Synthesis and Characterization of Functional Cationic Copolymers
As previously mentioned, our main goal was the design of water-soluble functional cationic copolymers, derived from the homopolymer PVBC (Scheme 1). The homopolymer was prepared through free radical polymerization in an organic solvent, since it is water-insoluble. Its weight-average molecular weight and polydispersity index were determined through gel permeation chromatography and found 43000 and 2.7, respectively. Partial quaternization of PVBC with triethylamine, TEAM, provided a series of cationic copolymers P(VBC-co-VBCTEAMx) (Scheme 1, Route 1) of VBC with VBCTEAM (namely VBC quaternized with TEAM), where x is the molar content of VBCTEAM units as found through 1H NMR characterization. From solubility tests it was verified that the copolymers with x > 30% maintain water-solubility. For comparison reasons, the cationic homopolymer PVBCTEAM was also synthesized. In order to avoid the presence of even marginal quantities of unreacted VBC units, quaternization of the monomer VBC with TEAM was now performed prior to the polymerization of the resulting cationic monomer VBCTEAM (Scheme 1, Route 2).
The copolymers were characterized through a variety of techniques. As example, the 1H NMR spectra of the homopolymers PVBC and PVBCTEAM are compared in Fig. 1 with the respective spectra of two P(VBC-co-VBCTEAMx) copolymers with x = 53 and 70%. In more detail, in the spectrum of the homopolymer PVBC in deuterated chloroform, the peaks attributed to the aromatic ring (b and c) are observed in the 6.3–7.3 ppm region, while the peak attributed to the CH2Cl group (a) is observed at 4.6 ppm and the peaks attributed to the backbone (d) are observed in the 1.2-2 ppm region. In the spectra of the homopolymer PVBCTEAM and the two P(VBC-co-VBCTEAMx) copolymers in D2O, the peak of -CH2- group of TEAM (e) appears at 3 ppm, while the strong peak, attributed to the CH3- groups of TEAM (d) is superimposed on the peaks attributed to the backbone in the 1–2 ppm region. From the ratio of the integral of this region (d, f) over that of the aromatic ring (b and c), the actual composition x of the copolymers could be determined. The results, summarized in Table 1, are in a rather good agreement with the feed composition of the polymer quaternization reaction.
The characteristic vibrational modes of VBC and VBCTEAM can be identified in the spectra of P(VBC-co-VBCTEAMx), shown in Fig. 2. The contribution of each component becomes evident if the spectra of PVBC (x = 0) and PVBCTEAM (x = 100) are compared. A detailed list of Raman and FTIR bands for these two homopolymers is reported in Table S1. The distinctive bands of the -CH2Cl chemical species dominate both FTIR and Raman spectra for the case of PVBC. More specifically, the strong bands at 3006 cm− 1 and 2958 cm− 1 in the Raman spectra and their weaker counterparts in the FTIR spectra are attributed to the CH2 stretching vibrations of these molecular species (Socrates 2004). Deformation, wagging, twisting and rocking vibrations of CH2 can be also resolved most probably at ~ 1445, 1265, 1180 and 745 cm− 1. The contribution of the first two in both Raman and FTIR spectra is considerable while the intensity of the rocking vibrations is weaker. The most characteristic band is the C-Cl stretching vibration (at ~ 675 cm− 1) which is very strong in the FTIR spectra. Although its contribution is considerable in the Raman spectra it lies rather close to the PVBCTEAM band at 681 cm− 1, which is of similar intensity. For the case of VBCTEAM, the vibrational bands of the aliphatic components of the tertiary amine are the most distinguishing. Hence, the CH3 asymmetric, symmetric stretching vibrations at ~ 2985 cm− 1, 2950 cm− 1 and the respective deformation vibrations at 1455 cm− 1, 1400 cm− 1 dominate the vibrational spectra. Additionally, the N+C4 asymmetric stretching vibration of the quaternary amine at ~ 930 cm− 1 (Berg 1978), the symmetric one (Kabisch 1980) at ~ 780 cm− 1 and the CH3 and CH2 rocking vibrations (1010 cm− 1 and 785 cm− 1), which, as expected, are stronger in the FTIR spectra, are the bands associated with PVBCTEAM. For symmetric quaternary species (e.g. tetramethylammonium) single symmetric and asymmetric vibrational modes are expected. In the case of VBCTEAM the lack of symmetry in quaternary ammonium may result in splitting of the asymmetric vibrational mode in two peaks most possibly the one at 903 cm1 (Mendoza 2012). From the assignment of the above-mentioned peaks observed in the Raman and FTIR spectra of VBC and VBCTEAM polymers, it becomes obvious that for the copolymers P(VBC-co-VBCTEAMx), the relative intensity of the VBCTEAM and VBC vibrations are essentially related to the relative content of each unit along the polymer chain, namely x.
Apart from the Raman and FTIR spectroscopic data, the success of the polymer synthesis is furthermore justified by XPS spectroscopy. The XPS spectra collected from PVBC, PVBCTEAM and P(VBC-co-VBCTEAMx) copolymers in the spectral range of C1s, N1s and Cl2p orbitals are given in Fig. 3. The C1s spectral region may be fitted using three peaks: (i) one attributed to C atoms participating in the benzene ring, (ii) a second peak attributed to the C atoms of the backbone chain and (iii) a third one associated with C atoms attached to the quaternary N+ atom or/and Cl atom (Moulder 1992; Topalovic 2007; Wang et al. 2021). The fitting indicates that the contribution of the third peak is more pronounced for the case of PVBCTEAM, its intensity decreases by a factor of nearly 0.5 for the case of P(VBC-co-VBCTEAM53) and nearly vanishes in the PVBC spectrum. The intensity of the first peak is greater for the PVBC polymer a fact that is reasonable since for each PVBC monomer the proportion of the carbon participating in the benzene ring is 6 out of 9. It is interesting to notice that in the N1s only one peak is detected at 402 eV which is attributed to quaternary nitrogen (Topalovic 2007; Wang et al. 2021). This peak is absent in PVBC. Finally, concerning the Cl2p spectral region, two doublets are recorded. The first at 197.0 eV is assigned to Cl− i.e. the quaternary N+ counter ion, while the second at 201.1 eV is assigned to neutral Cl of PVBC (Topalovic 2007; Wang et al. 2021). A weak band at 201.1 eV in the spectrum of PVBCTEAM indicates that there exists a small proportion of non-reacted monomers during the synthesis of the homopolymer. By following a typical fitting procedure of the XPS spectra, the elemental composition of all studied homo- and co-polymers may be derived and is summarized in Table 1.
Table 1
1Η NMR and XPS characterization of homopolymers and P(VBC-co-VBCTEAMx) copolymers
Polymer | TEAM content in feed, x | 1H NMR results, x | Element, XPS characterzation |
| | | %C | %N | %Cl |
| | | Theoretical* | Experimental | Theoretical* | Experimental | Theoretical* | Experimental |
| | | | | | | Cl− | Cl | Cl− | Cl |
PVBCTEAM | 100 | 100 | 88.2 | 90.0 | 5.9 | 4.2 | 5.9 | | 5.8 | |
PVBC | 0 | 0 | 90.0 | 94.0 | 0 | 0 | | 10.0 | | 6.0 |
P(VBC-co-VBCTEAM53) | 50 | 53 | 88.9 | 89.9 | 3.9 | 2.9 | 3.9 | 3.3 | 3.9 | 3.3 |
P(VBC-co-VBCTEAM70) | 70 | 70 | 88.6 | 90.1 | 4.7 | 4.0 | 4.7 | 2.0 | 4.9 | 1.6 |
* Calculated from 1H NMR data.
Modification of Cotton Fabrics
As a standard protocol, the modification of cotton fabrics in aqueous solutions containing the desired polymer and 0.25 M NaOH for 3h at 60oC was explored.
In fact, after drying the modified textile, no significant weight change has been observed. In addition, no indication of the presence of cationic polymers on the cotton surface was observable through ATR-FTIR investigation. However, the presence of cationic polymers on treated textiles was evident in the respective Raman spectra. As seen in Fig. 4, the only spectral feature originating from the cationic polymer in the Raman spectra of modified textiles, not interfering with any of the cellulose vibrational modes, is the band at ~ 1610 cm− 1, attributed to the para-substituted phenyl ring of the polymer units. The intensity of this band is rather low with respect to the strong cellulose vibrational bands, suggesting that the polymer mass within the scattering volume is small relative to cellulose.
XPS investigation offers additional evidence on the presence of polymers on the surface of modified textiles. In order to evaluate the XPS spectra in modified textiles the reference spectra of the free-standing cotton fabric are required. It has to be stressed that there is a noticeable effect of X-Ray irradiation, which is observed by the intensity decrease in the XPS spectra collected from modified and non-modified fabrics. The procedure followed so that the most reliable data could be obtained for the materials under investigation is given in detail in the supplementary material (Section S2).
Comparison with existing detailed XPS spectra of treated textiles45,46 indicates that the bleaching process of the textiles used in the current work may be considered mild. Hence in the respective XPS spectra significant contribution of C-C and C-H, as well as weak N peaks, are resolved (Fig. 5). Taking into account the general chemical formulae of the modifiers studied, the most characteristic XPS peaks are expected to be the quaternary N+ and Cl− bands associated with the VBCTEAM component, the Cl peak related to VBC component, while in the C spectral region the phenyl species of both VBC and VBCTEAM components are expected to contribute to the overall C band. Surface modification is demonstrated by the observation of peaks in the C-C and C = O as well as quaternary N+ and Cl− peaks at ~ 285, 402 and 197 eV respectively (Fig. 3). Similar peaks are observed in the XPS spectra of all modified fabrics using different types of homo-/co- polymers.
Apart from the direct observation of the presence of polymer modifiers on the surface of treated cotton fabrics through the aforementioned Raman and XPS studies, the modification was also indirectly observed through UV-Vis spectroscopy by the decrease of the polymer concentration in the aqueous treating solution. The latter is straightforward since all polymers exhibit a strong absorbance band at ~ 260 nm. As seen in Fig. 6, the absorbance of the polymer solution decreases considerably after treatment of the cotton fabrics. To quantify these observations, calibration curves associated with the 260 nm band were constructed for each polymer (Fig. S3). The polymer exhaustion from the modification bath and the % mass coverage of the textile with cationic polymer can thus be determined as
Polymer Exhaustion = (c0 – c) /c0 (1)
% Mass Coverage = Polymer Exhaustion x V/m x 100 (2)
where c0 and c is the polymer concentration (mg/mL) before and after treatment of cotton fabrics, V (mL) is the solution volume and m(mg) is the mass of cotton fabric.
Figure 6 UV-Vis absorption spectra of the polymer solutions, used for cotton modification, before (solid lines) and after the immersion of cotton fabrics (dashed line) for (a) P(VBC-co-VBCTEAM53) and (b) PVBCTEAM. With the exception of the more dilute solution, all mother solutions were adequately diluted to a final concentration c0 = 0.3 mg/mL. The solutions after cotton treatment were respectively diluted
The results concerning Polymer Exhaustion and % Mass Coverage of cotton fabrics treated with aqueous PVBCTEAM or P(VBC-co-VBCTEAM53) solutions are given in Fig. 7 as a function of the initial polymer concentration, c0. As seen, polymer exhaustion decreases sharply with the polymer concentration, already from low c0. This indicates a low surface coverage of the cotton fabric with the cationic polymers. Indeed, Fig. 7 shows that the % Mass Coverage of the fabric increases strongly at low polymer concentration and soon, at an initial polymer concentration somewhat higher than 1 mg/mL, tends to a plateau value of the order of 1-1.5%. It is noteworthy that a similar behavior concerning polymer exhaustion and % Mass Coverage of the fabric is observed for both polymers, with the homopolymer PVBCTEAM to be slightly more effective than the copolymer P(VBC-co-VBCTEAM53). This behavior indicates that polymer binding onto the negative cotton surface is primarily an ion exchange process. Since both polymers are characterized by a high linear charge density, no significant variations are expected, as it is the case. To modulate the extent of polymer binding, copolymers containing a much lower cationic content should be explored. However, this is not possible with the P(VBC-co-VBCTEAMx) copolymers family since the products are not water-soluble for x < 0.3.
Modification Kinetics of Cotton Fabrics
Τhe modification kinetics is easily monitored indirectly, following the characteristic absorption band at 260 nm in the respective UV-Vis spectra of the solutions used for the textile modification at various time intervals, covering a modification time of 24 h. Representative results for the modification with PVBCTEAM and P(VBC-co-VBCTEAM53) with an initial polymer concentration c0 = 0.5 mg/mL are shown in Fig. 8. For both polymers the absorbance decreases considerably in the first few hours, reaching a constant value for higher modification times.
These experiments were performed in individual aqueous polymer solutions for each treatment time. Characterization of the modified textiles through Raman spectroscopy offers direct monitoring of the modification kinetics. The spatial resolution offered by the technique indicated that the intensity of the characteristic stretching vibration of the para-substituted benzene ring at ~ 1610 cm− 1 was significantly altered at different spots on the same sample. This suggests that the modification is not uniform at least down to the level determined by the technique’s spatial resolution (in the order of 1 µm2). In order to confront the issue, average spectra obtained from at least 20 different spots on each sample were extracted. Averaged Raman spectra obtained from samples treated with PVBCTEAM for various modification times are shown in Fig. 9. The spectra are normalized with respect to the 1481 cm1 band attributed to cellulose, the intensity of which was found to be the less sensitive to the polarization of light and/or sample anisotropy (hence it was used as a reference band). It is seen that the intensity of the ~ 1610 cm1 band increases significantly during the first hours of modification.
The dependence of the extent of modification on treatment time could also be indirectly evaluated through the color of the samples after dyeing at fixed conditions. The appearance of the cotton fabrics prepared for the kinetics study, after dyeing with the reactive dye Novacron Ruby NRS-3B under basic conditions at room temperature is shown in Fig. 10. It should be noted that salt-free dyeing was attempted in this study. As a result, the unmodified textile (modification time is 0 min) is barely dyed, as seen. In contrast, it is clear that the textile treatment applied in the present study enables the effective salt-free dyeing of cotton. In fact, the modified samples are more strongly dyed with the increase of modification time. This is evident mostly for the first stages of modification (up to a modification time of about 2 h).
Figure 10 The appearance of the cotton fabrics modified for different time intervals using the polymer P(VBC-co-VBCTEAM53), after dyeing with the reactive dye Novacron Ruby NRS-3B under alkaline conditions. co = 0.5 mg/mL. Standard dyeing conditions were applied. The unmodified fabric was also dyed in the presence of NaCl at a 0.9M concentration
The remaining dye in the dyebath can be easily monitored through UV-Vis spectroscopy (Fig. 11). As seen, the absorption band of the dye, centered at ~ 545 nm, decreases strongly with the time of cotton modification with PVBCTEAM or PVBC-co-VBCTEAM53). The construction of the calibration curve of the dye under these conditions (Fig. S4) allows the quantification of the dye concentration in the dyebath and, thus, the determination of dye exhaustion, defined as
Dye Exhaustion = (C0- C)/C0 (3)
where C0 and C is the dye concentration (mg/mL), before and after dyeing of cotton, respectively.
Figure 12 summarizes the results for the modification kinetics of cotton samples, as determined through the three methodologies, i.e. polymer exhaustion, Raman spectroscopy (on modified cotton surface) and dye exhaustion. It is clear that the results of the three techniques are in a very good agreement, following the same kinetics curve: modification of cotton increases sharply within the first two hours of treatment, while a smooth increase is observed for higher modification times.
Figure 12 Modification kinetics of cotton fabrics using the polymers P(VBC-co-VBCTEAM53) (left) and PVBCTEAM (right), as monitored through polymer exhaustion dye exhaustion and Raman study
In agreement with the results reported in Fig. 7, the results presented in Fig. 12 evidence that extent of modification is similar regardless of x, since the cationic content is high enough. This is also clearly expressed in the dyeing capability of cotton fabrics modified for 2 hours with PVBCTEAM or P(VBC-co-VBCTEAMx) copolymers. As shown in Fig. 13, while the dye exhaustion of the unmodified sample in salt-free solutions is very low (indicating no dyeability), in the case of all modified fabrics the dye exhaustion in salt-free solutions is high, indicating effective dyeability, similar to that of unmodified samples in strongly saline solution.
Chemistry of modification
Of particular interest is the understanding of bonding of P(VBC-co-VBCTEAMx) modifiers onto the cellulose fabrics. In principle, the polymers possess two possible active sites that can interact with the substrate (Scheme 2). The first is the VBC unit, which could covalently bind to the cellulose fabrics and the second is the VBCTEAM counterpart which may electrostatically interact with the oppositely charged sites of cellulose during the modification process. However, possible hydrolysis of VBC units should also be considered, when studying the interaction between the modifier and cotton surface (Mabey and Mill 1978). The low quantity of modifier on the cotton surface complicates further the study, since it prohibits the monitoring of C-Cl vibrational modes due to the strong interference with the corresponding cellulose vibrational bands. Confocal measurements were applied which however did not suppress the cellulose signal to a level that the modifier peaks were observed.
Careful XPS measurements, on the other hand, could offer valuable infiromation on the physicochemical characterization of the cotton modification process, despite the disadvantage of samples’ degradation. The Cl2p spectral region of the P(VBC-co-VBCTEAM53) modifier is compared in Fig. 14 to that obtained from the respective modified fabric. As explained in Section 3.1 (Fig. 3, spectral region of Cl2p orbitals), the two doublets are assigned to Cl− of the VBCTEAM counterion (~ 197 eV) and the neutral Cl of the VBC unit (~ 200 eV) and are of approximately equal intensity indicating that the VBC/VBCTEAM analogy is basically the same. After the modification, the line profiles in the same spectral region are altered. More specifically, the neutral Cl bands are severely suppressed compared to the Cl− bands (Fig. 14). Since there is an overlap of the Cl− bands with the neutral Cl bands, a chemical protocol was followed that exchanges the counter Cl− anion with Br−. As verified, the Cl− band disappears now in the XPS spectra. On the other hand, a Br− band is clearly resolved in the respective spectral region. The elimination of the Cl− contribution reveals that the absence of the characteristic Cl band of the VBC segments is striking. In fact, a similar observation was also verified during the first stages of modification kinetics study, while VBC units were clearly observable in the modification solution, indicating that the origin of this complete disappearance of the characteristic Cl band of VBC is not just hydrolysis. In contrast, this is a strong indication that covalent attachment takes indeed place. Concerning electrostatic binding of the modifier with cellulose, the N+/Cl− atomic ratio should be monitored. This ratio may be used for the evaluation of the electrostatic binding between the VBCTEAM units and cellulose and it was found that it is considerably altered. However, the detection of peaks associated with the VBCTEAM counterions (either Cl− or Br−) in the modified fabrics suggests that this type of binding follows an ion exchange process.
This ion exchange process can be monitored by XPS spectroscopy on the samples, which were modified at selected time durations with the homopolymer PVBCTEAM. The N+/Cl− atomic ratio of the samples was calculated for all samples. As expected, the ratio is ~ 1 for the homopolymer PVBCTEAM, while its value increases with treatment time and soon reaches a plateau value of ~ 5–6 (Fig. 15). This indicates that the majority (more than 80%) of the compensating Cl− anions relative to the N+ cations is not present on the surface of the treated fabrics, as a consequence of electrostatic binding of the modifier’s polymer chains with cellulose anions. The detection of Cl− counterions, however, indicates that the surface of the fabric is positively charged, allowing the facile salt-free dyeing with the negatively charged reactive dyes, as evidenced in the previous section.