7.1 Aim of the work

This doctoral work aimed to explore the amine exchange of enaminones as a new dynamic bond for the synthesis of vitrimers. Vitrimers, only reported for the first time in 2011,1 show quite interesting features as they combine mechanical properties of thermosets with a glass-like malleability at elevated temperatures. This combination of properties was achieved through incorporation of thermally triggered exchangeable bonds that follow an associative exchange process, i.e. bonds are only broken after new bonds are formed (Figure 7.1). Because of this particular exchange mechanism, vitrimers are characterised by a constant crosslink density. This property also gives rise to a unique viscosity decrease upon heating, reminiscent of glass. As vitrimers combine highly desired properties that are not found back in other organic materials, they show great promise to have a considerable impact on industries that rely on cross-linked polymers. Figure 7.1: Associative exchange reactions enable network reorganisation without losing connectivity during exchange. The pioneering work relied on catalysed transesterification reactions in epoxy materials.1-5 Although transesterification was the first exchange chemistry used to demonstrate the concept of vitrimers, it provided already a high-level benchmark. Transesterification is a very simple and well-understood chemistry6 that can be easily implemented in materials with good properties, via the addition of a transesterification catalyst to known epoxy-acid and epoxy-anhydride formulations. Nevertheless, the system also showed limitations as high catalyst loadings are required, transesterification is intrinsically slow and the catalystsolubility impedes high-Tg materials. In this thesis, the amine exchange of vinylogous acyls, also referred to as enaminones, was examined as alternative exchangeable bond that would enhance vitrimer’ processability and resolve the limitations in glass-transition temperature. The amine exchange on these enaminone moieties was already reported in various reports for the synthesis of small molecules.7-10 Also in polymer science, enaminones and more precisely vinylogous urethanes (Vurethanes) were used in the field of water-borne coatings11 and adhesives12 due to their straightforward synthesis from amines and acetoacetates. Nevertheless, not one report in literature designed these enaminone containing networks to exploit its dynamic nature in bulk materials. Thus, in this PhD research, the different enaminone moieties were systematically researched for their potential in vitrimer materials (Figure 7.2). Figure 7.2: Amine exchange of enaminones was explored for new vitrimer materials. Simultaneous with this research, that spanned from 2012 to 2016, other academic contributions were made as well to this quickly emerging field of vitrimer materials. An overview is presented in Chapter 2 and also published as a minireview.

7.2 Overview of the results

In the first part of this research, presented in chapter 3, a systematic study on low-molecular-weight compounds was conducted to assess the kinetics of the amine exchange on enaminones. This study showed that the vinylogous transamination of enaminones spans over a very wide practical temperature window, ranging from room temperature for vinylogous urea (Vurea) to temperatures as high as 170°C for substituted vinylogous amides (Vamides), the relative order of enaminones is depicted in Figure 7.3. Furthermore, two possible exchange mechanisms were provided and it was rationalised that proton exchanges are crucial. Consequently, it was also demonstrated that the addition of a protic acid increased the exchange rate. Based on these results, Vurethanes and Vurea were selected as most promising enaminones for vitrimers as they exhibit the fastest exchange kinetics and a straightforward monomer synthesis. Figure 7.3: Relative order of the amine exchange of enaminones, with an indicative temperature when the exchange reaction starts to become reasonable fast. In Chapter 4, different approaches were examined to prepare Vurethane vitrimers. The first strategy relied on amine-functional prepolymers, which would be crosslinked with bis-acetoacetates. Yet, this strategy was abandoned as side-reactions complicated the preparation of amine-functional prepolymers. The second approach, starting from low-MW amine and acetoacetate monomers, demonstrated the first proof-of-concept vitrimers based on vinylogous transamination (Figure 7.4). Poly(vinylogous urethane) networks with a glass transition temperature of 87°C and a storage modulus of ~2.4 GPa were obtained. As expected for a polymer network, the samples were insoluble even at elevated temperature and a rubbery plateau was observed by DMA. Stress-relaxation and creep experiments showed a viscoelastic liquid behaviour. Due to the fast exchange reactions and high density of exchangeable bonds throughout the network, relaxation times as short as 85 s at 170°C were achieved without the use of any catalyst. Moreover, the poly(vinylogous urethane) networks can be recycled without loss of mechanical properties. These properties position these materials among the top-performing vitrimers currently known. Furthermore, a monomer screening showed that materials with a glass-transition up to 145°C and good mechanical properties are feasible. In our quest to improve the processability even more, the influence of catalyst on the amine exchange of Vurethanes was examined both in a low-MW model study as an elastomeric vitrimer (Chapter 5). Although working catalyst-free was an initial target, it was realised that addition of small amounts of additives provides a way to control the exchange kinetics. Indeed, this study showed that the amine exchange Vurethanes can easily be controlled and a good correlation between model reactions and mechanical relaxation times was observed. Addition of acids increased the exchange rate and allowed faster processing while a strong base such as TBD strongly decelerate exchange reactions, which could be interesting when a good creep resistance is aimed for. Moreover, using acid catalyst, relaxation times of the hard Vurethane vitrimers presented in chapter 4 could be reduced to 10 s, which significantly improves the processability of these materials. In the last part of this thesis (Chapter 6), Vinylogous urea were exploited for vitrimer materials as they showed the fastest exchange kinetics of all enaminones. Although an initial study hinted to a limited thermal stability, varying slightly the Vurea structure resulted in materials with a satisfactory thermostability. Networks were prepared by combination of 1,4-(piperazine)bis(acetoacetamide), 1,6-hexane diamine and tris(2-amineoethyl)amine. Simultaneously, also the same network catalysed with acid was investigated to find the boundaries for processability. Vitrimers with a glass transition of 110°C and a storage modulus of 2.2 GPa at room temperature were obtained. Although no significant differences were observed in material properties of the catalysed and uncatalysed samples, frequency sweep experiments indicated strongly reduced relaxation times for the acid catalysed networks. While the uncatalysed samples showed relaxation times in the order of minutes (130 s to 60 s from 140 to 170°C), the acid catalysed samples showed relaxation times in the order of seconds (6 to 3 s from 140 to 170°C). To the best of our knowledge, these catalysed Vurea vitrimers exhibit the shortest relaxation times reported up-to-date. Finally, Vurea vitrimers were explored as matrix for composites. After some adaptations of the matrix to enhance the glass-transition and allow for easy fiber impregnation, one-layer glass-fiber sheets were prepared that could serve as an alternative for prepregs. While normal prepregs require special storage conditions and anti-adhesive layers due to its partial cured matrix, the vitrimer prepregs do not require special storage and anti-adhesive layers and remain processable because of its vitrimer nature. Indeed, multilayer composites could be prepared through compression of stacked prepregs. The mechanical properties of the obtained materials were comparable to those of epoxy networks prepared via reaction injection moulding. Finally, a proof-of-concept of composite processing was presented through thermoforming a two-layer vitrimer composite (Figure 7.5).

7.3 Perspectives

In this project, an underexplored exchange reaction in polymer chemistry was researched for its potential in the application for vitrimers, a brand new class of materials that were only reported for the first time in 2011. Consequently, many challenges and opportunities remain. As mentioned before, one of the most promising applications of vitrimers would be composites. At the end of chapter 6, a proof-of-concept for Vurea composites was reported together with initial characterisation. While these initial efforts showed already potential, a more extensive assessment of the properties and the assets of its vitrimer matrix will be required to launch industrial projects. Moreover, academic research including this work often aims to demonstrate virtues of its research, which is essential to publish, intrigue industrial researchers and initiate industrial projects. Yet, knowing limitations is at least as important. Therefore, additional research on its limitations such as aging, UV/water-resistance and amine leaching will be indispensable. Besides composites, we presented an approach relying on the cross-linking of amine-functionalised prepolymers. Since this thesis focused mainly on materials with a high Tg, PDMS-based materials were not evaluated. Nevertheless, aminefunctionalised PDMS is commercially available and could be readily used to prepare enaminone-based vitrimers. In addition, as demonstrated in chapter 3 on model compounds, the useful temperature range could be shifted by choosing the different enaminone moieties. Also the use of a photobase to inhibit transamination and unwanted creep after processing, would be feasible on such matrices. Finally, the use of fillers, which is a common practice to enhance mechanical properties in elastomers, is an interesting factor to consider in vitrimeric elastomers as already demonstrated by Leibler et al.1 These various possibilities on PDMS-based enaminone vitrimers will be a part of the PhD-thesis of Yann Spieschaert. Finally, other parts of this doctoral thesis could be subject of further improvements. For example, other ways of vitrimer processing could be interesting as this work mainly made use of compression moulding. The initial extrusion experiments of Vurea showed already some potential but require considerable efforts to avoid material degradation. Next, the used cross-link densities and excess of amines - obligatory to enable amine exchange - were often chosen based on an educated guess. Probably, there is a firm margin of improvement possible, yet these optimisations would become only very interesting when real applications are considered in order to have a target.