processes
Article
Novel Technique for Coating of Fine Particles Using
Fluidized Bed and Aerosol Atomizer
Rongyi Zhang *, Torsten Hoffmann and Evangelos Tsotsas
Thermal Process Engineering, Otto von Guericke University, Universitäts Platz 2, 39106 Magdeburg, Germany;
torsten.hoffmann@ovgu.de (T.H.); evangelos.tsotsas@ovgu.de (E.T.)
* Correspondence: rongyi.zhang@ovgu.de
Received: 29 October 2020; Accepted: 19 November 2020; Published: 24 November 2020
Abstract: Fine particles are widely used in many industrial fields, and there are many techniques
applied for these particles, like electroplating, and chemical and physical vapor deposition. However,
in the food and pharmaceutical industries, most coating processes conducted with fluidized bed use
core particles with a diameter larger than 200 µm, otherwise agglomerates are formed. This study
contributes to the development of a new coating process for fine particles with diameters of around
50 µm. The innovation lies in the combined use of a Wurster fluidized bed and a novel aerosol atomizer.
The feasibility of the operation is based on the application of the aerosol atomizer, which generates
droplets smaller than 1 µm in diameter. A series of experiments with different coating solutions and
glass beads in a 150 mm fluidized bed fed with droplet aerosol supplied from the cone chamber
bottom is presented. The quality of the coating product is analyzed by scanning electron microscopy
and CAMSIZER® . In this way, the influence of different conditions and core material properties on
the product quality were determined. Experimental results showed the coating layer quality getting
worse as coating solution viscosity became lower, meanwhile moderate process temperature was
found to enhance coating layer formation and quality of that. It was also observed that lower aerosol
feed rates help improve the yield of the process.
Keywords: coating; aerosol generator; fluidized bed; fine particles
1. Introduction
The coating of particulate materials is an important process for various industries, including the
chemical, food and pharmaceutical industries [1]. Coating is frequently used for protective or functional
purpose [2,3]. Additionally, the visual properties of the particles can also be changed [4]. Therefore,
through coating processes, various properties can be achieved, such as stabilization of active substances,
enhanced mechanical properties, masking of taste and odor, improved optical appearance, and defined
drug release profile in the human body [5]. Among different particulate materials, fine particles have
come to be of major interest lately. Many industrial sectors are expected to find applications and take
advantage of the new functionalities and many desirable properties attributed to ultrafine particles,
e.g., their use as drug delivery carriers [6].
Application of the coating layer to the surface of a particle is a complex process [5]. Typically,
a suspension or solution is sprayed onto the particles, after that, every single particle is wrapped by
the liquid film. New solid layers are formed after the liquid in the film has evaporated, which results
in coating of the particles. Specifically, depending on the apparatus used and the coating material
conditions, the process may comprise the following steps: droplet formation, wetting, spreading,
evaporation, and drying [7].
Fluidization is one of the most important techniques in the particle formation process because it
possesses the advantages of high heat and mass transfer rates, good mixing and homogeneity. With the
Processes 2020, 8, 1525; doi:10.3390/pr8121525
www.mdpi.com/journal/processes
Processes 2020, 8, 1525
2 of 18
increasing progress of experimental methods and mathematical models in recent years, the spray
fluidized bed coating of particulate materials is becoming an established industrial process [8,9].
Despite this, fluidized bed coating without severe agglomeration is typically limited to particle sizes
larger than 100 µm [5]. On the one hand, the reason for this is that fine powders fluidize poorly
due to their strong interparticle cohesive forces, exhibiting channeling, lifting as a plug and forming
“rat holes” when aerated [10]. On the other hand, when using ordinary two-fluid nozzles, the spray is
composed of droplets with an average diameter of 40 µm or more [11], which can be considered to be
large droplets relative to the particle size. Hence, an over-wetted particle surface will result in strong
cohesive forces of liquid bridges that hold particles together after the particles collide with each other
during the drying process. These are converted to solid bridges upon drying, fixing the structure of
agglomerates that will not break in the subsequent steps [12].
To overcome these issues, conventional methods can be optimized by improving the particles or
modifying the fluidized bed. Thereby, powders can be granulated with larger particles in order
to enhance the fluidization, drying and coating process in the fluidized bed [13]. Otherwise,
fluidization can also be assisted by a variety of mechanisms such as vibration, mechanical agitation,
sound and magnetic force to improve the flowability of cohesive particles [6,14,15]. Such approaches are
effective in the formation of coating layers, but they also have the disadvantage of limited application
ranges. When granulating the fine particles with coarse ones, the high concentration of easily fluidized
material will finally influence the product purity. Additionally, most assistance techniques require
complicated modifications to the traditional fluidized bed configuration that might demand large
additional costs.
Recently, several researchers have tried to coat fine particles by using different apparatuses
for under various operating conditions. For instance, a Wurster fluidized bed was used by
Ichikawa et al. [16] along with 12 µm cornstarch as the core particles and composite of polymers
as the coating material. The corresponding product only contained 3 wt.% agglomerates. However,
there were still some ineluctable problems when using this method: for example, sufficiently prolonged
release properties could not be obtained. Watano et al. [6,17] tried the rotating fluidized bed using
fine cornstarch particles. In this case, the cohesive cornstarch particles behaved like particles that
could be well fluidized because the centrifugal forces on increased the apparent particle weight. This
led to cornstarch being successfully coated nearly without agglomeration. However, this method
required a complex apparatus and a significant amount of energy caused by the huge pressure drop
through the particles. Chen et al. [18,19] proposed a way in which core materials were pretreated by
deposition of nano-particles on the surface of fine particles. The addition of a nano-particle surface
coating reduces the cohesive behavior so that fine particles can fluidize in a conventional fluidized
bed and good quality coating product may be obtained. Nevertheless, the pretreatment process was
inefficient and time-consuming, which is a big drawback in large-scale industrial production. The
recent work of Hampel et al. [20] developed a new method that combined two particle formulation
processes: spray drying and coating. The idea was to spray fine particles to be coated together with
coating solution means of a two-fluid nozzle, aiming at the generation of droplets with only one
contained particle at the nozzle outlet. After evaporation of the water in such a droplet, a thin solid
film is built on the particle surface. This innovative, low-cost and easy-to-implement method could
avoid the occurrence of agglomeration to a certain extent, but partly coated or even uncoated products
appeared under certain process conditions.
As indicated by the previous experiments conducted in a fluidized bed, a difficulty lies in
successfully fluidizing the original fine particles. Previous research by Ichikawa et al. demonstrated
that the bottom-spray Wurster bed is an effective way of improving the fluidization behavior of
cohesive fine particles [16]. However, as mentioned earlier, large spray droplets lead to the formation
of agglomerates, so that individual coating of fine particles still be a challenge and this work is focus
on the topic.
Processes 2020, 8, 1525
3 of 18
A radical change from conventional spray to aerosol for coating process in fluidized bed is
Processes 2020, 8, x FOR PEER REVIEW
presented
in this work. The aerosol, generated by a recently developed method [21,22], was supplied 3 of 18
into the chamber of a Wurster fluidized bed from the bottom. It was recently shown that ultrathin
coating can be easily produced on large particles in a traditional fluidized bed by such an aerosol [23].
coating can be easily produced on large particles in a traditional fluidized bed by such an aerosol [23].
Here, the coating of fine particles was in focus. The working hypothesis was that aerosol with
Here, the coating of fine particles was in focus. The working hypothesis was that aerosol with diameters
diameters 10–100 times smaller than regular sprays would be more suitable for fine particles and
10–100 times smaller than regular sprays would be more suitable for fine particles and enable uniform
enable uniform coating layers and without the excessive agglomeration that traditional spray
coating layers and without the excessive agglomeration that traditional spray fluidized bed particle
fluidized bed particle coating systems show.
coating systems show.
In this work, investigations will be presented on the effect that operating conditions such as core
In this work, investigations will be presented on the effect that operating conditions such as
material size, air temperature, coating material concentration and the presence of the additive in
core material size, air temperature, coating material concentration and the presence of the additive in
coating solution have on efficacy of the presented process to produce coated fine particles. Finally,
coating solution have on efficacy of the presented process to produce coated fine particles. Finally,
the influence of operating conditions on the resulting product quality will be described and discussed;
the influence of operating conditions on the resulting product quality will be described and discussed;
the strengths and weaknesses of the novel aerosol coating process to its conventional counterpart will
the strengths and weaknesses of the novel aerosol coating process to its conventional counterpart will
be summarized.
be summarized.
2. Experimental
2. Experimental
2.1. Aerosol
Generation
2.1. Aerosol
Generation
A simple
and reliable
technique
for the generation
aerosolwas
droplets
was
recently developed
A simple
and reliable
technique
for the generation
of aerosolof
droplets
recently
developed
and
and
investigated
by
Mezhericher
[21,22].
Compared
to
traditional
liquid
atomization
technologies
investigated by Mezhericher [21,22]. Compared to traditional liquid atomization technologies [24],
[24], this
method
based on shear-driven
disintegration
gas jets
of thin
films
by
this method
is based
on isshear-driven
disintegration
by gas jets by
of thin
liquid
filmsliquid
formed
by formed
gas
on a liquid
The
main
part isofanthe
device
is antube
elastic
rubber
tube which is
bubblesgas
on bubbles
a liquid surface.
The surface.
main part
of the
device
elastic
rubber
which
is perforated
perforated
with
ostioles
along
the
tube
at
several
axial
positions.
The
tube
is
horizontally
placed and
with ostioles along the tube at several axial positions. The tube is horizontally placed and partially
partially
in with
a vessel
filledsolution:
with coating
theoflower
part is
of immersed
the tube is in
immersed
in
submerged
in asubmerged
vessel filled
coating
the solution:
lower part
the tube
the
the
liquid,
and
the
upper
portion
of
the
tube
is
exposed
to
the
air,
see
Figure
1a.
The
air
inlet
tube
is
liquid, and the upper portion of the tube is exposed to the air, see Figure 1a. The air inlet tube is
connected
to
compressed
air,
and
the
stable
compressed
air
is
discharged
through
the
perforated
connected to compressed air, and the stable compressed air is discharged through the perforated
ostioles.
many bubbles
small bubbles
are caused
the compressed
air through
the part
lower
ostioles.
Many small
which which
are caused
by the by
compressed
air through
the lower
ofpart
the of the
tube,
then
come
up
to
the
liquid
surface
and
create
a
foam
of
thin
spherical
liquid
films
near
the upper
tube, then come up to the liquid surface and create a foam of thin spherical liquid films near the
part
of
the
tube.
Soon
afterwards,
the
air
discharged
through
the
upper
ostioles
breaks
upper part of the tube. Soon afterwards, the air discharged through the upper ostioles breaks and and
disintegrates
these into
foams
into aerosol
droplets,
as schematically
illustrated
in Figure
1b. In
this study,
disintegrates
these foams
aerosol
droplets,
as schematically
illustrated
in Figure
1b. In this
study,
a
self-made
aerosol
generator
developed
based
on
this
principle
was
used
in
the
experiments.
a self-made aerosol generator developed based on this principle was used in the experiments.
(a)
(b)
Figure
1. (a) Simultaneous
production
of bubbles,
micro-sprays
fine
dropletsinina a crossFigure 1.
(a) Simultaneous
production
of bubbles,
gas gas
jets jets
andand
micro-sprays
of of
fine
droplets
cross-section
ofof
the
atomizing
tube
[25];
schematic
of of
novel
aerosol
generator.
section
the
atomizing
tube
[25](b)
; (b)
schematic
novel
aerosol
generator.
The central
part ofpart
the device
was a 220
polyvinyl
chloride
tube oftube
12 mm
The central
of the device
wasmm
a 220
mm polyvinyl
chloride
of 12outer
mm diameter.
outer diameter.
The 1.5The
mm1.5
walls
the tube
were
with 0.6
mm0.6
ostioles
along the
tube
at
mmofwalls
of the
tubeperforated
were perforated
with
mm ostioles
along
thecircumference
tube circumference
at
four axial
positions
(38
orifices
per
line,
144
in
total).
The
tube
was
horizontally
oriented
and
partially
four axial positions (38 orifices per line, 144 in total). The tube was horizontally oriented and partially
submerged
into a plastic
vessel filled
tap
water
coating
solution.
This vessel
dimensions
submerged
into a plastic
vesselwith
filled
with
tapor
water
or coating
solution.
This had
vessel
had dimensions
of 16.5 cm (diameter) × 18.5 cm (height) and a circular outlet of 4 cm at its upper horizontal wall. A
flexible PVC tube with a length of 165 cm connected the outlet of the container and the fluidized bed,
so that droplets and gas could flow out horizontally from the container.
Droplet size distributions produced by the aerosol generator were characterized separately by
means of a laser diffraction system (Particle Analyzer LA-960, Retsch Technologies, Haan, Germany)
Processes 2020, 8, 1525
4 of 18
of 16.5 cm (diameter) × 18.5 cm (height) and a circular outlet of 4 cm at its upper horizontal wall.
A flexible PVC tube with a length of 165 cm connected the outlet of the container and the fluidized bed,
so that droplets and gas could flow out horizontally from the container.
Processes
2020,
xxFOR
PEER
4 4ofof1818
Processes
2020,8,8,
FOR
PEERREVIEW
REVIEW
Droplet
size
distributions
produced by the aerosol generator were characterized separately
by
means of a laser diffraction system (Particle Analyzer LA-960, Retsch Technologies, Haan, Germany)
Measurement
results
obtained
for
different
coating
solutions
are
shown
ininFigure
2a.
ensure
Measurementresults
resultsobtained
obtainedfor
fordifferent
differentcoating
coatingsolutions
solutionsare
areshown
shownin
Figure2a.
2a.To
Toensure
ensureaaa
Measurement
Figure
To
constant
liquid
level,
solution
was
pumped
totothe
aerosol
atomizer
through
peristaltic
pump
from
constantliquid
liquidlevel,
level,solution
solutionwas
waspumped
pumpedto
theaerosol
aerosolatomizer
atomizerthrough
throughaaaperistaltic
peristalticpump
pumpfrom
from
constant
the
another
container.
Then,
the
flow
rate
of
the
droplets
was
calculated
by
measuring
the
amount
of
another
container.
Then,
the
flow
rate
of
the
droplets
was
calculated
by
measuring
the
amount
another container. Then, the flow rate of the droplets was calculated by measuring the amount of liquidof
liquid
left
in
the
storage
tank
and
the
corresponding
time
interval.
Figure
2b
indicates
the
flow
rate
liquid
leftstorage
in the storage
the corresponding
time interval.
Figure
2b indicates
flow
rate
left
in the
tank andtank
the and
corresponding
time interval.
Figure 2b
indicates
the flowthe
rate
of the
ofofthe
aerosol
generator
under
different
operating
conditions.
The
volumetric
size
distributions
of
the generator
aerosol generator
under different
conditions.
The volumetric
size distributions
aerosol
under different
operatingoperating
conditions.
The volumetric
size distributions
of Figure 2aof
Figure
2a
demonstrate
that
aerosol
droplet
diameters
range
from
0.1
μm
to
10
μm,
and
the
mean
Figure 2a demonstrate
aerosol
droplet
diameters
0.1 and
μm the
to 10
μm,droplet
and the
mean
demonstrate
that aerosol that
droplet
diameters
range
from 0.1range
µm tofrom
10 µm,
mean
size
of
droplet
size
ofofthe
aerosol
used
ininthe
experiments
isissmaller
than
11μm.
Figure
2b
shows
that
the
droplet
size
the
aerosol
used
the
experiments
smaller
than
μm.
Figure
2b
shows
that
the
the aerosol used in the experiments is smaller than 1 µm. Figure 2b shows that the liquid flow rate of
liquid
flow
rate
of
the
generated
aerosol
increases
with
atomizing
pressure.
liquid
flow rate
of theincreases
generated
aerosol
increases
with atomizing pressure.
the
generated
aerosol
with
atomizing
pressure.
(a)
(a)
(b)
(b)
Figure
2.2.(a)
(a)
Volumetric
droplet
size
distributions
aerosols
generated
55bar
Figure2.
(a)Volumetric
Volumetric
droplet
size
distributions
indifferent
different
aerosols
generated
baratomizing
atomizing
Figure
droplet
size
distributions
inin
different
aerosols
generated
at 5atat
bar
atomizing
air
air
pressure;
(b)
aerosol
flow
rate
vs.
atomizing
pressure
for
different
liquid
solutions.
air pressure;
(b) aerosol
rateatomizing
vs. atomizing
pressure
for different
solutions.
pressure;
(b) aerosol
flow flow
rate vs.
pressure
for different
liquidliquid
solutions.
2.2.
2.2.
2.2.Materials
Materials
2.2.1.
Core Material
2.2.1.
2.2.1.Core
CoreMaterial
Material
To
compare
the effects
of different particle
sizes on
the new
coating technology,
two different
To
Tocompare
comparethe
theeffects
effectsofofdifferent
differentparticle
particlesizes
sizeson
onthe
thenew
newcoating
coatingtechnology,
technology,two
twodifferent
different
sizes
of
glass
particles
(Cerablast
GmbH,
Löchgau,
Germany)
were
used
as
the
core
material
in the
sizes
of
glass
particles
(Cerablast
GmbH,
Löchgau,
Germany)
were
used
as
the
core
material
sizes of glass particles (Cerablast GmbH, Löchgau, Germany) were used as the core materialininthe
the
coating
experiments. Particle
Particle size distributions
distributions (PSD) were
were measured offline
by CAMSIZER-XT
coating
coatingexperiments.
experiments. Particlesize
size distributions(PSD)
(PSD) weremeasured
measuredoffline
offlineby
by CAMSIZER-XT
CAMSIZER-XT
(Retsch
Technologies, Germany).
Measured PSD
results for
each kind
of particles are
given inin
Figure 3.
(Retsch
(RetschTechnologies,
Technologies,Germany).
Germany).Measured
MeasuredPSD
PSDresults
resultsfor
foreach
eachkind
kindofofparticles
particlesare
aregiven
given inFigure
Figure
The
mean
diameter
was
63
µm
and
202
µm,
respectively.
In
both
cases,
the
density
of
the
core
particle
3.3.The
diameter
was
63
μm
and
202
μm,
respectively.
InInboth
cases,
the
density
ofofthe
core
Themean
mean
diameter
was
63
μm
and
202
μm,
respectively.
both
cases,
the
density
the
core
3 and sphericity is 0.94.
is
2500 kg/m
3 and sphericity is 0.94.
particle
is
2500
kg/m
particle is 2500 kg/m3 and sphericity is 0.94.
Figure 3. Initial normalized particle size distributions of the two groups of glass particles.
Figure
Figure3.3.Initial
Initialnormalized
normalizedparticle
particlesize
sizedistributions
distributionsofofthe
thetwo
twogroups
groupsofofglass
glassparticles.
particles.
2.2.2.
2.2.2.Coating
CoatingSolution
Solution
Sodium
Sodiumbenzoate
benzoatesolution
solutionwas
wasused
usedasasaacoating
coatingagent
agentininthese
theseexperiments.
experiments.Sodium
Sodiumbenzoate
benzoate
(NaB,
C
7H5NaO2, Trigon Chemie GmbH, Schlüchtern, Germany) is a water-soluble material which is
(NaB, C7H5NaO2, Trigon Chemie GmbH, Schlüchtern, Germany) is a water-soluble material which is
used widely as a preservative in the food industry, but also in academic coating research [26]. To
Processes 2020, 8, 1525
5 of 18
2.2.2. Coating Solution
Sodium benzoate solution was used as a coating agent in these experiments. Sodium benzoate
(NaB, C7 H5 NaO2 , Trigon Chemie GmbH, Schlüchtern, Germany) is a water-soluble material which is
Processes 2020, 8, x FOR PEER REVIEW
5 of 18
used widely as a preservative in the food industry, but also in academic coating research [26]. To analyze
the influence of viscosity of the coating solution, different amounts of hydroxy-propyl-methyl-cellulose
methyl-cellulose (HPMC, trade name Pharmacoat 606, from Shin-Etsu, Tokyo, Japan) were added
(HPMC, trade name Pharmacoat 606, from Shin-Etsu, Tokyo, Japan) were added into the solution.
into the solution. The coating solution density at 20 °C was measured by using a density measuring
The coating solution density at 20 ◦ C was measured by using a density measuring device DMA 58
device DMA 58 (Anton Paar GmbH, Graz, Austria). A rheometer MCR 302 (Anton Paar GmbH, Graz,
(Anton Paar GmbH, Graz, Austria). A rheometer MCR 302 (Anton Paar GmbH, Graz, Austria) was
Austria) was used to measure the dynamic viscosity of different coating solution with various NaB
used to measure the dynamic viscosity of different coating solution with various NaB content. Both of
content. Both of the devices were calibrated by using distilled water. The solution composition and
the devices were calibrated by using distilled water. The solution composition and measured properties
measured properties are summarized in Table 1.
are summarized in Table 1.
Table1.1. Material
Material properties
propertiesof
ofcoating
coatingsolution.
solution.
Table
Solution Composition 1 Density, ρcs(kg/m3)3 Viscosity, ηcs(Pa·S)
Viscosity, ηcs (Pa·S)
Solution Composition 1
Density, ρcs (kg/m )
30% NaB
1128.14
0.0039
1128.14
0.0039
30%30%
NaBNaB
+ 0.5% HPMC
1129.98
0.0062
30%30%
NaBNaB
+ 0.5%
HPMC
1129.98
0.0062
+ 1.0% HPMC
1130.01
0.0111
30%30%
NaBNaB
+ 1.0%
HPMC
1130.01
0.0111
+ 1.5%
HPMC
1131.90
0.0618
30% NaB + 1.5% HPMC
1131.90
0.0618
1 The composition of solution is based on mass fraction (wt%).
1
The composition of solution is based on mass fraction (wt%).
2.3. Fluidized Bed Coating Process
2.3. Fluidized Bed Coating Process
The coating process of fine particles was conducted in a laboratory-scale fluidized bed (GPCG 1
The coating process of fine particles was conducted in a laboratory-scale fluidized bed (GPCG 1 in
in Wurster version, Glatt GmbH, Weimar, Germany). The lab-scale set up use the distributer plate
Wurster version, Glatt GmbH, Weimar, Germany). The lab-scale set up use the distributer plate which
which has the diameter of 150 mm and a 550 mm cone fluidized chamber. This setup was modified
has the diameter of 150 mm and a 550 mm cone fluidized chamber. This setup was modified to
to incorporate
a bottom
inlet
into
chamberproviding
providingthe
theinflow
inflowofofaerosol
aerosoland
andmixing
mixing with
with the
the
incorporate
a bottom
inlet
into
thethe
chamber
fluidized particles.
particles. The
The aerosol
aerosol atomizer
atomizer in
in the
the experiments
experiments can
can well
well assume
assume the
the role
role of
of the
the nozzle
nozzle in
in
fluidized
the ordinary
ordinary Wurster
Wurster fluidized
fluidized bed
bed (see
(see Figure
Figure 4).
4). In
In the
the course
course of
of the
the study,
study, the
the distance
distance between
between
the
Wurster
rises
tube
and
distributer
plate
was
set
as
5cm.
Different
length
of
copper
tubes
(inner
Wurster rises tube and distributer plate was set as 5cm. Different length of copper tubes (inner diameter:
diameter:
2.2installed
cm) wereinside
installed
therises
Wurster
tubeplace
and take
place
of the
to introduce
2.2
cm) were
the inside
Wurster
tube rises
and take
of the
nozzle
to nozzle
introduce
solution
in
the
fluidized
bed.
solution in the fluidized bed.
Figure 4.
4. Scheme
Scheme of
of the
the experimental
experimental setup
setup using
using aerosols
aerosols for
for particle
particle coating
coating in
in fluidized
fluidizedbed.
bed.
Figure
During the
the experiments,
experiments, the
the aerosol
aerosol and
and compressed
compressed air
air coming
coming from
from the
the aerosol
aerosol generator
generator were
were
During
introduced
introduced into
into the
the fluidized
fluidized bed
bed system
system from
from the
the bottom
bottom through
through aa copper
copper inlet
inlet tube.
tube. To
To prevent
prevent
particles
from
falling
back
into
the
copper
tube,
its
top
of
was
closed.
Ten
openings
with
a
diameter
of
particles from falling back into the copper tube, its top of was closed. Ten openings with a diameter
0.5
cmcm
each
were
evenly
arranged
at at
a distance
of 0.5
each
were
evenly
arranged
a distanceofof1 1cm
cmfrom
fromthe
thetop.
top.To
Tostudy
studythe
theeffect
effectof
of different
different
opening heights on the experimental results, different lengths of tubes were used in the experiments.
Namely 6, 7, and 8 cm.
In the beginning, 1 kg of core particles were placed in the fluidized bed and fluidized by hot air
at a mass flow rate of 15 kg/h for 63 μm particles, 23 kg/h for 202 μm particles. After the temperature
in the product vessel was stabilized, the coating solution was sprayed into the fluidized bed. The
Processes 2020, 8, 1525
6 of 18
opening heights on the experimental results, different lengths of tubes were used in the experiments.
Namely 6, 7, and 8 cm.
In the beginning, 1 kg of core particles were placed in the fluidized bed and fluidized by hot air at
Processes
2020,
8, x of
FOR
of 18
a mass
flow
rate
15PEER
kg/hREVIEW
for 63 µm particles, 23 kg/h for 202 µm particles. After the temperature6 in
the product vessel was stabilized, the coating solution was sprayed into the fluidized bed. The aerosol
are listed
in operated
Table 2. Every
10 compressed
min, a sample
takentemperature.
from the fluidized
bed. Every
was
atomizer
was
by 5 bar
airwas
at room
A peristaltic
pumpsample
adjusted
kept
in
a
container
for
further
analysis.
After
3
h
of
operation,
the
experiment
was
finished.
the liquid level in the coating solution vessel. The operating parameters of each experiment are listed
in Table 2. Every 10 min, a sample was taken from the fluidized bed. Every sample was kept in a
2. Experimental
plan.
container for further analysis. After 3 h Table
of operation,
the experiment
was finished.
Coating
NaB
HPMC
Water
Table
2. Experimental
plan.
Material
Percentage
Percentage
Percentage
1
Glass (63 μm)
30%
0%
70%
NaB
HPMC
Water
Coating
2 Serial
Glass (202
μm)Material
30% Percentage 0%Percentage
70%
Percentage
3 1Glass (63Glass
μm)(63 µm) 30%
0%
70%70%
30%
0%
4 2Glass (63
μm)
30%
0%
70%70%
Glass (202 µm)
30%
0%
5 3Glass (63Glass
μm)(63 µm) 30%
0%
70%70%
30%
0%
30%
0%
6 4Glass (63Glass
μm)(63 µm) 30%
0%
70%70%
30%
70%
7 5Glass (63Glass
μm)(63 µm) 30%
0.5% 0%
69.5%
6
Glass (63 µm)
30%
0%
70%
8 7Glass (63Glass
μm)(63 µm) 30%
1% 0.5%
69%
30%
69.5%
9 8Glass (63Glass
μm)(63 µm) 30%
1.5% 1%
68.5%
30%
69%
30%
68.5%
10 9Glass (63Glass
μm)(63 µm) 30%
0% 1.5%
70%
30%
0%
11 10Glass (63Glass
μm)(63 µm) 30%
0%
70%70%
Serial
11
Glass (63 µm)
30%
0%
70%
Air Inlet
Copper Tube
Temperature
Length
70 °C
8 cm
Air Inlet
Copper Tube
70
°C
8
Temperature
Length cm
50
°C
8 cm
◦
70 C
8 cm
◦
60
°C
70 C
8 cm 8 cm
80◦ C
°C
50
8 cm 8 cm
◦
60
8 cm 8 cm
90 C
°C
80 ◦ C
8 cm
70◦ °C
8 cm
90 C
8 cm
70
°C
70 ◦ C
8 cm 8 cm
◦
70
°C
70 C
8 cm 8 cm
70
8 cm 7 cm
70◦ C
°C
◦
70
7 cm 6 cm
70 C
°C
70 ◦ C
6 cm
2.4. Characterization Methods
2.4. Characterization Methods
The PSD of each sample taken during the experiment was measured offline by using an optical
The PSD of device,
each sample
taken during the
experiment
was measured
offline
using
opticalan
measurement
the CAMSIZER-XT
(Retsch
Technologies,
Germany).
Theby
PSD
is inan
general
measurement
device, of
theproduct
CAMSIZER-XT
(Retsch Technologies,
Germany).
The PSD
is in general
an
important measure
quality. Moreover,
the accumulated
differences
in volume
frequency
important
measure
of
product
quality.
Moreover,
the
accumulated
differences
in
volume
frequency
between product PSD, fi,pr, and raw material PSD, fi,raw, above the point, m, of intersection of those
between
product
PSD,
fi,prto, and
raw material
PSD,
fi,raw , above theThis
point,
m, of intersection
of those
two curves
can be
used
evaluate
the degree
of agglomeration.
corresponds
to the blue
area of
two
curves
can
be
used
to
evaluate
the
degree
of
agglomeration.
This
corresponds
to
the
blue
area
Figure 5 and has been proposed in order to quantify agglomeration by Chen et al. [27] as well asofby
Figure
5 and
has[20].
beenIt proposed
in order be
to quantify
agglomeration
by Chen
et al. [27]
as well
as byto
Hample
et al.
should, however,
noted that
size enlargement
by coating
is also
captured
Hample
et al.
[20].by
It ϕ
should,
however,
be noted (1).
that size enlargement by coating is also captured to a
a certain
extent
obtained
from Equation
certain extent by φ obtained from Equation (1).
𝜙X
= (𝑓 , − 𝑓, ),
(1)
φ=
fi,pr − fi,raw ,
(1)
i≥m
Figure 5. Principle for calculating the agglomeration ratio.
Figure 5. Principle for calculating the agglomeration ratio.
Moreover, the extent of particle surface coating, which is named coating coverage here, was
determined from scanning electron microscopy (SEM) images, taken with a Phenom G2 Pro
microscope. Each image used for this purpose contained 35–40 small particles and had a resolution
of 1024 × 1024 pixels sufficient to identify the coated area. Three images with, in total, 100 particles
Processes 2020, 8, 1525
7 of 18
Moreover, the extent of particle surface coating, which is named coating coverage here,
was determined from scanning electron microscopy (SEM) images, taken with a Phenom G2 Pro
microscope. Each image used for this purpose contained 35–40 small particles and had a resolution of
1024 × 1024 pixels sufficient to identify the coated area. Three images with, in total, 100 particles were
Processes 2020, 8, x FOR PEER REVIEW
7 of 18
typically used for each sample. An analysis was conducted with image identification tools in Python.
The
image
analysis
procedure
shown
Figure
a first
step,
image
preprocessing
was
The
image
analysis
procedure
is is
shown
in in
Figure
6. 6.
In In
a first
step,
thethe
image
preprocessing
was
performed
by
Hough
circle
detection,
which
can
identify
and
locate
the
position
of
particles
in
performed by Hough circle detection, which can identify and locate the position of particles in thethe
original
image.
Particles
that
were
partially
outside
image
were
discarded.
Figure
shows
original
image.
Particles
that
were
partially
outside
of of
thethe
image
were
discarded.
Figure
6a6a
shows
thethe
identified
particles
in
the
original
SEM
picture.
Applying
the
particle
areas
obtained
by
the
pretreatment
identified particles in the original SEM picture. Applying the particle areas obtained by the
of the original
image,
the particles
were
isolatedwere
and the
background
cleared. Circle
areas were
then
pretreatment
of the
original
image, the
particles
isolated
and thewas
background
was cleared.
Circle
measured,
images
were converted
to 8-unit
images,to
and
a threshold
to binary)
was applied.
areas
were then
measured,
images were
converted
8-unit
images,(greyscale
and a threshold
(greyscale
to
The
above
step
successfully
removed
the
noise
caused
by
reflection.
Subsequently,
the
coating
coverage
binary) was applied. The above step successfully removed the noise caused by reflection.
was calculated
the proportion
black
pixels within
theproportion
circular image
areaspixels
that correspond
Subsequently,
theascoating
coverageofwas
calculated
as the
of black
within the to
identified
particles,
seecorrespond
Figure 6b. To
eliminate particles,
the impact
ofFigure
the inability
accurately
circular
image
areas that
to identified
see
6b. To to
eliminate
the identify
impact ofthe
edge
pixels
of
the
particles
on
the
results,
only
the
central
part
of
each
particle
in
the
original
image
the inability to accurately identify the edge pixels of the particles on the results, only the central part
in the
calculation
thatused
is, the
region
whose radius
was
80%
initial
radius.
ofwas
eachused
particle
in the
original process,
image was
in the
calculation
process,
that
is, of
thethe
region
whose
Please
note
that
SEM
images
are
planar,
so
there
would
be
errors
in
the
directly
calculated
coverage
radius was 80% of the initial radius. Please note that SEM images are planar, so there would be errors
the curvature
of coverage
the particles.
eliminate
this of
effect,
a geometrical
correction
was
applied
in due
the to
directly
calculated
due To
to the
curvature
the particles.
To eliminate
this
effect,
a
to
the
pictures.
The
respective
change
of
S
(surface
area
on
sphere)
to
S
(visible
area)
with
3D
2D
geometrical correction was applied to the pictures. The respective change of S3D (surface area onthe
distance
particle center can be seen in Figure 6c. In this study, to eliminate the blurring of the
sphere)
to Sfrom
2D (visible area) with the distance from particle center can be seen in Figure 6c. In this
circular
border, evaluation
hasof
been
by dividing
the central
area
(80% based
diameter)
study,
to eliminate
the blurring
the conducted
circular border,
evaluation
has been
conducted
byon
dividing
the of
each
particle
into
ten
parts,
see
Figure
6d.
central area (80% based on diameter) of each particle into ten parts, see Figure 6d.
(a)
(b)
(c)
(d)
Figure
6. (a)
Pretreatment
of SEM
pictures
byby
Hough
circle
detection.
(b)(b)
Processed
binary
picture
forfor
Figure
6. (a)
Pretreatment
of SEM
pictures
Hough
circle
detection.
Processed
binary
picture
to Sto2DSwith
respect
to the
distance
to particle
center.
thethe
calculation
of of
coating
coverage.
(c)(c)
Ratio
of of
S3DS3D
calculation
coating
coverage.
Ratio
respect
to the
distance
to particle
center.
2D with
Area
subdivision
image
error
elimination
method.
(d)(d)
Area
subdivision
of of
thethe
image
error
elimination
method.
For further analysis, the yield of the coating process Y was also calculated. For this purpose, at
first, each of the samples taken from the product was dried in an oven to get the dry mass M1 of
coated particles accurately. Subsequently, the coating layers were totally removed by soaking and
stirring the sample in distilled water for 24 h. The second desired mass M2 was obtained after the core
Processes 2020, 8, 1525
8 of 18
For further analysis, the yield of the coating process Y was also calculated. For this purpose,
at first, each of the samples taken from the product was dried in an oven to get the dry mass M1 of
coated particles accurately. Subsequently, the coating layers were totally removed by soaking and
stirring the sample in distilled water for 24 h. The second desired mass M2 was obtained after the core
particles dried again. Combined the total sprayed amount Ms and raw material mass Mr process yield
can then be calculated as:
(M1 − M2 )/M2
Y=
,
(2)
Rcoating,max
Rcoating,max =
Ms
,
Mr
(3)
3. Results and Discussion
The coating process in a fluidized bed is complex. It involves an ingenious balance between
mixing, wetting and drying, which depends on several parameters such as air temperature and flow
rate, sprayed solution rate, coating solution composition and many other factors [1,28–30]. All the
above variables have a decisive effect on the quality of the product. Comprehensive experimental and
numerical research has been done in the past few years for large particles coated in fluidized beds with
conventional nozzles to explain the effect of these variables.
From the past studies it is clear that: (1) the rise of inlet air temperature enhances the drying
of wet particles during the coating; (2) more uniform dispersion of coating solution droplets in the
equipment results in more uniform layers on the particle surfaces; (3) growth of coating layer thickness
depends on the flow rate of the coating solution. Despite the many results on the coating of large
particles (diameter > 200 µm) in the literature, less attention has been paid to the investigation of the
coating parameters of fine particles. However, with the help of the novel aerosol atomizer, the coating
of fine particles is easier, so it is possible to make a systematic investigation of the effects of important
operating parameters on the coating performance.
3.1. Effect of Particle Size
As mentioned earlier, large particles and small particles usually behave differently in fluidized
bed coating. This difference is usually caused by the imperfect fluidization of small particles. In this
study, in order to enhance the fluidization of small particles, the Wurster fluidized bed was used.
At the same time, the new aerosol generator was also applied in the coating process for both kinds of
particles. The experimental results show that both large and small particles could be coated under the
usage of aerosol, but there are still some differences.
Particle size distributions of uncoated and coated glass particles are plotted in Figure 7a.
Average particle sizes of small and large coated particles are 66 µm and 215 µm respectively, with the
original particles at 63 µm and 202 µm, as mentioned before. This implied that most of the particles
were individually coated. In addition, the agglomeration rate φ were 5.4% and 6.8%, respectively,
when using the calculation method from Section 2.4. These results all prove that both large particles
and small particles could be successfully coated without severe agglomeration by the novel method.
Figure 7b shows the evolution of the average particle diameter for the product samples.
The absolute growth value of large particles is higher than that of small particles. On the other
hand, the growth interval of small particles is mainly concentrated in the first 40 min and particle
diameter is stabilized after that, while large particles can continue to grow for nearly 90 min.
the original particles at 63 μm and 202 μm, as mentioned before. This implied that most of the
particles were individually coated. In addition, the agglomeration rate ϕ were 5.4% and 6.8%,
respectively, when using the calculation method from Section 2.4. These results all prove that both
large particles and small particles could be successfully coated without severe agglomeration by the
Processes 2020, 8, 1525
9 of 18
novel method.
Processes 2020, 8, x FOR PEER REVIEW
9 of 18
Figure 7b shows the evolution of the average particle diameter for the product samples. The
(a)of large particles is higher than that of small particles.
(b) On the other hand, the
absolute growth value
growth
interval
of small
is mainly
concentrated
in theparticle
first
40diameter
min
andd1,3
particle
diameter is
Figure
(a)Particle
Particle
sizeparticles
distributions,
and(b)
(b)
evolutionofofaverage
average
particle
diameter
d
1,3for
forparticles
particles
Figure
7.7.(a)
size
distributions,
and
evolution
stabilized
after
that, while large particles can continue to grow for nearly 90 min.
different
size.
ofofdifferent
size.
SEM images with different magnification of product (shown in Figure A1 of Appendix A) reveal
SEM
images
different
of product
in proving
Figure A1the
of Appendix
revealof
the larger
and with
smaller
glass magnification
beads were both
coated(shown
by NaB,
successful A)
coating
the
larger and
smaller
glass
beads
were both
byThe
NaB,
proving the successful
coating of
particles
by the
novel
coating
process
withcoated
aerosol.
corresponding
coating coverage
asparticles
shown in
byFigure
the novel
coating
process
with and
aerosol.
The
corresponding
coating
coveragecovered,
as shown
in Figure
8
8 reveals
that
both large
small
particles
were almost
completely
with
the mean
reveals
that
both
large
and
small
particles
were
almost
completely
covered,
with
the
mean
coverage
coverage value of large particles being 99.2% and that of small particles being 99%. Since this is close
value
of large
and that
of small
particles
99%. Since this is close to unity,
to unity,
the particles
coveragebeing
had a99.2%
very small
standard
deviation
inbeing
both cases.
the coverage had a very small standard deviation in both cases.
Figure 8. Coverage comparison based on particle size.
Figure 8. Coverage comparison based on particle size.
Smaller particle size results in a relatively lower yield, as shown in Table 3; 13.59% for glass
Smaller
size results
a relatively
lower
yield,
as shown
in Table
glass cores
cores of
63 µmparticle
as compared
with in
21.51%
for glass
cores
of 202
µm. The
loss 3;
of13.59%
coatingfor
material
is
of
63
μm
as
compared
with
21.51%
for
glass
cores
of
202
μm.
The
loss
of
coating
material
is
mainly
mainly attributed to the collected fines in the filter above the fluidized bed. Because the droplet size
collectedsmall,
fines itinwas
the easy
filter to
above
the fluidized
thetodroplet
size
the
ofattributed
the aerosoltoisthe
extremely
be dried
by hot airbed.
andBecause
entrained
the filter
byof
the
aerosol
is
extremely
small,
it
was
easy
to
be
dried
by
hot
air
and
entrained
to
the
filter
by
the
fluidization gas.
fluidization gas.
1
1
Table 3. Yield of aerosol coating process with different particles 1 .
Table 3. Yield of aerosol coating process with different particles 1.
Particle Type
Mass of Cores (g)
Coating Mass (g)
Yield (%) 2
Particle Type Mass of Cores (g) Coating Mass (g) Yield (%) 2
Small particle
2.5010 ± 0.55
0.0849 ± 0.017
13.59
Small particle
2.5010 ± 0.55
0.0849 ± 0.017
13.59
Large particle
2.0592 ± 0.16
0.1122 ± 0.012
21.51
Large particle
2.0592 ± 0.16
0.1122 ± 0.012
21.51
Total mass of glass beads: 1000 g; Mean aerosol spray rate: 282 g/h.
2
Average value from three samples.
Total mass of glass beads: 1000 g; Mean aerosol spray rate: 282 g/h. 2 Average value from three samples.
From the above results, it can be seen that both large and small particles can be properly
From the above results, it can be seen that both large and small particles can be properly
fluidized in the modified Wurster equipment, and uniform coating can be obtained when the aerosol
fluidized in the modified Wurster equipment, and uniform coating can be obtained when the aerosol
generator replaces the conventional two-fluid nozzle. However, it is worth noting that due to the
generator replaces the conventional two-fluid nozzle. However, it is worth noting that due to the
extremely small size of the aerosol, it takes a longer time for the core particles to obtain an almost
extremely small size of the aerosol, it takes a longer time for the core particles to obtain an almost
complete coating compare to the conventional coating process. Moreover, according to experimental
observations, small particles would rather stick to the wall of the fluidized bed due to electrostatic
and other micro-scale forces, which might explain why the coverage rate is slightly lower than that
of large particles.
Processes 2020, 8, 1525
10 of 18
complete coating compare to the conventional coating process. Moreover, according to experimental
observations, small particles would rather stick to the wall of the fluidized bed due to electrostatic
and other micro-scale forces, which might explain why the coverage rate is slightly lower than that of
large particles.
3.2. Effect of Air Inlet Temperature
Hot fluidized gas passing through the distributor plate and the fluidization chamber is responsible
for drying of the particles by heat and mass transfer to the fluidized bed. Based on previous experience,
higher temperature during the process favors drying, leading to rapid evaporation of droplets.
Processes 2020, 8, x FOR PEER REVIEW
10 of 18
In comparison, when temperature is lower, the evaporation rate of the coating solution decreased,
leading to slight agglomeration. A moderate coating process in the modified fluidized bed is only
coating solution decreased, leading to slight agglomeration. A moderate coating process in the
obtained when wetting and drying reached a proper balance. The above conclusions are also applicable
modified fluidized bed is only obtained when wetting and drying reached a proper balance. The
to the use of traditional nozzles for large particles. However, the balance is more difficult to maintain
above conclusions are also applicable to the use of traditional nozzles for large particles. However,
when using fine particles, and the temperature control the coating of fine fluidized particles using the
the balance is more difficult to maintain when using fine particles, and the temperature control the
aerosol generator is more challenging.
coating of fine fluidized particles using the aerosol generator is more challenging.
The particle size distribution results for small glass particles when air inlet temperature increased
The particle size distribution results for small glass particles when air inlet temperature
from 50 ◦ C to 90 ◦ C are plotted in Figure 9a. Product size does not have obvious differences as
increased from 50 °C to 90 °C are plotted in Figure 9a. Product size does not have obvious differences
temperature changes; mean particle size is 67.5 µm, 66.8 µm, 66 µm, 65.7 µm and 65.8 µm respectively.
as temperature changes; mean particle size is 67.5 μm, 66.8 μm, 66 μm, 65.7 μm and 65.8 μm
Figure 9a also indicates the slight difference of agglomeration ratios (13.1%, 8.6%, 5.4%, 3.6% and
respectively. Figure 9a also indicates the slight difference of agglomeration ratios (13.1%, 8.6%, 5.4%,
3.4% based on the PSD results). It can be seen that the overall agglomeration rate is at a relatively
3.6% and 3.4% based on the PSD results). It can be seen that the overall agglomeration rate is at a
low level, but there are still more agglomerated particles under low temperature conditions than at
relatively low level, but there are still more agglomerated particles under low temperature conditions
high temperature.
than at high temperature.
(a)
(b)
Figure
(a)Particle
Particlesize
size
distributions,
(b) evolution
of average
particle
diameter
d1,3 at
Figure9.9. (a)
distributions,
andand
(b) evolution
of average
particle
diameter
d1,3 at different
different
temperatures.
temperatures.
As
in Figure
9b, in
each
is achieved
at the beginning
the experiment.
Asdepicted
depicted
in Figure
9b,
in case,
each faster
case, growth
faster growth
is achieved
at the of
beginning
of the
After
60 min, average
size is still
increasing.
The increasing.
process is stopped
at 180 is
min
with fluctuating
experiment.
After 60particle
min, average
particle
size is still
The process
stopped
at 180 min
average
particle diameters
of approximately
µm in the product.
with fluctuating
average particle
diameters 66
of approximately
66 μm in the product.
SEM
SEMimages
imagesshown
shownininFigure
FigureA2
A2ofofAppendix
AppendixAAthat
thatclearly
clearlyillustrate
illustratethat
thatduring
duringthe
thecoating
coating
process
in
the
modified
fluidized
bed,
agglomeration
hardly
happened.
At
the
lowest
temperature
process in the modified fluidized bed, agglomeration hardly happened. At the lowest temperature of
◦ C, coating quality was though worse than at higher temperature. When particles are coated
of5050°C,
coating quality was though worse than at higher temperature. When particles are coated at a
athigher
a higher
temperature,
images
showa abetter
better coating
coating structure.
structure. To
To differentiate
temperature,
thethe
images
show
differentiate the
the coating
coating
performance
at
the
above
five
temperatures,
coating
coverage
calculation
based
on
SEM
images
performance at the above five temperatures, coating coverage calculation based on SEM imageshas
has
been
beencarried
carriedout
outand
andthe
theresults
resultsare
areshown
shownininFigure
Figure10.
10.ItItisisseen
seenthat
thatthe
thecoating
coatingcoverage
coveragetends
tendstoto
◦ (with coating coverage
increase
increasewith
withthe
thefluidized
fluidizedbed
bedtemperature,
temperature,though
thoughwith
withaamaximum
maximumatat70
70 C
°C (with coating coverage
◦
◦
◦
◦
◦ C).
ofof86%
for
50
C,
88%
for
60
C,
99%
for
70
C,
94%
for
80
C
and
92%
for
90
86% for 50 °C, 88% for 60 °C, 99% for 70 °C, 94% for 80 °C and 92% for 90 °C).
higher temperature, the images show a better coating structure. To differentiate the coating
performance at the above five temperatures, coating coverage calculation based on SEM images has
been carried out and the results are shown in Figure 10. It is seen that the coating coverage tends to
increase with the fluidized bed temperature, though with a maximum at 70 °C (with coating coverage
Processes
2020,
1525
11 of 18
of 86%
for8,50
°C, 88% for 60 °C, 99% for 70 °C, 94% for 80 °C and 92% for 90 °C).
Figure 10. Coverage comparison based on temperature.
Figure 10. Coverage comparison based on temperature.
The influence of temperature on the yield of the aerosol coating process is shown in Table 4.
The results indicate that when the temperature is too high, the yield may be much lower than at
moderate temperature; 13.59% for the temperature of 70 ◦ C as compared with 8.83% for the temperature
of 90 ◦ C. It is known that temperature may be a critical factor that can affect process yield even when
using a two-fluid nozzle. In fact, when the aerosol generator is used, because the aerosol has smaller
droplet size, the temperature causes a more noticeable effect on the yield. Higher temperature results
in the loss of coating material due to rapidly evaporation before deposition and coating can take place
on the surface of the cohesive fluidized particles, which allows the aerosol droplets to be directly dried
into fine powder and then quickly blown out of the fluidization chamber with the airflow.
Table 4. Yield of aerosol coating process at different temperatures 1 .
1
Temperature (◦ C)
Mass of Cores (g)
Coating Mass (g)
Yield (%) 2
50
60
70
80
90
2.0113 ± 0.07
2.2572 ± 0.37
2.5010 ± 0.55
1.8998 ± 0.16
1.3141 ± 0.05
0.0772 ± 0.015
0.0779 ± 0.002
0.0849 ± 0.017
0.0435 ± 0.012
0.0293 ± 0.041
14.48
14.22
13.59
9.08
8.83
Total mass of glass beads: 1000 g; Mean aerosol spray rate: 282 g/h.
2
Average value from three samples.
In summary, it can be seen that the evaluation indicators of coated products are susceptible to
temperature changes. At low temperature, some aerosol droplets cannot be dried in time. So even
though the droplets are very small when the aerosol generator is applied, liquid bridges between
the particles may still lead to agglomeration. On the contrary, agglomeration is negligible at high
temperature, but the loss of coating solution by overspray may cause lower solution yield. Therefore,
in the case of various process parameters being adjusted, it is particularly important to first determine
an appropriate range of mild temperatures to obtain high-quality products.
3.3. Effect of Coating Solution Viscosity
To investigate the influence of coating solution viscosity on coating speed and process time with
the same core particles as before, an additive was mixed into the coating solution. As mentioned above,
the additive used was HPMC, which is suitable for the agglomeration of excipients and also plays an
important role as a coating material in the pharmaceutical industry. The concentration of HPMC in the
solution should be carefully selected. Due to its extremely high viscosity, HPMC is usually used as an
agglomerating binder with more than 4 wt% in the solution [31]. In this study of coatings, less HPMC
was added to the NaB solution, namely zero to 1.5%.
The influence of coating solution viscosity on coating performance was experimentally investigated
by adding different mass fraction of HPMC (with a relationship between HPMC mass fraction and
Processes 2020, 8, 1525
12 of 18
solution viscosity, as can be seen in Table 1), and the corresponding results are discussed in the
following. Figure 11a indicates that more HPMC in the solvent makes the size distribution of coated
glass particles slightly different and somewhat narrows. The average size of the final coated particles
decreased monotonically from 66 µm to 64.8 µm when the HPMC mass fraction in the NaB solution
was increased from 0% to 1.5%. Figure 11b shows the evolution of average particle size with different
viscosity of the coating solution. In these experiments, a much more quicker particle growth is achieved
when the coating solution has a higher viscosity. A fluctuating steady state is reached in only 20 min
for 1.5 wt% HPMC, compared to 40 min for 0% HPMC. On the other hand, higher solution viscosity
results
smaller
final
particle
from
Processesin
2020,
8, x FOR
PEER
REVIEWsize. The process stops after 180 min with particle size growth 12
of 18
1.8 µm to 3 µm.
(a)
(b)
Figure
Figure11.
11.(a)
(a)Particle
Particlesize
sizedistributions
distributionsand
and(b)
(b)evolution
evolutionofofaverage
averageparticle
particlediameter
diameterd1,3
d1,3for
fordifferent
different
HPMC concentrations.
HPMC concentrations.
With traditional two-fluid nozzles, higher viscosity of the coating solution can increase the rate
With traditional two-fluid nozzles, higher viscosity of the coating solution can increase the rate
of undesired agglomeration in the course of coating processes. However, the novel aerosol atomizer
of undesired agglomeration in the course of coating processes. However, the novel aerosol atomizer
can generate small aerosol droplets, which makes it difficult to produce solid bridges at a suitable
can generate small aerosol droplets, which makes it difficult to produce solid bridges at a suitable
process temperature even when the solution viscosity is high. The present results confirm that the
process temperature even when the solution viscosity is high. The present results confirm that the
agglomeration rate remains small even at high viscosity of the coating solution (agglomeration rate of
agglomeration rate remains small even at high viscosity of the coating solution (agglomeration rate
2.99% for 0.5 wt% HPMC, 2.32% for 1 wt% HPMC and 1.24% for 1.5 wt% HPMC based on the PSD).
of 2.99% for 0.5 wt% HPMC, 2.32% for 1 wt% HPMC and 1.24% for 1.5 wt% HPMC based on the
SEM images of coated particles are shown in Figure A3 of the Appendix A. It appears that the
PSD).
increase in viscosity of the solution decreases significantly the overall average coverage of particles,
SEM images of coated particles are shown in Figure A3 of the Appendix A. It appears that the
with increasing differences among individual particles of the population. In general, island growth is
increase in viscosity of the solution decreases significantly the overall average coverage of particles,
observed, and patchy particles are obtained. In the case of 1.5 wt% HPMC, the overall coverage is only
with increasing differences among individual particles of the population. In general, island growth
21%. In addition to the decrease in coverage, the coating layers also gradually become compact and
is observed, and patchy particles are obtained. In the case of 1.5 wt% HPMC, the overall coverage is
brittle as the viscosity increases. Figure 12 shows the coverage results evaluated from the SEM images
only 21%. In addition to the decrease in coverage, the coating layers also gradually become compact
at different HPMC concentrations.
and brittle as the viscosity increases. Figure 12 shows the coverage results evaluated from the SEM
As the solution viscosity increases, the coverage shows a downward trend. One explanation is that
images at different HPMC concentrations.
after the high-viscosity liquid dries on the surface of the particles, the dense and compact coating layers
break off due to collisions between particles. Finally, the growth and breakage of the coating layer
reach an equilibrium at a certain surface coverage, which may be quite low. To verify this conjecture,
an additional experiment has been conducted with 5 wt% of HPMC. During this experiment, the yield
of the solution has been evaluated every 10 min. Results are plotted in Figure 13. It can be seen that the
yield decreases to zero 60 min, that is, the rate of coating layer formation becomes equal to the rate of
dried coating layer breakage. This trend is stronger when introducing HPMC to increase the viscosity
of the solution; it is, therefore, not easy to obtain a coating product with 100% coverage in such cases.
Figure 12. Coverage comparison based on HPMC concentration
As the solution viscosity increases, the coverage shows a downward trend. One explanation is
increase in viscosity of the solution decreases significantly the overall average coverage of particles,
with increasing differences among individual particles of the population. In general, island growth
is observed, and patchy particles are obtained. In the case of 1.5 wt% HPMC, the overall coverage is
only 21%. In addition to the decrease in coverage, the coating layers also gradually become compact
and
brittle
as8,the
SEM
Processes
2020,
1525viscosity increases. Figure 12 shows the coverage results evaluated from the 13
of 18
images at different HPMC concentrations.
Processes 2020, 8, x FOR PEER REVIEW
13 of 18
HPMC to increase the viscosity of the solution; it is, therefore, not easy to obtain a coating product
with 100% coverageFigure
in such
12.cases.
Coverage comparison based on HPMC concentration
Figure 12. Coverage comparison based on HPMC concentration
As the solution viscosity increases, the coverage shows a downward trend. One explanation is
that after the high-viscosity liquid dries on the surface of the particles, the dense and compact coating
layers break off due to collisions between particles. Finally, the growth and breakage of the coating
layer reach an equilibrium at a certain surface coverage, which may be quite low. To verify this
conjecture, an additional experiment has been conducted with 5 wt% of HPMC. During this
experiment, the yield of the solution has been evaluated every 10 min. Results are plotted in Figure
13. It can be seen that the yield decreases to zero 60 min, that is, the rate of coating layer formation
becomes equal to the rate of dried coating layer breakage. This trend is stronger when introducing
Figure 13. Coating solution yield versus process time.
Figure 13. Coating solution yield versus process time.
In summary, agglomeration can be avoided by the innovative coating system even when the
In are
summary,
can beisavoided
by thesolution
innovative
systemthe
even
when
the
particles
fine andagglomeration
the coating solution
sticky. Sticky
may coating
even increase
coating
rate
particles
are
fine
and
the
coating
solution
is
sticky.
Sticky
solution
may
even
increase
the
coating
rate
by promoting the deposition of aerosol droplets on the surface of core particles. At the same time,
by promoting
deposition
of aerosol
droplets
onthan
the surface
of core particles.
At coating
the same
time,
however,
coatingthe
layers
are denser
and more
compact
at low viscosity.
Such dense
layers
however,
coating
layers
are
denser
and
more
compact
than
at
low
viscosity.
Such
dense
coating
layers
seem to be brittle, and are more likely to break off after collisions between particles and with walls in
seem
to be fluidized
brittle, and
arewhich
moreresults
likely to
off after
collisions
between
and with
walls in
the
Wurster
bed,
in break
low product
coverage
and
unevenparticles
overall coating
of patchy
the
Wurster
fluidized
bed,
which
results
in
low
product
coverage
and
uneven
overall
coating
of
product particles.
patchy product particles.
3.4. Effect of Aerosol Inlet Position
3.4. Effect of Aerosol Inlet Position
In this study, the aerosol was introduced into the fluidized bed from the bottom by means of a
thistube
study,
the aerosol
was introduced
into the
fluidized
bedfluidized
from thebed
bottom
means of a
copperIninlet
through
the distributor
plate. When
using
a Wurster
withby
a traditional
copper inlet
tubethe
through
the between
distributor
When
using
Wurster
fluidized
bed with
traditional
two-fluid
nozzle,
distance
theplate.
nozzle
outlet
anda the
bottom
distributor
platea can
affect
two-fluid
nozzle,
the
distance
between
the
nozzle
outlet
and
the
bottom
distributor
plate
can
affect
the coating quality. Similarly, in the novel coating system, aerosol and fluidized particles move
the coating in
quality.
Similarly,
the novel
coating may
system,
aerosol
and fluidized
move
concurrently
riser, so
the inletinposition
of aerosol
affect
its contact
time withparticles
the particles.
concurrently
in
riser,
so
the
inlet
position
of
aerosol
may
affect
its
contact
time
with
the
particles.
Therefore, the influence of aerosol inlet position on the experimental results was explored using three
Therefore,
the influence
of lengths
aerosol of
inlet
position
on the
experimental
results was explored using three
different
copper
tubes with
6 cm,
7 cm and
8 cm,
respectively.
different
tubes
lengths size
of 6 distributions
cm, 7 cm andin8 Figure
cm, respectively.
As cancopper
be seen
fromwith
the particle
14, the inlet tube of 8 cm has a minor
As
can
be
seen
from
the
particle
size
distributions
in
Figure
14, the
inlet
tube change
of 8 cm in
hasaverage
a minor
effect in altering the particle size, showing scarce agglomeration and
only
a slight
effect insize.
altering
particle size,
showing
scarce
agglomeration
and
only a of
slight
change
in average
particle
Withthe
a decrease
of the
inlet tube
length
to 7 cm, the
amount
larger
agglomerates
particle size.
With
a decrease
of the
inlet tube
7 cm,that
thedecreasing
amount ofthe
larger
increased
slightly.
Results
based on
calculated
fromlength
graphstoverify
tubeagglomerates
length from
slightly.
basedincreased
on calculated
from
verify
that
decreasing
theproportion
tube length
8 increased
cm to 6 cm
results Results
in a slightly
particle
sizegraphs
from 66
µm to
67.5
µm, and the
8 cm to 6 increases
cm results
in a slightly
increased
sizereason
from for
66 μm
to 67.5 μm,
andthe
the
offrom
agglomeration
significantly
from
5.4% to particle
13.2%. The
this behavior
is that
proportion of agglomeration increases significantly from 5.4% to 13.2%. The reason for this behavior
is that the shorter tube increases the residence time of droplets in the Wurster riser, but it also
increases the humidity in regions with many particles, so agglomeration is more likely to occur.
As depicted in Figure 14b, the experiment reaches steady state after around 60 min with average
particle size of approximately 67.5 μm in the product when the copper tube is 6 cm. Compared to the
7 cm and 8 cm experiments, the growth rate is faster at beginning, and the average particle size is
Processes 2020, 8, x FOR PEER REVIEW
14 of 18
Processes 2020, 8, 1525
14 of 18
shorter
increases
theREVIEW
residence time of droplets in the Wurster riser, but it also increases14the
Processestube
2020, 8,
x FOR PEER
of 18
humidity in regions with many particles, so agglomeration is more likely to occur.
(b)
(a)
Figure 14. (a) Particle size distributions and (b) evolution of average particle diameter d1,3 for different
aerosol inlet tube lengths.
SEM images of the final product (Appendix A, Figure A4) show good coating of the glass
particles for all investigated aerosol inlet tube lengths. The related coating coverage shown in Figure
15 ranges from 99.8% to 99%. All three conditions have a small standard deviation, which means the
(b)
(a)
particle coverage is relatively
uniform.
Table 5 indicates that a longer aerosol tube results in lower yield; 13.59% for a copper tube of 8
Figure
14.
size
and (b)
evolution ofofaverage
particle
diameter
d1,3
different
Figure
14.(a)
(a)Particle
Particle
sizedistributions
distributions
average
particle
diameter
d1,3for
for
cm as
compared
with 23.13%
for 6 cm. Itand
can(b)beevolution
seen that the
closer
the aerosol
outlet
is different
to the bottom
aerosol
inlet
tube
lengths.
aerosol inlet tube lengths.
distributor plate, the less coating solution is lost; therefore, the use rate is the highest in this case.
As depicted in Figure 14b, the experiment reaches steady state after around 60 min with average
SEM images of the final product (Appendix A, Figure A4) show good coating
of the glass
1.
Table 5. Yield67.5
of aerosol
process
withthe
different
aerosol
tubesCompared
particle size of approximately
µm in coating
the product
when
copper
tube inlet
is 6 cm.
to the
particles for all investigated aerosol inlet tube lengths. The related coating coverage shown in Figure
7 cm and 8 cm experiments,
the growth
rate is
at beginning,
and (g)
the average
particle
size is
2
Inlet Tube
Length
Mass
of faster
Cores
(g)
Coating
Mass
Yield (%)
15 ranges from 99.8%
to 99%.
All (cm)
three conditions
have
a small
standard
deviation,
which
means the
slightly larger at steady state.
6 uniform.
1.6187 ± 0.51
0.0697 ± 0.015
23.13
particle coverage is relatively
SEM images of the final7 product (Appendix
A,± Figure
show good
of the glass particles
5.1553
0.21 A4) 0.0848
± 0.009coating 17.47
Table 5 indicates that a longer aerosol tube results in lower yield; 13.59% for a copper tube of 8
for all investigated aerosol inlet
tube lengths.
The ±related
coverage
in Figure 15 ranges
8
2.5010
0.55 coating
0.0849
± 0.017 shown13.59
cm as compared with 23.13% for 6 cm. It can be seen that the closer the aerosol outlet is to the bottom
2 Average
from199.8%
to 99%.
All beads:
three conditions
a small
deviation,
which
means
thesamples.
particle
Total mass
of glass
1000 g; Meanhave
aerosol
spray standard
rate: 282 g/h.
value from
three
distributor plate, the less coating solution is lost; therefore, the use rate is the highest in this case.
coverage is relatively uniform.
Table 5. Yield of aerosol coating process with different aerosol inlet tubes 1.
Inlet Tube Length (cm)
6
7
8
1
Mass of Cores (g)
1.6187 ± 0.51
5.1553 ± 0.21
2.5010 ± 0.55
Coating Mass (g)
0.0697 ± 0.015
0.0848 ± 0.009
0.0849 ± 0.017
Yield (%) 2
23.13
17.47
13.59
Total mass of glass beads: 1000 g; Mean aerosol spray rate: 282 g/h. 2 Average value from three samples.
Figure 15. Coverage comparison based on length of aerosol inlet tube.
Figure 15. Coverage comparison based on length of aerosol inlet tube.
Table 5 indicates that a longer aerosol tube results in lower yield; 13.59% for a copper tube of 8 cm
The above
to seen
the conventional
bottom
spray coating
when the
as compared
withresults
23.13%show
for 6that,
cm. similar
It can be
that the closer
the aerosol
outlet issystem,
to the bottom
coating
process
is
conducted
with
the
novel
aerosol
generator,
the
position
at
which
the
generated
distributor plate, the less coating solution is lost; therefore, the use rate is the highest in this case.
aerosol is introduced into the fluidized bed influences the results. Due to higher particle
1.
Yield ofofaerosol
coating fluidized
process with
different
aerosol
inlet tubes
concentration Table
at the5.bottom
the Wurster
bed
[32], when
aerosol
enters
the fluidized bed
at a lower position, the frequency of contacts with particles and the probability of particles
capturing
Inlet Tube Length (cm)
Mass of Cores (g)
Coating Mass (g)
Yield (%) 2
aerosol droplets rise. Eventually, the process yield will also increase. However, it is worth noting that
Figure
15. Coverage comparison
aerosol inlet tube.
6
1.6187 ± 0.51based on length
0.0697 of
± 0.015
23.13
7
5.1553 ± 0.21
0.0848 ± 0.009
17.47
8 show that, similar
2.5010
± 0.55
0.0849
± 0.017
13.59system, when the
The above results
to the
conventional
bottom
spray coating
mass is
of glass
beads: 1000
g; Mean
aerosolaerosol
spray rate:
282 g/h. 2 Average
value from
samples.
coating1 Total
process
conducted
with
the novel
generator,
the position
at three
which
the generated
aerosol is introduced into the fluidized bed influences the results. Due to higher particle
concentration at the bottom of the Wurster fluidized bed [32], when aerosol enters the fluidized bed
at a lower position, the frequency of contacts with particles and the probability of particles capturing
aerosol droplets rise. Eventually, the process yield will also increase. However, it is worth noting that
Processes 2020, 8, 1525
15 of 18
The above results show that, similar to the conventional bottom spray coating system, when the
coating process is conducted with the novel aerosol generator, the position at which the generated
aerosol is introduced into the fluidized bed influences the results. Due to higher particle concentration
at the bottom of the Wurster fluidized bed [32], when aerosol enters the fluidized bed at a lower
position, the frequency of contacts with particles and the probability of particles capturing aerosol
droplets rise. Eventually, the process yield will also increase. However, it is worth noting that a too-low
inlet position will also cause the humidity of the dense zone of bed to increase, with a potential increase
in the agglomeration.
4. Conclusions
A novel coating process in a Wurster fluidized bed combined with an aerosol atomizer was
presented in this study. It was shown by a series of experiments that different operation parameters
influence the particle growth behavior. The experiments have shown the feasibility of using aerosol
to produce a coating layer on fluidized particles. It should be noted that submicron aerosol droplets
have been used rather than conventional two-fluid nozzle spray with droplet diameters of tens of
microns. Due to the extremely small diameter of aerosol droplets, even very small core particles could
be individually coated and without severe agglomeration. In addition, in the industrial field, due to
the extremely small aerosol droplets, the resulting coating layer is ultrathin, and the yield of the new
coating process was determined to be 15–20% under different conditions, which is considered to be
improvable. The yield in this study may be enhanced by the recycling of entrained solids or better
equipment design, especially when scaling up the experiment plant.
Coating performance was characterized from both microscale (SEM) and macroscale (PSD, yield)
points of view. Different operation parameters such as particle size, air temperature, coating solution
viscosity and aerosol inlet position in the Wurster riser have an influence on product quality. The main
conclusions are as follows:
•
•
•
•
Both conventionally large particles and smaller particles that are difficult to coat in conventional
equipment can be coated without agglomeration in a Wurster fluidized bed with the help of an
aerosol generator. The highest growth rate is observed at the beginning of the novel process.
Over time, the growth rate decreases gradually until a quasi-steady state has been reached.
The temperature of the fluidizing gas has a significant impact on the quality of the coated
product. A decrease in temperature may increase the agglomeration rate, while excessively high
temperature will lower the use of coating solution and the coating rate. Moderate temperature
can maintain reasonable yield and low agglomeration ratio.
Particle growth rate was observed to increase with the increase of coating solution viscosity.
Higher viscosity aerosol is easier to adhere to the particles. Unfortunately, this is accompanied
by the creation of compact and dense coating layers, which are more likely to break down due
to collisions with other particles and walls in the Wurster fluidized bed. Equilibrium between
coating layer growth and breakage may result in lower coverage when the solution viscosity is
high, and in patchy product particles.
The position where coating aerosol is introduced into the fluidized bed has an impact on the use
rate of the coating solution. When aerosol is fed close to the bottom distributor plate, it interacts
with the particles in a dense zone of the equipment, hence the use rate is higher. Meanwhile,
the agglomeration rate also slightly increases.
In summary, the use of very small aerosol droplets enables the coating of fine particles without
severe agglomeration in a Wurster fluidized bed. This novel approach of potential interest for
pharmaceutical, food, catalyst and other industries, where it may contribute to the development of
new products based on coated fine particles.
Processes 2020, 8, 1525
16 of 18
Author Contributions: R.Z. carried out the experiment. R.Z. wrote the manuscript with support from T.H. and
E.T., T.H. and E.T. helped supervise the project. E.T. and R.Z. conceived the original idea. All authors have read
and agreed to the published version of the manuscript.
Funding: The first author gratefully acknowledges the funding of this work by the China Scholarship Council
(NO. 201708080202).
Acknowledgments: We also thank Jing Yang for his assistance in the conduction of the experiments and
measurements of samples in the frame of his M.Sc. thesis.
Processes2020,
2020,
8,xxFOR
FORPEER
PEER
REVIEWdeclare no conflict of interest.
Processes
Conflicts
of 8,
Interest:
The REVIEW
authors
16ofof18
18
16
AppendixAA
Appendix
(b)
(b)
(a)
(a)
FigureA1.
SEMimages
imagesfrom
fromexperiments
experimentsusing
using(a)
(a)large
largeparticles;
Figure
A1.SEM
particles;(b)
(b)small
smallparticles.
particles.
(b)
(b)
(a)
(a)
(c)
(c)
(e)
(e)
(d)
(d)
◦ C; (b) 60 ◦ C;
FigureA2.
A2.SEM
SEMimages
images
from
fluidization
temperatures
(a)(a)
5050
imagesfrom
fromexperiments
experimentsunder
underdifferent
different
fluidization
temperatures
(a)
50°C;
°C;(b)
(b)60
60
Figure
under
different
fluidization
temperatures
◦
◦
◦
(c)
70
C;°C;
(d)
80
C;°C;
(e)(e)
90
C.°C.
°C;(c)
(c)70
70
°C;(d)
(d)80
80
°C;
(e)90
90
°C.
°C;
(a)
(a)
(b)
(b)
(e)
(d)
Figure
A2. SEM images from experiments under different fluidization temperatures (a) 50 °C; (b) 60
Processes 2020,
8, 1525
°C; (c) 70 °C; (d) 80 °C; (e) 90 °C.
Processes 2020, 8, x FOR PEER REVIEW
Processes 2020, 8, x FOR PEER REVIEW
(a)
(b)
(c)
(c)
(d)
(d)
17 of 18
17 of 18
17 of 18
Figure
A3. SEM
SEM
images
from
experimentsusing
usingvarious
variousHPMC
HPMCmass
mass
fractions
coating
solution
images
from
experiments
using
various
fractions
ofof
the
coating
solution
Figure
A3. SEM
images
from
experiments
HPMC
mass
fractions
ofthe
the
coating
solution
(a)
0
wt%;
(b)
0.5
wt%;
(c)
1.0
wt%;
(d)
1.5
wt%.
(b) wt%;
0.5 wt%;
wt%;
(d)1.5
1.5wt%.
wt%.
(a) 0 wt%; (b) 0.5
(c) (c)
1.01.0
wt%;
(d)
(a)
(a)
(b)
(b)
(c)(c)
Figure
SEM
images
from
experiments
lengths
ofofthe
inlet
tube
(a)(a)
6 cm;
(b)(b)
Figure
A4.
SEM
images
fromexperiments
experimentsusing
using
various
lengths
the
aerosol
inlet
tube
6 cm;
Figure
A4. A4.
SEM
images
from
usingvarious
various
lengths
ofaerosol
the
aerosol
inlet
tube
(a)
6 cm;
7 cm;
(c)
88 cm.
7
cm;
(c)
cm.
(b) 7 cm; (c) 8 cm.
Author Contributions:
support
from
T.H.
and
Author
Contributions: R.Z.
R.Z. carried
carriedout
outthe
theexperiment.
experiment.R.Z.
R.Z.wrote
wrotethe
themanuscript
manuscriptwith
with
support
from
T.H.
and
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
E.T., T.H.
T.H. and
idea.
All
authors
have
read
E.T.,
and E.T.
E.T. helped
helped supervise
supervisethe
theproject.
project.E.T.
E.T.and
andR.Z.
R.Z.conceived
conceivedthe
theoriginal
original
idea.
All
authors
have
read
and agreed
agreed to
published
version
the
Kleinbach,
E.;the
Riede,
T. Coating
ofof
solids.
Chem. Eng. Process. Process. Intensif. 1995, 34, 329–337. [CrossRef]
and
to
the
published
version
of
themanuscript.
manuscript.
Felton,
L.;
Timmins,
G.
A
nondestructive
technique
to determinework
the rate of China
oxygen
permeation
into solid
Funding: The
Scholarship
Council
Funding:
The first
first author
author gratefully
gratefullyacknowledges
acknowledgesthe
thefunding
fundingofofthis
this workbybythe
the China
Scholarship
Council
dosage
forms. Pharm. Dev. Technol. 2006, 11, 141–147. [CrossRef] [PubMed]
(NO. 201708080202).
201708080202).
(NO.
Porter, S.C.; Felton, L.A. Techniques to assess film coatings and evaluate film-coated products. Drug Dev.
Acknowledgments: We also thank Jing Yang for his assistance in the conduction of the experiments and
Acknowledgments:
also thank
Jing Yang
for his assistance in the conduction of the experiments and
Ind.
Pharm. 2010, 36,We
128–142.
[CrossRef]
[PubMed]
measurements of samples in the frame of his M.Sc. thesis.
measurements
samples in
frame of his
thesis. U.; Smirnova, I.; Heinrich, S. Novel technique for
Goslinska,
M.;of Selmer,
I.; the
Kleemann,
C.;M.Sc.
Kulozik,
Conflicts
of
Interest:
The
authors
declare
no
conflict
of interest.
measurement
of coating
layer thickness
fine and
porous particles using focused ion beam. Particuology
Conflicts of Interest:
The authors
declare noofconflict
of interest.
2019, 42, 190–198. [CrossRef]
References
References K.; Huyghebaert, A. Fluidized bed coating in food technology. Trends Food Sci. Technol. 1999, 10,
Dewettinck,
1.
Kleinbach,
E.; Riede, T. Coating of solids. Chem. Eng. Process. Process. Intensif. 1995, 34, 329–337.
163–168.
[CrossRef]
1.
Kleinbach, E.; Riede, T. Coating of solids. Chem. Eng. Process. Process. Intensif. 1995, 34, 329–337.
2.
Felton,
L.;
Timmins,
G. A nondestructive
technique
to determine
the rate
of oxygen
permeation
into solid
Watano,
S.; Imada,
Y.; Hamada,
K.; Wakamatsu,
Y.; Tanabe,
Y.; Dave,
R.N.;
R. Microgranulation
2.
Felton,
L.; Timmins,
G. A nondestructive
technique
to determine
the rate
of Pfeffer,
oxygen permeation
into solid of
dosage forms. Pharm. Dev. Technol. 2006, 11, 141–147.
fine powders
by a Pharm.
novel rotating
fluidized
bed141–147.
granulator. Powder Technol. 2003, 131, 250–255. [CrossRef]
dosage forms.
Dev. Technol.
2006, 11,
3.
Porter, S.C.; Felton, L.A. Techniques to assess film coatings and evaluate film-coated products. Drug Dev.
Jones,
D.M. S.C.;
Factors
to consider
in fluid-bed
processing.
Pharm.
9, 50–55.
3.
Porter,
Felton,
L.A. Techniques
to assess
film coatings
andTechnol.
evaluate1985,
film-coated
products. Drug Dev.
Ind. Pharm. 2010, 36, 128–142.
Rieck,
C.;Pharm.
Bück,2010,
A.; 36,
Tsotsas,
E. Monte Carlo modeling of fluidized bed coating and layering processes.
Ind.
128–142.
4.
Goslinska, M.; Selmer, I.; Kleemann, C.; Kulozik, U.; Smirnova, I.; Heinrich, S. Novel technique for
4.
Goslinska,
M.;2670–2680.
Selmer, I.;[CrossRef]
Kleemann, C.; Kulozik, U.; Smirnova, I.; Heinrich, S. Novel technique for
AIChE
J. 2016, 62,
measurement
of coating layer
thickness of fine and porous particles using focused ion beam. Particuology
measurement
of
coating
layer
thickness
of fine and
particlesofusing
focused
ion
beam. Particuology
Pietsch,
S.;42,
Peter,
A.; Wahl, P.; Khinast,
J.; Heinrich,
S. porous
Measurement
granule
layer
thickness
in a spouted
2019,
190–198.
2019,
42,
190–198.
bed
process
via optical coherence
tomography.
Technol. 2019,
356,Food
139–147.
[CrossRef]
5. coating
Dewettinck,
K.; Huyghebaert,
A. Fluidized
bed coating Powder
in food technology.
Trends
Sci. Technol.
1999,
5.
6.
6.
7.
7.
8.
8.
Dewettinck,
10, 163–168. K.; Huyghebaert, A. Fluidized bed coating in food technology. Trends Food Sci. Technol. 1999,
10,
163–168.
Watano,
S.; Imada, Y.; Hamada, K.; Wakamatsu, Y.; Tanabe, Y.; Dave, R.N.; Pfeffer, R. Microgranulation of
Y.; Dave,
R.N.;
Pfeffer,
R. Microgranulation of
Watano,
S.; Imada,
Y.; Hamada,
Wakamatsu,
Y.; Tanabe,
fine powders
by a novel
rotating K.;
fluidized
bed granulator.
Powder
Technol.
2003,
131, 250–255.
fine
powders
by
a
novel
rotating
fluidized
bed
granulator.
Powder
Technol.
2003,
131,
Jones, D.M. Factors to consider in fluid-bed processing. Pharm. Technol. 1985, 9, 50–55. 250–255.
Jones,
Factors
consider
fluid-bed
Pharm.
Technol.
9, 50–55.
Rieck, D.M.
C.; Bück,
A.; to
Tsotsas,
E. in
Monte
Carloprocessing.
modeling of
fluidized
bed1985,
coating
and layering processes.
Rieck,
C.;
Bück,
A.;
Tsotsas,
E.
Monte
Carlo
modeling
of
fluidized
bed
coating
and layering processes.
AIChE J. 2016, 62, 2670–2680.
Processes 2020, 8, 1525
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
18 of 18
Geldart, D. Types of gas fluidization. Powder Technol. 1973, 7, 285–292. [CrossRef]
Rieck, C.; Hoffmann, T.; Bück, A.; Peglow, M.; Tsotsas, E. Influence of drying conditions on layer porosity in
fluidized bed spray granulation. Powder Technol. 2015, 272, 120–131. [CrossRef]
Ennis, B.J.; Tardos, G.; Pfeffer, R. A microlevel-based characterization of granulation phenomena.
Powder Technol. 1991, 65, 257–272. [CrossRef]
To, D.; Dave, R.N. Fluid bed film coating of fine ibuprofen particles. Powder Technol. 2016, 290, 102–113. [CrossRef]
Mawatari, Y.; Koide, T.; Tatemoto, Y.; Takeshita, T.; Noda, K. Comparison of three vibrational modes (twist,
vertical and horizontal) for fluidization of fine particles. Adv. Powder Technol. 2001, 12, 157–168. [CrossRef]
Zhu, Q.; Li, H. Study on magnetic fluidization of group C powders. Powder Technol. 1996, 86, 179–185.
Ichikawa, H.; Tokumitsu, H.; Jono, K.; HUKUDA, T.; Osako, Y.; Fukumori, Y. Coating of pharmaceutical
powders by fluidized bed process, Part 6: Microencapsulation using blend and composite latex of copoly (ethyl
acrylate-methyl methacrylate-2-hydroxyethyl methacrylate). Chem. Pharm. Bull. 1994, 42, 1308–1314. [CrossRef]
Watano, S.; Nakamura, H.; Hamada, K.; Wakamatsu, Y.; Tanabe, Y.; Dave, R.N.; Pfeffer, R. Fine particle
coating by a novel rotating fluidized bed coater. Powder Technol. 2004, 141, 172–176. [CrossRef]
Chen, Y.; Jallo, L.; Quintanilla, M.A.; Dave, R.N. Characterization of particle and bulk level cohesion reduction
of surface modified fine aluminum powders. Colloids Surf. A Physicochem. Eng. Asp. 2010, 361, 66–80. [CrossRef]
Chen, Y.; Yang, J.; Mujumdar, A.; Dave, R.N. Fluidized bed film coating of cohesive Geldart group C powders.
Powder Technol. 2009, 189, 466–480. [CrossRef]
Hampel, N.; Roydeva, E.; Bück, A.; Tsotsas, E. Coating of finely dispersed particles by two-fluid nozzle.
Particuology 2018, 38, 80–93. [CrossRef]
Mezhericher, M.; Ladizhensky, I.; Etlin, I. Atomization of liquids by disintegrating thin liquid films using gas
jets. Int. J. Multiph. Flow 2017, 88, 99–115. [CrossRef]
Mezhericher, M.; Nunes, J.K.; Guzowski, J.J.; Stone, H.A. Aerosol-assisted synthesis of submicron particles at
room temperature using ultra-fine liquid atomization. Chem. Eng. J. 2018, 346, 606–620. [CrossRef]
Mezhericher, M.; Rieck, C.; Razorenov, N.; Tsotsas, E. Ultrathin coating of particles in fluidized bed using
submicron droplet aerosol. Particuology 2020. [CrossRef]
Lefebvre, A.H.; McDonell, V.G. Atomization and Sprays; CRC Press: Boca Raton, FL, USA, 2017.
Mezhericher, M.; Razorenov, N.; Mazor, G.; Ju, Y.; Stone, H.A. Submicron aerosols of liquid fuels: Method of
production, experimental characterization and a semi-empirical model. Appl. Energy 2019, 235, 1651–1663.
[CrossRef]
Sondej, F.; Bück, A.; Tsotsas, E. Comparative analysis of the coating thickness on single particles using
X-ray micro-computed tomography and confocal laser-scanning microscopy. Powder Technol. 2016, 287, 330–340.
[CrossRef]
Chen, Y.; Yang, J.; Dave, R.N.; Pfeffer, R. Fluidization of coated group C powders. AIChE J. 2008, 54, 104–121.
[CrossRef]
Dewettinck, K.; Huyghebaert, A. Top-spray fluidized bed coating: Effect of Process variables on coating
efficiency. Lebensm. Wiss. Und Technol. 1998, 31, 568–575. [CrossRef]
Liu, L.X.; Litster, J.D. Spouted bed seed coating: The effect of process variables on maximum coating rate
and elutriation. Powder Technol. 1993, 74, 215–230. [CrossRef]
Takei, N.; Unosawa, K.; Matsumoto, S. Effect of the spray-drying process on the properties of coated films in
fluidized bed granular coaters. Adv. Powder Technol. 2002, 13, 333–342. [CrossRef]
Du, J.; Bück, A.; Tsotsas, E. Influence of process variables on spray agglomeration process in a continuously
operated horizontal fluidized bed. Powder Technol. 2020, 363, 195–206. [CrossRef]
Jiang, Z.; Bück, A.; Tsotsas, E. CFD–DEM study of residence time, droplet deposition, and collision velocity
for a binary particle mixture in a Wurster fluidized bed coater. Dry. Technol. 2018, 36, 638–650. [CrossRef]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional
affiliations.
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).