Real-time kinetic hard-modelling for the optimisation of reaction conditions and the detection of process upset in semi-batch reactors

Julien Léo Billeter, Konrad Hungerbühler
2008
Conference paper

Abstract

Process Analytical Technology (PAT) has greatly evolved in the last decades due to the development of multivariate online sensors that are able to monitor the properties of industrial processes in real time [1, 2]. The online monitoring of product quality and the detection of process upsets are important for the pharmaceutical and fine chemical industry in order to maintain their product specifications and their commitments regarding safety, health and environment. Most frequent sources of deviations from normal operating conditions in semi-batch processes are due to slightly imprecise initial conditions or impurities in the initial reactants causing unexpected side reactions [3]. Online monitoring of industrial processes usually relies on calibration methods, such Principal Component Regression (PCR), Partial Least Squares (PLS) or Black Box models (e.g. Neural Networks) [4, 5]. A drawback of these calibration methods is their poor behaviour regarding extrapolation, requiring a constant effort for the operator to maintain the calibration conditions. Kinetic modelling techniques [6] do not suffer from this drawback as they are based on first principal models and can also be adapted for the monitoring of highly fluctuating processes, e.g. semi-batch processes. In this contribution, we propose a method for the online monitoring of semi-batch processes based on kinetic hard-modelling. The proposed method assumes that the kinetic model and the associated rate constants have already been determined at an earlier stage in R&D. In a first phase, the algorithm corrects estimates for the initial concentrations from dosing a small amount of reagent and fitting the kinetic model to the measured signals, e.g. mid-IR, UV-vis or released/consumed heat. In a second phase, if no process upset is detected, the corrected initial concentrations are fed back into the kinetic model and the algorithm optimises the dosing rate of the reagent or the operating temperature by maximising a property of the process, e.g. yield, selectivity or conversion. When optimum operating conditions are found, the algorithm forces the reactor to work under these improved conditions and the process is continuously re-optimised to detect possible process upsets. The method will be discussed based on simulated and experimental data taken from our high performance small scale reaction calorimeter coupled to in-situ mid-IR and UV-vis ATR-spectroscopy [7]. [1] P. Gemperline, G. Puxty, M. Maeder, D. Walker, F. Tarczynski, M. Bosserman, Analytical Chemistry 76 (2004) 2575-2582. [2] J. Workman, M. Koch, D. Veltkamp, Analytical Chemistry 77 (2005) 3789-3806. [3] E.N.M. van Sprang, H.J. Ramaker, H.F.M. Boelens, J.A. Westerhuis, D. Whiteman, D. Baines, I. Weaver, Analyst 128 (2003) 98-102. [4] M. Spear, Chemical Processing 70 (2007) 20-26. [5] T.J. Thurston, R.G. Brereton, D.J. Foord, R.E.A. Escott, Journal of Chemometrics 17 (2003) 313-322. [6] M. Maeder, Y.M. Neuhold, Practical Data Analysis in Chemistry, Elsevier, Amsterdam NL, 2007. [7] F. Visentin, S.I. Gianoli, A. Zogg, O.M. Kut, K. Hungerbühler, Organic Process Research & Development 8 (2004) 725-737.

Official source

https://infoscience.epfl.ch/record/176226?ln=en

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related concepts

Related publications

Julien Léo Billeter, Konrad Hungerbühler
2008
Conference paper

Abstract

Official source

https://infoscience.epfl.ch/record/176226?ln=en

About this result

Related concepts (20)

Related concepts

Related publications (9)

Related publications

Process Analytical Chemistry/Technology has tremendously evolved in the last decades due to the development of multivariate online sensors that are able to monitor the properties of industrial processes in real time [1, 2]. Online monitoring of product quality and the detection of process upsets are important for the pharmaceutical and fine chemical industry in order to maintain product specifications and meet their commitments regarding safety, health and environment. Many methods exist to extract useful information from the vast amount of data produced by online sensors. Chemometric methods, such as Principal Component Regression (PCR) and Partial Least Squares (PLS) or Black Box modelling (e.g. Neural Networks) are commonly used during the monitoring of batch processes [3, 4]. However, for these data-driven methods, calibration conditions need to be maintained during the actual process and the calibration generally behaves poorly when extrapolated to different operating conditions. On the other hand, kinetic modelling techniques [5], based on first principal models, describing the kinetics of main and side products, do not encounter such drawbacks and can be adapted for the monitoring of highly fluctuating processes, e.g. under semi-batch conditions. During batch and semi-batch processes, deviations from standard operating conditions can have various origins. Most frequent sources of deviations are due to slightly imprecise initial conditions (e.g. initial concentrations) or impurities in the initial reactants causing unexpected side reactions [6]. In this contribution, we propose a method for the online monitoring of semi-batch processes based on a kinetic modelling approach in order to optimise operating conditions and reduce “batch to batch” deviations. To our knowledge, this option has not yet been considered in literature. The proposed method requires the kinetic model and the associated rate constants to be known, i.e. determined in an early phase of R&D. In the following, the different steps of the algorithm, currently implemented into Matlab, are outlined. The algorithm assumes a first small amount of reagent to be dosed into the reaction mixture inside the reactor. Corrected initial concentrations are then determined by fitting the kinetic model to measurements, such as UV-vis, IR or heat power, using the Newton-Gauss-Levenberg/Marquardt (NGL/M) optimiser. If the optimiser fails the operator has the option to dose more reagent. Possible failure can be due to an early process upset, or to the fact that too little reagent was dosed in order to follow the kinetics reliably. The corrected initial concentrations are then fed back into the kinetic model and the algorithm optimises the flow rate for the dosed reagent or the operating temperature in order to maximise under constraints user-defined properties of the process, such as yield, selectivity or conversion. For this constrained optimisation, nonlinear programming (NLP) is employed (Matlab’s fmincon function). As soon as optimum operating conditions are obtained by the algorithm, the reactor will automatically run at these improved settings. As an option, flow rate and temperature can continuously be re-optimised to adapt to possible fluctuations in operating conditions. During the whole procedure, the algorithm also tests for possible process upset. If such an incident is detected the operator is asked to take appropriate action, for example a reactor shut down. The algorithm will be demonstrated using simulated data from mid-IR and UV-vis spectroscopy as well as from calorimetry. [1] P. Gemperline, G. Puxty, M. Maeder, D. Walker, F. Tarczynski, M. Bosserman, Analytical Chemistry 76 (2004) 2575-2582. [2] J. Workman, M. Koch, D. Veltkamp, Analytical Chemistry 77 (2005) 3789-3806. [3] M. Spear, Chemical Processing 70 (2007) 20-26. [4] T.J. Thurston, R.G. Brereton, D.J. Foord, R.E.A. Escott, Journal of Chemometrics 17 (2003) 313-322. [5] M. Maeder, Y.M. Neuhold, Practical Data Analysis in Chemistry, Elsevier, Amsterdam NL, 2007. [6] E.N.M. van Sprang, H.J. Ramaker, H.F.M. Boelens, J.A. Westerhuis, D. Whiteman, D. Baines, I. Weaver, Analyst 128 (2003) 98-102.

2008

Heat Management for Process Intensification of Fast Exothermic Reactions in Microstructured Reactors

Julien Haber

Nowadays, the production in fine chemical and pharmaceutical industry is mostly carried out in large scale batch reactors having typically dimensions of a few meters to satisfy the demand of the market. Even though this technology has been widely used and developed for centuries, it is by far not optimal for every type of reaction. For example, when working with exothermic reactions, the produced heat can’t be always fully evacuated. To avoid run-aways, high amounts of solvents are used to increase the heat capacity of the mixture, or semi-batch mode with a slow addition of one of the reactants. In both cases, the space-time yield, i.e. mass of product produced per unit of time and per unit of volume, drastically diminishes. One of the main enabling technologies allowing process intensification are the microstructured devices, characterized by high heat and mass transport rates due to the small characteristic dimensions (< 1mm). Using this type of equipment, almost isothermal conditions can be achieved while carrying out fast exothermic reactions (with characteristic reaction times down to tr ≈ 10 s). Thereby, the target throughput is reached by numbering-up, i.e. parallel connection of several identical microreactors. For very fast exothermic reactions, especially for quasi-instantaneous reactions, dimensions smaller than 100 μm are needed to prevent the formation of unwanted hot spots. As such small dimensions are not suitable for industrial scale due to possible clogging and high pressure drops, other solutions are warranted. The aim of this thesis is to develop alternative microstructured reactors enabling quasi-instantaneous reactions to be carried out under intensified conditions while suppressing the large hot spots. The work is divided into two main parts: determination of suitable strategies for the microstructured reactor design via numerical simulations (Chapter 3) and the experimental validation of the best microstructured reactor concept (Chapter 4-6). Three strategies for enhanced temperature control within microstructured reactors for quasi-instantaneous reactions are taken for analysis using numerical simulation: 1) reduction of hot spot temperature by increased axial heat transfer in the reactor wall, 2) by injection of one reactant in multiple points along the reactor length and 3) by continuous injection of one reactant through a porous wall in a concentric reactor geometry. The multi-injection reactor (option 2) is the most effective design since with an optimized dosing with only 4 injection points the temperature rise is 5-fold smaller as compared to the adiabatic temperature rise. Furthermore, the key design requirements for an efficient multi-injection reactor are identified: 1) complete mixing after each injection and 2) evacuation of the produced heat before reaching the next injection point. To experimentally validate the simulation results, in the subsequent chapter, an experimental method to monitor temperature in microstructured reactors is developed (Chapter 4). To track axial temperature profiles quantitatively, a method based on non-intrusive infrared thermography is developed yielding a resolution of 100 points/mm2 and a precision of 1 °C. In the first validation experiments, the heat transfer coefficient determined in a micro heat exchanger (574 W/m2K) is in good agreement with prior estimations. While carrying out the hydrolysis of tetraethoxysilane as a fast model reaction, incomplete mixing of the reactants is detected via the temperature profile, and is ascribed to the high difference in density of the inlet flows. Applying the method of quantitative IR-thermography to a T-micromixer with circular cross section gives insight into the mixing phenomenon (Chapter 5). The latter is studied via the temperature profile of the reactions strongly controlled by mixing, i.e. dilution of sulfuric acid with water and cyclization of pseudoionone. The mass transfer coefficients determined are in the order of 0.1-9 1/s. It is shown that at high Fourier numbers Fo = tdiff/τ (mixing by shearing), the Damköhler number DaI= τ/tmix remains constant with respect to flow rate at the reactor outlet, as both, mixing time and residence time decrease proportionally with the latter parameter. To enhance the mixing performance, two approaches are applied: 1) the introduction of a carrier phase leading to travelling micro-batches with up to 4-fold faster mixing and 2) the structuring of the channel walls leading to the formation of vortices, and thus, to a substantially improved mixing efficiency. For efficient mixing in the multi-injection reactor (Chapter 6), two types of mixing structures, i.e. the tangential mixer and the herringbone mixer, are developed using low temperature co-fired ceramics, and compared using quantitative infrared thermography. The best mixing performance is obtained by the herringbone structure, providing efficient mixing in a large range of flow rates corresponding to Reynolds numbers Re = 20-130. Finally, a multi-injection reactor comprising three injection points and the herringbone microstructure is developed. Using the quasi-instantaneous and exothermic cyclisation of pseudoionone to α-ionone and β-ionone as model reaction, it is demonstrated that the temperature rise can be reduced 8-fold compared to the adiabatic temperature rise due to 1) the high volumetric heat transfer coefficient in the order of 4·106 W/(m3K), 2) the reduced overall transformation rate due to gradual mixing within the herringbone structure and 3) the injection of pseudoionone at three injection points. Yields of α-ionone and β-ionone above 98 % are achieved at a residence time of 3.7 s while efficiently avoiding the unwanted consecutive polymerization in a temperature range of 30-60 °C. Compared to the conventional semi-batch process, where such high yields can only be attained at temperatures below 10 °C, a 500-fold increased space-time- yield is achieved. In addition to the intensification of the process, the required mass of solvent is halved while maintaining good temperature control, rendering the overall process safe.

EPFL2013

Small-scale reactor for data oriented process development

Julien Léo Billeter, Konrad Hungerbühler

During the early stages of research and development chemical industry needs flexible and versatile tools to investigate chemical reaction systems. An important part of the optimisation of a process considering economic factors, risk analysis and environmental impacts is the determination of a reaction mechanism and its associated parameters (i.e. activation energies, rates and heat of reactions). We present here a new fully automated small-scale reaction calorimeter combining a power-compensation heater and a thermoelectrically regulated metal surrounding. This dual temperature control makes the reactor highly suitable for fast and exothermic reactions and eliminates the need for time-consuming calibration of heat transfer coefficients. With a working volume from 25 to 45ml the device is particularly suited for the fine and pharmaceutical chemical industries where only small amounts of test substances are available. An integrated ATR-IR probe coupled to an FT-IR spectrometer allows the investigation of complex reaction mechanisms. Moreover, the new reactor design allows the simultaneous use of various additional in-situ analytics such as UV-Vis, gas intake/uptake and particle size analysis. The performance of the new reaction calorimeter has already been successfully demonstrated based on several reactions [1, 2, 3]. Different analytical techniques may point to different optimum reaction parameters. Algorithms and mathematical tools, such as multi-objective multivariate kinetic modelling, become an important research area in data oriented process development. Only a short introduction into these methods will be given here. [1] Zogg, A., Fischer, U., & Hungerbühler, K. (2003). A new small-scale reaction calorimeter that combines the principles of power compensation and heat balance. Industrial & Engineering Chemistry Research, 42, 767-776. [2] Visentin, F., Gianoli, S. I., Zogg, A., Kut, O. M., & Hungerbühler, K. (2004). Pressure-resistant smallscale reaction calorimeter that combines the principles of power compensation and heat balance (CRC.v4). Organic Process Research & Development, 8(5), 725-737. [3] Visentin, F., Puxty, G., Kut, O. M., & Hungerbühler, K. (2006). Study of the hydrogenation of selected nitro compounds by simultaneous measurements of calorimetric, FT-IR, and gas-uptake signals. Industrial & Engineering Chemistry Research, 45(13), 4544-4553.

2007