.v- H, :—mwm—o~r~v—wm .u . . w ‘. . . a C rz~pxaxzyq : w—v‘m “-1. '1‘ “am. “W“ 7—...“ “3’1: ' . .r {L 4; F.- "'r‘i‘u . ‘ .3 .r. 2;. ,1:..;....x<. , u , > . 1-11 L . 3 . h x5035? \u, 7 14. v t. 76.. \ \HH 3“”: " C .. . . . ... . a. , Lynuvnv .skymwwun. .C :4 .17 .0... .. 5‘ ,\ II .01» 4 an 1. r T: ingrfiuhx . N. r t ‘1 in: 3A. .x .. O u a... 00...! . u r 'QI. € ‘0 ' . (IN; I‘I‘vi‘. . Ali‘vuha. ‘ :4: b [to I; llll‘ 3.“..90 \ 9.» ‘Io’ll ”Agnmlv. :01» z! .31.}... T 1.4.4. .05-.. .Lr . 1 . . [I n!‘ pg- .lplh'g? InnAHpUD'l-Iullwu’fo: . . w” b éunzln.wunl‘iflflt {I} % J}. 1.1. .'.¢| yr, 01.00"} négfil I w . .fitfii I a. '1‘ 33,-,[gl I\ I3: (1!)... o r1: . . 05:4 I ‘t~v I . i. .I . V0 .1: oomvlallo‘tlhhw.flva.or ..... I . l.‘ !|. ll . .. taxi. . 1.1..th ; . .rdfrflfl. 1:: ll. .vlli Iii-lb: {thou II..‘ III“ Id ‘ l ‘f.’ .r .. » u 0 ln\ ‘ s . I I (Jllf. I‘aorlotls‘.ltf I}. .r . I.wapltvlllf;1f!firs . (I |l| It’ll! Inn“ ll ‘Ilfv I." I .No‘l' P-O-O. P-O-O. + PH -—> P-O-O-H + P. propagation P. + P. -——~ products R. + P. ———> products R. + P-O-O. -——> products termination P. + PAC-O. -——9 products ZP-O-O. -—> 2P-O. + 0-0 c. Random and Specific Site Attack Naturally, specific site attack is expected if macromolecules which posses only a single or a few functional groups are brought in contact with a reagent capable of reacting only with these functional groups. Non-random main-chain scission has been observed with linear homopolymers subjected to mechanical forces: the center portions of the polymer chains are much more likely to undergo main-chain scissions than other parts of’ the macromolecules. Probably the most intriguing problems concerning non- random degradation processes refer to so-called "weak links", which are mostly identical with impurities, incorporated chemically in macromolecules. The polymers that are more likely to exhibit random degradation will be linear homopolymers (Schnabel, 1981). 4. Fracture Phenomena One of the most remarkable features of the fracture or rupture of polymers is the great variety of ways in which different materials respond to stress. The elongation at break varies from less than 1% to several thousand 8 percent ; breaking stresses vary from less than 10 2 10 2 dyne/cm to 2 x 10 dyne/cm : cracks may travel catastro- phically at near sonic velocities, or so slowly that little change can be observed in a day 3 the depth of residual deformation may be measured in centimeters or in microns. To classify different type of fracture one can use the shape of a load-deformation curve as the primary basis and supplement it by observations of deformed and broken specimens. There are 5 distinctive types of behavior for fracture: uniform extension, cold drawing, necking rupture, brittle fracture and necking rupture of the second kind (see Figure 1). In a tensile test it is necessary to choose the shape of the specimen. Different specimen profiles will lead to different results and can not be compared with one another. The specimen profile specified in ASTM D 638 often suffers from slippage and fracture near the clamps. As the specimen is extended in simple tension, the molecules become oriented toward the direction of the applied force. This molecular orientation makes the specimen harder and more difficult to extend, "orientation hardening." When a specimen deforms by cold drawing, the load- extension curve does not immediately represent the general behavior of the material but only the behavior of a specimen of a particular shape. For example, if a 10 cm long test specimen has been extended a further 10 cm, this Load Load Extension Extension (a) Uniform extension (b) Cold drawing Load Load Extension Extension (c) Necking rupture (d) Brittle fracture Load Extension (e) Necking rupture of the second kind Figure 1 : Fracture behavior of polymers (Bikaies, 1971) does not mean that the strain is 100% ; some of the specimen may be strained, for example, 400% and other parts less than 10%. Necking rupture is behavior in whiCh the specimen necks and then breaks without restabilization of the neck. Specimens which fail by necking rupture whiten in the neck. This is usually attributed to the occurrence of very small voids; the scattering of light at the microvoids is the cause of the white appearance. It is not ordinarily easy to demonstrate the existence of these microvoids in a photomicrograph because they are below the limit of resolution. The neck never restabilizes because the specimen breaks before the orientation hardening is sufficient. Necking shows an angle neck because yielding and necking are shear phenomena (Bikales, 1971). 5. Impact Resistance Impact resistance is a property of considerable importance where the use of plastic is concerned. The ability of a fabricated article to withstand shocks is often a decisive factor in replacement of a conventional material by a polymer. The impact resistance of a plastic article depends not only on the basic impact properties of the polymer but also on the following factors: 1) design of the object, 2) conditions during fabrication, 3) the nature of the blow, 4) environmental conditions, and 5) the frequency of the shock. Normally, thick sections produce areas of potential weakness in that the impact resistance is less than in thinner sections. This factor is of considerable importance in the design of blow-molded containers. It is essential to avoid drastic changes of thickness and complex detail in the object. The conditions of fabrication are more frequently responsible than any other factor for impact failure. The impact strength of polyethylene increases as the rate of cooling from melt increases and the size of the spherulites decreases. The impact strength undergoes no change once the polymer has completely solidified. The type of molding affects the impact strength. For example, in compression-molded SBR-modified PS, the stress builds up to a level at which catastrophic failure occurs and the material undergoes a brittle fracture. In an injection-molded sample, failure at the critical stress takes place with a much slower crack propagation and a certain amount of elongation takes place. The injection- molded sample, as a result of this difference in impact behavior, has an impact strength greater than that of the compression-molded material by a factor of at least 2. Factors that can contribute errors to the Izod Impact test are: a) variation of clamping pressure, b) failure to strike the specimens squarely, c) the state of the cutter and the cutting technique for machining notches. 10 B. EFFECTS OF REPROCESSING One of the factors that needs to be considered before making a decision to use recycled HDPE is how much change in the properties of the material will occur due to polymer degradation caused by container fabrication, grinding and contamination from the use cycle. Much research has been done to predict the performance of various kind of recycled plastics. ' Some properties are severely damaged and some properties only slightly changed. Each polymer has its own characteristic changes which depend on the nature of the particular polymer. Polymer degradation is the reaction that causes the reduction in polymer stability. Molecular weight and molecular weight distribution of the polymer have been changed. The question is how much it changes and how we can control it and what kind of detection can be done or will be the most sensitive measurement for each polymer. Schnabel (1981) concluded that there were 6 types of initiation of polymer degradation : thermal, mechanical, chemical, biological, radiation and photooxidation degradations, which were discussed in the preceding section. Rideal and Padget (1976) stated that there was change in the molecular structure of HDPE during processing. They found that high melt temperature (more than 290 C) would result in decreasing melt viscosity and a narrowing of molecular weight distribution, while lower melt temperature 11 resulted in an increase in melt viscosity. The increase in melt viscosity arose from a molecular enlargement reaction which was mainly attributable to the formation of long chain branches. The decrease in melt viscosity and narrowing of MWD were caused by scission of polymer back- bone. The scission and enlargement reaction are not mutually exclusive, but competitive to each other. During processing the shear degradation result was consistent with the theory that in a viscous matrix the action of high shear forces preferentially cuts a polymer chain at the center leading to a narrowing of MWD. Abbas (1980) studied the degradation of polycarbonate during recycling in a capillary rheometer. He found that at high constant shear stresses (0.15-0.95 MPa) and at temperatures between 275°and BZd’C the degradation kinetics were non-random chain scission. He concluded that bonds were more susceptible to scission the closer they were to the middle of the polymer molecule, and that the extent of degradation increased with an increase in molecular weight. Mellor et a1 (1973) stated that the lifetime of polyolefins in the presence of UV light was dependent on the degree of oxidation occurring during the prior processing operation. Luongo (1963) found that the rate of oxygen uptake in solid PE during oxidation was inversely proportional to the percent crystallinity. Only the amorphous regions of 12 semicrystalline polyethylene were sensitive to oxygen attack. He used two polyethylene samples: one was highly crystalline polyethylene of density 0.96 and the other a branched polyethylene of density 0.92. The samples were oxidized by exposing them to oxygen at 100 C under controlled conditions over a period of 400 hours, and were examined at intervals by infrared spectroscopy. Sadrmohaghegh and Scott (1980) studied the effects of reprocessing on low density polyethylene. The results of their experiment showed an increase in melt viscosity, rapid formation of gel, increase in tensile strength and elongation at break during the first 10 minutes of processing and then the beginning of a decline after 20 minutes of processing. They explained that allylic radicals were formed in the polymer by mechanochemical and oxidative reactions during processing that led to cross- 1inking and then later to chain-scission following thermolysis of the hydroperoxide. Mitterhofer (1980) stated that chain scission and cross linking occurred simultaneously during the processing of HDPE. Depending upon conditions of temperature, oxygen availability and polymer type, any one might prevail over the other. The cross-linking of an HDPE was detected by a drastic drop in the melt flow index after 10 minutes of residence time in the melt index apparatus. Scott (1976) suggested that stringent precautions needed to be taken to eliminate the effects of oxygen 13 during the reprocessing operation, otherwise the reprocessed product would be different both chemically and mechanically from that made from virgin material. For consumer waste plastic, the environmental exposure caused further rapid changes in the composition of the polymer which greatly affected its behavior on reprocessing and presented a technical problem in polymer stabilization. Consumer waste plastic, unless effectively cleaned, will contain small amounts of metal ions which are both thermal and UV pro-oxidants. The effect of light on thermally processed polymer is the introduction of unsaturated groups which themselves act as pro-oxidants in a reprocessing operation. In order to control degradation in reprocessed polymers, he suggested using more effective antioxidants which prevent the initiation process, such as metal deactivators, UV absorbers and peroxide decomposers. C. RESEARCH ON PROPERTIES OF POLYMER MATERIALS Properties of polymers are affected by polymer structure. In the same kind of polymer, Molecular Weight (MW) and Molecular Weight Distribution (MWD) will play important roles in the performance of the material. The physical and mechanical characteristics of recycled polymers such as flow behavior, tensile strength, elongation and impact strength are different from and generally inferior to those of virgin polymer. 14 For recycled material it was expected that there would be a reduction in molecular weight and narrowing of MWD caused by polymer degradation. Much research confirming this has been done on various polymer materials (Bevis et al. 1975, Scott 1976, Stamper and Connole 1984, Sayago and Petrie 1985). Perron and Lederman (1972) worked on polyethylene films that had different MWDs. They found that impact strength increased as the MWD was broadened. They explained this behavior by molecular entanglement theory. Broader MWD meant more of the high MW portion that would increase molecular entanglement. Bikales (1971) concluded that as the average MW decreased, there was a tendency for breaking stresses, strains and energies to decrease, but for little or no change in moduli and yield stresses. For HDPE, the material showed more ductility under stress as the MW decreased. Shenoy et al.(1983) proposed the use of MFI information as an indicator of the effect of processing history and suitable end use for a particular polymer material. The MFI value also indicates the average MW of the polymer. Blends of recycled and virgin material cause a reduction in mechanical properties. Normally, a maximum of 20% in-plant recycled materials was used to mix with virgin resin for production without significant differences in / product quality (Abbas et al., 1978). '15 III. MATERIALS AND METHODS A. Material Preparation Collected milk bottles were rinsed with cold water, then dried at room temperature. A lowline granulator model 68-913 from Polymer Machinery Corp. was used to chop the milk bottles into a flaked form. Next contaminants (milk, dirt etc.) were separated by passing the recycled material into an agitated cold water tank, overflowing through a screen, and then leaving them to dry at room temperature. The blends of virgin and recycled HDPE, in weight percent, were physically mixed by a propeller feed mixer for 30 minutes. B. Melt Flow Index Determination Materials and Apparatus: -Virgin HDPE "FORTIFLEX A60-70-119" from Soltex Polymer Corp. -Recycled HDPE milk bottles separated by brand -Regrind unused HDPE bottles from Heatherwood Farms Dairy that were made from pure virgin material which came from the same lot of virgin HDPE stock. -Recycled used HDPE milk bottles from Heatherwood Farms Dairy that came from the same lot as the unused bottles. (Milk bottles in half gallon size from the same lot as the unused bottles, filled by Heatherwood Farms Dairy, were purchased at a local food store. These samples were stored 16 under refrigeration until the expiration date, then were cleaned and chopped into flake form.) -Recycled used HDPE milk bottles mixed from different sources. -Melt Flow Indexer, Ray-Ran model 2A Digital Auto Procedure: ASTM standard 1238 was followed. The Melt Flow Indexer temperature control was set at 190’+/- 0.2’C with die and piston in the cylinder. Warm up usually took about 30 minutes to get a constant temperature. Sample resin (3-3.5 gm) was put in the cylinder with the charging tool. The sample was preheated for 6 minutes in order to allow the sample to completely melt and have a constant flow rate. The 2.16 kg dead load was used and run under the automatic mode that allows the piston to travel for 6.35 mm. The collected data were times in seconds. The MFI value was determined using the formula, 207 MFI (g/10 min) = time(second) where 207 is the factor for calculating the flow rate polyethylene from ASTM D 1238, from Flow rate (427 x L x d)/t weight/t 17 of where: L = length of calibrated piston travel, cm. d = density of resin at test temperature, g/cm3 t = time of piston travel for length L, second 427 = mean of areas of piston and cylinder x 600 (600 = 60 sec/min x 10 min) In this experiment, L = 0.635 cm. 3 a = 0.7636 g/cm (from ASTM D 1238) C. Estimating the change in Molecular Weight Distribution from virgin .HDPE to regrind HDPE by the Melt Flow Index Technique Materials and Apparatus: -Virgin HDPE "FORTIFLEX A60-70-119" of Soltex Polymer Corp. —Recycled HDPE milk bottles -Melt Flow Indexer, Ray-Ran model 2 A Procedure: The Melt Flow Indexer was operated at different shear rate by varying the dead load from 2.16 to 5.0, 7.5, and 10.0 kg. From the Melt Flow Index values, viscosity as a function of shear rate was obtained by, Rn F Shear stress 2 2 x 3.1416 Rp L 18 4 Q Shear rate = 3 3.1416 Rn From ASTM 1238 specifications for piston and die dimensions Rn = nozzle radius 1.0475 mm. Rp = piston radius 4.737 mm. L = nozzle length 8.00 mm. 5 F = test load x 9.807 x 10 Q = flow rate = MEI/(600 x density) From the above information, we can simplify the formula to, 4 Shear stress = 9.11 x 10 x W where W = test load in kg. 1.84 MFI Shear rate = density shear stress Viscosity shear rate 19 D. Tensile Properties Determination Materials and Apparatus -Materials were the same as Melt Flow Index determination, except recycled HDPE milk bottles separated by brand were not used. -Carver Laboratory Press compression molding machine, model M 25 ton. -Instron Tensile tester -Tensilkut cutting machine. Procedure: ASTM standard 638 was followed. Virgin granulated HDPE and recycled HDPE milk bottles were compression molded using a Carver Laboratory Press into 5 inches by 5 inches by 0.1 inch plastic sheets. The condition of molding was heating at 210° C for 6 minutes, then cooling to 50°C within 8 minutes. The plastic sheets were cut into 0.75 inch by 5 inch strips with a band saw, then shaped into dumbbell shape type 1 according to ASTM 638 by a Tensilkut machine. These samples were conditioned at 25°C, 50% RH for 2 days before running the tensile test. An Instron, model no. 1114, was used. An adjustable load cell with scale capacities of 200, 500, 1000, 5000, and 10000 lbs was installed. For testing, the Instron was set at 500 lbs range, 2 in/min. cross head speed, 5 in/min chart speed. A specimen was placed between the grips with 20 abrasive paper to prevent slippage. Five samples were tested and results from the chart recorder were analysed to calculate tensile strength, modulus of elasticity, elongation at yield and elongation at break. Sample mixtures of virgin material and recycled material in weight percents of 10%, 20%, 50%, and 80% were prepared and tested in the same manner. E. Impact Strength Determination Materials and Apparatus -Materials are the same as for the tensile properties determination. -Impact Tester from TMI, model 43-02, monitor/impact -Notching machine from TMI -Carver Laboratory Press, model M 25 ton -Band saw Procedure: ASTM standard 256-81 was followed. Samples were prepared by compression molding virgin pellets, chopped recycled material, or mixtures of recycled material and virgin material in proportions of 10%, 20%, 50%, and 80% by weight of recycled materials. The molding condition was 210° C, 25000 lbs for 6 minutes, then cooling to Sd’C within 8 minutes, producing a sheet 5 inches by 5 inches by 0.125 inches. The molded sheets were conditioned at 24°C for 2 21 days, and then cut into test samples 2.5 inches long and 0.5 inches wide. They were notched with a 0.1 inch deep cut. In the testing procedure the thickness of the original sheet is the width of the specimens. Eight samples were cut from each molded sheet. All tests were performed at 24°C. F. Data Analysis Procedure The software program M-Stat written by Scott P. Eisensmith Land Ken W. Rorick, Crop and Soil Science Department, Michigan State University, was used with an IBM-PC to analyse all data. 22 IV. RESULTS AND DISCUSSION A. Melt Flow Index Value (MFI) 1. Results: The average MFIs for 3 samples of virgin HDPE were 0.578, 0.592 and 0.727 g/10 min. These average values were obtained from at least 3 replications. The recycled HDPE materials were collected from different sources over time, and were separated by brand. (Note: the term ”recycled" will always mean post-consumer materials.) The MFI values ranged from 0.635 to 0.750 g/10 min. (see Table 1). Table 2 shows the MFI for those materials received from Heatherwood Farms, a local dairy producer. The MFI of virgin resins, blown but unused bottles from pure virgin materials, and post-consumer bottles that came from the same lot as the unused bottles are 0.727, 0.753, and 0.715 g/10 min. respectively. Table 3 shows MFI values for virgin material, recycled material, and mixtures between recycled and virgin HDPE. 2. Discussion: From the variation in the MFI value of recycled milk bottles, we can see that there is a range of flow properties in the materials that are used for producing milk bottles. This observation was confirmed by the difference in MFI for the first and second lots of virgin HDPE received from Heatherwood Farms. The mixtures of virgin and recycled HDPE did not show 23 any significant differences in MFI, and there was also no significant change in MFI between the blown but unused bottles and the post-consumer bottles, compared with virgin material from the same lot. This may be due to cross- linking between molecules balancing the molecular breakdown caused by mechanical and thermal degradation of the polymer by shear forces and elevated temperatures during container fabrication and grinding into flake form. As was discussed earlier, both these reactions are known to occur during processing of HDPE (Rideal and Padget 1976, Mitterhofer 1980). MFIs of virgin resin in Table 2 and 3 were different, because they were tested at different times. The value in Table 2 was obtained about 5 months before the experiment . for Table 3 was conducted. The aging of the polymer over that time period evidently resulted in a lower MFI value. 24 TABLE 1 MELT FLOW INDEX OF HDPE MILK BOTTLES IN LANSING MICHIGAN (g/10 min.) Material Mean SD. Replication Virgin resin (Heatherwood) lot 1 0.584 0.020 7 lot 2 0.727 0.021 5 Recycled bottles (Heatherwood) lot 1 0.726 0.011 4 lot 2 0.695 0.034 4 lot 3 0.635 0.094 8 lot 4 0.731 0.012 5 Recycled bottles (Meijer) lot 1 0.700 0.012 7 lot 2 0.701 0.011 4 Recycled bottles (Country Fresh)0.699 0.004 4 Recycled bottles (Sta-Fresh) 0.744 0.008 4 Recycled bottles (Springdale) 0.750 0.005 5 25 TABLE 2 EFFECTS OF CONTAINER FABRICATION & REGRINDING AND OF USE CYCLE ON MFI VALUE Material Mean SD. Replications (g/10 min.) Virgin Resin 0.727 0.021 5 Regrind unused bOttles 0.753 0.012 5 (in-plant) Recycled post-consumer 0.715 0.012 5 bottles TABLE 3 MFI OF MIXTURES OF VIRGIN AND RECYCLED HDPE MILK BOTTLES Material Mean SD. Replications Virgin resin 0.691 0.008 5 10% Recycle 0.686 0.013 5 20% Recycle 0.696 0.011 5 50% Recycle 0.685 0.003 5 80% Recycle 0.688 0.014 5 100% Recycle 0.681 0.004 5 '26 B. Estimation of changes in the Molecular Weight distribution 1. Results: ' The MFI values increased as the loads increased, as expected (see Table 4). The MFI changes were from 0.260 to 12.577 g/10 min. as the loads varied from 1.1 kg to 10 kg. The viscosity of virgin and recycled HDPE varied with the shear rate by about the same magnitude (see Table 5). 2. Discussion: Polyethylene is a shear sensitive material. Graessley (1984) stated that a lower molecular weight for a polymer would result in a higher MFI, and that a narrower MWD would result in lower viscosity at low shear rate (less than 0.1 sec-1) and a higher viscosity at high shear rate (more than 1.0 sec-1). In another words, broad MWD resins are more shear sensitive than narrow MWD resins (see Figure 2). For this experiment, we investigated the low shear rate range (0.005 to 0.2 sec-1). If there were changes resulting in a lower MW and narrower MWD, we expected lower viscosity at the same shear rate. From Table 5, an increase in shear rate resulted in a decrease in the melt viscosity of HDPE, but no significant differences in molecular weight and molecular weight distribution between virgin and recycled HDPE were demonstrated, as the curves of viscosity vs. shear rate of virgin and recycled HDPE were on almost the same line (see Figure 3). 27 Again, the masking of chain scission by cross-linking may be occurring. Viscosity o broad MWD A narrow MWD Shear rate Figure 2 : The effect of MWD on viscosity & shear rate (Graessley, 1984) 28 TABLE 4 MELT FLOW INDEX VS. LOAD Load MFI (g/10 min.) (kg) Virgin Recycle 1.1 0.260 0.267 2.16 0.727 - 0.704 3.8 1.925 _ 1.949 5.0 3.044 3.198 10.0 12.577 11.504 TABLE 5 EFFECTS OF SHEAR RATE ON VISCOSITY Virgin Recycle Load Shear rate Viscosity Shear rate Viscosity (kg) (1/sec) (g/cm sec) (1/sec) (g/cm sec) 1.1 0.0049 2.03 E+7 0.0051 1.969 E+7 2.16 0.0138 1.43 E+7 0.0134 1.472 E+7 3.8 0.0367 9.45 E+6 0.0390 8.9 E+6 5.0 0.0580 7.87 E+6 0.0610 7.49 E+6 10.0 0.2397 3.81 E+6 0.2193 4.163 E+6 29 Log of Viscosity (g/cm sec) 7.5 W 0 Virgin HDPE . Recycled HDPE q .1 6 T I I T -2.5 l ) Log of Shear rate (sec- Figure 3: Estimation of changes in MWD 30 C. Tensile Properties 1. Results: From Tables 6, 7, 8, and 9, we can see that the tensile strength, modulus of elasticity and elongation at yield do not significantly change from virgin material to blends to 100% recycled HDPE. Only elongation at break shows significant changes. In other words, the critical property that can be detected is elongation at complete break. The tensile strength averages about 4,900 psi. The elongation at yield averages about 17% and the modulus of elasticity averages about 93,000 psi. 2. Discussion: The assumption of polymer degradation in recycled milk bottles involving both polymer main—chain scission and cross-linking occurring at the same time can again be used to explain this behavior. These two reactions appear to balance each other, resulting in maintenance of the tensile strength, modulus of elasticity, and elongation at yield. The elongation at break is more complex. From Appendix B, the typical graphs of the tensile tests demonstrate the differences in fracture behavior of the polymer. The 80% and 100% recycled HDPE in the mixtures with virgin resin appear to be "necking rupture of the second kind", while 31 the 0, 10, 20, and 50% recycled HDPE in the mixtures are "necking rupture" according to the classifications discussed by Bikales (1971) and shown in Figure 1. TABLE 6 TENSILE STRENGTH (psi) Material Run 1 Run 2 Mean SD. Mean SD. Virgin HDPE 4890 158 4780 56 10% Recycled 4830 62 4960 156 20% Recycled 4800 51 4930 213 50% Recycled 4900 100 4990 243 80% Recycled 4960 175 5080 183 100% Recycled 4960 143 5020 252 TABLE 7 MODULUS OF ELASTICITY (psi) Material Run 1 Run 2 Mean . SD. Mean SD. Virgin HDPE 86500 12800 83800 5080 10% Recycled 87000 6810 88600 11700 20% Recycled 99900 20300 84400 13700 50% Recycled 97400 17800 92000 4550 80% Recycled 98700 23800 102000 18100 100% Recycled 92800 14700 98600 9910 32 TABLE 8 ELONGATION AT YIELD (%) Material Run 1 Run 2 Mean SD. Mean SD. Virgin HDPE 17 1.22 17.3 0.98 10% Recycled 18 0.93 17.2 0.67 20% Recycled 16.4 1.44 17.6 0.76 50% Recycled 17 1.7 17.4 0.86 30% Recycled 17.5 0.79 17.2 0.66 100% Recycled 16.2 1.68 17.4 0.33 TABLE 9 ELONGATION AT COMPLETE BREAK (%) Material Run 1 Run 2 Mean SD. Mean SD. Virgin HDPE 69.7 16.5 74.0 17.5 10% Recycled 62.7 10.1 62.4 6.54 20% Recycled 47.2 8.77 51.3 12.5 50% Recycled 48.9 18.7 41.4 19.4 80% Recycled 35.1 9.24 .34.6 9.44 100% Recycled 36.9 18.2 30.7 4.74 33 Tensile strength (psi) ‘ C) run 1 I'run 2 I I I I T r I % Recycled HDPE Figure 4: Tensile strength comparisons 34 GM. 15 Modulus of elasticity (psi x 104) O 100 % Recycled HDPE Figure 5: Modulus of elasticity comparisons 35 % Elongation at yield 20 1 0 run 1 0 run 2 <0 0 o e. _ II A. O ' O O 15q 10 3.1 I I I I I I I T I I 0 100 Z Recycled HDPE Figure 6: Percent elongation at yield comparisons 36 Z Elongation at break 100- 0 run 1 0 run 2 0 100 Z Recycled HDPE Figure 7: Percent elongation at break comparisons 37 D. Impact Strength 1. Results: Izod impact strength comparisons of virgin material, unused bottles and post-consumer bottles known to be from the same lot of HDPE show significant changes, as can be seen in Table 10. All failures were classified as partial break as defined in ASTM D 256. The reduction in strength compared to virgin resin is about 16% for unused bottles and about 36% for used bottles. For recycled blends, in Table 11, the Impact strength started to drop at about 20% recycled HDPE in the blend. In this case the recycled HDPE has an impact strength about 13% lower than the virgin resin. 2. Discussion: It has been demonstrated that Izod impact strength is one of the sensitive properties that can be used to detect changes in polymer materials. Several researchers did similar kinds of work and concluded that a lower molecular weight and narrower MWD resulted in decreasing impact strength. Perron and Lederman (1972) explained that the higher molecular weight material has longer chains, leading to more molecular entanglement, thus requiring higher energy to break the material. The broader MWD has more of the high molecular weight end that will dominate the resistance to impact force. The impact strength values of virgin HDPE in Tables 10 38 and 11 are different due to differences in experiment time periods and variation in sample preparation. TABLE 10 EFFECT OF USE CYCLE ON IZOD IMPACT STRENGTH (ft lb/in) Material Mean SD. Used bottles 1.883 0.243 1.720 0.141 Unused bottles 2.356 0.086 2.364 0.170 Virgin HDPE 2.690 0.232 2.913 0.260 TABLE 11 IZOD IMPACT STRENGTH OF MIXTURES OF VIRGIN AND RECYCLED HDPE (ft lb/in) Material Mean SD. Virgin HDPE 2.522 30.16 10% recycled 2.693 0.261 16.7% recycled 2.608 0.156 50% recycled 2.409 0.138 80% recycled 2.231 0.238 100% recycled 2.201 0.144 39 [] run 2 *** *.*1.* .w* *i.*.** .** *1.*.** * *1.* *.*1.* *_*1.* d 2 AGH\QH umV nuwsouum uommafi couH Virgin HDPE bottles Unused Used bottles Figure 8: Effect of used cycle on Izod impact strength. 40 Izod impact strength (ft lb/in) o o 3? o o o 2L 1 1 I r r 1 11 r11 0 100 Z Recycled HDPE Figure 9: Izod impact strength comparisons 41 E. DATA ANALYSIS To evaluate the significance of changes in properties between virgin, recycled and blended materials, we used Analysis of Variance (ANOVA) at the 95% and 99% confidence levels. Tensile strength, Elongation at yield, and Modulus of elasticity of all treatments showed no significant differences.‘ Elongation at break and Izod impact strength did show significant differences between treatments both at the 95% and 99% confidence levels. Results are summarized in Table 12. 42 TABLE 12 ANALYSIS OF VARIANCE FOR MIXTURES OF VIRGIN AND RECYCLED HDPE Variable d.f. F-value Prob. significance level 0.05 0.01 Tensile strength run 1 5 1.23 .329 ns. ns. run 2 5 1.33 .285 ns. ns. Elongation at break run 1 5 4.57 .004 * ** run 2 5 12.79 .000 * ** Elongation at yield run 1 5 1.23 .326 ns.. ns. run 2 5 0.25 - ns. ns. Modulus of elasticity run 1 5 1.72 .170 ns. ns. run 2 5 1.85 .143 ns. ns. Izod impact strength 5 9.00 .000 * ** 43 V. SUMMARY AND CONCLUSIONS Recycled HDPE milk bottles have been changed in some mechanical and physical properties due to polymer degradation and contamination. The Elongation at break and Izod impact strength are the most sensitive properties to detect the inferiority of recycled materials compared to virgin materials. Moreover, both properties have similar behaviors for mixtures of virgin and recycled HDPE. These observations may be useful for predicting one property from another. Other properties such as MFI, tensile strength, elongation at yield and modulus of elasticity showed little or no change. The estimation of changes in MWD by the MFI technique was not effective in this case, as we could not detect differences between virgin and recycled materials, although the mechanical properties illustrated that there were changes in the polymer. VI. SUGGESTIONS FOR FURTHER WORK The suggestions about MW and MWD changes that are provided in this paper should be confirmed by gel permeation chromatography which is not presently available in the School of Packaging, Michigan State University. Other properties that should be studied are stress crack resistance, material lifetime (aging, weathering, etc.) and brittleness temperature. 44 APPENDIX A DATA AND ANALYSIS OF VARIANCE DO YOU WANT VAR 10 11 TYPE numeric numeric numeric numeric text 4 numeric numeric numeric numeric numeric numeric A l 0 V A - 1 one way A1071 rev. 10/10/85 ISTAT Version 4.00/8! Revised by Scott P. Eisensmith A LIST OF THE VARIABLES DISPLAYED ON THE SCREEI? (Y or 1) LIST OF VARIABLES [AER/DESCRIPTIOI replications treatment Tensile strength run 1 Tensile strength run 2 sample description Zelongation at break run 1 z elongation at yield run 1 moludus of elasticity run 1 modulus of elasticity run 2 %elongation at break run 2 1 elongation at yield run 2 PRESS 1mm 70 00311101: 45 r— ISTAT DATEITRY 5.0 (C) 1986 Iichigan State University I Choose specific variables to be edited. I Blankcase Define lewtxt Variables Goto L_ Case 1 repl 2 tree 3 Tensile 4 Tensile 5 sample 6 Selenga 7 z elong 8 moludus 1 1.0 1.0 5163.0 4817.0 0%re 84.0 17.5 106707.0 2 2.0 1.0 4827.0 4756.0 ” 59.5 17.5 109756.0 3 3.0 1.0 4756.0 4695.0 ” 57.5 16.0 109756.0 4 4.0 1.0 4837.0 4817.0 ” 56.5 15.5 106707.0 5 5.0 1.0 4878.0 4827.0 ” 91.0 18.5 101626.0 6 1.0 2.0 4756.0 4817.0 10%r 61.3 17.0 102642.0 7 2.0 2.0 4776.0 5163.0 ” 61.8 17.5 99085.0 8 3.0 2.0 4878.0 5081.0 ” 52.5‘ 19.0 101626.0 9 4.0 2.0 4878.0 4898.0 ” 58.5 19.0 105014.0 10 5.0 2.0 4878.0 4827.0 ” 79.5 17.5 101626.0 11 1.0 3.0 4827.0 5203.0 20%r 38.8 17.5 96545.0 12 2.0 3.0 4827.0 4888.0 ” 40.7 17.5 105014.0 13 3.0 3.0 4827.0 4675.0 " 53.5 16.0 101626.0 14 4.0 3.0 4726.0 5081.0 ” 56.0 14.5 111789.0 15 5.0 3.0 4797.0 ” 16 1.0 4.0 4868.0 4980.0 502r 31.8 17.5 101626.0 17 2.0 4.0 5061.0 4959.0 ” 40.3 14.0 108401.0 18 3.0 4.0 4787.0 5386.0 ” 73.5 18.0 98238.0 19 4.0 4.0 4898.0 4898.0 ” 35.0 17.5 104675.0 20 5.0 4.0 4878.0 4726.0 ” 64.0 18.0 106707.0 46 F' ISTAT DATEITRY 5.0 (C) 1986 Michigan State University I Goto specific variable and case. I Blankcase Define lewtxt Variables Gate 1 Case 1 repl 2 tree 3 Tensile 4 Tensile 5 sample 6 telonga 7 2 along 8 moludus 21 22 23 24 25 26 27 20 29 30 1. 2.0 3. 5.0 5.0 5.0 5.0 5.0 6.0 6.0 6.0 6.0 6.0 4878. 5264. 4939. 4827. 4888. 4980. 5203. 4898. 4898. 4837. 47 5300.0 5051.0 4900.0 4900.0 5071.0 4900.0 5437.0 4040.0 4797.0 5020.0 80%r 24.5 44.5 39.5 25.8 41.0 40.5 34.1 64.0 13.5 32.5 16.5 17.5 18.0 17.0 18.5 18.0 16.0 16.5 13.5 17.0 108401. 112466. 108401. 101626. 111789. 105014. 104675. 108401. 101626. 111789. r— HSTAT DATEITRY 5.0 (C) 1986 Iichigan State University I Choose specific variables to be edited. I Blankcase Define lewtxt Variables Goto L Case 5 sample 9 modulus 10 lelong 11 2 elon 1 0%re 70219.0 07.0 17.0 2 7 00302.0 05.2 10.0 3 7 00302.0 72.0 15.0 4 7 00302.0 110.0 17.0 5 n 0 10%: 01301.0 54.2 17.4 7 7 70219.0 02.0 17.0 0 7 91403.0 00.4 17.0 9 7 100707.0 72.4 17.2 10 7 07390.0 02.4 10.0 11 20%r 70219.0 50.0 10.4 . 12 7 101020.0 35.0 17.4 13 7 71130.0 55.0 10.4 14 7 90545.0 41.4 17.0 15 7 70219.0 00.0 10.0 10 50:: 91403.0 22.0 10.0 17 7 90545.0 35.2 10.0 10 7 90545.0 01.0 10.0 19 7 00923.0 20.0 17.0 20 7 00302.0 02.4 17.0 48 r— HSTAT DATEFTRY 5.0 (C) 1986 lichigan State University I Goto specific variable and case. I Blankcase Define lewtxt Variables Gate 1 Case 5 sample 9 modulus 10 zelong 11 z elon 21 80%r 22 ” 101626.0 22.0 16.4 23 " 86382.0 43.2 16.8 24 " 91463.0 32.8 17.6 25 ” 127032.0 40.2 17.8 26 100% 101626.0 29.2 17.8 27 " 111789.0 36.0 17.6 28 ” 86382.0 26.8 17.2 29 " 91463.0 35.4 17.0 30 " 101626.0 26.0 17.2 49 Data file 111313}? Title: lechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Vithout selection One way ANOVA grouped over variable 2 treatment with values from 1 to 6 Variable 3 Tensile strength run 1 A H A L Y S I S O F V A R I A N C E T A B L E Degrees of Sun of Error Freedom Squares Kean Square F-value Prob. Between 5 97829.8845 19565.98 1.23 .329 Vithin 23 367141.1500 15962.66 Total 28 464971.0345 Coefficient of Variation= 50 2.58% Var. V A R I A B L E No. 3 2 lumber Sum Average SD SE 1 5.00 24461.000 4892.20 157.63 56.50 2 5.00 24166.000 4833.20 61.75 56.50 3 4.00 19207.000 4801.75 50.50 63.17 4 5.00 24492.000 4898.40 100.24 56.50 5 5.00 24796.000 4959.20 174.96 56.50 6 5.00 24816.000 4963.20 143.37 56.50 Total 29.00 141938.000 4894.41 128.86 23.93 Vithin 126.34 Bartlett’s Test Chi-square = 6.979116 Number of Degrees of Freedom Approximate Significance = 51 = 5 .2221 Data file HDPE Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Vithout selection One way AFOVA grouped over variable 2 treatment with values from 1 to 6 Variable 4 Tensile strength run 2 A N A L Y S I S O F V A R I A H C E T A B L E Degrees of Sum of Error Freedom Squares Mean Square F-value Prob. Between 5 253971.366? 50794.27 1.33 .285 Within 24 917225.6000 38217.73 Total 29 1171196.9667 Coefficient of Variation= 3.942 52 Var. V A R I A B L E No. 4 2 Number Sum Average SD SE 1 5.00 23912.000 4702.40 50.39 07.43 2 5.00 24700.000 4957.20 150.36 07.43 3 5.00 24044.000 4920.00 213.19 07.43 4 5.00 24949.000 4909.00 242.92 07.43 5 5.00 25390 000 5079.20 103.12 07.43 0 5.00 25002.000 5010.40 252.30 07.43 16:61 30.00 140769.000 4950.97 200.90 36.09 Vithin 195.49 Bartlett's Test Chi-square = 7.092797 Number of Degrees of Freedom = 5 Approximate Significance = 53 .2138 Data file HDPE Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Without selection One way AlOVA grouped over variable 2 treatment with values from 1 to 6 Variable 6 zelongation at break run 1 A N A L Y S I S O F V A R I A I C E T A B L E Degrees of Sun of Error Freedom Squares Mean Square F-value Prob. Between 5 4753.810? 950.76 4.57 .004 Within 23 4780.1139 207.83 Total 28 9533.9246 Coefficient of Variation= 28.732 54 Var. V A R I A B L E lo. 6 2 Number Sum Average SD SE 1 5.00 348.500 69.70 16.47 6.45 2 5.00 313.500 62.70 10.09 6.45 3 4.00 188.950 47.24 8.77 7.21 4 5.00 244.550 48.91 18.67 6.45 5 5.00 175.300 35.06 9.24 6.45 6 5.00 184.600 36.92 18.18 6.45 Total 29.00 1455.400 50.19 18.45 3.43 Vithin 14.42 Bartlett’s Test Chi-square = 3.999969 Number of Degrees of Freedom = Approximate Significance = 55 5 .5494 Data file 11131:]; Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Vithout selection One way AIOVA grouped over variable 2 treatment with values from 1 to 6 Variable 7 z elongation at yield run 1 A F A L Y S I S O F V A R I A H C E T A B L E Degrees of Sun of Error Freedom Squares Kean Square F-value Prob. Between 5 10.9780 2.20 1.23 .326 Within 23 40.9875 1.78 Total 28 51.9655 Coefficient of Variation= 7.84% 56 Var. V A R I A B L B IO. 2 Number Sum Average 1 5.00 85.000 17.00 2 5.00 90.000 18.00 3 4.00 65.500 16.38 4 5.00 85.000 17.00 5 5.00 87.500 17.50 6 5.00 81.000 16.20 Total 29.00 494.000 17.03 Vithin Bartlett's Test Chi-square 3.202316 Number of Degrees of Freedom = 5 Approximate Significance = .6688 57 Data file 111)}ng Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Without selection One way AFOVA grouped over variable 2 treatment with values from 1 to 6 Variable 8 moludus of elasticity run 1 A R A L Y S I S O F V A R I A F C E T A B L E Degrees of Sun of Error Freedom Squares Mean Square F-value Prob. Between 5 143930416.0621 28786083.21 1.72 .170 Vithin 23 385581625.8000 16764418.51 Total 28 529512041.8621 Coefficient of Variation= 58 3.89% Var. V A R 1 A B L E No. 8 2 lumber Sum Average SD SE 1 5.00 534552.000 106910.40 3324.25 1831.09 2 5.00 509993.000 101998.60 2137.37 1831.09 3 4.00 414974.000 103743.50 6393.92 2047.22 4 5.00 519647.000 103929.40 4060.91 1831.09 5 5.00 542683.000 108536.60 4295.68 1831.09 6 5.00 531505.000 106301.00 3895.00 1831.09 Total 29.00 73053354.000 105288.07 4348.69 807.53 Within 4094.44 Bartlett's Test Chi-square = 3.876682 Number of Degrees of Freedom = 5 Approximate Significance = .5673 59 Data file HDPE Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Without selection One way ABOVA grouped over variable 2 treatment with values from 1 to 6 Variable 9 modulus of elasticity run 2 A N A L Y S I S O F V A R I A F C E T A B L E Degrees of Sum of Error Freedom Squares lean Square F-value Prob. Between 5 1193258662.2071 238651732.44 1.85 .143 Within 22 2831730373.9000 Total 27 4024989036.1071 128715017.00 Coefficient of Variation= 12.41% 60 Var. V A R I A B L E lo. 9 2 Number Sum Average SD SE 1 4.00 335365.000 83841.25 5081.50 5672.63 2 5.00 443088.000 88617.60 11662.63 5073.76 3 5.00 421747.000 84349.40 13729.18 5073.76 4 5.00 459858.000 91971.60 4545.00 5073.76 5 4.00 406503.000 101625.75 18084.35 5672.63 6 5.00 492886.000 98577.20 9905.51 5073.76 Total 28.00 12559447.000 91408.82 12209.57 2307.39 within 11345.26 Bartlett's Test Chi-square = 7.832493 Number of Degrees of Freedom = 5 Approximate Significance = .1657 61 Data file HDPE Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Vithout selection One way AIOVA grouped over variable 2 treatment with values from 1 to 6 Variable 10 telongation at break run 2 A R A L Y S I S O F V A R I A H C E T A B L E Degrees of Sum of Error Freedom Squares Mean Square F-value Prob. Between 5 9892.7489 1978.55 12.79 .000 Within 22 3404.5081 154.75 Total 27 13297.2570 Coefficient of Variation= 24.502 62 Var. V A R I A B L E lo. 10 2 Number Sum Average SD SE 1 4.00 354.200 88.55 15.79 .22 2 5.00 312.000 62.40 6.54 .56 3 5.00 256.600 51.32 12.53 .56 4 5.00 207.200 41.44 19.37 .56 5 4.00 138.200 34.55 9.44 .22 6 5.00 153.400 30.68 4.74 .56 Total 28.00 1421.600 50.77 22.19 .19 Vithin 12.44 Bartlett’s Test Chi-square = 8.567118 lumber of Degrees of Freedom = Approximate Significance = 63 .1276 5 Data file 11131313 Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 30 Without selection One way AFOVA grouped over variable 2 treatment with values from 1 to 6 Variable 11 % elongation at yield run 2 A H A L Y S I S 0 F V A R I A H C E T A B L E Degrees of Sun of Error Freedom Squares Mean Square F-value Prob. Between 5 0.6711 0.13 0.25 Within 22 11.7560 0.53 Total 27 12.4271 Coefficient of Variation= 4.22% 64 Var. V A R I A B L E lo. 11 2 number Sum Average SD 1 4.00 69.000 17.25 .98 .37 2 5.00 85.800 17.16 .67 .33 3 5.00 88.000 17.60 .76 .33 4 5.00 00.000 17.30 .00 .33 5 4.00 68.600 17.15 .66 .37 6 5.00 86.800 17.36 .33 .33 Total 28.00 485.000 17.32 .68 .13 Vithin .73 Bartlett’s Test Chi-square = 3.830831 Number of Degrees of Freedom = 5 Approximate Significance = 65 .574 A l O V A - 1 one way AIOVA rev. 10/10/85 MSTAT Version 4.00/EM Revised by Scott P. Eisensmith DO YOU VAIT A LIST OF THE VARIABLES DISPLAYED OF THE SCREEF? (Y or M) y LIST OF VARIABLES VAR TYPE FAME/DESCRIPTION 1 numeric replications 2 numeric treatments 3 text 4 sample discriptions 4 numeric Izod impact strength PRESS RETURN TO CONTINUE 66 r— MSTAT DATEITRY 5.0 (C) 1986 Michigan State University I File command menu I File Edit Quit 1 Case 1 r 2 t 3 sample 4 120 1 1.0 1.0 0%re 2.663 2 2.0 1.0 7 2.094 3 3.0 1.0 7 2.555 4 4.0 1.0 7 2.019 5 5.0 1.0 7 2.500 6 0.0 1.0 7 2.444 7 7.0 1.0 7 2.399 0 0.0 1.0 7 2.217 9 1.0 2.0 107: 2.513 10 2.0 2.0 7 3.000 11 3.0 2.0 7 2.030 12 4.0 2.0 7 2.041 13 5.0 2.0 7 2.705 14 6.0 2.0 7 2.204 15 7.0 2.0 7 2.440 16 0.0 2.0 7 2.935 17 1.0 3.0 16.7 2.420 10 2.0 3.0 7 2.553 19 3.0 3.0 7 2.702 20 4.0 3.0 7 2.595 67 r— MSTAT DATEITRY 5.0 (C) 1986 Michigan State University Editin command menu. | 8 I File Edit Quit P Case 1 r 2 t 3 sample 4 120 21 5.0 3.0 ” 2.608 22 6.0 3.0 ” 2.597 23 7.0 3.0 ” 2.464 24 8.0 3.0 ” 2.927 25 1.0 4.0 50%r 2.491 26 2.0 4.0 ” 2.337 27 3.0 4.0 ” 2.132 28 4.0 4.0 ” 2.582 29 5.0 4.0 ” 2.499 30 6.0 4.0 " 2.359 31 7.0 4.0 ” 2.469 32 8.0 4.0 " 2:400 33 1.0 5.0 80%r 2.555 34 2.0 5.0 ” 1.877 35 3.0 5.0 " 2.138 36 4.0 5.0 " 2.243 37 5.0 5.0 ” 2.468 38 6.0 5.0 ” 2.267 39 7.0 5.0 ” 1.942 40 8.0 5.0 " 2.359 68 r— MSTAT DATEITRY 5.0 (C) 1986 Michigan State University I Editing command menu. I File Edit Quit L Case 1 r 2 t 3 sample 4 120 41 1.0 6.0 100% 2.214 42 2.0 6.0 ” 2.201 43 3.0 6.0 ” 2.178 44 4.0 6.0 ” 1.902 45 5.0 6.0 ” 2.286 46 6.0 6.0 " 2.379 47 7.0 6.0 " 2.310 48 8.0 6.0 ” 2.138 69 muffle HDPE—2 Title: Mechanical evaluation of HDPE Function: AlOVA-l Data case no. 1 to 48 Vithout selection One way AlOVA grouped over variable 2 treatments with values from 1 to 6 Variable 4 Izod impact strength A F A L Y S I S O F V A R I A H C E T A B L E Degrees of Sum of Error Freedom Squares Mean Square F—value Prob. Between 5 1.6069 0.32 9.00 .000 Vithin 42 1.4992 0.04 Total 47 3.1061 Coefficient of Variation= 7.73% 70 Bartlett’s Test Chi-square = 5.245091 lumber of Degrees of Freedom = 5 Approximate Significance = .3867 71 Var. V A R I A B L E lo. 4 2 Number Sum Average SD 1 8.00 20.179 2.52 .16 .07 2 8.00 21.546 2.69 .26 .07 3 8.00 20.866 2.61 .16 .07 4 ' 0:00 19.209 2.41 .14 .07 5 8.00 17.849 2.23 .24 .07 6 8.00 17.608 2.20 .14 .07 Total 48.00 117.317 2.44 .26 .04 ‘Vithin .19 APPENDIX B TYPICAL GRAPHS AND APPEARANCES OF SPECIMENS 0F TENSILE TEST .mpcman 050 c« mmom poaomoom SON 0H .o mo ummu oHHmsoH mo mnemuw Hmowmxh "0H madman 72 .mecmen 0cm as man: camosoan Nooa can .ow .om mo omou mammcoH mo msomuw HmoflamH ”Ha owowflm 73 ‘1: RECYCLED HDPE Figure 12: Appearances of specimens after Tensile test 74 BI BLIOGRAPHY LIST OF REFERENCES Abbas, K.B., 1980. Degradation of Polycarbonate During Recycling in a Capilllary Rheometer, Polymer Engineering and Science, 20(10):703-707. ASTM D 256-81, 1986. Impact Resistance of Plastics and Electrical Insulating Materials, Annual Book 92 ASTM Standards, Philadelphia, Pa., 8:99-120. ASTM D 638-82a, 1986. Tensile Properties of Plastics, Annual Book 9f ASTM Standards, Philadelphia, Pa., 8:231-247. ASTM D 1238-82, 1986. Flow Rates of Thermoplastics by Extrusion Plastometer, Annual Book 9; ASTM Standards, Philadelphia, Pa., 8:569-581. Bevis, M.; Owen, T.W.; Skallam, D. 1975. Recycling contaminated polyethylene, Polymer Age, 6(1/2):27,28,31. Bikales, N.M., 1971. Mechanical Properties 9; Polymer, Encyclopedia Reprint, Wiley-Interscience, New York, 268 pp. Luongo, J.P., 1963. Effect of oxidation on polyethylene morphology, Journal 9; Polymer Science: Part B, 1(3)141-143. Graessley, W.W. 1984. Viscoelasticity and Flow in Polymer Melts and Concentrated Solutions, in Physical Properties 9f Polymers by Mark, J.E.: Eisenberg, A,; Graessley, W.W.: Mandelkern, L.: and Koenig, J.L., American Chemical Society, Washington, D.C., 246 pp. Mellor, D.C.; Moir, A.B.; and Scott, G. 1973. The effect of processing condition on the UV stability of polyolefins, European Polymer Journal, 9(3):219-225. _ Mitterhofer, F. 1980. Processing Stability of Polyolefins, Polymer Engineering and Science, 20(10):692-695. Perron, P.J. and Lederman, P.B. 1972. The effect of molecular weight distribution on polyethylene film properties, Polymer Engineering and Science, 12(5):340-345. Rideal, G.R. and Padget, J.C. 1976. The Thermal-Mechanical Degradation of High Density Polyethylene, g; 9; Polymer Science: Polymer Symposium No.57, p.1-15. 75 Sadrmohaghegh, C. and Scott, G. 1980. The effect of reprocessing on polymer-1 low density polyethylene, European Polymer Journal, 16(11):1037-1042. Sayago, J.H. and Petrie, S.P. 1980. The Recycling of Low Density Polyethylene Film, Antec'85, p.96-99. Schnabel, W. 1981. Polymer Degradation Principles and Practical Applications, Hanser International, New York, 227 pp. Scott, G. 1976. Some chemical problem in the recycling of plastics, Resource Recovery and Conservation, 1:381-395. Selke, S.E.; Lai, C.C.; Johnson, D.: Yam, K.: Grulke, E.; Hernandez, R,; Drzal, R.: Pattanakul, C.; Kalyankar, V.; Toebe, J.; and Chou, S. 1987. Recycling 9: High Density Polyethylene Milk Bottles, Status Report to Center for Plastics Recycling Research, Rutgers University, Jan.-Mar., 18 pp. (unpub.) Shenoy, A.V. and Saini, D.R. 1984. Rheological models for unified curves for simplified design calculations in polymer processing,Rheologica Acta, 23(4):368-377. Shenoy, A.V.; Chattopadhyay, 8.: and Nadkarni, V.M. 1983. From melt flow index to rheologram, Rheologica Acta, 22(1):90-101. Stamper, L. and Connole, K. 1984. In-Plant Reworking of Polyethylene Materials, Wire Journal International, 17(10):46-48,51,53. 76 GENERAL REFERENCES Aklonis, J.J. and MacKnight, W.J. 1983. Introduction 39 Polymer Viscoelasticity, 2nd Edition, A Wiley- Interscience Publication. Bodyfelt, F.W.; Morgan, M.E.; Scanlan, R.A.; and Bill, D.D. 1976. A Critical Study of the Multiuse Polyethylene Plastic Milk Container System, i; Milk Food Technology, 39(7):401-405. Brydson, J.A. 1982. Plastics Materials, 4th Edition, Butterworth Scientific, London. Dealy, J.M. 1982. Rheometers for Molten Plastics, Van Nostrand Reinhold Co. Horio, M.; Fujii, T.: and Onogi, S. 1964. Rheological Properties of polyethylene Melts: Effects of Temperature and Blending, Journal 9: Physical Chemistry, 68(4):778-783. Kelen, T. 1983. Polymer Degradation, Van Nostrand Reinhold Co. Knutsson, A.: Abbas, K.B.: Berglund, S.H. 1978. New Thermoplastics from old, Chemtech, 8(8):502-508. Kresser, T.O.J. 1957. Reinhold Plastics Application Series: 1.Polyethylene, Reinhold Publishing Corp., New York, 217 pp. Leidner, J. 1981. Plastics Waste Recovery 9: Economic Value, Marcel Dekker, Inc., New York. Lever, A.E. and Rhys, J.A. 1968. The Properties and Testing 9; Plastics Materials, Third edition, Temple Press Books, 445 pp. Lietz, G.A. 1983. Reprocessing PE-film wastes, Translated from Kunststoffe, 73(8):414-418. Miller, E. 1981. Plastics products design Handbook part Bi Materials and Components, Marcel Dekker Inc., New York. Miltz, J. and Ram, A. 1973. Flow Behavior of Well Characterized Polyethylene Melts, Polymer Engineering and Science, 13(4):273-279. 77 Okamoto, T. and Takayanaki, M. 1968. Application of Two- Phase Mechanical Model to Viscoelastic Properties of High Density and Low-Density Polyethylene, Journal 9; Polymer Science: Part C, 23:597-606. Ram, A., and Shimon, G. 1984. Reprocessing and Shear Modification of Polyethylene, Journal 9; Applied Polymer Science, 29:2501-2515. Rokudai, M. 1979. Influence of Shearing History on the Rheological Properties and Processability of Branched Polymers, Journal 9; Applied Polymer Science, 23(2):463-471. Sedlacek, B.: Overberger, C.G.: Mark, H.F.: and Fox, T.G. 1976. Degradation and Stabilization of Polyolefins, Journal 9; Polymer Science, Polymer Symposia, No.57, 475 PP- Selke, S.E.; Grulke, E.A.; Johnson, D.I.; Lai, C.C.; and Miltz, J. 1986. Recycling 9; High Density Polyethylene Milk Bottles, Technical Report #15 to Center for Plastics Recycling Research, Rutgers University, Jan.1-June 30, 38 pp. Shenoy, A.V. and Saini, D.R. 1984. An Approach to the estimation of polymer elasticity, Rheologica Acta, 23(6):608-616. Shenoy, A.V.; Saini, D.R.: and Nadkarni, V.M. 1983. Estimation of the melt rheology of polymer waste from melt flow index, Polymer, 24(6):722-728. Teoh, S. and Cherry, B.W. 1984. Creep rupture of a linear polyethylene: Rupture and pre-rupture phenomena, Polymer, 25(5):727-734. Vink, P.: Rotteveel, R.T.; and Wisse, J.D.M. 1984. Recycling of Crate Material: Weatherability of Stabilised Recycled High Density Polyethylene, Polymer Degradation and Stability, 9:133-144. Wogrolly, E.G. 1975. Present situation of reuse and recycling within the plastics industry in Austria, Paper 11, Reclaim, Recycling and Reuse of Polymers and Plastics, Institution of Electrical Engineering, London WC2, 18/19 June, 6 pp. Yang, H.W.H.; Farris, R.; and Chien, C.W. 1979. Study of the Effect of Regrinding on the Cumulative Damage to the Mechanical Properties of Fiber-Reinforced Nylon 66., Journal 9; Applied Polymer Science, 23(11):3375-3382. 78 "IIIIIIIIIIIEVIL