Enhancement of the mechanical properties at the macro and nanoscale of thermosetting systems modified with a polystyrene-block-polymethyl methacrylate block copolymer

Abstract

A diglycidyl ether of bisphenol A (DGEBA) epoxy monomer-based thermosetting system modified with a varying content of a polystyrene-block-polymethyl methacrylate (PS-b-PMMA) block copolymer and cured with a 4,4′-methylenebis(3-chloro-2,6-diethylaniline) (MCDEA) curing agent was prepared by two different methods, without and with a solvent for the previous solution of the PS-b-PMMA block copolymer. The influence of the modifier content as well as the preparation method on the final properties of the PS-b-PMMA/(DGEBA-MCDEA) cured systems was investigated. DSC characterization confirmed the miscibility of the block copolymer with the thermoset matrix. Regardless of the preparation method, all analyzed thermosetting systems modified with PS-b-PMMA showed a nanostructured morphology, as confirmed by AFM measurement. The mechanical properties studied by a universal testing machine (MTS) at the macroscale were enhanced by the addition of the PS-b-PMMA block copolymer, especially in terms of the fracture toughness which improved considerably up to 25 wt% PS-b-PMMA content, although also the flexural modulus presented a slight increase with the modification with the PS-b-PMMA block copolymer. The quantitative nanomechanical (QNM) properties were studied at the nanoscale by AFM in PeakForce mode, and showed an improvement of the elastic modulus of the PMMA/(DGEBA-MCDEA) matrix up to 25 wt% PS-b-PMMA content. The QNM results allowed also the detection of a high difference between the elastic modulus values corresponding to the microseparated PS block rich phase and the PMMA/(DGEBA-MCDEA) matrix. A similar tendency was exhibited for both preparation methods, although the thermosetting systems prepared with the solvent reached slightly higher elastic moduli.

---

Introduction

Epoxy networks are a type of thermosetting material that often exhibit good mechanical properties, excellent chemical and corrosion resistance and good dimensional stability. Due to these desirable properties among others, epoxy networks have been extensively produced in the last decades to be utilized for coatings, corrosion protection, electrical insulation, fiber reinforced composites and adhesives.^1,2 Nevertheless, as a result of their high cross-link density achieved during the curing process, these materials tend to be rather brittle having low impact and fracture strengths, with this being the main drawback for some applications. In recent years, many researchers have focused their work on the enhancement of the toughness of epoxy networks.^2,3 One of the efficient ways to make the epoxy-based thermosets tougher is to modify the original epoxy monomer by the incorporation of a second phase into the continuous matrix of the epoxy-based thermoset through physical blending or chemical reactions. The addition of modifiers can convert the epoxy-based thermoset into multiphase systems and in the case when the modifier is suitably dispersed through the matrix, the fracture toughness could be significantly increased. Highly cross-linked thermosetting systems have a limited ability to be deformed by shielding, and the incorporation of another component could reduce the cross-linking density, leading to an improved toughness.

Many kinds of modifiers have been employed for this purpose, such as thermoplastic homopolymers, liquid rubbers, reactive diluents, inorganic particles, etc.^4–6 Here the attention is focused on block copolymers as modifiers to produce nanostructured domains in thermoset materials.^7–9 The blocks of a block copolymer usually present a different affinity towards a solvent, and also, they show a tendency to avoid the mixing of dissimilar blocks between each other. Consequently, they form well ordered structures such as spheres, worm-like micelles, vesicles and core–shell structures with domain sizes typically on the scale of nanometers, and this makes block copolymers excellent materials to create nanostructured thermosetting systems with improved toughness and without any considerable negative effect on the properties of the epoxy systems. One of the approaches to achieve a nanostructured morphology requires a block of the block copolymer to be immiscible with the thermosetting system and another block to be miscible with it up to a high content. Depending on the solubility of the different blocks of the block copolymer, the self-assembly of the block copolymer can take place before curing, or during the curing reaction, by a mechanism known as reaction induced phase separation (RIPS) where one of the blocks undergoes phase separation as the polymerization reaction proceeds due to the increasing immiscibility with the thermosetting matrix.

Some of the block copolymers most commonly used in the literature for the toughening of thermosetting systems have been poly(ethylene oxide)-b-poly(ethylene-alt-propylene) (PEO-b-PEP),⁷ poly(ethylene oxide)-b-poly(propylene oxide)-b-poly(ethylene oxide) (PEO-b-PPO-b-PEO),¹⁰ polystyrene-b-polybutadiene (PS-b-PB),¹¹ and polystyrene-b-polybutadiene-b-poly-(methyl methacrylate) (PS-b-PB-b-PMMA),¹² among others. In general, the most widely employed epoxy miscible blocks for this purpose have been poly(ethylene oxide) and poly(methyl methacrylate). Polystyrene-block-polymethyl methacrylate (PS-b-PMMA) block copolymer have also been employed in nanostructured thermosetting systems. Some authors have already investigated the miscibility and phase behavior of PS-b-PMMA block copolymers with different epoxy systems^13–15 and also the effect of the PS-b-PMMA content,^14,16 the PS-b-PMMA topological and sequential structures¹⁷ and the curing conditions¹³ on the morphologies obtained when modifying an epoxy system. Zucchi et al. also studied the mechanical properties of an epoxy-based thermoset modified with a PS-b-PMMA block copolymer, in terms of the elastic modulus and yield stress.¹⁸

The current work investigates the use of a PS-b-PMMA diblock copolymer as a modifier of a diglycidyl ether of bisphenol A (DGEBA)-based epoxy matrix with the aim of obtaining nanostructured thermoset systems with improved mechanical properties. Two different ways of preparing the epoxy-based thermosets modified with a PS-b-PMMA block copolymer were analyzed and compared with each other, one by using a solvent for the solution of the block copolymer and the second by dissolving the block copolymer directly into the epoxy monomer. The effect of the solvent as well as the effect of the PS-b-PMMA block copolymer content on the final properties of the materials have been investigated. The morphologies of all the cured thermosetting systems obtained with a varying content of the PS-b-PMMA block copolymer from 5 to 50 wt% were analyzed by atomic force microscopy (AFM). The glass transition temperatures were determined by differential scanning calorimetry (DSC). The mechanical properties at the macroscale were studied by the universal testing machine (MTS) in terms of the flexural behavior and fracture toughness measurements, about which no evidence was found in the literature, showing a clear improvement of both parameters thanks to the modification of the epoxy system with the block copolymer. Additionally, the quantitative nanomechanical (QNM) properties of the designed thermosetting systems at the nanoscale were studied using the PeakForce mode of AFM.

Experimental part

Materials

A diglycidyl ether of bisphenol A (DGEBA) (DER 330) epoxy monomer, with an epoxy equivalent weight of 176–185 g eq.⁻¹, was provided by Dow Chemical Company. The curing agent used to cure this epoxy monomer was 4,4′-methylenebis(3-chloro-2,6-diethylaniline) (MCDEA), supplied by Lonza. A polystyrene-block-polymethyl methacrylate (PS-b-PMMA) diblock copolymer, with polymethyl methacrylate rich in the syndiotactic content (>78%), was purchased from Polymer Source, Inc. It had a polydispersity index (M_w/M_n) of 1.09 and a number-average molecular weight of each of the PS and PMMA block of 80 000 g mol⁻¹. Chloroform was purchased from Labscan and used as solvent.

Blending protocol

The PS-b-PMMA/(DGEBA-MCDEA) cured systems were prepared by two different methods. In the first method (from now denoted as the non-solvent method), a certain amount of the PS-b-PMMA block copolymer was dissolved in DGEBA, by manual mixing and by heating the blend at 150 °C, in order to ease the solution. When complete homogenization was achieved, a stoichiometric amount of MCDEA was added and manually stirred for 5 minutes. In the second protocol (from now denoted as the solvent method), the PS-b-PMMA block copolymer was first dissolved in chloroform at a 10 mg mL⁻¹ concentration at ambient temperature and then mixed with DGEBA. The mixture was heated at 150 °C until evaporation of the solvent was reached. At this point, a stoichiometric amount of MCDEA was added and manually stirred for 5 minutes. The homogeneous mixtures obtained after the addition of the curing agent were poured into a glass mold and degassed at 150 °C under vacuum. All the thermosetting systems were cured at 190 °C for 6 h. Apart from the neat DGEBA-MCDEA system, four thermosetting systems were prepared with each of the two protocols, with 5, 15, 25 and 50 wt% of the PS-b-PMMA block copolymer. All the plates were 1.5 mm thick.

Techniques

Differential scanning calorimetry (DSC) measurements of the neat DGEBA-MCDEA cured system as well as the PS-b-PMMA/(DGEBA-MCDEA) cured systems were performed using a Mettler Toledo DSC 822^e differential scanning calorimeter. The thermal transition temperatures of the cured blends were determined as the midpoint of the glass transition temperature range, by dynamic scans performed from 20 to 200 °C with a heating rate of 5 °C min⁻¹. Prior to this scan, heating from 20 to 200 °C followed by cooling from 200 to 20 °C was carried out in order to delete the thermal history of the investigated materials. All experiments were conducted under a nitrogen flow of 10 mL min⁻¹ using 10–15 mg samples in aluminum pans.

The morphologies of the neat DGEBA-MCDEA cured system and PS-b-PMMA/(DGEBA-MCDEA) cured systems were studied by atomic force microscopy (AFM) under ambient conditions. AFM images were obtained using a scanning probe microscope (Nanoscope IIIa Multimode™, Digital Instruments). Tapping mode (TM) was employed in air using an integrated tip/cantilever (125 μm in length with ca. 300 kHz resonance frequency). The typical scan rates during recording were 0.7–1 line per s using a scan head with a maximum range of 16 μm × 16 μm. The transverse cross section surface of each investigated thermosetting system was cut using an ultramicrotome Leica Ultracut R with a diamond blade.

Regarding the mechanical properties of the neat DGEBA-MCDEA cured system and PS-b-PMMA/(DGEBA-MCDEA) cured systems, flexural tests were carried out following the ASTM D790-10 standard test method and using a universal testing machine (MTS, model Insight 10) provided with a 250 N load cell. Three-point bending tests without notch were performed with a support span of 22 mm at a rate of crosshead of 0.5 mm min⁻¹ using specimen dimensions of 27 × 6 × 1.5 mm³ (rectangular shape). The flexural modulus was determined from the slope of the load–displacement curve in the zone of linear elasticity. Fracture toughness tests were performed according to the ASTM D5045-99 standard test method using the same universal testing machine as that for the flexural tests. Fracture toughness was measured by determining the critical stress intensity factor (K_Ic) and the critical strain energy release rate (G_Ic) in three-point bending with a support span of 24 mm at a crosshead rate of 10 mm min⁻¹ using single edge notched specimens (SENBs) with dimensions of 27 × 6 × 1.5 mm³. Initially a sharp notch of around 2.7 mm was made by machining, and subsequently a natural crack was initiated using a razor blade. G_Ic was calculated from the energy derived from integration of the load versus displacement curve up to the same load point as that used for K_Ic. For both the flexural and fracture toughness tests more than five specimens for each system were tested. The 50 wt% PS-b-PMMA/(DGEBA-MCDEA) cured systems could not be analyzed in terms of their mechanical properties due to the difficulty in obtaining a continuous sheet after curing.

PeakForce quantitative nanomechanical property mapping (PeakForce QNM) was used to study the mechanical properties of the designed thermosetting systems at the nanometric scale. Height, adhesion and modulus PeakForce QNM images were simultaneously captured using a Dimension Icon AFM microscope from Bruker. Measurements were carried out in PeakForce mode under ambient conditions. A silicon tip with a nominal radius of 10 nm, a cantilever length of 125 μm, and a resonance frequency of 150 kHz was used. The measurements were performed with calibrated optical sensitivity. The exact spring constant of the tip was calculated using the Thermal Tune option and a defined tip radius was adjusted using PS as standard.

Results and discussion

Different contents of the PS-b-PMMA block copolymer were incorporated into the DGEBA-MCDEA system by following two different procedures. Before curing, all the mixtures of the epoxy monomer and the block copolymer were transparent and homogeneous, suggesting that there was no macrophase separation. It should be pointed out that in the case where no solvent was used, DGEBA acted as the solvent for the block copolymer, being a good solvent for a low content of the block copolymer, whereas for the highest content the stirring had to be done for a longer time to reach a homogeneous solution. After curing, all the investigated thermosetting systems continued to be transparent and homogeneous, except for the thermosetting systems containing 50 wt% content of the PS-b-PMMA block copolymer. The visual appearance of the neat DGEBA-MCDEA cured system and of the PS-b-PMMA/(DGEBA-MCDEA) cured systems up to a 25 wt% content of the block copolymer is shown in Fig. 1, where it is possible to see that the transparency of the thermosetting systems only undergoes a slight decrease with the addition of the PS-b-PMMA block copolymer, but in all cases they are still transparent. On the other hand, in the case of the 50 wt% PS-b-PMMA/(DGEBA-MCDEA) cured systems, it was not possible to obtain a continuous sheet after curing as the thermosetting systems prepared by both procedures had a high viscosity which prevented the formation of a continuous and homogeneous plate. However, it should be pointed out that the prepared non-continuous sheets were also transparent, indicating the absence of macrophase separation. In the cases of the samples with 15 wt% PS-b-PMMA (solvent method) and 25 wt% PS-b-PMMA (non-solvent method), it should be noted that, even if they are still transparent, a darker color is observed for them. This fact did not have any relation with the structure of the sample, but with the superficial effect of the curing when the mold of the samples was not hermetically sealed.

Fig. 1 Visual appearance of the neat DGEBA-MCDEA cured system (a), the 5 wt% (b), 15 wt% (d) and 25 wt% (f) PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the non-solvent method, and the 5 wt% (c), 15 wt% (e) and 25 wt% (g) PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the solvent method.

The miscibility and thermal transitions of the neat DGEBA-MCDEA cured system and PS-b-PMMA/(DGEBA-MCDEA) cured systems were studied by DSC analysis. Fig. 2 shows the DSC thermograms of the second heating scan applied to each investigated sample. As can be seen, the neat DGEBA-MCDEA cured system presented one glass transition temperature (T_g) at 174 °C, whereas the neat block copolymer presented two T_g values, one at 105 °C corresponding to the PS block and the other one at 131 °C corresponding to the PMMA block. The thermograms obtained for all the studied thermosetting systems prepared by the two employed methods presented one clear T_g, which indicated that the block copolymer is partially miscible with the thermosetting system and consequently the cured systems exhibited one T_g located somewhere between the one of the neat DGEBA-MCDEA cured system and the one of the block copolymer. In addition, the miscibility of the components was also confirmed by the fact that this T_g shifted to lower temperatures as the addition of the block copolymer increased, as is expected from a system based on a thermosetting system modified with a block copolymer.^16,19 Some authors have already reported that the miscibility of the PMMA block with the epoxy is higher than that of the PS block with the epoxy^14,15,17 and it has been proven that PMMA is miscible with the DGEBA-MCDEA system up to the end of the curing reaction²⁰ whereas the PS block microseparates before the gel point.²¹ Moreover, the Fox theoretical equation was used for the calculation of the theoretical T_g values of the PS-b-PMMA/(DGEBA-MCDEA) cured systems¹⁰ and these values are also shown in Fig. 2. First it should be pointed out that the employment of this equation is just an approximation, since it only considers binary mixtures and for this calculation only the PMMA block and neat DGEBA-MCDEA cured systems were contemplated, since the PMMA block is the one that is miscible with the thermosetting system. The fact that the PS T_g was not taken into account for the estimation could lead to an error in the calculated values. Nevertheless, here it should be pointed out that the main goal of this calculation was just to compare the tendency of the T_g values with the theory, and in this case it was observed that as well as in the experimental results, the theoretical values tended also to decrease with the addition of the block copolymer. For a low content of the block copolymer the experimental values fitted well with the theoretical ones. However, for the highest content like 50 wt%, the experimental values differed quite a lot from the theoretical ones, mainly because of the error in the application of the Fox equation as commented above. On the other hand, by comparing the systems prepared by the two methods, it could be pointed out that the T_g values of the systems prepared by the solvent method were slightly lower than the T_g values of the systems prepared without solvent up to a 25 wt% PS-b-PMMA block copolymer content, indicating that a higher miscibility was reached when the solvent method was employed, probably as a result of a better solution of the block copolymer in DGEBA. In the case of 50 wt% PS-b-PMMA, however, the T_g of the non-solvent system was 13 °C lower than the T_g of the sample prepared using the solvent, and both of them were lower than the T_g of all the neat components. This could be explained considering that at a 50 wt% PS-b-PMMA block copolymer content, it was not possible to obtain a homogeneous and continuous sheet, therefore the material might not show the expected thermal behavior for this block copolymer content. In addition to this, even if we did not study the kinetics of these investigated systems, we are sure that 6 h at 190 °C is enough time to complete the curing process. Here it should be pointed out that the low T_g can be related to the fact that a high content of the block copolymer delays the curing reaction, due to the dilution effect caused by the addition of the PS-b-PMMA block copolymer.^19,22 It should be also noted that in the systems with a content of the PS-b-PMMA block copolymer from 5 to 25 wt%, apart from the clearly detected T_g, another subtle increase in the heat flow could be perceived at temperatures lower than that of the clear T_g, which could be related to the glass transition of the microseparated PS block rich phase. Nevertheless, the content of this PS block in the final systems is so low that this transition cannot be detected by DSC.

Fig. 2 Dynamic DSC thermograms of the second heating scan of the neat DGEBA-MCDEA cured system, neat PS-b-PMMA block copolymer and the 5, 15, 25 and 50 wt% PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the non-solvent method (a), and by the solvent method (b). The dotted line in each graph indicates the theoretical T_g values calculated by the Fox equation.

The neat DGEBA-MCDEA cured system as well as the PS-b-PMMA/(DGEBA-MCDEA) cured systems were analyzed in terms of their morphology by means of AFM. The AFM phase image of the neat DGEBA-MCDEA system is shown in Fig. 3. It can be clearly observed that the morphology is regular and uniform with no visible separation at the micro and macroscale, as is expected from the neat DGEBA-MCDEA cured system.

Fig. 3 AFM phase image (1 μm × 1 μm and inset of 3 μm × 3 μm) of the neat DGEBA-MCDEA cured system.

The morphologies of the PS-b-PMMA/(DGEBA-MCDEA) cured systems containing 5, 15, 25 and 50 wt% of the block copolymer and prepared by following the two different procedures are presented in Fig. 4. The four images above correspond to the non-solvent method, whereas the images below correspond to the method that employs solvent for the mixing of the block copolymer and the DGEBA monomer.

Fig. 4 AFM phase images (1 μm × 1 μm and inset of 3 μm × 3 μm) of the 5 wt% (a), 15 wt% (b), 25 wt% (c) and 50 wt% (d) PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the non-solvent method and the 5 wt% (e), 15 wt% (f), 25 wt% (g) and 50 wt% (h) PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the solvent method.

It can be seen that all the analyzed thermosetting systems containing from 5 to 50 wt% PS-b-PMMA exhibited nanostructured morphologies. Before the curing reaction, it was observed that the PS-b-PMMA block copolymer was miscible with the DGEBA epoxy monomer. As mentioned above, DGEBA has a higher miscibility with the PMMA block than with the PS block,^14,15,17 as also was determined by the calculation of the Flory–Huggins interaction parameters of both pairs of compounds by means of the Hoftyzer and van Krevelen method.²³ Therefore, during the curing reaction, the PS block started to separate from the DGEBA-MCDEA matrix while the PMMA block remained miscible with it. It is due to the existence of the chemical bond between both the PS and PMMA blocks of the block copolymer that the macrophase separation of the PS block did not take place, occurring instead only a microphase separation. Looking at Fig. 4, it should be first mentioned that all images present some dark domains of varying shape and size depending on the content of PS-b-PMMA, separated from a clearer continuous phase. The separated dark domains were attributed to PS blocks, whereas the continuous matrix was attributed to a DGEBA-MCDEA cured phase mixed with PMMA blocks.¹⁶ The difference between the contrasts of the two phases is related to the difference in the viscoelastic character of them. Therefore, it can be claimed that the PMMA block/(DGEBA-MCDEA) rich phase is the one that presents the highest modulus.

The morphology started to be spherical for the lowest content of the block copolymer, namely for the 5 and 15 wt% PS-b-PMMA content, with the number of domains being slightly higher in the case of a 15 wt% PS-b-PMMA content and also in the thermosetting systems prepared by the non-solvent method in comparison to the thermosetting systems prepared by the solvent method. Regarding the size of the nanodomains, the ones in the thermosetting systems prepared by the non-solvent method (30–40 nm for 5 wt% PS-b-PMMA and 35–45 nm for 15 wt% PS-b-PMMA) were also slightly larger than those prepared with solvent (20–35 nm for 5 wt% PS-b-PMMA and 30–45 nm for 15 wt% PS-b-PMMA). These differences between the two preparation methods were somehow expected since the use of the solvent could have helped to reduce the viscosity and to achieve a better solution of the block copolymer, and therefore a better miscibility between the block copolymer and the epoxy monomer was reached, leading to a fewer number of separated domains and smaller ones. On the other hand, when the incorporated block copolymer amount increased to 25 and 50 wt%, the spherical morphology disappeared in the case where no solvent was used, showing a coexistence of bigger spheres and interconnected domains like cylinders for the 25 wt% content and a less clearly nanostructured morphology for the 50 wt% content of PS-b-PMMA. The increase in the size of the separated domains was due to the higher content of the block copolymer and consequently a higher amount of the PS block tended to separate from the matrix, taking with it part of the PMMA block present in the matrix, but always without reaching a macrophase separation as was confirmed by the transparency of the samples with a 50 wt% block copolymer content (not shown here) and by the DSC results. However, the employment of solvent in the preparation of the samples led to a still uniform and spherical structure for a high content of the block copolymer like 25 wt% (with bigger spheres of 55–75 nm) and almost spherical with some interconnected domains like cylinders for the 50 wt% content. Consequently, it should be pointed out that the use of the solvent maintained the regular nanostructured morphology up to a higher content of the block copolymer in comparison with the non-solvent method. The morphologies obtained by the addition of the PS-b-PMMA block copolymer to the thermosetting matrix by both employed methods have been schematically represented in Fig. 5.

Fig. 5 Schematic representation of the morphologies obtained for thermosetting systems modified with a low and high PS-b-PMMA block copolymer content.

The mechanical properties of all the thermosetting systems, except for the 50 wt% PS-b-PMMA/(DGEBA-MCDEA) cured systems, were studied in terms of the flexural modulus (E), the critical stress intensity factor (K_Ic) and the critical strain energy release rate (G_Ic). The flexural moduli of the neat DGEBA-MCDEA cured system and of the DGEBA-MCDEA cured systems modified with a different content of the PS-b-PMMA block copolymer are presented in Fig. 6a. It is observed that the cured DGEBA-MCDEA system presented a flexural modulus of 2500 MPa which is in accordance with the values reported in the literature for this system.^24,25 Moreover, the flexural modulus value increased when the thermosetting system was modified with the PS-b-PMMA block copolymer. In the case of the method without solvent, the highest value of the modulus was obtained for the 15 wt% PS-b-PMMA content, whereas in the case of the method using solvent, the 25 wt% PS-b-PMMA/(DGEBA-MCDEA) cured system was the one that reached the highest value of the flexural modulus. Taking into account the AFM phase images of these two systems (Fig. 4), it could be claimed that both systems were the ones that presented the most regular spherical morphology among all the studied systems, reaching a quasi hexagonal morphology. In any case, it should be also taken into account that the differences among the moduli obtained by the two employed methods and with the different concentrations of the block copolymer are not significant if the error bars are taken into account. However, the general tendency would be that the higher the amount of the block copolymer, the higher the obtained modulus. In general, the opposite effect has been observed in the literature, as it is known that the addition of a block copolymer in a content higher than 10 wt% tends to plasticize the thermoset matrix leading to a lower flexural modulus.^{10,11,19,24,26} In this case, an improvement in the flexural modulus occurred at least up to the 25 wt% PS-b-PMMA content, probably due to the fact that even if the addition of the PS-b-PMMA block copolymer reduced the cross-linking density of the epoxy network, the modifier employed in this case did not have a modulus much lower than that of the neat thermosetting system¹⁸ and it had a positive contribution in the flexural behavior of the epoxy. In addition, the increase in the flexural modulus could be also related to a decrease in the free volume provoked by the incorporation of the block copolymer.¹⁰ Therefore, these results confirm the miscibility between the PMMA block and the epoxy system, which is high enough to reach such an increase in the flexural behavior of the epoxy matrix.

Fig. 6 The flexural modulus (E) (a), the critical stress intensity factor (K_Ic) (b), and the critical strain energy release rate (G_Ic) (c), of the neat DGEBA-MCDEA cured system and all the PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the non-solvent and the solvent method.

Regarding the toughness of these analyzed systems, the values of K_Ic and G_Ic were calculated and are shown in Fig. 6b and c, respectively. The toughness in terms of K_Ic was maintained in respect to the neat DGEBA-MCDEA cured system with the addition of a 5 wt% PS-b-PMMA content to the thermoset matrix. However, with higher incorporated amounts of the PS-b-PMMA block copolymer, with both 15 and 25 wt% contents, the K_Ic values increased considerably in respect to the toughness of the neat DGEBA-MCDEA cured system. In this case, no appreciable difference was observed between the non-solvent and the solvent methods. The toughness of the samples not only increased with the addition of the block copolymer, but also increased more as the content of the block copolymer was higher. This confirms that the modification of the thermosetting system with the PS-b-PMMA block copolymer was worthy from the point of view of an enhancement in the mechanical properties of the neat DGEBA-MCDEA system, of both the flexural and toughness behaviors, but especially of the toughness, as a low toughness is one of the known drawbacks of epoxy matrices. Unmodified thermosetting systems are usually single-phase materials, meanwhile the addition of modifiers can turn them into multiphase systems, which is the case of the PS-b-PMMA block copolymer modifier. As reported by many authors, when the modifier domains are correctly dispersed throughout the epoxy matrix, the fracture toughness can be greatly improved.^2,10,11,27 Consequently, the existence of an important relation between the morphology in the nanoscale and the fracture toughness is also quite known.^7,10,26 In this case, the bigger spherical domains observed in the 25 wt% PS-b-PMMA content, which coexisted with some cylinders in the case of the non-solvent method, resulted in the highest toughness values. In addition, as can be seen in Fig. 6c, the critical strain energy release rate (G_Ic) showed a similar tendency in comparison to the K_Ic values, where G_Ic maintained the value of the neat DGEBA-MCDEA system for the 5 wt% content of the block copolymer and also increased considerably for the 15 and 25 wt% PS-b-PMMA block copolymer content.

The quantitative nanomechanical (QNM) properties of the thermosetting systems prepared without and with solvent were investigated using AFM in PeakForce mode. The height, the adhesion and the elastic modulus PeakForce QNM images of the investigated thermosetting systems are shown in Fig. 7 and 8 for the non-solvent and solvent methods, respectively. First of all it should be pointed out that for both the non-solvent and solvent preparation methods, the thermosetting systems modified with the PS-b-PMMA block copolymer showed similar morphologies as described above. The average roughness (R_a), extracted from the height PeakForce QNM images of the thermosetting systems prepared by the non-solvent and solvent methods, increased with the increase of the PS-b-PMMA block copolymer content in both series, being 0.2 nm for the neat DGEBA-MCDEA cured system and 2.5 nm and 4.9 nm for the thermosetting systems modified with 50 wt% PS-b-PMMA prepared by the non-solvent and solvent methods, respectively. This rise can be provoked by the increase of the amount and size of microphase separated PS block domains with the addition of a higher PS-b-PMMA block copolymer content.

Fig. 7 PeakForce QNM images (1 μm × 1 μm): the neat DGEBA-MCDEA cured system height (aI), adhesion (aII), and modulus (aIII); the PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared without solvent containing 5 wt% height (bI), adhesion (bII), and modulus (bIII); 15 wt% height (cI), adhesion (cII), and modulus (cIII); 25 wt% height (dI), adhesion (dII), and modulus (dIII); and 50 wt% height (eI), adhesion (eII), and modulus (eIII).

Fig. 8 PeakForce QNM images (1 μm × 1 μm): the neat DGEBA-MCDEA cured system height (aI), adhesion (aII), and modulus (aIII); the PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared with solvent containing 5 wt% height (bI), adhesion (bII), and modulus (bIII); 15 wt% height (cI), adhesion (cII), and modulus (cIII); 25 wt% height (dI), adhesion (dII), and modulus (dIII); and 50 wt% height (eI), adhesion (eII), and modulus (eIII).

As expected, the adhesion PeakForce QNM image of the neat DGEBA-MCDEA cured system showed a very low, homogeneously distributed adhesion value on the whole investigated surface as compared with the adhesion of the thermosetting systems modified with the PS-b-PMMA block copolymer prepared without and with solvent. Thus, the adhesion value of the neat DGEBA-MCDEA cured system (∼0.5 N m) was more than 10 times lower in comparison with the adhesion values of the PS-b-PMMA/(DGEBA-MCDEA) cured systems (adhesion values in the range 5–30 N m). All the thermosetting systems modified with the PS-b-PMMA block copolymer showed two different adhesion values. The one with the lowest adhesion corresponds to the microphase separated PS block rich phase (darker areas), whereas the one with the highest adhesion value is related to the PMMA block/(DGEBA-MCDEA) rich phase (brighter areas). Here, it should be pointed out that generally the thermosetting systems modified with the PS-b-PMMA block copolymer in the absence of solvent possessed lower adhesion values compared to the thermosetting systems with the same PS-b-PMMA block copolymer content prepared with solvent. For both preparation methods, the adhesion value increased up to the 15 wt% PS-b-PMMA block copolymer content, showing only a slight difference between the values of the two different areas. For this content of the PS-b-PMMA block copolymer, in the case of the non-solvent method, the values of the PMMA block/(DGEBA-MCDEA) rich phase and PS block rich phase were ∼27 N m and ∼24 N m, respectively, whereas in the case of the solvent method these values were ∼32 N m and ∼27 N m for the PMMA block/(DGEBA-MCDEA) rich phase and the PS block rich phase, respectively. The adhesion values of the thermosetting systems modified with the 25 wt% PS-b-PMMA and higher PS-b-PMMA block copolymer content decreased and showed a higher difference between the adhesions corresponding to the matrix and to the microphase separated domains. For the 25 wt% PS-b-PMMA/(DGEBA-MCDEA) cured system prepared with the non-solvent method, the lowest adhesion value was ∼5 N m (PS block rich phase) while the highest adhesion value was ∼15 N m (PMMA block/(DGEBA-MCDEA) rich phase). The adhesion values of the 25 wt% PS-b-PMMA/(DGEBA-MCDEA) cured system prepared with the solvent method were ∼5 N m and ∼17 N m, for the lowest and the highest adhesion areas, respectively.

The elastic modulus of the neat DGEBA-MCDEA cured system had almost the same value in every measured point in the modulus PeakForce QNM image, this value being ∼2.3 GPa. Similarly to the adhesion PeakForce QNM images, the modulus PeakForce QNM images of the thermosetting systems modified with the PS-b-PMMA block copolymer without and with solvent revealed two different areas with different elastic modulus values. The highest elastic modulus corresponded to the PMMA block/(DGEBA-MCDEA) rich phase and the lowest one to the microphase separated PS block domains. For the PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared without solvent, the elastic modulus corresponding to the PMMA block/(DGEBA-MCDEA) rich phase (∼2.5 GPa) was slightly higher than the elastic modulus of the neat DGEBA-MCDEA cured system. Simultaneously, the modulus PeakForce QNM images allowed the detection also of the elastic modulus values of the microseparated PS block phase, which was much lower than the elastic modulus values of the PMMA block/(DGEBA-MCDEA) matrix for the thermosetting systems with the same PS-b-PMMA block copolymer content. In addition, the elastic modulus values of the microseparated PS block phase did not undergo any significant variation among the different investigated thermosetting systems. In fact, its values varied between 1.2 and 1.4 GPa, confirming that the microseparated PS block had a two times lower modulus than the PMMA/(DGEBA-MCDEA) rich phase. Here it should be highlighted that the elastic modulus of the PMMA block/(DGEBA-MCDEA) rich phase is higher than the elastic modulus of the neat DGEBA-MCDEA cured system, confirming that the PMMA block is partially miscible with the thermosetting system, and that it had a strong effect on the increase of the elastic modulus at the macroscopic scale on the contrary to the microphase separated PS block domains, which resulted in a lower local elastic modulus.

The elastic modulus of the PMMA block/(DGEBA-MCDEA) rich phase of the thermosetting systems modified with the PS-b-PMMA block copolymer prepared with the solvent method increased considerably with the increase of the PS-b-PMMA block copolymer content, being ∼2.6 GPa, ∼2.7 GPa and ∼2.8 GPa for the 5, 15 and 25 wt% PS-b-PMMA/(DGEBA-MCDEA) cured systems, respectively. However, the elastic modulus of the PMMA block/(DGEBA-MCDEA) rich phase of the 50 wt% PS-b-PMMA/(DGEBA-MCDEA) cured system was lower than that of the 25 wt% PS-b-PMMA/(DGEBA-MCDEA) cured system, although it was still higher than the neat DGEBA-MCDEA cured system. If the elastic modulus of the PMMA block/(DGEBA-MCDEA) rich phase of the thermosetting systems prepared without and with solvent are compared among them, one can conclude that the use of the solvent for the preparation leads to a higher elastic modulus of the PMMA block/(DGEBA-MCDEA) rich phase probably due to higher miscibility between the PMMA block and the thermosetting system in these systems.

The elastic modulus values of the PS block rich phase in the thermosetting systems prepared with solvent varied from 1.5 GPa to 1.7 GPa, being slightly higher than the elastic modulus values of the PS block rich phases corresponding to the thermosetting systems prepared without solvent. The elastic modulus of the PS block rich phase in the 50 wt% PS-b-PMMA/(DGEBA-MCDEA) cured system with solvent decreased to 1 GPa.

Conclusions

A PS-b-PMMA block copolymer has been effectively employed as a modifier of an epoxy-based thermosetting system, leading to systems with clearly improved final properties. Two different methods were followed to incorporate different contents of the PS-b-PMMA block copolymer into the DGEBA-MCDEA thermoset matrix. The visual transparency of the prepared sheets indicated the absence of macrophase separation in all the analyzed PS-b-PMMA/(DGEBA-MCDEA) cured systems for both preparation methods. On the other hand, AFM results showed a clear microphase separation in all the thermosetting systems modified with the PS-b-PMMA block copolymer, obtaining morphologies dependent on both the block copolymer content and the preparation method. In general, the PS block microphase separated domains increased in size and quantity with a higher content of the PS-b-PMMA block copolymer, changing from spherical domains to almost cylindrical interconnected ones. The miscibility between the PMMA block and the DGEBA-MCDEA system was proven by DSC, where it was observed that the addition of the block copolymer to the matrix provoked a decrease in the T_g of the thermoset matrix.

The mechanical properties of the neat DGEBA-MCDEA cured system and the PS-b-PMMA/(DGEBA-MCDEA) cured systems, investigated at the macroscale and nanoscale, demonstrated that the addition of the PS-b-PMMA block copolymer was able to enhance the mechanical properties of the investigated thermosetting system. The flexural modulus and the fracture toughness measured by the MTS increased with the addition of the PS-b-PMMA block copolymer up to a 25 wt% PS-b-PMMA content, with the enhancement of the fracture toughness being higher than that of the flexural modulus. The quantitative nanomechanical properties, studied by PeakForce QNM, showed an improvement in the elastic modulus of the thermosetting system, and in particular of the PMMA block/(DGEBA-MCDEA) rich phase, when the PS-b-PMMA block copolymer was incorporated into the epoxy-based matrix up to a 25 wt% PS-b-PMMA content. The PS-b-PMMA/(DGEBA-MCDEA) cured systems prepared by the solvent method led to slightly higher values of the fracture toughness and elastic modulus, due to a higher miscibility reached in these systems.

Acknowledgements

Financial support from the Basque Government (Grupos Consolidados IT776-13) and from the Spanish Ministry of Economy and Competitiveness and European Union (MAT2012-31675) is gratefully acknowledged. L. C. thanks the Basque Government for the PhD Fellowship (BFI-2011-218) and A. T. acknowledges the MICINN for the Ramón y Cajal program (RYC-2010-05592). Moreover, we are grateful to the ‘Macrobehavior-Mesostructure-Nanotechnology’ SGIker unit of the UPV/EHU.

References

1. D. Ratna, Handbook of Thermoset Resins, iSmithers Rapra Technology, Shrewsbury, United Kingdom, 2009.
2. L. Ruiz-Pérez, G. J. Royston, J. P. A. Fairclough and A. J. Ryan, Polymer, 2008, 49, 4475–4488.
3. N. Domun, H. Hadavinia, T. Zhang, T. Sainsbury, G. H. Liaghat and S. Vahid, Nanoscale, 2015, 7, 10294–10329.
4. J. H. Park and S. C. Jana, Polymer, 2003, 44, 2091–2100.
5. M. R. Dadfar and F. Ghadami, Mater. Des., 2013, 47, 16–20.
6. A. J. Kinloch, S. H. Lee and A. C. Taylor, Polymer, 2014, 55, 6325–6334.
7. J. M. Dean, P. M. Lipic, R. B. Grubbs, R. F. Cook and F. S. Bates, J. Polym. Sci., Part B: Polym. Phys., 2001, 39, 2996–3010.
8. P. Sun, Q. Dang, B. Li, T. Chen, Y. Wang, H. Lin, Q. Jin and D. Ding, Macromolecules, 2005, 38, 5654–5667.
9. N. Hameed, Q. Guo, Z. Xu, T. L. Hanley and Y. W. Mai, Soft Matter, 2010, 6, 6119–6129.
10. M. Larrañaga, E. Serrano, M. D. Martin, A. Tercjak, G. Kortaberria, K. de la Caba, C. C. Riccardi and I. Mondragon, Polym. Int., 2007, 56, 1392–1403.
11. C. Ocando, A. Tercjak, M. D. Martin, J. A. Ramos, M. Campo and I. Mondragon, Macromolecules, 2009, 42, 6215–6224.
12. S. Ritzenthaler, F. Court, E. Girard-Reydet, L. Leibler and J. P. Pascault, Macromolecules, 2003, 36, 118–126.
13. H. E. Romeo, I. A. Zucchi, M. Rico, C. E. Hoppe and R. J. J. Williams, Macromolecules, 2013, 46, 4854–4861.
14. E. Girard-Reydet, A. Sévignon, J. P. Pascault, C. E. Hoppe, M. J. Galante, P. A. Oyanguren and R. J. J. Williams, Macromol. Chem. Phys., 2002, 203, 947–952.
15. T. Fine, F. Lortie, L. David and J. P. Pascault, Polymer, 2005, 46, 6605–6613.
16. M. Blanco, M. López, G. Kortaberria and I. Mondragon, Polym. Int., 2010, 59, 523–528.
17. W. Fan and S. Zheng, Polymer, 2008, 49, 3157–3167.
18. I. A. Zucchi, M. J. Galante and R. J. J. Williams, Polymer, 2005, 46, 2603–2609.
19. L. Cano, D. H. Builes and A. Tercjak, Polymer, 2014, 55, 738–745.
20. S. Ritzenthaler, E. Girard-Reydet and J. P. Pascault, Polymer, 2000, 41, 6375–6386.
21. C. C. Riccardi, J. Borrajo, L. Meynie, F. Fenouillot and J. P. Pascault, J. Polym. Sci., Part B: Polym. Phys., 2004, 42, 1351–1360.
22. S. M. George, D. Puglia, J. M. Kenny, P. Jyotishkumar and S. Thomas, Polym. Eng. Sci., 2012, 52, 2336–2347.
23. D. W. van Krevelen, Properties of Polymers, Elsevier Science B.V., Amsterdam, The Netherlands, 4th edn, 1990.
24. C. Ocando, A. Tercjak and I. Mondragon, Compos. Sci. Technol., 2010, 70, 1106–1112.
25. M. I. Giannotti, C. R. Bernal, P. A. Oyanguren and M. J. Galante, Polym. Eng. Sci., 2005, 45, 1312–1318.
26. D. H. Builes, H. Hernandez, I. Mondragon and A. Tercjak, J. Phys. Chem. C, 2013, 117, 3563–3571.
27. E. Serrano, A. Tercjak, C. Ocando, M. Larrañaga, M. D. Parellada, S. Corona-Galván, D. Mecerreyes, N. E. Zafeiropoulos, M. Stamm and I. Mondragon, Macromol. Chem. Phys., 2007, 208, 2281–2292.

---