Bio-based polyesters from cyclic monomers derived from carbohydrates
Abstract
Institutional repository that preserves and disseminates the academic and scientific output of the institution.
Full text
Bio-based polyesters from cyclic monomers derived from carbohydrates Ph.D. Thesis presented by Cristina Lavilla Aguilar Directed by Prof. Sebastián Muñoz Guerra Barcelona, September 2013 Tesi presentada per obtenir el títol de Doctor per la Universitat Politècnica de Catalunya Doctorat en Polímers i Biopolímers Departament d’Enginyeria Química Escola Tècnica Superior d’Enginyeria Industrial de Barcelona Universitat Politècnica de Catalunya
i Abstract Polyesters are extremely versatile polymers which can be used in a wide variety of applications ranging from high performance materials to recyclable and degradable polymers. The preparation of polyesters from renewable feedstock is currently receiving increasing attention in both industrial and academic research. This Thesis is specifically addressed to the development of aliphatic and aromatic polyesters with enhanced properties made from carbohydrate-based cyclic acetalized monomers, i.e. a bicyclic acetalized aldaric acid derived from D-galactose, bicyclic acetalized alditols derived from D-galactose and D-mannose, and a cyclic acetalized alditol derived from L-tartaric acid. The novel aliphatic polyesters derived from bicyclic acetalized galactaric acid and D-mannitol, which are biodegradable materials, are distinguished by presenting an enhanced rigidity compared to the aliphatic polyesters commonly used so far, which invariably influences their whole thermal and mechanical behavior. Bicyclic acetalized carbohydrate-based compounds are also used as comonomers in the preparation of random poly(alkylene terephthalate) copolyesters by melt polycondensation (MP). Since linear α,ω-alkanediols with varying length are employed, and the copolymerizations with bicyclic monomers are carried out for a wide range of compositions, a detailed structure-properties study is described. The effect of the presence of the carbohydrate-based comonomer on thermal and mechanical properties of the polyester is largely dependant on which unit, the diol or the diacid, is replaced. The incorporation of acetalized alditols increases the thermal stability, the glass-transition temperature and the mechanical modulus. On the contrary, these parameters are diminished when the terephthalate units are replaced by bicyclic acetalized galactaric acid. Also the hydrolytic and enzymatic degradability depends on the units introduced into the polyester backbone. The incorporation of cyclic acetalized carbohydrate-based alditols into the amorphous phase of poly(butylene terephthalate) by the solid-state modification (SSM) technique is reported, which leads to increases in glass-transition temperature. The resulting SSM-prepared copolyesters have a unique block-like chemical microstructure that endows them with superior thermal properties when compared to their random counterparts obtained by MP.
ii Given the structural proximity between isosorbide and bicyclic acetalized alditols, as well as their common potential use as polycondensation monomers, a comparative evaluation of their suitability for the synthesis of aromatic polyesters is carried out in this Thesis. The greater facility of bicyclic acetalized alditols compared to isosorbide to react under the conditions employed is highlighted. Also the influence of the symmetry and stiffness of the bicyclic structure on the thermal behavior of the copolyesters is discussed. Key words: polyester, carbohydrates, bio-based, cyclic acetalized monomers, D-galactose, D-mannose, L-tartaric acid, random copolyesters, block-like copolyesters
iii Resumen Los poliésteres son polímeros extremadamente versátiles que se usan en una amplia variedad de aplicaciones, desde materiales de altas prestaciones a polímeros reciclables o degradables. La preparación de poliésteres a partir de fuentes renovables genera cada vez más interés, tanto en la industria como en el mundo académico. En esta Tesis se emplean fuentes naturales para desarrollar nuevos poliésteres alifáticos y aromáticos con propiedades mejoradas, a partir de monómeros cíclicos derivados de carbohidratos, i.e. un ácido aldárico acetalizado bicíclico derivado de D-galactosa, alditoles acetalizados bicíclicos derivados de D-galactosa y D-manosa, y un alditol acetalizado cíclico derivado de ácido L-tartárico. Los nuevos poliésteres alifáticos derivados de ácido galactárico y D-mannitol acetalizado bicíclico, que son materiales biodegradables, destacan por presentar una rigidez mejorada respecto a los poliésteres alifáticos comúnmente usados hasta ahora, hecho que influencia invariablemente sus propiedades térmicas y mecánicas. También se reporta el uso de monómeros bicíclicos acetalizados derivados de carbohidratos para la preparación de copoliésteres poli(alquilen tereftalato)s con microestructura al azar, mediante policondensación en fundido (MP). Se emplean α,ωalcanodioles lineales con diferentes longitudes, y las copolimerizaciones con monómeros bicíclicos se llevan a cabo en un amplio rango de composiciones, lo que permite desarrollar un estudio estructura-propiedades detallado. El efecto de la unidad renovable en las propiedades térmicas y mecánicas del poliéster depende de qué unidad no renovable, diol o diácido, reemplaza. La incorporación de alditoles acetalizados aumenta la estabilidad térmica, la temperatura de transición vítrea y el módulo elástico. Por contra, estos parámetros disminuyen cuando las unidades tereftálicas se reemplazan por ácido galactárico acetalizado bicíclico. Las unidades renovables introducidas afectan también a la degradabilidad hidrolítica y enzimática de los poliésteres. Mediante la técnica de modificación en estado sólido (SSM), se incorporan alditoles cíclicos acetalizados a la fase amorfa del poli(butilen tereftalato), lo cual conlleva a un aumento de su temperatura de transición vítrea. Los copoliésteres resultantes presentan una microestructura en bloques, que los dota de propiedades térmicas superiores respecto a los copoliésteres análogos con microestructura al azar preparados mediante MP.
iv En esta Tesis también se comparan las características y propiedades del monómero isosorbide y los alditoles acetalizados bicíclicos para la síntesis de poliésteres aromáticos, debido a su proximidad estructural, así como a su uso potencial como monómeros en policondensaciones. Se destaca la mayor facilidad de reacción en las condiciones empleadas de los alditoles acetalizados bicíclicos comparados con isosorbide. También se estudia la influencia de la simetría y rigidez de la estructura bicíclica en las propiedades térmicas de los copoliésteres. Palabras clave: poliéster, carbohidratos, fuentes naturales, monómeros cíclicos acetalizados, D-galactosa, D-manosa, ácido L-tartárico, copoliésteres al azar, copoliésteres en bloque
v Resum Els polièsters són polímers extremadament versàtils que s’usen en una àmplia varietat d’aplicacions, des de materials d’altres prestacions a polímers reciclables o degradables. La preparació de polièsters a partir de fonts renovables genera cada vegada més interès, tant a la indústria com al món acadèmic. En aquesta Tesi s’utilitzen fonts naturals per desenvolupar nous polièsters alifàtics i aromàtics amb propietats millorades, a partir de monòmers cíclics derivats de carbohidrats, i.e. un àcid aldàric acetalitzat bicíclic derivat de D-galactosa, alditols acetalitzats bicíclics derivats de D- galactosa i D-manosa, i un alditol acetalitzat cíclic derivat d’àcid L-tartàric. Els nous polièsters alifàtics derivats d’àcid galactàric i D-mannitol acetalitzat bicíclic, que són materials biodegradables, destaquen per presentar una rigidesa millorada respecte als polièsters alifàtics comunament usats fins al dia d’avui, fet que influencia invariablement les seves propietats tèrmiques i mecàniques. També es reporta l’ús de monòmers bicíclics acetalitzats derivats de carbohidrats per a la preparació de copolièsters poli(alquilen tereftalat)s amb microestructura a l’atzar, mitjançant policondensació en estat fos (MP). S’utilitzen α,ωalcanodiols lineals de diferents longituds, i les polimeritzacions amb monòmers bicíclics es duen a terme en un ampli rang de composicions, fets que permeten desenvolupar un estudi estructura-propietats detallat. L’efecte de la unitat renovable en les propietats tèrmiques i mecàniques del polièster depèn de quina unitat no renovable, diol o diàcid, substitueix. La incorporació d’alditols acetalitzats augmenta l’estabilitat tèrmica, la temperatura de transició vítria i el mòdul elàstic. D’altra banda, aquests paràmetres disminueixen quan les unitats tereftàliques es substitueixen per àcid galactàric acetalitzat bicíclic. Les unitats renovables introduïdes afecten també la degradabilitat hidrolítica i enzimàtica dels polièsters. Mitjançant la tècnica de modificació en estat sòlid (SSM), s’incorporen alditols cíclics acetalitzats a la fase amorfa del poli(butilen tereftalat), fet que augmenta la seva temperatura de transició vítria. Els copolièsters resultants presenten una microestructura en blocs, que els proporciona unes propietats tèrmiques superiors respecte als copolièsters anàlegs amb microestructura a l’atzar preparats mitjançant MP. En aquesta Tesi també es comparen les característiques i propietats del monòmer isosorbide i els alditols acetalitzats bicíclics per a la síntesi de polièsters
vi aromàtics, degut a la seva proximitat estructural, així com el seu ús potencial com a monòmers en policondensacions. Es destaca la major facilitat de reacció en les condicions estudiades dels alditols acetalitzats, i també s’avalua la influència de la simetria i la rigidesa de l’estructura bicíclica en les propietats tèrmiques dels copolièsters. Paraules clau: polièster, carbohidrats, fonts naturals, monòmers cíclics acetalitzats, D-galactosa, D-manosa, àcid L-tartàric, copolièsters a l’atzar, copolièsters en bloc
vii Table of Contents Chapter 1. Aim and outline of this Thesis 1 Chapter 2. Polyesters: Traditional products and present bio-based tendencies 13 2.1. Introduction 14 2.2. Types and applications of thermoplastic polyesters 16 2.2.1. Aliphatic polyesters 16 2.2.2. Aromatic polyesters 18 2.2.3. Aliphatic-aromatic copolyesters 20 2.2.4. Fully aromatic polyesters 21 2.2.5. Thermoplastic elastomers 22 2.3. Synthetic methods 23 2.3.1. Melt polymerization 23 2.3.2. Solid-state polymerization 25 2.3.3. Solution polymerization from diacyl dichlorides 27 2.3.4. Ring-opening polymerization 27 2.3.5. Enzymatic polymerization 28 2.4. Bio-based polyesters 29 2.4.1. Polyesters partially or totally produced by bacteria 29 2.4.1.1. Polylactic acid (PLA) 29 2.4.1.2. Polyhydrohyalkanoates (PHAs) 30 2.4.2. Alternative routes from renewable sources to conventional monomers 31 2.4.3. Polyesters from carbohydrate-derived monomers with modified properties 33 2.4.3.1. Unprotected alditols and aldaric acids 33 2.4.3.2. Acyclic O-protected alditols and aldaric acids 34 2.4.3.3. 2,5-Furandicarboxylic acid (FDCA) 36 2.4.3.4. Dianhydrohexitols and derivatives 38 2.4.3.5. Cyclic acetalized alditols, aldonic and aldaric acids 47 2.5. References 49
xiv Glossary a Mark-Houwink parameter Ad Dimethyl adipate Ar 2,3,4-Tri-O-methyl-L-arabinitol ArH Aromatic proton Ar-OMe 2,3,4-Tri-O-methyl-L-arabinitol asym st Asymmetric stretching vibration ATR Attenuated total reflectance B 1,4-Butylene units BISHET Bis-(2-hydroxyethyl) terephthalate bs Broad singlet COSY 2D 1H-1H Homonuclear NMR spectra d Doublet DBTO Dibutyl tin oxide dd Doublet of doublets DFT Density functional theory dhkl Bragg spacings DMF N,N-Dimethyl formamide DMSO Dimethyl sulfoxide DMT Dimethyl terephthalate DSC Differential scanning calorimetry Ð Polydispersity δ Chemical shift (ppm) ΔHc Crystallization enthalpy ΔHcc Cold crystallization enthalpy ΔHm Melting enthalpy ΔHmº Melting enthalpy of a 100% crystalline polymer ΔT ‘Undercooling’ values required for crystallization (ΔT= Tm-Tc) E Elastic modulus E 1,2-Ethylene units ε Elongation at break Endo Secondary hydroxyl groups (1,4:3,6-dianhydrohexitols) or methylol groups (bicyclic acetalized alditols) oriented inside the bicyclic structure EG Ethylene glycol EtOH Ethanol Et2O Diethyl ether Exo Secondary hydroxyl groups (1,4:3,6-dianhydrohexitols) or methylol groups (bicyclic acetalized alditols) oriented outside the bicyclic structure F Furfural FA Furfuryl alcohol FDCA 2,5-Furandicarboxylic acid FTIR Fourier transform infrared spectroscopy
xv Galx Acetalized bicyclic non-fused 2,3:4,5-di-O-methylene structure with D- galacto configuration GPC Gel permeation chromatography HETCOR 2D 13C-1H Heteronuclear shift correlation NMR spectra HFIP 1,1,1,3,3,3-Hexafluoroisopropanol HMF Hydroxymethylfuraldehyde HMW High molecular weight HPLC High performance liquid chromatography [η] Intrinsic viscosity ICI Imperial Chemical Industries IIDCA 1-Carbon extended isoidide derivative with dicarboxylic functionality (chain extension at C2 and C5 of the isohexide skeleton), namely isoidide dicarboxylic acid IIDML 1-Carbon extended isoidide derivative with two primary hydroxyl groups (chain extension at C2 and C5 of the isohexide skeleton), namely isoidide hydroxymethylene diol IR Infrared spectroscopy Is Isosorbide; 1,4:3,6-dianhydro-D-glucitol Isoidide 1,4:3,6-Dianhydro-L-iditol Isomannide 1,4:3,6-Dianhydro-D-mannitol Isosorbide 1,4:3,6-Dianhydro-D-glucitol J Coupling constant k Avrami kinetic constant K Mark-Houwink constant LA Lactic acid LC Liquid crystalline LDH Lactate deshydrogenase Lit. Literature LMW Low molecular weight LNP Liquid nitrogen pump m In NMR spectra, multiplet peak; in WAXD measurements, medium intensity MAF Mobile amorphous fraction Manx Acetalized bicyclic fused 2,4:3,5-di-O-methylene structure with D-manno configuration Mn Number-average molecular weight Mp Peak molecular weight m.p. Melting point MP Melt polycondensation MPPBxThxyT Random copolyesters poly(butylene terephthalate-co- 2,3-di-O- methylene-L-threitol terephthalate)s, prepared by melt polycondensation Mv Viscosimetric molecular weight Mw Weight-average molecular weight N Relative molar amount of the dyads as obtained from 13C NMR
xvi n In isothermal crystallization studies, Avrami exponent; in chemical microstructure analysis by 13C NMR, number average sequence lengths NaTFA-HFIP 0.05 M sodium trifluoroacetate-hexafluoroisopropanol n.d. Not determined NMP N-Methyl-2-pyrrolidone NMR Nuclear magnetic resonance NOE Nuclear Overhauser effect NOESY Nuclear Overhauser effect spectroscopy PBF Poly(butylene 2,5-furandicarboxylate) PBGalx Poly(butylene 2,3:4,5-di-O-methylene-galactarate) PBS Poly(butylene succinate) PBT Poly(butylene terephthalate) PBTxGalxy Random copolyesters poly(butylene terephthalate-co- butylene 2,3:4,5- di-O-methylene-galactarate)s, prepared by melt polycondensation PBxGalxyT Random copolyesters poly(butylene terephthalate-co- 2,3:4,5-di-O- methylene-galactitol terephthalate)s, prepared by melt polycondensation PBxIsyT Random copolyesters poly(butylene terephthalate-co- isosorbide terephthalate)s, prepared by melt polycondensation PBxManxyS Random copolyesters poly(butylene succcinate-co-2,4:3,5-di-O- methylene-D-mannitol succinate)s, prepared by melt polycondensation PBxManxyT Random copolyesters poly(butylene terephthalate-co- 2,4:3,5-di-O- methylene-D-mannitol terephthalate)s, prepared by melt polycondensation PCL Poly(ε-caprolactone) PDGalx Poly(dodecamethylene 2,3:4,5-di-O-methylene-galactarate) PDT Poly(dodecamethylene terephthalate) PDTxGalxy Random copolyesters poly(dodecamethylene terephthalate-co- dodecamethylene 2,3:4,5-di-O-methylene-galactarate)s, prepared by melt polycondensation PEF Poly(ethylene 2,5-furandicarboxylate) PEI Poly(ethylene isophthalate) PE-nAd Poly(alkylene adipate)s, the n being the number of methylenes PE-nAdxGalxy Random copolyesters poly(alkylene adipate-co- alkylene 2,3:4,5-di-O- methylene-galactarate)s, the n being the number of methylenes, prepared by melt polycondensation PE-nGalx Poly(alkylene 2,3:4,5-di-O-methylene-galactarate)s, the n being the number of methylenes PET Poly(ethylene terephthalate) PExManxyT Random copolyesters poly(ethylene terephthalate-co- 2,4:3,5-di-O- methylene-D-mannitol terephthalate)s, prepared by melt polycondensation PGA Poly(glycolic acid) PGalxT Poly(2,3:4,5-di-O-methylene-galactitol terephthalate)
xvii PHAs Polyhydroxyalkanoates PHB Poly(3-hydroxybutyrate) PHF Poly(hexamethylene 2,5-furandicarboxylate) PHGalx Poly(hexamethylene 2,3:4,5-di-O-methylene-galactarate) PHH Poly(3-hydroxyhexanoate) PHT Poly(hexamethylene terephthalate) PHTxGalxy Random copolyesters poly(hexamethylene terephthalate-co- hexamethylene 2,3:4,5-di-O-methylene-galactarate)s, prepared by melt polycondensation PHV Poly(3-hydroxyvalerate) PIsT Poly(isosorbide terephthalate) PLA Polylactic acid PLLA Isotactic poly(L-lactic acid) PManxS Poly(2,4:3,5-di-O-methylene-D-mannitol succinate) PManxT Poly(2,4:3,5-di-O-methylene-D-mannitol terephthalate) PMMA Poly(methyl methacrylate) POF Poly(octamethylene 2,5-furandicarboxylate) Poly(HB-co-HH) Copolyester poly(3-hydroxybutyrate-co- 3-hydroxyhexanoate) Poly(HB-co-HV) Copolyester poly(3-hydroxybutyrate-co- 3-hydroxyvalerate) POM Polarizing optical microscopy PPF Poly(propylene 2,5-furandicarboxylate) PTA Purified terephthalic acid PThxT Poly(2,3-di-O-methylene-L-threitol terephthalate) PTT Poly(trimethylene terephthalate) py Pyridine θ Contact angle in degrees (º) R Randomness index of copolyesters statistically calculated on the basis of the 13C NMR analysis RAF Rigid amorphous fraction Rc Randomness index of the amorphous fraction of copolyesters rock Rocking vibration ROP Ring-opening polymerization s In NMR spectra, singlet peak; in WAXD measurements, strong intensity S Sugar-based units σ Tensile strength SA Succinic acid SEC Size-exclusion chromatography SEM Scanning electron microscopy SSM Solid-state modification SSMPBxGalxyT Block-like copolyesters poly(butylene terephthalate-co- 2,3:4,5-di-O- methylene-galactitol terephthalate)s, prepared by solid-state modification of PBT
xviii SSMPBxManxyT Block-like copolyesters poly(butylene terephthalate-co- 2,4:3,5-di-O- methylene-D-mannitol terephthalate)s, prepared by solid-state modification of PBT SSMPBxThxyT Block-like copolyesters poly(butylene terephthalate-co- 2,3-di-O- methylene-L-threitol terephthalate)s, prepared by solid-state modification of PBT SSP Solid-state polycondensation st Stretching vibration Sym st Symmetric stretching vibration t Triplet peak T Terephthalate units t0 Onset crystallization time t1/2 Half-crystallization time ºT5% In TGA, temperature at which 5% weight loss was observed TBT Titanium (IV) tetrabutoxide Tc Crystallization temperature Tcc Cold crystallization temperature Td or maxTd In TGA, temperature for maximum degradation rate TFA Trifluoroacetic acid TFA-d Deuterated trifluoroacetic acid Tg Glass-transition temperature TGA Thermogravimetry THF Tetrahydrofuran Th-OMe 2,3-Di-O-methyl-L-threitol Thx Acetalized cyclic 2,3-di-O-methylene structure with L-tartaric acid configuration Tm Melting temperature TMS Tetramethylsilane TPEs Thermoplastic elastomers tSSM SSM reaction time U One unit (U) was defined as that amount of enzyme which catalyzed the release of fatty acid from triglycerides at the rate of 1 µmol·min-1. Wavenumber (cm-1) w In WAXD measurements, weak intensity W In TGA, remaining weight at 600 ºC WAXD Wide angle X-ray diffraction X Molar composition determined by 1H NMR Xc Crystallinity index Xy 2,3,4-Tri-O-methylxylitol Xy-OBn 2,3,4-Tri-O-benzylxylitol XRD X-Ray diffraction
1 CHAPTER 1 AIM AND OUTLINE OF THIS THESIS Introduction The development of bio-based polymers is nowadays attracting a great deal of interest, due to the high price and future depletion of fossil fuel stocks, together with concerns regarding environmental sustainability. However, this approach is not new; biobased polymers have a history of more than a century, much longer than petrochemical polymers. In the 19th century, natural raw materials such as casein, shellac, gum, natural rubber, and cellulose were chemically modified to convert them into useful polymers with new properties. An important objective was to convert the infusible and frequently insoluble natural stocks into materials capable of being processed. The first horn-like plastic material was galalith, produced by reacting casein from milk with formaldehyde to form a stiff thermoset resin resembling stone. Although the biodegradable and watersoluble galalith was not moldable, sheets could be produced, thus enabling dyeing and machining. The latex of Brazilian rubber trees was collected, coagulated, dried, and vulcanized with sulfur to produce industrial rubber for making tires. When the highly flammable and explosive nitrocellulose, obtained by nitration of cellulose and used as smokeless gunpowder, was plasticized with camphor, it was rendered thermoplastic. Biobased nitrocellulose, marketed as Parkesine and Cellulloid, was rather hazardous owing to its explosive character. Nevertheless, moldable nitrocellulose became a successful substitute for ivory and was used for making billiard balls, thus saving the lives of thousands of elephants. Photographic films were possible at those times thanks to the use of nitrocellulose as supporting film for the silver nitrate emulsion, although they represented a severe fire and safety hazard for cinemas. Since then, numerous new compounds derived from renewable resources have been developed. One example is ethylene which was synthesized by the dehydration of bio-based ethanol in the 1940s. Also in that time the protein casein was used for paints and glues and later on, in formulation with formaldehyde, as plastic used for products like buttons, knives and letter openers. Soy protein was reacted with formaldehyde and cocondensed with phenol and urea; these soy plastics were used to produce several
Chapter 1 2 automotive parts such as steering wheels, glove-box doors and interior trim. Shellac was produced by extraction of the natural polymer excreted by the shell louse, and was used for paints and varnish next to small solid articles. Also regenerated cellulose, e.g. in the form of cellophane film or man-made cellulose fibers, was developed in those times and has been used in a wide range of applications, such as apparel, food (e.g. for sausages) and non-plastics (e.g. varnishes). However, many of the inventions in the 1930s and 1940s stayed in laboratory and were never used for commercial production. The main reason was the large-scale industrial use of petrochemical feedstock to manufacture a great variety of synthetic polymers since 1950s. Exploitation of oil and gas as fossil raw materials for the chemical industry and polymer production greatly improved costeffectiveness and simplified manufacturing of macromolecular materials. The attractive combination of low cost with facile processing represented the key feature of petrochemical polymers. Also some bio-based products, such as man-made cellulose fibers, continued being produced but their production did not grow at the rate of the newly emerging petrochemical products. On the other hand, starch derivatives used as paper and textile auxiliaries had a long period of growth and are still used nowadays. The oil price shocks of the 1970s led to renewed interest into the possibilities offered by non-petrochemical feedstock. However, some time after the interest in petrochemical polymers continued growing. Driven by concerns about rapidly increasing amounts of waste, in the 1990s new recycling technologies enabled effective reuse of polymer products that had completed their first lifecycle. Bio-based polymers have experienced a renaissance in the last few decades. Many new polymers from bio-based feedstock have been developed, e.g. polylactic acid (PLA) from carbohydrates. Today, public concerns about environment, climate change and limited fossil fuel resources have become more important drivers. The utilization of fossil fuels in the manufacture of plastics accounts today for about 7% of worldwide oil and gas. It is anticipated that this resource will be depleted in the coming decades; furthermore, the price of oil has increased rapidly in recent years, and it is subject to unpredictable socio-political influences. While the use of fossil reserves for transportation and heating is certainly the most serious concern, the chemical industry will also be faced with the problematic issue associated to the use of an essentially non-renewable feedstock for the majority of their products. Polymers based on renewable resources are very attractive, not only for coming from natural feedstock, but also for the benefits they provide from waste treatment point of view. In fact, many of the products derived from
Aim and outline of this Thesis 3 renewable resources can be degraded in the presence of microorganisms, therefore being able to battle against the problem of increasing amounts of waste and limited landfill capacity. In addition, polymers derived from natural sources not only provide important environmental benefits, but in some cases are also biocompatible materials, property essential for its application in biomedicine. Figure 1.1. Raw materials used by the chemical industry in hystorical perspective. Renewable feedstock (a); coal (b); oil and gas (c). Source: Lichtenthaler, F.W. Carbohydrates as Organic Raw Materials. Ullmann’s Encyclopedia of Industrial Chemistry, 2010. Carbohydrates are the most abundant natural organic compounds in plants, originated as product of photosynthesis, an endothermic reaction reducing carbon dioxide which requires the energy of the light and the pigment chlorophyll. Carbohydrates are an important source of metabolic energy, both for humans and animals. Nature produces annually about 140·109 tons of carbohydrates from carbon dioxide and water, making these compounds the most abundant organic material on earth. Humanity only uses 4% of this huge amount for food and non-food purposes. They can be looked therefore as an extraordinary source of chemicals capable of providing a wide diversity of building blocks for polycondensation polymers. Thus, the development of carbohydrate-based polymers has the potential to reduce the amount of petroleum consumed in the chemical industry and also to open new high-value-added markets to agriculture. However, there are only a few examples of carbohydrate-derived plastics commercially available, which is mostly due to the still high cost of these materials compared to petroleum-based counterparts. Another severe limitation in the use of carbohydrates as a source for monomers for the synthesis of linear polycondensates is their inherent multifunctionality; certain previous
Chapter 1 4 treatments are usually required to avoid the formation of highly cross-linked polymers. The most commonly used strategy is blocking the exceeding functional chemical groups with stable protecting groups, and thus retaining only two reactive functional groups to carry out linear polycondensation. Protection of the exceeding hydroxyl groups as methoxy ethers has led to obtaining hydrophilic linear polyesters, which in some cases were also biodegradable. However, given the flexibility of the acyclic molecular structure of these carbohydrate-based monomers, the resulting polyesters have a glass-transition temperature, mechanical strength and stiffness generally restricted. Isosorbide, along with their two less accessible 1,4:3,6-dianhydrohexitol stereoisomers isomannide and isoidide, are currently drawing a great interest in the polymer science field as bio-based monomers able to provide enhanced stiffness into the polymer chain they are incorporated. These three isohexides are composed of two fused tetrahydrofuran rings, with two secondary hydroxyl groups remaining free for reaction. Because of their fused bicyclic structure, 1,4:3,6-dianhydrohexitols are able to increase the glass-transition temperature of common aliphatic and aromatic polyesters. Nevertheless, the main shortcoming of isosorbide and its isomers is the limited reactivity of their secondary hydroxyl groups; in fact, this feature seriously hampers the polycondensation reaction in the melt, so polyesters from 1,4:3,6-diahydrohexitols obtained by this method display rather limited molecular weights. Aim and scope of this Thesis In this Thesis, the potential use of cyclic carbohydrate-based monomers to prepare polyesters is highlighted. Polyesters are extremely versatile polymers which can be used in a wide variety of applications ranging from high performance materials to recyclable and degradable polymers. The preparation of polyesters from renewable feedstock is currently receiving increasing attention in the industry and academia. For instance, bio-based succinic acid for aliphatic polyester preparation is industrially produced nowadays by anaerobic fermentation using various types of microorganisms and natural substrates. Also the bicyclic diol monomer isosorbide, which is prepared by dehydration of D-glucose coming from cereal starch, is already available at industrial scale today. This Thesis is specifically addressed to the development of aliphatic and aromatic polyesters with enhanced properties made from carbohydrate-based cyclic
Aim and outline of this Thesis 5 acetalized monomers. The main advantage of cyclic acetalized diols compared to 1,4:3,6- diahydrohexitols is that in the former case, the two hydroxyl groups free for reaction are primary and thus display enhanced reactivity compared to those of 1,4:3,6- diahydrohexitols. The aim of the Thesis is to prepare, characterize and evaluate the basic properties of linear aliphatic and aromatic polyesters made from cyclic acetalized monomers derived from carbohydrates, namely a bicyclic acetalized aldaric acid derived from D-galactose, bicyclic acetalized alditols derived from D-galactose and D-mannose, and a cyclic acetalized alditol derived from L-tartaric acid. An aldaric acid is an aldose derivative which has both the primary hydroxyl and the carbonyl groups oxidized to carboxylic groups. Aldaric acids are useful to obtain linear polyesters by reaction with diols provided that all the secondary hydroxyl groups are duly protected. In this work, the secondary hydroxyl groups are protected as methylene acetals, leading to carbohydrate-based diacids with a cyclic structure. An alditol is an aldose derivative which has the carbonyl group reduced to hydroxyl group. The protection of the secondary hydroxyl groups, leaving the two primary hydroxyl groups free for reaction, makes alditols usable as monomers for the production of linear polyesters by reaction with dicarboxylic acids. In this work, the secondary hydroxyl groups are protected as methylene acetals, leading to carbohydrate-based diols with a cyclic structure. The synthetic procedures followed for the preparation of the four carbohydratebased monomers used in this Thesis are depicted in Schemes 1.1, 1.2 and 1.3, summarized below and further detailed in Annex A. Scheme 1.1. Synthesis of dimethyl 2,3:4,5-di-O-methylene-galactarate and 2,3:4,5-di-O-methylene- galactitol from D-galactose.
Chapter 1 12
13 CHAPTER 2 POLYESTERS: TRADITIONAL PRODUCTS AND PRESENT BIO-BASED TENDENCIES Summary: Polyester is the general name given to a family of polymers which contain the ester group in their constitutional repeating unit. Since the pioneering experiments of Carothers in the 1930s, polyesters have found extensive use in a wide range of applications, including engineering plastics, high performance materials, and recyclable and degradable polymers, due to the enormous variety of structures and properties which constitute the polyester family. This introductory chapter reviews some aspects of polyesters, such as their main types and applications as well as their synthetic methods. It also describes the present bio-based tendencies regarding polyester preparation. Alternative routes from natural feedstock to produce monomers which are conventionally obtained from petrochemical sources are described; this approach entails no differences in terms of material properties between bio-based polyesters and their petrochemical counterparts since they share the same chemical structure. Finally, the use so far of carbohydrate-derived monomers to produce polyesters with modified properties is reviewed; this includes acyclic monomers as well as cyclic monomers such as 2,5- furandicarboxylic acid and 1,4:3,6-dianhydrohexitols.
Chapter 2 14 2.1. Introduction Polyesters are polymers in which the main chain is composed of aliphatic or aromatic moieties R1 and R2 linked together by ester groups (Scheme 2.1). Scheme 2.1. General formula of polyesters. The polyester family is extremely large and, depending on the nature of R1 and R2, exhibits an enormous variety of structures, properties and therefore, applications. It has been known since the early 19th century that heating carboxylic polyacids with glycerol resulted in resinous compounds. In the 1910-1920s, the General Electric Company led extensive research on the chemistry of phthalic anhydride-glycerol reaction and developed the technology of alkyd resins (glyptal resins), which are essentially polyesters of phthalic anhydride, glycerol, and monocarboxylic unsaturated fatty acids. These resins are still used in coatings, varnishes and paints. However, the modern history of polyesters began when Carothers performed experimental studies on reactions between aliphatic diacids and diols and established the relationships between degree of polymerization, conversion, functionality and gel point, that is, the base relationships of step-growth polymerization (Carothers, 1929, 1931 and 1936; Carothers and Arvin, 1929; Carothers and Dorough, 1930; Carothers et al., 1932). However, these polyesters had low melting point and were sensitive to hydrolysis, and therefore they could not compete with aliphatic polyamides (nylons), also discovered in the 1930s by Carothers at the DuPont Company. To increase the polyester melting point and to approach the thermomecanical properties obtained by nylons, it was necessary to stiffen the polyester chain by using rigid aromatic monomers instead of flexible aliphatic ones. In the early 1940s, high melting point fiber-forming polyesters from terephthalic acid and aliphatic diols were synthesized by Whinfield in the laboratories of the Calico Printers Association in the United Kingdom. After World War II the patent rights on these aromatic polyesters were shared between Imperial Chemical Industries (ICI) and DuPont, and several members of this family became, and are still today, major commercial polymers (Whinfield and Dickson, 1946). Poly(ethylene terephthalate) (PET) is now one of the most produced
Polyesters: Traditional products and present bio-based tendencies 15 polymers, primarily for textile and packaging applications. Poly(butylene terephthalate) (PBT) finds uses as solid-state molding resin. Poly(trimethylene terephthalate) (PTT), although was described in Whinfield’s original patent, is a newcomer in the commercial polyester family and has found its first applications in the textile industry. A new type of thermosetting resin, based on unsaturated polyesters, was developed at the end of the 1930s (Ellis, 1937). Unsaturated polyesters were synthesized by reacting mixtures of saturated and unsaturated diacids or anhydrides with aliphatic diols. The thermosetting resin was obtained by dissolving these polyesters in an unsaturated monomer, such as styrene, capable of undergoing free-radical copolymerization with the unsaturations in polyester chains. The liquid resin was transformed into a rigid and insoluble cross-linked polymer network after radical polymerization in the presence of heat or catalysts. Unsaturated polyester resins found their first applications in combination with glass fibers for protective radar domes during World War II, and are now one of the most important matrix resins for glass-fiber- reinforced composite materials (Fradet and Tessier, 2003). In the late 1950s and the 1960s, low-molecular weight (Mn= 1,000-3,000 g·mol-1) hydroxyl-ended aliphatic polyesters were used as macromonomers in the synthesis of polyurethane elastomers and flexible foams by reaction with diisocyanates. Although aliphatic polyesters were the first step-growth polymers fully characterized, they did not find commercial applications until then. In the 1970s, polyesterether block copolymers were commercialized by Dupont under the tradename Hytrel; these copolymers exhibited the characteristics of thermoplastic elastomers (Holden et al., 1996). During this period attention was also focused on high-performance fully aromatic polyesters; the commercial introduction of the first one, the amorphous poly(bisphenol-A isophthalate-terephthalate) (Union Carbide’s Ardel), took place in the mid 1970s (Dean et al., 1989). Despite their high cost, liquid crystalline thermotropic polyesters, such as poly(6-hydroxy-2-naphthoic acid-co-4-hydroxybenzoic acid) (Ticona’s Vectra), described at the end of the 1970s (Calundann, 1977), found and still have a number of applications in high-technology markets. In the 1990s, environmental concerns began to be gaining ground (Lichtenthaler, 2010). The versatility of the ester linkage, able to undergo hydrolysis in some conditions, makes polyesters the polymers of choice to fulfill the increasing demand for recyclable and biodegradable polymers (Albertsson and Varma, 2003). This has resulted in a renewed interest in aliphatic polyesters, such as poly(lactones), poly(lactides) or
Chapter 2 16 copolyesters containing aliphatic moieties. PET production is also strongly driven by the demand of recyclable polymers. 2.2. Types and applications of thermoplastic polyesters Polyesters are one of the most versatile classes of polymers ever produced, covering a wide range of properties and applications. As described above, polyester polyols are used in the synthesis of polyurethane elastomers and flexible foams by reaction with diisocyanates; they are also used for other applications such as coatings, paints, sealants and adhesives. Unsaturated polyesters have found extensive use to yield thermosetting polyester resins with a wide range of properties, and can be also reinforced with glass fibers or filled with large amounts of low-cost fillers. Furthermore, thermoplastic polyesters are used as fibers, engineering thermoplastics and high-performance polymers. The main properties and applications of thermoplastic polyesters which have a closer relation with this Thesis are discussed below. 2.2.1. Aliphatic polyesters The ester linkage can be easily cleaved by hydrolysis under alkaline, acid, or enzymatic catalysis. This feature makes aliphatic polyesters very attractive for those applications which require degradation of the polymers, i.e. biomedical devices and environmentally degradable polyesters. Poly(lactic acid) (PLA) is an aliphatic polyester whose monomer is produced from carbohydrates and is degraded by microorganisms to generate ultimately CO2 and H2O. This polymer can be therefore considered fully environmentally-friendly and it is an excellent representative of what is today so-called a green polymer (Scheme 2.2). PLA has found use in medical applications and it is receiving interest as a material suitable for replacing conventional commodity plastics in a wide diversity of applications (Tokiwa and Jarerat, 2004).
Polyesters: Traditional products and present bio-based tendencies 17 Scheme 2.2. Cycle of poly(lactic acid). Bioresorbable implants, for instance orthopedic fixations or sutures, are devices designed to slowly degrade in the body after implantation, so that a second surgical intervention is not required for implant removal after healing. Polyesters and copolyesters of lactic acid and glycolic acid have been used as bioresorbable sutures since the 1960s. These polyesters give no toxic or inflammatory response, can be sterilized, and are slowly and completely hydrolyzed into natural metabolic by-products that are eliminated by the body. Other aliphatic polyesters such as poly(ε-caprolactone), poly(dioxanone) and poly(trimethylene carbonate) have also been used for this application (Middleton and Tipton, 2000). Bioresorbable polymers can be processed by conventional methods such as injection-molding, compression-molding and extrusion, and sterilization is achieved by irradiation or by exposure to ethylene oxide. They are used in wound closure (sutures, staples), osteosynthetic materials (pins, screws, rods, bone plates), cardiovascular surgery (stents, grafts), and intestinal surgery (anastomosis rings) (Ikada and Tsuji, 2000). They also find application as matrix materials for implanted drug release devices or drug-containing microspheres or microcapsules (Arshady, 1991). The use of environmentally biodegradable harmless polymers, which disappear after a few weeks or months in soil, is an elegant way of dealing with solid waste disposal problem. Poly(ε-caprolactone) has been proposed as a soil-degraded container material, with applications for growing and transplanting trees as a thin-walled tree sedling container (Raghavan, 1995). In addition, its blends with starch and derivatives have been
Chapter 2 18 used in shopping bags (Darwis et al., 1998). During the last few years, a number of companies have put biodegradable polymers on the market; almost all these polymers are polyesters or copolyesters. For instance, copolyesters of poly(butylene succinate), poly(butylene adipate) and poly(ethylene succinate) have been commercialized under the trade mark of Bionolle. The potential applications of aliphatic polyesters in injectionmolded articles (cutlery, brushes), tubular films (composting bags, shopping bags), flexible packaging, food tray, cosmetic bottles and beverage bottles have been proposed, although for some of these applications the performance of the existing grades should be improved. The chemical structures of some commonly used aliphatic polyesters are depicted in Scheme 2.3. Scheme 2.3. Chemical structure of some common aliphatic polyesters. 2.2.2. Aromatic polyesters In terms of volume and economic importance, thermoplastic polyesters are dominated by poly(ethylene terephthalate) (PET); this polyester accounts for 8% of the global polymer production (Rappaport, 2010). The chemical structure of PET, together with other aromatic polyesters, is depicted in Scheme 2.4. PET has experienced an enormous development in fibers and molding resins, and has found extensive use in food packaging and bottle markets for glass replacement, where its capability of being recycled is a supplementary force driving PET consumption. Fibers are the largest end use for PET; the world production of PET fibers surpassed that of nylon fibers in the beginning of the 1970s and has reached a level close to that of cotton (Altun and Ulcay, 2004). The growth of PET fibers production has been caused by its low production costs and their ability to be blended with natural fibers such as cotton. PET resins, which
Polyesters: Traditional products and present bio-based tendencies 19 constitute the second largest application of PET, are mainly used for the packaging field, in reheat stretch blow-molded bottles for carbonated and non-carbonated soft drinks and water, and in food and cosmetics containers. PET resins are also used in sheet thermoforming applications, and as glass-fiber reinforced resin for injection-molding applications. Due to a slow crystallization rate in the unoriented state but rapid crystallization upon orientation, PET is very suitable for the biaxial film orientation process. In this process the amorphous film obtained after extrusion and quenching is successively drawn in two orthogonal directions, leading to stress-induced crystallization with biaxial orientation of polymer chains in the film plane. The resulting transparent film has approximately 50% crystallinity and exhibits excellent thermal, mechanical and electrical properties (Fradet and Tessier, 2003). PET film is the base film for some photographic films and magnetic tapes; it is also used as dielectric film in capacitors. Scheme 2.4. Chemical structure of some common aromatic polyesters. Poly(butylene terephthalate) (PBT) is mainly used as glass-fiber reinforced engineering thermoplastic, although PBT fibers can also be made. PBT crystallizes much more rapidly than PET and does not require nucleating agents for extrusion and injection molding applications; PBT crystallizes rapidly in mold at room temperature. Moreover, due to a lower melting point than PET, it can be molded at comparatively lower temperature (Gallucci and Patel, 2004). This permits fast processing and rapid production cycles. However, larger volumes and lower prices make PET a serious PBT competitor for resin applications. Poly(trimethylene terephthalate) (PTT) was already mentioned in Whinfield’s original patent that described PET and PBT (Whinfield and Dickson, 1946). However, no economically viable route to 1,3-propanediol monomer was available until the 1990s when Shell developed a route based on ethylene oxide hydroformylation (Drent, 1992) and Degussa a route based on acrolein hydration/hydrogenation (Breitkopf, 1991). PTT fibers present a combination of properties, including resilience, softness of touch and
Chapter 2 20 dyeability which makes them very attractive for the fabric, carpet and apparel markets. PTT is intended to compete for film and engineering thermoplastic applications with PET, PBT and polyamides. 2.2.3. Aliphatic-aromatic copolyesters Aliphatic polyesters can be degraded under relatively mild conditions, and are biodegradable products. However, their thermal and mechanical properties are often limited and exclude these materials from many applications. On the contrary, aromatic polyesters are distinguished by displaying an excellent pattern of physical properties; nevertheless, they are strongly resistant to hydrolysis as well as to bacterial and fungal attack, and usually remain unaltered in the environment. Trying to combine both the excellent material properties of aromatic polyesters and the potential biodegradability of the aliphatic ones, a large number of aliphatic-aromatic copolyesters have been developed during the last decades, and some products have been recently commercialized on a scale of a few thousand tons per year (Müller, 2005). A variety of random copolyesters made from a mixture of terephthalic acid and several aliphatic diacids with different diols have been prepared by melt polycondensation (Witt et al., 1994, 1996 and 1997; Atfani and Brisse, 1999; Nagata et al., 2000; Ki and Park, 2001). In the range between 40 and 50 mol% of terephthalic acid (referred to the acid components), random copolyesters combine sufficient biodegradability with adequate technical properties (Müller, 2005). Furthermore, reactive blending of homopolymers is another method to produce aliphatic-aromatic copolyesters with intermediate properties. Aromatic homopolyesters such as PET and PBT have been reacted by this technique with aliphatic homopolyesters like poly(butylene succinate), poly(butylene adipate), poly(ε-caprolactone), poly(lactic acid) and poly(glycolic acid) (Tokiwa and Suzuki, 1981; Chiellini et al., 1996; Iwamoto and Towika, 1994; Kim and Park, 1999; Kint et al., 2003). The aliphatic-aromatic copolyesters obtained by heating the homopolyesters in the molten state have a blocky chemical microstructure. Several aliphatic-aromatic copolyesters are nowadays commercialized by different companies. Ecoflex, produced by BASF (Warzelhan et al., 1996a and 1996b), and Eastar Bio, produced by Eastman (Ohta et al., 1995), are copolyesters from
Polyesters: Traditional products and present bio-based tendencies 21 terephthalic acid, adipic acid and 1,4-butanediol. These materials are suitable for biowaste bags and films for horticulture, agriculture, food packaging and household applications. Biomax (DuPont) is, according to the producer, a standard PET with addition of special monomers to allow degradation to take place, with applications in paper coatings, disposable cutlery, thermoformable cups and trays, and films (Gallagher et al., 1992a, 1992b and 1995). Ire Chemicals produces aliphatic-aromatic copolyesters made from terephthalic acid, adipic acid, succinic acid, ethylene glycol and/or 1,4-butanediol, with applications such as agricultural films and plastic bags (Chung et al., 2000, 2001a and 2001b). 2.2.4. Fully aromatic polyesters This class comprises amorphous high-Tg copolyesters, known as amorphous polyarylates, and semicrystalline polyesters that often exhibit anisotropic liquid crystalline melts. Copolyesters of bisphenol A and iso- and terephthalic acid (Scheme 2.5) are amorphous engineering thermoplastics with excellent heat and ultraviolet light resistance (Dean et al., 1989). Their amorphous morphology imparts on them properties of transparency and dimensional stability (Mahajan et al., 1996). Amorphous polyarylates find applications for injection-molded parts in automotive, electronics, safety and building equipment; for example, headlight housings, fire helmets, face shields, and transparent exterior parts such as solar energy components and glazing. Scheme 2.5. Chemical structure of random copolyesters of bisphenol A and iso- and terephthalic acid. Liquid crystalline (LC) polyesters are made from rigid para-oriented monomers such as 4-hydroxybenzoic acid, hydroquinone, terephthalic acid, or 2,6-hydroxynaphthoic acid (Calundann, 1977). They exhibit a LC mesophase in the molten state, usually a nematic one; due to high stiffness, polymer chains tend to line up parallel to each other, instead of forming random coils and isotropic melts. This structure is retained in the
Chapter 2 28 2.3.5. Enzymatic polymerization Polyesters can be also obtained by enzyme-catalyzed polymerization. Enzymes are proteins which exhibit a high catalytic efficiency and specificity for a given chemical reaction. Lipases and esterases belong to a class of enzymes referred to as hydrolyases which specifically catalyze the in vivo hydrolysis of esters. These enzymes, mainly lipases, were also found to catalyze a number of in vitro esterification and polyesterification reactions in organic medium (Kobayashi, 1999). Enzyme-catalyzed polyesterifications are carried out at 20-80 ºC in solution or in the bulk using free or immobilized enzymes, which are usually removed from reaction medium at the end of the reaction. Solvent hydrophobicity plays an important role in enzymatic activity, particulary for lipases which normally act at oil-water interfaces in living cells (Brady et al., 1990). The best reactivity is usually found for hydrophobic solvents such as hexane, toluene, diisopropylether or diphenylether. Hydrophilic polar solvents such as DMSO or methanol lead to significant modifications in enzyme conformation and therefore to a dramatic decrease in enzyme activity (Lalot et al. 1996). Polyesters have been obtained in organic medium by polyesterification of hydroxy acids (Shuai et al., 1999), hydroxy esters (Gutman et al., 1997), stoichiometric mixtures of diols and diacids (Binns et al., 1998 and 1999), diols and diesters (Mezoul et al., 1995 and 1996; Berkane et al., 1997) and cyclic anhydrides (Kobayashi and Uyama, 1993). Since lipases also catalyze the reverse reaction (i.e. hydrolysis or alcoholysis of polyester), lipase-catalyzed polyesterifications can be regarded as equilibrium polycondensations. Various methods have been employed to remove the reaction by-product, water or alcohol, in order to displace the equilibrium toward the formation of polyester: addition of molecular sieves, Dean-Stark distillation of toluene-water azeotrope, or bubbling inert gas in reaction medium. In the case of reactions carried out without solvent, vacuum can be applied to distill off the byproduct, i.e. water or alcohol. The use of vinyl esters is a convenient way to displace esterification equilibrium, since the theoretical by-product, vinyl alcohol, tautomerizes into acetaldehyde, which cannot participate in the reverse reaction (Uyama et al., 1999). Activated esters such as 2,2,2-trifuoroethyl, 2-chloroethyl and 2,2,2-trichloroethyl esters have been reported to yield high-molecular weight polyesters (Linko et al., 1995; Wang et al., 1996). The enzymatic synthesis of polyesters is mainly used for the preparation of soluble or low-melting point polyesters, mainly aliphatic ones. Although the synthesis of some aromatic polyesters of isophthalic acid has also been reported, lipases exhibit a very low catalytic activity for rigid aromatic monomers such as terepthalates or diphenols (Wu et al., 1998; Rodney et al., 1999).
Polyesters: Traditional products and present bio-based tendencies 29 2.4. Bio-based polyesters 2.4.1. Polyesters partially or totally produced by bacteria 2.4.1.1. Polylactic acid (PLA) Polylactic acid (PLA) is an aliphatic polyester used in biomedical applications such as sutures, stents, dental implants, vascular grafts, bone screws and pins (Albertsson and Varma, 2003). It has also been investigated as a vector for drug delivery, for example in the long-term delivery of antimicrobial drugs, contraceptives and prostate cancer treatments, and as a scaffold material to support cell and tissue growth (Williams, 2007). Moreover it is also used for food packaging and bottle packaging of milk, water and juices. For the PLA industry, lactic acid (LA) with high optical purity of over 98-99% of L-LA is required; D-LA is not desirable for food, drink and biomedical applications due to the metabolic problems that D-LA may cause. The chemical structures of L-LA and D-LA are depicted in Scheme 2.8. Isotactic poly(L-lactic acid) (PLLA) is a semicrystalline polymer with a glass-transition temperature near 60 ºC and a melting temperature near 180 ºC (Tang et al., 2004). Scheme 2.8. Chemical structure of L- and D-lactic acid. Lactic acid can be produced via either chemical synthesis or microbial fermentation. However, the chemical processes produce a racemic (50:50) mixture of D- LA and L-LA (D,L-LA). Glucose is transformed to pyruvic acid through the glycolysis pathway. Many bacteria contain an enzyme called lactate deshydrogenase (LDH) which converts pyruvic acid to lactic acid. Depending on the bacterial species and its LDH specificity, the lactic acid fermentation process can produce very pure D-LA or L-LA or a mixture of them. Molecular biology tools have been employed to delete the D-LDH genes to enhance the optical purity of its L-LA synthesis (Jem et al., 2010). Under high temperature (> 100 ºC), D-LA or L-LA may be converted to each other through a racemization process, which eventually results in a racemic mixture with inferior optical
Chapter 2 30 properties (Benninga, 1990). Therefore, processing at elevated temperature should be avoided. Polymerization of lactic acid can be conducted either by direct polycondensation of LA, or by a ring-opening polymerization of the lactide, a cyclic dimer of LA. The direct polycondensation process dehydrates LA to form oligomers which are further polymerized to PLA under simultaneous dehydration to avoid the degradation of polymer in the presence of moisture. Removal of water generated by the LA condensation is usually very difficult during the final stage of polymerization as the diffusion of moisture in a highly viscous polymeric melt is very slow. The residual water trapped will reduce PLA molecular weights. The direct polymerization process is reported to be used only by Tonji and Mitsui Chemicals companies (Chen and Patel, 2012). Most industrial PLA production processes employ the conversion of lactides to PLA via a ring-opening polymerization (ROP) catalyzed by organometallic catalysts (Du et al., 2007 and Pang et al., 2008). During the process, LA is dehydrated and condensed into its oligomers at high temperature under vacuum to remove moisture. Subsequently lactide is obtained from catalytic depolymerization of these short polylactic acid chains under reduced pressure. Residual LA is removed from lactide via distillation or crystallization. The purified lactide is polymerized into PLA at temperatures above the melting point of lactide and below the degradation temperatures of PLA. 2.4.1.2. Polyhydroxyalkanoates (PHAs) Whereas PLA production is a two-stage process (fermentation to monomer followed by a conventional polymerization step), polyhydroxyalkanoates (PHAs) are the only bioplastics completely synthesized by microorganisms. PHA can be synthesized by over 30% of soil inhabiting bacteria (Chen and Patel, 2012). Scheme 2.9. Chemical structure of polyhydroxyalkanoates.
Polyesters: Traditional products and present bio-based tendencies 31 The generic chemical structure for PHAs is depicted in Scheme 2.9, where x is 1 for all commercially relevant polymers and R can be either hydrogen or hydrocarbon chains. Poly(3-hydroxybutyrate) (PHB), with R=methyl, poly(3-hydroxyvalerate) (PHV), with R=ethyl, and poly(3-hydroxyhexanoate) (PHH), with R=propyl, are some of the most widely known PHAs. The bacteria can be fed a range of different carbon sources; for instance, E. coli fed with a range of oils produces different compositions of poly(HB-co- HH), whereas R. eutropha fed with a combination of glucose and propionate produces poly(HB-co-HV). Homopolymers, random copolymers and block copolymers of PHAs can be produced depending on the bacterial species and growth conditions (Gao, 2011). PHAs with Tg ranging from -50 to 4 ºC, and Tm between 60 and 170 ºC have been developed. Such diversity has allowed the development of a range of applications including environmentally friendly biodegradable plastics for packaging purposes, fibers, biodegradable and biocompatible implants and controlled drug release carriers (Wu et al., 2000). 2.4.2. Alternative routes from renewable sources to conventional monomers In this section, alternative routes from renewable sources to produce monomers which are currently obtained from petrochemical feedstock are described. The use of these bio-based monomers does not entail differences in terms of material properties between partially bio-based polyesters and their petrochemical counterparts, since they have exactly the same chemical structure. Succinic acid. Bio-based succinic acid may be produced by anaerobic fermentation using various types of microorganisms. For example, succinic acid can be produced from glucose by the rumen organism Actinobacillus succinogenes; and can also be produced by Anarobiospirillum succiniproducens using glucose or even lactose, sucrose, maltose or fructose as carbon sources (Ryu et al., 1999; Lee et al., 2003). Feedstocks including corn starch, corn steep liquor, whey, cane molasses, glycerol, lignocelluloses, cereals, and straw hydrolases could be utilized for succinic acid production (Xu and Guo, 2010). The French starch and starch derivative producer Roquette and the Dutch chemical company DSM have built a demonstration plant in France for production of several hundred tons of succinic acid per year from starch using fermentation technology. Also BASF and the Dutch company Purac are jointly pursuing succinic acid production.
Chapter 2 32 Mitsubishi Japan has also attempted to industrialize succinic acid microbial production (Chen and Patel, 2012). Adipic acid. The bio-based adipic acid is still in its nascent state and it is not commercially available nowadays. However, companies as Verdezyne and Rennovia are developing bio-based routes to produce adipic acid. The bio-based process carried out by Verdezyne uses genetically modified enzymes to ferment glucose to adipic acid. On the other hand, Rennovia uses air oxidation to convert glucose to glucaric acid, followed by hydrodeoxygenation to convert glucaric acid to adipic acid (IHS Chemical Process Economics Program, 2012). Terephthalic acid. Terephthalic acid has been reported as potentially being made from bio-based p-xylene produced by depolymerization of lignin (Bozell et al., 2007). Recently, the production of terephthalic acid from limonene has been patented. In this case biobased terephthalic acid is produced by the hydrogenation of limonene to p-cymene using zeolites and the subsequent oxidation of this compound (Berti et al., 2010). This technology needs, however, further development and does not offer an economically viable production route for the short to midterm, mainly due to the difficult accessibility of limonene. Ethylene glycol. Microbial ethanol for application as biofuel is being produced at the multimillion ton scale, from starch and from sugar. Bioethanol can be catalytically dehydrated to ethylene with activated clay, phosphoric acid, sulfuric acid, activated alumina, metal oxides and zeolites as catalysts (Huang et al., 2008). Using bio-based ethylene, bio-based ethylene glycol can be produced via the conventional route of direct oxidation to ethylene oxide followed by thermal hydrolysis (Rebsdat and Mayer, 2000). The large beverage company Coca-Cola Co. introduced in 2009 the PlantBottle as the first recyclable PET bottle using up to 30% bio-based monomers; bio-based ethylene glycol and recycled PET are employed in PET production. PepsiCo developed the first 100% bio-based bottle in 2011 (PepsiCo, 2011). It has also been announced a new generation of 100% bio-based PlantBottle from Coca-Cola Co. (Coca-Cola Co., 2011). 1,3-Propanediol. The production of bio-based 1,3-propanediol has been developed and commercialized by the joint venture DuPont Tate & Lyle LLC, with an annual capacity of 45,000 tons. Bio-based 1,3-propanediol is produced via the fermentation of glucose. There is a fermentation pathway in nature which consists of two steps: naturally occurring yeasts first ferment glucose to glycerol and then microbes ferment glycerol to 1,3-
Polyesters: Traditional products and present bio-based tendencies 33 propanediol. In the patented industrial bioprocess, glucose derived from wet-milled corn is metabolized by genetically engineered microorganism E. coli; this microorganism converts glucose to 1,3-propanediol in a single step (DuPont, 2013). 1,4-Butanediol. Genomatica Inc. developed a sucrose-based process for the manufacture of 1,4-butanediol by an engineered microorganism (Chen and Patel, 2012). On the other hand, 1,4-butanediol can also be produced from bio-based succinic acid involving three steps: corn-derived glucose is fermented to succinic acid, succinic acid is then purified by electrodialysis and it is finally reduced catalytically to 1,4-butanediol. Recently, an aqueous-phase hydrogenation of biomass-based succinic acid to 1,4- butanediol over supported bimetallic catalysts was reported (Minh et al., 2010). However, most 1,4-butanediol is still produced from chemical processes. 1,10-Decanediol. By reacting castor oil with NaOH or KOH at elevated temperatures and in the presence of catalysts, sebacic acid and 2-octanol are formed preferably. Then sebacic acid can be converted to 1,10-decanediol via hydrogenation. Thus, 1,10- decanediol has been reported as a potential bio-based derivative of the castor oil (Mutlu and Meier, 2010). 2.4.3. Polyesters from carbohydrate-derived monomers with modified properties 2.4.3.1. Unprotected alditols and aldaric acids Linear polycondensation of monomers derived from carbohydrates is not straightforward. Carbohydrate-based compounds usually possess an excess of functional groups that upon polycondensation would lead to undesirable cross-linking reactions unless special precautions are taken. Nevertheless, some linear polycondensates have been synthesized using sugar-derived monomers bearing free hydroxyl groups by a solution polycondensation approach (Kiely et al., 1994 and 2000). Lipases have been reported to provide regioselectivity during esterification reactions of alditols at mild temperatures. In 2000, the first example of an alditol used as monomer in enzyme-catalyzed polymerization was described; D-glucitol was polymerized through its positions C-1 and C-6 with divinyl sebacate in CH3CN (Uyama et al., 2000). The preparation of polyesters from 1,8-octanediol, adipic acid and several unprotected alditols, such as erythritol, xylitol, ribitol, D-mannitol, D-glucitol and galactitol, has also
Chapter 2 34 been reported. Polymerizations were carried out under vacuum, using the immobilized Lipase B from Candida antarctica (Novozyme 435) as catalyst (Hu et al., 2006). 2.4.3.2. Acyclic O-protected alditols and aldaric acids Although some syntheses leading to linear polyesters have been realized using sugar-derived monomers with free hydroxyl groups (see above), most of them are carried out with derivatives in which the secondary hydroxyl groups have been appropriately blocked (Scheme 2.10). Scheme 2.10. Modification of carbohydrates leading to O-protected alditols and aldaric acids. The pioneering work of Drew and Haworth in the 1920s described the polycondensation of 2,3,4-tri-O-methyl-L-arabinonic acid to give an oligomeric material not completely characterized (Drew and Haworth, 1927). Also in the field of aliphatic polyesters from acyclic O-protected carbohydrate-based monomers, several poly(alkylene dicarboxylates) have been recently obtained by polycondensation reaction of the alditols 2,3,4-tri-O-methyl-L-arabinitol and 2,3,4-tri-O-methyl-xylitol, and aldaric acids 2,3,4-tri-O-methyl-L-arabinaric acid and 2,3,4-tri-O-methyl-xylaric acid (García- Martín et al., 2006). The chemical structures of these O-protected carbohydrate-based monomers are depicted in Scheme 2.11. 1,4-Butanediol and adipic acid were also used as comonomers.
Polyesters: Traditional products and present bio-based tendencies 35 Scheme 2.11. Chemical structure of some monomers derived from L-arabinose and D-xylose. The synthesis of aromatic polyesters, specifically of poly(phthalate)s, using carbohydrate-based acyclic O-protected monomers has also been explored. Thus poly(ethylene terephthalate) (PET), poly(ethylene isophthalate) (PEI) and poly(butylene terephthalate) (PBT) have been chemically modified by insertion of different O-methyl alditols (Galbis and García-Martín, 2008 and 2010). Firstly, a series of PET copolyesters containing 2,3-di-O-methyl-L-threitol were prepared and duly characterized (Kint et al., 2001). Later PET and PEI analogues with total replacement of the ethylene glycol units by 2,3,4,5-tetra-O-methyl-hexitols having either D-manno or galacto configurations were synthesized (Zamora et al., 2005a). Polyesters analogous to PET, PEI and PBT have also been prepared by using 2,3,4-tri-O-methyl-L-arabinitol and 2,3,4-tri-O-methyl-xylitol (Zamora et al., 2005b; Alla et al., 2006). The 2,3,4-tri-O-benzyl ethers of L-arabinitol and xylitol were also used to synthesize polyesters analogous to PET and PBT (Zamora et al., 2008). Butylene copolyesters based on O-methyl aldaric acids as L-arabinaric and galactaric acids, and terephthalic acid were also prepared and characterized (Zamora et al., 2009). The chemical structures of some of these O-protected carbohydrate-based monomers are depicted in Scheme 2.12. Scheme 2.12. Chemical structure of some monomers derived from D-galactose and D-mannose.
Chapter 2 36 The hydrolytic degradation of a series of homo- and copolyesters analogous to PET and PEI based on O-methyl protected L-arabinitol, xylitol, D-mannitol and galactitol was relatively fast at temperatures 10 ºC above their respective Tg. The hydrolysis of copolyesters took place preferentially by cleavage of the ester groups of the sugar units (Zamora et al., 2006). The hydrolytic degradation of PBT copolyesters containing up to 50% of O-methyl protected pentitols derived from L-arabinose and D-xylose was also studied (Alla et al., 2006). It was shown that these copolyesters were much more sensitive to hydrolysis than was PBT. 2.4.3.3. 2,5-Furandicarboxylic acid (FDCA) The research strategy related to furan polymers relies on the use of two furan derivatives, i.e. furfural (F) and hydroxymethylfuraldehyde (HMF), as starting building blocks from which two sets of monomers can be synthesized: those containing double bonds suitable for chain polymerizations and copolymerizations, and those with carboxyl or hydroxyl functionalities associated with step-growth mechanisms (Gandini and Belgacem, 1997; Gandini, 2010 and 2011). Furfural (F) is produced as an industrial commodity, with a yearly production close to 300,000 tons. Its precursors are pentoses present in corn, rice, sugar and cotton, among others. Furfural production is based on the acid catalyzed hydrolytic depolymerization of C5 hemicelluloses, particulary xylanes. Most of the furfural is today converted into furfuryl alcohol (FA), which is extensively employed as a precursor to a variety of resins for several applications. The second furan derivative which can be prepared from the appropriate C6 polysaccharides is hydroxymethylfuraldehyde (HMF). The chemical structures of these furan-based products are depicted in Scheme 2.13. The acid-catalyzed dehydration of fructose to HMF can be perfomed in high-boiling polar solvents, in supercritical acetone and water, or by addition of phase modifiers (Moreau et al., 2004; Román-Leshkov et al., 2006; Zhao et al., 2007).
Polyesters: Traditional products and present bio-based tendencies 37 Scheme 2.13. Chemical structure of furfural (F), furfuryl alcohol (FA), hydroxymethylfuraldehyde (HMF) and 2,5-furandicarboxylic acid (FDCA). Especially 2,5-furandicarboxylic acid (FDCA), as a HMF derivative, has gained particular attention as a suitable monomer for step-growth polymerization. Moore and Kelly initially explored the scope and limitations of FDCA in polycondensation reactions and synthesized a series of polyesters based on FDCA and several diols (Moore and Kelly, 1978 and 1979). However, due to the limited availability of high-purity FDCA, the research interest in this field declined in the following years. Recent developments in the catalytic routes toward high-purity FDCA allowed the preparation of larger amounts of this compound, and led to a significant increase in the amount of research carried out towards FDCA polycondensates. The use of this compound to obtain bio-based polyesters has been extensively studied in the last years (Gandini et al., 2009a and 2009b; Jiang et al., 2011; Wu et al., 2012a; Gubbels et al., 2013b). The general chemical structure of the furan-based polyesters obtained upon polycondensation of FDCA with linear aliphatic diols is depicted in Scheme 2.14, and the thermal properties of this family of compounds are shown in Table 2.1. Scheme 2.14. Chemical structure of furan-based polyesters obtained upon polycondensation of FDCA with linear aliphatic diols. FDCA was used in conjunction with ethylene glycol to synthesize poly(ethylene 2,5-furandicarboxylate) (PEF), i.e. the furan analog of PET. The replacement of the terephthalate ring by the furan counterpart led to a decrease in melting temperature compared to PET, whereas maintained its glass-transition temperature (Gomes et al.,
Chapter 2 44 Storbeck and Ballauff performed polycondensation of terephthaloyl chloride with isosorbide and ethylene glycol. Intrinsic viscosities of up to 0.46 dL·g-1 were obtained. They studied the entire range of diol proportions, and found that isosorbide-based PET copolyesters were amorphous above 20 mol% of isosorbide (Storbeck and Ballauff, 1996). The classical industrial melt polycondensation from terephthalic acid or dimethyl terepthalate were covered by patent (Charbonneau et al., 1999). The content of isosorbide in the feed was not totally incorporated into the polyester chains; significant amounts of isosorbide were lost by distillation during the esterification or during the transesterification step. Fenouillot et al. performed also studies on melt polycondensation of such copolyesters to quantify the actual proportion of isosorbide inserted into the polyester chains, and found that about 25% of isosorbide was lost (Fenouillot et al., 2010). Kricheldorf et al. studied polycondensation of 1,4-butanediol with isosorbide and either dimethyl terepthalate in bulk using TBT as catalyst, or terephthaloyl chloride in solution using pyridine as catalyst and HCl acceptor (Kricheldorf et al., 2007). Copolyesters with very low molecular weights, i.e. inherent viscosities between 0.08 and 0.37 dL·g-1, were obtained by transesterification in the melt. However higher inherent viscosities, with values of up to 0.74 dL·g-1, could be attained by using polycondensation in solution with terepthaloyl chloride. Isosorbide-based PBT copolyesters were found to be amorphous for isosorbide contents above 32 mol%. The industrial applications of copolyesters containing low to moderate proportions of isosorbide are that of classical of PET or PBT taking advantage of the increased glass-transition temperature. Numerous patents described isosorbide-based PET and PBT copolyesters; some of them claimed the method of polymerization while several claimed specific polyester applications. For instance, the addition of isosorbide into PET to produce bottles able to resist hot-fill process (Charbonneau and Johnson, 1999) and thermoformable sheets where a low rate of crystallization is required (Khanarian et al., 2000) have been patented. Also fibers, films and polymer blends were cited (Khanarian et al., 2002). Furthermore, the incorporation of isosorbide by transesterifying poly(ethylene terephthalate) with poly(isosorbide terephthalate) oligomers in bulk at 285 ºC was also patented (Adelman et al., 2003). In this way they could incorporate 24 mol% of isosorbide relative to the total diol content, and the polymer obtained was amorphous with Tg= 139 ºC and an inherent viscosity of 0.57 dL·g-1. In this case the aim was not to compete with PET but rather with polycarbonate (PC) or
Polyesters: Traditional products and present bio-based tendencies 45 poly(methyl methacrylate) (PMMA). At this point it should be stated that all aromatic copolyesters containing isosorbide are yellowish, an important flaw for certain applications. The sensitivity of isosorbide to thermooxidation at the high temperatures needed to polymerize polyesters is the cause of yellowing. The efforts for solving this problem were highlighted in a recent patent from Charbonneau on poly(ethylene-co- isosorbide terepthalate); the synthetic routes cited were the typical PTA or DMT synthesis but with the observation that oxygen should absolutely be avoided in the reactive medium. Moreover, two antioxidants were added just before polycondensation: the primary one was a hindered phenol and the secondary one was a trivalent phosphorous compound (Charbonneau, 2006). Recently, Sablong et al. studied the incorporation of isosorbide into PBT by solid-state modification (SSM). Experiments using SSM for the direct incorporation of isosorbide into PBT were unsuccessful; isosorbide could not be incorporated under such conditions, due to the relatively low reactivity of its two secondary hydroxyl. The synthesis of a macrodiol bearing two primary hydroxyl groups, from isosorbide, terepthaloyl chloride and 1,4-butanediol was necessary to incorporate isosorbide into the PBT chain. Very high molecular weights (Mn= 80,000-140,000 g·mol-1) could be attained by using this technique (Sablong et al., 2008). Isosorbide and isoidide have also been polymerized with the dichloride derivative of 2,5-furandicarboxylic acid in solution. Amorphous polyesters with high glasstransition temperatures, i.e. 180 ºC for poly(isosorbide 2,5-furandicarboxylate) (Mn= 13,700 g·mol-1) and 140 ºC for poly(isoidide 2,5-furandicarboxylate) (Mn= 5,700 g·mol-1), were obtained (Gomes et al., 2011). The low reactivity of the secondary hydroxyl groups of the 1,4:3,6- dianhydrohexitols explains why the synthesis of more reactive derivatives such as amines (Thiem and Lüders, 1986; Bachmann et al., 1998; Jasinska et al., 2011) or isocyanates (Pfeffer et al., 2011a and 2011b), has been considered. Regarding polyester synthesis from isohexide-derived monomers with enhanced reactivity, another strategy is based on chain extension at C2 and C5 of the isohexide skeleton. Very recently the synthesis of a 1-carbon extended isohexide derivative with dicarboxylic functionality, namely isoidide dicarboxylic acid (IIDCA), and a 1-carbon extended diol with primary hydroxyl groups, namely isoidide hydroxymethylene diol (IIDML), has been reported (Wu et al., 2011). Their chemical structures are depicted in Scheme 2.19.
Chapter 2 46 Scheme 2.19. Chemical structure of the 1-carbon extended isohexide derivatives IIDCA and IIDML. The main properties of polyesters based on these 1-carbon extended isohexide derivatives are presented in Table 2.3. Aliphatic polyesters with Mw in the 13,000-34,000 g·mol-1 range were obtained by melt polycondensation from the dimethyl ester of IIDCA and aliphatic diols with 2 to 12 methylene groups (Wu et al., 2012b). These aliphatic polyesters were semicrystalline, with Tm values in the 40-125 ºC range. The glasstransition temperatures were between -33 ºC and 18 ºC. The synthesis of a series of fully isohexide-based polyesters has been recently reported (Wu et al., 2013). The combination of the dimethyl ester of the IIDCA with isosorbide, isomannide or isoidide resulted in semicrystalline low molecular weight polyesters (Mw= 1,700-3,600 g·mol-1) with Tm in the 80-170 ºC range. Tg values were 73, 30 and 85 ºC for polyesters obtained from the dimethyl ester of IIDCA and isosorbide, isomannide or isoidide, respectively. Also a fully isohexide-based polyester from the dimethyl ester of IIDCA and the extended diol IIDML was prepared using the same melt polycondensation conditions. An increase in the reactivity of the extended diol bearing two primary hydroxyl groups with respect to the secondary hydroxyl groups of isoidide was reported; the Mw of this IIDCA-IIDML polyester being 10,400 g·mol-1. This polyester was semicrystalline and its glass-transition temperature was 48 ºC (Wu et al., 2013). However the 1-carbon extension resulted in a significant decrease in Tg of the resulting polyesters compared to incorporation of the parent isohexide; the IIDCA-isoidide polyester showed a Tg value of 85 ºC depite having lower molecular weight. When compared to the polyester from terephthalic acid and isoidide, the analogous polyester containing the extended isodide diol IIDML also showed a dramatic drop in Tg of approx. 100 ºC (209 ºC vs. 105 ºC) (Wu, 2012; van Es, 2013).
Polyesters: Traditional products and present bio-based tendencies 47 Table 2.3. Thermal properties and molecular weights of polyesters from monomers based on chain extension at C2 and C5 of the isohexide skeleton. Thermal properties Molecular weights Diol Diacid Tg (ºC) Tm (ºC) Mn (g·mol-1) Mw (g·mol-1) Ref. ethylene glycol IIDCA 18 125 6,100 13,100 Wu et al., 2012b 1,4-butanediol IIDCA 0 47/62a 14,100 34,100 Wu et al., 2012b 1,6-hexanediol IIDCA -20 54 10,200 22,800 Wu et al., 2012b 1,8-octanediol IIDCA -30 43/51/55a 10,000 17,100 Wu et al., 2012b 1,10-decanediol IIDCA -33 43/54a 11,800 24,600 Wu et al., 2012b 1,12- dodecanediol IIDCA -30 57/64a 8,800 18,000 Wu et al., 2012b isosorbide IIDCA 73 125/168a 2,600 3,300 Wu et al., 2013 isomannide IIDCA 30 89 1,200 1,700 Wu et al., 2013 isoidide IIDCA 85 109/143a 2,500 3,600 Wu et al., 2013 IIDML IIDCA 48 118/150a 5,400 10,400 Wu et al., 2013 IIDML terephthalic 105 268/277/ 296a 7,700 17,700 Wu, 2012 a Multiple melting peak. 2.4.3.5. Cyclic acetalized alditols, aldonic and aldaric acids The use of cyclic acetalized alditols, aldonic and aldaric acid derivatives as building blocks for the synthesis of polycondensates is not new, but very few cases have been reported so far: The synthesis and crystal structure of linear polyamides from 2,3-di-O-methylene-L- tartaric acid and several diamines have been described (Rodríguez-Galán et al., 1992; Iribarren et al., 2000), and their biodegradability has been tested (Kimura et
Chapter 2 48 al., 1999). The incorporation of the dimethyl ester derivative into polyesters has been recently reported (Japu et al., 2013a). Low molecular weight polyesters made from 2,3-di-O-isopropylidene-L-tartaric acid and linear alkanediols and from 2,3-di-O-isopropylidene-L-threitol and aliphatic diacids have been reported (Dhamaniya and Jacob, 2010 and 2012). Polyurethanes have been synthesized from 2,3-di-O-isopropylidene-L-threitol and from 2,3-di-O-methylene-L-threitol (Marín and Muñoz-Guerra, 2008). In the 1950s, Mehltretter and Mellies reported on the polymerization of 2,4:3,5-di-O- methylene-D-gluconic acid in an attempt to benzoylate this compound with an equivalent amount of benzoyl chloride or benzoic anhydride in pyridine solution (Mehltretter and Mellies, 1955). Polyurethanes from 1-amino-1-deoxy-2,3:4,5-di-O-isopropylidene-D-galactitol have been prepared (Gómez and Varela, 2009). 2,4:3,5-di-O-methylene-D-glucitol has been used as monomer for the preparation of polyurethanes (Marín and Muñoz-Guerra, 2009), and its incorporation into polyesters is currently being studied in another PhD thesis in our research group (Japu et al., 2012 and 2013b). The chemical structure of these cyclic acetalized alditols, aldonic and aldaric acids are depicted in Scheme 2.20. Scheme 2.20. Chemical structure of cyclic acetalized alditols, aldonic and aldaric acids being studied so far. 2,3-di-O-methylene-L-threitol 2,3-di-O-methylene-L-tartaric acid 2,3-di-O-isopropylidene-L-tartaric acid 2,3-di-O-isopropylidene-L-threitol HO O O O O OH 2,4:3,5-di-O-methylene-D-glucitol HO O O O O NH2 1-amino-1-deoxy- 2,3:4,5-di-O-isopropylidene-D-galactitol HO O OH O HO O OH O O O HO O OH O HO O OH O O O HO O O O O OH 2,4:3,5-di-O-methylene-D-gluconic acid O
Polyesters: Traditional products and present bio-based tendencies 49 2.5. References Adelman, D.J.; Charbonneau, L.F.; Ung, S. US Patent 6656577 (to E.I. du Pont de Nemours and Co.), 2003. Albertsson, A.-C.; Varma, I.K. Biomacromolecules 2003, 4, 1466-1486. Alla, A.; Hakkou, K.; Zamora, F.; Martínez de Ilarduya, A.; Galbis, J.A.; Muñoz-Guerra, S. Macromolecules 2006, 39, 1410-1416. Altun, S.; Ulcay, Y. J. Polym. Environ. 2004, 12, 231-237. Arshady, R. J. Control. Release 1991, 17, 1-21. Atfani, M.; Brisse, F. Macromolecules 1999, 32, 7741-7752. Bachmann, F.; Reimer, J.; Ruppenstein, M.; Thiem, J. Macromol. Rapid Commun. 1998, 19, 21−26. Benninga, H.A. History of Lacid Acid Making: A Chapter in the History of Biotechnology; Kluwer: Dordrecht, Boston and London, 1990. Berkane, C.; Mezoul, G.; Lalot, T.; Brigodiot, M.; Maréchal, E. Macromolecules 1997, 30, 7729-7734. Berti, C.; Binassi, E.; Colonna, M.; Fiorini, M.; Kannan, G.; Karanam, S.; Mazzacurati, M.; Odeh, I.; Vannini, M. US Pat. Appl. Publ. 20100168371 (to SABIC-IP), 2010. Binns, F.; Harffey, P.; Roberts, S.M.; Taylor, A. J. Polym. Sci., Polym. Chem., 1998, 36, 2069-2079. Binns, F.; Harffey, P.; Roberts, S.M.; Taylor, A. J. Chem. Soc., Perkin Trans. 1, 1999, 19, 2671-2676. Bozell, J.J.; Holladay, J.E.; Johnson, D.; White, J.F. Top Value Added Chemicals from Biomass - Volume II: Results of Screening for Potential Candidates from Biorefinery Lignin. Report PNNL-16983., Pacific Northwest National Laboratory (PNNL) and National Renewable Energy Laboratory (NREL), 2007, http://www.pnl.gov/main/publications/external/technical_reports/PNNL-16983.pdf [Accessed on May 2, 2013] Brady, L.; Brzozowski, A.M.; Derewenda, Z.S.; Dodson, E.; Dodson, G.; Tolley, S.; Turkenburg, J.P.; Christiansen, L.; Huge-Jensen, B.; Norskov, L.; Thim, L.; Menge, U. Nature 1990, 343, 767-770. Braun, D.; Bergmann, M. J. Prakt. Chem. 1992, 334, 298-310. Breitkopf, N.; Dämbkes, G.; Bach, H. US Patent 5008473 (to Ruhrchemie Aktiengesellschaft), 1991. Buck, K.W.; Duxbury, J.M.; Foster, A.B.; Perry, A.R.; Webber, J.M. Carbohydr. Res. 1966, 2, 122-131.
Chapter 2 50 Calundann, G.W. US Patent 4161470 (to Celanese Corp.), 1977. Carothers, W.H. J. Am. Chem. Soc. 1929, 51, 2548-2559. (1929a) Carothers, W.H.; Arvin, J.A. J. Am. Chem. Soc. 1929, 51, 2560-2570. (1929b) Carothers, W.H.; Dorough, G.L. J. Am. Chem. Soc. 1930, 52, 711-721. Carothers, W.H. Chem. Rev. 1931, 8, 353-426. Carothers, W.H.; Dorough, G.L.; van Natta, F.J. J. Am. Chem. Soc. 1932, 54, 761-772. Carothers, W.H.; Trans. Faraday. Soc. 1936, 32, 39-53. Chang, T. Polym. Eng. Sci. 1970, 10, 364-368. Chang, S.; Sheu, M.; Chen, S. J. Appl. Polym. Sci. 1983, 28, 3289-3300. Charbonneau, L.F.; Johnson, R.E.; Witteler, H.B.; Khanarian, G. US Patent 5959066 (to HNA Holdings, Inc.), 1999. Charbonneau, L.F.; Johnson, R.E. WO Patent 9954533 (to HNA Holdings, Inc.), 1999. Charbonneau, L.F. WO Patent 2006032022 (to E.I. du Pont de Nemours and Co.), 2006. Chen, G.Q.; Patel, M.K. Chem. Rev. 2012, 112, 2082-2089. Chiellini, E.; Corti, A.; Giovannini, A.; Narducci, P.; Paparella, M.; Solaro, R. J. Environ. Polym. Degrad. 1996, 4, 37-50. Chung, H.S.; Lee, J.W.; Kim, D.H.; Jun, J.N.; Lee, S.W. WO Patent 0011063 (to Ire Chemical Ltd.), 2000. Chung, H.S.; Lee, J.W.; Kim, D.H.; Jun, J.N.; Lee, S.W. EP Patent 1106640A2 (to Ire Chemical Ltd.), 2001. (2001a) Chung, H.S.; Lee, J.W.; Kim, D.H.; Jun, J.N.; Lee, S.W. EP Patent 1108737A2 (to Ire Chemical Ltd.), 2001. (2001b) Coca-Cola Co., 2011. http://www.coca-colacompany.com/stories/plant-bottle-basics [Accessed on May 2, 2013] Cope, A.C.; Shen, T.Y. J. Am. Chem. Soc. 1956, 78, 3177-3182. Darwis, D.; Mitomo, H.; Enjoji, T.; Yoshii, F.; Makuuchi, K. Polym. Degrad. Stabil. 1998, 62, 259-265. Dean, B.D.; Matzner, M.; Tibbitt, J.M. The Synthesis, Characterization, Reactions and Applications of Polymers. In: Allen, G.; Bevington, J.C.; Eds. Comprehensive Polymer Science, Vol 5; Pergamon: Oxford, 1989, p. 317. Dhamaniya, S.; Jacob, J. Polymer 2010, 51, 5392-5399. Dhamaniya, S.; Jacob, J. Polym. Bull. 2012, 68, 1287-1304.
Polyesters: Traditional products and present bio-based tendencies 51 Drent, E. Eur. Patent 478850A1 (to Shell Oil Company), 1992. Drew, H.D.K.; Haworth, W.N. J. Chem. Soc. 1927, 775-779. Du, H.Z.; Pang, X.; Yu, H.Y.; Zhuang, X.L.; Chen, X.S.; Cui, D.M.; Wang, X.H.; Jing, X.B. Macromolecules 2007, 40, 1904-1913. Duh, B. J. Appl. Polym. Sci. 2001, 81, 1748-1761. Duh, B. Polymer 2002, 43, 3147-3154. DuPont. DuPont Renewable Sourced Materials website, 2013 http://www2.dupont.com/Renewably_Sourced_Materials/en_US/procbuildingblocks.html [Accessed on May 2, 2013] Ellis, C. US Patent 2225313 (to Ellis-Foster Co.), 1937. Fakirov, S.; Fakirov, C.; Fischer, E.W.; Stamm, M. Polymer 1991, 32, 1773-1180. Fenouillot, F.; Rosseau, A.; Colomines, G.; Saint-Loup, R.; Pascault, J.-P. Prog. Polym. Sci. 2010, 35, 578-622. Fortunato, B.; Pilati, F.; Manaresi, P. Polymer 1981, 22, 655-657. Fradet, A.; Maréchal, E. Adv. Polym. Sci. 1982, 43, 51-142. Fradet, A.; Tessier, M. Polyesters. In: Rogers, M.E.; Long, T.E.; Eds. Synthetic Methods in Step-Growth Polymers; John Wiley & Sons: Hoboken, 2003, pp. 17-134. Galbis, J.A.; García-Martín, M.G. Sugars as Monomers. In: Belgacem, M.N.; Gandini, A.; Eds. Monomers, Polymers and Composites from Renewable Resources; Elselvier: Oxford, 2008, 89-114. Galbis, J.A.; García-Martín, M.G. Chem. Curr. Top. 2010, 295, 147-176. Gallagher, F.G.; Hamilton, C.J.; Hansen, S.M.; Shin, H.; Huynkook; Tietz, R.F. US Patent 5171308 (to DuPont De Nemours and Co.), 1992. (1992a) Gallagher, F.G.; Hamilton, C.J.; Hansen, S.M.; Shin, H.; Huynkook; Tietz, R.F. US Patent 5171309 (to DuPont De Nemours and Co.), 1992. (1992b) Gallagher, F.G.; Gray, D.S.; Hamilton, C.J.; Tietz, R.F.; Wallenberger, F.T. WO Patent 9514740 (to DuPont De Nemours and Co.), 1995. Gallucci, R.R.; Patel, B.R. Poly(butylene terephthalate). In: Scheirs, J.; Long, T.E.; Eds. Modern Polyesters, Chemistry and Technology of Polyesters and Copolyesters; John Wiley & Sons: Chichester, 2004, pp. 293-321. Gandini, A.; Belgacem, M.N. Prog. Polym. Sci. 1997, 22, 1203-1397. Gandini, A.; Silvestre, A.J.D.; Pascoal Neto, C.; Sousa, A.F.; Gomes, M. J. Polym. Sci., Polym. Chem. 2009, 47, 295-298. (2009a)
Chapter 2 52 Gandini, A.; Coelho, D.; Gomes, M.; Reis, B.; Silvestre, A. J. Mater. Chem. 2009, 19, 8656-8664. (2009b) Gandini, A. Monomers and Macromonomers from Renewable Resources. In: Loos, K.; Ed. Biocatalysis in Polymer Chemistry; Wiley-VCH: Weinheim, 2010. Gandini, A. Green Chem. 2011, 13, 1061-1083. Gao, Q.; Nan-Xun, H.; Zhi-Lian, T.; Gerking, L. Chem. Eng. Sci. 1997, 52, 371-376. Gao, X.; Chen, J.C.; Wu, Q.; Chen, G.Q. Curr. Opin. Biotechnol. 2011, 22, 768-774. García-Martín, M.G.; Ruiz Pérez, R.; Benito Hernández, E.; Galbis, J.A. Macromolecules 2006, 39, 7941-7949. Gomes, M.; Gandini, A.; Silvestre, A.J.D.; Reis, B. J. Polym. Sci., Polym. Chem. 2011, 49, 3759-3768. Gómez, R.; Varela, O. Macromolecules 2009, 42, 8112-8117. Gross, S.; Flowers, D.; Roberts, G.; Kiserow, D.; DeSimone, J. Macromolecules 1999, 32, 3167-3169. Gross, S.; Roberts, G.; Kiserow, D.; DeSimone, J. Macromolecules 2000, 33, 40-45. Gubbels, E.; Jasinska-Walc, L.; Hermida Merino, D.; Goossens, H.; Koning, C. Macromolecules 2013, 46, 3975-3984. (2013a) Gubbels, E.; Jasinska-Walc, L.; Koning, C. J. Polym. Sci., Polym.Chem. 2013, 51, 890- 898. (2013b) Gutman, A.L.; Knaani, D.; Bravdo, T. Macromol. Symp. 1997, 122, 39-44. Holden, G.; Legge, N.R.; Quirk, R.; Schroeder, H.E. Thermoplastic Elastomers; Carl Hanser Verlag: Munich, 1996. Hu, J.; Gao, W.; Kulshrestha, A.; Gross, R.A. Macromolecules 2006, 39, 6789-6792. Huang, B.; Walsh, J. Polymer 1998, 39, 6991-6999. Huang, Y.M.; Li, H.; Huang, X.J.; Hu, Y.C.; Hu, Y. Chin. J. Bioprocess Eng. 2008, 6, 1-6. IHS Chemical Process Economics Program (PEP) Report #284 Bio-Based Adipic Acid, 2012. http://www.ihs.com/products/chemical/technology/pep/bio-based-adipic-acid.aspx [Accessed on May 2, 2013] Ikada, Y.; Tsuji, H. Macromol. Rapid Commun. 2000, 21, 117-132. Iribarren, J. I.; Martínez de Ilarduya, A.; Alemán, C.; Oraison, J. M.; Rodríguez-Galán, A.; Muñoz-Guerra, S. Polymer 2000, 41, 4869-4879. Iwamoto, A.; Tokiwa, Y. Polym. Degrad. Stabil. 1994, 45, 205-213.
Polyesters: Traditional products and present bio-based tendencies 53 Jackson, W.J. Br. Polym. J. 1980, 12, 154-162. Jansen, M.A.G.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Schick, C.; Koning, C.E. Macromolecules 2005, 38, 10658-10666. Jansen, M.A.G.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Koning, C.E. Anal. Chim. Acta 2006, 557, 19-30. Jansen, M.A.G.; Wu, L.H.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Koning, C.E. J Polym. Sci., Polym. Chem. 2007, 45, 882-899. Jansen, M.A.G.; Wu, L.H.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Koning, C.E; Portale, G. J Polym. Sci., Polym. Chem. 2008, 46, 1203-1217. Japu, C.; Alla, A.; Martínez de Ilarduya, A.; García-Martín, M.G.; Benito, E.; Galbis, J.A.; Muñoz-Guerra, S. Polym. Chem. 2012, 3, 2092-2101. Japu, C.; Martínez de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. Polymer 2013, 54, 1573- 1582. (2013a) Japu, C.; Martínez de Ilarduya, A.; Alla, A.; García-Martín, M.G.; Galbis, J.A.; Muñoz- Guerra, S. Polym. Chem. 2013, 4, 3524-3536. (2013b) Jasinska, L.; Villani, M.; Wu, J.; Van Es, D.; Klop, E.; Rastogi, S.; Koning, C. E. Macromolecules 2011, 44, 3458−3466. Jem, K.J.; van der Pol, J.F.; de Vos, S. Plastics from Bacteria, Natural Functions and Applications; Springer: Heidelberg, 2010, pp. 323-346. Jiang, M.; Liu, Q.; Zhang, Q.; Ye, C.; Zhou, G. J. Polym. Sci., Polym. Chem. 2011, 50, 1026-1036. Kampouris, E.; Papaspyrides, C. Polymer 1985, 26, 413-417. Khanarian, G.; Charbonneau, L.F.; Johnson, R.E.; Witteler, H.B.; Lee, R.G.; Sandor, R.B. US Patent 6025061 (to HNA Holdings, Inc.), 2000. Khanarian, G.; Charbonneau, L.F.; Witteler, H.B. US Patent 6359070 (to E.I. Du Pont de Nemours and Co.), 2002. Ki, H.C.; Park, O.O. Polymer 2001, 42, 1849-1861. Kiely, D.E.; Chen, L.; Lin, T.-H. J. Am. Chem. Soc. 1994, 116, 571-578. Kiely, D.E.; Chen, L.; Lin, T.-H. J. Polym. Sci., Polym. Chem. 2000, 38, 594-603. Kim, Y.J.; Park, O.O. J. Appl. Polym. Sci. 1999, 72, 945-951. Kimura, H.; Yoshinari, T.; Takeishi, M. Polym. J. 1999, 31, 388-392.
Chapter 3 60
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 61 3.1. Aim and scope of this Chapter Driven by increasing concerns on sustainable development and minimizing the impact of materials on the environment, biodegradable polymers have attracted a great deal of interest in the last decades. Among synthetic biodegradable polymers, aliphatic polyesters are the most widely studied and largely used. Aliphatic polyesters such as poly(butylene succinate) (PBS), poly(L-lactic acid) (PLA), poly(ε-caprolactone) (PCL) and poly(3-hydroxy butyrate) (PHB) have found intensive use in a broad variety of medical applications such as bioresorbable surgical sutures, prosthesis, dental implants, bone screw and plates for temporary internal fracture fixation, and controlled drug delivery systems. Aliphatic polyesters are also receiving attention as materials suitable for replacing conventional commodity plastics in injection-molded articles (cutlery, brushes), tubular films (composting bags, shopping bags), packaging and food tray. However, most of them have a glass-transition temperature (Tg) not high enough for its use in those applications where stiffness and thermal resistance are priority requisites. In this regard, various approaches such as blending and copolymerization with aromatic polyesters as poly(butylene terephthalate) (PBT) and poly(ethylene terephthalate) (PET) have been explored. The interest for cyclic monomers arises from their capacity for adding stiffness to the polymer chain with the subsequent increase in Tg; the disadvantage of using aromatic units for such purpose is that they are originated from fossil feedstocks and are reluctant to biodegradation. The use of carbohydrate-based monomers with a cyclic structure constitutes a scarcely explored approach for the preparation of renewable aliphatic polyesters with improved physical properties, especially those related to polymer chain stiffness. Moreover, carbohydrate-derived polycondensates typically display enhanced hydrophylicity, lower toxicity and higher susceptibility to hydrolytic degradation and biodegradation than those coming from petrochemical feedstocks. 1,4:3,6-Dianhydro-D-glucitol, known as isosorbide, along with their two less accessible stereoisomers isomannide and isoidide, are currently drawing an enormous interest in the polymer science field as bio-based monomers able to provide enhanced stiffness into the polymer chain they are incorporated. Nevertheless, the main shortcoming of isosorbide and its isomers is the limited reactivity of their secondary hydroxyl groups; in fact, this feature seriously hampers the polycondensation reaction in the melt, so aliphatic polyesters from 1,4:3,6-dianhydrohexitols and aliphatic dicarboxylic
Chapter 3 62 acids or dicarboxylic esters obtained by this method display rather limited molecular weights. Higher molecular weights are achievable via polycondensation with aliphatic dicarboxylic chlorides, but this method is not appropriate for industrial application. In this Chapter, aliphatic polyesters from bicyclic acetalized carbohydrate-based monomers will be obtained by polycondensation in the melt and in the total absence of solvents, to imitate as far as possible the conditions usually applied in the industry. The characterization of the new polyesters, the evaluation of their thermal and mechanical properties and their degradability and biodegradability will be reported.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 63 3.2. Carbohydrate-based polyesters made from bicyclic acetalized galactaric acid Summary: The dimethyl ester of 2,3:4,5-di-O-methylene-galactaric acid (Galx) was made to react in the melt with 1,n-alkanediols HO(CH2)nOH containing even numbers of methylenes (n from 6 to 12) to produce linear polycyclic polyesters. Two sets of poly(alkylene 2,3:4,5-di-O-methylene-galactarate) polyesters (PE-nGalx) with weightaverage molecular weights in the 5,000-10,000 and 35,000-45,000 g·mol-1 ranges were obtained using TBT and DBTO catalysts, respectively. For comparative purposes a set of poly(alkylene adipate) polyesters (PE-nAd) was also synthesized with molecular weights in the higher range using a similar procedure. The thermal stability of PE-nGalx was greater than that of PE-nAd although it notably decayed as molecular weight decreased. The replacement of Ad by Galx in the polyesters caused increases in Tg of up to 70 ºC, and almost doubled the tensile mechanical parameters. All PE-nGalx were semicrystalline but only those made from 1,12-dodecanediol were able to crystallize from the melt with a crystallization rate that diminished as the molecular weight increased. In general, the galactarate containing polyesters displayed higher solubility and wettability than polyadipates, they hydrolyzed faster and exhibited comparable sensitivity to the action of lipases. Publication derived from this work: Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. Biomacromolecules 2011, 12, 2642-2652.
Chapter 3 64 3.2.1. Introduction The development of bio-based polymers today is drawing an enormous amount of interest for their potential to reduce the utilization of petrochemicals and to increase the added-value of agriculture products and wastes (Wool and Sun, 2005). Among the renewable naturally-occurring sources, carbohydrates stand out in a privileged position due to their huge abundance and because they are inexhaustible and readily available. A good assortment of monomers based on carbohydrates has been recently explored for making different classes of polymers (Wang et al., 2002; Bou et al., 1996; Kricheldorf, 1997; Galbis and García-Martín, 2008). The hydroxyl and carboxylic rich functionality of these compounds makes them mostly appropriate for polycondensation in spite of the fact that certain chemical handling for group protection will be required if linear polymers are searched. Carbohydrate-based polycondensates may be provided with a great diversity of chemical structures and unusual properties; they typically display enhanced hydrophilicity, less toxicity, and higher susceptibility to biodegradation than petrochemical-based polycondensates, offering therefore a wide possibility of applications in food packaging and medical devices (García-Martín et al., 2005; Haider and Williams, 2008). Among the wide diversity of sugar-derived monomers that have been recently explored, 1,4:3,6-dianhydrohexitols have drawn particular attention for the synthesis of linear polycondensates (Fenouillot et al., 2010; Galbis and García-Martín, 2010). In these hexitols, the four exceeding hydroxyl groups are blocked by intramolecular etherification leading to a bicyclic structure with the secondary 2- and 5-hydroxyl groups standing free for reaction. The interest for monomers having a cyclic structure arises from their capacity for adding stiffness to the polymer chain with the subsequent increasing in the Tg. 1,4:3,6- Dianhydro-D-glucitol known as isosorbide (Is) is the only dianhydrohexitol industrially available today because it can be produced from cereal starch through an economically acceptable process. It is composed of two cis-fused nearly planar tetrahydrofuran rings with a dihedral angle of 120º and the 2- and 5-hydroxyl groups in endo and exo positions, respectively (Scheme 3.1). Isosorbide has been extensively investigated in the last decades as a convenient building block for diverse polycondensates (Chatti et al., 2006; Caouthar et al., 2007; Sablong et al., 2008; Noordover et al., 2008; Marín and Muñoz- Guerra, 2009). In this line, special attention is given to aromatic polyesters and copolyesters (Thiem and Lüders, 1984; Storbeck et al., 1993, Quintana et al., 2011), in
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 65 which Is has proven to replace successfully petrochemical diols without detriment, or even with improvement, of their properties. Scheme 3.1. Chemical structures of 1,4:3,6-dianhydrosorbitol (isosorbide, Is) and dimethyl 2,3:4,5- di-O-methylene galactarate (Galx). In this work we present a new bicyclic carbohydrate-based monomer useful for the preparation of linear polycondensates, specifically for the synthesis of aliphatic polyesters. Dimethyl 2,3:4,5-di-O-methylene-galactarate (Scheme 3.1), abbreviated as Galx, is the dimethyl ester of galactaric acid with the hydroxyl groups acetalized with formaldehyde. This compound is readily synthesized in one step from commercially available mucic acid (galactaric acid) which in turn is produced from naturally-occurring D- galactose or D-galactose-containing compounds by oxidation with nitric acid (Butler et al., 1958). Contrary to isosorbide, Galx is centrosymmetric and therefore does not generate regicity in the growing polymer chain, and its two carboxyl groups display the same reactivity because they are spatially undistinguishable. The use of acetalized aldaric acid derivatives as building blocks for the synthesis of polycondensates is not new, but most of the cases reported so far refer to polyamides and polyesters derived from acetalized tartaric acid, which contain a dioxolane ring forming part of the polymer backbone. Thus, Muñoz-Guerra et al. have described the synthesis and crystal structure of linear polyamides from di-O-methylene L-tartaric acid and several diamines (Rodríguez-Galán et al., 1992; Iribarren et al., 2000), and Kimura et al. studied their biodegradability (Kimura et al., 1999). More recently, low molecular weight polyesters made from di-O-isopropylidene L-tartaric acid and alkanediols have been reported (Dhamaniya and Jacob, 2010). To our knowledge, bicyclic acetalized aldaric acid derivatives have not been explored for the synthesis of polycondensates to date. It should be mentioned however that 1-deoxy-1-isocyanate-2,3:4,5-di-O- isopropylidene-D-galactitol has been synthesized and used as monomer for the preparation of [n]-polyurethanes (Gómez and Varela, 2009).
Chapter 3 66 3.2.2. Experimental section 3.2.2.1. Materials Dimethyl 2,3:4,5-di-O-methylene-galactarate was synthesized following the procedure reported by Stacey et al. (Butler et al., 1958). The reagents 1,6-hexanediol (97%), 1,8-octanediol (98%), 1,10-decanediol (98%), 1,12-dodecanediol (99%), dimethyl adipate (> 99%), and the catalysts titanium (IV) tetrabutoxide (TBT, 98%) and dibutyl tin oxide (DBTO, 98%) were purchased from Sigma-Aldrich. Solvents used for purification and characterization, as chloroform, methanol and diethyl ether, and solvents used in the solubility essays, were purchased from Panreac and were all of either technical or highpurity grade. The enzymes used in biodegradation experiments, lipase from porcine pancreas (activity 15-35 U·mg-1, pH 8.0, 37 ºC) and Amano lipase from Pseudomonas fluorescens (activity ≥ 20 U·mg-1, pH 8.0, 55 ºC) were also purchased from Sigma- Aldrich. One unit (U) was defined as that amount of enzyme which catalyzed the release of fatty acid from triglycerides at the rate of 1 µmol·min-1. All the reagents and solvents were used as received without further purification. 3.2.2.2. General methods 1H and 13C NMR spectra were recorded on a Bruker AMX-300 spectrometer at 25.0 ºC operating at 300.1 and 75.5 MHz, respectively. Polyesters and water degradation products were dissolved in deuterated chloroform or deuterated water, and spectra were internally referenced to tetramethylsilane (TMS) or the sodium salt of 3-(trimethylsilyl)- propanesulfonic acid. About 10 and 50 mg of sample dissolved in 1 mL of solvent were used for 1H and 13C NMR, respectively. A total of 64 scans were acquired for 1H and 1000-10000 for 13C, with 32 and 64-K data points as well as relaxation delays of 1 and 2 s, respectively. Elemental analyses were determined in the Microanalysis Laboratories of the CSIC, Barcelona, Spain. FTIR measurements were carried out in a Jasco 4100 FTIR spectrophotometer, coupled with an ATR accessory Specac MKII, with a single reflection Golden Gate diamond, ZnSe lenses and a high stability temperature driver West6100+. The absorbance of the sample was recorded in the range of 4000-550 cm-1 accumulating 32 scans for each run. Intrinsic viscosities of polyesters dissolved in chloroform were measured in an Anton Paar AMVn Automated Micro Viscosimeter at 25.00±0.01 ºC, using the VisioLab for AMVn software. Gel permeation chromatograms were acquired at 35.0 ºC with a Waters equipment provided with a refraction-index detector. The samples were chromatographed with 0.05 M sodium trifluoroacetate-hexafluoroisopropanol
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 67 (NaTFA-HFIP) using a polystyrene-divinylbenzene packed linear column with a flow rate of 0.5 mL·min-1. Chromatograms were calibrated against poly(methyl methacrylate) (PMMA) monodisperse standards. The thermal behavior of polyesters was examined by DSC using a Perkin Elmer DSC Pyris 1. DSC data were obtained from 3 to 5 mg samples at heating/cooling rates of 10 ºC·min-1 under a nitrogen flow of 20 mL·min-1. Indium and zinc were used as standards for temperature and enthalpy calibration. The glasstransition temperatures were determined at a heating rate of 20 ºC·min-1 from rapidly melt-quenched polymer samples. The treatment of the samples for isothermal crystallization experiments was the following: the thermal history was removed by heating the sample up to 150 ºC and left at this temperature for 5 min, and then it was cooled at 20 ºC·min-1 to the selected crystallization temperature, where it was left to crystallize until saturation. Isothermal crystallization under the same conditions was also carried out in an Olympus BX51 Polarizing Optical Microscope coupled to a THMS LINKAM heating plate and a cooling system LNP (Liquid Nitrogen Pump). Thermogravimetric analyses were performed under a nitrogen flow of 20 mL·min-1 at a heating rate of 10 ºC·min-1, within a temperature range of 30 to 600 ºC, using a Perkin Elmer TGA 6 equipment. Sample weights of about 10-15 mg were used in these experiments. Films for mechanical testing and contact angle measurements were prepared with a thickness of ~200 µm by casting from a chloroform solution at a concentration of 100 g·L-1. For evaluating mechanical properties, the films were cut into strips with a width of 3 mm, while the distance between testing marks was 10 mm. The tensile strength, elongation at break and Young’s modulus were measured at a stretching rate of 30 mm·min-1 on a Zwick 2.5/TN1S testing machine coupled with a compressor Dalbe DR 150, at 23 ºC. Contact angles between liquid and polyester film surfaces were measured by means of a Krüss DSA 100 contact angle measuring system. Angle values were registered after 3 min of dropping 15 µL of either water or ethylene glycol onto the polyester surface at 23 ºC. Scanning electron microscopy (SEM) images were taken with a field-emission JEOL JSM-7001F instrument (JEOL, Japan) from uncoated samples. 3.2.2.3. Polymer synthesis Synthesis of poly(alkylene 2,3:4,5-di-O-methylene galactarate)s (PE-nGalx). PE- nGalx polyesters were obtained from an aliphatic α,ω-diol (1,6-hexanediol, 1,8- octanediol, 1,10-decanediol or 1,12-dodecanediol) and dimethyl 2,3:4,5-di-O-methylene- galactarate.
Chapter 3 68 Procedure A. A mixture of dimethyl 2,3:4,5-di-O-methylene-galactarate and the corresponding diol at a molar ratio of 1/1 was prepared in a round-bottom flask, and several cycles of argon and vacuum were applied. The temperature was raised to 135 ºC for 30 min with stirring until homogenization of the mixture, and then TBT catalyst (0.5% mol respect to monomers) was added. Transesterification reactions were carried out under argon at heating at 140 ºC for 4 h followed by two periods of 1 h at 150 and 160 ºC, respectively. Polycondensation reactions were performed for 2 h at 175 ºC under vacuum (0.6-1.3 mbar). The polymerized mixture was then dissolved in dichloromethane, and the solution was poured dropwise into diethyl ether while stirring. The precipitated solid was filtered, washed with diethyl ether, and dried under vacuum. Procedure B. The reaction was performed in a three-necked, cylindrical-bottom flask equipped with a mechanical stirrer, a nitrogen inlet and a vacuum distillation outlet. A 1/1 molar ratio of the diol to dimethyl 2,3:4,5-di-O-methylene-galactarate and dibutyl tin oxide (DBTO) as catalyst (0.4% mol respect to monomers) were used. The apparatus was vented with nitrogen several times at room temperature to remove the air and avoid oxidation during the polymerization. Transesterification reactions were carried out under a low nitrogen flow for a period of 3 h at 140 ºC. Polycondensation reactions were left to proceed for 5 h at 140 ºC, under a 0.03-0.06 mbar vacuum. Then, the reaction mixture was cooled to room temperature, and the atmospheric pressure was recovered with nitrogen to prevent degradation. Polymers obtained were dissolved in chloroform and precipitated in excess of methanol to remove unreacted monomers and remaining oligomers. Finally, the polymer was collected by filtration, extensively washed with methanol and diethyl ether, and dried under vacuum. PE-6Galx. IR (film), (cm-1): 1734 (C=O st), 1260 (C-O asym st, ester), 1165 (C-O sym st, ester), 1093 (C-O asym st, acetal), 799 (C-O sym st, acetal), 724 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 5.25 and 5.08 (2 s, 4H, 2CH2), 4.61 (m, 2H, 2CH), 4.27 (m, 2H, 2CH), 4.20 (t, 4H, 2CH2), 1.70 (m, 4H, 2CH2), 1.40 (m, 4H, 2CH2). 13C (75.5 MHz, CDCl3), δ (ppm): 170.3 (CO), 96.8, 79.0, 75.3, 65.6, 28.4, 25.4. Anal. Calcd for C14H20O8· 0.25 H2O: C, 52.40; H, 6.37. Found: C, 52.08; H, 6.44. PE-8Galx. IR (film), (cm-1): 1733 (C=O st), 1259 (C-O asym st, ester), 1167 (C-O sym st, ester), 1092 (C-O asym st, acetal), 796 (C-O sym st, acetal), 719 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 5.26 and 5.08 (2s, 4H, 2CH2), 4.61 (m, 2H, 2CH), 4.28 (m, 2H, 2CH), 4.20 (t, 4H, 2CH2), 1.68 (m, 4H, 2 CH2), 1.34 (m, 8H, 4CH2). 13C (75.5 MHz,
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 69 CDCl3), δ (ppm): 170.3 (CO), 96.8, 79.0, 75.2, 65.8, 29.0, 28.5, 25.7. Anal. Calcd for C16H24O8·0.25 H2O: C, 55.09; H, 7.08. Found: C, 55.08; H, 7.02. PE-10Galx. IR (film), (cm-1): 1734 (C=O st), 1259 (C-O asym st, ester), 1168 (C-O st sym, ester), 1093 (C-O asym st, acetal), 797 (C-O sym st, acetal), 719 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 5.25 and 5.07 (2s, 4H, 2CH2), 4.60 (m, 2H, 2CH), 4.28 (m, 2H, 2CH), 4.19 (t, 4H, 2CH2), 1.67 (m, 4H, 2CH2), 1.30 (m, 12H, 6CH2). 13C (75.5 MHz, CDCl3), δ (ppm): 170.3 (CO), 96.8, 79.0, 75.2, 65.9, 29.4, 29.1, 28.5, 25.7. Anal. Calcd for C18H28O8·0.25 H2O: C, 58.05; H, 7.58. Found: C, 57.47; H, 7.47. PE-12Galx. IR (film), (cm-1): 1734 (C=O st), 1259 (C-O asym st, ester), 1168 (C-O sym st, ester), 1095 (C-O asym st, acetal), 801 (C-O sym st, acetal), 719 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 5.25 and 5.07 (2s, 4H, 2CH2), 4.60 (m, 2H, 2CH), 4.28 (m, 2H, 2CH), 4.19 (t, 4H, 2CH2), 1.67 (m, 4H, 2CH2), 1.27 (m, 16H, 8CH2). 13C (75.5 MHz, CDCl3), δ (ppm): 170.3 (CO), 96.8, 79.0, 75.1, 65.9, 29.5, 29.4, 29.2, 28.5, 25.8. Anal. Calcd for C20H32O8·0.25 H2O: C, 59.32; H, 8.09. Found: C, 58.60; H, 8.07. Synthesis of poly(alkylene adipate)s (PE-nAd). Dimethyl adipate and the corresponding aliphatic α,ω-diol (1,6-hexanediol, 1,8-octanediol, 1,10-decanediol or 1,12- dodecanediol) were introduced in a three-necked, cylindrical-bottom flask equipped with a mechanical stirrer, a nitrogen inlet and a vacuum distillation outlet. A molar excess of 16, 11, 3 and 2% of diol was used for 1,6-hexanediol, 1,8-octanediol, 1,10-decanediol and 1,12-dodecanediol, respectively. The apparatus was vented with nitrogen several times at room temperature, heated at 180 ºC under stirring until homogeneization of the mixture, and then titanium (IV) butoxide (TBT) catalyst (0.5% mol respect to dimethyl adipate) was added. Transesterification reactions were carried out under a low nitrogen flow for a period of 3 h at 180 ºC. The temperature was then increased to 200 ºC, and polycondensation reactions were performed at this temperature for 3 h, under a 0.03-0.06 mbar vacuum. After that, the reaction mixture was cooled to room temperature, and atmospheric pressure was recovered with nitrogen to prevent degradation. The solid mass was dissolved in chloroform, and the polymer was precipitated with methanol, collected by filtration, extensively washed with methanol and diethyl ether, and dried under vacuum. PE-6Ad. IR (film), (cm-1): 1724 (C=O st), 1256 (C-O asym st, ester), 1162 (C-O st sym ester), 733 and 725 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 4.06 (t, 4H, 2CH2), 2.32 (t, 4H, 2 CH2), 1.64 (m, 8H, 4CH2), 1.38 (m, 4H, 2CH2). 13C (75.5 MHz, CDCl3), δ
Chapter 3 76 there is a clear indication that the resistance to heat of PE-nGalx increases with molecular weight, and that the thermal stability of these polyesters is 20-30 ºC higher than that of PE-nAd having similar n and molecular weight values. The valuable conclusion drawn from this thermogravimetric study is that the insertion of Galx units in aliphatic polyesters instead of reducing their thermal stability contributes to a significant increase in their decomposition temperatures. Figure 3.2. TGA traces of PE-nGalx (HMW) and PE-nAd. Linear aliphatic polyesters without substituents are characterized by displaying low glass-transition temperatures. Our DSC Tg measurements of PE-nAd polyesters (Table 3.2) confirm indeed such general observation; these polyesters are so flexible and tend to crystallize so fast that Tg could be only determined for the relatively less flexible PE-6Ad which displayed a value as low as -61 ºC. The replacement of the adipate unit by the Galx unit in PE-nAd increased Tg of PE-6Ad by almost 70 ºC and renders values for the HMW series of PE-nGalx comprised in the -17 to 6 ºC range. It is well-known that the incorporation of rigid structures such as aromatic or aliphatic cyclic units in a polyester chain gives rise to a dramatic increase in Tg (Brandrup et al., 1999). Polyesters with Mn in the 10,000-23,000 g·mol-1 range made from isosorbide and aliphatic dicarboxylic acids with lengths similar to the alkanediols used in this work are reported to have Tg between -10 and 10 ºC (Braun and Bergmann, 1992), which are perfectly comparable to those displayed by HMW PE-nGalx. It seems therefore that Galx exerts a similar effect as Is on Tg, which is rather striking because a higher stiffness should be expected for the almost planar fused five-membered tetrahydrofurane rings than for the puckered and conformationally interconvertible 5,5’-bis(1,3-dioxolane) structure.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 77 Table 3.2. Thermal and mechanical properties of PE-nGalx and PE-nAd polyesters. Polyester TGA DSC Tensile test First heatinge Coolinge Second heatinge ºT5%a Tdb W c Tgd Tm ΔHm Xcg Tc ΔHc Tc ΔHc Tm ΔHm Xcg E (ºC) (ºC) (%) (ºC) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) (MPa) (MPa) (%) LMW PE-6Galx 280 356 18 -4 61/73f 34.8 - - - - - - - n.d. n.d. n.d. PE-8Galx 315 385 13 -17 52/66/74f 44.3 - - - - - - - n.d. n.d. n.d. PE-10Galx 294 375 12 -22 57/75f 48.5 - - 31 30.4 78 31.8 - n.d. n.d. n.d. PE-12Galx 276 369 10 -19 77 42.7 46 28.6 - - 79 28.2 - n.d. n.d. n.d. HMW PE-6Galx 325 398 13 6 51/70/73f 15.7 - - - - - - - 154 13 55 PE-8Galx 320 392 15 -8 52/69/73f 24.5 - - - - - - - 157 12 50 PE-10Galx 323 387 16 -13 53/77/81f 25.0 - - 40 20.1 81 20.2 - 151 13 60 PE-12Galx 327 387 13 -17 86 28.9 35 26.3 - - 86 26.9 - 153 13 60 PE-6Ad 300 367 1 -61 56 100.5 0.60 40 69.8 - - 57 73.1 0.44 100 7 201 PE-8Ad 297 372 2 n.d. 67 108.4 0.68 46 84.6 - - 68 85.3 0.53 88 8 210 PE-10Ad 305 363 1 n.d. 73 109.2 0.70 58 99.2 - - 74 99.7 0.64 95 7 203 PE-12Ad 311 375 1 n.d. 75 118.8 0.71 63 102.5 - - 75 109.1 0.65 87 6 215 aTemperature at which 5 % weight loss was observed. bTemperature for maximum degradation rate. cRemaining weight at 600 ºC. dGlass-transition temperature taken as the inflection point of the heating DSC traces of melt-quenched samples recorded at 20 ºC·min-1. eMelting (Tm) and crystallization (Tc) temperatures, and melting (ΔHm) and crystallization (ΔHc) enthalpies measured by DSC at heating/cooling rates of 10 ºC·min-1. f Multiple melting peak. g Crystallinity index based on 100 % crystalline polyester: ΔHmº PE-6Ad = 38.1 KJ·mol-1; ΔHmºPE-8Ad = 41.0 KJ·mol-1; ΔHmºPE-10Ad = 44.0 KJ·mol-1; ΔHmºPE-12Ad = 52.0 KJ·mol-1 (Maglio et al., 1979).
Chapter 3 78 It seems that the relatively elongated shape of Galx will allow a closer packing of the polymer chains, which implies less free volume and accordingly a higher Tg. With regard to the influence of the constitution of PE-nGalx on Tg, the effects observed were those according to expectations; Tg of the LMW set were around 10 ºC lower than of HMW set, and within each set, it decreased almost steadily with increasing values of n. The melting-crystallization behavior of the polyesters was characterized by DSC and the collected data are gathered in Table 3.2. PE-nAd polyesters are highly crystalline polymers displaying melting temperatures between 56 and 75 ºC and melting enthalpies greater than 100 J·g-1 which correspond to crystallinity indexes of 0.6-0.7, both parameters increasing steadily with the value of n. These unsubstituted polyesters crystallized very fast when cooled from the melt with a recovery of about 70-90% of their initial crystallinity and almost exact replication of their melting temperatures. PE-nGalx are also crystalline but they show remarkable differences compared to PE-nAd that evidence the detrimental influence exerted by the diacetalized galactarate residue on crystallinity and crystallizability. The heating DSC traces of PE-nGalx (HMW) are depicted in Figure 3.3. Those obtained from samples coming directly from synthesis displayed wide endotherms consisting of several melting peaks indicative of a defective crystallization of the polyester. Traces with similar profiles were obtained from PE-nGalx (LMW), which are not shown here but can be inspected on Annex B. Annealing of pristine samples of PE- nGalx (HMW) at temperatures between 60 and 75 ºC for 12 h homogenized the crystalline morphology; in fact, melting of the annealed polymer took place sharply and at high temperatures, as corresponds to a single population of crystallites of well-definite size. Tm of PE-nGalx (measured on annealed samples) are within the 70-85 ºC range, which is not far from the PE-nAd melting temperature range, with values increasing regularly with the value of n. Whereas neither molecular weight nor constitution seem to notably affect melting temperatures, their influence on melting enthalpy appears to be very remarkable. As can be seen in Table 3.2, Hm of PE-nGalx (HMW) polymers are several times smaller than those of PE-nAd and approximately the half of the values observed for PE-nGalx (LMW). Lastly, just to note that, with regard to polyesters made from isosorbide reported in the literature (Okada et al., 2000), melting of PE-nGalx takes place at higher temperatures but with similar associated enthalpies.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 79 Figure 3.3. DSC melting traces of samples of PE-nGalx (HMW) coming directly from synthesis (left) and annealed (right) at (a) 75 ºC for 12 h, (b) 70 ºC for 12 h, (c) 65 ºC for 12 h and (d) 60 ºC for 12 h. At difference with PE-nAd, PE-nGalx are in general unable to crystallize from the melt. In fact, only PE-12Galx, both LMW and HMW, which is the highest flexible PE- nGalx, showed exothermal crystallization peaks at the cooling DSC traces; the crystallized polymers display similar Tm but significantly lower Hm than their respective pristine samples. Given the extreme relevance of this property regarding polymer processing, the isothermal crystallization of LMW and HMW PE-12Galx was kinetically analyzed at different temperatures by using the Avrami approach to appraise quantitatively the effect of molecular weight on crystallizability. The relative crystallinity vs. time plot as well as the double logarithmic plot for both low and high molecular weight PE- 12Galx are compared in Figure 3.4, and kinetics data extracted from these results are gathered in Table 3.3. The conclusions that may be drawn from the kinetics study are according to expectations. LMW PE-12Galx crystallized much faster than HMW PE- Table 3.3. Isothermal crystallization data for LMW and HMW PE-12Galx. Tc t0 t1/2 n -logk Tm Polyester (ºC) (min) (min) (ºC) PE-12Galx (LMW) 50 0.28 2.08 2.28 0.75 78.2 55 0.38 3.16 2.51 1.27 78.5 60 0.41 5.16 2.96 2.14 79.2 PE-12Galx (HMW) 50 0.35 2.36 2.80 1.04 81.9 55 0.40 3.74 2.92 1.76 82.5 60 0.45 8.66 2.99 2.90 83.1
Chapter 3 80 12Galx, and in both cases crystallization rate diminished as temperature increased revealing that within the studied temperature range, the process is dominated by nucleating effects. Furthermore, the influence of temperature on crystallization rate is more pronounced in the case of high molecular weight PE-12Galx. Figure 3.4. Isothermal crystallization of PE-12Galx at the indicated temperatures. a) Relative crystallinity vs. time plot for PE-12Galx (LMW). b) Log-log plot for PE-12Galx (LMW). c) Relative crystallinity vs. time plot for PE-12Galx (HMW). d) Log-log plot for PE-12Galx (HMW). A preliminary evaluation of the mechanical behavior of PE-nGalx (HMW) compared to PE-nAd was made by tensile testing at room temperature. The obtained stress-strain curves are depicted in Figure 3.5, and the mechanical parameters extracted from them are listed in Table 3.2. The elastic modulus values of PE-nGalx oscillate in the 150-160 MPa range, whereas PE-nAd have moduli in the 85-100 MPa range. Tensile strengths reach values of 12-13 MPa and 6-8 MPa for PE-nGalx and PE-nAd, respectively, whereas elongations at break are around 50 and 200%. The higher rigidity and lower ductility observed for PE-nGalx seem to be the logical consequence of the higher Tg displayed by these polyesters compared to PE-nAd, which doubtless is due to the presence of the bicyclic acetal structure in the polymer chain. However, no correlation between the mechanical parameters and the length of polymethylene segment becomes
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 81 apparent, the small observed differences falling within the acceptable experimental error limit. Figure 3.5. Stress-strain curves for PE-nGalx (HMW) and PE-nAd. 3.2.3.3. Hydrolytic degradation and biodegradation It is well-known that polycondensates of synthetic origin are resistant to hydrolysis or even apparently inert for long time intervals when incubated under physiological conditions. It has been repeatedly shown that the incorporation of sugar residues in polyesters, polyamides or polyurethanes enhances the susceptibility of the polymer to hydrolysis (Ruiz-Donaire et al., 1995; Zamora et al., 2006; De Paz et al., 2007). Both aliphatic and aromatic polyesters containing isosorbide were reported to be hydrolyzed at higher rates than their unsubstituted analogous when incubated under similar conditions. Okada et al. focused on the biodegradability of polyesters made from aliphatic dicarboxylic acids and 1,4:3,6 dianhydrohexitols (Okada et al., 1996, 1997 and 1999), and concluded that all of them were biodegraded to an extent that largely depended on the configuration of the anhydroalditol. To evaluate the effect of Galx on the hydrolysis of aliphatic polyesters, a comparative study using PE-8Ad and PE-8Galx was carried out in this work. Samples of the two polymers were incubated in aqueous buffers at pH 2.0, 7.4 and 10.5 at room temperature for two months, and the evolution of the degradation process was followed by measuring the weight loss and the molecular weight of the residual polymer. The results of this comparative study are presented in Figure 3.6. The profiles obtained for PE-8Ad indicate that there is no loss of mass and that the decrease in molecular weight is practically negligible at any pH. On the contrary, PE- 8Galx showed an appreciable weight loss that was minimum (4%) at pH 7.4 and
Chapter 3 82 maximum (~15%) at pH 2.0, as well as a significant decreasing of both Mw and Mn, whereas polydispersity remained almost constant. Figure 3.6. a) Remaining weight vs. hydrolytic degradation time for PE-8Galx (HMW) and PE-8Ad at pH 2.0, 7.4 and 10.5. (b) Changes in Mw and Mn (solid symbols) and polydispersity index (empty symbols) vs. incubation time for PE-8Galx. (c) Changes in Mw and Mn (solid symbols) and polydispersity index (empty symbols) vs. incubation time for PE-8Ad. Figure 3.7. Degradation of PE-8Galx (HMW) and PE-8Ad in the presence of lipases from Pseudomonas fluorescens and porcine pancreas, and without lipase. (a) Remaining weight vs. degradation time. (b) Changes in Mw and Mn (solid symbols) and polydispersity index (empty symbols) vs. incubation time for PE-8Galx. (c) Changes in Mw and Mn (solid symbols) and polydispersity index (empty symbols) vs. incubation time for PE-8Ad.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 83 A similar study was then performed with PE-8Ad and PE-8Galx but adding either P.fluorescens or porcine pancreas lipases to the incubation medium at pH 7.4, and results obtained over an incubation period of three weeks at 37 ºC are presented in Figure 3.7. In the presence of lipases, both PE-8Ad and PE-8Galx showed appreciable greater weight losses and molecular weight reductions than in the absence of enzymes indicating that the hydrolysis of both polyesters was speeded by lipases. The activity of the two lipases was not noteworthy different as neither was the sensitivity of the two polymers to the enzymatic action; in fact, a comparison of the plots in Figure 3.7 clearly reveals that the greater degree of biodegradation apparently observed for PE-8Galx is the only consequence of its lower intrinsic stability against hydrolysis. The changes taking place in the micromorphology of PE-8Ad and PE-8Galx due to the degradation process were revealed by SEM analysis as it is illustrated in Figure 3.8. The surface of a fresh fractured pristine sample of PE-8Ad exhibited a spherulitic texture, which became more vivid after incubation in water at pH 2.0, probably due to the removal of some intra- and interspherulitic amorphous material. Upon incubation in the presence of porcine pancreas lipase, the surface became smoother and developed some small cracks, indicating that a certain degradation in bulk happened. The results obtained in the parallel analysis carried out on PE-8Galx were clearly different. In this case, the initial surface did not display any feature of crystalline morphology, and after incubation in acidic water, plenty of holes and grooves appeared delineating what seems to be a texture made of relatively small impinging spherulites. In the presence of lipase, large cracks causing the total fracture of the material became apparent. The SEM pictures support the conclusion that PE-8Galx becomes more severely degraded than PE-8Ad, both in the presence and in the absence of lipases, and that the degradation process does not follow the same mechanism in each case.
Chapter 3 84 Figure 3.8. SEM micrographs of PE-8Ad (top) and PE-8Galx (HMW) (bottom). (a, a’) Initial sample. (b, b’) After incubation at pH=2.0 for 56 days. (c, c`) After incubation in the presence of lipase from porcine pancreas at pH=7.4 for 21 days. Figure 3.9. Compared 1H NMR spectra for PE-8Galx (HMW) after incubation at pH=2.0 for 56 days (top) and initial sample (bottom). The arrow indicates the signal arising from -CH2OH end groups that appear upon hydrolysis of the main chain ester group.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 85 Figure 3.10. 1H NMR spectra in D2O of the product resulting after incubation of dimethyl 2,3:4,5-di- O-methylene galactarate at pH 2.0 for three months at room temperature (top) and of initial sample in CDCl3 (bottom). Finally, a NMR study was made to explore how the polyester chain is degraded at the molecular level. The changes taking place in the 1H NMR spectrum of a sample of PE-8Galx upon incubation in aqueous buffer at pH 2.0 are shown in Figure 3.9. All the signals present in the original spectrum remained unaltered after degradation, and the only changes observed after incubation was the appearing of a small signal corresponding to hydroxymetyl group at 3.6 ppm. 1NMR spectra of PE-8Galx samples incubated at pH 7.4 and 10.5 afforded similar results, indicating that degradation proceeded by splitting of the relatively weak ester group without modification of the polymer chain structure. The result is rather striking since the acetal group is known to be sensitive to acidic conditions (Smith and March, 2007) and the opening of the dioxolane rings should have therefore been expected to happen in some extent. To check the stability of the diacetal structure, dimethyl 2,3:4,5-di-O-methylene-galactarate was incubated at room temperature in aqueous buffer at pH 2.0, 7.4 and 10.5. The initial powdered nonsoluble sample dissolved as degradation proceeded, and the NMR analysis of the solution after three months revealed that hydrolysis of the ester groups had happened almost quantitatively at both pH 2.0 and 10.5, whereas at pH 7.4 about 5% of the initial diester was still remaining. What was really remarkable is that the acetal group stayed intact after the treatment with whatever were the incubation conditions. The
Chapter 3 92 3.3.2. Experimental section 3.3.2.1. Materials Dimethyl 2,3:4,5-di-O-methylene-galactarate was synthesized following the procedure reported by Stacey et al. (Butler et al., 1958). The reagents 1,12-dodecanediol (99%), 1,6-hexanediol (97%) and dimethyl adipate (> 99%), and the catalyst dibutyl tin oxide (DBTO, 98%) were purchased from Sigma-Aldrich. Solvents used for purification and characterization, as chloroform and methanol, as well as solvents used in the solubility essays, were purchased from Panreac and they all were of either technical or high-purity grade. All the reagents and solvents were used as received without further purification. 3.3.2.2. General methods 1H and 13C NMR spectra were recorded on a Bruker AMX-300 spectrometer at 25.0 ºC operating at 300.1 and 75.5 MHz, respectively. Polyesters were dissolved in deuterated chloroform or in deuterated benzene, and spectra were internally referenced to tetramethylsilane (TMS). About 10 and 50 mg of sample dissolved in 1 mL of solvent were used for 1H and 13C NMR, respectively. Sixty-four scans were acquired for 1H and 1,000-10,000 for 13C with 32 and 64-K data points as well as relaxation delays of 1 and 2 s, respectively. FTIR measurements were carried out in a Jasco 4100 FTIR spectrophotometer, coupled with an ATR accessory Specac MKII, with a single reflection Golden Gate diamond, ZnSe lenses and a high stability temperature driver West6100+. The absorbance of the sample was recorded in the range of 4000-550 cm-1 accumulating 32 scans for each run. Intrinsic viscosities of polyesters dissolved in chloroform were measured in an Anton Paar AMVn Automated Micro Viscosimeter at 25.00±0.01 ºC, using the VisioLab for AMVn software. Gel permeation chromatograms were acquired at 35.0 ºC with a Waters equipment provided with a refraction-index detector. The samples were chromatographed with 0.05 M sodium trifluoroacetate-hexafluoroisopropanol (NaTFA-HFIP) using a polystyrene-divinylbenzene packed linear column with a flow rate of 0.5 mL·min-1. Chromatograms were calibrated against poly(methyl methacrylate) (PMMA) monodisperse standards. The thermal behavior of polyesters was examined by DSC using a Perkin Elmer DSC Pyris 1. DSC data were obtained from 3 to 5 mg samples at heating/cooling rates of 10 ºC·min-1 under a nitrogen flow of 20 mL·min-1. Indium and zinc were used as standards for temperature and enthalpy calibration. The glass-
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 93 transition temperatures were determined at a heating rate of 20 ºC·min-1 from rapidly melt-quenched polymer samples. The treatment of the samples for isothermal crystallization experiments was the following: the thermal history was removed by heating the sample up to 150 ºC and left at this temperature for 5 min, and then it was cooled at 20 ºC·min-1 to the selected crystallization temperature, where it was left to crystallize until saturation. For morphological study, isothermal crystallizations under the same conditions were carried out in an Olympus BX51 Polarizing Optical Microscope coupled to a THMS LINKAM heating plate and a cooling system LNP (Liquid Nitrogen Pump). Thermogravimetric analyses were performed under a nitrogen flow of 20 mL·min-1 at a heating rate of 10 ºC·min-1, within a temperature range of 30 to 600 ºC, using a Perkin Elmer TGA 6 equipment. Sample weights of about 10-15 mg were used in these experiments. Films for mechanical testing measurements were prepared with a thickness of ~200 µm by casting from chloroform solution at a polymer concentration of 100 g·L-1; the films were then cut into strips with a width of 3 mm while the distance between testing marks was 10 mm. The tensile strength, elongation at break and Young’s modulus were measured at a stretching rate of 30 mm·min-1 on a Zwick 2.5/TN1S testing machine coupled with a compressor Dalbe DR 150, at 23 ºC. Powder X-ray diffraction patterns were recorded on the INEL CPS-120 diffractometer in Debye-Scherrer configuration using the Cu-Kα radiation of wavelength 0.1542 nm from powdered samples coming directly from synthesis. Wide-angle X-ray scattering of anisotropic samples was performed in the synchrotron beamline A2 at HASYLAB, Hamburg, Germany. Scattering patterns were collected by a two-dimensional position sensitive marccd 165 detector. Series of scattering patterns were recorded at 30 s cycle times including 20 s of exposure. 3.3.2.3. Polymer synthesis The homopolyesters PE-12Galx, PE-12Ad, PE-6Galx and PE-6Ad were synthesized following the procedure recently reported by Lavilla et al. (Lavilla et al., 2011 -Subchapter 3.2-). PE-nAdxGalxy copolyesters were obtained from the chosen aliphatic α,ω-diol (either 1,6-hexanediol or 1,12-dodecanediol) and a mixture of dimethyl 2,3:4,5- di-O-methylene-galactarate and dimethyl adipate with the selected composition. The reaction was performed in a three-necked, cylindrical-bottom flask equipped with a mechanical stirrer, a nitrogen inlet and a vacuum distillation outlet. A 5% molar excess of diol to the diester mixture, and dibutyl tin oxide (DBTO) as catalyst (0.8% w/w respect to monomers) were used. The apparatus was vented with nitrogen several times at room
Chapter 3 94 temperature in order to remove the air and avoid oxidation during the polymerization. The transesterification reaction was carried out under a low nitrogen flow at 140 ºC for 3 h. The polycondensation reaction was left to proceed at 140 ºC for 5 h, under a 0.03-0.06 mbar vacuum. Then, the reaction mixture was cooled to room temperature, and the atmospheric pressure was recovered with nitrogen to prevent degradation. The resulting polymers were dissolved in chloroform and precipitated in excess of methanol in order to remove unreacted monomers and formed oligomers. Finally, the polymer was collected by filtration, extensively washed with methanol, and dried under vacuum. PE-12AdxGalxy. IR (film), (cm-1): 1730 (C=O st), 1260 (C-O asym st, ester), 1160 (C-O sym st, ester), 1095 (C-O asym st, acetal), 800 (C-O sym st, acetal), 730 and 720 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 5.25 and 5.05 (2s, Galx, 2CH2), 4.60 (m, Galx, 2CH), 4.28 (m, Galx, 2CH), 4.19 (t, Galx, 2CH2), 4.06 (t, Ad, 2CH2), 2.32 (t, Ad, 2CH2), 1.65 (m, Ad, Galx, 6CH2), 1.28 (m, Ad, Galx, 16CH2). 13C NMR (75.5 MHz, C6D6), δ (ppm): 172.6 (CO), 170.3 (CO), 97.1, 79.9, 76.3, 65.6, 64.3, 34.1, 29.9, 29.8, 29.6, 29.5, 29.2, 28.9, 26.4, 26.1, 24.9. PE-6AdxGalxy. IR (film), (cm-1): 1730 (C=O st), 1260 (C-O asym st, ester), 1160 (C-O sym st, ester), 1095 (C-O asym st, acetal), 800 (C-O sym st, acetal), 730 and 720 (CH2 rock). 1H NMR (300.1 MHz, CDCl3), δ (ppm): 5.25 and 5.05 (2s, Galx, 2CH2), 4.60 (m, Galx, 2CH), 4.28 (m, Galx, 2CH), 4.20 (t, Galx, 2CH2), 4.06 (t, Ad, 2CH2), 2.32 (t, Ad, 2CH2), 1.65 (m, Ad, Galx, 6CH2), 1.39 (m, Ad, Galx, 4CH2). 13C NMR (75.5 MHz, CDCl3), δ (ppm): 173.4 (CO), 170.3 (CO), 96.8, 79.0, 75.3, 65.6, 64.3, 33.9, 28.5, 28.4, 25.5, 25.4, 24.4. 3.3.3. Results and discussion 3.3.3.1. Synthesis and chemical structure The synthesis of PE-12AdxGalxy and PE-6AdxGalxy copolyesters was carried out by reacting mixtures of dimethyl adipate and Galx in the selected proportions, with 1,12- dodecanediol or 1,6-hexanediol respectively, as depicted in Scheme 3.4. Polycondensation conditions were similar to those previously applied by us in the synthesis of PE-nAd and PE-nGalx homopolyester series, which included the two pairs of parent homopolyesters (PE-12Ad/PE12Galx and PE-6Ad/PE-6Galx) used in this work as
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 95 references. Thus, reactions were performed in the melt and in the total absence of solvents, and under a progressively diminishing reaction pressure in order to facilitate the releasing of volatile by-products. As before, the catalyst DBTO was used instead of the more familiar TBT, which allowed to keep the temperature at 140 ºC along the whole polycondensation process minimizing the decomposition of thermally sensitive sugar compounds. By these means, copolyesters could be prepared as white powders with acceptable molecular weights in yields close to 90%. The synthesis results obtained for the two copolyester series are gathered in Table 3.4 together with data previously reported for their corresponding parent homopolyesters. Scheme 3.4. Polymerization reactions leading to PE-nAdxGalxy copolyesters. The intrinsic viscosities of the copolyesters oscillate between 0.5 and 0.8 dL·g-1, with the lower values being generally observed for the PE-6 series. The GPC analysis afforded Mn values comprised in the ~14,000-16,000 g·mol-1 range with polydispersities between 2 and 3. In addition to GPC, the Mn of the polyesters was estimated by 1H NMR (end-group analysis) and a very good agreement was found between values provided by the two techniques. The chemical constitution and composition of the resulting polycondensates were ascertained by infrared and 1H NMR spectroscopies. Infrared spectra were very similar for the two series showing the typical carboxylate (~1730, ~1260 and ~1170 cm-1) and cyclic acetal (~1095 and ~800 cm-1) stretching absorptions as well as the methylene stretching (3000-2850 cm-1), bending (~1470 cm-1) and rocking (~730-720 cm-1) bands with intensities according to compositions.
Chapter 3 96 Table 3.4. Molecular weights and compositions of PE-12AdxGalxy and PE-6AdxGalxy copolyesters. Copolyester Molecular weight Molar composition Solubilitye Feed Copolyesterd Yield (%) [η]a Mnb Mnc Mwc Ðc XAd XGalx XAd XGalx H2O EtOH Et2O DMSO NMP DMF THF CHCl3 HFIP TFA PE-12Ad 91 0.66 16,000 16,900 38,000 2.2 100 0 100 0 - - - + + PE-12Ad90Galx10 90 0.73 15,200 14,200 37,000 2.6 90 10 89.6 10.4 - - - + + PE-12Ad70Galx30 88 0.76 16,900 16,300 45,600 2.8 70 30 68.2 31.8 - - + + + PE-12Ad50Galx50 89 0.68 15,500 14,300 38,700 2.7 50 50 48.6 51.4 - - + + + PE-12Ad30Galx70 89 0.72 16,300 15,500 42,000 2.7 30 70 28.2 71.8 - - + + + PE-12Ad10Galx90 90 0.69 15,800 14,500 37,800 2.6 10 90 9.6 90.4 - - + + + PE-12Galx 91 0.81 20,400 16,400 41,200 2.5 0 100 0 100 - - + + + PE-6Ad 92 0.68 23,800 21,700 42,200 1.9 100 0 100 0 - - - + + PE-6Ad90Galx10 90 0.80 18,600 16,200 45,300 2.8 90 10 89.0 11.0 - - - + + PE-6Ad70Galx30 88 0.67 17,900 15,100 39,400 2.6 70 30 68.2 31.8 - - + + + PE-6Ad50Galx50 89 0.52 16,500 15,400 35,000 2.3 50 50 48.3 51.7 - - + + + PE-6Ad30Galx70 88 0.53 16,900 16,000 36,200 2.3 30 70 28.1 71.9 - - + + + PE-6Ad10Galx90 91 0.62 17,200 15,300 38,100 2.5 10 90 11.7 88.3 - - + + + PE-6Galx 90 0.51 16,800 15,100 34,700 2.3 0 100 0 100 - - + + + aIntrinsic viscosity in dL·g-1 measured in chloroform at 25 ºC. b Number-average molecular weight in g·mol-1 determined by 1H NMR end group analysis. c Number and weight-average molecular weights in g·mol-1 and dispersities measured by GPC in HFIP against PMMA standards.d Molar composition determined by integration of the 1H NMR spectra.e (-) Insoluble, (+) soluble.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 97 The 1H NMR spectra corroborated the chemical structure of the copolyesters with all signals being correctly assigned to the different protons contained in their repeating units (Figure 3.11). No significant traces of other signals indicative of impurities were detected in these spectra. Furthermore, integration of the proton signals arising from Ad and Galx units led to quantify the composition of the copolyesters in such units. Data provided by this analysis are given in Table 3.4, where it can be seen that the copolyesters compositions are essentially similar to those of their respective feeds. A close comparison of the composition values reveals however, that the content of the copolymers in adipic units is slightly lower than in their corresponding feed in most cases. Such preferential incorporation of the Galx units may be attributed to slight losses of the relatively volatile dimethyl adipate taking place during polycondensation. Figure 3.11. Compared 1H NMR spectra of PE-12AdxGalxy and PE-6AdxGalxy copolyesters. The microstructure of PE-12AdxGalxy copolyesters was determined by 13C NMR analysis and the spectrum registered from PE-12Ad50Galx50 is shown in Figure 3.12 for illustration. The signals of all the magnetically different carbons contained in the repeating units of the copolyesters appear well resolved in the 13C NMR spectra and in particular those arising from the methylene carbons of the diol, which come out to be sensitive to sequence effects at the dyad level. As shown in Figure 3.13 for the case of the PE- 12AdxGalxy, series, each resonance of the two δ-methylene carbons appears split so that four peaks are seen in the 29.4-29.7 ppm chemical shift interval corresponding to the three types of dyads (AdAd, AdGalx and GalxAd, GalxGalx) that are feasible along the copolyester chain. The dyad contents were determined by area integration provided that
Chapter 3 98 relaxation times of the involved carbon atoms are comparable. The plot of the content in each type of dyad as a function of the copolyesters composition reveals that the microstructure of the copolyesters is clearly statistical and randomness was calculated to be quite near unity in all cases. The statistical values obtained from this microstructure analysis are given in Table 3.5. Similar experimental results (not shown) were obtained in the analysis of the PE-6AdxGalxy so the same conclusions regarding microstructure were drawn for this series. Figure 3.12. 13C NMR spectrum of PE-12Ad50Galx50 copolyester. Figure 3.13. 13C NMR signals used for the microstructure analysis of PE-12AdxGalxy copolyesters (left) with indication of the dyads to which they are assigned (right).
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 99 Table 3.5. Microstructure analysis of PE-12AdxGalxy copolyesters. Copolyester Dyads (mol %) Number Average Sequence Lengths Randomness AA AG/GA GG nAd nGalx R PE-12Ad90Galx10 81.17 17.34 1.50 10.36 1.17 0.95 PE-12Ad70Galx30 48.15 39.98 11.87 3.41 1.59 0.92 PE-12Ad50Galx50 22.71 51.43 25.86 1.88 2.01 1.03 PE-12Ad30Galx70 9.97 37.06 52.97 1.54 3.85 0.91 PE-12Ad10Galx90 2.30 15.28 82.42 1.30 11.79 0.85 3.3.3.2. Thermal properties The thermal behavior of both PE-12AdxGalxy and PE-6AdxGalxy copolyesters has been systematically studied by TGA and DSC taking as reference their parent homopolyesters containing exclusively adipate or 2,3:4,5-di-O-methylene galactarate units. Firstly, the thermal stability was evaluated by TGA under an inert atmosphere. The weight loss vs. temperature plots registered for the two series are shown in Figure 3.14 evidencing the similitude of behavior among the members of each series. As it is illustrated in Figure 3.15 for the specific cases of PE-12Ad70Galx30 and PE-12Ad10Galx90, the thermal decomposition happens through two or three steps depending on the copolyester composition. In all cases the main step takes place with a maximum rate Td located in the 350-400 ºC temperature range with values roughly increasing with the content in Galx units, and involving weight losses up to 90% of the initial weight. A second decomposition step taking place at temperatures around 450 ºC and involving weight losses of 30% as maximum is invariably accompanying the main peak. The copolyesters highly enriched in Galx units as well as Galx homopolyesters still show a third decomposition step near below 350 ºC with weight losses less than 15%. The concluding remark that can be drawn from the TGA analysis is that in general the replacement of adipate by Galx units increases the thermal stability of the polyesters in the two series. Such a trend is not applicable however to copolyesters with very high contents in sugar units ( 90%), which start to decompose at temperatures below 350 ºC, apparently due to the presence of long sequences of Galx units.
Chapter 3 100 Figure 3.14. TGA traces of PE-12AdxGalxy (a) and PE-6AdxGalxy (b) copolyesters. Figure 3.15. TGA traces (solid lines) and derivative curves (dashed lines) of PE-12Ad70Galx30 and PE-12Ad10Galx90 copolyesters.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 101 The DSC analysis revealed significant changes in the melting and glasstransitions of PE-12Ad and PE-6Ad polyesters upon the incorporation of cyclic acetalized galactaric units. Glass-transition and melting temperatures, and melting enthalpies measured in the DSC analysis of both pristine samples and samples crystallized from the melt are given in Table 3.6. A remarkable feature of Galx homopolyesters is that they have considerably higher Tg than their analogues made of Ad. Accordingly, the Tg of the copolyesters within each series increases steadily with the content in Galx units with minimum and maximum values corresponding to their respective reference homopolyesters (Figure 3.16). The trend observed for Tg with the copolyester composition is in full agreement with what should be expected from the stiffness/bulkiness of the bicyclic sugar unit compared to the relative flexible/slender tetramethylene segment of the adipic acid, once it is provided that the microstructure of the copolyesters is statistical. Figure 3.16. Glass-transition temperature (a) and melting temperature (b) vs. composition plots for PE-12AdxGalxy and PE-6AdxGalxy.
Chapter 3 108 3.3.3.3. Crystal structure It is known that the insertion of sugar units bearing side groups in linear polyamides and polyesters create bulky heterogeneities in the polymer hindering the close chain packing required for crystallization. The behavior usually observed is that both aliphatic and aromatic random copolyesters with minor amounts of sugar units, i.e. below ~30%, are able to crystallize retaining the crystal structure of the parent homopolyester with rejection of the sugar units from the crystalline phase. Whereas homopolyesters entirely made of sugar units are able to crystallize adopting specific crystal structures, the copolyesters enriched in sugar units are usually amorphous (Zamora et al., 2005; Alla et al., 2006 and 2008). DSC has shown that both PE- 12AdxGalxy and PE-6AdxGalxy copolyesters are crystalline for the whole range of compositions which is a rather unusual behavior. The X-ray diffraction analysis corroborated the DSC results since, as it is shown in Figure 3.22, discrete scattering profiles indicative of semicrystalline material were recorded for the whole series. The Bragg spacings and crystallinity indexes measured on such profiles are compared in Table 3.8. It is seen that homopolymer PE-12Ad and the copolymers PE-12Ad90Galx10 and PE-12Ad70Galx30 display the same diffraction pattern which is distinguished by the presence of two prominent reflections at 4.2 and 3.7 Å. Such a pattern is characteristic of poly(alkylene dicarboxylate) polyesters which are known to crystallize usually in an orthorhombic or monoclinic lattice with chains adopting a quasi all-trans conformation (Armelin et al., 2001). Accordingly, the two observed reflections should be related with the side-by-side packing of the chains and indexed therefore as hk0, and the absence of higher spacing reflections, which would arise from planes with l 0, may be interpreted as the result of the balancing shift of the chains along the c direction of the structure. A different pattern containing three sharp reflections at 4.0, 4.3 and 4.6 Å along with another broad one at 17-18 Å was recorded for PE-12AdxGalxy copolyesters with 50% or more of Galx units as well as for the homopolyester PE-12Galx. It is clear that a new crystal structure is adopted when the content in Galx units becomes the majority. The X-ray diffraction pattern shown in Figure 3.22, which is illustrative of the whole set, was recorded from a fiber of PE-12Ad10Galx90 obtained by stretching from the melt. The reflections at 4.0 and 4.6 Å are equatorially placed and interpreted therefore to arise from hk0 plane whereas the in-between one at 4.3 Å is off-meridional and should be related with the azimuthal arrangement of the chains. On the other hand, the reflection at 17.9 Å is observed in the fiber pattern at an off-meridian position (inset of Figure 3.22b). This
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 109 reflection may be indexed as either hk0 or hk1 for a crystal structure made of almost extended chains which are progressively or regularly shifted along the c-axis, respectively. Since the repeating unit length calculated for PE-12AdxGalxy with the chain in all-trans conformation is 24 Å, it is concluded that the shifting of the chains must be around 3 Å. Figure 3.22. a) Powder WAXD profiles of PE-12AdxGalxy copolyesters. b) Fiber pattern of PE- 12Ad10Galx90 (fiber axis vertical), Inset: Inner region showing medium-angle reflections.
Chapter 3 110 Table 3.8. Powder X-ray diffraction data and mechanical properties of PE-12AdxGalxy copolyesters. Copolyester X-ray diffraction data Mechanical Properties d a (Å) Xcb Elastic modulus Tensile strength Elongation at break (MPa) (MPa) (%) PE-12Ad 4.2 s 3.7 m 0.51 87 ± 5 6 ± 1 215 ± 9 PE-12Ad90Galx10 4.2 s 3.7 m 0.48 98 ± 4 7 ± 1 110 ± 8 PE-12Ad70Galx30 4.2 s 3.7 m 0.47 100 ± 4 8 ± 1 92 ± 7 PE-12Ad50Galx50 17.9 s 4.6 s 4.3 s 4.0 s 0.46 107 ± 3 9 ± 1 90 ± 8 PE-12Ad30Galx70 17.9 s 4.6 s 4.3 s 4.0 s 0.44 129 ± 5 10 ± 1 78 ± 8 PE-12Ad10Galx90 17.9 s 4.6 s 4.3 s 4.0 s 0.43 140 ± 5 12 ± 1 67 ± 6 PE-12Galx 17.9 s 4.6 s 4.3 s 4.0 s 0.42 155 ± 4 13 ± 1 59 ± 4 a Bragg spacings measured in powder diffraction patterns for samples coming directly from synthesis. Intensities visually estimated as follows: m, medium; s, strong; w, weak. b Crystallinity index calculated as the quotient between crystalline area and total area. Crystalline and amorphous areas were quantified using PeakFit v4.12 software. 3.3.3.4. Stress-strain behavior Aliphatic polyesters are known to display meager mechanical properties, which together with their rather low melting temperature, constitute one of the main reasons that have hampered their industrial expansibility. This has also restricted their use in biomedical applications in spite of their good biocompatibility and biodegradability. In this regard it is interesting to evaluate the effect of the replacement of the alkylene dicarboxylate by the galactarate units on the mechanical behavior of polyadipates. For this aim, comparative tensile essays of the PE-12AdxGalxy series were carried out using thin films prepared by casting from chloroform solution. The traces registered in these essays are shown in Figure 3.23 and the mechanical parameters values extracted from such traces are given in Table 3.8. The evolution followed by the stress-strain curves with the content in Galx units clearly illustrates their influence on the mechanical compliance. It is noteworthy how the behavior changes from flexible/ductile typical of semicrystalline aliphatic polyesters to stiff/brittle when adipate units are entirely replaced by Galx units; the elastic modulus becomes almost doubled and the elongation to break is reduced to near one fourth of its value. The copolyesters with intermediate compositions show a steadily changing stress-strain behavior with mechanical parameters spanning between those of the two homopolyesters. As expected, the observed changes in the stress-strain behavior closely follow the changes in Tg; both effects are the logical consequence of the
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 111 increasing stiffness of the polyester chain that is motivated by the progressive replacement of Ad by Galx. Figure 3.23. Stress-strain curves for PE-12AdxGalxy copolyesters. 3.3.4. Conclusions The bicyclic diester dimethyl 2,3:4,5-di-O-methylene galactarate (Galx) was successfully copolymerized in the melt with dimethyl adipate and alkanediols (1,6- hexanediol and 1,12-dodecanediol) to obtain bio-based polycyclic random copolyesters covering the full range of compositions. The Galx containing copolyesters had agreeable molecular weights and were thermally stable and semicrystalline for all compositions. The enriched adipate copolyesters take up the crystal structure typical of aliphatic linear polyesters while a second crystal form seems to be adopted for high contents in sugar units. The melting temperature of the copolyesters changes with composition according to a power function that goes to the minimum as the contents in the two comonomers approach to each other. Copolyesters made from 1,12-dodecanediol were able to crystallize from the melt at a rate depending on composition. The incorporation of the Galx unit in the polyadipate chain increased the Tg as well as the modulus of rigidity and the stress yield whereas diminished the ductility of the polymer. The suitability of the bicyclic acetalized dimethyl galactarate to prepare bio-based aliphatic copolyesters with overall satisfying properties and increased Tg has been demonstrated.
Chapter 3 112 3.3.5. References Alla, A.; Hakkou, K.; Zamora, F.; Martínez de Ilarduya, A.; Galbis, J.A.; Muñoz-Guerra, S., Macromolecules 2006, 39, 1410-1416. Alla, A.; Hakkou, K.; Zamora, F.; Martínez de Ilarduya, A.; Galbis, J.A.; Muñoz-Guerra, S., J Polym. Sci., Polym. Chem. 2008, 46, 5167–5179. Armelin, E.; Casas, M.T.; Puiggalí, J. Polymer 2001, 42, 5695-5699. Beaman, R.G.; Morgan, P.W.; Koller, C.R.; Wittbecker, E.L.; Magat, E.E. J. Polym. Sci., 1959, 40, 329-336. Belgacem, M.N.; Gandini, A. Monomers, Polymers and Composites from Renewable Resources; Elselvier: Oxford, 2008. Bird, T.P.; Black, A.P.; Dewar, E.T.; Rutherford, D. Chem. Ind. 1960, 1331-1332. Bird, T.P.; Black, W.A.P.; Dewar, E.T.; Hare, J.B. Chem. Ind. 1961, 1077. Bird, T.P.; Black, W.A.P; Dewar, E.T.; Hare, J.B. J. Chem. Soc. 1963, 1208-1212. Black, W.A.P; Dewar, E.T.; Hare, J.B. J. Chem. Soc. 1963, 5724-5727. Burden, I.; Stoddart, J. F. J. Chem. Soc. Perkin I 1975, 675-682. Butler, K.; Lawrance, R.L; Stacey, M. J. Chem. Soc. 1958, 740-743. Chatti, S.; Schwarz, G.; Kricheldorf, H.R. Macromolecules 2006, 39, 9064-9070. (2006a) Chatti, S.; Kricheldorf, H.R.; Schwarz, G. J. Polym. Sci., Polym. Chem. 2006, 44, 3616- 3628. (2006b) Chen, L.; Kiely, D.E. J. Org. Chem. 1996, 61, 5847-5851. Dhamaniya, S.; Jacob, J. Polymer 2010, 51, 5392-5399. Fenouillot, F.; Rousseau, A.; Colomines, G.; Saint-Loup, R.; Pascault, J.P. Prog. Polym. Sci. 2010, 35, 578-622. Gómez, R.; Varela, O. Macromolecules 2009, 42, 8112-8117. Galbis, J.A.; García-Martín, M.G. Top. Curr. Chem. 2010, 295, 147–176. Gandini, A. Green Chem. 2011, 13, 1061-1083. Iribarren, J.I.; Martínez de Ilarduya, A.; Alemán, C., Oraison, J.M., Rodríguez-Galán, A., Muñoz-Guerra, S. Polymer 2000, 41, 4869–4879. Juais, D.; Naves, A.F.; Li, C.; Gross, R.A.; Catalani, L.H. Macromolecules 2010, 43, 10315-10319. Kiely, D.E.; Chen, L.; Lin, T.H. J. Amer. Chem. Soc. 1994, 116, 571-578.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 113 Kimura, H.; Yoshinari, T.; Takeishi, M. Polym. J. 1999, 31, 388-392. Kint, D.P.R.; Wingströn, E., Martínez de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. J. Polym. Sci., Polym. Chem. 2001, 39, 3250–3262. Kricheldorf, H.R. J. Macromol. Sci., Rev. Macromol. Chem. Phys. 1997, C37, 599-631. Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. Biomacromolecules 2011, 12, 2642–2652. -Subchapter 3.2- Marín, R.; Muñoz-Guerra, S. J. Appl. Polym. Sci. 2009, 114, 3723-3736. Noordover, B.A.J.; van Staalduinen, V.G.; Duchateau, R.; Koning, C.E.; van Benthem, R.A.T.M.; Mak, M.; Heise, A.; Frissen, A.E.; van Haveren, J. Biomacromolecules 2006, 7, 3406-3416. Okada, M. Prog. Polym. Sci. 2001, 26, 67-104. Pillai, C.K.S. Design. Monom. Polym. 2010, 13, 87-121. Quintana, R.; Martínez de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. J. Polym. Sci., Polym. Chem. 2011, 49, 2252–2260. Rodríguez-Galán, A.; Bou, J.; Muñoz-Guerra, S. J. Polym. Sci., Polym. Chem. 1992, 30, 713–721. Sablong, R.; Duchateau, R.; Koning, C.E.; de Wit, G.; van Es, D.; Koelewijn, R.; van Haveren, J. Biomacromolecules 2008, 9, 3090-3097. Storbeck, R.; Rehahn, M.; Balluff, M. Makromol. Chem. 1993, 194, 53-64. Thiem, J.; Lüders, H. Polym. Bull. 1984, 11, 365-369. Varela, O. Adv. Carbohyd. Chem. Biochem. 2000, 55, 137-174. Varma, A.J.; Kennedy, J.F.; Galgali, P.Q. Carbohyd. Polym. 2004, 56, 429-445. Wool, R.; Sun, S. Biobased Polymers and Composites; Academic Press: New York, 2005. Zamora, F.; Hakkou, K.; Alla, A., Espartero, J.L.; Muñoz-Guerra, S.; Galbis, J.A. J. Polym. Sci., Polym. Chem. 2005, 43, 6394–6410.
Chapter 3 114
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 115 3.4. High Tg bio-based aliphatic polyesters from bicyclic D-mannitol Summary: The carbohydrate-based diol 2,4:3,5-di-O-methylene-D-mannitol (Manx) has been used to obtain aliphatic polyesters. Manx is a symmetric bicyclic compound consisting of two fused 1,3-dioxane rings and bearing two primary hydroxyl groups. In terms of stiffness it is comparable to the widely known isosorbide but it affords the additional advantages of being much more reactive in polycondensation and capable of producing stereoregular polymers with fairly high molecular weights. A fully bio-based homopolyester (PManxS) has been synthesized by polycondensation in the melt from dimethyl succinate and Manx. The high thermal stability of PManxS, its relatively high glass-transition temperature (Tg= 68 ºC) and elastic modulus, and its enhanced sensitivity to the action of lipases point to PManxS as a polyester of exceptional interest for those applications where biodegradability and molecular stiffness are priority requirements. In addition, random copolyesters (PBxManxyS) covering a broad range of compositions have been obtained using mixtures of Manx and 1,4-butanediol in the reaction with dimethyl succinate. All PBxManxyS were semicrystalline and displayed Tg values from -29 to +51 ºC steadily increasing with the content in Manx units. The stress-strain behavior of these copolyesters largely depended on their content in Manx and they were enzymatically degraded faster than PBS. Publication derived from this work: Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Muñoz-Guerra, S. Biomacromolecules 2013, 14, 781-793.
Chapter 3 116 3.4.1. Introduction Due to increasing concerns on sustainable development and minimizing the impact of materials on the environment, biodegradable polymers have attracted a great deal of interest in the last decades (Engelberg and Kohn, 1991; Albertsson and Varma, 2002; Okada, 2002; Nair and Laurencin, 2007). Aliphatic polyesters such as poly(butylene succinate) (PBS), poly(L-lactic acid), poly(ε-caprolactone) and poly(3- hydroxy butyrate) among others, constitute a distinguished bunch of biodegradable polymers that are commercially available in several forms. Some of these aliphatic polyesters have found intensive use in a broad variety of medical applications such as bioresorbable surgical sutures, prosthesis, dental implants, bone screw and plates for temporary internal fracture fixation, and controlled drug delivery systems (Wang et al., 2012; Coutinho et al., 2012). Nowadays PBS is receiving exceptional attention as a polymeric material suitable for replacing conventional commodity plastics in injectionmolded articles (cutlery, brushes), tubular films (composting bags, shopping bags), flexible packaging and food tray since it has a melting temperature similar to that of low density polyethylene, satisfactory mechanical properties and can be processed with the conventional equipments commonly used for polyolefins (Papageorgiou and Bikiaris, 2007; Gualandi et al., 2012). However, PBS has a glass-transition temperature (Tg) of -37 ºC, not high enough for its use in rigid packaging and cosmetic and beverage bottles where stiffness and thermal resistance are priority requisites. In this regard, various approaches such as blending and copolymerization with aromatic polyesters as poly(butylene terephthalate) (PBT) and poly(ethylene terephthalate) (PET) have been recently explored (Deng et al., 2004; Li et al., 2012). The interest for cyclic monomers arises from their capacity for adding stiffness to the polymer chain with the subsequent increase in Tg; the disadvantage of using aromatic units for such purpose is that they are originated from fossil feedstocks and are reluctant to biodegradation. The use of carbohydrate-based monomers with a cyclic structure constitute a scarcely explored approach for the preparation of renewable aliphatic polyesters with improved physical properties, especially those related to polymer chain stiffness (Galbis and García-Martín, 2010; Gandini, 2011). Moreover, carbohydrate-derived polycondensates typically display enhanced hydrophylicity, lower toxicity and higher susceptibility to hydrolytic degradation and biodegradation than those coming from petrochemical feedstocks. Thus 2,3-O-isopropylidene-L-tartaric acid (Dhamaniya and Jacob, 2010 and 2012) and 2,5-furandicarboxylic acid (Gandini et al., 2009; Gomes et al.,
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 117 2011; Wu et al., 2012a) have been polymerized with several diols to obtain bio-based aliphatic cyclic polyesters. 1,4:3,6-Dianhydro-D-glucitol, known as isosorbide, which is prepared by dehydration of D-glucose coming from cereal starch, is the only bicyclic carbohydrate-based monomer industrially available today. Isosorbide, along with their two less accessible stereoisomers isomannide and isoidide, are currently drawing an enormous interest in the polymer science field as bio-based monomers able to provide enhanced stiffness into the polymer chain they are incorporated (Fenouillot et al., 2010). These three isohexides are composed of two fused tetrahydrofuran rings, with two secondary hydroxyl groups remaining free for reaction. Due to their fused bicyclic structure, 1,4:3,6-dianhydrohexitols and their diacid derivatives are able to increase the glass-transition temperature of common aliphatic polyesters (Braun and Bergmann, 1992; Okada et al., 1996 and 2000; Noordover et al., 2006; Wu et al., 2012b). Nevertheless, the main shortcoming of isosorbide and its isomers is the limited reactivity of their secondary hydroxyl groups; in fact, this feature seriously hampers the polycondensation reaction in the melt, so aliphatic polyesters from 1,4:3,6-dianhydrohexitols and aliphatic dicarboxylic acids or dicarboxylic esters obtained by this method display rather limited molecular weights (Braun and Bergmann, 1992; Noordover et al., 2006). Higher molecular weights are achievable via polycondensation with aliphatic dicarboxylic chlorides (Okada et al., 1996 and 2000), but this method is not appropriate for industrial application (Fradet and Tessier, 2003). Recently, we have reported the synthesis and characterization of aliphatic polyesters made from dimethyl 2,3:4,5-di-O-methylene-galactarate (Galx), which is a bicyclic monomer obtained by internal acetalization of galactaric acid (Lavilla et al., 2011 -Subchapter 3.2- and 2012a -Subchapter 3.3-). This monomer has been shown to be very suitable to prepare aliphatic polyesters by polycondensation with aliphatic diols in the melt, since it is able to react at a rate similar to that of other acyclic conventional monomers. At difference with dianhydrohexitols, Galx is composed of two independent 1,3-dioxolane rings linked by a single C-C bond, and thence they were found to confer lower rigidity to the polyester backbone than a fused bicyclic structure is able to do. To our knowledge, the bicyclic diester Galx is the only bicyclic diacetalized carbohydratebased monomer that has been used up to date to obtain aliphatic polyesters. In this work bicyclic 2,4:3:5-di-O-methylene-D-mannitol, abbreviated as Manx, has been used to obtain aliphatic polyesters. Manx is obtained by internal acetalization of D-mannitol in a similar way to the bicyclic diester Galx is obtained from mucic acid.
Chapter 3 124 Table 3.9. Molar composition, molecular weight, microstructure and contact angle. Copolyester Molar composition Molecular weight Microstructure Contact anglee Yield (%) Feed Copolyestera Dyads Number Average Sequence Lengths Randomnessd XB XManx XB XManx [η]b Mnc Mwc Ðc BB BManx/ ManxB Manx Manx nB nManx R θwater (deg) PBS 90 100 0 100 0 0.97 19,500 49,100 2.5 113±4 PB95Manx5S 87 95 5 94.3 5.7 0.90 19,900 47,800 2.4 87.9 11.4 0.7 16.4 1.1 0.96 114±5 PB90Manx10S 89 90 10 88.7 11.3 0.72 16,900 39,100 2.3 78.0 20.5 1.5 8.6 1.1 0.99 114±4 PB80Manx20S 88 80 20 80.7 19.3 0.69 14,600 35,200 2.4 66.3 30.8 3.0 5.3 1.2 1.03 117±3 PB70Manx30S 86 70 30 69.1 30.9 0.70 14,200 35,000 2.5 47.0 44.3 8.8 3.1 1.4 1.04 113±2 PB50Manx50S 87 50 50 51.1 48.9 0.61 13,900 33,400 2.4 26.1 50.0 24.0 2.0 2.0 1.00 114±4 PB30Manx70S 89 30 70 29.7 70.3 0.60 13,500 33,500 2.5 8.4 43.4 48.2 1.4 3.2 1.03 116±8 PManxS 85 0 100 0 100 0.56 13,300 30,800 2.3 74±7 a a Molar composition determined by integration of the 1H NMR spectra. b Intrinsic viscosity in dL·g-1 measured in chloroform at 25 ºC. c Number and weightaverage molecular weights in g·mol-1 and dispersities measured by GPC in HFIP against PMMA standards. d Randomness index of copolyesters statistically calculated on the basis of the 13C NMR analysis. e Contact angle measured after 45 seconds of dropping 1.5 µL of water.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 125 Both the 1H and 13C NMR analysis ascertained the chemical constitution and composition of the polyesters. They are detailed in Section 3.4.2, and a selection of the illustrative spectra is provided in Annex C. The assignment of the signals arising from the sugar units was supported by COSY and HETCOR spectra, which are included in Annex C. Integration of the proton signals arising from B and Manx units led to quantify the composition of the copolyesters in such units, which appeared to be very close to those of their respective feeds (Table 3.9). The microstructure of PBxManxyS copolyesters was determined by 13C NMR analysis. As it is depicted in Figure 3.24, the 13C NMR spectra of copolyesters show one of the succinic signals with resolution enough as to make possible the elucidation of the copolyester microstructure as far as distribution of B and Manx units along the copolyester chain is concerned. As a consequence of the occurrence of different dyad types (BB, BManx, ManxB and ManxManx), this signal appears split into four peaks that spread within the 29.2-29.7 ppm chemical shift interval. By integration of these peaks, the dyad contents, the number-average sequence lengths and the degree of randomness were estimated (Table 3.9), leading to the conclusion that the microstructure of the copolyesters was random in all cases. Figure 3.24. 13C NMR signals used for the microstructure analysis of PBxManxyS copolyesters with indication of the dyads to which they are assigned.
Chapter 3 126 The solvent affinity of Manx containing polysuccinates was similar to that displayed by PBS. All they continue to be non-soluble in water but readily soluble in chloroform. The contact angle between water and polyester films was measured to estimate their hydrophilicity (Table 3.9); no significant differences were observed in the θwater of PBS (~113º) with the incorporation of Manx units except for the case of the homopolyester where the contact angle decreased to near 75º indicating the much higher hydrophilic character of this compound. 3.4.3.2. Thermal and mechanical properties The thermal behavior of PManxS homopolyester and PBxManxyS copolyesters has been comparatively studied by TGA and DSC; the thermal parameters resulting from these analyses are given in Table 3.10, where the corresponding data for the homopolyester PBS are also included for reference. First, the thermal stability was evaluated by TGA under inert atmosphere. The TGA traces of PManxS and PBS homopolyesters together with their derivative curves are shown in Figure 3.25a, and those recorded from the whole set of PBxManxyS copolyesters are comparatively depicted in Figure 3.25b. Thermal decomposition of PManxS occurs in a single stage with maximum rate taking place at 413 ºC (maxTd), and only 5% of the initial weight remains at 600 ºC; the maximum decomposition rate of PBS homopolyester occurs at 384 ºC (maxTd), the residual weight left upon heating at 600 ºC being 4%. Decomposition of PBxManxyS copolyesters takes place also in one stage, at temperatures ranging between maxTd of their two corresponding homopolyesters PBS and PManxS and steadily increasing as butanediol units are replaced by Manx units. The valuable conclusion drawn from this comparative thermogravimetric study is that the insertion of Manx in aliphatic polyesters instead of reducing their decomposition temperatures contributes significantly to increasing their thermal stability, which can broaden even more the application window of aliphatic polyesters. Since the melting in these copolyesters decreases with composition, a wider range of temperatures between melting and decomposition exists, allowing a more comfortable melt processing.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 127 Table 3.10. Thermal properties. Copolyester TGA DSC First heatinge Coolinge Second heatinge ºT5%a Tdb W c Tgd Tm ΔHm Tc ΔHc Tc ΔHc Tm ΔHm (ºC) (ºC) (%) (ºC) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) PBS 320 384 4 -37 111/114 (113) 75.0 (103.1) 81 60.7 - - 113 61.1 PB95Manx5S 320 386 6 -29 107 (107) 62.9 (69.3) 64 59.5 - - 107 60.7 PB90Manx10S 322 387 5 -20 94 (93) 45.3 (48.4) 42 16.9 35 18.7 95 36.2 PB80Manx20S 323 410 4 -8 50/74/84 (51/82) 44.7 (47.3) - - 59 16.0 85 16.1 PB70Manx30S 322 413 4 3 52/64 (49/63) 30.1 (38.7) - - - - - - PB50Manx50S 323 413 4 26 64/85 (62/80) 10.1 (10.9) - - - - - - PB30Manx70S 321 413 5 51 91 (90) 3.8 (4.6) - - - - - - PManxS 320 413 5 68 - (125)f - (14.0)f - - - - - - aTemperature at which 5% weight loss was observed. bTemperature for maximum degradation rate. cRemaining weight at 600 ºC. dGlasstransition temperature taken as the inflection point of the heating DSC traces of melt-quenched samples recorded at 20 ºC·min-1. eMelting (Tm) and crystallization (Tc) temperatures, and melting (ΔHm) and crystallization (ΔHc) enthalpies measured by DSC at heating/cooling rates of 10 ºC·min-1 of powdered samples coming directly from synthesis (without parentheses) and of films prepared by casting from solution (in parentheses). fAfter annealing for 1h at 60 ºC, Tm increased to 127 ºC and ΔHm to 21.5 J·g-1.
Chapter 3 128 Figure 3.25. (a) TGA traces (solid lines) and derivative curves (dashed lines) of PBS and PManxS homopolyesters. (b) TGA traces of PBxManxyS copolyesters. Another thermal property of prime importance in connection with the potential application of these polyesters is the glass-transition temperature (Tg). For instance, an increase in the Tg of aliphatic polyesters such as PBS could open their use in the rigid packaging field. The DSC analyses revealed that the Tg steadily increased as butanediol was replaced by Manx units going from -29 ºC for PB95Manx5S up to 51 ºC for PB30Manx70S (Figure 3.26a). These results are fully consistent with the Tg values of -37 ºC and 68 ºC observed for the PBS and PManxS homopolyesters, respectively, provided that the microstructure of the copolyesters is random. Such a remarkable positive effect of the incorporation of Manx on Tg makes this compound a bio-based comonomer very suitable for the preparation of aliphatic polyesters with enhanced Tg.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 129 Figure 3.26. (a) DSC heating traces of PBxManxyS copolyesters quenched from the melt. (b) DSC heating traces of PBxManxyS copolyesters coming directly from synthesis. DSC traces of polyesters coming directly from synthesis are depicted in Figure 3.26b. All PBxManxyS copolyesters as well as PBS homopolyester gave heating traces with melting endotherms indicating that they were semicrystalline, whereas PManxS appeared to be amorphous. Comparison of melting temperatures and enthalpies of PBxManxyS copolyesters with that of PBS leads to the conclusion that the insertion of Manx units gives place to a significant decrease in both Tm and ΔHm. The Tm did not follow a continuous trend with composition but it fell into a minimum for copolyester compositions around 30% of Manx units (Figure 3.27). This behavior suggested the occurrence of two different crystal structures depending on the unit, butanediol or Manx, that is predominant in the copolyester. To complement DSC data, powder X-ray diffraction analyses were performed for PBxManxyS copolyesters as well as for PBS and PManxS homopolyesters coming directly
Chapter 3 130 from synthesis. The X-ray diffraction profiles recorded from powders are compared in Figure 3.28, and the most prominent Bragg spacings present therein are listed in Table 3.11. Essentially the same diffraction pattern regarding both spacing and intensities is shared by PBS homopolyester and copolyesters containing up to 30% of Manx units. This pattern is distinguished by the presence of three sharp strong reflections at 4.5, 4.1 and 3.9 Å indicating that the monoclinic crystal structure of PBS (Ihn et al., 1995) is retained in such copolyesters. It must be assumed that the bicyclic structure in these copolyesters is segregated from the crystal lattice which is made of homogeneous sequences of butylene succinate units. Since the average length of these sequences diminishes with the content in Manx, the crystallite thickness decreases and consequently Tm goes down. Conversely, both PB50Manx50S and PB30Manx70S produced a scattering profile with a clearly different shape where the three most prominent reflections corresponding now to 4.7, 4.5 and 4.0 Å which also coincides with that exhibited by the PManxS homopolyester. The larger values observed for these spacings, which presumably arise from the interplanar distances that are defined by the side-by-side packing of the chains, are consistent with the bigger size of Manx compared to butanediol. In addition, several weaker reflections are observed at higher spacings in the 6-8 Å range, which might be associated to the axial repeat of the polyester chain. Apparently a second crystal form different from that of PBS, which is able to accommodate the bicyclic structure in the lattice, seems to be adopted by both the homopolyester and the copolyesters containing major amounts of Manx. In this case the butylene succinate would be the unit that is segregated from the crystalline phase; accordingly, the Tm in the Manx-enriched interval should be expected to increase with the content in Manx, as it is experimentally observed. Figure 3.27. Glass-transition and melting temperatures versus composition plot for PBxManxyS copolyesters.
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 131 Table 3.11. Powder X-ray diffraction data and mechanical properties. Copolyester X-ray diffraction data Mechanical Properties d a (Å) Xcb Elastic modulus Tensile strength Elongation at break (MPa) (MPa) (%) PBS 4.5 m 4.1 m 3.9 s 3.4 w 3.1 w 2.6 w 0.52 351±11 20±2 19±4 PB95Manx5S 4.5 m 4.1 m 3.9 s 3.4 w 3.1 w 2.6 w 0.48 355±13 22±2 14±3 PB90Manx10S 4.5 m 4.1 m 3.9 s 3.4 w 3.1 w 2.6 w 0.45 372±14 22±2 9±2 PB80Manx20S 4.5 m 4.1 m 3.9 s 3.4 w 3.1 w 2.6 w 0.43 403±15 24±2 7±3 PB70Manx30S 4.5 m 4.1 m 3.9 s 3.4 w 3.1 w 2.6 w 0.37 428±12 24±3 5±2 PB50Manx50S 7.0 w 6.2 w 5.6 w 4.7 s 4.5 m 4.2 w 4.0 m 3.4 w 3.1 w 2.9 w 2.8 w 0.28 471±15 26±2 3±1 PB30Manx70S 7.0 w 6.2 w 5.6 w 4.7 s 4.5 m 4.2 w 4.0 m 3.4 w 3.1 w 2.9 w 2.8 w 0.17 515±14 28±3 2±1 PManxS (7.0 w) (6.2 w) (5.6 w) (4.7 s) (4.5 m) (4.2 w) (4.0 m) (3.4 w) (3.1 w) (2.9 w) (2.8 w) (0.31) 573±15 31±3 2±1 a Bragg spacings measured in powder diffraction patterns. Intensities visually estimated as follows: m, medium; s, strong; w, weak. Data from samples coming directly from synthesis (without parentheses) and from PManxS film annealed for 1h at 60 ºC (in parentheses). b Crystallinity index calculated as the quotient between crystalline area and total area quantified in the X-ray diffraction profiles using the PeakFit v4.12 software.
Chapter 3 132 Figure 3.28. Powder WAXD profiles of PBS, PManxS and PBxManxyS copolyesters coming directly from synthesis. *The PManxS profile at the bottom was obtained from a film casted from chloroform that was annealed for 1 h at 60 ºC. Crystallinity was found however to decay steadily along the whole series as the content in Manx units increased, to the point that no discrete scattering could be detected for the homopolyester PManxS coming directly from synthesis. These observations are in full agreement with DSC results and reveal the depressing influence that the bicyclic structure has on crystallizability, which may be due to the opposite effect that it has on Tm and Tg. Increasing amounts of Manx led to polyesters with narrower crystallization window (Tm-Tg) and therefore exhibiting less crystallinity because the melt becomes frozen before the chains can crystallize upon supercooling. Nevertheless, PManxS was able to crystallize by casting from a chloroform solution. After annealing the cast film showed a melting peak at 127 ºC with an enthalpy of 21.5 J·g-1, and produced a WAXD profile identical to those recorded from the enriched Manx copolyesters (Figure 3.28). These results suggest that rather than hampering the close side-by-side packing of the polyester chains required to form the crystal lattice, the main repressing effect of Manx on crystallization should be due to the restricted mobility that the presence of the bicyclic structure confers to the polyester chain. This is in full agreement with the high delay that is observed in the isothermal crystallization time of PBxManxyS for slight rises in Manx contents, more than tenfold for a 5% of increment in Manx (see below).
Aliphatic polyesters from cyclic acetalized carbohydrate-based monomers 133 Figure 3.29. Isothermal crystallization of PBS, PB95Manx5S and PB90Manx10S at the indicated temperatures. Relative crystallinity versus time plot (a) and log-log plot (b). The cooling DSC traces obtained from molten samples revealed that PBxManxyS copolyesters with molar contents in Manx below or equal to 10%, as well as PBS homopolyester, were able to crystallize from the melt. After crystallization upon cooling at a constant rate of 10 ºC·min-1, PBS homopolyester and PB95Manx5S copolyester recovered about 80-97% of their initial crystallinity and displayed almost the same melting temperatures. Conversely PB90Manx10S copolyester crystallized under such conditions recovering only 37% of its original crystallinity. However, it presented cold crystallization at the second heating run with recovering about 80% of the initially crystallized material, and melting at nearly the same initial temperature. Given the relevance of the ability for crystallizing from the melt regarding polymer processing, the isothermal crystallization of PB95Manx5S and PB90Manx10S copolyesters and PBS homopolyester was comparatively studied in the 70-100 ºC interval. Unfortunately, not all of them could be compared at exactly the same crystallization temperature due to large differences in crystallization rates displayed by them; nevertheless, crystallization conditions were chosen as close as possible in order to be able to draw out meaningful conclusions. Avrami log[-ln(1-Xc)] versus log(t-t0) data plots and the evolution of the relative crystallinity, Xc, versus
Chapter 5 236 Figure 5.19. Compared 1H NMR after incubation with water at pH 2.0 at 80 ºC and initial spectra of PB50Galx50T (a) and PBT50Galx50 (b) copolyesters, and representation of Galx content vs. incubation time. SEM analysis revealed that upon incubation PBT spherulites showed cracks in a low proportion whereas spherulites of incubated PGalxT homopolyester appeared frequently cracked (Annex E); with regard to PB50Galx50T copolyester, the initial surface displayed also very apparent physical alterations (Figure 5.17), showing that hydrolytic degradation was largely enhanced by the presence of galactitol units. Both sample weighting and SEM observations indicated that PB50Galx50T copolyester became
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 237 apparently degraded in a higher degree than PGalxT homopolyester despite its lesser content in galactitol. The reason for such behavior may be the lower glass-transition temperature displayed by the copolyester. 5.3.3.2. PBT copolyesters containing galactarate units To investigate the influence of the incorporation of dimethyl 2,3:4,5-di-O- methylene-galactarate on degradation of PBT, a comparative study of the two homopolyesters PBT and PBGalx and the intermediate PBT50Galx50 copolyester was carried out under the same conditions used in the study of the PBT copolyesters containing galactitol units described above. 5.3.3.2.1. Hydrolytic degradation and biodegradation at pH 7.4 at 37 ºC The changes taking place in sample weight and molecular weight of PBT, PBT50Galx50 and PBGalx polyesters at increasing incubation times in aqueous pH 7.4 buffer at 37 ºC, with and without porcine pancreas lipase, are presented in Figure 5.20. The weight that was lost upon 8 weeks of incubation in the presence of lipase was about 20% and 35% for PBT50Galx50 and PBGalx, respectively; whereas only 5% and 7% was lost upon incubation of the same polyesters in the absence of enzymes, respectively. However, the sample weight and molecular weight of PBT homopolyester was maintained unchangeable through the whole process. On the contrary, a slight decay in Mw and Mn was observed for both PBT50Galx50 and PBGalx polyesters when incubated in the absence of enzymes. Furthermore, when these two polyesters were incubated with lipase, Mw and Mn decayed substantially, the more pronounced change being observed for the homopolyester in agreement with the expected ability of galactarate units for making PBT biodegradable. Melting enthalpies of PBT samples upon incubation were nearly the same than that of the initial sample according to the inexistence of degradation. However, it was noteworthy to observe that the crystallinity of both PBT50Galx50 and PBGalx increased as degradation proceeded, and that the raise in crystallinity was higher when samples were incubated in the presence of lipase (Figure 5.16). The SEM micrographs of PBT, PBT50Galx50 and PBGalx polyesters before and after incubation are shown in Figures 5.21 and 5.22 and in Annex E. This analysis confirmed that the surface of PBT remained unaltered after being incubated either in the presence or the absence of lipases. The results obtained in the parallel analysis carried out on PBT50Galx50 and PBGalx were
Chapter 5 238 clearly different. SEM micrographs of PBT50Galx50 showed a certain grade of destruction of the polymer surface when it was incubated in the presence of lipase, a feature that was absent when incubation was carried out without enzymes. Regarding PBGalx, the initial surface did not display any feature of crystalline morphology, and after incubation with lipase, plenty of curved lines appeared delineating what seems to be a spherulitic texture. The conclusion derived from these observations is that degradation of PBT50Galx50 and PBGalx polyesters under physiological conditions was greatly enhanced by the action of enzymes, a property that points to dimethyl 2,3:4,5-di-O-methylene-galactarate as a potential comonomer for obtaining biodegradable PBT copolyesters. Figure 5.20. Degradation of PBT, PBT50Galx50 and PBGalx. Remaining weight (a) and molecular weight (b) vs. degradation time with (solid symbols) and without (empty symbols) porcine pancreas lipase at pH 7.4 at 37 ºC. Remaining weight (a’) and molecular weight (b’) vs. degradation time at pH 2.0 at 80 ºC.
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 239 Figure 5.21. SEM micrographs of PBGalx. Initial sample (a). After incubation at pH=2.0 at 80 ºC for 7 days (b). After incubation at pH=7.4 at 37 ºC for 56 days with (c) and without (d) lipase from porcine pancreas. Figure 5.22. SEM micrographs of PBT50Galx50. Initial sample (a). After incubation at pH=2.0 at 80 ºC for 56 days (b). After incubation at pH=7.4 at 37 ºC for 56 days with (c) and without (d) lipase from porcine pancreas.
Chapter 5 240 5.3.3.2.2. Hydrolytic degradation at pH 2.0 at 80 ºC The changes taking place in sample weight and molecular weight of PBT, PBT50Galx50 and PBGalx polyesters upon incubation at pH 2.0 at 80 ºC are depicted in Figure 5.20. The profiles obtained for PBT indicate again that there was no loss of mass and that the decrease in molecular weight was practically negligible. Also the SEM analysis revealed that PBT remained almost unaltered after incubation, only showing very few small cracks and confirming the reluctance of this polyester to be hydrolyzed. On the contrary, PBGalx homopolyester showed a drastic loss of mass, about 85% upon only 2 weeks of incubation and achieving a total weight loss of almost 100% after 8 weeks, as well as a significant decrease in both Mw and Mn, which fell down to final values of 9,000 and 4,000 g·mol-1, respectively. Neither SEM observations nor DSC measurements could be carried out after the whole period of incubation since no sample remained; however, SEM observations of PBGalx incubated for one week (Figure 5.21) confirmed that degradation of PBGalx under these conditions had already taken place in some extend in such a short period of time. PBT50Galx50 copolyester degradation results were intermediate between those of their respective two parent homopolyesters, losing about 70% of the initial mass and showing Mw and Mn values of 16,000 and 7,000 g·mol-1 after 8 weeks of incubation. Furthermore, SEM micrographs (Figure 5.22) and DSC measurements (Figure 5.16 and Annex E) showed that the initial smooth surface of PBT50Galx50 was almost destroyed and crystallinity increased after incubation. In order to deep insight the degradation of the polyester chain at the molecular level a NMR study was undertaken. 1H NMR spectra of the products released to the aqueous medium and the residual material resulting after 8 weeks of incubation in acidic water are depicted in Figures 5.18, 5.19 and 5.23. As expected, PBT spectra of both the aqueous medium and the residual material did not show differences with the initial spectra. By contrast, the PBGalx spectrum of the incubation media showed signals corresponding to the released 2,3:4,5-di-O-methylene-galactaric acid and 1,4-butanediol, whereas the NMR spectrum of the residual material revealed no changes in the diacetal structure but showed -CH2OH signals of terminal groups of the polyester, indicating a decrease in the molecular weight. On the other hand, the NMR analysis of the incubation medium of PBT50Galx50 showed only signals characteristic of 2,3:4,5-di-O-methylene- galactaric acid and 1,4-butanediol, whereas the spectrum of the residual polymer showed a progressive decrease in galactaric content. In fact, integration of galactaric and terephthalic signals indicated that the residual material contained only 18% of the
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 241 galactaric units after 2 weeks of incubation and almost nothing after 8 weeks. The spectrum showed also prominent -CH2OH signals due to a notable increase of terminal groups. It was concluded therefore that after the whole incubation period, the residual material of PBT50Galx50 copolyester essentially consisted of 1,4-butyleneterepthalate oligomers. Figure 5.23. 1H NMR spectra in D2O of the products released to the aqueous medium after incubation of PBGalx (a) and PBT50Galx50 (b) at pH 2.0 at 80 ºC for 56 days. 5.3.3.3. Stability of the diacetal structure PBT copolyesters containing bicyclic acetalized units, as well as their respective sugar-based homopolyesters, have shown to be degraded by water through splitting of the ester group whereas the acetal group remains stable against hydrolysis (Figures 5.18, 5.19 and 5.23). This result is rather striking because the acetal group is known to be sensitive to acidic conditions (Smith and March, 2007) and the opening of the dioxolane
Chapter 5 242 rings might be expected to happen in some extend. A NMR study was undertaken in order to check the stability of the acetal groups of the bicyclic dioxolane structure, especially compared with the ester group. The dimethyl ester of the bicyclic acetalized galactaric acid was then incubated in aqueous buffer at pH 2.0, 7.4 and 10.5. The degradation products detected by NMR are depicted in Scheme 5.3 and the recorded spectra are provided in Annex E. After 12 weeks of incubation, degradation of the ester groups happened almost quantitatively in both acid and basic media. Upon incubation at pH 7.4, most of the bicyclic acetalized monomer was in the dicarboxylic form, a fourth part was as monoester form and only around 5% still remained as diester. What was really remarkable in this study is that the acetal structure stayed unchanged after the treatment under any applied conditions. Thus the stability against hydrolysis observed for the dioxolane rings of the sugar-based units contained in the PBT copolyesters was corroborated. Scheme 5.3. Resulting products of the hydrolysis of dimethyl 2,3:4,5-di-O-methylene-galactarate at pH 2.0, pH 7.4 and pH 10.5 after 90 days. 5.3.4. Conclusions PBT copolyesters in which the terephthalate units were replaced by dimethyl 2,3:4,5-di-O-methylene-galactarate appeared to be sensitive to enzymatic action showing a significant degradability under physiological conditions in the presence of lipases. When these polyesters were incubated at 80 ºC in acidic medium, hydrolytic degradation occurred so fast that no sample remained after incubation for a few weeks. Conversely, PBT analogs based on 2,3:4,5-di-O-methylene-galactitol were not degraded under
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 243 physiological conditions either with or without the concourse of enzymes, although they displayed enhanced hydrodegradability compared to PBT when incubated at 80 ºC in acidic medium. In all cases, degradation proceeded by splitting of the relatively weak ester group associated to the sugar moiety and without modification of the diacetal structure. The outstanding conclusion derived from this study is that whereas the incorporation of 2,3:4,5-di-O-methylene-galactitol units in PBT leads to copolyesters with controlled hydrodegradability, the incorporation of 2,3:4,5-di-O-methylene-galactarate units is a suitable option for obtaining rapidly hydrodegradable and biodegradable PBT copolyesters. 5.3.5. References Alla, A.; Hakkou, K.; Zamora, F.; Martínez de Ilarduya, A.; Galbis, J.A.; Muñoz-Guerra, S. Macromolecules 2006, 39, 1410-1416. Burden, I.J.; Stoddart, J.F. J. Chem. Soc. Perk. T. 1 1975, 675-682. Butler, K.; Lawrance, R.L.; Stacey, M. J. Chem. Soc. 1958, 740-743. Fenouillot, F.; Rosseau, A.; Colomines, G.; Saint-Loup, R.; Pascault, J.-P. Prog. Polym. Sci. 2010, 35, 578-622. Gallucci, R.R.; Patel, B.R. Poly(butylene terephthalate). In: Scheirs, J.; Long, T.E.; Eds. Modern Polyesters, Chemistry and Technology of Polyesters and Copolyesters; John Wiley & Sons: Chichester, 2004, pp. 293-321. Galbis, J.A.; García-Martín, M.G. Sugars as Monomers. In: Belgacem, M.N.; Gandini, A.; Eds. Monomers, Polymers and Composites from Renewable Resources; Elselvier: Oxford, 2008, 89-114. Gomes, M.; Gandini, A.; Silvestre, A.J.D.; Reis, B. J. Polym. Sci., Polym. Chem. 2011, 49, 3759-3768. Kiely, D.E.; Chen, L.; Lin, T.-H. J. Polym. Sci., Polym. Chem. 2000, 38, 594-603. Kijchavengkul, T.; Auras, R.; Rubino, M.; Alvarado, E.; Montero, J.C.R.; Rosales, J.M. Polym. Degrad. Stabil. 2010, 95, 99-107. (2010a) Kijchavengkul, T.; Auras, R.; Rubino, M.; Selke, S.; Ngouajio, M.; Fernandez, R.T. Polym. Degrad. Stabil. 2010, 95, 2641-2647. (2010b) Kulshrestha, A.S.; Gao, W.; Gross, RA. Macromolecules 2005, 38, 3193-3204. Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. Biomacromolecules 2011, 12, 2642-2652. -Subchapter 3.2-
Chapter 5 244 Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. Polymer 2012, 53, 3432-3445. -Subchapter 5.2- Nakajima-Kambe, T.; Ichihashi, F.; Matsuzoe, R.; Kato, S.; Shintani, N. Polym. Degrad. Stabil. 2009, 94, 1901-1905. Rychter, P.; Kawalec, M.; Sobota, M.; Kurcok, P.; Kowalczuk, M. Biomacromolecules 2010, 11, 839-847. Smith, M.B.; March, J. March’s Advanced Organic Chemistry; Wiley: Hoboken, 2007, pp. 523,1270. Tripathy, A.R.; MacKnight, W.J.; Kukureka, S.N. Macromolecules 2004, 37, 6793-6800. Yokoe, M.; Aoi, M.; Okada, M. J. Polym. Sci., Polym. Chem. 2005, 43, 3909-3919. Zaliz, C.L.R.; Varela, O. Tetrahedron Asymm. 2005, 16, 97-103.
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 245 5.4. Bio-based aromatic polyesters from a bicyclic diol derived from D-mannitol Summary: 2,4:3,5-di-O-methylene-D-mannitol, abbreviated as Manx, is a D-mannitol- derived compound with the secondary hydroxyl groups acetalized with formaldehyde. The bicyclic structure of Manx consists of two fused 1,3-dioxane rings, with two primary hydroxyl groups standing free for reaction. A homopolyester made of Manx and dimethyl terephthalate, as well as a set of copolyesters of poly(butylene terephthalate) (PBT) in which 1,4-butanediol was replaced by Manx up to 50% were synthesized and characterized. The polyesters had Mw in the 30,000-52,000 g·mol-1 range and a random microstructure, and were thermally stable up to nearly 370 ºC. They displayed outstanding high Tg with values from 55 to 137 ºC which steadily increased with the content in Manx. Copolyesters containing up to 40% of Manx were semicrystalline and adopted the crystal structure of PBT. Their stress-strain parameters were sensitively affected by the presence of carbohydrate-based units with elongation at break decreasing but tensile strength and elastic moduli steadily increasing with the degree of replacement. Publication derived from this work: Lavilla, C.; Martínez de Ilarduya, A.; Alla, A.; García-Martín, M.G.; Galbis, J.A.; Muñoz- Guerra, S. Macromolecules 2012, 45, 8257-8266.
Chapter 5 252 5.4.3. Results and discussion 5.4.3.1. Monomer synthesis and conformation The bicyclic diol 2,4:3,5-di-O-methylene-D-mannitol (Manx) was prepared from D- mannitol following the chemical synthetic route depicted in Scheme 5.4. Primary hydroxyl groups of D-mannitol were first protected as benzoyl esters and then the addition of paraformaldehyde led to cyclic acetalization of the secondary hydroxyl groups in the 2,4:3,5 arrangement with the two fused 1,3-dioxane rings. 1H and 13C NMR spectra of Manx are shown in Annex F and detailed in Section 5.4.2. The coupling between the two protons of the methylene acetalic group with a J= -6.2 Hz confirms the presence of the two fused six membered rings and then the 2,4:3,5 arrangement of the bicyclic structure (Burden and Stoddart, 1975). It should be stressed that the synthesis route used for the preparation of Manx, although effective for the purpose of the work, is not environmentally friendly. Obviously, an alternative synthetic route or an improvement of the existing one will need to be developed to make this monomer of fully practical use and to render benefits to the environment. Scheme 5.4. Synthesis of 2,4:3:5-di-O-methylene-D-mannitol. This bicyclic diol Manx possesses a twofold axis of symmetry and therefore its two hydroxyl groups will display the same reactivity in the polycondensation reaction producing regioregular polymer chains. Since the two rings are fused, a fairly high degree of rigidity should be in principle expected for Manx although it could be largely relieved if
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 253 interconversion between possible conformers takes place. The rigidity of the monomer is critical for the preparation of polymers with improved thermal and mechanical properties. To appraise the effect that this novel diol monomer will have on the stiffness of the polyesters thence produced, a conformational analysis of 1,6-di-O-benzoyl-2,4:3,5-di-O- methylene-D-mannitol was undertaken. This diester is an adequate model compound for aromatic polyesters and copolyesters made from Manx. The two most probable conformations (Mills, 1955) for the 2,4:3,5-di-O-methylene-D-manno bicyclic structure are depicted in Scheme 5.5. Scheme 5.5. Preferential conformers of 1,6-di-O-benzoyl-2,4:3,5-di-O-methylene-D-mannitol. In order to assess the relative preference of Manx for these two conformations, a variable temperature 1H NMR study was carried out in the +60 to -60 ºC range, and 2D- NOESY spectra were recorded at room temperature. In the case that I and II conformers were present, two signals for each proton should appear in the 1H NMR spectra at low temperatures since no interconversion is expected to take place under such conditions. On the contrary, single proton signals were observed along the whole interval and the only detected change was an upfield shift of the signals with heating (see Annex F). It can be concluded therefore that no conformational interconversion occurs and that only one conformer is present along the whole range of temperatures. The analysis of the coupling constants of the 1H NMR spectra suggested that conformer I is the preferred one for Manx. In fact NMR spin simulation using the SpinWorks 3 program provided the spin-spin coupling constants (see Annex F) for the ABXYY’X’A’B’ system, in which small long range couplings between Y and X’ protons were detected. The calculated coupling constants were: JAB= JA’B’= -11.98 Hz; JAY= JA’Y’= 2.91 Hz; JBY= JB’Y’= 6.71 Hz; JYY’= JY’Y= 5.3 Hz; JYX= JY’X’= 8.60 Hz; JYX’= JY’X= -0.58 Hz
Chapter 5 254 Applying these coupling constants to the Karplus equation derived for alditols (Jardetzky and Roberts, 1981; Franks et al., 1991), the torsional angles Ψ and Φ could be estimated. The values obtained were 160º and 48º respectively, which are very close to the angles measured for conformer I but very far from those of conformer II. The identity of conformer I as the preferred one was strongly supported by 2D NOESY spectrum (Figure 5.24). As expected for conformation I, strong NOE signals were detected for HY and one of the acetal methylene protons, whereas no NOE signals were observed for HX. In conformation II the short distances between HX and one of the protons of the acetal methylene should be reflected in the 2D spectra with strong NOE signals, which is not the case. It is worthy to note that crystal structure studies on 2,4:3,5- di-O-methylene-D-mannitol-1,6-di-trans-cinnamate (Bernstein et al., 1980) established a conformation in the solid state for the bicyclic unit similar to which is determined in this work. The conclusion derived from this conformational study is that the preferred conformation of Manx is I and that this conformation is well stable; such features will confer to this monomer an enhancing stiffness effect on the polymer chain in which it is inserted. Figure 5.24. 2D NOESY spectrum of 1,6-di-O-benzoyl-2,4:3,5-di-O-methylene-D-mannitol. Small arrows indicate NOE signals.
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 255 5.4.3.2. Polymer synthesis and chemical structure The polymerizations involving 2,4:3,5-di-O-methylene-D-mannitol were carried out in the total absence of solvents to imitate as far as possible the conditions usually applied in the industrial practice. PManxT homopolyester, from 2,4:3,5-di-O-methylene-D- mannitol and dimethyl terephthalate, as well as PBxManxyT copolyesters from 2,4:3,5-di- O-methylene-D-mannitol, 1,4-butanediol and dimethyl terephthalate, were prepared by a two-step melt-polycondensation process using DBTO as catalyst (Scheme 5.6). For comparison purposes, the parent PBT homopolyester was obtained from 1,4-butanediol and dimethyl terephthalate using the same polycondensation procedure. In the preparation of PBxManxyT copolyesters, the incorporation of the diols into the copolyesters was accurately controlled by using a 5% molar excess of the diol mixture to dimethyl terephthalate, and by initiating transesterification reactions at mild temperature to prevent volatilization of diols. Temperature was progressively increased to avoid crystallization of oligomers, and polycondensation reactions were performed at temperatures between 210 and 240 ºC and under vacuum to facilitate the removal of volatile by-products; lower temperatures and longer reaction times were used in copolyesters with higher contents in Manx to minimize the decomposition of this sugar compound. When polymerizations were ended, the reaction mixtures were dissolved either in chloroform or in a mixture of chloroform and trifluoroacetic acid, and precipitated in methanol to obtain the polymers in yields close to 85-90%. The synthesis results obtained for sugar-based PManxT homopolyester and PBxManxyT copolyesters are gathered in Table 5.7 together with data for the parent PBT homopolyester. Scheme 5.6. Polymerization reactions leading to PBxManxyT copolyesters.
Chapter 5 256 Table 5.7. Molar composition, molecular weight and microstructure of PBxManxyT copolyesters. Copolyester Molar composition Molecular weight Microstructure Yield (%) Feed Copolyestera Dyads Number Average Sequence Lengths Randomnessd XB XManx XB XManx [η]b Mnc Mwc Ðc BB BManx/ ManxB ManxManx nB nManx R PBT 90 100 0 100 0 0.93 17,100 (11,800) 41,300 (40,500) 2.4 (3.4) PB90Manx10T 86 90 10 91.0 9.0 1.18 20,300 (14,200) 51,200 (50,800) 2.5 (3.6) 75.6 21.7 2.7 8.0 1.3 0.92 PB80Manx20T 87 80 20 80.3 19.7 0.84 16,500 (11,500) 41,000 (40,200) 2.5 (3.5) 59.9 34.0 6.1 4.5 1.4 0.96 PB70Manx30T 86 70 30 69.2 30.8 0.75 16,600 (10,400) 38,500 (37,900) 2.3 (3.6) 44.9 44.0 11.1 3.0 1.5 0.99 PB60Manx40T 85 60 40 59.2 40.8 0.83 16,400 (11,600) 40,100 (39,800) 2.4 (3.4) 35.0 47.8 17.2 2.5 1.7 0.99 PB50Manx50T 88 50 50 49.0 51.0 0.77 16,300 (10,900) 38,900 (38,400) 2.4 (3.5) 23.1 48.4 28.5 2.0 2.2 0.97 PManxT 85 0 100 0 100 0.51 12,900 (8,100) 30,200 (29,300) 2.3 (3.6) a Molar composition determined by integration of the 1H NMR spectra. b Intrinsic viscosity in dL·g-1 measured in dichloroacetic acid at 25 ºC. c Number and weight-average molecular weights in g·mol-1 and dispersities measured by GPC in HFIP against PMMA standards after (without parentheses) and before purification (in parentheses). d Randomness index of copolyesters statistically calculated on the basis of the 13C NMR analysis.
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 257 The precipitated PBxManxyT copolyesters were obtained with intrinsic viscosities between 0.8 and 1.2 dL·g-1, weight-average molecular weights confined in the 38,000- 52,000 g·mol-1 interval, and dispersities in the 2.3-2.5 range. Molecular weights of PBxManxyT copolyesters slightly decay with the content in Manx, in accordance with the values observed for PManxT homopolyester. The GPC analysis of the reaction product (without being subjected to any treatment) afforded weight-average molecular weights very close to those obtained for the precipitated product but with significantly higher dispersities. Such differences indicate the presence of minor amounts of oligomeric fractions or remaining monomers in the raw product that were then removed by precipitation. The 1H NMR spectra corroborated the chemical structure of the copolyesters with all signals being correctly assigned to the different protons contained in their repeating units (Figure 5.25). Integration of the proton signals arising from B and Manx units led to quantify the composition of the copolyesters in such units. Data provided by this analysis are given in Table 5.7, showing that all copolyesters had compositions very close to those of their corresponding feeds. Figure 5.25. Compared 1H NMR spectra of PBxManxyT copolyesters.
Chapter 5 258 The microstructure of PBxManxyT copolyesters was determined by 13C NMR analysis. The signals of all the magnetically different carbons contained in the repeating units of the copolyesters appear well resolved in the 13C NMR spectra (see Annex F) and in particular those arising from the nonprotonated aromatic carbons, which come out to be sensitive to sequence effects at the dyad level. As shown in Figure 5.26 for the whole set of PBxManxyT copolyesters, each resonance of the nonprotonated aromatic carbons appears split so that four peaks are seen in the 133-135 ppm chemical shift interval corresponding to the three types of dyads (BB, BManx and ManxB, ManxManx) that are feasible along the copolyester chain. By integration of these peaks, the dyad contents, the number-average sequence lengths and the degree of randomness were estimated (Table 5.7), leading to the conclusion that the microstructure of the copolyesters was random in all cases. Figure 5.26. 13C NMR signals used for the microstructure analysis of PBxManxyT copolyesters with indication of the dyads to which they are assigned. 5.4.3.3. Thermal properties 2,4:3,5-di-O-methylene-D-mannitol is a thermally fairly stable solid compound which starts to volatilize above 150 ºC and displays a sublimation temperature of 280 ºC without perceivable decomposition. The thermal stability of PManxT homopolyester and PBxManxyT copolyesters made from this compound was evaluated by thermogravimetry
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 259 under inert atmosphere and compared with that displayed by PBT. TGA traces are depicted in Figure 5.27, and thermal data provided by this analysis are presented in Table 5.8. The weight loss profiles generated reveal that all polyesters start to decompose well above 360 ºC with a thermal degradation mechanism involving one main step with the maximum rate at temperatures steadily increasing with the content in Manx units. The same trend is observed for ºT5%, and these results are in accordance with the high thermal stability observed for PManxT homopolyester, with ºT5% and maxTd at 378 and 421 ºC, respectively. The valuable conclusion drawn from this thermogravimetric study is that the insertion of Manx units in aromatic polyesters instead of reducing their thermal stability contributes to a significant increase in their decomposition temperatures. The fact of improving the thermal stability of commonly used aromatic polyesters such as PBT can broaden even more the application window of these materials. Figure 5.27. (a) TGA traces (solid lines) and derivative curves (dashed lines) of PBT and PManxT homopolyesters. (b) TGA traces of PBxManxyT copolyesters.
Chapter 5 260 Table 5.8. Thermal properties of PBxManxyT copolyesters. Copolyester TGA DSC First heatinge Coolinge Second heatinge ºT5%a Tdb W c Tgd Tm ΔHm Tc ΔHc Tm ΔHm (ºC) (ºC) (%) (ºC) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) PBT 371 408 2 31 223 (223) 56.2 (54.3) 196 43.3 223 44.0 PB90Manx10T 371 408 3 55 197 (196) 34.1 (33.0) 159 30.4 198 31.3 PB80Manx20T 372 410 8 66 184 (183) 26.0 (27.0) 140 23.4 184 24.1 PB70Manx30T 374 411 8 77 162 (162) 5.5 (16.6) - - - - PB60Manx40T 376 411 9 88 122 (119) 5.0 (9.8) - - - - PB50Manx50T 377 412 11 100 - - - - - - PManxT 378 421 13 137 - - - - - - aTemperature at which 5% weight loss was observed. bTemperature for maximum degradation rate. cRemaining weight at 600 ºC. dGlass-transition temperature taken as the inflection point of the heating DSC traces of melt-quenched samples recorded at 20 ºC·min-1. eMelting (Tm) and crystallization (Tc) temperatures, and melting (ΔHm) and crystallization (ΔHc) enthalpies measured by DSC at heating/cooling rates of 10 ºC·min-1 of precipitated samples (without parentheses) and of films (in parentheses). Other thermal properties of prime importance in connection with the potential application of these polyesters are the glass-transition temperature (Tg) and the melting temperature (Tm). For instance, an increase in the Tg of these polyesters could open their use in the hot-filled and pasteurized-container fields. The glass-transition temperatures of the PManxT and PBT homopolyesters and PBxManxyT copolyesters were measured as the inflection point of the heating DSC traces of samples quenched from the melt. A remarkable feature of PManxT homopolyester is that it has a Tg more than 100 ºC higher than the aromatic PBT homopolyester. Regarding PBxManxyT copolyesters, their Tg increase steadily with the content in Manx units, with minimum and maximum values corresponding to their respective PBT and PManxT homopolyesters (Figure 5.28), as should be expected from their statistical microstructure. As expected from its cyclic stiff nature, the effect of Manx on Tg of PBT is largely greater than that exerted by the
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 261 incorporation of acyclic sugar units such as trimethoxy xylitol or L-arabinitol (Alla et al., 2006). It has even a much more pronounced effect than 2,3:4,5-di-O-methylene- galactitol, another acetalized sugar diol constituted by two non-fused dioxolane rings, which has been recently reported by us as capable of raising the Tg of PBT by near 30 ºC when it replaces 1,4-butanediol in a 40% (Lavilla et al., 2012b -Subchapter 5.2-). The effect of Manx turns to be similar to that of isosorbide; in fact, both PBT copolyesters containing 40% of either Manx or isosorbide (Kricheldorf et al., 2007; Sablong et al., 2008) display a Tg of around 90 ºC. No doubt the fused bicyclic structure characteristic of both Manx and isosorbide is unique in providing the polyester chain with a degree of stiffness not achievable with the insertion of rotable bicyclic units, as it is the case of the acetalized hexitol with galacto configuration. Figure 5.28. Glass-transition temperature versus composition plot for PBT copolyesters containing 2,4:3,5-di-O-methylene-D-mannitol (Manx), 2,3:4,5-di-O-methylene-galactitol (Galx) (Lavilla et al., 2012b -Subchapter 5.2-), 2,3,4-tri-O-methyl-L-arabinitol (Ar) or 2,3,4-tri-O-methylxylitol (Xy) (Alla et al., 2006). DSC traces of polyesters in the powdered form registered at first heating are depicted in Figure 5.29. PBxManxyT copolyesters with molar contents in Manx below or equal to 40%, as well as PBT homopolyester, gave heating traces with melting endotherms indicating that they are semicrystalline. Comparison of melting temperatures and enthalpies of semicrystalline PBxManxyT copolyesters with that of PBT homopolyester leads to the conclusion that the insertion of Manx units gives place to a significant decrease in both Tm and ΔHm (Table 5.8). PB50Manx50T copolyester as well as PManxT homopolyester appear to be amorphous. DSC data recorded from polyester
Chapter 5 268 5.4.5. References Alla, A.; Hakkou, K.; Zamora, F.; Martínez de Ilarduya, A.; Galbis, J.A.; Muñoz-Guerra, S. Macromolecules 2006, 39, 1410-1416. Bernstein, J.; Green, B.S.; Rejtö, M. J. Am. Chem. Soc. 1980, 102, 323-328. Brigl, P.; Gruner, H. Ber. 1932, 65B, 641-645. Burden, I.J.; Stoddart, J.F. J. Chem. Soc. Perk. T 1 1975, 7, 666-674. Chen, G.Q.; Patel, M.K. Chem. Rev. 2012, 112, 2082-2099. Fenouillot, F.; Rousseau, A.; Colomines, G.; Saint-Loup, R.; Pascault, J.P. Prog. Polym. Sci. 2010, 35, 578-622. Franks, F.; Dadok, J.; Ying, S.; Kay, R.L.; Grigera, J.R. J Chem. Soc. Faraday Trans. 1991, 87, 579-585. Gallucci, R.R.; Patel, B.R. Poly(butylene terephthalate). In: Scheirs, J.; Long, T.E.; Eds. Modern Polyesters, Chemistry and Technology of Polyesters and Copolyesters; John Wiley & Sons: Chichester, 2004, pp. 293-321. Gandini, A.; Coelho, D.; Gomes, M.; Reis, B.; Silvestre, A.J.D. J. Mater. Chem. 2009, 19, 8656-8664. Gandini, A. Green Chem. 2011, 13, 1061-1083. Gomes, M.; Gandini, A.; Silvestre, A.J.D.; Reis, B. J. Polym. Sci. Polym. Chem. 2011, 49, 3759-3768. Hall, I.H. The Determination of the Structures of Aromatic Polyesters from their Wideangle X-Ray Diffraction Patterns. In: Hall, I,H.; Ed. Structure of Crystalline Polymers; Elselvier Applied Science: UK, 1984, pp. 39-78. Haskins, W. T.; Hann, R. M.; Hudson, C. S. J. Am. Chem. Soc. 1943, 65, 67-70. Haworth, W. N.; Wiggins, L. F. J. Chem. Soc. 1944, 58-61. Jardetzky, O.; Roberts, G.C.K. Molecular Biology; Academic Press: New York, 1981. p. 201. Kiely, D.E.; Chen, L.; Lin, T.H. J. Am. Chem. Soc. 1994, 116, 571-578. Kiely, D.E.; Chen, L.; Lin, T.H. J. Polym. Sci., Polym. Chem. 2000, 38, 594-603. Kricheldorf, H.R.J. Macromol. Sci. 1997, C37, 599-631. Kricheldorf, H.R.; Behnken, G.; Sell, M. J. Macromol. Sci. Pure Appl. Chem. 2007, 44, 679-684. Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. Biomacromolecules 2011, 12, 2642-2652. -Subchapter 3.2-
Poly(butylene terephthalate) copolyesters from cyclic acetalized carbohydrate-based monomers 269 Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. J. Polym. Sci. Polym. Chem. 2012, 50, 1591-1604. (2012a) -Subchapter 3.3- Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, Polymer 2012, 53, 3432-3445. (2012b) -Subchapter 5.2- Metzke, M.; Guan, Z. Biomacromolecules 2008, 9, 208-215. Mills, J.A. Adv. Carbohydr. Chem. 1955, 10, 1-53. Okada, M. Prog. Polym. Sci. 2002, 27, 87-133. Pillai, C.K.S. Design. Monom. Polym. 2010, 13, 87-121. Sablong, R.; Duchateau, R.; Koning, C.E.; de Wit, G.; van Es, D.; Koelewijn, R.; van Haveren, J. Biomacromolecules 2008, 9, 3090-3097. Storbeck, R.; Ballauff, M. J. Appl. Polym. Sci. 1996, 59, 1199-1202. Tripathy, A.R.; MacKnight, W.J.; Kukureka, S.N. Macromolecules 2004, 37, 6793-6800. Williams, C.K. Chem. Soc. Rev. 2007, 36, 1573-1580. Wool, R.; Sun, S. Biobased Polymers and Composites; Academic Press: New York, 2005. Wu, J.; Eduard, P.; Thiyagarajan, S.; Jasinska-Walc, L.; Rozanski, A.; Fonseca Guerra, C.; Noordover, B.A.J.; van Haveren, J.; van Es, D.S.; Koning, C.E. Macromolecules 2012, 45, 5069-5080.
Chapter 5 270
271 CHAPTER 6 SSM-PREPARED POLY(BUTYLENE TEREPHTHALATE) COPOLYESTERS FROM CYCLIC ACETALIZED CARBOHYDRATEBASED MONOMERS
Chapter 6 272
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 273 6.1. Aim and scope of this Chapter Most of the approaches aimed at increasing the glass-transition temperature (Tg) of poly(butylene terephthalate) (PBT), such as reactive blending with more rigid polyesters and melt copolymerization with rigid comonomers, are successful in increasing the glass-transition temperature of PBT but at the same time severely compromise the melting and crystallization behavior. However, with solid-state modification (SSM) it is possible to exclusively modify the properties related to the amorphous phase (i.e. glasstransition temperature) while more or less retaining the crystallization behavior of the parent polyester. SSM consists of making to react a comonomer with a semicrystalline polymer in the solid state, under an inert gas flow or vacuum at 20-30 ºC below the Tm of the polymer but well above its Tg. SSM is based on the same principles as the commonly used solid-state polycondensation (SSP), which is usually employed to increase the molecular weight of semicrystalline polyesters and polyamides by transesterification and transamidation reactions, respectively, below the Tm but well above the Tg. Both techniques take advantage of the molecular mobility in the amorphous phase of the polymer above Tg. During SSM, only chain segments present in the mobile amorphous fraction (MAF) of PBT are modified with the monomer. Mobility restrictions prevent the crystalline phase as well as the rigid amorphous fraction (RAF) from taking part in transesterification reactions occurring during the SSM process. Among sugar-based monomers, especially those with a cyclic structure are of interest, due to their ability to add stiffness to the polymer backbone, resulting in enhanced Tg values. Recently, high molecular weight aromatic copolyesters containing isosorbide units have been prepared by solid-state modification of PBT. However, due to the relatively low reactivity of the secondary hydroxyl groups of isosorbide, the direct incorporation of this monomer into the polyester backbone using this technique was not possible. The synthesis of a macrodiol from isosorbide, terephthaloyl chloride and 1,4- butanediol was necessary to incorporate this bicyclic monomer into PBT via SSM. In this Chapter, the sugar-based cyclic acetalized diols 2,3:4,5-di-O-methylene- galactitol (Galx), 2,4:3,5-di-O-methylene-D-mannitol (Manx) and 2,3-di-O-methylene-L- threitol will be directly incorporated into PBT. The chemical microstructure and thermal
Chapter 6 274 properties of the resulting bio-based copolyesters will be discussed in detail, and compared to those of the melt polycondensation (MP)-prepared analogous copolyesters.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 275 6.2. Solid-state modification of PBT with cyclic acetalized galactitol and D- mannitol: Influence of composition and chemical microstructure on thermal properties Summary: Two bicyclic carbohydrate-based diols, 2,3:4,5-di-O-methylene-galactitol (Galx) or 2,4:3,5-di-O-methylene-D-mannitol (Manx), were introduced into the backbone of poly(butylene terephthalate) using the solid-state modification technique (SSM). The resulting copolyesters had a unique block-like chemical microstructure that endows them with superior thermal properties when compared with their random counterparts obtained by melt copolymerization. The materials prepared by SSM displayed higher melting points, crystallization temperatures and crystallinity, due to the presence of long PBT sequences in the copolyester. The glass-transition temperatures also increased upon incorporation of the bicyclic comonomers, this effect being more pronounced for Manx units. The melting points of these block-like copolyesters decreased after melting due to the occurrence of randomization but they remained higher than those of copolyesters prepared from the melt. SSM was demonstrated to be a very suitable technique for the incorporation of rigid monomers into the amorphous phase of PBT leading to bio-based non-random copolyesters with remarkable thermal properties. Publication derived from this work: Lavilla, C.; Gubbels, E.; Martínez de Ilarduya, A.; Noordover, B.A.J.; Koning, C.E.; Muñoz-Guerra, S. Macromolecules 2013, 46, 4335-4345.
Chapter 6 276 6.2.1. Introduction Poly(butylene terephthalate) (PBT) is a well-known semicrystalline aromatic polyester, characterized by a melting temperature near 225 ºC and a high crystallization rate, which is nowadays used in a wide variety of engineering plastic applications (Gallucci and Patel, 2004). There is a current growing interest in increasing the glasstransition temperature (Tg) of PBT to extend its use to new demanding fields like manufacturing of containers for pressurized gaseous beverages. In this regard, various approaches such as reactive blending with more rigid polyesters and melt copolymerization with rigid comonomers have been explored in order to obtain PBT- based copolyesters with enhanced Tg (Denchev and Sarkissova, 1996; Kelsey et al., 2000; Santos and Guthrie, 2006; Lotti et al., 2011). However these copolyesters show a random or almost random chemical microstructure with shorter PBT sequences, and consequently they exhibit a lower melting temperature, crystallization rate and crystallinity with respect to pure PBT. It would therefore be desirable to prepare PBT copolyesters having a higher Tg but with a crystallizability comparable to that of PBT. This may be achieved using the solid-state modification (SSM) technique, which derives from the solid-state polymerization (SSP) technique traditionally used as a post-polycondensation method for increasing the molecular weight of semicrystalline polymers (Chang, 1970; Fortunato et al., 1981; Papaspyrides and Vouyiouka, 2009). During SSM a comonomer is incorporated into the backbone of a semicrystalline polymer in the solid state, under an inert gas flow or vacuum at 20-30 ºC below the melting temperature (Tm) of the polymer but well above its glass-transition temperature (Tg); under these relatively mild conditions, side or thermal degradation reactions are minimized or even effectively suppressed. The SSM technique takes benefit from the molecular mobility present at temperatures above Tg to insert the comonomer into the mobile amorphous phase of the polyester, whereas the chain segments in the rigid amorphous fraction and the crystalline phase do not participate in the transesterification reactions (Jansen et al., 2005 and 2006). Thus, SSM allows for the exclusive modification of the properties concerning the amorphous phase and hence, the crystallization behavior of the resulting copolymers should be quite comparable with that of the starting material. Jansen et al. have reported on the incorporation of 2,2-bis-[4-(2-hydroxyethoxy)phenyl]propane, bis(2- hydroxyethyl)terephthalate and 2,2’-biphenyldimethanol into the amorphous phase of PBT by SSM leading to block-like copolyesters with unusual chemical microstructures
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 277 and morphologies (Jansen et al., 2005, 2006, 2007 and 2008). Sablong et al. have recently used this innovative approach to successfully incorporate the heating sensitive isosorbide into PBT (Sablong et al., 2008). In order to reduce the use of petrochemicals and to increase the added value of agriculture waste streams, the development of bio-based monomers and polymers are receiving exceptional attention nowadays (Wool and Sun, 2005; Pillai, 2010; Chen and Patel, 2012). Carbohydrates are very prevalent and readily available compounds that may afford a wide variety of monomers suitable for making different classes of polymers (Galbis and García-Martín, 2010; Gandini, 2011). For example, 2,5-furandicarboxylate is regarded as the potential replacement for terephthalic acid in engineering, bottle or coating applications (Gandini et al., 2009; Gomes et al., 2011; Gubbels et al., 2013). Carbohydrate-based difunctional compounds with a bicyclic molecular structure are receiving exceptional attention nowadays as comonomers for linear polycondensation because, in addition to their natural origin, they distinguish by their capacity to provide stiffness to the polymer backbone and to increase therefore the Tg. A well-known example is 1,4:3,6-dianhydro-D-glucitol, known as isosorbide, as well as other dianhydrohexitol derivatives used as monomers in the synthesis of polyesters with interesting properties and enhanced Tg (Sablong et al., 2008; Kricheldorf et al., 2007; Fenouillot et al., 2010). Another family of carbohydrate-based bicyclic monomers, which have recently emerged in the polycondensation field, is that obtained by internal acetalization of carbohydrates. These monomers have shown to be very suitable for the preparation of copolyesters with enhanced Tg values by melt polycondensation (Lavilla et al., 2011 -Subchapter 3.2-, 2012a -Subchapter 5.2-, 2012b -Subchapter 5.4-, 2013a -Subchapter 4.3- and 2013b -Subchapter 3.4-; Japu et al., 2012). In this work, two bicyclic acetalized carbohydrate-based diols, 2,3:4,5-di-O- methylene-galactitol (Galx) and 2,4:3,5-di-O-methylene-D-mannitol (Manx), have been incorporated into PBT via solid-state modification, leading to copolyesters with enhanced Tg values and a retained melting behavior. PBT appears to be a very convenient aromatic polyester to carry out the present study due to its relatively low Tm compared to PET, which allows the application of the SSM technique at lower temperatures. Galx is obtained by acetalization of galactaric acid with paraformaldehyde and subsequent reduction; the centrosymmetric structure of Galx consists of two non-fused 1,3-dioxolane rings with the two primary hydroxyl groups in exo positions. Manx is obtained by acetalization of 1,6-di-O-benzoyl-D-mannitol followed by hydrolysis of the benzoxy
Chapter 6 284 broad peak going out at lower elution time corresponds to the unmodified PBT. After 0.5 h of transesterification, a clear decrease in intensity of the Manx peak was observed while the peak originating from PBT shifted to higher elution times. This shift is indicative for the chain scission of the PBT caused by the the transesterification reaction with Manx. At tSSM= 1 h, the polymer was eluted at the highest elution time to subsequently shift back to lower values. Hence, recombination of the polymer chains takes place resulting in higher Mn and Mw values and liberation of 1,4-butanediol, which diffuses out of the polymer particles. All the SSM processes were followed using SEC, with the aim of obtaining SSMPBxGalxyT and SSMPBxManxyT copolyesters with comparable molecular weights, and also similar to molecular weights of samples prepared by MP (abbreviated here as MPPBxGalxyT and MPPBxManxyT copolyesters) (Lavilla et al., 2012a -Subchapter 5.2- and 2012b -Subchapter 5.4-), in order to perform an accurate and reliable comparison of their crystallization behavior. All the SSM reactions were stopped when Mn and Mw values of about 17-20 and 40-44 kg·mol-1, respectively, were reached (Table 6.1). The proton nuclear magnetic resonance (1H NMR) corroborated the expected chemical structure of the copolyesters (Figure 6.2), with all resonances being assigned to the various protons present in the copolyester backbone. Integration of the proton signals arising from 1,4-butanediol and sugar units led to the determination of the composition of the copolyesters. Data provided by this analysis is given in Table 6.1, where it can be seen that the bicyclic monomers Galx and Manx were successfully incorporated into the PBT by SSM. The final compositions of the copolyesters are comparable to the 1,4- butanediol:sugar feed compositions. This is in contrast to reported studies about the SSM of PBT using the sugar-based bicyclic diol isosorbide (Is). Experiments using SSM for the direct incorporation of Is into PBT were unsuccessful; isosorbide could not be incorporated under such conditions. The synthesis of a macrodiol from isosorbide, terephthaloyl chloride and 1,4-butanediol was necessary to incorporate this bicyclic monomer into PBT via SSM (Sablong et al., 2008). Nevertheless, in this present study the sugar-based diol was successfully incorporated without having to revert to the preparation of a sugar-based macrodiol. The reason for this difference is most probably that the primary hydroxyl groups of Galx and Manx have a much higher reactivity compared to the secondary hydroxyl groups of Is. By varying the PBT/Galx and PBT/Manx feed ratios, six copolyesters with various 1,4-butanediol:sugar ratios were obtained within each series (Table 6.1). The compositions and molecular weights of both SSM series are comparable and are also very similar to the fully random MPPBxGalxyT and MPPBxManxyT copolyesters obtained by polycondensation in the melt.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 285 Table 6.1. Molar composition, molecular weight and microstructure of SSMPBxGalxyT and SSMPBxManxyT copolyesters. Copolyester Molar composition Molecular weight Microstructure Feed Copolyestera Dyad content Number Average Sequence Lengths Randomnessc Randomnessd XB XSugar XB XSugar Mnb Mwb Ðb NBB NBS/SB NSS nB nS R Rc PB91Galx9T 90.9 9.1 91.0 9.0 17,800 41,300 2.3 85.6 12.2 2.2 15.0 1.4 0.80 0.92 PB89Galx11T 87.0 13.0 88.7 11.3 19,800 43,600 2.2 80.3 16.2 3.5 10.9 1.4 0.79 0.93 PB86Galx14T 83.3 16.7 86.4 13.6 19,400 44,800 2.3 76.2 18.3 5.5 9.3 1.6 0.73 0.89 PB81Galx19T 80.0 20.0 80.6 19.4 17,900 43,100 2.4 65.6 24.6 9.8 6.3 1.8 0.71 0.89 PB78Galx22T 76.9 23.1 77.9 22.1 18,800 44,700 2.4 63.4 24.9 11.7 6.1 1.9 0.68 0.89 PB71Galx29T 71.4 28.6 71.0 29.0 18,500 44,000 2.4 52.9 29.3 17.8 4.6 2.2 0.67 0.85 PB92Manx8T 90.9 9.1 92.0 8.0 18,800 42,000 2.2 85.7 11.4 2.9 16.0 1.5 0.73 0.80 PB88Manx12T 87.0 13.0 88.2 11.8 17,600 41,300 2.3 78.8 15.4 5.8 11.2 1.7 0.66 0.77 PB86Manx14T 83.3 16.7 85.9 14.1 18,700 43,900 2.3 72.8 19.7 7.5 8.4 1.8 0.69 0.83 PB82Manx18T 80.0 20.0 81.5 18.5 17,100 40,800 2.4 66.4 23.1 10.5 6.7 1.9 0.67 0.83 PB77Manx23T 76.9 23.1 77.0 23.0 19,500 44,800 2.3 59.7 27.8 12.5 5.3 1.9 0.72 0.87 PB70Manx30T 71.4 28.6 69.7 30.3 17,700 40,700 2.3 52.1 29.2 18.7 4.6 2.3 0.66 0.82 a Molar composition determined by integration of the 1H NMR spectra. b Number and weight-average molecular weights in g·mol-1 and dispersities measured by SEC in HFIP against PMMA standards. c Randomness index of copolyesters statistically calculated on the basis of the 13C NMR analysis. d Randomness index of the amorphous fraction of copolyesters, determined using the correction method reported by Jansen et al. (Jansen et al., 2005)
Chapter 6 286 Figure 6.2. 1H NMR spectra of SSMPBxGalxyT (a) and SSMPBxManxyT (b) copolyesters recorded in CDCl3/TFA at room temperature. 6.2.3.2. Chemical microstructure The chemical microstructure of SSMPBxGalxyT and SSMPBxManxyT copolyesters was elucidated using quantitative carbon nuclear magnetic resonance (13C NMR) spectroscopy. The resonances of all magnetically different carbons present in the repeat units of the copolyesters are well resolved in the 13C NMR spectra, which are shown in the Annex G. This is true in particular for the resonances arising from the non-protonated aromatic carbons, which prove to be sensitive to sequence distribution at the level of dyads (Kricheldorf, 1978; Newmark, 1980). The resonance of the quaternary carbon atom has split into four well resolved peaks which are observed in the 133-135 ppm chemical shift interval. These four resonances correspond to the different types of dyads (BB, BS and SB, SS) that are present along the copolyester backbone, and they display peak areas with a dependence on the copolyester composition (Figure 6.3).
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 287 Figure 6.3. 13C NMR signals used for the microstructure analysis of SSMPBxGalxyT (a) and SSMPBxManxyT (b) copolyesters with indication of the dyads to which they are assigned. The molar fractions (N) of the different dyads in the copolyesters were determined by the integration of their corresponding resonances. Based on these results, the number average sequence lengths n of the butylene-terephthalate (B) and sugarterephthalate (S) sequences, as well as the degree of randomness (R), are determined for each copolyester by using the equations given below (Yamadera and Murano, 1967;
Chapter 6 288 Randall, 1977). When the R is determined to be in the order of unity the chemical microstructure is fully random. If the R value decreases towards zero, the chemical microstructure becomes more blocky. nB = (NBB + 0.5·(NBS+NSB)) / (0.5·(NBS+NSB)) nS = (NSS + 0.5·(NBS+NSB)) / (0.5·(NBS+NSB)) R= (1/nB) + (1/nS) Results from these calculations (Table 6.1) show that the overall sequence distribution in SSMPBxGalxyT and SSMPBxManxyT copolyesters is non-random for the whole composition range used in this study. At low comonomer contents it was challenging to accurately determine the integral values for the comonomers. The R-values of copolyesters with medium to high content in sugar were all close to 0.70, indicating an overall non-random microstructure. Since the crystalline phase does not participate in the SSM process, long PBT sequences are expected to be retained in the copolyester, which probably lead to the observed deviation of the R-values from unity. In order to ascertain information about the chemical microstructure in the amorphous phase, a correction needs to be made for the crystalline phase. Jansen et al. have developed a calculation method to adjust the solution 13C NMR peak integral values of the dyad sequences in such a way that only the amorphous fraction is accounted for (Jansen et al., 2005). The chemical microstructures of the total amorphous phases of SSMPBxGalxyT and SSMPBxManxyT copolyesters were determined by using this calculation method, implementing data obtained by differential scanning calorimetry, which will be discussed below. The calculated R-values of the amorphous phase are shown in Table 6.1. It can be observed that all the R-values increase after the correction. This can be easily understood because long PBT sequences have been excluded by this correction. The corrected R-values from the two series are similar, all of them being closer to unity, thus indicating a practically random chemical microstructure in the amorphous phase. The difference in the R-values between the two series before correction is most likely caused by the higher crystallinity of the Galx-based samples. The corrected R-values are depicted in Figure 6.4, together with overall SSM and MP R-values, showing that the chemical microstructures in the amorphous phases of copolyesters prepared by SSM are comparable to those of materials prepared by MP. Hence, by SSM Galx and Manx are incorporated into the mobile amorphous phase in a similar way as through melt polymerization, even though in SSM only the most mobile chain segments are participating in the transesterification reactions. The main conclusion of this study is that,
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 289 by using SSM, partially bio-based copolyesters with unique chemical microstructures have been obtained, consisting of a blocky overall microstructure formed by PBT homopolymer sequences in the crystalline phase and a practically random microstructure in the amorphous part. Figure 6.4. Comparison of the degrees of randomness of PBxGalxyT (a) and PBxManxyT (b) copolyesters obtained by SSM and MP (Lavilla et al., 2012a -Subchapter 5.2- and 2012b - Subchapter 5.4-). 6.2.3.3. Thermal properties The thermal behavior of the PBT-based copolyesters prepared by SSM has been systematically studied by TGA and DSC (Table 6.2). The results obtained from the two series of copolyesters are compared with each other, with the parent homopolyester PBT, and with copolyesters prepared by MP. First, the thermal stability was evaluated by TGA under an inert atmosphere. The TGA traces of the complete sets of copolyesters are comparatively depicted in Annex G. The thermal properties derived from the shown thermograms, viz. onset decomposition temperatures (ºT5%) and temperatures for maximum degradation rate (Td), are listed in Table 6.2. Decomposition of all SSMPBxGalxyT and SSMPBxManxyT copolyesters takes place in one step, at higher temperatures with respect to neat PBT. Additionally, the residual weight left upon heating at 600 ºC is about 3-7% of the initial weight. These results are in full accordance with the enhanced thermal stability observed for the copolyesters prepared by MP compared to PBT (Lavilla et al., 2012a -Subchapter 5.2- and 2012b -Subchapter 5.4-). The valuable conclusion drawn from this comparative thermogravimetric study is that the insertion of sugar-based bicyclic Galx and Manx units into PBT by SSM increases the thermal stability of the PBT-based copolyesters.
Chapter 6 290 Table 6.2. Thermal properties of SSMPBxGalxyT and SSMPBxManxyT copolyesters. Copolyester TGA DSC First heatinge Coolinge Second heatinge ºT5%a Tdb W c Tgd Tm ΔHm Tc ΔHc Tc ΔHc Tm ΔHm (ºC) (ºC) (%) (ºC) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) (ºC) (J·g-1) PBT 348 385 0 45 225 60.5f 193 52.0 - - 224 49.0 PB91Galx9T 350 390 4 47 218 77.9 180 48.3 - - 212 46.6 PB89Galx11T 360 399 3 49 209 71.6 176 43.8 - - 209 38.3 PB86Galx14T 350 388 5 50 207 67.5 167 41.2 - - 205 37.1 PB81Galx19T 354 391 5 54 208 56.5 143 37.5 - - 191 26.9 PB78Galx22T 354 391 4 54 208 55.9 131 32.0 - - 185 25.4 PB71Galx29T 354 391 6 57 205 41.1 - - 121 9.6 169 11.9 PB92Manx8T 363 399 4 56 211 67.3 183 41.5 - - 212 41.4 PB88Manx12T 349 388 7 62 210 64.8 170 39.0 - - 205 32.3 PB86Manx14T 364 401 6 67 208 63.5 158 38.3 - - 200 30.3 PB82Manx18T 351 389 5 68 207 56.0 140 27.3 - - 190 27.0 PB77Manx23T 362 402 6 70 206 45.4 121 2.6 139 5.1 181 8.8 PB70Manx30T 354 392 7 76 205 37.7 - - - - - - a Temperature at which 5% weight loss was observed. b Temperature for maximum degradation rate. c Remaining weight at 600 ºC. d Glasstransition temperature taken as the inflection point of the heating DSC traces recorded at 20 ºC·min-1 during the 3rd heating run. e Melting (Tm) and crystallization (Tc) temperatures, and melting (ΔHm) and crystallization (ΔHc) enthalpies measured by DSC at heating/cooling rates of 10 ºC·min-1. f After applying a similar thermal treatment which was used for the copolyesters the ΔHm increased to 74.4 J/g.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 291 The DSC analysis revealed that the incorporation of the bicyclic Galx and Manx units by SSM induced significant changes in the glass-transition temperature (Tg) of PBT- based copolyesters (Table 6.2). In the SSMPBxGalxyT series, the Tg values steadily augmented as 1,4-butanediol was replaced by Galx units, reaching a value of 57 ºC when 29 mol% of Galx was incorporated, and following a trend similar to that observed for MPPBxGalxyT copolyesters (Lavilla et al., 2012a -Subchapter 5.2-). Regarding the SSMPBxManxyT series, the Tg also increased with increasing amount of incorporated comonomer; in this case the effect of the bicyclic diol on Tg was even more pronounced for all the compositions. A Tg value of 76 ºC was reached for a Manx content of 30 mol%. Apparently, the fused bicyclic 1,3-dioxane structure of Manx has a more significant influence on the Tg values than the more flexible structure of Galx, a difference that has also been noticed in copolyesters prepared by MP (Lavilla et al.,2012b -Subchapter 5.4-). These results evidence the effect that the rigid sugar-based comonomer exerts on restricting the mobility of the copolyester chain. Figure 6.5. Glass-transition temperature versus composition plots for SSM and MP (Lavilla et al., 2012a -Subchapter 5.2- and 2012b -Subchapter 5.4-) copolyesters. (a) PBxGalxyT and (b) PBxManxyT. The glass-transition temperatures were obtained from the third heating run. The triangle represents the Tg of the homopolyester PBT.
Chapter 6 292 Melting temperatures (Tm) and enthalpies (ΔHm), and crystallization temperatures (Tc) and enthalpies (ΔHc) were measured using differential scanning calorimetry (DSC) and are listed in Table 6.2. For illustration, the DSC traces of SSMPB81Galx19T, recorded upon heating and cooling, are depicted in Annex G. A decrease in Tm and ΔHm can be seen when the first and second heating runs are compared, this behavior will be discussed later in this article. All the prepared SSMPBxGalxyT and SSMPBxManxyT copolyesters are semicrystalline and exhibit a Tm value exceeding 205 ºC in the first heating run. This Tm corresponds to the crystalline portion of the material which did not participate in the SSM reaction. Melting temperature versus composition plots for SSMPBxGalxyT and SSMPBxManxyT copolyesters are comparatively depicted in Figure 6.6, together with data of the corresponding copolyesters obtained by MP. Whereas melting temperatures of MPPBxGalxyT and MPPBxManxyT sharply decrease with increasing amounts of sugar-based units, the Tm of the copolyesters prepared by SSM did not present such a tendency. Moreover, the difference between the Tm values of materials prepared by SSM and MP is more pronounced as the sugar content increases. These results proved that, by means of SSM, it is possible to prepare sugar-based PBT copolyesters with only moderately decreasing melting temperatures. Figure 6.6. Melting temperature versus composition plots for SSM and MP (Lavilla et al., 2012a -Subchapter 5.2- and 2012b -Subchapter 5.4-) copolyesters. (a) PBxGalxyT and (b) PBxManxyT. The triangle represents the melting temperature of the homopolyester PBT.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 293 The copolyesters obtained by SSM have relatively high melting enthalpies (Table 6.2), especially compared to those prepared by MP (Lavilla et al., 2012a - Subchapter 5.2- and 2012b -Subchapter 5.4-). This increase is due to annealing occurring during the SSM reaction. The results presented in Table 6.2 suggest that small amounts of the monomer present in the first SSM stages could act as a plasticizer, thus increasing the molecular mobility of the polymer chains and favoring annealing. However, a decrease in the melting enthalpy was observed when the amount of sugar-based comonomer was increased further. In order to determine the crystallinity of SSMPBxGalxyT and SSMPBxManxyT copolyesters, to elucidate if co-crystallization is taking place and to complement the DSC data, wide-angle X-ray diffraction (WAXD) was performed on the two series of copolyesters as well as on neat PBT. The powder WAXD profiles are shown in Figure 6.7, and the most prominent Bragg spacings present therein are listed in Table 6.3. Figure 6.7. Powder WAXD profiles of SSMPBxGalxyT (a) and SSMPBxManxyT (b) copolyesters with indication of dhkl in Å.
Chapter 6 300
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 301 6.3. Carbohydrate-based PBT copolyesters from a cyclic diol derived from naturally occurring tartaric acid: A comparative study regarding melt polycondensation and solid-state modification Summary: 2,3-di-O-methylene-L-threitol (Thx) is a cyclic carbohydrate-based diol prepared by acetalization and subsequent reduction of the naturally occurring tartaric acid. The structure of Thx consists of a 1,3-dioxolane ring with two attached primary hydroxyl groups. Two series of partially bio-based poly(butylene terephthalate) (PBT) copolyesters were prepared using Thx as comonomer by melt polycondensation (MP) and solid-state modification (SSM). Fully random copolyesters were obtained after MP using mixtures of Thx and 1,4-butanediol in the reaction with dimethyl terephthalate. Copolyesters with a unique block-like chemical microstructure were prepared by incorporation of Thx into the amorphous phase of PBT by SSM. The partial replacement of 1,4-butanediol units by Thx resulted in satisfactory thermal stabilities and gave rise to an increase in Tg values, and this effect was comparable for copolyesters prepared by MP and SSM. The partially bio-based materials prepared by SSM displayed higher melting points and easier crystallization from the melt, due to the presence of long PBT sequences in the backbone of the copolyester. The incorporation of Thx in the copolyester backbone enhanced the hydrolytic degradation of the materials with respect to the degradation of pure PBT. Publication derived from this work: Lavilla, C.; Gubbels, E.; Alla, A.; Martínez de Ilarduya, A.; Noordover, B.A.J.; Koning, C.E.; Muñoz-Guerra, S. (Submitted 2013)
Chapter 6 302 6.3.1. Introduction Nature produces approx. 140·109 tons of carbohydrates from carbon dioxide and water annually, making these compounds the most abundant organic materials on earth. Humanity only uses 4% of this feedstock for food and non-food purposes (Lichtenthaler, 2010). Therefore this class of compounds is an extraordinary source of chemicals, capable of providing a wide variety of building blocks for polycondensates (Wool and Sun, 2005; Galbis and García-Martín, 2008; Pillai, 2010). Thus, the development of carbohydrate-based polycondensates has the potential to significantly reduce the amount of petroleum consumed for polymer production. However, there are only a few examples of commercially available carbohydrate-derived polymers, which is mainly due to the relatively high price of these materials compared to their petroleum-based counterparts. Another limitation of carbohydrates as building blocks in the synthesis of linear polycondensates is their inherent multifunctionality. Although some linear polycondensates have been synthesized using carbohydrate-based monomers bearing pendant hydroxyl groups (Kiely et al., 1994 and 2000), the most commonly used strategy is blocking them with stable protecting groups, and thus retaining only two reactive functions to carry out linear polycondensation (Galbis and García-Martín, 2010; Gandini, 2011). Among carbohydrates, tartaric acid (2,3-dihydroxy-succinic acid) stands out by its easy accessibility and relative simplicity in terms of its molecular structure. This hydroxylated aldaric acid occurs in many plants and fermented grape juice (Blair and DeFraties, 2000). From tartaric acid, various monomers have been prepared and explored as carbohydrate-based building blocks in the synthesis of polycondensates, usually with the pendant hydroxyl groups protected in several forms (Alla et al., 2000 and 2005; Kint et al., 2001; Marín and Muñoz-Guerra, 2008; Japu et al., 2013). Aromatic polyesters are used in a wide array of applications due to their high performance as thermoplastic materials with good thermal and mechanical properties (Gallucci and Patel, 2004). The properties of such aromatic polyesters can be adjusted by copolymerization, preferably with bio-based comonomers, in order to make these materials more environmentally friendly and more sustainable (Li et al., 2012). However, the incorporation of bio-based acyclic monomers usually results in a lowered glass-transition temperature and reduced stiffness (Kint et al., 2001; Zamora et al., 2005; Li et al., 2012). For example, 2,3-di-O-methyl-L-threitol, an acyclic diol with its secondary hydroxyl groups protected as methoxy ethers, has been employed in the preparation of aromatic
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 303 copolyesters (Kint et al., 2001). On the other hand, carbohydrate-derived monomers with a cyclic structure distinguish themselves by providing polycondensates with improved properties, especially those related to polymer chain stiffness (Kricheldorf et al., 2007; Gandini, 2009; Fenouillot et al., 2010; Gubbels et al., 2013a; Lavilla et al., 2013a -Subchapter 4.3-). Two cyclic acetalized carbohydrate-based diols derived from tartaric acid, i.e. 2,3-di-O-isopropylidene-L-threitol and 2,3-di-O-methylene-L-threitol, with the pendant functional groups protected as isopropylidene acetal and methylene acetal, respectively, have been recently employed in the preparation of polyurethanes (Marín and Muñoz-Guerra, 2008). In this work, the cyclic diol 2,3-di-O-methylene-L-threitol (Thx) is employed as building block for the preparation of poly(butylene terephthalate) (PBT) copolyesters. Whereas isopropylidene acetals are known to be unstable and easily removed in acidic media (Marín and Muñoz-Guerra, 2008; Gómez and Varela, 2009; Dhamaniya and Jacob, 2010), the protection by methylene acetal groups allows for the preparation of polycondensates with a satisfactory stability during processing and handling, since the methylene acetal rings are much more resistant to opening (Lavilla and Muñoz-Guerra, 2012 -Subchapter 5.3-; Lavilla et al., 2013b -Subchapter 3.4-). Two different techniques, i.e. melt polycondensation (MP) and solid-state modification (SSM), will be used to prepare the novel bio-based copolyesters from the aforementioned Thx. The widely known MP technique consists of reacting the monomers Thx, 1,4-butanediol, and dimethyl terephthalate in the melt, viz. in two successive stages involving transesterification and polycondensation reactions. The SSM technique (Jansen et al., 2006) is based on the same principles as the commonly used solid-state polycondensation (SSP), which is usually employed to increase the molecular weight of semicrystalline polyesters and polyamides by transesterification and transamidation reactions, respectively (Vouyiouka et al., 2005; Papaspyrides and Vouyiouka, 2009). Both SSM and SSP techniques are carried out at a reaction temperature just (20-30 °C) below the melting temperature of the crystalline phase of the polymer, and they take advantage of the molecular mobility in the amorphous phase of the polymer relatively far above Tg. During the SSM process, the monomer Thx is inserted exclusively into the mobile amorphous fraction (MAF) of poly(butylene terephthalate); mobility restrictions prevent the crystalline phase and the rigid amorphous fraction (RAF) from taking part in transesterification reactions (Jansen et al., 2005; Gubbels et al., 2013b; Lavilla et al, 2013c -Subchapter 6.2-). The synthesis, chemical microstructure and thermal properties of the resulting MP- and SSM-prepared bio-based copolyesters will be studied and
Chapter 6 304 discussed in detail. Moreover, a hydrolytic degradation study will be carried out to evaluate the influence of the Thx units on the degradation of the copolyester, and to ascertain the stability of the acetal group forming part of the carbohydrate-based residues. 6.3.2. Experimental section 6.3.2.1. Materials The cyclic diol 2,3-di-O-methylene-L-threitol (Thx) was synthesized from commercially available dimethyl L-tartrate, as described elsewere (Marín and Muñoz- Guerra, 2008). The reagents 1,4-butanediol (99%), dimethyl terephthalate (99+%) and the catalyst dibutyl tin oxide (DBTO, 98%) were purchased from Sigma-Aldrich. PBT was a kind gift from DSM. Deuterated chloroform (CDCl3, 99.8% atom D) and deuterated trifluoroacetic acid (TFA-d, 99% atom D) were obtained from Cambridge Isotope Laboratories. Solvents used for purification and characterization were purchased from Panreac and Biosolve and these were all of either technical or high-purity grade. All chemicals were used as received unless stated otherwise. 6.3.2.2. General methods 1H and 13C NMR spectra were recorded on a Bruker AMX-300 spectrometer at 25.0 ºC operating at 300.1 and 75.5 MHz, respectively. Samples were dissolved in a mixture of deuterated chloroform and deuterated trifluoroacetic acid (9:1), and spectra were internally referenced to tetramethylsilane (TMS). Approximately 10 and 50 mg of sample dissolved in 1 mL of the solvent mixture were used for 1H and 13C NMR, respectively. Sixty-four scans were acquired for 1H and 1,000-10,000 for 13C with 32 and 64-K data points as well as relaxation delays of 1 and 2 s, respectively. Size-exclusion chromatography (SEC) was performed on a system equipped with a Waters 1515 Isocratic HPLC pump, a Waters 2414 refractive index detector working at 40 ºC, a Waters 2707 autosampler, and a PSS PFG guard column followed by a 2PFG-linear-XL (7 μm, 8·300 mm) columns in series at 40 ºC. HFIP with potassium trifluoroacetate (3 g·L-1) was used as eluent at a flow rate of 0.8 mL·min-1. The molecular weights were calculated against polymethyl methacrylate standards (Polymer Laboratories, Mp= 580 Da up to Mp= 7.1·106 Da). Differential scanning calorimetry (DSC) was performed using a Perkin Elmer DSC Pyris 1 equipment. DSC thermograms were obtained from 3 to 5 mg samples at
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 305 heating/cooling rates of 10 ºC·min-1 under a nitrogen flow of 20 mL·min-1. Indium and zinc were used as standards for temperature and enthalpy calibration. The glass-transition temperatures were determined from the third heating run at a heating rate of 20 ºC·min-1 from rapidly melt-quenched polymer samples. The treatment of the samples for isothermal crystallization experiments was the following: the thermal history was removed by heating the sample up to 250 ºC and keeping the sample at this temperature for 5 min, subsequently it was cooled at 20 ºC·min-1 to the selected crystallization temperature, where it was left to crystallize until saturation. Thermogravimetric analyses were performed under a nitrogen flow of 20 mL·min-1 at a heating rate of 10 ºC·min-1, within a temperature range of 30 to 600 ºC, using a Perkin Elmer TGA 6 equipment. Sample weights of about 10-15 mg were used in these experiments. Wide-angle X-ray diffraction (WAXD) measurements were performed on a Rigaku Geigerflex Bragg-Brentano Powder Diffractometer using Cu radiation, wavelength 1.54056 Å, at 40 kV and 30 mA. The scans were performed with 0.02 º steps and a dwell time of 3 s in the 2θ range from 10 º till 35 º. The analyses were performed on the crude copolyester samples. Scanning electron microscopy (SEM) images were taken with a field-emission JEOL JSM-7001F instrument (JEOL, Japan) from platinum/palladium coated samples. 6.3.2.3. Polymer synthesis 6.3.2.3.1. Melt polycondensation (MP) Thx-based copolyesters were obtained from a mixture of 1,4-butanediol, the cyclic diol 2,3-di-O-methylene-L-threitol (Thx) and dimethyl terephthalate (DMT) with a predetermined composition. PThxT homopolyester was obtained by reacting Thx with DMT. The reactions were performed in a three-necked, cylindrical-bottom flask equipped with a mechanical stirrer, a nitrogen inlet and a vacuum distillation outlet. A 5% molar excess of the diol mixture to DMT was used, and dibutyl tin oxide (DBTO, 0.6% molar with respect to monomers) was the catalyst of choice. The apparatus was purged with nitrogen several times at room temperature in order to remove the last traces of air. Transesterification reactions were carried out under a low nitrogen flow at the selected temperature. Polycondensation reactions were performed at a selected temperature under a 0.03-0.06 mbar vacuum. At the end of the reaction, the reaction mixture was cooled to room temperature under a nitrogen atmosphere. The resulting polymers were dissolved in a mixture of chloroform and trifluoroacetic acid (9:1) and precipitated in an excess of methanol. Finally, the polymer was collected by filtration, extensively washed with methanol and dried under vacuum. The detailed reaction conditions are given below.
Chapter 6 306 The Thx-based copolyesters obtained via MP are abbreviated as MPPBxThxyT, where x and y are the mole percentages (mol-%) of 1,4-butanediol and Thx, respectively, in the resulting copolyester. PThxT homopolyester. Transesterification reactions were performed at 160 ºC for 2 h and at 180 ºC for 1 h under a low nitrogen flow. Polycondensation reactions were performed at 180 ºC for 8 h under a 0.03-0.06 mbar vacuum. 1H NMR (300.1 MHz, CDCl3/TFA), δ (ppm): 8.1 (s, 4H, ArH), 5.3 (s, 2H, OCH2O), 4.6 (m, 4H, OCH2CH), 4.5 (m, 2H, OCH2CH). 13C NMR (75.5 MHz, CDCl3/TFA), δ (ppm): 167.0 (CO), 133.5, 130.3, 95.7, 75.9, 64.9. MPPBxThxyT copolyesters. The copolyesters were obtained by a similar procedure, with polymerization conditions slightly differing for each composition feed. MPPB96Thx4T. Transesterification reactions were performed at 160 ºC for 1 h, at 200 ºC for 1 h and at 240 ºC for 0.5 h under a low nitrogen flow. Polycondensation reactions were performed at 250 ºC for 2.5 h under a 0.03-0.06 mbar vacuum. MPPB91Thx9T. Transesterification reactions were performed at 160 ºC for 1 h, at 200 ºC for 1 h and at 240 ºC for 0.5 h under a low nitrogen flow. Polycondensation reactions were performed at 240 ºC for 3 h under a 0.03-0.06 mbar vacuum. MPPB84Thx16T and MPPB82Thx18T. Transesterification reactions were performed at 160 ºC for 1 h, at 200 ºC for 1 h and at 230 ºC for 0.5 h under a low nitrogen flow. Polycondensation reactions were performed at 230 ºC for 3.5 h under a 0.03-0.06 mbar vacuum. MPPB77Thx23T and MPPB71Thx29T. Transesterification reactions were performed at 160 ºC for 1 h, at 200 ºC for 1 h and at 220 ºC for 0.5 h under a low nitrogen flow. Polycondensation reactions were performed at 220 ºC for 4 h under a 0.03-0.06 mbar vacuum. NMR characterization of MPPBxThxyT copolyesters. 1H NMR (300.1 MHz, CDCl3/TFA), δ (ppm): 8.1 (s, 4H, ArH), 5.3 (s, y·2H, OCH2O), 4.6 (m, y·4H, OCH2CH), 4.5 (t, x·4H, OCH2CH2), 4.5 (m, y·2H, OCH2CH), 2.0 (t, x·4H, OCH2CH2). 13C NMR (75.5 MHz, CDCl3/TFA), δ (ppm): 168.0 (CO), 167.0 (CO), 134.2-133.2, 130.3, 130.1, 95.7, 75.9, 66.3, 64.9, 25.4.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 307 6.3.2.3.2. Solid-state modification (SSM) of PBT Physical mixtures of purified PBT (Mn= 23.3 kg·mol-1 and Mw= 47.1 kg·mol-1, determined by SEC), the cyclic diol Thx, and DBTO catalyst (0.36% molar with respect to PBT) were prepared from solution, using a common solvent approach as described elsewere (Lavilla et al., 2013c -Subchapter 6.2-), in which the PBT was first precipitated to remove any titanium-based catalysts. Different molar ratios of PBT and Thx were used in order to obtain copolyesters with varying compositions. The solid-state modification of PBT was performed in a reactor comprising a glass tube (inner diameter= 2.4 cm) with a sintered glass frit at the bottom. A heat exchange glass coil (inner diameter= 0.5 mm) surrounded the reactor and entered the inner glass tube at the bottom just below the glass frit. The nitrogen gas was heated by passing through this coil prior to entering the reactor, which was immersed in an oil bath kept at 160 ºC. The nitrogen flow was controlled by a flow-meter. Typically, 0.4 g of the PBT/Thx physical mixture was placed on the sintered glass plate. The powder was fixed in place by addition of glass pearls (diameter= 2 mm) on top of the powder, and the reactor was purged with a nitrogen flow of 0.5 L·min-1 during 30 min prior to the reaction. Reactions were left to proceed at 160 ºC under a 0.5 L·min-1 nitrogen flow until the desired molecular weights were attained, which were followed as function of reaction time using SEC. Reaction times were in the 4-20 h range. After completion of the reaction, the reactor was gradually cooled down to room temperature by removing the heat source and by continuing the purging of the reactor with nitrogen. Subsequently, the obtained polymer was dried under vacuum. The Thx copolyesters obtained via SSM are abbreviated as SSMPBxThxyT, where x and y are the mole percentages (mol-%) of 1,4-butanediol and Thx, respectively, in the resulting copolyester. NMR characterization of SSMPBxThxyT copolyesters. 1H NMR (300.1 MHz, CDCl3/TFA), δ (ppm): 8.1 (s, 4H, ArH), 5.3 (s, y·2H, OCH2O), 4.6 (m, y·4H, OCH2CH), 4.5 (t, x·4H, OCH2CH2), 4.5 (m, y·2H, OCH2CH), 2.0 (t, x·4H, OCH2CH2). 13C NMR (75.5 MHz, CDCl3/TFA), δ (ppm): 168.0 (CO), 167.0 (CO), 134.2-133.2, 130.3, 130.1, 95.7, 75.9, 66.3, 64.9, 25.4. 6.3.2.4. Hydrolytic degradation procedures Films for the hydrolytic degradation studies on the (co)polyesters were prepared with a thickness of approx. 200 µm by casting from solution (100 g·L-1) in a mixture of chloroform and HFIP (5:1). The films were cut into disks with a diameter of 10 mm and a
Chapter 6 308 weight of 20 to 30 mg, which were subsequently dried under vacuum to constant weight. For the hydrolytic degradation, samples were immersed in vials containing 10 mL of citric acid buffer (pH 2.0) at 80 ºC. After incubation for a predetermined period of time, the samples were rinsed thoroughly with distilled water, dried to constant weight and analyzed by SEC chromatography, NMR spectroscopy and SEM microscopy. For hydrolytic degradation studies of the monomer 2,3-di-O-methylene-L-threitol, samples of this diol (80 mg) were immersed in NMR tubes containing 1 mL of citric acid buffer (pH 2.0), sodium phosphate buffer (pH 7.4) or sodium carbonate buffer (pH 10.5), all of them prepared in D2O, and were incubated at 80 ºC for 6 weeks. The residue left after incubation was analyzed by NMR spectroscopy. 6.3.3. Results and discussion 6.3.3.1. Polymer synthesis The bio-based compound 2,3-di-O-methylene-L-threitol (Thx) was prepared following the route depicted in Scheme 6.2. The pendant hydroxyl groups from naturally occurring L-tartaric acid were protected by internal acetalization leading to the formation of a 1,3-dioxolane ring. The resulting product, dimethyl 2,3-di-O-methylene-L-tartrate, was subsequently reduced to obtain the desired Thx diol monomer. This cyclic diol possesses a 2-fold axis of symmetry, and therefore will produce stereoregular polymer chains and its two primary hydroxyl groups are expected to display the same reactivities towards a polycondensation reaction. Scheme 6.2. Preparation of the bio-based 2,3-di-O-methylene-L-threitol.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 309 Thx was used as diol comonomer in the synthesis of the bio-based copolyesters by two different approaches, viz. melt polycondensation (MP) and solid-state modification (SSM), as depicted in Scheme 6.3. Scheme 6.3. Preparation of MPPBxThxyT and SSMPBxThxyT copolyesters. MPPBxThxyT copolyesters were obtained by reacting dimethyl terephthalate (DMT) with mixtures with varying compositions of Thx and 1,4-butanediol in the melt. Transesterification reactions were initiated at 160 ºC in order to prevent volatilization of the diols. As the reaction progressed the temperature was progressively increased to avoid crystallization of oligomers. The subsequent polycondensation reactions were performed under reduced pressure at temperatures in the range of 220-250 ºC. Lower temperatures and longer reaction times were used for copolyesters with higher Thx contents, to avoid decomposition of this thermally sensitive compound. Dibutyl tin oxide (DBTO) was the catalyst of choice as replacement of the commonly used titanium (IV) tetrabutoxide (TBT). Results obtained from the synthesis of dimethyl 2,3:4,5-di-O- methylene-galactarate-based aliphatic polyesters by melt polycondensation demonstrated the higher activity of DBTO as catalyst compared to TBT (Lavilla et al., 2011 -Subchapter 3.2-). MPPBxThxyT copolyesters were obtained with weight average molecular weights between 39 and 43 kg·mol-1 and polydispersity indices in the range of 2.1-2.4 (Table 6.4). Besides the copolyesters prepared by MP, solid-state modification (SSM) was used to prepare copolyesters with similar overall chemical compositions. Usually, SSM is performed at 20-30 ºC below the melting point of a semicrystalline polymer (Jansen et al., 2005 and 2006; Gubbels et al., 2013b). However, in this present study the SSM of poly(butylene terephthalate) (PBT, Tm= 225 ºC) with the cyclic diol Thx was conducted at 160 ºC under a relatively low nitrogen flow, to minimize the volatilization of the carbohydrate-based monomer. DBTO was also used as catalyst for the SSM reactions, because of its high activity in the low temperature SSM processes (Lavilla et al., 2013c -Subchapter 6.2-). All the SSM reactions were monitored using sizeexclusion chromatography (SEC) to ensure that the SSMPBxThxyT copolyesters had similar molecular weights compared to their MP counterparts, in order to make a reliable
Chapter 6 316 The DSC analysis revealed that the incorporation of Thx into the backbone of PBT induced significant changes in the glass-transition temperature (Tg) when compared to neat PBT (Table 6.5). The dependence of the Tg values on the Thx content is plotted in Figure 6.11. Remarkably, PThxT has a Tg value 50 °C higher than PBT, proving a much more rigid main chain structure. Consequently, the partial replacement of the 1,4- butanediol units by Thx residues yields an increase in Tg, which shows no dependency on the copolyester preparation method (MP or SSM) . Figure 6.11. Glass-transition temperature versus composition plots for the MPPBxThxyT and the SSMPBxThxyT copolyesters. The triangle represents the Tg of the homopolyester PBT. Melting temperatures (Tm) and enthalpies (ΔHm) were measured by DSC and are listed in Table 6.5. PBT is a semicrystalline polyester with a melting temperature of 225 ºC. Copolymerization of PBT with other compounds commonly affects the thermal properties significantly. Endothermic peaks, indicative for the presence of a crystalline phase, were observed for all MPPBxThxyT and SSMPBxThxyT copolyesters, whereas PThxT did not show such a behavior and proves to be fully amorphous. The Tm of the copolyesters proved to be dependent on the Thx content and also on the preparation method. Since in the first heating run the samples prepared by MP or SSM have a different thermal history, the SSM samples being annealed during the SSM treatment, a comparative analysis of the Tm will be performed on the results obtained from the second heating run, so after crystallization from the melt (discussed below). Melting and crystallization enthalpies of the MP and SSM copolyesters, however, will be compared for first heating, cooling and second heating.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 317 From DSC analysis of the obtained copolyesters it is clear that the materials prepared by SSM exhibit higher first heating melting enthalpies compared to those prepared by MP exhibiting a similar overall chemical composition (Table 6.5). This higher enthalpy is due to the segmented nature of these SSM-prepared copolyesters and annealing of the copolyester during the SSM reaction. In order to determine the crystallinity of the Thx-based samples, to elucidate if co-crystallization had taken place and to complement the DSC data, wide-angle X-ray diffraction (WAXD) was performed on the two series of as synthesized copolyesters as well as on the PThxT and PBT homopolyesters. The WAXD profiles and the most important Bragg spacings present therein are shown Annex H. The pattern of PBT is characterized by five prominent reflections at 5.6, 5.1, 4.3, 3.8 and 3.6 Å. Essentially the same pattern, in terms of spacing and relative intensities, is shared by all MPPBxThxyT and SSMPBxThxyT copolyesters, revealing that the triclinic crystal structure of PBT (Hall, 1984) has been retained and that co-crystallization of ThxT repeat units in the PBT crystals does not occur. In agreement with DSC results, PThxT did not show any discrete reflections characteristic of crystalline material. The crystallinity index of the materials was estimated as the quotient between the crystalline and total area of the X-ray diffraction patterns (Annex H). A noticeable decrease in crystallinity with increasing Thx content was observed for the MP series, which was expected in view of the lack of co-crystallizability of the ThxT repeat units with BT repeat units. Regarding SSMPBxThxyT copolyesters, a slight decrease in crystallinity was also observed during first heating when the Thx content was increased, indicating that the crystalline phase was not completely retained during the SSM process. This phenomenon was observed before in other copolyesters prepared by SSM, and can be attributed to the participation of the outer side of the PBT crystallites in trans-reactions, thereby gradually dissolving part of the crystallites (Jansen et al., 2008; Lavilla et al., 2013c -Subchapter 6.2-). From the data presented in Table 6.5, it can be noted that all the MPPBxThxyT and SSMPBxThxyT copolyesters were able to crystallize from the melt, and that the crystallization temperature (Tc) and the crystallization enthalpy (ΔHc) both decrease with increasing Thx content. Besides the fact that Thx-based repeat units cannot (co)- crystallize (see earlier), the increasing amount of stiffer Thx containing segments makes crystallization from the melt more difficult. Nevertheless, when the results for the MP series are compared to those obtained for copolyesters prepared by SSM at similar compositions, the materials obtained by SSM showed higher Tc and ΔHc values. This was expected, since longer PBT sequences are present in the SSM-prepared copolyesters,
Chapter 6 318 which facilitate the crystallization of the materials. A further detailed crystallization study will be described below. The thermal properties obtained from the second heating run of the MP- and SSM-prepared samples can be comparatively studied since they have similar thermal histories and have all been crystallized from the melt. After this melt-crystallization, MPPBxThxyT copolyesters recovered about 70-95% of their initial crystallinity and displayed no changes in their melting temperatures. However, when the SSM copolyesters were crystallized from the melt and reheated, there was a definite change in their melting behavior. Both the Tm and ΔHm decreased significantly when the first and second heating were compared. A reason for this difference is the (partial) randomization of the chemical microstructure by transesterification reactions occurring in the melt. This randomization is an entropically driven process and is catalyzed by the presence of DBTO, which was added to the physical mixtures to facilitate the Thx incorporation. To study the randomization phenomenon, samples of SSMPB82Thx18T were kept in the melt at 250 °C for predetermined times and after this treatment their chemical microstructure and thermal properties were studied using 13C NMR and DSC, respectively (Annex H). A decrease of the melting temperature is observed as a function of the time during which the sample was kept in the melt, which was accompanied by an increase of the R-value from 0.73 to 0.92, showing that almost complete randomization of the chemical microstructure occurred after long residence times at 250 °C. Therefore, the changes in the thermal behavior between the first and second heating run are caused by the (at least partial) randomization of the chemical microstructure yielding shorter crystallizable PBT sequences. When the Tm values obtained from the second heating runs of the copolyesters prepared by SSM are compared to those prepared by MP, noticeable differences are observed and an overview is given in Figure 6.12. Interestingly, both Tm and ΔHm recorded during the second heating run of the samples prepared by SSM are higher than those measured for the MPPBxThxyT series, the higher melting points of the SSM samples being related to the possible formation of thicker lamellae from the longer PBT sequences. This implies that even though randomization of the blocky chemical microstructure of the SSM-prepared copolyesters does occur to some extent during recording the DSC traces, it is not yet so extensive as to yield fully random copolyesters for which only relatively thin lamellae with low melting points can be formed.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 319 Figure 6.12. Overview of the melting temperature during the second heating run of the Thx-based copolyesters prepared by MP and SSM. The triangle represents the homopolyester PBT. 6.3.3.4. Isothermal crystallization As mentioned above, all the MPPBxThxyT and SSMPBxThxyT copolyesters as well as PBT are able to crystallize from the melt. Since crystallization from the melt is relevant for processing, the isothermal crystallization of MPPB96Thx4T, MPPB91Thx9T, SSMPB96Thx4T, SSMPB91Thx9T and PBT were comparatively studied in the 195-205 ºC temperature interval. All the polyester materials could be isothermally crystallized at 200 °C. The evolution of the relative crystallinity, Xc, versus crystallization time is shown in Figure 6.13. Figure 6.13. Relative crystallinity versus time plot of PBT, MPPB96Thx4T, MPPB91Thx9T, SSMPB96Thx4T and SSMPB91Thx9T isothermally crystallized at 200 ºC.
Chapter 6 320 Data obtained from the isothermal crystallization experiments are displayed in Table 6.6. The observed onset and half-crystallization times, as well as the corresponding calculated Avrami parameters, are given for each experiment. It can be noted that for all the copolyesters the Avrami exponent n increases with temperature, the values obtained being in the 2.0-2.6 range. The double-logarithmic plot of these data indicated that only primary crystallization takes place in the selected time interval (Annex H). An increase in crystallization temperature yielded a delay in the onset of crystallization besides the decrease in crystallization rate. Moreover, it was observed that the presence of minor amounts of Thx noticeably depressed the crystallizability of PBT. Furthermore, the comparison of crystallization data for MPPB96Thx4T and SSMPB96Thx4T evidenced that crystallizability is more repressed for a fully random copolyester prepared by MP than for the more blocky SSM-prepared copolyester. In accordance, SSMPB91Thx9T was found to crystallize at 200 ºC with a higher crystallization rate than the analogous MPPB91Thx9T copolyester. The conclusion that can be drawn from this comparative study is that the presence of Thx units depresses the crystallizability of terephthalate-based polyesters, and that the crystallization rate depends on the chemical microstructure of the polymeric material. SSM allows for the preparation of more readily crystallizable copolyesters, since long PBT sequences are retained during the SSM process. Table 6.6. Isothermal crystallization data. Copolyester Tc t0 t1/2 N -logk Tm (ºC) (min) (min) (ºC) PBT 200 0.19 0.82 2.14 -0.25 223.9 205 0.51 2.68 2.45 1.02 225.5 MPPB96Thx4T 195 0.41 4.82 2.56 1.77 210.0 200 0.55 12.77 2.61 2.99 212.1 MPPB91Thx9T 195 1.58 9.52 2.06 2.04 207.0 200 6.64 33.01 2.08 3.14 209.5 SSMPB96Thx4T 195 0.19 1.54 2.31 0.48 215.8 200 0.36 4.74 2.62 1.82 217.9 SSMPB91Thx9T 195 0.23 5.95 2.01 1.68 213.0 200 1.74 17.29 2.17 2.74 215.2
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 321 6.3.3.5. Hydrolytic degradation To investigate the influence of the incorporation of Thx on the hydrolytic degradation of PBT, a comparative study of the two homopolyesters PThxT and PBT in addition to the copolyesters with 18% Thx prepared by MP and SSM was carried out at pH 2.0 at 80 ºC. Under these conditions, it is possible to study the degradation in a reasonable amount of time because of the relatively high hydrolysis rate. Changes in sample weight and molecular weight at increasing incubation times are depicted in Figure 6.14. Figure 6.14. Degradation of PBT, MPPB82Thx18T, SSMPB82Thx18T and PThxT at pH 2.0 at 80 ºC. Remaining weight (a) and molecular weight (b) versus time.
Chapter 6 322 PThxT underwent a weight loss of approx. 16% after six weeks of incubation time, which was accompanied by a decrease in the molecular weight of about 50%. This is in contrast to PBT, because the latter did not show any weight loss or significant decrease in molecular weights under these conditions. The decrease in the sample weight and the molecular weight were also noticeable for MPPB82Thx18T and SSMPB82Thx18T, evidencing the positive effect of the Thx units on the hydrolytic degradability. Furthermore, SEM analysis showed more apparent physical degradation of the surfaces of MPPB82Thx18T, SSMPB82Thx18T and PThxT compared to PBT (Annex H). The decreases in sample weight and molecular weight of the two copolyesters were found to be intermediate between those of PBT and PThxT homopolyesters, although the copolyesters prepared by MP showed more extensive hydrolytic degradation compared to the sample prepared by SSM. This can be explained by the higher crystallinity of SSMPB82Thx18T compared to MPPB82Thx18T. The relatively long PBT sequences in the SSM sample do limit the degradation. The melting enthalpies of the initial SSM and MP samples obtained by solvent casting were 45.9 and 28.2 J·g-1, respectively. It is noteworthy to mention that the melting enthalpies of the remaining parts of both the copolyesters increased up to 59.3 and 46.6 J·g-1 after 6 weeks of incubation, respectively. Such an increase in crystallinity is indicative that hydrolysis has taken place preferably in the amorphous phase, as is usually observed in semicrystalline polymers. Figure 6.15. 1H NMR spectra in CDCl3/TFA of PThxT after incubation at pH 2.0 at 80 ºC for 6 weeks (top) and initial sample (bottom).
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 323 To gain insight into the degradation mechanism of the polyester chain at the molecular level, 1H NMR spectra were recorded for the polymer residues after incubation (Figure 6.15 and Annex H). The spectrum of the residue of PThxT displayed, in addition to the signals characteristic for the polymer, an increase in the signal arising from CH2OH end groups. This is indicative of the reduced molecular weight, in full agreement with data provided by the SEC analysis. Furthermore, a full stability of the cyclic Thx structure against hydrolysis can be inferred from the total absence of any signal indicative of the hydrolysis of the acetal group in the MPPB82Thx18T, SSMPB82Thx18T and PThxT spectra. This result is rather striking because acetal groups are known to be sensitive to acidic conditions (Smith and March, 2007), and the opening of the dioxolane ring might be therefore expected to happen to some extent. To corroborate the stability of the cyclic structure against hydrolysis, 2,3-di-O-methylene-L-threitol was incubated in aqueous buffer at pH 2.0, 7.4 and 10.5 at 80 ºC for 6 weeks. The spectra recorded at the end of the incubation period are shown in Annex H; all spectra correspond to the structure of the original diol without any sign which would indicate hydrolysis of the acetal group taking place. 6.3.4. Conclusions The cyclic diol 2,3-di-O-methylene-L-threitol (Thx), derived from naturally occurring tartaric acid, has been used as carbohydrate-based building block in the preparation of poly(butylene terephthalate) (PBT) copolyesters. Two series of partially bio-based copolyesters with comparable molecular weights were prepared by either melt polycondensation (MP) of Thx with dimethyl terephthalate and 1,4-butanediol, or by the solid-state modification (SSM) of PBT with Thx. Copolyesters with compositions very close to the feed ratio and a random chemical microstructure were obtained by MP. Conversely, SSM yielded only the partial incorporation of the Thx in the PBT backbone and led to a limited range of compositions. These copolyesters had a block-like overall chemical microstructure, formed by PBT sequences present in the crystalline phase during the SSM process and a practically random amorphous phase, which is taking part in the transesterification reactions. However, the blocky microstructure of the copolyesters prepared by SSM randomized after prolonged times in the melt. The incorporation of the cyclic bio-based Thx afforded copolyesters displaying a satisfactory thermal stability and exhibiting increased Tg values. Furthermore, the unique
Chapter 6 324 block-like chemical microstructure obtained after SSM yielded copolyesters with remarkable thermal properties. The SSM-prepared materials showed better crystallizability from the melt because of the long PBT sequences present in the backbone of these copolyesters. Furthermore, they displayed higher Tm values and a higher crystallinity compared to their MP counterparts, although these properties were not completely recovered after crystallizing from the melt as found for the MP-prepared samples. The presence of the Thx significantly increased the hydrolytic degradability of PBT. This effect was more pronounced for the materials prepared by MP because of their lower crystallinity. The degradation of Thx-containing polyesters happened through hydrolysis of the main chain ester bonds without alteration of the acetal structure. 6.3.5. References Alla, A.; Rodríguez-Galán, A.; Muñoz-Guerra, S. Polymer 2000, 41, 6995-7002. Alla, A.; Oxelbark, J.; Rodríguez-Galán, A.; Muñoz-Guerra, S. Polymer 2005, 46, 2854- 2861. Blair, G.T.; DeFraties, J.J. Hydroxy Dicarboxylic Acids. Kirk-Othmer Encyclopedia of Chemical Technology, 2000. DOI: 10.1002/0471238961.0825041802120109.a01 Dhamaniya, S.; Jacob, J. Polymer 2010, 51, 5392-5399. Fenouillot, F.; Rousseau, A.; Colomines, G.; Saint-Loup, R.; Pascault, J.P. Prog. Polym. Sci. 2010, 35, 578-622. Galbis, J.A.; García-Martín, M.G. Sugars as Monomers. In: Belgacem, M.N.; Gandini, A.; Eds. Monomers, Polymers and Composites from Renewable Resources; Elselvier: Oxford, 2008, 89-114. Galbis, J.A.; García-Martín, M.G. Chem. Curr. Top. 2010, 295, 147-176. Gallucci, R.R.; Patel, B.R. Poly(butylene terephthalate). In: Scheirs, J.; Long, T.E.; Eds. Modern Polyesters, Chemistry and Technology of Polyesters and Copolyesters; John Wiley & Sons: Chichester, 2004, pp. 293-321. Gandini, A.; Coelho, D.; Gomes, M.; Reis, B.; Silvestre, A. J. Mater. Chem. 2009, 19, 8656-8664. Gandini, A. Green Chem. 2011, 13, 1061-1083. Gómez, R.; Varela, O. Macromolecules 2009, 42, 8112-8117.
SSM-prepared PBT copolyesters from cyclic acetalized carbohydrate-based monomers 325 Gubbels, E.; Jasinska-Walc, L.; Koning, C. J. Polym. Sci., Polym.Chem. 2013, 51, 890- 898. (2013a) Gubbels, E.; Jasinska-Walc, L.; Hermida Merino, D.; Goossens, H.; Koning, C. Macromolecules 2013, 46, 3975-3984. (2013b) Hall, I.H. The Determination of the Structures of Aromatic Polyesters from their Wideangle X-Ray Diffraction Patterns. In: Hall, I,H.; Ed. Structure of Crystalline Polymers; Elselvier Applied Science: UK, 1984, pp. 39-78. Jansen, M.A.G.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Schick, C.; Koning, C.E. Macromolecules 2005, 38, 10658-10666. Jansen, M.A.G.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Koning, C.E. Anal. Chim. Acta 2006, 557, 19-30. Jansen, M.A.G.; Wu, L.H.; Goossens, J.G.P.; de Wit, G.; Bailly, C.; Koning, C.E; Portale, G. J Polym. Sci., Polym. Chem. 2008, 46, 1203-1217. Japu, C.; Martínez de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. Polymer 2013, 54, 1573- 1582. Kiely, D.E.; Chen, L.; Lin, T.-H. J. Am. Chem. Soc. 1994, 116, 571-578. Kiely, D.E.; Chen, L.; Lin, T.-H. J. Polym. Sci., Polym. Chem. 2000, 38, 594-603. Kint, D.P.R.; Wigstrom, E.; Martínez de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. J. Polym. Sci., Polym. Chem. 2001, 39, 3250-3262. Kricheldorf, H.R. Makromol. Chem. 1978, 179, 2133-2143. Kricheldorf, H.R.; Behnken, G.; Sell, M. J. Macromol. Sci. Part A Pure Appl. Chem. 2007, 44, 679–684. Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Benito, E.; García-Martín, M.G.; Galbis, J.A.; Muñoz-Guerra, S. Biomacromolecules 2011, 12, 2642-2652. -Subchapter 3.2- Lavilla, C.; Muñoz-Guerra, S. Polym. Degrad. Stabil. 2012, 97, 1762-1771. -Subchapter 5.3- Lavilla, C.; Martínez de Ilarduya, A.; Alla, A.; Muñoz-Guerra, S. Polym. Chem. 2013, 4, 282-289. (2013a) -Subchapter 4.3- Lavilla, C.; Alla, A.; Martínez de Ilarduya, A.; Muñoz-Guerra, S. Biomacromolecules 2013, 14, 781-793. (2013b) -Subchapter 3.4- Lavilla, C.; Gubbels, E.; Martínez de Ilarduya, A.; Noordover, B.A.J.; Koning, C.E.; Muñoz-Guerra, S. Macromolecules 2013, 46, 4335-4345. (2013c) -Subchapter 6.2- Li, W.D.; Zeng, J.B.; Lou, X.L.; Zhang, J.J.; Wang, Y.Z. Polym. Chem. 2012, 3, 1344- 1353. Lichtenthaler, F.W. Carbohydrates as Organic Raw Materials. Ullmann’s Encyclopedia of Industrial Chemistry, 2010. DOI: 10.1002/14356007.n05_n07
Chapter 7 332 7.2.1. Introduction The urgent need to reduce the amount of petroleum consumed in the industry, as well as to minimize the impact of the use of plastics on the environment is attracting important efforts in research by following different approaches. Thus, significant increases in the recycling rates of polymers are taking place with facilities and changes in recent legislation that now permits even recycling of food contact polymers (Wool and Sun, 2005; Welle, 2011; Grause et al., 2011). On the other side substitution of petroleumbased monomers by those coming from renewable sources has become one of the most important challenges in current polymer research (Belgacem and Gandini, 2008; Williams and Hillmyer, 2008). Among the renewable naturally occurring sources, carbohydrates stand out because they are hugely abundant, readily available and able to provide great functional diversity; an increasing activity in the research of carbohydrate-based polymers is reflected in the recently reported literature (Galbis and García-Martín, 2008; Narain, 2011; Gandini, 2011). Aromatic polyesters, poly(ethylene terephthalate) (PET) and poly(butylene terephthalate) (PBT) in particular, are high performance thermoplastic materials that are massively used in a wide diversity of applications due to their excellent mechanical and thermal properties in addition to other more specific ones. The high resistance to atmospheric action and biological agents exhibited by these materials is an added value when they are intended to be used in long-term applications. On the contrary, they are considered unfriendly compounds when used in short-term applications because their non-renewable origin and great resistance to degradation (Müller et al., 2001; Rychter et al., 2010). In this regard, the development of aromatic polyesters from natural sources is currently drawing an enormous interest. To this purpose a really attractive approach is to unfold sustainable processes able to produce the traditional monomers implied in their synthesis. Thus ethylene glycol can be produced via oxidation and subsequent hydrolysis of bio-based ethylene (Chen and Patel, 2012), and 1,4-butanediol by hydrogenation of succinic acid obtained from corn (Delhomme et al., 2009); also the production of terephthalic acid from limonene has been recently patented (Berti et al., 2010). However the industrial application of these methods is still far from realization and monomers used today for the synthesis of industrial aromatic polyesters are almost entirely produced from petrochemical feedstocks. Another possibility of creating sustainable copolymers of PET and PBT is replacing either the diol or the diacid by other bio-based monomers. This is an
Cyclic acetalized alditols compared to isosorbide as comonomers for terephthalate copolyesters 333 extremely appealing option provided that the properties of the original polymers are improved, or at least not significantly impoverished. The high glass-transition temperature (Tg) is a greatly appreciated property in those polyesters addressed to packaging under heating or to the manufacture of containers for pressurized gaseous beverages; copolymerization with cyclic diols such as 1,4-cyclohexanedimethanol (Turner, 2004; González-Vidal et al., 2009) or 1,3- ciclobutanediol (Booth et al., 2006) has been a common strategy to obtain aromatic polyesters with enhanced Tg. Carbohydrate-derived monomers with a cyclic structure have achieved a privileged position because, in addition to their natural origin, they are able to provide polyesters with higher glass-transition temperatures while essentially retaining their original pattern of behavior. Thus, the olden known 2,5-furandicarboxylic acid is receiving in these last years a great deal of attention as a serious alternative to replace terephthalic acid in polyterephthalates (Gandini et al., 2009; Gomes et al., 2011). A decade ago, the bicyclic anhydride 1,4:3,6-dianhydro-D-glucitol, known as isosorbide, emerged with great pushing force as an adequate monomer for replacing aliphatic diols in PET and PBT (Storbeck and Ballauff, 1996; Kricheldorf, 1997). Very recently we have shown that internally diacetalized alditols with a bicyclic structure are also carbohydrate derivatives very suitable to prepare aromatic copolyesters displaying satisfactory general properties and enhanced Tg (Lavilla et al., 2012a -Subchapter 5.2-, 2012b -Subchapter 5.4- and 2013 -Subchapter 4.3-; Japu et al., 2012). Given the structural proximity between isosorbide and diacetalized alditols, as well as their common potential use as polycondensation monomers, it is encouraging to make a comparative evaluation of their suitability for the synthesis of aromatic polyesters. This paper has been thought with such aim and it reports a comparative study of PBT polyesters and copolyesters obtained by replacing 1,4-butanediol by 1,4:3,6-dianhydro-D- glucitol (Is), 2,3:4,5-di-O-methylene-galactitol (Galx), and 2,4:3,5-di-O-methylene-D- mannitol (Manx). They will be called PBxIsyT, PBxGalxyT and PBxManxyT, where x and y stand for the mole percentages (mol-%) of 1,4-butanediol and sugar-based bicyclic diol, respectively, in the resulting copolyester. The study examines the reactivity of the three bicyclic diols in the polycondensation with dimethyl terephthalate (DMT) under the same polymerization conditions and makes a comparative evaluation of the influence that each monomer exerts on the glass-transition temperature and crystallinity of the polyesters.
Chapter 7 334 Compared structure and properties of Is, Galx and Manx 1,4:3,6-dianhydro-D-glucitol (Is) is the dianhydride generated in the dehydration of D-glucitol (sorbitol). Sorbitol is prepared by hydrogenation of D-glucose coming from cereal starch. Is is the only bicyclic carbohydrate-based monomer commercially available today at industrial level (Fenouillot et al., 2010). It is composed of two cis-fused nearly planar tetrahydrofuran rings with a dihedral angle of 120º and the 2- and 5- hydroxyl groups in endo and exo positions, respectively. Since the two hydroxyl groups of Is are secondary, their accessibility to reagents is rather restricted, and due to their different spatial position in the molecule they will display different reactivity; in fact, a deficiency of the reactivity of the endo hydroxyl group has been detected in several polymerization studies recently reported (Noordover et al., 2006 and 2007; Fenouillot et al., 2010). 2,3:4,5-di-O-methylene-galactitol (Galx) is obtained from galactaric acid by acetalization of the four secondary hydroxyl groups with paraformaldehyde and subsequent reduction. The structure of Galx consists of two non-fused 1,3-dioxolane rings with the two primary hydroxyl groups in exo positions. Contrary to isosorbide, Galx is centrosymmetric so its two free hydroxyl groups display the same reactivity since they are spatially undistinguishable. 2,4:3,5-di-O-methylene-D-mannitol (Manx) is obtained by acetalization of 1,6-di-O-benzoyl-D-mannitol followed by hydrolysis of the benzoxy groups. Manx consists of two fused 1,3-dioxane rings and possesses a twofold axis of symmetry and, similarly to Galx, the two primary hydroxyl groups are in exo positions and display the same reactivity. A comparative summary of the main features of Is, Galx and Manx of relevance to this study is given in Table 7.1. Table 7.1. Main features of bicyclic diols. Compound Is Galx Manx Chemical name 1,4:3,6-dianhydro- D-glucitol 2,3:4,5-di-O- methylene-galactitol 2,4:3,5-di-O- methylene-D-mannitol Chemical structure O O H HOH HO OH HO O O O O O O O O OH HO Protecting group dianhydride diacetal diacetal OH groups secondary primary primary Cycle structure tetrahydrofuran 1,3-dioxolane 1,3-dioxane Symmetry 1 I C2 OH stereo positions exo/endo exo/exo exo/exo Melting point 60-63 ºC 100-102 ºC 139-140 ºC Origin D-glucose D-galactose D-fructose
Cyclic acetalized alditols compared to isosorbide as comonomers for terephthalate copolyesters 335 7.2.2. Experimental section The three series of polyesters, PBxIsyT, PBxGalxyT and PBxManxyT, subjected here to study were obtained according to the synthesis scheme depicted in Scheme 7.1. Scheme 7.1. Polymerization reactions leading to PBxIsyT, PBxGalxyT and PBxManxyT copolyesters. Five different copolymerizations, with molar feed ratios of 1,4-butanediol to sugar-based diol of 90:10, 80:20, 70:30, 60:40 and 50:50, were carried out for each bicyclic diol. Exactly the same reaction conditions were used to prepare copolyesters of different series with the same molar feed ratio (1,4-butanediol to sugar-based diol) for the three series. Specific conditions were used for each molar feed ratio, as they are detailed in Annex I. The synthesis and characterization of PBxGalxyT and PBxManxyT copolyesters and their parent homopolyesters has been reported previously in detail (Lavilla et al., 2012a -Subchapter 5.2- and 2012b -Subchapter 5.4-). The synthesis of the PBxIsyT family has been carried out now with the purpose of providing the data necessary for the comparative study aimed in this work. The common synthetic procedure was the following: The polymerization feed was a mixture of dimethyl terephthalate, 1,4-butanediol and the sugar-based diol in the selected compositions. A molar excess of diol mixture to dimethyl terepthalate was used, and dibutyl tin oxide (DBTO) catalyst in 0.6% molar respect to monomers as well as antioxidants Irganox 1010 (0.1% w/w) and Irgafos 126 (0.3% w/w) were added. The reaction was performed in the melt along two steps in a three-necked, cylindrical-bottom flask equipped with a mechanical stirrer, a nitrogen inlet and a vacuum distillation outlet. Firstly the transesterification reaction was carried out
Chapter 7 336 under a low nitrogen flow at temperatures in the 160-200 ºC range depending on composition. The polycondensation reaction was left to proceed at a temperature between 210 and 240 ºC under a 0.03-0.06 mbar vacuum. Then, the reaction mixture was cooled to room temperature, and the atmospheric pressure was recovered with nitrogen to prevent degradation. The resulting polymers were dissolved in chloroform or in a mixture of chloroform and trifluoroacetic acid (9:1), and precipitated in excess of methanol in order to remove unreacted monomers and formed oligomers. Finally, the polymer was collected by filtration, extensively washed with methanol, and dried under vacuum. The resulting polymers were extensively characterized and their thermal properties accurately evaluated in parallel. The chemical constitution, composition and microstructure of polyesters and copolyesters were determined by NMR and their thermal stability measured by thermogravimetry under inert atmosphere. Their melting and glasstransition behavior were examined by DSC and temperatures and enthalpies measured at heating using exactly the same recording conditions. The specific reaction conditions and results of the synthesis of the three series of polyesters as well as the technical and methodological details used in their characterization and properties evaluation are provided in Annex I. 7.2.3. Results and discussion 7.2.3.1. Synthesis: Molecular weight and composition The polymerization conditions used in the synthesis of homopolyesters and copolyesters were selected to imitate as far as possible those usually applied in the industrial practice; reactions were performed in the melt with total absence of solvents, and they were left to proceed through a two-steps sequence of increasing temperature and vacuum. Transesterification reactions were started at 160 ºC in order to prevent volatilization of diols, and left to proceed for 2-3 hours period along which temperature was progressively increased to avoid crystallization of early formed oligomers. Polycondensation reactions were carried out for 2-4 hours at temperatures in the 210-240 ºC range and under vacuum to facilitate the removal of volatile by-products; lower temperatures and longer reaction times were used for copolyesters with higher contents in sugar-based monomers. Antioxidants additive were added in the synthesis of PBxIsyT
Cyclic acetalized alditols compared to isosorbide as comonomers for terephthalate copolyesters 337 copolyesters to avoid discoloration, and dibutyl tin oxide (DBTO) was the catalyst of choice for the three series instead of the commonly used titanium (IV) tetrabutoxide (TBT). Our earlier results in the synthesis of aliphatic homopolyesters from the bicyclic diester dimethyl 2,3:4,5-di-O-methylene-galactarate and linear alkanediols demonstrated the higher activity of DBTO catalyst compared to TBT which allowed to proceed at lower reaction temperatures without increasing reaction times (Lavilla et al., 2011 -Subchapter 3.2-). Under softer reaction conditions, decomposition of the thermally-sensitive sugar compounds was minimized and higher molecular weights could be attained. The synthesis results obtained for PBT, the three copolyester series, PBxGalxyT, PBxManxyT and PBxIsyT, and their respective parent homopolyesters PGalxT, PManxT and PIsT are compared in Table 7.2. PBxGalxyT copolyesters were obtained from 1,4-butanediol, Galx and DMT with intrinsic viscosities of 0.7-0.9 dL·g-1 and weight-average molecular weights between 36,000 and 41,000 g·mol-1 (Lavilla et al., 2012a -Subchapter 5.2-). Correspondingly, PBxManxyT copolyesters were obtained when Manx was used as sugar-based comonomer instead (Lavilla et al., 2012b -Subchapter 5.4-). In this case, relatively high molecular weight copolymers were also achieved, with intrinsic viscosities between 0.8 and 1.2 dL·g-1 and weight-average molecular weights confined in the 38,000-52,000 g·mol-1 interval. On the contrary, the results obtained in the copolymerizations based on Is were not as satisfactory as those based on Galx and Manx; intrinsic viscosities and weight-average molecular weights of PBxIsyT copolyesters were in the 0.3-0.4 dL·g-1 and 15,000-20,000 g·mol-1 intervals, respectively. Although polycondensations in bulk involving Is have been usually carried out with TBT as catalyst (Noordover et al., 2006 and 2007; Kricheldorf et al., 2007; Inkinen et al., 2010); it was here replaced by DBTO by the reasons indicated above. The molecular weights reported for Is containing copolyesters obtained by using TBT catalyst are even lower than those obtained here with DBTO. It is worthy to note that molecular weight values are very similar within each series (with the only exception of the homopolyester entirely made of sugar-based diol), and that essentially the same molecular weight dispersity is obtained for all the prepared polyesters regardless of which series is concerned. This satisfactory agreement adds confidence to the comparative analysis of thermal properties that is carried out below.
[Document text truncated for crawler view.]