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
dr. A. dr. L. Broekhuis Prof. F.P.M. ir.B. dr. dr. H. Minnaard Prof. ir. Moscicki ISBN : 978‐90‐367‐3592‐6 ISBN : 978‐90‐367‐3593‐3 (electronic version) . Picchioni Beoordelingscommissie : Prof. dr. dr. A. L. Janssen Prof.J.J. Heeres Prof. Promotores : Prof.A.
and my sister . to: Tresna. my parents.
184.108.40.206.4.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (2) 27 28 26 27 27 27 2.3. Starch Esters 1. Thesis Outline 1. Starch ‐ biopolymer blends and graft co‐polymers 1. Cross‐linked starch 1.2.4.Table of Contents Chapter 1: Introduction 1.1. Product analyses . Introduction 220.127.116.11. Starch‐based blends by in situ polymerization 1.1. Biomaterials from starch 1. Typical example for the synthesis of polydioxanone end‐capped with 1.2. Methods 2.1.3. Starch modifications to improve product properties 18.104.22.168. The potential of biodegradable plastics 22.214.171.124.126.96.36.199. Thermoplasticized starch 1.3. Materials 2. Starch 1.1. Biodegradable plastics from starch 1.2. Starch production processes 1.3. Starch‐based blends by melt mixing 188.8.131.52. Materials and Methods 2. 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 184.108.40.206. Plastic applications and waste issues 1.5.2. Structure and properties of starch 220.127.116.11.1.
6.3. Effects of process conditions on the average chain length ( X n exp ) and end group distribution 28 29 29 30 36 38 39 40 42 43 44 2.1.2. References Chapter 3: Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors 18.104.22.168. Systematic studies 2.2.2. Nomenclature 22.214.171.124. Typical example of the starch silylation procedure 3.2. Conclusions 2. Typical example of in situ polymerization of ε‐CL with silylated starch 50 126.96.36.199.3. Nuclear Magnetic Resonance (NMR) 188.8.131.52.2.1. Peracetylation of silylated starch 3.2. Screening experiments 2. Synthesis of silylated starch viii . Methods 184.108.40.206.2. Product analyses 2.2. Analytical methods 3.2. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers 51 51 52 52 52 53 54 48 49 49 50 50 220.127.116.11.3. Materials and Methods 3.2.4. 2.3. Introduction 18.104.22.168.22.214.171.124. Mechanistic aspects 2.4. Calculation of average degree of polymerization 2. Results and Discussions 2.4. Materials 3.2.2. Product yield 2. Results and Discussions 3.1. Calculations 3.3.1.
Results and Discussion 4. Systematic studies 126.96.36.199.3.3.3. Determination of the Degree of Substitution (DS) 4.3. References ix .188.8.131.52.3. Effect of the addition of toluene as a co‐solvent 184.108.40.206.220.127.116.11.6. Catalysts screening 4. Effect of vinyl ester to AHG ratio on the product DS 4. Methods 4.1.2. Product characterisation 4.2.4. References Chapter 4: Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants 4. Materials and Methods 4. Materials 4.3. Typical example of the synthesis of laurate and stearate esters of corn starch 72 73 74 75 75 76 76 78 79 80 81 81 82 83 84 70 71 71 71 72 4.3. Analytical equipment 4. Conclusions 3.3.2. Peracetylation procedure 4.3. 3. Nomenclature 4.2.1. In situ ring opening polymerization of ε‐caprolactone with silylated starch 57 63 64 64 65 3. Conclusions 4.3.3. Deprotection of silylated‐starch‐g‐PCL 18.104.22.168. Exploratory experiments 22.214.171.124.126.96.36.199. Introduction 4.3. 1H‐ and 13C‐NMR analyses 4.2. Nomenclature 3. FT‐IR measurements 188.8.131.52.2.
Nomenclature 5. Methods 6. Mathematical modelling 5.3.1. Methods 5. Introduction 6.2. Preparation of PCL‐starch blends with the reactive compatibilisers 110 6. Analytical Methods 110 x .184.108.40.206.2. Conclusions 5.3.3. Materials 6.1. Analytical Equipment 5.4.4. References Chapter 6: Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 6. Product properties 5.2.2. Peracetylation procedure and Degree of Substitution (DS) determination 88 88 89 92 96 102 102 103 220.127.116.11.1. Materials 5.2. Materials and Methods 6.2.3.Chapter 5: Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch 18.104.22.168.2. Experimental Design 5.2. Work‐up of PCL‐g‐GMA products 22.214.171.124.6. Work‐up of PCL‐g‐DEM 108 109 109 109 109 110 110 126.96.36.199. Materials and Methods 5.3.1. Results and Discussion 5. Introduction 5. Typical example of the preparation of laurate and stearate starch Esters 88 86 87 87 87 88 5.2. Compatibilizer synthesis 188.8.131.52.184.108.40.206.
3.1. References Summary Samenvatting (Dutch Summary) Acknowledgements List of Publications 133 137 141 143 xi . Preparation of the ractive compatibilizers 6.2.6. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers 6. Statistical mdeling 6.3. Nomenclature 6. Effect of the BPO intake on the product FD 220.127.116.11.2.4.2. Ternary blends compatibilized with PCL‐g‐GMA 18.104.22.168. 6.3. 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 6.1. Results and Discussions 22.214.171.124. Effect of substrate (GMA/DEM) to PCL Ratio on the FD 6. Conclusions 126.96.36.199.3.3.2. Binary blends of starch and PCL 6.3.3. Synthesis and Properties of Starch‐ PCL Blends 188.8.131.52.3.1.1. Thermal properties of the compatibilizers 6. Ternary blends compatibilized with PCL‐g‐DEM 6.3.
Chapter 1 Introduction Abstract A general overview of starch properties will be given and the potential use of starch as a starting material for a wide range of green biomaterials will be reviewed and discussed. Finally. Different routes to modify starch to improve the product properties and to extend the application range will be provided. . an outline of this thesis is given.
wheat. .1. potato. rice. but other types of starch such as cassava. Starch is found in seeds (i. maize.9 ton starch per hectare . corn. An important polysaccharide is starch.Chapter 1 1. Starch 75% of all organic material on earth is present in the form of polysaccharides. with an average yield of 4. The worldwide production of starch in 2008 is estimated to be around 66 million tons . Most of the starch produced worldwide is derived from corn.e. potato or cassava) of the plants [2‐3]. 80 70 60 50 40 30 20 10 0 Starch Production (Million Tons) 1995 2000 Year 2005 2010 Figure 1. .e. Potato accumulates starch to approximately 75 % of the dry weight in the tubers with a yield up to 21 ton starch per hectare. It is generally deposited in the form of small granules or cells with diameters between 1‐100 µm . sweet potato. present and a forecast of starch production  o : Europe □ : USA : Total ∆ : the Rest of the World 2 .1. Plants synthesise and store starch in their structure as an energy reserve. current and future estimations of worldwide starch production are given in Figure 1.45 (estimated from the total volume and value of the US corn starch export ). or peas) and in tubers or roots (i. The second and third starch producer regions are Europe and Asia [4‐5]. while corn seeds consist of 65‐80% starch by weight. 4‐5].1. Past. Most of the starch is produced in the USA. The current price of corn starch is around $0. sorghum. and wheat starch are also produced in large amounts [2. barley. Most starch crops are very productive. Past.
the corn kernel is softened by steeping in warm water containing SO2 until the volume of the kernel increases with 55‐65%. As an example. The low density gluten is separated from the starch suspension by centrifugation.2. a. b. the mixture is fed to a hydrocyclone to separate the germ and the rest of the kernel.2. Corn starch production  a. Typical corn‐milling operation b.a. a typical corn starch production process is given in Figure 1. and starch. Cross‐sectional view of a corn kernel 3 . The fiber is removed using washing screens. The resulting starch is further washed in a cyclone and finally dried. The starch is present in the endosperms (floury and horny). After a coarse milling. After initial cleaning to remove cob. the debris is filtered from the slurry. and is embedded in a proteinaceous cellular matrix as is shown in Figure 1.b.2. and starch granules are obtained after centrifugation from the suspension.1. and other foreign materials. which is again fed to a second milling process. Introduction 1. sand. Figure 1. The resulting suspension from the mills contains fiber. gluten. Starch production processes Starch is generally extracted from the plant by wet milling processes [9‐10].1. The plant material is grounded in water.
The starch granules will swell rapidly to many times of its original volume... Amylopectin Starch is insoluble in cold water. Typically. H OH H H2COH H2COH H O H H OH H H2COH H O H H OH H O H OH H O .2. . Two types of AHG polymers are usually present in starch: amylose and amylopectin [2‐3.6)‐ glucosidic bonds .4)‐glucosidic bonds. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. 6].Chapter 1 1. The content of amylose and amylopectine in starch varies and largely depends on the starch source.. Each branch contains about 20‐30 anhydroglucose units. . Structure and properties of starch Starch is a polymer consisting of anhydroglucose (AHG) units (see Figure 1.a. but usually in the range of 1. Figure 1. H OH H O H OH H O H H OH H c..1.1.3.. the amylose content is between 18‐28% .. H2COH H H OH H O H OH O H OH O CH2 H O H H OH H O H OH H O . The amylose content of several common starches is given in Table 1. The molecular weight of amylopectin is higher than that of amylose and is typically 4‐5 x 108 Da . Amylose c.6‐7 x 105 Da ... a.. but it is very hygroscopic and binds water reversibly. Chemical Structure of Starch a.. H . The molecular weight of amylose is a function of the plant source and processing method.. The linear 4 . H OH H2COH H2COH H . H2COH H . H2COH H OH H O H H O b.3. Anhydroglucose (AHG) unit b.. Amylopectin is a branched polymer. Heating a starch solution leads to loss of hydrogen bonding in the interior of the starch granule and the starch will start to gelatinize.) .. H O H OH H OH H O H OH O H OH H O . containing periodic branches linked with the backbones through α‐D‐(1..
2. swollen granules.8 16. Table 1. depending on the amount of water present.2. and granule fragments. will form a thick paste or gel. The resulting suspension contains a mixture of linear amylose molecules. Biomaterials from starch 1.1. Starch gelatinization temperature range  Starch Potato Tapioca Corn Waxy corn Wheat Gelatinization Temperature Range [o C] 59‐68 184.108.40.206 25. Plastic Applications and Waste Issues Plastic is the general term for a wide range of synthetic or semisynthetic polymerisation products.8 17. 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 (%) 20.5‐70 62‐72 63‐72 58‐64 1. Cellulose nitrate was first prepared by A. The first generation of commercial plastics was derived from cellulose nitrate and is known as celluloid [12‐14].5 28 52 70‐75 27 15. The gelatinization temperature range of various starch sources is given in Table 1. and. and celluloid was 5 . Introduction amylose molecules leach out of the granules into the solution.7 20 18. Parker in 1838. Plastics are used in a wide range of applications and the demand is still increasing every year .2.7 26 Table 1. 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 .
A dramatic increase in demand for plastics began after World War II. Worldwide Plastics Production (1950‐2006)  6 . Polyurethanes. and polycarbonates (PC). polypropylene (PP). Coatings. Sealants. and therefore need less energy to shape it into useful materials .Chapter 1 patented by J. Elastomers. polystyrene (PS). Adhesives. Plastics have lower melting temperatures compared to glass and metals. In 2006. polyethylene terephtalate (PET). discovered by Baekeland in 1907 [12. space. Thermosets. Important polymers used for packaging are polyethylene (HDPE and LDPE). About 29% of the total plastics produced in the USA . the oldest purely synthetic plastic is Bakelite. Hyatt in 1870 .). for instance to be used in packaging and paper coatings . They have a low density and can be shaped in thin films while maintaining good properties.4. and PP-Fibers 2006: 245 2002: 200 200 Plastics Production (Million Tons) Not Included: PET-. The latter is important when using the material for packaging purposes to save weight. Plastics are also used for building materials and automotive. The largest application of plastics is for packaging purposes. polyvinyl chloride (PVC). Plastics are very attractive materials. the worldwide plastics production has reached 245 million ton per annum  (see Figure 1. and 37% of the total plastics demand in EU  is used as packaging materials. and consumer products (see Figure 1.5.Fibers 150 1989: 100 100 1976: 50 50 1950: 1. PE is a very versatile plastic because it can be shaped easily into various forms. PA-. and energy during transportation of goods.4. and PolyAcryl.) 250 Included: Thermoplastics.5 0 1950 1960 1970 1980 Year 1990 2000 2010 Figure 1. electrical. 15]. While celluloid is derived from a natural polymer (cellulose). when polyethylene (PE) was invented. The production and consumption of plastics has increased significantly with a rate of almost 10% every year since 1950.
hydropower plants operation. Plastic litter is hazardous to wildlife [1. Plastics have been polluting sea [1. soil.7% in 2006 . irrigation. however. In the USA. are not easy to recycle .5. consisting of 25%‐v of the total waste . the amount of plastics in MSW increased from less than 1% in the 1960 to 11. 6]. which are usually found in waste streams. The disposal of plastics materials in municipal solid waste (MSW) is a serious issue in many parts in the world. Most of the waste is still disposed (50%) by landfill or incinerated to recover the energy (30%) . and may be more than twice the weight fraction . rivers. It is not applicable for thermoset resins . In Europe (2006). threatening fishery. 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). Introduction Plastic waste. plastics waste was ranked as the second major source of MSW after paper and paperboard. Co‐mingled plastics. only 20% of the plastic is recycled.2. Figure 1. and lakes. Plastic applications and waste treatment in Europe  1. Biodegradable plastics are polymeric materials capable of decomposing when given an 7 .2. is causing serious environmental problems. ship navigation. In 1996. and is only effective for single plastic sources or simple plastic formulations. The volume fraction of plastic in MSW is much larger due to the low density of plastics. and other public works . and accumulation of its residues in soil cause significant reductions in agricultural yields. It has a high volume to weight ratio and is resistant to degradation. Recycling (part of) the plastics to reduce the amount of MSW also has limitations. 16].
Chapter 1 appropriate environment and sufficient amount of time . For instance. Common natural biopolymers are carbohydrates. and microorganisms) and biodegradable synthetic polymers . Nowadays new types of biodegradable plastics with improved properties and lower costs have been developed . The price of these biodegradable plastics is expected to be reduced considerably in the next decade . Biopolymers are inherently biodegradable. The change in chemical structure results from the action of naturally occurring micro‐ organisms. Several authorities have provided definitions for biodegradable plastics . and polyesters from micro‐organisms . Biodegradable plastics may be classified into two general groups. and since 2000 they have become competitive with traditional materials in some applications [20‐21]. Biodegradable plastics have gained considerable interest since the 1980s. biological and/or physical forces in the environment at a rate which leads to fragmentation or disintegration of the plastics. because they take part in nature’s cycle of renewal. biopolymers from nature (from plants. 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. 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. proteins (present abundantly in plants and animals). Biodegradable plastic waste may be treated in composting facilities.5/ 8 . DIN 103.96 proposal ‐ Degradable plastics are plastic materials that undergo bond scission in the backbone of a polymer through chemical. the price of PLA based plastics in 2010 is estimated to be around € 1.2 working group on biodegradable polymers ‐ Biodegradation of a plastic material is a process leading to naturally occurring metabolic end products. ASTM sub‐committee D20. 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 . together with food and yard waste as well as paper. animals. The market price of biodegradable plastics (mainly from starch and PLA) has decreased the last years. It may also be treated in sewage sludge water treatment plants or buried in the soil [20‐21].
50‐4. 1. hygienic materials (diapers. This price is much lower than conventional plastics derived from oil. These are summarized briefly in the next sections. such as polyethylene (PE. It cannot be shaped in films with adequate mechanical properties (high percentage elongation. 9 . Introduction kg. Biodegradable plastics are also used for food servicewares . € 1. biofillers for tires. starch must be modified. Bastioli [20‐21] showed that nearly all biodegradable plastics available in the market are derived from starch. towels.45 per kg . either by plasticization. blending with other materials. 20]. Starch may become an attractive raw material for plastics in the future. breathable fabrics. or when environmental impacts have to be minimized. 1. agricultural products (mulch films. 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 . 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 . before they can be applied as biodegradable plastics. chemical modification or combinations of them . Consequently. The price of starch in 2007 was about $ 0.4 or $ 1.4 per kg in 2007‐2008 ). and chewable items for pets [1. tensile and flexural strength) and is too sensitive to water [26‐27]. Virgin starch is not suitable as a packaging material. consumer products and food products. pots). napkins) .8‐ 2. either from starch‐based materials (slightly modified starch.50/ kg in 2005) is also expected to be lower in the future due to considerable reduction in modification costs . Biodegradable Plastics from Starch Starch is a very attractive source for the development of biodegradable plastics.2‐ 1.2.3. . when recycling is not practical or uneconomical [20‐21].3. alone or complexed with natural or synthetic biodegradable polymers) or from polylactic acid which originates from the fermentation of a starch feedstock. Biodegradable plastics are especially very useful for single‐use applications. utensils . The price of starch‐based biodegradable plastics (€ 1. Starch modifications to improve product properties Several techniques may be applied to develop starch based biomaterials with improved properties. Examples are the use as packaging material for carrier bags.
. plasticized. Starch can be modified to obtain materials which melt below the decomposition temperature . the semicrystalline structure of starch and its granular form are lost and the starch polymers are partially depolymerized. extrusion. approximately 230 oC) . and therefore are processable by conventional polymer processing techniques such as injection. sugars) [30. sorbitol. resulted in the formation of an amorphous mass [35. such as water and polyols (glycerol. 33]. or melted starch [29. phase separation may still occur. 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. which results in embrittlement. The use of other plasticizers (for example glycerol) results in a rubbery material. which will result in a continuous phase in the form of a viscous melt [29‐30. The modification involves break down of the starch granular structure by the use of plasticizers at high temperatures (90‐180 oC) and shear. 40]. and blow moulding [28. which is even above the thermal degradation temperature. For instance. 35. This problem is mainly caused by the presence of strong inter‐ and intra‐molecular hydrogen bonds between the starch macromolecules [28. 30]. glycol.1. destructed. 39]. 37]. glycerol. TPS synthesized from polyol and sugar plasticizers have the tendency to re‐crystallize (retrogradation) after being stored for a period of time. Although thermoplasticization seems to be a promising method. The use of water as a plasticizer is not preferable. because the resulting product will be brittle when equilibrated with ambient humidity . Yu. Thermoplasticized starch Virgin starch is brittle and difficult to be processed into articles due to its relatively high glass transition temperature (Tg. and hexylene glycol. At intermediate water levels. 40]. which still limits their use. The brittleness is known to increase in time due to free volume relaxation and retrogradation. The plasticizers are also usually hydrophilic and can be 10 . the thermal properties of glycerol‐plasticized starch are a function of water content [31. with better properties than virgin starch in various applications. 30. water content. The thermoplasticization process will decrease the interactions of the molecular chain and destruct the structure of the starch .3. and the properties and amount of plasticizers. There are several substances used as plasticizer for the preparation of thermoplastic starch (TPS). Another issue is the poor water resistance and low strength. 32. et al  discovered that the elongation of break of the thermoplastic starch is significantly improved by plasticization with glycol.Chapter 1 1. The plasticized starch properties may be tuned by changing the temperature of processing. As the result. 32‐ 36]. The modified products are known as thermoplastic.
The esterification reaction is catalyzed by H+ and proceeds with the formation of water . differing in Degree of Substitution (DS). organic chloro‐compounds. resulting in an increase of the Young’s modulus and tensile strength of the products and a decrease in elongation at break. et al  studied the mechanical properties of films from maize starch cross‐linked by sodium trimetaphosphate (SMTP). Higher processing temperatures generally led to higher cross‐linking levels. Schuetzenberger acetylated starch with acetic anhydride . These hydroxyls can react easily with a wide range of compounds such as acid anhydrides.2) are commercially available and used as food additives and in the textile industry. Silva. while highly substituted starch acetate (DS of 2‐3) are soluble in organic solvents. The cross‐linked products are therefore more rigid (and less elastic) materials than virgin starch . 39]. As a result. the use of formyl esters of starch is much lower than that of acetate esters. The most popular starch ester is starch acetate .3‐1) are still soluble in water. paper. aldehydes. textile. Cross‐linking results in a reduction of the solubility in water and to thickening. Most studies dealt with the synthesis of starch esters of C1‐C4 carboxylic acids.3. The simplest starch ester is starch formate (C1). High DS starch acetate can be easily casted 11 . 46]. The DS is defined as the moles of substituents per mole of AHG units . and ethylenic compounds. 1. Introduction washed out by water . Starch Esters The development of starch esters started in the mid 19th century [44. leading to higher viscosities [42‐43]. or coating with hydrophobic polymers [33. Low DS products (0. especially in the food. synthesized by direct addition of formic acid to starch at room temperature. The products are called cross‐linked starches.3. Chemicals of these classes having two or more of the reactive groups may react with two or more hydroxyls of the starch molecules.2. Three different types of starch acetates may be distinguished. The cross‐linked starches have found many applications. 1.01‐0. Solutions to improve the properties of TPS are blending. epoxy.3. and particularly with acetic acid . Substitution of the hydroxyl groups of starch with acetate groups makes the esters more hydrophobic than native starch. Medium DS starch acetates (0. Cross‐linked starch An anhydroglucose molecule of starch contains two secondary and one primary hydroxyl group. and adhesive industry. Breakdown of the starch and the formation of low molecular weight products occur to a significant extent . Already in 1865.
xylene.1. and phosphate) will result in C‐2.Chapter 1 into films using organic solvents. DMF. et al [55‐56]. Acetylation of starch in water and DMSO using vinyl acetate has been studied lately by Mormann and Al‐Higari  as well as Dicke . while alkaline catalysts (such as carbonate. hydrogencarbonate. and is shown in Scheme 1. Mechanism of starch acetylation using vinyl acetate 12 . The kinetics of the reaction between gelatinized aqueous potato starch and vinyl acetate was studied by De Graaf. or isopropanol) with acidic (hydrochloric or sulfuric acid) or alkaline (NaOH or triethylamine) catalysts . High DS starch acetates are thermoplastic materials suitable to be used as biodegradable plastics . Neutral and weak acid/ alkaline catalysts result in regioselective substitution at the C2 hydroxyl groups of starch. 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. Starch triacetate has been successfully synthesized using acetic anhydride in combination with pyridine‐gelatinized starch . acetate. The degree of acetylation of starch acetates can be easily controlled. Several synthetic routes have been developed for starch esters. allowing the polymer to be produced with a range of hydrophobicities . DMSO. Starch esters have also been synthesized using alkanoyl chlorides [49‐53]. Esterification may be performed using acid anhydrides in aqueous media or organic solvents (pyridine. but also C‐6 and C‐ 3 substitution . Another attractive route involves the use of vinyl esters as reagents .1.
Two main methodologies are applied for the production of polymer blends. a chemical reaction might take place upon processing (reactive extrusion). i. in order to achieve also a good molecular adhesion between the phases by melt mixing. dodecanoyl. it is frequently used in order to chemically graft the polymerized component on the other one (polystyrene on polybutadiene in our example above ). The introduction of longer acid chain is expected to reduce the brittleness of virgin starch and to increase its hydrophobicity .g. The fatty acid chloride reactants are.3. Despite this advantage. 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 .61) products could be obtained using this approach. 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 . octadecanoyl) chlorides [49‐53]. The first involves simple melt mixing of the two components for example by extrusion. compatibilizers) might be used. 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 . interfacial agents (e. relatively expensive and rather corrosive. melt‐mixing remains the preferred route to polymer blends mainly because of very practical reasons: low costs. availability of mixing equipment and no necessity to use any organic solvent (often employed for the in situ polymerization). in the simplest case of a binary blend. polymers are rarely miscible with each other  so that. they locate themselves at the interface between the 13 . However. one component will be dispersed into the other.4. Although the in situ polymerization process is not as technologically straightforward and economically convenient as melt‐mixing. the latter are mixed together. As a result. 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. Starch ‐ Biopolymer Blends and Graft Co‐Polymers Blending of different polymers is an established method to obtain products with improved properties. By working at temperatures above the melting point or glass transition temperature of the two components.e. however. the two components are chemically linked to each other. Their role is comparable to the one of a surfactant in emulsion formation . which in turn provide a very strong adhesion at the molecular level between the dispersed phase and the matrix. Fatty‐acid starch esters have been synthesized using fatty acid (octanoyl. 1. If the right combination of chemical groups on the two components is present along the polymers backbone.34‐0. but only relatively low‐DS (0. However. Introduction The synthesis of long chain fatty acid ester of starch has attracted much interest lately.
The products have been synthesized in the lab as well as on industrial scale [32.3. The starch/polyethylene blends are used for agricultural mulch [27. The blending and grafting of starch with synthetic polymers is usually performed to achieve higher hydrophobicity and to improve the mechanical and thermal properties as well as to obtain cheaper and more biodegradable products. 80‐ 81] or food packaging . 39. 66‐81]. Matrix Dispersed Phase Figure 1. 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. 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.Chapter 1 two components stabilizing the dispersion (most probably by a steric repulsion mechanism ) and providing an improved adhesion at the interface.4. 64‐98]. The most often used synthetic polymer for blending with starch is polyethylene [27.6. there have been efforts to blend as well as to graft synthetic polymers onto starch. Illustration of the role of interfacial agent in compatibilizing blends 1.6. In the past. Starch‐based blends by melt mixing Starch‐synthetic polymer blending has been studied as early as in 1973 [64‐65]. The uncompatibilized blends of starch and 14 .).1. The synthesis of compatibilizer and its use for starch/ synthetic polymer blending has also been studied.
To obtain completely biodegradable products. 15 . among which synthetic polyesters are considered very promising materials . Polycaprolactone (PCL) is a well‐known synthetic biodegradable polyester. which both can react in situ with the hydroxyl groups of starch [70. Examples of these biodegradable polyesters are poly‐glycolide. PCL‐g‐pyromellitic anhydride [94‐95]. the compatibilizer. 72‐73]. while the tensile strength of the compatibilized ones decreases only slightly with the starch content. the uncompatibilized blends of starch with PCL give coarse phase separation and a reduction in the mechanical properties when the starch content is increased. poly‐lactides. 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 tensile strength of the uncompatibilized blends is drastically reduced when starch content is increased. while PE‐g‐GMA posseses epoxide groups. which combines excellent biodegradability with acceptable mechanical properties. since the conventional synthetic polymers are usually poorly or non‐biodegradable. The use of conventional synthetic polymer such as PE for blending with starch will only result in a partially biodegradable material. 39. and poly‐caprolactone). As a result a graft copolymer (PE‐g‐Starch). thermal and water/gas permeability properties that are even comparable to bulk non‐biodegradable polymers such as PE and PP. poly‐valerolactone. dextran‐g‐PCL . 84‐98]. thus improving the dispersion of starch in the PE matrix. The ester bonds of these polymers are susceptible to attack by water and this leads to enhanced biodegradability. and poly‐lactones (such as poly‐butyrolactone. These biodegradable polyesters will finally decompose into non‐toxic products . The use of reactive compatibilizer precursors PCL‐g‐glycidyl methacrylate (PCL‐g‐GMA) [89‐90]. poly‐dioxanone. and EVOH . The mechanical properties of the blend are also improved.e. synthetic biodegradable polymers have been applied. 96]. Introduction polyethylene show a coarse phase separation due to differences in polarity of starch (hydrophilic) and polyethylene (hydrophobic). 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. PE‐g‐MA contains reactive anhydride sites. i. 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. As is the case of blends with PE. Despite these good results. Some of these polyesters also have very good mechanical. The mechanical properties of these blends (tensile strength and elongation at break) decrease at higher starch content. PCL‐g‐maleic anhydride (PCL‐g‐MA) [91. is formed upon mixing. Studies on the blending of starch with PCL have been already described [32.
Previous studies showed that common ROP catalysts such as tin octoate or aluminium isopropoxide gave low (0‐14%) grafting efficiencies (GE.3. Starch is a very hydrophilic material that always contains moisture. A new strategy for the in situ ROP of ε−caprolactone on starch with the use of common ROP catalyst is therefore highly desirable. Thesis Outline The objective of this thesis is to study synthetic pathways to obtain starch derivatives with the potential to be used as bioplastic. a very flammable by‐product.4.Chapter 1 1.2. In Chapter 3. during the reaction. the synthesis of starch‐g‐PCL is reported. Chapter 2 describes the ROP of p‐dioxanone initiated by a protected monosaccharide (1. The water in starch granules competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. starch‐g‐PCL has been synthesized using toxic materials such as isocyanates . leading to formation of PCL homopolymers rather that starch‐g‐PCL. 1. Another possible cause for the low GE values is the heterogenous nature of the reaction. 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. 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 . The highest GE value (up to 90%) has been achieved when using triethylaluminium as catalyst [97‐98]. In this reaction. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). In the past.4. thus resulting eventually in low GE values . the hydroxyl groups of starch are supposed to function as initiating sites. Three routes have been studied in detail: starch esterification. The results of this study have been used as input for the synthesis of the starch‐g‐PCL. The method applied basically consists of three steps (temporary partial protection of starch‐hydroxyl 16 . and the presence of liquid‐solid reaction system leads to reduced reaction rates between starch and ε−caprolactone. and starch‐g‐ biopolymer formation. and difficult to handle since it releases ethane.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) using i Al(O Pr)3 as the catalyst. Another approach consists of the Ring Opening Polymerization (ROP) of ε‐caprolactone monomer in the presence of starch.2. starch/ biopolymer blending. which however is extremely air‐ and water‐sensitive.3. defined as the percentage of grafted polyester to starch compared to the total amount of homopolymer and grafted polyester) .
UK. Young: Fractionation of starch.S.F.A. Breed. addition of co‐solvent. E. The Biocomposites Centre. Corn Refiners Association. Stevens: Green Plastics. 1984. 2007 17 . in Starch: Chemistry and Technology (Ed. 2004. C.B. The synthesis of two interesting compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) for starch/PCL blends is discussed in Chapter 6. Inc. 2008. . and polyhydroxyalkanoates. Denmark: http://www. Somerville: Mini review: Plant polymers for biodegradable plastics: cellulose. de Bragança. Mol.. O.. J. in Modified Starches: Properties and Uses (Ed. ROP of ε−CL on the remaining starch‐hydroxyl groups. Washington D. A. 1986.C. are also reported. and application of different catalysts) on starch esterification using vinyl fatty esters (vinyl laurate and vinyl stearate) is provided in Chapter 4. References .N. 1. O.M. York. Gywnedd. Wurzburg). Passchall). Whistler. Y. An Introduction to the New Science of Biodegradable Plastics.L. Academic Press. Website of International Starch Institute. Bemiller. and removal of the silyl groups). New Jersey. accessed on August 01. Central Science Laboratory. Inc. The model. Boca Raton. Poirier.html. Sand Hutton. . Wurzburg: Introduction. Based on these preliminary results. together with the mechanical and thermal properties of the synthesized starch esters are reported in Chapter 5.starch. including the mechanical and thermal properties. .B. 105‐122. London. E. Corn. CRC Press. Princeton University Press. . The use of these compatibilizers in starch/PCL blends. 2005. Bangor. . Introduction by trimethylsilyl groups.5. University of Wales.H. 2000‐2005. starch. . Corn Refiners Annual Report 2007. Fowler: Industrial markets for starch. Summary Report for the European Union.dk/isi/stat/rawmaterial. Part of A Global Economy. C. 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.. 1. The results were quantified using a statistical model. Agricultural and Rural Strategy Group. C. Nawrath. A preliminary study on the effect of several process variables (reactant ratio. UK. UK. USA. USA. 2002. R. USA. . P. 1995. R. Aarhus.
Munich. 18 .E. USA. Prog. 1990. 53 . in Kirk Othmer Encyclopedia of Chemical Technology Volume 22. Daniel: Starch. Wolf. Link. Patel. the Food Biopack Conference Proceeding (Ed. 11. Plastics Europe. G. . Processing.M. Strong: Plastics: Materials and Processing. Germany. 2005. Germany.M. Sci. Inc. X. Applications. Chandra. 1273‐1335 . 2008 . Ind. Weith: Plastics. Marscheider‐Weidemann. Hanser Publishers. in Starch: Chemistry and Technology (Ed.gov/garbage/plastic. B. C. G. UK. Brussels. Bastioli: Global status of the production of biobased packaging materials. Starch/ Starke 2001. J. demand and recovery for 2006 in Europe. A. Selke: Packaging and the Environment. Cleaner Prod. C. 557‐562. Prentice‐Hall. . and Regulations. 105‐109 . 2000. Association of Plastics Manufacturers. Ren: Biodegradable plastics: a solution or challenge?. R. 31. . London.Chapter 1 . E. Fifth Edition (Ed. Department of Dairy and Food Science. Passchall).L. Lancaster.epa.. Rustgi: Biodegradable polymers.F.. Eng. 53. Belgium. Frederiksberg. M. Properties. Braun: Environmental impacts of biobased/ biodegradable packaging. Selke. J.. Watson: Corn and sorghum starches: production. Institute for Prospective Technological Studies and European Science and Technology Observatory. 1984.J. New Jersey. F. Inc.351‐355. A.D. 2003. Englewood Cliffs. Properties. Applications. Seidel).R. S.htm. USA. Starch/ Starke 2001. accessed on August 19. O. 1998. Hernandez. U. J. Academic Press. .J. Culter: Plastics Packaging. 23. Inc. Weber). USA: http://www. Husing. 27‐40. Website of Environmental Protection Agency. Schleich. S. Munich. J. . R. . Polym.N. R. The compelling facts about plastics: An analysis of plastics production. Whistler.. Inc. Denmark. USA. . . 2000.J. C.E. .A. .W. J. R. Ehrenstein: Polymeric Materials: Structure. Techno‐economic Feasibility of Large‐Scale Production of Bio‐based Polymers in Europe. Angerer: Technical Report Series. 2000. John Wiley & Sons.B. 1939. R.L. 1996. M. Hanser Publishers. Whistler. J. . R. Crank. A. Washington DC. 2008. Fritz. Chem. Bastioli: Global status of the production of biobased packaging materials. Bemiller. S. The Royal Veterinary and Agricultural University. Technomic Publishing Company. Hoboken.
P. O. Yu.V. C. Weber). 2003. Forssella. K. Bagdi.G. Press Release: Major Price Hikes in Plastics. J. Denmark. a model for thermoplastic starch. Ind. Starch/ Starke 2001. 1980.P. Res. J. L.S. Schwach. Gandini: Surface chemical modification of thermoplastic starch: reactions with isocyanates. Polymer Sci. Curvelo. Muller. Doane: Starch‐based blown films. 275‐282 19 . R. . 21. Sci. . Moates. Carbohydr. Yu. Carvalho. Prod. Martin. 2005. Avérous. 2000.F. 4. Chambre Syndicale des Emballages en Matiere Plastique (French Plastic and Flexible Packaging Association). E. 2008. Carbohydr. Nielsen. 372–380 . Ind. properties. X. Physical and mechanical properties of biobased materials – starch.B. J. Y. Frederiksberg C. 1996. interaction. Compos. A.F. Polym. 494–504 . Wang: Properties of starch blends with biodegradable polymers. 1997. A. A. 385–409 . polylactate and polyhydroxybutyrate. Vliegenthart: Interaction between dry starch and plasticisers glycerol or ethylene glycol.L. P. 62. J. Ma: The effects of citric acid on the properties of thermoplastic starch plasticized by glycerol. 53. Part C: Polym. the Royal Veterinary and Agricultural University. July 15. J. epoxy functions and stearoyl chloride. Petersen.J. 53. Polym.M. N. van Soest. 53. .H. Mikkilti. 1491‐1494 . 2005. 34..M. K.H. Appl. Macromol. A. Smits. Gao.M. 13. 2003. Parker: Phase and glass transition behaviour. K. Kruiskamp. J. Eng. Biodegradable thermoplastic starch.M. Rev. 592‐595.. and Olsen. Crops Prod. P. and Couturier: Properties of biodegradable multilayer films based on plasticized wheat starch. Dev. and Pukanszky: Thermoplastic starch/layered silicate composites: structure. measured by differential scanning calorimetry and solid state NMR spectroscopy. report of Food Biopack Project (Ed. Yang. Spain. M. status and perspectives. J. J.K. P.J.G. T. Chem. 331–336 .A. Biobased packaging materials for the food industry. Westhoff.Lin: Note. G. Introduction European Commission Directorate General of Joint Research Centre.. R. 57. Department of Dairy and Food Science. Otey. 2005. B. W. of concentrated barley starch‐glycerol‐water mixtures. Starch/ Stärke 2001. Seville. 409–416 . Y. A Real Case of “Force Majeure” that should be passed through to Packaging Prices Immediately. 43. X. Wang. 356‐361 . J. 1–17 . . Starch/Stärke. Interfaces 2006. Wang. F.
Polymer 2003. 1986.A. . Jiang. . Mullen. 55‐78. H. 58.D. E. Myllarinen. J. Wurzburg. chapter 4 in book Modified Starches: Properties and Uses. E. Sci. Chem. Bertrand. K. Sun: Reinforced thermoplastic acetylated starch with layered silicates. J. Starch/Stärke 2005.C. I. destructuration and hydrolysis during O‐formylation. 1209‐1217. Inc. M. Qiao. Matzinos.T. D. Y. 581–586 . Eng. 11–17. P. 2004. J. Inc. Polym. Crops Prod. Divers. 35. 1517–1526 . 2006. CRC Press. M. USA. Tserki. Carvalho. Borredon. Appl.. O. Appl. 1. E.D. 1647‐1656. 2594–2604 . 381‐384 . Prinos. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Grohens: Starch modifications. Wurzburg). Boca Raton. Bikiaris. F. 1986. Ind. S. Gattin.A. X. Degrad. Ind.W. Ind. S. Copinet. Ibezim. 57. Ramsay. 2001. 139–147 . 17–24 . I. CRC Press. B. J. C.. Panayiotou: Processing and characterization of starch/polycaprolactone products. 705‐721. B. J. USA. A. J. 77. 20 . Feller. Wurzburg: Cross‐linked starches.C. 44. Pilin. O. in Modified Starches: Properties and Uses (Ed. Polym. Carbohydr.W. Levesque.D. Mullen.D. Polym. 58 (9). A. Pacsu: Starch studies: Possible industrial utilization of starch esters. Coutuier: Comparative biodegradation study of starch‐ and polylactic acid‐based materials. 2002.J. 74. E. Environ. E. Feil: Morphology and enzymatic degradation of thermoplastic starch– polycaprolactone blends. A. Aburto. O.B. . 1942. Rodriguez‐Gonzalez. 65. Hulleman. Kontoyiannis. Jarowenko: Acetylated starch and miscellanous organic esters. Favis: High performance LDPE/ thermoplastic starch blends: a sustainable alternative to pure polyethylene. V.H.B. 1997. 2004. M. B. F. Vikman. P. Pacsu: Starch studies: Preparation and properties of starch triesters.W.A.A. W. Boca Raton. C. Sci. J. 1943. Sagar. Ribeiro. C. Polym. 56.J.F. Silva. Polym. 24.B. van der Zee. T. T. 46–51 . Chem. 389‐398 . Rodriguez‐Gonzalez. Andrade: Reactive processing and mechanical properties of cross‐linked maize starch. 1999. Favis: Rheological and thermal properties of thermoplastic with high glycerol content. . Journal of Applied Polymer Science 1995.P.W. (editor). B. R. J. W. C. Merrill: Properties of fatty‐acid esters of starch. Alric.Chapter 1 . Ramsay. 34. G. Stab. C. Thiebaud. Eng. Y. .
and biodegradability of fatty‐acid esters of amylose and starch. Alric.F. Aburto. Appl. Polym. 118‐121. Panayiotou: Synthesis. J.K. Munich. E. C. Guegan. Marechal. Cellulose 2004. A. Wiley‐VCH.Coleman. . A. J.H. A straight way to regioselectively functionalized polysaccharide esters. W.828. Hanser Publishers. F. Bikiaris. Macromolecules 1996.J. . Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. C. 1997. Symp. 1999. Starch‐Starke 1999.E. H. M. Starch‐Starke 1999. Hu. J. A. Tuschoff: Acylation of hydroxy compounds with vinyl esters. Hu). Senocq. J. Starch‐Starke 2004. in Reactive Polymer Blending (Ed. E. P. Degee. W. E. Nakayama. Borredon. D. Carbohydr. . 56. 145‐152. 34.Stokes. C. 175. Dicke. 2001. Germany. Sci. Carbohydr. Hamaili. Macosko. . Prinos. Aburto. D. Technomic Publishing Company. US Patent 2.C. C. Aburto. Lancaster. M. 74. P. de Graaf. Starch‐ Starke 2005. Introduction. R. M. Polym.E. 2001. 51. . G.Fennel Evans: Fundamental of Interfacial Engineering.. E. C. J. Introduction . S. T. Alric. USA. Inoue: Compatibilizers for melt blending: Premade block copolymers. . 33‐43 21 .E. .W. Dubois: Synthesis and characterization of compatibilized poly(ε‐caprolactone)/ granular starch composites. Thiebaud. Smith. Borredon. Prinos. W. J. P. C. Bakker. 1995. 137‐144. Inc. Duquesne. P. A. Alric. Janssen. A. Rutot. 302‐307. D. Thesis. P. D.E. I.Graf. Khandpur. R. R. New York. Alric. S. P. . Rijksuniversiteit Groningen.V. 57. Aburto. USA. J. Bakker and G. Alric. Macromol. Polym. A.D.M. M. A. 1440‐1451. 5590‐5598 . R. E. E. G. 101‐112. . . Broekroelofs. I. J. Al‐Higari: Acylation of starch with vinyl acetate in water. B. . Mouysset‐Baziard. Mormann. 1991. I. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides.928. 51.Painter: Specific Interactions and the Miscibility of Polymer Blends. characterization. 132‐135. 1960 . Scott. 11. G. I. J. 1996 . L. de Graaf: Ph. A. J. 29. I. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. 255‐263. 1997. Beenackers: The kinetics of the acetylation of gelatinised potato starch.H. Thiebaud. Panayiotou: Properties of octanoated starch and its blends with polyethylene. Aburto. Bikiaris. 28.
1998. thermal and morphological characterization of recycled LDPE/corn starch blends. A. L. C. R. 287‐291 .D. Rosa: Mechanical.S. Proc. and tensile properties of Low‐Density Polyethylene (LDPE) and corn starch blends. 1841‐1846 . Danjaji. Effect of amylopectin to amylose ratio in starch.S. 75–81 . Polym.‐J. thermal and morphological properties of virgin and reprocessed LDPE/corn starch blends. Bikiaris. 1242‐1251 . Mani.M. 88. Appl. 1503–1521 . L. . 20.. 59. Mei: Study and development of LDPE/starch partially biodegradable compounds. U. Tech. Polym. Ismail.D. Danjaji. J. Polym. Ismail. A. U. D. 167–172 . N. I.S. Appl. Effect of compounding and starch modification on properties of starch‐filled Low‐ Density Polyethylene. Wang. R. H. A. Polym. Z. 1998. L. J. Nawang. St‐Pierre. Ramsay. R. Polym. Mater. 30. Chem. 88. J. Koutsopoulos.D. Res.A. Carbohydr. Appl. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends. Ribes‐Greus: Thermal and viscoelastic properties of some commercial starch products. 16.G.G. 1‐9 . Ishiaku. Pedroso. Pavlidou.I. 21. III. Sci. J. 59. L.A. Liu. H. morphology. 1467‐1475. 2005. B. 38. J. Polym. Nakamura. Contat‐Rodrigo. 236–241 . Polymer. M. 2003. 2005. .H. D. Frangis. Adv. N. Sung. Bikiaris. Almeida.L. Mechanical properties of sago starch‐filled linear low density polyethylene (LLDPE) composites. Ramsay. Mohd Ishak: Material behaviour. 1998.G. Ind. 1991. B. R. D. H. E. D. Testing. G. Mohd Ishak: Material properties.S. Panayiotou: LDPE/plasticized starch blends containing PE‐g‐MA copolymer as compatibilizer.A. Pedroso. Eur. Cordi. Nawang. I. Ishiaku. Z. Technol. J. J. Degradation studies and moisture uptake of sago‐starch‐ filled linear low‐density polyethylene composites. 647‐655 22 . Stab. Z. Rosa: Effects of the compatibilizer PE‐g‐GMA on the mechanical. Sci. Evangelista. Polym. 1997.A. 310–317 . Vouroutzis. Panayiotou: LDPE/Starch blends compatibilized with PE‐g‐ MA copolymers. Verhoogt: Processing and characterization of thermoplastic starch/polyethylene blends. W. N. Polvm. Polym. 2001. 2005.Chapter 1 . R. 70. Favis. Duran. 2003. 2002. Z. J.J. Y. Eng. Gelina. W. Prinos. Jane. E. K. Degr. N. C. Sun: Effects of polyethylene‐grafted maleic anhydride (PE‐g‐MA) on thermal properties. 162–163. Testing.M. Sci. 2904–2911. 34.S. Nikolov.
H. Manan. Park: Modification of aliphatic polyesters and their reactive blends with starch. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. D.M. J. Ranucci.P. 1507– 1516.. Sci. U. 3539–3546 . Moro. R. Sci. 89. F. 157. Polym. O.K. Ang.D. 43. Maiti: A biodegradable film. Y. Lindblad. J. Azemi: Effect of starch predrying on the mechanical properties of starch/poly(e‐caprolactone) composites. Tech. S. 1. B. M. A. 3976–3987 . U. Ind. 81. K.‐C. 284‐287 . 2004. 1298‐1305.S. Sankar. 1984. J. U. Makromol. J. Ishiaku. Cho. Appl. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. K. Adv. 4157–4167 . L. Prod.M. J. Sen. C. 26. C. Introduction . 2003. Sci. O. M. Ishiaku. 1997. 877–884. S.G.S. Odusanya. J. PHB‐V.S. K.H.C.S. Sci. Res. 41. dos Santos Rosa. . M. . glycerol modified.R. John. Sci. Guedes. H. in Polym. S. T. Sci. 16–19 . Ind. Res. Albertsson: Controlled Ring‐Opening Polymerization: Polymers with designed Macromolecular Architecture. Jana. Westhoff. 1659‐1663 .H. Adv. L. Chem.P. Dev. Chem. Odusanya. Adv. 12. Appl. Liu. F. cross‐linked and glycerol modified starch. Eur. Kammer: On mechanical properties of sago starch/poly(ε‐caprolactone) composites. Otey. Karlsson: Polymers from Renewable Resources. Kim. Ryner. R. and their blends with starch in soil compost. B.F. Stridsberg. 2004. Polymer 2000. M. Appl. Garg. Doane: Starch‐based blown films. Kim. 2002. Polym. Jana: Studies on the properties and characteristics of starch– LDPE blend films using cross‐linked. Bhattacharya Residual stresses and density gradient in injection molded starch/synthetic polymer blends Polymer 2000. . Eng. C. Averous. J. 32–45 . W. Polym. 41. Otey. J. Azemi. S. A‐C. 2. Fringant: Properties of thermoplastic blends: starch–polycaprolactone. 1987. Part A: Polym. Kim. N. . J. 2002. Polym. Polym. 2000. . 42‐65.K. 40.. M. A. Polym. T. S. J.Y. 139‐161. Tang. 87.M. Siddaramaiah. 157. Yang. Westhoff: Starch‐Based Films. Preliminary diffusion evaluation. D. 1999. 2007. Low density polyethylene/ starch blend films for food packaging applications. Sci. Environ. J. Pilot plant investigation for production of the biodegradable film. A. 9177–9190 .H. C. Rodrigues.W. P. Z. 2003. A.. B. 23 . Eng.M. Polym.N.M. Manan. Calil: Effect of thermal aging on the biodegradation of PCL. 2001. Chem. Chem. 179‐187. Jung. Albertsson. . Polym. Eng. B. 23. 35. 267. 23. M. Polym. M.K. Dole. E. 1139–1148. Raj.
J. P. . Degee. 80. R. Composite Interfaces. Narayan. 24 . 7. . M. R. D.Chapter 1 . P. Polymer 1999. 175. J. Macromol. E. Martuscelli.E. Dubois: Synthesis and characterization of compatibilized poly (e‐caprolactone)/ granular starch composites. 1998. P. R. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. Symp. 19. M. Polym. P. Avella.S. Rutot. 2003. 40. 2002. Macromol. D. Dubois. M. 2001. Rapid Commun. Polym. Krishnan. Sci. Bhattacharya: Synthesis and characterization of starch‐ graft‐polycaprolactone as compatibilizer for starch/polycaprolactone blends. M. 2000. E. Dubois: Aliphatic polyester‐grafted starch composites by in situ ring opening polymerization. 1432–1442. 215–225. M. . Appl. Polymer 2000. Degée. P. Mani. Avella.E. Errico. 127–134. P. Tang. 33–43 . 283–286. Duquesne. M. Errico. Rimedio. Sadocco: Preparation of biodegradable polyesters/high‐amylose‐starch composites by reactive blending and their characterization. Raimo. R. M. Degrad. 3091‐3100. Rimedio: Polymer communication: Preparation and characterisation of compatibilised polycaprolactone/ starch composites. . Laurienzo. and P. Rutot. 3875–3881 . Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. 41. C. R. 83. Stab.
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. and MALDI‐TOF mass spectrometry. ring‐opening polymerization . 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. also polydioxanone with an OiPr end‐group was formed (20‐30 %). Keywords: biodegradable.2. Systematic studies showed that the polymer yield is a function of the reaction temperature and the reaction time.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. The products were synthesized at 60‐100°C and characterized by 1H‐ and 13C‐NMR.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose)/Al(OiPr)3 initiator system to yield polydioxanone with a protected monosaccharide end‐ group is described. polyesters. with higher temperatures (100°C) leading to lower yields.
reaction of the metal‐alkoxide with an appropriate alcohol allows the synthesis of end‐capped poly‐lactones. tocopherol. polydioxanone also shows good performance. and polylactides. Polydioxanone (poly(p‐dioxanone) or poly(1. probably because the p‐dioxanone monomer has become commercially available only recently . whereas the high temperature for polylactides results in thermal degradation and undesirable transfer reactions during synthesis and processing steps . polyglycolide. which is a unique compromise between application and processing temperature. and is tougher than polylactides and even HDPE . This chapter describes experimental studies on the catalytic ROP of p‐ dioxanone in the presence of a protected glucose molecule (1. The catalysts are metal alkoxides with Lewis acidic character [8.Chapter 2 2. Nishida et al  reported that polydioxanone decomposes to non‐toxic gases (CO2 and H2O) by microorganisms. 9]. has interesting product properties compared to other aliphatic polyesters. pregnenolone. It is fully degraded in the body within a period of 180 days . particularly for nano‐ encapsulation systems for drug delivery .4‐dioxan‐2‐ one)). are interesting polymers because of their good product performance and biodegradability . From a biodegradability point of view. The relatively low melting point of polycaprolactone limits its applicability. Despite its good properties. Polydioxanone has a tensile strength close to 48. stigmasterol and ergocalciferol)  and also with protected monosaccharides (galactopyranose/ glucofuranose) [6‐7].3. The ring‐opening polymerization (ROP) of p‐dioxanone using metal catalysts such as aluminum isopropoxide [Al(OiPr)3]. such as polycaprolactone. This melting point is considerably higher than typically found for polycaprolactone (60 oC) and lower than that of polylactides (at least 175°C). Its melting temperature is close to 110 oC. accessible by the polymerization of p‐dioxanone (1). or zinc lactate has been reported. stannous octoate [Sn(Oct)2]. quinine.1. Several polymers with bioactive alcoholic and phenolic end‐groups of interest for drug‐related applications were synthesized (for example geraniol.g. Besides 26 . Introduction Aliphatic polyesters. Biodegradable aliphatic polyesters end‐capped with sugar molecules have been studied extensively for use in biomedical applications.2. 2). only limited information about the synthesis and properties of polydioxanone is available in the open literature. 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. which to the best of our knowledge is the first study to functionalise polydioxanone with a monosaccharide. The synthesis of protected monosaccharide end‐capped biodegradable polymers is usually performed via Ring Opening Polymerization (ROP).4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. Exchange of the alkoxide group by e.3 MPa and an elongation at break of 500‐600%. testosterone.
NMR) due to the poor solubility of the products in standard organic solvents.3.55 mmol of 2 and 0. Boehringer Ingelheim. The resulting clear solution was stirred for 2 h at 50 oC.g. 0.07 g. After the pre‐determined reaction time.8 g.1.1. Toluene (Labscan) was dried and stored on molecular sieves 3 Å (Labscan) under nitrogen. Methods All polymerization experiments were carried out under a protective nitrogen atmosphere using standard Schlenk‐ and glovebox techniques. Typical example for the synthesis of polydioxanone end‐capped with 1. grafting efficiencies are difficult to determine in this system by standard analytical techniques (e. hot dichloroethane (20‐25 ml) was added to completely dissolve the partly solid polymer at room temperature. 3]. CDCl3 was obtained from Sigma and was used as received. part of this solution (350 µl.2. Germany) was purified according to the procedure described by Raquez et al [2. heptane (Acros).3.2.2. poorly soluble heterogeneous systems. 2. The reaction mixture was clear and colorless during the reaction.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2) 2 (1. As such. Particularly interesting in this field are starch polymers grafted with polydioxanone. Subsequently. 1.44 g. relatively low molecular weight compounds reported in this study may be viewed as model systems for more complex. 2. Materials and Methods 2.08 mmol aluminum isopropoxide) was added to pure 1 (0. and diethylether (Labscan) were used as received. The hot solution was 27 .2. which was pre‐heated till about 60°C to obtain it in a liquid state. Analytical grade dichloroethane (Labscan). the mixture was brought to room temperature and several drops of HCl (1 N) were added to stop the reaction.5 mmol) was dissolved in toluene (1 ml) at 50 oC.2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2). (20 mmol) of aluminum isopropoxide to 20 ml of toluene. To this solution. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System potential applications in biomedical products. Next.2. 97% (Sigma) and aluminum isopropoxide. the synthetic pathways and the soluble. containing 0. Materials p‐Dioxanone monomer (1. 98+% (Aldrich) were used as received. 5. 8 mmol).2. However.8 ml of a solution of an aluminum isopropoxide stock solution was added. 2. This stock solution was prepared by adding 4. The polymerization was allowed to proceed for 16 h at 100 oC.2. the results of this study are also of interest for the preparation of starch/polydioxanone polymers using ROP.
Processing of the raw data was performed using VNMR software. 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). Product Analyses NMR analyses were performed in CDCl3.2. If the amount of sugar is in excess with respect to the aluminum catalyst.2.3. This leads to the following equation: Xn exp X n exp is = A A H-repeating unit H-galactopyranose end group +A (2. 1H‐ and 13C‐NMR spectra were recorded on a Varian AMX 400 NMR. MALDI‐TOF spectra were recorded on an Applied Biosystems Voyager DE‐PRO machine using dithranol/ NaI as the matrix (linear mode). X n exp of the product. Calculation of Average Degree of Polymerization The Theoretical Average Degree of Polymerization.2.) Here. 2. 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. two polymers with different end‐groups are present. 2D‐NMR spectra were recorded on a Varian Unity 500 NMR.Chapter 2 precipitated in a heptane/ ether mixture (300‐400 ml. As will be shown later.3.) H‐NMR was applied to determine the experimental average degree of polymerization.4. it is assumed that all available initiator is used effectively. 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. one initiated on a galactopyranose molecule and the other on an isopropoxide group. 2.1. 40 oC) until constant weight.) H-isopropoxide end group 28 . 4:1 by volume) at 4‐8 oC. The isolated yield at this condition was 68%.
7 ppm − 6 × A 5. 2. the protected mono‐saccharide 2 reacts with Al(OiPr)3 to form the actual catalyst for the polymerization reaction.) 2.1 ppm (2.7. eq. Screening Experiments Initial experiments to synthesize protected galactopyranose end‐capped polydioxanone (3) were performed using 1. is rewritten as: Xn exp = [ A 3.1 ppm (2.1.1.).) or.6 ppm ] / 6 A 5. 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. Results and Discussions 2. is calculated using: R gp = A H − galactopyranose end group H − isopropoxide end group A (2. in term of the NMR resonances : R gp = A 5.6 ppm + A 220.127.116.11.4− 4. Schematic representation of the polymerization reaction including atom numbering scheme 29 .3. Typically. R gp .5.3.6 ppm A 5. 2.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose (2) and Al(OiPr)3 as the catalyst precursor (Scheme 2.3. The exchange reaction typically takes place at 50°C for 2 h. The polymerization reaction was performed as a two step process.). To compensate for this effect. 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 2.6: 96 was applied. an Al ‐ monosaccharide 2 ‐ monomer 1 mol ratio of 1: 6. To avoid the formation of isopropoxide end groups. In the first step.1.) The ratio of the two different types of polymers (either end‐capped with galactopyranose or an isopropoxide group).6. an excess of monosaccharide 2 on Al was used (see eq.
H‐NMR spectra (Figure 2.1 and 2.) are not particularly informative. Although the spectra clearly indicate the presence of the monosaccharide 2.3. It is difficult to determine the end‐groups of the polymer on the basis of 1H‐NMR. Product analyses The products were analyzed using 1H‐ and 13C‐ NMR and MALDI‐TOF. 1 30 . of interest is the presence of a small multiplet at about 5. although it is evident that p‐dioxanone polymerisation occurred.1 ppm compared to the monomer.40 ppm) unit are broadened and shifted up to 0.1.51 ppm. However. Typical 1 H‐ and 13C‐NMR spectra of 3 are shown in Figure 2.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. The OiPr : 2 end‐group ratio for the standard experiment was 1: 2. 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.) In the next step.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).1.2. 2. respectively.1. Typical isolated product yields are 68% at these conditions.15‐1. The off‐white solid reaction product was collected after a dissolution‐reprecipitation process using dichloroethane and a heptane/ diethyl ether mixture. Proton resonances of the CH3 group of the isopropoxide end‐group.1 ppm. This multiplet is characteristic for the CH proton of an isopropoxide end‐group. together with the CH3 groups from the protecting groups of the sugar appear in the range 1. The typical proton resonances of the p‐dioxanone (3.67. the in‐situ formed catalyst was reacted with p‐dioxanone monomer 1 at 100 oC for 16 h.
68. 1.2 ppm. 1 (c).3 ppm are shifted considerably.2. The carbon resonances of the polydioxanone backbone are present at δ = 63. In addition to the mono‐saccharide end‐ group.1. Particularly the C2‐C6 carbons in the range δ = 66.3. H‐NMR Spectra of: (a).52 ppm (β’).17 and 69. 3 1 C‐NMR (Figure 2.28 ppm in 2 to δ 66.2 ppm. For instance.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose. p‐dioxanone monomer. 2 (b). 1. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.8.61 (γ’) and 73. C‐6 is shifted from δ = 62.2.2‐70.3.2 and 96.2. characteristic resonances of an ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end group are present at δ = 61. Carbon resonances arising from the monosaccharide end‐group are present between δ = 66.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.) is more informative and clearly shows the presence of a polydioxanone polymer backbone and a monosaccharide end group.2 ppm in product 3. 13 31 .
. 2 (b).1.3. 1 (c).4‐di‐O‐isopropylidene‐α‐D‐galactopyranose. 1. 2 32 .3.70 (>C‐(CH3)2) and between δ = 21. and 2.48 ppm (‐CH‐ (CH3)2) and between δ = 21. An overview of the data is given in Tables 2. the numbering scheme of carbons and protons is given in Scheme 2.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.97 (>C‐(CH3)2). A typical example of a part of the 2D‐NMR spectra is given in Figure 2.2.2. D‐NMR (HSQC) was applied for complete peak assignment of the product.18.104.22.168 and 109.3.76‐25. 1.Chapter 2 Figure 2. p‐dioxanone monomer. C‐ NMR Spectra of: (a).97 ppm (‐CH‐(CH3)2).76‐25. 3 13 Carbon resonances of the isopropoxide groups from the by‐product (isopropoxide end‐capped polydioxanone) are present at δ = 68. The resonances of the protecting group of the sugar appears at δ = 108.
around 1. which were determined from HSQC spectrum due to overlapping resonances.83 p‐dioxanone Monomer (1) H‐α H‐β H‐γ a 4.74 3.1.31 4.301 3. except values with *. 33 .40 all values were determined using 1H NMR. HSQC 13C‐1H spectra of 3 Table 2.78 4.8 Reactant peaks (ppm) Galactopyranose (2) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 5.60 4. 1 H‐NMR peak assignments a Product peaks (ppm) Poly(p‐dioxanone) end‐capped with protected galactopyranose (3) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 H‐α H‐β H‐β’ H‐γ H‐γ’ H‐OH 5.25 3.32 * 4.17 3.26 3.54 4.77 3.16* 4.732 broad.3.09 4.50 4.59 4.68 4. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.00 4.
2 68.4. 34 . An example of a MALDI‐TOF spectrum of 3 recorded in a dithranol/NaI matrix is given in Figure 2.52 63.07 Reactant Peaks (ppm) Galactopyranose (2) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 96.61 170. MALDI‐TOF was also applied to characterize the products. The molecular weight distribution of the major peaks may be represented by the following relation: m / z = 23 + 260 + 102n (2.21 NMR analyses also allow calculation of the molecular weight of the products. which is the molecular weight of a repeating dioxanone unit.11/ 170. A detailed procedure is given in the experimental section.09 65. 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. For this purpose.31 70.2.54 166.10 71. A typical molecular weight distribution is observed.33 70.17 69.58 68. 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. corresponding with an average number molecular weight (Mn) of about 1600.80 61.28 p‐dioxanone monomer (1) C‐α C‐β C‐γ C‐carbonyl 61.92 68.28 70. The difference in molecular weight between the main peaks is 102 g/mol.00 171.Chapter 2 Table 2.57 62.2 73.21 70.8. the ratio of the intensity of the end groups and the polymer backbone peaks is determined. The product obtained at screening conditions (16 h reaction time at 100°C) contains on average 13 monomer units.92 66.75 70.
10. Typical MALDI‐TOF spectra of 3 A second molecular weight distribution is also clearly visible when enlarging the spectra.5 +102 1400 1600 1800 2000 2200 2400 Mass (m/z) Figure 2.) and imply the presence of dioxanone polymers with carboxylic end groups.) (2. 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.) . This series may be described by the following relation: m / z = 23 + 60 + 102n (2. two other distributions are present. m / z = 23 + 18 + 102n m / z = 1 + 18 + 102n 35 (2.10.). in line with the NMR data.) and (2. ionized with either Na+ or H+.9.5. see Figure 2.4. for details.5.11. Furthermore. although both with a very low intensity (Figure 2.11.) This relation is indicative for the presence of a polydioxanone polymer containing an isopropoxide end‐group. These distributions may be represented by equation (2.
13].Chapter 2 50 40 792. 2‐(4′‐ hydroxybenzeneazo)benzoic acid (HABA)).1. the catalyst precursor is treated with monosaccharide 2 resulting in an alcohol exchange 36 . Enlarged MALDI‐TOF spectra for 3 Although MALDI‐TOF clearly demonstrated the presence of various types of end‐groups in the product.7 796. is not particularly suitable for polydioxanone.3. Various other matrices were tested (e. 2. Most likely the matrix (dithranol/NaI). In addition. In the first step. Various samples with.g.7 %-Intensity 30 20 858. different molecular weights were analysed. but in all cases poor quality. The observed differences in the molecular weight distributions of the various samples were only marginal. Mechanistic aspects Both NMR and MALDI‐TOF measurements imply that the main product is indeed the desired monosaccharide end‐capped polydioxanone.5. although successfully applied for galactopyranose‐end‐capped polycaprolactone . according to NMR.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. 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. [6. low resolution spectra were obtained.2.3 10 836. small but significant amounts (20‐30%‐mol) of polydioxanone chains with an OiPr end group are present.
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. termination of the chain growth may occur by reaction with an alcohol. (R = monosaccharide 2) 37 . Next. The resulting Al‐alkoxide may again initiate a polymerisation reaction.6. 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. Subsequent coordination and insertion of dioxanone molecules leads to the formation of a polymer chain with a monosaccharide end‐group. a p‐dioxanone molecule will coordinate to the Lewis acidic aluminium center followed by an insertion step. During the reaction. 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. Irreversible termination of the polymerisation is performed at the end of reaction period by adding dilute acid to the polymerisation mixture.
74 18.6 6.).55 Product Properties Avg.5 T (°C) 100 100 100 100 100 60 60 60 60 60 60 60 60 60 60 dioxanone/ 2 ratio (mol/mol) 3. two different reaction times (1.7.53 59. 2.04 19.1 3.5 1.3 and 62.6 was applied. These may be present in the reaction mixture because the exchange reaction between Al(OiPr)3 and 2 was incomplete.2.3 6.4 2/OiPr ratio Rgp (mol/mol) 2.50 25.3 81.74 52.2 91. will be formed when the polymerization starts with an aluminum alkoxide with a remaining Al‐OiPr group (eq.8 2. despite the excess of 2.3 6.4 3.5 67.6 6. 2.11 22.22 57.1 2. time and the mol ratio of monomer to monosaccharide) on the yield.6 89.30 62.50 35. A total of 15 experiments were performed at two polymerization temperatures (60 and 100 oC).22 13.08 30.9 1 7. Table 2.3 6.93 37.7.6 81. However.5 hrs and 16 hrs) and a p‐dioxanone to monosaccharide 2 molar ratio ranging between 3.6 6.3 6.3 6.6 86.6 6.5 1.12 42.8 3.96 2 3 a For each experiment.9 2.21 54.6 2.49 13.78 10. a (nearly) fixed Al(OiPr)3 : monosaccharide ratio of 1 : 6.70 58. 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 .3‐6.5.45 19.3 3.3.6 80.16 46.29 14. the end group may also be formed by a termination reaction with free isopropanol.7 2.93 8.6 t (h) 16 16 16 16 16 16 16 16 16 16 1. 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 96.5 81. 2.5 84.10 15.5 1. OiPr‐end‐capped poly(p‐dioxanone). formed in the first step of the polymerisation reaction (eq. An overview of the experiments and the results are given in Table 2.6 6.1 2.36 17.5 92. Systematic Studies The effect of important process variables (temperature.8 2.4 2.).3 6.3.6 6.81 16.85 14.6 6. In all cases.22 16.6 6.5 2. Chain Length (Xn exp ) Yieldb (%) 30.4 2.74 23. degree of polymerisation of the product and the end group distribution was determined.5 1.1 86.28 33.3.0 87.Chapter 2 The minor product.6 91.
3. To the best of our knowledge. 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. 39 . Another possibility is the occurrence of polymer crystallization. equilibrium conversion is not yet achieved within 1. For the two experiments with a dioxanone: 2 ratio higher than 16 (≅ 23 and 60).3. 1. the yields for reactions conducted at 60 oC are always higher than those performed at 100 oC. One of the experiments (S111) resulted in a very low yield (31%) compared to the other reaction. 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. no detailed explanation has been put forward to explain this anomalous behaviour.5 h. see Table 2. the monomer conversion reaches a maximum value before going down to the equilibrium monomer conversion. the yields are higher when performing the reaction at 16 h reaction times. However. For this particular experiment. when the reaction is performed at a low dioxanone: 2 ratio (≅ 16). Product yield The isolated yields of the reactions are all but one between 67 and 96% (see Table 2. The latter is due to the slight exothermicity of the reaction . The effect of the reaction time on the product yield may be derived from the data provided in Table 2.3.3. Evidently. will be required.5 h is higher than the yield at 16 h. 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. It is likely that this relatively low molecular weight compound dissolves partly during isolation/ purification procedure leading to lower isolated yields.5 and 16 h reaction time). Higher polymerisation yields were obtained at lower temperatures. A more detailed analysis on the actual nature and composition of the polymerization products. as suggested by Raquez et al .3. the yield at 1. 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. For instance. and particularly when comparing the data in set 2 and 3 (60oC. Clearly. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 2. which reduced to 75% when increasing the temperature to 120oC.2. It was shown that at lower monomer to catalyst ratios.).2.). the equilibrium conversion of dioxanone at 80oC was about 80%. et al .1. 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. a low dioxanone: monosaccharide 2 ratio was applied. which is expected to be more pronounced at lower temperatures. leading to a low average chain length of the dioxanone polymer (7. On the basis of our experimental data and in line with literature data. with an enthalpy of polymerization of approximately ‐15 kJ/ mol .
and dioxanone/ 2 mol ratio) is absent.). 2.7. as expected for a typical ring opening polymerisation . 60 Temperature: 100oC.) for all experiments are scattered randomly between 2. 2.2.. to reduce the number of OiPr endgroups in the product may be the removal of isopropyl alcohol formed in step 40 . reaction time and temperature on the X n exp of the products is shown in Figure 22.214.171.124. time: 16 hr 50 Temperature: 60oC. A clear trend between R gp and the process conditions (temperature.) as a function of the dioxanone : 2 ratio at 90% and 100% monomer conversion is also provided. A possible strategy to increase the Rgp values i. Effects of process conditions on the average chain length ( X n exp ) and end group distribution The effects of the p‐dioxanone: 2 ratio.2. and 2. the X n theo (eq. in line with the theoretical predictions. Average Chain Length ( X n exp ) of the product as a function of the dioxanone : 2 mol ratio The end group distribution was determined using NMR and is expressed in terms of R gp (eq. In Figure 2.7. The X n exp increases linearly with respect to the dioxanone : 2 ratio. time: 1.3.4 and 3. The R gp values (Table 2. time: 16 hr Temperature: 60oC.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. time.e.2.Chapter 2 2.3. Most experimental points are scattered along these lines.6.
The R2 value for the model is 0.3‐6.) where t and T are respectively the polymerization time and temperature.49 + 0. The low P‐value clearly indicates that the model is statistically significant.2.974). 2.703 F 173. indicating that the model describes the experimental data reasonably well. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 1 of the polymerisation process (eq.543 exp as a Model Error Total P‐value 3.459*10‐7 41 . Table 2.854 × (dioxanone : 2 ratio) + 0. 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. A linear model proved adequate to describe the effects of the independent variables on the X n exp : Xn exp = 8.097 × (T ) (2.g. vacuum distillation before adding the dioxanone monomer .4.7.3. time and dioxanone/ 2 ratio) on the X n exp has been performed by multivariable linear regression  on the data given in Table 2. A parity plot of the modeled versus experimental values of the average chain length Xn confirms this statement (Figure 2. 2.6) was too limited to draw sound conclusions. Statistical Data Analysis Quantification of the influence of the experimental factors (temperature.). The analysis of variance for the model is given in Table 2.3.) from the reaction mixture by e.977 (with an adjusted R2 value of 0.12.4.3. The 2/ Al ratio was not included in the model as the experimental range (6.177 × (t ) − 0.8.
17]. 16. 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. the X n exp is negatively influenced by temperature.3.2. Furthermore. the model allows determination of the process variables to obtain a product with the desired degree of polymerization. the model predicts that the X n exp is positively influenced by the polymerization time. with higher ratios leading to higher molecular weights. 2.Chapter 2 60 50 40 Xn model 30 20 10 0 0 10 20 30 Xn exp 40 50 60 Figure 2. Both NMR and MALDI‐TOF measurements indicate that the products are mixtures of polymers and significant 42 . Conclusions The ROP of p‐dioxanone in the presence of a monosaccharide (1.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. 11] and due to the fact that the reaction is an equilibrium polymerization with a slight exothermic effect. Within the experimental ranges. 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). In addition. 2) with Al(OiPr)3 as the catalyst is reported. The isolated yields of the off‐white solid products were between 30 and 96%. This is in line with literature data [5.8.4. with high ratios leading to a higher average chain length. which is in agreement with the available data on ring opening polymerization [15.
P. 1063‐1071. These studies will be reported in the next chapter. 199. Chem. P.M. Narayan. 1998.4‐dioxan‐2‐ one initiated by Al((OPr)‐Pr‐i)(3) in bulk.4‐dioxan‐2‐one. Raquez. Degee. Degee. 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. Zn‐lactate‐catalyzed polymerizations of 1. . Narayan. 2. References . C. D. 153‐166. R. Chu. 42. temperature and monomer: monosaccharide ratio) on the X n exp . 1089‐ 1097. Grubb: Hydrolytic degradation and morphologic study of poly‐p‐dioxanone. Dubois: Some thermodynamic. H.L. Raquez. P. Kricheldorf.C. 8419‐8425. starch). . Damrau: Polylactones.O. Mater. Biomed.M. Lin. Macromol.5. 2000. Dubois: ʺCoordination‐insertionʺ ring‐ opening polymerization of 1. Phys. 1993. . R. 27. kinetic.R.6. 43 . 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. H. Rapid Commun. Macromol.4‐dioxan‐2‐one and controlled synthesis of diblock copolymers with epsilon‐caprolactone. D. J. and mechanistic aspects of the ring‐opening polymerization of 1. P.J. A statistical model has been developed to quantify the effects of process variables (time. 21. 34. 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. Moreover. Res.g. J. Macromolecules 2001.
. 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.
grafting.7 was obtained. 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. polycaprolactone. Keywords: starch. The DS of the polycaprolactone graft is between 0. biodegradable polymers.72. silylation . The poly‐(ε)‐caprolactone side chains consist of 40‐55 monomer units and is a function of the reagent intakes. The silylation reaction was performed using hexamethyl disilazane (HMDS) as the reagent in DMSO at 70°C. 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. the remainder being homopolymers of ε‐caprolactone.21 and 0. Silylated starch with a degree of substitution (DS) between 0.45‐0. ε‐ Caprolactone is grafted to silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 in THF at 50oC.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 grafting efficiency varies between 28 and 58%.
There are two types of biopolymer in starch.6‐glucosidic bonds besides α‐D‐1. The development of green biodegradable polymers for e. amylose (a linear polymer of anhydroglucoses with α‐D‐1. The content of amylose in starches depends on the plant and typically varies between 18‐28%. Starch has also been thermoplasticized with the help of plasticizers such as glycerol and other polyalcohols . and polybutyrolactone). Introduction Worldwide. and this value increases with about 10% per year . 245 million tons of plastics are produced per year. it is water sensitive.4‐glucosidic bonds) and amylopectin (a branched polymer with α‐D‐1. roots) as granules or cells with typical particle sizes between 1‐100 µm. However. the used monomers and the corresponding grafted chains are not easily biodegradable. The polymeric structure of starch consists of repeating anhydroglucose units. Starch is present in the body of many plants (tubers. which are known to degrade with difficulty and cause serious environmental problems . Chapter 3 3. Well known examples are polyesters derived from cyclic lactones (polycaprolactone. Starch is abundantly available. is one of the potential candidates for future biodegradable polymer products. the future generation of packaging materials is highly desirable. A wide variety of synthetic biodegradable polymers have been prepared.1. Various modification strategies have been explored. 5]. resulting 48 . Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decreased rapidly in time and were fully degraded after 4 weeks . 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.g. a natural biopolymer. and the mechanical properties of starch‐based films are generally inferior to those derived from synthetic polymers [4. polyvalerolactone. Polycaprolactone (PCL) is easily degraded by micro‐organisms . in almost all cases. However.4‐ glucosidic bonds). 7]. The amylose‐amylopectin ratio in native as well as modified starches has a strong impact on the product properties. Blends of thermoplastic starch and PCL are not fully miscible. Starch. However. Starch films are known to have good oxygen barrier properties. Global production of starch is 60 million ton per year in 2004 . for instance grafting of monomers (like styrene and methyl methacrylate) to the starch backbone [6. Starch modification is therefore needed to meet the product properties in a number of application areas. as starch is highly hydrophilic. These plastics are mainly synthetic polymers from fossil resources. the product properties are in most cases still not up to standards and blending with other polymers is required .
49 . Acros) and methanol (Labscan) were used as received. the ring opening polymerisation occurs solely in the liquid phase and this is expected to lead to higher GE values.67 gram (8 mmol) Al(OiPr)3 in 50 ml of dry toluene in a glove box. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors in undesirable phase separation . it has been proposed to chemically graft caprolactone onto the hydroxyl groups of starch using ring‐opening polymerisation . This is rationalised by assuming that water competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. This approach has also been applied successfully to graft PCL and polylactide on dextran [19‐20]. Materials and Methods 3. This catalyst is extremely air and water sensitive. This is expected to lead to reduced reaction rates between starch and CL compared to CL homopolymerisation. an alternative method to synthesize poly‐(ε)‐caprolactone grafted starch co‐polymers (starch‐g‐PCL) is reported. thus to a reduction in the GE. Common Ring Opening Polymerization (ROP) catalysts such as tin octoate or aluminium isopropoxide gave low grafting efficiencies (GE. 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]. The highest GE (90%) was achieved using triethylaluminium as catalyst . DMSO (Acros) and toluene (Labscan) were dried overnight over molecular sieves 3 Å (Merck) and stored under a protective nitrogen atmosphere. ε‐Caprolactone monomer (Fluka) was dried over Calcium Hydride (CaH2) for 48 h. Aluminium triisopropoxide [Al(OiPr)3] (Acros) was used without further purification. Dry tetrahydrofuran (THF) and toluene for polymerization experiments were obtained in closed vessels from Aldrich and were used as received.2. a very flammable by‐product. 0‐14%). In this way. distilled under reduced pressure at 100 oC and stored under a protective nitrogen atmosphere.2. Hydrochloric acid (HCl) 1 N was prepared from Titrisol concentrated hydrochloric acid solution (Merck) and distilled (Milipore) water. therefore difficult to handle and releases ethane. To increase the miscibility of starch and polycaprolactone. Another possible cause for the low grafting efficiencies is the heterogenous nature of the reaction. Hexamethyldisilazane (HMDS. during the reaction.1. leading to a liquid‐solid system. thus leading to the formation of PCL homopolymers rather that starch‐g‐PCL . Materials Corn starch (Sigma) was dried at high vacuum (~1 mbar) at 100 oC for one day before use. 3. In this chapter. All available data indicate that the presence of water reduces the GE. A stock solution was prepared by dissolving 1.
After 2 and 4 h reaction time.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. For each experiment. After removal of the solvents under reduced pressure (0.5 ml) and a predetermined amount of the stock solution of Al(OiPr)3 in toluene were added. acetate‐CH3). After 24 h. 3‐6 ppm (m. 0. Traces of DMSO trapped in the product were removed by dissolving the product in a small amount of toluene and re‐precipitation in methanol. DS = 0. silyl‐CH3). The white‐to‐transparent solid products were stored under vacuum in a desiccator at 6‐8 oC. DS = 0. Typical example of the starch silylation procedure The procedure for corn starch silylation was adapted from that published for dextran [19‐23]. Subsequently.12 (m. 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. Sample SN‐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. broad peaks. 1 3.2.12 (m. 50 oC): δ 0.2. starch). 80 oC). toluene (40 ml) was added to solubilize the precipitated (partially silylated) starch. The samples were characterized by 1H‐NMR. The reaction was carried out at 70 oC. 50 oC): δ 0. The pre‐determined amount of HMDS (typically 24 ml. This procedure was repeated three times.111 mol) was added to the gelatinized mixture to initiate the silylation reaction. To this homogenous solution.2. Silylated Starch (1.60): H‐NMR (CDCl3. after peracetylation. a predetermined amount of ε‐caprolactone monomer (molar ratio of monomer to 50 .6 mL) at 50 oC (1‐2 h). 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. A mol ratio of ‐OH groups to catalyst of 10:1 was used.2. silyl‐CH3). the solvents were removed from the silylated starch product under reduced pressure (~ 20 mbar) at 70 oC. before peracetylation. broad peaks.2. Chapter 3 3. another portion of toluene (20 ml) was added. 3. After 6 h. 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.1 mbar.1. Sample SN‐3.60): 1 H‐NMR (CDCl3.2. starch).7‐2.5 (m. 40 oC) until constant weight. Silylated Starch (1. THF (4. 3‐6 ppm (m. the silylated starch (1) product was dried in a vacuum oven (~5 mbar.
Typically. β and δ‐PCL). 3. ‐CH.3‐4.98 (t. Silylated Starch‐g‐PCL (2.21): H‐NMR (DMSO d‐6. Starch‐g‐PCL (3. 1.5. Subsequently. The silylated starch‐g‐PCL (2) product was precipitated from the solution by the addition of heptane (about 250 ml) at ‐18 o C.5‐ 3.37 (t. The yield was measured gravimetrically and is based on the weight of the product and the total weight of reactants charged to the reactor.54 (m. 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. 1. 60 oC): δ 1. The solid precipitate was filtered and dried under vacuum (~ 5 mbar) at 40 oC for 48 h. 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.) was >99 %.16 (d. 1.34): H‐NMR (DMSO. silyl‐CH3).16 (d.2. filtrated. 1. 4. and vacuum dried at 40 oC for 24 h. DSsilylation = 0. the desilylated starch‐g‐polycaprolactone product (3) was precipitated using heptane. 3.31 (m. 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. iPr).37 (t. γ‐ PCL).3‐4. 3. 1 3. 2.88 ppm (m. 3. α‐PCL).5‐3. broad peaks. The total isolated yield at this condition (see Table 3. iPr). 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. Peracetylation of the remaining OH groups of modified starch is a well established procedure to improve characterisation of the products by NMR . ε‐PCL).54 (m.68. iPr). 3.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until it was fully dissolved (typically 3 h). 25]. 1.2. Deprotection product of Sample SN1CL2. DSPCL=0.2.2. γ‐PCL). 1 3.5. and 5‐5. 4.75. 4.2.31 (m. 60 oC): δ 0.3. The samples were characterized by 1H‐NMR. 2. starch). starch). α‐PCL). ‐CH.4 (m.88 ppm (m. ε’‐PCL). β and δ‐ PCL). broad peaks. The peracetylation procedure applied in this study was adapted from the literature [24. ‐CH3. the peracetylating reagents 51 . ε‐PCL). Sample SN1CL1. ‐CH3 .25 (t.68. 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). 1 (0. ε’‐PCL). DSPCL=0. 3. DSsilylation = 0. The product was collected as a white solid and characterised by 1H‐NMR.4.4 (m. After stirring for 2 h. 4.75.25 (t. iPr). and 5‐5.98 (t.
). ACH 3 silyl / 9 1 7 A−0. When assuming that all ε‐CL monomer is converted. 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. the DSPCL may be calculated using eq. This leads to the following equation: ACL = ACH 2 −ε + ACH 2 −ε ' ACH 2 −ε ' = A3.6−0. 3. Chapter 3 (DMAP.1. The product was precipitated by the addition of methanol and washed several times with methanol. 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. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.3.2. The ACL is calculated from 1H‐NMR spectra by comparing the peak area of protons attached to ε‐carbon atoms in a repeating CL unit with that of the characteristic ε’ protons of the last CL unit in a PCL chain  (see Figure 3.2.6 ppm DSsilylation = × = × 3 Astarch proton / 7 27 A3−5.3.1. The peracetylation reaction was conducted for 7 h at 50 oC.3−3. 3.2. In this calculation.4 ppm + 1 (3. 3. It was finally dried overnight in a vacuum oven at 70 oC and 5 mbar until constant weight.2 ppm A3. it is assumed that the average length of the grafted chain is equal to the chain length of the homopolymer.8− 4.1. Analytical Methods 3.3.) where Ax‐y stands for the 1H‐NMR peak area in the range δ x‐y ppm.2.4.) 52 .) 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. 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 ppm (3. The assumption of high conversions (>95%) was correct for all experiments (see Table 3.3.2. DSsilylation may be calculated using 1H‐NMR spectra of the products after peracetylation using eq. 1 3.
A3.8− 4 .CL/OH ratio ⎥ ⎣ ⎦ (3. T stands for absolute temperature.8− 4 .3−3.( A3.9 ⎡ 0.4 + A3.4 / 2 ⎟ ⎠ ⎝ ⎝ where Ax‐y stands for the peak area in the range δ x‐y ppm. 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. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors DS PCL = 0. This leads to the following equation: GE = ACH 2 −ε ' / 2 − ACH −iso ACH 2 −ε ' / 2 × 100 % (3. The values of ∆Hv and Vm were obtained from the SciFinder Scholar database (American Chemical Society. The Hildebrand solubility parameter of HMDS and DMSO were calculated using the following equation : ⎛ ∆H v − RT ⎞ ⎟ δ =⎜ ⎟ ⎜ Vm ⎠ ⎝ 1/ 2 where ∆Hv stands for heat of vaporization.4 − A4 .3. the focus of this chapter will be on the first two steps of the procedure.4.5.) 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).2 ) ⎤ 1 × ⎢ (3 − DSsilylation )⎥ ⎢ ε . 2007) 3.5.3−3.9 ⎟ × 100 % ⎜ ⎜ ⎟ ACH 2 −ε ' / 2 ⎠ A3. 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. Although all steps have been investigated. 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). and Vm stands for molar volume.3.) ⎛ ⎞ ⎞ ⎛ A A = ⎜1 − CH −iso ⎟ × 100 % = ⎜1 − 4.8− 4. 53 .3−3.
5.). However. The silylation procedure was adapted from that previously reported for dextran [19‐23]. a homogeneous reaction mixture was maintained throughout the reaction.5. The 1H‐ NMR peaks from the AHG unit after peracetylation were indeed considerably sharper and allowed more accurate DS calculations (Figure 3.a. 24 h).) Instead of using DMSO as solvent.1. the starch peaks are very broad and this feature hampers accurate determination of the DSsilylation. The 1H‐ NMR spectrum of silylated starch (Figure 3.b.) shows the presence of silyl groups at about δ 0 ppm. Using standard conditions (HMDS: AHG molar ratio of 3. Chapter 3 3.1. It is defined as the average number of silyl substituted OH groups on the anhydroglucose (AHG) unit of starch.3. C O H 2OR H 1 + NH3 2 OH H3C [R = H or Si(CH3)3] Starch Hexamethyldisilazane Silylated Starch (1) (3. 3. 54 .1. Synthesis of Silylated Starch The silylation of corn starch was performed with HMDS as the silylating agent (eq. The silylated products were characterised by NMR. In this way. 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 product with a DS of 0.60 was obtained. 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. 70 oC. The degree of substitution of the silyl groups (DSsilylation) was determined by 1H‐ NMR. mixtures of toluene and DMSO were applied to avoid precipitation of the silylated starch during the reaction.). as suggested by Einfeldt et al . An additional peracetylation procedure to substitute the free hydroxyl groups with acetate groups was performed to improve the quality of the NMR spectra.a.).
Typical 1H‐NMR Spectra of Silylated Starch (Sample SN3.5) on the product DS was investigated by performing experiments with a constant starch and a variable HMDS intake.60. and Figure 3. peracetylated The effect of the HMDS to starch molar ratio (1. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Figure 3.1.5‐4. in CDCl3. The results are given in Table 3.2.1. not peracetylated (b). 50 oC): (a). 55 . DSsilylation=0.
5:1.4 1.45 0. The experimentally observed lowering at higher HMDS intakes is likely due to a decrease in the polarity of the reaction mixture. mmol) 12 (56) 18 (83) 24 (111) 36 (167) HMDS: AHG mol ratio 1. Surprisingly.6 DSsilylation 0. a positive effect of higher reagent intake on the reaction rate and thus the DS is expected. Chapter 3 0. At a mol ratio of HMDS to AHG of 4.2.5 2 2. For all experiments. DS of the silylated products at different HMDS: AHG ratios (DMSO.46 Experiment SN1 SN2 SN3 SN4 a Experiments were performed in DMSO at 70°C.5 2.5 Figure 3.55 0. solubility parameter = 11. a fixed starch intake of 6 g (37 mmol AHG) starch was applied. Its presence will reduce the polarity of the reaction medium (DMSO.68 0.65 0. The silylation reaction 56 . Effect of HMDS: AHG mol ratio on DS of the silylation product a HMDS intake (ml.25 cal1/2cm‐3/2) due to the presence of the bulky apolar SiMe3 groups.36 cal1/2cm‐3/2) considerably.5 0.25 3 4. the DSsilylation decreases for higher intakes of HMDS.5 3 3.67 0.5 Product DS (DSsilylation) 0. the volumetric HMDS intake is about half of the DMSO intake.60 0. 70 oC) Table 3.1. For reactions with an order higher than zero. HMDS is a rather apolar compound (Hildebrand solubility parameter of 6.5 HMDS : AHG Ratio [mol/ mol] 4 4.
3.1 and 3. Similar reductions in the reaction rates when working at higher reactant intakes were observed for the esterification of starch with vinyl laurate and stearate . liquid ammonia.6. the silylation of starch in pure DMSO using HMDS as the reagent has not been reported to date. The products are soluble in DMSO as well as in less‐polar solvent such as chloroform and THF. 29]. Petzold et al  reported that silylation of starch with trimethylsilylchloride (TMSCl) in pyridine yielded trimethyl‐silyl substituted starch with DS values between 0.) 57 . 3.6‐glucosidic for dextran). Silylation of starch with HMDS in formamide.3‐2.6. DMA/LiCl. The DSsilylation (0.25‐5.2. sample SN‐1) in THF at 50 oC for 24 h using Al(OiPr)3 as catalyst.45‐0.7‐3. and DMSO/pyridine mixtures yielded silylated starch with DS values ranging between 0.68. 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. After precipitation with heptane and vacuum drying. pyridine. A schematic representation of the reaction is provided in eq.0 [18. The use of HMDS to silylate dextran (Mw=6000‐40000) in DMSO (HMDS to OH molar ratio of 0. 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) (3.7) of the products is in the range of those published for other starch silylation systems.0 . white solid products with isolated yield > 96% were obtained.0) gave silylated dextran with DS values between 1.3. O O H2C O R H H OR H O O H OR H O Catalyst: Al(OiPr)3 Toluene. DMF.2. 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. To the best of our knowledge. 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.
Further process optimization studies allowed the synthesis of products with less than 42% of homopolymers (vide infra).9 and 1. This is clearly indicated by resonances of the iPr end‐group of the PCL homopolymer at δ 4.8 and 5. Resonances from the starch peaks are observable as small. A typical spectrum is shown in Figure 3.4‐3. The peaks from the polycaprolactone units are clearly present in the range of δ 1.2‐4 ppm and imply that the ring‐opening polymerisation reaction of CL indeed occurred. Coding of the peaks is given in Figure 3.2 ppm.3. 26]. Typical 1H‐NMR spectrum of a silylated starch‐g‐PCL sample.3. not all of the caprolactone is grafted to starch and large amounts of PCL homopolymers (72%) were formed. Apparently. DSsilylation =0. as is shown in Figure 3. Chapter 3 The products were characterized using 1H‐NMR analysis in DMSO‐d‐6 as the solvent. The homopolymers are formed by direct polymerization of ε‐CL initiated on isoproproxide moieties attached to the Al catalyst. However.0‐5.68. the exchange reaction between Al(OiPr)3 and the OH groups of silylated starch is not quantitative under the conditions applied in this study. broad peaks in the region δ 3. Figure 3.21) in DMSO‐d6 at 60 oC.4.4. The formation of CL homopolymers for this type of reactions has been observed before [15. 58 . DSPCL=0.4 ppm. (Sample SN1CL1.
Exchange Reaction: iPr O Al O iPr iPr O + H2O iPr O iPr O Al OH + iPr OH 2.4.7. 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. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors 1. residual water may also initiate the polymerisation reaction. However.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. This leads to the formation of carboxylic end groups (see eq. 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. peaks arising from a ‐COOH unit could not be detected in 13C‐ NMR spectra (δ 175‐180 ppm). 3.) 59 .). Product formation for the reaction between silylated starch and ε‐ caprolactone Besides the ‐OH group of starch. 1.
34 0. This implies that the ε‐CL conversion is essentially quantitative in all cases.).5.5 Avg. This implies that grafting of caprolactone to starch also occurs to a significant extent.2 Total Yieldb [%] >99 >99 >99 99 96.5.3. the intensity of the resonance ε’ was considerably higher that that of the ‐CH‐ iPr peak.68) in THF at 50 oC with Al(OiPr)3 as the catalyst (1 mol Al(OiPr)3 per 10 mol‐starch‐OH groups).3 ppm) should be twice the intensity of the ‐CH‐ proton of the isopropoxide end group (δ 4. In the case of only homopolymerisation. Experiment SN1CL1 SN1CL2 SN1CL3 SN1CL4 SN1CL5 a. The yield of the products was measured gravimetrically and varies between 96. b.5 and 100%. and reaches 58% for a ε‐CL to starch–OH ratio of 29.2. 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. Chain Length [mon. 60 . units] 40 43 44 49 54 DSPCL 0.47 0.0 15. The results are given in Table 3.5 29. (SN1CL1) is 28%. Table 3. the intensity of the peak from the ‐CH2‐ end group of the homopolymer (ε’ at δ 3.2.21 0. Five experiments were performed to study the effect of different ε‐CL to silylated starch ratio. Overview of results for the grafting of ε‐CL on silylated starch a ε‐CL/ OH [mol/ mol] 13. Chapter 3 The ratio of homopolymerisation versus grafting on starch may be obtained by comparing the integrals of selected peaks in 1H‐NMR spectra. and Figure 3.9 ppm).2.9 22. The GE increases with higher ε‐CL intakes.72 Grafting Efficiency [%] 28 43 48 55 58 All reactions were performed using the same intake of SN1 silylated starch (DS=0. in all samples.58 0.0 18. However. The grafting efficiency (GE) for the sample given in Figure 3. Determined gravimetrically and defined as the total weight of the isolated product divided by the total intake of reactants (silylated starch and CL). 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)
4. the highest value was obtained with a ε‐CL‐AHG ratio of 29.2. 3. Conclusions The successful synthesis of poly‐(ε)‐caprolactone grafted corn starch co‐ polymers using a three step approach is reported. This was achieved by making the starch more hydrophobic by partial substitution of the OH groups by trimethylsilyl groups. Chapter 3 3.72.986 cal mol‐1 K‐1 : temperature [K] : R T Vm molar volume [cm3/ mol] 64 . 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.68) was obtained using a DMSO/toluene mixture as the solvent and HMDS as the silylating agent.21‐0. the percentage of PCL grafted to starch compared to the total amount (grafted and homopolymer) of polymerized CL [%] : gas constant.46‐0. Considerable amounts of ε‐CL hompolymers with isopropyl end‐groups were also formed. 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. 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. indicating that homogeneous reaction conditions are favorable for the grafting reaction. depending on the ε‐CL‐starch ratio. The key feature is the use of a homogeneous reaction mixture for the ROP of starch with ε‐CL. The DS of the PCL chains was between 0. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] A x − y ppm ACL DS DSsilylation : DS of silyl group substituents [‐] DSPCL GE : DS of PCL chain substituents [‐] : Grafting Efficiency. The grafting efficiency varied between 28‐58%. The products may have interesting applications as compatibilizers for starch‐polymer blends.5. The ROP with ε‐CL was performed using Al(OiPr)3 as catalyst in THF as the solvent. Silylated starch with a low‐medium DS (0. 1.
Eng. 2008.starch. Goldberg: Review of the biodegradability and utility of poly(caprolactone). 11. Brussels. 1977. K. Yavuz.6. report of Food Biopack Project (Ed.Sci. Wang: Properties of starch blends with biodegradable polymers. Website of International Starch Institute. demand and recovery for 2006 in Europe. Mater. . H. J.dk/isi/stat/rawmaterial. . 2000.A. A. . Starch‐Starke 2000. C. Sci. 1273‐1335. Babac: Preparation and biodegradation starch/polycaprolactone films. Polym. 107‐113. Weber). W. G. Starch‐Starke 2004. 43. Rev. . A. E. Wang. D. Bagley. A. References . Belgium. 1996. of .M. J. 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. . . 2003. M. C. Denmark. Aly.. 1995. 65 .K. The compelling facts about plastics: An analysis of plastics production. Burr. 1251‐1253. the Royal Veterinary and Agricultural University.F. 407‐412. Chandra. 3.B.. J. Aarhus. E. Fanta. 51. Krochta. L. Beliakova.M. Macromol.R. Chiellini. 305‐313. Environ. C. Sci. 52.A.C. Frederiksberg C.C. C. Prog. Sci. 2003. 23. J. Polym. Linke: Water vapor sorption determination of starch based porous packaging materials. Plastics Europe. 38. Albertsson: Biodegradable polymers and environmental interaction. 8. R. Polym. Part C: Polym. . . Department of Dairy and Food Science. 1998. 1998. J. De Mulder‐Johnston: Edible and biodegradable polymer films: Challenges and opportunities. Y. Eng. Russell: Graft copolymers of polysaccharides with thermoplastic polymers ‐ new type of filled plastic. 61‐67. accessed on August 01. Biobased packaging materials for the food industry. 385–409 . R. Yang. R.html. 311‐316. R. 2008. S. 61‐74. Environmen. status and perspectives. Polym. Food Technol. Abdel‐Mohdy: Grafting of poly(methacrylic acid) on starch and poly(vinyl alcohol). X. Polym. . Degrad. . F. Rustgi: Biodegradable polymers. Solaro: Biodegradable polymeric materials. Association of Plastics Manufacturers. 53‐57. Karlsson. 56. Doane. 1997. Adv. 17. Tsiapouris.A. Denmark: http://www.
Einfeldt. Monomers Polym. Macromol. R. Moro. selective syntheses. Polymer 1999. 10. Six: Partial or total silylation of dextran with hexamethyldisilazane. 40. 41. Krishnan. Dellacherie. Einfeldt. A. Nouvel. 251‐269. Polymer 2002. J. Duquesne. Hamaide. J. . 4157‐4167. D. Petzold. P. A. D. . 163. and subsequent reactions. high value applications. D. Dellacherie. D. 2001. D. . I. 1443‐1450. 6713‐ 6721. E. Fringant: Properties of thermoplastics blends: starch‐polycaprolactone. 4.L. E. Dole. 42. Ydens. Polymer 2000. D. Dubois. C. Averous. Klemm. Einfeldt: Structure design of polysaccharides: novel concepts. 73. Stein. Biomacromolecules 2001. L. . regioselectivity. Macromolecules 2000. Polym. Degee. Stab. Rutot. Degee. 341‐347. E. W. Polym. Sci. T. . Cellulose 2003. I. Petzold. Klemm. J. Six: Silylation reaction of dextran: Effect of experimental conditions on silylation yield. Heublein: Silylation of cellulose and starch ‐ selectivity. 2577‐2588. Koschella. P. Klemm: Regioselective functionalization of starch: Synthesis and 1H NMR characterization of 6‐O‐ silyl ethers. W. . Nouvel. A. 43. Chem. Dubois. 2. Rutot. K. Macromol.L. I. P.L. Klemm. K. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer 66 . P. Symp. P. structure analysis. H. D. 1. 3091‐3100. P. 35‐47. Gunther: Synthesis and NMR characterization of regiocontrolled starch alkyl ethers. 2004. J. A. Kussler. W. P. B. Gunther. Dubois: Controlled synthesis of poly(epsilon‐caprolactone)‐grafted dextran copolymers as potential environmentally friendly surfactants. K. . Klemm: Preparative and 1H NMR investigation on regioselective silylation of starch dissolved in Dimethyl Sulfoxide. D. . 965‐969. Dellacherie. . Petzold. Biomacromolecules 2003. Six: Controlled synthesis of amphiphilic biodegradable polylactide‐grafted dextran copolymers. 33. Chapter 3 . 2002. K. Stein. Gunther. M. 415‐426. Biosci. L. 1735‐ 1743. L. Ydens. L. Nouvel. and chemical stability of silylated dextrans. Des. Degrad. P. Dubois: Aliphatic polyester‐ based biodegradable materials: new amphiphilic graft copolymers. M. P. J. Part A: Polym. C. Fessi. 5. L. P. Six. M. . 561‐566. Dellacherie. Ydens. Stein. . Pantiru.L. Dubois. 2001. Dubois. Petzold. Degee. E. P. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. 2001. E. C. ..
2000. Biela. . H. 1978. Macromol Chem. . 2001. D. 659‐663. 67 . Damrau: Zn‐lactate catalyzed polymerizations of 1. Lehmann: Selective 6‐O‐acetylation of amylose. 61.R. Rapid Commun. Sugih. 1089‐1097. 1989 . Res. 29. 1399‐1406. F.B. Janssen. A. Picchioni. Junistia. Macromol. Rapid Commun. Duda: Polymerization of ε‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. Horton. 2008. Duda: Living polymerization with reversible chain transfer and reversible deactivation: the case of cyclic esters.4‐ dioxan‐2‐one. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors agents. 199.P. 941‐950. . Macromolecul. Macromolecules 1996. Kricheldorf. T. D. L. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. accepted for publication in Starch‐Starke. Manurung.O. D. presented at ACS Rubber Division meeting.J. .: Solubility parameter concepts – a new look. Mexico City. L. A novel route to functionalised nanoparticles. Phys.L. A. 22. Carbohydr. H. Penczek. 1998.M. 553‐556. J. Kinetics and mechanism. Hertz Jr. 21. S. A.K. R. .
When performing the reactions using Na2HPO4 as the catalyst. vinyl laurate. 13C‐NMR and FT‐IR. K2CO3. Keywords: corn starch. The fatty esters were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. For low vinyl‐ester to starch ratios. and Na‐acetate) in DMSO at 110°C.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. K2CO3. vinyl ester to starch ratio and the type of catalyst. At higher ratios. esterification. the DS for the starch‐laurate compounds is higher than for the corresponding starch‐ stearates. The DS of the products is a function of the carbon number of the fatty acid chain. an increase in the vinyl‐ester concentration leads to higher product DS values. the DS decreases.96). vinyl stearate . and Na‐acetate are superior catalysts with respect to activity compared to Na2HPO4 and products with DS values close to 3 were obtained. presumably due to a reduction of the polarity of the reaction medium. The yellowish products were characterized by 1 H‐.24 ‐ 2.
5]. An attractive field of application for these polymers is the use as packaging materials. For the current petrochemical based products recycling is often neither practical nor economically feasible . However. 7]. Relatively low‐DS (0. The hydrophobicity increases with the degree of acetate substitution (DS.61) products were obtained using this approach. In particular starch is attractive as it is relatively cheap and abundantly available. However. 5]. 5‐6] were introduced using fatty acid chloride reagents. Their 70 . and incompatibility with hydrophobic polymers.34‐0. and particularly acetic acid derivatives (C2) [2‐4]. defined as the moles of substituents per mole of anhydroglucose (AHG) units) . Esterification of starch with low molecular weight fatty acid derivatives is one of the oldest modification technologies to improve starch properties. poor processibility (high viscosity). The major obstacle is the pronounced brittleness of the materials.1. and even fatty acid derivatives (C12‐C18) have been used in the modification reaction [5.Chapter 4 4.7 [1. even after the addition of plasticizers . resulting in products with DS values up to 2. which are relatively expensive and rather corrosive . However. most of the studies performed to date use short chain carboxylic acids (C1‐C4). However. To improve the mechanical properties. An alternative method using methyl and glyceryl laurate esters in the absence of solvent has been recently developed . higher molecular weight carboxylic starch esters (C4‐C6) . The introduction of acetate groups on starch makes the product more hydrophobic. The mechanical properties and hydrophobicity of the products were significantly improved when using these longer chain fatty acid precursors [1. cellulose or proteins are potentially very interesting starting materials for biodegradable packaging materials. high brittleness. the fatty ester substituents [1. 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. more water‐resistant products may be obtained [3‐4]. Natural polymers such as starch. Introduction Green biodegradable polymers derived from natural resources are potentially very interesting substitutes for non‐biodegradable petroleum‐based polymers. Recently Mormann et al  explored the possibility of using vinyl esters and particularly vinylacetate as reagents for the preparation of starch esters. and consequently. Further modification of starch is therefore required to introduce hydrophobicity and to improve mechanical and moisture barrier properties. The first paper on the acetylation of starch was already published in 1865 . the use of native starch for packaging materials is limited due to its low moisture resistance.
The maximum attainable DS of starch acetate in water was below 1 and limited to 0. starch esters with a substantially higher DS value (up to 1. Germany) and acetic anhydride (Merck.1. and tetrahydrofuran (THF) were supplied by Acros (Belgium) and were also used as received. 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. Materials and Methods 4. 4.2. Potassium carbonate (Boom. 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. Germany) were used as received. vinyl laurate (Fluka. 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. Analytical grade vinyl stearate (Aldrich. as recommended by Laignel et al . the Netherlands). not been reported to date. Technical grade dimethyl sulfoxide (DMSO). 4. The synthesis of starch stearate esters using vinyl ester reagents has.01 when using vinyl‐laurate. to the best of our knowledge. sodium acetate (Merck. The products were placed directly on the diamond plate and 50 scans with a resolution of 4 cm‐1 were recorded. The effects of the starch to vinylester ratio on the reaction rates and DS have been explored. Germany) and disodium hydrogenphosphate (Merck. The reactions were either performed in water or in DMSO using a basic catalyst (Na2HPO4). leading to a moisture content of 2 %‐wt (measured gravimetrically). leading to higher reaction rates.2.2. In DMSO. 4‐N. 1 71 .N‐dimethylaminopyridine (DMAP).2. Germany) were used without further purification.6 for starch acetate ester) were obtained. The spectra were recorded at 50 oC. IR spectra were recorded on a Spectrum 2000 FT‐IR Spectrometer (Perkin Elmer). the use of basic catalysts other than Na2HPO4 has been investigated. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). Materials Corn starch (approx. ~1 mbar). Japan). Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. This solvent effect is likely caused by the higher solubility of the vinyl esters in DMSO compared to water. In addition.
3H.0 (C12). broad peaks.3. 3H. 2H.96) H‐NMR (before peracetylation. C4‐11). C2). broad peaks. The samples were characterized by 1H‐ and 13C‐NMR and FT‐IR.9 (t. Typical example of the synthesis of laurate and stearate esters of corn starch Corn starch (0. 1740 (C=O). 68‐74 (broad. 1. 7H.4 (m. After cooling. CDCl3): δ 0.3. C1S‐6S) 1 72 . 28‐32 (C4‐9).3. C3). 172‐174 ppm (C=O. 2. the product was precipitated using methanol (100 ml) and separated from the liquid phase by decantation.5 (m. CDCl3): δ 0. Table 4. 1. C4‐C17). broad peaks. 2H. (1H‐NMR). 3‐6 ppm (m. 61. vinyl laurate or vinyl stearate (3 mol/mol AHG units in starch) and potassium carbonate catalyst (2 %‐wt with respect to starch) were added and the mixture was stirred at 110 oC for 24 h. 720. broad peaks.0 (m.7 (C11).1. broad peaks.8‐2.1 (C11). 1. 12).9 (t. 76‐78 ppm.9 (broad. 2. C1S‐6S).3 (m. 24.1. 7H. 1350 (C‐H bending).52): H‐NMR (before peracetylation. CDCl3): δ 14. 1230 (C‐O stretching). The product was washed twice with methanol (50 and 25 ml. O‐C3S.6 (m.4. 2H. DS = 2. 3H.2. C2). C2S.5 g) was first gelatinized in DMSO (5 ml) at 70 oC for 3 h. 1.2. C1S‐6S). 1455 (CH2). Starch‐stearate (Sample 19. resulting in the formation of a homogenous transparent solution. 31. 3‐6 ppm (m. 2. 1020 (C‐O stretching). Finally. The atom numbering scheme is given in Figure 4.3 (m. broad peaks.4 (broad. C18). C4‐11). CDCl3): δ 0..1 (m. C3). 16H.9 (C3). 1 H‐NMR (after peracetylation.1. approximately 5 mbar) for 24 h until constant weight. broad peaks. 1410 (C‐H bending). 1.. 2H. 760. (FT‐IR). broad peaks. 5S). C3). C2’). 1150 (C‐O stretching). typical spectra are given in Figure 4. Subsequently.Chapter 4 4. the product was dried in a vacuum oven (70 oC. 2H. 3H.3 (m. Methods 4.2. 1 C‐NMR (before peracetylation.1.. 2H.5 (m. 95. 1110 (C‐O stretching). 22. 2850 (C‐H stretching). 16H. 34. 13 FT‐IR (cm‐1): 2920 (C‐H stretching). C6S). and O‐C6S). respectively). 28H. C1S). 1370 (C‐H bending). Starch‐laurate (Sample 17. 1295. Figure 4. overlap with CDCl3 (C4S). 3S. (13C‐NMR) and Figure 4. 7H. C12). 1. 3‐6 ppm (m. attached to O‐C2S.9 (t. Table 4.9 (C10). 935 (C‐O stretching).5 (m. DS = 2. C2). 1.
1.9 (t. O‐C3S. C2).5 (m. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants H‐NMR (after peracetylation.4 (broad. C1S‐6S) 1 C‐NMR (before peracetylation.4 (m. 865. 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.8‐2. CDCl3): δ 14. 1740 (C=O). 34. 2H. 32. 68‐74 (broad. 61. 1. 2H. 25. Numbering scheme for carbon atoms of products 4. 172‐174 ppm (C=O. 3H.3. 3‐6 ppm (m. 1370 (C‐H bending). 3S. C18). C1S). 760.7 (C17). C2’). 3H. 950 (C‐O stretching). attached to O‐C2S.0 (C18).7 (C4S). 1295. 75. C3). C4‐C17). 1350 (C‐H bending). broad peaks.3 (m. 1100 (C‐O stretching). 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. 1150 (C‐O stretching). 5S). 95.0 (C3). 1. CDCl3): δ 0. broad peaks.2. 22. 1020 (C‐O stretching).0 (C16). 720. 2. 2850 (C‐H stretching). 1. A peracetylation reaction to substitute all of the remaining 73 . C2S.2. 1410 (C‐H bending). 7H. 28H.6 (m. and O‐ C6S) 13 FT‐IR (cm‐1): 2920 (C‐H stretching).2 (C2).5 (broad. 1455 (CH2). 26‐32 (C4‐15). C6S).
4.) AC 2 = 2 × AH − fattyester The AC2 values of both the laurate and stearate side chains were calculated from the peak intensity in the range δ 0.8 ppm (protons attached to the fatty acid carbons C3‐C12 for laurate and C3‐C18 for stearate) using eq.6 7 (4.).) 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 1.6 − AC 2 3 (4.: 74 .) 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 3.6 ppm range) and a correction has to be made (eq 4. Typically.2. The DS of the fatty acid esters was calculated from the DS of the products after peracetylation (eq 4.3.3. The peracetylation reaction was conducted for 7 h at 50 oC. and 4.2.Chapter 4 hydroxyl groups with acetate groups was applied to obtain reliable DS data.6−5. It was finally dried overnight in a vacuum oven at 70oC and 5 mbar until constant weight.: DSacetate = AH − acetate AH − starch A3..4. and 4.7. AH − acetate = A1.6. The peracetylation procedure by Einfeldt et al  was applied. the peractylating reagents (DMAP.2.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until the starch was fully dissolved (typically 3 h).5.6 ppm).6‐5. Determination of the Degree of Substitution (DS) The DS of the product was calculated using 1H‐NMR spectra of the products after peracetylation. Subsequently.5‐1. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added. DS fatty ester = 3 − DS acetate (4.5. 4. 4.) (126.96.36.199− 2.8‐2. The product was precipitated by the addition of methanol and washed several times with methanol.) AH − starch = (4.3.1. the starch ester (0.1. The procedure was described earlier by Elomaa et al  and the relevant equations are shown in equation 4.).
the esterified starch started to separate from the medium in the form of a gel. the brownish gel was precipitated with methanol and the product was collected after vacuum drying in the form of a transparent. 4.) (4. A schematic representation of the esterification reaction of starch with the vinyl esters is provided in Scheme 188.8.131.52.8 = (laurate) 21 21 AC 3−C18 A0.8 = (stearate) 33 33 (4.. 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. The reaction was performed in two discrete steps. 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. 75 .5−1. the vinylester and the catalyst were added and the reaction mixture was heated to 110°C. Results and Discussion 4. The products of these exploratory reactions are insoluble in water and DMSO. Initially. After 24 h. After 2‐3 h.‐4.1. Ax‐y stands for the peak area in the range δ x‐y ppm.7. the starch was gelatinised in DMSO at 70°C for 3 h to make the hydroxyl groups of starch more accessible for reaction.3.7.1. Subsequently.1). 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.) In eq. 4. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants AH − fattyester = AH − fattyester = AC 3−C12 A0.1. light yellow solid. but swell in organic solvents such as toluene and THF.5−1.
For atom numbering scheme: see Figure 4.52 (Sample 17.1. even at higher temperatures (50 oC). 4.1. 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. Medium DS starch laurate and starch stearate (1 < DS < 2) dissolve poorly in DMSO‐d6 and CDCl3. DS = 2.3. A reaction with vinylstearate at similar conditions resulted in a stearate starch ester with a DS of 2. (b) starch laurate. Product Characterisation 4.52 was obtained. (c) peracetylated starch laurate. 76 .96. Table 4. Table 4.Chapter 4 The DS of the products was determined by using NMR (vide infra). DS = 2.). and 4. a product DS of 2.52 (Sample 17.) in CDCl3 at 50oC. Higher DS products have a higher solubility in CDCl3 and good quality 1H‐ and 13C‐NMR spectra could be obtained (Figure 184.108.40.206.2.3.) in CDCl3 at 50oC.1. Typical 1H‐NMR spectrum of: (a) native starch in DMSO d‐6 at 60oC.1. 4. Figure. When using a vinyl‐laurate : AHG molar ratio of 3 and K2CO3 as the catalyst.2.
3 ppm.c. DS = 2.3.96 (Sample 19) in CDCl3 at 50oC. 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. 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. Clearly visible are the peaks arising from starch and the aliphatic hydrogen atoms of the fatty acid chain (δ 0. (c) starch stearate. The resonances arising from the anhydroglucose unit of starch are broadened. NMR spectra of the peracetylated products are considerably improved in terms of peak resolution and allow a more reliable calculation of the DS. The proton signals of the acetate methyl group. Two of the carbon resonances (1S and 4S) are considerably shifted compared to native starch. in DMSO d‐6 at 60oC. required for DS determinations.8‐2. Figure 4. Typical 13C‐NMR spectra of the products are given in Figure 4. The same 77 . Typical 13C‐NMR spectra of: (a) native starch. For atom numbering scheme: see Figure 4.52 (Sample 17) in CDCl3 at 50oC.5 ppm). The 1H‐NMR spectrum of a typical peracetylated starch laurate is shown in Figure 4. 10‐11]. DS = 2. (b) starch laurate.1. The starch peaks (δ 3‐5.2.3. Clearly visible are the carbon resonances of the fatty ester chains (δ 10‐35 ppm) and C atom of the ester group (δ 170‐175 ppm).2. are together with the CH2 groups of the acid chain adjacent to the ester moiety in the range δ 1.8‐2.5 ppm) are broad and overlapping .
4.) show characteristic bands of the carbonyl group of the fatty esters in the 1750‐1700 cm‐1 region. FT‐IR measurements The FT‐IR spectra of starch laurate and starch stearate are shown in Figure 4.1. Characteristic peaks of the polysaccharide backbone are visible in the 1250‐900 cm‐1 region (C‐O stretching) .2.4.) is also included. the C‐H stretching vibrations of the alkyl groups of the fatty ester chain are clearly present at 2920 and 2850 cm‐1.). In addition.4. and 4. Table 4.3. 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. 4. Sample 19.Chapter 4 phenomenon was observed by Dicke for starch acetate . a spectrum of native starch (Figure 4.4.4.b.96.c. and 4. Table 4.1. c.c.52. native starch -OH C-H 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Figure 4. For comparison.b. Sample 17.a. starch laurate C-H C=O C-O C-O a.4.) and native starch FT‐IR spectra of both starch laurate and starch stearate (Figure 4. The near absence of remaining hydroxyl 78 . FT‐IR Spectra of starch laurate (DS= 2.4. starch stearate C-H C=O C-O Absorbance b.2. starch stearate (DS= 2.
In addition.24 2. The results are shown in Table 4. Systematic Studies The effect of important process variables like the vinyl‐ester to AHG ratio.91 0.90 0.3.05 0.01 0.52 2. in line with the NMR data.99 1.13 1. Overview of the esterification of starch using vinyl‐esters and basic catalysts a Catalyst Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 K2CO3 CH3COONa K2CO3 CH3COONa Vinyl Ester: AHG ratio [mol/ mol] 2 3 6 2 3 6 2 3 4 6 2 3 4 6 3 3 3 3 Amount of Toluene added [ml] ‐ ‐ ‐ 5 5 5 ‐ ‐ ‐ ‐ 5 5 5 5 ‐ ‐ ‐ ‐ DS 1. 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.90 1.44 Exp.2.1.08 1.96 2.07 0. type of catalyst and the effect of the addition of co‐solvents on the product DS was studied in more detail. 79 .54 2.1. Most of the experiments (14) were performed using Na2HPO4 as the catalyst.23 0.57 0.60 1.68 0. 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). Table 4. 4.
The DS values are increasing with higher vinyl‐ ester/AHG molar ratio until a certain maximum. The results are presented in Figure 4. Higher concentrations of vinyl esters are expected to lead to higher esterification reaction rates. This behaviour is likely the result of two opposing effects. The highest DS value was 1.23 for vinyl‐laurate at an intermediate vinyl‐ester‐AHG ratio of 3. At a ratio of 1:6. The DS of the products is a clear function of the vinyl laurate and stearate ester intakes (Figure 4. 1. 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. A further increase leads to a reduction in the DS.).5 0 0 1 2 3 4 Vinyl Ester: AHG Starch Ratio [mol/mol] 5 6 Figure 4. DS of the product as a function of the type of vinyl ester and the vinyl ester‐starch ratio (24 h reaction time. This is likely due to a reduction of the polarity of the reaction medium.3. 2 wt% catalyst intake on starch).5. A further increase in the vinyl ester intake leads to a reduction in the DS.1. 110 oC.Chapter 4 4. this positive effect dominates the reaction rate and the DS of the products will therefore increase at higher vinyl ester intakes.3. 7‐10). At low to medium vinyl ester/AHG ratios (0‐3).5. This reduced polarity is expected to lead to ○ : starch‐laurate □ : starch stearate (lines for illustrative purposes only) 80 . the ester intake is equal on a weight basis to the DMSO intake.5 Degree of Substitution (DS) 1 0.5.
).5. This effect is particularly evident at higher vinyl ester/ AHG ratios (>3) (Figure 4. 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. Thus. K2CO3 and Na‐ acetate were tested. 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. the starch laurate esters display higher DS values than the starch stearates. 4. et al  for the silylation of starch. 4. 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. For starch 81 . with reductions in polarity leading to lower reaction rates. The latter factors appears to have a strong effect on reaction rates (vide supra). 11‐14 in Table 4. Here. and illustrated in Figure 4. When using Na2HPO4 as the catalyst. 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 . it also results in a dilution of the reaction mixture and a reduction in the polarity.3. These findings were rationalised by assuming that the co‐solvents increase the solubility of the silylated products.e.220.127.116.11. i.3. with high carbon numbers leading to a reduction in the DS. the DS of the product is also a function of the chain length of the fatty acid.).3.1. Aburto.1. A similar trend in reactivity pattern was observed in this study and the DS decreased from 1.). Effect of the addition of toluene as a co‐solvent A number of reactions were performed using a co‐solvent.7 for lauroyl chloride to 0. A similar procedure was proposed by Nouvel. The results are given in Table 4. Although toluene may positively effect the reaction by (partly) re‐dissolving starch ester precipitates. 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. In this case. the addition of co‐solvents (toluene/ THF) led to an increase in the DS.2.8 for stearoyl chloride. It is clear that Na‐acetate and K2CO3 are considerably more active than Na2HPO4 and products with a significantly higher DS were obtained. Catalysts screening A number of alternative basic catalysts for Na2HPO4. These negative effects dominate the reaction performance at higher vinyl ester/AHG ratios and lead to a reduction in the DS values. leading to enhanced reactivity. The addition of toluene for the esterification of starch with vinyl‐esters surprisingly did not lead to improved DS values.
function of the type of catalyst. the statement that the DS for the laurate esters is always higher than for the stearate esters is not generally valid and a. and Figure 4.Chapter 4 laurate esterification.5 0 Laurate Vinyl-Ester Stearate Figure 4. The products were synthesised in DMSO using vinyl‐esters in the presence of basic catalysts (Na2HPO4. in line with the findings for NaH2PO4. Comparison of DS values with different catalysts (Vinylester: AHG molar ratio = 3:1.6.).44). 3 2. The DS of the products is a clear function of the chain length of the fatty ester and the type of catalyst. the DS of the product is also tunable by proper catalyst selection.96) compared to Na‐acetate (DS=2. when using K2CO3 as the catalyst.6. Apparently.5 Degree of Substitution 2 1.5 1 0. The DS of the laurate ester when using Na‐acetate is higher than for the stearate ester (Table 4. 110oC. Conclusions A study on the synthesis of corn starch fatty acid esters with high DS values is reported. K2CO3 gave products with a significantly higher DS (2. Thus. the two catalysts are equally effective and products with a DS of about 2. and Na‐acetate). The yellow products were characterized by 1H‐ and 13C‐NMR. 82 . However. the DS for the laurate ester is lower than the stearate ester. For starch stearate.5 were obtained. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4.o.1. DMSO). catalyst amount = 2%‐w. gray: Na2HPO4 black: K2CO3 white: Na‐acetate 4.4. and FTIR and confirm the presence of chemically bound fatty acid chains. K2CO3.
and biodegradability of fatty‐acid esters of amylose and starch. Ind. Prinos. Y. 1‐11. 2003. Bikiaris. . S. X.W. Xu. 74. The DS of the products may also be tuned with the vinyl ester/AHG molar ratio. Bengtsson. Dzenis. 1999. D. Sci. J. J. 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. Aburto. Sci. Borredon. 34.4 could be obtained for both laurate and stearate esters. Appl. the DS of the product increases at higher vinyl ester intakes. 1440‐1451.6. Polym. Merrill: Properties of Fatty‐Acid Esters of Starch. Pacsu: Starch Studies: Preparation and Properties of Starch Triesters. . Chem. E. presumably due to a reduction of the polarity of the reaction medium. J. Mullen and E. A. C. M. At low vinyl ester/ AHG ratio. Polym. J. J. Thiebaud. J. Sci. 1647‐1656. A. Alric. J. 1997. W. Y. Polym. I. Thiebaud. and P. . E. 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. 2005. Higher ratios led to a reduction in the DS. S. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. 58. Panayiotou: Synthesis. 1995.5. Prinos. thermal characteristics and biodegradability of extruded starch acetate foams. 54. Sagar. E. Carbohydr. J. Polym. Crops Prod. characterization. Koch. A maximum was observed at a vinyl ester: AHG ratio between 2 and 4. Appl. Borredon. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants With these catalysts. 1208‐1217. D. 705‐721. Alric. M. . Gatenholm: Surface octanoylation of high‐ amylose potato starch films. References . D. Important product properties will be described in the next chapter. C. Ind. products with a DS > 2. K. 361‐368. Appl. 83 . . Aburto. 4. 65. 21. Bikiaris. 1942. I. Hanna: Water solubility. Eng.
L. M. M. Asplund. Dubois. Mater. Six: Partial or total silylation of dextran with hexamethyldisilazane. Borredon: Free‐solvent Synthesis and Properties of Higher Fatty Esters of Starch – Part 2. 84 . Carbohydr. 261‐267. 43. 341‐347. Peltonen. 11. H‐1 NMR and TGA/IR. Klemm: Preparative and H‐1 NMR investigation on regioselective silylation of starch dissolved in dimethyl sulfoxide. Med. Bliard. 57. I. 1735‐1743. Nouvel. . and E. G. T. C. J. Kussler. Senocq. Gunther. C. Reis: Thermal properties of thermoplastic starch/synthetic polymer blends with potential biomedical applicability. J. Res. Cellulose 2004. A. Koniarova. Soininen. P. . G. 1997. Carbohydr. Macromol. E. Alric. Starch‐Starke 1999. Mater. 2004. J. Mouysset‐Baziard. . 251‐260. L. M. R. B. D. . Mormann. 56. Dicke: A straight way to regioselectively functionalized polysaccharide esters. R. S. L. H. R. W. Sci. Aburto. 145‐152. 57. I. S. 2003. Urtti: Determination of the degree of substitution of acetylated starch by hydrolysis. P. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. J. Starch‐Starke 2005. 1. D. I. Polymer 2002. Mano. 2001. Ydens. 255‐263. J. Starch‐Starke 2004. Stein. Hyvarinen.Chapter 4 . Petzold. Al‐Higari: Acylation of starch with vinyl acetate in water. Elomaa. .. W. 118‐121. J. . . Hamaili. E. Laignel. Laatikainen. 298. F. Degee. K. P. M. 51. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. 14(2) 127‐135. Biosci. Polym. Massiot. Dellacherie. A. . Einfeldt. Aburto. 302‐307. Alric.F.
45‐1.39) are totally amorphous whereas the low‐DS ones (DS= 1. K2CO3.5 MPa. trans‐esterification.96). catalyst intake.7‐3. High‐DS products (DS= 2. 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.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. The basicity of the catalyst and the reaction temperature have the highest impact on the product DS. elongation at break of 3‐26%. Keywords: starch esters. and Na‐acetate). The results were adequately modeled using a non‐linear multivariable regression model (R2≥0. The thermal stability of the esterified products is higher than that of native starch.26‐2. vinyl stearate . Mechanical tests show that the products have tensile strength (stress at break) between 2. The reactions were carried out in DMSO using basic catalysts (Na2HPO4. vinyl laurate. and modulus of elasticity of 46‐113 MPa.75) are still partially crystalline. The thermal and mechanical properties of some representative product samples were determined. The effect of the process variables (vinyl ester to starch ratio.
The high DS esters showed interesting properties. however. However. high brittleness. polyethylene  and polycaprolactone  and chemical modification by esterification with organic acids [11‐19]. and food service‐ware . most of the plastic waste is not biodegradable. 0. 3‐4]. The products were biodegradable. The global starch production was estimated at 60 million tons in 2004 . The use of native starch as a building block for novel biodegradable polymers. et al [15‐19] synthesized longer chain fatty esters (C8‐C18) of potato starch and corn amylose using fatty acid chlorides and pyridine. Examples are agricultural mulch . and incompatibility with hydrophobic polymers . The use of vinylesters has also been explored [20‐22].1. Unfortunately. and this causes serious environmental problems [1. Typical examples are blending with polyvinyl alcohol . Introduction Plastics are very useful materials and are used in large amounts (245 million ton per annum) [1‐2]. the research activities were mainly limited to the use of vinylacetate. Chemical or physical modification is required for successful applications as biodegradable polymers. and particularly on the synthesis of starch acetate (C2) [11‐12]. Most early studies mainly focused on the use of short‐chain carboxylic acid (C1‐C4). However. packaging material. 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. Alternative routes to avoid the use of acid‐chloride/pyridine combinations have been developed. however. Two examples using higher vinyl esters (vinyl laurate) were reported by Mormann. the mechanical properties were still not satisfactorily. Esterification is one of the oldest methods used to improve starch properties . 86 .65) . Thiebaud. The limited use is mainly because of a number of unfavourable properties of starch such as low moisture resistance. The largest application area of plastics is the use as packaging material (37%) . the physicochemical properties of the products were not mentioned. The mechanical properties of products with longer ester chains (C4 and C6) products and plasticized acetate esters were considerably improved compared to native starch. et al  and Aburto. Starch is an attractive feedstock for the synthesis of novel biodegradable plastics.Chapter 5 5. The development of novel biodegradable plastic materials capable of decomposing when given an appropriate environment and time is of utmost importance . Sagar and Merill  studied the synthesis and properties of C4‐C6 esters of high amylose starch. The products were hydrophobic and the mechanical properties were considerably improved compared to native starch. Mullen and Pacsu [11. is rather limited to date. et al . It is cheap and abundantly available . 12] studied the synthesis and properties of of C1‐C6 esters of starch.
Japan). Germany) were used without further purification. Technical grade dimethyl sulfoxide (DMSO). as recommended by Laignel et al .1. TGA measurements were performed using a Perkin Elmer TGA 7 Thermogravimetric Analyzer. sodium acetate. 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 . each sample was cooled to ‐50°C and then heated again to 200 oC (heating rate 10 oC/min). DSC analyses were performed on a TA Instruments DSC 2920. Germany). The experiments were modeled using non‐linear multivariable regression. It was dried for 48 hour at 105 oC under vacuum (~1 mbar) to reduce the moisture content below 2 % before use. and disodium hydrogenphosphate dodecahydrate (both Merck.2. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). 5. Materials Corn starch (approx. vinyl ester to anhydroglucose ratio. After a first heating run from room temperature up to 200 oC to delete the thermal history of the material. Germany). Analytical grade methanol. vinyl laurate (Fluka. Long fatty esters of corn starch with a broad range in degree of substitution (DS= 0. The samples were heated to 700 oC in a nitrogen atmosphere with a heating rate of 10°C/min.2. The spectra were recorded at 50 oC. and toluene were obtained from Labscan (Ireland). The samples (about 10 mg) were placed in sealed aluminum cells. All these chemicals were used as received. This chapter describes systematic studies using Design of Experiments (DOE) to gain insights in the effect of process variables (temperature. and acetic anhydride (Merck. catalyst type and intake) on the DS of the products. pyridine.96) were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts in DMSO. Tc. Materials and Methods 5. 5. Analytical grade vinyl stearate (Aldrich. the thermal and mechanical properties of representative examples of the highly hydrophobic materials are described and discussed.N‐ dimethylaminopyridine (DMAP). and tetrahydrofuran (THF) were supplied by Acros (Belgium). the Netherlands). In addition. 4‐N.2. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. The thermal properties (Tm. and Tg) of each sample were determined from the spectra related to the cooling run and the second heating one. The catalysts for the esterification reaction were analytical grade potassium carbonate (Boom. T‐Bones samples (with thickness of 2 mm) for determination of the tensile properties were prepared using a melt press 1 87 .2.24 ‐ 2.
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 18.104.22.168. 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. 22.214.171.124. 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 2.00 2.00 5.00 3.00 5.00 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 4.00 2.40 1.00 5.52 2.00 5.17 2.00 5.47 1.52 2.32 1.44 2.27 1.00 4.00 5.00 3.00 3.54 2.96 0.00 3.00 5.24 2.00 5.00 5.00 3. 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] 3.00 5.00 3.00 2.28 2.79 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 .00 2.00 3.59 2.00 3.00 5.70 2.00 Catalyst Temperature [%‐w] [˚C] 2.00 3.00 5.00 5.50 2.00 3.00 5.00 5.00 2.00 2.00 2.00 5.00 3.00 3.42 2.54 2.00 2.82 2.00 5.00 3.00 2.00 4.00 2.35 0.84 2.00 3.00 5.00 5.00 5.00 5.00 5.00 2.00 5.00 3.00 2.93 0.12 0.00 5.78 2.41 1.87 1.00 5.00 5.12 2.00 2.00 3.00 5. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.00 3.88 2.11 2.44 2.27 0.67 2.00 5.00 5.07 0.53 1.00 5.00 4.2.00 3.72 2.50 0.00 5.00 5.00 3.50 2.50 2.00 5.94 2.00 2.23 2.09 0.00 3.00 5.62 0.56 1.27 1.00 5.50 2.48 0.00 5.00 3.00 5.00 3.75 1.00 5.46 1.00 3.
3.3. Values for the coefficients of the DS model for starch laurate and stearate Coefficient Starch Laurate 9.3.0027 ‐0.0249 ‐0.64 2.0009 0.Chapter 5 45 46 47 48 49 50 51 52 53 54 a.41 mol ratio of the vinyl ester to AHG units of starch 5.).6939 0.90 2.0003 0.96 2.0857 ‐0. The type of catalyst was quantified using the basicity constant (pKb) in water [26‐27].0016 0.1748 0.1947 β0 β1 β2 β3 β4 β5 β6 β7 β8 β9 β10 β11 β12 β13 92 . stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 4. 5.).00 5.98 2.0327 ‐2.1979 Starch Stearate 17. The experimental data for both starch laurate and starch stearate esterification are best described with a quadratic model including interaction terms (eq.0066 0.00 5.0010 ‐0.) Table 5.50 2.0017 0.00 3.00 3.1082 ‐0.24 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 5.00 3.1.3.0732 ‐0.00 5.0008 ‐0.The modelled values for βij for the two esters are given in Table 5.55 2.00 5.0120 0.0011 0.00 3.0470 0.1.00 5.0008 0.1104 ‐0.2183 0.5650 0.1568 ‐2.59 2.00 3.0006 ‐0.0001 0.00 4. 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 (5.64 2.00 2.00 3.00 5.50 95 80 80 110 110 80 80 110 110 95 1.00 5.3.21 2.00 2.00 5.0023 0.00 2.
967 0. respectively and it may be concluded that the models describe the experimental data well. Analysis of variance for DS models for starch esterification Starch Laurate Model Error Total SS 16.).2903 P‐value 1.967 (starch stearate). The very low P‐ values indicate that the models are statistically significant.1. Also in this case the reasonable R2 values (0.5 1 1. Finally.8167 21.5 DS actual 2 2.805‐0.5 3 0 0 0.5 2.5 2 DS predicted DS predicted 2 1.5 1. which indicate  that all significant variables are included in the model.7699 DF 13 14 27 MS 1.939 0. Parity plots of the DS models for starch laurate and stearate Moreover.96 x 10‐8 R2 values R2 R2adjusted R2press 0.9531 0.4. 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 0 0 0.875 Starch Stearate Model Error Total SS 20.875) indicate that the model correctly predict the 93 .5 0. and confirm this statement.4.970 0.4986 16.2569 0. Table 5. starch laurate b.5 3 a. we also performed a PRESS analysis  (see corresponding R2 values in Table 5.4.026 x 10‐7 R2 values R2 R2adjusted R2press 0.0583 F 27.970 (starch laurate) and 0. the adjusted‐R2 values are very close to the R2 ones. starch stearate Figure 5.3399 0.5 1 1 0.0356 F 35. which represents an “internal” validation method for the model.5 1 1. The R2 values for the models are 0.805 3 3 2.5 DS actual 2 2.629 P‐value 1.945 0.1.8386 DF 13 14 27 MS 1. The parity plots of both models are given in Figure 5.6118 0.
This is indeed the case when comparing the performance of K2CO3 with CH3COONa.00) than for K2CO3 (2.8) is intermediate between that of the other two catalysts. the base strengths of the catalysts in DMSO are not known. The resulting anion will react with the vinyl ester to from the product.). To evaluate the effect of each variable on the product DS.66) than CH3COONa (pKb 9. the performance of Na2HPO4 does not follow this trend. Of all the variables studied.Chapter 5 products DS as function of the process variables within the range of experimental variables. The highest product DS values were obtained using K2CO3. It is assumed that the first step in the reaction sequence is activation of the starch OH groups by deprotonation by a base [21. Unfortunately.25) and this leads to higher product DS values for K2CO3.2. The pKb for the latter (6. whereas catalyst performance is considerable lower. Another explanation may be related to the regio‐chemistry of the reaction.82). K2CO3 is a stronger base (pKb 3. 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). This was not the case for alkaline catalysts. The product DS for Na2HPO4 was considerably lower (1. However. 22]. such as carbonate or acetate salts leading to higher DS esters.18) and CH3COONa (1. On the basis of the sequence. Catalyst performance of CH3COONa was slightly less than for K2CO3 whereas the lowest DS products were produced when using Na2HPO4 as the catalyst.2. A similar trend was observed by Dicke  for the acetylation of Hylon VII starch using vinylacetate. 94 . the type of catalyst has the largest effect on the product DS for both the laurate and stearate esters (Figure 5. 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. 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. These results may be rationalised by considering the role of the catalyst in the modification reaction. it can be rationalised that the rate of the reaction (and thus the product DS) will be higher when using a stronger base. for both starch laurate and starch stearate are given in Figure 5.
). starch stearate Figure 5. the catalyst intake and reaction temperature also affect the product DS (Figure 5. although to a lesser extent.2. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch a. 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. As expected and in 95 .2. starch laurate b.
DSstearate = 2. heating Exothermic Tm = 21 oC cooling Endothermic Tc = 12 oC -20 -10 0 10 Temperature ( C) o 20 30 40 Figure 5. the vinyl ester to AHG ratio has the smallest effect on the product DS. 5.Chapter 5 line with studies on starch acetates . . . higher reaction temperatures and catalyst intakes lead to higher product DS values.39) were determined. Product Properties The thermal and mechanical product properties for two starch‐laurate and stearate samples with medium (DSlaurate = 126.96.36.199.45 (.5. Representative DSC spectra for the starch stearate are given in Figure 5.26. and the results for all samples are summarised in Table 5.) DS ( 2. DSC analysis of starch stearate esters ( ) DS 1.3. DSstearate = 1.75. Of all variables studied.39) 96 .45) and high DS values (DSlaurate = 2.
5.d.d. in the present work. 17]. 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.4. For instance a Tm of 32 oC was reported for a potato starch stearate ester with a DS of 1. The melting point of the potato starch stearate was close to that of model compounds for the stearate side chains (methyl stearate. 40‐42 oC.= not detectable The thermal behavior of the products is a strong function of the DS.d. close to the value obtained in our study for the medium DS starch stearate (21 °C).). Thermal transitions at relatively high temperature (melting. these are the first examples of completely amorphous starch esters. octadecane.d.a 22 n. esterification of starch with vinyl laurate and stearate reduces the crystallinity of the starch considerably and leads to a lowering of the transition temperatures. 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. which is known to contain crystalline areas. To the best of our knowledge. or glass transition above 0 oC) were even absent for the high DS products.d. At moderate DS values the products still display a melting temperature (at 20‐30 oC) and a crystallization temperature (at 10‐25 oC). TGA analysis was performed to study the thermal degradation behavior of the starch laurate and stearate samples (Figure 5. In any case. Changes in the transition temperatures of native starch by esterification have been reported in the literature [13‐14. Unfortunately the thermal transitions for the virgin material lie above the degradation temperature and can therefore not be accurately determined . 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. 22]. In particular the behaviour at relatively high temperatures (>0 oC) is further discussed. n. crystallization.45 Tm [˚C] n. implying that these products are fully amorphous.a 12 a. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. These values are reduced considerably compared to virgin corn starch.26 1.75 2.8 . especially in the amylopectin part [15.a 21 Tc [˚C] n. 97 . However.a 27 n. 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).39 1.
TGA analysis of various starch samples.39) The TGA (Figure 5.6.4.): starch stearate (DS 2. .).Chapter 5 100 80 %-weight 60 40 20 0 20 100 200 300 400 o 500 600 700 Temperature ( C) Figure 5.26) (. .4.) curves clearly show starch esterification with either laurate or stearate results in products with enhanced thermal stability. The mechanical properties (stress at break. The difference in thermal stability between the laurate and stearate sample is limited. Native corn starch degrades at lower onset temperatures (250‐300 oC) than the corresponding esters (300‐350 oC). Similar trends were reported for amylose octanoate‐ stearate‐laurate esters  and potato starch octanoate‐ laurate esters . ( ): native starch ( ): starch laurate (DS 2. 98 . elongation at break and elasticity modulus) of the products were determined (Table 5.
9 46.e. The corresponding yield strain (roughly 0.5 ± 1.1 Elongation at break [%] 26.75 2. 17].5. 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.7 ± 1.26 1.0 112. For starch laurate.2 8. This product is more rigid compared to the product with medium DS value (higher tensile strength and elasticity modulus. The values of the yield strain.8 Starch Ester Starch laurate Starch laurate Starch stearate Starch stearate All materials generally show tensile strengths between 2. Starch stearate with a relatively low DS (1. Mechanical properties of the starch esters DS 2.1 82. starch stearate at relatively high DS values (2. An inverse behaviour is. elongation at break between 3‐26%. and lower elongation at break). determined from the stress‐strain curves as the stress at which the material ceases to be linearly elastic) and the modulus.2 ± 0.7‐3. but higher elongation at break.7 ± 3. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. independently of the DS values. 99 .3 21.39) does not show any plastic behavior but only an elastic one.3 3.45) displays a very similar behavior with respect to the laurate samples (i. an increase in product DS results in lower tensile strengths and modulus of elasticity. On the other hand. a clear plastic behavior with a maximum in the stress‐strain curves.0 3. This indicates that their mechanical behavior is comparable to that of typical engineering polymers .7 ± 4.025) lies still in the typical range of engineering polymers but also close to that typical of composite materials .9 ± 14.3 ± 2. however. calculated as the ratio between the stress at yield (σy. This is in agreement with previous studies regarding the mechanical properties of starch esters with different chain length of the fatty acid moiety [13. A deeper understanding of the mechanical behavior can be gained by investigating the shape of the stress‐strain curves (typical examples in Figure 5.0 3. is roughly 0.1 Modulus [MPa] 59. Both starch laurate samples display. thus indicating a relatively more rigid material with respect to all other samples.05).1 ± 1.7 ± 4.1 ± 15.5 MPa. plastic deformation and yield strain of about 0.39 1. observed for the high DS starch stearate sample.7 ± 14.05 for both laurate samples.45 Stress at break [MPa] 3.).6.0 ± 2. and modulus of elasticity between 46‐113 MPa.2 2.
.7. Stress‐strain curves for starch esters.): starch stearate (DS 1.45) (‐ .26) (.75) ( ): starch laurate (DS 2.). ( ): starch laurate (DS 1. . 100 . 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. ‐): starch stearate (DS 2.5.Chapter 5 6 5 4 stress (MPa) 3 2 1 0 0 5 10 15 20 strain (%) 25 30 35 Figure 5.39).
a. we show that although no crystallization of C18 occurred (see DSC data of the high DS material in Table 5.7 ±4.39) 2. as confirmed by DSC.1 High DS (2.3 10 ±2 Property Low DS (1.7 ±1. 9% for potato starch). 3.75) 3. While for corn starch (DS 2. 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.2 Stress at Break [MPa] Elongation at Break [%] a.0 ±2.7.5. 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 ±0. In agreement with our own data.1 ±1. On the other hand at low DS values comparable stress at break (3.1 ±15.7 ±3. They related this inverse property of starch stearate to crystallization of C18 side chains.0 26.8) 3.). 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.2 MPa for corn starch esters.6 9 ±2 High DS (2. In this respect the difference in stress and elongation at break for the high‐DS starch laurate sample is striking. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. 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.= not available Inspection of the mechanical properties of the starch esters as shown in Table 5.7 ±0.7) 1.5 MPa) is coupled with a low elongation (26 %).a. merely because of the structural property of the C18 side chain with respect to the starch backbone.26) 3.26) relatively high stress at break (3.6 Corn Starch Stearate Ester Low DS (1. Moreover. the high DS material is still more rigid. DS) which mainly determines the thermal and mechanical behaviour.3 Potato Starch Stearate Ester Low DS (1. n.a.7 MPa for potato starch ones) are coupled with significant differences in the elongation (22% for corn starch esters.9 ±0.a.a.2 ±0. n. High DS (2.5 ±1. Here.0 3. 101 .1 21.3 High DS (2.7) 0.2 8. but the interplay between those factors.45) 3.4 1500 ±8. an unexpected behavior was also observed for the mechanical properties of the starch esters from potato starch.7. for potato starch exactly the opposite is observed: relatively low stress (0. 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. confirms that the corn starch esters synthesized in our research are relatively rigid materials as compared to the potato starch esters.
The use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2.5 for starch laurate. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] 102 . catalyst intake. 1. 5. 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.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. The models may be used to determine the appropriate process conditions to obtain a product with a pre‐defined DS.4. including statistical modelling. reaction temperature.4‐3 for starch stearate). on the synthesis of corn starch esters with long chain fatty acids is described. and catalyst basicity) to the DS of the starch ester products were developed. Comparison with literature data indicates that the properties may be (fine) tuned also by the starch source. K2CO3. 0.9 for starch laurate.1‐2. The DS of the products is a strong function of the basicity of the catalyst.3‐1. Nomenclature β : regression coefficients obtained by a multiple regression procedure [various units. 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. Statistically adequate (R2 ≥0. while the use of Na2HPO4 resulted in low‐medium DS products (0. 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.5 for starch stearate).5. Conclusions A systematic study. Thus.07‐1. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. and Na‐acetate). Reaction temperature and catalyst intake also affect the product DS although to a lesser extent than the type of catalyst.
F. H. USA. and Biopolymers. Plastics Europe. demand and recovery for 2006 in Europe. .6. Stevens: Green Plastics. M. New Jersey.J. A.W. Processing. 1986. S. 103 . Bastioli: Global Status of the Production of Biobased Packaging Materials. S. The Royal Veterinary and Agricultural University. 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. 1980. Marsden: The challenge of domestic waste disposal. Hanser Publishers. J. W. 2008 . C. UK. 2002. R. Glasgow. Denmark. Brussels. Eng. Munich. Structure – Properties ‐ Applications. Frederiksberg C. E. Levy). Hernandez. Westhoff. Chem. 27‐29 August 2000 (Ed. 2000. References . Conference Proceedings. Doane: Starch‐based blown films. J. Association of Plastics Manufacturers. . Packaging. in Packaging in the Environment (Ed. 2000. . Dev. Properties. Selke. 592‐595. M. . Denmark. in The Food Biopack Conference: Foodstuffs. An Introduction to the New Science of Biodegradable Plastics. Otey. Hanser Publishers. . Ind. E. 19. Culter: Plastics Packaging. M. 2000. Denmark: http://www. Prod. G.dk/isi/stat/rawmaterial. D. Aarhus. G. 2008. Res. Copenhagen. Princeton University Press. Blackie Academic & Professional. Website of International Starch Institute. Ehrenstein: Polymeric Materials. Belgium. Munich.starch. Applications. Department of Dairy and Food Science. C. Weber). Germany. P. and Regulations. The compelling facts about plastics: An analysis of plastics production.html. Germany. R. accessed on August 01. .
Sci. S. Sci. J.. Pacsu: Starch studies: preparation and properties of starch triesters. Macromol. E. 1996 104 . 28. 1440‐1451. Mormann. W. Mouysset‐Baziard. report of Food Biopack Project (Ed. I. C. 65. 1209‐1217. Appl. M. 1995. Borredon. A. I. J. Prinos. 132‐135. K. 1995. Alric. S. H. 118‐121. Thiebaud. J. Janssen. Ind. Starch‐Starke 1999. J. K. Alric. Frederiksberg C. Senocq. 51. A. S. E. Weber). Borredon. Carbohydr. 51. . C. J. Wang: Properties of starch blends with biodegradable polymers. Polym. Broekroelofs. . . . 58. 74. Starch‐Starke 1999. I. W. Appl. 2003. Rev. Yang. 385– 409. Wang. 381‐384. J. Eng. M. 101‐112. Y. de Graaf. X. J. Starch‐Starke 2004. Panayiotou: Synthesis. Alric. Z. . Hamaili. Aburto. 56. 1943. . Rijksuniversiteit Groningen. Borredon. 1997. C43. F. Aburto. Bikiaris. M. W. J. Thiebaud. R. Alric. D. Prinos. E. C. Bikiaris. . I. Polym. J. J. and biodegradability of fatty‐acid esters of amylose and starch. J. characterization. Beenackers: The kinetics of the acetylation of gelatinised potato starch. G. Prinos. E. C. Aburto. I. E. Appl. Ind. de Graaf: Ph. Polym. J. Thesis. 1997. L. . R. Sci. status and perspectives. the Royal Veterinary and Agricultural University. Starch‐ Starke 2005. E. . E. 1647‐1656. 705‐721. 137‐144. E. 34. Part C: Polym. Merrill: Properties of fatty‐acid esters of starch. Panayiotou: Properties of octanoated starch and its blends with polyethylene. Mullen. Aburto. C. Aburto. Denmark. A. Mullen. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. D. 2000. 145‐152. Al‐Higari: Acylation of starch with vinyl acetate in water. 302‐307. Sci. .. 1999. Biobased packaging materials for the food industry. I.Chapter 5 . J. Alric. 57. A. 1942. J. Department of Dairy and Food Science. 34. E. Chem. Aburto. A. Polym. Carbohydr. D. . Eng. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. J. W. P. G. L. . Thiebaud. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Sagar.D. Bikiaris. B. A. J. . Alric. Chem. Polym. 35. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. Pacsu: Starch studies: possible industrial utilization of starch esters. D.
the Netherlands. 1997. Prentice‐Hall. B. Bliard. J. Materials: Engineering. 2008. USA. Picchioni. Ashby. Junistia. Dicke. 2007. R. 105 . Manurung. Janssen. Massiot. A straight way to regioselectively functionalized polysaccharide esters. A. G. Butterworth‐Heinemann. A. Res. M. K. L. Montgomery: Design and Analysis of Experiments 5th edition. Laignel. . 1986.. . Amsterdam. Cebon. New Jersey. L. H. C. P. R. A. 251‐260. USA. 298. L. Day. accepted for publication in Starch‐Starke. . R. 11. C. A. Underwood: Quantitative Analysis 5th edition. John Wiley & Sons Inc. B. . New York. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. D. Cellulose 2004. Carbohydr. M. Science. 255‐263. Processing and Design. F. J. M. H. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch . 2001. D. . Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Shercliff. Sugih.
glycidyl methacrylate. 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. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. 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. at relatively high GMA and BPO intakes. the mechanical behavior of the compatibilized blends is in general different from that of pure PCL and of the corresponding uncompatibilized blends. grafting . At relatively low starch intakes. despite the expected differences (favorable to PCL‐g‐GMA) in the in situ formation of the compatibilizers. Keywords: starch‐PCL blend. As a result. This discrepancy could be explained on the basis of the functional groups (GMA or DEM) distribution along the PCL backbone. In particular the elastic modulus for the compatibilized blends is significantly higher than that of uncompatibilized ones. The reactive interfacial agents were tested for their performance in PCL‐starch blends. PCL‐g‐ glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM) is decribed. Both act as compatibilizers for PCL/starch blends by improving the interfacial adhesion between the starch particles and the PCL matrix. diethyl maleate. PCL‐g‐DEM has at least a comparable performance with respect to PCL‐g‐GMA.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. compatibilizer. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %).
it would be actually more accurate to define the block copolymer as an “interfacial agent”. A well‐known biodegradable polyester.e. The latter displays the presence. upon mixing) by using a functionalized PCL. Such structure is rather difficult to achieve by simple copolymerization methods and it is usually prepared in situ (i. to an ideal compatibilizer having both PCL and starch blocks. for starch/PCL blends. When using ungelatinized starch as a component in the blend. However. along the backbone. 2]. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decrease rapidly in time . Blending of starch and synthetic biodegradable polyesters has been widely applied for the synthesis of fully‐biodegradable products. 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. a compatibilizer (i. This leads. as in this study. Unfortunately. Unfortunately. To overcome this issue. Among these. polyesters are considered very promising alternatives . Starch modifications to improve the product properties like enhanced hydrophobicity and mechanical properties were already reported in the early 19th century [1. such as polyethylene (PE) or polystyrene (PS) [3‐6]. 108 . To reduce the tendency for phase separation.e. blends of hydrophilic starch and hydrophobic biodegradable polyesters exhibit phase separation  due to differences in polarity of the building blocks. 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 ). is known to be degraded with ease by microorganisms widely distributed in nature . The ester bonds are susceptible to attack by water and this leads to enhanced biodegradability. 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. an interfacial agent) may be used to improve the interfacial association between the two polymer phases. synthetic biodegradable polymers have been applied. polycaprolactone (PCL). 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. PCL has gained considerable interest for possible applications in the fields of packaging materials and medical applications [10‐11]. these synthetic polymers are poorly or non‐biodegradable. As a consequence. This phenomenon is highly undesirable and limits the application range considerably .1. of polar groups (usually epoxides or anhydrides [13‐18]) able to react with the – OH groups along the starch backbone. Introduction Starch is a cheap and abundantly available natural polymer with very good application perspectives in the area of biodegradable plastics.
~1 mbar) prior to use. 109 . Mw=50000) from Solvay Caprolactone. Belgium. >99%) was obtained from Acros. This low molecular weight PCL grade was used without further purification. Tetrahydrofuran (THF. UK.2. 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. 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. After the PCL was melted (1‐2 minutes). 6. 6. Methods 6. The synthesis and application of PCL‐g‐DEM is. This chapter describes a systematic study on the synthesis of two reactive interfacial agents for starch‐PCL systems. 73% amylopectin and 27% amylase) was obtained from Sigma and high molecular weight PCL (CAPA 6503. Germany and methanol (99. after which the equipment was stopped and the chamber was opened to collect the samples. Glycidyl methacrylate 97% purity (Aldrich). Germany) were used as received. diethyl maleate ≥97% purity (Fluka).8%) from Labscan. The kneader was heated to 130 oC and PCL (CAPA 2304) was added while maintaining a rotational speed of 80 rpm.2.1.2. Corn starch (with approx. Mw=3000 from Solvay Caprolactones. and benzoyl peroxide 75% (Merck. xylene (99.2. The intake of reagents was maximally 75‐80% of the volumetric volume to ensure proper mixing. UK was used for the preparation of the interfacial agents. to the best of our knowledge. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM).2. Materials Polycaprolactone (PCL) CAPA 2304. Intakes for each experiment are given in Table 1 and 2.2. The various compatibilizers have been tested for their performance in PCL‐starch blends. Ireland.1. an absolute novelty of the present research. Materials and Methods 6. systematic studies and subsequent quantification of the functionalization reaction has not been reported to date. 16]. The materials were mixed for another 5 minutes. Compatibilizer synthesis The compatibilizers were prepared in a Brabender Plasticorder PL2000 batch‐ kneader (chamber volume 35 cm3). a solution of BPO in GMA or DEM was added drop by drop over a period of 5 minutes. however. Exploratory studies on the synthesis of PCL‐g‐GMA and its applications for starch/PCL blends have been published [13. The starch was dried for at least 24 h at 110oC under vacuum (approx.
2. The content was blended for 15 minutes.8%. 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). and then filtered. 6. The error on the transition temperature is assumed to be ± 1 oC and 5 % of the calculated values for the corresponding enthalpies. the samples were covered with a paladium/platinum conductive layer of 3 µm thickness. An operation temperature of 170°C and a rotation speed of 80 rpm were applied . Infrared spectra were collected with a FT‐IR apparatus in the ATR mode using a Spectrum 2000 instrument from Perkin Elmer. Subsequently.2. PCL‐g‐DEM (5 g) was dissolved in xylene and refluxed at 150 oC for 2 h.5 h. the chamber was opened and the resulting material was collected. 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 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). 6. created using a Cressington 208 sputter coater.2. Before analysis.Chapter 6 6. The resulting suspension was filtered and precipitated using methanol (450 mL) at 6‐8°C. Methanol (450 mL) was added to the filtrate. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. 5 mbar). stirred for 1. PCL was added to the chamber followed by the addition of the starch and the reactive compatibilizer.2. 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 . 5 mbar).3.3.2. Work‐up of PCL‐g‐GMA products  To remove unreacted GMA monomer and GMA homopolymer. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. Analytical Methods H‐NMR measurements were performed using a 400 MHz Varian AMX Oxford NMR apparatus with CDCl3 (99. Aldrich) as the solvent.4.2. Digital Scanning Calorimetry (DSC) measurements were performed using a Q1000 TA Instruments equipped with a TA Instruments DSC cooling system. Tensile tests were performed using an Instron 4301 1 110 . 6. PCL‐g‐GMA (5 g) was dissolved in 50 ml THF.2.2. and the product was precipitated at 6‐8 oC. Scanning Electron Microscopy (SEM) was performed using a Jeol 6320 F Scanning Electron Microscope.
6. The latter is a measure of the statistical significance of the model. 8 different T‐bones were used. 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) (6.2.) The FD was calculated using 1H‐NMR by comparing the area of protons belonging to the GMA (‐CH< proton at δ 3. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends machine. 6. As a result we were able to obtain a model for the FD of the reaction.0 ppm [13.2. leading to a 10% relative error in the FD values. This procedure is described in detail in the literature  and consists of calculating the sum of squares (SS) for the model and the error. 19]. On the basis of the latter values. 15‐16. For every T‐bone. The validity of the model was determined by performing an analysis of variance (ANOVA. the F‐ value for the model is determined followed by the P‐value.3.4. 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.1.2 ppm) side chains with that of a characteristic proton resonance of the PCL backbone (‐CH2‐ protons at δ 4. Table 3). The T‐bone samples were prepared using a Fontijne Holland TH 400 hot‐ press. 111 . it is possible to calculate the mean square (MS) for the model and the error.1. stress at break (σ) and modulus (E) were measured. For a given sample/blend. strain at break (ε).2 ppm) or DEM (‐CH2‐ protons at δ 4. When the relative degrees of freedom (DF) are known. A 5% relative error in the peak area of the NMR spectra was assumed. 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. 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).
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 6.1.1.).3.1. 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. 112 .Chapter 6 6. Results and Discussions 6.1.
Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends
Figure 6.1. Typical 1H‐NMR spectra of PCL‐g‐GMA (top) and PCL‐g‐DEM (bottom) Peak assignments were based on available data reported for related products [13, 15‐16, 20]. The FD values and the thermal properties of the products are shown in Table 6.1. (PCL‐g‐GMA) and Table 6.2. (PCL‐g‐DEM).
Overview of experiments for the PCL‐g‐GMA compatibilisers a
Intake FD (%) ‐ 5.7 15.2 27.9 3.2 45.2 c 6.3 T cryst. ( C)
(%‐mol) b GMA BPO ‐ 0.6 0.6 0.6 0.3 1.1 0.3
ΔH cryst. (J/ g‐PCL) 75 72 68 67 73 66 71
T melt. ( C)
ΔH melt. (J/ g‐PCL) 79 72 69 68 73 68 67
PCL PCL‐g‐GMA 1 PCL‐g‐GMA 2 PCL‐g‐GMA 3 PCL‐g‐GMA 4 PCL‐g‐GMA 5 PCL‐g‐GMA 6
‐ 12 24 36 6 24 12
27 22 19 29 25 20 28
51 47 47 48 47 46 47
Experiments were carried out at 130 °C. %‐mol with respect to the CL repeating units in the PCL c based on the soluble fraction of the compatibilizer
Table 6.2. Overview of experiments for the PCL‐g‐DEM compatibilisers a
Intake Sample (%‐mol) b DEM PCL PCL‐g‐DEM 1 PCL‐g‐DEM 2 PCL‐g‐DEM 3 PCL‐g‐DEM 4 PCL‐g‐DEM 5 PCL‐g‐DEM 6
FD (%) ‐ 3.5 2.1 7.2 3.6 6.5 0.9
T cryst. ( C)
ΔH cryst. (J/ g‐PCL) 75 70 68 64 61 63 70
T melt. ( C)
ΔH melt. (J/ g‐PCL) 79 72 62 66 60 66 69
BPO ‐ 0.6 0.6 0.6 1.1 1.1 0.3 27 27 30 21 26 25 30 51 44 47 45 42 41 47
‐ 15 30 45 30 60 7.5
Experiments were carried out at 130 °C %‐mol with respect to the CL repeating units in the PCL
One of the PCL‐g‐GMA compatibilizers (PCL‐g‐GMA 5, see Table 6.1.), was only partly soluble in CDCl3, and therefore the FD is based on the soluble fraction of the compatibilizer only. The presence of an insoluble fraction, combined with a relatively broad molecular weight distribution (as shown by GPC, but not shown here for brevity), suggests that cross‐linking occurred during this experiment. In general, the FD of the PCL‐g‐GMA (3.2‐45.2%) products is much higher than those of PCL‐g‐DEM (0.9‐7.2%). This difference may be either due to the difference in mutual solubility of the GMA and DEM in PCL or differences in the molecular
Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends
mechanism of the grafting reaction. The mutual solubility may be expressed in terms of the differences of solubility parameters of PCL and the reagents . The difference between the solubility parameters of GMA and PCL is 0.29 cal1/2cm‐3/2, while it is much higher (6.3 cal1/2cm‐3/2) for DEM and PCL . Thus, GMA is likely better soluble in PCL, leading to higher values for FD of the products, as confirmed by our experiments. However, the higher experimental grafting efficiencies for PCL‐GMA may also be rationalised by considering the reaction mechanism for the preparation of the compatibilisers. GMA molecules may either react directly with a radical at the PCL backbone or with a radical present on an already coupled GMA molecule. The latter leads to longer GMA grafts on a PCL backbone . A simplified representation of the reactivity of GMA is shown in Figure 6.2.
O (CH2)4 CH2 C O + RO
O (CH2)4 CH
+ GMA O (CH2)4 CH H2C H3C C
C O CH2 HC O O
Termination + H or + R or + RO (CH2)4
+ k GMA O CH C O
H2C H3C C
C O CH2 HC O O
Figure 6.2. Simplified scheme of the GMA Grafting Reaction Mechanism  The reactivity of DEM in radical reactions is expected to be different from that of GMA. Previous studies on maleic anhydride (MA), a compound resembling the
3. This trend is independent of the BPO amount and matches with data reported by other groups [13.2. BPO=0.1% mol/mol-CL units 10 5 0 10 20 30 40 50 60 Substrate Intake (mol% on CL units) Figure 6.3. 116 . BPO=1. subsequent reactions of MA to an already grafted MA molecule hardly occur.6% mol/mol-CL units 15 GMA. However. BPO=0. The results are graphically provided in Figure 6. Our experimental findings. 22]. 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 length of a MA graft is always unity whereas longer grafts are possible for GMA.Chapter 6 chemical structure of DEM.).6% mol/mol-CL units DEM. Thus an increase in the GMA intake will provide more monomer available for the growing of the grafted chains leading to higher FD values. showed that MA reacts easily with a radical on the PCL backbone. It is most probably related to the fact that GMA is able to propagate to longer grafted chains when reacted with PCL (Figure 6.3% mol/mol-CL units DEM.1.e.1. i. Effect of GMA and DEM to CL‐units ratio (mol/mol) on the FD of the products (constant PCL intake. Hence.3. higher FD values for GMA than for DEM are in line with this explanation and support the proposed molecular reaction mechanisms. 30 25 20 FD (%) GMA. 6. BPO=0. 130°C) Higher GMA intakes lead to higher FD values.
1. no detectable influence of the BPO amount on the FD is observed (Figure 6. at relatively lower GMA intakes. 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. These results are in line with earlier work [15. doubling the amount of initiator results in considerable higher product FD.).3. 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. constant substrate to PCL ratio) For high GMA to CL ratio (24%‐mol/ mol CL units).1 1. Effect of the BPO intake on the product FD (130°C. This will lead to a higher proportion of PCL radicals by hydrogen abstraction from the polymer backbone and thus to higher FD values.8 0.6 0. DEM=30% mol/mol-CL units 0. although a slight increase in the FD values might be appreciated. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The experimental trend for DEM is different.3 0.4. The use of higher initiator concentrations will result in an increase in the number of formed radicals.9 1 1. for which a levelling off and eventually a decrease of the FD values for relatively higher MA amounts has been observed. However.7 0. there 117 . GMA=12% mol/mol-CL units DEM. 15. 17].2.5 0.2 GMA. GMA=24% mol/mol-CL units GMA.4. Apparently.).4.2 BPO Intake (mol% on CL units) Figure 6. 17]. The FD values are within a rather narrow range.4 0. 50 45 40 35 30 FD (%) 25 20 15 10 5 0 0.
2. 6. If many macroradicals are created in the system (relatively high BPO intakes) at low GMA intake. In addition. 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). The effect is however much less pronounced than for GMA. defined as the difference in solubility parameters between PCl and the substrates.0325nmδ + 0.3.2729nm niδ (6.1. a statistical model has been developed by performing a multivariable linear regression on the data reported in Table 6.6022nm ni − 0.3. 17]. as well as several transfer reactions might be responsible for the observed trend . 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.2. This phenomenon is likely related to the different nature of the radical grafting mechanism of GMA and DEM on PCL as mentioned before. i. However. This leads to the following equation: FD = f (nm . the mutual solubility of DEM in PCL and DMA in PCL was included in the model by using a parameter δ. recombination of (macro) radicals. To quantify synergic effects of monomer and initiator molar intakes on the FD of GMA and DEM on the PCL backbone. 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). the BPO and monomer intakes are considered as independent parameters. and 6.3. 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.5431niδ + 1.) where nm is the molar amount of monomer in the feed. Here.3. gives the best description of the experimental data: FD = −1. The model provided in eq. the possibility of side reactions will become significant. 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. ni the molar amount of initiator in the feed and δ the difference in solubility parameters calculated using group contributions . In particular the occurrence of “cage effects”. ni .Chapter 6 is an optimum ratio between the BPO concentration and available monomer (GMA in this case) on the FD.e.1. δ ) (6. namely an increase in the BPO intake results in higher product FD.8875 + 0. 6.) 118 .
SS 1828 81. Model Error Total Analysis of variance for the FD model provided in eq 6. b. a.3 10 10 20 30 20 0.4 0. The very low P‐value implies that the model is statistically significant.908 10.957) and its closeness to the adjusted R2 (0. respectively.8 0.7 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 ). 1 0. b.2 10 15 20 25 30 35 40 GMA Intakes (mol% on CL units) 45 50 a.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 ).5. Figure 6. Table 6.3.941) also suggests that all important variables have been included in the model. and Figure 6.6.9 60 70 40 50 80 BPO Intake (mol% on CL units) 0.8 0.5.358 1909.358 DF 4 8 12 MS 456. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The analysis of variance data are given in Table 6.. 119 .4 BPO Intake (mol% on CL units) 0. 3D plot.6 0. The R2 value for the model (0. Graphical representation of the FD models for PCL‐g‐GMA and PCL‐g‐DEM are given in Figure 6. Graphical representation of the FD model for PCL‐g‐GMA.17 F 44.5 0.3. This was also confirmed by inspection of the residue distribution by a normal probability plot (not reported here for brevity) .3. Contour plot.6 20 10 20 40 50 30 60 FD (%) 40 FD (%) 30 40 20 0 1 0.2 10 20 GMA Intake (mol% on CL units) 40 30 50 0.
2. For relatively high DEM intakes (> 30 % mol/mol).3 5 6 0. which is expected to lead to a Tc reduction.4 0.6 0. the crystallization temperature and enthalpy are not a clear function of the FD values.1. a. and a favored nucleation 1 2 50 120 . Remarkable is once more the different trends for the two substrates.Chapter 6 1 0.4 BPO Intake (mol% on CL units) 0. For all samples. Contour plot. b. Such changes in the thermal properties compared to pure PCL may be caused by the introduction of grafts on the PCL chains. b. radical initiator and PCL intakes). et al  working with PCL‐g‐GMA. Similar observations have also been made by Kim. This induces irregularities and is expected to result in a lowering of the Tc and the Tm. Thermal properties of the compatibilizers The thermal properties (Table 6. 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.) of the compatibilizers were determined by DSC.7 3 4 6 FD (%) 4 0. From a more practical point of view.1. and 6.4. While for GMA an increase in BPO or monomer intakes invariably leds to higher FD values. 3D plot. 7 6 3 3 4 Figure 6. for DEM a clear transition is observed. the melting temperature and the relative enthalpy decrease with respect to pure PCL.2 10 20 DEM Intake (mol% on CL units) 40 30 50 6 0. 6.8 0.8 0. The random behavior of the Tc as function of the FD values is probably the result of two concurring effects: the presence of irregularities.3. a higher BPO intake leads to a reduction of the FD values while an opposite trend is observed at lower DEM intakes.9 4 5 8 BPO Intake (mol% on CL units) 0. 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.6 2 FD (%) 5 2 0 1 0.5 0.6.2 10 15 20 25 30 35 40 DEM Intake (mol% on CL units) 45 7 a. Graphical representation of the FD model for PCL‐g‐DEM. Furthermore. although both properties are significantly lower than those of pure PCL.
8 372. 6.5 ε (%) 640.4.1 13.0 341. 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.6 5.3 10.3 15.2 337.0 332. 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.5 489.9 E (MPa) 270.0 430.6 305.4 230.2 321. 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.8 329.5 425.2 342.9 368.2 9.1 11.4.7 %.0 401.9 379. 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) 16.6%. PCL‐g‐GMA has an FD of 9. An overview of all prepared blends together with their thermal and mechanical (tensile tests) properties is given in Table 6.6 168.3 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 1.9 424.5 7.4 431.2 343.6 11.8 371.6 386.8 384. A total of 12 blends were prepared: a series of binary ones (not containing any compatibilizers) constituting our reference points.2.1 10.0 9.7 10. 121 .9 11. 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). Table 6.5 357. 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.3 380.3.
43 MPa and 640. 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 .3.7. respectively from 16.2 MPa up to 341. The thermal and mechanical properties indicate no or very little interaction of the starch particles with the PCL ○ : ∆Hc □ : ∆Hm 122 .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 . 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. 75 70 65 60 55 50 45 40 35 ∆ H (J/ g-PCL) 0 5 10 15 20 Starch content (%) 25 30 Figure 6. 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 MPa and 341. Melting and crystallization enthalpy as function of the starch content for binary blends with PCL.7.2. The simultaneous increase in the modulus (from 270. 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 6.1.5 % for pure PCL to 7.8 MPa) clearly indicates that the rigidity of the blends increases at higher starch contents.Chapter 6 6.
Morphology of starch/PCL binary blends.). The latter trend is also valid for the corresponding enthalpies: i.) 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.).9. The presence of these voids was also observed for sago starch/PCL blends [25. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends matrix.3. This is visually confirmed by the morphology of the prepared blends (Figure 6. a. the starch particles are simply inserted into voids of the PCL matrix. However. in order to fully understand the role of PCL‐g‐DEM.8. In this respect the thermal properties remain substantially unchanged with the exception of the Tc. was used as compatibilizer for the preparation of ternary blends with starch and PCL. (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. for which a 6 oC drop is observed when using PCL‐g‐DEM. a decrease with respect to pure PCL is observed and then a substantial invariance as function of the PCL‐g‐DEM content. whose synthesis is described for the first time in this work. The thermal behavior is characterized (Table 6. As consequence one might expect a slight different structure of these ternary blends as compared to the corresponding binary one.4. the corresponding binary blend (S/PCL 20/80) contitutes a better reference point compared to pure PCL.e. 123 . This is confirmed by examination of the blend morphology by SEM (Figure 6. either by heating or in combination with mechanical stress . Ternary blends compatibilized with PCL‐g‐DEM PCL‐g‐DEM. c. b. and the lack of interface adhesion cause void formation.8. Evaporation of the water during blend preparation. Such behavior is consistent with the hypothesis that the starch particles hinder the melting and crystallization processes of the PCL chains. Figure 6. 26] and explained by assuming that the voids are formed by water in the starch. Thus.2. 6.2.
which will hinder the flowability and fibre forming ability of PCL matrix under cold drawing. 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.).). Morphology of ternary blends compatibilized with PCL‐g‐DEM. Figure 6.9. (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. 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. b. As a result of the different morphology also the mechanical behavior display differences with the virgin PCL and the binary blend.9. This can be explained by the improved interfacial adhesion  between PCL and starch.Chapter 6 a.). At 1 %‐wt of PCL‐g‐DEM in the blends the modulus increases with respect to the binary blend (0 %‐wt compatibilizer in Figure 6. resulting in more rigid material with higher modulus. c. The stress and strain at break remain constant as function of the compatibilizer intake while the modulus display an optimum as function of the compatibilizer intake (Figure 6.10.
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). In particular also in the case of PCL‐g‐GMA. 188.8.131.52. However. the crystallization temperature experiences a drop of about 5‐6 oC. 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. Modulus of ternary blends S/PCL/PCL‐g‐DEM as function of PCL‐g‐ DEM amount. 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.2. by taking as reference the corresponding binary blend (S/PCL 20/80). the stress and strain at break are lowered and hardly a function of the PCL‐g‐GMA amount. 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).e. 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. at all PCL‐g‐GMA contents) there is a clear increase of the modulus with respect to the 125 . in all cases (i. the modulus is higher and hardly a function of the compatibilizer amount.
9. This is surprising if one takes into account the different FD values: 9. 126 .8 % for PCL‐g‐GMA against 1. Thus. As a result. 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. and morphologies in Figures 6. partial interfacial adhesion) which render these ternary blends a kind of “intermediate” case between the binary ones and those compatibilized with PCL‐g‐DEM. Figure 6.11. (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. However. This confines all reactive GMA groups in relatively “isolated” spots along the PCL backbone.11. PCL‐g‐DEM is at least as active as compatibilizer (compare modulus values in Table 6. As given schematically in Figure 6. 28. presence of voids.12. Morphology of ternary blends S/PCL/PCL‐g‐GMA. and 6.) than PCL‐g‐GMA for blends containing 20 %‐wt of starch. 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.7 % for PCL‐g‐DEM. b. 29]. c. 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.).Chapter 6 binary blend.4. 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. a. These differences in compatibilizing effect are not yet fully understood. (left) the structure of PCL‐g‐GMA is inhomogeneous at the molecular level with long poly(GMA) branches pending from the PCL backbone.
On the other hand (right of Figure 6. once PCL‐g‐DEM has reacted with one –OH groups on the starch particle. for all starch amounts (Table 6.12.3 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 %) .4.) PCL‐g‐DEM has much less reactive groups (lower FD) than PCL‐g‐GMA. In order to check this we compared blends with the same amount of PCL‐g‐GMA (2 %‐wt) but with different starch intakes (10.12. at higher starch contents more –OH groups would be available for reaction with the GMA groups. Schematic representation of the in situ reaction between functionalized PCL and starch.) for further reaction. Concerning the thermal behavior (Table 6. other groups will presumably remain available (arrows in Figure 6. The latter result (from a modulus of 341. thus attenuating the effect discussed above. 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.8 MPa for S/PCL 30/70 up to 430. but since DEM is preferentially grafted as monomer.12. 20 and 30 %‐wt respectively).4.). The proposed explanation implies however that at relatively higher starch contents (> 20 %‐wt) the segregation of the poly(GMA) chains in “isolated” spot along the PCL backbone would become less important. the distribution of the reactive group along the PCL backbone is more “homogenous”. We can therefore conclude that the efficiency of PCL‐g‐GMA in the 127 . 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.).) and relating the “coverage” of the starch particle surface upon reaction with PCL‐g‐GMA with the compatibilization efficiency. Indeed. 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. This is in agreement with the hypothesis made above (Figure 6.12. when comparing ternary blends (compatibilized with PCL‐g‐ GMA) with the binary ones. Moreover. thus probably ensuring a better coverage of the surface. As consequence.
6. has been performed on the synthesis of two reactive compatibilisers. 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. The PCL‐g‐GMA and PCL‐g‐DEM compatibilizers display lower melting temperatures and melting enthalpies compared to virgin PCL. mutual solubility of the monomer into the molten polymer) on the FD adequately described the experimental results (R2=0. P‐value ≤10‐9).957. Nomenclature E FD : initial modulus [MPa] : functionalization degree.Chapter 6 compatibilization of starch/PCL blends can be significantly improved by changing the blend composition. The most important variable for the final product FD is the interaction between the amount of monomer and initiator used. moles of GMA or DEM present per mole of CL repeating units : : ni nm amount of initiator intake [mol% on CL units] amount of monomer GMA or DEM intake [mol% on CL units] 128 . The reactive compatibilizers were used in blends of starch with PCL. 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. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). 6. firstly reported in this work. can replace PCL‐g‐GMA as compatibilizer at relatively low starch contents offering at the same time the advantage of a less pronounced modification of the polymer backbone and a reduced consumtpion of polar groups to be grafted on PCL. 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. including statistical modeling. Conclusions A systematic study. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends.4. From a more practical point of view it can be concluded that the newly synthesized PCL‐g‐DEM.5. The proposed model to quantify the effects of process variables (monomer and initiator intake.
Albertsson: Biodegradable Polymers and Environmental Interaction. Pacsu: Starch Studies: Preparation and properties of starch triesters.S. Polym.D. M. . E.B. Eur. Davis: High performance LDPE/ thermoplastic starch blend: a sustainable alternative to pure polyethylene. Sci. J. Ramsay. . Polym.Eng. 129 . Albertsson. Effect of amylopectin to amylase ratio in starch.W. Jarowenko: Acetylated starch and miscellanous organic esters. Eng. Ranucci. Kawasaki: Biodegradable films made from low density polyethylene (LDPE). Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – III. 38. 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. J. B. A‐C. Y. 157. 33. 34. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – IV. Mani. . S. M. 227‐242. 1477‐1487. F. 44. . J. Mullen.. Wurzburg). Liu. Eur. . S. 1986. 1998. 1517‐1526. Ioannis. Polym. . Karlsson. Mani. O. Karlsson: Polymers from Renewable Resources. A‐C. CRC Press. Biliaderis. Carbohydr. Inc.6. 34. R. 1997. References . Psomiadou. 1998. 1998. 1209‐1217. E. W. Lindblad. Sci. H. Rodriguez‐Gonzales. M. wheat starch and soluble starch for food packaging applications. USA. 1251‐1253. Polym. 1942. in Modified Starches: Properties and Uses (Ed. A. 34. Thermal and morphological properties.Chem. R. 2002. . 139‐161. 1467‐1475. Polym. Part 2. Boca Raton. Ind.A. N. B. Polymer 2003. C. E.J.G. Adv.
12. A. Laurienzo. 29. J. Painter: Specific Interactions and the Miscibility of Polymer Blends. R. Park: Modification of Aliphatic Polyesters and Their Reactive Blends with Starch. Tang. Polym. 5135‐ 5142. Adv. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. 2001. Stab. . Part A: Polym. 1996. Sci. J. C. 42. 2001. J. . Errico. 305‐313. Technomic Publication. . Khandpur. 61‐67. 1991. 91‐95. Degrad. E. . M. E. Kim. Polymer 2000. K. 41. Kim. Tang: Functionalization of polyesters with maleic anhydride by reactive extrusion.H.S. Mater. C. 127. . J. 1139‐1148. M.M. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. . P. Polym. Kim. Environ. 127–134.. It. 1693‐1702. C. D.W. J. .E. Moro. 8. J. Polymer 2001. Park: Grafting of glycidyl methacrylate onto polycaprolactone: preparation and characterization. Nakayama. 3.H. 1273‐1335. Bhattacharya. Macromolecules 1996. P.Y. R. C. G.M. Passaglia. R. 1995. M. K. C.K.K. J. D. John Wiley & Sons Inc. M.K. 1997. Macosko. 130 . Goldberg: A review of the biodegradability and utility of poly(caprolactone). Marechal. Graf. 37. . Cho. Appl. Fringant: Properties of thermoplastic blends: starch‐polycaprolactone. 2004. C. Dole. Rimedio: Preparation and characterization of compatibilised polycaprolactone/ starch composites. Ruggeri. Mani. Prog. Rustgi: Biodegradable polymers. 1999. Polym. Sci.. Solaro: Biodegradable polymeric materials. 1507‐ 1516. J. 5590‐5598. E. . New York. Raimo. 81. Chiellini.F. P.H. Jung. Sci. J.C.. P. USA. . Cho. M.Chapter 6 . 3875‐3881. 80. Pennsylvania.S. Polym. Averous. J. Polym. Z. . R. Montgomery: Design and Analysis of Experiments 5th edition. A. Polym. Chim. USA. 1998. 4157‐4167. Polym. J. Chem. M. C. Yang. Sci. 35. L. Kim. Coleman. J. Environ. 2003. . Chandra. Chem. L. John. 41. M. T. Marrucci. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. Aglietto: Chemical reactions affecting the free‐radical grafting of diethyl maleate onto ethylene polymers. Gazz. J. 179‐187.K. Avella. M. 23. Polymer 2000. Martuscelli. 1997. Degrad. Inoue: Compatibilizers for melt blending: Premade block copolymers. . P. Part A: Polym.Y. R. . K. Guegan.
. G. R. Mauler: Comparative study of PP‐MA and PP‐GMA as compatibilizing agent on polypropylene/nylon 6 blends. Test. Kammer: On mechanical properties of sago starch/ poly(ε‐caprolactone) composites. Manan. M. S. 2003. . B. 2000.N. Aglietto.M.W.S. K.A. .S. Picchioni: Formation and compatibilizing effect of the grafted copolymer in the reactive blending of 2‐ diethylsuccinate containing polyolefins with poly‐ε‐caprolactam (nylon‐6).B. D.H. Polym. Polym.S. 21. Adv. 131 . Sci. H. U.W. Eur. O. Eng. Sci. Azemi: Effect of starch predrying on the mechanical properties of starch/ poly(ε‐caprolactone) composites. 877‐884.M.V. Polym. . S.N. Ruggeri. Ishiaku. R. Polym. 11‐15. D. 2006. A. 393‐401.Nasirtabrizi: Ring opening reactions of glycidyl methacrylate copolymers to introduce bulky organosilicon side chain substituents. 57. Odusanya. Azemi.Safa.S. 1998. W. Appl.M. Lee. 2002. 40. Technol. 293‐304. Barbosa. Nachtigall. E. 87. 9. Ishiaku. M. M. 273‐281. Pang. J. Tedesco. U. Passaglia. B. U.M.M.A. Ishak: Mechanical properties and enzymic degradation of thermoplastic and granular sago starch filled poly(ε‐ caprolactone). F. Polym. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends . O.D.S. 38.S. Odusanya. J. 2002. Manan. Polym. Ishiaku. K. Bull. . 1298‐1305.
Starch‐based biodegradable materials are considered interesting candidates to replace certain types of conventional plastics. such as thermoplasticization. Starch must therefore be modified before it can be applied as a biodegradable plastic. 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. there is a strong need for renewable alternatives for plastics from fossil resources. Starch is a polymer consisting of anhydroglucose (AHG) units. Its production has increased significantly since 1950. 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. The use of virgin starch for packaging materials is restricted because it cannot be shaped to films with adequate mechanical properties (high percentage elongation. There are several starch modification methods available. it has a high volume to weight ratio and is generally resistant to degradation.4)‐glucosidic bonds.6)‐glucosidic bonds. The content of amylose and amylopectine in starch varies and largely depends on the starch source. we have . To gain insight in the potential of this approach. blending with other materials. containing periodic branches linked with the backbones through α‐D‐(1. plastics may end up in the environment and cause serious environmental problems. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. For instance. 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. Starch is relatively cheap and available from a broad range of plants.o oil and gas) are very attractive materials for a broad range of applications. Starch is also too sensitive to water.Summary Plastics made from fossil resources (a. There are two types of polymers present in starch: amylose and amylopectin. with a rate of almost 10% every year. Examples are the use as packaging and construction materials. while amylopectin is branched polymer. After use. tensile and flexural strength). chemical modification or combinations thereof. 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. In combination with the current high prices for petrochemical products.
even in the presence of plasticizers. The yield of the polymers was a function of the reaction temperature and the reaction time.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose) using Al(OiPr)3 as the catalyst. 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. Esterification of starch with carboxylic acid derivatives is one of the oldest strategies to improve starch properties. The DS of the PCL chains was between 0.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose ratio (at constant Al(OiPr)3 intake). are described in Chapter 2. The synthetic procedure may be divided into three steps: hydrophobisation of starch by the introduction of SiMe3 groups followed by in situ ROP. depending on the ε‐CL to starch ratio. Most of the previous starch esterification studies involved the use of short chain carboxylic acids (C1‐C4).Summary initially performed research on the ROP with a simpler model system involving a simple protected mono‐saccharide instead of starch. with higher temperatures (100°C) leading to lower yields.2. The polymerizations were performed at 60‐100°C. temperature and monomer: monosaccharide ratio) on the average degree of polymerization.46‐0. Poly‐(ε)‐ caprolactone grafted silylated starch co‐polymers with average chain length of 40‐ 55 monomer units (molecular weight of 4500‐6300) were obtained. and subsequent removal of the silyl groups.21‐0. A statistical model has been developed to quantify the effects of process variables (time.68) was obtained. Considerable amounts of ε‐CL homopolymers with isopropyl end‐groups were also formed. The use of higher carboxylic (fatty) acid to esterify starch resulted in products with significantly improved 134 . The grafting efficiency for the desired reaction was 28‐58%. but the products are still very brittle. The silylation reaction was performed using hexamethyl disilazane (HMDS) in DMSO/ toluene mixtures at 70°C. 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. and particularly acetic acid derivatives (C2) have received considerable attention. The results for the model system involving the ROP of p‐dioxanone initiated by hydroxyl groups of a protected monosaccharide (1. The hydrophobicity of starch acetates is higher than virgin starch.3. and off‐white solid products with isolated yields of 30‐96% were obtained. 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. with higher ratios leading to higher molecular weights. 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. The results are described in detail in Chapter 184.108.40.206. Silylated starch with a low to medium DS (0.
the synthesis of two polycaprolactone (PCL)‐based reactive interfacial agents. a systematic study.07‐1. vinyl ester to starch ratio. usually performed using fatty acid chlorides which are relatively expensive and rather corrosive. while the use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2. 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. 1. K2CO3.5 MPa. and the type of catalyst.9 for starch laurate. however. High‐DS products (DS= 2. Statistically adequate (R2≥0. Reaction temperature and catalyst intake also affect the product DS but only to a lesser extent.26‐2.45‐1. The products have a broad range of degree of substitution (DS = 0. reaction temperature. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4.7‐3. The last part of this thesis deals with reactive blending of starch and biodegradable polymers using reactive interfacial agents. In Chapter 4‐5. an increase in the vinyl‐ester concentration led to higher product DS values. 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.39) are totally amorphous whereas the low‐DS ones (DS= 1. PCL‐g‐ 135 . presumably due to a reduction of the polarity of the reaction medium. The thermal and mechanical properties of some representative product samples are also described.96).1‐2. The thermal stability of the esterified products is higher than that of native starch. elongation at break of 3‐26%. catalyst intake. The use of Na2HPO4 resulted in low‐medium DS products (0. an alternative method for higher fatty starch esters by using fatty vinyl esters is reported.4 could be obtained for both laurate and stearate esters. and catalyst basicity) to the DS of the starch ester products were developed. The yellowish products were characterized by 1H‐. and Na‐ acetate).5 for starch laurate. The DS of the products was a clear function of the chain length of the fatty ester. and modulus of elasticity of 46‐113 MPa. The use of methyl and glyceryl esters results in products with only relatively low DS.75) are still partially crystalline. In Chapter 6. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. For low vinyl‐ester to starch ratios. K2CO3. At higher ratios.4‐3 for starch stearate).3‐1. In Chapter 5. The DS of the products is a strong function of the basicity of the catalyst. and Na‐acetate) in DMSO at 110°C. the DS decreased. 0. 13C‐NMR and FT‐IR. products with a DS > 2.24 ‐ 2.5 for starch stearate). With these catalysts.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐ starch mol ratio. Summary mechanical properties and hydrophobicities. The starch esters were prepared by reacting starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. The synthesis of these fatty starch esters is. Mechanical tests show that the products have tensile strength (stress at break) between 2.
At a fixed starch content (20 %‐wt). mutual solubility of the monomer into the molten polymer) on the functionalisation degree (FD) of the GMA and DEM units to the PCL backbone. PCL‐g‐DEM seems to have slightly more efficient compatibilising effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values.957. the mechanical behavior of the compatibilised blends is in general different from that of pure PCL and of the corresponding uncompatibilised blends. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. As the result. 136 . 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. 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). at relatively high GMA and BPO intakes. P‐value ≤10‐9) has been developed to quantify the effects of process variables (monomer and initiator intake. Both of the PCL‐based compatibilisers improve the interfacial adhesion between the starch particles and the PCL matrix in starch/ PCL blends. and their use in compatibilising starch‐PCL blends is described. 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.Summary glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. A statistically adequate model (R2=0.
v. De synthese van dergelijke producten met een in situ katalytische ringopening polymerisatie (ROP) van geschikte monomeren als b. Voorbeelden zijn thermo‐ plastificatie. Zetmeel is een polymeer bestaande uit anhydroglucose (AHG) eenheden. Na gebruik kunnen de materialen in het milieu eindigen en grote problemen veroorzaken. ε‐caprolactone op de hydroxy‐groepen van zetmeel is helaas niet eenvoudig. 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. Het amylopectine is een vertakt polymeer waarbij de vertakkingen via α‐D‐(1. Modificatie van zetmeel kan het aantal toepassingen sterk vergroten. Er zijn verschillende methodes beschikbaar om zetmeel te modificeren. Zetmeel is relatief goedkoop en te winnen uit een breed scala aan planten. chemische modificaties en combinaties van deze methoden.4)‐glucoside bindingen verbonden zijn. De toepassing van een alternatieve methode waarbij zetmeel eerst minder hydrofiel gemaakt wordt door de introductie van grote hydrofobe SiMe3 groepen. Biologisch afbreekbare polymeren op basis van zetmeel kunnen mogelijk bepaalde conventionele kunststoffen vervangen. heeft geleid tot een grote interesse voor biologisch afbreekbare alternatieven gemaakt uit hernieuwbare grondstoffen. Amylose is een lineair polymeer waarin de AHG eenheden hoofdzakelijk via α‐D‐(1. gevolgd door de ROP met een polyester precursor en verwijdering van de SiMe3 groepen . in combinatie met de huidige hoge prijzen voor de grondstoffen. Dit proefschrift beschrijft een aantal methodes om nieuwe polymere materialen op basis van zetmeel te maken. Het gebruik van zetmeel voor verpakking materialen is momenteel beperkt. 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. De verhouding amylose en amylopectine in zetmeel varieert en hangt grotendeels af van de plant waaruit het zetmeel gewonnen wordt.Samenvatting Kunststoffen gemaakt van fossiele grondstoffen als olie en gas zijn zeer aantrekkelijke materialen voor een breed scala aan toepassingen. Het zetmeel is daarnaast ook erg gevoelig voor water. mede omdat het niet mogelijk is om films met goede mechanische eigenschappen te maken. Dit. menging met andere materialen. De AHG eenheden kunnen op een aantal manieren aan elkaar gekoppeld worden.6)‐glucoside bindingen met de hoofdketen zijn verbonden. De productie van kunststoffen is sinds 1950 met bijna 10% per jaar gestegen. Twee type polymeren zijn dominant aanwezig in zetmeel: amylose en amylopectine.
46‐0. Een hogere verhouding leidde tot hogere molecuul massa’s. De gevolgde procedure bestaat uit drie stappen: hydrofobisering van het zetmeel door de introductie van SiMe3 groepen. gevolgd door in situ ROP met ε‐ caprolacton en daarna de verwijdering van de silyl groepen. 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%). waarin hogere temperaturen (100°C) tot lagere opbrengsten leiden. De effecten van procesvariabelen (tijd. De grafting efficiency van de gewenste reactie was 28‐58%. De opbrengst van de polymeren was een functie van de reactietemperatuur en de reactietijd.3. De resultaten worden in Hoofdstuk 3 in detail beschreven.68). 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. Er werden ook aanzienlijke hoeveelheden homopolymeer van ε‐ caprolacton met isopropyl eindgroepen gevormd. temperatuur en monomeer: monosaccharide verhouding) op de gemiddelde polymerisatie graad zijn gekwantificeerd met een statistisch model. De resultaten voor het modelsysteem waar de ROP van p‐dioxanone wordt geïnitieerd door de hydroxy‐groepen van een beschermde mono‐saccharide (1. De substitutiegraad van de SiMe3 groepen varieerde van laag tot middelhoog (degree of substitution.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (bij constante i Al(O Pr)3 concentratie). De grafting van ε‐ caprolactone aan het gesilyleerde zetmeel werd in THF bij 50°C uitgevoerd met Al(OiPr)3 als katalysator. De silylerings reactie werd bij 70°C uitgevoerd met hexamethyl disilazane (HMDS) in DMSO/tolueen mengsels. De kennis verkregen met het modelsysteem is toegepast om ε‐caprolactone te graften op zetmeel.3. De silyl groepen van het poly‐ε‐caprolactone gesilyleerde zetmeel werden uiteindelijk succesvol verwijderd door middel van een milde behandeling met verdund zoutzuur in THF bij kamertemperatuur. De 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. DS = 0.Samenvatting zou in theorie moeten leiden tot een hogere grafting efficiency.2.2.21‐0. 138 . De polycaprolacton eenheden (PCL) van het verkregen poly‐ε‐caprolactone gesilyleerde zetmeel hadden een gemiddelde ketenlengte van 40‐55 monomeereenheden (molecuul massa 4500‐6300). De DS van de PCL ketens lag tussen de 0.72 en was afhankelijk van de ε‐caprolacton/zetmeel verhouding.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) met Al(OiPr)3 als katalysator worden in Hoofdstuk 2 beschreven. De polymerisaties werden tussen 60‐100°C uitgevoerd en leverde vaste producten in geïsoleerde opbrengsten tussen de 30 en 96%.
13C‐NMR en FT‐IR gekarakteriseerd.4 verkregen.5 voor zetmeel laureaat. Bij lage vinyl‐ester/zetmeel verhoudingen leidde een verhoging van de vinyl‐ester concentratie in eerste instantie tot hogere DS waarden.5 voor zetmeel stearaat). In Hoofdstuk 4‐5. wordt een alternatieve methode voor de synthese van zetmeel esters met lange vetzuren beschreven. Echter de producten zijn nog steeds zeer bros. reactietemperatuur. inclusief statistische modellering. De synthese van deze esters wordt in het algemeen uitgevoerd met vetzuurchloriden die vrij duur en sterk corrosief zijn. Het gebruik van hogere (vet‐) zuren resulteerde in producten met beduidend betere mechanische eigenschappen en hydrofobiciteit. de vinyl‐ ester/zetmeel verhouding en het type katalysator. 0. De thermische en mechanische eigenschappen van sommige representatieve producten zijn uitgebreid bepaald. Het gebruik van Na2HPO4 resulteerde in producten met een laag tot middelhoge DS (0. Het gebruik van methyl en glyceryl esters resulteert in producten met een lage substitutiegraad. katalysator concentratie en katalysator basiciteit) op de DS zijn gekwantificeerd met een statistisch model (R2 ≥ 0. Met deze katalysatoren worden voor zowel de laureaat als stearaat esters producten met een DS van boven de 2.24 tot 2. vermoedelijk als gevolg van een verlaging van de polariteit van het reactiemengsel. K2CO3 en Na‐acetaat) in DMSO bij 110°C te laten reageren. De zetmeel esters werden bereid door zetmeel met vinyl laureaat of vinyl stearate in DMSO in de aanwezigheid van basische katalysatoren (Na2HPO4. De substitutiegraad varieert van 0. en Na‐acetaat) te laten reageren. zelfs in aanwezigheid van plastificeermiddelen. K2CO3 en Na‐acetaat vertonen superieure activiteiten vergeleken met Na2HPO4.07‐1. De effecten van belangrijke procesvariabelen (vinyl ester/zetmeel molverhouding.96. In de meeste zetmeel esterificatie studies wordt gebruikt gemaakt van kleine carbonzuren (C1‐C4). en in het bijzonder azijnzuur derivaten (C2). In Hoofdstuk 5. terwijl het gebruik van de K2CO3 en Na‐acetaat katalysatoren 139 . wordt een systematische studie. naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat. K2CO3. De zetmeel esters werden bereid door het zetmeel met vinyl laureaat of vinyl stearaat in aanwezigheid van basische katalysatoren (Na2HPO4. De DS van de producten is een sterke functie van de basiciteit van de katalysator. De DS van de producten was een duidelijke functie van de keten lengte van de vetzure vinylester. De temperatuur en de katalysator concentratie beïnvloeden de DS ook maar in mindere mate. Echter.96 en P‐ waarde van ≤ 10‐7). Samenvatting De esterificatie van zetmeel met carbonzuur derivaten is één van de oudste strategieën voor het verbeteren van zetmeeleigenschappen. bij verdere verhoging van de verhouding verminderde de DS. 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.3‐1. De producten werden met 1H‐. De acetaat esters zijn in het algemeen minder hydrofiel dan zetmeel.
140 . 1.9 voor zetmeel laureaat. De synthese van deze verbindingen wordt uitgebreid beschreven. gebruikt. een uitrekking tot breuk van 3‐26%. De FD waarden voor PCL‐g‐DEM waren aanzienlijk lager (tot 7%).4‐3. De thermische stabiliteit van de geësterificeerde producten is hoger dan die van zetmeel.5 MPa. De producten met een hoge DS (2. De hoogste FD waarden (tot 45%) werden gevonden voor PCL‐g‐GMA bij relatief hoge GMA en BPO concentraties. Dit leidt tot verschillen in mechanische eigenschappen tussen gecompatibiliseerde mengsels en van de niet‐ gecompatibiliseerde mengsels van zetmeel en PCL. Mechanische tests tonen aan dat de producten een treksterkte hebben variërend van 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). Als reactieve compatibilisers zijn twee polycaprolactone derivaten. De PCL gebaseerde compatibilisers werden bereid door reactie van laag moleculair PCL (Mw 3000) met glycidyl methacrylaat of diethyl maleaat in de aanwezigheid van een benzyolperoxide initiator bij 130°C.45‐1. 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. Beide reactieve compatibilisers verbeteren de interactie tussen de zetmeeldeeltjes en de PCL matrix. 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.0 voor zetmeel stearaat).957 en P‐ waarde van ≤ 10‐9) ontwikkeld waarmee de effecten van procesvariabelen (monomeer en initiator concentratie. en een elasticiteit modulus van 46‐113 MPa.1‐2.75 nog gedeeltelijk kristallijn zijn.7‐3.Samenvatting resulteerde in producten met een middelhoog tot hoge DS waarden (2. Het laatste hoofdstuk van dit proefschrift (Hoofdstuk 6) behandelt de menging van zetmeel en biologisch afbreekbare polymeren met reactieve compatibilisers.26‐2. Er is een statistisch model (R2 =0.
Broekhuis. I want to thank Erik Heeres for being more than a good supervisor. People from other groups have also been very kind and supportive to my research. and Laurens Bosgra for being always helpful with technical issues and Jan Henk Marsman for the help with analytical equipment. the raw material for fossil derived plastics. Mosiscki. Thank you Erik. for her hospitality and solving administrative matters. These include the current high oil prices. I want to thank Leon Janssen for his hospitality. I also want to give my deep appreciation to my reading committee. Thank you very much for spending your valuable time reading my thesis and for the constructive comments and corrections you have made. L. and for helping me arranging many things when I was not able to do it from Indonesia.Acknowledgements The synthesis of biodegradable materials is currently receiving a lot of attention. who was my first daily supervisor for about two months. I want to thank my three promoters. Minnaard. Anne Appeldoorn. First of all. as well as taking the SEM pictures. and for always encouraging me to finish this thesis. since the mechanical properties as well as hydrophobicity of this starting material is relatively poor. for your consistent support. together with difficulties in managing plastic waste. Prof. Although it is indeed a very interesting topic. and Prof. which improved the quality of this thesis. He has also been a good friend and always supported me during the hard times. Prof. Erwin Wilbers. 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. I want to thank Francesco Picchioni for being always there when I needed him to discuss anything. Leon Janssen. Erik Heeres. for his efforts to make every foreign students feel at home in Groningen. Marcel de Vries. for the inspiring ideas during the discussions. not in the last place because of the problems related with conventional plastics. 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. Francesco Picchioni. This thesis is a summary of my research on developing novel biodegradable polymers from starch. A. for valuable discussions during my early PhD period and for handing me over to the right hands.A. especially on his opinions about the statistical modelling and mechanical properties. Prof. Gert Alberda van Ekenstein and Harry Nijland helped me a lot with mechanical and thermal property measurements. Prof. and Prof. A. for upgrading my writing quality.J. I also want to thank Ineke Ganzeveld. Anno was also always there when . I want to thank Marya van der Duin‐de Jonge.
Ardiyanti. Thyara. Henky Muljana (my paranymph) and Anindhita Widyadhana. Endy. I want to thank Edwin A. and the ISD staff helped me a lot with visum and residence permit arrangements as well as tax papers. and Iwan Kustiawan for becoming my family. Taufik. and although we seldom met afterwards. pak Harry. Teddy. kak Roga. Oscar Rojas. Niels van Vegten. Anneke Toxopeus. Yongki. Mahfud. Yao Jie. visiting researchers.Acknowledgements I had questions or needed specific equipments. Tiara. Francesca Fallani. Puspita Kencana. Zhang Youchun. Francesca Gambardella. Tante Smith. Agnes R. Wahyudin. Farchad H. Erna Subroto. Wita Sondari. patience. Ignacio Melian Cabrera. in the office. for never leaving me alone. Judy Retti Witono. Egi. and Henk van de Bovenkamp for cheering my days and being my friends during summer and winter. Jelle Wildschut. Dado. for everything. My special thanks are for Laura Junistia and Jan‐Pieter Drijfhout. Teguh. Hans Heeres. Louis Daniel. Boelo Schuur. Anna Nizniowska. August Kurniawan. Tenny. I have met a lot of Indonesian students in Groningen. spring and autumn in Groningen. kak Atha. in the lab. Annette Korringa. M. Ais Jenie. Diana Jirjis. You all have always been my inspiration and motivation. A lot of thanks for Henk Stegeman and Poppy Sutanto. master students who worked with me. Hofman. support. Nidal Hammoud Hassan. Thank you Tresna. Taufik. and for my wife and family. Ita. Groningen is also a place where I found true friendship. Thanks also for Marcel and Connie. Asal Hamarneh. Albert Kieviet helped me with the MALDI‐TOF measurements.B. and many others for the good old times! Thank you Tante Ietje. 142 . Vincent Nieborg and Fesia Lestari. and on many occassions. and Tante Alma for being like my parents and family in Groningen. Samuel. Wangsa. Marcel Wiegman. Iwan Harsono. and helpful. and passion. Sari Fabianti. Arjan Kloekhorst. Tante Caroline. Wuri Raspati. Chalid. Rasrendra. Thank you God. Thank you my parents and sister. Buana Girisuta and Rina Karina. R. Many master and PhD students. mbak Ika. bu Ida Susanti. Jasper Huijsman. for your never‐ending blessings. Tante Oppie. C. friendly. and assistant professors have shared the days with me in Groningen. Patrick and Oppie. Our collaboration has also become part of this thesis. Mita. Nadia Gozali. Gerard Kraai. Togi. Ronny Sutarto. The chemical and glassware magazijn as well as the FWN library personnel were always patient. Kuslan. Diana Santangelo. mas Pandu. Abdul Osman. mbak Mia. for your love. Jaap Bosma. our memories will never fade away. Bima and Wisnu (also my paranymph). Thank you AW. and Mahesa for being my brothers and sisters in Groningen.
H.B. Sugih. L. Sugih. Appl. A. Janssen. L.P. L. Heeres. 3. Bratislava (Slovak Republic).Picchioni. H. F. starch‐based innovative product: The synthesis of grafted polycaprolactone‐starch by ring opening polymerisation.J.Janssen. Drijfhout. Sugih. A.K. R. H. Picchioni.P. Manurung.K. L. and F. 2008 (submitted to Carbohydr. J. South China University of Technology. F. Picchioni: Functionalization and blending of polycaprolactone. J. Synthesis of poly‐ (ε)‐caprolactone grafted starch co‐polymers by ring‐opening polymerisation using silylated starch precursors. F. Sci. A.M.K. Broekhuis.J. F. presented in 11th Asia Pacific Confederation of Chemical Engineering (APCChE) Congress.P. 2008 (accepted for publication in Starch‐Starke). L. Heeres: Green.1016/j. F. 17‐20 December 2006 (oral presentation).K. H. H. presented in International Workshop: “From Polymer Modification to Multicomponent Polymer Systems”.010). Heeres. F. Sugih. Sugih. A.List of Publications 1. presented in the 1st International Symposium on Sustainable Chemical Product and Process Engineering. L. and H.K. J.P. 2008. Polym. Experimental Studies on the ring opening polymerization of p‐dioxanone using an Al(OiPr)3‐monosaccharide initiator system. H. 2008 (accepted for publication in Starch‐ Starke).B.J. Picchioni.J.P. Junistia.). Experimental and modeling studies on the synthesis and properties of higher fatty esters of corn starch.Heeres. Kuala Lumpur (Malaysia). F. India. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants.Drijfhout. Pune Chapter. L.K. 27‐30 August 2006 (oral presentation) 7. 2008 (submitted to J. DOI: 10. 2.10. Heeres: Synthesis and properties of novel green biopolymers derived from starch and fatty acids. Guangzhou (China). A. A.2008. A. and F.A. Drijfhout. .M Janssen and H. Heeres. Picchioni. L. 25‐28 September 2007 (oral presentation).P.A. Sugih. J. Heeres. Junistia. Heeres. Picchioni.eurpolymj. Junistia.J. Broekhuis. Sugih.M. Junistia. Picchioni: Biodegradable polymer materials based on polyesters and starch/proteins. Manurung. A. Picchioni. Sugih.K.J. (in press.K. A. and H. 9th National Conference organized by the Society for Polymer Science.K. R. 9. Picchioni.M. Sugih. Synthesis and properties of reactive interfacial agents for polycaprolactone‐starch blends. 5.B. A.J.) 6. Mumbai (India).P. 26‐28 November 2006 (oral presentation). Janssen. presented in “Polymers for Advanced Technologies”. Polym. A.J. Eur.B. 4. Polym.J. 8.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.