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
P. Promotores : Prof. ir.M. H.B. dr. dr. F. dr. dr. Picchioni Beoordelingscommissie : Prof.J. dr. dr.J. ir. Moscicki ISBN : 978‐90‐367‐3592‐6 ISBN : 978‐90‐367‐3593‐3 (electronic version) . A. Heeres Prof. Janssen Prof. A.A. L. L. Broekhuis Prof. Minnaard Prof.
and my sister . to: Tresna. my parents.
3.2.Table of Contents Chapter 1: Introduction 1. Thermoplasticized starch 22.214.171.124.126.96.36.199.2. Methods 188.8.131.52. Starch Esters 1. Materials and Methods 2. Introduction 2.2.3. The potential of biodegradable plastics 1. Biodegradable plastics from starch 184.108.40.206.220.127.116.11. Starch production processes 18.104.22.168.1.2. Starch ‐ biopolymer blends and graft co‐polymers 1. Starch 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 2. Thesis Outline 1. Plastic applications and waste issues 22.214.171.124.2. Starch‐based blends by in situ polymerization 1.2. Starch‐based blends by melt mixing 1. Starch modifications to improve product properties 126.96.36.199. Structure and properties of starch 1. Product analyses . Cross‐linked starch 1.2. Biomaterials from starch 1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (2) 27 28 26 27 27 27 2.2.3. Materials 2.2. Typical example for the synthesis of polydioxanone end‐capped with 188.8.131.52.
3. 184.108.40.206.220.127.116.11. Typical example of the starch silylation procedure 3. Introduction 3.2.3. Materials and Methods 3. Calculation of average degree of polymerization 18.104.22.168.22.214.171.124.2.1. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers 51 51 52 52 52 53 54 48 49 49 50 50 3. Product yield 2. Methods 3. Peracetylation of silylated starch 3.3.2. Results and Discussions 2. Product analyses 2. Mechanistic aspects 2.4. Screening experiments 126.96.36.199. Analytical methods 3.1. Typical example of in situ polymerization of ε‐CL with silylated starch 50 3. Synthesis of silylated starch viii . Systematic studies 2. Nuclear Magnetic Resonance (NMR) 3.2. Nomenclature 188.8.131.52.1.1.2. 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 44 184.108.40.206.3.1. Results and Discussions 220.127.116.11.18.104.22.168.3. Conclusions 2. Materials 3. Calculations 22.214.171.124.3. References Chapter 3: Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors 3.
Product characterisation 4.3. Analytical equipment 4.3. Nomenclature 3.3.2. Conclusions 126.96.36.199. Catalysts screening 4. Nomenclature 188.8.131.52.3. 1H‐ and 13C‐NMR analyses 4.1.3. FT‐IR measurements 4.3.3. Effect of the addition of toluene as a co‐solvent 4.1. Systematic studies 4.3.2. Deprotection of silylated‐starch‐g‐PCL 184.108.40.206.1.2. References Chapter 4: Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants 220.127.116.11.18.104.22.168. 3.6. Exploratory experiments 4. Peracetylation procedure 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 22.214.171.124. Determination of the Degree of Substitution (DS) 4.2.3. Results and Discussion 4. Effect of vinyl ester to AHG ratio on the product DS 4.5. References ix . Conclusions 4.2.5. Methods 126.96.36.199.188.8.131.52. Materials 4.3. Materials and Methods 4.3. Introduction 4.3. In situ ring opening polymerization of ε‐caprolactone with silylated starch 57 63 64 64 65 3.
2.3. Mathematical modelling 5. Materials 6.3. Methods 6.1.3. Introduction 6.2. Product properties 184.108.40.206.5.2. Peracetylation procedure and Degree of Substitution (DS) determination 88 88 89 92 96 102 102 103 220.127.116.11. Compatibilizer synthesis 6. Experimental Design 5.2. Preparation of PCL‐starch blends with the reactive compatibilisers 110 18.104.22.168.2. References Chapter 6: Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 6. Conclusions 5. Materials 22.214.171.124.2.2.Chapter 5: Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch 5.4. Nomenclature 5.2. Materials and Methods 6.2. Results and Discussion 5.2. Analytical Equipment 5.2.2. Introduction 5. Work‐up of PCL‐g‐GMA products 6. Methods 126.96.36.199.2.4.2. Analytical Methods 110 x .1.2.2. Materials and Methods 5.3. Typical example of the preparation of laurate and stearate starch Esters 88 86 87 87 87 88 5.2. Work‐up of PCL‐g‐DEM 108 109 109 109 109 110 110 188.8.131.52.
3.2.2. Ternary blends compatibilized with PCL‐g‐DEM 6.3.3. 6. References Summary Samenvatting (Dutch Summary) Acknowledgements List of Publications 133 137 141 143 xi .1.1. Statistical mdeling 6.1. Results and Discussions 184.108.40.206.3.3. Synthesis and Properties of Starch‐ PCL Blends 6.3.4. Nomenclature 220.127.116.11.2. Thermal properties of the compatibilizers 18.104.22.168.4. 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 6. Effect of the BPO intake on the product FD 6.5.3. Ternary blends compatibilized with PCL‐g‐GMA 22.214.171.124.126.96.36.199. Effect of substrate (GMA/DEM) to PCL Ratio on the FD 6.3.1. Binary blends of starch and PCL 6. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers 6. Preparation of the ractive compatibilizers 6. Conclusions 6.
. 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. Finally.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.
potato or cassava) of the plants [2‐3]. or peas) and in tubers or roots (i. barley.1. current and future estimations of worldwide starch production are given in Figure 1. sorghum. corn.9 ton starch per hectare . wheat. Past. Most starch crops are very productive. Plants synthesise and store starch in their structure as an energy reserve. sweet potato. rice. Starch is found in seeds (i.1. 80 70 60 50 40 30 20 10 0 Starch Production (Million Tons) 1995 2000 Year 2005 2010 Figure 1. maize. Past.1. The current price of corn starch is around $0. .45 (estimated from the total volume and value of the US corn starch export ). and wheat starch are also produced in large amounts [2. Potato accumulates starch to approximately 75 % of the dry weight in the tubers with a yield up to 21 ton starch per hectare. The second and third starch producer regions are Europe and Asia [4‐5]. It is generally deposited in the form of small granules or cells with diameters between 1‐100 µm . while corn seeds consist of 65‐80% starch by weight. potato. The worldwide production of starch in 2008 is estimated to be around 66 million tons . .e. but other types of starch such as cassava. present and a forecast of starch production  o : Europe □ : USA : Total ∆ : the Rest of the World 2 . Most of the starch produced worldwide is derived from corn. Most of the starch is produced in the USA. 4‐5].e. An important polysaccharide is starch. with an average yield of 4.Chapter 1 1. Starch 75% of all organic material on earth is present in the form of polysaccharides.
and starch. the corn kernel is softened by steeping in warm water containing SO2 until the volume of the kernel increases with 55‐65%. The resulting suspension from the mills contains fiber.b. Figure 1.a. and starch granules are obtained after centrifugation from the suspension. The fiber is removed using washing screens. Starch production processes Starch is generally extracted from the plant by wet milling processes [9‐10]. After a coarse milling. Typical corn‐milling operation b. b.1. The low density gluten is separated from the starch suspension by centrifugation. The resulting starch is further washed in a cyclone and finally dried. the mixture is fed to a hydrocyclone to separate the germ and the rest of the kernel. a.2. After initial cleaning to remove cob. The starch is present in the endosperms (floury and horny). The plant material is grounded in water. As an example.2. which is again fed to a second milling process.1.2. and is embedded in a proteinaceous cellular matrix as is shown in Figure 1. a typical corn starch production process is given in Figure 1. and other foreign materials. the debris is filtered from the slurry. Cross‐sectional view of a corn kernel 3 . Corn starch production  a. Introduction 1. gluten. sand.
Amylopectin Starch is insoluble in cold water. Anhydroglucose (AHG) unit b. H OH H2COH H2COH H . 6]. H . The linear 4 . Figure 1. Each branch contains about 20‐30 anhydroglucose units. The amylose content of several common starches is given in Table 1. H2COH H H OH H O H OH O H OH O CH2 H O H H OH H O H OH H O .2. but it is very hygroscopic and binds water reversibly.1. Two types of AHG polymers are usually present in starch: amylose and amylopectin [2‐3..3. a. The molecular weight of amylose is a function of the plant source and processing method.6)‐ glucosidic bonds . Amylopectin is a branched polymer.......) .3... The molecular weight of amylopectin is higher than that of amylose and is typically 4‐5 x 108 Da . H OH H O H OH H O H H OH H c..6‐7 x 105 Da . The starch granules will swell rapidly to many times of its original volume.4)‐glucosidic bonds. The content of amylose and amylopectine in starch varies and largely depends on the starch source..a... Amylose c. Structure and properties of starch Starch is a polymer consisting of anhydroglucose (AHG) units (see Figure 1. but usually in the range of 1. Chemical Structure of Starch a. H O H OH H OH H O H OH O H OH H O .Chapter 1 1. .1. containing periodic branches linked with the backbones through α‐D‐(1. H OH H H2COH H2COH H O H H OH H H2COH H O H H OH H O H OH H O . Heating a starch solution leads to loss of hydrogen bonding in the interior of the starch granule and the starch will start to gelatinize. Typically. H2COH H OH H O H H O b. the amylose content is between 18‐28% .. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. . H2COH H .
2. The resulting suspension contains a mixture of linear amylose molecules. and. will form a thick paste or gel. Plastics are used in a wide range of applications and the demand is still increasing every year .7 20 18. 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 . Introduction amylose molecules leach out of the granules into the solution.8 16.5‐70 62‐72 63‐72 58‐64 1. Starch gelatinization temperature range  Starch Potato Tapioca Corn Waxy corn Wheat Gelatinization Temperature Range [o C] 59‐68 58. Plastic Applications and Waste Issues Plastic is the general term for a wide range of synthetic or semisynthetic polymerisation products. The first generation of commercial plastics was derived from cellulose nitrate and is known as celluloid [12‐14]. Biomaterials from starch 188.8.131.52 28 52 70‐75 27 15. and granule fragments. depending on the amount of water present.2. Parker in 18184.108.40.206 17.7 26 Table 1. 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. swollen granules. The gelatinization temperature range of various starch sources is given in Table 1.5 25. Table 1. and celluloid was 5 . Cellulose nitrate was first prepared by A.
Fibers 150 1989: 100 100 1976: 50 50 1950: 1. polypropylene (PP). polystyrene (PS). In 2006.). and PolyAcryl. and polycarbonates (PC).5 0 1950 1960 1970 1980 Year 1990 2000 2010 Figure 1. for instance to be used in packaging and paper coatings . Sealants. Plastics are very attractive materials.4. PA-. Polyurethanes. electrical. and energy during transportation of goods. space. Worldwide Plastics Production (1950‐2006)  6 . The production and consumption of plastics has increased significantly with a rate of almost 10% every year since 1950.5. the worldwide plastics production has reached 245 million ton per annum  (see Figure 1. Elastomers. and consumer products (see Figure 1. when polyethylene (PE) was invented. and 37% of the total plastics demand in EU  is used as packaging materials. Thermosets. Important polymers used for packaging are polyethylene (HDPE and LDPE).) 250 Included: Thermoplastics. Plastics are also used for building materials and automotive. Adhesives. Hyatt in 1870 . About 29% of the total plastics produced in the USA . The latter is important when using the material for packaging purposes to save weight.Chapter 1 patented by J.4. PE is a very versatile plastic because it can be shaped easily into various forms. Coatings. They have a low density and can be shaped in thin films while maintaining good properties. While celluloid is derived from a natural polymer (cellulose). the oldest purely synthetic plastic is Bakelite. The largest application of plastics is for packaging purposes. Plastics have lower melting temperatures compared to glass and metals. polyethylene terephtalate (PET). polyvinyl chloride (PVC). and therefore need less energy to shape it into useful materials . and PP-Fibers 2006: 245 2002: 200 200 Plastics Production (Million Tons) Not Included: PET-. A dramatic increase in demand for plastics began after World War II. discovered by Baekeland in 1907 [12. 15].
which are usually found in waste streams. 16]. 6]. threatening fishery. irrigation. Introduction Plastic waste.7% in 2006 . only 20% of the plastic is recycled. consisting of 25%‐v of the total waste . and is only effective for single plastic sources or simple plastic formulations. Co‐mingled plastics. and lakes. Plastic applications and waste treatment in Europe  1. hydropower plants operation. and accumulation of its residues in soil cause significant reductions in agricultural yields. are not easy to recycle . In the USA.5. rivers. however.2. Recycling (part of) the plastics to reduce the amount of MSW also has limitations. Most of the waste is still disposed (50%) by landfill or incinerated to recover the energy (30%) . 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). ship navigation. Figure 1. soil. plastics waste was ranked as the second major source of MSW after paper and paperboard. is causing serious environmental problems. It has a high volume to weight ratio and is resistant to degradation. In Europe (2006). Plastics have been polluting sea [1. the amount of plastics in MSW increased from less than 1% in the 1960 to 11. Plastic litter is hazardous to wildlife [1. The volume fraction of plastic in MSW is much larger due to the low density of plastics. It is not applicable for thermoset resins . Biodegradable plastics are polymeric materials capable of decomposing when given an 7 . and other public works .2. In 1996. The disposal of plastics materials in municipal solid waste (MSW) is a serious issue in many parts in the world. and may be more than twice the weight fraction .
biological and/or physical forces in the environment at a rate which leads to fragmentation or disintegration of the plastics. Several authorities have provided definitions for biodegradable plastics . Biodegradable plastics have gained considerable interest since the 1980s. The price of these biodegradable plastics is expected to be reduced considerably in the next decade . proteins (present abundantly in plants and animals). and microorganisms) and biodegradable synthetic polymers . DIN 103. For instance. The change in chemical structure results from the action of naturally occurring micro‐ organisms. Common natural biopolymers are carbohydrates. Nowadays new types of biodegradable plastics with improved properties and lower costs have been developed . the price of PLA based plastics in 2010 is estimated to be around € 1. 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. animals. It may also be treated in sewage sludge water treatment plants or buried in the soil [20‐21]. because they take part in nature’s cycle of renewal. and since 2000 they have become competitive with traditional materials in some applications [20‐21]. and polyesters from micro‐organisms . Biodegradable plastic waste may be treated in composting facilities.5/ 8 .Chapter 1 appropriate environment and sufficient amount of time . ASTM sub‐committee D20.2 working group on biodegradable polymers ‐ Biodegradation of a plastic material is a process leading to naturally occurring metabolic end products. Biopolymers are inherently biodegradable.96 proposal ‐ Degradable plastics are plastic materials that undergo bond scission in the backbone of a polymer through chemical. The market price of biodegradable plastics (mainly from starch and PLA) has decreased the last years. together with food and yard waste as well as paper. 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 . biopolymers from nature (from plants. Biodegradable plastics may be classified into two general groups. 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.
blending with other materials. starch must be modified. half of its price in 2003 (around 3 Euro/kg). because the price of oil based polymers may still increase due to the rise in the crude oil prices . Biodegradable plastics are also used for food servicewares . 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 . 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.50/ kg in 2005) is also expected to be lower in the future due to considerable reduction in modification costs . The price of starch‐based biodegradable plastics (€ 1. Examples are the use as packaging material for carrier bags.2. breathable fabrics. Virgin starch is not suitable as a packaging material. towels. such as polyethylene (PE. Biodegradable plastics are especially very useful for single‐use applications. Consequently. when recycling is not practical or uneconomical [20‐21].4 or $ 1. It cannot be shaped in films with adequate mechanical properties (high percentage elongation. 1. utensils . napkins) . tensile and flexural strength) and is too sensitive to water [26‐27]. and chewable items for pets [1. agricultural products (mulch films. chemical modification or combinations of them . Starch may become an attractive raw material for plastics in the future. pots). hygienic materials (diapers.3. 9 .4 per kg in 2007‐2008 ). Introduction kg.50‐4. either from starch‐based materials (slightly modified starch. biofillers for tires. 20]. The price of starch in 2007 was about $ 0. consumer products and food products.2‐ 1. This price is much lower than conventional plastics derived from oil.45 per kg . € 1. alone or complexed with natural or synthetic biodegradable polymers) or from polylactic acid which originates from the fermentation of a starch feedstock.8‐ 2. Biodegradable Plastics from Starch Starch is a very attractive source for the development of biodegradable plastics. either by plasticization. before they can be applied as biodegradable plastics. 1. or when environmental impacts have to be minimized. These are summarized briefly in the next sections. .3.
such as water and polyols (glycerol. extrusion. 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. and therefore are processable by conventional polymer processing techniques such as injection. This problem is mainly caused by the presence of strong inter‐ and intra‐molecular hydrogen bonds between the starch macromolecules [28. and the properties and amount of plasticizers. 40].3. 39]. resulted in the formation of an amorphous mass [35. 37]. the semicrystalline structure of starch and its granular form are lost and the starch polymers are partially depolymerized. 35. 32‐ 36]. Another issue is the poor water resistance and low strength. The modification involves break down of the starch granular structure by the use of plasticizers at high temperatures (90‐180 oC) and shear. The use of water as a plasticizer is not preferable. The brittleness is known to increase in time due to free volume relaxation and retrogradation. There are several substances used as plasticizer for the preparation of thermoplastic starch (TPS). or melted starch [29. 30]. glycol. 32. and hexylene glycol. because the resulting product will be brittle when equilibrated with ambient humidity . Although thermoplasticization seems to be a promising method. 33]. which results in embrittlement. The plasticizers are also usually hydrophilic and can be 10 . which is even above the thermal degradation temperature. Starch can be modified to obtain materials which melt below the decomposition temperature . which will result in a continuous phase in the form of a viscous melt [29‐30. which still limits their use. with better properties than virgin starch in various applications. 30. 40]. . plasticized. The use of other plasticizers (for example glycerol) results in a rubbery material. and blow moulding [28. water content.Chapter 1 1. At intermediate water levels. Yu. The modified products are known as thermoplastic. glycerol. The plasticized starch properties may be tuned by changing the temperature of processing.1. As the result. Thermoplasticized starch Virgin starch is brittle and difficult to be processed into articles due to its relatively high glass transition temperature (Tg. For instance. The thermoplasticization process will decrease the interactions of the molecular chain and destruct the structure of the starch . approximately 230 oC) . sorbitol. TPS synthesized from polyol and sugar plasticizers have the tendency to re‐crystallize (retrogradation) after being stored for a period of time. the thermal properties of glycerol‐plasticized starch are a function of water content [31. destructed. sugars) [30. et al  discovered that the elongation of break of the thermoplastic starch is significantly improved by plasticization with glycol. phase separation may still occur.
Medium DS starch acetates (0. Solutions to improve the properties of TPS are blending. resulting in an increase of the Young’s modulus and tensile strength of the products and a decrease in elongation at break. The DS is defined as the moles of substituents per mole of AHG units . and ethylenic compounds.01‐0. The esterification reaction is catalyzed by H+ and proceeds with the formation of water . Three different types of starch acetates may be distinguished. epoxy. or coating with hydrophobic polymers [33. The cross‐linked products are therefore more rigid (and less elastic) materials than virgin starch .3. These hydroxyls can react easily with a wide range of compounds such as acid anhydrides. Low DS products (0. Substitution of the hydroxyl groups of starch with acetate groups makes the esters more hydrophobic than native starch. and particularly with acetic acid . The products are called cross‐linked starches. paper. 39]. textile.3. leading to higher viscosities [42‐43]. while highly substituted starch acetate (DS of 2‐3) are soluble in organic solvents. The cross‐linked starches have found many applications. The most popular starch ester is starch acetate . and adhesive industry. The simplest starch ester is starch formate (C1). especially in the food. Cross‐linking results in a reduction of the solubility in water and to thickening. organic chloro‐compounds. High DS starch acetate can be easily casted 11 . Starch Esters The development of starch esters started in the mid 19th century [44. Breakdown of the starch and the formation of low molecular weight products occur to a significant extent . aldehydes. Introduction washed out by water . synthesized by direct addition of formic acid to starch at room temperature. Silva.2. the use of formyl esters of starch is much lower than that of acetate esters. et al  studied the mechanical properties of films from maize starch cross‐linked by sodium trimetaphosphate (SMTP). Cross‐linked starch An anhydroglucose molecule of starch contains two secondary and one primary hydroxyl group. Most studies dealt with the synthesis of starch esters of C1‐C4 carboxylic acids. 46]. Schuetzenberger acetylated starch with acetic anhydride . As a result. Chemicals of these classes having two or more of the reactive groups may react with two or more hydroxyls of the starch molecules.3‐1) are still soluble in water. 1. 1. Already in 1865. differing in Degree of Substitution (DS). Higher processing temperatures generally led to higher cross‐linking levels.2) are commercially available and used as food additives and in the textile industry.3.
Esterification may be performed using acid anhydrides in aqueous media or organic solvents (pyridine. or isopropanol) with acidic (hydrochloric or sulfuric acid) or alkaline (NaOH or triethylamine) catalysts . DMSO. Starch triacetate has been successfully synthesized using acetic anhydride in combination with pyridine‐gelatinized starch . Neutral and weak acid/ alkaline catalysts result in regioselective substitution at the C2 hydroxyl groups of starch.1. Several synthetic routes have been developed for starch esters. Mechanism of starch acetylation using vinyl acetate 12 . Starch esters have also been synthesized using alkanoyl chlorides [49‐53]. while alkaline catalysts (such as carbonate. xylene. acetate. Another attractive route involves the use of vinyl esters as reagents .Chapter 1 into films using organic solvents. The kinetics of the reaction between gelatinized aqueous potato starch and vinyl acetate was studied by De Graaf.1. High DS starch acetates are thermoplastic materials suitable to be used as biodegradable plastics . hydrogencarbonate. and phosphate) will result in C‐2. allowing the polymer to be produced with a range of hydrophobicities . The degree of acetylation of starch acetates can be easily controlled. et al [55‐56]. 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. but also C‐6 and C‐ 3 substitution . and is shown in Scheme 1. DMF. 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. However. it is frequently used in order to chemically graft the polymerized component on the other one (polystyrene on polybutadiene in our example above ). in order to achieve also a good molecular adhesion between the phases by melt mixing. 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 . compatibilizers) might be used. one component will be dispersed into the other. Starch ‐ Biopolymer Blends and Graft Co‐Polymers Blending of different polymers is an established method to obtain products with improved properties. in the simplest case of a binary blend. octadecanoyl) chlorides [49‐53]. The fatty acid chloride reactants are. As a result. However.4. dodecanoyl. relatively expensive and rather corrosive. a chemical reaction might take place upon processing (reactive extrusion).61) products could be obtained using this approach. 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.e. By working at temperatures above the melting point or glass transition temperature of the two components. the two components are chemically linked to each other. Introduction The synthesis of long chain fatty acid ester of starch has attracted much interest lately. polymers are rarely miscible with each other  so that. 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 . however.g. interfacial agents (e. but only relatively low‐DS (0. Fatty‐acid starch esters have been synthesized using fatty acid (octanoyl. 1. availability of mixing equipment and no necessity to use any organic solvent (often employed for the in situ polymerization). Their role is comparable to the one of a surfactant in emulsion formation . the latter are mixed together. which in turn provide a very strong adhesion at the molecular level between the dispersed phase and the matrix. The first involves simple melt mixing of the two components for example by extrusion. melt‐mixing remains the preferred route to polymer blends mainly because of very practical reasons: low costs. Although the in situ polymerization process is not as technologically straightforward and economically convenient as melt‐mixing.34‐0.3. they locate themselves at the interface between the 13 . The introduction of longer acid chain is expected to reduce the brittleness of virgin starch and to increase its hydrophobicity . 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 . Two main methodologies are applied for the production of polymer blends. If the right combination of chemical groups on the two components is present along the polymers backbone. i.
The starch/polyethylene blends are used for agricultural mulch [27.6. 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. 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.3. 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. In the past. The synthesis of compatibilizer and its use for starch/ synthetic polymer blending has also been studied.6. Illustration of the role of interfacial agent in compatibilizing blends 1. Matrix Dispersed Phase Figure 1. The most often used synthetic polymer for blending with starch is polyethylene [27. The products have been synthesized in the lab as well as on industrial scale [32. 39. there have been efforts to blend as well as to graft synthetic polymers onto starch. Starch‐based blends by melt mixing Starch‐synthetic polymer blending has been studied as early as in 1973 [64‐65].).4. 64‐98]. 66‐81]. 80‐ 81] or food packaging . The uncompatibilized blends of starch and 14 .1.Chapter 1 two components stabilizing the dispersion (most probably by a steric repulsion mechanism ) and providing an improved adhesion at the interface.
the uncompatibilized blends of starch with PCL give coarse phase separation and a reduction in the mechanical properties when the starch content is increased. i. and poly‐lactones (such as poly‐butyrolactone. poly‐valerolactone. thermal and water/gas permeability properties that are even comparable to bulk non‐biodegradable polymers such as PE and PP. PCL‐g‐pyromellitic anhydride [94‐95]. The mechanical properties of the blend are also improved. Some of these polyesters also have very good mechanical. 15 . The ester bonds of these polymers are susceptible to attack by water and this leads to enhanced biodegradability. 39. Polycaprolactone (PCL) is a well‐known synthetic biodegradable polyester. and poly‐caprolactone). 84‐98]. thus improving the dispersion of starch in the PE matrix. which both can react in situ with the hydroxyl groups of starch [70. while PE‐g‐GMA posseses epoxide groups. These biodegradable polyesters will finally decompose into non‐toxic products . among which synthetic polyesters are considered very promising materials . Examples of these biodegradable polyesters are poly‐glycolide. poly‐dioxanone. The tensile strength of the uncompatibilized blends is drastically reduced when starch content is increased. 96]. since the conventional synthetic polymers are usually poorly or non‐biodegradable. 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.e. dextran‐g‐PCL . To obtain completely biodegradable products. is formed upon mixing. PCL‐g‐maleic anhydride (PCL‐g‐MA) [91. 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. The use of conventional synthetic polymer such as PE for blending with starch will only result in a partially biodegradable material. Despite these good results. 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. and EVOH . The use of reactive compatibilizer precursors PCL‐g‐glycidyl methacrylate (PCL‐g‐GMA) [89‐90]. poly‐lactides. Introduction polyethylene show a coarse phase separation due to differences in polarity of starch (hydrophilic) and polyethylene (hydrophobic). As a result a graft copolymer (PE‐g‐Starch). PE‐g‐MA contains reactive anhydride sites. which combines excellent biodegradability with acceptable mechanical properties. The mechanical properties of these blends (tensile strength and elongation at break) decrease at higher starch content. the compatibilizer. while the tensile strength of the compatibilized ones decreases only slightly with the starch content. synthetic biodegradable polymers have been applied. 72‐73]. As is the case of blends with PE. Studies on the blending of starch with PCL have been already described [32.
2.3. thus resulting eventually in low GE values . which however is extremely air‐ and water‐sensitive. leading to formation of PCL homopolymers rather that starch‐g‐PCL. the hydroxyl groups of starch are supposed to function as initiating sites. Three routes have been studied in detail: starch esterification.Chapter 1 1.3. The water in starch granules competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction.4. Chapter 2 describes the ROP of p‐dioxanone initiated by a protected monosaccharide (1. defined as the percentage of grafted polyester to starch compared to the total amount of homopolymer and grafted polyester) . 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. In the past. A new strategy for the in situ ROP of ε−caprolactone on starch with the use of common ROP catalyst is therefore highly desirable. and the presence of liquid‐solid reaction system leads to reduced reaction rates between starch and ε−caprolactone. the synthesis of starch‐g‐PCL is reported. The method applied basically consists of three steps (temporary partial protection of starch‐hydroxyl 16 . 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 . Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF).2. 1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) using i Al(O Pr)3 as the catalyst.4. The highest GE value (up to 90%) has been achieved when using triethylaluminium as catalyst [97‐98]. Previous studies showed that common ROP catalysts such as tin octoate or aluminium isopropoxide gave low (0‐14%) grafting efficiencies (GE. a very flammable by‐product. In this reaction. starch/ biopolymer blending. Another approach consists of the Ring Opening Polymerization (ROP) of ε‐caprolactone monomer in the presence of starch. Thesis Outline The objective of this thesis is to study synthetic pathways to obtain starch derivatives with the potential to be used as bioplastic. In Chapter 3. starch‐g‐PCL has been synthesized using toxic materials such as isocyanates . The results of this study have been used as input for the synthesis of the starch‐g‐PCL. and starch‐g‐ biopolymer formation. and difficult to handle since it releases ethane. Starch is a very hydrophilic material that always contains moisture. during the reaction. Another possible cause for the low GE values is the heterogenous nature of the reaction.
Website of International Starch Institute. . 105‐122. Mol. 2004. C. New Jersey. Inc. Passchall). Boca Raton.N.B. York. Bemiller. Academic Press. Holmes: Interactive European Network for Industrial Crops and their Applications. a systematic experimental study on the effect of the process variables on the starch ester DS has been performed.. Corn Refiners Association. Aarhus. A preliminary study on the effect of several process variables (reactant ratio.C. Denmark: http://www. . Young: Fractionation of starch. C. de Bragança. 1984.M. 1. Wurzburg: Introduction. E.H. Gywnedd.html. together with the mechanical and thermal properties of the synthesized starch esters are reported in Chapter 5. 2008. 1995.B. Nawrath. and polyhydroxyalkanoates. A. E. J.starch. Bangor. UK. Introduction by trimethylsilyl groups. References .dk/isi/stat/rawmaterial. Sand Hutton. accessed on August 01. Whistler. The model. Princeton University Press. 2007 17 . The Biocomposites Centre. . in Starch: Chemistry and Technology (Ed. ROP of ε−CL on the remaining starch‐hydroxyl groups. University of Wales. Summary Report for the European Union. Inc. An Introduction to the New Science of Biodegradable Plastics. 1986. . Agricultural and Rural Strategy Group. O. 2002. Y. Poirier. and application of different catalysts) on starch esterification using vinyl fatty esters (vinyl laurate and vinyl stearate) is provided in Chapter 4. 2000‐2005. .. Wurzburg).A.L. CRC Press. UK. . Based on these preliminary results. The results were quantified using a statistical model. Somerville: Mini review: Plant polymers for biodegradable plastics: cellulose. addition of co‐solvent. UK. and removal of the silyl groups). . Fowler: Industrial markets for starch.. 2005. O. Corn. R.S. USA. are also reported. USA.F. P. Washington D. Stevens: Green Plastics. C.5. The synthesis of two interesting compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) for starch/PCL blends is discussed in Chapter 6. London. 1. The use of these compatibilizers in starch/PCL blends. Breed. Part of A Global Economy. in Modified Starches: Properties and Uses (Ed. including the mechanical and thermal properties. starch. R. Central Science Laboratory. USA. Corn Refiners Annual Report 2007.
Bastioli: Global status of the production of biobased packaging materials. G. Seidel).. J. Hanser Publishers. Wolf. and Regulations. . A.gov/garbage/plastic. Brussels. Chandra. USA. Hernandez. 1984. The Royal Veterinary and Agricultural University. A. G. Prog. Properties. Husing. X. 2000. Bastioli: Global status of the production of biobased packaging materials. . Angerer: Technical Report Series. 2008 . J. 2008.W. O.L. C. Ind. Department of Dairy and Food Science. A. R. S. Washington DC. the Food Biopack Conference Proceeding (Ed. 1990. Whistler. USA. Inc. 18 . Selke: Packaging and the Environment. Rustgi: Biodegradable polymers. Daniel: Starch. Marscheider‐Weidemann. Plastics Europe. Starch/ Starke 2001. Passchall).J.M. in Kirk Othmer Encyclopedia of Chemical Technology Volume 22. Schleich. Chem. Hoboken. Lancaster. . New Jersey. S. Germany.L. Inc. . 2000. . .. 11. Institute for Prospective Technological Studies and European Science and Technology Observatory. R. M. Prentice‐Hall. Strong: Plastics: Materials and Processing. U.R. Weith: Plastics.A.M.D. Englewood Cliffs. UK. R. ..F. Watson: Corn and sorghum starches: production.J. 1939. Applications. Cleaner Prod. Patel. Munich.N. 53. S. 1273‐1335 . 1996. . Techno‐economic Feasibility of Large‐Scale Production of Bio‐based Polymers in Europe. Selke. accessed on August 19. in Starch: Chemistry and Technology (Ed. Applications. Germany. Ren: Biodegradable plastics: a solution or challenge?. 105‐109 . Sci. Academic Press. B. London. USA: http://www. Fritz. J. demand and recovery for 2006 in Europe.htm. 2003.. J. C. The compelling facts about plastics: An analysis of plastics production. 31. Fifth Edition (Ed. USA.B. Crank. R. 557‐562. R. . C. Inc. R. Association of Plastics Manufacturers.epa. J. Frederiksberg. Properties. 27‐40. E. Belgium. Polym. . 1998. 53 . . Whistler. Ehrenstein: Polymeric Materials: Structure. Inc. Technomic Publishing Company. M. Bemiller. Website of Environmental Protection Agency. 23. 2005.E. J.351‐355. John Wiley & Sons. Hanser Publishers. Denmark. Processing. Munich.Chapter 1 . Link. F. 2000. Culter: Plastics Packaging. Weber).E.J. Eng. Braun: Environmental impacts of biobased/ biodegradable packaging. Starch/ Starke 2001.
2003. Polym. properties. July 15. Nielsen. Carbohydr. Carbohydr. J. polylactate and polyhydroxybutyrate. 592‐595. Starch/ Stärke 2001. T. Wang. Martin. Prod. K.Lin: Note. Part C: Polym. Westhoff. F. Ind. 409–416 . Kruiskamp. 1996. 275‐282 19 . Curvelo. status and perspectives. J. .A. and Pukanszky: Thermoplastic starch/layered silicate composites: structure. Res.H.G. Bagdi. Frederiksberg C. Physical and mechanical properties of biobased materials – starch. J. P. J.L. 1491‐1494 . 1997. J. 53. 372–380 . K. Starch/ Starke 2001. Avérous. M. A. and Couturier: Properties of biodegradable multilayer films based on plasticized wheat starch. 2005.B. A.J.M. epoxy functions and stearoyl chloride. van Soest. A. Polymer Sci. J. A Real Case of “Force Majeure” that should be passed through to Packaging Prices Immediately. 57. 53. Macromol. 43. Dev. Yang. Department of Dairy and Food Science. 2005. Sci. Gandini: Surface chemical modification of thermoplastic starch: reactions with isocyanates. Chem. C. 1980. 331–336 .S.H. Polym. Interfaces 2006. . N. Spain. a model for thermoplastic starch.M. Vliegenthart: Interaction between dry starch and plasticisers glycerol or ethylene glycol. Forssella. Compos.K. Biodegradable thermoplastic starch. A. Yu. X. W. 385–409 . 4. Denmark. 494–504 . the Royal Veterinary and Agricultural University.G. and Olsen. Weber). 2005. E. Schwach.J. L. 2008. Chambre Syndicale des Emballages en Matiere Plastique (French Plastic and Flexible Packaging Association). Y. Crops Prod. Otey. P. . Yu. Press Release: Major Price Hikes in Plastics. Eng. . 62. Ind. 13. P. report of Food Biopack Project (Ed. Starch/Stärke. P.F. Introduction European Commission Directorate General of Joint Research Centre. Biobased packaging materials for the food industry. 34. 2000. Y.. R. interaction.. Carvalho. Mikkilti. B.M. 21. Appl. X. Smits. Moates. 2003. J.P. Ma: The effects of citric acid on the properties of thermoplastic starch plasticized by glycerol. K. 1–17 . Doane: Starch‐based blown films. R. Rev. G. Petersen. of concentrated barley starch‐glycerol‐water mixtures. J. O. Seville. J. Wang. Muller.F. Parker: Phase and glass transition behaviour. Gao. 356‐361 . measured by differential scanning calorimetry and solid state NMR spectroscopy.V. Wang: Properties of starch blends with biodegradable polymers..M. 53.
Crops Prod. Y.D. Polym. S.W. 58 (9). .A. . C. Journal of Applied Polymer Science 1995.B. Chem. Tserki. E. J.P. Ibezim. 1986. Andrade: Reactive processing and mechanical properties of cross‐linked maize starch.A. Ind. S. Feil: Morphology and enzymatic degradation of thermoplastic starch– polycaprolactone blends. Qiao. Pilin.A. E. K. 65. Gattin. Panayiotou: Processing and characterization of starch/polycaprolactone products. Prinos. O.Chapter 1 . I.W.D. Copinet. Eng. Feller. Borredon. 58. Ramsay. 35. Favis: High performance LDPE/ thermoplastic starch blends: a sustainable alternative to pure polyethylene. Boca Raton. 1209‐1217. Stab. Starch/Stärke 2005. Polym. 381‐384 .J. CRC Press. Jiang.. Merrill: Properties of fatty‐acid esters of starch. Polym. Sun: Reinforced thermoplastic acetylated starch with layered silicates. 11–17. USA. Grohens: Starch modifications. 139–147 . . Ribeiro.J. 2001. 55‐78. J. Mullen. O. Alric. R. A. Rodriguez‐Gonzalez. V. 2004. destructuration and hydrolysis during O‐formylation. E. in Modified Starches: Properties and Uses (Ed.A. Levesque.C. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Sci.B. Divers. Favis: Rheological and thermal properties of thermoplastic with high glycerol content. C.B. Silva. Vikman. I. D. W. Mullen.H. G.T. 1943. 2002. USA. A. Thiebaud. H. J.W. 74. Jarowenko: Acetylated starch and miscellanous organic esters. Bertrand. Polymer 2003. 2006. 46–51 . Hulleman. CRC Press. T. P. F.C. Myllarinen. Coutuier: Comparative biodegradation study of starch‐ and polylactic acid‐based materials. Aburto. (editor). Y. M. Sagar. B. Polym. P. Environ. B. A. 1999. J. 1647‐1656. Ind. C. B.. 1997. van der Zee. O. . Ind. Ramsay. 34.D. Wurzburg. 1986. Pacsu: Starch studies: Preparation and properties of starch triesters. 44. 2004. 24. C. T. 57. B. Carvalho. 2594–2604 . 1942. Chem.W. E. F. chapter 4 in book Modified Starches: Properties and Uses. 20 . Appl. 1. Matzinos. M. M. Wurzburg: Cross‐linked starches. X. 389‐398 . J. Carbohydr. Polym. . Rodriguez‐Gonzalez.F. J. Pacsu: Starch studies: Possible industrial utilization of starch esters. Appl. Sci. 77. C. Kontoyiannis. 17–24 . Boca Raton. 581–586 . Wurzburg). E. Inc. J. Inc. Bikiaris. 56. Eng. J.D. 705‐721. 1517–1526 . W. Degrad.
E. A straight way to regioselectively functionalized polysaccharide esters.Fennel Evans: Fundamental of Interfacial Engineering. . 175.Coleman. A. Polym. .. Polym. E. 1991. Bakker.828. 56. R. Hu). Hu. R. L. . Hanser Publishers. Sci. Inc. 132‐135. A. C. Alric. A. de Graaf. Aburto. A. S. Alric. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. Bikiaris. Guegan. P. Broekroelofs. J. 101‐112. J. 5590‐5598 . E. Bikiaris.Graf. P. Wiley‐VCH.W. B. 2001. D. H. Smith. de Graaf: Ph. 34. Carbohydr. C. P. Degee. I. E.E. . Appl.K. Inoue: Compatibilizers for melt blending: Premade block copolymers. . Nakayama. 33‐43 21 . Thiebaud. Dicke. 28. Alric. Borredon. Introduction.D. 1996 . and biodegradability of fatty‐acid esters of amylose and starch. I. Mormann. Thesis. 145‐152. Aburto. Rijksuniversiteit Groningen. I. C. M. Cellulose 2004. P. A. Thiebaud. M. Macromol. Introduction . Marechal.Stokes. Panayiotou: Synthesis. G.V. J. Starch‐Starke 1999.F. 137‐144. P. Polym. Alric.E. 29. Senocq. J.H. D. Aburto. C. W. A. 51.Painter: Specific Interactions and the Miscibility of Polymer Blends. 302‐307. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. . S. I. Mouysset‐Baziard. Munich.M. F. D. 1960 . T. Starch‐Starke 1999. Symp. US Patent 2.E. Tuschoff: Acylation of hydroxy compounds with vinyl esters. 1440‐1451. E. Alric. Scott. Macosko. A. Beenackers: The kinetics of the acetylation of gelatinised potato starch. J. 1999. Bakker and G. Aburto. Duquesne. Dubois: Synthesis and characterization of compatibilized poly(ε‐caprolactone)/ granular starch composites. W. 255‐263.H. P. Germany. D. Carbohydr. M.E. J. C. . J. A. USA. characterization. C. J. Hamaili. . J. Rutot.J. 118‐121. Panayiotou: Properties of octanoated starch and its blends with polyethylene. 1997.928. 51. . . 11. I. R. Borredon. R. Aburto. G. Prinos. 1997. Starch‐Starke 2004. 74. Khandpur. Technomic Publishing Company. in Reactive Polymer Blending (Ed. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Prinos. E. M. Al‐Higari: Acylation of starch with vinyl acetate in water. Macromolecules 1996. J. 2001. Janssen. W. .C. 1995. 57. New York. G. Starch‐ Starke 2005. Lancaster. USA.
Technol. E. Polymer. Z. 2003. C. G. 59. Mater. 1998. Frangis. I.H. 59. H. J. Z. H. J. 75–81 . U. Danjaji. Polym.S. A. Rosa: Mechanical. Eur. Polym. Nikolov. A. B. Eng. Bikiaris. Degradation studies and moisture uptake of sago‐starch‐ filled linear low‐density polyethylene composites. 1998. Liu. 1242‐1251 . Adv. 88. 2005. J.D. Tech.A. J. 287‐291 . J. .. Ramsay.D. 2005. St‐Pierre. Mani. Cordi. Z. Appl. Testing. 647‐655 22 . 21.Chapter 1 . Vouroutzis. L. 2001. and tensile properties of Low‐Density Polyethylene (LDPE) and corn starch blends. J. R.A. Stab. thermal and morphological properties of virgin and reprocessed LDPE/corn starch blends. Sun: Effects of polyethylene‐grafted maleic anhydride (PE‐g‐MA) on thermal properties.G. D. Pavlidou. K. Danjaji. Jane.M. M. Res.‐J. N. H. Gelina. 1841‐1846 . L. Wang. 2003. Mohd Ishak: Material properties. morphology. Mohd Ishak: Material behaviour. 34. D. L. Nawang. Y. W. 70. . 2005. 2904–2911. Contat‐Rodrigo. Polym. Rosa: Effects of the compatibilizer PE‐g‐GMA on the mechanical. R. D.S. L. 162–163. Pedroso.S. 88. Ramsay. Sci. Favis. Pedroso. B. Z. 310–317 . D. Prinos. 1‐9 . Polym. Testing. Panayiotou: LDPE/Starch blends compatibilized with PE‐g‐ MA copolymers. Carbohydr.D. Duran. 30. E. Ismail. R. J. Appl. Polym. 20. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends. Ismail. 2002. Almeida. N. III. thermal and morphological characterization of recycled LDPE/corn starch blends. Sci. Polvm. J. 16.L. Polym. Ribes‐Greus: Thermal and viscoelastic properties of some commercial starch products. Nakamura. U. Effect of compounding and starch modification on properties of starch‐filled Low‐ Density Polyethylene. Chem. Bikiaris. Panayiotou: LDPE/plasticized starch blends containing PE‐g‐MA copolymer as compatibilizer.A. R.J. Polym. Polym.A. Ishiaku. 1998. Evangelista.G. 167–172 . R. Mei: Study and development of LDPE/starch partially biodegradable compounds. 1503–1521 . C. Koutsopoulos. Sung. Nawang. 1991. 1997. Degr. Mechanical properties of sago starch‐filled linear low density polyethylene (LLDPE) composites. 38. W.M.S.I. 1467‐1475.G. Ishiaku. Effect of amylopectin to amylose ratio in starch. A. N. Proc. Verhoogt: Processing and characterization of thermoplastic starch/polyethylene blends.S. N. Appl. 236–241 . I. Sci. Ind.
3539–3546 . Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. 139‐161. B. S.N. J. K. O. 32–45 . M.M. J. 35. 1139–1148. Calil: Effect of thermal aging on the biodegradation of PCL.M. R. Averous. 2007. Appl. H. D. Polym. .P. Cho. Jung. C. Raj. Karlsson: Polymers from Renewable Resources. M. A‐C.. Azemi. . 43. 1997. Kim. O. Chem. Liu. Introduction . J. S. and their blends with starch in soil compost. Garg.. Moro. Doane: Starch‐based blown films..M. Ryner.K. Polym. 157. J. A. Azemi: Effect of starch predrying on the mechanical properties of starch/poly(e‐caprolactone) composites. M.Y. Adv. Dole.S. . Appl. 2003. 81.H. F. Albertsson. C. A. 179‐187. Sen. B. J. Polym. Sci. 41. Sci. R.K. 23. Jana: Studies on the properties and characteristics of starch– LDPE blend films using cross‐linked. Sci. 1298‐1305. PHB‐V. 267. 23. 157. U. B.‐C. 1999. Manan. Manan. Stridsberg. Eng.S. Eng. J. Yang. Otey. M. 40. C. Chem. Odusanya. 2001. 2. Ishiaku. Polym. Sankar. B. Kim.H. Sci. E. 2002. Ang. N. in Polym. K. Eur. Res.M. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate.R. 877–884. S. 42‐65.S. 1987. J. Low density polyethylene/ starch blend films for food packaging applications. Bhattacharya Residual stresses and density gradient in injection molded starch/synthetic polymer blends Polymer 2000. P. Jana. Y. Dev. Maiti: A biodegradable film. Kim. 9177–9190 . 26. Z. Otey. Res. 12. Sci.D. 2002. Guedes. U. Ind.W. . Tech. J.H. Appl. K. Sci. T. dos Santos Rosa. S. A. Westhoff. glycerol modified. 89. L. M. S.G. 4157–4167 . Part A: Polym. Tang. Odusanya. Rodrigues. cross‐linked and glycerol modified starch. J.K. Polym.M. 1659‐1663 . 2004. 2000. 23 . . Preliminary diffusion evaluation. Ind. J. Albertsson: Controlled Ring‐Opening Polymerization: Polymers with designed Macromolecular Architecture. Eng. 1984.P. 2004. Polym. . T. 16–19 . Park: Modification of aliphatic polyesters and their reactive blends with starch.H. 3976–3987 . W. Adv. 41. Polym. 87. Lindblad. Environ. 1507– 1516. Chem. Polym. J. Prod. Adv. Ranucci. Ishiaku. Sci. 284‐287 . John. A. Chem. Westhoff: Starch‐Based Films. C. Kammer: On mechanical properties of sago starch/poly(ε‐caprolactone) composites. Makromol. Pilot plant investigation for production of the biodegradable film. Polym. F. Polymer 2000.S. M. Fringant: Properties of thermoplastic blends: starch–polycaprolactone. U. 2003. D. Siddaramaiah. 1. L. M.C.F.
M. 1998. R. Duquesne. M. Polym. P. Narayan. . 41. 3875–3881 . P. . Stab. Sci. Mani. Errico. 80. R. E. Bhattacharya: Synthesis and characterization of starch‐ graft‐polycaprolactone as compatibilizer for starch/polycaprolactone blends. 175. Krishnan. E. Avella. M. P. 2002. Degée. Degrad. 83. Tang. 19. P. Sadocco: Preparation of biodegradable polyesters/high‐amylose‐starch composites by reactive blending and their characterization. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. M.S. 2000. Rutot. R. P. Rimedio. . Laurienzo. Composite Interfaces. 33–43 . Raimo. Rimedio: Polymer communication: Preparation and characterisation of compatibilised polycaprolactone/ starch composites. 24 . Degee. R. D. Dubois: Aliphatic polyester‐grafted starch composites by in situ ring opening polymerization. Dubois: Synthesis and characterization of compatibilized poly (e‐caprolactone)/ granular starch composites. Polymer 1999. Martuscelli. 2001. R. 3091‐3100. Dubois. 7. Errico. J. Rapid Commun. 1432–1442. Macromol. M. M. 40.E. 215–225. 2003. M. Avella. Macromol. J.E. C. Rutot. Appl. Polymer 2000. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. P. . and P.Chapter 1 . Symp. 127–134. D. 283–286. Polym.
Keywords: biodegradable.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose)/Al(OiPr)3 initiator system to yield polydioxanone with a protected monosaccharide end‐ group is described. and MALDI‐TOF mass spectrometry. Besides the desired polydioxanone functionalised with a monosaccharide end‐group. 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. 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).3. ring‐opening polymerization .2. The products were synthesized at 60‐100°C and characterized by 1H‐ and 13C‐NMR. polyesters. with higher temperatures (100°C) leading to lower yields. Systematic studies showed that the polymer yield is a function of the reaction temperature and the reaction time. also polydioxanone with an OiPr end‐group was formed (20‐30 %).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.
whereas the high temperature for polylactides results in thermal degradation and undesirable transfer reactions during synthesis and processing steps . testosterone.1. pregnenolone. Several polymers with bioactive alcoholic and phenolic end‐groups of interest for drug‐related applications were synthesized (for example geraniol. stigmasterol and ergocalciferol)  and also with protected monosaccharides (galactopyranose/ glucofuranose) [6‐7]. 9]. This chapter describes experimental studies on the catalytic ROP of p‐ dioxanone in the presence of a protected glucose molecule (1. which to the best of our knowledge is the first study to functionalise polydioxanone with a monosaccharide. The ring‐opening polymerization (ROP) of p‐dioxanone using metal catalysts such as aluminum isopropoxide [Al(OiPr)3]. Besides 26 . Its melting temperature is close to 110 oC. 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. has interesting product properties compared to other aliphatic polyesters. polyglycolide. The relatively low melting point of polycaprolactone limits its applicability.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. The catalysts are metal alkoxides with Lewis acidic character [8. and is tougher than polylactides and even HDPE . Polydioxanone (poly(p‐dioxanone) or poly(1. which is a unique compromise between application and processing temperature.2. The synthesis of protected monosaccharide end‐capped biodegradable polymers is usually performed via Ring Opening Polymerization (ROP). accessible by the polymerization of p‐dioxanone (1).3 MPa and an elongation at break of 500‐600%. Introduction Aliphatic polyesters.g. It is fully degraded in the body within a period of 180 days . stannous octoate [Sn(Oct)2]. such as polycaprolactone. only limited information about the synthesis and properties of polydioxanone is available in the open literature. Polydioxanone has a tensile strength close to 48. particularly for nano‐ encapsulation systems for drug delivery . are interesting polymers because of their good product performance and biodegradability . or zinc lactate has been reported. quinine. Biodegradable aliphatic polyesters end‐capped with sugar molecules have been studied extensively for use in biomedical applications. reaction of the metal‐alkoxide with an appropriate alcohol allows the synthesis of end‐capped poly‐lactones. From a biodegradability point of view. This melting point is considerably higher than typically found for polycaprolactone (60 oC) and lower than that of polylactides (at least 175°C). polydioxanone also shows good performance. Exchange of the alkoxide group by e. tocopherol. Nishida et al  reported that polydioxanone decomposes to non‐toxic gases (CO2 and H2O) by microorganisms. probably because the p‐dioxanone monomer has become commercially available only recently . 2).4‐dioxan‐2‐ one)).Chapter 2 2. and polylactides.3. Despite its good properties.
Analytical grade dichloroethane (Labscan).1. Next.2.g.2. Toluene (Labscan) was dried and stored on molecular sieves 3 Å (Labscan) under nitrogen. 98+% (Aldrich) were used as received. To this solution.3.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2) 2 (1. poorly soluble heterogeneous systems. which was pre‐heated till about 60°C to obtain it in a liquid state. NMR) due to the poor solubility of the products in standard organic solvents. This stock solution was prepared by adding 4. 5.3.2. 3]. the results of this study are also of interest for the preparation of starch/polydioxanone polymers using ROP. Materials p‐Dioxanone monomer (1.2. 8 mmol). and diethylether (Labscan) were used as received. Germany) was purified according to the procedure described by Raquez et al [2. the synthetic pathways and the soluble. As such.8 ml of a solution of an aluminum isopropoxide stock solution was added.5 mmol) was dissolved in toluene (1 ml) at 50 oC. 97% (Sigma) and aluminum isopropoxide. The reaction mixture was clear and colorless during the reaction.2. hot dichloroethane (20‐25 ml) was added to completely dissolve the partly solid polymer at room temperature. CDCl3 was obtained from Sigma and was used as received. After the pre‐determined reaction time. However.44 g.2. grafting efficiencies are difficult to determine in this system by standard analytical techniques (e. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System potential applications in biomedical products. 0.8 g. Materials and Methods 2.2. Methods All polymerization experiments were carried out under a protective nitrogen atmosphere using standard Schlenk‐ and glovebox techniques.07 g. Subsequently. relatively low molecular weight compounds reported in this study may be viewed as model systems for more complex. The polymerization was allowed to proceed for 16 h at 100 oC. containing 0. Typical example for the synthesis of polydioxanone end‐capped with 1. (20 mmol) of aluminum isopropoxide to 20 ml of toluene. Particularly interesting in this field are starch polymers grafted with polydioxanone. The resulting clear solution was stirred for 2 h at 50 oC.08 mmol aluminum isopropoxide) was added to pure 1 (0. Boehringer Ingelheim. 2. the mixture was brought to room temperature and several drops of HCl (1 N) were added to stop the reaction.55 mmol of 2 and 0. 2. heptane (Acros). 1.1. 2. part of this solution (350 µl.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2). The hot solution was 27 .2.
) H‐NMR was applied to determine the experimental average degree of polymerization.2.Chapter 2 precipitated in a heptane/ ether mixture (300‐400 ml.3. 2.3. one initiated on a galactopyranose molecule and the other on an isopropoxide group. the above equation simplifies to Xn theo = monomer conversion × [monomer ]0 [ sugar ]0 (2. The white solid was finally separated from the liquid by decantation and dried in a vacuum oven (5 mbar.1. The isolated yield at this condition was 68%.2. X n exp of the product.) H-isopropoxide end group 28 . If the amount of sugar is in excess with respect to the aluminum catalyst. 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. As will be shown later. This leads to the following equation: Xn exp X n exp is = A A H-repeating unit H-galactopyranose end group +A (2. 40 oC) until constant weight. Processing of the raw data was performed using VNMR software. 4:1 by volume) at 4‐8 oC. Product Analyses NMR analyses were performed in CDCl3. 2D‐NMR spectra were recorded on a Varian Unity 500 NMR.) Here. Calculation of Average Degree of Polymerization The Theoretical Average Degree of Polymerization. MALDI‐TOF spectra were recorded on an Applied Biosystems Voyager DE‐PRO machine using dithranol/ NaI as the matrix (linear mode). it is assumed that all available initiator is used effectively. two polymers with different end‐groups are present. 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).4.2. 1H‐ and 13C‐NMR spectra were recorded on a Varian AMX 400 NMR.
6. Schematic representation of the polymerization reaction including atom numbering scheme 29 .3. R gp .2. 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. eq. The exchange reaction typically takes place at 50°C for 2 h.6 ppm A 5.3.7 ppm − 6 × A 5. 2. To avoid the formation of isopropoxide end groups.220.127.116.11. is rewritten as: Xn exp = [ A 3.6 ppm + A 5.) The ratio of the two different types of polymers (either end‐capped with galactopyranose or an isopropoxide group).4− 18.104.22.168‐di‐O‐isopropylidene‐α‐D‐ galactopyranose (2) and Al(OiPr)3 as the catalyst precursor (Scheme 2.1 ppm (2.1 ppm (2.). the protected mono‐saccharide 2 reacts with Al(OiPr)3 to form the actual catalyst for the polymerization reaction.6: 96 was applied. In the first step.5. To compensate for this effect. The polymerization reaction was performed as a two step process.) or.) 2.). is calculated using: R gp = A H − galactopyranose end group H − isopropoxide end group A (2. Typically. Screening Experiments Initial experiments to synthesize protected galactopyranose end‐capped polydioxanone (3) were performed using 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. 2. an Al ‐ monosaccharide 2 ‐ monomer 1 mol ratio of 1: 6. Results and Discussions 2.6 ppm ] / 6 A 5. an excess of monosaccharide 2 on Al was used (see eq. in term of the NMR resonances : R gp = A 5.1.
However.3.1 and 2.67..40 ppm) unit are broadened and shifted up to 0. Product analyses The products were analyzed using 1H‐ and 13C‐ NMR and MALDI‐TOF.1.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.) In the next step. together with the CH3 groups from the protecting groups of the sugar appear in the range 1. respectively. of interest is the presence of a small multiplet at about 5.15‐1. H‐NMR spectra (Figure 2. 2.1. The typical proton resonances of the p‐dioxanone (3. Although the spectra clearly indicate the presence of the monosaccharide 2.2.1 ppm compared to the monomer. Proton resonances of the CH3 group of the isopropoxide end‐group. although it is evident that p‐dioxanone polymerisation occurred.1 ppm. The OiPr : 2 end‐group ratio for the standard experiment was 1: 2. Typical isolated product yields are 68% at these conditions.51 ppm. It is difficult to determine the end‐groups of the polymer on the basis of 1H‐NMR.1.) are not particularly informative. Typical 1 H‐ and 13C‐NMR spectra of 3 are shown in Figure 2. the in‐situ formed catalyst was reacted with p‐dioxanone monomer 1 at 100 oC for 16 h.7.78 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). 1 30 . This multiplet is characteristic for the CH proton of an isopropoxide end‐group. The off‐white solid reaction product was collected after a dissolution‐reprecipitation process using dichloroethane and a heptane/ diethyl ether mixture. 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.
Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.2.2 ppm. p‐dioxanone monomer.52 ppm (β’).2 ppm in product 3.3 ppm are shifted considerably.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone. 13 31 . characteristic resonances of an ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end group are present at δ = 61.8. 1. H‐NMR Spectra of: (a).1. 1.3.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.3. 68.17 and 69. Carbon resonances arising from the monosaccharide end‐group are present between δ = 66. For instance.) is more informative and clearly shows the presence of a polydioxanone polymer backbone and a monosaccharide end group.2 ppm. 1 (c). C‐6 is shifted from δ = 62.28 ppm in 2 to δ 66. 2 (b). In addition to the mono‐saccharide end‐ group.2‐70.2.2 and 96. Particularly the C2‐C6 carbons in the range δ = 66.61 (γ’) and 73. 3 1 C‐NMR (Figure 2. The carbon resonances of the polydioxanone backbone are present at δ = 63.2.
1. The resonances of the protecting group of the sugar appears at δ = 108. C‐ NMR Spectra of: (a).2.73 and 109. 1. the numbering scheme of carbons and protons is given in Scheme 2.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone. 2 (b).97 ppm (‐CH‐(CH3)2).3. 2 32 .48 ppm (‐CH‐ (CH3)2) and between δ = 21.76‐25. 1 (c).2. D‐NMR (HSQC) was applied for complete peak assignment of the product.2. A typical example of a part of the 2D‐NMR spectra is given in Figure 2.3.1. 3 13 Carbon resonances of the isopropoxide groups from the by‐product (isopropoxide end‐capped polydioxanone) are present at δ = 68.1.76‐25.2. An overview of the data is given in Tables 2. p‐dioxanone monomer.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.Chapter 2 Figure 2. and 2..70 (>C‐(CH3)2) and between δ = 21.3.97 (>C‐(CH3)2).
31 4. around 1.59 4. 33 .16* 4.3.77 3.60 4.26 3.17 3.83 p‐dioxanone Monomer (1) H‐α H‐β H‐γ a 4.25 3.54 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 5.40 all values were determined using 1H NMR.09 4.1.32 * 4. HSQC 13C‐1H spectra of 3 Table 2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.50 4.732 broad.8 Reactant peaks (ppm) Galactopyranose (2) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 5. which were determined from HSQC spectrum due to overlapping resonances.68 4.00 4.301 3.78 4. except values with *.74 3.
2.09 65. For this purpose. A typical molecular weight distribution is observed. corresponding with an average number molecular weight (Mn) of about 1600.61 170.52 63.17 69.2 73.07 Reactant Peaks (ppm) Galactopyranose (2) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 96. 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 product obtained at screening conditions (16 h reaction time at 100°C) contains on average 13 monomer units.92 68.21 70.4.31 70. 34 .92 66.8. The molecular weight distribution of the major peaks may be represented by the following relation: m / z = 23 + 260 + 102n (2.80 61.00 171. The difference in molecular weight between the main peaks is 102 g/mol.2 68.58 68.57 62.11/ 170.75 70.21 NMR analyses also allow calculation of the molecular weight of the products.28 70. MALDI‐TOF was also applied to characterize the products. the ratio of the intensity of the end groups and the polymer backbone peaks is determined. The latter likely stems from the matrix used to ionize the sample.) This series represents a dioxanone polymer end capped with 2 and an additional Na ion.63 68. An example of a MALDI‐TOF spectrum of 3 recorded in a dithranol/NaI matrix is given in Figure 2.54 166.10 71.Chapter 2 Table 2.33 70. which is the molecular weight of a repeating dioxanone unit.28 p‐dioxanone monomer (1) C‐α C‐β C‐γ C‐carbonyl 61. A detailed procedure is given in the experimental section.
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. Furthermore.9. see Figure 2.) and (2.5.). ionized with either Na+ or H+.5. m / z = 23 + 18 + 102n m / z = 1 + 18 + 102n 35 (2. although both with a very low intensity (Figure 2.5 +102 1400 1600 1800 2000 2200 2400 Mass (m/z) Figure 2.) (2. Typical MALDI‐TOF spectra of 3 A second molecular weight distribution is also clearly visible when enlarging the spectra.11. in line with the NMR data.) This relation is indicative for the presence of a polydioxanone polymer containing an isopropoxide end‐group.10.4. This series may be described by the following relation: m / z = 23 + 60 + 102n (2. These distributions may be represented by equation (2.) . for details.10.11. two other distributions are present.) and imply the presence of dioxanone polymers with carboxylic end groups.
Most likely the matrix (dithranol/NaI).5.1 0 600 700 800 900 Mass (m/z) 1000 1100 1200 Figure 2. it proved not suitable to determine the average molecular weight of the products.6.13]. although successfully applied for galactopyranose‐end‐capped polycaprolactone . small but significant amounts (20‐30%‐mol) of polydioxanone chains with an OiPr end group are present. the catalyst precursor is treated with monosaccharide 2 resulting in an alcohol exchange 36 . Various samples with.3 10 836.2.Chapter 2 50 40 792. 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. but in all cases poor quality. according to NMR.g. different molecular weights were analysed. Various other matrices were tested (e.7 796. 2. is not particularly suitable for polydioxanone. 2‐(4′‐ hydroxybenzeneazo)benzoic acid (HABA)). Mechanistic aspects Both NMR and MALDI‐TOF measurements imply that the main product is indeed the desired monosaccharide end‐capped polydioxanone. Enlarged MALDI‐TOF spectra for 3 Although MALDI‐TOF clearly demonstrated the presence of various types of end‐groups in the product. [6. In the first step. low resolution spectra were obtained.1. In addition.3.7 %-Intensity 30 20 858. The observed differences in the molecular weight distributions of the various samples were only marginal.
Subsequent coordination and insertion of dioxanone molecules leads to the formation of a polymer chain with a monosaccharide end‐group. termination of the chain growth may occur by reaction with an alcohol. (R = monosaccharide 2) 37 . a p‐dioxanone molecule will coordinate to the Lewis acidic aluminium center followed by an insertion step. Next. The termination reaction is known to be reversible and the formed polymer may again react with an aluminum center and continue to grow . 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 resulting Al‐alkoxide may again initiate a polymerisation reaction. 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. Irreversible termination of the polymerisation is performed at the end of reaction period by adding dilute acid to the polymerisation mixture. During the reaction.6. Simplified reaction scheme for the ROP of p‐dioxanone catalyzed by Al(OiPr)3. The termination reactions lead to the formation of ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end groups.
2.5 1.1 2.04 19.Chapter 2 The minor product.36 17.12 42.55 Product Properties Avg. A total of 15 experiments were performed at two polymerization temperatures (60 and 100 oC).10 15.5 92. 2. two different reaction times (1.78 10.50 35.3 3.3.6 6. 2. OiPr‐end‐capped poly(p‐dioxanone).6 96.5 67.53 59.6 6.5 1.08 30. the end group may also be formed by a termination reaction with free isopropanol. An overview of the experiments and the results are given in Table 2.6 was applied.9 2.9 1 7.4 2.3 and 62.6 89. 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 .8 3.6 6.74 18. a (nearly) fixed Al(OiPr)3 : monosaccharide ratio of 1 : 6. degree of polymerisation of the product and the end group distribution was determined.16 46. In all cases.3. These may be present in the reaction mixture because the exchange reaction between Al(OiPr)3 and 2 was incomplete.5 T (°C) 100 100 100 100 100 60 60 60 60 60 60 60 60 60 60 dioxanone/ 2 ratio (mol/mol) 3.5 2.5 1.29 14.81 16.5 hrs and 16 hrs) and a p‐dioxanone to monosaccharide 2 molar ratio ranging between 3.6 81.8 2. However.49 13.93 37.3.3 6.3 6.4 2/OiPr ratio Rgp (mol/mol) 2.0 87.1 86.6 2.7. Systematic Studies The effect of important process variables (temperature.5.7 2. Chain Length (Xn exp ) Yieldb (%) 30.21 54.45 19.6 91.22 13.5 84.6 6.1 2.2 91.70 58.22 16.3 81.3 6.50 25.74 52.30 62.6 6.3‐6.96 2 3 a For each experiment.7. Table 2.6 86.2.5 1.3 6.85 14.11 22.6 6.3 6.3 6. 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 t (h) 16 16 16 16 16 16 16 16 16 16 1.1 3. despite the excess of 2.6 6.93 8.8 2.). will be formed when the polymerization starts with an aluminum alkoxide with a remaining Al‐OiPr group (eq.6 80.22 57.4 2. formed in the first step of the polymerisation reaction (eq.4 3.28 33.).74 23.5 81.6 6. time and the mol ratio of monomer to monosaccharide) on the yield.
2. For this particular experiment. will be required. Product yield The isolated yields of the reactions are all but one between 67 and 96% (see Table 2. which is expected to be more pronounced at lower temperatures. 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. the yield at 1. et al .). 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.2.). The latter is due to the slight exothermicity of the reaction . 39 .5 h. Evidently. A more detailed analysis on the actual nature and composition of the polymerization products.5 and 16 h reaction time). when the reaction is performed at a low dioxanone: 2 ratio (≅ 16).3. To the best of our knowledge. as suggested by Raquez et al . 1. For instance. leading to a low average chain length of the dioxanone polymer (7. and particularly when comparing the data in set 2 and 3 (60oC. the yields are higher when performing the reaction at 16 h reaction times. see Table 2. equilibrium conversion is not yet achieved within 1. the equilibrium conversion of dioxanone at 80oC was about 80%. It is likely that this relatively low molecular weight compound dissolves partly during isolation/ purification procedure leading to lower isolated yields. a low dioxanone: monosaccharide 2 ratio was applied. 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.3. The effect of the reaction time on the product yield may be derived from the data provided in Table 2. the yields for reactions conducted at 60 oC are always higher than those performed at 100 oC. the monomer conversion reaches a maximum value before going down to the equilibrium monomer conversion. For the two experiments with a dioxanone: 2 ratio higher than 16 (≅ 23 and 60). It was shown that at lower monomer to catalyst ratios. On the basis of our experimental data and in line with literature data.3.3.1. no detailed explanation has been put forward to explain this anomalous behaviour. Clearly. However. with an enthalpy of polymerization of approximately ‐15 kJ/ mol . One of the experiments (S111) resulted in a very low yield (31%) compared to the other reaction. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 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. which reduced to 75% when increasing the temperature to 120oC. Higher polymerisation yields were obtained at lower temperatures. Another possibility is the occurrence of polymer crystallization.3. The effect of the reaction temperature (60 and 100°C) on the product yield at three different dioxanone: 2 ratios is given in Table 2.5 h is higher than the yield at 16 h.
and dioxanone/ 2 mol ratio) is absent.7.) for all experiments are scattered randomly between 2. The X n exp increases linearly with respect to the dioxanone : 2 ratio. A clear trend between R gp and the process conditions (temperature. time: 16 hr Temperature: 60oC. The R gp values (Table 2.5 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. to reduce the number of OiPr endgroups in the product may be the removal of isopropyl alcohol formed in step 40 . time: 1. 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.). 2. A possible strategy to increase the Rgp values i.4 and 3. time: 16 hr 50 Temperature: 60oC. in line with the theoretical predictions. the X n theo (eq. 2. time.3.7.3. Most experimental points are scattered along these lines. and 22.214.171.124.Chapter 2 2.. reaction time and temperature on the X n exp of the products is shown in Figure 2.6.e.5. In Figure 2.2.3. 60 Temperature: 100oC.) as a function of the dioxanone : 2 ratio at 90% and 100% monomer conversion is also provided. Effects of process conditions on the average chain length ( X n exp ) and end group distribution The effects of the p‐dioxanone: 2 ratio. as expected for a typical ring opening polymerisation .
49 + 0.8. indicating that the model describes the experimental data reasonably well.4. vacuum distillation before adding the dioxanone monomer .).459*10‐7 41 .543 exp as a Model Error Total P‐value 3.854 × (dioxanone : 2 ratio) + 0. time and dioxanone/ 2 ratio) on the X n exp has been performed by multivariable linear regression  on the data given in Table 2.6) was too limited to draw sound conclusions.977 (with an adjusted R2 value of 0.3. The low P‐value clearly indicates that the model is statistically significant.7.703 F 173. The 2/ Al ratio was not included in the model as the experimental range (6.2. The analysis of variance for the model is given in Table 2.g.3‐6. The R2 value for the model is 0. Table 2. A parity plot of the modeled versus experimental values of the average chain length Xn confirms this statement (Figure 2.097 × (T ) (2. Statistical Data Analysis Quantification of the influence of the experimental factors (temperature.177 × (t ) − 0.12. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 1 of the polymerisation process (eq.4.3.) from the reaction mixture by e. 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. 2.974). A linear model proved adequate to describe the effects of the independent variables on the X n exp : Xn exp = 8. 2.) where t and T are respectively the polymerization time and temperature.3.
Furthermore. The isolated yields of the off‐white solid products were between 30 and 96%. 2) with Al(OiPr)3 as the catalyst is reported. 11] and due to the fact that the reaction is an equilibrium polymerization with a slight exothermic effect.3. 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.2.8.4. 17]. with high ratios leading to a higher average chain length. the model allows determination of the process variables to obtain a product with the desired degree of polymerization. Within the experimental ranges. Conclusions The ROP of p‐dioxanone in the presence of a monosaccharide (1. This is in line with literature data [5.Chapter 2 60 50 40 Xn model 30 20 10 0 0 10 20 30 Xn exp 40 50 60 Figure 2. with higher ratios leading to higher molecular weights. the X n exp is negatively influenced by temperature. In addition. 2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. Both NMR and MALDI‐TOF measurements indicate that the products are mixtures of polymers and significant 42 . the model predicts that the X n exp is positively influenced by the polymerization time. 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). which is in agreement with the available data on ring opening polymerization [15. 16.
Narayan. D. Chu. Macromol.4‐dioxan‐2‐ one initiated by Al((OPr)‐Pr‐i)(3) in bulk. J. P.O. Narayan. D. 153‐166.J. 2000. starch). 34. Biomed.L. 1089‐ 1097. 42. Mater. R. 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. and mechanistic aspects of the ring‐opening polymerization of 1. Chem. Grubb: Hydrolytic degradation and morphologic study of poly‐p‐dioxanone. C. Damrau: Polylactones. 1063‐1071. Macromol. temperature and monomer: monosaccharide ratio) on the X n exp . Degee.4‐dioxan‐2‐one and controlled synthesis of diblock copolymers with epsilon‐caprolactone. P. P. H. Degee.C. Phys. Res. Lin. Moreover. 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.M. A statistical model has been developed to quantify the effects of process variables (time. Zn‐lactate‐catalyzed polymerizations of 1.6. 21. References .5. Macromolecules 2001. Dubois: ʺCoordination‐insertionʺ ring‐ opening polymerization of 1. Raquez.4‐dioxan‐2‐one. .R. These studies will be reported in the next chapter. 43 . 199. 27. 1993. kinetic. 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. H. 8419‐8425. . Raquez. 2. J. P. R. Rapid Commun.M. Dubois: Some thermodynamic. Kricheldorf.g. . 1998.
. 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.
The silylation reaction was performed using hexamethyl disilazane (HMDS) as the reagent in DMSO at 70°C.72. grafting. biodegradable polymers.7 was obtained. Keywords: starch. polycaprolactone. silylation .21 and 0. The DS of the polycaprolactone graft is between 0. The poly‐(ε)‐caprolactone side chains consist of 40‐55 monomer units and is a function of the reagent intakes. 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. ε‐ Caprolactone is grafted to silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 in THF at 50oC. 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.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. the remainder being homopolymers of ε‐caprolactone. The grafting efficiency varies between 28 and 58%. Silylated starch with a degree of substitution (DS) between 0.45‐0.
1. polyvalerolactone. The polymeric structure of starch consists of repeating anhydroglucose units. The content of amylose in starches depends on the plant and typically varies between 18‐28%. in almost all cases. However. 245 million tons of plastics are produced per year. A wide variety of synthetic biodegradable polymers have been prepared. Starch has also been thermoplasticized with the help of plasticizers such as glycerol and other polyalcohols . for instance grafting of monomers (like styrene and methyl methacrylate) to the starch backbone [6. Starch. and this value increases with about 10% per year . 5]. Global production of starch is 60 million ton per year in 2004 .g. Introduction Worldwide.6‐glucosidic bonds besides α‐D‐1. Blends of thermoplastic starch and PCL are not fully miscible. The development of green biodegradable polymers for e. These plastics are mainly synthetic polymers from fossil resources. 7]. and the mechanical properties of starch‐based films are generally inferior to those derived from synthetic polymers [4.4‐glucosidic bonds) and amylopectin (a branched polymer with α‐D‐1. which are known to degrade with difficulty and cause serious environmental problems . the used monomers and the corresponding grafted chains are not easily biodegradable. Polycaprolactone (PCL) is easily degraded by micro‐organisms .4‐ glucosidic bonds). and polybutyrolactone). There are two types of biopolymer in starch. The amylose‐amylopectin ratio in native as well as modified starches has a strong impact on the product properties. a natural biopolymer. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decreased rapidly in time and were fully degraded after 4 weeks . However. Starch films are known to have good oxygen barrier properties. the product properties are in most cases still not up to standards and blending with other polymers is required . Starch modification is therefore needed to meet the product properties in a number of application areas. is one of the potential candidates for future biodegradable polymer products. Starch is present in the body of many plants (tubers. Starch is abundantly available. resulting 48 . Chapter 3 3. as starch is highly hydrophilic. 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. Well known examples are polyesters derived from cyclic lactones (polycaprolactone. the future generation of packaging materials is highly desirable. However. amylose (a linear polymer of anhydroglucoses with α‐D‐1. it is water sensitive. roots) as granules or cells with typical particle sizes between 1‐100 µm. Various modification strategies have been explored.
3.2.1. Hydrochloric acid (HCl) 1 N was prepared from Titrisol concentrated hydrochloric acid solution (Merck) and distilled (Milipore) water. To increase the miscibility of starch and polycaprolactone. This catalyst is extremely air and water sensitive. distilled under reduced pressure at 100 oC and stored under a protective nitrogen atmosphere. 49 .2. 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]. an alternative method to synthesize poly‐(ε)‐caprolactone grafted starch co‐polymers (starch‐g‐PCL) is reported. Aluminium triisopropoxide [Al(OiPr)3] (Acros) was used without further purification. In this way. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors in undesirable phase separation . the ring opening polymerisation occurs solely in the liquid phase and this is expected to lead to higher GE values. ε‐Caprolactone monomer (Fluka) was dried over Calcium Hydride (CaH2) for 48 h. This approach has also been applied successfully to graft PCL and polylactide on dextran [19‐20]. Materials Corn starch (Sigma) was dried at high vacuum (~1 mbar) at 100 oC for one day before use. during the reaction. it has been proposed to chemically graft caprolactone onto the hydroxyl groups of starch using ring‐opening polymerisation . Acros) and methanol (Labscan) were used as received. Materials and Methods 3. Common Ring Opening Polymerization (ROP) catalysts such as tin octoate or aluminium isopropoxide gave low grafting efficiencies (GE. All available data indicate that the presence of water reduces the GE. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). DMSO (Acros) and toluene (Labscan) were dried overnight over molecular sieves 3 Å (Merck) and stored under a protective nitrogen atmosphere. This is expected to lead to reduced reaction rates between starch and CL compared to CL homopolymerisation. Another possible cause for the low grafting efficiencies is the heterogenous nature of the reaction. thus to a reduction in the GE. therefore difficult to handle and releases ethane. A stock solution was prepared by dissolving 1. In this chapter. 0‐14%). This is rationalised by assuming that water competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. Dry tetrahydrofuran (THF) and toluene for polymerization experiments were obtained in closed vessels from Aldrich and were used as received. Hexamethyldisilazane (HMDS. thus leading to the formation of PCL homopolymers rather that starch‐g‐PCL . a very flammable by‐product. leading to a liquid‐solid system.67 gram (8 mmol) Al(OiPr)3 in 50 ml of dry toluene in a glove box. The highest GE (90%) was achieved using triethylaluminium as catalyst .
50 oC): δ 0. Silylated Starch (1. For each experiment.2. broad peaks. starch).1 mbar.5 ml) and a predetermined amount of the stock solution of Al(OiPr)3 in toluene were added. Traces of DMSO trapped in the product were removed by dissolving the product in a small amount of toluene and re‐precipitation in methanol. acetate‐CH3). This procedure was repeated three times. 0.12 (m. Typical example of the starch silylation procedure The procedure for corn starch silylation was adapted from that published for dextran [19‐23]. 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. The samples were characterized by 1H‐NMR. 3‐6 ppm (m. 3. 1. 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.12 (m. 80 oC). A mol ratio of ‐OH groups to catalyst of 10:1 was used. the solvents were removed from the silylated starch product under reduced pressure (~ 20 mbar) at 70 oC. silyl‐CH3). 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. 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.2. after peracetylation. a predetermined amount of ε‐caprolactone monomer (molar ratio of monomer to 50 .60): 1 H‐NMR (CDCl3.2. another portion of toluene (20 ml) was added. 40 oC) until constant weight. DS = 0. After 2 and 4 h reaction time.2. Subsequently. 50 oC): δ 0. The pre‐determined amount of HMDS (typically 24 ml. After removal of the solvents under reduced pressure (0.7‐2. toluene (40 ml) was added to solubilize the precipitated (partially silylated) starch.2. Typical example of in situ polymerization of ε‐CL with silylated starch The silylated product from the first step was dissolved in THF (0. 3‐6 ppm (m.60): H‐NMR (CDCl3. To this homogenous solution. before peracetylation. starch). silyl‐CH3).111 mol) was added to the gelatinized mixture to initiate the silylation reaction. Silylated Starch (1.6 mL) at 50 oC (1‐2 h).2. After 24 h. The white‐to‐transparent solid products were stored under vacuum in a desiccator at 6‐8 oC.1. DS = 0.2. Sample SN‐3. THF (4. Sample SN‐3. The reaction was carried out at 70 oC. 1 3. broad peaks. the silylated starch (1) product was dried in a vacuum oven (~5 mbar. After 6 h.5 (m. Chapter 3 3.
2. 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.12 (s. β and δ‐PCL).4. 1. iPr). The peracetylation procedure applied in this study was adapted from the literature [24. The solid precipitate was filtered and dried under vacuum (~ 5 mbar) at 40 oC for 48 h. The samples were characterized by 1H‐NMR.75. Typically. 4. 1 3. γ‐PCL).98 (t. ε’‐PCL). ε‐PCL).75. DSPCL=0.3. ε‐PCL). DSPCL=0. Subsequently. Sample SN1CL1. 25]. silyl‐CH3). 3. broad peaks. 1. ‐CH.37 (t. ‐CH3 . Starch‐g‐PCL (3.2. Peracetylation of the remaining OH groups of modified starch is a well established procedure to improve characterisation of the products by NMR .34): H‐NMR (DMSO. α‐PCL).5. The product was collected as a white solid and characterised by 1H‐NMR. iPr). The yield was measured gravimetrically and is based on the weight of the product and the total weight of reactants charged to the reactor. the desilylated starch‐g‐polycaprolactone product (3) was precipitated using heptane. iPr).88 ppm (m.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until it was fully dissolved (typically 3 h). 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.5‐3. Deprotection product of Sample SN1CL2. and 5‐5. starch). Silylated Starch‐g‐PCL (2.88 ppm (m.16 (d.3‐4.68. α‐PCL).5. the peracetylating reagents 51 .25 (t. After stirring for 2 h.98 (t.3‐4. 1. γ‐ PCL). DSsilylation = 0.68. 1 (0. β and δ‐ PCL).4 (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. 4.54 (m. filtrated. 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). 4.37 (t. broad peaks. 3. 2. 3.31 (m. 1 3. and vacuum dried at 40 oC for 24 h. 3. starch). 60 oC): δ 1. iPr). 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.2.4 (m.21): H‐NMR (DMSO d‐6.5‐ 3. ‐CH3. ε’‐PCL).54 (m. DSsilylation = 0.2. 4. and 5‐5. 60 oC): δ 0. 1. 3.) was >99 %.2. The total isolated yield at this condition (see Table 3.25 (t.16 (d. ‐CH.2. 3. 1.31 (m.
). Chapter 3 (DMAP.2.3. The assumption of high conversions (>95%) was correct for all experiments (see Table 3. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added. This leads to the following equation: ACL = ACH 2 −ε + ACH 2 −ε ' ACH 2 −ε ' = A3.2.) 52 .1. 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. In this calculation.4.3. The product was precipitated by the addition of methanol and washed several times with methanol. Analytical Methods 3.) where Ax‐y stands for the 1H‐NMR peak area in the range δ x‐y ppm.6 ppm DSsilylation = × = × 3 Astarch proton / 7 27 A3−5.4 ppm + 1 (3.2.) 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.6−0. 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.8− 4. 3. 126.96.36.199.8 ppm (3. It was finally dried overnight in a vacuum oven at 70 oC and 5 mbar until constant weight. 1 3. the DSPCL may be calculated using eq. DSsilylation may be calculated using 1H‐NMR spectra of the products after peracetylation using eq. it is assumed that the average length of the grafted chain is equal to the chain length of the homopolymer. 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. ACH 3 silyl / 9 1 7 A−0.3−3. When assuming that all ε‐CL monomer is converted.1.2 ppm A3.2. 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. The peracetylation reaction was conducted for 7 h at 50 oC. 3.
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). 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.8− 4. and Vm stands for molar volume.) 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).) ⎛ ⎞ ⎞ ⎛ A A = ⎜1 − CH −iso ⎟ × 100 % = ⎜1 − 4. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors DS PCL = 0.3−188.8.131.52. T stands for absolute temperature.5. The values of ∆Hv and Vm were obtained from the SciFinder Scholar database (American Chemical Society. 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.CL/OH ratio ⎥ ⎣ ⎦ (3. A3.2 ) ⎤ 1 × ⎢ (3 − DSsilylation )⎥ ⎢ ε . 53 . 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.8− 4 .3.4 − A4 .9 ⎡ 0.8− 4 .3−3.4 + A3.( A3.3−3. 2007) 3.9 ⎟ × 100 % ⎜ ⎜ ⎟ ACH 2 −ε ' / 2 ⎠ A3. Although all steps have been investigated. the focus of this chapter will be on the first two steps of the procedure. This leads to the following equation: GE = ACH 2 −ε ' / 2 − ACH −iso ACH 2 −ε ' / 2 × 100 % (3.4 / 2 ⎟ ⎠ ⎝ ⎝ where Ax‐y stands for the peak area in the range δ x‐y ppm.
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. a homogeneous reaction mixture was maintained throughout the reaction. The 1H‐ NMR peaks from the AHG unit after peracetylation were indeed considerably sharper and allowed more accurate DS calculations (Figure 3. as suggested by Einfeldt et al . The degree of substitution of the silyl groups (DSsilylation) was determined by 1H‐ NMR.a. In this way. a product with a DS of 0.).60 was obtained. mixtures of toluene and DMSO were applied to avoid precipitation of the silylated starch during the reaction. C O H 2OR H 1 + NH3 2 OH H3C [R = H or Si(CH3)3] Starch Hexamethyldisilazane Silylated Starch (1) (3. 54 .a. The silylated products were characterised by NMR.1.). Synthesis of Silylated Starch The silylation of corn starch was performed with HMDS as the silylating agent (eq. An additional peracetylation procedure to substitute the free hydroxyl groups with acetate groups was performed to improve the quality of the NMR spectra. Chapter 3 3.5. Using standard conditions (HMDS: AHG molar ratio of 3. The 1H‐ NMR spectrum of silylated starch (Figure 3.3. However.) shows the presence of silyl groups at about δ 0 ppm.). The silylation procedure was adapted from that previously reported for dextran [19‐23]. 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.1.1.1.) Instead of using DMSO as solvent. It is defined as the average number of silyl substituted OH groups on the anhydroglucose (AHG) unit of starch. 3. 24 h).b. 70 oC.5. the starch peaks are very broad and this feature hampers accurate determination of the DSsilylation.
55 .2.60. Typical 1H‐NMR Spectra of Silylated Starch (Sample SN3. not peracetylated (b).1.1.5‐4. The results are given in Table 3. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Figure 3. and Figure 3. 50 oC): (a).5) on the product DS was investigated by performing experiments with a constant starch and a variable HMDS intake. DSsilylation=0. peracetylated The effect of the HMDS to starch molar ratio (1. in CDCl3.
At a mol ratio of HMDS to AHG of 4.6 DSsilylation 0. For reactions with an order higher than zero.4 1. solubility parameter = 11. 70 oC) Table 3.25 3 4.1.36 cal1/2cm‐3/2) considerably.65 0.2. The silylation reaction 56 .55 0. HMDS is a rather apolar compound (Hildebrand solubility parameter of 6. the volumetric HMDS intake is about half of the DMSO intake.25 cal1/2cm‐3/2) due to the presence of the bulky apolar SiMe3 groups. mmol) 12 (56) 18 (83) 24 (111) 36 (167) HMDS: AHG mol ratio 1. the DSsilylation decreases for higher intakes of HMDS.60 0. a fixed starch intake of 6 g (37 mmol AHG) starch was applied.46 Experiment SN1 SN2 SN3 SN4 a Experiments were performed in DMSO at 70°C. Effect of HMDS: AHG mol ratio on DS of the silylation product a HMDS intake (ml.5 3 3.5 2 2.5 HMDS : AHG Ratio [mol/ mol] 4 4.68 0.5 Product DS (DSsilylation) 0.45 0. a positive effect of higher reagent intake on the reaction rate and thus the DS is expected. Surprisingly.5 Figure 3. Its presence will reduce the polarity of the reaction medium (DMSO.67 0. DS of the silylated products at different HMDS: AHG ratios (DMSO.5:1.5 0. Chapter 3 0.5 2. The experimentally observed lowering at higher HMDS intakes is likely due to a decrease in the polarity of the reaction mixture. For all experiments.
To the best of our knowledge. and DMSO/pyridine mixtures yielded silylated starch with DS values ranging between 0.3‐2.) 57 .1 and 3. O O H2C O R H H OR H O O H OR H O Catalyst: Al(OiPr)3 Toluene. DMF. 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. The use of HMDS to silylate dextran (Mw=6000‐40000) in DMSO (HMDS to OH molar ratio of 0. Petzold et al  reported that silylation of starch with trimethylsilylchloride (TMSCl) in pyridine yielded trimethyl‐silyl substituted starch with DS values between 0. Silylation of starch with HMDS in formamide. Similar reductions in the reaction rates when working at higher reactant intakes were observed for the esterification of starch with vinyl laurate and stearate .6‐glucosidic for dextran).0) gave silylated dextran with DS values between 1.68. 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. DMA/LiCl. sample SN‐1) in THF at 50 oC for 24 h using Al(OiPr)3 as catalyst. pyridine.6.7) of the products is in the range of those published for other starch silylation systems.25‐5. The DSsilylation (0.7‐3. white solid products with isolated yield > 96% were obtained.2.6. 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.0 [18. After precipitation with heptane and vacuum drying.3. 3.45‐0. 29]. liquid ammonia.0 . the silylation of starch in pure DMSO using HMDS as the reagent has not been reported to date. 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. The products are soluble in DMSO as well as in less‐polar solvent such as chloroform and THF.2.
The homopolymers are formed by direct polymerization of ε‐CL initiated on isoproproxide moieties attached to the Al catalyst. A typical spectrum is shown in Figure 3. DSsilylation =0.4 ppm.3. Coding of the peaks is given in Figure 3.4. as is shown in Figure 3. However. Further process optimization studies allowed the synthesis of products with less than 42% of homopolymers (vide infra).2 ppm.8 and 5.0‐5. Typical 1H‐NMR spectrum of a silylated starch‐g‐PCL sample. 58 . Apparently.3. (Sample SN1CL1.21) in DMSO‐d6 at 60 oC. broad peaks in the region δ 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.2‐4 ppm and imply that the ring‐opening polymerisation reaction of CL indeed occurred. The peaks from the polycaprolactone units are clearly present in the range of δ 1. Figure 3. not all of the caprolactone is grafted to starch and large amounts of PCL homopolymers (72%) were formed.4‐3. DSPCL=0.9 and 1.4. Resonances from the starch peaks are observable as small.68. The formation of CL homopolymers for this type of reactions has been observed before [15. Chapter 3 The products were characterized using 1H‐NMR analysis in DMSO‐d‐6 as the solvent. This is clearly indicated by resonances of the iPr end‐group of the PCL homopolymer at δ 4.
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. Exchange Reaction: iPr O Al O iPr iPr O + H2O iPr O iPr O Al OH + iPr OH 2. 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.) 59 . peaks arising from a ‐COOH unit could not be detected in 13C‐ NMR spectra (δ 175‐180 ppm). Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors 1. 3.7. 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 3.7. residual water may also initiate the polymerisation reaction. However.4. This leads to the formation of carboxylic end groups (see eq.). 1. Product formation for the reaction between silylated starch and ε‐ caprolactone Besides the ‐OH group of starch.
2 Total Yieldb [%] >99 >99 >99 99 96. and reaches 58% for a ε‐CL to starch–OH ratio of 29. However. b.5. The yield of the products was measured gravimetrically and varies between 96. Five experiments were performed to study the effect of different ε‐CL to silylated starch ratio. In the case of only homopolymerisation. The grafting efficiency (GE) for the sample given in Figure 3.0 15. The GE increases with higher ε‐CL intakes.5 and 100%.21 0.58 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. (SN1CL1) is 28%. This implies that the ε‐CL conversion is essentially quantitative in all cases.5. in all samples. Chain Length [mon.3.2. Determined gravimetrically and defined as the total weight of the isolated product divided by the total intake of reactants (silylated starch and CL).3 ppm) should be twice the intensity of the ‐CH‐ proton of the isopropoxide end group (δ 4.68) in THF at 50 oC with Al(OiPr)3 as the catalyst (1 mol Al(OiPr)3 per 10 mol‐starch‐OH groups).2.).72 Grafting Efficiency [%] 28 43 48 55 58 All reactions were performed using the same intake of SN1 silylated starch (DS=0.5 29. the intensity of the peak from the ‐CH2‐ end group of the homopolymer (ε’ at δ 3. units] 40 43 44 49 54 DSPCL 0. the intensity of the resonance ε’ was considerably higher that that of the ‐CH‐ iPr peak.0 18. Experiment SN1CL1 SN1CL2 SN1CL3 SN1CL4 SN1CL5 a.9 ppm). and Figure 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. Overview of results for the grafting of ε‐CL on silylated starch a ε‐CL/ OH [mol/ mol] 13. The results are given in Table 3. 60 .34 0. This implies that grafting of caprolactone to starch also occurs to a significant extent.47 0.5 Avg.2.9 22. Table 3. The ACL of the polymer and the DSPCL increase almost linearly with the ε‐ caprolactone intake (Figure 3.
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)
Considerable amounts of ε‐CL hompolymers with isopropyl end‐groups were also formed. 3. depending on the ε‐CL‐starch ratio. the highest value was obtained with a ε‐CL‐AHG ratio of 29. 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. The grafting efficiency varied between 28‐58%.2.4. Chapter 3 3. The ROP with ε‐CL was performed using Al(OiPr)3 as catalyst in THF as the solvent. The DS of the PCL chains was between 0. Conclusions The successful synthesis of poly‐(ε)‐caprolactone grafted corn starch co‐ polymers using a three step approach is reported. 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. indicating that homogeneous reaction conditions are favorable for the grafting reaction. This was achieved by making the starch more hydrophobic by partial substitution of the OH groups by trimethylsilyl groups. 1. The key feature is the use of a homogeneous reaction mixture for the ROP of starch with ε‐CL. the percentage of PCL grafted to starch compared to the total amount (grafted and homopolymer) of polymerized CL [%] : gas constant. 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. The products may have interesting applications as compatibilizers for starch‐polymer blends.72.46‐0. Silylated starch with a low‐medium DS (0.21‐0.986 cal mol‐1 K‐1 : temperature [K] : R T Vm molar volume [cm3/ mol] 64 . 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.5.68) was obtained using a DMSO/toluene mixture as the solvent and HMDS as the silylating agent.
M. X. The compelling facts about plastics: An analysis of plastics production. Biobased packaging materials for the food industry. 2003.A. Aarhus. Plastics Europe. C. . Association of Plastics Manufacturers. of . 107‐113. Beliakova. . Polym. Krochta. . S. Mater. Sci. W. R. . 23. Prog. J. 1995. Starch‐Starke 2004. 407‐412. 3. Sci. G. Polym. Polym. Doane. 1997. L. R. Food Technol. 385–409 . K. References . Eng.html.R. Wang.K. 1998. status and perspectives. Chiellini. Polym. Rev. Fanta. 1996. R. 8. E. 2000. 2008.dk/isi/stat/rawmaterial. Adv. Sci. demand and recovery for 2006 in Europe. A.Sci. Solaro: Biodegradable polymeric materials. Department of Dairy and Food Science. 61‐74. 38. 1977. Babac: Preparation and biodegradation starch/polycaprolactone films.. Part C: Polym. Linke: Water vapor sorption determination of starch based porous packaging materials. Goldberg: Review of the biodegradability and utility of poly(caprolactone). Starch‐Starke 2000. Environmen. Tsiapouris. 51. Wang: Properties of starch blends with biodegradable polymers. 56. De Mulder‐Johnston: Edible and biodegradable polymer films: Challenges and opportunities. J. . 17. Yavuz. accessed on August 01. Bagley. 1251‐1253. E. J. 53‐57. . A. J.A.M. Rustgi: Biodegradable polymers. R. report of Food Biopack Project (Ed. Denmark: http://www. . . Website of International Starch Institute. Burr. Aly. 52. Weber)..C. Environ. J. Yang. Denmark. H.A. Chandra. 11. 65 . Russell: Graft copolymers of polysaccharides with thermoplastic polymers ‐ new type of filled plastic. Abdel‐Mohdy: Grafting of poly(methacrylic acid) on starch and poly(vinyl alcohol). Brussels. 2003. 2008. Macromol. Y.F.B.starch. C. 1998. Degrad. Belgium. Eng. 305‐313. C. A. 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.C. . Frederiksberg C. . Karlsson. D. 1273‐1335. C. F. the Royal Veterinary and Agricultural University. Polym. M. Albertsson: Biodegradable polymers and environmental interaction. 311‐316. 61‐67. 43.6.
Heublein: Silylation of cellulose and starch ‐ selectivity. . 561‐566. Macromolecules 2000. structure analysis. D. 41. Moro. 2004. E.L. E. Sci. M. 43. P. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. Dellacherie. I. 6713‐ 6721. Petzold. Stein. Rutot. Degee. K. A. Polymer 1999. Dellacherie. selective syntheses. L. . Hamaide. Monomers Polym. Klemm. 1. Nouvel. Nouvel. high value applications.L. B. D. E. Stab. and subsequent reactions. Duquesne. Dole. J. Gunther. Nouvel. . Ydens. 965‐969. T. Degee. Kussler. Six: Partial or total silylation of dextran with hexamethyldisilazane. 415‐426. Ydens. W. W. Polym. Fessi. 10. 2001. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer 66 . Six. P. Biomacromolecules 2003. L. M. 2577‐2588. 251‐269. Des. J. K. P. Symp. D. Cellulose 2003. Dubois. C. Averous. Klemm. A. Part A: Polym. 40. Klemm: Regioselective functionalization of starch: Synthesis and 1H NMR characterization of 6‐O‐ silyl ethers. P. 163. 341‐347. D. . Einfeldt. D. Pantiru. Biomacromolecules 2001. J. Einfeldt: Structure design of polysaccharides: novel concepts. A. Petzold. D. M. 4. E. 2002. Krishnan. H. Dubois. Rutot. C. Dubois: Controlled synthesis of poly(epsilon‐caprolactone)‐grafted dextran copolymers as potential environmentally friendly surfactants. Ydens. 4157‐4167. 33. 2001. Dubois: Aliphatic polyester‐ based biodegradable materials: new amphiphilic graft copolymers. R. . P.L. . and chemical stability of silylated dextrans. Klemm. Petzold. . P.L. 42. Stein. . 2001. Biosci. Six: Controlled synthesis of amphiphilic biodegradable polylactide‐grafted dextran copolymers. A. Degee. D. 1735‐ 1743. . P. W. P. Petzold. 5. L. K. J. Einfeldt. Polymer 2002. Polym. Dubois. P. . C. Chapter 3 .. Dellacherie. . Dubois. P. I. J. Chem. . Klemm: Preparative and 1H NMR investigation on regioselective silylation of starch dissolved in Dimethyl Sulfoxide. Gunther. Koschella. E. Gunther: Synthesis and NMR characterization of regiocontrolled starch alkyl ethers. Polymer 2000. L. 3091‐3100. P. Macromol. 35‐47. Degrad. Dellacherie. regioselectivity. Fringant: Properties of thermoplastics blends: starch‐polycaprolactone. 1443‐1450. 73. Six: Silylation reaction of dextran: Effect of experimental conditions on silylation yield. 2. K. D. Stein. Macromol. L. I.
.K. S. D. Res. Rapid Commun.R. Macromolecul. 659‐663. L.L. 1399‐1406. 1089‐1097. Mexico City. accepted for publication in Starch‐Starke. 1978. 29. 553‐556. Biela. L. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. 941‐950. J. 67 . 2008. D. Picchioni. . 61. Macromolecules 1996. 21. 2001. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors agents. Duda: Living polymerization with reversible chain transfer and reversible deactivation: the case of cyclic esters. Macromol Chem. . A novel route to functionalised nanoparticles.O. Macromol. Lehmann: Selective 6‐O‐acetylation of amylose. Damrau: Zn‐lactate catalyzed polymerizations of 1. Penczek. Sugih. Rapid Commun. Hertz Jr. A.B. Horton. .P. T. R. 22. Carbohydr. H. A.J. A. 1989 .4‐ dioxan‐2‐one. F. Duda: Polymerization of ε‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. Kricheldorf. 199. Junistia. . 2000.M. H. Janssen.: Solubility parameter concepts – a new look. Kinetics and mechanism. Phys. Manurung. presented at ACS Rubber Division meeting. D. 1998.
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. When performing the reactions using Na2HPO4 as the catalyst. presumably due to a reduction of the polarity of the reaction medium. vinyl laurate. esterification.24 ‐ 2. the DS decreases. Keywords: corn starch. At higher ratios. The fatty esters were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. vinyl ester to starch ratio and the type of catalyst. vinyl stearate . For low vinyl‐ester to starch ratios. K2CO3. the DS for the starch‐laurate compounds is higher than for the corresponding starch‐ stearates.96). K2CO3. and Na‐acetate) in DMSO at 110°C. 13C‐NMR and FT‐IR. an increase in the vinyl‐ester concentration leads to higher product DS values. and Na‐acetate are superior catalysts with respect to activity compared to Na2HPO4 and products with DS values close to 3 were obtained. The yellowish products were characterized by 1 H‐. The DS of the products is a function of the carbon number of the fatty acid chain.
5]. most of the studies performed to date use short chain carboxylic acids (C1‐C4). defined as the moles of substituents per mole of anhydroglucose (AHG) units) .61) products were obtained using this approach. The mechanical properties and hydrophobicity of the products were significantly improved when using these longer chain fatty acid precursors [1. An attractive field of application for these polymers is the use as packaging materials. more water‐resistant products may be obtained [3‐4]. poor processibility (high viscosity).7 [1. cellulose or proteins are potentially very interesting starting materials for biodegradable packaging materials. high brittleness. 7]. Further modification of starch is therefore required to introduce hydrophobicity and to improve mechanical and moisture barrier properties. In particular starch is attractive as it is relatively cheap and abundantly available. resulting in products with DS values up to 2.34‐0.Chapter 4 4. However. and consequently. even after the addition of plasticizers . The hydrophobicity increases with the degree of acetate substitution (DS. Introduction Green biodegradable polymers derived from natural resources are potentially very interesting substitutes for non‐biodegradable petroleum‐based polymers. the fatty ester substituents [1.1. and particularly acetic acid derivatives (C2) [2‐4]. To improve the mechanical properties. The major obstacle is the pronounced brittleness of the materials. 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. However. Natural polymers such as starch. higher molecular weight carboxylic starch esters (C4‐C6) . However. Their 70 . 5]. The introduction of acetate groups on starch makes the product more hydrophobic. Recently Mormann et al  explored the possibility of using vinyl esters and particularly vinylacetate as reagents for the preparation of starch esters. the use of native starch for packaging materials is limited due to its low moisture resistance. and incompatibility with hydrophobic polymers. and even fatty acid derivatives (C12‐C18) have been used in the modification reaction [5. Relatively low‐DS (0. However. The first paper on the acetylation of starch was already published in 1865 . An alternative method using methyl and glyceryl laurate esters in the absence of solvent has been recently developed . Esterification of starch with low molecular weight fatty acid derivatives is one of the oldest modification technologies to improve starch properties. 5‐6] were introduced using fatty acid chloride reagents. For the current petrochemical based products recycling is often neither practical nor economically feasible . which are relatively expensive and rather corrosive .
6 for starch acetate ester) were obtained. and tetrahydrofuran (THF) were supplied by Acros (Belgium) and were also used as received. the use of basic catalysts other than Na2HPO4 has been investigated. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). In addition. to the best of our knowledge. ~1 mbar). Potassium carbonate (Boom. The effects of the starch to vinylester ratio on the reaction rates and DS have been explored. Germany) and acetic anhydride (Merck. The products were placed directly on the diamond plate and 50 scans with a resolution of 4 cm‐1 were recorded.1.2. 4. 4‐N. 1 71 . Materials Corn starch (approx. vinyl laurate (Fluka. 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. 4. sodium acetate (Merck.N‐dimethylaminopyridine (DMAP). In DMSO. The synthesis of starch stearate esters using vinyl ester reagents has. Germany) were used as received. starch esters with a substantially higher DS value (up to 1. Materials and Methods 4.01 when using vinyl‐laurate.2. IR spectra were recorded on a Spectrum 2000 FT‐IR Spectrometer (Perkin Elmer). Japan).2. not been reported to date.2. The spectra were recorded at 50 oC. The reactions were either performed in water or in DMSO using a basic catalyst (Na2HPO4). Germany) and disodium hydrogenphosphate (Merck. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. The starch was dried before use for 48 h at 105 oC under vacuum (approx. 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. leading to higher reaction rates. This solvent effect is likely caused by the higher solubility of the vinyl esters in DMSO compared to water. The maximum attainable DS of starch acetate in water was below 1 and limited to 0. the Netherlands). Germany) were used without further purification. leading to a moisture content of 2 %‐wt (measured gravimetrically). Analytical grade vinyl stearate (Aldrich. as recommended by Laignel et al . 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. Technical grade dimethyl sulfoxide (DMSO).
typical spectra are given in Figure 4.96) H‐NMR (before peracetylation. broad peaks.4 (broad. 28‐32 (C4‐9).9 (t. broad peaks. 3H.. 12).2. 31. 1. 1295. 3‐6 ppm (m.1 (m.3 (m.4 (m. respectively). 68‐74 (broad. 95. the product was precipitated using methanol (100 ml) and separated from the liquid phase by decantation. 184.108.40.206 (t. After cooling. 1230 (C‐O stretching). 13 FT‐IR (cm‐1): 2920 (C‐H stretching). Starch‐stearate (Sample 19. C12). CDCl3): δ 14.9 (C3). 1455 (CH2).Chapter 4 4. 3H. CDCl3): δ 0. 1. 1370 (C‐H bending).2. C1S). 172‐174 ppm (C=O. Table 4. 2H.6 (m.4. 1. the product was dried in a vacuum oven (70 oC.5 g) was first gelatinized in DMSO (5 ml) at 70 oC for 3 h. The product was washed twice with methanol (50 and 25 ml. 760. C2’). 16H. broad peaks. 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.9 (broad. C1S‐6S). 1. 1740 (C=O).52): H‐NMR (before peracetylation. C2). CDCl3): δ 0. 2H. DS = 2. The samples were characterized by 1H‐ and 13C‐NMR and FT‐IR.5 (m. 1020 (C‐O stretching). The atom numbering scheme is given in Figure 4. 2.3.3 (m. 1110 (C‐O stretching). 1.5 (m. resulting in the formation of a homogenous transparent solution. 3S. attached to O‐C2S. 1. DS = 2.1.3. 935 (C‐O stretching). 3‐6 ppm (m. 720. broad peaks. C6S). Figure 4.3 (m. broad peaks.. 2850 (C‐H stretching). 1410 (C‐H bending). C1S‐6S) 1 72 . 28H. Finally. C18). 3H. 34. 22. CDCl3): δ 0. 2H. Typical example of the synthesis of laurate and stearate esters of corn starch Corn starch (0. approximately 5 mbar) for 24 h until constant weight.2. 1 H‐NMR (after peracetylation. C4‐C17). 76‐78 ppm. broad peaks. C3).7 (C11). 3‐6 ppm (m. (FT‐IR). 2H.0 (C12). 7H. C3). 1150 (C‐O stretching). 7H.8‐2. (13C‐NMR) and Figure 4.. broad peaks. overlap with CDCl3 (C4S). C2S.9 (C10). 7H.0 (m. and O‐C6S). Methods 4. 61. C1S‐6S).9 (t. 2H. C4‐11). 1350 (C‐H bending). 2H. Subsequently. C2). (1H‐NMR). C2). 24. 2. O‐C3S. C4‐11). 5S).1 (C11). 16H. Starch‐laurate (Sample 17.3. 1 C‐NMR (before peracetylation. C3). 3H. Table 4. 1.1. broad peaks.5 (m.
3.0 (C16).9 (t. 3H. 1100 (C‐O stretching). 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. 1.2. C1S). 720.0 (C18). 1. 1150 (C‐O stretching). 28H. 3S.6 (m. 5S).4 (broad. 2. attached to O‐C2S. 760. 1020 (C‐O stretching). C2). 1350 (C‐H bending). 1295.5 (m. C6S).3 (m. 32.7 (C4S). 172‐174 ppm (C=O. 3‐6 ppm (m.1. CDCl3): δ 0. C18). and O‐ C6S) 13 FT‐IR (cm‐1): 2920 (C‐H stretching). 1370 (C‐H bending). 950 (C‐O stretching). C3).2 (C2).4 (m. 2850 (C‐H stretching). O‐C3S. 2H. C2’).7 (C17). A peracetylation reaction to substitute all of the remaining 73 . broad peaks. 22. 75. 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. 7H. 1. 26‐32 (C4‐15). 1410 (C‐H bending). 3H. 34. CDCl3): δ 14.5 (broad. 61.0 (C3). 95. broad peaks. C1S‐6S) 1 C‐NMR (before peracetylation. 25. 1455 (CH2). 865. 2H.2. 1740 (C=O). Numbering scheme for carbon atoms of products 4. C4‐C17).8‐2. 68‐74 (broad. C2S. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants H‐NMR (after peracetylation.
) 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. Subsequently.).) (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 3.8‐2. The peracetylation reaction was conducted for 7 h at 50 oC.4.6. the peractylating reagents (DMAP..3.) AH − starch = (4.6‐5. 4. and 4.3. The peracetylation procedure by Einfeldt et al  was applied. The product was precipitated by the addition of methanol and washed several times with methanol.).6 − AC 2 3 (4.1.6 7 (4. and 4.3. AH − acetate = A1.2.5.: 74 . DS fatty ester = 3 − DS acetate (4.: DSacetate = AH − acetate AH − starch A3. Determination of the Degree of Substitution (DS) The DS of the product was calculated using 1H‐NMR spectra of the products after peracetylation.4. the starch ester (0. The DS of the fatty acid esters was calculated from the DS of the products after peracetylation (eq 4.8 ppm (protons attached to the fatty acid carbons C3‐C12 for laurate and C3‐C18 for stearate) using eq.1. It was finally dried overnight in a vacuum oven at 70oC and 5 mbar until constant weight. Typically.7.3. 4. The procedure was described earlier by Elomaa et al  and the relevant equations are shown in equation 4.6 ppm).2. 4.6 ppm range) and a correction has to be made (eq 4.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until the starch was fully dissolved (typically 3 h).2.5.6−5.8− 2.Chapter 4 hydroxyl groups with acetate groups was applied to obtain reliable DS data.5‐1.) 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.
Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants AH − fattyester = AH − fattyester = AC 3−C12 A0. Results and Discussion 4. 75 .5−1.8 = (stearate) 33 33 (4. the esterified starch started to separate from the medium in the form of a gel. After 2‐3 h. 4. the vinylester and the catalyst were added and the reaction mixture was heated to 110°C. After 24 h.. 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. A schematic representation of the esterification reaction of starch with the vinyl esters is provided in Scheme 4.8 = (laurate) 21 21 AC 3−C18 A0.7. the starch was gelatinised in DMSO at 70°C for 3 h to make the hydroxyl groups of starch more accessible for reaction.) (4. but swell in organic solvents such as toluene and THF.1).) In eq. 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.1.1. the brownish gel was precipitated with methanol and the product was collected after vacuum drying in the form of a transparent. The products of these exploratory reactions are insoluble in water and DMSO.5−1. The reaction was performed in two discrete steps.3.3. light yellow solid. Subsequently. 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. Initially. Ax‐y stands for the peak area in the range δ x‐y ppm.6.3.‐4.1.7.
1. 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.3. Higher DS products have a higher solubility in CDCl3 and good quality 1H‐ and 13C‐NMR spectra could be obtained (Figure 4. Table 4.Chapter 4 The DS of the products was determined by using NMR (vide infra). 4. (b) starch laurate.2. Product Characterisation 4.1. even at higher temperatures (50 oC). DS = 2.52 (Sample 17. Medium DS starch laurate and starch stearate (1 < DS < 2) dissolve poorly in DMSO‐d6 and CDCl3.). 4. (c) peracetylated starch laurate. Typical 1H‐NMR spectrum of: (a) native starch in DMSO d‐6 at 60oC.2.2. Table 4.52 was obtained.3.52 (Sample 17.) in CDCl3 at 50oC.3. For atom numbering scheme: see Figure 4. DS = 2.1. A reaction with vinylstearate at similar conditions resulted in a stearate starch ester with a DS of 2. and 4.2. When using a vinyl‐laurate : AHG molar ratio of 3 and K2CO3 as the catalyst. a product DS of 2.96.) in CDCl3 at 50oC. Figure. 76 .1.
5 ppm) are broad and overlapping .5 ppm). required for DS determinations. 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.c. 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). DS = 2. NMR spectra of the peracetylated products are considerably improved in terms of peak resolution and allow a more reliable calculation of the DS.96 (Sample 19) in CDCl3 at 50oC. The proton signals of the acetate methyl group. (b) starch laurate. The resonances arising from the anhydroglucose unit of starch are broadened. The starch peaks (δ 3‐220.127.116.11.52 (Sample 17) in CDCl3 at 50oC. Clearly visible are the peaks arising from starch and the aliphatic hydrogen atoms of the fatty acid chain (δ 0. This feature hampers the DS determination by NMR. and therefore a peracetylation procedure to substitute all of the remaining OH groups with acetate groups was applied [8. For atom numbering scheme: see Figure 4. Typical 13C‐NMR spectra of the products are given in Figure 4. The 1H‐NMR spectrum of a typical peracetylated starch laurate is shown in Figure 4.8‐2. 10‐11]. Figure 4. DS = 2. The same 77 . Typical 13C‐NMR spectra of: (a) native starch.2. in DMSO d‐6 at 60oC. are together with the CH2 groups of the acid chain adjacent to the ester moiety in the range δ 1. (c) starch stearate.2.3 ppm. Two of the carbon resonances (1S and 4S) are considerably shifted compared to native starch.8‐2.
starch stearate (DS= 2. c.) and native starch FT‐IR spectra of both starch laurate and starch stearate (Figure 4. starch laurate C-H C=O C-O C-O a. Table 4. Sample 17. In addition. Table 4. Characteristic peaks of the polysaccharide backbone are visible in the 1250‐900 cm‐1 region (C‐O stretching) . starch stearate C-H C=O C-O Absorbance b.c.c.4. the C‐H stretching vibrations of the alkyl groups of the fatty ester chain are clearly present at 2920 and 2850 cm‐1. native starch -OH C-H 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Figure 4.4. a spectrum of native starch (Figure 4.a. For comparison.Chapter 4 phenomenon was observed by Dicke for starch acetate .2.2.) show characteristic bands of the carbonyl group of the fatty esters in the 1750‐1700 cm‐1 region. The near absence of remaining hydroxyl 78 .4. and 4.52. and 4.4.1. 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.).) is also included.b.4. FT‐IR measurements The FT‐IR spectra of starch laurate and starch stearate are shown in Figure 4. 4.3. Sample 19. FT‐IR Spectra of starch laurate (DS= 18.104.22.168.1.b.
90 1. Most of the experiments (14) were performed using Na2HPO4 as the catalyst.1. The results are shown in Table 4.05 0.54 2. Table 4.07 0.13 1.91 0.1.2. type of catalyst and the effect of the addition of co‐solvents on the product DS was studied in more detail.44 Exp.3. In addition.96 2.99 1. 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. four experiments were performed with two alternative basic catalysts (K2CO3 and Na‐acetate). 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 1.90 0.57 0.68 0.60 1.52 2. 79 . 4.24 2. 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.08 1.23 0.01 0. in line with the NMR data. Systematic Studies The effect of important process variables like the vinyl‐ester to AHG ratio.
). This is likely due to a reduction of the polarity of the reaction medium.3. The highest DS value was 1. the ester intake is equal on a weight basis to the DMSO intake. At low to medium vinyl ester/AHG ratios (0‐3).5. This behaviour is likely the result of two opposing effects.5 Degree of Substitution (DS) 1 0. The results are presented in Figure 4. 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 in the vinyl ester intake leads to a reduction in the DS.3.5. The DS of the products is a clear function of the vinyl laurate and stearate ester intakes (Figure 4. 110 oC.23 for vinyl‐laurate at an intermediate vinyl‐ester‐AHG ratio of 3. 1.5 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 . 2 wt% catalyst intake on starch). 7‐10).5. The DS values are increasing with higher vinyl‐ ester/AHG molar ratio until a certain maximum.1. DS of the product as a function of the type of vinyl ester and the vinyl ester‐starch ratio (24 h reaction time. A further increase leads to a reduction in the DS. At a ratio of 1:6. 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. Higher concentrations of vinyl esters are expected to lead to higher esterification reaction rates.
and illustrated in Figure 4. The latter factors appears to have a strong effect on reaction rates (vide supra). et al  for the silylation of starch.7 for lauroyl chloride to 0. Aburto. This effect is particularly evident at higher vinyl ester/ AHG ratios (>3) (Figure 4.3.8 for stearoyl chloride. The results are given in Table 4. 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. with reductions in polarity leading to lower reaction rates. 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. the addition of co‐solvents (toluene/ THF) led to an increase in the DS.2. 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. 4. the DS of the product is also a function of the chain length of the fatty acid. it also results in a dilution of the reaction mixture and a reduction in the polarity.3.e.3.1. Here.5. A similar trend in reactivity pattern was observed in this study and the DS decreased from 1. A similar procedure was proposed by Nouvel. These negative effects dominate the reaction performance at higher vinyl ester/AHG ratios and lead to a reduction in the DS values. i.3. It is clear that Na‐acetate and K2CO3 are considerably more active than Na2HPO4 and products with a significantly higher DS were obtained.). K2CO3 and Na‐ acetate were tested. The addition of toluene for the esterification of starch with vinyl‐esters surprisingly did not lead to improved DS values.). For starch 81 .1. These findings were rationalised by assuming that the co‐solvents increase the solubility of the silylated products. Although toluene may positively effect the reaction by (partly) re‐dissolving starch ester precipitates. 4. In this case. Catalysts screening A number of alternative basic catalysts for Na2HPO4. with high carbon numbers leading to a reduction in the DS. 11‐14 in Table 4. 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 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 4‐6.6. the starch laurate esters display higher DS values than the starch stearates. leading to enhanced reactivity. When using Na2HPO4 as the catalyst.3. Thus. Effect of the addition of toluene as a co‐solvent A number of reactions were performed using a co‐solvent.).1.
catalyst amount = 2%‐w.44). the DS of the product is also tunable by proper catalyst selection. The DS of the laurate ester when using Na‐acetate is higher than for the stearate ester (Table 4. the DS for the laurate ester is lower than the stearate ester. For starch stearate.5 0 Laurate Vinyl-Ester Stearate Figure 4.6.6. 110oC. K2CO3 gave products with a significantly higher DS (2. The yellow products were characterized by 1H‐ and 13C‐NMR. Apparently.o. when using K2CO3 as the catalyst. The products were synthesised in DMSO using vinyl‐esters in the presence of basic catalysts (Na2HPO4. the two catalysts are equally effective and products with a DS of about 2. Conclusions A study on the synthesis of corn starch fatty acid esters with high DS values is reported.5 1 0.5 were obtained.Chapter 4 laurate esterification. Thus. the statement that the DS for the laurate esters is always higher than for the stearate esters is not generally valid and a. Comparison of DS values with different catalysts (Vinylester: AHG molar ratio = 3:1. The DS of the products is a clear function of the chain length of the fatty ester and the type of catalyst.96) compared to Na‐acetate (DS=2. However. 3 2. function of the type of catalyst. DMSO).4. gray: Na2HPO4 black: K2CO3 white: Na‐acetate 4. 82 .). K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. and FTIR and confirm the presence of chemically bound fatty acid chains. K2CO3. and Figure 4.5 Degree of Substitution 2 1. in line with the findings for NaH2PO4.1. and Na‐acetate).
Important product properties will be described in the next chapter. 34. products with a DS > 2. Hanna: Water solubility. Polym. At low vinyl ester/ AHG ratio. 361‐368. References . Pacsu: Starch Studies: Preparation and Properties of Starch Triesters. 58. I. Higher ratios led to a reduction in the DS. Polym. 1997. Borredon. Thiebaud. Thiebaud. X. 65. 83 . E. A maximum was observed at a vinyl ester: AHG ratio between 2 and 4. 1647‐1656. The DS of the products may also be tuned with the vinyl ester/AHG molar ratio. S. 705‐721. Bikiaris. Prinos. Prinos. presumably due to a reduction of the polarity of the reaction medium. C. A. Aburto. J. I. Y.W. Alric. E. Appl. Xu. J.4 could be obtained for both laurate and stearate esters. Borredon. W. Crops Prod. Gatenholm: Surface octanoylation of high‐ amylose potato starch films. Panayiotou: Synthesis. A. 1999. Merrill: Properties of Fatty‐Acid Esters of Starch. . J. 54. Polym. 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. Koch. Alric. Chem. 1‐11. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. S. Appl. Appl. D. Carbohydr. E. Sci.5. . 21. D. Bikiaris. 2003. Aburto. M. J. thermal characteristics and biodegradability of extruded starch acetate foams. 1995. D. J. M. . Dzenis. Sagar. C. . J. Ind. 4. 1440‐1451. 1942. K. Sci. Ind. Polym. the DS of the product increases at higher vinyl ester intakes. J. 2005. 74. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants With these catalysts. and P. Eng. Mullen and E. Y. 1208‐1217. characterization. Sci. . J. and biodegradability of fatty‐acid esters of amylose and starch.6. 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. Bengtsson.
Mano. 56. A. J. 84 . Elomaa. Mater. Laatikainen. . Dubois. G. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. 57. Aburto. C. Med. Einfeldt.. Alric. 2004. R. Petzold. P. 11. . Starch‐Starke 2004.F. 14(2) 127‐135. E. . Carbohydr. 302‐307. 1735‐1743. Peltonen. Urtti: Determination of the degree of substitution of acetylated starch by hydrolysis. Alric. . J. W. 51. M. D. I. F. Massiot. W. S. M. C. Cellulose 2004. Nouvel. Al‐Higari: Acylation of starch with vinyl acetate in water. 251‐260. Stein. S. J. 298. Kussler. P. A. Bliard. Hamaili. Koniarova. Borredon: Free‐solvent Synthesis and Properties of Higher Fatty Esters of Starch – Part 2. M. Reis: Thermal properties of thermoplastic starch/synthetic polymer blends with potential biomedical applicability. . 2001. I. J. and E. 43. 1997. Mormann. P. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. . Carbohydr. J. Ydens. Aburto. B. 57. 255‐263. 2003. J. Macromol. Soininen. Polym. . H.Chapter 4 . Res. L. Asplund. M. 341‐347. Klemm: Preparative and H‐1 NMR investigation on regioselective silylation of starch dissolved in dimethyl sulfoxide. R. Biosci. T. H‐1 NMR and TGA/IR. G. L. E. D. 261‐267. 145‐152. Starch‐Starke 1999. I. Degee. . Laignel. Mouysset‐Baziard.L. Gunther. Mater. Hyvarinen. R. Six: Partial or total silylation of dextran with hexamethyldisilazane. K. Dellacherie. Dicke: A straight way to regioselectively functionalized polysaccharide esters. Starch‐Starke 2005. Polymer 2002. Sci. Senocq. 1. 118‐121.
trans‐esterification. catalyst intake. The reactions were carried out in DMSO using basic catalysts (Na2HPO4. and modulus of elasticity of 46‐113 MPa. and Na‐acetate). 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. The thermal stability of the esterified products is higher than that of native starch. The results were adequately modeled using a non‐linear multivariable regression model (R2≥0.39) are totally amorphous whereas the low‐DS ones (DS= 1. The thermal and mechanical properties of some representative product samples were determined. K2CO3.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.45‐1. elongation at break of 3‐26%. The basicity of the catalyst and the reaction temperature have the highest impact on the product DS. vinyl laurate. Keywords: starch esters. High‐DS products (DS= 2.96).7‐3. vinyl stearate . The effect of the process variables (vinyl ester to starch ratio.75) are still partially crystalline.5 MPa. Mechanical tests show that the products have tensile strength (stress at break) between 2.26‐2.
12] studied the synthesis and properties of of C1‐C6 esters of starch. 86 . 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. However. most of the plastic waste is not biodegradable. et al  and Aburto. and particularly on the synthesis of starch acetate (C2) [11‐12]. The products were biodegradable. The global starch production was estimated at 60 million tons in 2004 . polyethylene  and polycaprolactone  and chemical modification by esterification with organic acids [11‐19]. Chemical or physical modification is required for successful applications as biodegradable polymers. The high DS esters showed interesting properties.1. Mullen and Pacsu [11. however. Examples are agricultural mulch . Most early studies mainly focused on the use of short‐chain carboxylic acid (C1‐C4). Starch is an attractive feedstock for the synthesis of novel biodegradable plastics. 3‐4]. Alternative routes to avoid the use of acid‐chloride/pyridine combinations have been developed. the research activities were mainly limited to the use of vinylacetate. The development of novel biodegradable plastic materials capable of decomposing when given an appropriate environment and time is of utmost importance . Typical examples are blending with polyvinyl alcohol . Sagar and Merill  studied the synthesis and properties of C4‐C6 esters of high amylose starch. Two examples using higher vinyl esters (vinyl laurate) were reported by Mormann. The mechanical properties of products with longer ester chains (C4 and C6) products and plasticized acetate esters were considerably improved compared to native starch. It is cheap and abundantly available . However.Chapter 5 5. et al [15‐19] synthesized longer chain fatty esters (C8‐C18) of potato starch and corn amylose using fatty acid chlorides and pyridine. Thiebaud. et al . Introduction Plastics are very useful materials and are used in large amounts (245 million ton per annum) [1‐2]. The largest application area of plastics is the use as packaging material (37%) . the physicochemical properties of the products were not mentioned. and food service‐ware . The use of vinylesters has also been explored [20‐22]. Unfortunately. and this causes serious environmental problems [1. packaging material. The limited use is mainly because of a number of unfavourable properties of starch such as low moisture resistance. and incompatibility with hydrophobic polymers . The use of native starch as a building block for novel biodegradable polymers. the mechanical properties were still not satisfactorily.65) . however. The products were hydrophobic and the mechanical properties were considerably improved compared to native starch. 0. Esterification is one of the oldest methods used to improve starch properties . is rather limited to date. high brittleness.
and toluene were obtained from Labscan (Ireland). The thermal properties (Tm. All these chemicals were used as received. Japan). Germany). the thermal and mechanical properties of representative examples of the highly hydrophobic materials are described and discussed. The catalysts for the esterification reaction were analytical grade potassium carbonate (Boom. TGA measurements were performed using a Perkin Elmer TGA 7 Thermogravimetric Analyzer. and disodium hydrogenphosphate dodecahydrate (both Merck. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). T‐Bones samples (with thickness of 2 mm) for determination of the tensile properties were prepared using a melt press 1 87 . pyridine. 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 . Germany). DSC analyses were performed on a TA Instruments DSC 2920. vinyl laurate (Fluka. Materials Corn starch (approx. and tetrahydrofuran (THF) were supplied by Acros (Belgium). sodium acetate.2. In addition. Materials and Methods 5. catalyst type and intake) on the DS of the products.96) were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts in DMSO.2. The spectra were recorded at 50 oC. This chapter describes systematic studies using Design of Experiments (DOE) to gain insights in the effect of process variables (temperature. Tc. and acetic anhydride (Merck. The samples (about 10 mg) were placed in sealed aluminum cells. 5. Germany) were used without further purification. Analytical grade methanol. each sample was cooled to ‐50°C and then heated again to 200 oC (heating rate 10 oC/min). vinyl ester to anhydroglucose ratio. 4‐N. and Tg) of each sample were determined from the spectra related to the cooling run and the second heating one. The samples were heated to 700 oC in a nitrogen atmosphere with a heating rate of 10°C/min. It was dried for 48 hour at 105 oC under vacuum (~1 mbar) to reduce the moisture content below 2 % before use. After a first heating run from room temperature up to 200 oC to delete the thermal history of the material.24 ‐ 2. 5. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. the Netherlands). Analytical grade vinyl stearate (Aldrich. Technical grade dimethyl sulfoxide (DMSO).2. as recommended by Laignel et al .2.1. Long fatty esters of corn starch with a broad range in degree of substitution (DS= 0. The experiments were modeled using non‐linear multivariable regression.N‐ dimethylaminopyridine (DMAP).
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 22.214.171.124. 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. 126.96.36.199. 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 5.00 3.44 2.00 5.00 2.00 2.00 4.00 5.00 3.00 5.00 3.00 5.00 3.00 3.00 4.54 2.00 2.00 5. 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] 3.46 1.00 2.00 2.00 5.23 2.2.27 0.00 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 Catalyst Temperature [%‐w] [˚C] 2.75 1.27 1.00 5.53 1.44 2.00 2.00 3.59 2.00 3.62 0.67 2.00 2.50 2.00 3.17 2.93 0.00 3.84 2.00 5.00 3.00 5.50 2.00 5.52 2.00 5.00 5.00 5.00 2.00 5.00 5.12 0.79 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 .50 2.00 5.50 0.00 5.00 3.00 5.00 2.00 5.00 5.00 5.00 5.00 3.87 1.00 5.54 2.00 3.00 3.00 5.00 5. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.00 3.00 5.00 2.96 0.00 3.00 2.00 4.50 2.88 2.52 2.27 1.00 5.00 2.00 4.40 1.72 2.28 2.00 3.41 1.94 2.00 3.00 5.42 2.35 0.00 5.00 2.82 2.00 5.00 5.00 2.00 5.00 5.47 1.00 3.56 1.00 5.07 0.00 5.00 3.00 5.48 0.32 1.12 2.00 5.00 3.24 2.70 2.11 2.09 0.00 3.78 2.00 2.
00 5.00 2.0001 0. The type of catalyst was quantified using the basicity constant (pKb) in water [26‐27].0010 ‐0. 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 (5.1.1082 ‐0.50 2.00 5.).0732 ‐0.1947 β0 β1 β2 β3 β4 β5 β6 β7 β8 β9 β10 β11 β12 β13 92 .96 2.00 5.00 3.21 2.0470 0. stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 4.50 95 80 80 110 110 80 80 110 110 95 1.00 5.0327 ‐2. Values for the coefficients of the DS model for starch laurate and stearate Coefficient Starch Laurate 9.00 5.90 2.3.0011 0.1748 0.55 2. Mathematical Modeling Modelling of the DS data for both vinylesters was performed using non‐linear multivariable regression based on the 4 independent variables (Table 5.0006 ‐0.0008 ‐0.0009 0.00 3.0857 ‐0.00 4.6939 0.98 2.1568 ‐2.00 3.64 2.3.0017 0.0120 0.3.2183 0.59 2.24 2. The experimental data for both starch laurate and starch stearate esterification are best described with a quadratic model including interaction terms (eq.0003 0.00 3.0023 0.) Table 5.00 5.). 5.0066 0.00 2.00 3.41 mol ratio of the vinyl ester to AHG units of starch 5.0027 ‐0.5650 0.00 5.3.0016 0.00 2.00 5.1979 Starch Stearate 17.0008 0.Chapter 5 45 46 47 48 49 50 51 52 53 54 a.00 3.0249 ‐0.The modelled values for βij for the two esters are given in Table 5.1.1104 ‐0.64 2.3.
8167 21.5 1 1 0. The very low P‐ values indicate that the models are statistically significant.967 0.0583 F 27.026 x 10‐7 R2 values R2 R2adjusted R2press 0.1. which indicate  that all significant variables are included in the model.875 Starch Stearate Model Error Total SS 20. Also in this case the reasonable R2 values (0. Analysis of variance for DS models for starch esterification Starch Laurate Model Error Total SS 16. which represents an “internal” validation method for the model. the adjusted‐R2 values are very close to the R2 ones.4986 16.629 P‐value 1.4.5 2.939 0.945 0.5 3 0 0 0.5 0 0 0.0356 F 35.5 3 a. we also performed a PRESS analysis  (see corresponding R2 values in Table 5. and confirm this statement. The parity plots of both models are given in Figure 5.8386 DF 13 14 27 MS 1.5 2 DS predicted DS predicted 2 1.9531 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 5.5 DS actual 2 2.5 DS actual 2 2.2569 0. Table 5. The R2 values for the models are 0.2903 P‐value 1.5 0.970 0.1. starch laurate b.5 1 1.96 x 10‐8 R2 values R2 R2adjusted R2press 0. Finally.970 (starch laurate) and 0. Parity plots of the DS models for starch laurate and stearate Moreover.805 3 3 2.7699 DF 13 14 27 MS 1. starch stearate Figure 5.4. respectively and it may be concluded that the models describe the experimental data well.6118 0.805‐0.3399 0.875) indicate that the model correctly predict the 93 .5 1.5 1 1.967 (starch stearate).4.).
82).18) and CH3COONa (1. the type of catalyst has the largest effect on the product DS for both the laurate and stearate esters (Figure 5.2. A similar trend was observed by Dicke  for the acetylation of Hylon VII starch using vinylacetate.). 94 . Catalyst performance of CH3COONa was slightly less than for K2CO3 whereas the lowest DS products were produced when using Na2HPO4 as the catalyst. 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. for both starch laurate and starch stearate are given in Figure 5. such as carbonate or acetate salts leading to higher DS esters. 22]. To evaluate the effect of each variable on the product DS. 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. Of all the variables studied. However.2. Another explanation may be related to the regio‐chemistry of the reaction. 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. K2CO3 is a stronger base (pKb 3. whereas catalyst performance is considerable lower. The highest product DS values were obtained using K2CO3.00) than for K2CO3 (2. the base strengths of the catalysts in DMSO are not known. Unfortunately. On the basis of the sequence. The pKb for the latter (6.Chapter 5 products DS as function of the process variables within the range of experimental variables. This was not the case for alkaline catalysts.66) than CH3COONa (pKb 9. This is indeed the case when comparing the performance of K2CO3 with CH3COONa. it can be rationalised that the rate of the reaction (and thus the product DS) will be higher when using a stronger base. The product DS for Na2HPO4 was considerably lower (1.25) and this leads to higher product DS values for K2CO3. 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. two plots showing the dependence of the DS on reaction temperature and catalyst amount (in %‐w). the performance of Na2HPO4 does not follow this trend.8) is intermediate between that of the other two catalysts.
As expected and in 95 .2. starch stearate Figure 5.2. starch laurate b.). 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. 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. although to a lesser extent.
.2.39) were determined.45) and high DS values (DSlaurate = 2. the vinyl ester to AHG ratio has the smallest effect on the product DS. and the results for all samples are summarised in Table 5. Of all variables studied.3.5. Representative DSC spectra for the starch stearate are given in Figure 5. DSC analysis of starch stearate esters ( ) DS 1.3.3. higher reaction temperatures and catalyst intakes lead to higher product DS values. . DSstearate = 2. 5.) DS ( 2.39) 96 . heating Exothermic Tm = 21 oC cooling Endothermic Tc = 12 oC -20 -10 0 10 Temperature ( C) o 20 30 40 Figure 5. Product Properties The thermal and mechanical product properties for two starch‐laurate and stearate samples with medium (DSlaurate = 1. DSstearate = 1.75.Chapter 5 line with studies on starch acetates .26.45 (.
a 21 Tc [˚C] n.d. 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 2.39 1. these are the first examples of completely amorphous starch esters. The melting point of the potato starch stearate was close to that of model compounds for the stearate side chains (methyl stearate. n. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. especially in the amylopectin part [15. or glass transition above 0 oC) were even absent for the high DS products. At moderate DS values the products still display a melting temperature (at 20‐30 oC) and a crystallization temperature (at 10‐25 oC).a 12 a. 40‐42 oC. To the best of our knowledge. 17].26 1.d. octadecane. 97 . Thermal transitions at relatively high temperature (melting.8 . 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). esterification of starch with vinyl laurate and stearate reduces the crystallinity of the starch considerably and leads to a lowering of the transition temperatures.).a 22 n. These values are reduced considerably compared to virgin corn starch.45 Tm [˚C] n. In any case. For instance a Tm of 32 oC was reported for a potato starch stearate ester with a DS of 1. TGA analysis was performed to study the thermal degradation behavior of the starch laurate and stearate samples (Figure 5. 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.d. 22]. in the present work. Unfortunately the thermal transitions for the virgin material lie above the degradation temperature and can therefore not be accurately determined . crystallization.75 2.5. implying that these products are fully amorphous. However.a 27 n.= not detectable The thermal behavior of the products is a strong function of the DS. close to the value obtained in our study for the medium DS starch stearate (21 °C).d. In particular the behaviour at relatively high temperatures (>0 oC) is further discussed.4. which is known to contain crystalline areas. Changes in the transition temperatures of native starch by esterification have been reported in the literature [13‐14.
TGA analysis of various starch samples.Chapter 5 100 80 %-weight 60 40 20 0 20 100 200 300 400 o 500 600 700 Temperature ( C) Figure 188.8.131.52) The TGA (Figure 5. Similar trends were reported for amylose octanoate‐ stearate‐laurate esters  and potato starch octanoate‐ laurate esters .) curves clearly show starch esterification with either laurate or stearate results in products with enhanced thermal stability.26) (. Native corn starch degrades at lower onset temperatures (250‐300 oC) than the corresponding esters (300‐350 oC). . The mechanical properties (stress at break. The difference in thermal stability between the laurate and stearate sample is limited. elongation at break and elasticity modulus) of the products were determined (Table 5. .6.): starch stearate (DS 2.). 98 . ( ): native starch ( ): starch laurate (DS 2.
45) displays a very similar behavior with respect to the laurate samples (i. determined from the stress‐strain curves as the stress at which the material ceases to be linearly elastic) and the modulus.26 1. Both starch laurate samples display.2 8.39 1. An inverse behaviour is.9 46. This is in agreement with previous studies regarding the mechanical properties of starch esters with different chain length of the fatty acid moiety [13. On the other hand.6.0 3. This indicates that their mechanical behavior is comparable to that of typical engineering polymers . 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. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. however.3 3. and lower elongation at break).7‐3.0 112.).1 Elongation at break [%] 26. but higher elongation at break. observed for the high DS starch stearate sample. 17].1 82.7 ± 4.7 ± 3.75 2.9 ± 14. The values of the yield strain.7 ± 14.7 ± 1.5 ± 1. thus indicating a relatively more rigid material with respect to all other samples. independently of the DS values. is roughly 0.2 2.1 Modulus [MPa] 59.1 ± 1. starch stearate at relatively high DS values (2. a clear plastic behavior with a maximum in the stress‐strain curves. A deeper understanding of the mechanical behavior can be gained by investigating the shape of the stress‐strain curves (typical examples in Figure 5.39) does not show any plastic behavior but only an elastic one.5. This product is more rigid compared to the product with medium DS value (higher tensile strength and elasticity modulus.0 3.3 ± 2. plastic deformation and yield strain of about 0. 99 . For starch laurate. The corresponding yield strain (roughly 0.e. Starch stearate with a relatively low DS (1.025) lies still in the typical range of engineering polymers but also close to that typical of composite materials . Mechanical properties of the starch esters DS 2. elongation at break between 3‐26%. an increase in product DS results in lower tensile strengths and modulus of elasticity.7 ± 4.05).05 for both laurate samples.0 ± 2.8 Starch Ester Starch laurate Starch laurate Starch stearate Starch stearate All materials generally show tensile strengths between 2. calculated as the ratio between the stress at yield (σy.45 Stress at break [MPa] 3.1 ± 15.3 21.5 MPa.2 ± 0. and modulus of elasticity between 46‐113 MPa.
.75) ( ): starch laurate (DS 2. 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. Although there are many differences in the synthetics methods as well as in testing conditions and procedures.5. 100 .): starch stearate (DS 1. ‐): starch stearate (DS 2.26) (.7.).45) (‐ .39). Stress‐strain curves for starch esters. ( ): starch laurate (DS 1.Chapter 5 6 5 4 stress (MPa) 3 2 1 0 0 5 10 15 20 strain (%) 25 30 35 Figure 5. .
7 ±0.2 ±0.2 8.5 ±1. 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.a. DS) which mainly determines the thermal and mechanical behaviour.9 ±0.75) 3. the high DS material is still more rigid. merely because of the structural property of the C18 side chain with respect to the starch backbone. 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).7 ±3. n.39) 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. confirms that the corn starch esters synthesized in our research are relatively rigid materials as compared to the potato starch esters.7) 0.0 26. In this respect the difference in stress and elongation at break for the high‐DS starch laurate sample is striking.1 ±1. an unexpected behavior was also observed for the mechanical properties of the starch esters from potato starch.a.6 Corn Starch Stearate Ester Low DS (1. 101 . n. for potato starch exactly the opposite is observed: relatively low stress (0. Here. Moreover.3 Potato Starch Stearate Ester Low DS (1. On the other hand at low DS values comparable stress at break (3.26) relatively high stress at break (3. 9% for potato starch).7) 1.1 ±15.0 ±2. as confirmed by DSC.7 ±0.a.7 ±1. we show that although no crystallization of C18 occurred (see DSC data of the high DS material in Table 5.6 9 ±2 High DS (2. 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) .7 MPa) coupled with a significant (1500%) elongation.5 MPa) is coupled with a low elongation (26 %).2 Stress at Break [MPa] Elongation at Break [%] a. While for corn starch (DS 2.1 21.= not available Inspection of the mechanical properties of the starch esters as shown in Table 5.26) 3.a.5.4 1500 ±8. but the interplay between those factors.8) 3.7 ±4.1 High DS (2.7 MPa for potato starch ones) are coupled with significant differences in the elongation (22% for corn starch esters.7.a.2 MPa for corn starch esters.7. In agreement with our own data.). 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. They related this inverse property of starch stearate to crystallization of C18 side chains. High DS (2.3 High DS (2. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.3 10 ±2 Property Low DS (1.45) 3.0 3.
catalyst intake.1‐2. 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.3‐1. and Na‐acetate). 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. Nomenclature β : regression coefficients obtained by a multiple regression procedure [various units. 0. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] 102 . 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 stearate). 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. The models may be used to determine the appropriate process conditions to obtain a product with a pre‐defined DS. K2CO3. 1. Thus.5.5 for starch laurate. including statistical modelling. Statistically adequate (R2 ≥0.4.07‐1. 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. on the synthesis of corn starch esters with long chain fatty acids is described.Chapter 5 5.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐starch mol ratio. while the use of Na2HPO4 resulted in low‐medium DS products (0. reaction temperature.4‐3 for starch stearate). 5.9 for starch laurate. and catalyst basicity) to the DS of the starch ester products were developed. Conclusions A systematic study. Comparison with literature data indicates that the properties may be (fine) tuned also by the starch source.
Applications.6. Association of Plastics Manufacturers. The Royal Veterinary and Agricultural University. 2000. accessed on August 01. Hernandez. 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. New Jersey. M. in Packaging in the Environment (Ed. Munich.starch. demand and recovery for 2006 in Europe. Bastioli: Global Status of the Production of Biobased Packaging Materials. Res. 1986. C. Stevens: Green Plastics. Processing. Brussels. S. . Ehrenstein: Polymeric Materials. 1980. M. Glasgow. Culter: Plastics Packaging. C. Princeton University Press. Copenhagen. Hanser Publishers. UK.J. . E. Properties. Blackie Academic & Professional. . J. Dev. Denmark. 2000. H. Ind. Chem. W. Weber). Munich. Prod. Denmark: http://www. 592‐595. Otey. J. Structure – Properties ‐ Applications. 103 . E. USA. . The compelling facts about plastics: An analysis of plastics production. A. Frederiksberg C. F. Belgium. Doane: Starch‐based blown films. Website of International Starch Institute. Conference Proceedings. 2002. D. Germany. Denmark. Department of Dairy and Food Science. R. Hanser Publishers. R. G. Aarhus. Marsden: The challenge of domestic waste disposal. Levy). Plastics Europe. S. 2008.dk/isi/stat/rawmaterial. An Introduction to the New Science of Biodegradable Plastics.W. G. 2000. Selke. Germany. M. 2008 . . Packaging. References . and Biopolymers. and Regulations. 27‐29 August 2000 (Ed. Westhoff. 19. Eng.html. . in The Food Biopack Conference: Foodstuffs. P.
E. 51. Prinos. C. . 57. E. Alric. 1942. . Merrill: Properties of fatty‐acid esters of starch. 132‐135. I. E. E. M. and biodegradability of fatty‐acid esters of amylose and starch. C. W. Bikiaris. 28. . Al‐Higari: Acylation of starch with vinyl acetate in water.. Sci. Panayiotou: Properties of octanoated starch and its blends with polyethylene. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Polym. D. Department of Dairy and Food Science. Thiebaud. Chem. Panayiotou: Synthesis. A. B. 101‐112. I. Thiebaud. X. J. 35. Polym. Alric. G. A. Rev. 1995. R. Broekroelofs. Ind. Mullen. 1943. C.Chapter 5 . Mormann. Weber). 118‐121. J. . Biobased packaging materials for the food industry. 1997. Borredon. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. Thiebaud. 1999. Aburto. Eng. 1440‐1451. P. Starch‐ Starke 2005. G. 65. D. Pacsu: Starch studies: possible industrial utilization of starch esters. de Graaf: Ph. W. J. K. Alric. . Prinos. 302‐307. J. D. M. Mouysset‐Baziard. I. Ind. F. Starch‐Starke 1999. Macromol. E. Alric. Aburto. Mullen. Pacsu: Starch studies: preparation and properties of starch triesters. Appl. Chem. E. Starch‐Starke 2004. characterization. E. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. 34. A. M. Thesis. Aburto. Janssen. Part C: Polym. Hamaili. W. Starch‐Starke 1999. L. W.D. Yang. 381‐384. 137‐144. Sci. Aburto. 1996 104 . Sci. . 74.. A. Wang. Carbohydr. . 56. J. Alric. 34. Borredon. H. Prinos. J. J. J. A. 1995. the Royal Veterinary and Agricultural University. Polym. S. Appl. J. Y. J. Sagar. . Carbohydr. Bikiaris. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. J. E. I. Z. 58. D. . report of Food Biopack Project (Ed. 705‐721. J. Bikiaris. Wang: Properties of starch blends with biodegradable polymers. Frederiksberg C. Polym. 145‐152. L. A. J. J. C. Eng. 385– 409. I. J. Borredon. Sci. Senocq. . C. Aburto. Polym. S. 2003. J. E. Aburto. . R. 1647‐1656. de Graaf. I. 51. K. Appl. . Denmark. Alric. status and perspectives. 1997. 2000. S. C43. 1209‐1217. Beenackers: The kinetics of the acetylation of gelatinised potato starch. . Rijksuniversiteit Groningen.
Montgomery: Design and Analysis of Experiments 5th edition. A. L. the Netherlands. Picchioni. . Shercliff. A. 1986. C. Massiot. A straight way to regioselectively functionalized polysaccharide esters. 1997. B. 251‐260. C. 2007. B. L. Manurung. Carbohydr. Day. 255‐263. Cebon. R. 105 . Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch . Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. . Cellulose 2004. G. L. Laignel. New Jersey. Junistia. A. Materials: Engineering. accepted for publication in Starch‐Starke. John Wiley & Sons Inc. H. Sugih. F. R. J. 2008. Janssen. H. D. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. . R. Bliard. . 298. Prentice‐Hall. New York. J.. Science. Processing and Design. Ashby. USA. K. . Res. USA. M. M. Butterworth‐Heinemann. 11. Underwood: Quantitative Analysis 5th edition. Dicke. P. M. Amsterdam. D. A. 2001.
The reactive interfacial agents were tested for their performance in PCL‐starch blends. In particular the elastic modulus for the compatibilized blends is significantly higher than that of uncompatibilized ones. 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. As a result. This discrepancy could be explained on the basis of the functional groups (GMA or DEM) distribution along the PCL backbone. 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‐ glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM) is decribed. despite the expected differences (favorable to PCL‐g‐GMA) in the in situ formation of the compatibilizers. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. PCL‐g‐DEM has at least a comparable performance with respect to PCL‐g‐GMA. the mechanical behavior of the compatibilized blends is in general different from that of pure PCL and of the corresponding uncompatibilized blends. At relatively low starch intakes. grafting .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. Both act as compatibilizers for PCL/starch blends by improving the interfacial adhesion between the starch particles and the PCL matrix. compatibilizer. glycidyl methacrylate. at relatively high GMA and BPO intakes. Keywords: starch‐PCL blend.
108 . One of the well known modification strategies is blending the starch with polymers displaying a stronger hydrophobic character and better mechanical properties.Chapter 6 6. a compatibilizer (i. as in this study. Among these. synthetic biodegradable polymers have been applied. To overcome this issue. polycaprolactone (PCL). To reduce the tendency for phase separation. an interfacial agent) may be used to improve the interfacial association between the two polymer phases. Blending of starch and synthetic biodegradable polyesters has been widely applied for the synthesis of fully‐biodegradable products. of polar groups (usually epoxides or anhydrides [13‐18]) able to react with the – OH groups along the starch backbone. PCL has gained considerable interest for possible applications in the fields of packaging materials and medical applications [10‐11]. such as polyethylene (PE) or polystyrene (PS) [3‐6]. blends of hydrophilic starch and hydrophobic biodegradable polyesters exhibit phase separation  due to differences in polarity of the building blocks. these synthetic polymers are poorly or non‐biodegradable. This phenomenon is highly undesirable and limits the application range considerably . Unfortunately. is known to be degraded with ease by microorganisms widely distributed in nature . to an ideal compatibilizer having both PCL and starch blocks. polyesters are considered very promising alternatives . 2].e. When using ungelatinized starch as a component in the blend. 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.1. Introduction Starch is a cheap and abundantly available natural polymer with very good application perspectives in the area of biodegradable plastics. Starch modifications to improve the product properties like enhanced hydrophobicity and mechanical properties were already reported in the early 19th century [1. Such structure is rather difficult to achieve by simple copolymerization methods and it is usually prepared in situ (i. The ester bonds are susceptible to attack by water and this leads to enhanced biodegradability. As a consequence. along the backbone. This leads. for starch/PCL blends. it would be actually more accurate to define the block copolymer as an “interfacial agent”. upon mixing) by using a functionalized PCL.e. The latter displays the presence. Unfortunately. 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. 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 ). A well‐known biodegradable polyester. However. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decrease rapidly in time .
After the PCL was melted (1‐2 minutes). The starch was dried for at least 24 h at 110oC under vacuum (approx.8%) from Labscan.1. Mw=3000 from Solvay Caprolactones. Materials Polycaprolactone (PCL) CAPA 2304. Belgium.1. Intakes for each experiment are given in Table 1 and 2.2.2. 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. UK. an absolute novelty of the present research. Corn starch (with approx. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). Ireland. The synthesis and application of PCL‐g‐DEM is. and benzoyl peroxide 75% (Merck. Glycidyl methacrylate 97% purity (Aldrich). >99%) was obtained from Acros. Exploratory studies on the synthesis of PCL‐g‐GMA and its applications for starch/PCL blends have been published [13. to the best of our knowledge.2. The various compatibilizers have been tested for their performance in PCL‐starch blends.2. 109 . Mw=50000) from Solvay Caprolactone. diethyl maleate ≥97% purity (Fluka). however. Germany) were used as received. ~1 mbar) prior to use. This low molecular weight PCL grade was used without further purification. This chapter describes a systematic study on the synthesis of two reactive interfacial agents for starch‐PCL systems. Materials and Methods 6. systematic studies and subsequent quantification of the functionalization reaction has not been reported to date. xylene (99. 6. Methods 6. 6. Tetrahydrofuran (THF.2. The materials were mixed for another 5 minutes. after which the equipment was stopped and the chamber was opened to collect the samples. 73% amylopectin and 27% amylase) was obtained from Sigma and high molecular weight PCL (CAPA 6503.2. The kneader was heated to 130 oC and PCL (CAPA 2304) was added while maintaining a rotational speed of 80 rpm. a solution of BPO in GMA or DEM was added drop by drop over a period of 5 minutes. 16]. UK was used for the preparation of the interfacial agents. Germany and methanol (99. 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.8%) from Merck. Compatibilizer synthesis The compatibilizers were prepared in a Brabender Plasticorder PL2000 batch‐ kneader (chamber volume 35 cm3). The intake of reagents was maximally 75‐80% of the volumetric volume to ensure proper mixing.
2.2. created using a Cressington 208 sputter coater. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. Methanol (450 mL) was added to the filtrate. 5 mbar).4. 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 content was blended for 15 minutes. PCL‐g‐DEM (5 g) was dissolved in xylene and refluxed at 150 oC for 2 h.2. Each sample was first heated from 0 oC to 100 oC (heating rate 10 oC/min) to remove the thermal history of the material. The resulting suspension was filtered and precipitated using methanol (450 mL) at 6‐8°C. Scanning Electron Microscopy (SEM) was performed using a Jeol 6320 F Scanning Electron Microscope. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. the chamber was opened and the resulting material was collected. Infrared spectra were collected with a FT‐IR apparatus in the ATR mode using a Spectrum 2000 instrument from Perkin Elmer. Analytical Methods H‐NMR measurements were performed using a 400 MHz Varian AMX Oxford NMR apparatus with CDCl3 (99.3. Digital Scanning Calorimetry (DSC) measurements were performed using a Q1000 TA Instruments equipped with a TA Instruments DSC cooling system.5 h. 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 . 6.3. 6. 6. Tensile tests were performed using an Instron 4301 1 110 . Work‐up of PCL‐g‐GMA products  To remove unreacted GMA monomer and GMA homopolymer. the samples were covered with a paladium/platinum conductive layer of 3 µm thickness. Aldrich) as the solvent.2.2. An operation temperature of 170°C and a rotation speed of 80 rpm were applied . 5 mbar). Subsequently. stirred for 1.Chapter 6 6.8%. The error on the transition temperature is assumed to be ± 1 oC and 5 % of the calculated values for the corresponding enthalpies.2. Before analysis. and then filtered. PCL‐g‐GMA (5 g) was dissolved in 50 ml THF. 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. 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).2.2.
19]. leading to a 10% relative error in the FD values. stress at break (σ) and modulus (E) were measured.2 ppm) or DEM (‐CH2‐ protons at δ 4. For every T‐bone. 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.0 ppm [13. 6. 6. 111 . 8 different T‐bones were used.2 ppm) side chains with that of a characteristic proton resonance of the PCL backbone (‐CH2‐ protons at δ 4. 15‐16.1. When the relative degrees of freedom (DF) are known. A 5% relative error in the peak area of the NMR spectra was assumed. the F‐ value for the model is determined followed by the P‐value. This procedure is described in detail in the literature  and consists of calculating the sum of squares (SS) for the model and the error.3. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends machine. 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). As a result we were able to obtain a model for the FD of the reaction.4. The validity of the model was determined by performing an analysis of variance (ANOVA. For a given sample/blend. 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. 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. it is possible to calculate the mean square (MS) for the model and the error. On the basis of the latter values.) The FD was calculated using 1H‐NMR by comparing the area of protons belonging to the GMA (‐CH< proton at δ 3.2.2. Table 3). strain at break (ε). The latter is a measure of the statistical significance of the model.1. The T‐bone samples were prepared using a Fontijne Holland TH 400 hot‐ press.
1.1. Results and Discussions 6.1.1. 112 .Chapter 6 6.). 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 6.3. 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 6.3. 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.
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
However. BPO=0. Hence. higher FD values for GMA than for DEM are in line with this explanation and support the proposed molecular reaction mechanisms. Effect of GMA and DEM to CL‐units ratio (mol/mol) on the FD of the products (constant PCL intake. Our experimental findings.1% mol/mol-CL units 10 5 0 10 20 30 40 50 60 Substrate Intake (mol% on CL units) Figure 6. This trend is independent of the BPO amount and matches with data reported by other groups [13. 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. The results are graphically provided in Figure 6.3. 116 . Thus an increase in the GMA intake will provide more monomer available for the growing of the grafted chains leading to higher FD values. i.3.1.). showed that MA reacts easily with a radical on the PCL backbone. the length of a MA graft is always unity whereas longer grafts are possible for GMA. BPO=1. 30 25 20 FD (%) GMA.6% mol/mol-CL units DEM. It is most probably related to the fact that GMA is able to propagate to longer grafted chains when reacted with PCL (Figure 6. 22]. 130°C) Higher GMA intakes lead to higher FD values.e.6% mol/mol-CL units 15 GMA.1.Chapter 6 chemical structure of DEM.2. BPO=0.3% mol/mol-CL units DEM. subsequent reactions of MA to an already grafted MA molecule hardly occur.3. BPO=0.
6. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The experimental trend for DEM is different.8 0. for which a levelling off and eventually a decrease of the FD values for relatively higher MA amounts has been observed.9 1 1.1 1.7 0. doubling the amount of initiator results in considerable higher product FD. 15. This will lead to a higher proportion of PCL radicals by hydrogen abstraction from the polymer backbone and thus to higher FD values. 17]. Apparently. constant substrate to PCL ratio) For high GMA to CL ratio (24%‐mol/ mol CL units). Such discrepancy is probably related to the relatively low intake of DEM as well as to the different reactivity of DEM compared to MA . Such behavior is slightly in contrast to what observed for maleic anhydride [13.2.5 0. Effect of the BPO intake on the product FD (130°C. However. GMA=12% mol/mol-CL units DEM. These results are in line with earlier work [15.2 GMA.3.4. no detectable influence of the BPO amount on the FD is observed (Figure 6. The FD values are within a rather narrow range. although a slight increase in the FD values might be appreciated.3 0. 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 6.6 0.). there 117 .2 BPO Intake (mol% on CL units) Figure 6.4.).4 0. GMA=24% mol/mol-CL units GMA. 17].4.1. at relatively lower GMA intakes. DEM=30% mol/mol-CL units 0. The use of higher initiator concentrations will result in an increase in the number of formed radicals. 50 45 40 35 30 FD (%) 25 20 15 10 5 0 0.
6. Here. To quantify synergic effects of monomer and initiator molar intakes on the FD of GMA and DEM on the PCL backbone.0325nmδ + 0.3. ni the molar amount of initiator in the feed and δ the difference in solubility parameters calculated using group contributions .Chapter 6 is an optimum ratio between the BPO concentration and available monomer (GMA in this case) on the FD.1. 17].3.1. namely an increase in the BPO intake results in higher product FD. as well as several transfer reactions might be responsible for the observed trend . This phenomenon is likely related to the different nature of the radical grafting mechanism of GMA and DEM on PCL as mentioned before. 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. recombination of (macro) radicals. 6. In particular the occurrence of “cage effects”.8875 + 0. the possibility of side reactions will become significant. the BPO and monomer intakes are considered as independent parameters.e.6022nm ni − 0. However.) where nm is the molar amount of monomer in the feed. 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). defined as the difference in solubility parameters between PCl and the substrates. the mutual solubility of DEM in PCL and DMA in PCL was included in the model by using a parameter δ.3.2. The model provided in eq. a statistical model has been developed by performing a multivariable linear regression on the data reported in Table 6. In addition. and 6. This leads to the following equation: FD = f (nm . δ ) (6. 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.) 118 . The effect is however much less pronounced than for GMA. ni . gives the best description of the experimental data: FD = −1.5431niδ + 1.3. 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).2729nm niδ (6. If many macroradicals are created in the system (relatively high BPO intakes) at low GMA intake.2. 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. i.
908 10. This was also confirmed by inspection of the residue distribution by a normal probability plot (not reported here for brevity) . Table 6.6 20 10 20 40 50 30 60 FD (%) 40 FD (%) 30 40 20 0 1 0.358 DF 4 8 12 MS 456.8 0.3.6 0.4 BPO Intake (mol% on CL units) 0.3. 3D plot.17 F 44. b. respectively. 1 0.2 10 20 GMA Intake (mol% on CL units) 40 30 50 0.7 0.5 0.941) also suggests that all important variables have been included in the model.8 0. the interaction between (PCL‐monomer) intake and solubility parameter difference ( nmδ ) and the interaction between the latter factor and the initiator intake ( ni δ ). Contour plot. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The analysis of variance data are given in Table 6..6. The very low P‐value implies that the model is statistically significant.4 0. Graphical representation of the FD models for PCL‐g‐GMA and PCL‐g‐DEM are given in Figure 6.5. Figure 6.3 10 10 20 30 20 0.3. Model Error Total Analysis of variance for the FD model provided in eq 6.358 1909.2 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.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 ). 119 .957) and its closeness to the adjusted R2 (0. and Figure 6.9 60 70 40 50 80 BPO Intake (mol% on CL units) 0. SS 1828 81. The R2 value for the model (0. b.5. a.
and 6. 3D plot.Chapter 6 1 0. and a favored nucleation 1 2 50 120 . which is expected to lead to a Tc reduction.6 0. Graphical representation of the FD model for PCL‐g‐DEM. b. 7 6 3 3 4 Figure 6. radical initiator and PCL intakes).6 2 FD (%) 5 2 0 1 0. Thermal properties of the compatibilizers The thermal properties (Table 6. the crystallization temperature and enthalpy are not a clear function of the FD values.8 0. For relatively high DEM intakes (> 30 % mol/mol). 6. Similar observations have also been made by Kim. a higher BPO intake leads to a reduction of the FD values while an opposite trend is observed at lower DEM intakes.5 0.3. While for GMA an increase in BPO or monomer intakes invariably leds to higher FD values.4.2 10 20 DEM Intake (mol% on CL units) 40 30 50 6 0. Remarkable is once more the different trends for the two substrates.6. 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.8 0. b. although both properties are significantly lower than those of pure PCL.2 10 15 20 25 30 35 40 DEM Intake (mol% on CL units) 45 7 a. This induces irregularities and is expected to result in a lowering of the Tc and the Tm. For all samples. Furthermore. for DEM a clear transition is observed. From a more practical point of view.7 3 4 6 FD (%) 4 0. a.2. the melting temperature and the relative enthalpy decrease with respect to pure PCL.) of the compatibilizers were determined by DSC.1. The random behavior of the Tc as function of the FD values is probably the result of two concurring effects: the presence of irregularities. et al  working with PCL‐g‐GMA. Contour plot. 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. Such changes in the thermal properties compared to pure PCL may be caused by the introduction of grafts on the PCL chains.9 4 5 8 BPO Intake (mol% on CL units) 0.1.4 0.4 BPO Intake (mol% on CL units) 0.3 5 6 0.
4 230.9 368.2 337.3 15.6 168.7 10.4 431. PCL‐g‐GMA has an FD of 9.1 10. 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. 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).5 357.6%.1 13.8 372.6 5.0 332.0 430.5 7.6 386.3.6 11. 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.0 341.2 342. 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.0 401.0 9.2 343. 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) 16.9 424.9 E (MPa) 270. An overview of all prepared blends together with their thermal and mechanical (tensile tests) properties is given in Table 6.1 11. 6.6 305.9 11.2 9.5 ε (%) 640.9 379.3 10.4.8 371.3 380.5 489.5 425.7 %. Table 6. 121 .4. 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.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 1.2. A total of 12 blends were prepared: a series of binary ones (not containing any compatibilizers) constituting our reference points.8 329.8 384.2 321.
3. The thermal and mechanical properties indicate no or very little interaction of the starch particles with the PCL ○ : ∆Hc □ : ∆Hm 122 . respectively from 16. Melting and crystallization enthalpy as function of the starch content for binary blends with PCL. 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 6.7.1 MPa and 341. 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.1.2 MPa up to 341.2. 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 .43 MPa and 640.8 % at 30 % starch content. This is most probably due to a nucleation effect of the starch on PCL as observed also for other kinds of polymer blends .). 75 70 65 60 55 50 45 40 35 ∆ H (J/ g-PCL) 0 5 10 15 20 Starch content (%) 25 30 Figure 6.5 % for pure PCL to 7.Chapter 6 6.7. The simultaneous increase in the modulus (from 270.8 MPa) clearly indicates that the rigidity of the blends increases at higher starch contents. 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.
2. b. 6. c. a. Evaporation of the water during blend preparation.4. 123 .8. 26] and explained by assuming that the voids are formed by water in the starch. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends matrix. (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.e.). Figure 6. Morphology of starch/PCL binary blends. However.8.2. for which a 6 oC drop is observed when using PCL‐g‐DEM. The thermal behavior is characterized (Table 6.) 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. the starch particles are simply inserted into voids of the PCL matrix. As consequence one might expect a slight different structure of these ternary blends as compared to the corresponding binary one. This is confirmed by examination of the blend morphology by SEM (Figure 6. The latter trend is also valid for the corresponding enthalpies: i. in order to fully understand the role of PCL‐g‐DEM. Thus. In this respect the thermal properties remain substantially unchanged with the exception of the Tc. Such behavior is consistent with the hypothesis that the starch particles hinder the melting and crystallization processes of the PCL chains. Ternary blends compatibilized with PCL‐g‐DEM PCL‐g‐DEM. This is visually confirmed by the morphology of the prepared blends (Figure 6. the corresponding binary blend (S/PCL 20/80) contitutes a better reference point compared to pure PCL. and the lack of interface adhesion cause void formation. whose synthesis is described for the first time in this work. either by heating or in combination with mechanical stress .). The presence of these voids was also observed for sago starch/PCL blends [25.3. was used as compatibilizer for the preparation of ternary blends with starch and PCL.9.
Indeed starch particles in the ternary blends display a smoother interface with the PCL matrix compared to the corresponding binary blends (Figure 6. 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. c.). b. This can be explained by the improved interfacial adhesion  between PCL and starch. Figure 6.Chapter 6 a. 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. Morphology of ternary blends compatibilized with PCL‐g‐DEM.). (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 .).10. resulting in more rigid material with higher modulus.9. which will hinder the flowability and fibre forming ability of PCL matrix under cold drawing. As a result of the different morphology also the mechanical behavior display differences with the virgin PCL and the binary blend. At 1 %‐wt of PCL‐g‐DEM in the blends the modulus increases with respect to the binary blend (0 %‐wt compatibilizer in Figure 6.10. 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.9.
3.e. the modulus is higher and hardly a function of the compatibilizer amount.10. In particular also in the case of PCL‐g‐GMA. in all cases (i. 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.3. 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. the crystallization temperature experiences a drop of about 5‐6 oC. 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). 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). the stress and strain at break are lowered and hardly a function of the PCL‐g‐GMA amount. at all PCL‐g‐GMA contents) there is a clear increase of the modulus with respect to the 125 . 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. However.2. 6. Modulus of ternary blends S/PCL/PCL‐g‐DEM as function of PCL‐g‐ DEM amount. by taking as reference the corresponding binary blend (S/PCL 20/80).
As a result. a. This is surprising if one takes into account the different FD values: 9. c.8 % for PCL‐g‐GMA against 1. presence of voids. (left) the structure of PCL‐g‐GMA is inhomogeneous at the molecular level with long poly(GMA) branches pending from the PCL backbone. and morphologies in Figures 6. (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. 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 6.11. Morphology of ternary blends S/PCL/PCL‐g‐GMA.) than PCL‐g‐GMA for blends containing 20 %‐wt of starch. As given schematically in Figure 6.). Thus. 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. 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 [27.12. 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.11. partial interfacial adhesion) which render these ternary blends a kind of “intermediate” case between the binary ones and those compatibilized with PCL‐g‐DEM. This confines all reactive GMA groups in relatively “isolated” spots along the PCL backbone.9. and 6.7 % for PCL‐g‐DEM. PCL‐g‐DEM is at least as active as compatibilizer (compare modulus values in Table 6. These differences in compatibilizing effect are not yet fully understood.4.Chapter 6 binary blend. 28. 126 . b. Figure 6.11. 29]. However.
when comparing ternary blends (compatibilized with PCL‐g‐ GMA) with the binary ones.) and relating the “coverage” of the starch particle surface upon reaction with PCL‐g‐GMA with the compatibilization efficiency. 20 and 30 %‐wt respectively).4. other groups will presumably remain available (arrows in Figure 6. at higher starch contents more –OH groups would be available for reaction with the GMA groups. 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. Moreover. thus probably ensuring a better coverage of the surface.4.3 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 %) .12.). for all starch amounts (Table 6. but since DEM is preferentially grafted as monomer. Indeed. 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. This is in agreement with the hypothesis made above (Figure 6. the distribution of the reactive group along the PCL backbone is more “homogenous”. As consequence. The latter result (from a modulus of 341.) PCL‐g‐DEM has much less reactive groups (lower FD) than PCL‐g‐GMA. Concerning the thermal behavior (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.) for further reaction. once PCL‐g‐DEM has reacted with one –OH groups on the starch particle.12.8 MPa for S/PCL 30/70 up to 430.12. thus attenuating the effect discussed above. On the other hand (right of Figure 6. 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. In order to check this we compared blends with the same amount of PCL‐g‐GMA (2 %‐wt) but with different starch intakes (10. Schematic representation of the in situ reaction between functionalized PCL and starch.). We can therefore conclude that the efficiency of PCL‐g‐GMA in the 127 .
including statistical modeling. 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. 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.4. mutual solubility of the monomer into the molten polymer) on the FD adequately described the experimental results (R2=0.957.5. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends.Chapter 6 compatibilization of starch/PCL blends can be significantly improved by changing the blend composition. 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. 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 . 6. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). has been performed on the synthesis of two reactive compatibilisers. Nomenclature E FD : initial modulus [MPa] : functionalization degree. The proposed model to quantify the effects of process variables (monomer and initiator intake. The most important variable for the final product FD is the interaction between the amount of monomer and initiator used. Conclusions A systematic study. 6. P‐value ≤10‐9). 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. firstly reported in this work. From a more practical point of view it can be concluded that the newly synthesized PCL‐g‐DEM. The reactive compatibilizers were used in blends of starch with PCL. The PCL‐g‐GMA and PCL‐g‐DEM compatibilizers display lower melting temperatures and melting enthalpies compared to virgin PCL.
. Polymer 2003. Liu. Boca Raton. 1477‐1487. 1998. 157. 1986. 139‐161. S. Mani. 34. J. . Polym. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – IV. J. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – III. R. 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. Ogawa. Rodriguez‐Gonzales. Polym. Inc. M. Jarowenko: Acetylated starch and miscellanous organic esters. 1998. Eur. F. Thermal and morphological properties. USA. A‐C. CRC Press. References . . Adv. N.6. Davis: High performance LDPE/ thermoplastic starch blend: a sustainable alternative to pure polyethylene. 33. Polym. . Psomiadou. wheat starch and soluble starch for food packaging applications. Eur. M. Biliaderis.A. Kawasaki: Biodegradable films made from low density polyethylene (LDPE). Ramsay. E. Carbohydr.Chem. H. O. C. A‐C. Karlsson: Polymers from Renewable Resources. Polym. Wurzburg). Ind. J. E. 1251‐1253. A. Y.Eng. 2002. Pacsu: Starch Studies: Preparation and properties of starch triesters. S. 1209‐1217. Eng. M. . 1997. Ioannis. Polym.. 34. Effect of amylopectin to amylase ratio in starch. R.D. B. 1942.G. 34. Part 2. 227‐242. Mani. 1998.S. Albertsson. Sci. Sci. Mullen. Albertsson: Biodegradable Polymers and Environmental Interaction. in Modified Starches: Properties and Uses (Ed. 38. Ranucci. E. . Lindblad.J. 1467‐1475. 1517‐1526. W. . 44. Karlsson. B.B.W. 129 .
Sci. 29. 305‐313. R. Sci. Polym. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. USA. Cho. 61‐67. . M. . J. John. J. . 42. 2001. R.K. 1998. Appl. K. . J. Environ. M. M. Polym. Marechal. 41. . Rustgi: Biodegradable polymers.M. C. D. 1999. K. K.C. C. Kim.K. Part A: Polym. Z. Avella. 1139‐1148. G. 1997. P. Fringant: Properties of thermoplastic blends: starch‐polycaprolactone. Tang. P. 41. Degrad. Polym. Sci. Martuscelli. R. Jung. Dole. . 1507‐ 1516. Polym. . Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. Chem. Technomic Publication. Montgomery: Design and Analysis of Experiments 5th edition. Polym. C. Macromolecules 1996. Sci. Inoue: Compatibilizers for melt blending: Premade block copolymers.Chapter 6 . Kim. Polymer 2000. 2004. Kim.W.H. J. J. 5590‐5598. 8. Mater.K. . 179‐187. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. It. Aglietto: Chemical reactions affecting the free‐radical grafting of diethyl maleate onto ethylene polymers. Polymer 2001.Y.. M. 127–134.S. E. Kim.M. J. Guegan. J. P. 127.. J. Chiellini. E. Environ. M. 1991. M. Degrad.K. USA. John Wiley & Sons Inc. A. Gazz. J. Adv. . M.S.H. 37. Mani. J. Polym. Solaro: Biodegradable polymeric materials. Painter: Specific Interactions and the Miscibility of Polymer Blends. 130 . C.. 1997. Macosko. . Nakayama. A. Ruggeri. Park: Grafting of glycidyl methacrylate onto polycaprolactone: preparation and characterization. P.F. Pennsylvania. D. 12. P. Averous. 91‐95. Bhattacharya. New York. 1995. Raimo. J. 23. Chim. C. 1273‐1335.E. L.Y. 80. Errico. Graf. Marrucci. Chem. L. Goldberg: A review of the biodegradability and utility of poly(caprolactone). Tang: Functionalization of polyesters with maleic anhydride by reactive extrusion. Part A: Polym. Prog. 2003. . Coleman. 35. 1996. 1693‐1702. Khandpur. 4157‐4167. 81. . C. M. Laurienzo. R. Polym. Park: Modification of Aliphatic Polyesters and Their Reactive Blends with Starch. Passaglia. Stab. Yang.H. 3. J. E. J. T. Cho. . Polymer 2000. Moro. 5135‐ 5142. Chandra. 2001. C. Rimedio: Preparation and characterization of compatibilised polycaprolactone/ starch composites. R. . 3875‐3881.
Odusanya. O. Polym.W. M. R.W. Ruggeri. M.Nasirtabrizi: Ring opening reactions of glycidyl methacrylate copolymers to introduce bulky organosilicon side chain substituents. 2000. Polym. 21. 2002. 38. J. . 1998. 57. Technol. 131 . Polym.S. B. 1298‐1305. Sci. .N. U. H. 87. Eur. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends . W. O.S. Azemi. Polym. . Appl. Bull. Lee. Aglietto. 2006.V.Safa. M. D.D. 273‐281. E.N. A. K. 393‐401. 40.S. Test. Mauler: Comparative study of PP‐MA and PP‐GMA as compatibilizing agent on polypropylene/nylon 6 blends. Ishiaku. Sci. Kammer: On mechanical properties of sago starch/ poly(ε‐caprolactone) composites. Passaglia. Tedesco. Adv. 293‐304. Ishiaku. B.M. Ishiaku. 2002. F. Manan. J. Barbosa. 11‐15.S.M. G.H.A.S. D. Odusanya. 9. Picchioni: Formation and compatibilizing effect of the grafted copolymer in the reactive blending of 2‐ diethylsuccinate containing polyolefins with poly‐ε‐caprolactam (nylon‐6). R.A. . 2003.M. Ishak: Mechanical properties and enzymic degradation of thermoplastic and granular sago starch filled poly(ε‐ caprolactone).M. Polym. K. U. S.B. Azemi: Effect of starch predrying on the mechanical properties of starch/ poly(ε‐caprolactone) composites. 877‐884.S. U. .M. Eng. Nachtigall. Polym. S. Manan. Pang.
Starch is also too sensitive to water. chemical modification or combinations thereof. with a rate of almost 10% every year. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. Starch is relatively cheap and available from a broad range of plants. blending with other materials.Summary Plastics made from fossil resources (a. There are several starch modification methods available. there is a strong need for renewable alternatives for plastics from fossil resources. tensile and flexural strength).o oil and gas) are very attractive materials for a broad range of applications. Starch‐based biodegradable materials are considered interesting candidates to replace certain types of conventional plastics. For instance. 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.6)‐glucosidic bonds. After use. while amylopectin is branched polymer. we have . There are two types of polymers present in starch: amylose and amylopectin. To gain insight in the potential of this approach. Examples are the use as packaging and construction materials. The application of a number of prospective methods for developing starch‐based biomaterials will be described in this thesis. 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. it has a high volume to weight ratio and is generally resistant to degradation. Starch is a polymer consisting of anhydroglucose (AHG) units. Starch must therefore be modified before it can be applied as a biodegradable plastic. 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.4)‐glucosidic bonds. plastics may end up in the environment and cause serious environmental problems. Its production has increased significantly since 1950. 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. containing periodic branches linked with the backbones through α‐D‐(1. 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. such as thermoplasticization. The use of virgin starch for packaging materials is restricted because it cannot be shaped to films with adequate mechanical properties (high percentage elongation.
and subsequent removal of the silyl groups. The results for the model system involving the ROP of p‐dioxanone initiated by hydroxyl groups of a protected monosaccharide (1. 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. The yield of the polymers was a function of the reaction temperature and the reaction time. temperature and monomer: monosaccharide ratio) on the average degree of polymerization. Poly‐(ε)‐ caprolactone grafted silylated starch co‐polymers with average chain length of 40‐ 55 monomer units (molecular weight of 4500‐6300) were obtained. The synthetic procedure may be divided into three steps: hydrophobisation of starch by the introduction of SiMe3 groups followed by in situ ROP. Most of the previous starch esterification studies involved the use of short chain carboxylic acids (C1‐C4). 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. Considerable amounts of ε‐CL homopolymers with isopropyl end‐groups were also formed. The results are described in detail in Chapter 3.46‐0.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose ratio (at constant Al(OiPr)3 intake). depending on the ε‐CL to starch ratio. The polymerizations were performed at 60‐100°C.3. Silylated starch with a low to medium DS (0. A statistical model has been developed to quantify the effects of process variables (time.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose) using Al(OiPr)3 as the catalyst. with higher temperatures (100°C) leading to lower yields.2. even in the presence of plasticizers.2. and off‐white solid products with isolated yields of 30‐96% were obtained. The DS of the PCL chains was between 0. The silylation reaction was performed using hexamethyl disilazane (HMDS) in DMSO/ toluene mixtures at 70°C. with higher ratios leading to higher molecular weights. The grafting efficiency for the desired reaction was 28‐58%. The use of higher carboxylic (fatty) acid to esterify starch resulted in products with significantly improved 134 . are described in Chapter 2. Esterification of starch with carboxylic acid derivatives is one of the oldest strategies to improve starch properties. and particularly acetic acid derivatives (C2) have received considerable attention.72.68) was obtained. The hydrophobicity of starch acetates is higher than virgin starch. but the products are still very brittle. 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.Summary initially performed research on the ROP with a simpler model system involving a simple protected mono‐saccharide instead of starch.3. 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.21‐0.
an alternative method for higher fatty starch esters by using fatty vinyl esters is reported. Reaction temperature and catalyst intake also affect the product DS but only to a lesser extent. the DS decreased. 1. The DS of the products was a clear function of the chain length of the fatty ester. The thermal stability of the esterified products is higher than that of native starch.45‐1. The last part of this thesis deals with reactive blending of starch and biodegradable polymers using reactive interfacial agents. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. The DS of the products is a strong function of the basicity of the catalyst.9 for starch laurate. and the type of catalyst. while the use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2. 13C‐NMR and FT‐IR. 0.3‐1. and modulus of elasticity of 46‐113 MPa. At higher ratios. The products have a broad range of degree of substitution (DS = 0. PCL‐g‐ 135 .5 MPa. however.4 could be obtained for both laurate and stearate esters.5 for starch laurate. The yellowish products were characterized by 1H‐. an increase in the vinyl‐ester concentration led to higher product DS values. In Chapter 4‐5. and Na‐acetate) in DMSO at 110°C. and Na‐ acetate). The starch esters were prepared by reacting starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. Summary mechanical properties and hydrophobicities. 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. K2CO3.26‐2. and catalyst basicity) to the DS of the starch ester products were developed. High‐DS products (DS= 2.7‐3. presumably due to a reduction of the polarity of the reaction medium. In Chapter 5. a systematic study.39) are totally amorphous whereas the low‐DS ones (DS= 1. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4.4‐3 for starch stearate).5 for starch stearate). For low vinyl‐ester to starch ratios. Mechanical tests show that the products have tensile strength (stress at break) between 2. usually performed using fatty acid chlorides which are relatively expensive and rather corrosive.24 ‐ 2. Statistically adequate (R2≥0. 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 use of methyl and glyceryl esters results in products with only relatively low DS. the synthesis of two polycaprolactone (PCL)‐based reactive interfacial agents. The thermal and mechanical properties of some representative product samples are also described.1‐2. products with a DS > 2. catalyst intake. vinyl ester to starch ratio. With these catalysts.07‐1.96). In Chapter 6. K2CO3. elongation at break of 3‐26%.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐ starch mol ratio. reaction temperature.75) are still partially crystalline. The synthesis of these fatty starch esters is. The use of Na2HPO4 resulted in low‐medium DS products (0.
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. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. At a fixed starch content (20 %‐wt). PCL‐g‐DEM seems to have slightly more efficient compatibilising effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values. Both of the PCL‐based compatibilisers improve the interfacial adhesion between the starch particles and the PCL matrix in starch/ PCL blends. A statistically adequate model (R2=0. As the result. 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. and their use in compatibilising starch‐PCL blends is described. P‐value ≤10‐9) has been developed to quantify the effects of process variables (monomer and initiator intake. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). the mechanical behavior of the compatibilised blends is in general different from that of pure PCL and of the corresponding uncompatibilised blends.957. mutual solubility of the monomer into the molten polymer) on the functionalisation degree (FD) of the GMA and DEM units to the PCL backbone. 136 . Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. at relatively high GMA and BPO intakes.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).
mede omdat het niet mogelijk is om films met goede mechanische eigenschappen te maken. Dit.4)‐glucoside bindingen verbonden zijn. Twee type polymeren zijn dominant aanwezig in zetmeel: amylose en amylopectine. Amylose is een lineair polymeer waarin de AHG eenheden hoofdzakelijk via α‐D‐(1. De toepassing van een alternatieve methode waarbij zetmeel eerst minder hydrofiel gemaakt wordt door de introductie van grote hydrofobe SiMe3 groepen. 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. Zetmeel is een polymeer bestaande uit anhydroglucose (AHG) eenheden. chemische modificaties en combinaties van deze methoden. Zetmeel is relatief goedkoop en te winnen uit een breed scala aan planten. Voorbeelden zijn thermo‐ plastificatie.Samenvatting Kunststoffen gemaakt van fossiele grondstoffen als olie en gas zijn zeer aantrekkelijke materialen voor een breed scala aan toepassingen. menging met andere materialen.6)‐glucoside bindingen met de hoofdketen zijn verbonden. Na gebruik kunnen de materialen in het milieu eindigen en grote problemen veroorzaken. Het gebruik van zetmeel voor verpakking materialen is momenteel beperkt. De productie van kunststoffen is sinds 1950 met bijna 10% per jaar gestegen. heeft geleid tot een grote interesse voor biologisch afbreekbare alternatieven gemaakt uit hernieuwbare grondstoffen.v. Het zetmeel is daarnaast ook erg gevoelig voor water. in combinatie met de huidige hoge prijzen voor de grondstoffen. De synthese van dergelijke producten met een in situ katalytische ringopening polymerisatie (ROP) van geschikte monomeren als b. De AHG eenheden kunnen op een aantal manieren aan elkaar gekoppeld worden. Er zijn verschillende methodes beschikbaar om zetmeel te modificeren. Modificatie van zetmeel kan het aantal toepassingen sterk vergroten. Biologisch afbreekbare polymeren op basis van zetmeel kunnen mogelijk bepaalde conventionele kunststoffen vervangen. Dit proefschrift beschrijft een aantal methodes om nieuwe polymere materialen op basis van zetmeel te maken. gevolgd door de ROP met een polyester precursor en verwijdering van de SiMe3 groepen . De verhouding amylose en amylopectine in zetmeel varieert en hangt grotendeels af van de plant waaruit het zetmeel gewonnen wordt. Het amylopectine is een vertakt polymeer waarbij de vertakkingen via α‐D‐(1. 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. ε‐caprolactone op de hydroxy‐groepen van zetmeel is helaas niet eenvoudig.
De silylerings reactie werd bij 70°C uitgevoerd met hexamethyl disilazane (HMDS) in DMSO/tolueen mengsels. Er werden ook aanzienlijke hoeveelheden homopolymeer van ε‐ caprolacton met isopropyl eindgroepen gevormd. De opbrengst van de polymeren was een functie van de reactietemperatuur en de reactietijd. 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%).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.68).Samenvatting zou in theorie moeten leiden tot een hogere grafting efficiency. De grafting van ε‐ caprolactone aan het gesilyleerde zetmeel werd in THF bij 50°C uitgevoerd met Al(OiPr)3 als katalysator. De substitutiegraad van de SiMe3 groepen varieerde van laag tot middelhoog (degree of substitution.72 en was afhankelijk van de ε‐caprolacton/zetmeel verhouding. 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. 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.46‐0. De polymerisaties werden tussen 60‐100°C uitgevoerd en leverde vaste producten in geïsoleerde opbrengsten tussen de 30 en 96%. De resultaten voor het modelsysteem waar de ROP van p‐dioxanone wordt geïnitieerd door de hydroxy‐groepen van een beschermde mono‐saccharide (1.2. DS = 0. temperatuur en monomeer: monosaccharide verhouding) op de gemiddelde polymerisatie graad zijn gekwantificeerd met een statistisch model.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (bij constante i Al(O Pr)3 concentratie). De resultaten worden in Hoofdstuk 3 in detail beschreven. De polycaprolacton eenheden (PCL) van het verkregen poly‐ε‐caprolactone gesilyleerde zetmeel hadden een gemiddelde ketenlengte van 40‐55 monomeereenheden (molecuul massa 4500‐6300).21‐0. waarin hogere temperaturen (100°C) tot lagere opbrengsten leiden. gevolgd door in situ ROP met ε‐ caprolacton en daarna de verwijdering van de silyl groepen.3. De kennis verkregen met het modelsysteem is toegepast om ε‐caprolactone te graften op zetmeel. Een hogere verhouding leidde tot hogere molecuul massa’s. De DS van de PCL ketens lag tussen de 0.2.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) met Al(OiPr)3 als katalysator worden in Hoofdstuk 2 beschreven. De gevolgde procedure bestaat uit drie stappen: hydrofobisering van het zetmeel door de introductie van SiMe3 groepen. De effecten van procesvariabelen (tijd. 138 . De grafting efficiency van de gewenste reactie was 28‐58%.
In Hoofdstuk 5. Met deze katalysatoren worden voor zowel de laureaat als stearaat esters producten met een DS van boven de 2. De thermische en mechanische eigenschappen van sommige representatieve producten zijn uitgebreid bepaald. en in het bijzonder azijnzuur derivaten (C2). Het gebruik van Na2HPO4 resulteerde in producten met een laag tot middelhoge DS (0. De substitutiegraad varieert van 0. Echter de producten zijn nog steeds zeer bros.5 voor zetmeel laureaat. zelfs in aanwezigheid van plastificeermiddelen. Samenvatting De esterificatie van zetmeel met carbonzuur derivaten is één van de oudste strategieën voor het verbeteren van zetmeeleigenschappen. K2CO3 en Na‐acetaat vertonen superieure activiteiten vergeleken met Na2HPO4. 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. De zetmeel esters werden bereid door het zetmeel met vinyl laureaat of vinyl stearaat in aanwezigheid van basische katalysatoren (Na2HPO4. wordt een systematische studie. De effecten van belangrijke procesvariabelen (vinyl ester/zetmeel molverhouding. Bij lage vinyl‐ester/zetmeel verhoudingen leidde een verhoging van de vinyl‐ester concentratie in eerste instantie tot hogere DS waarden.3‐1. De producten werden met 1H‐. De zetmeel esters werden bereid door zetmeel met vinyl laureaat of vinyl stearate in DMSO in de aanwezigheid van basische katalysatoren (Na2HPO4.96 en P‐ waarde van ≤ 10‐7). de vinyl‐ ester/zetmeel verhouding en het type katalysator. wordt een alternatieve methode voor de synthese van zetmeel esters met lange vetzuren beschreven.24 tot 2. De acetaat esters zijn in het algemeen minder hydrofiel dan zetmeel. katalysator concentratie en katalysator basiciteit) op de DS zijn gekwantificeerd met een statistisch model (R2 ≥ 0. De DS van de producten was een duidelijke functie van de keten lengte van de vetzure vinylester. K2CO3. Echter. terwijl het gebruik van de K2CO3 en Na‐acetaat katalysatoren 139 . Het gebruik van hogere (vet‐) zuren resulteerde in producten met beduidend betere mechanische eigenschappen en hydrofobiciteit. inclusief statistische modellering. In de meeste zetmeel esterificatie studies wordt gebruikt gemaakt van kleine carbonzuren (C1‐C4). De temperatuur en de katalysator concentratie beïnvloeden de DS ook maar in mindere mate. bij verdere verhoging van de verhouding verminderde de DS. De DS van de producten is een sterke functie van de basiciteit van de katalysator. 0. 13C‐NMR en FT‐IR gekarakteriseerd.5 voor zetmeel stearaat).4 verkregen. naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat. Het gebruik van methyl en glyceryl esters resulteert in producten met een lage substitutiegraad. In Hoofdstuk 4‐5. De synthese van deze esters wordt in het algemeen uitgevoerd met vetzuurchloriden die vrij duur en sterk corrosief zijn. en Na‐acetaat) te laten reageren.96. vermoedelijk als gevolg van een verlaging van de polariteit van het reactiemengsel.07‐1. K2CO3 en Na‐acetaat) in DMSO bij 110°C te laten reageren. reactietemperatuur.
en een elasticiteit modulus van 46‐113 MPa. De thermische stabiliteit van de geësterificeerde producten is hoger dan die van zetmeel. 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.7‐3. Beide reactieve compatibilisers verbeteren de interactie tussen de zetmeeldeeltjes en de PCL matrix. Als reactieve compatibilisers zijn twee polycaprolactone derivaten. gebruikt. PCL‐g‐glycidyl methacrylaat (PCL‐GMA) en PCL‐g‐diethyl maleaat (PCL‐g‐DEM).39) zijn volledig amorf terwijl de producten met een DS van 1.957 en P‐ waarde van ≤ 10‐9) ontwikkeld waarmee de effecten van procesvariabelen (monomeer en initiator concentratie. Dit leidt tot verschillen in mechanische eigenschappen tussen gecompatibiliseerde mengsels en van de niet‐ gecompatibiliseerde mengsels van zetmeel en PCL.5 MPa. De synthese van deze verbindingen wordt uitgebreid beschreven. De FD waarden voor PCL‐g‐DEM waren aanzienlijk lager (tot 7%).Samenvatting resulteerde in producten met een middelhoog tot hoge DS waarden (2. De hoogste FD waarden (tot 45%) werden gevonden voor PCL‐g‐GMA bij relatief hoge GMA en BPO concentraties.4‐3.45‐1. 1.26‐2. Zo is bijvoorbeeld de elastische modulus van de gecompatibiliseerde mengsels beduidend hoger (tot 430 MPa) dan die van niet‐gecompatibiliseerde mengsels (320‐340 MPa). Het laatste hoofdstuk van dit proefschrift (Hoofdstuk 6) behandelt de menging van zetmeel en biologisch afbreekbare polymeren met reactieve compatibilisers.1‐2.75 nog gedeeltelijk kristallijn zijn. 140 .9 voor zetmeel laureaat. Mechanische tests tonen aan dat de producten een treksterkte hebben variërend van 2.0 voor zetmeel stearaat). 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. Er is een statistisch model (R2 =0. een uitrekking tot breuk van 3‐26%. De producten met een hoge DS (2.
Prof. Thank you Erik. People from other groups have also been very kind and supportive to my research. and Prof. Erwin Wilbers. L. Marcel de Vries. and for always encouraging me to finish this thesis. First of all. Erik Heeres. A. especially on his opinions about the statistical modelling and mechanical properties. I want to thank Marya van der Duin‐de Jonge. and Prof. since the mechanical properties as well as hydrophobicity of this starting material is relatively poor. for the inspiring ideas during the discussions. and for helping me arranging many things when I was not able to do it from Indonesia. as well as taking the SEM pictures. I also want to give my deep appreciation to my reading committee. Francesco Picchioni. the raw material for fossil derived plastics. This thesis is a summary of my research on developing novel biodegradable polymers from starch. who was my first daily supervisor for about two months. for upgrading my writing quality.J. A. I want to thank Erik Heeres for being more than a good supervisor. 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. together with difficulties in managing plastic waste. Mosiscki. for your consistent support. not in the last place because of the problems related with conventional plastics. I also want to thank Ineke Ganzeveld. I want to thank my three promoters. for her hospitality and solving administrative matters. Gert Alberda van Ekenstein and Harry Nijland helped me a lot with mechanical and thermal property measurements. which improved the quality of this thesis. Prof.Acknowledgements The synthesis of biodegradable materials is currently receiving a lot of attention. Prof. These include the current high oil prices. Although it is indeed a very interesting topic. I want to thank Francesco Picchioni for being always there when I needed him to discuss anything. and Laurens Bosgra for being always helpful with technical issues and Jan Henk Marsman for the help with analytical equipment. Leon Janssen. for valuable discussions during my early PhD period and for handing me over to the right hands. Anno was also always there when .A. Broekhuis. Prof. Anno Wagenaar and Wim Kruizinga taught me how to operate the NMR machines and to interpret the spectra. I found out that developing a better fully‐biodegradable‐material from starch is not an easy job. Anne Appeldoorn. for his efforts to make every foreign students feel at home in Groningen. He has also been a good friend and always supported me during the hard times. I want to thank Leon Janssen for his hospitality. Minnaard. Thank you very much for spending your valuable time reading my thesis and for the constructive comments and corrections you have made.
Tiara. Francesca Gambardella. Wahyudin. mbak Ika. Iwan Harsono. Vincent Nieborg and Fesia Lestari. Our collaboration has also become part of this thesis. in the office. Zhang Youchun. Thyara. Endy. I have met a lot of Indonesian students in Groningen. Ais Jenie. Groningen is also a place where I found true friendship. patience. Thanks also for Marcel and Connie. friendly. Jasper Huijsman. Hofman. Jaap Bosma. Ardiyanti. Jelle Wildschut. and for my wife and family. for your never‐ending blessings. Tante Smith. and Mahesa for being my brothers and sisters in Groningen. and Iwan Kustiawan for becoming my family. Sari Fabianti. Louis Daniel.Acknowledgements I had questions or needed specific equipments. Niels van Vegten. Dado. Francesca Fallani. and passion. Yao Jie. Wita Sondari. for your love. Bima and Wisnu (also my paranymph). Tenny. and many others for the good old times! Thank you Tante Ietje. You all have always been my inspiration and motivation. pak Harry. and although we seldom met afterwards. visiting researchers. Patrick and Oppie. Buana Girisuta and Rina Karina. Erna Subroto. Agnes R. bu Ida Susanti. Teguh. Samuel. Togi. Teddy. Abdul Osman. The chemical and glassware magazijn as well as the FWN library personnel were always patient. Thank you Tresna. Anna Nizniowska. 142 . and on many occassions. and the ISD staff helped me a lot with visum and residence permit arrangements as well as tax papers. August Kurniawan. and Henk van de Bovenkamp for cheering my days and being my friends during summer and winter. Oscar Rojas. Tante Caroline. Diana Santangelo. Marcel Wiegman. Diana Jirjis. and helpful. for everything. Rasrendra. Taufik. kak Atha. kak Roga. for never leaving me alone. Mahfud. master students who worked with me. Thank you my parents and sister. Anneke Toxopeus. Nadia Gozali. Ignacio Melian Cabrera. our memories will never fade away. Ita. Yongki. Albert Kieviet helped me with the MALDI‐TOF measurements. R. Puspita Kencana. Thank you God. Chalid. Thank you AW. My special thanks are for Laura Junistia and Jan‐Pieter Drijfhout. spring and autumn in Groningen. Judy Retti Witono. Mita. Ronny Sutarto. Boelo Schuur.B. Nidal Hammoud Hassan. mbak Mia. Wangsa. Asal Hamarneh. C. A lot of thanks for Henk Stegeman and Poppy Sutanto. Farchad H. mas Pandu. and assistant professors have shared the days with me in Groningen. Arjan Kloekhorst. Egi. Tante Oppie. I want to thank Edwin A. in the lab. and Tante Alma for being like my parents and family in Groningen. Henky Muljana (my paranymph) and Anindhita Widyadhana. M. Wuri Raspati. Gerard Kraai. Annette Korringa. Hans Heeres. Taufik. support. Many master and PhD students. Kuslan.
Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants.10. Heeres. Janssen.P.Heeres. A. 2008 (submitted to J. 2008 (accepted for publication in Starch‐ Starke). Polym. A. L. Sugih.J. 2.K.K.K. Experimental and modeling studies on the synthesis and properties of higher fatty esters of corn starch. F. Broekhuis. Eur. L. and H. F.1016/j. and H. 4. Picchioni. A. Heeres.K. presented in the 1st International Symposium on Sustainable Chemical Product and Process Engineering.P. Heeres. India. . L.P.J.). H. Guangzhou (China). Junistia. DOI: 10.A. Heeres. 2008 (submitted to Carbohydr. F.M.J. Picchioni. J. 17‐20 December 2006 (oral presentation). presented in “Polymers for Advanced Technologies”. Mumbai (India).J. 27‐30 August 2006 (oral presentation) 7.K.K.P. starch‐based innovative product: The synthesis of grafted polycaprolactone‐starch by ring opening polymerisation. and F. F. Sugih. Heeres. Synthesis and properties of reactive interfacial agents for polycaprolactone‐starch blends. Synthesis of poly‐ (ε)‐caprolactone grafted starch co‐polymers by ring‐opening polymerisation using silylated starch precursors. L.Janssen.B. 9. 9th National Conference organized by the Society for Polymer Science.M. J. A.010). Pune Chapter. Manurung. Polym. Picchioni. 26‐28 November 2006 (oral presentation). A.J.J. L. Drijfhout. Junistia.J.M Janssen and H.List of Publications 1. J.K. A. A. Heeres: Green. 5. 3. 2008 (accepted for publication in Starch‐Starke). Appl. F. Janssen. Sugih. Heeres: Synthesis and properties of novel green biopolymers derived from starch and fatty acids. Sugih. H. H. Sugih.eurpolymj. Sugih.) 6. Picchioni. F. presented in International Workshop: “From Polymer Modification to Multicomponent Polymer Systems”. R. A.B.M. Experimental Studies on the ring opening polymerization of p‐dioxanone using an Al(OiPr)3‐monosaccharide initiator system.P. South China University of Technology. H.K. 8.P. R. Sugih.B. Kuala Lumpur (Malaysia). presented in 11th Asia Pacific Confederation of Chemical Engineering (APCChE) Congress. 2008. Sugih. Drijfhout. L.J. Sci. A. Polym.2008. Bratislava (Slovak Republic). H. Picchioni. J. A.K. F. Junistia. L.Drijfhout. Picchioni: Functionalization and blending of polycaprolactone.J. A. H. Manurung. 25‐28 September 2007 (oral presentation).Picchioni. and F. Broekhuis.P. (in press. Junistia. Picchioni.A. L. Sugih. Picchioni: Biodegradable polymer materials based on polyesters and starch/proteins.B.