This work aims to investigate the vapor sensing behavior of conductive polymer composites (CPCs). In connection with the protection of the environment and human beings, sensing of different kinds of chemical vapors is of increasing importance. At the moment, four kinds of vapor sensors are widely investigated and reported, namely semiconducting metal oxide sensors (MO), conjugated polymer sensors, carbonaceous nanomaterial based sensors, and CPC based sensors. Due to their unique component systems, the different sensor types are based on different sensing mechanisms resulting in different potential application ranges.
In consideration of cost and processability, CPC based vapor sensors are promising owning to their low cost, excellent processability, and designable compositions. In terms of vapor sensing behavior of CPC sensors, the interaction between the polymer and the organic vapor is a decisive factor in determining the sensing performance of CPCs. Ideally, the chosen polymer matrix should be able to swell without dissolving during vapor exposure so that the conductive network within the matrix can be disconnected, giving rise to the resistance change of CPCs. In some reported cases, polymers such as PLA and polycaprolactone (PCL) are degradable polymers, which are not durable when being exposed to environmental conditions for a long time. Therefore, it is necessary to make sure whether the selected polymers are resistive to vapors or not. There are two options for the polymer selection. One is to select a polymer that is only swellable in a specific or few organic solvents; another one is to select a polymer that is swellable to a variety of solvents. Since CPC sensors are used for detecting as many as possible hazardous chemicals to human beings or environment, the second case is more desired because of its broader window of detection. The solubility parameter is effective to characterize the interaction of polymers and organic solvents/vapors, which was firstly proposed by Charles Hansen. Initially, the Hansen solubility parameter (HSP) was used to predict the compatibility between polymer partners, chemical resistance, permeation rates, and even to characterize the surface of fillers. Liquids with similar solubility parameter (δ) are miscible, and polymers will dissolve in solvents whose δ is similar to their own value. This behavior is recognized as “like dissolves like”. Based on the description above, CPCs that can be used as liquid/vapor sensor materials should meet the following two requirements: 1) the chosen polymer should be swellable to vapors; 2) the CPCs as sensor materials have to be electrically conductive. Therefore, the relationship between conductive network and vapor sensing behavior of CPCs was investigated from the following aspects:
1) According to the previous studies, CB/polymer composites exhibit poor reversibility in cyclic vapor sensing tests because of the susceptible conductive network formed by CB particles. Thus, there is a need to improve the reversibility and increase the relative resistance change (Rrel) of CPCs. MWCNTs, as 1-dimensional carbon fillers with high aspect ratio, have excellent electrical and mechanical properties. Therefore, a hybrid filler system (MWCNT and CB) was utilized and incorporated in polycarbonate (PC) via melt compounding. PC was selected as the polymer matrix of CPCs because it showed high affinity with many commercial organic solvents/vapors as well as high and fast volume change upon organic solvents/vapors. In order to discuss the effect of conductive network formation on the vapor sensing behavior of PC/MWCNT/CB composites, two MWCNT contents were selected, which were lower and higher than the electrical percolation threshold of the PC/MWCNT composites. In the following, three CB contents were selected for the mixtures with MWCNT. The conductive networks composed of either MWCNT or hybrid CB/MWCNT are compared. The morphology of CPCs with different hybrid filler ratios was observed and investigated using SEM and OM. Moreover, to quantify the vapor sensing behavior of CPCs, some organic solvents were chosen and characterized by Flory-Huggins interaction parameter to demonstrate the polymer-vapor interaction. Afterwards, the cyclic vapor sensing was applied to illustrate the vapor sensing behavior of CPCs with different conductive network formations.
2) At moment, the filler dispersion is still a big challenge for MWCNT filled polymer composites due to the fact that the strong Van der Waals force among nanotubes makes them easily to entangle with each other resulting in the formation of agglomerates. A good filler dispersion state is desirable to achieve CPCs with low φc and. In order to reduce the φc of CPCs, immiscible polymer blend systems are introduced, which can have different blend microstructures by adjusting the polymer component ratios. In the second section, an immiscible polymer blend system based on two amorphous component, namely PC and polystyrene (PS), was chosen aiming to explain the influence of the blend morphology on the sensing performance of CPCs. PC/PS blends with different compositions filled with MWCNT were fabricated by melt mixing. The selective localization of MWCNTs in the blends was predicted using the Young’s equation. Moreover, the composite morphology, filler dispersion, and distribution were characterized by SEM and TEM. In the following, three kinds of CPCs ranging from sea-island structure to co-continuous structure were selected for the cyclic sensing measurement. The relationship between composite microstructure and resulting vapor sensing behavior was evaluated and discussed.
3) The poor reversibility of CPCs towards good solvent vapors is still a problem that hinders the cyclic use of CPC sensor materials. As an important class of polymer, crystalline polymers are rigid and less affected by solvent penetration because of the well-arranged polymer chains. Therefore, the effect of polymer crystallinity on the vapor sensing behavior of CPCs is imperative to be studied. In the third section, poly(lactic acid) (PLA), a semi-crystalline polymer, was selected to melt-mixed with PS and MWCNTs with the aim to improve the sensing reversibility of CPCs towards organic vapors, especially good solvent vapors. Thermal annealing was utilized to tune the PLA crystallinity and the polymer blend microstructure of CPCs. The electrical, morphological, and thermal behavior of CPCs after different thermal annealing times is discussed. In the following, the effect of crystallinity on the vapor sensing behavior of the CPCs was studied in detail. Besides, the different sensing performances of the CPCs towards different vapors resulted from the selective localization of MWCNTs and increased polymer matrix crystallinity were investigated and compared.
4) As discussed for the amorphous polymer blends and crystalline polymer blends and their vapor sensing behavior. The comparison of compact and porous structure of CPCs is going to be studied. In the fourth section, studies to further improve the sensing performance and to find out the exact sensing mechanism of CPCs were performed. Therefore, poly(vinylidene fluoride) (PVDF), a solvent resistive polymer, was chosen to be melt-mixed with PC and MWCNTs. In order to compare the MWCNT dispersion and localization in the blends, three kinds of PCs with different molecular weights were selected; hence, the viscosity ratio of immiscible blends was varied. Rheological, morphological, and electrical properties of CPCs were characterized. After that, the cyclic sensing and long-term immersion tests of CPCs towards different vapors were carried out to evaluate the vapor sensing behavior of compact CPCs with different blend viscosity ratios. Moreover, porous CPC sensors were prepared by extracting the PC component. The same sensing protocols were also applied to these porous sensor materials. The sensing mechanisms between compact CPC sensor and porous CPC sensor were compared and investigated.
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:38069 |
| Date | 30 January 2020 |
| Creators | Li, Yilong |
| Contributors | Voit, Brigitte, Pötschke, Petra, Piontek, Jürgen, Leibniz Institute of Polymer Research Dresden |
| Publisher | Technische Universität Dresden |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0029 seconds