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
H. A.P. Broekhuis Prof. Picchioni Beoordelingscommissie : Prof. Moscicki ISBN : 978‐90‐367‐3592‐6 ISBN : 978‐90‐367‐3593‐3 (electronic version) . dr. F. dr.A. Promotores : Prof.M.J.B. ir. Heeres Prof. L. L. dr. Minnaard Prof. Janssen Prof. dr. ir.J. dr. dr. A.
to: Tresna. my parents. and my sister .
188.8.131.52.2.2.1. Plastic applications and waste issues 1. The potential of biodegradable plastics 1.2.5. Starch production processes 184.108.40.206.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (2) 27 28 26 27 27 27 2.3. Thesis Outline 1. Starch‐based blends by in situ polymerization 220.127.116.11.2. Biodegradable plastics from starch 1. Cross‐linked starch 18.104.22.168. Starch‐based blends by melt mixing 1.2.4. Thermoplasticized starch 1. Starch ‐ biopolymer blends and graft co‐polymers 1.2.1. Biomaterials from starch 1. Introduction 2. Methods 2.2. Typical example for the synthesis of polydioxanone end‐capped with 1.3. Starch 1. Structure and properties of starch 22.214.171.124.3.1.Table of Contents Chapter 1: Introduction 126.96.36.199.3.1. Product analyses .2.1.1. Starch modifications to improve product properties 1. Materials and Methods 2.4.4. References 2 3 4 5 5 7 9 9 10 11 11 13 14 16 16 17 Chapter 2: Experimental Studies on the Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐ Monosaccharide Initiator System 2.1. Materials 2. Starch Esters 1.
3. 188.8.131.52. Materials and Methods 184.108.40.206.2. Materials 3. Screening experiments 2. Product analyses 2.5. Typical example of the starch silylation procedure 220.127.116.11.4. Introduction 3. Nomenclature 18.104.22.168. Calculations 3.2.3. Calculation of average degree of polymerization 22.214.171.124. Analytical methods 3. References Chapter 3: Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors 3. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers 51 51 52 52 52 53 54 48 49 49 50 50 3. Synthesis of silylated starch viii .126.96.36.199. Nuclear Magnetic Resonance (NMR) 3. Methods 3. Typical example of in situ polymerization of ε‐CL with silylated starch 50 3. Systematic studies 188.8.131.52.2.2.6. Results and Discussions 2.2.2. Effects of process conditions on the average chain length ( X n exp ) and end group distribution 28 29 29 30 36 38 39 40 42 43 44 2.3.3. Results and Discussions 184.108.40.206. Peracetylation of silylated starch 220.127.116.11. Product yield 18.104.22.168. Conclusions 22.214.171.124. Mechanistic aspects 126.96.36.199.
2.2. 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 4.3.2. Systematic studies 4.2. In situ ring opening polymerization of ε‐caprolactone with silylated starch 57 63 64 64 65 188.8.131.52. Peracetylation procedure 4.3. Materials 4.2. Nomenclature 4.6.3. Catalysts screening 4.2. References ix . Effect of the addition of toluene as a co‐solvent 184.108.40.206. FT‐IR measurements 220.127.116.11. 18.104.22.168.22.214.171.124.126.96.36.199. Introduction 188.8.131.52.2.1.3. References Chapter 4: Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants 4.3. Exploratory experiments 4.3.3. Determination of the Degree of Substitution (DS) 4. Conclusions 4. Results and Discussion 184.108.40.206. Product characterisation 4. Effect of vinyl ester to AHG ratio on the product DS 4. Conclusions 3. 1H‐ and 13C‐NMR analyses 4. Methods 4. Materials and Methods 220.127.116.11. Nomenclature 3. Deprotection of silylated‐starch‐g‐PCL 18.104.22.168. Analytical equipment 4.5.
22.214.171.124. Compatibilizer synthesis 6.3.4. References Chapter 6: Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 126.96.36.199.2.4. Work‐up of PCL‐g‐GMA products 6. Peracetylation procedure and Degree of Substitution (DS) determination 88 88 89 92 96 102 102 103 5. Materials 188.8.131.52. Experimental Design 5.3. Materials and Methods 6.2. Introduction 6.3. Results and Discussion 5.2.1. Work‐up of PCL‐g‐DEM 108 109 109 109 109 110 110 6. Conclusions 184.108.40.206.3. Analytical Methods 110 x . Introduction 220.127.116.11. Nomenclature 18.104.22.168.3.2. Preparation of PCL‐starch blends with the reactive compatibilisers 110 6. Mathematical modelling 5.2.2. Methods 5.5.Chapter 5: Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch 5.2.2. Analytical Equipment 5. Materials and Methods 5. Product properties 22.214.171.124. Typical example of the preparation of laurate and stearate starch Esters 88 86 87 87 87 88 5.3. Methods 126.96.36.199. Materials 5.2.
2.3. Thermal properties of the compatibilizers 6.2. Results and Discussions 6.1.4. Modeling of the combined effects of the GMA/ DEM and BPO intakes on the FD 111 111 111 116 117 118 120 121 122 123 125 128 128 129 188.8.131.52. Ternary blends compatibilized with PCL‐g‐GMA 6.3. Preparation of the ractive compatibilizers 6. Statistical mdeling 184.108.40.206.3.3. 220.127.116.11.1. Effect of the BPO intake on the product FD 18.104.22.168. Conclusions 6. Synthesis and Properties of Starch‐ PCL Blends 22.214.171.124. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers 6.3. Effect of substrate (GMA/DEM) to PCL Ratio on the FD 126.96.36.199.188.8.131.52. Ternary blends compatibilized with PCL‐g‐DEM 6.2. Nomenclature 6. Binary blends of starch and PCL 6.1. References Summary Samenvatting (Dutch Summary) Acknowledgements List of Publications 133 137 141 143 xi .
Different routes to modify starch to improve the product properties and to extend the application range will be provided. Finally. .Chapter 1 Introduction Abstract A general overview of starch properties will be given and the potential use of starch as a starting material for a wide range of green biomaterials will be reviewed and discussed. an outline of this thesis is given.
Past. potato. or peas) and in tubers or roots (i.e.45 (estimated from the total volume and value of the US corn starch export ). 80 70 60 50 40 30 20 10 0 Starch Production (Million Tons) 1995 2000 Year 2005 2010 Figure 1. Most of the starch produced worldwide is derived from corn. Most starch crops are very productive. maize. barley. . Potato accumulates starch to approximately 75 % of the dry weight in the tubers with a yield up to 21 ton starch per hectare. An important polysaccharide is starch. with an average yield of 4. sorghum. Starch 75% of all organic material on earth is present in the form of polysaccharides. Most of the starch is produced in the USA. present and a forecast of starch production  o : Europe □ : USA : Total ∆ : the Rest of the World 2 . The worldwide production of starch in 2008 is estimated to be around 66 million tons . but other types of starch such as cassava. The current price of corn starch is around $0. . The second and third starch producer regions are Europe and Asia [4‐5].1.9 ton starch per hectare .e. sweet potato. corn.Chapter 1 1. while corn seeds consist of 65‐80% starch by weight. potato or cassava) of the plants [2‐3]. Plants synthesise and store starch in their structure as an energy reserve. Starch is found in seeds (i. Past.1. rice. current and future estimations of worldwide starch production are given in Figure 1. wheat. 4‐5]. It is generally deposited in the form of small granules or cells with diameters between 1‐100 µm . and wheat starch are also produced in large amounts [2.1.
1. the mixture is fed to a hydrocyclone to separate the germ and the rest of the kernel. The resulting suspension from the mills contains fiber. As an example. Figure 1. the debris is filtered from the slurry. After a coarse milling. The plant material is grounded in water. which is again fed to a second milling process.b. b. The starch is present in the endosperms (floury and horny).2. After initial cleaning to remove cob. The resulting starch is further washed in a cyclone and finally dried. and starch. gluten. the corn kernel is softened by steeping in warm water containing SO2 until the volume of the kernel increases with 55‐65%. sand. Starch production processes Starch is generally extracted from the plant by wet milling processes [9‐10]. Typical corn‐milling operation b. Introduction 1. Cross‐sectional view of a corn kernel 3 .2.2.a. a typical corn starch production process is given in Figure 1. a. Corn starch production  a. and starch granules are obtained after centrifugation from the suspension.1. The fiber is removed using washing screens. and other foreign materials. The low density gluten is separated from the starch suspension by centrifugation. and is embedded in a proteinaceous cellular matrix as is shown in Figure 1.
containing periodic branches linked with the backbones through α‐D‐(1.1..3.. The amylose content of several common starches is given in Table 1. H O H OH H OH H O H OH O H OH H O . H2COH H H OH H O H OH O H OH O CH2 H O H H OH H O H OH H O .6)‐ glucosidic bonds . but usually in the range of 1.. Structure and properties of starch Starch is a polymer consisting of anhydroglucose (AHG) units (see Figure 1.. H OH H2COH H2COH H . .1. Typically.Chapter 1 1. Chemical Structure of Starch a.a. H OH H O H OH H O H H OH H c.6‐7 x 105 Da . The linear 4 .3. The starch granules will swell rapidly to many times of its original volume... Figure 1. 6]... Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1.) . The content of amylose and amylopectine in starch varies and largely depends on the starch source. The molecular weight of amylopectin is higher than that of amylose and is typically 4‐5 x 108 Da . H OH H H2COH H2COH H O H H OH H H2COH H O H H OH H O H OH H O ..4)‐glucosidic bonds. the amylose content is between 18‐28% . H . Amylopectin Starch is insoluble in cold water... Anhydroglucose (AHG) unit b. The molecular weight of amylose is a function of the plant source and processing method. Two types of AHG polymers are usually present in starch: amylose and amylopectin [2‐3.. but it is very hygroscopic and binds water reversibly. a. Amylose c.. Amylopectin is a branched polymer. H2COH H OH H O H H O b.2. . H2COH H . Heating a starch solution leads to loss of hydrogen bonding in the interior of the starch granule and the starch will start to gelatinize. Each branch contains about 20‐30 anhydroglucose units..
and.5 28 52 70‐75 27 15. swollen granules. Biomaterials from starch 1. The first generation of commercial plastics was derived from cellulose nitrate and is known as celluloid [12‐14].7 20 18. and celluloid was 5 .1. Parker in 1838. Introduction amylose molecules leach out of the granules into the solution. depending on the amount of water present.2.2. Starch gelatinization temperature range  Starch Potato Tapioca Corn Waxy corn Wheat Gelatinization Temperature Range [o C] 59‐68 58. and granule fragments. Cellulose nitrate was first prepared by A. Plastic Applications and Waste Issues Plastic is the general term for a wide range of synthetic or semisynthetic polymerisation products. The gelatinization temperature range can be defined as the temperature at which granular swelling begins until the temperature when nearly 100% of the granules are gelatinized .5 25.7 26 Table 1. will form a thick paste or gel. The gelatinization temperature range of various starch sources is given in Table 1.8 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 (%) 20.8 184.108.40.206‐70 62‐72 63‐72 58‐64 1. Table 1. The resulting suspension contains a mixture of linear amylose molecules. Plastics are used in a wide range of applications and the demand is still increasing every year .2.
polystyrene (PS). The production and consumption of plastics has increased significantly with a rate of almost 10% every year since 1950.5. A dramatic increase in demand for plastics began after World War II.4. discovered by Baekeland in 1907 [12. PE is a very versatile plastic because it can be shaped easily into various forms. and consumer products (see Figure 1. Coatings. polyethylene terephtalate (PET). Worldwide Plastics Production (1950‐2006)  6 . Plastics are very attractive materials. Sealants. and PP-Fibers 2006: 245 2002: 200 200 Plastics Production (Million Tons) Not Included: PET-. 15]. Polyurethanes. and therefore need less energy to shape it into useful materials .). when polyethylene (PE) was invented.Fibers 150 1989: 100 100 1976: 50 50 1950: 1. Hyatt in 1870 .4. and PolyAcryl. the worldwide plastics production has reached 245 million ton per annum  (see Figure 1.Chapter 1 patented by J. PA-. Plastics have lower melting temperatures compared to glass and metals. Thermosets. and 37% of the total plastics demand in EU  is used as packaging materials. and energy during transportation of goods. The largest application of plastics is for packaging purposes. for instance to be used in packaging and paper coatings . space. About 29% of the total plastics produced in the USA . Elastomers.) 250 Included: Thermoplastics. polypropylene (PP). They have a low density and can be shaped in thin films while maintaining good properties. Plastics are also used for building materials and automotive.5 0 1950 1960 1970 1980 Year 1990 2000 2010 Figure 1. the oldest purely synthetic plastic is Bakelite. Important polymers used for packaging are polyethylene (HDPE and LDPE). While celluloid is derived from a natural polymer (cellulose). In 2006. and polycarbonates (PC). polyvinyl chloride (PVC). electrical. The latter is important when using the material for packaging purposes to save weight. Adhesives.
Introduction Plastic waste. Biodegradable plastics are polymeric materials capable of decomposing when given an 7 . In the USA. threatening fishery.5. which are usually found in waste streams. are not easy to recycle . however. plastics waste was ranked as the second major source of MSW after paper and paperboard. and may be more than twice the weight fraction . The disposal of plastics materials in municipal solid waste (MSW) is a serious issue in many parts in the world. It has a high volume to weight ratio and is resistant to degradation. hydropower plants operation. It is not applicable for thermoset resins .2. soil. 16]. consisting of 25%‐v of the total waste . In 1996. Figure 1. In Europe (2006). Plastic applications and waste treatment in Europe  1. irrigation. the amount of plastics in MSW increased from less than 1% in the 1960 to 11. Plastics have been polluting sea [1. 6]. Plastic litter is hazardous to wildlife [1. and other public works . only 20% of the plastic is recycled. and lakes. and is only effective for single plastic sources or simple plastic formulations. Recycling (part of) the plastics to reduce the amount of MSW also has limitations. The volume fraction of plastic in MSW is much larger due to the low density of plastics. The potential of biodegradable plastics The application of biodegradable plastics could be an attractive solution for the problems related to the use of conventional plastics (vide infra). Most of the waste is still disposed (50%) by landfill or incinerated to recover the energy (30%) .2. rivers. is causing serious environmental problems. ship navigation. and accumulation of its residues in soil cause significant reductions in agricultural yields.7% in 2006 . Co‐mingled plastics.
Biodegradable plastics may be classified into two general groups. Several authorities have provided definitions for biodegradable plastics . The price of these biodegradable plastics is expected to be reduced considerably in the next decade .5/ 8 . Biopolymers are inherently biodegradable. ASTM sub‐committee D20. 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. For instance. and microorganisms) and biodegradable synthetic polymers . The change in chemical structure results from the action of naturally occurring micro‐ organisms. Biodegradable plastic waste may be treated in composting facilities. 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 . and polyesters from micro‐organisms . The market price of biodegradable plastics (mainly from starch and PLA) has decreased the last years.Chapter 1 appropriate environment and sufficient amount of time . Biodegradable plastics have gained considerable interest since the 1980s. because they take part in nature’s cycle of renewal. and since 2000 they have become competitive with traditional materials in some applications [20‐21]. biological and/or physical forces in the environment at a rate which leads to fragmentation or disintegration of the plastics. together with food and yard waste as well as paper. Nowadays new types of biodegradable plastics with improved properties and lower costs have been developed . proteins (present abundantly in plants and animals). 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.96 proposal ‐ Degradable plastics are plastic materials that undergo bond scission in the backbone of a polymer through chemical. animals. the price of PLA based plastics in 2010 is estimated to be around € 1. biopolymers from nature (from plants. DIN 103. It may also be treated in sewage sludge water treatment plants or buried in the soil [20‐21]. Common natural biopolymers are carbohydrates.2 working group on biodegradable polymers ‐ Biodegradation of a plastic material is a process leading to naturally occurring metabolic end products.
Biodegradable Plastics from Starch Starch is a very attractive source for the development of biodegradable plastics. 20]. The price of starch in 2007 was about $ 0.2‐ 1.50‐4. either from starch‐based materials (slightly modified starch. biofillers for tires. Introduction kg.50/ kg in 2005) is also expected to be lower in the future due to considerable reduction in modification costs . . Examples are the use as packaging material for carrier bags.4 per kg in 2007‐2008 ). tensile and flexural strength) and is too sensitive to water [26‐27]. This price is much lower than conventional plastics derived from oil. Biodegradable plastics are especially very useful for single‐use applications.45 per kg . These are summarized briefly in the next sections. when recycling is not practical or uneconomical [20‐21]. 1. € 1. 9 . either by plasticization. alone or complexed with natural or synthetic biodegradable polymers) or from polylactic acid which originates from the fermentation of a starch feedstock. starch must be modified. and chewable items for pets [1. blending with other materials. napkins) . Virgin starch is not suitable as a packaging material. breathable fabrics. 1. It cannot be shaped in films with adequate mechanical properties (high percentage elongation. Bastioli [20‐21] showed that nearly all biodegradable plastics available in the market are derived from starch. because the price of oil based polymers may still increase due to the rise in the crude oil prices . chemical modification or combinations of them .8‐ 2. utensils . such as polyethylene (PE. towels. hygienic materials (diapers.4 or $ 1. Consequently. Biodegradable plastics are also used for food servicewares . agricultural products (mulch films. Starch may become an attractive raw material for plastics in the future. The price of starch‐based biodegradable plastics (€ 1. before they can be applied as biodegradable plastics.3. pots). Starch modifications to improve product properties Several techniques may be applied to develop starch based biomaterials with improved properties. 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 . consumer products and food products.3. half of its price in 2003 (around 3 Euro/kg). or when environmental impacts have to be minimized.2.
30. 39]. and therefore are processable by conventional polymer processing techniques such as injection. the semicrystalline structure of starch and its granular form are lost and the starch polymers are partially depolymerized. For instance. As the result. water content. Another issue is the poor water resistance and low strength. such as water and polyols (glycerol. or melted starch [29. et al  discovered that the elongation of break of the thermoplastic starch is significantly improved by plasticization with glycol.Chapter 1 1. plasticized. There are several substances used as plasticizer for the preparation of thermoplastic starch (TPS). Thermoplasticized starch Virgin starch is brittle and difficult to be processed into articles due to its relatively high glass transition temperature (Tg. which is even above the thermal degradation temperature. sugars) [30. Yu. which results in embrittlement. and blow moulding [28. Although thermoplasticization seems to be a promising method. 32‐ 36]. 33]. destructed. The brittleness is known to increase in time due to free volume relaxation and retrogradation. 35. 40]. which still limits their use. and the properties and amount of plasticizers. and hexylene glycol. glycerol. extrusion. The plasticizers are also usually hydrophilic and can be 10 . 32.3.1. 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. The thermoplasticization process will decrease the interactions of the molecular chain and destruct the structure of the starch . TPS synthesized from polyol and sugar plasticizers have the tendency to re‐crystallize (retrogradation) after being stored for a period of time. The use of other plasticizers (for example glycerol) results in a rubbery material. At intermediate water levels. The use of water as a plasticizer is not preferable. sorbitol. This problem is mainly caused by the presence of strong inter‐ and intra‐molecular hydrogen bonds between the starch macromolecules [28. phase separation may still occur. . approximately 230 oC) . The plasticized starch properties may be tuned by changing the temperature of processing. 37]. because the resulting product will be brittle when equilibrated with ambient humidity . which will result in a continuous phase in the form of a viscous melt [29‐30. Starch can be modified to obtain materials which melt below the decomposition temperature . resulted in the formation of an amorphous mass [35. glycol. The modified products are known as thermoplastic. 40]. the thermal properties of glycerol‐plasticized starch are a function of water content [31. with better properties than virgin starch in various applications. The modification involves break down of the starch granular structure by the use of plasticizers at high temperatures (90‐180 oC) and shear. 30].
Most studies dealt with the synthesis of starch esters of C1‐C4 carboxylic acids.3. especially in the food. Already in 1865. epoxy. 1. aldehydes. The cross‐linked products are therefore more rigid (and less elastic) materials than virgin starch . Schuetzenberger acetylated starch with acetic anhydride . Higher processing temperatures generally led to higher cross‐linking levels. The DS is defined as the moles of substituents per mole of AHG units . As a result. the use of formyl esters of starch is much lower than that of acetate esters. Solutions to improve the properties of TPS are blending.3. High DS starch acetate can be easily casted 11 . Chemicals of these classes having two or more of the reactive groups may react with two or more hydroxyls of the starch molecules. and ethylenic compounds. while highly substituted starch acetate (DS of 2‐3) are soluble in organic solvents. 1. Low DS products (0. 46]. differing in Degree of Substitution (DS). The most popular starch ester is starch acetate . et al  studied the mechanical properties of films from maize starch cross‐linked by sodium trimetaphosphate (SMTP). Cross‐linking results in a reduction of the solubility in water and to thickening. 39].2) are commercially available and used as food additives and in the textile industry.2.01‐0.3. textile. Introduction washed out by water . The esterification reaction is catalyzed by H+ and proceeds with the formation of water . Starch Esters The development of starch esters started in the mid 19th century [44. and particularly with acetic acid . or coating with hydrophobic polymers [33. paper. Breakdown of the starch and the formation of low molecular weight products occur to a significant extent . Medium DS starch acetates (0. organic chloro‐compounds. These hydroxyls can react easily with a wide range of compounds such as acid anhydrides. The products are called cross‐linked starches. The simplest starch ester is starch formate (C1). Cross‐linked starch An anhydroglucose molecule of starch contains two secondary and one primary hydroxyl group. resulting in an increase of the Young’s modulus and tensile strength of the products and a decrease in elongation at break. and adhesive industry. Three different types of starch acetates may be distinguished. leading to higher viscosities [42‐43]. The cross‐linked starches have found many applications. Substitution of the hydroxyl groups of starch with acetate groups makes the esters more hydrophobic than native starch. Silva.3‐1) are still soluble in water. synthesized by direct addition of formic acid to starch at room temperature.
or isopropanol) with acidic (hydrochloric or sulfuric acid) or alkaline (NaOH or triethylamine) catalysts . Another attractive route involves the use of vinyl esters as reagents . hydrogencarbonate. Mechanism of starch acetylation using vinyl acetate 12 . DMSO. The degree of acetylation of starch acetates can be easily controlled. 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. Esterification may be performed using acid anhydrides in aqueous media or organic solvents (pyridine. Acetylation of starch in water and DMSO using vinyl acetate has been studied lately by Mormann and Al‐Higari  as well as Dicke .1. and is shown in Scheme 1. DMF. Starch triacetate has been successfully synthesized using acetic anhydride in combination with pyridine‐gelatinized starch . The kinetics of the reaction between gelatinized aqueous potato starch and vinyl acetate was studied by De Graaf.1. Neutral and weak acid/ alkaline catalysts result in regioselective substitution at the C2 hydroxyl groups of starch. xylene. acetate. Several synthetic routes have been developed for starch esters. while alkaline catalysts (such as carbonate. but also C‐6 and C‐ 3 substitution . High DS starch acetates are thermoplastic materials suitable to be used as biodegradable plastics . et al [55‐56].Chapter 1 into films using organic solvents. allowing the polymer to be produced with a range of hydrophobicities . Starch esters have also been synthesized using alkanoyl chlorides [49‐53]. and phosphate) will result in C‐2.
one component will be dispersed into the other. Although the in situ polymerization process is not as technologically straightforward and economically convenient as melt‐mixing. dodecanoyl. the latter are mixed together. however. relatively expensive and rather corrosive. 1. The classical example of such process is represented by the production of High Impact Poystyrene (HIPS) obtained by styrene polymerization in the presence of polybutadiene . However. However. Two main methodologies are applied for the production of polymer blends.61) products could be obtained using this approach. The first involves simple melt mixing of the two components for example by extrusion. availability of mixing equipment and no necessity to use any organic solvent (often employed for the in situ polymerization).e. 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. the two components are chemically linked to each other. which in turn provide a very strong adhesion at the molecular level between the dispersed phase and the matrix. in order to achieve also a good molecular adhesion between the phases by melt mixing. Despite this advantage. a chemical reaction might take place upon processing (reactive extrusion). i. Introduction The synthesis of long chain fatty acid ester of starch has attracted much interest lately. The fatty acid chloride reactants are. By working at temperatures above the melting point or glass transition temperature of the two components. they locate themselves at the interface between the 13 . Their role is comparable to the one of a surfactant in emulsion formation . 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 . but only relatively low‐DS (0. polymers are rarely miscible with each other  so that.3.34‐0.g. melt‐mixing remains the preferred route to polymer blends mainly because of very practical reasons: low costs. octadecanoyl) chlorides [49‐53]. interfacial agents (e. If the right combination of chemical groups on the two components is present along the polymers backbone. 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 . Fatty‐acid starch esters have been synthesized using fatty acid (octanoyl. in the simplest case of a binary blend.4. Starch ‐ Biopolymer Blends and Graft Co‐Polymers Blending of different polymers is an established method to obtain products with improved properties. it is frequently used in order to chemically graft the polymerized component on the other one (polystyrene on polybutadiene in our example above ). compatibilizers) might be used. The introduction of longer acid chain is expected to reduce the brittleness of virgin starch and to increase its hydrophobicity . As a result.
The synthesis of compatibilizer and its use for starch/ synthetic polymer blending has also been studied.4.1. Starch‐based blends by melt mixing Starch‐synthetic polymer blending has been studied as early as in 1973 [64‐65]. In the past.3. The most often used synthetic polymer for blending with starch is polyethylene [27. The uncompatibilized blends of starch and 14 . Illustration of the role of interfacial agent in compatibilizing blends 1.Chapter 1 two components stabilizing the dispersion (most probably by a steric repulsion mechanism ) and providing an improved adhesion at the interface. The starch/polyethylene blends are used for agricultural mulch [27. The products have been synthesized in the lab as well as on industrial scale [32. 39. Matrix Dispersed Phase Figure 1.6. 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. 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. 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 . 66‐81]. there have been efforts to blend as well as to graft synthetic polymers onto starch. 64‐98].6.
dextran‐g‐PCL . 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‐dioxanone. poly‐lactides. since the conventional synthetic polymers are usually poorly or non‐biodegradable. the compatibilizer. Some of these polyesters also have very good mechanical. 15 . poly‐valerolactone. which both can react in situ with the hydroxyl groups of starch [70. among which synthetic polyesters are considered very promising materials . Introduction polyethylene show a coarse phase separation due to differences in polarity of starch (hydrophilic) and polyethylene (hydrophobic). PE‐g‐MA contains reactive anhydride sites. which combines excellent biodegradability with acceptable mechanical properties. PCL‐g‐pyromellitic anhydride [94‐95]. 84‐98]. The use of reactive compatibilizer precursors PCL‐g‐glycidyl methacrylate (PCL‐g‐GMA) [89‐90]. and EVOH . 96]. Studies on the blending of starch with PCL have been already described [32. 72‐73]. 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. is formed upon mixing. and poly‐caprolactone). As is the case of blends with PE. thermal and water/gas permeability properties that are even comparable to bulk non‐biodegradable polymers such as PE and PP. PCL‐g‐maleic anhydride (PCL‐g‐MA) [91. Examples of these biodegradable polyesters are poly‐glycolide. Despite these good results. The mechanical properties of these blends (tensile strength and elongation at break) decrease at higher starch content. The mechanical properties of the blend are also improved. The tensile strength of the uncompatibilized blends is drastically reduced when starch content is increased. The use of conventional synthetic polymer such as PE for blending with starch will only result in a partially biodegradable material. i. As a result a graft copolymer (PE‐g‐Starch). thus improving the dispersion of starch in the PE matrix. Polycaprolactone (PCL) is a well‐known synthetic biodegradable polyester. while PE‐g‐GMA posseses epoxide groups.e. To obtain completely biodegradable products. while the tensile strength of the compatibilized ones decreases only slightly with the starch content. systematic studies on the compatibilizer precursor synthesis as well as on the influence of the chain topology and chemical reactivity (for the in situ compatibilizer formation) have not yet been reported. The ester bonds of these polymers are susceptible to attack by water and this leads to enhanced biodegradability. These biodegradable polyesters will finally decompose into non‐toxic products . synthetic biodegradable polymers have been applied. and poly‐lactones (such as poly‐butyrolactone. 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. 39.
during the reaction. In this reaction. In the past. 1. A new strategy for the in situ ROP of ε−caprolactone on starch with the use of common ROP catalyst is therefore highly desirable. and difficult to handle since it releases ethane.4. 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 method applied basically consists of three steps (temporary partial protection of starch‐hydroxyl 16 . The results of this study have been used as input for the synthesis of the starch‐g‐PCL. and the presence of liquid‐solid reaction system leads to reduced reaction rates between starch and ε−caprolactone.2. 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 . Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). Starch is a very hydrophilic material that always contains moisture.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) using i Al(O Pr)3 as the catalyst. Three routes have been studied in detail: starch esterification. Starch‐based blends by in situ polymerization The synthesis of starch based graft copolymers by in situ polymerization represents not only an alternative route to melt blending (vide supra) for the production of novel biomaterials. In Chapter 3. Another approach consists of the Ring Opening Polymerization (ROP) of ε‐caprolactone monomer in the presence of starch. the hydroxyl groups of starch are supposed to function as initiating sites.2. the synthesis of starch‐g‐PCL is reported.4. leading to formation of PCL homopolymers rather that starch‐g‐PCL. Thesis Outline The objective of this thesis is to study synthetic pathways to obtain starch derivatives with the potential to be used as bioplastic. starch/ biopolymer blending. and starch‐g‐ biopolymer formation. defined as the percentage of grafted polyester to starch compared to the total amount of homopolymer and grafted polyester) . a very flammable by‐product.3. which however is extremely air‐ and water‐sensitive.3. Chapter 2 describes the ROP of p‐dioxanone initiated by a protected monosaccharide (1. Previous studies showed that common ROP catalysts such as tin octoate or aluminium isopropoxide gave low (0‐14%) grafting efficiencies (GE. thus resulting eventually in low GE values . The highest GE value (up to 90%) has been achieved when using triethylaluminium as catalyst [97‐98].Chapter 1 1. Another possible cause for the low GE values is the heterogenous nature of the reaction.
. 2005. C. Wurzburg). Corn Refiners Association. Central Science Laboratory. 105‐122.. Corn.N. 1. The synthesis of two interesting compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) for starch/PCL blends is discussed in Chapter 6. . Bangor.C..html. Fowler: Industrial markets for starch. . in Modified Starches: Properties and Uses (Ed. York. Academic Press. Wurzburg: Introduction. and removal of the silyl groups). Aarhus. CRC Press. . New Jersey. Bemiller. R. 1986.F. ROP of ε−CL on the remaining starch‐hydroxyl groups.. are also reported.B. Agricultural and Rural Strategy Group. Y. a systematic experimental study on the effect of the process variables on the starch ester DS has been performed. USA. 2007 17 . Boca Raton. The model. starch. R. An Introduction to the New Science of Biodegradable Plastics. P. 2008. Denmark: http://www. E. Somerville: Mini review: Plant polymers for biodegradable plastics: cellulose. Passchall).B.starch. addition of co‐solvent. 2004. References .L. Inc. A preliminary study on the effect of several process variables (reactant ratio.H. accessed on August 01. Gywnedd. Website of International Starch Institute.A. The use of these compatibilizers in starch/PCL blends. Washington D.M. C. Nawrath. UK. USA. . J. Inc. 1984. and application of different catalysts) on starch esterification using vinyl fatty esters (vinyl laurate and vinyl stearate) is provided in Chapter 4. in Starch: Chemistry and Technology (Ed. 1. O. London. A. University of Wales. Whistler. E. 1995. UK.5. Holmes: Interactive European Network for Industrial Crops and their Applications. de Bragança. 2002. Young: Fractionation of starch. and polyhydroxyalkanoates. The Biocomposites Centre. Sand Hutton. Introduction by trimethylsilyl groups. 2000‐2005. O. USA. UK. Based on these preliminary results. .S. The results were quantified using a statistical model. Poirier. . including the mechanical and thermal properties. Princeton University Press. together with the mechanical and thermal properties of the synthesized starch esters are reported in Chapter 5. Breed. Mol. Part of A Global Economy. C. Stevens: Green Plastics. Corn Refiners Annual Report 2007. Summary Report for the European Union.dk/isi/stat/rawmaterial.
E. J. Germany.M. Hanser Publishers. Passchall). X. C. Braun: Environmental impacts of biobased/ biodegradable packaging. Inc.Chapter 1 . B. Inc. Starch/ Starke 2001. A. Selke. Daniel: Starch. 105‐109 . 31. J. Inc. and Regulations. Prog.J. Munich. Wolf. UK.N. 53. 2000.B. M. Technomic Publishing Company. Applications. Weber). Department of Dairy and Food Science. Belgium. Inc. 2000. Institute for Prospective Technological Studies and European Science and Technology Observatory. 2008 . 18 . J. . Chem.R. The Royal Veterinary and Agricultural University. S.J.W. Hanser Publishers. Whistler. G. R. Selke: Packaging and the Environment. 23. Ind. Processing. . Denmark. Watson: Corn and sorghum starches: production. Seidel). R. Angerer: Technical Report Series. U. Germany. Bastioli: Global status of the production of biobased packaging materials. Fritz.F. J. .. Munich. Website of Environmental Protection Agency. Bemiller. E. Link. USA.D. . Techno‐economic Feasibility of Large‐Scale Production of Bio‐based Polymers in Europe. Husing. Properties. . accessed on August 19.J. Schleich. Sci. Applications. Weith: Plastics. the Food Biopack Conference Proceeding (Ed. 2005. London. A. .. Prentice‐Hall. Frederiksberg. 27‐40. 1984. 2003. 557‐562. Whistler.M. in Starch: Chemistry and Technology (Ed. Starch/ Starke 2001. demand and recovery for 2006 in Europe. Strong: Plastics: Materials and Processing. R. Washington DC. S. Marscheider‐Weidemann. Properties. USA. 1990.L. John Wiley & Sons. 1998. in Kirk Othmer Encyclopedia of Chemical Technology Volume 22. Hoboken. Academic Press. 11. R. Crank. Cleaner Prod. . 1996. 53 . A. USA. Culter: Plastics Packaging. 1939. 2000. S.htm. J. Polym. Chandra.. Lancaster. Fifth Edition (Ed.A. R. Bastioli: Global status of the production of biobased packaging materials.E. Ren: Biodegradable plastics: a solution or challenge?. USA: http://www. . New Jersey. 1273‐1335 . Eng. C. F. Plastics Europe. Ehrenstein: Polymeric Materials: Structure. Association of Plastics Manufacturers. . The compelling facts about plastics: An analysis of plastics production. G. 2008.epa.351‐355. Rustgi: Biodegradable polymers. Patel. . Brussels. O. J. Englewood Cliffs. R. . C. Hernandez.gov/garbage/plastic.L.. M.
G. Sci. X. Physical and mechanical properties of biobased materials – starch. report of Food Biopack Project (Ed. 2005. epoxy functions and stearoyl chloride. Muller. X. Starch/Stärke.F. 2005. K. P. Interfaces 2006.M. Smits. G. 1996. Curvelo. 2003. Carbohydr. Denmark. P. van Soest. K. J. Eng. Rev.H. J. interaction. Avérous. . 53. Polymer Sci. Seville.B. 494–504 .S.A. 1997. 13. Macromol. L. P..M. R. Y. 1980.L.M. and Pukanszky: Thermoplastic starch/layered silicate composites: structure.. 2008. A. 21. R. 1–17 .. W. . polylactate and polyhydroxybutyrate. J. of concentrated barley starch‐glycerol‐water mixtures. Parker: Phase and glass transition behaviour. Biodegradable thermoplastic starch.F. 331–336 .H. Yu. 372–380 . Otey. A. Spain. Appl. Compos. Doane: Starch‐based blown films. 385–409 . Polym. Ind.V. 53. J. 57. properties.G.M. Wang. a model for thermoplastic starch.K. Moates. Wang: Properties of starch blends with biodegradable polymers. 34. Wang. T. Weber). C. 1491‐1494 . 53. Ind. Chambre Syndicale des Emballages en Matiere Plastique (French Plastic and Flexible Packaging Association). J. Starch/ Starke 2001. B. Mikkilti. K. J. 2000. Chem. and Couturier: Properties of biodegradable multilayer films based on plasticized wheat starch. Press Release: Major Price Hikes in Plastics. 4. Petersen. Yu. Biobased packaging materials for the food industry. Dev. 2003. and Olsen. J. Res.J. N. status and perspectives. Vliegenthart: Interaction between dry starch and plasticisers glycerol or ethylene glycol. Part C: Polym. Frederiksberg C. 356‐361 . Crops Prod. Gao. . 2005. F. P. Forssella. Nielsen. O. Introduction European Commission Directorate General of Joint Research Centre. July 15. Starch/ Stärke 2001. 409–416 . M. Carvalho. Prod. A Real Case of “Force Majeure” that should be passed through to Packaging Prices Immediately. J. measured by differential scanning calorimetry and solid state NMR spectroscopy. Y. Gandini: Surface chemical modification of thermoplastic starch: reactions with isocyanates. the Royal Veterinary and Agricultural University.P. 592‐595. Martin. A. Ma: The effects of citric acid on the properties of thermoplastic starch plasticized by glycerol. Schwach.Lin: Note.J. Department of Dairy and Food Science. 62. 275‐282 19 . Carbohydr. 43. E. Westhoff. Polym. A. Bagdi. Yang. J. . Kruiskamp.
F. Thiebaud. Feil: Morphology and enzymatic degradation of thermoplastic starch– polycaprolactone blends. Inc. Qiao. Pacsu: Starch studies: Preparation and properties of starch triesters. Bikiaris. Crops Prod. C. Ind. T. 1647‐1656. K. H.P. C. Environ. Inc.H. 1986. Wurzburg. Ramsay. 1986. J.T. Appl. Sagar. Wurzburg).B. 35. I. B.A. Degrad. O. in Modified Starches: Properties and Uses (Ed. B.D. C. O. 11–17. . V. Wurzburg: Cross‐linked starches. Chem. Appl. 381‐384 . Mullen. destructuration and hydrolysis during O‐formylation.J. USA. Eng. A. Feller. . J. 1209‐1217. W. 74. A. Coutuier: Comparative biodegradation study of starch‐ and polylactic acid‐based materials. Aburto. Ind. 44. 55‐78. CRC Press. W. J. Journal of Applied Polymer Science 1995. 1942. 20 . Sci. Polym.B. .W. E. Ribeiro. Polym. 1997. J. J. Prinos. Boca Raton. Panayiotou: Processing and characterization of starch/polycaprolactone products. Polymer 2003. 65. Y. Grohens: Starch modifications.W. Gattin. (editor).Chapter 1 . . 2006. Ramsay.. Y. J. 46–51 . Andrade: Reactive processing and mechanical properties of cross‐linked maize starch. 389‐398 .A. 58. CRC Press. Sun: Reinforced thermoplastic acetylated starch with layered silicates. Alric. Mullen.W.A. 1517–1526 . Bertrand. F. Carvalho. chapter 4 in book Modified Starches: Properties and Uses. E. Rodriguez‐Gonzalez.C. Myllarinen. Favis: Rheological and thermal properties of thermoplastic with high glycerol content. 58 (9). 57. 2004. Polym. Silva.A. 56. Polym.J.C.W. Divers. Hulleman.D. P. Ibezim. G. A. Rodriguez‐Gonzalez. Copinet. 139–147 . D. T. Polym. Pilin. M. Borredon. Eng. Favis: High performance LDPE/ thermoplastic starch blends: a sustainable alternative to pure polyethylene. 1943.B. Boca Raton. Kontoyiannis. 581–586 .D. I.. Merrill: Properties of fatty‐acid esters of starch. Matzinos. Starch/Stärke 2005. B. Sci.F. O. S. J. X. Tserki. Ind. Jiang. 2002. 24. E. 1999. R. S. M. 2001. Pacsu: Starch studies: Possible industrial utilization of starch esters. Carbohydr. Stab. 17–24 . J.D. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. C. P. E. 705‐721. 77. 34. USA. van der Zee. 2594–2604 . B. 2004. . C. 1. Vikman. E. Levesque. M. Chem. Jarowenko: Acetylated starch and miscellanous organic esters.
Munich. Nakayama. Rijksuniversiteit Groningen.E. . J. . 5590‐5598 . D. USA.V. E. W. 145‐152. Germany. 1440‐1451. in Reactive Polymer Blending (Ed. I. Duquesne. A. Thiebaud. . de Graaf. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Hu).K. C. 56. 33‐43 21 . de Graaf: Ph. Macromolecules 1996. A.C. Lancaster. Sci. and biodegradability of fatty‐acid esters of amylose and starch. Starch‐Starke 2004. . characterization. Technomic Publishing Company. J. 51. US Patent 2. Beenackers: The kinetics of the acetylation of gelatinised potato starch. L. G. Prinos. M. I. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. S. 1960 . H. J. 137‐144. P. . J. D. 255‐263. Polym. Alric. E. A. Aburto. W. . Janssen. R.. 57. 175. Dubois: Synthesis and characterization of compatibilized poly(ε‐caprolactone)/ granular starch composites. Aburto. D. R. Inc. I.F. Polym.W. Hanser Publishers. 2001. New York. J. A. Alric. J. . Bikiaris. 28. 51. Carbohydr.Stokes. T. Marechal. . Aburto. . Alric. Guegan. P. A.828. Inoue: Compatibilizers for melt blending: Premade block copolymers.Coleman. 101‐112. F. S. Smith. Mouysset‐Baziard. Tuschoff: Acylation of hydroxy compounds with vinyl esters. Thesis. Panayiotou: Synthesis. Alric. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. B. Senocq. 1991. USA. Khandpur.E. Broekroelofs. R. 34. Panayiotou: Properties of octanoated starch and its blends with polyethylene.H. C. W. . Appl. Hu. E. Introduction. P. J. Introduction . Degee. A. Borredon. R.Graf. Symp. Bikiaris. J. 29. Thiebaud. E.J. Macromol. C. . I. P. 1996 . 1997. Starch‐Starke 1999.928.E. 11. A straight way to regioselectively functionalized polysaccharide esters. Starch‐ Starke 2005. P. D. E. E.D. Bakker and G. Starch‐Starke 1999. 1995. Mormann. C. 132‐135. 74. M. M. Cellulose 2004. Hamaili. Aburto. Wiley‐VCH.E.Fennel Evans: Fundamental of Interfacial Engineering. 118‐121. Al‐Higari: Acylation of starch with vinyl acetate in water. Rutot. I. 302‐307. Dicke. Scott. C. Bakker. Alric. J. C. Carbohydr. Borredon. Aburto. G. G. Macosko. 1997. 1999. J. A.M. Prinos. M. A.H. 2001.Painter: Specific Interactions and the Miscibility of Polymer Blends. Polym. P.
I. U. R. Bikiaris.A. Pavlidou. Testing. Mani. J. Cordi.G. Koutsopoulos. and tensile properties of Low‐Density Polyethylene (LDPE) and corn starch blends. 647‐655 22 . 1991. Polvm. Polymer. Effect of amylopectin to amylose ratio in starch. 88. B.‐J. Rosa: Mechanical. 1503–1521 . 162–163. Adv. N. J. Verhoogt: Processing and characterization of thermoplastic starch/polyethylene blends. Appl. Eur. Ramsay. Bikiaris. N.D. 1242‐1251 . L. Degradation studies and moisture uptake of sago‐starch‐ filled linear low‐density polyethylene composites. 2005. Nikolov. R. morphology. Danjaji. 16. Favis.M. Degr. Evangelista. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends. Rosa: Effects of the compatibilizer PE‐g‐GMA on the mechanical.S. 75–81 . A. Ishiaku. J. 1998. Appl. Mater. 2002. Duran. 1‐9 . Polym. Z. Carbohydr. Mohd Ishak: Material behaviour. 34. L. D. 30. W. D. L. N.A. Panayiotou: LDPE/plasticized starch blends containing PE‐g‐MA copolymer as compatibilizer. L. Liu. Ishiaku. I. Mohd Ishak: Material properties.S. 287‐291 .A. 236–241 . A. Proc. 1997. Panayiotou: LDPE/Starch blends compatibilized with PE‐g‐ MA copolymers. Tech.S. Almeida. Sci. Nawang. 2005. Chem. R. Z. Eng. Vouroutzis. J. W.M. Ind. Res. G. Gelina. J. Pedroso. Ismail. Mei: Study and development of LDPE/starch partially biodegradable compounds. Polym. Polym. 2003. C. Ribes‐Greus: Thermal and viscoelastic properties of some commercial starch products. 59. R. Mechanical properties of sago starch‐filled linear low density polyethylene (LLDPE) composites. Polym.J. Technol. Appl. R. Pedroso. 88. Nakamura. . 38.D. III.Chapter 1 .G. 310–317 . Wang. 21. 1998. K. U. J.H. 70. B. Z. 2904–2911. 59. thermal and morphological characterization of recycled LDPE/corn starch blends. . E. 2005.G. Danjaji. H.S. Stab. 1841‐1846 . 167–172 . I. C. Polym. J. Polym.L. Prinos. Sci. Ismail. Polym. Nawang. Z. D. J.S. Jane. Effect of compounding and starch modification on properties of starch‐filled Low‐ Density Polyethylene. Y. Frangis. St‐Pierre. Sun: Effects of polyethylene‐grafted maleic anhydride (PE‐g‐MA) on thermal properties. N. Testing.A. H. 20. Sci. E. A. M. Contat‐Rodrigo. 1467‐1475. Sung. Ramsay. 1998. thermal and morphological properties of virgin and reprocessed LDPE/corn starch blends. Polym.D. H. 2003.. 2001. D.
U. 2000. J. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. Tang. Appl. J. 157. M. Sankar. J. Lindblad. Otey. J. Low density polyethylene/ starch blend films for food packaging applications. B. Westhoff: Starch‐Based Films. A. 2004. 2003. 2004. F. Res. 3976–3987 . Bhattacharya Residual stresses and density gradient in injection molded starch/synthetic polymer blends Polymer 2000.G. Fringant: Properties of thermoplastic blends: starch–polycaprolactone. 1999. 23. Sen. R. 4157–4167 . Dole. Chem. Tech. 1. Park: Modification of aliphatic polyesters and their reactive blends with starch. E. 23 . J. 2007. 42‐65. F. Stridsberg. A. Polym. Adv. Ranucci. Polym. Raj.Y. Maiti: A biodegradable film. Manan. 23. 1984. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. Rodrigues. Pilot plant investigation for production of the biodegradable film. A‐C. Chem. Part A: Polym. Introduction . C. Kammer: On mechanical properties of sago starch/poly(ε‐caprolactone) composites. T. L. . Siddaramaiah. Manan. 1997. Odusanya. Sci. . 2003.P. C. Z. 877–884. J. M. 32–45 . 41. Environ. Jana: Studies on the properties and characteristics of starch– LDPE blend films using cross‐linked. M. D. 284‐287 . A. 2002. 9177–9190 . Eng. Guedes. P.N. 1139–1148. Odusanya. Jung. C. 1659‐1663 .M. Makromol.S. N.C. Adv.‐C. Kim. 1507– 1516. . Sci. 12. Sci. Karlsson: Polymers from Renewable Resources. Albertsson. S. Averous. S. Sci. . Eng. B. Appl. Ishiaku. Sci. 43. T. Garg.S. . Polym.H. J. 40. M. S.K.W. Cho. O. Liu. Otey. Preliminary diffusion evaluation. Polym. K. Calil: Effect of thermal aging on the biodegradation of PCL. Kim.. M. . U. B. glycerol modified.M. 1298‐1305. Kim. Sci. K. 2. O. 179‐187. Chem. M. Polym.H. PHB‐V. Polymer 2000. 16–19 . Y. Doane: Starch‐based blown films. Ind. Res. D. 3539–3546 .. J. dos Santos Rosa.K. Eng. cross‐linked and glycerol modified starch. J. S. Westhoff. Sci. Ind. Polym. Eur. Dev. Appl. Ryner. 139‐161.M. 26. 2002. Albertsson: Controlled Ring‐Opening Polymerization: Polymers with designed Macromolecular Architecture. 41. Moro. C.. in Polym. Polym. 1987.F.P. John. Chem. H. 89. Azemi. Azemi: Effect of starch predrying on the mechanical properties of starch/poly(e‐caprolactone) composites.K. B. Jana.M.D. U. J. 157. A. W. S. Ang. Yang. Ishiaku. and their blends with starch in soil compost. Adv. R. 35. Polym.H. 87. M. 81.S. K. 2001. Polym. L. 267.H. Prod.S. J.M.R.
Sadocco: Preparation of biodegradable polyesters/high‐amylose‐starch composites by reactive blending and their characterization. Composite Interfaces. J. 83. Rimedio. 19. Rimedio: Polymer communication: Preparation and characterisation of compatibilised polycaprolactone/ starch composites. Tang. Rutot. R. Mani. P. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. Errico. 215–225. 41. R. Polymer 1999. P. R. Laurienzo. Symp. Polymer 2000. and P. Polym. Duquesne. P. . . M. 3875–3881 . Degee. Dubois. M. Bhattacharya: Synthesis and characterization of starch‐ graft‐polycaprolactone as compatibilizer for starch/polycaprolactone blends. 7. Rutot. 2003. 1432–1442. 33–43 . Avella. M. Sci. Macromol. M. M. Polym. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. Errico. 175. 24 . R. Appl. Krishnan.Chapter 1 .E. Degrad.E. C. Martuscelli. E. Macromol. Dubois: Aliphatic polyester‐grafted starch composites by in situ ring opening polymerization. P. J. Rapid Commun. 2002. Degée.S. Narayan. 2000. Dubois: Synthesis and characterization of compatibilized poly (e‐caprolactone)/ granular starch composites. 283–286. Avella. . 3091‐3100. R. E. Raimo. . 2001. P. D. M. P. 40. M. Stab. 127–134. 1998. D. 80.
4‐di‐O‐isopropylidene‐α‐D‐galactopyranose)/Al(OiPr)3 initiator system to yield polydioxanone with a protected monosaccharide end‐ group is described. Systematic studies showed that the polymer yield is a function of the reaction temperature and the reaction time. The products were synthesized at 60‐100°C and characterized by 1H‐ and 13C‐NMR. also polydioxanone with an OiPr end‐group was formed (20‐30 %). polyesters.3. Keywords: biodegradable. with higher temperatures (100°C) leading to lower yields. 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. ring‐opening polymerization . Besides the desired polydioxanone functionalised with a monosaccharide end‐group. 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).2. and MALDI‐TOF mass spectrometry.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.
are interesting polymers because of their good product performance and biodegradability . The relatively low melting point of polycaprolactone limits its applicability. Polydioxanone has a tensile strength close to 48. has interesting product properties compared to other aliphatic polyesters. only limited information about the synthesis and properties of polydioxanone is available in the open literature. Despite its good properties.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. which is a unique compromise between application and processing temperature. quinine. 2). probably because the p‐dioxanone monomer has become commercially available only recently . such as polycaprolactone. The synthesis of protected monosaccharide end‐capped biodegradable polymers is usually performed via Ring Opening Polymerization (ROP). or zinc lactate has been reported. The catalysts are metal alkoxides with Lewis acidic character [8. Besides 26 . This chapter describes experimental studies on the catalytic ROP of p‐ dioxanone in the presence of a protected glucose molecule (1. Several polymers with bioactive alcoholic and phenolic end‐groups of interest for drug‐related applications were synthesized (for example geraniol. particularly for nano‐ encapsulation systems for drug delivery . The ring‐opening polymerization (ROP) of p‐dioxanone using metal catalysts such as aluminum isopropoxide [Al(OiPr)3]. stannous octoate [Sn(Oct)2]. and polylactides. testosterone. reaction of the metal‐alkoxide with an appropriate alcohol allows the synthesis of end‐capped poly‐lactones. This melting point is considerably higher than typically found for polycaprolactone (60 oC) and lower than that of polylactides (at least 175°C). pregnenolone. polydioxanone also shows good performance. stigmasterol and ergocalciferol)  and also with protected monosaccharides (galactopyranose/ glucofuranose) [6‐7]. and is tougher than polylactides and even HDPE .1.3 MPa and an elongation at break of 500‐600%. which to the best of our knowledge is the first study to functionalise polydioxanone with a monosaccharide. Introduction Aliphatic polyesters. Nishida et al  reported that polydioxanone decomposes to non‐toxic gases (CO2 and H2O) by microorganisms. accessible by the polymerization of p‐dioxanone (1). whereas the high temperature for polylactides results in thermal degradation and undesirable transfer reactions during synthesis and processing steps .4‐dioxan‐2‐ one)). tocopherol. polyglycolide.2.3. Exchange of the alkoxide group by e. The alkoxide group will end up as an ester end‐group in the polymer and in this way at least one of the end‐groups may be easily controlled and varied. Polydioxanone (poly(p‐dioxanone) or poly(1. It is fully degraded in the body within a period of 180 days . 9]. Its melting temperature is close to 110 oC. From a biodegradability point of view.Chapter 2 2. Biodegradable aliphatic polyesters end‐capped with sugar molecules have been studied extensively for use in biomedical applications.g.
Germany) was purified according to the procedure described by Raquez et al [2. Materials p‐Dioxanone monomer (1. As such. 3].g. Materials and Methods 2. CDCl3 was obtained from Sigma and was used as received. This stock solution was prepared by adding 4. The polymerization was allowed to proceed for 16 h at 100 oC. 97% (Sigma) and aluminum isopropoxide. which was pre‐heated till about 60°C to obtain it in a liquid state.3. 2. part of this solution (350 µl. (20 mmol) of aluminum isopropoxide to 20 ml of toluene. NMR) due to the poor solubility of the products in standard organic solvents.2. The hot solution was 27 . 2. 5. 98+% (Aldrich) were used as received.1. relatively low molecular weight compounds reported in this study may be viewed as model systems for more complex. After the pre‐determined reaction time. heptane (Acros).2.08 mmol aluminum isopropoxide) was added to pure 1 (0.55 mmol of 2 and 0.1.3. Typical example for the synthesis of polydioxanone end‐capped with 1. the results of this study are also of interest for the preparation of starch/polydioxanone polymers using ROP.5 mmol) was dissolved in toluene (1 ml) at 50 oC.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2). To this solution.2. poorly soluble heterogeneous systems. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System potential applications in biomedical products. The reaction mixture was clear and colorless during the reaction. the mixture was brought to room temperature and several drops of HCl (1 N) were added to stop the reaction.2. grafting efficiencies are difficult to determine in this system by standard analytical techniques (e.8 ml of a solution of an aluminum isopropoxide stock solution was added. and diethylether (Labscan) were used as received. 2. Boehringer Ingelheim.2. Methods All polymerization experiments were carried out under a protective nitrogen atmosphere using standard Schlenk‐ and glovebox techniques.2.44 g. However. 1. The resulting clear solution was stirred for 2 h at 50 oC.07 g. 8 mmol). the synthetic pathways and the soluble. containing 0. Toluene (Labscan) was dried and stored on molecular sieves 3 Å (Labscan) under nitrogen. hot dichloroethane (20‐25 ml) was added to completely dissolve the partly solid polymer at room temperature.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2) 2 (1. Next.8 g. 0.2.2. Analytical grade dichloroethane (Labscan). Particularly interesting in this field are starch polymers grafted with polydioxanone. Subsequently.
1H‐ and 13C‐NMR spectra were recorded on a Varian AMX 400 NMR. 2.3. the above equation simplifies to Xn theo = monomer conversion × [monomer ]0 [ sugar ]0 (2. 1 calculated by comparing the peak area of characteristic end‐group protons with that of a proton of the repeating unit in the polymer (A H‐polymer). As will be shown later. 2D‐NMR spectra were recorded on a Varian Unity 500 NMR. two polymers with different end‐groups are present. it is assumed that all available initiator is used effectively. 4:1 by volume) at 4‐8 oC. 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. If the amount of sugar is in excess with respect to the aluminum catalyst.) H-isopropoxide end group 28 .2.2. Calculation of Average Degree of Polymerization The Theoretical Average Degree of Polymerization. The white solid was finally separated from the liquid by decantation and dried in a vacuum oven (5 mbar.Chapter 2 precipitated in a heptane/ ether mixture (300‐400 ml. Processing of the raw data was performed using VNMR software. one initiated on a galactopyranose molecule and the other on an isopropoxide group.) Here.2. 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. The isolated yield at this condition was 68%. Product Analyses NMR analyses were performed in CDCl220.127.116.11. 2.) H‐NMR was applied to determine the experimental average degree of polymerization. This leads to the following equation: Xn exp X n exp is = A A H-repeating unit H-galactopyranose end group +A (2. 40 oC) until constant weight.
an Al ‐ monosaccharide 2 ‐ monomer 1 mol ratio of 1: 6.). Typically. an excess of monosaccharide 2 on Al was used (see eq.5. is rewritten as: Xn exp = [ A 3. To compensate for this effect. Schematic representation of the polymerization reaction including atom numbering scheme 29 .6 ppm A 5.4− 4.) or.3.6: 96 was applied. the protected mono‐saccharide 2 reacts with Al(OiPr)3 to form the actual catalyst for the polymerization reaction. Results and Discussions 2.7.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose (2) and Al(OiPr)3 as the catalyst precursor (Scheme 2.).1 ppm (2. R gp . Screening Experiments Initial experiments to synthesize protected galactopyranose end‐capped polydioxanone (3) were performed using 1. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System The NMR peaks of the repeating unit overlap partially with those of the galactopyranose end group. 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. in term of the NMR resonances : R gp = A 5. The exchange reaction typically takes place at 50°C for 2 h.4. The polymerization reaction was performed as a two step process.7 ppm − 6 × A 5.) 18.104.22.168. is calculated using: R gp = A H − galactopyranose end group H − isopropoxide end group A (2.6.) The ratio of the two different types of polymers (either end‐capped with galactopyranose or an isopropoxide group).6 ppm + A 5.1.2. To avoid the formation of isopropoxide end groups. In the first step. 2.6 ppm ] / 6 A 5.1 ppm (2.1. eq. 2.3.
Proton resonances of the CH3 group of the isopropoxide end‐group.67. 1 30 .3. Typical isolated product yields are 68% at these conditions. the in‐situ formed catalyst was reacted with p‐dioxanone monomer 1 at 100 oC for 16 h..1 ppm compared to the monomer. It confirms that polymer initiation not only occurs with the monosaccharide but also with the remaining isopropoxide group of the catalyst precursor (vide infra). of interest is the presence of a small multiplet at about 5. This multiplet is characteristic for the CH proton of an isopropoxide end‐group.1. 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.15‐1. Product analyses The products were analyzed using 1H‐ and 13C‐ NMR and MALDI‐TOF. The off‐white solid reaction product was collected after a dissolution‐reprecipitation process using dichloroethane and a heptane/ diethyl ether mixture.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 (2.2.) are not particularly informative. 2. It is difficult to determine the end‐groups of the polymer on the basis of 1H‐NMR. H‐NMR spectra (Figure 2. respectively.40 ppm) unit are broadened and shifted up to 0.78 to 4. 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 (3.1 ppm.) In the next step.1.1 and 2. The OiPr : 2 end‐group ratio for the standard experiment was 1: 2. Typical 1 H‐ and 13C‐NMR spectra of 3 are shown in Figure 2.51 ppm. Although the spectra clearly indicate the presence of the monosaccharide 2. However.7.1. although it is evident that p‐dioxanone polymerisation occurred.
2 and 96.8. 13 31 . For instance.17 and 69.3. 1. 1 (c). characteristic resonances of an ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end group are present at δ = 22.214.171.124 ppm. 2 (b).2.52 ppm (β’). 1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.1. Particularly the C2‐C6 carbons in the range δ = 66. 3 1 C‐NMR (Figure 2. Carbon resonances arising from the monosaccharide end‐group are present between δ = 66.3 ppm are shifted considerably. C‐6 is shifted from δ = 62.) is more informative and clearly shows the presence of a polydioxanone polymer backbone and a monosaccharide end group. H‐NMR Spectra of: (a).28 ppm in 2 to δ 66.2 ppm in product 3. In addition to the mono‐saccharide end‐ group. The carbon resonances of the polydioxanone backbone are present at δ = 63.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone. p‐dioxanone monomer.61 (γ’) and 73.2 ppm.2‐70. 68.2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.
the numbering scheme of carbons and protons is given in Scheme 2.2.97 ppm (‐CH‐(CH3)2).76‐25. 1 (c).2. 1. D‐NMR (HSQC) was applied for complete peak assignment of the product.97 (>C‐(CH3)2). and 2.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.48 ppm (‐CH‐ (CH3)2) and between δ = 21. 2 (b).1.3. 1. C‐ NMR Spectra of: (a). p‐dioxanone monomer.70 (>C‐(CH3)2) and between δ = 21.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.3. The resonances of the protecting group of the sugar appears at δ = 108..76‐25.3.Chapter 2 Figure 2. A typical example of a part of the 2D‐NMR spectra is given in Figure 2.1. 3 13 Carbon resonances of the isopropoxide groups from the by‐product (isopropoxide end‐capped polydioxanone) are present at δ = 68. 2 32 .2.2.73 and 109. An overview of the data is given in Tables 2.
54 4. 33 .40 all values were determined using 1H NMR.301 3.50 4.3. which were determined from HSQC spectrum due to overlapping resonances. except values with *.74 3.8 Reactant peaks (ppm) Galactopyranose (2) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 5.83 p‐dioxanone Monomer (1) H‐α H‐β H‐γ a 4.16* 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 5.60 4.68 4. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.17 3.09 4.77 3.59 4.1.732 broad.00 4.25 3. HSQC 13C‐1H spectra of 3 Table 2.26 3.78 4.32 * 4. around 1.31 4.
11/ 170. The difference in molecular weight between the main peaks is 102 g/mol.75 70. 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 96.31 70.2 73. An example of a MALDI‐TOF spectrum of 3 recorded in a dithranol/NaI matrix is given in Figure 2.07 Reactant Peaks (ppm) Galactopyranose (2) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 96.21 NMR analyses also allow calculation of the molecular weight of the products. 34 . A typical molecular weight distribution is observed.28 p‐dioxanone monomer (1) C‐α C‐β C‐γ C‐carbonyl 61.17 69.8. which is the molecular weight of a repeating dioxanone unit. The product obtained at screening conditions (16 h reaction time at 100°C) contains on average 13 monomer units. The molecular weight distribution of the major peaks may be represented by the following relation: m / z = 23 + 260 + 102n (2.2. the ratio of the intensity of the end groups and the polymer backbone peaks is determined.54 166. A detailed procedure is given in the experimental section.80 61.52 63.28 70.00 171.21 70.92 66.09 65.92 68.Chapter 2 Table 2. The latter likely stems from the matrix used to ionize the sample.61 170.63 68.33 70.) This series represents a dioxanone polymer end capped with 2 and an additional Na ion.2 68. MALDI‐TOF was also applied to characterize the products. For this purpose.10 71. corresponding with an average number molecular weight (Mn) of about 1600.58 68.57 62.4.
Typical MALDI‐TOF spectra of 3 A second molecular weight distribution is also clearly visible when enlarging the spectra. two other distributions are present.4. m / z = 23 + 18 + 102n m / z = 1 + 18 + 102n 35 (2. ionized with either Na+ or H+.5. in line with the NMR data.) and (2.11. These distributions may be represented by equation (2. although both with a very low intensity (Figure 2.5.) and imply the presence of dioxanone polymers with carboxylic end groups.11.5 +102 1400 1600 1800 2000 2200 2400 Mass (m/z) Figure 2. This series may be described by the following relation: m / z = 23 + 60 + 102n (2.) (2. Furthermore. see Figure 2.10.).) This relation is indicative for the presence of a polydioxanone polymer containing an isopropoxide end‐group.9. 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 1098.10.) . for details.
2.7 796. In addition.7 %-Intensity 30 20 858. different molecular weights were analysed.Chapter 2 50 40 792. according to NMR.2. Enlarged MALDI‐TOF spectra for 3 Although MALDI‐TOF clearly demonstrated the presence of various types of end‐groups in the product. The observed differences in the molecular weight distributions of the various samples were only marginal. In the first step.g. the catalyst precursor is treated with monosaccharide 2 resulting in an alcohol exchange 36 .1.3. but in all cases poor quality. small but significant amounts (20‐30%‐mol) of polydioxanone chains with an OiPr end group are present.5. 2‐(4′‐ hydroxybenzeneazo)benzoic acid (HABA)). Various other matrices were tested (e. it proved not suitable to determine the average molecular weight of the products. Mechanistic aspects Both NMR and MALDI‐TOF measurements imply that the main product is indeed the desired monosaccharide end‐capped polydioxanone. 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. low resolution spectra were obtained. Various samples with.3 10 836. is not particularly suitable for polydioxanone. [6.6.1 0 600 700 800 900 Mass (m/z) 1000 1100 1200 Figure 2. although successfully applied for galactopyranose‐end‐capped polycaprolactone . Most likely the matrix (dithranol/NaI).13].
(R = monosaccharide 2) 37 . Irreversible termination of the polymerisation is performed at the end of reaction period by adding dilute acid to the polymerisation mixture. The termination reactions lead to the formation of ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end groups. During the reaction. Subsequent coordination and insertion of dioxanone molecules leads to the formation of a polymer chain with a monosaccharide end‐group. 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. Next. termination of the chain growth may occur by reaction with an alcohol.6. The resulting Al‐alkoxide may again initiate a polymerisation reaction. The termination reaction is known to be reversible and the formed polymer may again react with an aluminum center and continue to grow . 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. Simplified reaction scheme for the ROP of p‐dioxanone catalyzed by Al(OiPr)3. a p‐dioxanone molecule will coordinate to the Lewis acidic aluminium center followed by an insertion step.
5 1.30 62.74 52.50 25. A total of 15 experiments were performed at two polymerization temperatures (60 and 100 oC).3.1 86.3 and 62.5 84.50 35. 2.6 6.6 89.5 92. However.5 1.Chapter 2 The minor product.3 6.).5 67. These may be present in the reaction mixture because the exchange reaction between Al(OiPr)3 and 2 was incomplete.9 2. An overview of the experiments and the results are given in Table 2.1 2. Table 2.6 96.3 6.3 81.12 42.22 16. OiPr‐end‐capped poly(p‐dioxanone). In all cases.74 18.04 19.6 6.8 3.7.6 6.4 2/OiPr ratio Rgp (mol/mol) 2.3.6 was applied.5 81. formed in the first step of the polymerisation reaction (eq. degree of polymerisation of the product and the end group distribution was determined.81 16.9 1 7.6 6.10 15.08 30.55 Product Properties Avg.78 10.6 6.6 6.6 86.6 91.6 t (h) 16 16 16 16 16 16 16 16 16 16 1. Systematic Studies The effect of important process variables (temperature.6 2.3‐6.3.21 54.2.3 3.6 80.96 2 3 a For each experiment.5 126.96.36.199 13.29 14.74 23. will be formed when the polymerization starts with an aluminum alkoxide with a remaining Al‐OiPr group (eq.16 46.93 37.11 22. a (nearly) fixed Al(OiPr)3 : monosaccharide ratio of 1 : 6. 2.6 81.8 2.5 1.).36 17. time and the mol ratio of monomer to monosaccharide) on the yield.6 6.53 59.22 13.7 2.4 3. two different reaction times (1.85 14.22 57.3 6. Set Sample S111 S112 S113 S114 S115 S211 S212 S213 S214 S215 S221 S222 S223 S224 S225 Overview of experiments a Processing Condition 2/Al ratio (mol/mol) 6.8 2.70 58.45 19.28 33.93 8.4 2. Chain Length (Xn exp ) Yieldb (%) 30.5 1.5 hrs and 16 hrs) and a p‐dioxanone to monosaccharide 2 molar ratio ranging between 3. 2.5 T (°C) 100 100 100 100 100 60 60 60 60 60 60 60 60 60 60 dioxanone/ 2 ratio (mol/mol) 3.1 2.6 6.4 2.3 6.3 6.2 91. 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 6.0 87. despite the excess of 2. the end group may also be formed by a termination reaction with free isopropanol.1 3.
5 h. It was shown that at lower monomer to catalyst ratios.2. It is likely that this relatively low molecular weight compound dissolves partly during isolation/ purification procedure leading to lower isolated yields.3.3. as suggested by Raquez et al . Product yield The isolated yields of the reactions are all but one between 67 and 96% (see Table 2. with an enthalpy of polymerization of approximately ‐15 kJ/ mol . the yield at 1. the monomer conversion reaches a maximum value before going down to the equilibrium monomer conversion. On the basis of our experimental data and in line with literature data. Higher polymerisation yields were obtained at lower temperatures. 39 . For instance. 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. 1. when the reaction is performed at a low dioxanone: 2 ratio (≅ 16). the yields are higher when performing the reaction at 16 h reaction times.5 h is higher than the yield at 16 h. 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. Evidently.3. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 2. Another possibility is the occurrence of polymer crystallization. see Table 2. 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. and particularly when comparing the data in set 2 and 3 (60oC. a low dioxanone: monosaccharide 2 ratio was applied. will be required. To the best of our knowledge.). 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. equilibrium conversion is not yet achieved within 1. no detailed explanation has been put forward to explain this anomalous behaviour. A more detailed analysis on the actual nature and composition of the polymerization products. Clearly.3. et al .5 and 16 h reaction time). For this particular experiment. One of the experiments (S111) resulted in a very low yield (31%) compared to the other reaction. 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. leading to a low average chain length of the dioxanone polymer (7. which reduced to 75% when increasing the temperature to 120oC. However. the yields for reactions conducted at 60 oC are always higher than those performed at 100 oC. 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.). the equilibrium conversion of dioxanone at 80oC was about 80%. For the two experiments with a dioxanone: 2 ratio higher than 16 (≅ 23 and 60).1. which is expected to be more pronounced at lower temperatures.3.2.
in line with the theoretical predictions.6.2.5 hr Xn theo theo (90%-conversion) (100%-conversion) 40 Xn exp Xn 30 20 10 0 0 10 20 30 40 50 dioxanone : 2 ratio (mol/mol) 60 70 Figure 2. time: 1.7. 2.) as a function of the dioxanone : 2 ratio at 90% and 100% monomer conversion is also provided. time: 16 hr 50 Temperature: 60oC. to reduce the number of OiPr endgroups in the product may be the removal of isopropyl alcohol formed in step 40 . The R gp values (Table 2.2. The X n exp increases linearly with respect to the dioxanone : 2 ratio. 2.Chapter 2 2. A clear trend between R gp and the process conditions (temperature.5. In Figure 2. 60 Temperature: 100oC.4 and 3. time.3.7. as expected for a typical ring opening polymerisation . time: 16 hr Temperature: 60oC. Effects of process conditions on the average chain length ( X n exp ) and end group distribution The effects of the p‐dioxanone: 2 ratio.. Most experimental points are scattered along these lines. the X n theo (eq. and 2.).e.2. and dioxanone/ 2 mol ratio) is absent. reaction time and temperature on the X n exp of the products is shown in Figure 2.3. 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.7. A possible strategy to increase the Rgp values i.) for all experiments are scattered randomly between 2.3.
977 (with an adjusted R2 value of 0.3‐6.8. Statistical Data Analysis Quantification of the influence of the experimental factors (temperature.2. indicating that the model describes the experimental data reasonably well.g. Table 2.4.6) was too limited to draw sound conclusions.4. The low P‐value clearly indicates that the model is statistically significant.459*10‐7 41 . A parity plot of the modeled versus experimental values of the average chain length Xn confirms this statement (Figure 2.543 exp as a Model Error Total P‐value 3. The R2 value for the model is 0.3.854 × (dioxanone : 2 ratio) + 0.177 × (t ) − 0. vacuum distillation before adding the dioxanone monomer .3. A linear model proved adequate to describe the effects of the independent variables on the X n exp : Xn exp = 8.7.).703 F 173.12.3.) where t and T are respectively the polymerization time and temperature.49 + 0. The 2/ Al ratio was not included in the model as the experimental range (6. 2. Analysis of variance (ANOVA) for linear model of X n function of experimental parameters SS 4103 93 4010 DF 3 12 15 MS 1337 7. 2.) from the reaction mixture by e.974).097 × (T ) (2. time and dioxanone/ 2 ratio) on the X n exp has been performed by multivariable linear regression  on the data given in Table 2. The analysis of variance for the model is given in Table 2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 1 of the polymerisation process (eq.
3. The isolated yields of the off‐white solid products were between 30 and 96%.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. 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. Furthermore. 16. with high ratios leading to a higher average chain length. which is in agreement with the available data on ring opening polymerization [15. In addition. the X n exp is negatively influenced by temperature.Chapter 2 60 50 40 Xn model 30 20 10 0 0 10 20 30 Xn exp 40 50 60 Figure 2.8.4. the model allows determination of the process variables to obtain a product with the desired degree of polymerization. Conclusions The ROP of p‐dioxanone in the presence of a monosaccharide (1. This is in line with literature data [5. the model predicts that the X n exp is positively influenced by the polymerization time. 11] and due to the fact that the reaction is an equilibrium polymerization with a slight exothermic effect. 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). Both NMR and MALDI‐TOF measurements indicate that the products are mixtures of polymers and significant 42 . 17]. 2. 2) with Al(OiPr)3 as the catalyst is reported.2. Within the experimental ranges. with higher ratios leading to higher molecular weights.
Narayan.L. 21. D. 8419‐8425. 1063‐1071.5. Macromol. P.g. . Dubois: ʺCoordination‐insertionʺ ring‐ opening polymerization of 1. 153‐166. C. and mechanistic aspects of the ring‐opening polymerization of 1. Lin.O. P. 2.J. H. temperature and monomer: monosaccharide ratio) on the X n exp .C. 34. P.4‐dioxan‐2‐one. References . 43 . Degee. Narayan. 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. . D. H.4‐dioxan‐2‐ one initiated by Al((OPr)‐Pr‐i)(3) in bulk. 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. Chem. Res. Macromol. These studies will be reported in the next chapter.R. . R. Kricheldorf. J. 42. 1998. Chu. 27. Damrau: Polylactones. P. Raquez. 199.M. R.4‐dioxan‐2‐one and controlled synthesis of diblock copolymers with epsilon‐caprolactone. Moreover. 1993. 2000.6. Zn‐lactate‐catalyzed polymerizations of 1. J. kinetic. Mater. Rapid Commun. 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. Raquez. 1089‐ 1097. Biomed. A statistical model has been developed to quantify the effects of process variables (time. Dubois: Some thermodynamic. starch). Degee. Grubb: Hydrolytic degradation and morphologic study of poly‐p‐dioxanone. Macromolecules 2001. Phys.M.
. 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.
polycaprolactone. The DS of the polycaprolactone graft is between 0.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. silylation . 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. ε‐ Caprolactone is grafted to silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 in THF at 50oC. The poly‐(ε)‐caprolactone side chains consist of 40‐55 monomer units and is a function of the reagent intakes. the remainder being homopolymers of ε‐caprolactone. The grafting efficiency varies between 28 and 58%. The silylation reaction was performed using hexamethyl disilazane (HMDS) as the reagent in DMSO at 70°C. Silylated starch with a degree of substitution (DS) between 0.72.45‐0.7 was obtained. grafting.21 and 0. Keywords: starch. biodegradable polymers. 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.
Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decreased rapidly in time and were fully degraded after 4 weeks . amylose (a linear polymer of anhydroglucoses with α‐D‐1.4‐ glucosidic bonds). Starch. A wide variety of synthetic biodegradable polymers have been prepared. Polycaprolactone (PCL) is easily degraded by micro‐organisms . Chapter 3 3.1. and the mechanical properties of starch‐based films are generally inferior to those derived from synthetic polymers [4. Various modification strategies have been explored. The development of green biodegradable polymers for e. for instance grafting of monomers (like styrene and methyl methacrylate) to the starch backbone [6. is one of the potential candidates for future biodegradable polymer products. However. a natural biopolymer. Starch has also been thermoplasticized with the help of plasticizers such as glycerol and other polyalcohols . roots) as granules or cells with typical particle sizes between 1‐100 µm. in almost all cases. the used monomers and the corresponding grafted chains are not easily biodegradable. There are two types of biopolymer in starch. Well known examples are polyesters derived from cyclic lactones (polycaprolactone. Starch is present in the body of many plants (tubers. PCL has gained much interest for possible applications in the medical field as well as in the area of packaging materials [12‐ 13] Several studies to combine the properties of starch and PCL have been performed to obtain fully biodegradable materials with improved product properties. 7]. 5]. However. Introduction Worldwide.g. The polymeric structure of starch consists of repeating anhydroglucose units. The amylose‐amylopectin ratio in native as well as modified starches has a strong impact on the product properties. Starch modification is therefore needed to meet the product properties in a number of application areas. and this value increases with about 10% per year . Starch films are known to have good oxygen barrier properties. the product properties are in most cases still not up to standards and blending with other polymers is required . 245 million tons of plastics are produced per year. resulting 48 . Global production of starch is 60 million ton per year in 2004 . Starch is abundantly available. However.4‐glucosidic bonds) and amylopectin (a branched polymer with α‐D‐1. These plastics are mainly synthetic polymers from fossil resources. Blends of thermoplastic starch and PCL are not fully miscible.6‐glucosidic bonds besides α‐D‐1. the future generation of packaging materials is highly desirable. as starch is highly hydrophilic. which are known to degrade with difficulty and cause serious environmental problems . and polybutyrolactone). polyvalerolactone. it is water sensitive. The content of amylose in starches depends on the plant and typically varies between 18‐28%.
A stock solution was prepared by dissolving 1. Hydrochloric acid (HCl) 1 N was prepared from Titrisol concentrated hydrochloric acid solution (Merck) and distilled (Milipore) water. In this chapter. 0‐14%). All available data indicate that the presence of water reduces the GE. 3. ε‐Caprolactone monomer (Fluka) was dried over Calcium Hydride (CaH2) for 48 h. 49 . distilled under reduced pressure at 100 oC and stored under a protective nitrogen atmosphere.1. the ring opening polymerisation occurs solely in the liquid phase and this is expected to lead to higher GE values. during the reaction.67 gram (8 mmol) Al(OiPr)3 in 50 ml of dry toluene in a glove box. Materials Corn starch (Sigma) was dried at high vacuum (~1 mbar) at 100 oC for one day before use. thus to a reduction in the GE. To increase the miscibility of starch and polycaprolactone. a very flammable by‐product. Hexamethyldisilazane (HMDS. This approach has also been applied successfully to graft PCL and polylactide on dextran [19‐20]. In this way.2. This is expected to lead to reduced reaction rates between starch and CL compared to CL homopolymerisation. Materials and Methods 3. thus leading to the formation of PCL homopolymers rather that starch‐g‐PCL . DMSO (Acros) and toluene (Labscan) were dried overnight over molecular sieves 3 Å (Merck) and stored under a protective nitrogen atmosphere. it has been proposed to chemically graft caprolactone onto the hydroxyl groups of starch using ring‐opening polymerisation . Acros) and methanol (Labscan) were used as received. This is rationalised by assuming that water competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors in undesirable phase separation . 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]. therefore difficult to handle and releases ethane. an alternative method to synthesize poly‐(ε)‐caprolactone grafted starch co‐polymers (starch‐g‐PCL) is reported. This catalyst is extremely air and water sensitive. The highest GE (90%) was achieved using triethylaluminium as catalyst .2. leading to a liquid‐solid system. 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. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). Another possible cause for the low grafting efficiencies is the heterogenous nature of the reaction.
5 ml) and a predetermined amount of the stock solution of Al(OiPr)3 in toluene were added.60): 1 H‐NMR (CDCl3. after peracetylation. starch). The reaction was carried out at 70 oC. Silylated Starch (1. Subsequently.2. After 2 and 4 h reaction time. acetate‐CH3).2. 3. 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. After removal of the solvents under reduced pressure (0.2. Typical example of the starch silylation procedure The procedure for corn starch silylation was adapted from that published for dextran [19‐23].12 (m. the silylated starch (1) product was dried in a vacuum oven (~5 mbar. the solvents were removed from the silylated starch product under reduced pressure (~ 20 mbar) at 70 oC.6 mL) at 50 oC (1‐2 h).2.2. This procedure was repeated three times. broad peaks. Silylated Starch (1. DS = 0. Sample SN‐3. Typical example of in situ polymerization of ε‐CL with silylated starch The silylated product from the first step was dissolved in THF (0. starch). 1 3. The samples were characterized by 1H‐NMR. before peracetylation. 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. Traces of DMSO trapped in the product were removed by dissolving the product in a small amount of toluene and re‐precipitation in methanol. 80 oC). The pre‐determined amount of HMDS (typically 24 ml. For each experiment. 40 oC) until constant weight.1. silyl‐CH3). 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.111 mol) was added to the gelatinized mixture to initiate the silylation reaction. 1. 0. toluene (40 ml) was added to solubilize the precipitated (partially silylated) starch. After 24 h.2. DS = 0. 50 oC): δ 0.2.60): H‐NMR (CDCl3.12 (m. 3‐6 ppm (m. THF (4. 50 oC): δ 0.5 (m. another portion of toluene (20 ml) was added. After 6 h. Sample SN‐3. The white‐to‐transparent solid products were stored under vacuum in a desiccator at 6‐8 oC. a predetermined amount of ε‐caprolactone monomer (molar ratio of monomer to 50 .1 mbar. To this homogenous solution. 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. broad peaks. 3‐6 ppm (m. Chapter 3 3. A mol ratio of ‐OH groups to catalyst of 10:1 was used.7‐2. silyl‐CH3).
the desilylated starch‐g‐polycaprolactone product (3) was precipitated using heptane. broad peaks. Silylated Starch‐g‐PCL (2. α‐PCL). After stirring for 2 h. The total isolated yield at this condition (see Table 3. broad peaks. β and δ‐PCL). ‐CH3 .88 ppm (m. 3.5‐ 3. γ‐PCL). DSPCL=0. 60 oC): δ 0. iPr). 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). starch).34): H‐NMR (DMSO. The silylated starch‐g‐PCL (2) product was precipitated from the solution by the addition of heptane (about 250 ml) at ‐18 o C.75. the peracetylating reagents 51 . Peracetylation of the remaining OH groups of modified starch is a well established procedure to improve characterisation of the products by NMR . The yield was measured gravimetrically and is based on the weight of the product and the total weight of reactants charged to the reactor. ε’‐PCL).16 (d.75. starch). 2.4 (m. 3.68. 1. The peracetylation procedure applied in this study was adapted from the literature [24. 4. iPr). 1 (0. The solid precipitate was filtered and dried under vacuum (~ 5 mbar) at 40 oC for 48 h. 3. Deprotection product of Sample SN1CL2. 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. 1.5‐3.) was >99 %. 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. Typically.31 (m.37 (t. 1. ε’‐PCL). Sample SN1CL1. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors OH‐groups was 13 : 1 for a standard experiment) was added and the ring opening polymerization reaction was allowed to proceed for 24 h at 50 oC.5. 1 3. ‐CH3.2.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until it was fully dissolved (typically 3 h). The product was collected as a white solid and characterised by 1H‐NMR.54 (m.98 (t. α‐PCL). 1. silyl‐CH3).31 (m. 3. iPr). 2.3.37 (t. 60 oC): δ 1. 1.2. 3. 3.68.12 (s. γ‐ PCL).5. 4. and 5‐5.21): H‐NMR (DMSO d‐6.2. iPr). DSPCL=0.54 (m. 4.2. Peracetylation of silylated starch Characterisation of the silylated starch by NMR proved very difficult due to the presence of very broad and overlapping resonances arising from starch.98 (t.2. DSsilylation = 0.3‐4.25 (t. Subsequently.16 (d. 1 3.88 ppm (m. β and δ‐ PCL). ε‐PCL). ε‐PCL). The samples were characterized by 1H‐NMR.4. DSsilylation = 0. 25]. 4. ‐CH. Starch‐g‐PCL (3. filtrated.25 (t.4 (m.3‐4. and vacuum dried at 40 oC for 24 h. and 5‐5. ‐CH.
This leads to the following equation: ACL = ACH 2 −ε + ACH 2 −ε ' ACH 2 −ε ' = A3. DSsilylation may be calculated using 1H‐NMR spectra of the products after peracetylation using eq. 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.1. ACH 3 silyl / 9 1 7 A−0. 3.1.6 ppm DSsilylation = × = × 3 Astarch proton / 7 27 A3−5.) 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. 3.2. When assuming that all ε‐CL monomer is converted.1.2.2 ppm A3. 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. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.8 ppm (3.3. The assumption of high conversions (>95%) was correct for all experiments (see Table 3. Calculations The DS of the silylated starch (DSsilylation) is defined as the average number of silyl groups present on an anhydroglucose unit (AHG) of starch.3. In this calculation.2.3.6−0.3−3.). The product was precipitated by the addition of methanol and washed several times with methanol.2.) 52 . 1 3.8− 4. It was finally dried overnight in a vacuum oven at 70 oC and 5 mbar until constant weight.4 ppm + 1 (3. Analytical Methods 3.) where Ax‐y stands for the 1H‐NMR peak area in the range δ x‐y ppm. it is assumed that the average length of the grafted chain is equal to the chain length of the homopolymer. the DSPCL may be calculated using eq.4. 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. Chapter 3 (DMAP.3. The peracetylation reaction was conducted for 7 h at 50 oC. 3.2.
CL/OH ratio ⎥ ⎣ ⎦ (3.4 / 2 ⎟ ⎠ ⎝ ⎝ where Ax‐y stands for the peak area in the range δ x‐y ppm.9 ⎟ × 100 % ⎜ ⎜ ⎟ ACH 2 −ε ' / 2 ⎠ A3. Although all steps have been investigated.4 + A3.9 ⎡ 0. 53 .3.8− 4 . 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.2 ) ⎤ 1 × ⎢ (3 − DSsilylation )⎥ ⎢ ε . and Vm stands for molar volume.3. Results and Discussions The overall procedure to synthesize poly‐(ε)‐caprolactone grafted starch co‐ polymers (3) consists of three steps and includes hydrophobization of starch by silylation of part of the hydroxyl groups of starch using hexamethyl disilazane (HMDS).5.3−3. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors DS PCL = 0. T stands for absolute temperature.) ⎛ ⎞ ⎞ ⎛ A A = ⎜1 − CH −iso ⎟ × 100 % = ⎜1 − 4. 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.) 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). 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. This leads to the following equation: GE = ACH 2 −ε ' / 2 − ACH −iso ACH 2 −ε ' / 2 × 100 % (3. A3. the focus of this chapter will be on the first two steps of the procedure.3−3.3−3.8− 4 .( A3.8− 188.8.131.52 − A4 . 2007) 3. The values of ∆Hv and Vm were obtained from the SciFinder Scholar database (American Chemical Society.
70 oC. a homogeneous reaction mixture was maintained throughout the reaction. Synthesis of Silylated Starch The silylation of corn starch was performed with HMDS as the silylating agent (eq. An additional peracetylation procedure to substitute the free hydroxyl groups with acetate groups was performed to improve the quality of the NMR spectra. It is defined as the average number of silyl substituted OH groups on the anhydroglucose (AHG) unit of starch.1. C O H 2OR H 1 + NH3 2 OH H3C [R = H or Si(CH3)3] Starch Hexamethyldisilazane Silylated Starch (1) (3. The 1H‐ NMR spectrum of silylated starch (Figure 3. However.b. 54 .) shows the presence of silyl groups at about δ 0 ppm.5.1. Using standard conditions (HMDS: AHG molar ratio of 3. Chapter 3 3. 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 3.1.a. mixtures of toluene and DMSO were applied to avoid precipitation of the silylated starch during the reaction. a product with a DS of 0.a. The silylated products were characterised by NMR.5.). The degree of substitution of the silyl groups (DSsilylation) was determined by 1H‐ NMR. 24 h). the starch peaks are very broad and this feature hampers accurate determination of the DSsilylation.3. 3. 6 H2C OH 4H 3 H5 OH H 6 H2C O R 4H H5 OR 3 H O H 1 H H3C CH3 CH3 CH3 + H3C Si NH Si DMSO 50 deg. The 1H‐ NMR peaks from the AHG unit after peracetylation were indeed considerably sharper and allowed more accurate DS calculations (Figure 3.).60 was obtained. as suggested by Einfeldt et al . The silylation procedure was adapted from that previously reported for dextran [19‐23]. In this way.) Instead of using DMSO as solvent.).1.
and Figure 3.1. The results are given in Table 3. 55 .60.5‐4. Typical 1H‐NMR Spectra of Silylated Starch (Sample SN3. peracetylated The effect of the HMDS to starch molar ratio (1.2. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Figure 3. in CDCl3. not peracetylated (b).5) on the product DS was investigated by performing experiments with a constant starch and a variable HMDS intake. 50 oC): (a). DSsilylation=0.1.
4 1. a fixed starch intake of 6 g (37 mmol AHG) starch was applied.45 0. mmol) 12 (56) 18 (83) 24 (111) 36 (167) HMDS: AHG mol ratio 1.5:1. For all experiments. Surprisingly. the volumetric HMDS intake is about half of the DMSO intake.5 2 2.2. the DSsilylation decreases for higher intakes of HMDS.5 Product DS (DSsilylation) 0.25 3 4. solubility parameter = 11.46 Experiment SN1 SN2 SN3 SN4 a Experiments were performed in DMSO at 70°C. Effect of HMDS: AHG mol ratio on DS of the silylation product a HMDS intake (ml. For reactions with an order higher than zero.60 0.55 0.68 0.25 cal1/2cm‐3/2) due to the presence of the bulky apolar SiMe3 groups. DS of the silylated products at different HMDS: AHG ratios (DMSO.6 DSsilylation 0.5 0.5 HMDS : AHG Ratio [mol/ mol] 4 4. The silylation reaction 56 . 70 oC) Table 3. a positive effect of higher reagent intake on the reaction rate and thus the DS is expected. HMDS is a rather apolar compound (Hildebrand solubility parameter of 6. The experimentally observed lowering at higher HMDS intakes is likely due to a decrease in the polarity of the reaction mixture. At a mol ratio of HMDS to AHG of 4.5 Figure 3.5 3 3.65 0.36 cal1/2cm‐3/2) considerably.5 2.1.67 0. Its presence will reduce the polarity of the reaction medium (DMSO. Chapter 3 0.
3.45‐0. A schematic representation of the reaction is provided in eq.) 57 . The DSsilylation (0.0 [18. and DMSO/pyridine mixtures yielded silylated starch with DS values ranging between 0.0) gave silylated dextran with DS values between 1. Silylation of starch with HMDS in formamide.0 . After precipitation with heptane and vacuum drying.68. DMF. 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.1 and 3. the silylation of starch in pure DMSO using HMDS as the reagent has not been reported to date. 50 oC H CH2 H OR H C (CH2)5 O n H O H OR + H (R = H or Si(CH3)3 Silylated Starch (1) ε-Caprolactone Monomer [R = H or Si(CH3)3 ] Silylated Starch-g-Polycaprolactone (2) (3. sample SN‐1) in THF at 50 oC for 24 h using Al(OiPr)3 as catalyst. Similar reductions in the reaction rates when working at higher reactant intakes were observed for the esterification of starch with vinyl laurate and stearate . 29].2. pyridine.3‐2. liquid ammonia.6. The much higher DS values obtained for dextran compared to starch may be related to differences in molecular weights between starch and dextran and the type of AHG linkages (mainly α‐1.6‐glucosidic for dextran).3. O O H2C O R H H OR H O O H OR H O Catalyst: Al(OiPr)3 Toluene.6.7) of the products is in the range of those published for other starch silylation systems. The use of HMDS to silylate dextran (Mw=6000‐40000) in DMSO (HMDS to OH molar ratio of 0.25‐5. The products are soluble in DMSO as well as in less‐polar solvent such as chloroform and THF. 3.2.7‐3. In situ Ring Opening Polymerization of ε‐Caprolactone with Silylated Starch A number of in situ ROP experiments with ε‐caprolactone (CL) were carried out using a typical silylated starch sample (DS=0. white solid products with isolated yield > 96% were obtained. DMA/LiCl. To the best of our knowledge. Petzold et al  reported that silylation of starch with trimethylsilylchloride (TMSCl) in pyridine yielded trimethyl‐silyl substituted starch with DS values between 0.
26]. not all of the caprolactone is grafted to starch and large amounts of PCL homopolymers (72%) were formed. as is shown in Figure 3. Typical 1H‐NMR spectrum of a silylated starch‐g‐PCL sample.68. broad peaks in the region δ 3. the exchange reaction between Al(OiPr)3 and the OH groups of silylated starch is not quantitative under the conditions applied in this study.3.3. Further process optimization studies allowed the synthesis of products with less than 42% of homopolymers (vide infra). Resonances from the starch peaks are observable as small. The homopolymers are formed by direct polymerization of ε‐CL initiated on isoproproxide moieties attached to the Al catalyst.4. This is clearly indicated by resonances of the iPr end‐group of the PCL homopolymer at δ 4.4‐3. 58 . Coding of the peaks is given in Figure 3. (Sample SN1CL1.4 ppm. Chapter 3 The products were characterized using 1H‐NMR analysis in DMSO‐d‐6 as the solvent.2‐4 ppm and imply that the ring‐opening polymerisation reaction of CL indeed occurred.8 and 5. A typical spectrum is shown in Figure 3. Apparently. However. The formation of CL homopolymers for this type of reactions has been observed before [15.21) in DMSO‐d6 at 60 oC. DSPCL=0.4.2 ppm.9 and 1.0‐5. Figure 3. DSsilylation =0. The peaks from the polycaprolactone units are clearly present in the range of δ 1.
Exchange Reaction: iPr O Al O iPr iPr O + H2O iPr O iPr O Al OH + iPr OH 2. However.) 59 .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. This leads to the formation of carboxylic end groups (see eq. Product formation for the reaction between silylated starch and ε‐ caprolactone Besides the ‐OH group of starch. 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. peaks arising from a ‐COOH unit could not be detected in 13C‐ NMR spectra (δ 175‐180 ppm). Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors 1. 1. 3. 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. residual water may also initiate the polymerisation reaction.4.7.
0 18. in all samples.5.58 0.2 Total Yieldb [%] >99 >99 >99 99 96.0 15. This implies that the ε‐CL conversion is essentially quantitative in all cases. b.9 22. Determined gravimetrically and defined as the total weight of the isolated product divided by the total intake of reactants (silylated starch and CL).3 ppm) should be twice the intensity of the ‐CH‐ proton of the isopropoxide end group (δ 4.5 29.5 and 100%. Overview of results for the grafting of ε‐CL on silylated starch a ε‐CL/ OH [mol/ mol] 13.21 0. In the case of only homopolymerisation. the intensity of the peak from the ‐CH2‐ end group of the homopolymer (ε’ at δ 3.2.5. The results are given in Table 3. The yield of the products was measured gravimetrically and varies between 96.3. The GE increases with higher ε‐CL intakes. 60 .72 Grafting Efficiency [%] 28 43 48 55 58 All reactions were performed using the same intake of SN1 silylated starch (DS=0. Five experiments were performed to study the effect of different ε‐CL to silylated starch ratio.47 0. and reaches 58% for a ε‐CL to starch–OH ratio of 29. The grafting efficiency (GE) for the sample given in Figure 3. units] 40 43 44 49 54 DSPCL 0. However. Table 3. This implies that grafting of caprolactone to starch also occurs to a significant extent.68) in THF at 50 oC with Al(OiPr)3 as the catalyst (1 mol Al(OiPr)3 per 10 mol‐starch‐OH groups). (SN1CL1) is 28%.2.). The ACL of the polymer and the DSPCL increase almost linearly with the ε‐ caprolactone intake (Figure 3. Chapter 3 The ratio of homopolymerisation versus grafting on starch may be obtained by comparing the integrals of selected peaks in 1H‐NMR spectra. Chain Length [mon.9 ppm).34 0. Experiment SN1CL1 SN1CL2 SN1CL3 SN1CL4 SN1CL5 a. the intensity of the resonance ε’ was considerably higher that that of the ‐CH‐ iPr peak.2.5 Avg. 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. and 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)
21‐0.2. 1. This was achieved by making the starch more hydrophobic by partial substitution of the OH groups by trimethylsilyl groups. the highest value was obtained with a ε‐CL‐AHG ratio of 29. The products may have interesting applications as compatibilizers for starch‐polymer blends. 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. the percentage of PCL grafted to starch compared to the total amount (grafted and homopolymer) of polymerized CL [%] : gas constant. Nomenclature A : peak area of certain proton in 1H‐NMR spectra [‐] : peak area of certain peak at x until y ppm in 1H‐NMR spectra [‐] : average number of CL repeating units in a grafted polymer chain or homopolymer [monomer units] : Degree of Substitution. Chapter 3 3. Conclusions The successful synthesis of poly‐(ε)‐caprolactone grafted corn starch co‐ polymers using a three step approach is reported. indicating that homogeneous reaction conditions are favorable for the grafting reaction.5. The DS of the PCL chains was between 0.986 cal mol‐1 K‐1 : temperature [K] : R T Vm molar volume [cm3/ mol] 64 . 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.46‐0.68) was obtained using a DMSO/toluene mixture as the solvent and HMDS as the silylating agent. The key feature is the use of a homogeneous reaction mixture for the ROP of starch with ε‐CL. Considerable amounts of ε‐CL hompolymers with isopropyl end‐groups were also formed. 3. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] A x − y ppm ACL DS DSsilylation : DS of silyl group substituents [‐] DSPCL GE : DS of PCL chain substituents [‐] : Grafting Efficiency. depending on the ε‐CL‐starch ratio. The grafting efficiency varied between 28‐58%. Silylated starch with a low‐medium DS (0.72. The ROP with ε‐CL was performed using Al(OiPr)3 as catalyst in THF as the solvent.4.
1273‐1335. De Mulder‐Johnston: Edible and biodegradable polymer films: Challenges and opportunities. D.F. 8. 1977.starch. 311‐316. Goldberg: Review of the biodegradability and utility of poly(caprolactone). 1251‐1253. Degrad.dk/isi/stat/rawmaterial. Biobased packaging materials for the food industry. . demand and recovery for 2006 in Europe. 2003. Chiellini. . the Royal Veterinary and Agricultural University. Denmark.B. 2003. Burr. Krochta.M. . Frederiksberg C. 52. Environ. 407‐412. The compelling facts about plastics: An analysis of plastics production. 1998. accessed on August 01. Linke: Water vapor sorption determination of starch based porous packaging materials. Eng. 65 . Environmen. Polym. . Plastics Europe. Polym. H. Sci. Bagley. Mater. Rustgi: Biodegradable polymers. J. Brussels. 38. Chandra. W. A. J. Wang: Properties of starch blends with biodegradable polymers. . F. Karlsson. Solaro: Biodegradable polymeric materials. Department of Dairy and Food Science. . Aly.K. X. Adv. 56. Polym. Association of Plastics Manufacturers. Abdel‐Mohdy: Grafting of poly(methacrylic acid) on starch and poly(vinyl alcohol).6. Yang. R. C.Sci. 385–409 . Food Technol. J. A.html. 1995. 2008. Belgium. Yavuz.A.. 43.C. 305‐313. 1998. S. status and perspectives. C. C. Starch‐Starke 2004. 3. 17. of . . M. E. Russell: Graft copolymers of polysaccharides with thermoplastic polymers ‐ new type of filled plastic. Wang. Denmark: http://www. Doane. Babac: Preparation and biodegradation starch/polycaprolactone films. R. Part C: Polym. Y. 61‐67. L. E. R. Rev. Polym. Starch‐Starke 2000. .. Weber). . Website of International Starch Institute. 1997.A. R. Prog. 51. Sci. Sci. J.A. A. report of Food Biopack Project (Ed. Eng. Tsiapouris. . Fanta. 1996. 23. 2008. References . 2000.C. Beliakova. 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. 11.R. 107‐113. K. Macromol. 61‐74. J. Polym. Aarhus. Albertsson: Biodegradable polymers and environmental interaction. 53‐57. G.M. C.
561‐566. W. Des.L. 4. Nouvel. P. Six. Macromol. Rutot. D. C.. D. . Degee. K. Gunther. Klemm: Regioselective functionalization of starch: Synthesis and 1H NMR characterization of 6‐O‐ silyl ethers. Gunther. L. J.L. P. M. Six: Controlled synthesis of amphiphilic biodegradable polylactide‐grafted dextran copolymers. selective syntheses. Degee. . Heublein: Silylation of cellulose and starch ‐ selectivity. Degrad. L. Polym. . K. A. Dellacherie. W. Macromolecules 2000. Averous. L. Fringant: Properties of thermoplastics blends: starch‐polycaprolactone. . P. C. Klemm. D. Klemm. 3091‐3100. H. Chapter 3 . Stab. Biomacromolecules 2001. Stein. Nouvel. Pantiru. Dubois. . I. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer 66 . Klemm. 43. Degee. P. P. Biosci. Dubois. 2001. L. Polym. Chem. A. P. high value applications. J. 2004. and subsequent reactions. 2577‐2588. I. Petzold. J. Cellulose 2003. Koschella. Polymer 2002. 10. L. Dellacherie. T.L. 1735‐ 1743. Duquesne. structure analysis. 251‐269. A. P. Polymer 2000. B. 41. 965‐969. Nouvel. Six: Partial or total silylation of dextran with hexamethyldisilazane. Ydens. . D. Six: Silylation reaction of dextran: Effect of experimental conditions on silylation yield. . 6713‐ 6721. . Monomers Polym. 33. Dubois. D. Dellacherie. 415‐426. Moro. Ydens. Kussler. Klemm: Preparative and 1H NMR investigation on regioselective silylation of starch dissolved in Dimethyl Sulfoxide. E. Einfeldt: Structure design of polysaccharides: novel concepts. . K. J. A. 1443‐1450. 40. P. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. 4157‐4167. D. 2. 5. R. Ydens. Rutot. Part A: Polym. Sci. Stein. E. Einfeldt. 2001. D. Fessi. E. Symp. Dubois: Controlled synthesis of poly(epsilon‐caprolactone)‐grafted dextran copolymers as potential environmentally friendly surfactants. Dubois: Aliphatic polyester‐ based biodegradable materials: new amphiphilic graft copolymers. 2002. Petzold. 42. and chemical stability of silylated dextrans. 163. . Einfeldt. 341‐347. M. J. Gunther: Synthesis and NMR characterization of regiocontrolled starch alkyl ethers. I. P. Stein. 35‐47. 1. . Dubois. Polymer 1999. Macromol. E. M. P. regioselectivity. . 2001. K. Krishnan. P. Dellacherie. D. Petzold. E. Hamaide. C.L. 73. Dole. Petzold. W. Biomacromolecules 2003.
61. L.4‐ dioxan‐2‐one. Penczek. Macromol. Damrau: Zn‐lactate catalyzed polymerizations of 1. . H. . A.K. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors agents.O. F. Macromolecul. R. Res. 1089‐1097. Janssen. Lehmann: Selective 6‐O‐acetylation of amylose. Horton. Duda: Living polymerization with reversible chain transfer and reversible deactivation: the case of cyclic esters. Carbohydr. A. 21. S.L. Rapid Commun. Macromolecules 1996. 941‐950. A novel route to functionalised nanoparticles. D. . 2001. Junistia. T. accepted for publication in Starch‐Starke. L.B. 199. 29. Picchioni. Kinetics and mechanism. Biela. Rapid Commun. Hertz Jr. Duda: Polymerization of ε‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. 1989 .: Solubility parameter concepts – a new look. D. Kricheldorf. 1978. J. 22. 67 . H. 2000. presented at ACS Rubber Division meeting.M. . 659‐663. A. Mexico City.J.P. Sugih. 2008. D. 1998. 553‐556. Macromol Chem.R. . Phys. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. 1399‐1406. Manurung.
K2CO3. 13C‐NMR and FT‐IR. The fatty esters were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. the DS for the starch‐laurate compounds is higher than for the corresponding starch‐ stearates. For low vinyl‐ester to starch ratios. The yellowish products were characterized by 1 H‐. Keywords: corn starch.96). the DS decreases. esterification.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. and Na‐acetate) in DMSO at 110°C. K2CO3. vinyl ester to starch ratio and the type of catalyst. vinyl stearate . an increase in the vinyl‐ester concentration leads to higher product DS values. vinyl laurate. When performing the reactions using Na2HPO4 as the catalyst. At higher ratios. presumably due to a reduction of the polarity of the reaction medium.24 ‐ 2. and Na‐acetate are superior catalysts with respect to activity compared to Na2HPO4 and products with DS values close to 3 were obtained. The DS of the products is a function of the carbon number of the fatty acid chain.
more water‐resistant products may be obtained [3‐4]. 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. Recently Mormann et al  explored the possibility of using vinyl esters and particularly vinylacetate as reagents for the preparation of starch esters. Relatively low‐DS (0. An attractive field of application for these polymers is the use as packaging materials. the use of native starch for packaging materials is limited due to its low moisture resistance. 5‐6] were introduced using fatty acid chloride reagents. high brittleness. Esterification of starch with low molecular weight fatty acid derivatives is one of the oldest modification technologies to improve starch properties. For the current petrochemical based products recycling is often neither practical nor economically feasible . However.61) products were obtained using this approach. Further modification of starch is therefore required to introduce hydrophobicity and to improve mechanical and moisture barrier properties. resulting in products with DS values up to 2. However. Their 70 . and even fatty acid derivatives (C12‐C18) have been used in the modification reaction [5. 5]. The mechanical properties and hydrophobicity of the products were significantly improved when using these longer chain fatty acid precursors [1. An alternative method using methyl and glyceryl laurate esters in the absence of solvent has been recently developed . higher molecular weight carboxylic starch esters (C4‐C6) .Chapter 4 4. 7]. However. The introduction of acetate groups on starch makes the product more hydrophobic. defined as the moles of substituents per mole of anhydroglucose (AHG) units) . The first paper on the acetylation of starch was already published in 1865 . 5]. Natural polymers such as starch. cellulose or proteins are potentially very interesting starting materials for biodegradable packaging materials.7 [1. Introduction Green biodegradable polymers derived from natural resources are potentially very interesting substitutes for non‐biodegradable petroleum‐based polymers. In particular starch is attractive as it is relatively cheap and abundantly available. even after the addition of plasticizers . which are relatively expensive and rather corrosive . most of the studies performed to date use short chain carboxylic acids (C1‐C4). and particularly acetic acid derivatives (C2) [2‐4].1.34‐0. and incompatibility with hydrophobic polymers. and consequently. The major obstacle is the pronounced brittleness of the materials. poor processibility (high viscosity). the fatty ester substituents [1. However. The hydrophobicity increases with the degree of acetate substitution (DS. To improve the mechanical properties.
The spectra were recorded at 50 oC. 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. and tetrahydrofuran (THF) were supplied by Acros (Belgium) and were also used as received. to the best of our knowledge. Materials and Methods 4. IR spectra were recorded on a Spectrum 2000 FT‐IR Spectrometer (Perkin Elmer). 4‐N. 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.2. leading to higher reaction rates.2. The products were placed directly on the diamond plate and 50 scans with a resolution of 4 cm‐1 were recorded. In DMSO.2. Japan). Analytical grade vinyl stearate (Aldrich.N‐dimethylaminopyridine (DMAP). the use of basic catalysts other than Na2HPO4 has been investigated. Potassium carbonate (Boom. Germany) and disodium hydrogenphosphate (Merck. vinyl laurate (Fluka. Germany) and acetic anhydride (Merck.1. The synthesis of starch stearate esters using vinyl ester reagents has. The effects of the starch to vinylester ratio on the reaction rates and DS have been explored. The reactions were either performed in water or in DMSO using a basic catalyst (Na2HPO4). starch esters with a substantially higher DS value (up to 1. Materials Corn starch (approx. The starch was dried before use for 48 h at 105 oC under vacuum (approx. leading to a moisture content of 2 %‐wt (measured gravimetrically). 4. the Netherlands).01 when using vinyl‐laurate. This solvent effect is likely caused by the higher solubility of the vinyl esters in DMSO compared to water. 4.2. Germany) were used as received. ~1 mbar). 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. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. not been reported to date. Technical grade dimethyl sulfoxide (DMSO). as recommended by Laignel et al . In addition.6 for starch acetate ester) were obtained. 1 71 . Germany) were used without further purification. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). sodium acetate (Merck. The maximum attainable DS of starch acetate in water was below 1 and limited to 0.
2. broad peaks. 3S. 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. C18). C2). 3H. broad peaks. 7H.3.0 (m.2.Chapter 4 4.5 (m. 1230 (C‐O stretching). 2.9 (t.4. C4‐11). 1. 61. 1 H‐NMR (after peracetylation. Table 4.3 (m. Finally. 1295. CDCl3): δ 0. 2H. 3H.52): H‐NMR (before peracetylation.9 (t.3. the product was dried in a vacuum oven (70 oC. 1..1. (13C‐NMR) and Figure 4. respectively). C2). C3).5 g) was first gelatinized in DMSO (5 ml) at 70 oC for 3 h. 1 C‐NMR (before peracetylation.3.96) H‐NMR (before peracetylation. C1S). 1.1 (C11)..6 (m.5 (m. 7H. C12). 1. 2H.1. 28H. 3H. 5S). 31. DS = 2. 1.9 (t. C1S‐6S). C4‐C17). The samples were characterized by 1H‐ and 13C‐NMR and FT‐IR.3 (m. Table 4. 2H. DS = 2. Starch‐stearate (Sample 19. 2H. The atom numbering scheme is given in Figure 4. 1455 (CH2). and O‐C6S). 1. broad peaks.4 (broad. C3).3 (m. 24. Subsequently. CDCl3): δ 0. Methods 4. 760. 28‐32 (C4‐9). 2. 68‐74 (broad. Starch‐laurate (Sample 17. broad peaks. 1. 935 (C‐O stretching).1 (m. Figure 4.7 (C11). C2’). 3‐6 ppm (m. typical spectra are given in Figure 4. C2S. 2H. broad peaks. 1150 (C‐O stretching).8‐2.9 (broad. C3).2. C6S). 76‐78 ppm. attached to O‐C2S. the product was precipitated using methanol (100 ml) and separated from the liquid phase by decantation. broad peaks. 7H. resulting in the formation of a homogenous transparent solution. 1410 (C‐H bending). O‐C3S. 1020 (C‐O stretching).1. 2850 (C‐H stretching). 1350 (C‐H bending). 2. 1740 (C=O).. broad peaks. 3‐6 ppm (m. approximately 5 mbar) for 24 h until constant weight. 12). C1S‐6S). 2H. C1S‐6S) 1 72 .4 (m. 3H. 1110 (C‐O stretching). After cooling.5 (m. CDCl3): δ 14. 16H. 34. broad peaks.9 (C10). C2). (FT‐IR). 22. overlap with CDCl3 (C4S). 13 FT‐IR (cm‐1): 2920 (C‐H stretching). 720. (1H‐NMR).9 (C3).0 (C12). 16H.1. 3‐6 ppm (m. CDCl3): δ 0. Typical example of the synthesis of laurate and stearate esters of corn starch Corn starch (0. 1370 (C‐H bending). 95. 172‐174 ppm (C=O. C4‐11). The product was washed twice with methanol (50 and 25 ml.
C1S‐6S) 1 C‐NMR (before peracetylation. 3S.5 (broad. 75. 68‐74 (broad. 3H. 760.2. 25.5 (m. 1350 (C‐H bending). 22. C2’). CDCl3): δ 14. attached to O‐C2S. 1.0 (C3). 1.0 (C18).9 (t.8‐2. O‐C3S.1. 1410 (C‐H bending). 1150 (C‐O stretching). 1455 (CH2). broad peaks. 28H.0 (C16).3. 3‐6 ppm (m. 2H. 172‐174 ppm (C=O. 32. C2S. C18). C2). 1740 (C=O). 720. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants H‐NMR (after peracetylation. 2. 5S). broad peaks. 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. 26‐32 (C4‐15). C1S). 1100 (C‐O stretching). 2850 (C‐H stretching). 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. A peracetylation reaction to substitute all of the remaining 73 . CDCl3): δ 0. C6S).7 (C4S). 1.6 (m. 95.2. 2H.3 (m. 1295.4 (m.2 (C2). and O‐ C6S) 13 FT‐IR (cm‐1): 2920 (C‐H stretching). 34. 1020 (C‐O stretching). C4‐C17).7 (C17). 950 (C‐O stretching). 7H.4 (broad. 3H. 61. C3). Numbering scheme for carbon atoms of products 4. 1370 (C‐H bending).
Typically.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until the starch was fully dissolved (typically 3 h). acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.). 4.) AH − starch = (4. 4.) The DS of the acetate groups of the products may be calculated by comparing the unit area of the acetate protons (AH‐acetate) to the unit area of the starch protons (calculated from the intensity of the starch peaks at δ 3.Chapter 4 hydroxyl groups with acetate groups was applied to obtain reliable DS data. the peractylating reagents (DMAP.5. The peracetylation reaction was conducted for 7 h at 50 oC..: 74 .3.8− 2.5.2. Determination of the Degree of Substitution (DS) The DS of the product was calculated using 1H‐NMR spectra of the products after peracetylation. It was finally dried overnight in a vacuum oven at 70oC and 5 mbar until constant weight.5‐1.6 ppm).) 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 0.7.). The peracetylation procedure by Einfeldt et al  was applied.6‐5. AH − acetate = A1.) 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. The procedure was described earlier by Elomaa et al  and the relevant equations are shown in equation 4.4.2. and 4.) (184.108.40.206.6 7 (4. The DS of the fatty acid esters was calculated from the DS of the products after peracetylation (eq 4.: DSacetate = AH − acetate AH − starch A3. The product was precipitated by the addition of methanol and washed several times with methanol.3.6 − AC 2 3 (4. 4. and 4. the starch ester (0.3.6−5.1.8‐2.6.8 ppm (protons attached to the fatty acid carbons C3‐C12 for laurate and C3‐C18 for stearate) using eq. DS fatty ester = 3 − DS acetate (4. Subsequently.6 ppm range) and a correction has to be made (eq 4.4.
the starch was gelatinised in DMSO at 70°C for 3 h to make the hydroxyl groups of starch more accessible for reaction. 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. but swell in organic solvents such as toluene and THF.8 = (stearate) 33 33 (4. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants AH − fattyester = AH − fattyester = AC 3−C12 A0. the brownish gel was precipitated with methanol and the product was collected after vacuum drying in the form of a transparent. The products of these exploratory reactions are insoluble in water and DMSO. 4.1. the esterified starch started to separate from the medium in the form of a gel. 4. A schematic representation of the esterification reaction of starch with the vinyl esters is provided in Scheme 4.1. Ax‐y stands for the peak area in the range δ x‐y ppm. After 24 h.3..) (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. 75 .1.7. 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. Initially.5−1. The reaction was performed in two discrete steps.‐4.1).5−1. the vinylester and the catalyst were added and the reaction mixture was heated to 110°C. Subsequently. After 2‐3 h.8 = (laurate) 21 21 AC 3−C18 A0.) In eq.220.127.116.11. Results and Discussion 4. light yellow solid.
Table 4.) in CDCl3 at 50oC.2.1. a product DS of 2. 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.96.) in CDCl3 at 50oC.52 (Sample 17. Table 4. 76 . DS = 2. Higher DS products have a higher solubility in CDCl3 and good quality 1H‐ and 13C‐NMR spectra could be obtained (Figure 4. 4.3. DS = 2. (c) peracetylated starch laurate.1. 4. When using a vinyl‐laurate : AHG molar ratio of 3 and K2CO3 as the catalyst.Chapter 4 The DS of the products was determined by using NMR (vide infra).52 was obtained.52 (Sample 17.2. Product Characterisation 18.104.22.168. Medium DS starch laurate and starch stearate (1 < DS < 2) dissolve poorly in DMSO‐d6 and CDCl3. For atom numbering scheme: see Figure 4. even at higher temperatures (50 oC).). Typical 1H‐NMR spectrum of: (a) native starch in DMSO d‐6 at 60oC. Figure.3.2. (b) starch laurate. and 4.1. A reaction with vinylstearate at similar conditions resulted in a stearate starch ester with a DS of 2.
Typical 13C‐NMR spectra of: (a) native starch. Typical 13C‐NMR spectra of the products are given in Figure 4. Clearly visible are the peaks arising from starch and the aliphatic hydrogen atoms of the fatty acid chain (δ 0. DS = 2. are together with the CH2 groups of the acid chain adjacent to the ester moiety in the range δ 1.1. 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 4.5 ppm) are broad and overlapping . The same 77 . (b) starch laurate. in DMSO d‐6 at 60oC. DS = 2. The starch peaks (δ 3‐5. Two of the carbon resonances (1S and 4S) are considerably shifted compared to native starch. For atom numbering scheme: see Figure 4.3 ppm. This feature hampers the DS determination by NMR.8‐2.3. The resonances arising from the anhydroglucose unit of starch are broadened. 10‐11].96 (Sample 19) in CDCl3 at 50oC. (c) starch stearate.2.52 (Sample 17) in CDCl3 at 50oC. Figure 4. and therefore a peracetylation procedure to substitute all of the remaining OH groups with acetate groups was applied [8.c. NMR spectra of the peracetylated products are considerably improved in terms of peak resolution and allow a more reliable calculation of the DS.8‐2. The proton signals of the acetate methyl group. 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). The 1H‐NMR spectrum of a typical peracetylated starch laurate is shown in Figure 4. required for DS determinations.5 ppm).3.2.
For comparison.2. c.2.52.3. native starch -OH C-H 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Figure 22.214.171.124.4. the C‐H stretching vibrations of the alkyl groups of the fatty ester chain are clearly present at 2920 and 2850 cm‐1. Table 4.Chapter 4 phenomenon was observed by Dicke for starch acetate .). FT‐IR measurements The FT‐IR spectra of starch laurate and starch stearate are shown in Figure 4. FT‐IR Spectra of starch laurate (DS= 2. a spectrum of native starch (Figure 4. Sample 19. 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 17.) and native starch FT‐IR spectra of both starch laurate and starch stearate (Figure 4. and 4.4. The near absence of remaining hydroxyl 78 .c. starch laurate C-H C=O C-O C-O a. Characteristic peaks of the polysaccharide backbone are visible in the 1250‐900 cm‐1 region (C‐O stretching) .b. Table 4. and 4.4.1. starch stearate C-H C=O C-O Absorbance b. In addition.96.c.) is also included. starch stearate (DS= 2.b. 4.) show characteristic bands of the carbonyl group of the fatty esters in the 1750‐1700 cm‐1 region.a.1.
07 0. 4. Systematic Studies The effect of important process variables like the vinyl‐ester to AHG ratio.2.96 2.23 0. 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.05 0. 79 .60 1.90 1.68 0.3. 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).90 0.99 1.1.01 0. 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 1.08 1.13 1.52 2.44 Exp.24 2. The results are shown in Table 4.1.54 2. Table 4. Most of the experiments (14) were performed using Na2HPO4 as the catalyst. In addition. type of catalyst and the effect of the addition of co‐solvents on the product DS was studied in more detail.91 0.57 0. in line with the NMR data.
5.5 Degree of Substitution (DS) 1 0. A further increase leads to a reduction in the DS. Higher concentrations of vinyl esters are expected to lead to higher esterification reaction rates. At a ratio of 1:6. This behaviour is likely the result of two opposing effects. DS of the product as a function of the type of vinyl ester and the vinyl ester‐starch ratio (24 h reaction time. 110 oC.5 0 0 1 2 3 4 Vinyl Ester: AHG Starch Ratio [mol/mol] 5 6 Figure 4.5. 1. The results are presented in Figure 4. Effect of vinyl ester to AHG ratio on the product DS The effect of the vinyl‐ester to AHG molar ratio on the product DS was determined for both types of vinyl‐esters with Na2HPO4 as catalyst (samples 1‐ 3. this positive effect dominates the reaction rate and the DS of the products will therefore increase at higher vinyl ester intakes.23 for vinyl‐laurate at an intermediate vinyl‐ester‐AHG ratio of 3. The highest DS value was 1.).1. This reduced polarity is expected to lead to ○ : starch‐laurate □ : starch stearate (lines for illustrative purposes only) 80 . 7‐10). The DS values are increasing with higher vinyl‐ ester/AHG molar ratio until a certain maximum. 2 wt% catalyst intake on starch). the ester intake is equal on a weight basis to the DMSO intake. The DS of the products is a clear function of the vinyl laurate and stearate ester intakes (Figure 4.3.Chapter 4 4.3.5. This is likely due to a reduction of the polarity of the reaction medium. At low to medium vinyl ester/AHG ratios (0‐3). A further increase in the vinyl ester intake leads to a reduction in the DS.
Although toluene may positively effect the reaction by (partly) re‐dissolving starch ester precipitates. and illustrated in Figure 4. Here. it also results in a dilution of the reaction mixture and a reduction in the polarity. 11‐14 in Table 4.3. with high carbon numbers leading to a reduction in the DS. leading to enhanced reactivity.6. Effect of the addition of toluene as a co‐solvent A number of reactions were performed using a co‐solvent. The addition of toluene for the esterification of starch with vinyl‐esters surprisingly did not lead to improved DS values. the DS of the product is also a function of the chain length of the fatty acid.). These negative effects dominate the reaction performance at higher vinyl ester/AHG ratios and lead to a reduction in the DS values.3.1. When using Na2HPO4 as the catalyst. The latter factors appears to have a strong effect on reaction rates (vide supra).1. i. 4.5. 4. It is clear that Na‐acetate and K2CO3 are considerably more active than Na2HPO4 and products with a significantly higher DS were obtained.3.3. A similar trend in reactivity pattern was observed in this study and the DS decreased from 1. Aburto. 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.e. For starch 81 .). The results are given in Table 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 4.1.8 for stearoyl chloride. with reductions in polarity leading to lower reaction rates. This effect is particularly evident at higher vinyl ester/ AHG ratios (>3) (Figure 4. K2CO3 and Na‐ acetate were tested. 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. 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. These findings were rationalised by assuming that the co‐solvents increase the solubility of the silylated products. et al  for the silylation of starch. Thus. Catalysts screening A number of alternative basic catalysts for Na2HPO4. the starch laurate esters display higher DS values than the starch stearates. A similar procedure was proposed by Nouvel.). the addition of co‐solvents (toluene/ THF) led to an increase in the DS.7 for lauroyl chloride to 0. 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 .3.2. In this case.
5 Degree of Substitution 2 1. when using K2CO3 as the catalyst. Thus.6.44). the DS of the product is also tunable by proper catalyst selection. 110oC. K2CO3. function of the type of catalyst.5 were obtained.5 0 Laurate Vinyl-Ester Stearate Figure 4. The DS of the products is a clear function of the chain length of the fatty ester and the type of catalyst.). and Na‐acetate). the statement that the DS for the laurate esters is always higher than for the stearate esters is not generally valid and a. 82 . in line with the findings for NaH2PO4. For starch stearate. The DS of the laurate ester when using Na‐acetate is higher than for the stearate ester (Table 4. the DS for the laurate ester is lower than the stearate ester. Conclusions A study on the synthesis of corn starch fatty acid esters with high DS values is reported.1.5 1 0.4. K2CO3 gave products with a significantly higher DS (2. and Figure 4. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4.96) compared to Na‐acetate (DS=2. catalyst amount = 2%‐w.Chapter 4 laurate esterification. the two catalysts are equally effective and products with a DS of about 2.o. However. 3 2. Apparently. and FTIR and confirm the presence of chemically bound fatty acid chains. Comparison of DS values with different catalysts (Vinylester: AHG molar ratio = 3:1.6. The yellow products were characterized by 1H‐ and 13C‐NMR. gray: Na2HPO4 black: K2CO3 white: Na‐acetate 4. The products were synthesised in DMSO using vinyl‐esters in the presence of basic catalysts (Na2HPO4. DMSO).
C. E. K. Bikiaris. 705‐721. S. 1999. 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. Higher ratios led to a reduction in the DS. Thiebaud. D. J. Sci. . and biodegradability of fatty‐acid esters of amylose and starch. J. M. 65. 361‐368. 1995. Polym. . Borredon. Merrill: Properties of Fatty‐Acid Esters of Starch. Aburto. M. the DS of the product increases at higher vinyl ester intakes. Appl. 58. 1997. Crops Prod.W. At low vinyl ester/ AHG ratio. Sci. Prinos. Eng. Panayiotou: Synthesis. E. J. Sagar. 83 . Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants With these catalysts. 21. Alric. characterization. Koch. J. Polym. 74. . 2003. The DS of the products may also be tuned with the vinyl ester/AHG molar ratio. X.5. Borredon. A. A. Y. E. S. I. 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. J. C. 1440‐1451. Important product properties will be described in the next chapter. W. I. . Sci. 4. 54. Gatenholm: Surface octanoylation of high‐ amylose potato starch films. Mullen and E. Carbohydr. Chem. Aburto. . J. A maximum was observed at a vinyl ester: AHG ratio between 2 and 4. Prinos. Xu. References . 1942. Dzenis. D. Y. J. presumably due to a reduction of the polarity of the reaction medium. 34. products with a DS > 2. 1647‐1656. Alric. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Appl. 1208‐1217. Pacsu: Starch Studies: Preparation and Properties of Starch Triesters. 2005. J. D.6. Ind.4 could be obtained for both laurate and stearate esters. Thiebaud. Bengtsson. thermal characteristics and biodegradability of extruded starch acetate foams. Hanna: Water solubility. Bikiaris. Appl. Polym. Polym. and P. Ind. 1‐11.
S. C. M. 298. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. C. E. Gunther. Dubois. 11. Senocq. . J. . Med. 1997. Alric. Nouvel. R. Bliard. G. Elomaa. R. L. H. Petzold. Ydens. Aburto. Mouysset‐Baziard.F. Reis: Thermal properties of thermoplastic starch/synthetic polymer blends with potential biomedical applicability. Starch‐Starke 1999. F. Stein. 84 . 57. G. D. P. Polymer 2002. Mano. I. Laignel. Macromol. Peltonen. I. M. 43. 2001. . 51. 251‐260. Mater. Einfeldt. P. and E.Chapter 4 . Laatikainen. Klemm: Preparative and H‐1 NMR investigation on regioselective silylation of starch dissolved in dimethyl sulfoxide. Dicke: A straight way to regioselectively functionalized polysaccharide esters. S. Massiot. Carbohydr. J. J. Biosci. L. 145‐152. 2004. 56. A. J. Cellulose 2004. W. 302‐307. Degee. 14(2) 127‐135. A.L. Hamaili. Aburto. Borredon: Free‐solvent Synthesis and Properties of Higher Fatty Esters of Starch – Part 2. 1. T. 57. M. Urtti: Determination of the degree of substitution of acetylated starch by hydrolysis. Al‐Higari: Acylation of starch with vinyl acetate in water. 261‐267. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. J. P. Mormann. 1735‐1743. . . H‐1 NMR and TGA/IR. E. D. . K. Sci. Res. W. 255‐263. Polym. Soininen. Starch‐Starke 2005. I. J. Carbohydr. 2003. Kussler. .. R. Koniarova. Mater. B. Alric. Dellacherie. Six: Partial or total silylation of dextran with hexamethyldisilazane. 118‐121. . M. Asplund. Starch‐Starke 2004. 341‐347. Hyvarinen.
vinyl laurate.39) are totally amorphous whereas the low‐DS ones (DS= 1. The reactions were carried out in DMSO using basic catalysts (Na2HPO4.96). Keywords: starch esters.5 MPa. trans‐esterification. The thermal and mechanical properties of some representative product samples were determined. High‐DS products (DS= 2. elongation at break of 3‐26%.45‐1. The basicity of the catalyst and the reaction temperature have the highest impact on the product DS.75) are still partially crystalline. The thermal stability of the esterified products is higher than that of native starch. vinyl stearate .7‐3. The effect of the process variables (vinyl ester to starch ratio. catalyst intake. The results were adequately modeled using a non‐linear multivariable regression model (R2≥0. Mechanical tests show that the products have tensile strength (stress at break) between 2.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. and Na‐acetate). and modulus of elasticity of 46‐113 MPa.26‐2. K2CO3. 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.
Mullen and Pacsu [11. Sagar and Merill  studied the synthesis and properties of C4‐C6 esters of high amylose starch. The global starch production was estimated at 60 million tons in 2004 . The largest application area of plastics is the use as packaging material (37%) . 12] studied the synthesis and properties of of C1‐C6 esters of starch. Unfortunately. et al  and Aburto. Esterification is one of the oldest methods used to improve starch properties . however. However. et al . 0. The use of native starch as a building block for novel biodegradable polymers. and incompatibility with hydrophobic polymers . The use of vinylesters has also been explored [20‐22]. Alternative routes to avoid the use of acid‐chloride/pyridine combinations have been developed. is rather limited to date. The limited use is mainly because of a number of unfavourable properties of starch such as low moisture resistance. 86 . and this causes serious environmental problems [1. and food service‐ware . Examples are agricultural mulch . Thiebaud. most of the plastic waste is not biodegradable. Starch is an attractive feedstock for the synthesis of novel biodegradable plastics. However. high brittleness. Most early studies mainly focused on the use of short‐chain carboxylic acid (C1‐C4). polyethylene  and polycaprolactone  and chemical modification by esterification with organic acids [11‐19]. Introduction Plastics are very useful materials and are used in large amounts (245 million ton per annum) [1‐2]. the mechanical properties were still not satisfactorily. the research activities were mainly limited to the use of vinylacetate. et al [15‐19] synthesized longer chain fatty esters (C8‐C18) of potato starch and corn amylose using fatty acid chlorides and pyridine.1. Two examples using higher vinyl esters (vinyl laurate) were reported by Mormann. Chemical or physical modification is required for successful applications as biodegradable polymers. The development of novel biodegradable plastic materials capable of decomposing when given an appropriate environment and time is of utmost importance . 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. It is cheap and abundantly available . Typical examples are blending with polyvinyl alcohol . however. and particularly on the synthesis of starch acetate (C2) [11‐12]. 3‐4]. The high DS esters showed interesting properties. The mechanical properties of products with longer ester chains (C4 and C6) products and plasticized acetate esters were considerably improved compared to native starch.65) . The products were hydrophobic and the mechanical properties were considerably improved compared to native starch.Chapter 5 5. the physicochemical properties of the products were not mentioned. packaging material. The products were biodegradable.
2. All these chemicals were used as received. pyridine. Japan). T‐Bones samples (with thickness of 2 mm) for determination of the tensile properties were prepared using a melt press 1 87 .N‐ dimethylaminopyridine (DMAP). Materials Corn starch (approx.2. The thermal properties (Tm. Analytical grade methanol. 5. This chapter describes systematic studies using Design of Experiments (DOE) to gain insights in the effect of process variables (temperature. Technical grade dimethyl sulfoxide (DMSO). as recommended by Laignel et al . 4‐N. catalyst type and intake) on the DS of the products. 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 . and acetic anhydride (Merck. Germany). Analytical grade vinyl stearate (Aldrich. and tetrahydrofuran (THF) were supplied by Acros (Belgium). After a first heating run from room temperature up to 200 oC to delete the thermal history of the material. Germany) were used without further purification.24 ‐ 2. and Tg) of each sample were determined from the spectra related to the cooling run and the second heating one.96) were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts in DMSO. 5. vinyl laurate (Fluka. Germany).1. sodium acetate.2. vinyl ester to anhydroglucose ratio. The samples (about 10 mg) were placed in sealed aluminum cells. Materials and Methods 5. the thermal and mechanical properties of representative examples of the highly hydrophobic materials are described and discussed. It was dried for 48 hour at 105 oC under vacuum (~1 mbar) to reduce the moisture content below 2 % before use. the Netherlands). and toluene were obtained from Labscan (Ireland). TGA measurements were performed using a Perkin Elmer TGA 7 Thermogravimetric Analyzer. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany).2. The spectra were recorded at 50 oC. and disodium hydrogenphosphate dodecahydrate (both Merck. In addition. The experiments were modeled using non‐linear multivariable regression. each sample was cooled to ‐50°C and then heated again to 200 oC (heating rate 10 oC/min). Tc. DSC analyses were performed on a TA Instruments DSC 2920. The catalysts for the esterification reaction were analytical grade potassium carbonate (Boom. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. 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.
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 126.96.36.199. 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. 188.8.131.52. 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 3.07 0.82 2.00 3.00 3.00 5.54 2. 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 3.00 3.00 2.00 5.00 3.00 2.70 2.47 1.00 5.00 5.00 3.00 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 0.87 1.00 5.24 2.00 5.00 5.96 0.00 5.50 0.00 5.50 2.00 2.00 5.79 Experiment Vinyl Ester 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate 91 .75 1.50 2.00 3.00 5.00 5.00 5.23 2.00 5.00 5.00 3.00 5.00 2.35 0.17 2.12 2.50 2.00 5.00 2.88 2.46 1.00 5.00 5.62 0.00 2.00 5.00 3.00 3.00 3.93 0.00 2.00 3.00 5.48 0.52 2.00 5.28 2.56 1.00 5.67 2.50 2.52 2. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.00 3.09 0.27 1.00 3.78 2.00 5.84 2.72 2.00 5.00 2.59 2.00 5.00 3.00 3.00 2.00 3.00 5.00 3.00 2.2.00 5.00 5.00 5.00 4.00 5.40 1.00 5.00 2.42 2.00 3.00 4.54 2.00 5.00 5.00 5.00 5.00 5.00 5.27 0.00 5.00 3.11 2.00 3.00 3.00 2.00 5.00 2.53 1.00 2.44 2.94 2.44 2.00 4.12 0.00 2.27 1.00 Catalyst Temperature [%‐w] [˚C] 2.41 1.32 1.00 4.
64 2.3.).0066 0.0023 0.0249 ‐0.98 2.6939 0.0006 ‐0.0009 0.00 3.00 3.00 3.1104 ‐0.64 2.1.3.0017 0.00 5.24 2.00 3.00 5.50 95 80 80 110 110 80 80 110 110 95 1.3.0027 ‐0.1. Mathematical Modeling Modelling of the DS data for both vinylesters was performed using non‐linear multivariable regression based on the 4 independent variables (Table 5.5650 0.00 5.1082 ‐0.00 5.0120 0.0470 0.00 2.0016 0.2183 0.00 2.0011 0.90 2.3. 5. stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 4.0008 ‐0.0327 ‐2.3.0003 0.The modelled values for βij for the two esters are given in Table 5.1947 β0 β1 β2 β3 β4 β5 β6 β7 β8 β9 β10 β11 β12 β13 92 . Values for the coefficients of the DS model for starch laurate and stearate Coefficient Starch Laurate 9.1568 ‐2.00 5.1979 Starch Stearate 17. 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 (5.0857 ‐0.0732 ‐0. The experimental data for both starch laurate and starch stearate esterification are best described with a quadratic model including interaction terms (eq.).59 2.21 2.0010 ‐0.00 5.00 5.96 2.) Table 5.Chapter 5 45 46 47 48 49 50 51 52 53 54 a.0008 0.00 5.00 2.41 mol ratio of the vinyl ester to AHG units of starch 5.50 2.1748 0. The type of catalyst was quantified using the basicity constant (pKb) in water [26‐27].00 3.0001 0.00 3.55 2.00 4.
Table 5.5 DS actual 2 2.4.96 x 10‐8 R2 values R2 R2adjusted R2press 0.5 0.2903 P‐value 1. starch laurate b.5 1 1 0. Analysis of variance for DS models for starch esterification Starch Laurate Model Error Total SS 16.5 2 DS predicted DS predicted 2 1.4. the adjusted‐R2 values are very close to the R2 ones.5 1 1.5 3 a. Finally.9531 0. which represents an “internal” validation method for the model. respectively and it may be concluded that the models describe the experimental data well.875 Starch Stearate Model Error Total SS 20.5 1.1.5 0 0 0. The very low P‐ values indicate that the models are statistically significant.4.939 0.629 P‐value 1.1.5 1 1.967 0. which indicate  that all significant variables are included in the model. The R2 values for the models are 0.6118 0.4986 16. starch stearate Figure 5. and confirm this statement.5 DS actual 2 2.5 2. Also in this case the reasonable R2 values (0.967 (starch stearate).5 3 0 0 0.8386 DF 13 14 27 MS 1. we also performed a PRESS analysis  (see corresponding R2 values in Table 5.3399 0.0356 F 35. 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 5.0583 F 27.805 3 3 2.2569 0.970 (starch laurate) and 0.970 0.945 0.).805‐0.875) indicate that the model correctly predict the 93 . Parity plots of the DS models for starch laurate and stearate Moreover.7699 DF 13 14 27 MS 1.8167 21. The parity plots of both models are given in Figure 5.026 x 10‐7 R2 values R2 R2adjusted R2press 0.
the performance of Na2HPO4 does not follow this trend. two plots showing the dependence of the DS on reaction temperature and catalyst amount (in %‐w). it can be rationalised that the rate of the reaction (and thus the product DS) will be higher when using a stronger base. A similar trend was observed by Dicke  for the acetylation of Hylon VII starch using vinylacetate.00) than for K2CO3 (2.). whereas catalyst performance is considerable lower. 94 . such as carbonate or acetate salts leading to higher DS esters. 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. Another explanation may be related to the regio‐chemistry of the reaction. However. This is indeed the case when comparing the performance of K2CO3 with CH3COONa. This was not the case for alkaline catalysts. The product DS for Na2HPO4 was considerably lower (1. 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. Unfortunately. These results may be rationalised by considering the role of the catalyst in the modification reaction. Catalyst performance of CH3COONa was slightly less than for K2CO3 whereas the lowest DS products were produced when using Na2HPO4 as the catalyst. the type of catalyst has the largest effect on the product DS for both the laurate and stearate esters (Figure 5. for both starch laurate and starch stearate are given in Figure 5.Chapter 5 products DS as function of the process variables within the range of experimental variables.82). To evaluate the effect of each variable on the product DS.2. K2CO3 is a stronger base (pKb 3. Of all the variables studied.2. The pKb for the latter (6. It is assumed that the first step in the reaction sequence is activation of the starch OH groups by deprotonation by a base [21.66) than CH3COONa (pKb 9. the base strengths of the catalysts in DMSO are not known. The highest product DS values were obtained using K2CO3. 22]. The resulting anion will react with the vinyl ester to from the product. On the basis of the sequence. 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.18) and CH3COONa (1.25) and this leads to higher product DS values for K2CO3.8) is intermediate between that of the other two catalysts.
As expected and in 95 .2. starch laurate b.). starch stearate Figure 5. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch a. although to a lesser extent. the catalyst intake and reaction temperature also affect the product DS (Figure 5. 3D contour plot of the DS as a function of reaction temperature and catalyst intake (at constant vinylester to AHG mol ratio of 4) Besides the type of catalyst.2.
Chapter 5 line with studies on starch acetates . higher reaction temperatures and catalyst intakes lead to higher product DS values.39) were determined.3.3.45 (. DSstearate = 2. . DSC analysis of starch stearate esters ( ) DS 1.39) 96 .2.) DS ( 2. . Of all variables studied.5.75. the vinyl ester to AHG ratio has the smallest effect on the product DS. 5.3.26. DSstearate = 1. Representative DSC spectra for the starch stearate are given in Figure 5.45) and high DS values (DSlaurate = 2. and the results for all samples are summarised in Table 5. Product Properties The thermal and mechanical product properties for two starch‐laurate and stearate samples with medium (DSlaurate = 1. heating Exothermic Tm = 21 oC cooling Endothermic Tc = 12 oC -20 -10 0 10 Temperature ( C) o 20 30 40 Figure 5.
d.39 1.a 21 Tc [˚C] n. especially in the amylopectin part [15. Thermal transitions at relatively high temperature (melting. 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.a 12 a. TGA analysis was performed to study the thermal degradation behavior of the starch laurate and stearate samples (Figure 5. 97 . Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. The melting point of the potato starch stearate was close to that of model compounds for the stearate side chains (methyl stearate. At moderate DS values the products still display a melting temperature (at 20‐30 oC) and a crystallization temperature (at 10‐25 oC). In particular the behaviour at relatively high temperatures (>0 oC) is further discussed. crystallization. 40‐42 oC. or glass transition above 0 oC) were even absent for the high DS products. Unfortunately the thermal transitions for the virgin material lie above the degradation temperature and can therefore not be accurately determined .d.75 2. 22]. implying that these products are fully amorphous. These values are reduced considerably compared to virgin corn starch. in the present work. 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).= not detectable The thermal behavior of the products is a strong function of the DS. octadecane. which is known to contain crystalline areas.).45 Tm [˚C] n.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.26 1. close to the value obtained in our study for the medium DS starch stearate (21 °C).d.a 27 n.d. 17].8 .d. However. To the best of our knowledge. these are the first examples of completely amorphous starch esters. For instance a Tm of 32 oC was reported for a potato starch stearate ester with a DS of 1. Changes in the transition temperatures of native starch by esterification have been reported in the literature [13‐14. In any case. 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. 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 2.5.
4. .39) The TGA (Figure 5.). elongation at break and elasticity modulus) of the products were determined (Table 5. . Native corn starch degrades at lower onset temperatures (250‐300 oC) than the corresponding esters (300‐350 oC).): starch stearate (DS 2.4. ( ): native starch ( ): starch laurate (DS 2.) curves clearly show starch esterification with either laurate or stearate results in products with enhanced thermal stability. The mechanical properties (stress at break. Similar trends were reported for amylose octanoate‐ stearate‐laurate esters  and potato starch octanoate‐ laurate esters . TGA analysis of various starch samples.Chapter 5 100 80 %-weight 60 40 20 0 20 100 200 300 400 o 500 600 700 Temperature ( C) Figure 5. 98 . The difference in thermal stability between the laurate and stearate sample is limited.6.26) (.
26 1.1 ± 1.0 3.1 Elongation at break [%] 26. an increase in product DS results in lower tensile strengths and modulus of elasticity. calculated as the ratio between the stress at yield (σy.0 3.2 8. however. Starch stearate with a relatively low DS (1. is roughly 0.0 ± 2. determined from the stress‐strain curves as the stress at which the material ceases to be linearly elastic) and the modulus.7 ± 4.5 ± 1.1 ± 15. observed for the high DS starch stearate sample. thus indicating a relatively more rigid material with respect to all other samples.).39 1. but higher elongation at break. On the other hand. The values of the yield strain.45 Stress at break [MPa] 3.39) does not show any plastic behavior but only an elastic one. and lower elongation at break).9 ± 14. An inverse behaviour is.6. 17].05).45) displays a very similar behavior with respect to the laurate samples (i.7 ± 3.3 3. The corresponding yield strain (roughly 0. a clear plastic behavior with a maximum in the stress‐strain curves. plastic deformation and yield strain of about 0.7 ± 4. independently of the DS values. This is in agreement with previous studies regarding the mechanical properties of starch esters with different chain length of the fatty acid moiety [13. This indicates that their mechanical behavior is comparable to that of typical engineering polymers .5 MPa.1 Modulus [MPa] 59.3 21. 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. This product is more rigid compared to the product with medium DS value (higher tensile strength and elasticity modulus. Mechanical properties of the starch esters DS 2. 99 .2 ± 0.7‐3. starch stearate at relatively high DS values (2.7 ± 1.e.8 Starch Ester Starch laurate Starch laurate Starch stearate Starch stearate All materials generally show tensile strengths between 2.1 82. For starch laurate. elongation at break between 3‐26%. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.0 112. A deeper understanding of the mechanical behavior can be gained by investigating the shape of the stress‐strain curves (typical examples in Figure 5.9 46.5. and modulus of elasticity between 46‐113 MPa.2 2.7 ± 14.025) lies still in the typical range of engineering polymers but also close to that typical of composite materials .75 2.3 ± 2.05 for both laurate samples. Both starch laurate samples display.
45) (‐ . ‐): starch stearate (DS 2. 100 . .5. ( ): starch laurate (DS 1.).39). . 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.Chapter 5 6 5 4 stress (MPa) 3 2 1 0 0 5 10 15 20 strain (%) 25 30 35 Figure 5.26) (.): starch stearate (DS 1. Stress‐strain curves for starch esters.7.75) ( ): starch laurate (DS 2. Although there are many differences in the synthetics methods as well as in testing conditions and procedures.
we show that although no crystallization of C18 occurred (see DSC data of the high DS material in Table 5.7 MPa) coupled with a significant (1500%) elongation.a. 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.7) 1. n.).75) 3.a. 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. In agreement with our own data.0 26. 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 ±0.7 ±0.5.a. 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) . Here. as confirmed by DSC. In this respect the difference in stress and elongation at break for the high‐DS starch laurate sample is striking. an unexpected behavior was also observed for the mechanical properties of the starch esters from potato starch.a.= not available Inspection of the mechanical properties of the starch esters as shown in Table 5. for potato starch exactly the opposite is observed: relatively low stress (0.26) 3. On the other hand at low DS values comparable stress at break (3.6 9 ±2 High DS (2.0 ±2.1 21.7) 0. Moreover.3 Potato Starch Stearate Ester Low DS (1.3 10 ±2 Property Low DS (1. 3.45) 3. merely because of the structural property of the C18 side chain with respect to the starch backbone. High DS (2.9 ±0.1 High DS (2. 9% for potato starch).39) 2. They related this inverse property of starch stearate to crystallization of C18 side chains.5 MPa) is coupled with a low elongation (26 %).5 ±1. confirms that the corn starch esters synthesized in our research are relatively rigid materials as compared to the potato starch esters. the high DS material is still more rigid.7 ±4.2 8.7 ±1.7 MPa for potato starch ones) are coupled with significant differences in the elongation (22% for corn starch esters. but the interplay between those factors. n.7 ±0.7. DS) which mainly determines the thermal and mechanical behaviour.1 ±15.26) relatively high stress at break (3.2 MPa for corn starch esters.7.4 1500 ±8.3 High DS (2. 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).8) 3. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.1 ±1. While for corn starch (DS 2.0 3. 101 .a.6 Corn Starch Stearate Ester Low DS (1.7 ±3.2 Stress at Break [MPa] Elongation at Break [%] a.
on the synthesis of corn starch esters with long chain fatty acids is described. 0. and Na‐acetate). catalyst intake.5. Conclusions A systematic study. The DS of the products is a strong function of the basicity of the catalyst. 5. and catalyst basicity) to the DS of the starch ester products were developed.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐starch mol ratio. K2CO3.07‐1. The use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2.5 for starch laurate. 1. The models may be used to determine the appropriate process conditions to obtain a product with a pre‐defined DS. Statistically adequate (R2 ≥0.3‐1.4‐3 for starch stearate).1‐2. while the use of Na2HPO4 resulted in low‐medium DS products (0.4. 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.Chapter 5 5. reaction temperature. 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.9 for starch laurate. Nomenclature β : regression coefficients obtained by a multiple regression procedure [various units. Thus. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] 102 . including statistical modelling. Comparison with literature data indicates that the properties may be (fine) tuned also by the starch source. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. 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). Reaction temperature and catalyst intake also affect the product DS although to a lesser extent than the type of catalyst.
Dev. Blackie Academic & Professional. 2000. USA. Copenhagen. 2008. 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.J. Packaging. J. References .html. Germany. .6. Culter: Plastics Packaging. Glasgow. Association of Plastics Manufacturers. 2002. Hanser Publishers. M. Marsden: The challenge of domestic waste disposal. H. Processing. Denmark. S. Weber). Germany.W. in Packaging in the Environment (Ed. J. W. and Regulations. 103 . . Plastics Europe. Ehrenstein: Polymeric Materials. C. . F. G. A. M. R. 592‐595. Denmark. Properties. Princeton University Press. demand and recovery for 2006 in Europe. Conference Proceedings. 2008 . Applications. The Royal Veterinary and Agricultural University. An Introduction to the New Science of Biodegradable Plastics. Department of Dairy and Food Science. Otey. 1980. New Jersey. M. 2000. 19. P. Hernandez. Hanser Publishers. Munich. accessed on August 01. The compelling facts about plastics: An analysis of plastics production. Aarhus. Belgium. S. Doane: Starch‐based blown films. Ind. Eng. . Brussels. . in The Food Biopack Conference: Foodstuffs. Prod. Munich. Chem. E. Denmark: http://www. D. R. Levy). 27‐29 August 2000 (Ed. Res. E. Westhoff. Bastioli: Global Status of the Production of Biobased Packaging Materials. Stevens: Green Plastics. Website of International Starch Institute. UK.starch. Structure – Properties ‐ Applications. and Biopolymers. Frederiksberg C. G. 2000. . C.dk/isi/stat/rawmaterial. Selke. 1986.
E. J. J. Senocq. E. E. S. S. 51. A. Starch‐Starke 2004. Hamaili. J. Pacsu: Starch studies: preparation and properties of starch triesters. I. characterization. E. Aburto. Beenackers: The kinetics of the acetylation of gelatinised potato starch. 302‐307. Sci. Alric. Aburto. 1647‐1656. status and perspectives. Prinos. Z. A. C. . . E. W. J. 35. J. Appl. C. 1942. K. Rijksuniversiteit Groningen. 57. Janssen. 1995. 1209‐1217. Sci. Starch‐Starke 1999. Aburto. . C. . Macromol. de Graaf: Ph. W.D. Borredon. Starch‐Starke 1999. Alric. Alric. Starch‐ Starke 2005. . Alric. R. Sci. Sci. Alric.Chapter 5 . 118‐121. 705‐721. Merrill: Properties of fatty‐acid esters of starch. Pacsu: Starch studies: possible industrial utilization of starch esters. 58. J. 1995. J. 132‐135. D. . 2003. M. 28. K. and biodegradability of fatty‐acid esters of amylose and starch. 65. Bikiaris. P. Bikiaris. Aburto. C. . Rev. L. Polym. E. J. B. 1999. J. I. C43. Weber). Thiebaud. the Royal Veterinary and Agricultural University. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Department of Dairy and Food Science. Carbohydr. Eng. Sagar. A. Thiebaud. Panayiotou: Synthesis. M. A. J. Polym. I. 381‐384. Appl. E. J. Polym. I. . . Part C: Polym. Polym. Alric. . Prinos. S. Thesis. E. Yang. Mouysset‐Baziard. E. 1997. Mullen. Carbohydr. 1440‐1451. G. Y. 51. 145‐152. Mullen. Appl. I. W. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. 385– 409. J. Aburto. 74. 56. D. J. Denmark. Chem. 137‐144. 1943. D.. A. Chem. D. J. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. J. M. de Graaf. Polym. Thiebaud. C. J. 101‐112. X. A. H. Borredon. Al‐Higari: Acylation of starch with vinyl acetate in water. F. 34. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Borredon. Panayiotou: Properties of octanoated starch and its blends with polyethylene. R. L. Wang: Properties of starch blends with biodegradable polymers. Biobased packaging materials for the food industry. 1997. 1996 104 . 2000. W. Mormann. Eng. Bikiaris. Broekroelofs. Ind. Wang. Frederiksberg C. .. Prinos. G. Aburto. I. Ind. . . 34. report of Food Biopack Project (Ed.
New Jersey. B. . Underwood: Quantitative Analysis 5th edition.. Carbohydr. Laignel. 105 . Montgomery: Design and Analysis of Experiments 5th edition. H. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Materials: Engineering. M. Massiot. Prentice‐Hall. J. H. Manurung. 1986. A. . Butterworth‐Heinemann. New York. M. D. C. R. Day. Ashby. 2008. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. A. Amsterdam. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch . 2001. Processing and Design. John Wiley & Sons Inc. B. G. F. USA. Res. Cellulose 2004. J. 2007. 251‐260. Science. C. L. 1997. R. M. USA. Junistia. 298. D. Dicke. A. Cebon. Janssen. the Netherlands. P. . R. Bliard. 11. accepted for publication in Starch‐Starke. . Picchioni. Shercliff. A. Sugih. . K. 255‐263. L. L. A straight way to regioselectively functionalized polysaccharide esters.
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. the mechanical behavior of the compatibilized blends is in general different from that of pure PCL and of the corresponding uncompatibilized blends. Keywords: starch‐PCL blend. In particular the elastic modulus for the compatibilized blends is significantly higher than that of uncompatibilized ones. despite the expected differences (favorable to PCL‐g‐GMA) in the in situ formation of the compatibilizers. grafting . The reactive interfacial agents were tested for their performance in PCL‐starch blends. As a result. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). PCL‐g‐ glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM) is decribed. at relatively high GMA and BPO intakes. glycidyl methacrylate. compatibilizer. This discrepancy could be explained on the basis of the functional groups (GMA or DEM) distribution along the PCL backbone. At relatively low starch intakes. 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. PCL‐g‐DEM has at least a comparable performance with respect to PCL‐g‐GMA. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. diethyl maleate. Both act as compatibilizers for PCL/starch blends by improving the interfacial adhesion between the starch particles and the PCL matrix. 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.
To reduce the tendency for phase separation. is known to be degraded with ease by microorganisms widely distributed in nature . synthetic biodegradable polymers have been applied. blends of hydrophilic starch and hydrophobic biodegradable polyesters exhibit phase separation  due to differences in polarity of the building blocks. to an ideal compatibilizer having both PCL and starch blocks. Blending of starch and synthetic biodegradable polyesters has been widely applied for the synthesis of fully‐biodegradable products. such as polyethylene (PE) or polystyrene (PS) [3‐6]. 108 .e. Starch modifications to improve the product properties like enhanced hydrophobicity and mechanical properties were already reported in the early 19th century [1.Chapter 6 6. 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. as in this study. The latter displays the presence. When using ungelatinized starch as a component in the blend. This phenomenon is highly undesirable and limits the application range considerably . As a consequence. polycaprolactone (PCL). these synthetic polymers are poorly or non‐biodegradable. Introduction Starch is a cheap and abundantly available natural polymer with very good application perspectives in the area of biodegradable plastics. However.1. 2]. for starch/PCL blends. of polar groups (usually epoxides or anhydrides [13‐18]) able to react with the – OH groups along the starch backbone.e. This leads. The ester bonds are susceptible to attack by water and this leads to enhanced biodegradability. upon mixing) by using a functionalized PCL. a compatibilizer (i. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decrease rapidly in time . 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. A well‐known biodegradable polyester. 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 ). along the backbone. Among these. it would be actually more accurate to define the block copolymer as an “interfacial agent”. PCL has gained considerable interest for possible applications in the fields of packaging materials and medical applications [10‐11]. Unfortunately. Such structure is rather difficult to achieve by simple copolymerization methods and it is usually prepared in situ (i. One of the well known modification strategies is blending the starch with polymers displaying a stronger hydrophobic character and better mechanical properties. Unfortunately. an interfacial agent) may be used to improve the interfacial association between the two polymer phases. To overcome this issue. polyesters are considered very promising alternatives .
Germany and methanol (99. systematic studies and subsequent quantification of the functionalization reaction has not been reported to date. 73% amylopectin and 27% amylase) was obtained from Sigma and high molecular weight PCL (CAPA 6503. Exploratory studies on the synthesis of PCL‐g‐GMA and its applications for starch/PCL blends have been published [13.2. After the PCL was melted (1‐2 minutes). Mw=50000) from Solvay Caprolactone.2. Intakes for each experiment are given in Table 1 and 2. Materials Polycaprolactone (PCL) CAPA 2304. The synthesis and application of PCL‐g‐DEM is.2. to the best of our knowledge. a solution of BPO in GMA or DEM was added drop by drop over a period of 5 minutes. The materials were mixed for another 5 minutes. 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. The effect of important process variables on the degree of grafting of the GMA and DEM units to the PCL backbone has been explored in detail and quantified using multivariable linear regression. The intake of reagents was maximally 75‐80% of the volumetric volume to ensure proper mixing. The various compatibilizers have been tested for their performance in PCL‐starch blends.1. Belgium. diethyl maleate ≥97% purity (Fluka). The kneader was heated to 130 oC and PCL (CAPA 2304) was added while maintaining a rotational speed of 80 rpm. Mw=3000 from Solvay Caprolactones. after which the equipment was stopped and the chamber was opened to collect the samples. Compatibilizer synthesis The compatibilizers were prepared in a Brabender Plasticorder PL2000 batch‐ kneader (chamber volume 35 cm3). UK. Tetrahydrofuran (THF. This low molecular weight PCL grade was used without further purification. 109 . Glycidyl methacrylate 97% purity (Aldrich).2. Corn starch (with approx. Ireland. xylene (99.2.8%) from Merck. This chapter describes a systematic study on the synthesis of two reactive interfacial agents for starch‐PCL systems. 6.8%) from Labscan. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM).1. Materials and Methods 6. UK was used for the preparation of the interfacial agents. The starch was dried for at least 24 h at 110oC under vacuum (approx. ~1 mbar) prior to use. an absolute novelty of the present research. however. 6. Methods 6. >99%) was obtained from Acros. and benzoyl peroxide 75% (Merck. Germany) were used as received.2. 16].
Work‐up of PCL‐g‐GMA products  To remove unreacted GMA monomer and GMA homopolymer. Each sample was first heated from 0 oC to 100 oC (heating rate 10 oC/min) to remove the thermal history of the material.2. and the product was precipitated at 6‐8 oC. stirred for 1. 5 mbar). 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). PCL‐g‐DEM (5 g) was dissolved in xylene and refluxed at 150 oC for 2 h. Infrared spectra were collected with a FT‐IR apparatus in the ATR mode using a Spectrum 2000 instrument from Perkin Elmer. the chamber was opened and the resulting material was collected. Aldrich) as the solvent. Digital Scanning Calorimetry (DSC) measurements were performed using a Q1000 TA Instruments equipped with a TA Instruments DSC cooling system.2. 5 mbar). 6.2. PCL‐g‐GMA (5 g) was dissolved in 50 ml THF. PCL was added to the chamber followed by the addition of the starch and the reactive compatibilizer. 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 . created using a Cressington 208 sputter coater.2. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. and then filtered.Chapter 6 6.3. The resulting suspension was filtered and precipitated using methanol (450 mL) at 6‐8°C.5 h. An operation temperature of 170°C and a rotation speed of 80 rpm were applied . Subsequently. Before analysis. Scanning Electron Microscopy (SEM) was performed using a Jeol 6320 F Scanning Electron Microscope.184.108.40.206%.2. the samples were covered with a paladium/platinum conductive layer of 3 µm thickness. 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).2.2. The content was blended for 15 minutes. 6. 6. Analytical Methods H‐NMR measurements were performed using a 400 MHz Varian AMX Oxford NMR apparatus with CDCl3 (99. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC. The error on the transition temperature is assumed to be ± 1 oC and 5 % of the calculated values for the corresponding enthalpies. Tensile tests were performed using an Instron 4301 1 110 . Methanol (450 mL) was added to the filtrate.
As a result we were able to obtain a model for the FD of the reaction.2. 15‐16. The T‐bone samples were prepared using a Fontijne Holland TH 400 hot‐ press.2 ppm) side chains with that of a characteristic proton resonance of the PCL backbone (‐CH2‐ protons at δ 4. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends machine. The latter is a measure of the statistical significance of the model.2. leading to a 10% relative error in the FD values.) The FD was calculated using 1H‐NMR by comparing the area of protons belonging to the GMA (‐CH< proton at δ 3. A 5% relative error in the peak area of the NMR spectra was assumed.3. 111 . strain at break (ε).0 ppm [13. stress at break (σ) and modulus (E) were measured.1. The validity of the model was determined by performing an analysis of variance (ANOVA. 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). 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. 6. For every T‐bone. 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) (6.4. 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.1. For a given sample/blend. 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]. the F‐ value for the model is determined followed by the P‐value. it is possible to calculate the mean square (MS) for the model and the error. 6.2 ppm) or DEM (‐CH2‐ protons at δ 4. 8 different T‐bones were used. On the basis of the latter values. When the relative degrees of freedom (DF) are known. Table 3).
1.Chapter 6 6. Results and Discussions 6.). 112 .1.1. COOEt COOEt * O O n DEM BPO n * PCL-g-DEM * O O * GMA BPO * O O O O O n * PCL-g-GMA Scheme 6.3. 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. Functionalization reactions (only showing reactivity for the >CH2 in α position on PCL backbone) Typical 1H‐NMR spectra for the products are shown in Figure 6.1.3.
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
6% mol/mol-CL units DEM. 6. higher FD values for GMA than for DEM are in line with this explanation and support the proposed molecular reaction mechanisms. The results are graphically provided in Figure 6.6% mol/mol-CL units 15 GMA. Effect of GMA and DEM to CL‐units ratio (mol/mol) on the FD of the products (constant PCL intake. However. BPO=0.1.1% mol/mol-CL units 10 5 0 10 20 30 40 50 60 Substrate Intake (mol% on CL units) Figure 6. This trend is independent of the BPO amount and matches with data reported by other groups [13.3% mol/mol-CL units DEM. BPO=0. 116 . Hence. showed that MA reacts easily with a radical on the PCL backbone.).e. i. 30 25 20 FD (%) GMA. the length of a MA graft is always unity whereas longer grafts are possible for GMA. 22].3. Our experimental findings.2. subsequent reactions of MA to an already grafted MA molecule hardly occur. BPO=0. 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. 130°C) Higher GMA intakes lead to higher FD values.3. 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. Thus an increase in the GMA intake will provide more monomer available for the growing of the grafted chains leading to higher FD values.Chapter 6 chemical structure of DEM. BPO=1.1.
8 0. at relatively lower GMA intakes.4 0.2. although a slight increase in the FD values might be appreciated. Apparently. no detectable influence of the BPO amount on the FD is observed (Figure 6.7 0. However. GMA=12% mol/mol-CL units DEM. Such behavior is slightly in contrast to what observed for maleic anhydride [13.9 1 1. This will lead to a higher proportion of PCL radicals by hydrogen abstraction from the polymer backbone and thus to higher FD values. These results are in line with earlier work [15. 17].).3. The FD values are within a rather narrow range.6 0.2 BPO Intake (mol% on CL units) Figure 6.3 0. Effect of the BPO intake on the product FD (130°C.4. 15.1.5 0. 6. DEM=30% mol/mol-CL units 0. The use of higher initiator concentrations will result in an increase in the number of formed radicals.2 GMA. GMA=24% mol/mol-CL units GMA. constant substrate to PCL ratio) For high GMA to CL ratio (24%‐mol/ mol CL units). for which a levelling off and eventually a decrease of the FD values for relatively higher MA amounts has been observed.1 1.4. there 117 . 50 45 40 35 30 FD (%) 25 20 15 10 5 0 0. 17]. Such discrepancy is probably related to the relatively low intake of DEM as well as to the different reactivity of DEM compared to MA . 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. doubling the amount of initiator results in considerable higher product FD.).4. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The experimental trend for DEM is different.
2. If many macroradicals are created in the system (relatively high BPO intakes) at low GMA intake.) 118 . Here.6022nm ni − 0. δ ) (6.2729nm niδ (6. In addition. This leads to the following equation: FD = f (nm . However. 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.Chapter 6 is an optimum ratio between the BPO concentration and available monomer (GMA in this case) on the FD. defined as the difference in solubility parameters between PCl and the substrates. the possibility of side reactions will become significant.) where nm is the molar amount of monomer in the feed. namely an increase in the BPO intake results in higher product FD.3. 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. 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).3.5431niδ + 1.1.8875 + 0. the BPO and monomer intakes are considered as independent parameters. To quantify synergic effects of monomer and initiator molar intakes on the FD of GMA and DEM on the PCL backbone.3. and 6. 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).3.2. The results described in the previous paragraph imply that the mutual interaction between these variables and not the absolute value itself determines the final FD values to a great extent. i. 17]. a statistical model has been developed by performing a multivariable linear regression on the data reported in Table 6. the mutual solubility of DEM in PCL and DMA in PCL was included in the model by using a parameter δ. 6.e. 6. recombination of (macro) radicals. gives the best description of the experimental data: FD = −1. ni .0325nmδ + 0.1. This phenomenon is likely related to the different nature of the radical grafting mechanism of GMA and DEM on PCL as mentioned before. as well as several transfer reactions might be responsible for the observed trend . The model provided in eq. In particular the occurrence of “cage effects”. The effect is however much less pronounced than for GMA. ni the molar amount of initiator in the feed and δ the difference in solubility parameters calculated using group contributions .
6. Graphical representation of the FD models for PCL‐g‐GMA and PCL‐g‐DEM are given in Figure 6..5 0.928 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 ). SS 1828 81. the interaction between (PCL‐monomer) intake and solubility parameter difference ( nmδ ) and the interaction between the latter factor and the initiator intake ( ni δ ).5. a. This was also confirmed by inspection of the residue distribution by a normal probability plot (not reported here for brevity) . respectively.957) and its closeness to the adjusted R2 (0. 3D plot.3 10 10 20 30 20 0.358 1909. b. Contour plot.4 BPO Intake (mol% on CL units) 0.3.8 0. and Figure 6. The very low P‐value implies that the model is statistically significant. Table 6.2 10 15 20 25 30 35 40 GMA Intakes (mol% on CL units) 45 50 a. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The analysis of variance data are given in Table 6.6 0.358 DF 4 8 12 MS 456.9 60 70 40 50 80 BPO Intake (mol% on CL units) 0.941) also suggests that all important variables have been included in the model.2 10 20 GMA Intake (mol% on CL units) 40 30 50 0. 1 0.5.17 F 44.6 20 10 20 40 50 30 60 FD (%) 40 FD (%) 30 40 20 0 1 0.8 0.4 0. Figure 6. b.7 0. Graphical representation of the FD model for PCL‐g‐GMA. 119 .3.3.908 10. The R2 value for the model (0. Model Error Total Analysis of variance for the FD model provided in eq 6.
5 0. a higher BPO intake leads to a reduction of the FD values while an opposite trend is observed at lower DEM intakes. b. radical initiator and PCL intakes). This induces irregularities and is expected to result in a lowering of the Tc and the Tm.8 0.Chapter 6 1 0. the melting temperature and the relative enthalpy decrease with respect to pure PCL. and a favored nucleation 1 2 50 120 . and 6. a. 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.4. 6.7 3 4 6 FD (%) 4 0. et al  working with PCL‐g‐GMA.2 10 15 20 25 30 35 40 DEM Intake (mol% on CL units) 45 7 a.1.4 BPO Intake (mol% on CL units) 0. Thermal properties of the compatibilizers The thermal properties (Table 6. Graphical representation of the FD model for PCL‐g‐DEM. Contour plot. While for GMA an increase in BPO or monomer intakes invariably leds to higher FD values. 3D plot. 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. For all samples. Remarkable is once more the different trends for the two substrates.2 10 20 DEM Intake (mol% on CL units) 40 30 50 6 0. From a more practical point of view. which is expected to lead to a Tc reduction.6.6 2 FD (%) 5 2 0 1 0. 7 6 3 3 4 Figure 6.4 0.9 4 5 8 BPO Intake (mol% on CL units) 0. Similar observations have also been made by Kim.8 0. for DEM a clear transition is observed. For relatively high DEM intakes (> 30 % mol/mol).2.1. the crystallization temperature and enthalpy are not a clear function of the FD values. Furthermore. The random behavior of the Tc as function of the FD values is probably the result of two concurring effects: the presence of irregularities.) of the compatibilizers were determined by DSC.3 5 6 0. Such changes in the thermal properties compared to pure PCL may be caused by the introduction of grafts on the PCL chains. although both properties are significantly lower than those of pure PCL. b.6 0.3.
3.2 337.2.5 489.5 7.1 11. 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) 16.4 230.6 386. An overview of all prepared blends together with their thermal and mechanical (tensile tests) properties is given in Table 6.5 425. A total of 12 blends were prepared: a series of binary ones (not containing any compatibilizers) constituting our reference points.3 15.6 11.0 401.2 342. 6.4.8 372.0 430.3 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 1.3 380.0 9.9 E (MPa) 270.8 371.5 357.6%. a series with PCL‐g‐DEM (different intakes of the latter at fixed starch/PCL ratio) and two series with PCL‐g‐GMA (one with different intakes of PCL‐g‐GMA at a fixed starch/PCL ratio and one with a fixed compatibilizer intake at three different starch/PCL ratios).6 305.4.1 10. PCL‐g‐GMA has an FD of 9. 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.8 329.2 321. 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 10.6 5.5 ε (%) 640.7 %.4 431.1 13.9 424.9 379.8 384.2 343.6 168.0 332.9 11.2 9. 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. Table 6. 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. 121 .0 341.7 10.9 368.
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 . Melting and crystallization enthalpy as function of the starch content for binary blends with PCL. The thermal and mechanical properties indicate no or very little interaction of the starch particles with the PCL ○ : ∆Hc □ : ∆Hm 122 .8 MPa) clearly indicates that the rigidity of the blends increases at higher starch contents. 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. This is most probably due to a nucleation effect of the starch on PCL as observed also for other kinds of polymer blends . The simultaneous increase in the modulus (from 270.2.43 MPa and 640.5 % for pure PCL to 7.2 MPa up to 341. 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 6.1.8 % at 30 % starch content.Chapter 6 6.7.3. 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.).1 MPa and 341. 75 70 65 60 55 50 45 40 35 ∆ H (J/ g-PCL) 0 5 10 15 20 Starch content (%) 25 30 Figure 6.7. respectively from 16.
8. The thermal behavior is characterized (Table 6.e. the starch particles are simply inserted into voids of the PCL matrix. 6.2. Such behavior is consistent with the hypothesis that the starch particles hinder the melting and crystallization processes of the PCL chains. in order to fully understand the role of PCL‐g‐DEM.3. a.2. a decrease with respect to pure PCL is observed and then a substantial invariance as function of the PCL‐g‐DEM content.9. 26] and explained by assuming that the voids are formed by water in the starch. (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. and the lack of interface adhesion cause void formation. for which a 6 oC drop is observed when using PCL‐g‐DEM. Thus. Evaporation of the water during blend preparation. Ternary blends compatibilized with PCL‐g‐DEM PCL‐g‐DEM. Figure 6. The latter trend is also valid for the corresponding enthalpies: i. This is visually confirmed by the morphology of the prepared blends (Figure 6. In this respect the thermal properties remain substantially unchanged with the exception of the Tc. 123 .). was used as compatibilizer for the preparation of ternary blends with starch and PCL.4.) 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. the corresponding binary blend (S/PCL 20/80) contitutes a better reference point compared to pure PCL. b. either by heating or in combination with mechanical stress .8. The presence of these voids was also observed for sago starch/PCL blends [25. However. As consequence one might expect a slight different structure of these ternary blends as compared to the corresponding binary one. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends matrix.). Morphology of starch/PCL binary blends. c. This is confirmed by examination of the blend morphology by SEM (Figure 6. whose synthesis is described for the first time in this work.
10. 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. 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.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 .Chapter 6 a. resulting in more rigid material with higher modulus.).10. 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.). Figure 6. Morphology of ternary blends compatibilized with PCL‐g‐DEM. 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. Indeed starch particles in the ternary blends display a smoother interface with the PCL matrix compared to the corresponding binary blends (Figure 6. As a result of the different morphology also the mechanical behavior display differences with the virgin PCL and the binary blend. which will hinder the flowability and fibre forming ability of PCL matrix under cold drawing. This can be explained by the improved interfacial adhesion  between PCL and starch. b.9.). c.
it can be observed that: 1) the melting temperature as well as the crystallization and melting enthalpies do not change significantly (discrepancies in the values are within the boundary defined by the experimental error). the crystallization temperature experiences a drop of about 5‐6 oC.10. In particular also in the case of PCL‐g‐GMA.2.e.3. 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). in all cases (i.3. However. the stress and strain at break are lowered and hardly a function of the PCL‐g‐GMA amount. 6. the modulus is higher and hardly a function of the compatibilizer amount. by taking as reference the corresponding binary blend (S/PCL 20/80). at all PCL‐g‐GMA contents) there is a clear increase of the modulus with respect to the 125 . Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 390 380 370 360 E (MPa) 350 340 330 320 0 1 2 3 PCL-g-DEM amount (wt%) 4 5 Figure 6. 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. 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. Modulus of ternary blends S/PCL/PCL‐g‐DEM as function of PCL‐g‐ DEM amount.
11. This confines all reactive GMA groups in relatively “isolated” spots along the PCL backbone.7 % for PCL‐g‐DEM. However. As a result.11.8 % for PCL‐g‐GMA against 1. PCL‐g‐DEM is at least as active as compatibilizer (compare modulus values in Table 6. These differences in compatibilizing effect are not yet fully understood. partial interfacial adhesion) which render these ternary blends a kind of “intermediate” case between the binary ones and those compatibilized with PCL‐g‐DEM.). As given schematically in Figure 6.) than PCL‐g‐GMA for blends containing 20 %‐wt of starch.11. upon reaction of PCL‐g‐GMA with one of the –OH groups on the surface of the starch particles the remaining GMA groups are actually scarcely available for further reaction with other –OH groups spatially distant from the reacted one. (a) 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. and 6. 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.12. (left) the structure of PCL‐g‐GMA is inhomogeneous at the molecular level with long poly(GMA) branches pending from the PCL backbone. 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. 126 . presence of voids. c. This is surprising if one takes into account the different FD values: 9. 29]. Morphology of ternary blends S/PCL/PCL‐g‐GMA. 28. Thus. and morphologies in Figures 6. b. Figure 6.9.4.Chapter 6 binary blend. 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.
) and relating the “coverage” of the starch particle surface upon reaction with PCL‐g‐GMA with the compatibilization efficiency.4. other groups will presumably remain available (arrows in Figure 6. once PCL‐g‐DEM has reacted with one –OH groups on the starch particle. 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. 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. In order to check this we compared blends with the same amount of PCL‐g‐GMA (2 %‐wt) but with different starch intakes (10. thus probably ensuring a better coverage of the surface. when comparing ternary blends (compatibilized with PCL‐g‐ GMA) with the binary ones. 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. On the other hand (right of Figure 6. The latter result (from a modulus of 341. the Tc decreases (with respect to the corresponding binary blends) while all other factors (Tm and enthalpies) remain virtually unchanged with respect to the binary blends and also as function of the starch intake. for all starch amounts (Table 6. As consequence. Concerning the thermal behavior (Table 6. Indeed.) PCL‐g‐DEM has much less reactive groups (lower FD) than PCL‐g‐GMA.). We can therefore conclude that the efficiency of PCL‐g‐GMA in the 127 .) for further reaction. Moreover. Schematic representation of the in situ reaction between functionalized PCL and starch.3 MPa for S/PCL/PCL‐g‐GMA 30/70/2) clearly indicates that the efficiency of PCL‐g‐GMA as compatibilizer becomes more relevant at relatively higher starch contents (>30 %) . the distribution of the reactive group along the PCL backbone is more “homogenous”. 20 and 30 %‐wt respectively).12. but since DEM is preferentially grafted as monomer. This is in agreement with the hypothesis made above (Figure 6. thus attenuating the effect discussed above. at higher starch contents more –OH groups would be available for reaction with the GMA groups.220.127.116.11.8 MPa for S/PCL 30/70 up to 430.).
The most important variable for the final product FD is the interaction between the amount of monomer and initiator used. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends.Chapter 6 compatibilization of starch/PCL blends can be significantly improved by changing the blend composition. Nomenclature E FD : initial modulus [MPa] : functionalization degree. The PCL‐g‐GMA and PCL‐g‐DEM compatibilizers display lower melting temperatures and melting enthalpies compared to virgin PCL. including statistical modeling. 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 . From a more practical point of view it can be concluded that the newly synthesized PCL‐g‐DEM. has been performed on the synthesis of two reactive compatibilisers. 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. mutual solubility of the monomer into the molten polymer) on the FD adequately described the experimental results (R2=0. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). firstly reported in this work. Conclusions A systematic study. 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.4.5. 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 reactive compatibilizers were used in blends of starch with PCL. The proposed model to quantify the effects of process variables (monomer and initiator intake. 6. P‐value ≤10‐9).957. 6. 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.
J. . Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – IV. 1998. 227‐242.D. References . E. Adv. Polym. . Polym. Karlsson. CRC Press. 1997.B. Albertsson. Eur. 34. J. 1942. F. B. Davis: High performance LDPE/ thermoplastic starch blend: a sustainable alternative to pure polyethylene. Ioannis. 1209‐1217. Ramsay. E. Ind. N. H. A. 38. Effect of amylopectin to amylase ratio in starch. 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. M. Rodriguez‐Gonzales. Mani.G.Chem. Sci. W. Y. Kawasaki: Biodegradable films made from low density polyethylene (LDPE). R. 34. Pacsu: Starch Studies: Preparation and properties of starch triesters. Ogawa. 129 . Wurzburg). 44. Part 2. . 34. C. Ranucci. Eng. M. Carbohydr. Thermal and morphological properties. S. Inc.J. Jarowenko: Acetylated starch and miscellanous organic esters. A‐C. Mani. B. Sci. Boca Raton. 33.S. Lindblad.Eng.6. . E. 1986. Eur. Biliaderis.A. Karlsson: Polymers from Renewable Resources. 1467‐1475.. O. M. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – III. 2002. Mullen. 1477‐1487. Liu. Polym. 1998. 1517‐1526. Polym. A‐C. J. Psomiadou.W. R. 1998. Polymer 2003. 1251‐1253. 139‐161. in Modified Starches: Properties and Uses (Ed. USA. Polym. 157. . . S. Albertsson: Biodegradable Polymers and Environmental Interaction. . wheat starch and soluble starch for food packaging applications.
2001. E. P. 2003. Polym. A. . J. Polym.W. Chem. Polymer 2001. Rustgi: Biodegradable polymers. John Wiley & Sons Inc. . Moro. C. . J. Cho. 1995. 2004. It. 3.S. Passaglia. Kim. Averous. J. M.Chapter 6 . 1997. Errico. M. R. J. Ruggeri. 3875‐3881.Y. Dole. G. P. R. J. 5135‐ 5142. 61‐67. L. Chim. Polym. Inoue: Compatibilizers for melt blending: Premade block copolymers. K. Khandpur. USA. E. 1139‐1148.Y. Kim. Sci. Sci. Z. . Macromolecules 1996. C. 12. 81. Environ. Bhattacharya. 41. 4157‐4167.S. 1991. J. Gazz. Environ. . C. M. A. 1997. Park: Modification of Aliphatic Polyesters and Their Reactive Blends with Starch. 35. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. C. . . D. L. Polym.K. T. 41. Polym. 130 . J. Coleman.M. Graf.C. Aglietto: Chemical reactions affecting the free‐radical grafting of diethyl maleate onto ethylene polymers. Sci. M..H. New York. K. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. Raimo. P.F. E. M. Polymer 2000. Polym. Nakayama. R. Stab.. Part A: Polym. USA. Kim. Tang. J. Solaro: Biodegradable polymeric materials. 305‐313. M. Painter: Specific Interactions and the Miscibility of Polymer Blends. 127. Guegan. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. 5590‐5598. R. Part A: Polym. Chiellini. Park: Grafting of glycidyl methacrylate onto polycaprolactone: preparation and characterization.K. 1693‐1702. Marechal. Degrad. J. Macosko. M. . C. Adv. K. Mani. Laurienzo. C. Appl. Sci. . 29.E. P. 8. Chandra. 179‐187. Montgomery: Design and Analysis of Experiments 5th edition. Cho. J. Marrucci. 37. Fringant: Properties of thermoplastic blends: starch‐polycaprolactone. Polym. 1273‐1335. Kim. 91‐95. 1996. . R. .M. 2001. 80. Polymer 2000. 1998. Rimedio: Preparation and characterization of compatibilised polycaprolactone/ starch composites. Tang: Functionalization of polyesters with maleic anhydride by reactive extrusion. . .K. J. 127–134. D. M. J. Prog. Avella. Mater.H. 23. John. C. Goldberg: A review of the biodegradability and utility of poly(caprolactone). J. Yang. 1507‐ 1516.H. 1999. Martuscelli.. Degrad. . Technomic Publication. Chem. Pennsylvania.K. Jung. 42. P.
Safa. Odusanya.M. 2006.Nasirtabrizi: Ring opening reactions of glycidyl methacrylate copolymers to introduce bulky organosilicon side chain substituents. 2002. Ishiaku. Kammer: On mechanical properties of sago starch/ poly(ε‐caprolactone) composites. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends . K. G. Polym. Eur. . Picchioni: Formation and compatibilizing effect of the grafted copolymer in the reactive blending of 2‐ diethylsuccinate containing polyolefins with poly‐ε‐caprolactam (nylon‐6). H. Ruggeri.M. D. 40.S. U. .A. Polym.S. 38. M. D.M. S.W. Ishak: Mechanical properties and enzymic degradation of thermoplastic and granular sago starch filled poly(ε‐ caprolactone). Sci. Passaglia. 293‐304. U. 2000. 393‐401. U. 131 . Sci.S. Aglietto.S. M. Polym.M. B. Azemi: Effect of starch predrying on the mechanical properties of starch/ poly(ε‐caprolactone) composites. 2003. 87. Manan. Tedesco.V. . 9.M.N. .W. Barbosa. 57. E. Technol. Eng. Bull. O. J. B. Polym. J.B. M. S. 1998. O.A. R. 21. Azemi.S. W. Polym. K. Pang. Polym. F. Ishiaku. 2002. 11‐15. Adv. Manan.D.N. Appl. Test. Mauler: Comparative study of PP‐MA and PP‐GMA as compatibilizing agent on polypropylene/nylon 6 blends. Ishiaku.H. .S. 273‐281. R. A. 1298‐1305. Lee. Nachtigall. 877‐884. Odusanya.
it has a high volume to weight ratio and is generally resistant to degradation. there is a strong need for renewable alternatives for plastics from fossil resources.Summary Plastics made from fossil resources (a. There are two types of polymers present in starch: amylose and amylopectin. containing periodic branches linked with the backbones through α‐D‐(1. Its production has increased significantly since 1950. Examples are the use as packaging and construction materials. 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. Starch is a polymer consisting of anhydroglucose (AHG) units. Starch‐based biodegradable materials are considered interesting candidates to replace certain types of conventional plastics. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. The application of a number of prospective methods for developing starch‐based biomaterials will be described in this thesis. In combination with the current high prices for petrochemical products. Starch is relatively cheap and available from a broad range of plants. while amylopectin is branched polymer. Starch must therefore be modified before it can be applied as a biodegradable plastic. To gain insight in the potential of this approach. we have . with a rate of almost 10% every year. After use.o oil and gas) are very attractive materials for a broad range of applications. plastics may end up in the environment and cause serious environmental problems. such as thermoplasticization. The use of virgin starch for packaging materials is restricted because it cannot be shaped to films with adequate mechanical properties (high percentage elongation. 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. Starch is also too sensitive to water. chemical modification or combinations thereof. For instance. blending with other materials.6)‐glucosidic bonds. The main reasons are the water sensitivity of common catalysts and the fact that starch is poorly or even insoluble in the common organic solvents used for ROP. The content of amylose and amylopectine in starch varies and largely depends on the starch source. tensile and flexural strength).4)‐glucosidic bonds. 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 are several starch modification methods available.
The results are described in detail in Chapter 3.68) was obtained. The silylation reaction was performed using hexamethyl disilazane (HMDS) in DMSO/ toluene mixtures at 70°C. Most of the previous starch esterification studies involved the use of short chain carboxylic acids (C1‐C4). even in the presence of plasticizers.21‐0. Poly‐(ε)‐ caprolactone grafted silylated starch co‐polymers with average chain length of 40‐ 55 monomer units (molecular weight of 4500‐6300) were obtained.2. 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. with higher ratios leading to higher molecular weights. The hydrophobicity of starch acetates is higher than virgin starch. The polymerizations were performed at 60‐100°C. temperature and monomer: monosaccharide ratio) on the average degree of polymerization. Characterization of the products using 1H‐ and 13C‐NMR as well as MALDI‐TOF mass spectrometry showed the presence of significant amounts of p‐dioxanone polymers with an isopropoxide end group (20‐30%) The knowledge obtained from the model system was applied for the synthesis of starch‐g‐poly‐ε‐caprolactone using hydrophobised silylated starch. The grafting efficiency for the desired reaction was 28‐58%. and off‐white solid products with isolated yields of 30‐96% were obtained. are described in Chapter 2. with higher temperatures (100°C) leading to lower yields. The results for the model system involving the ROP of p‐dioxanone initiated by hydroxyl groups of a protected monosaccharide (1.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose) using Al(OiPr)3 as the catalyst. The use of higher carboxylic (fatty) acid to esterify starch resulted in products with significantly improved 134 .Summary initially performed research on the ROP with a simpler model system involving a simple protected mono‐saccharide instead of starch. but the products are still very brittle. and particularly acetic acid derivatives (C2) have received considerable attention.2. Silylated starch with a low to medium DS (0.3.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose ratio (at constant Al(OiPr)3 intake).46‐0. The synthetic procedure may be divided into three steps: hydrophobisation of starch by the introduction of SiMe3 groups followed by in situ ROP. The DS of the PCL chains was between 0. Considerable amounts of ε‐CL homopolymers with isopropyl end‐groups were also formed. 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.72. and subsequent removal of the silyl groups. depending on the ε‐CL to starch ratio. Esterification of starch with carboxylic acid derivatives is one of the oldest strategies to improve starch properties.3. A statistical model has been developed to quantify the effects of process variables (time. 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 yield of the polymers was a function of the reaction temperature and the reaction time.
however.4‐3 for starch stearate). and Na‐acetate) in DMSO at 110°C. including statistical modelling on the synthesis of long fatty esters of corn starch (starch‐laurate and starch‐stearate) using corresponding vinyl fatty esters is reported. The synthesis of these fatty starch esters is.26‐2.96). The DS of the products is a strong function of the basicity of the catalyst. reaction temperature. In Chapter 4‐5. and the type of catalyst. K2CO3. the synthesis of two polycaprolactone (PCL)‐based reactive interfacial agents. vinyl ester to starch ratio. For low vinyl‐ester to starch ratios.45‐1. catalyst intake. usually performed using fatty acid chlorides which are relatively expensive and rather corrosive.24 ‐ 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.1‐2.75) are still partially crystalline. elongation at break of 3‐26%. High‐DS products (DS= 2. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. 0. The products have a broad range of degree of substitution (DS = 0. 13C‐NMR and FT‐IR. Mechanical tests show that the products have tensile strength (stress at break) between 2. K2CO3. The starch esters were prepared by reacting starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. The DS of the products was a clear function of the chain length of the fatty ester.39) are totally amorphous whereas the low‐DS ones (DS= 1. and modulus of elasticity of 46‐113 MPa. At higher ratios.3‐1.9 for starch laurate. 1. Summary mechanical properties and hydrophobicities. PCL‐g‐ 135 . an alternative method for higher fatty starch esters by using fatty vinyl esters is reported.5 for starch laurate. and catalyst basicity) to the DS of the starch ester products were developed. The use of methyl and glyceryl esters results in products with only relatively low DS. Reaction temperature and catalyst intake also affect the product DS but only to a lesser extent. In Chapter 6. a systematic study.5 MPa.5 for starch stearate). while the use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2. Statistically adequate (R2≥0. presumably due to a reduction of the polarity of the reaction medium. The yellowish products were characterized by 1H‐.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐ starch mol ratio. the DS decreased.07‐1. The thermal stability of the esterified products is higher than that of native starch. and Na‐ acetate). With these catalysts. The last part of this thesis deals with reactive blending of starch and biodegradable polymers using reactive interfacial agents. The thermal and mechanical properties of some representative product samples are also described. products with a DS > 2.4 could be obtained for both laurate and stearate esters. In Chapter 5. The use of Na2HPO4 resulted in low‐medium DS products (0.7‐3. an increase in the vinyl‐ester concentration led to higher product DS values. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4.
PCL‐g‐DEM seems to have slightly more efficient compatibilising effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values. 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.Summary glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). A statistically adequate model (R2=0. At a fixed starch content (20 %‐wt). The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). The PCL‐based compatibilisers were prepared by reacting a low molecular weight PCL (Mw 3000) with glycidyl methacrylate or diethyl maleate in the presence of benzyolperoxide initiator at 130°C. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. P‐value ≤10‐9) has been developed to quantify the effects of process variables (monomer and initiator intake. As the result. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. 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). mutual solubility of the monomer into the molten polymer) on the functionalisation degree (FD) of the GMA and DEM units to the PCL backbone. 136 . at relatively high GMA and BPO intakes. Both of the PCL‐based compatibilisers improve the interfacial adhesion between the starch particles and the PCL matrix in starch/ PCL blends. the mechanical behavior of the compatibilised blends is in general different from that of pure PCL and of the corresponding uncompatibilised blends. and their use in compatibilising starch‐PCL blends is described.957.
Voorbeelden zijn thermo‐ plastificatie. Zetmeel is een polymeer bestaande uit anhydroglucose (AHG) eenheden.4)‐glucoside bindingen verbonden zijn. Modificatie van zetmeel kan het aantal toepassingen sterk vergroten.v.6)‐glucoside bindingen met de hoofdketen zijn verbonden. Zetmeel is relatief goedkoop en te winnen uit een breed scala aan planten. mede omdat het niet mogelijk is om films met goede mechanische eigenschappen te maken. gevolgd door de ROP met een polyester precursor en verwijdering van de SiMe3 groepen .Samenvatting Kunststoffen gemaakt van fossiele grondstoffen als olie en gas zijn zeer aantrekkelijke materialen voor een breed scala aan toepassingen. Dit proefschrift beschrijft een aantal methodes om nieuwe polymere materialen op basis van zetmeel te maken. De AHG eenheden kunnen op een aantal manieren aan elkaar gekoppeld worden. Het gebruik van zetmeel voor verpakking materialen is momenteel beperkt. menging met andere materialen. De synthese van dergelijke producten met een in situ katalytische ringopening polymerisatie (ROP) van geschikte monomeren als b. Er zijn verschillende methodes beschikbaar om zetmeel te modificeren. De toepassing van een alternatieve methode waarbij zetmeel eerst minder hydrofiel gemaakt wordt door de introductie van grote hydrofobe SiMe3 groepen. Amylose is een lineair polymeer waarin de AHG eenheden hoofdzakelijk via α‐D‐(1. Het zetmeel is daarnaast ook erg gevoelig voor water. 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. Na gebruik kunnen de materialen in het milieu eindigen en grote problemen veroorzaken. De verhouding amylose en amylopectine in zetmeel varieert en hangt grotendeels af van de plant waaruit het zetmeel gewonnen wordt. De productie van kunststoffen is sinds 1950 met bijna 10% per jaar gestegen. in combinatie met de huidige hoge prijzen voor de grondstoffen. Het amylopectine is een vertakt polymeer waarbij de vertakkingen via α‐D‐(1. Twee type polymeren zijn dominant aanwezig in zetmeel: amylose en amylopectine. chemische modificaties en combinaties van deze methoden. 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. Dit. 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. ε‐caprolactone op de hydroxy‐groepen van zetmeel is helaas niet eenvoudig.
De DS van de PCL ketens lag tussen de 0. De gevolgde procedure bestaat uit drie stappen: hydrofobisering van het zetmeel door de introductie van SiMe3 groepen.3.2. 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 effecten van procesvariabelen (tijd. waarin hogere temperaturen (100°C) tot lagere opbrengsten leiden.72 en was afhankelijk van de ε‐caprolacton/zetmeel verhouding. De opbrengst van de polymeren was een functie van de reactietemperatuur en de reactietijd.Samenvatting zou in theorie moeten leiden tot een hogere grafting efficiency. De resultaten worden in Hoofdstuk 3 in detail beschreven. De substitutiegraad van de SiMe3 groepen varieerde van laag tot middelhoog (degree of substitution. Om inzicht in het potentieel van deze benadering te krijgen is onderzoek verricht naar de ROP met een eenvoudig modelsysteem bestaande uit een beschermde mono‐saccharide in plaats van zetmeel. De grafting efficiency van de gewenste reactie was 28‐58%. De kennis verkregen met het modelsysteem is toegepast om ε‐caprolactone te graften op zetmeel. De polymerisaties werden tussen 60‐100°C uitgevoerd en leverde vaste producten in geïsoleerde opbrengsten tussen de 30 en 96%. De silylerings reactie werd bij 70°C uitgevoerd met hexamethyl disilazane (HMDS) in DMSO/tolueen mengsels.46‐0. DS = 0.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (bij constante i Al(O Pr)3 concentratie).3.68). temperatuur en monomeer: monosaccharide verhouding) op de gemiddelde polymerisatie graad zijn gekwantificeerd met een statistisch model. gevolgd door in situ ROP met ε‐ caprolacton en daarna de verwijdering van de silyl groepen. De polycaprolacton eenheden (PCL) van het verkregen poly‐ε‐caprolactone gesilyleerde zetmeel hadden een gemiddelde ketenlengte van 40‐55 monomeereenheden (molecuul massa 4500‐6300). Een hogere verhouding leidde tot hogere molecuul massa’s. Er werden ook aanzienlijke hoeveelheden homopolymeer van ε‐ caprolacton met isopropyl eindgroepen gevormd. 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/1.21‐0. 138 .2. De grafting van ε‐ caprolactone aan het gesilyleerde zetmeel werd in THF bij 50°C uitgevoerd met Al(OiPr)3 als katalysator. De resultaten voor het modelsysteem waar de ROP van p‐dioxanone wordt geïnitieerd door de hydroxy‐groepen van een beschermde mono‐saccharide (1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) met Al(OiPr)3 als katalysator worden in Hoofdstuk 2 beschreven. 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%).
Met deze katalysatoren worden voor zowel de laureaat als stearaat esters producten met een DS van boven de 2. Echter. 0. zelfs in aanwezigheid van plastificeermiddelen. De substitutiegraad varieert van 0. De zetmeel esters werden bereid door zetmeel met vinyl laureaat of vinyl stearate in DMSO in de aanwezigheid van basische katalysatoren (Na2HPO4.24 tot 2. Het gebruik van Na2HPO4 resulteerde in producten met een laag tot middelhoge DS (0. De zetmeel esters werden bereid door het zetmeel met vinyl laureaat of vinyl stearaat in aanwezigheid van basische katalysatoren (Na2HPO4. In Hoofdstuk 4‐5. K2CO3 en Na‐acetaat vertonen superieure activiteiten vergeleken met Na2HPO4. Echter de producten zijn nog steeds zeer bros.07‐1. en in het bijzonder azijnzuur derivaten (C2). Het gebruik van hogere (vet‐) zuren resulteerde in producten met beduidend betere mechanische eigenschappen en hydrofobiciteit. reactietemperatuur. naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat.5 voor zetmeel laureaat. inclusief statistische modellering. De DS van de producten was een duidelijke functie van de keten lengte van de vetzure vinylester. terwijl het gebruik van de K2CO3 en Na‐acetaat katalysatoren 139 . De producten werden met 1H‐. Bij lage vinyl‐ester/zetmeel verhoudingen leidde een verhoging van de vinyl‐ester concentratie in eerste instantie tot hogere DS waarden. De synthese van deze esters wordt in het algemeen uitgevoerd met vetzuurchloriden die vrij duur en sterk corrosief zijn. K2CO3 en Na‐acetaat) in DMSO bij 110°C te laten reageren. De thermische en mechanische eigenschappen van sommige representatieve producten zijn uitgebreid bepaald. De DS van de producten is een sterke functie van de basiciteit van de katalysator. katalysator concentratie en katalysator basiciteit) op de DS zijn gekwantificeerd met een statistisch model (R2 ≥ 0. wordt een systematische studie.96 en P‐ waarde van ≤ 10‐7). vermoedelijk als gevolg van een verlaging van de polariteit van het reactiemengsel. De effecten van belangrijke procesvariabelen (vinyl ester/zetmeel molverhouding. Het gebruik van methyl en glyceryl esters resulteert in producten met een lage substitutiegraad. De acetaat esters zijn in het algemeen minder hydrofiel dan zetmeel. en Na‐acetaat) te laten reageren. In de meeste zetmeel esterificatie studies wordt gebruikt gemaakt van kleine carbonzuren (C1‐C4). wordt een alternatieve methode voor de synthese van zetmeel esters met lange vetzuren beschreven.96. In Hoofdstuk 5.3‐1.5 voor zetmeel stearaat). De temperatuur en de katalysator concentratie beïnvloeden de DS ook maar in mindere mate. 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.4 verkregen. K2CO3. de vinyl‐ ester/zetmeel verhouding en het type katalysator. 13C‐NMR en FT‐IR gekarakteriseerd.
Mechanische tests tonen aan dat de producten een treksterkte hebben variërend van 2. De synthese van deze verbindingen wordt uitgebreid beschreven. De hoogste FD waarden (tot 45%) werden gevonden voor PCL‐g‐GMA bij relatief hoge GMA en BPO concentraties. Als reactieve compatibilisers zijn twee polycaprolactone derivaten. een uitrekking tot breuk van 3‐26%.0 voor zetmeel stearaat). Er is een statistisch model (R2 =0.Samenvatting resulteerde in producten met een middelhoog tot hoge DS waarden (2.7‐3.26‐2. Dit leidt tot verschillen in mechanische eigenschappen tussen gecompatibiliseerde mengsels en van de niet‐ gecompatibiliseerde mengsels van zetmeel en PCL. 140 . De FD waarden voor PCL‐g‐DEM waren aanzienlijk lager (tot 7%).1‐2. 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. Beide reactieve compatibilisers verbeteren de interactie tussen de zetmeeldeeltjes en de PCL matrix.45‐1.957 en P‐ waarde van ≤ 10‐9) ontwikkeld waarmee de effecten van procesvariabelen (monomeer en initiator concentratie. PCL‐g‐glycidyl methacrylaat (PCL‐GMA) en PCL‐g‐diethyl maleaat (PCL‐g‐DEM).4‐3. gebruikt. Zo is bijvoorbeeld de elastische modulus van de gecompatibiliseerde mengsels beduidend hoger (tot 430 MPa) dan die van niet‐gecompatibiliseerde mengsels (320‐340 MPa).5 MPa. De thermische stabiliteit van de geësterificeerde producten is hoger dan die van zetmeel. Het laatste hoofdstuk van dit proefschrift (Hoofdstuk 6) behandelt de menging van zetmeel en biologisch afbreekbare polymeren met reactieve compatibilisers.9 voor zetmeel laureaat. 1. en een elasticiteit modulus van 46‐113 MPa.75 nog gedeeltelijk kristallijn zijn. wederzijdse oplosbaarheid van het monomeer in het gesmolten polymeer) op de functionalisatie graad (FD) van de GMA en DEM eenheden aan de PCL keten adequaat gekwantificeerd kunnen worden. De producten met een hoge DS (2.39) zijn volledig amorf terwijl de producten met een DS van 1.
and Prof. Anno Wagenaar and Wim Kruizinga taught me how to operate the NMR machines and to interpret the spectra. since the mechanical properties as well as hydrophobicity of this starting material is relatively poor. I want to thank Erik Heeres for being more than a good supervisor. L. Prof. which improved the quality of this thesis. Anne Appeldoorn. Minnaard. Thank you very much for spending your valuable time reading my thesis and for the constructive comments and corrections you have made. He has also been a good friend and always supported me during the hard times. Erik Heeres. First of all. Gert Alberda van Ekenstein and Harry Nijland helped me a lot with mechanical and thermal property measurements. Francesco Picchioni. Broekhuis. and for always encouraging me to finish this thesis. as well as taking the SEM pictures. for valuable discussions during my early PhD period and for handing me over to the right hands. especially on his opinions about the statistical modelling and mechanical properties. Prof. Anno was also always there when . Thank you Erik. I want to thank my three promoters. 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 also want to thank Ineke Ganzeveld. Erwin Wilbers. for your consistent support. Leon Janssen. for the inspiring ideas during the discussions. I want to thank Leon Janssen for his hospitality. not in the last place because of the problems related with conventional plastics. Prof. I want to thank Francesco Picchioni for being always there when I needed him to discuss anything.A. This thesis is a summary of my research on developing novel biodegradable polymers from starch. A. These include the current high oil prices. for upgrading my writing quality.Acknowledgements The synthesis of biodegradable materials is currently receiving a lot of attention. Although it is indeed a very interesting topic. and Prof. People from other groups have also been very kind and supportive to my research. I want to thank Marya van der Duin‐de Jonge. Mosiscki. for her hospitality and solving administrative matters.J. and for helping me arranging many things when I was not able to do it from Indonesia. the raw material for fossil derived plastics. Marcel de Vries. for his efforts to make every foreign students feel at home in Groningen. who was my first daily supervisor for about two months. Prof. and Laurens Bosgra for being always helpful with technical issues and Jan Henk Marsman for the help with analytical equipment. together with difficulties in managing plastic waste. A. I found out that developing a better fully‐biodegradable‐material from starch is not an easy job. I also want to give my deep appreciation to my reading committee.
mbak Mia. and Mahesa for being my brothers and sisters in Groningen. mas Pandu. friendly. Vincent Nieborg and Fesia Lestari. Ais Jenie. and Tante Alma for being like my parents and family in Groningen. Boelo Schuur. for your never‐ending blessings. The chemical and glassware magazijn as well as the FWN library personnel were always patient.B. You all have always been my inspiration and motivation. Chalid. Wuri Raspati. and Henk van de Bovenkamp for cheering my days and being my friends during summer and winter. Egi. Thank you AW. and assistant professors have shared the days with me in Groningen. Louis Daniel. Hofman. Tante Oppie. Nidal Hammoud Hassan. Groningen is also a place where I found true friendship. Bima and Wisnu (also my paranymph). Wangsa. and Iwan Kustiawan for becoming my family. 142 .Acknowledgements I had questions or needed specific equipments. Wahyudin. Patrick and Oppie. Teguh. Sari Fabianti. Francesca Gambardella. for everything. Taufik. Taufik. Ignacio Melian Cabrera. Ronny Sutarto. C. Rasrendra. My special thanks are for Laura Junistia and Jan‐Pieter Drijfhout. Agnes R. Wita Sondari. and the ISD staff helped me a lot with visum and residence permit arrangements as well as tax papers. for your love. Tante Smith. kak Roga. and passion. Thank you Tresna. Jaap Bosma. visiting researchers. Thank you my parents and sister. patience. mbak Ika. Many master and PhD students. Judy Retti Witono. Tante Caroline. Togi. Ita. Anneke Toxopeus. for never leaving me alone. Thank you God. Kuslan. Buana Girisuta and Rina Karina. Thyara. Endy. Erna Subroto. I have met a lot of Indonesian students in Groningen. Farchad H. Jelle Wildschut. Ardiyanti. kak Atha. Henky Muljana (my paranymph) and Anindhita Widyadhana. in the office. and for my wife and family. Diana Santangelo. Asal Hamarneh. and on many occassions. Mahfud. Puspita Kencana. Iwan Harsono. R. spring and autumn in Groningen. Zhang Youchun. Hans Heeres. Our collaboration has also become part of this thesis. Teddy. and many others for the good old times! Thank you Tante Ietje. August Kurniawan. our memories will never fade away. Gerard Kraai. Tiara. and helpful. Niels van Vegten. master students who worked with me. support. Anna Nizniowska. bu Ida Susanti. Thanks also for Marcel and Connie. Albert Kieviet helped me with the MALDI‐TOF measurements. Jasper Huijsman. Samuel. Arjan Kloekhorst. Dado. Yao Jie. Mita. pak Harry. Marcel Wiegman. Francesca Fallani. Annette Korringa. Abdul Osman. in the lab. Diana Jirjis. Nadia Gozali. Yongki. Oscar Rojas. I want to thank Edwin A. Tenny. M. and although we seldom met afterwards. A lot of thanks for Henk Stegeman and Poppy Sutanto.
P.P.). 3.Drijfhout.P.eurpolymj. F. Janssen. A. F. Picchioni. Bratislava (Slovak Republic). L.J. F. Guangzhou (China). presented in “Polymers for Advanced Technologies”. Kuala Lumpur (Malaysia).List of Publications 1. 2008 (accepted for publication in Starch‐ Starke). A. J. and F. Manurung. Pune Chapter. R. 2008 (accepted for publication in Starch‐Starke). Synthesis of poly‐ (ε)‐caprolactone grafted starch co‐polymers by ring‐opening polymerisation using silylated starch precursors.Janssen. 5. 8. Manurung. F.M. Polym.Heeres. starch‐based innovative product: The synthesis of grafted polycaprolactone‐starch by ring opening polymerisation. Sugih. A. Heeres. and H. Sugih. Heeres: Synthesis and properties of novel green biopolymers derived from starch and fatty acids.1016/j. Heeres.J.K. India. 2. presented in the 1st International Symposium on Sustainable Chemical Product and Process Engineering.) 6. F. A.J. Drijfhout. L. L.J. 2008 (submitted to J.B. DOI: 10. L.P. Sci.K. (in press. A. 9. Sugih. Broekhuis.P.Picchioni. 26‐28 November 2006 (oral presentation). Junistia.B. 9th National Conference organized by the Society for Polymer Science.K. 17‐20 December 2006 (oral presentation).M.010). Junistia. 25‐28 September 2007 (oral presentation). Sugih. Picchioni.M Janssen and H. Heeres.A. 27‐30 August 2006 (oral presentation) 7.P. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Junistia. A. A. Janssen. . Heeres.J. R. Picchioni. Drijfhout. A. L. Picchioni. Polym. H. Sugih.M.K. A. Sugih. J. Picchioni: Biodegradable polymer materials based on polyesters and starch/proteins. J. Picchioni. A.P.K. Sugih. Sugih. Eur.J. 4.B.K.K. presented in 11th Asia Pacific Confederation of Chemical Engineering (APCChE) Congress. Polym. H. Picchioni.K. 2008 (submitted to Carbohydr.B.J. and F. and H. presented in International Workshop: “From Polymer Modification to Multicomponent Polymer Systems”.A. Appl. J. Sugih. Synthesis and properties of reactive interfacial agents for polycaprolactone‐starch blends. L. A. Experimental Studies on the ring opening polymerization of p‐dioxanone using an Al(OiPr)3‐monosaccharide initiator system. L.10. H. H. Picchioni: Functionalization and blending of polycaprolactone. H. Mumbai (India).2008. H. Broekhuis. Heeres: Green.J. Junistia.J. South China University of Technology. F. F. L. 2008. Heeres. Experimental and modeling studies on the synthesis and properties of higher fatty esters of corn starch.K.
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 listening from where you left off, or restart the preview.