Analysis of Fiber Drawing in Wet Spinning for Surface Roughness

The drawing process behavior was investigated with focus on fiber stretching speed (spinning acceleration) to improve luster quality of fibers. Wet spinning method has a limitation of low luster at high spinning speed. It was determined that (1) luster quality could be evaluated by arithmetic average ( R ∆ a ) of fibers, which indicated the roughness, and (2) the roughness of the fiber was related to the spinning acceleration through the analysis of R ∆ a . Spinning acceleration was measured by chasing markers that were used to tie fibers. Regardless of the length of spinning bath, fibers were mainly stretched during the first stretching stage. Therefore, a multi-step drawing method was used. In the case where the drawing ratio was 220% by one-step, R ∆ a was 10.1°; however by multi-step drawing (1 st and 2 nd drawing ratios were 148.3%), R ∆ a decreased to 8.3°. The multi-step drawing method enabled the reduction in fiber roughness by preventing a sudden change in fiber stress. In addition, high temperatures improved the fiber roughness. At high temperatures, roughness decreased despite the high acceleration because the fiber was easier to stretch than at low temperatures.


Introduction
Wet spinning method has shown high productivities for a long time. This method with acrylic polymer is preferred to construct fibers such as human hair products because of good texture and feel. However, it is limited by low luster at high spinning speeds. Typically, wet spinning method involves two drawing processes for determining fiber strength. First drawing process is in the bath and the second drawing process is after drying, and the drawing ratio is approximately 200-800% (Ohe et al., 1967). In this report, drawing ratio means ratio of rolls speed between inlet and outlet. For example, When inlet rolls speed is 10 cm/s and outlet rolls speed is 20 cm/s, the drawing ratio is 200%.
When sample is taken at high draw ratio or high speed spinning, drawing fibers sometimes have many wrinkles on the surface. It has been reported that wrinkles on the coagulated fiber side surface increase in the stretching stage by drawing below secondary transition temperature (Sawanishi et al., 1998). The fiber roughness studies have focused on "dried fiber" in the drawing process. Drawing behavior in the bath was reported only for desolvation (Takeda et al., 1964). Desolvation in the bath is related to the fiber stretching ratio, but the relationship between roughness and drawing behavior has not yet been clarified.
In this study, we focused on roughness in drawing process in the bath and we investigated fiber behavior during drawing. By controlling the factors that result in fiber roughness in the bath can allow high productivity by wet spinning.

Materials
Polymer solution was prepared by mixing modacrylic copolymer (acrylonitrile : vinyl chloride = 50 : 50 (abt.)) powder, Dimethyl sulfoxide (DMSO), and pure water. Modacrylic copolymer was synthesized by emulsion polymerization. Figure 1 illustrates the schematic experimental apparatus in one-step drawing. Polymer solution was supplied by the gear pump from 3 L tank to the coagulation bath (1 st bath). Polymer solution was coagulated and fibers were formed in the 1 st bath. Extruded fibers from 10-hole nozzle were rolled up and introduced to the drawing bath (2 nd bath) connected to the 1 st bath. After drawing process, fiber structure is getting dense and that fiber strength is increased.  samples were collected by installing clover-shaped rolls after each bath. In one-step drawing, the drawing ratio was 150-290%, and in two-step drawing, two baths were assembled. Figure 2 illustrates the schematic experimental apparatus of two-step drawing. A bath of 2.0-5.0 m was used for drawing and DMSO concentration was 50 wt.% in each bath. For the evaluation of stretching behavior, the moving speed was measured by a digital camera. The stretching phenomena was expressed by equation (1). Acceleration α is typically affected by stretching ratio, temperature, and fiber tension. In this case, it was assumed that α was constant in any stretching ratio under constant temperature and fiber tension.

Fiber characterization
To investigate the factors that cause low luster at high spinning speed, scanning electron microscopy (SEM; HITACHI; S-4800) was utilized to observe the fiber side surface. Additionally, the fiber arithmetic average (R∆a), which indicated fiber roughness, was evaluated using a laser microscope (KEYENCE; VK-X100).
The drawability is affected by the drawing bath temperature. Therefore, thermal mechanical analysis (TMA; SII Seiko Instruments Inc.; TMA 150) was utilized to evaluate the relationship between temperature and the drawing response for a single fiber. DMSO concentration in the drawing bath was 50 wt.% and the load was 14 mN. Figure 3 shows the fiber side surface of fiber products of high and low luster level using laser microscope. To satisfy high luster level, it is necessary that R∆a is lower than 12° by evaluating standard fiber products.

Multi-step drawing
Thereafter, R∆a was measured in the drawing fiber side surface between the high and low productivities. As a result, R∆a was 8° at 8.3 cm/s and 21° at 16.7 cm/s ( Figure  4). It was speculated that excessive stretching force in a drawing process caused the roughness in fiber side surface; therefore, R∆a should be reduced to improve the luster in drawing process. Next, the stretching behavior was evaluated using parameter α. Figure 5 shows a comparison of the experimental stretching behavior in one-step stretching and that of the calculation when α is constant in equation (1). As a result, the stretching behavior could be expressed by equation (1) at constant α.  Figure 6 shows one-step drawing and multi-step (two step) drawing behavior. In this experiment, bath temperature was 90°C and drawing ratio was 220%. In the case of two-step drawing, drawing ratio in the 2 nd bath 1 and 2 nd bath 2 were 148.3%. In one-step stretching, the fibers were suddenly stretched in the first half of the bath, but two-step drawing showed slower stretching due to the control of drawing ratio in each bath.
In addition, the time required to achieve the target velocity (24.8 m/min) was measured. In one-step drawing, the required time was approximately 9 s and in two-step stretching, it was approximately 12 s. The values for α were calculated in both cases. It was determined that α was 3.5 cm/s 2 in one-step drawing and 0.93 cm/s 2 (first step) and 2.4 cm/s 2 (second step) in two-step drawing. Furthermore, R∆a decreased from 10.1° to 8.1° upon changing from one-step drawing to two-step drawing. Therefore, two-step drawing was effective in decreasing α and RΔa (Table 1) because α was decreased upon changing to two-step drawing.

Figure 6.
Comparison of stretching behavior in one-step and two-step drawing at 90°C.  Figure 7 shows the relationship between R∆a and α in one-step drawing at 50°C. When α was increased, R∆a also increased. It was necessary to decrease the roughness of fiber side surface to satisfy R∆a value below 12° in the drawing fiber. It was determined that for R∆a below 12°, α must be below 2.5 cm/s 2 .

Figure 7.
Relationship between R∆a and a in one-step drawing at 50℃.

Temperature dependency of the upper limit of α
We examined the temperature dependency of the upper limit of α, which satisfied the R∆a value of below 12°. Figure 8 illustrates the relationship between α and temperature, with R∆a as a parameter in one-step stretching. The value of R∆a was measures from 40-80°C at every 10°C interval. The dashed line represents the upper limit of α, which satisfied RΔa value of below 12°.
It was determined that the allowable range of acceleration (RΔa below 12°) significantly increased when the temperature exceeded 60°C. Therefore, it is important that fiber temperature is > 60°C in the drawing process for a decrease in roughness. To understand this phenomenon, strain was measured using TMA. Figure 9 shows the TMA result. Fiber strain gradually increased when fiber temperature is above 55°C.
It was assumed that increase in upper limit of α at 60°C was due to the fiber material characteristics. Therefore, an increase in the temperature during drawing process corresponds to a decrease in the fiber stress. Figure 9. Stretch response to temperature.

Optimal ratio of each bath in two-step drawing
We investigated optimal ratio of each bath in two-step drawing. In this experiment, we changed 2 nd bath 1 drawing ratio from 120% to 180% and at that time we changed 2 nd bath-1 temperature each condition.  Figure 10 illustrates the relationship R∆a and 2 nd bath 1 ratio in each temperature. As a result, we found that when T1 was lower, for example T1 = 50℃, 2 nd bath 1 ratio was better to decrease R∆a. Otherwise when T1 was over 60℃, 2 nd bath 1 ratio was not effective until around half ratio of total.
These results mean two-step drawing have adjustable function in proportion to 2 nd bath 1 temperature to decrease R∆a by prevent fiber stress.

Conclusion
To improve the fiber luster quality in wet spinning, we focused on the drawing process in the bath. It was determined that to decrease the roughness of fiber side surface, acceleration in drawing process should be decreased. In high speed spinning, selecting two-step drawing was an effective method to decrease the acceleration at stretching stage. Slow stretching prevented extra stress in the fiber such that the roughness of the fiber surface decreased and the fiber luster level increased. In addition, the temperature of the bath is also related to with fiber roughness. Secondary transition temperature in modacrylic polymer is approximately 60°C and the upper limit of α is above 60°C. At temperatures above 60°C, roughness decreased despite the high acceleration because the fiber was easier to stretch than at low temperatures. T 1 = 50℃, T 2 = 90℃ T 1 = 70℃, T 2 = 90℃ T 1 = 60℃, T 2 = 90℃