• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Mapping of Spin Finish Oils on Nylon 66 Fibers

Posted on: Friday, 2 April 2004, 06:00 CST

ABSTRACT

Mapping the distribution of spin finish oils on nylon carpet fibers is the focus of this investigation. Nylon 66 fibers of two modification ratios, 2.4 and 3.0, manufactured for the carpet industry, are compared to similar fibers pulled from dyed and finished carpet tufts. Fibers are treated with osmium tetraoxide to provide contrast in scanning electron microscopy. After wet processes of dyeing and finishing, the carpet fibers have a significantly lower amount of finish, but the concentration of oil on the carpet fibers is 47% that of the raw fiber. Higher relative concentrations of finish oil components are found at the fiber core compared to the surface. The finish is believed to diffuse into the core during application on the melt spinning line, and the morphology of the fibers in the core allows retention of more residual finish oil components. Capillary action appears to be a driving force for wicking of finish in this application system, because significantly higher relative concentrations of oil are found in the valleys compared to the tips of the multilobal fibers. Side-to-side variations exist in the finish distribution, which are attributed to the single applicator system, where the finish has to wick around the fibers and diffuse through them to coat the side opposite the applicator.

Spin finishes are applied to synthetic fibers to aid in processing into final products [6]. Both the uniformity and location of finishes have an impact on their effectiveness, but techniques to measure the distribution and uniformity of spin finishes on fibers are limited. A wettability scanning method was developed by Kamath et al. [2] to measure surface energies along the length of a single filament. They used changes in surface energy along a fiber length to indicate spin finish uniformity. This technique measures surface effects, but does not take into account variations that may be seen throughout the cross sections. Further research by Kamath et al. [3] at TRi Princeton led to the development of a microfluorometry technique that measures finish uniformity along the thread line. This technique involves mixing a water-soluble fluorescing agent with the finish before application and subsequently measuring the location of this fluorescing agent as an indicator of finish uniformity along the threadline. Neither of these methods determines the uniformity of the finish throughout the cross sections, especially for non-round cross sections such as the trilobal shapes often used in carpet fibers. Nor did these techniques detect finish absorbed into the interior of the fibers, which impacts manufacturing costs and possible changes in fiber properties such as when the finish acts as a plasticizer of the polymer.

Much work has been done to locate residual oils on apparel fibers, such as cotton, polyester, and nylon, by using chemical tagging and electron microscopy methodology [1, 4, 5]. X-ray microanalysis and backscattered images of tagged oils are very useful for mapping the distribution of oily soils on textiles. Trent et al. [10] used similar methodology for ruthenium tetraoxide staining of polymers including nylon 1 1 to enhance contrast in transmission electron microscopy to study micromorphology in semicrystalline structures. However, we have found nothing in the literature about using this technique to map the distribution of spin finish oils on textile fibers. Therefore, this is the focus of our research paper.

Experimental Procedures

SAMPLE PREPARATION

Nylon 66 staple fibers targeted for the carpet market and carpets that were dyed and finished (also staple fibers) were provided for this study by Nylon Textile Technology, Solutia, Inc. (Pensacola, FL). Fibers had mixed cross sections, 2.4 and 3.0 modification ratios, which were 18 and 2) denier filaments, respectively. The modification ratio is a number used to describe the shapes of fibers with trilobal cross sections. To obtain the modification ratio (MR), one draws two circles around the fiber, using the fiber tips for the points of one circle and the fiber valleys as the points of the other circle. The ratio of the outside circle to the inside circle is the MR. A 3.0 MR fiber (Figure Ia) has concave surfaces of small radius, compared to a 2.4 MR fiber (Figure Ib). In spinning the fibers in this research, both MR fibers were present in each spinning position at a ratio of 80:20 of 3.0 MR : 2.4 MR to create an intimate blend product. The spin finish to produce the fibers for this study is a commercial proprietary composition from Solutia, Inc.; it was applied by a metering pump to control the flow rate, using a single applicator with a target finish application of 1.2% by yarn weight.

FIGURE 1. Location of data collection for (a) high modification ratio fiber (upper left), (b) low modification ratio fiber (upper right), (c) illustration of the regions for valley, labeled 1, and tip, labeled 2 (lower left), and (d) illustration of the distance from the fiber surface into the core with zones labeled 1 through 4 (lower right).

The nylon staple fibers produced for this study and staple fibers drawn from nylon carpet tufts, randomly cut from the carpet, were exposed for 48 hours to vapors of 2% (w/v) aqueous osmium tetraoxide solution (OsO^sub 4^; E. M. grade obtained from Electron Microscopy Sciences, Fort Washington, PA). The osmium tetraoxide was used to chemically tag the finish oils by reacting with unsaturated double bonds. Typical spin finishes normally contain lubricants, emulsifiers, and antistatic agents, as well as other components that may be included to provide special characteristics, e.g., antimicrobial or cohesive properties [8]. Industrial grade chemicals used in spin finishes include components such as ethylene oxide- propylene oxide random copolymer, nonionic surfactants such ethoxylated aliphatic alcohols and ethoxylated nonyl phenols, hydrogenated castor oil, phosphate ester antistatic agent, coconut oil, and coconut oil ethoxylate. We experimentally determined that each of these industrial chemicals reacted with osmium tetraoxide.

MICROSCOPY

We obtained backscattered electron images for longitudinal and cross-sectional specimens of the chemically tagged fibers. To make the cross sections, the fiber samples were embedded in Spurr's low- viscosity resin (Electron Microscopy Services) and cured for 8 hours at 7O0C. Each block contained either one carpet tuft or several commercial raw nylon fibers. Resulting blocks were trimmed into a trapezoidal shape and then sliced into smooth sections with glass knives on a DuPont Sorvall MT-I Ultramicrotome (Research and Manufacturing Co. Inc., Tucson, AZ). Cross-sectional and longitudinal specimens were mounted on 3/8-in. carbon stubs with carbon tape and carbon-coated in an Edwards Auto 306 high vacuum evaporator (Edwards High Vacuum International, Wilmington, MA).

We used backscattered electron imaging and subsequent energy dispersive x-ray (EDX) analyses to study the location and distribution of finish oil within the chemically tagged fiber samples. The scanning electron microscope was a JEOL 733 superprobe (JEOL Ltd., Tokyo, Japan) equipped with a Tracer Northern (Middletown, WI) EDX analyzer. Longitudinal and cross-sectional images were obtained with an accelerating voltage of 15 kV. The specimen current was 5 10^sup -9^ A, and the working distance was 20 mm. To map the finish oil distribution and to obtain relative intensities of the Oslabeled oil within a single fiber, we used magnifications of 838 to 873. Net x-ray counts for osmium at an energy window of 1.85-2.05 keV (osmium M line) were recorded.

Undrawn nylon 66 fibers that had no spin finish were treated with osmium tetraoxide vapors and analyzed to determine any reactivity of the nylon 66 polymers with the selected tag. Backscattered electron images exhibited no contrast for these specimens. X-ray microanalyses of cross-sectional specimens of these osmium treated control fibers also showed no presence of osmium. The mean and standard deviation of the x-ray intensity for the energy window for osmium were 16 35 for the low modification ratio nylon fiber and 28 25 for the high modification ratio nylon fiber. Thus, we conclude that osmium tetraoxide can be used for contrast in electron microscopy to distinguish the location of spin finish components on the nylon 66 fibers.

To investigate the relative concentration of residual finish oils on both raw fibers and individual fibers from the carpet tufts (fiber type), EDX data were collected by placing the spot probe at selected locations of the cross sections of individual fibers (Figure 1). In all, forty fibers were randomly selected for data collection; we examined ten high MR raw fibers, ten low MR raw fibers, ten high MR carpet fibers, and ten low MR carpet fibers.

STATISTICAL ANALYSES

Our analysis of the results consisted of a cross comparison of relative intensities of Os within samples of carpet fibers and raw fibers (fiber type) for both low and high MR fibers (MR). Data were organized to create different regions within the fibers to analyze cross comparisons of the relative intensities of Os (Figure 1). The regions analyzed were orientation to determine if there were side- to-side variations (orientation) in finish distribution, the valleys as compared to the tips of the fibers \(valley-tip) (Figure 1c), and distance, where the data analyzed were a function of distance from the surface to the core of the fibers (distance) (Figure Id). Relative concentrations of finish oils at selected locations within the textile structure as represented by x-ray intensity data were statistically analyzed with Statistical Analysis Software (SAS, Cary, NC). Analyses of covariance were calculated using a PROC MIXED procedure, which is designed for analyses of random and fixed treatment effects [7].

FIGURE 2. Backscattered electron image cross sections obtained from the osmium treated raw libers (a, left) and carpet fibers (b, right). The benchmarks are 30 m.

Results and Discussion

SPIN FINISH DISTRIBUTION ON FIBERS

Backscattered electron images of cross sections for (a) raw fibers and (b) carpet fibers are shown in Figure 2. Light areas are related to high concentrations of osmium due to its high atomic number. There is a light halo effect surrounding the cross sections; This effect seems fairly uniform, with the lighter areas distributed evenly around the low MR raw fibers and carpet fibers of both MR. However, the intensity appears to be higher in the valleys of the high trilobal cross sections of the raw fibers.

We measured the relative concentrations of osmium, which tags components in the spin finish using energy dispersive x-ray microanalyses. We determined the distributions of oils around the fiber exteriors and within the interiors of both raw fibers and carpet fibers. However, with this methodology it is not possible to determine any differential partitioning of spin finish components that might occur due to molecular size or functionality differences. In the analysis of variance, the main effects of fiber type, valley- tip, distance, and orientation were significant at the 95% confidence level, while the main effect of MR (modification ratio) was not. In addition, all interaction terms are significant at this confidence level.

In the initial data collection phase, we noticed a definite side- to-side variation in relative concentrations of oil on each fiber. The relative difference between the two sides of the fibers was 230 counts versus 119 counts. We were surprised that one side was almost 50% lower in relative oil concentration than the other side. It is probable that one side of each fiber was oriented facing the finish applicator. In the single applicator system, the spin finish is pumped through a slot to coat the fibers as they move past the slot. The applicator is engineered so that a small pool of finish forms at the open slot, and the fibers are supposedly immersed in the liquid as they pass through the pool. Based on our data, however, the finish resides more heavily to one side of the fiber bundle. It seems reasonable to assume that the side oriented toward the slot that disperses the finish receives a significantly higher amount of oil compared to the side of the fibers where the finish has to wick around the fibers or diffuse through them to coat the opposite side. Even though the finish pump and applicator were engineered to coat the fibers uniformly, we observed a definite side-to-side effect.

Recently, some fiber producers have moved to using dual-opposed applicators, two applicators in series, or other specially designed application systems in attempts to improve coating uniformity. The fiber producer often relies on processing performance and other macroscopic testing methods to evaluate the effectiveness of the new application system for improving finish uniformity. It has been a difficult task for the fiber producer to effectively draw conclusions about finish uniformity with the standard evaluations in practice in the industry. Thus, the methodology that we used in this study may be a useful tool for examining coating uniformity during fiber production.

To evaluate the effects of the trilobal fiber shape, we selected regions related to surface morphology. Figure Ic is a schematic of a high MR fiber with those regions labeled 1 representing the valleys and those labeled 2 representing the tips; these regions were similarly defined for the low MR fibers. There were higher relative concentrations of oils in the valleys (212 counts) compared to the tips or legs of the trilobal fibers (151 counts), regardless of fiber type (raw or carpet) or modification ratio. Capillary forces probably influence flow of the finish, which preferentially distributes a higher concentration of oils in the fiber valleys during finish application. Workers in the fiber industry have often thought that finish migrates to the valleys of trilobal fibers. Our data support this hypothesis.

FINISH DIFFUSION INTO FIBERS

When exiting the spinnerette, nylon fibers are very dry and hot, with temperatures greater than 260C. Under these conditions, nylon readily absorbs water due to the high polarity of the polymer. When nylon first enters the finish applicator, it absorbs a significant amount of liquid, much higher than its equilibrium moisture content of 4-4.5%. The finish itself is applied as a low concentration of oils (12-18%) in water. The oils comprise lubricants and a high percentage of water-soluble components, including emulsifiers, cohesive agents, wetting agents, and anti-static agents. As evidenced by our data, water-soluble components of smaller sizes are probably diffused into the fibers with the water during the spinning operation.

Higher concentrations of residual oils at the fiber core may be attributed to the micromorphology of the fiber being different near its surface compared to its core. During quenching in the melt spinning process, fibers have a radial temperature gradient in the cross section [9], the outside of the fibers cooling more quickly than the interior. Crystallization steps of nucleation and growth are temperature-dependent, so there is often a difference in morphology at the fiber core compared with the surface [9]. Thus, the core may be a more open structure, allowing for a higher relative concentration of oil components than near the fiber surface.

The effect of morphology differences between the surface and interior of the fibers can further be substantiated by analyzing the orientation distance interaction term. In Figure 3, we have plotted the amount of finish oils as a function of distance from side 1, where distance 4 is the fiber core (center) and distance 7 is the other surface of the fiber. From these data, we argue that once the finish is inside the fiber, the concentration difference between distances 1 and 2 is due to a morphological difference from the surface into the core of the fiber; the finish distribution within the fiber at distances 3 through 7 then becomes diffusion-driven.

FIGURE 3. Concentrations of finish oils across the fiber cross section.

When the fiber is extruded from the spinnerette and exposed to finish at the finish applicator, using a single applicator and metering pump system, the finish enters the fiber at the side oriented towards the applicator. Finish oils preferentially go to the interior of the fiber, due to morphological differences, where the fiber core is a more open structure compared to the fiber surface. Once inside the fiber, the finish diffuses to the opposite side, giving the gradient in finish amount that we have observed, where the distance that is farthest from the finish applicator has the lowest concentration of finish oils.

EFFECTS OF MODIFICATION RATIO ON FINISH DISTRIBUTION

We found no significant differences in the average relative concentration of oil on fibers with different modification ratios (MR), 2.4 compared to 3.0. From our data, the finish does not appear to migrate preferentially to fibers of one MR versus the other. However, we could not determine the surface area differences in the two modification ratios with the analysis methodologies used in this research.

With further analysis, we observed that orientation or side-to- side differences in finish oil concentration differed with the modification ratio of the fiber (interaction terms of MR orientation). The 2.4 MR fibers had a larger side-to-side variation than the 3.0 MR fibers (Table I). During application, the finish probably wicks along the concave channels of the fibers going from one side to the other in the bundle. With the triangular shape of the low MR fiber, there is very little capillary effect compared to that with the longer legs of the high MR fiber. The finish most likely migrates to the valleys of the higher MR fiber, which tend to have capillaries with a smaller radius and thus higher capillary forces. Capillary action plays a large role in wicking of the finish, causing it to distribute preferentially to the higher modification ratio fiber on the side that is opposite the finish applicator. This explanation becomes more evident when looking at the differences in relative concentration in the valley and tip for fibers with different modification ratios (MR valley-tip).

TABLE I. Relative x-ray counts at fiber locations for fibers with two modification ratios.

FIGURE 4. Concentrations of finish oils across fiber cross sections of raw and processed carpel fibers.

The interaction term of MR valley-tip was significant at the 95% confidence level. The difference in finish oil concentration observed in the valleys compared to the tips of the high MR fibers was larger than that observed for the low MR fibers (Table I). This provides further evidence to support the hypothesis that capillary action is a key driving force that controls the wicking behavior of the finish application process in the spin line using this particular application technology. A fiber with an MR of 2.4 is virtually triangular in shape (Figure 2). The very gradual slope from the tips to the valleys of the low MR fiber results in lower capillary forces than expected for the high MR fiber with a distinct capillary formed in the \valley between two adjacent legs of the 3.0 MR fibers.

EFFECTS OF FINISHING OPERATIONS ON SPIN FINISH DISTRIBUTION

There was a lower relative concentration of oil on carpet fibers than on raw fibers (234 versus 110 counts). Spin finish is removed during dyeing and finishing operations, resulting in significantly lower concentrations after these processes. However, it is surprising that substantial residual oils were still present on the carpet fibers after processing, with the concentrations being 47% that of the original.

Interestingly, the interaction term of fiber type distance was significant. The data for the raw fibers indicate a spin finish oil distribution from the surface to the core of the fiber in a linear relationship (Figure 4). For the carpet fiber, the three data points on the interior of the fiber were not significantly different from each other (nonlinear), although they were higher than that observed at the surface. Carpet goes through significant wet finishing, such as dyeing, and stainblocker and fluorocarbon treatments. These wet processes may level out the finish components found inside the fiber core.

Conclusions

Our research has shown that the distribution of spin finish oils on nylon 66 can effectively be mapped in fiber cross sections by using osmium tetraoxide to tag the unsaturated double bonds in the oils, providing contrast for electron microscopy with the use of backscattered electron images and x-ray microanalysis.

There is a definite orientation (side-to-side) effect, probably due to the single applicator system used. Significantly higher relative concentrations of oils appeal' in the valleys of the fibers versus the tips, especially for the high MR fiber, giving evidence to support the hypothesis that capillary action is a driving force controlling the wicking of the finish. Higher concentrations of residual finish are located within the core of the fibers compared to the surface, probably due to the finish diffusing into the fibers at the initial finish application and to morphological differences of the surface compared to the core. After wet processing during carpet manufacturing, the relative concentration of finish oil is 50% of the relative concentration of oils found on raw fibers.

Further experimentation would be required to address partitioning of components of the spin finish that might occur due to differences in molecular size or chemistry. Also, additional research on the effect of spin finish as a plasticizer on the modulus of the fibers would provide complementary confirmation of the presence of any spin finish within the fibers.

ACKNOWLEDGMENTS

We wish to thank Nylon Textile Technology of SoIutia, Inc. (Pensacola, FL) for providing the fibers and carpets used in this study, and in particular, Annabella Anderson and Rusty Carter for their assistance in producing the samples. We thank Emil Delgado of Ethox Chemical for technical discussions on spin finish components, and John Hunt at the Center for Materials Research at Cornell University for his assistance with the scanning electron microscopy. This research was supported in part by the Cornell University Agricultural Experiment Station federal formula funds, projects no. NYC329407, received from the Cooperative State Research, Education, and Extension Service, U.S. Department of Agriculture, and by Nylon Textile Technology of Solutia, Inc.

Literature Cited

1. Breen, N. E., Durnam, D. J., and Obendorf, S. K., Residual oily Soil Distribution on Polyester/Cotton Fabric after Laundering with Selected Detergents at Various Wash Temperatures, Textile Res. J. 54, 198-204 (1984).

2. Kamath, Y. K., Dansizer, C. J., Hornby, S., and Weigmann, H.- D., Surface Wettability Scanning of Long Filaments by a Liquid Membrane Method, Textile Res. J. 57, 205-213 (1987).

3. Kamath, Y. K., Ruetsch, S. B., and Weigmann, H.-D., Microfluorometric Studies of the Distribution of Finishes on Fibers and Yarns, Textile Res. J. 63, 19-32 (1993).

4. Obendorf, S. K., and KIemash, N. A., Electron Microscopical Analysis of oily Soil Penetration into Cotton and Polyester/Cotton Fabrics, Textile Res. J. 52, 434-442 (1982).

5. Obendorf, S. K., Mejldal, R., Varanasi, A., and Thellersen, M., Kinetic Study of Lipid Distribution after Washing with Lipases: Microscopy Analysis, J. Surfactants Deterg. 4, 43-55 (2001).

6. Redston, J. P., Bernholz, W. F., and Schlatter, C., Chemicals Used as Spin-Finishes for Man-made Fibers, Textile Res. J. 43, 325- 333 (1973).

7. Singer, J. D., Using SAS PROC MIXED to lit multilevel models, hierarchical models, and individual growth models, J. Edu. Behav. Stat. 24, 323-355 (1998).

8. Slade, P. E., "Handbook of Fiber Finish Technology," Marcel Dekker, Inc., NY, 1998, p. 5.

9. Spruiell, J. E., Structure Formation during Melt Spinning, in "Structure Formation in Polymeric Fibers," David Salem, Ed., Hanser Publishers, Munich, Germany, 2001, pp. 17-28.

10. Trent, J. S., Scheinbeim, J. L, and Couchman, P. R., Ruthenium Tetraoxide Staining of Polymers for Electron Microscopy, Macromolecules 16, 589-598 (1983).

Manuscript received September 23, 2003; accepted May 5, 2003.

D. N. HILD,1 S. K. OBENDORF, AND W. Y. FOK

Department of Textiles and Apparel, New York State College of Human Ecology, Cornell University, Ithaca, New York 14853, U.S.A.

1 Corresponding author: Debra N. HiId, 521 Turnberry Road, Cantonment, FL 32533, 850-968-5443, email: dnhild@cox.net

Copyright Textile Research Institute Mar 2004

More News in this Category