Teijin (Netherlands)

Abstract The European Food Safety Authority (EFSA) initiated the “Roadmap for actions on Artificial Intelligence for evidence management in risk assessment” project in May 2021, aiming to optimize risk assessment processes by integrating Artificial Intelligence (AI) methodologies. A key focus is the automation of the Risk of Bias (RoB) assessment, a critical and time‐consuming step in systematic reviews that evaluates the methodological quality of studies. Implemented under Framework Contract OC/EFSA/AMU/2021/03 ‐ Specific Contract 3, this project built and piloted a prototype automation tool to support the risk of bias assessment process. This report details the steps taken to define current EFSA processes, select appropriate critical appraisal tools and use cases, and develop and validate a prototype automation tool based on a GPT model with zero‐shot learning (AutoCAT). The implementation phase includes integrating AutoCAT with existing systematic review platforms, and deployment within the EFSA environment. The assessment phase involved evaluating the tool with development and system test datasets, and user testing with EFSA scientific experts. The project's outcomes are discussed, highlighting the limitation of the pilot prototype and proposing future work and potential applications in other contexts.

Often researchers make use of Principal Component Analysis and Partial Least Squares Regression, an unsupervised and a supervised method, respectively, to extract the chemical information in the shape of one or more latent variables. However, when the research question is qualitative and requires a figure of merit, these two models will primarily focus on the quantitative and continuous information present in the data. In these cases, a valid approach may be to dichotomize the data and to analyze the resulting non-linear data via Non-Linear Principal Component Analysis. However, the results of the method are not always easy to interpret due to the possible multidimensionality of the solution. Here we introduce an alternative framework, composed of Rasch modeling and Generalized Linear Mixed Effect Models, to extract information from multivariate binary chemical data with an underlying design of experiment. The model obtained by this framework provides information in a unidimensional representation that can be easily translated into one-dimensional action and control. Furthermore, we show that, through Generalized Linear Mixed Effect Models, it is possible to extend the Rasch model to its multilevel form, which enables the consideration of each random factor possibly present.

Para-aramid is used as reinforcement fiber in many composite materials, including rubbers. Optimal fiber-matrix adhesion is essential in these materials. Since untreated aramid always shows poor adhesion, an adhesion activation process is required. This contribution describes and explains such processes, with focus on those applicable in industry. For rubber applications, adhesion can be achieved by resorcinol-formaldehyde latex (RFL) dipping of the fiber cord. This dipping has to be preceded by a treatment with an epoxy-based finish or epoxy-based pre-dip. Fundamental studies have improved the understanding of the maturation, application, and curing step of the epoxy treatment, including the resulting chemical and physical fiber surface structure. In addition, these studies have shown migration of curatives from rubber to RFL, during curing of the rubber. In this way, the complete aramid-rubber adhesion mechanism has been revealed. The epoxy treatment has also added value in applications of aramid fibers in thermoset matrices, such as epoxy, and in thermoplastic matrices, such as polyamide and polyurethane. This has been proven using several micro and macro-mechanical tests. Special attention is given to possible negative effects of processing oil on the adhesion performance. Some oil turns out to be necessary to reach the maximum adhesion and higher amounts do not directly lead to adhesion problems, both for rubber and for epoxy matrices. This is in line with long-standing industrial experience. Plasma treatment is potentially able to replace epoxy treatment of aramid fibers for rubber applications. However, considerable challenges are encountered in case the activation is carried out on bundles of filaments. On the other hand, plasma activation of cords seems to be feasible as an industrial alternative. There are also important applications where aramid is used as short-cut fiber. For rubber matrices, “dipped chopped fibers” are commonly used, resulting in a higher modulus of the reinforced material. This can be attributed to improved rubber adhesion and improved dispersion of the fibers throughout the rubber. For engineering plastics, the primary objective of adding short-cut fibers is to improve the abrasion resistance. This property does not seem to be affected by adhesion activation.

The crystal structure and phase behavior of bisamidegelators areinvestigated using differential scanning calorimetry (DSC), Fouriertransform infrared spectroscopy, X-ray diffraction (XRD), and molecularmodeling, aiming at a better understanding of bisamide gel systems.A homologous series of bisamide model compounds (nBAs) was prepared with the (CH2)n spacer between the two amide groups, where n varies from 5 to 10, and with two symmetric C17 alkyl tails. Withincreasing spacer length, the thermal properties show a clear odd–eveneffect, which was characterized using our newly developed analyticalmodel DSCN(T). Using XRD, all studied nBA compounds turn out to have a layer-like structure. The XRD patternsof the odd BA series are very similar but show marked differencescompared to the XRD patterns of the even series, which in turn arevery similar. The odd-membered 5BA molecules are nearly perpendicularto the stacked layers, as described by a pseudo-orthorhombic unitcell, whereas the even-membered 6BA molecules are tilted at an anglewith respect to the layer normal, as described by a triclinic unitcell. In both the odd and even series, the inter-layer interactionis the van der Waals interaction. The 6BA hydrogen bonding schemeis very similar to that of Nylon 6,10 α, unlike the 5BA H bondingscheme. The packing of the C17 alkyl tails in the 5BA layers is similarto polyethylene, and unlike 6BA. The slightly higher crystalline densityof 6BA (1.038 g cm–3) as compared to 5BA (1.018g cm–3) explains the higher melting point, higherenthalpy of fusion, and the observed shift of N–H stretch bandsto higher wave numbers. The structural differences observed betweenthe odd and even BA series reflect the different structure-directingeffect of parallel versus antiparallel amide hydrogen bonding motifs.These differences underlie the observed odd–even effect inthe thermal properties of nBA compounds.

SummaryThe activation of a bis‐phenoxyimine catalyst based on titanium (IV) using different aluminoxanes (MAO, PMAO and MMAO12) has been studied. The effect of a co‐catalyst modifier (BHT) used in combination with the MAO has been also tested. In particular, the effect of the activation time between the catalyst and the different aluminoxanes has been taken into consideration. On increasing the activation time between catalyst and the different aluminoxanes and TMA‐free MAO, differences in the catalyst activities have been observed. UHMWPEs having a reduced number of entanglements have been synthesized activating the FI catalyst with MAO and TMA‐free MAO. The obtained reactor powders can be solid‐state processed below the melting temperature in order to obtain high modulus/high tenacity tapes used for body armor and vehicle protection applications.

Aramid fibers are more and more used as reinforcing material in both tires and mechanical rubber goods, such as hoses, belts, etc. In these dynamic rubber applications, optimal adhesion between fiber and rubber is essential. This can be achieved by activating the aramid chemically with an epoxy formulation, followed by RFL (Resorcinol-Formaldehyde Latex) treatment. In the current study fundamental knowledge about the chemical adhesion activation process of aramid has been used to investigate a potential alternative for the epoxy treatment: plasma treatment. It is shown that plasma treatment introduces oxygen-containing functional groups at the surface of aramidyarn, but not to such an extent as epoxy treatment. Moreover, only a limited portion of the oxygencontaining groups can be attributed to hydroxyl groups, which are the key in the adhesion mechanism to, e.g., RFL/rubber. This explains the lower adhesion of plasma-treated yarn in comparison to epoxy-treated yarn. In addition, plasma treatment leads to surface roughening. However, this does not lead to higher adhesion,since the adhesion (Fibre–RFL–Rubber) is dominated by chemical interactions. So based on this experiment, plasma treatment cannot yet replace chemical treatment. Further investigation of ways to increase the number of reactive groups on the surface by plasma treatment is worthwhile since the chemical modification of the aramid surface turns out to be rather persistent in time, which is one of the prerequisites for future industrial application of plasma technology.