Characterization of Bio-Based Monomer
The number of double bonds, calculated from the iodine value, and the number of epoxy groups, expressed as the epoxy value, were determined in ME, EME, and BM (Table 2). The epoxy groups in EME were evidenced by an epoxy value, and their formation was also confirmed by a drop in the level of unsaturation, although some double bonds were found to be non-epoxidized. After the itaconation of EME, an increased iodine value was determined for BM, which can be attributed to the presence of methyl itaconate functions in BM molecules. A theoretical amount of MMI that could be bonded via epoxy groups in 100 g of EME is approximately 46.8 g (calculated based on the determined epoxy value for EME). However, the results of the iodine values for EME and BM indicate that the real amount of reacted MMI was much lower; approximately 11.1 g of MMI was bonded in 100 g of BM, which means that only approximately 25% of epoxy groups had been successfully itaconated.
Table 2
Iodine and epoxy values of intermediate products and final BM
Sample | ME | EME | BM |
Iodine value (g I2/100 g) | 180.2 ± 5.2 | 32.3 ± 1.2 | 51.7 ± 2.1 |
Moles of double bonds/100 g | 0.701 ± 0.020 | 0.127 ± 0.005 | 0.204 ± 0.008 |
Epoxy value (moles of epoxy groups/100 g) | 0 | 0.325 ± 0.010 | 0 |
The itaconation of EME was followed by NMR spectroscopy. 1H NMR spectrum of EME (Fig. 1) shows characteristic signals of residual vinylene function at 5.25–5.60 ppm (denoted a), the methyl ester group at 3.62 ppm (denoted b), and the CH group of epoxy cycles in the region 2.8–3.2 ppm (denoted c and d). Upon synthesis of BM, the high conversion of the epoxy function is evident from the disappearance of the signals c and d. The introduction of methyl itaconate into the BM structure is proven by the appearance of new signals at 6.34, 5.68, and 3.32 ppm (denoted e, f, and i), assigned to fully esterified itaconic acid. The signal of the methyl itaconate group (signal h) is shifted to a lower field compared to that of the fatty acid methyl esters (signal b) because of the electronic effects of the methylidene function. The signal at 4.88 ppm (signal g) was assigned to the CH group in the fatty acid tail in the α-position relative to the itaconate function. The integral intensities of the signals show that each fatty acid tail in EME contains ~ 0.78 itaconate functions on average. Note that the residues of the starting MMI were detected at 6.41, 5.79, and 3.35 ppm (signals denoted by *), but their content is only ~ 20 mol. % (~ 8 wt. %).
Successful itaconation was further supported by IR and Raman spectra of BM focusing on the appearance of the characteristic C = C stretching band at 1644 cm–1 and the C = O stretching band at 1719 cm–1 (Fig. 2). Note that the bands are significantly shifted in comparison to the fatty acid functions as a result of conjugation. The itaconate function also gives the characteristic C–H stretching band of the methylidene group at 3109 cm–1. The formation of hydroxy function after itaconation is evidenced by the increase in the O–H stretching band at ~ 3500 cm–1 in the IR spectrum.
Composition of Bio-Based Monomer According to Thermodynamic Calculations
The real composition of the synthesized BM cannot be easily determined due to the formation of many structurally similar compounds (theoretically above 30). The compounds differ in the number of methyl itaconate groups and their position in the fatty acid chain. (In the itaconation reaction step, the epoxy group is transformed into one itaconate and one hydroxyl group). These compounds are almost impossible to separate and identify by common chromatographic methods (standards are not commercially available). Therefore, the thermodynamic calculations of Gibbs energy for the individual itaconation reactions of EME intermediate product with MMI were carried out to estimate the compounds preferred to occur in the BM final product. The calculations were based on theoretical quantum chemistry (thermodynamic values of the individual reactants and products are not tabulated) implementing the actual reaction conditions (temperature of 293 K and relative permittivity of EME of 6.02). It should be mentioned that the thermodynamic calculations for the particular epoxidation reactions of ME intermediate product were not the subject of this work but were presented in the recent study [72]. Low Gibbs energies ranging from − 242 to − 264 kJ⋅mol–1 were calculated for the individual epoxidation reactions, suggesting their strongly spontaneous character under specified reaction conditions.
The illustrations of the compounds that may be formed in BM as a result of itaconation of the most represented epoxidized methyl esters of unsaturated fatty acids, namely oleic (C18:1), linoleic (C18:2), and linolenic (C18:3), with calculated Gibbs energy values are presented in Figs. 3−5. As for the double-epoxidized methyl esters of linolenic acid, all the possible reactions were not included because a high number of structurally similar compounds can be formed, requiring time-consuming calculations of Gibbs energies that are not expected to differ significantly. Compounds based on a triple-epoxidized methyl ester of linolenic acid were also not addressed as this intermediate product had not been proven to be formed [73].
In the case of the presented itaconation reactions, Gibbs energy values ranging from − 16 to − 58 kJ⋅mol–1 were calculated, suggesting that itaconation is less spontaneous compared to epoxidation. The probable reason is the strong exothermic character of epoxidation due to the decomposition of hydrogen peroxide. Focusing on the mono-itaconation reactions, the more spontaneous (i.e., advantageous) position of the methyl itaconate group appears the more distant bonding from the fatty acid head (i.e., the methyl ester group of fatty acid), probably for steric reasons. The calculated Gibbs energies of the mono-itaconation reactions were also found to decrease with the increasing number of double bonds present in the fatty acid tail. This result lets us assume that the presence of double bonds increases the spontaneity of the mono-itaconation reaction. Regarding the double-itaconation reactions, the trends are ambiguous. As for the number of methyl itaconate groups formed, Gibbs energies ranging from − 16 to − 35 kJ⋅mol–1 were calculated for the mono-itaconation reactions, while Gibbs energies in the range of − 37 to − 58 kJ⋅mol–1 were obtained for the double-itaconation reactions. These results indicate that the formation of compounds containing two itaconate functions in the final BM is more spontaneous, i.e., favorable according to the thermodynamic calculations.
Characterization of Latexes
The properties of prepared latexes are presented in Table 3. All synthesized latexes exhibited a low coagulum content which was slightly affected by the amount of copolymerized BM; the higher the BM concentration in the monomer mixture, the higher the coagulum content formed during latex synthesis. The amount of BM was also shown to have a considerable impact on monomer conversion; decreasing conversion was found with increasing BM content in the case of latexes synthesized using 20, 25, and 30 wt. % of BM in the monomer mixture. This phenomenon could be related to the lower water solubility of BM compared to standard acrylic monomers, which probably limited its transport in water and incorporation into the existing latex particles [74–76]. The decreased conversion and colloidal stability during latex synthesis were also the cause of why no further attempts were made to incorporate a higher BM content into the latex copolymer.
Further tests revealed that neither the viscosity nor the particle diameter of freshly prepared latexes were influenced pronouncedly by the amount of introduced BM. On the contrary, the zeta potential of the freshly prepared latexes (being inherently negative because of the presence of carboxyl and sulphate functions originating from copolymerized MAA, persulphate initiator, and adsorbed anionic surfactant) was found (meant in absolute value) to be proportional to the amount of introduced BM. This phenomenon can be attributed to partial hydrolysis of the methyl ester functions [77] in the BM molecules under acidic environment (pH ∼ 2) of emulsion polymerization, which caused the formation of additional carboxyl groups. The zeta potential values (in absolute terms) above −40 mV indicate that all the freshly prepared latexes exhibited colloidal stability [78]. After storage at 40°C for 3 months, neither visible coagulation nor a significant increase in the average hydrodynamic diameter of the particles were observed. However, a mild reduction in the zeta potential (the absolute value) was found, which can predicate the desorption of some emulsifiers from latex particles [79]. Despite this fact, all latexes can still be considered long-term stable.
Table 3
Sample | Coagulum content (%) | Conversion (%) | Viscosity | Hydrodynamic diameter (nm) | Zeta potential (mV) |
(mPa⋅s) | Initial state | After storage | Initial state | After storage |
REF | 0.7 ± 0.6 | 94.3 ± 1.0 | 11.1 ± 0.4 | 92.1 ± 1.9 | 94.0 ± 1.6 | −42.2 ± 0.2 | −38.5 ± 1.1 |
L_5 | 1.1 ± 0.2 | 93.5 ± 0.2 | 18.9 ± 0.3 | 87.4 ± 2.2 | 87.0 ± 1.3 | −40.1 ± 3.3 | −37.3 ± 2.1 |
L_10 | 0.7 ± 0.1 | 93.7 ± 0.6 | 19.5 ± 0.2 | 85.0 ± 1.1 | 85.0 ± 1.8 | −42.8 ± 0.9 | −36.7 ± 1.0 |
L_15 | 1.2 ± 0.1 | 93.1 ± 0.5 | 18.6 ± 0.2 | 84.8 ± 1.5 | 85.8 ± 1.4 | −48.1 ± 0.3 | −41.7 ± 0.6 |
L_20 | 1.3 ± 0.1 | 91.6 ± 0.8 | 19.2 ± 0.3 | 83.8 ± 0.8 | 84.3 ± 1.4 | −49.4 ± 1.1 | −42.9 ± 0.8 |
L_25 | 1.7 ± 0.2 | 88.8 ± 1.1 | 19.8 ± 0.1 | 80.7 ± 1.6 | 80.0 ± 1.6 | −48.9 ± 2.4 | −43.6 ± 0.1 |
L_30 | 2.6 ± 0.1 | 84.5 ± 0.6 | 19.3 ± 0.2 | 87.0 ± 0.9 | 87.5 ± 0.9 | −51.5 ± 2.3 | −43.8 ± 0.5 |
Characterization of Copolymer Structure
The weight-average molar mass (Mw), dispersity (Ð), the z-average RMS radius, and the polymer and nanogel fractions are summarized in Table 4. Similarly to our previous research on latex copolymers synthesized from acrylated BMs [30], the RI fractograms show two peaks (an example is shown in Fig. 6). The peak at lower retention times can be assigned to dissolved individual macromolecules, while that at higher retention times corresponds to swollen cross-linked latex particles (nanogels). As expected, with increasing BM content, nanogels become more compact, which is evident from the decreasing retention time and RMS radius, as shown in Fig. 7. Figure 7 also reveals a slightly decreasing RMS radius with increasing retention time. This tendency was found for all the tested samples. Since the retention time in AF4 increases with increasing hydrodynamic volume, decreasing RMS radius may suggest that the central parts of the nanogels become more compact despite their increasing hydrodynamic volumes.
The level of cross-linking incorporated into latex polymers was assessed, considering the nanogel content obtained from A4F and the cross-link density obtained through swelling experiments (Table 4). As expected, the nanogel content and cross-link density were related to the amount of introduced BM. This effect is a consequence of the copolymerization of multi-itaconated BM fractions represented mainly by linoleic and linolenic fatty acids, which has already been documented in our previous work on latex copolymers with acrylated BM counterparts [30].
Table 4
Results of the AF4-MALS analysis (the results are averages from three measurements with the measurement uncertainty below 10%) and cross-link density determined through swelling experiments
Sample | Soluble polymer | Nanogel | RMS radius (nm) | Cross-link density × 10− 6 (moles of crosslinks⋅cm–3) |
Mw (103 g⋅mol–1) | Ð | Fraction (%) | Mw (106 g⋅mol–1) | Fraction (%) |
REF | 6000 | 36.0 | ≅ 100 | -b | -b | 120 | -c |
L_5 | 4300 | 56.1 | 41.7 | 82 | 58.3 | 109 | 3.51 ± 0.2 |
L_10 | 6190 | 12.2 | 20.7 | 89 | 79.3 | 79 | 7.59 ± 0.1 |
L_15 | 339 | 20.4 | 19.1 | 88 | 80.9 | 71 | 12.6 ± 0.4 |
L_20 | 349 | 15.2 | 19.3 | 114 | 80.7 | 68 | 18.8 ± 0.6 |
L_25 | -a | -a | 22.0 | 81 | 78.0 | 60 | 34.5 ± 1.1 |
L_30 | -a | -a | 14.4 | 133 | 85.6 | 59 | 61.2 ± 2.9 |
a Value was impossible to determine due to a low signal of the MALS detector.
b The nanogel fraction was not present.
c Impossible to determine.
IR spectroscopy was employed to follow the incorporation of BM into latex polymer chains (Fig. 8). For this purpose, a protocol previously described for acrylated fatty acid methyl esters was used [60]. The increased content of BM in the starting monomer mixture resulted in polymers with increased intensity of the absorption bands at 2931 and 2855 cm–1 assigned to the antisymmetric and symmetric C–H stretching modes of the methylene groups, νa(C–H, CH2) and νs(C–H, CH2), respectively. Thus, the occurrence of methylene groups forming the fatty acid tails could be evidence of successful BM copolymerization. A similar effect was observed in the Raman spectra (Fig. 8), where the increased amount of incorporated BM well correlated with the increased intensity of the symmetric CH stretching band of the methylene groups, νs(C–H, CH2), at 2854 cm–1.
Evaluation of Coatings
The coating properties of the latex films are presented in Table 5. All coating samples appeared smooth, glossy, and transparent, with gloss and transmittance values comparable to the REF latex coating. It was found that copolymerized BM decreased coating wettability; almost all BM-comprising coating samples exhibited about 7 ° higher WCA than REF coating. This phenomenon could be attributed to the hydrophobic nature of the long fatty acid chains in the BM structure. However, the presence of hydrophilic hydroxyl groups, formed as a result of the ring-opening reaction of the epoxy group (see e.g. Figure 3), probably diminished this effect.
In our previous research on acrylated BMs derived from various vegetable oils, including CO [30, 60], a decrease in Tg and hardness of BM-comprising coatings was found revealing the plasticizing effect of the fatty acid tails [80, 81] attached to the acrylic polymer backbone. The present study found the opposite effect of copolymerized itaconated BM on the Tg and coating hardness; the higher the BM concentration, the higher the Tg and hardness. This phenomenon can be explained by the significant contribution of rigid methyl itaconate groups [59].
Table 5
Coating properties of latex films
Sample | Gloss at 60° (GU) | Transmittance at 500 nm (%) | WCA (°) | Hardness (%) | Tg (°C) |
REF | 83.5 ± 0.1 | 91.4 ± 0.3 | 64.9 ± 1.5 | 5.8 ± 0.2 | 1.3 ± 0.3 |
L_5 | 84.6 ± 0.1 | 91.2 ± 0.5 | 70.6 ± 3.5 | 8.4 ± 0.3 | 8.4 ± 0.4 |
L_10 | 84.5 ± 0.2 | 90.5 ± 0.4 | 72.7 ± 2.7 | 8.8 ± 0.6 | 9.3 ± 0.2 |
L_15 | 83.8 ± 0.1 | 90.0 ± 0.2 | 72.5 ± 1.9 | 10.3 ± 0.5 | 10.2 ± 0.3 |
L_20 | 84.6 ± 0.1 | 90.3 ± 0.7 | 72.6 ± 3.7 | 12.3 ± 0.9 | 11.9 ± 0.3 |
L_25 | 84.2 ± 0.5 | 90.2 ± 0.5 | 72.3 ± 0.7 | 10.6 ± 0.6 | 11.1 ± 0.6 |
L_30 | 84.1 ± 0.1 | 90.3 ± 0.3 | 72.6 ± 0.9 | 14.8 ± 1.1 | 12.8 ± 0.2 |
a Impossible to determine.
One of the most pressing shortcomings of common latex protective coating films is their poor water resistance, which manifests itself in the deterioration of protective properties, as well as the appearance of water-whitening effect. The latter phenomenon is caused by light scattering induced by water clusters formed in the film interior due to water penetration [82]. The results of the water-whitening measurements are presented in Fig. 9. It was found that the higher the content of the introduced BM, the lower the water-whitening of the coating films. Moreover, employing high amounts of BM (25 and 30 wt.%) can dramatically improve the resistance of coatings against water whitening, which extends the possibility of their use in various outdoor applications. This phenomenon can be related to the introduced cross-linking that does not allow the water clusters to grow large enough to scatter light visible to the human eye [83].