Twin-screw extrusion production and characterization of starch foam products for use in cushioning and insulation applications.
Foam plastic packaging is experiencing growing pressure from existing and proposed environmental and disposal regulations, and from market based sustainability initiatives. It presents a major disposal problem for companies and municipalities, as it is lightweight and bulky and so does not lend itself to a viable economic and environmentally responsible recycling operation due to high handling and transportation costs. It is not biodegradable, which makes disposal in soil or composting operations untenable. Further, issues such as sustainability, industrial ecology, biodegradability, and recyclability are becoming major considerations in a company's product packaging design, especially with single-use disposable packaging. There is, thus, a market need for bio-based, biodegradable foam plastic packaging that can be safely and effectively disposed of in soil or in composting operations, while retaining all of the current foam plastics performance requirements. In previous work, we have reported on the rationale, design, and engineering of bio-based, biodegradable polymer materials [1-4].
Extruded starch foams with polyvinyl alcohol (PVA) are patented by Lacourse and Altieri  and Roesser et al. . Altieri and Tessler  patented the water-resistant foams from blends of starch with starch esters. Bastioli et al.  patented foams from blends of starch with 10-30% of polymers such as PVA, poly (caprolactone), cellulose acetate, poly(ethylene vinyl alcohol), and poly (ethylene-co-acrylic acid). Neumann and Seib  patented the technology to make biodegradable starch-based foams using polyalkylene glycols.
The physical, mechanical, and/or thermal properties of starch foams with various additives have been studied by Hutchinson et al. , Chinnaswamy and Hanna , Warburton et al. [12, 13], Bhatnagar and Hanna , and Wang et al. . Hutchinson et al.  conducted tension, compression, and flexure tests, and developed a power-law correlation between the mechanical properties and bulk density of the foams made from maize grits. Chinnaswamy and Hanna  reported that the optimum temperature for maximum expansion of cornstarch extrudates increased from 130 to 160[degrees]C, as the amylose content increased from 0 to 70%, and the bulk density of the extrudates decreased as the amylose content in the starch increased. Warburton et al. [12, 13] showed that the Young's modulus  and fracture stress  decreases with increasing starch content. Bhatnagar and Hanna  extruded regular cornstarch with either polystyrene or poly (methyl methacrylate) at a 70:30 ratio with other additives in a single-screw extruder. Foam densities in the range of 29.5-132 kg/[m.sup.3] were obtained, with radial expansion ratios (ER; ratio of the cross sectional area of the foam to that of the die) of 8.8-40.1. Wang et al.  showed that cornstarch foams had better radial expansions but higher unit and bulk densities as compared to wheat starch foams.
Cha et al.  studied the moisture adsorption isotherms, bulk densities, and expansion ratios (ERs) of the starch-based foams involving about 33% synthetic polymers (polystyrene), which reduces its biodegradability. They showed that the bulk density of the starch foams decreased with an increase in extrusion temperature up to 160[degrees]C, but highest expansion was obtained at 140[degrees]C. Also, chemical blowing agents were used in addition to water as the physical blowing agent.
Shogren  reported that extruded foams made from acetylated high amylose starch had higher water resistance and bulk densities of 40-60 kg/[m.sup.3]. Willett and Shogren  extruded blends of normal cornstarch as well as high amylose cornstarch, wheat starch, and potato starch with resins such as PVA, cellulose acetate, and several biodegradable polyesters such as PLA, poly(hydroxyester ether) (PHEE), poly(caprolactone), poly(ester amide) (PEA), and poly(hydroxybutyrate-co-valerate), among others. Winkler et al.  extruded blends of starch-based thermoplastic hydroxy-functionalized polyetheramines, which were pelletized and formed into loose-fill packaging foams using a twin-screw or a single-screw extruder.
In this study, we report the use of a new polymer poly(hydroxyl aminoether) (PHAE) to provide flexibility and resilience, thereby eliminating one of the major problems associated with starch foams. PHAE offers the adhesion and durability of epoxy resins with the flexibility and processibility of thermoplastic resins. However, the biodegradability of PHAE has not yet been confirmed. We also report the engineering parameters developed on the ZSK-30 twin-screw extrusion (TSE) system, such as the screw configuration and optimum temperature, water, and PHAE contents necessary to make a portfolio of foam products with control of cell structure and die shapes, for applications in insulation and cushioning.
The optimum conditions determined on the ZSK-30 were then implemented on the Wenger-80, to scale up the process to produce foam sheets. The target applications were cushion packaging and insulation.
The type of starch used was hydroxypropylated high amylose cornstarch (70% amylose content). The starch was purchased from National Starch and Chemicals (Indianapolis, IN), under the trade name of HYLON 7. The density of HYLON 7 starch is 1.2 g/[cm.sup.3]. The inherent moisture content of the starch was 11.2% under ambient conditions. Water was used as the plasticizer as well as the blowing agent. Water content was maintained at 7-10% of the starch used. Talc (magnesium silicate), used as the nucleating agent, was obtained from Luzenac (Ontario, Canada). It has a specific gravity of 2.76 and a bulk density of 150 kg/[m.sup.3]. The talc content was maintained at 1% for all the experiments. PHAE is an additive that offers the adhesion and durability of epoxy resins with the flexibility and processibility of thermoplastic resins.
PHAE was purchased from Dow Chemicals (Midland, MI), under the trade name BLOX 110. PHAE has a melt temperature of 75[degrees]C, and is produced by reacting liquid epoxy resin (LER) with hydroxy functional dinucleophilic amines and diglycidyl ethers of bisphenol-A, hydroquinone, or resorcinol (RDGE) .
The experimental setup used in this study was a TSE system. The TSE system consisted of an extruder driver with a speed control gearbox, a Werner Pfleiderer ZSK-30 twin-screw co-rotating extruder with a screw diameter of 30 mm, an L/D of 32, a positive displacement pump for injecting water into the extruder, accurate single-screw feeders for feeding starch, and PHAE and talc could be fed individually or as a mixture. A cylindrical filament die 2.7 mm in diameter and 8.1 mm in length, with a cooling sleeve, was assembled to the extruder. The sensors were mounted on the die to measure the temperature and pressure of the melt. A high-speed cutter was used to get cylindrical foam samples of required size.
The foam sheets were produced on an industrial scale twin-screw food extruder, a Wenger-80, having a screw diameter of 80 mm and an L/D of 16. An annular die of width 2 mm was used. Talc was not used in the production of starch foam sheets in order to get the minimum density product. Also, the addition of talc rendered the product more brittle.
The objective to be attained in an extruder is to ensure proper plasticization of granular starch, thus disrupting its crystalline structure, and complete dissolution of the blowing agent in the plasticized starch, by promoting convective diffusion under high processing pressure, to form a single-phase plastic.
[FIGURE 1 OMITTED]
The screw configuration played an important role in this foaming process, because unlike conventional plastics that require only temperature to form a melt phase, plasticization of starch required thermal energy as well as mechanical energy in the form of shear to form a thermoplastic melt. Hence, the screw configuration had to be so designed to achieve the aforementioned objective, and to build up enough pressure at the die to cause uniform foaming. Figure 1 shows the foaming screw configuration on the small scale ZSK-30 TSE system. It is worth noting that two other screw configurations (not reported) were used for the foaming operation. These screw configurations had low and medium levels of shearing through kneading blocks. Also, the left-handed elements were not used in these configurations to build up pressure within the extruder. The control starch foam density obtained using the low shear screw configuration was 52.2 [+ or -] 2.3 kg/[m.sup.3], while it was 38.6 [+ or -] 1.8 kg/[m.sup.3] using the medium shear screw configuration. These values were higher as compared to those reported later in the results and discussions section.
Manipulating the screw configuration is a tedious process. Hence, when a density of ~30 kg/[m.sup.3] was obtained for the control starch foams using the screw configuration reported (Fig. 1), no further attempts were made to modify the screw configuration.
Foaming Screw Configuration
The foaming screw configuration is explained on the basis of the following zones.
Feed/Conveying Zone. The feed zone consisted of all right-handed conveying screw elements, also called as right-handed bushings. Unlike plastic resin processing, starch tends to clog up the feed zone. Hence, four large pitch (42-mm) right-handed conveying screw elements were used to quickly convey the starch and prevent build-up/clogging in the feed zone, which is a major issue in processes involving starch.
The starch, PHAE, and talc were added to the feed hopper, while the water was injected through a side tube.
The elements were so arranged that their pitch reduced along the length of the screw. This was done to increase the pressure along the length of the screw, and it also helped in increasing the residence time. The increase in pressure would help in increasing the solubility of water in the thermoplastic starch melt. Density of the feed is low because of the air trapped in the incoming granular raw material. The incoming material was compressed and the air was expelled. The water was injected in this feed zone to facilitate textural and viscosity development and to enhance conductive heat transfer. Decreasing the pitch increased the surface-to-volume ratio, thus increasing the conversion of mechanical energy to heat through friction. However, the positive pumping characteristic of these conveying elements limited the ability of the screws to effectively convert mechanical energy into heat through friction. Hence, kneading elements were added. These elements reduced the positive conveying feature and, thus, forced the extruder barrel to fill, which helps in building up pressure, which assists foaming at the die.
Kneading/Mixing Zone. The wide-flighted dispersive kneading blocks helped break the starch granules into smaller particles. The shearing action of the wide-flighted elements, between the barrel and the flights, was responsible for the breakup of the granules into finer particles. The narrow-flighted kneading blocks in this kneading zone provided better homogenizing properties, thus enhancing the blending of starch, water, and the additives.
The next set of screw elements were the neutral kneading blocks, which held the material for longer and thus increased the residence time. This would further facilitate the dissolution of water in the melt, and ensure complete plasticization.
The kneading zone continued the compression started in the feeding zone, and the flow channels of the extruder achieved a higher degree of fill as a result of the compression of the extrudate. Basically, in the kneading zone, starch granules were broken down into finer particles, and the hydrogen bonds are broken down in the starch matrix by the water molecules, resulting in a thermoplastic melt of plasticized starch. Within this zone of the extruder barrel, the extrudate began to lose its granular definition, the density began to increase and pressure began to develop in the barrel.
Within the kneading zone, the discrete particles of starch begin to agglomerate due to the increasing temperature resulting from conduction, and viscous energy dissipation.
The left-handed elements were mainly used for building up high pressures inside the extruder. They were included downstream, to build up higher pressures than the solubility pressure.
Conveying Zone. This conveying zone released the built-up pressure slightly, and helped convey the plasticized starch to the next kneading zone, thus preventing a build up of material in the extruder. A left-handed bushing was introduced at the end of the conveying zone to increase the pressure, so that the resulting melt is supersaturated with water.
Kneading Zone. Another kneading zone was incorporated in the configuration to ensure complete homogenization of the materials, and dissolution of the water in the starch matrix. This zone was sealed on both sides by left-handed elements, thus ensuring complete mixing of the material held in the kneading zone. The extrudate formed a more integral flowing dough mass as it moved through this kneading zone and typically reached its maximum compaction. This is the zone of the extruder, where the mass became amorphous and texturized. Temperature and pressure increased most rapidly in this region, and shear rates were highest because of the screw configuration and maximum compression of the extrudate.
Final Conveying Zone. The last conveying zone forced the supersaturated melt through the die. Pressure, temperature, and the resulting fluid viscosity were such that the extrudate is forced from the extruder, creating the desired final product texture, density, and functional properties.
The nucleating sites were generated at the sudden drop in pressure at the die exit, and the supersaturated blowing agent diffuses out through the matrix, giving rise to a foamed structure.
The L/D for the Wenger TX-80 was half of that of the ZSK-30 TSE used earlier. Food extruders typically use a lower L/D ratio [10-18], as compared to a plastics extruder, to minimize degradation of starches extruded. The screw configuration for the TX-80 was designed corresponding to the configuration for the ZSK-30 discussed earlier. The L/D for each conveying and kneading zones on the ZSK-30 were calculated as a fraction of the total L/D. These fractions were then used to determine the screw configuration of the Wenger TX-80.
The temperatures in the ZSK-30 extruder zones were set up to reach the required temperatures. The temperature profile for the best product obtained was as follows:</p> <pre> Feed (Zone 1): 20[degrees]C (cold feed) Zone 2: 100[degrees]C Zone 3: 110[degrees]C Zone 4: 120[degrees]C Zone 5: 120[degrees]C Zone 6: 110[degrees]C Die: 105[degrees]C Melt temperature: 105-107[degrees]C. </pre> <p>The feeder for starch was calibrated and set at a particular speed. The other feeder/feeders were calibrated and set at feeding rates accordingly. Initially, during startup, water was pumped into the system immediately after the feed throat, at 15-20% of the starch fed, and later its flow rate was reduced to about 7-10% of starch. The inherent moisture present in starch (11.2%) also helped in plasticization of starch. The different parameters were varied one at a time, while the other parameters remained those of the basic setting. When formulations were changed, extrusion was continued until the torque and the die pressure stabilized. Extrusions were carried out at a torque of 70-75%, and a pressure of 4.8-5.2 MPa. The foam was extruded at a rate of about 11-12 kg/hr, while the foam sheets were extruded at a throughput of 410-420 kg/hr.
Characterization and Analysis
The samples collected were conditioned as per ASTM D-4332 , in a constant environment room at 23 [+ or -] 1[degrees]C (73.4 [+ or -] 3.6[degrees]F) and 50 [+ or -] 2% RH (relative humidity) for at least 72 hr before testing.
This test method covers the determination of density of foam by calculation from the mass and volume of a regularly shaped specimen. The test method ASTM D-3575 (section 43, Method A)  was used. At least 10 specimens were measured for each formulation.
The dimensions of the sample were measured using Vernier Calipers graduated to permit measurements accurate to 0.0025 cm. ER was calculated as the ratio of the cross-sectional area of the foam to that of the die.
Compressive Strength and Resiliency
Compressive strength and resiliency describe the mechanical integrity of the foam. Compressive strength of the lab-scale specimens was measured according to the test method explained by Tatarka and Cunningham , on a UTS (ultimate tensile strength) tensile testing machine. Foam specimens were securely fastened lengthwise and compressed by a steel probe (0.635 cm diameter) with a hemispherical end-cap. By lowering the piston to the foam surface, an initial load of 0.5 N was applied on the specimen for approximately 5 s. From this point, the probe was lowered at a rate of 30 mm/min for a distance of 3 mm. The maximum load was recorded. After 60 s had elapsed, a relaxation load was recorded. Compressive strength was determined by dividing the maximum load by the cross-sectional area of the probe. Resiliency is the percentage of the compressive force after the 60 s hold period divided by the maximum force required to compress the foam by 3 mm. Averages were calculated from five sets of starch foam specimens.
Moisture Sorption Analysis
A total of 10 blocks of each formulation, collected at different times were placed in an environmental humidity chamber, subject to a relative humidity of 95 [+ or -] 5%, and a temperature of 38 [+ or -] 5[degrees]C. The weight and the dimensions (length and diameter) of the samples were monitored. They were measured at regular intervals using an accurate weighing balance and a pair of Vernier Calipers, to the third decimal place. The entities measured at different time intervals were normalized using the value measured before placing the samples in the humidity chamber (time, t = 0). The weight and dimensions of the samples were recorded until a steady state value was reached (approximately 30 days). The results for a formulation were obtained as an average over the 10 samples used for that formulation.
Environmental Scanning Electron Microscopy (ESEM)
Foam samples were sectioned with a razor blade and mounted on aluminum stubs with graphite filled tape, and examined with a Phillips Electroscan 2020 environmental scanning electron microscope (ESEM).
Thermal Resistance (R-Value)
The thermal resistance is a measure of resistance to heat flow. The ability of insulation to slow the transfer of heat is measured in R-values. The higher the R-value, the better is the insulation material's ability to resist the flow of heat through it. The C-value (C) is a measure of the thermal conductance of the material and is the reciprocal of R, or
C = 1/R. (1)
C is determined only when the thermal conductivity (k) of a material is known.
C = k/thickness. (2)
The test method ASTM C 177-97  was used to determine the thermal conductivity of the foam sheet samples. The thicknesses used were 2.5, 5, and 7.5 cm. Three samples of each thickness were run and the average was reported (Table 1).
Temperature Hold Time Test
A temperature hold time test was carried out to compare expanded polystyrene (EPS), polyurethane (PUR), and starch foams as insulation materials. The test was intended to be a qualitative test to compare the performance of multiple cases. Case configurations (sizing, insulation, amount of refrigerant) were determined by product requirements. Product considerations such as temperature requirements and product size and thermal properties mean that a hold time for one product may be different from another unique product. The purpose of this test was to compare hold times for identical case configurations in different cases, not to establish a "case hold time."
As this was a comparative test, there was some flexibility in testing procedure. The temperature probes were placed in direct contact with the product. The cases were lined with plastic bags to prevent water damage from condensation or possible leakage. Since the cases were all set up in the same method, the test produced useful results. All cases were taped closed with clear packaging tape. A 1000-ml Nalgene bottle with a water fill was used as the product, and was staged for 24 hr prior to testing. The refrigerant consisted of two large packs of 1 kg of gel-ice (Freez Pak brand, item no. 4984) per case. One package of gel ice was placed below the product and one above. The temperature probe was fed through a hole in the bottle so that it was in direct contact with its contents. The EPS, PUR, and starch foam cases were all 28 X 21.5 X 18 cm with a thickness of 5 cm at the wall. The temperatures were recorded over a length of 84 hr. The ambient temperature is always maintained at 22[degrees]C. Our criterion for comparison was the length of time the product stays under 8[degrees]C.
Dynamic Cushion Curve Testing
A dynamic cushion curve is a graphic representation of dynamic shock cushioning or transmitted shock in G's (dimensionless) over a variety of static loading conditions (psi or kPa) for a cushioning material thickness (or structure) at a specific equivalent free fall drop height. One cushion curve was generated for each material type, material thickness and drop height combination. The test procedure, as per ASTM D-1596 , was basically one of dropping a platen of specified weight from a known drop height onto a foam cushion of predetermined bearing area and thickness. The deceleration experienced by the platen at impact was monitored and recorded by an accelerometer. Also cushion curves were generated as per ASTM D-4168 , which is designed to evaluate foam-in-place cushioning materials in a manner in which the foam-in-place packaging material is used. In particular, the method included simultaneous use of the foam (cushion sample), and the box with other accessories used in this method of packaging. A platen of specified weight was placed in the box in contact with the foam cushion of predetermined bearing area and thickness. In this test, the entire packaged box was dropped from a known drop height. The accelerometer was in contact with the platen, which recorded the deceleration experienced by the platen.
It is important to identify the drop height from which the package will be expected to fall. There is, however, a certain inherent variability with the manner in which packages are handled. Generally, low level drops occur frequently, while very high-level drops are rare. The drop heights are assumed based on the weight of the product being handled and are presented in ASTM D-3332 . Typical fragility levels for different products are also discussed in ASTM D-3332. Extremely fragile products such as missile guidance systems, precision aligned test instruments, etc., have fragility levels in the range of 15-25 G, while rugged products such as industrial machinery have a fragility level greater than about 120 G. In this case, the drop height used was 0.75 m, assuming the product to be packaged would be in the range of 25-50 kg .
RESULTS AND DISCUSSION
The effect of process variables on the degree of foaming was investigated. These process variables included the content of water, the temperature of the thermoplastic starch melt, the PHAE content, and the screw speed.
[FIGURE 2 OMITTED]
Effect of the Amount of Water on the ER
The ER and density were very sensitive functions of the amount of water injected. The ER increased with an increase in the amount of water, but beyond a certain amount, it decreased (Fig. 2). The maximum ER was obtained when the water added is 7% of the starch used (wet basis). The inherent moisture content of the starch was 11.2%. Thus, the total moisture content of the starch in the process was 17.42%.
The ER decreased with the increase in the amount of water, due to enhanced gas loss, as a result of increased plasticizing effect. When water dissolved in the starch matrix, the viscosity of the melt dropped as a result of the free volume increase. The plasticizing effect increased with an increase in the amount of water injected. Similarly, the diffusivity of water in the starch matrix increased as well. Hence, the loss of moisture to the atmosphere by diffusion increased with an increase in the amount of water. Also, the use of high amount of water (>10%), led to high fluctuation of pressure in the barrel, due to which high processing pressures could not be achieved.
The maximum flow rate of water that can be injected to achieve a greater volume ER was limited by the solubility of the water in starch. When water did not completely dissolve in the starch matrix, it resulted in a nonuniform structure and much lower ER. The processing pressure should be higher than the solubility pressure to dissolve water in the starch matrix.
Effect of the Processing (Melt) Temperature on the ER
When stabilized processing conditions were obtained by completely dissolving the injected water (blowing agent), well-expanded starch foams were achieved in a certain temperature range. The maximum ER was obtained when the melt temperature was in the range of 100-110[degrees]C (Fig. 3). When the melt was overcooled below 90[degrees]C, the foam did not expand because the increased stiffness of the thermoplastic starch matrix was unfavorable for cell growth. Also, above 120[degrees]C, the ER was lower due to the escape of excess moisture to the environment during expansion. Thus, the foam product obtained was dry and brittle. The corresponding unit densities are shown in Fig. 4.
[FIGURE 3 OMITTED]
Effect of PHAE on ER
PHAE is a thermoplastic epoxy based resin offered by Dow Chemicals. It offered the adhesion and durability of epoxy resins with the flexibility and processibility of thermoplastic resins. When we included PHAE as an additive in our starch foam process, we obtained a foam structure with improved mechanical strength and toughness, weather and water resistance.
It was observed that the ER increased with an increase in the PHAE content, but only up to a certain level, and then decreased again. As shown in Fig. 5, the maximum ER and minimum density was obtained at a PHAE content of 7% of the starch used. The density of the starch foam was higher in the absence of PHAE. This is because thermoplastic starch did not have enough melt strength and flexibility to support the expansion. This lead to the rupture of cell walls and the surface of starch foams leading to the loss of water. The presence of a flexible polymer like PHAE, which is compatible with starch helped support foam expansion, and gave rise to a uniform cell size and structure, and hence a lower density material. However, at a higher PHAE content (greater than 7% PHAE), the foam density increased marginally. This was due to the fact that the pressure developed at the die was lower due to a reduction in melt viscosity of the extrudate. The die pressure was 5.4 [+ or -] 0.2 MPa in the absence of PHAE, while it was 5.0 [+ or -] 0.13 MPa and 4.4 [+ or -] 0.1 MPa in the presence of 7% and 15% PHAE, respectively. A more detailed study on the shear viscosities of the starch-based melts using PHAE in the extruder, and its effect on foaming, will be reported (Y.U. Nabar and R. Narayan, unpublished results). At higher PHAE concentrations, the expanded foams contracted after reaching a maximum radial expansion, leading to lower ERs. This contraction resulted in higher densities. This contraction may be due to a cooling rate not rapid enough to prevent cell collapse, as suggested by Willett and Shogren . With an increase in PHAE concentration, the specific length (cm/gm) increased too, implying an improvement in flow properties.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Effect of Screw Speed on the ER
Screw speed directly affected the degree of barrel fill, and hence the residence time distribution and the shear stress on the material being extruded. From Fig. 6 below, it was observed that there was no significant variation in the densities and the ERs of the starch foams at screw speeds varying from 200 rpm to 250 rpm. There was a marginal decrease in density at higher screw speeds. However at screw speeds lower than 175 rpm, the density increased considerably. Thus, screw speeds in excess of 175 rpm were normal. Figure 6 also shows that at higher screw speeds, the ER decreased too with a reduction in density. This implied a substantial increase in the specific length of the product. At these speeds, significant frictional heat was generated creating starch-melting phenomena with a reduction in the viscoelastic nature. One of the factors in determining the maximum volumetric output of the extruder is the screw speed. At the same starch feed rate, the torque loading on the motor is higher at lower screw speeds. Thus, the volumetric output at which we can process the starch foams is limited by the screw speed used. It is worth noting that a change in screw speed, however, would not change the volumetric throughput of the extruder since the twin-screw extruders are starve-fed. Also, the pressure developed at the die was lower at lower screw speeds. However, at higher screw speeds, there was an increase in wear rate of mechanical components such as screws and barrels. The specific mechanical energy also, increased with an increase in screw speed, and leveled off at higher screw speeds (Fig. 7). As the screw speed increased, the apparent melt viscosity can be assumed to decrease, which also decreased the net power input via viscous dissipation.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Compressive Strength and Resiliency
The compressive strength and resiliency of the foam samples is listed in Table 2. Typically, a power-law relationship is observed between compressive strength [[sigma].sub.c] and foam density [rho]([[sigma].sub.c] ~ [[rho].sup.n]). Denser foams tend to have thicker cell walls and hence resist deformation better than lower density foams with thinner cell walls . A strong relation existed between foam density and compressive strength (Fig. 8). The regression line in Fig. 8 was drawn using the data from Table 3, and a slope of 1.15 was obtained (n = 1.15). A value of n = 0.92 was obtained by Willett and Shogren , while Hutchinson et al.  reported exponents of 1.5-1.6 for compressive strengths of foams prepared from maize grits.
The addition of PHAE improved the resiliency considerably from 69.7% (control) to 93.5% at a PHAE content of 7% of the starch used.
Moisture Sorption Analysis
Starch foams gain weight as well as shrink in the presence of moisture. This gain in weight and the dimensional stability of the starch-based foams is important in packaging applications. The lower the loss in dimensions, the better is the dimensional stability. Table 3 shows the normalized steady state weight gains and dimensions of the starch foams (varying PHAE content), respectively. The hydrophobic behavior of the starch-based foams improved with an increase in the PHAE content. The hydroxypropylated high amylose starch foams gained about 13% of their original weight (time, t = 0), while shrinking by about 50% of their original length and diameter in the absence of PHAE. At the optimum PHAE content of 7%, the starch foams gained about 8% of the original weight, while shrinking by about 20% of their original dimensions. This was as a result of the high water-barrier properties of PHAE.
[FIGURE 8 OMITTED]
Scale-Up-Continuous Large-Scale Production of Starch Foam Sheets
The starch formulation developed and optimized for extrusion foam processing in the ZSK-30 was successfully employed on the large-scale Wenger TX-80. Table 4 shows the results obtained on the foam sheet line. The process was given sufficient time to reach a steady state, when any input parameters such as starch and PHAE feed rates, moisture content and temperature were changed. Once the steady state was attained, samples were collected every 2 minutes for 10 minutes, and the corresponding temperatures and torque readings were recorded. The results were consistent with the results obtained on the laboratory scale extruder.
Entries 1-3 show the effect of temperature on the process. Lower temperatures (entry 1) resulted in a denser product, and higher torques. Also the product obtained was hard and brittle. At higher temperatures (entry 3), though the product had a similar density, it was brittle with cracks developing on the surface. Also, surging of the product was observed at higher temperatures, leading to inconsistencies in the downstream cutting and laminating operations. In the higher temperature zones, the shear viscosity of the localized melt in those zones was lower as compared to the viscosity of the starch melt in the other zones. Thus, this surging was due to process instabilities resulting from inconsistent viscosities within the extruder.
Entries 4 and 5 show the effect of screw speed on the final product. At a screw speed of 450 rpm (entry 4), a lower density product was obtained, however, in the presence of surging. This was due to a lower degree of fill in the barrel, resulting in an inconsistent flow. At lower screw speeds, the viscosity of the material in the extruder was higher due to lower work, resulting in a slower, though stable throughput from the extruder.
Entry 6 shows an extrusion run with a lower starch feed rate, which resulted in lower torque and reduced pressure at the die. This resulted in a higher density product, which turned brittle over time due to the presence of excess moisture.
Entries 7 and 8 exhibit the effect of moisture content within the extruder. At a higher water content (entry 7), an effect similar to entry 6 was observed. The product seemed moist as it exited the extruder die, and showed low resiliency. The product hardened and became brittle over time, due to the presence of excess moisture. At low levels of moisture (entry 8), the torque was extremely high due to the highly viscous material in the extruder, and the product was dense in the absence of enough water (blowing agent). Also, the product had a rough texture (cracks on the surface), and the presence of hard spots in the foam implied incomplete gelatinization of the starch.
[FIGURE 9 OMITTED]
Entries 9 and 10 show the effect of PHAE content on the starch foams. As observed previously, the density of the starch foam was higher in the absence of PHAE (entry 9). The density of the starch foams reduced with the introduction of PHAE (entry 2). This entry shown in bold in Table 3 represents the current regular production run for the manufacture of starch foam sheets for sale. However, at a higher PHAE content (entry 10), the foam density increased marginally, This was due to the fact that the pressure developed at the die was lower due to a reduction in melt viscosity of the extrudate. This was obvious from the torque readings for the run. The product obtained, though having a slightly higher density than the product made by entry 2, was sellable because of its high flexibility.
The foam sheets thus obtained were targeted to apply to cushion packaging and insulation cooler applications. The following section deals with the determination of insulation and cushioning properties of the starch foam sheets. The ability to provide insulation was measured in terms of the R-value, while the ability to perform as cushion packaging was measured in terms of cushion curves.
Figure 9 shows the closed-cell structure and Fig. 10 shows the surface of the starch foam sheets without (Fig. 10a) and with 7% PHAE (Fig. 10b). PHAE provided a finer and a more stable surface preventing the rapid loss of moisture through the surface. This is probably due to the surface migration of a fraction of PHAE added during the production of starch foams, which is in good agreement with the results obtained with PHEE .
Thermal Insulation (R-Value)
All samples had an R-value ranging from 616.4 to 669.2 K-[m.sup.2]/kW (Table 4). The overall average R-value of the samples tested was 634 K-[m.sup.2]/kW. This was in the same class of performance as low-density polystyrene foam bead board, while provided much better insulation properties as compared to polyethylene. However, polyurethane exhibited the best insulation properties.
Temperature Hold Time Test
As shown in Fig. 11, it was observed that the starch foam sheets maintained the product under 8[degrees]C for 56.5 hr as compared to 51.5 hr for polystyrene and 83.5 hr for polyurethane foams. Thus, it proved to be a better insulator than polystyrene foam, polyurethane foams providing the best insulation. These results are also evident from the R-values obtained, polyurethane having the maximum R-value.
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
Dynamic Cushion Curve Testing
The cushion curves generated by ASTM D-1596 are shown in Fig. 12. It was observed that the cushion curves shifted downward with an increase in the thickness of the foam sheet, under identical conditions of static loading and drop height. The cushion curves indicated that the starch foam sheets tested (thickness of 2.5 cm and 5 cm) would provide enough protection for products having fragility levels of greater than 70 G, classified as moderately delicate, moderately rugged and rugged products. Foam sheets with thickness greater than 5 cm would be required for more delicate products having fragility levels of less than 70 G.
[FIGURE 12 OMITTED]
[FIGURE 13 OMITTED]
A similar trend was observed in the cushion curves generated from the testing standard ASTM D-4168 (Fig. 13). However, the peak deceleration values (G) generated were expected to be lower than those generated from ASTM D-1596. This is because the box and other accessories, apart from the foam cushion, take a part of the shock in the drop tests. More work needs to be done to correlate the thickness and foam cell structure with the cushioning properties.
A twin-screw extrusion process was successfully developed to produce biodegradable starch foams, where water functioned as the plasticizer as well as the blowing agent, talc as the nucleating agent, and PHAE imparted flexibility and processibility. The results obtained on a laboratory-scale extruder ZSK-30 (11-12 kg/hr) were successfully scaled-up to a continuous large-scale Wenger TX-80 (410-420 kg/hr) to manufacture starch foam sheets. Process engineering parameters such as screw configuration, temperature, input levels of water and PHAE were optimized. The ER increased with an increase in the amount of water injected into the system, but up to a certain extent only. The degree of foaming reduced when the water used was more than 7% of the starch used in the system. Minimum density of starch foams was obtained at a PHAE concentration of 7%, with sufficient flexibility. A large ER was obtained by optimizing the processing temperature, and the die temperature, which helps prevent the loss of moisture through the surface of the foam. The current technology to make starch based loose-fill was successfully translated to manufacture foam sheets. The extruded starch foam sheets provided excellent insulation properties. The R-values obtained suggested better insulation characteristics, as compared to expanded polystyrene, and efforts are in progress to enhance their thermal resistivity to make them comparable to polyurethane foams. Also, the dynamic cushioning data reveals that the starch foam sheets provided good cushioning or shock absorption properties. Future work is headed towards improving hydrophobic characteristics as well as the cushioning properties of these starch foams, in order to expand their use in the market place.
The author thanks KTM Industries, East Lansing, MI (www.ktmindustries.com) manufacturer of the starch foam products described in this publication.
1. R. Narayan, Kunstoffe, 79. 1022 (1989).
2. R. Rowell, T. Schultz, and R. Narayan, "Emerging Technologies for Materials and Chemicals from Biomass," in ACS Symp. Ser., No. 467 (1991).
3. R. Narayan, "Polymeric Materials from Agricultural Feedstocks," in Polymers From Agricultural Coproducts, in ACS Symp. Ser., M.L. Fishman, R.B. Friedman, and S.J. Huang, editors, No. 575, 2 (1994).
4. R. Narayan, "Commercialization Technology: A Case Study of Starch Based Biodegradable Plastics," in Paradigm for Successful Utilization of Renewable Resources, D.J. Sessa and J.L. Willett, editors, AOCS Press, Champaign. IL, 78 (1998).
5. N.L. Lacourse and P.A. Altieri, Biodegradable Packaging Material and the Method of Preparation Thereof, U.S. Patent 4,863,655 (Sep. 5, 1989); Biodegradable Shaped Products and the Method of Preparation Thereof, U.S. Patent 5,043,196 (Aug. 27, 1991).
6. D.S. Roesser, J. Nevling, D.C. Rawlins, and R.J. Billmers, Biodegradable Expanded Starch Products and the Method of Preparation, U.S. Patent 6,107,371 (Aug. 22, 2000).
7. P.A. Altieri and M.M. Tessler, Water Humidity Resistant Starch Foams, U.S. Patent 5,554,660 (Sept. 10, 1996).
8. C. Bastioli, V. Bellotti, L. Del Giudici, R. Lombi, and A. Rallis, Expanded Articles of Biodegradable Plastic Materials, U.S. Patent 5,360,830 (Nov. 1, 1994); Biodegradable Foamed Plastic Materials, U.S. Patent 5,736,586 (Apr. 7, 1998); Biodegradable Foamed Articles and Process for the Preparation Thereof, U.S. Patent 5,801,207 (Sept. 1, 1998).
9. P.E. Neumann and P.A. Seib, Starch-Based, Biodegradable Packing Filler and Method of Preparing Same. U.S. Patent 5,185,382 (Feb. 9, 1993); U.S. Patent 5,208,267 (May 4, 1993).
10. R.J. Hutchinson, G.D.E. Siodlak, and A.C. Smith, J. Mater. Sci., 22, 3956 (1987).
11. R. Chinnaswamy and M.A. Hanna, Cereal Chem., 65, 138 (1988).
12. S.C. Warburton, A.M. Donald, and A.C. Smith, J. Mater. Sci., 25, 4001 (1990).
13. S.C. Warburton, A.M. Donald, and A.C. Smith, J. Mater. Sci., 27, 1469 (1992).
14. S. Bhatnagar and M. A. Hanna, Trans ASAE, 38, 567-571, (1995).
15. W. Wang, R.A. Flores, and C.T. Huang, Cereal Chem., 72, 38 (1995).
16. J.Y. Cha, D.S. Chung, P.A. Seib, R.A. Flores, and M.A. Hanna, Ind. Crops and Prod., 14, 23 (2001).
17. R.L. Shogren, Carbohydr. Polym., 29(1), 57 (1996).
18. J.L. Willett and R.L. Shogren, Polymer, 43, 5935 (2002).
19. M.S. Winkler, T.S. Berry, and D.E. Kirkpatrick, Process for Preparing Starch and Epoxy-Based Thermoplastic Polymer Compositions, U.S. Patent 6,365,079 (Apr. 2, 2002).
20. C.H. Silvis and J.E. White, Hydroxyl Functionalized Polyetheramines as Barrier Packaging for Oxygen-Sensitive Materials, U.S. Patent 5,275,853 (1986).
21. American Standard Test Methods, "Standard Practice for Conditioning Containers, Packages, or Packaging Components for Testing," in Annual Book of ASTM Standards, D-4332, 15.09, ASTM, Philadelphia, PA (2002).
22. American Standard Test Methods, "Standard Test Methods for Flexible Cellular Materials Made From Olefin Polymers," in Annual Book of ASTM Standards, D-3575, 8.02, ASTM, Philadelphia, PA (2002).
23. P.D. Tatarka and R.L. Cunningham, J. Appl. Polym. Sci., 67, 1157 (1998).
24. American Standard Test Methods, "Standard Test Method for Steady-State Thermal Transmission Properties by Means of the Heat-Flow Meter Apparatus," in Annual Book of ASTM Standards, C-518, 4.06, ASTM, Philadelphia, PA (2002).
25. American Standard Test Methods, "Standard Test Methods for Dynamic Shock Cushioning of Packaging Materials," in Annual Book of ASTM Standards, D-1596, 15.09, ASTM, Philadelphia, PA (2002).
26. American Standard Test Methods, "Standard Test Methods for Transmitted Shock Characteristics of Foam-in-Place Cushioning Materials," in Annual Book of ASTM Standards, D-4168, 15.09, ASTM, Philadelphia, PA (2002).
27. American Standard Test Methods, "Standard Test Methods for Mechanical-Shock Fragility of Products, Using Shock Machines," in Annual Book of ASTM Standards, D-3332, 15.09, ASTM, Philadelphia, PA (2002).
Yogaraj Nabar, Ramani Narayan
Department of Chemical Engineering & Material Science, 2527 E.B., Michigan State University, East Lansing, Michigan 48824
Department of Biochemistry and Molecular Biology, Michigan State University, East Lansing, Michigan 48824
Correspondence to: R. Narayan; e-mail: firstname.lastname@example.org
TABLE 1. Thermal resistivity of foam sheets. Average R/m K-[m.sup.2]/kW Entry Sample ([+ or -]35.2) 1 Starch foam sheet (1.17" thick) 634.0 2 Starch foam sheet (2.08" thick) 616.4 3 Starch foam sheet (3.41" thick) 669.2 4 Polystyrene bead board 634.0 5 Polystyrene composite board 581.1 6 Expanded polystyrene (EPS) 686.8 7 Extruded polystyrene (XEPS) 880.5 8 Sprayed polyurethane foam 1215.1 9 Polyethylene (73[degrees]F) 352.2 10 Polyethylene (23[degrees]F) 422.6 TABLE 2. Compressive strength and resiliency of foam samples. PHAE content (%) Compressive strength (Pa) Resiliency (%) 0.0% 4.36 ([+ or -]0.03)E + 05 69.8 ([+ or -]2.1) 3.0% 3.50 ([+ or -]0.03)E + 05 90.9 ([+ or -]0.7) 5.0% 3.39 ([+ or -]0.01)E + 05 91.9 ([+ or -]0.6) 7.0% 3.12 ([+ or -]0.01)E + 05 93.5 ([+ or -]0.9) 10.0% 3.35 ([+ or -]0.01)E + 05 92.7 ([+ or -]0.9) 15.0% 3.64 ([+ or -]0.05)E + 05 92.9 ([+ or -]0.4) TABLE 3. Normalized steady state weight gains, diameters, lengths of the starch-based foams with different PHAE contents. Normalized Normalized steady state steady state weight gain diameter [[W - [W.sub.0]]/[W.sub.0]] [D/[D.sub.0]] PHAE content (%) ([+ or -]0.024) ([+ or -]0.034) 0.0% 0.128 0.564 3.0% 0.109 0.731 5.0% 0.092 0.780 7.0% 0.078 0.844 10.0% 0.058 0.887 15.0% 0.036 0.936 Normalized steady state length PHAE content (%) [L/[L.sub.0]] ([+ or -]0.038) 0.0% 0.498 3.0% 0.702 5.0% 0.746 7.0% 0.811 10.0% 0.861 15.0% 0.921 TABLE 4. Typical starch foam sheet production table. Starch Water Screw feed injection SME speed rate rate PHAE (KWhr/kg) No. (rpm) (kg/hr) (kg/hr) (kg/hr) ([+ or -]0.007) 1 400 356.1 28.6 27.8 0.135 2 (a) 400 356.1 28.5 27.8 0.129 3 400 356.1 28.9 27.8 0.123 4 450 356.1 28.5 27.8 0.132 5 375 356.1 28.4 27.8 0.129 6 400 338.3 28.8 27.8 0.119 7 400 356.1 39.2 27.8 0.107 8 400 356.1 25.1 27.8 0.136 9 400 356.1 28.5 0.0 0.141 10 400 356.1 28.7 40.1 0.119 Barrel temperature profile No. ([degrees]C) ([degrees]C) Density (kg/[m.sup.3]) 1 96 99 34.6 ([+ or -]0.7) 2 (a) 112 115 28.3 ([+ or -]0.4) 3 121 127 28.4 ([+ or -]1.7) 4 114 118 26.3 ([+ or -]1.3) 5 111 114 29.2 ([+ or -]0.6) 6 112 115 29.7 ([+ or -]0.7) 7 112 113 31.2 ([+ or -]0.4) 8 114 116 33.9 ([+ or -]1.6) 9 113 116 32.1 ([+ or -]1.1) 10 112 115 30.2 ([+ or -]0.2) (a) Regular production run.
|Printer friendly Cite/link Email Feedback|
|Author:||Nabar, Yogaraj; Narayan, Ramani; Schindler, Melvin|
|Publication:||Polymer Engineering and Science|
|Date:||Apr 1, 2006|
|Previous Article:||Manufacturing and foaming of high melt viscosity of polypropylene by using electron beam radiation technology.|
|Next Article:||Intercalation and exfoliation of clay nanoplatelets in epoxy-based nanocomposites: TEM and XRD observations.|