Molecular versus Silica-Supported Iridium Water Oxidation Catalysts

A stringent comparison between two pairs of molecular/immobilized water oxidation catalysts ([Cp^*Ir(Me-pica)Cl], 1 , versus 1_SiO₂ , Me-pica=κ²-N-methyl-picolinamide; [Cp^*Ir(pysa)NO₃], 2 , versus 2_SiO₂ , pysa=κ²-pyridine-2-sulfonamide]) reveals distinctive catalytic trends. While the molecular compound 1 exhibits a substantial higher activity than the analogous immobilized system 1_SiO₂ , under all the experimental conditions explored, the contrary is found with 2 that is far less active than its immobilized counterpart 2_SiO₂ . This is explained by the unique tendency of 2 to form dimeric complexes [Cp^*Ir-(κ²-μ₂-Hpysa)(κ²-μ₂-pysa)IrCp^*], 2 a , in phosphate buffered solution at pH 7, and [Cp^*Ir-(κ²-μ₂-Hpysa)₂IrCp^*], 2 b , in water. 2 a and 2 b have been completely characterized in solution by multinuclear and multidimensional NMR spectroscopy. They have been also isolated as single crystals and their structure in solid state determined by X-Ray diffractometry. 2 a and 2 b appear to be off-cycle species, whose formation is detrimental for water oxidation activity, as indicated by the observation of a long induction period when 2 a is used as catalytic precursor. In addition, TOF versus ΔE (E−E⁰=−RT/nF ln([IO₄⁻]/[IO₃⁻]) trends for the first two runs do not overlap for catalyst 2 and TOF_max is remarkably higher in the second run upon the addition of fresh NaIO₄. In the immobilized system 2_SiO₂ the detrimental associative processes are likely inhibited leading to an activity higher than that of 2 .     

Lost in a black-box? Interpretable machine learning for assessing Italian SMEs default

Academic research and the financial industry have recently shown great interest in Machine Learning algorithms capable of solving complex learning tasks, although in the field of firms' default prediction the lack of interpretability has prevented an extensive adoption of the black-box type of models. In order to overcome this drawback and maintain the high performances of black-boxes, this paper has chosen a model-agnostic approach. Accumulated Local Effects and Shapley values are used to shape the predictors' impact on the likelihood of default and rank them according to their contribution to the model outcome. Prediction is achieved by two Machine Learning algorithms (eXtreme Gradient Boosting and FeedForward Neural Networks) compared with three standard discriminant models. Results show that our analysis of the Italian Small and Medium Enterprises manufacturing industry benefits from the overall highest classification power by the eXtreme Gradient Boosting algorithm still maintaining a rich interpretation framework to support decisions.     

Protonated Porphyrins: Bifunctional Catalysts for the Metal-Free Synthesis of N-Alkyl-Oxazolidinones

The protonation of commercially available porphyrin ligands yields a class of bifunctional catalysts able to promote the synthesis of N-alkyl oxazolidinones by CO₂ cycloaddition to corresponding aziridines. The catalytic system does not require the presence of any Lewis base or additive, and shows interesting features both in terms of cost effectiveness and eco-compatibility. The metal-free methodology is active even with a low catalytic loading of 1 % mol, and the chemical stability of the protonated porphyrin allowed it to be recycled three times without any decrease in performance. In addition, a DFT study was performed in order to suggest how a simple protonated porphyrin can mediate CO₂ cycloaddition to aziridines to yield oxazolidinones.     

Engineered Green Fluorescence Proteins for Proteomics and Biomolecular Electronic Applications

The discovery of the photochromic characteristics of engineered green fluorescent proteins (GFPs) allows new proteomics and biomolecular electronic applications. In particular, photoreversibility among two distinct optical states can lead to the realization of a bio-optical high density storage memory. Here we review our recent work on an optically bistable GFP and we report the recent developments of self-assembly methods for spatial immobilization of proteins into well-definite 2D patterns.