This action might not be possible to undo. Are you sure you want to continue?
Asaf Kleopas Sugih
The author thanks the University of Groningen for the financial support through an Ubbo Emmius Scholarship.
Synthesis and Properties of Starch Based Biomaterials
Proefschrift ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. F. Zwarts in het openbaar te verdedigen op vrijdag 12 december 2008 om 13.15 uur door Asaf Kleopas Sugih geboren op 4 juli 1975 te Bandung, Indonesië
Janssen Prof. dr.M. dr. dr. dr. Heeres Prof. Broekhuis Prof. dr.P. ir.A. L. dr. Minnaard Prof.J. Moscicki ISBN : 978‐90‐367‐3592‐6 ISBN : 978‐90‐367‐3593‐3 (electronic version) . H. Promotores : Prof. ir.B. L. F.J. A. A. Picchioni Beoordelingscommissie : Prof.
and my sister . to: Tresna. my parents.
1. Starch 1. Product analyses . Thesis Outline 188.8.131.52. Biomaterials from starch 184.108.40.206.2. Starch‐based blends by in situ polymerization 1.1. Biodegradable plastics from starch 1. Thermoplasticized starch 220.127.116.11.2.Table of Contents Chapter 1: Introduction 1. Introduction 2. Starch Esters 1.2.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (2) 27 28 26 27 27 27 2.4. Cross‐linked starch 18.104.22.168.1.2.1. Starch modifications to improve product properties 1.3. Starch ‐ biopolymer blends and graft co‐polymers 1. References 2 3 4 5 5 7 9 9 10 11 11 13 14 16 16 17 Chapter 2: Experimental Studies on the Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐ Monosaccharide Initiator System 22.214.171.124. Structure and properties of starch 1.2.4. Materials and Methods 2.3. The potential of biodegradable plastics 1. Typical example for the synthesis of polydioxanone end‐capped with 1. Methods 126.96.36.199.2.3.3. Plastic applications and waste issues 1. Starch‐based blends by melt mixing 1.3.4. Materials 2.2. Starch production processes 188.8.131.52.3.3.
184.108.40.206.6.4.1. Peracetylation of silylated starch 3.2. Results and Discussions 2. Product yield 2.3. Analytical methods 220.127.116.11.4.1.2. Nuclear Magnetic Resonance (NMR) 3.2. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers 51 51 52 52 52 53 54 48 49 49 50 50 3.1.2. Materials 18.104.22.168.1. Calculations 3.1. Product analyses 2.3.2. Typical example of in situ polymerization of ε‐CL with silylated starch 50 22.214.171.124. Materials and Methods 3. Effects of process conditions on the average chain length ( X n exp ) and end group distribution 28 29 29 30 36 38 39 40 42 43 44 2. 126.96.36.199. Nomenclature 2.3. References Chapter 3: Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors 3. Results and Discussions 3.2. Systematic studies 2. Mechanistic aspects 188.8.131.52.184.108.40.206.220.127.116.11. Calculation of average degree of polymerization 18.104.22.168.5. Synthesis of silylated starch viii . Methods 3. Typical example of the starch silylation procedure 3.3. Conclusions 2. Introduction 3. Screening experiments 2.2.
2. Results and Discussion 22.214.171.124.3.1.2. In situ ring opening polymerization of ε‐caprolactone with silylated starch 57 63 64 64 65 3.3.4. Deprotection of silylated‐starch‐g‐PCL 3.2. Peracetylation procedure 4. Materials 4.2.1. Product characterisation 126.96.36.199. Introduction 4.3. Catalysts screening 4. Analytical equipment 4. Methods 4.2. Systematic studies 188.8.131.52.3.6.4. Typical example of the synthesis of laurate and stearate esters of corn starch 72 73 74 75 75 76 76 78 79 80 81 81 82 83 84 70 71 71 71 72 4. 1H‐ and 13C‐NMR analyses 4. Exploratory experiments 184.108.40.206.1.2.3. FT‐IR measurements 4. Effect of the addition of toluene as a co‐solvent 4. Nomenclature 4.3.3. Effect of vinyl ester to AHG ratio on the product DS 220.127.116.11. Determination of the Degree of Substitution (DS) 4.3. Conclusions 18.104.22.168.22.214.171.124.3. 3.1. References Chapter 4: Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants 4. References ix . Nomenclature 3.2.3. Materials and Methods 4.6.5. Conclusions 4.3.
Materials and Methods 6.Chapter 5: Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch 5. Materials 6. Introduction 126.96.36.199.2. Materials 5.1.2. Preparation of PCL‐starch blends with the reactive compatibilisers 110 188.8.131.52. Experimental Design 5.2.2. Compatibilizer synthesis 184.108.40.206. Nomenclature 5. Results and Discussion 5.2.6. Materials and Methods 5.1.3. Product properties 5.2.3. Conclusions 5.2. Typical example of the preparation of laurate and stearate starch Esters 88 86 87 87 87 88 220.127.116.11.2.4. Analytical Equipment 18.104.22.168.22.214.171.124.126.96.36.199.2.5. Introduction 5.4.2. Work‐up of PCL‐g‐DEM 108 109 109 109 109 110 110 6.2. Methods 5.2. Work‐up of PCL‐g‐GMA products 6. Methods 6.4. References Chapter 6: Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 6. Analytical Methods 110 x . Mathematical modelling 5. Peracetylation procedure and Degree of Substitution (DS) determination 88 88 89 92 96 102 102 103 5.
Effect of substrate (GMA/DEM) to PCL Ratio on the FD 6.5. Binary blends of starch and PCL 6. Ternary blends compatibilized with PCL‐g‐GMA 6. 6. Effect of the BPO intake on the product FD 6. Preparation of the ractive compatibilizers 188.8.131.52.1. References Summary Samenvatting (Dutch Summary) Acknowledgements List of Publications 133 137 141 143 xi . Statistical mdeling 6.3. Nomenclature 6.2.3. Ternary blends compatibilized with PCL‐g‐DEM 184.108.40.206.220.127.116.11.4.1.3. Synthesis and Properties of Starch‐ PCL Blends 6. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers 6.1. Conclusions 18.104.22.168.3.3. Results and Discussions 22.214.171.124.1.1. Thermal properties of the compatibilizers 6.3.2. Modeling of the combined effects of the GMA/ DEM and BPO intakes on the FD 111 111 111 116 117 118 120 121 122 123 125 128 128 129 126.96.36.199.
an outline of this thesis is given. Different routes to modify starch to improve the product properties and to extend the application range will be provided. .Chapter 1 Introduction Abstract A general overview of starch properties will be given and the potential use of starch as a starting material for a wide range of green biomaterials will be reviewed and discussed. Finally.
1. sweet potato. Past.9 ton starch per hectare . current and future estimations of worldwide starch production are given in Figure 1. Plants synthesise and store starch in their structure as an energy reserve. . The current price of corn starch is around $0. 80 70 60 50 40 30 20 10 0 Starch Production (Million Tons) 1995 2000 Year 2005 2010 Figure 1.e.1. Starch is found in seeds (i.45 (estimated from the total volume and value of the US corn starch export ). Most of the starch produced worldwide is derived from corn. 4‐5]. but other types of starch such as cassava. and wheat starch are also produced in large amounts [2. The second and third starch producer regions are Europe and Asia [4‐5]. Potato accumulates starch to approximately 75 % of the dry weight in the tubers with a yield up to 21 ton starch per hectare. It is generally deposited in the form of small granules or cells with diameters between 1‐100 µm . potato. wheat. Past. corn. or peas) and in tubers or roots (i. maize. while corn seeds consist of 65‐80% starch by weight. potato or cassava) of the plants [2‐3].e. sorghum. rice. The worldwide production of starch in 2008 is estimated to be around 66 million tons .1. Starch 75% of all organic material on earth is present in the form of polysaccharides. barley. with an average yield of 4. Most of the starch is produced in the USA. . An important polysaccharide is starch.Chapter 1 1. Most starch crops are very productive. present and a forecast of starch production  o : Europe □ : USA : Total ∆ : the Rest of the World 2 .
The plant material is grounded in water. The starch is present in the endosperms (floury and horny). and is embedded in a proteinaceous cellular matrix as is shown in Figure 1. a. which is again fed to a second milling process.2. Introduction 1.b. After initial cleaning to remove cob. The resulting starch is further washed in a cyclone and finally dried. The resulting suspension from the mills contains fiber.a. As an example. The low density gluten is separated from the starch suspension by centrifugation. b. the mixture is fed to a hydrocyclone to separate the germ and the rest of the kernel. gluten. Cross‐sectional view of a corn kernel 3 . a typical corn starch production process is given in Figure 1. Typical corn‐milling operation b. the debris is filtered from the slurry.2.1.1. and starch granules are obtained after centrifugation from the suspension. After a coarse milling. the corn kernel is softened by steeping in warm water containing SO2 until the volume of the kernel increases with 55‐65%.2. and other foreign materials. and starch. Starch production processes Starch is generally extracted from the plant by wet milling processes [9‐10]. sand. Figure 1. The fiber is removed using washing screens. Corn starch production  a.
. the amylose content is between 18‐28% .3.Chapter 1 1....3. containing periodic branches linked with the backbones through α‐D‐(1.. Typically.2. The starch granules will swell rapidly to many times of its original volume. Amylose c. The molecular weight of amylopectin is higher than that of amylose and is typically 4‐5 x 108 Da . Anhydroglucose (AHG) unit b. .4)‐glucosidic bonds. The content of amylose and amylopectine in starch varies and largely depends on the starch source. H OH H2COH H2COH H .1. H OH H O H OH H O H H OH H c. The molecular weight of amylose is a function of the plant source and processing method. H2COH H H OH H O H OH O H OH O CH2 H O H H OH H O H OH H O . Chemical Structure of Starch a. H2COH H OH H O H H O b..1. H .. Amylopectin Starch is insoluble in cold water.) .6)‐ glucosidic bonds .. Amylopectin is a branched polymer. H2COH H . Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. The linear 4 .6‐7 x 105 Da ..a.. Two types of AHG polymers are usually present in starch: amylose and amylopectin [2‐3. Heating a starch solution leads to loss of hydrogen bonding in the interior of the starch granule and the starch will start to gelatinize. Structure and properties of starch Starch is a polymer consisting of anhydroglucose (AHG) units (see Figure 1. Each branch contains about 20‐30 anhydroglucose units... but it is very hygroscopic and binds water reversibly. Figure 1. H O H OH H OH H O H OH O H OH H O . The amylose content of several common starches is given in Table 1.. but usually in the range of 1. a. H OH H H2COH H2COH H O H H OH H H2COH H O H H OH H O H OH H O . . 6]..
5‐70 62‐72 63‐72 58‐64 1. and granule fragments. The gelatinization temperature range of various starch sources is given in Table 1. Plastic Applications and Waste Issues Plastic is the general term for a wide range of synthetic or semisynthetic polymerisation products. Plastics are used in a wide range of applications and the demand is still increasing every year .1. The resulting suspension contains a mixture of linear amylose molecules.2. Parker in 1838. The first generation of commercial plastics was derived from cellulose nitrate and is known as celluloid [12‐14]. Cellulose nitrate was first prepared by A. Biomaterials from starch 1. swollen granules.7 20 18. and celluloid was 5 . and.5 28 52 70‐75 27 15.8 17.2. depending on the amount of water present.8 16. will form a thick paste or gel. The gelatinization temperature range can be defined as the temperature at which granular swelling begins until the temperature when nearly 100% of the granules are gelatinized .5 25. Amylose content of common starches  Starch Arrowroot Corn Hybrid amylomaize Class V Hybrid amylomaize Class VII Oat Manioc Potato Rice Sago Sweet potato Tapioca Wheat Amylose (%) 20.2.7 26 Table 1. Introduction amylose molecules leach out of the granules into the solution. Table 1.2. Starch gelatinization temperature range  Starch Potato Tapioca Corn Waxy corn Wheat Gelatinization Temperature Range [o C] 59‐68 58.1.
5 0 1950 1960 1970 1980 Year 1990 2000 2010 Figure 1. Thermosets. the oldest purely synthetic plastic is Bakelite. A dramatic increase in demand for plastics began after World War II. and 37% of the total plastics demand in EU  is used as packaging materials. Plastics are very attractive materials. polystyrene (PS). Polyurethanes. In 2006.). and polycarbonates (PC). Important polymers used for packaging are polyethylene (HDPE and LDPE). for instance to be used in packaging and paper coatings . and PolyAcryl. Elastomers. About 29% of the total plastics produced in the USA . when polyethylene (PE) was invented. 15]. The production and consumption of plastics has increased significantly with a rate of almost 10% every year since 1950. the worldwide plastics production has reached 245 million ton per annum  (see Figure 1. While celluloid is derived from a natural polymer (cellulose). The latter is important when using the material for packaging purposes to save weight. and energy during transportation of goods. and therefore need less energy to shape it into useful materials . polyethylene terephtalate (PET).4. and PP-Fibers 2006: 245 2002: 200 200 Plastics Production (Million Tons) Not Included: PET-. Plastics have lower melting temperatures compared to glass and metals. Hyatt in 1870 . Adhesives.5. space. Plastics are also used for building materials and automotive. polyvinyl chloride (PVC).Fibers 150 1989: 100 100 1976: 50 50 1950: 1. PA-. PE is a very versatile plastic because it can be shaped easily into various forms.4. and consumer products (see Figure 1. polypropylene (PP). The largest application of plastics is for packaging purposes.Chapter 1 patented by J. Coatings. discovered by Baekeland in 1907 [12. They have a low density and can be shaped in thin films while maintaining good properties. Sealants. electrical.) 250 Included: Thermoplastics. Worldwide Plastics Production (1950‐2006)  6 .
plastics waste was ranked as the second major source of MSW after paper and paperboard. Recycling (part of) the plastics to reduce the amount of MSW also has limitations. 6]. and is only effective for single plastic sources or simple plastic formulations. rivers.2. is causing serious environmental problems. In the USA.5. Figure 1. Introduction Plastic waste. Plastic applications and waste treatment in Europe  1. however. ship navigation. Biodegradable plastics are polymeric materials capable of decomposing when given an 7 . 16]. hydropower plants operation. threatening fishery. irrigation. and other public works . and accumulation of its residues in soil cause significant reductions in agricultural yields. only 20% of the plastic is recycled. The disposal of plastics materials in municipal solid waste (MSW) is a serious issue in many parts in the world. soil. In Europe (2006). consisting of 25%‐v of the total waste . The volume fraction of plastic in MSW is much larger due to the low density of plastics.7% in 2006 .2. Plastic litter is hazardous to wildlife [1. It has a high volume to weight ratio and is resistant to degradation. the amount of plastics in MSW increased from less than 1% in the 1960 to 11. The potential of biodegradable plastics The application of biodegradable plastics could be an attractive solution for the problems related to the use of conventional plastics (vide infra). Most of the waste is still disposed (50%) by landfill or incinerated to recover the energy (30%) . and may be more than twice the weight fraction . and lakes. It is not applicable for thermoset resins . Co‐mingled plastics. which are usually found in waste streams. are not easy to recycle . In 1996. Plastics have been polluting sea [1.
because they take part in nature’s cycle of renewal.5/ 8 . Nowadays new types of biodegradable plastics with improved properties and lower costs have been developed . Several authorities have provided definitions for biodegradable plastics . Biodegradable plastics may be classified into two general groups. ASTM sub‐committee D20. DIN 103.2 working group on biodegradable polymers ‐ Biodegradation of a plastic material is a process leading to naturally occurring metabolic end products. Biopolymers are inherently biodegradable. It may also be treated in sewage sludge water treatment plants or buried in the soil [20‐21]. animals. The price of these biodegradable plastics is expected to be reduced considerably in the next decade . biological and/or physical forces in the environment at a rate which leads to fragmentation or disintegration of the plastics. Common natural biopolymers are carbohydrates. and polyesters from micro‐organisms . the price of PLA based plastics in 2010 is estimated to be around € 1. Biodegradable plastic waste may be treated in composting facilities. Japanese Biodegradable Plastic Society draft proposal ‐ Biodegradable plastics are polymeric materials which are changed into lower molecular weight compounds where at least one step in the degradation process is through metabolism in the presence of naturally occurring organisms. Biodegradable plastics have gained considerable interest since the 1980s. and microorganisms) and biodegradable synthetic polymers . The market price of biodegradable plastics (mainly from starch and PLA) has decreased the last years. and these are shown below: ISO 472: 1998 ‐ A plastic designed to undergo a significant change in its chemical structure under specific environmental conditions resulting in a loss of some properties that may vary as measured by standard test methods appropriate to the plastics and application in a period of time that determines its classification.Chapter 1 appropriate environment and sufficient amount of time . and since 2000 they have become competitive with traditional materials in some applications [20‐21]. proteins (present abundantly in plants and animals). together with food and yard waste as well as paper. biopolymers from nature (from plants. For instance.96 proposal ‐ Degradable plastics are plastic materials that undergo bond scission in the backbone of a polymer through chemical. The considerable growth of interest in composting as a means to replace landfill due to the decreasing disposal spaces (especially in Europe) may also help the progress of biodegradable plastics development . The change in chemical structure results from the action of naturally occurring micro‐ organisms.
consumer products and food products. chemical modification or combinations of them . half of its price in 2003 (around 3 Euro/kg). Biodegradable plastics are especially very useful for single‐use applications.4 per kg in 2007‐2008 ). These are summarized briefly in the next sections. napkins) . because the price of oil based polymers may still increase due to the rise in the crude oil prices . Virgin starch is not suitable as a packaging material. either by plasticization. hygienic materials (diapers. such as polyethylene (PE.4 or $ 1. Starch may become an attractive raw material for plastics in the future. 1. This price is much lower than conventional plastics derived from oil. towels. biofillers for tires. Biodegradable Plastics from Starch Starch is a very attractive source for the development of biodegradable plastics. starch must be modified. Introduction kg. The price of starch‐based biodegradable plastics (€ 1. blending with other materials. when recycling is not practical or uneconomical [20‐21]. The price of starch in 2007 was about $ 0. Biodegradable plastics are also used for food servicewares . 20]. alone or complexed with natural or synthetic biodegradable polymers) or from polylactic acid which originates from the fermentation of a starch feedstock. agricultural products (mulch films. pots).8‐ 2.2.50‐4.3. 1. € 1. tensile and flexural strength) and is too sensitive to water [26‐27]. 9 . or when environmental impacts have to be minimized. either from starch‐based materials (slightly modified starch. Consequently. .3. The global production capacity of starch‐based bioplastics in 2010 is estimated to increase to 200‐300 kiloton per year from 77‐200 kiloton in 2003 .45 per kg . Examples are the use as packaging material for carrier bags. breathable fabrics. before they can be applied as biodegradable plastics.2‐ 1. Starch modifications to improve product properties Several techniques may be applied to develop starch based biomaterials with improved properties. Bastioli [20‐21] showed that nearly all biodegradable plastics available in the market are derived from starch. and chewable items for pets [1. utensils .50/ kg in 2005) is also expected to be lower in the future due to considerable reduction in modification costs . It cannot be shaped in films with adequate mechanical properties (high percentage elongation.
The use of water as a plasticizer is not preferable. phase separation may still occur. 30. The plasticized starch properties may be tuned by changing the temperature of processing. extrusion. The use of other plasticizers (for example glycerol) results in a rubbery material. which will result in a continuous phase in the form of a viscous melt [29‐30. the semicrystalline structure of starch and its granular form are lost and the starch polymers are partially depolymerized. and blow moulding [28. destructed. the thermal properties of glycerol‐plasticized starch are a function of water content [31. and therefore are processable by conventional polymer processing techniques such as injection. As the result. 30]. 35. glycol. 40]. A biodegradation study according to ISO/CEN 14852 and ASTM D5209‐92 standards  on films made from starch–glycerol–water mixtures confirmed that the films are easily biodegradable.3. 32‐ 36]. 37]. The brittleness is known to increase in time due to free volume relaxation and retrogradation. . sugars) [30.Chapter 1 1. or melted starch [29. 40]. Although thermoplasticization seems to be a promising method. sorbitol.1. which is even above the thermal degradation temperature. For instance. Thermoplasticized starch Virgin starch is brittle and difficult to be processed into articles due to its relatively high glass transition temperature (Tg. 39]. water content. plasticized. because the resulting product will be brittle when equilibrated with ambient humidity . The modified products are known as thermoplastic. Starch can be modified to obtain materials which melt below the decomposition temperature . 33]. such as water and polyols (glycerol. There are several substances used as plasticizer for the preparation of thermoplastic starch (TPS). and hexylene glycol. glycerol. At intermediate water levels. Yu. approximately 230 oC) . and the properties and amount of plasticizers. The thermoplasticization process will decrease the interactions of the molecular chain and destruct the structure of the starch . The plasticizers are also usually hydrophilic and can be 10 . This problem is mainly caused by the presence of strong inter‐ and intra‐molecular hydrogen bonds between the starch macromolecules [28. resulted in the formation of an amorphous mass [35. Another issue is the poor water resistance and low strength. which still limits their use. The modification involves break down of the starch granular structure by the use of plasticizers at high temperatures (90‐180 oC) and shear. TPS synthesized from polyol and sugar plasticizers have the tendency to re‐crystallize (retrogradation) after being stored for a period of time. with better properties than virgin starch in various applications. et al  discovered that the elongation of break of the thermoplastic starch is significantly improved by plasticization with glycol. which results in embrittlement. 32.
01‐0. These hydroxyls can react easily with a wide range of compounds such as acid anhydrides. Most studies dealt with the synthesis of starch esters of C1‐C4 carboxylic acids.3‐1) are still soluble in water. Introduction washed out by water . The simplest starch ester is starch formate (C1). Three different types of starch acetates may be distinguished. Cross‐linked starch An anhydroglucose molecule of starch contains two secondary and one primary hydroxyl group. The cross‐linked products are therefore more rigid (and less elastic) materials than virgin starch . 1. textile. Low DS products (0. The products are called cross‐linked starches.2.3. The cross‐linked starches have found many applications. and adhesive industry. Cross‐linking results in a reduction of the solubility in water and to thickening.2) are commercially available and used as food additives and in the textile industry. The esterification reaction is catalyzed by H+ and proceeds with the formation of water . while highly substituted starch acetate (DS of 2‐3) are soluble in organic solvents. As a result. Higher processing temperatures generally led to higher cross‐linking levels. 46]. Substitution of the hydroxyl groups of starch with acetate groups makes the esters more hydrophobic than native starch. Solutions to improve the properties of TPS are blending. Starch Esters The development of starch esters started in the mid 19th century [44. organic chloro‐compounds. and ethylenic compounds. 39]. the use of formyl esters of starch is much lower than that of acetate esters. The DS is defined as the moles of substituents per mole of AHG units . aldehydes. Chemicals of these classes having two or more of the reactive groups may react with two or more hydroxyls of the starch molecules. Schuetzenberger acetylated starch with acetic anhydride . resulting in an increase of the Young’s modulus and tensile strength of the products and a decrease in elongation at break. synthesized by direct addition of formic acid to starch at room temperature. Breakdown of the starch and the formation of low molecular weight products occur to a significant extent . and particularly with acetic acid . epoxy. Medium DS starch acetates (0. The most popular starch ester is starch acetate . Already in 1865.3. leading to higher viscosities [42‐43]. 1. or coating with hydrophobic polymers [33. especially in the food. differing in Degree of Substitution (DS). High DS starch acetate can be easily casted 11 .3. paper. et al  studied the mechanical properties of films from maize starch cross‐linked by sodium trimetaphosphate (SMTP). Silva.
The kinetics of the reaction between gelatinized aqueous potato starch and vinyl acetate was studied by De Graaf. DMF.1. Starch OH + OH - Starch O - + H2O O - O Starch O - + H3C C O CH O H3C C O CH2 H3C C O CH O Starch CH2 H2O O Starch + H3C C H + OH - O H3C C O Starch O - + OH - H3C C O OH Starch H2O O H3C C OH + Starch O - Scheme 1. and is shown in Scheme 1. Another attractive route involves the use of vinyl esters as reagents . but also C‐6 and C‐ 3 substitution . or isopropanol) with acidic (hydrochloric or sulfuric acid) or alkaline (NaOH or triethylamine) catalysts . Starch esters have also been synthesized using alkanoyl chlorides [49‐53].Chapter 1 into films using organic solvents. allowing the polymer to be produced with a range of hydrophobicities .1. Mechanism of starch acetylation using vinyl acetate 12 . and phosphate) will result in C‐2. acetate. while alkaline catalysts (such as carbonate. Esterification may be performed using acid anhydrides in aqueous media or organic solvents (pyridine. Neutral and weak acid/ alkaline catalysts result in regioselective substitution at the C2 hydroxyl groups of starch. et al [55‐56]. The degree of acetylation of starch acetates can be easily controlled. xylene. DMSO. High DS starch acetates are thermoplastic materials suitable to be used as biodegradable plastics . hydrogencarbonate. Starch triacetate has been successfully synthesized using acetic anhydride in combination with pyridine‐gelatinized starch . Several synthetic routes have been developed for starch esters. Acetylation of starch in water and DMSO using vinyl acetate has been studied lately by Mormann and Al‐Higari  as well as Dicke .
Despite this advantage. As a result. The first involves simple melt mixing of the two components for example by extrusion. The second methodology for producing polymer blends involves the in situ polymerization of one component (thus originally present in the blend in monomeric form) in the presence of the second one. By working at temperatures above the melting point or glass transition temperature of the two components. polymers are rarely miscible with each other  so that.e. However. the two components are chemically linked to each other. Starch ‐ Biopolymer Blends and Graft Co‐Polymers Blending of different polymers is an established method to obtain products with improved properties.4. Their role is comparable to the one of a surfactant in emulsion formation . a chemical reaction might take place upon processing (reactive extrusion). however. The use of methyl and glyceryl esters of fatty acid (in the absence of solvent) to synthesize starch fatty acid esters has also been studied . However. which in turn provide a very strong adhesion at the molecular level between the dispersed phase and the matrix. The fatty acid chloride reactants are. in the simplest case of a binary blend. dodecanoyl. compatibilizers) might be used. but only relatively low‐DS (0. The degree of adhesion (binding) between the dispersed phase and the matrix is dependent on the molcular interactions between the two components and represents a crucial factor in determining the morphology of the blends and eventually the product performance .g. Introduction The synthesis of long chain fatty acid ester of starch has attracted much interest lately. The classical example of such process is represented by the production of High Impact Poystyrene (HIPS) obtained by styrene polymerization in the presence of polybutadiene .61) products could be obtained using this approach. in order to achieve also a good molecular adhesion between the phases by melt mixing.3. relatively expensive and rather corrosive. melt‐mixing remains the preferred route to polymer blends mainly because of very practical reasons: low costs. i. it is frequently used in order to chemically graft the polymerized component on the other one (polystyrene on polybutadiene in our example above ). Two main methodologies are applied for the production of polymer blends.34‐0. interfacial agents (e. they locate themselves at the interface between the 13 . one component will be dispersed into the other. availability of mixing equipment and no necessity to use any organic solvent (often employed for the in situ polymerization). Although the in situ polymerization process is not as technologically straightforward and economically convenient as melt‐mixing. the latter are mixed together. Fatty‐acid starch esters have been synthesized using fatty acid (octanoyl. 1. octadecanoyl) chlorides [49‐53]. If the right combination of chemical groups on the two components is present along the polymers backbone. The introduction of longer acid chain is expected to reduce the brittleness of virgin starch and to increase its hydrophobicity .
Chapter 1 two components stabilizing the dispersion (most probably by a steric repulsion mechanism ) and providing an improved adhesion at the interface. The synthesis of compatibilizer and its use for starch/ synthetic polymer blending has also been studied. Matrix Dispersed Phase Figure 1. The uncompatibilized blends of starch and 14 . In the past. 66‐81]. The products have been synthesized in the lab as well as on industrial scale [32. there have been efforts to blend as well as to graft synthetic polymers onto starch. A suitable interfacial agent for the blends of two polymeric materials is a block copolymer for which the chemical structure of every block is the same (or very similar) to the one of the individual components to be blended  (illustrated in Figure 1.3. 39. Interfacial agents already available on the market can be used as such or can be produced upon mixing (in situ) by chemical reaction of the two components.6. The most often used synthetic polymer for blending with starch is polyethylene [27.). The blending and grafting of starch with synthetic polymers is usually performed to achieve higher hydrophobicity and to improve the mechanical and thermal properties as well as to obtain cheaper and more biodegradable products. 64‐98]. The starch/polyethylene blends are used for agricultural mulch [27. Illustration of the role of interfacial agent in compatibilizing blends 188.8.131.52. Starch‐based blends by melt mixing Starch‐synthetic polymer blending has been studied as early as in 1973 [64‐65]. 80‐ 81] or food packaging .
is formed upon mixing. since the conventional synthetic polymers are usually poorly or non‐biodegradable. i. Despite these good results. As is the case of blends with PE. systematic studies on the compatibilizer precursor synthesis as well as on the influence of the chain topology and chemical reactivity (for the in situ compatibilizer formation) have not yet been reported. The ester bonds of these polymers are susceptible to attack by water and this leads to enhanced biodegradability. These biodegradable polyesters will finally decompose into non‐toxic products . thus improving the dispersion of starch in the PE matrix. among which synthetic polyesters are considered very promising materials . and poly‐lactones (such as poly‐butyrolactone. Polycaprolactone (PCL) is a well‐known synthetic biodegradable polyester. PCL‐g‐pyromellitic anhydride [94‐95]. thermal and water/gas permeability properties that are even comparable to bulk non‐biodegradable polymers such as PE and PP. As a result a graft copolymer (PE‐g‐Starch). The use of conventional synthetic polymer such as PE for blending with starch will only result in a partially biodegradable material. Studies on the blending of starch with PCL have been already described [32. The mechanical properties of these blends (tensile strength and elongation at break) decrease at higher starch content. 39. 72‐73]. PE‐g‐MA contains reactive anhydride sites. dextran‐g‐PCL .e. and EVOH . 96]. 15 . The mechanical properties of the blend are also improved. while the tensile strength of the compatibilized ones decreases only slightly with the starch content. poly‐lactides. the uncompatibilized blends of starch with PCL give coarse phase separation and a reduction in the mechanical properties when the starch content is increased. poly‐valerolactone. synthetic biodegradable polymers have been applied. which both can react in situ with the hydroxyl groups of starch [70. while PE‐g‐GMA posseses epoxide groups. Some of these polyesters also have very good mechanical. the compatibilizer. and poly‐caprolactone). To obtain completely biodegradable products. The tensile strength of the uncompatibilized blends is drastically reduced when starch content is increased. and of a premade starch‐g‐PCL  for starch/PCL blends resulted in a better dispersion between the phases and in turn in improved mechanical properties compared to the uncompatibilized blends. 84‐98]. Polyethylene‐g‐maleic anhydride (PE‐g‐MA) and polyethylene‐g‐glycidyl methacrylate (PE‐g‐GMA) have been used as reactive compatibilizer for starch/PE blends. Examples of these biodegradable polyesters are poly‐glycolide. poly‐dioxanone. The use of reactive compatibilizer precursors PCL‐g‐glycidyl methacrylate (PCL‐g‐GMA) [89‐90]. PCL‐g‐maleic anhydride (PCL‐g‐MA) [91. Introduction polyethylene show a coarse phase separation due to differences in polarity of starch (hydrophilic) and polyethylene (hydrophobic). which combines excellent biodegradability with acceptable mechanical properties.
1. In the past.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) using i Al(O Pr)3 as the catalyst. Another possible cause for the low GE values is the heterogenous nature of the reaction. starch/ biopolymer blending. The water in starch granules competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. The method applied basically consists of three steps (temporary partial protection of starch‐hydroxyl 16 . thus resulting eventually in low GE values . leading to formation of PCL homopolymers rather that starch‐g‐PCL. The results of this study have been used as input for the synthesis of the starch‐g‐PCL. A new strategy for the in situ ROP of ε−caprolactone on starch with the use of common ROP catalyst is therefore highly desirable.3.4. Thesis Outline The objective of this thesis is to study synthetic pathways to obtain starch derivatives with the potential to be used as bioplastic. Starch is a very hydrophilic material that always contains moisture. In Chapter 3.4. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF).3. starch‐g‐PCL has been synthesized using toxic materials such as isocyanates . but it can also provide an efficient synthetic methodology for the production of a compatibilizer to be used for melt blending of starch with a biopolymer . during the reaction.Chapter 1 1. a very flammable by‐product. Three routes have been studied in detail: starch esterification. Another approach consists of the Ring Opening Polymerization (ROP) of ε‐caprolactone monomer in the presence of starch. and the presence of liquid‐solid reaction system leads to reduced reaction rates between starch and ε−caprolactone. and starch‐g‐ biopolymer formation. the synthesis of starch‐g‐PCL is reported.2. defined as the percentage of grafted polyester to starch compared to the total amount of homopolymer and grafted polyester) . In this reaction. which however is extremely air‐ and water‐sensitive. and difficult to handle since it releases ethane. The highest GE value (up to 90%) has been achieved when using triethylaluminium as catalyst [97‐98].2. Previous studies showed that common ROP catalysts such as tin octoate or aluminium isopropoxide gave low (0‐14%) grafting efficiencies (GE. Starch‐based blends by in situ polymerization The synthesis of starch based graft copolymers by in situ polymerization represents not only an alternative route to melt blending (vide supra) for the production of novel biomaterials. the hydroxyl groups of starch are supposed to function as initiating sites. Chapter 2 describes the ROP of p‐dioxanone initiated by a protected monosaccharide (1.
Nawrath.B.N. The Biocomposites Centre. .L. E. Y. Gywnedd. J. and removal of the silyl groups). in Modified Starches: Properties and Uses (Ed.A. Fowler: Industrial markets for starch. O. a systematic experimental study on the effect of the process variables on the starch ester DS has been performed. including the mechanical and thermal properties. 2000‐2005. 2005. 1984. London. 1995. and polyhydroxyalkanoates. Boca Raton. C. P.dk/isi/stat/rawmaterial. A. E. UK. R. Young: Fractionation of starch. 1986. Bemiller. accessed on August 01. Sand Hutton.. USA. Wurzburg). . Central Science Laboratory. An Introduction to the New Science of Biodegradable Plastics. Somerville: Mini review: Plant polymers for biodegradable plastics: cellulose. Princeton University Press. 1.H.. Stevens: Green Plastics.F. Poirier. . . Wurzburg: Introduction. Agricultural and Rural Strategy Group. ROP of ε−CL on the remaining starch‐hydroxyl groups. Bangor. The model. . Website of International Starch Institute. Summary Report for the European Union. addition of co‐solvent. 2004. 2002. 105‐122. starch. Breed. Holmes: Interactive European Network for Industrial Crops and their Applications. are also reported. York. References . 2008.starch. R. University of Wales. 1. Whistler. Part of A Global Economy.S. USA. Denmark: http://www. New Jersey. Corn Refiners Annual Report 2007. The results were quantified using a statistical model. in Starch: Chemistry and Technology (Ed.C. Corn Refiners Association.B. together with the mechanical and thermal properties of the synthesized starch esters are reported in Chapter 5. O. and application of different catalysts) on starch esterification using vinyl fatty esters (vinyl laurate and vinyl stearate) is provided in Chapter 4. Introduction by trimethylsilyl groups. Inc.html. UK. Passchall). The use of these compatibilizers in starch/PCL blends.. . Washington D.M. . UK. Corn. de Bragança. A preliminary study on the effect of several process variables (reactant ratio. C. Mol. USA. The synthesis of two interesting compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) for starch/PCL blends is discussed in Chapter 6. Academic Press. Inc. 2007 17 . CRC Press.5. C. Aarhus. Based on these preliminary results.
Association of Plastics Manufacturers. Patel.R. . in Starch: Chemistry and Technology (Ed.htm.L. Germany. R. 2003. Ren: Biodegradable plastics: a solution or challenge?. Watson: Corn and sorghum starches: production.J. The compelling facts about plastics: An analysis of plastics production. 1273‐1335 . Plastics Europe. Link. accessed on August 19. 2008. USA: http://www. 1998. Inc. Schleich. Starch/ Starke 2001. Seidel).. M. 1939. Whistler. Munich. Daniel: Starch. Applications. Bemiller. 1996. R.gov/garbage/plastic. A.J.B. Belgium. Hanser Publishers. C. Germany. USA. . R.epa. C. Whistler. 27‐40. 2000. Bastioli: Global status of the production of biobased packaging materials. Husing.E. X. Hanser Publishers. Techno‐economic Feasibility of Large‐Scale Production of Bio‐based Polymers in Europe. O. Fifth Edition (Ed.J. S. Eng. 31. Inc. 557‐562. . Strong: Plastics: Materials and Processing. Inc.. 23. The Royal Veterinary and Agricultural University. J. . Academic Press.L. R. Rustgi: Biodegradable polymers. 2000. S. Passchall). Prog. USA.N. UK. . .M. C. and Regulations.A.W. Starch/ Starke 2001. Washington DC. . Fritz. Crank. J. Chem. John Wiley & Sons. Cleaner Prod. . 53. Angerer: Technical Report Series. . 1990. Selke: Packaging and the Environment. Marscheider‐Weidemann. Polym. Properties. demand and recovery for 2006 in Europe. Institute for Prospective Technological Studies and European Science and Technology Observatory. Selke. 11. J. . Sci. A.. Properties. S.D. 2008 . Department of Dairy and Food Science. Wolf.351‐355. U.. Braun: Environmental impacts of biobased/ biodegradable packaging. J. Processing. 53 . 18 . Chandra. Inc. R. London. Ind. Technomic Publishing Company. Bastioli: Global status of the production of biobased packaging materials. 2005. Frederiksberg. J. 1984. Applications. Munich. A. Culter: Plastics Packaging.M. the Food Biopack Conference Proceeding (Ed. R. Lancaster. J. New Jersey. G. Hernandez. B.Chapter 1 . Website of Environmental Protection Agency. Ehrenstein: Polymeric Materials: Structure. Hoboken. Englewood Cliffs. F.E. Prentice‐Hall. Weith: Plastics. .F. Denmark. M. G. 2000. in Kirk Othmer Encyclopedia of Chemical Technology Volume 22. USA. 105‐109 . Weber). E. Brussels.
275‐282 19 . 592‐595. A. N. Parker: Phase and glass transition behaviour. J. 53. Compos. K. 34. Forssella. P. Res. 13.M. 53. Interfaces 2006. Gandini: Surface chemical modification of thermoplastic starch: reactions with isocyanates.H. and Olsen. Starch/ Starke 2001. M. Petersen. Moates. Chambre Syndicale des Emballages en Matiere Plastique (French Plastic and Flexible Packaging Association). polylactate and polyhydroxybutyrate. Schwach. L. Seville. K.G. J. a model for thermoplastic starch. Ind. report of Food Biopack Project (Ed. status and perspectives. interaction. 2005. 57. 2003. J. E. C. 43. and Pukanszky: Thermoplastic starch/layered silicate composites: structure. Physical and mechanical properties of biobased materials – starch. X.K. 1491‐1494 . Starch/Stärke. G. and Couturier: Properties of biodegradable multilayer films based on plasticized wheat starch. Rev. Part C: Polym. 356‐361 .A. R. Polym.. Gao.V. A. Nielsen. Wang. measured by differential scanning calorimetry and solid state NMR spectroscopy. 2005. the Royal Veterinary and Agricultural University. Denmark. Vliegenthart: Interaction between dry starch and plasticisers glycerol or ethylene glycol. 331–336 . van Soest. A. 1980. Y. P. J. B. Sci.F. Kruiskamp. Carvalho.M. Ind.J. Press Release: Major Price Hikes in Plastics. Ma: The effects of citric acid on the properties of thermoplastic starch plasticized by glycerol. Prod. 409–416 . Biobased packaging materials for the food industry. O. Biodegradable thermoplastic starch. . epoxy functions and stearoyl chloride. 2008. Yang. Muller. Mikkilti. Crops Prod. Starch/ Stärke 2001. Chem. Carbohydr.Lin: Note. A Real Case of “Force Majeure” that should be passed through to Packaging Prices Immediately. . Bagdi. T. 62. Spain. 4.S. Westhoff. Introduction European Commission Directorate General of Joint Research Centre. Wang. X. 1–17 .L. Otey. Smits. Curvelo. Dev. A. Carbohydr. Macromol. Doane: Starch‐based blown films. properties. J.. 2000. Appl. 53. J. Martin.B. 2005. Department of Dairy and Food Science. P. Y. Polymer Sci. Yu.M. . 21. J. 494–504 .J. 2003. . Avérous. Eng. of concentrated barley starch‐glycerol‐water mixtures.P. J. J. Weber). F.G. 372–380 . July 15.H. R. 385–409 . 1997. Yu..M. 1996. W. P. Frederiksberg C. K. Wang: Properties of starch blends with biodegradable polymers.F. Polym.
Wurzburg). P. Silva. 1997. Wurzburg. Favis: High performance LDPE/ thermoplastic starch blends: a sustainable alternative to pure polyethylene. B. 581–586 . 1209‐1217. C. CRC Press. F. P. 44. Hulleman..C. Tserki. Sci. 20 . Chem. Polymer 2003. Sun: Reinforced thermoplastic acetylated starch with layered silicates. Vikman. Polym. C. W.W. Ind. CRC Press. Mullen. 58 (9). Eng.T. Environ. Eng.J. Andrade: Reactive processing and mechanical properties of cross‐linked maize starch. 2594–2604 . X. Qiao. Chem. (editor). I. 1. Pacsu: Starch studies: Preparation and properties of starch triesters. M. 57. 1943. 1986. J. E.D. B. S. Wurzburg: Cross‐linked starches. A. 58. van der Zee. Appl. J. 705‐721. Prinos. Grohens: Starch modifications. E. C. B. J. Polym. Polym. Merrill: Properties of fatty‐acid esters of starch. A. 74. Inc. Sagar. E. 1942. Levesque. D. Ramsay. M. Boca Raton.B. . G. Polym.A. Matzinos. Kontoyiannis. I.P. W. 65. Appl. Ind. Carbohydr. Panayiotou: Processing and characterization of starch/polycaprolactone products. F. Carvalho. T. J. . C. Y. Y. Crops Prod. Coutuier: Comparative biodegradation study of starch‐ and polylactic acid‐based materials.C. Thiebaud. Rodriguez‐Gonzalez. .D. O. Polym. 139–147 .D. Ramsay. Ribeiro.A. Favis: Rheological and thermal properties of thermoplastic with high glycerol content. 34. Bikiaris. Borredon. Copinet. E. Feil: Morphology and enzymatic degradation of thermoplastic starch– polycaprolactone blends. 389‐398 .B. J. V.F. Rodriguez‐Gonzalez. J. 1517–1526 . O.J. H.W. 2004. J. 77. 1986. Myllarinen. in Modified Starches: Properties and Uses (Ed. . 55‐78. Pilin. T.W.. 2001. 381‐384 . K. Ibezim. Divers.B. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE.H. A. Aburto. Alric. 46–51 . C. Journal of Applied Polymer Science 1995.D. S. . Bertrand. E. Feller. Pacsu: Starch studies: Possible industrial utilization of starch esters. 24. Jarowenko: Acetylated starch and miscellanous organic esters. 11–17. Mullen.W. 2002. 2004. 1647‐1656. USA. 35. chapter 4 in book Modified Starches: Properties and Uses. B. Inc. destructuration and hydrolysis during O‐formylation. 17–24 . 2006. Jiang.A. J.Chapter 1 . USA. Stab. O. Sci.A. Gattin. 1999. Starch/Stärke 2005. R. Degrad. 56. Ind. M. Boca Raton.
C. Aburto.E. US Patent 2. Alric.E. USA. P. Mormann. Degee. D. Aburto. Macromolecules 1996. Beenackers: The kinetics of the acetylation of gelatinised potato starch.K. I. Alric. 2001. Carbohydr. 1999. Smith. 118‐121. . C. 57.Fennel Evans: Fundamental of Interfacial Engineering. S. A straight way to regioselectively functionalized polysaccharide esters. J. Alric. Polym. 34. I. . A. Wiley‐VCH.Coleman. Duquesne. G. 175. D. . Munich. A.H. Hu. Inc. Technomic Publishing Company. Starch‐Starke 1999. . Carbohydr. Starch‐Starke 1999. Hanser Publishers. . R. Panayiotou: Synthesis. in Reactive Polymer Blending (Ed. P. J. P. F. Symp. A. I. Aburto. New York. Introduction. Borredon. P. R. B. USA. 1960 .C. Guegan. E. C. W. Aburto. 1997. Thiebaud. . Introduction . 101‐112. H. Thesis. 1995. . 1996 . 302‐307. 51. Alric. J. Tuschoff: Acylation of hydroxy compounds with vinyl esters.928. M. E. P. Rutot. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Janssen. 56. R. Bikiaris. J. Prinos. 255‐263. Bikiaris. Macosko. S.E. C. Alric. 1997. Marechal. W. characterization. A. E.H. A. Inoue: Compatibilizers for melt blending: Premade block copolymers. Borredon. . 29. Germany. Cellulose 2004. Dubois: Synthesis and characterization of compatibilized poly(ε‐caprolactone)/ granular starch composites. R. de Graaf.W. 1991. Rijksuniversiteit Groningen. 2001.J. C. M.F. 33‐43 21 . E. E. Senocq. M. Starch‐ Starke 2005. G. L. J. P. Bakker and G.828. J. 1440‐1451.Painter: Specific Interactions and the Miscibility of Polymer Blends. Polym. W. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. Scott. 51. D. .Stokes. Appl. Aburto. A. G. Hu). Polym. 137‐144. J. 5590‐5598 .E. . Panayiotou: Properties of octanoated starch and its blends with polyethylene. 74. Bakker. D. Broekroelofs. 132‐135. C. . Nakayama. E.. Prinos. Lancaster. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. Khandpur. 11. de Graaf: Ph. and biodegradability of fatty‐acid esters of amylose and starch. Macromol. Starch‐Starke 2004. I. M. Sci. Dicke. 145‐152.D.M. Al‐Higari: Acylation of starch with vinyl acetate in water. J. 28.V. Hamaili. A. I.Graf. T. A. Thiebaud. Mouysset‐Baziard. J. J.
Polym. Eng. Sci. . Degr. A. Res.S. U. 2001. Favis. 1997. J. Polym. Bikiaris. Mohd Ishak: Material behaviour. D. D. L. Duran. Adv. 38. Danjaji. 2003. J. Mei: Study and development of LDPE/starch partially biodegradable compounds. Nawang. L. A. 167–172 .M. Ribes‐Greus: Thermal and viscoelastic properties of some commercial starch products. Gelina. Pedroso. J. 647‐655 22 . Nikolov. R. N. N. Appl. J. 30. 2005. Rosa: Effects of the compatibilizer PE‐g‐GMA on the mechanical. Polym. Ismail. 21. Carbohydr.S. St‐Pierre. Ramsay. III. 310–317 . Polymer. Prinos. 59. W. 1503–1521 . Liu. Danjaji. Appl. I.M. 75–81 . Eur.G. Cordi. 70. 236–241 . D. 1998.G. K. J. Mechanical properties of sago starch‐filled linear low density polyethylene (LLDPE) composites. M. 2005. 1‐9 . J. Ishiaku. H. Proc. Contat‐Rodrigo. N. Nakamura. Testing. Effect of compounding and starch modification on properties of starch‐filled Low‐ Density Polyethylene. Sci. 1841‐1846 . E. Degradation studies and moisture uptake of sago‐starch‐ filled linear low‐density polyethylene composites. Ishiaku.D. Nawang. R. Pavlidou. 88. Panayiotou: LDPE/plasticized starch blends containing PE‐g‐MA copolymer as compatibilizer. Panayiotou: LDPE/Starch blends compatibilized with PE‐g‐ MA copolymers. 162–163. . N. thermal and morphological characterization of recycled LDPE/corn starch blends.‐J. 2002. and tensile properties of Low‐Density Polyethylene (LDPE) and corn starch blends.A. L.G. Rosa: Mechanical. Jane. Appl. 1242‐1251 . Ind. Z. Koutsopoulos. C. Frangis. Technol.I. Pedroso. W.S. Mani.S. J. Stab. R. Z. Polym. E. Wang. J. Testing. Polym.D. Z.A. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends. 2904–2911. morphology. 20. Ismail. 2005.A. Effect of amylopectin to amylose ratio in starch. Chem. R.S. U. 2003.. Vouroutzis. 1991. Sci. Z. Verhoogt: Processing and characterization of thermoplastic starch/polyethylene blends. 287‐291 . Polvm. Ramsay. C. Polym. I. 88. Sun: Effects of polyethylene‐grafted maleic anhydride (PE‐g‐MA) on thermal properties. B. 1998. 59. thermal and morphological properties of virgin and reprocessed LDPE/corn starch blends.A.L.Chapter 1 . Mater. L. G. Sung. H. D. 34. Y. Evangelista.H. 16.D.J. Bikiaris. Polym. B. Almeida. Mohd Ishak: Material properties. 1467‐1475. A. R. Tech. Polym. H. 1998.
Polym.S. M. 43. Environ. Manan. Rodrigues. C. Kammer: On mechanical properties of sago starch/poly(ε‐caprolactone) composites. Z. S. Adv. F. Adv. M.P. S. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone.M. Eng. A.H. Appl. 4157–4167 . 2000. 35. Otey. Ishiaku. Cho. Sci. 877–884. 2004. Low density polyethylene/ starch blend films for food packaging applications. 3976–3987 . 1139–1148. 16–19 . Stridsberg. Y.G. J. Otey.D. Jana. Odusanya. Westhoff.Y. Averous. A. F. . N.R. S. Introduction . Polymer 2000. Bhattacharya Residual stresses and density gradient in injection molded starch/synthetic polymer blends Polymer 2000. 2002. Garg. Sci. Chem.M. Polym. 267. Appl. H. Lindblad. J. Ind. J. 2003.S. D. 1999. B. Eng. 157. 32–45 . 1659‐1663 . T. Ishiaku.. Azemi: Effect of starch predrying on the mechanical properties of starch/poly(e‐caprolactone) composites. R.S. Sen. Sci. M. Preliminary diffusion evaluation. 157. M. Manan. E. Albertsson. 42‐65. P.P. Jung. S. 1507– 1516. Siddaramaiah. 23 . 1997.C. Calil: Effect of thermal aging on the biodegradation of PCL. 40. C.K. 284‐287 . S.K. Maiti: A biodegradable film. K. 2007. Makromol. cross‐linked and glycerol modified starch. Sankar. Kim. Raj. John.. Azemi. Eur. O. 2001. Res. Chem. 41. J. Karlsson: Polymers from Renewable Resources. . 23. Sci. J. A‐C. 2002. Res. Tech. J. R. U. Doane: Starch‐based blown films. C. 179‐187. Yang. Liu. Chem. 87. Odusanya. J.. Adv. D. K. Eng. 26.M.S. Polym. Moro. 1987. L. B. 2003. Polym. Kim. U. Ryner. 81.K. in Polym. Ranucci.H. Prod. J. A. Albertsson: Controlled Ring‐Opening Polymerization: Polymers with designed Macromolecular Architecture. K. 139‐161. Fringant: Properties of thermoplastic blends: starch–polycaprolactone. W. 2. . . Tang. B. Dole. U. Park: Modification of aliphatic polyesters and their reactive blends with starch. glycerol modified. Ang. PHB‐V. B.F.N. .H.H. 1298‐1305. Polym. Sci. A. J. Dev. Polym.M. 41. Polym. 2004. Pilot plant investigation for production of the biodegradable film. Part A: Polym. Polym. M. . 1984. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. M. J. and their blends with starch in soil compost. 23.M. Sci.W. Jana: Studies on the properties and characteristics of starch– LDPE blend films using cross‐linked. Appl. C. Chem.‐C. 12. L. dos Santos Rosa. 89. Guedes. 1. Sci. J. O. 9177–9190 . M. Westhoff: Starch‐Based Films. T. Ind. Kim. Polym. 3539–3546 .
M. Errico. Polymer 1999. 40. Degrad. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. E. Raimo. Dubois. D. Dubois: Aliphatic polyester‐grafted starch composites by in situ ring opening polymerization. 24 . C.E. . Degée. Sci. R. . Avella. 1432–1442. P. P. 175. Macromol. Polym. and P. Rutot. Dubois: Synthesis and characterization of compatibilized poly (e‐caprolactone)/ granular starch composites. Mani. M. 127–134. P. Composite Interfaces. 7. Krishnan. M. M.E. 2001. R. Appl. 19. Avella. D. Rapid Commun. Degee. Rimedio: Polymer communication: Preparation and characterisation of compatibilised polycaprolactone/ starch composites. M. P. Polym. R. R. 80. 3875–3881 . R. 215–225. Bhattacharya: Synthesis and characterization of starch‐ graft‐polycaprolactone as compatibilizer for starch/polycaprolactone blends. 3091‐3100. Polymer 2000. Errico. 2000. J.S. 33–43 . Rutot. Symp. M. P. Stab. 41. 2002. Tang. Rimedio. P. 1998. J. 2003. Laurienzo. Sadocco: Preparation of biodegradable polyesters/high‐amylose‐starch composites by reactive blending and their characterization. E. Martuscelli. Duquesne. Narayan. . 283–286. . 83. M. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization.Chapter 1 . Macromol.
The products were synthesized at 60‐100°C and characterized by 1H‐ and 13C‐NMR. also polydioxanone with an OiPr end‐group was formed (20‐30 %). Besides the desired polydioxanone functionalised with a monosaccharide end‐group. Keywords: biodegradable.2. The average chain length of the polymers is between 7 and 58 repeating units and may be tuned by the monomer to monosaccharide ratio (at constant Al(OiPr)3 intake).Chapter 2 Experimental Studies on the Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Abstract The ring opening polymerization (ROP) of p‐dioxanone using a protected monosaccharide (1.3.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose)/Al(OiPr)3 initiator system to yield polydioxanone with a protected monosaccharide end‐ group is described. with higher temperatures (100°C) leading to lower yields. and MALDI‐TOF mass spectrometry. ring‐opening polymerization . polyesters. A statistical model has been developed that successfully describes the experimentally observed relation between the average chain length of the functionalized polymer and reaction parameters. Systematic studies showed that the polymer yield is a function of the reaction temperature and the reaction time.
Polydioxanone (poly(p‐dioxanone) or poly(1. reaction of the metal‐alkoxide with an appropriate alcohol allows the synthesis of end‐capped poly‐lactones. 2). and is tougher than polylactides and even HDPE . only limited information about the synthesis and properties of polydioxanone is available in the open literature. From a biodegradability point of view.1. The synthesis of protected monosaccharide end‐capped biodegradable polymers is usually performed via Ring Opening Polymerization (ROP).Chapter 2 2. Besides 26 . stannous octoate [Sn(Oct)2].3 MPa and an elongation at break of 500‐600%.2. particularly for nano‐ encapsulation systems for drug delivery .g. Several polymers with bioactive alcoholic and phenolic end‐groups of interest for drug‐related applications were synthesized (for example geraniol. The ring‐opening polymerization (ROP) of p‐dioxanone using metal catalysts such as aluminum isopropoxide [Al(OiPr)3]. has interesting product properties compared to other aliphatic polyesters. polyglycolide. whereas the high temperature for polylactides results in thermal degradation and undesirable transfer reactions during synthesis and processing steps . The catalysts are metal alkoxides with Lewis acidic character [8. pregnenolone. Introduction Aliphatic polyesters. Nishida et al  reported that polydioxanone decomposes to non‐toxic gases (CO2 and H2O) by microorganisms. accessible by the polymerization of p‐dioxanone (1). or zinc lactate has been reported. stigmasterol and ergocalciferol)  and also with protected monosaccharides (galactopyranose/ glucofuranose) [6‐7]. are interesting polymers because of their good product performance and biodegradability . This melting point is considerably higher than typically found for polycaprolactone (60 oC) and lower than that of polylactides (at least 175°C). Despite its good properties. It is fully degraded in the body within a period of 180 days . 9].3. which to the best of our knowledge is the first study to functionalise polydioxanone with a monosaccharide. testosterone.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. Its melting temperature is close to 110 oC. quinine. tocopherol. Polydioxanone has a tensile strength close to 48.4‐dioxan‐2‐ one)). The relatively low melting point of polycaprolactone limits its applicability. and polylactides. Exchange of the alkoxide group by e. The alkoxide group will end up as an ester end‐group in the polymer and in this way at least one of the end‐groups may be easily controlled and varied. This chapter describes experimental studies on the catalytic ROP of p‐ dioxanone in the presence of a protected glucose molecule (1. such as polycaprolactone. polydioxanone also shows good performance. which is a unique compromise between application and processing temperature. Biodegradable aliphatic polyesters end‐capped with sugar molecules have been studied extensively for use in biomedical applications. probably because the p‐dioxanone monomer has become commercially available only recently .
After the pre‐determined reaction time. 98+% (Aldrich) were used as received. containing 0.44 g. The resulting clear solution was stirred for 2 h at 50 oC. Methods All polymerization experiments were carried out under a protective nitrogen atmosphere using standard Schlenk‐ and glovebox techniques.2. 2. 0.5 mmol) was dissolved in toluene (1 ml) at 50 oC. The polymerization was allowed to proceed for 16 h at 100 oC. poorly soluble heterogeneous systems. Toluene (Labscan) was dried and stored on molecular sieves 3 Å (Labscan) under nitrogen.1.08 mmol aluminum isopropoxide) was added to pure 1 (0. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System potential applications in biomedical products. CDCl3 was obtained from Sigma and was used as received.g. The reaction mixture was clear and colorless during the reaction.55 mmol of 2 and 0. 2. As such. 97% (Sigma) and aluminum isopropoxide.2. To this solution. This stock solution was prepared by adding 4. the synthetic pathways and the soluble. which was pre‐heated till about 60°C to obtain it in a liquid state. relatively low molecular weight compounds reported in this study may be viewed as model systems for more complex. 2. NMR) due to the poor solubility of the products in standard organic solvents. Materials and Methods 2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2) 2 (1. Materials p‐Dioxanone monomer (1.2. Subsequently.2. 8 mmol).3. 5. Typical example for the synthesis of polydioxanone end‐capped with 1. Next.2. the mixture was brought to room temperature and several drops of HCl (1 N) were added to stop the reaction. (20 mmol) of aluminum isopropoxide to 20 ml of toluene.2. 3]. hot dichloroethane (20‐25 ml) was added to completely dissolve the partly solid polymer at room temperature.8 ml of a solution of an aluminum isopropoxide stock solution was added.2. 1. grafting efficiencies are difficult to determine in this system by standard analytical techniques (e. and diethylether (Labscan) were used as received. Germany) was purified according to the procedure described by Raquez et al [2. Analytical grade dichloroethane (Labscan). Particularly interesting in this field are starch polymers grafted with polydioxanone. part of this solution (350 µl. Boehringer Ingelheim.1.07 g.2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2).3. The hot solution was 27 . However. heptane (Acros).8 g. the results of this study are also of interest for the preparation of starch/polydioxanone polymers using ROP.
4. 1H‐ and 13C‐NMR spectra were recorded on a Varian AMX 400 NMR.2. The white solid was finally separated from the liquid by decantation and dried in a vacuum oven (5 mbar. The isolated yield at this condition was 68%.2. MALDI‐TOF spectra were recorded on an Applied Biosystems Voyager DE‐PRO machine using dithranol/ NaI as the matrix (linear mode).Chapter 2 precipitated in a heptane/ ether mixture (300‐400 ml. Processing of the raw data was performed using VNMR software. If the amount of sugar is in excess with respect to the aluminum catalyst. Product Analyses NMR analyses were performed in CDCl3.3. 2D‐NMR spectra were recorded on a Varian Unity 500 NMR. 2. 1 calculated by comparing the peak area of characteristic end‐group protons with that of a proton of the repeating unit in the polymer (A H‐polymer). 2. two polymers with different end‐groups are present.) H-isopropoxide end group 28 . Calculation of Average Degree of Polymerization The Theoretical Average Degree of Polymerization. it is assumed that all available initiator is used effectively. the above equation simplifies to Xn theo = monomer conversion × [monomer ]0 [ sugar ]0 (2.1. one initiated on a galactopyranose molecule and the other on an isopropoxide group. As will be shown later. X n theo in terms of number of monomer units is calculated as follows : Xn theo = monomer conversion × [monomer ]0 [total initiator ] [monomer ]0 = monomer conversion × 3[ Al (O i Pr) 3 ] + [ sugar ]0 (2. 40 oC) until constant weight.3. 4:1 by volume) at 4‐8 oC.2.) Here. This leads to the following equation: Xn exp X n exp is = A A H-repeating unit H-galactopyranose end group +A (2.) H‐NMR was applied to determine the experimental average degree of polymerization. X n exp of the product.
) or. Results and Discussions 2.4− 4. Schematic representation of the polymerization reaction including atom numbering scheme 29 .6 ppm + A 5.1 ppm (2. The exchange reaction typically takes place at 50°C for 2 h. in term of the NMR resonances : R gp = A 5. is calculated using: R gp = A H − galactopyranose end group H − isopropoxide end group A (2.4. eq.2.6 ppm A 5.1.5.). an excess of monosaccharide 2 on Al was used (see eq. In the first step. R gp . the protected mono‐saccharide 2 reacts with Al(OiPr)3 to form the actual catalyst for the polymerization reaction. 2. is rewritten as: Xn exp = [ A 3. To compensate for this effect.) The ratio of the two different types of polymers (either end‐capped with galactopyranose or an isopropoxide group).). an Al ‐ monosaccharide 2 ‐ monomer 1 mol ratio of 1: 6.3.1. CH2OH 6 O 4 H H5 O 3 H O O H 2 O H O 1 α O β O γ Al(OiPr)3 catalyst O O α γ γ' β' β α' H2C O [ C CH2 O CH2 CH2 O ]n C CH2 O CH2 CH2 OH 6 O H O H5 1 4 O H O H 2 3 H O + n+1 (2) (1) (3) Scheme 2.3.6: 96 was applied.1 ppm (2. Screening Experiments Initial experiments to synthesize protected galactopyranose end‐capped polydioxanone (3) were performed using 1. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System The NMR peaks of the repeating unit overlap partially with those of the galactopyranose end group.6.3.6 ppm ] / 6 A 5.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose (2) and Al(OiPr)3 as the catalyst precursor (Scheme 2. The polymerization reaction was performed as a two step process.7 ppm − 6 × A 5.1.7. To avoid the formation of isopropoxide end groups.3. 2.) 2. Typically.
Typical isolated product yields are 68% at these conditions.Chapter 2 H H CH2OH iPr O Al O iPr iPr O O O H O H H O H O O H O H O O H O Al O CH2 O O H O H H O O H O H H O + O 3 H H2C H2C O H H O H + H O 3 iPr OH O (2.1 ppm. The typical proton resonances of the p‐dioxanone (3. Typical 1 H‐ and 13C‐NMR spectra of 3 are shown in Figure 2.2.) are not particularly informative.67. Although the spectra clearly indicate the presence of the monosaccharide 2. together with the CH3 groups from the protecting groups of the sugar appear in the range 1. The OiPr : 2 end‐group ratio for the standard experiment was 1: 2. This multiplet is characteristic for the CH proton of an isopropoxide end‐group. it is not possible to determine whether this is truly an end‐group due to overlapping peaks with protons from the poly(p‐ dioxanone) backbone. It is difficult to determine the end‐groups of the polymer on the basis of 1H‐NMR.15‐1. although it is evident that p‐dioxanone polymerisation occurred.3.7. Product analyses The products were analyzed using 1H‐ and 13C‐ NMR and MALDI‐TOF. However. 2.1 and 2.51 ppm.78 to 4. It confirms that polymer initiation not only occurs with the monosaccharide but also with the remaining isopropoxide group of the catalyst precursor (vide infra). respectively.. The off‐white solid reaction product was collected after a dissolution‐reprecipitation process using dichloroethane and a heptane/ diethyl ether mixture.1 ppm compared to the monomer. 1 30 . the in‐situ formed catalyst was reacted with p‐dioxanone monomer 1 at 100 oC for 16 h.1. Proton resonances of the CH3 group of the isopropoxide end‐group.1. of interest is the presence of a small multiplet at about 5.) In the next step.1. H‐NMR spectra (Figure 2.40 ppm) unit are broadened and shifted up to 0.
4‐di‐O‐isopropylidene‐α‐D‐galactopyranose. 1 (c).2. 1.2 ppm.2.52 ppm (β’). Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2. p‐dioxanone monomer.8.) is more informative and clearly shows the presence of a polydioxanone polymer backbone and a monosaccharide end group.3.28 ppm in 2 to δ 66. H‐NMR Spectra of: (a).2‐70. In addition to the mono‐saccharide end‐ group.17 and 69.1.2 and 96.3. 13 31 . Particularly the C2‐C6 carbons in the range δ = 66.2 ppm in product 3.3 ppm are shifted considerably. characteristic resonances of an ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end group are present at δ = 61.2. The carbon resonances of the polydioxanone backbone are present at δ = 63. 2 (b). 3 1 C‐NMR (Figure 2. Carbon resonances arising from the monosaccharide end‐group are present between δ = 66.2 ppm. For instance. C‐6 is shifted from δ = 62. 68. 1.61 (γ’) and 73.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.
C‐ NMR Spectra of: (a).Chapter 2 Figure 2. D‐NMR (HSQC) was applied for complete peak assignment of the product. 1 (c).76‐25.2. The resonances of the protecting group of the sugar appears at δ = 108. p‐dioxanone monomer.76‐25. 2 (b).2. 3 13 Carbon resonances of the isopropoxide groups from the by‐product (isopropoxide end‐capped polydioxanone) are present at δ = 68.70 (>C‐(CH3)2) and between δ = 21.2. and 2.3.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.1. An overview of the data is given in Tables 2.48 ppm (‐CH‐ (CH3)2) and between δ = 21.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.73 and 109. 1. 2 32 . 184.108.40.206. the numbering scheme of carbons and protons is given in Scheme 2. A typical example of a part of the 2D‐NMR spectra is given in Figure 2.3..97 (>C‐(CH3)2).97 ppm (‐CH‐(CH3)2).
which were determined from HSQC spectrum due to overlapping resonances. except values with *.77 3.31 4. HSQC 13C‐1H spectra of 3 Table 2.732 broad. around 1.60 4. 1 H‐NMR peak assignments a Product peaks (ppm) Poly(p‐dioxanone) end‐capped with protected galactopyranose (3) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 H‐α H‐β H‐β’ H‐γ H‐γ’ H‐OH 5.83 p‐dioxanone Monomer (1) H‐α H‐β H‐γ a 4.16* 4.1.8 Reactant peaks (ppm) Galactopyranose (2) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 5.32 * 4.50 4.68 4.74 3. 33 .54 4.09 4.78 4.40 all values were determined using 1H NMR.59 4.00 4.26 3.17 3.25 3.301 3.3. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.
The latter likely stems from the matrix used to ionize the sample.28 70.) This series represents a dioxanone polymer end capped with 2 and an additional Na ion. MALDI‐TOF was also applied to characterize the products. which is the molecular weight of a repeating dioxanone unit.21 70.52 63.8.17 69. corresponding with an average number molecular weight (Mn) of about 1600.2 68. The product obtained at screening conditions (16 h reaction time at 100°C) contains on average 13 monomer units.63 68.61 170.92 68. A detailed procedure is given in the experimental section.4. 13 C‐NMR peak assignments Product Peaks (ppm) Poly(p‐dioxanone) end‐capped with protected galactopyranose (3) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 C‐α C‐β C‐β’ C‐γ C‐γ’ C‐carbonyl C‐carbonyl‐iPr/gal 96. The molecular weight distribution of the major peaks may be represented by the following relation: m / z = 23 + 260 + 102n (2. 34 .31 70.09 65.21 NMR analyses also allow calculation of the molecular weight of the products. the ratio of the intensity of the end groups and the polymer backbone peaks is determined.28 p‐dioxanone monomer (1) C‐α C‐β C‐γ C‐carbonyl 61.92 66.07 Reactant Peaks (ppm) Galactopyranose (2) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 96. An example of a MALDI‐TOF spectrum of 3 recorded in a dithranol/NaI matrix is given in Figure 2.Chapter 2 Table 2.00 171.2 73. A typical molecular weight distribution is observed.54 166.75 70.57 62.2.10 71. For this purpose.80 61.58 68.33 70. The difference in molecular weight between the main peaks is 102 g/mol.11/ 170.
m / z = 23 + 18 + 102n m / z = 1 + 18 + 102n 35 (2.). two other distributions are present. although both with a very low intensity (Figure 2.) and (2.10. Typical MALDI‐TOF spectra of 3 A second molecular weight distribution is also clearly visible when enlarging the spectra.5.) This relation is indicative for the presence of a polydioxanone polymer containing an isopropoxide end‐group.11.5 +102 1400 1600 1800 2000 2200 2400 Mass (m/z) Figure 2. see Figure 2. This series may be described by the following relation: m / z = 23 + 60 + 102n (2. for details.5. ionized with either Na+ or H+.) .) (2. These distributions may be represented by equation (220.127.116.11. Furthermore. in line with the NMR data.10. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 100 90 80 70 60 %-Intensity 50 40 30 20 10 0 600 800 1000 1200 1098.) and imply the presence of dioxanone polymers with carboxylic end groups.
g.1 0 600 700 800 900 Mass (m/z) 1000 1100 1200 Figure 2. different molecular weights were analysed.7 796. Mechanistic aspects Both NMR and MALDI‐TOF measurements imply that the main product is indeed the desired monosaccharide end‐capped polydioxanone. but in all cases poor quality.13]. according to NMR. Various samples with. Enlarged MALDI‐TOF spectra for 3 Although MALDI‐TOF clearly demonstrated the presence of various types of end‐groups in the product. Most likely the matrix (dithranol/NaI). it proved not suitable to determine the average molecular weight of the products.5. the catalyst precursor is treated with monosaccharide 2 resulting in an alcohol exchange 36 .7 %-Intensity 30 20 858. Various other matrices were tested (e. although successfully applied for galactopyranose‐end‐capped polycaprolactone . 2‐(4′‐ hydroxybenzeneazo)benzoic acid (HABA)). small but significant amounts (20‐30%‐mol) of polydioxanone chains with an OiPr end group are present.6. is not particularly suitable for polydioxanone.3 10 836.2.Chapter 2 50 40 792. low resolution spectra were obtained.3. The observed differences in the molecular weight distributions of the various samples were only marginal. A mechanistic proposal for the ROP of p‐dioxanone with Al(OiPr)3 as the catalyst precursor leading to the desired monosaccharide end‐ capped polydioxanone is given in Figure 2. In the first step. [6. 2.1. In addition.
Irreversible termination of the polymerisation is performed at the end of reaction period by adding dilute acid to the polymerisation mixture. iPr iPr O Al O iPr + R OH (in excess) O O Al O R + O iPr OH O Al O O iPr + O + O O O O n O l O O O Al [O CH2 CH2 O CH2 C ] n OR O O O O Al [O CH2 CH2 O CH2 C ] O n+m OR + m R-OH O + HX (termination) O Al O R O O Al X O O + + O O H [O CH2 CH2 O CH2 C ] l by-product O iPr H [O CH2 CH2 O CH2 C ]n OR H [O CH2 CH2 O CH2 C ] n+m OR product Figure 2. The termination reaction is known to be reversible and the formed polymer may again react with an aluminum center and continue to grow . (R = monosaccharide 2) 37 . a p‐dioxanone molecule will coordinate to the Lewis acidic aluminium center followed by an insertion step. termination of the chain growth may occur by reaction with an alcohol. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System reaction and the formation of the desired active catalyst with preferentially all three OiPr exchanged with 2. The resulting Al‐alkoxide may again initiate a polymerisation reaction. The termination reactions lead to the formation of ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end groups. Simplified reaction scheme for the ROP of p‐dioxanone catalyzed by Al(OiPr)3. During the reaction. Next. Subsequent coordination and insertion of dioxanone molecules leads to the formation of a polymer chain with a monosaccharide end‐group.6.
degree of polymerisation of the product and the end group distribution was determined.29 14.7.08 30.10 15.70 58.78 10.). 2. Set Sample S111 S112 S113 S114 S115 S211 S212 S213 S214 S215 S221 S222 S223 S224 S225 Overview of experiments a Processing Condition 2/Al ratio (mol/mol) 6.6 6.81 16. a (nearly) fixed Al(OiPr)3 : monosaccharide ratio of 1 : 6.1 2.1 3.6 91.85 14.5 84.7 2. An overview of the experiments and the results are given in Table 2.4 3.5.6 96.1 86.8 3.6 t (h) 16 16 16 16 16 16 16 16 16 16 1.50 35.).4 2/OiPr ratio Rgp (mol/mol) 2.12 42.3 81.Chapter 2 The minor product. In all cases.6 6.74 18.104.22.168 2.6 86.5 T (°C) 100 100 100 100 100 60 60 60 60 60 60 60 60 60 60 dioxanone/ 2 ratio (mol/mol) 3.3 6.22 13.3 3. OiPr‐end‐capped poly(p‐dioxanone).5 hrs and 16 hrs) and a p‐dioxanone to monosaccharide 2 molar ratio ranging between 3.93 37.50 25. 2.74 18.3 6.5 67.5 81. the exchange reaction between the catalyst and the protected sugar was performed for 2 hours at 50°C The yield is the isolated yield of product 3 b 38 .1 2.74 52.6 6.9 1 7.36 17.5 1.55 Product Properties Avg.2.6 81.16 46.6 6.6 6.5 1.6 was applied.49 13. 2. A total of 15 experiments were performed at two polymerization temperatures (60 and 100 oC).8 2.4 2.21 54.3 6.04 19.6 80. despite the excess of 2.9 2. Table 2.30 62.6 6.45 19.5 1.2 91. Systematic Studies The effect of important process variables (temperature.28 33.5 1.0 87.6 6. time and the mol ratio of monomer to monosaccharide) on the yield.3 6.11 22.3 6.7. These may be present in the reaction mixture because the exchange reaction between Al(OiPr)3 and 2 was incomplete.6 89.22 16.6 6. Chain Length (Xn exp ) Yieldb (%) 30.93 8.3 and 62.96 2 3 a For each experiment.53 59. However.4 2. will be formed when the polymerization starts with an aluminum alkoxide with a remaining Al‐OiPr group (eq.3‐6.6 2.22 57. two different reaction times (1. the end group may also be formed by a termination reaction with free isopropanol.5 92.8 2.3 6. formed in the first step of the polymerisation reaction (eq.3.
with an enthalpy of polymerization of approximately ‐15 kJ/ mol . Evidently. which is expected to be more pronounced at lower temperatures. For this particular experiment.3. et al . On the basis of our experimental data and in line with literature data. will be required. we conclude that equilibrium conversion is achieved after 16 h and that the lower polymer yields at higher temperatures are due to the slight exothermicity of the reaction. Clearly. The effect of the reaction time on the product yield may be derived from the data provided in Table 2. Similar observations were made by Raquez et al  and Kricheldorf et al  for dioxanone polymerisations using Al(OiPr)3 in the absence of a external alcohol or using benzyl alcohol.3. the equilibrium conversion of dioxanone at 80oC was about 80%. It was shown that at lower monomer to catalyst ratios.3. The latter is due to the slight exothermicity of the reaction . the yield at 1. when the reaction is performed at a low dioxanone: 2 ratio (≅ 16).2. and particularly when comparing the data in set 2 and 3 (60oC.3. a low dioxanone: monosaccharide 2 ratio was applied. For instance. as suggested by Raquez et al . To the best of our knowledge. the monomer conversion reaches a maximum value before going down to the equilibrium monomer conversion.5 and 16 h reaction time).2. Product yield The isolated yields of the reactions are all but one between 67 and 96% (see Table 2. The effect of the reaction temperature (60 and 100°C) on the product yield at three different dioxanone: 2 ratios is given in Table 2. the yields are higher when performing the reaction at 16 h reaction times. the yields for reactions conducted at 60 oC are always higher than those performed at 100 oC. Another possibility is the occurrence of polymer crystallization. One of the experiments (S111) resulted in a very low yield (31%) compared to the other reaction.). 39 . no detailed explanation has been put forward to explain this anomalous behaviour. It is likely that this relatively low molecular weight compound dissolves partly during isolation/ purification procedure leading to lower isolated yields.1.). A more detailed analysis on the actual nature and composition of the polymerization products. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 2. 1. For the two experiments with a dioxanone: 2 ratio higher than 16 (≅ 23 and 60). equilibrium conversion is not yet achieved within 1. These results were explained by assuming that the reaction is an equilibrium polymerisation and that the equilibrium is shifted to the monomer side at higher temperatures. Higher polymerisation yields were obtained at lower temperatures. The effect of temperature on the yield of the bulk polymerization of dioxanone using Sn(Oct)2 and Al(OiPr)3 as catalysts (without the use of a second alcohol) has been studied by Nishida et al  and Esteves.5 h is higher than the yield at 16 h. However.3. see Table 2. leading to a low average chain length of the dioxanone polymer (7. which reduced to 75% when increasing the temperature to 120oC.5 h.
time: 1.e. Effects of process conditions on the average chain length ( X n exp ) and end group distribution The effects of the p‐dioxanone: 2 ratio.6.Chapter 2 2. and 2. time: 16 hr 50 Temperature: 60oC.) for all experiments are scattered randomly between 2. A possible strategy to increase the Rgp values i. the X n theo (eq. in line with the theoretical predictions. Average Chain Length ( X n exp ) of the product as a function of the dioxanone : 2 mol ratio The end group distribution was determined using NMR and is expressed in terms of R gp (eq. reaction time and temperature on the X n exp of the products is shown in Figure 2. Most experimental points are scattered along these lines. 2. In Figure 2. time: 16 hr Temperature: 60oC. The R gp values (Table 2.7.) as a function of the dioxanone : 2 ratio at 90% and 100% monomer conversion is also provided. A clear trend between R gp and the process conditions (temperature.3.7. 2.2. to reduce the number of OiPr endgroups in the product may be the removal of isopropyl alcohol formed in step 40 .3.5.. and dioxanone/ 2 mol ratio) is absent. 60 Temperature: 100oC. The X n exp increases linearly with respect to the dioxanone : 2 ratio.3.2. time.5 hr Xn theo theo (90%-conversion) (100%-conversion) 40 Xn exp Xn 30 20 10 0 0 10 20 30 40 50 dioxanone : 2 ratio (mol/mol) 60 70 Figure 2.).7.4 and 3. as expected for a typical ring opening polymerisation .2.
Table 2. A parity plot of the modeled versus experimental values of the average chain length Xn confirms this statement (Figure 2. time and dioxanone/ 2 ratio) on the X n exp has been performed by multivariable linear regression  on the data given in Table 2.8. vacuum distillation before adding the dioxanone monomer .854 × (dioxanone : 2 ratio) + 0.6) was too limited to draw sound conclusions.). Analysis of variance (ANOVA) for linear model of X n function of experimental parameters SS 4103 93 4010 DF 3 12 15 MS 1337 7.) where t and T are respectively the polymerization time and temperature. The low P‐value clearly indicates that the model is statistically significant.459*10‐7 41 .3‐6.543 exp as a Model Error Total P‐value 3.12. 2.977 (with an adjusted R2 value of 0. A linear model proved adequate to describe the effects of the independent variables on the X n exp : Xn exp = 8.4.3. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 1 of the polymerisation process (eq. The analysis of variance for the model is given in Table 2.703 F 22.214.171.124 × (t ) − 0.974). 2.3.4.) from the reaction mixture by e.49 + 0. The 2/ Al ratio was not included in the model as the experimental range (6.097 × (T ) (2. The R2 value for the model is 0.3.g. indicating that the model describes the experimental data reasonably well. Statistical Data Analysis Quantification of the influence of the experimental factors (temperature.
8. which is in agreement with the available data on ring opening polymerization [15. 2) with Al(OiPr)3 as the catalyst is reported.Chapter 2 60 50 40 Xn model 30 20 10 0 0 10 20 30 Xn exp 40 50 60 Figure 2. the X n exp is negatively influenced by temperature. with higher ratios leading to higher molecular weights. with high ratios leading to a higher average chain length. 2. the model predicts that the X n exp is positively influenced by the polymerization time. the model allows determination of the process variables to obtain a product with the desired degree of polymerization.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. Both NMR and MALDI‐TOF measurements indicate that the products are mixtures of polymers and significant 42 . Molecular weights (NMR) of the product were between 970 and 6200 and are a clear function of the p‐dioxanone/2 ratio (at constant Al(OiPr)3 intake). Modeled versus experimental values for the average chain length Xn The model predicts that the X n is a clear function of the p‐dioxanone : 2 mol ratio. In addition.3. 17]. Furthermore. 11] and due to the fact that the reaction is an equilibrium polymerization with a slight exothermic effect. This is in line with literature data [5. The isolated yields of the off‐white solid products were between 30 and 96%.2.4. 16. Within the experimental ranges. Conclusions The ROP of p‐dioxanone in the presence of a monosaccharide (1.
21.4‐dioxan‐2‐one and controlled synthesis of diblock copolymers with epsilon‐caprolactone. starch). P. Chem. Raquez. 2000. D. temperature and monomer: monosaccharide ratio) on the X n exp .4‐dioxan‐2‐ one initiated by Al((OPr)‐Pr‐i)(3) in bulk.J. 199. D. 153‐166.5. H. 1993. 42. J. Macromol.g. Rapid Commun. J.C.4‐dioxan‐2‐one.M. Dubois: Some thermodynamic. Kricheldorf. Degee. 1063‐1071. Macromolecules 2001. Nomenclature AH A x − y ppm : peak area of certain proton in 1H‐NMR spectra [‐] : peak area of certain peak at x until y ppm in 1H‐NMR spectra [‐] : mol ratio of galactopyranose end‐capped polydioxanone and isopropoxide‐end‐capped polydioxanone [‐] : number of monomer unit in the polymer products [‐] : time [hour] : temperature [oC] : experimental average degree of polymerization the polymer products [monomer units] : theoretical average degree of polymerization of the polymer products [monomer units] R gp n t T X n exp X n theo 2. Macromol. C. 8419‐8425. 2. R. Grubb: Hydrolytic degradation and morphologic study of poly‐p‐dioxanone. . H. Damrau: Polylactones.R.O. Chu. 34. Moreover. 1089‐ 1097. Res. the findings of this study have proven to be valuable input for synthetic studies on the preparation of novel biodegradable polymers consisting of polydioxanones and polycaprolactones grafted on oligo‐ and polysaccharides (e. . kinetic. 1998. . Biomed. Lin.M.L. P. Phys.6. Dubois: ʺCoordination‐insertionʺ ring‐ opening polymerization of 1. 27. Narayan. Raquez. Narayan. A statistical model has been developed to quantify the effects of process variables (time. Zn‐lactate‐catalyzed polymerizations of 1. 43 . These studies will be reported in the next chapter. and mechanistic aspects of the ring‐opening polymerization of 1. P. Mater. P. References . Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System amounts of p‐dioxanone polymers with an isopropoxide end group (20‐30%) were present. Degee. R.
. H. Nishida, M. Yamashita, T. Endo, Y. Tokiwa: Equilibrium polymerization behavior of 1,4‐dioxan‐2‐one in bulk. Macromolecules 2000, 33, 6982‐6986. . T. Hamaide, M. Pantiru, H. Fessi, P. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer agents. A novel route to functionalised nanoparticles. Macromol. Rapid Commun. 2001, 22, 659‐663. . K. Bernard, P. Degee, P. Dubois: Regioselective end‐functionalization of polylactide oligomers with D‐glucose and D‐galactose. Polym. Int. 2003, 52, 406‐411. . H.R. Kricheldorf: Syntheses and application of polylactides. Chemosphere 2001, 43, 49‐54. . H.R. Kricheldorf, M. Berl, N. Scharnagl: Poly(lactones). 9. Polymerization mechanism of metal alkoxide initiated polymerizations of lactide and various lactones. Macromolecules 1988, 21, 286‐293. . H.R. Kricheldorf, I. Kreisersaunders: Polylactones. 30. Vitamins, hormones and drugs as co‐Initiators of AlEt3‐initiated polymerizations of lactide. Polymer 1994, 35, 4175‐4180. . L.M. Esteves, L. Marquez, A.J. Muller: Optimization of the coordination‐ insertion ring‐opening polymerization of poly(p‐dioxanone) by programmed decrease in reaction temperatures. J. Appl. Polym. Sci. 2005, 97, 659‐665. . D.C. Montgomery: Design and Analysis of Experiments, 5th Edition, John Wiley & Sons Inc., New York, USA, 2001. . P. Dubois, R. Jerome, P. Teyssie: Aluminum alkoxides ‐ A family of versatile initiators for the ring‐opening polymerization of lactones and lactides. Makromol. Chem. Macromol. Symp. 1991, 42/43, 103‐116. . A. Duda, A. Kowalski, J. Libiszowski, S. Penczek: Thermodynamic and kinetic polymerizability of cyclic esters. Macromol. Symp. 2005, 224, 71‐84. . A. Kowalski, A. Duda, S. Penczek: Polymerization of L,L‐lactide initiated by aluminum isopropoxide trimer or tetramer. Macromolecules 1998, 31, 2114‐ 2122. . A. Duda, A. Kowalski, S. Penczek, H. Uyama, S. Kobayashi: Kinetics of the ring‐opening polymerization of 6‐, 7‐, 9‐, 12‐, 13‐, 16‐, and 17‐membered lactones. Comparison of chemical and enzymatic polymerizations. Macromolecules 2002, 35, 4266‐4270. . A. Duda: Polymerization of epsilon‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. Kinetics and mechanism. Macromolecules 1996, 29, 1399‐1406.
Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System
. S. Penczek, T. Biela, A. Duda: Living polymerization with reversible chain transfer and reversible deactivation: The case of cyclic esters. Macromol. Rapid Commun. 2000, 21, 941‐950.
Keywords: starch. polycaprolactone. the remainder being homopolymers of ε‐caprolactone. The grafting efficiency varies between 28 and 58%. Removal of the silyl groups of the poly‐(ε)‐ caprolactone grafted starch co‐polymers is possible using a mild acid treatment with diluted hydrochloric acid in THF at room temperature. silylation . The silylation reaction was performed using hexamethyl disilazane (HMDS) as the reagent in DMSO at 70°C.Chapter 3 Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors Abstract Poly‐(ε)‐caprolactone grafted corn starch co‐polymers were synthesized using a hydrophobised silylated starch precursor. Silylated starch with a degree of substitution (DS) between 0. The poly‐(ε)‐caprolactone side chains consist of 40‐55 monomer units and is a function of the reagent intakes. The DS of the polycaprolactone graft is between 0. ε‐ Caprolactone is grafted to silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 in THF at 50oC. grafting. Experiments with native starch under similar conditions do not result in the desired poly‐(ε)‐caprolactone grafted corn starch co‐polymers and unreacted starch was recovered after work‐up. biodegradable polymers.7 was obtained.72.21 and 0.45‐0.
Chapter 3 3. a natural biopolymer. as starch is highly hydrophilic.g. polyvalerolactone. These plastics are mainly synthetic polymers from fossil resources. Starch has also been thermoplasticized with the help of plasticizers such as glycerol and other polyalcohols . There are two types of biopolymer in starch. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decreased rapidly in time and were fully degraded after 4 weeks . Starch modification is therefore needed to meet the product properties in a number of application areas. Blends of thermoplastic starch and PCL are not fully miscible.4‐glucosidic bonds) and amylopectin (a branched polymer with α‐D‐1.6‐glucosidic bonds besides α‐D‐1. Global production of starch is 60 million ton per year in 2004 . in almost all cases. Starch films are known to have good oxygen barrier properties. the future generation of packaging materials is highly desirable. The amylose‐amylopectin ratio in native as well as modified starches has a strong impact on the product properties. it is water sensitive. However. is one of the potential candidates for future biodegradable polymer products.1. A wide variety of synthetic biodegradable polymers have been prepared. Polycaprolactone (PCL) is easily degraded by micro‐organisms . However. PCL has gained much interest for possible applications in the medical field as well as in the area of packaging materials [12‐ 13] Several studies to combine the properties of starch and PCL have been performed to obtain fully biodegradable materials with improved product properties. 7]. 5]. Introduction Worldwide. which are known to degrade with difficulty and cause serious environmental problems . 245 million tons of plastics are produced per year. and the mechanical properties of starch‐based films are generally inferior to those derived from synthetic polymers [4. amylose (a linear polymer of anhydroglucoses with α‐D‐1. and this value increases with about 10% per year . the used monomers and the corresponding grafted chains are not easily biodegradable. Starch is abundantly available. the product properties are in most cases still not up to standards and blending with other polymers is required . The development of green biodegradable polymers for e. However. for instance grafting of monomers (like styrene and methyl methacrylate) to the starch backbone [6. Starch is present in the body of many plants (tubers.4‐ glucosidic bonds). The polymeric structure of starch consists of repeating anhydroglucose units. Well known examples are polyesters derived from cyclic lactones (polycaprolactone. resulting 48 . roots) as granules or cells with typical particle sizes between 1‐100 µm. Various modification strategies have been explored. Starch. and polybutyrolactone). The content of amylose in starches depends on the plant and typically varies between 18‐28%.
thus to a reduction in the GE. This is expected to lead to reduced reaction rates between starch and CL compared to CL homopolymerisation. This catalyst is extremely air and water sensitive.1. leading to a liquid‐solid system. A stock solution was prepared by dissolving 1. a very flammable by‐product. This approach has also been applied successfully to graft PCL and polylactide on dextran [19‐20]. Aluminium triisopropoxide [Al(OiPr)3] (Acros) was used without further purification. 49 . 3. To increase the miscibility of starch and polycaprolactone. This is rationalised by assuming that water competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. 0‐14%). Materials Corn starch (Sigma) was dried at high vacuum (~1 mbar) at 100 oC for one day before use. Materials and Methods 3. The highest GE (90%) was achieved using triethylaluminium as catalyst .2.67 gram (8 mmol) Al(OiPr)3 in 50 ml of dry toluene in a glove box. therefore difficult to handle and releases ethane. DMSO (Acros) and toluene (Labscan) were dried overnight over molecular sieves 3 Å (Merck) and stored under a protective nitrogen atmosphere. Hexamethyldisilazane (HMDS. All available data indicate that the presence of water reduces the GE. thus leading to the formation of PCL homopolymers rather that starch‐g‐PCL . Acros) and methanol (Labscan) were used as received. an alternative method to synthesize poly‐(ε)‐caprolactone grafted starch co‐polymers (starch‐g‐PCL) is reported. Hydrochloric acid (HCl) 1 N was prepared from Titrisol concentrated hydrochloric acid solution (Merck) and distilled (Milipore) water. distilled under reduced pressure at 100 oC and stored under a protective nitrogen atmosphere.2. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). it has been proposed to chemically graft caprolactone onto the hydroxyl groups of starch using ring‐opening polymerisation . during the reaction. the ring opening polymerisation occurs solely in the liquid phase and this is expected to lead to higher GE values. Common Ring Opening Polymerization (ROP) catalysts such as tin octoate or aluminium isopropoxide gave low grafting efficiencies (GE. In this chapter. In this way. ε‐Caprolactone monomer (Fluka) was dried over Calcium Hydride (CaH2) for 48 h. The starch source is made less hydrophilic and thus more soluble in organic solvents by substituting part of the OH groups of starch by a bulky silyl group [16‐18]. Another possible cause for the low grafting efficiencies is the heterogenous nature of the reaction. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors in undesirable phase separation . Dry tetrahydrofuran (THF) and toluene for polymerization experiments were obtained in closed vessels from Aldrich and were used as received.
Typical example of in situ polymerization of ε‐CL with silylated starch The silylated product from the first step was dissolved in THF (0. 0.1 mbar.111 mol) was added to the gelatinized mixture to initiate the silylation reaction.60): H‐NMR (CDCl3. Sample SN‐3. To this homogenous solution. the silylated starch (1) product was dried in a vacuum oven (~5 mbar. 3. 40 oC) until constant weight. broad peaks. Sample SN‐3. Typical example of the starch silylation procedure The procedure for corn starch silylation was adapted from that published for dextran [19‐23]. 3‐6 ppm (m. This procedure was repeated three times. Methods All reactions and manipulations with air‐sensitive materials were carried out under a protective nitrogen atmosphere either using standard Schlenk techniques or in a glove box.2. starch). THF (4.12 (m. The samples were characterized by 1H‐NMR. a predetermined amount of ε‐caprolactone monomer (molar ratio of monomer to 50 .5 ml) and a predetermined amount of the stock solution of Al(OiPr)3 in toluene were added.6 mL) at 50 oC (1‐2 h). Chapter 3 3. after peracetylation.60): 1 H‐NMR (CDCl3. silyl‐CH3). The white‐to‐transparent solid products were stored under vacuum in a desiccator at 6‐8 oC. After 24 h.2. broad peaks. After removal of the solvents under reduced pressure (0. before peracetylation. 80 oC).2. After 2 and 4 h reaction time. A mol ratio of ‐OH groups to catalyst of 10:1 was used. The mixture was stirred at 50 oC for 4 h to promote the exchange reaction between the isopropoxide groups of Al(OiPr)3 and the free ‐OH groups of starch.5 (m.2. 1. DS = 0. pre‐dried corn starch (6 g) and dry DMSO (75 ml) were stirred at 70oC for about 3 h until a clear solution was formed. Silylated Starch (1. Subsequently. The reaction was carried out at 70 oC. DS = 0. another portion of toluene (20 ml) was added. toluene (40 ml) was added to solubilize the precipitated (partially silylated) starch.2. 3‐6 ppm (m. the solvents were removed from the silylated starch product under reduced pressure (~ 20 mbar) at 70 oC. silyl‐CH3).2. Silylated Starch (1.1. For each experiment.2. 50 oC): δ 0. starch).7‐2. After 6 h. The intake of 1 depended on its DS and was adjusted to obtain a solution with 5 x 10‐5 mol free‐OH groups/ml of THF. Traces of DMSO trapped in the product were removed by dissolving the product in a small amount of toluene and re‐precipitation in methanol.12 (m. 50 oC): δ 0. acetate‐CH3). The pre‐determined amount of HMDS (typically 24 ml. 1 3.
4 (m. β and δ‐PCL). and vacuum dried at 40 oC for 24 h. The total isolated yield at this condition (see Table 3. broad peaks. The peracetylation procedure applied in this study was adapted from the literature [24. iPr). filtrated. 60 oC): δ 1. 2.2. Typically. α‐PCL). iPr). 1.2. 1 3.5. starch). Starch‐g‐PCL (3. 4. and 5‐5. 60 oC): δ 0. 1.34): H‐NMR (DMSO. 2. iPr). ε’‐PCL). Peracetylation of silylated starch Characterisation of the silylated starch by NMR proved very difficult due to the presence of very broad and overlapping resonances arising from starch.54 (m.98 (t.25 (t. DSsilylation = 0.68.2. Deprotection product of Sample SN1CL2. 4.) was >99 %.98 (t. 3.54 (m. 1. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers Desilylation of the 2 was performed using a procedure described by Ydens et al  for desilylation of silylated dextran‐g‐PCL. The silyl group was removed by adding a slight excess (with respect to the number of the silyl functionalities) of 1N HCl to a solution of ‐starch‐silylated‐g‐PCL in THF (10 % w/v).5. γ‐ PCL). The solid precipitate was filtered and dried under vacuum (~ 5 mbar) at 40 oC for 48 h.75. 25].31 (m. The silylated starch‐g‐PCL (2) product was precipitated from the solution by the addition of heptane (about 250 ml) at ‐18 o C. 3.16 (d.37 (t. silyl‐CH3).3‐4.12 (s. 4. 1.2.88 ppm (m.68.31 (m. the peracetylating reagents 51 . The reaction was stopped by cooling down the mixture to room temperature and the addition of 2‐3 drops of 1 N HCl to deactivate the catalyst. ε’‐PCL). iPr). After stirring for 2 h. starch). DSsilylation = 0. α‐PCL). ‐CH.4. The samples were characterized by 1H‐NMR.37 (t. DSPCL=0.88 ppm (m.3. β and δ‐ PCL). 3. 3.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until it was fully dissolved (typically 3 h). 4. broad peaks. ‐CH3 .25 (t.2. Sample SN1CL1. Subsequently. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors OH‐groups was 13 : 1 for a standard experiment) was added and the ring opening polymerization reaction was allowed to proceed for 24 h at 50 oC. 1 (0. and 5‐5.16 (d. The product was collected as a white solid and characterised by 1H‐NMR. The yield was measured gravimetrically and is based on the weight of the product and the total weight of reactants charged to the reactor. 1 3. Peracetylation of the remaining OH groups of modified starch is a well established procedure to improve characterisation of the products by NMR . Silylated Starch‐g‐PCL (2. ‐CH3. ε‐PCL). the desilylated starch‐g‐polycaprolactone product (3) was precipitated using heptane. ε‐PCL).3‐4. γ‐PCL).21): H‐NMR (DMSO d‐6.75. 3.5‐ 3. DSPCL=0. 1.5‐3. 3. ‐CH.4 (m.
3. 1 3. The peracetylation reaction was conducted for 7 h at 50 oC.4.3. 3. Calculations The DS of the silylated starch (DSsilylation) is defined as the average number of silyl groups present on an anhydroglucose unit (AHG) of starch. DSsilylation may be calculated using 1H‐NMR spectra of the products after peracetylation using eq.6 ppm DSsilylation = × = × 3 Astarch proton / 7 27 A3−5.1.2. In this calculation.) The degree of substitution of the PCL graft on 2 (DSPCL) is defined as the average number of PCL polymer chains present on an AHG unit of starch. When assuming that all ε‐CL monomer is converted. 3. The assumption of high conversions (>95%) was correct for all experiments (see Table 3. the DSPCL may be calculated using eq.) 52 .3.4 ppm + 1 (3. The ACL is calculated from 1H‐NMR spectra by comparing the peak area of protons attached to ε‐carbon atoms in a repeating CL unit with that of the characteristic ε’ protons of the last CL unit in a PCL chain  (see Figure 3.2 ppm A3.). Nuclear Magnetic Resonance (NMR) H‐ NMR spectra were recorded in CDCl3 at 50oC or in DMSO d‐6 at 60 oC on a Varian AMX 400 NMR machine.3.1.2. Analytical Methods 3.3−3.2. It was finally dried overnight in a vacuum oven at 70 oC and 5 mbar until constant weight.8− 4.1.2. it is assumed that the average length of the grafted chain is equal to the chain length of the homopolymer.2. ACH 3 silyl / 9 1 7 A−0. Chapter 3 (DMAP. The product was precipitated by the addition of methanol and washed several times with methanol.) where Ax‐y stands for the 1H‐NMR peak area in the range δ x‐y ppm. This leads to the following equation: ACL = ACH 2 −ε + ACH 2 −ε ' ACH 2 −ε ' = A3.8 ppm (3. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added. The Average Chain Length (ACL) of the Poly‐(ε)‐caprolactone chain is defined as the average number of CL repeating units in a grafted polymer chain.3.6−0.
and Vm stands for molar volume. 2007) 3. This leads to the following equation: GE = ACH 2 −ε ' / 2 − ACH −iso ACH 2 −ε ' / 2 × 100 % (3.( A3.CL/OH ratio ⎥ ⎣ ⎦ (3. the focus of this chapter will be on the first two steps of the procedure.4.2 ) ⎤ 1 × ⎢ (3 − DSsilylation )⎥ ⎢ ε . Results and Discussions The overall procedure to synthesize poly‐(ε)‐caprolactone grafted starch co‐ polymers (3) consists of three steps and includes hydrophobization of starch by silylation of part of the hydroxyl groups of starch using hexamethyl disilazane (HMDS).5. 53 .8− 4.4 − A4 . Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors DS PCL = 0. It is calculated by comparing the area of protons related to the PCL grafted to starch with the area of the protons of all PCL chains present in the product.3−3.8− 4 . The values of ∆Hv and Vm were obtained from the SciFinder Scholar database (American Chemical Society. The Hildebrand solubility parameter of HMDS and DMSO were calculated using the following equation : ⎛ ∆H v − RT ⎞ ⎟ δ =⎜ ⎟ ⎜ Vm ⎠ ⎝ 1/ 2 where ∆Hv stands for heat of vaporization. followed by an in‐situ Ring Opening Polymerization (ROP) of ε‐ caprolactone monomer on the hydrophobized starch and subsequent silyl group removal by a mild acid treatment.5. A3. T stands for absolute temperature.3.) ⎛ ⎞ ⎞ ⎛ A A = ⎜1 − CH −iso ⎟ × 100 % = ⎜1 − 4. Although all steps have been investigated.3−3.3−3.3.4 + A3.8− 4 .) The grafting efficiency (GE) is defined as the percentage of PCL grafted to starch compared to the total amount of polymerized CL (grafted and PCL homopolymer).4 / 2 ⎟ ⎠ ⎝ ⎝ where Ax‐y stands for the peak area in the range δ x‐y ppm.9 ⎟ × 100 % ⎜ ⎜ ⎟ ACH 2 −ε ' / 2 ⎠ A3.9 ⎡ 0.
). The silylated products were characterised by NMR. The silylation procedure was adapted from that previously reported for dextran [19‐23]. mixtures of toluene and DMSO were applied to avoid precipitation of the silylated starch during the reaction. It is defined as the average number of silyl substituted OH groups on the anhydroglucose (AHG) unit of starch.a.1.) Instead of using DMSO as solvent. The degree of substitution of the silyl groups (DSsilylation) was determined by 1H‐ NMR.60 was obtained. 3. 6 H2C OH 4H 3 H5 OH H 6 H2C O R 4H H5 OR 3 H O H 1 H H3C CH3 CH3 CH3 + H3C Si NH Si DMSO 50 deg. An additional peracetylation procedure to substitute the free hydroxyl groups with acetate groups was performed to improve the quality of the NMR spectra. a product with a DS of 0. Chapter 3 3. However. the starch peaks are very broad and this feature hampers accurate determination of the DSsilylation. C O H 2OR H 1 + NH3 2 OH H3C [R = H or Si(CH3)3] Starch Hexamethyldisilazane Silylated Starch (1) (3.5.1. a homogeneous reaction mixture was maintained throughout the reaction. The 1H‐ NMR spectrum of silylated starch (Figure 3. 54 .).5.1. The 1H‐ NMR peaks from the AHG unit after peracetylation were indeed considerably sharper and allowed more accurate DS calculations (Figure 3. 24 h). Very broad peaks of starch protons at δ 3‐5 ppm and a sharp peak of the methyl substituents of the silyl group at about δ 0 ppm were observed (see Figure 3. as suggested by Einfeldt et al .1. In this way.b. Synthesis of Silylated Starch The silylation of corn starch was performed with HMDS as the silylating agent (eq. 70 oC.).) shows the presence of silyl groups at about δ 0 ppm.a.3. Using standard conditions (HMDS: AHG molar ratio of 3.
60. and Figure 3. Typical 1H‐NMR Spectra of Silylated Starch (Sample SN3. not peracetylated (b).2.1. peracetylated The effect of the HMDS to starch molar ratio (1.5) on the product DS was investigated by performing experiments with a constant starch and a variable HMDS intake. The results are given in Table 3.5‐4. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Figure 3. 50 oC): (a).1. in CDCl3. 55 . DSsilylation=0.
Surprisingly. For reactions with an order higher than zero.67 0.5 2 2. the volumetric HMDS intake is about half of the DMSO intake.68 0.25 3 4. DS of the silylated products at different HMDS: AHG ratios (DMSO.5 0. HMDS is a rather apolar compound (Hildebrand solubility parameter of 6.55 0. a fixed starch intake of 6 g (37 mmol AHG) starch was applied. Its presence will reduce the polarity of the reaction medium (DMSO. mmol) 12 (56) 18 (83) 24 (111) 36 (167) HMDS: AHG mol ratio 1. Chapter 3 0.65 0.6 DSsilylation 0. the DSsilylation decreases for higher intakes of HMDS. a positive effect of higher reagent intake on the reaction rate and thus the DS is expected.60 0. Effect of HMDS: AHG mol ratio on DS of the silylation product a HMDS intake (ml.5:1.2.46 Experiment SN1 SN2 SN3 SN4 a Experiments were performed in DMSO at 70°C. 70 oC) Table 3.5 Product DS (DSsilylation) 0. solubility parameter = 11.36 cal1/2cm‐3/2) considerably.1. The experimentally observed lowering at higher HMDS intakes is likely due to a decrease in the polarity of the reaction mixture.5 2. At a mol ratio of HMDS to AHG of 4. For all experiments.5 HMDS : AHG Ratio [mol/ mol] 4 4.5 Figure 3.5 3 3.4 1.25 cal1/2cm‐3/2) due to the presence of the bulky apolar SiMe3 groups. The silylation reaction 56 .45 0.
7) of the products is in the range of those published for other starch silylation systems. The products are soluble in DMSO as well as in less‐polar solvent such as chloroform and THF.0 .3.6‐glucosidic for dextran).68. DMA/LiCl.45‐0.1 and 3.2. the silylation of starch in pure DMSO using HMDS as the reagent has not been reported to date.0 [18. 3.6. 3. Similar reductions in the reaction rates when working at higher reactant intakes were observed for the esterification of starch with vinyl laurate and stearate . 50 oC H CH2 H OR H C (CH2)5 O n H O H OR + H (R = H or Si(CH3)3 Silylated Starch (1) ε-Caprolactone Monomer [R = H or Si(CH3)3 ] Silylated Starch-g-Polycaprolactone (2) (3. and DMSO/pyridine mixtures yielded silylated starch with DS values ranging between 0. 29]. liquid ammonia. Silylation of starch with HMDS in formamide. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors involves charged‐ionic species  and a reduction in the polarity of the reaction medium is expected to lead to a lowering of the silylation reaction rates.3‐2.6. The much higher DS values obtained for dextran compared to starch may be related to differences in molecular weights between starch and dextran and the type of AHG linkages (mainly α‐1. Petzold et al  reported that silylation of starch with trimethylsilylchloride (TMSCl) in pyridine yielded trimethyl‐silyl substituted starch with DS values between 0.25‐5.2. After precipitation with heptane and vacuum drying.0) gave silylated dextran with DS values between 1. DMF.7‐3. In situ Ring Opening Polymerization of ε‐Caprolactone with Silylated Starch A number of in situ ROP experiments with ε‐caprolactone (CL) were carried out using a typical silylated starch sample (DS=0. A schematic representation of the reaction is provided in eq. O O H2C O R H H OR H O O H OR H O Catalyst: Al(OiPr)3 Toluene. pyridine. white solid products with isolated yield > 96% were obtained. To the best of our knowledge. sample SN‐1) in THF at 50 oC for 24 h using Al(OiPr)3 as catalyst. The use of HMDS to silylate dextran (Mw=6000‐40000) in DMSO (HMDS to OH molar ratio of 0.) 57 . The DSsilylation (0.
(Sample SN1CL1.3. 26]. the exchange reaction between Al(OiPr)3 and the OH groups of silylated starch is not quantitative under the conditions applied in this study.8 and 5.4.0‐5. broad peaks in the region δ 3. The peaks from the polycaprolactone units are clearly present in the range of δ 1.4.2 ppm.68.4 ppm. Coding of the peaks is given in Figure 3. The formation of CL homopolymers for this type of reactions has been observed before [15. However.21) in DMSO‐d6 at 60 oC. A typical spectrum is shown in Figure 3. This is clearly indicated by resonances of the iPr end‐group of the PCL homopolymer at δ 4.2‐4 ppm and imply that the ring‐opening polymerisation reaction of CL indeed occurred. as is shown in Figure 3.3. Chapter 3 The products were characterized using 1H‐NMR analysis in DMSO‐d‐6 as the solvent. Typical 1H‐NMR spectrum of a silylated starch‐g‐PCL sample. DSsilylation =0. Resonances from the starch peaks are observable as small. DSPCL=0. 58 .4‐3. Figure 3.9 and 1. Further process optimization studies allowed the synthesis of products with less than 42% of homopolymers (vide infra). Apparently. not all of the caprolactone is grafted to starch and large amounts of PCL homopolymers (72%) were formed. The homopolymers are formed by direct polymerization of ε‐CL initiated on isoproproxide moieties attached to the Al catalyst.
7.7. residual water may also initiate the polymerisation reaction. Exchange Reaction: iPr O Al O iPr iPr O + H2O iPr O iPr O Al OH + iPr OH 2. 3. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors 1.) 59 . Polymerization Reaction: iPr O Al O iPr H2C O O H O H OR H H OR H O O O O α β χ δ ε α' β' χ' δ' ε' H2C [O C CH2 CH2 CH2 CH2 CH2 ]n O C CH2 CH2 CH2 CH2 CH2 OH O O H O H OR H H OR H O [R = H or Si(CH3)3 ] + n+1 Silylated Starch-g-PCL (2) O iPr O Al O iPr iPr O + O n+1 iPr O α β χ δ ε O α' β' χ' δ' ε' [O C CH2 CH2 CH2 CH2 CH2 ]n O C CH2 CH2 CH2 CH2 CH2 OH PCL homopolymer Figure 3. Product formation for the reaction between silylated starch and ε‐ caprolactone Besides the ‐OH group of starch.). peaks arising from a ‐COOH unit could not be detected in 13C‐ NMR spectra (δ 175‐180 ppm). Exchange Reaction: iPr O Al O iPr H2C OH iPr O Al O iPr iPr O H2C O O O H O H OR H H OR [R = H or Si(CH3)3 ] H O + O H H OH H H OH H O + iPr OH Silylated Starch (1) 2. However. Polymerization Reaction: O iPr O Al OH iPr O PCL homopolymer with carboxylic end-group + O O α H β χ δ ε O α' β' χ' δ' ε' [O C CH2 CH2 CH2 CH2 CH2 ]n O C CH2 CH2 CH2 CH2 CH2 OH (3. This leads to the formation of carboxylic end groups (see eq.4. 1.
0 15. The results are given in Table 3. in all samples.34 0. The yield of the products was measured gravimetrically and varies between 96. Determined gravimetrically and defined as the total weight of the isolated product divided by the total intake of reactants (silylated starch and CL).72 Grafting Efficiency [%] 28 43 48 55 58 All reactions were performed using the same intake of SN1 silylated starch (DS=0. This indicates that higher ε‐CL concentrations during the reaction lead to longer PCL grafts as well as to higher levels of initiation of the grafting reaction on free hydroxyl group of silylated starch.5 and 100%. Experiment SN1CL1 SN1CL2 SN1CL3 SN1CL4 SN1CL5 a.). In the case of only homopolymerisation. Chain Length [mon. units] 40 43 44 49 54 DSPCL 0. Table 3. Chapter 3 The ratio of homopolymerisation versus grafting on starch may be obtained by comparing the integrals of selected peaks in 1H‐NMR spectra. This implies that grafting of caprolactone to starch also occurs to a significant extent. the intensity of the resonance ε’ was considerably higher that that of the ‐CH‐ iPr peak.2. This implies that the ε‐CL conversion is essentially quantitative in all cases.2.3.9 ppm).0 18. The GE increases with higher ε‐CL intakes.9 22.5 29.5.68) in THF at 50 oC with Al(OiPr)3 as the catalyst (1 mol Al(OiPr)3 per 10 mol‐starch‐OH groups). and Figure 3.3 ppm) should be twice the intensity of the ‐CH‐ proton of the isopropoxide end group (δ 4.2. Five experiments were performed to study the effect of different ε‐CL to silylated starch ratio. the intensity of the peak from the ‐CH2‐ end group of the homopolymer (ε’ at δ 3.5 Avg. The ACL of the polymer and the DSPCL increase almost linearly with the ε‐ caprolactone intake (Figure 3. b. The grafting efficiency (GE) for the sample given in Figure 3. (SN1CL1) is 28%. Overview of results for the grafting of ε‐CL on silylated starch a ε‐CL/ OH [mol/ mol] 13. and reaches 58% for a ε‐CL to starch–OH ratio of 29.5.47 0.58 0.21 0. However.2 Total Yieldb [%] >99 >99 >99 99 96. 60 .
Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors
Average Polymer Length [mon. units]
50 0.7 45 0.6 0.5 40 0.4 35 0.3 0.2 30 12 14 16 18 20 22 24 26 28 30 DSPCL
ε-CL : Starch-OH Ratio [mol/ mol]
Figure 3.5. Average Chain Length (ACL) and the DS for as a function of the CL‐ starch intake. ∆ : Average Chain Length (ACL); □ : DSPCL The mechanism of ROP of cyclic esters such as caprolactone in the presence of an alcohol (ROH, silylated starch in our case) is provided in Scheme 3.1. [30‐31]. Higher monomer intakes are expected to lead to higher polymerization rates as shown in Scheme 3.1.b. This will result in longer PCL chains in the final product, in line with the experimental observations (see Table 3.2. and Figure 3.5.). At higher ε‐caprolactone intakes not only the ACL of the grafted chain increases but also higher values for the DSPCL are observed. This finding may be rationalised by assuming that the rate of chain transfer (Scheme 3.1.c.) with starch is increased at higher ε‐caprolactone intakes. The rate of this reaction is expected to be a function of both the starch and the concentrations of Al‐species with a growing PCL chain. The starch intake for all experiments was equal, meaning that the concentration of Al‐species with a growing PCL chain should be higher at higher ε‐CL intakes. This is indeed predicted by the mechanism given in Scheme 3.1.b.; higher caprolactone intakes will increase the rate of this reaction and lead to higher concentration of Al‐species with a growing PCL chain.
Al O iPr
R OH O O kp kd
Al O R
O Al O [(CH2)5 C O]n R
Al O R
O Al O [(CH2)5 C O]n R + R OH
O Al O R
+ HO [(CH2)5 C O]n R
Scheme 3.1. The observation that a higher monomer to alcohol ratio leads to higher amounts of PCL chains with an alcohol end group and thus a higher DS was also reported for the polymerization of p‐dioxanone with Zn‐lactate as catalyst and α‐tocopherol as the alcohol . To show the potential of our approach to use hydrophobised starch instead of native starch for the ring opening polymerisation of cyclic esters, several ring openings polymerisations of native starch with ε‐CL monomer were performed either in pure ε‐CL or in a mixture of ε‐CL and toluene (80‐100oC, 24 hr). At the start of the reaction, the starch was always insoluble in the reaction medium. After reaction the product was isolated, washed thoroughly with toluene and dried. The weight of the product, however, was very close to the initial starch intake. Examination of the products by FT‐IR does not show the presence of caprolactone vibrations. Thus, it may be concluded that solubilisation of starch is of key importance to obtain poly‐(ε)‐caprolactone grafted starch co‐polymers. These findings are in line with earlier studies on the ROP of ε‐caprolactone on native granular starch using Al(OiPr)3 as catalyst . Here, caprolactone polymerization did not occur after 18‐24 h reaction time and only liquid ε‐CL was recovered. Only when performing the reaction with high concentrations of the aluminium catalyst, a product with a GE of about 13% was obtained. This low GE was explained by assuming that the reaction between starch and Al(OiPr)3 is slow and due to the heterogeneous nature of the reaction mixture. Our study, together with the result of Dubois et al  indicate that homogenous reaction conditions are required for the successful ROP of ε‐CL to obtain poly‐(ε)‐caprolactone grafted starch co‐polymers when using Al(OiPr)3 as the catalyst. When performing the reaction under heterogenous conditions, a high grafting efficiency is only achievable when using triethylaluminium as the catalyst .
Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors
3.3.3. Deprotection of Silylated‐Starch‐g‐PCL A preliminary experiment was performed to remove the silylated groups of the silylated starch‐g‐PCL (eq. 3.8.) using a mild acid treatment with diluted hydrochloric acid in THF at room temperature. All silyl groups were removed successfully, as is clearly seen from an NMR spectrum given in Figure 3.6.
Figure 3.6. H‐ NMR Spectrum of Starch‐g‐PCL (reaction product after desilylation of sample SN1CL2, DSsilylation =0.68, DSPCL=0.34), in DMSO‐d6
O O H2C H H OR H C (CH2)5 O
O H O CH2 H H OH H C (CH2)5 O
O H OR
THF, 2 hr H
O H OH
[R = H or Si(CH3)3 ]
Silylated Starch-g-Polycaprolactone (2)
986 cal mol‐1 K‐1 : temperature [K] : R T Vm molar volume [cm3/ mol] 64 .5. Silylated starch with a low‐medium DS (0. The ROP with ε‐CL was performed using Al(OiPr)3 as catalyst in THF as the solvent.21‐0. The products may have interesting applications as compatibilizers for starch‐polymer blends. This was achieved by making the starch more hydrophobic by partial substitution of the OH groups by trimethylsilyl groups. indicating that homogeneous reaction conditions are favorable for the grafting reaction.46‐0. Considerable amounts of ε‐CL hompolymers with isopropyl end‐groups were also formed. 3. The DS of the PCL chains was between 0.4.68) was obtained using a DMSO/toluene mixture as the solvent and HMDS as the silylating agent. Chapter 3 3. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] A x − y ppm ACL DS DSsilylation : DS of silyl group substituents [‐] DSPCL GE : DS of PCL chain substituents [‐] : Grafting Efficiency. the highest value was obtained with a ε‐CL‐AHG ratio of 29. The key feature is the use of a homogeneous reaction mixture for the ROP of starch with ε‐CL. depending on the ε‐CL‐starch ratio. The grafting efficiency varied between 28‐58%.72. the percentage of PCL grafted to starch compared to the total amount (grafted and homopolymer) of polymerized CL [%] : gas constant. Poly‐(ε)‐caprolactone grafted silylated starch co‐polymers with average chain length of 40‐55 monomer units (polymer molecular weight of 4500‐6300) were obtained. Nomenclature A : peak area of certain proton in 1H‐NMR spectra [‐] : peak area of certain peak at x until y ppm in 1H‐NMR spectra [‐] : average number of CL repeating units in a grafted polymer chain or homopolymer [monomer units] : Degree of Substitution. 1. Conclusions The successful synthesis of poly‐(ε)‐caprolactone grafted corn starch co‐ polymers using a three step approach is reported.2. Control ROP experiments of ε‐CL with native starch under similar conditions did not produce the desired poly‐(ε)‐caprolactone grafted corn starch co‐polymers.
385–409 . The compelling facts about plastics: An analysis of plastics production. 56. K. J. 1998. L. Chandra. D. Frederiksberg C. C. 61‐67. report of Food Biopack Project (Ed. . R. Belgium.dk/isi/stat/rawmaterial. A. 43. . Y. 61‐74. Sci. 2003. demand and recovery for 2006 in Europe.html. 2003. A. Eng. 8. . Goldberg: Review of the biodegradability and utility of poly(caprolactone). Department of Dairy and Food Science. H. 1998. 1996. Babac: Preparation and biodegradation starch/polycaprolactone films. A. 38. 52. C. . Website of International Starch Institute. Polym. Sci. Part C: Polym. E. Albertsson: Biodegradable polymers and environmental interaction. 1977. Adv. C. 2008. Prog. R. . 407‐412. J. J. Polym. 2000. Starch‐Starke 2000. Rev. 1995. Rustgi: Biodegradable polymers. Chiellini.B. 311‐316. Eng. Fanta. 17.6. C. Polym. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Greek symbols: ∆H v : heat of vaporization [kJ/mol] : δ Hildebrand solubility parameter [cal1/2cm‐3/2] 3. of . G.A. J.K. 2008.. 65 .. Association of Plastics Manufacturers. Beliakova.C.A. Linke: Water vapor sorption determination of starch based porous packaging materials. Solaro: Biodegradable polymeric materials.C.starch. Yang. Aly. 11. Degrad. accessed on August 01. Russell: Graft copolymers of polysaccharides with thermoplastic polymers ‐ new type of filled plastic. Doane. . Food Technol. R. Polym. X. Macromol. 23. W. Polym. Krochta. S. . . 1251‐1253. Weber). Yavuz.A. 53‐57.M. J. De Mulder‐Johnston: Edible and biodegradable polymer films: Challenges and opportunities. Environ. Plastics Europe.R. 3. R. Denmark: http://www. Biobased packaging materials for the food industry. . Starch‐Starke 2004. . F. 107‐113. References . the Royal Veterinary and Agricultural University. Mater. 1273‐1335. 305‐313. Wang: Properties of starch blends with biodegradable polymers. Karlsson. Abdel‐Mohdy: Grafting of poly(methacrylic acid) on starch and poly(vinyl alcohol).Sci. E. Sci. 1997. Environmen. status and perspectives. Bagley.F.M. Aarhus. Denmark. M. 51. Tsiapouris. Wang. Brussels. Burr.
Six: Partial or total silylation of dextran with hexamethyldisilazane. Stein. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer 66 . Duquesne. 40. Klemm. K. 163. 43. 1. Six: Controlled synthesis of amphiphilic biodegradable polylactide‐grafted dextran copolymers. Chem. Biosci. 1443‐1450. Biomacromolecules 2003. K. . Einfeldt. P. Gunther. J. K. Petzold. Gunther: Synthesis and NMR characterization of regiocontrolled starch alkyl ethers. . 2577‐2588. 2002. Dubois. Polym. C. M. Klemm.L. 2001. Dole. Averous. 2. Degee. . Moro. Stab. 35‐47. A. E. D. Kussler. D. Six. H. P. C. W. 10. Dubois: Aliphatic polyester‐ based biodegradable materials: new amphiphilic graft copolymers. 415‐426. Dellacherie. T. Klemm: Preparative and 1H NMR investigation on regioselective silylation of starch dissolved in Dimethyl Sulfoxide. 6713‐ 6721. Krishnan. Dubois. Nouvel. L. Petzold. I. Chapter 3 . Heublein: Silylation of cellulose and starch ‐ selectivity. Hamaide. Macromol. L. M. P. A. Pantiru. 251‐269. Klemm: Regioselective functionalization of starch: Synthesis and 1H NMR characterization of 6‐O‐ silyl ethers. J. R. L. Dubois. C. Nouvel. D. Symp. Degrad. 1735‐ 1743. Macromol. Rutot. Dellacherie. 965‐969. structure analysis. J. L. . Monomers Polym. Ydens. high value applications. Six: Silylation reaction of dextran: Effect of experimental conditions on silylation yield. P. E. P. Polymer 1999. 5.L. Ydens. Degee. 2004. Einfeldt: Structure design of polysaccharides: novel concepts. Part A: Polym. Biomacromolecules 2001. Ydens. 561‐566. 33. Cellulose 2003. P. . Nouvel. Polym. Fringant: Properties of thermoplastics blends: starch‐polycaprolactone. regioselectivity. D. Stein. and subsequent reactions. Klemm. A. Dubois. E. 4. 4157‐4167. M. Polymer 2002. Stein. P. P. Polymer 2000. and chemical stability of silylated dextrans. 2001. . D. J. D. . 42. 73. Dellacherie. Degee. Des. W.L. 3091‐3100. 2001. Dubois: Controlled synthesis of poly(epsilon‐caprolactone)‐grafted dextran copolymers as potential environmentally friendly surfactants. E. P. I. Koschella. . selective syntheses. Rutot. Dellacherie. 341‐347. Sci. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. D. P. Gunther. Macromolecules 2000. W. . 41.L. . Petzold. A. Petzold. . B. . J. K. L. Fessi. E. D.. I. P. Einfeldt.
Penczek. Kinetics and mechanism. 1399‐1406. A. Mexico City.: Solubility parameter concepts – a new look. H. 199. Janssen. 941‐950. H. Macromolecules 1996. T. 553‐556. S. Kricheldorf. D. . 1978. 1998.B. 22.K. Picchioni. Macromol.4‐ dioxan‐2‐one. Rapid Commun. L. 67 . accepted for publication in Starch‐Starke. 29. .P.L. 1089‐1097. Damrau: Zn‐lactate catalyzed polymerizations of 1. Macromol Chem. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. D. Biela. Rapid Commun. Lehmann: Selective 6‐O‐acetylation of amylose. 2001. Macromolecul. A. 2008. R. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors agents. . Sugih. L.J. Duda: Living polymerization with reversible chain transfer and reversible deactivation: the case of cyclic esters. A. Hertz Jr. 21. 2000. Manurung.R.M. . 1989 . Phys. A novel route to functionalised nanoparticles. D. . Horton. 61. presented at ACS Rubber Division meeting. J. Duda: Polymerization of ε‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. Carbohydr.O. F. Junistia. Res. 659‐663.
The yellowish products were characterized by 1 H‐.24 ‐ 2. vinyl laurate. At higher ratios. the DS for the starch‐laurate compounds is higher than for the corresponding starch‐ stearates. vinyl ester to starch ratio and the type of catalyst. presumably due to a reduction of the polarity of the reaction medium. esterification. K2CO3. Keywords: corn starch. an increase in the vinyl‐ester concentration leads to higher product DS values. the DS decreases. and Na‐acetate are superior catalysts with respect to activity compared to Na2HPO4 and products with DS values close to 3 were obtained.96). When performing the reactions using Na2HPO4 as the catalyst. K2CO3. For low vinyl‐ester to starch ratios. 13C‐NMR and FT‐IR. The fatty esters were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4.Chapter 4 Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants Abstract This chapter describes the synthesis of long fatty esters of corn starch (starch‐ laurate and starch‐stearate) with a broad range in degree of substitution (DS= 0. The DS of the products is a function of the carbon number of the fatty acid chain. vinyl stearate . and Na‐acetate) in DMSO at 110°C.
and particularly acetic acid derivatives (C2) [2‐4]. However. 5]. cellulose or proteins are potentially very interesting starting materials for biodegradable packaging materials. 7]. Further modification of starch is therefore required to introduce hydrophobicity and to improve mechanical and moisture barrier properties. most of the studies performed to date use short chain carboxylic acids (C1‐C4). the mechanical properties of high‐ DS starch derivatives of low chain carboxylic acids still need considerable improvements before large scale application as packaging materials becomes within reach. and incompatibility with hydrophobic polymers. The major obstacle is the pronounced brittleness of the materials. and consequently. The first paper on the acetylation of starch was already published in 1865 . The mechanical properties and hydrophobicity of the products were significantly improved when using these longer chain fatty acid precursors [1. An alternative method using methyl and glyceryl laurate esters in the absence of solvent has been recently developed . 5]. more water‐resistant products may be obtained [3‐4]. Relatively low‐DS (0. which are relatively expensive and rather corrosive . resulting in products with DS values up to 2.7 [1. In particular starch is attractive as it is relatively cheap and abundantly available. Natural polymers such as starch. the fatty ester substituents [1. higher molecular weight carboxylic starch esters (C4‐C6) . Esterification of starch with low molecular weight fatty acid derivatives is one of the oldest modification technologies to improve starch properties. even after the addition of plasticizers . Recently Mormann et al  explored the possibility of using vinyl esters and particularly vinylacetate as reagents for the preparation of starch esters. However. and even fatty acid derivatives (C12‐C18) have been used in the modification reaction [5. An attractive field of application for these polymers is the use as packaging materials. poor processibility (high viscosity). 5‐6] were introduced using fatty acid chloride reagents.34‐0. Their 70 .Chapter 4 4. the use of native starch for packaging materials is limited due to its low moisture resistance. For the current petrochemical based products recycling is often neither practical nor economically feasible . The hydrophobicity increases with the degree of acetate substitution (DS. To improve the mechanical properties.61) products were obtained using this approach. Introduction Green biodegradable polymers derived from natural resources are potentially very interesting substitutes for non‐biodegradable petroleum‐based polymers. The introduction of acetate groups on starch makes the product more hydrophobic. high brittleness. However. However. defined as the moles of substituents per mole of anhydroglucose (AHG) units) .1.
Technical grade dimethyl sulfoxide (DMSO). and tetrahydrofuran (THF) were supplied by Acros (Belgium) and were also used as received. Japan). vinyl laurate (Fluka. Germany) were used as received. 4‐N.2. IR spectra were recorded on a Spectrum 2000 FT‐IR Spectrometer (Perkin Elmer). 1 71 . not been reported to date. The effect of the addition of a non‐polar solvent (toluene) to the reaction medium to solubilise the products and thus to enhance the reaction rates has also been studied. Germany) and acetic anhydride (Merck.2. the use of basic catalysts other than Na2HPO4 has been investigated. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. Germany) were used without further purification. The synthesis of starch stearate esters using vinyl ester reagents has. In DMSO. the Netherlands). Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants research focused on the synthesis of starch acetates and only two examples with a higher molecular weight fatty acid vinyl ester were reported. The starch was dried before use for 48 h at 105 oC under vacuum (approx. ~1 mbar).6 for starch acetate ester) were obtained. leading to higher reaction rates. The spectra were recorded at 50 oC. We here report an investigation on the synthesis of higher fatty acid esters of starch with an emphasis on the introduction of laurate and stearate ester side chains. The effects of the starch to vinylester ratio on the reaction rates and DS have been explored. Potassium carbonate (Boom. 4. This solvent effect is likely caused by the higher solubility of the vinyl esters in DMSO compared to water. Materials and Methods 4.2.N‐dimethylaminopyridine (DMAP). leading to a moisture content of 2 %‐wt (measured gravimetrically). The reactions were either performed in water or in DMSO using a basic catalyst (Na2HPO4). 4. The maximum attainable DS of starch acetate in water was below 1 and limited to 0. as recommended by Laignel et al .1. Analytical grade vinyl stearate (Aldrich. to the best of our knowledge.01 when using vinyl‐laurate. Germany) and disodium hydrogenphosphate (Merck. sodium acetate (Merck.2. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). In addition. starch esters with a substantially higher DS value (up to 1. Materials Corn starch (approx. The products were placed directly on the diamond plate and 50 scans with a resolution of 4 cm‐1 were recorded.
broad peaks.0 (m.5 g) was first gelatinized in DMSO (5 ml) at 70 oC for 3 h. 5S). C2S. Typical example of the synthesis of laurate and stearate esters of corn starch Corn starch (0. 2H.7 (C11). Table 4. 22.1.6 (m. and O‐C6S). 3S.52): H‐NMR (before peracetylation. C2). 1230 (C‐O stretching). approximately 5 mbar) for 24 h until constant weight.2. After cooling. 3‐6 ppm (m. broad peaks. O‐C3S.0 (C12). typical spectra are given in Figure 4. C2). 2.3. 2H.1 (m. C3). (1H‐NMR). C6S). 1020 (C‐O stretching). Starch‐stearate (Sample 19. 1410 (C‐H bending).8‐2. 1350 (C‐H bending). 3‐6 ppm (m. C2). 2H. DS = 2. 2.2. The samples were characterized by 1H‐ and 13C‐NMR and FT‐IR. 28H. 1..4 (broad. 28‐32 (C4‐9).5 (m. 1 C‐NMR (before peracetylation. 760. 935 (C‐O stretching).3 (m. broad peaks. attached to O‐C2S. 3H. 1.1. 12). broad peaks. Table 4. broad peaks. 3H. broad peaks.3. CDCl3): δ 0. 16H. C1S). C4‐11). the product was precipitated using methanol (100 ml) and separated from the liquid phase by decantation. The atom numbering scheme is given in Figure 4.4 (m. 1295. CDCl3): δ 14.4. 1.3. C3). 31. 1455 (CH2). broad peaks. DS = 2. 172‐174 ppm (C=O. 3H.96) H‐NMR (before peracetylation. 7H. 1 H‐NMR (after peracetylation. 2H.2.. (13C‐NMR) and Figure 4.5 (m. Figure 4. C12). Methods 4. 1150 (C‐O stretching). 1. 1. 61. 13 FT‐IR (cm‐1): 2920 (C‐H stretching). C3).1. CDCl3): δ 0. C2’). C4‐C17). Finally. C1S‐6S) 1 72 ..5 (m. C1S‐6S). resulting in the formation of a homogenous transparent solution. 95. 68‐74 (broad. Starch‐laurate (Sample 17. Subsequently.9 (t. 1740 (C=O). C1S‐6S). C18).1 (C11). 16H. (FT‐IR). The product was washed twice with methanol (50 and 25 ml.9 (t. 2H. 24.3 (m. CDCl3): δ 0.9 (broad. 7H. 1110 (C‐O stretching). 76‐78 ppm. 2H. overlap with CDCl3 (C4S). 1370 (C‐H bending).9 (t. respectively). 3H. 1.3 (m. broad peaks. 2850 (C‐H stretching).Chapter 4 4. 2. 1. vinyl laurate or vinyl stearate (3 mol/mol AHG units in starch) and potassium carbonate catalyst (2 %‐wt with respect to starch) were added and the mixture was stirred at 110 oC for 24 h. 720. 34.1. 7H. the product was dried in a vacuum oven (70 oC.9 (C10).9 (C3). 3‐6 ppm (m. C4‐11).
1100 (C‐O stretching). C1S). 25. 5S). C2S.0 (C3). 2H. 1410 (C‐H bending). A peracetylation reaction to substitute all of the remaining 73 . 3S. C2). C3).4 (broad. 75.0 (C16). CDCl3): δ 14. 1. attached to O‐C2S. 3H.5 (m. 28H.2. 32. 22.8‐2.9 (t. 2850 (C‐H stretching).1.7 (C17). 1150 (C‐O stretching). 950 (C‐O stretching).7 (C4S). C4‐C17). 1350 (C‐H bending). O‐C3S.2 (C2). and O‐ C6S) 13 FT‐IR (cm‐1): 2920 (C‐H stretching). 2. 6S O R 5S O R O 4S O 1S 2S 1 3S O O 2 3 4 5 6 7 8 9 10 11 12 CH3 Starch-Laurate O 1' 6S O CH3 2' O O R O 4S 5S O 1S 2S 3S O O 1 2 3 4 5 6 7 8 9 10 11 12 CH3 Starch-Laurate peracetylated 6S 4S O O R 5S O R O O 1S 2S 3S O O 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 CH3 Starch-Stearate O Figure 4. 3H. 1740 (C=O). C6S). 95. 68‐74 (broad. 61.5 (broad.3 (m. 3‐6 ppm (m. 1370 (C‐H bending). 1. 26‐32 (C4‐15). broad peaks. CDCl3): δ 0.2.3. 1020 (C‐O stretching). 34.0 (C18). 1455 (CH2). 1. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants H‐NMR (after peracetylation. broad peaks. C1S‐6S) 1 C‐NMR (before peracetylation. C2’). 760. Peracetylation procedure The presence of remaining hydroxyl groups in the products resulted in broad and overlapping starch resonances in 1H‐NMR spectra  and hampered calculation of the DS. 720. C18). 172‐174 ppm (C=O. 1295.6 (m. 865. Numbering scheme for carbon atoms of products 4.4 (m. 2H. 7H.
The peracetylation procedure by Einfeldt et al  was applied.4.. 4.8 ppm (protons attached to the fatty acid carbons C3‐C12 for laurate and C3‐C18 for stearate) using eq. and 4.) The DS of the acetate groups of the products may be calculated by comparing the unit area of the acetate protons (AH‐acetate) to the unit area of the starch protons (calculated from the intensity of the starch peaks at δ 3.5‐1.3.6.) Determination of the unit area of the acetate H‐atoms (AH‐acetate) is hampered by peak overlap with the H‐atoms attached to C2 (AC2) of the fatty acid chains (δ 1. 4. It was finally dried overnight in a vacuum oven at 70oC and 5 mbar until constant weight.6−5.2. The DS of the fatty acid esters was calculated from the DS of the products after peracetylation (eq 4.: 74 .3. The product was precipitated by the addition of methanol and washed several times with methanol.2.8− 2.6 7 (4. The peracetylation reaction was conducted for 7 h at 50 oC. Determination of the Degree of Substitution (DS) The DS of the product was calculated using 1H‐NMR spectra of the products after peracetylation.).1.5.3.) AH − starch = (4. and 4.7.8‐2. Typically. the starch ester (0.: DSacetate = AH − acetate AH − starch A3.6 − AC 2 3 (4.6 ppm range) and a correction has to be made (eq 4.2.Chapter 4 hydroxyl groups with acetate groups was applied to obtain reliable DS data. The procedure was described earlier by Elomaa et al  and the relevant equations are shown in equation 4. 4. Subsequently.5.3.).6‐5.1. AH − acetate = A1.4.) AC 2 = 2 × AH − fattyester The AC2 values of both the laurate and stearate side chains were calculated from the peak intensity in the range δ 0. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.6 ppm).1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until the starch was fully dissolved (typically 3 h).) (4. DS fatty ester = 3 − DS acetate (4. the peractylating reagents (DMAP.
After 2‐3 h. The reaction was performed in two discrete steps. 4.8 = (stearate) 33 33 (4.8 = (laurate) 21 21 AC 3−C18 A0. the starch was gelatinised in DMSO at 70°C for 3 h to make the hydroxyl groups of starch more accessible for reaction. 75 .7. A schematic representation of the esterification reaction of starch with the vinyl esters is provided in Scheme 4. OH O HO OH O OH + R O CH2 O HO O R + n H3C H O n O (1) R = laurate (C12) (2) R = stearate (C18) Scheme 4. Subsequently. while ACx‐Cy is the area of the H‐atoms attached to carbons in the range Cx‐Cy (carbon numbering scheme is given in Figure 4.) (4. 4. the esterified starch started to separate from the medium in the form of a gel. the vinylester and the catalyst were added and the reaction mixture was heated to 110°C. the brownish gel was precipitated with methanol and the product was collected after vacuum drying in the form of a transparent. After 24 h. The products of these exploratory reactions are insoluble in water and DMSO.1. Ax‐y stands for the peak area in the range δ x‐y ppm..) In eq.5−126.96.36.199).3. but swell in organic solvents such as toluene and THF. Results and Discussion 4.6. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants AH − fattyester = AH − fattyester = AC 3−C12 A0.7. light yellow solid.5−1. Initially. Exploratory Experiments A number of exploratory experiments were carried out with vinyl‐laurate and vinyl stearate (vinyl ester: AHG molar ratio of 1 : 3) at 110°C for 24 h in DMSO using K2CO3 as the catalyst.3.‐4.1.
1.2.2. 4. Typical 1H‐NMR spectrum of: (a) native starch in DMSO d‐6 at 60oC.52 (Sample 17.3.3.). For atom numbering scheme: see Figure 4. A reaction with vinylstearate at similar conditions resulted in a stearate starch ester with a DS of 2.) in CDCl3 at 50oC. Higher DS products have a higher solubility in CDCl3 and good quality 1H‐ and 13C‐NMR spectra could be obtained (Figure 4. Table 4. 1H‐ and 13C‐NMR analyses The solubility of the products in common NMR solvents (DMSO‐d6 or CDCl3) is a function of the product DS.52 was obtained.3.1.1. 76 .96. Table 4.) in CDCl3 at 50oC. (b) starch laurate.1. a product DS of 2. 4. DS = 2. Medium DS starch laurate and starch stearate (1 < DS < 2) dissolve poorly in DMSO‐d6 and CDCl3. even at higher temperatures (50 oC).2.2.Chapter 4 The DS of the products was determined by using NMR (vide infra). (c) peracetylated starch laurate. When using a vinyl‐laurate : AHG molar ratio of 3 and K2CO3 as the catalyst. DS = 2.52 (Sample 17. and 4. Product Characterisation 4. Figure.
and therefore a peracetylation procedure to substitute all of the remaining OH groups with acetate groups was applied [8. are together with the CH2 groups of the acid chain adjacent to the ester moiety in the range δ 1.c.52 (Sample 17) in CDCl3 at 50oC. required for DS determinations. DS = 2. 10‐11]. The resonances arising from the anhydroglucose unit of starch are broadened. NMR spectra of the peracetylated products are considerably improved in terms of peak resolution and allow a more reliable calculation of the DS. Typical 13C‐NMR spectra of: (a) native starch.8‐2.5 ppm). Two of the carbon resonances (1S and 4S) are considerably shifted compared to native starch. DS = 2.3. For atom numbering scheme: see Figure 4. This feature hampers the DS determination by NMR. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants A typical 1H‐NMR spectrum of starch laurate is shown in Figure 4. Clearly visible are the peaks arising from starch and the aliphatic hydrogen atoms of the fatty acid chain (δ 0. (c) starch stearate. (b) starch laurate.3. Clearly visible are the carbon resonances of the fatty ester chains (δ 10‐35 ppm) and C atom of the ester group (δ 170‐175 ppm).1. The starch peaks (δ 3‐5.8‐2. in DMSO d‐6 at 60oC. The proton signals of the acetate methyl group. The same 77 .96 (Sample 19) in CDCl3 at 50oC.3 ppm.2.2. Figure 4.5 ppm) are broad and overlapping . The 1H‐NMR spectrum of a typical peracetylated starch laurate is shown in Figure 4. Typical 13C‐NMR spectra of the products are given in Figure 4.
2. The shift of the starch peaks and the presence of peaks arising from the fatty ester chains clearly indicate that the esterification reaction with vinyl laurate and vinyl stearate was successful.96.c.Chapter 4 phenomenon was observed by Dicke for starch acetate . c. Table 4. Sample 19.4. a spectrum of native starch (Figure 4.4.4. Sample 17. starch stearate (DS= 2. Characteristic peaks of the polysaccharide backbone are visible in the 1250‐900 cm‐1 region (C‐O stretching) .4. Table 4.2. the C‐H stretching vibrations of the alkyl groups of the fatty ester chain are clearly present at 2920 and 2850 cm‐1. For comparison.) show characteristic bands of the carbonyl group of the fatty esters in the 1750‐1700 cm‐1 region. starch laurate C-H C=O C-O C-O a. 4.4.a. starch stearate C-H C=O C-O Absorbance b. The near absence of remaining hydroxyl 78 . FT‐IR Spectra of starch laurate (DS= 2. In addition.b.52. and 4. native starch -OH C-H 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Figure 4. FT‐IR measurements The FT‐IR spectra of starch laurate and starch stearate are shown in Figure 4.) is also included.4.).1.4.b.3.) and native starch FT‐IR spectra of both starch laurate and starch stearate (Figure 4.1. and 4.c.
01 0. The results are shown in Table 4. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants vibrations in the range 3000‐3600 cm‐1 and at 1640 cm‐1 indicates that the DS of the product is high.68 0.52 2.23 0.1. 79 . 4.07 0.13 1. Overview of the esterification of starch using vinyl‐esters and basic catalysts a Catalyst Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 K2CO3 CH3COONa K2CO3 CH3COONa Vinyl Ester: AHG ratio [mol/ mol] 2 3 6 2 3 6 2 3 4 6 2 3 4 6 3 3 3 3 Amount of Toluene added [ml] ‐ ‐ ‐ 5 5 5 ‐ ‐ ‐ ‐ 5 5 5 5 ‐ ‐ ‐ ‐ DS 1.57 0. four experiments were performed with two alternative basic catalysts (K2CO3 and Na‐acetate).90 0.96 2. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 a Vinyl Ester laurate laurate laurate laurate laurate laurate stearate stearate stearate stearate stearate stearate stearate stearate laurate laurate stearate stearate All experiments were performed at 24 h at 110°C in DMSO with a catalyst intake of 2 wt% based on starch. In addition.60 1.1.05 0.2.99 1. Most of the experiments (14) were performed using Na2HPO4 as the catalyst.44 Exp. Systematic Studies The effect of important process variables like the vinyl‐ester to AHG ratio.54 2.08 1.24 2.91 0. Table 4. type of catalyst and the effect of the addition of co‐solvents on the product DS was studied in more detail. in line with the NMR data.3.90 1.
23 for vinyl‐laurate at an intermediate vinyl‐ester‐AHG ratio of 3. Effect of vinyl ester to AHG ratio on the product DS The effect of the vinyl‐ester to AHG molar ratio on the product DS was determined for both types of vinyl‐esters with Na2HPO4 as catalyst (samples 1‐ 3. This behaviour is likely the result of two opposing effects. 2 wt% catalyst intake on starch).5. At a ratio of 1:6. DS of the product as a function of the type of vinyl ester and the vinyl ester‐starch ratio (24 h reaction time. Higher concentrations of vinyl esters are expected to lead to higher esterification reaction rates. The highest DS value was 1.3.5 Degree of Substitution (DS) 1 0.3. the ester intake is equal on a weight basis to the DMSO intake. The DS of the products is a clear function of the vinyl laurate and stearate ester intakes (Figure 4.5. The DS values are increasing with higher vinyl‐ ester/AHG molar ratio until a certain maximum. A further increase in the vinyl ester intake leads to a reduction in the DS.1. At low to medium vinyl ester/AHG ratios (0‐3). 1. 110 oC.5 0 0 1 2 3 4 Vinyl Ester: AHG Starch Ratio [mol/mol] 5 6 Figure 4. This reduced polarity is expected to lead to ○ : starch‐laurate □ : starch stearate (lines for illustrative purposes only) 80 .). 7‐10).5. this positive effect dominates the reaction rate and the DS of the products will therefore increase at higher vinyl ester intakes.Chapter 4 4. A further increase leads to a reduction in the DS. This is likely due to a reduction of the polarity of the reaction medium. The results are presented in Figure 4.
the DS of the product is also a function of the chain length of the fatty acid.1.).3.6. Catalysts screening A number of alternative basic catalysts for Na2HPO4.3. A similar procedure was proposed by Nouvel. Although toluene may positively effect the reaction by (partly) re‐dissolving starch ester precipitates. 11‐14 in Table 4. The addition of toluene for the esterification of starch with vinyl‐esters surprisingly did not lead to improved DS values. and illustrated in Figure 4. These findings were rationalised by assuming that the co‐solvents increase the solubility of the silylated products. 4.3. 4. with reductions in polarity leading to lower reaction rates. It is clear that Na‐acetate and K2CO3 are considerably more active than Na2HPO4 and products with a significantly higher DS were obtained. This effect is particularly evident at higher vinyl ester/ AHG ratios (>3) (Figure 4. the addition of co‐solvents (toluene/ THF) led to an increase in the DS. et al  reported the synthesis of fatty esters of starch using alkanoyl chloride reactants (C8‐C18) with reactant ratio of 6 mol alkanoyl chloride/mol AHG.1.3. In this case. the reactions were initiated in DMSO and toluene was added after 12 h reaction time to re‐dissolve the poorly soluble partially‐esterified starch products (entry 4‐6.e. et al  for the silylation of starch. Aburto.2.3. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants a lowering of the reaction rates due to a reduction of the solubility and degree of ionization of the starch reactant as well as the base catalysts. For starch 81 . A similar trend in reactivity pattern was observed in this study and the DS decreased from 1. it also results in a dilution of the reaction mixture and a reduction in the polarity. leading to enhanced reactivity. i. The latter factors appears to have a strong effect on reaction rates (vide supra).8 for stearoyl chloride. The results are given in Table 4.). Thus. These negative effects dominate the reaction performance at higher vinyl ester/AHG ratios and lead to a reduction in the DS values. the starch laurate esters display higher DS values than the starch stearates.). Here. The difference was explained by assuming that the reaction rate is reduced for larger reagents due to steric effects and this explanation likely also holds for the reactions with the vinylesters . The products have about the same DS value for vinyl‐laurate when using only DMSO and even reduced DS values for vinyl‐stearate (see Table 4.1.7 for lauroyl chloride to 0. K2CO3 and Na‐ acetate were tested. with high carbon numbers leading to a reduction in the DS. When using Na2HPO4 as the catalyst. Effect of the addition of toluene as a co‐solvent A number of reactions were performed using a co‐solvent.5.
and FTIR and confirm the presence of chemically bound fatty acid chains. the DS of the product is also tunable by proper catalyst selection. Conclusions A study on the synthesis of corn starch fatty acid esters with high DS values is reported. DMSO). However. The DS of the laurate ester when using Na‐acetate is higher than for the stearate ester (Table 4.6. 82 . the statement that the DS for the laurate esters is always higher than for the stearate esters is not generally valid and a. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. The yellow products were characterized by 1H‐ and 13C‐NMR.Chapter 4 laurate esterification. Thus.).5 were obtained. the DS for the laurate ester is lower than the stearate ester. For starch stearate. K2CO3 gave products with a significantly higher DS (2. in line with the findings for NaH2PO4.1.6. when using K2CO3 as the catalyst.5 0 Laurate Vinyl-Ester Stearate Figure 4. and Na‐acetate).96) compared to Na‐acetate (DS=2. Apparently.o. 3 2. 110oC. function of the type of catalyst. The DS of the products is a clear function of the chain length of the fatty ester and the type of catalyst.4.44). the two catalysts are equally effective and products with a DS of about 2. gray: Na2HPO4 black: K2CO3 white: Na‐acetate 4. The products were synthesised in DMSO using vinyl‐esters in the presence of basic catalysts (Na2HPO4. catalyst amount = 2%‐w.5 1 0.5 Degree of Substitution 2 1. K2CO3. and Figure 4. Comparison of DS values with different catalysts (Vinylester: AHG molar ratio = 3:1.
Chem. M. 21. Prinos. Alric. Y. 58. 74. . Y. Polym. I. References . Pacsu: Starch Studies: Preparation and Properties of Starch Triesters. 1995. Appl. Mullen and E. Bengtsson. K. . and biodegradability of fatty‐acid esters of amylose and starch. 361‐368. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants With these catalysts. Polym. Polym. 34. Dzenis. E. J. . C. 65. S. 1208‐1217. 83 . J.5. 1‐11. and P. Borredon. Eng. At low vinyl ester/ AHG ratio.W. X. Appl. characterization. C. Bikiaris. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] DS DS fatty ester : DS of fatty ester (laurate or stearate) group substituents [‐] DSacetate : DS of acetate group substituents after peracrtylation [‐] 4. Crops Prod. Panayiotou: Synthesis. 1997. Koch. Alric. D. I. J. Nomenclature A A x − y ppm : peak area of certain proton in 1H‐NMR spectra [‐] : peak area of certain peak at x until y ppm in 1H‐NMR spectra [‐] : Degree of Substitution. J. 1942. A maximum was observed at a vinyl ester: AHG ratio between 2 and 4. Appl. Carbohydr. Ind. Thiebaud. J. Hanna: Water solubility. E. Sci. . Bikiaris. Aburto. A. Thiebaud. D. Merrill: Properties of Fatty‐Acid Esters of Starch. Aburto. S. 1647‐1656. D. 2003. Xu. The DS of the products may also be tuned with the vinyl ester/AHG molar ratio. Sci. Borredon. Gatenholm: Surface octanoylation of high‐ amylose potato starch films. J. 705‐721. J. Prinos. 54. Higher ratios led to a reduction in the DS. thermal characteristics and biodegradability of extruded starch acetate foams. 1440‐1451. the DS of the product increases at higher vinyl ester intakes. W. products with a DS > 2. Ind. Polym. M. . Sagar. 2005. Sci.6. Important product properties will be described in the next chapter. 1999. J.4 could be obtained for both laurate and stearate esters. A. E. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. 4. presumably due to a reduction of the polarity of the reaction medium.
S. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Aburto. K. Laignel. M. Einfeldt. Bliard. Alric. Starch‐Starke 2004. J. Elomaa. Biosci. Nouvel. 43. Res.F. . Asplund. Senocq. 2004. L. 261‐267. T. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Polymer 2002. J. 11. R. Kussler. 57. M. . F. Alric. Dicke: A straight way to regioselectively functionalized polysaccharide esters. 84 . C. 1. Laatikainen. W. . H‐1 NMR and TGA/IR. J. Al‐Higari: Acylation of starch with vinyl acetate in water. Urtti: Determination of the degree of substitution of acetylated starch by hydrolysis. Degee. 145‐152.L. Carbohydr. C. G. Hamaili. Med. M. 2003. 56. . Gunther. Carbohydr. P. R. B. 118‐121. J. Starch‐Starke 1999. Sci. Hyvarinen. 57. 14(2) 127‐135. Dellacherie. Mater. . Peltonen. A. . 341‐347. D. Klemm: Preparative and H‐1 NMR investigation on regioselective silylation of starch dissolved in dimethyl sulfoxide. Aburto. P. Starch‐Starke 2005. and E. Soininen. A. 1997. Stein. R. I. . J. S. Polym. D. Six: Partial or total silylation of dextran with hexamethyldisilazane. Petzold. 302‐307. 51. W. Mouysset‐Baziard. Mano. L. 2001. E. P. Massiot. 298.. I.Chapter 4 . Koniarova. H. M. 251‐260. J. . Dubois. Borredon: Free‐solvent Synthesis and Properties of Higher Fatty Esters of Starch – Part 2. G. Macromol. Cellulose 2004. Reis: Thermal properties of thermoplastic starch/synthetic polymer blends with potential biomedical applicability. E. 1735‐1743. Mater. Mormann. Ydens. 255‐263. I.
The reactions were carried out in DMSO using basic catalysts (Na2HPO4. The effect of the process variables (vinyl ester to starch ratio. The results were adequately modeled using a non‐linear multivariable regression model (R2≥0.7‐3.39) are totally amorphous whereas the low‐DS ones (DS= 1. The thermal stability of the esterified products is higher than that of native starch. catalyst intake. trans‐esterification.75) are still partially crystalline. The basicity of the catalyst and the reaction temperature have the highest impact on the product DS.96). elongation at break of 3‐26%.Chapter 5 Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Abstract This chapter describes a systematic study on the synthesis of higher fatty esters of corn starch (starch‐laurate and starch‐stearate) by using the corresponding vinyl esters. reaction temperature and type of the catalyst) on the Degree of Substitution (DS) of the starch laurate and starch stearate esters was determined by performing a total of 54 experiments.26‐2. and Na‐acetate). vinyl laurate. The thermal and mechanical properties of some representative product samples were determined. Keywords: starch esters.45‐1. vinyl stearate .5 MPa. K2CO3. High‐DS products (DS= 2. Mechanical tests show that the products have tensile strength (stress at break) between 2. and modulus of elasticity of 46‐113 MPa.
et al . The use of methyl and glyceryl esters to prepare starch laurate (C12) ester was investigated but only yielded products with relatively low DS values (max. Sagar and Merill  studied the synthesis and properties of C4‐C6 esters of high amylose starch. 0. packaging material. and this causes serious environmental problems [1. Most early studies mainly focused on the use of short‐chain carboxylic acid (C1‐C4). Two examples using higher vinyl esters (vinyl laurate) were reported by Mormann. The use of vinylesters has also been explored [20‐22]. the mechanical properties were still not satisfactorily. Examples are agricultural mulch . however. The global starch production was estimated at 60 million tons in 2004 . However. the physicochemical properties of the products were not mentioned. Mullen and Pacsu [11. 12] studied the synthesis and properties of of C1‐C6 esters of starch. and food service‐ware . Thiebaud. The development of novel biodegradable plastic materials capable of decomposing when given an appropriate environment and time is of utmost importance . Alternative routes to avoid the use of acid‐chloride/pyridine combinations have been developed. however. and incompatibility with hydrophobic polymers . most of the plastic waste is not biodegradable. However. Starch is an attractive feedstock for the synthesis of novel biodegradable plastics.Chapter 5 5. The mechanical properties of products with longer ester chains (C4 and C6) products and plasticized acetate esters were considerably improved compared to native starch. 86 . polyethylene  and polycaprolactone  and chemical modification by esterification with organic acids [11‐19]. It is cheap and abundantly available . The use of native starch as a building block for novel biodegradable polymers. Esterification is one of the oldest methods used to improve starch properties . is rather limited to date. and particularly on the synthesis of starch acetate (C2) [11‐12]. The limited use is mainly because of a number of unfavourable properties of starch such as low moisture resistance. The high DS esters showed interesting properties. et al  and Aburto. Typical examples are blending with polyvinyl alcohol . high brittleness. The largest application area of plastics is the use as packaging material (37%) . 3‐4]. et al [15‐19] synthesized longer chain fatty esters (C8‐C18) of potato starch and corn amylose using fatty acid chlorides and pyridine. The products were biodegradable. Chemical or physical modification is required for successful applications as biodegradable polymers. Unfortunately. The products were hydrophobic and the mechanical properties were considerably improved compared to native starch. Introduction Plastics are very useful materials and are used in large amounts (245 million ton per annum) [1‐2].65) .1. the research activities were mainly limited to the use of vinylacetate.
and disodium hydrogenphosphate dodecahydrate (both Merck.1.2. In addition. and tetrahydrofuran (THF) were supplied by Acros (Belgium). Tc. 5.24 ‐ 2. The catalysts for the esterification reaction were analytical grade potassium carbonate (Boom. pyridine. The spectra were recorded at 50 oC. Analytical grade methanol. vinyl laurate (Fluka. Germany). catalyst type and intake) on the DS of the products. Analytical grade vinyl stearate (Aldrich.2. This chapter describes systematic studies using Design of Experiments (DOE) to gain insights in the effect of process variables (temperature. Japan). and toluene were obtained from Labscan (Ireland). Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. the Netherlands). Technical grade dimethyl sulfoxide (DMSO). each sample was cooled to ‐50°C and then heated again to 200 oC (heating rate 10 oC/min). DSC analyses were performed on a TA Instruments DSC 2920. vinyl ester to anhydroglucose ratio. and acetic anhydride (Merck. as recommended by Laignel et al .96) were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts in DMSO. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch We have recently performed exploratory studies on the synthesis of starch esters of higher fatty acids using vinyl laurate and vinyl stearate as the reagents . T‐Bones samples (with thickness of 2 mm) for determination of the tensile properties were prepared using a melt press 1 87 . It was dried for 48 hour at 105 oC under vacuum (~1 mbar) to reduce the moisture content below 2 % before use. All these chemicals were used as received. The samples were heated to 700 oC in a nitrogen atmosphere with a heating rate of 10°C/min. After a first heating run from room temperature up to 200 oC to delete the thermal history of the material.N‐ dimethylaminopyridine (DMAP).2. Materials and Methods 5. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). TGA measurements were performed using a Perkin Elmer TGA 7 Thermogravimetric Analyzer. 4‐N. Materials Corn starch (approx. Germany) were used without further purification. 5. the thermal and mechanical properties of representative examples of the highly hydrophobic materials are described and discussed. and Tg) of each sample were determined from the spectra related to the cooling run and the second heating one. The thermal properties (Tm. The experiments were modeled using non‐linear multivariable regression. Long fatty esters of corn starch with a broad range in degree of substitution (DS= 0. The samples (about 10 mg) were placed in sealed aluminum cells.2. Germany). sodium acetate.
apparatus (Fontijne, Holland), operated at 150 oC and 150 bar for 3 minutes. The tensile tests were performed using an Instron Series IX Automated Materials Testing System 1.09 at 20oC and a crosshead speed of 30 mm/min. 5.2.3. Methods 188.8.131.52. Typical example of the preparation of laurate and stearate starch esters The corn starch (0.5 g) was gelatinized in DMSO (5 ml) at 70oC for 3 h. After the mixture became a homogenous, colourless solution, vinyl laurate or vinyl stearate (3‐5 mol/mol with respect to the AHG units) and the catalyst (potassium carbonate, or sodium acetate, or disodium hydrogenphosphate, 2‐5%‐w with respect to starch) were added. The reactor content was kept at 80‐110oC for 24 h. The product was precipitated using methanol (100 ml), and the liquid phase was removed by decantation. This precipitation‐decantation procedure was repeated twice using 50 and 25 ml of methanol, respectively, to purify the reaction product. Finally, the product was dried in a vacuum oven (70oC, 5 mbar) for 24 h or until constant weight. 184.108.40.206. Peracetylation procedure and Degree of Substitution (DS) determination The presence of un‐substituted hydroxyl groups in the products resulted in broad and overlapping starch resonances in 1H‐NMR spectra. A peracetylation reaction to substitute all of the remaining hydroxyl groups with acetate groups was therefore applied to obtain reliable DS values. Typically, the starch ester (0.1 g) was added to THF (4%‐w/v) and stirred at 55 oC until dissolution (typically 3 h). Subsequently, the peracetylating reagent (1 mol DMAP, 10 mol acetic anhydride, and 22 mol pyridine per mol AHG units) were added. The peracetylation reaction was conducted for 7 h at 50 oC. The product was precipitated by the addition of methanol and washed several times with methanol before finally dried overnight in a vacuum oven at 70oC and 5 mbar. The DS of the products was calculated using a procedure given in previous work . 5.2.4. Experimental Design The experiments were carried out in 6 blocks. The three variables used in each block were the vinyl ester to AHG starch mol ratio ( x1 ), catalyst intake ( x 2 ), and reaction temperature ( x3 ). All experiments in a block were conducted using a 3‐ variable, 2‐level Full‐Factorial Experimental Design with one center point, giving a total of 23+1 = 9 experiments per block. For each block, a given type of catalyst (Na2HPO4, K2CO3 or Na‐acetate) and vinyl ester (either vinyllaurate or vinylstearate) was applied. This gave a total of 54 experiments (2 types of vinyl
Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch
esters x 3 types of catalysts x 9 experiments per block). In a later stage, the type of catalyst was quantified using the pKb of the catalyst ( x 4 ). The ranges for the individual variables ( x1 ‐ x 4 ) are shown in Table 5.1. The mathematical analysis of the experimental data was performed with the software package “Matchad 13” (Mathsoft). The response (y, DS of the products) was modelled using the following expression:
y = β 0 + ∑ β i xi + ∑ β ii xi2 + ∑∑ β jk x j x k
i i j k≠ j
Here, βi, βii and βjk are the regression coefficients obtained by a multiple 2 regression procedure. One of the terms, namely x 4 , caused singularity of the matrix used for the modelling, and was therefore excluded. A t‐statistic was used to rank the individual regression coefficients (βi, βii, or βjk) according to their relative importance . An analysis of variance (ANOVA) was performed to check the adequacy of the model. The applied procedure is well described in the literature  and consists of calculating the sum of squares (SS) for the model and the error, together with the total sum of squares. In combination with the relative degrees of freedom (DF) it is possible to calculate the mean square (MS) for the model and the corresponding error. On the basis of the latter values, the F‐value for the model is calculated. With this information the P‐value for the model is determined. The latter value is related to the statistical significance of the model.
5.3. Results and Discussion
A schematic representation of the esterification reaction of starch with a vinyl ester is provided in eq. 5.2. The starch was gelatinized before the addition of the vinylester to make the starch OH groups more accessible for reaction. The reactions were carried out in DMSO for 24 h using three different basic salts (Na2HPO4, K2CO3, or Na‐acetate) as the catalysts. The products of the reaction were brownish gels which became lighter in colour after product precipitation and washing with methanol. After drying, the products were isolated as transparent, yellowish solids. The products were soluble in organic solvents such as toluene and THF.
OH O HO OH
R O CH2
O HO O R O
(1) R = laurate (C12) (2) R = stearate (C18)
The products were characterized by 1H‐NMR in CDCl3. The peaks of starch (δ 3‐ 5.5 ppm) and aliphatic hydrogen atoms of the fatty acid chain (δ 0.8‐2.5 ppm) were clearly present. The starch peaks were broad and overlapping, and only after the starch esters were peracetylated, the resolution of these peaks was considerably improved. The DS of the products was determined by using 1H‐NMR spectra of the peracetylated starch esters. The effect of three reaction parameters (molar ratio of vinyl ester to AHG units of starch x1 , catalyst intake x2 , and reaction temperature x3 ) on the DS of starch was studied using a full factorial experimental design with one center point. The ranges of the values of the independent variables ( x1 ‐ x3 ) are shown in Table 5.1. An additional variable x 4 (related to the basicity of the catalyst) was also included in order to obtain a general model for starch esterification. A total of 54 experiments were conducted. Table 5.1. Experimental Design Variables
Independent Variables Full Factorial Variables Vinyl ester to AHG‐ starch ratio [mol/mol] Catalyst intake [%‐w]a Reaction temperature [°C] Additional Variable Catalyst basicity (pKb)b
Low (‐1) Middle (0) High (+1)
2.0 3.5 5.0
80 95 110
K2CO3 (3.66) Na2HPO4 (6.8) CH3COONa (9.25)
3 4 5
in %‐wt based on starch intake, in the model (eq. 5.3.) it is transformed to 105 x catalyst mol amount. taken from ref .
The results of the experiment are given in Table 5.2.
00 5.00 3.00 2.00 5.00 5.00 5.00 5.50 0.00 Catalyst Temperature [%‐w] [˚C] 2.00 3.00 2.84 2.78 2.00 3.54 2.00 5.00 5.00 2.00 2.52 2.28 2.12 2.27 1.00 5.00 3.00 5.00 2.88 2.67 2.00 4.00 3.00 5.00 3.00 3.00 5.47 1.11 2.00 3.23 2.00 3.00 5.00 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 DS 0.00 2.87 1.00 3.00 5.00 5.00 2. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.00 5.00 5.41 1.00 5.00 3.53 1.00 3.46 1. Overview of experimental data Catalyst Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa Vinyl : AHGa [mol/mol] 3.44 2.50 2.00 4.00 2.54 2.00 3.50 2.00 3.94 2.00 5.00 2.00 5.12 0.00 5.24 2.00 5.00 3.00 3.00 3.00 5.27 0.44 2.00 3.56 1.50 2.00 5.79 Experiment Vinyl Ester 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate 91 .96 0.00 5.09 0.00 5.00 2.00 5.52 2.00 5.40 1.35 0.00 3.00 5.00 2.00 5.00 3.27 1.00 5.00 5.00 2.00 5.00 3.00 5.00 5.17 2.93 0.00 5.2.00 5.00 2.75 1.00 5.00 5.32 1.00 4.00 2.72 2.00 3.82 2.00 5.00 4.48 0.07 0.62 0.00 3.59 2.50 2.70 2.00 2.42 2.00 5.
00 5.0017 0.00 3.0011 0.1568 ‐2.24 2.The modelled values for βij for the two esters are given in Table 5.98 2.0016 0.1.0120 0.59 2.00 2.).0023 0. Mathematical Modeling Modelling of the DS data for both vinylesters was performed using non‐linear multivariable regression based on the 4 independent variables (Table 5.00 4.00 5.3.55 2.1082 ‐0.00 5.Chapter 5 45 46 47 48 49 50 51 52 53 54 a.3.1748 0.1104 ‐0.6939 0.00 3.0857 ‐0.64 2.00 3.00 5.50 2.0249 ‐0.) Table 5.00 3.0732 ‐0.50 95 80 80 110 110 80 80 110 110 95 1.0066 0.0327 ‐2.0010 ‐0. stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 4. Values for the coefficients of the DS model for starch laurate and stearate Coefficient Starch Laurate 9.00 3. The type of catalyst was quantified using the basicity constant (pKb) in water [26‐27].21 2.3.).5650 0.1947 β0 β1 β2 β3 β4 β5 β6 β7 β8 β9 β10 β11 β12 β13 92 .64 2.00 5.0009 0.90 2. 5.0006 ‐0.00 2. The experimental data for both starch laurate and starch stearate esterification are best described with a quadratic model including interaction terms (eq.0008 0.2183 0.00 2.0470 0.0008 ‐0.00 5.0003 0.00 3.3.00 5.1979 Starch Stearate 17.3.1.0001 0.0027 ‐0.96 2. DS = β 0 + β1 x1 + β 2 x 2 + β 3 x3 + β 4 x 4 + β 5 x1 x 2 + β 6 x1 x3 + β 7 x1 x 4 + β 8 x 2 x3 + β 9 x 2 x 4 2 2 2 + β10 x3 x 4 + β11 x 2 + β12 x3 + β13 x 4 (5.00 5.41 mol ratio of the vinyl ester to AHG units of starch 5.
Finally.6118 0. the adjusted‐R2 values are very close to the R2 ones.5 2.970 0. respectively and it may be concluded that the models describe the experimental data well. which indicate  that all significant variables are included in the model. The parity plots of both models are given in Figure 5. Parity plots of the DS models for starch laurate and stearate Moreover.5 3 0 0 0.4. Analysis of variance for DS models for starch esterification Starch Laurate Model Error Total SS 16.8167 21. Also in this case the reasonable R2 values (0.026 x 10‐7 R2 values R2 R2adjusted R2press 0. we also performed a PRESS analysis  (see corresponding R2 values in Table 5.5 DS actual 2 2.5 0 0 0.1. The very low P‐ values indicate that the models are statistically significant.2903 P‐value 1.0356 F 35.5 3 a.4.5 1 1.805‐0. which represents an “internal” validation method for the model.805 3 3 2.629 P‐value 1.3399 0.5 DS actual 2 2.4986 16.945 0.96 x 10‐8 R2 values R2 R2adjusted R2press 0.5 2 DS predicted DS predicted 2 1.1.970 (starch laurate) and 0.7699 DF 13 14 27 MS 1. and confirm this statement.5 1. starch stearate Figure 5.8386 DF 13 14 27 MS 1.875 Starch Stearate Model Error Total SS 20.).939 0.4. starch laurate b.5 1 1.5 1 1 0. Table 5.967 (starch stearate).2569 0.875) indicate that the model correctly predict the 93 .967 0.0583 F 27.9531 0.5 0. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch The result of the analysis of variance is given in Table 5. The R2 values for the models are 0.
for both starch laurate and starch stearate are given in Figure 5. K2CO3 is a stronger base (pKb 3. Dicke  showed that Na2HPO4 has a strong tendency to selectively deprotonate the OH group at the C2 position of starch leading to C2 substituted acetate esters. This was not the case for alkaline catalysts. the base strengths of the catalysts in DMSO are not known.18) and CH3COONa (1. The product DS for Na2HPO4 was considerably lower (1. The pKb for the latter (6. A similar trend was observed by Dicke  for the acetylation of Hylon VII starch using vinylacetate.Chapter 5 products DS as function of the process variables within the range of experimental variables.). whereas catalyst performance is considerable lower. The highest product DS values were obtained using K2CO3. the type of catalyst has the largest effect on the product DS for both the laurate and stearate esters (Figure 5. Catalyst performance of CH3COONa was slightly less than for K2CO3 whereas the lowest DS products were produced when using Na2HPO4 as the catalyst. To evaluate the effect of each variable on the product DS. Of all the variables studied. The explanation for the high preference of C2 substitution for the Na2HPO4 catalyst is not yet known and needs to be established by mechanistic studies. Unfortunately. The resulting anion will react with the vinyl ester to from the product. A possible explanation of this peculiar behaviour of Na2HPO4 is the fact that the pKb values in water are used for quantification whereas the actual solvent for the reaction is DMSO. On the basis of the sequence.8) is intermediate between that of the other two catalysts. it can be rationalised that the rate of the reaction (and thus the product DS) will be higher when using a stronger base. This is indeed the case when comparing the performance of K2CO3 with CH3COONa.00) than for K2CO3 (2.82). two plots showing the dependence of the DS on reaction temperature and catalyst amount (in %‐w). It is assumed that the first step in the reaction sequence is activation of the starch OH groups by deprotonation by a base [21. These results may be rationalised by considering the role of the catalyst in the modification reaction. However. the performance of Na2HPO4 does not follow this trend.66) than CH3COONa (pKb 9.2. Another explanation may be related to the regio‐chemistry of the reaction. such as carbonate or acetate salts leading to higher DS esters. 22]. 94 .25) and this leads to higher product DS values for K2CO3.2.
starch stearate Figure 5. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch a. the catalyst intake and reaction temperature also affect the product DS (Figure 5. As expected and in 95 . 3D contour plot of the DS as a function of reaction temperature and catalyst intake (at constant vinylester to AHG mol ratio of 4) Besides the type of catalyst.2. although to a lesser extent.2. starch laurate b.).
higher reaction temperatures and catalyst intakes lead to higher product DS values.45) and high DS values (DSlaurate = 2. DSC analysis of starch stearate esters ( ) DS 1. and the results for all samples are summarised in Table 5. . heating Exothermic Tm = 21 oC cooling Endothermic Tc = 12 oC -20 -10 0 10 Temperature ( C) o 20 30 40 Figure 5. the vinyl ester to AHG ratio has the smallest effect on the product DS.5.) DS ( 2.45 (.39) were determined.75.3. Product Properties The thermal and mechanical product properties for two starch‐laurate and stearate samples with medium (DSlaurate = 1. DSstearate = 1. Of all variables studied. DSstearate = 2.3. Representative DSC spectra for the starch stearate are given in Figure 5. .Chapter 5 line with studies on starch acetates .39) 96 .2.3. 5.26.
a 21 Tc [˚C] n.a 27 n. these are the first examples of completely amorphous starch esters. esterification of starch with vinyl laurate and stearate reduces the crystallinity of the starch considerably and leads to a lowering of the transition temperatures.39 1.8 . Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. Melting point (Tm) and crystallization temperature (Tc) of representative starch laurate and stearate samples Starch ester Starch laurate Starch laurate Starch stearate Starch stearate DS 2.). For instance a Tm of 32 oC was reported for a potato starch stearate ester with a DS of 1. This clearly indicates that the observed thermal transitions are not only determined by the side chains but are an interplay between that of the starch backbone and ester side chains. the Tm for the medium DS starch laurate (27 oC) is much higher than the corresponding side chain model compounds (‐10 oC for dodecane and 5 oC for methyl laurate). These values are reduced considerably compared to virgin corn starch. implying that these products are fully amorphous. close to the value obtained in our study for the medium DS starch stearate (21 °C). which is known to contain crystalline areas. TGA analysis was performed to study the thermal degradation behavior of the starch laurate and stearate samples (Figure 5. 22].= not detectable The thermal behavior of the products is a strong function of the DS.d. in the present work. However.26 1.5. or glass transition above 0 oC) were even absent for the high DS products.a 12 a.d. n. Changes in the transition temperatures of native starch by esterification have been reported in the literature [13‐14. 28‐30 oC) and the authors concluded that the transition temperatures are solely determined by the side chain without any significant contribution of the starch backbone.d. crystallization. To the best of our knowledge.75 2. At moderate DS values the products still display a melting temperature (at 20‐30 oC) and a crystallization temperature (at 10‐25 oC). The melting point of the potato starch stearate was close to that of model compounds for the stearate side chains (methyl stearate.4. Unfortunately the thermal transitions for the virgin material lie above the degradation temperature and can therefore not be accurately determined .d. In any case. especially in the amylopectin part [15. 17].45 Tm [˚C] n. In particular the behaviour at relatively high temperatures (>0 oC) is further discussed. octadecane. 40‐42 oC. 97 .d.a 22 n. Thermal transitions at relatively high temperature (melting.
( ): native starch ( ): starch laurate (DS 2.26) (. The mechanical properties (stress at break.4. elongation at break and elasticity modulus) of the products were determined (Table 5.) curves clearly show starch esterification with either laurate or stearate results in products with enhanced thermal stability. . .).): starch stearate (DS 2.Chapter 5 100 80 %-weight 60 40 20 0 20 100 200 300 400 o 500 600 700 Temperature ( C) Figure 5.4. 98 . TGA analysis of various starch samples. Native corn starch degrades at lower onset temperatures (250‐300 oC) than the corresponding esters (300‐350 oC).6. The difference in thermal stability between the laurate and stearate sample is limited.39) The TGA (Figure 5. Similar trends were reported for amylose octanoate‐ stearate‐laurate esters  and potato starch octanoate‐ laurate esters .
A deeper understanding of the mechanical behavior can be gained by investigating the shape of the stress‐strain curves (typical examples in Figure 5.45 Stress at break [MPa] 3. and lower elongation at break).6.1 Modulus [MPa] 59.1 ± 1. elongation at break between 3‐26%. plastic deformation and yield strain of about 0. These considerations strongly point out the fact that the mechanical behavior can be coarsely (plastic vs elastic behavior) but also finely (stress and strain at break but also modulus values) tuned by the chemical structure of the fatty acid chains as well as the DS values.3 21. and modulus of elasticity between 46‐113 MPa. Starch stearate with a relatively low DS (1. The corresponding yield strain (roughly 0.9 46. An inverse behaviour is. thus indicating a relatively more rigid material with respect to all other samples.1 82.1 ± 15.025) lies still in the typical range of engineering polymers but also close to that typical of composite materials .7 ± 4. The values of the yield strain. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.1 Elongation at break [%] 26. an increase in product DS results in lower tensile strengths and modulus of elasticity.2 ± 0. Mechanical properties of the starch esters DS 2.9 ± 14.). For starch laurate.7 ± 4. 17].5 MPa.7 ± 3. a clear plastic behavior with a maximum in the stress‐strain curves.26 1.7 ± 1. starch stearate at relatively high DS values (2.8 Starch Ester Starch laurate Starch laurate Starch stearate Starch stearate All materials generally show tensile strengths between 2. This is in agreement with previous studies regarding the mechanical properties of starch esters with different chain length of the fatty acid moiety [13. This indicates that their mechanical behavior is comparable to that of typical engineering polymers .7‐3. however. On the other hand. This product is more rigid compared to the product with medium DS value (higher tensile strength and elasticity modulus.0 112.2 2.75 2. calculated as the ratio between the stress at yield (σy.5. but higher elongation at break. Both starch laurate samples display.3 ± 2. observed for the high DS starch stearate sample.5 ± 1.7 ± 14.0 3. determined from the stress‐strain curves as the stress at which the material ceases to be linearly elastic) and the modulus.0 3.e.45) displays a very similar behavior with respect to the laurate samples (i.39) does not show any plastic behavior but only an elastic one.3 3.2 8.0 ± 2.05). independently of the DS values. 99 .05 for both laurate samples.39 1. is roughly 0.
26) (. ‐): starch stearate (DS 2.39).5. 100 .7.75) ( ): starch laurate (DS 2.Chapter 5 6 5 4 stress (MPa) 3 2 1 0 0 5 10 15 20 strain (%) 25 30 35 Figure 5. Stress‐strain curves for starch esters.): starch stearate (DS 1.45) (‐ . Although there are many differences in the synthetics methods as well as in testing conditions and procedures.). . a rough comparison can be made between the mechanical properties of the esters described in this chapter and those for reported for related starch esters synthesized with alkanoyl chloride as reagent and pyridine as catalyst  (Table 5. . ( ): starch laurate (DS 1.
3 Potato Starch Stearate Ester Low DS (1.a. DS) which mainly determines the thermal and mechanical behaviour.6 9 ±2 High DS (2. an unexpected behavior was also observed for the mechanical properties of the starch esters from potato starch.4 1500 ±8. for potato starch exactly the opposite is observed: relatively low stress (0. They related this inverse property of starch stearate to crystallization of C18 side chains. In this respect the difference in stress and elongation at break for the high‐DS starch laurate sample is striking.5 MPa) is coupled with a low elongation (26 %).7 ±3.7) 0.6 Corn Starch Stearate Ester Low DS (1.2 ±0.7 MPa for potato starch ones) are coupled with significant differences in the elongation (22% for corn starch esters. n. confirms that the corn starch esters synthesized in our research are relatively rigid materials as compared to the potato starch esters. On the other hand at low DS values comparable stress at break (3.7) 1. 101 . merely because of the structural property of the C18 side chain with respect to the starch backbone. we show that although no crystallization of C18 occurred (see DSC data of the high DS material in Table 5.26) relatively high stress at break (3. In agreement with our own data. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. the high DS material is still more rigid.7 ±0.).0 ±2.7 MPa) coupled with a significant (1500%) elongation. Comparison with the literature data implies that the mechanical behaviour of starch esters of higher fatty acids is also a clear function of the type of starch used. 3. n.5 ±1.7 ±1.a.0 3.5.= not available Inspection of the mechanical properties of the starch esters as shown in Table 5.7.2 MPa for corn starch esters. Here.a.9 ±0.7 ±4. The same considerations are valid for starch stearate at high DS values (higher stress and lower elongation when using corn starch instead of potato one).a.7.2 Stress at Break [MPa] Elongation at Break [%] a.39) 2.45) 3. High DS (2.a. 9% for potato starch).3 High DS (2.0 26.8) 3. as confirmed by DSC.75) 3. this comparison also confirms our conclusion derived from the analysis of thermal properties (vide supra): it is not only the individual factors (kind of starch and fatty acid.3 10 ±2 Property Low DS (1.1 21.7 ±0. but the interplay between those factors. Their medium or high DS stearate esters of potato starch are more rigid compared to the starch esters from lower chain fatty acids (octanoate and laurate) .1 ±15. While for corn starch (DS 2.2 8.1 ±1.1 High DS (2. Comparison of the mechanical properties of corn starch ester (this study) with potato starch ester  Corn Starch Laurate Ester Potato Starch Laurate Ester Low DS n.26) 3. Moreover.
The use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2. The DS of the products is a strong function of the basicity of the catalyst. the chemical composition of the starting materials (either starch or the ester precursor) as well as the processing parameters affect the final product DS values and as such provide an effective toolbox to modulate the desired product properties for a given application.3‐1.5 for starch stearate). average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] 102 . and Na‐acetate). 0. 5.9 for starch laurate. Conclusions A systematic study. K2CO3.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐starch mol ratio. The models may be used to determine the appropriate process conditions to obtain a product with a pre‐defined DS.4.07‐1. Thermal and mechanical behaviour of the samples of different DS values clearly shows that the DS represents one of the crucial factors affecting the final product properties. Comparison with literature data indicates that the properties may be (fine) tuned also by the starch source. depend on the unit of related variable x] : constant regression coefficients obtained by a multiple regression procedure [‐] : regression coefficients of variable xi obtained by a multiple regression procedure [various units] : regression coefficients of quadratic variable xi2 obtained by a multiple regression procedure [various units] : regression coefficients of interaction variable xjxk obtained by a multiple regression procedure [various units] DS β0 βi β ii β jk : Degree of Substitution. Reaction temperature and catalyst intake also affect the product DS although to a lesser extent than the type of catalyst. 1. catalyst intake. reaction temperature.1‐2. on the synthesis of corn starch esters with long chain fatty acids is described. Nomenclature β : regression coefficients obtained by a multiple regression procedure [various units. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4.5 for starch laurate. while the use of Na2HPO4 resulted in low‐medium DS products (0. Thus. and catalyst basicity) to the DS of the starch ester products were developed. Statistically adequate (R2 ≥0.4‐3 for starch stearate).Chapter 5 5. including statistical modelling.5.
starch. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch DS fatty ester : DS of fatty ester (laurate or stearate) group substituents [‐] pKb T : base dissociation constant [‐] : temperature [oC] : crystallization temperature [oC] : melting temperature [oC] : experimental variables used for mathematical modelling using multiple regression procedure [various units] Tc Tm xi Greek symbols: σy : stress at yield [MPa] 5. E. G. 19. Stevens: Green Plastics. Properties. USA. Ind. The compelling facts about plastics: An analysis of plastics production. in Packaging in the Environment (Ed. Otey. Brussels. Hanser Publishers. M. M. W. Prod. 2002. Marsden: The challenge of domestic waste disposal. C. Culter: Plastics Packaging.dk/isi/stat/rawmaterial. Dev. S. Denmark: http://www. Doane: Starch‐based blown films. Germany. Munich. An Introduction to the New Science of Biodegradable Plastics. Conference Proceedings. Selke. R. S. Structure – Properties ‐ Applications.W. 2000. . 2008. Frederiksberg C. G. The Royal Veterinary and Agricultural University. A. Plastics Europe.J. 592‐595. Hernandez. demand and recovery for 2006 in Europe.6. . Association of Plastics Manufacturers. R. Weber). Hanser Publishers. Princeton University Press. . Department of Dairy and Food Science. 2000. 1980. Belgium. Glasgow. Packaging.html. References . . 2008 . Applications. Aarhus. Bastioli: Global Status of the Production of Biobased Packaging Materials. Chem. F. Denmark. UK. Westhoff. Website of International Starch Institute. Res. H. 27‐29 August 2000 (Ed. and Biopolymers. Levy). 103 . Munich. accessed on August 01. M. Processing. Blackie Academic & Professional. . and Regulations. New Jersey. Denmark. 1986. Ehrenstein: Polymeric Materials. Germany. 2000. P. Copenhagen. in The Food Biopack Conference: Foodstuffs. D. C. E. . J. J. Eng.
. . Thiebaud. E. . P. Beenackers: The kinetics of the acetylation of gelatinised potato starch. Hamaili. and biodegradability of fatty‐acid esters of amylose and starch. E. Aburto. D. I. S. 145‐152. . 35. 51. Aburto. L. the Royal Veterinary and Agricultural University. R. Polym. I. Ind. Thesis. K. 132‐135. 1999. 101‐112. C. Eng. Janssen. Sci. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. D. H. de Graaf. Senocq. status and perspectives. M. Aburto. J. D. de Graaf: Ph. C. . E. Appl. C43. Alric. J. Polym. C. F.D. 1997. Aburto. A. 56. 2003. E. Macromol. Pacsu: Starch studies: possible industrial utilization of starch esters. . A. C. M. Appl. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. X. . J. 58. E. J. Aburto. Panayiotou: Properties of octanoated starch and its blends with polyethylene. Borredon. 34. Appl. Chem. 74. Prinos. Merrill: Properties of fatty‐acid esters of starch. L. Panayiotou: Synthesis. E. Alric. 302‐307. Mullen. Mouysset‐Baziard. G. Carbohydr. report of Food Biopack Project (Ed. 34. J. S. Part C: Polym. Chem. 28. J. 2000. C. . Sci. Eng. 385– 409. J. Pacsu: Starch studies: preparation and properties of starch triesters. Mormann. 1647‐1656. J. 1995. A. D. A. Al‐Higari: Acylation of starch with vinyl acetate in water. K. Thiebaud. I. Wang. 1997. Z. Polym. Rev. Alric. Sci. E. J. Alric. Borredon. . J. J. 1942. 118‐121. Starch‐Starke 1999. Rijksuniversiteit Groningen. 705‐721. E. Weber). . Department of Dairy and Food Science. Sagar. Thiebaud. J. . Carbohydr. R. 1995. 1996 104 . Borredon. 65. Polym. Polym. Starch‐Starke 2004.. I. 1209‐1217. characterization. Frederiksberg C. J. 51. Starch‐Starke 1999. M. 381‐384. B. A. Bikiaris. Denmark. J. I. 1943. . 1440‐1451. .Chapter 5 . A. Bikiaris. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. G. Alric. Biobased packaging materials for the food industry.. Prinos. J. 137‐144. E. Broekroelofs. W. Sci. Y. Alric. Aburto. I. Wang: Properties of starch blends with biodegradable polymers. W. Mullen. Bikiaris. Ind. W. Starch‐ Starke 2005. J. 57. Yang. W. S. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Prinos.
New York. the Netherlands. Junistia. Shercliff. 2007. H. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Butterworth‐Heinemann. R. Dicke. . . Underwood: Quantitative Analysis 5th edition. USA. USA. A straight way to regioselectively functionalized polysaccharide esters. D. A. B. K. Carbohydr. 2008. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch . Processing and Design. C. Cebon. accepted for publication in Starch‐Starke. Res. Picchioni. F. Montgomery: Design and Analysis of Experiments 5th edition. Massiot. G. Sugih. 105 . M. L. Prentice‐Hall. 1997. 298. J. Bliard. Laignel. Day. Janssen. A. H. Materials: Engineering. New Jersey. Ashby. A. P. 255‐263. 251‐260. M. Science. D. . R. J. C. John Wiley & Sons Inc. 11. L. . A. M. R. B. Amsterdam. 1986.. . 2001. Cellulose 2004. Manurung. L.
Keywords: starch‐PCL blend.Chapter 6 Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends Abstract The synthesis of two reactive interfacial agents for starch‐PCL blends. In particular the elastic modulus for the compatibilized blends is significantly higher than that of uncompatibilized ones. compatibilizer. The effect of important process variables (intiator and monomer intakes as well as estimated solubility of the monomer in molten PCL) on the degree of grafting (FD) of the GMA and DEM units to the PCL backbone was explored in detail and quantified using multivariable linear regression. diethyl maleate. PCL‐g‐DEM has at least a comparable performance with respect to PCL‐g‐GMA. despite the expected differences (favorable to PCL‐g‐GMA) in the in situ formation of the compatibilizers. glycidyl methacrylate. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. At relatively low starch intakes. at relatively high GMA and BPO intakes. The reactive interfacial agents were tested for their performance in PCL‐starch blends. The compounds were prepared by reacting a low molecular weight PCL (Mw 3000) with glycidyl methacrylate or diethyl maleate in the presence of benzoylperoxide at 130°C. the mechanical behavior of the compatibilized blends is in general different from that of pure PCL and of the corresponding uncompatibilized blends. grafting . The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). This discrepancy could be explained on the basis of the functional groups (GMA or DEM) distribution along the PCL backbone. As a result. Both act as compatibilizers for PCL/starch blends by improving the interfacial adhesion between the starch particles and the PCL matrix. PCL‐g‐ glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM) is decribed.
When using ungelatinized starch as a component in the blend. The ester bonds are susceptible to attack by water and this leads to enhanced biodegradability. A well‐known biodegradable polyester. polyesters are considered very promising alternatives . blends of hydrophilic starch and hydrophobic biodegradable polyesters exhibit phase separation  due to differences in polarity of the building blocks. This leads. polycaprolactone (PCL). synthetic biodegradable polymers have been applied. The latter displays the presence. of polar groups (usually epoxides or anhydrides [13‐18]) able to react with the – OH groups along the starch backbone. Among these. 108 . these synthetic polymers are poorly or non‐biodegradable. This phenomenon is highly undesirable and limits the application range considerably . Unfortunately.e. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decrease rapidly in time . Blending of starch and synthetic biodegradable polyesters has been widely applied for the synthesis of fully‐biodegradable products. upon mixing) by using a functionalized PCL. To overcome this issue. for starch/PCL blends. Unfortunately. an interfacial agent) may be used to improve the interfacial association between the two polymer phases. Such structure is rather difficult to achieve by simple copolymerization methods and it is usually prepared in situ (i. However. the native material is very hydrophilic and important mechanical properties are inferior compared to most synthetic polymers and this hampers its direct use as packaging materials. One of the well known modification strategies is blending the starch with polymers displaying a stronger hydrophobic character and better mechanical properties. 2]. To reduce the tendency for phase separation. is known to be degraded with ease by microorganisms widely distributed in nature . In general a compatibilizer is a block‐copolymer where each block displays a chemical structure equal or very similar to that of the polymers to be mixed. along the backbone. As a consequence. Introduction Starch is a cheap and abundantly available natural polymer with very good application perspectives in the area of biodegradable plastics.Chapter 6 6. as in this study. it would be actually more accurate to define the block copolymer as an “interfacial agent”. PCL has gained considerable interest for possible applications in the fields of packaging materials and medical applications [10‐11].e. a compatibilizer (i. such as polyethylene (PE) or polystyrene (PS) [3‐6]. It must be stressed here that the word “compatibilizer” is correctly used only when the block copolymer is actually able to significantly influence the dispersion of the polymers to be mixed (most probably through a steric stabilization mechanism ). Starch modifications to improve the product properties like enhanced hydrophobicity and mechanical properties were already reported in the early 19th century [1. to an ideal compatibilizer having both PCL and starch blocks.1.
Tetrahydrofuran (THF. The effect of important process variables on the degree of grafting of the GMA and DEM units to the PCL backbone has been explored in detail and quantified using multivariable linear regression. The materials were mixed for another 5 minutes. Germany) were used as received. a solution of BPO in GMA or DEM was added drop by drop over a period of 5 minutes. xylene (99.8%) from Labscan. 6.2.1. Materials and Methods 6. Methods 6. Intakes for each experiment are given in Table 1 and 2. After the PCL was melted (1‐2 minutes). Compatibilizer synthesis The compatibilizers were prepared in a Brabender Plasticorder PL2000 batch‐ kneader (chamber volume 35 cm3). UK. Belgium.2. The intake of reagents was maximally 75‐80% of the volumetric volume to ensure proper mixing. systematic studies and subsequent quantification of the functionalization reaction has not been reported to date.1. Corn starch (with approx.2. to the best of our knowledge. Mw=3000 from Solvay Caprolactones.8%) from Merck. and benzoyl peroxide 75% (Merck. 6.2. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). The starch was dried for at least 24 h at 110oC under vacuum (approx. an absolute novelty of the present research. Exploratory studies on the synthesis of PCL‐g‐GMA and its applications for starch/PCL blends have been published [13. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends which is able to mainly improve the interfacial adhesion between the polymer and starch itself. after which the equipment was stopped and the chamber was opened to collect the samples.2. UK was used for the preparation of the interfacial agents. The kneader was heated to 130 oC and PCL (CAPA 2304) was added while maintaining a rotational speed of 80 rpm. 109 . This low molecular weight PCL grade was used without further purification. Mw=50000) from Solvay Caprolactone. diethyl maleate ≥97% purity (Fluka). The synthesis and application of PCL‐g‐DEM is. This chapter describes a systematic study on the synthesis of two reactive interfacial agents for starch‐PCL systems. >99%) was obtained from Acros. Materials Polycaprolactone (PCL) CAPA 2304. Glycidyl methacrylate 97% purity (Aldrich). however. Germany and methanol (99. 16]. The various compatibilizers have been tested for their performance in PCL‐starch blends. Ireland. ~1 mbar) prior to use.2. 73% amylopectin and 27% amylase) was obtained from Sigma and high molecular weight PCL (CAPA 6503.
Aldrich) as the solvent. and then filtered. The content was blended for 15 minutes. Analytical Methods H‐NMR measurements were performed using a 400 MHz Varian AMX Oxford NMR apparatus with CDCl3 (99. Methanol (450 mL) was added to the filtrate. Work‐up of PCL‐g‐GMA products  To remove unreacted GMA monomer and GMA homopolymer. 6. Scanning Electron Microscopy (SEM) was performed using a Jeol 6320 F Scanning Electron Microscope.5 h. PCL was added to the chamber followed by the addition of the starch and the reactive compatibilizer. and the product was precipitated at 6‐8 oC.2. 6.8%.Chapter 6 6. Each sample was first heated from 0 oC to 100 oC (heating rate 10 oC/min) to remove the thermal history of the material. created using a Cressington 208 sputter coater. 6. Subsequently. Preparation of PCL‐starch blends with the reactive compatibilisers The PCL‐starch blends were prepared in a Brabender Plasticorder PL2000 batchkneader (chamber volume 35 cm3).2. 5 mbar).2. stirred for 1.2. An operation temperature of 170°C and a rotation speed of 80 rpm were applied .4. Work‐up of PCL‐g‐DEM products Purification of the PCL‐g‐DEM product was performed according to a modified procedure for PCL‐g‐maleic anhydride .2.2. The resulting suspension was filtered and precipitated using methanol (450 mL) at 6‐8°C.2. the chamber was opened and the resulting material was collected. PCL‐g‐GMA (5 g) was dissolved in 50 ml THF. The transition temperatures of each sample were further determined by first cooling down the samples from 100 oC to 0 oC and subsequently heating up back to 100 oC (cooling and heating rate were 10 oC/min). The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. 5 mbar). Infrared spectra were collected with a FT‐IR apparatus in the ATR mode using a Spectrum 2000 instrument from Perkin Elmer. Before analysis.3.2.3. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. The error on the transition temperature is assumed to be ± 1 oC and 5 % of the calculated values for the corresponding enthalpies. Tensile tests were performed using an Instron 4301 1 110 . PCL‐g‐DEM (5 g) was dissolved in xylene and refluxed at 150 oC for 2 h. Digital Scanning Calorimetry (DSC) measurements were performed using a Q1000 TA Instruments equipped with a TA Instruments DSC cooling system. the samples were covered with a paladium/platinum conductive layer of 3 µm thickness.
2. The FD is defined as: FD = number of moles of GMA/DEM attached to PCL (mol) × 100% number of repeating units of the PCL backbone (mol) (6. leading to a 10% relative error in the FD values. 6. it is possible to calculate the mean square (MS) for the model and the error. 111 . The latter is a measure of the statistical significance of the model.4.) The FD was calculated using 1H‐NMR by comparing the area of protons belonging to the GMA (‐CH< proton at δ 3. The validity of the model was determined by performing an analysis of variance (ANOVA. 19]. 15‐16.0 ppm [13. Statistical Modeling The influence of different processing parameters on the FD values has been determined by performing a multivariable regression procedure on the available data.3. A 5% relative error in the peak area of the NMR spectra was assumed.2 ppm) side chains with that of a characteristic proton resonance of the PCL backbone (‐CH2‐ protons at δ 4. This procedure is described in detail in the literature  and consists of calculating the sum of squares (SS) for the model and the error. the F‐ value for the model is determined followed by the P‐value. For a given sample/blend. stress at break (σ) and modulus (E) were measured. The T‐bone samples were prepared using a Fontijne Holland TH 400 hot‐ press. When the relative degrees of freedom (DF) are known.1.1. Table 3). The corresponding value for every blend was calculated as an average of the 8 measurements while the standard deviation was taken as absolute error on the average values. On the basis of the latter values.2.2 ppm) or DEM (‐CH2‐ protons at δ 4. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends machine. 6. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers The number of moles of GMA or DEM present on the PCL backbone was quantified using the degree of functionalisation (FD). For every T‐bone. As a result we were able to obtain a model for the FD of the reaction. strain at break (ε). 8 different T‐bones were used.
3.Chapter 6 6.1. Preparation of the Reactive Compatibilizers Twelve compatibilizers were prepared by either reacting glycidyl methacrylate (GMA) or diethyl maleate (DEM) with low molecular PCL as the starting polymer and benzoyl peroxide (BPO) as the radical initiator (Scheme 6. 112 .3.1. Results and Discussions 6. COOEt COOEt * O O n DEM BPO n * PCL-g-DEM * O O * GMA BPO * O O O O O n * PCL-g-GMA Scheme 6.). Functionalization reactions (only showing reactivity for the >CH2 in α position on PCL backbone) Typical 1H‐NMR spectra for the products are shown in Figure 6.1.1.
Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends
Figure 6.1. Typical 1H‐NMR spectra of PCL‐g‐GMA (top) and PCL‐g‐DEM (bottom) Peak assignments were based on available data reported for related products [13, 15‐16, 20]. The FD values and the thermal properties of the products are shown in Table 6.1. (PCL‐g‐GMA) and Table 6.2. (PCL‐g‐DEM).
Overview of experiments for the PCL‐g‐GMA compatibilisers a
Intake FD (%) ‐ 5.7 15.2 27.9 3.2 45.2 c 6.3 T cryst. ( C)
(%‐mol) b GMA BPO ‐ 0.6 0.6 0.6 0.3 1.1 0.3
ΔH cryst. (J/ g‐PCL) 75 72 68 67 73 66 71
T melt. ( C)
ΔH melt. (J/ g‐PCL) 79 72 69 68 73 68 67
PCL PCL‐g‐GMA 1 PCL‐g‐GMA 2 PCL‐g‐GMA 3 PCL‐g‐GMA 4 PCL‐g‐GMA 5 PCL‐g‐GMA 6
‐ 12 24 36 6 24 12
27 22 19 29 25 20 28
51 47 47 48 47 46 47
Experiments were carried out at 130 °C. %‐mol with respect to the CL repeating units in the PCL c based on the soluble fraction of the compatibilizer
Table 6.2. Overview of experiments for the PCL‐g‐DEM compatibilisers a
Intake Sample (%‐mol) b DEM PCL PCL‐g‐DEM 1 PCL‐g‐DEM 2 PCL‐g‐DEM 3 PCL‐g‐DEM 4 PCL‐g‐DEM 5 PCL‐g‐DEM 6
FD (%) ‐ 3.5 2.1 7.2 3.6 6.5 0.9
T cryst. ( C)
ΔH cryst. (J/ g‐PCL) 75 70 68 64 61 63 70
T melt. ( C)
ΔH melt. (J/ g‐PCL) 79 72 62 66 60 66 69
BPO ‐ 0.6 0.6 0.6 1.1 1.1 0.3 27 27 30 21 26 25 30 51 44 47 45 42 41 47
‐ 15 30 45 30 60 7.5
Experiments were carried out at 130 °C %‐mol with respect to the CL repeating units in the PCL
One of the PCL‐g‐GMA compatibilizers (PCL‐g‐GMA 5, see Table 6.1.), was only partly soluble in CDCl3, and therefore the FD is based on the soluble fraction of the compatibilizer only. The presence of an insoluble fraction, combined with a relatively broad molecular weight distribution (as shown by GPC, but not shown here for brevity), suggests that cross‐linking occurred during this experiment. In general, the FD of the PCL‐g‐GMA (3.2‐45.2%) products is much higher than those of PCL‐g‐DEM (0.9‐7.2%). This difference may be either due to the difference in mutual solubility of the GMA and DEM in PCL or differences in the molecular
Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends
mechanism of the grafting reaction. The mutual solubility may be expressed in terms of the differences of solubility parameters of PCL and the reagents . The difference between the solubility parameters of GMA and PCL is 0.29 cal1/2cm‐3/2, while it is much higher (6.3 cal1/2cm‐3/2) for DEM and PCL . Thus, GMA is likely better soluble in PCL, leading to higher values for FD of the products, as confirmed by our experiments. However, the higher experimental grafting efficiencies for PCL‐GMA may also be rationalised by considering the reaction mechanism for the preparation of the compatibilisers. GMA molecules may either react directly with a radical at the PCL backbone or with a radical present on an already coupled GMA molecule. The latter leads to longer GMA grafts on a PCL backbone . A simplified representation of the reactivity of GMA is shown in Figure 6.2.
O (CH2)4 CH2 C O + RO
O (CH2)4 CH
+ GMA O (CH2)4 CH H2C H3C C
C O CH2 HC O O
Termination + H or + R or + RO (CH2)4
+ k GMA O CH C O
H2C H3C C
C O CH2 HC O O
Figure 6.2. Simplified scheme of the GMA Grafting Reaction Mechanism  The reactivity of DEM in radical reactions is expected to be different from that of GMA. Previous studies on maleic anhydride (MA), a compound resembling the
Thus an increase in the GMA intake will provide more monomer available for the growing of the grafted chains leading to higher FD values.Chapter 6 chemical structure of DEM.1.3. BPO=1.6% mol/mol-CL units DEM. BPO=0. However. BPO=0.3.1% mol/mol-CL units 10 5 0 10 20 30 40 50 60 Substrate Intake (mol% on CL units) Figure 6. subsequent reactions of MA to an already grafted MA molecule hardly occur. 116 . Our experimental findings. 22]. i. Effect of GMA and DEM to CL‐units ratio (mol/mol) on the FD of the products (constant PCL intake. BPO=0.e. the length of a MA graft is always unity whereas longer grafts are possible for GMA.1. 30 25 20 FD (%) GMA. This trend is independent of the BPO amount and matches with data reported by other groups [13.3% mol/mol-CL units DEM. 6. Effect of substrate (GMA/DEM) to PCL ratio on the FD The effect of the substrate (GMA or DEM) to PCL ratio on the FD was studied by changing the GMA/DEM intake at different BPO amounts. higher FD values for GMA than for DEM are in line with this explanation and support the proposed molecular reaction mechanisms. showed that MA reacts easily with a radical on the PCL backbone. It is most probably related to the fact that GMA is able to propagate to longer grafted chains when reacted with PCL (Figure 6.3. 130°C) Higher GMA intakes lead to higher FD values.2.6% mol/mol-CL units 15 GMA. Hence.). The results are graphically provided in Figure 6.
3. However. The use of higher initiator concentrations will result in an increase in the number of formed radicals. Effect of the BPO intake on the product FD The effect of the BPO intake on the FD was studied by using different intakes of BPO (Figure 6. Such behavior is slightly in contrast to what observed for maleic anhydride [13. 17]. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The experimental trend for DEM is different.3 0.5 0. Effect of the BPO intake on the product FD (130°C.1. These results are in line with earlier work [15. 50 45 40 35 30 FD (%) 25 20 15 10 5 0 0. GMA=12% mol/mol-CL units DEM.4 0. The FD values are within a rather narrow range.2 BPO Intake (mol% on CL units) Figure 6. doubling the amount of initiator results in considerable higher product FD. 17].9 1 1. constant substrate to PCL ratio) For high GMA to CL ratio (24%‐mol/ mol CL units). 15. This will lead to a higher proportion of PCL radicals by hydrogen abstraction from the polymer backbone and thus to higher FD values. DEM=30% mol/mol-CL units 0.2.7 0. for which a levelling off and eventually a decrease of the FD values for relatively higher MA amounts has been observed.6 0. 6.).1 1. although a slight increase in the FD values might be appreciated. Such discrepancy is probably related to the relatively low intake of DEM as well as to the different reactivity of DEM compared to MA . there 117 .4. at relatively lower GMA intakes.4.). GMA=24% mol/mol-CL units GMA.4. no detectable influence of the BPO amount on the FD is observed (Figure 6. Apparently.2 GMA.8 0.
1. ni . the mutual solubility of DEM in PCL and DMA in PCL was included in the model by using a parameter δ.Chapter 6 is an optimum ratio between the BPO concentration and available monomer (GMA in this case) on the FD. If many macroradicals are created in the system (relatively high BPO intakes) at low GMA intake. Here. these investigations focused on a better understanding of the individual variables by studying the effect of higher peroxide and monomer intakes while keeping for example their ratio constant. The model provided in eq.6022nm ni − 0. A doubling of the initiator intake for PCL‐g‐DEM only results in a 70% increase in the FD (compared to 300% for PCL‐g‐GMA). δ ) (6. and 6. Modeling of the combined effects of the GMA/ DEM and BPO intakes on the FD Previous studies showed the importance of the initiator to monomer ratio on the FD values [15.3.2. The effect is however much less pronounced than for GMA. This leads to the following equation: FD = f (nm . ni the molar amount of initiator in the feed and δ the difference in solubility parameters calculated using group contributions . the BPO and monomer intakes are considered as independent parameters. as well as several transfer reactions might be responsible for the observed trend . namely an increase in the BPO intake results in higher product FD. However. The results described in the previous paragraph imply that the mutual interaction between these variables and not the absolute value itself determines the final FD values to a great extent. 6.3. This phenomenon is likely related to the different nature of the radical grafting mechanism of GMA and DEM on PCL as mentioned before.) 118 . The data for PCL‐g‐DEM shows a similar trend as for the experiments with a high GMA to CL ratio (24%‐mol/ mol CL units).8875 + 0.0325nmδ + 0. a statistical model has been developed by performing a multivariable linear regression on the data reported in Table 6.5431niδ + 1. 17]. In particular the occurrence of “cage effects”.2. the possibility of side reactions will become significant.3.2729nm niδ (6.e. In addition. 6. recombination of (macro) radicals. gives the best description of the experimental data: FD = −1. i.3.1. To quantify synergic effects of monomer and initiator molar intakes on the FD of GMA and DEM on the PCL backbone. defined as the difference in solubility parameters between PCl and the substrates.) where nm is the molar amount of monomer in the feed.
4 BPO Intake (mol% on CL units) 0.3 10 10 20 30 20 0.5. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The analysis of variance data are given in Table 6. 119 .4 0. and Figure 6.358 DF 4 8 12 MS 456. Figure 6. SS 1828 81.957) and its closeness to the adjusted R2 (0. a.6 20 10 20 40 50 30 60 FD (%) 40 FD (%) 30 40 20 0 1 0.8 0.. Table 6. Contour plot. b.17 F 44. Model Error Total Analysis of variance for the FD model provided in eq 6.5.941) also suggests that all important variables have been included in the model.8 0.9 60 70 40 50 80 BPO Intake (mol% on CL units) 0.3.7 0.358 1909.6 0.3.908 10.2 10 20 GMA Intake (mol% on CL units) 40 30 50 0.3. Graphical representation of the FD models for PCL‐g‐GMA and PCL‐g‐DEM are given in Figure 6. The R2 value for the model (0. This was also confirmed by inspection of the residue distribution by a normal probability plot (not reported here for brevity) .928 P‐value <10‐9 The value of the coefficients in the model imply that the FD is positively influenced by the interaction between monomer and initiator intake ( nm ni ). 1 0. b. The very low P‐value implies that the model is statistically significant. the interaction between (PCL‐monomer) intake and solubility parameter difference ( nmδ ) and the interaction between the latter factor and the initiator intake ( ni δ ).2 10 15 20 25 30 35 40 GMA Intakes (mol% on CL units) 45 50 a. Graphical representation of the FD model for PCL‐g‐GMA.6. respectively.5 0. 3D plot.
Contour plot. Graphical representation of the FD model for PCL‐g‐DEM. the crystallization temperature and enthalpy are not a clear function of the FD values. although both properties are significantly lower than those of pure PCL. The random behavior of the Tc as function of the FD values is probably the result of two concurring effects: the presence of irregularities.2.6 0.) of the compatibilizers were determined by DSC. and 6.4 BPO Intake (mol% on CL units) 0.9 4 5 8 BPO Intake (mol% on CL units) 0.1. From a more practical point of view. radical initiator and PCL intakes).2 10 15 20 25 30 35 40 DEM Intake (mol% on CL units) 45 7 a. 7 6 3 3 4 Figure 6.8 0. Furthermore.3.5 0. For all samples. b. b. Similar observations have also been made by Kim. the statistical model provides besides a reliable description of the experimental data also a good mathematical description to modulate the FD values of the two substrates by changing the chemical composition (monomer.7 3 4 6 FD (%) 4 0. Such changes in the thermal properties compared to pure PCL may be caused by the introduction of grafts on the PCL chains.3 5 6 0. a.4. et al  working with PCL‐g‐GMA.6. and a favored nucleation 1 2 50 120 . 6. For relatively high DEM intakes (> 30 % mol/mol). While for GMA an increase in BPO or monomer intakes invariably leds to higher FD values. This induces irregularities and is expected to result in a lowering of the Tc and the Tm.1. 3D plot.2 10 20 DEM Intake (mol% on CL units) 40 30 50 6 0. a higher BPO intake leads to a reduction of the FD values while an opposite trend is observed at lower DEM intakes.8 0.4 0. which is expected to lead to a Tc reduction. Thermal properties of the compatibilizers The thermal properties (Table 6. the melting temperature and the relative enthalpy decrease with respect to pure PCL. These differences in dependency of the FD values on the DEM and GMA intakes may be explained on the basis of the different grafting mechanisms as previously discussed. Remarkable is once more the different trends for the two substrates.Chapter 6 1 0.6 2 FD (%) 5 2 0 1 0. for DEM a clear transition is observed.
4 431. Table 6.0 430.2.8 372.4.3 380. Synthesis and Properties of Starch‐ PCL Blends The two compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) were further used as interfacial agents in blends of PCL with starch.5 425. a series with PCL‐g‐DEM (different intakes of the latter at fixed starch/PCL ratio) and two series with PCL‐g‐GMA (one with different intakes of PCL‐g‐GMA at a fixed starch/PCL ratio and one with a fixed compatibilizer intake at three different starch/PCL ratios).6 11.6 305.5 ε (%) 640.2 321. Our experimental design allows comparisons of thermal and mechanical properties as function of the starch content for binary blends (no compatibilizer) and intake of PCL‐g‐DEM and of PCL‐g‐GMA. The lower crystallization and melting enthalpy of the compatibilizers compared to virgin PCL is possibly caused by the presence of side chains on the PCL backbone which are expected to have a negative effect on the crystallinity of the products.9 379. 6.0 401.5 489.5 357.4.3 15.8 384.5 7.1 11.9 424. An overview of all prepared blends together with their thermal and mechanical (tensile tests) properties is given in Table 6. PCL‐g‐GMA has an FD of 9.2 9.3 Tc (oC) 35 36 36 36 30 30 30 29 31 30 29 28 29 ∆Hc (J/gPCL) 55 69 51 40 52 54 55 53 58 57 63 54 55 T m (oC) 57 57 57 57 57 57 57 57 58 57 57 58 56 ∆Hm (J/gPCL) 62 68 53 43 46 50 52 51 54 50 60 51 48 PCL‐g‐DEM has an FD of 1.6 168. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends of the PCL chains in the compatibilizers compared to virgin PCL (higher Tc) induced by the presence of polar groups. A total of 12 blends were prepared: a series of binary ones (not containing any compatibilizers) constituting our reference points.9 368.1 13.7 %.9 E (MPa) 270. Thermal and mechanical properties of starch/PCL blends a Sample PCL S/PCL 10/90 S/PCL 20/80 S/PCL 30/70 S/PCL/PCL‐g‐DEM 20/80/1 S/PCL/PCL‐g‐DEM 20/80/2 S/PCL/PCL‐g‐DEM 20/80/5 S/PCL/PCL‐g‐GMA 20/80/1 S/PCL/PCL‐g‐GMA 20/80/2 S/PCL/PCL‐g‐GMA 20/80/5 S/PCL/PCL‐g‐GMA 20/80/10 S/PCL/PCL‐g‐GMA 10/90/2 S/PCL/PCL‐g‐GMA 30/70/2 a σ (MPa) 16.2 337.7 10.6 386. 121 .8 329.9 11.6%.2 343.0 9.2 342.0 332.8 371.0 341.3.4 230.6 5.1 10.3 10.
3.). 75 70 65 60 55 50 45 40 35 ∆ H (J/ g-PCL) 0 5 10 15 20 Starch content (%) 25 30 Figure 6.43 MPa and 640.8 MPa) clearly indicates that the rigidity of the blends increases at higher starch contents.1 MPa and 341.7.2 MPa up to 3220.127.116.11.5 % for pure PCL to 7. respectively from 16.8 % at 30 % starch content. Binary blends of starch and PCL For binary blends (no compatibilizer is added) a monotonous decrease of the stress and strain at break is observed as function of the starch intake. The thermal and mechanical properties indicate no or very little interaction of the starch particles with the PCL ○ : ∆Hc □ : ∆Hm 122 . This is in agreement with the data previously reported on PCL/starch blends and related to the lack of interfacial adhesion between the starch particles and the PCL matrix . This is most probably due to a nucleation effect of the starch on PCL as observed also for other kinds of polymer blends .Chapter 6 6. The thermal behavior is characterized by no significant changes in the Tm and Tc values but unreported trends of the corresponding enthalpies as function of the starch content (Figure 6. The simultaneous increase in the modulus (from 270. The two trends are remarkably mirroring each other and are both characterized by an increase of ∆Hm and ∆Hc with respect to pure PCL at 10 % starch content followed by a monotonous decrease of both quantities as function of the starch intake. Melting and crystallization enthalpy as function of the starch content for binary blends with PCL.
). was used as compatibilizer for the preparation of ternary blends with starch and PCL. 123 . c. either by heating or in combination with mechanical stress . b. Figure 6.). whose synthesis is described for the first time in this work. 6. Ternary blends compatibilized with PCL‐g‐DEM PCL‐g‐DEM. and the lack of interface adhesion cause void formation. for which a 6 oC drop is observed when using PCL‐g‐DEM. 26] and explained by assuming that the voids are formed by water in the starch. the corresponding binary blend (S/PCL 20/80) contitutes a better reference point compared to pure PCL.) by constant values of the Tm (with respect to pure PCL) while the Tc is slightly lower (30 vs 35 oC) than that of pure PCL and it is independent of the compatibilizer content. a decrease with respect to pure PCL is observed and then a substantial invariance as function of the PCL‐g‐DEM content.3.2. This is confirmed by examination of the blend morphology by SEM (Figure 6.8. The presence of these voids was also observed for sago starch/PCL blends [25. However. in order to fully understand the role of PCL‐g‐DEM.9. the starch particles are simply inserted into voids of the PCL matrix.e. Such behavior is consistent with the hypothesis that the starch particles hinder the melting and crystallization processes of the PCL chains.2.8. (a) S/PCL 10/90 (b) S/PCL 20/80 (c) S/PCL 30/70 Indeed starch particles are clearly dispersed in the PCL matrix but no interaction (adhesion) between the two phases is actually detected.4. As consequence one might expect a slight different structure of these ternary blends as compared to the corresponding binary one. Morphology of starch/PCL binary blends. Evaporation of the water during blend preparation. This is visually confirmed by the morphology of the prepared blends (Figure 6. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends matrix. a. The latter trend is also valid for the corresponding enthalpies: i. Thus. The thermal behavior is characterized (Table 6. In this respect the thermal properties remain substantially unchanged with the exception of the Tc.
which will hinder the flowability and fibre forming ability of PCL matrix under cold drawing. At higher PCL‐g‐DEM intakes the lower average molecular weight of the compatibilizer as compared to the one of the unmodified PCL used in the blends (3000 vs 50000) is probably responsible for the observed decrease in the E values. Morphology of ternary blends compatibilized with PCL‐g‐DEM. Figure 6.).). (a) S/PCL/PCL‐g‐DEM 20/80/1 (b) S/PCL/PCL‐g‐DEM 20/80/2 (c) S/PCL/PCL‐g‐DEM 20/80/5 124 . A closer inspection of the SEM pictures reveals that the starch particles are clearly embedded in the PCL matrix with almost no void spaces at the interface. At 1 %‐wt of PCL‐g‐DEM in the blends the modulus increases with respect to the binary blend (0 %‐wt compatibilizer in Figure 6. resulting in more rigid material with higher modulus.). c. Indeed starch particles in the ternary blends display a smoother interface with the PCL matrix compared to the corresponding binary blends (Figure 6.Chapter 6 a. As a result of the different morphology also the mechanical behavior display differences with the virgin PCL and the binary blend. The stress and strain at break remain constant as function of the compatibilizer intake while the modulus display an optimum as function of the compatibilizer intake (Figure 6.10. b. This can be explained by the improved interfacial adhesion  between PCL and starch.9.9.10.
Ternary blends compatibilized with PCL‐g‐GMA When using PCL‐g‐GMA as compatibilizer two possible comparisons can actually be made: one at fixed starch/PCL ratio and changing the amount of PCL‐ g‐GMA and one at different starch/PCL ratios but with fixed amount of PCL‐g‐ GMA (2 %‐wt). 2) 3) 4) Remarkable is that in the case of PCL‐g‐GMA no optimum is found in the modulus as function of the compatibilizer intake. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 390 380 370 360 E (MPa) 350 340 330 320 0 1 2 3 PCL-g-DEM amount (wt%) 4 5 Figure 6.e.10. at all PCL‐g‐GMA contents) there is a clear increase of the modulus with respect to the 125 . it can be observed that: 1) the melting temperature as well as the crystallization and melting enthalpies do not change significantly (discrepancies in the values are within the boundary defined by the experimental error).3.3. the stress and strain at break are lowered and hardly a function of the PCL‐g‐GMA amount.2. the modulus is higher and hardly a function of the compatibilizer amount. in all cases (i. However. The first comparison as function of the PCL‐g‐GMA intake (ternary blends with starch/PCL ratio of 20/80) results in quite similar considerations as the ones made for PCL‐g‐DEM. by taking as reference the corresponding binary blend (S/PCL 20/80). 6. Modulus of ternary blends S/PCL/PCL‐g‐DEM as function of PCL‐g‐ DEM amount. the crystallization temperature experiences a drop of about 5‐6 oC. In particular also in the case of PCL‐g‐GMA.
) than PCL‐g‐GMA for blends containing 20 %‐wt of starch. This is surprising if one takes into account the different FD values: 9. b. As a result.4.11. These differences in compatibilizing effect are not yet fully understood.9. and morphologies in Figures 6. clearly indicating that also PCL‐g‐GMA (like PCL‐g‐DEM) acts as compatibilizer in the blends Such hypothesis is partially and qualitatively confirmed by the blends morphology (Figure 6. 29].Chapter 6 binary blend. Figure 6.12. and 6. As given schematically in Figure 6.8 % for PCL‐g‐GMA against 1. This confines all reactive GMA groups in relatively “isolated” spots along the PCL backbone. 28. despite a much more favourable FD value for PCL‐g‐GMA with respect to PCL‐g‐ DEM and even despite a relative low reactivity of ester groups towards the –OH groups on starch and nucleophilic groups in general [27. 126 . upon reaction of PCL‐g‐GMA with one of the –OH groups on the surface of the starch particles the remaining GMA groups are actually scarcely available for further reaction with other –OH groups spatially distant from the reacted one. a.). Morphology of ternary blends S/PCL/PCL‐g‐GMA. presence of voids. (a) S/PCL/PCL‐g‐GMA 20/80/1 (b) S/PCL/PCL‐g‐GMA 20/80/2 (c) S/PCL/PCL‐g‐GMA 20/80/3 In this case we observe structural features (partially smooth interface between the components. one might speculate that the longer length of the GMA grafts compared to DEM (vide supra) plays a negative role in the in situ formation of the compatibilizer. PCL‐g‐DEM is at least as active as compatibilizer (compare modulus values in Table 6. Thus. However. (left) the structure of PCL‐g‐GMA is inhomogeneous at the molecular level with long poly(GMA) branches pending from the PCL backbone. c. partial interfacial adhesion) which render these ternary blends a kind of “intermediate” case between the binary ones and those compatibilized with PCL‐g‐DEM.7 % for PCL‐g‐DEM.11.11.
As consequence. Moreover. Concerning the thermal behavior (Table 6. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends PCL-g-DEM PCL-g-GMA OH O OH O OH Starch particle OH Starch particle Figure 6.12. Schematic representation of the in situ reaction between functionalized PCL and starch. 20 and 30 %‐wt respectively).4.4. once PCL‐g‐DEM has reacted with one –OH groups on the starch particle. In order to check this we compared blends with the same amount of PCL‐g‐GMA (2 %‐wt) but with different starch intakes (10. for all starch amounts (Table 6. the Tc decreases (with respect to the corresponding binary blends) while all other factors (Tm and enthalpies) remain virtually unchanged with respect to the binary blends and also as function of the starch intake. Indeed. at higher starch contents more –OH groups would be available for reaction with the GMA groups. but since DEM is preferentially grafted as monomer.8 MPa for S/PCL 30/70 up to 430.12. The latter result (from a modulus of 341.12.12.) PCL‐g‐DEM has much less reactive groups (lower FD) than PCL‐g‐GMA. thus attenuating the effect discussed above. On the other hand (right of Figure 6. This is in agreement with the hypothesis made above (Figure 6. thus probably ensuring a better coverage of the surface.) for further reaction. when comparing ternary blends (compatibilized with PCL‐g‐ GMA) with the binary ones.3 MPa for S/PCL/PCL‐g‐GMA 30/70/2) clearly indicates that the efficiency of PCL‐g‐GMA as compatibilizer becomes more relevant at relatively higher starch contents (>30 %) .). The proposed explanation implies however that at relatively higher starch contents (> 20 %‐wt) the segregation of the poly(GMA) chains in “isolated” spot along the PCL backbone would become less important.). the distribution of the reactive group along the PCL backbone is more “homogenous”.) and relating the “coverage” of the starch particle surface upon reaction with PCL‐g‐GMA with the compatibilization efficiency. other groups will presumably remain available (arrows in Figure 6. We can therefore conclude that the efficiency of PCL‐g‐GMA in the 127 . the stress and strain at break decrease while the modulus is unchanged at low starch contents and increases significantly for S/PCL blends with 30 %‐wt starch.
4. The most important variable for the final product FD is the interaction between the amount of monomer and initiator used. moles of GMA or DEM present per mole of CL repeating units : : ni nm amount of initiator intake [mol% on CL units] amount of monomer GMA or DEM intake [mol% on CL units] 128 .5. mutual solubility of the monomer into the molten polymer) on the FD adequately described the experimental results (R2=0. From a more practical point of view it can be concluded that the newly synthesized PCL‐g‐DEM. At a fixed starch content (20 %‐wt) PCL‐g‐DEM seems to have sligthly more efficient compatibilizing effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values. Conclusions A systematic study. Nomenclature E FD : initial modulus [MPa] : functionalization degree. including statistical modeling. The PCL‐g‐GMA and PCL‐g‐DEM compatibilizers display lower melting temperatures and melting enthalpies compared to virgin PCL. 6. The reactive compatibilizers were used in blends of starch with PCL. The proposed model to quantify the effects of process variables (monomer and initiator intake. has been performed on the synthesis of two reactive compatibilisers. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. 6. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). P‐value ≤10‐9). This is in contrast with chemical reactivity and amount of chemical groups along the PCL backbone (both factors favorable to GMA as compared to DEM) but it is explainable on the basis of the group distribution along the PCL backbone. firstly reported in this work. can replace PCL‐g‐GMA as compatibilizer at relatively low starch contents offering at the same time the advantage of a less pronounced modification of the polymer backbone and a reduced consumtpion of polar groups to be grafted on PCL.957.Chapter 6 compatibilization of starch/PCL blends can be significantly improved by changing the blend composition. This represents an unreported mathematical confirmation of the fact that these kinds of functionalization reactions are mainly governed by the synergy between the different process variables and only slightly by their individual values.
Davis: High performance LDPE/ thermoplastic starch blend: a sustainable alternative to pure polyethylene. Ramsay.G. A‐C. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – III. Jarowenko: Acetylated starch and miscellanous organic esters. Ind. Sci. J. Pacsu: Starch Studies: Preparation and properties of starch triesters.Chem. 34. Karlsson. 1998. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – IV. 1997.W. Thermal and morphological properties. 1942. 227‐242.6. H. Mullen. M. 1998. R. E. Lindblad. . B. Mani. C. Albertsson: Biodegradable Polymers and Environmental Interaction. E. Polymer 2003. . 2002. Carbohydr. 44. wheat starch and soluble starch for food packaging applications. 1986. Effect of amylopectin to amylase ratio in starch. Ogawa. Karlsson: Polymers from Renewable Resources. 38. in Modified Starches: Properties and Uses (Ed.Eng. Sci.A. 34. M. . Polym. 1477‐1487. Boca Raton. J. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends T : temperature [oC] : crystallization temperature [oC] : melting temperature [oC] Tc Tm Greek symbols: ∆Hc ∆Hm : enthalpy of crytallization [J/g‐PCL] : enthalpy of melting [J/g‐PCL] : solubility parameter [cal1/2cm‐3/2] : stress at break [MPa] : strain at break [%] δ σ ε 6. Biliaderis. W.D.S. Part 2. S. Inc. 139‐161. Psomiadou. 1209‐1217.. CRC Press. Mani.J. . E. Ioannis. B. . 1251‐1253. USA. A. 1998. Polym. 1517‐1526. 129 . Kawasaki: Biodegradable films made from low density polyethylene (LDPE). M. Polym. R. 1467‐1475. Eur. Adv. 157. N. Wurzburg). 33. Rodriguez‐Gonzales. Eur. S. . Polym. Liu. A‐C. References . Y. Eng. J. . Albertsson. O.B. Ranucci. Polym. F. 34.
R.. 1998. Macosko. Sci. E. Coleman. 1997. 5135‐ 5142. Kim. Martuscelli. M.F. 1693‐1702. C. . Averous.. 1996. Degrad. 61‐67. R. New York. 2001. Laurienzo. J. Degrad.E. J. Raimo. John. Stab. Goldberg: A review of the biodegradability and utility of poly(caprolactone). Polymer 2000. Dole. L. Technomic Publication. J. Sci. 130 . 37. Pennsylvania. Khandpur. K. . Fringant: Properties of thermoplastic blends: starch‐polycaprolactone. J. Painter: Specific Interactions and the Miscibility of Polymer Blends. Guegan. Tang: Functionalization of polyesters with maleic anhydride by reactive extrusion. 3. C. 12. M. . . Polymer 2000. Yang. Part A: Polym. R. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. J. Polymer 2001. Adv. . J. Polym. 5590‐5598. 80. 81.W. Avella. Macromolecules 1996. Ruggeri. Solaro: Biodegradable polymeric materials. G.C.H. Marrucci. Tang. Prog. P. . Chiellini. Kim.. E. Sci. Chem. D. Environ. M. Rimedio: Preparation and characterization of compatibilised polycaprolactone/ starch composites.H. . 1273‐1335. A. Polym. C. K.K. Rustgi: Biodegradable polymers. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. 2004.M. Cho. . 23. USA. Polym. R. L. 305‐313. 1995. Polym. 127. 3875‐3881. C. J. P. J. 8. . D. Errico. Passaglia. It. 41. A.S. Chem. 2001. M. 1991. J. M. Inoue: Compatibilizers for melt blending: Premade block copolymers. 1507‐ 1516.Y. M. 41.K. USA. . 4157‐4167. E. 2003.K. . J. P. Part A: Polym. Bhattacharya.Y. M. Sci. Polym. Polym. Mani. 1139‐1148.K. Kim. M. Environ. .S. Cho. C. Aglietto: Chemical reactions affecting the free‐radical grafting of diethyl maleate onto ethylene polymers. Park: Grafting of glycidyl methacrylate onto polycaprolactone: preparation and characterization. John Wiley & Sons Inc. J. 179‐187. Gazz. P. Z.H. J. Chandra. Kim. Appl. T. . J. C. Graf. Marechal. 91‐95. K.M. R.Chapter 6 . 1997. Moro. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. Nakayama. Chim. P. . 127–134. Polym. Park: Modification of Aliphatic Polyesters and Their Reactive Blends with Starch. 35. 1999. Mater. Jung. 42. 29. C. Montgomery: Design and Analysis of Experiments 5th edition.
F. Tedesco. Odusanya. Passaglia. Ishiaku. 57. 21. Polym. B. 87.S. 1298‐1305. 293‐304. 2003.A. Polym. H. A. 11‐15.S. Pang. S.M. Eur. Appl. M. Polym.B. O.V.M. W. D. 9. Sci. 2000. 40. M. Polym.M. K. . 1998. U.A.Safa. Eng. M.S. Lee. R. Picchioni: Formation and compatibilizing effect of the grafted copolymer in the reactive blending of 2‐ diethylsuccinate containing polyolefins with poly‐ε‐caprolactam (nylon‐6). J. Ruggeri. Test. Kammer: On mechanical properties of sago starch/ poly(ε‐caprolactone) composites.N. D. Polym. 877‐884. Sci. . 393‐401. Ishak: Mechanical properties and enzymic degradation of thermoplastic and granular sago starch filled poly(ε‐ caprolactone). Ishiaku. Azemi. Azemi: Effect of starch predrying on the mechanical properties of starch/ poly(ε‐caprolactone) composites. Technol. 2006.Nasirtabrizi: Ring opening reactions of glycidyl methacrylate copolymers to introduce bulky organosilicon side chain substituents. E. O. 273‐281.D.N. R.M.S. B. Polym. Bull. U.S. S. Aglietto. K.W. U. Ishiaku. 2002. 38. Odusanya. . J.H. G. .M. 2002. Manan. Nachtigall.S. Manan.W. Adv. 131 . . Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends . Barbosa. Mauler: Comparative study of PP‐MA and PP‐GMA as compatibilizing agent on polypropylene/nylon 6 blends.
Starch‐based biodegradable materials are considered interesting candidates to replace certain types of conventional plastics. with a rate of almost 10% every year. it has a high volume to weight ratio and is generally resistant to degradation. such as thermoplasticization. There are several starch modification methods available. After use.Summary Plastics made from fossil resources (a. containing periodic branches linked with the backbones through α‐D‐(1. tensile and flexural strength). To gain insight in the potential of this approach. Starch is also too sensitive to water. The use of virgin starch for packaging materials is restricted because it cannot be shaped to films with adequate mechanical properties (high percentage elongation.6)‐glucosidic bonds. The main reasons are the water sensitivity of common catalysts and the fact that starch is poorly or even insoluble in the common organic solvents used for ROP. while amylopectin is branched polymer. plastics may end up in the environment and cause serious environmental problems. The synthesis of such grafted products by the in situ ring opening polymerization (ROP) of the monomers on the hydroxyl groups of starch is unfortunately not very straightforward. blending with other materials. chemical modification or combinations thereof. The application of a number of prospective methods for developing starch‐based biomaterials will be described in this thesis. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1.o oil and gas) are very attractive materials for a broad range of applications. The application of an alternative method involving hydrophobisation of starch by the introduction of large hydrophobic SiMe3 groups followed by a ROP with a polyester precursor and subsequent removal of the SiMe3 groups is expected to result in higher grafting efficiencies. Chemical grafting of biodegradable polyesters on the starch backbone is expected to result in less hydrophilic and thus less water sensitive materials with improved mechanical properties.4)‐glucosidic bonds. The content of amylose and amylopectine in starch varies and largely depends on the starch source. In combination with the current high prices for petrochemical products. there is a strong need for renewable alternatives for plastics from fossil resources. Starch is a polymer consisting of anhydroglucose (AHG) units. Examples are the use as packaging and construction materials. Starch is relatively cheap and available from a broad range of plants. For instance. Starch must therefore be modified before it can be applied as a biodegradable plastic. Its production has increased significantly since 1950. There are two types of polymers present in starch: amylose and amylopectin. we have .
The use of higher carboxylic (fatty) acid to esterify starch resulted in products with significantly improved 134 . The hydrophobicity of starch acetates is higher than virgin starch.2. but the products are still very brittle. The polymerizations were performed at 60‐100°C. The grafting efficiency for the desired reaction was 28‐58%. Poly‐(ε)‐ caprolactone grafted silylated starch co‐polymers with average chain length of 40‐ 55 monomer units (molecular weight of 4500‐6300) were obtained.68) was obtained. are described in Chapter 2. and off‐white solid products with isolated yields of 30‐96% were obtained. Most of the previous starch esterification studies involved the use of short chain carboxylic acids (C1‐C4). The results are described in detail in Chapter 3.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose) using Al(OiPr)3 as the catalyst. and particularly acetic acid derivatives (C2) have received considerable attention. Considerable amounts of ε‐CL homopolymers with isopropyl end‐groups were also formed.46‐0. The results for the model system involving the ROP of p‐dioxanone initiated by hydroxyl groups of a protected monosaccharide (1. with higher ratios leading to higher molecular weights.3. The grafting of ε‐caprolactone to the silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 was performed in THF at 50oC.2. The yield of the polymers was a function of the reaction temperature and the reaction time. with higher temperatures (100°C) leading to lower yields. Esterification of starch with carboxylic acid derivatives is one of the oldest strategies to improve starch properties.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose ratio (at constant Al(OiPr)3 intake).72. The silylation reaction was performed using hexamethyl disilazane (HMDS) in DMSO/ toluene mixtures at 70°C. depending on the ε‐CL to starch ratio.21‐0. and subsequent removal of the silyl groups.3. Average molecular weights of the products were between 970 and 6200 (7 – 58 monomer units) and were a clear function of the p‐ dioxanone/ 1. even in the presence of plasticizers. temperature and monomer: monosaccharide ratio) on the average degree of polymerization. The silyl groups of the poly‐(ε)‐caprolactone grafted starch co‐polymers were finally successfully removed using a mild dilute hydrochloric acid treatment in THF at room temperature. Characterization of the products using 1H‐ and 13C‐NMR as well as MALDI‐TOF mass spectrometry showed the presence of significant amounts of p‐dioxanone polymers with an isopropoxide end group (20‐30%) The knowledge obtained from the model system was applied for the synthesis of starch‐g‐poly‐ε‐caprolactone using hydrophobised silylated starch. The DS of the PCL chains was between 0.Summary initially performed research on the ROP with a simpler model system involving a simple protected mono‐saccharide instead of starch. A statistical model has been developed to quantify the effects of process variables (time. The synthetic procedure may be divided into three steps: hydrophobisation of starch by the introduction of SiMe3 groups followed by in situ ROP. Silylated starch with a low to medium DS (0.
The starch esters were prepared by reacting starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. PCL‐g‐ 135 . In Chapter 4‐5. however.5 MPa.1‐2.9 for starch laurate.5 for starch laurate.75) are still partially crystalline.3‐1. presumably due to a reduction of the polarity of the reaction medium. The DS of the products was a clear function of the chain length of the fatty ester.39) are totally amorphous whereas the low‐DS ones (DS= 1. and Na‐ acetate). including statistical modelling on the synthesis of long fatty esters of corn starch (starch‐laurate and starch‐stearate) using corresponding vinyl fatty esters is reported. The yellowish products were characterized by 1H‐. a systematic study.45‐1. The synthesis of these fatty starch esters is. 0. The use of methyl and glyceryl esters results in products with only relatively low DS. and the type of catalyst. an increase in the vinyl‐ester concentration led to higher product DS values. vinyl ester to starch ratio. The use of Na2HPO4 resulted in low‐medium DS products (0.24 ‐ 2. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. The products have a broad range of degree of substitution (DS = 0. while the use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐ starch mol ratio. and catalyst basicity) to the DS of the starch ester products were developed. The thermal and mechanical properties of some representative product samples are also described. 1. Chapter 4 describes the results of a preliminary study on the synthesis of long fatty esters of corn starch (starch‐laurate and starch‐stearate) using vinyl fatty esters.07‐1. With these catalysts. Mechanical tests show that the products have tensile strength (stress at break) between 2. K2CO3. High‐DS products (DS= 2. catalyst intake. products with a DS > 2. and Na‐acetate) in DMSO at 110°C. Reaction temperature and catalyst intake also affect the product DS but only to a lesser extent. 13C‐NMR and FT‐IR. and modulus of elasticity of 46‐113 MPa.26‐2. The DS of the products is a strong function of the basicity of the catalyst. Summary mechanical properties and hydrophobicities.5 for starch stearate). reaction temperature.96). In Chapter 6.7‐3. K2CO3. The last part of this thesis deals with reactive blending of starch and biodegradable polymers using reactive interfacial agents. the synthesis of two polycaprolactone (PCL)‐based reactive interfacial agents.4 could be obtained for both laurate and stearate esters. For low vinyl‐ester to starch ratios. Statistically adequate (R2≥0. elongation at break of 3‐26%. the DS decreased. The thermal stability of the esterified products is higher than that of native starch. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. In Chapter 5. At higher ratios. an alternative method for higher fatty starch esters by using fatty vinyl esters is reported.4‐3 for starch stearate). usually performed using fatty acid chlorides which are relatively expensive and rather corrosive.
PCL‐g‐DEM seems to have slightly more efficient compatibilising effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). P‐value ≤10‐9) has been developed to quantify the effects of process variables (monomer and initiator intake. mutual solubility of the monomer into the molten polymer) on the functionalisation degree (FD) of the GMA and DEM units to the PCL backbone.Summary glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). In particular the elastic modulus for the compatibilised blends is significantly higher (reaching up to 430 MPa) than that of uncompatibilised ones (320‐340 MPa). 136 . The PCL‐based compatibilisers were prepared by reacting a low molecular weight PCL (Mw 3000) with glycidyl methacrylate or diethyl maleate in the presence of benzyolperoxide initiator at 130°C.957. the mechanical behavior of the compatibilised blends is in general different from that of pure PCL and of the corresponding uncompatibilised blends. and their use in compatibilising starch‐PCL blends is described. A statistically adequate model (R2=0. As the result. At a fixed starch content (20 %‐wt). This is in contrast with chemical reactivity and amount of chemical groups along the PCL backbone (both factors favorable to GMA as compared to DEM) but it is explainable on the basis of the group distribution along the PCL backbone. at relatively high GMA and BPO intakes. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. Both of the PCL‐based compatibilisers improve the interfacial adhesion between the starch particles and the PCL matrix in starch/ PCL blends.
De AHG eenheden kunnen op een aantal manieren aan elkaar gekoppeld worden. Voorbeelden zijn thermo‐ plastificatie. ε‐caprolactone op de hydroxy‐groepen van zetmeel is helaas niet eenvoudig.4)‐glucoside bindingen verbonden zijn. Het zetmeel is daarnaast ook erg gevoelig voor water.v. De productie van kunststoffen is sinds 1950 met bijna 10% per jaar gestegen. menging met andere materialen. Zetmeel is een polymeer bestaande uit anhydroglucose (AHG) eenheden. Twee type polymeren zijn dominant aanwezig in zetmeel: amylose en amylopectine. De verhouding amylose en amylopectine in zetmeel varieert en hangt grotendeels af van de plant waaruit het zetmeel gewonnen wordt. Het chemisch grafting van biologisch afbreekbare polyesters op de zetmeel ketens zou kunnen leiden tot minder hydrofiele en dus minder water gevoelige materialen met betere mechanische eigenschappen. gevolgd door de ROP met een polyester precursor en verwijdering van de SiMe3 groepen .6)‐glucoside bindingen met de hoofdketen zijn verbonden. mede omdat het niet mogelijk is om films met goede mechanische eigenschappen te maken.Samenvatting Kunststoffen gemaakt van fossiele grondstoffen als olie en gas zijn zeer aantrekkelijke materialen voor een breed scala aan toepassingen. Na gebruik kunnen de materialen in het milieu eindigen en grote problemen veroorzaken. De toepassing van een alternatieve methode waarbij zetmeel eerst minder hydrofiel gemaakt wordt door de introductie van grote hydrofobe SiMe3 groepen. in combinatie met de huidige hoge prijzen voor de grondstoffen. Er zijn verschillende methodes beschikbaar om zetmeel te modificeren. Modificatie van zetmeel kan het aantal toepassingen sterk vergroten. Dit proefschrift beschrijft een aantal methodes om nieuwe polymere materialen op basis van zetmeel te maken. Het amylopectine is een vertakt polymeer waarbij de vertakkingen via α‐D‐(1. Biologisch afbreekbare polymeren op basis van zetmeel kunnen mogelijk bepaalde conventionele kunststoffen vervangen. Amylose is een lineair polymeer waarin de AHG eenheden hoofdzakelijk via α‐D‐(1. De synthese van dergelijke producten met een in situ katalytische ringopening polymerisatie (ROP) van geschikte monomeren als b. Zetmeel is relatief goedkoop en te winnen uit een breed scala aan planten. Het gebruik van zetmeel voor verpakking materialen is momenteel beperkt. De belangrijkste redenen zijn de watergevoeligheid van de katalysatoren en het feit dat zetmeel slecht of zelfs onoplosbaar is in de typische organische oplosmiddelen die voor de ROP gebruikt worden. heeft geleid tot een grote interesse voor biologisch afbreekbare alternatieven gemaakt uit hernieuwbare grondstoffen. Dit. chemische modificaties en combinaties van deze methoden.
Samenvatting zou in theorie moeten leiden tot een hogere grafting efficiency. De effecten van procesvariabelen (tijd. waarin hogere temperaturen (100°C) tot lagere opbrengsten leiden.2. De gevolgde procedure bestaat uit drie stappen: hydrofobisering van het zetmeel door de introductie van SiMe3 groepen. gevolgd door in situ ROP met ε‐ caprolacton en daarna de verwijdering van de silyl groepen. Er werden ook aanzienlijke hoeveelheden homopolymeer van ε‐ caprolacton met isopropyl eindgroepen gevormd. Om inzicht in het potentieel van deze benadering te krijgen is onderzoek verricht naar de ROP met een eenvoudig modelsysteem bestaande uit een beschermde mono‐saccharide in plaats van zetmeel. De silylerings reactie werd bij 70°C uitgevoerd met hexamethyl disilazane (HMDS) in DMSO/tolueen mengsels. temperatuur en monomeer: monosaccharide verhouding) op de gemiddelde polymerisatie graad zijn gekwantificeerd met een statistisch model.3. De silyl groepen van het poly‐ε‐caprolactone gesilyleerde zetmeel werden uiteindelijk succesvol verwijderd door middel van een milde behandeling met verdund zoutzuur in THF bij kamertemperatuur. De grafting efficiency van de gewenste reactie was 28‐58%. De resultaten voor het modelsysteem waar de ROP van p‐dioxanone wordt geïnitieerd door de hydroxy‐groepen van een beschermde mono‐saccharide (1. Een hogere verhouding leidde tot hogere molecuul massa’s. De substitutiegraad van de SiMe3 groepen varieerde van laag tot middelhoog (degree of substitution.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) met Al(OiPr)3 als katalysator worden in Hoofdstuk 2 beschreven. De grafting van ε‐ caprolactone aan het gesilyleerde zetmeel werd in THF bij 50°C uitgevoerd met Al(OiPr)3 als katalysator. Karakterisering van de producten met 1 H‐ en 13C‐NMR en MALDI‐TOF massaspectrometrie toonden de aanwezigheid aan van de gewenste producten en daarnaast significante hoeveelheden van p‐ dioxanone polymeren met een isopropoxide eindgroep (20‐30%). De resultaten worden in Hoofdstuk 3 in detail beschreven. 138 .21‐0. De polymerisaties werden tussen 60‐100°C uitgevoerd en leverde vaste producten in geïsoleerde opbrengsten tussen de 30 en 96%.68). De gemiddelde molecuul massa van de producten lag tussen de 970 en 6200 (7 ‐ 58 monomeereenheden) en was een functie van de verhouding p‐ dioxanone/1.2. De polycaprolacton eenheden (PCL) van het verkregen poly‐ε‐caprolactone gesilyleerde zetmeel hadden een gemiddelde ketenlengte van 40‐55 monomeereenheden (molecuul massa 4500‐6300).4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (bij constante i Al(O Pr)3 concentratie). DS = 0. De kennis verkregen met het modelsysteem is toegepast om ε‐caprolactone te graften op zetmeel.46‐0. De opbrengst van de polymeren was een functie van de reactietemperatuur en de reactietijd. De DS van de PCL ketens lag tussen de 0.72 en was afhankelijk van de ε‐caprolacton/zetmeel verhouding.3.
Echter de producten zijn nog steeds zeer bros. De producten werden met 1H‐. zelfs in aanwezigheid van plastificeermiddelen.96. vermoedelijk als gevolg van een verlaging van de polariteit van het reactiemengsel.24 tot 2. K2CO3 en Na‐acetaat) in DMSO bij 110°C te laten reageren. Echter. 0.96 en P‐ waarde van ≤ 10‐7). De substitutiegraad varieert van 0.07‐1. naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat. De effecten van belangrijke procesvariabelen (vinyl ester/zetmeel molverhouding. In Hoofdstuk 4‐5. In Hoofdstuk 5. De synthese van deze esters wordt in het algemeen uitgevoerd met vetzuurchloriden die vrij duur en sterk corrosief zijn.4 verkregen. K2CO3 en Na‐acetaat vertonen superieure activiteiten vergeleken met Na2HPO4. De thermische en mechanische eigenschappen van sommige representatieve producten zijn uitgebreid bepaald. De temperatuur en de katalysator concentratie beïnvloeden de DS ook maar in mindere mate. Bij lage vinyl‐ester/zetmeel verhoudingen leidde een verhoging van de vinyl‐ester concentratie in eerste instantie tot hogere DS waarden. en Na‐acetaat) te laten reageren. Het gebruik van Na2HPO4 resulteerde in producten met een laag tot middelhoge DS (0. bij verdere verhoging van de verhouding verminderde de DS. 13C‐NMR en FT‐IR gekarakteriseerd. Hoofdstuk 4 beschrijft de resultaten van een voorstudie naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat) met de vinyl esters van de vetzuren al reagens. Met deze katalysatoren worden voor zowel de laureaat als stearaat esters producten met een DS van boven de 2. De acetaat esters zijn in het algemeen minder hydrofiel dan zetmeel. wordt een systematische studie. inclusief statistische modellering.3‐1. De DS van de producten is een sterke functie van de basiciteit van de katalysator. wordt een alternatieve methode voor de synthese van zetmeel esters met lange vetzuren beschreven. de vinyl‐ ester/zetmeel verhouding en het type katalysator. De zetmeel esters werden bereid door zetmeel met vinyl laureaat of vinyl stearate in DMSO in de aanwezigheid van basische katalysatoren (Na2HPO4. terwijl het gebruik van de K2CO3 en Na‐acetaat katalysatoren 139 .5 voor zetmeel stearaat). De DS van de producten was een duidelijke functie van de keten lengte van de vetzure vinylester. K2CO3. Het gebruik van hogere (vet‐) zuren resulteerde in producten met beduidend betere mechanische eigenschappen en hydrofobiciteit. In de meeste zetmeel esterificatie studies wordt gebruikt gemaakt van kleine carbonzuren (C1‐C4). De zetmeel esters werden bereid door het zetmeel met vinyl laureaat of vinyl stearaat in aanwezigheid van basische katalysatoren (Na2HPO4.5 voor zetmeel laureaat. Samenvatting De esterificatie van zetmeel met carbonzuur derivaten is één van de oudste strategieën voor het verbeteren van zetmeeleigenschappen. en in het bijzonder azijnzuur derivaten (C2). reactietemperatuur. Het gebruik van methyl en glyceryl esters resulteert in producten met een lage substitutiegraad. katalysator concentratie en katalysator basiciteit) op de DS zijn gekwantificeerd met een statistisch model (R2 ≥ 0.
Als reactieve compatibilisers zijn twee polycaprolactone derivaten. De PCL gebaseerde compatibilisers werden bereid door reactie van laag moleculair PCL (Mw 3000) met glycidyl methacrylaat of diethyl maleaat in de aanwezigheid van een benzyolperoxide initiator bij 130°C.1‐2. wederzijdse oplosbaarheid van het monomeer in het gesmolten polymeer) op de functionalisatie graad (FD) van de GMA en DEM eenheden aan de PCL keten adequaat gekwantificeerd kunnen worden. De hoogste FD waarden (tot 45%) werden gevonden voor PCL‐g‐GMA bij relatief hoge GMA en BPO concentraties. een uitrekking tot breuk van 3‐26%. De producten met een hoge DS (2. Mechanische tests tonen aan dat de producten een treksterkte hebben variërend van 2.Samenvatting resulteerde in producten met een middelhoog tot hoge DS waarden (2. gebruikt. 140 . PCL‐g‐glycidyl methacrylaat (PCL‐GMA) en PCL‐g‐diethyl maleaat (PCL‐g‐DEM).0 voor zetmeel stearaat). Het laatste hoofdstuk van dit proefschrift (Hoofdstuk 6) behandelt de menging van zetmeel en biologisch afbreekbare polymeren met reactieve compatibilisers. Zo is bijvoorbeeld de elastische modulus van de gecompatibiliseerde mengsels beduidend hoger (tot 430 MPa) dan die van niet‐gecompatibiliseerde mengsels (320‐340 MPa).26‐2.7‐3. Dit leidt tot verschillen in mechanische eigenschappen tussen gecompatibiliseerde mengsels en van de niet‐ gecompatibiliseerde mengsels van zetmeel en PCL. en een elasticiteit modulus van 46‐113 MPa. De synthese van deze verbindingen wordt uitgebreid beschreven. Er is een statistisch model (R2 =0.39) zijn volledig amorf terwijl de producten met een DS van 1. De thermische stabiliteit van de geësterificeerde producten is hoger dan die van zetmeel. De FD waarden voor PCL‐g‐DEM waren aanzienlijk lager (tot 7%). 1.5 MPa.45‐1.957 en P‐ waarde van ≤ 10‐9) ontwikkeld waarmee de effecten van procesvariabelen (monomeer en initiator concentratie.4‐3.9 voor zetmeel laureaat.75 nog gedeeltelijk kristallijn zijn. Beide reactieve compatibilisers verbeteren de interactie tussen de zetmeeldeeltjes en de PCL matrix.
L. the raw material for fossil derived plastics. Marcel de Vries. Erwin Wilbers. Erik Heeres. Leon Janssen. Prof. He has also been a good friend and always supported me during the hard times. I want to thank the persons of Building 18 in Nijenborgh 4 for making me feel at home in Groningen and for their help during my research. for the inspiring ideas during the discussions. Minnaard. A. and for helping me arranging many things when I was not able to do it from Indonesia. Anne Appeldoorn. I want to thank Erik Heeres for being more than a good supervisor. as well as taking the SEM pictures. for upgrading my writing quality. for his efforts to make every foreign students feel at home in Groningen. I also want to give my deep appreciation to my reading committee. and Prof. Anno was also always there when . not in the last place because of the problems related with conventional plastics. for her hospitality and solving administrative matters. and for always encouraging me to finish this thesis. People from other groups have also been very kind and supportive to my research. I want to thank my three promoters. I want to thank Marya van der Duin‐de Jonge. I want to thank Francesco Picchioni for being always there when I needed him to discuss anything. which improved the quality of this thesis. since the mechanical properties as well as hydrophobicity of this starting material is relatively poor. Thank you very much for spending your valuable time reading my thesis and for the constructive comments and corrections you have made. Prof.Acknowledgements The synthesis of biodegradable materials is currently receiving a lot of attention. Thank you Erik. who was my first daily supervisor for about two months. I also want to thank Ineke Ganzeveld. for your consistent support. for valuable discussions during my early PhD period and for handing me over to the right hands. These include the current high oil prices. I found out that developing a better fully‐biodegradable‐material from starch is not an easy job. A. I want to thank Leon Janssen for his hospitality. Mosiscki. This thesis is a summary of my research on developing novel biodegradable polymers from starch. Francesco Picchioni. Broekhuis. Anno Wagenaar and Wim Kruizinga taught me how to operate the NMR machines and to interpret the spectra. and Laurens Bosgra for being always helpful with technical issues and Jan Henk Marsman for the help with analytical equipment. and Prof. Prof. Although it is indeed a very interesting topic. First of all. Prof. together with difficulties in managing plastic waste.A.J. especially on his opinions about the statistical modelling and mechanical properties. Gert Alberda van Ekenstein and Harry Nijland helped me a lot with mechanical and thermal property measurements.
Jaap Bosma. Our collaboration has also become part of this thesis. our memories will never fade away. Togi. and Tante Alma for being like my parents and family in Groningen. Abdul Osman. Chalid. and for my wife and family. Thank you my parents and sister. Anneke Toxopeus. You all have always been my inspiration and motivation. Ita. Wahyudin. mas Pandu. patience. and passion. Tiara. Ais Jenie. Mita. Patrick and Oppie. My special thanks are for Laura Junistia and Jan‐Pieter Drijfhout. and helpful. kak Roga. and Henk van de Bovenkamp for cheering my days and being my friends during summer and winter. Farchad H. Wuri Raspati. Hofman. Dado. Sari Fabianti. and although we seldom met afterwards. Francesca Gambardella. Taufik. Wita Sondari. Thanks also for Marcel and Connie. and on many occassions. and many others for the good old times! Thank you Tante Ietje. kak Atha. I have met a lot of Indonesian students in Groningen. Yao Jie. Zhang Youchun. pak Harry. Nidal Hammoud Hassan. Agnes R. Ignacio Melian Cabrera. Hans Heeres. Iwan Harsono. Endy. M. Thank you God. Niels van Vegten. visiting researchers. Diana Santangelo. Francesca Fallani. Thyara. Rasrendra. Jelle Wildschut. Henky Muljana (my paranymph) and Anindhita Widyadhana. Ronny Sutarto. Wangsa. A lot of thanks for Henk Stegeman and Poppy Sutanto. and the ISD staff helped me a lot with visum and residence permit arrangements as well as tax papers. The chemical and glassware magazijn as well as the FWN library personnel were always patient. mbak Ika. Ardiyanti. Tante Oppie. for never leaving me alone. Gerard Kraai. Buana Girisuta and Rina Karina. Teddy. in the lab. and Iwan Kustiawan for becoming my family. Puspita Kencana. Egi. Thank you AW. friendly. Diana Jirjis.B. support. C. and assistant professors have shared the days with me in Groningen. Arjan Kloekhorst. Albert Kieviet helped me with the MALDI‐TOF measurements. Marcel Wiegman. Nadia Gozali. mbak Mia. I want to thank Edwin A. spring and autumn in Groningen. Many master and PhD students. for everything. Samuel. Groningen is also a place where I found true friendship. bu Ida Susanti. Erna Subroto. R. August Kurniawan. Annette Korringa.Acknowledgements I had questions or needed specific equipments. Teguh. Tenny. Judy Retti Witono. Yongki. for your love. Jasper Huijsman. Asal Hamarneh. Louis Daniel. Mahfud. Boelo Schuur. master students who worked with me. Vincent Nieborg and Fesia Lestari. Tante Smith. Tante Caroline. 142 . Oscar Rojas. and Mahesa for being my brothers and sisters in Groningen. Thank you Tresna. Anna Nizniowska. Kuslan. Bima and Wisnu (also my paranymph). in the office. Taufik. for your never‐ending blessings.
(in press.2008.eurpolymj. 5. Sugih. H. Eur. starch‐based innovative product: The synthesis of grafted polycaprolactone‐starch by ring opening polymerisation.A.J. F. L.K. A. L. J. A. and H.J.P.P.J.Janssen.List of Publications 1. Sci.J. Manurung. A. L.K. J. L. Bratislava (Slovak Republic). Sugih. Sugih. A.K. Heeres. Drijfhout. Broekhuis. Experimental Studies on the ring opening polymerization of p‐dioxanone using an Al(OiPr)3‐monosaccharide initiator system. A. A. J. H. India.P. Synthesis and properties of reactive interfacial agents for polycaprolactone‐starch blends. Junistia.1016/j. Picchioni: Biodegradable polymer materials based on polyesters and starch/proteins. Picchioni. presented in “Polymers for Advanced Technologies”. 9. presented in 11th Asia Pacific Confederation of Chemical Engineering (APCChE) Congress. . 2. F. Heeres.P.K. Janssen. H. R.J. Broekhuis. Janssen. presented in International Workshop: “From Polymer Modification to Multicomponent Polymer Systems”. A. DOI: 10.B. A. Polym.010). Experimental and modeling studies on the synthesis and properties of higher fatty esters of corn starch.M. Picchioni. H.K.K. Sugih. 9th National Conference organized by the Society for Polymer Science.) 6.P. 2008 (submitted to Carbohydr. Drijfhout. F. 2008 (submitted to J.B. Pune Chapter. Picchioni. Picchioni. Polym. Sugih. Picchioni: Functionalization and blending of polycaprolactone. Kuala Lumpur (Malaysia). Sugih. F. H. 26‐28 November 2006 (oral presentation).K. F.M Janssen and H. 2008 (accepted for publication in Starch‐ Starke). Appl. L. A. F. R. L. A.J. Heeres: Green. Junistia.M. H. 4. 3. Picchioni. 8. 25‐28 September 2007 (oral presentation). L.J. Sugih. Manurung. and H.).J. 2008. Polym. and F. Synthesis of poly‐ (ε)‐caprolactone grafted starch co‐polymers by ring‐opening polymerisation using silylated starch precursors. 17‐20 December 2006 (oral presentation). Mumbai (India). 2008 (accepted for publication in Starch‐Starke).P. Heeres.Drijfhout. Sugih.K.M.Picchioni.K. South China University of Technology. Heeres: Synthesis and properties of novel green biopolymers derived from starch and fatty acids.10. presented in the 1st International Symposium on Sustainable Chemical Product and Process Engineering. Heeres.Heeres. F. Sugih.B. 27‐30 August 2006 (oral presentation) 7. J. Picchioni.P. and F. Guangzhou (China). Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Heeres.B. A. Junistia. Junistia.A. L.J.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.