Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane

PROGRAMA DE DOCTORADO EN FÍSICA TESIS DOCTORAL: “Synthesis, Foaming Kinetics and Physical Properties of Cellular Nanocomposites Based on Rigid Polyurethane” Presentada por Mercedes Santiago Calvo para optar al grado de Doctora por la Universidad de Valladolid Dirigida por: Dr. Miguel Ángel Rodríguez Pérez Dr. Fernando Villafañe González Index 0- Resumen de la tesis en castellano 0.1 Introducción ................................................................................................................... 3 0.2 Marco de esta tesis ......................................................................................................... 4 0.3 Objetivos ......................................................................................................................... 8 0.4 Principales resultados y conclusiones............................................................................... 9 0.5 Estructura y metodología de la tesis ............................................................................... 12 0.6 Publicaciones, conferencias y actividades complementarias .......................................... 14 0.7 Referencias ................................................................................................................... 19 1- Introduction 1.1 Introduction .................................................................................................................. 27 1.2 Framework of this thesis ............................................................................................... 28 1.3 Objectives ...................................................................................................................... 32 1.4 Main novelties ............................................................................................................... 33 1.5 Structure of the thesis ................................................................................................... 36 1.6 Publications, conferences and activities ......................................................................... 37 1.7 References ..................................................................................................................... 42 2- Background and state of the art 2.1 Introduction .................................................................................................................. 49 2.2 Discover of polyurethane: Polyurethane foams ............................................................. 49 2.3 Basic chemistry, polymeric morphology and cellular structure of polyurethane foams .. 50 2.3.1 Basic chemistry ..................................................................................................... 50 3.3.2 Polymer morphology ............................................................................................ 56 3.3.3 Cellular structure and density ............................................................................... 57 2.4 Common raw materials of polyurethane foams ............................................................ 58 2.4.1 Polyol .................................................................................................................... 59 2.4.2 Isocyanate ............................................................................................................. 61 2.4.3 Surfactant .............................................................................................................. 63 2.4.4 Catalyst .................................................................................................................. 64 2.4.5 Blowing agent ........................................................................................................ 65 2.5 Foaming process of polyurethane foams ....................................................................... 66 Index 2.6 Properties and applications of polyurethane foams ...................................................... 68 2.7 Polyurethane foam composites ..................................................................................... 69 2.8 Monitoring of foaming process of polyurethane foams ................................................. 73 2.9 Thermoplastic polyurethane foams ............................................................................... 80 2.9.1 Thermoplastic polyurethanes................................................................................ 80 2.9.2 Foaming of thermoplastic polyurethanes ............................................................. 82 2.10 References ................................................................................................................... 84 3- Materials, foam production and experimental techniques 3.1 Introduction .................................................................................................................. 93 3.2 Polyurethane formulations ............................................................................................ 93 2.2.1 Commercial formulation of water-blown rigid polyurethane foam ..................... 94 3.2.2 Commercial formulation of cyclopentane-water blown rigid polyurethane foam 94 3.2.3 CellMat formulations of water-blown rigid polyurethane foams ......................... 95 3.1.4 CellMat formulations of thermoplastic polyurethane foams ............................... 98 3.3 Fillers ............................................................................................................................. 99 3.4 Foams production ........................................................................................................ 101 3.4.1 Rigid polyurethane foams .................................................................................. 101 3.4.2 Thermoplastic polyurethane foams .................................................................... 103 3.5 Experimental techniques ............................................................................................. 106 3.5.1 Summary of experimental techniques ............................................................... 106 3.5.2 Experimental techniques for the monitoring of foaming process ..................... 108 3.5.2.1 FTIR spectroscopy ............................................................................... 108 3.5.2.2 Infrared expandometry ....................................................................... 110 3.5.2.3 X-ray radioscopy .................................................................................. 111 3.5.2.4 Adiabatic temperature rise .................................................................. 113 3.6 References .................................................................................................................. 113 4- Development of a methodology to follow the foaming process of rigid polyurethane foams 4.1 Introduction ................................................................................................................ 119 “The effects of functional nanofillers on the reaction kinetics, microstructure, thermal and mechanical properties of water blown rigid polyurethane foams” .................................... 122 Index “Infrared expandometry: a novel methodology to monitor the expansion kinetics of cellular materials produced with exothermic foaming mechanisms” ............................................. 151 “X-ray radioscopy validation of a polyol functionalized with graphene oxide for producing rigid polyurethane foams with improved cellular structures” ........................................... 172 5- Effect of fillers on cellular structure and physical properties of rigid polyurethane foams 5.1 Introduction ................................................................................................................ 191 “Evaluation of the thermal conductivity and mechanical properties of water blown polyurethane rigid foams reinforced with carbon nanofibers” ........................................... 194 “Long-term thermal conductivity of cyclopentane-water blown rigid polyurethane foams reinforced with different types of fillers” ........................................................................... 216 6- Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties 6.1 Introduction ................................................................................................................ 243 “Synthesis, characterization and physical properties of rigid polyurethane foams prepared with poly(propylene oxide) polyols containing graphene oxide” ........................................ 246 “Improvement of thermal and mechanical properties by control of formulations in rigid polyurethane foams from polyols functionalized with graphene oxide” ............................ 270 7- International research stay: Thermoplastic polyurethane foams 7.1 Introduction ................................................................................................................ 291 “Synthesis, characterization, and foaming of thermoplastic polyurethane with different hard segment contents” ............................................................................................................ 293 8- Conclusions and Future work 8.1 Conclusions.................................................................................................................. 313 8.1.1 Rigid polyurethane foams .................................................................................. 313 8.1.2 Thermoplastic polyurethane foams .................................................................... 325 8.2 Future work ................................................................................................................. 328 Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane Mercedes Santiago Calvo CHACTER 0: RESUMEN DE LA TESIS EN CASTELLANO Resumen de la tesis en castellano estudio de las espumas de TPU es complementario a este y se realizó principalmente durante los tres meses de estancia y además se desarrolló durante el último año de esta tesis. Figure 0.4. Esquema simplificado de las investigaciones desarrolladas en esta tesis. 0.3- Objetivos El objetivo de la presente tesis es adquirir nuevos conocimientos sobre las espumas de PU termoestables (particularmente las espumas de RPU), centrándose en el estudio del proceso de espumado, especialmente en la cinética de reacción de la generación de gas (reacción de espumado) y polimerización (reacción de polimerización) cuando las micro- y/o nanopartículas se incorporan a la matriz de PU para mejorar las propiedades térmicas y mecánicas. El segundo aspecto importante a estudiar es el efecto que la modificación de la reacción podría tener en la estructura y propiedades de los materiales finales. El tercer aspecto importante es optimizar las formulaciones de los sistemas RPU que contienen partículas utilizando la información obtenida en el estudio cinético. Para alcanzar estos objetivos generales, se han considerado los siguientes objetivos específicos: Espumas de poliuretano (PU) Termoestable Termoplástico Espumado reactivo Espumado de disolución de gas Sintesisy caracterización de materiales de PU termoplásticos (TPU) Espumado y caracterización de espumasde TPU Composites de espumas de PU termoestables: •Síntesis y caracterización •Propiedades térmicas y mecánicas •Estudio del proceso de espumado, especialmentelas cinéticas de reacción •Optimizaciónde la formulación de PU 8 Chapter 0 • Modelización de la conductividad térmica de espumas de RPU modificadas con micropartículas y/o nanopartículas para estudiar el efecto de las partículas (Capítulo 5). Desarrollar nuestras formulaciones propias de PU para poder controlar las características finales de los sistemas RPU. Desarrollar y validar la metodología que permitirá estudiar la cinética de las principales reacciones químicas (espumado y polimerización) y la morfología de la matriz polimérica durante el proceso de espumación de las espumas RPU (Capítulo 4). Desarrollar y validar la metodología que permitirá estudiar las cinéticas de expansión que se producen en la producción de espumas RPU (Capítulo 4). Aplicar otras metodologías existentes que permitan estudiar la estructura microcelular interna durante el proceso de espumación de las espumas RPU (Capítulo 4). Aplicar las metodologías de monitorización del proceso de espumado a los sistemas RPU modificados con micropartículas y/o nanopartículas para evaluar el efecto de las partículas en el proceso de espumado, especialmente en la cinética de reacción y la morfología final de la matriz polimérica (Capítulo 6). Establecimiento de correlaciones entre la cinética de reacción, la morfología final de la matriz polimérica, la estructura celular y las propiedades físicas de los sistemas de RPU modificados con micropartículas y/o nanopartículas (Capítulo 6). Optimización de la formulación de RPU modificada con micropartículas y/o nanopartículas a partir de la información cinética obtenida (Capítulo 6). Para sistemas RPU modificados con micropartículas y / o nanopartículas, establecer correlaciones entre la temperatura de espumación y la evolución de su conductividad térmica con el tiempo (Capítulo 5). Modelización de la conductividad térmica de espumas de RPU modificadas con micropartículas y/o nanopartículas para estudiar el efecto de las partículas (Capítulo 5). Otro objetivo de la presente tesis es estudiar el comportamiento de espumado de los materiales de TPU utilizando la espumación por disolución de gas, para generar conocimiento sobre las espumas de TPU (Capítulo 7). 0.4- Principales resultados y conclusiones 1. La principal novedad de la tesis. Hasta ahora, la literatura científica actual [59, 62-75], que describe espumas de PU reforzadas con micro y/o nanopartículas, sigue el enfoque descrito en la Figura 0.5, que se centra principalmente en la producción y caracterización de las espumas, sin poner especial atención al efecto de estos aditivos en las reacciones de polimerización y espumación. 9 Resumen de la tesis en castellano Figura 0.5. Esquema seguido por la literatura científica actual sobre espumas de PU reforzadas con micro- y/o nanopartículas. Sin embargo, una comprensión clara del efecto de las partículas en la cinética de reacción es esencial, ya que afecta al proceso de formación de la espuma, pero también a la morfología final de la matriz de PU, y por lo tanto a las propiedades finales de la espuma resultante. Por esta razón, nuestra investigación se basa en un enfoque novedoso representado en la Figura 0.6, basado en el estudio simultáneo de la modificación de la cinética de la reacción y de la microestructura y propiedades físicas de los sistemas de RPU que contienen micro y/o nanopartículas. Figura 0.6. Novedoso esquema seguido en esta tesis para el estudio de espumas RPU reforzadas con nanopartículas. 2. Novedades sobre el estudio del proceso de espumado para espumas de RPU (Capítulo 4). Como punto de partida, se requirió el diseño de una metodología basada en la espectroscopia in situ FTIR que consiste en la monitorización y la posterior deconvolución de la región carbonílica (que contiene los productos de reacción). Esto permitiría saber cómo las principales reacciones se ven afectadas por la incorporación de partículas en la matriz de PU. Esta metodología se ha probado en espumas de RPU reforzadas con diferentes tipos de nanosilicas o nanoclays (5% en peso), demostrando cómo los diferentes tipos y superficies (hidrofóbicas Síntesisde nanocomposites Estructuracelular Propiedadesfinales de espumas de PU Nanoarcillas Nanosilicas Puro 5%A200 5%R812 5%R974 5%CNa+ 5%C30B 0 200 400 600 800 1000 Tamaño de celda (micras) 1640 1660 1680 1700 1720 1740 Absorbancia (a.u) Número de onda (cm-1) Síntesis de nanocomposites de espumas de PU Estructuracelular Cinética de reacción Propiedades finales Puro 5%A200 5%R812 5%R974 5%CNa+ 5%C30B 40 41 42 43 44 45 Conductividad térmica (mW/mK) Espuma PU 10 Chapter 0 y/o hidrofílicas) de las nanopartículas afectan a la cinética de la reacción de diferentes maneras. Además, se discute el efecto de la modificación de la cinética de reacción sobre la densidad, la estructura celular, la conductividad térmica y las propiedades mecánicas de las espumas de RPU. Además, en esta tesis se ha desarrollado una metodología novedosa llamada “Expandometría infrarroja”, para caracterizar simultáneamente la cinética de expansión (altura frente al tiempo y volumen frente al tiempo) y la evolución de la temperatura de la superficie de las espumas de PU. Esto se logra mediante el uso de la radiación infrarroja emitida por la reacción exotérmica de la espuma de PU. Para probar esta metodología, se ha elegido una formulación de RPU donde el agente espumante (agua) y/o el catalizador de espumado se cambian sistemáticamente. Finalmente, el proceso de espumado también se ha estudiado mediante radioscopia de rayos X, que mide la densidad relativa, el tamaño de celda y la densidad de nucleación celular en función del tiempo. Esta técnica se ha utilizado para comparar el comportamiento de espumado de dos series de espumas: las preparadas a partir de polioles funcionalizados con óxido de grafeno (GO) y las que contienen GO dispersado en el poliol mediante mezclado a alta cizalla (0.017 y 0.083% en peso). Los resultados obtenidos con esta técnica permiten demostrar que el uso de poliol funcionalizado con GO mejora mucho la estructura celular y también hace que los resultados sean más reproducibles. 3. Novedades sobre las propiedades físicas de las espumas RPU reforzadas con partículas (Capítulo 5). El efecto de las nanofibras de carbono (CNF) en el mecanismo de conducción de calor de un sistema de PUR espumado con agua se ha estudiado en profundidad, midiendo el coeficiente de extinción y modelizando la conductividad térmica. Se logra una mejora térmica del 2% con solo el 0,1% en peso de CNFs, ya que los CNFs como absorbentes de radiación infrarroja aumentando el coeficiente de extinción de la espuma y, como consecuencia, hay una clara reducción de la contribución radiativa. Este es un resultado notable teniendo en cuenta que las espumas en estudio tienen densidades relativamente altas (56 kg/m3) y, por lo tanto, la contribución radiativa tiene un peso bajo. Además, se ha estudiado el efecto de diferentes partículas sobre la conductividad térmica de una espuma de RPU con ciclopentano y agua a lo largo de aproximadamente tres años. En los primeros días después de la producción de espuma, las espumas que contienen partículas mejoran en gran medida la conductividad térmica en comparación con el material de referencia, principalmente debido a la disminución significativa en el tamaño de la celda promovida por las partículas que reducen la contribución radiativa. Sin embargo, después de un tiempo, esta primera mejora se pierde en muchos sistemas, porque la composición del gas dentro de las celdas evoluciona de manera diferente. Por primera vez, la diferente evolución de la conductividad térmica en las primeras medidas se ha relacionado con la temperatura de espumación alcanzada por cada sistema. Las partículas que alcanzan una temperatura de espumación más alta presentan una mayor evolución de la conductividad térmica, ya que esto 11 Resumen de la tesis en castellano causa un mayor gradiente de presión entre las celdas y la atmósfera, lo que genera una mayor velocidad de difusión del gas. 4. Novedades sobre la optimización de las propiedades físicas de las espumas de RPU reforzadas con partículas a partir de los resultados cinéticos (Chapter 6). Se ha investigado el efecto de los polioles funcionalizados con GO en el proceso de espumado, la morfología de la matriz de PU, la estructura celular, la conductividad térmica y las propiedades mecánicas de compresión de las espumas de RPU. En este estudio, el uso de partículas de GO unidas químicamente a las cadenas de poliol impide la aglomeración de nanofillers, evitando los problemas habituales de dispersión de partículas en nanocompuestos poliméricos. Solo con un 0,033% en peso de GO, la conductividad térmica se incrementa en alrededor del 4% sin modificaciones decisivas de otros aspectos importantes de la estructura celular, mientras que las propiedades mecánicas empeoran, lo que podría deberse a una modificación negativa de la morfología del polímero. Por este motivo, se lleva a cabo un estudio en profundidad del efecto de las partículas GO en el proceso de espumado y la morfología del PU utilizando la expandometría infrarroja, la espectroscopia de in situ FTIR y las medidas de la temperatura de espumado. En conclusión, la presencia de GO disminuye la conversión de isocianato y favorece la reacción de polimerización (generación de uretanos), dando lugar a una reticulatión inferior dentro de la espuma de RPU. Basándose en los resultados del estudio cinético, las propiedades mecánicas de la espuma con 0.033% en peso de GO podrían mejorarse finalmente cambiando la formulación de PU, mientras que al mismo tiempo la conductividad térmica se mantiene o mejora. 5. La novedad de la estancia de investigación en la Universidad de Manchester (Capítulo 7). El comportamiento de espumado de una nueva serie de TPUs sintetizados con diferentes contenidos de segmentos duros (HS) se ha analizado exhaustivamente en diferentes condiciones de saturación y espumación mediante un proceso de espumado por disolución de gas en un solo paso. El estudio también incluye la densidad y la estructura celular de las espumas resultantes. La mayoría de los trabajos anteriores publicados en la literatura han estudiado la espumación en TPU comerciales en los que la composición no suele conocerse, pero esta investigación permite relacionar la composición del TPU con el comportamiento de espumación. 0.5- Estructura y metodología de la tesis Esta tesis se escribe como un compendio de 8 artículos científicos, y se divide en 7 capítulos, en los que se incluyen los artículos (ver Tabla 0.1). Además, esta tesis cumple con los requisitos para ser acreditada con la Mención Internacional. Cada capítulo incluye la siguiente información: 12 Chapter 0 -Capítulo 1: Introducción. Se presenta la introducción, el marco de la investigación, los objetivos, las principales novedades, la estructura de la tesis y la lista de las publicaciones resultantes, comunicaciones en conferencias y proyectos de investigación. -Capítulo 2: Antecedentes y estado del arte. Revisa los conceptos básicos relacionados con las espumas de PU, como su química básica, morfología polimérica, estructura celular, materias primas comunes, proceso de espumación, propiedades y aplicaciones. Además, se describe el estado del arte del seguimiento del proceso de espumado de los materiales compuestos de espuma de PU. -Capítulo 3: Materiales. Describe los componentes de las formulaciones de PU utilizadas en este trabajo, como los polioles, isocianatos, catalizadores, surfactantes y agentes espumantes, así como los diferentes rellenos agregados a las formulaciones de PU. -Capítulo 4: Desarrollo de una metodología para seguir el proceso de espumado de espumas rígidas de poliuretano. Explica la metodología desarrollada durante esta tesis para estudiar el proceso de espumado de las espumas RPU que permite controlar la morfología química y, en consecuencia, las propiedades de la espuma. Esta metodología se basa en el uso de espectroscopia FTIR, expandometría infrarroja y radioscopia de rayos X, cuyos resultados son complementarios. Este capítulo contiene tres artículos que tratan cada uno de una técnica diferente para monitorear el proceso de formación de espuma. -Capítulo 5: Efecto de las partículas sobre la estructura celular y las propiedades físicas de las espumas rígidas de poliuretano. Este capítulo contiene dos artículos. El primero describe el efecto de las CNFs en las propiedades morfológicas, térmicas y mecánicas de las espumas de PUR espumadas con agua. El segundo artículo describe el efecto de diferentes tipos de partículas (talco, tierra de diatomeas y sílice no porosa) sobre el envejecimiento de la conductividad térmica durante tres años para las espumas de RPU que contienen ciclopentano y agua como agentes espumantes. -Capítulo 6: La optimización de la formulación de poliuretano a partir de los resultados cinéticos para obtener espumas de poliuretano rígidas reforzadas con óxido de grafeno con mejores propiedades térmicas y mecánicas. Contiene dos artículos. El primero investiga el efecto de un poliol funcionalizado con GO en la estructura celular y las propiedades termomecánicas de las espumas de RPU obtenidas. Además, se lleva a cabo un estudio cinético detallado utilizando la metodología desarrollada en esta tesis, que permite identificar el efecto del GO en la cinética de reacción y, por consiguiente, en la morfología del polímero final. El segundo artículo se basa en la información recopilada del estudio cinético anterior y describe cómo se puede ajustar la formulación de PU que contiene GO para mejorar las propiedades térmicas y mecánicas al mismo tiempo. -Capítulo 7: Estancia internacional de investigación: Espumas de poliuretano termoplástico. Incluye el trabajo realizado durante la estadía de investigación internacional y se ocupa de las espumas de TPU producidas por el proceso de espumado por disolución de gas. 13 Resumen de la tesis en castellano -Capítulo 8: Conclusiones y trabajo futuro. Resume las conclusiones finales obtenidas durante toda la tesis y también el trabajo futuro. 0.6- Publicaciones, conferencias y actividades complementarias Un total de 8 artículos científicos, enumerados en la Tabla 0.1, se han redactado durante el desarrollo de esta tesis. Varios de ellos han sido publicados y otros han sido enviados (pendientes de publicación), todos en revistas internacionales. El capítulo en el que se incorporan estos artículos también se especifica en la Tabla 0.1. Tabla 0.1. Artículos científicos de esta tesis. ARTÍCULOS CIENTÍFICOS Capítulo 1 M. Santiago-Calvo, J. Tirado-Mediavilla, J.L. Ruiz-Herrero, M.A. Rodríguez-Pérez, F. Villafañe. The effects of functional nanofillers on the reaction kinetics, microstructure, thermal and mechanical properties of water blown rigid polyurethane foams. Polymer 150 (2018) 138-149.https://doi.org/10.1016/j.polymer.2018.07.029 4 2 M. Santiago-Calvo, S. Pérez-Tamarit, J. Tirado-Mediavilla, F. Villafañe , M.A. Rodríguez- Pérez. Infrared expandometry: a novel methodology to monitor the expansion kinetics of cellular materials produced with exothermic foaming mechanisms. Polymer Testing 66 (2018) 383-393. https://doi.org/10.1016/j.polymertesting.2018.02.004 4 3 M. Santiago-Calvo, S. Pérez-Tamarit, P. Cimavilla-Román, V. Blasco, C. Ruiz, R. París, F. Villafañe , M.A. Rodríguez-Pérez. X-ray radioscopy validation of a polyol functionalized with graphene oxide for producing rigid polyurethane foams with improved cellular structures. European polymer journal 118 (2019) 404-411. https://doi.org/10.1016/j.eurpolymj.2019.06.012 4 4 M. Santiago-Calvo, J. Tirado-Mediavilla, J.C. Rauhe, L.R. Jensen, J.L. Ruiz-Herrero, F. Villafañe, M.A. Rodríguez-Pérez. Evaluation of the thermal conductivity and mechanical properties of water blown polyurethane rigid foams reinforced with carbon nanofibers. European polymer journal 108 (2018) 98-106. https://doi.org/10.1016/j.eurpolymj.2018.08.051 5 5 M. Santiago-Calvo, J. Tirado-Mediavilla, J.L. Ruiz-Herrero, M.A. Rodríguez-Pérez, F. Villafañe. Long-thermal conductivity aging of cyclopentane-water blown rigid polyurethane foams reinforced with different types of fillers. Polymer International (2019).https://doi.org/10.1002/pi.5893 5 6 M. Santiago-Calvo, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.A. Rodríguez-Pérez. Synthesis, characterization and physical properties of rigid polyurethane foams prepared with poly(propylene oxide) polyols containing graphene oxide. European Polymer Journal 97 (2017) 230–240.https://doi.org/10.1016/j.eurpolymj.2017.10.013 6 7 M. Santiago-Calvo, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.A. Rodríguez-Pérez. Improvement of thermal and mechanical properties by control of formulations in rigid polyurethane foams from polyols functionalized with graphene oxide. Journal of Applied Polymer Science, 136 (2019) 47474. DOI: 10.1002/app.47474 6 14 Chapter 0 ARTÍCULOS CIENTÍFICOS Capítulo 8 M. Santiago-Calvo, H. Naji, V. Bernardo, J. Martín-de León, A. Saiani, F. Villafañe, M.A. Rodríguez-Pérez. Synthesis, characterization, and foaming of thermoplastic polyurethane with different hard segment contents. Pending publication 7 Además, el trabajo desarrollado en esta tesis ha sido presentado en conferencias nacionales e internacionales, como se recoge en la Tabla 0.2. Table 0.2. Comunicaciones en conferencias relacionadas con el trabajo de esta tesis. CONTRIBUCIONES EN CONGRESOS 1 M. Santiago-Calvo, J. Tirado-Mediavilla, J.L. Ruiz-Herrero, M.A. Rodríguez-Pérez, F. Villafañe. FTIR studies to characterize the kinetics of nanocomposite polyurethane foams. XII Simposio de Investigadores Jóvenes RSEQ-Sigma Aldrich. November 2015, Barcelona, Spain. (Poster) 2 M. Santiago-Calvo, J. Tirado-Mediavilla, J.L. Ruiz-Herrero, M.A. Rodríguez-Pérez, F. Villafañe. An approach to follow the kinetics of nanocomposite polyurethane foam formation by FTIR spectroscopy. VIII Jornadas de Jóvenes Investigadores en Física Atómica y Molecular (J2IFAM). February 2016, Valladolid, Spain. (Talk) 3 M. Santiago-Calvo, J. Tirado-Mediavilla, J.L. Ruiz-Herrero, M.A. Rodríguez-Pérez, F. Villafañe. FTIR studies to characterize the kinetics of nanocomposite polyurethane foams. XIV Reunión del Grupo Especializado de Polímero (GEP). September 2016, Burgos, Spain. (Poster) 4 M. Santiago-Calvo1, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.A. Rodríguez-Pérez. Synthesis, properties and kinetic study of rigid polyurethane foams obtained from poly(propylene oxide) polyols functionalized with graphene oxide. 15th International Conference on Advances in Foam Materials & Technology (FOAMS). October 2017, Bayreuth, Germany. (Poster) 5 M. Santiago-Calvo, H. Naji, V. Bernardo, J. Martín-de León, A. Saiani, F. Villafañe, M.A. Rodríguez- Pérez. Synthesis, characterization and foaming of thermoplastic polyurethane synthesized with different soft/hard segment ratio and graphene nanoplatelet contents. 1st International congress PDFA, Polymers: Design, Function and Application. March 2018, Barcelona, Spain. (Poster) 6 M. Santiago-Calvo1, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.A. Rodríguez-Pérez. Synthesis, properties and kinetic study of rigid polyurethane foams obtained from poly(propylene oxide) polyols functionalized with graphene oxide. 5th International Conference on Cellular Materials (CellMAT). October 2018, Bad Staffelstein, Germany. (Talk) 7 M. Santiago-Calvo, H. Naji, V. Bernardo, J. Martín-de León, A. Saiani, F. Villafañe, M.A. Rodríguez- Pérez. Synthesis, characterization and foaming of thermoplastic polyurethane synthesized with different soft/hard segment ratio and graphene nanoplatelet contents. 5th International Conference on Cellular Materials (CellMAT). October 2018, Bad Staffelstein, Germany. (Poster) Ya que se presenta esta tesis para obtener un doctorado con Mención Internacional, se ha llevado a cabo una estancia en una institución extranjera. Todos los detalles están incluidos en la Tabla 0.3. 15 Resumen de la tesis en castellano Tabla 0.3. Estancias en otro centro de investigación durante la investigación. ESTANCIA DE INVESTIGACIÓN INTERNACIONAL Polymers & Peptides Research Group, The University of Manchester (UK). From June to August 2017 (3 months). Under the supervision of Prof. Alberto Saiani. Research work: Synthesis and characterization of thermoplastic polyurethane (TPU) in order to produce TPU foams by gas dissolution foaming process. Uno de los trabajos de investigación desarrollados durante la tesis ganó el Premio al Mejor Póster 2017 en la 15th International Conference on Advances in Foam Materials & Technology (FOAMS 2017), celebrada en Bayreuth (Alemania). Esto se enumera en la Tabla 0.4. Table 0.4. Trabajos de investigación premiados. PREMIOS M. Santiago-Calvo1, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.A. Rodríguez-Pérez. Synthesis, properties and kinetic study of rigid polyurethane foams obtained from poly(propylene oxide) polyols functionalized with graphene oxide. 5th International Conference on Cellular Materials (CellMAT). October 2017, Bayreuth, Germany. Best Poster selected by the Society of Plastics Engineers: Thermoplastic Materials & Foams Division. El conocimiento generado en el presente trabajo se ha desarrollado y aplicado en diferentes proyectos de investigación en colaboración con empresas privadas, recogidos en la Tabla 0.5. Este es un indicador decisivo de la importancia principal de esta investigación para importantes sectores industriales como la refrigeración, la automoción, la construcción, etc. Además, la investigación aquí desarrollada también se ha financiado con fondos públicos (también recogidos en la Tabla 0.5). 16 Chapter 0 Tabla 0.5. Proyectos privados y públicos que han financiado la investigación realizada en el presente trabajo. PROYECTOS DE INVESTIGACIÓN 1 Título del proyecto: “Improvement of short and long term thermal conductivity for rigid polyurethane foams through the use of specific additives and the control of kinetic reactions”. Fecha inicial: Dicienbre 2013 Fecha final: Noviembre 2016 Entidad financiadora: BOSH AND SIEMENS HOME APPLIANCES (Navarra). Investigador principal: Miguel Ángel Rodríguez Pérez 2 Título del proyecto: “Innovative additives for foams with better thermal insulation performance and fire behavior (NEOADFOAM)”. Fecha inicial: Septiembre 2015 Fecha final: Diciembre 2018 Entidad financiadora: MINECO, PROYECTO RETOS EN COLABORACIÓN CON TOLSA SA E IMDEA (Madrid). Investigador principal: Miguel Ángel Rodríguez Pérez 3 Título del proyecto: “Development of advanced cellular materials”. Fecha inicial: Abril 2016 Fecha final: Diciembre 2016 Entidad financiadora: Parque Científico. Universidad de Valladolid (Valladolid). Investigador principal: Miguel Ángel Rodríguez Pérez 4 Título del proyecto: “Analysis of a rigid polyurethane system: manufacturing, study of the reaction kinetics and characterization of the structure and properties”. Fecha inicial: Mayo 2016 Fecha final: Diciembre 2016 Entidad financiadora: REPSOL S.A (Madrid). Investigador principal: Miguel Ángel Rodríguez Pérez 5 Título del proyecto: “Comparative analysis of the cellular structure and reaction kinetics for three systems based on polyurethane”. Fecha inicial: Abril 2016 Fecha final: Diciembre 2018 Entidad financiadora: GRUPO ANTOLÍN INGENIERÍA (Burgos). Investigador principal: Miguel Ángel Rodríguez Pérez 6 Título del proyecto: “Study of polyurethane formulations by X-ray radioscopy”. Fecha inicial: Enero 2017 Fecha final: Junio 2017 Entidad financiadora: SOPREMA (France). Investigador principal: Miguel Ángel Rodríguez Pérez 7 Título del proyecto: “Research on applications of advanced carbon nanomaterials to improved performance of polyurethane matrices with interest in the automotive sector”. Fecha inicial: Abril 2017 Fecha final: 2019 Entidad financiadora: GRUPO ANTOLÍN INGENIERÍA (Burgos). Investigador principal: Miguel Ángel Rodríguez Pérez 8 Título del proyecto: “Study of polyisocyanurate foam formulations”. Fecha inicial: Enero 2019 Fecha final: 2019 Entidad financiadora: DUPONT (USA). Investigador principal: Miguel Ángel Rodríguez Pérez Finalmente, la Tabla 0.6 muestra las actividades adicionales desarrolladas durante esta tesis relacionadas con el campo de las espumas de PU. 17 24 Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane Mercedes Santiago Calvo CHAPTER 1: INTRODUCTION Chapter 1 1.1- Introduction From its early beginnings during and after World War II, plastic production has resulted in a rapid and continuous growth (Figure 1.1.A). The needs to develop synthetic polymers and relevant scientific advances have allowed the expansion of a complete set of polymers which are now usually recognized as "plastics". These polymers were easily commercialized due to the aim to lighten the weight of products and to offer economic alternatives to natural and metallic materials. In 2017, the world plastic production reached over 350 million tonnes of which 64.4 were manufactured in Europe [1]. One of the most versatile polymers is polyurethane (PU), since their unique properties have allowed its usage in a wide range of different industrial sectors, such as automotive, construction, furniture, packaging, medical, sports equipment, etc [2]. Figure 1.1.B shows that PU polymer represented 7.7% of the European plastic consumption in 2017 [1]. The business area of PU involves billions of U.S. dollars throughout the World and it is constantly growing, being estimated at $60.5 billion in 2017, and it is forecasted that it will reach over $79 billion in 2021 [3]. Figure 1.1. A) Global plastic production after World War II to the present (Adapted from [1]). B) European plastic demand by polymer types in 2017: Polyethylene (PE), Polypropylene (PP), Polyvinyl chloride (PVC), Polyurethane (PU), Polyethylene terephthalate (PET), Polystyrene (PS), others (Adapted from [1]). Because of the diversity of ways in which PUs can be synthesized, many different types of PU materials are available: flexible and rigid foams, elastomers, adhesives and sealants, coatings, among others [2]. Figure 1.2 clearly shows that the PU foams referred to thermoset PU foams cover the majority of the PU market, occupying 67% as flexible, rigid and molded foams, whereas the remaining 33% includes the rest of PU types, such as adhesives, elastomers, coating and other [4, 5]. The maximum usage (31% of PU market) is allocated to flexible foams, which are used principally in mattresses, cushions and others. They are followed by rigid foams (25% of PU market), which are widely used in thermal insulation of buildings, refrigerators, pipe insulation and refrigerated transport. Molded foams (11% of PU market) are mainly used in automotive and furniture industries. On the other hand, it is important to note that 27 Introduction elastomers (6% of PU market) include thermoplastic polyurethane (TPU) materials, which also enclose the TPU foams. Figure 1.2. World consumption of polyurethane by types in 2016 (Adapted from [4, 5]). The preceding data clearly show that most of the PUs industrially obtained are PU foams. The polymer foam market is divided into PU, polystyrene (PS), polyvinylchloride (PVC), phenolic, polyolefin, melamine, and other foams. PU foams are one of the largest segments of polymeric foam markets, representing more than 50% of the global volume of polymeric foams [6]. The cause of this high value is mainly due to the high applicability of PU foams, what is mainly based on the possibility of tailoring the final properties of the material by adequately changing the types or quantities of the initial components: isocyanate, polyol, surfactants, catalysts and blowing agents. In addition, the simplicity of the foaming technology has made these materials very competitive from an economic perspective. Thus, the investigation of these materials has attracted a lot of interest from their discovery in 1941. 1.2- Framework of this thesis This thesis, entitled “Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane”, has been carried out in the Cellular Materials Laboratory (CellMat) of the Department of Física de la Materia Condensada of the Universidad of Valladolid (UVa) during the last five years. The main target of the work is gaining new scientific knowledge in the field of rigid polyurethane (RPU) foams. This thesis has been supervised by professors Dr. Miguel Ángel Rodríguez-Pérez and Dr. Fernando Villafañe. CellMat was established in 1999 at the University of Valladolid by professors Dr. Jóse Antonio de Saja and Dr. Miguel Ángel Rodríguez-Pérez, after the defense of PhD thesis of Miguel Ángel Rodríguez Pérez dealing with the properties of polyolefins based foams, with the purpose of studying cellular materials [7]. As a result, at the beginning, the research activity was focused on the characterization of the microstructure and physical properties of commercial cellular 28 Chapter 1 polymers based on polyolefins, in order to evaluate their structure-properties relationships [8- 32]. In the following years, the research topics have been extended to the development of novel polymeric foams and their production routes. Over the years, a new research line started focused on the production and characterization of metal foams [33-37]. The work of our group has allowed to increase the scientific knowledge about the foaming mechanisms involved in the manufacturing processes, as well as about the production-structure-properties relationships of the materials studied [38, 39]. The final applications of the materials developed in CellMat have also been always considered, because one of the main principles of our laboratory is transferring knowledge and technology between the University and the industry. At present, five main research topics are being developed in CellMat laboratory, all supervised by Prof. Dr. Miguel Ángel Rodríguez-Pérez: nanocellular polymers [40-45], multifunctional cellular materials [46-48], cellular nanocomposites [49-52], bioplastic cellular materials [53- 57], and PU foams [58-61]. Furthermore, the research group has participated in many research projects funded both by public and private institutions, and it maintains collaborations with a significant number of research groups all around the world. As a result of this work, a total of 30 PhD theses and more than 200 scientific papers have been published so far on cellular materials, what makes CellMat laboratory an international reference in this field. The present thesis is part of the work developed in CellMat laboratory in the field of PU foams. This research line was started in 2014 with the thesis later defended by Sergio Estravis, dealing with thermoset PU foams [59]. One of the key reasons what led CellMat to establish this new topic was the very high demand of these materials in the global foams market, due to their diverse applications and economical weight. CellMat had already knowledge on the incorporation of nanoparticles to different foams, what is an interesting approach for the material industry, and in particular for the field of PU foams. These fillers can reinforce the PU matrix, increasing the mechanical properties, and also can act as nucleating agents and/or infrared blockers, thus improving their thermal insulating properties among other effects. Dr. Estravis carried out a detailed study on the effects of the addition of nanoclays to the microstructure, mechanical and thermal properties of commercial RPU foams, whose main applications were the thermal management in buildings, refrigeration systems and transport. Notably, the main conclusion of this work was that the thermal conductivity of the RPU foams may improve significantly with the presence of very low amounts of nanoclays (1 wt% or even less). At the same time, the presence of nanoclays in the PU systems slightly improved the elastic modulus and decreased the compression strength. The presence of nanoclays did not worsen the cellular structure, in fact smaller cell size were obtained, and therefore the observed decrease of the mechanical properties could be attributed to the alteration of the reaction kinetics involved in the PU foam formation causing a different polymer morphology. This indicated that the addition of particles affected the chemical modification of the PU matrix. However, the effect of particles on the main chemical reactions of PU foams (blowing and gelling) could not be studied in this work, but it was underlined as a main key for future works. This thesis was the first contact of CellMat laboratory with PU foams and it allowed establishing the methodology needed for manufacturing and characterizing PU foams in future 29 Introduction projects and theses. The main issues addressed by CellMat may be summarized as a tetrahedron (Figure 1.3), and Dr. Estravis thesis is a definite contribution to the four three vertexes of its base. Figure 1.3. Tetrahedron of materials referring to PU foams. The research work described in the present thesis aims to continue the research based on PU foams, but this time mainly focusing the research on the chemical aspects. The thermoset PU foams are produced by reactive foaming, and the balance between the reactions implicated in the PU foam formation is crucial in order to obtain the desired properties. As it was already observed by Dr. Estravis, the inclusion of nanoparticles into the RPU foams may also modify the kinetics of the reactions. This would give rise to unexpected consequences on the properties of the final PU matrix, since some foam properties may be improved, but others may be worsened. Therefore, understanding these effects becomes essential to make a better use of nanoparticles, and finally to get better cellular structure and foam properties. One of the main targets of the present thesis was the development of a methodology which allows to follow the chemical reactions occurring during the PU formation. The final purpose was controlling the PU formulation in order to obtain the best performance of the thermoset PU foam composite. Hence, with the beginning of the present thesis (end of 2014) CellMat started the study of PU foams from a chemical point of view. The first step was developing the methodology to monitor the thermoset PU foaming, which are key to understand the properties of final foam Applications Properties Structure Production 0.00 0.02 0.04 0.06 0.08 36 37 38 39 40 Thermal conductivity (mW/mK) Graphene oxide (wt%) 0.00 0.02 0.04 0.06 0.08 0.00 0.05 0.10 0.15 0.20 0.25 Relative Young's Modulus Graphene oxide (wt%) Foaming mechanisms Modelling 30 Chapter 1 and also to optimize them. This methodology is based on the use of the following complementary techniques: FTIR spectroscopy, infrared expandometry, temperature measurements, and X-ray radioscopy. Moreover, for the first time we could develop our own PU formulations, instead of the commercial formulations previously used in our laboratory. The design of an own formulation has been an additional way to control the final properties of the foams. Therefore, in this thesis we have been able to improve significantly the quality of our research on PU foams, having a better control on all the key aspects of the process, including the raw material used. In addition to thermoset PU foams, it is possible to manufacture cellular materials based on TPU. In order to gain knowledge in this field, a three-months stay was carried out from June to August 2017 in the Polymers & Peptides Research Group at the School of Materials of the University of Manchester (UK), under the supervision of Professor Alberto Saiani. This research group was established in 2004 by Profs. Aline F. Miller and Alberto Saiani and is focused on the characterization of polymer, biopolymer and peptide materials across different length scales, highlighting the study of TPU materials. Their research is mainly intended to the understanding of the chemical architecture-thermodynamic-structure-physical properties correlations in complex systems in order to control processes and products. During the research stay in Manchester, the work was focused on the synthesis and characterization of TPUs, which were later on foamed in CellMat by gas dissolution foaming. This is the first time that CellMat has achieved foaming of synthesized TPU. This research stay allows fulfilling the requirements to obtain a PhD with the International Mention. In summary, the work carried out in this thesis has generated knowledge in the two methods that can be used to produce foamed materials based on PU, that is, by reactive foaming (thermoset PU foams) and by gas dissolution foaming (TPU foams) (Figure 1.4). The first of them is the main part of the Thesis, since the study of thermoset PU foams was carried out during the first three years. The study of TPU foams is complementary to this and was mainly conducted during the three-months stay and further developed during the last year of this thesis. 31 Introduction Figure 1.4. Simplified scheme of the investigations developed in this thesis. 1.3- Objectives The aim of the present thesis is to acquire new knowledge about the thermoset PU foams (particularly RPU foams), focusing on the study of the foaming process, especially the reaction kinetics of the gas generation (blowing reaction) and polymerization (gelling reaction) when micro- and/or nanoparticles are incorporated into the PU matrix to improve the thermal and mechanical properties. The second important aspect to study is the effect that modifying the reaction could have on the structure and properties of the final materials. The third important aspect is to optimize the formulations of RPU systems containing particles using the kinetic information obtained. In order to achieve these general objectives, the following specific targets have been considered: Developing our own PU formulations in order to can control the final characteristics of the RPU systems. Polyurethane (PU) foams Thermoset Thermoplastic By reactive foaming By gas dissolution foaming Synthesis and characterization of thermoplastic PU (TPU) material Foaming and characterization of TPU foams Thermoset PU foams composites: •Synthesis and characterization •Thermal and mechanical properties •Study of foaming process, especially the reaction kinetics •Optimization of PU formulation 32 Chapter 1 Developing and validating the methodology which will allow to study the kinetics of the main chemical reactions (blowing and gelling) and the morphology of the polymeric matrix during the foaming process of RPU foams (Chapter 4). Developing and validating the methodology which will allow to study the expansion kinetics occurring in the production of RPU foams (Chapter 4). Applying other existing methodologies which will allow to study the internal microcellular structure during foaming process of RPU foams (Chapter 4). Applying the methodologies of monitoring the foaming process to RPU systems modified with microparticles and/or nanoparticles in order to evaluate the particles effect on the foaming process, especially on the reaction kinetics and final morphology of the polymer matrix (Chapter 6). Establishing correlations between the reaction kinetics, the final morphology of the polymer matrix, the cellular structure and the physical properties of the RPU systems modified with microparticles and/or nanoparticles (Chapter 6). Optimizing the RPU formulation modified with microparticles and/or nanoparticles from kinetic information obtained (Chapter 6). For RPU systems modified with microparticles and/or nanoparticles, establishing correlations between the foaming temperature and the evolution of their thermal conductivity with time (Chapter 5). Modeling of the thermal conductivity of RPU foams modified with microparticles and/or nanoparticles to study the particles effect (Chapter 5). Other objective of the present thesis is to study the foaming behavior of TPU materials using gas dissolution foaming in order to generate knowledge about TPU foams (Chapter 7). 1.4- Main novelties 1. The main novelty of the thesis. So far, the current scientific literature [59, 62-75], describing PU foams reinforced with micro- and/or nanoparticles, follows the approach depicted in Figure 1.5, which is mainly focused on the production and characterization of the foams, without paying particular attention to the effect of these additives on the gelling and blowing reactions. Synthesis of PU foams Cellular structure Final properties nanocomposites 33 Introduction One of the research works developed during the thesis won the 2017 Best Poster Award at the 15th International Conference on Advances in Foam Materials & Technology (FOAMS 2017), celebrated in Bayreuth (Germany). This is listed in Table 1.4. Table 1.4. Research works awarded. AWARDS M. Santiago-Calvo1, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.A. Rodríguez-Pérez. Synthesis, properties and kinetic study of rigid polyurethane foams obtained from poly(propylene oxide) polyols functionalized with graphene oxide. 5th International Conference on Cellular Materials (CellMAT). October 2017, Bayreuth, Germany. Best Poster selected by the Society of Plastics Engineers: Thermoplastic Materials & Foams Division. The knowledge generated in the present work has developed and applied in different research projects in collaboration with private companies, collected in Table 1.5. This is a decisive indicator of the main importance of this research for important industrial sectors such as refrigeration, automotive, construction, etc. Moreover, the research herein developed has also been financed by public funding (also collected in Table 1.5). Table 1.5. Private and public projects which have financed the investigation performed in the present work. RESEARCH PROJECTS 1 Project Title: “Improvement of short and long term thermal conductivity for rigid polyurethane foams through the use of specific additives and the control of kinetic reactions”. Initial date: December 2013 Final date: November 2016 Financial entity: BOSH AND SIEMENS HOME APPLIANCES (Navarra). Principal researcher: Miguel Ángel Rodríguez Pérez 2 Project Title: “Innovative additives for foams with better thermal insulation performance and fire behavior (NEOADFOAM)”. Initial date: September 2015 Final date: December 2018 Financial entity: MINECO, PROYECTO RETOS EN COLABORACIÓN CON TOLSA SA E IMDEA (Madrid). Principal researcher: Miguel Ángel Rodríguez Pérez 3 Project Title: “Development of advanced cellular materials”. Initial date: April 2016 Final date: December 2016 Financial entity: Parque Científico. Universidad de Valladolid (Valladolid). Principal researcher: Miguel Ángel Rodríguez Pérez 4 Project Title: “Analysis of a rigid polyurethane system: manufacturing, study of the reaction kinetics and characterization of the structure and properties”. Initial date: May 2016 Final date: December 2016 Financial entity: REPSOL S.A (Madrid). Principal researcher: Miguel Ángel Rodríguez Pérez 40 Chapter 1 RESEARCH PROJECTS 5 Project Title: “Comparative analysis of the cellular structure and reaction kinetics for three systems based on polyurethane”. Initial date: April 2016 Final date: December 2018 Financial entity: GRUPO ANTOLÍN INGENIERÍA (Burgos). Principal researcher: Miguel Ángel Rodríguez Pérez 6 Project Title: “Study of polyurethane formulations by X-ray radioscopy”. Initial date: January 2017 Final date: June 2017 Financial identity: SOPREMA (France). Principal researcher: Miguel Ángel Rodríguez Pérez 7 Project Title: “Research on applications of advanced carbon nanomaterials to improved performance of polyurethane matrices with interest in the automotive sector”. Initial date: April 2017 Final date: 2019 Financial entity: GRUPO ANTOLÍN INGENIERÍA (Burgos). Principal researcher: Miguel Ángel Rodríguez Pérez 8 Project Title: “Study of polyisocyanurate foam formulations”. Initial date: January 2019 Final date: 2019 Financial entity: DUPONT (USA). Principal researcher: Miguel Ángel Rodríguez Pérez Finally, Table 1.6 shows additional activities developed during this thesis related to the field of PU foams. Table 1.6. Additional activities carried out during this thesis. RESEARCH ARTICLES S. Estravís, J. Tirado-Mediavilla, M. Santiago-Calvo, J.L. Ruiz-Herrero, M.A. Rodríguez-Pérez, F. Villafañe. Rigid polyurethane foams with infused nanoclays: Relationship between cellular structure and thermal conductivity. European Polymer Journal 80 (2016) 1–15. https://doi.org/10.1016/j.eurpolymj.2016.04.026 P. Acuña, M. Santiago-Calvo, F. Villafañe, M.A. Rodríguez-Perez, J. Rosas, De-Yi Wanga. Impact of Expandable Graphite on Flame Retardancy and Mechanical Properties of Rigid Polyurethane Foam. Polymer composites (2018). https://doi.org/10.1002/pc.25127 P. Acuña, Z. Li, M. Santiago-Calvo, F. Villafañe, M.A. Rodríguez-Perez, De-Yi Wanga. Influence of the characteristics of expandable graphite on the morphology, thermal properties, fire behaviour and compression performance of a rigid polyurethane foam. Polymers 11 (2019) 168. https://doi.org/10.3390/polym11010168 A. Galakhova, M. Santiago-Calvo, J. Tirado-Mediavilla, F. Villafañe, M.A. Rodríguez-Pérez and G. Riess. Identification and Quantification of Cell Gas Evolution in Rigid Polyurethane Foams by Novel GCMS Methodology. Polymers 11 (2019) 1192. https://doi.org/10.3390/polym11071192 41 Introduction CONTRIBUTIONS TO CONGRESS P. Cimavilla-Román, S. Pérez-Tamarit, M. Santiago-Calvo, M.A. Rodríguez-Pérez. In situ analysis, by X-ray radioscopy, of the foaming process of aerogel-polyurethane cellular composites. 10th European School on Molecular Nanoscience (ESMolNa2017). May 2017, El Escorial, Madrid, Spain. (Talk) P. Cimavilla-Román, S. Pérez-Tamarit, M. Santiago-Calvo, M.A. Rodríguez-Pérez. In-situ physicochemical analysis of the foaming process of aerogel-rigid polyurethane composite foams. 12th European Conference on Foams, Emulsions and Applications (EUFOAM). July 2018, Liege (Belgium). (Talk) ACADEMIC ACTIVITIES Final Master Project: Characterization and in-situ analysis of the cellular structure and foaming process of aerogel-polyurethane composites by means of X-ray radioscopy and tomography. Author: P. Cimavilla-Román. Tutors: M.A. Rodríguez-Pérez, S. Pérez-Tamarit, M. Santiago-Calvo. Document published: 2017. University qualification: Máster en Nanociencia y Nanotecnología Molecular. http://uvadoc.uva.es/handle/10324/26203 DIVULGATION TALKS Cycle of conferences: Woman Researchers of the University of Valladolid in the Adventure of Science and Technology. -In 2016: Study of formation reactions of polyurethane foam nanocomposites by FTIR spectroscopy. (Talk) -In 2017: Polyurethane foams from polyols functionalized with graphene oxide. (Talk) -In 2018: Synthesis and characterization of thermoplastic polyurethanes (TPUs) with different contents of hard segments in order to prepare TPU foams. (Talk) RESEARCH PROJECTS Project Title: “Evolution: The Electric Vehicle revolution enhabled by advanced materials highly into ligh”. Initial date: October 2013 Duration: 1 year Financial entity: EUROPEAN UNION (FPVII). Principal researcher: Miguel Ángel Rodríguez Pérez Project Title: “ACTIBIOPACK: Envasado activo y biodegradable de champiñón en fresco y carne fresca”. Initial date: October 2014 Duration: 3 months Financial entity: MINECO. Principal researcher: Miguel Ángel Rodríguez Pérez 1.7- References [1] Plastics Europe (the Association of Plastics Manufacturers in Europe). https://www.plasticseurope.org/en/resources/market-data. [2] M. Szycher, Szycher's Handbook of Polyurethanes, Second ed., CRC Press Boca Raton, Florida, USA, 2012. [3] Polyurethane Global Market Size Forecast 2021. Available online: https://www.statista.com/statistics/720449/global-polyurethane-market-size-forecast/ (accessed on 22 February 2019). [4] Plastics Insight: https://www.plasticsinsight.com/resin-intelligence/resinprices/polyurethane/. [5] Statista: https://www.statista.com/statistics/615265/distribution-of-polyurethane- consumption-worldwide-by-end-use/. 42 Chapter 1 [6] Polymer Foams Market Expected to Consume 25.3 Million Tonnes by 2019. http://www.smithersrapra.com/news/2014/may/polymer-foam-market-to-consume-25-3- million-tonnes, 2014 (accessed 16 July 2019). [7] M.Á. Rodríguez-Pérez, PhD Thesis: Thermal and mechanical properties of polyolefin foams University of Valladolid, 1999. [8] M.A.Rodríguez-Pérez, O.Alonso, J. Souto, J.A.d. Saja, Thermal Conductivity of Crosslinked Closed Cell Polyolefin Foams, Polymer Testing, 16 (1997) 287-298. [9] M.A. Rodríguez-Pérez, S. Rodríguez-Llorente, J.A.d. Saja, Dynamic Mechanical Properties of Polyolefin Foams Studied by DMA Techniques, Polym. Eng. Sci., 37 (1997) 959-966. [10] M.A.Rodríguez-Pérez, A.Duijsens, J.A.d. Saja, Effect of Addition of EVA on the Technical Properties of Extruded Foam Profiles of Low-Density Polyethylene/EVA Blends, J.Applied Polymer Science (1998) 1237-1244 [11] M.A.Rodriguez, S.Diez, J.A.d. Saja, The recovery behaviour of crosslinked closed cell polyolefin foams, Polymer Eng. & Sci, (1998) 831-838. [12] M.A.Rodríguez-Pérez, O.Alonso, A.Duijsens, J.A.d. Saja, Thermal Expansion of Crosslinked Closed-Cell Polyethylene Foams, J.Polymer Sci. , 36 (1998) 2587-2596. [13] M.A. Rodríguez-Pérez, J.A.D. Saja, The effect of blending on the physical properties of crosslinked closed cell polyrthylene foams, Cellular Polymers 18 (1999) 1-20. [14] M.A. Rodríguez‐Pérez, O. Almanza, J.A.D. Saja, Anomalous thickness increase in crosslinked closed cell polyolefin foams during heat treatments. , Journal of Applied Polymer Science, 73 (1999) 2825–2835. [15] J.I. Velasco, A.B. Martinez, D. Arencon, M.A. Rodriguez-Perez, J.A.D. Saja, Application of instrumented falling dart impact to the mechanical characterization of thermoplastic foams, Journal of Materials Science, 34 (1999) 431–438. [16] O.A. Almanza, M.A. Rodríguez-Pérez, J.A.D. Saja, Prediction of the radiation term in the thermal conductivity of crosslinked closed cell polyolefin foams, Journal of Polymer Science Part B: Polymer Physics, 38 (2000) 993–1004. [17] M.A. Rodriguez-Perez, J.I. Velasco, D. Arencon, O. Almanza, J.A.D. Saja, Mechanical Characterization of Closed-CellPolyolefin Foams, Journal of Applied Polymer Science, 75 (2000) 156 –166. [18] M.A. Rodríguez-Pérez, J.A.d. Saja, Dynamic mechanical analysis applied to the characterisation of closed cell polyolefin foams, Polymer Testing, 19 (2000) 831–848. [19] J.I. Velasco, A.B. Martinez, D. Arencon, M.A. Rodriguez-Perez, J.A.d. Saja, Rigidity characterisation of flexible foams by falling dart robound tests, Cellular Polymers, 19 (2000) 115–133. [20] O. Almanza, M.A. Rodriguez-Perez, J.A.D. Saja, The microestructure of polyethylene foams produced by a nitrogen solution process, Polymer, 42 (2001) 7117–7126. [21] O. Almanza, L.O.A.y. Rabago, M.A.Rodriguez-Perez, A. Gonzalez, J.A.d. Saja, Strucutreproperty relationships in polyolefin foams, Journal of Macromolecular Science, Part B 40 (2001) 603–613. [22] J.A. Martinez-Diez, M.A. Rodriguez-Perez, J.A.d. Saja, L.O.A.y. Rabago, O.A. Almanza, The thermal conductivity of a polyethylene foam block produced by a compression molding process, Journal of Cellular Plastics, 37 (2001) 21–42. [23] N.J. Mills, M.A. Rodriguez-Perez, Modelling the Gas-loss Creep Mechanism in EVA Foam from Running Shoes, Cellular Polymers, 20 (2001) 79–100. [24] M.A. Rodriguez-Perez, O. Almanza, J.L.D. Valle, A. Gonzalez, J.A.D. Saja, Improvement of the measurement process used for the dynamic mechanical characterization of polyolefin foams in compression, Polymer Testing, 20 (2001) 253–267. [25] M.A. Rodriguez-Perez, The effect of chemical composition, density and cellular structure on the dynamic mechanical response of polyolefin foams, Cellular Polymers, 21 (2002) 117– 136. 43 Introduction [26] M.A. Rodriguez-Perez, J.I. Gonzalez-Peña, N. Witten, J.A.D. Saja, Morphology of semicrystalline foams based on polyethylene, Journal of Macromolecular Science: Part B – Physics 41 (2002) 761–775. [27] M.A. Rodriguez-Perez, J.I. Gonzalez-Peña, N. Witten, J.A.D. Saja, The Effect of Cell Size on the Physical Properties of Crosslinked Closed Cell Polyethylene Foams Produced by a High Pressure Nitrogen Solution Process, Cellular Polymers, 21 (2002) 165–194. [28] O. Almanza, M.A. Rodríguez-Pérez, J.A. de Saja, Applicability of the Transient Plane Source Method To Measure the Thermal Conductivity of Low-Density Polyethylene Foams, Journal of Polymer Science: Part B: Polymer Physics, 42 (2004) 1226–1234. [29] O. Almanza, Y. Masso-Moreu, N.J. Mills, M. A.Rodriguez-Perez, Thermal expansion coefficient and bulk modulus of polyethylene closed-cell foams, Journal of Polymer Science, Part B: Polymer Physics, 42 (2004) 3741–3749. [30] O. Almanza, M.A. Rodriguez-Perez, J.A.d. Saja, Measurement of the thermal diffusivity andspecific heat capacity of polyethylene foamsusing the transient plane source technique, Polymer International 2004, 53, , 53 (2004) 2038–2044. [31] R.A. Campo-Arnáiz, M.A. Rodríguez-Pérez, B. Calvo, J.A. de Saja, Extinction coefficient of polyolefin foams, Journal of Polymer Science Part B: Polymer Physics, 43 (2005) 1608-1617. [32] O. Almanza, M.A. Rodríguez-Pérez, B. Chernev, J.A.d. Saja, P. Zipper, Comparative study on the lamellar structure of polyethylene foams, European Polymer Journal, 41 (2005) 599–609. [33] J.A. Reglero-Ruiz, PhD Thesis: Manufacture and characterization of aluminium foams: Applications in the aeronautical sector, University of Valladolid, 2007. [34] E. Solórzano, PhD Thesis: Aluminium foams: Foaming process, cellular structure and properties, University of Valladolid, 2008. [35] J.A. Reglero, M.A. Rodriguez-Perez, D. Lehmhus, M. Windmann, J.A.d. Saja, A. Fernandez, An experimental study on the inhomogeneities of aluminum foams measuring the thermal conductivity by using the transient plane source method, Materials Science Forum 480-481 (2005) 133–138. [36] E. Solórzano, M.A. Rodríguez-Pérez, J.A. Reglero, J.A.d. Saja, Density gradients in aluminium foams: characterisation by computed tomography and measurements of the effective thermal conductivity, Journal of Materials Science, 42 (2007) 2557–2564. [37] E. Solórzano, J.A. Reglero, M.A. Rodríguez-Pérez, J.A.d. Saja, M.L. Rodríguez-Méndez, Improvement of the foaming process for 4045 and 6061 aluminium foams by using the Taguchi methodology, Journal Material Science, 42 (2007) 7227–7238. [38] C. Saiz-Arroyo, PhD thesis: Fabricación de Materiales Celulares Mejorados Basados en Poliolefinas. Relación procesado-composición-estructura-propiedades, University of Valladolid, 2012. [39] S. Pardo, PhD Thesis: X-Ray Imaging Applied to the Characterization of Polymer Foams’ Cellular Structure and its Evolution, University of Valladolid, 2014. [40] J. Pinto, PhD Thesis: Fabrication and characterization of nanocellular polymeric materials from nanostructured polymers., University of Valladolid, 2014. [41] B. Notario-Collado, PhD Thesis: Fabrication and characterization of the physical properties of nanocellular polymers: The transition from the micro to the nanoscale., University of Valladolid, 2016. [42] J.A.R. Ruiz, M. Dumon, J. Pinto, M.A. Rodriguez-Perez, Low-Density Nanocellular Foams Produced by High-Pressure Carbon Dioxide, Macromol. Mater. Eng., 296 (2011) 752-759. [43] J. Pinto, M. Dumon, M. Pedros, J. Reglero, M.A. Rodriguez-Perez, Nanocellular CO2 foaming of PMMA assisted by block copolymer nanostructuration, Chem. Eng. J., 243 (2014) 428-435. [44] B. Notario, J. Pinto, E. Solorzano, J.A. de Saja, M. Dumon, M.A. Rodriguez-Perez, Experimental validation of the Knudsen effect in nanocellular polymeric foams, Polymer, 56 (2015) 57-67. 44 Chapter 1 [45] J. Martin-de Leon, V. Bernardo, M.A. Rodriguez-Perez, Key Production Parameters to Obtain Transparent Nanocellular PMMA, Macromol. Mater. Eng., 302 (2017) 5. [46] B. Notario, A. Ballesteros, J. Pinto, M.A. Rodriguez-Perez, Nanoporous PMMA: A novel system with different acoustic properties, Mater. Lett., 168 (2016) 76-79. [47] J.A. Reglero-Ruiz, C. Saiz-Arroyo, M. Dumon, M.A. Rodríguez-Perez, L. Gonzalez, Production, cellular structure and thermal conductivity ofmicrocellular (methyl methacrylate)– (butyl acrylate)–(methyl methacrylate) triblock copolymers, Polymer International, 60 (2011) 146-152. [48] M. Dumon, J.A.R. Ruiz, J.P. Sanz, M.A.R. Perez, J.-M. Tallon, M. Pedros, E. Cloutet, P. Viot, Block Copolymer-Assisted Microcellular Supercritical CO2 Foaming of Polymers and Blends, Cellular Polymers, 31 (2012) 207-222. [49] J.I. Velasco, M. Antunes, O. Ayyad, J.M. López-Cuesta, P. Gaudon, C. Saiz-Arroyo, M.A. Rodríguez-Pérez, J.A. de Saja, Foaming behaviour and cellular structure of LDPE/hectorite nanocomposites, Polymer, 48 (2007) 2098-2108. [50] C. Saiz-Arroyo, M.A. Rodriguez-Perez, J.I. Velasco, J.A. de Saja, Influence of foaming process on the structure-properties relationship of foamed LDPE/silica nanocomposites, Compos. Pt. B-Eng., 48 (2013) 40-50. [51] E. Laguna-Gutierrez, J. Escudero, V. Kumar, M.A. Rodriguez-Perez, Microcellular foaming by using subcritical CO2 of crosslinked and non-crosslinked LDPE/clay nanocomposites, Journal of Cellular Plastics, 54 (2018) 257-282. [52] M. Ardanuy, J.I. Velasco, M. Antunes, M.A. Rodriguez-Perez, J.A. de Saja, Structure and Properties of Polypropylene/Hydrotalcite Nanocomposites, Polymer Composites, 31 (2010) 870-878. [53] M.A. Rodriguez-Perez, R.D. Simoes, C.J.L. Constantino, J.A.D. Saja, Structure and Physical Properties of EVA/Starch Precursor Materials for Foaming Applications, Journal of Applied Polymer Science, 121 (2011) 2324-2330. [54] M.A. Rodriguez-Perez, R.D. Simoes, S. Roman-Lorza, M. Alvarez-Lainez, C. Montoya-Mesa, C.J.L. Constantino, J.A.d. Saja, Foaming of EVA/Starch Blends: Characterization of the Structure, Physical Properties and Biodegradability, Polymer Enginnering and Science, 52 (2012) 62-70. [55] A. Lopez-Gil, F. Silva-Bellucci, D. Velasco, M. Ardanuy, M.A. Rodriguez-Perez, Cellular structure and mechanical properties of starch-based foamed blocks reinforced with natural fibers and produced by microwave heating, Industrial Crops and Products, 66 (2015) 194-205. [56] A. López-Gil, PhD Thesis: Development of environmentally friendly cellular polymers for packaging and structural applications. Study of the relationship cellular structure-mechanical properties, University of Valladolid, 2016. [57] H. Ventura-Casellas, PhD Thesis: Development of new lightweight green composites reinforced with nonwoven structures of flax fibers, University of Valladolid, 2017. [58] S. Pardo-Alonso, E. Solórzano, S. Estravís, M.A. Rodríguez-Perez, J.A. de Saja, In situ evidence of the nanoparticle nucleating effect in polyurethane–nanoclay foamed systems, Soft Matter, 8 (2012) 11262. [59] S. Estravis, PhD Thesis: Cellular nonocomposites based on rigid polyurethane and nanoclays: fabrication, caharacterization and modelling of the mechanical and thermal properties, University of Valladolid, 2014. [60] S. Pardo-Alonso, E. Solórzano, M.A. Rodriguez-Perez, Time-resolved X-ray imaging of nanofiller-polyurethane reactive foam systems, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 438 (2013) 119-125. [61] S. Pardo-Alonso, E. Solórzano, L. Brabant, P. Vanderniepen, M. Dierick, L. Van Hoorebeke, M.A. Rodríguez-Pérez, 3D Analysis of the progressive modification of the cellular architecture in polyurethane nanocomposite foams via X-ray microtomography, European Polymer Journal, 49 (2013) 999–1006. 45 Introduction [62] L. Madaleno, R. Pyrz, A. Crosky, L.R. Jensen, J.C.M. Rauhe, V. Dolomanova, A.M.M.V. de Barros Timmons, J.J. Cruz Pinto, J. Norman, Processing and characterization of polyurethane nanocomposite foam reinforced with montmorillonite–carbon nanotube hybrids, Composites Part A: Applied Science and Manufacturing, 44 (2013) 1–7. [63] X. Cao, L. James Lee, T. Widya, C. Macosko, Polyurethane/clay nanocomposites foams: processing, structure and properties, Polymer, 46 (2005) 775–783. [64] G. Harikrishnan, T.U. Patro, D.V. Khakhar, Polyurethane Foam-Clay Nanocomposites: Nanoclays as Cell Openers, Industrial and Engineering Chemistry Research, 45 (2006) 7126– 7134. [65] M.M.A. Nikje, Z.M. Tehrani, Thermal and mechanical properties of polyurethane rigid foam/modified nanosilica composite, Polymer Engineering & Science, 50 (2010) 468–473. [66] M.M.A. Nikje, Z.M. Tehrani, Polyurethane Rigid Foams Reinforced by Doubly Modified Nanosilica, Journal of Cellular Plastics, 46 (2010) 159–172. [67] T. Widya, C. Macosko, Nanoclay‐Modified Rigid Polyurethane Foam, Journal of Macromolecular Science, Part B: Physics, 44 (2005) 897-908. [68] M.C. Saha, M.E. Kabir, S. Jeelani, Enhancement in thermal and mechanical properties of polyurethane foam infused with nanoparticles, Materials Science and Engineering: A, 479 (2008) 213-222. [69] M.M.A. Nikje, Z.M. Tehrani, The Effects of Functionality of the Organifier on the Physical Properties of Polyurethane Rigid Foam/Organified Nanosilica, Designed Monomers and Polymers, 14 (2012) 263-272. [70] M.M.A. Nikje, Z.M. Tehrani, Novel Modified Nanosilica-Based on Synthesized Dipodal Silane and Its Effects on the Physical Properties of Rigid Polyurethane Foams, Designed Monomers and Polymers, 13 (2012) 249-260. [71] M.E. Kabir, M.C. Saha, S. Jeelani, Effect of ultrasound sonication in carbon nanofibers/polyurethane foam composite, Materials Science and Engineering: A, 459 (2007) 111-116. [72] C. Caglayan, I. Gurkan, S. Gungor, H. Cebeci, The Effect of CNT-Reinforced Polyurethane Foam Cores to Flexural Properties of Sandwich Composites, Composites: Part A, (2018). [73] S.-X. Wang, H.-B. Zhao, W.-H. Rao, S.-C. Huang, T. Wang, W. Liao, Y.-Z. Wang, Inherently flame-retardant rigid polyurethane foams with excellent thermal insulation and mechanical properties, Polymer, 153 (2018) 616-625. [74] S. Sathiyamoorthy, G. Girijakumrai, P. Kannan, K. Venugopal, S.T. Shanmugam, P. Veluswamy, K.D. Wael, H. Ikeda, Tailoring the functional properties of polyurethane foam with dispersions of carbon nanofiber for power generator applications, Applied Surface Science, (2018). [75] N. Nazeran, J. Moghaddas, Synthesis and characterization of silica aerogel reinforced rigid polyurethane foam for thermal insulation application, Journal of Non-Crystalline Solids, 461 (2017) 1-11. 46 Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane Mercedes Santiago Calvo CHAPTER 2: BACKGROUND AND STATE OF THE ART Chapter 2 2.1- Introduction Polyurethane (PU) foams (referred to thermoset PU foams) are discussed in more detail in most sections of this chapter, since the largest part of this thesis is centered in them. The following sections summarize these aspects about PU foams: - Their discovery (section 2.2). - Their basic chemistry (main reactions), polymer morphology and cellular structure (section 2.3). - The common raw materials used to produce them (section 2.4). - The steps of foaming process and cell degeneration mechanisms (section 2.5). - Their properties and applications (section 2.6). - PU foam composites (section 2.7). - The monitoring of foaming process of PU foams (section 2.8). The last section (2.9) is focused on thermoplastic polyurethane (TPU) foams, whose study was carried during the international stay. It is also important to mention that a detailed state of the art of the different topics considered during the thesis is discussed in each paper included in chapters 4 to 7. 2.2- Discover of polyurethane: Polyurethane foams In 1849, Charles Adolphe Wurtz discovered the reaction of urethane formation (-NH-CO-O-) by reacting aliphatic monofunctional isocyanates with alcohols [1]. However, it was not until the 1930s that Professor Dr. Otto Bayer and his co-workers at the laboratories of I.G. Farben in Leverkusen, Germany rediscovered and patented the novel polymerization reaction called polyaddition reaction for the synthesis of PUs by reacting diisocyanates and compounds containing at least two hydroxyl and/or amine groups [2]. Initially, their work was focused on the production of fibers to be used for textile applications, but unfortunately these fibers based on PU were not long enough for their initial purpose. This is why they decided to study PU moldable masses, what gave instead products containing undesired bubbles, as a result of the use of an impure polyester component which contained carboxyl groups, which react with isocyanates forming amides and CO2 [3]. Therefore, it was by chance that Professor Dr. Otto Bayer and his team obtained thermoset PU foams in 1941. Subsequently, they used a small amount of water to obtain PU foams with the desired amount of CO2 [3]. Figure 2.1 shows a photograph of Bayer showing his discovery of PU foam. The first patent of a flexible polyurethane (FPU) foam preparation was assigned to Zaunbrecher and Barth in 1942 [4]. But the commercial production of FPU foam had to wait until the early fifties, and it was based on the reaction between toluene diisocyanate (TDI) and polyester polyols. These reactants were also used to produce other types of PUs, such as rigid foams, gum rubber, or elastomers. In 1956 DuPont launched poly(tetramethylene glycol) (PTMG), the first commercial polyether polyol. Polyether polyols became more popular than 49 Background and state of the art As a conclusion, a number of side reactions may take place, depending predominantly on the reaction conditions, such as temperature, the ratio between polyol and isocyanate, and the presence of catalysts. Moreover, these side products may suppose a supplementary source of crosslinking to the PU network. Since the PU formulations used in the present thesis are for rigid foams in which the excess of isocyanate is not high, the amount of products corresponding to the side reactions is very small in comparison with that of the main reactions (gelling and blowing) and thus the side reactions can be despised. 2.3.2- Polymer morphology PU foams present a complex polymer morphology composed of alternating hard and soft segments. Soft segments (SS) are composed of polyol chains such as polyether or polyester polyols, which offer flexibility and elastomeric properties at room temperature. Hard segments (HS) are derived from the main simultaneous reactions of isocyanate with polyol and water, which provide rigidity [17]. The polymerization reaction creates progressively a chemical crosslinking network based on urethane structures, while the blowing reaction generates CO2 and polyurea-based HS which are chemical incompatibility with polyol SS and tends to form microphase separation where polyureas are aggregated via hydrogen bonding [17, 18]. The urea groups in foams have stronger specific hydrogen-bonding interaction and high stiffness compared to those of the urethane groups, making them more appropriate for microphase separation as compared to the urethane groups [17]. For this reason the hydrogen-bonded urea, in particular ordered urea (bidentate urea), is frequently used for estimating the degree of microphase separation [19, 20]. The formation of the microphase separation into urea HS and polyol SS depends on the kinetics and thermodynamic effects in foam formation, and it is favoured in flexible foams [21, 22]. On the one hand, flexible foams have a wide meshed cross-linked structure formed by the reactions of low functionalized long-chain polyols with water and diisocyanate [23]. Moreover, the flexible foams are characterized by the phase separation into urea HS and polyol SS (Figure 2.2.A). Consequently, the polymer morphology of the flexible foams consists of an elastic network of strong covalent cross-links in the soft domains and weaker physical cross-links in the hard domains. On the other hand, rigid foams have a close meshed cross-linked structure generated by the reaction between highly functionalized low-chain polyols with polyfunctional isocyanates [23]. Moreover, the excess of isocyanate used in RPU or PIR foams generates additional cross-linking points such as allophanate, biuret or isocyanurate groups (Figure 2.2.B) [23]. 56 Chapter 2 Figure 2.2. A) Polymer morphology of flexible polyurethane foam and B) rigid polyurethane foam [24]. 2.3.3. Cellular structure and density RPU and FPU foams differ not only on the polymer morphology but also on the type of cellular structure. Two types of cellular structures are present in PU foams: closed cell structure and open cell structure. The cellular structure is predominantly closed for rigid foams (Figure 2.3.A), which means that the cells present cell walls and contain the dispersed gas completely enclosed. However, in FPU, the cellular structure is predominantly open (Figure 2.3.B), which means that the cell walls are interconnected and are filled with air. These types of cellular structures depend on the starting components and their amounts, as it will be explained in section 2.4. In addition to the open or closed cellular structure, the cell size is very relevant since many applications require either large or small cells. The cell shape (isotropy or anisotropy) and the uniformity of cellular structure are other important factors. It is very common to have PU foams with anisotropic cellular structures, with cells elongated (with a large cell site) in the growing direction (Figure 2.3.C). As a consequence of the type of structure, rigid foams have lower thermal conductivities and high mechanical properties, whereas flexible foams have excellent shock absorption, acoustic absorption, absorptive capacity for water and moisture and higher permeability to gas and vapor [5, 25]. B 57 Background and state of the art Figure 2.3. SEM micrographs of polyurethane foams: A) rigid and B) flexible and C) rigid with anisotropic cells in the growing direction. The density of the PU foams is another very important parameter, because it determines the performance and cost of a foam, by measuring how much of a foam is air and how much is solid polymer [26]. The final density of PU foams, which determines a wide range of applications, can be easily modified by changing the amount of blowing agents. RPU foams may have densities between 20 and 800 kg/m3 [5]. RPU foams with lower densities (lower than 50 kg/m3) are used as insulating materials, while RPU foams with higher densities are most used as structural materials. FPU foams are produced with densities ranging from 20 to 45 kg/m3 [27]. 2.4- Common raw materials of polyurethane foams Polyols and isocyanates are the most important raw materials to build the complex structure of a PU polymer. Isocyanates are highly reactive groups, as already indicated in section 2.2, but their reaction with polyols is relatively slow at room temperature due to the phase incompatibility between the relatively non-polar and denser isocyanate phase, and the more polar and less-dense polyol phase [8]. A similar situation occurs when isocyanate reacts with water. Moreover, the successful production of PU foams requires an adequate balance of the kinetics of the reaction. It is therefore necessary to use surfactants to improve the phase compatibilization, and also catalysts to enhance or regulate the rate of blowing and gelling reactions. Hence, the PU foam formulations usually include many different components, such as polyols, isocyanates, catalysts, surfactants, blowing agents, and additives (cell openers, flame retardants, antioxidants, and light stabilizers, among others) in order to obtain a final C 58 Chapter 2 product with the desired properties. The essential components of the PU foam formulation are discussed next. 2.4.1- Polyol Polyols contain multiple hydroxyl functional groups intended to react with isocyanate groups, to give urethane linkages (scheme 2.1). There are two main types of polyols available: polyether polyols and polyester polyols. Polyether polyols contain ether and hydroxyl groups in their backbone, and are normally produced by the reaction of an alkylene oxide (epoxide) with an initiator in the presence of an acid or a base acting as a catalyst. The initiator is an active hydrogen compound containing functionality from 2 to 8 (scheme 2.9). Some examples of initiators are ethylene glycol (EG), propylene glycol, glycerin, pentaerythritol, trimethylolpropane, sucrose, or sorbitol, while ethylene oxide (EO), propylene oxide (PO), mixtures of EO and PO, or tetrahydrofuran (THF) are examples of alkylene oxides commonly used. Polyether polyols are widely used for producing foams and other PU products because they present several advantages over polyester polyols: their functionalities and equivalent weight can be widely changed, lower cost production, lower viscosities, and resulting foams which are resistant to hydrolysis [28]. However, a disadvantage of polyether polyols is their resistance to oxidation, which is lower compared to that of polyester polyols based foams. Scheme 2.9. Example of anionic polymerization of alkylene oxides, such as ethylene or propylene oxide, to obtain polyether polyols. "fn" is the functionality of the initiator, "R-(OH)f" is the initiator, n is the number of moles of alkylene oxide molecules, and R′ is H or CH3 (Adapted from [28]). Polyester polyols contain ester and hydroxyl groups in the backbone, and are generally produced by the polycondensation reaction of multifunctional carboxylic acids and polyols (scheme 2.10), or by the ring opening polymerization of a lactone with a polyol (scheme 2.11). Some examples of compounds used to produce polyester polyols are glycerine, trimethylolpropane (TMP), EG or diethyleneglycol (DEG) as polyols, and adipic acid or azelaic acid as a carboxylic acid. A large number of ester groups, which are highly polar, may give rise to polyester-based PUs with strong cohesive strength and adhesion, and thus to high strength and wear resistance. Initiator Alkylene oxide Polyether polyol 59 Background and state of the art Scheme 2.10. Example of polycondensation reaction to obtain polyester polyols. n is the number of moles of carboxylic acid molecules. Scheme 2.11. Example of the ring-opening polymerization of lactones to obtain polyester polyols. n is the number of moles of lactone molecules. Polyols may be designed to have different structures and characteristics by varying the types of starting materials. Some important features for polyols are: hydroxyl number, functionality, molecular weight, acid value, moisture, viscosity, density, and color. The main important feature is the hydroxyl number or hydroxyl index, which is the quantitative value defined by the amount of hydroxyl groups available for the reaction with isocyanates. It is expressed as milligrams of potassium hydroxide by gram of sample (mg KOH/g). Lower hydroxyl numbers indicate longer polyol chain lengths and higher molecular weights and vice versa. Functionality is the second important characteristic of polyols, and is defined as the number of hydroxyl groups by each molecule of polyol. The functionality is based on the initiator or on the mixture of initiators used for producing the polyols. Hydroxyl number, functionality and molecular weight are the main polyol features which determine the rigid or flexible behavior of the final foam (Table 2.2), but the rest of characteristics must also be considered for the foam formulation. In general, all the foams are crosslinked, being the rigid foams those highly crosslinked. Hence, rigid foams are produced from polyols with high hydroxyl number, high functionality and low molecular weight. Instead, flexible foams are obtained from polyols with low hydroxyl number, low functionality and high molecular weight. Table 2.2. Characteristics of polyols used for both rigid and flexible PU foams (Adapted from [12, 28]). Polyol Rigid foam Flexible foam Hydroxyl number (mg KOH/g) 350-560 5.6-70 Functionality 3.0-8.0 2.0-3.1 Molecular weight range (Da) 150-1000 1000-6500 Alcohol Carboxylic acid Polyester polyol Lactone Polyester polyol Alcohol 60 Chapter 2 2.4.2-Isocyanate The isocyanate component reacts with functional groups from the polyol (scheme 2.1), with water (scheme 2.2), with themselves (schemes 2.5-7), and with other components on the formulation (schemes 2.3, 2.4 and 2.8). The isocyanates used in the production of PU foams contain at least two NCO groups per molecule, and are usually aromatic isocyanates, such as toluene diisocyanate (TDI) or diphenyl methane diisocyanate (MDI). They present higher reactivity and thermal stability compared to aliphatic isocyanates, such as isophorone diisocyanate (IPDI) or hexamethylene iscocyanate (HDI) [29]. The phosgenation of amines, reported by Hentschel in 1884, is the most important commercial method for isocyanate production (Scheme 2.12). Isocyanate is obtained by the reaction of gaseous phosgene with amines or amine salt precursors. The main trouble about this commercial process of isocyanate production is the use of phosgene, which is a colorless, very reactive, and a highly toxic gas. Exposure to this toxic gas can cause severe respiratory effects, ocular irritation and burns to the eye and to the skin, and eventually to death [30]. Moreover, the isocyanates monomers thus obtained (MDI and TDI) also are harmful [30, 31]. As a consequence, the Occupational Safety and Health Administration of the United States and the European Agency for Safety and Health at Work have released regulations about the hazards and exposure limits to isocyanates. Scheme 2.12. Example of isocyanate production by the reaction of gaseous phosgene with amines. Compared to polyols, there are fewer types of commonly used isocyanates, being TDI and MDI in their oligomeric forms those most employed. Commercial TDI is a liquid of low viscosity at room temperature, and contains a mixture of 2,4- and 2,6-isomers (scheme 2.13) in an either 80/20 or 65/35 w/w ratio, being the latter mixture less reactive. TDI is the most common isocyanate used for flexible foams. Modified TDI or undistilled TDI can be used for preparing rigid foams and semirigid foams. However, TDI is not suitable for PIR foams. Scheme 2.13. Chemical Structure of TDI. Amine Phosgene Isocyanate 2,6-Isomer 2,4-Isomer 61 Background and state of the art Pure MDI (or monomeric MDI) is difunctional, is a solid at room temperature, and is used for elastomers and coatings. Polymeric and oligomeric MDI are commercialized as liquids, what enhaces processing and/or properties. The 4,4′-isomer, which usually contains a small amount of 2,2′-isomer and up to 10% of the 2,4-isomer (Scheme 2.14) is the mixture more commonly used. Its average functionality is in the range 2.3-3.0. Its functionality and its liquid nature make pMDI to be widely used to obtain rigid and semirigid foams, as well as PIR foams. Currently new products based on pMDI are being developed in order to substitute the TDI of the flexible foams, due to the higher toxicity levels of TDI compared to pMDI. Scheme 2.14. Chemical structures of MDI isomers and pMDI. The isocyanates are characterized by their NCO content, functionality, viscosity and acidity (HCl percent). NCO content, normally given in % by weight of NCO, represents the weight percent of an isocyanate comprised of NCO groups. Functionality is defined as the number of NCO groups by molecule of isocyanate. Acidity is a measure of the acid content derived from the synthesis of isocyanate (Scheme 2.12), which can influence its reactivity. It is very important to employ the correct stoichiometry in order to carry out the polymerization reaction. Normally, the total number of isocyanate groups must be higher than, or equal to, the sum of the active hydrogen-containing groups in the reacting system (hydroxyl groups of polyols plus water in water-blown PU foams). This concept is expressed by the isocyanate index or the isocyanate/hydroxyl equivalent ratio (equation 2.1) [32]. Depending on the foam system to be manufactured (for example, rigid or flexible), the isocyanate index is fixed, and thus the amount of isocyanate required to react with the polyol and with any other reactant in the formulation must be calculated in terms of theoretically stoichiometric equivalents. Therefore, the actual amount of isocyanate used in the foam is obtained from these two terms. The isocyanate index of flexible foams usually goes from 90 to 110 [26], whereas the range for rigid foams is usually 105–120 [12]. Isocyanate indexes above 100 imply 4,4’-Isomer 2,4’-Isomer 2,2’-Isomer Polymeric MDI 62 Chapter 2 the use of an excess of isocyanate groups relative to the active hydrogen-containing groups in the reacting mixture, which may generate isocyanate reactions with itself (schemes 2.5-7). Obviously, PIR foams present higher isocyanate indexes, up to 300 even more in order to form the isocyanurate group [33]. 2.4.3-Surfactant Surfactants or foam stabilizers are frequently used to improve the properties of PU foams. Surfactants consist of block copolymers of either polydimethylsiloxane-polyoxyalkylene, nonylphenol ethoxylates, silicone oils or some other organic compounds [29]. The most used surfactants for PU foams are polydimethylsiloxane-polyoxyalkylene copolymers. These surfactants are copolymers of poly(dimethylsiloxane) or PDMS [-Si(CH3)2-O-]n and oxyalkylene chains, such as polyethylene oxide chain (EO)x and polypropylene oxide chains (PO)y (Scheme 2.15). These copolymers can be linear or branched (pendant). Structural parameters of the silicone surfactants, such as the balance/content of hydrophobic PDMS and hydrophilic oxyalkylene chains, significantly affect their functions [28]. Scheme 2.15. Chemical structure of polydimethylsiloxane-polyoxyalkylene copolymers (Adapted from [28]). In PU foams, surfactants act mainly in three ways: (1) lowering the surface tension of the PU- air interface, thus promoting the generation of bubbles which evolve into the foam cells; (2) offering emulsification for the whole components, what affects to the speed of the reactions, because the concentration of reagents available is then increased; and (3) controlling the stability of the foam structure, thus preventing the collapse and regulating the voids at the Silicone Surfactant 63 Background and state of the art sub-surface [11, 12, 29]. The surfactants requirements for rigid and flexible foams are different, since the compatibility of the starting materials and the cellular structure of the final rigid and flexible foams are different, but the general surfactant behavior is similar. In rigid foams the starting components are incompatible and cells are completely closed, whereas in flexible foams the starting components are homogeneously miscible and the cells are open. The process of cell opening consists of draining the walls towards the struts, and the emulsion of components depends mainly on the surfactant structure. Flexible foams require surfactants which promote improved cell-wall drainage in order to open the cell walls during the foaming reaction, whereas rigid foams require the opposite. However, it should be emphasized that rigid foams present more stable growth than flexible foams, because the drainage effect is reduced due to the high increase in viscosity during the PU formation. In this case there is no danger of collapsing by separation of a solid phase of polyurea, as it happens in the formation of flexible foam [12]. 2.4.4-Catalyst Catalysts are required in PU foam formulations to accelerate the reactions involved in the foaming process at room temperature. At the same time, catalysts are able to maintain an adequate balance between the gelling and blowing reactions, as well as the side reactions. PU catalysts can be classified into two categories: amine compounds and organometallic compounds. Common amine catalysts are tertiary amines, such as dimethylcyclohexylamine (DMCHA), dimethylethanolamine (DMEA) and triethylenediamine [29]. Organometallic compounds usually contain metals such as bismuth, lead, zinc, tin or mercury. The basic mechanism of any of these two types of catalysts consists on increasing polarization of either the isocyanate or the hydroxyl compounds by polar interactions (Scheme 2.16) [11]. According to this, the catalyst enhances de polarity of the bonds and therefore its reactivity. Scheme 2.16. Examples of bond polarization mechanisms by tertiary amines (NR3) or organometallic compounds as catalysts (Adapted from [11]). - + - + + - + - 64 Chapter 2 The catalytic activity of tertiary amines is controlled by their structure and their basicity, increasing with the increasing basicity and decreasing with the steric hindrance of the substituents of the nitrogen atom [7]. Amine catalysts are able to promote both gelling and blowing reactions (urethane and urea products) at the same time, as well as different side reactions, depending on isocyanate excess, catalyst concentration, temperature, etc [11]. However, specific amine catalysts can be designed to preferably catalyze either the gelling or the blowing reactions, thus their appropriate combinations may help to achieve the desired properties of the final foam. On the other hand, organometallic catalysts favor almost exclusively the gelling reaction. Some examples of organometallic catalysts are stannous octoate, dibutyltindilaurate, and tin mercaptides. The advantages of these catalysts respect to tertiary amines are based on their lower volatility, what facilitates their maintenance in the foam without causing odors which is an important issue in automotive applications, for example. The odor problem of amine catalysts can be also solved by using amines containing functional groups able to react with isocyanate, such as OH- or NH2, which react during the foaming process and are eventually bonded to the foams [12]. Organometallic catalysts can be used either alone or in combination with amine catalysts in order to achieve the expected balance of reaction rates in the PU foam formation. 2.4.5-Blowing agent PU foam expansion occurs due to the presence of either a chemical blowing agent, a physical blowing agent, or a mixture of both. Physical blowing agents (such as solvents with low boiling point: pentane, hexane, acetone, hydrochlorofluorocarbons (HCFCs) or hydrofluoroolefins (HFOs)) provide gas for PU expansion by vaporization during the foaming process, what is an endothermic process [34]. Chemical blowing agents are compounds which release gas during the foam formation, due to either a thermal decomposition or to a chemical reaction. This is the role of water, which is the most common chemical blowing agent in PU formation [26]. Until recently, chlorofluorocarbons (CFCs) and HCFCs were commonly used as physical blowing agents, but they have been recently prohibited because they play a decisive role in the destruction of the ozone layer. For this reason, more environmentally friendly blowing agents, such as cyclopentane (CP), HFOs or water, are currently being used in most of the industrial applications. The CP evaporated during the foaming process as a result of the heat produced during the polymerization reaction may be occluded as a physical blowing agent inside the cells. As indicated before (Schemes 2.2, 2.7, and 2.8) water acts as a chemical blowing agent expanding PU due to the carbon dioxide generated during different exothermic reactions, during water-blown RPU foams production. The blowing agents play an important role in both the production and performance of the PU foams. Thus, the amount of blowing agent greatly determines the foam density of the final foam, and influences the cellular microstructure and the polymer morphology of the PU foams: these are essential parameters in order to describe their final behavior [26]. Thermal 65 Background and state of the art Ref. System studied (filler used) Study of the foaming process Study of the cellular structure parameters Thermal conductivity and mechanical properties [67] Water-blown RPU foam with nanosilica (Aerosil 380) and nanosilica suspension of nanosilica and ionic liquid (0.5, 1, and 2 wt% of nanosilica). Yes. The characteristic foaming times (cream, gel and tack-free times) and the maximum foaming temperature were measured. Yes. Cell diameter, wall thickness and cell size distribution were calculated from SEM images. 0.5 and 1 wt% of nanosilica and nanosilica suspension improved the physic-mechanical properties (compressive strength, three-point bending test, apparent density) and thermal conductivity. However, 2 wt% of content was not improved these properties. [68] Water-blown RPU foam with two types of polyhedral oligomeric silsesquioxanes (POSSs) (0.5, 1.5 and 5 wt%). Yes. The characteristic foaming times (cream, gel and tack-free times) and the maximum foaming temperature were measured. Yes. Cell diameter, wall thickness and cell size distribution were calculated from SEM images. The addition of both fillers in an amount of 5 wt% led to samples with reduced compression modulus, compressive strength, thermal transitions, and storage modulus with respect to the foams containing 0.5 and 1.5 wt% of the fillers, mainly due to damaging changes induced by the fillers. [69] Water-blown RPU foam reinforced with nanoclay and Silica aerogel (2 and 4 wt% in polyol). No. No. Only SEM images were commented without calculating cellular structure parameters. Thermal conductivity was highly enhanced only for nanoclay addition. In general, the compressive and tensile properties improved with the fillers. [70] Water-blown RPU foam with raw and functional MWCNTs (up to 0.2 wt%). No. Yes. Cell size, cell wall thickness, cell edge length and cell density were calculated from SEM images. Fillers showed an enhancement of 13% in compressive strength compared to neat PU foam. [71] Water-blown RPU foam filled with MWCNTs and poly (ethylene oxide) grafted MWCNTs (up to 3 wt%). Yes. The characteristic foaming times (cream, gel and tack-free times). Yes. Cell size, cross section of cells and anisotropy were calculated from SEM images. The specific compressive properties (property/density) were improved with the fillers, improving more with poly (ethylene oxide) grafted MWCNTs. [72] Water-blown semi-rigid PU foam with aluminum microfibers (0.5, 1, 1.5 and 2 wt%). No. No. This increase in properties (61.81%-compressive strength and 71.29%-energy absorption) was obtained by adding up to 1.5%. [73] Water-blown RPU foam filled with silanized MWCNTs (1.5 and 3 wt%) No. Yes. SEM images are shown and cell density was calculated. Tensile mechanical properties were improved with 1.5 wt% of fillers, whereas foams with 3 wt% showed poor results. [62] Water-blown RPU foam reinforced with fullerene soot, modified fullerene soot, AL2O3 and CrO3 (up to 0.6 wt%). Yes. The degree of microphase separation in polyurethanepolyurea matrix was determined by deconvolution of carbonyl band in FTIR spectra of the final foams. Yes. Optical images analysis allowed determining cell size and cell size distribution. The introduction of nanofillers leads to increase of mechanical strength of the foam with the exception of CrO3. The predicted thermal conductivity increased with the nanofillers addition. [74] Water-blown RPU foam with various contents of graphene oxide (GO) (up to 0.7 wt%) and graphite (up to 5 wt%). No. No. Only SEM images were commented without calculating cellular structure parameters. The compressive strength improved by 5% and by 10% for 1 wt% of graphite and 0.05 wt% of GO, respectively. Thermal conductivity was not reduced with GO, but it was slightly reduced by 0.1% with 1 wt% of graphite. [75] Cyclopentanewater blown RPU foam reinforced with TiO2 nanoparticles and halloysite clay nanotubes (uo to 10 wt%). No. Yes. Cell size and cell density were calculated by SEM analysi. The compressive mechanical properties were enhanced only with high contents of fillers (6, 8 and 10 wt%). Thermal conductivity was reduced with fillers addition after 3 days of foam production, but this worsen in all the foams after 10 days. 72 Chapter 2 Ref. System studied (filler used) Study of the foaming process Study of the cellular structure parameters Thermal conductivity and mechanical properties [76] A series of RPU foams with different types of blowing agents (CFC-11, cyclopentane, pentane or hexane) and with silica aerogel (1–5 wt%). No. FTIR spectroscopy was used in order to observe if there is reaction between silica aerogel and isocyanate or polyol matrix. No. Only SEM images were commented without calculating cellular structure parameters. The thermal conductivity of nanocomposite foams for all blowing agents was less than the pure foams. Also, the mechanical properties improved for nanocomposite foams, particularly for lower silica aerogel contents (1 wt.% and 3 wt%). [77] Three different types of water-blown RPU foams with granular silica aerogel and powdered silica aerogel (up to 4 wt%). No. No. Only SEM images were commented without calculating cellular structure parameters. Thermal conductivity of foams did not enhance by adding silica aerogel. [78] Water-blown RPU foam reinforced with nanoporous graphene (0.5–5 wt%). No. No. Only SEM images were commented without calculating cellular structure parameters. Only 0.25 wt % of fillers improved compressive strength and modulus respectively by 10.7% and 66.5%. [79] Cyclopentane-water blown RPU foam reinforced with glass fibre (5, 10 and 20 wt%). No. No. Only SEM images were commented without calculating cellular structure parameters. Fibre reinforcements increased the mechanical properties of the foam in a significant manner (up to 121% increase in stiffness and 101% increase in strength), proportionally with the increase in fibre mass content. [80] Water-blown RPU foam with titanium dioxide, zinc oxide, and magnetite nanofillers (1 or 10 wt% in polyol) Yes. Reaction kinetics were followed by FTIR spectroscopy (isocyanate conversion was studied). Yes. Some parameters of cellular structure were measured or commented from SEM images. Thermal and mechanical properties did not study. [81] RPU foam blown HCFC- 141 b and filled with MWCNTs (0.1, 1 and 2 wt%) No. Yes. Optical and SEM images were commented and cell size and cellular density were calculated. The thermal conductivity showed a modest increase with increased MWCNTs content. 2.8- Monitoring of foaming process of polyurethane foams As indicated previously (section 2.3.1), the reaction kinetics of PU foams are very complex because they involve simultaneous reactions which are responsible for the final morphology of the PU matrix and microstructural characteristics, and therefore affect to final properties of the resulting foams. The balance of the PU reactions can be modified by the addition of the fillers, by changes in PU components, or by foaming conditions, among others, and for this reason understanding how these reactions take place is a decisive factor. Historically, Fourier transform infrared (FTIR) spectroscopy has proven to be a very useful tool to investigate both the reaction kinetics and the morphology development during the foaming process [82-85]. In PU foams, analysis of the FTIR spectra must be carried out by studying specific bands. The most studied bands are the isocyanate band at 2270 cm-1, corresponding to asymmetric stretching vibrations of the isocyanate group, and the amide I or carbonyl region (1800-1600 cm-1) corresponding to stretching vibrations of the carbonyl groups present in reaction products. However, the carbonyl region may contain many hidden bands corresponding mainly to the gelling reaction (urethane compounds such as free urethane and hydrogen-bonded urethanes), and blowing reaction (urea compounds such as free urea and hydrogen-bonded 73 Background and state of the art ureas, which include both monodentate, or disordered, and bidentate, ordered) (see Table 2.4). The frequencies of the hidden bands in the carbonyl region can be established from the second-derivative spectra, and then each absorption can be identified, separated, and quantified by deconvolution [82, 83, 86-89]. Given the difficulty of the deconvolution analysis of the carbonyl region, most of the investigations studied only the intensity changes of the total bands (amide I). Table 2.4. Carbonyl stretching assignments in PU foams. Functional group Frequency (cm-1) Band assignment Structure ν (C=O) Urethane (Gelling reaction) 1740-1730 Free urethane (Urethane groups free from hydrogen bonds). [19, 85, 90-94] 1730-1725 Hydrogen-bonded urethane hard segment/soft interaction (Ether polyol hydrogenbonded urethane). [90-94] 1715-1700 Hydrogen-bonded urethane hard segment/ hard segment interaction [90-94] ν (C=O) Urea (Blowing reaction) 1700-1690 Free Urea (Urea groups free from hydrogen-bonds). [84, 85, 90, 95] 1690-1650 Monodentate or disordered hydrogen-bonded urea [19, 90] 1670-1660 C=O Hydrogen-bonded and N-H bonded to an ether group [95] 1650-1640 Bidentate or ordered hydrogen-bonded urea [19, 84, 85, 90, 95-97] 74 Chapter 2 Since the inclusion of fillers in the PU matrix is an attractive strategy to enhance the properties of PU foams, most of the published articles have been focused on the production and characterization of the foams. Some of these studies are focused on the effect of fillers on the reaction kinetics and on the development of the PU morphology during the foaming process [62, 80, 96, 98, 99]. The following paragraphs collect a brief summary of those studies focused on the effect of fillers on the reaction kinetics and/or the morphology of PU foams monitoring the isocyanate and/or carbonyl bands by ATR-FTIR spectroscopy. Wilkinson et al.(2007) [98] prepared water-blown FPU foams infused with unmodified Na+ montmorillonite (MMT) at a concentration of 5 or 10 wt% in the final foam. FTIR data showed a significant increase in the rate of isocyanate decay as MMT content was increased, in agreement with the adiabatic temperature rise profiles that were also measured. The absorbance changes in soluble and hydrogen-bonded urea group contents in the HS phases were used to determine the microphase-separation times. These times decreased for nanocomposites foams, but the formation of hydrogen-bonded urea of HS phase was higher for reference foam (without MMT) at longer times. Bernal et al.(2011) [96] investigated the influence of chemically inert multi-walled CNTs (MWCNTs) and functionalized MWCNTs (f-MWCNTs) bearing oxygen-containing groups over the surface on the foaming process of water-blown FPU foams. They studied foams containing 0.1 or 0.5 phpp (parts by hundred parts of polyol), which correspond to 0.06 or 0.33 wt% of the final foam. A decrease of isocyanate conversion rates at the early stages was observed when the MWCNTs amount increases, what was explained considering that the high initial viscosity of the systems with MWCNTs induced a lower mobility of the molecules during the reaction. Moreover, the onset of the microphase separation (the formation of ordered hydrogen-bonded urea) was influenced by the presence of functional groups on the MWCNT surface. The same group (2012) [99] later reported a kinetic study of water-blown FPU foams with two contents of HS (25% or 32.5%) filled with MWCNTs, f-MWCNTs or functionalized graphene sheets (FGS) (0.5 phpp which correspond to 0.33 wt% of the final foam). Figure 2.6 shows the deceleration of the rate of polymerization due to the increase of the HS content and to the inclusion of the nanoparticles. In the case of HS systems, the decrease in the rate and isocyanate conversion was attributed to an increase in the viscosity of the system and thus to a restriction of the chain mobility due to the larger HS content. In the case of the nanofillers system, the chain mobility was also affected by both the viscosity effect and by the hindrance to the formation of hydrogen bonds between the HS. 75 Background and state of the art Figure 2.6. Isocyanate conversion, pNCO, for FPU nanocomposite foams with 25%HS (A) and for FPU nanocomposite foams with 32.5%HS (B). [100] Akkoyun et al.(2016) [80] studied the effects of different different types, shapes and concentrations of titanium dioxide, zinc oxide, and magnetite nanofillers (1 or 10 wt% in polyol) on the RPU foam formation reaction. The results of the isocyanate conversion showed that, regardless of the nanofiller type, the reaction rate increased as the surface area increased beyond a critical surface area of nanofillers (30 m2 in this study). Moreover, the nanofillers reduced the final cell size below the critical surface area, whereas above this critical value the cell size distribution was wider. Thus, the increase of the reaction rate facilitated uncontrolled cell nucleation, growth, and hence coalescence, which resulted in an uncontrolled foam structure. Pikhurov et al.(2018) [62] prepared water-blown RPU foams reinforced with nanofillers of different nature (fullerene soot containing fullerenes with hydrophobic surface, modified fullerene soot with neutral hydrophilic surfaces, Al2O3 nanoparticles with acidic or basic hydrophilic surfaces, or Cr2O3 nanoparticles with acidic hydrophilic surfaces) at concentrations 0.075–0.6 wt%. The effects of these fillers on the foam morphology were investigated by using ATR-FTIR spectroscopy. The different hidden peaks of the free and hydrogen-bonded groups were obtained by deconvolution of the carbonyl region in order to calculate the degree of phase separation (DPS) for urethane bonds and for urea bonds (equation 2.2), thus obtaining a measurement of the micro-phase separation (Table 2.5 collects the DPS values). Ab and Af are the deconvoluted area of hydrogen-bonded and free molecular groups, respectively. The FTIR results indicate that the foams with hydrophilic nanofillers (modified fullerene soot, Al2O3 and Cr2O3), able to form hydrogen bonds with urethane and urea groups, increased DPS for urethane bonds in all concentrations, compared to the reference foam. However, DPS for the urea bonds were practically the same as that in the reference and did A B 76 Chapter 2 not depend on concentration or type of filler. However, no correlations could be found between mechanical performance and DPS. Table 2.5. DPS values obtained from FTIR analysis (Adapted from [62]). Loading (wt%) DPS for urethane bonds DPS for urea bonds Reference 0.915 0.557 Fullerene soot 0.075 0.799 0.214 0.15 0.733 0.23 0.3 0.747 0.225 0.6 0.858 0.437 Modified fullerene soot 0.075 0.988 0.589 0.15 0.990 0.625 0.3 0.989 0.621 0.6 0.993 0.63 Al2O3 0.075 0.959 0.547 0.15 0.956 0.545 0.3 0.959 0.534 0.6 0.959 0.525 CrO3 0.075 0.966 0.535 0.15 0.981 0.544 0.3 0.953 0.547 0.6 0.960 0.573 In addition to FTIR spectroscopy, other methods can be used to obtain additional kinetic information, such as measuring the adiabatic temperature raise. It is directly measured during the foaming process by thermocouples inside the foam, since the gelling and blowing reactions are exothermic. Once the cellular structure of the PU foam has been formed, the center of foam is isolated from its surroundings, so it may be considered as an adiabatic system. Thus, the adiabatic temperature monitoring of the PU foam may be related to the kinetics of the reaction, and hence to the consumption of isocyanate in both gelling and blowing reactions [101]. The foaming temperature as a function of time has been measured in several articles to determine the effect caused on the reaction kinetics of PU foams by: different blowing agents [102, 103], different contents of bio-polyols [9, 104] or nanofillers [59, 98, 105]. Some studies of nanocomposite foams described an increase of the foaming temperature when nanofillers are incorporated in water-blown RPU foams [59, 98, 105]. Bernal et all. (2014) [59] studied RPU foams filled with MWCNTs, f-MWCNTs or FGS (0.4 or 0.8 phpp, which corresponded to 0.17 or 0.35 wt% of the final foam) to obtain electromagnetic interference (EMI) shielding materials. The extent of conversion of RPU nanocomposite foams was followed by adiabatic temperature raise (Figure 2.7). The final degree of conversion was similar for all 77 Background and state of the art samples, but those with nanofillers showed faster conversion rates during the initial stages of the reaction than the reference foam. The acceleration of the reaction was favoured by the high thermal conductivity in the system with MWCNTs, and by the low viscosities and surface functionalities in the system with f-MWCNTs and FGS. Similar results were obtained by Wilkinson et al.(2007) [98] using the adiabatic temperature rise method for RPU foams with MMT (0, 5 or 10 wt% in foam). In the initial stage, a significant increase in the rate of reaction when increasing MMT content was detected, due possibly to a surface catalytic effect related to the high-surface area of the MMT. Ahn et al. (2015) [105] also observed a faster and higher temperature raise with time for RPU foams containing MWCNTs (0.005-0.1 phpp). This could be understood as a result of the creation of cells with smaller sizes and of a more uniform cell size distribution due to the nucleating effect of MWCNT. Figure 2.7. (A) Variation of the experimental temperature (Texp) with time for RPU nanocomposite foams. (B) Variation of the conversion x with time for RPU nanocomposite foams [59]. Finally, X-ray radioscopy allows to collect a great deal of information of the PU foaming process. The use of this technique in combination with image analysis allows determining the evolution of relative density and cell size, as well as to calculate cell nucleation density during the whole foaming process (Figure 2.8). This technique was applied for the first time by our group (Pardo-Alonso et al, 2012) [106] in order to evaluate the effect of nanoclays content (0.5, 1, 3 or 5 wt% of Cloisite®30B) on the foaming process of water-blown RPU foams. No significant changes of the relative density were found (Figure 2.8.A). However, a noticeable cell size reduction was observed as the amount of nanoclays increased, being reduced to a half for the foam with 5 wt% nanoclays (Figure 2.8.B). The cell nucleation density also increased with the content of nanoclays, and it remained constant with time, thus confirming the absence of cell coalescence (Figure 2.8.C). Our group (2013) [107] also analyzed by X-ray A B 78 Chapter 2 radioscopy the effect of different nanofillers (1 or 5 wt% contents of nanoclays (Cloisite®30B) or nanosilicas (Aerosil®A200 and Aerosil®R812)) on the cellular structure of water-blown RPU foams. Both nanoclays and nanosilicas showed a significant cell size reduction, being nanoclays those promoting higher cell nucleation density. Moreover, none of the nanoparticles induced coalescence in the RPU system. Another X-ray radioscopy study was carried out by Bernal et al. (2014) [108]. The cellular structure and density evolution were monitored for water-blown FPU foams with MWCNTs, f-MWCNTs or FGS (0.5 phpp, which corresponded to 0.33 wt% of the final foam). At early stage, a significant cell nucleation enhancement was detected for MWCNT (hydrophobic surface) and FGS (hydrophobic/hydrophilic surface), but these particles induced higher coalescence and thus the cell size reduction was lost at the final stage. Instead, f- MWCNT (hydrophilic surface) promoted a similar number of cells than neat FPU foam without coalescence. Thus, the surface nature of the carbon nanoparticles affected the physical events occurring on the PU nanocomposite foams, what is correlated with the chemical modifications observed for these systems by in situ FTIR [99]. Figure 2.8. (A) Relative density evolution; (B) cell size evolution and (B) cell density evolution during the PU foaming process for neat PU foam and foams containing 0.5, 1, 3 or 5 wt% of nanoclays [106]. A B C 79 Background and state of the art The review of the different techniques used to study the foaming process allows concluding that they provide complementary information. FTIR spectroscopy allows to follow the reaction kinetics and the composition of PU matrix during the foam formation. The measurements of foaming temperature also provide information on the reactivity of the system. Finally, the X- ray radioscopy allows analysing the foaming mechanisms produced (cell nucleation and growth and degeneration mechanisms such as nucleation, cell growth, pore coalescence and coarsening) at all intermediate stages during the foaming process. Thus, the combination of the information provided by each technique can help us to understand the final properties of the PU foams, and thus to improve the PU formulation. As it was explained in Chapter 1, this is one of the main objectives of this research. 2.9- Thermoplastic polyurethane foams 2.9.1- Thermoplastic polyurethanes One of the most important types of PU material is TPU, which is a thermoplastic elastomer discovered in 1958 [109]. TPU combines advantages of both thermoplastics (meltprocessability) and elastomers (elasticity). Hence, TPUs are very versatile materials, which can exhibit properties ranging from very soft thermoplastic elastomers to strong, rigid thermoplastics, depending on their chemical composition, backbone structures and resultant microphase separation [88]. Given the possible combinations of TPU properties, these materials have many applications in automotive components, textile fibers, adhesives, wire and cable sheathings, footwear, and biomaterials [27, 29]. TPUs are usually produced by the polyaddition reaction of a diisocyanate with two other reactants: a macrodiol and a low chain diol or diamine, that acts as chain extender (scheme 2.1). TPUs consist of linear segmented block copolymers composed of alternating SS and HS which are flexible (SS) and rigid (HS) polymeric chains which resemble the behavior of an elastomer, but they lack permanent cross-links. The thermodynamic incompatibility between SS and HS gives rise to a two-phase microstructure, in which the HS from the diisocyanate and the chain extender segregate into semicrystalline domains through physical cross-links, whereas the SS from the macrodiol chains form amorphous domains in which the HS are dispersed (Figure 2.9) [110]. Therefore, HS offers rigidity, and SS provides flexibility to the TPU material, and their ratio and distribution are determined by different chemical and structural factors, including the polymerization procedure used for their synthesis and the processing conditions [111]. Moreover, the final properties of TPUs are generally attributed to their morphological structure, and they can be adequately tuned by changing the components, the ratio of SS and HS and the reaction steps. 80 Chapter 2 Figure 2.9. Schematic illustration of TPU morphology [112]. TPUs with different properties can be manufactured due to the commercial availability of a large number of starting compounds. These are the following: diisocyanate, macrodiol and chain extender. -Isocyanates: are difunctional and monomeric with a low molecular weight of around 150-250 [113]. Isocyanates in TPUs have a double function: (1) reacting with the macrodiol to produce the urethane-sparse TPU SS, and (2) reacting with the chain extender to generate HS in the TPU backbone [113]. The isocyanates most used are TDI and MDI (see section 2.4.2). -Macrodiols: are polyester or polyether diols with long chains with molecular weights of around 500-4000 [113]. Macrodiols comprise the SS, so as for example longer macrodiols give more flexible TPUs. Some examples of macrodiol are polyethylene oxide glycol, oxytetramethylene glycol and poly (tetramethylene adipate) glycol. -Chain extenders: are typically diols or diamines with low chains of molecular weight around 100-350 [113]. For example, shorter chain extenders favor more crystalline TPUs, that are tougher and with a higher strength [114]. Some examples of chain extenders are 1-ethanediol, 1,3-propanediol, 1,5-pentanediol or ethylene diamine. TPUs can be synthesized by two basic polymerization methods: one-step process and two-step process (the latter also called pre-polymer process) [113]. - One-step process: All of TPU components (diisocyanate, macrodiol and chain extender) are mixed together at the same time (Scheme 2.17). Scheme 2.17. Synthesis of TPU by one-step process [113]. -Two-steps process: In a first step, the macrodiol reacts with an excess of the diisocyanate to form an isocyanate-terminated prepolymer. Then, the isocyanate-terminated prepolymer Diisocyanate Macrodiol Chain extender TPU 81 Background and state of the art [76] N. Nazeran, J. Moghaddas, Synthesis and characterization of silica aerogel reinforced rigid polyurethane foam for thermal insulation application, Journal of Non-Crystalline Solids, 461 (2017) 1-11. [77] A. Dourbash, C. Buratti, E. Belloni, S. Motahari, Preparation and characterization of polyurethane/silica aerogel nanocomposite materials, Journal of applied Polymer Science, 134 (2017) 44521. [78] M. Hoseinabadi, M. Naderi, M. Najafi, S. Motahari, M. Shokri, A study of rigid polyurethane foams: The effect of synthesized polyols and nanoporous graphene, Journal of Applied Polymer Science, 134 (2017) 45001. [79] D.-A. Şerban, O. Weissenborn, S. Geller, L. Marşavina, M. Gude, Evaluation of the mechanical and morphological properties of long fibre reinforced polyurethane rigid foams, Polymer Testing, 49 (2016) 121-127. [80] M. Akkoyun, E. Suvaci, Effects of TiO2, ZnO, and Fe3O4 nanofillers on rheological behavior,microstructure, and reaction kinetics of rigid polyurethane foams, Journal of Applied Polymer Science, 133 (2016). [81] J.J. Espadas-Escalante, F. Avilés, P.I. Gonzalez-Chi, A.I. Oliva, Thermal conductivity and flammability of multiwall carbon nanotube/polyurethane foam composites, Journal of Cellular Plastics, 53 (2016) 215-230. [82] M.M. Coleman, K.H. Lee, D.J. Skrovanek, P.C. Painter, Hydrogen bonding in polymers. 4. Infrared temperature studies of a simple polyurethane, Macromolecules, 19 (1986) 2149– 2157. [83] R.D. Priester, J.V. McClusky, R.E. O'Neill, R.B. Turner, M.A. Harthcock, B.L. Davis, FT-IR-A Probe into the Reaction Kinetics and Morphology Development of Urethane Foams, Journal of Cellular Plastics, 26 (1990) 346-367. [84] M.J. Elwell, A.J. Ryan, H.J.M. Grünbauer, H.C.V. Lieshout, In-Situ Studies of Structure Development during the Reactive Processing of Model Flexible Polyurethane Foam Systems Using FT-IR Spectroscopy, Synchrotron SAXS, and Rheology, Macromolecules, 29 (1996). [85] M.J. Elwell, A.J. Ryan, An FT i.r. study of reaction kinetics and structure development in model flexible polyurethane foam systems, Polymer, 37 (1996) 1353–1361. [86] Y. Shi, X. Zhan, Z. Luo, Q. Zhang, F. Chen, Quantitative IR characterization of urea groups in waterborne polyurethanes, Journal of Polymer Science Part A: Polymer Chemistry, 46 (2008) 2433-2444. [87] D.P. Queiroz, M.N. de Pinho, C. Dias, ATR-FTIR Studies of Poly(propylene oxide)/Polybutadiene Bi-Soft Segment Urethane/Urea Membranes, Macromolecules, 36 (2003) 4195–4200. [88] I. Yilgor, E. Yilgor, I.G. Guler, T.C. Ward, G.L. Wilkes, FTIR investigation of the influence of diisocyanate symmetry on the morphology development in model segmented polyurethanes, Polymer, 47 (2006) 4105-4114. [89] L.-S. Teo, C.-Y. Chen, J.-F. Kuo, Fourier Transform Infrared Spectroscopy Study on Effects of Temperature on Hydrogen Bonding in Amine-Containing Polyurethanes and Poly(urethaneurea)s, Macromolecules, 30 (1997) 1793-1799. [90] A.L. Daniel-da-Silva, J.C.M. Bordado, J.M. Martín-Martínez, Moisture curing kinetics of isocyanate ended urethane quasi-prepolymers monitored by IR spectroscopy and DSC, Journal of Applied Polymer Science, 107 (2008) 700–709. [91] Michael M. Coleman, Maria Sobkowiak, George J. Pehlert, P.C. Painter, Infrared temperature studies of a simple polyurea, Macromolecular Chemistry Physis, 198 (1997) 117– 134. [92] L. Ning, W. De-Ning, Y. Sheng-Kang, Hydrogen-Bonding Properties of Segmented Polyether Poly(urethane urea) Copolymer, Macromolecules, 30 (1997) 4405–4409. 88 Chapter 2 [93] V. Zharkov, A. Strikovsky, T. Verteletskaya, Amide I absorption band: description of the urethane group association scheme in polyether urethane elastomers, Polymer, 34 (1993) 935–941. [94] A. Strikovsky, V. Zharkov, Infra-red spectroscopy study of equilibrium association of urethane groups in poly(ether urethane)s, Polymer, 34 (1993) 3397. [95] A. Marcos-Fernández, A.E. Lozano, L. González, A. Rodríguez, Hydrogen Bonding in Copoly(ether-urea)s and Its Relationship with the Physical Properties, Macromolecules, 30 (1997) 3584–3592. [96] M.M. Bernal, M.A. Lopez-Manchado, R. Verdejo, In situ Foaming Evolution of Flexible Polyurethane Foam Nanocomposites, Macromolecular Chemistry and Physics, 212 (2011) 971– 979. [97] R. Versteegen, R. Sijbesma, E. Meijer, Synthesis and Characterization of Segmented Copoly(ether urea)s with Uniform Hard Segments, Macromolecules, 38 (2005) 3176–3184. [98] A.N. Wilkinson, N.H. Fithriyah, J.L. Stanford, D. Suckley, Structure Development in Flexible Polyurethane Foam-Layered Silicate Nanocomposites, Macromolecular Symposia, 256 (2007) 65-72. [99] M.M. Bernal, M. Martin-Gallego, L.J. Romasanta, A.-C. Mortamet, M.A. López-Manchado, A.J. Ryan, R. Verdejo, Effect of hard segment content and carbon-based nanostructures on the kinetics of flexible polyurethane nanocomposite foams, Polymer, 53 (2012) 4025-4032. [100] M.M. Bernal, PhD Thesis: Estudio de nanocompuestos de espumas de poliuretano reforzadas con nanocargas en base carbono, Polytechnic University of Valencia, 2012. [101] L.D. Artavia, C.W. Macosko, Foams kinetics, Journal of Cellular Plastics, 26 (1990) 490- 511. [102] M. Modesti, V. Adriani, F. Simioni, Chemical and Physical Blowing Agents in Structural Polyurethane Foams: Simulation and Characterization, Polymer Engineering & Science, 40 (2000). [103] M. Kuranska, A. Prociak, S. Michalowski, K. Zawadzinska, The influence of blowing agents type on foaming process and properties of rigid polyurethane foams, Polimery, 63 (2018) 672- 678. [104] M. Kurańska, A. Prociak, The influence of rapeseed oil-based polyols on the foaming process of rigid polyurethane foams, Industrial Crops and Products, 89 (2016) 182-187. [105] W. Ahn, J.-M. Lee, Effect of CNT as a Nucleating Agent on Cell Morphology and Thermal Insulation Property of the Rigid Polyurethane Foams, Journal of Nanoscience and Nanotechnology, 15 (2015) 9125-9130. [106] S. Pardo-Alonso, E. Solórzano, S. Estravís, M.A. Rodríguez-Perez, J.A. de Saja, In situ evidence of the nanoparticle nucleating effect in polyurethane–nanoclay foamed systems, Soft Matter, 8 (2012) 11262. [107] S. Pardo-Alonso, E. Solorzano, M.A. Rodriguez-Perez, Time-resolved X-ray imaging of nanofiller-polyurethane reactive foam systems, Colloid Surf. A-Physicochem. Eng. Asp., 438 (2013) 119-125. [108] M. Mar Bernal, S. Pardo-Alonso, E. Solórzano, M.Á. Lopez-Manchado, R. Verdejo, M.Á. Rodriguez-Perez, Effect of carbon nanofillers on flexible polyurethane foaming from a chemical and physical perspective, RSC Advances, 4 (2014) 20761. [109] C.S. Schollenberger, H. Scott, G.R. Moore, Rubber Worls, 137 (1958) 549. [110] C. Prisacariu, Polyurethane elastomers, Wien New York: Springer, 2011. [111] Y. He, D. Xie, X. Zhang, The structure, microphase-separated morphology, and property of polyurethanes and polyureas, Journal of Materials Science, 49 (2014) 7339-7352. [112] I. Javni, O. Bilić, N. Bilić, Z.S. Petrović, E.A. Eastwood, F. Zhang, J. Ilavský, Thermoplastic polyurethanes with controlled morphology based on methylenediphenyldiisocyanate/isosorbide/butanediol hard segments, Polymer International, 64 (2015) 1607-1616. 89 Background and state of the art [113] Handbook of Elastomers, in: A.K. Bhowmick, H. Stephens (Eds.), CRC Press2000. [114] K.E. Kear, Developments in Thermoplastic Elastomers, in: R.R. Reports (Ed.), 2003. [115] X. Sun, H. Kharbas, L.-S. Turng, Fabrication of highly expanded thermoplastic polyurethane foams using microcellular injection molding and gas-laden pellets, Polymer Engineering & Science, 55 (2015) 2643-2652. [116] C. Dai, C. Zhang, W. Huang, K.-C. Chang, L.J. Lee, Thermoplastic polyurethane microcellular fibers via supercritical carbon dioxide based extrusion foaming, Polymer Engineering & Science, 53 (2013) 2360-2369. [117] X.-C. Wang, X. Jing, Y.-Y. Peng, Z.-K. Ma, C.-T. Liu, L.-S. Turng, C.-Y. Shen, The effect of nanoclay on the crystallization behavior, microcellular structure, and mechanical properties of thermoplastic polyurethane nanocomposite foams, Polymer Engineering & Science, 56 (2016) 319-327. [118] H.-Y. Mi, X. Jing, M.R. Salick, X.-F. Peng, L.-S. Turng, A novel thermoplastic polyurethane scaffold fabrication method based on injection foaming with water and supercritical carbon dioxide as coblowing agents, Polymer Engineering & Science, 54 (2014) 2947-2957. [119] Baralu Jagannatha Rashmi, D. Rusu, K. Prashantha, M.F. Lacrampe, P. Krawczak, Development of Water-Blown Bio-Based Thermoplastic Polyurethane Foams Using Bio-Derived Chain Extender, Journal of Applied Polymer Science, 128 (2013) 292–303. [120] S.-K. Yeh, Y.-C. Liu, W.-Z. Wu, K.-C. Chang, W.-J. Guo, S.-F. Wang, Thermoplastic polyurethane/clay nanocomposite foam made by batch foaming, Journal of Cellular Plastics, 49 (2013) 119-130. [121] C.-C. Chu, S.-K. Yeh, S.-P. Peng, T.-W. Kang, W.-J. Guo, J. Yang, Preparation of microporous thermoplastic polyurethane by low-temperature supercritical CO2 foaming, Journal of Cellular Plastics, 53 (2017) 135-150. [122] X. Dai, Z. Liu, Y. Wang, G. Yang, J. Xu, B. Han, High damping property of microcellular polymer prepared by friendly environmental approach, The Journal of Supercritical Fluids, 33 (2005) 259-267. [123] S. Ito, K. Matsunaga, M. Tajima, Y. Yoshida, Generation of microcellular polyurethane with supercritical carbon dioxide, Journal of Applied Polymer Science, 106 (2007) 3581-3586. [124] R. Rizvi, H. Naguib, Porosity and composition dependence on electrical and piezoresistive properties of thermoplastic polyurethane nanocomposites, Journal of Materials Research, 28 (2013) 2415-2425. [125] A. Prasad, G. Fotou, S. Li, The effect of polymer hardness, pore size, and porosity on the performance of thermoplastic polyurethane-based chemical mechanical polishing pads, Journal of Materials Research, 28 (2013) 2380-2393. [126] S.-K. Yeh, Y.-C. Liu, C.-C. Chu, K.-C. Chang, S.-F. Wang, Mechanical Properties of Microcellular and Nanocellular Thermoplastic Polyurethane Nanocomposite Foams Created Using Supercritical Carbon Dioxide, Industrial & Engineering Chemistry Research, 56 (2017) 8499-8507. [127] S.-K. Yeh, Y.-R. Chen, T.-W. Kang, T.-J. Tseng, S.-P. Peng, C.-C. Chu, S.-P. Rwei, W.-J. Guo, Different approaches for creating nanocellular TPU foams by supercritical CO2 foaming, Journal of Polymer Research, 25 (2017) 30. 90 Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane CHAPTER 3: MATERIALS, FOAM PRODUCTION AND EXPERIMENTAL TECHNIQUES Mercedes Santiago Calvo Chapter 3 3.1- Introduction This chapter describes in detail the polyurethane (PU) formulations, the fillers and the procedures for foams production employed in this research work. A summary of the information included in this chapter is collected in Table 3.1. The experimental techniques, enumerated together with the standards used in each case, will be collected later (Tables 3.12- 15). Table 3.1. Summary of PU commercial and own formulations, fillers and foaming process employed in the different chapters of this research. RESEARCH ARTICLES Chapter Formulations Fillers Foaming process 1 The effects of functional nanofillers on the reaction kinetics, microstructure, thermal and mechanical properties of water blown rigid polyurethane foams. 4 Commercial formulation of water-blown RPU foam Nanosilicas and nanoclays Reactive foaming: Free expansion in a plastic cup 2 Infrared expandometry: a novel methodology to monitor the expansion kinetics of cellular materials produced with exothermic foaming mechanisms. 4 Own formulations of water-blown RPU foams Reactive foaming: Free expansion in a plastic cup 3 X-ray radioscopy validation of a polyol functionalized with graphene oxide for producing rigid polyurethane foams with improved cellular structures. 4 Own formulations of water-blown RPU foams from polyols functionalized with GO and polyols with GO dispersed GO Reactive foaming: Free expansion in a plastic cup 4 Evaluation of the thermal conductivity and mechanical properties of water blown polyurethane rigid foams reinforced with carbon nanofibers. 5 Commercial formulation of water-blown RPU foam CNFs Reactive foaming: Free expansion using a mould 5 Long-thermal conductivity of cyclopentanewater blown rigid polyurethane foams reinforced with different types of fillers. 5 Commercial formulation of cyclopentane-water blown RPU foam Talc (T), Non-porous silicas (NPS), Diatomeous earth (DE) Reactive foaming: Restricted expansion using a mould 6 Synthesis, characterization and physical properties of rigid polyurethane foams prepared with poly(propylene oxide) polyols containing graphene oxide. 6 Own formulations of water-blown RPU foams from polyols functionalized with GO GO Reactive foaming: Free expansion in a plastic cup 7 Improvement of thermal and mechanical properties by control of formulations in rigid polyurethane foams from polyols functionalized with graphene oxide. 6 Own formulations of water-blown RPU foams from a polyol functionalized with 1000 ppm of GO GO Reactive foaming: Free expansion in a plastic ckkup 8 Synthesis, characterization, and foaming of thermoplastic polyurethane with different hard segment contents. 7 Own formulation of TPU foam Gas dissolution foaming by 1-step in a autoclave 3.2- Polyurethane formulations Both commercial and own formulations of rigid polyurethane (RPU) foams were used in this thesis (Table 3.1). In addition, the study of TPU foams was carried out with our own 93 Materials, foam production and experimental techniques formulations of thermoplastic polyurethane (TPU) materials (Table 3.1). In the case of commercial formulations, we do not have a complete information about the characteristics and the amounts of the components included in each formulation. We only have the information provided by the company producing these formulations. 3.2.1- Commercial formulation of water-blown rigid polyurethane foam The commercial formulation of water-blown RPU foam, supplied by "BASF Poliuretanos Iberia S.A.", is formed by the polyol component (Elastopor H 1501/2) and by the isocyanate component (IsoPMDI 92140) [1, 2]. This formulation was chosen for two of the studies of this thesis because this is a typical PU formulation commonly used for the core material of sandwich insulating panels. Firstly, this commercial formulation was employed in order to evaluate the modification of the reaction kinetics by FTIR spectroscopy when different functional nanofillers (nanosilicas and nanoclays) were incorporated into water-blown RPU formulation (article 1 included in chapter 4, Table 3.1). Secondly, this commercial formulation was used to study both the thermal and mechanical properties of RPU foams reinforced with different contents of carbon nanofibers (CNFs) (article 4 included in chapter 5, Table 3.1). The main characteristics of this PU formulation, provided by BASF, are summarized in Table 3.2. Table 3.2. Features of commercial formulation of water-blown RPU foam from BASF Poliuretanos Iberia S.A (article 1 included in chapter 4 and article 4 included in chapter 5). Product Features Elastopor H 1501/2 (Polyol component) Chemical composition Polyether polyols, catalysts, flame retardant agents, stabilizers and blowing agent (water) OH index (mg KOH/g) 651 Viscosity (20°C)(mPa·s) 650 Density (20°C) (g/cm3) 1.07 IsoPMDI 92140 (Isocyanate component) Chemical composition Polymeric diphenylmethane diisocyanate (pMDI) Isocyanate content Weight (%) 31.5 Viscosity (20°C)(mPa·s) 170-250 Density (20°C) (g/cm) 1.23 Foam Polyol/isocyanate ratio (by weight) 100/160 Density (Kg/m3, free foaming) 52±5 Cream time (s) 45±7 Gel time (s) 145±18 Rise time (s) 232±21 3.2.2- Commercial formulation of cyclopentane-water blown rigid polyurethane foam The same formulation included in Table 3.2 has been the starting point to develop a cyclopentane-water blown RPU foam. The proportions of the three components used were set at 100/160/13 by weight for the polyol (Elastopor H 1501/2), isocyanate (IsoPMDI 92140) and cyclopentane (99.9% purity from Sigma Aldrich) in order to have a free foaming density of 30 kg/m3. This RPU system has a low thermal conductivity due to its low density, and also to the 94 Chapter 3 use of CP as physical blowing agent. Due to its excellent insulating properties, this formulation was chosen to produce RPU foams with different types of fillers, such as talc (T), Diatomaceous earth (DE) or non-porous silica (NPS), and to study the thermal conductivity aging during these years (article 5 included in chapter 5, Table 3.1). 3.2.3- CellMat formulations of water-blown rigid polyurethane foams RPU formulations were developed in our laboratory in order to have a better control on the foam properties and to have the possibility of understanding the relation between formulation, reaction kinetics, cellular structure and properties. The chemical and physical characteristics of several formulation components (isocyanate, surfactant, catalysts and blowing agent) used to produce the RPU foams are collected in Table 3.3. On the other hand, Table 3.4 shows the characteristics of Alcupol® R4520, a commercial polyol supplied by Repsol S.A., and collects four non-commercial polyols synthesized by Repsol in order to carry out several works of this thesis: a polyol without GO (to obtain the reference material), and three polyols chemically functionalized with low amounts of GO (GO-f) (500, 1000, and 2500 rpm). Table 3.3. General features of formulation components used to produce own RPU foams. Components Commercial product Features of PU components Isocyanate IsoPMDI 92140 Supplier BASF Poliuretanos Iberia S.A Chemical composition Polymeric diphenylmethane diisocyanate (pMDI) Isocyanate content Weight (%) 31.5 Viscosity (20°C)(mPa·s) 170-250 Density (20°C) (g/cm3) 1.23 Surfactant TEGOSTAB® B 8522 Supplier Evonik Chemical composition Non - hydrolysable poly-ether - polydimethyl-siloxane Specific gravity (25 °C) (g/cm³) 1.035 - 1.055 Viscosity (25°C)(mPa·s) 600 - 1 000 pH value (4 % hydrous solution) 6.0 - 8.0 Catalyst TEGOAMIN® DMCHA Supplier Evonik Chemical composition N,N-dimethylcyclohexylamine (DMCHA) Content (%) ≥ 98.0 Water content (%) ≤ 0.25 TEGOAMIN® PMDETA Supplier Evonik Chemical composition N,N,N',N'',N''- Pentamethyldiethylenetriamine (PMDETA) Blowing agent Distilled water 95 Materials, foam production and experimental techniques Table 3.4. General features of a commercial polyol from Repsol and non-commercial polyols from Repsol used to produce RPU foams with own formulations. Commercial product Features of polyols Alcupol® R4520 Supplier Repsol S.A. Chemical composition Polyether polyol Starter Sucrose/ Glycerol Functionality 4.5 OH index (mg KOH/g) 455 Viscosity (25°C)(mPa·s) 5250 Molecular weight (g/mol) 555 Density (25°C) (g/cm3) 1.08 Water content (%) 0.1 Alkaline content (ppm) ≤ 130 Non-commercial product Features of polyols Polyol: 0 ppm GO-f (pure poliol) Supplier Repsol S.A. Chemical composition Polyether polyol Starter Glycerol Functionality 3 OH index (mg KOH/g) 426.4 Viscosity (25°C)(mPa·s) 476.85 Molecular weight (g/mol) 488 Polyol: 500 ppm GO-f Supplier Repsol S.A. Chemical composition Polyether polyol Starter Glycerol Functionality 3 OH index (mg KOH/g) 399.3 Viscosity (25°C)(mPa·s) 553.30 Molecular weight (g/mol) 494 Polyol: 1000 ppm GO-f Supplier Repsol S.A. Chemical composition Polyether polyol Starter Glycerol Functionality 3 OH index (mg KOH/g) 358.6 Viscosity (25°C)(mPa·s) 547.59 Molecular weight (g/mol) 570 Polyol: 2500 ppm GO-f Supplier Repsol S.A. Chemical composition Polyether polyol Starter Glycerol Functionality 3 OH index (mg KOH/g) 377.5 Viscosity (25°C)(mPa·s) 4463.23 Molecular weight (g/mol) 675 Table 3.5 summarizes four formulations developed during the research work. From the formulation collected in Table 3.5, four different RPU foams were obtained, which differ in the amount of water (2 or 5 ppw) and in the blowing catalyst used (TEGOAMIN® PMDETA) (0 or 2 ppw). These materials were selected in order to describe a new methodology called “Infrared Expandometry”, appropriate to characterize simultaneously the expansion kinetics (height vs. time, and volume vs. time) and the surface temperature evolution of cellular materials (article 2 included in chapter 4, Table 3.1). Table 3.6 shows two series of RPU foams formulations: one contains polyols functionalized with GO (GO-f), and the second one GO dispersed (GO-d) into the pure polyol. A pure polyol without GO was used to obtain a pure material as a reference for both series. These formulations were used in order to study the effect of both GO-f and GO-d, on the foaming mechanisms by using X-ray radioscopy (article 3 included in chapter 4, Table 3.1). 96 Chapter 3 This previous work allowed concluding that incorporating GO chemically linked to the polyol chains clearly improves the characteristics of RPU foams nanocomposites. Therefore, we decided to produce water-blown RPU foams from polyols functionalized with low amounts of GO-f (0.017, 0.033 and 0.088 wt. % in the foam) following the formulations collected in Table 3.7. The influence of the amount of GO-f on density, cellular structure, thermal conductivity and mechanical properties were investigated (article 6 included in chapter 6, Table 3.1). Furthermore, a detailed study of foaming process was carried out by FTIR spectroscopy, infrared expandometry and temperature evolution, in order to evaluate the effect of GO-f on the reaction kinetics and foam formation process. The main conclusion of this work was that the better performance was displayed by the foam containing 0.033 wt% GO-f, because it reduced the thermal conductivity. However the mechanical properties were not improved in comparison to the reference foam. Finally, this optimum system with 0.033 wt% GO-f was further studied in the last work herein collected, in order to improve their mechanical properties without loss of their thermal properties (article 7 included chapter 6 shown in Table 3.1). For this purpose, the data obtained from the previous modifications of the reaction kinetics induced by the presence of GO in the polyol, led us to obtain a series of RPU foams by varying the isocyanate index, the amounts of catalyst, the amounts of surfactant, or a combination of these components. The formulations used for this purpose are included in Table 3.8. Table 3.5. Own formulations of water-blown RPU foams (article 2 included in chapter 4). Isocyanate component Polyol component (ppw) Isocyanate index Polyol Surfactant Catalyst Blowing agent Samples IsoPMDI 92140 Alcupol® R4520 TEGOSTAB® B 8522 TEGOAMIN® DMCHA TEGOAMIN® PMDETA Water PUR1 110 100 1 1 0 2 PUR2 110 100 1 1 2 2 PUR3 110 100 1 1 0 5 PUR4 110 100 1 1 2 5 Table 3.6. Own formulations of water-blown RPU foams from polyols functionalized with GO and polyols with GO dispersed (article 3 included in chapter 4). Isocyanate component Polyol component (ppw) Isocyanate index Polyol Surfactant Catalyst Blowing agent Samples IsoPMDI 92140 0 ppp GO-f 500 ppm GO-f 2500 ppm GO-f TEGOSTAB® B 8522 TEGOAMIN® DMCHA Water Pure (Reference) 120 100 1 0.3 5 0.017%GO-f 120 100 1 0.3 5 0.083%GO-f 120 100 1 0.3 5 0.017%GO-f 120 100 1 0.3 5 0.083%GO-d 120 100 1 0.3 5 97 Materials, foam production and experimental techniques the additional amount of MDI and the prepolymer previously obtained in DMAc (240 mL), was added dropwise from the addition funnel to a preheated mixture of 1,5-PDO and DABCO (0.3 g, 0.003 mmol, 0.3%) in DMAc (60 mL). This reaction mixture was stirred at 400 rpm for 2h into an oil bath at 80°C. Finally, the solution containing the TPU was poured into silicone moulds without exclusion of air and maintained in an oven at 80°C for 3 days in order to obtain TPU casts. Figure 3.2. Scheme of TPU synthesis by pre-polymer method. •Samples preparation. Extruded samples with 1.5 ± 0.5 mm of diameter were prepared by TWELVindex extrusion plastometer from ATS Faar using a temperature above melting temperature (between 165-180 °C) in order to carry out the foaming tests. Compression molded samples were prepared by using a hot plate press in order to characterize their using different techniques: shore hardness, WAXD, DMTA and rheology (Table 3.15). The material was first heated at a temperature above melting temperature (between 165-180 °C) for 3 min, raising the pressure to 10 MPa at 0.05 MPa/s. Then the samples were pressed under a constant pressure of 10 MPa for 7 min, lowering the temperature to 60°C at 25°C/min. MDI excess Polyol (PEG-PPG-PEG) Prepolymer Prepolymer MDI DMAc Chain extender DABCO DMAc TPU solution in DMAc 80°C for 2 hours N 2 , 400 rpm 80°C for 2 hours N 2 , 400 rpm 1 st STEP 2 nd STEP TPU casts in a silicone mold SOLVENT CASTING 80°C for 3 days in a oven 104 Chapter 3 •Foamed of solid TPUs. The extruded samples were foamed by a one-step gas dissolution foaming process using CO2 as physical blowing agent (Figure 3.3). A high pressure vessel provided by Parr Instrument Company (model PARR 4681), with a capacity of 1 L, and capable of operating at a maximum temperature of 350° C and a maximum pressure of 41 MPa was used to obtain the TPU foams. The pressure was automatically controlled by an accurate pressure pump controller (model SFT-10) provided by Supercritical Fluid Technologies Inc. The vessel is equipped with a clamp heater of 1200 W, and the temperature is controlled by a CAL 3300 temperature controller. This set up has been used to carry out a set of experiments by using a one-step foaming process. Samples were first introduced in the pressure vessel under certain pressure and temperature conditions for the saturation stage. After saturation, pressure was rapidly released and the samples expanded in the pressure vessel after depressurization. Two sets of experiments were performed. First, the effect of the foaming temperature was analyzed by fixing the saturation pressure to 20 MPa and the saturation time to 1 h, whereas varying the foaming temperature at 140, 150, 160, 170 and 180 °C. Second, the influence of the saturation pressure was evaluated by choosing four different saturation pressures: 10, 15, 20 and 25 MPa at fixed temperature and time: 170 °C for 1 h. Extruded TPU samples were foamed and then the remaining CO2 was allowed to desorb before measuring their density and characterizing their cellular structure. Figure 3.3. Scheme of TPU foam production by 1-step gas dissolution foaming process in an autoclave. Saturation Pressure release CO2 Foaming in the autoclave Saturation pressure: 20 MPa Saturation temperature: 140-180°C Time:1h Temperature: 140-180°CFoaming temperature: 140-180°C 1-Step foaming Autoclave Autoclave Autoclave Saturation pressure: 10-25 MPa Saturation temperature: 170°C Time:1h Temperature: 170°C 1 Foaming temperature: 170°C 2 Effect of foaming temperature (TPUs with 40, 50 and 60 wt.%HS) Effect of saturation pressure (TPU with 50 wt.%HS) Foamed TPU TPU 105 Materials, foam production and experimental techniques 3.5- Experimental techniques 3.5.1- Summary of experimental techniques Several standard and well-known experimental techniques were employed in order to characterize the materials studied in this research work. A full description of these experimental techniques has been already provided in the scientific articles included in this thesis, therefore only a brief description of them is collected in Tables 3.12 to 3.15. Instead, detailed descriptions of those experimental techniques used to monitor the reaction kinetics of the RPU foams (FTIR spectroscopy, X-ray radioscopy, infrared expandometry and adiabatic temperature rise, Table 3.12.) are herein detailed, since these techniques are the most important ones in this research. Table 3.12. Summary of the experimental techniques used for the characterization and the monitoring of foaming process of RPU foams. CHARACTERIZATION OF RPU FOAMS EXPERIMENTAL TECHNIQUE Purpose Chapter Density determination (Geometric method) ASTM D1622/D1622M-14 [20] Obtain the foam density 4, 5 and 6 Open cell content determination (Gas picnometry) ASTM D6226-10 [21] Gas pycnometer Accupyc II 1340 from Micromeritics Calculate the open cell content 4, 5 and 6 Scanning Electron Microscopy (SEM) JEOL JSM-820 microscope. Determine the average cell size (Ф 3D ), anisotropy ratio (AR), the standard deviation (SD) of the cell size distribution and the normalized standard deviation (NSD) 4, 5 and 6 Optical Microscopy (OM) Leica DM2500M microscope Locate the position of CNFs in the foam structure 5 Thermal conductivity by transient method ISO 22007 ‐ 2:2008 [22] Hot-disk transient plane source (TPS) thermal constant analyzer Measure the thermal conductivity 5 Thermal conductivity by stationary method UNE12667 [47] Rapid K heat flowmeter from Holometrix Measure the thermal conductivity 5 Mechanical tests in compression ASTM D1621-10 [44] Instron Machine (model 5.500R6025) Measure Young´s modulus and collapse stress. 4, 5 and 6 Fourier transform infrared (FTIR) spectroscopy by transmission Bruker Tensor 27 spectrometer Calculate the spectral extinction coefficient 5 Fourier transform infrared (FTIR) spectroscopy by transmission and attenuated total reflectance (ATR) Bruker Tensor 27 spectrometer Characterize the premix of the nanoparticles with the isocyanate component 4 Dynamical mechanical analysis (DMA). PerkinElmer DMA7 dynamic mechanical analzer Analyze the viscoelastic behavior of foamed samples. Measure the sample thickness to calculate the spectral extinction coefficient 4 and 5 FOLLOW THE FOAMING PROCESS OF RPU FOAMS EXPERIMENTAL TECHNIQUE Purpose Chapter FTIR spectroscopy by ATR Bruker ALPHA spectrometer Follow the isocyanate conversion and the generation of products during the foaming process 4 and 6 Infrared expandometry Infrared camera (model Hotfind L) from SDS Follow the foam expansion to determine the surface temperature, height and volume evolution 4 X-ray radioscopy Home-designed equipment Follow the cellular structure evolution, measuring the relative density, cell size and cell nucleation density evolution 4 and 6 Adiabatic temperature rise Follow the temperature evolution to extract complementary information about the reactions 4 and 6 106 Chapter 3 Table 3.13. Summary of the experimental techniques used for the characterization of polyol functionalized with GO. CHARACTERIZATION OF POLYOL FUNCTIONALIZED WITH GO EXPERIMENTAL TECHNIQUE Purpose Chapter Hydroxyl number determination ASTM D-4274 [23] Determine hydroxyl numbers 6 Viscosity determination Brookfield DV-III ULTRA Rheometer. Determine viscosity 6 Differential scanning calorimeter (DSC) DSC TA Instruments Q2000. Measure the glass transition temperature (Tg). 6 Gel-permeation chromatography (GPC) Bruker 3800 equipped with a deflection RI detector. Measure the weigh-averaged molecular weights (M W ) and polydispersity index (PDI). 6 Table 3.14. Summary of the experimental techniques used for fillers characterization. CHARACTERIZATION OF FILLERS EXPERIMENTAL TECHNIQUE Purpose Chapter Fourier transform infrared (FTIR) spectroscopy by attenuated total reflectance (ATR) method Bruker Tensor 27 spectrometer Identify the surface groups of nanoclays and nanosilicas 4 Thermogravimetric analysis (TGA) Mettler Toledo TGA/SDTA 851 Analyze the thermal decomposition of nanoclays and nanosilicas 4 Environmental scanning electron microscope (ESEM) Quanta 200 FEG Examine the morphologies of nanoclays and nanosilicas 4 Scanning Electron Microscopy (SEM) Zeiss EVO LS15 Characterize the dimensions of CNFs 5 Table 3.15. Summary of the experimental techniques used for solid TPU and TPU foams. CHARACTERIZATION OF SOLID TPU AND TPU FOAMS EXPERIMENTAL TECHNIQUE Purpose Chapter Density determination (Gas picnometry) Gas pycnometer Accupyc II 1340 from Micromeritics Density of the solid TPUs 7 Shore hardness A and D ISO 868:2003 [24] Bareiss U72 durometer Analyze the difference in the shore hardness of the solid TPUs 7 Gel-permeation chromatography (GPC) Analyze the difference in the molecular weight of the solid TPUs 7 Differential scanning calorimetry (DSC) DSC30 Mettler Toledo Instrument. Detect variations in the glass transition temperature of the solid TPUs 7 Wide angle X-ray diffraction (WAXD) PANalytical X’Pert Pro (XRD 5) Instrument. Analyze the difference in the crystallinity of the solid TPUs 7 Dynamic mechanical thermal analysis (DTMA) Q800 DMA instrument (TA) Analyze the viscoelastic properties of the solid TPUs 7 Shear rheology ARES-G2, TA Instruments Characterize the rheological properties of the solid TPUs 7 Density determination (Archimedes method) (Precision Balance AT 261 from Mettler) Obtain the density of TPU foams 7 Scanning Electron Microscopy (SEM) JEOL JSM-820 microscope. Determine the average cell size (Ф 3D ), anisotropy ratio (AR), the standard deviation (SD) of the cell size distribution, the normalized standard deviation (NSD) and cell nucleation density (N0) of TPU foams. 7 107 Materials, foam production and experimental techniques 3.5.2- Experimental techniques for the monitoring of foaming process 3.5.2.1- FTIR spectroscopy Aiming at investigating the reaction kinetics of RPU foams, a methodology based on in-situ FTIR spectroscopy by ATR was established in the article 1 included in chapter 4 of this thesis and was also applied in article 6, included in chapter 6 of this thesis. The RPU foam formation may be monitored by in-situ FTIR by following the isocyanate consumption from the isocyanate asymmetric stretching vibration at 2270 cm⁻1. The generation of urethane and urea products associated with the gelling and blowing reactions can be also deduced from the carbonyl stretching vibrations of the Amide I or carbonyl region (in the range of 1610-1760 cm⁻1). Therefore, only the main reactions, that are the most critical ones in most of the cases, have been considered for the analysis of products. A Bruker ALPHA spectrometer working in ATR method was employed in order to collect the insitu FTIR spectra of the samples. The measurements were carried out following the next steps (Figure 3.4): 1. Sample preparation: A low amount of reacting foam (ca. 1 mL of the reactants mixture) extracted from a mixture made with the usual amounts to produce the foams with higher dimensions in a plastic cup (as explained in the section 3.3.1 of this chapter), was poured on the ATR surface. The addition of the isocyanate was taken as time t = 0. 2.Measurement parameters of FTIR spectra: FTIR spectra were measured during 30 min each 20 or 30 sec. (16 scans, 4 cm-1 of resolution in the range 4000-400 cm⁻1). The sample was maintained during all the experiment at 70°C in order to reproduce the high temperatures characteristic of the exothermic foaming process of RPU foam with higher dimensions. 3. Processing of FTIR spectra: First, a background spectrum was deduced from each spectrum. Second, baseline correction was conducted in order to correct the intensity shifts at lower frequencies. The asymmetric CH stretching band at 2972 cm⁻1 (whose concentration remains constant during the reaction) was used as internal reference band to correct the concentration or density changes during the foaming process [25-28]. 4.Isocyanate analysis: The isocyanate conversion during the foaming process was quantified by the area decay of the isocyanate absorption band. The limits chosen to carry out these measurements are 2500 and 2000 cm⁻1 (explained in the article 1 included in chapter 4). 5.Amide I or carbonyl region analysis: The carbonyl absorptions of the urethane and urea groups obtained from the blowing and gelling reactions were identified by their second derivative technique, and they were separated and quantified by deconvolution using PeakFit program (Figure 3.5). The deconvoluted region (1601-1760 cm⁻1) was chosen in order to avoid the absorption effects of other side bands from the aromatic C=C ring stretching and/or from amide II bands. Gaussian bands were used to be deconvoluted [29, 30], which finished once a correlation coefficient with at least three nines was obtained in the curve fittings. The parameter used to quantify the different bands contained in the carbonyl region and the 108 Chapter 3 global urea and urethane content are detailed in the article 1 included in chapter 4. Quantitative analysis was performed in this study assuming that the urethane and urea compounds in carbonyl region present similar extinction coefficients. Figure 3.4. Scheme of the procedure followed to carry out the FTIR measurements with time. Figure 3.5. Example of deconvolution of Amide I region. Hidden bands in the carbonyl region associated with blowing and gelling reactions and their corresponding structure. Reactive mixtureFTIR spectra evolution with time Reacting foam poured over the ATR surface Gelling reaction peaks Free urea (1700-1690 cm-1) C=O hydrogen-bonded and N-H bonded to an ether group (1670-1660 cm-1) Ordered hydrogen-bonded urea (1650-1640 cm-1) Disordered hydrogenbonded urea (1690-1650 cm-1) Free urethane (1740-1730 cm-1) Hydrogen-bonded urethane hard segment/soft segment interaction (1715-1700 cm-1) Hydrogen-bonded urethane hard segment/soft segment interaction (1730-1725 cm-1) Blowing reaction peaks 109 Materials, foam production and experimental techniques 3.5.2.2- Infrared expandometry A novel experimental technique named “Infrared Expandometry” was defined and introduced in the article 2 included in chapter 4 of this thesis. It allows monitoring simultaneously the expansion kinetics (height vs. time and volume vs. time), and the evolution of surface temperature of the RPU foams. A infrared camera detects the infrared radiation emitted by the foam during the exothermic foaming process, and transforms the infrared radiation into luminous images visible by the human eye. The acquired images are then analyzed in order to obtain quantifiable information. This technique was also applied in the article 6 included in chapter 6 of this thesis in order to study the differences of the expansion kinetics of RPU foams with different types of GO. In order to record the foam expansion, an infrared camera (model HotFind L) from SDS Infrared was employed. Other experimental details of this technique for the expansion characterization of RPU foams are detailed next (Figure 3.6): 1. Sample preparation: The foams are produced in a plastic cup, as explained in the section 3.3.1 of this chapter, and situated at 1,00 m of the infrared camera in order to record the foam expansion. The addition of the isocyanate to the polyol component was taken as time t = 0. From this moment, 25 frames per second (fps) were recorded during 7 minutes for every sample. 2. Temperature calibration: The infrared camera detects the surface temperature, which depends on the atmosphere transmittance. The latter is conditioned by the environmental conditions (mainly temperature and relative humidity of the surrounding atmosphere) and by the emissivity of the samples. Therefore, the environmental conditions were measured for each experiment, and the emissivity of the PU foams under study was fixed at 0.82 during the whole process. 3. Image analysis: Once the infrared images sequence of each sample has been acquired (see an example in Figure 3.7), an image analysis methodology based on ImageJ/Fiji software tool [31, 32] was implemented in order to extract the quantitative information of the foaming process. The main steps of the image analysis methodology are the following: (1) the 25 frames corresponding to one second were condensed into an average for these 25 images, (2) a binarization process after a filter application, and (3) binarized images were used to determine the quantitative parameters of the foaming process. Therefore, this technique allows evaluating the evolution of height, volume, expansion rate, expansion acceleration, and superficial temperatures (minimum, maximum and average), as well as the inhomogeneities in the sample surface. More details of the image analysis protocol and the parameters determined are detailed in the article 2 included in chapter 4 of this thesis. 110 Chapter 3 Figure 3.6. Scheme of the infrared radiation detection system in order to record the expansion of PU foam in a plastic cup. Figure 3.7. Example of infrared images at different times of the foaming process. 3.5.2.3- X-ray radioscopy X-ray radioscopy, in which sequences of radiographies are acquired during foaming, has been demonstrated to be a useful tool for the in-situ inspection of the foaming process of PU materials. Several articles have been previously published in our laboratory on this methodology [33-35], which was applied in the article 4 included in chapter 4 of this thesis. A home-designed equipment of X-ray was employed in order to follow the foaming process of RPU foams by X-ray radioscopy. The main experimental aspects of this technique are detailed next (Figure 3.8): 1. Sample preparation: The usual amount of isocyanate and polyol components were mixed in a plastic cup, as explained in the section 3.3.1 of this chapter. Then 0.020 ± 0.009 mL of the reaction mixture were deposited on a specific foaming mould designed to perform the X-ray radioscopy experiments. This specific foaming mould is made of stainless steel, and consists on a central cylindrical hole (Ø = 6mm, thickness = 0.6 mm) where the reaction mixture was 111 Materials, foam production and experimental techniques poured. The hole is enclosed between two polypropylene plastic covers (25 μm thick) and the sample thickness is invariable during the process due to two lateral evacuation conduits. Two materials of known relative density (air, relative density = 0, and a TPU disc of the same thickness, relative density = 1) are situated into two smaller holes at each side of the main one, and scanned with the evolving sample to calibrate in-situ the X-ray acquired signal. Typically, lower amount of catalyst (0.3 ppw) is used in these studies because the cream time must be increased, in order to properly observe the foaming behaviour with the X-ray setup. 2. Measurement parameters of X-ray radioscopy: The X-ray tube was adjusted at 40 kV and 120 μA in order to obtain the optimum contrast of the acquired radiographies. Moreover, an exposure time of 800 ms was selected in order to record motionless and sharp images. A total of 700 radiographies were acquired, corresponding to ca. 9 min of the foaming process. A initial period of 50s is needed in order to place the foaming frame inside the X-ray cabinet. 3. Image Analysis: Once the radiographies of each sample were acquired (see an example in Figure 3.9), an image analysis protocol based on ImageJ/Fiji software tool [37,38] was applied to quantify the relative density, the cell size, and the nucleation cell density. All the details of the image analysis protocol and the parameters determined are detailed in the article 3 included in chapter 4 of this thesis. Figure 3.8. Scheme of the system employed to perform the X-ray radioscopy experiments. 112 Chapter 3 Figure 3.9. Example of X-ray radiographies obtained for two RPU foams at 75 and 420 seconds of the foaming process. 3.5.2.4- Adiabatic temperature rise Adiabatic temperature rise is a technique used to follow the foaming temperature evolution with time. It is directly measured during the foaming process by thermocouples inside the foam, since the gelling and blowing reactions are exothermic. Because this is a fast and easy technique, several articles have been used to obtain information about the foaming process [36-40]. This technique was applied in the article 2 (chapter 4), in the article 5 (chapter 5) and in the article 6 (chapter 6). In order to measure the foaming temperature evolution, several thermocouples type K were introduced in the plastic cup which contains the reactive mixture (produced as explained in section 3.3.1 of this chapter). The thermocouples are placed vertically in the center of plastic cup at the different heights from the base (Figure 3.8). The data collected by the thermocouples were registered in a computer. Figure 3.8. Scheme of the system employed to follow the foaming temperature reached with time. The position and number of thermocouples were fixed for each formulation of this thesis. 3.6- References [1] IsoPMDI 92140 Technical Data Sheet. BASF Poliuretanos Iberia S.A. [2] Elastopor H 1501/2 Technical Data Sheet. BASF Poliuretanos Iberia S.A. 2000 μm 160 s 2000 μm 46 s 2000 μm 90 s 2000 μm 450 s Foam expansion Thermocouples at different heights 113 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams Figure 4.1. Scheme of the techniques used to follow the foaming process of RPU foams. In the first article, the first experimental method to follow the foaming process is discussed. Insitu FTIR spectroscopy is used to monitor the evolution of both the reactions and the morphology of the PU matrix for RPU foams reinforced with different types of nanosilicas or nanoclays (5 wt%). For this purpose, a low amount of reacting foam (around 1 mL) extracted from a higher amount of reactive mixture, is poured on the ATR surface. The contribution of both blowing and gelling reactions (urea and urethane compounds) are studied by deconvolution of the carbonyl region. The FTIR results demonstrate how the different types and surfaces (hydrophobic and/or hydrophilic) of nanoparticles affect the reaction kinetics on different ways. The foams with nanosilicas show higher isocyanate conversion than those with nanoclays, being the conversion higher when hydrophobic groups are present on their surface. Moreover, hydrophobic nanoparticles promote the gelling reaction generating higher urethane groups, whereas hydrophilic nanoparticles enhance the blowing reactions increasing the urea groups. On the other hand, the effect of the modification of the reaction kinetics on the density, cellular structure, thermal conductivity and mechanical properties of RPU foams are also discussed. Infrared expandometry, a novel experimental technique, developed in the second article, for monitoring the expansion process of a RPU foam, by taking advantage of their exothermic foaming process. In order to test this methodology, a RPU formulation has been chosen where the blowing agent (water) and/or the blowing catalyst are systematically changed. Thus, the expansion evolution of the foam can be followed from the initial to final times without any restrictions in the sample size. For example, the foaming process is monitored in this article in a plastic cup (15 cm of height by 10 cm of diameter). Important parameters of the foaming process are determined using the approach developed in this paper, such as the evolution of TECHNIQUES TO MONITOR THE FOAMING PROCESS OF RIGID POLYURETHANE FOAMS Chemical reactions and morphology of PU matrix In-situ FTIR spectroscopy Infrared Expandometry In-situ X-ray radioscopy Expansion process Internal microstructure Isocyanate Carbonyl region Article 1 Article 2 Article 3 Pure 75s 420s 0.017% GO-f 0.017% GO-d 0.083% GO-f 0.083% GO-d 75s 420s 75s 420s Pure 75s 420s 0.017% GO-f 0.017% GO-d 0.083% GO-f 0.083% GO-d 75s 420s 75s 420s 1500 μm Wavenumber (cm-1) Time (s) Absorbance (a.u) 120 Chapter 4 height, volume, expansion rate, acceleration rate and surface temperature, among others. Moreover, the measurement of surface temperature can be quantified at certain areas of the surface or calculated for the whole sample surface, in both cases without interfering with the expansion process, what is a remarkable advantage compared to the traditional measurement of internal temperature using thermocouples. Thus, this article demonstrates the potential of this technique in order to follow the expansion kinetics of PU foams. In-situ X-ray radioscopy is used in the third article to visualize the evolution of the internal microstructure of the foams during their production, obtaining relative density, cell size and cell nucleation density vs. time. These data are obtained by using a very low amount of the reacting components (around 0.02 mL), which is poured on the foaming mold after being extracted from a higher amount of reactive mixture. In the article, this technique has been used in order to compare the foaming behavior of two series of foams: those prepared from polyols functionalized with GO (GO-f) and those containing GO dispersed in the polyol (GO-d) by high shear mixing (loading of 0.017 and 0.083 wt%). The results obtained allow us evaluating which is the preponderant mechanism (nucleation or degeneration) in these systems. The foams containing GO-f present a cell size reduction as the amount of GO-f increases, whereas in the foams containing GO-d there is a decrease of cell size at low contents, because it is most difficult to achieve a proper dispersion of the fillers when the amount increases. Although the systems with GO-f present a high coalescence, the enhanced nucleation experienced by these foams prevails at final stages over the cell degeneration processes. The results obtained demonstrate that the use of polyol functionalized with GO-f highly improves the cellular structure and also makes the results more reproducible. Taking into account the objectives of this thesis (chapter 1, section 1.2), the results obtained from these three works have allowed establishing and validating a methodology for monitoring the reaction and expansion kinetics, the morphology of the polymeric matrix, and the internal structure during foaming process of PU foams. Moreover, this methodology has allowed us understanding the effect produced by the fillers on the foaming process. This knowledge is fundamental in order to analyze and to understand the cellular structure and properties of the foamed materials. 121 Chapter 4 Polymer 150 (2018) 138-149 https://doi.org/10.1016/j.polymer.2018.07.029 The effects of functional nanofillers on the reaction kinetics, microstructure, thermal and mechanical properties of water blown rigid polyurethane foams Mercedes Santiago-Calvo1,*, Josías Tirado-Mediavilla1, José Luis Ruiz-Herrero1, Miguel Ángel Rodríguez-Pérez1, Fernando Villafañe2 1 Cellular Materials Laboratory (CellMat), Condensed Matter Physics Department, Faculty of Science, University of Valladolid, Campus Miguel Delibes, Paseo de Belén 7, 47011 Valladolid, Spain 2 GIR MIOMeT-IU Cinquima-Química Inorgánica. Faculty of Science, University of Valladolid, Campus Miguel Delibes, Paseo de Belén 7, 47011 Valladolid, Spain * Corresponding author: [email protected]a.es Abstract The use of functional nanofillers to improve the properties of rigid polyurethane (PU) foams has caused the need for a better understanding of how these nanofillers modify the reaction kinetic of the PU system. In this study, different nanoclays and nanosilicas are used as functional nanofillers. Analysis of the kinetic data obtained by in-situ FTIR spectroscopy monitoring allows to correlate the isocyanate consumption with the type of nanoparticles. The quantification of urethane and urea, obtained by deconvolution of the carbonyl region absorptions, enables to follow the blowing and gelling reactions during the foaming process. These reactions are correlated to the nature of the chemical groups present on the surface of the nanoparticles added. In addition, the effect of the modification of the reaction kinetics on the density, cellular structure, thermal conductivity and mechanical properties is herein discussed. Keywords: Polyurethane foam; nanoclay; nanosilica; reaction kinetics; thermal properties; mechanical properties 122 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams 1. Introduction Polyurethanes (PUs) are multipurpose polymers, since they are present in a wide range of industrial sectors, like automotive, medical, construction, furniture, appliances, etc [1]. PUs can be found commercially as solid thermoplastics (known as TPU), coatings, adhesives, sealants, binders, elastomers, and foams [2]. PUs foams are usually classified according to their mechanical behavior as flexible foams (such as those used for furniture, mattresses or automotive seats), or rigid foams (which can be applied for insulation and structural materials) [3]. PU foams represent almost a 50% of the global market of foams, and their use is still substantially growing over the years [4]. Thermal management in buildings and transportation insulation or refrigeration systems are the core applications of rigid PU foams with closed cells [Figure 1(a)]. These applications have been extensively studied both theoretically [5-7] and experimentally [8-11]. Incorporating additives to the PU formulation is an excellent method to improve the mechanical, thermo-mechanical, or thermal properties of the foam. Nanoscale size fillers are becoming a promising option in order to obtain improved properties [12, 13]. For a PU foam with closed cells, with a density of 50 kg/m3, the cell wall thickness is usually in the range of 1 to 2.5 microns [Figure 1(c)], and therefore the typical sizes of cells and struts are slightly higher [Figures 1(a) and 1(b)] [14, 15]. These features should be taken into account in order to select an appropriate filler for modifying its physical properties, and a logical approach is adding fillers with at least one of the dimensions clearly below the thickness of the cell walls (i.e. smaller than 1 micron). 123 Chapter 4 Figure 1. SEM micrographs of the microstructure of a closed cell rigid PU foam: (a) Microstructure formed by closed cells. The components of closed cells are (b) struts, and (c) cell walls with thickness in the range of 1 to 2.5 microns. These reasons have encouraged the research on the incorporation of nanofillers into a variety of materials and foams [16, 17]. These works have allowed concluding that the properties on the final composite depend, among other aspects, on the nanofiller chemical composition and on the size and shape of the nanoparticles [18, 19]. So far, the reinforcing inorganic materials probably more extensively used are layered nanoclays, especially those based on montmorillonite, [(Na,Ca)0.33(Al,Mg)2(Si4O10)(OH)2•nH2O]. This is due to their cation exchange capacity, high surface and reactivity, and adsorptive properties. The structure of montmorillonite is based on a central octahedral sheet of alumina, sandwiched between two external silica tetrahedral sheets. The interlayer galleries (of about 1 nm in thickness and 100 nm in width and length) are occupied by cations (usually Na+ and Ca2+). The presence of these cations in the interlayer spaces makes montmorillonite hydrophilic, which is the main reason for its poor compatibility with organic polymers. Thus, metallic cations must be substituted by hydrophobic organic cations (alkylammonium or - phosphonium quaternary salts) in order to both enhance a better layer separation, and to improve compatibility with the polymer matrix. Introducing nanoclays based on montmorillonite into PUs allows overcoming some of the weaker aspects of these foams, such as their relatively poor thermal stability, or their low gas-barrier properties. Thus, these type of PU-montmorillonite composites have been studied,[14, 20-22] and many investigations have been focused on their methods of preparation, structure, hydrogen bonding, degree of clay dispersion or exfoliation, microphase morphology, rheology, mechanical properties, thermal stability, flame retardancy, water sorption, or barrier properties [23]. On the other hand, the number of papers describing polymers reinforced with dispersed silica nanoparticles is also high [24, 25]. Nanosilica surface is hydrophilic due to the presence of hydroxyl groups, which generate hydrogen bonding and therefore significant interactions between particles. For this reason, nanoparticles tend to agglomerate, what should be precluded in order to favor the dispersion of nanoparticles into the polymer matrix. The degree of nanoparticles dispersion and the interfacial adhesion are important drawbacks to obtain appropriate materials, [26] and surface modification is the best way to reduce the preference for agglomeration of nanosilica particles [27]. Thus, the replacement of silanol groups by non-polar silyl groups removes the active hydrogen bonds on the surface, and changes the hydrophilicity of the filler surface into hydrophobic. The studies carried out on PU/nanosilica composites reveal that the addition of surface modified nanosilicas can improve the thermal, rheological, mechanical, and adhesion properties of PUs [17]. The properties of composites resulting from the addition of either nanoclays or nanosilicas to PU foams have been studied to some extent, [21, 24] but most of the reports are mainly focused on the production and characterization of the foams, paying particular attention to the nanoparticles dispersion. However, the effect of these additives on the chemical reactions occurring during foaming has been very scarcely explored. PUs are obtained when isocyanate 124 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams react with active hydrogen containing compounds, such as the hydroxyl groups of long polyether or polyester chains (polymerization or gelling reaction), or water (blowing reaction) to give urethane or urea linkages respectively. These two simultaneous processes should be adequately controlled in order to obtain a polymer foam with the desired cellular structure and physical properties [3, 28, 29]. The presence of additives such as nanofillers may modify the kinetics of these chemical reactions, by increasing or decreasing reaction ratios leading to a modification of urethane and/or urea linkages, thus affecting the foaming process and the final morphology of the PU matrix, and consequently the properties of the final foam. The modification of the reaction kinetics, microstructure and physical properties of PU systems containing nanofillers is a topic which depends on many interconnected factors, and therefore requires a high degree of systematization. The aim of this work is to study systematically the effect of different nanoparticles on the reaction kinetics of rigid PU foams formation, and to analyze how these modifications affect the cellular structure and physical properties of the foams. 2. Experimental 2.1. Reactants A commercial, bi-component formulation of rigid PU foam from "BASF Poliuretanos Iberia S.A." was used in this investigation. The polyol component, Elastopor H 1501/2 (OH index 651 mg KOH/g, density 1.07 g cm-3, viscosity 650 mPa·s) is a mixture of components containing polyether polyol, catalysts and stabilizers [30]. The formulation uses water as blowing agent. The isocyanate component, IsoPMDI 92140 (31.5% NCO, density 1.23 g cm-3, viscosity 170-250 mPa·s), is a polymeric diphenylmethane diisocyanate (pMDI) [30, 31]. According to supplier technical data sheet, the proportions of the two component formulation were set at 100/160 by weight for the polyol and isocyanate. Other features provided for BASF formulation are: free foaming density of 52±5 kg/m3, cream time (45±7), gel time (145±18) and rise time (232±21). 2.2. Nanofillers Reactants Table 1 collects the nanofillers used: two nanoclays (Cloisite®Na+ and Cloisite®30B) and three nanosilicas (Aerosil®A200, Aerosil®R812 and Aerosil®R974). The hydrophobic or hydrophilic character of these particles has been previously established in several studies referenced in Table 1, and has been confirmed in this research through the characterization of nanoparticles by FTIR spectroscopy and Thermogravimetric analysis (TGA). AEROSIL® fumed silica supplied by Evonik Industries AG (Evonik Degussa) are: (a) Aerosil®A200 (without modifier) with a specific surface area of 175-225 m2/g; (b) Aerosil® R812 (post-treated 125 Chapter 4 with hexamethyldisilazane) with a specific surface area of 230-290 m2/g; and (c) Aerosil® R974 (post-treated with dimethyldichlorosilane) with a specific surface area of 150-190 m2/g [32]. Nanoclays supplied by Southern Clay Products Inc are: Cloisite®Na+ with a specific surface area of 700-786 m2/g (CNa+, natural montmorillonite), and Cloisite®30B (organically modified montmorillonite with methyl tallow bis-2-hydroxyethyl quaternary ammonium chloride) with a specific surface area of 760 m2/g. Cloisite®30B contains both hydrophilic (hydrogen bonded water molecules in the silicate layer surfaces) and hydrophobic groups (hydrocarbon chains of either 18, 16 or 14 carbon atoms in the methyl-bis-2-hydroxyethyl-tallow quaternary ammonium cation) [33, 34]. Table 1. Additives used in this study describing the surface groups and their hydrophilic/hydrophobic nature. Additive Surface groups Type of surface Aerosil®A200 (A200) Silica nanoparticles without any further treatment [32, 35, 36] Hydrophilic Aerosil®R812 (R812) Silica nanoparticles post-treated with hexamethyldisilazane (90%) [32, 37] Hydrophobic Aerosil®R974 (R974) Silica nanoparticles post-treated with dimethyldichlorosilane (50%) [32, 35, 36] Partially hydrophilic/partially hydrophobic Cloisite®Na+ (CNa+) Natural Montmorillonite [33, 38, 39] Hydrophilic Cloisite®30B (C30B) Organically modified montmorillonite with methyl tallow bis-2-hydroxyethyl quaternary ammonium chloride (30%) [33, 38, 39] Partially hydrophilic/partially hydrophobic 2.3. Synthesis of the PU Foams Six different PU foam systems were studied: the pure material, and those containing either 5 wt% A200, 5 wt% R812, 5 wt% R974, 5 wt% CNa+ or 5 wt% C30B. “Pure” corresponds to the PU foam without additives that will be considered as the reference material. The nanoparticles (5% in weight) were previously dried under vacuum at 50°C overnight before being dispersed into the isocyanate component. An overhead stirrer (EUROSTAR Power control-visc P1, IKA) with a 50 mm diameter Lenart disc stirrer was used to premix the nanoparticles with the isocyanate component during 300 seconds for an optimal dispersion (at 1200 rpm during the first minute and the remaining time at 250 rpm). The polyol and isocyanate were mixed in a plastic cup with a ratio of 100:160 by weight at 1200 rpm for 10 seconds to promote the chemical reactions and the foaming process. Foams with cylindrical sizes of 100 mm x 200 mm (diameter x height) were produced. After the foaming and curing at room temperature for 2 days, samples were characterized. 126 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams 2.4. Nanofillers characterization Three methods were used for the characterization of the nanoparticles. Surface groups were identified by Fourier transform infrared (FTIR) spectroscopy, using a Bruker Tensor 27 spectrometer in the attenuated total reflectance (ATR) method. Thermogravimetric analysis (TGA) was performed with a Mettler Toledo TGA/SDTA 851 from 50 to 1000 °C at a heating rate of 20 °C/min, under inert atmosphere (N2). Morphological characterization was carried out using a Quanta 200 FEG environmental scanning electron microscope (ESEM). 2.5. Foam characterization 2.5.1. Density Foam density was measured as described by ASTM D1622/D1622M-14 [40]. Three different cylindrical samples of 30 mm x 25 mm (diameter x height) for each material were measured. 2.5.2. Open cell content The open cell content (OC) was measured by using a gas pycnometer Accupyc II 1340 from Micromeritics, according to ASTM D6226-10 [41]. The OC was measured in three cylindrical samples of each material after measuring their densities. 2.5.3. Scanning electron microscopy (SEM) The cellular structure of the foam and SEM micrographs were acquired by Scanning electron microscopy (SEM) with a JEOL JSM-820 microscope. The cured foams were cut in order to ensure a smooth surface, which was examined by SEM after vacuum coating with a gold monolayer. SEM micrographs were obtained for the growth plane (z plane) of the foam. An image analysis technique [42] was used to determine the main characteristics of the cellular structure: average cell size (Ф3D), anisotropy ratio (AR)) and the homogeneity of the cell size distribution measured using the parameter normalized standard deviation (NSD), which is the ratio of the standard deviation (SD) of the cell size distribution and the Ф3D. 127 Chapter 4 2.5.4. Thermal Conductivity measurements The thermal conductivity was determined at 20°C using a hot-disk transient plane source (TPS) thermal constant analyzer according to ISO 22007‐2:2008 method [43]. Transient methods are based on the analysis of the thermal response of a sample submitted to a controlled transient heat flow. The measurements were performed using two cylindrical samples of 30 mm x 25 mm (diameter x height) for each material. A disk shaped TPS sensor with a radius of 3.189 mm was used in all measurements, after being located in contact with the xy plane (perpendicular to growth plane) of the two samples. 2.5.5. Dynamical mechanical analysis (DMA) Dynamical mechanical analysis (DMA) was carried out using a PerkinElmer DMA7 dynamic mechanical analyzer using a heating rate of 3°C/min from 25 to 200°C. A parallel-plate system with a top plate of 12 mm in diameter was used on a cylindrical sample of 12 mm x 7 mm (diameter x height). The storage modulus (E’), loss modulus (E’’), and loss factor (tan δ) curves were obtained from DMA measurements. The glass transition temperature (Tg) of the polymer was calculated from the curve of tanδ vs. temperature. 2.5.6. Mechanical Tests Mechanical tests in compression were performed according to ASTM D1621-10 [44]. An Instron Machine (model 5.500R6025) was used for this purpose. Stress ()–strain () curves in compression were obtained at room temperature at a strain rate of 10 mm/min. The maximum static strain was 75% for all the experiments, and the compression tests were always performed in a direction parallel to the growing direction. The samples were prepared with rigorously parallel contact surfaces. Young´s modulus (E) and collapse stress (σ) were measured in three cylindrical samples of 30 mm x 25 mm (diameter x height) for each material. 2.6. Foaming kinetics: FTIR measurements In-situ FTIR spectra of the samples were collected using a Bruker ALPHA spectrometer working in ATR method. 1 mL of the reacting foam was poured on the surface of the ATR cell to obtain the variation of FTIR spectra vs time. In order to ensure the uniformity of the reacting foam, a higher amount of polyol (30 g) and isocyanate (48 g) (similar amounts to those used to produce the foams with the dimensions mentioned in section 2.2) were mixed at 1200 rpm for 10 seconds, and then 1 mL of this reacting foam was poured on the ATR surface. This method was designed in order to make uniform the low amount of reaction mixture used in FTIR experiments. The addition of the isocyanate was taken as time t = 0. The maximum elapsed time between the addition of isocyanate and the acquisition of the first scan was 128 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams approximately 1 minute. Each FTIR spectrum was obtained after 16 scans, with a resolution of 4 cm-1 in the range 4000-400 cm-1. In order to reproduce the high temperatures characteristics of the exothermic foaming process of PU, FTIR experiments were carried out at 70°C, using the plate available in the FTIR equipment. A background spectrum was deduced from each reaction spectra, and a total of 90 spectra were run for each experiment, which lasted 30 minutes. Baseline correction was conducted in order to correct the intensity shifts at lower frequencies. According to the literature [45, 46], the asymmetric CH stretching band at 2972 cm-1 (which remains constant during the reaction) was used as internal reference band to correct the concentration or density changes during the foaming process. The results herein reported are the average of three experiments. The maximum error reached for the isocyanate conversion was a 3% and a 5% for urea and urethane quantification. The overlapped absorptions in the amide I region (carbonyl region) were deconvoluted using Gaussian bands. Curve fittings were adjusted using methods previously reported.[47, 48] Hidden bands were identified following the second derivative technique, and correlation coefficient with at least three nines was obtained in the fittings. The deconvolution range was chosen to give appropriate adjustment within the range 1790-1570 cm-1. These limits were chosen in order to avoid the absorption effects of side bands from the aromatic C=C ring stretching and from amide II bands. In addition, the premix of the nanoparticles with the isocyanate component was characterized by FTIR spectra (blank experiments), using a Bruker Tensor 27 spectrometer in ATR method. 3. Results and discussion 3.1. Nanoparticles characterization Table 1 summarizes the nanoparticles used in this study, as well as the hydrophilic and/or hydrophobic nature of the surface groups present in them. These groups are critical on several aspects of the foams production, such as the dispersion of the fillers, their interaction with the PU matrix, and the possible nucleating effect during foaming. In addition, these surface groups can interact both with the functional groups of the reactants, and with the polymer molecules formed during the reaction. All these aspects may affect the reaction kinetics, so we decide to carry out a complete characterization of the selected additives in order to understand the phenomena occurring during the reactions. 3.1.1. FTIR spectroscopy FTIR spectra of nanosilicas are shown in Figure S1 (supporting information). The spectrum of A200 (unmodified nanosilica) shows an absorption peak ca. 3334 cm (O-H stretching vibrations), which may be assigned to the silanol groups on the nanosilica surface. The spectra 129 Chapter 4 Figure 6. Relative area percentage of the total absorbance associated to urea and urethane detected in the amide I region after a time of 30 minutes. Table 3 and Figure 6 clearly reflect how the presence of nanoparticles severely affects the overall reaction kinetics. A clear trend in the first five minutes is observed in Table 3, since the urethane/urea ratio decreases for all the systems, showing that more blowing reaction is occurring (more urea products) because the foam is expanding. Hence, the final urethane/urea ratio depends on the type of nanoparticle incorporated into the PU foam. As indicated in Table 3 the urethane/urea ratio of the system containing the A200 nanosilica is the lowest from 3 to 30 minutes. These data indicate that the blowing reaction, (responsible of both urea linkages formation and gas formation) is highly enhanced when A200 is added. The high imbalance between blowing and gelling reactions (38.4% of ureas and 61.6% urethanes, as shown in Figure 6) precludes the production of a foam with a low density and a homogeneous cellular structure. This will be discussed in the next section. The data show that A200 (the more hydrophilic nanosilica), and R974 (which contains both hydrophilic and hydrophobic groups on their surface), behave similarly: low values of the urethane/urea ratio and high proportion of ureas at 30 minutes, although proper foam is obtained in this case. For R974 the urethane/urea ratio is the second lower considering all the formulations used (see Table 3), thus the blowing reaction is again enhanced. There are previous reports on the somehow anomalous behavior of the nanosilica R974, what has been attributed to its dual behavior, which sometimes evinces a "split personality", due to the presence of both hydrophilic and hydrophobic groups in its surface [48]. In our experiments, R974 hydrophilic behavior seems to prevail. Instead, R812, which is the nanosilica containing only hydrophobic groups on the surface (90% methyl groups), gives rise to a slight increase of the urethane/urea ratio with respect to the 136 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams pure material (see Table 3). This occurs during the whole reaction process, and the final urethane/urea ratio for the PU containing R812 is the highest among those containing nanosilicas. Table 3 and Figure 6 also show the effects of adding nanoclays. The natural nanoclay CNa+, which contains hydrophilic groups, shows urethane/urea ratios similar to those of the pure foam (all the values between 3 and 30 minutes are similar, some of them higher, some lower, see Table 3). Moreover, the final content of urea groups for this foam is slightly higher than that of the reference system. However, the modified nanoclay C30B, which contains both hydrophilic and hydrophobic groups on their surface, induces higher values of the urethane/urea ratio during all the experiment (Table 3) and the highest increase of the urethane/urea ratio values after 30 minutes (Figure 6). The gelling reaction is greatly enhanced when this additive is present. As it is shown in Figure 6, this is the system more affected by the inclusion of nanoparticles. In summary, nanoparticles containing hydrophilic groups on their surface (A200 and CNa+) lead to a slight increase of the urea groups (and therefore of the blowing reaction) with respect to the pure foam. Moreover, R974 also exhibits a blowing reaction increase, due to its dual behavior with hydrophilic and hydrophobic groups on its surface. One of the possible reasons for the increase of blowing reaction could be the water present on these particles, which could be released during the foaming process. This water could be the adsorbed water (higher ratios on A200 and R974 nanosilicas surface), and the chemically bonded water (higher amount in the CNa+ nanoclays), as was observed in TGA analysis. However, establishing further relationships between the nature or surface of the nanoparticles added and their effect on the reaction is rather difficult. These data neither clarify whether a chemical interaction (weak, through hydrogen bonds, or strong, through covalent bonds) between the nanoparticles and the reactants (isocyanate, polyol, catalysts, surfactants and water) and/or the final product (PU), occurs or not. This is a very controversial point according to literature precedents, as some reports support a chemical reaction between the starting isocyanate and the OH groups present in nanoclays, [58-61] whereas other reports conclude that hydrogen bonds are the only type of interactions present,[62, 63] or that interactions do not occur at all [64-67]. In order to clarify this question, blank experiments were carried out by mixing the nanoparticles with isocyanate carrying out FTIR experiments. Figures 7 and S5-S8 (supporting information) show the isocyanate band and the Amide I region in the FTIR spectra of the isocyanate and of the starting premixture of isocyanate with nanoparticles. The spectra clearly indicate that the intensity of isocyanate absorption does not decrease, what indicates that there is no consumption of isocyanate reactant when the nanoparticles are dispersed into it. However, the intensity of the isocyanate band slightly increases, what might be due to weak hydrogen bonds interactions between the isocyanate groups and the hydroxyl groups present in the nanoparticles. The absence of reaction between the nanoparticles and the isocyanate is also shown by the lack of urethane absorptions observed in the Amide I region. Therefore, we can conclude that the hydroxyl groups present in nanoparticles do not react with the 137 Chapter 4 isocyanate groups of the isocyanate component in the conditions used in this experiment. This also means that the water included in the nanoparticles neither reacts with the isocyanate groups, very possibly because a higher temperatures would be needed in order to release the water from the nanoparticles surface. Figure 7. Spectra of isocyanate component and of the dispersion of 5% A200 in isocyanate: (a) isocyanate band and (b) Amide I region. Clearly, the modifications induced in the kinetics of the reactions by the nanoparticles will play a role on the final structure and properties of the foams. This is the main focus of the next sections. 3.3. Cellular structure and properties of final foams The density and the main characteristics of the cellular structure of the foams are collected in Table 4. The density may be related to the amount of gas generated during the foaming process and with the stability of the foam. The higher density value is reached for system containing A200. This foam, as indicated above, collapsed due to the strong imbalance between the two main reactions (i.e. part of the gas was released too soon, when the viscosity of the reactive mixture was not enough to hold the gas phase inside). The other two materials based on partially hydrophilic particles, R974 and CNa+, show a lower density than that of the pure foam. This interesting effect can be explained taking into account that the blowing reaction is favored for these systems. This means that more gas is available when the viscosity is not too high in order to promote a higher expansion ratio. Finally, the foams containing R812 and C30B nanoparticles, where the urethane/urea ratio is higher (more gelling reaction) than that of the pure foam (see Table 4), have a higher density than that of the pure system. This result can be understood taking into account that in this case the gelling reaction was quicker, 138 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams and the viscosity of the system was also higher during foaming, giving as a consequence a lower expansion ratio. On the other hand, there is an evident increase of the open cell content (OC) (which range from ca. 8 to 23%) with the addition of nanoparticles, as compared to that found in the pure foam (which is only ca. 5%). Clearly, nanosilicas give rise to higher OC compared to those of nanoclays, as shown in Table 4. Moreover, nanosilicas containing more OH groups on their surface give higher values of OC, and the same trend is observed for nanoclays. This significant increase of the open cell content should be connected to additional phenomena taking place during foaming, because it occurs in a higher or lower extent for all the nanoparticles considered. One possible reason is the high amount of particles used (5 wt%), which leads to possible agglomerations, what could promote opening of the cell walls, as previously proved for polypropylene foams containing clays [68]. This effect would be more important in those systems based on hydrophilic particles, in which a worse dispersion is expected. The analysis of the SEM micrographs (Figure 8) in the growth plane allows measuring cell size (Ф3D), anisotropy ratio (AR), and the homogeneity of the cell size distribution (NSD) (Table 4). Cell size is highly reduced up to 34% by nanoclays, and up to 22% by nanosilicas, compared to the value of the pure foam. As previously reported, the cell size reduction for similar systems to the those herein studied is due to a nucleation effect of the nanofillers [69]. Thus, this cell size reduction is due to a physical phenomenon not directly connected to the different reaction kinetics caused by the presence of the different particles. From this perspective, nanoclays seem to be more effective than nanosilicas, possibly because the specific surface area of nanoclays is higher than those of nanosilicas. All the foams (pure and those containing nanoclays) can be considered isotropic, as their anisotropy is close to one in all the cases. However, there are several differences between samples, due to the inaccuracy of the measurements. Moreover, all the samples display a high homogeneity, as their NSD values (related to the width of the cell size distributions) are small. All the particles increase the value of NSD, what is a typical result for systems in which the particles act as nucleating sites. This effect is associated to the presence of a combination of cells (typically smaller), nucleated in the particles and cells nucleated in a similar way to those nucleated in the reference material [69]. 139 Chapter 4 Table 4. Density, OC, Ф3D, AR and NSD and AC for the foams under study. Sample A200 was not characterized due to its very deteriorated cellular structure. Material Density (Kg/m3) OC (%) Ф3D (µm) AR NSD Pure 57.15±0.31 5.00±0.53 619±118 1.10±0.23 0.19 5%A200 67.52±3.15 - - - - 5%R812 57.43±0.65 15.22±0.78 452±152 1.34±0.37 0.34 5%R974 55.92±0.65 22.67±1.32 484±170 1.23±0.38 0.35 5%CNa+ 51.00±0.15 11.56±0.18 408±100 1.10±0.18 0.25 5%C30B 59.03±1.55 8.13±0.83 409±169 1.06±0.21 0.41 Figure 8. SEM micrographs of PU foam: (a) Pure material, (b) with 5% wt.A200, (c) with 5% wt.R812, (d) with 5% wt.R974, (e) with 5% wt. CNa+ and (f) with 5% C30B. 3.4. Thermal conductivity Figure 9 shows the thermal conductivity of the pure foam and of those containing nanofillers (except for the A200 material; as its poor quality precluded to characterize the physical properties of this material). The values have been measured when the cells contain only atmospheric air, that is, once the carbon dioxide produced in the blowing reaction has completely diffused outside the cells (this occurs ca. 2 weeks after production), being substituted by atmospheric air [11]. The conductivity of the two systems containing nanoclays is lower than that of the reference system. This result is due to a combination of facts. As previously reported [11], a cell size reduction in these systems promotes the reduction of the heat transfer by radiation. In addition to this, the open cell content of these two systems does not increase significantly in comparison with that of the pure foam, so the expected increase of thermal conductivity due to the higher open cell content does not compensate the effect of the reduced cell size. Finally, 140 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams for the system containing the CNa+ clay, the density reduction due to the different kinetics of this material (see previous section) also contributes to the reduction of the thermal conductivity. The two systems containing nanosilica particles present similar (R812) or higher (R974) thermal conductivity than of the reference material. This is probably due to the high open cell content of these two materials, which induces a significant increase of the heat transfer by radiation in spite of their reduced cell size. Considering these two foams the one with a smaller thermal conductivity (R812) is showing a lower open cell content. Figure 9. Thermal conductivity values for PU nanocomposite foams. 3.5. Dynamic mechanical analysis (DMA) Table 5 summarizes the viscoelastic properties of the PU foams measured by DMA. The first column in Table 5 collects the glass transition temperatures (Tg), which are associated with the α relaxation of soft domains in the foams. Both Tg and the maximum of tan δ account for the polymer chain mobility. The foam with the nanosilica R812 (hydrophobic) displays the highest Tg, whereas the hydrophilic nanoclay CNa+ affords the lowest Tg value. These values are in accordance with the isocyanate conversion (Figure 4), and show how the formation of more products (higher isocyanate conversion) gives rise to higher Tg values, and therefore to restrict the chain mobility. Obviously, lower products formation (low isocyanate conversion) leads to lower Tg, and to a higher chain mobility. Therefore, these results show that the reaction kinetics is directly connected to the chain mobility and to the viscoelastic behavior of the foams. 141 Chapter 4 Table 5. The DMA data for pure foam and the foams containing nanoparticles. Material Tg (°C) Maximun Tan δ Pure 149.9±1.7 0.549±0.031 5%R812 157.5±2.6 0.292±0.023 5%R974 161.2±2.6 0.333±0.007 5%CNa+ 118.6±2.7 0.491±0.031 5%C30B 149.1±2.2 0.495±0.022 3.6. Mechanical properties The effect of the functional nanofillers on the mechanical properties of the PU foams obtained was also studied. The mechanical properties of the PU foams mainly depend on the foam density, the cellular morphology (cell size, OC, cell wall thickness, etc), the polymeric matrix morphology (reaction products generated, etc), and the compatibility of nanoparticles with the matrix. As the foams studied have slightly different densities. The Gibson and Ashby relationship between mechanical properties and density [70] is used to exclude these differences in densities. Figure 10 shows the compressive modulus and collapse stress divided by the relative density of the foam. These values allow to compare the mechanical properties of the PU foams containing different nanofillers, and therefore with different densities. Figure 10. Relative compressive modulus (a) and relative collapse stress (b) for pure foam and foams with particles correspond to compression test. As shown in Figure 10, the foams containing nanosilicas show a significant decrease of the compression properties, whereas this decrease for the foams containing nanoclays is smaller. Compared to the pure PU foam, those containing nanoclays present a decrease of 12% (for CNa+) and of 14% (for C30B) in compressive modulus, and a decrease of 21% and 13% (respectively) in their collapse stress. The higher decrease for foams containing nanosilicas 142 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams reaches a 64% for R812 and a 46% for R974 in their compressive modulus, whereas the reduction of their collapse stress is 61% and 53% (respectively). The reduction of the final mechanical properties for the four systems compared to the reference foam is mainly due to the increase of the open cell content, due to the presence of the particles. In fact, the materials containing nanosilica with a higher open cell content present reduced compressive properties. In addition to this, the better properties of the nanosilica systems are displayed by R794, which contains higher amounts of urea in the final foam (see Figure 6). The increase of the mechanical properties due to the increase of urea content has been already described [71]. 4. Conclusions The effects of nanofillers (nanosilicas and nanoclays) on the reaction kinetics, microstructure, thermal conductivity, viscoelastic properties and compressive mechanical properties of rigid PU foams are herein systematically studied. The reaction kinetics are studied by in-situ FTIR spectroscopy. The isocyanate conversion depends on the type of particles used in PU formulation, and on their superficial groups. Therefore, the foams containing nanosilicas show higher isocyanate conversions than those containing nanoclays, whereas those nanoparticles containing hydrophilic groups afford lower isocyanate conversions. The quantification of urea and urethane groups by deconvolution of the carbonyl region can be also correlated to the chemical groups present on the surface of the nanoparticles added, what allows to conclude that hydrophilic nanoparticles enhance the blowing reaction, giving higher ratios of urea groups. The modifications of the reaction kinetics induced by the nanoparticles have consequences on the density, cellular structure and physical properties of the final foams. Density is reduced for systems in which the blowing reaction is enhanced (foams with R974 and CNa+). However, when the blowing reaction is accelerated in a significant extent the strong decoupling between blowing and gelling promotes a deterioration of the cellular structure, as observed for the A200 system. The conversion degree clearly affects the viscoelastic properties of the foams, since a higher conversion degree promotes both a higher glass transition temperature and a lower degree of mobility of the polymer chains. In addition, the presence of a higher urea ratio in the final foams allows improving the compressive properties. In addition to these effects, the particles also modify the foaming mechanisms. In this particular case, all the analyzed nanoparticles act as nucleating agents during the process, reducing the average cell size and decreasing the homogeneity of the cell size distribution. Moreover, the high content of particles used induces an increase of the open cell content. These modifications of the cellular structure lead to a reduction of the thermal conductivity for the systems containing clays, and to a decrease of the compressive properties for all the analyzed foams. 143 Chapter 4 Acknowleggements Financial assistance from MINECO, FEDER, UE (MAT2015-69234-R) and the Junta de Castile and Leon (VA011U16) are gratefully acknowledged. References [1] U. Meier-Westhues, Polyurethanes: Coatings, Adhesives and Sealants., Vincentz Network, Hannover, 2007. [2] M. Szycher, Szycher's Handbook of Polyurethanes, Second ed., CRC Press Boca Raton, Florida, USA, 2012. [3] D. Klempner, V. Sendijarevic, Handbook of Polymeric Foams and Foam Technology, Hanser 1991. [4] Polymer Foams Market Expected to Consume 25.3 Million Tonnes by 2019. [5] L. Glicksman, Low Density Cellular Plastics: Physical Basis of Behaviour, Springer Netherlands, London, 1994. [6] R.J.J. Williams, C. Aldao, Thermal Conductivity of Plastic Foams, Polymer Engineering and Science, 23 (1983) 293–298. [7] O.A. Almanza, M.A. Rodríguez-Pérez, J.A.D. Saja, Prediction of the radiation term in the thermal conductivity of crosslinked closed cell polyolefin foams, Journal of Polymer Science Part B: Polymer Physics, 38 (2000) 993–1004. [8] J.W. Wu, W.F. Sung, H.S. Chu, Thermal conductivity of polyurethane foams, International Journal of Heat and Mass Transfer, 31 (1999) 1100–1106. [9] M.S. Han, S.J. Choi, J.M. Kim, Y.H. Kim, W.N. Kim, H.S. Lee, J.Y. Sung, Effects of Silicone Surfactant on the Cell Size and Thermal Conductivity of Rigid, Macromolecular Research, 17 (2009) 44–50. [10] C.-j. Tseng, M. Vamaguch, T. Ohmori, Thermal conductivity of polyurethane foams from room temperature to 20 K, Cryogenics, 37 (1997) 305–312. [11] S. Estravís, J. Tirado-Mediavilla, M. Santiago-Calvo, J.L. Ruiz-Herrero, F. Villafañe, M.A. Rodríguez-Pérez, Rigid polyurethane foams with infused nanoclays: Relationship between cellular structure and thermal conductivity, European Polymer Journal, 80 (2016) 1–15. [12] E.P. Giannelis, Polymer Layered Silicate Nanocomposites, Advanced Materials, 8 (1996) 29–35. [13] Y. Chen, S. Zhou, H. Yang, L. Wu, Structure and properties of polyurethane/nanosilica composites, Journal of Applied Polymer Science, 95 (2005) 1032–1039. [14] S. Estravis, PhD Thesis: Cellular nonocomposites based on rigid polyurethane and nanoclays: fabrication, caharacterization and modelling of the mechanical and thermal properties, University of Valladolid, 2014. [15] S. Pardo-Alonso, E. Solórzano, L. Brabant, P. Vanderniepen, M. Dierick, L. Van Hoorebeke, M.A. Rodríguez-Pérez, 3D Analysis of the progressive modification of the cellular architecture in polyurethane nanocomposite foams via X-ray microtomography, European Polymer Journal, 49 (2013) 999–1006. [16] A.J. Crosby, J.Y. Lee, Polymer Nanocomposites: The “Nano” Effect on Mechanical Properties, Polymer Reviews, 47 (2007) 217–229. [17] M.A. Samir, F. Alloin, A. Dufresne, Review of Recent Research into Cellulosic Whiskers, Their Properties and Their Application in Nanocomposite Field, Biomacromolecules, 6 (2005) 612–626. 144 Development of a methodology to follow the reaction kinetics of rigid polyurethane foams [18] Y. Rao, J.M. Pochan, Mechanics of Polymer-Clay Nanocomposites, Macromolecules 40 (2007) 290–296. [19] R. Van Hooghten, S. Gyssels, S. Estravis, M.A. Rodriguez-Perez, P. Moldenaers, Understanding the effect of particle surface free energy on the structural and mechanical properties of clay-laden rigid polyurethane foams, European Polymer Journal, 60 (2014) 135– 144. [20] L. Madaleno, R. Pyrz, A. Crosky, L.R. Jensen, J.C.M. Rauhe, V. Dolomanova, A.M.M.V. de Barros Timmons, J.J. Cruz Pinto, J. Norman, Processing and characterization of polyurethane nanocomposite foam reinforced with montmorillonite–carbon nanotube hybrids, Composites Part A: Applied Science and Manufacturing, 44 (2013) 1–7. [21] X. Cao, L. James Lee, T. Widya, C. Macosko, Polyurethane/clay nanocomposites foams: processing, structure and properties, Polymer, 46 (2005) 775–783. [22] G. Harikrishnan, T.U. Patro, D.V. Khakhar, Polyurethane Foam-Clay Nanocomposites: Nanoclays as Cell Openers, Industrial and Engineering Chemistry Research, 45 (2006) 7126– 7134. [23] I.V. Khudyakov, D.R. Zopf, N.J. Turro, Polyurethane Nanocomposites, Designed Monomers & Polymers, 12 (2009) 279–290. [24] M.M.A. Nikje, Z.M. Tehrani, Thermal and mechanical properties of polyurethane rigid foam/modified nanosilica composite, Polymer Engineering & Science, 50 (2010) 468–473. [25] M.M.A. Nikje, Z.M. Tehrani, Polyurethane Rigid Foams Reinforced by Doubly Modified Nanosilica, Journal of Cellular Plastics, 46 (2010) 159–172. [26] D. Qi, Y. Bao, Z. Weng, Z. Huang, Preparation of acrylate polymer/silica nanocomposite particles with high silica encapsulation efficiency via miniemulsion polymerization, Polymer, 47 (2006) 4622–4629. [27] O. Aso, J.I. Eguiazábal, J. Nazábal, The influence of surface modification on the structure and properties of a nanosilica filled thermoplastic elastomer, Composites Science and Technology, 67 (2007) 2854–2863. [28] S.T. Lee, N.S. Ramesh, Polymeric foams: mechanisms and materials, Boca Raton, Florida, USA, 2004. [29] G. Woods, Flexible Polyurethane Foams: Chemistry and Technology, London, 1982. [30] Elastopor H 1501/2 Technical Data Sheet. BASF Poliuretanos Iberia S.A. [31] IsoPMDI 92140 Technical Data Sheet. BASF Poliuretanos Iberia S.A. [32] Technical data sheet: Products AEROSIL® fumed silica from Evonik Industries AG (Evonik Degussa). www.aerosil.com/product/aerosil/en/products. [33] Technical data sheet: Cloisite nanoclays from Southern Clay Products. [34] N.K. Borse, M.R. Kamal, Melt processing effects on the structure and mechanical properties of PA-6/clay nanocomposites, Polymer Engineering & Science, 46 (2006) 1094– 1103. [35] H.J. Walls, J. Zhou, J.A. Yerian, P.S. Fedkiw, S.A. Khan, M.K. Stowe, G.L. Baker, Fumed silicabased composite polymer electrolytes: synthesis, rheology, and electrochemistry, Journal of Power Sources, 89 (2000) 156–162. [36] S.R. Raghavan, J. Hou, G.L. Baker, S.A. Khan, Colloidal Interactions between Particles with Tethered Nonpolar Chains Dispersed in Polar Media: Direct Correlation between Dynamic Rheology and Interaction Parameters, Langmuir 16 (2000) 1066–1077. [37] A. Chuiko, Y. Gorlov, Chemistry of a Silica Surface, Kiev, 1992. [38] J.M. Cervantes-Uc, J.V. Cauich-Rodríguez, H. Vázquez-Torres, L.F. Garfias-Mesías, D.R. Paul, Thermal degradation of commercially available organoclays studied by TGA–FTIR, Thermochimica Acta, 457 (2007) 92–102. 145 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties 36% and 32%, respectively. However the thermal conductivity was not improved. Bernal et al. [13] produced RPU nanocomposite foams (with density ca. 80 Kg/m3) containing 0.17 and 0.35 wt% of functionalized graphene sheets (FGS) to obtain electromagnetic interference (EMI) shielding materials, which showed a clear reduction of the cell size by 22%. Lorenzetti et al. [14] studied the effect of 0.3 and 0.5 wt% of reduced GO (1-2 layers) on the thermal conductivity of RPU foams of density ca. 35 kg/m3. The foam containing 0.3 wt% of graphene presented cell size reduction, although an increase of the graphene content up to 0.5 wt% did not decrease the cell size. The RPU-graphene nanocomposite foams showed a reduction of the thermal conductivity and a slower aging rate with respect to the pure material. In addition, the extinction coefficient was increased due to role played by the particles as IR- blockers. In all these papers graphene or GO was added as a nanoparticle to the polyol blend and was dispersed using mechanical stirring or ultrasonication. This is a simple approach that has a main drawback: it is necessary to disperse properly the nanoparticles in the polyol. In addition, it is well known that dispersing nanoparticles in most of the cases is not a simple task, since it requires specific methods (such as high shear mixing, ultrasounds, or functionalization of the particles, for example) to reach a certain degree of dispersion. The approach followed herein is completely different and it avoids the dispersion problem previously mentioned. The polyols used have been polymerized in situ in the presence of GO, what allows attaching the particles to the polymer chains. This is a relatively new approach and in fact we are aware of only one previous report on polyols functionalized with graphene which are used to form RPU foams [15]. This patent describes a chemical bond method in order to prepare polyether polyol modified with reduced GO. Different RPU foams were prepared by changing the ratio between pure polyol and graphene modified polyol. The flame retardant and mechanical properties were evaluated in the materials obtained. The results of this patent show that all RPU foams reached V-0 in the UL94 test, whereas the smoke release rate of RPU foams containing graphene was significantly reduced with respect to the reference material. In addition, the tensile strength of the RPU foams was significantly improved. In spite of these promising results, this work do not report in detail how density, cellular structure and thermal conductivity and the PU reaction kinetics are affected by the presence of graphene in the polyol. Taking the previous ideas in mind, herein we present a detailed study on RPU foams produced from a polyol functionalized with GO. Low amounts of GO (0.017, 0.033 and 0.083 wt%) are used in order to maintain a viscosity of the polyol allowing the production of RPU foams. The influence of the amount of GO in the polyol on density, cellular structure, thermal conductivity and mechanical properties are herein described. Moreover, a kinetic study is carried out by FTIR spectroscopy, infrared expandometry and temperature evolution, in order to evaluate the effects of GO on the foam formation process. 248 Chapter 6 2. Experimental 2.1. Materials IsoPMDI 92140 (31.5% NCO, density 1.23 g cm-3, viscosity 170-250 mPas), a polymeric diphenylmethane diisocyanate (pMDI) from "BASF Poliuretanos Iberia S.A”, was used to produce the rigid polyurethane foams. TEGOAMIN® DMCHA (N,N-dimethylcyclohexylamine) from Evonik, was used as gelling and foaming catalyst. TEGOSTAB® B 8522 (a non-hydrolysable poly-ether-polydimethyl-siloxane–stabilizer) from Evonik, was used as a surfactant. Distilled water was used as a blowing agent. 2.2. Synthesis of polyols functionalized with GO Graphene Oxide (Graphenea S.A.) was dispersed in the triol initiator using UltraTurrax for 15 minutes at room temperature. The mixture was then added to the polymerization reactor (2 L, 600 rpm) and heated to 50 °C in presence of Propylene Oxide (Repsol) at 1 Bar of pressure and 2000 ppm of acid catalyst, according to Appel et al. [16]. Polymerization was stopped when a constant pressure was reached. Four poly(propylene oxide) polyols were synthesized: the pure polyol, and those containing 500, 1000 and 2500 ppm of GO. 2.3. Characterization of polyol functionalized with GO The hydroxyl number in mg of KOH per grams of polyol and polyols functionalized with GO was calculated according to ASTM D-4274. Brookfield viscosity was determined at 25°C using a Brookfield DV-III ULTRA Rheometer. Glass transition temperature (Tg) was taken from the second heating in differential scanning calorimeter (DSC) experiments. Non isothermal (10°C/min from -85 to 200°C) experiments were carried out using a DSC TA Instruments Q2000 under nitrogen flow, operating with an intra-cooler under nitrogen flow. Temperature and heat flow calibrations were performed with indium as standard. Weigh-averaged molecular weights (MW) and polydispersity index (PDI) were determined against PEG standards by gel-permeation chromatography (GPC) using a Bruker 3800 equipped with a deflection RI detector. Tetrahydrofuran at 1 mL/min flow rate was used as eluent at 30°C. 2.4. Preparation of rigid polyurethane/graphene oxide nanocomposite foams Four different RPU foam systems were studied: the pure material (without GO, obtained as reference material), and those containing 0.017 wt% GO (500 ppm in polyol), 0.033 wt% GO 249 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties (1000 ppm in polyol) and 0.083 wt% GO (2500 ppm in polyol) with respect to the final mass of RPU foams. An overhead stirrer (EUROSTAR Power control-visc P1, IKA) with a 50 mm diameter Lenart disc stirrer was used to premix the polyol (100 parts by weight (ppw)) with catalyst (1 ppw), surfactant (1 ppw), and blowing agent (5 ppw) during 2 minutes at 250 rpm to obtain homogenous polyol blends (Part B). Part B and the isocyanate (Part A) at an isocyanate index of 120 were mixed in a plastic cup at 1200 rpm for 10 seconds. After the foaming, the curing process was extended for 2 days at room temperature, and then the foam samples were cut for characterization. 2.5. Foams characterization Foam density was measured as described by ASTM D1622/D1622M-14 [17]. Density was determined in three different samples for each material, with a diameter of 30 mm and a height of 25 mm. After measuring the densities in the samples, open cells content (OC%) was measured by using a gas pycnometer Accupyc II 1340 from Micromeritics, according to ASTM D6226-10 [18]. The cellular morphology of the foams was observed by Scanning Electron Microscopy (SEM) with a JEOL JSM-820 microscope. The growth plane (z plane) of cured foams was examined by SEM after vacuum coating with a gold monolayer. An image analysis technique [19] of SEM micrographs was used to determine the main characteristics of the cellular foam structure: mean cell size (Ф3D), anisotropy ratio (AR) and cell size distribution. The calculated statistical parameters for cell size distribution were: standard deviation (SD), normalized standard deviation (NSD) (i.e. SD/ Ф3D), and asymmetry coefficient (AC). NSD provides information about the homogeneity of the cell size distribution: homogeneous cell distributions present small values of this parameter. AC measures the degree of asymmetry of the distribution, so as a negative AC indicates that the smaller cells are more separated from the mean cell size than bigger ones, and vice versa. More than 150 cells of different areas of each material were used for this analysis. The detailed calculation of the cellular structure parameters are explained in the work by Pinto et al. [19]. FTIR spectra of the cured foams were collected using a Bruker ALPHA spectrometer by attenuated total reflectance (ATR) method. Polymer morphology and polymer composition of RPU foam was analyzed using deconvolution of amide I region (carbonyl region) of FTIR spectrum. The deconvolution was performed using the same method described in the following section dealing with the kinetic studies by FTIR spectroscopy. 250 Chapter 6 Thermal conductivity was determined at room temperature by using a hot-disk transient plane source (TPS) thermal constant analyzer, according to ISO 22007‐2:2008 method [20]. The measurements were performed using two cylindrical samples for each material with a diameter of 30 mm and a height of 25 mm. A disk shaped TPS sensor with a radius of 3.189 mm was used for all measurements, after being located in contact with the xy plane (perpendicular to growth plane) of the two samples [21]. Several measurements were performed until 40 days after the foams were produced. At that time the samples have reached the stationary state, considering that all the CO2 generated during foaming have diffused outwards the foam by then, and it has been replaced by atmospheric air [5]. Mechanical properties in compression were measured at room temperature by using an Instron Machine (model 5.500R6025), according to ASTM D1621-10 [22]. Stress ()–strain () curves were obtained at a strain rate of 10 mm/min and the maximum static strain was 75% for all the experiments. The compression tests were performed in a direction parallel to the growing direction for three different samples of each material, with a diameter of 30 mm and a height of 25 mm. The samples were prepared with rigorously parallel contact surfaces. Young´s modulus (E) and collapse stress (σC) were calculated from the stress-strain curves. 2.6. Kinetics studies Reaction kinetics for the reference material and for the material containing 0.033 wt% GO (1000 ppm in polyol) were studied by in situ FTIR, temperature evolution and IR expandometry. In situ FTIR spectra of the samples were collected using a Bruker ALPHA spectrometer by attenuated total reflectance (ATR) method. 1 mL of the reacting foam was poured on an ATR cell to obtain the variation of FTIR spectra versus time. The addition of the isocyanate was taken as t = 0 for the reaction, and the maximum elapsed time between the addition of isocyanate and the acquisition of the first scan was 40 seconds. Each FTIR spectrum was obtained after 16 scans, with a resolution of 4 cm-1 for the range 4000-400 cm-1. In order to reproduce the large scale foam temperature profile, the experiment was carried out at 70°C, as the spectrometer contains a heating device. A background spectrum was deduced from each reaction spectra, and a total of 60 spectra were taken for each experiment, which lasted 30 minutes. Baseline correction was conducted in order to correct the shifts from temperature changes in each spectrum. According to the literature,[23, 24] the asymmetric CH stretching band at 2972 cm-1 (which remains constant during the reaction) was used as internal reference band. The results herein reported are the average of three kinetics experiments. The overlapped absorptions in the amide I region (carbonyl region) (1601-1760 cm-1) were deconvoluted using Gaussian bands in order to follow separately the blowing and gelling reactions occurring during the foaming process. Curve fittings were adjusted by literature methods [25, 26] and the hidden bands were identified following the second derivative technique. The deconvolution range (1790-1570 cm-1) allows an admissible adjustment and 251 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties avoids the absorption effects of side bands from the aromatic C=C ring stretching and from amide II bands. The temperature profiles during foaming process were measured. Four thermocouples type K were introduced in a plastic cup of 11.5 cm of diameter and 14 cm of height in order to measure the temperature during 30 minutes at four different positions. As seen in Figure 1, thermocouples were placed vertically in the centre of plastic cup at the following heights from the base of the cylinder (Figure 1): 0.5 cm (thermocouple 1, T1), 2.0 cm (thermocouple 2, T2), 6.5 cm (thermocouple 3, T3) and 12.5 cm (thermocouple 4, T4). The data collected by the thermocouples were registered in a computer. 12.5 cm 0.5 cm 2.0 cm 6.5 cm Thermocouples T1 T2 T3 T4 Figure 1. Schematic view of the system of thermocouples inside the plastic cup. IR expandometry was used to measure expansion height and volume, expansion rate, expansion acceleration and maximum surface temperature reached. In this technique the foaming process is followed by using an infrared camera that records the variations of the surface temperature during foaming. The IR camera used in this paper was HotFind L from SDS Infrared. 25 images are taken every 1 second. Once all the images are taken and by using image analysis it is possible to measure the aforementioned characteristics [27]. 3. Results and Discussion 3.1. Polyol characterization Typically, poly(propylene oxide) polyols are produced by reacting a triol initiator with propylene oxide (PO) as shown in Scheme 1 for a 3-arm polyol. The presence of GO in the reaction mixture may favour polyol linear chain grafting, since the hydroxyl groups of GO could 252 Chapter 6 react with PO [16]. Therefore, polyether polyols grafting onto the GO surface are obtained. Due to the incorporation of the GO in the polyol during the synthesis it is possible to obtain an excellent dispersion of the particles solving the always difficult challenge of dispersing the particles. Scheme 1. Synthesis of polyether polyol. The properties of the resulting polyether polyols are summarized in Table 1. The first column shows the decrease of the the hydroxyl number with the inclusion of GO, being the polyol with 1000 ppm GO that showing a lower value. The decrease of the hydroxyl number for polyols functionalized with GO could be due the reaction of some hydroxyl groups of the polyether polyol components with GO during the synthesis of the polyether polyols. In addition, the viscosity slightly increases for the polyols containing 500 ppm and 1000 ppm GO, and it is obviously higher for the polyol containing 2500 ppm GO. In general, the viscosity is related to molecular weight. As can be appreciated on Table 1, the viscosity raises with the increase of the weigh-averaged molecular weight (Mw) of the polyol. The Mw value grows with the amount of GO, indicating that the GO promotes the polyol linear chain grafting, as has been mentioned previously [16]. Furthermore, the distribution of the molecular weights in the polyol is determined by the polydispersity index (PDI), which is almost constant in all the polyols prepared. On the other hand, Table 1 collects the glass transition temperatures (Tg), which slightly decreases when GO is added. The polyol with a higher amount of GO presents the smallest value of Tg, being this polyol that showing the higher polydispersity index. 253 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties Table 1. Properties of polyol functionalized with GO compared with the pure material. Polyol Hydroxyl number (mg KOH/g) Viscosity (mPa·s at 25°C) Tg (°C) Mw (Dalton) PDI Pure 426.4 476.9 -59.6 488 1.2 500 ppm GO 399.3 553.3 -60.4 494 1.1 1000 ppm GO 358.6 547.6 -61.0 570 1.1 2500 ppm GO 377.5 4463.2 -62.8 675 1.3 3.2. Density and Cellular structure characterization Different physical properties of the foams were determined, and the values obtained are collected in Table 2. The density of the foams containing 0.017 and 0.033 wt% GO increases 1 kg/m3 compared to that of the pure material, whereas the density of the foam containing 0.083 wt% GO increases around 3-4 kg/m3. This could be related to the significant increase of the polyol viscosity when a high GO content is incorporated (Table 1). The open cell content (OC) for the foam with 0.017 wt% GO is similar to that obtained for the reference foam. However, there is an evident increase of the OC with the GO addition, being the foam with the highest GO content (0.083 wt%) that showing the highest value (ca. 42%). The SEM micrographs in the growth plane (Figure 2) allow obtaining the main characteristics of the cellular structure (Table 2). It is remarkable that the incorporation of small amounts of GO into polyol promotes an important cell size reduction of up to 33% for the material containing 0.083 wt% GO, demonstrating its effectiveness as cell nucleating agent. This reduction is clearly observed in the SEM micrographs shown in Figure 2 b-d. In addition to the clear modification of cell size, samples with low GO contents (equal or below 0.033 wt%) present similar anisotropy ratios (AR) than the reference foam. However, this anisotropy ratio is reduced for the sample with higher GO content (0.083 wt% GO). This is probably due to some cells degeneration occurring in these samples which also promotes a large open cell content, higher NSD and AC and a higher density. The histograms and gaussian fitting depicted in Figure 3, as well as other quantitative information collected in Table 2, give extra information about the cell size distribution. Samples up to 0.033 wt% GO show a good homogeneity of the cell size distribution (small values of NSD) and a symetric distribution, with values of AC near to zero. However, the cell size distribution of the foam with 0.083 wt% GO is less homogeneous, considering its high value of NSD (0.55), and its highly asymmetric distribution, with a high and positive value of AC (1.33). As previosly mentioned, this is probably due to the degenation of the cellular structure due to cells coalescence. 254 Chapter 6 Table 2. Density, open cell content (OC), mean cell size (Ф3D), standard deviation (SD), normalized standard deviation (NSD), asymmetry coefficient (AC) and anisotropy (AR) for each foam obtained. Material Density (Kg/m3) OC (%) Ф3D (µm) SD NSD AC AR Pure (reference) 30.1±0.8 11.03±1.01 529 178 0.34 0.30 1.46±0.47 0.017 wt% GO 31.1±0.9 10.45±0.04 489 139 0.28 -0.32 1.54±0.38 0.033 wt% GO 31.7±0.7 14.05±1.93 427 140 0.33 0.15 1.57±0.36 0.083 wt% GO 34.0±2.0 41.58±10.79 355 177 0.55 1.33 1.25±1.31 (a) Pure (b) 0.017 wt% GO (c) 0.033 wt% GO (d) 0.083 wt% GO Figure 2. SEM micrographs for the growth plane of the foam: (a) Pure material (reference), (b) with 0.017 wt% GO, (c) with 0.033 wt% GO and (d) 0.083 wt% GO. 255 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties Figure 3. Cell size distributions and gaussian fits for each distribution. 3.3. Polymer morphology characterization RPU foams present a complex polymer morphology composed of alternant soft and hard segments. Hard segments derive from the reactions of isocyanate with water and/or hydroxyl groups give rise to foam stiffness, while soft segments from polyols provide foam elasticity. The polymerization reaction creates progressively a chemical crosslinking network of RPU foams, while blowing reaction generates polyurea segments whose different polarity and chemical incompatibility with polyol segments tends to aggregate via hydrogen bonding, thus originating microphase separation domains [28]. The polyurea hard segments in foams have stronger specific hydrogen-bonding interaction and high stiffness compared to the urethane hard segments found in most elastomers. For this reason the hydrogen-bonded urea, in particular ordered urea, is frequently used for estimating the degree of microphase separation [29, 30], and has a significant influence on the mechanical properties of RPU foams. Therefore, we decided to study the composition of the PU matrix by the deconvolution of the carbonyl region of the cured RPU foams [30]. The carbonyl region comprises the different urea and urethane groups formed during the foaming process. The data collected in Table 3 shows how the presence of GO alters the main PU reactions, and create changes in the hydrogen bonding 256 Chapter 6 network of PU nanocomposites. Table 3 shows that cured foams containing GO present a total percentage of urea slightly higher than that of the reference foam. This indicates that the blowing reaction has been favoured, being the isocyanate conversion around 90% for all the foams studied. A more detailed examination of the different types of hydrogen bonding may be inferred from these data. In general the percentage of ordered urea is slightly suppressed by the increase of GO, what indicates a lower degree of microphase separation. However, the development of the free urea in foams with GO increases and free urea is dissolved in polyolrich soft domains, while the amount of disordered hydrogen-bonded urea enhances, and can be found both within the phase mixed and as phase-separated. Furthermore, the percentage of hydrogen-bonded urethane increases, whereas free urethane goes down with the incorporation of GO in the RPU foams. The analysis of the polymer morphology allows concluding that the addition of GO generates a slight decrease of the polymerization reaction and, consequently, lower formation of chemical crosslinking network inside RPU foam. In addition to that, the incorporation of GO favours the blowing reaction, but there are more urea hard segments dissolved in polyol-rich soft domains than urea microphase separation domains. It is well known that these two modifications should have a negative influence on the mechanical properties of RPU foams with GO [28, 31, 32], as discussed below. Table 3. Relative area percentage of the absorbances of urea and urethane carbonyls detected in the amide I region after one month when foams are cured [33]. Material Urea relative area % Urethane relative area % Main reaction relative area % % Ordered H-Bonded urea carbonyl % Disordered H-Bonded urea carbonyl % Free Urea carbonyl % H-Bonded urethane carbonyl % Free urethane carbonyl % Total Urea % Total Urethane Pure (reference) 23.1 17.6 18.1 29.4 29.4 41.2 58.8 0.017 wt% GO 23.0 17.7 20.6 29.6 26.8 43.6 56.4 0.033 wt% GO 23.4 17.9 20.3 29.7 26.7 43.6 56.4 0.083 wt% GO 22.3 17.3 19.9 29.5 28.2 42.3 57.7 3.4. Thermal conductivity Thermal conductivity is a crucial parameter for RPU foams, given their applications as insulation materials. The values of the thermal conductivity of PUR foams (pure and containing GO) is shown in Figure 4. These thermal measurements were performed 40 days after the foams production, when the foam has reached a stationary state, what means that the 257 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties urethane percentage of the foam containing GO is higher than that of the pure foam. However, after 30 minutes the final urea percentage is increased when GO is incorporated to the foam compared to the pure material. Finally, one month after the foam production the cured foam containing GO presents a urea percentage slightly higher (44%) than the cured reference material (41%). Therefore, the foam containing GO slightly favours the blowing reaction (responsible of both urea linkages formation and gas formation) in comparison to the pure foam. This result may be explained considering that the GO layers are able to retain water on its layer structure, what has been experimentally evidenced previously [39]. Figure 12. Relative area percentage of the absorbances of urea or urethane detected in the amide I region at initial time, after 30 minutes and one month after the foams production. Reaction temperature measurements Gelling and blowing reactions generated during foaming are exothermic reactions. As the cellular structure of the RPU foam is being formed, the center of foam is isolated from its surroundings, so it may be considered as an adiabatic system. Thus, the adiabatic temperature monitoring of the RPU foam may be related to the reactions kinetics, and hence to the 264 Chapter 6 consumption of isocyanate in both gelling and blowing reactions [40]. Figure 12 shows the adiabatic temperature evolution versus time at the different heights where the four thermocouples are placed in the plastic cup during RPU formation (Figure 1). Both foams show similar temperatures rising profiles, independently on the zone studied. As expected, higher temperatures are reached at thermocouples placed at higher heights in comparison with those placed below. Moreover, the maxima temperatures reached at the different heights are greater for the pure material compared to those for the foam with GO. This may be related to the higher isocyanate consumption occurring during the formation of the pure foam (as shown by FTIR measurements) which generates more amount of products. Figure 12. Temperature evolution during foaming process detected by thermocouples for all the analysed samples. Kinetic studies: a summary Considering the amount of information provided by the techniques used, we can make a proposal about how the reaction kinetics is modified when polyol functionalized with GO is used to obtain RPU foams. At the first stages of the foaming process, the infrared expandometry shows a faster growing of the foam containing GO respect to the pure foam, reaching earlier the maxima expansion rate. On the other hand, the FTIR spectroscopy shows more urethane formation for the foam with GO, and the polyol functionalized with GO has also higher viscosity. All this could make the viscosity in the polymer appropriate to prevent the escape of carbon dioxide generated in the blowing reaction, what could explain the rapid grow of the foam containing GO at the beginning of the reaction. In addition to the main reactions involved in PU foams, additional processes may occur. The isocyanate might also react with carboxyl and/or hydroxyl groups of the GO surface, giving rise to the formation of amides and CO2 and/or urethanes, respectively [41]. Therefore, the higher urethane percentage obtained when GO is present may be due not only to the gelling reaction between isocyanate and polyol, but also to the reaction between isocyanate and the hydroxyl groups on the GO surface [41]. 265 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties At the final stages of the foaming process, the FTIR results show that the presence of GO slightly enhances the blowing reaction, thus giving higher percentages of urea groups, and hence a higher expansion for the foam containing 0.033 wt% of GO. This result may be explained considering that the GO layers are able to retain water on its layer structure [39], and that the retained water is more easily released at high reaction temperatures, increasing the blowing reaction at the late stages of the foaming process. Considering isocyanate consumption, it is slightly higher for the pure material, indicating that the reactions to obtain the pure foam generate more products. This is also consistent with the higher temperatures reached in this system. 4. Conclusions Polyols functionalized with GO and their use for obtaining water-blown RPU foams are herein described. The use of polyol functionalized with GO precludes the agglomeration of fillers, making unnecessary its dispersion in the PU matrix. This solves one of the main challenges of making polymer nanocomposites. The optimum content of GO in the final foam is 0.033 wt%, since it leads to reduce the cell size, and consequently the thermal conductivity, without decisive modifications of other important aspects of the cellular structure, such as density, open cell content or anisotropy. The mechanical properties are not improved when GO is present, mainly because there is a negative modification of the polymer morphology in all foams, and also a deterioration of the cellular structure occurs in the case of the foam with higher GO content (0.083 wt%). The kinetic studies show how the reaction kinetics are modified by the presence of GO. In summary, at the beginning of the foaming process the system with GO presents more urethane generation and, due to its higher viscosity, this material is able to entrap the evolving carbon dioxide, resulting in a fast expansion. On the other hand, at the late stages of the foaming process, the presence of GO enhances the blowing reaction, due to the release of the retained water on GO layers. Moreover, the foam containing GO presents a slightly decrease of isocyanate conversion, and this is also consistent with the lower temperatures reached for this system. 266 Chapter 6 Acknowledgements Financial assistance from MINECO, FEDER, UE (MAT2015-69234-R) and the Junta de Castile and Leon (VA011U16) are gratefully acknowledged. Authors thank Graphenea S.A. for supplying us with Graphene Oxide and for the technical discussion. References [1] M. Szycher, Szycher's Handbook of Polyurethanes, Second ed., CRC Press Boca Raton, Florida, USA, 2012. [2] G. Harikrishnan, S.N. Singh, E. Kiesel, C.W. Macosko, Nanodispersions of carbon nanofiber for polyurethane foaming, Polymer, 51 (2010) 3349-3353. [3] M.C. Saha, M.E. Kabir, S. Jeelani, Enhancement in thermal and mechanical properties of polyurethane foam infused with nanoparticles, Materials Science and Engineering: A, 479 (2008) 213-222. [4] S.H. Kim, M.C. Lee, H.D. Kim, H.C. Park, H.M. Jeong, K.S. Yoon, B.K. Kim, Nanoclay reinforced rigid polyurethane foams, Journal of Applied Polymer Science, 117 (2010) 1992- 1997. [5] S. Estravís, J. Tirado-Mediavilla, M. Santiago-Calvo, J.L. Ruiz-Herrero, F. Villafañe, M.A. Rodríguez-Pérez, Rigid polyurethane foams with infused nanoclays: Relationship between cellular structure and thermal conductivity, European Polymer Journal, 80 (2016) 1–15. [6] M.M.A. Nikje, Z.M. Tehrani, Thermal and mechanical properties of polyurethane rigid foam/modified nanosilica composite, Polymer Engineering & Science, 50 (2010) 468–473. [7] L. Zhang, E.D. Yilmaz, J. Schjødt-Thomsen, J.C. Rauhe, R. Pyrz, MWNT reinforced polyurethane foam: Processing, characterization and modelling of mechanical properties, Composites Science and Technology, 71 (2011) 877-884. [8] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Electric Field Effect in Atomically Thin Carbon Films, Science, 190 (2004) 666. [9] S. Eigler, A. Hirsch, Chemistry with graphene and graphene oxide-challenges for synthetic chemists, Angewandte Chemie, 53 (2014) 7720-7738. [10] K. Hu, D.D. Kulkarni, I. Choi, V.V. Tsukruk, Graphene-polymer nanocomposites for structural and functional applications, Progress in Polymer Science, 39 (2014) 1934-1972. [11] W. Hummers, R. Offeman, Preparation of Graphitic Oxide, Journal of American Chemistry Society, 80 (1958) 1339. [12] D. Yan, L. Xu, C. Chen, J. Tang, X. Ji, Z. Li, Enhanced mechanical and thermal properties of rigid polyurethane foam composites containing graphene nanosheets and carbon nanotubes, Polymer International, 61 (2012) 1107-1114. [13] M.M. Bernal, M. Martin-Gallego, I. Molenberg, I. Huynen, M.A. López Manchado, R. Verdejo, Influence of carbon nanoparticles on the polymerization and EMI shielding properties of PU nanocomposite foams, RSC Advances, 4 (2014) 7911. [14] A. Lorenzetti, M. Roso, A. Bruschetta, C. Boaretti, M. Modesti, Polyurethane-graphene nanocomposite foams with enhanced thermal insulating properties, Polymers for Advanced Technologies, 27 (2016) 303-307. [15] A Graphene chemically-modified hard polyurethane foam. China Patent CN 104844781 A. 2015. [16] A.-K. Appel, R. Thomann, R. Mülhaupt, Hydroxyalkylation and Polyether Polyol Grafting of Graphene Tailored for Graphene/Polyurethane Nanocomposites, Macromolecular Rapid Communications, 34 (2013) 1249-1255. [17] ASTM D1622-08: Standard Test Method for Apparent Density of Rigid Cellular Plastics. 267 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties [18] ASTM D6226-10: Standard Test Method for Open Cell Content of Rigid Cellular Plastics. [19] J. Pinto, E. Solorzano, M.A. Rodriguez-Perez, J.A. de Saja, Characterization of the cellular structure based on user-interactive image analysis procedures, Journal of Cellular Plastics, 49 (2013) 555–575. [20] ISO 22007‐2:2008. Plastics‐Determination of Thermal Conductivity and Thermal Diffusivity—Part 2: Transient Plane Heat Source (Hot Disc) Method. [21] O. Almanza, M.A. Rodríguez-Pérez, J.A. de Saja, Applicability of the Transient Plane Source Method To Measure the Thermal Conductivity of Low-Density Polyethylene Foams, Journal of Polymer Science: Part B: Polymer Physics, 42 (2004) 1226–1234. [22] ASTM D1621: Standard Test Method for Compressive Properties Of Rigid Cellular Plastics. [23] M.J. Elwell, A.J. Ryan, H.J.M. Grünbauer, H.C.V. Lieshout, In-Situ Studies of Structure Development during the Reactive Processing of Model Flexible Polyurethane Foam Systems Using FT-IR Spectroscopy, Synchrotron SAXS, and Rheology, Macromolecules, 29 (1996). [24] M.J. Elwell, A.J. Ryan, An FT i.r. study of reaction kinetics and structure development in model flexible polyurethane foam systems, Polymer, 37 (1996) 1353–1361. [25] D.P. Queiroz, M.N. de Pinho, C. Dias, ATR-FTIR Studies of Poly(propylene oxide)/Polybutadiene Bi-Soft Segment Urethane/Urea Membranes, Macromolecules, 36 (2003) 4195–4200. [26] A. Marcos-Fernández, A.E. Lozano, L. González, A. Rodríguez, Hydrogen Bonding in Copoly(ether-urea)s and Its Relationship with the Physical Properties, Macromolecules, 30 (1997) 3584–3592. [27] M. Santiago-Calvo, S. Pérez-Tamarit, J. Tirado-Mediavilla, F. Villafañe, M.A. Rodríguez- Pérez, Infrared expandometry: a novel methodology for monitoring blowing kinetics of cellular materials with exothermic foaming mechanisms, Polymer Testing, 66 (2018) 383–393. [28] J. Zou, Y. Chen, M. Liang, H. Zou, Effect of hard segments on the thermal and mechanical properties of water blown semi-rigid polyurethane foams, Journal of Polymer Research, 22 (2015). [29] W. Li, A.J. Ryan, Effect of Chain Extenders on the Morphology Development in Flexible Polyurethane Foam, Macromolecules, 35 (2002) 6306–6312. [30] A.M. Heintz, D.J. Duffy, C.M. Nelson, Y. Hua, S.L. Hsu, W. Suen, C.W. Paul, A Spectroscopic Analysis of the Phase Evolution in Polyurethane Foams, Macromolecules, 38 (2005) 9192-9199. [31] D.V. Dounis, G.L. Wilkes, Structure-property relationships of flexible polyurethane foams, Polymer 38 (1997) 2819-2828. [32] M.F. Sonnenschein, B.L. Wendt, Design and formulation of soybean oil derived flexible polyurethane foams and their underlying polymer structure/property relationships, Polymer, 54 (2013) 2511-2520. [33] L. Ning, W. De-Ning, Y. Sheng-Kang, Hydrogen-Bonding Properties of Segmented Polyether Poly(urethane urea) Copolymer, Macromolecules, 30 (1997) 4405–4409. [34] M.A. Rodríguez-Pérez, O. Alonso, J. Souto, J.A. de Saja, Thermal Conductivity of Crosslinked Closed Cell Polyolefin Foams, Polymer Testing, 16 (1997) 287-298. [35] O. Almanza, M.A. Rodríguez-Pérez, J.A. de Saja, The Thermal Conductivity of Polyethylene Foams Manufactured by a Nitrogen Solution Process, Cellular Polymers, 18 (1999) 385-401. [36] E. Solórzano, M.A. Rodriguez-Perez, J. Lázaro, J.A. de Saja, Influence of Solid Phase Conductivity and Cellular Structure on the Heat Transfer Mechanisms of Cellular Materials: Diverse Case Studies, Advanced Engineering Materials, 11 (2009) 818-824. 268 Chapter 6 [37] T. Kuilla, S. Bhadra, D. Yao, N.H. Kim, S. Bose, J.H. Lee, Recent advances in graphene based polymer composites, Progress in Polymer Science, 35 (2010) 1350-1375. [38] L. Gibson, M. Ashby, Cellular solids: structure and properties, Pergamon Press, Oxford, 1988. [39] D.R. Dreyer, A.D. Todd, C.W. Bielawski, Harnessing the chemistry of graphene oxide, Chemical Society reviews, 43 (2014) 5288-5301. [40] L.D. Artavia, C.W. Macosko, Foams kinetics, Journal of Cellular Plastics, 26 (1990) 490-511. [41] S. Stankovich, R.D. Piner, S.T. Nguyen, R.S. Ruoff, Synthesis and exfoliation of isocyanatetreated graphene oxide nanoplatelets, Carbon, 44 (2006) 3342-3347. 269 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties Applied polymer science 136 (2019) 47474 https://doi.org/10.1002/app.47474 Improvement of thermal and mechanical properties by control of formulations in rigid polyurethane foams from polyols functionalized with graphene oxide Mercedes Santiago-Calvo1,*, Victoria Blasco2, Carolina Ruiz2, Rodrigo París2, Fernando Villafañe3, Miguel-Ángel Rodríguez-Pérez1 1 Cellular Materials Laboratory (CellMat), Condensed Matter Physics Department, Faculty of Science, University of Valladolid, Campus Miguel Delibes, Paseo de Belén 7, 47011 Valladolid, Spain 2 D.C. Technology and Corporate Venturing, Repsol S.A. C/ Agustín de Betancourt s/n, 28935 Móstoles, Spain 3 GIR MIOMeT-IU Cinquima-Química Inorgánica, Faculty of Science, University of Valladolid, Campus Miguel Delibes, Paseo de Belén 7, 47011 Valladolid, Spain * Corresponding author: [email protected]a.es Abstract This paper addresses the optimization of water-blown rigid polyurethane foams (RPU) obtained from a polyol functionalized with graphene oxide (GO). For this purpose, a series of RPU foams are herein synthesized by varying either the isocyanate index, the contents of catalyst or the contents of surfactant, or a combination of these three components. The modifications introduced in the formulation are based on the effect of GO on the reaction kinetics. These strategies are mainly focused on the increase of both isocyanate conversion and polymerization reaction, which decrease for the foams containing GO. Density, cellular structure, thermal conductivity, and mechanical properties of the resulting foams are herein investigated. The results show how controlling PU formulation allows to improve both the thermal and the mechanical behaviour in these RPU foams containing GO. The highest cell size reduction of 25% and the lowest thermal conductivity are obtained for the sample with a simultaneous increase of isocyanate index, catalyst content, and surfactant content. 270 Chapter 6 Moreover, the adequate combination of these components leads to a high improvement of 59% of the relative Young’s modulus, and of 54% of the relative collapse stress. Keywords: Polyurethane foam; graphene oxide; formulation; thermal conductivity; mechanical properties 1. Introduction Rigid polyurethane (RPU) foams are one of the most important thermal insulating materials, since they can also be used simultaneously as structural materials, due to their combination of low weight, low thermal conductivity and good mechanical strength. Therefore, RPU foams play an essential role in many industries, such as construction, refrigeration, and piping [1]. In recent years, the promising characteristics displayed by polymer nanocomposite foams [2] have led them to gain significance in the group of lightweight materials. The addition of tiny amounts of nanoparticles to the polymer matrix may give rise to excellent improvements of a wide variety of physical features, such as thermal, mechanical, and barrier properties. This topic is being widely studied, in particular by several works reporting the effects of nanoparticles such as nanoclays, nanosilicas, carbon nanotubes, carbon nanofibers, graphene or graphene oxide (GO) on PU nanocomposites [3-11]. A proper dispersion of the nanoparticles in the PU matrix is essential to optimize their behaviour, what requiring special surface treatments and/or specific dispersion methods. The inclusion of nanoparticles into PU foams may also modify the kinetics of the reactions, which would give rise to unexpected consequences on the final properties of the final PU matrix, since some properties may be improved, but others may be worsened. The chemistry of PU foams is based on two simultaneous reactions: the polymerization or gelling reaction, where an isocyanate and a polyol generate the three-dimensional polyurethane network, and the foaming or blowing reaction, where a gas is generated by the addition of physical or chemical blowing agents [12]. Water is commonly used as chemical blowing agent in RPU foams, since it reacts with isocyanate giving urea groups and CO2, which contributes to the foam expansion. Physical blowing agents, such as cyclopentane or isopentane, can also be added since their vaporization contributes to the foam expansion. The PU formulation determines the final density, morphology and rigidity of the foam, thus defining which PU foams are appropriate for a desired application. In a previous study we described the formation of water-blown RPU foams from polyols functionalized with low amounts of graphene oxide (GO) (0.017, 0.033 and 0.083 wt%) [13]. One of the main advantages of this approach is that GO particles are chemically incorporated into polyol chains, which precludes the agglomeration of fillers. Among the RPU foams therein obtained, that containing 0.033 wt% GO (optimum system) gave the better insulating properties. This foam displayed a cell size decrease of 19%, and consequently a thermal 271 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties conductivity reduction of ca. 4% compared with that of the pure material without GO particles. However, the Young`s modulus and collapse stress did not improve with the presence of GO. This reduction of the mechanical properties for the optimum system was explained considering a different polymer morphology in the matrix of the foams, since the cellular structure did not worsen. Therefore, a detailed kinetic study was carried out by infrared expandometry [14], FTIR spectroscopy, and temperature measurements in order to identify the effect of the polyol functionalized with GO on the reaction kinetics, and consequently on the final polymer morphology. This study showed how the presence of GO increased the amount of urethane groups when the foaming process started, and how the higher viscosity of this firstly formed foam favoured the entrapment of the evolving CO2, resulting in a fast expansion. However, at the late stages of the foaming process, the presence of GO enhanced the urea generation, due to the release of the retained water on the GO layers. Furthermore, the foams containing GO showed a slight decrease of isocyanate conversion, and consequently lower temperatures were achieved with respect to pure material. In summary, it was concluded that the presence of GO in the polyol modified the reaction kinetics and therefore the properties. Consequently, this formulation could be more finely tuned in order to improve certain properties. The present work is based on the information collected from the aforementioned research [13], the main target being to improve the mechanical properties of RPU foams functionalized with GO, whereas the thermal conductivity is maintained or improved. For this purpose, we have chosen as starting material the better foam previously obtained, i.e. that containing 0.033 wt%. Using the previous data referring to the modifications of the reaction kinetics induced by the presence of GO in the polyol, a series of RPU foams are herein synthesized by varying the isocyanate index, the amounts of catalyst, the amounts of surfactant, or a combination of these components. Density, cellular structure, thermal conductivity and mechanical properties of the resulting foams are herein evaluated. The results show that, by a proper control of the PU formulation, the use of polyol functionalized with GO allows improving both the thermal and the mechanical properties simultaneously. 2. Experimental 2.1. Materials IsoPMDI 92140 (31.5% NCO, density 1.23 g cm-3, viscosity 170-250 mPa·s), a polymeric diphenylmethane diisocyanate (pMDI) from BASF Poliuretanos Iberia S.A (Rubí, Spain), was used to produce the RPU foams. TEGOAMIN® DMCHA (N,N-dimethylcyclohexylamine) from Evonik Nutrition & Care GmbH (Essen, Germany), was used as gelling and foaming catalyst. TEGOSTAB® B 8522 (a non-hydrolysable poly-ether-polydimethyl-siloxane–stabilizer) from 272 Chapter 6 Evonik Nutrition & Care GmbH (Essen, Germany), was used as surfactant. Distilled water was used as a blowing agent. Polyol functionalized with 1000 ppm of GO (OH index 358 mg KOH/g, viscosity 547.6 mPa·s, 570 Da) was synthesized by Repsol S.A (Móstoles, Spain) [13]. 2.2. Preparation of rigid polyurethane/graphene oxide nanocomposite foams RPU foam containing 0.033 wt% GO (1000 ppm in polyol) is the reference material whose components were modified in order to study their effect on the RPU properties [13]. Isocyanate index was varied from 115 to 130 (Table 1), gelling and foaming catalyst from 0.5 parts per weight (ppw) to 2 ppw (Table 2), and surfactant from 1 ppw to 2 ppw (Table 3). Moreover, a combination of the above strategies was also considered (Table 4). An overhead stirrer (EUROSTAR Power control-visc P1, IKA) with a 50 mm diameter Lenart disc stirrer was used to premix the polyol with the catalyst, surfactant, and blowing agent during 2 minutes at 250 rpm, to obtain a homogenous polyol blend (labelled as Part B = Polyol component). Part B and the isocyanate (Part A) were mixed in a plastic cup at 1200 rpm for 10 seconds. Free-rise foams were cured for 2 days at room temperature, and then the foam samples were cut for characterization. Table 1. PUR formulations with different isocyanate index from 115 to 130. Isocyanate component Polyol component Samples Isocyanate index Polyol [ppw] Surfactant [ppw] Catalyst [ppw] Blowing agent (H2O) [ppw] PU-115ISO 115 100 1 1 5 PU-120ISO (Reference) 120 100 1 1 5 PU-125ISO 125 100 1 1 5 PU-130ISO 130 100 1 1 5 273 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties On the other hand, the sample with the highest amount of catalyst (PUR_2CAT) affords the larger cell size reduction (15%), what might be attributed to the formation of important amount of gas bubbles [25]. This reduction is clearly observed in the SEM micrographs, shown in Figure 4. Furthermore, the sample with the highest amount of catalyst content (PUR_2CAT) is the most homogeneous and displays a highly symmetric cell size distribution, as indicated by the very small value of NSD and by a AC value near to zero. Additionally, AR increases with the catalyst content, being the sample with the highest content (PUR_2CAT) being the one with the higher AR value. Table 7. Density, open cell content (OC), mean cell size (Ф3D), normalized standard deviation (NSD), asymmetry coefficient (AC) and anisotropy ratio (AR) for foams with different catalyst content. Samples Density (Kg/m3) OC (%) Ф3D (µm) NSD AC AR PU-0.5CAT 33.7±0.3 8.12±0.62 427±98 0.23 0.39 1.30±0.28 PU-1CAT (Reference) 33.1±0.1 7.09±0.45 406±103 0.25 0.56 1.38±0.31 PU-1.5CAT 31.8±0.4 8.12±0.62 361±87 0.24 1.11 1.40±0.30 PU-2CAT 31.9±0.7 8.43±0.31 346±63 0.18 -0.22 1.49±0.45 Figure 4. SEM micrographs of foams with different catalyst content. 280 Chapter 6 The relative collapse stress and relative Young’s modulus of foams with different content of catalyst (Figure 5.a) evidence a slight increase of the mechanical properties with the increasing content of catalyst in the PU formulation. Therefore, the sample with the highest content (PUR_2CAT) achieves the highest enhancement. Compared to the reference foam, that obtained with 2 ppw of catalyst exhibits an increase of a 19% of the relative Young’s modulus, and a 15% of the relative collapse stress. As in the case of the foams with different isocyanate index, the cellular structure for the foams is not deteriorated when the catalyst content increases. Therefore, the improvement of the mechanical behaviour could be due to the higher isocyanate conversion promoted by the catalyst increase, and also to the decrease of the cell size. As a general trend, the value of thermal conductivity decreases with the catalyst content (Figure 5.b). As mentioned above, the increase of the catalyst content produces a significant cell size reduction in the foams, which would imply a decrease of the radiation contribution to the thermal conductivity [6, 23, 24]. Furthermore, the density decrease of the foams with higher catalyst content involves a decrease of the solid phase conduction contribution to the thermal conductivity. Therefore, the result obtained in figure 4b can be clearly explained taking into account the density and cellular structure results. Figure 5. (a) Mechanical and (b) thermal properties for the foams with different catalyst content. 3.3. Influence of surfactant The presence of surfactants has an important role in the PU formulation because they stabilize the gas bubbles formed during nucleation, thus inhibiting coalescence. For these reasons, the surfactant content has been increased from 1 ppw to 2 ppw, and the effect of this increase on the RPU foam properties has been evaluated. As shown in Table 8, no changes of density and of cellular structure parameters are detected when the surfactant content is increased. Figure 6 shows how the cells of the RPU foam remain small and uniform when the surfactant content is increased to 2 ppw. 281 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties Table 8. Density, open cell content (OC), mean cell size (Ф3D), normalized standard deviation (NSD), asymmetry coefficient (AC) and anisotropy ratio (AR) for foams with different surfactant content. Samples Density (Kg/m3) OC (%) Ф3D (µm) NSD AC AR PUR-1SURF (Reference) 33.1±0.1 7.09±0.45 406±103 0.25 0.56 1.38±0.31 PUR-2SURF 32.0±1.4 6.86±1.28 393±96 0.24 -0.26 1.48±0.28 Figure 6. SEM micrographs of foams with different surfactant content. The effect of the surfactant on the mechanical properties of RPU foam is observed in Figure 7.a. The compressive properties slightly increase when the surfactant content rises from 1 to 2 ppw, increasing the relative Young’s modulus by 10% and the relative collapse stress by 17%. Since the cellular structure is maintained when the surfactant content increases, this slight improvement of the mechanical properties could be due to either the modification of the polymer matrix morphology either to a different distribution of the material in cell walls and struts. 282 Chapter 6 Figure 7. (a) Mechanical and (b) thermal properties for the foams with different surfactant content. On the other hand, the thermal conductivity of the RPU foam containing 2 ppw of surfactant is similar to that of the reference foam, as shown in Figure 7.b. Thus, the thermal properties hardly change when the surfactant content is increased mainly because the cellular structure and density are not modified. 3.4. Combination of modifications Based on the results described above, where the effect of isocyanate index, catalyst or surfactant on the cellular structure and properties of the foam were separately studied, we have made a combination of the contents of each component with the aim of generating a substantial improvement in the thermal and mechanical foam properties. Therefore, two new materials (PUR-130ISO-2CAT-1SURF and PUR-130ISO-2CAT-2SURF) with changes in the PU formulation, have been designed and their properties have been compared with those of the reference material (PUR-120ISO-1CAT-1SURF). The isocyanate index (130) and the catalyst content (2ppw) have been increased and the surfactant content is maintained (1 ppw) in the PUR-130ISO-2CAT-1SURF sample. On the other hand, the isocyanate index (130), the catalyst content (2ppw), and also the surfactant content (2 ppw) have been increased in the PUR- 130ISO-2CAT-2SURF sample. Table 9 collects densities and cellular structure characteristics for these new RPU foams. The density of the sample with higher isocyanate index and catalyst content increases by 3 kg/m3 compared to that of the reference foam, whereas the density of the foam with higher isocyanate index, catalyst and surfactant contents decreases by around 1 kg/m3. The open cell content (OC) value is slightly lower for the PUR-130ISO-2CAT-1SURF sample than that of the reference foam. However, an evident decrease of the OC value is observed when the surfactant content is also increased, since the PUR-130ISO-2CAT-2SURF sample gives a lower OC value. Furthermore, a decisive cell size reduction is obtained for these new foams, being 22% for PUR-130ISO-2CAT-1SURF, and 25% for the PUR-130ISO-2CAT-2SURF sample. This 283 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties reduction is clearly observed in the SEM micrographs shown in Figure 8. The PUR-130ISO- 2CAT-2SURF sample presents similar AR value than that of the reference material, although this AR value is reduced for the PUR-130ISO-2CAT-1SURF sample. All foams studied have a high homogeneity (with small NSD) and a high symmetry (AC near to zero). Table 9. Density, open cell content (OC), mean cell size (Ф3D), normalized standard deviation (NSD), asymmetry coefficient (AC) and anisotropy ratio (AR) for foams with different combination of components. Samples Density (Kg/m3) OC (%) Ф3D (µm) NSD AC AR PUR-120ISO- 1CAT-1SURF (Reference) 33.1±0.1 7.09±0.45 406±103 0.25 0.56 1.38±0.31 PUR-130ISO- 2CAT-1SURF 36.2±0.1 6.37±0.56 318±53 0.17 0.07 1.11±0.20 PUR-130ISO- 2CAT-2SURF 31.9±0.2 4.32±0.03 306±65 0.21 0.39 1.39±0.28 Figure 8. SEM micrographs of foams with different combination of components. The values of relative Young’s modulus and relative collapse stress of these foams are collected in Figure 9.a, and show a noticeable improvement of the mechanical properties. Compared to the reference foam, the sample with higher isocyanate index and catalyst (PUR- 130ISO-2CAT-1SURF) presents an increase of 60% in the compressive modulus, and of 40% in the collapse stress. On the other hand, the sample with higher isocyanate index, catalyst and surfactant (PUR-130ISO-2CAT-2SURF) reaches an increase of 59% in the compressive modulus, 284 Chapter 6 whereas the increase of the collapse stress is 54%. This enhancement of the mechanical properties could be mainly due to the higher isocyanate conversion promoted by the catalyst and isocyanate increase, and also to the improvement of the cellular structure by these components combination. The values of the thermal conductivity of these two RPU foams are shown in Figure 9.b. As expected, thermal conductivity decreases with cell size reduction, because the radiative contribution to thermal conductivity is also reduced [6, 23, 24]. The significant cell size reduction achieved with these combined formulations can be appreciated from Table 9. The sample with higher isocyanate index and catalyst (PUR-130ISO-2CAT-2SURF) displays smaller cell size, but its thermal conductivity is not significantly reduced because it presents a higher density, which instead contributes to the solid phase conductivity increase [23, 24]. However, the thermal conductivity of the sample with higher isocyanate index, catalyst and surfactant (PUR-130ISO-2CAT-2SURF) is reduced by ca. 4%, compared to that of the reference material. In this case, this may be explained not only by the cell size reduction, but also by the decrease of the density and of the OC value, as well as a by its homogenous cellular structure. Figure 9. Mechanical and thermal properties for the foams with different combination of components. The main purpose of this study was improving the mechanical properties of the RPU foam containing GO, and at the same time maintaining or reducing the thermal conductivity. In conclusion, the synergy between the adequate combination of these three components in the PU formulation (isocyanate index, catalyst and surfactant) made possible to achieve a maximum enhancement of the mechanical and thermal properties. 285 Optimization of rigid polyurethane formulation from kinetics results to obtain rigid polyurethane foams reinforced with graphene oxide with better thermal and mechanical properties 4. Conclusions The formulation of a water-blown RPU foam from a polyol functionalized with GO have been modified in order to improve their mechanical properties. Starting from the experimental data on the reaction kinetics of the system containing GO in comparison with the system without GO, a series of modifications has been defined and tested. In particular, the effects of either the isocyanate index, catalyst, or surfactant contents, as well as a combination of these modifications of these three components, on the density, cellular structure, and mechanical and thermal properties of the final foams have been evaluated. The best performing formulation is that containing 130 of isocyanate index, 2 ppw of catalyst and 2 ppw of surfactant, which gives rise to a final foam where the cell size is reduced by 25%, as well as to significant reductions of density, and of open cell content. Correspondingly, the thermal conductivity is reduced by 4%. Moreover, the mechanical properties are highly improved, so as the Young’s modulus and the collapse stress are increased by 59% and 54% respectively. In conclusion, the strategies used in this paper has resulted in significant improvements of both mechanical and thermal properties for a RPU foam obtained from a polyol functionalized with GO. Acknowledgments Financial assistance from Spanish Ministry of Economy and Competitiveness, European Regional Development Fund, RETOS COLABORACION Project: NUMASTA. RTC-2016-5285-5 is gratefully acknowledged. Authors thank Graphenea S.A. for supplying us with Graphene Oxide and for the technical discussion. References [1] M. Szycher, Szycher's Handbook of Polyurethanes, Second ed., CRC Press Boca Raton, Florida, USA, 2012. [2] L. Chen, D. Rende, L.S. Schadler, R. Ozisik, Polymer nanocomposite foams, Journal of Materials Chemistry A, 1 (2013) 3837. [3] G. Harikrishnan, S.N. Singh, E. Kiesel, C.W. Macosko, Nanodispersions of carbon nanofiber for polyurethane foaming, Polymer, 51 (2010) 3349-3353. [4] M.C. Saha, M.E. Kabir, S. Jeelani, Enhancement in thermal and mechanical properties of polyurethane foam infused with nanoparticles, Materials Science and Engineering: A, 479 (2008) 213-222. 286 Chapter 6 [5] S.H. Kim, M.C. Lee, H.D. Kim, H.C. Park, H.M. Jeong, K.S. Yoon, B.K. Kim, Nanoclay reinforced rigid polyurethane foams, Journal of Applied Polymer Science, 117 (2010) 1992- 1997. [6] S. Estravís, J. Tirado-Mediavilla, M. Santiago-Calvo, J.L. Ruiz-Herrero, F. Villafañe, M.A. Rodríguez-Pérez, Rigid polyurethane foams with infused nanoclays: Relationship between cellular structure and thermal conductivity, European Polymer Journal, 80 (2016) 1–15. [7] M.M.A. Nikje, Z.M. Tehrani, Thermal and mechanical properties of polyurethane rigid foam/modified nanosilica composite, Polymer Engineering & Science, 50 (2010) 468–473. [8] L. Zhang, E.D. Yilmaz, J. Schjødt-Thomsen, J.C. Rauhe, R. Pyrz, MWNT reinforced polyurethane foam: Processing, characterization and modelling of mechanical properties, Composites Science and Technology, 71 (2011) 877-884. [9] D. Yan, L. Xu, C. Chen, J. Tang, X. Ji, Z. Li, Enhanced mechanical and thermal properties of rigid polyurethane foam composites containing graphene nanosheets and carbon nanotubes, Polymer International, 61 (2012) 1107-1114. [10] M.M. Bernal, M. Martin-Gallego, I. Molenberg, I. Huynen, M.A. López Manchado, R. Verdejo, Influence of carbon nanoparticles on the polymerization and EMI shielding properties of PU nanocomposite foams, RSC Advances, 4 (2014) 7911. [11] A. Lorenzetti, M. Roso, A. Bruschetta, C. Boaretti, M. Modesti, Polyurethane-graphene nanocomposite foams with enhanced thermal insulating properties, Polymers for Advanced Technologies, 27 (2016) 303-307. [12] G. Oertel, Polyurethane Handbook 2nd ed., Hanser Publishers, Munich, 1993. [13] M. Santiago-Calvo, V. Blasco, C. Ruiz, R. París, F. Villafañe, M.Á. Rodríguez-Pérez, Synthesis, characterization and physical properties of rigid polyurethane foams prepared with poly(propylene oxide) polyols containing graphene oxide, European Polymer Journal, 97 (2017) 230-240. [14] M. Santiago-Calvo, S. Pérez-Tamarit, J. Tirado-Mediavilla, F. Villafañe, M.A. Rodríguez- Pérez, Infrared expandometry: a novel methodology for monitoring blowing kinetics of cellular materials with exothermic foaming mechanisms, Polymer Testing, 66 (2018) 383–393. [15] ASTM D1622-08: Standard Test Method for Apparent Density of Rigid Cellular Plastics. [16] ASTM D6226-10: Standard Test Method for Open Cell Content of Rigid Cellular Plastics. [17] J. Pinto, E. Solorzano, M.A. Rodriguez-Perez, J.A. de Saja, Characterization of the cellular structure based on user-interactive image analysis procedures, Journal of Cellular Plastics, 49 (2013) 555–575. [18] ISO 22007‐2:2008. Plastics‐Determination of Thermal Conductivity and Thermal Diffusivity—Part 2: Transient Plane Heat Source (Hot Disc) Method. [19] O. Almanza, M.A. Rodríguez-Pérez, J.A. de Saja, Applicability of the Transient Plane Source Method To Measure the Thermal Conductivity of Low-Density Polyethylene Foams, Journal of Polymer Science: Part B: Polymer Physics, 42 (2004) 1226–1234. [20] ASTM D1621: Standard Test Method for Compressive Properties Of Rigid Cellular Plastics. [21] H. Fan, A. Tekeei, G.J. Suppes, F.-H. Hsieh, Physical Properties of Soy-Phosphate Polyol- Based Rigid Polyurethane Foams, International Journal of Polymer Science, 2012 (2012) 1-8. [22] L. Gibson, M. Ashby, Cellular solids: structure and properties, Pergamon Press, Oxford, 1988. [23] O. Almanza, M.A. Rodríguez-Pérez, J.A. de Saja, The Thermal Conductivity of Polyethylene Foams Manufactured by a Nitrogen Solution Process, Cellular Polymers, 18 (1999) 385-401. [24] E. Solórzano, M.A. Rodriguez-Perez, J. Lázaro, J.A. de Saja, Influence of Solid Phase Conductivity and Cellular Structure on the Heat Transfer Mechanisms of Cellular Materials: Diverse Case Studies, Advanced Engineering Materials, 11 (2009) 818-824. [25] K.H. Choe, D.S. Lee, W.J. Seo, W.N. Kim, Properties of Rigid Polyurethane Foams with Blowing Agents and Catalysts, Polymer Journal, 36 (2004) 368-373. 287 288 Synthesis, foaming kinetics and physical properties of cellular nanocomposites based on rigid polyurethane Mercedes Santiago Calvo CHAPTER 7: INTERNATIONAL RESEARCH STAY: THERMOPLASTIC POLYURETHANE FOAMS International research stay: Thermoplastic polyurethane foams 2. Materials and Methods 2.1. Materials The reactants used to obtain TPU are: MDI with a molecular weight of 250.25 g/mol and a functionality of 2, PEG-PPG-PEG as macrodiol with an average Mn of ~2000 g/mol and OH index of 56-59 mg KOH/g, 1,5-PDO as chain extender with a molecular weight of 104.15 g/mol and a functionality of 2, 1,4-Diazabicyclo[2.2.2]octane (DABCO) as catalyst and N,N- Dimethylacetamide (DMAc) as solvent. All of them were supplied by Sigma Aldrich and used as received, except PEG-PPG-PEG and 1,5-PDO, which were dried in a vacuum oven overnight at 80°C and then were stored in sealed glass jars with molecular sieves, and DABCO, which was deoxygenated by bubbling nitrogen before use. Medical grade CO2 (99.9% purity) was used as blowing agent for the gas dissolution foaming experiments. 2.2. TPU synthesis Three samples of TPU were synthesized with different weight fraction of HS: 40%HS, 50%HS and 60%HS (% indicated is by weight). The synthesis of TPUs was carried out in a reaction flask by a two-step, pre-polymer method under dry nitrogen atmosphere. Firstly, PEG-PPG-PEG was added drop wise from a funnel to an excess of MDI and this mixture was heated to in an oil bath at 80°C for 2 h with stirring at 400 rpm. In the second step, the mixture of an additional amount of MDI and the prepolymer previously obtained in DMAc (240 mL), was added drop wise from the funnel to a preheated mixture of 1,5-PDO and DABCO (0.3 g, 0.003 mmol, 0.3%) in DMAc (60 mL). This reaction mixture was stirred at 400 rpm for 2h in an oil bath at 80°C. Finally, the solution containing the TPU was poured into silicone moulds and maintained in an oven at 80°C for 3 days to obtain TPU casts. The amount of each component used for the different TPUs is collected in Table 1. The molar ratio of NCO/OH was adjusted as 1.02. The pre-polymer formulation (first step of synthesis) used is calculated having a molar ratio of the number of moles of MDI and PEG-PPG-PEG polyol as 6:1 for TPU 50%HS and TPU 60%HS, and as 4:1 for TPU 40%HS. 100 g of TPU material were produced in each synthesis. 296 Chapter 7 Table 1. The amount of each component used for first and second step of TPU synthesis. FIRST STEP SECOND STEP TPU SAMPLES MDI Isocyanate PEG-PPG-PEG Polyol PREPOLYMER MDI Isocyanate 1,5-PDO Chain extender 40%HS 43.8 g, 0.175 mol 85.8 g, 0.043 mol 90.6 g 0 g, 0 mol 9.4 g, 0.090 mol 50%HS 64.2 g, 0.257 mol 85.8 g, 0.043 mol 87.4 g 0 g, 0 mol 12.6 g, 0.121 mol 60%HS 64.2 g, 0.257 mol 85.8 g, 0.043 mol 69.9 g 13.4 g, 0.054 mol 16.7 g, 0.160 mol 2.3. Samples production of solid TPU Compression molded samples were prepared by using a hot plate press. The material was first heated at a temperature above melting temperature (between 165-180 °C) for 3 min, raising the pressure to 100 bar at 0.5 bar/s. Then the samples were pressed under a constant pressure of 10 MPa for 7 min, lowering the temperature to 60°C at 25°C/min. These compression molded samples were prepared to characterize their shore hardness, WAXD, DMTA and rheology. Extruded samples with 1.5±0.5 mm of diameter were also prepared by TWELVindex extrusion plastometer from ATS Faar using a temperature above melting temperature (between 165-180 °C). These extruded samples were prepared to carry out the foaming tests. All the samples were dried in a vacuum oven overnight at 80°C before characterization and foaming process. 2.4. Solid TPU characterization 2.4.1. Density Density of the solid TPUs (ρs) (extruded samples) was measured with a gas pycnometer Accupyc II 1340 from Micromeritics. 2.4.2. Shore hardness Shore A and B hardness tenting was measured with a Bareiss U72 durometer following the “ISO 868:2003: Plastics and ebonite. Determination of indentation hardness by means of a duromer (Shore hardness)” procedure. The measurements were taken for 1 s at the temperature of 23°C. Three samples of each material from the compressed sheet with dimensions 10 x 10 x 2 mm3 were used. 297 International research stay: Thermoplastic polyurethane foams 2.4.3. Gel permeation chromatography (GPC) Number-averaged molecular weights (Mn), weigh-averaged molecular weights (Mw) and polydispersity index (PDI) were determined by GPC. A dilute solution of 1 mg/ml of synthesized TPU was prepared in tetrahydrofuran (THF) and stirred for 2 hours. The solutions were filtered through a 0.2 μm polyamide filter. Three measurements of each synthesized TPU were carried out in order to obtain an average. 2.4.4. Differential scanning calorimetry (DSC) DSC was performed using a DSC30 Mettler Toledo Instrument. Tests were carried out under nitrogen atmosphere at a heating rate of 10 °C min−1 from -90 to 220 °C. All DSC measurements were done in aluminium pans with 8 mg of extruded TPU. 2.4.5. Wide angle X-ray diffraction (WAXD) WAXD patterns were obtained using a PANalytical X’Pert Pro (XRD 5) instrument employing Cu Kα radiation (λ = 1.54 Å) as the X-ray source, and a voltage of 40 KV and 40 mA of current. The diffraction angle was scanned from 5° to 90°. The samples were cut from the compressed sheet, with dimensions 10 x 10 x 2 mm3. 2.4.6. Dynamic mechanical thermal analysis (DTMA) DTMA tests were carried out using a Q800 DMA instrument (TA) in single cantilever mode at a heating rate of 3 °C/min, an amplitude of 15 µm and a frequency of 1 Hz. The samples were cut from the compressed sheet, with dimensions 17 x 6 x 2 mm3. Three samples of each sample were measured to obtain an average. The storage modulus (E’), loss modulus (E’’) and tan δ were recorded and glass transition temperature (Tg) of hard segments (HS) and soft segments (SS) were calculated from the peak tan δ. 2.4.7. Shear rheology Cylindrical samples obtained by compression molding with a thickness of 1 mm and a diameter of 25 mm were used for the rheological tests. Rheological measurements were carried out on a strain-controlled rheometer (ARES-G2, TA Instruments) with parallel plate geometry (diameter 25 mm, gap 1 mm) under a nitrogen atmosphere. The rheological characterization was performed in frequency sweep tests at 200 °C, a constant strain of 10% and an angular 298 Chapter 7 frequency range from 0.1 to 100 rad s−1. Strain was chosen in order to measure all the samples in the linear viscoelastic region (LVR), and was estimated by an initial study through amplitude sweep test at strains ranging from 0.01% to 10000%, an angular frequency of 1 rad/s and 200 °C. 2.5. Gas dissolution foaming experiments A high pressure vessel (model PARR 4681) provided by Parr Instrument Company with a capacity of 1 L was used for the foaming tests. The maximum temperature and pressure reached by this device are 350 °C and 41 MPa, respectively. The pressure was automatically controlled by an accurate pressure pump controller (model SFT-10) provided by Supercritical Fluid Technologies Inc. The vessel is equipped with a clamp heater of 1200 W, and its temperature is controlled by a CAL 3300 temperature controller. This set up has been used to get a set of experiments by using a one-step foaming process. Firstly, samples were introduced in the pressure vessel under certain pressure and temperature conditions for the saturation stage. After saturation, pressure was rapidly released and the samples expanded in the pressure vessel after depressurization. Two sets of experiments were performed for the extruded samples. Firstly, the effect of the foaming temperature was analysed by fixing the saturation pressure to 20 MPa and the saturation time to 1 h, and varying the foaming temperature between 140, 150, 160, 170 and 180 °C. Secondly, the influence of the saturation pressure was evaluated by choosing four different saturation pressures: 10, 15, 20 and 25 MPa at 170 °C for 1 h. Extruded TPU samples were foamed and then were left to desorb the remaining CO2 before characterizing their density and cellular structure. 2.6. Characterization of TPU foams 2.6.1. Density Density of the TPU foams was determined with the water-displacement method based on Archimedes’ principle using a density determination kit with an AT261 Mettler-Toledo balance. The density of TPU foams was measured when all CO2 was desorbed. Relative density (ρr) has been calculated by the ratio between the density of TPU foam (ρf) and the density of the solid TPU (extruded samples) with the same chemical composition (ρs). 2.6.2. Cellular structure With the aim of maintaining the cellular structure for the microscopic visualization, samples were cooled in liquid nitrogen and fractured. The cellular morphology of the extruded samples was observed by Scanning Electron Microscopy (SEM) with a JEOL JSM-820 microscope. The perpendicular plane of the extrusion direction was examined by SEM after vacuum coating 299 International research stay: Thermoplastic polyurethane foams with a gold monolayer. An image analysis technique [14] of SEM micrographs was used to determine the average cell size (Ф) of the cellular structure of the TPU foamed. More than 100 cells of different areas of each cellular material have been used for this analysis. Cell nucleation density (N0) [14] has been determined according to Equation 1 where Nv is the cell density, Ф is the average cell size and ρr is the relative density. Ф Equation 1 3. Results and Discussion 3.1. Characterization of Solid TPU A complete characterization of the TPU samples allows to relate the TPU properties with the gas dissolution foamability. Relevant properties of the samples studied are collected in Table 2. The density slightly increases when the HS content increases in the TPU material. The measurement of Shore hardness allows to test the TPU resistance into two categories: Shore A is used for more flexible types of TPU whereas Shore D is referred to more rigid varieties. Table 2 shows that the shore hardness increases with HS content. TPU 60%HS is not classified in Shore A scale, thus being the most rigid material of all those studied. In Table 2, GPC measurements of the samples with different contents of HS show that the molecular weights are higher for TPUs with low contents of HS (40%HS and 50%HS) than for TPU 60%HS. The polydispersity index, it is close to 2 in all TPU studied. Table 2. Properties of TPU samples: the density of the solid TPU (ρs), shore hardness, numberaveraged molecular weights (Mn), weigh-averaged molecular weights (Mw) and polydispersity index (PDI). TPU SAMPLES ρs (Kg/m3) Hardness (Shore A) Hardness (Shore D) Mn (g/mol) Mw (g/mol) PDI 40%HS 1.068±0.002 78.8±0.4 22.6±0.5 18583±1046 36556±1576 1.97±0.06 50%HS 1.125±0.003 88.6±0.5 37.6±0.9 26403±1469 51468±1274 1.95±0.06 60%HS 1.101±0.003 - 45.6±0.5 8552±1602 18325±535 2.18±0.32 The DSC thermograms of the TPU under study are shown in Figure 1 and the results obtained are collected in Table 3. DSC thermograms of all TPU samples display a first transition at ca. - 45°C associated with the glass transition of the SS (TgSS), which is similar for all the materials studied. The second transition is observed at ca. 90°C, and is associated with the annealing 300 Chapter 7 temperatures, since the samples had been dried in a vacuum oven overnight at 80°C before the characterization and the foaming experiments. Several endothermic peaks associated with the melting temperatures (Tm 1-3) of the HS crystalline are observed at higher temperatures, between 150°C and 180°C (Table 3). The observation of several Tm peaks is related to the different order of crystalline structures. Tm of TPU systems is shifted to higher temperatures when the HS content increases, and the TPU with high HS content (60%HS) presents a sharp endothermic peak at 180°C, which may be related to the formation of highly ordered crystalline structures. Moreover, the HS crystallinity of the samples grows from 5.1% to 14.3% when the HS content increases from 40 wt. % to 60 wt. % (Table 3). Figure 1. DSC thermograms of solid TPUs with different HS%. Table 3. DSC data of solid TPUs with different HS%. TPU SAMPLES Tg SS (°C) Tm 1 (°C) Tm 2 (°C) Tm 3 (°C) HS crystallinity (%) 40%HS -47 - 154 165 5.1 50%HS -41 154 169 177 10.9 60%HS -49 - 162 180 14.3 WAXD patterns for TPU systems are shown in Figure 2. The diffraction peaks near 2θ = 19°, 20°, 21°, 22°, 24° and 26° are obviously attributed to the presence of TPU crystals. The intensity of these peaks increases when TPU samples include more HS contents (50%HS and 60%HS), because these samples present higher crystallinities. In the case of TPU 40%HS, the material is more amorphous what explains the absence crystalline peaks. The decrease of crystallinity shown in the WAXD pattern for the TPUs under study is also observed in the DSC results. 301 International research stay: Thermoplastic polyurethane foams Figure 2. WAXD pattern of solid TPUs with different HS%. The viscoelastic properties of the TPU samples were measured by DMTA. Figure 3 shows the storage modulus, the loss modulus, and damping factor (tan δ), as a function of the temperature for the TPU samples prepared. The increase of HS% leads to an increase of the storage modulus, and to a decrease of the peak intensity maximum of both the loss modulus and the tan δ [15]. The increase of HS implies an increase in cristallinity which provide rigidity to the material, and consequently the storage modulus increases [16]. Moreover, the onset of the peak of Tg measured by DMTA shows similar temperatures to those measured by DSC. Figure 3. Dynamic mechanical analysis of solid TPUs with different HS%. 302 Chapter 7 A rheological characterization was performed for the TPU samples with different HS% to determine the complex viscosity curves shown in Figure 4. TPU 40%HS and TPU 60%HS show small complex viscosities at low frequencies, whereas TPU 50%HS displays a considerably increase of complex viscosity. The difference in melt viscosity becomes smaller at relatively high shear frequency. Clearly, the TPU with 50 wt. % HS greatly increases the melt viscosity in all the angular frequencies respect to the rest of TPU systems, what could favor its foamability capability. These complex viscosities are in concordance with the molecular weights of each material (Table 2), being the TPU with 50 wt. % HS that has higher molecular weight and complex viscosity than the rest of materials. Moreover, TPU with 50 wt. % HS exhibits plastic behavior, since the slope of the regression line is -0.96 [17]. However, TPU 40%HS and TPU 60%HS have a Newtonian behavior, because the slopes of their regression lines are -0.06 and - 0.03, respectively [17]. Figure 4. Complex viscosity as function of the angular frequency for TPUs with different HS%. 3.2. Study of TPU foaming The TPUs previously prepared were foamed by gas dissolution foaming process in one-step, in order to study the effect of HS content (40 wt%, 50 wt% and 60 wt. %) in the foaming behaviour. We selected low contents in HS since previous studies showed that most of the CO2 used for foaming process is dissolved in SS [7]. In the first part of the study, we evaluated the influence of the foaming temperature by using five different foaming temperatures: 140 °C, 150°C, 160 °C, 170 °C and 180 °C. The saturation pressure and foaming time were kept constant at 20 MPa and 1h respectively. Figure 5 shows the SEM micrographs of the TPU samples after the foaming process. Only the TPU with 50 wt. % HS shows a stable cellular structure for the different foaming conditions used. However, the TPU with 60 wt. % HS does not foam, probably due to its high HS content, that results in a high crystallinity (DSC results in Table 3 and WAXD pattern in Figure 2) and a high rigidity (Shore harness in Table 2), and also its low viscosity (Figure 4). Conversely, the TPU with 40 wt. % HS with low crystallinity (DSC results in Table 3) gives rise to a highly deteriorated cellular 303 International research stay: Thermoplastic polyurethane foams structure, clearly seen in Figure 5, especially for high foaming temperatures. The low quality of this cellular structure can be attributed mainly to the low viscosity of the TPU matrix (Figure 4), which could reduce the stability of the cellular structure, thus producing cell coalescence and/or collapsing. Hence, the characteristics of the TPU with 50 wt. % HS are optimum in order to get a better foamability. The relative density, average cell size, and cell nucleation density of the samples foamed with 50 wt. % HS were measured, and the values obtained are presented in Figure 6. Figure 6.A shows that the relative density decreases as the foaming temperature increases, reaching maximum values of expansion at 170°C, when the saturation pressure and foaming time are maintained. On the one hand, the average cell size decreases when the foaming temperature increases (Figure 5 and Figure 6.B), the cellular structure being less homogeneous at lower foaming temperatures (140°C and 150°C) possibly due to the crystalline part of TPU is not melted at these temperatures. The TPU foamed at 170 °C displays a small cell size of 5 microns, and a homogeneous cellular structure. The cell nucleation density increases as the foaming temperature increases (Figure 6.C), what points to a higher number of cells in the foams produced. A maximum cell nucleation density of 3.5x109 nuclei/cm3 is reached at 170 °C for all the foaming experiments, which coincides with the lowest cell size of 5 microns and with the lowest relative density of 0.814. These results indicate that the most efficient foaming is reached at 170°C. At 180°C, the cell nucleation density is slightly reduced, and consequently both cell sizes and relative densities slightly increase. These results are a consequence of the softening of the TPU polymer, which foams above the melting temperature. Since the TPU 50%HS is the optimum content of HS, it was selected in order to optimize the foaming of our TPU system in the second part of this study. For this purpose, the influence of the saturation pressure for TPU 50%HS was evaluated using four different saturation pressures: 10, 15, 20 and 25 MPa, whereas the foaming temperature and foaming time were kept constant at 170 °C and 1 h respectively. These conditions were chosen since the TPU foams with the best characteristics had been previously obtained with this foaming temperature. SEM micrographs and parameters of the TPU foams with 50 wt. % HS are collected in Figure 7 and Figure 8. Figure 8.A shows the relative density values, which decrease with the saturation pressure, achieving the lowest relative density when using 25 MPa. Moreover, Figure 8.B indicates that a saturation pressure increase leads to important cell size reduction, what is clearly observed in the SEM micrographs shown in Figure 7. Thus, the lowest cell size reduction is obtained for TPU foamed at 25 MPa. All these results are in concordance with the trend of the cell nucleation densities shown in Figure 8.C, which increase when increasing the saturation pressure. In general, the increase of the saturation pressure causes a higher amount of gas dissolved in TPU matrix, what leads to smaller relative densities and cell sizes, and also a to a higher cell nucleation density. Nevertheless, the TPU material is scarcely foamed at low saturation pressure (10 MPa), as observed in Figure 7. 304 Chapter 7 In conclusion, the optimum foaming and saturation conditions for the TPU with 50 wt. % HS are 170 °C of saturation pressure, 25 MPa of foaming pressure and 1h of saturation time. The resulting TPU foam has the lowest relative density (0.739), the smallest cell size (4 microns), and the maximum cell nucleation density (8.0x109). In the literature [13], a TPU synthesized with similar components to that used in our study (MDI as isocyanate, macrodiol as polyol and diol as chain extender) and with a similar HS% (48 wt. % HS) was foamed in one-step, and it presented a heterogeneous cellular structure with the smallest cell size of ca. 0.5 µm, its relative density was 0.91 and the cell density was 4 x 109 cells/cm3. Therefore, the relative density (0.739) of TPU foam under our study is lower than that of the TPU foam found in the literature (0.91) [13] and also our material has an homogeneous cellular structure. Figure 6. Relative density (A), cell size (B) and cell nucleation density (C) of TPU 50%HS foamed by 1-step between 140-180°C, 20 MPa and 1h. 305 [Document text truncated for crawler view.]