Birefringent approach for assessing the influence of the extrusion temperature profile on the overall Orientation of as-spun aliphatic-aromatic co-polyester fibers.
Synthetic biodegradable polyesters are produced in two broad types, highly amorphous, flexible polyesters with properties similar to low-density polyethylene (LDPE) and semicrystalline, rigid polyesters with properties similar to polypropylene (PP), polyester (PE), or/and polyethylenete-lephthalate (PET). Many scientists have worked to produce and develop new biodegradable polymers (1-5).
Biodegradable polymers are being investigated to solve environmental waste problems, environmental pollution. and the harming of wildlife by reducing the solid waste from plastics (6). The use of composite materials is increasing at more than 10% per year (7). As a result, the replacement of chemical fibers with environmentally friendly fibers is a major goal of the textile industries (8). The main types of biopolymers are based on starch, sugar, cellulose, or synthetic materials.
Both aliphatic polyesters and aromatic polyesters and their co-polyesters are degradable polymers. Compared to aliphatic polyester, the aromatic form is often based on terephthalic diacid (9), (10). They can made from petroleum with stable physical and chemical properties with comparable cost to other plastic polymers (11). A number of scientists have tried to increase the strength of bioplastics and their properties without affecting their biodegradability (12). A balance between the improvement of mechanical properties and the biodegradability needs to be investigated (13).
The commercial uses of biodegradable polyesters are for paper coating, fibers, and refuse bags. To improve biodegradable polyester properties, researchers have made crimp staple fibers from polyiactic acid, thermoplastic starch, polycaprolactones, and natural fibers. Other biodegradable products include disposable wipes, seed mats, and erosion control items (14). Aliphatic-aromatic co-polyesters are biodegradable under certain conditions leaving no environmental footprint (14). Aliphatic-aromatic co-polyesters are potential candidates to make staple fibers for various nonwoven materials particularly for expendable uses in medicine and agriculture (8), (15-18).
Melt spinning with its inherent high processing speeds and capacities has many economic advantages (8). The orientation in melt spun fibers increases with draw-down ratio. There are different reigns in the fiber structure that are randomized amorphous chain, molecular chains parallel to each other but not oriented, and oriented crystalline areas (19). Overorientation increases with shear rate associated with a small spinneret orifice size, resulting in an increase in fiber strength and reduction in elongation at break. Crystallinity and orientation affect strength, elongation, moisture absorption, absorption resistance, and dye-ability. High orientation reduces molecular mobility and the dyeabilily (19), (20).
In this research, different samples of as-spun aliphatic-aromatic co-polyester fibers were spun at different extrusion temperature profiles. The effect of the extrusion temperature profile on the optical birefringence of the spun fibers was determined. Accordingly, the overall orientation of the spun filaments was modeled. Other properties of these fibers will be reported in further work.
A fully biodegradable petroleum aromatic-aliphatic co-polyester (Solanyl flexibility component), based on butandiol, adipic acid, and terephthalic acid, supplied by Rodenburg (Netherlands) was used in this work.
The polymer shape is spherical granule resin, diameter 3-5 mm, with density of 1.2 [g/[cm.sup.3]]. The range of melting temperatures is 110-115[degrees]C. The material is referred to (21) as 1,4-benzenedicarboxylic acid, polymer with 1,4-butanediol and hexanedioic acid. Due to the low water solubility and high molecular weight of the polymer, it is unlikely to bioaccumulate (22), no ecotoxicity data were submitted (23), and no significant toxicological effect was observed (24).
Fibers were extruded via melt spinning on a lab-spin machine, Extrusion Systems (UK). The lab-spin machine (see Fig. 1) consists of an extruder, a metering pump, a die head, an air cooling window, a spin finish application system, and a winding system. Polymer granules are fed through the hopper into the extruder, then mechanically compressed, and melted. The molten polymer is forced through the spinneret (55 holes) as fine jets with speed adjusted by the metering pump (fixed at 4 rpm) which generates the high pressure during metering. The air cooling quench was set at 37%. The filaments then cool down and harden progressively to emerge as solid filaments. The spin finish (0.4 rpm) was diluted fivefold with water before use. The filaments were collected from the godets set at 36 m/min without tension between them and the winder.
[FIGURE 1 OMITTED]
Optical birefringence was determined using a polarizing interference microscope depending on the strong relationship between birefringence and molecular orientation (25). The value of the measured birefringence reflects the overall orientation or the alignment of the fiber molecules around its axis. As is known, low birefringence fibers have low orientation and would be more drawable than the higher values and vice versa.
The double refracting polarizing interference (Pluta) microscope was designed, developed, and applied by Pluta (26). It is especially suitable for microinterferometry of birefringent fibers. This microscope was used to conduct observations of various micro-objects that produce a shift either in the phase or amplitude of light waves being transmitted. This microscope also measures the optical path difference, gradient of the optical path difference thickness, refractive index, birefringence, and other physical quantities.
In this work the subtractive position is applied to give the nonduplicated images for the direct measurement of the fiber's birefringence ([DELTA]n). The direct measurement of [DELTA]n is most desirable using the Pluta microscope (subtractive position). For accurate and less time-consuming measurements, this Pluta microscope was equipped with a computerized unit consisting of a CCD microcamera, PC computer, and a digital monitor. The measured parameter can be analyzed online using this unit and the microinterferogram of fibers (27). The difference between the refractive indices ([n.sup.||] - [n.sup.[perpendicular to]]) of the fiber defines its birefringence ([DELTA]n) which can be determined using the following equation (28):
[DELTA]n = [[Z * [lambda]]/[b * t]], (1)
where Z is the apart of the fringe shift, [lambda] is the wave-length of the light used, b is the interfringe, spacing, and t is the fiber diameter. The error of measuring the optical path difference (Z [lambda]/b]) using the interference Pluta microscope is [+ or -]0.003 (29).
RESULTS AND DISCUSSION
Factorial Experimental Design and Results
The extrusion experiments conducted involve three factors at two levels as given in Table 1. A fraction factorial experimental design (L8 Array) with random order was used (30-32) as shown in Tables 2 and 3. These selected control factors covered the six temperature zones (see Fig. 2) by combining two zones and considering them as one in order to simplify the experiment i.e. Zone 1: Z1 = T1&T2, Zone 2: Z2 = T3&T4, and Zone 3: Z3 = T5&T6. The temperature zones in the extruder are Barrel zones (T1, T2, and T3), metering pump zone (T4) and die head zones (T3 and T6).
[FIGURE 2 OMITTED]
TABLE 1. Factors and their levels for the experiments. Factors Z1 Z2 Z3 Experiment Level (a) T1 T2 T3 T4 T5 T6 1 -1 110 115 120 120 125 125 1 115 120 125 125 135 135 2 -1 110 115 120 125 130 130 1 115 120 125 130 145 145 (a) -1: low level, 1: high level. TABLE 2. Experimental array for the first design. Factors Z1 Z2 Z3 Trial number T1 T2 T3 T4 T5 T6 Birefringence X [10.sup.-3] 1 115 120 120 120 135 135 17.2 2 110 115 120 120 135 135 20.5 3 115 120 125 125 125 125 26.3 4 115 120 120 120 125 125 23.9 5 115 120 125 125 135 135 10.9 6 110 115 125 125 125 125 20.8 7 110 115 125 125 135 135 17.6 8 110 115 120 120 125 125 33.9 TABLE 3. Experimental array for the second design. Factors Z1 Z2 Z3 Trial number T1 T2 T3 T4 T5 T6 Birefringence X [10.sup.-3] 1 115 120 120 125 145 145 3.7 2 110 115 120 125 145 145 9.3 3 115 120 125 130 130 130 21.7 4 115 120 120 125 130 130 17.1 5 115 120 125 130 145 145 3.0 6 110 115 125 130 130 130 10.5 7 110 115 125 130 145 145 0.0 8 110 115 120 125 130 130 22.1
The parameters (Eq. 1) were measured. The values are repeated measurements for each sample represented by the random order number in the first column in both design tables. It can be seen that the birefringence data from the three specimens for the second design have smaller standard deviations than that of the first design, for reasons such as: (i) blocked nozzles in the spinneret because of the nature of this polymer and the nonuniform flow which reduces at high temperature as in the second design, (ii) the tension, or (iii) some tension during preparing the sample for the test. The detailed experimental arrangement of the sixteen trials for the first and the second design are shown in Tables 2 and 3, respectively. The experiments were conducted in one block; the eight trials in each experiment design were listed in random order. Therefore, Figs. 3 and 4 show the recorded micro-interferograms of the two designed sets of based fibers using the polarizing (Pluta) microscope.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Statistical Analysis of the Effect of the Temperature on the Overall Orientation
Figure 5 shows the effect plots of statistical analysis of the effects caused by the main factors and their interactions on birefringence. All the effect plots are constructed directly from the raw data using the computer software available. From these effects plots it can be seen that the first and the second sets of data generally agree with each other (note the difference between the scales in Fig. 5). The main factors of the first design are not confounded with the other main factors of the second design.
[FIGURE 5 OMITTED]
Depending on the main effect plots, the most significant effect on overall orientation is Z3 presenting die head temperature (T4 and T5). The other main factors, Z2 and Z3, exhibit little prominence. The interactions of Zl and Z2 are quite more prominent in the second design [Fig. 6(b)] than that in the first design [Fig. 6(a)]. The noted interaction between Zl and Z2 will be discussed separately below to find the significance that may arise from normal experimental error. The interactions Z1Z3 and Z2Z3 show no significant interaction effect. In conclusion, there are weak interactions between the factors (Z1Z3 and Z2Z3) because of the parallel between the lines in the interaction plots as shown in Figs. 5(b) and (c).
[FIGURE 6 OMITTED]
Alternative to an effects plot, a Daniel's plot can be used to assess the significance of the factors effects. Figure 7 displays the normal probability (Daniel's) plots and shows the percentage and standardized effects for the first and the second designs. The effect from Z3 is again prominent in both designs; the effects from the Zl and Z2 are prominent in the second design. Fig. 7(a), but not in the first design, Fig. 7(b).
[FIGURE 7 OMITTED]
Analysis of Variance for Birefringence (ANOVA)
In order to determine the factor effects in terms of statistical significance, analysis of variance (ANOVA) of the birefringence data was conducted. A probability or P value used in ANOVA provides quantitative and objective criteria for judging the statistical significance of the effects. In case P-values are less than 0.05, the factors effect is significantly different from zero at the 95.0% confidence level. The F-ratio is a new statistic in ANOVA; if the ratio is much larger than the circuital value F(l,7) = 5.59, which was obtained from F tables at the appropriate level [alpha] = 0.05 (32), the factor has significant effect.
The ANOVA table partitions the variability in birefringence into separate parts for each effect and then tests the statistical significance of each effect by comparing the mean square against the experimental error. The results of both first and second designs are listed in Tables 4 and 5, respectively. To provide quantitative and objective criteria for judging the statistical significance of the effects, one factor (Z3) has P-values (0.0426) less than 0.05 (see Table 4), indicating that Z3 is significantly different from zero at the 95.0% confidence level; F-values of 22, 23. and the interaction Z1Z2 are greater than 0.05 and thus are non significant. In Table 5, (23) has P-values (0.0041) less than 0.05, indicating that 23 is significantly different from zero at the 95.0% confidence level, P-values of 22 and 23 are greater than 0.05 and of no significant effect; the interaction ZIZ2 has P-value of 0.0374, lower than 0.05, and so there is a significant effect for the interaction 2122 in the second design. The interaction 2122 plays a role in the higher temperature and that belongs to the heat flowing because of the continuous flow of the molten polymer, which is critical because there are no significant effects of the factors 21 and 22 and may be related to the normal experimental error. In the two cases, it is clear, from Tables 4 and 5, that the effect of 23 is statistically significant at the smallest risk level [alpha] = 0.05, the P-values for the effects of the other factors are considerably greater than 0.05, and as the result they are insignificant as indicated clearly by the effects plots. The quantitative ANOVA results are consistent with quantitative conclusions derived from the effects plots and Daniel's plots.
TABLE 4. ANOVA results identifying the statistical significance of factors effects on birefringence for the first design. Source Sum of squares Df Mean square F-ratio P-value Z1 26.2813 1 26.2813 1.62 0.2930 Z2 49.5012 1 49.5012 3.05 0.1791 Z3 187.211 1 187.211 11.53 0.0426 Z1Z2 18.3012 1 18.3012 1.13 0.3663 Total error 48.7137 3 16.2379 Total (corr.) 330.009 7 Df: degree of freedom; P-values are at the smallest risk level x at which the data are significant; F(1, 7) = 5.59 at [alpha] = 0.05 under the conditions. TABLE 5. ANOVA Results identifying the statistical significance of factors effects on birefringence for the second design. Source Sum of squares Df Mean square F-ratio P-value Zl 1.62 1 1.62 0.27 0.6395 Z2 36.125 1 36.125 6.01 0.0915 Z3 383.645 1 383.645 63.85 0.0041 Z1Z2 76.88 1 76.88 12.80 0.0374 Total error 18.025 3 6.00833 Total (corr.) 516.295 7
The Regression Equation and Estimation Results for Birefringence
Based on the analysis, the simplified model was fitted by the regression equation (Eqs. 2 and 3) which has been fitted to the data for the first and second designs, respectively. The regression equation in terms of the coded values is the following:
[DELTA][n.sub.1]([10.sup.-3]) = 21.3875 - 1.8125 * Z1 - 2.4875 * Z2 - 4.8375 * Z3 + 1.5125 * Z1 * Z2 - 0.6875 * Z1 * Z3 + 0.1875 * Z2 * Z3, (2)
[DELTA][n.sub.2]([10.sup.-3]) = 10.925 + 0.45 * Z1 - 2.125 * Z2 - 6.925 * Z3 + 3.1 * Z1 * Z2 - 1.1 * Z1 * Z3 - 0.375 * Z2 * Z3. (3)
From Eqs. 2 and 3, the most significant effect is 23 which has a larger constant than that of 21 and Z2 and the interactions. Figure 8 shows the observed results and fitted results generated using the last fitted models for the first (a) and the second design (b) for each trial. In Fig. 9, each value corresponds to the values of the experimental factors in a specific row of the data file in the cube plot depending on the regression equation.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
The pattern of estimated responses is based on the assumed model derived from the experimental observations. The geometric result of plotting a response variable is as a function of two factors. In order to determine the direction of the interaction ZIZ2, the estimated response surfaces of the response (birefringence) for the two designs (a and b) are constructed as shown in Fig. 10. The twist in the surface (3D surface response diagrams) in both plots seen in the figure confirms the interaction as mentioned before. The highest points of the both surfaces are associated with the low levels of Zl and Z2. From Fig. 10(a) the expected lowest value of the birefringence will be at high levels of Zl and Z2 but the lowest one in Fig. 10(b) will be at high level of Zl (T1 and T2) and high level of Z2 (T3 and T4), that is due to the heat diffusion between the second T2 and the third 73 zones of the barrel as shown previously in Fig. 2 as a result of the continuous flowing of the molten polymer.
[FIGURE 10 OMITTED]
In the optimization of birefringence, Table 6 shows the combination of factor levels which maximize birefringence over the indicated region. The optimum birefringence value will be 0.032 and 0.021 for the first and the second designs, respectively. When the temperatures of Zl, Z2, and Z3 increase, the birefringence decreases and that relates to the change in the polymer rheology at high temperatures. Table 6 .summarizes the main conclusion of the birefringence results in a concise statistical model. The model covers the identified significant main and interaction factors and specifies the combinations of factor levels for enhancing birefringence and thus the overall orientation of as-spun aliphatic-aromatic co-polyester filaments. Such (bicomponent) fibers can be used as the total fiber content in the fabric in agricultural and horticultural applications such as seed mats, erosions, and seasonal weed control ground covers and more of nontraditional fibers and fabric applications (14).
TABLE 6. The combination of factor levels: first design (a) and second design (b). Maximum 32 Minimum 13 Optimum value Factor Optimum model Actual value Optimum model Actual value (a) First design The combination of factor levels (birefringence X [10.sup.-3]) Z1 Low level 110-115 High level 115-120 Z2 Low level 120-120 High level 125-120 Z3 Low level 125-125 High level 135-135 Optimum value Maximum 21 Minimum 0 Factor Optimum model Actual value Optimum model Actual value (b) Second design The combination of factor levels (birefringence x [10.sup.-3]) Z1 Low level 110-115 Low level 110-115 Z2 Low level 120-125 High level 120-125 Z3 Low level 130-130 High level 145-145
(1.) HJ. Kang and S.S. Park, J. Appl. Polym, Sci., 72(4), 593 (1999).
(2.) S.S. Park, S.H. Chae, and S.S. Im, J. Polym, Sci. Polym. Chem., 36(1), 147 (1998).
(3.) S.S. Park and H.J. Kang. Polym. J., 31(3), 238 (1999).
(4.) R.A. Hayes, U.S. Patent 6.485,819 (2002).
(5.) N.C. Hoppens, T.W. Hudnall. A. Foster, and C.J. Booth, J. Polym. Sci. Polym. Chem. 42(14), 3473 (2004).
(6.) A. Steinbuchel, Ed., Biopolymers. Vol. 10, Wiley, New York (2003).
(7.) R.S. Blackburn, Biodegradable and Sustainable Fibers, The Textile Institute, Boca Raton & Woodhead Publishing, Manchester, UK (2005).
(8.) K. T. Schmidt, Fib. Text. East. Fur.. 12(2). 15 (2004).
(9.) Environment Australia, NOLAN-ITU Pty., Canberra, Australia (2002).
(10.) U. Witt, R.-J. Muller, and W.-D. Deckwer. J. Environ. Polym. Degrad., 3(4). 215 (1995).
(11.) Q. Fang and M.A. Hanna, Bioresource Technol., 78(2), 115 (2001).
(12.) E.S. Stevens, Green Plastics, Princeton University Press, Princeton, NJ and Oxford University Press, Oxford, UK (2002).
(13.) E.S. Stevens, Green Plastics. Plastics and the Environment, Princeton University Press, Princeton, NJ (2001).
(14.) Eastman Polymers for Fibers, Eastman Chemical, Kingsport, TN (2002).
(15.) L. Fumin. W.A. Haile, M.E. Tincher. and W.S. Harris, Fur. Patent EP 1330,350(2003).
(16.) J.H. Wang and H. Aimin, The World Intellectual Property Organization (WIPO) Patent WO/2008/008068 (2008).
(17.) F. Lu, W. Ahaile. M. Etincher, and S.H. Wiley. The World Intellectual Property Organization (WIPO) Patent WO/2002/028626 (2002).
(18.) K. Twarowska-Schmidt and M. Ratajska, Fib. Tex. East. Eur., 13(1), 71 (2005).
(19.) S.J. Kadolph and A.L. Langford, Textiles, Pearson Education. Upper Saddle River. NJ (2002).
(20.) H. Brady, Synthetic Fibre Materials, Longman, London (1994).
(21.) National Industrial Chemicals Notification and Assessment Scheme (NICNAS), Canberra, Australia (2003).
(22.) D.W. Council, General Characteristics of Organic Compounds Which Exhibit Bioaccumulation. In Bioaccumulation of Xenobiotic Compounds, CRC Press, Boca Raton. FL (1990).
(23.) J.V. Nabholz, P. Miller, and M. Zeeman, "Environmental Risk Assessment of the New Chemicals Under the Toxic Substances Control Act (TSCA) Section Five," in Environmental Toxicology and Risk Assessment. ASTM STP1179. Wayne G. Landis, Jane S. Hughes, and Michael A. Lewis, Eds., ASTM, Philadelphia, 40 (1993).
(24.) U. Witt, T. Einig, M. Yamamoto, I. Kleeberg. W.-D. Deckwer, and R.-J. Muller, Chemosphere, 44(2), 289 (2001).
(25.) H. Hermans, Contribution to the Physics of Cellulose Fibers, Elsevier, Amsterdam (1946).
(26.) M. Pluta, .J. Microsc, 93(2), 83 (1971).
(27.) T.Z.N. Sokkar, H.M. El-Dessouky, M.A. Shams-Eldin, and M.A. El-Morsy, Opt. Laser Eng., 45(3). 431 (2007).
(28.) S.C. Simmens, Nature, 181, 1260 (1958).
(29.) M. Pluta, Advanced Light Microscopy: Measuring Techniques. Vol. 3, PWN-Polish Scientific Publishers, Warsaw, Poland (1993).
(30.) R.H. Lochner and J.E. Mater, Design for Quality, Chapman and Hall, London (1990).
(31.) MINITAB, Minitab, State College. PA (1994).
(32.) M.S. Phadke. Quality Engineering Robust Design. Bell Laboratories, Murray, NJ (1989).
Basel Younes, (1) Alex Fotheringham, (1) Hassan M. El-Dessouky (2), (3)
(1) School of Textiles & Design, Heriot-Watt University, Scottish Borders Campus, Netherdale, Galashiels, TD1 3HF, UK
(2) Fibers Research Laboratory, School of Design, University of Leeds, LS2 9JT, UK
(3) Physics Department, Faculty of Science, Mansoura University, Mansoura, Egypt
Correspondence to: Alex Fotheringham; e-mail: email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||Younes, Basel; Fotheringham, Alex; Dessouky, Hassan M. El-|
|Publication:||Polymer Engineering and Science|
|Date:||Dec 1, 2009|
|Previous Article:||Online simulation-based process control for injection molding.|
|Next Article:||Characterization of contamination effects for two polypropylene-based materials.|