An efficient numerical prediction of the crust onset of a drying colloidal drop
Introduction
The evaporation of colloidal drops is a subject of great interest with respect to its fundamental aspect involving diffusion with moving front and for its practical aspect in many technological fields such as spray drying [1], [2], [3], [4], [5], spray pyrolysis [6], fluidized bed drying [7], freeze-drying [8]. A colloid droplet evaporation process is generallly subdivided into three stages: crust formation, crust growth and drying of the resulting wet particle. During the first phase, which is of particular interest to the present study, the evaporation of the liquid, constituting the colloid, causes a shrinking of the drop, consequently particles accumulates in vicinity of the droplet interface. The first stage ends with the formation of a first crust layer, which occurs when the volume fraction at the interface reaches the close packing range. This critical instant is commonly known as the ”locking point”. The crust formation is the common step between all mathematical models describing this process [9], [10], [11], [12], and most of them considered its determination important in the understanding of the drying process, to predict, sometimes, the total drying time and the final size of the dried particle [13], [14], [15], [16].
The problem complexity includes a moving boundary whose location, in general, is a priori unknown, in addition of the particles concentration field whose knolewdge in particularely of interest in the vicinity of the interface. The interface movement is governed by a mass balance equation coupling the particule transport and the liquid evaporation. This equation, which contains all information with respect to the agglomeration of particles at the interface, is nonlinear and similar to the well known Stefan’s condition [17]. Consequently, analytical solutions are often difficult to implement.
The lack of exact analytical solutions for Stefan’s like problem, similar to this one, leads usually to semi-analytical or numerical methods. Semi-analytical methods are generally integral methods based on energy or mass balance [18], [19], [20], [21], [22], [23]. These inexpensive methods would provide explicit, though not very accurate, approximate analytical solutions. However, one should note such methods would work better in cartesian coordinates were several approaches exist [24]. More accurate solutions can be obtained through numerical methods associated with interface localization techniques. Two techniques, commonly used, are the variable space grid method (VSGM) and the boundary immobilization method (BIM) respectively developed by Murray and Landis [25] and Crank [26]. These two methods were used for solving Stefan’s problem [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], as well as for droplet evaporation [39]. Nevertheless, Sadoun et al. [40] showed that despite differences in the approach, both techniques lead to the same numerical scheme. The numerical procedure adopted hereafter is based on the boundary immobilization method (BIM).
Up to our knowledge, the use of finite volume methods, highly appreciated for the treatment of transfer problems due to its conservative properties, is not possible by the boundary immobilization. The methods, available for solving this type of problems, are the finite-difference [27], [28], [29], [33], [34], [41], [42] and finite element methods [35]. These methods have a low order accuracy, often linear and rarely quadratic namely through the use of a finite difference Keller Box scheme [32], [39], [43].
The spectral collocations method is a numerical method for solving a wide variety of PDEs in physics. Its high convergence rate allows achieving a great accuracy with a reduced number of nodes. Its performance makes it a valuable method to solve transfer phenomena problems including radiative heat transfer [44], [45], [46], [47], [48], [49], magnetohydrodynamics and computational fluid dynamics [50], [51]. In case of moving boundary problems, this method showed its ability to deal with Stefan’s problem in cartesian coordinates [52], [53] and radial coordinates [54], with high accuracy and convergence rate, compared to those obtained by finite difference methods.
A mathematical model, describing the drying of a colloidal drop, was proposed on the bases of heat and mass balances [11]. This model, with the objective of improving the fundamental understandings on the evaporation process of a colloidal drop, identified the governing key parameter as being the Péclet number . The authors discussed the existence of two drying phases of the drop. The first phase includes concentration of particles, at the interface, up to a maximum value, corresponding to the formation of the first skin of the crust, while the growth and consolidation of the crust define the second phase. The durations of these two phases will be used to identify the drying regimes of the drop. Nevertheless, the numerical resolution was done by classical finite difference method whose convergence and precision constraints, mentioned before, are particularly felt for fast drying, due to strong gradients at the interface.
In order to overcome the reduced accuracy of the finite difference method [11] and to include the moving boundary, a numerical approach has been developed on the basis of a spectral collocation method, for the determination of the onset locking point of a drying colloidal drop. Similarly to the suggestion of Mitchell et al. [39], on the complete evaporation time of pure liquid droplet, the resulting error on the locking point was used as an indicator of the overall behaviour on the numerical scheme for a large range of Péclet numbers.
Section snippets
Problem statement
The drying of a colloidal drop of initial radius consisting of a suspension of solid particles in a pure liquid, was considered. The drop was suspended in a medium, made of air and vapour of the drop’s liquid (Fig. 1) maintained, far from the drop, at a constant temperature and molar fraction .
Evaporation of the liquid constituting the colloid causes a decrease of the drop radius over time resulting in an accumulation of particles in the neighbourhood of the inner drop
Numerical methods:
The solution was approached by two numerical methods: a spectral collocation method and a finite difference method. In order to solve this problem, BIM technique was used. The Landau variable [58] was introduced into the inner region equations, setting the spatial variable from 0 to 1. Eq. (12) and its boundary conditions ((14), (15)) read as:
Substituting the drop radius (19) into ((23), (24), (25)) leads to:
Numerical results
The spectral collocation method provides a set of ODEs, for the evolution of the particle volume fraction field evaluated at CGL points, solved by a variable time step solver based on Adams predictor-corrector method for the non-stiff differential equations and the Gear method for the stiff equations.
To show the convergence properties of this numerical method, the system of equations was integrated with respect to time up to the crust onset: ( reaching its maximum value ). In the
Conclusion
The spectral collocation method was applied for the determination of a crust onset, of a drying isothermal colloidal droplet. The mathematical model describing this process was based on mass diffusion equation with moving boundary. The treatment of the latter by the BIM technique led to a system of a PDE and an ODE characterizing the evolution the concentration field within the drop and the droplet radius.
The spectral collocation method using Chebyshev-Gauss-Lobatto points was proposed to solve
CRediT authorship contribution statement
Zakaria Larbi: Methodology, Investigation, Software, Validation, Formal analysis, Writing - original draft. Nacer Sadoun: Supervision, Methodology, Writing - review & editing. El-khider Si-Ahmed: Supervision, Formal analysis, Writing - review & editing. Jack Legrand: Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (59)
- et al.
Spray drying modelling based on advanced droplet drying kinetics
Chemical Engineering and Processing: Process Intensification
(2010) - et al.
On the particle formation in spray drying process for bio-pharmaceutical applications: Interrogating a new model via computational fluid dynamics
International Journal of Heat and Mass Transfer
(2018) - et al.
Spray drying of pharmaceuticals and biopharmaceuticals: Critical parameters and experimental process optimization approaches
European Journal of Pharmaceutical Sciences
(2019) - et al.
Predictive cfd modeling of whey protein denaturation in skim milk spray drying powder production
Advanced Powder Technology
(2017) - et al.
Cfd simulations of droplet and particle agglomeration in an industrial counter-current spray dryer
Advanced Powder Technology
(2018) - et al.
Synthesis of spherical limn2o4 microparticles by a combination of spray pyrolysis and drying method
Powder Technology
(2008) - et al.
Modeling of semicontinuous fluid bed drying of pharmaceutical granules with respect to granule size
Journal of pharmaceutical sciences
(2019) - et al.
Modelling the morphological evolution of nanosuspension droplet in constant- rate drying stage
Chemical Engineering Science
(2011) - et al.
Mathematical modeling of the drying of a spherical colloidal drop
Colloids and Surfaces A: Physicochemical and Engineering Aspects
(2019) - et al.
Modeling and simulation of solid-containing droplet drying and different-structure particle formation
International Journal of Heat and Mass Transfer
(2020)
Drying behavior and locking point of single droplets containing functional oil
Advanced Powder Technology
Numerical simulation of evaporation and drying of a bi-component droplet
International Journal of Heat and Mass Transfer
A cubic heat balance integral method for one-dimensional melting of a finite thickness layer
International Journal of Heat and Mass Transfer
On the refined integral method for the one-phase stefan problem with time-dependent boundary conditions
Applied mathematical modelling
Optimizing the exponent in the heat balance and refined integral methods
International Communications in Heat and Mass Transfer
Optimising the heat balance integral method in spherical and cylindrical stefan problems
Applied Mathematics and Computation
A modified variable time step method for the one-dimensional stefan problem
Computer Methods in Applied Mechanics and Engineering
The numerical solution of one-phase classical stefan problem
Journal of computational and applied mathematics
Numerical schemes for one-dimensional stefan-like problems with a forcing term
Applied mathematics and computation
Finite-difference methods with increased accuracy and correct initialization for one-dimensional stefan problems
Applied Mathematics and Computation
Finite difference solution of one-dimensional stefan problem with periodic boundary conditions
International journal of heat and mass transfer
One-dimensional solidification of pure materials with a time periodically oscillating temperature boundary condition
Applied Mathematics and Computation
A one-phase stefan problem with size-dependent thermal conductivity
Applied Mathematical Modelling
An accurate numerical solution for the transient heating of an evaporating spherical droplet
Applied Mathematics and Computation
On the boundary immobilization and variable space grid methods for transient heat conduction problems with phase change: Discussion and refinement
Comptes Rendus Mécanique
On the numerical solution of a stefan problem with finite extinction time
Journal of Computational and Applied Mathematics
Analysis of radiative transfer in a one-dimensional nonlinear anisotropic scattering medium with space-dependent scattering coefficient using spectral collocation method
International Journal of Heat and Mass Transfer
Thermal radiation heat transfer in one-and two-dimensional enclosures using the spectral collocation method with full spectrum k-distribution model
International Journal of Heat and Mass Transfer
Spectral collocation method for transient thermal analysis of coupled conductive, convective and radiative heat transfer in the moving plate with temperature dependent properties and heat generation
International Journal of Heat and Mass Transfer
Cited by (3)
Microparticle encapsulation of a tuberculosis subunit vaccine candidate containing a nanoemulsion adjuvant via spray drying
2021, European Journal of Pharmaceutics and BiopharmaceuticsNumerical Modelling of Formation of Highly Ordered Structured Micro-and Nanoparticles – A Review
2022, KONA Powder and Particle JournalHeating and Evaporation of Multi-component Droplets
2022, Mathematical Engineering