Search

Sorption-induced deformation is ubiquitous in nanoporous media, but underlying mechanisms are not yet fully understood, and a reliable modeling of this phenomenon is absent. Moreover hysteresis in sorption and swelling isotherms is observed but its origin not yet fully understood and not modeled. In this thesis the sorption-induced deformation of nanoporous media is studied systematically with different approaches. Three different nanoporous materials are considered: microporous polymers, microporous polymer-based composites and mesoporous materials. With the help of molecular simulations, the coupling mechanisms between sorption and deformation are revealed and the sorption and strain isotherms, as well as their hysteresis, are quantitatively modeled. With the knowledge gained at molecular level, a macroscopic description of sorption-induced deformation is given with the help of a dependent domain model. Molecular simulations demonstrate that microporous polymers swell upon water sorption as water molecules have a tendency to create more space between the flexible polymer chains for accommodating their presence. Sorption hysteresis is found to be related to deformation: polymers swell to form water–polymer hydrogen bonds upon adsorption but these bonds do not break upon desorption at the same chemical potential, which leads to sorption hysteresis. This hysteresis also manifests itself in other physical properties such as heat of sorption and bulk modulus. The influence of temperature and stress state on the coupled behavior is also examined. It is found that, when relating observable variables to the correct independent variables, hysteresis disappears as such explaining the actual origin of hysteresis. As a statement, hysteresis does not exist when looking at it from the correct driving potential. With the knowledge acquired on the bulk microporous polymer, the sorption-induced deformation of a microporous polymer-based composite, with cellulose nanocrystal (CN) as reinforcement and amorphous cellulose (AC) as matrix, is studied. Two competitive mechanisms are found regarding the coupling between sorption and deformation. The first mechanism is the reinforcing effect through CN-AC mechanical interaction, which constrains the sorption-induced swelling of the matrix and results in a reduction of sorption amount and of hysteresis in both sorption and deformation. The second mechanism is the CN-water interaction, enhancing water sorption in the matrix at the CN-matrix interface, increasing the sorption-induced swelling of the matrix and increasing the resulting hysteresis in sorption and deformation. Sorption-induced deformation in mesoporous materials is studied at single pore level with two atomistic models, a slit pore and a cylindrical pore. Two driving mechanisms are revealed for both slit and cylindrical pore models. At high relative vapor pressure, pore deformation is governed by Laplace pressure as the pore gets filled with liquid due to capillary condensation. At low pressure, when liquid films are formed on the pore surfaces and the pore remains mainly filled by vapor phase, the strain is driven by the attractive solid-fluid forces and the in-plane pressure within the film. Because of the interplay of these deformation mechanisms, the strain changes from shrinkage to expansion upon increase of pressure. The thesis ends with a Dependent Domain Model (DDM), developed to describe the coupled behavior at the macroscale of microporous polymers. The DDM is based on poromechanics taking into account the mechanical behavior of the solid and the influence of different pore sizes. The proposed dependent domain model captures the governing mechanism of the coupled behavior and provides a deeper understanding in sorption-induced deformation. Sorption-induced deformation is simulated and addressed systematically in this thesis regarding different materials, different coupling mechanisms and different scales. The simulation results agree with experiments and the proposed mechanism can explain the experimental results well. The outcome of this research provides a theoretical framework for modeling sorption-induced deformation of a great variety of nanoporous materials. Though the emphasis is laid at molecular scale, an upscaling approach is provided to connect the information at different scales.

Hydrocarbon polymers belong to the basic structures in polymer architecture. Poly(phenylene methylene) (PPM) represents one of the simplest structured polymer of this category. Before this PhD thesis, however, poly(phenylene methylene) was not available with molar masses sufficiently high to attract interest in materials science, in spite of numerous attempts to synthesize this polymer. Yet, based on the attractive properties of related polymers such as polyethylene, poly(para-phenylene), and pol(para-xylylene), it has been expected that poly(phenylene methylene) possesses interesting properties, such as high hydrophobicity, excellent thermal stability, and good barrier properties. This dissertation demonstrates the successful isolation of poly(phenylene methylene)s with broad range of molar masses by optimization of the catalytic polymerization of benzyl chloride with SnCl4, FeCl3, or organometallic tungsten(II) compounds, followed by fractionation. Low molar mass products were also obtained by quenching the reaction at moderate monomer conversions. Thus, products with number average molar masses (Mn) ranging from 200 g mol−1 – 167,900 g mol−1 were isolated. The glass transition temperature (Tg) of these polymers with different molar masses follows the Flory-Fox equation. The onset of decomposition temperature of higher molar mass products proceeds above 450 °C, according to TGA. Furthermore, the substitution pattern of PPM was discussed by study of chemical shifts of the methylene group by extensive NMR spectroscopy (1H, 13C, DEPT and HSQC) and by comparison with two mono-substituted derivatives of PPM – poly(2,4,6-trimethylphenylene methylene) and poly(2,3,5,6-tetramethylphenylene methylene). The analysis revealed that the signal group of the methylene unit in 13C NMR spectra are sensitive to the substitution pattern of the two adjacent phenylene rings, which is utilized to compare the substitution patterns of different PPM samples. Moreover, it was demonstrated that poly(phenylene methylene) exhibits pronounced blue fluorescence in solutions as well as in the solid state despite its non-p-conjugated nature. Optical spectroscopy was used to explore the characteristics and the physical origin of its unexpected optical properties, namely absorption in the 350 nm – 450 nm and photoluminescence in the 400 nm – 600 nm spectral regions. PPM possesses two discrete optically-active species, and a relatively long photoluminescence lifetime (>8 ns) in the solid-state. Phenomena like p-stacking and aggregation/crystallization, as well as impurities, are excluded as the origins of the optical properties. Instead there is sufficient evidence that PPM supports homoconjugation. Poly(2-methylphenylene methylene) and poly(2,4,6- trimethylphenylene methylene) – two derivatives of PPM – were synthesized and found to exhibit comparable spectroscopic properties, confirming the generality of the findings reported for PPM. Finally, the processability of fluorescent poly(phenylene methylene) has been investigated by employing various processing techniques. Thin fibers over one kilometer in length and short thick fibers (diameter of ~1 mm) were produced by melt spinning. The fibers exhibited birefringence and the ability to guide light waves (red light). Furthermore, a broad thickness range for films (few nm to few mm) was achieved with different approaches such as spin- coating, hot pressing, and die casting. The films featured crack-free and very smooth surfaces. Additionally, freestanding foams of the polymer were obtained by foaming highly concentrated solutions and near quasi-monodisperse microspheres were prepared by a microfluidic high-throughput emulsification. The materials properties of these morphologies were investigated and discussed for implementation as potential products from plastic optical fibers and light emitting diodes, to protective coatings and packaging, to insulators and separation membranes.

Encapsulation systems for the on-demand release of cargo molecules find applications in several fields, ranging from agricultural and construction to pharmaceuticals, food and materials science. In the construction sector, encapsulation systems have been used in the form of phase change materials for smart insulation of buildings and have recently been envisioned as a possible approach to deliberately control the setting reaction of cementitious materials. Controlling the setting of cement in building materials is crucial to prevent losses and high costs associated with early or late hardening of concretes and mortars. Current solutions are not satisfactory because accelerators and retarder molecules cannot be activated on demand in a homogeneous manner throughout the cementitious material. In this thesis, encapsulation systems for the on-demand release of molecules in cementitious materials are designed and investigated. On-demand release is triggered remotely using an external alternating magnetic field to locally heat the microcapsules. Heating leads to capsule bursting and local release of cargo molecules. Microcapsules with such thermally-triggered response were produced from double emulsions templates made in glass capillary microfluidic devices. The middle oil phase of the water-oil-water double emulsion contains a mixture of monomers, photoinitiator and inert liquids, which can be polymerized under UV light to generate monodisperse capsules with microstructured shells. The shell microstructure is determined by the phase separation of the polymer from the inert diluent during conversion of the double emulsion into microcapsules. The composition of the Summary ii oil phase directly impacts the size of polymerized particles formed within the shell, providing a useful control parameter to tune the mechanical properties of the microcapsules. Interestingly, the use of an amphiphilic inert liquid such as undecanol enables the formation of shells with open pores upon removal of the liquid. By contrast, microstructured shells with a polymer skin in the outer and inner walls are generated if the inert liquid displays no surface activity. This leads to the entrapment of the liquid inside the shell. If a volatile inert liquid is used, local heating of the microcapsule in an alternating magnetic field causes bursting and release of the cargo molecules encapsulated in the core. To ensure local heating of the capsules without an undesirable increase of the temperature of the surrounding cementitious paste, millimeter-sized metal spheres are used as inductive heat sources. In this approach, the metal spheres are coated with the thermo-sensitive microcapsules to generate a raspberry-like encapsulation system that can be activated using an external magnetic field. Experimental results obtained using such model raspberry-like system demonstrate that this is a feasible strategy to enable the on-demand release of an accelerator compound in an aqueous medium. The use of metallic fibers already used in concrete formulations as heat sources can potentially allow for the implementation of this concept in real applications.