3.1. Characterization of viscose with deposited chitosan
Viscose, as cellulose fiber, was firstly converted to dialdehyde cellulose using sodium-periodate, in incremental oxidation time (60 and 120 min), to promote the binding of chitosan via formation of Schiff base between aldehyde groups of cellulose and amino groups of chitosan (Kim et al. 2017; Korica et al. 2019).
However, it must be underlined that this is one of mechanisms for binding of CS to cellulose, the other one being hydrogen bonding between OH groups of CS and cellulose. After oxidations, chitosan was also physically deposited onto fibers surface and SEM/EDX analysis confirmed the presence of CS on viscose fibers (Fig. 1). Surface EDX analysis detected the nitrogen in samples with CS (Table 1).
The measurement of free positively charged -NH3+ groups after deposition of CS onto non-oxidized and oxidized CV was performed with CI acid orange 7 dye test (Fras Zemljič et al. 2009; Strnad et al. 2010; Sauperl et al. 2014) and results are given in Table 2, along with the results of quantitative determination of aldehyde groups in fibers; at the same time, estimation of interaction between functionalized viscose and CI acid orange 7 was determined knowing that pristine one does not bind dye at all, as can be seen in Fig. S1.
There was no clear correlation between quantity of aldehyde groups and free amino groups available to bind CI acid orange 7 dye.
3.2. Dyeing of functionalized viscose with NP4
Dyeing with NP4 was performed with 0.5 % o.w.f. (on the weight of fabric) in the dyeing system containing acetone and water (1:99), corresponding to 100 mg/L concentration. It should be pointed out that in the previous investigation (Kramar et al. 2014) dyeing was performed in methanol:water solution (50:50), and an increase in the UV-VIS absorbance after dyeing occurred, thus indicating that initially not all dye was dissolved until the temperature was elevated during dyeing. In this work, improved solubility of dye extract in the solvent system acetone:water was found. Comparative UV-VIS spectra in different solvent systems is given in Fig. S2.
The absorbance peak at 535 nm detected in UV-VIS corresponds to a previously reported prodigiosin (Kramar et al. 2014; Ren et al. 2018). After the dyeing, the absorbance and pH of residual solutions was measured (Fig. 2). The deposition of chitosan onto cellulose significantly increases exhaustion of the dyebath, even though there is no clear correlation between free –NH3+ groups in samples with CS (Table 2) and exhaustion of the dyebath. Furthermore, the oxidation itself leads to a better exhaustion (samples CV60 and CV120) which suggests that aldehyde groups in cellulose are involved in binding the NP4.
The pH of the dye solution was also measured before and after dyeing (Fig. 2b). From starting pH 5.5, after dyeing of pristine cellulose, an increase of value to pH 6.2 was measured. In all other samples, a decrease of pH was measured following the trend that the longer time of cellulose oxidation induced a greater decrease of pH and the addition of chitosan decreased pH even further.
Color coordinates of dyed samples were measured using spectrophotometer under illuminant D65 and standard 10° observer in CIELab color space (Table 3).
Parameter L* corresponds to the lightness of material (in the range 100-0 white-black), a*corresponds to red-green axis (from positive to negative, respectively) and b*corresponds to yellow-blue axis (from positive to negative, respectively). K/S value denotes color strength, derived using Kubelka-Munk equation (Sharma 2003) from reflectance R measured at the 535 nm.
Oxidation of viscose prior dyeing provides higher K/S values and lower L* parameter, indicating stronger color and darker samples. Samples with chitosan provided darker (lower L*), redder (higher a*) and less blue color (lower b*).
Since there is a very similar exhaustion of the dye solution in samples CV60/CS and CV120/CS and similar color strength, and evaluating benefit from shorter oxidation time, all further analysis was performed on samples oxidized for 60 min.
3.3. Structural analysis of functionalized viscose dyed with NP4
ATR-FTIR analysis of viscose before (CV) and after addition of chitosan (CV/CS) and dyeing (CV/CS/NP4) in Fig. 3 shows the typical cellulose II peaks, vibration of the β-glycosidic ring or deformation at C1 in cellulose II at 890 cm-1, C–O stretching of cellulose II at 1035 cm-1, CH2 wagging vibration at 1316 cm-1 and C-H deformation in cellulose II at 1364 cm-1; a broad band at 1640cm-1 originates from absorbed water, 1735 cm-1 from carbonyl signal (C=O), 2880 cm-1 represents C–H stretching in cellulose II and amorphous cellulose, and, a broad band in the region of 3000-3600 cm-1 originates from hydrogen-bonded OH groups in cellulose (Schwanninger et al. 2004; Široký et al. 2010; Kim et al. 2017; Korica et al. 2019).
The addition of chitosan (CV/CS) causes the shift from 890 cm-1 to peak at 893 cm-1, which is known from literature as CS peak (Kim et al. 2017; Korica et al. 2019). The band at 989 cm-1 corresponds to C–O valence vibration at C6 atom (Schwanninger et al. 2004; Korica et al. 2019) and a shift to 993 cm-1 indicates that OH group on C6 atom is involved in interaction with chitosan during deposition, probably via hydrogen bonding. Additional specific peaks of chitosan have been reported in other publications, 1560 cm-1 (stretching vibration of amino groups), 1590 cm-1 (N–H bending vibration of amide II) and 1651-1654 cm-1 related to amide I (Kim et al. 2017; Yang et al. 2018).
However, in Fig. 3, some of these peaks are overlapped with band of absorbed water at 1640 cm-1. In CV/CS, the 2880 cm-1 (CH stretching in cellulose II) is also shifted towards higher wavenumber, 2890 cm-1, while intensity of the band in the region 3000-3600 cm-1 decreased and a shift occurred from 3270 cm-1 to 3330 cm-1 confirming the formation of intermolecular hydrogen bonds between CS and CV, as it was reported in other publications (Yang et al. 2018; Korica et al. 2019).
The peak corresponding to carbonyls (1735 cm-1) does not change after addition of CS, even though the Schiff base formation between dialdehyde cellulose and chitosan should occur as a mechanism of crosslinking (Strnad et al. 2010; Sauperl et al. 2015; Kim et al. 2017; Yang et al. 2018; Pratama et al. 2019). This probably means that in this work dominant mechanism of cellulose-chitosan interaction is physical deposition, the hydrogen bonding as seen in the region 3000-3600 cm-1, and probably in some part Schiff’s base formation, but this is overlapped with a broad band at 1640 cm-1.
Furthermore, quantitative estimation of the functional groups content (Table 2) shows that quantity of aldehydes per g of fiber is much higher than that of amino groups; therefore a significant change in this peak at ATR-FTIR spectra could not be expected. ATR-FTIR did confirm that, beside chitosan, aldehyde groups are involved in interaction with NP4. The dyeing with NP4 causes in all samples a decrease of the peak at 1735 cm-1. The absence of this peak indicates that interaction occurred between NP4, i.e. prodigiosin as the main constituent of crude extract NP4 (Kramar et al. 2014) and aldehyde groups.
It should be stressed that characteristic vibrations of the NP4 in ATR-FTIR of crude extract (Fig. 4) could not be observed in spectra of dyed viscose fibers (Fig. 3), since they are overlapped by the cellulose peaks.
ATR-FTIR spectrum of NP4 extract (Fig. 4) exhibits characteristic vibrations (in cm-1) confirming the presence of prodigiosin based constituents and is in accordance with literature: 3250 (N–H stretching), 2923 (asymmetric CH2 stretching), 2852 (symmetric CH2 stretching), 1613 (C=C stretching + CNH bending), 1413 (CH3 bending), 1239 (ring stretching and bending), 1152 (C–O stretching), 993 (ring stretching and bending) (Jehlička et al. 2016). The NP4 extract is deprotonated at pH 10, i.e. proton from free pyrrol ring is removed, thus causing the rearrangement of electron density within the molecule which is observed as the bathochromic shift of absorption bands in UV-VIS spectra followed by the change of solution color from pink to blue (Fig. 4).
Taking into account the molecular structure of prodigiosin (Fig. 5) one can note that the presence of N–H groups makes this molecule sensitive to pH-value.
Generally, pKa value of prodogisions (Rizzo et al. 1999; La et al. 2007) varies between 7.2 and 7.98 indicating that during the dyeing (pH 5.5) pigment adopts protonated form. Due to the pronounced propensity of the protonated prodigiosin to bind anions via hydrogen bonds with all three pyrrole NH sites (Davis 2010), it is justified to assume that NP4 forms hydrogen bonds with aldehyde groups of cellulose. Having in mind that fibers with chitosan also bear positively charged -NH3+ at pH 5.5, the possible mechanism of NP4 binding is depicted at Fig. 6.
The orientation of the molecule is driven by the demand to minimize energy of the repulsion between two positive charges. In such way, two NH groups could be involved in the interaction with two individual aldehyde groups, or can form bifurcated hydrogen bond with one aldehyde group, while the third terminal pyrrole bearing alkyl side chains remains free. Accordingly, the binding of NP4 to the oxidized cellulose (without CS) also proceeds via O·····N–H hydrogen bonds. In this case, since no repulsion occurs, all three NH groups can form hydrogen bonds thus lowering the availability of aldehyde groups for interaction with other dye molecules, i.e. the number of possibly bounded dye molecules per aldehyde group is lower with regard to the samples with CS. Also, in this case, the steric hindrance of alkyl side chains on the surface of the fiber also affects the lowering of the exhaustion of dyebath. Suggested mechanism is also supported by the fact that NP4 shows affinity towards fibers bearing carbonyl groups (polyamide, triacetate) (Kramar et al. 2014).
3.4. Halochromic effect of dyed viscose
pH sensitivity was measured in the region pH 4-10, by measuring color coordinates and reflectance after the samples have been exposed to a buffer solution of pH 4, 7 and 10 (Fig. 7). The samples were tested for 10 color changing cycles. The change of color is evident both visually and spectrophotometrically and the change from blue to red/pink occurs fast (1-2 sec), while the change from red to blue lasts longer (10-15 sec). The reflectance curves of samples exposed to pH 10 indicate the shift of minimum towards a higher wavelength (580 nm). After exposure to neutral and acidic pH, curves have the same minimum (535 nm), but different reflectance, meaning that the color intensity is changing. In samples with CS, a difference in reflectance between samples exposed to all three values of pH was found, unlike sample CV60 whose reflectance curves measured at pH 4 and pH 7 are almost overlapping (Fig. 7). This suggest that for a cellulose to be used in all measuring ranges presented here (pH 4-7-10) chitosan deposition is crucial.
The oxidation and addition of CS lead to a higher red and less negative blue color coordinate when samples are exposed to pH 4 and pH 7 (Fig. 8). Alkali pH leads to a strong decrease of red and an increase of blue (Fig. 8a). To examine distinctive colors at different pH, the color difference ΔE* was calculated for textile samples exposed to pH 10 or pH 4 and pH 7 (Fig. 8b).
Samples with chitosan (CV/CS and CV60/CS) show an almost equal color difference in both directions, with the higher ΔE* in sample CV/CS. In sample CV60 there is a noticeable color difference towards pH 10, but towards acidic, ΔE* is slightly above 2, which is, according to literature (Sharma 2003) defined as “just noticeable difference” and this visually can be seen in Figure 8c.
The color of samples with CS exposed to various pH could be the consequence of binding mechanism. Proposed mechanism (Fig. 6) suggests that the dye molecule, i.e. its pyrrolyl pyrromethene chromophore is not all immobilized by bonds with cellulose and chitosan, and can respond to pH changes. Furthermore, the new shoulder was detected at 1590 cm-1 in ATR-FTIR after samples’ exposure to pH 10 (Fig. 8d).
This peak is ascribed to chitosan, specifically to NH2 groups arising from the deprotonated -NH3+, as reported in other works (Zhang et al. 2019), again confirming that ammonium groups do not directly participate in the binding of the NP4, i.e. prodigiosin, rather on its conformation when bonded to fibers surface, as suggests proposed mechanism on scheme in Fig. 6. Reversible pH-mediated color change of the fibers is also related to the structural changes of the dye solution at different pH values caused by protonation/deprotonation of the NP4 molecule (Fig. 5). This allows us to conclude that halochromic properties of the investigated fibers are influenced by the specific behavior of the NP4 at different pH values and specific bonding to functionalized cellulose with chitosan.
3.5. Cytotoxicity of dyed viscose
Cytotoxicity was evaluated by in vivo test using a human keratinocytes cell line (HaCaT) and human fibroblasts cell line (MRC5). According to this method, materials extracts were prepared by immersing the viscose samples (10 mg mL-1) in RPMI medium at 37 °C for 72 h under dynamic conditions. The cell viability exposed to viscose extracts of different concentrations was assessed using a standard MTT assay.
Histogram in Fig. 9 demonstrates the percentage of exposed cell growth compared to non-exposed cells (control) after 48 h.
The results presented in Fig. 9 indicate that 100% of materials extracts can cause modest toxic effect to keratinocytes and fibroblasts. It was shown that 100% concentrations of viscose materials did not induce significant decrease in HaCaT and MRC5 cells viability after 48h-treatmens.
Viscose with NP4 extract caused 30% decrease, whereas oxidized viscose and viscose with chitosan and NP4 caused approximately 40% and 50% decrease in MRC5 and HaCaT cells proliferation, respectively. Results also pointed out that over 80% of cells survived after being in contact with viscose materials extracts that were diluted by 50%. According to literature, materials are considered safe when the cells viability is over 70% (Andreani et al. 2017).
In conclusion, no significant cytotoxic effect on healthy human fibroblast (MRC-5) and human keratinocytes (HaCaT) cell lines of viscose with chitosan and NP4 was detected in the indirect assay, confirming their non-toxic nature.