Bayesian optimization (BO) is a widely-used method for optimizing expensive (to evaluate) problems. At the core of most BO methods is the modeling of the objective function using a Gaussian Process (GP) whose covariance is selected from a set of standard covariance functions. From a weight-space view, this models the objective as a linear function in a feature space implied by the given covariance K, with an arbitrary Gaussian weight prior ${\bf w} \sim \mathcal{N} ({\bf 0}, {\bf I})$. In many practical applications there is data available that has a similar (covariance) structure to the objective, but which, having different form, cannot be used directly in standard transfer learning. In this paper we show how such auxiliary data may be used to construct a GP covariance corresponding to a more appropriate weight prior for the objective function. Building on this, we show that we may accelerate BO by modeling the objective function using this (learned) weight prior, which we demonstrate on both test functions and a practical application to short-polymer fibre manufacture.

Real world experiments are expensive, and thus it is important to reach a target in a minimum number of experiments. Experimental processes often involve control variables that change over time. Such problems can be formulated as functional optimisation problem. We develop a novel Bayesian optimisation framework for such functional optimisation of expensive black-box processes. We represent the control function using Bernstein polynomial basis and optimise in the coefficient space. We derive the theory and practice required to dynamically adjust the order of the polynomial degree, and show how prior information about shape can be integrated. We demonstrate the effectiveness of our approach for short polymer fibre design and optimising learning rate schedules for deep networks.

The scale-up of laboratory procedures to industrial production is the main challenge standing between ideation and the successful introduction of novel materials into commercial products. Retaining quality while ensuring high per-batch production yields is the main challenge. Batch processing and other dynamic strategies that preserve product quality can be applied, but they typically involve a variety of experimental parameters and functions that are difficult to optimize because of interdependencies that are often antagonistic. Adaptive Bayesian optimization is demonstrated here as a valuable support tool in increasing both the per-batch yield and quality of short polymer fibers, produced by wet spinning and shear dispersion methods. Through this approach, it is shown that short fiber dispersions with high yield and a specified, targeted fiber length distribution can be obtained with minimal cost of optimization, starting from sub-optimal processing conditions and minimal prior knowledge. The Bayesian function optimization demonstrated here for batch processing could be applied to other dynamic scale-up methods as well as to cases presenting higher dimensional challenges such as shape and structure optimization. This work shows the great potential of synergies between industrial processing, material engineering, and machine learning perspectives.

The paper presents a novel approach to direct covariance function learning for Bayesian optimisation, with particular emphasis on experimental design problems where an existing corpus of condensed knowledge is present. The method presented borrows techniques from reproducing kernel Banach space theory (specifically m-kernels) and leverages them to convert (or re-weight) existing covariance functions into new, problem-specific covariance functions. The key advantage of this approach is that rather than relying on the user to manually select (with some hyperparameter tuning and experimentation) an appropriate covariance function it constructs the covariance function to specifically match the problem at hand. The technique is demonstrated on two real-world problems - specifically alloy design and short-polymer fibre manufacturing - as well as a selected test function.

This paper presents a novel approach to kernel tuning. The method presented borrows techniques from reproducing kernel Banach space (RKBS) theory and tensor kernels and leverages them to convert (re-weight in feature space) existing kernel functions into new, problem-specific kernels using auxiliary data. The proposed method is applied to accelerating Bayesian optimisation via covariance (kernel) function pre-tuning for short-polymer fibre manufacture and alloy design.

Experimental design is a process of obtaining a product with target property via experimentation. Bayesian optimization offers a sample-efficient tool for experimental design when experiments are expensive. Often, expert experimenters have 'hunches' about the behavior of the experimental system, offering potentials to further improve the efficiency. In this paper, we consider per-variable monotonic trend in the underlying property that results in a unimodal trend in those variables for a target value optimization. For example, sweetness of a candy is monotonic to the sugar content. However, to obtain a target sweetness, the utility of the sugar content becomes a unimodal function, which peaks at the value giving the target sweetness and falls off both ways. In this paper, we propose a novel method to solve such problems that achieves two main objectives: a) the monotonicity information is used to the fullest extent possible, whilst ensuring that b) the convergence guarantee remains intact. This is achieved by a two-stage Gaussian process modeling, where the first stage uses the monotonicity trend to model the underlying property, and the second stage uses `virtual' samples, sampled from the first, to model the target value optimization function. The process is made theoretically consistent by adding appropriate adjustment factor in the posterior computation, necessitated because of using the `virtual' samples. The proposed method is evaluated through both simulations and real world experimental design problems of a) new short polymer fiber with the target length, and b) designing of a new three dimensional porous scaffolding with a target porosity. In all scenarios our method demonstrates faster convergence than the basic Bayesian optimization approach not using such `hunches'.

This study outlines a novel thermal extrusion system and methodologies for evaluating the potential to manufacture carbon fiber directly from thermoplastic coals. It is envisioned that the intermediate product will be further refined by spinning down to commercial fiber sizes and thermal annealing. Commercial melt spinning is used for manufacturing carbon fibers from pitch-based feed materials, and a similar approach for plasticized coal is likely to be a lower risk option. However, the critical aspect of using coal for this purpose is its behavior inside a higher pressure extrusion unit and the need to characterize its rheology. This work has evaluated the thermoplastic development needed for extrusion of a single coking coal in terms of the heating rate and residence time and characterized the extruded fiber product. It was observed that the coal underwent a preliminary softening phase prior to extruding at significant speed. This phase appeared necessary to develop the critical viscosity for extrusion and was affected by the heating rate. The size of the orifice that the coal was extruded through also impacted the point of extrusion, with the smaller 0.5 mm hole requiring lower viscosity to be developed to flow at steady state. Other operating modes were developed to examine the thermoplastic properties of the coal over an extended residence time, and it was found that the coal could be maintained up to 60 min at selected temperatures. The product fiber was larger than the commercial size, appearing slightly larger than the orifice size. Internal porosity and surface roughness were observed as coal-based fiber qualities in need of controlling, along with the mineral content and size.

The discovery of processes for the synthesis of new materials involves many decisions about process design, operation, and material properties. Experimentation is crucial but as complexity increases, exploration of variables can become impractical using traditional combinatorial approaches. We describe an iterative method which uses machine learning to optimise process development, incorporating multiple qualitative and quantitative objectives. We demonstrate the method with a novel fluid processing platform for synthesis of short polymer fibers, and show how the synthesis process can be efficiently directed to achieve material and process objectives.

Prevailing batch Bayesian optimisation methods allow all control variables to be freely altered at each iteration. Real-world experiments, however, often have physical limitations making it time-consuming to alter all settings for each recommendation in a batch. This gives rise to a unique problem in BO: in a recommended batch, a set of variables that are expensive to experimentally change need to be fixed, while the remaining control variables can be varied. We formulate this as a process-constrained batch Bayesian optimisation problem. We propose two algorithms, pc-BO(basic) and pc-BO(nested). pc-BO(basic) is simpler but lacks convergence guarantee. In contrast pc-BO(nested) is slightly more complex, but admits convergence analysis. We show that the regret of pc-BO(nested) is sublinear. We demonstrate the performance of both pc-BO(basic) and pc-BO(nested) by optimising benchmark test functions, tuning hyper-parameters of the SVM classifier, optimising the heat-treatment process for an Al-Sc alloy to achieve target hardness, and optimising the short polymer fibre production process.

Questions about how to regulate nanoenhanced products regularly arise as researchers determine possible nanoparticle transformation(s). Focusing concern on the incorporation and subsequent release of nano-Ag in fabrics often overshadows the fact that many "conventional silver" antimicrobials such as ionic silver, AgCl, metallic Ag, and other forms will also form different species of silver. In this study we used a laboratory washing machine to simulate the household laundering of a number of textiles prepared with known conventional Ag or nano-Ag treatments and a commercially available fabric incorporating yarns coated with bulk metallic Ag. Serial filtration allowed for quantification of total Ag released in various size fractions (>0.45 μm, < 0.45 μm, <0.1 μm, and <10 kDa), while characterization of particles with TEM/EDX provided insight on Ag transformation mechanisms. Most conventional Ag additives yielded more total Ag and more nanoparticulate-sized Ag in the washing liquid than fabrics that used nano-Ag treatments. Incorporating nano-silver into the fiber (as opposed to surface treatments) yielded less total Ag during fabric washing. A variety of metallic Ag, AgCl, and Ag/S particles were observed in washing solution by TEM/EDX to various extents depending on the initial Ag speciation in the fabrics. Very similar particles were also observed when dissolved ionic Ag was added directly into the washing liquid. On the basis of the present study, we can state that all silver-treated textiles, regardless of whether the treatment is "conventional" or "nano", can be a source of silver nanoparticles in washing solution when laundering fabrics. Indeed, in this study we observed that textiles treated with "conventional" silver have equal or greater propensity to form nano-silver particles during washing conditions than those treated with "nano"-silver. This fact needs to be strongly considered when addressing the risks of nano-silver and emphasizes that regulatory assessment of nano-silver warrants a similar approach to conventional silver.

Many antimicrobial technologies are available for textiles. They may be used in many different textile applications to prevent the growth of microorganisms. Due to the biological activity of the antimicrobial compounds, the assessment of the safety of these substances is an ongoing subject of research and regulatory scrutiny. This review aims to give an overview on the main compounds used today for antimicrobial textile functionalization. Based on an evaluation of scientific publications, market data as well as regulatory documents, the potential effects of antimicrobials on the environment and on human health were considered and also life cycle perspectives were taken into account. The characteristics of each compound were summarized according to technical, environmental and human health criteria. Triclosan, silane quaternary ammonium compounds, zinc pyrithione and silver-based compounds are the main antimicrobials used in textiles. The synthetic organic compounds dominate the antimicrobials market on a weight basis. On the technical side the application rates of the antimicrobials used to functionalize a textile product are an important parameter with treatments requiring lower dosage rates offering clear benefits in terms of less active substance required to achieve the functionality. The durability of the antimicrobial treatment has a strong influence on the potential for release and subsequent environmental effects. In terms of environmental criteria, all compounds were rated similarly in effective removal in wastewater treatment processes. The extent of published information about environmental behavior for each compound varies, limiting the possibility for an in-depth comparison of all textile-relevant parameters across the antimicrobials. Nevertheless the comparative evaluation showed that each antimicrobial technology has specific risks and benefits that should be taken into account in evaluating the suitability of different antimicrobial products. The results also indicated that nanoscale silver and silver salts that achieve functionality with very low application rates offer clear potential benefits for textile use. The regular care of textiles consumes lots of resources (e.g. water, energy, chemicals) and antimicrobial treatments can play a role in reducing the frequency and/or intensity of laundering which can give potential for significant resource savings and associated impact on the environment.

ADVERTISEMENT RETURN TO ISSUEPREVCorrespondence/Rebut...Correspondence/RebuttalNEXTReply to Comments on "120 Years of Nanosilver History: Implications for Policy Makers"Bernd Nowack†*, Harald F. Krug†, and Murray Height‡View Author Information† Empa, Swiss Federal Laboratories for Materials Science and Technology, Lerchenfeldstrasse 5, CH-9014 St. Gallen, Switzerland‡ HeiQ Materials AG, CH-5330 Bad Zurzach, Switzerland phone +41 (0)58 765 76 92; e-mail: [email protected]Cite this: Environ. Sci. Technol. 2011, 45, 17, 7593–7595Publication Date (Web):August 5, 2011Publication History Published online5 August 2011Published inissue 1 September 2011https://pubs.acs.org/doi/10.1021/es2017895https://doi.org/10.1021/es2017895article-commentaryACS PublicationsCopyright © 2011 American Chemical Society. This publication is available under these Terms of Use. Request reuse permissions This publication is free to access through this site. Learn MoreArticle Views899Altmetric-Citations20LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InRedditEmail PDF (676 KB) Get e-AlertscloseSUBJECTS:Environmental pollution,Materials,Nanomaterials,Silver,Toxicology Get e-Alerts

Nanosilver is one nanomaterial that is currently under a lot of scrutiny. Much of the discussion is based on the assumption that nanosilver is something new that has not been seen until recently and that the advances in nanotechnology opened completely new application areas for silver. However, we show in this analysis that nanosilver in the form of colloidal silver has been used for more than 100 years and has been registered as a biocidal material in the United States since 1954. Fifty-three percent of the EPA-registered biocidal silver products likely contain nanosilver. Most of these nanosilver applications are silver-impregnated water filters, algicides, and antimicrobial additives that do not claim to contain nanoparticles. Many human health standards for silver are based on an analysis of argyria occurrence (discoloration of the skin, a cosmetic condition) from the 1930s and include studies that considered nanosilver materials. The environmental standards on the other hand are based on ionic silver and may need to be re-evaluated based on recent findings that most silver in the environment, regardless of the original silver form, is present in the form of small clusters or nanoparticles. The implications of this analysis for policy of nanosilver is that it would be a mistake for regulators to ignore the accumulated knowledge of our scientific and regulatory heritage in a bid to declare nanosilver materials as new chemicals, with unknown properties and automatically harmful simply on the basis of a change in nomenclature to the term "nano".

ABSTRACT Nanotechnology enables development and production of novel silver-based composite materials. We used in vitro tests to demonstrate the antimicrobial activity of a silver-silica nanocomposite compared to the activities of conventional materials, such as silver nitrate and silver zeolite. A silver-silica-containing polystyrene material was manufactured and shown to possess strong antimicrobial properties.