Design of Modified Polymer Membranes Using Machine Learning.

Surface modification is an attractive strategy to adjust the properties of polymer membranes. Unfortunately, predictive structure-processing-property relationships between the modification strategies and membrane performance are often unknown. One possibility to tackle this challenge is the application of data-driven methods such as machine learning. In this study, we applied machine learning methods to data sets containing the performance parameters of modified membranes. The resulting machine learning models were used to predict performance parameters, such as the pure water permeability and the zeta potential of membranes modified with new substances. The predictions had low prediction errors, which allowed us to generalize them to similar membrane modifications and processing conditions. Additionally, machine learning methods were able to identify the impact of substance properties and process parameters on the resulting membrane properties. Our results demonstrate that small data sets, as they are common in materials science, can be used as training data for predictive machine learning models. Therefore, machine learning shows great potential as a tool to expedite the development of high-performance membranes while reducing the time and costs associated with the development process at the same time.

Immobilization of Bi2WO6 on Polymer Membranes for Photocatalytic Removal of Micropollutants from Water - A Stable and Visible Light Active Alternative.

In this work, bismuth tungstate Bi2WO6 is immobilized on polymer membranes to photocatalytically remove micropollutants from water as an alternative to titanium dioxide TiO2. A synthesis method for Bi2WO6 preparation and its immobilization on a polymer membrane is developed. Bi2WO6 is characterized using X-ray diffraction and UV-vis reflectance spectroscopy, while the membrane undergoes analysis through scanning electron microscopy, X-ray photoelectron spectroscopy, and degradation experiments. The density of states calculations for TiO2 and Bi2WO6, along with PVDF reactions with potential reactive species, are investigated by density functional theory. The generation of hydroxyl radicals OH• is investigated via the reaction of coumarin to umbelliferone via fluorescence probe detection and electron paramagnetic resonance. Increasing reactant concentration enhances Bi2WO6 crystallinity. Under UV light at pH 7 and 11, the Bi2WO6 membrane completely degrades propranolol in 3 and 1 h, respectively, remaining stable and reusable for over 10 cycles (30 h). Active under visible light with a bandgap of 2.91 eV, the Bi2WO6 membrane demonstrates superior stability compared to a TiO2 membrane during a 7-day exposure to UV light as Bi2WO6 does not generate OH• radicals. The Bi2WO6 membrane is an alternative for water pollutant degradation due to its visible light activity and long-term stability.

Highly Efficient One-Step Protein Immobilization on Polymer Membranes Supported by Response Surface Methodology.

Immobilization of proteins by covalent coupling to polymeric materials offers numerous excellent advantages for various applications, however, it is usually limited by coupling strategies, which are often too expensive or complex. In this study, an electron-beam-based process for covalent coupling of the model protein bovine serum albumin (BSA) onto polyvinylidene fluoride (PVDF) flat sheet membranes was investigated. Immobilization can be performed in a clean, fast, and continuous mode of operation without any additional chemicals involved. Using the Design of Experiments (DoE) approach, nine process factors were investigated for their influence on graft yield and homogeneity. The parameters could be reduced to only four highly significant factors: BSA concentration, impregnation method, impregnation time, and electron beam irradiation dose. Subsequently, optimization of the process was performed using the Response Surface Methodology (RSM). A one-step method was developed, resulting in a high BSA grafting yield of 955 mg m-2 and a relative standard deviation of 3.6%. High efficiency was demonstrated by reusing the impregnation solution five times consecutively without reducing the final BSA grafting yield. Comprehensive characterization was conducted by X-ray photoelectron spectroscopy (XPS), Fourier-transform infrared spectroscopy (FTIR), and measurements of zeta potential, contact angle and surface free energy, as well as filtration performance. In addition, mechanical properties and morphology were examined using mercury porosimetry, tensile testing, and scanning electron microscopy (SEM).

Enhanced Removal and Toxicity Declineof Diclofenac by Combining UVA Treatmentand Adsorption of Photoproductsto Polyvinylidene Difluoride.

The occurrence of micropollutants in the environment is an emerging issue. Diclofenac, a non-steroidal anti-inflammatory drug, is one of the most frequently detected pharmaceuticals in the environment worldwide. Diclofenac is transformed by UVA light into different products with higher toxicity. The absorbance of the transformation products overlaps with the absorbance of diclofenac itself and inhibits the ongoing photoreaction. By adding polyvinylidene difluoride (PVDF), the products adsorb to the surface of PVDF. Therefore, phototransformation of diclofenac and total organic carbon (TOC) removal is enhanced and the toxicity decreased. At 15 min and 18 h of UVA treatment, removal of diclofenac and TOC increases from 56% to 65% and 18% to 54%, respectively, when PVDF is present. The toxicity of a UVA treated (18 h) diclofenac solution doubles (from 5 to 10, expressed in toxicity units, TU), while no toxicity was detectable when PVDF is present during UVA treatment (TU = 0). PVDF does not need to be irradiated itself but must be present during photoreaction. The adsorbent can be reused by washing with water or ethanol. Diclofenac (25 mg L-1) UVA light irradiation was monitored with high performance liquid chromatography (HPLC), UV-Vis spectroscopy and by analysing the decrease of TOC. The toxicity towards Vibrio fischeri was examined according to DIN EN ISO 11348-1: 2009-05. Density functional theory (DFT) was used to simulate the phototransformation products known in literature as well as further products identified via gas chromatography-mass spectrometry (GC-MS). The absorption spectra, reaction enthalpies (ΔH) and Gibbs free energy of reactions (ΔG) were calculated. The combination of UVA irradiation of diclofenac with adsorption of photoproducts to PVDF is unique and opens up new possibilities to enhance removal of pollutants from water.