|
|
Home » Publications » Articles
ArticlesVisit ReserachGate to find some additional materials 28. A Kossoy, Some thoughts about statistical evaluation of the results of kinetic analysis (A. Kossoy) DOI: 10.13140/RG.2.2.31782.80962 ABSTRACT When studying kinetics, as with any other measurements, it is necessary to evaluate the accuracy of the results. In kinetic analysis, this comes down to estimating the errors of the model parameters and assessing the quality of the model used to fit the data by applying certain statistical methods. Typically, classical regression analysis is used for this purpose. Due to the specificity of the results of kinetic experiments, this well-known and rather simple to use method turns out to be inapplicable. One of the main reasons is that in many cases the results of a kinetic experiment are represented not by a series of point measurements, but by the so-called time series or, more simply, by kinetic curves that represent the change of some characteristic quantity in time. Neglecting this feature leads to unreliable or even completely wrong results. Our aim here is to draw attention to this serious problem and to discuss (albeit on the dilettante’s level) possible solutions. 27. A. Kossoy, Yu. Akhmetshin, Applying reaction calorimetry for studying reaction kinetics, Report, June 2023 ABSTRACT Reaction calorimetry (RC) is a powerful experimental method, widely used for the reaction hazard assessment, in chemical engineering for chemical processes design, and in other fields. 26. A. Kossoy, Reaction calorimetry: Main types, simple theory, and application for kinetic study — A review. Process Safety Progress, April 2023, ABTSRACT Reaction calorimetry is a powerful experimental method widely used for reaction hazard assessment in chemical engineering for chemical process design and in other fields. Currently, several types of reaction calorimeters are available, each with its own features. Many articles and several monographs have been published dealing with various aspects of reaction calorimetry, but it is difficult to find an overview of different types of instruments and the theoretical basis of reaction calorimetry in one review in a compact form. The purpose of this article is to fill this gap as much as possible and to help in understanding the principles of this useful experimental method and in choosing the type of instrument most suitable for the purposes of a particular study. The review concludes with some recommendations for the proper organization of an experiment in reaction calorimetry, with an emphasis on the kinetic study. The usefulness of kinetics based on reaction calorimetry data is illustrated by an example of using kinetics to design an industrial process. 25. A. Kossoy, Effect of thermal inertia-induced distortions of DSC data on the correctness of the kinetics evaluated. J. of Thermal Analysis and Calorimetry, V.143, p. 599–608 (2021) ABSTRACT Influence of certain experimental factors and methods of data processing on the correctness of the resultant kinetics is considered on the basis of dynamic model of differential scanning calorimeter, DSC. Firstly, essential effect of sample temperature deviation from linearity or from constant outer temperature due to heat accumulation in the sample on reaction proceeding is discussed. Disregard of this effect may and will be one of the reasons of obtaining the inadequate kinetics. The method for calculation of sample temperature deviation (reconstructing sample temperature) is proposed. Secondly, influence of a DSC curve distortion due to thermal inertia of the cell on the results of kinetic analysis is considered. The method for correction of this distortion (data deconvolution) is presented which can be safely used for processing of data of kinetic experiments. Special attention is paid to the importance of data deconvolution when the kinetic experiment is carried out under isoperibolic conditions. Finally the combined effect of temperature deviation and inertia-induced distortions is considered. At first all the matters are discussed on the basis of simulated data. This allows elimination of various uncertainties of real experiment and influence of unknown kinetic model, and gives vivid illustration of the subject. Then the effects are demonstrated on the basis of real experimental data. Both the model-free and model-based kinetics evaluation methods were used. The simple method is proposed which allows vivid cross-verification of both these types of kinetics. 24. A. Kossoy, An in-depth analysis of some methodical aspects of applying pseudo-adiabatic calorimetry. Thermochimica Acta, V.683, (2020) ABSTRACT The Accelerating Rate Calorimeter (ARC) - the first representative of the family of pseudo-adiabatic calorimeters - was designed for assessing reactive hazards of chemical processes. Over time the ARC and similar instruments have found much wider applications such as testing of reactive substances, study of reaction kinetics, design of the emergency relief systems. In all these cases precise knowledge about the current state of the sample is required. There are numerous publications regarding the methodical aspects of pseudo-adiabatic calorimetry but many problems are still waiting for the solution. The aim of this paper is to consider in detail the following serious problems:
The results presented here demonstrate that in many cases the basic assumptions used for interpretation of adiabatic data can be violated and due attention should be paid to the conditions of an experiment that allow it to remain within the applicability limits of the method. 23. A. Kossoy, Applying adiabatic calorimetry for study of energetic materials - is it possible? ABSTRACT Nowadays the use of adiabatic calorimetry for study of energetic materials becomes quite popular. It is not surprising considering that testing of these materials under extreme conditions can give valuable information. Unfortunately the correctness of such experiments is more than questionable. The origin is simple - energetic materials can generate huge amount of energy which doesn't allow applying sample masses that adiabatic calorimeters were designed for because temperature rise would be well above the permissible limits of the instruments. To avoid this problem one has to use very small samples - often dozens of milligrams instead of typical 3-6 grams (figures are given for the ARC instrument). With such small samples the effective thermal inertia of a calorimetric bomb becomes very big (more than 10-20) and temperature rise becomes acceptable. This method gives rise to serious problems:
These and some other related problems that impede application of adiabatic calorimetry are the subject of discussion.
22. A. Kossoy, Effect of peculiarities of DSC experiment on correctness of the kinetics created ABSTRACT Influence of certain experimental factors and methods of data processing on the correctness of the resultant kinetics is considered on the basis of dynamic model of DSC. Firstly, essential effect of sample temperature deviation from linearity or from constant outer temperature due to heat accumulation in the sample on reaction proceeding is discussed. Disregard of this effect may and will be one of the reasons of obtaining the inadequate kinetics. The method for calculation of temperature deviation (reconstruction of sample temperature) is proposed. Secondly, influence of a DSC curve distortion due to thermal inertia of the cell on the results of kinetic analysis is considered. The method for correction of this distortion (data deconvolution) is presented which can be safely used for processing of data of kinetic experiments. Special attention is paid to the importance of data deconvolution when the kinetic experiment is carried out under isoperibolic conditions. Finally the combined effect of temperature deviation and inertia-induced distortions is considered. It should be specifically emphasized that the matters under consideration are especially important when energetic materials are tested due to big heat release and rate of heat generation. Firstly all the matters are discussed on the basis of simulated data. This allows elimination of various uncertainties of real experiment and effect of unknown kinetic model and gives vivid illustration of the subject. Then the effects are demonstrated on the basis of real experimental data.
21. A. Kossoy, P. Grinberg, O. Antonov, Kinetics-based simulation of thermal explosion – some examples of experimental validation (full text is available here - Newsletter #13) ABSTRACT Advantages of kinetics-based simulation as the method for analysis of reactive hazards are acknowledged. It is the most universal and in many cases the only method applicable for hazard analysis. It specifically concerns simulation of thermal explosion when large-scale experiments are very expensive if at all possible. Nevertheless one can often meet certain disbelief in reliability of simulation results. No doubt there are serious reasons for such a disbelief because kinetics is evaluated from experimental data for very small samples (several milligrams in DSC experiments), the simplified physical models are used for explosion simulation, and simulation comes to quite faraway extrapolation. That is why every case when simulation results can be validated by experiment is important for strengthening of the confidence in simulation. As the examples of this kind are still rare it seemed to be of interest to collect and present some existing ones which is the aim of the presentation. Two examples are based on the authors' own results, the third one presents the results of the CISP® TSS software users.
ABSTRACT To evaluate thermal hazards of benzaldehyde oxime (BO), dynamic experiments were carried out by differential scanning calorimeter (DSC) to obtain thermodynamic parameters. A kinetic model was evaluated by fitting experimental curves. Finally, thermal behaviors under isothermal, adiabatic and conditions of limited intensity of heat exchange were simulated. The results indicate that BO decomposes rapidly in liquid phase, and releases a large amount of thermal energy. The reaction model of full autocatalysis has been created comprising two parallel stages: initiation stage of the n-order type, and the autocatalytic stage. Contribution of the two stages are also presented. Simulation results demonstrate low stability of BO in liquid phase, it ecomposes at low temperature right above melting and results in thermal explosion even for a small container. Estimation of time to maximum rate (TMR ad ) demonstrates the operational temperature should not be higher than 42 °C during production and usage. object. 19. A. Kossoy, P. Misharev, V. Belochvostov, Peculiarities of Calorimetric Data Processing for Kinetics Evaluation in Reaction Hazard Assessment, Presented at 53rd Annual Calorimetry Conference, Midland, Michigan, USA, August 9-14, 1998 PREFACE (of July 2016) This material has been presented at 53 Annual Calorimetric Conference almost 20 years ago. Unfortunately because of certain circumstances it has not been published. We say "unfortunately" as the proceedings of this conference can hardly be found freely whereas the matters discussed were important and could be useful for researchers involved in experiments based on DSC or adiabatic technique. As a matter of fact the main ideas of the report remain topical ones and deserve attention of wider audience. We are planning to publish the updated and more in-depth material in the near future, nevertheless it seemed reasonable to make this paper available right now. Authors are indebted to Tom Hofelich - some ideas presented here came to life after discussions with him and thanks to his valuable comments and suggestions. ABSTRACT Calorimetry, as applied to reaction hazard assessment, provides valuable data for evaluating reaction kinetics, which allows solution of miscellaneous problems of thermal safety by using mathematical modeling. A peculiarity of kinetic study for hazard assessment is that the resultant kinetics is intended mostly for runaway simulation, and the results are sensitive to small variations of the kinetic parameters. The reliability of kinetics, in its turn, depends on the elaboration of the methodology of calorimetric experiment. Therefore, some inaccuracies in methodology, often inessential in general practice, become crucial when reaction hazards are investigated. Although numerous publications are devoted to this subject, many problems are still unsolved or are not given due consideration. In particular, the influence of small errors in interpretation of calorimetric data on the resultant kinetics has not been practically discussed. The aim of this article is to bridge this gap by considering some peculiarities of DSC and adiabatic data. Concerning DSC data, two topics are discussed – sample overheating and heat inertia of a cell. Both result in rather small distortions of a calorimetric response but, if not taken into account, may lead to unsafe kinetics and incompatibility of data. Regarding adiabatic data, the problem of uncertainty of initial conditions at the onset temperature is considered. ABSTRACT The advantages of simulation-based method for the SADT determination are widely recognized. Nevertheless, active introduction of this method in practice requires careful verification. The project proposed by the Federal Institute for Materials Research and Testing, BAM, pursued this very object. Decomposition of 2,2’-azobis(isobutyronitrile), AIBN, has been studied by DSC in isothermal mode and series of large-scale This paper represents the results achieved by “ChemIinform” Ltd. Firstly the formal kinetic model has been created that provided appropriate fit of calorimetric data. Then this model was used for simulation of the conditions of H.1 and H.4 experiments. The results demonstrate good correspondence with experimental data. The materials presented show the potential of the simulation-based method as very useful addition to the methods recommended by international regulations. The CISP TSS software was used for implementing all the steps of the study. 17. A. Kossoy. Late response to an old article or some comments on how to implement comparative analysis ABSTRACT About 13 years ago the article “Comparison of several computational procedures for evaluating the kinetics of thermally stimulated condensed phase reactions” published in Chemometrics and Intelligent Laboratory Systems compared model-free method and model-based method presented by ForK - one of the CISP- developed programs for kinetics evaluation – which resulted in unfavorable conclusions with regard to ForK capabilities. We found it important to analyze the origins of such an unfavorable opinion of the authors, to show that the comparison had been implemented without due attention to all the details, and to give the results of more objective and accurate comparison. We believe that this paper may be of general interest. ABSTRACT The paper represents some results of comparative analysis of the methods used for processing and interpreting data of adiabatic calorimetry as well as applying it to practical situations. Specifically two approaches are compared – approximate method based on evaluation of simplified kinetics and a more comprehensive, simulation-based method that utilizes the evaluation of more detailed kinetic models. The analysis is focused on two important types of data processing – correction of experimental results on thermal inertia (phi-factor correction) and estimation of adiabatic time to maximum rate (TMR). The most widely cited method for phi-factor correction is considered and its improvement is proposed to enable more precise prediction of the adiabatic time scale. A procedure for phi-factor correction of pressure response is also proposed. The limitations of this enhanced Fisher’s method are discussed by comparison with simulation-based method. All the illustrative materials are based on real examples. As an example of application, the simplified method will be used to predict TMR and its limitations will be discussed. ABSTRACT Safety of chemical processes and plants is a matter of high priority. The design of an inherently safer process is one of very beneficial ways of achieving this goal. The paper describes the method of designing an inherently safer process for a chosen set of equipment and materials involved by applying non-linear optimization. The optimization is aimed at finding an operational mode, which guarantees safety of the process under normal conditions and provides maximal attainable safety in case of one typical accident scenario – cooling failure. Discussion covers problem statement, choice of the optimization criteria, appropriate methods for defining control variables. An important practical challenge is stability analysis of the optimized process mode with respect to permissible deviations of control parameters and variables from the estimated values. The original method for the stability analysis of a non-stationary process is proposed. It comprises simplified preliminary evaluation method followed by the more detailed numerical optimization-based analysis. Several examples illustrate application of the methods proposed. Abstract The thermal behavior of sample cells (bombs) of the ARC and VSP adiabatic calorimeters has been investigated by applying mathematical simulation. Influence of temperature gradient in a calorimetric bomb on the inaccuracy of kinetic parameters evaluated from adiabatic data has been analyzed. Then possible errors in kinetics-based predictions caused by the inaccuracy of kinetic parameters were identified by the example of two important hazard indicators – adiabatic time to maximum rate, TMR, and the self-accelerating decomposition temperature, SADT. A new control method for maintaining sample adiabaticity is proposed that provides obtaining the most correct experimental data suitable for creation of reliable kinetics. Abstract A hierarchical set of kinetic models were proposed and discussed for simulation of autocatalytic decomposition of cumene hydroperoxide (CHP) in cumene at low temperatures. The hierarchy leads from a formal model of full autocatalysis, which is based on conversion degree as a state variable, through a two-stage autocatalytic concentration-based model to a meticulous multi-stage model of the reaction. By the ForK (Formal Kinetics) and DesK (Descriptive Kinetics) software, developed by ChemInform Saint Petersburg (CISP) Ltd., the related kinetic parameters and their significance have also been estimated and elucidated. Through this best-fit approach, it is possible to formulate a systematic methodology on the kinetic studies for thermal decomposition of typical organic peroxides with autocatalytic nature, specifically at low temperature ranges. ABSTRACT The assessment, control and mitigation of reaction hazards is primarily based on the use of kinetic models. These kinetic models are used for the assessment of reaction hazards, the operation and control of the reactor, the design of emergency relief systems, and estimation of the consequences of a reaction runaway, to name a few. The validity of these assessments depends highly on the validity of the kinetic model employed. Several steps are required to identify a suitable kinetic model. This includes:
This paper discusses each of these steps in detail and identifies problems associated with each step. Several practical examples are used to demonstrate these problems. The results show that: 1) the results are sensitive to a number of assumptions, 2) mistakes may originate from misinterpretation of the thermal data, and 3) computational methods do exist to provide suitable kinetic models for hazard assessment. The analysis employed assumes a batch reaction system, since most of the kinetic data available is derived from batch calorimetric equipment. ABSTRACT The self-accelerating decomposition temperature (SADT) is an important parameter that characterizes thermal safety at transport of self-reactive substances. A great many articles were published focusing on various methodological aspects of SADT determination. Nevertheless there remain several serious problems that require further analysis and solution. Some of them are considered in the paper. Firstly four methods suggested by the United Nations “Recommendations on the Transport of Dangerous Goods” (TDG) are surveyed in order to reveal their features and limitations. The inconsistency between two definitions of SADT is discussed afterwards. One definition is the basis for the US SADT test and the heat accumulation storage test (Dewar test), another one is used when the Adiabatic storage test or the Isothermal storage test are applied. It is shown that this inconsistency may result in getting different and, in some cases, unsafe estimates of SADT. Then the applicability of the Dewar test for determination of SADT for solids is considered. It is shown that this test can be restrictedly applied for solids provided that the appropriate scale-up procedure is available. The advanced method based on the theory of regular cooling mode is proposed, which ensures more reliable results of the Dewar test application. The last part of the paper demonstrates how the kinetics-based simulation method helps in evaluation of SADT in those complex but practical cases (in particular, stack of packagings) when neither of the methods recommended by TDG can be used. ABSTRACT Historically, methyl ethyl ketone peroxide (MEKPO), a universal hardener in the rubber industries, has caused many serious explosions and fires in Taiwan, Japan, Korea and China. This study used certain thermal analytical methods to thoroughly explore both why MEKPO resulted in these accidents and what happened during the upset conditions. Potential process contaminants, such as H2SO4, KOH and Fe2O3, were deliberately selected to mix with MEKPO in various concentrations. Differential scanning calorimetry (DSC) was employed to calculate the thermokinetic parameters. Kinetics evaluation was also implemented by means of the methods and software developed by ChemInform St. Petersburg, Ltd. The results indicate that MEKPO was highly hazardous as mixed with any of the above-mentioned contaminants. The hazard of fires and explosions could be effectively controlled to a lesser extent only if safety parameters and thermokinetic parameters are properly imbedded into the manufacture processes. ABSTRACT Reactive hazards remain a significant safety challenge in the chemical industry despite continual attention devoted to this problem. The application of various criteria, which are recommended by the guidelines for assessment of reactive hazards, often causes unsafe results to be obtained. The main origins of such failures are as follows:
Four well-known hazard indicators – time to certain conversion limit, TCL, adiabatic time to maximum rate, TMR, adiabatic temperature rise, and NFPA reactivity rating number, Nr - are analyzed in the paper. It was ascertained that they could be safely used for preliminary assessment of reactive hazards provided that:
The applicability limits for every indicator were determined and the advanced kinetics-based simulation approach, which allows reliable determination of the indicators, is proposed. The technique of applying this approach is illustrated by two practical examples. 8. Kossoy A., Sheinman I., Evaluating thermal explosion hazard by using kinetics-based simulation approach, Process Safety and Envir. Protection. Trans IchemE, V. 82, Issue B6, November 2004, p.421-430. (B6 Special Issue: Risk Management); A. Benin, A. Kossoy, I. Sheinman and P.Grinberg. Evaluating Thermal Explosion Hazard of Self-Reactive Substances by Using Kinetics-Based Simulation Approach.// International Journal of Self-Propagating High-Temperature Synthesis.- 2006, V.15, N 4, P. 297 - 307. abstract Analysis of possible development of runaway at production, storage and use of a chemical product, and subsequent choice of measures that can prevent an accident or mitigate its consequences is one of the main tasks of reaction hazards assessment. A kinetic model evaluated from calorimetric data gives the reliable basis for implementing the analysis by means of numerical simulation. The purpose of this paper is to discuss some features of the approach as applied to such typical problems as determination of critical conditions of thermal explosion and the SADT for solid and liquid reactive chemicals. Firstly the brief survey of some popular simplified theories is discussed to reveal their main limitations. Secondly the mathematical models of thermal explosion in solid and liquid reacting systems are presented followed by a basic sketch of the numerical methods chosen for solving the problems. Finally the practical usefulness of the kinetics-based simulation approach for analyzing influence of various factors on explosion development is illustrated with several examples. The discussed models and methods were embodied in the ThermEx and ConvEx program packages developed by CISP. All the presented results have been obtained by means of this software. ABSTRACT To resolve various problems in creating a process, or conducting stability analysis, and\or yazard assessment, one needs to know the reactivity of a chemical system. The National Fire Protection Association (NFPA) requires use of a reactivity rating number (RRN) to describe such reactivity potentials as thermal stability, interaction with water, and gas generation. For assessing thermal stability of a substance, NFPA has recently a new quantitative approach based on the idea of "Instantaneous Power Density”. Though it has many advantages compared to the largely qualitative previous approach, the new method has one serious drawback - it doesn’t take into account the peculiarities of such complex cases as self-accelerating or multi-stage reactions. This, in turn, can lead to a less-than-safe, or unsafe, design. In this paper, we propose a method to generalize the concept of instantaneous power density by considering the maximal power density as the quantitative measure of the reactivity, allowing one to take proper account of kinetics complexity. We also briefly discuss the ReRank software which was developed to assess reactivity ratings in general, and, specifically, to calculate reactivity rating numbers. 6. Misharev P., Kossoy A., Benin A. Methodology and software for numerical simulation of thermal explosion. Process Safety and Envir. Protection. Trans IChemE, v.74, part B, February (1996)17. ABSTRACT The importance of the computer simulation for the prediction of a thermal explosion in reacting substances is beyond any doubt nowdays. Special software for solving this problem has been developed. This article describes comprehensively the THERMAL EXPLOSION program that provides simulation of explosion for systems with conductive heat transfer and forms a part of this software. Mathematical formulation, numerical method, several examples are discussed. ABSTRACT This paper is dedicated to the problem of the adequacy of the kinetics evaluation methods used in calorimetric investigation of reaction kinetics. The problem is especially important for adiabatic calorimetry because in this case application of usual methods may lead to obtaining non-correct kinetic models and, hence, to serious mistakes in hazard assessment of runaway reactions. The essence of the problem is being considered by method of mathematical simulation. The basic features and advantages of the appropriate method are discussed on the basis of real experimental data processing for kinetics evaluation. ABSTRACT The method of mathematical simulation was used for analysis of some methodological problems related with adiabatic calorimetry application: correctness of the procedure of initial temperature determination, influence of thermal inertia on temperature distribution in a reacting system, features of data interpretation in the case of a complicated reaction mechanism and some others. The efficiency of the approach based on the use of mathematical simulation and appropriate software has been illustrated by several examples. ABSTRACT Ensuring thermal safety of chemical processes is an important practical problem. Thermal safety means the processes' safety from the view point of possible development of thermal explosion caused by heat evolving during the chemical processes. The computerized system developed for solving this complicated problem is described as based on the complex of thermoanalytical and calorimetric devices of "SETARAM". Structure, purpose and possibilities of the system are considered. Methodological questions of kinetic experiments, kinetic analysis, thermal explosion simulation and organization of software are also discussed. ABSTRACT Correctness of kinetic experiment is an essential condition for obtaining of reliable results in kinetic investigation. Methods for provision and testing of thermo-physical and concentrational correctness are discussed in the present article. Problems connected with non-isothermal mode of real thermoanalytical experiment caused by programming as well as by heat release in the sample are considered. Application analysis of combined partial-linear heating laws in kinetic investigations is given. Results of correctness analysis are presented in relation to heat flux calorimeters "SETARAM". ABSTRACT Kinetic research with employment of thermal analysis methods comprises a complicated multi-stage procedure. Its successive implementation is impossible without the automation of all the stages with regard to their interconnections. Development of the automated system of kinetic researches (ASKR) in thermal analysis is the solution of this problem. ASKR is described as based on the set of thermoanalytical devices of "SETARAM" firm. The system allows shortening of time of a study and provides high quality and reliability of the results. Structure, purpose and possibilities of ASKR are considered. Methodological questions of kinetic experiments and kinetic data analysis, organization of software are also discussed. |
