This action might not be possible to undo. Are you sure you want to continue?
Asaf Kleopas Sugih
The author thanks the University of Groningen for the financial support through an Ubbo Emmius Scholarship.
Synthesis and Properties of Starch Based Biomaterials
Proefschrift ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. F. Zwarts in het openbaar te verdedigen op vrijdag 12 december 2008 om 13.15 uur door Asaf Kleopas Sugih geboren op 4 juli 1975 te Bandung, Indonesië
H. dr. Minnaard Prof. dr.J.B. ir. ir. Moscicki ISBN : 978‐90‐367‐3592‐6 ISBN : 978‐90‐367‐3593‐3 (electronic version) . dr. Broekhuis Prof.A. Janssen Prof. dr.M. Heeres Prof. dr. dr. Promotores : Prof. A.J. Picchioni Beoordelingscommissie : Prof. L. A. F. L.P.
to: Tresna. my parents. and my sister .
184.108.40.206.2.2. Plastic applications and waste issues 220.127.116.11.3. Starch‐based blends by melt mixing 1.2.2. Introduction 2.3.1. Materials and Methods 2. Starch 1. Structure and properties of starch 1. Starch‐based blends by in situ polymerization 18.104.22.168.2.1. Biomaterials from starch 1. Methods 22.214.171.124. Starch modifications to improve product properties 1. Starch ‐ biopolymer blends and graft co‐polymers 1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (2) 27 28 26 27 27 27 2. Cross‐linked starch 1. Biodegradable plastics from starch 1.2. The potential of biodegradable plastics 1.3.4. Thermoplasticized starch 1.2.1.Table of Contents Chapter 1: Introduction 1. Materials 126.96.36.199. Starch Esters 1. Thesis Outline 188.8.131.52.184.108.40.206. Starch production processes 220.127.116.11.1. Product analyses .3. 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. Typical example for the synthesis of polydioxanone end‐capped with 1.
1. Typical example of the starch silylation procedure 18.104.22.168.2.1.4. 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 22.214.171.124.2. Typical example of in situ polymerization of ε‐CL with silylated starch 50 3. Materials 126.96.36.199.6. Product yield 188.8.131.52.184.108.40.206. Introduction 220.127.116.11.2.2.3. References Chapter 3: Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers by Ring Opening Polymerisation using Silylated Starch Precursors 3.3. Conclusions 18.104.22.168. Materials and Methods 3. Mechanistic aspects 2. Calculation of average degree of polymerization 2.1. Results and Discussions 2.4. Product analyses 2. Methods 3. Analytical methods 3.2.1. Results and Discussions 3. Calculations 3.2.3. Systematic studies 2.3. Screening experiments 22.214.171.124.2. Nuclear Magnetic Resonance (NMR) 3.3. De‐silylation of poly‐(ε)‐caprolactone grafted silylated starch co‐polymers 51 51 52 52 52 53 54 48 49 49 50 50 3. Nomenclature 126.96.36.199.2. Peracetylation of silylated starch 3.2. Synthesis of silylated starch viii .1. 2.
2. Nomenclature 3. Materials 4.2. In situ ring opening polymerization of ε‐caprolactone with silylated starch 57 63 64 64 65 3. Effect of vinyl ester to AHG ratio on the product DS 4. 3.3.5. Effect of the addition of toluene as a co‐solvent 188.8.131.52. Analytical equipment 4. Methods 4. Systematic studies 4. Results and Discussion 184.108.40.206. Introduction 220.127.116.11.18.104.22.168.22.214.171.124.3.3. FT‐IR measurements 126.96.36.199.188.8.131.52. References Chapter 4: Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants 4.3. References ix .184.108.40.206.3.1.1. 1H‐ and 13C‐NMR analyses 220.127.116.11. Conclusions 3.3. Exploratory experiments 4. Nomenclature 4.2. Product characterisation 4.2. Catalysts screening 4.2. Materials and Methods 4. Determination of the Degree of Substitution (DS) 4.3. Deprotection of silylated‐starch‐g‐PCL 3.5. 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. Conclusions 4.6.2. Peracetylation procedure 4.
Preparation of PCL‐starch blends with the reactive compatibilisers 110 18.104.22.168. Typical example of the preparation of laurate and stearate starch Esters 88 86 87 87 87 88 5. Conclusions 5. Methods 6. Compatibilizer synthesis 6. Analytical Methods 110 x .22.214.171.124.126.96.36.199. Nomenclature 5.4.2. Mathematical modelling 5. Work‐up of PCL‐g‐DEM 108 109 109 109 109 110 110 6. Materials and Methods 5.2. Introduction 6.3.2. Methods 188.8.131.52. Materials 6.1. Materials and Methods 6. Experimental Design 184.108.40.206.2.2. References Chapter 6: Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends 6.2. Product properties 5. Peracetylation procedure and Degree of Substitution (DS) determination 88 88 89 92 96 102 102 103 5.1.2. Work‐up of PCL‐g‐GMA products 220.127.116.11.18.104.22.168.2. Analytical Equipment 5. Materials 5. Results and Discussion 5.1.Chapter 5: Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch 22.214.171.124.4. Introduction 126.96.36.199.
3.3.2. Statistical mdeling 6. Binary blends of starch and PCL 188.8.131.52. Synthesis and Properties of Starch‐ PCL Blends 6. Effect of substrate (GMA/DEM) to PCL Ratio on the FD 6. Modeling of the combined effects of the GMA/ DEM and BPO intakes on the FD 111 111 111 116 117 118 120 121 122 123 125 128 128 129 6.4. Ternary blends compatibilized with PCL‐g‐DEM 6. Preparation of the ractive compatibilizers 184.108.40.206. Results and Discussions 6.4.1. Nomenclature 220.127.116.11. 18.104.22.168.3. Ternary blends compatibilized with PCL‐g‐GMA 22.214.171.124.3. References Summary Samenvatting (Dutch Summary) Acknowledgements List of Publications 133 137 141 143 xi .2.2. Thermal properties of the compatibilizers 6. Conclusions 126.96.36.199.3. Calculation of the Degree of Functionalization (FD) of the reactive compatibilizers 6. Effect of the BPO intake on the product FD 188.8.131.52.1.1.
Different routes to modify starch to improve the product properties and to extend the application range will be provided. Finally. an outline of this thesis is given. .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.
but other types of starch such as cassava. sweet potato. sorghum. Past. potato or cassava) of the plants [2‐3]. present and a forecast of starch production  o : Europe □ : USA : Total ∆ : the Rest of the World 2 .e. with an average yield of 4.1. 80 70 60 50 40 30 20 10 0 Starch Production (Million Tons) 1995 2000 Year 2005 2010 Figure 1. The worldwide production of starch in 2008 is estimated to be around 66 million tons . Starch 75% of all organic material on earth is present in the form of polysaccharides. An important polysaccharide is starch.e. or peas) and in tubers or roots (i. corn. and wheat starch are also produced in large amounts [2. rice. Most starch crops are very productive.45 (estimated from the total volume and value of the US corn starch export ). . wheat. It is generally deposited in the form of small granules or cells with diameters between 1‐100 µm . Plants synthesise and store starch in their structure as an energy reserve. potato.Chapter 1 1. The current price of corn starch is around $0. Most of the starch produced worldwide is derived from corn. The second and third starch producer regions are Europe and Asia [4‐5].1. Past. Most of the starch is produced in the USA. maize.1. Starch is found in seeds (i. barley.9 ton starch per hectare . . while corn seeds consist of 65‐80% starch by weight. current and future estimations of worldwide starch production are given in Figure 1. Potato accumulates starch to approximately 75 % of the dry weight in the tubers with a yield up to 21 ton starch per hectare. 4‐5].
Typical corn‐milling operation b.1. which is again fed to a second milling process. b. The resulting starch is further washed in a cyclone and finally dried. Starch production processes Starch is generally extracted from the plant by wet milling processes [9‐10]. Corn starch production  a. the debris is filtered from the slurry.2. After a coarse milling. sand. The starch is present in the endosperms (floury and horny).b. the corn kernel is softened by steeping in warm water containing SO2 until the volume of the kernel increases with 55‐65%. After initial cleaning to remove cob. Figure 1. Introduction 1. and starch granules are obtained after centrifugation from the suspension. Cross‐sectional view of a corn kernel 3 .2. The fiber is removed using washing screens. and other foreign materials. a.1.a.2. and is embedded in a proteinaceous cellular matrix as is shown in Figure 1. a typical corn starch production process is given in Figure 1. the mixture is fed to a hydrocyclone to separate the germ and the rest of the kernel. The resulting suspension from the mills contains fiber. gluten. The low density gluten is separated from the starch suspension by centrifugation. The plant material is grounded in water. As an example. and starch.
3. H OH H O H OH H O H H OH H c. Structure and properties of starch Starch is a polymer consisting of anhydroglucose (AHG) units (see Figure 1. . H O H OH H OH H O H OH O H OH H O . H2COH H OH H O H H O b. Anhydroglucose (AHG) unit b. Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1.. The amylose content of several common starches is given in Table 1....4)‐glucosidic bonds. H2COH H H OH H O H OH O H OH O CH2 H O H H OH H O H OH H O .2. H OH H2COH H2COH H .. Amylopectin is a branched polymer. The linear 4 . Amylose c. the amylose content is between 18‐28% .1. Figure 1. 6].) .. H2COH H . .. The content of amylose and amylopectine in starch varies and largely depends on the starch source. The starch granules will swell rapidly to many times of its original volume. Amylopectin Starch is insoluble in cold water. but it is very hygroscopic and binds water reversibly..1..6‐7 x 105 Da . Each branch contains about 20‐30 anhydroglucose units. H OH H H2COH H2COH H O H H OH H H2COH H O H H OH H O H OH H O .Chapter 1 1. Two types of AHG polymers are usually present in starch: amylose and amylopectin [2‐3. Heating a starch solution leads to loss of hydrogen bonding in the interior of the starch granule and the starch will start to gelatinize. a... H ... containing periodic branches linked with the backbones through α‐D‐(1. The molecular weight of amylopectin is higher than that of amylose and is typically 4‐5 x 108 Da . Typically.6)‐ glucosidic bonds . but usually in the range of 1..3. Chemical Structure of Starch a.a. The molecular weight of amylose is a function of the plant source and processing method.
and granule fragments.5 28 52 70‐75 27 15. Parker in 1838. Table 1.2. and celluloid was 5 . Biomaterials from starch 1.2. and. swollen granules. 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 . Cellulose nitrate was first prepared by A. Amylose content of common starches  Starch Arrowroot Corn Hybrid amylomaize Class V Hybrid amylomaize Class VII Oat Manioc Potato Rice Sago Sweet potato Tapioca Wheat Amylose (%) 20.2.5‐70 62‐72 63‐72 58‐64 1. The resulting suspension contains a mixture of linear amylose molecules. Plastic Applications and Waste Issues Plastic is the general term for a wide range of synthetic or semisynthetic polymerisation products. Plastics are used in a wide range of applications and the demand is still increasing every year .5 25. Introduction amylose molecules leach out of the granules into the solution. The first generation of commercial plastics was derived from cellulose nitrate and is known as celluloid [12‐14]. depending on the amount of water present.8 16.7 26 Table 1. Starch gelatinization temperature range  Starch Potato Tapioca Corn Waxy corn Wheat Gelatinization Temperature Range [o C] 59‐68 58.7 20 18.2.1. The gelatinization temperature range of various starch sources is given in Table 1. will form a thick paste or gel.1.8 17.
Plastics have lower melting temperatures compared to glass and metals. and consumer products (see Figure 1.Chapter 1 patented by J. the oldest purely synthetic plastic is Bakelite.4. and PolyAcryl. PA-. The largest application of plastics is for packaging purposes. and energy during transportation of goods. Polyurethanes.Fibers 150 1989: 100 100 1976: 50 50 1950: 1.4. when polyethylene (PE) was invented. discovered by Baekeland in 1907 [12. space. Plastics are also used for building materials and automotive.).5 0 1950 1960 1970 1980 Year 1990 2000 2010 Figure 1. and therefore need less energy to shape it into useful materials .5. A dramatic increase in demand for plastics began after World War II. Worldwide Plastics Production (1950‐2006)  6 . the worldwide plastics production has reached 245 million ton per annum  (see Figure 1. The latter is important when using the material for packaging purposes to save weight. Thermosets. electrical. Elastomers. Hyatt in 1870 . They have a low density and can be shaped in thin films while maintaining good properties. About 29% of the total plastics produced in the USA . Sealants. Important polymers used for packaging are polyethylene (HDPE and LDPE). Plastics are very attractive materials.) 250 Included: Thermoplastics. Coatings. polypropylene (PP). 15]. The production and consumption of plastics has increased significantly with a rate of almost 10% every year since 1950. and 37% of the total plastics demand in EU  is used as packaging materials. polystyrene (PS). and polycarbonates (PC). While celluloid is derived from a natural polymer (cellulose). and PP-Fibers 2006: 245 2002: 200 200 Plastics Production (Million Tons) Not Included: PET-. for instance to be used in packaging and paper coatings . polyethylene terephtalate (PET). PE is a very versatile plastic because it can be shaped easily into various forms. In 2006. polyvinyl chloride (PVC). Adhesives.
irrigation. however. Figure 1. It is not applicable for thermoset resins . Recycling (part of) the plastics to reduce the amount of MSW also has limitations. hydropower plants operation. Biodegradable plastics are polymeric materials capable of decomposing when given an 7 .5. plastics waste was ranked as the second major source of MSW after paper and paperboard. 16]. It has a high volume to weight ratio and is resistant to degradation. The disposal of plastics materials in municipal solid waste (MSW) is a serious issue in many parts in the world.2. 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).7% in 2006 . which are usually found in waste streams. Plastic applications and waste treatment in Europe  1. In Europe (2006). threatening fishery.2. In the USA. 6]. and may be more than twice the weight fraction . and lakes. Plastics have been polluting sea [1. Plastic litter is hazardous to wildlife [1. are not easy to recycle . only 20% of the plastic is recycled. The volume fraction of plastic in MSW is much larger due to the low density of plastics. rivers. ship navigation. and other public works . and is only effective for single plastic sources or simple plastic formulations. the amount of plastics in MSW increased from less than 1% in the 1960 to 11. and accumulation of its residues in soil cause significant reductions in agricultural yields. Co‐mingled plastics. Introduction Plastic waste. is causing serious environmental problems. soil. In 1996. consisting of 25%‐v of the total waste . Most of the waste is still disposed (50%) by landfill or incinerated to recover the energy (30%) .
and polyesters from micro‐organisms . 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. biological and/or physical forces in the environment at a rate which leads to fragmentation or disintegration of the plastics. It may also be treated in sewage sludge water treatment plants or buried in the soil [20‐21]. Biodegradable plastics have gained considerable interest since the 1980s. and microorganisms) and biodegradable synthetic polymers . proteins (present abundantly in plants and animals). Japanese Biodegradable Plastic Society draft proposal ‐ Biodegradable plastics are polymeric materials which are changed into lower molecular weight compounds where at least one step in the degradation process is through metabolism in the presence of naturally occurring organisms. For instance.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 . The market price of biodegradable plastics (mainly from starch and PLA) has decreased the last years.5/ 8 . ASTM sub‐committee D20. Biodegradable plastics may be classified into two general groups. biopolymers from nature (from plants. Biopolymers are inherently biodegradable. Several authorities have provided definitions for biodegradable plastics . Nowadays new types of biodegradable plastics with improved properties and lower costs have been developed . Common natural biopolymers are carbohydrates. Biodegradable plastic waste may be treated in composting facilities. the price of PLA based plastics in 2010 is estimated to be around € 1. together with food and yard waste as well as paper. animals.Chapter 1 appropriate environment and sufficient amount of time . The considerable growth of interest in composting as a means to replace landfill due to the decreasing disposal spaces (especially in Europe) may also help the progress of biodegradable plastics development . The change in chemical structure results from the action of naturally occurring micro‐ organisms. DIN 103. and since 2000 they have become competitive with traditional materials in some applications [20‐21].96 proposal ‐ Degradable plastics are plastic materials that undergo bond scission in the backbone of a polymer through chemical. because they take part in nature’s cycle of renewal.
Examples are the use as packaging material for carrier bags. alone or complexed with natural or synthetic biodegradable polymers) or from polylactic acid which originates from the fermentation of a starch feedstock. 9 . Virgin starch is not suitable as a packaging material. biofillers for tires. Biodegradable plastics are also used for food servicewares . either from starch‐based materials (slightly modified starch. because the price of oil based polymers may still increase due to the rise in the crude oil prices . tensile and flexural strength) and is too sensitive to water [26‐27]. utensils .8‐ 2. blending with other materials. 1. 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 .50/ kg in 2005) is also expected to be lower in the future due to considerable reduction in modification costs . Consequently. It cannot be shaped in films with adequate mechanical properties (high percentage elongation.4 or $ 1.4 per kg in 2007‐2008 ). starch must be modified. towels. consumer products and food products. The price of starch in 2007 was about $ 0. Biodegradable plastics are especially very useful for single‐use applications.3. These are summarized briefly in the next sections. agricultural products (mulch films. and chewable items for pets [1. 20]. € 1.50‐4.3. breathable fabrics.2.2‐ 1. napkins) . Introduction kg. Starch may become an attractive raw material for plastics in the future. Biodegradable Plastics from Starch Starch is a very attractive source for the development of biodegradable plastics. Bastioli [20‐21] showed that nearly all biodegradable plastics available in the market are derived from starch. before they can be applied as biodegradable plastics. hygienic materials (diapers. such as polyethylene (PE. chemical modification or combinations of them . or when environmental impacts have to be minimized. 1. pots). Starch modifications to improve product properties Several techniques may be applied to develop starch based biomaterials with improved properties. .45 per kg . half of its price in 2003 (around 3 Euro/kg). when recycling is not practical or uneconomical [20‐21]. This price is much lower than conventional plastics derived from oil. The price of starch‐based biodegradable plastics (€ 1. either by plasticization.
and hexylene glycol. 40]. water content. The use of water as a plasticizer is not preferable. 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 . TPS synthesized from polyol and sugar plasticizers have the tendency to re‐crystallize (retrogradation) after being stored for a period of time. with better properties than virgin starch in various applications. which results in embrittlement. . the thermal properties of glycerol‐plasticized starch are a function of water content [31. destructed. 30. such as water and polyols (glycerol. The brittleness is known to increase in time due to free volume relaxation and retrogradation. Thermoplasticized starch Virgin starch is brittle and difficult to be processed into articles due to its relatively high glass transition temperature (Tg. glycerol. 32.3. The plasticizers are also usually hydrophilic and can be 10 . At intermediate water levels. The modified products are known as thermoplastic. For instance. Although thermoplasticization seems to be a promising method. The use of other plasticizers (for example glycerol) results in a rubbery material. Another issue is the poor water resistance and low strength. the semicrystalline structure of starch and its granular form are lost and the starch polymers are partially depolymerized. plasticized. 35. sorbitol. glycol. and therefore are processable by conventional polymer processing techniques such as injection.Chapter 1 1. Yu. 40]. and blow moulding [28. 32‐ 36]. resulted in the formation of an amorphous mass [35.1. There are several substances used as plasticizer for the preparation of thermoplastic starch (TPS). This problem is mainly caused by the presence of strong inter‐ and intra‐molecular hydrogen bonds between the starch macromolecules [28. 33]. phase separation may still occur. et al  discovered that the elongation of break of the thermoplastic starch is significantly improved by plasticization with glycol. 39]. 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. extrusion. approximately 230 oC) . sugars) [30. Starch can be modified to obtain materials which melt below the decomposition temperature . The plasticized starch properties may be tuned by changing the temperature of processing. and the properties and amount of plasticizers. which still limits their use. 30]. which will result in a continuous phase in the form of a viscous melt [29‐30. As the result. which is even above the thermal degradation temperature. 37]. or melted starch [29. The thermoplasticization process will decrease the interactions of the molecular chain and destruct the structure of the starch .
The products are called cross‐linked starches. 46].3‐1) are still soluble in water. The most popular starch ester is starch acetate . Cross‐linking results in a reduction of the solubility in water and to thickening.3. the use of formyl esters of starch is much lower than that of acetate esters. 39]. and particularly with acetic acid . while highly substituted starch acetate (DS of 2‐3) are soluble in organic solvents. Higher processing temperatures generally led to higher cross‐linking levels. These hydroxyls can react easily with a wide range of compounds such as acid anhydrides. aldehydes. Starch Esters The development of starch esters started in the mid 19th century [44. 1. Most studies dealt with the synthesis of starch esters of C1‐C4 carboxylic acids.2) are commercially available and used as food additives and in the textile industry. Schuetzenberger acetylated starch with acetic anhydride . Chemicals of these classes having two or more of the reactive groups may react with two or more hydroxyls of the starch molecules. Substitution of the hydroxyl groups of starch with acetate groups makes the esters more hydrophobic than native starch.3. Solutions to improve the properties of TPS are blending. leading to higher viscosities [42‐43]. 1.3. Already in 1865. The esterification reaction is catalyzed by H+ and proceeds with the formation of water . The cross‐linked products are therefore more rigid (and less elastic) materials than virgin starch . Three different types of starch acetates may be distinguished. et al  studied the mechanical properties of films from maize starch cross‐linked by sodium trimetaphosphate (SMTP). Cross‐linked starch An anhydroglucose molecule of starch contains two secondary and one primary hydroxyl group. textile. paper. epoxy. The simplest starch ester is starch formate (C1). or coating with hydrophobic polymers [33. especially in the food. High DS starch acetate can be easily casted 11 .01‐0. and adhesive industry. Silva. organic chloro‐compounds. The cross‐linked starches have found many applications. Low DS products (0. differing in Degree of Substitution (DS). Medium DS starch acetates (0. The DS is defined as the moles of substituents per mole of AHG units . synthesized by direct addition of formic acid to starch at room temperature. Breakdown of the starch and the formation of low molecular weight products occur to a significant extent .2. Introduction washed out by water . and ethylenic compounds. resulting in an increase of the Young’s modulus and tensile strength of the products and a decrease in elongation at break. As a result.
The kinetics of the reaction between gelatinized aqueous potato starch and vinyl acetate was studied by De Graaf. Starch triacetate has been successfully synthesized using acetic anhydride in combination with pyridine‐gelatinized starch .1. acetate.Chapter 1 into films using organic solvents. et al [55‐56]. hydrogencarbonate. Another attractive route involves the use of vinyl esters as reagents . DMF. Starch esters have also been synthesized using alkanoyl chlorides [49‐53]. 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. or isopropanol) with acidic (hydrochloric or sulfuric acid) or alkaline (NaOH or triethylamine) catalysts . Acetylation of starch in water and DMSO using vinyl acetate has been studied lately by Mormann and Al‐Higari  as well as Dicke . and is shown in Scheme 1. High DS starch acetates are thermoplastic materials suitable to be used as biodegradable plastics . and phosphate) will result in C‐2. Esterification may be performed using acid anhydrides in aqueous media or organic solvents (pyridine. DMSO. xylene. Mechanism of starch acetylation using vinyl acetate 12 . Neutral and weak acid/ alkaline catalysts result in regioselective substitution at the C2 hydroxyl groups of starch. while alkaline catalysts (such as carbonate. allowing the polymer to be produced with a range of hydrophobicities . but also C‐6 and C‐ 3 substitution . The degree of acetylation of starch acetates can be easily controlled.1. Several synthetic routes have been developed for starch esters.
relatively expensive and rather corrosive. in the simplest case of a binary blend. however. Their role is comparable to the one of a surfactant in emulsion formation . interfacial agents (e. i. polymers are rarely miscible with each other  so that. If the right combination of chemical groups on the two components is present along the polymers backbone. melt‐mixing remains the preferred route to polymer blends mainly because of very practical reasons: low costs. availability of mixing equipment and no necessity to use any organic solvent (often employed for the in situ polymerization). 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 . a chemical reaction might take place upon processing (reactive extrusion). 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 .4. which in turn provide a very strong adhesion at the molecular level between the dispersed phase and the matrix. 1. Although the in situ polymerization process is not as technologically straightforward and economically convenient as melt‐mixing. However. Fatty‐acid starch esters have been synthesized using fatty acid (octanoyl. Two main methodologies are applied for the production of polymer blends. The first involves simple melt mixing of the two components for example by extrusion. 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 .34‐0. the latter are mixed together.3. dodecanoyl. 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 . As a result. but only relatively low‐DS (0. in order to achieve also a good molecular adhesion between the phases by melt mixing.e. octadecanoyl) chlorides [49‐53]. the two components are chemically linked to each other. one component will be dispersed into the other.g. Despite this advantage. compatibilizers) might be used. Introduction The synthesis of long chain fatty acid ester of starch has attracted much interest lately. By working at temperatures above the melting point or glass transition temperature of the two components. it is frequently used in order to chemically graft the polymerized component on the other one (polystyrene on polybutadiene in our example above ). However. they locate themselves at the interface between the 13 . 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.61) products could be obtained using this approach.
4. Matrix Dispersed Phase Figure 1. The synthesis of compatibilizer and its use for starch/ synthetic polymer blending has also been studied.1. The most often used synthetic polymer for blending with starch is polyethylene [27. 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. The blending and grafting of starch with synthetic polymers is usually performed to achieve higher hydrophobicity and to improve the mechanical and thermal properties as well as to obtain cheaper and more biodegradable products. 64‐98]. The products have been synthesized in the lab as well as on industrial scale [32. 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 uncompatibilized blends of starch and 14 . there have been efforts to blend as well as to graft synthetic polymers onto starch. The starch/polyethylene blends are used for agricultural mulch [27. 66‐81]. In the past. 80‐ 81] or food packaging . Starch‐based blends by melt mixing Starch‐synthetic polymer blending has been studied as early as in 1973 [64‐65].6. 39.Chapter 1 two components stabilizing the dispersion (most probably by a steric repulsion mechanism ) and providing an improved adhesion at the interface.6.3.
The use of reactive compatibilizer precursors PCL‐g‐glycidyl methacrylate (PCL‐g‐GMA) [89‐90]. 84‐98]. 15 . while PE‐g‐GMA posseses epoxide groups. To obtain completely biodegradable products. Studies on the blending of starch with PCL have been already described [32. dextran‐g‐PCL . PE‐g‐MA contains reactive anhydride sites. the compatibilizer. 96]. 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. among which synthetic polyesters are considered very promising materials . which both can react in situ with the hydroxyl groups of starch [70. the uncompatibilized blends of starch with PCL give coarse phase separation and a reduction in the mechanical properties when the starch content is increased. which combines excellent biodegradability with acceptable mechanical properties.e. 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. synthetic biodegradable polymers have been applied. Introduction polyethylene show a coarse phase separation due to differences in polarity of starch (hydrophilic) and polyethylene (hydrophobic). since the conventional synthetic polymers are usually poorly or non‐biodegradable. while the tensile strength of the compatibilized ones decreases only slightly with the starch content. Despite these good results. Some of these polyesters also have very good mechanical. is formed upon mixing. and EVOH . As a result a graft copolymer (PE‐g‐Starch). and poly‐lactones (such as poly‐butyrolactone. The tensile strength of the uncompatibilized blends is drastically reduced when starch content is increased. poly‐lactides. PCL‐g‐pyromellitic anhydride [94‐95]. The ester bonds of these polymers are susceptible to attack by water and this leads to enhanced biodegradability. and poly‐caprolactone). poly‐valerolactone. thermal and water/gas permeability properties that are even comparable to bulk non‐biodegradable polymers such as PE and PP. The mechanical properties of these blends (tensile strength and elongation at break) decrease at higher starch content. 39. thus improving the dispersion of starch in the PE matrix. The mechanical properties of the blend are also improved. These biodegradable polyesters will finally decompose into non‐toxic products . poly‐dioxanone. Examples of these biodegradable polyesters are poly‐glycolide. PCL‐g‐maleic anhydride (PCL‐g‐MA) [91. The use of conventional synthetic polymer such as PE for blending with starch will only result in a partially biodegradable material. As is the case of blends with PE. 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. i. 72‐73]. Polycaprolactone (PCL) is a well‐known synthetic biodegradable polyester.
4. 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. 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. starch‐g‐PCL has been synthesized using toxic materials such as isocyanates . the hydroxyl groups of starch are supposed to function as initiating sites. The water in starch granules competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. Three routes have been studied in detail: starch esterification. and the presence of liquid‐solid reaction system leads to reduced reaction rates between starch and ε−caprolactone. 1. leading to formation of PCL homopolymers rather that starch‐g‐PCL. which however is extremely air‐ and water‐sensitive.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) using i Al(O Pr)3 as the catalyst. In Chapter 3. and starch‐g‐ biopolymer formation. Another approach consists of the Ring Opening Polymerization (ROP) of ε‐caprolactone monomer in the presence of starch.2. Thesis Outline The objective of this thesis is to study synthetic pathways to obtain starch derivatives with the potential to be used as bioplastic. Starch is a very hydrophilic material that always contains moisture. The results of this study have been used as input for the synthesis of the starch‐g‐PCL.3. Previous studies showed that common ROP catalysts such as tin octoate or aluminium isopropoxide gave low (0‐14%) grafting efficiencies (GE. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). The highest GE value (up to 90%) has been achieved when using triethylaluminium as catalyst [97‐98]. during the reaction. Another possible cause for the low GE values is the heterogenous nature of the reaction. The method applied basically consists of three steps (temporary partial protection of starch‐hydroxyl 16 . a very flammable by‐product. and difficult to handle since it releases ethane. In the past.3.4.Chapter 1 1. thus resulting eventually in low GE values . defined as the percentage of grafted polyester to starch compared to the total amount of homopolymer and grafted polyester) .2. 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 . starch/ biopolymer blending. Chapter 2 describes the ROP of p‐dioxanone initiated by a protected monosaccharide (1.
Washington D. Holmes: Interactive European Network for Industrial Crops and their Applications. USA. Academic Press.. ROP of ε−CL on the remaining starch‐hydroxyl groups. Mol. . CRC Press. . de Bragança. Young: Fractionation of starch. Denmark: http://www. A. Poirier. UK. Part of A Global Economy. starch. in Modified Starches: Properties and Uses (Ed. Breed.C.html. Agricultural and Rural Strategy Group. Bemiller. C. Website of International Starch Institute.M. New Jersey. York. O. Summary Report for the European Union. O. 2002. Wurzburg: Introduction. P.L. London. Inc. 2005. University of Wales. accessed on August 01. Boca Raton. are also reported. 2007 17 . and removal of the silyl groups). References . Based on these preliminary results. R. R. A preliminary study on the effect of several process variables (reactant ratio. and application of different catalysts) on starch esterification using vinyl fatty esters (vinyl laurate and vinyl stearate) is provided in Chapter 4. Wurzburg).starch. Aarhus. UK. E. The synthesis of two interesting compatibilizers (PCL‐g‐GMA and PCL‐g‐DEM) for starch/PCL blends is discussed in Chapter 6. 1. . including the mechanical and thermal properties.5. 2000‐2005. Princeton University Press. Sand Hutton. E. and polyhydroxyalkanoates. The model. C.B. in Starch: Chemistry and Technology (Ed. 1. J. . Fowler: Industrial markets for starch.S.B. 1986. USA. Passchall). Bangor..dk/isi/stat/rawmaterial.F. 105‐122. Somerville: Mini review: Plant polymers for biodegradable plastics: cellulose. Central Science Laboratory. a systematic experimental study on the effect of the process variables on the starch ester DS has been performed. The Biocomposites Centre. UK. .H. 1984. addition of co‐solvent. Nawrath. Gywnedd. . 2008. An Introduction to the New Science of Biodegradable Plastics. Corn Refiners Association. together with the mechanical and thermal properties of the synthesized starch esters are reported in Chapter 5. . Introduction by trimethylsilyl groups.. USA. C. The results were quantified using a statistical model. 1995.N.A. Corn. 2004. Y. Stevens: Green Plastics. The use of these compatibilizers in starch/PCL blends. Inc. Whistler. Corn Refiners Annual Report 2007.
F. 2000. Passchall). New Jersey.R. UK. Academic Press. Strong: Plastics: Materials and Processing. Munich. Brussels. Polym. 2008. . Ren: Biodegradable plastics: a solution or challenge?.. Applications. R. Fifth Edition (Ed. Angerer: Technical Report Series. Website of Environmental Protection Agency. . Inc. accessed on August 19. USA: http://www. 2008 . 18 . Bastioli: Global status of the production of biobased packaging materials.. USA. Washington DC. Inc. C. J. A. Starch/ Starke 2001. Prog. R. M.N. Starch/ Starke 2001. Whistler. . The compelling facts about plastics: An analysis of plastics production. Rustgi: Biodegradable polymers. Hanser Publishers. Ehrenstein: Polymeric Materials: Structure. demand and recovery for 2006 in Europe. in Starch: Chemistry and Technology (Ed. 27‐40.M. O. 31. E.htm. Denmark. Seidel). J. Weith: Plastics. Culter: Plastics Packaging.D. Institute for Prospective Technological Studies and European Science and Technology Observatory. R. Chem. Germany. 1984. . 53 .351‐355. J. M. G. Bastioli: Global status of the production of biobased packaging materials. Marscheider‐Weidemann. 1273‐1335 .E. Lancaster. Hanser Publishers. 2005.J. Link.L.Chapter 1 . 105‐109 . Germany. .L. C. 2000. U. S. 557‐562. J. Selke: Packaging and the Environment. 1996. Prentice‐Hall. Association of Plastics Manufacturers. The Royal Veterinary and Agricultural University. 2003.W. . Eng. F. in Kirk Othmer Encyclopedia of Chemical Technology Volume 22.M. . 11..J. USA. Technomic Publishing Company. Cleaner Prod. 2000.J. S. Department of Dairy and Food Science. Frederiksberg. Inc. John Wiley & Sons. 53. London. Bemiller. X. G. J. . Watson: Corn and sorghum starches: production. Hernandez. J.A. S. Sci. R. . Whistler. 23. 1939. Hoboken.epa. Patel. Properties. Properties. . Schleich. B.. Fritz. Wolf. . R. Husing. 1990. Inc. the Food Biopack Conference Proceeding (Ed. Englewood Cliffs. Chandra. Plastics Europe. A. Applications. C. Munich. Daniel: Starch. Weber).B. Processing. Crank. A.gov/garbage/plastic. Braun: Environmental impacts of biobased/ biodegradable packaging. Belgium. and Regulations.E. Selke. Techno‐economic Feasibility of Large‐Scale Production of Bio‐based Polymers in Europe. 1998. Ind. USA. R.
Westhoff. Mikkilti. Yu. properties. Physical and mechanical properties of biobased materials – starch.M. Res. Otey. X. F. Ind. A Real Case of “Force Majeure” that should be passed through to Packaging Prices Immediately. 4. Nielsen. Muller.B. Y. 385–409 . and Couturier: Properties of biodegradable multilayer films based on plasticized wheat starch.J. Eng. 592‐595. J. J. 62. N. Interfaces 2006. Starch/ Stärke 2001. Doane: Starch‐based blown films.K. Smits. Spain. Polym. report of Food Biopack Project (Ed.H. J. 2005. 275‐282 19 . 1996.S. Dev. Kruiskamp. B. Avérous. Starch/Stärke. 1980.H.G. J. Petersen. Crops Prod.G. Forssella. R. Compos. Polym. Frederiksberg C. 57. K. Biodegradable thermoplastic starch. 34. Martin. . 409–416 . A. Part C: Polym. C. A. Macromol. P. Parker: Phase and glass transition behaviour. J. E. P. Wang: Properties of starch blends with biodegradable polymers. J. K. Chambre Syndicale des Emballages en Matiere Plastique (French Plastic and Flexible Packaging Association). O. M. Ma: The effects of citric acid on the properties of thermoplastic starch plasticized by glycerol. A. Sci. 331–336 . 53.M. interaction. 2000. Starch/ Starke 2001. Gandini: Surface chemical modification of thermoplastic starch: reactions with isocyanates.. J. . and Olsen.V.Lin: Note. 2003. Rev. and Pukanszky: Thermoplastic starch/layered silicate composites: structure. 21. 43. Moates. 1997. Curvelo. 53. Denmark.M. epoxy functions and stearoyl chloride. Polymer Sci. 13. P. the Royal Veterinary and Agricultural University. 2005. T. Prod.M. Gao. L. P.P. Appl. Carvalho. 1–17 . R.. a model for thermoplastic starch. of concentrated barley starch‐glycerol‐water mixtures. . Seville.L. . 1491‐1494 . Ind. Yu. 2008. 372–380 . J. status and perspectives. Schwach. Wang. Chem. W. Wang. Carbohydr.. van Soest. Y. measured by differential scanning calorimetry and solid state NMR spectroscopy. polylactate and polyhydroxybutyrate. Introduction European Commission Directorate General of Joint Research Centre. Press Release: Major Price Hikes in Plastics. A. July 15.F. J. K. Carbohydr. Vliegenthart: Interaction between dry starch and plasticisers glycerol or ethylene glycol. 494–504 .J. G. 2005.F. 356‐361 . Biobased packaging materials for the food industry.A. 2003. 53. Yang. Department of Dairy and Food Science. X. Bagdi. Weber).
destructuration and hydrolysis during O‐formylation. Mullen. 58 (9). 139–147 . 56. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Feller. B. Sci. Chem. 1942. Y.A. 65. 1517–1526 . P. Divers. S. Stab.D. J. Gattin. 24. Ramsay. . 74. Andrade: Reactive processing and mechanical properties of cross‐linked maize starch. Kontoyiannis. C. 46–51 . Boca Raton. Alric. K. 2004. X. A. Polym. B. Degrad. 1986. Appl. Jiang. . Polym. Matzinos. 1.T. J. E. 2004. Eng. USA. Coutuier: Comparative biodegradation study of starch‐ and polylactic acid‐based materials. B.A. Pacsu: Starch studies: Preparation and properties of starch triesters. 11–17. . Panayiotou: Processing and characterization of starch/polycaprolactone products. Grohens: Starch modifications. Carvalho. Vikman. Inc. Environ. Polym.C. Ind.H. Y. Ind. C. Chem. Wurzburg: Cross‐linked starches. F.A.B. O. Polym. 55‐78. J. V. Hulleman. Carbohydr. P.F. M. 2594–2604 . Mullen. Thiebaud. Pacsu: Starch studies: Possible industrial utilization of starch esters. D. 2002. Favis: Rheological and thermal properties of thermoplastic with high glycerol content. J. Polym. Silva. E. CRC Press.C. 77. in Modified Starches: Properties and Uses (Ed. 34. B. J. Bertrand. Ramsay.W. 2001.A. Qiao. J. 705‐721. E. Borredon. 1647‐1656. Sun: Reinforced thermoplastic acetylated starch with layered silicates.P. Crops Prod. E. Aburto. 35. Jarowenko: Acetylated starch and miscellanous organic esters.. Tserki. 17–24 . Sagar. Pilin. Sci. Boca Raton. Rodriguez‐Gonzalez.W. 1209‐1217. . 389‐398 . A. C. J. C. Inc. Polymer 2003.D. 44. W.J. Feil: Morphology and enzymatic degradation of thermoplastic starch– polycaprolactone blends. O. Ribeiro. W. I. Favis: High performance LDPE/ thermoplastic starch blends: a sustainable alternative to pure polyethylene. 1997. Bikiaris. Myllarinen. 381‐384 . Ind. USA. Prinos. A. 1986.B. T. 581–586 . 1943. Journal of Applied Polymer Science 1995. 1999.Chapter 1 . O. T.W.W. Starch/Stärke 2005. 57.D. S. R. Levesque. Copinet. J. Merrill: Properties of fatty‐acid esters of starch. Eng. 20 .D.J. F. chapter 4 in book Modified Starches: Properties and Uses. (editor).. M.B. 2006. Wurzburg). Ibezim. Wurzburg. E. van der Zee. . 58. M. H. Rodriguez‐Gonzalez. Appl. C. I. CRC Press. G.
A. 175. Panayiotou: Properties of octanoated starch and its blends with polyethylene. Prinos. E. C. 57. Starch‐ Starke 2005. I. Starch‐Starke 1999. . F. Beenackers: The kinetics of the acetylation of gelatinised potato starch.828. A straight way to regioselectively functionalized polysaccharide esters. Wiley‐VCH. Appl. . B. Dubois: Synthesis and characterization of compatibilized poly(ε‐caprolactone)/ granular starch composites. . D. J. 33‐43 21 . P. 302‐307. Duquesne. . D. H. Aburto. Starch‐Starke 2004. Mouysset‐Baziard. A. Janssen.J. 118‐121. 1991. Smith. A. W. P. Starch‐Starke 1999. C. Inoue: Compatibilizers for melt blending: Premade block copolymers. 11. Bikiaris. Technomic Publishing Company. . . W. Rutot. USA. Bakker. S. T. Al‐Higari: Acylation of starch with vinyl acetate in water. 34. Macromol. Tuschoff: Acylation of hydroxy compounds with vinyl esters.M. 1997. C. Bikiaris. Carbohydr.H. M. R. A. . J. 5590‐5598 .H. 145‐152. de Graaf. J. Nakayama. I. Aburto. . G. M. Munich. 101‐112. P. I.Stokes. Borredon. Polym. 1997. Symp.W. Thiebaud. characterization. Broekroelofs. J. M. Dicke. 51. I. Bakker and G. 1999.. New York. J. Alric.Graf.Coleman. US Patent 2. 1960 . A. Hu. G.K. de Graaf: Ph. 74. Hu). E. R. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Carbohydr. P. A. L. E. I. G. Alric. Inc.E. Scott.D. Alric. P. 28. R. M. 1440‐1451. Polym. A. S. Macromolecules 1996. P. Macosko. .E.928. J. 137‐144. E. and biodegradability of fatty‐acid esters of amylose and starch. D. E. J. USA. 51. 132‐135.C. J. W. Cellulose 2004. 56. Aburto. C. 1996 . Prinos. Introduction . Aburto. . 2001.Painter: Specific Interactions and the Miscibility of Polymer Blends. Senocq. Khandpur. A. Guegan. C. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. in Reactive Polymer Blending (Ed. Alric. R. 2001. 255‐263. Aburto.V. 29. Panayiotou: Synthesis. J. C. Rijksuniversiteit Groningen. Sci. J. Germany. 1995. Thiebaud. Introduction. Hanser Publishers.F. Lancaster. Hamaili. . Degee. Borredon. E. Mormann. Polym. D. Thesis.E.E. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. Marechal. Alric.Fennel Evans: Fundamental of Interfacial Engineering.
R. J. W. Pavlidou. . Ishiaku. 1242‐1251 . Chem. 167–172 . R.S. 647‐655 22 . 38. L. D. U. I. Vouroutzis. R. Y. 34. Almeida. Nikolov. Mani. Degradation studies and moisture uptake of sago‐starch‐ filled linear low‐density polyethylene composites. M. C. J. Bikiaris. Nakamura. Polym. R. 1998. Tech. 1467‐1475. 1997. G. St‐Pierre. Polym.‐J. Verhoogt: Processing and characterization of thermoplastic starch/polyethylene blends.A. Polymer.G.S. Z. 2904–2911. Ishiaku. Wang.D. Danjaji. III.Chapter 1 .H. Effect of compounding and starch modification on properties of starch‐filled Low‐ Density Polyethylene.D. 88. 236–241 . L. H.I.M. K. thermal and morphological properties of virgin and reprocessed LDPE/corn starch blends. Ramsay. . 59. N. N. Favis. Cordi. Danjaji. E. Rosa: Mechanical. D. 1841‐1846 . Z. H. 88. and tensile properties of Low‐Density Polyethylene (LDPE) and corn starch blends. Nawang. Polym. Eur. 2001. Polym. 1‐9 . J. thermal and morphological characterization of recycled LDPE/corn starch blends. 21. morphology. H. Res. 2003. Adv. 162–163. Rosa: Effects of the compatibilizer PE‐g‐GMA on the mechanical.A. B. Ismail. Sci. Stab. Koutsopoulos.M. Liu.G. Sci. J. 2005. Carbohydr. Panayiotou: LDPE/Starch blends compatibilized with PE‐g‐ MA copolymers. E. Appl. Eng. 2005. 310–317 . Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends.A. Mohd Ishak: Material properties. Mohd Ishak: Material behaviour.S. Sung. Z.J. 1991. Ribes‐Greus: Thermal and viscoelastic properties of some commercial starch products. Testing. 70. Pedroso. J. Prinos. A. J. A. Mechanical properties of sago starch‐filled linear low density polyethylene (LLDPE) composites. Pedroso. Mater. Duran.L.. 30. Gelina. Mei: Study and development of LDPE/starch partially biodegradable compounds. Testing. Polym. Ramsay. 287‐291 . D. I. Polvm. A. Nawang.S. D. R. Jane. Ind. Contat‐Rodrigo. Appl. N. Appl. W. Panayiotou: LDPE/plasticized starch blends containing PE‐g‐MA copolymer as compatibilizer. 1998. Technol. 16. Sci.D. Polym.S. 59. L. Polym. Evangelista. 1503–1521 . U. J.A. L. Bikiaris. Sun: Effects of polyethylene‐grafted maleic anhydride (PE‐g‐MA) on thermal properties. 2002. Z. 75–81 . C. Ismail. 20. B. Frangis. 2005.G. N. Proc. Effect of amylopectin to amylose ratio in starch. Polym. J. 2003. 1998. Degr.
M. Sci. Low density polyethylene/ starch blend films for food packaging applications. K. 1984. 139‐161. 267. 1. Yang. Jung.N. Appl. .Y. Jana. PHB‐V.M. A. A‐C. Makromol. Odusanya.. 87. J. Rodrigues.K. J.P. 1999. C.H.C. 35. Eng. Westhoff: Starch‐Based Films. R. E. Environ. T. Odusanya. 1997. J. 23. Chem. Azemi: Effect of starch predrying on the mechanical properties of starch/poly(e‐caprolactone) composites.M. cross‐linked and glycerol modified starch. Jana: Studies on the properties and characteristics of starch– LDPE blend films using cross‐linked. Polym. M. L.D.S. 16–19 . 43.S. Cho. Maiti: A biodegradable film. dos Santos Rosa. H. 2004. 40. Polym. Dole. Sci. 89. Averous. Res. R. Appl. . 4157–4167 . Polym.H. C. J. Tech. Garg. in Polym.W. B. Polym. N. Polym. 9177–9190 . Res. 42‐65. 2004. Polym. J. J. U. P. Stridsberg. Preliminary diffusion evaluation. Tang. 284‐287 . Dev. Park: Modification of aliphatic polyesters and their reactive blends with starch.P.M. Adv. 2. B. O. Karlsson: Polymers from Renewable Resources. 26. Lindblad. Calil: Effect of thermal aging on the biodegradation of PCL. Polym. . 2002. S. Pilot plant investigation for production of the biodegradable film. and their blends with starch in soil compost. Westhoff. Introduction . F. Albertsson. Sankar. Ind. .K. Fringant: Properties of thermoplastic blends: starch–polycaprolactone. Eng. O. M. Sci. K. L. Kim. 157. C. Albertsson: Controlled Ring‐Opening Polymerization: Polymers with designed Macromolecular Architecture. 2002. Sci. D. U. Y.H. 1987. Adv. A.M. . Appl. Sci. Raj. Ishiaku. Guedes. 1298‐1305. Otey. Manan.G. 41. Sen. Eur. Otey. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate.K. J. 41. Kammer: On mechanical properties of sago starch/poly(ε‐caprolactone) composites.. Sci.F. Polym. 2001. D. 12. M.S. C. . Liu. W. A. Prod. J. Ang. S. Moro. S. A. Kim. 2003. Kim. M. 23. B. Z. K. Part A: Polym. 1507– 1516.. 1659‐1663 . John. 2007. Manan. Ranucci. 81. 2000. F. Azemi. S. J. Ryner. Eng. Doane: Starch‐based blown films. Bhattacharya Residual stresses and density gradient in injection molded starch/synthetic polymer blends Polymer 2000. 23 . Ind. Siddaramaiah. Polymer 2000. 3539–3546 . 877–884. U. M. Chem.S. Adv. 1139–1148. T. B. J. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. glycerol modified.R. Chem. Polym. Chem.H.‐C. J. 2003. M. S.M. 157. Sci. Ishiaku. 32–45 . 3976–3987 . 179‐187.
Mani. R. Sci. Errico. .E. M. Dubois. . Macromol. Laurienzo. Dubois: Aliphatic polyester‐grafted starch composites by in situ ring opening polymerization. Avella. Composite Interfaces. Sadocco: Preparation of biodegradable polyesters/high‐amylose‐starch composites by reactive blending and their characterization. R. Dubois: Synthesis and characterization of compatibilized poly (e‐caprolactone)/ granular starch composites. P. Polym. Krishnan. 2002. Symp. Rutot. Macromol. Degee. J. 1998. M. 83. 3091‐3100. Appl. Stab. C. Raimo. E. 80. Errico.Chapter 1 . M. Rimedio: Polymer communication: Preparation and characterisation of compatibilised polycaprolactone/ starch composites. Rapid Commun. Avella. 1432–1442. P. 24 . M. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. D. Polym. 283–286. Narayan.E. 2001. J. . Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite. M. 2003. 2000. 41. Polymer 2000. P. M. 7. 3875–3881 . Bhattacharya: Synthesis and characterization of starch‐ graft‐polycaprolactone as compatibilizer for starch/polycaprolactone blends. and P. Degée. 40. P. Degrad. E. Martuscelli. D. Polymer 1999. P. P. R. 175. 19. R. Rimedio. 127–134. M. 215–225. Tang. 33–43 . . Rutot.S. Duquesne. R.
The products were synthesized at 60‐100°C and characterized by 1H‐ and 13C‐NMR.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 (1.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose)/Al(OiPr)3 initiator system to yield polydioxanone with a protected monosaccharide end‐ group is described. and MALDI‐TOF mass spectrometry. Besides the desired polydioxanone functionalised with a monosaccharide end‐group. Keywords: biodegradable.2. ring‐opening polymerization . 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.3. polyesters. also polydioxanone with an OiPr end‐group was formed (20‐30 %). 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. 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).
Despite its good properties. stannous octoate [Sn(Oct)2]. This chapter describes experimental studies on the catalytic ROP of p‐ dioxanone in the presence of a protected glucose molecule (1. Exchange of the alkoxide group by e. Polydioxanone (poly(p‐dioxanone) or poly(1. Nishida et al  reported that polydioxanone decomposes to non‐toxic gases (CO2 and H2O) by microorganisms. From a biodegradability point of view. such as polycaprolactone.Chapter 2 2. are interesting polymers because of their good product performance and biodegradability .3 MPa and an elongation at break of 500‐600%. testosterone. pregnenolone. The ring‐opening polymerization (ROP) of p‐dioxanone using metal catalysts such as aluminum isopropoxide [Al(OiPr)3]. Several polymers with bioactive alcoholic and phenolic end‐groups of interest for drug‐related applications were synthesized (for example geraniol. 9]. Besides 26 . The synthesis of protected monosaccharide end‐capped biodegradable polymers is usually performed via Ring Opening Polymerization (ROP).g. 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.1. Polydioxanone has a tensile strength close to 48. It is fully degraded in the body within a period of 180 days . polydioxanone also shows good performance. 2). quinine.4‐dioxan‐2‐ one)). stigmasterol and ergocalciferol)  and also with protected monosaccharides (galactopyranose/ glucofuranose) [6‐7]. Its melting temperature is close to 110 oC.2. The catalysts are metal alkoxides with Lewis acidic character [8. probably because the p‐dioxanone monomer has become commercially available only recently . particularly for nano‐ encapsulation systems for drug delivery . which is a unique compromise between application and processing temperature. and polylactides. whereas the high temperature for polylactides results in thermal degradation and undesirable transfer reactions during synthesis and processing steps . This melting point is considerably higher than typically found for polycaprolactone (60 oC) and lower than that of polylactides (at least 175°C). which to the best of our knowledge is the first study to functionalise polydioxanone with a monosaccharide. accessible by the polymerization of p‐dioxanone (1). reaction of the metal‐alkoxide with an appropriate alcohol allows the synthesis of end‐capped poly‐lactones.3. Biodegradable aliphatic polyesters end‐capped with sugar molecules have been studied extensively for use in biomedical applications. or zinc lactate has been reported. only limited information about the synthesis and properties of polydioxanone is available in the open literature. The relatively low melting point of polycaprolactone limits its applicability. has interesting product properties compared to other aliphatic polyesters. and is tougher than polylactides and even HDPE . Introduction Aliphatic polyesters. tocopherol. polyglycolide.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose.
2. 8 mmol). 5.2.2. Subsequently. To this solution.07 g. Analytical grade dichloroethane (Labscan). the synthetic pathways and the soluble. However.1. The polymerization was allowed to proceed for 16 h at 100 oC.2.5 mmol) was dissolved in toluene (1 ml) at 50 oC.3. 0. part of this solution (350 µl. hot dichloroethane (20‐25 ml) was added to completely dissolve the partly solid polymer at room temperature.2. Germany) was purified according to the procedure described by Raquez et al [2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2) 2 (1. Materials p‐Dioxanone monomer (1.3. poorly soluble heterogeneous systems. CDCl3 was obtained from Sigma and was used as received. Typical example for the synthesis of polydioxanone end‐capped with 1. (20 mmol) of aluminum isopropoxide to 20 ml of toluene.8 ml of a solution of an aluminum isopropoxide stock solution was added. 2. 98+% (Aldrich) were used as received. NMR) due to the poor solubility of the products in standard organic solvents. 97% (Sigma) and aluminum isopropoxide. The reaction mixture was clear and colorless during the reaction. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System potential applications in biomedical products. 1. The resulting clear solution was stirred for 2 h at 50 oC.g. heptane (Acros).2. and diethylether (Labscan) were used as received. Toluene (Labscan) was dried and stored on molecular sieves 3 Å (Labscan) under nitrogen.44 g. Next. the mixture was brought to room temperature and several drops of HCl (1 N) were added to stop the reaction. Particularly interesting in this field are starch polymers grafted with polydioxanone.2. As such.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose (2). grafting efficiencies are difficult to determine in this system by standard analytical techniques (e. After the pre‐determined reaction time. the results of this study are also of interest for the preparation of starch/polydioxanone polymers using ROP.08 mmol aluminum isopropoxide) was added to pure 1 (0. relatively low molecular weight compounds reported in this study may be viewed as model systems for more complex. This stock solution was prepared by adding 4. 3]. containing 0.1. The hot solution was 27 . 2. Boehringer Ingelheim. which was pre‐heated till about 60°C to obtain it in a liquid state. Methods All polymerization experiments were carried out under a protective nitrogen atmosphere using standard Schlenk‐ and glovebox techniques.2.55 mmol of 2 and 0.8 g. Materials and Methods 2.2.
2.) H-isopropoxide end group 28 .4. 2. 2D‐NMR spectra were recorded on a Varian Unity 500 NMR. the above equation simplifies to Xn theo = monomer conversion × [monomer ]0 [ sugar ]0 (2. 2.) Here. 4:1 by volume) at 4‐8 oC. two polymers with different end‐groups are present.3. 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). If the amount of sugar is in excess with respect to the aluminum catalyst. 1H‐ and 13C‐NMR spectra were recorded on a Varian AMX 400 NMR. Calculation of Average Degree of Polymerization The Theoretical Average Degree of Polymerization.) H‐NMR was applied to determine the experimental average degree of polymerization. MALDI‐TOF spectra were recorded on an Applied Biosystems Voyager DE‐PRO machine using dithranol/ NaI as the matrix (linear mode).Chapter 2 precipitated in a heptane/ ether mixture (300‐400 ml. The isolated yield at this condition was 68%. 40 oC) until constant weight.1. The white solid was finally separated from the liquid by decantation and dried in a vacuum oven (5 mbar.3.2. X n theo in terms of number of monomer units is calculated as follows : Xn theo = monomer conversion × [monomer ]0 [total initiator ] [monomer ]0 = monomer conversion × 3[ Al (O i Pr) 3 ] + [ sugar ]0 (2.2. Product Analyses NMR analyses were performed in CDCl3. one initiated on a galactopyranose molecule and the other on an isopropoxide group. it is assumed that all available initiator is used effectively. X n exp of the product. As will be shown later. This leads to the following equation: Xn exp X n exp is = A A H-repeating unit H-galactopyranose end group +A (2. Processing of the raw data was performed using VNMR software.
CH2OH 6 O 4 H H5 O 3 H O O H 2 O H O 1 α O β O γ Al(OiPr)3 catalyst O O α γ γ' β' β α' H2C O [ C CH2 O CH2 CH2 O ]n C CH2 O CH2 CH2 OH 6 O H O H5 1 4 O H O H 2 3 H O + n+1 (2) (1) (3) Scheme 2. In the first step.1.) 2. Screening Experiments Initial experiments to synthesize protected galactopyranose end‐capped polydioxanone (3) were performed using 184.108.40.206. in term of the NMR resonances : R gp = A 5.1. Typically.6 ppm + A 5.). R gp . To compensate for this effect.3. To avoid the formation of isopropoxide end groups. an excess of monosaccharide 2 on Al was used (see eq.2.6: 96 was applied. an Al ‐ monosaccharide 2 ‐ monomer 1 mol ratio of 1: 6.1 ppm (2. the protected mono‐saccharide 2 reacts with Al(OiPr)3 to form the actual catalyst for the polymerization reaction.6 ppm A 5. Schematic representation of the polymerization reaction including atom numbering scheme 29 .).4− 4.7 ppm − 6 × A 5.) or.5. The exchange reaction typically takes place at 50°C for 2 h. is rewritten as: Xn exp = [ A 3.1 ppm (2. 2. eq.4.3. 2.6 ppm ] / 6 A 5. The polymerization reaction was performed as a two step process. is calculated using: R gp = A H − galactopyranose end group H − isopropoxide end group A (2. Results and Discussions 2.3.4‐di‐O‐isopropylidene‐α‐D‐ galactopyranose (2) and Al(OiPr)3 as the catalyst precursor (Scheme 2.7. 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.) The ratio of the two different types of polymers (either end‐capped with galactopyranose or an isopropoxide group).
It confirms that polymer initiation not only occurs with the monosaccharide but also with the remaining isopropoxide group of the catalyst precursor (vide infra)..1.1. although it is evident that p‐dioxanone polymerisation occurred. the in‐situ formed catalyst was reacted with p‐dioxanone monomer 1 at 100 oC for 16 h.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. together with the CH3 groups from the protecting groups of the sugar appear in the range 1.15‐1. Typical isolated product yields are 68% at these conditions. of interest is the presence of a small multiplet at about 5. Although the spectra clearly indicate the presence of the monosaccharide 2. Proton resonances of the CH3 group of the isopropoxide end‐group. it is not possible to determine whether this is truly an end‐group due to overlapping peaks with protons from the poly(p‐ dioxanone) backbone. Product analyses The products were analyzed using 1H‐ and 13C‐ NMR and MALDI‐TOF.40 ppm) unit are broadened and shifted up to 0. Typical 1 H‐ and 13C‐NMR spectra of 3 are shown in Figure 2.78 to 4.1 ppm. 220.127.116.11 ppm compared to the monomer.) In the next step.3. 1 30 . The OiPr : 2 end‐group ratio for the standard experiment was 1: 2.1 and 2.7. H‐NMR spectra (Figure 2. The typical proton resonances of the p‐dioxanone (3. 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.) are not particularly informative. respectively. However. The off‐white solid reaction product was collected after a dissolution‐reprecipitation process using dichloroethane and a heptane/ diethyl ether mixture.51 ppm.
28 ppm in 2 to δ 66. 13 31 . For instance. p‐dioxanone monomer.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose.2. 1.3.17 and 69.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.52 ppm (β’).3 ppm are shifted considerably.3.2. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2. 2 (b). Particularly the C2‐C6 carbons in the range δ = 66.1.2 ppm in product 3.2‐70. 1. 68. C‐6 is shifted from δ = 62. H‐NMR Spectra of: (a). characteristic resonances of an ‐O‐C(=O)‐CH2‐O‐CH2‐CH2‐OH end group are present at δ = 61.) is more informative and clearly shows the presence of a polydioxanone polymer backbone and a monosaccharide end group. 3 1 C‐NMR (Figure 2.8. 1 (c).2.2 ppm.61 (γ’) and 73.2 and 96. Carbon resonances arising from the monosaccharide end‐group are present between δ = 66. In addition to the mono‐saccharide end‐ group.2 ppm. The carbon resonances of the polydioxanone backbone are present at δ = 63.
3.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose. 2 32 . and 2.97 ppm (‐CH‐(CH3)2). 1 (c).97 (>C‐(CH3)2). The resonances of the protecting group of the sugar appears at δ = 108.70 (>C‐(CH3)2) and between δ = 21. A typical example of a part of the 2D‐NMR spectra is given in Figure 2.2.1. 3 13 Carbon resonances of the isopropoxide groups from the by‐product (isopropoxide end‐capped polydioxanone) are present at δ = 68. the numbering scheme of carbons and protons is given in Scheme 2.3. 1..1. D‐NMR (HSQC) was applied for complete peak assignment of the product.76‐25.48 ppm (‐CH‐ (CH3)2) and between δ = 21.2. 2 (b).Chapter 2 Figure 2.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose‐end‐capped polydioxanone.76‐25.2. C‐ NMR Spectra of: (a).73 and 109. An overview of the data is given in Tables 2.2. p‐dioxanone monomer. 1.3.
68 4.25 3.8 Reactant peaks (ppm) Galactopyranose (2) H‐1 H‐2 H‐3 H‐4 H‐5 H‐6 5.78 4.40 all values were determined using 1H NMR.1.50 4.3. 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 5.09 4.60 4.31 4.59 4. except values with *. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System Figure 2.83 p‐dioxanone Monomer (1) H‐α H‐β H‐γ a 4.00 4. around 1.732 broad. which were determined from HSQC spectrum due to overlapping resonances.16* 4.77 3.74 3.17 3.26 3. HSQC 13C‐1H spectra of 3 Table 2.32 * 4. 33 .301 3.54 4.
For this purpose.21 NMR analyses also allow calculation of the molecular weight of the products.63 68.61 170. A detailed procedure is given in the experimental section.28 70. which is the molecular weight of a repeating dioxanone unit.52 63. The difference in molecular weight between the main peaks is 102 g/mol. the ratio of the intensity of the end groups and the polymer backbone peaks is determined.80 61.58 68.00 171. corresponding with an average number molecular weight (Mn) of about 1600.28 p‐dioxanone monomer (1) C‐α C‐β C‐γ C‐carbonyl 61.2 68.2.54 166. An example of a MALDI‐TOF spectrum of 3 recorded in a dithranol/NaI matrix is given in Figure 2. 34 .75 70. The latter likely stems from the matrix used to ionize the sample.) This series represents a dioxanone polymer end capped with 2 and an additional Na ion.31 70.21 70.4.33 70.11/ 170.10 71.92 68. The product obtained at screening conditions (16 h reaction time at 100°C) contains on average 13 monomer units. A typical molecular weight distribution is observed.17 69.2 73.57 62.07 Reactant Peaks (ppm) Galactopyranose (2) C‐1 C‐2 C‐3 C‐4 C‐5 C‐6 96.92 66.8. 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.Chapter 2 Table 2. The molecular weight distribution of the major peaks may be represented by the following relation: m / z = 23 + 260 + 102n (2.09 65. MALDI‐TOF was also applied to characterize the products.
for details. in line with the NMR data.5 +102 1400 1600 1800 2000 2200 2400 Mass (m/z) Figure 2.) (2. ionized with either Na+ or H+.) and imply the presence of dioxanone polymers with carboxylic end groups.11.) This relation is indicative for the presence of a polydioxanone polymer containing an isopropoxide end‐group.11. This series may be described by the following relation: m / z = 23 + 60 + 102n (2.) and (2. Furthermore.).5. 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. Typical MALDI‐TOF spectra of 3 A second molecular weight distribution is also clearly visible when enlarging the spectra. These distributions may be represented by equation (2. m / z = 23 + 18 + 102n m / z = 1 + 18 + 102n 35 (18.104.22.168.) . two other distributions are present.10. see Figure 2. although both with a very low intensity (Figure 2.
the catalyst precursor is treated with monosaccharide 2 resulting in an alcohol exchange 36 . Enlarged MALDI‐TOF spectra for 3 Although MALDI‐TOF clearly demonstrated the presence of various types of end‐groups in the product. but in all cases poor quality.1.g.3. although successfully applied for galactopyranose‐end‐capped polycaprolactone . The observed differences in the molecular weight distributions of the various samples were only marginal. small but significant amounts (20‐30%‐mol) of polydioxanone chains with an OiPr end group are present. In the first step. Various other matrices were tested (e.1 0 600 700 800 900 Mass (m/z) 1000 1100 1200 Figure 2.7 796. Most likely the matrix (dithranol/NaI). is not particularly suitable for polydioxanone. Various samples with.13].3 10 836. according to NMR. it proved not suitable to determine the average molecular weight of the products. 2‐(4′‐ hydroxybenzeneazo)benzoic acid (HABA)).5. [6. 2. different molecular weights were analysed. In addition. low resolution spectra were obtained. 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 2.7 %-Intensity 30 20 858.Chapter 2 50 40 792. Mechanistic aspects Both NMR and MALDI‐TOF measurements imply that the main product is indeed the desired monosaccharide end‐capped polydioxanone.2.6.
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. 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. a p‐dioxanone molecule will coordinate to the Lewis acidic aluminium center followed by an insertion step. 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. The termination reaction is known to be reversible and the formed polymer may again react with an aluminum center and continue to grow . Next. During the reaction. (R = monosaccharide 2) 37 . 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 2.6. termination of the chain growth may occur by reaction with an alcohol.
1 86.5 67.6 6.11 22.4 2.5.3. 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 .6 6.5 1.3 6.6 2.22 13.21 54.5 1. a (nearly) fixed Al(OiPr)3 : monosaccharide ratio of 1 : 6.6 6.70 58.3. 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) 6.08 30.6 6.45 19. Systematic Studies The effect of important process variables (temperature. 2.8 3.1 2.6 t (h) 16 16 16 16 16 16 16 16 16 16 1. despite the excess of 2.3 6.36 17.96 2 3 a For each experiment. In all cases.55 Product Properties Avg.6 6.8 2.3 6.5 2.5 92. the end group may also be formed by a termination reaction with free isopropanol.74 52.6 6.74 23.2 91. OiPr‐end‐capped poly(p‐dioxanone).4 2.5 81.29 14.2. Table 2.4 3.3 81.04 19. However.5 1.78 10.85 14. will be formed when the polymerization starts with an aluminum alkoxide with a remaining Al‐OiPr group (eq.8 2.7. 2. An overview of the experiments and the results are given in Table 2.6 89.5 hrs and 16 hrs) and a p‐dioxanone to monosaccharide 2 molar ratio ranging between 3.3.93 8.81 16.50 25.6 96.7 2.9 1 7.74 18. A total of 15 experiments were performed at two polymerization temperatures (60 and 100 oC).3 3.50 35.).6 6.6 6. time and the mol ratio of monomer to monosaccharide) on the yield.3 6.53 59. These may be present in the reaction mixture because the exchange reaction between Al(OiPr)3 and 2 was incomplete.4 2/OiPr ratio Rgp (mol/mol) 2.49 13.22 57.).6 80.3 and 62.5 1. 2.12 42.3‐6. degree of polymerisation of the product and the end group distribution was determined.1 3.6 91.10 15.6 81.30 62. two different reaction times (1. formed in the first step of the polymerisation reaction (eq.16 46.3 6.0 87.5 84.6 86.Chapter 2 The minor product.9 2.28 33.1 2.93 37.7.6 was applied.5 T (°C) 100 100 100 100 100 60 60 60 60 60 60 60 60 60 60 dioxanone/ 2 ratio (mol/mol) 3. Chain Length (Xn exp ) Yieldb (%) 30.3 6.22 16.
). Another possibility is the occurrence of polymer crystallization. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 2. no detailed explanation has been put forward to explain this anomalous behaviour. 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. 39 . the yields are higher when performing the reaction at 16 h reaction times.3. will be required. the monomer conversion reaches a maximum value before going down to the equilibrium monomer conversion. equilibrium conversion is not yet achieved within 1.3. see Table 2. Clearly. with an enthalpy of polymerization of approximately ‐15 kJ/ mol .2. It was shown that at lower monomer to catalyst ratios. For the two experiments with a dioxanone: 2 ratio higher than 16 (≅ 23 and 60). the yields for reactions conducted at 60 oC are always higher than those performed at 100 oC. On the basis of our experimental data and in line with literature data.3.1. One of the experiments (S111) resulted in a very low yield (31%) compared to the other reaction. The effect of the reaction time on the product yield may be derived from the data provided in Table 2. Evidently. For this particular experiment. which reduced to 75% when increasing the temperature to 120oC. and particularly when comparing the data in set 2 and 3 (60oC. The effect of the reaction temperature (60 and 100°C) on the product yield at three different dioxanone: 2 ratios is given in Table 2.2.). the equilibrium conversion of dioxanone at 80oC was about 80%.5 and 16 h reaction time). the yield at 1. et al .3. To the best of our knowledge. A more detailed analysis on the actual nature and composition of the polymerization products.3. For instance. The latter is due to the slight exothermicity of the reaction . Similar observations were made by Raquez et al  and Kricheldorf et al  for dioxanone polymerisations using Al(OiPr)3 in the absence of a external alcohol or using benzyl alcohol. Product yield The isolated yields of the reactions are all but one between 67 and 96% (see Table 2. which is expected to be more pronounced at lower temperatures.5 h. However. a low dioxanone: monosaccharide 2 ratio was applied. 1.5 h is higher than the yield at 16 h. as suggested by Raquez et al . It is likely that this relatively low molecular weight compound dissolves partly during isolation/ purification procedure leading to lower isolated yields. leading to a low average chain length of the dioxanone polymer (7. when the reaction is performed at a low dioxanone: 2 ratio (≅ 16). 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. Higher polymerisation yields were obtained at lower temperatures. These results were explained by assuming that the reaction is an equilibrium polymerisation and that the equilibrium is shifted to the monomer side at higher temperatures.
A clear trend between R gp and the process conditions (temperature.3.2. and 2. time. The X n exp increases linearly with respect to the dioxanone : 2 ratio. time: 16 hr Temperature: 60oC. in line with the theoretical predictions.6. The R gp values (Table 2.5.Chapter 2 2. Most experimental points are scattered along these lines. A possible strategy to increase the Rgp values i. 60 Temperature: 100oC.) for all experiments are scattered randomly between 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. 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. as expected for a typical ring opening polymerisation . 2. reaction time and temperature on the X n exp of the products is shown in Figure 2.3.). 2. to reduce the number of OiPr endgroups in the product may be the removal of isopropyl alcohol formed in step 40 .7.) as a function of the dioxanone : 2 ratio at 90% and 100% monomer conversion is also provided. the X n theo (eq. and dioxanone/ 2 mol ratio) is absent.2. In Figure 2. Effects of process conditions on the average chain length ( X n exp ) and end group distribution The effects of the p‐dioxanone: 2 ratio.7.4 and 3.. time: 16 hr 50 Temperature: 60oC.7. time: 1.e.2.3.
49 + 0. The R2 value for the model is 0. The low P‐value clearly indicates that the model is statistically significant.459*10‐7 41 .2.4.g.7.6) was too limited to draw sound conclusions. indicating that the model describes the experimental data reasonably well.) where t and T are respectively the polymerization time and temperature. A parity plot of the modeled versus experimental values of the average chain length Xn confirms this statement (Figure 2. Statistical Data Analysis Quantification of the influence of the experimental factors (temperature. Analysis of variance (ANOVA) for linear model of X n function of experimental parameters SS 4103 93 4010 DF 3 12 15 MS 1337 7.3.3. A linear model proved adequate to describe the effects of the independent variables on the X n exp : Xn exp = 8.703 F 173.177 × (t ) − 0.974).097 × (T ) (2. vacuum distillation before adding the dioxanone monomer .12.) from the reaction mixture by e. Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System 1 of the polymerisation process (eq.854 × (dioxanone : 2 ratio) + 0.543 exp as a Model Error Total P‐value 3. time and dioxanone/ 2 ratio) on the X n exp has been performed by multivariable linear regression  on the data given in Table 2.3. Table 2.977 (with an adjusted R2 value of 0. 2. The analysis of variance for the model is given in Table 2.).3‐6.4.8. 2. The 2/ Al ratio was not included in the model as the experimental range (6.
2. Modeled versus experimental values for the average chain length Xn The model predicts that the X n is a clear function of the p‐dioxanone : 2 mol ratio. the model predicts that the X n exp is positively influenced by the polymerization time. with high ratios leading to a higher average chain length. Conclusions The ROP of p‐dioxanone in the presence of a monosaccharide (1. the model allows determination of the process variables to obtain a product with the desired degree of polymerization. Both NMR and MALDI‐TOF measurements indicate that the products are mixtures of polymers and significant 42 . Furthermore. 16. Within the experimental ranges. In addition. which is in agreement with the available data on ring opening polymerization [15. 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).Chapter 2 60 50 40 Xn model 30 20 10 0 0 10 20 30 Xn exp 40 50 60 Figure 2. the X n exp is negatively influenced by temperature. 2) with Al(OiPr)3 as the catalyst is reported. with higher ratios leading to higher molecular weights.4. The isolated yields of the off‐white solid products were between 30 and 96%. This is in line with literature data [5.3. 11] and due to the fact that the reaction is an equilibrium polymerization with a slight exothermic effect. 17].2.4‐di‐O‐ isopropylidene‐α‐D‐galactopyranose.8.
temperature and monomer: monosaccharide ratio) on the X n exp .4‐dioxan‐2‐one and controlled synthesis of diblock copolymers with epsilon‐caprolactone. H. These studies will be reported in the next chapter. .O. R. 43 . and mechanistic aspects of the ring‐opening polymerization of 1. 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 2.5. 1993. Rapid Commun. Macromol. A statistical model has been developed to quantify the effects of process variables (time. D. Narayan. Macromolecules 2001. Res. 199. Macromol. Dubois: ʺCoordination‐insertionʺ ring‐ opening polymerization of 1.J. References . 27. Chem. Mater. Chu. Biomed. Narayan. Raquez.g. 153‐166.6. Raquez. Dubois: Some thermodynamic. Degee. R. 1063‐1071.M. P. Kricheldorf. 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. . 21. D. J. Lin.M. 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. 2000. 2. J. C. 8419‐8425. Grubb: Hydrolytic degradation and morphologic study of poly‐p‐dioxanone. Zn‐lactate‐catalyzed polymerizations of 1. 34. Moreover.4‐dioxan‐2‐one.C.4‐dioxan‐2‐ one initiated by Al((OPr)‐Pr‐i)(3) in bulk. P. Phys. kinetic. 1089‐ 1097.R. 42. 1998. . H. starch). Damrau: Polylactones. P.L. P. Degee.
. H. Nishida, M. Yamashita, T. Endo, Y. Tokiwa: Equilibrium polymerization behavior of 1,4‐dioxan‐2‐one in bulk. Macromolecules 2000, 33, 6982‐6986. . T. Hamaide, M. Pantiru, H. Fessi, P. Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer agents. A novel route to functionalised nanoparticles. Macromol. Rapid Commun. 2001, 22, 659‐663. . K. Bernard, P. Degee, P. Dubois: Regioselective end‐functionalization of polylactide oligomers with D‐glucose and D‐galactose. Polym. Int. 2003, 52, 406‐411. . H.R. Kricheldorf: Syntheses and application of polylactides. Chemosphere 2001, 43, 49‐54. . H.R. Kricheldorf, M. Berl, N. Scharnagl: Poly(lactones). 9. Polymerization mechanism of metal alkoxide initiated polymerizations of lactide and various lactones. Macromolecules 1988, 21, 286‐293. . H.R. Kricheldorf, I. Kreisersaunders: Polylactones. 30. Vitamins, hormones and drugs as co‐Initiators of AlEt3‐initiated polymerizations of lactide. Polymer 1994, 35, 4175‐4180. . L.M. Esteves, L. Marquez, A.J. Muller: Optimization of the coordination‐ insertion ring‐opening polymerization of poly(p‐dioxanone) by programmed decrease in reaction temperatures. J. Appl. Polym. Sci. 2005, 97, 659‐665. . D.C. Montgomery: Design and Analysis of Experiments, 5th Edition, John Wiley & Sons Inc., New York, USA, 2001. . P. Dubois, R. Jerome, P. Teyssie: Aluminum alkoxides ‐ A family of versatile initiators for the ring‐opening polymerization of lactones and lactides. Makromol. Chem. Macromol. Symp. 1991, 42/43, 103‐116. . A. Duda, A. Kowalski, J. Libiszowski, S. Penczek: Thermodynamic and kinetic polymerizability of cyclic esters. Macromol. Symp. 2005, 224, 71‐84. . A. Kowalski, A. Duda, S. Penczek: Polymerization of L,L‐lactide initiated by aluminum isopropoxide trimer or tetramer. Macromolecules 1998, 31, 2114‐ 2122. . A. Duda, A. Kowalski, S. Penczek, H. Uyama, S. Kobayashi: Kinetics of the ring‐opening polymerization of 6‐, 7‐, 9‐, 12‐, 13‐, 16‐, and 17‐membered lactones. Comparison of chemical and enzymatic polymerizations. Macromolecules 2002, 35, 4266‐4270. . A. Duda: Polymerization of epsilon‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. Kinetics and mechanism. Macromolecules 1996, 29, 1399‐1406.
Ring Opening Polymerization of p‐dioxanone using an Al(OiPr)3‐Monosaccharide Initiator System
. S. Penczek, T. Biela, A. Duda: Living polymerization with reversible chain transfer and reversible deactivation: The case of cyclic esters. Macromol. Rapid Commun. 2000, 21, 941‐950.
The DS of the polycaprolactone graft is between 0. biodegradable polymers. 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. Keywords: starch. 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. Silylated starch with a degree of substitution (DS) between 0. The silylation reaction was performed using hexamethyl disilazane (HMDS) as the reagent in DMSO at 70°C.7 was obtained. silylation .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. polycaprolactone. grafting.72. The grafting efficiency varies between 28 and 58%.21 and 0.45‐0. the remainder being homopolymers of ε‐caprolactone. ε‐ Caprolactone is grafted to silylated starch by a ring‐opening polymerisation catalysed by Al(OiPr)3 in THF at 50oC. The poly‐(ε)‐caprolactone side chains consist of 40‐55 monomer units and is a function of the reagent intakes.
the used monomers and the corresponding grafted chains are not easily biodegradable. and the mechanical properties of starch‐based films are generally inferior to those derived from synthetic polymers [4. it is water sensitive. as starch is highly hydrophilic. 5]. resulting 48 . The polymeric structure of starch consists of repeating anhydroglucose units. Global production of starch is 60 million ton per year in 2004 . 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. for instance grafting of monomers (like styrene and methyl methacrylate) to the starch backbone [6. A wide variety of synthetic biodegradable polymers have been prepared. The content of amylose in starches depends on the plant and typically varies between 18‐28%. Polycaprolactone (PCL) is easily degraded by micro‐organisms . a natural biopolymer. Various modification strategies have been explored. These plastics are mainly synthetic polymers from fossil resources. However. Chapter 3 3. 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. which are known to degrade with difficulty and cause serious environmental problems . in almost all cases. roots) as granules or cells with typical particle sizes between 1‐100 µm.g. Starch has also been thermoplasticized with the help of plasticizers such as glycerol and other polyalcohols . 245 million tons of plastics are produced per year. 7]. Well known examples are polyesters derived from cyclic lactones (polycaprolactone. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decreased rapidly in time and were fully degraded after 4 weeks . the product properties are in most cases still not up to standards and blending with other polymers is required . The development of green biodegradable polymers for e. Starch.4‐glucosidic bonds) and amylopectin (a branched polymer with α‐D‐1.1. Starch is present in the body of many plants (tubers. Blends of thermoplastic starch and PCL are not fully miscible. is one of the potential candidates for future biodegradable polymer products. and this value increases with about 10% per year . Starch is abundantly available. The amylose‐amylopectin ratio in native as well as modified starches has a strong impact on the product properties. polyvalerolactone. Introduction Worldwide. Starch films are known to have good oxygen barrier properties. However. However. the future generation of packaging materials is highly desirable. There are two types of biopolymer in starch.4‐ glucosidic bonds). and polybutyrolactone).
49 .67 gram (8 mmol) Al(OiPr)3 in 50 ml of dry toluene in a glove box. This is rationalised by assuming that water competes with the hydroxyl groups of starch in the initiation step of the polymerization reaction. In this way. Aluminium triisopropoxide [Al(OiPr)3] (Acros) was used without further purification. To increase the miscibility of starch and polycaprolactone. 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]. Hydrochloric acid (HCl) 1 N was prepared from Titrisol concentrated hydrochloric acid solution (Merck) and distilled (Milipore) water. Starch is insoluble in the typical organic solvents used for ROP (such as toluene or THF). Acros) and methanol (Labscan) were used as received. it has been proposed to chemically graft caprolactone onto the hydroxyl groups of starch using ring‐opening polymerisation . In this chapter. ε‐Caprolactone monomer (Fluka) was dried over Calcium Hydride (CaH2) for 48 h. This approach has also been applied successfully to graft PCL and polylactide on dextran [19‐20]. distilled under reduced pressure at 100 oC and stored under a protective nitrogen atmosphere. Common Ring Opening Polymerization (ROP) catalysts such as tin octoate or aluminium isopropoxide gave low grafting efficiencies (GE. during the reaction. Dry tetrahydrofuran (THF) and toluene for polymerization experiments were obtained in closed vessels from Aldrich and were used as received. A stock solution was prepared by dissolving 1. Another possible cause for the low grafting efficiencies is the heterogenous nature of the reaction.2. the ring opening polymerisation occurs solely in the liquid phase and this is expected to lead to higher GE values. Hexamethyldisilazane (HMDS. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors in undesirable phase separation . This is expected to lead to reduced reaction rates between starch and CL compared to CL homopolymerisation. 0‐14%). leading to a liquid‐solid system. thus leading to the formation of PCL homopolymers rather that starch‐g‐PCL . an alternative method to synthesize poly‐(ε)‐caprolactone grafted starch co‐polymers (starch‐g‐PCL) is reported. a very flammable by‐product. This catalyst is extremely air and water sensitive. Materials and Methods 3. therefore difficult to handle and releases ethane. Materials Corn starch (Sigma) was dried at high vacuum (~1 mbar) at 100 oC for one day before use.2. The highest GE (90%) was achieved using triethylaluminium as catalyst .1. DMSO (Acros) and toluene (Labscan) were dried overnight over molecular sieves 3 Å (Merck) and stored under a protective nitrogen atmosphere. thus to a reduction in the GE. All available data indicate that the presence of water reduces the GE. 3.
A mol ratio of ‐OH groups to catalyst of 10:1 was used. After 24 h. Silylated Starch (1.2.111 mol) was added to the gelatinized mixture to initiate the silylation reaction. The reaction was carried out at 70 oC. broad peaks. 3. Sample SN‐3. 50 oC): δ 0. the silylated starch (1) product was dried in a vacuum oven (~5 mbar. DS = 0. After 6 h.2.2.2. Subsequently. Traces of DMSO trapped in the product were removed by dissolving the product in a small amount of toluene and re‐precipitation in methanol.2. starch). broad peaks. Sample SN‐3. 50 oC): δ 0.1 mbar.6 mL) at 50 oC (1‐2 h). the solvents were removed from the silylated starch product under reduced pressure (~ 20 mbar) at 70 oC.12 (m. silyl‐CH3). The white‐to‐transparent solid products were stored under vacuum in a desiccator at 6‐8 oC.2.2. Silylated Starch (1.60): H‐NMR (CDCl3. another portion of toluene (20 ml) was added. Chapter 3 3. before peracetylation.7‐2. 1 3. after peracetylation. For each experiment. toluene (40 ml) was added to solubilize the precipitated (partially silylated) starch. The samples were characterized by 1H‐NMR. 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. After removal of the solvents under reduced pressure (0. This procedure was repeated three times. Typical example of the starch silylation procedure The procedure for corn starch silylation was adapted from that published for dextran [19‐23].60): 1 H‐NMR (CDCl3. 40 oC) until constant weight.5 (m. 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.5 ml) and a predetermined amount of the stock solution of Al(OiPr)3 in toluene were added. a predetermined amount of ε‐caprolactone monomer (molar ratio of monomer to 50 . 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. 3‐6 ppm (m. DS = 0. pre‐dried corn starch (6 g) and dry DMSO (75 ml) were stirred at 70oC for about 3 h until a clear solution was formed. 3‐6 ppm (m. acetate‐CH3). starch). The pre‐determined amount of HMDS (typically 24 ml.1. THF (4. 1. Typical example of in situ polymerization of ε‐CL with silylated starch The silylated product from the first step was dissolved in THF (0.12 (m. 0. To this homogenous solution. silyl‐CH3). 80 oC). After 2 and 4 h reaction time.
The yield was measured gravimetrically and is based on the weight of the product and the total weight of reactants charged to the reactor. Silylated Starch‐g‐PCL (2.75. DSsilylation = 0.98 (t. 2. iPr).5.37 (t. starch).21): H‐NMR (DMSO d‐6. 1 3.5‐ 3. The product was collected as a white solid and characterised by 1H‐NMR. ε‐PCL). and vacuum dried at 40 oC for 24 h.31 (m.88 ppm (m. and 5‐5. γ‐PCL). 4. ‐CH.3‐4.5‐3. The peracetylation procedure applied in this study was adapted from the literature [24. 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). Subsequently.31 (m. The silylated starch‐g‐PCL (2) product was precipitated from the solution by the addition of heptane (about 250 ml) at ‐18 o C.) was >99 %. filtrated.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until it was fully dissolved (typically 3 h). Sample SN1CL1. 1 (0.16 (d.5. Typically. β and δ‐PCL).4 (m.75. 2. α‐PCL). Peracetylation of silylated starch Characterisation of the silylated starch by NMR proved very difficult due to the presence of very broad and overlapping resonances arising from starch.2. 3. 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. After stirring for 2 h. silyl‐CH3). broad peaks. DSsilylation = 0. and 5‐5. 3. iPr). 1 3. 3.54 (m. 4. Deprotection product of Sample SN1CL2.3. 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.88 ppm (m.2.2.68. DSPCL=0. β and δ‐ PCL).98 (t.2.25 (t. 60 oC): δ 0. 1. ‐CH3. 60 oC): δ 1. ε’‐PCL). DSPCL=0. broad peaks.2. The solid precipitate was filtered and dried under vacuum (~ 5 mbar) at 40 oC for 48 h. 3. 3. 1. ε’‐PCL). ‐CH. The samples were characterized by 1H‐NMR.25 (t. 3.3‐4. iPr).12 (s.16 (d.54 (m. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors OH‐groups was 13 : 1 for a standard experiment) was added and the ring opening polymerization reaction was allowed to proceed for 24 h at 50 oC.34): H‐NMR (DMSO. iPr). γ‐ PCL).68. 4. 4. 1. the desilylated starch‐g‐polycaprolactone product (3) was precipitated using heptane. the peracetylating reagents 51 . 1. α‐PCL). ε‐PCL). starch). Peracetylation of the remaining OH groups of modified starch is a well established procedure to improve characterisation of the products by NMR . The total isolated yield at this condition (see Table 3. Starch‐g‐PCL (3.4 (m. ‐CH3 .37 (t.4. 25]. 1.
3. 1 3.3.2. In this calculation. When assuming that all ε‐CL monomer is converted. 3. This leads to the following equation: ACL = ACH 2 −ε + ACH 2 −ε ' ACH 2 −ε ' = A3. ACH 3 silyl / 9 1 7 A−0.) 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. DSsilylation may be calculated using 1H‐NMR spectra of the products after peracetylation using eq.2. The peracetylation reaction was conducted for 7 h at 50 oC.8− 4. The assumption of high conversions (>95%) was correct for all experiments (see Table 3.4.3. It was finally dried overnight in a vacuum oven at 70 oC and 5 mbar until constant weight.6 ppm DSsilylation = × = × 3 Astarch proton / 7 27 A3−5. Analytical Methods 3. the DSPCL may be calculated using eq.8 ppm (3.2. The product was precipitated by the addition of methanol and washed several times with methanol.) where Ax‐y stands for the 1H‐NMR peak area in the range δ x‐y ppm. Chapter 3 (DMAP.3.2. 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.2. Nuclear Magnetic Resonance (NMR) H‐ NMR spectra were recorded in CDCl3 at 50oC or in DMSO d‐6 at 60 oC on a Varian AMX 400 NMR machine.3.1.).1.4 ppm + 1 (3. The ACL is calculated from 1H‐NMR spectra by comparing the peak area of protons attached to ε‐carbon atoms in a repeating CL unit with that of the characteristic ε’ protons of the last CL unit in a PCL chain  (see Figure 3. 3.2 ppm A3. 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.) 52 .1. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.3−3.6−0. it is assumed that the average length of the grafted chain is equal to the chain length of the homopolymer.
CL/OH ratio ⎥ ⎣ ⎦ (3. The values of ∆Hv and Vm were obtained from the SciFinder Scholar database (American Chemical Society.4 / 2 ⎟ ⎠ ⎝ ⎝ where Ax‐y stands for the peak area in the range δ x‐y ppm.8− 4 .2 ) ⎤ 1 × ⎢ (3 − DSsilylation )⎥ ⎢ ε . This leads to the following equation: GE = ACH 2 −ε ' / 2 − ACH −iso ACH 2 −ε ' / 2 × 100 % (3. 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.) 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).5. the focus of this chapter will be on the first two steps of the procedure. 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).9 ⎡ 0.3. A3. and Vm stands for molar volume.4 − A4 .3−3.3−3.5.( A3.3−3.4. Although all steps have been investigated.4 + A3. followed by an in‐situ Ring Opening Polymerization (ROP) of ε‐ caprolactone monomer on the hydrophobized starch and subsequent silyl group removal by a mild acid treatment. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors DS PCL = 0.3. T stands for absolute temperature. The Hildebrand solubility parameter of HMDS and DMSO were calculated using the following equation : ⎛ ∆H v − RT ⎞ ⎟ δ =⎜ ⎟ ⎜ Vm ⎠ ⎝ 1/ 2 where ∆Hv stands for heat of vaporization.8− 4. 2007) 3. 53 .8− 4 .) ⎛ ⎞ ⎞ ⎛ A A = ⎜1 − CH −iso ⎟ × 100 % = ⎜1 − 4.9 ⎟ × 100 % ⎜ ⎜ ⎟ ACH 2 −ε ' / 2 ⎠ A3.
the starch peaks are very broad and this feature hampers accurate determination of the DSsilylation. The silylation procedure was adapted from that previously reported for dextran [19‐23].3.).) shows the presence of silyl groups at about δ 0 ppm. 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. 24 h).) Instead of using DMSO as solvent.). An additional peracetylation procedure to substitute the free hydroxyl groups with acetate groups was performed to improve the quality of the NMR spectra. The 1H‐ NMR spectrum of silylated starch (Figure 3. 54 . a homogeneous reaction mixture was maintained throughout the reaction. C O H 2OR H 1 + NH3 2 OH H3C [R = H or Si(CH3)3] Starch Hexamethyldisilazane Silylated Starch (1) (3.5. It is defined as the average number of silyl substituted OH groups on the anhydroglucose (AHG) unit of starch.a. Synthesis of Silylated Starch The silylation of corn starch was performed with HMDS as the silylating agent (eq.1. mixtures of toluene and DMSO were applied to avoid precipitation of the silylated starch during the reaction. However.5. Chapter 3 3. 70 oC. The silylated products were characterised by NMR. 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. In this way.60 was obtained.b.a. as suggested by Einfeldt et al . The degree of substitution of the silyl groups (DSsilylation) was determined by 1H‐ NMR.1. The 1H‐ NMR peaks from the AHG unit after peracetylation were indeed considerably sharper and allowed more accurate DS calculations (Figure 3. 3.1. Using standard conditions (HMDS: AHG molar ratio of 3.1.). a product with a DS of 0.
55 . The results are given in Table 3. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors Figure 3. peracetylated The effect of the HMDS to starch molar ratio (1.2. in CDCl3.5) on the product DS was investigated by performing experiments with a constant starch and a variable HMDS intake. DSsilylation=0. 50 oC): (a).60.5‐4.1. not peracetylated (b). Typical 1H‐NMR Spectra of Silylated Starch (Sample SN3.1. and Figure 3.
46 Experiment SN1 SN2 SN3 SN4 a Experiments were performed in DMSO at 70°C.25 cal1/2cm‐3/2) due to the presence of the bulky apolar SiMe3 groups.4 1. the DSsilylation decreases for higher intakes of HMDS.45 0.60 0.6 DSsilylation 0.65 0. At a mol ratio of HMDS to AHG of 4.67 0.5 Figure 3.2.55 0. Its presence will reduce the polarity of the reaction medium (DMSO.5 HMDS : AHG Ratio [mol/ mol] 4 4.25 3 4. mmol) 12 (56) 18 (83) 24 (111) 36 (167) HMDS: AHG mol ratio 1.36 cal1/2cm‐3/2) considerably.5 2. The silylation reaction 56 . Effect of HMDS: AHG mol ratio on DS of the silylation product a HMDS intake (ml. a positive effect of higher reagent intake on the reaction rate and thus the DS is expected. 70 oC) Table 3. a fixed starch intake of 6 g (37 mmol AHG) starch was applied. HMDS is a rather apolar compound (Hildebrand solubility parameter of 6. the volumetric HMDS intake is about half of the DMSO intake.5:1. DS of the silylated products at different HMDS: AHG ratios (DMSO.5 0. Chapter 3 0.5 3 3.68 0. Surprisingly. For all experiments. solubility parameter = 11.5 2 2.1.5 Product DS (DSsilylation) 0. For reactions with an order higher than zero. The experimentally observed lowering at higher HMDS intakes is likely due to a decrease in the polarity of the reaction mixture.
Silylation of starch with HMDS in formamide.25‐5. The DSsilylation (0. 3.45‐0.7‐3. Petzold et al  reported that silylation of starch with trimethylsilylchloride (TMSCl) in pyridine yielded trimethyl‐silyl substituted starch with DS values between 0. DMA/LiCl.0 . sample SN‐1) in THF at 50 oC for 24 h using Al(OiPr)3 as catalyst.2.0) gave silylated dextran with DS values between 1.68.3. O O H2C O R H H OR H O O H OR H O Catalyst: Al(OiPr)3 Toluene. liquid ammonia. The use of HMDS to silylate dextran (Mw=6000‐40000) in DMSO (HMDS to OH molar ratio of 0.3‐2. 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. pyridine. To the best of our knowledge.6‐glucosidic for dextran).6. A schematic representation of the reaction is provided in eq. Similar reductions in the reaction rates when working at higher reactant intakes were observed for the esterification of starch with vinyl laurate and stearate .2. the silylation of starch in pure DMSO using HMDS as the reagent has not been reported to date.0 [18. white solid products with isolated yield > 96% were obtained. 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.) 57 .7) of the products is in the range of those published for other starch silylation systems. 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. The products are soluble in DMSO as well as in less‐polar solvent such as chloroform and THF. DMF. After precipitation with heptane and vacuum drying. 3. In situ Ring Opening Polymerization of ε‐Caprolactone with Silylated Starch A number of in situ ROP experiments with ε‐caprolactone (CL) were carried out using a typical silylated starch sample (DS=0.1 and 3.6. and DMSO/pyridine mixtures yielded silylated starch with DS values ranging between 0.
However. The homopolymers are formed by direct polymerization of ε‐CL initiated on isoproproxide moieties attached to the Al catalyst.22.214.171.124‐4 ppm and imply that the ring‐opening polymerisation reaction of CL indeed occurred.4 ppm. broad peaks in the region δ 3.0‐5. This is clearly indicated by resonances of the iPr end‐group of the PCL homopolymer at δ 4. the exchange reaction between Al(OiPr)3 and the OH groups of silylated starch is not quantitative under the conditions applied in this study. The peaks from the polycaprolactone units are clearly present in the range of δ 1.8 and 5. Resonances from the starch peaks are observable as small. not all of the caprolactone is grafted to starch and large amounts of PCL homopolymers (72%) were formed.4‐3. Figure 3. 26]. DSPCL=0. Further process optimization studies allowed the synthesis of products with less than 42% of homopolymers (vide infra).9 and 1. (Sample SN1CL1.68.2 ppm. Typical 1H‐NMR spectrum of a silylated starch‐g‐PCL sample.3. Apparently. DSsilylation =0. as is shown in Figure 3. The formation of CL homopolymers for this type of reactions has been observed before [15.21) in DMSO‐d6 at 60 oC. 58 . A typical spectrum is shown in Figure 3. Coding of the peaks is given in Figure 3. Chapter 3 The products were characterized using 1H‐NMR analysis in DMSO‐d‐6 as the solvent.
7. Exchange Reaction: iPr O Al O iPr H2C OH iPr O Al O iPr iPr O H2C O O O H O H OR H H OR [R = H or Si(CH3)3 ] H O + O H H OH H H OH H O + iPr OH Silylated Starch (1) 2.). 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. Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors 1. 1. 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. Exchange Reaction: iPr O Al O iPr iPr O + H2O iPr O iPr O Al OH + iPr OH 2. 3. However. residual water may also initiate the polymerisation reaction. This leads to the formation of carboxylic end groups (see eq.7. Product formation for the reaction between silylated starch and ε‐ caprolactone Besides the ‐OH group of starch. peaks arising from a ‐COOH unit could not be detected in 13C‐ NMR spectra (δ 175‐180 ppm).4.) 59 .
Overview of results for the grafting of ε‐CL on silylated starch a ε‐CL/ OH [mol/ mol] 13. The ACL of the polymer and the DSPCL increase almost linearly with the ε‐ caprolactone intake (Figure 3. This implies that grafting of caprolactone to starch also occurs to a significant extent. However.21 0.5 Avg.5 and 100%.34 0. Experiment SN1CL1 SN1CL2 SN1CL3 SN1CL4 SN1CL5 a.5. Chapter 3 The ratio of homopolymerisation versus grafting on starch may be obtained by comparing the integrals of selected peaks in 1H‐NMR spectra. the intensity of the resonance ε’ was considerably higher that that of the ‐CH‐ iPr peak.0 15.5 29. and Figure 3.47 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. Chain Length [mon. the intensity of the peak from the ‐CH2‐ end group of the homopolymer (ε’ at δ 3. The results are given in Table 3. The GE increases with higher ε‐CL intakes. 60 . The grafting efficiency (GE) for the sample given in Figure 3.5. Determined gravimetrically and defined as the total weight of the isolated product divided by the total intake of reactants (silylated starch and CL).2. units] 40 43 44 49 54 DSPCL 0.2.2 Total Yieldb [%] >99 >99 >99 99 96. and reaches 58% for a ε‐CL to starch–OH ratio of 29.9 ppm). b.9 22.2.72 Grafting Efficiency [%] 28 43 48 55 58 All reactions were performed using the same intake of SN1 silylated starch (DS=0.3 ppm) should be twice the intensity of the ‐CH‐ proton of the isopropoxide end group (δ 4. In the case of only homopolymerisation. This implies that the ε‐CL conversion is essentially quantitative in all cases. in all samples. Table 3. The yield of the products was measured gravimetrically and varies between 96.3.68) in THF at 50 oC with Al(OiPr)3 as the catalyst (1 mol Al(OiPr)3 per 10 mol‐starch‐OH groups).0 18.58 0. (SN1CL1) is 28%. Five experiments were performed to study the effect of different ε‐CL to silylated starch ratio.
Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors
Average Polymer Length [mon. units]
50 0.7 45 0.6 0.5 40 0.4 35 0.3 0.2 30 12 14 16 18 20 22 24 26 28 30 DSPCL
ε-CL : Starch-OH Ratio [mol/ mol]
Figure 3.5. Average Chain Length (ACL) and the DS for as a function of the CL‐ starch intake. ∆ : Average Chain Length (ACL); □ : DSPCL The mechanism of ROP of cyclic esters such as caprolactone in the presence of an alcohol (ROH, silylated starch in our case) is provided in Scheme 3.1. [30‐31]. Higher monomer intakes are expected to lead to higher polymerization rates as shown in Scheme 3.1.b. This will result in longer PCL chains in the final product, in line with the experimental observations (see Table 3.2. and Figure 3.5.). At higher ε‐caprolactone intakes not only the ACL of the grafted chain increases but also higher values for the DSPCL are observed. This finding may be rationalised by assuming that the rate of chain transfer (Scheme 3.1.c.) with starch is increased at higher ε‐caprolactone intakes. The rate of this reaction is expected to be a function of both the starch and the concentrations of Al‐species with a growing PCL chain. The starch intake for all experiments was equal, meaning that the concentration of Al‐species with a growing PCL chain should be higher at higher ε‐CL intakes. This is indeed predicted by the mechanism given in Scheme 3.1.b.; higher caprolactone intakes will increase the rate of this reaction and lead to higher concentration of Al‐species with a growing PCL chain.
Al O iPr
R OH O O kp kd
Al O R
O Al O [(CH2)5 C O]n R
Al O R
O Al O [(CH2)5 C O]n R + R OH
O Al O R
+ HO [(CH2)5 C O]n R
Scheme 3.1. The observation that a higher monomer to alcohol ratio leads to higher amounts of PCL chains with an alcohol end group and thus a higher DS was also reported for the polymerization of p‐dioxanone with Zn‐lactate as catalyst and α‐tocopherol as the alcohol . To show the potential of our approach to use hydrophobised starch instead of native starch for the ring opening polymerisation of cyclic esters, several ring openings polymerisations of native starch with ε‐CL monomer were performed either in pure ε‐CL or in a mixture of ε‐CL and toluene (80‐100oC, 24 hr). At the start of the reaction, the starch was always insoluble in the reaction medium. After reaction the product was isolated, washed thoroughly with toluene and dried. The weight of the product, however, was very close to the initial starch intake. Examination of the products by FT‐IR does not show the presence of caprolactone vibrations. Thus, it may be concluded that solubilisation of starch is of key importance to obtain poly‐(ε)‐caprolactone grafted starch co‐polymers. These findings are in line with earlier studies on the ROP of ε‐caprolactone on native granular starch using Al(OiPr)3 as catalyst . Here, caprolactone polymerization did not occur after 18‐24 h reaction time and only liquid ε‐CL was recovered. Only when performing the reaction with high concentrations of the aluminium catalyst, a product with a GE of about 13% was obtained. This low GE was explained by assuming that the reaction between starch and Al(OiPr)3 is slow and due to the heterogeneous nature of the reaction mixture. Our study, together with the result of Dubois et al  indicate that homogenous reaction conditions are required for the successful ROP of ε‐CL to obtain poly‐(ε)‐caprolactone grafted starch co‐polymers when using Al(OiPr)3 as the catalyst. When performing the reaction under heterogenous conditions, a high grafting efficiency is only achievable when using triethylaluminium as the catalyst .
Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors
3.3.3. Deprotection of Silylated‐Starch‐g‐PCL A preliminary experiment was performed to remove the silylated groups of the silylated starch‐g‐PCL (eq. 3.8.) using a mild acid treatment with diluted hydrochloric acid in THF at room temperature. All silyl groups were removed successfully, as is clearly seen from an NMR spectrum given in Figure 3.6.
Figure 3.6. H‐ NMR Spectrum of Starch‐g‐PCL (reaction product after desilylation of sample SN1CL2, DSsilylation =0.68, DSPCL=0.34), in DMSO‐d6
O O H2C H H OR H C (CH2)5 O
O H O CH2 H H OH H C (CH2)5 O
O H OR
THF, 2 hr H
O H OH
[R = H or Si(CH3)3 ]
Silylated Starch-g-Polycaprolactone (2)
21‐0. the percentage of PCL grafted to starch compared to the total amount (grafted and homopolymer) of polymerized CL [%] : gas constant. Nomenclature A : peak area of certain proton in 1H‐NMR spectra [‐] : peak area of certain peak at x until y ppm in 1H‐NMR spectra [‐] : average number of CL repeating units in a grafted polymer chain or homopolymer [monomer units] : Degree of Substitution. Chapter 3 3. The grafting efficiency varied between 28‐58%. depending on the ε‐CL‐starch ratio.72.2. 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. Conclusions The successful synthesis of poly‐(ε)‐caprolactone grafted corn starch co‐ polymers using a three step approach is reported. This was achieved by making the starch more hydrophobic by partial substitution of the OH groups by trimethylsilyl groups. indicating that homogeneous reaction conditions are favorable for the grafting reaction.5. The products may have interesting applications as compatibilizers for starch‐polymer blends. Poly‐(ε)‐caprolactone grafted silylated starch co‐polymers with average chain length of 40‐55 monomer units (polymer molecular weight of 4500‐6300) were obtained.986 cal mol‐1 K‐1 : temperature [K] : R T Vm molar volume [cm3/ mol] 64 . Control ROP experiments of ε‐CL with native starch under similar conditions did not produce the desired poly‐(ε)‐caprolactone grafted corn starch co‐polymers.46‐0. Considerable amounts of ε‐CL hompolymers with isopropyl end‐groups were also formed. Silylated starch with a low‐medium DS (0. The DS of the PCL chains was between 0. 1. The ROP with ε‐CL was performed using Al(OiPr)3 as catalyst in THF as the solvent.68) was obtained using a DMSO/toluene mixture as the solvent and HMDS as the silylating agent.4. 3. the highest value was obtained with a ε‐CL‐AHG ratio of 29. The key feature is the use of a homogeneous reaction mixture for the ROP of starch with ε‐CL.
M. 61‐67. 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. Adv. . Polym. Rustgi: Biodegradable polymers. 51.F. R. J. R.. C. Food Technol. Eng. De Mulder‐Johnston: Edible and biodegradable polymer films: Challenges and opportunities. Chiellini. G. . Goldberg: Review of the biodegradability and utility of poly(caprolactone). Beliakova. 1995. 43. 385–409 . 11. H. 52. accessed on August 01. Babac: Preparation and biodegradation starch/polycaprolactone films. A. Yang. 1996. D. 61‐74. Polym. . Wang: Properties of starch blends with biodegradable polymers. Biobased packaging materials for the food industry. .Sci. Brussels.C. J. 17. Sci. Fanta. Abdel‐Mohdy: Grafting of poly(methacrylic acid) on starch and poly(vinyl alcohol). Chandra.dk/isi/stat/rawmaterial. demand and recovery for 2006 in Europe. . status and perspectives. Rev. 2008.A. Linke: Water vapor sorption determination of starch based porous packaging materials. Starch‐Starke 2000. Environ.R.6. Karlsson. Prog. J. 407‐412. S. K. Tsiapouris. . R. 65 . Environmen. W. Polym. Wang. X. report of Food Biopack Project (Ed. Solaro: Biodegradable polymeric materials. Russell: Graft copolymers of polysaccharides with thermoplastic polymers ‐ new type of filled plastic. Association of Plastics Manufacturers.starch. 56. Polym.B.A. . the Royal Veterinary and Agricultural University. 107‐113. 2008. 23. 38. 1273‐1335.C. E. Frederiksberg C. 1998. M. F. Mater. 2003.M. J. Bagley. Aarhus. L. Macromol. R. C. . Burr. 1998. Krochta.html. Plastics Europe. 8. Department of Dairy and Food Science. 2003. 1997. C. Yavuz. References ..A. Y. Belgium. 305‐313. 1251‐1253. Sci. Part C: Polym. 53‐57. .K. Denmark. . C. 3. Doane. Aly. Eng. Starch‐Starke 2004. Polym. Website of International Starch Institute. Sci. Denmark: http://www. The compelling facts about plastics: An analysis of plastics production. A. 2000. Weber). of . Degrad. Albertsson: Biodegradable polymers and environmental interaction. 1977. E. J. A. 311‐316.
L. D. D. 40. . high value applications. Biosci. Symp. regioselectivity. Petzold. Dellacherie. . Rutot. P. 2002. D. D. C. Narayan: Aliphatic polyester‐grafted starch‐like polysaccharides by ring‐opening polymerization. P. K. 3091‐3100. Stein. L. Degrad. Hamaide.L. Krishnan. Sci. D. Klemm. 561‐566. Klemm. Stab. 73. Dubois. Six: Controlled synthesis of amphiphilic biodegradable polylactide‐grafted dextran copolymers. C. structure analysis. and subsequent reactions.L. Gunther: Synthesis and NMR characterization of regiocontrolled starch alkyl ethers. H. Dubois: Aliphatic polyester‐ based biodegradable materials: new amphiphilic graft copolymers. 965‐969. 251‐269. L. Einfeldt: Structure design of polysaccharides: novel concepts. Dellacherie. 41. Dole. 1443‐1450. Petzold. Chem. E. 1735‐ 1743. L. 2001. Koschella. E. E. P. E. . 42. Petzold. Klemm. Nouvel. Des. . Averous. D. . Petzold. I. Einfeldt.. 163. W. Degee. . D. 5. Cellulose 2003. K. Monomers Polym. Part A: Polym. A. 4157‐4167. M. Polym. Dellacherie. 2001. Nouvel. J. Ydens. M. 35‐47. 2. Klemm: Preparative and 1H NMR investigation on regioselective silylation of starch dissolved in Dimethyl Sulfoxide. J. Chapter 3 . J. Polym. B. K. Heublein: Silylation of cellulose and starch ‐ selectivity. P. W. Macromol. Six. Gunther. Klemm: Regioselective functionalization of starch: Synthesis and 1H NMR characterization of 6‐O‐ silyl ethers. Polymer 2002. Gunther. 2577‐2588. Dubois. 341‐347. A. T. P. Degee. Einfeldt. Rutot. Ydens. I. P. K. . Boullanger: Ring‐opening polymerisation of epsilon‐caprolactone with monosaccharides as transfer 66 . J. P. D. selective syntheses. Stein. Fringant: Properties of thermoplastics blends: starch‐polycaprolactone. R. W. Degee. 33. Polymer 1999. Dubois. 2004. L. C. P. Dubois: Controlled synthesis of poly(epsilon‐caprolactone)‐grafted dextran copolymers as potential environmentally friendly surfactants. . Biomacromolecules 2003. J. Fessi. Six: Silylation reaction of dextran: Effect of experimental conditions on silylation yield. Pantiru. Dubois. Polymer 2000. Dellacherie. 10. Six: Partial or total silylation of dextran with hexamethyldisilazane. A. Stein. L. 2001. . Biomacromolecules 2001. I. and chemical stability of silylated dextrans. A. Macromol. . Ydens. 1. P. Duquesne. 4. 43. . Macromolecules 2000. P. Kussler. 415‐426. M. E. 6713‐ 6721.L. P. . Moro. Nouvel.
2008.J. . A novel route to functionalised nanoparticles. D. Kricheldorf.M.: Solubility parameter concepts – a new look. 1399‐1406. Res. R. 21. T.P.O. D. 659‐663. . Synthesis of Poly‐(ε)‐caprolactone Grafted Starch Co‐polymers using Silylated Starch Precursors agents. Biela. Junistia. A. 199. F. 1998. 1978. . Sugih. L. Duda: Living polymerization with reversible chain transfer and reversible deactivation: the case of cyclic esters. Manurung. 67 . J. 1089‐1097. Rapid Commun. Macromolecul. D. 29. Macromol.4‐ dioxan‐2‐one. . Mexico City. S. Macromolecules 1996. presented at ACS Rubber Division meeting. H. L. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. 2000. Horton.R. H. 941‐950. A. Penczek. 61. accepted for publication in Starch‐Starke. 22. 2001. . Lehmann: Selective 6‐O‐acetylation of amylose. Duda: Polymerization of ε‐caprolactone initiated by aluminum isopropoxide carried out in the presence of alcohols and diols. Rapid Commun. Kinetics and mechanism. A. Janssen.K. Macromol Chem. Picchioni. Phys. 553‐556. Hertz Jr.L. 1989 . Carbohydr. Damrau: Zn‐lactate catalyzed polymerizations of 1.B.
Keywords: corn starch. vinyl ester to starch ratio and the type of catalyst. 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. and Na‐acetate) in DMSO at 110°C. The DS of the products is a function of the carbon number of the fatty acid chain. the DS for the starch‐laurate compounds is higher than for the corresponding starch‐ stearates. the DS decreases. When performing the reactions using Na2HPO4 as the catalyst.96). The yellowish products were characterized by 1 H‐. For low vinyl‐ester to starch ratios. 13C‐NMR and FT‐IR. esterification. The fatty esters were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. vinyl stearate . an increase in the vinyl‐ester concentration leads to higher product DS values. K2CO3.Chapter 4 Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants Abstract This chapter describes the synthesis of long fatty esters of corn starch (starch‐ laurate and starch‐stearate) with a broad range in degree of substitution (DS= 0. K2CO3. At higher ratios.24 ‐ 2.
7]. In particular starch is attractive as it is relatively cheap and abundantly available. An attractive field of application for these polymers is the use as packaging materials. To improve the mechanical properties. the use of native starch for packaging materials is limited due to its low moisture resistance. An alternative method using methyl and glyceryl laurate esters in the absence of solvent has been recently developed . Relatively low‐DS (0. most of the studies performed to date use short chain carboxylic acids (C1‐C4). higher molecular weight carboxylic starch esters (C4‐C6) . 5].34‐0. poor processibility (high viscosity). However. The hydrophobicity increases with the degree of acetate substitution (DS. However. even after the addition of plasticizers . Introduction Green biodegradable polymers derived from natural resources are potentially very interesting substitutes for non‐biodegradable petroleum‐based polymers. 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. Further modification of starch is therefore required to introduce hydrophobicity and to improve mechanical and moisture barrier properties. 5‐6] were introduced using fatty acid chloride reagents. and even fatty acid derivatives (C12‐C18) have been used in the modification reaction [5. Esterification of starch with low molecular weight fatty acid derivatives is one of the oldest modification technologies to improve starch properties. The major obstacle is the pronounced brittleness of the materials. Natural polymers such as starch. more water‐resistant products may be obtained [3‐4]. resulting in products with DS values up to 2. defined as the moles of substituents per mole of anhydroglucose (AHG) units) . The first paper on the acetylation of starch was already published in 1865 . The mechanical properties and hydrophobicity of the products were significantly improved when using these longer chain fatty acid precursors [1.7 [1.61) products were obtained using this approach. Their 70 . 5]. which are relatively expensive and rather corrosive . The introduction of acetate groups on starch makes the product more hydrophobic. and particularly acetic acid derivatives (C2) [2‐4].1. cellulose or proteins are potentially very interesting starting materials for biodegradable packaging materials.Chapter 4 4. and consequently. Recently Mormann et al  explored the possibility of using vinyl esters and particularly vinylacetate as reagents for the preparation of starch esters. and incompatibility with hydrophobic polymers. However. high brittleness. However. the fatty ester substituents [1. For the current petrochemical based products recycling is often neither practical nor economically feasible .
Germany) and acetic anhydride (Merck. 4.2. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. IR spectra were recorded on a Spectrum 2000 FT‐IR Spectrometer (Perkin Elmer). 4. not been reported to date. The products were placed directly on the diamond plate and 50 scans with a resolution of 4 cm‐1 were recorded. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany). leading to a moisture content of 2 %‐wt (measured gravimetrically). The synthesis of starch stearate esters using vinyl ester reagents has. The spectra were recorded at 50 oC. sodium acetate (Merck. The reactions were either performed in water or in DMSO using a basic catalyst (Na2HPO4). the use of basic catalysts other than Na2HPO4 has been investigated. Technical grade dimethyl sulfoxide (DMSO). the Netherlands).2. 1 71 . Germany) and disodium hydrogenphosphate (Merck. Germany) were used without further purification. 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. Potassium carbonate (Boom. Analytical grade vinyl stearate (Aldrich. and tetrahydrofuran (THF) were supplied by Acros (Belgium) and were also used as received. The starch was dried before use for 48 h at 105 oC under vacuum (approx. In DMSO.N‐dimethylaminopyridine (DMAP). Japan).2. Materials and Methods 4. The effects of the starch to vinylester ratio on the reaction rates and DS have been explored. leading to higher reaction rates. In addition. vinyl laurate (Fluka. 4‐N. We here report an investigation on the synthesis of higher fatty acid esters of starch with an emphasis on the introduction of laurate and stearate ester side chains. as recommended by Laignel et al . Materials Corn starch (approx. 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.1. ~1 mbar). The maximum attainable DS of starch acetate in water was below 1 and limited to 0. starch esters with a substantially higher DS value (up to 1. This solvent effect is likely caused by the higher solubility of the vinyl esters in DMSO compared to water.01 when using vinyl‐laurate.2. Germany) were used as received. to the best of our knowledge.6 for starch acetate ester) were obtained.
C2). The atom numbering scheme is given in Figure 4. 34. 3‐6 ppm (m. 1230 (C‐O stretching).9 (C3). 1455 (CH2). C3). 1. The product was washed twice with methanol (50 and 25 ml. Table 4. C1S‐6S).7 (C11). The samples were characterized by 1H‐ and 13C‐NMR and FT‐IR. 12). the product was dried in a vacuum oven (70 oC..1. 2.9 (broad. C4‐C17). Table 4. the product was precipitated using methanol (100 ml) and separated from the liquid phase by decantation. 1295.. 3H. 28‐32 (C4‐9). approximately 5 mbar) for 24 h until constant weight. CDCl3): δ 0. 2850 (C‐H stretching).8‐2. 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.4 (m.96) H‐NMR (before peracetylation. 172‐174 ppm (C=O. Finally. Starch‐laurate (Sample 17. C4‐11).3 (m. C12). 24. 1740 (C=O). broad peaks. C6S). C2). 1 H‐NMR (after peracetylation.52): H‐NMR (before peracetylation. typical spectra are given in Figure 4. broad peaks. C3). 2H. 16H. 5S).4. 3‐6 ppm (m. C1S‐6S). 1. broad peaks.5 g) was first gelatinized in DMSO (5 ml) at 70 oC for 3 h. 1370 (C‐H bending). 760.4 (broad.2.. 1020 (C‐O stretching). CDCl3): δ 14. 3H. Subsequently. 1150 (C‐O stretching). and O‐C6S). 720. Methods 4. 1110 (C‐O stretching). broad peaks.5 (m. 7H. 1410 (C‐H bending). C3). 61.2.0 (C12). 95. 2.Chapter 4 4.5 (m.9 (C10). 1. C2).1. attached to O‐C2S. (FT‐IR).3 (m. 1. broad peaks. 1. 1. broad peaks. 7H. (1H‐NMR). 68‐74 (broad. C2S.1 (C11). broad peaks. C1S‐6S) 1 72 . 2H. C4‐11).1. (13C‐NMR) and Figure 4. 3S.9 (t. CDCl3): δ 0.3 (m. 28H.3. 2. C18). overlap with CDCl3 (C4S).9 (t. 2H. Typical example of the synthesis of laurate and stearate esters of corn starch Corn starch (0. broad peaks.3. 3‐6 ppm (m. 935 (C‐O stretching).0 (m. 3H. 13 FT‐IR (cm‐1): 2920 (C‐H stretching).2. Starch‐stearate (Sample 19. 2H. 16H. CDCl3): δ 0. After cooling. C2’).5 (m. O‐C3S. Figure 4. C1S). DS = 2.9 (t. 76‐78 ppm. 1350 (C‐H bending). 2H. DS = 2. 1. 7H. 22.1. 1 C‐NMR (before peracetylation. resulting in the formation of a homogenous transparent solution. 2H. 3H.1 (m.6 (m.3. 31. respectively).
2. 6S O R 5S O R O 4S O 1S 2S 1 3S O O 2 3 4 5 6 7 8 9 10 11 12 CH3 Starch-Laurate O 1' 6S O CH3 2' O O R O 4S 5S O 1S 2S 3S O O 1 2 3 4 5 6 7 8 9 10 11 12 CH3 Starch-Laurate peracetylated 6S 4S O O R 5S O R O O 1S 2S 3S O O 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 CH3 Starch-Stearate O Figure 4. 2H. 25. 1410 (C‐H bending).2.2 (C2).1. 3‐6 ppm (m. 3H.9 (t. 950 (C‐O stretching). 1. 68‐74 (broad. 1350 (C‐H bending). 26‐32 (C4‐15). 3S. 1100 (C‐O stretching). Numbering scheme for carbon atoms of products 4. O‐C3S. 1. 3H.5 (m. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants H‐NMR (after peracetylation. C2S.0 (C3).4 (broad. 61. 760. CDCl3): δ 0. 75. 1020 (C‐O stretching). C2). 32. broad peaks. 1295.5 (broad.8‐2. 2850 (C‐H stretching).4 (m. C1S‐6S) 1 C‐NMR (before peracetylation. 1740 (C=O). C6S). 5S). 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. C18).7 (C17). C3). 865. 22. C1S). 34. 1150 (C‐O stretching). 28H. CDCl3): δ 14.0 (C16). A peracetylation reaction to substitute all of the remaining 73 . 1455 (CH2). 720. 7H. and O‐ C6S) 13 FT‐IR (cm‐1): 2920 (C‐H stretching). broad peaks. 172‐174 ppm (C=O.2. C2’). 95. 1.7 (C4S). attached to O‐C2S. 1370 (C‐H bending). 2H. C4‐C17).3 (m.6 (m.0 (C18).3.
) AH − starch = (4.6 − AC 2 3 (4.: DSacetate = AH − acetate AH − starch A3.4.). The peracetylation reaction was conducted for 7 h at 50 oC.3.2.3. 4.8 ppm (protons attached to the fatty acid carbons C3‐C12 for laurate and C3‐C18 for stearate) using eq. The procedure was described earlier by Elomaa et al  and the relevant equations are shown in equation 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 1.7.) 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. and 4.8‐2.5. acetic anhydride and pyridine in a DMAP : acetic anhydride : pyridine : AHG molar ratio of 1: 10: 22 : 1) were added.1.) 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. Subsequently.3.6. the peractylating reagents (DMAP. The product was precipitated by the addition of methanol and washed several times with methanol. It was finally dried overnight in a vacuum oven at 70oC and 5 mbar until constant weight. DS fatty ester = 3 − DS acetate (4.1 g) was suspended in THF (4%‐w/v) and stirred at 55 oC until the starch was fully dissolved (typically 3 h). Typically.5‐1.) (4. AH − acetate = A1. 4.2.Chapter 4 hydroxyl groups with acetate groups was applied to obtain reliable DS data.8− 2.6 ppm range) and a correction has to be made (eq 4. Determination of the Degree of Substitution (DS) The DS of the product was calculated using 1H‐NMR spectra of the products after peracetylation..6 7 (4. The DS of the fatty acid esters was calculated from the DS of the products after peracetylation (eq 126.96.36.199−5. the starch ester (0.).6‐5. and 4.5. 4. The peracetylation procedure by Einfeldt et al  was applied.3.4.: 74 .6 ppm).
8 = (stearate) 33 33 (4.7. Subsequently. 75 . the esterified starch started to separate from the medium in the form of a gel. 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.5−1.5−1.8 = (laurate) 21 21 AC 3−C18 A0. 4. After 2‐3 h. The products of these exploratory reactions are insoluble in water and DMSO.1.1. Ax‐y stands for the peak area in the range δ x‐y ppm. but swell in organic solvents such as toluene and THF. OH O HO OH O OH + R O CH2 O HO O R + n H3C H O n O (1) R = laurate (C12) (2) R = stearate (C18) Scheme 4.) In eq.) (4. light yellow solid. After 24 h. the vinylester and the catalyst were added and the reaction mixture was heated to 110°C.3.3.6. Results and Discussion 4.3..‐4. The reaction was performed in two discrete steps. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants AH − fattyester = AH − fattyester = AC 3−C12 A0.1). A schematic representation of the esterification reaction of starch with the vinyl esters is provided in Scheme 4. the brownish gel was precipitated with methanol and the product was collected after vacuum drying in the form of a transparent.7.1. Initially. the starch was gelatinised in DMSO at 70°C for 3 h to make the hydroxyl groups of starch more accessible for reaction. while ACx‐Cy is the area of the H‐atoms attached to carbons in the range Cx‐Cy (carbon numbering scheme is given in Figure 4. 4.
(b) starch laurate.96. DS = 2.Chapter 4 The DS of the products was determined by using NMR (vide infra).52 (Sample 17. Medium DS starch laurate and starch stearate (1 < DS < 2) dissolve poorly in DMSO‐d6 and CDCl3.3. 76 . 4. Table 4. Higher DS products have a higher solubility in CDCl3 and good quality 1H‐ and 13C‐NMR spectra could be obtained (Figure 4.1. Typical 1H‐NMR spectrum of: (a) native starch in DMSO d‐6 at 60oC. When using a vinyl‐laurate : AHG molar ratio of 3 and K2CO3 as the catalyst.3. Table 4. Figure. even at higher temperatures (50 oC).52 was obtained.).2. and 4. Product Characterisation 4.) in CDCl3 at 50oC.) in CDCl3 at 50oC. a product DS of 2. DS = 188.8.131.52. 4. A reaction with vinylstearate at similar conditions resulted in a stearate starch ester with a DS of 2. (c) peracetylated starch laurate.52 (Sample 17.2.2. 1H‐ and 13C‐NMR analyses The solubility of the products in common NMR solvents (DMSO‐d6 or CDCl3) is a function of the product DS. For atom numbering scheme: see Figure 4.3.1.
required for DS determinations. The starch peaks (δ 3‐5. 10‐11].5 ppm). in DMSO d‐6 at 60oC. DS = 2. Figure 4.2. The 1H‐NMR spectrum of a typical peracetylated starch laurate is shown in Figure 4. 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).8‐2.3 ppm. DS = 2.c.2. The same 77 . are together with the CH2 groups of the acid chain adjacent to the ester moiety in the range δ 1. This feature hampers the DS determination by NMR.8‐2.3. The resonances arising from the anhydroglucose unit of starch are broadened. Clearly visible are the peaks arising from starch and the aliphatic hydrogen atoms of the fatty acid chain (δ 0. Typical 13C‐NMR spectra of: (a) native starch. For atom numbering scheme: see Figure 4.52 (Sample 17) in CDCl3 at 50oC.5 ppm) are broad and overlapping . Two of the carbon resonances (1S and 4S) are considerably shifted compared to native starch. (b) starch laurate. The proton signals of the acetate methyl group. and therefore a peracetylation procedure to substitute all of the remaining OH groups with acetate groups was applied [8. NMR spectra of the peracetylated products are considerably improved in terms of peak resolution and allow a more reliable calculation of the DS. Typical 13C‐NMR spectra of the products are given in Figure 4.1. (c) starch stearate.96 (Sample 19) in CDCl3 at 50oC.3. 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.
b. a spectrum of native starch (Figure 184.108.40.206.1.b.) is also included. In addition.4. FT‐IR Spectra of starch laurate (DS= 2.96.c. starch laurate C-H C=O C-O C-O a.Chapter 4 phenomenon was observed by Dicke for starch acetate . and 4.4.c. The near absence of remaining hydroxyl 78 . 4. and 4.4. the C‐H stretching vibrations of the alkyl groups of the fatty ester chain are clearly present at 2920 and 2850 cm‐1.4.) show characteristic bands of the carbonyl group of the fatty esters in the 1750‐1700 cm‐1 region.) and native starch FT‐IR spectra of both starch laurate and starch stearate (Figure 4. native starch -OH C-H 4000 3500 3000 2500 2000 1500 1000 500 Wavenumber (cm-1) Figure 4. Sample 17.a.). starch stearate C-H C=O C-O Absorbance b. Table 4. c. Table 4. starch stearate (DS= 220.127.116.11. Characteristic peaks of the polysaccharide backbone are visible in the 1250‐900 cm‐1 region (C‐O stretching) . FT‐IR measurements The FT‐IR spectra of starch laurate and starch stearate are shown in Figure 4.4. 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. For comparison. Sample 19.4.
52 2.07 0.54 2.1. The results are shown in Table 4. four experiments were performed with two alternative basic catalysts (K2CO3 and Na‐acetate).91 0.60 1.05 0.24 2. in line with the NMR data.90 1.2.68 0.01 0.96 2.3. 4. 79 . Systematic Studies The effect of important process variables like the vinyl‐ester to AHG ratio. Table 4. In addition.44 Exp.90 0.1.08 1. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants vibrations in the range 3000‐3600 cm‐1 and at 1640 cm‐1 indicates that the DS of the product is high.99 1.23 0.13 1. type of catalyst and the effect of the addition of co‐solvents on the product DS was studied in more detail. 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.57 0. Most of the experiments (14) were performed using Na2HPO4 as the catalyst. 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.
5. 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. this positive effect dominates the reaction rate and the DS of the products will therefore increase at higher vinyl ester intakes. The results are presented in Figure 4. DS of the product as a function of the type of vinyl ester and the vinyl ester‐starch ratio (24 h reaction time. 7‐10). 110 oC. At a ratio of 1:6. At low to medium vinyl ester/AHG ratios (0‐3). 2 wt% catalyst intake on starch).Chapter 4 4. 1. Higher concentrations of vinyl esters are expected to lead to higher esterification reaction rates.5. This behaviour is likely the result of two opposing effects. This is likely due to a reduction of the polarity of the reaction medium.5 0 0 1 2 3 4 Vinyl Ester: AHG Starch Ratio [mol/mol] 5 6 Figure 18.104.22.168. The highest DS value was 1. Effect of vinyl ester to AHG ratio on the product DS The effect of the vinyl‐ester to AHG molar ratio on the product DS was determined for both types of vinyl‐esters with Na2HPO4 as catalyst (samples 1‐ 3.23 for vinyl‐laurate at an intermediate vinyl‐ester‐AHG ratio of 3.3. This reduced polarity is expected to lead to ○ : starch‐laurate □ : starch stearate (lines for illustrative purposes only) 80 . A further increase leads to a reduction in the DS.). the ester intake is equal on a weight basis to the DMSO intake.5 Degree of Substitution (DS) 1 0. The DS values are increasing with higher vinyl‐ ester/AHG molar ratio until a certain maximum.
A similar trend in reactivity pattern was observed in this study and the DS decreased from 1. The results are given in Table 4. For starch 81 .e.1.3.7 for lauroyl chloride to 0. When using Na2HPO4 as the catalyst. Here. The latter factors appears to have a strong effect on reaction rates (vide supra). Aburto. leading to enhanced reactivity. The addition of toluene for the esterification of starch with vinyl‐esters surprisingly did not lead to improved DS values.8 for stearoyl chloride. Thus. 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 . 4. Although toluene may positively effect the reaction by (partly) re‐dissolving starch ester precipitates.1. Effect of the addition of toluene as a co‐solvent A number of reactions were performed using a co‐solvent. These negative effects dominate the reaction performance at higher vinyl ester/AHG ratios and lead to a reduction in the DS values. This effect is particularly evident at higher vinyl ester/ AHG ratios (>3) (Figure 4.6. K2CO3 and Na‐ acetate were tested.2.3. Catalysts screening A number of alternative basic catalysts for Na2HPO4. 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.3. et al  for the silylation of starch.5. These findings were rationalised by assuming that the co‐solvents increase the solubility of the silylated products. 4.).1. In this case. with reductions in polarity leading to lower reaction rates. the DS of the product is also a function of the chain length of the fatty acid. and illustrated in Figure 4. 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.3. the starch laurate esters display higher DS values than the starch stearates. with high carbon numbers leading to a reduction in the DS. It is clear that Na‐acetate and K2CO3 are considerably more active than Na2HPO4 and products with a significantly higher DS were obtained.3. A similar procedure was proposed by Nouvel. 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. 11‐14 in Table 4. it also results in a dilution of the reaction mixture and a reduction in the polarity. i. the addition of co‐solvents (toluene/ THF) led to an increase in the DS.). 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.).
5 were obtained. The DS of the products is a clear function of the chain length of the fatty ester and the type of catalyst.5 Degree of Substitution 2 1.). 82 . the DS of the product is also tunable by proper catalyst selection.4. catalyst amount = 2%‐w.Chapter 4 laurate esterification. gray: Na2HPO4 black: K2CO3 white: Na‐acetate 4. Apparently. when using K2CO3 as the catalyst. Comparison of DS values with different catalysts (Vinylester: AHG molar ratio = 3:1. and Na‐acetate). 3 2. 110oC. and FTIR and confirm the presence of chemically bound fatty acid chains. However.1.5 1 0. the statement that the DS for the laurate esters is always higher than for the stearate esters is not generally valid and a. The products were synthesised in DMSO using vinyl‐esters in the presence of basic catalysts (Na2HPO4. the DS for the laurate ester is lower than the stearate ester. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4. Conclusions A study on the synthesis of corn starch fatty acid esters with high DS values is reported. function of the type of catalyst. For starch stearate.5 0 Laurate Vinyl-Ester Stearate Figure 4. in line with the findings for NaH2PO4. The yellow products were characterized by 1H‐ and 13C‐NMR. K2CO3 gave products with a significantly higher DS (2. DMSO).o. K2CO3.6.96) compared to Na‐acetate (DS=2.44). and Figure 4. the two catalysts are equally effective and products with a DS of about 2.6. Thus. The DS of the laurate ester when using Na‐acetate is higher than for the stearate ester (Table 4.
Ind. E. and P. Sagar. References . . M. Polym. X. C. C. 1647‐1656. M. Bengtsson. A. Prinos. 1208‐1217. the DS of the product increases at higher vinyl ester intakes. Appl. Sci. . . Y. products with a DS > 2. 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. Polym. 1‐11. 1440‐1451. Borredon. The DS of the products may also be tuned with the vinyl ester/AHG molar ratio. 1942. . Crops Prod. 58. Borredon. Prinos. At low vinyl ester/ AHG ratio. J. presumably due to a reduction of the polarity of the reaction medium. Dzenis. 1995. I. Polym. Thiebaud. D.4 could be obtained for both laurate and stearate esters.W.6. D. I. J. 361‐368. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Sci. 54. Eng. Polym. characterization. Pacsu: Starch Studies: Preparation and Properties of Starch Triesters. Mullen and E. J. Merrill: Properties of Fatty‐Acid Esters of Starch. Alric. 705‐721. 2005. Y. 74. 34. 1997. 4. Aburto.5. A. Gatenholm: Surface octanoylation of high‐ amylose potato starch films. and biodegradability of fatty‐acid esters of amylose and starch. J. Sci. A maximum was observed at a vinyl ester: AHG ratio between 2 and 4. 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. Higher ratios led to a reduction in the DS. Bikiaris. W. 65. K. 83 . Bikiaris. Appl. S. Aburto. . Chem. Carbohydr. Thiebaud. D. J. 21. J. thermal characteristics and biodegradability of extruded starch acetate foams. E. Xu. Hanna: Water solubility. Synthesis of Higher Fatty Acid Starch Esters using Vinyl Laurate and Stearate as Reactants With these catalysts. 1999. J. 2003. S. Alric. Ind. Appl. J. Panayiotou: Synthesis. E. Koch. Important product properties will be described in the next chapter.
Asplund. 43.Chapter 4 . 298. 255‐263. Al‐Higari: Acylation of starch with vinyl acetate in water. 1997. 57. 14(2) 127‐135. A. S. J. Klemm: Preparative and H‐1 NMR investigation on regioselective silylation of starch dissolved in dimethyl sulfoxide. . C. Laatikainen. I. . D. I. .F. W. R. H. Polym.. 57. R. J. Hyvarinen. Ydens. Einfeldt. Starch‐Starke 2004. Polymer 2002. Carbohydr. 1735‐1743. Soininen. Dellacherie. W. M. J. Elomaa. Six: Partial or total silylation of dextran with hexamethyldisilazane. Med. C. R. Macromol. S. D. Mano. Bliard. and E. M. L. J. Mater. . 2003. Res. P. 56. M. Reis: Thermal properties of thermoplastic starch/synthetic polymer blends with potential biomedical applicability. Mouysset‐Baziard. Urtti: Determination of the degree of substitution of acetylated starch by hydrolysis. Aburto. 51. Cellulose 2004. Kussler. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Senocq. J. B. 11. . M. Mormann. Biosci. Stein. G. Sci.L. T. Hamaili. H‐1 NMR and TGA/IR. Starch‐Starke 1999. 302‐307. G. Petzold. Aburto. Dicke: A straight way to regioselectively functionalized polysaccharide esters. 1. Carbohydr. Dubois. . F. Degee. Massiot. 2004. . A. E. Borredon: Free‐solvent Synthesis and Properties of Higher Fatty Esters of Starch – Part 2. Koniarova. Gunther. Alric. Nouvel. 84 . Laignel. E. L. 261‐267. 118‐121. J. 145‐152. I. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. P. 341‐347. 251‐260. K. Starch‐Starke 2005. 2001. Alric. . Mater. Peltonen. P.
45‐1. catalyst intake. vinyl stearate .5 MPa. Keywords: starch esters.26‐2. The effect of the process variables (vinyl ester to starch ratio. The basicity of the catalyst and the reaction temperature have the highest impact on the product DS. trans‐esterification. High‐DS products (DS= 2. elongation at break of 3‐26%. The thermal stability of the esterified products is higher than that of native starch.7‐3. and modulus of elasticity of 46‐113 MPa. K2CO3. Mechanical tests show that the products have tensile strength (stress at break) between 2. The reactions were carried out in DMSO using basic catalysts (Na2HPO4. The thermal and mechanical properties of some representative product samples were determined. The results were adequately modeled using a non‐linear multivariable regression model (R2≥0. vinyl laurate.75) are still partially crystalline.39) are totally amorphous whereas the low‐DS ones (DS= 1.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.96). and Na‐acetate). reaction temperature and type of the catalyst) on the Degree of Substitution (DS) of the starch laurate and starch stearate esters was determined by performing a total of 54 experiments.
86 . the physicochemical properties of the products were not mentioned. Most early studies mainly focused on the use of short‐chain carboxylic acid (C1‐C4). The high DS esters showed interesting properties. et al [15‐19] synthesized longer chain fatty esters (C8‐C18) of potato starch and corn amylose using fatty acid chlorides and pyridine. The development of novel biodegradable plastic materials capable of decomposing when given an appropriate environment and time is of utmost importance .Chapter 5 5. and particularly on the synthesis of starch acetate (C2) [11‐12].65) . most of the plastic waste is not biodegradable. The products were hydrophobic and the mechanical properties were considerably improved compared to native starch. Esterification is one of the oldest methods used to improve starch properties . 12] studied the synthesis and properties of of C1‐C6 esters of starch. It is cheap and abundantly available . The largest application area of plastics is the use as packaging material (37%) . Two examples using higher vinyl esters (vinyl laurate) were reported by Mormann. the research activities were mainly limited to the use of vinylacetate. packaging material. Typical examples are blending with polyvinyl alcohol . The mechanical properties of products with longer ester chains (C4 and C6) products and plasticized acetate esters were considerably improved compared to native starch. Unfortunately. The use of native starch as a building block for novel biodegradable polymers. Introduction Plastics are very useful materials and are used in large amounts (245 million ton per annum) [1‐2]. However. Sagar and Merill  studied the synthesis and properties of C4‐C6 esters of high amylose starch.1. The limited use is mainly because of a number of unfavourable properties of starch such as low moisture resistance. The global starch production was estimated at 60 million tons in 2004 . the mechanical properties were still not satisfactorily. et al  and Aburto. and food service‐ware . However. Alternative routes to avoid the use of acid‐chloride/pyridine combinations have been developed. 0. Mullen and Pacsu [11. 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. 3‐4]. Thiebaud. and incompatibility with hydrophobic polymers . Chemical or physical modification is required for successful applications as biodegradable polymers. Starch is an attractive feedstock for the synthesis of novel biodegradable plastics. however. and this causes serious environmental problems [1. The products were biodegradable. Examples are agricultural mulch . however. high brittleness. is rather limited to date. The use of vinylesters has also been explored [20‐22]. et al . polyethylene  and polycaprolactone  and chemical modification by esterification with organic acids [11‐19].
and acetic anhydride (Merck. Japan). The spectra were recorded at 50 oC.2. the thermal and mechanical properties of representative examples of the highly hydrophobic materials are described and discussed. All these chemicals were used as received. 5. Materials and Methods 5.2. Technical grade dimethyl sulfoxide (DMSO). After a first heating run from room temperature up to 200 oC to delete the thermal history of the material.1. Analytical grade vinyl stearate (Aldrich. The samples (about 10 mg) were placed in sealed aluminum cells. sodium acetate. TGA measurements were performed using a Perkin Elmer TGA 7 Thermogravimetric Analyzer. 73% amylopectine and 27% amylose) was purchased from Sigma (Germany).24 ‐ 2. 4‐N. vinyl laurate (Fluka. Analytical Equipment H‐ and 13C‐NMR spectra were recorded in CDCl3 on a 400 MHz Varian AMX NMR machine. Tc. Analytical grade methanol. 5. In addition. Germany). catalyst type and intake) on the DS of the products. and Tg) of each sample were determined from the spectra related to the cooling run and the second heating one. The samples were heated to 700 oC in a nitrogen atmosphere with a heating rate of 10°C/min. It was dried for 48 hour at 105 oC under vacuum (~1 mbar) to reduce the moisture content below 2 % before use.N‐ dimethylaminopyridine (DMAP).2. 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 . as recommended by Laignel et al . vinyl ester to anhydroglucose ratio. Long fatty esters of corn starch with a broad range in degree of substitution (DS= 0. T‐Bones samples (with thickness of 2 mm) for determination of the tensile properties were prepared using a melt press 1 87 . The catalysts for the esterification reaction were analytical grade potassium carbonate (Boom. Germany) were used without further purification. the Netherlands).2. Materials Corn starch (approx. and tetrahydrofuran (THF) were supplied by Acros (Belgium). each sample was cooled to ‐50°C and then heated again to 200 oC (heating rate 10 oC/min). This chapter describes systematic studies using Design of Experiments (DOE) to gain insights in the effect of process variables (temperature. and toluene were obtained from Labscan (Ireland). Germany). DSC analyses were performed on a TA Instruments DSC 2920. The thermal properties (Tm. pyridine. and disodium hydrogenphosphate dodecahydrate (both Merck. The experiments were modeled using non‐linear multivariable regression.96) were prepared by reacting the starch with vinyl laurate or vinyl stearate in the presence of basic catalysts in DMSO.
apparatus (Fontijne, Holland), operated at 150 oC and 150 bar for 3 minutes. The tensile tests were performed using an Instron Series IX Automated Materials Testing System 1.09 at 20oC and a crosshead speed of 30 mm/min. 5.2.3. Methods 22.214.171.124. Typical example of the preparation of laurate and stearate starch esters The corn starch (0.5 g) was gelatinized in DMSO (5 ml) at 70oC for 3 h. After the mixture became a homogenous, colourless solution, vinyl laurate or vinyl stearate (3‐5 mol/mol with respect to the AHG units) and the catalyst (potassium carbonate, or sodium acetate, or disodium hydrogenphosphate, 2‐5%‐w with respect to starch) were added. The reactor content was kept at 80‐110oC for 24 h. The product was precipitated using methanol (100 ml), and the liquid phase was removed by decantation. This precipitation‐decantation procedure was repeated twice using 50 and 25 ml of methanol, respectively, to purify the reaction product. Finally, the product was dried in a vacuum oven (70oC, 5 mbar) for 24 h or until constant weight. 126.96.36.199. Peracetylation procedure and Degree of Substitution (DS) determination The presence of un‐substituted hydroxyl groups in the products resulted in broad and overlapping starch resonances in 1H‐NMR spectra. A peracetylation reaction to substitute all of the remaining hydroxyl groups with acetate groups was therefore applied to obtain reliable DS values. Typically, the starch ester (0.1 g) was added to THF (4%‐w/v) and stirred at 55 oC until dissolution (typically 3 h). Subsequently, the peracetylating reagent (1 mol DMAP, 10 mol acetic anhydride, and 22 mol pyridine per mol AHG units) were added. The peracetylation reaction was conducted for 7 h at 50 oC. The product was precipitated by the addition of methanol and washed several times with methanol before finally dried overnight in a vacuum oven at 70oC and 5 mbar. The DS of the products was calculated using a procedure given in previous work . 5.2.4. Experimental Design The experiments were carried out in 6 blocks. The three variables used in each block were the vinyl ester to AHG starch mol ratio ( x1 ), catalyst intake ( x 2 ), and reaction temperature ( x3 ). All experiments in a block were conducted using a 3‐ variable, 2‐level Full‐Factorial Experimental Design with one center point, giving a total of 23+1 = 9 experiments per block. For each block, a given type of catalyst (Na2HPO4, K2CO3 or Na‐acetate) and vinyl ester (either vinyllaurate or vinylstearate) was applied. This gave a total of 54 experiments (2 types of vinyl
Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch
esters x 3 types of catalysts x 9 experiments per block). In a later stage, the type of catalyst was quantified using the pKb of the catalyst ( x 4 ). The ranges for the individual variables ( x1 ‐ x 4 ) are shown in Table 5.1. The mathematical analysis of the experimental data was performed with the software package “Matchad 13” (Mathsoft). The response (y, DS of the products) was modelled using the following expression:
y = β 0 + ∑ β i xi + ∑ β ii xi2 + ∑∑ β jk x j x k
i i j k≠ j
Here, βi, βii and βjk are the regression coefficients obtained by a multiple 2 regression procedure. One of the terms, namely x 4 , caused singularity of the matrix used for the modelling, and was therefore excluded. A t‐statistic was used to rank the individual regression coefficients (βi, βii, or βjk) according to their relative importance . An analysis of variance (ANOVA) was performed to check the adequacy of the model. The applied procedure is well described in the literature  and consists of calculating the sum of squares (SS) for the model and the error, together with the total sum of squares. In combination with the relative degrees of freedom (DF) it is possible to calculate the mean square (MS) for the model and the corresponding error. On the basis of the latter values, the F‐value for the model is calculated. With this information the P‐value for the model is determined. The latter value is related to the statistical significance of the model.
5.3. Results and Discussion
A schematic representation of the esterification reaction of starch with a vinyl ester is provided in eq. 5.2. The starch was gelatinized before the addition of the vinylester to make the starch OH groups more accessible for reaction. The reactions were carried out in DMSO for 24 h using three different basic salts (Na2HPO4, K2CO3, or Na‐acetate) as the catalysts. The products of the reaction were brownish gels which became lighter in colour after product precipitation and washing with methanol. After drying, the products were isolated as transparent, yellowish solids. The products were soluble in organic solvents such as toluene and THF.
OH O HO OH
R O CH2
O HO O R O
(1) R = laurate (C12) (2) R = stearate (C18)
The products were characterized by 1H‐NMR in CDCl3. The peaks of starch (δ 3‐ 5.5 ppm) and aliphatic hydrogen atoms of the fatty acid chain (δ 0.8‐2.5 ppm) were clearly present. The starch peaks were broad and overlapping, and only after the starch esters were peracetylated, the resolution of these peaks was considerably improved. The DS of the products was determined by using 1H‐NMR spectra of the peracetylated starch esters. The effect of three reaction parameters (molar ratio of vinyl ester to AHG units of starch x1 , catalyst intake x2 , and reaction temperature x3 ) on the DS of starch was studied using a full factorial experimental design with one center point. The ranges of the values of the independent variables ( x1 ‐ x3 ) are shown in Table 5.1. An additional variable x 4 (related to the basicity of the catalyst) was also included in order to obtain a general model for starch esterification. A total of 54 experiments were conducted. Table 5.1. Experimental Design Variables
Independent Variables Full Factorial Variables Vinyl ester to AHG‐ starch ratio [mol/mol] Catalyst intake [%‐w]a Reaction temperature [°C] Additional Variable Catalyst basicity (pKb)b
Low (‐1) Middle (0) High (+1)
2.0 3.5 5.0
80 95 110
K2CO3 (3.66) Na2HPO4 (6.8) CH3COONa (9.25)
3 4 5
in %‐wt based on starch intake, in the model (eq. 5.3.) it is transformed to 105 x catalyst mol amount. taken from ref .
The results of the experiment are given in Table 5.2.
00 5.00 5.00 5.00 4.70 2.00 3.00 3.50 0.12 2.93 0.00 3.00 5.00 2.00 3.00 3.00 2.00 5.00 2.54 2.00 5.72 2.00 5.27 1.00 3.12 0.00 2.50 2.00 3.00 5.40 1.00 3.00 5.00 Catalyst Temperature [%‐w] [˚C] 2.00 4.2.00 5.00 5.00 3.79 Experiment Vinyl Ester 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate laurate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate 91 .00 3.00 2.35 0.50 2.00 3.75 1.00 5.52 2.00 5.50 2.00 3.00 5.00 5.00 5.87 1.00 5.78 2.48 0.00 5.44 2.00 2. Overview of experimental data Catalyst Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 Na2HPO4 CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa CH3COONa Vinyl : AHGa [mol/mol] 3.67 2.00 2.00 4.41 1.28 2.52 2.07 0.00 5.00 5.23 2.53 1.00 2.00 2.00 5.00 3.00 3.96 0.00 5.82 2.47 1.00 5.00 5.00 5.00 5.27 1. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5.00 2.32 1.00 5.00 3.00 5.00 5.00 5.00 5.00 2.00 2.00 3.54 2.42 2.24 2.00 2.00 5.00 3.50 2.00 3.84 2.46 1.00 4.00 3.44 2.11 2.00 2.00 3.27 0.56 1.88 2.62 0.00 5.00 3.00 5.00 3.09 0.00 5.00 5.94 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 5.00 3.17 2.00 5.00 2.59 2.00 5.
0857 ‐0.5650 0.00 5.21 2.00 5.3.Chapter 5 45 46 47 48 49 50 51 52 53 54 a.96 2.41 mol ratio of the vinyl ester to AHG units of starch 5.).0120 0.0732 ‐0.3.64 2.0009 0.).0003 0.24 2.0023 0.00 2.2183 0.0470 0.0017 0.00 3.) Table 5.The modelled values for βij for the two esters are given in Table 5.1.00 5.0011 0.0010 ‐0. The type of catalyst was quantified using the basicity constant (pKb) in water [26‐27].00 5.0006 ‐0.90 2.0327 ‐2.1.00 5. Values for the coefficients of the DS model for starch laurate and stearate Coefficient Starch Laurate 9. stearate stearate stearate stearate stearate stearate stearate stearate stearate stearate CH3COONa K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 4.1979 Starch Stearate 17.64 2.98 2.50 95 80 80 110 110 80 80 110 110 95 1. 5.1748 0.1104 ‐0.55 2. Mathematical Modeling Modelling of the DS data for both vinylesters was performed using non‐linear multivariable regression based on the 4 independent variables (Table 5.0016 0.00 3.00 3.0027 ‐0.3.0008 ‐0.00 5.00 3.0001 0.6939 0.00 5.0249 ‐0.1568 ‐2.00 4.3.50 2.00 2.0066 0.00 3.3.00 5. DS = β 0 + β1 x1 + β 2 x 2 + β 3 x3 + β 4 x 4 + β 5 x1 x 2 + β 6 x1 x3 + β 7 x1 x 4 + β 8 x 2 x3 + β 9 x 2 x 4 2 2 2 + β10 x3 x 4 + β11 x 2 + β12 x3 + β13 x 4 (5.00 2.00 3.0008 0. The experimental data for both starch laurate and starch stearate esterification are best described with a quadratic model including interaction terms (eq.1082 ‐0.1947 β0 β1 β2 β3 β4 β5 β6 β7 β8 β9 β10 β11 β12 β13 92 .59 2.
Parity plots of the DS models for starch laurate and stearate Moreover.5 3 a.5 1 1. The parity plots of both models are given in Figure 5.0583 F 27.5 1. Analysis of variance for DS models for starch esterification Starch Laurate Model Error Total SS 16. Also in this case the reasonable R2 values (0.967 0.5 2 DS predicted DS predicted 2 1. Table 5.5 1 1.4.5 DS actual 2 2. the adjusted‐R2 values are very close to the R2 ones.5 1 1 0.0356 F 35. we also performed a PRESS analysis  (see corresponding R2 values in Table 5.5 0.2903 P‐value 1. Finally.8167 21.2569 0.5 DS actual 2 2. The R2 values for the models are 0.939 0.5 3 0 0 0.3399 0.875) indicate that the model correctly predict the 93 .6118 0. starch stearate Figure 5.970 (starch laurate) and 0.4986 16.7699 DF 13 14 27 MS 1.4.970 0.96 x 10‐8 R2 values R2 R2adjusted R2press 0.026 x 10‐7 R2 values R2 R2adjusted R2press 0. which represents an “internal” validation method for the model. respectively and it may be concluded that the models describe the experimental data well.967 (starch stearate).9531 0.805‐0.629 P‐value 1.805 3 3 2. starch laurate b.).1. which indicate  that all significant variables are included in the model.875 Starch Stearate Model Error Total SS 20.8386 DF 13 14 27 MS 1.5 2.4.1. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch The result of the analysis of variance is given in Table 5.5 0 0 0. and confirm this statement.945 0. The very low P‐ values indicate that the models are statistically significant.
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. A similar trend was observed by Dicke  for the acetylation of Hylon VII starch using vinylacetate.2.66) than CH3COONa (pKb 9. This was not the case for alkaline catalysts.25) and this leads to higher product DS values for K2CO3. the base strengths of the catalysts in DMSO are not known. 94 . The resulting anion will react with the vinyl ester to from the product. the performance of Na2HPO4 does not follow this trend.2. To evaluate the effect of each variable on the product DS. This is indeed the case when comparing the performance of K2CO3 with CH3COONa. the type of catalyst has the largest effect on the product DS for both the laurate and stearate esters (Figure 5. The pKb for the latter (6.82). Unfortunately.). K2CO3 is a stronger base (pKb 3. The product DS for Na2HPO4 was considerably lower (1.Chapter 5 products DS as function of the process variables within the range of experimental variables. it can be rationalised that the rate of the reaction (and thus the product DS) will be higher when using a stronger base. 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. two plots showing the dependence of the DS on reaction temperature and catalyst amount (in %‐w). such as carbonate or acetate salts leading to higher DS esters. However. 22]. The highest product DS values were obtained using K2CO3. These results may be rationalised by considering the role of the catalyst in the modification reaction. 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.8) is intermediate between that of the other two catalysts. On the basis of the sequence. Of all the variables studied. 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.18) and CH3COONa (1. for both starch laurate and starch stearate are given in Figure 5. Catalyst performance of CH3COONa was slightly less than for K2CO3 whereas the lowest DS products were produced when using Na2HPO4 as the catalyst.00) than for K2CO3 (2. whereas catalyst performance is considerable lower.
starch laurate b. As expected and in 95 . Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch a. starch stearate Figure 5. although to a lesser extent. 3D contour plot of the DS as a function of reaction temperature and catalyst intake (at constant vinylester to AHG mol ratio of 4) Besides the type of catalyst.2.2. the catalyst intake and reaction temperature also affect the product DS (Figure 5.).
the vinyl ester to AHG ratio has the smallest effect on the product DS.3.39) were determined.45) and high DS values (DSlaurate = 188.8.131.52. higher reaction temperatures and catalyst intakes lead to higher product DS values.2.26.75. Of all variables studied. . . heating Exothermic Tm = 21 oC cooling Endothermic Tc = 12 oC -20 -10 0 10 Temperature ( C) o 20 30 40 Figure 5.Chapter 5 line with studies on starch acetates .45 (. Representative DSC spectra for the starch stearate are given in Figure 5. DSC analysis of starch stearate esters ( ) DS 1. DSstearate = 1. 5. Product Properties The thermal and mechanical product properties for two starch‐laurate and stearate samples with medium (DSlaurate = 1.39) 96 . and the results for all samples are summarised in Table 5. DSstearate = 2.) DS ( 2.
= not detectable The thermal behavior of the products is a strong function of the DS.45 Tm [˚C] n.a 22 n. To the best of our knowledge. in the present work.). These values are reduced considerably compared to virgin corn starch. which is known to contain crystalline areas. esterification of starch with vinyl laurate and stearate reduces the crystallinity of the starch considerably and leads to a lowering of the transition temperatures.39 1. n. Thermal transitions at relatively high temperature (melting. octadecane. crystallization. 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. Changes in the transition temperatures of native starch by esterification have been reported in the literature [13‐14. In particular the behaviour at relatively high temperatures (>0 oC) is further discussed.d. At moderate DS values the products still display a melting temperature (at 20‐30 oC) and a crystallization temperature (at 10‐25 oC). 17].4. close to the value obtained in our study for the medium DS starch stearate (21 °C).26 1.a 27 n. The melting point of the potato starch stearate was close to that of model compounds for the stearate side chains (methyl stearate.d.75 2. In any case.a 21 Tc [˚C] n.a 12 a. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. 97 . 22].8 . 28‐30 oC) and the authors concluded that the transition temperatures are solely determined by the side chain without any significant contribution of the starch backbone.d.d. However.d. 40‐42 oC. implying that these products are fully amorphous. these are the first examples of completely amorphous starch esters. 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.5. TGA analysis was performed to study the thermal degradation behavior of the starch laurate and stearate samples (Figure 5. Unfortunately the thermal transitions for the virgin material lie above the degradation temperature and can therefore not be accurately determined . 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). For instance a Tm of 32 oC was reported for a potato starch stearate ester with a DS of 1. or glass transition above 0 oC) were even absent for the high DS products. especially in the amylopectin part [15.
Chapter 5 100 80 %-weight 60 40 20 0 20 100 200 300 400 o 500 600 700 Temperature ( C) Figure 5. The mechanical properties (stress at break. ( ): native starch ( ): starch laurate (DS 2.): starch stearate (DS 2. Native corn starch degrades at lower onset temperatures (250‐300 oC) than the corresponding esters (300‐350 oC). .).26) (. The difference in thermal stability between the laurate and stearate sample is limited. elongation at break and elasticity modulus) of the products were determined (Table 5.6. . 98 . TGA analysis of various starch samples.39) The TGA (Figure 5. Similar trends were reported for amylose octanoate‐ stearate‐laurate esters  and potato starch octanoate‐ laurate esters .) curves clearly show starch esterification with either laurate or stearate results in products with enhanced thermal stability.4.4.
These considerations strongly point out the fact that the mechanical behavior can be coarsely (plastic vs elastic behavior) but also finely (stress and strain at break but also modulus values) tuned by the chemical structure of the fatty acid chains as well as the DS values.7 ± 4.1 Modulus [MPa] 59.0 112. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. The values of the yield strain.9 46.1 ± 15. This indicates that their mechanical behavior is comparable to that of typical engineering polymers .9 ± 14. calculated as the ratio between the stress at yield (σy.75 2. but higher elongation at break. This product is more rigid compared to the product with medium DS value (higher tensile strength and elasticity modulus.5 ± 1. observed for the high DS starch stearate sample. determined from the stress‐strain curves as the stress at which the material ceases to be linearly elastic) and the modulus. Both starch laurate samples display. Mechanical properties of the starch esters DS 2. A deeper understanding of the mechanical behavior can be gained by investigating the shape of the stress‐strain curves (typical examples in Figure 5.1 ± 1. 17].26 1. independently of the DS values.45) displays a very similar behavior with respect to the laurate samples (i. an increase in product DS results in lower tensile strengths and modulus of elasticity.3 21.2 2.e.7 ± 4.45 Stress at break [MPa] 3.).025) lies still in the typical range of engineering polymers but also close to that typical of composite materials . thus indicating a relatively more rigid material with respect to all other samples. and lower elongation at break).1 Elongation at break [%] 26.3 ± 2.0 ± 2.7 ± 1. is roughly 0.6.05 for both laurate samples.1 82.0 3.8 Starch Ester Starch laurate Starch laurate Starch stearate Starch stearate All materials generally show tensile strengths between 2. elongation at break between 3‐26%.2 8. starch stearate at relatively high DS values (2.05).7 ± 3.5 MPa. For starch laurate.2 ± 0.3 3. The corresponding yield strain (roughly 0. and modulus of elasticity between 46‐113 MPa. On the other hand. This is in agreement with previous studies regarding the mechanical properties of starch esters with different chain length of the fatty acid moiety [13.5.39 1. however. Starch stearate with a relatively low DS (1. a clear plastic behavior with a maximum in the stress‐strain curves. plastic deformation and yield strain of about 0. An inverse behaviour is. 99 .0 3.7‐3.39) does not show any plastic behavior but only an elastic one.7 ± 14.
a rough comparison can be made between the mechanical properties of the esters described in this chapter and those for reported for related starch esters synthesized with alkanoyl chloride as reagent and pyridine as catalyst  (Table 5.): starch stearate (DS 1.7.Chapter 5 6 5 4 stress (MPa) 3 2 1 0 0 5 10 15 20 strain (%) 25 30 35 Figure 5. ‐): starch stearate (DS 2. Although there are many differences in the synthetics methods as well as in testing conditions and procedures. 100 .45) (‐ . .5. ( ): starch laurate (DS 1.75) ( ): starch laurate (DS 2.39).26) (. Stress‐strain curves for starch esters. .).
confirms that the corn starch esters synthesized in our research are relatively rigid materials as compared to the potato starch esters.7 ±0.7.6 9 ±2 High DS (2.7) 0. we show that although no crystallization of C18 occurred (see DSC data of the high DS material in Table 5. 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.a.8) 3. n.3 High DS (2. 101 .5 MPa) is coupled with a low elongation (26 %). While for corn starch (DS 2.1 ±1.3 Potato Starch Stearate Ester Low DS (1.0 ±2. the high DS material is still more rigid.2 MPa for corn starch esters. In this respect the difference in stress and elongation at break for the high‐DS starch laurate sample is striking.7) 1. 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) . merely because of the structural property of the C18 side chain with respect to the starch backbone.a.7.4 1500 ±8.0 26.26) 3. n. 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.a.26) relatively high stress at break (3.7 ±3.5 ±1.a.5. High DS (2.2 Stress at Break [MPa] Elongation at Break [%] a.7 ±1.7 ±0.1 High DS (2.2 8.7 ±4.7 MPa) coupled with a significant (1500%) elongation.= not available Inspection of the mechanical properties of the starch esters as shown in Table 5. In agreement with our own data.75) 3.0 3. 9% for potato starch).3 10 ±2 Property Low DS (1.1 21.1 ±15.). 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). On the other hand at low DS values comparable stress at break (3.9 ±0. They related this inverse property of starch stearate to crystallization of C18 side chains.7 MPa for potato starch ones) are coupled with significant differences in the elongation (22% for corn starch esters. an unexpected behavior was also observed for the mechanical properties of the starch esters from potato starch.2 ±0.a.45) 3. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch Table 5. Here. 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. for potato starch exactly the opposite is observed: relatively low stress (0.6 Corn Starch Stearate Ester Low DS (1. Moreover. as confirmed by DSC.39) 2. DS) which mainly determines the thermal and mechanical behaviour. 3. but the interplay between those factors.
07‐1. the chemical composition of the starting materials (either starch or the ester precursor) as well as the processing parameters affect the final product DS values and as such provide an effective toolbox to modulate the desired product properties for a given application.5 for starch laurate. K2CO3.5. and catalyst basicity) to the DS of the starch ester products were developed.Chapter 5 5. including statistical modelling. 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.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐starch mol ratio. reaction temperature. 0. 5. while the use of Na2HPO4 resulted in low‐medium DS products (0. catalyst intake. Statistically adequate (R2 ≥0. Thermal and mechanical behaviour of the samples of different DS values clearly shows that the DS represents one of the crucial factors affecting the final product properties. The models may be used to determine the appropriate process conditions to obtain a product with a pre‐defined DS. Nomenclature β : regression coefficients obtained by a multiple regression procedure [various units. Conclusions A systematic study. Reaction temperature and catalyst intake also affect the product DS although to a lesser extent than the type of catalyst. and Na‐acetate).4‐3 for starch stearate). 1.4. The use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2.9 for starch laurate.1‐2. Thus.5 for starch stearate). The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. The DS of the products is a strong function of the basicity of the catalyst. on the synthesis of corn starch esters with long chain fatty acids is described. average value of mole of substituted –OH per mole of anhydroglucose (AHG) units [‐] 102 .3‐1. Comparison with literature data indicates that the properties may be (fine) tuned also by the starch source.
F. A. G. The Royal Veterinary and Agricultural University. Selke. Denmark. Prod. W. Germany. S. 103 . G. 2002. E. . R. 2000. Packaging. Applications.dk/isi/stat/rawmaterial. D. Marsden: The challenge of domestic waste disposal. 2008. Westhoff. Hanser Publishers. Plastics Europe. Department of Dairy and Food Science. 1986. 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. Copenhagen. Levy). The compelling facts about plastics: An analysis of plastics production. accessed on August 01. References . Aarhus. . Culter: Plastics Packaging. UK.6. S.html. Ehrenstein: Polymeric Materials. An Introduction to the New Science of Biodegradable Plastics. and Regulations. demand and recovery for 2006 in Europe. USA. J. 1980. Denmark: http://www. Dev.starch. Doane: Starch‐based blown films. Frederiksberg C. Website of International Starch Institute.W. Belgium. Hernandez. Stevens: Green Plastics. and Biopolymers. Weber). M. 27‐29 August 2000 (Ed. H. Processing. Structure – Properties ‐ Applications. Res. Chem. Princeton University Press. Conference Proceedings. P.J. R. C. . J. Denmark. Association of Plastics Manufacturers. Hanser Publishers. 19. . 2000. . in Packaging in the Environment (Ed. E. Brussels. Munich. Bastioli: Global Status of the Production of Biobased Packaging Materials. Otey. Eng. M. 2008 . C. 592‐595. Glasgow. Blackie Academic & Professional. Germany. 2000. New Jersey. Munich. in The Food Biopack Conference: Foodstuffs. M. . Ind. Properties.
. Sci. . A. J. G. E. Starch‐ Starke 2005. Z. . . L. M. J. Pacsu: Starch studies: preparation and properties of starch triesters. Polym. 1999. Biobased packaging materials for the food industry. characterization. Aburto. 145‐152. Prinos. 35. I. Polym. F. M. Chem. Thiebaud. Starch‐Starke 2004. 56. .Chapter 5 . Sci. 1209‐1217. Prinos. 34. .. R. A. 1996 104 . Sci. . Thesis. Polym. Macromol. Aburto. report of Food Biopack Project (Ed. Y. H. S. Carbohydr. D. Alric. D. J. . Thiebaud. J. W. K. 57. J. Senocq. 1997. A. 28. A. D. 137‐144. Merrill: Properties of fatty‐acid esters of starch. . E. C. Prinos. 2000. W. C43. 65. Alric. Carbohydr. B. C. E. Ind. 51. Panayiotou: Properties of fatty‐acid esters of starch and their blends with LDPE. Borredon. Hamaili. A. I. Rev. Janssen. 101‐112. Eng. Starch‐Starke 1999. 2003. Aburto. de Graaf. C. Borredon: Preparation of long‐chain esters of starch using fatty acid chlorides in the absence of an organic solvent. J. Beenackers: The kinetics of the acetylation of gelatinised potato starch. Wang: Properties of starch blends with biodegradable polymers. the Royal Veterinary and Agricultural University. Appl. 58. 1942. Pacsu: Starch studies: possible industrial utilization of starch esters. Sagar. E. Alric. D. I. S. Polym. M. 132‐135. de Graaf: Ph. Borredon. J. 51. 1997. Wang. S. Bikiaris. L. and biodegradability of fatty‐acid esters of amylose and starch. Eng. Mouysset‐Baziard. Panayiotou: Properties of octanoated starch and its blends with polyethylene. status and perspectives. Rijksuniversiteit Groningen. 1995. Department of Dairy and Food Science. J. 302‐307. K. J. 381‐384. Alric. Frederiksberg C. Chem. 74. Polym. Alric. Mormann. 34. Aburto.D. Borredon: Organic solvent‐free transesterification of various starches with lauric acid methyl ester and triacyl glycerides. Starch‐Starke 1999. 118‐121. I. Sci. Appl. P. Part C: Polym. Aburto. Appl. J. W. 1943. Borredon: Free‐solvent synthesis and properties of higher fatty esters of starch ‐ Part 2. Mullen. J. . J. G. J. J. E. Yang. Ind. 1440‐1451. E. R. Borredon. Broekroelofs. 1647‐1656. Weber). Aburto. E. 705‐721. . I. I. J. C. 1995. X. Thiebaud. E. A. E. Bikiaris. . W. 385– 409. C. Panayiotou: Synthesis. Alric. Denmark. . Mullen. J. Bikiaris.. Al‐Higari: Acylation of starch with vinyl acetate in water.
M. 2007. 255‐263. Massiot. A. F. Prentice‐Hall. Ashby. 2001. Amsterdam. M. L. P. R. A. . Manurung. Day. Processing and Design. Montgomery: Design and Analysis of Experiments 5th edition. . H. Dicke. Underwood: Quantitative Analysis 5th edition. John Wiley & Sons Inc. Butterworth‐Heinemann. B. J. M. R. A. 1986. the Netherlands.. B. Science. Junistia. D. . 105 . R. Experimental and Modeling Studies on the Synthesis and Properties of Higher Fatty Esters of Corn Starch . G. Res. D. L. 298. C. C. A. 1997. 251‐260. J. Carbohydr. USA. Materials: Engineering. Nuzillard: Proton NMR spectroscopy assignment of D‐glucose residues in highly acetylated starch. Bliard. 2008. H. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Cebon. New York. . USA. accepted for publication in Starch‐Starke. K. . Janssen. New Jersey. 11. Cellulose 2004. Laignel. L. A straight way to regioselectively functionalized polysaccharide esters. Shercliff. Picchioni. Sugih.
Keywords: starch‐PCL blend. at relatively high GMA and BPO intakes. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. the mechanical behavior of the compatibilized blends is in general different from that of pure PCL and of the corresponding uncompatibilized blends. This discrepancy could be explained on the basis of the functional groups (GMA or DEM) distribution along the PCL backbone. glycidyl methacrylate. diethyl maleate.Chapter 6 Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends Abstract The synthesis of two reactive interfacial agents for starch‐PCL blends. The 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. In particular the elastic modulus for the compatibilized blends is significantly higher than that of uncompatibilized ones. At relatively low starch intakes. PCL‐g‐ glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM) is decribed. despite the expected differences (favorable to PCL‐g‐GMA) in the in situ formation of the compatibilizers. The compounds were prepared by reacting a low molecular weight PCL (Mw 3000) with glycidyl methacrylate or diethyl maleate in the presence of benzoylperoxide at 130°C. As a result. compatibilizer. PCL‐g‐DEM has at least a comparable performance with respect to PCL‐g‐GMA. The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). grafting .
Unfortunately. polycaprolactone (PCL). Starch modifications to improve the product properties like enhanced hydrophobicity and mechanical properties were already reported in the early 19th century [1. As a consequence. However. The ester bonds are susceptible to attack by water and this leads to enhanced biodegradability. of polar groups (usually epoxides or anhydrides [13‐18]) able to react with the – OH groups along the starch backbone. these synthetic polymers are poorly or non‐biodegradable. PCL has gained considerable interest for possible applications in the fields of packaging materials and medical applications [10‐11]. upon mixing) by using a functionalized PCL. This leads. To overcome this issue. 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 ).e. synthetic biodegradable polymers have been applied. Introduction Starch is a cheap and abundantly available natural polymer with very good application perspectives in the area of biodegradable plastics. Unfortunately. Blending of starch and synthetic biodegradable polyesters has been widely applied for the synthesis of fully‐biodegradable products. is known to be degraded with ease by microorganisms widely distributed in nature . When using ungelatinized starch as a component in the blend.1. Such structure is rather difficult to achieve by simple copolymerization methods and it is usually prepared in situ (i. A well‐known biodegradable polyester. such as polyethylene (PE) or polystyrene (PS) [3‐6]. along the backbone.Chapter 6 6. 2]. blends of hydrophilic starch and hydrophobic biodegradable polyesters exhibit phase separation  due to differences in polarity of the building blocks. as in this study. Among these. to an ideal compatibilizer having both PCL and starch blocks. One of the well known modification strategies is blending the starch with polymers displaying a stronger hydrophobic character and better mechanical properties. The latter displays the presence. an interfacial agent) may be used to improve the interfacial association between the two polymer phases. Aerobic soil‐burial experiments showed that the mechanical properties of PCL films decrease rapidly in time . for starch/PCL blends. 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.e. This phenomenon is highly undesirable and limits the application range considerably . 108 . it would be actually more accurate to define the block copolymer as an “interfacial agent”. a compatibilizer (i. To reduce the tendency for phase separation. 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. polyesters are considered very promising alternatives .
Methods 6. Mw=50000) from Solvay Caprolactone.8%) from Labscan. and benzoyl peroxide 75% (Merck. 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. Compatibilizer synthesis The compatibilizers were prepared in a Brabender Plasticorder PL2000 batch‐ kneader (chamber volume 35 cm3). The starch was dried for at least 24 h at 110oC under vacuum (approx. Materials Polycaprolactone (PCL) CAPA 2304. UK was used for the preparation of the interfacial agents.2. 6.2. Exploratory studies on the synthesis of PCL‐g‐GMA and its applications for starch/PCL blends have been published [13. Germany and methanol (99. 109 . an absolute novelty of the present research. 16].8%) from Merck. after which the equipment was stopped and the chamber was opened to collect the samples. This low molecular weight PCL grade was used without further purification. The materials were mixed for another 5 minutes. Materials and Methods 6. xylene (99. The synthesis and application of PCL‐g‐DEM is. to the best of our knowledge. however. Glycidyl methacrylate 97% purity (Aldrich). >99%) was obtained from Acros. The kneader was heated to 130 oC and PCL (CAPA 2304) was added while maintaining a rotational speed of 80 rpm. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM). Belgium. This chapter describes a systematic study on the synthesis of two reactive interfacial agents for starch‐PCL systems. 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. Tetrahydrofuran (THF.2. After the PCL was melted (1‐2 minutes). diethyl maleate ≥97% purity (Fluka).2. Corn starch (with approx. The intake of reagents was maximally 75‐80% of the volumetric volume to ensure proper mixing. a solution of BPO in GMA or DEM was added drop by drop over a period of 5 minutes. 73% amylopectin and 27% amylase) was obtained from Sigma and high molecular weight PCL (CAPA 6503.2. ~1 mbar) prior to use. Ireland. Mw=3000 from Solvay Caprolactones.2. UK. systematic studies and subsequent quantification of the functionalization reaction has not been reported to date. Germany) were used as received.1. 6. Intakes for each experiment are given in Table 1 and 2. The various compatibilizers have been tested for their performance in PCL‐starch blends.1.
The solvent was decanted and the solid product was dried in a vacuum oven (40 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. Before analysis. The resulting suspension was filtered and precipitated using methanol (450 mL) at 6‐8°C. Infrared spectra were collected with a FT‐IR apparatus in the ATR mode using a Spectrum 2000 instrument from Perkin Elmer. 6. 5 mbar). Work‐up of PCL‐g‐GMA products  To remove unreacted GMA monomer and GMA homopolymer.8%. PCL was added to the chamber followed by the addition of the starch and the reactive compatibilizer. Analytical Methods H‐NMR measurements were performed using a 400 MHz Varian AMX Oxford NMR apparatus with CDCl3 (99. Scanning Electron Microscopy (SEM) was performed using a Jeol 6320 F Scanning Electron Microscope. The solvent was decanted and the solid product was dried in a vacuum oven (40 oC.2. the samples were covered with a paladium/platinum conductive layer of 3 µm thickness. 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). Tensile tests were performed using an Instron 4301 1 110 .5 h.2.2.4. and then filtered. created using a Cressington 208 sputter coater. 184.108.40.206. the chamber was opened and the resulting material was collected. and the product was precipitated at 6‐8 oC.2. Subsequently. Digital Scanning Calorimetry (DSC) measurements were performed using a Q1000 TA Instruments equipped with a TA Instruments DSC cooling system. Methanol (450 mL) was added to the filtrate. PCL‐g‐GMA (5 g) was dissolved in 50 ml THF. stirred for 1.Chapter 6 6.3. 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).3. 5 mbar). 6. An operation temperature of 170°C and a rotation speed of 80 rpm were applied .2. The content was blended for 15 minutes. Aldrich) as the solvent. 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 . The error on the transition temperature is assumed to be ± 1 oC and 5 % of the calculated values for the corresponding enthalpies. PCL‐g‐DEM (5 g) was dissolved in xylene and refluxed at 150 oC for 2 h.
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).2 ppm) side chains with that of a characteristic proton resonance of the PCL backbone (‐CH2‐ protons at δ 4. For a given sample/blend. 19].0 ppm [13. This procedure is described in detail in the literature  and consists of calculating the sum of squares (SS) for the model and the error.) The FD was calculated using 1H‐NMR by comparing the area of protons belonging to the GMA (‐CH< proton at δ 3. On the basis of the latter values. The corresponding value for every blend was calculated as an average of the 8 measurements while the standard deviation was taken as absolute error on the average values. 6. The latter is a measure of the statistical significance of the model. it is possible to calculate the mean square (MS) for the model and the error.2.1. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends machine. strain at break (ε). A 5% relative error in the peak area of the NMR spectra was assumed. 111 . When the relative degrees of freedom (DF) are known. Statistical Modeling The influence of different processing parameters on the FD values has been determined by performing a multivariable regression procedure on the available data.1. 15‐16.2 ppm) or DEM (‐CH2‐ protons at δ 4. 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. As a result we were able to obtain a model for the FD of the reaction.4. 8 different T‐bones were used. For every T‐bone.3. stress at break (σ) and modulus (E) were measured. The validity of the model was determined by performing an analysis of variance (ANOVA. 6. The T‐bone samples were prepared using a Fontijne Holland TH 400 hot‐ press. the F‐ value for the model is determined followed by the P‐value.2. Table 3). leading to a 10% relative error in the FD values.
1. Results and Discussions 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.Chapter 6 6.).1. 112 . 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 6.3.1. Preparation of the Reactive Compatibilizers Twelve compatibilizers were prepared by either reacting glycidyl methacrylate (GMA) or diethyl maleate (DEM) with low molecular PCL as the starting polymer and benzoyl peroxide (BPO) as the radical initiator (Scheme 6.3.
Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends
Figure 6.1. Typical 1H‐NMR spectra of PCL‐g‐GMA (top) and PCL‐g‐DEM (bottom) Peak assignments were based on available data reported for related products [13, 15‐16, 20]. The FD values and the thermal properties of the products are shown in Table 6.1. (PCL‐g‐GMA) and Table 6.2. (PCL‐g‐DEM).
Overview of experiments for the PCL‐g‐GMA compatibilisers a
Intake FD (%) ‐ 5.7 15.2 27.9 3.2 45.2 c 6.3 T cryst. ( C)
(%‐mol) b GMA BPO ‐ 0.6 0.6 0.6 0.3 1.1 0.3
ΔH cryst. (J/ g‐PCL) 75 72 68 67 73 66 71
T melt. ( C)
ΔH melt. (J/ g‐PCL) 79 72 69 68 73 68 67
PCL PCL‐g‐GMA 1 PCL‐g‐GMA 2 PCL‐g‐GMA 3 PCL‐g‐GMA 4 PCL‐g‐GMA 5 PCL‐g‐GMA 6
‐ 12 24 36 6 24 12
27 22 19 29 25 20 28
51 47 47 48 47 46 47
Experiments were carried out at 130 °C. %‐mol with respect to the CL repeating units in the PCL c based on the soluble fraction of the compatibilizer
Table 6.2. Overview of experiments for the PCL‐g‐DEM compatibilisers a
Intake Sample (%‐mol) b DEM PCL PCL‐g‐DEM 1 PCL‐g‐DEM 2 PCL‐g‐DEM 3 PCL‐g‐DEM 4 PCL‐g‐DEM 5 PCL‐g‐DEM 6
FD (%) ‐ 3.5 2.1 7.2 3.6 6.5 0.9
T cryst. ( C)
ΔH cryst. (J/ g‐PCL) 75 70 68 64 61 63 70
T melt. ( C)
ΔH melt. (J/ g‐PCL) 79 72 62 66 60 66 69
BPO ‐ 0.6 0.6 0.6 1.1 1.1 0.3 27 27 30 21 26 25 30 51 44 47 45 42 41 47
‐ 15 30 45 30 60 7.5
Experiments were carried out at 130 °C %‐mol with respect to the CL repeating units in the PCL
One of the PCL‐g‐GMA compatibilizers (PCL‐g‐GMA 5, see Table 6.1.), was only partly soluble in CDCl3, and therefore the FD is based on the soluble fraction of the compatibilizer only. The presence of an insoluble fraction, combined with a relatively broad molecular weight distribution (as shown by GPC, but not shown here for brevity), suggests that cross‐linking occurred during this experiment. In general, the FD of the PCL‐g‐GMA (3.2‐45.2%) products is much higher than those of PCL‐g‐DEM (0.9‐7.2%). This difference may be either due to the difference in mutual solubility of the GMA and DEM in PCL or differences in the molecular
Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends
mechanism of the grafting reaction. The mutual solubility may be expressed in terms of the differences of solubility parameters of PCL and the reagents . The difference between the solubility parameters of GMA and PCL is 0.29 cal1/2cm‐3/2, while it is much higher (6.3 cal1/2cm‐3/2) for DEM and PCL . Thus, GMA is likely better soluble in PCL, leading to higher values for FD of the products, as confirmed by our experiments. However, the higher experimental grafting efficiencies for PCL‐GMA may also be rationalised by considering the reaction mechanism for the preparation of the compatibilisers. GMA molecules may either react directly with a radical at the PCL backbone or with a radical present on an already coupled GMA molecule. The latter leads to longer GMA grafts on a PCL backbone . A simplified representation of the reactivity of GMA is shown in Figure 6.2.
O (CH2)4 CH2 C O + RO
O (CH2)4 CH
+ GMA O (CH2)4 CH H2C H3C C
C O CH2 HC O O
Termination + H or + R or + RO (CH2)4
+ k GMA O CH C O
H2C H3C C
C O CH2 HC O O
Figure 6.2. Simplified scheme of the GMA Grafting Reaction Mechanism  The reactivity of DEM in radical reactions is expected to be different from that of GMA. Previous studies on maleic anhydride (MA), a compound resembling the
116 . The results are graphically provided in Figure 6.e. BPO=1.3% mol/mol-CL units DEM. 22].1. the length of a MA graft is always unity whereas longer grafts are possible for GMA. 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. subsequent reactions of MA to an already grafted MA molecule hardly occur. However. 6. BPO=0.3.2.1. 30 25 20 FD (%) GMA.Chapter 6 chemical structure of DEM. Our experimental findings. It is most probably related to the fact that GMA is able to propagate to longer grafted chains when reacted with PCL (Figure 6.3.6% mol/mol-CL units DEM.). This trend is independent of the BPO amount and matches with data reported by other groups [13. Hence. higher FD values for GMA than for DEM are in line with this explanation and support the proposed molecular reaction mechanisms. i. 130°C) Higher GMA intakes lead to higher FD values. BPO=0. Effect of GMA and DEM to CL‐units ratio (mol/mol) on the FD of the products (constant PCL intake.6% mol/mol-CL units 15 GMA. Thus an increase in the GMA intake will provide more monomer available for the growing of the grafted chains leading to higher FD values. BPO=0.1% mol/mol-CL units 10 5 0 10 20 30 40 50 60 Substrate Intake (mol% on CL units) Figure 6. showed that MA reacts easily with a radical on the PCL backbone.3.
However. DEM=30% mol/mol-CL units 0.7 0.).6 0. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The experimental trend for DEM is different.4.1. Such behavior is slightly in contrast to what observed for maleic anhydride [13.3.4 0.4. doubling the amount of initiator results in considerable higher product FD. The FD values are within a rather narrow range. constant substrate to PCL ratio) For high GMA to CL ratio (24%‐mol/ mol CL units).9 1 1.3 0. no detectable influence of the BPO amount on the FD is observed (Figure 6. 6.8 0. for which a levelling off and eventually a decrease of the FD values for relatively higher MA amounts has been observed.5 0. This will lead to a higher proportion of PCL radicals by hydrogen abstraction from the polymer backbone and thus to higher FD values. at relatively lower GMA intakes. Effect of the BPO intake on the product FD (130°C. although a slight increase in the FD values might be appreciated.2. 17]. Apparently.2 GMA. GMA=12% mol/mol-CL units DEM. The use of higher initiator concentrations will result in an increase in the number of formed radicals.).1 1. there 117 .2 BPO Intake (mol% on CL units) Figure 6. GMA=24% mol/mol-CL units GMA. 15. Such discrepancy is probably related to the relatively low intake of DEM as well as to the different reactivity of DEM compared to MA . Effect of the BPO intake on the product FD The effect of the BPO intake on the FD was studied by using different intakes of BPO (Figure 6.4. 50 45 40 35 30 FD (%) 25 20 15 10 5 0 0. 17]. These results are in line with earlier work [15.
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).5431niδ + 1. The model provided in eq.2.3. gives the best description of the experimental data: FD = −1. the BPO and monomer intakes are considered as independent parameters. The effect is however much less pronounced than for GMA.1. This leads to the following equation: FD = f (nm .) where nm is the molar amount of monomer in the feed. the possibility of side reactions will become significant. defined as the difference in solubility parameters between PCl and the substrates.2. Here.0325nmδ + 0. i.6022nm ni − 0.1.8875 + 0.e. namely an increase in the BPO intake results in higher product FD. 6. recombination of (macro) radicals. ni the molar amount of initiator in the feed and δ the difference in solubility parameters calculated using group contributions .3. and 6. as well as several transfer reactions might be responsible for the observed trend .Chapter 6 is an optimum ratio between the BPO concentration and available monomer (GMA in this case) on the FD. 17]. However. This phenomenon is likely related to the different nature of the radical grafting mechanism of GMA and DEM on PCL as mentioned before.) 118 . To quantify synergic effects of monomer and initiator molar intakes on the FD of GMA and DEM on the PCL backbone. δ ) (6. Modeling of the combined effects of the GMA/ DEM and BPO intakes on the FD Previous studies showed the importance of the initiator to monomer ratio on the FD values [15. In addition. ni . In particular the occurrence of “cage effects”.3. 6. these investigations focused on a better understanding of the individual variables by studying the effect of higher peroxide and monomer intakes while keeping for example their ratio constant. 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. If many macroradicals are created in the system (relatively high BPO intakes) at low GMA intake.2729nm niδ (6.3. A doubling of the initiator intake for PCL‐g‐DEM only results in a 70% increase in the FD (compared to 300% for PCL‐g‐GMA). the mutual solubility of DEM in PCL and DMA in PCL was included in the model by using a parameter δ. a statistical model has been developed by performing a multivariable linear regression on the data reported in Table 6.
5 0. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends The analysis of variance data are given in Table 6.2 10 15 20 25 30 35 40 GMA Intakes (mol% on CL units) 45 50 a.2 10 20 GMA Intake (mol% on CL units) 40 30 50 0.17 F 44.928 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 ).358 1909. 1 0.908 10.6 0. SS 1828 81.941) also suggests that all important variables have been included in the model.6. This was also confirmed by inspection of the residue distribution by a normal probability plot (not reported here for brevity) . respectively. Table 6.5.9 60 70 40 50 80 BPO Intake (mol% on CL units) 0.3. and Figure 6.5..957) and its closeness to the adjusted R2 (0. Graphical representation of the FD models for PCL‐g‐GMA and PCL‐g‐DEM are given in Figure 6. b.3 10 10 20 30 20 0.6 20 10 20 40 50 30 60 FD (%) 40 FD (%) 30 40 20 0 1 0. 119 . the interaction between (PCL‐monomer) intake and solubility parameter difference ( nmδ ) and the interaction between the latter factor and the initiator intake ( ni δ ).358 DF 4 8 12 MS 456.3.4 0. Graphical representation of the FD model for PCL‐g‐GMA. a.8 0. The R2 value for the model (0. b. Figure 6.4 BPO Intake (mol% on CL units) 0. 3D plot. The very low P‐value implies that the model is statistically significant.3. Model Error Total Analysis of variance for the FD model provided in eq 6.7 0.8 0. Contour plot.
5 0. This induces irregularities and is expected to result in a lowering of the Tc and the Tm. which is expected to lead to a Tc reduction.) of the compatibilizers were determined by DSC. Furthermore. 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. 6. a.4. Similar observations have also been made by Kim.6 2 FD (%) 5 2 0 1 0. and a favored nucleation 1 2 50 120 . For all samples.2 10 15 20 25 30 35 40 DEM Intake (mol% on CL units) 45 7 a. the melting temperature and the relative enthalpy decrease with respect to pure PCL. a higher BPO intake leads to a reduction of the FD values while an opposite trend is observed at lower DEM intakes. From a more practical point of view. the crystallization temperature and enthalpy are not a clear function of the FD values.3 5 6 0. b.6 0. Such changes in the thermal properties compared to pure PCL may be caused by the introduction of grafts on the PCL chains. 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. The random behavior of the Tc as function of the FD values is probably the result of two concurring effects: the presence of irregularities. and 6. for DEM a clear transition is observed. Remarkable is once more the different trends for the two substrates. For relatively high DEM intakes (> 30 % mol/mol). While for GMA an increase in BPO or monomer intakes invariably leds to higher FD values.2 10 20 DEM Intake (mol% on CL units) 40 30 50 6 0. 7 6 3 3 4 Figure 6. radical initiator and PCL intakes).4 BPO Intake (mol% on CL units) 0.2.9 4 5 8 BPO Intake (mol% on CL units) 0.7 3 4 6 FD (%) 4 0.4 0. Thermal properties of the compatibilizers The thermal properties (Table 6.1.8 0.3.8 0.Chapter 6 1 0. et al  working with PCL‐g‐GMA. although both properties are significantly lower than those of pure PCL. Contour plot.6. 3D plot. Graphical representation of the FD model for PCL‐g‐DEM.1.
2 9.9 E (MPa) 270.3 15.5 425. 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) 16.0 430. PCL‐g‐GMA has an FD of 9.9 11. 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.6 305.5 7.2.4 431.6 386. 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.0 341.9 379.2 337.0 9.3 380. Table 6.7 %. 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.8 329.4.1 13.3 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 1.5 357.1 10. 121 .5 489.0 401.3 10. An overview of all prepared blends together with their thermal and mechanical (tensile tests) properties is given in Table 6.8 372.1 11.8 371.4.9 424.7 10. 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.3.6%.8 384.0 332. 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).4 230.6 5.5 ε (%) 640.2 343.9 368. 6.2 342.6 11.6 168.2 321. A total of 12 blends were prepared: a series of binary ones (not containing any compatibilizers) constituting our reference points.
2. This is most probably due to a nucleation effect of the starch on PCL as observed also for other kinds of polymer blends .7.8 % at 30 % starch content.).1 MPa and 341. 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. respectively from 16.5 % for pure PCL to 7. 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 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 6.8 MPa) clearly indicates that the rigidity of the blends increases at higher starch contents.43 MPa and 640. The simultaneous increase in the modulus (from 270. The two trends are remarkably mirroring each other and are both characterized by an increase of ∆Hm and ∆Hc with respect to pure PCL at 10 % starch content followed by a monotonous decrease of both quantities as function of the starch intake.7.3.2 MPa up to 341.Chapter 6 6. 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 thermal and mechanical properties indicate no or very little interaction of the starch particles with the PCL ○ : ∆Hc □ : ∆Hm 122 .
either by heating or in combination with mechanical stress . In this respect the thermal properties remain substantially unchanged with the exception of the Tc.e.) 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. This is visually confirmed by the morphology of the prepared blends (Figure 220.127.116.11. a. However. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends matrix. c. a decrease with respect to pure PCL is observed and then a substantial invariance as function of the PCL‐g‐DEM content.2. Evaporation of the water during blend preparation.2.8. whose synthesis is described for the first time in this work. 26] and explained by assuming that the voids are formed by water in the starch.3. Such behavior is consistent with the hypothesis that the starch particles hinder the melting and crystallization processes of the PCL chains. Figure 6. for which a 6 oC drop is observed when using PCL‐g‐DEM.). Morphology of starch/PCL binary blends. in order to fully understand the role of PCL‐g‐DEM. b. the corresponding binary blend (S/PCL 20/80) contitutes a better reference point compared to pure PCL. 123 . (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. the starch particles are simply inserted into voids of the PCL matrix. was used as compatibilizer for the preparation of ternary blends with starch and PCL. This is confirmed by examination of the blend morphology by SEM (Figure 6. Thus.). As consequence one might expect a slight different structure of these ternary blends as compared to the corresponding binary one. and the lack of interface adhesion cause void formation. The presence of these voids was also observed for sago starch/PCL blends [25. 6. Ternary blends compatibilized with PCL‐g‐DEM PCL‐g‐DEM. The latter trend is also valid for the corresponding enthalpies: i. The thermal behavior is characterized (Table 6.
). b. As a result of the different morphology also the mechanical behavior display differences with the virgin PCL and the binary blend.). At 1 %‐wt of PCL‐g‐DEM in the blends the modulus increases with respect to the binary blend (0 %‐wt compatibilizer in Figure 6. (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 . c. 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. resulting in more rigid material with higher modulus.Chapter 6 a.9. Morphology of ternary blends compatibilized with PCL‐g‐DEM. Figure 6.9. 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. This can be explained by the improved interfacial adhesion  between PCL and starch.). Indeed starch particles in the ternary blends display a smoother interface with the PCL matrix compared to the corresponding binary blends (Figure 6. which will hinder the flowability and fibre forming ability of PCL matrix under cold drawing. 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.10.10.
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. 6. it can be observed that: 1) the melting temperature as well as the crystallization and melting enthalpies do not change significantly (discrepancies in the values are within the boundary defined by the experimental error). in all cases (i. 2) 3) 4) Remarkable is that in the case of PCL‐g‐GMA no optimum is found in the modulus as function of the compatibilizer intake. Modulus of ternary blends S/PCL/PCL‐g‐DEM as function of PCL‐g‐ DEM amount. 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 crystallization temperature experiences a drop of about 5‐6 oC. Ternary blends compatibilized with PCL‐g‐GMA When using PCL‐g‐GMA as compatibilizer two possible comparisons can actually be made: one at fixed starch/PCL ratio and changing the amount of PCL‐ g‐GMA and one at different starch/PCL ratios but with fixed amount of PCL‐g‐ GMA (2 %‐wt). the stress and strain at break are lowered and hardly a function of the PCL‐g‐GMA amount.2.3.3. by taking as reference the corresponding binary blend (S/PCL 20/80). However. at all PCL‐g‐GMA contents) there is a clear increase of the modulus with respect to the 125 .10.e. the modulus is higher and hardly a function of the compatibilizer amount. In particular also in the case of PCL‐g‐GMA.
8 % for PCL‐g‐GMA against 1. partial interfacial adhesion) which render these ternary blends a kind of “intermediate” case between the binary ones and those compatibilized with PCL‐g‐DEM. 28. PCL‐g‐DEM is at least as active as compatibilizer (compare modulus values in Table 6.11.11. (left) the structure of PCL‐g‐GMA is inhomogeneous at the molecular level with long poly(GMA) branches pending from the PCL backbone.11. This is surprising if one takes into account the different FD values: 9.7 % for PCL‐g‐DEM. As a result. 126 . 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. presence of voids. 29].12. a. These differences in compatibilizing effect are not yet fully understood. and morphologies in Figures 6.9. (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. 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.).Chapter 6 binary blend. Figure 6.) than PCL‐g‐GMA for blends containing 20 %‐wt of starch. b. c. This confines all reactive GMA groups in relatively “isolated” spots along the PCL backbone. As given schematically in Figure 6. and 6. Morphology of ternary blends S/PCL/PCL‐g‐GMA. Thus.4. However. 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. 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.
12. the stress and strain at break decrease while the modulus is unchanged at low starch contents and increases significantly for S/PCL blends with 30 %‐wt starch.8 MPa for S/PCL 30/70 up to 430.12.4. thus probably ensuring a better coverage of the surface. Moreover. The proposed explanation implies however that at relatively higher starch contents (> 20 %‐wt) the segregation of the poly(GMA) chains in “isolated” spot along the PCL backbone would become less important. This is in agreement with the hypothesis made above (Figure 6. Schematic representation of the in situ reaction between functionalized PCL and starch. The latter result (from a modulus of 341. In order to check this we compared blends with the same amount of PCL‐g‐GMA (2 %‐wt) but with different starch intakes (10. Concerning the thermal behavior (Table 6. thus attenuating the effect discussed above. at higher starch contents more –OH groups would be available for reaction with the GMA groups.4.). once PCL‐g‐DEM has reacted with one –OH groups on the starch particle.) and relating the “coverage” of the starch particle surface upon reaction with PCL‐g‐GMA with the compatibilization efficiency. the distribution of the reactive group along the PCL backbone is more “homogenous”. Indeed. On the other hand (right of Figure 6. when comparing ternary blends (compatibilized with PCL‐g‐ GMA) with the binary ones. 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.12.3 MPa for S/PCL/PCL‐g‐GMA 30/70/2) clearly indicates that the efficiency of PCL‐g‐GMA as compatibilizer becomes more relevant at relatively higher starch contents (>30 %) . 20 and 30 %‐wt respectively).) for further reaction. for all starch amounts (Table 6. As consequence. 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. other groups will presumably remain available (arrows in Figure 6.). but since DEM is preferentially grafted as monomer. We can therefore conclude that the efficiency of PCL‐g‐GMA in the 127 .12.) PCL‐g‐DEM has much less reactive groups (lower FD) than PCL‐g‐GMA.
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. P‐value ≤10‐9). Nomenclature E FD : initial modulus [MPa] : functionalization degree.4. 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. PCL‐g‐glycidyl methacrylate (PCL‐g‐ GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM).Chapter 6 compatibilization of starch/PCL blends can be significantly improved by changing the blend composition. 6.5. mutual solubility of the monomer into the molten polymer) on the FD adequately described the experimental results (R2=0. The proposed model to quantify the effects of process variables (monomer and initiator intake. This is in contrast with chemical reactivity and amount of chemical groups along the PCL backbone (both factors favorable to GMA as compared to DEM) but it is explainable on the basis of the group distribution along the PCL backbone. The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends.957. The PCL‐g‐GMA and PCL‐g‐DEM compatibilizers display lower melting temperatures and melting enthalpies compared to virgin PCL. From a more practical point of view it can be concluded that the newly synthesized PCL‐g‐DEM. Conclusions A systematic study. moles of GMA or DEM present per mole of CL repeating units : : ni nm amount of initiator intake [mol% on CL units] amount of monomer GMA or DEM intake [mol% on CL units] 128 . 6. firstly reported in this work. 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. has been performed on the synthesis of two reactive compatibilisers. including statistical modeling. The reactive compatibilizers were used in blends of starch with PCL. The most important variable for the final product FD is the interaction between the amount of monomer and initiator used.
B. . Albertsson: Biodegradable Polymers and Environmental Interaction. Ind.Eng. B. S. 33. R. A.W. C. 2002. Ramsay. Psomiadou. 1998. 38. 157. Adv. 1477‐1487. Polym. Karlsson. Polym. Sci. 1942. Pacsu: Starch Studies: Preparation and properties of starch triesters. R.G. 1467‐1475. Eng. W. Mullen.Chem. Karlsson: Polymers from Renewable Resources.J. 1209‐1217. Eur. in Modified Starches: Properties and Uses (Ed. . O. 227‐242. S. Polymer 2003.6. Ogawa. Liu. Jarowenko: Acetylated starch and miscellanous organic esters. Biliaderis. J. 1251‐1253. . Y. A‐C. 129 . Albertsson. Lindblad. Inc. Polym. H. 1997. F. 34. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – III. Mani. 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. Mani. References .S. 44. N. 34. . M. 34. Thermal and morphological properties. 1998. Carbohydr. Sci. ..A. Eur. Rodriguez‐Gonzales. M. 139‐161. M. 1986. Boca Raton. Polym. E. Effect of amylopectin to amylase ratio in starch. . Polym. Ioannis. E. E. J. 1517‐1526. Bhattacharya: Properties of injection moulded starch/ synthetic polymer blends – IV. J. Part 2. Kawasaki: Biodegradable films made from low density polyethylene (LDPE). Ranucci. USA. B. 1998. Davis: High performance LDPE/ thermoplastic starch blend: a sustainable alternative to pure polyethylene. wheat starch and soluble starch for food packaging applications. Wurzburg). CRC Press. A‐C. .D.
K. Sci. Passaglia.K.W. D. M. 3875‐3881. Khandpur. Sci.. 127–134.H. Moro. . Painter: Specific Interactions and the Miscibility of Polymer Blends. . 42. Tang: Functionalization of polyesters with maleic anhydride by reactive extrusion. John Wiley & Sons Inc. Ruggeri. C.S. Polymer 2000. . 179‐187. 1997. 1273‐1335. 35. . Polym. Marechal. Kim. P. 41.F. 2004. Park: Reactive blends of gelatinized starch and polycaprolactone‐g‐glycidyl methacrylate. Martuscelli. Avella. Park: Grafting of glycidyl methacrylate onto polycaprolactone: preparation and characterization. 2001. Bhattacharya. Stab.. Polym. Solaro: Biodegradable polymeric materials. . 5135‐ 5142. 2001. Chiellini. Polym. Laurienzo. J. J. Degrad. 12. Marrucci. L. Chandra. . 29. 1998. Cho. Technomic Publication. 81. Fringant: Properties of thermoplastic blends: starch‐polycaprolactone. 41.Y. Part A: Polym. K. E. P. Rimedio: Preparation and characterization of compatibilised polycaprolactone/ starch composites. Aglietto: Chemical reactions affecting the free‐radical grafting of diethyl maleate onto ethylene polymers.S. Sci. P. C. 1999. A.K. Chem. 127. Environ. Polym. R. Polym. 1991. Park: Modification of Aliphatic Polyesters and Their Reactive Blends with Starch. C. Sci. 1139‐1148. . R. R. M. Appl. Guegan.M. Cho. Kim. 23. Chem. 305‐313. J. 91‐95. Part A: Polym. . 1693‐1702. 1507‐ 1516. Kim. Dole. Kim. John. Jung. Polym. Goldberg: A review of the biodegradability and utility of poly(caprolactone). 5590‐5598..H. 3. . J.E.K. Prog. P. M. 4157‐4167. 1997. 61‐67. Polymer 2000. New York. R. J. USA. D. Mani. Averous. Coleman. J. Polymer 2001. M. . Chim.C. Raimo. Environ.Chapter 6 . Gazz. C. 80. 37. 1996. . M. Tang. J. Inoue: Compatibilizers for melt blending: Premade block copolymers. Degrad. Macromolecules 1996. Pennsylvania. Errico. K. Adv. M. M.M. . R. T. Yang. L. USA.H. C. J. Rustgi: Biodegradable polymers. Bhattacharya: Synthesis and characterization of anhydride‐functional polycaprolactone. Macosko. Nakayama. J. J. J. J. It. Z. 8. Montgomery: Design and Analysis of Experiments 5th edition. E. A. 130 . 1995. C. 2003. Graf.K. Mater. . M.Y. . Polym. C. G. E. J. P. Wu: Physical properties and biodegradability of maleated‐ polycaprolactone/ starch composite.
Appl. S. Eur. Nachtigall. Synthesis and Properties of Reactive Interfacial Agents for Polycaprolactone‐Starch Blends . O.N. 38.Safa. 293‐304.D.S. Azemi. . Ruggeri. D. E. U. Adv. Polym.W. M. J. M. Bull. Ishiaku. 21.S. W. 1298‐1305. . Pang.M. . 2002. R. 393‐401. 2000. 11‐15. .S. K. H. . Azemi: Effect of starch predrying on the mechanical properties of starch/ poly(ε‐caprolactone) composites. 877‐884.M. Ishak: Mechanical properties and enzymic degradation of thermoplastic and granular sago starch filled poly(ε‐ caprolactone). R. Odusanya. Mauler: Comparative study of PP‐MA and PP‐GMA as compatibilizing agent on polypropylene/nylon 6 blends. 9. Aglietto. 87. Ishiaku. Odusanya. U.A.N. Passaglia. G. Barbosa.M. Polym. Polym. 57. Sci. Eng. S. Kammer: On mechanical properties of sago starch/ poly(ε‐caprolactone) composites. Manan. 2003. Ishiaku.A. Technol. B.S. F. Manan. Polym.M. M. Test.Nasirtabrizi: Ring opening reactions of glycidyl methacrylate copolymers to introduce bulky organosilicon side chain substituents. Polym.H. 40. D.M. 131 . K. Tedesco. 273‐281. 2006. A. Sci. Lee.S. 1998. U. Polym. 2002. B. Picchioni: Formation and compatibilizing effect of the grafted copolymer in the reactive blending of 2‐ diethylsuccinate containing polyolefins with poly‐ε‐caprolactam (nylon‐6). J.S.B.W.V. O.
4)‐glucosidic bonds. After use. containing periodic branches linked with the backbones through α‐D‐(1.Summary Plastics made from fossil resources (a. it has a high volume to weight ratio and is generally resistant to degradation. Starch is relatively cheap and available from a broad range of plants. In combination with the current high prices for petrochemical products. The application of a number of prospective methods for developing starch‐based biomaterials will be described in this thesis. Its production has increased significantly since 1950. while amylopectin is branched polymer. Examples are the use as packaging and construction materials. There are several starch modification methods available. there is a strong need for renewable alternatives for plastics from fossil resources. such as thermoplasticization. with a rate of almost 10% every year. The use of virgin starch for packaging materials is restricted because it cannot be shaped to films with adequate mechanical properties (high percentage elongation.6)‐glucosidic bonds. Starch is also too sensitive to water. For instance. To gain insight in the potential of this approach. The content of amylose and amylopectine in starch varies and largely depends on the starch source. blending with other materials. tensile and flexural strength). 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 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. we have . Amylose is essentially a linear polymer in which AHG units are predominantly connected through α‐D‐(1. There are two types of polymers present in starch: amylose and amylopectin. Starch must therefore be modified before it can be applied as a biodegradable plastic. chemical modification or combinations thereof. Starch‐based biodegradable materials are considered interesting candidates to replace certain types of conventional plastics. Starch is a polymer consisting of anhydroglucose (AHG) units. 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. plastics may end up in the environment and cause serious environmental problems.o oil and gas) are very attractive materials for a broad range of applications.
2. Esterification of starch with carboxylic acid derivatives is one of the oldest strategies to improve starch properties. The results for the model system involving the ROP of p‐dioxanone initiated by hydroxyl groups of a protected monosaccharide (1. but the products are still very brittle. and particularly acetic acid derivatives (C2) have received considerable attention. Characterization of the products using 1H‐ and 13C‐NMR as well as MALDI‐TOF mass spectrometry showed the presence of significant amounts of p‐dioxanone polymers with an isopropoxide end group (20‐30%) The knowledge obtained from the model system was applied for the synthesis of starch‐g‐poly‐ε‐caprolactone using hydrophobised silylated starch.Summary initially performed research on the ROP with a simpler model system involving a simple protected mono‐saccharide instead of starch. and off‐white solid products with isolated yields of 30‐96% were obtained. depending on the ε‐CL to starch ratio. 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.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose ratio (at constant Al(OiPr)3 intake).2.3. The silyl groups of the poly‐(ε)‐caprolactone grafted starch co‐polymers were finally successfully removed using a mild dilute hydrochloric acid treatment in THF at room temperature. even in the presence of plasticizers.72. Considerable amounts of ε‐CL homopolymers with isopropyl end‐groups were also formed. The yield of the polymers was a function of the reaction temperature and the reaction time.21‐0. The silylation reaction was performed using hexamethyl disilazane (HMDS) in DMSO/ toluene mixtures at 70°C. are described in Chapter 2. and subsequent removal of the silyl groups.46‐0. The DS of the PCL chains was between 0. The use of higher carboxylic (fatty) acid to esterify starch resulted in products with significantly improved 134 . Most of the previous starch esterification studies involved the use of short chain carboxylic acids (C1‐C4). The synthetic procedure may be divided into three steps: hydrophobisation of starch by the introduction of SiMe3 groups followed by in situ ROP. The polymerizations were performed at 60‐100°C. with higher ratios leading to higher molecular weights.68) was obtained. 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. Poly‐(ε)‐ caprolactone grafted silylated starch co‐polymers with average chain length of 40‐ 55 monomer units (molecular weight of 4500‐6300) were obtained. The results are described in detail in Chapter 3.3. Average molecular weights of the products were between 970 and 6200 (7 – 58 monomer units) and were a clear function of the p‐ dioxanone/ 1. The hydrophobicity of starch acetates is higher than virgin starch. The grafting efficiency for the desired reaction was 28‐58%. with higher temperatures (100°C) leading to lower yields. Silylated starch with a low to medium DS (0.
elongation at break of 3‐26%. The starch esters were synthesized by reacting gelatinized starch with vinyl laurate or vinyl acetate in DMSO in the presence of basic catalysts (Na2HPO4. In Chapter 4‐5. products with a DS > 2.3‐1.5 for starch stearate). while the use of K2CO3 and CH3COONa catalysts resulted in medium‐high DS products (2. 13C‐NMR and FT‐IR.96). In Chapter 5. At higher ratios. 0. and catalyst basicity) to the DS of the starch ester products were developed.26‐2.07‐1.7‐3. usually performed using fatty acid chlorides which are relatively expensive and rather corrosive. The thermal and mechanical properties of some representative product samples are also described.96 and P‐value of ≤10‐7) second‐order mathematical models correlating the effect of process variables (vinyl to AHG‐ starch mol ratio. The yellowish products were characterized by 1H‐. With these catalysts. K2CO3. 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. catalyst intake. For low vinyl‐ester to starch ratios. reaction temperature. a systematic study. and modulus of elasticity of 46‐113 MPa. Statistically adequate (R2≥0. Mechanical tests show that the products have tensile strength (stress at break) between 2. the DS decreased. The DS of the products is a strong function of the basicity of the catalyst. The thermal stability of the esterified products is higher than that of native starch. 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. Summary mechanical properties and hydrophobicities. vinyl ester to starch ratio. an increase in the vinyl‐ester concentration led to higher product DS values. and the type of catalyst. K2CO3. In Chapter 6.5 for starch laurate.4 could be obtained for both laurate and stearate esters.45‐1. an alternative method for higher fatty starch esters by using fatty vinyl esters is reported. High‐DS products (DS= 2. PCL‐g‐ 135 . The last part of this thesis deals with reactive blending of starch and biodegradable polymers using reactive interfacial agents. The use of methyl and glyceryl esters results in products with only relatively low DS. The DS of the products was a clear function of the chain length of the fatty ester. The use of Na2HPO4 resulted in low‐medium DS products (0.1‐2. K2CO3 and Na‐acetate are superior with respect to activity when compared with Na2HPO4.5 MPa. The starch esters were prepared by reacting starch with vinyl laurate or vinyl stearate in the presence of basic catalysts (Na2HPO4. and Na‐acetate) in DMSO at 110°C.39) are totally amorphous whereas the low‐DS ones (DS= 1. The synthesis of these fatty starch esters is. the synthesis of two polycaprolactone (PCL)‐based reactive interfacial agents. however.24 ‐ 2. Reaction temperature and catalyst intake also affect the product DS but only to a lesser extent.75) are still partially crystalline.4‐3 for starch stearate). presumably due to a reduction of the polarity of the reaction medium. 1.9 for starch laurate. and Na‐ acetate). The products have a broad range of degree of substitution (DS = 0.
mutual solubility of the monomer into the molten polymer) on the functionalisation degree (FD) of the GMA and DEM units to the PCL backbone.Summary glycidyl methacrylate (PCL‐GMA) and PCL‐g‐diethyl maleate (PCL‐g‐DEM).957. the mechanical behavior of the compatibilised blends is in general different from that of pure PCL and of the corresponding uncompatibilised blends. P‐value ≤10‐9) has been developed to quantify the effects of process variables (monomer and initiator intake. 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. Highest values of the FD (up to 45 %) were observed for PCL‐g‐GMA. 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. PCL‐g‐DEM seems to have slightly more efficient compatibilising effect than PCL‐g‐GMA as shown by blends morphology and elastic modulus values. A statistically adequate model (R2=0. 136 . The latter hypothesis is indirectly confirmed by the observation that PCL‐g‐GMA becomes more efficient at relatively higher starch contents in the blends. Both of the PCL‐based compatibilisers improve the interfacial adhesion between the starch particles and the PCL matrix in starch/ PCL blends. and their use in compatibilising starch‐PCL blends is described. As the result. at relatively high GMA and BPO intakes. 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). The FD values for PCL‐g‐DEM were considerably lower (up to 7 %). At a fixed starch content (20 %‐wt).
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. Biologisch afbreekbare polymeren op basis van zetmeel kunnen mogelijk bepaalde conventionele kunststoffen vervangen. mede omdat het niet mogelijk is om films met goede mechanische eigenschappen te maken. Het amylopectine is een vertakt polymeer waarbij de vertakkingen via α‐D‐(1. Het zetmeel is daarnaast ook erg gevoelig voor water.Samenvatting Kunststoffen gemaakt van fossiele grondstoffen als olie en gas zijn zeer aantrekkelijke materialen voor een breed scala aan toepassingen. De verhouding amylose en amylopectine in zetmeel varieert en hangt grotendeels af van de plant waaruit het zetmeel gewonnen wordt. Na gebruik kunnen de materialen in het milieu eindigen en grote problemen veroorzaken.4)‐glucoside bindingen verbonden zijn. Voorbeelden zijn thermo‐ plastificatie. Dit proefschrift beschrijft een aantal methodes om nieuwe polymere materialen op basis van zetmeel te maken.v. menging met andere materialen. Zetmeel is een polymeer bestaande uit anhydroglucose (AHG) eenheden. Amylose is een lineair polymeer waarin de AHG eenheden hoofdzakelijk via α‐D‐(1. De toepassing van een alternatieve methode waarbij zetmeel eerst minder hydrofiel gemaakt wordt door de introductie van grote hydrofobe SiMe3 groepen. Zetmeel is relatief goedkoop en te winnen uit een breed scala aan planten. ε‐caprolactone op de hydroxy‐groepen van zetmeel is helaas niet eenvoudig. De AHG eenheden kunnen op een aantal manieren aan elkaar gekoppeld worden.6)‐glucoside bindingen met de hoofdketen zijn verbonden. in combinatie met de huidige hoge prijzen voor de grondstoffen. Twee type polymeren zijn dominant aanwezig in zetmeel: amylose en amylopectine. De synthese van dergelijke producten met een in situ katalytische ringopening polymerisatie (ROP) van geschikte monomeren als b. Het gebruik van zetmeel voor verpakking materialen is momenteel beperkt. De productie van kunststoffen is sinds 1950 met bijna 10% per jaar gestegen. Het chemisch grafting van biologisch afbreekbare polyesters op de zetmeel ketens zou kunnen leiden tot minder hydrofiele en dus minder water gevoelige materialen met betere mechanische eigenschappen. gevolgd door de ROP met een polyester precursor en verwijdering van de SiMe3 groepen . Dit. heeft geleid tot een grote interesse voor biologisch afbreekbare alternatieven gemaakt uit hernieuwbare grondstoffen. Er zijn verschillende methodes beschikbaar om zetmeel te modificeren. Modificatie van zetmeel kan het aantal toepassingen sterk vergroten. chemische modificaties en combinaties van deze methoden.
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 resultaten voor het modelsysteem waar de ROP van p‐dioxanone wordt geïnitieerd door de hydroxy‐groepen van een beschermde mono‐saccharide (1. De DS van de PCL ketens lag tussen de 0. De opbrengst van de polymeren was een functie van de reactietemperatuur en de reactietijd.2. 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. 138 .72 en was afhankelijk van de ε‐caprolacton/zetmeel verhouding. Er werden ook aanzienlijke hoeveelheden homopolymeer van ε‐ caprolacton met isopropyl eindgroepen gevormd.68).3. gevolgd door in situ ROP met ε‐ caprolacton en daarna de verwijdering van de silyl groepen. DS = 0. De resultaten worden in Hoofdstuk 3 in detail beschreven. De silylerings reactie werd bij 70°C uitgevoerd met hexamethyl disilazane (HMDS) in DMSO/tolueen mengsels.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose (bij constante i Al(O Pr)3 concentratie). 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. temperatuur en monomeer: monosaccharide verhouding) op de gemiddelde polymerisatie graad zijn gekwantificeerd met een statistisch model. De grafting efficiency van de gewenste reactie was 28‐58%.46‐0.2. Karakterisering van de producten met 1 H‐ en 13C‐NMR en MALDI‐TOF massaspectrometrie toonden de aanwezigheid aan van de gewenste producten en daarnaast significante hoeveelheden van p‐ dioxanone polymeren met een isopropoxide eindgroep (20‐30%). waarin hogere temperaturen (100°C) tot lagere opbrengsten leiden. De grafting van ε‐ caprolactone aan het gesilyleerde zetmeel werd in THF bij 50°C uitgevoerd met Al(OiPr)3 als katalysator. De gevolgde procedure bestaat uit drie stappen: hydrofobisering van het zetmeel door de introductie van SiMe3 groepen. De effecten van procesvariabelen (tijd.21‐0. De polycaprolacton eenheden (PCL) van het verkregen poly‐ε‐caprolactone gesilyleerde zetmeel hadden een gemiddelde ketenlengte van 40‐55 monomeereenheden (molecuul massa 4500‐6300).Samenvatting zou in theorie moeten leiden tot een hogere grafting efficiency.3. Een hogere verhouding leidde tot hogere molecuul massa’s. De polymerisaties werden tussen 60‐100°C uitgevoerd en leverde vaste producten in geïsoleerde opbrengsten tussen de 30 en 96%. De substitutiegraad van de SiMe3 groepen varieerde van laag tot middelhoog (degree of substitution.4‐di‐O‐isopropylidene‐α‐D‐galactopyranose) met Al(OiPr)3 als katalysator worden in Hoofdstuk 2 beschreven. De kennis verkregen met het modelsysteem is toegepast om ε‐caprolactone te graften op zetmeel.
reactietemperatuur. K2CO3 en Na‐acetaat) in DMSO bij 110°C te laten reageren. zelfs in aanwezigheid van plastificeermiddelen. Echter. K2CO3 en Na‐acetaat vertonen superieure activiteiten vergeleken met Na2HPO4. katalysator concentratie en katalysator basiciteit) op de DS zijn gekwantificeerd met een statistisch model (R2 ≥ 0. Het gebruik van hogere (vet‐) zuren resulteerde in producten met beduidend betere mechanische eigenschappen en hydrofobiciteit. Met deze katalysatoren worden voor zowel de laureaat als stearaat esters producten met een DS van boven de 2. wordt een alternatieve methode voor de synthese van zetmeel esters met lange vetzuren beschreven. De producten werden met 1H‐. en in het bijzonder azijnzuur derivaten (C2). In Hoofdstuk 4‐5. De acetaat esters zijn in het algemeen minder hydrofiel dan zetmeel. K2CO3. de vinyl‐ ester/zetmeel verhouding en het type katalysator. Samenvatting De esterificatie van zetmeel met carbonzuur derivaten is één van de oudste strategieën voor het verbeteren van zetmeeleigenschappen. Bij lage vinyl‐ester/zetmeel verhoudingen leidde een verhoging van de vinyl‐ester concentratie in eerste instantie tot hogere DS waarden. De DS van de producten was een duidelijke functie van de keten lengte van de vetzure vinylester. 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. terwijl het gebruik van de K2CO3 en Na‐acetaat katalysatoren 139 . In Hoofdstuk 5. wordt een systematische studie. vermoedelijk als gevolg van een verlaging van de polariteit van het reactiemengsel. In de meeste zetmeel esterificatie studies wordt gebruikt gemaakt van kleine carbonzuren (C1‐C4). 0.3‐1. Echter de producten zijn nog steeds zeer bros. De substitutiegraad varieert van 0. 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 (Na2HPO4.96. bij verdere verhoging van de verhouding verminderde de DS.5 voor zetmeel stearaat).96 en P‐ waarde van ≤ 10‐7).5 voor zetmeel laureaat. De effecten van belangrijke procesvariabelen (vinyl ester/zetmeel molverhouding. De synthese van deze esters wordt in het algemeen uitgevoerd met vetzuurchloriden die vrij duur en sterk corrosief zijn. Het gebruik van methyl en glyceryl esters resulteert in producten met een lage substitutiegraad. De zetmeel esters werden bereid door het zetmeel met vinyl laureaat of vinyl stearaat in aanwezigheid van basische katalysatoren (Na2HPO4.07‐1. De DS van de producten is een sterke functie van de basiciteit van de katalysator. 13C‐NMR en FT‐IR gekarakteriseerd.4 verkregen.24 tot 2. inclusief statistische modellering. De thermische en mechanische eigenschappen van sommige representatieve producten zijn uitgebreid bepaald. Het gebruik van Na2HPO4 resulteerde in producten met een laag tot middelhoge DS (0. naar de synthese van lange vetzuur esters uit maïszetmeel (zetmeel laureaat en zetmeel stearaat. en Na‐acetaat) te laten reageren.
Er is een statistisch model (R2 =0. 140 . Dit leidt tot verschillen in mechanische eigenschappen tussen gecompatibiliseerde mengsels en van de niet‐ gecompatibiliseerde mengsels van zetmeel en PCL. De thermische stabiliteit van de geësterificeerde producten is hoger dan die van zetmeel. De producten met een hoge DS (2.4‐3.Samenvatting resulteerde in producten met een middelhoog tot hoge DS waarden (2.9 voor zetmeel laureaat.26‐2. Het laatste hoofdstuk van dit proefschrift (Hoofdstuk 6) behandelt de menging van zetmeel en biologisch afbreekbare polymeren met reactieve compatibilisers. een uitrekking tot breuk van 3‐26%.1‐2.957 en P‐ waarde van ≤ 10‐9) ontwikkeld waarmee de effecten van procesvariabelen (monomeer en initiator concentratie. Mechanische tests tonen aan dat de producten een treksterkte hebben variërend van 2. gebruikt. en een elasticiteit modulus van 46‐113 MPa. De synthese van deze verbindingen wordt uitgebreid beschreven. De FD waarden voor PCL‐g‐DEM waren aanzienlijk lager (tot 7%). 1.45‐1. De hoogste FD waarden (tot 45%) werden gevonden voor PCL‐g‐GMA bij relatief hoge GMA en BPO concentraties. 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.75 nog gedeeltelijk kristallijn zijn.39) zijn volledig amorf terwijl de producten met een DS van 1. Zo is bijvoorbeeld de elastische modulus van de gecompatibiliseerde mengsels beduidend hoger (tot 430 MPa) dan die van niet‐gecompatibiliseerde mengsels (320‐340 MPa).7‐3. Als reactieve compatibilisers zijn twee polycaprolactone derivaten.0 voor zetmeel stearaat). 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. Beide reactieve compatibilisers verbeteren de interactie tussen de zetmeeldeeltjes en de PCL matrix.5 MPa. PCL‐g‐glycidyl methacrylaat (PCL‐GMA) en PCL‐g‐diethyl maleaat (PCL‐g‐DEM).
Acknowledgements The synthesis of biodegradable materials is currently receiving a lot of attention. for his efforts to make every foreign students feel at home in Groningen. together with difficulties in managing plastic waste. A. Leon Janssen. Anno Wagenaar and Wim Kruizinga taught me how to operate the NMR machines and to interpret the spectra. Prof. Broekhuis. for her hospitality and solving administrative matters. I also want to thank Ineke Ganzeveld. I also want to give my deep appreciation to my reading committee. for upgrading my writing quality. Francesco Picchioni. I want to thank Marya van der Duin‐de Jonge. A. Erwin Wilbers. Prof. the raw material for fossil derived plastics. and Prof. L. especially on his opinions about the statistical modelling and mechanical properties.J. People from other groups have also been very kind and supportive to my research. Anne Appeldoorn. Marcel de Vries. I want to thank Leon Janssen for his hospitality. and Prof. This thesis is a summary of my research on developing novel biodegradable polymers from starch. which improved the quality of this thesis. I want to thank Francesco Picchioni for being always there when I needed him to discuss anything. for the inspiring ideas during the discussions. These include the current high oil prices. since the mechanical properties as well as hydrophobicity of this starting material is relatively poor. Mosiscki. Thank you very much for spending your valuable time reading my thesis and for the constructive comments and corrections you have made. and for helping me arranging many things when I was not able to do it from Indonesia. Anno was also always there when .A. Erik Heeres. as well as taking the SEM pictures. First of all. not in the last place because of the problems related with conventional plastics. Minnaard. Prof. for your consistent support. He has also been a good friend and always supported me during the hard times. I want to thank my three promoters. Prof. I want to thank Erik Heeres for being more than a good supervisor. for valuable discussions during my early PhD period and for handing me over to the right hands. I found out that developing a better fully‐biodegradable‐material from starch is not an easy job. Gert Alberda van Ekenstein and Harry Nijland helped me a lot with mechanical and thermal property measurements. Thank you Erik. 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. Although it is indeed a very interesting topic. and Laurens Bosgra for being always helpful with technical issues and Jan Henk Marsman for the help with analytical equipment. who was my first daily supervisor for about two months. and for always encouraging me to finish this thesis.
Tante Caroline. Ronny Sutarto. pak Harry. and although we seldom met afterwards. and on many occassions. Nadia Gozali. spring and autumn in Groningen. Vincent Nieborg and Fesia Lestari. Thank you God. Yongki. Patrick and Oppie. and the ISD staff helped me a lot with visum and residence permit arrangements as well as tax papers. Thank you AW. Tante Oppie. M. for never leaving me alone. Rasrendra. mbak Ika. Buana Girisuta and Rina Karina. Niels van Vegten. Egi. Nidal Hammoud Hassan. Yao Jie. and helpful.Acknowledgements I had questions or needed specific equipments. Jaap Bosma. I have met a lot of Indonesian students in Groningen. Thank you my parents and sister. Mita. kak Atha. Chalid. Ignacio Melian Cabrera. for your love. Anneke Toxopeus. Oscar Rojas. and Mahesa for being my brothers and sisters in Groningen. Wita Sondari. bu Ida Susanti. Endy. Jasper Huijsman. Tenny. Erna Subroto. Thank you Tresna. You all have always been my inspiration and motivation. Taufik. for your never‐ending blessings. August Kurniawan. Marcel Wiegman. and assistant professors have shared the days with me in Groningen. and Tante Alma for being like my parents and family in Groningen. Thanks also for Marcel and Connie. A lot of thanks for Henk Stegeman and Poppy Sutanto. and many others for the good old times! Thank you Tante Ietje. and Iwan Kustiawan for becoming my family. and Henk van de Bovenkamp for cheering my days and being my friends during summer and winter. master students who worked with me. Wahyudin. visiting researchers. in the lab. The chemical and glassware magazijn as well as the FWN library personnel were always patient. and passion. Wangsa.B. Hans Heeres. in the office. Francesca Gambardella. kak Roga. Samuel. My special thanks are for Laura Junistia and Jan‐Pieter Drijfhout. Tante Smith. Wuri Raspati. Arjan Kloekhorst. Kuslan. Tiara. Teguh. Diana Santangelo. Asal Hamarneh. Thyara. and for my wife and family. 142 . Henky Muljana (my paranymph) and Anindhita Widyadhana. Puspita Kencana. Abdul Osman. Louis Daniel. Bima and Wisnu (also my paranymph). C. Taufik. for everything. Ardiyanti. Farchad H. Ita. friendly. Annette Korringa. Anna Nizniowska. Dado. support. mbak Mia. mas Pandu. Judy Retti Witono. Francesca Fallani. Agnes R. Many master and PhD students. Mahfud. Hofman. our memories will never fade away. Gerard Kraai. patience. Diana Jirjis. Our collaboration has also become part of this thesis. Groningen is also a place where I found true friendship. Jelle Wildschut. Albert Kieviet helped me with the MALDI‐TOF measurements. Boelo Schuur. Teddy. Sari Fabianti. Iwan Harsono. I want to thank Edwin A. Togi. Zhang Youchun. Ais Jenie. R.
Experimental and modeling studies on the synthesis and properties of higher fatty esters of corn starch. South China University of Technology. 2008 (submitted to Carbohydr.10.P. F. 17‐20 December 2006 (oral presentation). H. .).2008. Sugih. Manurung. Picchioni. Sci. Heeres.010). A. J. L. Sugih.J. Sugih. Picchioni. Sugih. 5. Janssen. J.Heeres. L. 26‐28 November 2006 (oral presentation). L.List of Publications 1. Drijfhout. F.Drijfhout. A.M. J. H. and F. J.K. Eur. A. DOI: 10.J. R.P. H. Picchioni. Junistia.Picchioni. A. Bratislava (Slovak Republic). Drijfhout. Picchioni: Functionalization and blending of polycaprolactone.P.K. L.Janssen. Sugih.M. Mumbai (India). Picchioni. Appl. 2008.J. Heeres.J. and H.K. 2008 (accepted for publication in Starch‐Starke). Picchioni. presented in International Workshop: “From Polymer Modification to Multicomponent Polymer Systems”. H. 2008 (accepted for publication in Starch‐ Starke).P. 3. H.M. and F. and H. Picchioni: Biodegradable polymer materials based on polyesters and starch/proteins. Janssen. 8. presented in the 1st International Symposium on Sustainable Chemical Product and Process Engineering. A.K.K.eurpolymj. Picchioni. Junistia. India. Heeres. A. L. Broekhuis.J.J. Junistia.) 6. 4. A.K. F. Sugih. Heeres.J. Junistia. Sugih.J. F. L. F. Kuala Lumpur (Malaysia). A. Heeres: Synthesis and properties of novel green biopolymers derived from starch and fatty acids. Broekhuis. Synthesis of poly‐ (ε)‐caprolactone grafted starch co‐polymers by ring‐opening polymerisation using silylated starch precursors.P.K. Sugih.B.B. starch‐based innovative product: The synthesis of grafted polycaprolactone‐starch by ring opening polymerisation. 25‐28 September 2007 (oral presentation). F. Sugih. 2008 (submitted to J. 9. 27‐30 August 2006 (oral presentation) 7. presented in 11th Asia Pacific Confederation of Chemical Engineering (APCChE) Congress.B. 2. Guangzhou (China).M Janssen and H. Heeres: Green. Polym. A. 9th National Conference organized by the Society for Polymer Science.J. F.1016/j. presented in “Polymers for Advanced Technologies”. Polym. Polym. H. (in press. Synthesis and properties of reactive interfacial agents for polycaprolactone‐starch blends. L.A. Experimental Studies on the ring opening polymerization of p‐dioxanone using an Al(OiPr)3‐monosaccharide initiator system. A. Heeres: Synthesis of higher fatty acid starch esters using vinyl laurate and stearate as reactants. Manurung.K.K. A. Pune Chapter. R.P. L.B.A.P. Heeres.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue listening from where you left off, or restart the preview.