Biomacromolecules 2002, 3, 84-91
84
Crystallinity and Morphology in Films of Starch, Amylose and Amylopectin Blends Åsa Rindlav-Westling,† Mats Stading,‡ and Paul Gatenholm*,† Department of Polymer Technology, Chalmers University of Technology, S-412 96 Go¨teborg, Sweden; and Chalmers University of Technology and SIK, The Swedish Institute for Food and Biotechnology, P.O. Box 5401, S-402 29 Go¨teborg, Sweden Received July 9, 2001
Films of potato starch, amylose, and amylopectin and blends thereof were prepared by solution casting and examined using X-ray diffraction, light microscopy, transmission electron microscopy, and differential scanning calorimetry. Amylose films had a relative crystallinity of about 30% whereas amylopectin films were entirely amorphous. Blending of amylose and amylopectin resulted in films with a considerably higher degree of crystallinity than could be predicted. This is explained by cocrystallization between amylose and amylopectin and possibly by crystallization of amylopectin. The crystallized material gave rise to an endotherm detected with differential scanning calorimetry. The enthalpy and peak temperature of the transition also increased as the water content decreased. When the amylose proportion in the blends was low, separate phases of amylose and amylopectin were observed by light microscopy. At higher amylose proportions, however, the phase separation was apparently prevented by amylose gelation and the formation of a continuous amylose network. The amylose network in the films, observed with transmission electron microscopy, consisted of stiff strands and open pores and became less visible as the amylose proportion decreased. The water content of the films was dependent on the microstructure and the crystallinity. Introduction In addition to being a major nutrient, starch is widely used in technical applications and has recently gained interest as a renewable and biodegradable plastic. Films can be made from starch or its components, amylose and amylopectin, by various techniques such as thermoplastic processing and solution casting. Starch is a (1f4) linked poly-R-D-glucan and is a native blend of the linear amylose and the (1f6) branched amylopectin. These two components can be separated from each other and new blends with various proportions can thus be made. Blending two different polymers is common in work with synthetic polymers to achieve the desired properties for plastic applications and in biopolymers in food applications. When blending two polymers, it is important to know whether they are miscible or whether they will phase separate. Different analytical techniques such as microscopy, differential scanning calorimetry, NMR, IR, and scattering techniques can be used to study miscibility and phase morphology. For solutions of two polymers, it is useful to employ a phase diagram that shows the phase separation behavior as the concentrations are varied. Phase separation between different biopolymers in solution is common. However, many biopolymers also form gels when the macromolecules aggregate and create a physical network.1 †
Department of Polymer Technology, Chalmers University of Technol-
ogy. ‡ Chalmers University of Technology and SIK, The Swedish Institute for Food and Biotechnology.
Competition between gel formation and phase separation may occur in mixed biopolymer-water systems. If the relative rate of phase separation is lower than the relative rate of gelation, a non-phase-separated system is formed upon cooling from solution.2 The homogeneous network structures thus created prevent phase separation. The specific biopolymer pair, concentration, temperature, and other conditions determine this behavior.1 Amylose and amylopectin have been shown to phase separate in water solutions, provided the concentrations are high enough,3 and both amylose and amylopectin may form gels. Amylose is known to have a fast gelation, where the turbidity and shear modulus reach a maximum plateau after about 40 min.4 The gelation of amylopectin, on the other hand, is a slow process taking several weeks.5 In addition to the gelation of amylose, crystallinity is also formed.4 The A and B type crystalline structures are made up of specific arrangements of double helices. Amylose reaches a relatively high final crystallinity when dried from a water solution to a film.6 In contrast to amylose, amylopectin forms totally amorphous films under the same conditions but can crystallize, for instance when glycerol is added.6 The crystallization process of polymers proceeds in two steps: nucleation and then growth of the crystal. The resulting crystallinity and morphology depend on the crystallization conditions.7 If two polymers in a blend can crystallize there may be some extent of cocrystallization between the polymers, although this is rare. An example of cocrystallization between a linear fraction of polyethylene (HDPE) and a branched polyethylene (LDPE) is given by Conde Bran˜a and Gedde.8
10.1021/bm010114i CCC: $22.00 © 2002 American Chemical Society Published on Web 11/29/2001
Films of Starch, Amylose, and Amylopectin Blends
The behavior and properties of blends of amylose and amylopectin have been studied by several authors. The studies of Wolff et al.9 and Lourdin et al.10 on the mechanical properties of solution cast films showed increasing elongation and strength with an increasing amount of amylose. Van Soest and Essers11 studied the mechanical properties and crystallinity of extruded films and found increasing crystallinity with increasing amylose concentration. Indications of the occurrence of cocrystallization between amylose and amylopectin have been found by a few authors. Leloup et al.12 found that a fraction of amylopectin in amylose/ amylopectin mixed gels had become resistant to acid hydrolysis and suggested cocrystallization as a possible explanation. A lack of birefringence from amylose crystallites in extrusion cooked starch gels led Mestres et al.13 to believe that cocrystallization had taken place. High crystallinity in compression molded starch was explained by cocrystallization by Hulleman et al.,14 and Gudmundsson and Eliasson15 also found indications of cocrystallization when mixtures with low proportions of amylopectin retrograded to a higher extent than expected as observed with DSC. The latter two authors also mentioned the possibility of amylose inducing the crystallization of amylopectin. Leloup et al.12 suggested that a phase inversion takes place as the amylose content in mixed amylose/amylopectin gels increases. They based this explanation on studies of mechanical compression of gels where a sharp increase in the apparent modulus was found after the inversion point. The gels were said to transform from a continuous amylopectin matrix with dispersed amylose domains into a continuous amylose matrix with amylopectin domains at an amylose/amylopectin ratio of 30/70, although no microscopic studies were conducted. Doublier and Llamas16 also suggested phase inversion at an amylose/ amylopectin ratio of 15/85 on the basis of rheological measurements. When purified amylose was added to a molecularly dispersed potato starch in the study by Svegmark et al.,17 the phase separation observed with light microscopy became less visible as the amylose proportion increased and a bicontinuous system was proposed. The rheological data indicate that this inversion appeared at an amylose proportion of 25%. The objective of this research project is to gain knowledge about the starch behavior to be used in the development of new plastic materials. Previous studies showed improvements in material properties when amylose and amylopectin films were allowed to crystallize.6 The influence of amylose/ amylopectin proportions on the properties is of importance when choosing raw materials among different starches and considering newly developed starches with various amylose contents. The aim of this work was to study how the proportions of amylose and amylopectin affect the morphology, crystallinity, and endotherms in amylose, amylopectin, and starch-blended films. Materials and Methods Materials. Potato amylose (104561) was purchased from Merck. The amylopectin was in the form of granular amylopectin potato starch and was kindly supplied by
Biomacromolecules, Vol. 3, No. 1, 2002 85
Lyckeby Sta¨rkelsen. This potato was developed by Lyckeby Sta¨rkelsen and Svalo¨f Weibull using genetic engineering in order to suppress amylose synthesis.18 Native granular potato starch was also supplied by Lyckeby Sta¨rkelsen. Film Formation. The amylose was dissolved in ultrafiltered deionized water (3% w/w) by heating in a pressuretight autoclave in an oven. The autoclave interior temperature was kept above 135 °C for 2 h and eventually reached a maximum of 145 °C. A clear, viscous solution was obtained. Higher temperatures resulted in degradation such that no coherent films were formed. The granular amylopectin starch and native potato starch were gelatinized during stirring and heating to 75 °C prior to dissolution in an autoclave in the same way as the amylose. Amylose and amylopectin were mixed in different weight proportions: amylose/amylopectin 8/92, 25/75, 50/50, and 75/25. Starch was also mixed with amylose and amylopectin in the weight proportions of starch/ amylopectin 40/60 and starch/amylose 67/33. Appropriate amounts of amylose, amylopectin, and starch were mixed, gelatinized, and dissolved in an autoclave in the same way as described above. The total polysaccharide concentration was in all cases 3% w/w. The solutions obtained after autoclaving were poured onto polystyrene Petri dishes and left to dry at 23 °C and 50% RH in a climate room. An equilibrium water content of the films was reached in less than 3 days. A mixture of amylose/amylopectin in the ratio of 8/92 was also dried at 23 °C in 90% RH. To prepare samples for X-ray diffraction, DSC and water content measurements, the films were powdered by grinding in a mortar, 2 × 10 s. The films were immersed in N2(l) during grinding in order to make them brittle and minimize damage from heat and shear. The powders were reconditioned in 54% RH for at least 3 days before any measurements were made to ensure equilibrium water content. X-ray Diffraction. A wide-angle X-ray diffractometer (Siemens D5000) was used in the reflection geometry with nickel-filtered Cu KR radiation. A variable divergence slit was used, giving an irradiated area with a diameter of 20 mm as the sample holder was rotated at 30 rpm. An antiscatter slit of 0.6 mm, a detector slit of 0.2 mm, and a scintillation detector were used. Diffractograms were taken between 5 and 30°(2θ) at a rate of 1°(2θ) min-1 and with a step size of 0.1°(2θ) and were thereafter smoothed. Measurements were made in triplicate. To compare the crystallinities of the samples, a relative crystallinity was calculated; no reference for 100% crystalline material was needed. The background was subtracted from the diffractogram by drawing a straight baseline tangentially to the curve minimum obtained at 7°(2θ). A relative crystallinity can be determined by comparing the area under the crystalline peaks with the area of the amorphous halo under the peaks.19 In our case, the most distinct peak, located at 17°(2θ), was chosen to represent the crystalline peaks. Microscopy. Light microscopy photos were taken in a Nikon FXA microscope. The films were first stained with a droplet of 1:1 Lugols solution:water. Amylose was thereby colored blue and amylopectin purple, making it possible to visualize a phase separation between the two. The color
86
Biomacromolecules, Vol. 3, No. 1, 2002
images have been converted here into a gray scale, where amylose appears darker than amylopectin. Transmission electron microscopy (TEM) was used to study the films in a JEOL 100 CX-II microscope or LEO 906e at an acceleration voltage of 80-100 kV. The films were prepared by fixation in 2% glutaraldehyde and stepwise dehydration in an ethanol series. The ethanol was then replaced in steps, first by propylene oxide and thereafter by a cured epoxy resin (Polybed, Polyscience Inc.). The embedded samples were sectioned in ∼70 nm thick slices using a diamond knife in a Reichert-Jung Ultracut E. The sections were transferred onto Formvar-supported gold grids and stained with periodic acid, thiosemicarbazide, and silver proteinate (PA-TSC-SP) by the method described by Thie´ry.20 It was not possible to embed the films of amylopectin or phase separated films with a continuous amylopectin matrix, probably because the amylopectin partly dissolved in the preparation steps. Water Content. The films were ground in liquid nitrogen and conditioned in 54% air humidity before the water content of the samples was determined gravimetrically. Measurements were made in triplicate by drying in a vacuum oven, 60 °C for 28 h. The air humidity, 54% RH, was obtained using an Mg(NO3)2 saturated salt solution. Differential Scanning Calorimetry. A Perkin-Elmer DSC-7 was used and measurements were made in triplicate. Scans were made by heating from 0 to 160 °C, cooling to 0 °C, and reheating from 0 to 160 °C at rates of 10 °C/min. Closed stainless steel cups with about 20 mg of sample were used with an empty cup as reference. Powdered samples were used to achieve better contact and heat transfer from sample cups. The samples were reconditioned in 54% RH before any measurements were made. Starch films were also reconditioned in various air humidities to obtain varying water contents before DSC measurements. Saturated salt solutions of MgCl2, Mg(NO3)2, NaCl, and K2SO4 were used to obtain air humidities of 33, 54, 75, and 97% RH, and the water contents were determined gravimetrically by drying in a vacuum oven, 60 °C for 28 h. Results and Discussion Crystallinity. Films with varying amounts of amylose and amylopectin were prepared by solution casting from a 3% water solution. Their relative crystallinity was measured by wide-angle X-ray diffraction. All crystalline films had a B-type crystalline structure, which is the natural structure for starch drying from a dilute solution.21 Films made of amylose had a relative crystallinity of 33%, whereas films made of amylopectin were entirely amorphous. The development of crystallinity in amylopectin gels takes weeks,5 and the amylopectin in the studied films did not crystallize before the water content, and thereby the mobility, was too low. The crystallinity of the films of amylose/amylopectin blends did not increase linearly with the amount of amylose present, as could be expected from a theoretical blend of amorphous amylopectin and 33% crystalline amylose; see Figure 1 and Table 1. There was instead a dramatic increase in the crystallinity up to a proportion of 25% amylose, but no
Rindlav-Westling et al.
Figure 1. Relative crystallinity as measured by WAXS for films of amylose/amylopectin blends and starch blends. Dashed line symbolizes a theoretical blend of amorphous amylopectin and 33% crystalline amylose. Table 1. Table of Total Amylose Content and Relative Crystallinity of Films of Starch, Amylose, and Amylopectin Blendsa material
% amylose
crystallinity, relative %
amylopectin (AP) AM/AP 8/92 AM/AP 25/75 AM/AP 50/50 AM/AP 75/25 amylose (AM) starch starch/AP 40/60 starch/AM 67/33
0 8 25 50 75 100 22 9 48
0 23.5 36.1 35.2 33.3 33.3 21.7
