UNIVERSIDAD COMPLUTENSE DE MADRID FACULTAD DE CIENCIAS QUÍMICAS Departamento de Ciencia de los Materiales e Ingeniería Metalúrgica HIGH STRAIN-RATE BEHAVIOR OG MAGNESIUM ALLOYS (COMPORTAMIENTO MECÁNICO A ALTA VELOCIDAD DE DEFORMACIÓN DE ALEACIONES DE MAGNESIO) MEMORIA PARA OPTAR AL GRADO DE DOCTOR PRESENTADA POR Nathamar Valenthina Dudamel Caballero Bajo la dirección de los doctores María Teresa Pérez-Prado Francisco Gálvez Díaz-Rubio Madrid, 2013 © Nathamar Valenthina Dudamel Caballero, 2012 UNIVERSIDAD COMPLUTENSE DE MADRID FACULTAD DE CIENCIAS QUÍMICAS DEPARTAMENTO DE CIENCIAS DE LOS MATERIALES E INGENIERÍA METALÚRGICA HIGH STRAIN-RATE BEHAVIOR OF MAGNESIUM ALLOYS (COMPORTAMIENTO MECÁNICO A ALTA VELOCIDAD DE DEFORMACIÓN DE ALEACIONES DE MAGNESIO) A thesis submitted in fulfillment of the requirements for the degree of “Doctor por la Universidad Complutense de Madrid”. –European Mention- NATHAMAR VALENTHINA DUDAMELL CABALLERO SUPERVISED BY: DRA. MARÍA TERESA PÉREZ-PRADO PROF. FRANCISCO GÁLVEZ DÍAZ-RUBIO Madrid, julio 2012 UNIVERSIDAD COMPLUTENSE DE MADRID FACULTAD DE CIENCIAS QUÍMICAS DEPARTAMENTO DE CIENCIAS DE LOS MATERIALES E INGENIERÍA METALÚRGICA HIGH STRAIN-RATE BEHAVIOR OF MAGNESIUM ALLOYS (COMPORTAMIENTO MECÁNICO A ALTA VELOCIDAD DE DEFORMACIÓN DE ALEACIONES DE MAGNESIO) A thesis submitted in fulfillment of the requirements for the degree of “Doctor por la Universidad Complutense de Madrid”. –European Mention- NATHAMAR VALENTHINA DUDAMELL CABALLERO SUPERVISED BY: DRA. MARÍA TERESA PÉREZ-PRADO PROF. FRANCISCO GÁLVEZ DÍAZ-RUBIO Madrid, julio 2012 Lo importante es no dejar de hacerse preguntas... (The important thing is not to stop questioning…) (Albert Einstein, 1879-1955) A Christophe, mi mayor fortaleza. Mi compañero de aventuras. A mis hermanas y a mi familia, mi gran apoyo. A mis sobrinos y a mis pequeñas Hilda y María T. para que este logro les sirva de ejemplo de que, a pesar de las adversidades, con esfuerzo podemos lograr nuestras metas. A mi madre, con nostalgia. A mi padre, con cariño. AGRADECIMIENTOS Antes de iniciar el desarrollo de mi trabajo de investigación quisiera extender mi más profundo agradecimiento a las siguientes personas e instituciones:  A mis tutores, Dra. María Teresa Pérez-Prado y Prof. Francisco Gálvez Díaz-Rubio, por su apoyo absoluto, por su gran calidad profesional, por todo el conocimiento que me han trasmitido, por sus consejos durante todo el desarrollo de este trabajo y sobre todo, por el entusiasmo y las palabras de ánimo que me brindan en todo momento. Para mí ha sido un gran honor tenerlos de tutores.  Al Prof. Javier Llorca por abrirme las puertas del Instituto Madrileño de Estudios Avanzados de Materiales (IMDEA Materiales) y por darme la oportunidad de iniciar mi carrera de investigación en este excelente instituto.  Al departamento de Ciencia de Materiales de la ETSI Caminos Canales y Puertos, especialmente a la Ing. María Jesús Pérez y a la Ing. Noemí García Lepetit por toda su colaboración técnica en el desarrollo de esta tesis doctoral.  Al Dr. Gaspar González-Doncel, al Ing. Jesús Reales y al Ing. Marcos Angulo-Martin del Centro Nacional de Investigaciones Metalúrgicas (CENIM). A todos ellos mil gracias por su gran colaboración en el desarrollo experimental de esta investigación.  Al grupo de profesores del programa de doctorado “Ciencia y Tecnología de Materiales” de la Facultad de Ciencias Químicas de la Universidad Complutense de Madrid, por sus aportes y enseñanzas.  Al Dr. Dietmar Letzig, al Dr. Jan Bohlen, y al Dr. Sangbong Yi del Magnesium Innovation Centre (MagIC), de Helmholtz-Zemtrum Geesthacht (Alemania) por los conocimientos transmitidos, por abrirme las puertas de su prestigioso instituto y ofrecerme la oportunidad de una estancia que ha significado para mi una gran experiencia profesional y personal.  Al Dr. Ibai Ulacia, de la Universidad de Mondragón por el apoyo científico.  A mis compañeros de IMDEA Materiales por estos excelentes años y por el buen ambiente de trabajo.  A Dios porque en él todo lo puedo.  Desde el plano personal, quisiera agradecer muy especialmente a mis hermanas Nathaly, Marialberth y a Yadira por su apoyo incondicional. Al Ing. Miguel Ángel Contreras y a la Dra. Aurora Medina por facilitarme el camino. Además, de forma especial a: Christophe Ortiz, Yvonne Boutillier, Juan Ortiz, Katia Tamargo, Ana Fernández, Antoine Jérusalem, Juan Carlos Rubalcaba, Nora Cueto, Vanesa Martínez, Enrique Martínez, Berta Herrero, Eva Moreno, Teresa Pérez-Prado, Inés Leiva, Vanessa Fernández, Mariana Huerta, Eduardo Ciudad-Real, Lester López, Laura Castro, Patricia Ávila, Silvia Henández, Yinett Quesada y Lorena Andreu, por todos los momentos compartidos, por las alegrías en las buenas y el apoyo en las malas, por marcar la diferencia, por hacer que fuera de mi país, las cosas sigan siendo sencillas. Por ser mis amigos con todo lo que ello conlleva. ¡Gracias Chicos! TABLE OF CONTENTS viii TABLE OF CONTENTS PAG. 1. PRÓLOGO Y OBJETIVO DEL TRABAJO 11 2. INTRODUCCIÓN. ALEACIONES DE MAGNESIO 15 2.1. Características generales 16 2.2. Ventajas y desventajas 19 2.3. Aplicaciones 20 2.4. Estructura cristalina 22 2.5. Textura cristalina 24 2.6. Mecanismos de deformación 26 2.6.1. Mecanismos de deformación a bajas temperaturas 29 2.6.1.1. Policristal orientado al azar 29 2.6.1.2. Chapas laminadas 29 2.6.1.3. Barras extruidas 31 2.6.2. Mecanismos de deformación a alta temperatura 32 2.6.3. Mecanismos de deformación en aleaciones de magnesio con tierras raras (Mg-RE) 35 2.7. Aleaciones de magnesio procesadas mediante colada por inyección a alta presión 37 2.8. Objetivo de la investigación 38 3. RESEARCH PAPERS 43 3.1. Mechanical behavior and microstructural evolution of a Mg AZ31 sheet at dynamic strain rates 44 3.2. Twinning and grain subdivision during dynamic deformation of Mg AZ31 sheet alloy at room temperature 56 3.3. Influence of texture on the recrystallization mechanisms in an AZ31 Mg sheet alloy at dynamic rates 71 TABLE OF CONTENTS ix PAG. 4. COMPLEMENTARY STUDIES 80 4.1. Complementary Study I. Mechanical Behavior at quasi-static and dynamic rates of a magnesium alloy containing neodymium 81 4.1.1. Introduction. 82 4.1.2. Materials and Experimental Procedure 82 4.1.2.1. Material: initial microstructure 82 4.1.2.2. Mechanical testing 83 4.1.2.3. Microstructure examination 84 4.1.3. Results and Discussion 85 4.1.3.1. As-extruded and heat treated material 85 4.1.3.2. Mechanical behavior at quasi-static rates 86 4.1.3.3. Mechanical behavior at dynamic rates 94 4.1.3.4. Microstructural evolution under quasi- static and dynamic conditions 96 4.1.4. Conclusions 108 4.2. Complementary Study II. Dynamic deformation of high pressure die-cast Mg alloys 110 4.2.1. Introduction 111 4.2.2. Materials and Experimental Procedure 111 4.2.3. Results and Discussion 112 4.2.4. Conclusions 118 5. GENERAL DISCUSSION 120 5.1. Strain rate sensitivity of the yield stress in the Mg AZ31 alloy. 121 5.2. Strain rate dependence of the critical resolved shear stress of non-basal slip systems and of tensile twinning in the Mg AZ31 alloy. 122 TABLE OF CONTENTS x PAG. 5.3. Influence of strain rate on the twinning activity. 123 5.4. Influence of the strain rate on grain subdivision. 126 5.5. Dynamic recrystallization at high strain rates. 127 5.6. Influence of Rare Earth (RE) atoms in the incidence of basal slip and twinning at quasi-static rates. 130 5.7. Influence of Rare Earth (RE) atoms in the incidence of basal slip and twinning under dynamic deformations. 132 6. CONCLUSIONS (ENGLISH VERSION) 135 CONCLUSIONES (SPANISH VERSION) 138 7. FURTHER WORK 142 8. REFERENCES 144 9. APPENDICES 158 LIST OF FIGURES 178 LIST OF TABLES 187 PARTE I. PRÓLOGO Y OBJETIVO 11 PARTE I PRÓLOGO Y OBJETIVO PARTE I. PRÓLOGO Y OBJETIVO 12 1. PRÓLOGO Y OBJETIVO DEL TRABAJO. En la última década las aleaciones de magnesio se han estudiado profundamente ya que, debido a su baja densidad, se perfilan como materiales con gran potencial para reducir el peso de los vehículos. Esto conllevaría una disminución en el consumo de los combustibles fósiles y, a su vez, una reducción en la emisión de CO2. Sin embargo, para que las aleaciones de magnesio puedan ser ampliamente comercializadas, es necesario aumentar su resistencia mecánica y su ductilidad, disminuir su anisotropía para mejorar su capacidad para ser conformadas y aumentar su resistencia a la corrosión. Así mismo, es vital conocer sus propiedades mecánicas a velocidades de impacto, especialmente si se quiere fabricar con ellas determinados componentes estructurales de los automóviles. En la actualidad son muchos los trabajos que explican el comportamiento mecánico de las aleaciones de magnesio a velocidades de deformación cuasi-estáticas. Esta investigación pretende llenar la laguna conceptual existente en cuanto al comportamiento mecánico de estas aleaciones a alta velocidad de deformación. Por tanto, esta tesis doctoral, que se presenta con la finalidad de obtener el título de Doctor en Ciencia y Tecnología de los Materiales, tiene como objetivo principal estudiar el comportamiento mecánico así como los mecanismos de deformación y de recristalización en condiciones dinámicas de una chapa laminada de la aleación comercial Mg-3%pAl-1%pZn (AZ31), de una barra extruida de una aleación de magnesio con tierras raras de última generación (Mg-1%pNd-1%pMn, MN11) y de las aleaciones comerciales Mg-6%pAl-0.5%pMn (AM60B) y Mg-9%pAl-1%pZn (AZ91D) procesadas mediante colada por inyección. Con este propósito se realizaron ensayos de tensión y compresión en PARTE I. PRÓLOGO Y OBJETIVO 13 una barra Hopkinson a una velocidad de deformación de aproximadamente 103 s-1 y a temperaturas comprendidas entre 25ºC y 400ºC. También se realizaron ensayos de compresión a velocidades cuasi-estáticas (5x10-4 s-1, 5x10-3 s-1 y 5x10-2 s-1) con la finalidad de comparar los resultados obtenidos en ambos regímenes de deformación. Los resultados de este trabajo están recogidos en los siguientes 3 artículos publicados en revistas internacionales de elevado índice de impacto:  Acta Materialia. Vol. 58, Issue 8, Mayo 2010, Pag. 2988-2998.  Acta Materialia. Vol. 59, Issue 18, October 2011, Pag. 6949-6962.  Materials Science and Engineering: A,Vol. 532,Issue15, January 2012, Pag. 528-535. La estructura de este trabajo es la siguiente: primero se introduce el tema en el apartado 2. A continuación, en el apartado 3, se presentan los artículos de investigación que componen el cuerpo de esta tesis doctoral. En el apartado 4 se presentan dos estudios complementarios donde se recogen resultados adicionales obtenidos en esta investigación pero que no se encuentran recopilados en los artículos de investigación del apartado 3. Estos dos estudios complementarios darán lugar a sendos artículos científicos, que se encuentran en fase de preparación. En el apartado 6 se exponen las conclusiones más relevantes de este trabajo, el cual ha permitido establecer los mecanismos de deformación y recristalización predominantes durante la deformación de las aleaciones de magnesio estudiadas en condiciones dinámicas. Estos resultados son de gran utilidad para la industria de la automoción, ya que proporcionan información acerca de la aplicabilidad de este material en componentes que podrían estar sometidos a impacto. En el apartado 7 se presenta el trabajo futuro. En la sección 8 se enumeran las referencias utilizadas para escribir este trabajo de investigación. Por último, en el apartado 9 se presenta, a manera de apéndice, una base de datos de propiedades mecánicas de aleaciones de magnesio que PARTE I. PRÓLOGO Y OBJETIVO 14 consideramos de gran utilidad tanto para la industria automotriz como para trabajos futuros. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 15 PARTE II INTRODUCCIÓN. ALEACIONES DE MAGNESIO PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 16 2. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 2.1. Características Generales. El magnesio es un elemento químico que pertenece al Grupo IIa de la tabla periódica, es decir, forma parte de los metales alcalinotérreos. Su número atómico es 12 y su masa atómica es 24.3050 uma. Es el octavo elemento más abundante de la corteza terrestre. De hecho, el planeta Tierra contiene aproximadamente 2.5% en peso de este metal. En estado natural nunca se encuentra como metal puro ya que fácilmente reacciona con otros elementos para formar compuestos. Las principales fuentes de magnesio en la naturaleza son la magnesita, la dolomita y la carnalita. La magnesita es una mezcla de carbono, oxígeno y magnesio tal y como se puede observar en su fórmula química MgCO3. La dolomita está constituida por calcio, carbono, oxígeno y magnesio y su fórmula química es CaMg(CO3)2. La carnalita es un mineral compuesto de cloruro doble de potasio y de magnesio (KMgCl3·6H2O). Además, los océanos poseen gran cantidad de magnesio; cada kilogramo de agua de mar contiene 1.2 gramos de magnesio [1]. Las dos vías más utilizadas para obtener magnesio metálico mediante su separación del mineral son:  La reducción térmica: El mineral se calienta hasta alrededor de los 1600 ºC con reductores como aleaciones de hierro y silicio (ferrosilicio), carbono, CaC2, etc. Luego el magnesio se convierte en gas. Cuando el gas se enfría se transforma en magnesio líquido y posteriormente en magnesio sólido [1].  La electrólisis: Esta técnica utiliza corriente eléctrica para separar el magnesio de su mineral. Consiste en promover la PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 17 electrólisis del cloruro de magnesio en mezclas de sales fundidas [1]. Figura 2.1. Tabla periódica de los elementos. En su forma más pura el magnesio es un metal blanco plateado. Cuando se encuentra en forma de polvo es altamente reactivo y como sólido puede reaccionar, aunque muy lentamente, con el agua y oxidarse con el aire. En la gran mayoría de sus aplicaciones el magnesio se encuentra aleado con otros elementos tales como, por ejemplo, aluminio, cinc, manganeso, entre otros, para obtener materiales con propiedades mecánicas más atractivas. En comparación con otros metales estructurales de gran uso, como el acero y el aluminio, el magnesio tiene muy baja densidad (1.74 g/cm3) (Ver figura 2.2). Esta característica ha sido el motor que ha impulsado los estudios realizados en aleaciones de magnesio en la última década, ya que se ha considerado su gran potencial para contribuir en la reducción de peso en vehículos y otras estructuras [2,3]. Los materiales ligeros, como el Mg, tienen especial relevancia en una era como la actual en la que se sabe que los recursos energéticos de origen fósil son PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 18 limitados. La demanda de este tipo de materiales pretende reducir el consumo energético y, a la vez, disminuir el impacto medioambiental [4]. Introducir piezas de magnesio en el mercado requiere, dependiendo de su aplicación específica, mejorar su resistencia a la corrosión, incrementar su resistencia mecánica y su ductilidad y, a la vez, disminuir su anisotropía con la finalidad de mejorar su conformabilidad [5]. Además, es importante optimizar el comportamiento de las aleaciones de magnesio en condiciones de impacto [2], (es decir, a altas velocidades de deformación) especialmente cuando se diseñan piezas de magnesio para ser utilizadas en automóviles como componentes de seguridad crítica. Figura 2.2. Densidades de algunos metales estructurales. La tabla 2.1 recopila algunas de las propiedades físicas del magnesio y, a su vez, las compara con las de otros metales de gran uso estructural. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 19 Tabla 2.1. Comparación de las propiedades físicas del Mg puro con las de otros metales estructurales [6]. Gravedad Específica Punto de Fusión (ºC) Punto de Ebullición (ºC) Calor latente de fusión (kJ/kg.K; J/cm3.K) Calor específico (kJ/kg; J/cm3) Coeficiente de expansión lineal x106 Resistencia a la tracción (MPa) Elongación (%) Dureza (HB) Mg 1.74 650 1110 368; 640 1.05; 1.84 25.5 98 5 30 Al 2.74 660 2486 398; 1088 0.88; 2.43 23.9 88 45 23 Fe 7.86 1535 2754 272; 213 0.46; 3.68 11.7 265 45 67 2.2. Ventajas y desventajas. A continuación se enumeran algunas de las principales ventajas del Mg y sus aleaciones [3,6]:  Baja densidad. Es el metal estructural más ligero. Su densidad es 2/3 la del aluminio y 1/4 la del hierro.  Alta resistencia específica.  Buenas características de colabilidad. Es apto para colada por inyección a alta presión.  Buena soldabilidad en atmósfera controlada.  Abundancia tanto en la corteza terrestre como en los océanos.  Alta reciclabilidad además de bajo consumo de energía durante el proceso.  Capacidad de amortiguación de las vibraciones.  Buena capacidad para ser mecanizado.  Baja toxicidad en humanos.  Capacidad para inhibir ondas electromagnéticas  Si se compara con los materiales poliméricos: - Mejores propiedades mecánicas. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 20 - Mejor conductividad eléctrica. - Mejor conductividad térmica. Por otra parte, el magnesio también tiene algunos inconvenientes que han limitado su comercialización. Estos han sido y son objeto de muchas investigaciones cuyo desafío es superar, en lo posible, dichos inconvenientes. A continuación se presenta una lista de algunas de las desventajas más significativas [3,6]:  Bajo módulo elástico.  Baja tenacidad y baja ductilidad a temperatura ambiente. Esto va en detrimento de su aptitud para ser trabajado en frío.  Resistencia mecánica limitada a altas temperaturas.  Baja resistencia a la fluencia.  Alto porcentaje de grietas de contracción durante la solidificación.  Alta reactividad química.  Resistencia a la corrosión limitada (dependiendo de la aplicación). 2.3 Aplicaciones. Las aleaciones de magnesio son de gran interés tanto en el sector automovilístico como en el sector aeronáutico. En automoción, por ejemplo, al disminuir el peso de los coches se disminuye la cantidad de combustible necesario para ponerlo en marcha y, a la vez, los niveles de CO2 que se expulsan a la atmósfera también decrecen considerablemente [3,6]. La figura 2.3 muestra algunos componentes de automóviles que se fabrican ya en la actualidad con aleaciones de magnesio: la armadura del volante, la parte interna de puerta del maletero, la estructura interna de la puerta, la estructura del asiento, la carcasa de la caja de PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 21 transmisión, el colector de admisión, la carcasa de los cilindros y la columna de la dirección [6]. Figura 2.3. Aplicaciones de las aleaciones de magnesio en automóviles [6]. Sin embargo, las aplicaciones de las aleaciones de magnesio van más allá de las mencionadas anteriormente. Por ejemplo, en el área de los biomateriales ya se ha avanzado mucho en la fabricación de extensores coronarios biodegradables a partir de aleaciones de Mg. Estos componentes se introducen en el interior de una arteria coronaria obstruida con el fin de que se produzca la dilatación de la misma para así facilitar el flujo sanguíneo. La ventaja de utilizar Mg en este tipo de aplicaciones es que el cuerpo humano posee magnesio naturalmente por lo que es muy improbable que se produzca una infección debido a sus productos de corrosión [7]. El magnesio se utiliza también para fabricar material deportivo y dispositivos electrónicos. En el área deportiva encontramos, por PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 22 ejemplo, marcos de bicicletas, raquetas, patines, gafas, binoculares, entre otros. En el área electrónica se pueden mencionar, por ejemplo, carcasas de portátiles, de móviles, de cámaras fotográficas y de video. La Figura 2.4 muestra algunas de estas aplicaciones. Figura 2.4. Aplicaciones de las aleaciones de magnesio para la fabricación de dispositivos biomédicos, electrónicos y material deportivo. 2.4. Estructura cristalina. La estructura cristalina del magnesio puro, en condiciones de presión atmosférica y temperatura ambiente es hexagonal compacta (HCP) [8]. La figura 2.5 muestra una representación, mediante esferas reducidas, de la celdilla unidad HCP. En dicha figura se puede observar que las bases superior e inferior (planos basales) son hexágonos regulares con un átomo en cada vértice y otro en el centro. Los planos perpendiculares a los basales reciben el nombre de planos prismáticos. Además, hay otro plano que provee tres átomos adicionales a la celdilla unidad y está ubicado entre los planos basales. Cada celda unitaria HCP tiene 6 átomos: cada átomo de los 12 vértices superiores e inferiores contribuye a la celda unidad con 1/6 de átomo mientras que los tres átomos del plano central contribuyen enteramente. En la figura PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 23 II.5 también se observa que “c” y “a” son las dimensiones de la celda unitaria. La relación c/a ideal es de 1.633; no obstante, para la mayoría de los metales HCP esta relación se desvía de la idealidad. Los valores de los parámetros de red del magnesio son c=0.52105 nm y a=0.32095 [8]. Esto quiere decir que la relación c/a para el magnesio a temperatura ambiente es 1.6235; por ello se le podría considerar un metal con estructura hexagonal muy próxima a la ideal. Los vértices a1, a2, a3 (separados entre sí 120º) y el eje C, son los índices de Miller que permiten definir los planos de las celda (a1, a2, a3, C), cumpliéndose la regla de que a1+a2=-a3 [8-10]. Figura 2.5. Estructura cristalina hexagonal compacta (HCP): representación de la celdilla unidad mediante esferas reducidas. Ahora, en este contexto, se hace necesario introducir el concepto de textura cristalina (o cristalográfica), ya que debido a la baja disponibilidad de sistemas de deslizamiento independientes para cada modo de deslizamiento (este aspecto se tratará en detalle en el apartado 2.6), la deformación de las aleaciones de magnesio es altamente dependiente de la textura [11-12]. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 24 2.5 Textura cristalina. Se entiende por textura cristalina la distribución de las orientaciones cristalográficas de un metal policristalino. Atendiendo a su textura, los materiales pueden clasificarse en:  Sin Textura: Cuando los cristales están orientados completamente al azar.  Con textura: Cuando existen una o varias orientaciones preferentes (componentes de la textura). Dependiendo de la fracción de volumen de material que está orientada según las distintas componentes se dice que éste tiene textura débil, moderada o fuerte. Debido a que la orientación preferente de los granos es un fenómeno bastante común, la textura cristalina juega un rol fundamental en el comportamiento mecánico y físico del material. La textura evoluciona durante el procesado (colada, forja, laminación, extrusión, etc.), la soldadura y los tratamientos térmicos. Durante la deformación de un material policristalino el flujo plástico produce una reorientación de la red de cada grano y esto da lugar a la estabilización de orientaciones preferentes. La mayoría de los principios que gobiernan el desarrollo de una determinada textura durante los procesos convencionales de conformado han sido ampliamente estudiados y por tanto, en la actualidad es posible predecir la textura que tendrá un material fabricado mediante un determinado proceso [13,14]. La figura 2.6 muestra algunos ejemplos de las texturas que se desarrollan en las aleaciones de magnesio en tres procesos de fabricación distintos: la colada, la laminación y la extrusión. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 25 Colada Textura: Orientada al Azar Laminación + recocido Textura: Textura fuerte de fibra (0001) ND RD Extrusión Textura: Textura fuerte de fibra 0110 ED Sin Textura <0001> Figura 2.6. Tipos de textura que se desarrollan en los materiales procesados mediante tres procesos de fabricación: colada, laminación más un recocido posterior y extrusión. Como se puede observar en la figura anterior, el tipo de procesado influye en la textura del material. Así, por ejemplo, durante un proceso de colada los granos cristalinos se orientan al azar. En cambio, el flujo plástico generado durante un proceso de laminación seguido de un tratamiento de recocido favorece el desarrollo de una textura fuerte de fibra en la que los hexágonos se encuentran con el eje “c” perpendicular al plano de laminación. Finalmente, en las barras extruidas de magnesio también se genera una textura fuerte en la cual la dirección 0110 de la celda cristalina se alinea con el eje de extrusión. La red cristalina HCP, altamente anisótropa, confiere a las aleaciones de magnesio unas propiedades mecánicas particulares. Esto es así porque el número de sistemas de deslizamiento equivalentes es limitado. Así, las aleaciones de magnesio texturadas (esto es, cuyos granos no están orientados al azar), como es el caso de las aleaciones forjadas, presentan una fuerte anisotropía mecánica, una asimetría tensión- compresión en el límite elástico y baja ductilidad. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 26 En la sección que se presenta a continuación se describirán los principales mecanismos de deformación del magnesio para distintas aleaciones con diversas texturas y se discutirán las condiciones bajo las cuales se activan unos u otros. 2.6. Mecanismos de deformación. Los metales con estructura cristalina HCP tienen un número menor de sistemas de deslizamiento independientes para cada modo de deformación que los que poseen los metales con estructura cúbica centrada en el cuerpo (BCC) y cúbica centrada en las caras (FCC) [15,16]. En el caso particular del magnesio y sus aleaciones, los mecanismos de deformación que operan a bajas velocidades de deformación han sido exhaustivamente investigados en los últimos años [11-12,17-44]. Existen evidencias experimentales que muestran que en las aleaciones de Mg el deslizamiento de dislocaciones o “slip” puede producirse a lo largo de la dirección 0211 , también llamada dirección a , tanto en los planos basales ( 0001 ) como en los planos no basales (en los prismáticos  0110 , y en los piramidales  1110 ). Además, se ha observado deslizamiento a lo largo de la dirección ac  en planos piramidales  1110 y  2211 [34]. La figura 2.7 muestra los principales sistemas de deslizamiento que se pueden activar en las aleaciones de magnesio. Los mecanismos de deformación que operan a velocidades de deformación dinámicas aún no han sido estudiados en profundidad. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 27 Figura 2.7. Sistemas de deslizamiento de las aleaciones de magnesio. La deformación también pude tener lugar mediante la activación del maclado, principalmente a lo largo de los planos  2110 y  1110 y  3110 . Éste juega un papel importante durante la deformación de aleaciones de magnesio, y lo hace muy particularmente a bajas temperaturas [21,26,29,32,34,44] y altas velocidades de deformación. El maclado es un mecanismo polar [45]. Por ejemplo, el maclado  2110 (también llamado maclado de extensión) se activa en un grano sólo cuando las condiciones de deformación son tales que se produce la extensión del eje C de la red cristalográfica hexagonal [38]. Análogamente, el maclado  1110 , o maclado de compresión se activa cuando la deformación aplicada promueve la compresión del eje C de la red HCP [46]. La figura 2.8 muestra los dos tipos de maclado más comunes en las aleaciones de magnesio. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 28 <1012> (1012) {1012} <1011> C {1011} C (1011) Figura 2.8. Sistemas de maclado más comunes en las aleaciones de magnesio. La tensión de cizalladura crítica resuelta o “Critical Resolved Shear Stress” (CRSS) se define como el esfuerzo cortante mínimo que se debe aplicar sobre un plano de deslizamiento específico a lo largo de una dirección de deslizamiento concreta para que se produzca el deslizamiento cristalográfico. La CRSS varía considerablemente para cada uno de los sistemas de deslizamiento [16]. A pesar de que se ha publicado un amplio intervalo de valores de tensión de cizalladura crítica resuelta para los diferentes sistemas de deslizamiento y maclado [21, 26, 29, 32,34 ,44], se acepta generalmente que: CRSSBasal < CRSSMaclado < CRSSPrismático < CRSSPiramidal En [47] se recoge un resumen de las CRSS obtenidas de experimentos en monocristales, aunque los valores absolutos aún son un tema de debate [48]. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 29 2.6.1. Mecanismos de deformación a bajas temperaturas. 2.6.1.1. Policristal orientado al azar. Durante la deformación uniaxial de policristales de magnesio, orientados al azar, con tamaño de grano convencional (~10-50 µm), a baja velocidad de ensayo y a temperatura ambiente, el deslizamiento en planos basales y el maclado de extensión  2110 son los principales mecanismos de deformación [17-22]. En estas condiciones también se activa el deslizamiento no basal, pero en una proporción mucho menor [31,34,36 ]. 2.6.1.2. Chapas laminadas. Como se explicó anteriormente, las chapas laminadas poseen una textura basal, esto es, en la mayoría de los granos el eje C es paralelo a la dirección normal de laminación (ND). Cuando se comprime una placa de magnesio a lo largo de ND, es decir, cuando el eje de compresión es paralelo al eje C para la mayoría de los granos, el deslizamiento basal y el deslizamiento prismático y el maclado  2110 no están favorecidos (ya que su factor de Schmid es prácticamente cero) y, por esta razón, el deslizamiento piramidal [34] y el maclado  1110 se activan desde los primeros estadios de deformación [36]. El maclado se produce mayoritariamente en los planos  1110 , causando una reorientación de la red de aproximadamente 56º alrededor del eje 0211 . Ocasionalmente, áreas que han sufrido maclado de contracción se encuentran en una orientación favorable para un subsecuente maclado de extensión. Cuando ocurren de forma consecutiva estos dos mecanismos de maclado, se produce lo que se conoce como mecanismo de maclado secundario [46]. El ángulo de desorientación de las fronteras de macla que rodean un área con maclado secundario es de PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 30 aproximadamente 38º alrededor del eje 0211 . La importancia del maclado de contracción y del maclado secundario para promover la compresión del eje C en aleaciones de Mg no se conoce con certeza. De hecho, las evidencias experimentales disponibles sugieren que su contribución neta a la deformación total no es muy importante. En particular, se ha afirmado que el maclado de contracción y el maclado secundario se activan en monocristales de magnesio ensayados con el eje de tensión paralelo al plano basal [49-51]. Sin embargo, los estudios en monocristales fueron llevados a cabo hace varias décadas y mayoritariamente utilizando exclusivamente microscopía óptica. Además, los indicios de esos mecanismos han sido hallados solamente en regiones muy localizadas (por ejemplo, muy cerca de la superficie de fractura o de la misma superficie de la muestra). Finalmente, la ductilidad de los monocristales raras veces excede el 2% y, por lo tanto, no queda muy claro si este mecanismo puede ser responsable de las elevadas deformaciones a fractura que son típicamente observadas en policristales de magnesio. Algunos estudios recientes han mostrado también la activación del maclado de contracción y del maclado secundario en aleaciones de magnesio policristalinas durante la compresión del eje C [12,36,52-54]. Sin embargo, nuevamente, los indicios de estos tipos de maclado son observados únicamente en regiones muy pequeñas, con frecuencia muy cerca de las superficies de fractura. Por todo ello, en la actualidad se acepta de forma general que el deslizamiento cristalográfico piramidal acomoda la mayoría de la deformación de compresión a lo largo del eje C [29,55]. Algunos autores han destacado el papel de las maclas de compresión en el endurecimiento por deformación (strain hardening) de las aleaciones de Mg [55]. Los sistemas de deslizamiento prismáticos se activan cuando se realizan ensayos de tracción a lo largo de la dirección de laminación (RD) o a lo largo de la dirección transversal a la de laminación (TD) PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 31 [34,36]. Por el contrario, si a temperatura ambiente se comprime una lámina de magnesio a lo largo de RD, es decir, de tal manera que el eje de compresión es perpendicular al eje “c” en la mayoría de los granos, entonces es el maclado  2110 el que predomina desde las primeras etapas de la deformación, dando lugar a una rotación de la red cristalina de los granos de aproximadamente 86º, de forma que el eje “c” gira hasta casi alinearse con ND. Cuando un número suficiente de granos se ha orientado así, comienza a predominar el deslizamiento piramidal y el maclado  1110 , como se explicó anteriormente. Como consecuencia, se produce un importante endurecimiento por deformación debido a las interacciones dislocación–dislocación y macla– dislocación. La actividad de los distintos mecanismos de deformación debido a los diferentes ángulos entre la dirección de carga y el eje “c” genera una asimetría en el límite elástico [33,43,44]. 2.6.1.3. Barras extruidas. En barras de magnesio extruidas, en las cuales el eje C se alinea con la dirección radial (perpendicular al eje de extrusión) durante el procesado, el maclado de extensión predomina cuando se aplica una carga de compresión en la dirección de extrusión. Si una barra extruida se deforma en tracción a lo largo del eje de extrusión los mecanismos predominantes son el deslizamiento prismático y, en menor medida, el piramidal. En los últimos años algunos estudios han investigado el comportamiento mecánico dinámico de aleaciones de magnesio obtenidas mediante procesos de conformado, como extrusión o laminación [56-66]. Estos trabajos analizan el efecto del incremento de la velocidad de deformación en el límite elástico, la tensión máxima, y la elongación a fractura, así como en los mecanismos de deformación que PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 32 se activan. Se ha sugerido que, en condiciones dinámicas, tiene lugar un incremento de la sensibilidad a la velocidad de deformación, de la ductilidad y de la capacidad de absorción de energía. También se ha observado un aumento de la actividad del maclado de extensión con la velocidad de deformación y se sugiere que este mecanismo opera incluso a temperaturas bastante elevadas, a las cuales no se encuentra activo en condiciones de ensayo cuasi-estáticas. Sin embargo, en general, los mecanismos de deformación predominantes a velocidades de impacto en aleaciones de Mg son todavía poco conocidos. 2.6.2. Mecanismos de deformación a alta temperatura. A medida que aumenta la temperatura, a bajas velocidades de deformación y en aleaciones con tamaños de grano convencional (d~5- 50 m), la CRSS del deslizamiento basal y la del maclado de extensión  2110 permanecen constantes [32], pero las de los sistemas de deslizamiento no basales y la del maclado de contracción disminuyen gradualmente [32]. De esta forma, a altas temperaturas la actividad de estos sistemas de deslizamiento no basal aumenta, facilitando la compatibilidad intergranular [20, 23, 34, 35,37,42]. En la figura 2.9 se muestra la variación de la CRSS de los distintos modos de deformación con la temperatura, según un modelo ideado por Barnett [32]. Las curvas fueron calculadas utilizando las ecuaciones que se presentan en la tabla 2.2, donde Z es el parámetro de Zener-Hollomon,       RT Z 147000exp , R es la constante de los gases ideales y T, la temperatura. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 33 Tabla 2.2 Valores de la CRSS para los diferentes sistemas de deslizamiento y maclado en Mg según Barnett [32]. Parámetro Valor basalCRSS MPa5 prismaticCRSS MPaZ 38)ln(5.2  acCRSS  MPaZ 32)ln(1.2  twinningCRSS Twinning Region MPa32 112107  sxZ A temperaturas mayores que 250ºC la CRSS de los sistemas no basales se hace finalmente menor que la del maclado de extensión y comparable con la del deslizamiento basal. A temperaturas ligeramente superiores (T > ≈300ºC) la CRSS para el maclado de compresión también disminuye significativamente, aunque sigue manteniendo valores más altos que los de los otros modos de deformación. Como resultado, con el incremento de temperatura, la deformación se va acomodando progresivamente por deslizamiento múltiple [34,67,68]. Algunos autores han resaltado también la importancia del deslizamiento de fronteras de grano a temperaturas superiores a los 300ºC [39,42]. Debido a la activación de un mayor número de modos de deformación, la ductilidad aumenta y el comportamiento mecánico de las aleaciones de magnesio, incluso las de textura fuerte, se vuelve más isótropo. La deformación cuasi-estática a alta temperatura de aleaciones de magnesio con tamaño de grano convencional viene acompañada usualmente por recristalización dinámica (DRX). Hasta la fecha se han llevado a cabo bastantes estudios con el propósito de dilucidar el mecanismo de DRX predominante a distintas temperaturas y velocidades de deformación [69,70-82]. Ion et al. [70] publicaron en PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 34 1982 un estudio pionero en esta materia, en el cual observaron que la DRX era altamente dependiente de la temperatura de deformación. A T > 300-350ºC predominaría la recristalización dinámica discontinua (DDRX) [83,84,85], es decir, la nucleación de granos libres de deformación y el crecimiento de los mismos a expensas de las regiones deformadas mediante la migración a larga distancia de las fronteras de grano de ángulo alto. Esto ha sido confirmado, por ejemplo, en [76]. A temperaturas inferiores prevalecerían procesos de tipo continuo (CDRX), que consisten en la formación de límites de subgrano por acumulación de dislocaciones y el aumento gradual de la desorientación de éstos ( ) con la deformación [83,84]. Numerosos estudios han descrito la presencia de procesos continuos a temperaturas moderadas [69,74- 75,77-78,80-82]. Más recientemente, las condiciones específicas bajo las cuales predominan la DDRX y la CDRX, han sido asociadas con determinados valores críticos del parámetro de Zener-Hollomon (Z), así como con la velocidad de deformación que también desempeña un papel importante (opuesto al de la temperatura) [28]. Dependiendo de la composición del material y de las características microestructurales, la DDRX y la CDRX pueden coexistir en un amplio intervalo de valores de Z. Los mecanismos de deformación y recristalización de aleaciones de magnesio a velocidades de deformación de impacto (~103 s-1) todavía no se conocen, ya que se han llevado a cabo muy pocos estudios en este área [56-62]. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 35 Figure 2.9. Variación de la CRSS de los distintos sistemas de deslizamiento y maclado con la temperatura [32,58]. 2.6.3. Mecanismos de deformación en aleaciones de magnesio con tierras raras (Mg-RE). Uno de los obstáculos a la entrada masiva de aleaciones de Mg en el mercado automovilístico es su baja capacidad para ser conformadas que, como habíamos mencionado anteriormente, viene originada por la gran anisotropía mecánica presente en las aleaciones comerciales convencionales. Recientemente se ha descubierto que esta anisotropía se puede reducir en gran medida mediante la aleación de Mg con elementos de tierras raras (RE), puesto que estas aleaciones poseen texturas de extrusión y laminado significativamente débiles [86-99]. A pesar de los grandes esfuerzos puestos en los últimos años en investigar aleaciones de Mg-RE, aún no se conoce claramente el origen de esta textura débil. Tampoco se conocen las contribuciones relativas a la deformación de los diferentes sistemas de deslizamiento y maclado en PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 36 aleaciones de Mg-RE a velocidades cuasi-estáticas. Algunos estudios han sugerido que la actividad del deslizamiento no basal aumenta debido a la interacción de elementos de tierras raras con las dislocaciones [87,88,96,98,99], lo que hace que los valores de la CRSS en el deslizamiento basal y no basal se encuentran mucho más próximos entre sí que en el caso de las aleaciones de Mg que no contienen RE. Otra prueba de la interacción entre los átomos de soluto de RE y las dislocaciones es la observación del fenómeno de envejecimiento dinámico por deformación en varias aleaciones de Mg- RE [93,94,97,100-101]. Los mecanismos de deformación que operan a velocidades dinámicas en estas aleaciones todavía no han sido investigados. Como se había mencionado anteriormente, hasta la fecha se han llevado a cabo muchos estudios con la finalidad de elucidar los mecanismos de recristalización dinámica (DRX) en aleaciones de magnesio convencionales [70-82]. Sin embargo, los mecanismos de recristalización predominantes en aleaciones de Mg-RE no se conocen todavía. Se ha demostrado que los elementos de tierras raras debilitan drásticamente las texturas de recristalización. Esto fue inicialmente atribuido al fenómeno de “particle stimulated nucleation” (PSN) [43]. Sin embargo, después se ha mostrado que adiciones muy pequeñas de elementos de tierras raras, que no resultan en la precipitación de partículas de segunda fase, también pueden alterar notablemente las texturas de recristalización [86,93,102]. Se ha propuesto que la segregación de átomos de tierras raras en las fronteras de grano puede influir en las texturas de recristalización modificando la movilidad relativa de las fronteras [95,103]. Es necesario llevar a cabo estudios con mayor profundidad para clarificar estos aspectos. Los mecanismos de recristalización dinámica que operan a velocidades de deformación de impacto nunca han sido investigados. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 37 2.7. Aleaciones de magnesio procesadas mediante colada por inyección a alta presión. La colada por inyección a alta presión (“high pressure die casting”, HPDC) es una de las técnicas más ampliamente extendidas para la fabricación de componentes de las aleaciones de Mg para aplicaciones automovilísticas debido a su bajo coste y a que permite fabricar grandes volúmenes de productos semi-terminados [10]. Las aleaciones AZ91 (Mg-9%pAl-1%pZn) y AM60 (Mg-6%pAl-0.5%pMn) son las que se utilizan habitualmente para fabricar piezas mediante moldeo por inyección; la primera posee excelentes propiedades de colabilidad y resistencia y la segunda está dotada de una ductilidad y una capacidad de absorción de energía extraordinarias. Muchos estudios se han centrado en relacionar la microestructura con el comportamiento mecánico de las aleaciones de Mg obtenidas mediante HPDC, en especial las de las series AZ y AM, a bajas velocidades de deformación [104-113]. Las propiedades mecánicas de las piezas de Mg obtenidas mediante HPDC son menos predecibles que las de piezas procesadas mediante procesos de conformado ya que parámetros microestructurales clave como la distribución de segundas fases y su fracción de volumen, el tamaño de grano y, la forma y distribución de los poros depende fuertemente de las condiciones de solidificación [114-115], las cuales cambian durante el procesado y son, de este modo, difíciles de reproducir y controlar. Para que los componentes de Mg obtenidos mediante moldeo por inyección se puedan utilizar de forma masiva en aplicaciones estructurales en la industria del automóvil, éstos deben cumplir estrictos requisitos de absorción de energía frente a condiciones de impacto. Sin embargo, el comportamiento mecánico a alta velocidad de deformación de las aleaciones de Mg moldeadas por inyección no se PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 38 conoce con certeza [116]. 2.8. Objetivo de la investigación. El comportamiento de las aleaciones de Mg a altas velocidades de deformación (rango dinámico (~103 s-1)) aún no ha sido investigado en profundidad. En particular, la influencia de la temperatura y de la textura en los mecanismos de deformación y de recristalización, en las propiedades mecánicas y en las anisotropías del límite elástico y de la tensión máxima todavía no ha sido suficientemente estudiado. Los estudios en condiciones dinámicas llevados a cabo hasta la fecha han versado principalmente sobre aleaciones de Mg de las series AZ y AM, procesadas por extrusión o por colada, y los ensayos se han realizado predominantemente en compresión [56-,57,58,60,63,66,117- 123]. Estos han concluido que la ductilidad aumenta con la velocidad de deformación [60,118-,119,120,121] debido a que en condiciones dinámicas es mayor la sensibilidad a la velocidad de deformación. Asimismo, se ha observado que el maclado contribuye significativamente a la deformación inclusive a altas temperaturas, a las cuales es mayoritariamente suprimido a velocidades de deformación cuasi-estáticas [56-57,58,117]. También se ha mostrado que la tensión de flujo, o tensión máxima, en ensayos a alta velocidad de deformación es muy poco sensible a las variaciones de temperatura [56,57]. Solamente un estudio ha analizado el comportamiento mecánico dinámico de una aleación AZ31 laminada mediante ensayos de compresión a temperatura ambiente [66]. El propósito de esta tesis doctoral es contribuir a clarificar tanto el comportamiento mecánico, como los mecanismos de deformación y de recristalización de las aleaciones de Mg a velocidades de impacto. En PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 39 particular, se ha estudiado una chapa laminada de una aleación comercial (Mg-3%pAl-1%pZn, AZ31), una barra extruida de una aleación con tierras raras (Mg-1%pMn-1%pNd, MN11) y varias piezas de las aleaciones comerciales Mg-9%pAl-1%pZn (AZ91) y Mg-6%pAl-0.5%pMn (AM60) obtenidas mediante colada por inyección. Los resultados de este trabajo de investigación se han publicado en 3 artículos en revistas internacionales de SCI (Science Citation Index), los cuales constituyen el cuerpo de este escrito. A continuación se describe el contenido científico de cada uno de ellos. En el primer artículo (Ulacia I., N.V Dudamell, F. Gálvez, S. Yi, M.T. Pérez-Prado, I. Hurtado. Acta Materialia 58 (2010), 2988-2998) se ha estudiado el comportamiento mecánico en compresión a lo largo de RD y ND de una lámina de la aleación AZ31 a alta velocidad de deformación (~103 s-1) y en un amplio rango de temperaturas (de 25 ºC a 400 ºC) y éste se ha comparado con el observado tanto en estudios previos realizados a alta velocidad en tracción a lo largo de RD y TD como a bajas velocidades de deformación en tracción y en compresión a lo largo de RD, TD, y ND. Con este propósito se han realizado dos tipos de ensayos de compresión en muestras extraídas de una lámina de la aleación AZ31: por una parte, en una barra Hopkinson a temperaturas entre 25 ºC y 400 ºC a ~103 s-1 y, por otra, en una máquina convencional de ensayos Instron a 25 ºC, 200 ºC y 400 ºC y a velocidades de deformación de 5x10-4 s-1, 5x10-3 s-1 y 5x10-2 s-1. La anisotropía tanto del límite elástico como de la tensión máxima fuera del plano (esto es, comparando los valores correspondientes a ensayos de compresión realizados a lo largo de RD y ND) y la asimetría tensión- compresión se han evaluado en función de la temperatura. Estos datos han permitido investigar los mecanismos de deformación y recristalización predominantes y analizar la variación de la CRSS con la temperatura a alta velocidad de deformación. PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 40 En el segundo artículo de investigación (N.V. Dudamell, I. Ulacia, F. Gálvez, S. Yi, J. Bohlen, D. Letzig, I. Hurtado, M.T Pérez-Prado. Acta Materialia 59 (2011), 6949-6962) se ha estudiado la evolución de la microestructura y de la textura de una chapa de aleación de Mg AZ31, laminada y recocida, durante deformación dinámica a temperatura ambiente en función de la deformación y de la orientación de la carga aplicada. En particular, las condiciones analizadas incluyen, por una parte, compresión tanto a lo largo de la dirección de laminación (RD) como en la dirección normal (ND) y, por otra, tensión a lo largo de RD. Se realizaron ensayos en una barra Hopkinson controlando la deformación máxima alcanzada con el fin de evaluar la actividad de los distintos mecanismos de deformación y recristalización a medida que aumenta la deformación. La microestructura de las muestras, antes y después de deformar, fue caracterizada exhaustivamente mediante microscopía óptica, mediante difracción de electrones retrodispersados (“electron backscatter difraction” EBSD) y mediante difracción de neutrones. Se ha prestado especial atención a la actividad de diferentes modos de maclado y a la evolución de la distribución de la desorientación de los límites de grano, los cuales dan información valiosa sobre la presencia de mecanismos de restauración. La evolución microestructural a temperatura ambiente de esta aleación laminada de magnesio AZ31 a altas velocidades de deformación se ha comparado con aquéllas que se producen a velocidades de deformación cuasiestáticas. En el tercer artículo de investigación (N.V. Dudamell, I. Ulacia, F. Gálvez, S. Yi, J. Bohlen, D. Letzig, I. Hurtado, M.T Pérez-Prado. et al. Materials Science and Engineering A 532 (2012), 528-535), se han investigado los mecanismos de DRX que predominan durante la deformación a alta temperatura y a velocidades de deformación dinámicas en una chapa laminada de la aleación AZ31 en función del modo de carga y de la orientación relativa entre el eje de carga y el eje C del agregado PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 41 policristalino. Con este propósito se ha examinado la evolución de la microestructura, la textura y la distribución de la desorientación de las fronteras de grano con la deformación a 250ºC en una chapa laminada de AZ31 comprimida a lo largo de las direcciones de laminación (RD) y normal (ND) y también ensayada en tensión a lo largo de la RD. La microestructura de las muestras, antes y después de deformar, fue caracterizada exhaustivamente mediante microscopía óptica, mediante difracción de electrones retrodispersados (EBSD) y mediante difracción de neutrones. El comportamiento mecánico, la evolución de la microestructura y la textura y los mecanismos de DRX se han comparado con los que prevalecen a velocidades de deformación cuasi- estáticas a temperaturas similares. Además, en el apartado 4 se presentan dos estudios complementarios a los incluidos en los artículos descritos anteriormente. En el primero (apartado 4.1) se ha analizado el comportamiento mecánico, a temperaturas comprendidas entre 25ºC y 400ºC, tanto a velocidades cuasi-estáticas como dinámicas, de una barra extruida de una aleación de última generación de Mg con adición de tierras raras (Mg-1%pMn- 1%pNd (MN11)). Los ensayos se han realizado a lo largo de la dirección paralela al eje de extrusión y de una perpendicular a esta con la finalidad de evaluar la anisotropía mecánica. La evolución de la microestructura y de la textura se caracterizaron exhaustivamente mediante microscopía óptica y mediante difracción de electrones retrodispersados (EBSD) durante la deformación con el propósito de elucidar los mecanismos de deformación y recristalización predominantes. Los resultados se comparan con los obtenidos en la aleación comercial AZ31. Finalmente, se presenta un segundo estudio complementario (sección 4.2.) en el que se describe el comportamiento mecánico de las aleaciones comerciales de Mg moldeadas por inyección AZ91D y AM60B PARTE II. INTRODUCCIÓN. ALEACIONES DE MAGNESIO. 42 cuando se deforman a en el régimen dinámico en un amplio intervalo de temperaturas, en tensión y en compresión. Se ha estudiado la influencia de la velocidad de deformación en el límite elástico, en la tensión máxima, y en la elongación a fractura. Además, se describe la variabilidad de las propiedades mecánicas. Se ha realizado una comparación entre las propiedades obtenidas a alta y a baja velocidad de deformación. PART III. RESEARCH PAPERS 43 PART III RESEARCH PAPERS Important Note: Figures and tables are numbered as 1, 2, 3... inside each research paper, but they are labeled in the list of figures and tables with their corresponding article number, following the notation: chapter.article.figure(or table). For example, the figure 3.1.4 corresponds to the fourth figure of the first research paper located in chapter 3. Similarly, the table 3.2.1 corresponds to the first table of the second research paper located in chapter 3. PART III. RESEARCH PAPERS 44 3.1. Mechanical behavior and microstructural evolution of a Mg AZ31 sheet at dynamic strain rates. Acta Materialia 58 (2010) 2988-2998 I. Ulaciaa, N.V. Dudamellb, F. Gálvezc, S. Yid, M.T. Pérez-Pradob, I. Hurtadoa a Mondragon Goi Eskola Politeknikoa, Mondragon Unibertsitatea, 20500 Mondragón, Spain b Madrid Institute for Advanced Studies in Materials, IMDEA Materials, C/Profesor Aranguren s/n, 28040 Madrid, Spain c ETS Ingenieros de Caminos, Universidad Politécnica de Madrid, 28040 Madrid, Spain d GKSS Research Center, 21502 Geesthacht, Germany Received 10 December 2009; received in revised form 19 January 2010; accepted 19 January 2010 Available online 17 February 2010 PART III. RESEARCH PAPERS 45 PART III. RESEARCH PAPERS 46 PART III. RESEARCH PAPERS 47 PART III. RESEARCH PAPERS 48 PART III. RESEARCH PAPERS 49 PART III. RESEARCH PAPERS 50 PART III. RESEARCH PAPERS 51 PART III. RESEARCH PAPERS 52 PART III. RESEARCH PAPERS 53 PART III. RESEARCH PAPERS 54 PART III. RESEARCH PAPERS 55 PART III. RESEARCH PAPERS 56 3.2. Twinning and grain subdivision during dynamic deformation of Mg AZ31 sheet alloy at room temperature Acta Materialia 59 (2011) 6949-6962 N.V. Dudamella, I. Ulaciab, F. Gálvezc, S. Yid, J. Bohlend, D. Letzigd, I. Hurtadob, M.T. Pérez-Pradoa, a Madrid Institute for Advanced Studies in Materials, IMDEA Materials, C/Profesor Aranguren s/n, 28040 Madrid, Spain b Mondragon Goi Eskola Politeknikoa, Mondragon Unibertsitatea, 20500 Mondragón, Spain c ETS Ingenieros de Caminos, Universidad Politécnica de Madrid, 28040 Madrid, Spain d Magnesium Innovation Centre. Helmholtz-Zentrum Geesthacht, 21502 Geesthacht, Germany Received 1 June 2011; received in revised form 19 July 2011; accepted 21 July 2011 Available online 29 August 2011 PART III. RESEARCH PAPERS 57 PART III. RESEARCH PAPERS 58 PART III. RESEARCH PAPERS 59 PART III. RESEARCH PAPERS 60 PART III. RESEARCH PAPERS 61 PART III. RESEARCH PAPERS 62 PART III. RESEARCH PAPERS 63 PART III. RESEARCH PAPERS 64 PART III. RESEARCH PAPERS 65 PART III. RESEARCH PAPERS 66 PART III. RESEARCH PAPERS 67 PART III. RESEARCH PAPERS 68 PART III. RESEARCH PAPERS 69 PART III. RESEARCH PAPERS 70 PART III. RESEARCH PAPERS 71 3.3. Influence of texture on the recrystallization mechanisms in an AZ31 Mg sheet alloy at dynamic rates Materials Science and Engineering A 532 (2012) 528-535 N.V. Dudamella, I. Ulaciab, F. Gálvezc, S. Yid, J. Bohlend, D. Letzigd, I. Hurtadob, M.T. Pérez-Pradoa, a IMDEA Materials Institute, C/Profesor Aranguren s/n, 28040 Madrid, Spain b Mondragon Goi Eskola Politeknikoa, Mondragon Unibertsitatea, 20500 Mondragón, Spain c ETS Ingenieros de Caminos, Universidad Politécnica de Madrid, 28040 Madrid, Spain d Magnesium Innovation Centre. Helmholtz-Zentrum Geesthacht, 21502 Geesthacht, Germany Received 19 August 2011; received in revised form 2 November 2011; accepted 3 November 2011 Available online 11 November 2011 PART III. RESEARCH PAPERS 72 PART III. RESEARCH PAPERS 73 PART III. RESEARCH PAPERS 74 PART III. RESEARCH PAPERS 75 PART III. RESEARCH PAPERS 76 PART III. RESEARCH PAPERS 77 PART III. RESEARCH PAPERS 78 PART III. RESEARCH PAPERS 79 PART IV. COMPLEMENTARY STUDIES 80 PART IV COMPLEMENTARY STUDIES PART IV. COMPLEMENTARY STUDIES 81 4.1. COMPLEMENTARY STUDY I PART IV. COMPLEMENTARY STUDIES 82 4.1. Complementary study I. Mechanical behavior at quasi-static and dynamic rates of a magnesium alloy containing neodymium. 4.1.1. Introduction. In this chapter the mechanical behavior of an extruded Mg-1%wtMn- 1%wtNd (MN11) alloy during deformation at temperatures ranging from 25°C to 400°C at both quasi-static and dynamic rates is analyzed. Tests are performed along two perpendicular directions in order to evaluate the mechanical anisotropy. The evolution of the microstructure and the texture during deformation is investigated with the aim of elucidating the predominant deformation and recrystallization mechanisms. The behavior of this MN11 alloy is compared throughout the text to that of conventional Mg alloys, such as AZ31. 4.1.2. Materials and Experimental Procedure. 4.1.2.1. Material: initial microstructure. The material under study is an extruded bar of alloy MN11 (Mg- 1%wtMn-0.94%wtNd) fabricated in the Magnesium Innovation Center (MagIC) of the Helmholtz Zemtrum Geesthacht (Germany). The alloy was first gravity cast in order to produce billets for extrusion which were machined up to a diameter of 93 mm. Secondly and before extrusion, the billets were homogenized at 350 ºC during 15 h. Then, indirect extrusion was performed at 300 ºC to fabricate round bars of 17 mm of diameter, which corresponds to an extrusion ratio of 1:30. The extrusion speed used was of 10 m/min. The initial microstructure of the MN11 extruded bar is formed by equiaxed grains with average sizes of 10 µm and a very weak texture. Further considerations about the MN11 alloy can be found in [95]. PART IV. COMPLEMENTARY STUDIES 83 4.1.2.2. Mechanical testing. High strain rate tests ( 1310~ s ) were carried out in compression with the loading axis parallel (PL) and perpendicular (PP) to the extrusion axis (EA) and in tension parallel to the EA. Compression tests were performed in 3x3x4.5 mm3 prisms with the long axis parallel and perpendicular to the extrusion direction. Tensile tests along ED were performed in dog-bone shaped specimens with a gage length of 12 mm and a width of 4 mm. A Split Hopkinson Pressure Bar (SHPB) machine, provided with a radiant furnace and a high speed camera, was used to carry out these compression dynamic tests. Two nickel superalloy (RENE-41) bars, 1 m in length and 19.3 (incident bar) and 10 (output bar) mm in diameter were machined for this purpose. The compression tests were performed at temperatures ranging from 25ºC (RT) up to 400ºC, at intervals of 50 ºC, until failure and also up to several intermediate strains in order to observe the microstructure and texture evolution of the MN11 alloy with increasing strain level and temperature. With that purpose, the specimens were dynamically loaded to a predefined strain level using a stop-ring technique which restricts the movement of the incident bar (pushed by the impact of the striker bar or projectile) toward the sample. Ring stoppers of specific thickness were machined out of F522 steel. These rings were quenched and tempered in order to have a maximum control of the strain level achieved. A Split Hopkinson Tension Bar (SHTB) machine, provided with the same radiant furnace and the same high speed camera, was used to carry out tensile dynamic tests. The incident bar used in tensile tests consisted of two parts. The first one, made of steel with 20 mm of diameter and 4100 mm of length and upon which the projectile slide inside the barrel. The second part of the incident bar is located outside the barrel and it is made of RENE-41 with 19.3 mm of diameter and 3850 mm of length. PART IV. COMPLEMENTARY STUDIES 84 Both parts were attached with a threaded link. Output bar was made of RENE-41 with 19.3 mm of diameter and 3850 mm of length. In tension the MN11 alloy was tested until failure at room temperature (RT), 200 ºC and 300ºC. Tests were not carried out at 400ºC because it was observed that at this temperature samples crept, i.e. they deformed before the application of an external load. Quasi-static tests at room temperature (RT), 200ºC and 400ºC and at strain rates of 5x10-4 s-1, 5x10-3 s-1, and 5x10-2 s-1 were also carried out in compression until failure with the loading axis parallel and perpendicular to the ED in order to compare the variations in the microstructural evolution with the strain rate. Tests at 400ºC were stopped once a true strain of 0.4 was reached. All quasi-static tests were performed in a conventional universal testing machine (Instron) using the same geometries for the compression specimens described above. Two tests were carried out for each condition. 4.1.2.3. Microstructure examination. Optical microscopy (OM) was utilized for microstructural examination of selected samples. Surface preparation for OM included grinding with increasingly finer SiC papers, several diamond polishing steps, and surface finishing using a colloidal silica solution. The specimens prepared were additionally chemically etched during 15 s with a solution of 50 ml of ethanol, 0.5 g of picric acid, 0.5 ml of acetic acid, and 1 ml of distilled water. The microstructure and the microtexture of the samples compressed parallel to the EA and tensile tested along the EA were, additionally, analyzed by electron backscatter diffraction (EBSD) using the TSL- OIMTM software in a Zeiss Ultra 55TM FEG-SEM. Sample preparation for EBSD included grinding with 4000 SiC paper, mechanical polishing PART IV. COMPLEMENTARY STUDIES 85 with a 0.05 µm silica suspension and final electro-chemical polishing for 90 s at 33 V using the AC2TM commercial electrolyte at -20ºC approximately. The microstructure is represented by EBSD orientation maps showing the orientation of the EA and the texture by pole figures recalculated from the EBSD orientation data. Additionally, the grain size of various samples was calculated by the linear intercept method in the EBSD orientation maps using only high angle boundaries (misorientations higher than 15°). The macrotexture of some samples was also measured by the Schulz reflection method in a Philips X´pert-Pro Panalytical X-ray diffractometer furnished with a PW3050/60 goniometer, located at the CAI X-ray Diffraction of the Complutense University in Madrid, Spain. The radiation used was β-filtered Cu Kα. The surface area examined was about 2mm2. The polar angle ranged from 0° to 75° in steps of 3°. Irradiation time at each step was 2 s. The measured incomplete pole figures were corrected for background and defocusing using the Philips X´pert software. The orientation distribution function (ODF) and the complete pole figures were calculated using the MTex software [124]. Sample preparation for texture measurement included grinding with increasingly finer SiC papers, whose grit size ranged from 320 to 2000. 4.1.3. Results and Discussion. 4.1.3.1. As-extruded and heat treated material. Figure 4.1 illustrates the microstructure and the texture of the as- extruded MN11 bar. The grain structure is homogeneous and grains are mostly equiaxed, with an average size of 10 µm. The texture, as described previously [95], is significantly weaker than that characteristic of wrought conventional Mg alloys, such as AZ31, but not random. There is a slight tendency for the extrusion axes to cluster PART IV. COMPLEMENTARY STUDIES 86 around directions tilted away from the 0110 - 0211 boundary of the inverse stereographic triangle. This texture has been termed “RE texture” [86-99]. Figure 4.2 shows the effect of heat treatments on the microstructure and the texture of the MN11 alloy. In particular, the material was treated at 200°C (Figs. 4.2 a,b) and 400°C (Figs. 4.2 c,d) for 1 h, as this is the time it takes approximately to stabilize the temperature before mechanical tests. In both cases the grain size remains equal to approximately 10 µm. The texture after annealing is still weak and the tilt angle between the extrusion axes and the 0110 - 0211 boundary of the inverse stereographic triangle seems to increase slightly. 4.1.3.2. Mechanical behavior at quasi-static rates. The stress-strain curves corresponding to the MN11 alloy tested in compression at room temperature, 200°C, and 400°C under quasi-static rates (5x10-4 s-1, 5x10-3 s-1, and 5x10-2 s-1) are depicted in Figs. 4.3 a-f. Tests were performed both along a direction parallel to the EA (PL samples) (Figs. 4.3 a,c,e) as well as in a direction perpendicular to it (PP samples) (Figs. 4.3 b,d,f). Two tests were carried out for each temperature and both stress-strain curves are plotted in Fig. 4.3. Table 4.1 summarizes the yield stresses corresponding to the different tests. A) Room temperature. The “concave-up” shape of the curves at room temperature suggests that tensile twinning plays a major role during the first stages of deformation [33]. It is remarkable that, despite the weakness of the initial texture of the MN11 alloy, these curves resemble those corresponding to conventional rolled Mg alloys compressed along the rolling direction (RD) or to conventional extruded alloys compressed PART IV. COMPLEMENTARY STUDIES 87 along the EA, where twinning plays a major role at strains smaller than 0.06. Since the texture of the initial material is weak, it can be unambiguously stated that basal slip will also have a contribution to strain. (a) {0001} {1010}- {1011}- {1012}- {1013}- {1120}- EA (b) (100)(1100)(1100) (0001)(0001) (1210)(1210) (c) Figure 4.1. Microstructure and texture of the as-extruded MN11 bar. (a) Micrograph obtained by optical microscopy, (b) X-ray pole figures and (c) X-ray inverse pole figure showing the orientation of the extrusion axis (EA). The plane of observation is perpendicular to the extrusion axis (EA). 2 4 6 8 10 12 14 16 18 20 PART IV. COMPLEMENTARY STUDIES 88 (a) (b) (c) (d) Figure 4.2. Effect of heat treatment on the microstructure and the texture: (a,b) 200 ºC, 1h; (c,d) 400 ºC, 1h. Both samples where heated at 10ºC/min up to the corresponding heat treatment temperature. (100) (100) (1100)(1100) (0001)(0001) (1210)(1210) (1100)(1100) (0001)(0001) (1210)(1210) 2 4 6 8 10 12 14 16 18 20 PART IV. COMPLEMENTARY STUDIES 89 Figure 4.3. Stress-strain curves corresponding to the MN11 alloy deformed in compression at room temperature, 200 ºC and 400 ºC at: (a) 5x10-4 s-1 along the EA, (b) 5x10-4 s-1 perpendicular to the EA, (c) 5x10-3 s-1 along the EA, (d) 5x10-3 s-1 perpendicular to the EA, (e) 5x10-2 s-1 along the EA, (f) 5x10-2 s-1 perpendicular to the EA, (g) 103 s-1 along the EA, and (h) 103 s-1 perpendicular to the EA. 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True StrainRT 200 ºC 400 ºC a) b) c) d) e) f) g) h) 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 50 100 150 200 250 300 0 0.05 0.1 0.15 0.2 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 50 100 150 200 250 0 0.02 0.04 0.06 0.08 0.1 0.12 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 50 100 150 200 250 300 0 0.02 0.04 0.06 0.08 0.1 0.12 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Tr ue S tr es s (M Pa ) True Strain 50 100 150 200 250 0 0.05 0.1 0.15 0.2 0.25 0.3 PART IV. COMPLEMENTARY STUDIES 90 Table 4.1. Average yield stress values (in MPa). (a) Quasi-static deformation; (b) dynamic deformation. (a) (b) Significant differences exist between PL and PP tests, particularly at the two lowest strain rates. In particular, in PL tests the strain hardening is larger and the ductility is smaller. These observations may be explained taking into account that, due to the RE texture present in the as- received MN11 alloy, the incidence of twinning is smaller in PP tests. Figure 4.4 illustrates two inverse pole figures showing the orientation of the compression axis in two different PP tests (i.e., it shows the orientation of two radial directions, and , of the extruded bar). First, it can be seen that the orientation of the compression axis is not the same for all radial directions. This would explain the scatter in ductility values observed in PP tests (figure 4.5). Second, the fraction of orientations in the region close to the 0110 - 0211 boundary of the stereographic triangles is smaller than in the PL case (figure 4.1 c). This effect is more pronounced in the direction. Thus, the likelihood of twinning in PP tests is smaller than in PL ones. This is consistent with the lower strain hardening rate observed in the first stages of deformation in PP tests, which is usually attributed to slip/twin and Compression 5x10-4 s-1 5x10-3 s-1 5x10-2 s-1 RT 200ºC 400ºC RT 200ºC 400ºC RT 200ºC 400ºC PL 93 81 25 75 87 35 97 82 56 PP 95 91 22 95.5 85 36 92 88 43 Compression 103 s-1 Tension 103 s-1 RT 200ºC 400ºC RT 200ºC 300ºC PL 100 99 56 134 134 134 PP 104 86 66 PART IV. COMPLEMENTARY STUDIES 91 twin/twin interactions [125]. Consequently, a higher activation of basal slip would justify the higher ductility observed. The differences between PL and PP test are less noticeable at 5x10-2 s-1, probably because, as the strain rate increases, twinning is enhanced and it takes place in grains with lower Schmid factors, and thus the texture differences become less significant. Figure 4.4. Inverse pole figures showing the orientation of two radial directions of the as-received extruded bar of MN11. The two directions have been labeled and . Both are perpendicular to the EA. 2 4 6 8 10 (0001) (1100) - (1210 )- PART IV. COMPLEMENTARY STUDIES 92 0 50 100 150 200 250 300 350 400 0 100 200 300 400 500 5x10-4 s-1  m ax (M Pa ) Temperature (ºC) 0 0.1 0.2 0.3 >0.4 0 100 200 300 400 500 5x10-4 s-1  f Temperature (ºC) 0 0.1 0.2 0.3 >0.4 0 100 200 300 400 500 5x10-3 s-1  f Temperature (ºC) 0 50 100 150 200 250 300 350 400 0 100 200 300 400 500 5x10-3 s-1  m ax (M Pa ) Temperature (ºC) 0 0.1 0.2 0.3 >0.4 0 100 200 300 400 500 5x10-2 s-1  f Temperature (ºC) 0 50 100 150 200 250 300 350 400 0 100 200 300 400 500 5x10-2 s-1  m ax (M Pa ) Temperature (ºC) 0 50 100 150 200 250 300 350 400 0 100 200 300 400 500 103 s-1  m ax (M Pa ) Temperature (ºC)Parallel to the EA Perpendicular to the EA a) b) c) d) e) f) g) h) 0 0.1 0.2 0.3 0.4 0 100 200 300 400 500  f Temperature (ºC) 103 s-1 Figure 4.5. Evolution of the strain to failure ( f ) and the maximum flow stress ( max ) with temperature at different strain rates: a), b) 5x10-4 s-1; c),d) 5x10-3 s-1; e),f) 5x10-2 s-1, g),h) 103 s-1. PART IV. COMPLEMENTARY STUDIES 93 B) Elevated temperature. The variation with temperature of the strain to failure ( f ) and the maximum strength ( max ) corresponding to all the quasi-static tests are summarized in figure 4.5. As expected, max decreases with increasing temperature. No significant scatter in max values was observed in the two tests carried out at each temperature and strain rate. Figures 4.3 and 4.5 suggest that the MN11 alloy presents many manifestations of dynamic strain aging (DSA) [126] at strain rates of 5x10-4 s-1 and 5x10-3 s-1 and at 200°C (named hereafter “DSA” conditions). First, serrations are visible in the stress-strain curves (figures 4.3a-d). Second, the strain rate sensitivity, m, calculated at a strain of 0.1, adopts negative values (Table 4.2). Third, the strain to failure goes through a minimum (Fig. 4.5). Fourth, a larger strain hardening rate is observed at the lowest strain rate (compare Figs. 4.3a and 4.3c). This increased strain hardening rate cannot be related to a higher incidence of twinning, as this mechanism is less frequent as strain rate decreases. The appearance in the MN11 alloy of all the above DSA characteristics reveals a strong interaction between Nd atoms and dislocations. DSA has been recently reported in Mg alloys with Ce and Gd additions at similar temperatures [97,100]. In these studies, however, not all the DSA characteristics were observed (for example, a decrease in ductility was not reported and serrations were occasionally not visible) and this was attributed to a masking effect of twinning and dynamic recrystallization. In agreement with these reports, DSA characteristics in the MN11 alloy are, in general, slightly less pronounced in PL tests than in PP tests. For example, serrations of larger amplitude and a more pronounced ductility decrease can be noted in figures 4.3b than in PART IV. COMPLEMENTARY STUDIES 94 figures 4.3a. Also, a close look at the two curves at 200°C of figures 4.3d reveals that the amplitude of serrations is smaller in the curve with a more “concave-up” shape (and, thus, where twinning plays a more important role). At 5x10-2 s-1 and 200°C the signs of DSA disappear, probably because the waiting time of dislocations at obstacles decreases and solute atoms are no longer able to diffuse fast enough to form atmospheres around them. Compared to the tests at room temperature at the same strain rate (5x10-2 s-1), differences in PL and PP tests become more noticeable because the overall incidence of twinning decreases with temperature and, thus, the existing texture differences have an impact in the activation of this mechanism. Table 4.2. Strain rate sensitivity (m) values corresponding to the different quasi-static testing conditions investigated. Note: In the PP tests, the stress values utilized for the calculation of m are the average values between the two tests performed at each temperature. At 400°C the shape of the stress-strain curves is “concave-down” due to the predominance of crystallographic slip. f increases dramatically and tests were stopped at a strain of 0.4. The yield strength increases with increasing strain rate. 4.1.3.3. Mechanical behavior at dynamic rates. The stress-strain curves corresponding to the MN11 alloy tested in compression at room temperature, 200°C, and 400°C under dynamic conditions (103 s-1), both parallel and perpendicular to the EA, are RT 200ºC 400ºC 13 14 105 105   sx sx 12 13 105 105   sx sx 13 14 105 105   sx sx 12 13 105 105   sx sx 13 14 105 105   sx sx 12 13 105 105   sx sx PL 0.05 0.015 -0.10 -0.08 0.20 0.20 PP 0.07 0.03 0 -0.06 0.18 0.18 PART IV. COMPLEMENTARY STUDIES 95 depicted in figures 4.3g and 4.3h. The values of the stress at a strain of 0.005 (0.005) are summarized in Table 4.1. In dynamic tests it is difficult to measure the true yield stress at the nominal strain rate, since in the early stages of deformation the strain rate increases gradually. Thus, 0.005 is used in the present study as a best approximation to this material property. The shape of the compression stress-strain curves have still a “concave- up” shape, but this shape is significantly less pronounced than at lower strain rates. This is surprising, as it is known that tensile twinning is enhanced in Mg alloys deformed under impact loading at temperatures up to 400°C [56-58,117]. The strain to failure increases slightly with temperature from about 0.27 at RT up to 0.33 at 400°C. The maximum flow stress ( max ) values decrease with increasing temperature. No significant differences were found between the ductility and the maximum flow stress in PL and PP tests at these high strain rates (figures 4.5g and 4.5h). At RT and 200°C the strain to failure in compression is equal or higher under impact loading than at lower strain rates. A yield point effect is observed in the dynamic compression tests at 400°C, probably caused by the migration of solute atoms to dislocation cores during the temperature stabilization treatment prior to testing. Figure 4.6 illustrates the tension stress-strain curves corresponding to tests carried out under dynamic conditions at RT, 200 °C and 300°C. The shape of the curves is now “concave-down”, consistent with the predominance of crystallographic slip during the first stages of deformation. A yield point effect is observed during compression at 400°C and during tension at 300°C, revealing, again, the strong interaction between RE atoms and dislocations. PART IV. COMPLEMENTARY STUDIES 96 0 50 100 150 200 250 300 350 400 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 RT 200 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain Figure 4.6. Stress-strain curves corresponding to the MN11 alloy deformed in tension parallel to the extrusion axis (EA) at high strain rate at room temperature, 200 ºC and 300 ºC. 4.1.3.4. Microstructural evolution under quasi-static and dynamic conditions. Figures 4.7, 4.8 and 4.9 depict the microtexture of the MN11 alloy deformed in compression at 5x10-3 s-1 at RT, 200 °C and 400°C (PL samples). At the two lowest temperatures the c-axes tend to align with the EA. This texture is consistent with the occurrence of tensile twinning and basal slip [127]. The appearance of the six-fold symmetry in the prismatic pole figures reveals further the large incidence of twinning and suggests that, in most grains, only one twin variant or a single twin variant pair is active. Similar observations have been reported for an AZ31 rolled sheet deformed in compression along RD [128,129]. The operation of only one twin variant pair promotes the formation of parallel twin boundaries and rapid twin growth [129]. The activation of tensile twinning is also evidenced by the presence of tensile PART IV. COMPLEMENTARY STUDIES 97 twin boundaries in Figs. 4.7b and 4.8b. The definition of the six-fold symmetry is less clear at 200 °C, as twinning is less frequent (figure 4.8c). Compression and double twins are not frequent at RT but their incidence increases notably at 200 °C. This is in agreement with previous reports of a decrease in the CRSS of these twinning modes as temperature raises [32]. Finally, a second intensity maximum is also evident at the center of the  3110 pole figure at RT and 200 ºC, suggesting the simultaneous activation of pyramidal slip. No evidence of dynamic recrystallization has been found at 200 °C. At 400 °C the microstructure is completely recrystallized and a very weak texture is obtained. Table 4.3 compares the values of the average grain size measured for each condition (dM), without taking into account twin boundaries, and a theoretical grain size value (dT) calculated from the original grain size (10 µm) taking into account the strain attained. It must be noted that all grain size values are calculated and/or measured in a plane perpendicular to the extrusion axis. At RT and 200°C dM is slightly lower than dT, revealing the occurrence of grain subdivision as a consequence of dislocation interactions. At 400°C dM is half the value of dT, consistent with a significant reduction of grain size due to DRX. PART IV. COMPLEMENTARY STUDIES 98 (c) (d) (a) (b) Figure 4.7. Microtexture of the MN11 alloy tested at low strain rate (5x10-3 s-1), at room temperature in compression along the EA up to a strain of 0.12 (fracture) (a) EBSD inverse pole figure map showing the orientation of the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 - - - - - PART IV. COMPLEMENTARY STUDIES 99 (c) (d) (a) (b) Figure 4.8. Microtexture of the MN11 alloy tested at low strain rate (5x10-3 s-1) at 200 ºC in compression along the EA up to a strain of 0.11 (fracture) (a) EBSD inverse pole figure map in the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 - - - - - PART IV. COMPLEMENTARY STUDIES 100 (c) (d) (a) (b) Figure 4.9. Microtexture of the MN11 alloy tested at low strain rate (5x10-3 s-1) at 400 ºC in compression along the EA up to a strain of 0.6 (test stopped) (a) EBSD inverse pole figure map in the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 - - - - - PART IV. COMPLEMENTARY STUDIES 101 Figures 4.10 and 4.11 illustrate the evolution of the microtexture of the MN11 alloy (PL sample) with increasing strain during high strain rate (103 s-1) compression at RT and at 400°C. The microstructural development at room temperature is similar to that observed under quasi-static conditions. After a true strain of 0.095, the c-axes tend to align with the EA (Fig. 4.10c), consistent with the activation of tensile twinning and basal slip [127]. This texture becomes more intense with increasing strain (Fig. 4.10f). Simultaneously, a six-fold symmetry develops in the prismatic pole figures with increasing strain (figures 4.10f), suggesting that tensile twinning continues to operate up to true strains as high as 0.15. The activation of tensile twinning is also evidenced by the presence of tensile twin boundaries in Figs. 4.10b and 10e. A texture intensity maximum appears also at the center of the  3110 pole figure, revealing the operation of pyramidal slip. At 400°C the six maxima developed with increasing strain in the prismatic pole figures are more poorly defined at 400 °C (figure 4.11f) and compression and double twins are more frequent at 400 °C than at RT (figures 4.11b and 4.11e); finally, at a strain of 0.16 a small number of DRX grains are located at grain or twin boundaries (figures 4.11d), but their volume fraction is too small to have a significant impact on the texture development. Table 4.3 reveals, again, that, under dynamic conditions, the measured grain sizes (dM) are smaller than the calculated ones (dT) and, furthermore, that dM decreases with strain at all the temperatures investigated. This is consistent with the occurrence of grain subdivision processes due to dislocation interactions. Grain subdivision has been previously reported in an AZ31 alloy under quasi-static and dynamic conditions. Grain subdivision was observed to be favored when dislocations with high stacking fault energy (such as prismatic ones) carried most of the strain. In the MN11 alloy, grain subdivision might be favored by the interaction between basal and pyramidal dislocations. PART IV. COMPLEMENTARY STUDIES 102 Table 4.3. Theoretical grain size (dT) and measured grain size (dM) for MN11 samples tested in compression and in tension at high and low strain rate at different temperatures. The grain sizes are measured and calculated along a plane perpendicular to the extrusion axis. (*fracture, *test stopped) Compression Tension 5x10-3 s-1 103 s-1 103 s-1 RT 200ºC 400ºC RT 200ºC 400ºC RT 200ºC 300ºC 0.12* 0.11* 0.60* 0.095* 0.15* 0.10* 0.16* 0.11* 0.16* 0.25* 0.28* 0.4* dT 10.5 10.5 18 10.5 11 10.5 11 10.5 11 9 9 8.5 dM 7.5 8 9 7.5 6.5 8 6.5 8 7 6.5 5 3 PART IV. COMPLEMENTARY STUDIES 103 (a) (b) (d) (e) (c) (f) Figure 4.10. Evolution of the microtexture of the MN11 alloy during compression at high strain rate (103 s-1) along the EA at RT with increasing strain. (a) ε= 0.095; EBSD inverse pole figure map showing the orientation of the EA; (b) ε= 0.095; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) ε= 0.095; Pole figures and inverse pole figure showing the orientation of the EA. (d) ε= 0.15; EBSD inverse pole figure map showing the orientation of the EA; (e) ε= 0.15; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (f) ε= 0.15; Pole figures and inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. - - - - - 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 PART IV. COMPLEMENTARY STUDIES 104 (a) (b) (d) (e) (c) (f) Figure 4.11. Evolution of the microtexture of the MN11 alloy during compression at high strain rate (103 s-1) along the EA at 400 ºC with increasing strain. (a) ε= 0.11; EBSD inverse pole figure map showing the orientation of the EA; (b) ε= 0.11; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) ε= 0.11; Pole figures and inverse pole figures showing the orientation of the EA. (d) ε= 0.16; EBSD inverse pole figure map showing the orientation of the EA; (e) ε= 0.16; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (f) ε= 0.16; Pole figures and inverse pole figures showing the orientation of the EA. The plane shown is perpendicular to the EA. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 - - - - - - - - - - PART IV. COMPLEMENTARY STUDIES 105 Figures 4.12 and 4.13 illustrate the evolution of the microtexture of the MN11 alloy deformed in tension under dynamic conditions at RT and at 300°C. At both temperatures c-axes tend to align preferentially with the prismatic 0110 pole figure (figures 4.12c and 4.13c). The c-axes are located along the 0110 - 0211 boundary, except in the vicinity of the 0211 pole. Calnan and Clews [127] have predicted that a Mg randomly oriented polycrystal, deformed in tension by basal slip and tensile twinning, should, given enough strain, develop a texture in which all c- axes are parallel to all prismatic directions. The clustering of c-axes around the 0110 pole in our MN11 alloy is consistent with the additional activation of pyramidal  0110 slip [127]. Given the as- received texture of the MN11 alloy, it is logical that the contribution of basal slip is higher in tension than in compression and that, therefore, the shape of the tensile curves is “concave-down”. Figure 4.14 shows the superposition of the inverse stereographic triangle (extrusion axis) corresponding to the as-extruded MN11 alloy and the schematic triangle showing contours of constant resolved shear stress, i.e., of constant Schmid factor for basal slip, reported by Calnan and Clews. The occurrence of grain subdivision at room temperature is evidenced by the lower value of dM versus dT (Table 4.3). Finally, at 300°C a significant volume fraction of small DRX grains can be observed to form mainly at grain boundaries, giving rise to a dramatic decrease in dM with respect to dT at this temperature (Table 4.3). PART IV. COMPLEMENTARY STUDIES 106 (c) (d) (d) (a) (b) Figure 4.12. Microtexture of the MN11 alloy tested in tension along the EA at high strain rate (103 s-1) at RT up to a strain of 0.25 (fracture). (a) EBSD inverse pole figure map showing the orientation of the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 - - - - - PART IV. COMPLEMENTARY STUDIES 107 (c) (d) (a) (b) Figure 4.13. Microtexture of the MN11 alloy tested in tension along the EA at high strain rate (103 s-1) at 300 ºC up to a strain of 0.40 (fracture). (a) EBSD inverse pole figure map showing the orientation of the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures.; (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 - - - - - PART IV. COMPLEMENTARY STUDIES 108 Figure 4.14. Superposition of the inverse stereographic triangle (extrusion axis) corresponding to the as-extruded MN11 alloy and the schematic triangle showing contours of constant resolved shear stress, i.e., of constant Schmid factor for basal slip, reported by Calnan and Clews [127]. 4.1.4. Conclusions. The mechanical behavior of an extruded Mg-based alloy containing Mn and Nd additions (Mg-1%Mn-1%Nd) was investigated at quasi-static and dynamic rates and at temperatures ranging from 25°C to 400°C. Tests were carried out along the extrusion axis and along a direction perpendicular to it. The evolution of the texture and of the nature of the grain boundaries during testing was examined using electron backscattered diffraction (EBSD). The main conclusions of the present work are the following:  The strong interaction between Nd atoms and dislocations is evidenced by the occurrence of dynamic strain aging (DSA) at 200°C at quasi-static rates and of a yield point phenomenon at 300°C and 400°C at dynamic rates. PART IV. COMPLEMENTARY STUDIES 109  The predominant deformation mechanisms under all the testing conditions investigated are basal slip and tensile twinning. The incidence of each mechanism varies significantly with temperature and strain rate.  The critical resolved shear strength of basal slip appears to be similar or perhaps higher than that of tensile twinning. This is consistent with a higher yield strength value in tension than in compression and with the pronounced “concave-up” shape of the stress-strain curves observed at quasi-static rates. The increase in the CRSS of basal slip might be attributed to the preferential location of Nd atoms along basal planes.  Discontinuous dynamic recrystallization takes place during quasi- static compression at 400°C, leading to a weak texture. During dynamic deformation a small fraction of DRX grains are formed at grain or twin boundaries at temperatures higher than 300°C in both tension and compression. Their volume fraction is too small to have a significant impact in the texture development. Grain subdivision by dislocation interaction is observed both under quasi-static and dynamic rates. PART IV. COMPLEMENTARY STUDIES 110 4.2. COMPLEMENTARY STUDY II PART IV. COMPLEMENTARY STUDIES 111 4.2. complementary study II. Dynamic deformation of high pressure die-cast Mg alloys. 4.2.1. Introduction. The aim of this work is to investigate the mechanical behavior and the operative deformation mechanisms of die-cast AZ91 and AM60 alloys at impact strain rates under a wide range of temperatures in tension and in compression. The influence of the strain rate on the yield strength, the maximum stress, the elongation to failure and the property variability will be described. A comparison between high strain and low strain properties will be established. 4.2.2. Materials and Experimental Procedure The alloys utilized for this study are AZ91D and AM60B. They were purchased in the form of ingots. The latter were melted and die-cast into dog-bone mechanical testing specimens of circular cross-section with a gage length of 12 mm and a gage diameter of 4 mm. The liquid metal gate velocity was 8-9 m/s and the solidification pressure 400-500 bar. The liquid alloy temperature was 630°C-670°C and the temperature of the mould 190°C-230°C. The microstructure of the die- cast materials was investigated by optical (OM) and scanning electron microscopy (SEM). Sample preparation for these two techniques included grinding with increasingly finer SiC papers, several diamond polishing steps, and surface finishing using a colloidal silica solution. The specimens prepared for OM were additionally chemically etched during 15 s with a solution of 50 ml of ethanol, 0.5 g of picric acid, 0.5 ml of acetic acid, and 1 ml of distilled water. The macrotextures of the die-cast AZ91 and AM60 alloys were measured by the Schulz reflection method in a Philips X´pert-Pro Panalytical X-ray diffractometer PART IV. COMPLEMENTARY STUDIES 112 furnished with a PW3050/60 goniometer. The radiation used was - filtered Cu K. The measured incomplete pole figures were corrected for background and defocusing using the Philips X´pert software. The orientation distribution function (ODF), the complete pole figures and the volume fraction of selected texture components were calculated using the MTex software [124]. Sample preparation for texture measurement included grinding with SiC papers of grit sizes ranged from 320 to 2000. High strain rate mechanical tests (~103 s-1) were carried out using a tensile split Hopkinson bar, equipped with a radiant furnace. Testing temperatures ranged from room temperature to 400°C, at intervals of 50°C. Low strain rate tests were performed at room temperature, 200°C and 400°C at 5x10-3 s-1 in a conventional electromechanical Instron testing machine. Two samples were tested for each condition. The heads of the dog-bone die-cast AZ91 and AM60 coupons were threaded in order to be able to screw them to the grips of the testing machines for tension tests. Cylinders of 6 mm in length were cut out of the gage length of the die-cast dog-bone specimens for the compression tests. 4.2.3. Results and Discussion Fig. 4.15 illustrates the microstructure and the texture of the as-die- cast AZ91 and AM60 alloys. The plane examined is perpendicular to the loading axis of the coupons. Microanalysis at the SEM revealed that the phases present in both alloys alloy are mainly -Mg (light areas in figure 4.15), and Al12Mg17 (dark areas in figure 4.15). A very small fraction of Mg-Al-Mn compounds were also detected. Both materials have a weak texture. The texture of die-cast materials is usually deemed to be completely random. However, here the volume fraction of material with PART IV. COMPLEMENTARY STUDIES 113 c-axes tilted more than 45° away from the loading axis (Vf>45°) in both the AM60 and AZ91 alloys is ~70%. Figure 4.15. Microstructure and macrotexture (X-ray) of the as-received die- cast Mg alloy: (a) AM60, (b) AZ91. The plane examined is perpendicular to the loading axis. Figure 4.16 summarizes the values of the yield (y) and maximum strengths (max), as well as of the strain to failure (f), corresponding to the tests performed in tension (T) and in compression (C) at dynamic and quasi-static rates in both alloys. A note must be made here to state that f in compression corresponds to the strain until the maximum flow stress is reached. In dynamic tests it is difficult to measure the true yield stress at the nominal strain rate, since in the early stages of deformation the strain rate increases gradually. Thus, the stress at a true strain of 0.005 (0.005) is used in the present study as a best approximation to this material property. The two die-cast alloys exhibit an excellent dynamic mechanical behavior. In general, the room temperature yield and maximum strengths are slightly higher in the dynamic tests than at quasi-static rates. With increasing temperature the values of these two magnitudes decrease, but the change is less PART IV. COMPLEMENTARY STUDIES 114 pronounced at high strain rate. Thus, the yield and maximum strengths at dynamic rates are significantly higher than those obtained at quasi- static rates at temperatures higher than 200°C. At 400°C and quasi- static rates none of the two alloys experience any work hardening under all the conditions investigated while at dynamic rates a significant amount of work hardening is present. In both alloys the strain to failure increases dramatically with temperature at low strain rates while it remains rather constant at high strain rates. In the AM60 alloy, in the dynamic range, f retains values close to 20% in compression and between 10 and 15% in tension. In the AZ91 alloy the dynamic strain to failure in compression remains in the range between 15% and 20% until 350°C. In tension, it varies between 7% at room temperature and 5% at 400°C. In summary, the two die-cast alloys have a good energy absorption capacity at dynamic rates within all the temperature range investigated. This is because the decrease in the yield and maximum strengths with temperature are not as pronounced as during quasi- static tests and the strain to failure does not change substantially. A rather large scatter has been found in the tensile ductility of the AM60 alloy in low strain rate tests (figure 4.16d). This is consistent with the results of earlier works [104], which have reported a similar variability of the tensile ductility in a die-cast AM50 alloy deformed in tension at temperatures ranging from 25°C to 120°C. A strong quantitative correlation between the tensile ductility and the area fraction of the porosity in the corresponding fracture surfaces was found. However, the ductility was not reported to be related to the pore volume fraction. Thus, higher ductility values are predicted to be present in tensile specimens with a smaller number of large gas pores and/or of clustered small pores. In [104] the extent of the variability was observed to decrease with temperature. The authors did not report the reason for this decrease but they put forward several possible explanations such as variations in the strain hardening rate with PART IV. COMPLEMENTARY STUDIES 115 temperature or plastic relaxation due to the activation of secondary slip systems at high temperatures. In the present work the tensile ductility shows great variability until temperatures as high as 400°C. The differences between the two studies might be due to the use of different solidification conditions, which might result in drastic variations in the porosity distribution. Additionally, no significant variability was found in the tensile ductility of the AZ91 alloy at quasi-static rates (figure 4.16h). It has been reported that the presence of a network of low melting point second phases (Al12Mg17) improves the casting properties of the material by suppressing microporosity. Since the AZ91 alloy has a higher alloying content, this effect should be enhanced in this material when compared to the AM60 alloy [130]. Finally, in the dynamic tests the scatter in the tensile ductility is significantly smaller. This might suggest that the formation of adiabatic shear bands, the precursors to failure in most metals at impact strain rates, is rather independent of the porosity distribution. Figure 4.17 compares the compression stress-strain curves corresponding to the die-cast AM60 (figures 4.17a and 4.17b) and AZ91 (figures 4.17c and 4.17d) specimens tested at dynamic (figures 4.17a and 4.17c) and quasi-static rates (figures 4.17b and 4.17d) and at temperatures ranging from 25°C to 400°C. In both alloys, the low strain rate curves at room temperature have a “concave-up” shape at the early stages of deformation, revealing the activation of tensile twinning and slip. As temperature increases and secondary slip systems become operative, the twinning activity decreases significantly and the shape of the curve changes to “concave-down”. This effect has been observed in numerous occasions. Under dynamic conditions, the curves retain the “concave-up” shape up to temperatures as high as 400°C in the AM60 alloy and 350°C in the AZ91 alloy. This is consistent with the enhancement of twinning at high temperatures observed earlier [56,57]. PART IV. COMPLEMENTARY STUDIES 116 0 100 200 300 400 500 600 0 5 10 15 20 25 30 0 100 200 300 400  0. 00 5 ,  m ax ( M Pa ) f (% ) Temperature (ºC) (a) 0 100 200 300 400 500 600 0 5 10 15 20 25 >30 0 100 200 300 400  y ,  m ax ( M Pa ) f (% ) Temperature (ºC) (b) (e) 0 100 200 300 400 500 600 0 5 10 15 20 25 30 0 100 200 300 400  0. 00 5 ,  m ax ( M Pa ) f (% ) Temperature (ºC) (d) 0 100 200 300 400 500 600 0 5 10 15 20 25 30 0 100 200 300 400  ys ,  m ax ( M Pa ) f (% ) Temperature (ºC) AM60B, =5x10-3 s-1,T. 0 100 200 300 400 500 600 0 5 10 15 20 25 >30 0 100 200 300 400  y ,  m ax ( M Pa ) f (% ) Temperature (ºC) (f) 0 100 200 300 400 500 600 0 5 10 15 20 25 30 0 100 200 300 400  0. 00 5 ,  m ax ( M Pa ) f (% ) Temperature (ºC) (g) 0 100 200 300 400 500 600 0 5 10 15 20 25 30 0 100 200 300 400  0. 00 5 ,  m ax ( M Pa )  f (% ) Temperature (ºC) (c) 0 100 200 300 400 500 600 0 5 10 15 20 25 30 0 100 200 300 400  y ,  m ax ( M Pa ) f (% ) Temperature (ºC) (h) Figure 4.16. Variation of the yield stress (0.005) and of the flow stress (max) with temperature during high (103 s-1, left column) and low (5x10-3 s-1, right column) strain rate tests in die-cast AM60B and AZ91B alloys: (a) AM60B, 103 s-1 in compression; (b) AM60B, 5x10-3 s-1 in compression; (c) AM60B, 103 s-1 in tension, (d) AM60B, 5x10-3 s-1 in tension, (e) AZ91D, 103 s-1 in compression; (f) AZ91D, 5x10-3 s-1 in compression; (g) AZ91D, 103 s-1 in tension, (h) AZ91D, 5x10-3 s-1 in tension.  y  máx  ys  máx  ys  máx  f PART IV. COMPLEMENTARY STUDIES 117 It must be noted that the shape of the tension stress-strain curves, both at dynamic and quasi-static rates, was always “concave-down” in both alloys (figure 4.18). Thus, the activity of tensile twinning seems to be higher in compression than in tension tests. This is consistent with the fact that the volume fraction of material with c-axes tilted more than 45° away from the loading axis is higher than the fraction of material with tilt angles smaller than 45° (70% vs. 30%), as twinning is a polar mechanism [45]. (a) (b) (c) (d) Figure 4.17. True stress-true strain curves corresponding high (103 s-1) and low (5x10-3 s-1) strain rate tests performed in compression at different temperatures in die-cast Mg alloys : (a) AM60B, 103 s-1, (b) AM60B, 5x10-3 s-1 (c) AZ91D, 103 s-1, (d) AZ91, 5x10-3 s-1 . 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 300 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain AM60B, 103 s-1, C 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 RT 200 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain AM60B, 5x10-3 s-1, C 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 350 ºC Tr ue S tr es s (M Pa ) True Strain AZ91D, 103 s-1, C 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 RT 200 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain AZ91D, 5x10-3 s-1, C PART IV. COMPLEMENTARY STUDIES 118 (a) (b) Figure 4.18. True stress-true strain curves corresponding high strain rate (103 s-1) tests performed in compression at different temperatures in die-cast Mg alloys : (a) AM60B (b) AZ91D. 4.2.4. Conclusions. The dynamic behavior of the die-cast AM60 and AZ91 alloys has been investigated in a wide range of temperatures and compared to the quasi-static behavior. The main conclusions of the present work are the following:  The texture of the die-cast alloys is not random and this results in an anisotropic mechanical behavior.  It has been found that both alloys possess very good energy absorption capacity, since the yield and maximum strengths do not decrease with increasing temperature as dramatically as at quasi-static rates and the strain to failure values remain fairly constant. 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60 B, (103 s-1), T RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 300 ºC 350 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91D, (103 s-1), T RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 300 ºC 350 ºC Tr ue S tr es s (M Pa ) True Strain PART IV. COMPLEMENTARY STUDIES 119  The absence of any significant variability in the tensile ductility at dynamic rates is attributed to the independence of the adiabatic shear failure mechanism of the porosity distribution.  Tensile twinning is observed to be enhanced at dynamic rates in compression even at temperatures as high as 400°C. PART V. GENERAL DISCUSSION 120 PART V GENERAL DISCUSSION PART V. GENERAL DISCUSSION 121 5. GENERAL DISCUSSION 5.1. Strain rate sensitivity of the yield stress in the Mg AZ31 alloy. It is well known that the crystalline structure influences the strain rate response of metallic materials [131]. For body centered cubic (bcc) metals the strain rate sensitivity is in the yield stress dependence, whereas for face centered cubic (fcc) metals it is mainly in the strain hardening [132] . In the latter, the yield stress is virtually independent of the strain rate and of temperature. Hexagonal close packed metals are known to follow either bcc or fcc behavior. In particular, magnesium alloys with a random texture are reported to follow the fcc pattern [132]. Our results presented in the first research paper (Ulacia I., N.V Dudamell, et al. Acta Materialia 58 (2010), 2988-2998) show that the strain rate sensitivity of the yield stress of the AZ31 sheet alloy may follow the fcc or the bcc patterns depending on the initial relative orientation between the tensile/compression axis and the c-axes. A close look at the room temperature data of Figures 4a, 4c, and 4e of the mentioned paper reveals that the yield strength does indeed not vary with strain rate when testing the material in compression along RD (RD-C) due to the preferential activation of twinning, whose CRSS is strain rate insensitive, or when testing the material in compression along ND (ND-C), where pyramidal slip and compression twinning play a major role (fcc pattern). However, there is a significant variation of the yield strength with strain rate when testing the material in tension along in-plane directions (RD-T and TD-T), where prismatic and basal slip predominate (bcc pattern). PART V. GENERAL DISCUSSION 122 5.2. Strain rate dependence of the critical resolved shear stress of non-basal slip systems and of tensile twinning in the Mg AZ31 alloy. It has been found that the yield stress ( y ) is related to the CRSS of the various operating deformation mechanisms as 2 1  mkdm oy  , where m is the Taylor factor, o is the CRSS for the operative slip systems and k is the microstructural shear stress intensity characterizing the average grain boundary resistance to plasticity spreading between the grains [132]. Thus, analyzing the variation of y with strain rate and temperature (figures 4a,c, and e of the first paper) allows inferring roughly the sensitivity of the CRSS of the operative slip systems to these two parameters. Our results suggest, firstly, that at room temperature the strain rate dependency of the CRSS of prismatic slip systems is higher than that of pyramidal systems. Consistently with earlier findings, twinning is seen to be strain rate independent. Secondly, it appears that the temperature dependence of the CRSS of non-basal slip systems is significantly less pronounced at dynamic strain rates than at quasi-static strain rates. Barnett [32] has calculated the evolution of the CRSS of several Mg deformation mechanisms as a function of the Zener-Hollomon parameter (Z) (table 2.2). The model shows good correlation with experimental data for values of Z between 109 s-1 and 1014 s-1 , for temperatures between approximately 180°C and 380°C at 10-3 s-1 and higher than 380°C at 103 s-1. Although this model does not cover well the strain rate/temperature conditions utilized in the present work, it does point toward a higher variation of the CRSS of non-basal slip systems at lower strain rates. Thirdly, at dynamic strain rates the CRSS for {10-12} twinning seems to remain constant with temperature up to 400°C. PART V. GENERAL DISCUSSION 123 Figure 5a of the first research paper (Ulacia I., N.V Dudamell, et al. Acta Materialia 58 (2010), 2988-2998) shows that under dynamic conditions the yield stress tension-compression asymmetry and the out-of-plane yield stress anisotropy remain high up to 400°C. This suggests that the CRSS of non-basal slip systems remains higher than that of twinning, even at these high temperatures. Also, the in-plane yield stress anisotropy does not change with temperature, revealing that the CRSS of non-basal slip remains higher than that of basal slip up to temperatures as high as 400°C. Both observations are in full agreement with Barnett´s model prediction [32]. 5.3. Influence of strain rate on the twinning activity. The strain rate has also an important influence on the twinning activity. It is now well known that twinning is enhanced in Mg alloys during dynamic deformation to such an extent that it is present even at very high temperatures (400 ºC) [56-58,117], at which it is absent at low strain rates. A profuse twinning activity at dynamic rates has also been reported in other hexagonal close-packed metals, such as Ti and Zr [133-135]. The results of the second research paper presented (N.V. Dudamell et al. Acta Materialia 59 (2011), 6949-6962) additionally reveals that the influence of the strain rate on the AZ31 alloy consists of the enhancement of extension twinning, but that compression and double twinning are not only not enhanced, but are even hindered under dynamic conditions. In particular, it has been observed that the strain rate enhances extension twin propagation in grains that are favorably oriented for this mechanism to operate, and that it promotes the nucleation of extension twins even in grains that would not be favorably oriented for extension twinning at low strain rates. PART V. GENERAL DISCUSSION 124 The enhancement of extension twinning at high strain rates can be rationalized as follows. Arguments based on variations in the CRSS for twinning with strain rate must be ruled out, as it is well known that this parameter is strain rate and temperature insensitive [32,45]. The choice of twin systems was classically based on the criterion of minimum shear put forward by Jaswon and Dove [136] (see also Ref. [45]). This model assumes that the atoms in the twinned area reach their final positions exclusively by pure shear, which excludes atomic translations (shuffles). The twinning shears associated to extension and compression twinning in Mg (s) are -0.1289 and 0.1377, respectively [45]. These values are very similar, but usually extension twins are abundant and thick, while the compression twins are often observed only in localized regions. Furthermore, this criterion does not explain why extension twinning is preferentially enhanced at high strain rates, since shear is not diffusion assisted. A more sophisticated model for determining the operative twin systems was proposed by Bilby and Crocker [137] (see also Ref. [45]). This model considers twinning as a combination of shear plus individual atomic shuffles. The latter are dependent on both temperature and strain rate [138]. Bilby and Crocker suggested that, besides having a small shear, the operative twin mode should require only simple shuffles. In their theory, the “simplicity” of the atomic shuffles is related to a positive integer parameter  q associated with each twinning mode, which is the number of lattice invariant planes ( 1K ) of spacing d traversed by the primitive lattice vector in the shear direction ( 2 ) [45]. The lower the value of q , the simpler the shuffles involved. For extension twinning 4q , while for  1110 contraction twinning 8q [45]. Thus, contraction twinning involves more complicated atom shuffling. Since atomic translations (shuffling) are activated thermally, an increase in the strain rate would hinder these PART V. GENERAL DISCUSSION 125 atomic movements – and even more so if they are complex. Therefore contraction twinning is expected to be less favored than extension twinning at high strain rates. This is consistent with our observations. The enhancement of extension twinning under dynamic conditions is especially dramatic during compression along the ND, where the polarity of twinning is reversed in a rather large number of grains. It is remarkable that an inversion of the polarity of tensile twinning is preferred to an enhancement of compression twinning in grains in which the c-axes are being compressed. The reason for this surprising phenomenon is not entirely clear. Twin polarity is explained in the following way. The shear associated to a specific twinning system ( s ) is calculated as the ratio between the modulus of the corresponding twinning dislocation ( TDb ) and the step height, h ( h bs TD ) [45]. For twinning modes with q higher than 2, q bs TD2 ~ [45]. TDb , and therefore s , may also be expressed as a function of the c/a ratio ( ). For  2110 twins,     2 1 2 3 32   q bs TD . The shear thus changes dramatically with the c/a ratio. In Mg,  is equal to 1.624 and the resulting shear is negative. This makes this twinning mode a tension twin [22,45]. Within this framework, the observation of  2110 twins during c-axis compression at dynamic rates implies that the associated twinning shear adopts a positive value. This could only be achieved if the c/a ratio were to increase by approximately 5% during elastic deformation, which is not observed experimentally, or if a different mechanism (other than the propagation of the above-mentioned twinning dislocation) would dominate  2110 twinning under the mentioned conditions. Some controversy does indeed exist regarding the fundamental mechanism responsible for  2110 twinning in Mg alloys [139-140]. PART V. GENERAL DISCUSSION 126 5.4. Influence of the strain rate on grain subdivision. The strain rate has also influence on dynamic recovery. It is well known that the dislocations generated during shock loading in high stacking fault energy materials arrange in ordered cell structures whose size and misorientation angle depend on the applied pressure and pulse time [131]. Since dislocation reorganization into cell structures requires some thermal activation, usually cells become better developed as the strain rate decreases. In low stacking fault energy metals, however, dislocation rearrangement during dynamic deformation is more sluggish and thus homogeneous dislocation distributions develop. It has been observed that significant dislocation rearrangement to form new boundaries subdividing the original grains takes place when the AZ31 alloy is tested both at quasi-static and dynamic rates (see Figs. 3, 4 and 8 of the research paper N.V. Dudamell et al. Acta Materialia 59 (2011), 6949-6962). However, the misorientation of these GNBs can reach very high angles in AZ31, in some cases exceeding 15º, whereas the typical misorientation of a cell boundary is rarely higher than 1 or 2º. Such highly misoriented GNBs have been observed only in rare cases, such as when deforming severely ( 1 ) pure Al by equal channel angular pressing [141]. It is remarkable that dislocation rearrangement can take place to such an extent at dynamic rates, where thermal activation is very limited. The explanation for this phenomenon might be related to the very high values of the stacking fault energies corresponding to both ac  and prismatic 0110 dislocations in Mg alloys, which are activated under all the conditions investigated here, as described in the previous sections. These values are compared in Table 1 of section 3.2. PART V. GENERAL DISCUSSION 127 The phenomenon of grain subdivision is most pronounced in the sample deformed quasi-statically in tension along the RD, resulting in a large reduction of the grain size with increasing deformation (from 13 to 2.5 µm). This can be rationalized as follows. First, the predominant deformation mechanism is the slip of prismatic dislocations, which have high stacking fault energy [143]. Second, prismatic slip predominates during all the deformation stages, and thus carries a large amount of the total strain (not all, as basal slip also takes place to some extent). This is not the case of the compression test along the RD, where twinning carries quite a large portion of the strain during the first stages of deformation, and pyramidal ac  slip, also with a high stacking fault energy [143], becomes predominant only in the latter stages of deformation. This is not the case of the compression test along ND, since the total strain to failure is small (0.07). The dynamic fracture energy corresponding to the tension test along the RD is higher than that corresponding to the other two tests (see Section 3.1 of the research paper N.V. Dudamell et al. Acta Materialia 59 (2011), 6949- 6962)). This energy is presumably employed in the formation of a larger fraction of GNBs. Finally, grain subdivision is more pronounced in the test in tension along the RD at quasi-static rates than at dynamic rates due to the easier thermal activation in the former case. 5.5. Dynamic recrystallization at high strain rates. Crystallographic texture plays a key role on the recrystallization mechanisms of magnesium alloys. With the purpose of going into this matter in depth, the research paper entitled “Influence of texture on the recrystallization mechanisms in an AZ31 Mg sheet alloy at dynamic rates” (N.V. Dudamell et al. Materials Science and Engineering A 532 (2012), 528-535) was published. The results reveal that there is a higher resistance to DRX when testing dynamically in tension along the RD PART V. GENERAL DISCUSSION 128 than during compression along the RD and along the ND. These observations can be rationalized as follows. The influence of texture on DRX at quasi-static rates has been analyzed in a few earlier studies. In a pioneering work on the subject, Kaibyshev et al. [72] reported that the kinetics of DRX in an hot-pressed MA14 Mg rod (Mg-0.5wt%Zn-5%Zr), with a typical prismatic fiber texture and deformed in compression at 300°C and 2.8 x 10-3 s-1, were dependent on the crystallographic texture. In particular, they observed that if the compression axis was parallel or perpendicular to the basal planes, DRX took place readily and an almost fully recrystallized structure was obtained at moderate strains. However, when the compression axis was tilted 45° with respect to the basal planes, DRX took place more slowly. In the former, the reported operative slip systems at the strain at which recrystallization started are prismatic, pyramidal and basal. In the latter, only basal and prismatic slip are found to be active. Thus, Kaibyshev et al. [72] concluded that the kinetics of DRX were accelerated when all three main slip modes (basal, prismatic and pyramidal a ) are active. This, reportedly, is due to the fact that the operation of many slip modes favors the cross-slip of basal dislocations into non-basal planes. These non-basal dislocations have high stacking fault energies ( 50basalSFE mJ.m-2; 354prismaticSFE mJ.m-2; 452pyramidalSFE mJ.m-2 [142,143]) and, thus, can easily climb and arrange into new boundaries by the Friedel-Escaig mechanism. Del Valle et al. [75] also observed an enhancement of DRX in AZ31 samples oriented for multiple slip versus others initially oriented only for single (basal) slip. Finally, Barnett [28] reported a delay in DRX in an AZ31 sheet deformed in plane strain under conditions in which prismatic slip would predominate versus others in which ac  and basal slip are operative. PART V. GENERAL DISCUSSION 129 Our results at dynamic rates are consistent with the above observations at quasi-static rates and also support the fact that the kinetics of DRX is related to the operative deformation mechanisms. In particular, we observe a higher resistance to recrystallization when prismatic slip and basal slip operate (tension along the RD) than when a combination of ac  and basal slip are active (compression along the ND). This can be understood taking into account that pyramidal slip has more independent systems (5) than prismatic slip (2), and that the SFE of pyramidal dislocations is also significantly higher. Together, these two factors favor the occurrence of cross-slip and climb when pyramidal slip is active. We have also observed that the most favorable conditions for DRX at high strain rates are reached during compression along the RD, when extensive tensile twinning is followed by a combination of basal and pyramidal ac  slip. Under these conditions DRX takes place homogeneously throughout the microstructure, the final grain size is smaller and the strain hardening is lower than when compressing along the ND, where only basal and pyramidal ac  slip operate from the first stages of deformation. Thus, our results suggest that the operation of twinning during high temperature deformation at dynamic rates in the AZ31 alloy enhances DRX. The beneficial effect of twinning on DRX in Mg alloys as has been observed previously under quasi-static conditions [eg. 85,144-147] and it has been attributed to the large stresses accumulated close to twin boundaries, which promote the operation of multiple slip modes. Figure 1 of the research paper mentioned above reveals that DRX takes place more easily at quasi-static than at dynamic strain rates, resulting in a much more pronounced softening. This may be explained by the easiness of diffusion at low strain rates, as well as by the more pronounced decrease of the CRSS of ac  slip with temperature under such conditions. Thus, for example, at 250°C and 10-3 s-1, it is expected PART V. GENERAL DISCUSSION 130 that prismatic, ac  and basal slip operate, giving rise to the onset of recrystallization. However, at that same temperature and dynamic rates, the CRSS of pyramidal slip is still significantly higher than that of prismatic slip, and therefore only the latter and basal slip are active, resulting in a strong resistance to DRX, as is observed in the current study. 5.6. Influence of Rare Earth (RE) atoms in the incidence of basal slip and twinning at quasi-static rates. It is well accepted that, in weakly textured Mg alloys with a grain size similar to that of the MN11 studied here (~10 µm), the CRSS of the various slip and twinning systems increase following the order   pyramidalprismatictwinningbasal CRSSCRSSCRSSCRSS  2110 [32,44,47] and, therefore, at room temperature and quasi-static rates, these materials deform mainly by basal slip and, to a smaller extent, by tensile twinning. The incidence of non-basal systems is, comparatively, negligible. The corresponding stress-strain curves have either a “concave-down” shape or a very subtle “concave-up” shape, depending on the relative twinning activity [104,148]. Our results reveal that the predominant deformation mechanisms during room temperature quasi-static deformation of the MN11 alloy are also tensile twinning and basal slip. However, the pronounced “concave-up” shape of the stress-strain curves indicates that tensile twinning is the main contributor to strain during the first stages of deformation [149]. We propose that this dramatically enhanced activity of tensile twinning may be due to the fact that, in this MN11 alloy, the critical resolved shear stress of basal slip is higher than that of tensile twinning. The increase in the CRSS of basal slip could be caused by the preferential location of Nd atoms along basal slip planes. Indeed, we have observed that there is a strong interaction between Nd atoms and dislocations, evidenced by the observation of DSA at 200°C PART V. GENERAL DISCUSSION 131 at quasi-static rates and of yield point phenomena at 300°C and 400°C at dynamic rates. The strong interaction could be due to the large radius of the Nd atoms (181 pm), comparable to that of other potent Mg strengtheners such as Ce (182.5 pm) and Gd (180.2 pm) [97]. As the testing temperature increases to 200°C the activity of twinning decreases, as thermal activation promotes diffusion of the Nd atoms, resulting in a decrease of the CRSS for basal slip. Our results also reveal the activation of pyramidal slip during room temperature quasi-static deformation. This is evidenced, first, by the appearance of an intensity maximum at the center of the  3110 pole figure after compression straining, which suggests ac  slip is active and, second, by the occurrence of grain subdivision during deformation. The formation of high angle boundaries by the progressive increase of misorientation of dislocation arrays was observed to be favored when dislocations with high stacking fault energy (such as prismatic or pyramidal ones) carried most of the strain. Galiyev et al. [85] have reported that the creation of a 3D network of high angle grain boundaries in Mg alloys is favored by the mutual interaction of dislocations of two different Burgers vectors ( a and ac  ). The fact that grain subdivision takes place readily in the MN11 alloy, reveals, thus, the interaction between basal and pyramidal ac  dislocations and constitutes, therefore, further evidence of the activation of non- basal slip. In conventional, weakly textured, Mg alloys, deformed at room temperature and quasi-static rates, the contribution of non-basal slip systems to strain is very minor, since the CRSS of basal slip is at least an order of magnitude smaller than those of non-basal slip. However, an enhanced activation of non-basal slip has been reported in several Mg alloys containing RE elements [87-88,96-98]. The origin of this PART V. GENERAL DISCUSSION 132 increased activity is still not clear. Chino et al. [87-88,96] and Sandlöbes et al. [96] suggested that the stacking fault energies of certain types of dislocations changed by the addition of Ce and Y, respectively, thus altering the critical resolved shear stresses of a and ac  slip. Robson et al. [98] carried out VPSC modeling to simulate the behavior of a Mg–6Y–7Gd–0.5wt% Zr alloy and found that the best fit was obtained when the relative CRSS of prismatic slip was significantly reduced with respect to that of a conventional alloy such as AZ31. Our results are consistent with these studies, in that they suggest an increased activity of non-basal slip with the addition of Nd but suggest, instead, that this is due to an increase of the CRSS of basal slip. 5.7. Influence of Rare Earth (RE) atoms in the incidence of basal slip and twinning under dynamic deformations. The analysis of the mechanical behavior as well as of the microstructural evolution of this MN11 alloy deformed at high strain rates also reveals that the main deformation mechanisms are basal slip and tensile twinning. The additional contribution of pyramidal slip is also evidenced by the appearance of the texture intensity maxima at the center of the  3110 pole figure in the samples deformed in compression and by the clustering of orientations near the 0110 pole in the samples deformed in tension. The incidence of slip and twinning is different in compression and in tension tests, as evidenced by the “concave-up” shape of the compression stress-strain curves and the “concave-down” shape of the tension curves, which reveal that basal slip is a more important contributor to strain in tension and, in turn, twinning is enhanced in compression. This can be attributed to geometrical reasons as illustrated in figure 4.14, in which the inverse pole figure (extrusion PART V. GENERAL DISCUSSION 133 axis) of the as-received MN11 alloy has been superimposed to a schematic triangle showing the contours of constant resolved shear stress, i.e., constant Schmid factor, for basal slip. The contour corresponding to a Schmid factor of 0.5 denotes the orientations tilted 45° with respect to the 0110 - 0211 boundary and with respect to the 0001 pole. In compression, twinning would be favored in those grains in which the extrusion axis is parallel to the 0110 - 0211 boundary and the likelihood of twinning will gradually decrease until it becomes zero in grains tilted 45° from this boundary due to the twinning polarity. Figure 4.14 shows that a large fraction of grains have orientations within the area of the triangle where twinning is favored (darker blue areas). However, in tension, twinning is favored in those grains in which the extrusion axis is parallel to the 0001 boundary and, again, the likelihood of twinning decreases gradually until it becomes zero in grains tilted 45° from this boundary due to the twinning polarity. As can be seen in figure 4.14, twinning is thus not favored in tension in this MN11 alloy, even when the CRSS of basal slip would increase significantly. Table 4.1 reveals that the yield strength in tension at room temperature and 103 s-1 is significantly higher than the compression yield strength under the same temperature and strain rate conditions (134 MPa vs. 100 MPa). Since basal slip appears the to be the main contributor to strain in tension, this difference in yield strengths is, again, consistent with an increase of the CRSS of basal slip due to the addition of Nd. PART VI. CONCLUSIONS 134 PART VI CONCLUSIONS (CONCLUSIONES) PART VI. CONCLUSIONS 135 6. CONCLUSIONS This doctoral thesis constitutes a study of the mechanical properties as well as on the deformation and recrystallization mechanisms of the commercial magnesium alloy Mg- Mg-3%wtAl-1%wtZn (AZ31) at high strain rates. Additionally, the effect of rare earth alloying additions on the mechanical behavior and on the incidence of the different deformation mechanisms has been analyzed at high and quasi-static strain rates. Finally, this study has been extended to two commercial alloys, Mg-9%wtAl-1%wtZn (AZ91D) and Mg-6%wtAl-0.5%wtMn (AM60B), processed by high pressure die casting, as this is currently the most relevant processing technology for Mg alloys. The main conclusions from this research work are the following: 6.1. In the AZ31 alloy, at high strain rates, the yield stress asymmetry as well as the in-plane and out-of-plane yield stress anisotropies retain relatively high values even at the highest temperatures investigated (400°C). 6.2. The rate of decrease of the CRSS of non-basal slip systems with temperature is much smaller at high strain rates than at low strain rates in the AZ31 Mg alloy. The CRSS of non- basal systems is still significantly higher than that of twinning and of basal systems at 400°C. At room temperature, the strain rate dependency of the CRSS of prismatic slip systems seems to be higher than that of pyramidal slip systems. 6.3. Tensile twinning is enhanced at high strain rates and remains the predominant deformation mechanism at the early stages of deformation in compression tests along the RD of an AZ31 Mg sheet even at very high temperatures (400°C). Compression and double twinning, however, have PART VI. CONCLUSIONS 136 been found to be basically strain-rate insensitive. The choice of tensile twinning at high strain rates may be attributed to the simpler atom shuffles involved. 6.4. The polarity of twinning may be inverted at dynamic rates. In particular, 2110 twinning, which is well known to be a tension twin in the AZ31 alloy, has been observed to take place in grains in which the c-axes are compressed during dynamic compression along the normal direction. 6.5. Grain subdivision by the formation of geometrically necessary boundaries takes place readily in the AZ31 alloy during both quasi-static and dynamic deformation. The extent to which grain subdivision takes place depends on the relative orientation between the applied stress and the c-axes of the crystallites. This phenomenon is most pronounced in the sample tested in tension along RD since prismatic dislocations, which have high stacking fault energy, carry most of the strain up until failure. Grain subdivision is enhanced at low strain rates due to the easiness of thermal activation. 6.6. The DRX mechanisms and their kinetics depend on the operative deformation mechanisms and thus vary for different loading modes (tension, compression) as well as for different relative orientations between the loading axis and the c-axes of the grains. The highest resistance to dynamic recrystallization was found when the AZ31 Mg sheet was tested in tension along the RD, i.e., when prismatic slip predominates. Under these conditions, some recrystallized grains do appear at the latest stages of deformation (ε~0.20) as a result of rotational dynamic recrystallization (RDRX). DRX took place more readily under compression along the PART VI. CONCLUSIONS 137 RD, when twinning, ac  and basal slip operate. In this case, the main recrystallization mechanism is discontinuous dynamic recrystallization (DDRX). 6.7. The addition of rare earth elements (Nd) results in significant changes in the predominant deformation and recrystallization mechanisms. In particular, in the MN11 alloy, the critical resolved shear strength of basal slip increases to values similar or perhaps higher than that of tensile twinning. The increase in the CRSS of basal slip might be attributed to the preferential location of Nd atoms along basal planes. 6.8. The AZ91 and AM60 alloys, processed by high pressure die casting, possess good energy absorption capacity at high strain rates since the yield and maximum strengths do not decrease with increasing temperature as dramatically as at quasi-static rates and the strain to failure values remain fairly constant. The absence of any significant variability in the tensile ductility of these alloys at dynamic rates of, is attributed to the independence of the adiabatic shear failure mechanism of the porosity distribution. PART VI. CONCLUSIONES 138 6. CONCLUSIONES En esta tesis doctoral se han estudiado las propiedades mecánicas así como los mecanismos de deformación y recristalización de la aleación comercial de magnesio Mg-3%pAl-1%pZn (AZ31) a alta velocidad de deformación. Así mismo, se ha analizado el efecto de la adición de tierras raras en el comportamiento mecánico y en la actividad de diferentes mecanismos de deformación en condiciones dinámicas y cuasiestáticas. Finalmente, se ha investigado el comportamiento mecánico en condiciones dinámicas en dos aleaciones comerciales, Mg- 9%pAl-1%pZn (AZ91D) and Mg-6%pAl-0.5%pMn (AM60B), procesadas mediante colada por inyección a alta presión, la tecnología de procesado más importante para aleaciones de Mg. Las principales conclusiones de la presente tesis doctoral son las siguientes: 6.1. En la aleación AZ31, a altas velocidades de deformación, la asimetría en el límite elástico y las anisotropías del límite elástico en el plano y fuera del plano retienen valores relativamente altos incluso a las temperaturas más altas investigadas (400 ºC). 6.2. La tensión crítica de cizalla resuelta (CRSS) de los sistemas de deslizamiento no basales disminuye con la temperatura más lentamente a altas velocidades de deformación que a velocidades cuasi-estáticas para la aleación de Mg AZ31. Así, la CRSS de los sistemas no basales es aún significativamente mayor que la del maclado y la de los sistemas basales a 400ºC a velocidades de impacto. A temperatura ambiente, la dependencia de la CRSS de los sistemas de deslizamiento prismáticos con la velocidad de PART VI. CONCLUSIONES 139 deformación parece ser mayor que la de los sistemas de deslizamiento piramidales. 6.3. El maclado de extensión se favorece a altas velocidades de deformación y permanece como mecanismo de deformación predominante durante los primeros estadios de deformación en ensayos de compresión a lo largo de RD de una chapa laminada de aleación de Mg AZ31 incluso a temperaturas muy altas (400 ºC). Sin embargo la actividad del maclado de compresión y la del maclado secundario parecen no ser tan sensibles a la velocidad de deformación. La preferencia del maclado de extensión a altas velocidades de deformación se puede atribuir a que este mecanismo conlleva un menor reajuste atómico que los otros dos. 6.4. La polaridad del maclado de extensión se puede invertir a altas velocidades de deformación. En particular, éste se ha observado en algunos granos en los cuales se produce una deformación de compresión a lo largo del eje “c” durante ensayos de compresión dinámica a lo largo de la dirección normal de laminación para una chapa de aleación AZ31. 6.5. El proceso de subdivisión de granos debida a la formación de fronteras geométricamente necesarias tiene lugar en la aleación AZ31 deformada tanto en condiciones dinámicas como cuasiestáticas. El grado de subdivisión de los granos depende de la orientación relativa entre la carga aplicada y el eje “c” de los cristales. Así, este fenómeno es más pronunciado en muestras ensayadas en tensión a lo largo de la dirección de laminación (RD) debido a que el mecanismo de deformación predominante hasta la fractura es el deslizamiento prismático y a que las dislocaciones que se deslizan en planos prismáticos tienen alta energía de PART VI. CONCLUSIONES 140 falla de apilamiento. La subdivisión de grano se favorece a bajas velocidades de deformación debido a que, en estas condiciones se favorecen los procesos de deslizamiento cruzado, que se activan térmicamente. 6.6. Los mecanismos de recristalización dinámica (DRX) y su cinética dependen de los mecanismos de deformación operativos y, así, varían según sea el modo de carga (tensión, compresión) y la orientación relativa entre el eje de carga y el eje “c” de los granos. La mayor resistencia a la recristalización dinámica fue hallada cuando el material se ensayó en tracción a lo largo de RD, es decir, cuando predomina el deslizamiento prismático. En estas condiciones, algunos granos recristalizados aparecen en los últimos estadios de la deformación (ε~0.20) como resultado de la recristalización dinámica rotacional (RDRX). La recristalización dinámica se produce más fácilmente cuando la aleación AZ31 se deforma en compresión a lo largo de la dirección de laminación, es decir, cuando se activan el maclado, el deslizamiento y el deslizamiento basal. En este caso, el mecanismo de recristalización predominante es recristalización dinámica discontinua (DDRX). 6.7. La adición de elementos de tierras raras (Nd) da lugar a un cambio significativo de los mecanismos de deformación y de recristalización predominantes. En particular, en la aleación MN11, la tensión crítica de cizalla resuelta del deslizamiento basal aumenta hasta alcanzar valores comparables a los del maclado de extensión. El incremento en la CRSS del deslizamiento basal se puede atribuir a la ubicación preferencial de los átomos de Nd a lo largo de los planos basales. PART VI. CONCLUSIONES 141 6.8. Las aleaciones AZ91 y AM60, procesadas mediante colada por inyección a alta presión, poseen buena capacidad de absorción de energía a altas velocidades de deformación ya que el límite elástico y la tensión máxima no disminuyen cuando aumenta la temperatura tan drásticamente como a velocidades cuasi-estáticas y los valores de deformación a fractura permanecen prácticamente constantes. La poca variabilidad de la elongación a fractura de estas aleaciones a alta velocidad de deformación se puede atribuir a la independencia de los mecanismos de fallo de cizalla adiabática de la distribución de la porosidad. PART VII. FURTHER WORK 142 PART VII FURTHER WORK PART VII. FURTHER WORK 143 7. FURTHER WORK. This work has raised new questions and has motivated numerous opportunities for further research in magnesium alloys. Some areas of further work currently contemplated are listed below:  Study of the incidence of different twinning modes and of the occurrence of specific recrystallization mechanisms at dynamic rates in Mg as a function of alloy composition and grain size.  Analysis of the deformation and recrystallization mechanisms of Mg alloys with different amounts of selected rare earth elements under a wide range of temperatures and strain rates. Various alloys will be processed using different extrusion conditions including several extrusion ratios at selected temperatures. The dependency of the critical resolved shear stresses of basal slip and twinning on the processing conditions and on the alloy composition will be investigated.  Investigation of the relationship between the processing conditions, the microstructure and the mechanical behavior of high pressure die cast alloys. An attempt will be made to quantify mechanical properties such as the hardness and the yield stress as a function of the grain size, the distribution of the beta phase and the porosity distribution. PART VIII. REFERENCES 144 PARTE VIII REFERENCES PART VIII. REFERENCES 145 8. REFERENCES [1] UTTLEY C. “The elements: Magnesium”. Marshal Cavendish Corporation. Nueva York. (2000). Pp. 4. [2] EASTON M, BEER A, BARNETT M, DAVIES C, DUNLOP G, DURANDET Y, BLACKET S, HILDITCH T, BEGGS P. “Magnesium Alloy Applications in Automotive Structures” JOM 2008;60:57. [3] MORDIKE B.L., EBERT T. “Magnesium Alloy Applications in Automotive Structures” . Mater Sci Eng 2001;302:37. [4] DOEGE E, DRÖDER K., JANSSEN ST. “Deformation of Magnesium“. En: Magnesium Alloys and Technologies. Kainer KU [Ed.] Wiley-VCH. Geesthacht. 2003. (285 p.) Pp. 73. [5] BAMBERGER M, DEHM G. “Trends in the Development of New Mg Alloys”. Annu Rev Mater Res 2008; 38:505. [6] WATARAI H. “Trend of research and Development for Magnesium Alloys”. Science & Technology Trends. 2006;18:84 [7] MANI G., FELDMAN M.D., PATEL D., MAULI AGRAWAL C. “Coronary stents: A materials perspective”. Biomat. 2007; 28:1689 [8] AVEDESIAN, M. “Magnesium and Magnesium Alloys”. American Society for Metals. United States of America. 1999. (314 p.) Pp. 7-15, 258-263. [9] CALLISTER W., “Introducción a la Ciencia e Ingeniería de los Materiales” Vol.1. Ed. Reverté. Barcelona (2009). (459 p.) Pp. 37 [10] FRIEDRICH H E.; MORDIKE, B L (Eds.) “Magnesium Technology: metallurgy, design data, applications”. Ed. Springer. (2006), XXII. (677 p.) Pp. 146 [11] GEHRMANN R., FROMMERT M.M., GOTTSTEIN G. “Texture effects on plastic deformation of magnesium”. Mater Sci Eng 2005;395:338. [12] AL-SAMMAN T, GOTTSTEIN G. “Room temperature formability of a magnesium AZ31 alloy: Examining the role of texture on the deformation mechanisms”. Mater Sci Eng 2008;488:406. [13] KIM W.J., HONG S.I., KIM Y.S., MIN S.H., JEONG H.T., LEE JD. “Texture development and its effect on mechanical properties of an AZ61 Mg alloy fabricated by equal channel angular pressing”. Acta mater 2003; 51:3293 PART VIII. REFERENCES 146 [14] WANG Y.N., HUANG J.C. “Texture analysis in hexagonal materials”. Mater Chem and Phys 2003; 81:11 [15] GRAFF S. “Micromechanical modeling of the deformation of HCP Metals”. Ed. GKSS. Geesthacht (2008). (117p.) Pp. 1-3 [16] MAYAMA T, AIZAWA K, TADANO Y, KURODA M. “Influence of twinning deformation and lattice rotation on strength differential effect in polycrystalline pure magnesium with rolling texture” Comp. Mat. Sci 2009; 47: 448 [17] COULING S.L.,PASHAK J.F., STURKEY L. “Unique deformation and aging characteristics of certain magnesium-based alloys”. Trans ASM 1959;51:94. [18] KOCKS U.F., WESTLAKE D.G “The Importance of Twinning for the Ductility of CPH Polycrystals”. Trans AIME 1967;239:1107. [19] Kelley EW, Hosford WF. “Deformation characteristics of textured magnesium” Trans AIME 1968;242:654. [20] COURET A, CAILLARD D. “An in situ study of prismatic glide in magnesium—II. Microscopic activation parameters”. Acta metal 1985;33:1455. [21] CHIN GY, MAMMEL WL. “Competition among basal, prism, and pyramidal slip modes in HCP metals”. Metall Trans 1970;1:357. [22] YOO MH. “Slip, twinning and fracture of in hexagonal closed-packed metals” Metall Trans 1981;12A:409. [23] VAGARALI SS, LANGDON TG. “Deformation mechanisms in h.c.p. metals at elevated temperatures—I. Creep behavior of magnesium” Acta Metall 1981;29:1969. [24] ZELIN MG, YANG HS, VALIEV RZ, MUKHERJEE AK. “Interaction of high temperature deformation mechanisms in a magnesium alloy with mixed fine and coarse grains”. Metall Trans 1992;23:3135. [25] MUNROE N, TAN X. “Orientation dependence of slip and twinning in HCP metals”. Scripta Mater 1997;36:1383. [26] AGNEW SR, YOO MH, TOMÉ CN. “Application of texture simulation to understanding mechanical behavior of Mg and solid solution alloys containing Li or Y”. Acta mater 2001;49:4277. PART VIII. REFERENCES 147 [27] WATANABE H, TSUTSUI H, MUKAI T, KOHZU M, TANABE S, HIGASHI K. “Deformation mechanism in a coarse-grained Mg–Al–Zn alloy at elevated temperatures” Int J Plasticity 2001;17:387. [28] Barnett MR. “Influence of deformation conditions and texture on the high temperature flow stress of magnesium AZ31”. J Light Metals 2001;1:167. [29] AGNEW S.R., TOMÉ C.N., BROWN D.W., HOLDEN T.M., VOGEL S.C. “Study of slip mechanisms in a magnesium alloy by neutron diffraction and modeling”. Scripta Mater 2003; 48:1003. [30] GALIYEV A, SITDIKOV O, KAIBYSHEV R. “Deformation behavior and controlling mechanisms for plastic flow of magnesium and magnesium alloy”. Mater Trans 2003; 44:426. [31] KOIKE J, KOBAYASHI T, MUKAI T, WATANABE H, SUZUKI M, MARUYAMA K, HIGASHI K. “The activity of non-basal slip systems and dynamic recovery at room temperature in fine-grained AZ31B magnesium alloys” Acta mater 2003;51:2055. [32] BARNETT MR. “A Taylor model based description of the proof stress of magnesium AZ31 alloy during hot working,” Metall Mater Trans 2003; 34:1799. [33] BARNETT M.R., KESHAVARZ Z., BEER A.G., ATWELL D. “Influence of grain size on the compressive deformation of wrought Mg–3Al–1Zn”. Acta Mater 2004; 52:5093. [34] AGNEW SR, DUYGULU Ö. “Plastic anisotropy and the role of non-basal slip in magnesium alloy AZ31B”. Int J Plasticity 2005;21:1161. [35] DEL VALLE J.A., PÉREZ-PRADO M.T., RUANO O.A. “Deformation mechanisms responsible for the high ductility in a Mg AZ31 alloy analyzed by electron backscattered diffraction”. Metall Mater Trans 2005; 36:1427. [36] KESHAVARZ Z., BARNETT M.R. “EBSD analysis of deformation modes in Mg–3Al–1Zn”. Scripta mater 2006;55:915. [37] MEZA-GARCÍA E., DOBROŇ P., BOHLEN J., LETZIG D., CHMELIK F., LUKÁČ P., KAINER K.U. “Deformation mechanisms in an AZ31 cast magnesium PART VIII. REFERENCES 148 alloy as investigated by the acoustic emission technique”.Mater Sci Eng 2007; 462: 297. [38] BARNETT M.R. “Twinning and the ductility of magnesium alloys: Part I: “Tension” twins”. Mater Sci Eng 2007;464:1. [39] DEL VALLE J.A., RUANO O.A. “Separate contributions of texture and grain size on the creep mechanisms in a fine-grained magnesium alloy”. Acta mater 2007;55:455. [40] CHINO Y., KIMURA K., MABUCHI M. “Twinning behavior and deformation mechanisms of extruded AZ31 Mg alloy”. Mater Sci Eng 2008;486:481. [41] JAIN A., DUYGULU O., BROWN D.W., TOMÉ C.N., AGNEW S.R. “Grain size effects on the tensile properties and deformation mechanisms of a magnesium alloy, AZ31B, sheet”. Mater Sci Eng 2008;486:545. [42] HUTCHINSON B, BARNETT MR, GHADERI A, CIZEK P, SABIROV I. “Deformation modes and anisotropy in magnesium alloy AZ31”. Int J Mat Res 2009;100:556. [43] BALL E.A., PRANGNELL P.B. “Tensile-compressive yield asymmetries in high strength wrought magnesium alloys”. Scripta Metall Mater 1994;31:111. [44] LOU X.Y., LI M., BOGER R.K., AGNEW S.R., WAGONER R.H. “Hardening evolution of AZ31B Mg sheet”. Int J Plasticity 2007; 23:44. [45] CHRISTIAN J.W., MAHAJAN S. “Deformation Twinning”. Prog Mater Sci 1995;39:1. [46] BARNETT M.R. “Twinning and the ductility of magnesium alloys: Part II: “Contraction” twins”. Mater Sci Eng 2007;464:8. [47] CHAPUIS A., DRIVER J.H. “Temperature dependency of slip and twinning in plane strain compressed magnesium single crystals”. Acta Mater 2011; 59:1986. [48] HUTCHINSON B., BARNETT M.R. “Effective values of critical resolved shear stress for slip in polycrystalline magnesium and other hcp metals”. Scripta Mater 2010;63:737. PART VIII. REFERENCES 149 [49] WONSIEWICZ B.C., BACKOFEN W.A. “Plasticity of magnesium crystal”. Trans AIME. 1967; 239:1422. [50] HARTT W.H., REED-HILL R.E. “ Internal deformation and frature of second order [1011]-[1112] twins in magnesium”. Trans AIME. 1968; 242:1127. [51] HARTT W.H., REED-HILL R.E. “Irrational habit of second-order (1011)- (1012) twins in magnesium”. Trans AIME. 1967; 239:1511. [52] NAVE M.D., BARNETT M.R., “Microstructures and textures of pure magnesium deformed in plane-strain compression”. Scripta Mater 2004; 51:881. [53] BARNETT M.R., NAVE M.D., BETTLES C.J. “Deformation microstructures and textures of some cold rolled Mg alloys”. Mater Sci Eng 2004; 386:205. [54] CIZEK P., BARNETT M.D. “Characteristics of the contraction twins formed close to the fracture surface in Mg-3Al-1Zn alloy deformed in tension”. Scripta Mater 2008; 59:959. [55] KNEZEVIC M., LEVINSON A., HARRIS R., MISHRA R.K., DOHERTY R.D., KALIDINDI S.R. “Deformation twinning in AZ31: Influence on strain hardening and texture evolution”. Acta Mater 2010; 58:6230 [56] ISHIKAWA K, WATANABE H, MUKAI T. “High strain rate deformation behavior of an AZ91 magnesium alloy at elevated temperatures”. Mat Lett 2005;59:1511. [57] ISHIKAWA K., WATANABE H., MUKAI T. “High temperature compressive properties over a wide range of strain rates in an AZ31 magnesium alloy”. J Mat Sci 2005;40:1577. [58] WATANABE H, ISHIKAWA K. “Effect of texture on high temperature deformation behavior at high strain rates in a Mg–3Al–1Zn alloy”. Mater Sci Eng 2009;523:304. [59] ULACIA I., SALISBURY C.P., HURTADO I., WORSWICK M.J. “Tensile characterization and constitutive modeling of AZ31B magnesium alloy sheet over wide range of strain rates and temperatures” J Mater Proc Technol 2011; 211:830. PART VIII. REFERENCES 150 [60] EL-MAGD E, ABOURIDOUANE M. “Influence of Strain Rate and temperature on the compressive ductility of Al, Mg and Ti alloys”. J. Phys IV France. 2003;110:15. [61] MURR L.E., PIZAÑA C. “Dynamic recrystallization: The dynamic deformation regime”. Metall Trans A 2007; 38:2611 [62] ZOU D.L., ZHEN L., ZHU Y., XU C.Y., SHAO W.Z., PANG B.J. “Deformed microstructure and mechanical properties of AM60B magnesium alloy under hypervelocity impact at a velocity of 4 km s−1”. Mater Sci Eng 2010; 527:3323. [63] RITTEL D., WANG Z.G. “Thermo-mechanical aspects of adiabatic shear failure of AM50 and Ti6Al4V alloys”. Mech Mater 2008; 40: 629. [64] WAN G, WU BL, ZHAO YH, ZHANG YD, ESLING C. “Strain-rate sensitivity of textured Mg–3.0Al–1.0Zn alloy (AZ31) under impact deformation”. Scripta mater 2011; 65:461-464. [65] WU BL, WAN G, ZHANG YD, ESLING C. “Twinning characteristics in textured AZ31 alloy under impact loading along specified direction”. Mat Lett 2010; 64:636-639. [66] TUCKER M.T., HORSTEMEYER M.F., GULLETT P.M., EL KADIRI H., WHITTINGTON W.R. “Anisotropic effects on the strain rate dependence of a wrought magnesium alloy”. Scripta mater 2009;60:182. [67] STANFORD N., SOTOUDEH K., BATE P.S. “Deformation mechanisms and plastic anisotropy in magnesium alloy AZ31”. Acta Mater 2011; 59:4866. [68] KHAN A.S., PANDEY A., GNÄUPEL-HEROLD T., MISHRA R.K. “Mechanical response and texture evolution of AZ31 alloy at large strains for different strain rates and temperatures”. Int J Plas 2011; 27:688 [69] AL-SAMMAN T., LI X., CHOWDURY S.G. “Orientation dependent slip and twinning during compression and tension of strongly textured magnesium AZ31 alloy” Mater Sci Eng A 2010; 527:3450. [70] ION S.E., HUMPHREYS F.J., WHITE S.H. “Dynamic recrystallisation and the development of microstructure during the high temperature deformation of magnesium” Acta metall 1982; 30:1909. [71] KAIBYSHEV R., GALIYEV A.M., SOKOLOV B.K, “Effect of grain size on the plastic deformation and dynamic recrystallization of a magnesium alloy”.Phys Met Metallog 1994; 78:209. PART VIII. REFERENCES 151 [72] KAIBYSHEV R., SOKOLOV B., GALIYEV A. “The Influence of Crystallographic Texture on Dynamic Recrystallization”.Textures and Microstructures. 1999;32:47. [73] MYSHLYAEV M.M., MCQUEEN H.J., MWEMBELA A., KONOPLEVA E. “Twinning, dynamic recovery and recrystallization in hot worked Mg–Al– Zn alloy”. Mater Sci Eng 2002;337:121. [74] TAN J.C., TAN M.J. “Dynamic continuous recrystallization characteristics in two stage deformation of Mg–3Al–1Zn alloy sheet” Mater Sci Eng A 2003; 339:124 [75] DEL VALLE J.A., PÉREZ-PRADO M.T., RUANO O.A. “Texture evolution during large-strain hot rolling of the Mg AZ61 alloy ”. Mat Sci Eng 2003; 355:68. [76] BARNETT M.R., BEER A.G., ATWELL D., OUDIN A. “Influence of grain size on hot working stresses and microstructures in Mg–3Al–1Zn”. Scripta Mater 2004; 51:19 [77] YI S., ZAEFFERER S., BROKMEIER H.G. “Mechanical behavior and microstructural evolution of magnesium alloy AZ31 in tension at different temperatures” Mater Sci Eng A 2006; 424:275. [78] AL-SAMMAN T., GOTTSTEIN G. “Dynamic recrystallization during high temperature deformation of magnesium”.Mater Sci Eng 2008;490:411. [79] DEL VALLE J.A., RUANO O.A. “Influence of texture on dynamic recrystallization and deformation mechanisms in rolled or ECAPed AZ31 magnesium alloy”. Mater Sci Eng 2008;487:473. [80] MARTIN E., JONAS J.J. “Evolution of microstructure and microtexture during the hot deformation of Mg–3% Al”. Acta mater 2010; 58:4253. [81] SUN K., CHANG C.P., KAO P.W. “Microstructural study of strain localization in hot compressed Mg–3Al–1Zn alloy” . Mater Sci Eng A 2010; 527:7050. [82] LIU Z., BAI S., KANG S. “Low-temperature dynamic recrystallization occurring at a high deformation temperature during hot compression of twin-roll-cast Mg–5.51Zn–0.49Zr alloy”. Scripta mater 2009; 60:403. PART VIII. REFERENCES 152 [83] DOHERTY R.D., HUGHES D.A., HUMPHREYS F.J., JONAS J.J., JUUL JENSEN D., KASSNER M.E., KING W.E., MCNELLEY T.R., MCQUEEN H.J., ROLLETT A.D. “Current issues in recrystallization: a review” . Mater Sci Eng A 1997; 238:219. [84] HUMPHREYS F.J., HATHERLY M. “Recrystallization and related annealing phenomena”. Pergamon, 1995. [85] GALIYEV A., KAIBYSHEV R., GOTTSTEIN G. “Correlation of plastic deformation and dynamic recrystallization in magnesium alloy ZK60”. Acta mater 2001;49:1199. [86] BOHLEN J., NÜRNBERG M.R., SENN J.W., LETZIG D., AGNEW S. “The texture and anisotropy of magnesium–zinc–rare earth alloy sheets“. Acta Mater 2007; 55:2101. [87] CHINO Y., KADO M., MABUCHI M.“ Compressive deformation behavior at room temperature – 773 K in Mg–0.2 mass%(0.035at.%)Ce alloy“. Acta Mater 2008;56:387. [88] CHINO Y., KADO M., MABUCHI M. “Enhancement of tensile ductility and stretch formability of magnesium by addition of 0.2 wt%(0.035 at%)Ce”. Mater Sci Eng 2008;494:343. [89] MISHRA R.K., GUPTA A.K., RAMA RAO P., SACHDEV A.K., KUMAR A.M., LUO A.A.“Influence of cerium on the texture and ductility of magnesium extrusion”. Scripta mater 2008;59:562. [90] MACKENZIE L.W.F., PEKGULERYUZ M.O. “The recrystallization and texture of magnesium–zinc–cerium alloys” Scripta Mater 2008;59:665. [91] COTTAM R., ROBSON J., LORIMER G., DAVIS B. “Dynamic recrystallization of Mg and Mg–Y alloys: Crystallographic texture development”. Mater Sci Eng 2008;485:375. [92] STANFORD N., ATWELL D., BEER A., DAVIS C. AND BARNETT M.R. “Effect of microalloying with rare-earth elements on the texture of extruded magnesium-based alloys” Scripta Mater 2008; 59:772. [93] STANFORD N. “Micro-alloying Mg with Y, Ce, Gd and La for texture modification—A comparative study”. Mater Sci Eng 2010; 527:2669. [94] STANFORD N., ATWELL D., BARNETT R. “The effect of Gd on the recrystallisation, texture and deformation behavior of magnesium-based alloys”. Acta Mater 2010; 58:6773. PART VIII. REFERENCES 153 [95] BOHLEN J., YI S., LETZIG D., KAINER K.U. “Effect of rare earth elements on the microstructure and texture development in magnesium–manganese alloys during extrusion”. Mater Sci Eng A 2010; 527:7092. [96] SANDLÖBES S., ZAEFFERER S., SCHESTAKOW I., YI S., GONZÁLEZ- MARTÍNEZ R. “On the role of non-basal deformation mechanisms for the ductility of Mg and Mg–Y alloys”. Acta Mater 2011; 59:429. [97] JIANG L., JONAS J.J., MISHRA R. “Effect of dynamic strain aging on the appearance of the rare earth texture component in magnesium alloys”. Mater Sci Eng 2011;528:6596. [98] ROBSON J.D., TWIER A.M., LORIMER G.W., ROGERS P. “Effect of extrusion conditions on microstructure, texture, and yield asymmetry in Mg–6Y–7Gd–0.5 wt%Zr alloy”Mater Sci Eng 2011;528:7247 [99] AL-SAMMAN T., LI X. “Sheet texture modification in magnesium-based alloys by selective rare earth alloying” Mater Sci Eng 2011; 528:3809. [100] STANFORD N., SABIROV I., SHA G., LA FONTAINE A., RINGER S.P., BARNETT M.R. “Effect of Al and Gd solutes on the strain rate sensitivity of magnesium alloys. Metall Mater Trans 2010;41:734. [101] ZHU S.M., NIE J.F. “Serrated flow and tensile properties of a Mg–Y– Nd alloy”. Scripta Mater 2004;50:51. [102] AGNEW S.R., SENN J.W., HORTON J.A."Mg Sheet Metal Forming: Lessons Learned from deep drawing Li and Y Solid-Solution Alloys". JOM 2006;58:62. [103] STANFORD N., SHA G., XIA J., RINGER S.P., BARNETT M.R. “Solute segregation and texture modification in an extruded magnesium alloy containing gadolinium” Scripta Mater 2011;65:919. [104] LEE S.G., PATEL G.R., GOKHALE A.M., SREERANGANATHAN A., HORSTEMEYER M.F. “Variability in the tensile ductility of high-pressure die- cast AM50 Mg-alloy”. Scripta Mater 2005; 53:851-856. [105] WEILER J.P., WOOD J.T. “Modeling fracture properties in a die-cast AM60B magnesium alloy: I-Analytical failure model”. Mater Sci Eng 2009; 527:25-31. PART VIII. REFERENCES 154 [106] WEILER J.P., WOOD J.T. “Modeling fracture properties in a die-cast AM60B magnesium alloy: II-The effects of the size and location of porosity determined using finite element simulations”. Mater Sci Eng 2009; 527:32-37. [107] LEE C.D. “Dependence of tensile properties of AM60 magnesium alloy on microporosity and grain size”. Mater Sci Eng 2007; 454- 455:575-580. [108] WEILER J.P., WOOD J.T., KLASSEN R.J., MAIRE E., BERKMORTEL R., WANG G. “Relationship between internal porosity and fracture strength of die-cast magnesium AM60B alloy”. Mater Sci Eng 2005; 395:315-322. [109] WEI Y.H., HOU L.F., YANG L.J., XU B.S., KOZUKA M., ICHINOSE H. “Microstructures and properties of die casting components with various thicknesses made of AZ91D alloy”. J Mater Proc Tech 2009; 209:3278- 3284. [110] PRAKASH D.G.L., REGENER D., VORSTER W.J.J. “Effect of position on the tensile properties in high-pressure die cast Mg alloy”. J Alloys Comp 2009; 470:111-116. [111] SONG J., XIONG S.M., LI M., ALLISON J. “The correlation between microstructure and mechanical properties of high-pressure die-cast AM50 alloy”. J Alloys Comp 2009; 477:863-869. [112] DØRUM C., HOPPERSTAD O.S., BERSTAD T., DISPINAR D. “Numerical modelling of magnesium die-castings using stochastic fracture parameters”. Eng Fract Mech 2009; 76:2232-2248. [113] MAYER H., PAPAKYRIACOU M., ZETTL B., STANZL-TSCHEGG S.E. “Influence of porosity on the fatigue limit of die cast magnesium and aluminium alloys”. Int J Fatigue 2003; 35:245-256. [114] LEE SG, GOKHALE AM, PATEL GR, EVANS M. “Effect of process parameters on porosity distributions in high-pressure die-cast AM50 Mg- alloy”. Mater Sci Eng 2006; 427:99-111. [115] LEE SG, GOKHALE AM. “Formation of gas induced shrinkage porosity in Mg-alloy high-pressure die-castings”. Scripta Mater 2006; 55:387-390. [116] SONG QW, BEGGS P, EASTON M. “Compressive strain-rate sensitivity of magnesium–aluminum die casting alloys”. Materials & Design 2009; 30:642-648. PART VIII. REFERENCES 155 [117] WATANABE H., ISHIKAWA K., MUKAI T. “High strain rate deformation behavior of Mg–Al–Zn alloys at elevated temperatures”. Key Eng Mater 2007;340-341:107 [118] MUKAI T., MOHRI T., MABUCHI M., NAKAMURA M., ISHIKAWA K., HIGASHI K. “Experimental study of a structural magnesium alloy with high absorption energy under dynamic loading”. Scripta mater 1998;39:1249. [119] MUKAI T., YAMANOI M., YIGASHI K. “ductility enhancement in magnesium alloys under dynamic loading”. Mater Sci Forum 2000;350- 351:97. [120] YOKOHAMA T. “Impact Tensile Stress–strain Characteristics of Wrought Magnesium Alloys”. Strain 2003;39:167. [121] SHU D.W., ZHOU W., MA G.W. “Tensile Mechanical Properties of AM50A Alloy by Hopkinson Bar”. Key Eng Mater 2007;340-341:247. [122] EL-MAGD E., ABOURIDOUANE M. “Characterization, modelling and simulation of deformation and fracture behavior of the light-weight wrought alloys under high strain rate loading”. Int J Impact Eng 2006;32:741. [123] LI B., JOSHI S., AZEVEDO K., MA E., RAMESH K.T., FIGUEIREDO R.B., LANGDON T.G. “Dynamic testing at high strain rates of an ultrafine- grained magnesium alloy processed by ECAP”. Mater Sci Eng 2009;517:24. [124] Hielscher R., Mtex Software, TU Chemnitz, Germany. [125] Fernández A., Pérez-Prado M.T., Jérusalem A. “Continuum modeling of the response of a Mg alloy AZ31 rolled sheet during uniaxial deformation” . Int J Plast 2011; 27:1739-1757 [126] ROBINSON J.M., SHAW M.P. “Microstructure and mechanical influences on dynamic strain aging phenomena”. Int Mater Rev 1994; 39(3):113-122. [127] Calnan E.A., Clews C.J.B. “The development of deformation textures in metals Part III. Hexagonal structures”. Phil Mag. Ser. 7 1951; 42(331):919-931 [128] JÉRUSALEM A, FERNÁNDEZ A, PÉREZ-PRADO MT. “Modelización del maclado  2110 de una chapa laminada de Mg-3%Al-1%Zn” Rev. Metal. Madrid. 2010;46:133. PART VIII. REFERENCES 156 [129] HONG S.G., PARK S.H., LEE C.S. “Role of  2110 twinning characteristics in the deformation behavior of a polycrystalline magnesium alloy” Acta mater 2010; 58:5873-5885. [130] POLMEAR I.J. “Light Alloys”. Ed. Edward Arnold, London, UK, 1989, p.190 [131] MEYERS M.A. “Dynamic behavior of materials”. JJohn Wiley & Sons, New York, 1994. [132] ARMSTRONG R.W., WALLEY S.M. “High strain rate properties of metals and alloys” Int Mater Rev 2008;53:105-128. [133] PADILLA H.A., SMITH C.D., LAMBROS J., BEAUDOIN A.J., ROBERTSON I.M., “Effects of Deformation Twinning on Energy Dissipation in High Rate Deformed Zirconium” Metall Mater Trans 2007; 38A:2916-2927. [134] GURAO N.P., KAPOOR R, SUWAS S. “Deformation behaviour of commercially pure titanium at extreme strain rates” Acta Mater 2011; 59: 3431-3446 [135] NEMAT-NASSER S., GUO W.G., CHENG J.Y. “Mechanical properties and deformation mechanisms of a commercially pure titanium” Acta Mater 1999; 47:3705-3720. [136] JASWON M. A.; DOVE D. B. "Twinning properties of lattice planes”. Acta Cryst 1956; 9:621-626 [137] BILBY B.A., CROCKER A.G. “The Theory of the Crystallography of Deformation Twinning” Proc Roy Soc 1965; 288:240-255 [138] SERRA A., BACON D.J. “Modelling the motion of  2211 twinning dislocations in the HCP metals” Mater Sci Eng A 2005; 400–401: 496- 498. [139] LI B., MA E. “Atomic Shuffling Dominated Mechanism for Deformation Twinning in Magnesium”. Phys Rev Lett 2009; 103:035503. [140] SERRA A., BACON D.J., POND R.C., Comment on “Atomic Shuffling Dominated Mechanism For Deformation Twinning In Magnesium” Phys Rev Lett 2010; 104:029603. [141] CABIBBO M., BLUM W., EVANGELISTA E., KASSNER M.E. AND MEYERS M.A., "Transmission Electron Microscopy Study of Strain-Induced Low- and High-Angle Boundary Development in Equal-Channel Angular- PART VIII. REFERENCES 157 Pressed Commercially Pure Aluminum" Metall. and Mater. Trans. 39, pp. 181-189, 2008. [142] WEN L., CHEN P., TONG Z.F., TANG B. Y., PENG L.M., DING W.J., “A systematic investigation of stacking faults in magnesium via first- principles calculation”. Eur Phys J B 2009; 72:397-403. [143] SMITH A.E., “Surface, interface and stacking fault energies of magnesium from first principles calculations”. Surface Sci 2007; 601: 5762-5765 [144] YI S., SCHESTAKOW I., ZAEFFERER S. “Twinning-related microstructural evolution during hot rolling and subsequent annealing of pure magnesium”. Mater Sci ·ng 2009; 516:58-64. [145] ZHU S.Q., YAN H.G., CHEN J.H., WU Y.Z., LIU J.Z., TIAN J. “Effect of twinning and dynamic recrystallization on the high strain rate rolling process”. Scripta Mater 2010; 63:985-988. [146] SITDIKOV O., KAIBYSHEV R., SAKAI T. ”Dynamic Recrystallization Based on Twinning in Coarse-Grained Mg”. Mater Sci Forum 2003; 419- 422: 521-526 [147] SITDIKOV O., KAIBYSHEV R. “Dynamic recrystallization in pure magnesium” Mater Trans 2001; 42:1928-1937. [148] RAEISINIA B., AGNEW S.R. “Using polycrystal plasticity modeling to determine the effects of grain size and solid solution [149] BARNETT MR, NAVE MD, GHADERI A. “Yield point elongation due to twinning in a magnesium alloy” Acta mater 2012;60:1433-1443. PART IX. APPENDICES 158 PART IX APPENDICES PART IX. APPENDICES 159 9. APPENDICES. This appendix comprises, first, the stress-strain curves of several tests that were finally not included in the manuscript (section 9.1). In particular, data are presented for AM60 samples machined out of an automotive component processed by die casting and AM60 and AZ91 samples machined out of ingots processed by gravity casting. Additionally, it includes the tabulated mechanical properties corresponding to all the tests carried out (section 9.2). This information is thought useful for anyone wishing to know the mechanical response of several Mg alloys under a wide range of testing conditions. PART IX. APPENDICES 160 9.1. Stress-strain curves (a) (b) (c) (d) Figure 9.1. True stress – true strain curves corresponding to an AM60 automotive part processed by die casting and tested in compression at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, die casting, (5x10-4 s-1) RT RT 200 ºC 400ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, die casting, (5x10-3 s-1) RT 200 ºC 200ºC 300ºC 300ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, die casting, (5x10-2 s-1) RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, die casting, (103 s-1) RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 300 ºC 350 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 161 (a) (b) (c) Figure 9.2. True stress – true strain curves corresponding to samples of an AM60 automotive part processed by die casting and tested in compression at different temperatures: a) RT, b) 200 ºC and c) 400 ºC. 5x10-4 s-1 5x10-4 s-1 5x10-3 s-1 5x10-3 s-1 5x10-2 s-1 5x10-2 s-1 103 s-1 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, Die casting, RT Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, Die casting, 200ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, Die casting, 400ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 162 (a) (b) (c) (d) Figure 9.3. True stress – true strain curves corresponding to an AM60 ingot alloy tested in compression at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (5x10-2 s-1) RT RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (5x10-4 s-1) RT RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (5x10-3 s-1) RT RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot, (103 s-1) RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 300ºC 350 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 163 (a) (b) (c) Figure 9.4. True stress – true strain curves corresponding to an AM60 ingot alloy tested in compression at different temperatures: a) RT, b) 200 ºC and c) 400 ºC. 5x10-4 s-1 5x10-4 s-1 5x10-3 s-1 5x10-3 s-1 5x10-2 s-1 5x10-2 s-1 103 s-1 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60 ingot, 200 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60 ingot, 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60 ingot, RT Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 164 (a) (b) (c) (d) Figure 9.5. True stress – true strain curves corresponding to an AM60 ingot alloy processed by gravity casting and tested in tension at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (5x10-4 s-1) RT RT 200 ºC 200 ºC 300 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (5x10-3 s-1) RT RT 200 ºC 200 ºC 300 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (5x10-2 s-1) RT RT 200 ºC 200 ºC 300 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot (103 s-1) RT 200 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 165 (a) (b) (c) Figure 9.6. True stress – true strain curves corresponding to an AM60 ingot alloy processed by gravity casting and tested in tension at different temperatures: a) RT, b) 200 ºC and c) 300 ºC. 5x10-4 s-1 5x10-4 s-1 5x10-3 s-1 5x10-3 s-1 5x10-2 s-1 5x10-2 s-1 103 s-1 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot, RT Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot, 200 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AM60, ingot, 300 ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 166 (a) (b) (c) (d) Figure 9.7. True stress – true strain curves corresponding to an AZ91 ingot alloy processed by gravity casting and tested in compression at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot (5x10-2 s-1) RT RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot (103 s-1) RT 50 ºC 100 ºC 150 ºC 200 ºC 250 ºC 300 ºC 350 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot (5x10-4 s-1) RT RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot (5x10-3 s-1) RT RT 200 ºC 200 ºC 400 ºC 400 ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 167 (a) (b) (c) Figure 9.8. True stress – true strain curves corresponding to an AZ91 ingot alloy tested in compression at different temperatures: a) RT, b) 200 ºC and c) 400 ºC. 5x10-4 s-1 5x10-4 s-1 5x10-3 s-1 5x10-3 s-1 5x10-2 s-1 5x10-2 s-1 103 s-1 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, RT Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, 400ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, 200ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 168 (a) (b) (c) (d) Figure 9.9. True stress – true strain curves corresponding to an AZ91 ingot alloy processed by gravity casting and tested in tension at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, (5x10-4 s-1) RT RT 200 ºC 200 ºC 300 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, (5x10-3 s-1) RT RT 200 ºC 200 ºC 300 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, (5x10-2 s-1) RT RT 200 ºC 200 ºC 300 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, (103 s-1) RT 200 ºC 300 ºC Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 169 (a) (b) (c) Figure 9.10. True stress – true strain curves for corresponding to an AZ91 ingot alloy processed by gravity casting and tested in tension at different temperatures: a) RT, b) 200 ºC and c) 300 ºC. 5x10-4 s-1 5x10-4 s-1 5x10-3 s-1 5x10-3 s-1 5x10-2 s-1 5x10-2 s-1 103 s-1 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, 300 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, 200 ºC Tr ue S tr es s (M Pa ) True Strain 0 100 200 300 400 500 0 0.05 0.1 0.15 0.2 0.25 0.3 AZ91, ingot, RT Tr ue S tr es s (M Pa ) True Strain PART IX. APPENDICES 170 9.2 Mechanical properties data Table 9.1. Mechanical Properties of the AZ31 Mg alloy Alloy: AZ31. Fabrication Process: Rolling and Annealing. (Compression tests along rolling direction (RD) and normal direction (ND))  (s-1) T (ºC) 0.005 (MPa) 0.005 (MPa) max (MPa) max (MPa) f f RD ND RD ND RD ND 5x10-4 RT 64 276 334 320 0.16 0.06 200 75 100 90 122 0.60* 0.44* 400 15 19 21 21 0.70* 0.68* 5x10-3 RT 64 230 338 329 0.135 0.06 200 75 125 152 176 0.47* 0.45* 400 24 30 26 33 0.74* 0.68* 5x10-2 RT 62 278 350 333 0.09 0.04 200 65 214 233 241 0.53* 0.40* 400 31 41 37 51 0.78* 0.55* 103 RT 65 279 408 393 0.25 0.47 50 65 230 393 366 0.23 0.40 100 50 238 331 331 0.25 0.36 150 52 233 276 265 0.26 0.38 200 52 185 229 229 0.27 0.35 250 53 172 208 215 0.29 0.39 300 62 150 192 196 0.22 0.37 350 45 126 174 167 0.26 0.37 400 46 126 175 159 0.22 0.39 *Test stopped PART IX. APPENDICES 171 Table 9.2. Mechanical Properties of the MN11 Mg alloy. Alloy: MN11. Fabrication Process: extrusion. (Tests parallel to the extrusion axis)  (s-1) T (ºC) 0.005 (MPa) 0.005 (MPa) max (MPa) max (MPa) f f Compression Tension Compression Tension Compression Tension 5x10-4 RT 93 --- --- --- 264 --- --- --- 0.17 --- --- --- 200 81 81 --- --- 263 284 --- --- 0.12 0.10 --- --- 400 23 28 --- --- 25 30 --- --- 0.40* 0.50* --- --- 5x10-3 RT 69 80 --- --- 290 290 --- --- 0.11 0.12 --- --- 200 88 86 --- --- 211 222 --- --- 0.12 0.11 --- --- 400 34 37 --- --- 44 42 --- --- 0.60* 0.60* --- --- 5x10-2 RT 100 95 --- --- 304 302 --- --- 0.19 0.15 --- --- 200 89 76 --- --- 206 229 --- --- 0.18 0.13 --- --- 400 56 56 --- --- 73 83 --- --- 0.50* 0.50* --- --- 103 RT 100 --- 134 --- 346 --- 363 --- 0.29 --- 0.25 --- 50 88 --- --- --- 326 --- --- --- 0.33 --- --- --- 100 80 --- --- --- 297 --- --- --- 0.33 --- --- --- 150 93 --- --- --- 243 --- --- --- 0.31 --- --- --- 200 99 --- 134 --- 201 --- 226 --- 0.33 --- 0.19 --- 250 92 --- --- --- 177 --- --- --- 0.34 --- --- --- 300 68 --- 134 --- 151 --- 176 --- 0.34 --- 0.40 --- 350 51 --- --- --- 140 --- --- --- 0.34 --- --- --- 400 56 --- --- --- 127 --- --- --- 0.33 --- --- --- *Test stopped PART IX. APPENDICES 172 Table 9.3. Mechanical Properties of the MN11 Mg alloy. Alloy: MN11. Fabrication Process: extrusion. (Tests perpendicular to the extrusion axis)  (s-1) T (ºC) 0.005 (MPa) max (MPa) f Compression Compression Compression 5x10-4 RT 95 95 271 260 0.29 0.38 200 85 97 250 285 0.19 0.19 400 22 23 25 30 0.65* 0.66* 5x10-3 RT 97 94 294 310 0.30 0.14 200 87 82 250 239 0.13 0.27 400 39 34 43 41 0.60* 0.60* 5x10-2 RT 86 97 300 316 0.15 0.23 200 88 89 235 227 0.31 0.30 400 43 42 80 80 0.60* 0.6* 103 RT 104 --- 351 --- 0.31 --- 50 108 --- 328 --- 0.25 --- 100 101 --- 345 --- 0.25 --- 150 94 --- 252 --- 0.31 --- 200 86 --- 205 --- 0.33 --- 250 85 --- 184 --- 0.32 --- 300 77 --- 157 --- 0.34 --- 350 62 --- 143 --- 0.22 --- 400 66 --- 140 --- 0.33 --- *Test stopped PART IX. APPENDICES 173 Table 9.4. Mechanical Properties of the AM60 Mg alloy. Alloy: AM60. Fabrication Process: Die Casting (tensile coupon)  (s-1) T (ºC) 0.005 (MPa) 0.005 (MPa) max (MPa) max (MPa) f f Compression Tension Compression Tension Compression Tension 5x10-4 RT 77 --- 149 --- 193 --- 238 --- 0.23 --- 0.08 --- 200 98 94 90 91 110 108 102 101 0.82* 0.53* 0.18 0.10 400 9 13 7 6 19 22 9 8 0.80* 0.60* 0.28 0.53 5x10-3 RT 161 --- 73 149 406 --- 124 230 0.24 --- 0.09 0.07 200 117 116 98 96 166 165 123 113 0.71* 0.72* 0.20 0.10 400 30 29 22 22 31 30 22 23 0.6* 0.60* 0.28 0.18 5x10-2 RT 162 159 153 152 390 398 251 265 0.22 0.24 0.09 0.12 200 124 --- 117 113 230 --- 163 161 0.64* --- 0.18 0.16 400 45 49 40 38 46 51 41 40 0.60* 0.75* 0.21 0.26 103 RT 169 --- 208 --- 386 --- 281 --- 0.21 --- 0.14 --- 50 160 --- 128 --- 359 --- 267 --- 0.22 --- 0.11 --- 100 151 --- 133 --- 311 --- 263 --- 0.20 --- 0.14 --- 150 148 --- 116 --- 282 --- 239 --- 0.21 --- 0.14 --- 200 139 --- 145 --- 253 --- 207 --- 0.22 --- 0.14 --- 250 129 --- 128 --- 230 --- 203 --- 0.22 --- 0.16 --- 300 113 --- 105 --- 216 --- 172 --- 0.22 --- 0.16 --- 350 85 --- 95 --- 177 --- 137 --- 0.14 --- 0.13 --- 400 85 --- 108 --- 186 --- 120 --- 0.23 --- 0.10 --- *Test stopped PART IX. APPENDICES 174 Table 9.5. Mechanical Properties of the AM60 Mg alloy. Alloy: AM60. Fabrication Process: Die Casting (automotive part)  (s-1) T (ºC) 0.005 (MPa) max (MPa) f Compression Compression Compression 5x10-4 RT 122 133 337 350 0.25 0.26 200 89 --- 104 --- 0.57* --- 400 9 11 16 15 0.65* 0.65* 5x10-3 RT 133 --- 336 --- 0.23 --- 200 95 83 160 156 0.43* 0.39* 400 26 25 27 26 0.68* 0.73* 5x10-2 RT 132 --- 361 --- 0.24 --- 200 107 110 246 236 0.46* 0.47* 400 35 33 41 39 0.66 0.66* 103 RT 117 --- 316 --- 0.29 --- 50 125 --- 292 --- 0.21 --- 100 115 --- 273 --- 0.26 --- 150 102 --- 223 --- 0.27 --- 200 105 --- 225 --- 0.28 --- 250 75 --- 189 --- 0.29 --- 300 99 --- 187 --- 0.19 --- 350 76 --- 180 --- 0.26 --- 400 55 --- 161 --- 0.28 --- *Test stopped PART IX. APPENDICES 175 Table 9.6. Mechanical Properties of the AM60 Mg alloy. Alloy: AM60. Fabrication Process: gravity casting.  (s-1) T (ºC) 0.005 (MPa) 0.005 (MPa) max (MPa) max (MPa) f f Compression Tension Compression Tension Compression Tension 5x10-4 RT 71 72 85 88 181 193 147 152 0.22 0.34 0.07 0.05 200 48 60 65 71 112 122 110 119 0.47* 0.38* 0.23 0.20 300 --- --- 47 45 --- --- 58 55 --- --- 0.22 0.22 400 20 24 --- --- 23 25 --- --- 0.79* 0.83* 5x10-3 RT 70 78 82 74 200 224 118 152 0.26 0.12 0.04 0.07 200 60 49 61 70 160 138 112 124 0.39* 0.44* 0.18 0.18 300 --- --- 48 46 --- --- 67 66 --- --- 0.20 0.28 400 34 34 --- --- 41 35 --- --- 0.76* 0.65* --- --- 5x10-2 RT 85 81 83 100 191 211 148 159 0.4* 0.4* 0.05 0.05 200 81 70 71 64 165 181 150 160 0.4* 0.4* 0.16 0.19 300 --- --- 54 51 --- --- 90 86 --- --- 0.22 0.29 400 41 30 --- --- 64 48 --- --- 0.65* 0.65* --- --- 103 RT 45 --- 73 --- 253 --- 158 --- 0.29 --- 0.07 --- 50 57 --- --- --- 239 --- --- --- 0.28 --- --- --- 100 64 --- --- --- 224 --- --- --- 0.25 --- --- --- 150 55 --- --- --- 188 --- --- --- 0.28 --- --- --- 200 39 --- 60 --- 169 --- 173 --- 0.28 --- 0.15 --- 250 30 --- --- --- 151 --- --- --- 0.25 --- --- --- 300 24 --- 57 --- 134 --- 154 --- 0.24 --- 0.15 --- 350 40 --- --- --- 126 --- --- --- 0.22 --- --- --- 400 26 --- --- --- 85 --- --- --- 0.28 --- --- --- *Test stopped PART IX. APPENDICES 176 Table 9.7. Mechanical Properties of the AZ91 Mg alloy. Alloy: AZ91. Fabrication Process: Die Casting (tensile coupon)  (s-1) T (ºC) 0.005(MPa) 0.005 (MPa) max (MPa) max (MPa) f f Compression Tension Compression Tension Compression Tension 5x10-4 RT 221 217 186 184 486 447 250 234 0.19 0.17 0.05 0.04 200 106 109 91 97 113 115 98 105 0.34* 0.34* 0.14 0.17 400 9 9 3 4 16 16 4 5 0.34* 0.34* 0.54 0.58 5x10-3 RT 214 206 189 187 446 444 254 261 0.17 0.19 0.06 0.06 200 128 132 113 116 144 152 132 130 0.60* 0.60* 0.19 0.20 400 25 26 19 19 26 30 20 19 0.60* 0.60* 0.30 0.28 5x10-2 RT 210 209 189 188 445 431 248 241 0.18 0.18 0.06 0.05 200 148 162 138 134 220 231 184 168 0.65* 0.62* 0.13 0.13 400 44 44 37 36 44 49 37 37 0.6* 0.70* 0.32 0.21 103 RT 221 --- 157 --- 436 --- 298 --- 0.17 --- 0.06 --- 50 214 --- 259 --- 402 --- 302 --- 0.20 --- 0.09 --- 100 191 --- 127 --- 342 --- 238 --- 0.20 --- 0.05 --- 150 186 --- 157 --- 322 --- 236 --- 0.21 --- 0.07 --- 200 187 --- 136 --- 306 --- 233 --- 0.22 --- 0.05 --- 250 157 --- 143 --- 264 --- 203 --- 0.22 --- 0.08 --- 300 --- --- 126 --- --- --- 154 --- --- --- 0.05 --- 350 107 --- 189 --- 206 --- 191 --- 0.16 --- 0.04 --- 400 94 --- --- --- 148 --- --- --- 0.24 --- --- --- *Test stopped PART IX. APPENDICES 177 Table 9.8. Mechanical Properties of the AZ91 Mg alloy. Alloy: AZ91. Fabrication Process: gravity casting.  (s-1) T (ºC) 0.005 (MPa) 0.005 (MPa) max (MPa) max (MPa) f f Compression Tension Compression Tension Compression Tension 5x10-4 RT 180 135 139 135 268 248 147 158 0.06 0.10 0.01 0.02 200 100 48 86 85 133 112 102 118 0.54* 0.46* 0.11 0.12 300 --- --- 56 56 --- --- 63 59 --- --- 0.16 0.13 400 21 24 --- --- 21 24 --- --- 0.80* 0.80* --- --- 5x10-3 RT 129 150 133 156 254 241 134 174 0.12 0.11 0.02 0.03 200 86 111 107 95 155 166 145 137 0.53* 0.50* 0.13 0.10 300 --- --- 66 66 --- --- 75 75 --- --- 0.13 0.15 400 36 39 --- --- 37 40 --- --- 0.80* 0.80* --- --- 5x10-2 RT 106 119 120 153 242 237 138 174 0.24 0.27 0.02 0.03 200 157 114 97 93 211 201 126 126 0.60* 0.50* 0.06 0.04 300 --- --- 75 70 --- --- 91 88 --- --- 0.18 0.10 400 43 44 --- --- 50 54 --- --- 0.61* 0.73* --- --- 103 RT 115 --- 83 --- 289 --- 179 --- 0.28 --- 0.04 --- 50 96 --- --- --- 237 --- --- --- 0.29 --- --- --- 100 110 --- --- --- 262 --- --- --- 0.27 --- --- --- 150 79 --- --- --- 218 --- --- --- 0.26 --- --- --- 200 62 --- 114 --- 205 --- 169 --- 0.28 --- 0.06 --- 250 65 --- --- --- 186 --- --- --- 0.29 --- --- --- 300 69 --- 80 --- 175 --- 138 --- 0.23 --- 0.05 --- 350 55 --- --- --- 128 --- --- --- 0.27 --- --- --- 400 28 --- --- --- 70 --- --- --- 0.26 --- --- --- *Test stopped LIST OF FIGURES 178 LIST OF FIGURES PAG. PART II. INTRODUCCIÓN. EL MAGNESIO Figure 2.1. Tabla periódica de los elementos 17 Figure 2.2. Densidades de algunos metales estructurales 18 Figure 2.3. Aplicaciones de las aleaciones de magnesio en automóviles 21 Figure 2.4. Aplicaciones de las aleaciones de magnesio para la fabricación de dispositivos biomédicos, electrónicos y material deportivo 22 Figure 2.5. Estructura cristalina hexagonal compacta (HCP): representación de la celdilla unidad mediante esferas reducidas 23 Figure 2.6. Tipos de textura que se desarrollan en los materiales procesados mediante tres procesos de fabricación: colada, laminación más un recocido posterior y extrusión 25 Figure 2.7. Sistemas de deslizamiento de las aleaciones de magnesio 27 Figure 2.8. Sistemas de maclado más comunes en las aleaciones de magnesio 28 Figure 2.9. Variación de la CRSS de los distintos sistemas de deslizamiento y maclado con la temperatura 35 PART III. RESEARCH PAPERS 3.1 Mechanical behavior and microstructural evolution of a Mg AZ31 sheet at dynamic strain rates Figure 3.1.1. Microstructure and microtexture of the as-received Mg AZ31-O alloy obtained by EBSD 47 Figure 3.1.2. Schematic illustrating the relative orientation of the tension and compression axes with respect to the c-axes and the sheet reference system (RD, TD, ND) 47 Figure 3.1.3. True stress-true strain curves corresponding to HSR (103 s-1) tests performed at room temperature and at 250°C.: (a) RD-T versus RD-C; (b) RD-T versus TD-T; (c) RD-C versus ND-C 48 LIST OF FIGURES 179 PAG. Figure 3.1.4. Variation of the true stress at ε=0.005 (0.005) and of the flow stress (max) with temperature during HSR (103 s-1, closed symbols) and LSR (10-3 s-1, open symbols) strain rate tests 49 Figure 3.1.5. (a) Variation of the yield stress tension- compression asymmetry (RD-T)/(RD-C)), the yield stress in-plane anisotropy (TD-T)/RD-T)) and the yield stress out-of-plane anisotropy (ND-C)/RD-C)) with temperature at both high strain rates (HSR) and low strain rates (LSR). (b) Modulus-compensated flow stress (max/G) as a function of temperature corresponding to tests performed in tension along in- plane directions (RD-T/TD-T) and in compression along RD and ND (RD-C and ND-C) at both high strain rate (HSR) and low strain rate (LSR) 50 Figure 3.1.6. EBSD inverse pole figure maps in the ND and textures measured in the samples loaded in tension at HSR under the following conditions: a) room temperature, tension along RD; b) room temperature, tension along TD; c) 250ºC, tension along RD. d)Kernel average misorientation map at the area c) 52 Figure 3.1.7. Microstructure and microtexture of the AZ31 sheet deformed in tension along RD at 10-3 s-1 and 250ºC. a) EBSD inverse pole figure maps in the ND and b) Kernel Average misorientation map of the same area 53 Figure 3.1.8. Discrete orientations of the recrystallized grains in the sample deformed at high strain rates (RD-T at 103 s-1 and 250ºC) 54 LIST OF FIGURES 180 PAG. 3.2 Twinning and grain subdivision during dynamic deformation of Mg AZ31 sheet alloy at room temperature Figure 3.2.1. Texture of the as-received sheet alloy 59 Figure 3.2.2. Room temperature stress-strain curves corresponding to the AZ31 alloy deformed at a high strain rate (103 s-1) and a quasi-static strain rate in (a) compression along the RD; (b) tension along the RD; and (c) compression along the ND. Images of the specimen after fracture, taken in real time with a high speed camera, are included next to each plot. The strains to failure corresponding to each tests, εf, are indicated by arrows 60 Figure 3.2.3. EBSD inverse pole figure maps in the ND and microtextures measured in the samples deformed quasi-statically at room temperature in compression along the RD up to strains of (a) 0.05 and (b) 0.13. (c) Twin boundary map corresponding to the sample strained up to 0.13 62 Figure 3.2.4. EBSD inverse pole figure maps in the ND and microtextures measured in the samples dynamically loaded at room temperature in compression along the RD up to strains of (a) 0.05, (b) 0.10 and (c) 0.13. (d) Boundary map corresponding to the sample strained up to 0.13 63 Figure 3.2.5. Misorientation distribution histograms corresponding to AZ31 compressed at dynamic strain rates along the RD at room temperature up to strains of 0.05, 0.10 and 0.13 64 Figure 3.2.6. Texture of the AZ31 alloy deformed to failure at room temperature in tension along the RD at a strain rate of 10-3 s-1. Direct pole figures, measured by neutron diffraction 64 LIST OF FIGURES 181 PAG. Figure 3.2.7. EBSD inverse pole figure maps in the ND and direct pole figures, measured by neutron diffraction, corresponding to the samples dynamically loaded at room temperature in tension along the RD up to strains of (a) 0.10 and (b) 0.20. (c) Twin boundary map corresponding to the sample deformed to 0.20 65 Figure 3.2.8. Boundary maps corresponding to the AZ31 alloy deformed in tension along the RD: (a) at a high strain rate up to a strain of 0.10; (b) at a high strain rate up to a strain of 0.20; (c) at a low strain rate up to failure (failure strain approximately equal to 0.20). (d) Twin boundary map corresponding to the sample deformed at a low strain rate until failure 66 Figure 3.2.9. (a) EBSD inverse pole figure map in the ND, macrotexture (X-ray) and microtexture (EBSD) corresponding to the AZ31 alloy dynamically loaded at room temperature in compression along the ND up to a strain of 0.05; (b) twin boundary map corresponding to the same sample; (c) EBSD inverse pole figure map in the ND, macrotexture (X-ray) and microtexture (EBSD) corresponding to the AZ31 alloy dynamically loaded at room temperature in compression along the ND up to a strain of 0.10 67 3.3 Influence of texture on the recrystallization mechanisms in an AZ31 Mg sheet alloy at dynamic rates Figure 3.3.1. Stress-strain curves corresponding to the Mg AZ31 alloy deformed at high strain rate (103 s-1, solid line) and at quasi-static strain rates (10-3 s-1, dotted line) at 250 ºC in (a) compression along RD; (b) tension along RD; and (c) compression along NDThe images below each graph show the corresponding specimen before testing (upper image) and after dynamic deformation up to a true strain of approximately 0.25 (lower image) 73 Figure 3.3.2. Work hardening behavior corresponding to the AZ31 sheet alloy deformed at 250°C and 103 s-1. 74 LIST OF FIGURES 182 PAG. Figure 3.3.3. EBSD inverse pole figure maps in the ND and microtextures corresponding to samples dynamically loaded at 250ºC in compression along RD up to strains of (a) 0.06, (b) 0.10, and (c) 0.15. (d) Boundary map corresponding to the sample strained up to 0.15 74 Figure 3.3.4. Misorientation distribution histograms corresponding to the AZ31 sheet alloy compressed along RD at 103 s-1 and 250 ºC up to strains of 0.10 and 0.15 75 Figure 3.3.5. EBSD inverse pole figure maps in the ND and microtextures corresponding to samples dynamically loaded at 250ºC in tension along RD up to strains of (a) 0.10, and (c) 0.20. (b) Twin boundary map corresponding to the sample deformed up to a strain of 0.10. (d) Boundary map corresponding to the sample strained up to 0.20 76 Figure 3.3.6. EBSD inverse pole figure maps in the ND and microtextures corresponding to samples dynamically loaded at 250ºC in compression along ND up to strains of (a) 0.07, (b) 0.095, and (c) 0.17. (d) Boundary map corresponding to the sample strained up to 0.17 77 Figure 3.3.7. Misorientation distribution histograms corresponding to the AZ31 sheet alloy compressed along ND at 103 s-1 and 250 ºC up to strains of 0.095 and 0.17 78 Figure 3.3.8. Kernel average misorientation (KAM) map corresponding to the AZ31 alloy deformed (a) in compression along RD up to a strain of 0.15, (b) in tension along RD up to a strain of 0.20 and (c) in compression along ND up to a strain of 0.17 78 PART IV. ADDITIONAL AND COMPARATIVE STUDIES Figure 4.1. Microstructure and texture of the as- extruded MN11 bar. (a) Micrograph obtained by optical microscopy, (b) X-ray pole figures and (c) X-ray inverse pole figure showing the orientation of the extrusion axis (EA). The plane of observation is perpendicular to the extrusion axis (EA). 87 LIST OF FIGURES 183 PAG. Figure 4.2. Effect of heat treatment on the microstructure and the texture: (a,b) 200 ºC, 1h; (c,d) 400 ºC, 1h. Both samples where heated at 10ºC/min up to the corresponding heat treatment temperature. 88 Figure 4.3. Stress-strain curves corresponding to the MN11 alloy deformed in compression at room temperature, 200 ºC and 400 ºC at: (a) 5x10-4 s-1 along the EA, (b) 5x10-4 s-1 perpendicular to the EA, (c) 5x10-3 s-1 along the EA, (d) 5x10-3 s-1 perpendicular to the EA, (e) 5x10-2 s-1 along the EA, (f) 5x10-2 s-1 perpendicular to the EA, (g) 103 s-1 along the EA, and (h) 103 s-1 perpendicular to the EA. 89 Figure 4.4. Inverse pole figures showing the orientation of two radial directions of the as-received extruded bar of MN11. The two directions have been labeled and . Both are perpendicular to the EA. 91 Figure 4.5. Evolution of the strain to failure ( f ) and the maximum flow stress ( max ) with temperature at different strain rates: a), b) 5x10-4 s-1; c),d) 5x10-3 s-1; e),f) 5x10-2 s-1, g),h) 103 s-1. 92 Figure 4.6. Stress-strain curves corresponding to the MN11 alloy deformed in tension parallel to the extrusion axis (EA) at high strain rate at room temperature, 200 ºC and 300 ºC. 96 Figure 4.7. Microtexture of the MN11 alloy tested at low strain rate (5x10-3 s-1), at room temperature in compression along the EA up to a strain of 0.12 (fracture) (a) EBSD inverse pole figure map showing the orientation of the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 98 Figure 4.8. Microtexture of the MN11 alloy tested at low strain rate (5x10-3 s-1) at 200 ºC in compression along the EA up to a strain of 0.11 (fracture) (a) EBSD inverse pole figure map in the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 99 LIST OF FIGURES 184 PAG. Figure 4.9. Microtexture of the MN11 alloy tested at low strain rate (5x10-3 s-1) at 400 ºC in compression along the EA up to a strain of 0.6 (test stopped) (a) EBSD inverse pole figure map in the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 100 Figure 4.10. Evolution of the microtexture of the MN11 alloy during compression at high strain rate (103 s-1) along the EA at RT with increasing strain. (a) ε= 0.095; EBSD inverse pole figure map showing the orientation of the EA; (b) ε= 0.095; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) ε= 0.095; Pole figures and inverse pole figure showing the orientation of the EA. (d) ε= 0.15; EBSD inverse pole figure map showing the orientation of the EA; (e) ε= 0.15; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (f) ε= 0.15; Pole figures and inverse pole figure showing the orientation of the EA. The plane shown is perpendicular to the EA. 103 Figure 4.11. Evolution of the microtexture of the MN11 alloy during compression at high strain rate (103 s-1) along the EA at 400 ºC with increasing strain. (a) ε= 0.11; EBSD inverse pole figure map showing the orientation of the EA; (b) ε= 0.11; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) ε= 0.11; Pole figures and inverse pole figures showing the orientation of the EA. (d) ε= 0.16; EBSD inverse pole figure map showing the orientation of the EA; (e) ε= 0.16; Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (f) ε= 0.16; Pole figures and inverse pole figures showing the orientation of the EA. 104 Figure 4.12. Microtexture of the MN11 alloy tested in tension along the EA at high strain rate (103 s-1) at RT up to a strain of 0.25 (fracture). (a) EBSD inverse pole figure map showing the orientation of the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures. (d) Inverse pole figure showing the orientation of the EA. 106 LIST OF FIGURES 185 PAG. Figure 4.13. Microtexture of the MN11 alloy tested in tension along the EA at high strain rate (103 s-1) at 300 ºC up to a strain of 0.40 (fracture). (a) EBSD inverse pole figure map showing the orientation of the EA; (b) Twin boundary map (Red: tensile twins; Blue: compression twins; Green: double twins); (c) Pole figures.; (d) Inverse pole figure showing the orientation of the EA. 107 Figure 4.14. Superposition of the inverse stereographic triangle (extrusion axis) corresponding to the as- extruded MN11 alloy and the schematic triangle showing contours of constant resolved shear stress, i.e., of constant Schmid factor for basal slip, reported by Calnan and Clews [127] 108 Figure 4.15. Microstructure and macrotexture (X-ray) of the as-received die-cast Mg alloy: (a) AM60, (b) AZ91. The plane examined is perpendicular to the loading axis. 113 Figure 4.16. Variation of the yield stress (0.005) and of the flow stress (max) with temperature during high (103 s-1, left column) and low (5x10-3 s-1, right column) strain rate tests in die-cast AM60B and AZ91B alloys: (a) AM60B, 103 s-1 in compression; (b) AM60B, 5x10-3 s-1 in compression; (c) AM60B, 103 s-1 in tension, (d) AM60B, 5x10-3 s-1 in tension, (e) AZ91D, 103 s-1 in compression; (f) AZ91D, 5x10-3 s-1 in compression; (g) AZ91D, 103 s-1 in tension, (h) AZ91D, 5x10-3 s-1 in tension. 116 Figure 4.17. True stress-true strain curves corresponding high (103 s-1) and low (5x10-3 s-1) strain rate tests performed in compression at different temperatures in die-cast Mg alloys : (a) AM60B, 103 s-1, (b) AM60B, 5x10-3 s-1 (c) AZ91D, 103 s-1, (d) AZ91, 5x10-3 s-1 . 117 Figure 4.18. True stress-true strain curves corresponding high strain rate (103 s-1) tests performed in compression at different temperatures in die-cast Mg alloys : (a) AM60B (b) AZ91D. 118 PART IX. APPENDICES Figure 9.1. True stress – true strain curves corresponding to an AM60 automotive part processed by die casting and tested in compression at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 160 LIST OF FIGURES 186 PAG. Figure 9.2. True stress – true strain curves corresponding to samples of an AM60 automotive part processed by die casting and tested in compression at different temperatures: a) RT, b) 200 ºC and c) 400 ºC. 161 Figure 9.3. True stress – true strain curves corresponding to an AM60 ingot alloy tested in compression at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 162 Figure 9.4. True stress – true strain curves corresponding to an AM60 ingot alloy tested in compression at different temperatures: a) RT, b) 200 ºC and c) 400 ºC. 163 Figure 9.5. True stress – true strain curves corresponding to an AM60 ingot alloy processed by gravity casting and tested in tension at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 164 Figure 9.6. True stress – true strain curves corresponding to an AM60 ingot alloy processed by gravity casting and tested in tension at different temperatures: a) RT, b) 200 ºC and c) 300 ºC. 165 Figure 9.7. True stress – true strain curves corresponding to an AZ91 ingot alloy processed by gravity casting and tested in compression at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 166 Figure 9.8. True stress – true strain curves corresponding to an AZ91 ingot alloy tested in compression at different temperatures: a) RT, b) 200 ºC and c) 400 ºC. 167 Figure 9.9. True stress – true strain curves corresponding to an AZ91 ingot alloy processed by gravity casting and tested in tension at different strain rates: a) 5x10-4 s-1, b) 5x10-3 s-1, c) 5x10-2 s-1, d) 103 s-1. 168 Figure 9.10. True stress – true strain curves for corresponding to an AZ91 ingot alloy processed by gravity casting and tested in tension at different temperatures: a) RT, b) 200 ºC and c) 300 ºC. 169 LIST OF TABLES 187 LIST OF TABLES PÁG. PART II. INTRODUCCIÓN. EL MAGNESIO Tabla 2.1. Comparación de las propiedades físicas del Mg puro con las de otros metales estructurales 19 Tabla 2.2 Valores de la CRSS para los diferentes sistemas de deslizamiento y maclado en Mg según Barnett [32]. 33 PART III. RESEARCH PAPERS 3.2 Twinning and grain subdivision during dynamic deformation of Mg AZ31 sheet alloy at room temperature. Table 3.2.1. Comparison of the stacking fault energies corresponding to non-basal and basal dislocations in Mg with those of other metals 69 PART IV. ADDITIONAL AND COMPARATIVE STUDIES Table 4.1. Average yield stress values (in MPa). (a) Quasi-static deformation; (b) dynamic deformation. 90 Table 4.2. Strain rate sensitivity (m) values corresponding to the different quasi-static testing conditions investigated. Note: In the PP tests, the stress values utilized for the calculation of m are the average values between the two tests performed at each temperature. 94 Table 4.3. Theoretical grain size (dT) and measured grain size (dM) for MN11 samples tested in compression and in tension at high and low strain rate at different temperatures. The grain sizes are measured and calculated along a plane perpendicular to the extrusion axis. 102 PART IX. APPENDICES Table 9.1. Mechanical Properties of AZ31 Mg alloy 170 Table 9.2. Mechanical Properties of MN11 Mg alloy. 171 Table 9.3. Mechanical Properties of MN11 Mg alloy. 172 Table 9.4. Mechanical Properties of AM60 Mg alloy. 173 LIST OF TABLES 188 PAG. Table 9.5. Mechanical Properties of AM60 Mg alloy. 174 Table 9.6. Mechanical Properties of AM60 Mg alloy. 175 Table 9.7. Mechanical Properties of AZ91 Mg alloy. 176 Table 9.8. Mechanical Properties of AZ91 Mg alloy. 177 Tesis Nathamar Valenthina Dudamel Caballero PORTADA AGRADECIMIENTOS TABLE OF CONTENTS PARTE I: PROLOGO Y OBJETIVO PARTE II: INTRODUCCIÓN. ALEACIONES DE MAGNESIO PART III: RESEARCH PAPERS PART IV: COMPLEMENTARY STUDIES PART V: GENERAL DISCUSSION PART VI: CONCLUSIONS PART VII: FURTHER WORK PARTE VIII: REFERENCES PART IX: APPENDICES LIST OF FIGURES LIST OF TABLES