Energy efficient, intensified spray drying
A code, based on the generalised Lorenz-Mie theory merged with Huygens-Fresnel theory, has been developed. The code rigorously computes the image of homogeneous spherical particles, taking into account: the particle diameter, the particle complex refractive index, the polarisation and shape of the incident beam(s), the particle location inside the beam(s), the imaging optics (focal length and diameter), the detector location. For shadow Doppler technique, the code has been used to quantify the deep and width of the shadow recorded by the detector with as parameters: the beams polarisation and intersection angle, the particle size (0.1, 1, 5, 10, 20, 40 and 80 microns) and the particle locations (longitudinal and transverse) in the optical probe. From such data, the size measurement accuracy is determined according with the signal processing parameter adjustments.
The developed Shadow Doppler Velocimeter's ability to allow measurements in an industrial spray dryer is based on the design of a more flexible and compact system. The use of fibre optics and the design of a compact measuring probe able to be inserted into the flow enable local measurements within the spray dryer. The SDV allows the optimisation of flows in industrial equipment. The availability of data on the accuracy of the instrument, from Rouen, as well as the availability of an industrial flow to test the instrument, from Bayer, enabled the development of the SDV. The use of an SDV for research purposes enables Imperial College to continue providing cutting-edge research to the industry. A probe for Shadow Doppler Velocimeter (SDV) measurements inside an industrial spray dryer has been developed. The Transmitting optics is located outside the spray dryer and the laser beams are transmitted to the probe with the use of fibre optics. The resulting image and scattered light are collected within the probe and transmitted by lenses down the length of the probe to the receiving optics. The receiving optics is mounted in an aligned position at the rear of the measurement probe outside the spray dryer. All electronics involved in collecting and processing data are located outside the spray dryer. Such equipment may be used to measure the size distribution, velocities and other related data of particles in large-scale industrial flows.
The CFD code that was already in use at the Laboratory of Aerodynamics of NTUA has been extended and adapted for the CFD simulation of the spray drying process. A droplet agglomeration and a slurry droplet evaporation model have been validated against experimental data and introduced to the NTUA CFD code. Furthermore, these models could be introduced in commercial CFD codes, because they have been compared with commercial CFD codes providing direct conclusions regarding their reliability. A new Index number (HDIN) has been incorporated and introduced into the NTUA CFD code. It seems to be a reliable parameter in giving concrete conclusions regarding optimised spray dryer designs. The NTUA CFD code can be applied to other industrial applications, which have physical modelling similarities with spray drying and can be used to form other collaborations with industry as well as a marketable design tool. The computer code includes the numerical solution of the Navier Stokes equations using local grid refinement for improved accuracy in regions of flow complexity. Turbulence is modelled using the standard k-e model while heat transfer is also taken into account and gravitational effects due to density differences are included. The spray injection is modelled using a Lagrangian approach for liquid or solid particles with mechanisms for two way coupling between the flow and the particulate phase, heat transfer between the two phases, droplet evaporation, collision-coalescence and the ability for calculating 'slurry' (with solid burdens) particles. The code has been extensively validated against experimental measurements and comparisons of the results with those of commercial CFD codes have led to further improvement of the modelling. Moreover, a very significant parameter in assessing spray dryer performance (HDIN) has been introduced and tested.
The new computer code is able to simulate droplet distributions within a flow, using an Eulerian formulation. The code considers the droplet size and density distribution as a continuous variable within the flow, and calculates these distributions as the flow develops. Droplet agglomeration and coalescence is accounted for, by the use of a statistical probabilistic calculation of droplet/droplet impacts. The module has been demonstrated within a Langrangian simulation, and compared with a full Lagrangian calculation. The module has been implemented in an in-house Eulerian code. In this code, the much reduced computational time used means that the CFD can be used very effectively as input to a CFD based design procedure, in which a quick answer can be obtained to investigate the result of different design changes and philosophies. A computer simulation, validated against experimental data, shows that a newly developed model is able to give a rapid and accurate result for the droplet distribution within spray dryers. The new computer code is able to simulate droplet distributions within a flow, using an Eulerian formulation. The code considers the droplet size and density distribution as a continuous variable within the flow and calculates these distributions as the flow develops. Droplet agglomeration and coalescence is accounted for, by the use of a statistical probabilistic calculation of droplet/droplet impacts. The module has been demonstrated within a Lagrangian simulation, and compared with a full Lagrangian calculation. The module has been implemented in an in-house Eulerian code. In this code, the much-reduced computational time used means that the CFD can be used very effectively as a part of a design procedure. The ability to obtain quick answers will allow designers to investigate the result of different design changes and philosophies.