Designing optical glasses by machine learning coupled with a genetic algorithm

Engineering new glass compositions have experienced a sturdy tendency to move forward from (educated) trial-and-error to data- and simulation-driven strategies. In this work, we developed a computer program that combines data-driven predictive models (in this case, neural networks) with a genetic algorithm to design glass compositions with desired combinations of properties. First, we induced predictive models for the glass transition temperature (T_g) using a dataset of 45,302 compositions with 39 different chemical elements, and for the refractive index (n_d) using a dataset of 41,225 compositions with 38 different chemical elements. Then, we searched for relevant glass compositions using a genetic algorithm informed by a design trend of glasses having high n_d (1.7 or more) and low T_g (500 °C or less). Two candidate compositions suggested by the combined algorithms were selected and produced in the laboratory. These compositions are significantly different from those in the datasets used to induce the predictive models, showing that the used method is indeed capable of exploration. Both glasses met the constraints of the work, which supports the proposed framework. Therefore, this new tool can be immediately used for accelerating the design of new glasses. These results are a stepping stone in the pathway of machine learning-guided design of novel glasses.     

Hydrogen production through dry reforming of biogas using a porous electrochemical cell: Effects of a cobalt catalyst in the electrode and mixing of air with biogas

This paper reports the performance of porous Gd-doped ceria (GDC) electrochemical cells with Co metal in both electrodes (cell No. 1) and with Ni metal in the cathode and Co metal in the anode (cell No. 2) for CO₂ decomposition, CH₄ decomposition, and the dry reforming reaction of a biogas with CO₂ gas (CH₄ + CO₂ → 2H₂ + 2CO) or with O₂ gas in air (3CH₄ +?1.875CO₂ +?1.314O₂ → 6H₂ +?4.875CO +?0.7515O₂). GDC cell No. 1 produced H₂ gas at formation rates of 0.055 and 0.33?mL-H₂/(min?m²-electrode) per 1?mL-supplied gas/(min?m²-electrode) at 600?°C and 800?°C, respectively, by the reforming of the biogas with CO₂ gas. Similarly, cell No. 2 produced H₂ gas at formation rates of 0.40?mL-H₂/(min?m²) per 1?mL-supplied gas/(min?m²) at 800?°C from a mixture of biogas and CO₂ gas. The dry reforming of a real biogas with CO₂ or O₂ gas at 800?°C proceeded thermodynamically over the Co or Ni metal catalyst in the cathode of the porous GDC cell. Faraday's law controlled the dry reforming rate of the biogas at 600?°C in cell No. 2. This paper also clarifies the influence of carbon deposition, which originates from CH₄ pyrolysis (CH₄ → C + 2H₂) and disproportionation of CO gas (2CO → C + CO₂), on the cell performance during dry reforming. The dry reforming of a biogas with O₂ molecules from air exhibits high durability because of the oxidation of the deposited carbon by supplied air.     

The influence of oxygen partial pressure on growth of the (Hg,Re)-1223 intergrain junction

Hg/sub 0.82/Re/sub 0.18/Ba/sub 2/Ca/sub 2/Cu/sub 3/O/sub 8+/spl delta// polycrystalline samples were successfully obtained by using different oxygen partial pressure in the annealing treatment of the precursor ceramic. The doping state was confirmed by X-ray powder diffraction pattern analysis and by observing distinct thermopower values at room temperature. Also, the intergrain regions have shown an improvement in the critical current density when using the precursor preparation with 10% O/sub 2/ and 90% Ar (optimal doped). The optimal doped sample has presented the highest /spl alpha/ exponent of the J/sub c//spl prop/[1-(T/T/sub c/)/sup 2/]/sup /spl alpha// dependence. For the case of (Hg,Re)-1223 polycrystalline superconductor applications, the /spl alpha/ exponent can be used as a junction quality parameter.     

Binary Adsorption Isotherms of Water and Ethyl Acetate Vapor for Materials Contained in a Box of a Tobacco Product

An experimental study has been carried out on the characteristics of binary adsorption isotherms of water and volatile flavor for typical materials used in a box of a tobacco product or cigarettes. Ethyl acetate chosen as a model for water-soluble volatile flavor. Binary adsorption isotherms for the tobacco, papers, filters, and activated carbons were measured with a flow-type multi-component adsorption system under the canstant conditions of temperature a t 303 K and vapor pressure of water a t 2.5 kPain the vapor pressure range of ethyl acetate from 0 to 4.2 kPa. A linear equation was applied to express the binary adsorption isotherms for the tobacco, papers and filters, while a Dubinin-Astakhov equation was applied for the activated carbons. The binary adsorption was characterized into three groups, depending on the selectivity as well as the mechanism of adsorption; i.e.(l) for tobacco and papers, water was adsorbed much greater rather than ethyl acetate, (2) for filters, ethyl acetate was adsorbed on the surface as great as water, (3) for activated carbons, ethyl acetate was adsorbed much greater onto their micropores rather than water. The results showed that ethyl     

A 160-kDa Protein Is Essential for Hemocyanin-derived Melanosis of Prawn

ABSTRACT A purified 160-kDa protein, isolated from the cuticle of kurama prawn and named melanosis collaborating factor (MCF), was a key factor in promoting melanosis by a cooperative reaction with hemocyanin-derived phenoloxidase but hemocyanin itself was incapable of producing black pigment. The enzymatic activity of MCF in the body of the prawn was very stable against the process of freezing and thawing; the enzyme maintained more than 80% of its activity after 3 mo of storage at −2 5 °C. These results indicated that MCF, acting in conjunction with hemocyanin-derived phenoloxidase, played a crucial role in postharvest melanogenesis in prawn.