Estimation of Polymerization Conditions Needed to Make Ethylene/1‐olefin Copolymers with Specific Microstructures Using Artificial Neural Networks

Two artificial neural network models (forward and inverse) are developed to describe ethylene/1‐olefin copolymerization with a catalyst having two site types using training and testing datasets obtained from a polymerization kinetic model. The forward model is applied to predict the molecular weight and chemical composition distributions of the polymer from a set of polymerization conditions, such as ethylene concentration, 1‐olefin concentration, cocatalyst concentration, hydrogen concentration, and polymerization temperature. The results of the forward model agree well with those from the kinetic model. The inverse model is applied to determine the polymerization conditions to produce polymers with desired microstructures. Although the inverse model generates multiple solutions for the general case, unique solutions are obtained when one of the three key process parameters (ethylene concentration, 1‐olefin concentration, and polymerization temperature) is kept constant. The proposed model can be used as an efficient tool to design materials from a set of polymerization conditions.

     

Determination of Operating Conditions of Ethylene/1-Octene Copolymerization Using Artificial Neural Network (ANN)

An artificial neural network (ANN) is applied to determine appropriate parameters in copolymerization of ethylene and 1-octene via metallocene catalytic system for producing a copolymer with desired chain microstructures. The polymerization parameters of interests are polymerization temperature, ethylene pressure, and the amount of hydrogen used. The ANN used is a feed-forward network with a back propagation learning method and has a 5-6-6-3 architecture. When comparing with both training and testing experimental data sets, it was found that ANN can provide a good guesstimation of polymerization parameters.     

Evaluation of chemical equilibria with the use of artificial neural networks

Multivariate calibration with experimental design (ED) and artificial neural networks (ANN) modeling can be used to estimate equilibria constants from any kind of protonation or metal–ligand equilibrium data like potentiometry, polarography, spectrophotometry, extraction, etc. The method was tested on evenly or randomly distributed experimental error-free data and data with random noise and the results show that even rather higher experimental errors do not influence significantly the prediction power and correctness of ANN prediction. ANN with appropriate ED can provide accurate prediction of stability constants with the relative errors in the range of ±4% or smaller while the approach is very robust. Comparison with a hard model evaluation based on non-linear regression techniques shows excellent agreement. Proposed ANN method is of a general nature and, in principal, can be adopted to any analytical technique used in equilibria studies.     

Modeling of Swelling Behaviors of Acrylamide-Based Polymeric Hydrogels by Intelligent System

Hydrogels based on acrylamide (AAm) were synthesized by free radical polymerization in an aqueous solution using N,N’-methylenebisacrylamide (MBAAm) as crosslinker. To obtain anionic hydrogels, 2-acrylamido-2-methylpropanesulfonic acid sodium salt (AMPS) and acrylic acid (AAc) were used as comonomers. The swelling behaviors of all hydrogel systems were modeled using an artificial neural network (ANN) and compared with a multivariable least squares regression (MLSR) model and phenomenal model. The predictions from the ANN model, which associated input parameters, including the amounts of crosslinker (MBA) and comonomer, and swelling values with time, produce results that show excellent correlation with experimental data. The parameters of swelling kinetics and water diffusion mechanisms of the hydrogels were calculated using the obtained experimental data. Model analysis indicated that the ANN models could accurately describe complex swelling behaviors of highly swellable hydrogels.     

Estimation of Average Comonomer Content of Ethylene/1-Olefin Copolymers Using Crystallization Analysis Fractionation (Crystaf) and Artificial Neural Network (ANN)

Summary: An artificial neural network (ANN) with a 4-3-3-1 architecture was developed to estimate average comonomer content of ethylene/1-olefin copolymers from crystallization analysis fractionation (Crystaf) results. The ANN was trained with a back propagation algorithm. It was found that average comonomer contents predicted from ANN agree well with experimental results for both training and testing data sets. The developed ANN was also used to systematically investigate the effects of chain microstructures and Crystaf operating conditions on Crystaf calibration curves.     

Application of artificial neural network in food classification

Artificial neural network (ANN) classifiers have been successfully implemented for various quality inspection and grading tasks of diverse food products. ANN are very good pattern classifiers because of their ability to learn patterns that are not linearly separable and concepts dealing with uncertainty, noise and random events. In this research, the ANN was used to build the classification model based on the relevant features of beer. Samples of the same brand of beer but with varying manufacturing dates, originating from miscellaneous manufacturing lots, have been represented in the multidimensional space by data vectors, which was an assembly of 12 features (% of alcohol, pH, % of CO(2) etc.). The classification has been performed for two subsets, the first that included samples of good quality beer and the other containing samples of unsatisfactory quality. ANN techniques allowed the discrimination between qualities of beer samples with up to 100% of correct classifications.     

Model Development in Thermal Styrene Polymerization

Summary: The thermal polymerization of styrene is usually modeled by relying on a reaction scheme and a set of equations that were developed more than three decades ago by Hui and Hamielec. Many detailed models of styrene polymerization are available in the open literature and they are mostly based on the work of Hui and Hamielec, which nearly makes this the standard to follow in explaining the behavior of polystyrene reactors. The model of Hui and Hamielec does a very nice job of describing monomer conversion data but discrepancies are seen between observed and predicted values of number and weight average molecular weights, M_n and M_w. Discrepancies in number average molecular weight seem to be the result of random noise. Discrepancies in weight average molecular weight grow as the polymerization temperature decreases and some of the trends observed in the residuals over the entire temperature range cannot be attributed to random noise. Hui and Hamielec attributed the observed deficiencies to a standard deviation of ±10% in their GPC measurements. A new data set with an experimental error of 2% for average molecular weights is presented. The set contains measured values of M_n, M_w and M_z, so the polymerization scheme has been extended to include third order moments. The data set also includes the effect of ethylbenzene as a chain transfer agent. We present the results of comparing model predictions to our measurements and the adjustments made in the original set of kinetic parameters published by Hui and Hamielec.     

偏氟乙烯-三氟乙烯共聚物的分子链结构与多晶型现象

偏氟乙烯-三氟乙烯共聚物(PVDF/TrFE)的合成及物性研究,是改善聚偏氟乙烯(PVDF)压电性的一种尝试。据文献报道,PVDF/TrFE 有明显的铁电性,以及强的压电性     

Steady State Simulation of Ethylene Polymerization Using Multiple-Site Coordination Catalysts

Summary: Simulation models are important tools for the development and optimization of polymerization processes because they can describe catalyst performance and polymer properties as a function of polymerization kinetics and process conditions. As the polyolefin industry moves towards the production of resins with more complex microstructures, these models become essential for process understanding and product design. A simulation model has been developed for the polymerization of ethylene in a process with n reactors working in series. The model can predict raw material conversions and product properties like the molecular weight distribution (MWD) coupled with short chain branching distribution (SCBD), melt index, density and fluff morphology. Model parameters have been obtained from laboratory data. The model predictions are in good agreement with experimental results.     

Model Reduction in Emulsion Polymerization Using Hybrid First‐Principles/Artificial Neural Network Models

A first‐principles mathematical model for emulsion polymerization was reduced by using a hybrid mathematical model composed by artificial neural networks (ANN) and material balances. The goal was to have an accurate model that may be integrated fast enough to be used for online optimization purposes. In the reduced model the polymerization rate and the instantaneous weight‐average molecular weight were calculated by means of artificial neural networks. These ANNs were incorporated to first‐principles material balances. The accuracy of the reduced model under a wide range of conditions was assessed. Savings in computer time were achieved by using the reduced model, which makes it suitable for online optimization purposes.

Effect of the temperature on the cumulative weight‐average molecular weight: first principles mathematical model (—); (ANN2) and hybrid model predictions: (▵) 50 °C, (▪) 60 °C^(training), (▿) 70 °C^(validation), (•) 80 °C, (○) 90 °C.     

Use of block copolymer surfactants in the inverse-suspension polymerization of acrylamide

A comprehensive experimental investigation of the inverse microsuspension polymerization of acrylamide using an oil-soluble initiator and a block copolymeric surfactant whose hydrophobic miety is poly(12-hydroxystearic acid) and whose hydrophilic moeity is polyethylene oxide was carried out. It was found that the initial polymerization rate was first order with respect to molar monomer concentration, first order with respect to molar initiator concentration and zeroth order with respect to molar emulsifier concentration. Based on these experimental findings, a mechanism was proposed which includes initiation, propagation transfer to monomer and termination. It also includes transfer to impurities which are believed to be found in the surfactant. The kinetic model developed from the proposed mechanism is found to be in good agreement with the experimental conversion and weight-average molecular weight data. Comparing with sorbitan esters of fatty acids, the copolymeric surfactant provides higher polymerization rate and very high and linear molecular weight comparable to those obtained by solution polymerization.     

Simulation of Non‐Linear Free‐Radical Polymerization Using a Percolation Kinetic Gelation Model

A new kinetic gelation model that incorporates the kinetics of representative non‐linear free‐radical polymerization is presented. Specifically, free‐radical homopolymerization, polymerization in the presence of a chain‐transfer agent (CTA, CTA‐induced polymerization), and copolymerization of a mixture of the bi‐ and tetrafunctional monomer is used to simulate kinetic effects on polymerization statistics and microstructures. An algorithm for random next‐step selection in a self‐avoiding random walk and efficient mechanisms of a component's mobility are introduced to improve the generality of the predictions by removing commonly occurring deficiencies due to early trapping of radicals. The model has the capability to take into account into several free‐radical polymerization mechanisms such as crosslinking, branching, and transfer reaction, and also to predict the onset of the sol–gel transition, and the effect of chemical composition on the transition point. It is shown that a better understanding of microstructure evolution during polymerization and chemical gelation is attained. Lastly, one important benefit of the model is to simulate very highly packed random chains or microgels within a polymer network.     

Aqueous solution properties of polyampholytes: Effect of the net charge distribution

Viscometric and light scattering studies have been performed on aqueous solutions of polyampholyte terpolymers based on sodium-2-acrylamido-2-methylpropanesulfonate (Na-AMPS), 2-(methacryloyloxy)-ethyltrimethylammonium chloride (MADQUAT), and acrylamide (AM), prepared by an inverse microemulsion polymerization technique. The distribution of net charges among the chains was varied by adjusting the initial monomer composition and the degree of conversion. The effect of this distribution on the solubility of the samples and on the chain conformation was studied. It was found that samples with a narrow distribution of net charges were soluble in pure water even if the average net charge is small. Addition of salt induces a transition from an extended conformation to a more compact one, in qualitative agreement with theoretical predictions. A practically alternated NaAMPS-MADQUAT copolymer prepared by polymerization in homogeneous solution and with a small average net charge shows a behavior quite similar to that of the terpolymers. © 1996 John Wiley & Sons, Inc.     

The inverse emulsion polymerization of acrylamide using polystyrene-graft-polyoxyethylene as the stabilizer

Inverse emulsion polymerization of aqueous solution of acrylamide (AM) in toluene is carried out using polystyrene-graft-polyoxyethylene (PSt-g-PEO) as an emulsifier. The kinetics of polymerization, morphology of the particle, and particle size of the inverse emulsion have been investigated. The rates of polymerization are found to be proportional to the initiator concentration, the monomer concentration, and the emulsifier concentration. The morphology of the particle shows a spherical structure. The effects of amphipathic graft copolymer structure on the average molecular weight of polyacrylamide are studied. The mechanism of the inverse emulsion polymerization using amphipathic graft copolymer as emulsifier is proposed. © 1999 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 37: 2719–2725, 1999     

Modeling of elementary reactions and kinetics of radical‐initiated methyl methacrylate polymerization in the presence of ferrocene

On the basis of quantum chemical modeling, a kinetic scheme of methyl methacrylate polymerization initiated by benzoyl peroxide in the presence of ferrocene was proposed. The process runs by mechanism, which includes the reactions of free radical polymerization, and the reactions leading to formation and operability of two type coordination active sites that are capable of converting into each other. On the basis of the proposed scheme, a kinetic model was developed. This model quantitatively described the following: the experimentally determined time dependences of the methyl methacrylate conversion, the conversion dependencies of the number‐average and weight‐average molar masses of poly(methyl methacrylate), the stereoregularity values of poly(methyl methacrylate), and the time dependencies of the methyl methacrylate conversion upon its polymerization on poly(methyl methacrylate) macroinitiators obtained in radical‐initiated polymerization in the presence of ferrocene. As a result of solving the inverse kinetic problem, the parameters of temperature dependences of the reaction rate coefficients of the proposed kinetic scheme were found.     

主成分-人工神经网络在近红外光谱定量分析中的应用

近红外光谱的主成分由非线性迭代偏最小二乘法（ＮＩＰＡＬＳ）求出。主成分作标准化处理后,作为Ｂ－Ｐ神经网络的输入结点进行非线性迭代。该法的优点是,充分利用了全光谱的数据,得到消除噪声后的最佳主成分,能建立非线性模型,Ｂ－Ｐ神经网络迭代时间显著缩短。用该法对大麦中的淀粉含量进行了定量分析研究。结果为：校准和预测的相关系数分别为０．９８１和０．９５３,校准和预测的相对标准偏差分别为１．７０％和２．４８％。     

Chemical recycling of poly(methyl methacrylate) by pyrolysis. Potential use of the liquid fraction as a raw material for the reproduction of the polymer

Pyrolysis of poly(methyl methacrylate) (PMMA) was studied as an effective way to recycle this polymer and recover its monomer methyl methacrylate (MMA). Experiments were carried out in a laboratory fixed bed reactor using either a model polymer or a commercial product based on PMMA as feedstock. Gaseous and liquid products obtained from polymer degradation were analysed and it was found that the oil fraction constituted mainly of the MMA monomer. Thus, the possibility of directly using the liquid product for the reproduction of the polymer was further investigated. Polymerizations accomplished in a differential scanning calorimeter using azo-bis-isobutyronitrile as initiator and different reaction temperatures. Results obtained were compared to corresponding data from polymerization of neat monomer. It was found that the pyrolysis liquid fraction can be polymerized and produce a polymer similar to the original PMMA. However, even small amounts of other organic compounds (mainly methyl esters) included in this fraction act as non-ideal reaction retarders, altering the reaction rate curve and lowering the glass transition temperature and the average molecular weight of the polymer produced.     

Classification of time-of-flight secondary ion mass spectrometry spectra from complex Cu–Fe sulphides by principal component analysis and artificial neural networks

Artificial neural network (ANN) and a hybrid principal component analysis-artificial neural network (PCA-ANN) classifiers have been successfully implemented for classification of static time-of-flight secondary ion mass spectrometry (ToF-SIMS) mass spectra collected from complex Cu–Fe sulphides (chalcopyrite, bornite, chalcocite and pyrite) at different flotation conditions. ANNs are very good pattern classifiers because of: their ability to learn and generalise patterns that are not linearly separable; their fault and noise tolerance capability; and high parallelism. In the first approach, fragments from the whole ToF-SIMS spectrum were used as input to the ANN, the model yielded high overall correct classification rates of 100% for feed samples, 88% for conditioned feed samples and 91% for Eh modified samples. In the second approach, the hybrid pattern classifier PCA-ANN was integrated. PCA is a very effective multivariate data analysis tool applied to enhance species features and reduce data dimensionality. Principal component (PC) scores which accounted for 95% of the raw spectral data variance, were used as input to the ANN, the model yielded high overall correct classification rates of 88% for conditioned feed samples and 95% for Eh modified samples.     

Synthesis of copolymers containing acetals

The polymerization of cyclic acetals such as dioxolane and dioxepane, initiated by triflic acid in the presence of poly(ethylene oxide)glycol (PEOG), exhibits special features: One can produce triblock copolymers of PEOG and of one cyclic acetal; the molecular weight of the copolymer increases linearly with time like in a living system; a hydroxyl group is found at both ends of the copolymer obtained; macrocycle formation, usually inevitable in large quantity when a classical onium mechanism operates, can be very much reduced and even practically suppressed depending on the experimental conditions. When we tried to prepare a pentablock from the triblock by adding a second acetal under similar conditions (after recovery and purification of the triblock copolymer) we obtained fast transacetalization instead, leading to random placements of the two cyclic acetals, while the PEOG remained as one block, and that only a very small amount of cycles was formed if any. These results are interpreted in terms of multiple transfer reactions of hydroxyl groups onto activated monomer molecules or onto small growing oligomers.     

Novel approaches for the prediction of density of glycol solutions

Two new approaches for the accurate prediction of densities of the commonly used glycol solutions in the gasprocessing industry are presented in the article. The first approach is based on developing a simple-to-use polynomial correlation for an appropriate prediction of density of glycol solutions as a function of temperature and weight percent of glycols in water, where the obtained results show very good agreement with the reported experimental data. The second approach, however, is based on the artificial neural networks （ANN） methodology, wherein the results demonstrate the ability of the introduced method to predict reasonably accurate densities of glycols under operating conditions. Comparisons of the two novel approaches indicated that the simple-to-use correlation appears to be superior owing to its simplicity and clear numerical background, wherein the relevant coefficients can be retuned if new and more accurate data are available in the future. The average deviation of the new proposed polynomial correlation results from reported data is 0.64 kg/m^3 whereas the average deviation of artificial neural networks （ANN） methodology from reported data is 1.1 kg/m^3.