Influence of N-vynilcarbazole on the photopolymerization process and properties of epoxy-acrylate interpenetrating polymer networks

2.1 Materials

To synthesize simultaneous epoxy-acrylate IPNs cycloaliphatic-aliphatic epoxide 1-(2′,3′-epoxypropoxymethyl)-1-(2″,3″-epoxypropoxymethyl)-3,4-epoxy-cyclohexane (Epoxide) (UkrSRIPM, Donetsk, Ukraine) with two types of epoxy groups [cycloaliphatic and aliphatic (glycidyl) ones] was used as an epoxy component, triethylene glycol dimethacrylate (TEGDM) (95%, Sigma Aldrich, USA) – as an acrylate component, the mixture of triphenilsulfonium hexafluorophosphate salts (TPSHFPS) (Sigma Aldrich, USA, 50% solution in propylene carbonate) – as a photoinitiator, and NVC (98%, Sigma Aldrich, USA) – as a photosensitizer. The chemical structures of given substances are depicted below.

All materials were used as received.

2.2 Sample preparation

Epoxy-acrylate formulations were prepared in a weight ratio of epoxy and acrylate components of 50:50 with the formation of 50:50 IPN (for DMA analysis also 25:75 IPN and 75:25 IPN, and neat polymer networks), the amount of TPSHFPS was 4 wt%, and NVC – 2 wt%. The films with the thickness of 160 μm on the base of IPNs were formed between two glass plates previously covered with an antiadhesive agent. The formulations were irradiated with a set of four UV-lamps Philips TL2001 8W (I=1.0 mW/cm² at a sample position) in laboratory conditions and with sunlight in natural conditions (50°27′00″N, 30°31′25″E). In autumn cloudy weather the average intensity of UV-light was low (I=0.15±0.05 mW/cm²), but the temperature was very high (t=21±1°C). The intensity of UV-radiation (I, mW/cm²) was measured by UV-radiometer “UV Light Meter UV 340 B” in a range of 280–380 nm.

2.3 Methods

3.1 Investigation of photopolymerization kinetics

As a photosensitizer NVC provides electron transfer sensitization and additionally increases the rate and the completeness of epoxy polymerization due to its own intensive polymerization via cationic mechanism (21), (22), (23). The interaction of NVC with a photoinitiator under UV-radiation is depicted using the example of triphenilsulfonium hexafluorophosphate.

Absorbing UV-light by the molecule of NVC leads to its photoexcitation [absorbance spectrum of photosensitizer is given in a Figure 1 (24)]. Furthermore, its interaction with the molecule of the initiator results in the formation of an exciplex that undergoes an electron transfer to yield the carbazole cation radical and the triphenilsulfonium radical. Rapid recombination of free radicals prevents reverse electron transfer (21), (22). Accordingly, the electron-transfer photosensitization implies the redox reaction between the sensitizer and initiator.

Figure 1:

UV spectrum of NVC.

Furthermore, NVC is known to be able to generate free radical species, one part of which polymerizes, and another part is oxidized by the initiator with the formation of cationic species. These carbazole cations can induce the ring-opening reaction of epoxy groups or polymerize via cationic mechanism with the formation of NVC polymer (21), (23):

The additional amount of cations and free radicals except those generated during the photolysis of initiator is expected to speed up the polymerization both via cationic and free radical mechanisms (21). Therefore, by the example of epoxy-acrylate formulation in the component ratio of 50:50 the kinetics of IPNs formation in the presence of NVC was studied. For this purpose the sample was irradiated with sunlight in autumn cloudy weather during 1 hour. A parallel line assay was also conducted for similar system without NVC.

From IR-spectra obtained (given for IPNs before and after sunlight-irradiation in Supplemental Material) the degree of conversion of different functional groups (α, %) were calculated. The resulting kinetic curves for double bonds and two types of epoxy groups are demonstrated in Figure 2.

Figure 2:

Polymerization profiles of: double bonds (blue); cycloaliphatic epoxy groups (red); aliphatic epoxy groups (black) with NVC addition (solid lines) and without it (dash lines).

As it can be seen from Figure 2, the polymerization of acrylate component proceeds very intensively even without a photosensitizer, the final conversion degree of double bonds reaches 90%. Contrasted with acrylate polymerization, the polymerization profiles of both types of epoxy groups are characterized with a slow rate, the presence of an extended induction period, and low final conversion degrees: 19% for aliphatic and 38% for cycloaliphatic epoxy groups. However, by addition of NVC the results of epoxy groups photopolymerization are improved significantly. Particularly, in the system with the addition of NVC (Figure 2, solid lines) conversion degrees increase to 67 and 93% for aliphatic and cycloaliphatic epoxy groups, respectively. Moreover, the rate of epoxy photopolymerization rises, and the duration of induction period shortens. At the same time, despite the fact that NVC is able to speed up the polymerization via free radical mechanism, the obtained kinetic profiles of double bonds in the presence and absence of a photosensitizer are very similar. This can be attributed to the low reduction potential of NVC-radical and almost complete its oxidation to a cation (23).

3.2 Phase morphology of IPNs

According to the data obtained using the method of optical microscopy there are no phase heterogeneities in the investigated specimens of IPNs and neat polymer networks. Such results are frequently observed for single-phase systems. However, the absence of the features of phase heterogeneity in data obtained by optical microscopy data can also be inherent for phase-separated system in two cases: in systems with similar refractive indices of the neat components of the mixture or in systems with the dimensions of phase heterogeneities being less than the wavelength of visible light (≈500 nm). Therefore, the further investigation of IPNs phase morphology was conducted using SEM. The digital images of the fractured surfaces of the samples of neat polymer networks and of the samples of IPNs with and without the addition of NVC that were obtained by using SEM are presented in the Figure 3.

Figure 3:

SEM images of the fractured surfaces of: (A) Epoxide; (B) TEGDM; (C) 50:50 IPN; (D) 50:50 IPN (+NVC).

On the obtained images one can see elongated chips, small rounded particles of the film fractured near these chips, and some surface roughness, but not the structure heterogeneities. This fact can be explained by the absence of phase separation or by too small dimensions of such phase heterogeneities for registration by given microscope resolution.

For more detailed investigation of phase morphology specificities of IPNs the method of DMA was used. The results of DMA, which specification is to study the viscoelastic behavior of materials, can be used for T_g determination by α-relaxation transitions (from loss modulus or tan δ dependencies). These data may be useful in the characterization of phase morphology of multicomponent systems, particularly IPNs (25), (26), (27). Temperature dependencies of tan δ for neat polymer networks and for IPNs without NVC, which for better comparison and reproducibility were synthesized under stable laboratory conditions and irradiation of a UV-lamp, are given in Figure 4.

Figure 4:

Tan δ versus temperature for: 1 – epoxide (red); 2 – 50:50 IPN (black); 3 – TEGDM (blue).

As it can be seen from the Figure 4, the value of T_g of IPN (138°C) is positioned between T_g of neat components (119°C for neat epoxide and 177°C for neat acrylate). The presence of single α-relaxation transition of IPN confirms the formation of single-phase system (1). This can be attributed to the appearance of the forced compatibility of components in the process of their mutual fast curing. For quantitative estimation of components compatibility the Fox equation (28) is generally used:

where ω₁ and ω₂ are the mass fractions of polymer 1 and 2, respectively, Tg10 and Tg2o are glass transition temperatures of the individual components 1 and 2.

The results of calculations for investigated samples and for the samples with NVC content, which tan δ-T dependencies will be presented later (Figure 5), are given in a Table 1 together with their final conversion degrees (after 40 min of irradiation) according to IR-spectroscopy measurements.

Figure 5:

Tan δ versus temperature for NVC-containing: 1 – epoxide; 2 –75:25 IPN; 3 –50:50 IPN; 4 –25:75 IPN; 5 –TEGDM.

The obtained theoretical T_g calculated for 50:50 IPN (Table 1) is very close to the experimental one that implies the good compatibility of the epoxy and acrylate components.

The effect of NVC addition on the phase morphology of IPNs was analyzed for IPNs with different component ratios. Temperature dependencies of tan δ are presented in Figure 5.

Analyzing the systems with the presence of NVC it should be mentioned that for 75:25 IPN experimental and theoretical T_g completely coincide (Table 1), whereas for 25:75 IPN they are very different. This fact may be explained taking into account the difference between T_g of the neat polymer networks with and without the addition of NVC (Table 1). It indicates that the presence of NVC has less influence on the topology of the acrylate network (the difference in respective T_g is only 2°C) than on topology of the epoxy one (the difference in T_g is 13°C). Such a phenomenon may be observed due to the potential copolymerization of NVC with the epoxide. Thus, neglecting the effect of NVC for 25:75 IPN with the low content of epoxide, the calculation of T_g using the Tg10 in equation 1 of the neat epoxy polymer network without NVC gives the meaning of T_g (shown in Table 1 in brackets) almost equal to the experimental one.

But the most important result of the addition of NVC to the composition of initially single-phase IPN is that it evokes the appearance of two distinct maxima on the curve of tan δ only in the 50:50 IPN. At the same time the values of T_g for both epoxy and acrylate phases shift to higher temperatures, which is also demonstrated in a slight increase of the samples stiffness. As was evidenced before, the addition of NVC significantly increases the rate of cationic polymerization. Such a change in the kinetics of IPN formation leading to the faster polymerization of the components and subsequent earlier appearance of their incompatibility may promote phase separation in the system. The other reasons of phase separation are assumed to be not kinetic but rather thermodynamic ones and connected with the change in topology of the epoxy polymer network. Such changes may be attributed to the possible copolymerization of NVC with the epoxide and the appearance of bulk carbazole fragments decreasing the compatibility with the acrylate network. The occurrence of phase separation in the system of only one certain composition (50:50 IPN), where the fractions of both components are large enough to form their own individual phases, confirms the fact that the change in composition of IPNs is the effective way to control the material morphology and properties.