The role of the automation development group in analytical research and development at Dupont Merck

Laboratory robotics has been firmly established in many non-QC laboratories as a valuable tool for automating pharmaceutical dosage form analysis. Often a single project or product line is used to justify an initial robot purchase thus introducing robotics to the laboratory for the first time. However, to gain widespread acceptance within the laboratory and to justify further investment in robotics, existing robots must be used to develop analyses for existing manual methods as well as new projects beyond the scope off the original purchase justification. The Automation Development Group in Analytical Research and Development is a team of analysts primarily devoted to developing new methods and adapting existing methods for the robot. This team approach developed the expertise and synergy necessary to significantly expand the contribution of robotics to automation in the authors'' laboratory.     

Isolation kernel: the X factor in efficient and effective large scale online kernel learning

Large scale online kernel learning aims to build an efficient and scalable kernel-based predictive model incrementally from a sequence of potentially infinite data points. Current state-of-the-art large scale online kernel learning focuses on improving efficiency. Two key approaches to gain efficiency through approximation are (1) limiting the number of support vectors, and (2) using an approximate feature map. They often employ a kernel with a feature map with intractable dimensionality. While these approaches can deal with large scale datasets efficiently, this outcome is achieved by compromising predictive accuracy because of the approximation. We offer an alternative approach that puts the kernel used at the heart of the approach. It focuses on creating a sparse and finite-dimensional feature map of a kernel called Isolation Kernel. Using this new approach, to achieve the above aim of large scale online kernel learning becomes extremely simple—simply use Isolation Kernel instead of a kernel having a feature map with intractable dimensionality. We show that, using Isolation Kernel, large scale online kernel learning can be achieved efficiently without sacrificing accuracy.

     

Conspiracy vs science: A large-scale analysis of online discussion cascades

World Wide Web - With the emergence and rapid proliferation of social media platforms and social networking sites, recent years have witnessed a surge of misinformation spreading in our daily life....     

Designing biomass supply chains within planetary boundaries

The design of sustainable supply chains, which recently emerged as an active area of research in process systems engineering, is vital to ensure sustainable development. Despite past and ongoing efforts, the available methods often overlook impacts beyond climate change or incorporate them via standard life cycle assessment metrics that are hard to interpret from an absolute sustainability viewpoint. We here address the design of biomass supply chains considering critical ecological limits of the Earth—planetary boundaries—which should never be surpassed by anthropogenic activities. Our method relies on a mixed-integer linear program that incorporates a planetary boundaries-based damage model to quantify absolute sustainability precisely. We apply this approach to the sugarcane-to-ethanol industry in Argentina, identifying the optimal combination of technologies and network layout that minimize the impact on these ecological boundaries. Our framework can find applications in a wide range of supply chain problems related to chemicals and fuels production, energy systems, and agriculture planning.     

Iron recovery from zinc mine tailings by magnetic separation followed by carbothermal reduction of self‐reducing briquettes

The volume of tailings produced by the extractive industry has been increasing due to the processing of the low‐grade ore. This issue can cause environmental accidents and require significant investment to control the disposal of tailings. Therefore, this study aims to recover iron from zinc mine tailings by wet magnetic separation followed by the carbothermal reduction of self‐reducing briquettes. Two magnetic separation routes were investigated to concentrate iron. Zinc mine tailings were processed by route I, in a rougher stage followed by a scavenger stage; and route II, in a rougher stage followed by a cleaner stage. The carbothermal reductions were performed using self‐reducing briquettes composed of Fe concentrate from the route with high Fe content and charcoal. The products were analyzed by scanning electron microscopy with energy dispersive spectroscopy (SEM‐EDS), x‐ray diffraction (XRD), inductively coupled plasma optical emission spectrometry (ICP‐OES), and volumetric chemical analysis. Magnetic separation route II provided the highest‐grade Fe concentrate, 52% Fe, while route I provided 33% Fe. In the carbothermal reductions, a metallization degree of 98% in the Fe concentrate briquette, 97% in the briquette with a 10% replacement of its raw material by Fe concentrate, and 99% in the hematite briquette was reached. The replacement of raw material by Fe concentrate showed no significant change in Fe recovery. Considering the whole process, magnetic separation and carbothermal reduction, the recovery of Fe from the zinc mine tailings was 67%. Therefore, the process route suggested in this study will not only reduce tailings disposal and consequently the risk of environmental accidents, but it will also provide profitable raw material for the steel industry.     

Fabrication of hybrid proton-exchange membranes using a brandnew high temperature ionic liquid as charge transporting and clay modifier

The search for more compatibility between ionic liquids (ILs) and polymer matrices in proton-exchange membrane fuel cells (PEMFCs) is one of the ways in which IL leaking from proton-exchange membranes could be minimized. In this work, it is presented the synthesis of an aromatic high temperature ionic liquid (HTIL), which, incorporated into an aromatic matrix such as sulfonated polyether ether ketone (sPEEK), is expected to diminish the IL leaking that normally affects PEMFC. Phenylethylammonium trifluoromethane sulfonate (PhetaTfO) was successfully synthesized and characterized. Its melting point of 88°C makes it to classify as a HTIL and it was employed as modifier of natural Montmorillonite, forming the phenylethylammonium intercalated montmorillonite (MmtPheta) and thus, ternary membranes containing PhetaTfO, MmtPheta, and sPEEK were prepared and characterized. Immersion tests demonstrated a higher compatibility of PhetaTfO with matrix when compared to the reference DemaTfO, which was reflected in up to 30% lower IL loss by the synthesized IL than the DemaTfO; X-rays diffraction (XRD) patterns demonstrated that the modified clay was properly dispersed inside the membranes, while dynamic mechanical analyses (DMA) results indicated a strong plasticizer effect along the increase of PhetaTfO content inside the membrane, while at the same time, the conductivity increased in an exponential manner, which permitted to identify an empiric exponential equation to evaluate the effect of concentration on ionic conductivity. The maximum conductivity obtained at IL concentrations of around 38 wt% was 0.2 mS/cm. It could expect high ionic conductivities of 10 mS/cm when the concentration of this IL is 60%; nevertheless, in order to achieve that, crosslinking treatments should be done to give the membranes enough mechanical resistance.     

Extracting Microfacet‐based BRDF Parameters from Arbitrary Materials with Power Iterations

We introduce a novel fitting procedure that takes as input an arbitrary material, possibly anisotropic, and automatically converts it to a microfacet BRDF. Our algorithm is based on the property that the distribution of microfacets may be retrieved by solving an eigenvector problem that is built solely from backscattering samples. We show that the eigenvector associated to the largest eigenvalue is always the only solution to this problem, and compute it using the power iteration method. This approach is straightforward to implement, much faster to compute, and considerably more robust than solutions based on nonlinear optimizations. In addition, we provide simple conversion procedures of our fits into both Beckmann and GGX roughness parameters, and discuss the advantages of microfacet slope space to make our fits editable. We apply our method to measured materials from two large databases that include anisotropic materials, and demonstrate the benefits of spatially varying roughness on texture mapped geometric models.