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
A. Minnaard Prof. L. dr. dr. ir. dr.A. ir. dr.M. Heeres Prof. dr.J. Picchioni Beoordelingscommissie : Prof.B. Janssen Prof. Broekhuis Prof.J. A.P. Promotores : Prof. dr. F. H. Moscicki ISBN : 978‐90‐367‐3592‐6 ISBN : 978‐90‐367‐3593‐3 (electronic version) . L.
to: Tresna. and my sister . my parents.
Starch 1.2.2. References 2 3 4 5 5 7 9 9 10 11 11 13 14 16 16 17 Chapter 2: Experimental Studies on the Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐ Monosaccharide Initiator System 2.1. Materials 2.1. Biomaterials from starch 1. Starch‐based blends by in situ polymerization 1.1. Methods 2.2. Starch production processes 1.4.2. Product analyses .22.214.171.124.1.1.2. Thesis Outline 1. Starch Esters 1. Thermoplasticized starch 1. Plastic applications and waste issues 1.3.2. Cross‐linked starch 1.1. Starch‐based blends by melt mixing 1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (2) 27 28 26 27 27 27 2. Structure and properties of starch 1. Introduction 126.96.36.199.1. Biodegradable plastics from starch 1.2. The potential of biodegradable plastics 1.3.3. Typical example for the synthesis of polydioxanone end‐capped with 188.8.131.52.2.2.Table of Contents Chapter 1: Introduction 184.108.40.206. Starch ‐ biopolymer blends and graft co‐polymers 220.127.116.11.4. Materials and Methods 18.104.22.168.2. Starch modifications to improve product properties 1.2.
2.4.2. Materials 3.1. Mechanistic aspects 22.214.171.124.126.96.36.199. Analytical methods 3.4.3. Synthesis of silylated starch viii . Methods 3.2.4. Results and Discussions 2.2. Peracetylation of silylated starch 3.2. Screening experiments 2.2. Materials and Methods 3.3.3. Calculation of average degree of polymerization 188.8.131.52. Product yield 2.1.5. Nomenclature 2. Introduction 3. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers 51 51 52 52 52 53 54 48 49 49 50 50 184.108.40.206. Nuclear Magnetic Resonance (NMR) 220.127.116.11. Product analyses 18.104.22.168.22.214.171.124. Typical example of the starch silylation procedure 3. Results and Discussions 3. 2.3. Systematic studies 2. Calculations 3. Conclusions 126.96.36.199.3.1.2. References Chapter 3: Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors 3.6.2. 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 44 188.8.131.52. Typical example of in situ polymerization of ε‐CL with silylated starch 50 3.
2. 1H‐ and 13C‐NMR analyses 184.108.40.206. Exploratory experiments 220.127.116.11.18.104.22.168.3.2. Deprotection of silylated‐starch‐g‐PCL 3. Analytical equipment 22.214.171.124.2.3.2. Conclusions 3. Materials 126.96.36.199. Product characterisation 4. Nomenclature 3.3. Peracetylation procedure 4.3.2. 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. References Chapter 4: Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants 4.2. Methods 4. Conclusions 4. Effect of vinyl ester to AHG ratio on the product DS 4.1. Introduction 188.8.131.52.2.2. Results and Discussion 4.1. Determination of the Degree of Substitution (DS) 184.108.40.206.4. Systematic studies 4.2. References ix .5. Nomenclature 4.3. FT‐IR measurements 4.3. 3. Materials and Methods 4.2.3. Catalysts screening 4.2. Effect of the addition of toluene as a co‐solvent 220.127.116.11. In situ ring opening polymerization of ε‐caprolactone with silylated starch 57 63 64 64 65 3.6.3.
18.104.22.168.2. Typical example of the preparation of laurate and stearate starch Esters 88 86 87 87 87 88 5. Materials and Methods 5.2. Introduction 6.1. Nomenclature 5. Product properties 5.2.4. Analytical Equipment 5.1.2. Introduction 5. Experimental Design 5. Preparation of PCL‐starch blends with the reactive compatibilisers 110 22.214.171.124. Work‐up of PCL‐g‐DEM 108 109 109 109 109 110 110 6.2.1. Materials 126.96.36.199. Materials 188.8.131.52.2.2.2. Results and Discussion 5. References Chapter 6: Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 184.108.40.206. Materials and Methods 6.1. Compatibilizer synthesis 6.4. Mathematical modelling 5.1.2. Peracetylation procedure and Degree of Substitution (DS) determination 88 88 89 92 96 102 102 103 5. Methods 6. Analytical Methods 110 x .220.127.116.11.2.2.1. Work‐up of PCL‐g‐GMA products 18.104.22.168. Conclusions 5.4.Chapter 5: Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch 5. Methods 5.2.
4. Nomenclature 22.214.171.124. Effect of substrate (GMA/DEM) to PCL Ratio on the FD 6.3.4. Results and Discussions 6.5. References Summary Samenvatting (Dutch Summary) Acknowledgements List of Publications 133 137 141 143 xi .2. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers 6.3. Conclusions 126.96.36.199.2.3. 6.1.1. Synthesis and Properties of Starch‐ PCL Blends 188.8.131.52.3.1.3. Preparation of the ractive compatibilizers 6.3. Ternary blends compatibilized with PCL‐g‐DEM 6. Thermal properties of the compatibilizers 6.2.1. Modeling of the combined effects of the GMA/ DEM and BPO intakes on the FD 111 111 111 116 117 118 120 121 122 123 125 128 128 129 184.108.40.206.2. Effect of the BPO intake on the product FD 6. Statistical mdeling 6.2.3. Ternary blends compatibilized with PCL‐g‐GMA 6. Binary blends of starch and PCL 220.127.116.11.
an outline of this thesis is given. 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. Different routes to modify starch to improve the product properties and to extend the application range will be provided. .
1. 80 70 60 50 40 30 20 10 0 Starch Production (Million Tons) 1995 2000 Year 2005 2010 Figure 1. 4‐5].1. potato. wheat. maize. The current price of corn starch is around $0. The second and third starch producer regions are Europe and Asia [4‐5]. . Plants synthesise and store starch in their structure as an energy reserve. and wheat starch are also produced in large amounts [2.e. while corn seeds consist of 65‐80% starch by weight. . Most of the starch is produced in the USA. current and future estimations of worldwide starch production are given in Figure 1.45 (estimated from the total volume and value of the US corn starch export ).e. It is generally deposited in the form of small granules or cells with diameters between 1‐100 µm . Starch is found in seeds (i. Most of the starch produced worldwide is derived from corn. The worldwide production of starch in 2008 is estimated to be around 66 million tons . sweet potato. corn. An important polysaccharide is starch. Starch 75% of all organic material on earth is present in the form of polysaccharides. Most starch crops are very productive. rice. but other types of starch such as cassava. sorghum.9 ton starch per hectare . present and a forecast of starch production  o : Europe □ : USA : Total ∆ : the Rest of the World 2 .1. potato or cassava) of the plants [2‐3]. with an average yield of 4.Chapter 1 1. Potato accumulates starch to approximately 75 % of the dry weight in the tubers with a yield up to 21 ton starch per hectare. Past. Past. barley. or peas) and in tubers or roots (i.
Figure 1. As an example. The fiber is removed using washing screens. a typical corn starch production process is given in Figure 1. the mixture is fed to a hydrocyclone to separate the germ and the rest of the kernel. Corn starch production  a. and is embedded in a proteinaceous cellular matrix as is shown in Figure 1. gluten. and other foreign materials. Typical corn‐milling operation b. a. and starch granules are obtained after centrifugation from the suspension. The resulting suspension from the mills contains fiber. After a coarse milling. The starch is present in the endosperms (floury and horny).b.2.2. the debris is filtered from the slurry. Starch production processes Starch is generally extracted from the plant by wet milling processes [9‐10].a. the corn kernel is softened by steeping in warm water containing SO2 until the volume of the kernel increases with 55‐65%. The plant material is grounded in water. After initial cleaning to remove cob. Cross‐sectional view of a corn kernel 3 . sand. which is again fed to a second milling process. Introduction 1. and starch. b.1.1. The low density gluten is separated from the starch suspension by centrifugation.2. The resulting starch is further washed in a cyclone and finally dried.
6‐7 x 105 Da . Amylose c. H OH H O H OH H O H H OH H c. 6]. Figure 1. Amylopectin is a branched polymer. Anhydroglucose (AHG) unit b.. The content of amylose and amylopectine in starch varies and largely depends on the starch source..2.3. The starch granules will swell rapidly to many times of its original volume. H2COH H OH H O H H O b.. The amylose content of several common starches is given in Table 1. The linear 4 . The molecular weight of amylose is a function of the plant source and processing method. containing periodic branches linked with the backbones through α‐D‐(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 . H2COH H .. Amylopectin Starch is insoluble in cold water. Heating a starch solution leads to loss of hydrogen bonding in the interior of the starch granule and the starch will start to gelatinize. but it is very hygroscopic and binds water reversibly..4)‐glucosidic bonds. Two types of AHG polymers are usually present in starch: amylose and amylopectin [2‐3. H .3. a. Each branch contains about 20‐30 anhydroglucose units.1. .. The molecular weight of amylopectin is higher than that of amylose and is typically 4‐5 x 108 Da .. the amylose content is between 18‐28% . Chemical Structure of Starch a. H OH H2COH H2COH H .1...6)‐ glucosidic bonds .. but usually in the range of 1. H OH H H2COH H2COH H O H H OH H H2COH H O H H OH H O H OH H O . Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1.a..) . . Structure and properties of starch Starch is a polymer consisting of anhydroglucose (AHG) units (see Figure 1....Chapter 1 1. Typically.
5‐70 62‐72 63‐72 58‐64 1.8 16. and celluloid was 5 . Table 1. swollen granules. depending on the amount of water present. and. 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 . Plastic Applications and Waste Issues Plastic is the general term for a wide range of synthetic or semisynthetic polymerisation products. The gelatinization temperature range of various starch sources is given in Table 1.8 18.104.22.168. 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. Parker in 1838. and granule fragments. will form a thick paste or gel.2.7 26 Table 1. Introduction amylose molecules leach out of the granules into the solution.5 25. The resulting suspension contains a mixture of linear amylose molecules.1. Starch gelatinization temperature range  Starch Potato Tapioca Corn Waxy corn Wheat Gelatinization Temperature Range [o C] 59‐68 58. Plastics are used in a wide range of applications and the demand is still increasing every year . Cellulose nitrate was first prepared by A.5 28 52 70‐75 27 15.7 20 18. Biomaterials from starch 1.1. The first generation of commercial plastics was derived from cellulose nitrate and is known as celluloid [12‐14].
Adhesives.) 250 Included: Thermoplastics.). for instance to be used in packaging and paper coatings . Plastics are very attractive materials.Chapter 1 patented by J. Sealants. 15]. PE is a very versatile plastic because it can be shaped easily into various forms. the oldest purely synthetic plastic is Bakelite. polypropylene (PP). Worldwide Plastics Production (1950‐2006)  6 . Important polymers used for packaging are polyethylene (HDPE and LDPE). and consumer products (see Figure 1. space. A dramatic increase in demand for plastics began after World War II.5. The production and consumption of plastics has increased significantly with a rate of almost 10% every year since 1950. They have a low density and can be shaped in thin films while maintaining good properties. Coatings. Plastics have lower melting temperatures compared to glass and metals. and PP-Fibers 2006: 245 2002: 200 200 Plastics Production (Million Tons) Not Included: PET-. Elastomers. and PolyAcryl. electrical. and polycarbonates (PC). polyvinyl chloride (PVC). polyethylene terephtalate (PET).4.Fibers 150 1989: 100 100 1976: 50 50 1950: 1. The latter is important when using the material for packaging purposes to save weight. and 37% of the total plastics demand in EU  is used as packaging materials. Hyatt in 1870 .5 0 1950 1960 1970 1980 Year 1990 2000 2010 Figure 1. About 29% of the total plastics produced in the USA . Thermosets. the worldwide plastics production has reached 245 million ton per annum  (see Figure 1. discovered by Baekeland in 1907 [12. and energy during transportation of goods. The largest application of plastics is for packaging purposes.4. While celluloid is derived from a natural polymer (cellulose). polystyrene (PS). and therefore need less energy to shape it into useful materials . when polyethylene (PE) was invented. Polyurethanes. Plastics are also used for building materials and automotive. In 2006. PA-.
Most of the waste is still disposed (50%) by landfill or incinerated to recover the energy (30%) . consisting of 25%‐v of the total waste . 16]. It is not applicable for thermoset resins . are not easy to recycle . and accumulation of its residues in soil cause significant reductions in agricultural yields.2. In Europe (2006).2. Plastics have been polluting sea [1. The volume fraction of plastic in MSW is much larger due to the low density of plastics. The disposal of plastics materials in municipal solid waste (MSW) is a serious issue in many parts in the world. and is only effective for single plastic sources or simple plastic formulations. and may be more than twice the weight fraction . only 20% of the plastic is recycled.7% in 2006 . Introduction Plastic waste. Plastic applications and waste treatment in Europe  1. Co‐mingled plastics. soil. rivers. is causing serious environmental problems. and lakes. Recycling (part of) the plastics to reduce the amount of MSW also has limitations. The potential of biodegradable plastics The application of biodegradable plastics could be an attractive solution for the problems related to the use of conventional plastics (vide infra). ship navigation. Plastic litter is hazardous to wildlife [1. hydropower plants operation. however. Figure 1. and other public works . the amount of plastics in MSW increased from less than 1% in the 1960 to 11.5. threatening fishery. irrigation. plastics waste was ranked as the second major source of MSW after paper and paperboard. 6]. In the USA. Biodegradable plastics are polymeric materials capable of decomposing when given an 7 . It has a high volume to weight ratio and is resistant to degradation. In 1996. which are usually found in waste streams.
96 proposal ‐ Degradable plastics are plastic materials that undergo bond scission in the backbone of a polymer through chemical. Nowadays new types of biodegradable plastics with improved properties and lower costs have been developed . the price of PLA based plastics in 2010 is estimated to be around € 1. 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 .2 working group on biodegradable polymers ‐ Biodegradation of a plastic material is a process leading to naturally occurring metabolic end products. The price of these biodegradable plastics is expected to be reduced considerably in the next decade . Biopolymers are inherently biodegradable. Biodegradable plastics have gained considerable interest since the 1980s. Several authorities have provided definitions for biodegradable plastics . ASTM sub‐committee D20. biological and/or physical forces in the environment at a rate which leads to fragmentation or disintegration of the plastics. because they take part in nature’s cycle of renewal. It may also be treated in sewage sludge water treatment plants or buried in the soil [20‐21]. 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. DIN 103. For instance.Chapter 1 appropriate environment and sufficient amount of time . together with food and yard waste as well as paper. and microorganisms) and biodegradable synthetic polymers . Biodegradable plastic waste may be treated in composting facilities. and since 2000 they have become competitive with traditional materials in some applications [20‐21]. The market price of biodegradable plastics (mainly from starch and PLA) has decreased the last years. and polyesters from micro‐organisms . The change in chemical structure results from the action of naturally occurring micro‐ organisms. proteins (present abundantly in plants and animals). Common natural biopolymers are carbohydrates. biopolymers from nature (from plants. animals. Biodegradable plastics may be classified into two general groups. 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.5/ 8 .
before they can be applied as biodegradable plastics. half of its price in 2003 (around 3 Euro/kg). Bastioli [20‐21] showed that nearly all biodegradable plastics available in the market are derived from starch. Consequently. chemical modification or combinations of them . 20].2. 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 . Starch may become an attractive raw material for plastics in the future. agricultural products (mulch films.4 per kg in 2007‐2008 ). breathable fabrics. napkins) . . pots). 1.3. either by plasticization. biofillers for tires. utensils .50/ kg in 2005) is also expected to be lower in the future due to considerable reduction in modification costs .50‐4. The price of starch in 2007 was about $ 0.8‐ 2. 1. because the price of oil based polymers may still increase due to the rise in the crude oil prices . either from starch‐based materials (slightly modified starch. € 1.2‐ 1. blending with other materials. starch must be modified. hygienic materials (diapers.45 per kg . This price is much lower than conventional plastics derived from oil. alone or complexed with natural or synthetic biodegradable polymers) or from polylactic acid which originates from the fermentation of a starch feedstock. The price of starch‐based biodegradable plastics (€ 1. tensile and flexural strength) and is too sensitive to water [26‐27].3. such as polyethylene (PE. Virgin starch is not suitable as a packaging material. towels. Examples are the use as packaging material for carrier bags. Biodegradable Plastics from Starch Starch is a very attractive source for the development of biodegradable plastics. These are summarized briefly in the next sections. 9 . Introduction kg. consumer products and food products. or when environmental impacts have to be minimized. and chewable items for pets [1.4 or $ 1. Biodegradable plastics are especially very useful for single‐use applications. It cannot be shaped in films with adequate mechanical properties (high percentage elongation. Biodegradable plastics are also used for food servicewares . when recycling is not practical or uneconomical [20‐21].
The use of other plasticizers (for example glycerol) results in a rubbery material. the thermal properties of glycerol‐plasticized starch are a function of water content [31. such as water and polyols (glycerol. . There are several substances used as plasticizer for the preparation of thermoplastic starch (TPS). which is even above the thermal degradation temperature. approximately 230 oC) . glycerol. which results in embrittlement. glycol. plasticized. and blow moulding [28.Chapter 1 1. This problem is mainly caused by the presence of strong inter‐ and intra‐molecular hydrogen bonds between the starch macromolecules [28. Thermoplasticized starch Virgin starch is brittle and difficult to be processed into articles due to its relatively high glass transition temperature (Tg. 40]. and hexylene glycol. 30. et al  discovered that the elongation of break of the thermoplastic starch is significantly improved by plasticization with glycol.1. sorbitol. Starch can be modified to obtain materials which melt below the decomposition temperature . 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 brittleness is known to increase in time due to free volume relaxation and retrogradation. At intermediate water levels. 30]. 35. and the properties and amount of plasticizers. Although thermoplasticization seems to be a promising method. or melted starch [29. The plasticized starch properties may be tuned by changing the temperature of processing. The thermoplasticization process will decrease the interactions of the molecular chain and destruct the structure of the starch . 32‐ 36].3. 40]. with better properties than virgin starch in various applications. and therefore are processable by conventional polymer processing techniques such as injection. which still limits their use. 33]. which will result in a continuous phase in the form of a viscous melt [29‐30. For instance. TPS synthesized from polyol and sugar plasticizers have the tendency to re‐crystallize (retrogradation) after being stored for a period of time. water content. 39]. 32. The modification involves break down of the starch granular structure by the use of plasticizers at high temperatures (90‐180 oC) and shear. because the resulting product will be brittle when equilibrated with ambient humidity . The plasticizers are also usually hydrophilic and can be 10 . extrusion. sugars) [30. As the result. Yu. 37]. Another issue is the poor water resistance and low strength. The use of water as a plasticizer is not preferable. the semicrystalline structure of starch and its granular form are lost and the starch polymers are partially depolymerized. phase separation may still occur. destructed. resulted in the formation of an amorphous mass [35. The modified products are known as thermoplastic.
especially in the food. Chemicals of these classes having two or more of the reactive groups may react with two or more hydroxyls of the starch molecules. High DS starch acetate can be easily casted 11 . Three different types of starch acetates may be distinguished. The DS is defined as the moles of substituents per mole of AHG units . The simplest starch ester is starch formate (C1). The products are called cross‐linked starches. synthesized by direct addition of formic acid to starch at room temperature. differing in Degree of Substitution (DS). or coating with hydrophobic polymers [33. et al  studied the mechanical properties of films from maize starch cross‐linked by sodium trimetaphosphate (SMTP). These hydroxyls can react easily with a wide range of compounds such as acid anhydrides. paper. Already in 1865.3. Most studies dealt with the synthesis of starch esters of C1‐C4 carboxylic acids. Breakdown of the starch and the formation of low molecular weight products occur to a significant extent . Medium DS starch acetates (0. Solutions to improve the properties of TPS are blending.3‐1) are still soluble in water. leading to higher viscosities [42‐43]. Silva. Low DS products (0. aldehydes. Higher processing temperatures generally led to higher cross‐linking levels. Introduction washed out by water . 46].3.2. and ethylenic compounds. 39]. resulting in an increase of the Young’s modulus and tensile strength of the products and a decrease in elongation at break. textile. Substitution of the hydroxyl groups of starch with acetate groups makes the esters more hydrophobic than native starch.01‐0. Starch Esters The development of starch esters started in the mid 19th century [44. and adhesive industry. and particularly with acetic acid . organic chloro‐compounds.3. The cross‐linked starches have found many applications.2) are commercially available and used as food additives and in the textile industry. the use of formyl esters of starch is much lower than that of acetate esters. epoxy. while highly substituted starch acetate (DS of 2‐3) are soluble in organic solvents. As a result. 1. Schuetzenberger acetylated starch with acetic anhydride . 1. The most popular starch ester is starch acetate . The esterification reaction is catalyzed by H+ and proceeds with the formation of water . Cross‐linking results in a reduction of the solubility in water and to thickening. Cross‐linked starch An anhydroglucose molecule of starch contains two secondary and one primary hydroxyl group. The cross‐linked products are therefore more rigid (and less elastic) materials than virgin starch .
allowing the polymer to be produced with a range of hydrophobicities . acetate. hydrogencarbonate. and is shown in Scheme 1. Neutral and weak acid/ alkaline catalysts result in regioselective substitution at the C2 hydroxyl groups of starch. Esterification may be performed using acid anhydrides in aqueous media or organic solvents (pyridine. et al [55‐56]. DMSO. while alkaline catalysts (such as carbonate.1. Acetylation of starch in water and DMSO using vinyl acetate has been studied lately by Mormann and Al‐Higari  as well as Dicke . and phosphate) will result in C‐2. or isopropanol) with acidic (hydrochloric or sulfuric acid) or alkaline (NaOH or triethylamine) catalysts . but also C‐6 and C‐ 3 substitution .Chapter 1 into films using organic solvents. Starch triacetate has been successfully synthesized using acetic anhydride in combination with pyridine‐gelatinized starch .1. The kinetics of the reaction between gelatinized aqueous potato starch and vinyl acetate was studied by De Graaf. Several synthetic routes have been developed for starch esters. Another attractive route involves the use of vinyl esters as reagents . 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. The degree of acetylation of starch acetates can be easily controlled. Starch esters have also been synthesized using alkanoyl chlorides [49‐53]. DMF. xylene. Mechanism of starch acetylation using vinyl acetate 12 . High DS starch acetates are thermoplastic materials suitable to be used as biodegradable plastics .
melt‐mixing remains the preferred route to polymer blends mainly because of very practical reasons: low costs. Starch ‐ Biopolymer Blends and Graft Co‐Polymers Blending of different polymers is an established method to obtain products with improved properties. The introduction of longer acid chain is expected to reduce the brittleness of virgin starch and to increase its hydrophobicity . 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 fatty acid chloride reactants are. which in turn provide a very strong adhesion at the molecular level between the dispersed phase and the matrix. i. Introduction The synthesis of long chain fatty acid ester of starch has attracted much interest lately. interfacial agents (e.61) products could be obtained using this approach. polymers are rarely miscible with each other  so that. The classical example of such process is represented by the production of High Impact Poystyrene (HIPS) obtained by styrene polymerization in the presence of polybutadiene . The first involves simple melt mixing of the two components for example by extrusion. Fatty‐acid starch esters have been synthesized using fatty acid (octanoyl. Although the in situ polymerization process is not as technologically straightforward and economically convenient as melt‐mixing. However. availability of mixing equipment and no necessity to use any organic solvent (often employed for the in situ polymerization). they locate themselves at the interface between the 13 . in the simplest case of a binary blend. one component will be dispersed into the other. However. compatibilizers) might be used. If the right combination of chemical groups on the two components is present along the polymers backbone.e.34‐0.4. As a result. but only relatively low‐DS (0. octadecanoyl) chlorides [49‐53]. 1. it is frequently used in order to chemically graft the polymerized component on the other one (polystyrene on polybutadiene in our example above ). the latter are mixed together. Despite this advantage. a chemical reaction might take place upon processing (reactive extrusion). in order to achieve also a good molecular adhesion between the phases by melt mixing.g. Two main methodologies are applied for the production of polymer blends. the two components are chemically linked to each other.3. however. dodecanoyl. 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 . relatively expensive and rather corrosive. 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 . Their role is comparable to the one of a surfactant in emulsion formation . By working at temperatures above the melting point or glass transition temperature of the two components.
3.). 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. The products have been synthesized in the lab as well as on industrial scale [32.6.4. The starch/polyethylene blends are used for agricultural mulch [27. 39. there have been efforts to blend as well as to graft synthetic polymers onto starch. The most often used synthetic polymer for blending with starch is polyethylene [27. Starch‐based blends by melt mixing Starch‐synthetic polymer blending has been studied as early as in 1973 [64‐65]. The synthesis of compatibilizer and its use for starch/ synthetic polymer blending has also been studied. 66‐81]. Illustration of the role of interfacial agent in compatibilizing blends 1. A suitable interfacial agent for the blends of two polymeric materials is a block copolymer for which the chemical structure of every block is the same (or very similar) to the one of the individual components to be blended  (illustrated in Figure 1.6. 64‐98]. 80‐ 81] or food packaging . In the past. Matrix Dispersed Phase Figure 1. The uncompatibilized blends of starch and 14 .Chapter 1 two components stabilizing the dispersion (most probably by a steric repulsion mechanism ) and providing an improved adhesion at the interface.1. Interfacial agents already available on the market can be used as such or can be produced upon mixing (in situ) by chemical reaction of the two components.
The use of conventional synthetic polymer such as PE for blending with starch will only result in a partially biodegradable material.e. and poly‐lactones (such as poly‐butyrolactone. the compatibilizer. and of a premade starch‐g‐PCL  for starch/PCL blends resulted in a better dispersion between the phases and in turn in improved mechanical properties compared to the uncompatibilized blends. The ester bonds of these polymers are susceptible to attack by water and this leads to enhanced biodegradability. The tensile strength of the uncompatibilized blends is drastically reduced when starch content is increased. and poly‐caprolactone). which combines excellent biodegradability with acceptable mechanical properties. thermal and water/gas permeability properties that are even comparable to bulk non‐biodegradable polymers such as PE and PP. while PE‐g‐GMA posseses epoxide groups. As a result a graft copolymer (PE‐g‐Starch). Polycaprolactone (PCL) is a well‐known synthetic biodegradable polyester. Despite these good results. while the tensile strength of the compatibilized ones decreases only slightly with the starch content. among which synthetic polyesters are considered very promising materials . 96]. PCL‐g‐maleic anhydride (PCL‐g‐MA) [91. which both can react in situ with the hydroxyl groups of starch [70. 72‐73]. The use of reactive compatibilizer precursors PCL‐g‐glycidyl methacrylate (PCL‐g‐GMA) [89‐90]. 39. poly‐lactides. is formed upon mixing. PCL‐g‐pyromellitic anhydride [94‐95]. dextran‐g‐PCL . Studies on the blending of starch with PCL have been already described [32. As is the case of blends with PE. Introduction polyethylene show a coarse phase separation due to differences in polarity of starch (hydrophilic) and polyethylene (hydrophobic). 15 . i. To obtain completely biodegradable products. 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. thus improving the dispersion of starch in the PE matrix. poly‐dioxanone. 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. Examples of these biodegradable polyesters are poly‐glycolide. since the conventional synthetic polymers are usually poorly or non‐biodegradable. and EVOH . poly‐valerolactone. synthetic biodegradable polymers have been applied. 84‐98]. PE‐g‐MA contains reactive anhydride sites. 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. the uncompatibilized blends of starch with PCL give coarse phase separation and a reduction in the mechanical properties when the starch content is increased. These biodegradable polyesters will finally decompose into non‐toxic products . Some of these polyesters also have very good mechanical.
during the reaction.3. The highest GE value (up to 90%) has been achieved when using triethylaluminium as catalyst [97‐98]. which however is extremely air‐ and water‐sensitive.2. 1. defined as the percentage of grafted polyester to starch compared to the total amount of homopolymer and grafted polyester) . thus resulting eventually in low GE values . and difficult to handle since it releases ethane. Chapter 2 describes the ROP of p‐dioxanone initiated by a protected monosaccharide (1. and the presence of liquid‐solid reaction system leads to reduced reaction rates between starch and ε−caprolactone.4. Starch is a very hydrophilic material that always contains moisture. Previous studies showed that common ROP catalysts such as tin octoate or aluminium isopropoxide gave low (0‐14%) grafting efficiencies (GE. leading to formation of PCL homopolymers rather that starch‐g‐PCL. a very flammable by‐product. The water in starch granules competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. Another approach consists of the Ring Opening Polymerization (ROP) of ε‐caprolactone monomer in the presence of starch. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). The method applied basically consists of three steps (temporary partial protection of starch‐hydroxyl 16 . the synthesis of starch‐g‐PCL is reported. 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 hydroxyl groups of starch are supposed to function as initiating sites. Another possible cause for the low GE values is the heterogenous nature of the reaction. 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.Chapter 1 1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) using i Al(O Pr)3 as the catalyst.4. and starch‐g‐ biopolymer formation. Starch‐based blends by in situ polymerization The synthesis of starch based graft copolymers by in situ polymerization represents not only an alternative route to melt blending (vide supra) for the production of novel biomaterials. The results of this study have been used as input for the synthesis of the starch‐g‐PCL.3. In this reaction. A new strategy for the in situ ROP of ε−caprolactone on starch with the use of common ROP catalyst is therefore highly desirable. Three routes have been studied in detail: starch esterification. In the past. starch‐g‐PCL has been synthesized using toxic materials such as isocyanates .2. In Chapter 3.
Y. and removal of the silyl groups). Denmark: http://www. Breed. 2002. Holmes: Interactive European Network for Industrial Crops and their Applications.html. Corn. R. C. The model.B. Corn Refiners Annual Report 2007. in Modified Starches: Properties and Uses (Ed. Passchall). .F.N. Wurzburg: Introduction. Central Science Laboratory.. 1. E.H. O. . Nawrath.M. Fowler: Industrial markets for starch. E. USA. Young: Fractionation of starch. Bemiller. a systematic experimental study on the effect of the process variables on the starch ester DS has been performed.starch. J..5. Based on these preliminary results. 1. New Jersey. 2008. C. UK. Whistler. Academic Press. A preliminary study on the effect of several process variables (reactant ratio. Somerville: Mini review: Plant polymers for biodegradable plastics: cellulose. ROP of ε−CL on the remaining starch‐hydroxyl groups.C. Princeton University Press. . Wurzburg). are also reported. in Starch: Chemistry and Technology (Ed. Summary Report for the European Union. Part of A Global Economy. C. accessed on August 01. London. Gywnedd. CRC Press. Corn Refiners Association. Mol. A. O. References .. . USA.L. starch. 2007 17 . 2004. 2000‐2005. The use of these compatibilizers in starch/PCL blends.S. The synthesis of two interesting compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) for starch/PCL blends is discussed in Chapter 6. The results were quantified using a statistical model. The Biocomposites Centre. Aarhus. Introduction by trimethylsilyl groups. York. UK. and application of different catalysts) on starch esterification using vinyl fatty esters (vinyl laurate and vinyl stearate) is provided in Chapter 4. 2005. . R.B. An Introduction to the New Science of Biodegradable Plastics. 1986. and polyhydroxyalkanoates. P. Poirier. addition of co‐solvent. Sand Hutton. including the mechanical and thermal properties. USA. Boca Raton. Website of International Starch Institute. Stevens: Green Plastics. . Inc. Washington D. Agricultural and Rural Strategy Group. Bangor. 1995.A. University of Wales. Inc. UK. 105‐122. together with the mechanical and thermal properties of the synthesized starch esters are reported in Chapter 5. de Bragança. 1984. .dk/isi/stat/rawmaterial.
Starch/ Starke 2001. R. 2000. 557‐562. . UK.. 2000. Ren: Biodegradable plastics: a solution or challenge?. . X. John Wiley & Sons.E. 105‐109 . S. S. 18 . Academic Press. C. Wolf.L. Techno‐economic Feasibility of Large‐Scale Production of Bio‐based Polymers in Europe. Sci. Whistler. accessed on August 19.M. Bastioli: Global status of the production of biobased packaging materials. Whistler.R.J. J. M. Passchall).. Cleaner Prod. J. 27‐40. R. O. 2000. R. Strong: Plastics: Materials and Processing. Germany. M. 23. G. Fritz. J.F. Culter: Plastics Packaging. Chandra. Inc. Weith: Plastics. F. Prog.D. Watson: Corn and sorghum starches: production.351‐355. Denmark. 1998. S. Braun: Environmental impacts of biobased/ biodegradable packaging. . Link. 31. Seidel). in Kirk Othmer Encyclopedia of Chemical Technology Volume 22. Munich. . C. Applications. 2008 . J. A. Hernandez. Prentice‐Hall. Inc. Marscheider‐Weidemann. in Starch: Chemistry and Technology (Ed. Weber). Fifth Edition (Ed. The Royal Veterinary and Agricultural University. USA: http://www. . R. USA. 2008. 1273‐1335 .W. Website of Environmental Protection Agency. . USA. Patel. Germany.L. 1990.J. E. . Chem.J.gov/garbage/plastic. Rustgi: Biodegradable polymers.N. demand and recovery for 2006 in Europe. . New Jersey. Processing. Bastioli: Global status of the production of biobased packaging materials. B. R.epa. Hanser Publishers. Belgium. and Regulations. A. . The compelling facts about plastics: An analysis of plastics production. U. Daniel: Starch. Technomic Publishing Company. Brussels. Starch/ Starke 2001. Frederiksberg. 11.E. Properties. Selke: Packaging and the Environment. Inc. Department of Dairy and Food Science. . Schleich. Lancaster. G. Hanser Publishers. 1939. Applications. 2003. J.M. A. . Plastics Europe. Washington DC. 2005. Polym. J. Ind.. London. Eng. Englewood Cliffs. Inc. Bemiller. USA. Crank. Hoboken. Association of Plastics Manufacturers.A. Institute for Prospective Technological Studies and European Science and Technology Observatory. R. 53.B. 1984. Husing. C. Munich. 1996. 53 . Angerer: Technical Report Series. Properties.Chapter 1 . the Food Biopack Conference Proceeding (Ed. Selke.htm.. Ehrenstein: Polymeric Materials: Structure.
Kruiskamp. Spain.J. E. 1980. X.P.S. Smits. Denmark. 53. Ind.B. . Mikkilti. Chem. N. Martin. 409–416 . J.K. P. T. 34. 2003. properties.F.J. Nielsen. M. P. Physical and mechanical properties of biobased materials – starch. Muller. J. Res. 275‐282 19 . and Pukanszky: Thermoplastic starch/layered silicate composites: structure. Biobased packaging materials for the food industry. Polym. Chambre Syndicale des Emballages en Matiere Plastique (French Plastic and Flexible Packaging Association). Otey. and Couturier: Properties of biodegradable multilayer films based on plasticized wheat starch. and Olsen. Carvalho. Moates.. 53. Carbohydr. Yu. July 15. Bagdi. 331–336 . Ma: The effects of citric acid on the properties of thermoplastic starch plasticized by glycerol. K.M. Crops Prod. K.M. J. Westhoff. Rev. Ind. van Soest. A Real Case of “Force Majeure” that should be passed through to Packaging Prices Immediately. Dev. Schwach. W. polylactate and polyhydroxybutyrate. Wang. A. P. R. 2005. Biodegradable thermoplastic starch. Weber).Lin: Note. Wang: Properties of starch blends with biodegradable polymers. Gao. Forssella. 356‐361 . B. 2005. 385–409 . 43. 1996.F.H. J. 1491‐1494 . a model for thermoplastic starch. O. Macromol. X. Frederiksberg C. Vliegenthart: Interaction between dry starch and plasticisers glycerol or ethylene glycol. Polymer Sci.A. 21. interaction. Parker: Phase and glass transition behaviour. J. Appl. the Royal Veterinary and Agricultural University. of concentrated barley starch‐glycerol‐water mixtures. Polym.G. F. . Starch/ Stärke 2001. Sci. 2003. Carbohydr. Department of Dairy and Food Science. A. .V. J. Compos. Y. Introduction European Commission Directorate General of Joint Research Centre. 4.M. Avérous.. 2008. report of Food Biopack Project (Ed.G. Doane: Starch‐based blown films. L. Interfaces 2006. J. 2000. 57. 494–504 . Press Release: Major Price Hikes in Plastics. A. 53. 62. Petersen. Gandini: Surface chemical modification of thermoplastic starch: reactions with isocyanates. Part C: Polym. P. Seville. 1–17 . 2005. Eng. 372–380 . Y. J. epoxy functions and stearoyl chloride. G.L. Starch/Stärke. Yu. A. Prod. J. R.H.M. Wang. Yang. C. Starch/ Starke 2001. 1997.. Curvelo. . K. status and perspectives. 592‐595. 13. measured by differential scanning calorimetry and solid state NMR spectroscopy.
. 2594–2604 . 705‐721.P. 2001.. E. Inc. O. Ribeiro. 46–51 . A.A. J. Myllarinen. Boca Raton. P. Inc. CRC Press. Crops Prod. Carbohydr. W. Pacsu: Starch studies: Possible industrial utilization of starch esters. B. P. R. Panayiotou: Processing and characterization of starch/polycaprolactone products.A. Kontoyiannis. Feller. Appl. 56. Ind.W. A. Silva. 17–24 . 20 .F. Sagar. E. 581–586 . J. Carvalho. 65. Polym. 77.B. Grohens: Starch modifications. 1986. E. Tserki. 11–17. T. Mullen. Sci. Polymer 2003. Starch/Stärke 2005. . Coutuier: Comparative biodegradation study of starch‐ and polylactic acid‐based materials. H. T. 2002. K. . Matzinos.J. Levesque. E. 58. Sun: Reinforced thermoplastic acetylated starch with layered silicates. Ind.. 2004. 58 (9). 1942.D. F.C. V. Polym.H. B. Y. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. B. I.C. Pacsu: Starch studies: Preparation and properties of starch triesters. Andrade: Reactive processing and mechanical properties of cross‐linked maize starch. Mullen. Appl. 1. Journal of Applied Polymer Science 1995. X. Chem. Alric. Gattin. 1209‐1217. Jiang.B. F. .W. Ramsay. A. destructuration and hydrolysis during O‐formylation. Aburto. M.W. 55‐78. Rodriguez‐Gonzalez. 1986. G. 44. Y. 24. CRC Press. Favis: High performance LDPE/ thermoplastic starch blends: a sustainable alternative to pure polyethylene.D. Borredon. Merrill: Properties of fatty‐acid esters of starch. O. van der Zee. C. Polym. M. USA. J. C. chapter 4 in book Modified Starches: Properties and Uses. Boca Raton. 381‐384 . S.A. Bikiaris. O. 139–147 . Ibezim. 2006. Thiebaud. Sci. . J. Prinos. 1999. 74. Hulleman. J. Polym.Chapter 1 . Degrad. C.D. J. Qiao. M. Ramsay. Eng. E. 34. Jarowenko: Acetylated starch and miscellanous organic esters. C. S. I.A.W. 1517–1526 . D.J. Feil: Morphology and enzymatic degradation of thermoplastic starch– polycaprolactone blends. J. USA. Ind. (editor).D.B. Wurzburg. Stab. Eng. 35. W. 1997. Chem. Wurzburg: Cross‐linked starches. Copinet. 57. Vikman.T. 2004. 1943. Environ. Divers. Pilin. C. B. J. Polym. 389‐398 . in Modified Starches: Properties and Uses (Ed. Wurzburg). Rodriguez‐Gonzalez. Bertrand. Favis: Rheological and thermal properties of thermoplastic with high glycerol content. 1647‐1656.
D. 51. R. . Broekroelofs. Thesis. C. . Duquesne. Bikiaris. Hu. Carbohydr. Aburto. Starch‐ Starke 2005. 74.Coleman. 1997. A. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides.W. L. Panayiotou: Synthesis. Rutot. Lancaster. Khandpur. . P.928. C. A. 29. J. Germany. . Dicke. F.E. B. C. S. P. 1996 .E. Appl. I. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2.Stokes. G. 101‐112. de Graaf: Ph.Fennel Evans: Fundamental of Interfacial Engineering. Nakayama. M. J.K. Starch‐Starke 1999. Bakker. I. D. J. . J. . Carbohydr. Scott. 132‐135. J. Aburto. Macosko. Mormann. . Inoue: Compatibilizers for melt blending: Premade block copolymers. Symp. M. 2001. Thiebaud. Bakker and G. 28. 1960 . Macromolecules 1996. 302‐307. . Hu).F. Starch‐Starke 1999.E. US Patent 2. de Graaf. Beenackers: The kinetics of the acetylation of gelatinised potato starch.H. Dubois: Synthesis and characterization of compatibilized poly(ε‐caprolactone)/ granular starch composites. Sci. Senocq.828. 1999. Prinos. Mouysset‐Baziard. Borredon. A. Polym. USA. Janssen. C. J. E. Starch‐Starke 2004. 11. Wiley‐VCH. Aburto. S. Smith. 34. in Reactive Polymer Blending (Ed. Introduction . 1440‐1451. R. Munich. Inc.. 1991. Prinos. Alric. New York.Graf.M. Alric. Alric. Hamaili. Polym.C. A. W. I. 2001. A. A.J. E. H.D. I. 175. Aburto. D. . A. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. characterization. Hanser Publishers. Thiebaud. Tuschoff: Acylation of hydroxy compounds with vinyl esters. T. Aburto. W. 1995. Borredon. Degee. P. Panayiotou: Properties of octanoated starch and its blends with polyethylene. D. Alric. P. I. Alric. G. Al‐Higari: Acylation of starch with vinyl acetate in water.E. Rijksuniversiteit Groningen. and biodegradability of fatty‐acid esters of amylose and starch. 1997. R. R. J. Guegan. C. E. USA. 5590‐5598 . . 51. P. Introduction. G. 255‐263. E.H. 57. C. M. A. 137‐144. W. .V. Technomic Publishing Company. M. P. 145‐152. E. Polym. J. A straight way to regioselectively functionalized polysaccharide esters. E. 56.Painter: Specific Interactions and the Miscibility of Polymer Blends. Macromol. Marechal. 118‐121. Bikiaris. Cellulose 2004. J. 33‐43 21 . J.
B. 167–172 .M. 1997. 1841‐1846 . W. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends. 236–241 . 2005. Verhoogt: Processing and characterization of thermoplastic starch/polyethylene blends. 20.S.A. 1991. N. Mechanical properties of sago starch‐filled linear low density polyethylene (LLDPE) composites. Polvm.. Bikiaris.D. Chem. Appl. 2003. Pedroso. U. Testing.A. III.A. Adv. Polym.L. Danjaji.Chapter 1 . 59. E. L. Polym. Z. and tensile properties of Low‐Density Polyethylene (LDPE) and corn starch blends. H. 1998.S. Stab. Danjaji. thermal and morphological properties of virgin and reprocessed LDPE/corn starch blends. 30. Effect of compounding and starch modification on properties of starch‐filled Low‐ Density Polyethylene. Y. E. B. Frangis. H. Polym.S. Sci. Duran. Effect of amylopectin to amylose ratio in starch. Favis. Z. Proc.I. Prinos. J. D. Testing. J. 1998. Panayiotou: LDPE/Starch blends compatibilized with PE‐g‐ MA copolymers. Appl.G. 70. N. R. K. St‐Pierre.H.M. R. J. morphology.G. 2003. Ind. Nawang. I. Mater. Pedroso. . D. Wang. Mohd Ishak: Material behaviour. 287‐291 . Koutsopoulos. Mohd Ishak: Material properties. Gelina. Polym. 1242‐1251 . Eng.S. Ramsay. J. Technol. A.G. 310–317 . Polym. A.D. 1467‐1475. Almeida. Panayiotou: LDPE/plasticized starch blends containing PE‐g‐MA copolymer as compatibilizer. A. 88. N. 1‐9 . Polym. Mei: Study and development of LDPE/starch partially biodegradable compounds. thermal and morphological characterization of recycled LDPE/corn starch blends. Res. Carbohydr. Evangelista. N. 21. M. R. Polym. C. Sung. 2005.D. W. J. Liu. Vouroutzis. L.S. L. 1503–1521 . Ramsay. 88. 2001. Degradation studies and moisture uptake of sago‐starch‐ filled linear low‐density polyethylene composites. 38. Mani. D. Ishiaku. Nakamura. 34. 162–163. Nikolov. Sci. R. Z. 59. C. H. .J. J. 1998. 75–81 . Ribes‐Greus: Thermal and viscoelastic properties of some commercial starch products. I. Contat‐Rodrigo. Cordi. Sci. Eur. 647‐655 22 . Polymer. Sun: Effects of polyethylene‐grafted maleic anhydride (PE‐g‐MA) on thermal properties. Jane. 16.‐J. Polym. Nawang. Bikiaris. G. Z. Ismail. J. Appl. J. Rosa: Mechanical. Ismail. 2005. 2904–2911. Pavlidou.A. R. Tech. Rosa: Effects of the compatibilizer PE‐g‐GMA on the mechanical. 2002. U. Ishiaku. Degr. L. D.
G. B. T. Polym. Chem. 1987. Appl. S. Low density polyethylene/ starch blend films for food packaging applications.P. 2004. Odusanya. Kim. 2.M. . Polym. Ind. Sci. Liu. A‐C. 41. Doane: Starch‐based blown films. Stridsberg. Manan. A.H.M. Eur. 179‐187. Environ.R. O. R. 139‐161. . Rodrigues. 23 . . 157.H. A. 32–45 . Z. U. Res. 87. E. Sci.S.. Eng. H. Chem. J. A. 35. M. Westhoff. Calil: Effect of thermal aging on the biodegradation of PCL. Chem. Otey. Westhoff: Starch‐Based Films. B.M. Ang. Appl. Introduction . 284‐287 . and their blends with starch in soil compost. Lindblad. J. 1507– 1516. . dos Santos Rosa. Jung. 267.M. Siddaramaiah. Sci. Chem. Bhattacharya Residual stresses and density gradient in injection molded starch/synthetic polymer blends Polymer 2000. Polym. 42‐65. 89. Makromol. Averous. Ranucci. M.M. . 2002. J. Polym. 2003.W. S. C. O. 1659‐1663 . Adv. P. C. Tang. 40. J. John. S. U. 2007. Preliminary diffusion evaluation. Ind. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. 1997. M. Garg.P. 26. S. 2003. 157. K. Sci. L. Ishiaku. Jana: Studies on the properties and characteristics of starch– LDPE blend films using cross‐linked. . Sci.C. Sankar. 1984. Adv.S. Cho.‐C.H. Polymer 2000. 43. 1298‐1305. Ryner. Maiti: A biodegradable film. K. Kammer: On mechanical properties of sago starch/poly(ε‐caprolactone) composites. Albertsson. Dev. in Polym. F. 12. R. D. Prod. 2004. 41. J. 16–19 . Moro. Polym. Otey. F. 9177–9190 . Azemi: Effect of starch predrying on the mechanical properties of starch/poly(e‐caprolactone) composites. cross‐linked and glycerol modified starch. Polym. Y. D. 81.F. W. J. Azemi. Res. PHB‐V.K. Pilot plant investigation for production of the biodegradable film. Dole. 2002. Odusanya. Eng.K. J. Raj. 3976–3987 . Part A: Polym. Polym. B. B. M. M. C. J. 23. J. T. 23. Polym. J. 877–884. 1139–1148. K. U. Appl. Sci. N. 3539–3546 . Albertsson: Controlled Ring‐Opening Polymerization: Polymers with designed Macromolecular Architecture. J. Sci. C. Jana. Park: Modification of aliphatic polyesters and their reactive blends with starch.D. S.S. Karlsson: Polymers from Renewable Resources. Manan. Adv.H. Sen. Ishiaku. glycerol modified. A. 2000. Kim. 1999.. Tech.Y. Guedes. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. M. Kim. L. 4157–4167 . 2001. M. Eng. Yang.. 1. Fringant: Properties of thermoplastic blends: starch–polycaprolactone. Polym.N.S.K.
Rutot. and P. Rutot. D. D. P. . E. 1432–1442. Composite Interfaces. 215–225. Dubois. Raimo. Appl. Krishnan. 2003. Polym. P. Bhattacharya: Synthesis and characterization of starch‐ graft‐polycaprolactone as compatibilizer for starch/polycaprolactone blends. Degée. 3875–3881 . R. R. Polym. 127–134. Avella. Dubois: Synthesis and characterization of compatibilized poly (e‐caprolactone)/ granular starch composites. M. 1998. 2002. E. 2001. 41. 283–286. Macromol. Duquesne.E. Polymer 1999. J.Chapter 1 . R. Martuscelli. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite.E. Rimedio: Polymer communication: Preparation and characterisation of compatibilised polycaprolactone/ starch composites. Symp. . Sci. Narayan. Dubois: Aliphatic polyester‐grafted starch composites by in situ ring opening polymerization. Errico. 2000. Rimedio. 175. J. Sadocco: Preparation of biodegradable polyesters/high‐amylose‐starch composites by reactive blending and their characterization. Degee. Tang. P. M. C. P. Degrad. . 83. Rapid Commun. 7. 19. M. 24 . 40. M. P. 33–43 . Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. M. Mani. Macromol. Laurienzo. Avella. M. Stab. R. P.S. . R. M. 80. Errico. Polymer 2000. 3091‐3100.
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. polyesters. Besides the desired polydioxanone functionalised with a monosaccharide end‐group. The products were synthesized at 60‐100°C and characterized by 1H‐ and 13C‐NMR.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose)/Al(OiPr)3 initiator system to yield polydioxanone with a protected monosaccharide end‐ group is described. ring‐opening polymerization .3. and MALDI‐TOF mass spectrometry. with higher temperatures (100°C) leading to lower yields. Systematic studies showed that the polymer yield is a function of the reaction temperature and the reaction time. Keywords: biodegradable. also polydioxanone with an OiPr end‐group was formed (20‐30 %).2. The average chain length of the polymers is between 7 and 58 repeating units and may be tuned by the monomer to monosaccharide ratio (at constant Al(OiPr)3 intake). 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.
This melting point is considerably higher than typically found for polycaprolactone (60 oC) and lower than that of polylactides (at least 175°C). probably because the p‐dioxanone monomer has become commercially available only recently . polydioxanone also shows good performance. Introduction Aliphatic polyesters. reaction of the metal‐alkoxide with an appropriate alcohol allows the synthesis of end‐capped poly‐lactones. tocopherol. particularly for nano‐ encapsulation systems for drug delivery . are interesting polymers because of their good product performance and biodegradability . accessible by the polymerization of p‐dioxanone (1). The ring‐opening polymerization (ROP) of p‐dioxanone using metal catalysts such as aluminum isopropoxide [Al(OiPr)3]. and is tougher than polylactides and even HDPE . Polydioxanone has a tensile strength close to 48. Biodegradable aliphatic polyesters end‐capped with sugar molecules have been studied extensively for use in biomedical applications. From a biodegradability point of view. such as polycaprolactone. and polylactides. This chapter describes experimental studies on the catalytic ROP of p‐ dioxanone in the presence of a protected glucose molecule (1. stigmasterol and ergocalciferol)  and also with protected monosaccharides (galactopyranose/ glucofuranose) [6‐7]. Polydioxanone (poly(p‐dioxanone) or poly(1. only limited information about the synthesis and properties of polydioxanone is available in the open literature. testosterone. 9]. has interesting product properties compared to other aliphatic polyesters. The catalysts are metal alkoxides with Lewis acidic character [8. 2). Several polymers with bioactive alcoholic and phenolic end‐groups of interest for drug‐related applications were synthesized (for example geraniol. quinine.3 MPa and an elongation at break of 500‐600%. stannous octoate [Sn(Oct)2]. Nishida et al  reported that polydioxanone decomposes to non‐toxic gases (CO2 and H2O) by microorganisms. which is a unique compromise between application and processing temperature. Besides 26 .g.1.Chapter 2 2. The synthesis of protected monosaccharide end‐capped biodegradable polymers is usually performed via Ring Opening Polymerization (ROP). or zinc lactate has been reported.3. which to the best of our knowledge is the first study to functionalise polydioxanone with a monosaccharide.2. It is fully degraded in the body within a period of 180 days . Its melting temperature is close to 110 oC.4‐dioxan‐2‐ one)). polyglycolide. Exchange of the alkoxide group by e. The relatively low melting point of polycaprolactone limits its applicability. whereas the high temperature for polylactides results in thermal degradation and undesirable transfer reactions during synthesis and processing steps . Despite its good properties.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. pregnenolone. 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.
8 g.5 mmol) was dissolved in toluene (1 ml) at 50 oC.2.2. 3]. The reaction mixture was clear and colorless during the reaction.3. Next. Boehringer Ingelheim.8 ml of a solution of an aluminum isopropoxide stock solution was added. However. After the pre‐determined reaction time. Subsequently. 97% (Sigma) and aluminum isopropoxide. part of this solution (350 µl.2. CDCl3 was obtained from Sigma and was used as received. To this solution. 2.2.g. NMR) due to the poor solubility of the products in standard organic solvents.08 mmol aluminum isopropoxide) was added to pure 1 (0. As such. The resulting clear solution was stirred for 2 h at 50 oC. the results of this study are also of interest for the preparation of starch/polydioxanone polymers using ROP. containing 0.2. 1.44 g. Methods All polymerization experiments were carried out under a protective nitrogen atmosphere using standard Schlenk‐ and glovebox techniques.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2) 2 (1. Toluene (Labscan) was dried and stored on molecular sieves 3 Å (Labscan) under nitrogen. 98+% (Aldrich) were used as received. 5. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System potential applications in biomedical products. heptane (Acros). and diethylether (Labscan) were used as received. Particularly interesting in this field are starch polymers grafted with polydioxanone.2. the synthetic pathways and the soluble.1. The polymerization was allowed to proceed for 16 h at 100 oC. poorly soluble heterogeneous systems. 0. relatively low molecular weight compounds reported in this study may be viewed as model systems for more complex. the mixture was brought to room temperature and several drops of HCl (1 N) were added to stop the reaction.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2). Materials and Methods 2. which was pre‐heated till about 60°C to obtain it in a liquid state. hot dichloroethane (20‐25 ml) was added to completely dissolve the partly solid polymer at room temperature. (20 mmol) of aluminum isopropoxide to 20 ml of toluene.1. 2. The hot solution was 27 . This stock solution was prepared by adding 4.55 mmol of 2 and 0. Analytical grade dichloroethane (Labscan). Germany) was purified according to the procedure described by Raquez et al [2. 2.2. Materials p‐Dioxanone monomer (1.07 g.3. 8 mmol).2. Typical example for the synthesis of polydioxanone end‐capped with 1. grafting efficiencies are difficult to determine in this system by standard analytical techniques (e.
2. two polymers with different end‐groups are present. it is assumed that all available initiator is used effectively.3. 1H‐ and 13C‐NMR spectra were recorded on a Varian AMX 400 NMR.1. 2D‐NMR spectra were recorded on a Varian Unity 500 NMR. Product Analyses NMR analyses were performed in CDCl3.3. X n exp of the product. 2. Processing of the raw data was performed using VNMR software. MALDI‐TOF spectra were recorded on an Applied Biosystems Voyager DE‐PRO machine using dithranol/ NaI as the matrix (linear mode). As will be shown later. one initiated on a galactopyranose molecule and the other on an isopropoxide group. The isolated yield at this condition was 68%. 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. 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).) H‐NMR was applied to determine the experimental average degree of polymerization. If the amount of sugar is in excess with respect to the aluminum catalyst.4. 4:1 by volume) at 4‐8 oC. The white solid was finally separated from the liquid by decantation and dried in a vacuum oven (5 mbar.) Here. the above equation simplifies to Xn theo = monomer conversion × [monomer ]0 [ sugar ]0 (2.2. Calculation of Average Degree of Polymerization The Theoretical Average Degree of Polymerization.) H-isopropoxide end group 28 . 40 oC) until constant weight.2.Chapter 2 precipitated in a heptane/ ether mixture (300‐400 ml. This leads to the following equation: Xn exp X n exp is = A A H-repeating unit H-galactopyranose end group +A (2.
2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System The NMR peaks of the repeating unit overlap partially with those of the galactopyranose end group. Schematic representation of the polymerization reaction including atom numbering scheme 29 .). an Al ‐ monosaccharide 2 ‐ monomer 1 mol ratio of 1: 6.6 ppm + A 5.) 2.7. Results and Discussions 2.). The polymerization reaction was performed as a two step process. To avoid the formation of isopropoxide end groups.3.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose (2) and Al(OiPr)3 as the catalyst precursor (Scheme 2.4− 4.3.6: 96 was applied.1.4. 2.5.) The ratio of the two different types of polymers (either end‐capped with galactopyranose or an isopropoxide group). is calculated using: R gp = A H − galactopyranose end group H − isopropoxide end group A (2. R gp .) or.1. In the first step.1 ppm (2.6 ppm ] / 6 A 5. the protected mono‐saccharide 2 reacts with Al(OiPr)3 to form the actual catalyst for the polymerization reaction.3.3. is rewritten as: Xn exp = [ A 3.6 ppm A 5. an excess of monosaccharide 2 on Al was used (see eq. eq.6. Screening Experiments Initial experiments to synthesize protected galactopyranose end‐capped polydioxanone (3) were performed using 1.1. in term of the NMR resonances : R gp = A 5. Typically. 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 2.7 ppm − 6 × A 5.1 ppm (2. To compensate for this effect. The exchange reaction typically takes place at 50°C for 2 h.2.
. The typical proton resonances of the p‐dioxanone (3. Proton resonances of the CH3 group of the isopropoxide end‐group. Typical isolated product yields are 68% at these conditions.78 to 4. Although the spectra clearly indicate the presence of the monosaccharide 2. The off‐white solid reaction product was collected after a dissolution‐reprecipitation process using dichloroethane and a heptane/ diethyl ether mixture.40 ppm) unit are broadened and shifted up to 0.1. However. 2. the in‐situ formed catalyst was reacted with p‐dioxanone monomer 1 at 100 oC for 16 h.3. of interest is the presence of a small multiplet at about 5.51 ppm.1 and 2.Chapter 2 H H CH2OH iPr O Al O iPr iPr O O O H O H H O H O O H O H O O H O Al O CH2 O O H O H H O O H O H H O + O 3 H H2C H2C O H H O H + H O 3 iPr OH O (2.1. although it is evident that p‐dioxanone polymerisation occurred. respectively.1 ppm compared to the monomer. together with the CH3 groups from the protecting groups of the sugar appear in the range 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. Product analyses The products were analyzed using 1H‐ and 13C‐ NMR and MALDI‐TOF. Typical 1 H‐ and 13C‐NMR spectra of 3 are shown in Figure 2.7. The OiPr : 2 end‐group ratio for the standard experiment was 1: 2.) are not particularly informative.) In the next step.15‐1.1 ppm. 1 30 .67. It confirms that polymer initiation not only occurs with the monosaccharide but also with the remaining isopropoxide group of the catalyst precursor (vide infra). H‐NMR spectra (Figure 2. This multiplet is characteristic for the CH proton of an isopropoxide end‐group. It is difficult to determine the end‐groups of the polymer on the basis of 1H‐NMR.1.2.
4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.52 ppm (β’).61 (γ’) and 73. H‐NMR Spectra of: (a).17 and 69. 68.2 ppm. In addition to the mono‐saccharide end‐ group.8. 1. The carbon resonances of the polydioxanone backbone are present at δ = 63. 3 1 C‐NMR (Figure 2. Carbon resonances arising from the monosaccharide end‐group are present between δ = 66.28 ppm in 2 to δ 66. 1 (c).2 ppm.3 ppm are shifted considerably.2.3. p‐dioxanone monomer. 1.2 and 96.2.) is more informative and clearly shows the presence of a polydioxanone polymer backbone and a monosaccharide end group. 13 31 . Particularly the C2‐C6 carbons in the range δ = 66.2 ppm in product 3.2‐70.1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone. 2 (b). C‐6 is shifted from δ = 62. For instance.2.3. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2. characteristic resonances of an ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end group are present at δ = 61.
the numbering scheme of carbons and protons is given in Scheme 2. C‐ NMR Spectra of: (a).97 ppm (‐CH‐(CH3)2).73 and 109.76‐25.Chapter 2 Figure 2. 3 13 Carbon resonances of the isopropoxide groups from the by‐product (isopropoxide end‐capped polydioxanone) are present at δ = 68. 1 (c).3.2. 1. An overview of the data is given in Tables 2.2.97 (>C‐(CH3)2).2. 2 (b).1. A typical example of a part of the 2D‐NMR spectra is given in Figure 2.1.3. 2 32 . p‐dioxanone monomer.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone. The resonances of the protecting group of the sugar appears at δ = 108. and 2.76‐25. 1.2.70 (>C‐(CH3)2) and between δ = 21. D‐NMR (HSQC) was applied for complete peak assignment of the product.48 ppm (‐CH‐ (CH3)2) and between δ = 21..4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.3.
31 4.77 3. 33 .83 p‐dioxanone Monomer (1) H‐α H‐β H‐γ a 4.74 3.40 all values were determined using 1H NMR.54 4.17 3.09 4. which were determined from HSQC spectrum due to overlapping resonances.732 broad.78 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 5.3.1.00 4.60 4.68 4.16* 4. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2. except values with *.301 3.50 4.32 * 4.26 3.59 4.8 Reactant peaks (ppm) Galactopyranose (2) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 5.25 3. HSQC 13C‐1H spectra of 3 Table 2. around 1.
52 63. MALDI‐TOF was also applied to characterize the products.61 170.80 61.63 68.8. 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.) This series represents a dioxanone polymer end capped with 2 and an additional Na ion.92 68.21 NMR analyses also allow calculation of the molecular weight of the products.10 71. corresponding with an average number molecular weight (Mn) of about 1600.2 73. The product obtained at screening conditions (16 h reaction time at 100°C) contains on average 13 monomer units.09 65. For this purpose. which is the molecular weight of a repeating dioxanone unit. The molecular weight distribution of the major peaks may be represented by the following relation: m / z = 23 + 260 + 102n (2. the ratio of the intensity of the end groups and the polymer backbone peaks is determined.17 69. The latter likely stems from the matrix used to ionize the sample.Chapter 2 Table 2.75 70.07 Reactant Peaks (ppm) Galactopyranose (2) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 96.28 70. 34 . A typical molecular weight distribution is observed.28 p‐dioxanone monomer (1) C‐α C‐β C‐γ C‐carbonyl 61.2. A detailed procedure is given in the experimental section.11/ 170.21 70.31 70.54 166.33 70.2 68.92 66.57 62.00 171. An example of a MALDI‐TOF spectrum of 3 recorded in a dithranol/NaI matrix is given in Figure 2.58 68.4. The difference in molecular weight between the main peaks is 102 g/mol.
m / z = 23 + 18 + 102n m / z = 1 + 18 + 102n 35 (2. for details. see Figure 2.) and imply the presence of dioxanone polymers with carboxylic end groups. ionized with either Na+ or H+.11. Typical MALDI‐TOF spectra of 3 A second molecular weight distribution is also clearly visible when enlarging the spectra.5 +102 1400 1600 1800 2000 2200 2400 Mass (m/z) Figure 2.4.) (2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 100 90 80 70 60 %-Intensity 50 40 30 20 10 0 600 800 1000 1200 1098.5.10.9.) and (2.) . Furthermore.5.10.11. These distributions may be represented by equation (2. This series may be described by the following relation: m / z = 23 + 60 + 102n (2. in line with the NMR data.). two other distributions are present.) This relation is indicative for the presence of a polydioxanone polymer containing an isopropoxide end‐group. although both with a very low intensity (Figure 2.
[6.3 10 836. but in all cases poor quality. is not particularly suitable for polydioxanone. 2. Enlarged MALDI‐TOF spectra for 3 Although MALDI‐TOF clearly demonstrated the presence of various types of end‐groups in the product.2.7 796.3. although successfully applied for galactopyranose‐end‐capped polycaprolactone . different molecular weights were analysed. Various other matrices were tested (e.5. Mechanistic aspects Both NMR and MALDI‐TOF measurements imply that the main product is indeed the desired monosaccharide end‐capped polydioxanone. The observed differences in the molecular weight distributions of the various samples were only marginal. In addition. according to NMR.Chapter 2 50 40 792. 2‐(4′‐ hydroxybenzeneazo)benzoic acid (HABA)).7 %-Intensity 30 20 858. low resolution spectra were obtained. Various samples with.6.1. the catalyst precursor is treated with monosaccharide 2 resulting in an alcohol exchange 36 .13]. 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. it proved not suitable to determine the average molecular weight of the products. In the first step. small but significant amounts (20‐30%‐mol) of polydioxanone chains with an OiPr end group are present. Most likely the matrix (dithranol/NaI).1 0 600 700 800 900 Mass (m/z) 1000 1100 1200 Figure 2.g.
6. a p‐dioxanone molecule will coordinate to the Lewis acidic aluminium center followed by an insertion step. iPr iPr O Al O iPr + R OH (in excess) O O Al O R + O iPr OH O Al O O iPr + O + O O O O n O l O O O Al [O CH2 CH2 O CH2 C ] n OR O O O O Al [O CH2 CH2 O CH2 C ] O n+m OR + m R-OH O + HX (termination) O Al O R O O Al X O O + + O O H [O CH2 CH2 O CH2 C ] l by-product O iPr H [O CH2 CH2 O CH2 C ]n OR H [O CH2 CH2 O CH2 C ] n+m OR product Figure 2. The resulting Al‐alkoxide may again initiate a polymerisation reaction. 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. termination of the chain growth may occur by reaction with an alcohol. 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. Subsequent coordination and insertion of dioxanone molecules leads to the formation of a polymer chain with a monosaccharide end‐group. Simplified reaction scheme for the ROP of p‐dioxanone catalyzed by Al(OiPr)3. Next. (R = monosaccharide 2) 37 .
5 hrs and 16 hrs) and a p‐dioxanone to monosaccharide 2 molar ratio ranging between 3.6 6. despite the excess of 2. An overview of the experiments and the results are given in Table 2. time and the mol ratio of monomer to monosaccharide) on the yield.81 16.1 86. two different reaction times (1. Set Sample S111 S112 S113 S114 S115 S211 S212 S213 S214 S215 S221 S222 S223 S224 S225 Overview of experiments a Processing Condition 2/Al ratio (mol/mol) 6.74 23.70 58.21 54.5 2.11 22.5 84.6 89.6 6.22 57.5 1.2.5. formed in the first step of the polymerisation reaction (eq.04 19.6 6.3 6.6 6.50 35.6 6.5 1.7.49 13.4 2.1 2. Chain Length (Xn exp ) Yieldb (%) 30.9 1 7. a (nearly) fixed Al(OiPr)3 : monosaccharide ratio of 1 : 6.1 2.29 14.10 15. A total of 15 experiments were performed at two polymerization temperatures (60 and 100 oC).3 3.30 62.6 6.5 67.36 17.5 1.1 3.45 19. 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 . 2.5 92.8 3.16 46.3.4 3.3 6.9 2. These may be present in the reaction mixture because the exchange reaction between Al(OiPr)3 and 2 was incomplete. 2.6 was applied.3 6.2 91.3 6.6 6.3 81.8 2.6 81. Systematic Studies The effect of important process variables (temperature.50 25.6 2.).3.Chapter 2 The minor product. However.55 Product Properties Avg. the end group may also be formed by a termination reaction with free isopropanol. Table 2.22 13.3 6.6 91.6 6.28 33. In all cases.5 1.93 8.96 2 3 a For each experiment.4 2. degree of polymerisation of the product and the end group distribution was determined.8 2.08 30.22 16. will be formed when the polymerization starts with an aluminum alkoxide with a remaining Al‐OiPr group (eq.93 37. 2.3 6.6 96.6 t (h) 16 16 16 16 16 16 16 16 16 16 1.4 2/OiPr ratio Rgp (mol/mol) 2.7 2.78 10.3 and 62.53 59.3‐6.85 14.0 87.6 86.5 81. OiPr‐end‐capped poly(p‐dioxanone).).74 22.214.171.124 42.74 52.5 T (°C) 100 100 100 100 100 60 60 60 60 60 60 60 60 60 60 dioxanone/ 2 ratio (mol/mol) 3.6 80.
Clearly. The effect of the reaction time on the product yield may be derived from the data provided in Table 2.3.5 and 16 h reaction time).5 h is higher than the yield at 16 h. 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. et al . It is likely that this relatively low molecular weight compound dissolves partly during isolation/ purification procedure leading to lower isolated yields. However. 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. On the basis of our experimental data and in line with literature data. which is expected to be more pronounced at lower temperatures. the monomer conversion reaches a maximum value before going down to the equilibrium monomer conversion.3. 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. 1.3. no detailed explanation has been put forward to explain this anomalous behaviour. Evidently. One of the experiments (S111) resulted in a very low yield (31%) compared to the other reaction. 39 . For the two experiments with a dioxanone: 2 ratio higher than 16 (≅ 23 and 60). Product yield The isolated yields of the reactions are all but one between 67 and 96% (see Table 2. equilibrium conversion is not yet achieved within 1. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 2.3.5 h. Higher polymerisation yields were obtained at lower temperatures. the yields for reactions conducted at 60 oC are always higher than those performed at 100 oC. when the reaction is performed at a low dioxanone: 2 ratio (≅ 16). leading to a low average chain length of the dioxanone polymer (7.3. as suggested by Raquez et al .). and particularly when comparing the data in set 2 and 3 (60oC. Another possibility is the occurrence of polymer crystallization. It was shown that at lower monomer to catalyst ratios. the equilibrium conversion of dioxanone at 80oC was about 80%.1. the yield at 1. For instance. with an enthalpy of polymerization of approximately ‐15 kJ/ mol . a low dioxanone: monosaccharide 2 ratio was applied. the yields are higher when performing the reaction at 16 h reaction times. will be required. we conclude that equilibrium conversion is achieved after 16 h and that the lower polymer yields at higher temperatures are due to the slight exothermicity of the reaction.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. see Table 2.). A more detailed analysis on the actual nature and composition of the polymerization products. To the best of our knowledge. which reduced to 75% when increasing the temperature to 120oC. The latter is due to the slight exothermicity of the reaction . For this particular experiment.2.
time: 1. Average Chain Length ( X n exp ) of the product as a function of the dioxanone : 2 mol ratio The end group distribution was determined using NMR and is expressed in terms of R gp (eq. A possible strategy to increase the Rgp values i.3. 2.5. and dioxanone/ 2 mol ratio) is absent. The X n exp increases linearly with respect to the dioxanone : 2 ratio. time: 16 hr Temperature: 60oC.3.2.7. to reduce the number of OiPr endgroups in the product may be the removal of isopropyl alcohol formed in step 40 .2.Chapter 2 2.. The R gp values (Table 2. in line with the theoretical predictions. time. Most experimental points are scattered along these lines. 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. reaction time and temperature on the X n exp of the products is shown in Figure 2. and 2. time: 16 hr 50 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. as expected for a typical ring opening polymerisation . In Figure 2. the X n theo (eq.7.7.3.e.) for all experiments are scattered randomly between 2. A clear trend between R gp and the process conditions (temperature.4 and 3.6.2. 60 Temperature: 100oC.) as a function of the dioxanone : 2 ratio at 90% and 100% monomer conversion is also provided.
3.8.3.).459*10‐7 41 . The low P‐value clearly indicates that the model is statistically significant.49 + 0.543 exp as a Model Error Total P‐value 3.4.) where t and T are respectively the polymerization time and temperature.) from the reaction mixture by e.977 (with an adjusted R2 value of 0.854 × (dioxanone : 2 ratio) + 0. A parity plot of the modeled versus experimental values of the average chain length Xn confirms this statement (Figure 2.4. indicating that the model describes the experimental data reasonably well.097 × (T ) (2.2. Statistical Data Analysis Quantification of the influence of the experimental factors (temperature. Table 2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 1 of the polymerisation process (eq.12. A linear model proved adequate to describe the effects of the independent variables on the X n exp : Xn exp = 8.974). 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 7.177 × (t ) − 0.6) was too limited to draw sound conclusions. The R2 value for the model is 0.3‐6.g.7. time and dioxanone/ 2 ratio) on the X n exp has been performed by multivariable linear regression  on the data given in Table 2.703 F 173. vacuum distillation before adding the dioxanone monomer . The analysis of variance for the model is given in Table 2. 2.3.
2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose. 17]. Furthermore. Modeled versus experimental values for the average chain length Xn The model predicts that the X n is a clear function of the p‐dioxanone : 2 mol ratio. In addition. Both NMR and MALDI‐TOF measurements indicate that the products are mixtures of polymers and significant 42 .4. 11] and due to the fact that the reaction is an equilibrium polymerization with a slight exothermic effect.3. the X n exp is negatively influenced by temperature. the model predicts that the X n exp is positively influenced by the polymerization time. 2. Within the experimental ranges. with high ratios leading to a higher average chain length.8.Chapter 2 60 50 40 Xn model 30 20 10 0 0 10 20 30 Xn exp 40 50 60 Figure 2. 16. Molecular weights (NMR) of the product were between 970 and 6200 and are a clear function of the p‐dioxanone/2 ratio (at constant Al(OiPr)3 intake). which is in agreement with the available data on ring opening polymerization [15. Conclusions The ROP of p‐dioxanone in the presence of a monosaccharide (1. This is in line with literature data [5. 2) with Al(OiPr)3 as the catalyst is reported. The isolated yields of the off‐white solid products were between 30 and 96%. the model allows determination of the process variables to obtain a product with the desired degree of polymerization. with higher ratios leading to higher molecular weights.
8419‐8425. Chem. D. 43 . Kricheldorf. P. starch).g. 1993. Rapid Commun.6. Chu. temperature and monomer: monosaccharide ratio) on the X n exp . Phys.R. A statistical model has been developed to quantify the effects of process variables (time.M. Raquez. Dubois: Some thermodynamic.J. Moreover. J. Degee. P. Grubb: Hydrolytic degradation and morphologic study of poly‐p‐dioxanone. Macromol.C. These studies will be reported in the next chapter. C. 2000. H. H. Biomed. Macromolecules 2001.4‐dioxan‐2‐one. 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. .4‐dioxan‐2‐ one initiated by Al((OPr)‐Pr‐i)(3) in bulk.M. 34. Zn‐lactate‐catalyzed polymerizations of 1.4‐dioxan‐2‐one and controlled synthesis of diblock copolymers with epsilon‐caprolactone. Degee. R. . Lin. Dubois: ʺCoordination‐insertionʺ ring‐ opening polymerization of 1.O. Mater. Narayan. Narayan.5. J. References . R. P. P. kinetic. Raquez. Damrau: Polylactones. 1089‐ 1097. 42. 21. 1998. 199. 27. and mechanistic aspects of the ring‐opening polymerization of 1.L. 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. 2. . D. Macromol. 1063‐1071. 153‐166. 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. Res.
. 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. silylation . Keywords: starch. the remainder being homopolymers of ε‐caprolactone. 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.Chapter 3 Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors Abstract Poly‐(ε)‐caprolactone grafted corn starch co‐polymers were synthesized using a hydrophobised silylated starch precursor. The DS of the polycaprolactone graft is between 0. Silylated starch with a degree of substitution (DS) between 0.21 and 0. grafting. Experiments with native starch under similar conditions do not result in the desired poly‐(ε)‐caprolactone grafted corn starch co‐polymers and unreacted starch was recovered after work‐up. 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.45‐0. The poly‐(ε)‐caprolactone side chains consist of 40‐55 monomer units and is a function of the reagent intakes.72.7 was obtained. ε‐ Caprolactone is grafted to silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 in THF at 50oC. biodegradable polymers.
which are known to degrade with difficulty and cause serious environmental problems . the used monomers and the corresponding grafted chains are not easily biodegradable. The amylose‐amylopectin ratio in native as well as modified starches has a strong impact on the product properties. These plastics are mainly synthetic polymers from fossil resources. Global production of starch is 60 million ton per year in 2004 . roots) as granules or cells with typical particle sizes between 1‐100 µm. it is water sensitive. and this value increases with about 10% per year .4‐ glucosidic bonds). for instance grafting of monomers (like styrene and methyl methacrylate) to the starch backbone [6. Blends of thermoplastic starch and PCL are not fully miscible. Various modification strategies have been explored.g. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decreased rapidly in time and were fully degraded after 4 weeks . polyvalerolactone. resulting 48 . The polymeric structure of starch consists of repeating anhydroglucose units. Starch is present in the body of many plants (tubers. Polycaprolactone (PCL) is easily degraded by micro‐organisms . There are two types of biopolymer in starch. is one of the potential candidates for future biodegradable polymer products. and polybutyrolactone).1. However.4‐glucosidic bonds) and amylopectin (a branched polymer with α‐D‐1. 245 million tons of plastics are produced per year. The content of amylose in starches depends on the plant and typically varies between 18‐28%. However. the future generation of packaging materials is highly desirable. Well known examples are polyesters derived from cyclic lactones (polycaprolactone. The development of green biodegradable polymers for e. However. Starch has also been thermoplasticized with the help of plasticizers such as glycerol and other polyalcohols . in almost all cases. the product properties are in most cases still not up to standards and blending with other polymers is required . 7]. Introduction Worldwide. Starch. a natural biopolymer. as starch is highly hydrophilic. A wide variety of synthetic biodegradable polymers have been prepared. 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. Chapter 3 3. Starch is abundantly available. and the mechanical properties of starch‐based films are generally inferior to those derived from synthetic polymers [4. Starch films are known to have good oxygen barrier properties. 5]. Starch modification is therefore needed to meet the product properties in a number of application areas. amylose (a linear polymer of anhydroglucoses with α‐D‐1.6‐glucosidic bonds besides α‐D‐1.
Acros) and methanol (Labscan) were used as received. the ring opening polymerisation occurs solely in the liquid phase and this is expected to lead to higher GE values. Hydrochloric acid (HCl) 1 N was prepared from Titrisol concentrated hydrochloric acid solution (Merck) and distilled (Milipore) water. To increase the miscibility of starch and polycaprolactone. A stock solution was prepared by dissolving 1. All available data indicate that the presence of water reduces the GE. during the reaction. Materials and Methods 3.1. Dry tetrahydrofuran (THF) and toluene for polymerization experiments were obtained in closed vessels from Aldrich and were used as received. This approach has also been applied successfully to graft PCL and polylactide on dextran [19‐20]. In this way.2. an alternative method to synthesize poly‐(ε)‐caprolactone grafted starch co‐polymers (starch‐g‐PCL) is reported. ε‐Caprolactone monomer (Fluka) was dried over Calcium Hydride (CaH2) for 48 h.2.67 gram (8 mmol) Al(OiPr)3 in 50 ml of dry toluene in a glove box. 49 . This is expected to lead to reduced reaction rates between starch and CL compared to CL homopolymerisation. a very flammable by‐product. In this chapter. Aluminium triisopropoxide [Al(OiPr)3] (Acros) was used without further purification. 3. Another possible cause for the low grafting efficiencies is the heterogenous nature of the reaction. Materials Corn starch (Sigma) was dried at high vacuum (~1 mbar) at 100 oC for one day before use. it has been proposed to chemically graft caprolactone onto the hydroxyl groups of starch using ring‐opening polymerisation . therefore difficult to handle and releases ethane. distilled under reduced pressure at 100 oC and stored under a protective nitrogen atmosphere. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors in undesirable phase separation . DMSO (Acros) and toluene (Labscan) were dried overnight over molecular sieves 3 Å (Merck) and stored under a protective nitrogen atmosphere. Hexamethyldisilazane (HMDS. leading to a liquid‐solid system. This is rationalised by assuming that water competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. 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]. This catalyst is extremely air and water sensitive. 0‐14%). thus to a reduction in the GE. thus leading to the formation of PCL homopolymers rather that starch‐g‐PCL . Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). The highest GE (90%) was achieved using triethylaluminium as catalyst . Common Ring Opening Polymerization (ROP) catalysts such as tin octoate or aluminium isopropoxide gave low grafting efficiencies (GE.
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. starch).5 ml) and a predetermined amount of the stock solution of Al(OiPr)3 in toluene were added. broad peaks. Chapter 3 3.5 (m. after peracetylation. before peracetylation.2. Subsequently. The reaction was carried out at 70 oC. 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.60): 1 H‐NMR (CDCl3. silyl‐CH3). For each experiment. the silylated starch (1) product was dried in a vacuum oven (~5 mbar.7‐2. Typical example of in situ polymerization of ε‐CL with silylated starch The silylated product from the first step was dissolved in THF (0. Silylated Starch (1. broad peaks. 50 oC): δ 0.111 mol) was added to the gelatinized mixture to initiate the silylation reaction. The mixture was stirred at 50 oC for 4 h to promote the exchange reaction between the isopropoxide groups of Al(OiPr)3 and the free ‐OH groups of starch. A mol ratio of ‐OH groups to catalyst of 10:1 was used.2.6 mL) at 50 oC (1‐2 h). THF (4. 80 oC). acetate‐CH3). Sample SN‐3. 1 3. After 24 h. toluene (40 ml) was added to solubilize the precipitated (partially silylated) starch.60): H‐NMR (CDCl3. 0. 3. Sample SN‐3. 50 oC): δ 0. DS = 0. starch). 1. The intake of 1 depended on its DS and was adjusted to obtain a solution with 5 x 10‐5 mol free‐OH groups/ml of THF.12 (m.1 mbar. After 2 and 4 h reaction time.1.2. 3‐6 ppm (m.2. 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. 3‐6 ppm (m. a predetermined amount of ε‐caprolactone monomer (molar ratio of monomer to 50 . The samples were characterized by 1H‐NMR. 40 oC) until constant weight.2. The pre‐determined amount of HMDS (typically 24 ml. To this homogenous solution. silyl‐CH3). Silylated Starch (1. This procedure was repeated three times. another portion of toluene (20 ml) was added. Traces of DMSO trapped in the product were removed by dissolving the product in a small amount of toluene and re‐precipitation in methanol. the solvents were removed from the silylated starch product under reduced pressure (~ 20 mbar) at 70 oC.2. After removal of the solvents under reduced pressure (0. DS = 0. The white‐to‐transparent solid products were stored under vacuum in a desiccator at 6‐8 oC. After 6 h.
‐CH3.5. and vacuum dried at 40 oC for 24 h. Peracetylation of the remaining OH groups of modified starch is a well established procedure to improve characterisation of the products by NMR . 1.25 (t. 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). broad peaks.25 (t. ε’‐PCL).5.5‐3. The silylated starch‐g‐PCL (2) product was precipitated from the solution by the addition of heptane (about 250 ml) at ‐18 o C. 60 oC): δ 1.12 (s.75. 25].34): H‐NMR (DMSO. 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. iPr).21): H‐NMR (DMSO d‐6. iPr). β and δ‐PCL). 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. Silylated Starch‐g‐PCL (2. 3. DSPCL=0. ‐CH.16 (d. 1 3. 2.2. iPr). DSPCL=0. The samples were characterized by 1H‐NMR. the desilylated starch‐g‐polycaprolactone product (3) was precipitated using heptane.37 (t. 4. and 5‐5. 1. starch). The peracetylation procedure applied in this study was adapted from the literature [24. ‐CH. the peracetylating reagents 51 .68.31 (m.4 (m. 2. Subsequently. 1. 4.2. 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. ε‐PCL). The product was collected as a white solid and characterised by 1H‐NMR. Deprotection product of Sample SN1CL2. The total isolated yield at this condition (see Table 3. 3. 3. iPr). ε‐PCL).1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until it was fully dissolved (typically 3 h). γ‐ PCL).37 (t.2. The yield was measured gravimetrically and is based on the weight of the product and the total weight of reactants charged to the reactor.4 (m. α‐PCL).54 (m. filtrated. DSsilylation = 0. Starch‐g‐PCL (3. 1 3. 3. γ‐PCL).5‐ 3. starch). 3. 3. broad peaks.2.2. 4.68. Sample SN1CL1. 60 oC): δ 0.75.98 (t.98 (t. α‐PCL). 4. After stirring for 2 h. The solid precipitate was filtered and dried under vacuum (~ 5 mbar) at 40 oC for 48 h.88 ppm (m.31 (m.54 (m. β and δ‐ PCL).88 ppm (m. Typically. The reaction was stopped by cooling down the mixture to room temperature and the addition of 2‐3 drops of 1 N HCl to deactivate the catalyst. 1. ε’‐PCL).3‐4. 1 (0.4. silyl‐CH3).16 (d. 1.) was >99 %. ‐CH3 .3. and 5‐5. DSsilylation = 0.3‐4.
3. 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.1.4. The product was precipitated by the addition of methanol and washed several times with methanol. The assumption of high conversions (>95%) was correct for all experiments (see Table 3.8 ppm (3.6−0. Analytical Methods 3.) The degree of substitution of the PCL graft on 2 (DSPCL) is defined as the average number of PCL polymer chains present on an AHG unit of starch.6 ppm DSsilylation = × = × 3 Astarch proton / 7 27 A3−5. 3. 3. 1 3. 3. Chapter 3 (DMAP.2.2 ppm A3.).3. This leads to the following equation: ACL = ACH 2 −ε + ACH 2 −ε ' ACH 2 −ε ' = A3.1.3. the DSPCL may be calculated using eq. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.1. The ACL is calculated from 1H‐NMR spectra by comparing the peak area of protons attached to ε‐carbon atoms in a repeating CL unit with that of the characteristic ε’ protons of the last CL unit in a PCL chain  (see Figure 3. 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.4 ppm + 1 (3.) 52 .2. When assuming that all ε‐CL monomer is converted.2. The Average Chain Length (ACL) of the Poly‐(ε)‐caprolactone chain is defined as the average number of CL repeating units in a grafted polymer chain. it is assumed that the average length of the grafted chain is equal to the chain length of the homopolymer. DSsilylation may be calculated using 1H‐NMR spectra of the products after peracetylation using eq.) where Ax‐y stands for the 1H‐NMR peak area in the range δ x‐y ppm.2. It was finally dried overnight in a vacuum oven at 70 oC and 5 mbar until constant weight.3. ACH 3 silyl / 9 1 7 A−0.3−3.2.8− 4. In this calculation. The peracetylation reaction was conducted for 7 h at 50 oC.
) ⎛ ⎞ ⎞ ⎛ A A = ⎜1 − CH −iso ⎟ × 100 % = ⎜1 − 4. 53 . 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. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors DS PCL = 0. T stands for absolute temperature.4 + A3.8− 4.( A3. the focus of this chapter will be on the first two steps of the procedure.8− 4 .5. The values of ∆Hv and Vm were obtained from the SciFinder Scholar database (American Chemical Society.3.4 − A4 .2 ) ⎤ 1 × ⎢ (3 − DSsilylation )⎥ ⎢ ε .) 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).3−3. Although all steps have been investigated.9 ⎡ 0. This leads to the following equation: GE = ACH 2 −ε ' / 2 − ACH −iso ACH 2 −ε ' / 2 × 100 % (3. 2007) 3.3.9 ⎟ × 100 % ⎜ ⎜ ⎟ ACH 2 −ε ' / 2 ⎠ A3. A3.4. and Vm stands for molar volume. 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.CL/OH ratio ⎥ ⎣ ⎦ (3. 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.3−3.8− 4 .4 / 2 ⎟ ⎠ ⎝ ⎝ where Ax‐y stands for the peak area in the range δ x‐y ppm. 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).3−3.5.
In this way.60 was obtained. mixtures of toluene and DMSO were applied to avoid precipitation of the silylated starch during the reaction. The silylated products were characterised by NMR. Using standard conditions (HMDS: AHG molar ratio of 3.) Instead of using DMSO as solvent. as suggested by Einfeldt et al .) shows the presence of silyl groups at about δ 0 ppm. 54 .). 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 3.1.a.a. The silylation procedure was adapted from that previously reported for dextran [19‐23].). However. It is defined as the average number of silyl substituted OH groups on the anhydroglucose (AHG) unit of starch. a product with a DS of 0. An additional peracetylation procedure to substitute the free hydroxyl groups with acetate groups was performed to improve the quality of the NMR spectra.1. Very broad peaks of starch protons at δ 3‐5 ppm and a sharp peak of the methyl substituents of the silyl group at about δ 0 ppm were observed (see Figure 3. a homogeneous reaction mixture was maintained throughout the reaction. 70 oC. the starch peaks are very broad and this feature hampers accurate determination of the DSsilylation. The degree of substitution of the silyl groups (DSsilylation) was determined by 1H‐ NMR.1.). Synthesis of Silylated Starch The silylation of corn starch was performed with HMDS as the silylating agent (eq.3. Chapter 3 3. 126.96.36.199.b. The 1H‐ NMR spectrum of silylated starch (Figure 3. C O H 2OR H 1 + NH3 2 OH H3C [R = H or Si(CH3)3] Starch Hexamethyldisilazane Silylated Starch (1) (3. 24 h).
and Figure 3. 55 .5‐4. in CDCl3. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Figure 3. The results are given in Table 3. not peracetylated (b). Typical 1H‐NMR Spectra of Silylated Starch (Sample SN188.8.131.52) on the product DS was investigated by performing experiments with a constant starch and a variable HMDS intake. peracetylated The effect of the HMDS to starch molar ratio (1. DSsilylation=0. 50 oC): (a).60.1.
6 DSsilylation 0.5 2.5 Figure 3. 70 oC) Table 3.25 cal1/2cm‐3/2) due to the presence of the bulky apolar SiMe3 groups.55 0. HMDS is a rather apolar compound (Hildebrand solubility parameter of 6.5:1. The experimentally observed lowering at higher HMDS intakes is likely due to a decrease in the polarity of the reaction mixture.5 2 2. For all experiments.2.65 0. a positive effect of higher reagent intake on the reaction rate and thus the DS is expected.5 HMDS : AHG Ratio [mol/ mol] 4 4.5 Product DS (DSsilylation) 0. At a mol ratio of HMDS to AHG of 4. Surprisingly. the DSsilylation decreases for higher intakes of HMDS.4 1. solubility parameter = 11.67 0.5 3 3.25 3 4. Chapter 3 0.45 0. DS of the silylated products at different HMDS: AHG ratios (DMSO.1.5 0. the volumetric HMDS intake is about half of the DMSO intake. Its presence will reduce the polarity of the reaction medium (DMSO. mmol) 12 (56) 18 (83) 24 (111) 36 (167) HMDS: AHG mol ratio 1. Effect of HMDS: AHG mol ratio on DS of the silylation product a HMDS intake (ml.60 0.36 cal1/2cm‐3/2) considerably.68 0. The silylation reaction 56 . For reactions with an order higher than zero.46 Experiment SN1 SN2 SN3 SN4 a Experiments were performed in DMSO at 70°C. a fixed starch intake of 6 g (37 mmol AHG) starch was applied.
Petzold et al  reported that silylation of starch with trimethylsilylchloride (TMSCl) in pyridine yielded trimethyl‐silyl substituted starch with DS values between 0. The use of HMDS to silylate dextran (Mw=6000‐40000) in DMSO (HMDS to OH molar ratio of 0.7) of the products is in the range of those published for other starch silylation systems.6.6‐glucosidic for dextran). sample SN‐1) in THF at 50 oC for 24 h using Al(OiPr)3 as catalyst. DMA/LiCl. A schematic representation of the reaction is provided in eq.2.3‐2.0 . white solid products with isolated yield > 96% were obtained.1 and 3. 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. 29]. 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. DMF. 3. Silylation of starch with HMDS in formamide.6. In situ Ring Opening Polymerization of ε‐Caprolactone with Silylated Starch A number of in situ ROP experiments with ε‐caprolactone (CL) were carried out using a typical silylated starch sample (DS=0. pyridine.25‐5.3. The products are soluble in DMSO as well as in less‐polar solvent such as chloroform and THF. Similar reductions in the reaction rates when working at higher reactant intakes were observed for the esterification of starch with vinyl laurate and stearate .) 57 .2. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors involves charged‐ionic species  and a reduction in the polarity of the reaction medium is expected to lead to a lowering of the silylation reaction rates. To the best of our knowledge. and DMSO/pyridine mixtures yielded silylated starch with DS values ranging between 0. The DSsilylation (0.68.7‐3.45‐0.0 [18. liquid ammonia. After precipitation with heptane and vacuum drying. 3.0) gave silylated dextran with DS values between 1. O O H2C O R H H OR H O O H OR H O Catalyst: Al(OiPr)3 Toluene. the silylation of starch in pure DMSO using HMDS as the reagent has not been reported to date.
2 ppm. 58 . Figure 3. The formation of CL homopolymers for this type of reactions has been observed before [15.2‐4 ppm and imply that the ring‐opening polymerisation reaction of CL indeed occurred.4‐3.3. The peaks from the polycaprolactone units are clearly present in the range of δ 1. DSsilylation =0.4. Typical 1H‐NMR spectrum of a silylated starch‐g‐PCL sample. 26]. Further process optimization studies allowed the synthesis of products with less than 42% of homopolymers (vide infra). However.0‐5. (Sample SN1CL1. DSPCL=0. not all of the caprolactone is grafted to starch and large amounts of PCL homopolymers (72%) were formed.4 ppm. Apparently.8 and 5.68.21) in DMSO‐d6 at 60 oC. The homopolymers are formed by direct polymerization of ε‐CL initiated on isoproproxide moieties attached to the Al catalyst. This is clearly indicated by resonances of the iPr end‐group of the PCL homopolymer at δ 4. Coding of the peaks is given in Figure 3.4. broad peaks in the region δ 3.3. Chapter 3 The products were characterized using 1H‐NMR analysis in DMSO‐d‐6 as the solvent.9 and 1. as is shown in Figure 3. Resonances from the starch peaks are observable as small. A typical spectrum is shown in Figure 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.
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 (3.7.). 1. 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. Product formation for the reaction between silylated starch and ε‐ caprolactone Besides the ‐OH group of starch. However. 3.) 59 . peaks arising from a ‐COOH unit could not be detected in 13C‐ NMR spectra (δ 175‐180 ppm). This leads to the formation of carboxylic end groups (see eq. Exchange Reaction: iPr O Al O iPr iPr O + H2O iPr O iPr O Al OH + iPr OH 2.7. Polymerization Reaction: iPr O Al O iPr H2C O O H O H OR H H OR H O O O O α β χ δ ε α' β' χ' δ' ε' H2C [O C CH2 CH2 CH2 CH2 CH2 ]n O C CH2 CH2 CH2 CH2 CH2 OH O O H O H OR H H OR H O [R = H or Si(CH3)3 ] + n+1 Silylated Starch-g-PCL (2) O iPr O Al O iPr iPr O + O n+1 iPr O α β χ δ ε O α' β' χ' δ' ε' [O C CH2 CH2 CH2 CH2 CH2 ]n O C CH2 CH2 CH2 CH2 CH2 OH PCL homopolymer Figure 3. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors 1.4. residual water may also initiate the polymerisation reaction.
2. Chain Length [mon. Chapter 3 The ratio of homopolymerisation versus grafting on starch may be obtained by comparing the integrals of selected peaks in 1H‐NMR spectra. (SN1CL1) is 28%.68) in THF at 50 oC with Al(OiPr)3 as the catalyst (1 mol Al(OiPr)3 per 10 mol‐starch‐OH groups). The GE increases with higher ε‐CL intakes.2 Total Yieldb [%] >99 >99 >99 99 96.5. Table 3.9 22. The results are given in Table 3. Experiment SN1CL1 SN1CL2 SN1CL3 SN1CL4 SN1CL5 a.5 and 100%. The ACL of the polymer and the DSPCL increase almost linearly with the ε‐ caprolactone intake (Figure 3.47 0.5. Five experiments were performed to study the effect of different ε‐CL to silylated starch ratio.58 0. However.9 ppm).2. This implies that grafting of caprolactone to starch also occurs to a significant extent. the intensity of the peak from the ‐CH2‐ end group of the homopolymer (ε’ at δ 3. and Figure 3. In the case of only homopolymerisation.0 18. Determined gravimetrically and defined as the total weight of the isolated product divided by the total intake of reactants (silylated starch and CL). Overview of results for the grafting of ε‐CL on silylated starch a ε‐CL/ OH [mol/ mol] 13. 60 .).0 15.2.21 0. the intensity of the resonance ε’ was considerably higher that that of the ‐CH‐ iPr peak. units] 40 43 44 49 54 DSPCL 0. and reaches 58% for a ε‐CL to starch–OH ratio of 29.3 ppm) should be twice the intensity of the ‐CH‐ proton of the isopropoxide end group (δ 4. b.5 Avg. The grafting efficiency (GE) for the sample given in Figure 3.34 0.3.72 Grafting Efficiency [%] 28 43 48 55 58 All reactions were performed using the same intake of SN1 silylated starch (DS=0. This indicates that higher ε‐CL concentrations during the reaction lead to longer PCL grafts as well as to higher levels of initiation of the grafting reaction on free hydroxyl group of silylated starch. in all samples. The yield of the products was measured gravimetrically and varies between 96.5 29. This implies that the ε‐CL conversion is essentially quantitative in all cases.
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)
5.2. 1.72. Considerable amounts of ε‐CL hompolymers with isopropyl end‐groups were also formed. the highest value was obtained with a ε‐CL‐AHG ratio of 29.4. Silylated starch with a low‐medium DS (0. 3. Control ROP experiments of ε‐CL with native starch under similar conditions did not produce the desired poly‐(ε)‐caprolactone grafted corn starch co‐polymers. The ROP with ε‐CL was performed using Al(OiPr)3 as catalyst in THF as the solvent. This was achieved by making the starch more hydrophobic by partial substitution of the OH groups by trimethylsilyl groups. Chapter 3 3. The products may have interesting applications as compatibilizers for starch‐polymer blends. The key feature is the use of a homogeneous reaction mixture for the ROP of starch with ε‐CL. depending on the ε‐CL‐starch ratio.68) was obtained using a DMSO/toluene mixture as the solvent and HMDS as the silylating agent. indicating that homogeneous reaction conditions are favorable for the grafting reaction. The grafting efficiency varied between 28‐58%.986 cal mol‐1 K‐1 : temperature [K] : R T Vm molar volume [cm3/ mol] 64 . 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. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] A x − y ppm ACL DS DSsilylation : DS of silyl group substituents [‐] DSPCL GE : DS of PCL chain substituents [‐] : Grafting Efficiency. The DS of the PCL chains was between 0.21‐0. the percentage of PCL grafted to starch compared to the total amount (grafted and homopolymer) of polymerized CL [%] : gas constant. Conclusions The successful synthesis of poly‐(ε)‐caprolactone grafted corn starch co‐ polymers using a three step approach is reported.46‐0. 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.
Goldberg: Review of the biodegradability and utility of poly(caprolactone). R. 1995. Beliakova. Polym. 2003. Denmark. J. .html. Wang.. accessed on August 01. S. Linke: Water vapor sorption determination of starch based porous packaging materials. Wang: Properties of starch blends with biodegradable polymers. Weber). 1997. Brussels. 311‐316. 8. . .A. Starch‐Starke 2000. Russell: Graft copolymers of polysaccharides with thermoplastic polymers ‐ new type of filled plastic.C.K. 61‐74. . Plastics Europe. Macromol. . . J. 52. Babac: Preparation and biodegradation starch/polycaprolactone films. 2003. 65 . Denmark: http://www. A. . Sci. Mater.dk/isi/stat/rawmaterial. Polym. J. Frederiksberg C.A.M. 1251‐1253. W. Environmen. Tsiapouris. R. Adv. C. Biobased packaging materials for the food industry. R.starch.M. 385–409 .A. References .R. 3. K.Sci. C. Degrad. J. 2000. 23.6. 1998. H. Burr. . Prog. E. Doane. Polym. C. Karlsson. demand and recovery for 2006 in Europe. Aly. . Sci. 305‐313. Yang. 1977. of . R. Sci. Environ. Chandra. Food Technol. D. Part C: Polym. Yavuz.. 407‐412. . Starch‐Starke 2004. Eng. 11. M. Polym. 56. report of Food Biopack Project (Ed. The compelling facts about plastics: An analysis of plastics production. Chiellini. Abdel‐Mohdy: Grafting of poly(methacrylic acid) on starch and poly(vinyl alcohol). A. the Royal Veterinary and Agricultural University. 38. status and perspectives. Rev. De Mulder‐Johnston: Edible and biodegradable polymer films: Challenges and opportunities. C. 107‐113. 51. Aarhus. Department of Dairy and Food Science. F.F. G. Solaro: Biodegradable polymeric materials. Polym. Fanta. X. L. E. 17. 2008. 43. Y. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Greek symbols: ∆H v : heat of vaporization [kJ/mol] : δ Hildebrand solubility parameter [cal1/2cm‐3/2] 3.C. 2008. 1996. J.B. A. Albertsson: Biodegradable polymers and environmental interaction. Website of International Starch Institute. Belgium. Eng. Rustgi: Biodegradable polymers. 53‐57. 1273‐1335. 1998. Bagley. Krochta. Association of Plastics Manufacturers. 61‐67.
I. Six: Silylation reaction of dextran: Effect of experimental conditions on silylation yield. Macromol. P. structure analysis. 251‐269. J. Polym. L. Part A: Polym. Ydens. Six. and chemical stability of silylated dextrans. P. . Six: Controlled synthesis of amphiphilic biodegradable polylactide‐grafted dextran copolymers. T. P.L. Dubois. Klemm. 561‐566. 41. K. W. 1443‐1450. Des. Fringant: Properties of thermoplastics blends: starch‐polycaprolactone. Dubois. Klemm. L. E. Ydens. C. E. high value applications. Moro. 2577‐2588. B. Klemm. 42. D. Koschella. P. Macromolecules 2000. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. Nouvel. D. K. P. 35‐47. Hamaide. Stab. K. M. Polymer 2000. 43. Petzold. Klemm: Regioselective functionalization of starch: Synthesis and 1H NMR characterization of 6‐O‐ silyl ethers. I. Polym. Stein. P. H. 1. Macromol. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer 66 . Dubois: Aliphatic polyester‐ based biodegradable materials: new amphiphilic graft copolymers. D. 3091‐3100. M. Averous. . C. Duquesne. P. Petzold. D. J. 5. P. L. and subsequent reactions. P. Dole. Dellacherie. Dellacherie. . Klemm: Preparative and 1H NMR investigation on regioselective silylation of starch dissolved in Dimethyl Sulfoxide. A.L. . Six: Partial or total silylation of dextran with hexamethyldisilazane. W. M. I.L. . Einfeldt: Structure design of polysaccharides: novel concepts. Biomacromolecules 2001. E. 6713‐ 6721. Polymer 2002. Degee. R. 33. Einfeldt. K. Biomacromolecules 2003. Monomers Polym. . E. Stein. Dellacherie. Biosci. D. Krishnan. Rutot. J. 10. Gunther. Petzold. 2001. 40. Dubois. Nouvel. J. 2. Petzold. regioselectivity. Chapter 3 . 163. . Sci. L. . Ydens. 1735‐ 1743. Degee. Symp. A. Rutot. Stein. . . 415‐426. Chem. . W. P. Kussler. Fessi. . Degee. D. D. 2001. Dubois. Pantiru. Nouvel. 4. 2002. D. Polymer 1999. A. 341‐347. selective syntheses.. L. Gunther. P. C. 965‐969. 2001. J.L. Cellulose 2003. 2004. Heublein: Silylation of cellulose and starch ‐ selectivity. Dubois: Controlled synthesis of poly(epsilon‐caprolactone)‐grafted dextran copolymers as potential environmentally friendly surfactants. Einfeldt. 73. Dellacherie. E. Gunther: Synthesis and NMR characterization of regiocontrolled starch alkyl ethers. 4157‐4167. Degrad. A.
Penczek. A novel route to functionalised nanoparticles.M. Rapid Commun. 941‐950. 2001.J.B. F. S. . J. A. Kricheldorf. Carbohydr.4‐ dioxan‐2‐one. L. Hertz Jr. T. Sugih. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors agents. Manurung. 553‐556. D. Macromolecules 1996. Biela. Duda: Living polymerization with reversible chain transfer and reversible deactivation: the case of cyclic esters. Res. 659‐663. . H. Junistia.L. 1998. . Damrau: Zn‐lactate catalyzed polymerizations of 1.R. D. Kinetics and mechanism. 2000. . Macromolecul. H. Macromol. 61. Rapid Commun. R. A. 1399‐1406. 21. presented at ACS Rubber Division meeting. 199. Lehmann: Selective 6‐O‐acetylation of amylose. Picchioni.P. Mexico City. D. 1989 . L.: Solubility parameter concepts – a new look. Horton. 2008. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. 22. Macromol Chem. 1978. 29. A.K. 1089‐1097. Janssen. Phys.O. 67 . accepted for publication in Starch‐Starke. Duda: Polymerization of ε‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. .
The yellowish products were characterized by 1 H‐. and Na‐acetate) in DMSO at 110°C. vinyl ester to starch ratio and the type of catalyst. the DS for the starch‐laurate compounds is higher than for the corresponding starch‐ stearates. 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. Keywords: corn starch. the DS decreases.96). presumably due to a reduction of the polarity of the reaction medium. and Na‐acetate are superior catalysts with respect to activity compared to Na2HPO4 and products with DS values close to 3 were obtained. vinyl laurate. When performing the reactions using Na2HPO4 as the catalyst. The DS of the products is a function of the carbon number of the fatty acid chain. K2CO3. K2CO3.24 ‐ 2. At higher ratios. an increase in the vinyl‐ester concentration leads to higher product DS values. vinyl stearate . 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. For low vinyl‐ester to starch ratios.
5]. and incompatibility with hydrophobic polymers. cellulose or proteins are potentially very interesting starting materials for biodegradable packaging materials. In particular starch is attractive as it is relatively cheap and abundantly available.34‐0. Relatively low‐DS (0. high brittleness. The hydrophobicity increases with the degree of acetate substitution (DS. more water‐resistant products may be obtained [3‐4].Chapter 4 4. 5‐6] were introduced using fatty acid chloride reagents. 7]. Recently Mormann et al  explored the possibility of using vinyl esters and particularly vinylacetate as reagents for the preparation of starch esters. the fatty ester substituents [1. the use of native starch for packaging materials is limited due to its low moisture resistance. However. The major obstacle is the pronounced brittleness of the materials. 5]. and even fatty acid derivatives (C12‐C18) have been used in the modification reaction [5. Natural polymers such as starch. The mechanical properties and hydrophobicity of the products were significantly improved when using these longer chain fatty acid precursors [1. Their 70 . An attractive field of application for these polymers is the use as packaging materials. Further modification of starch is therefore required to introduce hydrophobicity and to improve mechanical and moisture barrier properties. The first paper on the acetylation of starch was already published in 1865 .1. defined as the moles of substituents per mole of anhydroglucose (AHG) units) . The introduction of acetate groups on starch makes the product more hydrophobic. An alternative method using methyl and glyceryl laurate esters in the absence of solvent has been recently developed . and consequently. most of the studies performed to date use short chain carboxylic acids (C1‐C4). the mechanical properties of high‐ DS starch derivatives of low chain carboxylic acids still need considerable improvements before large scale application as packaging materials becomes within reach. To improve the mechanical properties. even after the addition of plasticizers . For the current petrochemical based products recycling is often neither practical nor economically feasible . which are relatively expensive and rather corrosive .7 [1. However. Esterification of starch with low molecular weight fatty acid derivatives is one of the oldest modification technologies to improve starch properties. However. However. resulting in products with DS values up to 2. higher molecular weight carboxylic starch esters (C4‐C6) . and particularly acetic acid derivatives (C2) [2‐4].61) products were obtained using this approach. poor processibility (high viscosity). Introduction Green biodegradable polymers derived from natural resources are potentially very interesting substitutes for non‐biodegradable petroleum‐based polymers.
The spectra were recorded at 50 oC. 4.2. Analytical grade vinyl stearate (Aldrich. not been reported to date. In DMSO. Germany) were used without further purification. In addition. IR spectra were recorded on a Spectrum 2000 FT‐IR Spectrometer (Perkin Elmer). The starch was dried before use for 48 h at 105 oC under vacuum (approx. Materials and Methods 4.2. leading to higher reaction rates. The synthesis of starch stearate esters using vinyl ester reagents has. 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.2. This solvent effect is likely caused by the higher solubility of the vinyl esters in DMSO compared to water. to the best of our knowledge. the use of basic catalysts other than Na2HPO4 has been investigated. The reactions were either performed in water or in DMSO using a basic catalyst (Na2HPO4). vinyl laurate (Fluka. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants research focused on the synthesis of starch acetates and only two examples with a higher molecular weight fatty acid vinyl ester were reported. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine.01 when using vinyl‐laurate. The effects of the starch to vinylester ratio on the reaction rates and DS have been explored. The maximum attainable DS of starch acetate in water was below 1 and limited to 0. and tetrahydrofuran (THF) were supplied by Acros (Belgium) and were also used as received.2. Materials Corn starch (approx.1. 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. Potassium carbonate (Boom. Germany) were used as received.6 for starch acetate ester) were obtained. 4‐N.N‐dimethylaminopyridine (DMAP). Japan). The products were placed directly on the diamond plate and 50 scans with a resolution of 4 cm‐1 were recorded. Germany) and disodium hydrogenphosphate (Merck. as recommended by Laignel et al . 4. Technical grade dimethyl sulfoxide (DMSO). Germany) and acetic anhydride (Merck. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). 1 71 . ~1 mbar). leading to a moisture content of 2 %‐wt (measured gravimetrically). the Netherlands). sodium acetate (Merck. starch esters with a substantially higher DS value (up to 1.
2H..3 (m.6 (m. 1. 1. C2’). 3‐6 ppm (m. broad peaks. 3H. C6S).1 (m.2.3. 34. 935 (C‐O stretching). 76‐78 ppm. 3H.. 3‐6 ppm (m. 28‐32 (C4‐9). C2). 2H. 760. the product was dried in a vacuum oven (70 oC. broad peaks. 1740 (C=O). 1230 (C‐O stretching). 28H.0 (C12). 68‐74 (broad. 3S. 720.5 (m. 1410 (C‐H bending). 1. 24.Chapter 4 4.3 (m. Subsequently. 95. O‐C3S. C3).5 (m. Table 4.9 (t.1. broad peaks. approximately 5 mbar) for 24 h until constant weight. 2H. Starch‐laurate (Sample 17.1. C1S‐6S). (13C‐NMR) and Figure 4. Starch‐stearate (Sample 19. 22. 1020 (C‐O stretching). 2H. 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. 1. CDCl3): δ 0. C12).2.5 g) was first gelatinized in DMSO (5 ml) at 70 oC for 3 h. 1370 (C‐H bending). 2.9 (t. 2. 1150 (C‐O stretching). Figure 4. 1. CDCl3): δ 0. Methods 4. The atom numbering scheme is given in Figure 4. C4‐11). and O‐C6S). C2). respectively).0 (m. 2. The samples were characterized by 1H‐ and 13C‐NMR and FT‐IR.52): H‐NMR (before peracetylation. 7H. After cooling.1 (C11). broad peaks.. DS = 2.4 (m. C2). attached to O‐C2S. 2850 (C‐H stretching). 2H.3 (m. broad peaks. 1350 (C‐H bending). 13 FT‐IR (cm‐1): 2920 (C‐H stretching). 5S). 61. 1. Table 4.9 (t.9 (C10). C4‐C17).5 (m. C3). C1S).4 (broad.2. overlap with CDCl3 (C4S).3. (1H‐NMR). 1295.3. 31. The product was washed twice with methanol (50 and 25 ml. C1S‐6S). 16H. 1. 3H. the product was precipitated using methanol (100 ml) and separated from the liquid phase by decantation. 16H.96) H‐NMR (before peracetylation. 7H.7 (C11).9 (broad. broad peaks. 3H.1.1. broad peaks. C1S‐6S) 1 72 . 1 C‐NMR (before peracetylation. DS = 2. 12). 1455 (CH2). 1 H‐NMR (after peracetylation. 3‐6 ppm (m. Finally. C4‐11). Typical example of the synthesis of laurate and stearate esters of corn starch Corn starch (0. 2H. CDCl3): δ 14.9 (C3). 1110 (C‐O stretching). 172‐174 ppm (C=O. C2S. CDCl3): δ 0. 7H.8‐2. (FT‐IR). broad peaks. typical spectra are given in Figure 4. resulting in the formation of a homogenous transparent solution. C18).4. C3).
3S. 1350 (C‐H bending). 760. C18). 2H. 7H. 1. broad peaks. 95. C1S‐6S) 1 C‐NMR (before peracetylation. 720.9 (t.2. Numbering scheme for carbon atoms of products 4. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants H‐NMR (after peracetylation. Peracetylation procedure The presence of remaining hydroxyl groups in the products resulted in broad and overlapping starch resonances in 1H‐NMR spectra  and hampered calculation of the DS. 1.8‐2. and O‐ C6S) 13 FT‐IR (cm‐1): 2920 (C‐H stretching). C3). broad peaks.0 (C3).5 (broad. C2S.2. 26‐32 (C4‐15). 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 4.4 (m. attached to O‐C2S. 3H.3 (m. O‐C3S. C1S). 1740 (C=O). 32. C2). 172‐174 ppm (C=O.3. 1295.1. 1020 (C‐O stretching). 5S). 3H. 34. 2850 (C‐H stretching). 1150 (C‐O stretching). 950 (C‐O stretching). 25.6 (m.5 (m. A peracetylation reaction to substitute all of the remaining 73 . 1100 (C‐O stretching). 1410 (C‐H bending). 865. CDCl3): δ 14. C6S). 2.0 (C16). CDCl3): δ 0. 2H. 75. C2’). 3‐6 ppm (m. 61.2 (C2). 68‐74 (broad. 1. 1455 (CH2).7 (C17). C4‐C17). 22. 28H.7 (C4S). 1370 (C‐H bending).4 (broad.0 (C18).
7.) AH − starch = (4.8‐2.Chapter 4 hydroxyl groups with acetate groups was applied to obtain reliable DS data.6 − AC 2 3 (4.) AC 2 = 2 × AH − fattyester The AC2 values of both the laurate and stearate side chains were calculated from the peak intensity in the range δ 0. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.4.8− 2. Determination of the Degree of Substitution (DS) The DS of the product was calculated using 1H‐NMR spectra of the products after peracetylation.: DSacetate = AH − acetate AH − starch A3.5.5‐1.2.5. the starch ester (0.4.: 74 . the peractylating reagents (DMAP.6.3.) The DS of the acetate groups of the products may be calculated by comparing the unit area of the acetate protons (AH‐acetate) to the unit area of the starch protons (calculated from the intensity of the starch peaks at δ 3.6 ppm range) and a correction has to be made (eq 4. AH − acetate = A1.3. 4. and 4.6−5.3. and 4.).3.2. 4.) (4.8 ppm (protons attached to the fatty acid carbons C3‐C12 for laurate and C3‐C18 for stearate) using eq.1.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until the starch was fully dissolved (typically 3 h).6 ppm). The product was precipitated by the addition of methanol and washed several times with methanol. The peracetylation reaction was conducted for 7 h at 50 oC.1. The peracetylation procedure by Einfeldt et al  was applied.6‐5. The DS of the fatty acid esters was calculated from the DS of the products after peracetylation (eq 4. DS fatty ester = 3 − DS acetate (4.) Determination of the unit area of the acetate H‐atoms (AH‐acetate) is hampered by peak overlap with the H‐atoms attached to C2 (AC2) of the fatty acid chains (δ 1. It was finally dried overnight in a vacuum oven at 70oC and 5 mbar until constant weight. Subsequently. The procedure was described earlier by Elomaa et al  and the relevant equations are shown in equation 4.6 7 (4.. 4.). Typically.2.
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.7. A schematic representation of the esterification reaction of starch with the vinyl esters is provided in Scheme 4.) In eq. 4. 4. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants AH − fattyester = AH − fattyester = AC 3−C12 A0.8 = (laurate) 21 21 AC 3−C18 A0.3..1. Initially. the vinylester and the catalyst were added and the reaction mixture was heated to 110°C. After 2‐3 h. the esterified starch started to separate from the medium in the form of a gel.5−1. OH O HO OH O OH + R O CH2 O HO O R + n H3C H O n O (1) R = laurate (C12) (2) R = stearate (C18) Scheme 4.5−1.1. the starch was gelatinised in DMSO at 70°C for 3 h to make the hydroxyl groups of starch more accessible for reaction. light yellow solid. Ax‐y stands for the peak area in the range δ x‐y ppm. the brownish gel was precipitated with methanol and the product was collected after vacuum drying in the form of a transparent.7. 75 .3. but swell in organic solvents such as toluene and THF.1. After 24 h.) (4.6. 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 4.8 = (stearate) 33 33 (4.3.‐4. Results and Discussion 4. Subsequently. The reaction was performed in two discrete steps. The products of these exploratory reactions are insoluble in water and DMSO.1).
4.2. Higher DS products have a higher solubility in CDCl3 and good quality 1H‐ and 13C‐NMR spectra could be obtained (Figure 4. Typical 1H‐NMR spectrum of: (a) native starch in DMSO d‐6 at 60oC. (b) starch laurate. For atom numbering scheme: see Figure 4. A reaction with vinylstearate at similar conditions resulted in a stearate starch ester with a DS of 2.3. 4. Figure.1.3. and 4.2.) in CDCl3 at 50oC. DS = 2. When using a vinyl‐laurate : AHG molar ratio of 3 and K2CO3 as the catalyst.1. Table 4.52 (Sample 17.96.2. Medium DS starch laurate and starch stearate (1 < DS < 2) dissolve poorly in DMSO‐d6 and CDCl3. Product Characterisation 4.2.1. (c) peracetylated starch laurate.52 was obtained. even at higher temperatures (50 oC).3. 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. 76 . Table 4. DS = 2. a product DS of 2.).) in CDCl3 at 50oC.Chapter 4 The DS of the products was determined by using NMR (vide infra).1.52 (Sample 17.
The starch peaks (δ 3‐5.2.5 ppm).5 ppm) are broad and overlapping . 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. Figure 4. are together with the CH2 groups of the acid chain adjacent to the ester moiety in the range δ 1. The resonances arising from the anhydroglucose unit of starch are broadened.c. This feature hampers the DS determination by NMR. required for DS determinations. 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).3. Two of the carbon resonances (1S and 4S) are considerably shifted compared to native starch. The same 77 . 10‐11].52 (Sample 17) in CDCl3 at 50oC. Clearly visible are the peaks arising from starch and the aliphatic hydrogen atoms of the fatty acid chain (δ 0.2. The proton signals of the acetate methyl group. (b) starch laurate.8‐2.8‐2.1. DS = 2. For atom numbering scheme: see Figure 4. NMR spectra of the peracetylated products are considerably improved in terms of peak resolution and allow a more reliable calculation of the DS.3. and therefore a peracetylation procedure to substitute all of the remaining OH groups with acetate groups was applied [8. in DMSO d‐6 at 60oC.3 ppm.96 (Sample 19) in CDCl3 at 50oC. The 1H‐NMR spectrum of a typical peracetylated starch laurate is shown in Figure 4. (c) starch stearate. DS = 2. Typical 13C‐NMR spectra of the products are given in Figure 4. Typical 13C‐NMR spectra of: (a) native starch.
starch stearate C-H C=O C-O Absorbance b.1.c.4. the C‐H stretching vibrations of the alkyl groups of the fatty ester chain are clearly present at 2920 and 2850 cm‐1. a spectrum of native starch (Figure 4.4. c.) is also included.c. 4.). Sample 19.2. Sample 17.4.) show characteristic bands of the carbonyl group of the fatty esters in the 1750‐1700 cm‐1 region.b. FT‐IR measurements The FT‐IR spectra of starch laurate and starch stearate are shown in Figure 4. native starch -OH C-H 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Figure 4.96.b. 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.2.1.Chapter 4 phenomenon was observed by Dicke for starch acetate . Characteristic peaks of the polysaccharide backbone are visible in the 1250‐900 cm‐1 region (C‐O stretching) .) and native starch FT‐IR spectra of both starch laurate and starch stearate (Figure 4.a.3.4.4. The near absence of remaining hydroxyl 78 . starch laurate C-H C=O C-O C-O a. and 4. and 4. FT‐IR Spectra of starch laurate (DS= 2.4. Table 4. Table 4. For comparison. In addition. starch stearate (DS= 2.52.4.
07 0.68 0.90 1.08 1.13 1. 4. four experiments were performed with two alternative basic catalysts (K2CO3 and Na‐acetate).96 2. 79 .2. 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. 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.23 0. Systematic Studies The effect of important process variables like the vinyl‐ester to AHG ratio.3.1.44 Exp.91 0.52 2. Most of the experiments (14) were performed using Na2HPO4 as the catalyst.57 0.54 2. in line with the NMR data.90 0. The results are shown in Table 4.99 1.1. In addition.05 0.60 1. Table 4. type of catalyst and the effect of the addition of co‐solvents on the product DS was studied in more detail.24 2. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants vibrations in the range 3000‐3600 cm‐1 and at 1640 cm‐1 indicates that the DS of the product is high.01 0.
The results are presented in Figure 4. This behaviour is likely the result of two opposing effects. the ester intake is equal on a weight basis to the DMSO intake. At a ratio of 1:6.5.23 for vinyl‐laurate at an intermediate vinyl‐ester‐AHG ratio of 3.1. 110 oC.5. At low to medium vinyl ester/AHG ratios (0‐3). Higher concentrations of vinyl esters are expected to lead to higher esterification reaction rates. DS of the product as a function of the type of vinyl ester and the vinyl ester‐starch ratio (24 h reaction time.5 0 0 1 2 3 4 Vinyl Ester: AHG Starch Ratio [mol/mol] 5 6 Figure 4. The DS of the products is a clear function of the vinyl laurate and stearate ester intakes (Figure 4. A further increase in the vinyl ester intake leads to a reduction in the DS.Chapter 4 4. This is likely due to a reduction of the polarity of the reaction medium. this positive effect dominates the reaction rate and the DS of the products will therefore increase at higher vinyl ester intakes.3. 2 wt% catalyst intake on starch). 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‐ 3.5 Degree of Substitution (DS) 1 0. A further increase leads to a reduction in the DS.3.5. The highest DS value was 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. 1.
Although toluene may positively effect the reaction by (partly) re‐dissolving starch ester precipitates. with reductions in polarity leading to lower reaction rates. The products have about the same DS value for vinyl‐laurate when using only DMSO and even reduced DS values for vinyl‐stearate (see Table 4. the addition of co‐solvents (toluene/ THF) led to an increase in the DS. Here.7 for lauroyl chloride to 0. 4.6. In this case. Catalysts screening A number of alternative basic catalysts for Na2HPO4.1. When using Na2HPO4 as the catalyst.3. 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 negative effects dominate the reaction performance at higher vinyl ester/AHG ratios and lead to a reduction in the DS values. 4. The results are given in Table 4. Effect of the addition of toluene as a co‐solvent A number of reactions were performed using a co‐solvent. the starch laurate esters display higher DS values than the starch stearates.1. it also results in a dilution of the reaction mixture and a reduction in the polarity.3. 11‐14 in Table 4. K2CO3 and Na‐ acetate were tested. For starch 81 .2.).).3. et al  for the silylation of starch.3. A similar trend in reactivity pattern was observed in this study and the DS decreased from 1. 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.5. and illustrated in Figure 4. The addition of toluene for the esterification of starch with vinyl‐esters surprisingly did not lead to improved DS values. 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. i. Aburto.1. This effect is particularly evident at higher vinyl ester/ AHG ratios (>3) (Figure 4.).8 for stearoyl chloride. A similar procedure was proposed by Nouvel. The latter factors appears to have a strong effect on reaction rates (vide supra).3. 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 . with high carbon numbers leading to a reduction in the DS. Thus. These findings were rationalised by assuming that the co‐solvents increase the solubility of the silylated products. the DS of the product is also a function of the chain length of the fatty acid. leading to enhanced reactivity.e. It is clear that Na‐acetate and K2CO3 are considerably more active than Na2HPO4 and products with a significantly higher DS were obtained.
the two catalysts are equally effective and products with a DS of about 2. For starch stearate.5 were obtained.o.96) compared to Na‐acetate (DS=2. The products were synthesised in DMSO using vinyl‐esters in the presence of basic catalysts (Na2HPO4. gray: Na2HPO4 black: K2CO3 white: Na‐acetate 4. the DS for the laurate ester is lower than the stearate ester. function of the type of catalyst.5 Degree of Substitution 2 1. The DS of the laurate ester when using Na‐acetate is higher than for the stearate ester (Table 4.Chapter 4 laurate esterification. catalyst amount = 2%‐w.6. and FTIR and confirm the presence of chemically bound fatty acid chains.1.5 1 0.6. Conclusions A study on the synthesis of corn starch fatty acid esters with high DS values is reported. The yellow products were characterized by 1H‐ and 13C‐NMR. K2CO3 gave products with a significantly higher DS (2. However. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. Apparently. the statement that the DS for the laurate esters is always higher than for the stearate esters is not generally valid and a. 3 2. and Na‐acetate). and Figure 4. Thus. 110oC. 82 . The DS of the products is a clear function of the chain length of the fatty ester and the type of catalyst.5 0 Laurate Vinyl-Ester Stearate Figure 4. in line with the findings for NaH2PO4. DMSO).4. Comparison of DS values with different catalysts (Vinylester: AHG molar ratio = 3:1. the DS of the product is also tunable by proper catalyst selection.). when using K2CO3 as the catalyst.44). K2CO3.
1‐11. E. Polym. S. C. Alric. E. 65. The DS of the products may also be tuned with the vinyl ester/AHG molar ratio. 54. A. W. . Prinos. S. and P. 1647‐1656. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. 1995. 58. J. J. . products with a DS > 2. the DS of the product increases at higher vinyl ester intakes. 83 . A maximum was observed at a vinyl ester: AHG ratio between 2 and 4. Important product properties will be described in the next chapter. Hanna: Water solubility. 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. thermal characteristics and biodegradability of extruded starch acetate foams. 1999. Bengtsson. Panayiotou: Synthesis. Eng. Thiebaud. Merrill: Properties of Fatty‐Acid Esters of Starch. 74. characterization. A. presumably due to a reduction of the polarity of the reaction medium. Sagar. J. At low vinyl ester/ AHG ratio. Bikiaris. J. K. 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. Xu.4 could be obtained for both laurate and stearate esters. Aburto. Y. 1997. Sci. Polym. Borredon. and biodegradability of fatty‐acid esters of amylose and starch. 1440‐1451. 21. J. D. 361‐368. M. References . Carbohydr. Mullen and E. Appl. Polym. . X. E. Dzenis. C. Higher ratios led to a reduction in the DS. 2003. 1208‐1217. Y. J. J. Appl. . M.5. 2005. Gatenholm: Surface octanoylation of high‐ amylose potato starch films. Aburto.W. Appl. I. . D. Prinos. 1942. Polym. J. Sci. I. Crops Prod. Borredon. Pacsu: Starch Studies: Preparation and Properties of Starch Triesters. 705‐721. Thiebaud. 4. 34. Alric. D. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants With these catalysts. Ind. Sci. Chem. Koch. Bikiaris. Ind.6.
Dubois. . T. Bliard. C. J. Stein. Cellulose 2004. Einfeldt. H. . I. R. Mano. M. . Borredon: Free‐solvent Synthesis and Properties of Higher Fatty Esters of Starch – Part 2. 255‐263. Med. Dicke: A straight way to regioselectively functionalized polysaccharide esters. Soininen. Carbohydr. 2001. .. E. S. Polym. Gunther. Klemm: Preparative and H‐1 NMR investigation on regioselective silylation of starch dissolved in dimethyl sulfoxide. Reis: Thermal properties of thermoplastic starch/synthetic polymer blends with potential biomedical applicability. L. Alric. Hamaili. J. 2003. Laatikainen. Mater. 261‐267. . W. Carbohydr. Peltonen. Starch‐Starke 2005. 2004. 118‐121. R. Mormann. G. I. M. Al‐Higari: Acylation of starch with vinyl acetate in water. . . Alric. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Aburto. Mouysset‐Baziard. Starch‐Starke 2004. J. Six: Partial or total silylation of dextran with hexamethyldisilazane.F. H‐1 NMR and TGA/IR. 251‐260. 1997. Aburto. 84 . Ydens. I. Dellacherie. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. A. Macromol. E. 51. Degee. F. W. 145‐152. M. 1735‐1743. D. Nouvel. 57. 43. B. Starch‐Starke 1999.L. Sci. and E. K. 56. P. Mater. Petzold. 341‐347. P. A. M. 302‐307. C. Biosci. Polymer 2002. Hyvarinen. 14(2) 127‐135. Asplund.Chapter 4 . P. Laignel. 1. Res. Urtti: Determination of the degree of substitution of acetylated starch by hydrolysis. Kussler. J. Elomaa. L. G. 298. Massiot. Senocq. Koniarova. J. 57. R. . S. J. D. 11.
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. trans‐esterification.75) are still partially crystalline.45‐1.5 MPa. vinyl laurate. vinyl stearate .7‐3. elongation at break of 3‐26%. and Na‐acetate).39) are totally amorphous whereas the low‐DS ones (DS= 1. The results were adequately modeled using a non‐linear multivariable regression model (R2≥0. Keywords: starch esters. The effect of the process variables (vinyl ester to starch ratio.96).26‐2. K2CO3. The thermal and mechanical properties of some representative product samples were determined. The basicity of the catalyst and the reaction temperature have the highest impact on the product DS. The reactions were carried out in DMSO using basic catalysts (Na2HPO4. and modulus of elasticity of 46‐113 MPa. High‐DS products (DS= 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. The thermal stability of the esterified products is higher than that of native starch. catalyst intake. Mechanical tests show that the products have tensile strength (stress at break) between 2.
et al  and Aburto. Starch is an attractive feedstock for the synthesis of novel biodegradable plastics.1. The products were biodegradable. Mullen and Pacsu [11. The development of novel biodegradable plastic materials capable of decomposing when given an appropriate environment and time is of utmost importance .65) . The use of vinylesters has also been explored [20‐22]. Examples are agricultural mulch . Chemical or physical modification is required for successful applications as biodegradable polymers. 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. 12] studied the synthesis and properties of of C1‐C6 esters of starch.Chapter 5 5. and particularly on the synthesis of starch acetate (C2) [11‐12]. however. Esterification is one of the oldest methods used to improve starch properties . packaging material. The largest application area of plastics is the use as packaging material (37%) . The limited use is mainly because of a number of unfavourable properties of starch such as low moisture resistance. The mechanical properties of products with longer ester chains (C4 and C6) products and plasticized acetate esters were considerably improved compared to native starch. polyethylene  and polycaprolactone  and chemical modification by esterification with organic acids [11‐19]. and this causes serious environmental problems [1. high brittleness. Thiebaud. Unfortunately. 0. The high DS esters showed interesting properties. The use of native starch as a building block for novel biodegradable polymers. the physicochemical properties of the products were not mentioned. and incompatibility with hydrophobic polymers . However. and food service‐ware . 3‐4]. The global starch production was estimated at 60 million tons in 2004 . Two examples using higher vinyl esters (vinyl laurate) were reported by Mormann. Sagar and Merill  studied the synthesis and properties of C4‐C6 esters of high amylose starch. Introduction Plastics are very useful materials and are used in large amounts (245 million ton per annum) [1‐2]. et al . However. It is cheap and abundantly available . is rather limited to date. The products were hydrophobic and the mechanical properties were considerably improved compared to native starch. the research activities were mainly limited to the use of vinylacetate. Most early studies mainly focused on the use of short‐chain carboxylic acid (C1‐C4). Typical examples are blending with polyvinyl alcohol . the mechanical properties were still not satisfactorily. et al [15‐19] synthesized longer chain fatty esters (C8‐C18) of potato starch and corn amylose using fatty acid chlorides and pyridine. most of the plastic waste is not biodegradable. Alternative routes to avoid the use of acid‐chloride/pyridine combinations have been developed. 86 . however.
5. Germany) were used without further purification. 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 . Materials and Methods 5.2. The samples (about 10 mg) were placed in sealed aluminum cells. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). 5. The experiments were modeled using non‐linear multivariable regression. Germany). and disodium hydrogenphosphate dodecahydrate (both Merck. T‐Bones samples (with thickness of 2 mm) for determination of the tensile properties were prepared using a melt press 1 87 .2.24 ‐ 2. vinyl ester to anhydroglucose ratio. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine.96) were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts in DMSO. and Tg) of each sample were determined from the spectra related to the cooling run and the second heating one. the Netherlands). catalyst type and intake) on the DS of the products.2. 4‐N. sodium acetate. After a first heating run from room temperature up to 200 oC to delete the thermal history of the material. Tc. and acetic anhydride (Merck. The thermal properties (Tm. Japan). the thermal and mechanical properties of representative examples of the highly hydrophobic materials are described and discussed. and toluene were obtained from Labscan (Ireland). Analytical grade methanol. vinyl laurate (Fluka. pyridine. This chapter describes systematic studies using Design of Experiments (DOE) to gain insights in the effect of process variables (temperature.1. Materials Corn starch (approx. each sample was cooled to ‐50°C and then heated again to 200 oC (heating rate 10 oC/min). Germany). The catalysts for the esterification reaction were analytical grade potassium carbonate (Boom.2. All these chemicals were used as received. The spectra were recorded at 50 oC. The samples were heated to 700 oC in a nitrogen atmosphere with a heating rate of 10°C/min. Analytical grade vinyl stearate (Aldrich. In addition. Long fatty esters of corn starch with a broad range in degree of substitution (DS= 0. as recommended by Laignel et al . TGA measurements were performed using a Perkin Elmer TGA 7 Thermogravimetric Analyzer. It was dried for 48 hour at 105 oC under vacuum (~1 mbar) to reduce the moisture content below 2 % before use. DSC analyses were performed on a TA Instruments DSC 2920.N‐ dimethylaminopyridine (DMAP). Technical grade dimethyl sulfoxide (DMSO). and tetrahydrofuran (THF) were supplied by Acros (Belgium).
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 184.108.40.206. 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. 220.127.116.11. Peracetylation procedure and Degree of Substitution (DS) determination The presence of un‐substituted hydroxyl groups in the products resulted in broad and overlapping starch resonances in 1H‐NMR spectra. A peracetylation reaction to substitute all of the remaining hydroxyl groups with acetate groups was therefore applied to obtain reliable DS values. Typically, the starch ester (0.1 g) was added to THF (4%‐w/v) and stirred at 55 oC until dissolution (typically 3 h). Subsequently, the peracetylating reagent (1 mol DMAP, 10 mol acetic anhydride, and 22 mol pyridine per mol AHG units) were added. The peracetylation reaction was conducted for 7 h at 50 oC. The product was precipitated by the addition of methanol and washed several times with methanol before finally dried overnight in a vacuum oven at 70oC and 5 mbar. The DS of the products was calculated using a procedure given in previous work . 5.2.4. Experimental Design The experiments were carried out in 6 blocks. The three variables used in each block were the vinyl ester to AHG starch mol ratio ( x1 ), catalyst intake ( x 2 ), and reaction temperature ( x3 ). All experiments in a block were conducted using a 3‐ variable, 2‐level Full‐Factorial Experimental Design with one center point, giving a total of 23+1 = 9 experiments per block. For each block, a given type of catalyst (Na2HPO4, K2CO3 or Na‐acetate) and vinyl ester (either vinyllaurate or vinylstearate) was applied. This gave a total of 54 experiments (2 types of vinyl
Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch
esters x 3 types of catalysts x 9 experiments per block). In a later stage, the type of catalyst was quantified using the pKb of the catalyst ( x 4 ). The ranges for the individual variables ( x1 ‐ x 4 ) are shown in Table 5.1. The mathematical analysis of the experimental data was performed with the software package “Matchad 13” (Mathsoft). The response (y, DS of the products) was modelled using the following expression:
y = β 0 + ∑ β i xi + ∑ β ii xi2 + ∑∑ β jk x j x k
i i j k≠ j
Here, βi, βii and βjk are the regression coefficients obtained by a multiple 2 regression procedure. One of the terms, namely x 4 , caused singularity of the matrix used for the modelling, and was therefore excluded. A t‐statistic was used to rank the individual regression coefficients (βi, βii, or βjk) according to their relative importance . An analysis of variance (ANOVA) was performed to check the adequacy of the model. The applied procedure is well described in the literature  and consists of calculating the sum of squares (SS) for the model and the error, together with the total sum of squares. In combination with the relative degrees of freedom (DF) it is possible to calculate the mean square (MS) for the model and the corresponding error. On the basis of the latter values, the F‐value for the model is calculated. With this information the P‐value for the model is determined. The latter value is related to the statistical significance of the model.
5.3. Results and Discussion
A schematic representation of the esterification reaction of starch with a vinyl ester is provided in eq. 5.2. The starch was gelatinized before the addition of the vinylester to make the starch OH groups more accessible for reaction. The reactions were carried out in DMSO for 24 h using three different basic salts (Na2HPO4, K2CO3, or Na‐acetate) as the catalysts. The products of the reaction were brownish gels which became lighter in colour after product precipitation and washing with methanol. After drying, the products were isolated as transparent, yellowish solids. The products were soluble in organic solvents such as toluene and THF.
OH O HO OH
R O CH2
O HO O R O
(1) R = laurate (C12) (2) R = stearate (C18)
The products were characterized by 1H‐NMR in CDCl3. The peaks of starch (δ 3‐ 5.5 ppm) and aliphatic hydrogen atoms of the fatty acid chain (δ 0.8‐2.5 ppm) were clearly present. The starch peaks were broad and overlapping, and only after the starch esters were peracetylated, the resolution of these peaks was considerably improved. The DS of the products was determined by using 1H‐NMR spectra of the peracetylated starch esters. The effect of three reaction parameters (molar ratio of vinyl ester to AHG units of starch x1 , catalyst intake x2 , and reaction temperature x3 ) on the DS of starch was studied using a full factorial experimental design with one center point. The ranges of the values of the independent variables ( x1 ‐ x3 ) are shown in Table 5.1. An additional variable x 4 (related to the basicity of the catalyst) was also included in order to obtain a general model for starch esterification. A total of 54 experiments were conducted. Table 5.1. Experimental Design Variables
Independent Variables Full Factorial Variables Vinyl ester to AHG‐ starch ratio [mol/mol] Catalyst intake [%‐w]a Reaction temperature [°C] Additional Variable Catalyst basicity (pKb)b
Low (‐1) Middle (0) High (+1)
2.0 3.5 5.0
80 95 110
K2CO3 (3.66) Na2HPO4 (6.8) CH3COONa (9.25)
3 4 5
in %‐wt based on starch intake, in the model (eq. 5.3.) it is transformed to 105 x catalyst mol amount. taken from ref .
The results of the experiment are given in Table 5.2.
00 5.00 3.00 2.00 4.00 5.44 2.46 1.00 2.00 3.94 2.00 5.00 3.00 2.00 5.42 2.12 2.00 5.00 5.00 5.82 2.00 5.00 5.28 2.54 2.56 1.32 1.00 2.00 5.27 1.00 5.41 1.00 2.27 1.00 3.00 2.00 3.00 5.00 5.00 3.00 5.00 5.00 Catalyst Temperature [%‐w] [˚C] 2.2.00 2.00 3.50 0.00 3.67 2.00 3. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.00 3.00 2.24 2.00 5.00 5.00 2.88 2.00 3.50 2.78 2.00 3.00 3.00 2.00 5.00 3.53 1.00 5.00 4.00 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 .48 0.00 2.75 1.00 3.00 5.72 2.00 5.17 2.00 5.00 2.00 5.00 5.47 1.00 3.00 5.44 2.27 0.00 5.50 2.00 2.35 0.00 5. Overview of experimental data Catalyst Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa Vinyl : AHGa [mol/mol] 3.00 3.00 5.23 2.00 5.00 3.00 5.00 5.96 0.07 0.00 5.62 0.11 2.09 0.00 5.59 2.50 2.00 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 95 80 80 110 110 80 80 110 110 DS 0.00 5.00 3.70 2.00 3.12 0.00 5.87 1.40 1.00 2.00 5.00 5.00 2.00 3.52 2.00 5.52 2.00 4.50 2.00 5.00 4.00 3.93 0.84 2.54 2.00 3.
0027 ‐0.2183 0.50 95 80 80 110 110 80 80 110 110 95 1.1947 β0 β1 β2 β3 β4 β5 β6 β7 β8 β9 β10 β11 β12 β13 92 . 5.0470 0.00 5.64 2.5650 0.0001 0.).00 3.0327 ‐2.1748 0.41 mol ratio of the vinyl ester to AHG units of starch 5.0008 0.00 3.0857 ‐0.00 5.00 5.21 2.0008 ‐0.0023 0.00 3.) Table 5. Values for the coefficients of the DS model for starch laurate and stearate Coefficient Starch Laurate 9.3.1568 ‐2.0011 0.00 3.0120 0.24 2.00 5.1104 ‐0.3. The type of catalyst was quantified using the basicity constant (pKb) in water [26‐27].Chapter 5 45 46 47 48 49 50 51 52 53 54 a.00 5.00 5. Mathematical Modeling Modelling of the DS data for both vinylesters was performed using non‐linear multivariable regression based on the 4 independent variables (Table 5.3.98 2.00 2.1979 Starch Stearate 17.0006 ‐0.1.0010 ‐0.1.0066 0.00 3.59 2.00 4.00 5. stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 4.0732 ‐0.0003 0. DS = β 0 + β1 x1 + β 2 x 2 + β 3 x3 + β 4 x 4 + β 5 x1 x 2 + β 6 x1 x3 + β 7 x1 x 4 + β 8 x 2 x3 + β 9 x 2 x 4 2 2 2 + β10 x3 x 4 + β11 x 2 + β12 x3 + β13 x 4 (5.0016 0.50 2.00 3.00 5.0009 0. The experimental data for both starch laurate and starch stearate esterification are best described with a quadratic model including interaction terms (eq.00 2.55 2.).3.6939 0.64 2.1082 ‐0.90 2.00 2.0249 ‐0.0017 0.The modelled values for βij for the two esters are given in Table 5.96 2.3.
The very low P‐ values indicate that the models are statistically significant.6118 0.4.970 (starch laurate) and 0.5 1.9531 0.875) indicate that the model correctly predict the 93 . we also performed a PRESS analysis  (see corresponding R2 values in Table 5.875 Starch Stearate Model Error Total SS 20.5 0 0 0.5 1 1.3399 0.5 3 0 0 0.629 P‐value 1. which represents an “internal” validation method for the model. respectively and it may be concluded that the models describe the experimental data well.96 x 10‐8 R2 values R2 R2adjusted R2press 0. The parity plots of both models are given in Figure 5.967 (starch stearate). The R2 values for the models are 0.945 0. starch laurate b. which indicate  that all significant variables are included in the model.5 3 a. Finally.0583 F 27. Analysis of variance for DS models for starch esterification Starch Laurate Model Error Total SS 16.5 1 1 0.939 0.4986 16.967 0.2569 0. Parity plots of the DS models for starch laurate and stearate Moreover.5 DS actual 2 2.5 DS actual 2 2. the adjusted‐R2 values are very close to the R2 ones.). starch stearate Figure 5. and confirm this statement.805‐0.026 x 10‐7 R2 values R2 R2adjusted R2press 0. Table 5.8386 DF 13 14 27 MS 1.4.5 1 1.805 3 3 2.7699 DF 13 14 27 MS 1.2903 P‐value 1.5 2.1.970 0.5 2 DS predicted DS predicted 2 1.8167 21. 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 5.4.1.0356 F 35.5 0. Also in this case the reasonable R2 values (0.
These results may be rationalised by considering the role of the catalyst in the modification reaction. Unfortunately. Another explanation may be related to the regio‐chemistry of the reaction. It is assumed that the first step in the reaction sequence is activation of the starch OH groups by deprotonation by a base [21. The highest product DS values were obtained using K2CO3.Chapter 5 products DS as function of the process variables within the range of experimental variables.8) is intermediate between that of the other two catalysts. 22]. On the basis of the sequence. the base strengths of the catalysts in DMSO are not known. whereas catalyst performance is considerable lower. Of all the variables studied.2. 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. The product DS for Na2HPO4 was considerably lower (1. the type of catalyst has the largest effect on the product DS for both the laurate and stearate esters (Figure 5. 94 . 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. The pKb for the latter (6.82). two plots showing the dependence of the DS on reaction temperature and catalyst amount (in %‐w). The resulting anion will react with the vinyl ester to from the product. such as carbonate or acetate salts leading to higher DS esters.2. However.18) and CH3COONa (1. This was not the case for alkaline catalysts. K2CO3 is a stronger base (pKb 3. A similar trend was observed by Dicke  for the acetylation of Hylon VII starch using vinylacetate.66) than CH3COONa (pKb 9.). 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.00) than for K2CO3 (2.25) and this leads to higher product DS values for K2CO3. for both starch laurate and starch stearate are given in Figure 5. To evaluate the effect of each variable on the product DS. Catalyst performance of CH3COONa was slightly less than for K2CO3 whereas the lowest DS products were produced when using Na2HPO4 as the catalyst. This is indeed the case when comparing the performance of K2CO3 with CH3COONa. the performance of Na2HPO4 does not follow this trend. it can be rationalised that the rate of the reaction (and thus the product DS) will be higher when using a stronger base.
2. starch laurate b. 3D contour plot of the DS as a function of reaction temperature and catalyst intake (at constant vinylester to AHG mol ratio of 4) Besides the type of catalyst. As expected and in 95 . Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch a.2. although to a lesser extent.). starch stearate Figure 5. the catalyst intake and reaction temperature also affect the product DS (Figure 5.
2. Of all variables studied. Product Properties The thermal and mechanical product properties for two starch‐laurate and stearate samples with medium (DSlaurate = 1. 5.3. higher reaction temperatures and catalyst intakes lead to higher product DS values. the vinyl ester to AHG ratio has the smallest effect on the product DS.39) 96 .5.Chapter 5 line with studies on starch acetates . and the results for all samples are summarised in Table 5.3. . DSstearate = 1. Representative DSC spectra for the starch stearate are given in Figure 5.39) were determined.45 (. . DSC analysis of starch stearate esters ( ) DS 1. heating Exothermic Tm = 21 oC cooling Endothermic Tc = 12 oC -20 -10 0 10 Temperature ( C) o 20 30 40 Figure 5.45) and high DS values (DSlaurate = 2.26. DSstearate = 2.) DS ( 2.75.3.
especially in the amylopectin part [15. In particular the behaviour at relatively high temperatures (>0 oC) is further discussed.d. To the best of our knowledge. At moderate DS values the products still display a melting temperature (at 20‐30 oC) and a crystallization temperature (at 10‐25 oC).a 21 Tc [˚C] n. crystallization. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.).d. close to the value obtained in our study for the medium DS starch stearate (21 °C). Unfortunately the thermal transitions for the virgin material lie above the degradation temperature and can therefore not be accurately determined . 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. or glass transition above 0 oC) were even absent for the high DS products. 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). Changes in the transition temperatures of native starch by esterification have been reported in the literature [13‐14. In any case.d.d.8 . which is known to contain crystalline areas.75 2. These values are reduced considerably compared to virgin corn starch.45 Tm [˚C] n. TGA analysis was performed to study the thermal degradation behavior of the starch laurate and stearate samples (Figure 5.a 27 n. 40‐42 oC. For instance a Tm of 32 oC was reported for a potato starch stearate ester with a DS of 1.39 1. Thermal transitions at relatively high temperature (melting.26 1. esterification of starch with vinyl laurate and stearate reduces the crystallinity of the starch considerably and leads to a lowering of the transition temperatures.5. implying that these products are fully amorphous. octadecane. 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. However. these are the first examples of completely amorphous starch esters. n.a 12 a.= not detectable The thermal behavior of the products is a strong function of the DS. 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.d. 17]. in the present work.4. 97 .a 22 n. The melting point of the potato starch stearate was close to that of model compounds for the stearate side chains (methyl stearate. 22].
Chapter 5 100 80 %-weight 60 40 20 0 20 100 200 300 400 o 500 600 700 Temperature ( C) Figure 5.4. Native corn starch degrades at lower onset temperatures (250‐300 oC) than the corresponding esters (300‐350 oC).26) (. TGA analysis of various starch samples. . ( ): native starch ( ): starch laurate (DS 2.): starch stearate (DS 2. Similar trends were reported for amylose octanoate‐ stearate‐laurate esters  and potato starch octanoate‐ laurate esters .6.). elongation at break and elasticity modulus) of the products were determined (Table 5. The difference in thermal stability between the laurate and stearate sample is limited.39) The TGA (Figure 5. 98 .) curves clearly show starch esterification with either laurate or stearate results in products with enhanced thermal stability. The mechanical properties (stress at break.4. .
17].7 ± 1.7 ± 3. On the other hand.45) displays a very similar behavior with respect to the laurate samples (i.05 for both laurate samples. a clear plastic behavior with a maximum in the stress‐strain curves.05).1 ± 15.e.5 MPa. 99 .0 3.1 Elongation at break [%] 26. however.025) lies still in the typical range of engineering polymers but also close to that typical of composite materials .2 ± 0.75 2. elongation at break between 3‐26%.2 2. and lower elongation at break).45 Stress at break [MPa] 3. is roughly 0.0 112.0 ± 2. The values of the yield strain.39 1. thus indicating a relatively more rigid material with respect to all other samples.0 3. observed for the high DS starch stearate sample.8 Starch Ester Starch laurate Starch laurate Starch stearate Starch stearate All materials generally show tensile strengths between 2.2 8. plastic deformation and yield strain of about 0. Starch stearate with a relatively low DS (1. For starch laurate. An inverse behaviour is. determined from the stress‐strain curves as the stress at which the material ceases to be linearly elastic) and the modulus.1 ± 1.1 Modulus [MPa] 59. This indicates that their mechanical behavior is comparable to that of typical engineering polymers .7 ± 14. This product is more rigid compared to the product with medium DS value (higher tensile strength and elasticity modulus.7 ± 4. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. an increase in product DS results in lower tensile strengths and modulus of elasticity.3 ± 2.5 ± 1. calculated as the ratio between the stress at yield (σy. but higher elongation at break.7 ± 4.9 ± 18.104.22.168‐3. A deeper understanding of the mechanical behavior can be gained by investigating the shape of the stress‐strain curves (typical examples in Figure 5.26 1.1 82.39) does not show any plastic behavior but only an elastic one. This is in agreement with previous studies regarding the mechanical properties of starch esters with different chain length of the fatty acid moiety [13. Mechanical properties of the starch esters DS 2. starch stearate at relatively high DS values (2.3 21. Both starch laurate samples display.3 3.).9 46. and modulus of elasticity between 46‐113 MPa. The corresponding yield strain (roughly 0. These considerations strongly point out the fact that the mechanical behavior can be coarsely (plastic vs elastic behavior) but also finely (stress and strain at break but also modulus values) tuned by the chemical structure of the fatty acid chains as well as the DS values. independently of the DS values.
26) (. . Although there are many differences in the synthetics methods as well as in testing conditions and procedures.5.Chapter 5 6 5 4 stress (MPa) 3 2 1 0 0 5 10 15 20 strain (%) 25 30 35 Figure 5. 100 . .45) (‐ . Stress‐strain curves for starch esters.).7. 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 5.39). ‐): starch stearate (DS 2.): starch stearate (DS 1. ( ): starch laurate (DS 1.75) ( ): starch laurate (DS 2.
6 Corn Starch Stearate Ester Low DS (1.8) 3.1 ±15. 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.7 ±0.3 High DS (2. confirms that the corn starch esters synthesized in our research are relatively rigid materials as compared to the potato starch esters.a.26) 3.26) relatively high stress at break (3.7) 1.0 ±2. we show that although no crystallization of C18 occurred (see DSC data of the high DS material in Table 5. merely because of the structural property of the C18 side chain with respect to the starch backbone. 3.7 MPa) coupled with a significant (1500%) elongation.75) 3.7.7 ±3. Moreover. 101 .2 8.7) 0.). n. as confirmed by DSC.= not available Inspection of the mechanical properties of the starch esters as shown in Table 5. n. In agreement with our own data.3 Potato Starch Stearate Ester Low DS (1. Comparison with the literature data implies that the mechanical behaviour of starch esters of higher fatty acids is also a clear function of the type of starch used. DS) which mainly determines the thermal and mechanical behaviour. 9% for potato starch).2 ±0.45) 3. While for corn starch (DS 2.6 9 ±2 High DS (2.1 High DS (2. They related this inverse property of starch stearate to crystallization of C18 side chains. an unexpected behavior was also observed for the mechanical properties of the starch esters from potato starch.7 ±0.a.39) 2.a.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. Here.7.5 ±1. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.0 3. On the other hand at low DS values comparable stress at break (3.7 ±4. High DS (2.9 ±0. but the interplay between those factors.0 26.5 MPa) is coupled with a low elongation (26 %).7 MPa for potato starch ones) are coupled with significant differences in the elongation (22% for corn starch esters.1 ±1. In this respect the difference in stress and elongation at break for the high‐DS starch laurate sample is striking.5. for potato starch exactly the opposite is observed: relatively low stress (0.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) .4 1500 ±8.2 MPa for corn starch esters. the high DS material is still more rigid.3 10 ±2 Property Low DS (1.1 21. The same considerations are valid for starch stearate at high DS values (higher stress and lower elongation when using corn starch instead of potato one).7 ±1.2 Stress at Break [MPa] Elongation at Break [%] a.
reaction temperature. The models may be used to determine the appropriate process conditions to obtain a product with a pre‐defined DS.4‐3 for starch stearate). Comparison with literature data indicates that the properties may be (fine) tuned also by the starch source. Nomenclature β : regression coefficients obtained by a multiple regression procedure [various units. Reaction temperature and catalyst intake also affect the product DS although to a lesser extent than the type of catalyst. and catalyst basicity) to the DS of the starch ester products were developed. and Na‐acetate). 1. The use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2.5. 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. K2CO3. 5. Thus. while the use of Na2HPO4 resulted in low‐medium DS products (0.1‐2.5 for starch stearate). 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.Chapter 5 5.5 for starch laurate. catalyst intake. on the synthesis of corn starch esters with long chain fatty acids is described. including statistical modelling. The DS of the products is a strong function of the basicity of the catalyst. Conclusions A systematic study. 0.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐starch mol ratio.3‐1. the chemical composition of the starting materials (either starch or the ester precursor) as well as the processing parameters affect the final product DS values and as such provide an effective toolbox to modulate the desired product properties for a given application.4.9 for starch laurate. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] 102 .07‐1. Statistically adequate (R2 ≥0.
H. Germany. Marsden: The challenge of domestic waste disposal. F. An Introduction to the New Science of Biodegradable Plastics. UK. Copenhagen. Culter: Plastics Packaging. Hernandez. Weber). Website of International Starch Institute. W.starch. Res. demand and recovery for 2006 in Europe. Chem. . Selke. Hanser Publishers. G. 1986. E. 2000. 2008.6. . Otey. Association of Plastics Manufacturers. M. G. 1980. 27‐29 August 2000 (Ed. Structure – Properties ‐ Applications. M. Department of Dairy and Food Science. J. in Packaging in the Environment (Ed. Denmark: http://www. Levy). Eng. Properties. M. Denmark. Plastics Europe. S. E. C. 2000. J. USA. Applications. Packaging. Munich. Processing.dk/isi/stat/rawmaterial. Glasgow.html. Westhoff. Conference Proceedings. References . 2002. New Jersey. Denmark. R. Belgium. Prod. Stevens: Green Plastics.J.W. and Regulations. Germany. Brussels. 592‐595. 2000. Princeton University Press. . 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. 2008 . 19. 103 . . R. Aarhus. Munich. C. in The Food Biopack Conference: Foodstuffs. Dev. A. Ehrenstein: Polymeric Materials. Ind. Blackie Academic & Professional. . The compelling facts about plastics: An analysis of plastics production. The Royal Veterinary and Agricultural University. . S. Hanser Publishers. Frederiksberg C. Bastioli: Global Status of the Production of Biobased Packaging Materials. Doane: Starch‐based blown films. and Biopolymers. accessed on August 01. P. D.
Aburto. 385– 409. characterization. Mullen. R. Panayiotou: Synthesis. J. Prinos. K. Panayiotou: Properties of octanoated starch and its blends with polyethylene. Alric. Sci. Part C: Polym. Carbohydr. Carbohydr. 1209‐1217. J. L. E. Ind. 381‐384. 35. W. Sagar. J. 28. Hamaili. Bikiaris. Pacsu: Starch studies: possible industrial utilization of starch esters. . X. Appl. de Graaf: Ph. Starch‐ Starke 2005. C. Polym. Macromol. 65. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. 1440‐1451. 145‐152. Appl. Department of Dairy and Food Science. F. Yang. Thiebaud. Mouysset‐Baziard. . Y. I. 51. G. . Broekroelofs. E. Appl. Janssen. G. Alric. E. E. M. P. Aburto. E. Aburto. and biodegradability of fatty‐acid esters of amylose and starch. de Graaf. . . 1995. 51. . Starch‐Starke 2004. . Frederiksberg C. Al‐Higari: Acylation of starch with vinyl acetate in water. Z. J. 118‐121. . Rev. 132‐135. report of Food Biopack Project (Ed. S. Ind. M. W. A. . . J. Borredon. Borredon. 56. Starch‐Starke 1999. W. A. D. 1996 104 . status and perspectives. Chem. J. C.. 705‐721. C. 2000. I. Borredon. Senocq. E. J. Thiebaud. A. B. Aburto. A. C. Mullen. 2003. D. the Royal Veterinary and Agricultural University. C. I. J. R. 34. Weber). S. 58. Starch‐Starke 1999. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. A. I. Thiebaud. J. 101‐112. Bikiaris. Biobased packaging materials for the food industry. I. Polym. H. Pacsu: Starch studies: preparation and properties of starch triesters. Merrill: Properties of fatty‐acid esters of starch. I. Alric. Polym. W. Alric. 1997.Chapter 5 . 57. 1999. Sci. Rijksuniversiteit Groningen.. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. Wang: Properties of starch blends with biodegradable polymers. Bikiaris. M. 1997. Alric. Denmark. Polym. Thesis. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Chem. K. Sci. 1943. . 302‐307. 137‐144. E. 1942. . 1995. C43. E. 74. 1647‐1656. J. D. L. Beenackers: The kinetics of the acetylation of gelatinised potato starch. Sci. Aburto. Polym. J. J. J. E. J. Wang. Alric. J. S. Mormann. Aburto. 34. . D. Prinos.D. A. Eng. Eng. J. Prinos.
New York. H.. . USA. L. 255‐263. Picchioni. L. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. J. 11. B. New Jersey. 251‐260. R. Shercliff. John Wiley & Sons Inc. Science. A. K. 1986. Laignel. Dicke. . C. Junistia. 105 . A. Sugih. M. A. 2001. . accepted for publication in Starch‐Starke. 298. A. . USA. Cellulose 2004. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Materials: Engineering. Janssen. Manurung. M. B. Processing and Design. Res. R. F. Bliard. J. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch . . P. D. C. Butterworth‐Heinemann. Massiot. Cebon. Amsterdam. D. G. Day. Underwood: Quantitative Analysis 5th edition. Carbohydr. Ashby. M. L. A straight way to regioselectively functionalized polysaccharide esters. 1997. the Netherlands. 2007. Prentice‐Hall. 2008. R. Montgomery: Design and Analysis of Experiments 5th edition. H.
compatibilizer. 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. glycidyl methacrylate. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). the mechanical behavior of the compatibilized blends is in general different from that of pure PCL and of the corresponding uncompatibilized blends. As a result. PCL‐g‐ glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM) is decribed. 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. The reactive interfacial agents were tested for their performance in PCL‐starch blends. Both act as compatibilizers for PCL/starch blends by improving the interfacial adhesion between the starch particles and the PCL matrix. despite the expected differences (favorable to PCL‐g‐GMA) in the in situ formation of the compatibilizers. diethyl maleate. grafting . This discrepancy could be explained on the basis of the functional groups (GMA or DEM) distribution along the PCL backbone. In particular the elastic modulus for the compatibilized blends is significantly higher than that of uncompatibilized ones.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. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. Keywords: starch‐PCL blend. at relatively high GMA and BPO intakes. At relatively low starch intakes.
1. such as polyethylene (PE) or polystyrene (PS) [3‐6]. these synthetic polymers are poorly or non‐biodegradable. To overcome this issue. To reduce the tendency for phase separation. upon mixing) by using a functionalized PCL. is known to be degraded with ease by microorganisms widely distributed in nature . However. synthetic biodegradable polymers have been applied. for starch/PCL blends. Among these. A well‐known biodegradable polyester. as in this study. the native material is very hydrophilic and important mechanical properties are inferior compared to most synthetic polymers and this hampers its direct use as packaging materials. it would be actually more accurate to define the block copolymer as an “interfacial agent”. 2]. Unfortunately. Blending of starch and synthetic biodegradable polyesters has been widely applied for the synthesis of fully‐biodegradable products. 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 ). 108 . The ester bonds are susceptible to attack by water and this leads to enhanced biodegradability. This leads. an interfacial agent) may be used to improve the interfacial association between the two polymer phases. blends of hydrophilic starch and hydrophobic biodegradable polyesters exhibit phase separation  due to differences in polarity of the building blocks. of polar groups (usually epoxides or anhydrides [13‐18]) able to react with the – OH groups along the starch backbone.e. 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. polycaprolactone (PCL). polyesters are considered very promising alternatives . The latter displays the presence. Introduction Starch is a cheap and abundantly available natural polymer with very good application perspectives in the area of biodegradable plastics. to an ideal compatibilizer having both PCL and starch blocks. Starch modifications to improve the product properties like enhanced hydrophobicity and mechanical properties were already reported in the early 19th century [1. When using ungelatinized starch as a component in the blend. This phenomenon is highly undesirable and limits the application range considerably .e. Unfortunately. along the backbone. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decrease rapidly in time .Chapter 6 6. a compatibilizer (i. As a consequence. PCL has gained considerable interest for possible applications in the fields of packaging materials and medical applications [10‐11]. 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.
8%) from Merck. an absolute novelty of the present research. Ireland. The synthesis and application of PCL‐g‐DEM is. UK. 109 . after which the equipment was stopped and the chamber was opened to collect the samples. Belgium.1. 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. ~1 mbar) prior to use.2. The starch was dried for at least 24 h at 110oC under vacuum (approx. The kneader was heated to 130 oC and PCL (CAPA 2304) was added while maintaining a rotational speed of 80 rpm. Glycidyl methacrylate 97% purity (Aldrich). and benzoyl peroxide 75% (Merck. This low molecular weight PCL grade was used without further purification. >99%) was obtained from Acros.2. Germany) were used as received. After the PCL was melted (1‐2 minutes).1.8%) from Labscan. Methods 6. This chapter describes a systematic study on the synthesis of two reactive interfacial agents for starch‐PCL systems. however. The materials were mixed for another 5 minutes. Corn starch (with approx. The various compatibilizers have been tested for their performance in PCL‐starch blends. a solution of BPO in GMA or DEM was added drop by drop over a period of 5 minutes. Mw=3000 from Solvay Caprolactones. Germany and methanol (99.2. to the best of our knowledge. Tetrahydrofuran (THF. The intake of reagents was maximally 75‐80% of the volumetric volume to ensure proper mixing. Exploratory studies on the synthesis of PCL‐g‐GMA and its applications for starch/PCL blends have been published [13. 6. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). UK was used for the preparation of the interfacial agents.2. Compatibilizer synthesis The compatibilizers were prepared in a Brabender Plasticorder PL2000 batch‐ kneader (chamber volume 35 cm3). 6. xylene (99.2. Materials and Methods 6. Intakes for each experiment are given in Table 1 and 2. The effect of important process variables on the degree of grafting of the GMA and DEM units to the PCL backbone has been explored in detail and quantified using multivariable linear regression. 16]. Mw=50000) from Solvay Caprolactone. systematic studies and subsequent quantification of the functionalization reaction has not been reported to date.2. diethyl maleate ≥97% purity (Fluka). 73% amylopectin and 27% amylase) was obtained from Sigma and high molecular weight PCL (CAPA 6503. Materials Polycaprolactone (PCL) CAPA 2304.
8%. Before analysis. The resulting suspension was filtered and precipitated using methanol (450 mL) at 6‐8°C.Chapter 6 6. the samples were covered with a paladium/platinum conductive layer of 3 µm thickness. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC.2. and then filtered. Methanol (450 mL) was added to the filtrate.2. the chamber was opened and the resulting material was collected.3. An operation temperature of 170°C and a rotation speed of 80 rpm were applied .5 h. The content was blended for 15 minutes.2. 5 mbar). Subsequently. stirred for 1.2. 5 mbar). 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 . and the product was precipitated at 6‐8 oC. Each sample was first heated from 0 oC to 100 oC (heating rate 10 oC/min) to remove the thermal history of the material. Tensile tests were performed using an Instron 4301 1 110 . Analytical Methods H‐NMR measurements were performed using a 400 MHz Varian AMX Oxford NMR apparatus with CDCl3 (99. Work‐up of PCL‐g‐GMA products  To remove unreacted GMA monomer and GMA homopolymer. 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. created using a Cressington 208 sputter coater.2. The error on the transition temperature is assumed to be ± 1 oC and 5 % of the calculated values for the corresponding enthalpies. Scanning Electron Microscopy (SEM) was performed using a Jeol 6320 F Scanning Electron Microscope. 6.2. 6. PCL was added to the chamber followed by the addition of the starch and the reactive compatibilizer.4. 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). Infrared spectra were collected with a FT‐IR apparatus in the ATR mode using a Spectrum 2000 instrument from Perkin Elmer. PCL‐g‐GMA (5 g) was dissolved in 50 ml THF.2. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC.3. 6. PCL‐g‐DEM (5 g) was dissolved in xylene and refluxed at 150 oC for 2 h. Digital Scanning Calorimetry (DSC) measurements were performed using a Q1000 TA Instruments equipped with a TA Instruments DSC cooling system. Aldrich) as the solvent.
strain at break (ε). 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). 111 . As a result we were able to obtain a model for the FD of the reaction. 8 different T‐bones were used. stress at break (σ) and modulus (E) were measured. This procedure is described in detail in the literature  and consists of calculating the sum of squares (SS) for the model and the error.2 ppm) or DEM (‐CH2‐ protons at δ 4. 15‐16. When the relative degrees of freedom (DF) are known.0 ppm [13. The latter is a measure of the statistical significance of the model.1. Table 3). The corresponding value for every blend was calculated as an average of the 8 measurements while the standard deviation was taken as absolute error on the average values. leading to a 10% relative error in the FD values.3. 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.) The FD was calculated using 1H‐NMR by comparing the area of protons belonging to the GMA (‐CH< proton at δ 3. 6. The T‐bone samples were prepared using a Fontijne Holland TH 400 hot‐ press. For every T‐bone.2 ppm) side chains with that of a characteristic proton resonance of the PCL backbone (‐CH2‐ protons at δ 4.1. The validity of the model was determined by performing an analysis of variance (ANOVA. it is possible to calculate the mean square (MS) for the model and the error.2. For a given sample/blend. the F‐ value for the model is determined followed by the P‐value.4.2. 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. A 5% relative error in the peak area of the NMR spectra was assumed. 19]. On the basis of the latter values. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends machine. 6.
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 6.1.1. Functionalization reactions (only showing reactivity for the >CH2 in α position on PCL backbone) Typical 1H‐NMR spectra for the products are shown in Figure 6.3. 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. Results and Discussions 6.3.1.Chapter 6 6. 112 .
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
1. The results are graphically provided in Figure 6. BPO=0. i. higher FD values for GMA than for DEM are in line with this explanation and support the proposed molecular reaction mechanisms.Chapter 6 chemical structure of DEM. It is most probably related to the fact that GMA is able to propagate to longer grafted chains when reacted with PCL (Figure 6. However.3% mol/mol-CL units DEM. Effect of GMA and DEM to CL‐units ratio (mol/mol) on the FD of the products (constant PCL intake. Hence. 22].1% mol/mol-CL units 10 5 0 10 20 30 40 50 60 Substrate Intake (mol% on CL units) Figure 6. 6. Our experimental findings.6% mol/mol-CL units 15 GMA. 116 .).3. 130°C) Higher GMA intakes lead to higher FD values.2.6% mol/mol-CL units DEM.1. This trend is independent of the BPO amount and matches with data reported by other groups [13. subsequent reactions of MA to an already grafted MA molecule hardly occur. BPO=0. BPO=1.e. 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. 30 25 20 FD (%) GMA. the length of a MA graft is always unity whereas longer grafts are possible for GMA. showed that MA reacts easily with a radical on the PCL backbone. BPO=0. Thus an increase in the GMA intake will provide more monomer available for the growing of the grafted chains leading to higher FD values.3.3.
no detectable influence of the BPO amount on the FD is observed (Figure 6.).4. 15. Such behavior is slightly in contrast to what observed for maleic anhydride [13. Apparently. The use of higher initiator concentrations will result in an increase in the number of formed radicals. Such discrepancy is probably related to the relatively low intake of DEM as well as to the different reactivity of DEM compared to MA .3 0. for which a levelling off and eventually a decrease of the FD values for relatively higher MA amounts has been observed.4 0.2. constant substrate to PCL ratio) For high GMA to CL ratio (24%‐mol/ mol CL units). Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The experimental trend for DEM is different.8 0.). at relatively lower GMA intakes. 17].6 0. doubling the amount of initiator results in considerable higher product FD.5 0. GMA=24% mol/mol-CL units GMA. 17]. 50 45 40 35 30 FD (%) 25 20 15 10 5 0 0. although a slight increase in the FD values might be appreciated. This will lead to a higher proportion of PCL radicals by hydrogen abstraction from the polymer backbone and thus to higher FD values. Effect of the BPO intake on the product FD (130°C. GMA=12% mol/mol-CL units DEM.4. The FD values are within a rather narrow range.4. DEM=30% mol/mol-CL units 0.2 GMA. However.2 BPO Intake (mol% on CL units) Figure 6. 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. These results are in line with earlier work [15.9 1 1.1 1. there 117 .7 0.1.3. 6.
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).1. This phenomenon is likely related to the different nature of the radical grafting mechanism of GMA and DEM on PCL as mentioned before. 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. The effect is however much less pronounced than for GMA. namely an increase in the BPO intake results in higher product FD. This leads to the following equation: FD = f (nm . the mutual solubility of DEM in PCL and DMA in PCL was included in the model by using a parameter δ. recombination of (macro) radicals.e. In addition. 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). as well as several transfer reactions might be responsible for the observed trend . 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. δ ) (6. and 6.2729nm niδ (6. However. the possibility of side reactions will become significant.0325nmδ + 0.3.) 118 . To quantify synergic effects of monomer and initiator molar intakes on the FD of GMA and DEM on the PCL backbone.3. i. ni . 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 [15.1. ni the molar amount of initiator in the feed and δ the difference in solubility parameters calculated using group contributions .6022nm ni − 0. a statistical model has been developed by performing a multivariable linear regression on the data reported in Table 6.2. Here. 17].3.5431niδ + 1. defined as the difference in solubility parameters between PCl and the substrates.2.) where nm is the molar amount of monomer in the feed. gives the best description of the experimental data: FD = −1. If many macroradicals are created in the system (relatively high BPO intakes) at low GMA intake. 6.Chapter 6 is an optimum ratio between the BPO concentration and available monomer (GMA in this case) on the FD. In particular the occurrence of “cage effects”.8875 + 0. 6. the BPO and monomer intakes are considered as independent parameters.3. The model provided in eq.
3D plot. b. SS 1828 81.17 F 44. Graphical representation of the FD models for PCL‐g‐GMA and PCL‐g‐DEM are given in Figure 6. Table 6..3 10 10 20 30 20 0.957) and its closeness to the adjusted R2 (0. respectively. Model Error Total Analysis of variance for the FD model provided in eq 6.3.6 20 10 20 40 50 30 60 FD (%) 40 FD (%) 30 40 20 0 1 0.8 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 ). 1 0.3. Figure 6.941) also suggests that all important variables have been included in the model. The very low P‐value implies that the model is statistically significant. Graphical representation of the FD model for PCL‐g‐GMA.6 0.9 60 70 40 50 80 BPO Intake (mol% on CL units) 0. the interaction between (PCL‐monomer) intake and solubility parameter difference ( nmδ ) and the interaction between the latter factor and the initiator intake ( ni δ ).4 0. and Figure 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 6.8 0.5 0.4 BPO Intake (mol% on CL units) 0. The R2 value for the model (0.7 0.3.6. a.358 1909.5. b.908 10.5. Contour plot.2 10 20 GMA Intake (mol% on CL units) 40 30 50 0. This was also confirmed by inspection of the residue distribution by a normal probability plot (not reported here for brevity) . 119 .358 DF 4 8 12 MS 456.
7 6 3 3 4 Figure 6. Thermal properties of the compatibilizers The thermal properties (Table 6. the crystallization temperature and enthalpy are not a clear function of the FD values. which is expected to lead to a Tc reduction.9 4 5 8 BPO Intake (mol% on CL units) 0. although both properties are significantly lower than those of pure PCL.8 0.8 0.6. Similar observations have also been made by Kim. radical initiator and PCL intakes).5 0. From a more practical point of view. b.4.1. and a favored nucleation 1 2 50 120 .2. the melting temperature and the relative enthalpy decrease with respect to pure PCL. for DEM a clear transition is observed. Furthermore.3.) of the compatibilizers were determined by DSC. et al  working with PCL‐g‐GMA. Contour plot. 3D plot. For all samples. Graphical representation of the FD model for PCL‐g‐DEM.Chapter 6 1 0. a.7 3 4 6 FD (%) 4 0. This induces irregularities and is expected to result in a lowering of the Tc and the Tm. Remarkable is once more the different trends for the two substrates.4 0. The random behavior of the Tc as function of the FD values is probably the result of two concurring effects: the presence of irregularities. 6. While for GMA an increase in BPO or monomer intakes invariably leds to higher FD values.1.2 10 15 20 25 30 35 40 DEM Intake (mol% on CL units) 45 7 a. Such changes in the thermal properties compared to pure PCL may be caused by the introduction of grafts on the PCL chains. a higher BPO intake leads to a reduction of the FD values while an opposite trend is observed at lower DEM intakes. For relatively high DEM intakes (> 30 % mol/mol).3 5 6 0. and 6. These differences in dependency of the FD values on the DEM and GMA intakes may be explained on the basis of the different grafting mechanisms as previously discussed. b. the statistical model provides besides a reliable description of the experimental data also a good mathematical description to modulate the FD values of the two substrates by changing the chemical composition (monomer.6 2 FD (%) 5 2 0 1 0.2 10 20 DEM Intake (mol% on CL units) 40 30 50 6 0.4 BPO Intake (mol% on CL units) 0.6 0.
An overview of all prepared blends together with their thermal and mechanical (tensile tests) properties is given in Table 6.0 430.1 10.7 10.4.4 230.8 384.6%.2 321.3 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. 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.2.5 357.9 11.5 ε (%) 640.9 368.6 168.6 11.8 372.4 431.5 7.3 380.9 E (MPa) 270.3. The lower crystallization and melting enthalpy of the compatibilizers compared to virgin PCL is possibly caused by the presence of side chains on the PCL backbone which are expected to have a negative effect on the crystallinity of the products.9 424.2 342.8 329. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends of the PCL chains in the compatibilizers compared to virgin PCL (higher Tc) induced by the presence of polar groups. A total of 12 blends were prepared: a series of binary ones (not containing any compatibilizers) constituting our reference points.2 343.9 379. 6.3 10.2 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.1 13.0 401.8 371.6 386.6 5. 121 . 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) 16.0 9.6 305.0 332.7 %.5 489.0 341.1 11.3 15. PCL‐g‐GMA has an FD of 9. a series with PCL‐g‐DEM (different intakes of the latter at fixed starch/PCL ratio) and two series with PCL‐g‐GMA (one with different intakes of PCL‐g‐GMA at a fixed starch/PCL ratio and one with a fixed compatibilizer intake at three different starch/PCL ratios). Table 6.2 337.5 425.4.
8 MPa) clearly indicates that the rigidity of the blends increases at higher starch contents.2.8 % at 30 % starch content.7.2 MPa up to 341.3.43 MPa and 640. Melting and crystallization enthalpy as function of the starch content for binary blends with PCL. 75 70 65 60 55 50 45 40 35 ∆ H (J/ g-PCL) 0 5 10 15 20 Starch content (%) 25 30 Figure 6. 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.7. This is most probably due to a nucleation effect of the starch on PCL as observed also for other kinds of polymer blends .1 MPa and 341. The simultaneous increase in the modulus (from 270.5 % for pure PCL to 7. 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 .).1. Binary blends of starch and PCL For binary blends (no compatibilizer is added) a monotonous decrease of the stress and strain at break is observed as function of the starch intake. The thermal and mechanical properties indicate no or very little interaction of the starch particles with the PCL ○ : ∆Hc □ : ∆Hm 122 .Chapter 6 6. The thermal behavior is characterized by no significant changes in the Tm and Tc values but unreported trends of the corresponding enthalpies as function of the starch content (Figure 6. respectively from 16.
However. The thermal behavior is characterized (Table 6.3.). for which a 6 oC drop is observed when using PCL‐g‐DEM. In this respect the thermal properties remain substantially unchanged with the exception of the Tc. was used as compatibilizer for the preparation of ternary blends with starch and PCL. 26] and explained by assuming that the voids are formed by water in the starch. The latter trend is also valid for the corresponding enthalpies: i. The presence of these voids was also observed for sago starch/PCL blends [25.2. Figure 6. Evaporation of the water during blend preparation. a. the starch particles are simply inserted into voids of the PCL matrix.8.8. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends matrix. either by heating or in combination with mechanical stress . (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.9. Morphology of starch/PCL binary blends. a decrease with respect to pure PCL is observed and then a substantial invariance as function of the PCL‐g‐DEM content. 123 . whose synthesis is described for the first time in this work. and the lack of interface adhesion cause void formation.) 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.2.). the corresponding binary blend (S/PCL 20/80) contitutes a better reference point compared to pure PCL. As consequence one might expect a slight different structure of these ternary blends as compared to the corresponding binary one.4. This is confirmed by examination of the blend morphology by SEM (Figure 6. in order to fully understand the role of PCL‐g‐DEM. b. c. 6. Thus.e. This is visually confirmed by the morphology of the prepared blends (Figure 6. Such behavior is consistent with the hypothesis that the starch particles hinder the melting and crystallization processes of the PCL chains. Ternary blends compatibilized with PCL‐g‐DEM PCL‐g‐DEM.
Chapter 6 a. The stress and strain at break remain constant as function of the compatibilizer intake while the modulus display an optimum as function of the compatibilizer intake (Figure 6. which will hinder the flowability and fibre forming ability of PCL matrix under cold drawing. At higher PCL‐g‐DEM intakes the lower average molecular weight of the compatibilizer as compared to the one of the unmodified PCL used in the blends (3000 vs 50000) is probably responsible for the observed decrease in the E values.). 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. 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.10.9. resulting in more rigid material with higher modulus. Morphology of ternary blends compatibilized with PCL‐g‐DEM. 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. b. This can be explained by the improved interfacial adhesion  between PCL and starch.). c. (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 .9.10.
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. In particular also in the case of PCL‐g‐GMA. 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). the stress and strain at break are lowered and hardly a function of the PCL‐g‐GMA amount. However. 6.2. 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.e. Modulus of ternary blends S/PCL/PCL‐g‐DEM as function of PCL‐g‐ DEM amount. by taking as reference the corresponding binary blend (S/PCL 20/80). it can be observed that: 1) the melting temperature as well as the crystallization and melting enthalpies do not change significantly (discrepancies in the values are within the boundary defined by the experimental error).3. in all cases (i. the crystallization temperature experiences a drop of about 5‐6 oC.10.3. 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. the modulus is higher and hardly a function of the compatibilizer amount. at all PCL‐g‐GMA contents) there is a clear increase of the modulus with respect to the 125 .
9. partial interfacial adhesion) which render these ternary blends a kind of “intermediate” case between the binary ones and those compatibilized with PCL‐g‐DEM.7 % for PCL‐g‐DEM. As given schematically in Figure 6. 29]. c. 28. 126 . PCL‐g‐DEM is at least as active as compatibilizer (compare modulus values in Table 6.) than PCL‐g‐GMA for blends containing 20 %‐wt of starch.12.). As a result.4. a. However. Thus. 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. This is surprising if one takes into account the different FD values: 9. Figure 6. Morphology of ternary blends S/PCL/PCL‐g‐GMA.11.8 % for PCL‐g‐GMA against 1. 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. presence of voids.11. 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. and morphologies in Figures 6. b.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. and 6.Chapter 6 binary blend. These differences in compatibilizing effect are not yet fully understood. This confines all reactive GMA groups in relatively “isolated” spots along the PCL backbone. (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. (left) the structure of PCL‐g‐GMA is inhomogeneous at the molecular level with long poly(GMA) branches pending from the PCL backbone.
at higher starch contents more –OH groups would be available for reaction with the GMA groups.12. other groups will presumably remain available (arrows in Figure 6. 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. but since DEM is preferentially grafted as monomer.12. thus attenuating the effect discussed above.).) PCL‐g‐DEM has much less reactive groups (lower FD) than PCL‐g‐GMA.12. 20 and 30 %‐wt respectively). In order to check this we compared blends with the same amount of PCL‐g‐GMA (2 %‐wt) but with different starch intakes (10. Moreover.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”.8 MPa for S/PCL 30/70 up to 430.4. 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. when comparing ternary blends (compatibilized with PCL‐g‐ GMA) with the binary ones.) and relating the “coverage” of the starch particle surface upon reaction with PCL‐g‐GMA with the compatibilization efficiency. once PCL‐g‐DEM has reacted with one –OH groups on the starch particle. Indeed.). Concerning the thermal behavior (Table 6. Schematic representation of the in situ reaction between functionalized PCL and starch. The latter result (from a modulus of 341. As consequence.4.) for further reaction. for all starch amounts (Table 6. the Tc decreases (with respect to the corresponding binary blends) while all other factors (Tm and enthalpies) remain virtually unchanged with respect to the binary blends and also as function of the starch intake. On the other hand (right of Figure 6.12. thus probably ensuring a better coverage of the surface. 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. We can therefore conclude that the efficiency of PCL‐g‐GMA in the 127 . This is in agreement with the hypothesis made above (Figure 6.
4. The most important variable for the final product FD is the interaction between the amount of monomer and initiator used. From a more practical point of view it can be concluded that the newly synthesized PCL‐g‐DEM. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). mutual solubility of the monomer into the molten polymer) on the FD adequately described the experimental results (R2=0.Chapter 6 compatibilization of starch/PCL blends can be significantly improved by changing the blend composition. 6. The reactive compatibilizers were used in blends of starch with PCL. firstly reported in this work. P‐value ≤10‐9). Conclusions A systematic study. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. The PCL‐g‐GMA and PCL‐g‐DEM compatibilizers display lower melting temperatures and melting enthalpies compared to virgin PCL. including statistical modeling. has been performed on the synthesis of two reactive compatibilisers. 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 . 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. The proposed model to quantify the effects of process variables (monomer and initiator intake. 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. 6. 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.5.957. Nomenclature E FD : initial modulus [MPa] : functionalization degree. 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.
A.S. Ogawa.A. B. E. Effect of amylopectin to amylase ratio in starch.D. Carbohydr. Karlsson. Karlsson: Polymers from Renewable Resources.6. R. N. M. Adv. Sci. Pacsu: Starch Studies: Preparation and properties of starch triesters. . Albertsson: Biodegradable Polymers and Environmental Interaction. 1209‐1217. Eur. CRC Press. Wurzburg). 1998. Ind.B. E.G. M. Jarowenko: Acetylated starch and miscellanous organic esters. 129 . A‐C. Polym.W. Mullen. Eur. Albertsson. Polymer 2003. 1998. .. 157. Boca Raton. F. Sci. 1467‐1475. Polym. Ramsay. Y. J. 34. Eng. 139‐161. C. . 34. B. 1986. 1251‐1253. Inc. . Mani. J. . J. 44. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – IV. 38. Ioannis. Part 2. A‐C. Polym. Polym. . Mani. 1942. Kawasaki: Biodegradable films made from low density polyethylene (LDPE). S. W. Lindblad. Davis: High performance LDPE/ thermoplastic starch blend: a sustainable alternative to pure polyethylene. 2002. S. Rodriguez‐Gonzales. E. Polym. Psomiadou. 227‐242. 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.J. in Modified Starches: Properties and Uses (Ed. 1997. USA. 1477‐1487. 34. wheat starch and soluble starch for food packaging applications. Thermal and morphological properties. Liu.Eng. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – III. . M. Ranucci.Chem. R. 1998. 33. 1517‐1526. H. O. References . Biliaderis.
Cho. J. John. 37. New York. A. Prog. Part A: Polym. USA. 81. . Polymer 2000.. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. USA. P. Montgomery: Design and Analysis of Experiments 5th edition. M. 1991. Inoue: Compatibilizers for melt blending: Premade block copolymers. J. 2001. K. Mani. R. Part A: Polym.E. Stab.Y. Ruggeri. Environ.. . Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. Chim. Nakayama. C. 1996. L. Coleman. Laurienzo. A. M. . M. Kim. Kim. 91‐95. Marrucci. Polym. R. . Tang: Functionalization of polyesters with maleic anhydride by reactive extrusion. . J.K. 42. 41. 127. Polym. 1997. C. Tang.S. 61‐67. J. Graf. Sci.S. C. 5135‐ 5142. 1273‐1335.M. R. C. 1507‐ 1516. 29. 1999. C. 127–134. E. K. Passaglia. 41.W. Chiellini.H. Environ. . Sci. Solaro: Biodegradable polymeric materials. Technomic Publication. Chandra. . Sci. 23. Bhattacharya. Yang. . T. M. Polym. Moro. Dole. . Painter: Specific Interactions and the Miscibility of Polymer Blends. Pennsylvania. 35. Avella. Polymer 2001. P. Mater. 80. 1995. 2003. 4157‐4167.K.H. Fringant: Properties of thermoplastic blends: starch‐polycaprolactone. Gazz. D. Khandpur. Marechal. John Wiley & Sons Inc. 3. It. Guegan. D. M. Polym. K. J.H. E. P. J. 3875‐3881. Raimo. 5590‐5598. Macromolecules 1996. . M. Goldberg: A review of the biodegradability and utility of poly(caprolactone).C. 130 . J. Errico. E.Chapter 6 . Z. 2001. 179‐187. 1997. C. Macosko. Kim. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. J. 1139‐1148. M. .K. Park: Grafting of glycidyl methacrylate onto polycaprolactone: preparation and characterization. R. P. C. Rustgi: Biodegradable polymers. 12. Appl. Rimedio: Preparation and characterization of compatibilised polycaprolactone/ starch composites. Kim. Degrad.. Degrad. Jung. Adv.K. J. J. Polym. 2004. 8. R.F. . Averous. . M. Aglietto: Chemical reactions affecting the free‐radical grafting of diethyl maleate onto ethylene polymers.Y. G. Chem. Polymer 2000. L. J. 1693‐1702. Martuscelli. Cho. J. Park: Modification of Aliphatic Polyesters and Their Reactive Blends with Starch. Polym. J.M. 1998. . Sci. Polym. 305‐313. P. Chem.
Polym.M. B. S. U. B. . Ishiaku. R. A. Picchioni: Formation and compatibilizing effect of the grafted copolymer in the reactive blending of 2‐ diethylsuccinate containing polyolefins with poly‐ε‐caprolactam (nylon‐6). R. H. 11‐15. Kammer: On mechanical properties of sago starch/ poly(ε‐caprolactone) composites.A. Ishiaku. 1298‐1305. Manan.N. Odusanya.M. O.M. 87. 57. Ruggeri. 2000. 9. U. 131 . Azemi. Mauler: Comparative study of PP‐MA and PP‐GMA as compatibilizing agent on polypropylene/nylon 6 blends. Polym. 2003. 2002.A. Polym. Ishak: Mechanical properties and enzymic degradation of thermoplastic and granular sago starch filled poly(ε‐ caprolactone). G. K. Odusanya. Passaglia. Adv. . W. Polym. Eng. Nachtigall. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends .Safa. Aglietto. .V. M. Appl.S. Azemi: Effect of starch predrying on the mechanical properties of starch/ poly(ε‐caprolactone) composites.S.D. 21. Test. K. 2006. D.M. 393‐401.B. 877‐884.M.S.S. Polym. S. 1998. 2002. Manan. 38.W. U. Sci.S.W. Pang. Sci.Nasirtabrizi: Ring opening reactions of glycidyl methacrylate copolymers to introduce bulky organosilicon side chain substituents. Barbosa. 293‐304. Eur. J. . J. F. D.S. Lee. M. O. E.N. M. Technol. Polym. Ishiaku. 40. Tedesco. . Bull.H. 273‐281.
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. The application of a number of prospective methods for developing starch‐based biomaterials will be described in this thesis. The content of amylose and amylopectine in starch varies and largely depends on the starch source. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. Starch must therefore be modified before it can be applied as a biodegradable plastic. such as thermoplasticization.6)‐glucosidic bonds.o oil and gas) are very attractive materials for a broad range of applications. For instance. Examples are the use as packaging and construction materials. Starch‐based biodegradable materials are considered interesting candidates to replace certain types of conventional plastics. with a rate of almost 10% every year. there is a strong need for renewable alternatives for plastics from fossil resources. chemical modification or combinations thereof. plastics may end up in the environment and cause serious environmental problems. After use.4)‐glucosidic bonds. The application of an alternative method involving hydrophobisation of starch by the introduction of large hydrophobic SiMe3 groups followed by a ROP with a polyester precursor and subsequent removal of the SiMe3 groups is expected to result in higher grafting efficiencies. Chemical grafting of biodegradable polyesters on the starch backbone is expected to result in less hydrophilic and thus less water sensitive materials with improved mechanical properties. while amylopectin is branched polymer. it has a high volume to weight ratio and is generally resistant to degradation. blending with other materials. In combination with the current high prices for petrochemical products. tensile and flexural strength). 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. There are several starch modification methods available. Its production has increased significantly since 1950. There are two types of polymers present in starch: amylose and amylopectin. Starch is also too sensitive to water. we have . To gain insight in the potential of this approach. Starch is a polymer consisting of anhydroglucose (AHG) units.Summary Plastics made from fossil resources (a. containing periodic branches linked with the backbones through α‐D‐(1. Starch is relatively cheap and available from a broad range of plants. The use of virgin starch for packaging materials is restricted because it cannot be shaped to films with adequate mechanical properties (high percentage elongation.
2. The DS of the PCL chains was between 0.3.72. but the products are still very brittle.3. 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 results are described in detail in Chapter 3.2. The polymerizations were performed at 60‐100°C. Esterification of starch with carboxylic acid derivatives is one of the oldest strategies to improve starch properties.68) was obtained. 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. The hydrophobicity of starch acetates is higher than virgin starch. Poly‐(ε)‐ caprolactone grafted silylated starch co‐polymers with average chain length of 40‐ 55 monomer units (molecular weight of 4500‐6300) were obtained. even in the presence of plasticizers.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose ratio (at constant Al(OiPr)3 intake). and subsequent removal of the silyl groups.21‐0. Considerable amounts of ε‐CL homopolymers with isopropyl end‐groups were also formed. Silylated starch with a low to medium DS (0. The yield of the polymers was a function of the reaction temperature and the reaction time. Average molecular weights of the products were between 970 and 6200 (7 – 58 monomer units) and were a clear function of the p‐ dioxanone/ 1. The results for the model system involving the ROP of p‐dioxanone initiated by hydroxyl groups of a protected monosaccharide (1.Summary initially performed research on the ROP with a simpler model system involving a simple protected mono‐saccharide instead of starch. are described in Chapter 2. The use of higher carboxylic (fatty) acid to esterify starch resulted in products with significantly improved 134 . and off‐white solid products with isolated yields of 30‐96% were obtained. The grafting efficiency for the desired reaction was 28‐58%. The synthetic procedure may be divided into three steps: hydrophobisation of starch by the introduction of SiMe3 groups followed by in situ ROP. temperature and monomer: monosaccharide ratio) on the average degree of polymerization. A statistical model has been developed to quantify the effects of process variables (time.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose) using Al(OiPr)3 as the catalyst.46‐0. Most of the previous starch esterification studies involved the use of short chain carboxylic acids (C1‐C4). with higher temperatures (100°C) leading to lower yields. depending on the ε‐CL to starch ratio. and particularly acetic acid derivatives (C2) have received considerable attention. with higher ratios leading to higher molecular weights. The silylation reaction was performed using hexamethyl disilazane (HMDS) in DMSO/ toluene mixtures at 70°C. 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.
1‐2. products with a DS > 2. The thermal stability of the esterified products is higher than that of native starch. and Na‐acetate) in DMSO at 110°C. reaction temperature. The products have a broad range of degree of substitution (DS = 0. 0.3‐1. The synthesis of these fatty starch esters is. Mechanical tests show that the products have tensile strength (stress at break) between 2. 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.9 for starch laurate. K2CO3. the DS decreased. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. catalyst intake. In Chapter 4‐5. and modulus of elasticity of 46‐113 MPa.4 could be obtained for both laurate and stearate esters. K2CO3. In Chapter 5. 13C‐NMR and FT‐IR. usually performed using fatty acid chlorides which are relatively expensive and rather corrosive.24 ‐ 2. while the use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2. 1. and the type of catalyst. PCL‐g‐ 135 . Reaction temperature and catalyst intake also affect the product DS but only to a lesser extent.45‐1. In Chapter 6.5 MPa. presumably due to a reduction of the polarity of the reaction medium.39) are totally amorphous whereas the low‐DS ones (DS= 1. High‐DS products (DS= 2. however. the synthesis of two polycaprolactone (PCL)‐based reactive interfacial agents. The last part of this thesis deals with reactive blending of starch and biodegradable polymers using reactive interfacial agents. and catalyst basicity) to the DS of the starch ester products were developed. The use of Na2HPO4 resulted in low‐medium DS products (0. an alternative method for higher fatty starch esters by using fatty vinyl esters is reported. Statistically adequate (R2≥0.7‐3.5 for starch stearate).4‐3 for starch stearate). At higher ratios. Chapter 4 describes the results of a preliminary study on the synthesis of long fatty esters of corn starch (starch‐laurate and starch‐stearate) using vinyl fatty esters.07‐1.26‐2. elongation at break of 3‐26%. The thermal and mechanical properties of some representative product samples are also described. The DS of the products was a clear function of the chain length of the fatty ester.5 for starch laurate. 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. With these catalysts. an increase in the vinyl‐ester concentration led to higher product DS values. The DS of the products is a strong function of the basicity of the catalyst.75) are still partially crystalline. The starch esters were prepared by reacting starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. and Na‐ acetate).96). a systematic study. Summary mechanical properties and hydrophobicities. vinyl ester to starch ratio. For low vinyl‐ester to starch ratios. The use of methyl and glyceryl esters results in products with only relatively low DS. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4.
This is in contrast with chemical reactivity and amount of chemical groups along the PCL backbone (both factors favorable to GMA as compared to DEM) but it is explainable on the basis of the group distribution along the PCL backbone. and their use in compatibilising starch‐PCL blends is described. A statistically adequate model (R2=0. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends.957. Both of the PCL‐based compatibilisers improve the interfacial adhesion between the starch particles and the PCL matrix in starch/ PCL blends. 136 . at relatively high GMA and BPO intakes. 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. As the result. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). PCL‐g‐DEM seems to have slightly more efficient compatibilising effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values. At a fixed starch content (20 %‐wt). mutual solubility of the monomer into the molten polymer) on the functionalisation degree (FD) of the GMA and DEM units to the PCL backbone. the mechanical behavior of the compatibilised blends is in general different from that of pure PCL and of the corresponding uncompatibilised blends. 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).Summary glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. P‐value ≤10‐9) has been developed to quantify the effects of process variables (monomer and initiator intake.
Biologisch afbreekbare polymeren op basis van zetmeel kunnen mogelijk bepaalde conventionele kunststoffen vervangen. De verhouding amylose en amylopectine in zetmeel varieert en hangt grotendeels af van de plant waaruit het zetmeel gewonnen wordt. Zetmeel is een polymeer bestaande uit anhydroglucose (AHG) eenheden. menging met andere materialen. ε‐caprolactone op de hydroxy‐groepen van zetmeel is helaas niet eenvoudig.6)‐glucoside bindingen met de hoofdketen zijn verbonden. Er zijn verschillende methodes beschikbaar om zetmeel te modificeren. gevolgd door de ROP met een polyester precursor en verwijdering van de SiMe3 groepen . mede omdat het niet mogelijk is om films met goede mechanische eigenschappen te maken. 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. heeft geleid tot een grote interesse voor biologisch afbreekbare alternatieven gemaakt uit hernieuwbare grondstoffen. Zetmeel is relatief goedkoop en te winnen uit een breed scala aan planten.v. Amylose is een lineair polymeer waarin de AHG eenheden hoofdzakelijk via α‐D‐(1. De AHG eenheden kunnen op een aantal manieren aan elkaar gekoppeld worden. Dit.Samenvatting Kunststoffen gemaakt van fossiele grondstoffen als olie en gas zijn zeer aantrekkelijke materialen voor een breed scala aan toepassingen. De synthese van dergelijke producten met een in situ katalytische ringopening polymerisatie (ROP) van geschikte monomeren als b. 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. De toepassing van een alternatieve methode waarbij zetmeel eerst minder hydrofiel gemaakt wordt door de introductie van grote hydrofobe SiMe3 groepen. Voorbeelden zijn thermo‐ plastificatie. De productie van kunststoffen is sinds 1950 met bijna 10% per jaar gestegen. Na gebruik kunnen de materialen in het milieu eindigen en grote problemen veroorzaken. Dit proefschrift beschrijft een aantal methodes om nieuwe polymere materialen op basis van zetmeel te maken. Het amylopectine is een vertakt polymeer waarbij de vertakkingen via α‐D‐(1. in combinatie met de huidige hoge prijzen voor de grondstoffen. Het zetmeel is daarnaast ook erg gevoelig voor water. Modificatie van zetmeel kan het aantal toepassingen sterk vergroten. chemische modificaties en combinaties van deze methoden.4)‐glucoside bindingen verbonden zijn. Twee type polymeren zijn dominant aanwezig in zetmeel: amylose en amylopectine. Het gebruik van zetmeel voor verpakking materialen is momenteel beperkt.
De DS van de PCL ketens lag tussen de 0.3. De resultaten worden in Hoofdstuk 3 in detail beschreven.Samenvatting zou in theorie moeten leiden tot een hogere grafting efficiency. De polycaprolacton eenheden (PCL) van het verkregen poly‐ε‐caprolactone gesilyleerde zetmeel hadden een gemiddelde ketenlengte van 40‐55 monomeereenheden (molecuul massa 4500‐6300). De grafting efficiency van de gewenste reactie was 28‐58%.72 en was afhankelijk van de ε‐caprolacton/zetmeel verhouding. gevolgd door in situ ROP met ε‐ caprolacton en daarna de verwijdering van de silyl groepen. De gemiddelde molecuul massa van de producten lag tussen de 970 en 6200 (7 ‐ 58 monomeereenheden) en was een functie van de verhouding p‐ dioxanone/1. DS = 0.68). 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 substitutiegraad van de SiMe3 groepen varieerde van laag tot middelhoog (degree of substitution. waarin hogere temperaturen (100°C) tot lagere opbrengsten leiden.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) met Al(OiPr)3 als katalysator worden in Hoofdstuk 2 beschreven.46‐0. De gevolgde procedure bestaat uit drie stappen: hydrofobisering van het zetmeel door de introductie van SiMe3 groepen. De grafting van ε‐ caprolactone aan het gesilyleerde zetmeel werd in THF bij 50°C uitgevoerd met Al(OiPr)3 als katalysator. Er werden ook aanzienlijke hoeveelheden homopolymeer van ε‐ caprolacton met isopropyl eindgroepen gevormd. De resultaten voor het modelsysteem waar de ROP van p‐dioxanone wordt geïnitieerd door de hydroxy‐groepen van een beschermde mono‐saccharide (1.2. De effecten van procesvariabelen (tijd. 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.21‐0. 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%).4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (bij constante i Al(O Pr)3 concentratie). 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. temperatuur en monomeer: monosaccharide verhouding) op de gemiddelde polymerisatie graad zijn gekwantificeerd met een statistisch model. De kennis verkregen met het modelsysteem is toegepast om ε‐caprolactone te graften op zetmeel. Een hogere verhouding leidde tot hogere molecuul massa’s.3.2. 138 . De opbrengst van de polymeren was een functie van de reactietemperatuur en de reactietijd.
Echter de producten zijn nog steeds zeer bros. naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat. In Hoofdstuk 5. Het gebruik van hogere (vet‐) zuren resulteerde in producten met beduidend betere mechanische eigenschappen en hydrofobiciteit. K2CO3 en Na‐acetaat) in DMSO bij 110°C te laten reageren. Bij lage vinyl‐ester/zetmeel verhoudingen leidde een verhoging van de vinyl‐ester concentratie in eerste instantie tot hogere DS waarden. Echter. reactietemperatuur.96. wordt een alternatieve methode voor de synthese van zetmeel esters met lange vetzuren beschreven.07‐1. 0.5 voor zetmeel laureaat. Het gebruik van methyl en glyceryl esters resulteert in producten met een lage substitutiegraad.24 tot 2. Met deze katalysatoren worden voor zowel de laureaat als stearaat esters producten met een DS van boven de 2. terwijl het gebruik van de K2CO3 en Na‐acetaat katalysatoren 139 . De producten werden met 1H‐. De DS van de producten was een duidelijke functie van de keten lengte van de vetzure vinylester. zelfs in aanwezigheid van plastificeermiddelen. en in het bijzonder azijnzuur derivaten (C2). K2CO3. bij verdere verhoging van de verhouding verminderde de DS. de vinyl‐ ester/zetmeel verhouding en het type katalysator. 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.96 en P‐ waarde van ≤ 10‐7). en Na‐acetaat) te laten reageren. De zetmeel esters werden bereid door het zetmeel met vinyl laureaat of vinyl stearaat in aanwezigheid van basische katalysatoren (Na2HPO4. 13C‐NMR en FT‐IR gekarakteriseerd. De acetaat esters zijn in het algemeen minder hydrofiel dan zetmeel. De temperatuur en de katalysator concentratie beïnvloeden de DS ook maar in mindere mate. De zetmeel esters werden bereid door zetmeel met vinyl laureaat of vinyl stearate in DMSO in de aanwezigheid van basische katalysatoren (Na2HPO4.5 voor zetmeel stearaat). vermoedelijk als gevolg van een verlaging van de polariteit van het reactiemengsel. De synthese van deze esters wordt in het algemeen uitgevoerd met vetzuurchloriden die vrij duur en sterk corrosief zijn. inclusief statistische modellering. katalysator concentratie en katalysator basiciteit) op de DS zijn gekwantificeerd met een statistisch model (R2 ≥ 0. De thermische en mechanische eigenschappen van sommige representatieve producten zijn uitgebreid bepaald. In Hoofdstuk 4‐5.3‐1. De substitutiegraad varieert van 0. Het gebruik van Na2HPO4 resulteerde in producten met een laag tot middelhoge DS (0. De DS van de producten is een sterke functie van de basiciteit van de katalysator. wordt een systematische studie. In de meeste zetmeel esterificatie studies wordt gebruikt gemaakt van kleine carbonzuren (C1‐C4). K2CO3 en Na‐acetaat vertonen superieure activiteiten vergeleken met Na2HPO4. Samenvatting De esterificatie van zetmeel met carbonzuur derivaten is één van de oudste strategieën voor het verbeteren van zetmeeleigenschappen. De effecten van belangrijke procesvariabelen (vinyl ester/zetmeel molverhouding.4 verkregen.
9 voor zetmeel laureaat. Dit leidt tot verschillen in mechanische eigenschappen tussen gecompatibiliseerde mengsels en van de niet‐ gecompatibiliseerde mengsels van zetmeel en PCL. Het laatste hoofdstuk van dit proefschrift (Hoofdstuk 6) behandelt de menging van zetmeel en biologisch afbreekbare polymeren met reactieve compatibilisers.0 voor zetmeel stearaat).4‐3.45‐1. PCL‐g‐glycidyl methacrylaat (PCL‐GMA) en PCL‐g‐diethyl maleaat (PCL‐g‐DEM). 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. De thermische stabiliteit van de geësterificeerde producten is hoger dan die van zetmeel. Zo is bijvoorbeeld de elastische modulus van de gecompatibiliseerde mengsels beduidend hoger (tot 430 MPa) dan die van niet‐gecompatibiliseerde mengsels (320‐340 MPa).39) zijn volledig amorf terwijl de producten met een DS van 1.957 en P‐ waarde van ≤ 10‐9) ontwikkeld waarmee de effecten van procesvariabelen (monomeer en initiator concentratie. gebruikt.7‐3. De hoogste FD waarden (tot 45%) werden gevonden voor PCL‐g‐GMA bij relatief hoge GMA en BPO concentraties. De producten met een hoge DS (2. Er is een statistisch model (R2 =0.75 nog gedeeltelijk kristallijn zijn.5 MPa.26‐2. 1. De synthese van deze verbindingen wordt uitgebreid beschreven. De FD waarden voor PCL‐g‐DEM waren aanzienlijk lager (tot 7%). Als reactieve compatibilisers zijn twee polycaprolactone derivaten. 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. 140 . Beide reactieve compatibilisers verbeteren de interactie tussen de zetmeeldeeltjes en de PCL matrix. een uitrekking tot breuk van 3‐26%.Samenvatting resulteerde in producten met een middelhoog tot hoge DS waarden (2.1‐2. en een elasticiteit modulus van 46‐113 MPa. Mechanische tests tonen aan dat de producten een treksterkte hebben variërend van 2.
This thesis is a summary of my research on developing novel biodegradable polymers from starch. Leon Janssen. I want to thank my three promoters. Mosiscki. for his efforts to make every foreign students feel at home in Groningen. for upgrading my writing quality. since the mechanical properties as well as hydrophobicity of this starting material is relatively poor. Prof. First of all. the raw material for fossil derived plastics. Prof. Prof. who was my first daily supervisor for about two months.J. especially on his opinions about the statistical modelling and mechanical properties. Anno was also always there when . for the inspiring ideas during the discussions. for your consistent support. together with difficulties in managing plastic waste. These include the current high oil prices. Gert Alberda van Ekenstein and Harry Nijland helped me a lot with mechanical and thermal property measurements. and Laurens Bosgra for being always helpful with technical issues and Jan Henk Marsman for the help with analytical equipment. L. A. Erwin Wilbers. Erik Heeres. Marcel de Vries. Thank you Erik. which improved the quality of this thesis. I also want to give my deep appreciation to my reading committee. and for helping me arranging many things when I was not able to do it from Indonesia. Broekhuis. Prof.Acknowledgements The synthesis of biodegradable materials is currently receiving a lot of attention. Anno Wagenaar and Wim Kruizinga taught me how to operate the NMR machines and to interpret the spectra. Minnaard. I found out that developing a better fully‐biodegradable‐material from starch is not an easy job. and for always encouraging me to finish this thesis. Although it is indeed a very interesting topic. People from other groups have also been very kind and supportive to my research. I also want to thank Ineke Ganzeveld. not in the last place because of the problems related with conventional plastics. I want to thank Leon Janssen for his hospitality. and Prof. for her hospitality and solving administrative matters. 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. Francesco Picchioni. A. Anne Appeldoorn. I want to thank Francesco Picchioni for being always there when I needed him to discuss anything. for valuable discussions during my early PhD period and for handing me over to the right hands. I want to thank Marya van der Duin‐de Jonge. I want to thank Erik Heeres for being more than a good supervisor.A. and Prof. as well as taking the SEM pictures. 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.
Rasrendra. and helpful. Diana Jirjis. Tiara.Acknowledgements I had questions or needed specific equipments. Thyara. Wangsa. C. Togi. Anna Nizniowska. Chalid. Francesca Fallani. Jaap Bosma. spring and autumn in Groningen. Francesca Gambardella. Buana Girisuta and Rina Karina. mas Pandu. August Kurniawan. master students who worked with me. Albert Kieviet helped me with the MALDI‐TOF measurements. Wuri Raspati. Thank you AW. Nadia Gozali.B. in the lab. Farchad H. Oscar Rojas. Wahyudin. Agnes R. Iwan Harsono. My special thanks are for Laura Junistia and Jan‐Pieter Drijfhout. Annette Korringa. and the ISD staff helped me a lot with visum and residence permit arrangements as well as tax papers. and for my wife and family. Gerard Kraai. kak Atha. Tenny. The chemical and glassware magazijn as well as the FWN library personnel were always patient. Zhang Youchun. You all have always been my inspiration and motivation. and Mahesa for being my brothers and sisters in Groningen. Egi. Thank you God. Tante Smith. and Iwan Kustiawan for becoming my family. Teguh. Samuel. kak Roga. Thank you Tresna. Boelo Schuur. and assistant professors have shared the days with me in Groningen. I have met a lot of Indonesian students in Groningen. and on many occassions. 142 . for never leaving me alone. I want to thank Edwin A. Arjan Kloekhorst. mbak Ika. Ignacio Melian Cabrera. A lot of thanks for Henk Stegeman and Poppy Sutanto. Mita. Taufik. patience. Hofman. Taufik. friendly. Many master and PhD students. Yongki. Bima and Wisnu (also my paranymph). Groningen is also a place where I found true friendship. bu Ida Susanti. visiting researchers. mbak Mia. R. Wita Sondari. Thank you my parents and sister. in the office. Vincent Nieborg and Fesia Lestari. Judy Retti Witono. Louis Daniel. Anneke Toxopeus. Ita. Mahfud. Ardiyanti. and Tante Alma for being like my parents and family in Groningen. Tante Oppie. Patrick and Oppie. Our collaboration has also become part of this thesis. for your love. Puspita Kencana. Yao Jie. Marcel Wiegman. Tante Caroline. Nidal Hammoud Hassan. Jasper Huijsman. Jelle Wildschut. M. our memories will never fade away. for everything. Hans Heeres. and passion. and many others for the good old times! Thank you Tante Ietje. Ais Jenie. Kuslan. Diana Santangelo. support. and Henk van de Bovenkamp for cheering my days and being my friends during summer and winter. pak Harry. Asal Hamarneh. Abdul Osman. Endy. Erna Subroto. Thanks also for Marcel and Connie. Henky Muljana (my paranymph) and Anindhita Widyadhana. Dado. Teddy. Ronny Sutarto. Niels van Vegten. Sari Fabianti. for your never‐ending blessings. and although we seldom met afterwards.
8. Kuala Lumpur (Malaysia). Guangzhou (China).A. Sugih. Heeres: Synthesis and properties of novel green biopolymers derived from starch and fatty acids. presented in the 1st International Symposium on Sustainable Chemical Product and Process Engineering. Eur. 9. A. F.K. A. F.J. South China University of Technology. L. Heeres: Green.1016/j. Sugih. India. Broekhuis.eurpolymj. H. J. F. 26‐28 November 2006 (oral presentation). presented in “Polymers for Advanced Technologies”. H.P.K. H. A.Heeres.B. Bratislava (Slovak Republic). 9th National Conference organized by the Society for Polymer Science.B.Picchioni.K. Sugih.J.). Polym. Broekhuis.J.P. Manurung. Pune Chapter. and F. 2008. 4.010).M. L. F. Picchioni. Synthesis and properties of reactive interfacial agents for polycaprolactone‐starch blends. F. Janssen. presented in International Workshop: “From Polymer Modification to Multicomponent Polymer Systems”. F. Appl. 5. Sugih.K. A. Junistia.M.B. Heeres.J. and F. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Picchioni.K.P. R. and H. Sugih. Janssen.J. Drijfhout. A.K. A. Picchioni. Manurung. Experimental and modeling studies on the synthesis and properties of higher fatty esters of corn starch. (in press. Synthesis of poly‐ (ε)‐caprolactone grafted starch co‐polymers by ring‐opening polymerisation using silylated starch precursors. R. A. Heeres. L.Drijfhout.M. 2008 (accepted for publication in Starch‐Starke). Heeres.P. and H.List of Publications 1. Sugih. 2008 (accepted for publication in Starch‐ Starke). Picchioni. Junistia.J. presented in 11th Asia Pacific Confederation of Chemical Engineering (APCChE) Congress. L. J. 27‐30 August 2006 (oral presentation) 7.Janssen. Drijfhout. A. Picchioni: Functionalization and blending of polycaprolactone. Heeres. F. Picchioni. Picchioni: Biodegradable polymer materials based on polyesters and starch/proteins. Sugih. Heeres. J. 2008 (submitted to J. L.K.10. Junistia. Experimental Studies on the ring opening polymerization of p‐dioxanone using an Al(OiPr)3‐monosaccharide initiator system. starch‐based innovative product: The synthesis of grafted polycaprolactone‐starch by ring opening polymerisation. Sci. L. H. 2. J. Sugih. 3. .P. Polym.J. Mumbai (India).J. A.P. 17‐20 December 2006 (oral presentation). Polym. Picchioni.J. 2008 (submitted to Carbohydr. 25‐28 September 2007 (oral presentation). L. Sugih.2008.K. H. H.B.K. A.) 6.A. A. DOI: 10.P. Junistia.M Janssen and H. L.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.