Mathematical modelling of flow-injection systems

doi:10.1016/0003-2670(94)00574-6

Analytica Chimica Acta

Volume 308, Issues 1–3, 9 June 1995, Pages 36-66

https://doi.org/10.1016/0003-2670(94)00574-6 Get rights and content

Abstract

Analysis of the great variety of flow-injection (FI) manifolds used in analytical practice nowadays has shown that most of them can be decomposed into two basic flow configurations, i.e., the single-line and the conjugated two-line system. The former system has one influent and one effluent stream through which it can contact with the environment. The conduit walls are totally impermeable. The most distinctive characteristic of a conjugated two-line system is the existence of a flow-through section with two separate streams (e.g., donor and acceptor) which exchange matter continuously along a common semipermeable interface (e.g., membrane). It can be concluded that two of the cornerstones in the modelling of FI manifolds are the successful mathematical description of the two basic flow systems mentioned above. Numerous mathematical models of FI systems employing ideas from different scientific areas (e.g., statistics, chemical engineering, artificial intelligence, chromatography) have been developed so far. It should be pointed out that the majority of them describe only single-line FI systems. A classification of all these models based on the main principles on which they are built, is proposed. The models have been compared with respect to their predictive power, the complexity of their mathematical treatment, and the requirements for computation time when applied to single-line and conjugated two-line FI systems. It is concluded that the axially dispersed plug flow model deserves special attention because it offers an acceptable compromise between the conflicting requirements for maximal possible mathematical simplicity and maximal possible precision. It can be used as the basis for an unified approach to the modelling of FI systems.

References (286)

C.C. Painton et al.
Anal. Chim. Acta
(1983)
J. Ruzicka et al.
Anal. Chim. Acta
(1986)
J. Ruzicka et al.
Anal. Chim. Acta
(1977)
S. Angelova et al.
Anal. Chim. Acta
(1983)
A. Fernández et al.
Anal. Chim. Acta
(1984)
M.A. Gómez-Nieto et al.
Talanta
(1985)
C.García de Maria et al.
Anal. Chim. Acta
(1989)
J. Toei
Talanta
(1988)
J. Wang et al.
Anal. Chim. Acta
(1984)
J.T. Vanderslice et al.
Talanta
(1981)

P.L. Kempster et al.

Talanta

(1989)

S.D. Kolev et al.

Anal. Chim. Acta

(1988)

S.R. Bysouth et al.

Anal. Chim. Acta

(1986)

A. Fernández et al.

Anal. Chim. Acta

(1987)

A. Ríos et al.

Talanta

(1987)

M. del Valle et al.

Anal. Chim. Acta

(1990)

G.D. Clark et al.

Talanta

(1991)

S.D. Kolev et al.

Anal. Chim. Acta

(1988)

Y. Li et al.

Anal. Chim. Acta

(1994)

Y. Narusawa et al.

Anal. Chim. Acta

(1994)

I.C. van Nugteren-Osinga et al.

Anal. Chim. Acta

(1988)

I.C. van Nugteren-Osinga et al.

Anal. Chim. Acta

(1989)

I.C. van Nugteren-Osinga et al.

Anal. Chim. Acta

(1990)

B.F. Johnson et al.

Talanta

(1992)

F. Dondi et al.

J. Chromatogr.

(1984)

F. Dondi et al.

J. Chromatogr.

(1984)

A.U. Ramsing et al.

Anal. Chim. Acta

(1981)

D. Chen et al.

Anal. Chim. Acta

(1990)

H.C. Smit et al.

Anal. Chim. Acta

(1988)

J.M. Reijn et al.

Anal. Chim. Acta

(1980)

J.H.M. van den Berg et al.

Anal. Chim. Acta

(1980)

S.H. Brooks et al.

Anal. Chim. Acta

(1990)

H. Poppe

Anal. Chim. Acta

(1980)

J.M. Reijn et al.

Anal. Chim. Acta

(1981)

J.M. Reijn et al.

Anal. Chim. Acta

(1983)

D. Betteridge et al.

Anal. Chim. Acta

(1984)

J. Ruzicka et al.

Anal. Chim. Acta

(1990)

P.D. Wentzell et al.

Anal. Chim. Acta

(1993)

M.J.E. Golay et al.

J. Chromatogr.

(1979)

J.G. Atwood et al.

J. Chromatogr.

(1981)

S.R. Bysouth et al.

Anal. Chim. Acta.

(1992)

C.C. Painton et al.

Anal. Chim. Acta

(1984)

J.T. Vanderslice et al.

Anal. Chim. Acta

(1986)

H. Wada et al.

Anal. Chim. Acta

(1986)

J. Ruzicka et al.

Flow Injection Analysis

(1981)

J. Ruzicka et al.

Flow Injection Analysis

(1988)

M. Valcŕcel et al.

Flow-Injection Analysis: Principles and Applications

(1987)

M. Valcárcel et al.

Automatic Methods of Analysis

(1988)

B. Karlberg et al.

Flow Injection Analysis: A Practical Guide

(1989)

Cited by (34)

Flow injection analysis: An approach via linear non-equilibrium thermodynamics
2018, Talanta
Citation Excerpt :
The presented material is the first attempt uniting these factors into the alone logical whole using LNET ideas in contrast to the classical equations of FIA theory based on mass and momentum conservation law. Different approaches to the numerical solutions of such equations of the FIA theory are discussed in the literature [3–6,16,17,33]. Expanding of the developed approach based on LNET will require additional preliminary in-depth-analysis of forces and fluxes arising in flow analysis systems.
A novel approach to flow injection analysis (FIA) was proposed based on the main principles of linear non-equilibrium thermodynamics (LNET). The basic principles of I. Prigogine theory for dissipative structures, internal entropy production rates, thermodynamic forces and fluxes arising in flow systems were shown to be applicable to FIA. The practical application of this novel FIA approach allowed the use of the extent of analytical reaction and the entropy production rates for flow system optimization, and in-depth understanding of the steady state. The FIA approach was also found to be a suitable technique for and characterizes its quality, explaining the peculiarities of short-term and long-term steady states in a FIA system and their role for reproducibility of practical measurements. The practical application o the FIA approach was found to support its theoretical principles and allow formulating an original manner to derive a basic equation in FIA theory.
A Gaussian Decomposition Method and its applications to the prediction of shale gas production
2018, Fuel
History matching is normally used to predict the gas production and guide refracturing. This can be achieved either through a mathematics-based approach such as decline curve fitting or through a physics-based approach such as reservoir simulation. When applied to the case of shale gas, both approaches are not working well. In this study, a Gaussian Decomposition Method (GDM), as an alternative approach, is developed and applied to the investigation of shale gas production. In this approach, an auto-compute program is developed and applied to a spectrum of scales from the core scale to the reservoir scale. Specific steps are as follows: (1) we use the experimental measurements to determine the initial gas content distribution; (2) we use the gas production history to decompose the evolving contributions of different gas components in a shale gas reservoir; and (3) we extend the history matching to predict the production of shale gas under similar extraction conditions. For the core scale, we use the automatically decomposed Gaussian components to illustrate the evolving contributions of different gas components including the free-phase gas in pores, the adsorbed gas and the diffused gas to the overall gas production. In the reservoir study, GDM is applied to the production data history matching and real-time prediction. Firstly, GMD is verified against a commercial software on daily, monthly and annual gas production rates. Then a group of daily and monthly field data are history matched by GDM. Finally, GDM is applied to predict the real-time gas production rate. Application results indicate: (a) the early gas production is mainly from big pores/fractures while the late production is from kerogen/clay components; (b) The period of gas production in the early stage is relatively short while the period in the late stage is long.
Dispersion in cylindrical channels on the laminar flow at low Fourier numbers
2015, Analytica Chimica Acta
A numerical solution of the uniform dispersion model in cylindrical channels at low Fourier numbers is presented. The presented setup allowed to eliminate experimental non-idealities interfering the laminar flow. Double-humped responses measured in a flow injection system with impedance detection agreed with those predicted by theory. Simulated concentration profiles as well as flow injection analysis (FIA) responses show the predictive and descriptive power of the numerical approach. A strong dependence of peak shapes on Fourier numbers, at its low values, makes the approach suitable for determination of diffusion coefficients. In the work, the uniform dispersion model coupled with the Levenberg–Marquardt method of optimization allowed to determine the salt diffusion coefficient for KCl, NaCl, KMnO₄ and CuSO₄ in water. The determined values (1.83, 1.53, 1.57 and 0.90) × 10⁻⁹ m² s⁻¹, respectively, agree well with the literature data.
Flow injection analysis simulations and diffusion coefficient determination by stochastic and deterministic optimization methods
2013, Analytica Chimica Acta
Citation Excerpt :
The obtained results confirm the applied assumptions: the negligible axial diffusion for the numerical dispersion model and the normal distribution and the concentration determination for the random walk model. Following other authors [4,11,18], differences between the experimental and the present results can be credited to: changes in channel geometry and cross section area at the detector, coiling and mixing effects giving rise to radial dispersion, non-ideality of solutions, mixing at the tube entrance creating the diffuse and curved interface rather than the sharp plane assumed in simulations. The present methods are general and can be used for the determination of diffusion coefficients in cases where analytical solutions are invalid, offering relatively low computational time and flexibility.
Stochastic and deterministic simulations of dispersion in cylindrical channels on the Poiseuille flow have been presented. The random walk (stochastic) and the uniform dispersion (deterministic) models have been used for computations of flow injection analysis responses. These methods coupled with the genetic algorithm and the Levenberg–Marquardt optimization methods, respectively, have been applied for determination of diffusion coefficients. The diffusion coefficients of fluorescein sodium, potassium hexacyanoferrate and potassium dichromate have been determined by means of the presented methods and FIA responses that are available in literature. The best-fit results agree with each other and with experimental data thus validating both presented approaches.
Mathematical modeling of a Nafion membrane based optode incorporating 1-(2′-pyridylazo)-2-naphthol under flow injection conditions
2010, Talanta
A general mathematical model of a flow-through optical chemical sensor prepared by the immobilization of 1-(2′-pyridylazo)-2-naphthol (PAN) into a commercial Nafion^® membrane was developed. The model takes into account the preparation of the optode membrane and in our opinion the most important chemical and physical processes involved in the generation of the analytical signal. The following model parameters were determined separately from the experimental verification of the model: aqueous diffusion coefficient of CuSO₄ – 8.75 × 10⁻¹⁰ m² s⁻¹; membrane self-diffusion coefficient of the Cu²⁺–PAN complex and Cu²⁺ – 1.87 × 10⁻¹⁶ and 6.00 × 10⁻¹⁵ m² s⁻¹, respectively; Nafion/water ion-exchange equilibrium constants for the Cu²⁺–PAN complex and Cu²⁺ – 109.2 and 3.65 × 10⁻³, respectively. Very good agreement was obtained between the experimental optode response and the model predictions thus showing that the model developed could be used successfully for the mathematical description and optimization of the PAN/Nafion optode as well as of other ion-exchange membrane based optodes having a similar response mechanism.
Mathematical modeling of dispersion in single interface flow analysis
2010, Analytica Chimica Acta
This work describes the optimization of the recently proposed fluid management methodology single interface flow analysis (SIFA) using chemometrics modelling. The influence of the most important physical and hydrodynamic flow parameters of SIFA systems on the axial dispersion coefficients estimated with the axially dispersed plug-flow model, was evaluated with chemometrics linear (multivariate linear regression) and non-linear (simple multiplicative and feed-forward neural networks) models. A D-optimal experimental design built with three reaction coil properties (length, configuration and internal diameter), flow-cell volume and flow rate, was adopted to generate the experimental data. Bromocresol green was used as the dye solution and the analytical signals were monitored by spectrophotometric detection at 614 nm. Results demonstrate that, independent of the model type, the statistically relevant parameters were the reactor coil length and internal diameter and the flow rate. The linear and non-linear multiplicative models were able to estimate the axial dispersion coefficient with validation r² = 0.86. Artificial neural networks estimated the same parameter with an increased accuracy (r² = 0.93), demonstrating that relations between the physical parameters and the dispersion phenomena are highly non-linear. The analysis of the response surface control charts simulated with the developed models allowed the interpretation of the relationships between the physical parameters and the dispersion processes.

View all citing articles on Scopus

View full text