From Regenerated Wood Pulp Fibers to Cationic Cellulose: Preparation, Characterization and Dyeing Properties

The heterogenous modification of the regenerated wood pulp cellulose fibers was performed with the aim of fine-tuning target surface properties of the fibers, namely to enhance their dye uptake capacity. These regenerated cellulose fibers (CellReg) were chosen for their uniform morphological, mechanical and physical properties, biodegradability, COneutrality and low density [ 1 ], as well as their high reactivity, while GTAC was selected as the cationizing agent for making possible the improvement of the reactive dyeing of the fibers. Therefore, the modification was focused on the cationization of CellReg by grafting cationic pending groups via the reaction with a tetraalkylammonium derivative containing reactive epoxy moieties. The cationization reaction, displayed in Figure 1 A, was performed in a THF-water (90:10) mixture, since this was reported to enhance the cationization efficiency and the integrity of the cellulose fibers by suppressing the occurrence of side-reactions, namely the hydrolysis of GTAC in aqueous alkaline conditions [ 32 ]. Three distinct GTAC/AGU molar ratios, namely 1.5:1, 3.0:1 and 4.5:1, were studied to modulate the degree of substitution of the modified cellulose fibers [ 33 ]. The non-modified and modified regenerated cellulose samples were thoroughly characterized regarding their DS based on elemental analysis, molecular structure by FTIR-ATR and solid-stateC NMR spectroscopy, crystallinity by XRD, surface charge by zeta potential measurements, surface wettability by contact angle (with water) measurements, morphology by SEM, thermal stability by TGA and dye uptake by dyeing the cellulosic samples with a commercial reactive dye (i.e., Remazol brilliant Orange 3R), as discussed in the following paragraphs.

The crystallinity of the non-modified and modified regenerated cellulose samples was evaluated by XRD as illustrated in Figure 2 C. As foreseen, the CellReg sample exhibits the diffraction pattern of cellulose II consisting of a small broad peak centered at 2= 12.2° corresponding to the (10) diffraction plane and two overlapped broad peaks at 2= 20.2° and 21.5° allocated to the (110) and (020) diffraction planes, which are characteristic of regenerated cellulose [ 13 44 ]. The diffraction pattern of the regenerated cellulose is different from that of cellulose I [ 37 ] because the process of dissolution followed by regeneration changes the crystal structure of this polysaccharide [ 18 44 ]. The XRD diffractograms of the modified regenerated cellulose samples ( Figure 2 C), i.e., CellReg/GTAC_1.5, CellReg/GTAC_3.0 and CellReg/GTAC_4.5, are similar to that of CellReg, although showing some broadening of the peaks and a decrease in the intensity of the (110) and (020) reflections relative to that of (10), both observed for increasing DS values, which is consistent with a small increase in the amorphous domains of the samples and their macroscopic aspect. This slight increase in the amorphous character of the modified samples is in consonance, not only with the data described by Ho et al. [ 19 ] for GTAC-modified nanofibrillated cellulose, but also for other polysaccharides (e.g., chitosan) modified with GTAC [ 45 ].

Regarding the regenerated cellulose samples modified with the different GTAC/AGU ratios (1.5, 3.0 and 4.5), apart from the typical absorption bands of cellulose, the three spectra show the emergence of a new band at ca. 1476 cm, which is assigned to the trimethyl groups of the quaternized ammonium moiety of GTAC [ 32 42 ]. Thus, this absorption band confirms the presence of the grafted moieties and, therefore, the successful cationization of the regenerated cellulose fibers. As anticipated, a higher GTAC/AGU ratio translates into an increase in the intensity of the absorption bands characteristic of the substituent, which is in accordance with the DS values ( Table 1 ).

The cationization of the regenerated cellulose fibers was further confirmed by FTIR-ATR and solid stateC NMR spectroscopy, as shown in Figure 2 A,B. The FTIR-ATR spectrum of the non-modified regenerated cellulose (CellReg, Figure 2 A) show the typical absorption bands of cellulosic substrates at ca. 3320 cm(O–H stretching vibration of the primary and secondary OH groups), 2890 cm(C–H stretching vibration), 1306, 1366 and 1420 cm(O–H in-plane bending vibration bands of the primary and secondary OH groups), 1154 cm(C–O–C antisymmetric stretching vibration of the glycosidic bonds) and 1020 cm(C–O stretching vibration) [ 37 39 ].

The ensuing cellulose samples are shown in Figure 1 B and their distinct macroscopic aspect is the first indication that the cationization reaction predictably took place to different extents, leading to distinct DS values ( Table 1 ). While for the CellReg/GTAC_1.5 the fibrous aspect of the material was preserved, for the other samples, namely CellReg/GTAC_3.0 and CellReg/GTAC_4.5, a reduction of the fibrillar appearance was detected, particularly for the sample with the highest DS where a film-like material was obtained. The DS values of the modified fibers increased with the increasing GTAC/AGU ratio from 0.13 ± 0.004 for CellReg/GTAC_1.5 to 0.33 ± 0.002 for CellReg/GTAC_4.5. Therefore, the resultant DS values can be modulated by varying the GTAC/AGU ratio [ 19 36 ], and are comparable with the data described by Courtenay et al. [ 33 ], who achieved DS values of 0.188 and 0.230 for GTAC/AGU molar ratios of 1:1 and 3:1, respectively.

The cationization reaction was investigated with the goal of preserving the bulk properties of the fibers, but at the same time adequate to attain enhanced surface properties, namely superior dye uptake capacity. Therefore, the ratio of GTAC to AGU was varied from 1.5 (CellReg/GTAC_1.5) to 3.0 (CellReg/GTAC_3.0) and 4.5 (CellReg/GTAC_4.5), as listed in Table 1

The surface wettability of the non-modified and modified regenerated cellulose samples was studied by contact angle measurements with water [ 49 ] via the sessile drop method. As depicted in Table 1 , the non-modified regenerated cellulose fibers (CellReg) revealed the lowest contact angle value of 50.7 ± 6.3°, in accordance with data recorded by Pang et al. [ 39 ] for regenerated cellulose from cotton linters using different ionic liquids as solvents. In the case of the modified cellulose samples ( Table 1 ), the contact angles increased from 65.7 ± 3.4° for CellReg/GTAC_1.5 to 71.5 ± 2.2° for CellReg/GTAC_3.0 and 82.5 ± 3.8° for CellReg/GTAC_4.5. These results provide evidence that the cationization reaction slightly increased the hydrophobicity of the modified regenerated cellulose fibers, although remaining in the hydrophilic range (10° << 90°). This might be associated, on one hand, with the reduction of the fibrillar nature of the cellulosic materials, particularly for the sample with the highest DS (CellReg/GTAC_4.5) where a film-like material was obtained ( Figure 1 B); and, on the other hand, with the presence of the more hydrophobic methyl (−CH) groups of GTAC at the surface of the cellulosic materials, alongside the more hydrophilic hydroxyl (−OH) moieties [ 50 ]. In fact, a similar trend was described by Kallio and co-workers, who have shown that the water contact angle of cellulose nanofibrils increased from 32.6° to 67.7° after cationization with GTAC (DS of 0.35) [ 51 ], which, as conjectured by the authors, is due to the fact that the hydrophobic methyl groups of the (CHmoieties do not establish hydrogen bonds with water, thus, rendering the surface of the cellulose nanofibrils slightly less hydrophilic than the pristine cellulose nanofibrils [ 50 ].

The surface charge of the non-modified and modified regenerated cellulose samples was assessed by zeta potential measurements. According to Table 1 , CellReg presents a negative apparent zeta potential value of −15.4 ± 5.0 mV, as a result of the presence of carboxyl and hydroxyl groups [ 46 47 ], in accordance with, e.g., the value reported for Lyocell fibers of ca. −15 mV (at pH 7) [ 48 ]. After cationization, the surface charge turns positive with apparent zeta potential values of 8.4 ± 3.9 mV for CellReg/GTAC_1.5, 15.2 ± 3.2 mV for CellReg/GTAC_3.0 and 26.9 ± 5.5 mV for CellReg/GTAC_4.5. The change in the surface charge of CellReg after modification with GTAC reflects the success of the cationization and the different DS values, which is consistent with the charge reversal observed by Odabas et al. [ 32 ] for cationically modified bleached Kraft pulp and by Courtenay et al. [ 33 ] for cationically modified cellulose nanofibrils.

The morphology of the non-modified and modified regenerated cellulose samples was examined by optical and SEM microscopy as depicted in Figure 3 . The optical micrographs ( Figure 3 A) evince a fibrous morphology for all samples but with a considerable reduction of visible individualized fibers in the case of CellReg/GTAC_4.5, in line with their macroscopic appearance shown in Figure 1 B. In fact, a closer look at the optical micrograph of CellReg/GTAC_4.5 shows a large decline in the number of available fibers and a tremendous lowering of their length size, as well as a decrease in the amount of visible light that passes through the sample, which is congruous with a cellulose derivative with a film-like morphology. This is further corroborated by the SEM micrographs of the surface of the samples presented in Figure 3 B. After modification of CellReg with GTAC, only the CellReg/GTAC_1.5 sample with the lowest DS (0.13 ± 0.004) retained a total fibrillar morphology. On the other hand, the CellReg/GTAC_3.0 and CellReg/GTAC_4.5 samples exhibit a film-like morphology with a smooth surface with only a few visible fibers. This film-like morphology is a clear indication of the higher DS values of those two samples, again in accordance with their visual aspect ( Figure 1 B) and XRD data ( Figure 2 C). This was anticipated to some extent since, as the DS increases, the modification starts to occur beyond the surface of the fibers, thus starting to destroy the characteristic fibrillar microstructure of the cellulose fibers and originating a cellulose derivative with a film-like morphology. Therefore, given that the goal was to fine-tune specific surface properties without changing the bulk properties of the fibers, the CellReg/GTAC_1.5 sample is, thus far, the most promising one given its fibrillar morphology.

Overall, these results clearly prove that the modified regenerated cellulose samples have lower thermal stability than the non-modified regenerated cellulose fibers, which agrees with the data reported for cationic cellulose derivatives [ 25 52 ]. Despite the smaller thermal stability, the modified regenerated cellulose samples are still thermally stable up to about 220 °C, which demonstrates that these cationic fibers can, if necessary, withstand temperatures below 100 °C during textile manufacturing processes [ 10 12 ].

Regarding the CellReg/GTAC_4.5 sample with a DS value of 0.33 ± 0.002, the thermograms followed a one-step weight-loss with a maximum decomposition temperature at 280 °C and a total mass loss of about 81% at 800 °C ( Figure 2 D). When compared with the other two modified fibers (i.e., CellReg/GTAC_1.5 and CellReg/GTAC_3.0), this single step degradation profile is indicative of a higher DS and, hence, of the complete decomposition of the cationic cellulose, as noted by Yan et al. [ 36 ].

The thermal stability of the non-modified and modified regenerated cellulose samples was assessed by TGA and the corresponding thermograms are shown in Figure 2 D. Apart from the volatilization of water below 100 °C, the regenerated cellulose fibers exhibited a one-step weight-loss degradation profile in the range 273–370 °C with a maximum decomposition temperature of 344 °C and a total mass loss of ca. 89% at 800 °C, aligned with data described elsewhere [ 39 ]. On the other hand, the CellReg/GTAC_1.5 and CellReg/GTAC_3.0 samples displayed a two-step weight-loss degradation profile, alongside the initial loss allocated to the release of adsorbed water below 100 °C (loss of ca. 7 wt %). In relation to CellReg/GTAC_1.5, the first degradation stage took place at 243–319 °C with a maximum decomposition temperature of 288 °C, while the second one occurred at 319–362 °C with a maximum decomposition temperature of 331 °C ( Figure 2 D). For the CellReg/GTAC_3.0 sample, the first stage emerged at 243–314 °C with a maximum decomposition temperature of 286 °C, and the second stage appeared at 314–366 °C with a maximum decomposition temperature of 340 °C. According to Li et al. [ 52 ], the first degradation step is related to the thermal decomposition of the GTAC functional groups, whereas the second one is mainly linked with the decomposition of the cellulose backbone. However, this is questionable because the amount of material that was degraded in the first stage is higher than that in the second stage for both fiber samples. In fact, as hypothesized by Odabas et al. [ 25 ], the first stage is most likely associated with the simultaneous degradation of the substituent groups and the cellulose functional moieties to which they are bonded.

3.4. Dye Uptake Tests

Bacillus stratosphericus

SCA1007 [

The goal of the modification of the regenerated cellulose fibers via cationization reaction was to improve their dye uptake capacity envisioning their application as alternative textile fibers. Therefore, dye uptake tests were performed using the dischargeable diazo dye Remazol Brilliant Orange 3R (Reactive Orange 16), which is a water soluble reactive anionic dye that fixates into the fibers, thus enduring harsher conditions and creating a more permanent color [ 53 ]. Noteworthy from an environmental impact perspective is the fact that this reactive dye can be biologically degraded by using, for instance, the bacteriumSCA1007 [ 54 ]. These dye uptake tests will deliver information on the dye exhaustion (E) and dye fixation (F), which are, respectively, the amount of dye that is taken up by the fibers, and the amount of dye that is fixated into the fibers [ 31 53 ].

−2) dyed with Reactive Orange 16 (F = 63.1% ± 1.5%) and by Björquist et al. [® Brilliant Blue E-BRA Macrolat reactive dye (E≈ 96%).Figure 4 A shows the dye exhaustion and dye fixation data obtained for the non-modified and modified regenerated cellulose. As expected, the sample that shows the lowest dye exhaustion and dye fixation values is the non-modified regenerated cellulose (CellReg). This means that 66.8% ± 3.4% (E) of the total amount of dye was taken up by the fibers (and concomitantly ca. 33.2% remained in the dyebath solution), of which only 45.5% ± 5.0% (F) was fixated into the CellReg fibers. When compared with the literature, these values are lower than those obtained by Wang et al. [ 20 ] for cotton fibers dyed with Remazol Brilliant Orange 3R (E = 71.6%, F = 74.9%), by Shu et al. [ 55 ] for cotton fabrics (176 g m) dyed with Reactive Orange 16 (F = 63.1% ± 1.5%) and by Björquist et al. [ 23 ] for regenerated cellulose fibers from cotton waste pulp dyed with LevafixBrilliant Blue E-BRA Macrolat reactive dye (E≈ 96%).

Concerning the cationized regenerated cellulose, all three samples exhibited higher dye exhaustion and dye fixation abilities than CellReg, although the values of these two parameters for the three modified samples are not significantly different, reaching what seems to be maxima values of 89.3% ± 0.9% (E) and 80.6% ± 1.3% (F) for the CellReg/GTAC_1.5 ( Figure 4 A). This implies that 89.3% of the total amount of dye was taken up by the cationic fibers (with ca. 10.7% remaining in the dyebath), of which 80.6% (F) was fixated to the CellReg/GTAC_1.5 fibers. Therefore, the cationization of the regenerated cellulose fibers improved both the dye exhaustion and dye fixation because this anionic reactive dye can simultaneously interact with the hydroxyl and the tetraalkylammonium groups of the modified cellulose samples [ 21 ]. Furthermore, it is important to mention that CellReg/GTAC_1.5 generated the maxima values of dye exhaustion and dye fixation, despite displaying the lowest DS value, probably because of the conjunction of its fibrillar structure, greater surface area and higher availability of the surface cationic groups when compared with the other two modified samples with a film-like morphology.

−2) with C.I. Reactive Blue 5, which is higher than the value obtained in the present study.

If one compares these results with literature data, it is clear that they are superior to those achieved by Wang et al. [ 20 ] for cationic cotton fibers, where only 84.4% of the reactive dye (Remazol Brilliant Orange 3R) was exhausted and 63.5% was fixated. Khatri et al. [ 56 ] showed that, when compared with the non-modified cellulose nanofibers, the cationization of electrospun cellulose nanofibers with GTAC generated higher dye fixation values independently of the reactive dye used, namely C.I. Reactive Black 5 (F = 85–91%), C.I. Reactive Red 195 (F = 85–89%) and C.I. Reactive Blue 19 (F = 81–82%). Furthermore, Correia et al. [ 21 ] extensively studied the effect of four different (poly)electrolytes containing quaternary ammonium groups, including GTAC, on the dyeing process of cotton with the reactive red 195 dye. The authors concluded that the cationization of cotton with GTAC was the most advantageous in terms of dye fixation and exhaustion since it generated the lowest mass of residual dye in both exhaustion and washing baths [ 21 ]. More recently, Wang et al. [ 57 ] reported a dye fixation of 95% for cationic cotton fabrics manufactured via a single-step pad-steam cationization and dyeing process of plain-woven pure cotton fabric (126 g m) with C.I. Reactive Blue 5, which is higher than the value obtained in the present study.

The dye exhaustion and fixation data are further corroborated by the macroscopic appearance ( Figure 4 B), as well as the colorimetric coordinates ( Table 2 ) of CellReg, CellReg/GTAC_1.5, CellReg/GTAC_3.0 and CellReg/GTAC_4.5 after the dye uptake tests. As portrayed in Figure 4 B, the color of the non-modified fibers after the dyeing tests is much less intense than those of the modified fiber samples, and the color intensities of the modified fibers appear similar among them, which validates the fact that the values of the dye exhaustion and dye fixation of the modified fiber samples are not significantly different among them ( Figure 4 A). The optical micrographs of the cellulose samples after dyeing ( Figure 4 C) reveal that the dyeing process with a reactive dye did not affect their morphology since their micrographs are analogous to those portrayed in Figure 3 A, the only difference being in the orange/reddish color of the samples after dyeing.

p

< 0.0001) and, when compared with CellReg, the modified fibers present higher values except for the CellReg/GTAC_4.5. Moreover, there is a decrease in the ISO brightness of the cationic fibers with higher DS values, credited to their translucent aspect, which translates into a lower amount of reflectance of blue light, viz. lesser brightness [

The colorimetric coordinates of the fiber samples before dyeing ( Table 2 ) show that CellReg exhibits the highest values of L* (83.9 ± 2.0, lightness), a* (2.3 ± 0.3, redness) and b* (17.5 ± 1.0, yellowness) coordinates; and that their combination points to straw colored fibers, in accordance with the macroscopic aspect of these unbleached regenerated pulp fibers ( Figure 1 A) and similarly to data reported for non-regenerated unbleached kraft pulps of Eucalyptus globulus [ 58 ]. Regarding the modified samples before dyeing ( Table 2 ), there is a negative contribution to the a* coordinate (red (+a*) to green (−a*)) and a reduction of the b* (yellowness/blueness) and L* (white (100) to black (0)) coordinates with the cationization of the regenerated cellulose with GTAC. In terms of total color difference from the CellReg sample, this parameter increased with the rising DS of the modified fibers, in line with the macroscopic appearance of the fibers displayed in Figure 1 B. The ISO brightness values of the non-modified and modified fibers are significantly different (< 0.0001) and, when compared with CellReg, the modified fibers present higher values except for the CellReg/GTAC_4.5. Moreover, there is a decrease in the ISO brightness of the cationic fibers with higher DS values, credited to their translucent aspect, which translates into a lower amount of reflectance of blue light, viz. lesser brightness [ 28 ].

After the dyeing uptake tests with the reactive dye, the colorimetric coordinates of the non-modified and modified regenerated cellulose fibers changed for superior values of the coordinate +a* (redness) as outlined in Table 2 , which was expected given the orange/red color of the reactive dye (Remazol Brilliant Orange 3R or Reactive Orange 16). The same trend was disclosed, for instance, by Correia et al. [ 21 ] for cotton and cationic cotton fibers dyed with Reactive Red 195. The dyed CellReg is the fiber sample with the highest values of the coordinates L* and b*, and ISO brightness, which concurs with their lowest dye exhaustion and dye fixation values ( Figure 4 A) as well as their visual aspect ( Figure 4 B). When comparing the dyed non-modified with the dyed modified regenerated cellulose, it is evident that the values of lightness (L*) diminished with increasing DS values, except for CellReg/GTAC_4.5 whose value is not significantly different from that of CellReg/GTAC_3.0. This means that, after dyeing, the CellReg/GTAC_3.0 and CellReg/GTAC_4.5 samples seem to have a darker color and less brightness than the other samples. Moreover, the total color difference decreased with the increase in DS, which concurs with the higher values of dye exhaustion and fixation of CellReg/GTAC_1.5 ( Table 2 ).

Worth noting here is the fact that, despite the similar dye exhaustion and dye fixation data between the modified cellulose samples (not statistically different, Figure 4 A), it is perceptible that the colorimetric coordinates slightly differ among them, particularly for the CellReg/GTAC_1.5 whose coordinates (L*, a*, b*) are statistically different. Since the objective of the present study was to adjust target surface properties without altering the bulk properties of the fibers, the CellReg/GTAC_1.5 sample is indeed the most promising one in terms of retaining the fiber morphology, while having adequate surface charge, surface wettability and dye uptake capacity.