Dyeing of Recycled Electrospun Polyamide 6 Nanofibers: Implications of Dye Particle Size

Disperse Red 167 (DR167) dye was supplied by DyStar (Cairo, Egypt) and its structure is shown in Fig. 1. PA 6 100% knitted fabric wastes (135 g/m²) were collected from Salamtex Company, Cairo, Egypt. PA 6 wastes were scoured in a solution containing 1 g/L nonionic detergent for 20 min at 80 °C to remove impurities and waxes, and then rinsed thoroughly with cold tap water, and air-dried [39]. Formic acid 90% was supplied by BDH Chemicals (England). Hydrochloric acid and sodium hydroxide were provided by Biochim (Cairo, Egypt).

Based on our experimental work (data not shown), polyamide fabric wastes were dissolved in formic acid at a concentration of 21% wt/wt. The mixture was continuously stirred using a magnetic stirrer at 650 rpm until homogeneous mixing was achieved at room temperature (25 ± 3 °C) A 10 ml plastic syringe with a metallic needle was used for the electrospinning process. With an operating voltage of 20 kV, the positive electrode was connected to the syringe needle, while the negative electrode was attached to the metallic collector to fabricate the PA-NWNF. The distance between the needle tip and the collector was set at 25 cm and the solution flow rate was adjusted to 0.5 ml/h using a programmable syringe pump (NE-4000, USA).

The different particle sizes of DR 167 dye were prepared by intermittent milling, which included short periods of milling lasting 30 min followed by rest intervals lasting 15 min, using a ball mill machine (Planetary Ball Mill PM 100, RETSCH, Germany). Prior to the milling process, the dye undergoes a drying procedure for a duration of 15 min, at a temperature of 60 °C, with the aim of eliminating any remaining moisture content that could lead to an increase in particle aggregation and a decline in the quality of the dye. A total of 200 g of DR 167 was placed in a 500 mL stainless steel jar, along with 7 balls (2 cm in diameter) and 40 balls (0.5 cm in diameter). The jar was subjected to intermittent milling at a fixed rotational speed of 450 RPM, until a total milling time of 2 h was achieved. Following this, the milled dye was passed through a sieve with a mesh size of 38 µm, resulting in the collection of particles larger than 38 µm, which were labeled as DR 167-B. The remaining particles, with sizes smaller than 38 µm, underwent an additional 8 hours of milling to produce a new sample, designated as DR 167-C.

The dyeing of PA-NWNF samples was carried out with DR 167, DR 167-B, and DR 167-C using distilled water. Five dye concentrations (3, 5, 10, 15, and 20% o.w.f) of the dyes mentioned above were prepared with a liquor ratio of 1:500. To optimize the dyeing conditions, the dyeing was carried out at different temperatures (i.e., RT, 60, 80, 100, and 120 °C) for various dyeing times (i.e., 15, 30, 60, 90, and 120 min) at pH 6. After hot rinsing and air drying, the samples were prepared for color yield and fastness properties (i.e., wash, light, and perspiration) measurements.

SEM images of the PA-NWNF morphology before and after dyeing were taken at 20,000 × , while those of the applied dyes were taken at 400 × (Figs. 1 and 2). The undyed PA-NWNF showed a regular and smooth surface, in addition to the finest homogeneity of fiber diameter with no beads (Fig. 1a). Also, it had a high surface area of about 587.48 m²/g, as shown in Fig. 3, indicating that the polymer concentration of PA 6 and the employed electrospinning variables were the optimum conditions. The particles of DR 167 and DR 167-B exhibited aggregation on the fiber surface of the dyed PA-NWNF (Figs. 1c and 2b), meanwhile DR 167 (Fig. 1b) showed a higher aggregation in comparison to DR 167-B (Fig. 2a). Such superior aggregation can be correlated with the larger size of DR 167 particles [48]. Moreover, DR 167-B showed a scaly surface of a few fibers which may be due to the aggregation of scale-like particles. Thus, a high reduction in the surface area of the PA-NWNF (161.90 m²/g) occurred as a result of coating the pores and increasing the fiber diameter of PA-NWNF (Fig. 3). In contrast, the DR 167-C particles (Fig. 2c) became homogenously dispersed and had smaller sizes on the fiber surface (Fig. 2d). This confirms that the enhancement of its dispersion in water is correlated with its smaller particle size [48]. Thus, the dyed DR 167-C has the lowest surface area reduction (227.60 m²/g), as shown in Fig. 3.

Usually, PA 6 exhibited two crystalline phases, α (monoclinic, high crystallinity) and γ (pseudo-hexagonal, low crystallinity) [49]. PA 6 molecules undergo hydrogen bonding in both anti-parallel and parallel chain arrangements for the amide groups producing the α- and γ-polymorph phases at 2θ = 20.3 and 23.9°, as well as 21.5, respectively [50‐54]. XRD spectra of the PA-NWNF and dyed PA-NWNF together with the applied dyes were illustrated in Figs. 4, 5, 6. The PA-NWNF spectrum had a dominant broader diffraction peak at 2θ ≃ 21.55° with a d-spacing value of 4.12 Å, confirming the phase transition of PA 6 to the γ-phase during the electrospinning (Figs. 4, 5, 6) [55]. This is due to the rapid evaporation of solvent during the electrospinning processing of PA 6. Hence, the development of the γ-phase appeared, and the α-phase disappeared [55‐58].

After dyeing with DR 167, DR 167-B, and DR 167-C, the complete formation of the α-phase was observed, at 2θ ≃ 20.27 and 23.95° with the d-spacing value of 4.37 and 3.74 Å, respectively, beside the γ-phase at 21.5° with the d-spacing value 4.12 Å (Figs. 4, 5, 6). This can be attributed to the slow heating during the dyeing process, where the existent γ-phase of PA-NWNF melts and then the molecules are recrystallized into α-form (thermodynamically stable) together with γ-form [59‐61]. Furthermore, another new diffraction peak at 2θ ≃ 32.64° with a d-spacing value of 2.77 Å is inherited from the applied dyes with slight shifting in the distinguishing 2θ (Figs. 4, 5, 6). This was assured by matching peak intensity of dyed PA-NWNF with that of its applied dye (Figs. 4, 5, 6).

The effects of dying process on both chemical structures of PA-NWNF and applied dyes were examined by FTIR analysis and illustrated in Figs. 7, 8, 9. As well the functional groups of PA-NWNF were gathered in Table 1. The spectra of PA-NWNF revealed that the absorption bands at 1640.38 and 1543.11 cm⁻¹ (amide I and amide II, respectively) are related to the nature of polyamide materials [62], whereas the peaks at 3297.1 and 3087.13 cm⁻¹ are correlated to N–H stretching vibration mode (Free and hydrogen bonded N–H stretching, respectively) [63‐66]. The asymmetrical and symmetrical stretching, as well as the scissoring vibration of the CH₂ group in the polyamide backbone, can be assigned to the absorption bands at 2931.80, 2860.92 and 1462.71 cm⁻¹, orderly [67, 68]. Furthermore, the vibration of CH₃ groups related to terminal chains and C–N stretching were observed at 1262.33 and 1075.00 cm⁻¹, respectively [69]. Also, the adsorption band of the C=O group overlapped with the band of amide I at 1640.38 cm⁻¹ [50, 69]. We can also see the γ-phase of PA-NWNF predominating through recorded bands at 1437.46 cm⁻¹ (CH₂ bending vibration) [70‐72]. Moreover, twist-wag vibration of the CH₂ group was observed at 1368.48 and 1234.77 cm⁻¹ [71, 73]. Also, band at 1168.08 cm⁻¹ revealed the bending of the CH₂ group [74, 75]. The vibration band at 973.90 cm⁻¹ is correlated to CO–NH in γ-plane [62, 71, 73]. This is due to the γ-phase transition of PA 6 during the electrospinning process [55, 57, 73, 76]. In contrast, the α-peaks observed in the FTIR results of PA-NWNF reflect the pseudo-crystalline form of α-phase. Figures 7, 8, 9 also showed the FTIR spectra of applied dyes, illustrating differences in the intensities with slight peak shifting, as a function of milling duration. While DR-167 had the β-crystal form, DR 167-B had an overall amorphous nature with initial transformation to the α-crystal form and the complete transformation to the α-crystal form occurred in DR-167-C. After the dyeing process, the PA-NWNF exhibited the two α- and γ-phases. This was corroborated by the emergence of new α peaks at 831.80, 928.86, 958.55, 1027.20, and 1476.21 cm⁻¹, in addition to the presence of γ-phase peaks at 712, 974.21, and 1437.74 cm⁻¹ [54, 70, 77‐79]. In addition, the bands of amides V and VI were related to α- form at 690.40 and 578.00 cm⁻¹, respectively [56, 80]. Also, it can be observed that the amide II peak shifted from 1543.11 to ≃1540.40 cm⁻¹. The shifting of the amide II peak towards lower frequency indicates the formation of thermodynamically stable α-phase in PA-NWNF [63, 73, 80]. These findings matched with XRD results. The interaction between the dyes and PA-NWNF is not evident in the IR spectra (Figs. 7, 8, 9). This may be due to the small quantities of dyes employed in the experiments, as previously reported in other studies [81‐83].

The effects of dye concentration on color yield were studied at 100 °C for 60 min. Five concentrations (3, 5, 10, 15, and 20% o.w.f) of each dye were utilized and the color yield property was measured in terms of K/S values, as shown in Fig. 10. The results showed an increase in the dye concentration corresponding to an increase in K/S values for all dyes. Comparing between the K/S values of dyed samples, it was found that a 5% concentration of DR 167-C produced color yield almost equal to a 15% concentration of DR 167 and DR 167-B. In addition, 15% of DR 167-C produced about 55.1 and 33.18% color yield higher than DR 167 and DR 167-B, respectively. This demonstrates that DR 167-C dye exhibited an excellent ability to build color on PA-NWNF. Dyeing synthetic textiles with disperse dyes is generally based on the swelling of the amorphous region in the fiber under high temperature and pressure, which makes adsorbed dye particles on the fiber surface easily penetrate into the fiber inside [42, 84]. Accordingly, dye particles larger than the fiber diameter cannot penetrate into the fiber. Therefore, the difficulty of dyeing nanofibers with disperse dyes is not only because of the high surface area of nanofibers but also the large diameters of the conventional dye particles (average size ≃ 42.2 µm), which are larger than nanofibers diameters. Hence, it cannot penetrate into the fibers and thus be removed with washing. So, the increase in K/S of DR 167-C is attributed to the large number of small particles of dye that penetrated into nanofibers. In contrast, the decrease in K/S of DR 167 and DR 167-B is attributed to the small number of dye particles much smaller than the nanofiber diameter. This was confirmed by results of the SEM (Figs. 1b and 2a). 15% was chosen as the optimum dye concentration.

To optimize the dyeing temperature, PA-NWNF samples were dyed with 15% dye concentration of each dye for 60 min. Dyeing was carried out at five different temperatures (R.T., 60, 80, 100, and 120 °C). Figure 11 displayed the color yield of dyed PA-NWNF as a function of temperature. The color yield of dyed DR 167-C increased rapidly as the dyeing temperature increased from room temperature to 80 °C, whereas the color yield of dyed DR 167 and DR 167-B improved slowly at this temperature rising region. This indicates that particles of DR 167-C dye need low activation energy of diffusion to penetrate the polymer chains compared to the others [85]. From 80 to 100 °C, all dyes showed a steep increase in color yield. This confirms the dye's high dispersion ability in the absence of dyeing auxiliaries, which is consistent with SEM results (Fig. 2d). Furthermore, increasing the dyeing temperature after 100 °C resulted in a decrease in dye substantivity, resulting in a lower color yield [86, 87]. So, the dyeing is optimized at 100 °C.

To investigate the effects of dyeing time on the color yield of dyed PA-NWNF, the dyeing process was carried out using 15% dye concentration at 100 °C for five different dyeing times (15, 30, 60, 90, and 120 min) as presented in Fig. 12. For all dyes, it was noticed that with increasing the time, the color yield increased and beyond 60 min no further increase in color yield was observed. However, DR 167-C achieved the highest color yield compared to DR 167 and DR 167-B. More specifically, at 60 min, the K/S of 3.04 was attained for dyed DR 167-C, while the K/S of 2.2 and 1.96 were achieved for dyed DR 167 and DR 167-B, respectively. This was attributed to the rapid movement of small particles of DR 167-C dye, which diffused more quickly into PA-NWNF than the other dyes with larger particle sizes. Thus, the dyeing process was optimized for 60 min.

The resistance of dyed fabrics to color bleed against various factors, such as washing, perspiration, and light, is essential. At previously optimized conditions of the dyeing process, the color fastness to washing and perspiration (i.e., shade change and staining), furthermore, the light fastness of dyed PA-NWNF with the addressed dyes were evaluated, as shown in Table 2. The results demonstrated that the color fastness to washing has a very good grade (i.e., shade change) for DR 167-C, while it decreased to a good grade for DR 167-B and DR 167-C. All dyes showed a moderate to excellent range of colorfastness to washing in terms of staining, where staining rates increased for DR 167 and DR 167-B dyes. This finding matched with the data given in Figs. 1 and 2, since the highly dispersed and penetrated particles were obtained in dyed DR 167-C, suggesting that more dye molecules offered high resistance to washing on the test specimen. However, DR 167 and DR 167-B that aggregated on PA-NWNF surface can be easily removed during the test. Moreover, the ratings of perspiration fastness (i.e., shade change and staining) were the same for all dyes. The color change rate was recorded as 4/5, which is an excellent rate. In addition, the color staining on multifiber fabric was rated from moderate to very good. Generally, polyamide and acetate were always found to be the highly stained parts of the multifiber fabrics dyed PA-NWNF. The fastness to light showed the same ratings for all dyes, where overall light fastness had moderate properties.