four point probes method data analysis – Resistance
454749515355
TA MA BAElec
trica
l res
istiv
ity
(µΩ
m)
Position of anode in the furnace
One way Anova analysis
454749515355
TA MA BA
Elec
trica
l res
istiv
ity
(µΩ
m)
Anode position in the furnace
Two way ANOVA analysis
CS 1CS 2CS 3
21Suivi in situ de l’atmosphère d’un four pendant la cuisson d’anodes pour
l’électrolyse de l’aluminiumIn situ monitoring of pit gas atmosphere during anode baking
of anodes for aluminum electrolysis
The refractory lining in the anode baking furnace is found to be greatly affected by the chemical atmosphere in the anode pit. In situ experiments measuring the pit gas composition were carried out in during anode baking. A KanthalTM probe was inserted in the pit, between the anodes and the flue wall, extracting gas which was analyzed by an FTIR spectrometer and a gas chromatograph. Detailed information on the changes in pit gas atmosphere during a baking cycle was obtained. During heating, CH
4 and H
2 dominated the pit atmosphere. At maximum
temperature (1150 °C) CO was the dominating carbon compound, while CO2 dominates below ~700 °C. During
the last part of the baking cycle, the concentration ratio of CO and CO2 follows the behavior predicted by thermo-
dynamic calculations. Water was collected from pit gas during the first 24 hours of the baking cycle, and at max-imum temperature significant amounts of HF was detected. Water is predicted to react with the bath in the green anodes to form HF, which subsequently reacts with the refractory lining forming SiF
4. The results from the pit
gas analysis is discussed in relation to observed wear patterns in spent refractory lining from autopsy samples.
Le revêtement réfractaire d’un four de cuisson d’anodes est fortement affecté par la composition chimique de l’atmosphère du four. Dans cette étude, la composition des gaz durant la cuisson des anodes a été mesurée in situ à l’aide d’une sonde KanthalTM insérée entre les anodes et la paroi du four. Cela permettait d’extraire du gaz qui était ensuite analysé par spectrométrie IRTF et par chromatographie en phase gazeuse. Les analyses ont démontré que, durant le chauffage, les gaz dominants étaient le CH
4 et le H
2. À la température maximale
(1150 °C), le CO était le composé carboné dominant alors que le CO2 domine pour une température inférieure
à ~700 °C. Durant la dernière phase de cuisson, le ratio CO/CO2 suit ce qui est prédit par la thermodynamique.
L’analyse de la vapeur d’eau collectée des gaz durant les 24 premières heures de cuisson a révélé qu’à la température maximale, une quantité significative de HF était présente. Il est prédit que l’eau réagisse avec le bain contenu dans les anodes crues pour former du HF qui, à son tour, réagit avec le revêtement réfractaire, formant ainsi du SiF
4. Les résultats sont discutés en relation avec les modes d’usure de revêtements réfractaires
de pièces ayant été en service.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Trond BranvikNorwegian University of Science and Technology
H. Gaertner - SINTEFA. P. Ratvik - SINTEFT. Grande - NTNUT. A. Aarhaug - SINTEF
Journée des étudiants – REGAL8 et 9 octobre 2019
Trond Brandvik1, Heiko Gaertner2, Arne P. Ratvik2, Tor Grande1 and Thor A. Aarhaug2
1 NTNU Norwegian University of Science and Technology, Trondheim, Norway2 SINTEF Industry, Trondheim, Norway
In situ monitoring of pit gas composition during baking of
carbon anodes
Objective
Extraction and analysis of pit gas atmosphere during anode baking. Assessing the effect of the atmosphere
on degradation of refractory lining
Extraction and analysis of pit gas Setup• KanthalTM probe and thermocouple in the
packing coke• Condensation vessel and filter unit• Heating tube (200 ºC)• FTIR and GCAnalysis• Fourier transformed infrared spectroscopy• Gas chromatography
Challenges
• Non-continuous flow• Overlapping IR signal• Too strong IR signal
Summary of findings
Reducing atmosphere
CO2 (g) + C (s) = 2CO (g)
Degradation of refractory lining
• Temperature cycling• Reduction of silica in the lining• Water in the anodes: Formation of HF • Corrosion of lining by reaction with HF
• Densification by creep
Acknowledgements: Thanks to the CaRMa project with its partners SINTEF, NTNU, Hydro Aluminium, Alcoa Norway, Elkem Carbon, Skamol A/S and the Norwegian Researchcouncil for financial support. Thanks to the people at Alcoa Mosjøen for help and guidance during the measurements at the plant.
Reducing atmosphere
SiO2 (s) + CO (g) = SiO (g) + CO2 (g)
Corrosion by HF
Al6Si2O13 + 4HF = 2SiAl2F2O4 + Al2O3 + 2H2O4SiAl2F2O4 = 2SiF4 + Al2O3 + Al6Si2O13
2222Étude sur la réaction entre la chaux et le SO2 émis au cours du procédé d'électrolyse dans la production de l'aluminium primaireStudy on the reaction between lime and SO2 released during the electrolysis in the production of primary aluminum
Lors du procédé électrolytique pour la production de l’aluminium, le SO2 est émis. Il est un gaz nocif pour
l’environnement et une source des pluies acides. L’unité de traitement des gaz d’échappement actuel favorise de manière préférentielle l’élimination du HF (plus de 99%), de sortes qu’il est nécessaire d’ajouter une autre unité pour la captation du SO
2. Différentes méthodes existent pour l’élimination du SO
2 émanant des gaz effluant
de divers procédés. Il y a déjà certaines technologies proposées pour l’industrie de l’aluminium. Dans ce projet, l’objectif principal est d’étudier fondamentalement et de bien comprendre la réaction chimique du SO
2 avec la
chaux sous les conditions à la sortie du centre de traitement des gaz. Cela permettra de déterminer les conditions favorables pour la réaction chimique chaux/SO
2. Une analyse critique des systèmes disponibles pour la désul-
furisation de différentes sources de produit sulfureux provenant de diverses industries servira à concevoir un système de captation du SO
2. La méthode prévoit de déterminer l’effet de la variation de différentes conditions
d’expérimentation, de comprendre les aspects thermodynamiques du système et d’étudier la cinétique de la réaction chaux/SO
2.
During the electrolytic process for the aluminum production, SO2 is released. It is a harmful gas for the environ-
ment and a source of acid rain. The current gas treatment center preferentially favors the removal of HF (more than 99%) such that it is necessary to add another unit in order to capture SO
2. Different methods exist for SO
2
removal from effluent gases from various processes. There are already some technologies proposed for the aluminum industry. In this project, the main objective is to fundamentally study and understand the chemical reaction of SO
2 with lime under the conditions at the exit of the gas treatment center. This will help determine
the favorable conditions for the lime/SO2 chemical reaction. A critical analysis of the available systems for the
desulfurization of different sources of sulfurous product from various industries will be carried out to design a SO2
removal system. The method aims to determine the impact of the variation in different experimental conditions, to understand the thermodynamic aspects of the system, and to study the kinetics of the lime/SO
2 reaction.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Julie Bureau Université du Québec
à Chicoutimi
D. Kocaefe - UQACY. Kocaefe - UQAC
S. Gaboury - Rio Tinto Y. Dargis - Graymont
Journée des étudiants – REGAL8 et 9 octobre 2019
Julie Bureau1, Duygu Kocaefe1, Yasar Kocaefe1, Simon Gaboury2 et Yves Dargis31 Centre universitaire de recherche sur l’aluminium (CURAL), Université du Québec à Chicoutimi,
555 boulevard de l’Université, Chicoutimi, Québec G7H 2B12 Rio Tinto, Arvida Research and Development Center (ARDC), 1955, Boulevard Mellon, C.P 1250, Jonquière, Quebec, Canada G7S 4K8
3 Graymont, 206-25, rue De Lauzon, Boucherville, Québec, Canada, J4B 1E7
Étude sur la réaction entre la chaux et le SO2 émis au cours du procédé
d'électrolyse dans la production de l'aluminium primaire
Study on the reaction between lime and SO2emitted during the electrolysis process in the production of primary aluminum
1. Introduction et problématiqueObjectif principal: Étudier fondamentalement et comprendre la réaction chimique chaux/SO2 sous les conditions à la sortie du CTG.Objectifs spécifiques:1. Déterminer quels systèmes sont
susceptibles de capter efficacement le SO2;
2. Comprendre les aspects thermodynamiques, la cinétique de la réaction chaux/SO2 et les conditions d’utilisation les plus favorables du système conçu à l’échelle laboratoire et au milieu industriel;
3. Effectuer une analyse des résultatsafin de déterminer les conditionsd’utilisation les plus favorables pourla captation de SO2.
4. Méthodologie
6. Remerciements5. Travaux futurs Nous tenons à remercier vivement Rio Tinto etGraymont pour leur soutien financier ettechnique ainsi que le soutien financier duConseil de recherches en sciences naturelleset en génie du Canada (CRSNG) et del'Université du Québec à Chicoutimi (UQAC) .
7. Références• Compléter la revue de littérature.• Se procurer différentes chaux et conception des différents systèmes
expérimentales• Étude de la réaction chaux/SO2 selon les paramètres choisis.• Analyse des données expérimentales afin de déterminer les conditions
favorables pour maximiser la captation de SO2.
1. lca_report_2015_final.pdf.2015; disponible au : http://www.world-aluminium.org/media/filerpublic/2017/06/28/lca_report_2015_final.pdf.
2. Kocaefe, D., An experimental and theoretical study on the kinetics of basic oxide sulfation with SO2 and SO3. 1986, University of New Brunswick: Ottawa.
3. Ciullo, P.A., Industrial minerals and their uses: a handbook and formulary. 1996, Westwood, New Jersey, USA: Noyes Publications.
2. Objectifs
Images modifiées à partir de : https://www.ec.gc.ca/lcpe-cepa/272730B3-996D-4903-89E2-DAF0A12A588F/CdP-CoP-Aluminium_fra.pdf et http://www.ecotoxicologie.fr/acides.php
Figure 1. Schéma de principe du procédé électrolytique Figure 2. Pluies acides (sources, formation et conséquences)
Contenant du soufre
SO2, SO3, HF, …
SO2SO3NOx
H2O Acides formés: H2SO3H2SO4HNO3
Entraînement de métaux ayant des effets toxiques
pH ↓
Étude de la réaction chaux/SO2
Paramètres étudiés :• Conditions expérimentales
et caractéristiques des particules de chaux;
• Aspects thermodynamiques;
• Cinétique de la réaction;• Efficacité de captation.
Partie A(en laboratoire)
Étude préliminaire de la réaction chaux/SO2
Thermogravimétrie
Étude de la réaction chaux/SO2:À l’aide d’un lit granulaire
Caractérisation des particules
Partie B(milieu industriel, CTG)
Étude de la réaction chaux/SO2:À l’aide d’un lit granulaire
Caractérisation des particules
Analyse des résultats
Détermination de la réactivité du SO2 en fonction de:
Conditions expérimentales :• Températures;• Humidités;• Débits du gaz; • Compositions du gaz.
Chaux:• Tailles des particules;• Types de particules
(structure).
3. Revue de littératureIl existe trois principaux types de procédés de désulfurisation:A. Procédés humides
- Coûteux et requière le traitement desrésidus obtenus- Sous-produit contenant une grandequantité d’eau
B. Procédés semi-humidesC. Procédés secs
- Sous-produit sec pouvant être captédans un filtre et facilement manipulable- Ne produit pas de rejet liquide
• La majorité des études rapportées dans la littératureest applicable aux conditions associées à l’éliminationdes gaz émis par les centrales thermiques.
• Sous les conditions au CTG, aucune étudefondamentale de la réaction chimique entre la chauxet le SO2 n’est publiée à notre connaissance.
• Le SO2 en présence de chaux peut former du gypse(CaSO4). Ce produit peut être utilisé: en construction(panneaux muraux, Stucco, ciment Portland), enagriculture comme engrais, et d’autres usages: enverrerie ou comme agent de remplissage blanc dansla peinture et le papier.3 Figure 3. Certains des appareils de mesure utilisés pour l’étude de la réaction
Micro-thermogravimétrie Chromatographie en phase gazeuse
Isotherme d’adsorption de BET Microscopie optique FTIRMEB
Schéma du système pour le lit granulaire
Gaz àtraiter
(préalablement caractérisé)
Gaz àla sortie
(à caractériser)
Fournaise Particules de chaux
Au centre de traitement des gaz (CTG), l’unité actuelle capte préférentiellement le HF (> 99%) libérant ~15,28 kg de SO2/t.de Al1 (un des contributeurs des pluies acides). Différentes technologies de captation existes sur le marché2, mais la plupartsont coûteuses et génèrent des résidus qui nécessitent d’être traités ou qui ne peuvent pas être revalorisés. Certainestechnologies pour l’industrie de l’aluminium sont utilisées (majoritairement des procédés humides).
23Contributions à la caractérisation du comportement thermomécanique
des matériaux thermo-réactifs dans les cellules d'électrolyseCContributions to the characterization of thermomechanical
behavior of thermo-reactive materials in aluminum reduction cells
The life span of the electrolysis cells is strongly influenced by the nature and the interaction of the different materials in the lining. A good cell design requires a good knowledge of the behavior of these materials. Con-sequently, a research program was established in collaboration with Aluminerie Alouette Inc. to study their elec-trolysis cells during the start-up and in-service phases in order to understand better the cell behavior. Following a 4-step approach (numerical, experimental, in situ measurement, and operation), the present project focuses on the experimental component to investigate the evolution of the behavior of two thermo-reactive materials, which play a significant role in the electrolysis cell behavior. More specifically, the evolution of the mechanical and thermal properties of the NeO
2 monolithic paste in relation to its baking level as well as the regular refractory
brick (BRO) in its original form and after contamination with sodium vapor will be studied. The research results obtained will be used as an input to the numerical models to represent the behavior of these two materials realistically in their respective use.
La durée de vie des cellules d’électrolyse est fortement conditionnée par la nature ainsi que par l’interaction des différents matériaux constituant leur revêtement. La qualité du design passe donc impérativement par la bonne connaissance du comportement desdits matériaux. En ce sens, un programme de recherche en partenariat avec Aluminerie Alouette Inc. a été mis sur pied afin de permettre une meilleure compréhension du comportement de leurs cellules d’électrolyse en phase de démarrage et en service. Conséquent d’une approche en 4 volets (numé-rique, expérimental, mesure in situ, exploitation), le projet dont il est question ici porte sur le volet expérimental et s’attaque à l’étude de l’évolution du comportement de deux matériaux thermo-réactifs jouant un rôle significatif dans le comportement de la cellule d’électrolyse. Plus spécifiquement, on s’attardera à l’évolution des propriétés mécaniques et thermiques de la pâte monolithique de type NeO
2 en fonction de son niveau de cuisson, ainsi que
de la brique réfractaire ordinaire (BRO) dans ses phases saine et contaminée par le sodium en phase vapeur. Les résultats obtenus devront permettre d’alimenter les modèles numériques afin de permettre une représentation réaliste du comportement de ces deux matériaux dans leur contexte d’utilisation respectif.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Lucie CazenaveUniversité du Québec à Chicoutimi
D. Marceau - UQACD. Kocaefe - UQACS.-O. Tremblay - UQACS. Charest - Aluminerie Alouette Inc.J. Côté - Aluminerie Alouette Inc.
2424Évolution des propriétés des anodes pendant la cuissonAnode paste properties evolution during the baking process
L’aluminium est produit via le procédé d’électrolyse Hall-Héroult dans lequel l’anode est utilisée comme une électrode positive. La qualité de l’anode précuite joue un rôle crucial dans l’efficacité du procédé d’électrolyse de l’aluminium. Pour produire des anodes de haute qualité dans l’industrie de l’aluminium, le procédé de cuisson des anodes nécessite une compréhension approfondie des mécanismes qui régissent l’évolution des propriétés de l’anodes dans des conditions de température élevée. L’objectif de ce projet est de développer un modèle de comportement thermo-chimico-poromécanique pour la pâte d’anode pendant la cuisson et ce, en utilisant la théorie des milieux poreux réactifs dans un cadre thermodynamique. À cette fin, une variable d’état interne, appelée "indice de contraction", et un nouvel indice de cuisson sont définis pour caractériser la progression chimique de la pyrolyse du brai dans le squelette de l’anode. En considérant une énergie libre, dans un cadre Lagrangien, et un potentiel de dissipation, un modèle constitutif, liant la pyrolyse chimique au comportement mécanique, est développé. Les équations régissant la diffusion des volatiles et le transfert de chaleur à travers le mélange sont dérivées. Des essais expérimentaux à haute température (thermogravimétrie, fluage, compression, etc.) seront effectués pour obtenir l’évolution de la perte de masse, des comportements mécaniques et des paramètres essentiels (Module de Young, perméabilité, coefficients de dilatation thermique, etc.).
Aluminium is reduced from alumina by Hall-Héroult electrolysis process in which the anode is utilized as the posi-tive electrode. The quality of the prebaked anode plays a crucial rule in the efficiency of the aluminium electrolysis process. To produce high-quality anodes in aluminium industry, the anode baking process calls for a deep under-standing of mechanisms that govern the evolution of the anode mixture properties under the high-temperature condition. Therefore, the aim of this project is to develop a thermo-chemo-poromechanical behavior model for the baking anode paste by using the theory of reactive porous media and within the thermodynamic framework. For this purpose, an internal state variable "shrinking index" and a new baking index are defined to characterize the chemical progress of the pitch pyrolysis in the anode skeleton. By formulating a Lagrangian free energy and a dissipation potential, a constitutive model linking the chemical pyrolysis with the mechanical behavior is achieved. Field equations governing the volatile diffusion and the heat transfer through the mixture body are derived. Experimental characterizations at high temperature (TGA, Creep, Compression, etc.) will be performed to obtain the evolution of the mass loss, the mechanical behavior and essential parameters (Young’s modulus, permeability, thermal expansion coefficients, etc.).
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Anode paste properties evolution during baking process
Évolution des propriétés de la pâte d’anode durant la cuisson
B. Chen1, D. Picard1, G. Gauvin2, H. Chaouki1, H. Alamdari2, J. Tessier3, M. Fafard1
1 Department of Civil and Water Engineering, Université Laval2 Department of Mining, Metallurgical and Material Engineering, Université Laval
3 Alcoa Corporation, Aluminum Center of Excellence, Aluminerie de Deschambault
In the industrial aluminum production, the anode baking processis a complex process of high-energy consumption. It is also themost expensive stage in the anode manufacturing process. Afterthe removal of prebaked anodes from the baking furnace (Figure1), cracks can be observed on their surfaces (Figure 2), whichaffects the efficiency of the Hall-Héroult process.
Figure 1 Anode baking furnace
Figure 2 Cracks on the surface
Figure 3 Baking process with swelling and shrinkage effect
( )( )( )
t
ch
vp
REV: Representative elementary volume
V : Total swelling volume
V : Chemical shrinkage volume
V : Viscoplastic deformation volume
Irreversible volume expansion
Total swelling volume
Figure 4 Representative elementary volume (c-closed pores; t-transport pores; b-blind pores)
AggregatesBinder
Gas void
Figure 5 Carbon structure [Hulse K.L., 2000]
Porosity in the anode paste can be categorized into different structural types (e.g. closed, transport andblind) (Figure 4). During the baking process, three mechanisms interfere in the elementary volume:• Pitch carbonization that carbonizes the pitch into coke;• Devolatilization process that the solid carbon skeleton loses the volatile;• Structural changes of pores inside the anode block.
i) Stress state:
Figure 8 Evolution of the baking index [Zaglafi S., 2019]
Thermogravimetric analysis (TGA)
( )
( )( )
( )
sa
o
1a
1o
1 1
Arrhenius law
exp 1
82.7 kJ mol Activation energy factor
7.28e+4 h Pre-exponential factor
8.314 J mol K Ideal gas constant
Ed X k Xdt RT
E
k
R
−
−
− −
= − −
= =
=
Figure 7 Evolution of compressive strength[Zaglafi S., 2019]
Mechanical properties evolution
Figure 6 Evolution of the Young Modulus [Zaglafi S., 2019]
apvv
kp
= − xwiv) Darcy’s law:
ii) Volatile diffusion equation:
iii) Thermal balance equation:
v v v volv vˆ
n n n Ep p p cRT t t RT RT t
+ + + = x w
Local pressure
Pressure-porosity coupling
Pressure-strain coupling
Volatile diffusion
Mass supply
s-v c r= + +S S S S
Solid-volatile interaction
Chemical shrinkage
Thermo-mechanical
( )s
d vε v t
sd TC T T J rdt T
+ − + =
x xw q
Local temperature variation
Thermal convection
Energy source
Thermal conduction
Energy sinking
T= − xqv) Fourier’s law:
Mechanical properties (Young Modulus and Compressive Strength) can be affected by the chemical reaction of pitch pyrolysisand the evolution of the baking index can be used to observe, estimate and control the state of the baking anode.
The following methods are used for this project:• Development of a thermo-chemo-poromechanical model for
the baking process of the anode paste to predict property evolutions;
• Sample fabrication;• Baking of samples;• Mechanical tests: compression tests at different baking
temperatures (Young’s Modulus, compressive strength, Poisson’s ratio, etc.);
• Thermogravimetric analysis (baking index);• Permeability and dilatometry tests (permeability and volume
change);• Model calibration and numerical simulation.
In this project, a theoretical model is established for modeling theanode baking process. Experimental results identifying theYoung’s Modulus and compression strength are obtained. Themass loss of the anode are used to identify involving parametersin the Arrhenius law. In future works, other parameters will beexperimentally identified (permeability, volume change, etc.) andnumerical models will be built and implemented in the COMSOLsoftware.
• Hulse K.L., 2000, Anode Manufacture: Raw materials, formulation and processing parameters, R&D Carbon.• Zaglafi S., 2019, Caractérisation thermo-chimio-mécanique de l’anode de carbone et identification des paramètres des lois
de comportement, Mémoire de Maîtrise, Université Laval.
Context
ObjectiveThis project is to investigate the evolution of mechanicalproperties of the anode during the baking process by consideringeffects as follows:• Swelling of the anode paste due to the pore pressure increase
and the thermal expansion (Figure 3 and Figure 5);• Chemical shrinkage due to the carbonization of the pitch;• Viscous effects on the mechanical deformations of the anode at
higher temperatures.
Methodology
Thermo-chemo-poromechanical model
Conclusion and Future works
References
As seen in Figures 6-7, mechanical properties ofthe anode paste are influenced by the soaking(20h) and both the Young modulus and thecompressive strength show a similar trendduring the baking process. Their valuesdecrease below 200 ºC because of the pitchsoftening and further fluctuate between 200 ºCand 400 ºC probably due to the viscosity changeof the anode material. Above 500 ºC, theyincrease, which results from the solidification ofthe pitch in the paste.
Structural transformation of the anode paste
+
Bowen Chen Université Laval
D. Picard - ULG. Gauvin - UL
H. Chaouki - UL H. Alamdari - ULJ. Tessier - Alcoa
M. Fafard - UL
25Le bio-pitch comme liant pour l’anode de carbone dans la production de l’aluminium
Bio-pitch as a binder for carbon anodes in aluminum production
Coal tar pitch is the only recognized binder for anode production. However, this material is a fossil-based, non-re-newable binder, contains polycyclic aromatic hydrocarbons (PAH) which are considered as carcinogenic compon-ents. This situation created a need to produce a renewable and environmentally friendly binder to replace coal tar pitch for anode production. Bio-oil, one of the bio-mass pyrolysis products, can be heat-treated to produce a pitch-like material, named as bio-pitch. This material could be an attractive binder due its low sulfur and PAH contents. In this work, bio-pitch was mixed with the coke fines to determine its ability to produce a uniform and deformable binder matrix. Comparing to the coal tar pitch-based binder matrix, bio-pitch binder matrix showed similar density and specific electrical resistivity. The results of this work pointed out the possibility of using the bio-pitch to replace the coal tar pitch without impairing the anode performance.
Le brai de goudron de houille (pitch) est le seul liant connu pour la production des anodes. Cependant, ce matériau issu des fossiles est non-renouvelable et contient des hydrocarbures aromatiques polycycliques (HAP), qui sont des composants cancérogènes. Cette situation a créé le besoin de produire un liant renouvelable pour remplacer le brai de goudron de houille dans la production des anodes. La bio-huile, l'un des produits de pyrolyse de la biomasse, peut être soumise à un traitement thermique afin de produire un matériau semblable au brai de goudron de houille, nommé le bio-pitch. Ce matériau pourrait être intéressant en raison de ses faibles teneurs en soufre et en HAP. Dans ce travail, le bio-pitch a été mélangé aux fines particules de coke pour déterminer sa ca-pacité à produire une matrice liante uniforme et déformable. Comparativement à la matrice liante à base de brai de goudron de houille, la matrice liante à base de bio-pitch a présenté une densité et une résistivité électrique similaires. Les résultats de ce travail ont mis en évidence la possibilité d'utiliser le bio-pitch pour remplacer le brai de goudron de houille sans nuire aux performances de l'anode.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Asem HusseinUniversité Laval
H. Alamdari - ULD. Ziegler - AlcoaJ. Tessier - Alcoa
Journée des étudiants – REGAL8 et 9 octobre 2019
1. IntroductionCoal tar pitch (CTP), is the only recognized binder for anode production. However, it is a non-renewable binder, contains polycyclic aromatic hydrocarbons (PAH) which areconsidered as carcinogenic components. Thus, there is a need to produce a renewable and environmentally friendly binder to replace the CTP for the anode production.Bio-oil, one of the bio-mass pyrolysis products, can be heat-treated to produce a pitch-like material, named as bio-pitch. This material could be an attractive binder due itsinteresting binding properties, low sulfur and low PAH contents. In this work, bio-pitch was produced in the laboratory, then mixed with the coke fines to determine itsability to produce a deformable, dense and electrically conductive binder matrix. Compared to the CTP-based binder matrix, the BP-based one showed similar density andspecific electrical resistivity.
2. Objectives• To identify the optimum heat treatment conditions to convert the bio-oil
to a bio-pitch with acceptable properties.• To determine the ability of bio-pitch to mix with fine coke to produce a
dense and electrically conductive binder matrix.
3. Experimental work3.1 Preparation of bio-pitch:
3.2 Preparation of binder matrix samples:
Furnace
Mechanical stirrer
Bio-oil
(0.5 C /min)
4.2 Properties of binder matrix samples:• Green and baked apparent density
• Specific electrical resistivity
• Optical microscopic observations
5. Conclusion• Bio-pitch was successfully produced by heat treatment of bio-oil.• Comparing to coal tar pitch, bio-pitch showed acceptable viscosity, softening point
and sulfur content.• Bio-pitch was able to cover the coke particles forming a uniform and deformable
binder-matrix.• Baked bio-pitch binder matrix samples showed similar density and specific
electrical resistivity to that of the coal tar pitch binder matrix samples.
Compaction (60 MPa)
Green Sample
Baking (1100 ºC)
Mixing (178 C )Coke fines (75 %)
Bio-pitch (25 %)
During the anode paste production, bio-pitch must be able to mix with the coke fines to produce ahomogenous and deformable binder matrix.
The compacted binder matrix samples were characterized in terms of apparent density, mass lossand volume changes during the baking process. The specific electrical resistivity was determinedusing the Van der Pauw method.
1,591,58 1,59
1,53
1,54
1,55
1,56
1,57
1,58
1,59
1,60
1,61
1,62
1,63
BMA-CTP BMA-160 BMA-180
Appa
rent
den
sity
(g·c
m-3
)
Baked apparent densityGreen apparent density
-15
-10
-5
0
5
10
15
BMA-CTP BMA-160 BMA-180
Volu
me
shrin
kage
(%) M
ass
loss
(%)
Volume shrinkage Mass loss
Green and baked apparent density of binder matrix samples Mass loss and volume shrinkage after baking
50,6
49,0
48,4
45
46
47
48
49
50
51
52
53
BMA-CTP BMA-160 BMA-180
Spec
ific e
lect
rical
resis
tivity
(µΩ·
m)
200 𝜇𝜇m 200 𝜇𝜇m
BMA-CTP= Coal tar pitch binder matrixBMA-160= Bio-pitch (160 C) binder matrixBMA-180= Bio-pitch (180 C) binder matrix
SER measurement setup
4. Results4.1 Effects of final heat treatment temperature on the bio-pitch properties:Sample PY(%) SP(°°C) Vis (Pa·s) CV(%) QI(%) Density(gcm-3) O(%) S(%) P(ppm)
CTP - 109.5 2.54 61.8 16.5 1.363 - 0.62 11Bio-160 60 79.8 0.4 33.6 0.8 1.306 32.5 0.01 15Bio-180 57 91.5 0.8 34.9 1.8 1.316 30.1 0.01 14Bio-200 53 115.2 8.7 38.3 2.2 1.322 29.3 0.01 16Comparing to coal tar pitch, bio-pitch produced at 160 C and 180 C (Bio-160 and Bio-180) showed lowersoftening point and viscosity. These samples are expected to mix with coke fines to form homogeneousbinder matrix.
7. Benefits • To provide an alternative binder for carbon anodes.• To reduce the carbon footprint of the aluminum industry.
6. Future work• Lab scale anodes will be produced using bio-pitch as a binder. Properties of these
anodes will be compared to coal tar pitch anodes (reference anodes).
Bio-oil was heated at 160 C, 180 C and 200 C to determine the optimum treatment temperature.
The optimum treatment temperature was determined according to the bio-pitch properties such assoftening point, viscosity and coking value.
Asem Hussein1,2, Houshang Alamdari1,2, Donald P. Ziegler3, Jayson Tessier4
1NSERC/Alcoa Industrial Research Chair MACE3 and Aluminium Research Centre – REGAL, Laval University2 Department of Mining, Metallurgical and Material Engineering, Université Laval
3Alcoa Primary Metals, Alcoa Technical Centre, 859 White Cloud Road, New Kensington, PA, 15068, USA4Alcoa Corporation, Aluminium Center of Excellence, Aluminerie de Deschambault
Bio-pitch as a binder for carbon anodes in aluminium production
Bio-pitch comme liant pour l’anode de carbone dans la production de l’aluminium
Bio-pitch binder matrix Coal tar pitch binder matrix
100 mm
50 mm
Binder matrix slices
Prix Award
ex æquo
2626 Estimation des propriétés physiques du coke à l’aide de techniques d’émission acoustiqueEstimation of coke properties using acoustic emission techniques
La production d'aluminium par le procédé de Hall-Héroult est influencée, entre autres, par la variation des pro-priétés de l'anode en carbone. Celles-ci sont plus difficiles à contrôler en raison de la dégradation de la qualité et l'augmentation de la variabilité des particules de coke de pétrole utilisées pour fabriquer les anodes en carbone. D'où l'importance d'une méthode rapide et non destructive qui pourra être intégrée à la chaîne de production d'anodes et qui mesurera en temps réel la densité et la porosité des particules de coke. Un montage d'analyse acoustique a été préparé et testé pour estimer la densité et la porosité des particules de coke en utilisant la trans-formée de Fourier (FFT) pour la décomposition en fréquences couplée à des méthodes de régression multivariées pour l'estimation de modèles prédictifs. Cette étude vise à améliorer les résultats de prédiction par l’utilisation de la transformée en ondelettes en continue (CWT). Cette méthode de décomposition temps-fréquence permettrait d’obtenir une signature acoustique plus spécifique des particules de coke de différentes tailles et provenant de plusieurs sources. Le potentiel de l’approche proposée semble très prometteur en raison de la bonne capacité prédictive de la densité et porosité des particules obtenues jusqu’à présent, sur des échantillons non-mélangés provenant de différentes sources et tamisés en plusieurs tailles granulométriques.
The production of primary aluminum using the Hall-Héroult process is influenced by the variations in carbon anode properties. It is a challenge to control these properties due to the raw material degradation and variability. Petroleum coke is one of the main raw materials used to produce the carbon anodes. Therefore, it is important to develop a fast and non-destructive method that can be integrated into the anode production line in order to measure the density and porosity of the coke particles in real time. An experimental set-up based on an acoustic analysis was developed and tested to estimate the density and the porosity of coke particles. Fast Fourier Transform (FFT) was used for frequency decomposition, and multivariate regression methods to analyze the acoustic signals and predict coke properties. This study aims at improving the predictive performance of the model by using the Continuous Wavelet Transform (CWT). This time-frequency decomposition method will allow the capture of the acoustic signature of coke particles of different sizes coming from several sources. So far, the approach gave promising results for predicting the density and the porosity of coke particles obtained from different suppliers and sieved into several size classes.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Elias Ishak1, Donald Picard2, Carl Duchesne1 et DuyguKocaefe3
1 Department of Chemical Engineering, Laval University2 Department of Civil Engineering, Laval University
3 Department of Applied Sciences, University of Québec at Chicoutimi
Estimation of coke properties using acoustic emission
techniquesEstimation des propriétés physiques
du coke à l’aide de technique d’émission acoustique
1-Context: Degradation of the quality and increase in the variability of raw materials
can affect the quality of the carbon anode Hence it is important to have a fast and non-destructive estimation
method that can be integrated into the anode production line in order to take real-time measurements of coke density and porosity
Acoustic emissions have already been tested¹ to estimate the density and porosity of coke particles
2-Objectives: The main objective is to continue the development of a soft-sensor¹
which will complement the physical sensor (microphone) Demonstrate that coke density and porosity can be predicted from sound
measurements ¹ Rouget, G., Manolescu, P., Picard, D., Duchesne, C., Alamdari, H., Kocaefe, D. “Acoustic Emission Techniques to Measure the
Properties of Coke Particles–A First Foray”. ICSOBA 2017.
Figure 1. Schematic view of the sound acquisition set-up⁴.
4-Results: (Single and mixture of coke particle size and supplier) Supervised classification of coke particles size with PLS model:
Prediction of apparent density:
Prediction of open porosity:
5-Conclusion:By measuring the sound of impact coming from coke particles of same size or mixture of different sizes obtained from one supplier or number of suppliers, it can be concluded that it is possible to predict coke apparent density and open porosity. These properties are important for adjusting anode formulation.
Density Predicted
3-Methodology:
Sound Signal
PredictiveModel
PredictedDensity
PredictedPorosity
Extraction Of Sound Signal
CharacteristicsCluster By
Size
Cluster By Supplier
Coke Sample
Extraction Of Physical
Measurements
Scott Volumetric SD, VBD, HR
Apparent Density
Open Porosity
Helium DensityHelium Pycnometer
Water Pycnometer
FilteringFrequency
Domain Decomposition
Table 3
Matlab
DSP
Figure 2. Schematic view of the methodology.
Figure 1. Laboratory setup.Table 1. Samples used for experiments.
Figure 3. Coke particles score plot of single and mixture of coke particle size and supplier.
Size 4
Size 3
Size 2
Size 1
Mixture of coke particlesize and supplier
Bigger size of particlesSmaller size of particles
Figure 4. Predicted vs Observed of apparent density of single and mixture of coke particle sizes and suppliers.De
nsity
Obs
erve
dDensity Predicted
Apparent density g/cm³
Figure 5. Predicted vs Observed of open porosity of single and mixture of coke particle size and supplier.Porosity Predicted
Poro
sity
Obse
rved
Elias IshakUniversité Laval
D. Picard - ULC. Duchesne - UL
D. Kocaefe - UQAC
27Spéciation d’impuretés dans le coke de pétrole par spectrométrie d’absorption des rayons X (XANEX et EXAFS)
Impurity speciation in petroleum cokes using XAS Techniques XANES and EXAFS
The aluminium industry is facing challenges with quality changes in the calcined petroleum cokes available, where changes include higher metal and sulfur impurities. Sulfur in calcined petroleum cokes used in aluminum industry is present in the percent range and will result in formation of environmentally unfriendly gases COS and SO
2 during production of Al. Metals are known catalyzers for the reactions between carbon, air and CO
2 and are
undesirable in the coke, but the high-sulfur cokes also tend to have a very low CO2 reactivity. The mechanism
behind these effects are still unknown. Possible reactions and mechanisms where sulfur is participating during cell operation need to consider the chemical form of sulfur in the structure. This can be investigated by X-ray Ab-sorption Spectroscopy (XAS) techniques like X-ray Absorption Near Edge Structure (XANES) and Extended X-ray Absorption Fine Structure (EXAFS). In this study, sulfur speciation was discovered to vary significantly between industrial cokes and mainly present as organic sulfur and a specie containing S-S bonds. Ni, V and Fe is most likely present as hexagonal metal sulfides.
L’industrie de l’aluminium fait face à plusieurs défis dus aux changements de qualité du coke de pétrole calciné disponible, qui est maintenant plus concentré en impuretés métalliques et en soufre. Le coke de pétrole utilisé pour la production de l’aluminium contient du soufre à l’échelle du pourcent, ce qui engendre la formation de gaz nocifs pour l’environnement tels que le COS et le SO
2. Les métaux, connus comme des catalyseurs pour
les réactions entre le carbone, l’air et le CO2, sont également non-désirés dans le coke, mais les cokes à haute
teneur en soufre tendent à avoir une très basse réactivité au CO2. Le mécanisme derrière ces effets est toujours
inconnu. Les réactions et mécanismes possibles mettant en jeu le soufre durant l’électrolyse doivent considérer la forme chimique du soufre dans la structure. Cela peut être étudié par spectroscopie d’absorption des rayons X, notamment par les techniques XANES et EXAFS. Dans cette étude, il a été observé que la spéciation du soufre varie significativement entre les cokes industriels et qu’il est majoritairement présent sous forme de soufre organique et d’une espèce contenant des liens S-S. Les métaux Ni, V et Fe sont principalement présents sous la forme de sulfures métalliques de structure hexagonale.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Gøril JAHRSENGENE1, Arne Petter RATVIK2, Richard G. HAVERKAMP3 and Ann Mari SVENSSON1
1 NTNU Norwegian University of Science and Technology, Department of Materials Science and Engineering, Trondheim, Norway
2 SINTEF Industry, Trondheim, Norway3 Massey University, School of Engineering and Advanced Technology, Palmerston North, New Zealand
Impurity Speciation in Petroleum Cokes using XAS Techniques XANES
and EXAFS
Sulfur in calcined petroleum cokes used in aluminium industry is present in the percentrange, and will result in formation of environmentally unfriendly gases COS and SO2during production of Al. The industry is facing challenges with quality changes in thecalcined petroleum cokes available, where changes include higher metal and sulfurimpurities [1]. Metals are known catalysers for the reactions between carbon, air andCO2 and are undesirable in the coke [2], but the high-sulfur cokes also tend to have avery low CO2 reactivity. The mechanism behind this effect is still unknown. Possiblereactions and mechanisms where sulfur is participating during cell operation need totake into account the chemical form of sulfur in the structure. This can be investigated byX-ray Absorption Spectroscopy (XAS) techniques like X-ray Absorption Near EdgeStructure (XANES) and Extended X-ray Absorption Fine Structure (EXAFS).
Background
The normalised XANES spectra for five industrial petroleum cokes with varying sulfurlevels show clear variations in sulfur speciation independent of the total sulfur in thecoke [3]. The differences are much larger in this selection of industrial cokes than incokes investigated previously [4].
Sulfur K-edge XANES
Energy [eV]
2465 2470 2475 2480 2485 2490
Nor
mal
ised
Abs
orpt
ion
Coke A - 1.42 % S
Coke B - 3.56 % S
Coke C - 5.54 % S
Coke D - 3.82 % S
Coke E - 4.42 % S
2473.12471.8
Energy [eV]
2465 2470 2475 2480 2485 2490
Nor
mal
ised
Abs
orpt
ion
Iron sulfide (FeS)
Sulfur (S8)
L-methionine
Dibenzothiophene
Phenothiazine
2473.0
2475.0
2472.6
2471.8
2470.0
2472.82473.6
2474.4
(a) (b)
Fig. 1. Normalised (to step-edge = 1) sulfur K-edge XANES for (a) five industrial cokes with varying sulfurlevels (in wt%) and (b) reference samples.
XANES is a powerful tool for fingerprinting of chemical species, and relative amountscan be determined by linear combination fitting (LCF).• The peak at 2473.1 eV, the main peak in all five cokes, is likely an organic sulfur
specie, while the peak 2471.8 eV, most visible for coke E, is located at the sameposition as the peak for elemental sulfur, S8.
• LCF resulted in approx. 20 % S8 in cokes A, B and C, 41 % in D and 53 % in E.• The organic specie is likely a mixture of sulfur as a part of a five or six-membered
ring, i.e. dibenzothiophene and phenothiazine
Energy [eV]
2465 2470 2475 2480 2485 2490
Nor
mal
ised
TE
Y
0.0
0.5
1.0
1.5
2.0
2.5Coke C LCF Coke C Phenothiazine Dibenzothiophene Sulfur (S8)
Energy [eV]
2465 2470 2475 2480 2485 2490
Nor
mal
ised
TE
Y
0.0
0.5
1.0
1.5
2.0
2.5Coke ELCF Coke ECoke C Sulfur (S8)
(a) (b)
Fig. 2. The LCF for (a) coke C fitted with two organic references and S8 and (b) coke E fitted with coke CandS8.
The amount S8 (or S-S) detected increased with increasing pore area of pores withdiameter < 50 nm obtained by N2 adsorption and BJH calculations.
Sulfur Speciation and Porosity
Sulfur identified as S8 [wt%]
0.0 0.5 1.0 1.5 2.0 2.5
Pore
are
a fo
r por
e di
amet
er <
50
nm [m
2 /g]
4.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
AB
CD
E
Fig. 3. The amount of sulfur evaluated as S8 with respect to the pore area of small pores.
Metal sulfides cannot be detected with sulfur K-edge in the cokes due to the lowconcentration relative to the total sulfur content and the position of the peaks.XANES/EXAFS investigation and fingerprinting of Ni, V, and Fe shows that these metalsare present mainly as sulfides, possibly in combination with organic metal porphyrins.
Metal K-edge XANES and EXAFS
Fig. 4. Nickel XAS on coke D. (a) Experimental XANES spectrum of coke, porphyrin and sulfide.Experimental EXAFS spectrum and FEFF model fit for hexagonal NiS in (b) k-space and (c) R-space.
Energy [eV]
8300 8350 8400 8450 8500
Nor
mal
ised
Abs
orpt
ion
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Coke D Nickel Porphyrin Ni3S2
Wavenumber [Å-1]0 2 4 6 8
k2 χ
(k) [
Å-2 ]
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
Coke D Fit NiS
Radial distance [Å]0 1 2 3 4 5 6
|χ(R
)| [Å
-3]
0.0
0.5
1.0
1.5
Coke D Fit NiS
(a) (b)
(c)
• Sulfur XANES revealed that the sulfur speciation in some industrial cokes aresignificantly different than in others.
• The organic sulfur specie seemed to vary little between the cokes, and is likely in adibenzothiophene configuration.
• The difference between cokes was the amount of sulfur identified as S8. This sulfurtype may correlate to the pore size distribution, as cokes with a larger amount ofsmall pores had more of this type of sulfur.
• The sulfur speciation may be important when considering the CO2 reactivity.• Using both XANES and EXAFS, Ni, V and Fe was found to most likely be present as
dispersed, hexagonal metal sulfides in the cokes.
Conclusions
Financial support from the Norwegian Research Council and the partners HydroAluminium, Alcoa, Elkem Carbon, and Skamol through the project “Reactivity of Carbonand Refractory Materials used in Metal Production Technology” (CaRMa) isacknowledged. The sulfur XAS measurements were done at the Canadian Light Source(CLS), and the metals were investigated at the Australian Synchrotron.
Acknowledgements
[1] L. Edwards (2015): JOM, vol. 67, pp. 308–21.[2] G.J. Houston and H.A. Øye (1985): Aluminium, vol. 61, pp. 251–54, 346–49, 426–28.[3] G. Jahrsengene, H.C. Wells, S. Rørvik, A.P. Ratvik, R.G. Haverkamp, A.M. Svensson (2018).Metall. Mater. Trans. B. vol. 49, pp. 1434-1443.[4] S.J. Hay, J. Metson, and M.M. Hyland (2004): Ind. Eng. Chem. Res., vol. 43, pp. 1690–1700.
References
Gøril JahrsengeneNorwegian University of Science and Technology
A. P. Rarvik - SINTEFR. G. Haverkamp - Massey University Ann Mari Svensson - NTNU
Prix Award
2828 Mise en forme des anodes - Quantification de l’effet de la vibro-compactionAnode manufacturing for aluminum smelters: optimization of vibro-compaction process
La qualité des anodes utilisées dans le procédé de production d’aluminium constitue un facteur important au niveau de la minimisation des coûts et de la maximisation de l’efficacité énergique. L’amélioration des carac-téristiques mécaniques des anodes prolongera la vie de celles-ci dans la cellule d’électrolyse et améliorera sa performance. En effet, le but de ce projet est d’améliorer les caractéristiques mécaniques des anodes (plus denses et plus homogènes). Pour atteindre cet objectif, le processus de vibro-compaction sera optimisé. Pour pouvoir améliorer ce processus, le comportement de la pâte d’anode durant la compaction doit être investiguée. Des essais de compaction cycliques et monotones ont été exécutés sur une pâte carbonée pour caractériser son comportement et comprendre les mécanismes qui entrainent sa densification. Les différentes mesures prises durant les essais ont permis : (1) de comprendre l’effet de compacter par cycle par rapport à la compaction monotone, (2) d’identifier l’effet de la vitesse de chargement sur la compaction et (3) d’identifier l’évolution des propriétés élastiques durant la compaction cyclique. Les résultats montrent qu’il est plus efficace de compacter ce matériau sous des charges cycliques, et que diminuer la vitesse de chargement et l’amplitude de déformation diminue la contrainte engendrée par la compaction.
The quality of anodes used in aluminum smelters is a very important factor in terms of minimizing costs and maximizing the energetic efficiency. Improving the thermal, electrical and mechanical characteristics of the an-odes will extend their life in the electrolysis cell and improve their performance. The aim of this project is to improve the mechanical characteristics of the anodes (denser, more homogeneous). To reach this goal, the vibro-compacting process will be optimized. To be able to improve this process, the behavior of the anode paste during compaction must be investigated. Cyclic and monotonic compaction tests were performed on a carbonaceous paste to characterize its behavior and understand the mechanisms that lead to its densification. The various measurements taken during the tests allowed: (1) to understand the effect of cyclic compaction compared to the monotonic one, (2) to identify the effect of the loading rate on the compaction and (3) to identify the evolution of elastic properties during cyclic compaction. The result of this work showed that it is more efficient to compact this material under cyclic loads, and that decreasing the loading rate and the deformation amplitude decreases the stress generated by the compaction.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Zahraa Kansoun1, Donald Picard2, Hicham Chaouki2, Julien Lauzon-Gauthier3, Houshang Alamdari4 and Mario Fafard1,2
1 Department of Civil and Water Engineering, Laval University2 Aluminium Research Centre – REGAL, Department of Civil and water Engineering Laval University
3 Alcoa Corporation, Aluminium Center of Excellence, Aluminerie de Deschambault4 Department of Mining, Metallurgical and Material Engineering, Université Laval
Anode Manufacturing for Aluminum Smelters: Optimization of the Vibro-
Compaction.Mise en Forme des anodes - Quantification
de l’Effet de la Vibro-Compaction.
Optimize the vibro-compaction process
↑ Density & homogeneity
↑Pot life
↑Mechanical properties
↓ Electrical resistivity
↓ Reactivity with CO2 and air
↓ GHG emissions
Experimental characterization of green
anode paste (GAP) behavior
Model the rheological behavior of the GAP
Model the VC process and optimized it through numerical
simulations
Test 2: Monotone Tests at different strain rates
Set-up and Material
Material: Ramming paste (CTP binder)
To work at room temperature
Figure 1: Flexible mold usedfor experimental work.
conclusion
I. The cycling allows the paste densification at lower stresses than the monotonous compaction (Figure 3)
II. To decrease the stress during the compaction:
• Decrease the strain amplitude (Figure 3)
• Decrease the strain rate (Figure 4)
Figure 3: Stress response as a function of the density for a monotone test and for the encircled
peaks of the cyclic tests.
Figure 4: Typical behavior of the axial stress as a function of the axial strain during a cyclic test.
• Model the different parts of the vibrocompactor and couple it with the rheological model to the ramming paste
• Identify the influence of different VC parameters on the densification process
• Where the deformation is recoverable? Before a certain elasticity limit ?
• What kind of permanent deformation? Viscous, visco-plastic, plasticity with hardening, etc. ?
Test 1: Quasi-static cyclic tests
Figure 2: Load cell displacement.
Figure 4: Stress response as a function of the density for the different strain rates.
Zahraa Kansoun Université Laval
D. Picard - UL H. Chaouki - UL
J. Lauzon-Gauthier - AlcoaH. Alamdari - UL
M. Fafard - UL
29Modèle multi-échelle FEM – DEM pour les réactions de gazéification des anodes de carbone
Multi scale FEM–DEM model for gasification reactions of carbon anodes
A comprehensive FEM–DEM Model has been developed to simulate the carbon anode gasification process. The methodology is based on a Eulerian–Lagrangian concept. A Eulerian method for gas phase, including inside and outside of anode particle. In addition, a Lagrangian method for porous solid particle phase. Each particle is individually tracked and associated with multiple physical properties (porosity, size, density, composition, and temperature). Particle shrinkage effect is considered in this model. During the anode gasification process, heat and mass transfer, particle shrinkage, and non-catalytic chemical reactions are all considered. The dynamic concentration and temperature profile of the reactant and product gases as well as the solid conversion are performed in all the nodes (between and inside the particle) as a result. In addition, structural changes of the carbon anode such as permeability and porosity during the gasification were obtained. In order to validate the model, experimental tests were performed using a bed of particles. Meanwhile, the calculated results compare well with the experimental data. This indicates the developed FEM–DEM model can applied for the multi-scale modeling of anode gasification.
Un modèle complet FEM-DEM a été mis au point pour simuler le processus de gazéification des anodes de carbone. La méthodologie est basée sur un concept Eulerian-Lagrangian. Une méthode eulérienne pour la phase gazeuse, y compris à l'intérieur et à l'extérieur de la particule d'anode. En outre, une méthode lagrangienne pour la phase particulaire solide poreuse. Chaque particule est suivie individuellement et associée à de multiples propriétés physiques (porosité, taille, densité, composition et température). L'effet de retrait de particules est considéré dans ce modèle. Au cours du processus de gazéification des anodes, les transferts de chaleur et de masse, le retrait des particules et les réactions chimiques non catalytiques sont tous pris en compte. La concen-tration dynamique et le profil de température des gaz de réactif et de produit, ainsi que la conversion du solide sont effectués dans tous les nœuds (entre et à l'intérieur de la particule). De plus, des modifications structurelles de l'anode en carbone telles que la perméabilité et la porosité au cours de la gazéification ont été obtenues. Afin de valider le modèle, des tests expérimentaux ont été réalisés sur un lit de particules. Pendant ce temps, les résultats calculés se comparent bien avec les données expérimentales. Cela indique que le modèle FEM – DEM développé peut s'appliquer à la modélisation multi-échelle de la gazéification des anodes.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Mohammad Kavand Université Laval
R. Mollaabbasi - ULJ. Tessier - AlcoaF. Larachi - ULD. Picard - ULH. Alamdari - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
M. Kavand1, R. Mollaabbasi1, J. Tessier2, F. Larachi3, D. Picard4, H. Alamdari11 Department of Mining, Metallurgical and Material Engineering, Université Laval2Alcoa Corporation, Aluminum Center of Excellence, Aluminerie de Deschambault
3 Department of Chemical Engineering, Université Laval4 Alcoa Industrial Research Chair MACE3 and Aluminium Research Centre – REGAL, Laval University
Multi Scale FEM–DEM Model for Gasification Reactions of Carbon Anodes
Modèle multi-échelle FEM – DEM pour les réactions de gazéification des anodes de carbone
Avant de débuter votre travail, vérifiez que le bandeau inférieur apparaisse bien au bas de l’affiche.
Before starting your work, check thatthe bottom banner appears effectivelyat the bottom of the poster.
Questions : [email protected]
1- U.S. Geological Survey, 2008, Copper statistics, in Kelly, T.D., and Matos, G.R., comps., Historical statistics for mineral and material commodities in the United States Available: http://pubs.usgs.gov/ds/2005/ 140 /.2- B.A. Sadler, and S.H. Algie, Porosimetric study of sub-surface carboxy oxidation in anodes. in Proceedings of the 120th TMS Annual Meeting, New Orleans, LA, USA, 17-21 February 1991, Publ by Minerals, Metals & Materials Soc (TMS).3- J.C. Gottifredi and E.E. Gonzo, An approximate expression for predicting concentration and temperature profiles inside a catalyst pellet, Chemical Engineering Science. Vol. 51, No 5, (1996), 835–837.
6. References:
Introduction
2. Hypotheses and objectives 4. Results (continued)
5. Conclusions
1. Introduction
Objectives
Modeling the gasification reaction of anode with CO2
Identifying the limiting phenomena
Optimizing the model parameters
Simplifying the anode structure by considering
a bed of particles
Electrolysis reaction in the Hall- Héroult process 2 Al2O3(diss)+3C(s)→ 4 Al(m)+3CO2(g)
Air reactivityChemical reaction,
C + O2 →CO2Mass transport of O2 (Diffusion)
CO2 reactivityChemical reaction, C + CO2 →2 CO
Mass transport of CO2 (Diffusion + Convection)
Problem statement: Carbon dust in Hall Héroult cells is anoperational problem that is linked to reaction of the anodes withcarbon dioxide. The general understanding is that the reaction takesplace at the surface and the inside of anodes, below the bath surfaceand rather more toward the sides of the anodes than the bottom. Thecarbon-CO2 reaction becomes significant at temperatures above 800°C.
3. Methodology
The anode overconsumption is controlled by chemical reaction and mass transport phenomenon (diffusion and convection).
Solid phase micro-scale
A comprehensive FEM–DEM numerical model has been developed to simulate the anode gasification process in a fixed bed reactor Experimental data for fixed bed reactor was obtained. The integrated model further considers particle collision, gas–particle flow in fixed beds, heat and mass transfer, particle shrinkage, as well as chemical reactions. The model has also been validated by comparing the calculated results with the experimental data. Future works: writing papers (single particle, multi scale fixed bed reactor)
Inlet gas
Fluid phasemacro-scale
4. Results
Particle shrinks
Distribution of the reactant and product gas concentration in the bed and particles
Temperature distribution Particle tracking
Particle shrinkage;Average of the porosity change-Bed height
Figure 1: schematic of anode reactivity [3]
Figure 2: Reaction-diffusion definition
Figure 3: Particle-bed shrinkage
Figure 4: Distribution of Temperature in the fixed bedreactor a) 2 l/min b)1 l/min C) 0.5 l/min
Figure 5: Particle tracking in the fixed bed reactor a) 0.5 l/min,r=1mm b) 0.5 l/min, r=0.5 mm C) 1.0 l/min r=1mm d) 2 l/minr=1mm
Figure 6: Distribution of the porosity
a b c a b c d
The calculation of the trajectoryof a moving Lagrangian elementwith the knowledge of anEulerian flow field
Figure 7: Particle radius shrinkage
Large pores: could affect mass transport (permeability, convection) Small pores: are important in providing large surface area
Figure 8: Distribution ofCO2 concentration in thebed and particles
Figure 9: Distribution ofCO concentration in thebed and particles
Model flowchart
Multi scale modelAnode particle
BinderMatrix
Prix Award
3030Optimisation de la conductivité électrique et des propriétés mécaniques des alliages conducteurs de la série 6xxx avec divers rapports Mg / SiOptimization of electrical conductivity and mechanical properties in 6xxx conductor alloys with various Mg/Si ratios
Dans cette étude, l’effet des rapports Mg/Si (de 0,85 à 2) sur la conductivité électrique (CE) et les propriétés mécaniques sont étudiés pour les alliages 6201. Tout en utilisant un microscope optique, électronique à balayage et à transmission, une calorimétrie différentielle à balayage, des tests de traction, de microdureté et de CE. Il est constaté que la cinétique de précipitation et le durcissement par précipitation étaient améliorés avec la réduction du rapport Mg/Si (ou avec l'augmentation en teneur de Si). Puisque les alliages 6201 doivent avoir une CE mini-male (52,5 % IACS), tous les alliages ont été soumis aux conditions de survieillissement. Avec une CE supérieure à 52,5 % IACS, il a été observé que l'alliage avec un rapport Mg/Si de 1 présentait la plus haute résistance à la traction ultime (UTS) (346 MPa), tandis que l’UTS des alliages d’un rapport de 2 et 0,85 étaient respectivement de 306 et 318 MPa. Bien que l'alliage du rapport 0,85 ait montré le pic UTS le plus élevé, une augmentation du temps de survieillissement est requis pour extraire le Si de la matrice pour atteindre le minimum de CE, provo-quant simultanément la détérioration des précipitations et le durcissement par solution solide.
In this research, the effect of the various Mg/Si ratios (ranging from 0.85 to 2) on the electrical conductivity (EC) and mechanical properties was studied in 6201 aluminum conductor alloys. Investigations were carried out using optical microscope, scanning electron microscope, transmission electron microscope, differential scanning calorimetry, microhardness and tensile tests, and electrical conductivity test. It was found that the precipitation kinetics and precipitation hardening were enhanced with reducing Mg/Si ratio (or with increasing Si level). How-ever, given the fact that 6201 aluminum conductor alloys ought to meet the minimum EC (52.5 %IACS), all alloys were subjected to overaging condition. In EC above 52.5 %IACS, it was observed that the alloy with Mg/Si ratio of one exhibited the highest ultimate tensile strength (UTS) (346MPa) while the UTS levels for the alloys with Mg/Si ratio of 2 and 0.85 were 306 and 318 MPa, respectively. Although the alloy with Mg/Si ratio of 0.85 showed the highest peak UTS among alloys, more overaging time is required to extract Si out of the matrix to reach the minimum EC, simultaneously causing the deterioration in precipitation and solute strengthening in this alloy.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Siamak Nikzad Khangholi 1, X. Grant Chen1, Mousa Javidani 2
1 Research Chair in Metallurgy of Aluminum Transformation, University of Québec at Chicoutimi, Saguenay, Québec 2 Rio Tinto, Arvida Research and Development Center (ARDC), Jonquière, Quebec
Optimization of electrical conductivity and mechanical properties in 6xxx conductor alloys with
various Mg/Si ratios
Optimisation de la conductivité électrique (CE) et des propriétés mécaniques des alliages conducteurs de la
série 6xxx avec divers rapports Mg / Si
1. INTRODUCTION
Recently, the demand for aluminum conductors has been increasingly grown sincetheir density and cost are lower in relative to copper conductors.
Aluminum conductors are divided into two categories:1. aluminum conductor steel reinforced (ACSR) (AA1350)
2. all-aluminum alloy conductor (AAAC) (AA6201 and AA6101)
AAAC shows better mechanical properties due to precipitation hardening.
In AAAC, excessive silicon in 6xxx alloys promotes the precipitation hardening.
Steel reinforced
Aluminum conductors
2. OBJECTIVES
1. To enhance the ultimate tensile strength (UTS) and electrical conductivity (EC) usingexcessive Si relative to the Mg/Si ratio of β” in AA6201
2. To determine the contribution of each strengthening mechanism in Si-excessive alloys
3. To clarify the mechanism for the optimization of electrical conductivity (EC) and ultimatetensile strength (UTS) in Si-excessive 6201 alloys
3. EXPERIMENTAL PROCEDURES
Alloydesignation
Compositions (wt.%) Mg/Si (in at.%)Mg Si Fe Mn Al
Si4 0.67 0.4 0.09 <0.003 Bal. 1.94Si5 0.68 0.53 0.10 <0.003 Bal. 1.47Si6 0.66 0.66 0.13 <0.003 Bal. 1.15Si9 0.68 0.89 0.10 <0.003 Bal. 0.86
1. CastingTable1. The chemical composition for various AA6201
2. Wire production process 3. Microstructural and mechanical characterizations
Scalping the cast alloys, homogenization at 560°C for 6 hour, followed by hot rolling process at 350-480°C
Cutting and machining the hot-rolled samples into the square rods
Solution treatment at 540°C for 2hr, followed by wire drawing (area reduction: 50%)
Applying the aging treatment at 180°C at different aging time
Optical microscope
Transmission electron microscope
tensile test
Electrical conductivity test using SIGMASCOPE
differential scanning calorimetry
4. RESULTS AND DISCUSSIONS
050
100150200250300350400450
Si4-3hr Si6-5hr Si9-34hr
Gra
in S
ize
(µm
)
Alloys
5. Conclusions1. Si6 alloy (Mg/Si = 1) showed the better
optimization of EC and UTS due to precipitation,solute, and strain strengthening mechanisms.
2. In Si9-34hr alloy (Mg/Si = 0.86), longer agingtime is required to extract Si out of the matrix,causing the overgaing and lower contribution ofprecipitation hardening, but strain hardeningcontribution is high.
3. Si4 alloy (Mg/Si = 2) exhibited the lower UTSabove 52.5%IACS.
a bc
Si4-3hr a Si6-5hr b Si9-34hr c
Figure 1. (a) DSC results of the as-drawn samples (b) zoomed-in β” precipitation peaksFigure 2. (a) ultimate tensile strength (UTS) and (b) electrical conductivity as a function of aging time (c) the curve of UTS-EC above
the minimum EC (52.5%IACS) and UTS (305Mpa) based on ASTM B398
Figure 3. TEM micrographs for (a) Si4-3hr (b) Si6-5hr (c) Si9-34h alloys
Figure 4. quantitative analysis for density number of precipitates in Si4-3hr, Si6-5hr, Si9-34h alloys
Figure 7. the contribution of each strengthening mechanism for Si4-3hr, Si6-5hr, Si9-34h alloys
Figure 6. The strain hardening contribution from cold wire drawing in all alloys
Figure 5. quantitative analysis for grain sizes for Si4-3hr, Si6-5hr, Si9-34h alloys along the
drawing direction
a
UT
S >
305
MPa
(AST
M B
398)
b
EC
> 5
2.5%
IAC
S(A
STM
B39
8)
4920 5177
3171
0
1000
2000
3000
4000
5000
6000
7000
Si4-3hr Si6-5hr Si9-34hr
Ave.
Den
sity
(µm
-3)
Alloys
79.0 83.0
106.0
0.0
20.0
40.0
60.0
80.0
100.0
120.0
Si4 Si6 Si9
Stra
in h
arde
ing
from
dra
win
g (M
Pa)
Alloys
6. Future work
1. Investigation on the effect of pre-
natural aging on mechanical and
electrical properties.
2. Study on the effect of Cu, Ag, and Sr
on mechanical and electrical
properties in 6xxx conductors.
Siamak Nikzad Khangholi Université du Québec
à Chicoutimi
X. G. Chen - UQAC M. Javidani - Rio Tinto
31Optimisation de la mise en forme des anodes des cuves Hall-Héroult
Optimization of the anode forming process for Hall-Héroult cells
The green anode vibrocompaction forming process has a significant influence on the anode’s quality and proper-ties. The forming process is not optimal, however, as density gradients are present in zones where the paste is subjected to major distortions, primarily around the stub holes, the grooves and the different corners and edges of the anode. These gradients affect the anode’s performance in the cell negatively and can cause cracks to appear during the baking process or in operation. The objective of this project is to reduce the negative effects of these density gradients and improve the properties of the anode using numerical simulation tools to improve the density uniformity of the green anode. A numerical model of the forming process will be developed to study the effects of different parameters on the anode’s density distribution. The anode’s geometry, such as the depth and width of the grooves and the addition of chamfers or fillets to the corners and edges, the mould’s orientation, the initial paste distribution, lubrication and temperature of the mould will be examined. An optimization of the anode and the mould’s geometry will be performed based on the results obtained from the various stages of the numerical investigation.
Le procédé de mise en forme des anodes vertes par vibrocompactage influence directement leur qualité ainsi que leurs propriétés. Le procédé actuel n’est toutefois pas optimal puisqu’on retrouve des gradients de densité importants dans les zones de grandes distorsions, notamment autour des tourillons, des rainures anodiques, des coins et des différentes arêtes. Ces gradients, en plus d’affecter les performances de l’anode dans la cuve, peuvent causer l’apparition de fissures lors de la cuisson ou même en opération. L’objectif du projet est de réduire les gradients de densité à l’aide d’outils de simulation numérique dans le but d’uniformiser la densité de l’anode verte. Un modèle numérique du procédé de mise en forme sera développé et permettra d’étudier l’effet de différents paramètres sur la distribution de la densité dans l’anode. La géométrie de l’anode (notamment la forme des rainures et l’ajout de chanfreins ou d’arrondis aux coins et aux arêtes), l’orientation dans le moule, la distribution initiale de la pâte dans le moule, la lubrification ainsi que la température du moule seront examinées. Une optimisation de la géométrie de l’anode et de son moule sera finalement effectuée en se basant sur les résultats obtenus lors des différentes étapes d’investigation numérique.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Optimization of the anode forming process for Hall-
Héroult cellsOptimisation de la mise en forme
des anodes des cuves Hall-Héroult
O. Lacroix1, H. Chaouki1, J. Tessier3, J. Lauzon-Gauthier3, H. Alamdari2, M. Fafard1
1 Department of Civil and Water Engineering, Laval University2 Department of Mineral, Metallurgical and Materials Engineering Engineering, Laval University
3 Alcoa Corporation, Aluminium Center of Excellence, Aluminerie de Deschambault
Ongoing workStub hole compaction test
Stub hole compaction test setup
Slot compaction test setup
Slot compaction test
Motivations & Problematic• Anodes are formed by compaction of vibrocompaction
and their quality plays a crucial role in the efficiency of the electrolysis process.
• Presence of density gradients around the stub holes, the slots and the different edges and corners of the anode.
• These gradients can cause cracks in the anode duringthe baking process and reduce the anode’sperformance in the cell.
Cracks in a baked carbon anode [1]X-ray computed tomography
of a baked anode
Objectives• Reduce the density gradients inside the
anode to improve the anode’s performance and life expectancy as well as the efficiencyof the electrolysis process.
• Understand the effect of the anode geometry and process parameters on the density distribution in the anode usingnumerical modelling. Density distribution in a small scale anode
[1] M. B. Boubaker, et al., Non-destructive testing of baked anodes based on modal analysis and principal component analysis, Light Metals 2017, 1289-1298.
References
Methodology
Generic anode geometry
• Finite element and/or smooth particle hydrodynamicsmodel of the anode forming process in Abaqus.
• Compaction of the anode paste is used instead of vibrocompaction to simplify the model. Similardensity gradients are achieved with both methods.
• Develop a behaviour law for the anode paste basedon experiments and prior literature.
• Incorporate the effect of the temperature in the behaviour in the anode paste behaviour law to capture the effect of the mould’s temperature on the surrounding paste.
A
B• Study the effect of the anode’s geometry on the density distribution numerically.
C• Investigate the effect of different forming conditions and parameters on the density distribution with the numerical model.
D• Corroborate the results obtained with experimental compaction tests.• Optimization of the anode’s geometry and forming conditions based on the
results obtained during the investigation phase.
Mould temperatureand lubrification Inverted anode forming Initial distribution of the
anode paste
Chamfers and rounds at the external edges and corners
Slot dimensions, position and geometry Draft angles
Conclusions & future work
The authors would like to thank Alcoa Corporation as well as the Natural and Engineering ResearchCouncil of Canada (NSERC) for their technical and financial support.
Acknowledgments
• The density gradients can be reproduced by the finite element model.• The differences in density on both sides of the slot could be due to parameters not yet
taken into account by the model, such as the mould’s temperature or a different friction coefficient between the paste and the mould.
• The possibility of using an easier to implement constitutive behaviour law for the anode paste will be explored.
Olivier Lacroix Université Laval
H. Chaouki - ULJ. Tessier - Alcoa J. Lauzon-Gauthier - AlcoaH. Alamdari - ULM. Fafard - UL
3232Synthèse et caractérisation du brai biologique produit à partir du bio-huileSynthesis and characterization of bio-pitch produced from bio-oil
Le brai de houille est un matériau fossile actuellement utilisé comme liant dans la fabrication d’anode en car-bone. Outre les nombreux avantages du brai d’huile, il est connu qu'il contient des hydrocarbures aromatiques polycycliques (HAP), qui sont cancérigènes pour l'homme et qui endommagent l'environnement. Ma précédente étude montre que le brai d’huile pourrait être un potentiel liant pour la fabrication d’anode en carbone. Le brai biologique possède les propriétés rhéologiques requises pour remplir les pores de coke et pour mouiller les agrégats de coke. Cependant, les propriétés du brai biologique diffèrent de manière significative en fonction de leur origine et des conditions de pyrolyse. Afin de déterminer les propriétés appropriées du bio-brai en tant que liant pour la fabrication d'anode, il est important de comprendre ses propriétés chimiques et physiques. Cette étude se concentre sur les bio-brais produits à partir de la même bio-huile en utilisant différentes conditions de pyrolyse sous vide. Les propriétés physiques et chimiques des bio-brais ont été caractérisées, telles que la densité, le point de ramollissement, la valeur de cokéfaction, la quinoléine insoluble, la teneur en HAP, le poids moléculaire, la viscosité, la composition en éléments et les structures chimiques. Le liquide condensé de la bio-huile au cours de la pyrolyse sous vide pour obtenir les bio-brais a également été étudié.
Coal tar pitch is a fossil material currently used as binder in carbon anode manufacture. In addition to the many benefits of coal tar pitch, it is known to contain polycyclic aromatic hydrocarbons (PAHs), which are carcinogenic for humans and damaging for the environment. My previous study showed that bio-pitches, produced from bio-oils, could be a potential alternative binder for carbon anode manufacture. Bio-pitch has the required rheological properties to fill coke pores and wet coke aggregates. However, bio-pitch properties are significantly different depending on their origins and pyrolysis conditions. In order to determine the suitable properties of bio-pitch as binder for anode making, both chemical and physical properties are important to understand. This study focuses on bio-pitches produced from the same bio-oil using different vacuum pyrolysis conditions. Both physical and chemical properties of bio-pitches were characterized, such as density, softening point, coking value, quinoline insoluble, PAH content, molecular weight, viscosity, elements composition and chemical structures. The con-densed liquid from bio-oil during the vacuum pyrolysis to obtain the bio-pitches was also investigated.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
6. Synthesis process of BP from BO
Intéraction du brai écologique et coke (Synthèse et caractérisation du brai biologique
produit à partir du bio-huile)
The technical and financial support of Alcoa Corporation (Alcoa) as well as the financial support of the Natural Sciences and Engineering Research Council of Canada (NSERC), Université Laval, and the Fonds de recherche nature et technologies Québec are greatly appreciated.
• Coal tar pitch (CTP) is a fossil material currently used as binder in carbon anode manufacture. However, it contains polycyclic aromatic hydrocarbons (PAHs), which are carcinogenic for humans.
• Bio-pitch (BP), produced from bio-oils (BO), has the required rheological properties to fill coke pores and wet coke aggregates, and could be a potential environmental-friendly binder for carbon anode manufacture.
• Bio-pitch properties are significantly different depending on pyrolysis conditions. In order to determine the suitable properties of bio-pitch as binder for anode making, both chemical and physical properties are important to understand.
• This study focuses on bio-pitches synthesis from the same bio-oil using different vacuum pyrolysis conditions. Both physical and chemical properties of bio-pitches were characterized.
1. Introduction
1. The yield of bio-pitch reduced associating the releasing of volatile matter content as increasing in reaction temperature and time and decreasing in vacuum pressure in range of this study.
2. Bio-pitches have increasing aromaticity and decreasing oxygen-containing functionality chemical groups (COOH/OH) with the increasing holding time, this could influence to chemistry of pitch/coke interactions.
3. The properties of bio-pitch are interrelated and mainly determined by molecular structure identified by various analysis techniques. 1HNMR and ESI-TOF mass were used to characterize the bio-pitch, which confirmed the presence of multiple functional groups in the polymer, such as aromatic, alkenyl and alkoxyl groups.
4. The molecular weight of the produced bio-pitches was determined, scale-up of molecular suggesting the polymerization and condensation reaction during synthesis process.
9. Acknowledgements
2. Materials and Methodology
5. Mass analysis
ESI-TOF mass spectrum of bio-pitch (BP-2).
Ø Study wettability of bio-pitch on coke.ØModify bio-pitch properties by changing pyrolysis conditions and increase QI content by adding additives
such as carbon black/bio-char in order to optimise bio-pitch properties.
8. Future work
Y. Lu1,2, D. Li3, D. Picard2, R.Mollaabbasi2, J. Tessier4, T. Ollevier3, H.Alamdari1,2
1 Department of Mining, Metallurgical and Materials Engineering, Université Laval, Canada 2 NSERC/Alcoa Industrial Research Chair MACE3 and Aluminium Research Centre – REGAL, Université Laval, Canada
3 Department of Chemistry, Université Laval, Canada 4 Alcoa Corporation, Aluminum Center of Excellence, Aluminerie de Deschambault,Deschambault-Grondines, QC, Canada
3. Pitch properties
7. Conclusions and Perspective
Measured bio-pitch properties from different pyrolysis conditions
(a) 1H NMR, and (b) FTIR of bio-oil and bio-pitches from three different holding time.
Experimental conditions BP-1 BP-2 BP-3 BP-4 BP-5 BP-6 BP-7 BP-8 BP-9 BP-10
Final temperature (°C) 160 180 200 220 240 260 180 180 180 180
Holding time (h) 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 9.5 25.5
Pressure (Torr) 35 35 35 35 35 35 10 50 35 35
Pitch properties: üASTM D3104−14a (Softening point)üHelium gas pycnometer (Density)üASTM D4715-98 (Coking value) üASTM D2318 (Quinoline insoluble)
üGel permeation chromatography (GPC) (Molecular weight)
üGas chromatography (GC) (PAH content)
Chemical structure:ü1HNMR (Aromaticity)ü FTIR (Chemical groups)
üElectrospray ionization time-of-flight mass spectrometry (ESI-TOF MS)
(Molecular structure)
Characterization
Experimental conditions for synthesis bio-pitch samples
Diagram of the vacuum distillation of bio-oil
Bio-oil Bio-pitch
Synthesis
Pitch properties CTP BP-1 BP-2 BP-3 BP-4 BP-5 BP-6 BP-7 BP-8 BP-9 BP-10
Yield (wt %) - 70.4 65.3 61.3 54.9 50.6 48.8 65.2 66.3 60.7 47.8Softening point (°C) 110 53 75 87 121 149 162 80 68 89 168
Density (g/cm3) 1.35 1.25 1.26 1.27 1.28 1.27 1.27 1.27 1.27 1.27 1.27Coking value (wt %) 65.2 28.6 29.2 33.0 38.0 38.4 42.8 31.9 31.9 33.0 44.8
Quinoline insoluble (wt %) 16.00 0.36 0.71 0.77 0.92 1.02 1.39 1.44 0.97 0.77 1.22PAH content (wt %) 12.19 1.71 1.50 1.20 1.14 0.83 0.36 1.22 1.64 1.14 0.35
4. Chemical analysis
((aa))
((bb))
0.550
0.555
0.560
0.565
0.570
0.575
0.580
BP-2 BP-9 BP-10
Arom
atic
ity fr
om 1 H
NM
R
Bio-pitch samples
0.0
1.0
2.0
3.0
4.0
BP-2 BP-9 BP-10Area
COO
H/O
H/ A
rea
Alip
hatic
from
FTI
R
Bio-pitch samples
Polymerization of bio-oil showing several possible linkages.
Pitch properties BO BP-1 BP-2 BP-3 BP-4 BP-5 BP-6 BP-7 BP-8 BP-9 BP-10
GPCMw (g/mol) 310 669 949 1130 2875 13827 48891 2259 792 4649 52891
Polydispersity Index 2 3 4 4 9 31 121 10 3 39 122
Molecular weight of BO and BP from different pyrolysis conditions from GPC
Aliphatic Aromatic COOH/OH
Ying Lu Université Laval
D. Li - ULD. Picard - Alcoa
R. Mollaabbasi - ULJ. Tessier - Alcoa
T. Ollevier - ULH. Alamdari - UL
33Analyse thermogravimétrique d’isolants thermiques exposés
à la vapeur de sodium Thermogravimetric analysis of thermal insulating materials exposed
to sodium vapor
The bottom thermal insulating layer in aluminum electrolysis cells is normally well protected from chemical attacks by a refractory layer. However, autopsies of spent cathode lining have shown that sodium vapor may reach and react with the thermal insulating layer, thus affecting the thermal balance and structural integrity of the entire electrolysis cell. Here, we report on a new laboratory test, named Na-tg test, where commercial thermal insulating materials are exposed to sodium vapor at controlled temperature and sodium partial pressure. The weight change of three commercial insulation materials was measured as a function of time at various temperatures and partial pressures of sodium. The exposed materials were investigated with respect to changes in the microstructure, chemical and mineralogical composition, due to the reaction with sodium. The findings are discussed with respect to the chemical durability of the three thermal insulating materials.
La couche isolante inférieure d’une cellule d’électrolyse d’aluminium est généralement bien protégée contre les attaques chimiques par une couche réfractaire. Cependant, l’observation de cathodes ayant été en service a démontré que la vapeur de sodium peut atteindre et réagir avec la barrière thermique, affectant donc la balance thermique et l’intégrité de la cellule d’électrolyse. Dans cette étude, nous évaluons un nouveau test de laboratoire nommé Na-tg test, au cours duquel des matériaux isolants commerciaux sont exposés à la vapeur de sodium à une température et une pression de vapeur contrôlées. Le changement de masse de trois matériaux isolants commerciaux est mesuré en fonction du temps à différentes températures et pressions de vapeur de sodium. Les matériaux exposés sont étudiés selon les changements de microstructure et de composition chimique et minéralogique due à la réaction avec le sodium. Les résultats sont analysés en termes de durabilité chimique des trois matériaux isolants.
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Journée des étudiants – REGAL8 et 9 octobre 2019
Raymond LUNENG1, Zhaohui WANG1,2, Arne Petter RATVIK3 , Tor GRANDE1
1 Department of Materials Science and Engineering, NTNU Norwegian University of Science and Technology, NO-7491 Trondheim, Norway
2 SINTEF Industry, NO-7465 Trondheim, Norway
TThheerrmmooggrraavviimmeettrriicc AAnnaallyyssiiss ooff TThheerrmmaall IInnssuullaattiinngg MMaatteerriiaallss EExxppoosseedd ttoo
SSooddiiuumm VVaappoorr
Acknowledgements: Thanks to the CaRMa project with partners SINTEF, NTNU, Hydro Aluminium, Alcoa Norway, Elkem Carbon, and Skamol A/S, and the Norwegian Research Council for financial support.
Experimental setup
Challenge
Conclusions• Test method provides parameter
tunability and real-time measurements
• Mineralogical changes in reacted parts of all materials
• Limited attack on calcium silicate, excellent performance
• Glassy reaction layer on Moler, shrinkage
• Severe microstructure coarsening of vermiculite, shrinkage
Materials
Results
8 cm
2.4 cm
Na is known to be at the forefront of bottom lining degradation. How will it affect insulating materials?
Raymond LunengNorwegian University of Science and Technology
Z. Wang - NTNUA. P. Rarvik - SINTEFT. Grande - NTNU
3434
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
NL’estimation net de consommation de carbone dans la cuve d’électrolyse d’aluminium avec des méthodes multivariéesEstimation of net carbon consumption in aluminum electrolysis using multivariate analysis
Dans la littérature actuelle, la formule utilisée pour déterminer la consommation nette de carbone (NCC) dans les cuves d'électrolyse d'aluminium a été développée par R&D Carbon il y a plus de 25 ans. Elle fait le lien entre l'efficacité de courant, la température du bain et à quelques propriétés d'anode cuite. Les performances de cette formule ont été comparées à une approche statistique multivariée qui prend en compte les propriétés des anodes vertes et cuites, les paramètres de formulation et les performances des cuves. Pour ce faire, un système de suivi des anodes est utilisé pour faire le lien entre les différentes bases de données et permet de calculer le NCC pour chaque anode. Plus de 200 variables ont été utilisées dans un modèle des moindres carrés partiels (PLS), ce qui a donné un R2 de 66,6 %. Un modèle de régression multilinéaire (MLR) a également été créé en utilisant uniquement les variables trouvées dans la formule de R&D Carbon, ce qui a donné un R2 de 22 %. L’efficacité de prédiction de NCC est donc augmenté de 44 % par opposition au modèle précédemment développé.
Within the present literature the formula used to determine the net carbon consumption (NCC) in aluminum electrolysis was developed by R&D Carbon, over 25 years ago. It links the current efficiency of the reduction cell with bath temperature and few baked anode properties. The performance of this formula was compared to a multivariate statistical approach that considers green and baked anode properties, formulation parameters and reduction cell performance. The underlying equipment is an anode tracking system that links the different data-bases together and can thus calculate the NCC on an anode basis. Over 200 variables were used in the partial least squares (PLS) model which resulted in an R2 of 66.6%. A multilinear regression (MLR) model was also created using only the variables found in the R&D Carbon formula which resulted in an R2 of 22%. The prediction power of NCC is thus increased by 44% as opposed to using the previously developed model.
Petre Manolescu Université Laval
C. Duchesne - ULJ. Tessier - Alcoa
G. Saevarsdottir - Reykjavik University
Journée des étudiants – REGAL8 et 9 octobre 2019
Petre Manolescu1,2,4, Carl Duchesne1, Jayson Tessier3, Gudrun Saevarsdottir41 Department of Chemical Engineering, Université Laval, 1065 Av. de la Médecine, Québec, QC, G1V 0A6
2 Alcoa Corporation, Manufacturing Intelligence, Aluminerie de Deschambault, 1 Boul. des Sources, Deschambault-Grondines, QC, G0A 1S0
3 Alcoa Corporation, Aluminum Center of Excellence, Aluminerie de Deschambault, 1 Boul. des Sources, Deschambault-Grondines, QC, G0A 1S0
4 Reykjavik University, School of Science and Engineering, Menntavegur 1, 101 Reykjavik, Iceland
Estimation of Net Carbon Consumption in Aluminum Electrolysis Using Multivariate
AnalysisEstimation de la consummation nette de carbone
dans les cuves d’électrolyse de l’aluminium à l’aide de méthodes statistiques multivariées
2. ObjectivePredict the NCC on an anode basis using the anode tracking system (ATS) at Alcoa.The ATS consists of a series of weighing apparatus and cameras throughout the plant at Deschambault which provide information on individual anode properties, placement within the potline and butt shape. A link can thus be made between the differentdatabases thus enabling multivariate analysis.
1. IntroductionThe declining quality and increased variability of raw materials require new methods tomonitor and predict their impact on cell behaviour to maintain optimum performanceNet carbon consumption (NCC) in aluminum production can be defined as:
4. Methodology- Existing data structure for approximately 1200 individual anodes, AP30 technology,running approximately 380kA
3. Previous work on NCCR&D Carbon formula developed in 1991 based on full line operation data with celltechnology AP18 or older.NCC = 𝐶𝐶 + 334
𝐶𝐶𝐶𝐶 + 1.2 𝐵𝐵𝐵𝐵 − 960 − 1.7𝐶𝐶𝐶𝐶𝐶𝐶 + 9.3𝐴𝐴𝐴𝐴 + 8𝐵𝐵𝐶𝐶 − 1.5𝐴𝐴𝐶𝐶𝐶𝐶How good is this correlation on an individual anode basis ?
5. Models built1) PLS on the fullscale data incorporating approximately 200 variables. 2) Partial LeastSquares (PLS) on a matrix containing the same variables as the R&D Carbon formula.3) Multilinear Regression (MLR) on a matrix containing the same variables as the R&DCarbon formula.
- Approximately 200 variables:Formulation parameters, green & baked anode properties, pot operation data
- Pot operation data was averaged to remove the time dimension, variability present inthe dataset as standard deviation.- Butt information was used to calculate NCCAssumptions for calculating NCC:- Even consumption of the anode throughout its life- All anodes in the cell are consumed at the same rate
X-Space VIP[4] X-Space VIP[4]SumTapSchedule 4.60 SumResistanceModifierTotalDay 1.34Anode Set Cycle 3.85 avgAgeDay 1.33SumAluminaFed 3.49 avgBaseResistanceTargetDay 1.26334/CE 3.34 avgResistanceModifierTotalDay 1.24StdevAgeDay = SetCycle 3.29 BakedAnodeSulphur 1.19avgTapAmount 3.23 BakedAnodeCompressiveStrength 1.18avgMetalScheduledForTappingDay 2.02 StdevAmpsControlDay 1.15StdevTapAmount 1.72 CICB GEN 3 = Design 1.13BathTemp-960 1.69 StdevBathXsalf3Day 1.11AvgBathTemperatureDay 1.69 avgChampignonsTalons 1.09avgMetalSodiumNaDay 1.64 avgAluminaInCoverMix 1.08AvgResistanceDay 1.60 avgEventTalonsChampignons 1.07Count_TempBelow960 1.51 avgGreenAnodePitchTemperature 1.05AvgVoltsDay 1.51 StdevResBaseResistTargetDay 1.04AvgBathXsAlF3TargetDay 1.49 StdevBathTemperatureDay 1.04AvgStpBathXsalf3Day 1.48 BakedAnodeFractureEnergy 1.03AvgBathLevelTarget 1.47 avgMeasMetalPhosphorusPDay 1.02AvgMetalPadTargetDay 1.38 BakedAnodeCO2LossReactivity 1.01StDevMetalScheduledForTappingDay 1.36 StdevChampignonsTalons 1.00
GreenAnode Resistivity HLS 1.00
6. Model comparison & performanceThe prediction power of each of the three methods was compared. The best choicewould be the model that has the highest variance explained (R2) and lowest RMSE(prediction error) here given by the full scale PLS model.
Model performanceThe full scale PLS model explains 66.6% of the variance within the dataset, thepredicted values fitting well with the observed values.
8. AcknowledgementsThe authors would like to acknowledge the financial support of the Natural Sciences and Engineering Research Council (NSERC), Fonds de Recherche du Québec – Natureet Technologies (FRQNT), Alcoa and the Aluminum Research Centre – REGAL.
7. Next stepsExplore how the model could be improved by adding raw materials properties and pot tending information. Assess potential operationalization to improve plant performance.
Variables important to prediction in the PLS fullscaleThe set cycle and metal production rank highest. Pot operation parameters dominate the model with influence from core sample analysis, green anode properties and formulation, metal impurities and cover material integrity, all highlighted below. The main driver for the other two models created is the cell’s current efficiency.
PLS model Observed vs Predicted NCC
Green Anode PropertiesCore Samples AnalysisPot Operation ParametersMetal Production
J. C. Fischer, 27th International Aluminium Conference, 2012
Figure 1: Net Carbon Consumption in aluminum electrolysis
Figure 2: Data Structure
Figure 3: Performance of the methods: a) RMSE comparison b) R squared comparison
PredictedFigure 4: Observations vs Predictions of NCC given by the full scale PLS model
Obse
rved
Table 1: Variables Important to Prediction in the fullscale PLS model
35
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Propriétés rhéologiques de la matrice lianteRheological properties of the binder matrix
Green anodes consist of coarse coke particles surrounded by a binder matrix, which is a viscoelastic material and is made of fine coke particles and coal-tar pitch. During the compaction process, the coarse particles re-arrange while the binder matrix plastically deforms and reduces the overall porosity of the coarse particles. The rheological properties of the binder matrix are among the most important parameters that affects the final quality of the anode with respect to the anode paste recipe, the interaction of the binder matrix with the coarse particles and operational conditions. Different parameters such as particle size distribution of fine particles, concentration of fine particles and forming temperature are governing the rheological properties of the binder matrix. Here, the effects fine particle concentrations are considered to simulate real conditions and improve the understanding of the viscoelastic behaviour of the binder matrix.
Les anodes vertes sont constituées de grosses particules de coke entourées d’une matrice liante, constituée d’un matériau viscoélastique composé de fines particules de coke et de brai de houille. Pendant le processus de compactage, les particules grossières se réarrangent tandis que la matrice liante se déforme de manière plas-tique et réduit la porosité globale des particules grossières. Les propriétés rhéologiques de la matrice liante font partie des paramètres les plus importants qui affectent la qualité finale de l'anode en ce qui concerne la recette de la pâte anodique, l'interaction de la matrice liante avec les particules grossières et les conditions de fonc-tionnement. Différents paramètres tels que la distribution granulométrique des particules fines, la concentration en particules fines et la température de formation déterminent les propriétés rhéologiques de la matrice liante. Ici, les effets des concentrations de particules fines sont considérés comme simulant des conditions réelles et améliorant la compréhension du comportement viscoélastique de la matrice liante.
Roozbeh MollaabbasiUniversité Laval
L.J. Hansen - NTNU T. S. Barry - UL D. Ziegler - AlcoaT. Grande - NTNU S. M. Taghavi - ULH. Alamdari - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
R. Mollaabbasi1, L.J. Hansen2, T. S. Barry1 D. Ziegler3,T. Grande2 , S. M. Taghavi1 and H. Alamdari1
1Aluminium Research Centre- REGAL, Université Laval, Québec, QC, CANADA2 Department of Materials Science and Engineering, NTNU Norwegian University of Science and Technology
3 Alcoa Primary Metals, Alcoa Technical Centre, 859 White Cloud Road, New Kensington, PA, 15068, USA
Rheological Properties of BinderMatrix
Propriétés rhéologiques de la matrice liante
1: Context
Fig. 2: Compation of green anodes with (a) press-compactor (60-70 MPa) and (b) vibro-compactor (1 MPa).
(a) (b)Coarse particles
Binder matrixAnode Paste
Pitch
Fine particles
Density gradients in anodes:• Preferential current paths• Non-uniform anode consumption• Possible cracking during baking and aluminium production
Serious challenges:• High temperature• Pitch sensitivity to temperature variation
2: Objectives• Characterize the rheological properties of the binder matrix at forming temperature• Find fluid model with same rheological properties of binder matrix at room temperature
4: Results Oscillation testsT=166ºC T=178ºC T=190ºC
Rotation testsT=166ºC T=178ºC T=190ºC
5: Conclusion and Future works 6: Acknowledgements• Binder matrix is a visco-elastic material.• Fine particle concentration Viscous/elastic properties of the binder matrix • Temperature Viscous/elastic properties of the binder matrix
This research has been carried out at Université Laval,supported financially by RÉGAL and NSERC. The authorsacknowledge Alcoa for the support of this work. We greatlythank. Dr. D. Picard, A. Eslami and G. Gauvin (LavalUniversity) for very useful discussions.
Fig. 1: Density distribution of baked anode.
• Temperature gradient• Opaque materials
3: MethodologyPerforming Rotation and Oscillation tests at different
temperatures and concentrations to extract the rheological properties of binder matrix
at high temperature.
𝜼𝜼∗ = 𝑮𝑮∗𝝎𝝎
τ = κ ሶ𝜸𝜸𝒏𝒏𝝁𝝁 = κ ሶ𝜸𝜸𝒏𝒏−𝟏𝟏Viscosity
Shear stress
Rotation Oscillation
Fig. 3: The Oscillation results of binder matrix.
Fig. 4: The Rotation results of binder matrix.
Adjust the fluid model Perform tests at room temperature:
Studying effects of coke granulometry.Studying effects of vibration..
Fig. 5: The results of model fluid (Poly Ethylen Oxide solution at room temperature).
Increasing concentration of PEO
Complex viscosity
Complex modulus𝑮𝑮∗ = 𝑮𝑮′𝟐𝟐 + 𝑮𝑮′′𝟐𝟐
3636
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Étude des facteurs influençant les résultats de la flexion des carottes d’anodeStudy of the factors influencing the flexural strength results of anode cores
Des anodes de carbone sont utilisées dans le procédé électrolytique de la production de l’aluminium. Leur qualité a un impact important sur plusieurs aspects de la production, de la consommation énergétique aux émissions environnementales. La qualité d’une anode est définie par ses propriétés. La résistance à la flexion est une des propriétés importantes qui influence le comportement de l’anode durant l’électrolyse. La mesure de la flexion des carottes d’anode sous-entend que la cohésion entre les particules de coke calciné augmente à la suite de la cuisson des anodes. Pourtant, des résultats de flexion différents existent pour une même recette de pâte d’anode. La revue des paramètres de vibrocompaction et de la structure de l’anode a été réalisée sur des carottes ayant des résultats de flexion faible et élevé. La technique de l’analyse d’image a été utilisée pour l’étude de la structure crue et cuite de ces carottes d’anode.
Carbon anodes are used in the electrolytic production of aluminum. Their quality has an important impact on many aspects of the production process, from energy consumption to environmental emissions. The quality of an anode is defined by its properties. The flexural strength is one of the important properties that affects the anode behavior during the electrolysis. The flexural strength results of the anode cores indicate how well the cohesion between calcined coke particles increased as a result of the anode baking. However, different flexural strength results are found for the same anode paste recipe. A review of vibrocompaction parameters and the anode structure was carried out on cores with low and high flexural strength results. The image analysis technic was used to study the green and baked structures of the anode cores.
Brigitte MoraisUniversité du Québec
à Chicoutimi
D. Kocaefe - UQAC Y. Kocaefe - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
Brigitte Morais, Duygu Kocaefe, Yasar KocaefeChaire institutionnelle sur les matériaux industriels (CHIMI), Centre universitaire de recherche sur l’aluminium (CURAL),
Université du Québec à Chicoutimi, 555 boulevard de l’Université, Chicoutimi, Québec G7H 2B1
Étude des facteurs influençant les résultats de la flexion des
carottes d’anodeStudy of the factors influencing the flexural strength results of
anode cores
Problématique
Méthodologie
Analyse
Discussion
IntroductionLa dissociation de l’alumine Al2O3 en aluminium et en l’oxygène dans leprocédé d’électrolyse survient lorsqu’une quantité d’énergie électrique estimposée en passant de la tige de l’anode vers la barre cathodique à la base dela cuve d’électrolyse à environ 960 °C. La mesure de la résistance à la flexionindique la capacité de l’anode à maintenir son intégrité durant sa manutention,sa mise en régime en température et en courant électrique au démarraged’une cuve d’électrolyse ou en service.
La mesure de la flexion des carottes d’anode sous-entend que la cohésion entre les particules de coke calciné augmente à la suite de la cuisson des anodes. Pourtant des résultats de flexion différents existent pour une même recette de pâte d’anode.
4 Al + 3 CO22 Al2O3 + 3 C
Objectif Déterminer ce qui influence le résultat de la résistance à la flexion.
Étudier les paramètres de vibrocompaction en variant les pressions au-dessus eten dessous du vibrocompacteur de 17 anodes en utilisant l’échantillonnagedouble standard. Étudier la structure de l’anode sur des carottes ayant desrésultats de flexion faible et élevée en utilisant les échantillons des carottes 05Aet 05B, rang 1 et 2, et en position 5 de cuisson. La technique de l’analysed’image a été utilisée pour l’étude de la structure crue et cuite de ces carottes.
Pression du haut
Pression du bas
PHBas PHMoy PHHaut VIBRO A
1 PBBas
2 PBMoy
3 PBHaut
3 4 5
Rang
de
cuiss
on
Pres
sion
du b
as
Position de cuisson
Pression du hautÉchantillonnage Double standard :
carottes 5 et 7 (8 petites carottes)
No. d’anode
1
3 5
6
742
Std
Anode CrueAnode cuite
Paramètres de vibrocompaction Mosaïque au microscope optique
Les valeurs > 100% s’explique par la normalisation des moyennes de
l’échantillonnage double standard. La même valeur de normalisation a été
utilisée pour les carottes individuelles 05A & 05B
PHBas
PHMoy
PHHaut
7580859095
100
PBBasPBMoy
PBHaut
Accélération Tige- Vibro A
100-101
95-100
90-95
85-90
80-85
75-80
Rang 3 Position 3
Position 4
Position 5
Rang 1Rang 2
PHBas
PHMoy
PHHaut
84889296
100
PBBasPBMoy
PBHaut
Accélération Table- Vibro B
100-102
96-100
92-96
88-92
84-88
Rang 1 Position 3
Position 4
Position 5
Rang 3Rang 2
PHBas
PHMoy
PHHaut
7580859095
100
PBBasPBMoy
PBHaut
Accélération Tige- Vibro B
100-101
95-100
90-95
85-90
80-85
75-80Rang 3
Rang 2 Rang 1 Position 3
Position 4
Position 5
PHBas
PHMoy
PHHaut
84889296
100
PBBasPBMoy
PBHaut
Accélération Table- Vibro A
100-102
96-100
92-96
88-92
84-88
Rang 3 Position 3
Position 4
Position 5
Rang 1
Rang 2
Figure 3 Structure du coke pour un résultat de flexion faible et fort : le coke avec sa périphérie et la matrice libre (crue) et externe (cuit) en noire
Flexion
Faible
Forte
État crue État cuit : les nouvelles particules de coke formées
État cuit : la connexion entre les nouvelles particules de coke
formées
= Coke= Liant
= Pore
= Matrice libre
Terminologie de la structure d’anode
Liant = Mélange de brai et de fines particules de coke
à l’état cru ou cuit (couleur verte)
Si le liant une fois cuitdemeure de couleur verteil représente alors, unetransformation vers le cokemoins avancée que celle dela structure environnante.
Situation A À l’état cru (haute – faible flexion) (-) de liant et (+) de coke dans la structure générale (+) de cavités en périphérie des particules de coke Une faible quantité de la matrice libre (-) d’efficacité de remplissage, d'enrobage ou de malaxage
puisqu’il y a moins de liant
Conclusion
Autres Situations Un plus grand nombre de fissures induit des résultats de flexion faible. L’accélération de la table du vibrocompacteur influence les résultats de
flexion seulement dans les meilleures conditions de cuisson (rang 2) etavec l’usage de pression moyenne au bas du vibro. (Acc. Flexion )
Le confinement et le poids au-dessus de l’anode améliorent les résultats,quelle que soit la pâte utilisée au jour 1 ou 2 lors des tests en usine.
Situation C L’échantillon avec le résultat le plus faible de flexion
Augmentation de 3,4% des pores au lieu de diminuer après lacuisson
La plus grande quantité de liant et peu de pores (+) de pores comblés à l’intérieur des particules de coke (+) de liant dans la matrice libre qui possède peu de pores
Figure 1 Accélérations des tables et des tiges
0,0
0,2
0,4
0,6
0,8
1,0
Rang 1 Rang 2 Rang 3
R2Fi
ssur
e vs
Flex
ion
Le rang de cuisson 1 étant le dessus du four
R2 de la relation du nombre de fissures par rapport aux résultats de la flexion
Figure 2 Relation entre le nombre de fissures et les résultats de flexion
Situation B À l’état cuit (haute – faible flexion) La même quantité de structure moins transformée La même augmentation de la quantité de nouvelle structure de coke et
de cavités à l’intérieur Plus de liant dans les cavités du coke Moins de pores en périphérie du coke et l’indice d’enrobage a augmenté
Figure 4 Le nombre de particules de coke après la cuisson
Par rapport aux résultats de haute flexion
Analyse d’imageTest en usine
Une anode avec une structure possédant trop deliant et pas assez de pores peut ne pas avoir letemps nécessaire pour se transformer en unestructure solide à la cuisson. La structure del'anode peu changée pour une même recette.
L’échantillon, Vibro B 1-5, avait un nombre élevé de particules, mais la quantité de la matricelibre était juste assez pour toutes les lier efficacement les unes aux autres. La composition dela pâte peut changer les résultats de flexion, quel que soit l’ajustement des autres paramètresdu procédé de fabrication de l’anode. L’analyse d’image de la structure de l’anode crue et cuiteest un outil efficace pour statuer sur les résultats de flexion.
Rang 1 et 3 sont intervertis pour le vibro B
Tableau 5
Table Tige Table-Tige Table Tige Table-TigeRang 1Rang 2 0,8696Rang 3
Position 3Position 4Position 5
= Pas de corrélation
R 2 Flexion et fissure avec les accélérations des tables et des tiges des vibrocompacteurs
Fissure Flexion
0
20
40
60
80
100
0
20
40
60
80
100
120
Flexio
n
Nom
bre
de p
artic
ules
Vibro, rang et position de cuisson
Dénombrement de particules de coke dans la structure de l'anode crue et cuite
Particule de coke naturel Particule de coke avec périphérie Flexion Flexion
Anode crue Anode cuite
Tamis Grand Moyen Petit P. Broyé PoussièreJour (2-1) 78 42 -144 100 -75
Différence entre les recettes Vibro B (Jour2) et Vibro A (Jour 1)Tableau 6
Tableau 1 Tableau 2
Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro BRang 1 95,3 95,2 87,5 87,2 0 0 Rang 1 58,7 79,3 81 29 99,3 98,4 96,3 95,7Rang 2 97,2 97,7 95,2 96,1 1 1 Rang 2 71,8 85,9 21 10 99,3 98,0 96,3 95,6Rang 3 91,1 89,8 81,2 78,8 2 2 Rang 3 79,3 98,2 16 2 99,4 98,7 96,2 96,1
* 07D
Tableau 3 Tableau 4
Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro B Vibro A Vibro BRang 1 94,7 94,5 88,0 89,4 0 0 Rang 1 50,0 84,6 7 0 98,5 100,3 97,9 96,7Rang 2 97,0 96,1 89,1 90,2 1 1 Rang 2 53,8 88,5 1 0 102,1 101,5 97,9 96,0
* 07D
Densité cuiteDensité crue
Densité cuite
Conditions d'expérience pour les 17 anodesFlexion Fissure
Résultats pour les 17 anodes
Flexion Fissure (Nombre) Densité crue
T de cuisson Anode au dessusLien cristallin*
T de cuissonConditions d'expérience pour les 2 carottes 05A et 05B
Anode au dessusRésultats pour les 2 carottes 05A et 05B
Lien cristallin*
Tableau 7 La structure de l'anode par l'analyse d'image : Les résultats de la forte valeur de flexion après la soustraction de la faible valeur de flexion
% Pore % Liant % Coke % Cavité % Pore % Liant % Cavité % Pore % LiantTotal Total Total % Coke % Matrice
(libre/externe) % Pore % Liant
Vibro A 2,9 -11,0 8,1 -1,8 1,0 -2,9 3,0 2,0 0,9 9,3 -9,3 -0,2 -9,0 -12,0 -22,0 -8,6Vibro B -0,2 -8,4 8,6 0,0 1,4 -1,4 2,9 2,4 0,5 11,5 -11,5 -4,0 -7,5 -12,5 -20,1 -10,2
Vibro A -3,2 0,5 2,7 3,5 -0,5 4,0 -4,4 -2,2 -2,2 1,8 -1,8 -0,5 -1,3 5,8 7,5 4,5Vibro B -1,8 -0,5 2,3 3,2 0,7 2,5 -5,1 -2,2 -2,9 0,4 -0,4 -0,3 -0,1 2,5 0,3 4,0
Tableau 8 La structure de l'anode par l'analyse d'image : La différence entre le plus faible résultat de flexion et les trois autres valeurs Vibro A 8,4 42,3 49,3 8,1 1,5 6,7 14,9 2,6 12,3 72,3 27,7 4,4 23,4 82,5 81,9 82,8Vibro A 11,3 31,3 57,4 6,3 2,5 3,8 17,8 4,6 13,2 81,5 18,5 4,1 14,3 70,4 59,9 74,2ViBro B 12,7 36,8 50,5 6,8 2,1 4,7 14,9 2,9 12,0 72,2 27,8 7,7 20,1 77,1 69,4 80,6ViBro B 12,4 28,4 59,2 6,8 3,4 3,3 17,8 5,3 12,5 83,7 16,3 3,7 12,5 64,6 49,3 70,4
Vibro A 1-5 Flexion la plus faible 11,8 37,0 51,2 14,1 3,3 10,8 30,2 7,1 23,1 95,6 4,4 1,4 3,1 76,5 76,8 76,4Vibro A 2-5 8,6 37,5 54,0 17,6 2,8 14,8 25,8 4,9 20,9 97,4 2,6 0,9 1,7 82,3 84,4 80,9ViBro B 1-5 12,1 30,5 57,4 24,5 5,9 18,6 17,5 5,8 11,7 99,5 0,5 0,3 0,2 72,0 75,9 66,7ViBro B 2-5 10,3 30,0 59,7 27,7 6,6 21,1 12,4 3,6 8,8 99,8 0,2 0,0 0,1 74,5 76,2 70,7
Efficacité Malaxage
Efficacité Remplissage
Efficacité Enrobage
Cavité périphérie du Coke
2-5 Crue
Région du Coke
Matrice libre (crue) ou matrice externe (cuit)
Dans le coke En périphérie du coke
Rang-Position Comparaison des
résultats de flexion
Structure Générale Cavité dans le CokeÉtat
Cuit
Crue
Cuit
1-5 Flexion la plus faible
Situation
B
Vibro
C
X-5 (X=Rang 2 -Rang 1)X-5 (X=Rang 2 -Rang 1)
X-5 (X=Rang 2 -Rang 1)X-5 (X=Rang 2 -Rang 1)
1-5 2-5
A
37
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Transport de masse par les vagues à l’interface bain-métal dans une cuve d’électrolyse
Mass transport by waves along the bath-metal interface in electrolysis cells
In the aluminium electrolysis cell, there are two liquid layers, the cryolite bath on the top and the molten alum-inum at the bottom. Particles and solid bodies (undissolved alumina, cryolite snow) float on the bath-metal inter-face. These particles can move because of the movements of the fluids in the cell, induced by the magnetic field, by the CO
2 gas bubbles and by the natural convection. These movements generate more or less regular waves at
the bath-metal interface which can contribute to the horizontal displacement of floating particles. The transport of solids by surface waves is called Stokes drift, a phenomenon that can be observed on the ocean surfaces (e.g. drifting boats) but it has not been studied for the case of the bath-metal interface. The purpose of this work is therefore to review the literature of the research on surface and interface waves, and then to estimate what is the contribution of the Stokes drift to the movement of solids floating on the interface.
Dans les cuves d’électrolyse d’aluminium, il y a deux couches de liquide, soit un bain de cryolite en haut et de l’aluminium en fusion au-dessous. Des particules et des corps solides (alumine non-dissous, neige de cryolite) peuvent flotter au niveau de l’interface bain-métal. Ces particules peuvent se déplacer à cause des mouvements des fluides dans la cuve, induits par le champ magnétique, par le mouvement de bulle de gaz CO
2 ou par la
convection naturelle. Ces mouvements génèrent des vagues plus ou moins régulières à l’interface bain-métal. Ces vagues peuvent contribuer aux déplacements horizontaux des particules flottant à l’interface. Le phénomène de transport de matière par les ondes de surface est appelé mouvement de dérive de Stokes, un phénomène qu’on peut observer au niveau des surfaces des océans (p. ex. dérive des bateaux) mais n’est pas encore étudié pour le cas de l’interface bain-métal. L’objet de ce travail est donc de faire une revue de littérature des travaux effectués sur les ondes de surface et ondes d’interface, puis d’estimer quel est l’apport de la dérive de Stokes au mouvement des corps flottant à l’interface.
Lovatiana RakotondramananaUniversité du Québec à Chicoutimi
L. Kiss - UQACS. Poncsák - UQACS. Guérard - Rio TintoJ-F. Bilodeau - Rio Tinto
Journée des étudiants – REGAL8 et 9 octobre 2019
L. Rakotondramanana1, L. Kiss1, S. Poncsák1, S. Guérard2, J-F. Bilodeau2
1 GRIPS, Université du Québec à Chicoutimi, 555 boulevard de l’Université, Chicoutimi, QC, CANADA, G7H 2B1
2 Centre de recherche et de développement d’Arvida, Rio Tinto Aluminium, 1955 boulevard Mellon, Jonquière, QC, G7S 4K8, Canada
Transport de masse par les ondes à l’interface bain-métal
dans une cuve d’électrolyseMass transport by waves on the bath-
metal interface in electrolysis cell
ProblématiqueLes particules à la surface d’un milieu fluide animé par une onde de surface se déplacent dans le sens de propagation de l’onde, c’est la dérive deStokes. Ce mouvement de dérive contribue aux déplacement horizontal des corps flottant à la surface. Durant l’électrolyse de l’aluminium, desradeaux formés de bain solidifié et d’alumine non dissout peuvent flotter à l’interface bain métal (IBM). Sachant que l’IBM est animé par desoscillations plus ou moins régulières (MHD, bulles de 𝐶𝐶𝐶𝐶2, convection), les radeaux peuvent être transportés par ces ondes.Objectifs- Simuler le mouvement d’un radeau flottant à l’interface bain métal à l’aide du modèle de Rumer.- Estimer l’influence des paramètres tel que la longueur d’onde, l’amplitude, les coefficients de traînée sur le déplacement du radeau.
ConclusionLe mouvement d’un radeau flottant à l’interface bain métal est étroitement lié à la pente del’onde et aux vitesses des particules de fluide. Plus la pente est important, plus la gravité aune plus grande influence et le radeau se déplace plus vite. Pour une grande longueurd’onde telle qu’on retrouve à l’interface bain métal, le radeau finit par se déplacer à la mêmevitesse que la propagation de l’onde quand le coefficient de traînée est relativement grand.
𝐹𝐹𝑔𝑔𝑥𝑥 = 12 𝑚𝑚 𝑔𝑔 + 𝜕𝜕2𝜂𝜂
𝑑𝑑𝑡𝑡2 𝑠𝑠𝑠𝑠𝑠𝑠 2𝜃𝜃
𝐹𝐹𝑑𝑑𝑥𝑥 𝑏𝑏 = 𝜌𝜌𝑏𝑏 𝐶𝐶𝑑𝑑 𝐴𝐴 𝑣𝑣𝑥𝑥 − 𝑢𝑢𝑥𝑥 (𝑣𝑣𝑥𝑥 − 𝑢𝑢𝑥𝑥)𝐹𝐹𝑑𝑑𝑥𝑥 𝑚𝑚 = 𝜌𝜌𝑚𝑚 𝐶𝐶𝑑𝑑 𝐴𝐴 𝑣𝑣𝑥𝑥 − 𝑢𝑢𝑥𝑥 (𝑣𝑣𝑥𝑥 − 𝑢𝑢𝑥𝑥)
𝑢𝑢𝑥𝑥 : composante horizontale de la vitesse du radeau;𝑣𝑣𝑥𝑥 : composante horizontale de la vitesse du fluide𝐹𝐹𝑔𝑔𝑥𝑥: contribution horizontale de la force de gravité due à la pente𝐹𝐹𝑑𝑑𝑥𝑥 𝑏𝑏 : projection horizontale de la force de traînée due au bain𝐹𝐹𝑑𝑑𝑥𝑥 𝑚𝑚 : projection horizontale de la force de traînée due au métal𝐶𝐶𝑑𝑑: coefficient de traînée; 𝐶𝐶𝑎𝑎: coefficient d’ajout de masse𝐻𝐻 : amplitude de l’onde; 𝐿𝐿 : longueur d’onde; 𝑇𝑇 : période de l’onde𝑚𝑚: masse de l’objet𝐴𝐴: section de l’objet dans le plan perpendiculaire à sa vitesse
Déplacement horizontal d’un objet flottant 𝜌𝜌 = 1000𝑘𝑘𝑔𝑔/𝑚𝑚3 𝐶𝐶𝑑𝑑 = 0.08
𝑚𝑚 = 15𝑔𝑔; 𝐿𝐿 = 0.5𝑚𝑚; T = 0.57 𝑠𝑠; 𝐻𝐻 = 1.7𝑐𝑐𝑚𝑚; 𝐴𝐴 = 5𝑐𝑐𝑚𝑚 × 2𝑐𝑐𝑚𝑚𝑚𝑚 = 100𝑔𝑔; T = 0.6 𝑠𝑠; 𝐻𝐻 = 1.5𝑐𝑐𝑚𝑚; 𝐴𝐴 = 10𝑐𝑐𝑚𝑚 × 3𝑐𝑐𝑚𝑚𝑚𝑚 = 200𝑔𝑔; T = 0.57 𝑠𝑠 𝐿𝐿 = 0.5𝑚𝑚; 𝐴𝐴 = 10𝑐𝑐𝑚𝑚 × 3𝑐𝑐𝑚𝑚𝐿𝐿 = 0.5𝑚𝑚; T = 0.57 𝑠𝑠; 𝐻𝐻 = 1.5𝑐𝑐𝑚𝑚; 𝐴𝐴 = 10𝑐𝑐𝑚𝑚 × 3𝑐𝑐𝑚𝑚
𝑚𝑚 = 1 𝐾𝐾𝑔𝑔; 𝐿𝐿 = 5 𝑚𝑚; T = 40 𝑠𝑠; 𝐻𝐻 = 4 𝑐𝑐𝑚𝑚; 𝐴𝐴 = 15 𝑐𝑐𝑚𝑚 × 4𝑐𝑐𝑚𝑚; 𝜌𝜌𝑏𝑏𝑎𝑎𝑏𝑏𝑏𝑏 = 2100𝑘𝑘𝑔𝑔/𝑚𝑚3; 𝜌𝜌𝑚𝑚𝑚𝑚𝑚𝑚𝑎𝑎𝑚𝑚 = 2300𝑘𝑘𝑔𝑔/𝑚𝑚3
L’étude paramétrique est effectuée sur unobjet flottant à la surface de l’eau car lespropriétés de la propagation d’onde de surfacedans ce liquide sont bien connues.À 𝑡𝑡 = 0, l’objet est posé à une vitesse nulle surla surface de l’eau animé par une onde desurface. L’objet commence par undéplacement s’opposant à la propagation del’onde, mais finit par suivre son sens.
• Quand la longueur d’onde diminue,la vitesse de l’objet flottantaugmente.
• Quand l’amplitude de l’ondeaugmente, la vitesse de l’objetaugmente.
• L’augmentation de la masse del’objet diminue sa vitesse.
Dérive de StokesLes particules dans un milieu fluide animé d’uneonde de surface suivent une trajectoire trochoïdale.La vitesse du fluide induit une force de traînée quiinfluence le mouvement du corps flottant.
Résultats
• Quand la force de traînée estnulle (𝐶𝐶𝑑𝑑 = 0), l’objet se déplacedans le sens opposé au sens depropagation de l’onde. Pour un𝐶𝐶𝑑𝑑 non nul, quand le systèmeatteint son régime permanent,l’objet se déplace à la mêmevitesse que la propagation del’onde, l’objet reste au creux del’onde. Plus le coefficient detraînée est grand, plus la périodetransitoire est courte.
Modèle de Rumer
Trajectoires d’un radeau flottant à l’interface 𝐶𝐶𝑑𝑑
𝒎𝒎 𝟏𝟏 + 𝑪𝑪𝒎𝒎𝒅𝒅𝒖𝒖𝒙𝒙𝒅𝒅𝒅𝒅 = 𝑭𝑭𝒈𝒈𝒙𝒙 + 𝑭𝑭𝒅𝒅𝒙𝒙 𝒃𝒃 + 𝑭𝑭𝒅𝒅𝒙𝒙 𝒎𝒎
où
La projection des forces sur l’axe horizontal donne:
Ces équations sont résolues numériquement sur MATLAB pour obtenir la vitesse 𝑢𝑢𝑥𝑥(𝑡𝑡) et la position de l’objet flottant 𝑥𝑥(𝑡𝑡)
[R. R. Rumer, A. Wake and R. Crissman, Ice transport in Great Lakes, Great Lakes Environmental Research Laboratory, National Oceanic and Atmospheric Administration, U.S. Dept. of Commerce, 1979]
3838
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Amélioration de la qualité de l’anode par la modification du braiImprovement of anode quality through pitch modification
Dans l’industrie de l’aluminium, la qualité de l’anode a un impact direct sur le coût du métal, la consommation d'énergie et les émissions environnementales. Afin d’atteindre les propriétés finales des anodes nécessaires à l’efficacité du procédé électrolytique, une liaison satisfaisante coke/brai doit être obtenue. Une étude a démontré qu’il est possible de favoriser cette liaison par la modification du brai à l’aide d’additifs et d’améliorer certaines propriétés de l’anode. Ce projet est une continuation de cette étude antérieure pour l’amélioration de la qualité de l’anode par la modification du brai en utilisant des additifs. Les additifs sont choisis pour augmenter les groupes fonctionnels à la surface du brai afin d’améliorer la mouillabilité. En outre, la sélection des produits chimiques était basée sur leur impact économique pour l’application possible en industrie, leur impact sur l’environnement et la santé/sécurité des travailleurs. La méthodologie inclut la caractérisation des matériaux utilisés dans ce pro-jet en étudiant leur composition chimique par FTIR, les interactions coke/brai en utilisant les tests de mouillabilité, et d’autres méthodes afin de mieux comprendre les mécanismes sous-jacents, ce qui permettra de sélectionner l’additif et sa concentration appropriés pour fabriquer des anodes.
In aluminum industry, the quality of anodes has a direct impact on the cost of metal, energy consumption, and environmental emissions. In order to achieve the final properties of anodes required for an efficient electrolysis process, pitch must sufficiently bind with coke. A study showed that it is possible to enhance this binding through the modification of pitch using additives and improve certain anode properties. This project is a continuation of that previous study for the improvement of anode quality by modifying the pitch with additives. The additives are chosen with the objective of increasing the functional groups on the surface of pitch in order to improve the wettability. In addition, the economic impact for an eventual industrial application, their effect on the environment and the health and safety of employees were taken into account during the selection of the chemicals. The meth-odology involves the characterization of materials used in this project by studying their chemical composition with FTIR, the coke-pitch interactions using the wettability test, and other methods in order to understand the coke/pitch interaction mechanisms to be able to select the type and the concentration of the additive to be used in anode fabrication.
Armita Rastegari Université du Québec
à Chicoutimi
J. Bureau - UQACD. Kocaefe - UQAC Y. Kocaefe - UQAC
H. Darmstadt - Rio Tinto
Journée des étudiants – REGAL8 et 9 octobre 2019
In aluminum production, coal tar pitch is used as the binder material for producing carbon anodes whichare mostly composed of coke. The coke/pitch interactions should be strong since they considerablyaffect the quality of anodes. The quality of raw materials are variable and affect the bonding betweencoke and pitch. It is well-known that some of the major challenges of the aluminum industry are toincrease production and reduce energy consumption and environmental emissions. In order to achievethese objectives and continuously improve the process, it is important to find new avenues. Therefore,the modification of pitch properties using additives can be carried out to enhance the binding betweencoke and pitch, which will help produce better quality carbon anodes.
Results and Discussion Introduction and Statement of the problem
Objectives Improve coke/pitch interactions by modifying pitch using additives in order to enhance the wettability
of coke by pitch. Determine the impact of the additive by comparing: (a) the FTIR spectra of non-modified and modified
pitches (with different additives at different percentages) and (b) their wetting behavior. Understand the underlying mechanism to be able to select an appropriate additive and its concentration
for better coke/pitch interactions.
AcknowledgementsThe technical and financial support of Rio Tinto Aluminum as well as the financial support of theNatural Sciences and Engineering Research Council of Canada (NSERC), and the University ofQuebec at Chicoutimi (UQAC) is greatly appreciated.
MethodologyRaw Material: Pitch Coke
Characterization:• FTIR• Wettability (Sessile drop)
Additive Selection
Pitch Modification
Anode fabrication
Device for measuringcompressive strength
FTIR
Add3c2
A: B:
C:
D:
Heteroatoms
Aromatics and Aliphatic
Aromatics and Aliphatic
Heteroatoms
Aromatics and Aliphatic
Heteroatoms
Device for measuring resistivity
Characterization of anodes:1. Apparent density
2. Electrical resistivity
3. Air and CO2 reactivities
4. Compressive strength
5. Others if necessary
Wettability system in UQAC
Abs calculated = Absλ × 100 −%additi𝑣𝑣𝑣𝑣100 pitch
+ Absλ ×%additi𝑣𝑣𝑣𝑣100 additive
Aromaticity index (Ratio A) =Abs𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 ÷ ∑Abs𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎
Heteroatomes index (Ratio H) =∑Abs𝑎𝑣𝑣𝑎𝑎𝑣𝑣𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑣𝑣𝑠𝑠 ÷ ∑Abs𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎𝑎𝑎𝑎 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎
Non-Modified
Add1c2
Add2c2
a b
c d
e f
g h
Ɵ = 107.7 ° Ɵ = 46.82 °
Ɵ = 114.22 ° Ɵ = 33.91 °
Ɵ = 112.85 ° Ɵ = 19.56 °
Ɵ = 102.29 ° Ɵ = 28.49 °
Non-Modified
Add1c2
Add2c2
Add3c2
t1 = 0 s t2 ~ 60 s
Armita Rastegari1, Julie Bureau1, Duygu Kocaefe1, Yasar Kocaefe1 et Hans Darmstadt2
1Chaire institutionnelle sur les matériaux industriels (CHIMI), Centre universitaire de recherche sur l’aluminium (CURAL),Université du Québec à Chicoutimi, 555 boulevard de l’Université, Chicoutimi, Québec G7H 2B1
2Rio Tinto, Arvida Research and Development Center (ARDC), 1955, Boulevard Mellon, C.P 1250, Jonquière, Quebec, Canada G7S 4K8
Amélioration de la qualité de l’anode par la modification du brai
Improvement of Anode Quality Through Pitch Modification
1.Kocaefe et al., Light metals, TMS 2013 pp.1045-1050.2. Rocha et al., Fuel, 84, 2005 pp, 1550-1558.3. Bureau J., "Étude de l’amélioration de la qualité des anodes par la modification des propriétés du brai", Master’s Thesis, University of Québec at Chicoutimi, 2017.
References
Conclusions The three additives selected improved the coke wettability by pitch. The additive Add2 at a concentration of c2 possibly represents the best choice for pitch modification among the additives
and concentrations considered. The increase in ratios A and H with the use of the additive Add2 is greater compared to those of additive Add1. This may
explain why the use of additive Add2 improves the wettability better compared to the use of additive Add1.
Anodes will be fabricated and characterized. Their density, electrical resistivity, air reactivity, and flexural strength will bemeasured to evaluate the impact of additives on anode properties.
• It was observed that all modified pitches wet the coke better than the non-modified pitch. Thus, the addition of alladditives tested improves coke/pitch interactions.
• It can also be seen that Add2 and Add3 modify the wetting behavior of pitch similarly and this behavior is slightlybetter compared to that of the pitch modified with Add1.
• The results show that contact angles decrease and thus consequently the wettability increases with increasing additiveconcentration.
• Ratios A and H were determined for measured and calculated spectra in order to evaluate whether or not the additiveinteracted with the pitch. A difference between measured and calculated results means that an interaction took placeand the pitch surface was modified due to the presence of additives.
AdditiveConcentration
(g/g)PitchName
Coking Value(%)
Total Wetting Time (s)
Non None Non-Modified 59.1 182
Add1
C1
C2
C3
Add1c1
Add1c2
Add1c3
58.75
58.5
58.1
114
115
72
Add2
C1
C2
C3
Add2c1
Add2c2
Add2c3
59
58.2
57.5
115
91
57
Add3
C1
C2
C3
Add3c1
Add3c2
Add3c3
58.8
58.26
57.7
115
106
72
Table 1: Coking values and total wetting times for differentconcentrations and different additives (for example: Add1c1 meanspitch modified with additive 1 using an additive concentration of c1)(0.5% <c1<c2<c3≤5% g/g).
Figure 1: Wettability results for : a, b) Non-modified pitch; c, d) Add1c2; e, f) Add2c2; g, h) Add3c2
at t1 = 0 s and t2 ~ 60 s
Figure 2: FTIR Spectra for non-modified pitch and modified pitches at different concentrations (c1, c2, c3) of three different additives (Add1. Add2, Add3) at room temperature.
FTIR analysis: Non-modified pitch and pitches modified with Add1 at different concentrations
FTIR analysis: Non-modified pitch and pitches modified with Add2 at different concentrations
FTIR analysis: Non-modified pitch and pitches modified with Add3 at different concentrations
Figure 3: Comparison of measured and calculated FTIR results for non-modified pitch and modified pitches with different additives (Add1. Add2, Add3) at different concentrations (c1, c2, c3) and at room temperature.
Future Work
39
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Établissement de la région de spécification de qualité multivarié pour des anodes cuites de charbon
Establishing multivariate quality specification region for baked carbon anodes
Multiple quality attributes are required for baked carbon anodes, which include both chemical properties such as air reactivity, impurities content, and physical properties as porosity and thermal shock resistance. During the last couple decades, oil refineries have had to adapt to the new market, causing not only considerable variation but an overall reduction in quality of calcined coke. Laboratory characterization of physical and chemical characteristics of all anodes manufactured at a plant is impractical as the measurements are long, expensive, and potentially not representative of the whole anode block and potential issues may occur. A fast and non-destructive test (NDT) becomes a requirement to increase the fraction of analyzed anodes and understand what is needed and can be done to attain a stable and efficient operation in smelter units. This work aims to develop baked carbon anode quality specifications based on acousto-ultrasonic and modal analyses. It will attain its objective through three specific ones. To adapt this NDT technique previously developed in a lab environment to an anode plant. Then to record quality, process, and performance data on the plant for model building. Finally, to use the model as a tool to evaluate and specify the anodes quality requirements.
Plusieurs attributs de qualité sont requis pour les anodes de charbon cuites, notamment les propriétés chimiques comme la réactivité à l’air, le contenu d’impuretés et les propriétés physiques comme la porosité et la résistivité à un choc thermique. Dans les dernières décennies, les raffineries ont eu la nécessité de s’adapter au nouveau marché, ce qui a causé une variation considérable et une réduction générale de la qualité du coke calciné. La ca-ractérisation des propriétés physiques et chimiques en laboratoire de toutes les anodes manufacturées dans une usine est impraticable parce que les mesures sont longues, chères et potentiellement non-représentatives du bloc d’anode et de ses problèmes potentiels (par exemple, des fissures). Un test non destructif (TND) rapide est nécessaire pour augmenter la fraction des anodes analysées et afin de comprendre les correctifs à apporter pour atteindre une opération stable et efficiente dans une aluminerie. Ce projet vise à développer des spécifications de qualité pour les anodes cuites basées sur une analyse modale et acousto-ultrasonique. L’objectif principal sera atteint à travers de trois objectifs spécifiques. Adapter cette technique développée dans nos laboratoires pour une application en usine. Enregistrer les données de qualité, le procédé et la performance de l’usine pour développer un modèle. Finalement, utiliser ce modèle pour évaluer et spécifier les requis de qualité des anodes.
Daniel RodriguesUniversité Laval
C. Duchesne - UL J. Tessier - Alcoa D. Picard - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Daniel Rodrigues1, Carl Duchesne1, Jayson Tessier2, Donald Picard31 Department of Chemical Engineering, Université Laval
2 Alcoa Corporation, Aluminum Center of Excellence, Aluminerie de Deschambault3 Aluminium Research Centre-REGAL, Université Laval
Establishing multivariate quality specification region for baked carbon
anodes.Établissement de la région de
spécification de qualité multivarié pour des anodes cuites de charbon.
1. Optimized System 2. Sampling Campaign 3. Model Building 4. Quality Specification
EmittersReceivers
Got Questions?Contact me on
LinkedIn→
Introduction Problem Statement Objectives
Start
Sampling Campaign
Data
SMB-PLS Model Building
Are there irrelevant Variables? YES
Eliminate Irrelevant Variables
NO
Validation of Literature and Model Results
End of Modelling
• Data cleansing;
• Model building per se;
• Elimination of noise variables;
• Cross-validation with literature.
• Sampling of baked anodes;
• Analysis before and after rodding;
• Tracking and monitoring their behavior.
• Evaluate frequency-time domain techniques;
• Simulate conditions where not all data is available;
• Analyze signal echoes.
• Random sampling of all variables not related to the anode quality;
• Combination of anode latent space with historical data;
• Determination of “stable operation” region using machine learning.
Receivers’ possible positions
range
“stable operation” zone
“unstable operation” zone
We can enhance the system by: Using CWT; Analyzing the echoes; Changing the modulated signal;
The method can deal with defects positioned far from the measurement point;
Next step is to correlate them to CT-scan data.
Model Building Workflow. Example of a resulting contour plotfor a performance indicator.
Ultrasound signal moving from emittersto receivers in an anode.Acousto-Modal measurement device enhancement.
• ‘Bump’ continuous wavelet
transform filter bank better
isolates frequency bands and
does not lose time resolution.
• Even if the defects are not
precisely under the measurement
device, they are still detectable
with other measurements data;
• It is difficult to differentiate
the echoes data from the
main signal because of its
modulation.
Acknowledgements
• Pursue Acousto-Ultrasonic and modal non-destructive testing (NDT);
• Build a model to predict anode performance based on (NDT);
• Account for process variability on the quality specifications.
• Raw materials quality has generally decayed and increased variability;
• Actual anode quality assurance is based on sampling 2% of anodes;
• Testing procedures are slow, and not representative of the whole anode.
Analyze fast and effectively
Air Permeability
ThermalConductivity
ApparentDensity
CO2
Electricalresistivity
X-Ray Fluorescence
Other Analyses
• Multiple Anode properties affect anode performance and overall process efficiency;
• Lack of anode quality affects production performance in three different ways:
Energy consumption Carbon consumption Manpower;
• It has been shown possible to analyze them using Acousto-ultrasound and modal methods.Anode uneven consumption caused by defects.
Anode sampling simplified diagram.
‘db5’ DWT filter banks in 1MHz sampling. ‘bump’ CWT filter banks in 1MHz sampling. Example modulated signal used in previous Works.
-5
0
5
10
0 10 20 30
Cla
ssifi
er
Anode Number
Classifier Prediction
Classifier prediction without stub hole measurements
The authors would like to thank the support ofAlcoa Canada, the National Sciences andEngineering Research Council of Canada(NSERC), the Fonds de Recherche du Québec –Nature et Technologies (FRQNT), and the REGAL.
Methodology
Preliminary results – Focus on Step 1 Conclusions and future work
Prix Award
4040
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Analyse thermique de l’injection de poudre refroidit dans un liquideThermal analysis of the injection of cooled powder in a liquid
L’industrie de l’aluminium cherche à optimiser l’injection d’alumine dans la cuve d’électrolyse pour accroître la productivité et baisser les coûts. L’effet thermique a un impact important sur la performance de l’injection (formation de bain solidifié, flottaison du radeau, dispersion des particules). L’expérience décrite sur ce poster représente une analyse analogue simulant l’injection de particules dans l’eau. L’utilisation d’une méthode ana-logue permet de faciliter la conception et l’observation, contrairement aux conditions difficiles dans les cuves. L’expérimentation consiste à l’injection de particules, refroidies à l’azote liquide, dans l’eau. Pour cela, plusieurs conditions d’états des particules et d’eau ont été étudiées telles que les températures du liquide et des grains injectés, le type d’injecteur et le fait que les particules aient été mouillées préalablement, ou non, avec de l’azote. Des thermocouples insérés sous la surface du liquide permettent de mesurer l’évolution du refroidissement du liquide induit par l’injection des particules froides. L’expérience permet d’observer une injection similaire aux injections d’alumine dans les cuves d’électrolyse. La formation de glace sur les particules en frontière du radeau peut être observée. L’expérience montre l’impact des différentes conditions sur la forme du radeau, de sa température ou du nombre de particules immergées directement lors de l’impact à la surface du liquide.
The aluminum industry is seeking to optimize the injection of alumina into the electrolytic cell in order to in-crease productivity. The thermal effect has a significant impact on the performance of the injection (frozen bath formation, dispersion of the particles). The experiment described on this poster represents an analogue study, representing the injection of particles into a liquid. The analogue method facilitates the design and observation, compared to the critical conditions in the industrial cells. The experiment involves the injection of particles, pre-cooled by liquid nitrogen, into the water. Several conditions of the particles and water were studied such as the temperature of the liquid and the injected grains, the type of injector, particles wetted and non-wetted by nitrogen. Thermocouples inserted below the surface of the liquid made possible to follow the cooling of the liquid by the injected cold particles. The experiment permitted to observe an injection similar to the industrial process. Ice formation on the particles at the raft boundary could be observed. The experiment showed the impact of the different conditions on the shape of the raft, on its temperature history and on the fraction of particles sunk directly during the impact on the liquid surface.
Thomas RogerUniversité du Québec
à Chicoutimi
L. I. Kiss - UQAC S. Poncsák - UQAC
K. Fraser - CNRC-CTA J-F. Bilodeau - Rio Tinto
S. Guérard - Rio Tinto
Journée des étudiants – REGAL8 et 9 octobre 2019
Thomas Roger1, László I. Kiss1, Sándor Poncsák1, Kirk Fraser2, Jean-François Bilodeau3 et Sébastien Guérard3
1 Université du Québec à Chicoutimi, Saguenay2 Centre des technologies de l'aluminium, Conseil national de recherches Canada, 501 boulevard
de l’Université, Chicoutimi, QC, Canada, G7H 8C33 Centre de recherche et développement Arvida, Rio Tinto Alcan, Saguenay
Analyse thermique de l’injection de poudre refroidit dans un
liquideThermal analysis of the injection
of a cooled powder in a liquid
Partenaires et remerciementsNous remercions les partenaires pour leur confiance et leur support financière.
ConclusionCette affiche montre le développement d’un montage analogue pour reproduire lesradeaux d’alumine dans les cuves d’électrolyse. L’imagerie par la méthode schlierenpermet d’observer les gradients de température dans l’eau suite à l’injection departicules refroidis. Cette méthode met en évidence que le volume perturbéthermiquement par le radeau est très faible et que la couche limite est finecomparée à l’épaisseur du radeau.
IntroductionDans l’objectif d’améliorer l'injection de l’alumine dans les cuves d’électrolyse,l'industrie de l'aluminium cherche à comprendre tous les mécanismes présents(transfert thermique, déplacement des particules, dissolution). Dans ce poster,une méthode analogue est présentée pour observer la zone perturbéethermiquement par l’injection d’un matériaux granulaire dans un liquide.L’utilisation de capteurs a l’inconvénient d’influer sur les écoulements desparticules et du liquide. Pour cela, l’imagerie avec la méthode de schlieren estutilisée pour observer les gradients de température dans le liquide.
Méthodologie- Injection de particules organique refroidit dans de l’azote liquide à une température avoisinant les -180°C dans de l’eau à température ambiante- Observer les gradients horizontaux et verticaux de température sous le radeau à l’aide de l’imagerie par la méthode schlieren.
Zone de test
Source de lumière
Caméra
Filtre de couleur
Miroir
Miroir
Faisceaux de lumière
Fente
Particules organiques
Entonnoir
Eau
L’imagerie par la méthode schlierenLa méthode schlieren est basée sur la réfraction de la lumière engendrée parune variation de la densité d’un fluide transparent. La densité peut varier enfonction de la température ou de la concentration d’un soluté dans le fluide.Notre étude porte sur les échanges thermiques entre des particules et de l’eau,donc la méthode schlieren permettra d’observer les gradient de température.
La lumière est projetée dans un premier miroir pour rendre tous les faisceauxlumineux parallèles les uns aux autres. Après la zone de test, les faisceaux sontfocussés sur un filtre grâce à un deuxième miroir. La diffraction de la lumièreest observée grâce au filtre constitué de fine tranche de différentes couleurs.Une caméra est placée en arrière de celui-ci pour capturer l’image. Le filtrepeut-être placé de façon horizontale pour voir les gradients de températuresuivant y et verticale pour ceux suivant x.
La zone de test est composée d’un réservoir de plexiglass rempli d’eau àtempérature ambiante. Les particules sont refroidies dans un contenant plongédans de l’azote liquide pour atteindre une température avoisinant les -180°C.Elles sont par la suite injectées dans l’eau à l’aide d’un entonnoir.
RésultatsAfin d’obtenir les deux composantes de gradients de température(horizontale et verticale), il est nécessaire de répéter l'expériencedeux fois. Malheureusement les injections consécutives ne sont pasnécessairement identiques comme le montre la forme de l’ombredu radeau. Mais des points similaires sont rencontrés comme lesparticules libres qui coulent directement suite à l’injection.
Le filtre verticale montre que le gradient de température horizontalest très faible. Cette position du filtre permet d’observer une trainéederrière les particules coulées durant une partie de leur chute. Cecisignifie que leur température n’a pas atteint celle du liquide.
Le filtre horizontale montre la couche limite thermique en dessousdu radeau. Malgré un forte variation de température entre lesparticules injectées et le liquide (200°C), l’épaisseur de la couchelimite reste inférieur à l’épaisseur du radeau. Après un certaintemps, nous pouvons observer le décrochage de cette couche limite(Lignes horizontales sur la dernière photo ).
Temps après
injection
1s
2s
3s
10s
Filtre vertical*dT/dx
Filtre horizontal*dT/dy
* Les deux injections sont différentes
x
y
41
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
NÉtude des effets de la distribution granulométrique sur le comportement mécanique des agrégats de coke via la méthode des éléments discrets
Investigation of particle size distribution effects on the mechanical behavior of coke aggregates via the discrete element method
Understating and predicting the final properties of carbon anodes are of great importance in the production of consistently high-quality anodes. Anodes are composed of calcined coke particles, recycled anode butts, and coal-tar pitch, as a binder. The size distribution of the coke particles has direct effects on the density and the mechanical properties of anodes. The aim of this work is to model the effect of the size distribution of coke par-ticles on the compaction behavior of coke aggregates. Discrete Element Method (DEM) was used, as a simulation tool, to reveal the behavior of dry coke aggregates during the compaction process with different strain rates and confining stresses. The results show that the overall mechanical behavior of carbon coke aggregates is governed by big particles. Moreover, it can be deduced that strain rate and confining pressure have significant effects on the homogeneous or inhomogeneous deformation of dry coke aggregates.
La compréhension et la prévision des propriétés finales des anodes en carbone sont d’une grande importance pour la production d'anodes de haute qualité. Les anodes sont composées de particules de coke calcinées, d’anodes recyclées et de brai de goudron, servant de liant. La distribution de taille des particules de coke a des effets directs sur la densité et les propriétés mécaniques des anodes. Le but de ce travail est de modéliser l'effet de la distribution de taille des particules de coke sur le comportement en compactage des agrégats de coke. La méthode des éléments discrets (DEM) a été utilisée comme outil de simulation pour expliquer le comportement des agrégats de coke sec pendant le processus de compactage avec différents taux de chargements et de contraintes de confinement. Les résultats montrent que le comportement mécanique global des agrégats de coke de carbone est régi par les grosses particules. De plus, on peut en déduire que le taux de chargement et la pression de confinement ont des effets importants sur la déformation homogène ou non homogène des agrégats de coke secs.
AlirezaSadeghi-ChahardehUniversité Laval
R. Mollaabbasi - UL D. Picard - UL D. Ziegler - AlcoaS. M. Taghavi - ULH. Alamdari - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Modeling of carbon anode paste behavior via discrete element method
Modélisation du comportement de la pâte d'anode en carbone par la méthode des éléments discrets
A. Sadeghi Chahardeh1,2, R. Mollaabbasi1,2, D. Picard2, D. Ziegler4, S. M. Taghavi1,3, H. Alamdari1,2
1 Department of Mining, Metallurgical, and Materials Engineering, Université Laval, Québec, QC, CANADA2 NSERC/Alcoa Industrial Research Chair MACE3 and Aluminium Research Centre- REGAL, Université Laval, Québec, QC,
CANADA3 Department of Chemical Engineering, Université Laval, Québec, QC, CANADA
4 Alcoa Primary Metals, Alcoa Technical Centre, 859 White Cloud Road, New Kensington, PS, 15068, USA
• The density and homogeneity of carbonanodes have effects directly on theperformance and economy of thealuminum production.
• Any physical defects within the anodeblocks, such as cracks or heterogeneityin density, may adversely affect theanode performance, and consequentlythe performance of the reduction cells.
• Investigation the of compaction of the anode paste aggregate during thevibro-compaction process.
• Determination of the reasons of the crack generation in the stub-holeregion.
Figure 1. Pre-baked carbon anode [1]
Figure 2. Horizontal cracks in the baked carbon anodes [2]
Figure 3. Compaction band, characterized by vertical dimension shortening
• Strain localization can generate an area in which tensile stresses haveexisted after the external load is removed.
• This area is susceptible to crack generation.
Blue indicates tensile stresses Red indicates compression stresses
1. Determination of Representative Volume Element (RVE)• Optimum number of particles
2. Simulation of compaction process
• Discrete Element Method is used to simulate the anode paste behavior.
1. http://www.jncarbon.com/en/product/162738_for_detail.htm2. M. B. Boubaker, et al., Non-destructive testing of baked anodes based on modal analysis and principal
component analysis, Light Metals, Springer, (2017), 1289-1298.3. B. Majidi, et al., Packing density of irregular shape particles: DEM simulations applied to anode-grade
coke aggregates, Advanced Powder Technology, 26 (2015), 1256-1262.
• DEM simulation will be done to capture the strain localization during thecompaction process.
• Performing experimental tests to prove the numerical results.
1. Determination of the RVE
3000 particles2000 particles
1000 particles500 particles
Figure 4. Effect of number of the coke particles on the stress-strain diagram(Radius=5mm, Boundary= periodic, Isotropic pressure=100 kPa, Strain rate= 0.1 )-1s
2. Strain Localization Analysis
Figure 6. Mono-size RVE with 2000 particle and solid boundary condition(Radius=5mm, Confining pressure=100 kPa, Strain rate= 0.1 )-1s
Figure 5. The maximum error of the stress-strain fluctuation versus D/d for the different RVEs
(%)
4242
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Usure par piqûres sur les cathodes dans les cellules d'électrolyse en aluminiumPitting wear on cathodes in aluminium electrolysis cells
L’étude de six brasques usagées, opérés avec des cathodes de différents grades de blocs de carbone, ampé-rages et technologies de cellules a été menée dans trois différentes fonderies pour révéler les mécanismes causant l’usure du carbone de la cathode. La microstructure des échantillons de cathodes autopsiées a été étudiée par microscope optique et électronique ainsi que par tomographie par rayons X. Les résultats obtenus ont montré que la surface de la cathode est caractérisée par un motif d’usure par piqûres et il a été discuté que ce motif d’usure par piqûres sur les anodes de carbone dans les cellules d’électrolyse résulte d’un processus d’usure chimique ou électrochimique.
Investigations of six spent pot linings operated with different carbon cathode block grades, amperage regimes and cell technologies were conducted at three separate smelters to reveal the mechanisms causing carbon cath-ode wear. The microstructure of the cathode samples from the autopsies was investigated by optical and electron microscopy as well as X-ray computed tomography. The present findings revealed that the cathode surface was characterized by a pitting wear pattern and it is discussed that the pitting wear pattern on carbon cathodes in Al electrolysis cells results from a chemical/electrochemical wear process.
Samuel Senanu SINTEF
Z. Wang - SINTEFA. P. Ratvik - SINTEF
T. Grande - NTNU
Journée des étudiants – REGAL8 et 9 octobre 2019
Investigations of six spent potlinings operated with different carbon cathode block grades, amperage regimes and cell technologies were conducted at three separate smelters to reveal the mechanisms causing carbon cathodewear. The microstructure of the cathode samples from the autopsies was investigated by optical and electron microscopy and X-ray computed tomography. The present findings revealed that the cathode surface is characterizedby a pitting wear pattern and it is discussed that the pitting wear pattern on carbon cathodes in Al electrolysis cells results from a chemical/electrochemical wear process.
Carbon cathode wear is an important issue for the primary aluminium industry due to its crucial role indetermining the service life of the electrolysis cells used in the production of Al [1]. Autopsy observations ofspent potlinings show that the highest wear is located at the cathode block ends [2]. Figure 1 shows a laserscanning image revealing the locations with highest wear (red spots).
The wear at the locations with highest wear is reported to be characterized by pitting wear [3]. This posterdiscusses the occurrence and possible mechanism(s) behind the pitting wear on the carbon cathode surface.
Six spent potlinings were autopsied to understand the mechanisms behind cathode wear including the pittingwear pattern. The wear patterns were characterized by means of photography and wear measurements including3D optical profilometry. Microstructures at the wear surface as well as within carbon autopsy samples wereanalysed by means of optical microscopy, SEM and CT scanning. Table 1. summarizes the pots autopsied.
Pitting wear was observed on cathode surface of all pots. A relatively uniform wear surface, presence of aggregates within carbon matrix at pitting locations and a wear surface indicative of a chemically etched process suggest pitting is a chemical/electrochemical wear process.
Abstract
Introduction
Experimental
Results and Discussion
Conclusion
Figure 2. Cathode surface of pot 3. Highlighted image showing pitting at the upstream and downstream ends of a spent potlining.
Samuel Senanu 1, Zhaohui Wang 1, Arne Petter Ratvik 1 and Tor Grande 2
1)SINTEF 2)NTNU
Pitting Wear on Cathodes in Al Electrolysis Cells
Figure 1. Wear pattern across a spent potlining showing high wear (red spots) at cathode block ends.
Pot Technology Amperage (kA)
Cathode Current Density(A/cm2)
Pot Age (days)
Carbon Type Apparent density (g/cm3)
1 Prebaked- Side by Side
313 0.8 2461 Graphitized and Imp.
1.73±0.03
2 Prebaked- End to End
175 0.8 3154 Graphitic (100 % Gr.)
1.58
3 Prebaked- Side by Side
313 0.8 1731 Graphitized (HD & iso.)
1.78
4 Prebaked- End to End
175 0.8 2849 Graphitic (100 % Gr.)
1.65±0.03
5 Prebaked- End to End
175 0.8 1028 Graphitic (100 % Gr.)
1.61
6 Søderberg- End to End
128 0.6 2732 Semi-Graphitic (30 % Gr.)
1.53
Table 1. Summary of pots autopsied showing amperage, carbon type and some properties of cathode blocks
Figure 3. Images of the wear surface. a, b and c. Optical micrographs. d. CT image.
Results from the prebaked pots showed the same trends reported by other authors [1,2,3]. The highest wear wasobserved close to the cathode block ends while the centre channels areas showed the least wear. The Søderbergpot on the other hand showed a relatively uniform wear. Close observation of the locations with highest wearsuch as the cathode block ends and pothole locations of all the autopsied pots revealed a pitted wear surface.Figure 2 is a photograph of the cathode surface showing pitting wear at the upstream and downstream ends.
Results and Discussion
d
[1] M. Sørlie, H.A. Øye. Cathodes in aluminium electrolysis, 3rd ed., Aluminium-Verlag, Düsseldorf, 2010.
[2] E. Skybakmoen, et.al."Measurement of Cathode Surface Wear Profiles by Laser Scanning,' Light Metals 2011.
[3] A.T. Tabereaux, et.al."Erosion of cathode blocks in 180 kA prebake cells," Light Metals 1999.
• .
Reference
Acknowledgement
Contact Information Samuel Senanu (PhD student)Research Scientist Sintef Industry, Metal Production and [email protected]
Microscopic analysis of the pitted wear surface revealed that the aggregate and binder matrix are worn outuniformly with no indications of particle detachment. Figure 3 show images of the wear surface. Highmagnification images (Figures 3b and 3c) suggest a chemical etching or wear process.
43
AX
E 1 : P
RO
DU
CTIO
N D
E L’A
LU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
NÉtude de la microstructure et des propriétés mécaniques des pâtes
de ciment avec LCLLInvestigation on microstructure and mechanical properties
of cement pastes made with LCLL
Spent Pot Liner (SPL) is an industrial by-product generated from aluminium electrolysis cell of which disposal is an industrial challenge. In particular, LCLL ash is the inert by-product of the hydrometallurgical process de-veloped at the Arvida Research and Development Centre (Québec, Canada) called “Low Caustic Leaching and Liming”. This research project aims to valorize the LCLL transforming into a safe cementitious additive. LCLL ash has been ground to the necessary fineness to substitute a part of cement in cement pastes. This study will present investigation of the microstructure and micromechanical properties of LCLL ash blended cement pastes by multiple techniques including X-ray diffraction and micro-indentation at the level of the cement paste. The water to binder ratio is 0.35. The effect of curing ages and the different dosage of LCLL on the microstructure and mechanical properties of blended cements are presented and discussed with relation to ordinary Port-land cement.
Les brasques traitées sont un sous-produit industriel issu d’une cellule d'électrolyse de l'aluminium dont leur élimination représente un défi industriel. En particulier, LCLL est le sous-produit inerte du procédé hydrométal-lurgique développé au Centre de Recherche et Développement d'Arvida (Québec, Canada) appelé Low Caustic Leaching and Liming. Ce projet de recherche vise à valoriser les brasques traitées en les transformant par un ajout cimentaire sécuritaire. LCLLL a été broyé à la finesse nécessaire pour remplacer une partie du ciment dans les pâtes de ciment. Cette étude porte sur l'étude de la microstructure et des propriétés micromécaniques des pâtes de ciment mélangées avec LCLL par des techniques incluant la diffraction des X-rayons et la micro-inden-tation. Le rapport eau/liant est de 0,35. Les effets de la durée de cure et ainsi que le pourcentage de LCLL sur les propriétés mécaniques et la microstructure des ciments mélangés sont présentés et discutés en relation avec le ciment Portland ordinaire.
Thi Hang TranUniversité Laval
V. Brial - ETS T. Sanchez - UL L. Sorelli - UL C. O. Plamondon - ETSD. Conciatori - UL H. D. Alamdari - ULM. Fafard - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Hang T.TRAN1,3, Victor BRIAL2, Thomas Sanchez1, Luca SORELLI1, Claudiane O. PLAMONDON2, David Conciatori1, Houshang D.Alamdari1,3, Mario FAFARD1,3
1Université Laval, Québec, QC, Canada2École de technologie supérieure, Montréal, QC, Canada
3Aluminium Research Centre – REGAL, Québec, QC, Canada
Investigation on microstructure and mechanical properties of cement pastes made
with LCLL-SPLÉtude de la microstructure et des propriétés
mécaniques des pâtes de ciment avec SPL-LCLL
Problematic• Spent potlining (SPL) is an industrial by-product generated from aluminium electrolysis cells, which is classified as a hazardous waste
and no acceptable landfilling• About 20 kt of SPL is generated per year from Rio Tinto [1]• After treatment by “Low-Caustic Leaching and Liming” process, SPL become an inert by-product, called SPL-LCLL in this research,
which could be reused in the cement industry [1]• SPL-LCLL contains higher contents of silica and alumina compared to Portland cement → the final composition, the morphology and
the mechanical properties of the binder hydrates (e.g. the calcium-[aluminum]-silicate hydrates) with important effect on thedurability of concrete
Research objectives Characterization the physic-chemical of SPL-LCLL Study microstructure and mechanical properties of the blended cement pastes
with SPL-LCLL
http://www.regainmaterials.com/aluminium-smelting-industry/spent-potlining-challenges
Methodology X-ray fluorescence (XRF), X-ray diffraction (XRD), Scanning Electron
Microscopy (SEM) Micro-indentation technique
Conclusions
- The optimal proportion of SPL-LCLL in blended pastes mixed is approximately 10% to
achieved the highest strength and stiffness at 28 days of curing
- SPL-LCLL can be used as “a filler” in this case. It provides enhanced material properties of
SPL-LCLL to optimise it into concrete mix designs
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O TiO2 P2O5 L.O.I36.55 35.81 7.99 3.04 0.38 0.07 7.98 0.76 0.11 6.44
Name of samplesCement (wt.%)
LCLL(wt.%)
Anhydrite (wt.%)
Control-OPC 100 0 010LCLL 90 10 020LCLL 80 20 010LCLL-1.25AH 88.75 10 1.2520LCLL-2.5AH 77.5 20 2.5
Materials and mixture- SPL-LCLL provided from Rio Tinto- Cement was from Cement Quebec (cement type GU)- Anhydrite (CaSO4) was used as an additive material- The w/b ratio = 0.35
Results and discussions
Table 1. Details of mix proportions
Figure 2. SEM picture of SPL-LCLL powder
Figure 1. XRD spectrum of SPL-LCLL and a slag
Table 2. Chemical composition cement and LCLL by XRF
Figure 3. X-ray diffraction pattern of pastes at 28 days
Figure 4. Micromechanical properties of the blended cement pastes
Figure 4. Schematic view of (1) the indentation instrument, (2) P-h curve, (3) Indenter imprints
References[1]L. Birry, S. Leclerc, and S. Poirier, “The LCL&L Process: A Sustainable Solution for the Treatment and Recycling of Spent Potlining,” in Light Metals 2016, E. Williams, Ed. Cham: Springer International Publishing, 2016, pp. 467–471.
TRANSFORMATIONS ET APPLICATIONSTRANSFORMATION AND APPLICATIONS
AXE 2
Axe | Axis 2Répertoire des affiches | Posters directory
Les étudiants dont le nom est suivi d’un astérisque (*) sont récipiendaires d’un prix d’excellence pour leur affiche. Students whose name is followed by an asterisk (*) are recipients of an award of excellence for their poster.
Agbe, Henry Revêtements nanocomposites à base de Silver sur aluminium dotés de propriétés antibactériennes et antifouling ajustables 46
Ahmed, Mohamed Effet des éléments d'alliage sur les propriétés mécaniques et chimiques des alliages d’apport 4xxx pour le soudage de l'aluminium 47
Algendy, Ahmed L’évolution des dispersoïdes lors du traitement thermique des alliages Al-Mg avec différentes teneurs en Mg 48
Attia, Mohamed Étude de la conception et le traitement thermique d’un bras de suspension d’automobile en alliage d’Aluminium A357 Semi-solide 49
Beudon, Camille Conception et analyse économique du cycle de vie d’un pont poutres en bois/ platelage en aluminium 50
Chen, Zhixing La calorimétrie à balayage différentiel (DSC) et sa caractérisation des états de traitement thermique dans les alliages extrudés AA6082 51
David, Guillaume Conception d’un banc de mesure de la friction appliqué au procédé HSBF (High Speed Blow Forming) 52
Dhillon, Jaskaranpal Singh Fabrication additive d'alliage d'aluminium F357 par fusion sur lit de poudre au laser 53
Djedid, Amar Conception d’un prototype de platelage en aluminium pour les ponts aluminium / bois 54
Elashery, Ali* Effet du traitement de l’homogénéisation sur l'évolution des dispersoïdes Zr dans les alliages Al-Mg-Si avec deux teneurs en Mg et Si 55
Espiritu, Eileen Ross L.* Absorption d’humidité par de la poudre d’AlSi10Mg et effet sur son écoulement 56
Farid, Redouane* Développement des revêtements non chromatiques écologique et mécaniquement résistant pour la protection de l’aluminium contre la corrosion 57
Fu, An* Application de la calorimétrie à refroidissement rapide à la fabrication additive d'alliages d'aluminium 58
Grandmont, Philippe Influence des paramètres de soudage par friction malaxage sur les propriétés mécaniques locales et la microstructure des alliages Al-Mg 59
Jaberi, Ali Résistance en fatigue des pièces en alliage d’aluminium produits par fusion sélective laser 60
Luiz Felipe, Leitao, Martins Résistance en fatigue des pièces en alliage d’aluminium produits par fusion sélective laser 61
Li, Cong L’évolution de la conductivité électrique dans les composites Al-B4C avec différentes teneurs en B4C 62
Li, Dong L’influence des éléments de transition (V, Zr et Mo) et de la vitesse de refroidissement sur l'évolution des dispersoïdes dans l'alliage de fonderie Al-Si 356 63
Lorans, Anaïs L’influence de la texture cristalline et de la taille des grains sur l’atténuation des ondes ultrasonores dans des échantillons d’aluminium 6061-F et 6061 HIP+T6 produits par énergie concentrée (DED) 64
Javidikia, Mahshad Usinage à grande vitesse des alliages d'aluminium AA6061-T6 : Influence du rayon d'arête de l'outil sur les caractéristiques d'usinage 65
Mofarrehi, Mohammadreza Le rôle de la composition eutectique des alliages 5xxx dans la fissuration des bords induite par le sodium au cours du processus de laminage à chaud 66
Nodeh, Marzieh Évaluation expérimentale et numérique de joints adhésifs en aluminium à simple recouvrement pour l’industrie automobile 67
Nommeots-Nomm, Amy Fusion sur lit de poudre de l'alliage Al-Cu-TiB2 alloy 68
Plante, Justin* Caractérisation du comportement en extrusion d’un alliage d’aluminium expérimental de la série 6xxx destiné aux applications haute-performance de l’industrie automobile 69
Reichert, Emma Sue* Comparaison de l'écoulement de la poudre AlSi10Mg avec l’entonnoir Hall/Carney et le GranuFlow 70
Qassem, Mohamed Étude des propriétés à haute température des alliages 6111 coulés par refroidissement direct en ce qui concerne les phénomènes de la fissuration à chaud 71
Rakhmonov, Jovic L’effet de l’amélioration des caractéristiques des dispersoïdes sur la microstructure, la texture et les propriétés mécaniques à température ambiante / élevée des alliages extrudés 6082 72
Robitaille, Bénédicte* Impact du brossage d’oxydes de surface sur les propriétés mécaniques de joints à recouvrement en aluminium 2024 soudés par friction-malaxage 73
Salehi, Pouria Propriétés en fatigue des joints soudés au laser continu à fil froid en AA6005-T6 74
Tremblay, Maude L’effet de l’azote comme gaz transporteur lors de la fabrication additive par dépôt sous énergie dirigée d’aluminium 75
Trimech, Mahmoud Optimisation de la fatigue et contrôle-qualité des joints soudés par friction malaxage dans les platelages des ponts routiers en aluminium 76
Tumulu, Satish Kumar* Fabrication additive de AA2219 à l'aide du procédé de fusion par lit de poudre au laser 77
Voisard, Frédéric Analyse d’alliage d’aluminium par spectroscopie en perte d’énergie d’électron à 30keV 78
Wang, Shuai Effet de l'ajout de Mo sur le comportement à la fatigue thermomécanique de l'alliage de fonderie Al-Si 319 79
Wawrzyniak, Nicolas* Méthodologie de caractérisation locale en traction monotone à l’aide d’échantillons miniatures 80
4646
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
NRevêtements nanocomposites à base d’argent sur aluminium dotés de propriétés antibactériennes et antisalissure ajustablesSilver-based nanocomposite coatings on aluminium with tunable antibacterial and anti-fouling properties
La formation de biofilms sur des surfaces animées et inanimées constitue un réservoir bactérien idéal pour la propagation des infections nosocomiales. Par an, plus de 200 000 patients contractent des infections nosoco-miales au Canada, dont 8 000 en décèdent. Par conséquent, la fabrication de surfaces dotées de propriétés antisalissure et antibactériennes ajustables peut aider à réduire le biofilm de fixation bactérienne et conséquem-ment, à la propagation ultérieure d'infections nosocomiales. Dans ce travail, un nanocomposite à base d'argent à matrice polymère doté de propriétés antisalissure et antibactériennes ajustables a été obtenu. Le polymère était chargé avec une certaine quantité de nanomatériaux à base d’Ag et soumis aux ultrasons pour obtenir une dispersion uniforme, suivi d’un dépôt par étirage vers le haut sur un alliage d’aluminium. L'aluminium revêtu de nanocomposites d'argent a été caractérisé par des mesures au SEM, EDS, XRD, FTIR et à l'angle de contact. Le nanocomposite à base d’argent tel que synthétisé présentait une excellente propriété antibactérienne avec des valeurs de concentration inhibitrice minimale (CMI) de 6,25 μg / mL, 6,25 μg / mL et 3,125 μg / mL pour Staphylococcus aureus (gram +), Pseudomonas aeruginosa (gram -) et Escherichia coli (gram -), respectivement. En outre, l'aluminium revêtu de nanocomposites à base d'argent présente une propriété antisalissure avec des taux de réduction d'adhérence de 99,92 %, 99,98 % et 100% pour Staphylococcus aureus, Pseudomonas aeruginosa et Escherichia coli, respectivement. Cet aluminium recouvert de nanocomposites à base d'argent est idéal pour une utilisation en tant que surface tactile antimicrobienne en milieu hospitalier afin de réduire la prévalence des infections nosocomiales.
Biofilms formation on animate and inanimate surfaces provide perfect bacterial reservoir for the spread of noso-comial infections. More than 200,000 patients contract nosocomial infections in Canada per year, of which 8,000 result in death. Therefore, fabricating surfaces with tunable antibacterial and anti-fouling properties can help reduce bacterial attachment on biofilm and subsequent spread of nosocomial infections. In this work, a polymer matrix silver-based nanocomposite with tunable antibacterial and anti-fouling properties has been achieved. The polymer was loaded with certain amount of silver-based nanocomposite and sonicated for uniform dispersion, followed by upward drawing deposition on aluminium alloy. The silver-based nanocomposite coated aluminium has been characterised by SEM, EDS, XRD, FTIR and contact angle measurements. The as-synthe-sized silver-based nanocomposite showed excellent antibacterial property with minimum inhibitory concentration (MIC) values of 6.25 µg/mL, 6.25 µg/mL and 3.125 µg/mL for Staphylococcus aureus (gram +), Pseudomonas aeruginosa (gram -) and Escherichia-coli (gram -) respectively. Furthermore, the silver-based nanocomposite coated aluminium exhibits anti-fouling property with an adhesion reduction rate of 99.92 %, 99.98% and 100 % for Staphylococcus aureus, Pseudomonas aeruginosa and Escherichia-coli, respectively. This silver-based nano-composite coated aluminium is ideal for use as antimicrobial touch surfaces in hospital environment for reducing the prevalence of nosocomial infections.
Henry AgbeUniversité du Québec
à Chicoutimi
D. K. Sarkar - UQAC X. - G. Chen - UQAC
N. Faucheux - Université de Sherbrooke
G. Soucy - Université de Sherbrooke
J. L. Bernier - A3 Surfaces
Journée des étudiants – REGAL8 et 9 octobre 2019
Silver-based nanocomposite coatings on aluminium with antibacterial and anti-
fouling properties
Revêtements nanocomposites à base de Silversur aluminium dotés de propriétés
antibactériennes et antifouling
H. Agbe 1, D. K. Sarkar 1, X. - G. Chen 1, N. Faucheux 2 , G. Soucy 2 , J. L. Bernier 3
1 Department of Applied Science, University of Québec at Chicoutimi2 Department of Chemical and Biotechnological Engineering, Université de Sherbrooke, 3 A3 Surfaces, Chicoutimi
IntroductionHealthcare associated
Infections (HAI)
An infection acquired in a hospital, that was not present or incubating prior to the patient being admitted to the hospital
Bacterial strain of importance in HAI:o Clostridium difficile;o Vancomycin Resistant Enterococci (VRE) –Escherichia colio Methicillin Resistant Staphylococcus aureus (MRSA)o Carbapenemase-Producing Gram (-ve) Bacilli- Pseudomonas
aeruginosa
Objectiveso Fabricate silver-based nanocomposite polymeric coatings on aluminium
using simple dip coating deposition process.
o Evaluate the antibacterial and antifouling properties of the silver-based nanocomposite polymeric coated aluminium on clinically relevant HAI bacteria.
MethodsSamples Ag3PO4/Polymer (w/v%)
1 0.52 1.03 1.54 2.05 2.56 3.07 3.58 4.09 4.5
Ag3PO4 was loaded in Ultra-Ever Dry polymeric top coatings, followed by dip coating on
aluminium.
MBC Determination Bacterial adhesion Test
Results and DiscussionSEM EDS and XRD Characterisation ATR-FTIR Characterisation
Fig.1: (A) SEM of As-received Aluminium; (B) SEM micrographs of 0.5 w/v. % Ag3PO4 Ultra-Ever Dry polymer coated Aluminium; (C) EDS of Ag3PO4 Ultra-Ever Dry polymer coated Aluminium; (D) XRD of as- synthesized Ag3PO4. Inset: Contact angles of (A) and (B) respectively.
Fig. 2: ATR−FTIR spectra of: (I) Ultra-Ever Dry bottom coating; (II) 0 w/v % Ag3PO4 loading in Ultra-Ever Dry top Coating; (III) 0 w/v % Ag3PO4 loading in Ultra-Ever Dry top coating + bottom
coating; and (IV) 0.5 w/v % Ag3PO4 loading in Ultra-Ever Dry top coating + bottom coating.
Conclusiono The Ag3PO4 Nanocomposite coated Al
demonstrates an antibacterial and anti-fouling properties.
o Ag3PO4 nanocomposite shows excellent antibacterial property with MBC values of 6.25 ppm, 6.25 ppm and 3.125 ppm against S. aureus, P. aeruginosa and E-coli respectively.
o Anti-fouling property with bacterial adhesion percent reduction of 99.02 %, 99.00% and 99.99999% for S. aureus, P. aeruginosa and E-coli respectively.
o The Ag3PO4 nanocomposite coated aluminium is ideal for engineering touch surfaces for reducing the spread of Healthcare Associated Infections (HAI).
o Minimum Bactericidal concentration(MBC) is the “gold standard” inantimicrobial susceptibility testing (AST).
o The MBC values falls within theacceptable range (0.125 - 32 ppm) perClinical and Laboratory StandardsInstitute (CLSI) standards.
o Silver ions at low concentration (10 ppm)are known to exert biocidal effect withoutadverse effect on human cells.
Minimum Bactericidal Concentration Results Bacteria MBC μg/mL (ppm)
S.A 6.25
P.A 6.25
E-Coli 3.12
Future workFuture work will focus on stability, durabilityand toxicity study of the silver-basednanocomposite polymeric coated aluminium.
Bacterial adhesion ResultsBacteri
a
Bacteria Colony
on Uncoated
(CFU)
Bacteria Colony on 0.5 w/v
Ag3PO4 coated Al
S.A Initial number
[mL]
1.33 x 107 1.0 x 105
Reduction [%] - 99.02
P.A Initial number
[mL]
3.3 x 106 6.7 x 104
Reduction [%] - 99.00
E-Coli Initial number
[mL]
1.1 x 107 0.00
Reduction [%] - 99.99999
47Effet des éléments d'alliage sur les propriétés mécaniques et chimiques
des alliages d’apport 4xxx pour le soudage de l'aluminiumThe effect of alloying elements on the mechanical and chemical properties
of 4xxx filler alloys for aluminum welding
4xxx aluminum welding wires are widely used for welding most of aluminum alloys, especially 6xxx series, owing to excellent weldability, high hot cracking resistance, low melting points, and capability of strength modification by making it heat treatable and/or dispersion strengthened. The strength of these filler alloys is relatively low and modification on the strength is needed to make it suitable for welding of structural parts in automobiles and aerospace applications. In the present study, the influence of alloying elements (Mn, Mg and Zr) to 4xxx filler wires will be studied in both of as-welded and post-weld-heat-treatment conditions by using 6061 as the base metal and gas metal arc welding (GMAW) as welding technique. The macrostructure and microstructure will be revealed by optical microscopy, scanning electron microscopy, and transmission electron microscopy. The mechanical properties will be characterized by microhardness, tensile, and fatigue testing. Corrosion resistance of the new welded materials will be investigated.
Les fils de soudage en aluminium 4xxx sont largement utilisés pour souder la plupart des alliages d'aluminium, en particulier ceux de la série 6xxx, en raison de leur excellente soudabilité, de leur résistance élevée à la fissuration à chaud, de leurs points de fusion bas et de la capacité de modification de la dureté en les rendant traitable thermiquement et/ou durcis par dispersion. La dureté de ces alliages est relativement faible et une modification de la résistance est nécessaire pour le rendre apte au soudage de pièces de structure dans les applications automobiles et aérospatiales. Dans la présente étude, l'influence des éléments d'alliage (Mn, Mg et Zr) sur les fils d'apport 4xxx sera étudiée dans des conditions telles que soudés ainsi que traités thermiquement après soudage en utilisant le 6061 comme métal de base et le soudage à l'arc métallique sous gaz comme technique. La macrostructure et la microstructure seront révélées par microscopie optique, microscopie électro-nique à balayage et microscopie électronique à transmission. Les propriétés mécaniques seront caractérisées par des essais de microdureté, de traction et de fatigue. La résistance à la corrosion des nouveaux matériaux soudés sera étudiée.
Mohamed AhmedUniversité du Québec à Chicoutimi
X.-G. Chen - UQACM. Javidani - Rio Tinto
Journée des étudiants – REGAL8 et 9 octobre 2019
M. Ahmed1, X.-Grant Chen1, Mousa Javidani2
1 Research Chair in Metallurgy of Aluminum Transformation, University of Québec at Chicoutimi, Saguenay, Québec G7H 2B1
2 Rio Tinto, Arvida Research and Development Center, Jonquière, Quebec, G7S 4K8
The effect of alloying elements on the mechanical and chemical properties of 4xxx filler alloys for
aluminum welding
Effet des éléments d'alliage sur les propriétés mécaniques et chimiques des alliages d’apport 4xxx pour le soudage de
l'aluminium
1. Introduction 3. Experimentalwelding of aluminum alloys used in thetransportation, such as cars, airplanes, andspaceships, is a very important matter toform the assembly of the structure with highmechanical and chemical properties.4xxx filler alloys have many advantages overthe other types such as high weldability, hotcracking resistivity, low melting points, lessprone to porosity, and better choice for 6xxxwelding.
2. Objective• Enhance the efficiency and the strength
in as-welded and PWHT conditions bymaking a modification in the chemicalcomposition of the 4xxx filler alloys.
• Study the effect of Mn and Mg on themechanical and chemical properties of4xxx filler alloys.
- Material composition
Alloy Si Mg Cu Fe Cr Mn Ti ZnAA6061 0.7 0.84 0.24 0.43 0.17 0.13 0.04 0.07
ER 4043 4.5 - 6.0Max.
0.05
Max.
0.3
Max.
0.8---
Max.
0.05
Max.
0.20
Max.
0.10
ER 4943 5-6 0.1-0.5 0.1 0.4 --- 0.05 0.15 0.10
Voltage (V) 16
Current (Amp.) 80Wire feed speed
(ipm) 350
Travel speed (ipm) 40
Wire diameter (mm) 0.89
Welding technique GMAW
Shielding gas He
4. Results
6. Future work• The contained phases in both as welded and PWHT
condition for ER 4043 and ER 4943 filler alloys are thesame and include the base α-Al, Mg2Si, eutectic Al-Si, andsome Fe bearing intermetallic. The difference in thephase contents is based on the difference in the contentof the alloying elements.
• The HAZ in both filler alloys has coarse and semi-continuous Mg2Si at the grain boundary with a littleamount of Fe intermetallics in as welded case.
• Post weld heat treated samples shows recovery in themicrohardness profile.
• ER-4943 filler alloy shows better mechanical propertiesthan ER-4043.
5. conclusion
4043 – as welded 4043 – PWHT
4943 – as welded 4943 – as welded
1.At constant Si, Fe, and Mn contents, 6.2, 0.15, and 0.25% respectively, three different alloys withdifferent Mg content, 0.6, 1, 1.4% respectively, will be designed to study the effect of Mg content onthe microstructure and the mechanical properties of the weld zones in both of as welded and PWHTconditions,
2.Then, another three alloys with the same composition as the first three ones with addition of aconstant Zr content,0.2%, to study the effect of Mg and Zr contents on the as weld and post weldheat treatment (PWHT) mechanical properties such as tensile strength, yield strength, ductility, andthe fatigue strength of the welded samples, and
3.Perform the tensile, fatigue, and corrosion tests for the welded samples.
Material Young Modulus(GPa)
Yield Strength(MPa)
Tensile Strength(MPa)
Elongation(Pct)
6061-T6 69.8 309 360 9.88
4043- as welded 65 110 185 10.95
4943- as welded 69.5 140 195 4.37
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
4848L’évolution des dispersoïdes lors du traitement thermique des alliages Al-Mg avec différentes teneurs en MgEvolution of dispersoids during heat treatment of Al-Mg alloys with different Mg levels
De nos jours, l’importance technologique des alliages Al-Mg est accrue, particulièrement pour les applications à des températures élevées telles que les industries automobile et aérospatiale. Les dispersoïdes thermique-ment stables améliorent les propriétés du matériau à température élevée. Dans ce travail, le comportement de précipitation des dispersoïdes pendant le traitement thermique est étudié pour des alliages Al-Mg avec teneurs différentes de Mg (2-5% en poids). Un traitement en deux étapes (425 °C/2h + 500 °C/2-5h) a été réalisé pour clarifier l'évolution des dispersoïdes. Pendant les traitements thermiques, un grand nombre de dispersoïdes pré-cipitent dans la matrice. La microdureté et la conductivité électrique augmentent progressivement avec le temps de maintien tandis que le barème 500 °C/6h reflétait les conditions optimales pour des densités de dispersoïdes les plus élevées. La distribution et le type de dispersoïdes varient avec la teneur en Mg. Les alliages avec 5 % de Mg présentaient une fraction de surface des dispersoïdes inférieure, avec un seul type de dispersoïdes Al
6(Mn,
Fe, Cr), comparés aux alliages de 2 à 3 % en poids de Mg, dans lesquels deux types différents de dispersoïdes “Al
6(Mn, Fe, Cr), α-Al (Fe, Mn, Cr)Si’’ ont été observés.
Nowadays, the technological importance of Al-Mg alloys is markedly increased, particularly in applications operated at elevated temperatures such as automotive and aerospace industries. Thermally stable dispersoids enhance the material properties at elevated temperatures. In this work, the precipitation behavior of dispers-oids during heat treatment was investigated in Al-Mg alloys with different Mg levels (2-5 wt.%). Two-step heat treatment (425°C/2h + 500°C/2-24 h) was performed to clarify the evolution of dispersoids. During the heat treatments, large number of dispersoids precipitate in the matrix. The microhardness and electrical conductivity are gradually increased with increasing time, while 500°C/6h reflected the optimal condition for the formation of higher number density of dispersoids. The distribution and type of dispersoids varies with Mg contents. Alloys containing 5 wt.% Mg showed lower area fraction of dispersoids zone with only one type of dispersoids (Al
6(Mn-
FeCr)) compared to alloys with 2-3 wt.% Mg level, in which two different types of dispersoid “Al6(Mn, Fe, Cr) and
α-Al-(Fe, Mn, Cr)Si” were observed.
Ahmed AlgendyUniversité du Québec
à Chicoutimi
K. Liu - UQAC X. - G. Chen - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
EEvvoolluuttiioonn ooff DDiissppeerrssooiiddss dduurriinngg HHeeaatt TTrreeaattmmeenntt ooff AAll--MMgg AAllllooyyss wwiitthh DDiiffffeerreenntt MMgg LLeevveellss
LL’’éévvoolluuttiioonn ddeess ddiissppeerrssooïïddeess lloorrss dduu ttrraaiitteemmeenntt tthheerrmmiiqquuee ddeess aalllliiaaggeess AAll--MMgg aavveecc ddiifffféérreenntteess tteenneeuurrss eenn MMgg
Ahmed Y. AlGendy, Kun Liu, X. Grant ChenResearch Chair in Metallurgy of Aluminium
Transformation, UQAC
Nowadays, the vast technological importance of Al-Mg alloys is markedly increased particularly inapplications operated at elevated temperatures such as automotive and aerospace industries.Thermally stable dispersoids enhance the material properties at room and elevated temperatures.Commercial Al-Mg Alloy with variant Mg level were fabricated by casting and the effect of heattreatment and Mg content on the evolution of dispersoids were investigated.
IInnttrroodduuccttiioonnStudy the evolution of dispersoids during heat treatment in terms of type,
distribution and morphology.
Optimal a desirable microstructure with higher density and uniformdistribution of dispersoids by mean “suitable heat treatment”.
OObbjjeeccttiivvee
MMeetthhooddoollooggyy11.. MMaatteerriiaallss ccoommppoossiittiioonn
Si Fe Cu Mn Mg Cr AlAlloy 2% Mg 0.28 0.3 0.1 0.81 2 0.14 Bal.
Alloy 3% Mg 0.26 0.25 0.1 0.81 3.32 0.16 Bal.
Alloy 5% Mg 0.26 0.31 0.11 0.78 5.23 0.18 Bal.
22.. HHeeaatt TTrreeaattmmeenntt CCyyccllee
TimeTe
mpe
ratu
re
8h30 2h
1h 24h
Water Quenching
2 6 12 24
425oC
500oC
33.. AAllllooyy PPrrooppeerrttyy
Electrical Conductivity (EC) Using a sigmascope SMP 10 %IACS unit.
Microhardness (HV) Using NG-1000 CCD machine, with 25g load for 20s dwelling time.
44.. MMiiccrroossttrruuccttuurraall OObbsseerrvvaattiioonn Optical Microscopes: 0.5% HF for 35 Sec for clearly observation of
Dispersoid Zones (DZ) and Dispersoids Free Zones (DFZ).
Clemex imaging software
Scanning Electron Microscope (SEM) equipped with energy-
dispersive X-ray Spectrometer (EDS)
Transimision Electron Microscope (TEM)
The optimal condition for the formation of higher number density of dispersoids is soaking at 500oC for 6h.As the heating time increase, the electrical conductivity and hardness keep going up, which indicate that the precipitation
of dispersoids during heat treatment.Alloys containing 5 wt.% Mg showed lower area fraction of dispersoids zone with only one type of dipsersoids
(Al6(MnFeCr)) compered to alloys with 2-3 wt.% Mg level, in which two different types of dispersoid “Al6(Mn, Fe, Cr)and α-Al-(Fe, Mn, Cr)Si” were observed.
CCoonncclluussiioonnss
RReessuullttssiiii.. OOppttiiccaall OObbsseerrvvaattiioonn
3% Mg 5% Mg
(a) (b) (c)
AAllllooyy 22%%
AAllllooyy 33%%
AAllllooyy 55%%
2% Mg
3% Mg
FFuuttuurree WWoorrkk Study the mechanical properties at room and
elevated temperature. Improve the mechanical properties of Al-Mg alloys
at elevated temperature by increasing the volumefraction of fine and thermally stable dispersoids.
ii.. AAllllooyy PPrrooppeerrttiieess
(a) (b)
(d)
E
C
a b c
d e f
g h i
(a) (b) (c)
(d)
Figure 1 (a) Electrical conductivity (b) Vickers hardness of experimental alloys after heat-treated at 500 oC/6h
Figure 2 a, b and c Optical micrographs show the dispersoid Zone of the three alloys with different Mg level after heat-treated at 500 oC/6h, (d) The area percent of dispersoid zone and dispersoids free zone of different alloys after heat-treated at
500oC/6h.
Figure 3 Typical SEM images of dispersoids of different alloys after heat-treated at 500 oC for different time (a, d, g) at the beginning of holding (b, e, h) 6 h (c, f, i) 24 h
Figure EDS analysis of rod dispersoidsafter heat-treated at 500 oC/6h
Figure 5 The morphology of dispersoids of experimental alloys with different Mg content after heat-treated at 500 oC/6h (a) 2% Mg (b) 3% Mg (c) 5% Mg, (d) Number density of dispersoids of three alloys
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
49Étude de la conception et le traitement thermique d’un bras de suspension
d’automobile en alliage d’aluminium A357 semi-solideStudy of design and thermal treatment of automotive control arm
fabricated from A357 semi-solid alloy
Lower control arm is one of the main components of automotive suspension system. It is responsible of linking the wheels of the vehicle to the chassis. A common automotive problem, which cause road accidents, is due to fatigue failure of suspension control parts that are subjected to severe conditions of dynamic loadings. A new trend is to manufacture this automotive part from aluminum semi-solid alloys due to their significant charac-teristics and qualities. This work studies the performance of lower control arm manufactured from A357 alloys using semi-solid Rheocasting with SEED technology. The study aims to investigate a new heat treatment process capable of enhancing the fatigue characteristics of the alloy. It also proposes a new control arm design that is lighter and stronger than the traditional ones. The fatigue life is affected by thermal treatment and design parameters applied on the suspension control arm of A357 semi-solid alloys. The results revealed that the inter-rupted thermal aging process as well as the modified trussed-design showed superior quality and performances concerning the fatigue life.
Le bras de suspension inférieur est l’un des principaux composants du système de suspension automobile. Il est chargé de relier les roues du véhicule au châssis. Un problème courant dans l’automobile, qui cause des accidents de la route, est dû à la défaillance en fatigue de pièces de contrôle de la suspension soumises à des conditions extrêmes de chargement dynamique. Une nouvelle tendance consiste à fabriquer cette pièce automobile à partir d'alliages d'aluminium semi-solides en raison de leurs caractéristiques et de leurs qualités. Ce travail étudie les performances du bras de suspension inférieur fabriqué à partir d'alliages A357 à l'aide de Rhéocasting Semi-solide avec la technologie SEED. L’étude vise à étudier un nouveau procédé de traitement thermique capable d’améliorer les caractéristiques de fatigue de l’alliage. Il propose également une nouvelle conception de bras de suspension plus léger et plus résistant que le traditionnel. Le traitement thermique et les paramètres de conception appliqués au bras de suspension des alliages A357 semi-solides ont une incidence sur la durée de vie en fatigue. Les résultats ont révélé que le processus de vieillissement thermique interrompu ainsi que la conception en treillis modifiée présentaient une qualité et des performances supérieures en ce qui concerne la résistance à la fatigue.
Mohamed AttiaUniversité du Québec à Chicoutimi
Khaled Ragab - UQACM. Bouazara - UQACX.-G. Chen - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
Mohamed Attia1, Khaled Ragab1 ,Mohamed Bouazara1 et X-Grant Chen1,2
1 Department of Applied Sciences, University of Quebec at Chicoutimi2 Research Chair in Metallurgy of Aluminum Transformation, UQAC
Study of design and thermal treatment of automotive control arm fabricated from
A357semi-solid alloy
Étude de la conception et le traitement thermique d’un bras de suspension d’automobile en alliage
d’aluminium A357 semi-solide
Introduction
Courtesy of Federal Mogul Motorparts
• Automotive lower control arm is a major mechanical component in automotive suspension system.
• Links wheels to the chassis of the vehicle.• Subjected to dynamic loading due to vehicle
acceleration and deceleration.
A. Upper & lower control arms.B. Wheel spindle.C. Spring damper assembly. D. Rack and pinion assembly.
Semi-solid Rheocasting using SEED* technology
1. Higher forming speed.2. Less mold erosion.3. Less shrinkage porosity.4. Less oxides.5. Better surface finish.6. Lower flow stress than
conventional casting.
1. More sophisticated process. 2. Time consuming than forging.3. Accurate temperature and
time measurements.
* SEED: Swirled Enthalpy Equilibration Device
ObjectivesEnhancement of the total fatigue life of the automotive lower control arm by:
1. Application of unconventional thermal treatments. 2. Development of an innovative design to reduce part’s weight and
improve stress distribution.
Experimental procedures
0
100
200
300
400
500
600
0 2 4 6 8
Tem
pera
ture
(⁰C)
Time (Hr)0
50
100
150
200
250
0 5 10 15Time (Hr)
A B0
50
100
150
200
250
0 2 4 6 8Time (Hr)
C
2Hrs
2Hrs 0,1,3,5,8 Hrs
0,1,3,5,8 Hrs
C1(1Hr,1Hr,24Hrs@RT,2Hrs)
C2(3Hrs,2Hrs,1Hr)
C3(3Hrs, 24Hrs@RT,2Hrs,1Hr)
1Hr, @470 ⁰C
5Hrs, @540 ⁰C
T6/T7 T7/T6Interrupted aging,T4/T6/T7
Tensile testing
Selection
Heat treatment
Real part fatigue
F=7000 Nf=1 Hz
Selection
Selection
Cantilever bending fatigue
samplesf=12 Hz
δ= 6.35 mm
Design modifications
Finite element analysis
Merging of results
Results and observations
Conclusions• WC3 shows a remarkable superior yield strength of 288 MPa compared to other aging cycles, as well as, a good elongation of 9.4%.• WC3 is observed to have superior fatigue life in both the real part fatigue (72,000 cycles) and the cantilever fatigue testing (82,000) compared to the T6
standard. • The modified trussed design (Design 4) shows enhanced stress distribution with a max. Von-Mises stress of 198 MPa weighting 1040 gm.• Appling C3 to the Trussed design is expected to withstand more than 78,000 cycles. (117% enhancement of conventional design ).• Trussed design weights less than original design by 160 g (86% of conventional design).
AverageT6
53,000
AverageWA0
54,667
AverageWA1
59,250
AverageWB0
25,000
AverageWC1
53,000
AverageWC3
82,000
-
10,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
T6 WA0 WA1 WB0 WC1 WC3
No. o
f cyc
les
Aging cycle
Cantilever bending fatigue samples
Desig
n m
odifi
catio
ns
Design Finite element results
Design 1: Reinforced inclined web
Design 2: Planar thin web
Design 4: Trussed design
Design 3: Hollow design
σ𝑚𝑚𝑚𝑚𝑚𝑚 = 213 𝑀𝑀𝑀𝑀𝑀𝑀𝛿𝛿𝑚𝑚𝑚𝑚𝑚𝑚 = 1.45𝑆𝑆𝑆𝑆𝑆𝑆𝑚𝑚𝑚𝑚𝑚𝑚 = 8.0
σ𝑚𝑚𝑚𝑚𝑚𝑚 = 199 𝑀𝑀𝑀𝑀𝑀𝑀𝛿𝛿𝑚𝑚𝑚𝑚𝑚𝑚 = 1.46𝑆𝑆𝑆𝑆𝑆𝑆𝑚𝑚𝑚𝑚𝑚𝑚 = 9.95
σ𝑚𝑚𝑚𝑚𝑚𝑚 = 287 𝑀𝑀𝑀𝑀𝑀𝑀𝛿𝛿𝑚𝑚𝑚𝑚𝑚𝑚 = 1.69𝑆𝑆𝑆𝑆𝑆𝑆𝑚𝑚𝑚𝑚𝑚𝑚 = 5.2
σ𝑚𝑚𝑚𝑚𝑚𝑚 = 198 𝑀𝑀𝑀𝑀𝑀𝑀𝛿𝛿𝑚𝑚𝑚𝑚𝑚𝑚 = 1.78
𝑆𝑆𝑆𝑆𝑆𝑆𝑚𝑚𝑚𝑚𝑚𝑚 = 2.475
- 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000
WC3
WA0
T6
WC1
WB0
WA8
WB5
WA1
No. of cycles
Agi
ng c
ycle
Real part fatigue
WC3 T6 WA1Accepted Rejected
SHT WA0 WA1 WA3 WA5 WA8 WB0 WB1 WB3 WB5 WB8 WC1 WC2 WC3Ductility (%) 23.0% 17.3% 7.6% 8.0% 8.7% 9.3% 10.0% 8.2% 6.4% 5.3% 7.1% 12.0% 10.5% 9.4%
Yeild Strength (MPa) 136.619 201.53 263.8 252.3 208 191 225.476 233 233.75 243.4475 235 235.459 253.843 288
UTS (MPa) 261.33 270.00 285.80 277.75 252.00 238.20 255.60 261.40 272.40 278.20 265.85 272.30 291.80 302.80
0
50
100
150
200
250
300
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
Stre
ss (M
Pa)
Heat Treatmemt plan
Duct
ility
(%)
Mechanical Behaviour
m= 1141 g
m= 1198 g
m= 1082 g
m= 1040 g
• Conventional design mass = 1200 g [1]. • Conventional design max. Von-Mises stress = 193 MPa [1].[1] Ragab, K.A., Bouaicha, A., Bouazara, M., 2017. Optimization of Casting Design Parameters on Fabrication of Reliable Semi-Solid Aluminum Suspension Control Arm. J. Mater. Eng.
Perform. 26, 4450–4461. https://doi.org/10.1007/s11665-017-2878-1
Aluminum billet in the semi-solid state (Courtesy of formcast, Inc.)*E. J. Vinarcik, High Integrity Die Casting Processes. 2002.
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
5050Conception et analyse économique du cycle de vie d’un pont poutres en bois/ platelage en aluminiumDesign and economic life cycle assessment of a bridge built from wooden beams and aluminum deck
Aujourd’hui, les matériaux de construction les plus couramment utilisés pour les ponts au Québec sont le béton et l’acier. La production de ces matériaux a un fort impact environnemental et contribue à la raréfaction des ressources non renouvelables. D’autres matériaux de construction sont envisageables pour la construction de ponts et ont a priori une empreinte environnementale plus faible. Par exemple, le bois apparaît comme une res-source renouvelable qui est utilisée comme puits de carbone. L’aluminium, quant à lui, est un matériau recyclable produit à l’aide d’hydro-électricité, ce qui lui permet d’avoir un impact carbone relativement faible. En dépit de leur potentiel en termes de développement durable, l’impact environnemental de ponts utilisant uniquement du bois et de l’aluminium n’est pas quantifié. L’analyse de cycle de vie de tels ponts est une donnée de grande importance qui permettra de choisir des solutions de construction durables. Le projet débute par la conception d’un pont en bois lamellé-collé avec un platelage en aluminium. Une analyse de cycle de vie de ce pont fictif sera ensuite réalisée. Les résultats obtenus, notamment les impacts environnementaux, seront ensuite comparés avec ceux d’études analogues réalisées sur des ponts avec des poutres d’acier et une dalle de béton.
Nowadays, concrete and steel are the most commonly used materials in bridges construction in Quebec. The pro-duction of these materials has a significant environmental impact and contributes to the scarcity of non-renew-able resources. Other materials that can be used for bridges construction, have a lower environmental footprint a priori. For instance, wood is a renewable resource that can be used as carbon sink. Moreover, aluminum is a recyclable material which is produced thanks to hydroelectricity which makes it possible to have a relatively low environmental impact. Regardless of it potential in terms of sustainable development, the environmental impact of such bridges is not yet quantified. The assessment of life cycle of such bridges is therefore an important indicator that will allow selecting lasting building solutions. The project begins with the design of a bridge made of glue-laminated timber beams and aluminum deck. A life cycle analysis of this bridge will then be carried out. Achieved outcomes, especially the environmental impacts, will be compared to the outcomes of similar studies that have been carried out on bridges made of steel beams and concrete deck.
Camille BeudonUniversité Laval
M. Oudjene - UL A. Djedid - UL
C.-D. Annan - UL M. Fafard - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Camille BEUDON1, Marc OUDJENE1 , Amar DJEDID1, Charles-Darwin ANNAN1, Mario FAFARD1,2
1Département de génie civil et génie des eaux, Université Laval, 1065 Av. de la Médecine, Quebec, QC, G1V 0A62Centre de recherche sur l’aluminium-REGAL et Département de génie civil et génie des eaux, Laval University, 1065 Av. de la Médecine, Quebec, QC, G1V 0A6
Conception et analyse du cycle de vie d’un pont poutres en bois/ platelage en
aluminium
Design and life cycle assessment of a bridgebuilt from wooden beams and aluminium deck
MISE EN CONTEXTEAujourd’hui, les matériaux de construction les plus couramment utilisés pour les ponts au Québec sont le béton et l’acier. La production deces matériaux a un fort impact environnemental et contribue à la raréfaction des ressources non renouvelables. De plus, les coûts liés àleur entretien et leur installation (fermeture des routes) sont élevés.
D’autres matériaux de construction sont envisageables pour la construction de ponts et ont à priori une empreinte environnementale plus faible.Par exemple, le bois apparaît comme une ressource renouvelable qui est utilisée comme puits de carbone. L’aluminium, quant à lui, est unmatériau recyclable, produit à l’aide d’hydro-électricité ce qui lui permet d’avoir un impact carbone relativement faible.Hormis le bois en tant que surface de roulement, l’utilisation de ces matériaux de construction est encore embryonnaire au Québec, puisqueseuls quelques ponts utilisent du bois pour le système structural (poutres). Seul un pont utilise de l’aluminium pour sa superstructure et un seull’utilise pour le platelage.
PROBLÉMATIQUEEn dépit de leur potentiel en termes de développement durable, l’impact environnemental de pontsutilisant uniquement du bois et de l’aluminium n’est pas quantifié. L’analyse de cycle de vie de telsponts est une donnée de grande importance qui permettra de choisir des solutions de constructionéconomiques et durables.
Figure 2: Le bois et l’aluminium, deux matériaux complémentaires
Figure 1 : Estimation des émissions de C02 des différents types de platelage1
1 Estimation des émissions de C02 rejetées au cours de la fabrication des différents platelages. Ce calcul est réalisé sur un platelage de trois voies carrossables et d’une portée de 20m. L’hypothèse est faite que le coefficient d’émission de la productionde l’acier est de 1,06 tonnes de C02 émises par tonne d’acier produit, celui du béton de 0,12 et celui de l’aluminium de 2. À l’inverse, 1 m3 de bois absorbe quant à lui 1 tonne de C02 . Les données pour les platelages dalle de béton et acier-bois sont issuesd’études réalisées par l’entreprise Nordic Structures. Seules les émissions produites au cours de la production des différents matériaux ont été prises en compte. Un calcul plus précis nécessiterait la prise en compte des émissions au cours de chaque étape duchantier à savoir: la production des matériaux, l’assemblage , le transport jusqu’au site et la mise en place des différents éléments, ainsi que la remise en état des lieux.
ÉTUDE DES CONNECTEURS
TEST THERMIQUE/ TEST HUMIDITÉ
Où? Dans la chambre climatique (possibilité defaire des gradients de température positifs etnégatifs)
ObjectifObserver les déformations du connecteur
Matériel nécessaire- Jauge de déformation placée sur le connecteur- Jauge de déformation placée sur la poutre- Connecteur- Plaque rigide- Poutre
Protocole- Application d’une précontrainte à l’aide des
boulons aveugles- Essai à température fixe et variation du taux
d’humidité- Essai à taux d’humidité fixe et variation de
température
Plage de variation de températuresde - 55 °C à 68 °C
Plage de variation du taux d’humiditéDe 0 % à 100 % d’humidité
Nombre d’essais à réaliser3 essais par mesure
TEST D’ARRACHEMENT EN TRACTION
Où? Grâce à une presse mécanique
ObjectifTest de la résistance du connecteur àl’arrachement à l’ultime
Matériel nécessaire- Capteur LVDT (mesure du glissement du
connecteur)- Connecteur- Plaque rigide- Poutre
Protocole- La plaque rigide est tirée par la presse jusqu’à
l’arrachement- Réalisation d’un certain nombre de cycles
charge-décharge jusqu’à la rupture
Figure 3: Détail de l’assemblage du connecteur
Traction sur le connecteurÉcrasement
Test jusqu’à la ruptureMesure du déplacement conduisant à la
rupture ultime
Mise en contexteLe bois est un matériau hygroscopique et sujet à des variations dimensionnelles dues àl’humidité. En revanche, ses dimensions ne sont pas sensibles aux variations detempérature (dilatation thermique) contrairement à l’aluminium. C’est pourquoi l’actioncomposite entre ces deux matériaux n’est pas adaptée. L’assemblage bois/ aluminium adonc un comportement différent en fonction de la température et du taux d’humidité.
ObjectifTester les connecteurs (assemblages et connexion) sur l’assemblage bois/ aluminium
Choix du connecteurConnecteur en aluminium
L’ALUMINIUM ET LE BOIS,DEUX MATÉRIAUX PROMETTEURS
CONCEPTION D’UN PONT POUTRES EN BOIS / PLATELAGE EN ALUMINIUMSelon la norme CAN/CSA S6-2014. (2014). Code canadien sur le calcul des ponts routiers. Association canadienne denormalisation.
PONT EN BOIS,FORET DE MONTMORENCYDate de construction: 2010Portée: entre 10 et 38,4 mMatériaux utilisés:Système structural composé d’untablier en bois et de poutres en lamellécolléContexte de création:La Charte du bois du gouvernement duQuébec encourage l’utilisation du boisdans la construction de bâtiments etd’infrastructures afin qu’une industrieet qu’une expertise québécoises sedéveloppent. La filière bois est, de plus,un secteur économique majeur auQuébec.Conclusion:Le bois apparaît comme une alternativepossible et s’intègre bien dans un cadrede développement durable.Même si plusieurs mesures ont étéprises lors de la conception de cesouvrages pour assurer leur durabilité,la construction de ponts en boisprésente pour l’instant une certaineincertitude relative au boncomportement à long terme du boissoumis à des surcharges routières et auclimat rigoureux du Québec.
ANALYSE DU CYCLE DE VIE D’UN PONT POUTRES EN BOIS / PLATELAGE EN ALUMINIUMRÉALISATION D’UNE ANALYSE DE CYCLE DE VIE ÉCONOMIQUE ET ENVIRONNEMENTALE SUR LE PONT CONCU
PONT EN ALUMINIUM,SAINT AMBROISE, LAC SAINT JEANDate de construction: 2013-2014Portée: 9,2mMatériaux utilisés:Le platelage en aluminium est supportépar une structure typique de pontacier-bois avec 5 poutres en acier quireposent sur deux culées en béton.Contexte de création:En 1990, le Ministère des Transports duQuébec (MTQ) en collaboration avecl’Université Laval décide de réaliser uneétude d’opportunité pour leremplacement d’un platelage de boissur un pont à ferme métallique. Lasolution proposée est un platelage enaluminium déjà réalisé en Suède. Acette époque, l’aluminium n’a pas étéretenu à cause des difficultés et descoûts d’importation. Des avancéestechniques ont permis la réalisationd’un platelage tout en aluminiumquelques années plus tard.Conclusion:Les tabliers en aluminium sontintéressants pour leur étanchéité etleur durée de vie élevée (75 ans).
Analyse de cycle de vie de plusieurs ponts fictifs, de courte portée, susceptibles d’être installés sur des routes locales à faible trafic, afin de comparer lesimpacts environnementaux et les coûts globaux de chacune des solutions:- Pont à platelage en bois sur poutres en bois lamellé-collé- Pont à platelage en aluminium sur poutres en bois lamellé-collé- Pont à platelage en bois sur poutres d’acier- Pont à dalle de béton sur poutres d’acier
Étude de chacune des étapes du cycle de vie: phase de production (extraction et transformation des matières premières, transport jusqu’au fabricant etfabrication), phase de processus de construction (transport jusqu’au client et installation), phase d’utilisation (maintenance, réparation, remplacement etsubstitution, rénovation), phase de fin de vie (déconstruction, transport jusqu’au centre de traitement des déchets, traitement des déchets pour récupération,réutilisation et/ ou recyclage, élimination)
PONT RETENU
- Portée: 20 m Largeur: 11,6 m 3 voies carrossables- Porte-à-faux d’une largeur de 633 mm- Platelage en aluminium d’une épaisseur de 110 mm- 9 poutres en bois lamellé-collé espacées de 1300 mm
Étape 1Conception du pont sur un logiciel de modélisation par éléments finis
Étape 2Détermination des différentes charges pondérées appliquées sur le tablier du pont
• Calcul de la surcharge routière et du facteur d’essieu (charge uniformément répartie + charge ponctuelle)
• Calcul du poids propre total (somme du poids propre dû au platelage en aluminium, despoutres en bois lamellé-collé et des autres poids morts comme les glissières de sécurité dont lemodèle est validé par le MTQ) (charge uniformément répartie)
Une fois que les différentes charges sont pondérées, des calculs sont faits pour obtenir lemoment fléchissant, l’effort tranchant et la flèche associés à la poutre en lamellé-collé laplus sollicitée (poutre de rive). En traçant le diagramme des moments et des efforts, le momentfléchissant maximal ainsi que l’effort tranchant maximal et la flèche maximale de chaque poutresont obtenus.
Étape 3Détermination des résistance maximales admissibles en flexion et cisaillement
Détermination de la flèche maximale admissiblePour une poutre en bois lamellé-collé (1054*327*20000 mm)
• Utilisation de la norme canadienne (Wood Design Manual. Ottawa, Ontario: Canadian Wood Council, 2010)
• Utilisation de la norme européenne (NF EN 1995-1-1. Norme européenne, Eurocode 5,Conception et calcul des structures en bois : Comité Européen de normalisation, 2009)
• Comparaison des normes canadienne et européenne
Étape 4Comparaison des normes françaises et canadiennes
• Comparaison des résistances maximales théoriques en flexion et en cisaillement par rapport aux contraintes calculées.
• Détermination des taux de services des poutres en bois lamellé-colléATTENTION
La norme canadienne de calcul des ponts routiers ne mentionne que des ponts à dalle sur poutres enacier et non sur poutres en bois. Les calculs sont donc réalisés avec des formules associées à despoutres en acier. Une certaines incertitude est donc à prendre en compte lors de l’étape 2 etnotamment lors du calcul du facteur d’essieu.
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
51La calorimétrie à balayage différentiel et sa caractérisation des états de traitement thermique dans les alliages extrudés AA6082
DSC fingerprints of heat treatment tempers in extruded AA6082 alloys
6xxx wrought aluminum alloys have been widely used in automobile industries due to their good strength and corrosion resistance. During their applications, the differential scanning calorimetry (DSC) technique can be a convenient tool in order to rapidly acquire a diagnostic for the heat-treatment-related problems on the products, but limited systematic work is performed. In the current work, extruded 6082 aluminum alloys have received dif-ferent heat treatments and characterized by the DSC technique. Characteristic peaks on DSC curves have been analyzed to identify the precipitation / dissolution processes of different metastable phases. Results show that both the peak number and the peak shape varies with thermal treatment, indicating their different precipitation / dissolution behaviors. Meanwhile, the electric conductivity and microhardness at various tempering conditions are investigated together with the microstructures observation by the transmission electron microscope to well support the changes of DSC curves. As a result, different heat treatment conditions of extruded 6082 alloy has been successfully identified with the established DSC database in the present work.
Les alliages d'aluminium corroyés 6xxx ont été largement utilisés dans l'industrie automobile en raison de leur dureté et de leur résistance à la corrosion. Durant leurs applications, la technique de calorimétrie à balayage différentiel (DSC) peut être un outil de diagnostic pratique et rapide concernant des problèmes liés au traitement thermique des produits, mais un travail systématique limité est effectué. Dans les travaux en cours, les alliages d'aluminium extrudés 6082 ont subi différents traitements thermiques et ont été caractérisés par la technique DSC. Les pics caractéristiques des courbes DSC ont été analysés pour identifier les processus de précipitation / dissolution de différentes phases métastables. Les résultats montrent que le nombre et la forme des pics varient avec les différents états de traitement thermique, indiquant leurs différents comportements de précipitation / dissolution. Parallèlement, la conductivité électrique et la microdureté pour différents états de traitement ther-mique ont été étudiées avec l'observation des microstructures au microscope électronique à transmission afin de bien interpréter les modifications des courbes DSC. En conséquence, différentes conditions de traitement thermique de l'alliage 6082 extrudé ont été identifiées avec succès à l’aide de la base de données DSC établie dans le présent travail.
Zhixing ChenUniversité du Québec à Chicoutimi
K. Liu - UQACX.-G. Chen - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
DSC fingerprints of heat treatment tempers in extruded 6082 alloys
Calorimétrie à balayag différentiel (DSC) et sa caractérisation des états de traitement
thermique dans les alliages extrudés 6082
Zhixing Chen, Kun Liu, X. Grant Chen
Research Chair in Metallurgy of Aluminium Transformation, UQAC
• To perform DSC tests on alloys with various heattreatment to determine their characteristics.
• To build up a connection between DSC characteristicsand properties.
• To establish a fingerprint database for identifyingvarious tempers of extruded 6082 alloy.
2. OBJECTIVES
6xxx wrought aluminum alloys have been widely used inautomobile industries. During their applications, heat-treatment related problems on products can be found bythe differential scanning calorimetry (DSC) technique,which is a convenient tool in order to rapidly acquire adiagnostics since it is aiming for investigating a samplethat has undergone an unknown or ill defined heattreatment, but limited systematic work is performed.
1. INTRODUCTIONMaterial Composition
Heat Treatment
3. METHODOLOGY
Various tempers can be identified from the peak number and peak height in 6082 alloy: Peak number: F temper and solution states, natural aging states, and peak aging states by the DSC peak existence of β”, β’, β;
Peak height: F temper and solution states by peak height of β”, natural aging states by peak height of β”, β’, β;
Different DSC curves of various tempers are supported by the evolution of electric conductivityand microhardness, indicating the relations between DSC characteristics and properties;
The results of TEM observation and quantification offer evidence to the properties of 6082 alloyswith different tempers.
5. CONCLUSIONS Quantifying the DSC peak with phase
volume to find the relationship betweenDSC and microstructure;
Studying the precipitation kinetics in orderto explain the different temperatures ofDSC peaks, which can also be one of thecharacteristics of DSC curves.
6. FUTURE WORK
4. RESULTS
In the current work, extruded 6082aluminum alloys have been heattreated with different tempers andcharacterized by the DSC technique.
Alloy Al Si Mg Mn Fe Cr6082 Bal. 0.9 0.7 0.5 0.2 0.1
100
200
300
400
500
Tem
pera
ture
()
T im e
WQ AQ WQT4 T5 WQT6 AQT6
100 200 300 400 5002
4
6
8
10
12
14
16
18
20
22 DSC Curve
Hea
t Flo
w E
ndo
Up
(mW
)
Sample Temperature (°C)
Machine Baseline
(a)
Baseline subtraction
100 200 300 400 500
0.0
0.5
1.0
1.5
2.0(b)
DSC Curve baseline subtracted Polynomial fit of heat effect
Hea
t Flo
w E
ndo
Up
(mW
)
Sample Temperature (°C)100 200 300 400 500
-1.5
-1.0
-0.5
0.0
0.5(c)
Heat effect corrected Curve
Hea
t Flo
w E
ndo
Up
(mW
)
Sample Temperature (°C)
Heat effect correction
DSC
Properties EvaluationElectric
ConductivitySigmascope
SMP10Unit: IACS%
Vickers Hardness
NG-1000 CCD25gF, 15s
TEM Observation & Quantification
𝝆𝝆 = 𝟑𝟑𝑵𝑵∥𝑨𝑨 𝒕𝒕 + 𝝀𝝀
Exo. Peak a Exo. Peak b Exo. Peak c Endo. Peak I Exo. Peak d Exo. Peak e Endo. Peak II
F GP zones, 160 β", 240 β', 320 β, 410 β dissolution
WQ Clusters, 100 GP zones, 160 β", 240 β', 320 β, 410 β dissolution
AQ GP zones, 175 β", 240 β', 320 β, 390 β dissolution
WQT4_1 β", 240 β', 320 β, 430 β dissolution
WQT4_7 β", 240 β', 320 β, 430 β dissolution
WQT4_28 β", 240 β" dissolution β', 320 β, 430 β dissolution
T5 β", 225 β', 330 β, 410 β dissolution
WQT6 β", 225 β" dissolution β', 330 β, 400 β dissolution
AQT6 β", 225 β" dissolution β', 330 β, 390 β dissolution
100 200 300 400 500
II'
ed
WQ
Temperature ()
Nor
mal
ized
Hea
t Flo
w (W
/g)
Exot
h.
End
oth.
(a)0.01 W/g
b c
II
e'd'a'
c'
b'
b"c"
d" e"II"
F
AQ
100 200 300 400 500
WQT4_1
Nor
mal
ized
Hea
t Flo
w (W
/g)
Exot
h.
End
oth.
Temperature ()
0.01 W/g
WQT4_7
I28II28
II7c28
c7
WQT4_28
(c)
c1
II1
d28
d7
d1
e28
e7
e1
100 200 300 400 500
WQT6
Nor
mal
ized
Hea
t Flo
w (W
/g)
Exot
h.
End
oth.
0.005 W/g
Temperature ()
IIT5
eT5dT5cT5
(b)
AQT6cA
cW dW
dA
eW
eA
IW
IA
IIW
IIA
T5
0 7 14 21 28
4
6
8
10
1230
32
34
36(d)
Peak c Peak d Peak e
Peak
hei
ght (
10-3
W/g
)
Natural aging time (Day)
WQT6 AQT6 T50.0
5.0x103
1.0x104
1.5x104
2.0x104
17706.54
3350.64
16724.6
53.1
87.3
75.3
Number Density
Num
ber D
ensi
ty (μ
m-3
)
0
20
40
60
80
100
120
140 Precipitate Length
Prec
ipita
te L
engt
h (n
m)
F WQ AQ T4_1 T4_7 T4_28 T5 WQT6 AQT640
42
44
46
48
50
52
41.57
42.44
49.32
42.35 42.31 42.33
46.66
47.96
50.50 Electric Conductivity
Elec
tric
Con
duct
ivity
(IA
CS%
)
(a)
F WQ AQ T4_1 T4_7 T4_28 T5 WQT6 AQT640
50
60
70
80
90
100
110
120
130
93.57
75.91
57.19
69.56
77.14
97.93
116.88122.91
63.27
Vikers hardness
Mic
roha
rdne
ss (H
V0.0
25)
(b)
Fig. 1 DSC curves of different tempers: (a) F & SHT, (b) peak aging, (c) natural aging; and (d) peak heights of natural aging DSC curves.Table. 1 Summary of peaks on DSC curves.
Fig. 2 Physical & mechanical property evaluations of different tempers: (a) Electric conductivity, (b) Vickers hardness.
Fig. 3 TEM observations of peak aging tempers: (a) WQT6, (b) AQT6, (c) T5.
Fig. 4 TEM quantification result summary.
𝑵𝑵∥: Number of precipitate cross-sections;
𝑨𝑨: Image area;
𝒕𝒕: Thickness in the area;
𝝀𝝀: Average true needle length.
JOEL JEM-2100
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
5252Conception d’un banc de mesure de la friction appliqué au procédé High Speed Blow FormingDesign of a friction measurement bench applied to the High Speed Blow Forming process
Le High Speed Blow Forming (HSBF) est un procédé novateur de mise en forme superplastique de l’aluminium développé par Verbom qui, contrairement à la déformation superplastique classique, présente la particularité d’être beaucoup plus rapide, tout en conservant une capacité de déformation très élevée. Il consiste à injecter, dans la partie inférieure d’un moule prenant en tenaille une feuille d’aluminium portée à haute température (en-viron 500 °C), un gaz sous pression afin de plaquer la feuille contre la matrice supérieure pour lui en donner la forme. Cependant, les modèles classiques de friction aluminium-acier ne sont plus valables à cette température et entrainent des imprécisions dans les simulations du procédé et de fait, des défauts sur les pièces. De plus, aucun banc d’essai n’existe dans la littérature pour ce type d’application. Le but de ce projet est donc, dans un premier temps, de concevoir un banc de mesure de la friction à haute température permettant de prendre en compte l’influence de plusieurs variables comme la température, la pression ou la vitesse. Dans un second temps, une fois le banc fonctionnel, l’objectif sera de modéliser le coefficient de frottement en fonction des variables les plus significatives afin d’implémenter ce modèle aux logiciels d’éléments finis.
The High Speed Blow Forming (HSBF) is an innovative process for superplastic shaping of aluminum developed by Verbom which, unlike conventional superplastic deformation, has the particularity of being much faster, while maintaining a very high deformation capacity. It consists in injecting a pressurized gas into the lower part of a mold clamping an aluminum foil at high temperature (about 500°C) in order to press it against the upper matrix to give it its shape. However, conventional aluminum-steel friction models are no longer valid at this temperature leading to inaccuracies in process simulations and, consequently, defects on the parts. In addition, there is no test bench in the literature for this type of application. The aim of this project is therefore, to initially design a bench for measuring friction at high temperature, allowing the influence of several variables such as temperature, pressure or velocity to be taken into account. And then, once the bench is functional, the objective will be to model the friction coefficient according to the most significant variables in order to implement this model to finite element software.
Guillaume David ÉTS
N. Bombardier - VerbomM. Jahazi - ÉTS
Journée des étudiants – REGAL8 et 9 octobre 2019
Guillaume David1, Nicolas Bombardier2
Mohammad Jahazi31 M. Sc. Génie mécanique, École Technique Supérieure (ÉTS), 1100 Rue
Notre-Dame Ouest, Montréal, QCH3C1K32 Verbom, 5066, Route 222 Valcourt, QC, Canada, J0E 2L0
3 Directeur de recherche, ÉTS
Conception d’un banc de mesure de la friction appliqué au procédé HSBF (High Speed
Blow Forming)Design of a friction measurement bench applied to the HSBF (High
Speed Blow Forming) process
Figure 1 : Tesla model X (www.tesla.com/modelx)
Le coefficient de frottement à cette température ne suit plus les lois classiques utilisées dans leslogiciels de simulation du procédé. Les simulations ne sont plus suffisamment précises, onobserve l’apparition de défauts sur les pièces.
Grippage entre le moule etla plaque d’aluminium(galing)
Accumulations de lubrifiantqui laissent une empreintedans la pièce
Le hayon est composé d’une pièce d’aluminium mise en forme par une variante de la déformationsuperplastique développée par l’entreprise Verbom, le HSBF (High Speed Blow Forming).
Figure 3 : défaut de type galing Figure 4 : défaut sur la pièce causé par une accumulation de lubrifiant
Identifier les variables qui ont le plus d’influence (Pression, vitesse, température, usure dulubrifiant, …)
Réaliser un modèle mathématique du coefficient de friction qui pourra être implémenté dans lasimulation du procédé
Concevoir et fabriquer un banc permettant defaire des essais dans des conditions prochesdu procédé.
Châssis aluminium et
acier rigide
Actuateur linéaire
Piston fournissant un effort constant
Ressort quiabsorbe leslégères varia-tions de hauteur
Chariot oscillant et chauffant
Tige rigide pour prévenir la
déformation du piston
Outil monté sur rotule permettant un contact plan
Procédé HSBF
Problématique Solution
Objectifs
Principe de fonctionnement du banc
Variables étudiées
Figure 5 : Banc en cours de montage, 09/2019
L’effort normal et l’efforttangentiel dans la zone decontact sont mesurés pourobtenir le coefficient defrottement :
𝐹𝐹𝑇𝑇 = 𝜇𝜇 ∗ 𝐹𝐹𝑁𝑁𝐹𝐹𝑇𝑇 ∶ 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑒𝑒𝑡𝑡𝑡𝑡𝑡𝑡𝑒𝑒𝑡𝑡𝑒𝑒𝑡𝑡𝑒𝑒𝑡𝑡𝐹𝐹𝑁𝑁 ∶ 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑡𝑡𝑒𝑒𝑒𝑒𝑛𝑛𝑡𝑡𝑡𝑡𝜇𝜇 ∶ 𝑐𝑐𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑡𝑡𝑐𝑐𝑡𝑡𝑒𝑒𝑡𝑡𝑒𝑒 𝑑𝑑𝑒𝑒 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑛𝑛𝑒𝑒𝑡𝑡𝑒𝑒
Gaz sous pression
Il permet un taux d’élongation de 380% pour une durée de cycle 6 fois plus rapide que la mise enforme superplastique classique. La déformation se fait à une température de environ 500°C, et pourlimiter la friction la pièce est recouverte d’un lubrifiant solide, du nitrure de bore hexagonal.
Matrice supérieure
Pièce en aluminium
Matrice inférieure
Contrairement à la majorité des bancs de mesure il n’est paspossible d’utiliser une surface de contact ponctuelle (commeune sphère ou une pointe) car le lubrifiant est réparti sur lasurface en gouttelettes. Il est nécessaire d’utiliser une surfacede contact suffisamment grande pour qu’elle repose sur cesgouttelettes et ne rentre pas en contact direct avec l’aluminium,comme c’est le cas lors du procédé.Or il n’y a à ce jour aucun banc existant ou disponible quipermettent de réaliser des essais suffisamment similaires auprocédé HSBF et qui respectent ces conditions. C’est pourquoiil a fallu concevoir un nouveau banc d’essai.
Variables étudiées dans unpremier temps :- Vitesse de frottement- Nombre de passes- Pression- Température
Permet de fournir unemodélisation du coefficient
Variables étudiées dans un second temps :- Vieillissement du nitrure de bore (celui-ci se dégrade
progressivement à haute température)- Adhérence du nitrure de bore à la pièce lors de la déformation
(il n’a pas la même élasticité que la pièce ce qui peutentrainer son décrochage lors de la déformation)
- Influence de l’épaisseur et du taux de recouvrement du nitrurede bore
Permet une meilleure compréhension du procédé
Figure 2 : schéma explicatif du procédé HSBF
Outil (même acier que les matrices, déplacement vertical)
Chariot chauffant (déplacement horizontal)
Echantillon d’aluminium
Figure 6 : schéma du principe de mesure du coefficient de frottement
Figure 7 : CAD du banc d’essai
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
53Fabrication additive d'alliage d'aluminium F357 par fusion sur lit de poudre au laser
Additive manufacturing of F357 aluminum alloy using laser powder bed fusion
Aluminium alloys processed by additive manufacturing (AM) using laser powder bed fusion (LPBF) attracted industrial and research efforts striving from their lightweight advantage and combination of good mechanical and corrosion properties. In this study, manufacturing optimization followed by mechanical properties assessment was conducted. The typical features examined include the presence of defects, density, melt-pool signature (including width of the coarsened zone), microstructure evolution and microhardness. The tensile testing re-vealed superior performance in terms of yield strength, ultimate tensile strength and percentage elongation when compared to its cast counterpart, which is attributed to the fine microstructural morphology developed during processing.
Les alliages d'aluminium traités par fabrication additive à l'aide de la fusion au lit de poudre au laser ont attiré les efforts de l'industrie et de la recherche en raison de leur faible poids et de la combinaison de leurs propriétés mé-caniques et corrosives. Dans cette étude, l'optimisation de la fabrication suivie d'une évaluation des propriétés mécaniques a été réalisée. Les caractéristiques typiques examinées incluent la présence de défauts, la densité, la dimension du bain de fusion (y compris la largeur de la zone à grain grossiers), l'évolution de la microstructure et la microdureté. Les essais de traction ont révélé des performances supérieures en termes de limite d'élasticité, de résistance à la rupture et de pourcentage d'allongement par rapport à leur homologue moulé, ce qui peut être attribué à la morphologie microstructurale fine développée au cours du traitement.
Journée des étudiants – REGAL8 et 9 octobre 2019
Jaskaranpal Singh Dhillon1, José Alberto Muñiz-Lerma1, Mathieu Brochu1
1Centre de recherche sur l'aluminium REGAL, Département de génie des mines et des matériaux, Université McGill, Montréal, Québec, Canada H3A 0C5
Additive manufacturing of F357 aluminum alloy using laser powder
bed fusion.Fabrication additive d'alliage
d'aluminium F357 par fusion sur lit de poudre au laser.
Introduction
Advantages of laser powder bed fusion (LPBF)additive manufacturing techniques:• Complex shapes can be fabricated.• Design freedom but not beyond certain limits.• Ability to process different alloys.• Rapid solidification is one of the critical aspects of LPBF
which usually proves helpful in improving mechanicalproperties because of the resulting fine microstructure.
F357 aluminium alloy:• This alloy has lesser iron content as compared to A356
and A357 aluminium alloys.• Light weight, good mechanical properties and corrosion
resistance.• Can be processed using LPBF owing to its nearly eutectic
composition and also it has good weldability andcastability.
• Applications in automotive and aerospace industries.
LPBF processing
Figure 2: Schematic of part production using LPBF. Imagestaken from reference [2] and [3].
Figure 1: A schematic overview of selective laser melting process (Laser powder bed fusion process) both at the machine and powder scales [1].
Materials and methods• F357 powder.• Renishaw AM 400 is used for fabrication
purpose.• Powder characterization.• Parameter optimization.• Optical microscopy.• Scanning electron microscopy.• Microhardness measurements.
Table 1: Chemical composition of F357 powder
Results and Discussion
Figure 3: Scanning electron micrographs of F357 powder elucidatingpowder particle surface morphology.
D10 29.7 ± 0.4 µm
D50 46.0 ± 0.9 µm
D90 72.4 ± 1.8 µm
Particle size analysisFlow rate of 50 g powder,
s
Angle of repose, o
Apparent density, %
Hall 92.7 2.8 28.1 2.0 56.0Carney 10.9 0.1 29.3 0.2 56.2
Figure 4: Cohesive index (Rotating drum experiment)
Part density results using optical microscopy
Figure 4: Optical micrographs of samples with 99.96 % density
Figure 6: Width of coarsened zone variations
• Microhardness ofsamples producedusing different setof parametersseems to varyingover a same zone .
Figure 7: Optical micrograph depicting melt-pool features
(a) (b)
(c)
Figure 8: Optical micrographs showing melt-pool produced during LPBFelucidating melt-pool boundary and also depicting differentmicrostructural morphology at melt-pool boundaries as compared to themicrostructure evolved inside the melt-pool. (a), (b) and (c) As-builtcondition
Figure 9: SEM micrographs showing Silicon morphologies andpresence of Si and Mg particles as matrix strengtheners in Al-matrix
Conclusions• Part density of nearly 100 % is achieved.• Microhardness of as-built sample produced
using various energy densities has not variedsignificantly, however after heat treatmentmicrohardness has shown increase.
• Cellular microstructure is evolved.
References• King, W. et al. (2015). Laser powder bed fusion additive
manufacturing of metals; physics, computational, andmaterials challenges. Applied Physics Reviews, 2(4), 041304.
• https://www.royalmetalpowders.com/products/aluminum-/• https://www.sculpteo.com/en/materials/metals/slm-
aluminum-material/• Rao, J. H. et al. (2019). Multiple precipitation pathways in an
Al-7Si-0.6 Mg alloy fabricated by selective lasermelting. Scripta Materialia, 160, 66-69.
Microhardness measurements
Powder Characterization
Figure 5: Microhardness variations
Width of coarsened zone measurements
Melt-pool and microstructural morphology
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Jaskaranpal Singh DhillonMcGill University
J. A. Muñiz-Lerma - McGillM. Brochu - McGill
5454 Conception d’un prototype de platelage en aluminium pour les ponts aluminium / boisDesign of a prototype aluminum deck for aluminum / wood bridges
Il existe au Québec plusieurs ponts routiers dont le platelage et les poutres sont en bois et dont la durée de vie dépend de l’étanchéité du platelage. Le Ministère des Transports a installé en 2015 un pont totalement étanche constitué d’un platelage en aluminium sur des poutres d’acier. Le présent projet vise à concevoir un platelage en aluminium sur des poutres en bois en se basant sur les précédents projets réalisés à l’Université Laval. Les avantages de combiner ces deux matériaux sont : une empreinte environnementale faible, une totale étanchéité du platelage protégeant les poutres en bois, un porte-à-faux plus large en rive protégeant le bois contre les intempéries et les UV, une légèreté du platelage en aluminium réduisant ainsi le poids mort et le fluage des poutres en bois ainsi que l’abondance du bois et de l’aluminium au Québec. Ce projet s’inscrit dans les efforts économiques du gouvernement du Québec visant à augmenter l’utilisation du bois et de l’aluminium dans les infrastructures. De plus, une glissière de sécurité sera conçue et vérifiée par un test de crash numérique dans LS-DYNA. La conception se fera par la méthode des éléments finis en grands déplacements, plasticité et contact dans le logiciel NX.
In Quebec, there are several road bridges whose deck and beams are made of wood and whose service life depends on the impermeability of the deck. In 2015, the Ministry of Transport installed a completely waterproof bridge made of aluminum deck on steel girders. This project aims to design an aluminum deck on wooden beams based on the previous projects achieved at Laval University. The advantages of combining these two materials are: low environmental footprint; total impermeability of the deck protecting the wooden beams; wider cantilever protecting the wood against the weather and UV; lightness of the aluminum deck thus reducing the dead weight and the creep of the wooden beams; and the abundance of wood and aluminum in Quebec. This project is part of the Quebec government's economic efforts to increase the use of wood and aluminum in infrastructures. In addition, a guardrail will be designed and verified by a numerical simulation crash test in LS-DYNA. The design will be done using the finite element method taking into account the large displacements, plasticity and contact using NX software.
Amar DjedidUniversité Laval
C.-D. Annan - UL M. Oudjene - UL C. Beudon - UL M. Fafard - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Amar DJEDID1, Charles-Darwin ANNAN1, Marc OUDJENE1, Camille BEUDON1, Mario FAFARD1,2
1 Départment de Génie Civil et Génie de Eaux, Université Laval, 1065 Av. de la Médecine, Québec, QC, G1V 0A62 Centre d’expertise sur l’aluminium, Centre de commerce mondial 413, Saint-Jacques Street, Office 500, Montréal, Québec H2Y 1N9
Conception d’un platelage enaluminium sur poutres en bois, soudé
en FSW et sans effet composite
Design of an FSW welded aluminum deck on wood beam and without a composite effet
INTRODUCTIONIl existe au Québec plusieurs ponts routiers, en milieu rural, entièrement en bois et dont la durée de vie dépend de l’étanchéité du platelage.
Le Ministère des Transports a installé en 2015 un pont totalement étanche constitué d’un platelage en aluminium sur des poutres d’acier.
Le présent projet vise à concevoir un platelage modulaire en aluminium pour pont routier (autoroute), soudé en friction-malaxage, sur des poutres en bois en se basant sur les précédents projets de l’université Laval. La modularité permet de fabriquer les panneaux dans un environnement contrôlé, afin d’en garantir la qualité, mais aussi permettre une installation rapide sur site. Ceci réduit grandement les coûts et temps de construction, d’immobilisation ainsi que des déviations temporaires des routes. Les avantages de combiner ces deux matériaux sont : • empreinte environnementale faible ; • totale étanchéité du platelage protégeant les poutres en bois ;• porte-à-faux plus large en rive protégeant le bois contre les intempéries et les UV; • légèreté du platelage en aluminium réduisant ainsi le poids mort et le fluage des poutres en bois ; • Créer des débouchés commerciaux à ces deux matériaux abondants au Québec.
De plus, une glissière de sécurité sera conçue et vérifiée par un test de crash numérique dans LS-DYNA. La conception se fera par la méthode des éléments finis en grands déplacements, plasticité et contact dans le logiciel NX.
PROBLÉMATIQUELa conception des extrusions doit satisfaire à toutes les exigences des états limites de la norme CAN/CSA-S6-14, à savoir:• L’état limite de service ;• L’état limite en fatigue ;• L’état limite ultime. L’extrusion doit être conçue de façon à amoindrir les erreurs dimensionnelles lors de la production. Ceci permet de réduirel’ampleur des contraintes résiduelles lors du soudage. Aussi, les extrusions doivent être conçues pour pouvoir être soudées par la technique du soudage par friction-malaxage, sansavoir recours à toute forme de support (épaulement, pièce supplémentaire, etc..). Le but est d’avoir des soudures aboutées, sans le risque d’initiation de fissure due aux épaulements.Un mécanisme d’ancrage (connecteurs) du platelage aux poutres doit pouvoir s’adapter au déplacement relatif entre le platelage et les poutres en bois, dû à la différence entre le coefficient de dilatation thermique entre le bois (quasiment nul) et l’aluminium.De plus, le platelage doit être protégé lors d’un impact de véhicule sur les glissières de sécurité. Pour se faire, une extrusion sacrificielle, située entre les poteaux des glissières de sécurité et le platelage, agit comme fusible lors d’un impact pour protéger le platelage.Finalement, le platelage est conçu de façon à être extrudable et soudable au Québec, en fonction des capacités des équipements disponible dans la province.
VUE D’ENSEMBLE DU PONT PLATELAGE MODULAIRE (PANNEAUX PRÉFABRIQUÉS: P1 À P5)
P4P3
SOUDAGURES ABOUTÉES EN FSW SANS SUPPORT : GRACE AU BOBBING-TOOL
Le soudage FSW affecte thermiquement le métal de bases de façon largement moins sévère que le soudage MIG, augmentant ainsi considérablement toutes les limites de résistance des zones affectées thermiquement. De plus, souder avec un bobbing-tool permet d’obtenir des joints aboutés sans avoir recours à un support. Par conséquent, ça élimine les épaulements dont la partie non soudée donne souvent naissance aux amorces de fissure, faisant ainsi considérablement chuter la durée de vie en fatigue.Le combo : soudage FSW + bobbing tool permet de résistance largement supérieure au soudage MIG.
CONCEPTION ENTIÈREMENT EFFECTUÉES PAR DES ANALYSES EN ÉLÉMENTS FINIS VOLUMÉTRIQUES (3D)
Contraintes de von Mises à l’ÉLUL 1
Les analyses en éléments finis prennent en compte:• le comportement élastoplastique de l’aluminium ;• le caractère orthotrope du bois ;• le contact par friction là où applicable.La conception a été effectuée pour un pont routier, dont le débit de 4000 camions CL-625 par jour (poids total / camion est de 625kN) sur une durée de 75 ans.
L’ORTHOTROPIE DU BOIS LA COURBE ÉLASTOPLASTIQUE DE L’ALUMINIUM
Les panneaux fabriqués et pré-percés en usines sont imbriqués les uns aux autres en chantier. Ils sont reliés par des boulons aveugles: grand gain de temps et d’argent et qualité contrôlée.
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
55Effet du traitement de l’homogénéisation sur l'évolution des dispersoïdes Zr
dans les alliages Al-Mg-Si avec deux teneurs en Mg et SiEffect of homogenization treatment on Zr dispersoids evolution
in Al-Mg-Si alloys with two Mg and Si levels
Nano-sized dispersoids can control the grain structure during hot deformation process of many commercial aluminum alloys. A Comparative investigation was carried out on the precipitation behavior of Zr dispersoids in two 6xxx alloys having two Mg and Si levels with maintaining of Mg/Si atomic ratio at 1.0. The aim of this study is to find out an effective homogenization practice in terms of dispersoids density and size. For this purpose, Dif-ferent homogenization treatments were carried out at temperatures (450-550 ºC) and soaking times (2,6,12 and 24 hrs). The microstructure was characterized using optical microscopy and transmission electron microscopy. Electrical conductivity measurements used as an indication of the dispersoids precipitation out of the matrix. The results showed that there is a significant influence of higher Mg and Si content on the precipitation behavior of Zr dispersoids. While the effect is much less in case of low Mg and Si containing alloy. Furthermore, the type and the size of dispersoids changed from Al
3Zr to (Al,Si)
3Zr with increasing the temperature and soaking time.
Les dispersoïdes de taille nanométrique peuvent contrôler la structure granulaire lors de la déformation à chaud de nombreux alliages commerciaux d'aluminium. Une comparaison est réalisée sur la précipitation de disper-soïdes de Zr dans deux alliages 6xxx ayant deux teneurs en Mg et Si avec le même rapport atomique Mg/Si de 1,0. Le but de cette étude est de trouver l’homogénéisation efficace en termes de densité et de taille des dis-persoïdes. À cette fin, différents traitements d'homogénéisation ont été effectués à des différentes températures (450 550 ºC) et durées de trempage (2, 6, 12 et 24 heures). La microstructure a été caractérisée par microscopie optique et électronique à transmission. Les mesures de conductivité électrique sont utilisées pour indiquer la précipitation des dispersoïdes en dehors de la matrice. Les résultats montrent qu'il existe une influence signifi-cative des teneurs élevées en Mg et Si sur la précipitation des dispersoïdes de Zr. Tandis que l'effet est beaucoup moins important dans un alliage contenant peu de Mg et Si. De plus, le type et la taille des dispersoïdes sont transformés de Al
3Zr à (Al, Si)
3Zr avec l’augmentation de la température et de la durée de trempage.
Ali ElasheryUniversité du Québec à Chicoutimi
X.-G. Chen - UQACN. Parson - Rio Tinto
Journée des étudiants – REGAL8 et 9 octobre 2019
4. Results
Ali Elashery 1, X Grant Chen1, Nick parson 2
1 Research Chair in Metallurgy of Aluminum Transformation, University of Québec at Chicoutimi
2 Rio Tinto, Arvida Research and Development Center, Jonquière
Effect of homogenization treatment on Zr dispersoids evolution in Al-Mg-Si alloys
with two Mg and Si levels.Effet du traitement de l’homogénéisation sur l'évolution
des dispersoïdes Zr dans les alliages Al-Mg-Si avec deux teneurs en Mg et Si.
1. Introduction 2. Objectives
3. Experimental procedures
Si level
Zr level
Alloy ID.
Chemical composition, wt.%Mg/Siatomic
0.4% 0.2% L-Si Al - 0.2Fe – 0.4Si – 0.34 Mg – 0.2 Zr
1.001.00% 0.2% H-Si Al - 0.2Fe – 1.0 Si – 0.88 Mg –0.2 Zr
Cast alloys
5. Conclusions 6. Future work
Homogenization
Cycles
450
550
500
35
Tempe, 0C
Time
2h 6h 12h 24h
Water quench
Characterization
As-cast As-cast450 0C
500 0C
550 0C
2hr 12hr
450 0C
500 0C
550 0C
• Electrical conductivity measurements
• Optical Microscope
• Transmission Electron Microscope
(a) (b)
(c) (d)
(e) (f)
(a)
Fig.1 Microstructure evolution at differenthomogenization treatments of L-Si alloy
Fig.2 Electrical conductivity as a function ofhomogenization temperature and soaking time
Fig.3 TEM images showing the size and the type of dispersoids in case of L-Si alloy homogenized at500 0C for (a,b) 2hr and (c) 24 hr.
(C)
Fig.6 TEM images showing the size and the type of dispersoids in case of H-Si alloy homogenized at500 0C for (a,b) 2hr and (c) 24 hr.
Fig.5 Microstructure evolution at differenthomogenization treatments of H-Si alloy
Fig.4 Electrical conductivity as a function ofhomogenization temperature and soaking time
(b)(a)(b)
Al3Zr(Al,Si)3Zr Al3Zr
1. Based on optical micrographs, Zr dispersoids evolution is morepronounced in case of higher Mg and Si content.
2. The type of dispersoids changed from sphere Al3Zr to larger and elongated(Al,Si)3Zr with increasing the temperature and soaking time.
3. In case of high silicon alloys, the transformation from Al3Zr to (Al,Si)3Zroccurs at higher rate than in case of low silicon alloys.
4. The electrical conductivity decreased with increasing the homogenizationtemperature due to the increasing of solid solution level by Mg and Si.
1. Studying the effect of homogenization regime on flow stress curvebehavior.
2. Investigate the optimum homogenization treatment based on dynamic andstatic recrystallization resistance.
3. Further investigation to understand the transformation behavior of Zrdispersoids.
2hr 12hr(a) (b)
(c) (d)
(e) (f)
Zr-dispersoids
Mg2Si
Zr-dispersoids
Recrystallization resistance
laminar microstructure which is beneficial to the mechanical properties of the profiles
Nano-sized Zr-dispersoids Mg and Si amounts
• Study the effect of different homogenization treatments onthe precipitation behaviour of Zr dispersoids
• Investigate the effect of different Mg and Si amounts onthe precipitation behavior of Zr dispersoids.Affected
by
L-Si Alloy H-Si Alloy
(Al,Si)3Zr
(Al,Si)3Zr
Mg2Si
(C)
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Prix Award
5656Absorption d’humidité par de la poudre d’AlSi10Mg et effet sur son écoulementMoisture uptake of AlSi10Mg powder and the effect on its flowability
Les alliages d'aluminium sont largement utilisés dans les industries de l'aérospatiale et de l'automobile en raison de leur bonne conductivité thermique, de leur bonne résistance à la corrosion, de leur légèreté et de leur rapport coût / efficacité. Cherchant à accroître la complexité des pièces dans ces industries, la fusion sur lit de poudre au laser devient un moyen attrayant pour la fabrication de pièces en aluminium sophistiquées. Cependant, les alliages d'aluminium posent un défi dans ce type de procédé en raison de leur affinité avec l'hu-midité. Les molécules d'eau adsorbées entraînent des interactions plus fortes entre les particules qui affectent le comportement d'écoulement de la poudre. La qualité de la poudre étant un des facteurs clés pour obtenir des impressions de bonne qualité, la compréhension de l’écoulement de la poudre devient encore plus importante pour ce système métallique. Une caractérisation complète de la poudre AlSi10Mg a été réalisée pour comprendre l'effet de l'humidité adsorbée par les poudres d'aluminium. Une corrélation entre l’écoulement de la poudre et le comportement d'absorption de l'eau sera présentée dans ce travail.
Aluminium alloys are widely used in aerospace and automotive industries due to their good thermal conductivity, good corrosion resistance, lightweight and buy-to-fly ratio. Striving for increased complexity in parts in these industries, laser powder bed fusion becomes an attractive route to manufacture advanced Al parts. However, aluminium alloys pose a challenge in this type of process due to their affinity with moisture. Adsorbed water mol-ecules lead to stronger particle interactions affecting the flow behavior of the powder. As the powder feedstock quality is one of the key factors in obtaining good quality prints, understanding powder flow becomes even more important for this metallic system. Comprehensive powder characterization using AlSi10Mg was conducted to understand the effect of adsorbed moisture on aluminium powders. A correlation between powder flow and water sorption behavior will be presented in this work.
Eileen Ross L. EspirituMcGill University
K. E. Waters - McGill M. Brochu - McGill
Journée des étudiants – REGAL8 et 9 octobre 2019
q Adsorbed moisture increases particle – particle adhesion and particle – wall adhesion
Eileen Ross L. Espiritu1, Kristian E. Waters1, Mathieu Brochu1
1 Department of Mining and Materials Engineering, McGill University3610 University Street, Montreal, Quebec, Canada H2X 2C6
Moisture uptake of AlSi10Mg powder and the effect on its flowability
Absorption d’humidité par de la poudred’AlSi10Mg et effet sur son écoulement
P2[AM]2 Powder Processing and Additive Manufacturing of Advanced Materials
A B S T R A C TAluminium alloys are widely used in aerospace and automotive industries due to their good thermal conductivity, good corrosion resistance, lightweight and buy-to-fly ratio.Striving for increased complexity in parts in these industries, laser powder bed fusion becomes an attractive route to manufacture advanced Al parts. However, aluminiumalloys pose a challenge in this type of process due to their affinity with moisture. Adsorbed water molecules lead to stronger particle interactions affecting the flow behaviorof the powder. As the powder feedstock quality is one of the key factors in obtaining good quality prints, understanding powder flow becomes even more important for thismetallic system. Comprehensive powder characterisation using AlSi10Mg was conducted to understand the effect of adsorbed moisture on aluminium powders. A correlationbetween powder flow and water sorption behavior will be presented in this work.
BACKGROUND
OBJECTIVE To investigate the effect of moisture uptake of AlSi10Mg powder on its flowability
1. ALUMINIUM ALLOYS
q Properties: good thermal conductivity, corrosion resistant, lightweight and low buy-to-fly ratio
BACKGROUND [7]
q Forces affecting powder flow includes gravitational and interparticle forces
q Interparticle forces involve friction forces, mechanical interlocking, adhesion (liquid bridges) and cohesion
q Popular material in aerospace and automotive industries
q Its affinity to moisture poses a challenge in laser powder bed fusion (LPBF) process
q Adsorbed moisture could affectthe powder bed spreading
Issue with aluminium alloys
EXPERIMENTAL METHODPowder characterization
(AlSi10Mg)Sequential flow test at different RH conditions
Dynamic vapor sorption test
Particle size distribution [3]
Particle morphology [4]
Hall/Carney funnel [5]
Atmosphere controlled glove box Dynamic vapor sorption equipment [6]
RESULTS AND DISCUSSION
50 µm
Particle diameter, µµm
D10 47
D50 58
D90 68Adsorbed moisture /
gas porosity
100 µm
1. PARTICLE MORPHOLOGY AND PARTICLE SIZE DISTRIBUTION
2. POWDER FLOW BEHAVIOR AT DIFFERENT RELATIVE HUMIDITIES
3. POWDER’S WATER SORPTION BEHAVIOR
Particle morphology and particle size distribution of the powder
2. POWDER CHARACTERIZATION METHODS
Schematic of melt pool phenomena during powder bed melting [1]
q Several methods are standardized to characterize powder behavior during the LPBF process
q The most popular is the Hall/Carney funnel test due to its simple and straightforward operation
q This equipment could quantify the powder flowability by measuring its flow rate through a certain orifice size SORPTION, DESORPTION and
HYSTERESIS
q Water sorption increases with the exposure RH.
q Presence of hysteresis indicate irreversible surface reaction
Ø formation of surface oxide or hydroxide
q Particles are mostly round with small number of satellites
q Particles are relatively coarser than typical LPBF-size powders (15–45 µm)
OBSERVATIONS
SUMMARY AND CONCLUSIONS
References
q Powder flowability is affected by adsorbed moisture: ↑ exposure RH ↑ flow timeq Water sorption of powder increases with exposure RH; for example, powder exposed to
40% RH during the flow test, adsorbs the most during the vapor sorption test.q Powder flowability corresponds well with the results of vapor sorption test.
SORPTION
1. Hebert, R.J. J Mater Sci (2016) 51: 1165. https://doi.org/10.1007/s10853-015-9479-x2. ASTMB213 Metal Powders Flow Rate Apparatus. Date accessed: Sept. 06, 2019 from http://www.labulk.com/astmb213-metal-powders-flow-rate-apparatus/3. Camsizer X2. Date accessed: June 20, 2019 from https://www.retsch-technology.com/products/dynamic-image-analysis/camsizer-x2/function-features/4. Scanning Electron Microscope SU3800/SU3900. Date accessed: June 20, 2019 from https://www.hitachi-hightech.com/eu/product_detail/?pn=em-
su3800_su3900&version=5. ASTM B964 Metal Powders Carney Funnel Flowrate Meter. Date accessed: June 20, 2019 from http://www.labulk.com/astm-b964-metal-powders-carney-
funnel-flowrate-meter/6. Zhuravleva, M., Stand, L., Wei, H., Hobbs, C., Boatner, L. A., Ramey, J. O., ... & Tupitsyn, E. (2013). Hygroscopicity evaluation of halide scintillators. In 2013 IEEE
Nuclear Science Symposium and Medical Imaging Conference (2013 NSS/MIC) (pp. 1-5). IEEE.7. Mechanisms of particle interaction. Date accessed: Sept. 06, 2019 from https://www.freemantech.co.uk/_powders/powder-flowability-mechanisms-of-
particle-interaction
Hall funnel [2]
DisclaimerPart of the results was extracted from the work presented at the AMPM2019 Conference held lastJune 23-26, 2019 at Arizona, USA. It is published as a conference paper: E.R.L. Espiritu, J.A. MunizLerma, T. Turner, M.J. Conlon, D. Brouse and M. Brochu (2019). Understanding the factorsinfluencing powder spreadability for laser powder bed fusion. In AMPM2019. Arizona, USA.
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Prix Award
ex æquo
P2 [AM]2
5757
This work focuses on the protection of aluminum (Al) against corrosion using non-chromates coatings and inhibitors. As far as concerning coatings, two silicate-based coatings have been studied, namely, sodium silicate and zirconium silicate. Sodium silicate coatings were prepared by a simple ultrasonic dip-coating process on Al substrates. The effect of the pre-treatment of the Al substrate and the concentration of sodium silicate on the corrosion properties were investigated. Similarly, zirconium silicate coatings were prepared by a sol-gel process and the effect of the concentrations of precursors as well as the effect of the aging time of the sols on the corrosion properties were studied. The polarization resistance (R
p) for as-received Al was found to be 10 kΩ.cm2.
compared to 593 and 446 kΩ.cm2 for Sodium silicate and zirconium silicate, respectively. On the other hand, the corrosion properties of Al were studied by modifying the corrosive electrolyte with water-soluble compounds as inhibitors. Three inhibitors such as sodium silicate (Na2SiO3), manganese sulfate monohydrate (MnSO
4.H
2O),
and ammonium metavanadate (NH4VO
3) were tested. The R
p of the Al substrate in the presence of the inhibitors of
Na2SiO
3, Na
2SiO
3/MnSO
4.H
2O, and Na
2SiO
3/NH
4VO
3 was found to be 100, 133 and 679 kΩ.cm2, respectively. The
obtained results show that the studied compounds present a promising alternative for chromate-based coatings.
Ce travail porte sur la protection de l'aluminium (Al) contre la corrosion par des revêtements et d'inhibiteurs non-chromatiques. En ce qui concerne les revêtements, deux à base de silicate ont été étudiés : le silicate de sodium et le silicate de zirconium. Les revêtements de silicate de sodium ont été préparés par un simple procédé de revêtement par trempage par ultrasons sur Al. L'effet du prétraitement du substrat en Al et la concentration de silicate de sodium sur les propriétés de corrosion ont été étudiés. De même, des revêtements de silicate de zirconium ont été préparés par un procédé sol-gel et l’effet des concentrations de précurseurs ainsi que l’effet du temps de vieillissement des sols sur les propriétés de corrosion ont été étudiés. D’une part, la résistance de polarisation (R
p) de l’Al tel que reçu s’est révélée être de 10 kΩ.cm2 comparée à 593 et 446 kΩ.cm2 pour le
silicate de sodium et le silicate de zirconium, respectivement. D'autre part, les propriétés de corrosion de l’Al ont été étudiées en modifiant l'électrolyte corrosif avec des composés solubles dans l'eau comme inhibiteurs. Trois inhibiteurs tels que le silicate de sodium (Na
2SiO
3), le sulfate de manganèse monohydraté (MnSO
4.H
2O) et
le métavanadate d'ammonium (NH4VO
3) ont été testés. Le R
p du substrat en Al en présence des inhibiteurs de
Na2SiO
3, Na
2SiO
3/MnSO
4.H
2O et Na
2SiO
3/NH
4VO
3 s’établit à 100, 133 et 679 kΩ.cm2, respectivement.
Redouane FaridUniversité du Québec à Chicoutimi
D. K. Sarkar - UQACD. Levasseur - CMQ
Journée des étudiants – REGAL8 et 9 octobre 2019
Redouane FARID1, Dilip Kumar SARKAR1 et David LEVASSEUR2
1 Centre Universitaire de recherche sur l’Aluminium (CURAL) Université de Québec à Chicoutimi
2 Centre de métallurgie du Québec, 3095 Rue Westinghouse, Trois-Rivières
Development of mechanically durable ecologic and corrosion resistance non-
chromate coatings for aluminum
Développement des revêtements non chromatiques écologique et mécaniquement
résistant pour la protection de l’aluminium contre la corrosion
Introduction
• Textile manufacturing • Aircraft manufacturing• Metal finishing • Photography industry
Chromium-based
Ecologic coatings
Chromium-based coatings
Objectives1. Synthesis and characterization of non-chromate thinfilms with optimum preparation conditions on aluminum.2. Evaluation of the corrosion inhibition performance ofnon-chromate corrosion inhibitors for aluminum.
Experimental
Aluminum substrates were degreased using soap water inultrasonic bath and then rinsed with distilled water andethanol. The substrates were dried in the oven for night.The protection of aluminum against corrosion wasinvestigated in this work using two different methods.
Sodium silicate, zirconium silicate and zirconium-Stearicacid (Zr-SA) thin films were deposited on aluminumsubstrates and their morphology and corrosion propertieswere investigated.
The effect of silicate-based, manganese-based andvanadium-based inhibitors on the corrosion properties ofaluminum was investigated.
The morphology of the samples was studied usingscanning electron microscopy (SEM). The corrosionresistance properties of the samples was investigated viapotentiodynamic polarization curves acquired byelectrochemical experiments in NaCl solution (pH of 5.9).The electrochemical tests were performed using a PGZ100potentiostat and 300 cm3
–EC&G PAR flat cell,equipped with a standard three-electrodesystem with Ag/AgClreference electrode,a platinum mesh as acounter electrode andthe sample as a workingelectrode.
Results Protection of aluminum against corrosion by thin films
Protection of aluminum against corrosion by inhibitors
Conclusions
Sample Polarisationresistance Rp(kΩ.cm2)
Corrosion current density Icorr (nA/cm2)
Aluminum (NaCl) 23 422
Sodium silicate 340 79
Zirconium silicate 72 388
Zr-SA 932 43
Sample Polarisationresistance Rp(kΩ.cm2)
Corrosion current density Icorr (nA/cm2)
Without inhibitors 22 1300
Silicate-based 73 400
Manganese-based 126 180
Vanadium-based 545 5
(d)
(a)
(c)
(b)
SEM images of (a) as received aluminum substrate in NaCl solution, (b) Sodium silicate, Zirconium silicate, and Zr-SA thin films on aluminum substrate
The corrosion properties of non-chromates thin films and inhibitors were investigated.
Zr-SA thin films shows the better corrosion protection with Rp of 932 kΩ.cm2.
Vanadium-based inhibitor shows the best corrosion inhibition with Rp of 545 kΩ.cm2.
SEM images of (a) as received aluminum substrate in NaCl solution, (b) Sodium silicate, Zirconium silicate, and Zr-SA thin films on aluminum substrate
Samples preparation
Corrosion protection by thin films
Corrosion protection by inhibitors
Characterization tools
Image of water droplet on the surface of Zr-SAthin film
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Prix Award
Développement des revêtements non chromatiques écologique et mécaniquement résistant pour la protection de l’aluminium contre la corrosion
Development of mechanically durable ecologic and corrosion resistance non-chromate coatings for aluminum
5858
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Al alloys with small additions of rare elements such as Er and Sc exhibit significant improvement in mechanical properties that can be attributed to the precipitates formed by Al and the strengthening elements. However, upon equilibrium solidification, these strengthening elements have a very limited solid solubility in the Al matrix, which means that less precipitates can be formed from the solid solution. To achieve high strengthening effect, our attempt is to extend the solid solubility of the strengthening elements via the rapid solidification in additive manufacturing. Therefore, more precipitates are likely to be formed during processing or post-treatment and the strengthening effect can be increased. Differential Scanning Calorimetry (DSC) measurements of Al alloys with cooling rates up to 12.5 ºC/s were performed to understand the solidification behavior (microstructural evolution, phase transformation) within relatively low solidification rates. The results can be used for additive manufacturing technologies such as Arc Additive Manufacturing and Direct Energy Deposition.
Les alliages d'aluminium avec de petites additions d'éléments rares tels que Er et Sc présentent une amélioration significative des propriétés mécaniques qui peut être attribuée aux précipités formés par l’Al et aux éléments de renforcement. Cependant, lors de la solidification à l'équilibre, ces éléments de renforcement ont une solubilité solide très limitée dans la matrice d'Al, ce qui signifie que moins de précipités peuvent être formés à partir de la solution solide. Pour obtenir un effet de renforcement élevé, nous essayons d'étendre la solubilité solide des éléments de renforcement via la solidification rapide de la fabrication additive. Par conséquent, davantage de précipités risquent de se former pendant le traitement ou après et l'effet de renforcement peut être accru. Des mesures de calorimétrie différentielle à balayage (DSC) d’alliages d'Al avec des vitesses de refroidissement allant jusqu'à 12,5 ºC/s ont été effectuées pour comprendre le comportement de solidification (évolution de la micros-tructure, transformation de phase) à des vitesses de solidification relativement faibles. Les résultats peuvent être utilisés pour des technologies de fabrication additive, telles que la fabrication additive à arc électrique et le dépôt sous énergie concentrée.
An Fu McGill University
P. Hudon - McGillP. Bishop - Dalhousie
University M. Brochu - McGill
Application de la calorimétrie à refroidissement rapide à la fabrication additive d'alliages d'aluminiumApplication of fast cooling calorimetry in additive manufacturing of Al alloys
Journée des étudiants – REGAL8 et 9 octobre 2019
Near-eutectic temperature vscooling rate
Onset solidification temperature vscooling rate
Measured onset melting point: 567 °C
An Fu1, Pierre Hudon1 Paul Bishop2 Mathieu Brochu1
1 Department of Mining and Materials Engineering, McGill University, Montreal, Quebec, H3A 0C5, Canada2 Department of Mechanical Engineering, Dalhousie University, Halifax, Nova Scotia, B3H 4R2, Canada
Application of fast cooling calorimetry in additive manufacturing of Al alloys
DSC test of Al alloys
Small additions of rare elements e.g. Er, Sc induce appreciable increase in mechanical properties of Al alloys, which is due to the precipitation strengthening by Al3Er, Al3Sc for example. However, at equilibrium condition, these strengthening elements have limited solid solubilities in Al matrix. Therefore, the strengthening effect is limited.
Extend solid solubility and optimize microstructure via rapid solidification in AM.
ExperimentDSC test of Al alloys, Microstructural analysis of the tested samples& Thermodynamic modelling
Elements Er Sc ZrMaximum solid solubility in Al in equilibrium state (wt.%)
0.4 0.33 0.28
UTS (MPa) YS (MPa)
5083 290 192
5083+0.3Er+0.1Zr 305 202
Perkin Elmer DSC 8000
Normal DSC Hyper DSC Temperature range (°C) -150 to 2400 -180 to 750
Cooling rate (°C/s) 1 10
Coolant Air Liquid nitrogen
Tested alloys
Pure Al Al-Sc, Fe, NiUndercooling controlled only by thermal diffusion Precipitates strengthened alloys
AlSi7MgCasting alloy with characteristic structure affected by cooling rates1. 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 = 𝑎𝑎 ∙ 𝑆𝑆−𝑛𝑛 : cooling rate , : material constants Secondary Dendrite Arm Spacing2. TMK model (high cooling rates)𝛿𝛿∆𝑇𝑇 = 1 + 𝜑𝜑2 𝐾𝐾 𝛿𝛿 eutectic spacing, ∆𝑇𝑇 undercooling , φ constants for lamellar eutectic
Background
Welded joints of Al 5083
Rapid solidificationThe cooling rate far exceeds the equilibrium cooling rate Effects:Extend solid solubilityPromote chemical homogeneity Refine microstructure Enlarge undercooling
Additive manufacturing
AM technologies Normal cooling rate /sArc Additive Manufacturing 10-102
Direct Laser Deposition 103-104
Laser Powder Bed Fusion 105-106
Objective
Solidifying the fully melted alloys at the cooling rates ranging from 2.5 to 10/s
Results and summaryPure Al (Al0.2Sc, Al-1Fe,Ni)
AlSi7Mg
1. Measured onset melting point: 661 °C2. Measured onset solidification temperature
Onset solidification temperaturevs cooling rate
According to the equilibrium phase diagram constructed by Thermo-Calc , upon cooling, the first peak corresponds with the nucleation of primary α-Al and the second peak represents the near-eutectic reaction.
With increased cooling rate, the undercooling corresponding to nucleation of primary α-Al exhibits linear increase, while that for near-eutectic reaction shows parabolic increase.
For all the tested alloys, only one peak is detected, which represents melting or solidification. The undercooling regarding solidification exhibits linear increase upon increased cooling rate.
ReferencesTrivedi, R., F. Jin, and I. E. Anderson. "Dynamical evolution of microstructure in finely atomized droplets of Al-Si alloys." Acta materialia 51.2 (2003): 289-300.
Dongxia, Yang, et al. "Effect of minor Er and Zr on microstructure and mechanical properties of Al–Mg–Mn alloy (5083) welded joints." Materials Science & Engineering A 561 (2013): 226-231.
Prix Award
P2 [AM]2
59Influence des paramètres de soudage par friction malaxage sur les propriétés
mécaniques locales et la microstructure des alliages Al-MgInfluence of friction stir welding process parameters on local mechanical
properties and microstructure of Al-Mg alloys
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Philippe Grandmont ÉTS
L. St-Georges - UQACA. Rahem - UQAC
The objective of the project is to establish a quantitative link between the welding parameters (heat input) and the local mechanical properties of the different zones affected by the welding process in ECO Al-Mg alloys. The local mechanical properties of the areas affected by welding are not very well understood to this date due to the inability of conventional tensile tests to precisely target a single area of the weld. A method of mechanical testing by shear-punch is therefore used to solve this problem. The mechanical shear properties from the test are then used to estimate the equivalent tensile properties through an empirical relationship specific to the tested material and the shear-punch apparatus. Preliminary samples seem to indicate that the heat input has an effect on the local mechanical properties of the welds. This research project is carried out in collaboration with the Korean Institute of Industrial Technologies (KITECH) which provide us with the studied ECO Al-Mg material. These alloys containing 6 to 8% magnesium are the results of a new method developed by KITECH to produce Al-Mg alloys in a more ecological way. There is very few information on FSW of these alloys.
L'objectif du projet est d'établir un lien quantitatif entre les paramètres de soudage (apport de chaleur) et les propriétés mécaniques locales des différentes zones affectées par le soudage des alliages ECO Al-Mg. Les propriétés mécaniques locales des zones affectées par le soudage constituent un aspect très peu étudié à ce jour en raison de l’incapacité des essais de tractions conventionnels à cibler précisément une seule zone de la soudure. Une méthode d’essais mécanique par poinçonnage est donc utilisée pour remédier à ce problème. Les propriétés mécaniques en cisaillement issues du test sont ensuite utilisées pour estimer les propriétés équiva-lentes en tension par l’entremise d’une relation empirique spécifique au matériau et au montage des essais. Les échantillons préliminaires produits semblent indiquer que l’apport de chaleur a un effet sur les propriétés mécaniques locales des soudures. Ce projet de recherche est effectué en collaboration avec l’institut coréen de technologies industrielles (KITECH) qui nous fournissent le matériau ECO Al-Mg étudié. Ces alliages contenant 6 à 8 % de magnésium sont issus d’une nouvelle méthode développée par KITECH pour produire des alliages Al-Mg de façon plus écologique. Il existe très peu d'information sur le soudage FSW de ces alliages.
Journée des étudiants – REGAL8 et 9 octobre 2019
Philippe Grandmont
1 Department de génie Mécanique, École de technologie supérieure, 1100, rue Notre-Dame Ouest, Montréal, QC, H3C 1K3
Influence des paramètres de soudage par friction- malaxage sur les propriétés
mécaniques locales des alliages Al-Mg
Influence of friction stir welding (FSW) process parameters on local mechanical properties of Al-Mg
alloys
Introduction
Objectifs et défis
aMéthodologie
Résultats et discussions
Prochaines étapes et pistes à explorer
Les alliages Al-Mg corroyés (série 5000) sont des alliages non thermiquement traitables largement utilisés dans l’industrie automobile et navale en raison de leur haute résistance à la corrosion, leurs bonnes propriétés mécaniques et leur excellente formabilité. Ces propriétés rendent les alliages Al-Mg parfaits pour des applications comme les structures d’automobiles faites à partir de tôles assemblées par soudage puis embouties.
Les essais de poinçonnage (Shear Punch Test ou SPT) permettent déterminer les propriétés mécaniques locales en cisaillement en ciblant chacune des zones affectées. Ensuite, une estimation des propriétés équivalente en tension (YS, UTS, El%) est effectuée à partir des données en cisaillement.
La microscopie optique et MEB sont utilisés pour analyser la microstructure et la microscopie EBSD est utilisée pour l’identification des phases, la cartographie de la taille des grains et la texture cristallographique.
Pour finir, des essais de microdureté Vickers sont effectués pour comparer les contraintes d’écoulement (YS) mesurées au poinçonnage et la taille des grains observés dans chaque zone.
Objectifs
Établir un lien quantitatif entre les paramètres de soudage (apport de chaleur) et les propriétés mécaniques locales mesurées dans les zones affectées par le soudage (NZ, TMAZ, HAZ)
Lier les propriétés mécaniques locales aux microstructures observées dans ces zones Déterminer les paramètres optimaux pour le soudage des Al-6Mg et Al-8Mg
https://www.kirchhoff-automotive.com/products/
Essais préliminaires:
Une soudure froide et une soudure chaude sont effectuées dans un Al-6Mg à 120 mm/min et 50 mm/min respectivement. Une vitesse de rotation de 800 RPM est utilisée dans les deux cas.
Les soudures sont ensuite découpées pour produire des spécimens pour la microscopie optique et EBSD. Deux spécimens minces (0.65 mm) sont aussi produits pour les tests de poinçonnage.
Afin d’évaluer la formabilité des joints, il est important de connaître les propriétés mécaniques locales dans les zones affectées par le soudage FSW.
À ce jour, il existe très peu d’études portant sur les propriétés mécaniques locales des soudures FSW et sur le soudage des alliages Al-Mg non commerciaux contenant plus de 5% de magnésium. Dans ce projet, l’effet des paramètres de soudage FSW sur les propriétés mécaniques locales des alliages Al-6Mg et Al-8Mg sera analysé.
Les alliages étudiés sont issus d’une nouvelle méthode de production développée par KITECH (Korea Institute of Industrial Thechnology) permettant de produire des alliages Al-Mg de façon plus écologique, sans gaz à effet de serre.
Défis
Les essais de traction conventionnels ne peuvent pas être utilisés pour évaluer les propriétés locales en raison de la petite taille et de la géométrie des zones étudiées
Les Al-Mg sont difficiles à souder en raison de leurs résistances à chaud qui augmente drastiquement avec la vitesse de déformation, très élevée en FSW
La rigidité de la machine et du montage est critique Zones typiques d’un joint FSW
a. Métal de base (BM)b. Zone affectée thermiquement (HAZ)c. Zone affectée thermo mécaniquement (TMAZ)d. Zone malaxée (NZ)
Soudure froide – 160 mm/minSoudure chaude – 50 mm/min
Zones de poinçonnage à différentes distances
du centre du joint
IPF dans la zone thermo mécaniquement affectée (TMAZ)Soudure chaude – 50 mm/min
Taille moyenne des grains : 43.8 μmSoudure froide – 160 mm/min
Taille moyenne des grains : 42.3 μm
Figures de pôles inverses (IPF) dans la zone malaxée (NZ)Soudure chaude – 50 mm/min
Taille moyenne des grains : 9.91 μmSoudure froide – 160 mm/min
Taille moyenne des grains : 6.63 μm Dans la NZ, on retrouve une structure de grains fins recristallisés typique en FSW. La soudure chaude produit de plus gros grains (environ 30%) que la soudure froide
Dans la TMAZ, on retrouve une structure de grains déformés parfois entourés de grains recristallisés. La soudure froide produit des grains plus déformés alors que la soudure chaude produit une certaine recristallisation autour des grains déformés et une meilleure relaxation des contraintes.
Discussions Les deux soudures comportent des défauts volumétriques importants. Ceux-ci proviennent
probablement du manque de rigidité de la machine utilisée et/ou d’un manque d’agitation du matériau par l’outil.
Dans les deux soudures, la résistance dans la NZ est plus élevée (18 MPa) que celle du métal de base. La variation de la vitesse ne semble pas avoir d’effet significatif sur la résistance cette zone.
La microdureté de la NZ est plus élevée ( 8 HV) dans la soudure froide que dans la soudure chaude, probablement en raison de la plus petite taille de grains
La dureté et la résistance dans la TMAZ de la soudure froide sont légèrement plus élevées (4 HV et 12 MPa ) que dans la soudure chaude, possiblement causée par l’écrouissage plus important des grains dans la soudure plus froide.
Les propriétés mécaniques dans la HAZ ne semblent pas affectées significativement par le soudage
Afin d’éliminer les défauts, un outil fileté plus petit a été conçu afin d’aider au malaxage du matériau et de réduire les forces de soudage. De plus, l’inclinaison du montage semble aussi aider à l’obtention d’un meilleur contact de l’épaulement du côté avançant de la soudure.
Lorsque des paramètres viables seront trouvés, un plan expérimental complet sera effectué pour déterminer une plage de vitesses produisant des soudures sans défauts dans les Al-6Mg et Al-8Mg
Afin de procéder à l’évaluation des contraintes d’écoulement et d’allongement à la rupture, une calibration du test de poinçonnage sera effectuée pour réduire l’erreur associée aux déformations du poinçon et du montage. Ces propriétés serviront à quantifier la formabilité des joints FSW.
Des tests de pliage et d’emboutissage pourraient aussi servir à l’évaluation de la formabilité des joints.
L’évolution des précipités (Al3Mg2) sera aussi étudiée dans les différentes zones de la soudure. Une simulation du cycle thermique FSW sera effectuée dans un dilatomètre afin d’isoler les effets thermiques des effets mécaniques sur la microstructure est les propriétés mécaniques.
RemerciementsTout d’abord, merci à KITECH de nous avoir fourni le matériau étudié dans ces essais préliminaires ainsi quedans les prochaines étapes du projet. Je tiens aussi à remercier le REGAL, sans qui un projet académiquecomme celui-ci serait difficile à mettre en place. Finalement, je remercie mon directeur maîtrise,Professeur Mohammad Jahazi, pour son soutient infaillible dans mon processus d’apprentissage.
6060
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Aluminum has a wide range of applications from preserving food to engine blocks, making this metal and its re-lated alloys an inseparable part of modern life. In order to improve aluminum for these applications, its chemical and structural properties should be investigated, which is achievable by using electron microscopes. However, the electron microscopes may damage the specimen during the investigation process. When electrons beam undergoes high angle scattering events, the law of conservation of momentum required momentum transfer to the atoms in the sample. In cases where the transferred energy is large enough to overcome a barrier called displacement energy (Ed), the atom can be displaced from its original lattice site. This displacement of atoms is called knock-on displacement, a significant source of electron beam damage. Thus, knowing the displacement energy is vital to avoid knock-on damage during the electron-microscopic investigation. In this study the dis-placement energy of aluminum atoms is calculated with the Density Functional Theory (DFT) based method called Nudged Elastic Band (NEB). The displacement energies are calculated for several types of atomic displacement in the crystal and these values are then applied to find the optimum working condition of electron-microscopes.
L'aluminium possède un large éventail d'applications, allant de la conservation des aliments aux blocs-moteurs, ce qui en fait un élément inséparable de la vie moderne. Afin d'améliorer l'aluminium pour ces applications, il convient d'étudier ses propriétés chimiques et structurelles, ce qui est réalisable à l'aide de microscopes élec-troniques. Cependant, les microscopes électroniques peuvent endommager l'échantillon au cours du processus d'enquête. Lorsque les électrons du faisceau subissent une dispersion à angle élevé, la loi de conservation de la quantité de mouvement exige un certain transfert aux atomes de l’échantillon. Dans les cas où l'énergie transférée est suffisamment importante pour surmonter une barrière appelée énergie de déplacement (Ed), l'atome peut être déplacé de son site d'origine. Ce déplacement d'atomes est appelé premier atome frappé, source importante de dommages causés par un faisceau d'électrons. Il est donc essentiel de connaître l’énergie de déplacement pour éviter des dommages en chaîne lors de l’investigation au microscope électronique. Dans cette étude, l’énergie de déplacement des atomes d’aluminium est calculée à l’aide de la méthode basée sur la théorie fonctionnelle de la densité (DFT) appelée « Nudged Elastic Band » (NEB). Les énergies de déplacement sont calculées pour plusieurs types de déplacement atomique dans le cristal.
Ali Jaberi McGill University
R. Gauvin - McGill
Étude des dommages induits par le faisceau d'électrons dans l'aluminiumInvestigation the electron-beam induced damage in Aluminum
Journée des étudiants – REGAL8 et 9 octobre 2019
Ali Jaberi1, Raynald Gauvin11 Department of Mining and Materials Engineering, McGill University, 845
Sherbrooke Street West, Montreal, Quebec, Canada
Investigation of electron-beam induced damage in AluminiumÉtude sur l’endommagement de
l’aluminium dû au faisceau d’électrons
Introduction
During electron microscopic analysis, the high energy electron beam candamage the materials and this makes the characterization inaccurate. One ofthe most significant beam damage in conductive materials is knock-ondamage [1]. In this damage, the atom within the material is knocked outfrom its original site. This pross can be occurred in four forms as follow[2]:
Method
Through the law of conservation of momentum, the required energy ofbeam-electron to displace the atoms can be calculated by [3]:
So, the key parameter that should be determined is the displacement(barrier) energy which can be calculated by finding the saddle point inatomic diffusion process:
A : atomic weight Ed : displacement energy
18
from its site in the sample, then it will be ejected from that site. Knock-on damage can
occur in four forms, (Figure 2.1).
Figure 2.1. Schematic diagram of different forms of knock-on damage. a) Surface
sputtering: electron beam sputters atoms from back surface of sample, causing mass
loss. b) Bulk displacement: electron beam displaces atoms into interstitial sites, or into
other atoms, causing atomic cascades. c) Vacancy enhanced displacement: electron
beam displaces atoms into vacant sites. d) Surface adatom displacement.
Results
The displacement energies of all types of knocked-on damage for pureAluminum were calculated with Nudged Elastic Band method which isbased on Density functional theory and is illustrated bellow :
Since the barrier energy of adatom displacement is sufficiently small, thistype of displacement can be already thermally mobile and should not beconsidered for knocked-on damage [4].
Ener
gy
Traveling Distance
1
2
3
Discussion and Conclusion
Putting the barrier energy values into the conservation of momentumequation, the threshold displacement energy of electron beam can becalculated:
The results suggest that, to prevent the knock-on damage, the beam energyshould be below 13.8 KeV since exactly above this energy we have thevacancy enhance displacement. Also, if we are using TEM, we can have allbulk displacement, sputtering and vacancy displacement if the beam energyis above 228 KeV.
In this study, we have found the appropriate beam energy of electronmicroscopes in order to prevent knock-on damage which is vital for anaccurate characterization.
Bulk Displacement
Vacancy Displacement
Sputtering
Bea
m E
nerg
y (K
eV)
0
50
100
150
200
250
300
48.86
13.81
228.12
References:[1] Levin and D. A. Muller, “Physical limitations on transmission electron microscope imaging of lithiumbattery materials”, Microscopy and Microanalysis, vol. 19, no. S2, pp. 1228–1229, 2013[2] D. A. Muller and J. Silcox, “Radiation damage of ni3al by 100 keV electrons” Philosophical MagazineA , vol. 71, no. 6, pp. 1375–1387, 1995[3] J. Hren, Introduction to analytical electron microscopy. Springer Science & Business Media, 2013.[4] R. Egerton, “Beam-induced motion of adatoms in the transmission electron microscope”, Microscopyand Microanalysis, vol. 19, no. 2, pp. 479–486, 2013
61Résistance en fatigue des pièces en alliage d’aluminium produits par fusion sélective laser
Fatigue strength of aluminum alloys parts manufactured by selective laser melting
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Luiz Felipe Leitao MartinsÉcole Polytechnique de Montréal
P. R. Provencher - École Polytechnique de Montréal Mathieu Brochu - McGillMyriam Brochu - École Polytechnique de Montréal
Fatigue strength is one of the critical properties for some parts manufactured by additive manufacturing. Stress concentrations generated by internal defects and surface roughness intrinsic to the process constitute prefe-rential sites for crack initiation. The objective in this project is to quantify the fatigue strength of aluminum alloys produced by selective laser melting. The experimental strategy involves the study of the building direction, hea-ting of the building platform and possible thermal post-treatments. The staircase methodology is used to quantify the probability of failure for a stress amplitude leading to 106 cycles in rotation bending fatigue. The results will be analyzed by studying the microstructure and the damage mechanisms. Ways to improve the fatigue behavior of these alloys dedicated to additive manufacturing will be proposed.
La résistance en fatigue est une des propriétés critiques pour certaines pièces produites par fabrication additive. Les concentrateurs des contraintes générées par les défauts internes et la rugosité de surface intrinsèques au procédé constituent des sites préférentiels pour l’amorçage des fissures. L’objectif du projet de recherche est de quantifier la résistance en fatigue d’alliages d’aluminium produits par fusion sélective par laser. La stratégie expérimentale implique l’étude de la direction d’impression, du chauffage de la plateforme d’impression et de possibles post-traitements thermiques. La méthodologie de l'escalier est utilisée afin de quantifier la probabi-lité de rupture pour une amplitude de contrainte menant à 106 cycles en flexion rotative. Les résultats seront analysés en étudiant la microstructure et les mécanismes d’endommagement observés. Ainsi, des pistes pour améliorer la tenue en fatigue des alliages dédiés à la fabrication additive pourront être proposées.
Journée des étudiants – REGAL8 et 9 octobre 2019
Luiz Felipe LEITAO MARTINS1, Paul R. PROVENCHER1, Mathieu BROCHU2 et Myriam BROCHU1
1 Polytechnique Montréal2 McGill University
Résistance en fatigue des pièces en alliage d’aluminium produites
par fusion sélective laserFatigue strength of aluminum alloy parts manufactured by selective laser melting
Fabrication des éprouvettesAfin d’évaluer les conditions de conception et les paramètres importants pour les piècesFA en fatigue flexion rotative à HCF, il a été produit 15 éprouvettes en alliage AlSi7Mg (Al- 7,5%Si - 0,6%Mg) à partir de poudres sphériques. La poudre a été générée dans unenvironnement inerte. Les pièces ont été imprimées dans les directions X-Y (horizontale)sur une plateforme à 30 °C d’une machine SLM.
Méthodologie statistique de l’escalierLa méthode de l’Escalier sert à trouver le seuil de la fatigue endurance de manièreéconomique [4] (Fig. 6). Ceci se fait en effectuant des essais autour du seuil de la fatigueendurance en variant la contrainte, selon un certain pas. Certains facteurs sont àconsidérer dans les essais fatigue flexion-rotative: vibrations, fléchissement du montage,glissement dans les mors et un facteur statistique de pas (s) de l’escalier.La vibration a été contrôlée par le diamètre de la section réduite. Le pas a été déterminéà partir des valeurs typiques des bandes d’amplitudes de contrainte à ce nombre decycles [5]. Le nombre d’échantillons et le temps totale disponible pour les essais ont étéconsidérés comme des contraintes pour son choix. La contrainte a été variée d’un écartconstant de 5 MPa.
Paramètres d’essai fatigueR = - 1 ; f = 50 Hz; N = 106 cycles (cible); pas = 5 MPa.Design des éprouvettes (Fig. 5).
3. MÉTHODOLOGIE DES ESSAIS DE FATIGUE
Fusion sélective par laser (Selective Laser Melting - SLM)La fusion sélective par laser est une méthode de mise en forme par FA dans laquelle unlaser est utilisé pour la fusion d’une couche de poudre, de façon que chaque couche estdéposée sur la précédente. Des discontinuités ponctuelles sont formées lors de laconsolidation par fusion d’un lit de poudre à partir de l’emprisonnement du gaz, de lamanque de fusion, des effets de faible mouillabilité et de la vaporisation dans le bain defusion. Dans le cas des alliages d’aluminium, un film d'oxyde qui empêche la diffusionpeut se former à la surface des particules (Fig. 2).Afin de produire une pièce finale sans défauts, il faut posséder une compréhension del'interaction de la poudre avec les paramètres liés au processus de fabrication.
Caractéristiques de la poudrePendant la production d’une poudre par pulvérisation, le gaz résiduel de la poudre peutaussi générer des pores. La granulométrie, la taille et la morphologie de la poudre (Fig. 3)ont donc une influence sur les gaz emprisonnés et par conséquent la densité relative despièces produites (Figs. 3 et 4).Une granulométrie fine expose généralement une plus grande surface de fusion, unefusion plus complète et des pièces plus denses. En revanche, la granulométrie grossièremène à formation des pores dans les pièces produites.Ces pores sont des sites préférentiels d’amorce de fissures en fatigue.
2. REVUE DE LITTÉRATURE
D’après les résultats obtenus de résistance à la fatigue, la méthode de l’escalier a étévalidée par comparaison avec la littérature existante.
Un pas a été déterminé pour la méthode de l’escalier, car il a été possible de faire uneestimation de la moyenne et de valider les conditions du bris ou de la survie deséprouvettes. Ce pas servira donc pour les essais futurs.
Pour bénéficier de l'amélioration de tenue en fatigue qui pourrait résulter de l'utilisationd'une poudre monodispersée, il est nécessaire d‘améliorer les paramètres de dépôt afinde produire des pièces saines. Le prochain jalon vise à quantifier la résistance en fatiguede poudres monodispersées produites dans conditions nominalement optimisées.
4. CONCLUSIONS ET AVENIR
manque de fusion
[1] Chaman Lall et al., Aluminum Powder MetallurgyMaterials and Processes, Aluminum Science and Technology, ASM International, 2018.
[5] N. E. Uzan, R. Shneck, O. Yeheskel, and N. Frage, Fatigue of AlSi10Mg specimens fabricated by additive manufacturing selective laser melting, Materials Science and Engineering: A, vol. 704, pp. 229-237, 2017.
[3] N. Aboulkhair, I. Maskery, I. Ashcroft, C. Tuck, and N. M Everitt, The role of powder properties on the processability of Aluminium alloys in selective laser melting, 2015.
[2] Louvis Eleftherios, Fox Peter, Sutcliffe Christopher, Selective laser melting of aluminium components, Journal of Materials Processing Technology, 2011.
Mise en contexte:À partir du dépôt couche par couche par fabrication additive (FA) il est possible d’obtenir des pièces à géométries complexes en un temps relativement courtde production [1]. L’économie de matière brute est remarquable. La bonne combinaison d’une haute résistance mécanique, de faible densité et d’uneexcellente résistance à la corrosion fait des pièces en alliage d'aluminium produites par FA un choix attrayant dans la sélection des matériaux pour lafabrication des véhicules légers comme les aéronefs (Fig. 1).Objectif du projet: Quantifier la résistance en fatigue de pièces en alliage d’aluminium produites par FA à partir de poudres monodispersées. Sachant que lagranulométrie des poudres utilisées pour la mise en forme a une influence importante sur la qualité des pièces produites par FA et aussi sur leur résistance enfatigue, le projet vise à maîtriser ces poudres et contrôler le comportement en fatigue des pièces qui en sont produites.
1. INTRODUCTION
Figure 3: Morphologie des particules de poudres [A] et [B] de l’alliage AlSi10Mg [3].
Figure 5: Design des éprouvettes.
Figure 6: Résultats préliminaires du comportement en fatigue de l’AlSi7Mg par la méthode de l’escalier
Figure 8: Site d’amorce de la rupture à 500X de l’éprouvette #3.
Figure 1: Exemple d’application de conception des pièces structurales critiques
d’avion fait par des alliages d’aluminium
[A] [B]
Fiabilité des résultatsLes bandes à la Fig. 6 comparent lesessais à la littérature [5], confirmant lajustesse des résultats. La Fig. 8 illustreun site d’amorce d’une fissure due à undéfaut de manque de fusion. Pourconfirmer la fidélité des essais, il a étévérifiée que les éprouvettes onttoujours été rompues au centre de lasection réduite (Fig. 7).
Moyenne = 110.8 MPaÉcart-type = 7.4 MPa
Figure 7: Rupture de l’éprouvette #8 en flexion
rotative.
[4] W. J. Dixon, The Up-and-Down Method for Small Samples, Journal of the American Statistical Association, vol. 60, pp. 967-978, 1965.
Figure 4: Densité relative des pièces (A) et (B) [3].
Figure 2: Solidification et dissociation des oxydes lors de la fusion des poudres [2].
6262 L’évolution de la conductivité électrique dans les composites Al-B4C avec différentes teneurs en B4CEvolution of electrical conductivity in Al-B4C composites with various B4C contents
Les composites à matrice métallique renforcée par des particules (CMM) sont largement utilisés en tant que composants de structure fonctionnant à haute température et la conductivité thermique (CT) devient de plus en plus critique afin d’assurer leurs excellentes performances. La conductivité électrique (CE) a été bien développée en tant qu'indicateur sur la CT dans les alliages d'aluminium, mais peu de travaux sur la relation entre la CE et la CT dans les CMM ont été publiés. Par conséquent, ce projet vise à établir la relation entre la CE et la CT dans les composites à matrice d’Al. Au début du projet, l’évolution de la CE dans les composites Al-B4C contenant 10 à 21 % en volume de particules de B4C a été mesurée dans ce travail. Les résultats montrent que la CE du composite diminue avec l’augmentation de la fraction volumique des particules de renforcement. Parallèlement, l’évolution de la microstructure, y compris la distribution des particules dans la matrice, la couche d’interaction entre les particules et la matrice, ainsi que les différentes productions de réactions, ont été bien caractérisées pour dévoiler leurs influences sur l’évolution de la CE.
Particle-reinforced metal-matrix composites (MMCs) are widely used as structure components working at high temperature and the thermal conductivity (TC) becomes more and more critical to ensure their excellent per-formance. Electrical conductivity (EC) has been well developed as the indicator for TC in aluminum alloys, but little work on the relationship between EC and TC in MMCs is published. Therefore, this project is aiming to establish the relationship between EC and TC in Al matrix composites. As the start-up of project, the evolution of EC in Al-B4C composites with 10-21 vol.% B4C particles has been measured in this work. The results show that the EC of the composite decreases with the increase of the volume fraction of the reinforcement particles. Meanwhile, the evolution of microstructure, including the distribution of particles in the matrix, interaction layer between particles and matrix as well as the various reaction productions have been well characterized to discover their influence on the evolution of EC.
Cong LiUniversité du Québec
à Chicoutimi
L. Kiss - UQACK. Liu - UQAC
X.-G. Chen - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
Cong Li1, Laszlo Kiss2, Kun Liu1, X. Grant Chen1
1 Research Chair in Metallurgy of Aluminum Transformation, UQAC2 Department of Applied Science, University of Quebec at Chicoutimi
Evolution of electrical conductivity in Al-B4C composites with various B4C
contents
L’évolution de la conductivité électrique dans les composites Al-B4C avec
différentes teneurs en B4C
Particle-reinforced metal-matrix composites (MMCs) are widely used as structure
components working at high temperature and the thermal conductivity (TC)
becomes more and more critical to ensure their excellent performance. Electrical
conductivity (EC) has been well developed as the indicator for TC in aluminum alloys,
but little work on the relationship between EC and TC in MMCs is published.
Therefore, this project is aiming to establish the relationship between EC and TC in Al
matrix composites. As the start-up of project, the evolution of EC in Al-B4C
composites with 11-21 vol.% B4C particles has been measured in this work.
Materials Compositions ::
Microstructural characteristics and property testOptical microscopy, electrical conductivity tester are used in this study. The CLEMET image analysis software is used to calculate the volume fraction of B4C particles and reaction products.Investigating the effect of Al-B4C composites with 11-21 vol.% B4C particles and
various reaction productions on the evolution of EC.
1. To study the relationship between EC and TC and the factors affecting TC.
2. To improve the TC by heat treatment (Solid solution, aging, annealing), hot
deformation and any other potential methods.
1. The EC of the composite decreases with the increase of the volume fraction of the B4C particles.
2. The EC of MMCs is mainly affected by the B4C content. TiB2 and Al3BC also have an effect on the
EC.
MMCsVolume of reinforced
particles (Vol. %)Condition
1100-B4C
11Billet slice(as-cast) and Rolled
plate
19Billet slice(as-cast) and Rolled
plate
21Billet slice(as-cast) and Rolled
plate
Fig. 3. Volume fraction of the B4C particles and reaction products. (a) 1100-11%B4C; (b) 1100-19%B4C; (c) 1100-21%B4C; (d) Relationship between
different volume fractions of B4C Electrical conductivity.
(d)
(b)
(c)
(a)
(a) (b)
(d) (e) (f)
Fig. 1. Microstructure of B4C with different volume fractions.(a) 1100-11%B4C; (b) 1100-19%B4C; (c) 1100-21%B4C
Fig. 2. CLEMET image analysis the volume fraction of the B4C particles and reaction products.
(c)
Al3BC
TiB2Al3BC
TiB2
Al3BCTiB2
B4CB4C
B4C
10 12 14 16 18 20 2240
42
44
46
48
Ele
ctri
cal c
ondu
ctiv
ity (
IACS
)
B4C composition (%)
0
5
10
15
20
25
Volu
me
frac
tion
(%)Partical types
B4C
Al3BC
TiB2
0
5
10
15
20
25
Volu
me
frac
tion
(%)
Partical types
B4C
Al3BC
TiB2
0
5
10
15
20
25
Vo
lum
e fr
actio
n (%
)
Partical types
B4C
Al3BC
TiB2
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
63L’influence des éléments de transition (V, Zr et Mo) et de la vitesse de
refroidissement sur l'évolution des dispersoïdes dans l'alliage de fonderie Al-Si 356 Influence of transition elements (V, Zr and Mo) and cooling rate
on the evolution of dispersoids in Al-Si 356 foundry alloy
Replacing cast iron with Al-Si foundry alloys in automotive engines can result in significant weight reduction and consequently better fuel efficiency. However, their mechanical properties sharply drop with increasing temper-ature (above 250°C) due to the coarsening of precipitates. Therefore, the way to improve the elevated-temper-ature properties of Al-Si foundry alloys via dispersion strengthening is significant for their development. In the present work, individual/combined additions of transition elements (V, Zr and Mo) were introduced into Al-Si 356 foundry alloy at different cooling rates to discover their influence on the precipitation behavior of dispersoids. Results show that both individual and combined addition of V, Zr, Mo lead to the formation of dispersoids but with different composition, morphology and density after applied heat treatment while combined additions of V, Zr and Mo has the best strengthening effect. Higher cooling rate can refine the dispersoid size but leads to a larger dispersoid free zone.
Le remplacement de la fonte par des alliages de fonderie Al-Si dans les moteurs automobiles peut entraîner une réduction significative du poids, par conséquent, une meilleure efficacité énergétique. Cependant, leurs propriétés mécaniques chutent fortement avec l'augmentation de la température (supérieure à 250 °C) à cause du grossis-sement des précipités. Par conséquent, la manière d'améliorer les propriétés à température élevée des alliages de fonderie Al-Si à travers le durcissement par dispersion est importante pour leur développement. Dans le présent travail, des ajouts individuels/combinés des éléments de transition (V, Zr et Mo) ont été introduites dans l'alliage de fonderie Al-Si 356 avec des vitesses différentes de refroidissement pour découvrir leurs influences sur la précipita-tion des dispersoïdes. Les résultats montrent que les ajouts individuels et combinés de V, Zr, Mo conduisent à la for-mation des dispersoïdes mais avec différentes compositions, morphologies et densités après traitement thermique appliqué, tandis que les ajouts combinés de V, Zr et Mo ont le meilleur effet de durcissement. Une vitesse de refroi-dissement plus élevée peut affiner la taille du dispersoïde mais conduire à une zone sans dispersoïde plus grande.
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Journée des étudiants – REGAL8 et 9 octobre 2019
ResultsTEM dispersoid observation (×10k):
Experimental ProcedureAlloys composition:
Casting molds: Y-shape Thin plate(unit: mm) (90×40×30) (100×80×4)
Heat treatment: 500×2h + 520×4h
Microstructure analysis: Property evaluation:Optical Microscope and TEM Hardness and electrical conductivity
Dong Li, Kun Liu, X. Grant ChenResearch Chair in Metallurgy of Aluminum Transformation,
University of Québec at Chicoutimi, Saguenay, QC, G7H 2B1
Influence of transition elements (V, Zr and Mo) and casting methods on the elevated-temperature
properties of Al-Si 356 foundry alloy
L’influence des éléments de transition (V, Zr et Mo) et de la vitesse de refroidissement sur l'évolution des
dispersoïdes dans l'alliage de fonderie Al-Si 356
Alloy Si Fe Mg Mn Ti Cu Sr V Zr Mo Al356 7.00 - 0.35 0.20 0.20 0.60 0.01 - - - Bal.
356V 7.42 0.154 0.35 0.20 0.19 0.79 0.0138 0.28 - - Bal.356Zr 7.19 0.133 0.35 0.18 0.14 0.75 0.0065 - 0.227 - Bal.356Mo 7.35 0.146 0.35 0.20 0.12 0.72 0.0051 - - 0.31 Bal.
356VZrMo 6.81 0.123 0.36 0.20 0.15 0.49 0.0074 0.48 0.17 0.37 Bal.
ObjectiveTo discover the influence of transition elements (V, Zr and Mo) on the precipitation behavior of dispersoids and related elevated-temperature properties .
IntroductionReplacing cast iron with Al-Si foundry alloys in automotive engines can result in significant weight reduction and consequently better fuel efficiency. However, their mechanical properties sharply dropped with increasing temperature (above 250) due to the coarsening of precipitates. Therefore, how to improve the elevated-temperature properties of Al-Si foundry alloys via dispersion strengthening is significant for their development. In the present work, individual/combined additions of transition elements (V, Zr and Mo) were introduced into Al-Si 356 foundry alloy at different cooling rates to discover their influence on the precipitation behavior of dispersoids.
Conclusions1. Both individual and combined addition of V, Zr, Mo lead to the formation of
dispersoids but with different composition, morphology and number density after applied heat treatment, while combined additions of V, Zr and Mo has the best strengthening effect.
2. Higher cooling rate can effectively increase the dispersoid number density of all the alloys, and refine the dispersoid size of 356Zr and 356Mo alloys, but lead to a larger dispersoid free zone in 356Mo and 356VZrMo alloys.
Future work1. Elevated temperature strength test after 300 annealing for 100h.
2. Using lower first-step heat treatment temperature to produce nanoscale Al3Zr precipitates.
3. Modification of aging and solution heat treatment to get best high temperature performance.
356V 356Zr 356Mo 356VZrMo
Y
S
OM observation (after etching):
Precipitation zone
Precipitation-free zone
Property measurements:
AlloyConfirmed dispersoids
α-Al(Mn,Fe)Si series Al3M/α-Mg/α-Ti series
356Base Al(Mn,Fe)Si -
356V Al(V,Mn,Fe)Si, Al(Mn,Fe)Si Al3(V,Ti)
356Zr Al(Mn,Fe)Si Al3(Zr,Ti), Al(Mn,Mg,Fe)Si
356Mo Al(Mo,Mn,Fe)Si, Al(Mn,Fe)Si -
356VZrMo Al(V,Mo,Mn,Fe)Si and above Al3(Zr,V,Ti), Al(Zr,V,Ti,Mn,Fe)Si, Al(Zr,V,Ti,Mg)Si and above
TEM-EDS dispersoid confirmation:
200111
11ത1
200111
11ത1
200
020
220
200
020220
200
020220
200
020
220
200
111
11ത1
200
111
11ത1
198nm
179nm
207nm
182nm
213nm
249nm
250nm
228nm
Dong LiUniversité du Québec à Chicoutimi
K. Liu - UQACX.-G. Chen - UQAC
6464L’influence de la texture cristalline et de la taille des grains sur l’atténuation des ondes ultrasonores dans des échantillons d’aluminium 6061 F et 6061 HIP+T6 produits par énergie concentrée (DED)The influence of crystalline texture on ultrasonic wave attenuation in 6061-F et 6061 HIP+T6 aluminum samples produced by directed energy deposition
Dans le cadre du contrôle non destructif par ultrasons, l’atténuation désigne une perte d’intensité du signal. Dans un matériau parfaitement homogène, l’intensité d’un signal ultrasonore diminue à cause de l’étalement spatial. Dans un matériau polycristallin, la dispersion de l’énergie acoustique aux joints de grains entraîne une atténuation encore plus importante du signal ultrasonore, ce qui complique les contrôles non destructifs. Or, l’at-ténuation diffère selon la taille et l’orientation des grains. Les pièces produites par fabrication additive possèdent une microstructure souvent cellulaire à la base de l’échantillon, dans la zone proche du substrat, puis dendritique dans les couches supérieures. De plus, le traitement HIP augmente la taille des grains. Cette étude porte donc sur l’atténuation du signal ultrasonore au sein d’échantillons d’aluminium 6061 imprimés par dépôt sous énergie concentrée (directed energy deposition, DED). Des mesures ultrasonores sont réalisées par immersion avec une sonde mono-élément de 5 MHz et 3 1MHz. Des métallographies des échantillons permettent de caractériser la microstructure. Le but de ce travail est de corréler la texture cristalline ainsi que la taille des grains propre aux pièces produites par fabrication additive, à l’atténuation des ondes ultrasonores, afin de déterminer une direction préférentielle d’inspection avec un bruit structural minimal.
In the context of non-destructive ultrasonic testing, a loss of signal intensity is known as wave attenuation. In a perfectly homogeneous material, the signal’s intensity decreases because of spatial spreading. In a polycrys-talline material, the wave attenuation is greater due to the scattering of the signal at grains boundaries. This structural noise is detrimental to ultrasonic testing because the signal is difficult to read and the penetration of the wave in the material is superficial. Yet, wave attenuation is related to the grain size and orientation. The microstructure of additively manufactured parts is often cellular near the substrate and then dendritic in higher layers. Moreover, grains size grows with HIP treatment. The present work focuses on ultrasonic wave attenua-tion in aluminum 6061 samples produced by directed energy deposition (DED). Immersion ultrasonic testing is performed with a 5 MHz and a 3 MHz single-element probe. Microstructure characterization is done through metallography. This study aims to correlate additive manufacturing crystalline texture to the ultrasonic wave attenuation, in order to identify inspection directions with minimal structural noise.
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Anaïs Lorans École Polytechnique
de Montréal
A. Bois-Brochu - CMQ P. Provencher - École
Polytechnique de Montréal M. Brochu - École
Polytechnique de Montréal
Journée des étudiants – REGAL8 et 9 octobre 2019
Anaïs Lorans1, Alexandre Bois-Brochu2, Paul Provencher1 et Myriam Brochu1
1 Département de génie mécanique, Polytechnique Montréal, 2900 Boulevard Edouard Monpetit
2 Centre de Métallurgie du Québec, 3095 rue Westinghouse
L’influence de la microstructure sur l’atténuation des ondes ultrasonores dans
un échantillon d’aluminium AlSi7Mg produitpar dépôt sous énergie concentrée (DED)
Impact of aluminium microstructure on ultrasonic wave propagation in a AlSi7Mg
aluminium DED sample
1. Introduction & Problématique
3. Méthodologie
6. Conclusion
Le contrôle non destructif par ultrasons est basé sur la détection d’un changement d’impédance du milieu qui indique la présence d’un défaut. Or, plus l’atténuation intrinsèque du matériau estforte moins il sera perméable aux ultrasons. L’atténuation intrinsèque provient de la conversion de l’énergie vibratoire en chaleur (absorption), et de la déviation d’une partie de l’ondeincidente par réflexion (dispersion). Le coefficient α caractérise l’atténuation intrinsèque d’un matériau, via la mesure de l’amplitude des échos successifs, soient A1 et A2 sur la figure 1.
Dans le cas des pièces en aluminium AlSi7Mg produites par dépôt sous énergie concentrée (DED), les grains sont colonnaires, et parallèlement orientés à la direction de construction [2].L’influence de la taille des grains sur l’atténuation des ondes ultrasonores longitudinales a été étudiée dans la littérature [5] mais, l’influence de la texture reste peu investiguée. Il est doncintéressant d’évaluer l’atténuation par dispersion sur un échantillon texturé tel que ceux produits par DED.
2. Objectifs• Quantifier l’atténuation ultrasonore via la mesure
du coefficient d’atténuation sur les trois faces.• Analyser le lien entre la microstructure et
l’atténuation des ultrasons.• Établir une ou plusieurs directions d’inspection à
privilégier pour les pièces produites par DED.
1. Balayage en immersion des trois faces d’un cube d’aluminium AlSi7Mg de 15 mm delongueur, produit par DED, avec une sonde de 3,5 et 5 MHz d’après la figure 2. Lecoefficient d’atténuation est calculé à partir de l’amplitude des multiples échos de fondselon la norme ASTM E664 tel que montré à la figure 1.
2. Métallographies des trois faces pour caractériser la morphologie des grains.
3. Simulation du bruit de structure par les grains sur Matlab avec Field II [3] [4].
[1] B. Ellyson, Étude du comportement vibratoire de l’alliage Ti-6Al-4V produit par fabrication additive, Université de Montréal, 2015=6[2] J. H. Martin, B. D. Yahata, J. M. Hundley, J. A. Mayer, T. A. Schaedler & T. M. Pollock, 3D printing of high-strength aluminium alloys, Nature,Vol 549, 2017[3] J.A. Jensen: Field: A Program for Simulating Ultrasound Systems, Paper presented at the 10th Nordic-Baltic Conference on Biomedical Imaging Published in Medical & Biological Engineering & Computing, pp. 351-353, Volume 34, Supplement 1, Part 1, 1996. [4] J.A. Jensen and N. B. Svendsen: Calculation of pressure fiels from arbitrarily shaped, apodized, and excited ultrasound transducers, IEEE Trans. Ultrason., Ferroelec., Freq. Contr., 39, pp. 262-267, 1992.[5] A. V. Pamel, Ultrasonic inspection of highly scattered materials, Imperial College London, 2015
Nos résultats sont les premiers à démontrer l'effet de la microstructure produite par DED surl'atténuation au sein d'un alliage d'aluminium. Il serait donc utile de prendre en considérationla direction de construction des pièces produites par DED lors de contrôles par ultrasons afinde les rendre plus performants.
Ces résultats sont toutefois à relativiser, il serait nécessaire d’effectuer ces mesures à plusgrande échelle et sur des échantillons variés pour corroborer ces premières constatations.
Figure 6: Observation de la microstructurede la face 3 au MO tel que poli à 1µm etattaqué au réactif de Keller
Figure 4: Observation de la microstructurede la face 1 au MO tel que poli à 1µm etattaqué au réactif de Keller
Figure 5: Observation de la microstructurede la face 2 au MO tel que poli à 1µm etattaqué au réactif de Keller
4. Résultats
Figure 3: Coefficient d’atténuation des ondes ultrasonores longitudinales au traversdes 3 faces d’un cube d’aluminium AlSi7Mg sondé à 3,5 et 5 Mhz
Les résultats de la figure 3 montrent que :
• Les mesures du coefficient d’atténuation selon lestrois faces d’inspection sont comparables à 3,5 MHz.
• Le coefficient d’atténuation est presque deux foisplus élevé lorsque l’onde traverse la face 3 quelorsqu’elle traverse les faces 1 et 2 à 5 MHz.
𝐴𝐴1
𝐴𝐴4
Figure 1: Méthode de mesure du coefficient d’atténuationselon la norme ASTM E664
∝ 𝑒𝑒−α𝑥𝑥
Sonde ultrasonore
Face 3
Face 2Face 1
Figure 2: Dispositif expérimental
Cuve d’immersion
Dire
ctio
n de
cons
truc
tion
par D
ED e
t dire
ctio
n d’
élon
gatio
n de
s gra
ins
• Les observations métallographiques présentées sur les figures 4, 5 et 6 dévoilent unestructure colonnaire allongée selon la direction de construction.
5. Analyse & DiscussionLorsque le faisceau ultrasonore frappe la pièce en traversant la face 3 il y a une probabilitéplus importante pour qu’il rencontre un joint de grains perpendiculaire à son axe, quelorsqu’il traverse la face 1 ou 2. Or, la part d’énergie déviée est maximale lorsque le faisceaufrappe un réflecteur perpendiculaire. Ainsi les multiples joints de grains orientés à 90finissent par former une large interface fortement réflectrice (figure 7 et 8), ce qui crée uneplus forte atténuation de l’onde.
4.3 Visualisation du bruit de structures simulées
4.2 Observations métallographiques4.1 Mesures du coefficient d’atténuation
Une anisotropie potentielle d’atténuation estrévélée expérimentalement.
Une anisotropie microstructurale, issue du procédé DED, est mise en évidence.
Les contours blancs sont des joints de grain
Le jaune est associé à une forte atténuation
Les figures 6 et 7 présentent une simulation d’une microstructure respectivement équiaxe puis colonnaire au moyen d’undiagramme de Voronoï ainsi que l’image reconstruite par la méthode de focalisation en tout point (TFM).
La simulation montre que l’anisotropie microstructurale provoque une anisotropied’atténuation des ondes ultrasonores.
Figure 7: Fantôme simulé d’une microstructure équiaxe (à gauche) ettransformée de Hilbert de la reconstruction TFM (à droite) pour unesonde de 64 éléments à 5MHz (simulation des faces 1 et 2)
Dire
ctio
n du
faisc
eau
ultra
sono
re
Figure 8: Fantôme simulé d’une microstructure colonnaire (à gauche) ettransformée de Hilbert de la reconstruction TFM (à droite) pour unesonde de 64 éléments à 5MHz (simulation de la face 3)
Dire
ctio
n du
faisc
eau
ultra
sono
re
Un pore de 5 µm de diamètre est un site d’amorce de fissure [1].
Source de l’image : www.machine-outil.com
• Les points bleus représentent les joints de grains.
• Plus une zone est jaune plus elle est réflectrice.
65Usinage à grande vitesse des alliages d'aluminium AA6061-T6 : influence
du rayon d'arête de l'outil sur les caractéristiques d'usinage High speed machining of aluminum alloys AA6061-T6: influence
of tool edge radius on machining characteristics
The parts quality of the machined metallic components depends highly on cutting conditions, forces, stress, strain and temperature generated during the cutting process. A large cutting temperature leads to quick cut-ting tool degradation and to inaccuracies in components quality. Also, high machining forces raise the stress experienced by the machined part and could result in an excessive tool degradation and breakage. It is there-fore necessary to find means for reducing these negative effects. In the present research, a 2D finite element model for orthogonal machining of 6061-T6 aluminum alloy is developed and experimentally validated to study the interactions between the geography of a turning tool (cutting edge radius and cutting rake angle) and the machining parameters such as cutting speed and feedrate on machining forces and cutting temperature. This research can help to determine the machining conditions providing lower machining forces and temperature. In the future, a 3D finite element model of machining processes will be developed to obtain the optimal values of tool geometry and cutting conditions to improve machining characteristics and machined part surface integrity.
La qualité des pièces métalliques usinées dépend fortement des conditions de coupe, forces, contraintes, dé-formations et de la température générée lors de la coupe. Une température de coupe élevée entraîne une dé-gradation rapide de l'outil de coupe et une perte de précision des pièces. De plus, des forces d'usinage élevées augmentent les contraintes subies par la pièce usinée et peuvent entraîner une dégradation et une rupture excessive de l'outil. Il est donc nécessaire de trouver les moyens de réduire ces effets négatifs. Dans la présente recherche, un modèle 2D d’éléments finis pour l'usinage orthogonal de l'alliage d'aluminium 6061 T6 est déve-loppé et validé expérimentalement pour étudier les interactions entre la géométrie d'un outil de tournage (rayon de l’arête de coupe et angle de coupe) et les paramètres d'usinage tels que la vitesse de coupe et l'avance des forces et températures de coupe. Cette recherche a permis de déterminer les conditions d'usinage qui diminuent les forces d'usinage et la température. Dans le futur, un modèle 3D par éléments finis des processus d'usinage sera développé pour déterminer les valeurs optimales de la géométrie de l'outil et des conditions de coupe afin d'améliorer les caractéristiques d'usinage et l'intégrité de surface des pièces produites.
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Mahshad JavidikiaÉTS
M. Sadeghifar - Université Concordia V. Songmene - ÉTS M. Jahazi - ÉTS
Journée des étudiants – REGAL8 et 9 octobre 2019
Mahshad Javidikia1, Morteza Sadeghifar2, Victor Songmene1, and Mohammad Jahazi11 Department of Mechanical Engineering, École de Technologie Supérieure, Montreal, Canada
2 Department of Mechanical, Industrial and Aerospace Engineering, Concordia University, Montreal, Canada
High Speed Machining of Aluminum Alloys AA6061-T6: Influence of Tool Edge Radius
on Machining CharacteristicsUsinage à grande vitesse des alliages
d'aluminium AA6061-T6: Influence du rayon d'arête de l'outil sur les caractéristiques
d'usinage
cFtFct ( )cF N( )tF N )(mmtc
CCoonncclluussiioonnss:: This research demonstrates that a smaller tool edge radius produces lower cutting forces andtemperatures, which can reduce residual stresses in the machined part, inaccuracies in component dimensionsand quick cutting tool degradation and breakage. In the future, a 3D FE model will be developed to obtain theoptimal values of tool geometry and cutting conditions.
A 2D finite element (FE) model for orthogonal machining of 6061-T6 aluminum alloy is developedand experimentally validated to study the interactions between the geometry of a turning toolincluding cutting edge radius and rake angle and the machining parameters such as cutting speedand feedrate on machining forces and cutting temperature. CCuuttttiinngg CCoonnddiittiioonnss aanndd TTooooll GGeeoommeettrryyTest No.
Cutting speed [m/min]
Feed rate [mm]
Edge radius [mm]
Rake angle [o]
Clearance angle [o]
1 650 0.16 0.005 0 112 950 0.16 0.005 +8 113 950 0.2 0.005 0 11
TThhee ddeevveellooppeedd FFEE mmooddeellTTeesstt NNoo.. 11 TTeesstt NNoo.. 22 TTeesstt NNoo.. 33
RReessuullttss
0.4270.378
0.4410.413
0.351
0.465
00.05
0.10.15
0.20.25
0.30.35
0.40.45
0.5
1 2 3
Chip
thick
ness
[mm
]
Test No. EXP. FE
407.9344
472.1
395.6
334.1
467.86
050
100150200250300350400450500
1 2 3
Cutti
ng fo
rce
[N]
Test No. EXP. FE
TThhee sstteeaaddyy ssttaattee ccoonnddiittiioonn iinn tthhee FFEE mmooddeellss
Tool edge radius [mm] Tool edge radius [mm] Tool edge radius [mm]
Tool edge radius [mm] Tool edge radius [mm] Tool edge radius [mm]
V = Cutting speedf = Feed a = Rake angle
VVaalliiddaattiioonn ooff tthhee FFEE mmooddeellss ((eexxppeerriimmeennttss bbyy MMoonnzzeerr eett aall.. 22001144))
6666Le rôle de la composition eutectique des alliages 5xxx dans la fissuration des bords induite par le sodium au cours du processus de laminage à chaudThe role of the eutectic constituents of 5xxx alloys in the sodium induced edge cracking during the hot rolling process
L’effet néfaste du sodium sur la fissuration des bords des tôles en alliage d’aluminium AA5182 laminées à chaud constitue un obstacle majeur à l’efficacité de la production depuis des décennies. Gérer ce problème peut réduire les coûts pour inciter l’industrie automobile à utiliser ce type d’alliage pour alléger davantage la carrosserie au-tomobile. Dans cette étude, deux échantillons d'alliages contaminés au sodium : l’un de 0,3 et l’autre de 1 ppm, ont été homogénéisés et soumis à un laminage à chaud pour atteindre une réduction de 86 % et 95 %. Ensuite, d’un point de vue microstructural, nous avons étudié l’interaction entre les fissures et les constituants afin d’identifier le mécanisme de fissuration approprié à l’échelle pratique industrielle pour une solution ultérieure. En ce qui concerne nos observations, les composés intermétalliques contenant du fer ne correspondent pas à la fissuration du bord, tandis que les constituants de Mg2Si sont des sites de nucléation vides. Ce phénomène peut être considéré comme le principe fondamental des modèles de mesure de la cavitation pour prédire la fracture ductile, d’où l’initiation de la fissure au bord.
Detrimental influence of sodium on the edge cracking of the hot rolled AA5182 aluminum alloy sheets has been a remarkable hindrance of production efficiency for decades. Dealing with this issue can reduce the costs to attract the automotive industry to use this type for the car body lightening purposes more than before. In this study, two samples of 0.3 and 1ppm sodium contaminated alloys were homogenized and subjected to the hot rolling to reach 86% and 95% reductions. Subsequently, from the microstructural point of view, we studied the interaction of cracks and constituents to identify the suitable cracking mechanism in the industrial practice for the further solution. Regarding our observations, the iron-bearing intermetallic do not correspond to the edge cracking while the Mg2Si constituents are the void nucleation sites. This phenomenon can be assumed as the fundamental of cavitation measurement models to predict the ductile fracture, hence the edge crack initiation.
Mohammadreza MofarrehiUniversité du Québec
à Chicoutimi
X.-G. Chen - UQACM. Javidani - Rio Tinto
Journée des étudiants – REGAL 8 et 9 octobre 2019
Mohammadreza Mofarrehi1, X. Grant Chen1, Mousa Javidani2
1 Research Chair in Metallurgy of Aluminum Transformation, UQAC 2 Arvida Research and Development Centre of Rio Tinto
The role of the constituent phases of 5xxx
alloys in the sodium induced edge cracking during hot rolling
Le rôle des phases constitutives des alliages 5xxx dans la fissuration des bords induite par le sodium
au cours du processus de laminage à chaud
The AA5182 sheets are favorite of automotive industry to produce go-green autos. In this study, we try to help them via preventing edge crack occurrence during the hot rolling process to reduce the final production costs.
Alloy %Mg %Mn %Fe %Si %Cu %Ti Na(ppm) %Al
High Sodium 4.82 0.25 0.14 0.05 <0.01 0.023 1 B Low Sodium 4.88 0.25 0.15 0.06 <0.01 0.023 <0.5 B
Characterization
SEM, EDS
As-cast and Homogenized Microstructure
Dimensions in mm
Homogenization
#11 Hot Rolling #18 Hot Rolling
Phase As-Cast Homogenized
Al3(Mn,Fe)
Al6(Mn,Fe)
β-Al8Mg5
Mg2Si
#11 Rolled
For the both Na contents
#18 Rolled
HS
LS
HS
LS
The Mg2Si constituents evidently contribute in void nucleation.
Edge cracking during the hot rolling is attributed to the ductile failure which is dependent on cavitation volume.
The higher reduction promotes void growth and coalescence. In the higher sodium content, it is more severe due to the higher void nucleation.
From the microstructural point of view, the rolling process can be designed considering the ratio of Na content/Mg2Si volume fraction and the applied reduction to avoid edge cracking.
Inverse segregation of the alloying elements to the surface needs to be concerned which might escalate the volume of Mg2Si and Na concentration on the surface.
The quantitative measurements of the Mg2Si volume and Na amount will be carried out to achieve a model for prediction of a crack occurrence at various reductions.
A hot tensile test will be designed for a lab-scale analysis of a ductile failure caused by Na presence to improve prediction precision.
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
67Évaluation expérimentale et numérique de joints adhésifs en aluminium
à simple recouvrement pour l’industrie automobile Experimental and numerical evaluation of aluminum single lap adhesively
bonded joints for the automotive industry
Single lap adhesive joints are increasingly being used in the manufacturing of automotive structures. The type of the adhesive (brittle or ductile), the overlap length, the thickness of the adhesive and the curing conditions are the main factors which have an influence on the shear strength of bonded joints. The objective of this work is to investigate the influence of these factors on the joint resistance. Single lap joints were prepared according to ASTM D1002 standard and their strength was determined under tensile loading. The adhesives were epoxy and methacrylate and the adherend was aluminum 5052. The overlap length has been changed in the range from 12.5 to 20 mm and two values of adhesive thicknesses were used: 0.25 and 0.6mm. Adhesives were cured under two different conditions; the first one was conventional, as suggested by the manufacturer, and the second one was under a specific heating cycle. In order to determine the state of the stresses and their distribution, the bonded joint was also simulated using the finite element method. The results show that increasing the overlap length, decreasing the adhesive thickness and heating cycles as curing condition improve the joint strength. Epoxy provides more resistance joints than methacrylate.
Les joints adhésifs à un tour sont de plus en plus utilisés dans la fabrication de structures automobiles. Le type d'adhésif (fragile ou ductile), la longueur de recouvrement, l'épaisseur de l'adhésif et les conditions de durcis-sement sont les principaux facteurs qui influent sur la résistance au cisaillement des joints collés. L'objectif de ce travail est d'étudier l'influence de ces facteurs sur la résistance des articulations. Des joints à recouvrement simples ont été préparés selon la norme ASTM D1002 et leur résistance a été déterminée sous traction. Les adhésifs étaient de l'époxy et du méthacrylate et l'adhérant était de l'aluminium 5052. La longueur de chevau-chement a été modifiée dans la plage de 12,5 à 20 mm et deux valeurs d'épaisseurs d'adhésif ont été utilisées: 0,25 et 0,6 mm. Les adhésifs ont été durcis dans deux conditions différentes : le premier était conventionnel, comme suggéré par le fabricant, et le second était sous un cycle de chauffage spécifique. Afin de déterminer l'état des contraintes et leur distribution, le joint collé a également été simulé à l'aide de la méthode des éléments finis. Les résultats montrent que l'augmentation de la longueur de recouvrement, la réduction de l'épaisseur de l'adhésif et les cycles de chauffage à mesure que la condition de durcissement améliore la résistance du joint. L'époxy fournit plus de joints de résistance que le méthacrylate.
Marzieh NodehUniversité de Sherbrooke
A. Maslouhi - Université de Sherbrooke A. Desrochers - Université de Sherbrooke
Journée des étudiants – REGAL8 et 9 octobre 2019
Marzieh Nodeh1, Ahmed Maslouhi2 and Alain Desrochers21 Center Technologies Avancées (CTA), 3000 Boulevard de l'Université, Sherbrooke, QC J1K 0A5 Le
2 Department of Mechanical Engineering, Sherbrooke University, 2500, boul. de l'Université Sherbrooke (Québec), J1K 2R1
Experimental and numerical evaluation of aluminum single lap adhesively bonded joints for
the automotive industry
Évaluation expérimentale et numérique de joints adhésifs en aluminium à simple recouvrement pour
l’industrie automobile
Context & Objective
Methodology
Results
Conclusion
0.00
2.00
4.00
6.00
8.00
10.00
0 0.2 0.4 0.6
Load
(KN)
Displacement (mm)
Epoxy
1
2
30.00
2.00
4.00
6.00
8.00
10.00
0 0.5 1
Load
(KN)
Displacement (mm)
Methacrylate
1
2
3
0.00
2.00
4.00
6.00
8.00
10.00
0 0.5 1
Load
(KN)
Displacement (mm)
Overlap length: 12.5 mm
1
2
3 0.002.004.006.008.00
10.00
0 0.5 1 1.5
Load
(KN)
Displacement (mm)
Overlap length: 15mm
1
2
30.002.004.006.008.00
10.00
0 0.5 1 1.5
Load
(KN)
Displacement (mm)
Overlap length: 17.5mm
1
2
3
0.002.004.006.008.00
10.0012.00
0 1 2
Load
(KN)
Displacement (mm)
overlap length: 20mm
1
2
3
4
6
8
10
12
10 15 20
Aver
age
ultim
ate
load
(KN)
Displacement (mm)
Joint strength increases by increasing the overlap length with a linear trend
02468
1012141618
-10 -5 0 5 10
Shea
r stre
ss (M
pa)
overlap length
Distribution of Shear stress
Numerical Model
Analytical model
-15
-10
-5
0
5
10
15
20
25
-10 -5 0 5 10peel
stre
ss (M
pa)
overlap length
Distribution of peel stress
Numerical Model
Analytical model
Maximum peel and shear stresses along the centerline occur at a very small distance from the free edge.
0
2
4
6
0 0.2 0.4 0.6
Load
(KN)
Displacement (mm)
RT
1
2
3
0.00
2.00
4.00
6.00
8.00
10.00
0 0.5 1
Load
(KN)
Displacement (mm)
Heating cycle
1
2
3
Fig 1. Example of a recreational product-Spyder
• Investigating the influence of adhesive type, overlap length and curing temperature on the joint strength.• Investigating the distribution of shear and peel stress along the overlap.
Fig. 8. Influence of overlap length on the joint strength
Distribution of stresses
Overlap lengthAdhesive type
Fig.6. Influence of adhesive type on the failure load and displacement
Fig. 7. Fracture surface of a) epoxy & b) Methacrylate
Curing temperature
Fig.9. Influence of curing temperature on the joint resistance
Fig. distribution of a) shear stress and b) peel stress Fig. a) total deformation and b) equivalent stress
a
ab b
Experimental Procedure• The surface of the samples were cleaned with alcohol and Kim wipe.• The substrates were put together in the assembly jig and tightened in the place.• The single lap joints (Fig.2) were tested in an INSTRON test machine with a load cell of 100 KN at a rate of
1mm/min at room temperature and the displacement were measured with an extensometer (Fig. 3).
Numerical study• The 3D FE analysis was performed in ANSYS.• Table 2 presents the mechanical properties of adhesive and adherend.• The adhesive layer was divided into six elements along the thickness (Fig. 4).• The boundary conditions considered in this study are presented in Fig. 5.• The shear and peel stress distribution along the middle layer of adhesive will
investigated and validated with the model of Goland and Reissner.
Table 1. investigated parameters
Fig 2. The geometry of Single lap joint (dimension in mm)
Fig 3. Static test on an Instron test machine
Context:• The present study is part of the ATLAS project that aims to design an optimized aluminum alloy recreational chassis to reduce weight, the number of parts and cost
while providing good performance and reliability.• Adhesive bonding is considered as key technologies for assembling aluminum body structures in automotive industry.• Compared to classical fastening techniques such as welding and riveting, adhesive bonding leads to lighter body structure, less stress concentration, more aesthetic
joints, better corrosion resistance, cost reduction and lower noise and vibration harshness levels.• For an adhesively bonded joint design that meets the industry requirements, the effect of different parameters on the joint performance should be studied.
Objective:
Table 2. Material PropertiesFig 4. FE mesh for the single lap joint
Fig 5. Boundary condition
a
b
• The epoxy makes more strength joints with less strain at failure. However, it doesn’t make a good cohesion with the surface of Al.• The joint strength increased by increasing the overlap length linearly.• Curing during heating cycles resulted in the more resistant joints and the larger displacement at failure. These samples exhibited a considerable plastic deformation at the end of the joint.• The overlap length and curing temperature didn’t have any influence on the failure mode of joints bonded with methacrylate. • The finite element prediction for the distribution of shear and peel stresses agree well with the analytical model.
Samples bonded with methacrylate shows cohesive failure that means thefailure occurs within the adhesive layer. However, the fracture of epoxy occurs atthe interface of adhesive and adherend that is called adhesive failure.
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
6868Fusion sur lit de poudre de l'alliage Al-Cu-TiB2Powder bed fusion of Al-Cu-TiB2 alloy
Les alliages d'aluminium de la série 2000 présentent un grand intérêt dans l'industrie aérospatiale en raison de leur capacité à améliorer les propriétés mécaniques grâce au traitement thermique. L’un des inconvénients de la série 2000 est sa propension aux défauts de solidification ; des facteurs tels que le déchirement à chaud et les complications de porosité ont réduit leur capacité à être utilisés pour des pièces à géométrie complexe. Récemment, les alliages d'aluminium renforcés avec du diborate de titane sont devenus courants dans la littéra-ture. La phase TiB2 agit en tant que raffineur de grain, tout en améliorant simultanément la dureté et la stabilité thermique. Ici, nous avons étudié la possibilité d’utiliser la fabrication additive par fusion sur lit de poudre pour de tels alliages. Nous avons caractérisé les propriétés et le débit de la poudre, comparé les poudres atomisées au gaz avec les microstructures imprimées et lié les effets de différents paramètres du laser sur le lit de poudre au développement de la microstructure.
2000 series aluminum alloys are of great interest in the aerospace industry due to their capacity to enhance mechanical properties with heat treatment. One drawback to the 2000 aluminum series is associated with their propensity for solidification defects; factors such as hot tearing and porosity complications has reduced their ability to be used for complex geometry parts. Recently, aluminum alloys reinforced with titanium diborate have become prevalent in the literature. The TiB2 phase acts as a grain refiner, while simultaneously enhancing the hardness and thermal stability. Here we have investigated the possibility of utilizing powder bed fusion additive manufacturing for such alloys. We characterized the powder properties and flow, compared the gas atomized powders to the as-printed microstructures and linked the effects of different laser powder bed parameters with microstructure development.
Amy Nommeots-Nomm McGill University
S. K. Tumulu - McGillM. Brochu - McGill
Journée des étudiants – REGAL8 et 9 octobre 2019
A. Nommeots-Nomm1, S. K. Tumulu1, M. Brochu1
1 REGAL Aluminium Research Centre, Department of Mining and Materials Engineering, McGill University, Montreal, Quebec, Canada H3A 0C5
Powder Bed Fusion of Al-Cu-TiB2 alloy
Fusion sur lit de poudre de l'alliage Al-Cu-TiB2 alloy
2000 series aluminium alloys are of great interest in the aerospace industry due to their capacity to enhancemechanical properties with heat treatment. One drawback to the 2000 aluminium series is associated withtheir propensity for solidification defects; factors such as hot tearing and porosity complications has reducedtheir ability to be used for complex geometry parts. Recently, 2000 series aluminium alloys reinforced with TiB2have become prevalent in the literature. The TiB2 acts as a grain refiner, while simultaneously enhancing thehardness and thermal stability. Here we have investigated the possibility of utilizing powder bed fusionadditive manufacturing for such alloys.Methodology
A20X of composition; 4.6 Cu – 3.4 Ti – 1.4B – 0.75 Ag – 0.27 Mg was investigated.Particles were generally spherical withsome satellites present. Consisting of Al-Cu rich cells of size 1.6 ± 0.67 µm, withTiB2 rich rod formations of size 1.2 ± 0.49µm by 0.4 ± 0.17 µm. Once printed themicrostructure becomes refined with acell size of 0.96 ± 0.09 µm in the refinedzone and 1.47 ± 0.13 µm in the coarse.The TiB2 rods have not changed in size 1.2± 0.74 µm by 0.4 ± 0.26 µm. Suggestingthat they were not melted during theprinting process
Results
Introduction
Conclusions
Al B Cu Mg Ti Ag
5 µm5 µm5 µm5 µm 5 µm 5 µm
Powder Microstructure
Preliminary analysis of
printed samples
Printed Microstructure
5 µm5 µm5 µm 5 µm 5 µm 5 µm
AgMg TiCuAl B
A20X of composition; 4.6 Cu – 3.4 Ti –1.4 B – 0.75 Ag – 0.27 Mg wassuccessfully printed without crackingdefects. Microstructural analysissuggests no change in the TiB2 phase,but a refining of the microstructureduring LBPF processing.
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
P2 [AM]2
69Caractérisation du comportement en extrusion d’un alliage d’aluminium
expérimental de la série 6xxx destiné aux applications haute-performance de l’industrie automobile
Characterization of the extrusion behavior of an experimental 6xxx series aluminum alloy intended for high performance automotive applications
The growing demand for fuel efficient vehicles made aluminum a top candidate for automotive industry. In order to fulfill current need for mass reduction, Al is introduced into design of critical components, which require high strength and toughness. Currently, some of those components are produced from 6xxx (Al-Mg-Si) and 7xxx (Al-Zn) series Al extrusions. However, poor recyclability of Al-Zn alloys and high cost make them uninteresting to use. In this work, two variants of an experimental Al-Mg-Si alloy either with or without Zr addition were studied as an alternative to 7xxx alloys. Plant trials were carried out to evaluate their extrusion behavior. Extruded profiles were characterized in terms of mechanical properties and microstructure. Yield strength up to 380 MPa was obtained upon heat treatment. Results are used to optimize aging parameters in order to surpass strength of commercially available 7003, 6082 and 6008 Al alloys.
La demande croissante pour les véhicules à haute efficacité énergétique a fait de l’aluminium un candidat de premier choix pour l’industrie automobile. Dans le but de rencontrer les objectifs actuels de réduction de masse, l’Al est introduit dans le design de composants critiques requérant une résistance et ténacité élevées. Actuel-lement, certains de ces composants sont produits à partir d’extrusions des alliages des séries 6xxx (Al-Mg-Si) et 7xxx (Al-Zn). Cependant, l’utilisation d’alliages Al-Zn difficilement recyclables est coûteuse et s’oppose à l’optimisation du cycle de vie. Dans cette étude, deux variantes d’un alliage expérimental Al-Mg-Si avec et sans addition de Zr ont été étudiées comme alternative aux alliages de la série 7xxx. Des essais d’extrusion en usine ont été effectués afin d’en étudier le comportement. Les profilés produits ont été caractérisés en termes de propriétés mécaniques et microstructurales. Une limite élastique atteignant 380 MPa a été obtenue suite au traitement thermique. Les résultats de ces essais sont utilisés pour optimiser les paramètres de vieillissement dans le but de surpasser la résistance des alliages d’Al commerciaux 7003, 6082 et 6008.
Justin PlanteUniversité Laval
N. Giguère - CMQ C. Blais - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Alliage B
b)
Alliage Z
c)
Justin Plante1, Nicolas Giguère2 et Carl Blais1
1 Département de Génie des mines, des matériaux et de la métallurgie, Université Laval, QC2 Centre de métallurgie du Québec, QC
Caractérisation du comportement en extrusion d’un alliage d’aluminium expérimental de la série 6xxx destiné aux applications haute-performance de
l’industrie automobileCharacterization of the extrusion behaviour of an
experimental 6xxx series aluminium alloy intended for high performance automotive applications
IntroductionAl dans l’industrie automobile• Réduction de masse• Consommation de carburant
réduite• Autonomie améliorée
Fig. 1 : Matériaux typiques pour la structure d’une automobile moderne [1]
Feuille AlExtrusion AlCoulage AlHSS
Fig. 2: Répartition des alliages d’Al de corroyage d’une automobile [2]
Problématique : Alliages 7xxx• Coûteux $$• Difficiles à recycler
6060
6016
3003
51825754
6082
7072
But du projet• Développer une nouvelle géneration d’alliages haute
performance de la série 6xxx destinée à l’industrie automobile
Objectifs• Surpasser la performance des alliages 7003, 6082 et 6008• Limite élastique : 400 MPa ↑ Allongement : 10% ↑
MéthodologieMatériaux• Alliage B : Al-Mg-Si +…• Alliage Z : Al-Mg-Si+…+Zr
Extrusion• R = 35
Vieillissementnaturel• 0 – 96 hr
Vieillissement artificiel• 170 – 190°C• 10 – 12 hr
Propriétés mécaniques• LE + RM + All.
Résultats
Travaux futursTempérature (°C)
542517
492
467
443
Y
XZ
• Effet de la position• Courbes de vieillissement• Extrusion en laboratoire• Comportement en fatigue• Comportement en corrosion
RemerciementsCe travail a été rendu possible grâce au financement du Conseil de Recherche en Sciences Naturelles et Génie du Canada et à la participation de Alcoa et du Centre de métallurgie du Québec.
Références[1] « Audi puts steel back in the new A8 », Green Car Congress. [Online].[2] J. Cui et H. J. Roven, « Recycling of automotive aluminum », Transactions of Nonferrous Metals Society of China, vol. 20, no 11, p. 2057-2063, nov. 2010.[3] C. Bruni, A. Forcellese, et F. Gabrielli, « Flow Modelling of AA 6082 Aluminium Alloy », in AMST’02 Advanced Manufacturing Systems and Technology, 2002, p. 367-374.[4] T. Sheppard, Extrusion of Aluminium Alloys. Springer Science & Business Media, 2013.[5] H. Zhu, M. J. Couper, et A. K. Dahle, « Effect of process variables on Mg-Si particles and extrudability of 6xxx series aluminum extrusions », JOM, vol. 63, no 11, p. 66-71, nov. 2011.[6] H. J. McQueen, Hot deformation and processing of aluminum alloys. Boca Raton: CRC Press, 2011.
ND
RD
RD
200 µm
001
111
101
001
111
101
ND
500 µm
Fig. 4 : Courbes contrainte vraie - déformation vraie obtenues d’essais de compression à chaud
Fig. 3 : Macrographie d’une soudure longitudinale (a) et macrographie des alliages B (b) et Z (c) dans la direction longitudinale avec micrographies et cartographies EBSD correspondantes
Fig. 5 : Effet du temps de vieillissement naturel sur les propriétés mécaniques (vieilli artificiellement à 180°C – 10 hr)
Fig. 6 : Effet du vieillissement artificiel sur les propriétés mécaniques (vieilli naturellement pendant 24 hr)
Soudure longitudinale
a)
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Prix Award
7070Comparaison de l'écoulement de la poudre AlSi10Mg avec l’entonnoir Hall/Carney et le GranuFlowComparison of AlSi10Mg powder flowability using Hall/Carney funnel and GranuFlow
La fusion sur lit de poudre métallique est un processus de fabrication additive qui nécessite que chaque couche de poudre soit uniformément répartie sur une surface avant qu'une source de chaleur au laser ne fusionne les particules. Ce processus est répété jusqu’à ce qu’un design préprogrammé soit terminé. La qualité du produit final est affectée non seulement par les paramètres de l’imprimante, mais également par la capacité d'étalement de la poudre sur le lit. Ceci se caractérise par la compréhension de l’écoulement de la poudre grâce à des stan-dards établis tels que l’entonnoir de Hall/Carney. Cependant, une erreur humaine est associée à ce type de mé-thode en raison de la nature du processus. Le GranuFlow, un débitmètre introduit récemment, peut réduire cette erreur en raison de son mécanisme semi-automatisé. Dans ce travail, le comportement de la poudre AlSi10Mg en termes d'écoulement a été comparé à l'aide de l'entonnoir Hall/Carney et le GranuFlow.
Laser powder bed fusion is an additive manufacturing method that requires each layer of powder to be evenly distributed on a surface before a laser heat source fuses together the particles. This process is repeated until a preprogrammed design is completed. The final product quality is affected, not only by the machine parameters, but also by the powder spreadability across the bed. This is characterized by understanding the powder flowa-bility through established standards such as the Hall/Carney funnel. However, there is associated human error in this type of method due to the nature of the process. The GranuFlow, a flowmeter more recently introduced, may reduce this error because of its semi-automated mechanism. In this work, the flow behavior of AlSi10Mg powder was compared using the Hall/Carney funnel and GranuFlow.
Emma Sue ReichertMcGill University
E. R. L. Espiritu - McGillM. Brochu - McGill
Journée des étudiants – REGAL8 et 9 octobre 2019
Emma Sue REICHERT1, Eileen Ross L. ESPIRITU1, Mathieu BROCHU1
1Department of Mining and Materials Engineering, McGill University, 3610 Rue University, Montréal, QC H3A 0C5
Comparison of AlSi10Mg powder flowabilityusing Hall/Carney funnel and GranuFlow
Comparaison de l'écoulement de la poudre AlSi10Mg avec l’entonnoir Hall/Carney et le
GranuFlow
Introduction Results
Problematic and Objective
Conclusions & Future Work
References
§ In additive manufacturing, Laser Powder Bed Fusion depends on powder flowability
§ Powder must be evenly spread on bed to ensure optimal build quality [1]
§ The Hall/Carney flowmeter is traditionally used to characterize powder flowability [2]
Methodology
§ Hall/Carney flowmeter may not be as accurate due to human error
§ GranuFlow is semi-automatic and can reduce human error [3]
§ Experiments to determine accuracy of GranuFlowcompared to that of the Hall/Carney flowmeter
1. Hall/Carney§ 50g AlSi10Mg
powder§ Measure time
required to flow through funnel with stopwatch
§ Repeat 3 times for Hall (2.5 mm) and Carney (5 mm)
2. GranuFlow§ 50g AlSi10Mg
powder§ Measure flow rate
at orifice diameters 2.5mm and 5mm
§ Perform test once
Orifice diameter (mm) Flowmeter Mass flow
rate (g/s) Time (s)
2.5 Hall 1.01 50.29GanuFlow 0.86 57.94
5.0 Carney 5.53 9.07GranuFlow 6.13 8.16
0
1
2
3
4
5
6
7
2.5 5.0
FLOWRATE(G/S)
ORIFICEDIAMETER(MM)
FlowrateofHall/Carneyvs.GranuFlow
Hall/Carney GranuFlow
0
10
20
30
40
50
60
70
2.5 5.0
TIME(S)
ORIFICEDIAMETER(MM)
FlowtimeofHall/Carneyvs.GranuFlow
Hall/Carney GranuFlow
§ No immediate trends in human error can be seen from using the Hall/Carney vs the GranuFlow
§ Humidity causes powder to agglomerate and reduce flow, therefore affecting results
§ Can repeat experiment with multiple powders to seek further trends
§ Mass flow rate and time for Hall/Carney and Granuflow, respectively, were calculated using:
𝐹𝐹 = 𝑔𝑔/𝑠𝑠
[1] S. Sun, M. Brandt, and M. Easton, "2 - Powder bed fusion processes: An overview," in Laser Additive Manufacturing, M. Brandt, Ed., ed: Woodhead Publishing, 2017, pp. 55-77.
[2] M. Leary, "4 - Surface roughness optimisation for selective laser melting (SLM): Accommodating relevant and irrelevant surfaces," in Laser Additive Manufacturing, M. Brandt, Ed., ed: Woodhead Publishing, 2017, pp. 99-118.
[3] GranuTools. (2018). GranuFlow.
Hall/Carney flowmeter (above)GranuFlow (below)
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Prix de la relève Next generation
Award
71Étude des propriétés à haute température des alliages 6111 coulés par
refroidissement direct en ce qui concerne les phénomènes de la fissuration à chaud Study of high temperature properties of DC cast 6111 alloys regarding
hot tearing phenomena
High temperature tensile testing is one of the most effective methods in determining the onset of hot tearing in terms of temperature, load and solid fraction range. Among 6xxx alloys, 6111 alloys are the most prone ones to hot tearing. This work aims at developing a new criterion for evaluation of hot tearing based on stress-strain relationship obtained by Gleeble testing near solidus temperature between 540 and 600 oC. The new setup of Gleeble testing includes following up the evolution of fracture during test using high resolution camera and then correlating the obtained results by GOM software. In this study, the influence of chemical composition was studied using two DC cast ingots with the variation of Si, Cu and Mn levels. Microstructure investigations show the difference in intermetallic volume fractions and morphologies along the ingot thickness or between the two ingots. High volume fraction of β-intermetallic concentrate near the edge of the cracked region. Moreover, some differences in the fracture surfaces, such as smooth regions and intergranular failure, were observed for hot tearing appearance.
Le test de traction à haute température est l’une des méthodes les plus efficaces pour déterminer le début de la fissuration à chaud en termes des intervalles de température, de charge et de fraction solide. Les alliages 6111 sont les plus exposés à la fissuration à chaud. Ce travail vise à développer un nouveau critère d'évaluation de la fissuration à chaud, basé sur la relation contrainte-déformation obtenue par tests de Gleeble à une température proche du solidus entre 540 et 600 °C. La nouvelle configuration des tests Gleeble inclut le suivi de l’évolution de la fracture par une caméra haute résolution, puis la corrélation des résultats avec le logiciel GOM. L’influence de la composition chimique est étudiée sur deux lingots coulés par refroidissement direct avec différentes teneurs en Si, Cu et Mn. Les analyses microstructurales montrent une différence entre les fractions volumiques intermé-talliques et les morphologies le long de l'épaisseur du lingot ou entre les deux lingots. Une fraction volumique élevée de l’intermétallique β se concentre près du bord de la fissure. De plus, certaines différences entre les surfaces de fracture, comme les régions planes et les défaillances intergranulaires, ont été observées.
Mohamed QassemUniversité du Québec à Chicoutimi
J. Rakhomonov - UQAC D. Larouche - UL M. Javidani - Rio Tinto X.-G. Chen - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
Mohamed Qassem1, Jovid Rakhomonov1, Daniel Larouche2, Mousa Javidani3, X. Grant Chen1
1Research Chair in Metallurgy of Aluminum Transformation, University of Québec at Chicoutimi2 Department of Mining, Metallurgy and Materials Engineering, Laval University
3 Rio Tinto, Arvida Research and Development Center
Study of High Temperature Properties of DC Cast 6111 Alloys Regarding Hot Tearing
PhenomenaEtude des propriétés à haute température des alliages
6111 coulés par refroidissement direct en ce qui concerne les phénomènes de la fissuration à chaud
Introduction
Methodology
Results
Conclusion
Future Work
High temperature tensile testing is one of the most effective methods in determining theonset of hot tearing in terms of temperature, load and solid fraction range. Production of6111 ingots free from hot tearing is a challenge needed to be faced. Several studies shed lighton the role of eutectic phases in healing hot cracking during solidification. Relatingappearance of hot tearing to stress-strain relationship in semisolid region requires precisemeasurements as stress and strain values are so small to be distinguished.
Metallographic Examination
Sample Preparation
Temperature Gradient Measurements
Fig.1: Microstructure of DC cast 6111 alloy at different regions (a) surface, (b) mid-center and (c) Center
1. Fe-intermetallic appear in the form of α-intermetallicnear the surface whereas β-phase is dominant in center
2. Mn-rich α-intermetallic appears as constituent ofeutectic near the surface of 6111
3. Cu-rich phases concentrated near the surface of theblock and appears slightly at center as spheroids (Q-phase) and tiny blocky phase (θ-phase)
4. Microstructure changes from globular at surface toequiaxed dendrites at center of billet
5. Fracture surface reveals inter-granular fracture andsmooth areas which are indications for hot tearing.
6. Sample design with total length equal 120 mm is themost proper for semisolid testing based on temperaturegradient measurements
1. Operating Gleeble test at selected temperatures near soliduswith continuous follow up of evolution of macrostructureusing high resolution camera
2. Use GOM correlate software in measuring the strain of thesample in semisolid region and get stress-strain curve bycombining this data with those extracted from Gleeble
3. Changing strain rate to notice the difference in stress-straincurve behavior
4. Studying the behavior of alloy at necking by the aid oflocalized strains detected by GOM correlate software
Fig.3: Volume fraction of intermetallic
Fig.4: Fracture surface of cracked ingot representing smooth area in (a) and inter-granular fracture in (b) which indicates that hot tearing may be the
reason
Fig.5: Temperature gradient curves for different sample designs at different temperatures
200,300,400,500 and 600oC
Objectives1. Choosing the most proper sample design for semisolid
tensile testing2. Identifying existed phases in DC cast 6111 alloy and
variation of microstructure with distance from surface3. Ensuring that the reason for cracking is hot tearing by
fractography examination
Type of intermetallic SURFACE MID-CENTER CENTER
Mg2Si 1.86±0.78 0.56±0.22 0.61±0.61
Fe-intermetallic 6.09±1.82 1.05±0.37 1.35±0.44
Al2Cu 0.51±0.44 0.05±0.08 0.1±0.09
Q-Phase 1.05±1 0.08±0.12 0.08±0.11
a b c
a b
Fig.2: Microstructure of DC cast 6111 alloy at different regions (a) surface, (b) mid-center and (c) Center
a b c
α-intermetallic Mg2Si θ-phase Q-phase
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
7272L’effet de l’amélioration des caractéristiques des dispersoïdes sur la microstructure, la texture et les propriétés mécaniques à température ambiante / élevée des alliages extrudés 6082Effect of dispersoids with enhanced characteristics on microstructure, texture and room-/high-temperature mechanical properties of 6082 extrusions
Dans cette étude, l'effet des dispersoïdes fins et denses de Al(FeMn)Si sur l'évolution de la microstructure, de la texture et des propriétés mécaniques à température ambiante et élevée de l'alliage extrudé 6082 a été étudié. Les caractéristiques améliorées des dispersoïdes ont été obtenues en appliquant le traitement thermique de précipitation (400 oC / 5 h) à des billettes en alliage 6082 coulées par la technique de refroidissement direct, qui ont ensuite été extrudées pour produire des tiges. Des études de microstructure des alliages tels qu'extrudés ont révélé que les dispersoïdes fins et denses provoquaient la dissolution d'un plus grand nombre de particules de Mg
2Si dans la matrice α-Al, comparé à la pratique traditionnelle d'homogénéisation. De plus, les dispersoïdes
avec caractéristiques améliorées réduisaient considérablement l'épaisseur des grains allongés présentant prin-cipalement une texture fibreuse de type <111>. Les tiges d’extrusion ont également été traitées thermiquement par T5 : lors d’une exposition thermique à 300 °C, l’alliage avec des dispersoïdes fins et denses présentait une limite d'élasticité presque deux fois supérieure, que ce soit à la température ambiante ou à haute température, comparé à l'alliage contenant des dispersoïdes traditionnels.
In this study, the effect of fine and dense Al(FeMn)Si dispersoids on the evolution of the microstructure, texture and room-/high-temperature mechanical properties of 6082 extrusion alloy was investigated. Enhanced dispers-oids characteristics were achieved by applying the precipitation heat treatment (400 oC/5h) to direct-chill cast 6082 alloy billets, which were then extruded to produce rods. Microstructural investigations of as-extruded alloys revealed that fine and dense dispersoids caused the dissolution of more Mg
2Si constituent particles into the α-Al
matrix, when compared with the traditional homogenization practice. Moreover, the dispersoids with enhanced characteristics substantially reduced the thickness of elongated grains exhibiting predominantly a <111> type fiber texture. The extrusion rods were also heat treated under T5, and during thermal exposure at 300 oC, the alloy with fine and dense dispersoids exhibited almost twice higher yield strength at both room and high temper-atures, when compared to the alloy containing traditional dispersoids.
Jovic Rakhmonov Université du Québec
à Chicoutimi
K. Liu - UQAC X.-G. Chen - UQAC
P. Rometsch - Rio TintoN. Parson - Rio Tinto
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
73Impact du brossage d’oxydes de surface sur les propriétés mécaniques de
joints à recouvrement en aluminium 2024 soudés par friction-malaxageThe impact of oxide brushing on the mechanical properties
of aluminum 2024 friction-stir-welded lap joints
The mechanical properties of friction-stir-welded joints are affected by the state of the mating surfaces. The effects of oxides have not been extensively covered in the literature for friction-stir-welded lap joints. This pro-ject focuses on the preparation of the faying surfaces of aluminum overlap joints. The impact of this step in the welding process on the mechanical resistance and the fatigue life of the joints is assessed. AA2024-T3 AlClad plates were friction-stir-welded with two types of surface preparation: as-clad (unbrushed) and manually brushed. Tensile and constant-amplitude fatigue tests were conducted. Mechanical strength is similar for both surface preparation (brushed and as-clad), with failure of all tested specimens occurring through the welded zone, in shear. The failure path coincides with the cladding fragments in the welded zone and pure aluminum was detected on the fracture surfaces. This suggests that the cladding material fragments offered a preferred crack path for failure in shear. Fatigue specimens failed on the retreating side of the weld and fatigue lives for both preparations are similar. These results suggest that manual brushing to remove the oxide scale does not affect the mechanical properties of the joint, but that extensive brushing to remove partially the cladding material could.
Les propriétés mécaniques des joints soudés par friction-malaxage sont affectées par l’état des surfaces avant le soudage. L’effet des oxydes n’a pas été couvert de manière exhaustive dans la littérature. Ce projet étudie l’impact de la préparation de surface sur les propriétés mécaniques de joints à recouvrement lors de sollicita-tions monotone et cyclique. Des plaques d’AA2024-T3 AlClad ont été soudées et deux types de préparation de surface ont été effectués : non brossés et brossés manuellement. Des essais de traction et de fatigue à ampli-tude constante ont été faits. La résistance mécanique est semblable pour tous les échantillons. La défaillance se produit en cisaillement dans la zone soudée. Le chemin de fissure coïncide avec des fragments de AlClad dispersés dans la zone de soudure. Ces fragments offriraient donc un chemin préférentiel lors de la défaillance en cisaillement des échantillons de traction. Les échantillons de fatigue ont tous subi une défaillance en mode I, la fissure se propageant dans la plaque du dessus. La vie en fatigue est similaire pour les échantillons non brossés et brossés. Ces résultats suggèrent que le brossage manuel pour retirer les oxydes n’affecte pas les propriétés mécaniques du joint, mais qu’un brossage permettant le retrait partiel du AlClad pourrait avoir un effet.
Journée des étudiants – REGAL8 et 9 octobre 2019
Bénédicte Robitaille1, Lyne St-Georges2, Paul R. Provencher1 et Myriam Brochu1
1 Polytechnique Montréal2 Université du Québec à Chicoutimi
Impact du brossage d’oxydes de surface sur les propriétés mécaniques de joints à recouvrement en aluminium 2024 soudés
par friction-malaxageThe impact of oxide brushing on the mechanical properties of aluminum 2024 friction-stir-welded
lap joints
Le soudage par friction-malaxage (FSW) est un procédé à l’état solide de plus en plus populaire dans l’industrie du transport [1]. La diminution dupoids des assemblages soudés par FSW ainsi que leurs bonnes propriétés mécaniques font de cette technique une candidate au remplacement desboulons et des rivets.Les normes sur le FSW offrent certaines indications sur la préparation de surface des pièces avant le soudage. Elles ne contiennent toutefois pasd’indications spécifiques sur le brossage des oxydes de surface. Ce brossage est par contre essentiel dans les procédés de soudage conventionnel.Certains auteurs ont aussi établi que les oxydes de surface dans des joints FSW pouvaient affecter la résistance mécanique [2].Objectifs1. Évaluer l’impact du brossage sur la résistance mécanique et la vie en fatigue de joints à recouvrement en aluminium 2024-T3 soudés par FSW2. Émettre des recommandations quant à la préparation de surface pour les joints à recouvrement soudés par FSW
Production des joints soudés• Matériau : 2024-T3 AlClad• Soudure : 61 mm/min et 600 RPM, paramètres optimisés pour minimiser les défauts• Préparation de surface des plaques, deux types :
1. Brossées manuellement avec une brosse en acier inoxydable2. Simple nettoyage à l’acétone
Essais de traction et de fatigue• Échantillons découpés dans l’assemblage soudé (Fig. 1)• Essais sur une machine servo-hydraulique• Traction : déplacement du vérin de 1,3 mm/min• Fatigue : R = 0,1, f = 20 Hz, amplitude de contrainte (σa) de 19,5; 12,0 et 10,3 MPa
Macrostructure du joint soudé• Pas de différence significative entre les géométries des discontinuités des
remontées d’interface non soudée• Aucune trace d’oxydes détectée par le spectromètre à dispersion d’énergie
dans la zone soudéeEssais de traction• Défaillance en cisaillement à travers la zone soudée pour tous les échantillons
(Fig. 2)• La résistance mécanique moyenne en cisaillement des échantillons brossés
est de 102,9 MPa et celle des échantillons non brossés de 113,7 MPa, cettedifférence n’est pas considérée significative
• Défaillance à travers les fragments de AlClad et présence d’aluminium pur surles faciès de rupture
Essais de fatigue• Défaillance à travers la plaque du haut pour tous les échantillons (Fig. 3)• Pas de différence significative pour la vie en fatigue des échantillons brossés
et non brossés à toutes les amplitudes de contrainte
La largeur de la zone soudée serait théoriquement suffisante pour prévenir unerupture en cisaillement. Toutefois, des fragments de AlClad sont présents dansla zone soudée et offriraient un chemin préférentiel de fissure lors duchargement uniaxial en tension de l’échantillon (Fig. 5).
Un gradient de ductilité est observable sur les faciès de rupture en traction (Fig.6), la section du haut du faciès (RS) est plus ductile que celle du bas (AS). Cegradient peut s’expliquer par la présence de AlClad dans la zone soudée.Le brossage n’a pas d’effet sur l’épaisseur de la couche de revêtement, donc larépartition et la taille des fragments n’est pas affectée.
Le brossage des oxydes n’affecte pas la vie en fatigue des échantillons soudéspar FSW. Lors de la sollicitation cyclique, les fragments de AlClad ne sont pasdans le chemin de la fissure principale. L’épaisseur résiduelle effective, quiinfluence la rupture en mode I, n’est pas non plus affectée par le brossage.
• Le brossage des surfaces en contact avant le FSW n’affecte pas de manièresignificative la résistance mécanique des joints ou leur vie en fatigue
• Les fragments de revêtement AlClad dans la zone soudée ont un impact négatifsur la résistance mécanique des joints lors d’une sollicitation monotone
Des travaux futurs pourraient être orientés vers un traitement permettant deréduire significativement la couche de AlClad. Cela pourrait augmenter larésistance mécanique des échantillons.
1. Introduction
2. Méthodologie60 mm
147 mm
234 mm
33 mm
3,175 mmCale
Cale
x
z
y
Direction de chargement
3. Résultats 4. Discussion
5. Conclusion
x
z
yx
z
y
Fig. 2 : Rupture en traction d’un échantillon
Fig. 4 : Nombre de cycles à la rupture pour des échantillons soudés par FSW testés à différentes amplitudes de contrainte
RS
AS
z
x
y
Fig. 5 : Revêtement AlClad (blanc) dans la zone soudée – rupture en cisaillement
[1] P. L. Threadgill, A. J. Leonard, H. R. Shercliff, et P. J. Withers, “Friction stir welding of aluminium alloys,” Int. Mater. Rev., vol. 54, no. 2, pp. 49–93, 2013.[2] Q. Chu et al., “Impact of surface state in probeless friction stir spot welding of an Al–Li alloy,” Sci. Technol. Weld. Join., pp. 1–9, 2018.
Fig. 6 : Faciès de rupture en traction
© Obtenu de spacex.com
Fig. 1 : Échantillon soudé par FSW utilisé pour les essais mécaniques
Fig. 3 : Rupture en fatigue d’un échantillon
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Bénédicte RobitailleÉcole Polytechnique de Montréal
L. St-Georges - UQACP. R. Provencher - École Polytechnique de Montréal M. Brochu - École Polytechnique de Montréal
Prix Award
ex æquo
7474Propriétés en fatigue des joints soudés au laser continu à fil froid en AA6005-T6Fatigue properties of continuous laser cold-wire welding of AA6005-T6
L’utilisation d’alliages d’aluminium dans le soudage laser est en croissance par les efforts continus de réduire le poids et d’améliorer l’efficacité dans les industries du transport en raison de sa capacité d’automatisation et de flexibilité, de sa précision et sa répétabilité, de son faible apport de chaleur, de sa vitesse de soudage élevée, de sa zone affectée thermiquement étroite en raison du très fort gradient de température et de la faible distorsion de la soudure. Dans cette étude, le comportement en fatigue du soudage au laser à fil froid en alliage d'aluminium AA6005 T6 de 4,8 mm d'épaisseur a été étudié. Les performances en fatigue du joint soudé ont été quantifiées à l'aide de la technique à force contrôlée à amplitude constante afin d'obtenir des courbes S N. Dans la plage de faible contrainte, la résistance à la fatigue correspondant aux conditions d’interruption du test (107 cycles dans cette étude) varie de 25 à 35 MPa. À des contraintes élevées (60 MPa), la durée de vie des joints fabriqués est d'environ 105 cycles. Ces résultats de fatigue ont été recoupés avec des images métallographiques et une ana-lyse fractographique afin de mieux comprendre les mécanismes de nucléation et de propagation de la fissure. Toute rupture par la fatigue s’est produite dans la zone de soudure près de la ligne de fusion.
The use of aluminum alloys within laser beam welding is growing with the continuing effort to reduce the weight and improve the efficiency in transportation industries because of its capacity for automation and flexibility, precision and repeatability, low heat input, high welding speed, narrow heat affected zone due to extremely sharp temperature gradient, and low weld distortion. In this study, the fatigue properties of laser cold-wire welding made of 4.8 mm thick AA6005-T6 aluminum alloy were studied. The fatigue performances of the welded joint were quantified using the constant amplitude force-controlled technique to obtain S-N curves. At low-stress range, the fatigue strength corresponding to the run-out conditions (i.e., 107 cycles in this study) vary from 25 to 30 MPa. At high-stress range (around 60 MPa), the lifespan of the joints is about 105 cycles. These fatigue results were cross-referenced with metallographic, and fractographic analysis to better understand the crack nucleation and crack propagation mechanisms. All fatigue rupture occurred in the weld zone near the fusion line.
Pouria Salehi ÉTS
A. Maki - ÉTS Y. Zedan - ÉTS
M. Iben Houria - ÉTS N. Vanderesse - ÉTS
F. Atmani - ÉTS F. Mirakhorli - NRC
F. Nadeau - NRCV. Demers - ÉTS P. Bocher - ÉTS
Journée des étudiants – REGAL8 et 9 octobre 2019
P. Salehi1, A. Maki1, Y. Zedan1, M. Iben Houria1, N. Vanderesse1, F. Atmani1, F. Mirakhorli2, F. Nadeau2, V. Demers1, P. Bocher1
1 École de technologie supérieure, Department of Mechanical Engineering, 1100 Notre-Dame West, Montreal, H3C 1K3, QC, Canada2 National Research Council Canada, Aluminium Technology Centre, 501 University Boulevard East, Saguenay, QC, G7H 8C3, Canada
Fatigue properties of laser cold-wire welding of AA6005-T6
Propriétés en fatigue des joints soudés au laser à fil froid en
AA6005-T6
In recent years, the development of welding techniques was oriented troughthe fabrication of critical assemblies delivering high mechanical propertieseven for aluminum assembly having thick section higher than 3 mm [1]. Interms of strain induced by heat, the laser welding (LW) process forassembling thick aluminum components can be a potential candidate inreplacement of the conventional fusion welding [2]. Despite several studieswere performed on laser welding, most of them focused on the effect ofwelding parameters on keyhole, depth of penetration, weld pool geometry,porosity, and solidification crack without links with their fatigue performance[3, 4]. Liu et al. [5] explained the decrease in fatigue crack resistance of thefusion zone (FZ) by a grain boundary liquation accompanied by thedissolution of second-phase particles for LWed AA6156 joint. The role ofsuch defects directly derived from the LW will be studied in detail in thepresent work as they play important roles in the fatigue properties of weldedjoints. Fatigue properties of AA6005-T6 thick assemblies made from singlepass laser cold-wire welding process will be cross-referenced withmetallographic images, and fractographic analysis to investigate the cracknucleation and propagation mechanisms in the high-cycle fatigue regime.
Introduction
MaterialTable 1: Chemical composition of the base material (AA6005-T6) and the filler material (AWS ER5356) (wt. %)
Fig. 1: Joint design and configuration.
Fig. 2: Specimen used of fatigue tests according to ASTM-E466 standard
Si Fe Cu Mn Mg Cr Zn Ti AlAA6005 0.50 - 0.90 0.35 0.30 0.50 0.40 – 0.70 0.30 0.20 0.10 Bal.AWS ER5356 0.25 0.40 0.10 0.05 - 0.20 4.5 - 5.5 0.05 - 0.20 0.1 0.06 - 0.20 Bal.
At low-stress range, the fatiguestrength corresponding to the run-out conditions (i.e., 107 cycles)vary from 25 to 30 MPa. At high-stress range (around 60 MPa), thelifespan of the joints is about 105
cycles. All fatigue ruptureoccurred in the weld zone nearthe fusion line.
Conclusions
Bibliography1. Cornacchia, G., S. Cecchel, and A. Panvini, A comparative study of mechanical properties of metal inert
gas (MIG)-cold metal transfer (CMT) and fiber laser-MIG hybrid welds for 6005A T6 extruded sheet. TheInternational Journal of Advanced Manufacturing Technology, 2018. 94(5): p. 2017-2030.
2. El-Batahgy, A. and M. Kutsuna, Laser Beam Welding of AA5052, AA5083, and AA6061 AluminumAlloys. Advances in Materials Science and Engineering, 2009.
3. Wang, Z., et al., Laser beam oscillating welding of 5A06 aluminum alloys: Microstructure, porosity andmechanical properties. Optics & Laser Technology, 2019. 111: p. 58-65.
4. Zhang, C., M. Gao, and X. Zeng, Effect of microstructural characteristics on high cycle fatigue propertiesof laser-arc hybrid welded AA6082 aluminum alloy. Journal of Materials Processing Technology, 2016.231:p.479-487.
5. Liu, H., et al., Fatigue crack growth property of laser beam welded 6156 aluminium alloy. Fatigue &Fracture of Engineering Materials & Structures, 2014. 37(8): p. 937-944.
6. Zhou, J. and H.-L. Tsai, Porosity Formation and Prevention in Pulsed Laser Welding. Journal of HeatTransfer, 2006. 129(8): p. 1014-1024.
Fatigue properties of laser cold-wire welding on the 4.8 mm thick AA6005-T6aluminum alloy were studied.• All fatigue rupture occurred in the weld zone near the fusion line.• Fractography of specimens revealed that the fatigue properties improvement
pass through the minimization of the geometrical defects specially undercutdefect.
• Fatigue life depends not only on the applied stress but also on the defect size.
Abstract: The use of aluminum alloys within laser beam welding is growing with the continuing effort to reduce the weight and improve the efficiency in transportation industriesbecause of its capacity for automation and flexibility, precision and repeatability, low heat input, high welding speed, narrow heat affected zone due to extremely sharp temperature gradient,and low weld distortion. In this study, the fatigue properties of laser cold-wire welding made of 4.8 mm thick AA6005-T6 aluminum alloy were studied. The fatigue performances of thewelded joint were quantified using the constant amplitude force-controlled technique to obtain S-N curves. At low-stress range, the fatigue strength corresponding to the run-out conditions(i.e., 107 cycles in this study) vary from 25 to 30 MPa. At high-stress range (around 60 MPa), the lifespan of the joints is about 105 cycles. These fatigue results were cross-referenced withmetallographic, and fractographic analysis to better understand the crack nucleation and crack propagation mechanisms. All fatigue rupture occurred in the weld zone near the fusion line.
Fig. 4: Fractographic analysis of typical welded joint
Fig. 3: S/N curve at stress ratio R = 0.67
The optical microscope observations of the whole fracture surface (Fig. 4)revealed that the fatigue cracks were initiated from the top and root of the jointwere induced by the underfill and undercut defect (blue and yellow marks inFig. 4) with a Probability of Failure (P.O.F) less than 17 and 90% respectively.Therefore, the undercut defect seems to control the crack initiation during cyclicloading. Fig. 4 confirms that cracks nucleated from multi-initiations sites locatedat the root coalesce (indicated by red lines in Fig. 4) leading in a macroscopiccrack producing the final fracture.
Results and Discussion
Microvoids were observed in theFZ at the center and next to thefusion line (yellow circles in Fig.5) and according to Zhou et al.[6], were probably formedbetween low-melted eutecticphase and matrix at the last stageof the solidification due to highthermal shrinkage strains.
Fig. 5: Microstructure evolution, HAZ to FZ
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
75L’effet de l’azote comme gaz transporteur lors de la fabrication additive par
dépôt sous énergie dirigée d’aluminiumThe effect of nitrogen as a carrier gas for directed energy deposition
additive manufacturing of aluminum
Directed energy deposition (DED) is an additive manufacturing (AM) method that uses a high energy source to melt and deposit metal particles layer by layer to manufacture a part. It is commonly performed in a closed chamber filled with inert gas, usually argon, which is also used to feed the metal powder. Up to now, AM of metal components by DED focused mainly on stainless steels and titanium alloys. The lack of work on aluminium (Al) parts produced by DED is also amplified by the fact that there is a limited number of commercial Al alloys available under the form of powder for AM. It is well known that precipitates formed during solidification and/or heat treatment increase the mechanical properties of Al parts. This work studies the formation of nitrides in Al and Al-Ti alloys by using nitrogen as a carrier gas. Its aim is to improve strength, hardness and wear resistance of Al components produced by DED and increase the range of properties obtainable with the limited number of commercial Al alloys developed for AM.
Le dépôt par énergie dirigée (DED) est un procédé de fabrication additive (FA) utilisant une source à haute énergie pour fusionner et déposer une poudre métallique, couche par couche, pour fabriquer une pièce. Ce procédé est typiquement effectué dans une chambre fermée remplie de gaz inerte, normalement l’argon, qui est aussi utilisé comme gaz trans-porteur pour la poudre métallique. Jusqu’à maintenant, la FA de pièces métalliques par DED s’est concentrée majori-tairement sur les aciers inoxydables et les alliages de titane. Le manque de travaux portant sur les pièces d’aluminium (Al) produites par DED est amplifié par le nombre limité d’alliages d’Al disponibles commercialement sous forme de poudres. Il est connu que les précipités formés lors de la solidification et/ou d’un traitement thermique améliorent les propriétés mécaniques des pièces d’Al. Cette étude porte sur la formation de nitrures dans l’Al et des alliages d’Al-Ti en utilisant de l’azote comme gaz transporteur. Le but est d’améliorer la résistance mécanique, la dureté et la résistance à l’abrasion de pièces d’Al produites par DED et d’augmenter l’étendue des propriétés mécaniques possibles avec le nombre limité d’alliages commerciaux développés pour la FA.
Maude TremblayUniversité Laval
A. Bois-Brochu - CMQJ. N. Rousseau - CMQ C. Blais - UL
Journée des étudiants – REGAL8 et 9 octobre 2019
Maude Tremblay1, Alexandre Bois-Brochu2, Jean Nicolas Rousseau2 et Carl Blais1
1 Département des mines, de la métallurgie et des matériaux, Université Laval, 1065 Av. de la Médecine, Québec, QC, G1V 0A62 Centre de Métallurgie du Québec, 3095 Rue Westinghouse, Trois-Rivières, Qc, G9A 5E1
L’effet de l’azote comme gaz porteur lors de la fabrication additive par dépôt sous
énergie dirigée d’aluminiumThe Effect of Nitrogen as a Carrier Gas for
Directed Energy Deposition Additive Manufacturing of Aluminum
1
Introduction
L’aluminum en FA(2-3)
• Secteur principale: aéronautique• Nombre limité d’alliages
disponibles en poudre pour la FA
Pourquoi l’azote?(4)
• Coût inférieur à l’Ar• Formation in situ de
nitrures• Précipités↑
propriétés mécaniques des pièces d’Al
Ajout de Titane• Réactif avec l’azote• Forme des nitrures
Équipement LENS au CMQ
Figure 1: Tête de déposition LENS et schéma
• Laser: 1 kW• Poudres métalliques:
-150/+45 µm• Argon typiquement
utilisé pour chambre inerte et gaz transporteur de la poudre métallique
LentilleFaisceau Laser
Bain de fusion
Buses Coaxiales
ZX
Y
Alimentation
DED: Avantages et• Réparations ou ajouts fonctionnels• Gradient de composition• ↑ Densité• Poudre réutilisée
Faiblesses(1)
• ↓Précision• ↓ Fini de
surface
• Étudier l’effet de l’azote sur les pieces d’Al imprimées par DED• Former des nitrures et affiner les grains avec l’ajout de Ti et N
pour diminuer l’anisotropie• Améliorer la résistance et la dureté des pièces de DED en Al• Augmenter l’étendue des propriétés mécaniques possibles avec
le nombre limité d’alliages commerciaux développés pour la FA
Objectifs
• Poudres sphériques: AlSi7Mg -125/+65 µm et Ti64 -65/+45 µm• 0% et 10% Ti64 pré-mélangé avec la poudre d’AlSi7Mg• Chambre purgée à l’Ar < 100 ppm oxygène• Paramètres d’impression constants1-
Figure 2: Échantillons pour duretés Brinell et plan d’expérience
2-
Figure 3: Impression d’éprouvettes de traction réduites ASTM E8/E8M pour
essais de traction et micro-duretés dans les 4 conditions (Fig. 2)
Gas d’apport: Ar (0%N)
Al
Gas d’apport: 100% N
AlAl+ 10% Ti
Gas d’apport: Ar (0%N)
Al+10% Ti
Gas d’apport: 100% N
Résultats et Discussion
L’azote permet une impression sans interruption dû aux buses qui bouchent (Ar), améliorant l’efficacité du procédé
Figure 8: Distribution, par rayon-X EDS, d’Al et Ti dans les particules
a) sphérique b) fusionnées
a)
b)
Al Ti TiAl Ti(Al2.4Si0.6)TiNx Ti2NFigure 9: Cartographie EBSD des phases
dans les particules de Ti fusionnées
Figure 6: Cartographie EBSD avec diamètre équivalent des grains (d) et concentration d’azote ([N]) de pièce de DED en Al a) 0%N-0%Ti b) 100%N-0%Ti c) 0%N-10%Ti d) 100%N-10%Ti
a)
d)c)
b)d = 29.0 ± 26.0 µm [N] = 134 ± 8 ppm
d = 32.3±37.4 µm[N] = 125±3 ppm
d = 7.2 ± 2.7 µm[N] = 40 ± 11 ppm
d = 5.5 ± 2.1 µm[N] = 1141 ± 52 ppm • 10% Ti pré-mélangé avec l’AlSi7Mg affine les
grains et l’azote les affine d’avantage• L’azote comme gaz transporteur en DED d’Al:
•pas d’effet sur les propriétés mécanique de l’Al •↑ l’efficacité du procédé et ↓ coût •↑ la dureté des particules de Ti fusionnées •↑ les propriétés mécaniques de l’Al+10%Ti
Conclusions[1] S. Fale, A. Likhite, J. Bhatt,Transactions of the IndianInstitute of Metals 64 (2011)111-115. [2] Y. Mass, O. Amir,Additive Manufacturing 18(2017) 58-73. [3] J.R. Davis, ASMInternational, 2001. [4] Q. Hou,R. Mutharasan, M. Koczak,Materials Science andEngineering 195 (1995) 121-129.
Références
Matériaux et Méthodes
Figure 7: Propriétés en traction d’AlSi7Mg et AlSi7Mg+10%Ti produits sous Ar et N par DED
107 110142
163
207 208239
261
12.3 13.0
4.63.2
0246810121416
0
50
100
150
200
250
300
0%N-0%Ti 100%N-0%Ti 0%N-10%Ti 100%N-10%Ti
Allo
ngem
ent (
%)
σ(M
Pa)
Limite d'élasticité Résistance à la traction Allongement
Figure 4: Dureté d’échantillons d’Al et Al+10%Ti produit sous Ar et N par DED
Figure 5: Particules de Ti (MEB) a) sphériques
b) fusionnées c) distribution
a) b)
c)
61 62 72 81
298 311
513
631
72 7687
94
0
20
40
60
80
100
120
0%N-0%Ti 100%N-0%Ti 0%N-10%Ti 100%N-10%Ti0
100
200
300
400
500
600
700
Dure
té B
rinel
l (HB
W)
Micr
o-Du
reté
(HV)
Matrice Ti (sphériques) Ti (fusionnées) Alliage
• Ti ↑ dureté de l’Al• N ↑ dureté des
particules de Ti fusionnées et de l’alliage Al+10%Ti
• N effet non significatif sur l’Al
• Particules sphériques de Ti, non durcies
• 10% Ti dans l’AlSi7Mg ↑ la dureté et la résistance mais ↓ l’allongement
• La variabilité entre les réplicas d’une même condition diminue lorsque l’azote est le gaz transporteur
• Pas d'absorption de N par l'Al
• N n’a pas d’effet significatif sur les grains dans l’Al
• L'ajout de Ti affine les grains d'Al
• L'Al+10%Ti absorbe le N• N affine les grains dans
l'Al+10%Ti
• L’Al et le Ti diffusent l’un dans l’autre seulement dans les particules fusionnées
• Des nitrures de titane et des intermétalliques entre le Ti et l’Al sont formés dans les particules fusionnées
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
7676Optimisation de la fatigue et contrôle qualité des joints soudés par friction malaxage dans les platelages des ponts routiers en aluminiumFatigue optimization and quality control of friction stir welded joints in aluminum highway bridge decks
L'application limitée des alliages d’aluminium dans la réhabilitation des tabliers des ponts défectueux est prin-cipalement due à l'absence des outils de conception adéquats et des lignes directives normalisées. Certains phénomènes importants étudiés lors de la conception, tels que la fatigue, sont insuffisamment normalisés. Souvent, la fatigue gouverne la conception structurale des platelages des ponts routiers en aluminium (PPRA) dû à l’application des joints soudés présentant des fissurations microscopiques qui pourraient se propager sous l’effet des charges cycliques. Dans ce contexte, le soudage par friction-malaxage (FSW) promet une meilleure qualité de soudage ainsi que des meilleures résistances à la traction et à la fatigue. Cependant, les courbes S N caractérisant la résistance à la fatigue des joints FSW dans les PPRA n'ont pas été normalisées encore et ont été principalement extraites des résultats des essais à petite échelle non réalistes. L'objectif principal de cette phase du projet de recherche est de développer des courbes S N pour les configurations des joints FSW dans les PPRA en recourant à des essais de fatigue à grande échelle présentant des dimensions et des étendues de soudage pratiques. Des simulations par éléments finis ont été menées afin de définir le protocole expérimental et l’instrumentation correspondants.
The limited application of aluminum alloys in the rehabilitation of defected bridge decks is mainly due to the absence of adequate design tools and standard guidelines. Some important phenomenon studied in the design process such as fatigue are not sufficiently standardized. Fatigue governs the structural design of aluminum highway bridge decks (AHBD) in several cases due to the application of welded joints presenting microscopic de-fects, which could propagate under the effect of cyclic loads. In this context, Friction stir welding (FSW) promises a better welding quality and higher tensile and fatigue strengths. However, S-N curves characterizing the fatigue resistance of FSW joints in AHBD have not been standardized yet and were mostly extracted from non-realistic small-scale fatigue testing results. The main objective of this research project phase is to develop S-N curves for FSW joints configurations in AHBD through fatigue testing of large-scale specimens presenting practical dimensions and weld extent. Finite element simulations were conducted in order to define the corresponding experimental protocol and instrumentation.
Mahmoud Trimech Université Laval
C.-D. Annan - UL S. Walbridge - University
of Waterloo M. Fafard - CeAl
S. Amira - CQRDA
Journée des étudiants – REGAL8 et 9 octobre 2019
Mahmoud Trimech1, Charles-Darwin Annan1, Scott Walbridge2, Mario Fafard3 and Sofiene Amira4
1 Department of civil and water engineering, Université Laval2 Department of civil and environmental engineering, University of Waterloo
3 Centre d’expertise sur l’aluminium (CeAl) d’AluQuébec4 Centre québécois de recherche et de développement de l’aluminium – CQRDA
Fatigue optimization of friction stir welded joints in aluminum highway bridge decks
Optimisation de la performance en fatigue des joints soudés par friction malaxage dans les platelages des
ponts routiers en aluminium
1-Introduction
• Limited application of aluminum alloys in bridgeconstruction is due to the lack of design tools andstandard guidelines covering particular limit statesincluding fatigue
• Fatigue cracks initiate from vulnerable details suchas welds which are heavily applied in aluminumhighway bridge decks (AHBD)
• Recently, relatively new welding technology knownas friction stir welding (FSW) has been suggested toreplace conventional welding processes due to itsenhanced welding quality and fatigue strength [1]
Figure 1. Fatigue cracks initiation from welded joints in AHBD
2-Problem Statement
• Absence of S-N curves of FS welds in currentdesign codes and standards (CSA-S6, CSA-S157, ADM, AASHTO, Eurocode etc.)
• Available S-N curves in the literature are basedon small-scale fatigue tests [2,3]
• Previous large-scale fatigue tests are notrepresentative of current applications of FSW inhighway bridge decks (e.g. direction of bending)
• Lack of quality control criteria of FSW joints inAHBD
Figure 2. FSW AHBD in Saint Ambroise, Québec
3-Objectives
• Determine S-N curves of FSW joints in AHBDunder practical loading and boundary conditions
• Investigate the effect of mismatch on the fatigueperformance of FSW joints in AHBD
• Optimize the design of currently available FSwelded AHBD for enhanced fatigue strength
4-Methodology
• Design and fabricate representative specimensin order to assess the fatigue performance ofboth butt (Fig. 3) and lap joints (Fig. 4), typical inAHBD
• Fatigue tests will be carried out to simulate thepositive flexure in the bottom flange and thenegative flexure in the upper flange
• Finite element simulations to be used todetermine the test parameters and dimensions ofthe specimens, the load patch and the supports
Figure 3. Fatigue testing of butt joint specimensmade of AA6063-T6 (dimensions in mm)
Figure 5. Distribution of transversal stresses to the weld line S11; positive flexure in the bottom flange under
a load of 300 KN (stresses in MPa)
Figure 4. Fatigue testing of lap joint specimens made of AA6063-T6 (dimensions in mm)
• Specimens will be subjected to both constantamplitude (CA) and variable amplitude (VA)load ranges during the tests
• Fatigue tests will cover both optimally weldedand practical defective specimens (e.g.mismatch) to evaluate the tolerance levels
• Fracture mechanics analysis will be performedto extend the S-N curves to further tolerancelevels
Acknowledgements• Les fonds de recherche du Québec – Nature et
technologies (FRQNT)• SAFI Inc.
• Construction PROCO Inc.
References[1] J. Maljaars et al. “Structural design of
aluminum bridge decks for existing traffic
bridges”, creating and renewing urban
structures, 2005
[2] S. Guo et al. “Effect of quality control
parameter variations on the fatigue
performance of aluminum friction stir welded
joints,” International Journal of Fatigue, 2018
[3] R. Ranjan et al. “Fatigue analysis of friction stir
welded butt joints under bending and tension
load,” Eng. Fract. Mech., 2018
Aluminum Bridge Superstructure
Exagerated deformations
5-Upcoming Challenges
• Overcome the technical challenges during FSWprocess (design a backing plate to avoid thebending of the bottom thin flange during thewelding process)
• Implemented numerical simulations on Abaquswill be performed for the fatigue optimization ofthe welded joints (the placement, the applicationof radii in the weld vicinity etc.)
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
77Fabrication additive de AA2219 à l'aide du procédé de fusion par lit de poudre au laser
Additive manufacturing of AA2219 using the laser powder bed fusion process
Aluminum alloys are in high demand in the aerospace and automotive industries, mainly because of their high specific strength. Laser power bed fusion (LPBF) route of additive manufacturing (AM) has successfully fabricated aerospace brackets and nozzles with smart architecture and fine features. The use of aluminum alloys would further increase the buy-to-fly ratio of such components but the bulk of the literature and knowledge on AM focuses on the Al-Si-Mg system. Other opportunities for high strength alloys include the patented Scalmalloy by Airbus SE, for example. However, due to the high cost of scandium, there is a demand to find a compatible alternative to AM with lower cost and comparable properties. Aluminum-copper (Al-Cu) alloys could meet this opportunity but are subject to hot cracking and are thus less compatible with AM. The objective of this research is to study the unexplored heat-treatable 2219 Al-Cu alloy, using the LPBF process. Its solidification behavior has been investigated to control microstructure and reduce solidification cracking through process optimization. Finally, suitable combinations of solution heat treatment and ageing to be performed to enhance the strength.
Les alliages d'aluminium sont en forte demande dans les industries de l'aérospatiale et de l'automobile, prin-cipalement en raison de leur résistance spécifique élevée. La fusion par lit de poudre au laser (LPBF) pour la fabrication additive (FA) a permis de fabriquer des supports et des buses, pour l’industrie aérospatiale, dotés d’une architecture intelligente et des fonctionnalités raffinées. L'utilisation d'alliages d'aluminium augmenterait encore le rapport achat/vol de ces composants, mais l'essentiel de la littérature et des connaissances sur la FA concernent le système Al Si Mg. Parmi les autres possibilités offertes, citons le Scalmalloy breveté par Airbus SE. Cependant, en raison du coût élevé du scandium, il existe une demande pour trouver une alternative à un coût inférieur et des propriétés comparables. Les alliages aluminium¬¬ cuivre (Al Cu) pourraient répondre à cette demande mais sont sujets à la fissuration à chaud et sont moins compatibles avec la FA. L'objectif de cette recherche est d'étudier l'alliage inexploré Al¬ Cu 2219 pouvant être traité thermiquement, en utilisant le procédé LPBF. Son comportement en matière de solidification a été étudié pour contrôler la microstructure et réduire la fissuration par solidification grâce à l'optimisation des processus. Enfin, des combinaisons appropriées de traitement thermique en solution et de vieillissement doivent être effectuées pour améliorer la résistance.
Satish Kumar TumuluMcGill University
R. Tejas - McGill A. Nommeots-Nomm - McGill M. Brochu - McGill
Journée des étudiants – REGAL8 et 9 octobre 2019
S. K. Tumulu, R. Tejas, A. Nommeots-Nomm, M. BrochuREGAL Aluminum Research Centre, Department of Mining and Materials Engineering,
McGill University, Montreal, Quebec, Canada H3A 0C5
Additive Manufacturing of AA2219 using the Laser Powder Bed Fusion
processFabrication additive de AA2219 à l'aide du procédé de fusion par lit de poudre
au laser
Materials and Methods
Conclusion References
Results
• Powder characterisation showedpurchased AA2219 powder was not idealfor Laser Powder Bed Fusion.
• Coarsened melt pool boundaries and adiscontinuous secondary phase wereobserved in the manufactured samples.
Laser Powder Bed Fusion(LPBF) process• Layer by layer additive manufacturing (AM) process• Selectively melted using a laser beam in each layer• Near net shape complex geometry fabrication
Why Aluminum-Copper alloy?• High specific strength and low buy to fly ratio• High Toughness and solution heat treatable• But high crack sensitivity and hot cracking[3]
[1]. W.E. Frazier, Metal Additive Manufacturing: A Review, Journal of MaterialsEngineering and Performance 23(6) (2014) 1917-1928.[2].http://design-engineering.polimi.it[3]. DebRoy, T., et al., Additive manufacturing of metallic components – Process,structure and properties. Progress in Materials Science, 2018. 92: p. 112-224.[4]. C. E. Cross, "Weldability of aluminium-Lithium Alloys: An Investigation of HotTearing Mechanism," PhD, Colorado School of Mines, 1987.
• Gas atomized AA2219 powder from TLS Germany• Particle size distribution: particle size analyzer (Horiba LA-920)• Flow characteristics: Hall and Carney flowmeter• LPBF Manufacturing: Custom built LPBF machine at P2[AM]2
laboratory with a 6 kW pulsed laser source• Microstructure: Keyence VHX5000 digital optical microscope
and Hitachi SU3500 scanning electron microscope.Figure 6: (a)Particle size analysis and (b) Scanning electron
micrograph of the as received AA2219 powders
• Poor Flowability and presence of satellite • D10: 36.25 µm, D50: 58.27 µm and D90: 92.92 µm• Cu segregation in both powder & sample• Presence of delamination and porosity in sample
Figure 7: Microstructure: (a) optical microscopy and (b) scanning electron micrographs of the trial samples
(a) (b)
100µm
Introduction
Figure 4: Powder pile of as received AA2219 powder
Figure 1: Schematic representation of Laser Powder Bed Fusion process[1]
Figure 2: A heat exchanger made of aluminum alloy using LPBF process [2]
(a) (b)
Figure 5: Laser Characteristics used for LPBF AMFigure 3: Flow chart of the present research study
AcknowledgementWe thank the sponsor FRQNT-LU, for the project entitled: Additive Manufacturing of Aluminum Copper Alloys (AMACA).
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Prix Award
P2 [AM]2
7878 Analyse d’alliage d’aluminium par spectroscopie en perte d’énergie d’électron à 30keV30keV EELS analysis of aluminium alloys
La spectroscopie par perte d’énergie d’électrons (EELS) est une méthode d’analyse élémentaire qui permet la détection d’éléments légers lors d’observation en microscopie électronique en transmission. Cette technique permet donc l’étude de la composition de précipités contenants des éléments de masse atomique faible, tel que le lithium, qui jouent un rôle clé dans l’optimisation des propriétés mécaniques de certains alliages, tel l’Al2099. Un spectromètre EELS ainsi qu’un filtre en énergie ont été couplés à un microscope électronique en transmission à balayage à basse énergie. Le filtre en énergie de ce microscope permet d’obtenir des images de précipités à haut contraste grâce à l’excitation de différents modes de résonnance de plasmons. Les plasmons de surface peuvent être simulés par une méthode appelée dipôle discret approximation (DDA). Cette méthode requiert la fonction diélectrique des matériaux prenants part à l’interface. La DFT est utilisée pour calculer la fonction diélectrique de matériaux lorsque celle-ci n’est pas disponible dans la littérature scientifique.
Electron Energy Loss Spectroscopy (EELS), is an elemental analysis technique which allows the detection of light elements in a transmission electron microscope. EELS therefore allows the chemical analysis of light elements within precipitates, such as Lithium, which plays a crucial role in the strength of aluminum alloys such as AL2099. A EELS spectrometer as well as an energy filter have been coupled to a low-energy scanning transmis-sion electron microscope. The energy filtered imaging mode allows the rapid acquisition of high-contrast images of precipitates due to the local differences in plasmon resonance modes. Surface plasmons can be simulated using the discrete dipole approximation (DDA). This method requires the dielectric function of the materials taking part at the interface. Dielectric function that are not available from literature can be calculated using DFT.
Frédéric Voisard McGill University
N. Brodusch - McGillH. Demers -
Hydro-Québec M. L. Trudeau -
Hydro-Québec R. Gauvin - McGill
AX
E 2
: T
RA
NS
FO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
Journée des étudiants – REGAL8 et 9 octobre 2019
Frédéric VOISARD1, Nicolas BRODUSCH1, Hendrix DEMERS2, Michel L. TRUDEAU2 et Raynald GAUVIN1
1 Département de Génie des Matériaux, Université McGill, 3610 University, Room 2140, Montréal, QC H3A 0C52 Institut de recherche d’Hydro-Québec (IREQ), Sciences des Matériaux, 1800 Boul. Lionel-Boulet, Varennes, QC J3Z 1S1
Analyse d’alliage d’aluminium par spectroscopie en perte d’énergie d’électron
à 30keV30keV EELS Analysis of Aluminium Alloys
IntroductionLes alliages d'Al-Li-Cu ont d'excellents propriétés mécaniques enraison du réseau complexe de précipités qui se forment lors desprocessus de traitement thermique[1]. Il est donc primordial depouvoir mesurer la taille et la répartition des différents précipitésdans la matrice d’aluminium Les microscopes électroniques enbalayage (MEB) sont surtout reconnus pour leurs capacités àproduire des images de surfaces nettes. Les avancéestechnologiques récentes des microscopes ont grandement agrandi lechamp de compétences de ces instruments. Il est maintenantpossible d’effectuer des mesures qui était au paravent le domaineexclusif des microscopes électroniques en transmission (MET).
Diagramme simplifié du spectromètre, montrant le détecteur CCD de 1024 pixel, ainsi que le détecteur3 segments pour l’imagerie EF-STEM
Camera diffraction rétractable
Échantillion
Prismemagnetique
Détecteur de chand foncé annulaire à grand angle
Applications Les propriétés mécaniques des alliages métalliques sont dictées parleur microstructure. C’est d'ailleurs en contrôlant l’évolution de lamicrostructure des alliages qui permet les traitements thermiques. Ilest donc important de caractériser ces microstructures. Les méthodesillustrées ici offrent la possibilité de caractériser plusieurs aspect, doncla taille, le ratio d’aspect, la densité et même éventuellement lacomposition des précipités présent dans les alliages d’aluminium.
MéthodesSTEM à bas voltageLe SU9000EA est un microscope électronique à balayage en transmission(MEB-T) à tension d’accélération de 30 kV. Contrairement à la plupart desMEB, le SU9000EA est doté d’un porte échantillon intra-lentille, ce quipermet d’obtenir des images à facteur de grossissent de 3,000,000X.Comme ce microscope est surtout utilisé transmission, un spectromètrepermet de récolter le signal de pertes d’énergie d’électrons (EELS).
Spectroscopie de pertes d’énergie d’électron (EELS)En interagissent avec un spécimen, les électrons du faisceau d’unmicroscope perdent de l’énergie. Pour les échantillons minces, deuxphénomènes rentrent en jeux: l’ionisation et les plasmons [2]. Ces deuxphénomènes sont dépendent de la composition et de la structurecristallographique des matériaux. Il y a donc un contraste entre desmatériaux disparate.
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
-5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
Inte
nsity
(a.u
.)
Energy Loss (eV)
Plasmon Harmonics of Al-2099
Références[1] M. A. Muñoz-Morris and D. G. Morris, “Severe plastic deformation processing of Al–Cu–Li alloy for enhancing strength while maintaining ductility,” Scr. Mater., vol. 63, no. 3, pp. 304–307, Aug. 2010.[2] R. F. Egerton, Electron energy-loss spectroscopy in the electron microscope, Third edition. New York: Springer, 2011.[3] N. Brodusch, F. Voisard, and R. Gauvin, “About the contrast of δ’ precipitates in bulk Al–Cu–Li alloys in reflection mode with a field-emission scanning electron microscope at low accelerating voltage,” J. Microsc.
79Effet de l'ajout de Mo sur le comportement à la fatigue thermomécanique
de l'alliage de fonderie Al-Si 319Effect of Mo addition on the thermal-mechanical fatigue behavior
of Al-Si 319 foundry alloy
Al-Si 319 foundry alloys are widely used as engine blocks and cylinder heads in automobile industries. These engine components are often exposed to cyclic mechanical stress and cyclic operation temperature change, which can lead to the fatigue failure, especially the thermal-mechanical fatigue (TMF) failure. However, limited work has been performed on the TMF behaviors of Al-Si foundry alloys. In the present work, the influence of Mo addition on the TMF behaviors of Al-Si 319 foundry alloys has been investigated in the out-of-phase mode (OP-TMF) under cyclic temperature range of 60-300 ° and various strain amplitudes of 0.2-0.6%. Results showed that both 319 base and 319Mo alloy present the cyclic softening under all studied strain amplitudes. With Mo addition, high number of dispersoids formed and the higher fatigue strength as well as the longer fatigue life is observed in 319Mo alloy. The microstructure, including the intermetallic, precipitates and dispersoids are preliminary characterized to establish the relationship between the alloys and OP-TMF behavior.
Les alliages de fonderie Al-Si 319 sont largement utilisés dans les blocs moteurs et les culasses dans l'industrie automobile. Ces composants sont souvent exposés à des contraintes mécaniques cycliques et à des changements de température de fonctionnement cycliques, ce qui entraîne une défaillance en fatigue, en particulier une défaillance par fatigue thermomécanique (FTM). Cependant, peu de travaux ont été effectués sur le comportement de ces alliages à la FTM. Dans le présent travail, l’influence de l’ajout de Mo sur le comportement à la FTM des alliages Al-Si 319 est étudiée en mode de déphasage dans une plage de températures cycliques 60-300 ° et des amplitudes de déformation de 0,2-0,6 %. Les résultats ont montré que les alliages 319 de base et 319Mo présentent l’adoucissement cyclique pour toutes les amplitudes de déformation étudiées. Avec l'ajout de Mo, un nombre élevé de dispersoïdes sont formés et une résistance à la fatigue plus élevée ainsi qu'une limite d'endurance plus longue pour l’alliage 319Mo. La micros-tructure, y compris les composés intermétalliques, les précipités et les dispersoïdes, est caractérisée préalablement pour établir la relation entre les alliages et le comportement à FTM en déphasage.
Shuai WangUniversité du Québec à Chicoutimi
K. Liu - UQACX.-G. Chen - UQAC
Journée des étudiants – REGAL8 et 9 octobre 2019
Shuai Wang, Kun Liu ,X. Grant ChenNSERC/Rio Tinto Industrial Research Chairin Metallurgy of Aluminium Transformation,
University of Québec at Chicoutimi
Effect of Mo addition on the thermal-mechanical fatigue behavior of Al-Si 319
foundry alloy
Effet de l'ajout de Mo sur le comportement à la fatigue thermomécanique de l'alliage de
fonderie Al-Si 319
3. EXPERIMENTAL PROCEDURE
5. CONCLUSIONS 6. FUTURE WORK
Materials composition:
Alloy Si Cu Mg Mn Fe Ti Sr Mo Al319 5.93 3.34 0.12 0.284 0.307 0.11 0.0106 - Bal.
319Mo* 6 3.5 0.1 0.3 0.3 0.1 0.01 0.3 Bal.
Heat treatment:495°C/4h+515°C/2h, water quench, aged at 200°C for 5h.Sample preparation:Standard metallographic preparation techniques were applied.Micro-Hardness test:Vicker hardness test at room temperature with 25g F and 20s.Thermal-Mechanical Fatigue test:Test were performed under strain rate 10-3/s at temperature range of 60-300°C and under out-of-phase (OP) loading.
1. To analysis microstructure of 319 &319Mo via OM,SEM and TEM.
2. To study 356 & 356Mo TMF properties.3. To compare TMF & LCF properties of 319 &356 alloys.
1. INTRODUCTIONThermal-mechanical fatigue (TMF) behavior is
becoming more important in evaluating materialselevated-temperature properties, which used as thecritical components in automotive industries. Elevatedoperation temperature, dramatic temperaturegradient by cooling system, all lead to a complex cyclicchanges of load and temperature and give rise to TMFand limit use life of such critical components. One wayto overcome this problem is to design thermally stabledispersoids by adding Mo.
2. OBJECTIVES1. To study microstructure by adding Mo to 319 foundry alloys.2. To study TMF behavior of 319 foundry alloys.3. Characterize their TMF behavior by the typical hysteresis loops, cyclic stress/strain response and TMF life assessment.
The cyclic softening is generally observed in all strainamplitudes studied, in which the maximum tensile stressdecreases while the minimum compressive stress increaseswith increasing cycles.
At the cycle when sharp drop on the stress curve during TMFtest, or the stress falls by 30% below the initial min/max stresswas defined as the TMF life.
1. Thermal-mechanical fatigue tests between 60-300°C both 319 and 319Mo alloys show cyclicsoftening.
2. 319Mo alloy shows better fatigue life than 319 alloy.3. 319Mo shows higher strength after T7 heat treatment, but get lower strength than 319 with
increasing strain amplitude.
Accordingly, the TMF life of studied 319 Al alloy at 0.2%,0.4% and 0.6% strain amplitude are 402, 189 and 64 cycles, respectively, showing the decreasing TMF life with increasing strain amplitude.While the TMF life of studied 319Mo Al alloy at 0.2%, 0.4% and 0.6% strain amplitude are 1746, 245 and 107 cycles, respectively.
4.RESULTS
Fig.2 Variation of maximum and minimum stress of 319 Al alloy under various strain amplitudes vs. number of cycles.
Fig.1 Stress/strain hysteresis loops at various mechanical strain amplitudes of 319 Al alloy.
Fig.4 Variation of maximum and minimum stress of 319Mo Al alloy under various strain amplitudes vs. number of cycles.
Fig.3 Stress/strain hysteresis loops at various mechanical strain amplitudes of 319M Al alloy.
Fig.5 Micro-Hardness of 319 and
319Mo under different steps
AX
E 2
: TR
AN
SFO
RM
ATIO
N E
T A
PP
LIC
ATIO
NS
TR
AN
SFO
RM
ATIO
N A
ND
AP
PLIC
ATIO
NS
8080Méthodologie de caractérisation locale en traction monotone à l’aide d’échantillons miniaturesMethodology for local monotonic tensile characterization using miniature specimens
Cette étude s’inscrit dans le contexte de la fabrication additive (FA) des métaux. Avec la méthode de fusion sélec-tive laser sur lit de poudre, le taux de refroidissement est influencé par la géométrie locale des composants. Les structures de solidification des alliages d’aluminium sont donc hétérogènes, ce qui mène à des variations locales des propriétés mécaniques. Les spécimens de grande taille, conçus selon la norme ASTM-E8, ne permettent pas d’échantillonner ces variations. De plus, les grands spécimens sont mal adaptés au volume limité de la chambre de fabrication et ne sont pas représentatifs des sections fines des pièces technologiques de FA. Le but du projet est de développer une méthodologie de caractérisation en traction monotone avec des échantillons miniatures de dimensions proportionnelles et 3,5 fois plus petite que celles recommandées par la norme ASTM-E8. Une section calibrée de 1,5 mm2 par 7 mm de long a été utilisée. Il a été démontré que ces spécimens miniatures peuvent être éprouvés à l’aide de mors conventionnels. L’allongement a été mesuré sans contact en suivant l’écartement de repères par analyse d’images. Associée à des observations microstructurales, cette méthodologie permettra de comprendre l’évolution des propriétés mécaniques locales au sein de pièces de FA.
The context of this study is additive manufacturing (AM) of metals. The cooling rates induced by the laser powder bed fusion (LPBF) process depend upon local part geometry. The solidification structures of aluminum alloys are thus heterogeneous and cause local variations of the mechanical properties. The specimen dimensions dictated by the ASTM-E8 standard are too large to study such mechanical properties at this scale. Furthermore, those large specimens are a bad fit for the limited dimensions of the building chamber. They are also an inaccurate representation of the thin sections of AM parts. The aim of the project is to develop a methodology for tensile characterization using miniature specimens. The specimens are proportional but 3.5 times smaller than those specified by standard ASTM-E8. A reduced section of 1,5 mm² by 7 mm long was used. It was shown that these miniature specimens can be tested with conventional clamps. Tensile elongation was measured by tracking markings on the specimens by analysis of optical images. With the addition of microstructural observations, this methodology will enable the study of heterogeneous local mechanical properties of AM parts.
AX
E 1
: P
RO
DU
CTIO
N D
E L
’ALU
MIN
IUM
A
LU
MIN
IUM
PR
OD
UC
TIO
N
Nicolas Wawrzyniak École Polytechnique
de Montréal
Mathieu Brochu - Mc Gill P. R. Provencher - École
Polytechnique de Montréal Myriam Brochu - École
Polytechnique de Montréal
Prix Award
Journée des étudiants – REGAL8 et 9 octobre 2019
Nicolas Wawrzyniak1, Mathieu Brochu2, Paul R. Provencher1 et Myriam Brochu1
1 Polytechnique Montréal, Génie Mécanique, C.P. 6079, Succ. Centre-Ville, Montréal, Qc H3C 3A7, Canada2 Université McGill, 845 Rue Sherbrooke O., Montréal, Qc H3A 0G4, Canada
Méthodologie de caractérisation locale en traction monotone à
l’aide d’échantillons miniatures Methodology for local monotonic
tensile characterization usingminiature specimens
INTRODUCTIONLa fabrication additive permet la production rapide de pièces métalliques complexes. Toutefois, des variations locales des propriétés mécaniques sont observées au sein des pièces enalliages d’aluminium [1] et à différentes positions dans la chambre de fabrication [2]. De petits échantillons sont nécessaires pour caractériser ces variations, et sont plus représentatifsdes sections fines des pièces de fabrication additive. L’originalité du projet réside dans l’absence de norme pour la caractérisation en traction monotone avec des spécimens miniatures.OBJECTIF :L’objectif de cette étude est de développer une méthodologie de caractérisation locale en traction monotone avec des échantillons miniatures. Cette méthodologie sera employée dansla suite du projet de recherche pour comprendre les hétérogénéités de propriétés mécaniques au sein de pièces de fabrication additive.
1. MÉTHODOLOGIE1.1 Conception d’échantillonsDes échantillons miniatures, de dimensions proportionnelles mais environ 3,5 foisinférieures à celles spécifiées par la norme ASTM-E8 [3], ont été utilisés.
1.3 Mesure de l’allongementLes spécimens étant trop petits pour utiliser un extensomètre conventionnel,l’allongement a été mesuré sans contact. Des marques ont été peintes sur la sectionréduite (Figure 2). Le suivi de l’écartement entre les repères après traitement de la vidéode l’essai mécanique a permis le calcul de la déformation.
Figure 2. Principe de la méthodologie d’essai de traction monotone et de mesure de l’allongement.
CONCLUSIONS
3.2 Comportement des repèresLa mesure de l’allongement est basée surle suivi de la position des marques sur lasection réduite. Le glissement de certainsrepères a été observé après le bris del’échantillon (Figure 5), ce qui peut être àl’origine d’erreurs d’évaluation de leurposition. Le changement de méthoded’application des repères et la réductionde leur taille permettra de minimiser ceproblème.
3. DISCUSSION
Figure 4. Découpe d’un échantillon de fabrication additive pour mesures de la section
Figure 5. Glissement au cours de l’essai d’un repère sur la section réduite du spécimen
• La méthodologie proposée permet la caractérisation en traction monotoned’échantillons miniatures de fabrication additive.
• L’ajustement des différents paramètres d’essai permettra d’améliorer la courbe.• Le protocole de mesure des sections et d’application des repères devra être modifié.
• On devra évaluer la fiabilité de la méthodologie proposée, en étudiant lareproductibilité des résultats qu’elle permet d’obtenir.
• On devra évaluer la justesse de la méthodologie, en comparant les résultats obtenusavec des échantillons miniatures et normalisés, pour un matériau homogène.
2. RÉSULTATS
Figure 3. Courbe normalisée de la contrainte en fonction de la déformation obtenue.
3.1 Mesure des sectionsLa mesure précise des sections deséchantillons de fabrication additive estnécessaire pour un calcul exact descontraintes après l’essai mécanique. Uneméthode destructive a été utilisée, sur unéchantillon semblable à celui testé(Figure 4). Il sera nécessaire pour la suitedu projet de développer une méthode demesure non destructive. Une optionpourrait être la tomographie 3D.
Traitement de la
séquence d’images
Mesure de l’allongement par analyse d’image
La méthodologie d’essai a permis de tracer la courbe de contrainte en fonction de ladéformation de l’échantillon miniature éprouvé (Figure 3).
Une légère relaxation a été causée par la mise en pause momentanée de l’essai, forcéepar un problème avec la caméra. La suite de l’essai n’a apparemment pas été affectée.
[1] Dong, Z., Zhang, X., Shi, W., Zhou, H., Lei, H., & Liang, J. (2018). Study of Size Effect on Microstructure and Mechanical Properties of AlSi10Mg Samples Made by Selective Laser Melting. Materials (Basel), 11(12). doi:10.3390/ma11122463[2] Mishurova, T., Artzt, K., Haubrich, J., Requena, G., & Bruno, G. (2019). New aspects about the search for the most relevant parameters optimizing SLM materials. Additive Manufacturing, 25, 325-334. doi:10.1016/j.addma.2018.11.023 [3] ASTM (2016). E8/E8M – 15a; Standard Test Methods for Tension Testing of Metallic Materials
1.2 Mesure de la sectionLa section des échantillons miniatures defabrication additive a été mesurée paranalyse d’image après découpe. Unedifférence significative a été notée entrela consigne, la surface calculée à partir demesures avec un pied à coulisse, et lasurface mesurée par analyse d’image surles tranches (Figure 1).
Figure 1. Mesure des sections d’échantillons tels que fabriqués par fabrication additive.
Préparation et mesures de l’échantillon
Essai de traction monotone filmé A
B
𝜺𝜺 𝒕𝒕 = 𝑨𝑨𝑨𝑨 𝒕𝒕 − 𝑨𝑨𝑨𝑨(𝒕𝒕𝟎𝟎)𝑨𝑨𝑨𝑨(𝒕𝒕𝟎𝟎)
≈ 3 mm ≈ 3 mm
≈ 1 mm
Section 1
Section 2
≈ 0,5 mm
REMERCIEMENTSACKNOWLEDGEMENTS
8282
La mise en œuvre de l’ensemble des projets présentés dans cette encyclopédie nécessite des investissements majeurs et ce, tant au niveau des milieux universitaires et gouvernementaux que de la part des secteurs industriels concernés. C’est en parcourant cet ouvrage que vous réaliserez le dynamisme et l’ingéniosité de ces étudiants et professeurs, chercheurs passionnés, visant non seulement l’excellence, mais le développement d’un pôle québécois de recherche sur l’aluminium reconnu au niveau international.
Le Centre de recherche sur l’aluminium - REGAL tient à remercier les participants de la Journée des étudiants du REGAL qui, en acceptant la reproduction de leurs affiches, ont permis la création de cette encyclopédie.
The realization of the projects presented in this synopsis required major funding from key players working in the aluminium industry, including university, governments and various industrial sectors. When reading through this work, you will realise how dynamic and ingenious these passionate students, professors, and researchers are. They not only aim to excel, they wish to develop an internationally-recognised aluminium research hub in Quebec.
Aluminium Research Centre – REGAL would like to thank the participants of the REGAL Students’ Day who, by accepting to have their posters reproduced, made the creation of this synopsis possible.
Membres du bureau de direction du REGAL / Members of REGAL Steering Committee
Mario Fafard, directeur REGAL, Université LavalDaniel Marceau, directeur adjoint REGAL, UQACX-Grant Chen, UQAC
Houshang Alamdari, Université LavalFlorence Paray, McGill UniversityMyriam Brochu, École Polytechnique de Montréal
Victor Songmene, École de technologie supérieureAhmed Maslouhi, Université de SherbrookeGheorghe Marin, Cégep de Trois-Rivières
Le Centre de recherche sur l’aluminium - REGAL tient également à remercier l’ensemble de ses collaborateurs qui,de près ou de loin, ont contribué à la réalisation de cet ouvrage. Aluminium Research Centre - REGAL would also like to thank every one of their collaborators who, near or far, contributed to the production of this work.
SINTEFAarhaug, T. A.Gaertner, H.Ratvik, A. P.
Senanu, S.Wang, Z.
Massey UniversityHaverkamp, R. G.
RioTintoBilodeau, J.-F.Darmstadt, H.Gaboury, S.Guérard, S.
Javidani, M.Rometsch, P.Parson, N.Santerre, R.
Aluminerie AlouetteCôté, J. Charest, S.
Boisaco inc.Girard, A.
GraymontDargis, Y.
VerbomN. Bombardier
CQRDAAmira, S.
University of WaterlooWalbridge, S.
A3 SurfacesBernier, J. L.
AlcoaLauzon-Gauthier, J.Tessier, J.
Ziegler, D.
Hydro-QuébecDemers, H. Trudeau, M.L.
CNRC-CTAFraser, K.
NRCMirakhorli, F. Nadeau, F.
McGill UniversityBrochu, M.Dhillon, J. S.Espiritu, E. R. L.Fu, A.Gauvin, R.Hudon, P.Jaberi, A.
Muniz-Lerma, J. A.Nommeots-Nomm, A.Reichert, E. S.Tejas, R.Tumulu, S. K.Voisard, F.Waters, K. E.
Université de SherbrookeDesrochers, A.Faucheux, N.Maslouhi, A.
Nodeh, M.Soucy, G.
École de technologie supérieureAtmani, F.Bocher, P.Brial, V.David, G.Demers, V.Grandmont, P.Houria, M. I.Jahazi, M.
Javidikia, M.Maki, A.Plamondon, C. O.Salehi, P.Songmene, V.Vandererre, N.Zedan, Y.
École Polytechnique de MontréalBrochu, M.Leitao Martins, L. F.Lorans, A.
Provencher, P. R.Robitaille, B.Wawrzyniak, N.
Cégep de Trois-Rivières / CMQBois-Brochu, A.Giguère, N.
Levasseur, D.Rousseau, J.-N.
Université de ConcordiaSageghifar, M.
Dalhousie UniversityBishop, P.
Université LavalAlamdari, H.Annan, C.-D.Baiteche, A.Barry, T. S.Beudon, C.Blais, C.Chaouki, H.Conciatori, D.De Araujo Costa Rodri-gues, D.Djedid, A.Duchesne, C.Fafard, F.Gauvin, G.Hussein, A.Ishak, E.Kansoun, Z.Kavand, M.
Lacroix, O.Larachi, F.Larouche, D.Li, D.Lu, Y.Manolescu, P.Mollaabbasi, R.Ollevier, T.Oudjene, M.Rodrigues, D.Picard, D.Plante, J.Taghavi, S. M.Tran, T. H.Sanchez, T.Sorelli, L.Tremblay, M.Trimech, M.
Université du Québec à ChicoutimiAgbe, H.Ahmed, M.Alarie, J.Algendy, A.Attia, M.Belkacem, A.Bhattacharyay, D.Bouazara, M.Bureau, J.Cazenave, L.Chen, X. G.Chen, Z.Elashery, A.Farid, R.Khangholi, S. N.Kiss, L.Kocaefe, D.Kocaefe, Y.
Li, C.Li, D.Liu, K.Marceau, D.Mofarrehi, M.Morais, B.Poncsak, S.Ragab, K.Rahem, A.Rakhomonov, J.Rakotondramanana, L.Rastegari, A.Roger, T.Sadeghi-Chahardeh, A.Sarkar, D. K.St-Georges, L.Tremblay, S.-O.Wang, S.
Norwegian University of Science and Technology (NTNU)Branvik, T.Grande, T.Hansen, L. J.Jahrsengene, G.
Luneng, R.Svensson, A. M.Wang, Z.
Aluminium
Commanditaires / Sponsors
Alumine Bauxite
RemerciementsAcknowledgements
ICSOBA
Le Centre de recherche sur l’aluminium - REGAL est financé par le FRQNT