Experimental methods in chemical engineering: Process simulation

Process simulation software designs equipment, simulates operations, optimizes a plant's configuration (heat exchangers network, for example), estimates operating and capital expenses, and serves as an educational tool. However, mastering the theoretical background minimizes common mistakes such as applying an incorrect thermodynamic method, selecting improper algorithms in the case of tear systems, and setting irrational system specifications. Engineers and researchers will exploit this tool more often in the future as constant advancements in simulation science as well as new models are released continually. Process simulators make it easier to build digital twins and thus will facilitate the implementation of the industry 4.0 guidelines. We highlight the mathematical and technical features of process simulators, as well as the capabilities and the fields of application. A bibliometric map of keywords from articles citing Aspen+, Aspen plus, Hysys, and Pro/II indexed by Web of Science between 2017 and 2020 identified the main research clusters, such as design, optimization, energy or exergy, biomass; H₂ and CO₂ capture, thermodynamics; and separations and techno-economic analysis.     

Experimental methods in chemical engineering: Micro‐reactors

Whereas the bulk chemical industry has historically sought economic advantage through economies of scale, a paradigm shift has researchers developing systems on smaller scales. Nano‐cages and nano‐actuators increase selectivity and robustness at the molecular scale. In parallel, micro‐contactors with sub‐millimetre lateral dimensions are decreasing boundary layers that restrict heat and mass transfer and thus meet the objectives of process intensification with great increases in productivity with a smaller footprint. These contactors continue to serve chemical engineers and chemists to synthesize fine chemicals and characterize catalysts; however, they have now been adopted for sensors in biological and biochemical systems. A bibliometric analysis of articles indexed in the Web of Science in 2016 and 2017 identified five major clusters of research: catalysis and bulk chemicals; nanoparticles; organic synthesis and flow chemistry; systems and micro‐fluidics applied to biochemistry; and micro‐channel reactors and mass transfer. In the early 1990s, less than 100 articles a year mentioned micro‐reactors, while over 943 articles mentioned it in 2017. Here, we introduce micro‐reactors and their role in the continuous synthesis of fine chemicals across the various scales to commercialization.     

Experimental methods in chemical engineering: Rheometry

Rheometry is an experimental technique that measures the relationship between stresses, strains, and their derivatives of fluids and deformable materials. Capillary and Couette viscometers were among the first instruments used to study rheology but now sophisticated shear and extensional rheometers are widely available for quality control, research, and product development. Here, we introduce the basics of rheology, define material functions, and describe conventional instruments, physical principles, applications, and uncertainties. In 2016 and 2017, the Web of Science indexed 8400 articles that mention rheometry and a bibliometric map assigned the keywords into five clusters of research: behaviour and viscosity, mechanical properties and morphology, rheological properties and microstructure, polymers and nano-particles, and visco-elastic properties and mixtures. Journal of Rheology, Construction Building Materials, Food Hydrocolloids, Soft Matter, Rheologica Acta, and Journal of Applied Polymer Science publish at least 80 articles per year that mention rheometry.     

Experimental methods in chemical engineering: Artificial neural networks–ANNs

Artificial neural networks (ANNs) are one of the most powerful and versatile tools provided by artificial intelligence and they have now been exploited by chemical engineers for several decades in countless applications. ANNs are computational tools providing a minimalistic mathematical model of neural functions. Coupled with raw data and a learning algorithm, they can be applied to tasks such as modelling, classification, and prediction. Recently, their popularity has grown remarkably and they now constitute one of the most relevant research areas within the fields of artificial intelligence and machine learning. ANNs are large collections of simple classifiers called neurons. Chemical engineers apply them to model complex relationships, predict reactor performance, and to automate process controllers. ANNs can leverage their ability to learn and exploit large data sets, but they can also get stuck in local minima or overfit and are difficult to reverse engineer. In 2016 and 2017, ANNs were cited in 13 245 Web of Science (WoS) articles, 538 of which were in chemical engineering; the top WoS categories were electrical & electronic engineering (1615 occurrences) artificial intelligence (1253), and energy & fuels (980). The top 4 journals mentioning ANNs were Neural Computing & Applications (117), Neurocomputing (84), Energies (76), and Renewable & Sustainable Energy Reviews (76). In the near future, as larger data sets become available (and arduous to analyze), chemical engineers will be able to apply and leverage more sophisticated ANN architectures.     

Experimental methods in chemical engineering: Raman spectroscopy

When photons impinge on a substrate, most scatter with the same frequency (elastic scattering or Rayleigh dispersion) and only 10^?7 scatter with a different energy (inelastic scattering). This inelastic interaction (Raman scattering) exchanges energy in the region of molecular vibrational transitions for crystalline and amorphous materials. Raman bands in a spectra represent vibrational transitions, like infrared, however the selection rules are different. Typically, the vibrations that are intense in Raman are weak in infrared and vice versa. A remarkable feature of the Raman effect is that it is highly sensitive to nanocrystals, even below 4 nm, which are too small to generate XRD patterns. Plasmonic enhancement, like surface‐enhanced Raman spectroscopy (SERS) boost the Raman signal by 10⁴, providing single‐molecule detection capability. Glass, quartz, and sapphire are transparent to Raman effect (depending on the energy of the incident excitation radiation), which makes it ideal to examine materials under reaction conditions (in‐situ cells and operando reactors that operate over a broad range of temperature, pressures, and environments). Raman spectroscopy emerged in the 1930s; however, infrared spectrometry displaced it. With the advent of powerful lasers in the 1970s, more researchers began to apply Raman routinely. In 2019, the Web of Science indexed 20 400 articles mentioning Raman against 50 000 articles mentioning infrared. Chemical engineers apply Raman less frequently than in material science, physical chemistry, and applied physics, with 887 articles vs 6250, 3700, and 3510 for the other disciplines. A bibliometric analysis identified four research clusters centred on thin films and optics, graphene and nanocomposites, nanoparticles and SERS, and photocatalyst.     

Experimental methods in chemical engineering: Fluorescence emission spectroscopy

Fluorescence is a luminescence phenomenon in which a compound emits light after absorption of electromagnetic irradiation. Specialized terms such as photoluminescence, cathodoluminescence, anodoluminescence, radioluminescence, and x‐ray fluorescence sometimes are used to indicate the type of exciting radiation. Fluorescence spectroscopy provides reliable quantitative and qualitative data. It precisely tracks chemical reactions from fluorescent materials compounds with aromatic groups, or conjugated planar, or cyclic molecules. It is up to 1000 times more sensitive than UV‐vis or infrared spectroscopy. Fluorescence intensity depends on the fluorophore (compound that fluoresces), its concentration, excitation and emission wavelengths, temperature and contamination. We adjust the slit dimensions, photomultiplier tube voltage and bandpass filter cutoff to maximize the signal while avoiding saturating the detector. Together with x‐ray diffraction, it is the most common spectroscopic technique with applications in geology, chemistry, medicine, and astronomy. A bibliometric analysis of the top 10 000 cited papers identified 5 clusters based on keywords centered around: (1) cancer, cells, and proteins; (2) aggregation induced emission, LED, and complexes; (3) live cells, sensors, and probes; (4) quantum dots, DNA, and biosensors; and (5) nanoparticles, in vivo, and drug delivery. Chemical engineers have yet to fully embrace fluorescence spectroscopy as the category is ranked 16th among all scientific categories that exploit it.     

基于数据特征提取的AANN-ELM研究及化工应用

针对极限学习机不能有效解决化工过程中高维数据建模的问题,本文将其与自联想神经网络结合,通过自联想神经网络过滤输入数据中存在的冗余信息、提取特征分量,并对所提取的特征分量采用极限学习机进行训练,由此形成了一种基于数据特征提取的AANN-ELM(auto-associative neural network-extreme learning machine)神经网络。同时,以UCI标准数据集进行测试,以精对苯二甲酸(PTA)溶剂系统进行验证,结果表明,AANN-ELM在处理高维数据时具有学习速度快、网络稳定性强、建模精度高的特点,为神经网络在复杂化工生产中的应用提供了新思路。     

Experimental methods in chemical engineering: Reactors—fluidized beds

Industry relies on fluidized beds to synthesize chemicals (acrylonitrile, maleic anhydride, titanium dioxide, vinyl chloride), combust coal, dry powders, and treat waste. Fluidized bed folklore declares that they are hard to scale‐up and the gas phase is backmixed. Commercial failures that disregard standard design criteria around powder management, gas/solids injection, and mixing reinforce this belief. However, engineers select fluidized beds for processes that are impractical with conventional technologies to achieve economies of scale for highly exothermic, endothermic, or explosive reactions, for catalysts that deactivate in seconds (or minutes), and for chemistry that requires multiple dosing cycles. Failures are more frequent for these challenging applications. For this reason, researchers study reaction kinetics in fixed beds despite internal mass transfer limitations and axial and radial temperature and concentration gradients. Fluidized bed hydrodynamics vary with powder properties (particle diameter, size distribution, density, sphericity), operating conditions (gas density, viscosity, temperature, pressure), reactor geometry (diameter, height, mass, grid geometry). The minimum fluidization velocity (U_mf) is a property that identifies the transition from the fixed bed regime to the fluidized bed regime and equals the gas velocity at which the upward drag force equals the weight of the powder. At the experimental scale, fluidized beds operate isothermally, solids are completely backmixed, and the gas phase is close to plug flow (). Here, we describe the relationship between powder properties and fluidization quality, list experimental techniques, describe recent applications, and gas phase hydrodynamics and uncertainties.     

Experimental methods in chemical engineering: Electron paramagnetic resonance spectroscopy-EPR/ESR

Electron paramagnetic resonance (EPR) spectroscopy, also known as electron spin resonance spectroscopy (ESR), utilizes absorption of microwave radiation by unpaired electrons in a magnetic field. The interaction between the unpaired electron(s) and nearby magnetic nuclei helps identify paramagnetic species and can provide information about the motion of the molecule and the local polarity, pH, viscosity, concentration, and accessibility to other paramagnetic species. This mini-review discusses the fundamental underpinnings of EPR needed to correctly interpret EPR spectra. We describe various types of EPR spectra encountered by chemical engineers, and use application examples drawn from the chemical engineering literature to illustrate the information available from the technique. Few chemical engineering departments or even chemistry departments have EPR instruments, which contributes to the significant barrier that prevents this being adopted as a routine measurement technique. However, in 2016 and 2017, Web of Science indexed 7000 articles that applied EPR spectroscopy. A bibliometric map categorized the keywords in four categories based on co-occurrences: magnetic properties, films, and luminescence; crystal structure, complexes, and ligands; nanoparticles, oxidation, and degradation; and, systems, radicals, and H₂ O₂ .     

Smart systems engineering contributing to an intelligent carbon-neutral future: opportunities,challenges, and prospects

This communication paper provides an overview of multi-scale smart systems engineering (SSE) approaches and their applications in crucial domains including materials discovery, intelligent manufacturing, and environmental management. A major focus of this interdisciplinary field is on the design, operation and management of multi-scale systems with enhanced economic and environmental performance. The emergence of big data analytics, internet of things, machine learning, and general artificial intelligence could revolutionize next-generation research, industry and society. A detailed discussion is provided herein on opportunities, challenges, and future directions of SSE in response to the pressing carbon-neutrality targets.     

基于数据属性划分的递阶ELM研究及化工应用

针对极限学习机（ELM）不能有效处理化工过程中强耦合、带噪声的高维数据建模问题,提出了一种基于数据属性划分的递阶ELM神经网络DHELM。该神经网络采用数据属性划分（DAD）方法对高维输入进行聚类、建立自联想子网,并将自联想子网所提取的特征分量作为极限学习机的输入进行建模。同时,利用UCI标准数据集进行了测试,通过工业应用实例进行了验证,并进行了模型对比。结果表明,DHELM网络在处理复杂高维数据时具有收敛速度快、建模精度高、网络稳定性强的特点,为神经网络发展及其化工应用提供了新思路。     

Reconciling deep learning and first-principle modelling for the investigation of transport phenomena in chemical engineering

The use of machine learning in chemical engineering has the potential to greatly improve the design and analysis of complex systems. However, there are also risks associated with its adoption, such as the potential for bias in algorithms and the need for careful oversight to ensure the safety and reliability of machine learning-powered systems. This paper explores the opportunities and risks of using machine learning in chemical engineering and provides a perspective on how it may be integrated into engineering practices in a responsible and effective manner. We generated the text of this abstract with GPT-3, OpenAI's large-scale language-generation model. Upon generating the draft, we ensured that the language was to our liking, and we take ultimate responsibility for the content of this publication.     

Experimental methods in chemical engineering: X‐ray photoelectron spectroscopy‐XPS

X‐ray photoelectron spectroscopy (XPS) is a quantitative surface analysis technique used to identify the elemental composition, empirical formula, chemical state, and electronic state of an element. The kinetic energy of the electrons escaping from the material surface irradiated by an x‐ray beam produces a spectrum. XPS identifies chemical species and quantifies their content and the interactions between surface species. It is minimally destructive and is sensitive to a depth between 1–10nm. The elemental sensitivity is in the order of 0.1 atomic %. It requires ultra high vacuum ( Pa) in the analysis chamber and measurement time varies from minutes to hours per sample depending on the analyte. XPS dates back 50 years ago. New spectrometers, detectors, and variable size photon beams, reduce analysis time and increase spatial resolution. An XPS bibliometric map of the 10 000 articles indexed by Web of Science^[1] identifies five research clusters: (i) nanoparticles, thin films, and surfaces; (ii) catalysis, oxidation, reduction, stability, and oxides; (iii) nanocomposites, graphene, graphite, and electro‐chemistry; (iv) photocatalysis, water, visible light, and ; and (v) adsorption, aqueous solutions, and waste water.     

Experimental methods in chemical engineering: X-ray absorption spectroscopy—XAS,XANES, EXAFS

Although X-ray absorption spectroscopy (XAS) was conceived in the early 20th century, it took 60 years after the advent of synchrotrons for researchers to exploit its tremendous potential. Counterintuitively, researchers are now developing bench type polychromatic X-ray sources that are less brilliant to measure catalyst stability and work with toxic substances. XAS measures the absorption spectra of electrons that X-rays eject from the tightly bound core electrons to the continuum. The spectrum from 10 to 150 eV (kinetic energy of the photoelectrons) above the chemical potential—binding energy of core electrons—identifies oxidation state and band occupancy (X-ray absorption near edge structure, XANES), while higher energies in the spectrum relate to local atomic structure like coordination number and distance, Debye-Waller factor, and inner potential correction (extended X-ray absorption fine structure, EXAFS). Combining XAS with complementary spectroscopic techniques like Raman, Fourier transform infrared (FTIR), X-ray photoelectron spectroscopy (XPS), and electron paramagnetic resonance (EPR) elucidates the nature of the chemical bonds at the catalyst surface to better understand reaction mechanisms and intermediates. Because synchrotrons continue to be the light source of choice for most researchers, the number of articles Web of Science indexes per year has grown from 1000 in 1991 to 1700 in 2020. Material scientists and physical chemists publish an order of magnitude articles more than chemical engineers. Based on a bibliometric analysis, the research comprises five clusters centred around: electronic and optical properties, oxidation and hydrogenation catalysis, complementary analytical techniques like FTIR, nanoparticles and electrocatalysis, and iron, metals, and complexes.     

化工过程的智能混合建模方法及应用

随着人工智能技术和配套数据系统的快速发展,化工过程建模技术达到了新的高度,将多个机理模型和数据驱动模型以合理的结构加以组合的智能混合建模方法,可以综合利用化工过程的第一性原理及过程数据,结合人工智能算法以串联、并联或者混联的形式解决化工过程中的模拟、监测、优化和预测等问题,建模目的明确,过程灵活,形成的混合模型有着更好的整体性能,是近年来过程建模技术的重要发展趋势。本文围绕近年来针对化工过程的智能混合建模工作进行了总结,包括应用的机器学习算法、混合结构设计、结构选择等关键问题,重点论述了混合模型在不同任务场景下的应用。指出混合建模的关键在于问题和模型结构的匹配,而提高机理子模型性能,获取高质量宽范围的数据,深化对过程机理的理解,形成更有效率的混合建模范式,这些都是现阶段提高混合建模性能的研究方向。     

Experimental methods in chemical engineering: specific surface area and pore size distribution measurements—BET,BJH, and DFT

Gas physisorption is an experimental technique based on equilibrium Van der Waals interactions between gas molecules and solid particles, that quantifies the specific surface area (SSA), pore size distribution (PSD), and pore volume of solids and powders. The performance of catalysts, absorbents, chromatography column materials, and polymer resins depends on these morphological properties. Here we introduce the basic principles and procedures of physical adsorption, especially nitrogen physisorption, as a guide to students and researchers unfamiliar with the field. The Brunauer‐Emmett‐Teller theory (BET) is a common approach to estimate SSA that extends the Langmuir monolayer molecular adsorption model to multilayer layers. It relies on an equilibrium adsorption isotherm, measured at the normal boiling point of the adsorbate, eg, 77 K or 87 K for N₂ and Ar, respectively. Web of Science indexed 45 400 articles in 2016 and 2017 that mentioned N₂ adsorption porosimetry—BET and BJH (Barrett‐Joyner‐Halenda) keywords. The VOSViewer bibliometric tool grouped these articles into four research clusters: adsorption, activated carbon in aqueous solutions for removal of heavy metal ions; synthesis of nanoparticles and composites; catalysts performance in oxidation and reduction processes; and photocatalytic degradation with TiO₂. According to the literature, the accuracy of the density function theory (DFT) method is higher than with the BJH theory and it is more reliable.