1,721,098 research outputs found
AI3SD video: development of a full stack for digital R&D in chemistry and chemical process development
No full textIn order to enable seamless access to AI tools in research, it is necessary to transform how our laboratories are equipped. AI requires access to data, and it takes too long to gain access and to clean up datasets. Our experimental hardware is not ‘wired’ and is not accessible to algorithms. What is required is a development of data architecture that enables access to experimental and literature data both to a ‘human in the middle’ and fully algorithmic research tasks. In this talk I’ll present our joint effort with the group of Prof Markus Kraft to implement knowledge graph for ML workflow in chemical synthesis development, and the work @ iDMT centre in Cambridge on expanding this to a fully digital R&D in molecular sciences
Machine learning-guided space-filling designs for high throughput liquid formulation development
No full textLiquid formulation design involves using a relatively limited experimental budget to search a high-dimensional space, owing to the combinatorial selection of ingredients and their concentrations from a larger subset of available ingredients. This work investigates alternative shampoo formulations. A space-filling design is desired for screening relatively unexplored formulation chemistries. One of the few computationally efficient solutions for this mixed nominal-continuous design of experiments problem is the adoption of maximum projection designs with quantitative and qualitative factors (MaxProQQ). However, such purely space-filling designs can select experiments in infeasible regions of the design space. Here, stable products are considered feasible. We develop and apply weighted-space filling designs, where predictive phase stability classifiers are trained for difficult-to-formulate (predominantly unstable) sub-systems, to guide these experiments to regions of feasibility, whilst simultaneously optimising for chemical diversity by building on MaxProQQ. This approach is extendable to other mixed-variable design problems, particularly those with sequential design objectives
Multi-objective reaction optimization under uncertainties using expected quantile improvement
No full textMulti-objective Bayesian optimization (MOBO) has shown to be a promising tool for reaction development. However, noise is usually inevitable in experimental and chemical processes, and finding reliable solutions is challenging when the noise is unknown or significant. In this study, we focus on finding a set of optimal reaction conditions using multi-objective Euclidian expected quantile improvement (MO-E-EQI) under noisy settings. First, the performance of MO-E-EQI is evaluated by comparing with some recent MOBO algorithms in silico with linear and log-linear heteroscedastic noise structures and different magnitudes. It is noticed that high noise can degrade the performance of MOBO algorithms. MO-E-EQI shows robust performance in terms of hypervolume-based metric, coverage metric and number of solutions on the Pareto front. Finally, MO-E-EQI is implemented in a real case to optimize an esterification reaction to achieve the maximum space-time-yield and the minimal E-factor. The algorithm identifies a clear trade-off between the two objectives.</p
Highly selective Pd/titanate nanotube catalysts for the double-bond migration reaction
No full textPd(II) and Pd(0) catalysts supported onto titanate nanotubes (H2Ti3O7) were prepared by an ion-exchange technique. The catalysts are characterised by narrow size distribution of metal nanoparticles on the external surface of the nanotubes. Pd(II) catalysts show high selectivity toward double-bond migration reaction versus hydrogenation in linear olefins. The catalytic activity exhibits a volcano-type dependence on the metal loading, with the maximum activity observed at ca. 8 wt%. The Pd(II) was shown to be rapidly reduced to Pd(0) by appropriate choice of solvent. Prereduced Pd(0) catalysts were found to be less active toward double-bond migration and more selective toward hydrogenation. The DBM reaction was faster in protic solvents, such as methanol or ethanol
Selective oxidation of alcohols in a continuous multifunctional reactor: ruthenium oxide catalysed oxidation of benzyl alcohol
No full textA multifunctional compact reactor was designed, built, and tested, using as a model reaction the selective oxidation of benzyl alcohol to benzaldehyde with molecular oxygen. The reactor contains static mixers, heat-transfer channels and mm-scale packed-bed reaction channels, i.e., integrating mixing, heat transfer and reaction functionalities. Integrated compact reactor technology should be particularly attractive to the fine chemistry and pharmaceutical industries. The reactor was shown to operate in the kinetic regime over a broad range of operating conditions due to intensified mass transfer. The reactor was also shown to operate isothermally despite high reaction rate, an appreciable heat effect and high reactant concentration (TOF = 300 h-1, ΔH°r = -183.7 kJ mol-1). Staged injection of oxygen along the length of the reactor was investigated as a method of increasing the selectivity of oxidation reactions. Staged injection was shown to be beneficial; however, this was likely to be due to the development of a more uniform hydrodynamic regime of two-phase flow along the packed reaction channel. Experiments were performed with the ruthenium(III) hydrated oxide catalyst supported on alumina (0.9 wt% Ru/Al2O3). High activity and selectivity were observed even when working with reactant concentrations approaching industrial conditions (1 mol L-1)
Application of a structured multifunctional reactor for the oxidation of a liquid organic feedstock
No full textExperiments are performed in a structured packed bed reactor that has been developed for the selective oxidation of organic liquids to produce pharmaceutical products. Molecular oxygen is used as the oxidant, providing a “cleaner processing route”, than in established methods. The structured assembly consists of an integrated heat exchange system, a gas/liquid mixing zone, and provision for reactant injection. To illustrate the application, the selective oxidation of benzyl alcohol to benzaldehyde was studied. The reactor consisted of parallel reaction channels (10 cm long and square shaped) that were packed with Ru/Al2O3 (0.9 wt% Ru) catalyst particles. From experiments in three different sizes of channel: 2 mm × 2 mm, 3 mm × 3 mm, and 5 mm × 5 mm (at P = 8 bar, T = 388 K, with liquid and gas flows L = 3.2 kg m?2 s?1, G = 1.35 × 10?2 kg m?2 s?1), it was shown that in a short 10 cm length of channel, a product yield up to ca. 29% (with minimum 99.5% selectivity) could be obtained. Conversions were highest in the 3 mm × 3 mm channel. From the experiments with the reactor mounted both in horizontal and vertical planes, it was shown that gravitational influences are minor.The choice of solvent was shown to affect the yield of benzaldehyde. The highest yield was with toluene; the application of dioxane/water mixture (70/30 by vol.) and N-methyl pyrrolidone (NMP) not only decreased the yield by factor 2, but NMP also caused deactivation of the catalyst.Preliminary results are presented of work with Pt/C (Pt 3 wt%) catalyst
TiO2 nanotube-supported ruthenium(III) hydrated oxide: a highly active catalyst for selective oxidation of alcohols by oxygen
No full textRuthenium(III) hydrated oxide was deposited onto the surface of TiO2 nanotubes via ion exchange followed by alkali treatment. The catalytic activity of Ru(III)/TiO2 catalyst for the selective oxidation of alcohols by oxygen was studied in a continuous multichannel, compact reactor. It was found that the activity of Ru(III)/TiO2 (turnover frequency [TOF] ? 450 h?1) is greater than that of a recently reported Ru(III)/Al2O3 catalyst (TOF = 335 h?1), and that the ruthenium loading could exceed 8 wt% without any significant loss of specific catalytic activity. High-resolution transmission electron microscopy and X-ray diffraction data show that ruthenium is highly dispersed on both the inner and the outer surfaces of the mesoporous TiO2 nanotubes with the diameter of nanoparticles between 1.2 and 2.4 nm, and is easily accessible by reactant molecules
Reversible storage of molecular hydrogen by sorption into multilayered TiO2 nanotubes
No full textThe sorption of hydrogen between the layers of the multilayered wall of nanotubular TiO2 was studied in the temperature range of -195 to 200 °C and at pressures of 0 to 6 bar. Hydrogen can intercalate between layers in the walls of TiO2 nanotubes forming host-guest compounds TiO2·xH2, where x ? 1.5 and decreases at higher temperatures. The rate of hydrogen incorporation increases with temperature and the characteristic time for hydrogen sorption in TiO2 nanotubes is several hours at 100 °C. The rate of intercalate formation is limited by the diffusion of molecular hydrogen inside the multilayered walls of the TiO2 nanotube. 1H NMR-MAS and XRD data confirm the incorporation of hydrogen between the layers in the walls of TiO2 nanotubes. The nature and possible applications of the observed intercalates are considered
Heterogeneous benzaldehyde nitration in batch and continuous flow microreactor
No full textOrtho- and meta-nitrobenzaldehydes are two important intermediates in the chemical industry. They are the main products of benzaldehyde nitration by mixed acid, which is a hazardous chemical process since reaction rates are very sensitive to variations of the operating conditions. In the present investigation, a previously developed kinetic model under homogeneous conditions is validated and extended for heterogeneous liquid-liquid systems. A comparison between performances of a lab-scale batch reactor and a commercial microreactor is also presented. The results show that microreactors with embedded static mixer can outperform scaled-up batch reactors, achieving the theoretical limit of conversion/yield attainable under kinetic control
Liquid-phase oxidation of organic feedstock in a compact multichannel reactor
No full textA millimeter-scale compact multifunctional reactor has been developed and tested for the selective oxidation of benzyl alcohol, using molecular oxygen as the oxidant. The reactor consists of parallel reaction channels that are packed with catalyst particles. Within the structured assembly are an integrated heat-exchange system, a gas/liquid mixing zone, and a provision for reactant injection. Experiments were performed in square channels that are 10 cm long with a cross-sectional area of 9 mm2. These were packed either with Ru/Al2O3 (0.9 wt % of Ru) or Ru/TiO2 (2 wt % of Rt) catalysts. Hydrodynamic characteristics of the system were evaluated to establish pressure drop and effectiveness of gas/liquid mixing and to confirm the nature of the gas-liquid flow regime. Operating the reactor at 8 bar and 388 K with a liquid flow L = 3.2 kg m-2 s-1 and gas flows G > 2.5 × 10-2 kg m-2 s-1, it was shown that, even in a short 10 cm length of channel, a product yield of up to 55% (with 99.7% selectivity) could be obtained. Although the adiabatic temperature rise at 55% yield is estimated to be ca. 180 K, the reactor was shown to operate isothermally, because of the efficient removal of heat through the integrated micro-heat-exchange system. It was concluded that this structured design of a reactor showed considerable promise for the development of cleaner oxidation processes
- …
