Frank Bien,
Berthold Wiege,
Siegfried Warwel
H. P. Kaufmann-lnstitute, Federal Centre for Cereal, Potato and Lipid ResearchMunster, Germany
Hydrophobic Modification of Starch by Alkali-Catalyzed Addition of 1,2-Epoxyalkanes
An efficient method of preparing hydrophobic a-hydroxy starch ethers using aqueous alkaline conditions is described. α-Hydroxy starch ethers were synthesized by addition of 1,2-epoxyalkanes to an aqueous alkaline starch gel in the presence of sodium sul-fate as a co-catalyst. The reaction, carried out in a stirred autoclave at 140 °C and 3.9 bar, was optimized with respect to the concentrations of sodium hydroxide and 1,2-epoxyalkane. Optimum yields and molar substitutions (MS) were obtained at molar ratios of sodium hydroxide to anhydroglucose unit (AGU) of 0.5 to 1.0. The amount of molar substitution could be controlled by 1,2-epoxyalkane concentration. Thus, a series of a-hydroxyoctyl starch ethers with MS from 0.7 to 2.4 were synthesized in yields up to 90% by using these conditions. Starches with different amylose contents were also converted to the corresponding ethers using a threefold excess of 1,2-epoxyoctane and an equimolar ratio sodium hydroxide : AGU. The reaction is hardly effected by the origin of the starch and its amylose content. The influence of the 1,2-epoxyalkane chain length was investigated by performing the conversion with a series of terminal epoxyalkanes from 1,2-epoxyhexane to 1,2-epoxydodecane. The results indicated that the hydrophobic character of the starch ethers increased by increasing the molar substitution and alkyl chain length. All products were insoluble in water, but soluble in mixtures of methanol and methylene chloride. Furthermore the starch ethers can be converted into shaped articles by extrusion technology without the addition of plasticizers.
Keywords: Hydrophobic modification; 1,2-Epoxyalkanes; Etherification
1 Introduction
Starch is one of the most important renewable resources because of its abundant supply from potatoes and cereals. About 40% of the German production of starch was used for non-food applications, mainly in paper (308,0001) and corrugated cardboard (84,000 t) processing in 1996 [1], but only 5,000 t of starch were utilized for the production of biodegradable plastics [2]. A growing economical interest in bioplastics has been observed triggered by the increasing expenses for waste disposal. Therefore, the increasing demand for biodegradable packing materials and films has caused a considerable increase in applications for starch and its derivatives in plastics manufacturing.
Natural starch can be incorporated as a filler into petrochemical-based polymers in order to improve their biodegradability, but these blends have only poor mechanical properties [3]. Furthermore, starch can be processed like a thermoplastic material by addition of additives after dissolving the grain structure [4]. For instance, bioplastics based on natural starch were distrib-
Correspondence: Siegfried Warwel, Institute for Biochemistry and Technology of Lipids, H. P. Kaufmann-lnstitute, Federal Centre for Cereal, Potato and Lipid Research. Piusallee 68. D-48147 Munster, Germany. Phone: +49-251-4816720, Fax: +49-251-519275, E-mail: ibfett @ uni-muenster.de.
uted by the Biotec company (Emmerich, Germany) [5, 6]. However, the products partly maintain the hydrophilic character of native starch.
Alternatively, the properties of starch can be modified by introducing hydrophobic groups such as alkyl ester or ether moities or by direct grafting of polymer chains onto the starch backbone [7]. Especially, alkyl starch ethers and esters show improved processing properties because of their increased hydrophobic character. The es-terification of starch, which has been performed with different acylating agents and methods, has been intensively studied in the last years [8-12]. In contrast investigations into the etherification of starch were mainly confined to the synthesis of hydroxyethyl-, hydroxypropyl- and cationic starches [13-15], which have been preferably used in the food sector, the pulp and paper processing and textile processing industry.
A synthesis of α-hydroxyalkyl starch ethers was described by reacting starch with a series of terminal epoxyalkanes from 1,2-epoxyhexane to 1,2-epoxydodecane in dimethyl sulfoxide [16]. Using sodium hydride as a base for the activation, starch ethers were obtained with a molar substitution (MS) of up to 2.3, dependent on the reaction time. However, this approach seems to be unsuitable for the preparation of α-hydroxyalkyl starch ethers on an industrial scale.
Fig. 1. Alkali-catalyzed addition of 1,2-epoxyalkanes tostarch.
In view of a large scale application, our objective was the conversion of starch with 1,2-epoxyalkanes under alkaline catalysis in order to achieve a hydrophobically modified polymer.
The hydrophobic character of α-hydroxy starch ethers should depend both on the chain length of the 1,2-epoxyalkane and on the degree of molar substitution. Furthermore, the thermoplastic properties of starch derivatives should be influenced by the amylose and amy-lopectin ratio of the starch used in the synthesis.
In the following, a study is presented investigating the influence of the concentration of sodium hydroxide, the concentration and the chain length of the 1,2-epoxyalka-ne and the amylose content on the yield and the molar substitution of the α-hydroxyalkyl starch ethers (Fig. 1).
2 Materials and Methods
2.1 Materials
Potato starch was supplied by Emsland-Starke. Amylose from potatoes (Fluka), Hylon 7 (National Starch), maize starch (Cerestar), 1,2-epoxyhexane (Aldrich), 1,2-epoxy-octane, 1,2-epoxydecane and 1.2-epoxydodecane (Lancaster), sodium hydroxide (Merck), sodium sulfate (Riedel de Haen), chloroform (analytical grade, Merck), and Hydranal composite 5 (Riedel de Haen) were used as received. Methanol (Merck) was dried by destination over magnesium and stored over molecular sieves 4 Å.
2.2 Methods
2.2.1 Analytical methods
The water content of the starch ethers was determined by Karl-Fischer titration using a Metrohm Titrino 718. Before being titrated with Hydranal Composite 5 (Riedel de Haen), the starch ether had been suspended for 15 min in dry methanol/chloroform 3 : 1 or methanol/chloroform 1:1 at 45 °C. Elemental analysis was carried out on a Heraeus CHN-O-Rapid analyzer. 1H NMR/13C NMR-spectra were recorded at room temperature on a 300 MHz Bruker AMX spectrometer operating at 300 MHz/75 MHz. The chemical shifts were expressed in
ppm, CDCI3, dimethylsulfoxide-d6 or pyridine-d5 were used as reference, depending on the solubility of the compound. IR spectra were recorded with a Bruker IFS 28 FT-IR spectrometer (KBr).
2.2.2 Synthesis of α-hydroxyalkyl starch ethers
Starch (5.79 g, 30 mmol; water content 16% w/w) and a solution of 2.13g (15 mmol) sodium sulfate was introduced into a stirred autoclave and the respective amount of 3 M aqueous NaOH solution (0.1-2.0 eq. sodium hy-droxide/anhydroglucose unit (AGU)) and 1,2-epoxyalka-ne (1.2 eq.-5.0 eq. 1,2-epoxyalkane/AGU) were added. The water content was limited to 40 mL in all conversions. The vigorously stirred reaction mixture was heated at 140 °C and 3.9 bar for 4-6 h. The precipitated starch ether was then suspended in water (300 mL) at room temperature and neutralized with 1 M aqueous hydrochloric acid. The solid was filtered through a Buchner funnel and washed several times with water until the conductivity was reduced to 10μS/cm. The crude starch ether was extracted with diethyl ether or isohexane in order to remove low molecular weight impurities and dried in a vaccum oven at 50 °C overnight. The yield of the corresponding a-hydroxyalkyl starch ether was calculated from the amount of product obtained, taking account of the molar substitution.
2.2.3 Determination of the molar substitution
The molar substitution was determined by 1H NMR [17-19] and elemental analysis [20]. For determination of the MS value by 1H NMR, the sample was dissolved in CDCI3, pyridine-d5 or dimethyl sulfoxide-d6. The peaks between 0.75 and 1.60 ppm resulted from CH3- and CH2-groups of the alkyl side chains of the starch ethers, the signals between 3.00 and 6.00 ppm corresponded to the protons of the glycosidic structure and the O-CH2-CHOH-groups of the side chains. The MS value was calculated from the ratio of the integrals of the alkyl protons and the protons between 3.00 and 6.00 ppm using the following equation (1):
MS = 10K/(2n-3-3K) (1)
n = number of carbon atoms of the 1,2-epoxyalkane
K = Ialk/ Iremain
Ialk= integral of the alkyl protons
Iremain = integral of the remaining protons
For the determination of the molar substitution by elemental analysis the water content of the sample had to be determined by Karl-Fischer titration [21]. The molar substitution was calculated by comparing the percentages of C and H obtained by elemental analysis with the theoreti-
cal values calculated for molar substitutions ranging from 0 to 3.0. Taking into account the water content of the sample, the MS was taken as the value which produced the closest match between measured and theoretical value. Due to the good agreement of MS values (± 0.2) obtained by the two independent methods, the average of both results was used.
3 Results and Discussion
The synthetic yield of a-hydroxyalkyl starch ethers is mainly influenced by the concentration of sodium hydroxide, the concentration and chain length of the 1,2-epoxyalkane as well as by the reaction time and temperature. Former investigations concerning the alkali-catalyzed addition of 1,2-epoxyoctane have shown that a reaction temperature of 140 °C is necessary to achieve a sufficient conversion within 4 h, while 1,2-epoxyhexane already reacted with starch at 80 °C. However, the reaction time had to be extended to five days in order to achieve a sufficient yield.
Therefore, the conversion of 1,2-epoxyoctane with potato starch carried out in a stirred autoclave at 140 °C and 3.9 bar was used as model reaction for the optimization of the reaction parameters.
3.1 Variation of the sodium hydroxide concentration
concentration of sodium hydroxide was varied at a constant molar ratio of 1,2-epoxyoctane : sodium sul-fate : AGU = 3.0 : 0.5 : 1.0 (Tab. 1).
Using 0.1 eq. sodium hydroxide per AGU, an a-hydroxy-octyl starch ether with MS = 0.2 was isolated in a low yield (18%). Under these conditions, the amount of sodium hydroxide was not high enough to achieve a sufficient activation of the starch. Maximum yields (89-97%) and molar substitutions (MS: ~1.8) were obtained at molar ratios of sodium hydroxide: AGU from 0.5 up to 1.0. The increase
Tab. 1. Etherification of potato starch with 1,2-epoxyoctance: variation of the sodium hydroxide concentration.a)
a) Molar ratio of the reactants: 1,2-epoxyoctane/AGU/Na2SO4 = 3.0 : 1.0 : 0.5; solvent: water (total volume: 40 mL); reaction conditions: 140 °C, 3.9 bar, 4 h; potato starch containing 16% (w/w) water.
b) Determined by 1H NMR and elemental analysis.
of the sodium hydroxide concentration to 2.0 eq. per AGU gave a comparable yield, but led to a decrease of the MS of the product to 1.3. Obviously, the hydrolysis of the 1,2-epoxyalkane, always observed as side reaction, increased with rising sodium hydroxide concentration. Due to the constant concentration of sodium hydroxide during the conversion, higher NaOH concentrations should be avoided.
3.2 Influence of the concentration of 1,2-epoxyoctane
By variation of the concentration of 1,2-epoxyoctane from 1.2 to 5.0 eq. per AGU, a series of starch ethers was synthesized using the previously determined optimal amount of sodium hydroxide (0.5 eq. per AGU) (Tab. 2).
Tab. 2. Etherification of potato starch with 1,2-epoxyoctane; variation of the 1,2-epoxyoctane concentration.a)
a> Molar ratio of the reactants: AGU/NaOH/Na2SO4 = 1.0 : 0.5 : 0.5; solvent: water (total volume: 40 mL); reaction conditions: 140 °C. 3.9 bar, 4 h: potato starch containing 16% (w/w) water.
b) Determined by 1H NMR and elemental analysis.
The MS of the products, which were isolated in yields of 79%-89%, increased with increasing concentration of 1,2-epoxyoctane. The amount of hydrolysis products was not decreased by reducing the 1,2-epoxyoctane concentration and remained constant at approximately 45% for all conversions. For example, a highly hydrophobic starch ether with MS = 2.4 was synthesized using 5.0 eq. 1,2-epoxyoctane per AGU.
The properties of the prepared starch ethers could be distinctly influenced by the MS, which depended on the epoxyalkane concentration. From an approximate MS of 0.5, onward the compounds were insoluble in water, but soluble in mixtures of methanol and methylene chloride. The solubility in less polar organic solvents was increased with increasing MS, reflecting the increased hydrophobic character of the starch ether.
3.3 Influence of the starch source on molar substitution and yield of α-hydroxyoctyl starch ethers
Starches from different native sources can be distinguished by their characteristics such as crystallinity, mor-
Tab. 3. Etherification of starch with 1.2-epoxyoctane; variation of the starch.a)
a) Molar ratio of the reactants: AGU/NaOH/Na2SO4 = 1.0 : 1.0 : 0.5; solvent: water (total volume: 40 mL); reaction conditions: 140 °C, 3.9 bar. 4 h.
b) Amylose from potatoes (Fluka).
c) Hylon 7: amylo maize starch containing 70% amylose.
d) Determined by 1H NMR and elemental analysis.
Tab. 4. Etherification of potato starch with 1,2-epoxyalkanes. a)
a) Molar ratio of the reactants: 1,2-epoxyalkane/AGU/NaOH/ Na2SO4 = 3.0 : 1.0 : 1.0 : 0.5; solvent: water (total volume: 40 mL); reaction conditions: 140 °C, 3.9 bar, 4 h; potato starch containing 16% (w/w) water.
b) Molar ratio of the reactants: 1,2-epoxyhexane/AGU/NaOH/ Na2SO4 = 3.0: 1.0 : 0.5 : 0.5.
c) Determined by 1H NMR and elemental analysis.
phology of the starch granule and composition of the starch fraction. As different starches are gelatinized by the same quantity of sodium hydroxide [22], the crys-tallinity and morphology of the starch granule should not affect the etherification with 1,2-epoxyalkanes. Hence, starches with different amylose contents were also converted to the corresponding ethers using a threefold excess of 1,2-epoxyoctane and an equimolar amount of sodium hydroxide to AGU (Tab. 3).
The starch ethers were obtained in high yields (79-96%) with MS of 1.3 up to 1.8. The reaction is hardly affected by the composition of the starches used, although a tendency was observed that starch ethers prepared from potato starch showed increased MS.
Amylose enriched starches have been of particular interest because of their linear structure, which is characteristic for synthetic fibers. Films prepared from amylose and amylose esters showed improved mechanical properties [23]. Therefore, it can be assumed that the amylose content has an influence on the mechanical and thermoplastic properties of α-hydroxyalkyl starch ethers.
3.4 Variation of the 1,2-epoxyalkane
The influence of the 1,2-epoxyalkane chain length on the properties of starch ethers was also investigated (Tab. 4).
To this end the reaction was carried out using a series of terminal epoxides from 1,2-epoxyhexane to 1,2-epoxydo-decane. For all conversions, constant molar ratios of 1,2-epoxyalkane : sodium sulfate : AGU = 3.0 : 0.5 : 1.0 were used. Due to the higher reactivity of 1,2-epoxyhexane, the molar ratio of sodium hydroxide : AGU was reduced to 0.5. whereas other reactions were performed using equimolar amounts of sodium hydroxide : AGU. The starch ethers obtained from the conversions of 1,2-epoxyoctane, 1,2-epoxydecane and 1.2-epoxydodecane
showed MS values of 1.8, entailing a decrease in the yield by an increase in the 1,2-epoxyalkane chain length. In the case of the more reactive 1,2-epoxyhexane. a starch ether with a MS = 1.1 was achieved under these conditions. Obviously, the hydrolysis of the 1,2-epoxyhexane was preferred at 140 °C. Due to the lower reactivity of the 1,2-epoxydodecane, the reaction time had to be extended to 6 h in order to achieve complete conversion of the epoxide. The main part of the hydrolysis products was incorporated in the precipitated α-hydroxydodecyl starch ether, which needed an intensive purification using iso-hexane as solvent.
The hydrophobic character of the α-hydroxyalkyl starch ethers increased with increasing 1,2-epoxyalkane chain length, as reflected by a higher solubility in less polar organic solvents of starch ethers with equal MS values.
3.5 Material properties
The material properties were investigated at the Fraun-hofer Institute for Process Engineering and Packaging (Freising, Germany).
First of all starch ethers were characterized by the melt flow index (MFI) in dependence of temperature and by differential scanning calorimetry (DSC) for determination of the glass transition temperatures. The MFI of an a-hydroxydodecyl starch ether with MS = 1.6, for example, was approximately 3 g/10 min at 150 °C and 6 g/10 min at 160 °C with a load of 5 kg. This flow behavior is comparable with that of low density polyethylene (LDPE). According to the MFI-values. the hydrophobic starch ethers were converted by a co-rotating twin screw laboratory extruder into 5 cm wide and 1-2 mmthick ribbons at processing temperatures in the range of 130-160 °C. These ribbons have a modulus of elasticity in tension comparable to LDPE but an insufficient elasticity. All ribbons show
a very strong permeability for gases. For example the permeability for water vapor of a α-hydroxydodecyl starch ether with MS = 1.6 is about 500 times larger than the permeability of LDPE. The permeability for oxygen is 40 times larger.
A detailed description of the material properties of a-hy-droxystarch ethers will be given in a publication by the Fraunhofer Institute, which is in preparation.
4 Conclusions
A series of α-hydroxyalkyl starch ethers (C6 to C12) was prepared by a two-phase reaction of an aqueous alkaline starch gel with 1,2-epoxyalkanes at 140 °C and 3.9 bar in the presence of sodium sulfate as a co-catalyst. Thus, starch ethers with various MS values were obtained in high yields by optimization of the reaction parameters such as the sodium hydroxide concentration. It was possible to adjust the molar substitution using the 1,2-epoxyalkane concentration and, in this way, to influence specifically the hydrophobic character of the products.
Furthermore, the hydrophobic character of the starch ethers depends on the chain length of the 1,2-epoxyalka-ne used. In view of thermoplastic processing, the material properties have been investigated at the Fraunhofer Institute for Process Engineering and Packaging (Freising, Germany).
5 Acknowledgement
We would like to thank Dr. K. Bergander and Mrs. K. Voss. University of Munster, for 1H NMR/13C NMR spectra, Mrs. Gottschalk, University of Münster, for the elemental analysis, Dr. K. Huberand Dr. H.-J. Bach, Fraunhofer Institute for Process Engineering and Packaging, Freising, for investigation of the material properties. We gratefully acknowledge financial support of this work by the Bundesministerium für Ernährung, Landwirtschaft und Forsten, Bonn and the Fachagentur Nachwachsende Rohstoffe e.V., Gülzow.
Bibliography
[1] Fachverband der Stärke-lndustrie e.V.: Zahlen und Fakten zur Starke-lndustrie, Bonn 1997.
[2] A. Vetter, T. Graf: Markt und Innovationspotential pflanzli-cher Inhaltstoffe fur die Landwirtschaft und Industrie. 2. Internationales Symposium "Werkstoffe aus nachwachsen-den Rohstoffen". Erfurt, Sept. 1999.
[3] J. L. Willet: Mechanical properties of LDPE/granular starch composites. J.Appl. Polym. Sci. 1994, 54, 1685-1695.
[4] W. Aichholzer, H.-G. Fritz: Charakterisierung der Starke-destrukturierung bei der Aufbereitung von bioabbaubaren Polymerwerkstoffen. Starch/Starke 1996. 48. 434-444.
[5] Biotec GmbH: Product information. Emmerich 1999. Germany.