Received: 25 July 2024 Revised: 3 September 2024 Accepted: 26 September 2024

DOI: 10.1002/cjce.25526

REVIEW ARTICLE

A comprehensive review of cyclone separator technology

Ece Aylı | Eyup Kocak

Department of Mechanical Engineering,

Cankaya University, Ankara, Turkey Abstract
This review article examines the working principles, optimal dimensions,
Correspondence
effects of key parameters, and the results of experimental/numerical studies on
Ece Aylı, Department of Mechanical
Engineering, Cankaya University, Ankara, cyclone separators. Investigations have been conducted on the effects of
Turkey. parameters such as vortex finder diameter, conical part diameter, cyclone sepa-
Email: [email protected]
rator diameter, cylinder height, inlet height, inlet width, vortex finder length,
and cyclone total length on efficiency, performance, and pressure drop.
Furthermore, the article explores current modifications and efforts to improve
efficiency. These modifications include adding water nozzles, inserting ribs,
employing double-stage cyclones, incorporating additional inlets, using finned
cylinder bodies, adding extra top inlets, introducing liquid jets, employing heli-
cal roof inlets, adding laminarizers, incorporating internal spiral vanes, and
employing slotted vortex finders. While serving as a guide to optimize the
design and performance of cyclone separators, this article emphasizes new and
innovative approaches to enhance their industrial applicability. By compiling
studies conducted from conceptual birth to the present, the aim of this article
is to serve as a guidebook.

KEYWORDS
CFD, cyclone separator, DEM, experiment, geometric modification, particle, pressure drop,
separation efficiency

1 | INTRODUCTION can function safely for lengthy periods of time without

requiring maintenance. Particles such as dust, contami-
Cyclone separators are utilized in numerous processes nants, and other particles carried in gas streams during
where separation and filtration are requisite. Particularly industrial processes can damage equipment and the
prevalent in gas–solid particle separation, this mechanism surrounding environment. Applications for cyclone separa-
has been frequently employed since the 1930s for its tors include abrasive processing; solid fuel processing;
low operating costs, longevity, and adaptability to harsh grinding, screening, and separating processes; dusty chemi-
conditions, making it a commonly employed method in cal, pharmaceutical, and cosmetic operations; packaging
particle separation. Cyclone separators are devices that use machines; food processing and hygienic facilities; paper
the spin effect to create centrifugal airflow inside them, industry; and PVC stabilizers, among others. Cyclone sepa-
which allows for the separation of airborne particles. rators effectively eliminate particulates from gas streams,
The process of centrifugal separation causes the particles to lowering pollution levels, increasing equipment longevity,
lose their momentum, drop to the tank bottom, and then and enhancing energy economy. They are also a low-cost,
be released, while the air rises once more to resume its low-maintenance option that can be applied to a variety of
intended course. Cyclone separators are a preferred option situations due to their lack of moving parts and ability to
since they are less expensive than industrial filters and function in challenging environments.

Can J Chem Eng. 2024;1–39. wileyonlinelibrary.com/journal/cjce © 2024 Canadian Society for Chemical Engineering. 1
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 AYLI and KOCAK

In 1917, Strutt[1] laid the groundwork for fundamen- was more accurate compared to others. Instead of focus-
tal definitions regarding the stability of rotating flows. In ing on a central core like Barth’s theory,[11] the Leith–
the absence of axial and radial motions, Strutt expressed Licht theory[12] calculates efficiency by assuming dust
the equilibrium requirement as the rotational velocity ‘v’ particles are spread evenly throughout the cyclone. It
being solely a function of radius ‘r’. Taylor[2–4] analyti- then considers the average amount of time particles
cally elucidated the formation of symmetric vortex rings. spend inside (residence time) to determine how effec-
These studies also revealed the tendencies of solid parti- tively they are collected.
cles within the fluid to rotate. Lippmann and Chan[13] conducted tests with three
In 1949, Linden[5] provided a detailed presentation different commercially available cyclone separators. They
of the experimental studies conducted on cyclone sepa- investigated the relationship between cyclone collection
rator design prior to the 1939–1945 war at the laboratory performance and dust concentration.
of the Technical High School at Delft. The aim of this Beeckmans and Kim,[14] in one of the first experimen-
study is to better understand the gas flow of cyclone sep- tal works in the literature, investigated whether the
arators, which have become increasingly popular and Reynolds number is included in the analytical equations
newly used in recent years, to determine the most effi- of cyclone performance. Within the range of variables
cient shape of a cyclone, and to identify the type of dust covered by the experimental data, it was concluded that
recommended for a cyclone collector. It was found in the flow fields within the cyclone depend on the Reyn-
this study that the use of blades, spiral ducts, or passive olds number, and therefore, contrary to the assumptions
methods to rotate gases adversely affected efficiency. of most theoretical models, cyclone efficiency is not solely
Therefore, it is desirable to equip the cyclone with a dust a function of Stokes number.
bunker and an exhaust pipe where the gases can con- Lee et al.[15] conducted experiments to see how gas
tinue their rotation. density affects cyclone performance. Their findings
In 1951, Stairmand[6] determined that in order to showed that cyclones work better (higher efficiency) with
maximize cyclone efficiency, a high-efficiency cyclone denser gases. They also tried to develop a formula (empir-
with a standard design would handle approximately half ical correlation) based on their data, which involved com-
the amount of gas compared to a high-throughput design paring the size of particles that get separated efficiently
of the same diameter and overall height, operating at the (cut size) with a property called the Reynolds number.
same pressure drop. Kim and Lee[16] conducted an experimental study
In 1962, Smith[7] examined the theory of rotating in 1990 to investigate the particle collection character-
fluids dating back to 1919 in the context of cyclone sepa- istics of small cyclones. They demonstrated that geo-
rators. As a result of the study, the large angular momen- metric parameters have a significant impact on cyclone
tum gradients at the core surface confer upon it performance.
properties akin to a free surface, including weir flow and In a 1996 study, Griffiths and Boysan[17] used compu-
wave propagation. Wave motions within the core eluci- tational fluid dynamics (CFD) to model cyclone perfor-
date how the core flow can be highly unsteady without mance and compared it to real-world experiments. They
turbulent mixing. looked at three existing theories to predict cyclone effi-
Leith and Mehta[8] extensively compared the methods ciency. Out of these, Barth’s theory[11] worked well for
for calculating pressure drop and total efficiency of two small cyclone designs, but only if their shapes and
cyclone separators in the literature. They concluded that sampling rates were similar to a standard version. For the
the methods proposed by Alexander in 1949[9] and First biggest cyclone, only data from specific experiments by
in 1950[10] were superior to those by Barth[11] and Stair- Iozia and Leith[18] accurately reflected real-life perfor-
mand.[6] Barth’s theory simplifies cyclone performance mance. Overall, using existing performance data from
by imagining a hollow core running from the bottom to similar cyclones was not very reliable.
the outlet. Particles are categorized based on their size When examining the historical development of
compared to a critical size at the edge of this core. Smal- cyclone separators, we can discern significant milestones
ler particles experience a drag force that pushes them that represent key advancements:
towards the core, eventually exiting the cyclone. Larger
particles, due to their greater mass, experience a stronger • 1910s: Conceptual birth: The foundation for cyclone
centrifugal force flinging them outwards and collecting separators was laid in this era with the establishment
them on the cyclone wall.[11] Based on a limited compari- of the theory of cyclonic flow. This theory elucidates
son between experimental data and theoretical calcula- the fundamental principles governing the swirling
tions, they also found that the method proposed by Leith motion of gas or liquid within the separator, which is
and Licht[12] for predicting cyclone collection efficiency essential for particle separation.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 3

• 1950s: Theory into practice: This period saw the emer- The particles they handle can be abrasive, wearing down
gence of theoretical models for cyclone separator per- the separator over time. Choosing the right model is
formance. These models were subsequently validated important, as a poorly selected one may not perform well.
through rigorous experimental studies, solidifying the They also have slightly higher pressure losses compared
theoretical understanding with empirical evidence. to other separation methods.
• 1980s: The rise of CFD: The 1980s witnessed a pivotal This ongoing pursuit of knowledge and innovation
moment with the introduction of CFD simulations. ensures that cyclone separators remain a dependable and
These simulations enabled researchers and engineers efficient technology for particle separation across a wide
to analyze the intricate internal flow dynamics of range of industrial applications. Based on the data
cyclone separators virtually. This facilitated the optimi- obtained from Web of Science, the statistics of academic
zation of designs and improvements in performance studies on cyclone separators from 1981 to the present
that were previously challenging or unattainable using are presented in Figure 1. As seen, there is a significant
conventional methods. increase in the number of studies on the subject since the
• 1990s–Present: Continued evolution: Since the incep- 2010s. The primary reason for this increase is the possi-
tion of CFD, research on cyclone separators has bility of modelling solid particles using CFD methods.
remained robust. The focus has shifted towards the CFD assists designers and researchers in optimizing per-
following: formance by simulating the complex internal fluid

Enhanced design processes: Exploring new geome- dynamics of cyclone separators.
tries and configurations to achieve greater separa- The main objective of this study is to compile aca-
tion efficiency while minimizing pressure drop demic research related to cyclone separators and facilitate
within the separator. access to information on the subject. This will enable

Advanced techniques in CFD: Investigating the appli- researchers and designers to easily access up-to-date
cability of sophisticated methods such as two-phase information on the fundamentals, design, optimization,
flow models (which consider both gas and particle and applications of cyclone separators.
behaviour) in CFD applications. However, the accu-
racy of these simulations depends on the precise
modelling of the complex interactions between parti- 2 | WORKING PRINCIPLE OF
cles and fluid flow. CYCLONE S EPARATORS

Pressure drop management: Ongoing research aims
to predict and control pressure drop under various In cyclones, the gas and particle mixture is subjected to a
operating conditions. Accurate pressure drop con- spinning motion. This rotational movement causes the
trol is crucial for optimizing energy consumption particles to be thrown towards the wall. This throwing
and overall system efficiency. motion makes the particles heavier than air. The
increased weight of the particles also increases the mini-
In addition to their simple designs, ease of fabrication mum gas velocity required for them to remain airborne.
using readily available industrial materials, and low This minimum gas velocity is known as the saltation
maintenance costs, cyclones also have drawbacks such as velocity. The increase in the virtual mass of the particles
the potential impact of harsh and abrasive conditions on makes it difficult to keep them airborne. Therefore, the
surface roughness over time. Furthermore, the impor- particles fall out of the air and settle at the bottom of
tance of selecting an appropriate model for cyclones the cyclone. The clean air, on the other hand, exits from
cannot be overstated, as failure to do so may result in a separate tube at the top of the cyclone.
improper operation. A cyclone consists of an outlet, a body part, and an
Cyclone separators offer several benefits: they are inlet, which creates the necessary swirling motion to cre-
simple, efficient at separating solids from liquids, and can ate the vortex inside this as illustrated in Figure 2. The
also be used for mixing. They are affordable to produce rotation can be initiated by introducing the gas flow tan-
and maintain, and in some cases, the separated particles gentially into the unit or axially through a series of swirl
can be reused. They can withstand high temperatures vanes. Once entrained in the vortex, the particles are sep-
and pressures, and are resistant to certain chemicals. arated from the gas stream by centrifugal force. The
Their lack of moving parts and simple design make them denser particles tend to be entrained towards the outer
reliable and easy to maintain. They can be coated to pre- walls of the cyclone body. As shown, gravitational forces
vent corrosion, and are made from common materials move the particles towards the discharge section at the
like aluminium, plastic, steel alloys, and cast iron. bottom of the unit. Lighter and relatively particle-free
However, cyclone separators also have some drawbacks. gases reach the base of the vortex.[19]
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 AYLI and KOCAK

F I G U R E 1 Distribution of the
number of articles by year (generated
using the Web of Science database).

the centrifugal force in this region are also intensively

investigated. Moreover, methods for enhancing the
centrifugal effect by making geometric modifications
within this region have become a prominently studied
topic in recent years.
3. Particles and/or droplets are found in the outer layer
and gas in the inner layer, resulting from density dif-
ferences. Clean gas gathers to the centre portion dur-
ing the downward rotational movement.
4. The cleaned air is discharged through the outlet pipe.
The impact of the dimensions of this pipe on perfor-
mance is also a researched topic.
5. Particles and/or droplets settle to the bottom with the
gas flow. The region where they are collected is called
the dust hopper, and there are intensive studies on
the dust hopper as well.

In Bohnet’s study,[21] the flow inside the cyclone is

[20]
FIGURE 2 Velocity profiles inside of the cyclone separator. divided into five main regions. First, the boundary layer
flow occurs at the cover plate and the outer side of the
outlet pipe. The second region is the entrance region,
If the operating principle is described step-by-step, it which lies below the primary region. The third area is the
can be summarized as follows: downstream flow region formed along the wall edges.
The fourth region is the re-entrainment region before the
1. Air containing solid particles enters axially through hopper. The fifth region is the upstream flow region for
an inlet. The axial inlet creates a high rotational the air separated from the particles. The regions are illus-
speed. Studies on the effect and enhancement of this trated in Figure 3.
axial speed on inlet geometric configurations are Safikhani and Mehrabian[22] visually depicted flow
intensively conducted in the literature. Additionally, profiles within the cyclone separator through numerical
the effect of particle size is also a prominent focus of simulations (Figure 4). Similar to previous studies, the
research. velocity field comprises three primary velocity compo-
2. Second, the particle-laden flow entering the system nents: circumferential, axial, and radial. Due to the
axially is thrown to the wall due to the strong centrifu- strong tendency of the flow to swirl, the circumferential
gal force field and rotates downward along the cylin- velocity component is more significant compared to the
der wall. Geometric parameters’ effects on increasing axial and radial components. Peng et al.,[23] in their
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 5

FIGURE 3 Five regions of the flow area for cyclone

separator.[21] FIGURE 4 Velocity vectors for the cyclone separator on the
mid-plane[22]

study, argued that the radial velocity component could be

neglected since it is much lower than the axial and cir- where vθ represents the tangential velocity, C is a con-
cumferential velocities. The first definition of these veloc- stant, and r denotes the radial distance from the cyclone
ity components was made by Barth.[11] Barth’s model[11] centreline. For ideal forced vortex flow (or ‘solid body
and the studies of other researchers building upon it[23,24] rotation’), n = 1, and for ideal free (or ‘loss-free’) vor-
separated the velocity components as axially constant tex flow, n = 1. The axial velocity is highest near the cen-
tangential and radial velocities on the surface dividing tre of the cyclone and gets slower as it approaches the
the inner and outer vortices. outer walls. This creates two zones: a fast-moving inner
Due to the strong swirling nature of the flow, the tan- region called the ‘inner vortex’ and a slower-moving
gential velocity component holds more significance com- outer region sometimes called the ‘outer vortex’. The
pared to the axial and radial components. The radial boundary between the fast-moving inner zone (inner
component is comparatively smaller and can be disre- vortex) and the slower-moving outer zone (outer vortex)
garded in fundamental calculations.[25] is estimated by imagining a cylinder extending down-
The tangential velocity distribution exhibits two wards from the wall that guides the swirling flow (vortex
distinct regions: an outer area characterized by quasi- finder wall). This imaginary cylinder is assumed to mark
free vortex flow and encircling an inner zone of the area where the air flow speed becomes zero. In
quasi-forced vortex flow. This vortex type is com- Figure 5, Peng et al.[23] illustrate the velocity plots that
monly known as a ‘Rankine’-type vortex. The velocity are obtained with the numerical methods.
distributions in both regions can be described as Song et al.[27] conducted a numerical analysis of a
follows[23,25,26]: classic cyclone separator using the Reynolds stress model.
Figure 5 presents visualizations of the tangential velocity,
C radial velocity, axial velocity, static pressure, and turbu-
vθ ¼ ð1Þ
rn lence intensity. As shown in Figure 6, the high-speed gas
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 AYLI and KOCAK

entering the flow area at point A increases to 1.5–2 times restriction of the vortex finder and the collision between
the inlet velocity at point A. The velocity subsequently the inlet flow and the internal rotating flow.
decreases as the gas rotates downward along the wall. It is observed that, similar to Peng et al.,[23] the dom-
When examining the radial velocity distribution, positive inant velocity is the tangential velocity. The swirling
values occur on one side and negative values occur on motion within the cyclone impels particles towards the
the other side of the symmetry axis. The radial velocity wall due to centrifugal forces. The cross-section indi-
significantly changes below the inlet at point B due to the cates that the tangential velocity exhibits a composite
vortex structure, where the outer vortex is nearly free
and the inner vortex is forced. A change in velocity with
positive values on one side and symmetric negative
values on the other side is observed at one edge of the
cyclone separator, indicating axial symmetry.
In a study by Gao et al.,[28] the researchers visual-
ized airflow patterns using different numbers of sam-
pling points. Their findings, illustrated in Figure 7,
show that the flow patterns became clearer (with more
defined inner and outer vortex structures) as the num-
ber of sampling points increased. As the airflow moved
through the cone-shaped section, the narrowing walls
caused the speed to increase. The pressure on the outer
wall also rose steadily as the cone section narrowed. In
the centre of the cone, a low-pressure zone formed. This
pressure difference between the outer high pressure
and the inner low pressure pushed the airflow towards
the centre, where it turned upward at the bottom. This
F I G U R E 5 Velocity profiles obtained with computational fluid upward flow then spiralled continuously until it exited
dynamics (CFD).[23] through the vortex finder.

F I G U R E 6 (A) Tangential velocity (positive: clockwise, negative: counter-clockwise), (B) Radial velocity (positive: Outward, negative:
inward), (C) Axial velocity (positive: Upward, negative: downward), (D) Static pressure, and (E) Turbulence intensity within the cyclone
separator.[27]
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 7

F I G U R E 7 The stream-traces
diagram with different numbers of
samples.[28]

Cyclone separators gained the concept of a precessing vortex core diameter by determining the values of rota-
vortex core (PVC) thanks to Hoekstra et al.[29] The PVC tional velocity in the radial direction for various sections.
is a hydrodynamic instability observed in strongly swir- Another important aspect in cyclone separator studies
ling flows. PVC is a vortex core located at the centre of is expressed as the total separation efficiency or collection
the flow that exhibits a type of predetermined movement efficiency. Collection efficiency is calculated as the ratio of
pattern while rotating. This instability causes the vortex the amount of dust captured in the dust collector to the
core to move with a slight precession, leading to a regular amount of dust entering the system. This allows us to gain
variation in the flow. This formation is observed in information about cyclone collection efficiency. However,
cyclone separators. In a cyclone, the core region of the the sizes of dust particles are also among the most impor-
flow is particularly strongly affected by the PVC. The pri- tant parameters affecting collection efficiency. Therefore,
mary reason for this is the large magnitudes of the tan- when making calculations, it is important to try to calcu-
gential and axial velocities. PVC also influences the mean late at the same dust particle sizes as much as possible. In
velocity profiles. Later research by Derksen and col- general, as the dust particle size decreases, a decrease in
leagues[30,31] revealed that the airflow within the cyclone collection efficiency is expected. This is because dust parti-
is not perfectly steady but rather fluctuates in a predict- cles of very small size leave the cyclone without touching
able, repeating pattern. This fluctuation shows up as a any point in the cyclone separator or cyclone. The collec-
peak in the analysis of the flow behaviour. tion separation formula can be given as follows and is
Dehdarinejad et al.[32] tested six different types of valid only if all the injected particles are identical[33]:
cyclones (C1 to C6). In their study, they used helical
guide vanes with different pitch length. Although the mass of particle inside the dustbin
ηcollection ¼  100 ð2Þ
position of the maximum magnitude of the total velocity mass of total particle
(or rotational velocity) was different, it was observed that
for each case, 1.1D, 1.5D, and 2.5D (D is vortex diameter) The fundamental parameters affecting the performance
were the same on three cross-sections. The Rankine vor- of cyclone separators can be listed as cyclone dimensions,
tex, including the inner forced vortex and outer free vor- properties of the fluid used, and characteristics of the parti-
tex, are identified in same manner for all of the cles. Figure 8 presents the fundamental parameters
geometries. Therefore, it is possible to calculate the influencing the performance of cyclone separators.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 AYLI and KOCAK

FIGURE 8 Fundamental parameters that affect the cyclone separator performance.

3 | INVESTIGATION OF THE these parameters on pressure drop and efficiency has

E F F E C T OF G E OM E TR I C AL been intensively studied in the literature over the past
P A R A M E T E R S ON CY C L O N E 25 years.
SEPARATOR PERFORMANCE

Although the geometry of cyclone separators is simple, 3.1 | The effect of cyclone separator inlet
the flow within the separator is quite complex. Cyclone geometry
geometry is generally visualized with the same set of
parameters in the literature (Figure 9). Typically, a is 3.1.1 | a/b effect
the inlet height, b is the inlet width, D is the vortex
diameter, S is the total vortex finder length, h is the In recent years, there has been a surge in research on
cylinder length, Ht is the cyclone total length, and Bc how the design of cyclone separators, especially the
is the cone tip diameter are named. The effect of all inlet design pioneered by Alexander,[9] affects how well
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 9

FIGURE 9 A typical cyclone cylinder and parameter representation.[34]

they perform. Scientists have done numerous experi- where there is also a relationship between De/D, a/b
ments and computer simulations to examine how factors ratio, and the vortex length[40]:
like inlet size, air-particle speed at the inlet, and the

2
number of inlets influence cyclone performance. These Lc De D2
studies have mainly focused on comparing how much ¼ 2:3 ð3Þ
D D ab
pressure is lost (pressure drop), how well particles are col-
lected (collection efficiency), and overall efficiency. This
2:25
2 0:361
Lc De D
section reviews some recent studies on how the design of ¼ 2:4 ð4Þ
D D ab
cyclone separator inlets affects their performance.[35–37]
Elsayed and Lacor[34] investigated the effect of changing
the b/a ratio numerically on performance. As cyclone In their numerical study, Qian and Zhang[41]
dimensions increased, the maximum tangential velocity observed that as the diameter of the vortex finder
decreased. Furthermore, as both cyclone inlet width increased, the length of the vortex formed inside the sep-
and height increased, the maximum tangential velocity arator also increased. However, they noted a tendency for
decreased. The optimum b/a ratio was found to be the natural vortex length to decrease when the diameter
between 0.5 and 0.7. Yang et al.[36] observed that decreas- of the vortex finder reached its maximum value point.
ing the inlet dimensions reduces the maximum efficiency Gao et al.,[28] Yang et al.,[36] Ji et al.,[40], and Bryant
value, while on the other hand, Elsayed and Lacor[34] et al.[42] have also demonstrated a dominant relationship
observed that increasing the inlet dimensions decreases between vortex length and inlet dimension.
the total efficiency. Wang et al.[38] concluded that the Table 1 has been prepared to summarize all the stud-
cyclone pressure drop is independent of the cyclone ies reviewed in this section. This table allows for a quick
diameter. overview of the geometric modifications made by each
In the literature, the natural vortex length is defined researcher and the significant effects observed as a result.
as the distance between the bottom end of the vortex
finder and the section where the free vortex zone disap-
pears.[39] Alexander[9] has demonstrated that the natural 3.1.2 | Vortex finder dimension effect
vortex length is a function of the De/D ratio, as depicted
in correlation given by Equation (3). Another correlation Wei et al.[43] varied the vortex finder diameter for a con-
is obtained by Ji et al.,[40] as shown in Equation (4), figuration with a cyclone length of 1080 mm between
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 AYLI and KOCAK

T A B L E 1 Summary of the research works that investigate the effect of cyclone separator inlet geometry on the cyclone separator
performance.

The effect of cyclone separator inlet geometry

References Modification Effect on the performance

Alexander[9] Inlet size and air-particle speed at the Natural vortex length is a function of the De/D ratio.
inlet are varied.
Elsayed and Effect of changing the b/a ratio is The optimum b/a ratio was found to be between 0.5 and 0.7.
Lacor[34] studied.
Yang et al.[36] Inlet dimensions are varied. Decreasing the inlet dimensions reduces the maximum efficiency.
[38]
Li and Parnel Cyclone diameter is varied. Pressure drop is independent of the cyclone diameter.
Qian and Diameter of the vortex finder is Diameter of the vortex finder increased the length of the vortex formed
Zhang[41] varied. inside the separator.
Yang et al.[36] Inlet dimensions are varied. Dominant relationship between vortex length and inlet dimension
[40]
Ji et al. Inner dimensions are varied. The vortex length is a function of the distance between the bottom end of the
vortex finder and the section at which the free vortex zone disappears.
Gao et al.[28] Vortex motion in a reverse Stairmand At the base of the vortex finder, an uneven ring-shaped region forms around
cyclone separator is examined. the wall, representing the vortex distribution of the longitudinal circulation.

115 and 150 mm. Additionally, they examined the varia- gas cyclones can lead to improved performance with
tion of the vortex finder diameter in cases with different identical geometry. In another study by the same group,
inlet dimensions. As shown in Figure 10, when the Der they employed genetic algorithm and multi-objective
value (vortex finder diameter to the cyclone diameter) is optimization techniques to investigate the converging–
increased to 0.4, stagnation occurs in the internal region diverging vortex finder. As a result of this study, better
of the vortex finder and the upper region of the cyclone performance than the Stairmand cyclone was once again
cylinder. If the Der value is further increased, first to achieved.[46]
0.467 and then to 0.5, stagnation becomes apparent in all Researchers explored how changing the lengths of the
regions of the vortex finder, the cylinder, and the cone of overflow pipe (vortex finder) and underflow pipe together
the cyclone (the entire main regions). affects a cyclone’s performance. They defined a split ratio
El-Batsh[44] aimed to optimize the vortex finder value (F) to measure how much of the incoming flow exits
numerically to minimize pressure drop and maximize from the bottom (underflow). Their results showed that a
efficiency. They utilized a Eulerian–Lagrangian numeri- longer vortex finder increased the tangential flow within
cal approach. According to their results, increasing the the cone. This stronger swirling motion is believed to be
vortex finder diameter, which does not have a dominant a key reason why longer vortex finders improve separa-
effect on pressure drop, leads to a decrease in efficiency. tion efficiency.[47]
Dehdarinejad et al.[32] transformed the vortex finder In their study, Ficici et al.[48] experimented with three
geometry into a converging–diverging shape and altered different vortex finder diameter values. When they pro-
its dimensions to numerically investigate its effects on cessed the experimental results using the statistical
cyclone pressure drop and energy loss. They positioned regression method, they observed that an increase in vor-
the vortex finder with a standard geometry in the form of tex finder diameter resulted in an increase in pressure
an inverted cone within the flow area. The contours loss. However, El-Batsh[44] achieved the opposite result
depicted in Figure 11 indicate that the tangential velocity in the literature. An increase in the exit pipe diameter
in the dustbin region is lower in the case of the reduces pressure drop. Additionally, it does not have a
converging–diverging vortex finder compared to the con- dominant effect on efficiency.
ventional vortex finder. This discrepancy could potentially In some similar studies,[33,49,50] it is claimed that the
hinder the movement of small particles into the main vortex finder diameter has a dominant effect on the col-
flow, thereby reducing cyclone separation efficiency. lection efficiency, and the decrease in the De value leads
Kumar and Jha[45] conducted a similar study investi- to higher energy loss, that is, higher pressure drop. In
gating the effects of converging–diverging vortex finder the studies conducted by Zhang et al.[33] and Raoufi
numerically. They concluded that using converging– et al.,[49] they examined the effects of vortex finder
diverging instead of uniform diameter vortex finders in diameter on tangential velocity. Despite the completely
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 11

F I G U R E 1 0 Axial velocity
distribution patterns across various
main regions of cyclones with diverse
vortex finder diameters.[43]

FIGURE 11 Contours of tangential velocity normalized by inlet velocity along with streamlines for various scenarios.[32]

different conditions between the two studies, similar the letter ‘M’. While the change in De does not alter the
velocity distributions were obtained, as shown in velocity distribution trend, the lowest tangential veloc-
Figure 12. Symmetric profiles are formed around the ity values were obtained at the highest diameter values
symmetry centre, resulting in a distribution resembling in both studies.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 AYLI and KOCAK

F I G U R E 1 2 Tangential velocity distribution for different vortex finder diameters obtained from the studies of (A) Zhang et al.[33] and
(B) Raoufi et al.[49]

Table 2 has been prepared to summarize the studies cyclone structure provides high efficiency. In another
reviewed in this section, offering a quick overview of the study, the number of inlets was increased to five, and the
vortex finder diameter effect made by each researcher inlet angles were varied as 15
, 32
, 45
, 61
, and 72
. A
and the significant effects observed as a result. As seen linear relationship between angle and efficiency was not
from the results, studies in the literature often show simi- found, with the efficiency reaching its maximum value at
lar modifications but yield different, sometimes opposite, a radial angle of 61
. As the velocity value increased, the
results. The existence of conflicting views in the literature efficiency value also increased.[55]
regarding the effects of key parameters on cyclone effi- In the study by Khazaee,[56] it was observed that
ciency suggests that the results obtained may not be uni- increasing the number of cyclone inlets also increased
versally applicable. This situation highlights the need for the efficiency value. It was also found that the rectangu-
further research to clarify these inconsistencies. lar inlet had a collection efficiency 4%–6% higher than
the elliptical inlet. Similarly, Chen et al.[57] also demon-
strated that the efficiency of the two-stage inlet cyclone
3.1.3 | Multi-inlet effect separator is higher compared to the single inlet cyclone.
In another study, the location of the added new inlet has
Wang et al.[51] experimentally investigated the efficiency been modified. This secondary inlet was confined to a
and pressure drop of single and double inlet cyclone sep- region common to both the vortex finder tube and the
arators. According to the results, increasing the vortex cylindrical section. Similarly, in this system, higher effi-
finder diameter reduces the collection efficiency and ciency was calculated as well.[58] Many researchers have
pressure drop. In experiments comparing single and dual observed as a common consensus that the dual-inlet
inlet cyclone designs, the dual inlet cyclone outperformed structure outperforms the single-inlet in terms of perfor-
the single inlet cyclone in both collection efficiency mance. In our literature review, we found that all studies
(higher) and pressure drop (lower). Notably, the dual yielded the same result, and no research claiming other-
inlet cyclone achieved a collection efficiency exceed- wise was encountered.[59–61]
ing 95%.[52] Table 3 has been prepared to summarize the studies
Lim et al.[53] concluded that, similarly, the collection reviewed in this section, offering a quick overview of the
efficiency of double inlet cyclones is higher compared to multi-inlet cyclone effect made by each researcher and
single inlets. Another research group, Zhu et al.,[54] the significant effects observed as a result. Based on the
experimentally observed in their studies that the dual results, the research in this area is consistent, and
inlet structure reduces pressure drop. Especially at high the overall conclusion is that a dual inlet system is more
flow rate values, the study also notes that the dual-inlet efficient compared to a single inlet system.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 13

3.2 | The effect of cylinder and cone study to investigate the effect of cylindrical and conical
height height variations on the flow field. In the cyclone config-
uration with a diameter of D, the body height was varied
Research on cylinder/cone height has been ongoing between D and 2D, and the conical height was
since the 1950s, but it has gained momentum in recent varied between 2D and 3D. The study emphasizes that
years.[62,63] Demir et al.[64] conducted an experimental the body and conical heights have a dominant effect on

TABLE 2 Summary of the research works that investigate the effect of vortex finder dimension on the cyclone separator performance.

The effect of vortex finder dimension

References Modification Effect on the performance

[43]
Wei et al. Vortex finder diameter is varied. As the vortex finder diameter to the cyclone diameter value increases to 0.4,
stagnation begins in the vortex finder. Further increasing it results in
stagnation throughout the vortex finder, cylinder, and cone regions.
El-Batsh[44] Vortex finder diameter is varied. Increasing the vortex finder diameter, which does not have a dominant
effect on pressure drop, leads to a decrease in efficiency.
Dehdarinejad The vortex finder geometry turned to Tangential velocity in the dustbin region is lower in the case of the
et al.[32] converging–diverging shape. converging–diverging vortex finder.
Kumar and Jha[45] Effects of converging–diverging Using converging–diverging instead of uniform diameter vortex finders in
vortex finder. gas cyclones can lead to improved performance.
Kumar and Jha[46] Effects of converging–diverging Better performance than the Stairmand cyclone.
vortex finder.
Zheng et al.[47] The lengths of the overflow pipe are Longer vortex finder increased the tangential flow within the cone.
varied.
Ficici et al.[48] Different vortex finder diameter Increase in vortex finder diameter resulted in an increase in pressure loss.
values.
Zhang et al.[33] Vortex finder diameter is varied. Change in De does not alter the velocity distribution trend; the lowest
tangential velocity values were obtained at the highest diameter values.
Patterson and Vortex finder diameter is varied. Dominant effect on the collection efficiency, and the decrease in the De
Munz[50] value leads to higher energy loss.

TABLE 3 Summary of the research works that investigate the effect of multi inlet on the cyclone separator performance.

The effect of multi-inlet cyclone

References Modification Effect on the performance

[51]
Wang et al. Single and double inlet cyclone separators Increasing the vortex finder diameter reduces the collection
comparison. efficiency and pressure drop.
Lee and Single and double inlet cyclone separators The dual inlet cyclone achieved a collection efficiency
Yook[52] comparison. exceeding 95%.
Lim et al.[53] Single and double inlet cyclone separators The collection efficiency of double inlet cyclones is higher
comparison. compared to single inlets.
Zhu et al.[54] Single and double inlet cyclone separators Dual inlet structure reduces pressure drop.
comparison.
Fetuga et al.[55] The number of inlets was increased up to 5 and A linear relationship between angle and efficiency was not
inlet angles are varied. found.
Khazaee[56] Number of cyclone inlets is varied. The collection efficiency increases with the increment in the
cyclone inlet number.
Chen et al.[57] Single and double inlet cyclone separators Two-stage inlet cyclone separator is higher compared to the
comparison. single inlet cyclone.
Wasilewski and Location of the added new inlet has been Higher efficiency was calculated with changing the location.
Brar[58] modified.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 AYLI and KOCAK

F I G U R E 1 3 The tangential velocity profile inside the cyclone cone at different axial from left to right: Hb = D, 1.5D, 2D; from top to
bottom: Hc = 2D, 2.5D, 3D. Distances from the bottom, with an inlet velocity of 18.5 m/s.[64]

pressure drop, suggesting that the body height should be efficiency. In a study by Pandey et al.,[66] increasing the
less than 1.5D and the conical height should be less than cylinder height (H/D ratio) from 0.5 to 1.5 had a slight
2.5D for optimal performance. Raising the cyclone height negative effect on collection efficiency at a constant inlet
results in decreased tangential velocities along the cylin- velocity (Uin). The CC variant showed a marginally better
der, consequently resulting in a reduction in pressure efficiency compared to the standard ST variant for all H/D
drop as given in Figure 13. ratios. The CV variant exhibited the lowest efficiency over-
Another research group[65] numerically varied the H/D all. Details of the geometric configurations and the corre-
ratio between 1.05 and 4. Similarly to the results of Demir sponding efficiency curves can be found in Figure 14 of
et al.,[64] the increase in cylinder height significantly con- the referenced study (the CC, CV, and ST configurations
tributes to the pressure drop. Furthermore, the increase can be seen in Figure 14).
in the length of the cylindrical segment reduces the Brar et al.[67] claimed that increasing the cylinder
rotational speed, leading to a decrease in the collection height by 5.5 times the diameter improved the pressure
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 15

FIGURE 14 Models of the cyclone in the work of Pandey et al.[66] and separation efficiency distributions.

drop and enhanced efficiency by 9.5%. Increasing the Table 4 has been prepared to summarize the studies
cone length provided an improvement in efficiency. reviewed in this section, providing a quick overview of
Hoffmann et al.[68] discovered that lengthening a cyclone the effects of cylinder and cone height modifications
cylinder initially boosts its performance significantly. made by each researcher and the significant results
However, they also found that extending the cylinder observed. Based on the results, the research in this area is
beyond a certain point leads to a sharp decline in separa- consistent, with the overall conclusion being that a dual
tion efficiency. Some argue that lengthening the cylinder inlet system is more efficient compared to a single inlet
beyond the core of the swirling air (vortex length) offers system.
no benefit. They propose designing the cylinder based on
the vortex length itself. Ito et al.[69] found a connection
between the natural vortex length and the overall cyclone 3.3 | The effect of cyclone diameter
length, and Akiyama and Marui’s research supports this
concept.[70] Shastri et al.,[71] while keeping the total Brar and Sharma[73] numerically varied the diameter of
cyclone length constant, varied the lengths of the cylin- the cyclone as 0.8D, 1D, and 1.2D. They observed that
der and cone. The Hc/D ratio was varied between 0.5 and while the pressure drop values increased with the
2.5. Efficiency was maximized at the smallest Hc/D ratio enlargement of the body diameter, there was also an
and minimized at the largest. As in other studies, effi- increase in efficiency. In Figure 14, efficiency curves are
ciency increased as the Hc/D ratio decreased. Decreasing given for different diameters where C1 denotes the smal-
the Hc/D ratio increases the rotational velocity. lest and C3 is the largest diameter. Faulkner et al.[74]
In another study, Shastri et al.[72] varied the Hc/D ratio investigated the experimental influence of cyclone diame-
between 0.5 and 4. Increasing the length of the cylinder ter. In this study, a linear regression equation describing
reduced the pressure drop and resulted in a decrease in the relationship between cyclone diameter and efficiency
overall efficiency. Static pressure distributions for different is defined as given below.[75]
segment length configurations are provided in Figure 15.
The pressure distributions are similar for each configura- η ¼ 0:0009d2 þ 0:0314d þ 99:1587 ð5Þ
tion. Static pressure reaches its maximum value at the wall
edges and decreases radially towards the centre. In the This equation is only applicable for particulate matter
core region, however, it minimizes. (PM) with particle size distributions similar to that used
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
16 AYLI and KOCAK

F I G U R E 1 5 Mean static
distribution for different Hc/D
ratios.[72]

TABLE 4 Summary of the research works that investigate the effect of cylinder and cone height on the separator performance.

The effect of cylinder and cone height

References Modification Effect on the performance

[64]
Demir et al. Cylindrical and conical The body and conical heights significantly impact pressure drop, indicating that the
height variations. body height should be kept below 1.5D and the conical height should be below 2.5D.
Safikhani et al.[65] 1.05 < H/D < 4 Increase in cylinder height significantly contributes to the pressure drop
[66]
Pandey et al. 0.5 < H/D < 1.5 Increasing the cylinder height slightly reduced the collection efficiency.
Brar et al.[67] Cylinder height is varied. Increasing the cylinder height to 5.5 times the diameter improved pressure drop and
increased efficiency by 9.5%.
Hoffman et al.[68] Cyclone cylinder length is Lengthening the cyclone cylinder initially improves its performance significantly.
varied. However, extending the cylinder beyond a certain point leads to a sharp decline in
separation efficiency.
Ito et al.[69] Cyclone cylinder length is A connection was found between the natural vortex length and the overall cyclone
varied. length.
Shastri et al.[71] Varied the lengths of the Efficiency was maximized at the smallest Hc/D ratio.
cylinder and cone
Shastri et al.[72] Hc/D ratio is varied. Increasing the length of the cylinder reduced the pressure drop and resulted in a
decrease in overall efficiency

in this experiment. The curve shape they obtained for diameter body) exhibited the lowest efficiency among the
cyclone diameter versus efficiency is given in Figure 16A. tested configurations. Interestingly, the pressure drop
The results are contrary to those of Brar and Sharma’s also followed the same trend, with Cyclone 2D2Dn hav-
study.[73] In a study by Modabberifar et al.,[75] the rese- ing the highest pressure drop.
archers investigated how expanding the diameter of a In another study, both numerical and experimental
cyclone cylindrical body affects its performance. They investigations indicated that the pressure drop, velocity,
compared two configurations: one with a body diameter and turbulence demonstrate nearly negligible correlation
equal to the cyclone diameter and another with a double with the cyclone body diameter (with an error margin of
diameter. Their goal was to improve collection efficiency less than 11%), with the 1D2D configuration exhibiting
and pressure drop by lengthening the vortex inside the the least reliance on body diameter. Among the cyclone
cyclone. The results, presented in Figure 16B, showed types considered, the 1D3D configuration exhibits the
that Cyclone 2D2Dn (double diameter body) achieved highest efficiency.[76] The Texas A&M cyclone design
the highest collection efficiency. Cyclone 1D3Dn (one (TCD) method, pioneered by Parnell in 1996,[76] estab-
and a half diameter body) followed closely behind, with lishes cyclone dimensions by selecting a desired diameter
only a 2% efficiency difference. Cyclone 1D2Dn (equal (D) to attain a specific airflow velocity. For cyclones of
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 17

F I G U R E 1 6 Efficiency values for cyclone separator with different body diameters (A) from Brar and Sharma[73] and (B) from
Modabberifar et al.[75]

the 1D3D type, the TCD method aims for a design inlet cyclone wall triggered by centrifugal force, the agglom-
velocity of 975 ± 120 m/min, as depicted in Figure 17. eration of particles at the wall and in the dust bin, and
The diameter (D) is determined based on the required the separation within the dust hopper. Therefore, the
volumetric airflow rate of the application, calculated by geometry of the dust bin also has an impact on the sepa-
multiplying the barrel diameter by half itself (D/2) and ration performance. In line with this purpose, both
then by a quarter of itself (D/4) to represent the inlet numerical and experimental studies are available in the
cross-sectional area. The cyclone designation, such as literature. Huang et al.’s[78] experimental studies have
‘1D3D’, indicates the relative lengths of the cyclone sec- demonstrated the impact of back-mixing from the dust
tions, with ‘1’ representing the barrel length equal to the hopper on the separation performance. The design of
diameter (D), and ‘3’ indicating a cone length three times the dust outlet shouldn’t mess up the swirling flow (vor-
the diameter. In essence, the TCD method facilitates the tex) inside the cyclone. It should not slow down this
design of cyclones with appropriate diameters to achieve swirling motion. On the other hand, the dust outlet
desired airflow velocities, particularly tailored for optimal should help smooth out the flow within the cyclone and
performance in 1D3D cyclone configurations. prevent anything that might mess up the separation pro-
When the results of Lapple,[62] Pant et al.,[77] and cess, like highly turbulence motions. Separation surface
Barth[11] are examined, it is observed that as the cyclone of the cyclone should be increased with the appropriate
diameter increases, the collection efficiency decreases. In dust geometry.[79]
Pant et al.’s study,[77] the decrease in efficiency was found In the study by Obermair et al.,[80] the impact of dif-
to be much more predominant compared to the others, ferent dust geometries on performance was experimen-
while in the studies of Lapple[62] and Faulkner et al.,[74] tally examined. The examined geometries are shown in
the effect of diameter on efficiency was found to be Figure 16A. According to the obtained results, velocity
around 10%. magnitudes depend on the dust geometry, while the vor-
A summary table (Table 5) has been included to tex structure remains independent. Additionally, the
showcase the cyclone diameters investigated by various addition of a downcomer tube increases friction, leading
researchers and their corresponding results. The findings to decreased velocities and reduced separation efficiency.
from these studies exhibit a degree of consistency. The highest separation efficiency is achieved with the
cyclone equipped with a downcomer tube.
Zhu et al.[81] numerically investigated the effects of dif-
3.4 | The effect of dust hopper ferent dust hopper structures on the flow. As the diameter
configurations of the dust hopper decreases, the characteristic frequency
of the dust hopper increases. However, the increase in fre-
In cyclone separators, the gas–particle separation process quency leads to asymmetric flow, which in turn reduces
occurs due to the movement of particles towards the the efficiency of the separator. Kang and Kwak[82] have
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
18 AYLI and KOCAK

F I G U R E 1 7 Configuration
one and a half diameter body
(1D3D) and double diameter
body (2D2D) cyclone
separators.[76]

TABLE 5 Summary of the research works that investigate the effect cyclone diameter on the separator performance.

The effect of cyclone diameter

References Modification Effect on the performance

Brar and 0.8D < D < 1.2D While the pressure drop values increased with the enlargement of the body diameter, there
Sharma[73] was also an improvement in efficiency.
Faulkner Diameter is varied. A linear regression equation that describes the relationship between cyclone diameter and
et al.[74] efficiency is defined.
Modabberifar D, 2D Cyclone 2D2Dn (double diameter body) achieved the highest collection efficiency, with
et al.[75] Cyclone 1D3Dn (one and a half diameter body) close behind, showing only a 2% efficiency
difference.
Parnell[76] Barrel and cone Optimal performance in 1D3D cyclone configurations.
length are varied.
Lapple[62] Diameter is varied. Cyclone diameter increases, the collection efficiency decreases.

created three models as shown in Figure 18B by numeri- Some numerical and experimental studies on cyclone
cally varying the diameter and length of the dust. separators are summarized in Table 3.
Figure 19 compares three cyclone models (1, 2, and 3) and
their ability to separate particles. Model 2 outperformed
the base model (Model 1) by capturing particles more effi- 4 | EFFICIENCY IMPROVEMENT
ciently, especially the smaller ones. This improvement in ST UDI E S U SI NG P A SSI V E M E T HO D S
Model 2 is because it allows particles to settle out of the
air stream faster by increasing the velocity at which they The initial studies on classical cyclones date back to the
exit the hopper. To facilitate a comprehensive overview, 1940s,[5,6] but recent research has focused on studies con-
Table 6 summarizes the studies reviewed, outlining the ducted in recent years. Despite being invented over a cen-
cyclone diameters employed by each researcher and the tury ago, separators operating with cyclonic effect
significant outcomes observed. continue to attract the interest of researchers due to their
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 19

F I G U R E 1 8 (A) Examined dust geometries in the study of Obermair et al.[80,83,84] and (B) Three different hopper models used in the
study of Kang and Kwak.[82]

FIGURE 19 Separation efficiency curves for different models used in Zhu et al.[82]

simplicity, ease of manufacture, and the performance Section 3, the effects of basic cyclone geometric parame-
they can exhibit. Before the advancements in computa- ters on performance continue to be investigated in detail
tional technology, limited options for processors and both experimentally and numerically. Another evolving
memory, as well as the lack of widespread availability of process during this period is the strengthening and
computers with high memory and processor capacities, improvement of gas structures within separators by add-
led to the initial numerical studies of cyclones focusing ing features. Towards this end, some of the improve-
on geometries with axial symmetry.[85] As detailed in ments and experimental methods include:
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
20 AYLI and KOCAK

TABLE 6 Summary of the research works related to investigate the effect of dust hopper configurations.

The effect of dust hopper configurations

References Aim of the study Effect on the performance

[78]
Huang et al. Dust hopper effect on Dominant effect of back-mixing from the dust hopper on the separation
performance. performance.
Obermair et al.[80] The impact of different dust Velocity magnitudes depend on the dust geometry, while the vortex structure
geometries. remains independent.
Zhu et al.[81] The effects of different dust The increase in frequency leads to asymmetric flow, which in turn reduces the
hopper structures. efficiency of the separator.
Kang and Kwak[82] The diameter and length of Model 2 surpassed the base model (Model 1) in particle capture efficiency,
the dust are varied. particularly with smaller particles. This enhancement in Model 2 is due to its
design, which accelerates the particle exit velocity from the hopper, allowing them
to settle out of the air stream more quickly.

• Adding water nozzles, roof (WCR) are added to the classical Stairmand type
• Inserting ribs, cyclone separator. Results show that the ACR cyclone
• Employing double-stage cyclones, separator and the WCR cyclone separator have more
• Incorporating additional inlets, regular vortices and a lower degree of internal vortex
• Using finned cylinder bodies, breakdown than the FCR cyclone separator. However,
• Adding extra top inlets, introducing liquid jets, the iso-vortex surface on the wall of the exhaust pipe is
• Employing helical roof inlets, seriously broken as shown in Figure 21. Total separation
• Adding laminarizers, efficiency for FCR, ACR, and WCR are calculated as
• Incorporating internal spiral vanes, and 56.32%, 80.98%, 70.38%, respectively.
• Employing slotted vortex finders. Yuan et al.[108] used flat-roof instead of spiral-roof
and examined its effect on pressure drop and efficiency
Some of these studies are summarized in this section, numerically. They add spiral guide vanes into the spiral-
highlighting the types of modifications that have proven roof cyclone separator annular space as shown in
effective. Several studies about the geometric modifica- Figure 20C. The results demonstrate a more than 20%
tions on cyclone seperators are depicted in Table 7. reduction in pressure drop compared to the flat-roof
cyclone. The new design exhibits 3%–5% higher effi-
ciency than the conventional spiral-roof cyclone and
4.1 | Roof structure effect surpasses the flat-roof cyclone by 2%–3%. Also, Misiulia
et al.[109,110] conducted a comparison between the
One of the new methods used to improve separation per- spiral-roof cyclone separator and the flat-roof cyclone
formance is the use of a wedge-shaped roof. Li et al.[86] separator, both having identical inlet areas. The simula-
numerically integrated a wedge-shaped roof into the system tion findings revealed that the spiral-roof cyclone sepa-
and investigated the effects of the height, top diameter, and rator exhibited greater aerodynamic efficiency in
bottom diameter of the wedge-shaped roof on performance. comparison to the flat-roof cyclone separator.
Using a multi-objective optimization method, they exam- Wasilewski and Brar[58] conducted a numerical com-
ined the influence of these parameters on performance. parison of efficiency by opening air holes on the roof and
Geometric variants are presented in Figure 20. According the Stairmand separator. The maximum efficiency was
to the results obtained, the design with the wedge-shaped obtained with a value of 88.3% at 0.6 m/s. The pressure
roof leads to an increase in efficiency. Furthermore, in the drop is not very different for the base case and modifica-
best designs, separation efficiency is improved by 4.2% after tions. While Eu was 6.32 in the base case, it increased to
optimization, while the pressure drop is increased by 11%. 6.73 in the case with the highest pressure drop.
In another investigation,[107] three distinct configura- Table 8 presents a summary of the studies examining
tions for cylinder roofs were suggested as shown in roof structure modifications. The most popular methods
Figure 19B. Utilizing CFD technology, researchers exam- are the spiral roof and wedge-shaped roof designs. Both
ined the impact of these varied cylinder roof structures designs have consistently achieved higher efficiencies
on the cyclone vortex. Flat cylinder roof (FCR), arc- compared to the traditional design. This trend is reflected
shaped cylinder roof (ACR), and wedge-shaped cylinder across all the studies reviewed.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 21

F I G U R E 2 0 Variants of cyclone separator with wedged roof structured design in the studies of (A) Li et al.,[86] (B) Li et al.,[107] and
(C) Yuan et al.[108] ACR, arc-shaped cylinder roof; FCR, flat cylinder roof; WCR, wedge-shaped cylinder roof.

4.2 | Rod insertion effect tangential velocity and centrifugal force while decreasing
the Euler number. Figure 22 provides visual representa-
Another modification method is to add a rod inside the tions of fluid pathlines for configurations with rods.
cyclone separator to increase the tangential velocities, Yohana et al.[114] demonstrated that the insertion of a
thereby enhancing efficiency through turbulence effects metal rod increased the heat transfer and collection effi-
and reducing pressure drop. In their study, Houben ciency by 22% and 13%, respectively. The effect of rod
et al.[111] observed that adding a rod to the system positively usage on efficiency is illustrated in Figure 23. As observed,
affects the separation efficiency of small particles due to the the addition of the rod had a positive effect on efficiencies.
increase in tangential velocity values. Additionally, the use Table 9 summarizes the studies that have investigated
of a vortex stabilization rod reduced the pressure fluctua- the impact of rod insertion on performance. The overall
tion from 5%–8% to 0%–3%. Moreover, they observed that results indicate that rods enhance the centrifugal force by
increasing the rod length also increased the vortex length. increasing the rotational movement of the fluid. This
On the other hand, the use of a thinner rod positively enables even smaller particles to collide with the walls
affected both the pressure drop and vortex stabilization. and be separated. Furthermore, rods reduce turbulence
Wasilewski et al.[112] examined a rod with a circular by decreasing the irregular movements of the fluid. Con-
cross-section along the axis of the cyclone separator. Simi- sequently, particles move more stably, leading to more
larly, they observed that adding a rod stabilized the flow, efficient separation.
as noted by Houben et al.[111] The maximum pressure drop
reduction was around 23% compared to the base model.
Sripriya et al.[113] numerically investigated the effect of the 4.3 | Fin attachment effect
diameter and insertion height of a metal rod placed inside
the cyclone on the pressure drop. Consistent with other Yang et al.[91] placed helical fins on the cylindrical body
researchers, they observed that the rod increased the and investigated the effect of the size of these fins and
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
22 AYLI and KOCAK

TABLE 7 Several studies about cyclone separator.

Particle size/ Performance

Exp/ Cyclone mass Notes about criteria note
References Cyclone type Modification CFD diameter flow rate pressure drop (Eu, St, h)
Kumar et al.[87] Cyclone (a/D, b/D) = CFD 0.205 m 0.3, 0.6, 1, 1.5, 1.27–42 (change) h: 6–30 (increment)
separators (0.51, 0.49), 2, 3, 4, 6, 8, and
with elliptical (0.52, 0.48), 10 μm
cross- sections (0.53, 0.47),
(0.54, 0.46), and
(0.55, 0.45)
Yao et al.[88] The multi- Multiple CFD 6m 80 μm Increases the From 94.77% to
stage straight line pressure drop by 96.11%
contracted segments on its 187.04 Pa
inlet duct inner side wall
separator
Cao et al.[89] Reversed Iron powder & Exp 0.16 m 30, 50, 70, and No information The averaged
cyclone Catalyst Powder 100 g/m3 velocity of catalyst
separator or iron powder
increases as the inlet
velocity increases
Cao et al.[90] PV-type Eight-vane CFD 0.3 m Concentration: Reduced by about Increased by 0.94%
cyclone deswirler are 30 g/m3 25.74 26.29%
separator added
Yang et al.[91] 2D2D type Helical CFD 200 mm 0. 1, 0.5, 1, 2, The pressure drop Increased by 7.4%,
cyclone triangular fins and 3 μm is increased by 15.9%, 20.1%, 10.9%,
separator are added 6.6%. and 14.8%
Guo et al.[92] Stairmand Three new CFD 0.29 m No information No information Better for
cyclone models with enhancement of
separator variation of flow stability,
vortex stabilizer providing about
length and 18.4% maximum
diameter reduction of
normalized flow
angle
Li et al.[86] Stairmand Wedge-shaped CFD 0.29 m <10 μm Increases by 11% Increased by 4.2%
cyclone roof geometry
separator
Guo et al.[93] Stairmand Slotted vortex CFD 0.29 m 1,3,5 μm 20% reduction of Efficiency is
cyclone finder pressure drop improved with the
separator new design.
El-Emam Stairmand Varied inlet CFD 0.32 m 500: 2500 seed: Noteworthy 4%–14% increase
et al.[94] cyclone dimensions and leaf particles variation in
separator positions, pressure drop
featuring both values between the
single (SI) and single and double
double spiral inlet designs
axisymmetric
spiral (DI)
inlets
Fatahian Square Inlet CFD 0.2 m 1–32 μm No information Particle diameter of
et al.[95] cyclone temperature, 16 μm achieved 76%
separator inlet velocity,
particle mass
flow rate is
varied
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 23

TABLE 7 (Continued)

Particle size/ Performance

Exp/ Cyclone mass Notes about criteria note
References Cyclone type Modification CFD diameter flow rate pressure drop (Eu, St, h)
Prasanna Swift, Lapple, Vortex finder CFD No 5 g/s (mass Swift HE, 4-holes Swift HE, 4-holes
et al.[96] and Stairmand holes (4–8–12– information flow rate) design: 60% design: 2.6% lowered
16) lowered StairmandHT,
StairmandHT, 4-holes design:
4-holes design: 8% negligible change
lowered
Gopalakrishnan Stairmand Blade angle, CFD No 10,000 Increase in blade Increase in the
et al.[97] cyclone blade length, information particles, angle: number of blades:
separator blade-tube 2 g/m3 unfavourable increase in
distance and increase in efficiency, efficiency
number of pressure drop, is sensitive on blade
blades are number of blades length
varied >5, increment in
pressure drop
unfavourably
Katare et al.[98] Hydrocyclones No modification CFD 0.15 m 10–15 μ m Maximum and 76.19%
minimum
pressures: 9.76e04
Pa and 4.62e04 Pa
Rathish et al.[99] No Vane is used in CFD 0.32 m No information Vane angle of 10 10
: suitable angle to
information the cyclone degree gives the use to remove the
separator. maximum pressure moisture efficiently
Angle is varied- drop in cyclone separator
0
to 25

Venkatesh Square Body height CFD 0.2 m Concentration: Highly influenced Cut-off diameter of
et al.[100] cyclone and inlet 8.8 g/m3 by s inlet height, this new design is
separator height, total inlet width and 2.09 μm
height, inlet outlet diameter
width and
outlet diameter
Chen et al.[101] Divergent Spiral guide CFD 0.2 m 1 3.5 mm Has lowered the 44.4% better
cyclone vane and a & pressure drop by
separator central column Exp an average of
29.8%
Venkatesh Multi-cyclone Four parallel CFD 0.203 m 1–7 μm Increasing the inlet Increasing the
et al.[102] separators cyclone & velocity increases pressure drop
separators Exp the pressure drop decreases the
collection efficiency
of the cyclone
Wang et al.[103] Stairmand Inlet angle is CFD 0.205 m 2.2 μm The negative inlet The separation
cyclone varied angle brings about efficiency is
separator a larger pressure increased with
drop comparing to anincrease of
positive inlet angle. particle size,
efficiency firstly
increases and then
declined as inlet
angle changes from
negative to positive.
(Continues)
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
24 AYLI and KOCAK

TABLE 7 (Continued)

Particle size/ Performance

Exp/ Cyclone mass Notes about criteria note
References Cyclone type Modification CFD diameter flow rate pressure drop (Eu, St, h)
Misiulia Industrial Gas outlet CFD 0.3 m 0.2–4 μm Outlet scroll and Gas outlet
et al.[104] scroll-inlet configuration radial bend configurations had
high- increase the almost no effect on
efficiency pressure losses by the cyclone grade
cyclone 5.1% and 6.4% efficiency.
separator
Wasilewski and Stairmand Six different CFD 0.1 m 0.25–3 μm n increase of Eu At v = 0.6 m/s, for
Brar[58] cyclone values of clean occurred by about which value of Stk
separator gas inlet 6% 50 η increased by as
velocities much as 4.7%
Chu et al.[105] Lapple Effect of solid CFD 0.2 m 0–0.5–1–1.5–2– Pressure drop No information
cyclone loading ratio 2.5 kg solid/kg decreases with the
separator air increase of solid
loading steadily
Eyvazi et al.[106] Stairmand Particle CFD 0.127 m 0.15–5 μm Reducing the For the finest
cyclone diameter & size & swirling angle of particles (mean
separator Exp the channels leads 0.327 μm) reached
to a decrease in 86%
pressure drop

Abbreviation: CFD, computational fluid dynamics.

when helical fins were added, which was considered as a

contributing factor to the efficiency compared to the tra-
ditional configuration (Figure 24).
Dasar and Patil[115] also added helical fins to the
cylindrical part of the traditional separator. They varied
the fin size and fin pitch and examined their effects on
performance. Like Yang et al.,[91] it has been observed
that structures with helical fins offer improved collection
efficiency. Similar to Yang et al.,[91] they observed an
increase in particle adhesion to the wall due to the
increase in centrifugal force, resulting in enhanced col-
lection. In another study, Dasar and Patil[116] altered the
geometries of the helical fins placed in the cylindrical
region into triangle, semicircle, and rectangle shapes. It
has been observed that cyclone separators based on trian-
gular helical fins exhibit an improvement in collection
efficiency ranging from 5% to 10% compared to tradi-
tional cyclone separators. Additionally, they have demon-
F I G U R E 2 1 Iso-vortex structures of cyclone separators with strated through heat transfer analyses that triangular fin
different roof geometry.[107] ACR, arc-shaped cylinder roof; FCR, structures could also be utilized as energy-saving heat
flat cylinder roof; WCR, wedge-shaped cylinder roof.
exchangers.
In another study,[117] they converted the geometry of
the 2D2D Lapple type cyclone separator into finned
their spacing on performance. The separation efficiency cyclone separators by attaching semi-circular cross-
of the cyclone separator equipped with helical triangular section helical fins along the height of the barrel wall.
fins was calculated to be higher than that of the tradi- Semi-circular helical fins vary in size (5, 7.5, and 10 mm)
tional design. Its effect on pressure drop was considered and pitch (30 and 50 mm). Improved collection efficiency
to be low. Higher tangential velocities were observed ranging from 5% to 11% compared to conventional
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 25

TABLE 8 Studies related to investigate the effect of roof structure.

The effect of roof structure effect

References Modification Effect on the performance

[86]
Li et al. A wedge-shaped roof The design with the wedge-shaped roof leads to an
increase in efficiency.
Li et al.[107] A wedge-shaped roof with different heights, top and Arc-Shaped cylinder roof has more regular vortices.
bottom diameters
Yuan et al.[108] Spiral-roof with spiral guide vanes New design exhibits 3%–5% higher efficiency
[109,110]
Misiulia et al. Spiral-roof cyclone separator and the flat-roof cyclone Spiral-roof cyclone separator exhibited greater
separator comparison aerodynamic efficiency
Wasilewski and Air holes on the roof Eu is increased with the new design.
Brar[59]

F I G U R E 2 2 Fluid path
lines free vortex finder
clearances of (i) 17% and (ii) 60%
with a spigot blocking ratio of
0.4.[113]

the performance of the cyclone separator. The removal

efficiency of porosity = 0.2 was calculated to be 9.2 per-
centage points less than that of porosity = 0.
By incorporating fins of various shapes and dimen-
sions into cyclone separators, engineers have found a reli-
able method to enhance separation efficiency. These fins
modify the internal fluid dynamics of the cyclone, conse-
quently influencing the mechanism by which particles
are collected. Table 10 provides a compilation of studies
that have delved into this subject.

4.4 | The effect of separator with vane

The general structure of the axial inlet cyclone separator

is similar to that of the circular flow cyclone separator,
but an important internal component, the guide vane, is
FIGURE 23 Efficiency curves for different cases.[114] added to the cylinder of the axial flow cyclone separator.
The gas containing dust is introduced axially into the sep-
cyclone separators was observed with the attachment of arator cylinder. Then, it rotates and accelerates under the
semi-circular fins along the height of the barrel wall. Xu action of the guide vanes, forming a high-speed rotating
et al.[118] examined the effect of porosities in the fins on gas, so that the solid particles are separated under the
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
26 AYLI and KOCAK

TABLE 9 Summary of the research works investigating the effect of rod insertion.

Rod insertion effect

References Modification Effect on the performance

[111]
Houben et al. Insertion of vortex stabilization rod. Pressure fluctuation reduces from 5%–8% to 0%–3%.
Wasilewski et al.[112] Rod with a circular cross-section along the axis Adding a rod stabilized the flow and pressure drop is reduced.
of the cyclone separator.
Sripriya et al.[113] The effect of the diameter and insertion height Rod increased the tangential velocity and centrifugal force
of a metal rod are varied. while decreasing the Euler number.
Yohana et al.[114] Insertion of a metal rod. Increased the heat transfer and collection efficiency.

swirling flow and undesired instabilities. The new design

has increased collection efficiency.
Celis et al.[123] investigated the performance of a
gas–liquid cyclonic separator by adding swirl vanes
inside. Swirl vanes, commonly used in axial separators,
can be utilized to organize the internal flow, reduce
pressure losses, attenuate turbulence, stabilize the pre-
cession of the vortex core, and enhance the separation
process. Yan et al.[124] added guide vane structures
to the outlet of the cyclone. In the new design, the sepa-
ration efficiency has exceeded 80% for particle sizes
below 10 μm.
Xiong et al.[125] employed a guide vane structure fea-
turing eight blades, illustrated in Figure 25A. Their find-
ings revealed that the incorporation of a reflux cone
resulted in an efficiency increase of 5.2–6.7 percentage
points. The reflux cone serves as a barrier that prevents
particles from escaping through the entrance, thereby
contributing significantly to the enhanced separation effi-
ciency facilitated by the guide vanes. Figure 25B illus-
trates the efficiency changes for various types of cyclone
separators designed in the study.
Table 11 summarizes studies on the effect of separa-
tors with vanes. Vanes are elements used to regulate the
internal fluid dynamics of cyclone separators, thus
F I G U R E 2 4 Tangential velocity distribution for traditional
enhancing separation efficiency. The general consensus
configuration and configuration with helical fins.[91]
is that efficiency is increased.

action of centrifugal force. The advantages of the axial

inlet cyclone separator include compact structure, stable 4.5 | The effect of vortex finder with a
internal flow field and small pressure loss. In particular, deswirler
it is suitable for situations with large processing capacity,
high inlet concentration, and high gas velocity. In the Misiulia et al.,[126] converting the guide vane structure
axial inlet cyclone separator, the structure and size of into a deswirler structure, demonstrated that it prevented
the guide vanes play an important role in ensuring effec- reverse flow and thus reduced pressure loss. In another
tive separation performance. Therefore, researchers have study, they have shown that the positioning of the des-
developed various guide vane structures and achieved wirler structure is an influential parameter on perfor-
certain results.[119–121] Xu et al.[122] designed a slotted mance.[127] In other studies of Misiulia et al.,[128,129] it
vortex finder and conducted numerical analyses. This has been observed that deswirler reduces pressure drop
design reduces pressure drop by reducing the intensity of while also decreasing overall efficiency. However, it is
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 27

TABLE 10 Summary of the research works investigating the effect of fin attachment effect.

Fin attachment effect

References Modification Effect on the performance

Yang Helical fins are inserted on the The cyclone separator with helical triangular fins demonstrated superior
et al.[90] cylindrical body. separation efficiency compared to the conventional design.
Dasar and Added helical fins to the cylindrical Structures with helical fins offer improved collection efficiency.
Patil[115] part of the traditional separator.
Dasar and Alteration of the geometries of the Triangular helical fins exhibit an improvement in collection efficiency.
Patil[116] helical fins.
Dasar and Lapple type cyclone separator The attachment of semi-circular fins to the barrel wall of the cyclone separator
Patil[117] modified to finned cyclone separators. led to a notable enhancement in collection efficiency, ranging from 5% to 11%
higher than traditional designs.
Xu et al.[118] Effect of porosities in the fins. A porosity of 0.2 resulted in a 9.2% decrease in removal efficiency compared to a
porosity of 0.

FIGURE 25 (A) Designed cyclone separators and (B) efficiency variation for different configurations.[126]

important for the deswirler to be positioned far from the the installation of deswirler compensates for the pressure
vortex finder to preserve the flow topology. drop in the vortex finder while also reducing efficiency to
In the study conducted by Cao et al.,[90] it was a certain extent. Comparatively, the use of deswirler sig-
observed that, similar to Misiulia et al.’s research,[126–129] nificantly decreases the pressure loss in the vortex finder
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
28 AYLI and KOCAK

of the separator and makes the static pressure distribu- structures are widely recognized in the literature. Addi-
tion at the outlet more uniform, in contrast to traditional tionally, researchers have experimented with various
designs as given in Figure 26. other approaches. Table 13 provides a summary of some
Table 12 presents a summary of research on the modifications that have been attempted and proven suc-
impact of vortex finders and deswirlers. These compo- cessful. Although studies for this purpose began in the
nents play a significant role in controlling fluid flow, mid-1900s, they continue to be actively pursued to
minimizing turbulence, and improving separation effi- this day.
ciency. The studies conclude that these components can
significantly enhance separation efficiency by promoting
a more organized and stable fluid motion. 5 | NUMERICAL M ETHODS
EMPLOYED IN CYCLONE
SEPARATOR DESIGN FRAMEWORK
4.6 | Other modifications and their
impact on performance In the numerical modelling of cyclone separators, the best
approach for accurately modelling the flow field is through
To increase separator efficiency, methods such as roof numerical methods capable of modelling particle–fluid
structure alteration, rod insertion, fin attachment, the interactions. When numerically modelling how particles
addition of vanes, and the utilization of deswirler in a cyclone separator interact with the air, accurately

TABLE 11 Summary of the research works investigating the effect of vanes.

Separator with vane effect

References Modification Effect on the performance

[122]
Xu et al. Slotted vortex finder The new design effectively lowers pressure drop by suppressing swirling flow and unwanted
instabilities, leading to improved collection efficiency.
Celis Swirl vane insertion Incorporating swirl vanes into axial separators, it is possible to regulate internal flow, decrease
et al.[123] pressure losses, attenuate turbulence, stabilize the vortex core’s precession, and enhance the
separation process.
Yan Insertion of guide vane The separation efficiency has exceeded 80%
et al.[124] structures
Xiong Insertion of guide vane Incorporation of a reflux cone resulted in an efficiency increase of 5.2–6.7 percentage points.
et al.[125] structures

FIGURE 26 Variation of pressure drop and efficiency for precessing vortex (PV), deswirler, and hollow-type deswirler cyclone
separator.[90]
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 29

T A B L E 1 2 Summary of the research workss investigating the These combinations are used to analyze different turbu-
effect of vortex finder with a deswirler. lence characteristics and scales, selected depending on
the characteristics of the phenomena being studied and
Vortex finder with a Deswirler effect
the requirements of the analysis.[144,145] The fundamen-
Effect on the tal models utilized are presented in Table 14.
References Modification performance The technological boom has ushered in a powerful
Misiulia Conversion of the Pressure loss is new approach for understanding particle-laden flows:
et al.[126] guide vane structure reduced. CFD-DEM. This method merges the strengths of
into a deswirler
CFDand DEM, creating a detailed, multiphase analysis
structure
tool. Unlike other, often cumbersome options, CFD-
Misiulia Location of deswirler Position has a DEM shines in its computational efficiency and numeri-
et al.[128] structure dominant effect on
cal ease. CFD-DEM tackles the continuous phase, the
the performance.
fluid itself, using the well-established Navier–Stokes
Misiulia[129] Deswirler insertion Deswirler reduce
equations. Meanwhile, the discrete phase, the individual
pressure drop while
particles, are tracked by applying Newton’s second law
also decreasing
overall efficiency to each one within the fluid system.[146] The governing
equations for CFD-DEM coupling are given in the Sec-
Cao Installation of Pressure drop
et al.[90] deswirler reduction tions 5.1 and 5.2.

5.1 | Fluid phase

predicting the trajectories of the particles and the collection
efficiency becomes possible. Otherwise, it would be impos- The Navier–Stokes equations have been solved in three
sible to calculate particle distribution and efficiency accu- dimensions to analyze the fluid phase movement within
rately. In this context, the two main approaches used are the cyclone separator. Equations (6) and (7) represent the
the Euler–Lagrange and Euler–Euler models. The main dif- mass and momentum conservation equations, respec-
ference between the two is how gases and particles are cal- tively. The flow should be solved as a continuum, three-
culated and how their interactions are handled. Both dimensional, incompressible, and time-dependent.[146]
models track gases and particles separately, but in the
Euler–Lagrange approach, the gas phase is expressed by ∂α ∂αui
þ ¼0 ð6Þ
Euler flow equations, while the particle phase is calculated ∂t ∂xi
using Lagrange equations, whereas in the other approach,
both flows are calculated using Euler flow equations. The ∂αui ∂αui 1 ∂αp ∂ 2 αui fp
Euler–Lagrange approach is preferred for more accurately þ uj ¼ þv þ αgi  i ð7Þ
∂t ∂xj ρ ∂xi ∂xi ∂xj ρ
calculating particle motion, generally preferred for analyz-
ing the movement of large particles. The Euler–Lagrange
model is more extensively used in cyclone separators. The The variables are defined as follows: α represents the
discrete element method (DEM) is typically used in con- volume fraction of fluid in each cell, ui is the fluid veloc-
junction with the Euler–Lagrange model to model the ity field in direction i, p denotes fluid pressure, ν stands
behaviour of solid particles with more detail. DEM is uti- for kinematic viscosity of fluid, gi represents gravitational
lized to track the movement of solid particles in detail, acceleration in direction i, and ρ denotes fluid density.
allowing for a more accurate assessment of their effects on The fluid phase solution can be classified into three types:
the gas phase.[142,143] direct numerical simulation (DNS), large eddy simulation
On the other hand, particularly in older studies, dif- (LES), and Reynolds-averaged Navier–Stokes equations
ferent types of turbulence modelling approaches such as (RANS). This classification depends on the required flow
RANS, DNS, and LES are used in cyclone separator ana- resolution level and the computational resources at hand.
lyses. Especially, the RANS model is frequently used in
cyclone separator analysis in the literature due to its
computational cost advantages over other options and 5.1.1 | Direct numerical simulation
not requiring advanced technology. Different types of
turbulence modelling approaches like RANS, DNS, In DNS utilizing the point-particles (PP) approach, the
and LES can be used in conjunction with multiphase fluid velocity at the particle position can be directly
flow models like Euler–Lagrange and Euler–Euler. derived from the DNS solution.[146]
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
30 AYLI and KOCAK

TABLE 13 Some modification techniques for cyclone separator design.

Particle
Exp/ Cyclone size/ mass
References Modification Schematic view CFD diameter flow rate Important results
Pandey and Curved profile of CFD 0.205 m 0.5–10 μm • Pressure drop and the collection
Brar[130] the conical efficiency decrease with an
segment increasing curvature size.
• Decrease in the Euler number by
34.57% and an increase in the
collection efficiency by 4.32%, along
with a reduction in the cut-off size
by 5.53%.
Dong Water-only Exp 0.15 m Ash content • CCS improved the separation of
et al.[131] cyclones with 0. 5– coarse coal.
single angle of 70
, 0.25 mm • CCS is a highly efficient separation
90
, 110
, and 120
and 0.25– equipment for coarse coal slim.
0.125 mm

Pei Lapple cyclone CFD& 0.2 m 0.1 and • Pressure drop decreases by 16.41%.
et al.[132] separator with the Exp 140 μm • Separation efficiency increases
inserted cross- by 0.64%.
shaped metal • Reduction of the vortexcore
blades diameter decreases the
pressure drop.
Chen Spiral guide vane CFD & 0.2 m 1–5 μm • A novel divergent cyclone separator
et al.[101] and a central Exp with combined internals was
column introduced to enhance separation
efficiency in this study.
• The use of internals increases
velocity and pressure inside the
separator, enabling effective
separation of ultra-fine particles.
Ganegama Down-comer tubes Exp 0.33 m 0–8.0 g/m3 • Usage of down-comer tubes
Bogodage significantly reduces fine particle
and re-entrainment in cyclone
Leung[133] separators, improving collection
efficiency and reducing
pressure drop.
• Particle concentration has a
dominant effect on the
pressure drop.
Le and Four-inlet cyclone CFD 0.29 m 2–10 μm • C4 has a higher tangential velocity
Yoon[134] separator than N4 and conventional design.
C4: sequential • The average Euler number of
positioning design N4 is 4.46, 0.9% higher than
N4: stair S1 (4.42) and 1.1% lower than C4.
positioning • Both designs have better
performance than the conventional
design.
Parvaz Inner cone located CFD 0.205 m 10 μm • Increasing the inner cone height
et al.[135] at the bottom of increases the performance
the cyclone • The collection efficiency decreases
when the erosion effect
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 31

TABLE 13 (Continued)

Particle
Exp/ Cyclone size/ mass
References Modification Schematic view CFD diameter flow rate Important results
Xie Cartridge filter CFD & 0.1 m 5–10 μm • The initial separation in the
et al.[136] Exp cartridge-filtering cyclone involves
swirling airflow to separate larger
particles, followed by the filtration
of fine particles using filter
cartridges.
• 99.86% efficiency, which is 15.32%
and 0.05% higher than that of the
common cyclone
Fu et al.[137] Vortex finder with CFD & 0.31 m 15 g/m3 • The pressure drop was reduced up
slots on the side Exp to 27.9%
wall • Increase in collection efficiency
was 5.45

Venkatesh Series CFD & 0.2 m 2.2–10 μm • Efficiency of series arrangement

et al.[138] arrangement Exp cyclone is 61%, 10.2% higher than
Square cyclone the collection efficiency of single
separator square cyclone
• Pressure drop is reduced to 14.3%
Venkatesh Multi-cyclones CFD 0.203 m 1 to 7 μm • Higher collection efficiency
et al.[102] • Greater size particles are collected
in first cyclone and the smaller size
particles are collected in the
adjacent cyclones

Sakin New design of CFD 0.15 m 0.5, 0.8, 1, • In reverse flow mode, there is a
et al.[139] conic and flat 1.25, 1.5, 2, greater pressure drop and collection
outlet tube axial 3, 5, 7, 8, efficiency compared to the axial
flow cyclone and 10 μm flow operation.
• Axial flow cyclones appear to
exhibit higher efficiency compared
to reverse flow cyclones for small
particles
Guo Slotted vortex CFD 0.29 m 1–3-5 μm • The SVF_D2 model has the
et al.[92] finder potential to enhance cyclone
separation efficiency, resulting in
approximately a 20% decrease in
pressure drop and a 4.2% reduction
in cut-off size.
• It could lead to a significant
reduction of 40% in the central
vortex rotation frequency.
Zhao Direct symmetrical Exp. 0.3 m 5.97 Am • DSSI cyclone and CSSI demonstrate
et al.[140] spiral inlet, superior performance, with
converging enhancements ranging from 0.15%
symmetrical spiral to 1.15% and 0.40% to 2.40%.
inlet
(Continues)
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
32 AYLI and KOCAK

TABLE 13 (Continued)

Particle
Exp/ Cyclone size/ mass
References Modification Schematic view CFD diameter flow rate Important results
Su et al.[141] Double square CFD 0.2 m 46.63 μm • The separator with single normal
inlet separator and inlet had the best separation
double declining efficiency and the maximum
inlets pressure drop.

Chen Vent pipe Exp. 0.45 m 15 μm • The separation efficiencies of the

et al.[120] insertion modified cyclone separator exceed
those of the traditional cyclone
separator.
• Both the separation efficiency and
the pressure drop vary depending
on the orientation of the sloped
edge, following the same pattern of
change. They reach their maximum
at 90
and minimum at 270
.

Abbreviation: CFD, computational fluid dynamics.

TABLE 14 Fundamental models used in the analysis of cyclone separator.

Model type Description Main advantage Main disadvantage

Euler–Lagrange Gas and particles are tracked separately. Gas Suitable for accurately modelling Can be computationally
model phase is represented by Euler flow equations, complex flow fields and particle– expensive.
while particle phase is calculated using Lagrange gas interactions.
equations.
Euler–Euler Gas and particles are tracked separately, but both This approach is used to calculate Less suitable for modelling
model are represented by Euler flow equations. the interactions of multiphase complex flow fields than the
flow. Euler–Lagrange model.
Reynolds Statistically parameterizes turbulence Less expensive to calculate Can be difficult to
averaged interactions and takes the time average of turbulence effects. accurately model small-
Navier–Stokes turbulence. scale turbulence structures.
(RANS) model
Direct Aims to resolve the entire turbulence scale. The highest resolution model and Computationally very
numerical gives the most accurate results. expensive and can only be
simulation used for simple geometries.
(DNS) model
Large eddy Large turbulence structures are resolved directly, Provides more accurate results Less suitable for complex
simulation while small structures are parameterized using than the RANS model and is less geometries than the DNS
(LES) model subgrid modelling. expensive than the DNS model. model.

    DNS-DEM coupling method. While DEM solution itself

sðk Þ
ui ¼ uDNS
i x pðkÞ ðt Þ, t ¼ ui x pðkÞ ðt Þ, t is already costly, our research found no literature on
studies that consider the cost of DNS-DEM solution.
Due to the diverse time and length scales inherent in
turbulent flows, obtaining an exact solution (DNS)
demands significant computational resources and metic- 5.1.2 | Large eddy simulation
ulous mesh refinement. Consequently, its usage is lim-
ited. A review of the Web of Science database revealed no RANS and LES models find more frequent use in engineer-
studies conducting cyclone separator analysis using the ing applications. LES directly resolves large eddies while
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 33

modelling small eddies, representing a middle ground where m is the mass of the particle and u and ω are the
between DNS and RANS in terms of the fraction of resolved translation and rotational velocity of the particle. Fij
scales.[147] The filtering of LES is given as follows[147]: and Mij are the force and torque on particle i by
Z particle j. Ii is the moment of inertia, and Fijc contact
  force on particle i by particle j. When particles I and
φ¼ φðx 0 ÞG x, x 0 ;Δ dx 0 ð8Þ
J come into contact with each other, the contact
does not occur at just a single point; there is an over-
where D is the domain of integration, G is the specified fil- lapping region that also has a significant impact on
ter function, Δ is the filter width, and x indicates the fil- the results. Thus, the overlapping is not neglected in
tered spatial coordinates. G(x;x0 ) equals 1/V if x0 is within the analysis.[149,150]
the computational volume. Otherwise, G(x;x0 ) is equal to
0.[148] The structure and size of the small scales that are
collected are determined by the filter function. 5.2.1 | Force analysis

In two-phase flow scenarios, particles experience two

5.1.3 | Reynolds-averaged Navier–Stokes distinct aerodynamic forces: constant force and vari-
(RANS) equations able aerodynamic force. The force resulting from the
fluctuating velocity within the fluid layer adhering to
RANS exclusively resolves the mean flow statistics, the solid surface is referred to as the Basset force. The
meaning the fluid velocity at the particle location cannot Basset equation is presented below.[149]
be directly derived from the RANS solution for fluid flow
Z  
fields. In RANS, the instantaneous flow field u(x, t) 3 pffiffiffiffiffiffiffiffiffiffiffi t
d uf  vp =dτ
F Basset ¼ d2p ρf μf π pffiffiffiffiffiffiffiffiffi dτ ð13Þ
decomposes into a time-averaged component hu(x, t)i 2 0 tτ
and a fluctuating component u0 (x, t).[146]

ui ðx, t Þ ¼ < ui ðx,t Þ > þ u0 i ðx, tÞ ð9Þ Another influential force is the pressure gradient
force. The magnitude of this force varies over the surface
Z T and is expressed as shown in Equation (14).[149]
< ui ðx, t Þ ≥ lim ui ðx, t Þdt ð10Þ
T!∞ 0
F rP ¼ V p rP ð14Þ

The averaging process introduces additional terms, The Saffman force is based on the Bernoulli principle.
<u0 iu0 j>, known as Reynolds stresses, into the original It is associated with the lift force arising from the particle
Navier–Stokes equations, which are also modelled using experiencing higher pressure on one side and lower pres-
eddy viscosity. These terms, <u0 iu0 j>, referred to as Reyn- sure on the other side in the flow. The equations are as
olds stresses, represent stress units only when multiplied follows[149]:
by the fluid density ρ.
 0:5d2 P
F Saffman ¼ 6:46 ρf μf ur x ðrxuÞjrxuj0:5 ð15Þ
4
5.2 | Solid phase
In order to calculate the Saffman force coefficient
DEM has been utilized to track particles within the which is only applicable for spherical particle equation is
cyclone separator system over time. The translational and developed by Mei[151] is given as follows:
rotational motion of the particles are resolved using
8

< 1  0:3314χ 0:5  exp  Re þ 0:3314χ 0:5 Re ≤ 40
Newton’s second law of motion and Euler’s law, respec- >
tively, as outlined below[149,150]:
Cl ¼ 10
>
:  
nc
0:0524 Reχ 0:5 Re > 40
∂ui X i

mi ¼ F cij þ F fi þ F gi ð11Þ ð16Þ

∂t J¼1

nc Surface forces are composed of adhesion and con-

∂ωi X i

Ii ¼ M ij ð12Þ tact forces, both of which significantly influence the

∂t J¼1 rotational motion of particles. Particle collisions
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
34 AYLI and KOCAK

FIGURE 27 Flow chart of computational fluid dynamics—discrite element method (CFD-DEM) coupling method.

8  
generate contact forces, the viscoelasticity of which >
< min K nl stn , F tΔt
ij þ K nu Δsn if Δsn ≥ 0
is crucially related to time. In the DEM model, F nij ¼   ð17Þ
forces between particles are calculated tangentially >
: max F tΔt þ K nu Δsn , λK nl stn if Δsn < 0
ij
and normally to the contact plane using a rigid
spring. Particle diameter, particle density, and the
fluid physical properties all play vital roles in adhe- where Δt is the time step, and F nij and F tΔt
ij are the nor-
sion forces. mal elastoplastic contact forces at time t and time t  Δt,
The most sophisticated approach for computing respectively.
energy loss during contact between two solid particles
is the Hysteretic Linear Spring Model. This model
offers the primary advantage of energy loss calculation 5.3 | DEM-CFD coupling
without the need for viscous damping terms. It is
independent of the relative velocities of colliding par- The CFD-DEM procedure typically involves three pri-
ticles and can be computed for overlap quantities mary steps: CFD, coupling, and DEM. During coupling,
that are otherwise challenging to capture. The mat- solutions derived from CFD are exchanged with DEM,
hematical formulation of this model is described in and vice versa, amalgamating solid and liquid phase
Equation (17). solutions to achieve a hybrid approach. In CFD-DEM
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 35

coupling problems, the initial phase commences with subject remains relatively limited worldwide. The average
the modelling of Navier–Stokes equations in the fluid number of studies per year in the 2000s is around 50.
domain using the finite volume method. Subsequently, Another issue is the challenging hybridization of aca-
initial fluid calculations, encompassing fluid density, vis- demic research with industry, especially regarding the
cosity, pressure, velocity, and temperature obtained via lack of industrial implementation of methods aimed at
CFD, are transmitted to the DEM solver through the cou- enhancing efficiency. One of the primary objectives of
pling module. Upon time step advancement, data such as this review article is to enhance the accessibility of aca-
interaction forces, particle velocity, and volume fraction demic research, compile studies conducted on the subject
acquired from DEM are conveyed to CFD. The fluid up to the present day, create a guide for the industry,
velocity, pressure, and temperature are then updated by and facilitate easy access to information. The key areas of
CFD for the subsequent time step and transferred back to guidance provided by this study are presented in
DEM. This iterative process continues, with data sets Table 15. These areas include the conceptual framework,
continuously exchanged between CFD and DEM, until design criteria, potential improvements to enhance per-
the simulation time is completed. The schematic repre- formance, case studies, numerical and empirical research,
sentation of the CFD-DEM coupling procedure is methodologies used in design, and future research direc-
depicted in Figure 27.[149,152] tions. This review article is designed to assist researchers,
engineers, and industry professionals by offering valuable
insights and practical guidance in the field.
6 | C ON C L U S I ON This study examines the working mechanisms of
cyclone separators initially. Academic studies unani-
Cyclone separators are widely utilized in various fields mously focus on fluid dynamics, and a common under-
such as energy and environmental applications, chemical standing of flow within cyclone separators has been
processes, and mining operations, owing to their superior adopted. The majority of initial studies on the subject
attributes such as low-cost installations and extended already delve into the analysis of fluid dynamics. Follow-
service life, along with a broad operating range. Since ing this foundational understanding, the literature began
their conceptual birth in the 1910s, studies on cyclone to scrutinize the effects of key parameters, especially geo-
separators have been conducted continuously up to the metric parameters, on the flow. However, contradictory
present day. Initially, research focused on the flow mech- findings regarding the effects of the same parameter can
anism and physics, followed by parameter optimization be found in the literature. This article addresses these
studies to enhance separator efficiency, investigations contradictory views. Therefore, the existence of such con-
into optimal geometrical ranges, and explorations of tradictory views signifies the necessity for further
novel modifications. These studies have been carried out research on the subject. While studies on classical separa-
both experimentally and numerically. Particularly in tor geometry have continued from the 1940s to the pre-
recent years, with the rapid advancement of technology, sent, there has been a parallel exploration of whether
significant progress has been made in numerical studies, overall efficiency can be improved through novel modifi-
driven by the increased affordability of particle-fluid cations and geometric alterations. In Chapter 4, research
modelling and the development of DEM models. How- on this topic is compiled, and detailed information is
ever, as depicted in Figure 1, academic research on this gathered on modifications that positively contribute to

TABLE 15 Practical guidance for cyclone separator research and applications.

Guidance area Details Target audience

Conceptual framework An overview of the development of cyclone separator technology from its Researchers, engineers
inception to the present.
Design considerations Key design parameters and considerations for optimizing cyclone separators. Engineers, industry
professionals
Performance evaluation Techniques and measurements for assessing cyclone separator efficacy and Researchers, engineers
efficiency.
Case studies and Successful cyclone separator implementations with using numerical, Industry professionals,
examples experimental methods researchers
Potential areas for future Outlined deficiencies and proposed areas of investigation to further enhance Researchers, industry
research cyclone separator technology. professionals
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
36 AYLI and KOCAK

FIGURE 28 PESTEL analysis for cyclone separator industry. PESTEL, political, economic, social, technological, environmental and
legal.

efficiency. In recent years, studies have predominantly higher accuracy and speed compared to experimental or
been conducted in numerical environments. Chapter 5 numerical methods. By using machine learning algorithms
outlines the foundation of numerical studies and dis- (such as artificial neural networks or support vector
cusses the mathematical framework and historical devel- machines) to create a performance prediction model, it is
opment in this field. possible to increase data accuracy without the time costs
In Figure 28, a PESTEL (political, economic, social, associated with traditional methods. This allows for a
technological, environmental and legal) analysis was con- quick and easy examination of the effects of various
ducted for the cyclone separator industry. The political, parameters on performance, enabling the identification of
economic, social, technological, environmental, and legal optimal design parameters more efficiently and accurately
components were briefly outlined using this analytical than conventional methods. Research efforts should be
method. This review article particularly emphasized tech- directed towards exploring these avenues.
nological advancements, as already highlighted. However, Upon examining the present condition of cyclone
on the other hand, each parameter in the PESTEL analysis separator technology, it is apparent that there exists a
represents important aspects for the development of the significant disparity between academic research and its
cyclone separator industry. Especially in terms of being an practical use in industry. The existence of this gap arises
environmentally friendly product, the use of cyclone sepa- from the challenge of converting theoretical progress into
rators is crucial for environmental regulations, waste man- tangible implementations. Our review discusses advance-
agement, energy efficiency, and sustainability efforts. ments and suggested alterations in cyclone separator
Research indicates that there is still a gap in the design, emphasizing the necessity for additional research
understanding of cyclone separators. One of the most sig- to fill this gap. Future research should prioritize the
nificant challenges is the low implementation and adop- development of practical integration strategies and real-
tion of academic advancements by the industry, resulting world applications to ensure the successful implementa-
in a lack of follow-up. In this context, review studies tion of cyclone technology breakthroughs in industrial
serve as crucial guides and are of great importance. environments. It is essential to bridge this gap in order to
Further research should be conducted on the industrial improve the actual usefulness and effectiveness of
applications of cyclone separators and their usage in vari- cyclone separators in different industrial applications.
ous sectors. This includes potential applications
in industries such as mining, food production, energy gen- AUTHOR CONTRIBUTIONS
eration, and waste management. Additionally, the integra- Ece Aylı: Conceptualization; writing – original draft;
tion of advanced technologies such as nanotechnology, 3D methodology; validation; visualization; writing – review
printing, or artificial intelligence is foreseen as a method and editing; formal analysis; supervision. Eyup Kocak:
that could enhance the performance and efficiency of Writing – review and editing; visualization; validation.
cyclone separators. Artificial intelligence is considered to
be a particularly useful tool for enhancing cyclone separa- A C KN O WL ED G EME N T S
tor performance. For instance, AI tools can reveal the We acknowledge the use of various figures sourced from
impact of parameters affecting cyclone performance with other publications throughout this review. Permissions
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 37

for the reproduction of these figures have been obtained [26] W. Peng, P. J. A. J. Boot, A. C. Hoffmann, H. W. A. Dries,
from the respective publishers and copyright holders. J. Kater, A. Ekker, Ind. Eng. Chem. Res. 2001, 40, 5649.
[27] C. Song, B. Pei, M. Jiang, B. Wang, D. Xu, Y. Chen, Powder
Technol. 2016, 294, 437.
CONFLICT OF INTEREST STATEMENT
[28] Z.-W. Gao, Z. X. Liu, Y. D. Wei, C. X. Li, S. H. Wang,
The authors declare that they have no known competing X. Y. Qi, W. Huang, Pet. Sci. 2022, 19, 848.
financial interests or personal relationships that could [29] A. J. Hoekstra, J. J. Derksen, H. E. A. Van Den Akker, Chem.
have appeared to influence the work reported in this Eng. Sci. 1999, 54, 2055.
paper. [30] J. J. Derksen, H. E. A. Van den Akker, AIChE J. 2000, 46,
1317.
P EE R R EV IE W [31] J. J. Derksen, Comput. Fluids 2005, 34, 301.
[32] E. Dehdarinejad, M. Bayareh, F. Parvaz, S. H. Hosseini, G.
The peer review history for this article is available at
Ahmadi, Chem. Eng. Res. Des. 2023, 193, 587.
https://www.webofscience.com/api/gateway/wos/peer-re
[33] Z. W. Zhang, Q. Li, Y. H. Zhang, H. L. Wang, Purif. Technol.
view/10.1002/cjce.25526. 2022, 286, 120394.
[34] K. Elsayed, C. Lacor, Appl. Math. Model. 2011, 35, 1952.
DATA AVAILABILITY STATEMENT [35] Q. Wei, G. Sun, J. Yang, Chem. Eng. Sci. 2019, 204, 287.
The data that support the findings of this study are avail- [36] J. Yang, G. Sun, C. Gao, Purif. Technol. 2013, 105, 15.
able from the corresponding author upon reasonable [37] M. Siadaty, S. Kheradmand, F. Ghadiri, Adv. Powder Technol.
request. 2017, 28, 1459.
[38] L. Wang, C. B. Parnell, B. W. Shaw, in Proceedings of 2001
Beltwide Cotton Production Conferences, National Cotton
ORCID Council, Anaheim, CA 2001.
Ece Aylı https://orcid.org/0000-0002-6209-161X [39] X. Li, J. Song, G. Sun, M. Jia, C. Yan, Z. Yang, Y. Wei, Can.
Eyup Kocak https://orcid.org/0000-0002-1544-2579 J. Chem. Eng. 2016, 94, 2373.
[40] Z. Ji, X. Wu, M. Shi, Acta Petrolei Sinica Petroleum Processing
R EF E RE N C E S Section, Nanyang Technological University, Singapore
[1] J. W. Strutt, Proc. R. Soc. London 1917, 93, 148. 1993, p. 86.
[2] G. I. Taylor, Proc. R. Soc. London 1923, 104, 213. [41] F. Qian, M. Zhang, Comput. Chem. Eng. 2005, 29, 2155.
[3] G. I. Taylor, Proc. R. Soc. London 1921, 100, 114. [42] H. S. Bryant, R. W. Silverman, F. A. Zenz, Hydrocarbon
[4] G. I. Taylor, Philos. Trans. R. Soc. London 1923, 223, 289. Process. 1983, 62, 87.
[5] A. J. ter Linden, Proc. - Inst. Mech. Eng. 1949, 160, 233. [43] Q. Wei, G. Sun, C. Gao, Powder Technol. 2020, 369, 321.
[6] C. J. Stairmand, Trans. Inst. Chem. Eng. 1951, 29, 356. [44] H. M. El-Batsh, Appl. Math. Model. 2013, 37, 5286.
[7] J. L. Smith, J. Basic Eng. 1962, 84, 609. [45] V. Kumar, K. Jha, Simulation 2020, 96, 31.
[8] D. Leith, D. Mehta, Atmos. Environ. 1973, 7, 527. [46] V. Kumar, K. Jha, Purif. Technol. 2019, 215, 25.
[9] R. M. Alexander, Proc. - Australas. Inst. Min. Metall. 1949, [47] Y. Zheng, X. Liu, L. Ni, Chem. Eng. Process. Process Intensif.
152, 202. 2022, 181, 109125.
[10] M. W. First, Fundamental Factors in the Design of Cyclone [48] F. Ficici, A. Vedat, M. Kapsiz, Int. J. Phys. Sci. 2010, 5, 804.
Dust Collectors, Harvard University, Cambridge 1950. [49] A. Raoufi, M. Shams, M. Farzaneh, R. Ebrahimi, Chem. Eng.
[11] W. Barth, Brennstow-Wäerme-Kraft 1956, 8, 1. Process. Process Intensif. 2008, 47, 128.
[12] D. Leith, W. Licht, AIChE Symp. Ser. 1972, 68, 196. [50] P. A. Patterson, R. J. Munz, Can. J. Chem. Eng. 1996, 74, 213.
[13] M. Lippmann, T. L. Chan, Am. Ind. Hyg. Assoc. J. 1974, [51] Z. Wang, G. Sun, Y. Jiao, Chem. Eng. Process. Process Intensif.
35, 189. 2020, 158, 108188.
[14] J. M. Beeckmans, C. J. Kim, Can. J. Chem. Eng. 1977, 55, 640. [52] H. Lee, S. J. Yook, Powder Technol. 2014, 262, 36.
[15] K. W. Lee, J. A. Gieseke, W. H. Piispanen, Atmos. Environ. [53] K. S. Lim, S. B. Kwon, K. W. Lee, J. Aerosol Sci. 2003, 34, 1085.
1985, 19, 847. [54] Y. Zhu, M. C. Kim, K. W. Lee, Y. O. Park, M. R. Kuhlman,
[16] J. C. Kim, K. W. Lee, Aerosol Sci. Technol. 1990, 12, 1003. Aerosol Sci. Technol. 2001, 34, 373.
[17] W. D. Griffiths, F. Boysan, J. Aerosol Sci. 1996, 27, 281. [55] I. A. Fetuga, O. T. Olakoyejo, S. M. Abolarin, A. O. Adelaja,
[18] D. L. Iozia, D. Leith, Aerosol Sci. Technol. 1989, 10, 491. S. O. Ologunoba, O. Oluwatusin, O. Fetuata, O. O. Adewumi,
[19] R. Thorn, Met. Finish. 1998, 96, 30. U. D. Robinson, A. M. de Oliveira Siqueira, Alexandria
[20] H. Mothes, F. Loeffler, Ger. Chem. Eng. 1985, 8, 223. Eng. J. 2023, 78, 453.
[21] M. Bohnet, Chem. Eng. Process. Process Intensif. 1995, 34, 151. [56] I. Khazaee, Heat Mass Transfer 2017, 53, 2009.
[22] H. Safikhani, P. Mehrabian, Adv. Powder Technol. 2016, [57] J. Chen, Z. Jiang, B. Yang, Y. Wang, F. Zeng, Chin. J. Chem.
27, 379. Eng. 2022, 44, 8.
[23] W. Peng, A. C. Hoffmann, P. J. A. J. Boot, A. Udding, [58] M. Wasilewski, L. S. Brar, Powder Technol. 2023, 428,
H. W. A. Dries, A. Ekker, J. Kater, Powder Technol. 2002, 118849.
127, 212. [59] B. Zhao, Y. Su, J. Zhang, Chem. Eng. Res. Des. 2006, 84, 1158.
[24] P. W. Dietz, AIChE J. 1981, 27, 888. [60] H. Yoshida, Y. Inada, K. Fukui, T. Yamamoto, Powder Tech-
[25] H. Mothes, F. Loeffler, Int. Chem. Eng. 1988, 28, 231. nol. 2009, 193, 6.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
38 AYLI and KOCAK

[61] L. S. Brar, K. Elsayed, Powder Technol. 2017, 311, 465. [96] N. Prasanna, K. Subramanian, S. Ajay, T. Rajagopal,
[62] C. E. Lapple, Chem. Eng. 1951, 58, 144. V. Vigneshwaran, Mater. Today: Proc. 2020, 43, 1960.
[63] P. Swift, Steam Heat. Eng. 1969, 38, 453. [97] B. Gopalakrishnan, G. Saravana Kumar, K. A. Prakash, Adv.
[64] S. Demir, A. Karadeniz, M. Aksel, Powder Technol. 2016, Powder Technol. 2023, 34, 103959.
295, 209. [98] P. Katare, A. Krupan, A. Dewasthale, A. Datar, A. S. Dalkilic,
[65] H. Safikhani, M. A. Akhavan-Behabadi, M. Shams, Case Stud. Therm. Eng. 2021, 28, 101384.
M. H. Rahimyan, Adv. Powder Technol. 2010, 21, 435. [99] R. Rathish, P. Krishnakumar, G. Anandh, V. B. Kumar,
[66] S. Pandey, I. Saha, O. Prakash, T. Mukherjee, J. Iqbal, T. Bharathkumar, S. Dhanapal, Mater. Today: Proc. 2020, 37, 537.
A. K. Roy, M. Wasilewski, L. S. Brar, Appl. Sci. 2022, 12, 4904. [100] S. Venkatesh, R. Suresh Kumar, S. P. Sivapirakasam,
[67] L. S. Brar, R. P. Sharma, K. Elsayed, Powder Technol. 2015, M. Sakthivel, D. Venkatesh, S. Yasar Arafath, Powder Tech-
286, 668. nol. 2020, 371, 115.
[68] A. C. Hoffmann, M. De Groot, W. Peng, H. W. A. Dries, J. [101] G. Chen, J. Fan, P. Zhang, W. Wang, J. Taiwan Inst. Chem.
Kater, AIChE J. 2001, 47, 2452. Eng. 2020, 115, 160.
[69] S. Ito, K. Ogawa, C. Kuroda, J. Chem. Eng. Jpn. 1980, 13, 6. [102] S. Venkatesh, M. Sakthivel, H. Saranav, N. Saravanan,
[70] T. Akiyama, T. Marui, Powder Technol. 1989, 58, 181. M. Rathnakumar, K. K. Santhosh, Powder Technol. 2020, 361,
[71] R. Shastri, R. P. Sharma, L. S. Brar, Part. Sci. Technol. 2022, 1070.
40, 337. [103] S. Wang, H. Li, R. Wang, X. Wang, R. Tian, Q. Sun, Adv. Pow-
[72] R. Shastri, L. S. Brar, Purif. Technol. 2020, 249, 117149. der Technol. 2019, 30, 227.
[73] L. S. Brar, R. P. Sharma, Mater. Today: Proc. 2015, 2, 3230. [104] D. Misiulia, S. Antonyuk, A. G. Andersson, T. S. Lundström,
[74] W. B. Faulkner, M. D. Buser, D. P. Whitelock, B. W. Shaw, Powder Technol. 2020, 364, 943.
Trans. ASABE 2008, 51, 287. [105] K. W. Chu, B. Wang, D. L. Xu, Y. X. Chen, A. B. Yu, Chem.
[75] M. Modabberifar, H. Nazaripoor, H. Safikhani, Adv. Powder Eng. Sci. 2011, 66, 834.
Technol. 2021, 32, 4295. [106] M. Eyvazi, F. Asaadi, S. Akhbarifar, M. R. Yousefi,
[76] C. B. Parnell, in Proc. 1996 Beltwide Cott. Prod. Conf., 1996, M. Shirvani, S. H. Hashemabadi, A. K. Rezaei, Chem. Eng.
pp. 1666–1670. Res. Des. 2023, 199, 192.
[77] K. Pant, C. T. Crowe, P. Irving, Powder Technol. 2002, 125, 260. [107] W. Li, Z. Huang, G. Li, C. Ye, Purif. Technol. 2022, 296, 121370.
[78] Y. Huang, M. Zhang, J. Lyu, Z. Liu, H. Yang, Chem. Eng. Res. [108] S. Yuan, G. Sun, G. Cao, Y. Wu, Y. Yue, Z. Song, Powder
Des. 2018, 136, 906. Technol. 2024, 438, 119605.
[79] E. Feifel, Forsch. Geb. Ingenieurwes. 1938, 9, 68. [109] D. Misiulia, A. G. Andersson, T. S. Lundström, Chem. Eng.
[80] S. Obermair, J. Woisetschläger, G. Staudinger, Powder Tech- Technol. 2015, 38, 1425.
nol. 2003, 138, 239. [110] D. Misiulia, A. G. Andersson, T. S. Lundström, Chem. Eng.
[81] L. Zhu, S. Wang, Y. Ru, J. Wang, P. Yang, A. Li, Z. Ma, Res. Des. 2015, 102, 307.
Z. Wang, Particuology 2023, 82, 134. [111] J. J. H. Houben, C. Weiss, E. Brunnmair, S. Pirker, J. Appl.
[82] Y. R. Kang, J. B. Kwak, Machines 2023, 11, 440. Fluid Mech. 2016, 9, 487.
[83] S. Obermair, G. Staudinger, Chem. Ing. Tech. 2001, 73, 203. [112] M. Wasilewski, L. Singh Brar, G. Ligus, Purif. Technol. 2021,
[84] S. Obermair, G. Staudinger, Chem. Eng. Technol. 2001, 24, 1259. 274, 119020.
[85] B. Malgil, Design and Analysis of Coaxial Cyclone Separator [113] R. Sripriya, N. Suresh, S. Chandra, D. Bhattacharjee, Miner.
which Is Intended for Separation of Fire Causing Big Particles Eng. 2013, 42, 1.
in Dust Collectors and Evaluation of the Improvement [114] E. Yohana, M. Tauviqirrahman, B. Yusuf, K. H. Choi,
Possibilities, Istanbul Teknik University, Turkey 2019. V. Paramita, Powder Technol. 2021, 377, 464.
[86] W. Li, Z. Huang, G. Li, Chem. Eng. Sci. 2023, 268, 118404. [115] M. Dasar, R. S. Patil, Powder Technol. 2020, 363, 541.
[87] D. Kumar, K. Jha, V. Kumar, L. S. Brar, Powder Technol. [116] M. Dasar, R. S. Patil, J. Heat Transfer 2020, 142, 042104.
2024, 438, 119660. [117] M. Dasar, R. S. Patil, Proc. Inst. Mech. Eng., Part C 2020, 234,
[88] Y. Yao, M. Shang, X. Ke, J. Zhang, Z. Huang, T. Zhou, J. Lyu, 3509.
Purif. Technol. 2024, 332, 125753. [118] M. Xu, L. Yang, X. Sun, J. Wang, L. Gong, J. Taiwan Inst.
[89] M. Q. Cao, J. Y. Chen, B. Hu, H. Meng, S. Y. Li, Y. D. Wei, Chem. Eng. 2019, 96, 419.
Purif. Technol. 2024, 337, 126437. [119] W.-E. Conrad, H.-G. Conrad, T. Szylowlee (U.S. Patent and
[90] G. Cao, G. Sun, Y. Yue, Z. Wang, Adv. Powder Technol. 2024, Trademark Office), US6,312,594, 2002.
35, 104298. [120] J. Chen, X. Lu, H. Liu, C. Yang, Chem. Eng. J. 2007, 129, 85.
[91] H. Yang, N. Wang, Y. Cao, X. Meng, L. Yao, Adv. Powder [121] T. C. Hsiao, D. Chen, P. S. Greenberg, K. W. Street, J. Aerosol
Technol. 2023, 34, 103929. Sci. 2011, 42, 78.
[92] M. Guo, Y. Lu, C. Xue, X. Sun, J. Y. Yoon, Adv. Powder Tech- [122] W. Xu, Q. Li, J. Wang, Y. Jin, Purif. Technol. 2016, 160, 112.
nol. 2023, 34, 104149. [123] G. E. O. Celis, J. B. R. Loureiro, P. L. C. Lage, A. P. Silva
[93] M. Guo, H. Xue, J. Pang, D. K. Le, X. Sun, J. Y. Yoon, Powder Freire, Chem. Eng. Sci. 2022, 248, 117099.
Technol. 2023, 416, 118236. [124] Z. Yan, J. Zhou, Y. Liu, C. Lu, Chem. Eng. Process. Process
[94] M. A. El-Emam, L. Zhou, A. I. Omara, Biosyst. Eng. 2023, Intensif. 2022, 179, 109077.
233, 125. [125] Z. Xiong, Z. Ji, X. Wu, Powder Technol. 2013, 253, 644.
[95] H. Fatahian, E. Fatahian, R. Erfani, Powder Technol. 2023, [126] D. Misiulia, A. G. Andersson, T. S. Lundström, Chem. Eng.
428, 118789. Technol. 2017, 40, 709.
 1939019x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cjce.25526 by Firat University, Wiley Online Library on [15/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
AYLI and KOCAK 39

[127] D. Misiulia, S. Antonyuk, A. G. Andersson, T. S. Lundström, [140] B. Zhao, H. Shen, Y. Kang, Powder Technol. 2004, 145, 47.
Powder Technol. 2018, 336, 45. [141] Y. Su, A. Zheng, B. Zhao, Powder Technol. 2011, 210, 293.
[128] D. I. Misiulia, V. V. Kuz’Min, V. A. Markov, Theor. Found. [142] W. P. Adamczyk, A. Klimanek, R. A. Białecki, G. Węcel,
Chem. Eng. 2013, 47, 274. P. Kozołub, T. Czakiert, Particuology 2014, 15, 129.
[129] D. Misiulia, presented at 13th Workshop on two-phase flow [143] Y. Liu, M. Ersson, H. Liu, P. Jönsson, Y. Gan, Steel Res. Int.
predictions: Halle (Saale), Germany, 2012. 2019, 90, 1800494.
[130] S. Pandey, L. S. Brar, Powder Technol. 2023, 425, 118584. [144] S. Heinz, Prog. Aeronaut. Sci. 2020, 114, 100597.
[131] L. Dong, M. Fan, H. Yang, Int. J. Miner. Process. 2013, [145] C. H. Zhou, Comput. Fluids 2017, 143, 73.
122, 43. [146] A. Jaiswal, M. D. Bui, P. Rutschmann, Fluids 2022, 7, 337.
[132] B. Pei, L. Yang, K. Dong, Y. Jiang, X. Du, B. Wang, Powder [147] ANSYS, Fluent Theory Guide (Realease 18.2), Release 18,
Technol. 2017, 313, 135. Ansys Inc., Canonsburg 2011.
[133] S. Ganegama Bogodage, A. Y. T. Leung, J. Hazard. Mater. [148] E. Bumrungthaichaichan, Powder Technol. 2022, 396, 327.
2016, 311, 100. [149] Z. Zhao, L. Zhou, L. Bai, B. Wang, R. Agarwal, Arch. Comput.
[134] D. K. Le, J. Y. Yoon, Chem. Eng. Process. Process Intensif. Methods Eng. 2024, 31, 871.
2020, 150, 107867. [150] M. A. El-Emam, L. Zhou, W. Shi, C. Han, Powder Technol.
[135] F. Parvaz, S. H. Hosseini, K. Elsayed, G. Ahmadi, Purif. Tech- 2021, 385, 357.
nol. 2018, 201, 223. [151] R. Mei, Int. J. Multiph. Flow 1992, 18, 145.
[136] B. Xie, S. Li, H. Jin, S. Hu, F. Wang, F. Zhou, Powder Technol. [152] F. Chen, H. Xiong, X. Wang, Z.-Y. Yin, Acta Geotech. 2023,
2018, 339, 695. 18, 335.
[137] S. Fu, F. Zhou, G. Sun, H. Yuan, J. Zhu, Adv. Powder Technol.
2021, 32, 931.
[138] S. Venkatesh, S. P. Sivapirakasam, M. Sakthivel, S. How to cite this article: E. Aylı, E. Kocak, Can.
Ganeshkumar, M. Mahendhira Prabhu, M. Naveenkumar,
J. Chem. Eng. 2024, 1. https://doi.org/10.1002/cjce.
Powder Technol. 2021, 383, 93.
[139] A. Sakin, I. Karagoz, A. Avci, Chem. Eng. Process. Process
25526
Intensif. 2019, 144, 107630.