Extending Applications of High-Pressure Homogenization by Using Simultaneous
Emulsification and Mixing (SEM).An Overview
Vanessa Gall 1,*, Marc Runde 2 and Heike P. Schuchmann 1
1 Institute of Process Engineering in Life Sciences, Section I: Food Process Engineering,
Karlsruhe Insitute of Technology, Kaiserstraße 12, 76131 Karlsruhe, Germany; heike.schuchmann@kit.edu
2 mixolutions engineering, Oskar-von-Miller-Str. 23, 60314 Frankfurt, Germany; marc.runde@mixolutions.de
* Correspondence: vanessa.gall@kit.edu; Tel.: +49-721-608-42196
Academic Editor: Andreas Hakansson
Received: 30 September 2016; Accepted: 17 November 2016; Published: 26 November 2016
Abstract: Conventional high-pressure homogenization (HPH) is widely used in the pharmaceutical,
chemical, and food industries among others. In general, its aim is to produce micron or sub-micron
scale emulsions with excellent product characteristics. However, its energy consumption is still
very high. Additionally, several limitations and boundaries impede the usage of high-pressure
homogenization for special products such as particle loaded or highly concentrated systems. This article
gives an overview of approaches that have been used in order to improve the conventional high-pressure
homogenization process. Emphasis is put on the ‘Simultaneous Emulsification and Mixing’ process
that has been developed to broaden the application areas of high-pressure homogenization.
Keywords: high-pressure homogenization; mixing; process modifications; process intensification;
energy efficiency
1. Introduction
Emulsions are systems of at least two immiscible liquids in which one of the liquids is dispersed
in the other as small droplets. They can be produced by using different emulsification systems such
as rotor-stator systems, membrane systems, ultrasonic systems, and high-pressure systems [1.3].
High-pressure homogenizers are widely used in the pharmaceutical, chemical, and food industries [4].
They consist of a high-pressure pump and a disruption unit and enable a continuous homogenization.
In general, only high-pressure systems can achieve the energy density needed to produce submicron
emulsions [2,5,6] while at the same time providing the high throughputs required in industrial
processes. Small droplet sizes lead to retarded creaming or sedimentation and improve product
characteristics such as creaminess, texture, viscosity, color, bioavailability of active ingredients,
and shelf life stability [4,7.9].
As the industry increasingly demands emulsions with smaller mean droplet diameters and
narrower droplet size distributions, the objectives of the current research are:
.
Enhancing the stabilization of the disrupted droplets [2,10,11].
.
Decreasing the required energy input [2,10,11].
.
Implementing more durable materials for the construction of the disruption unit which is
commonly prone to wear by cavitation and particle abrasion [12].
.
Enhancing the understanding of the HPH process in order to increase its efficiency [11].
Furthermore, conventional high-pressure homogenizers do have specific limitations and boundaries:
Processes 2016, 4, 46; doi:10.3390/pr4040046
www.mdpi.com/journal/processes
Processes 2016, 4,46
2of15
.
The composition of an emulsion to be processed is limited. For example, particle-loaded systems
such as particle stabilized emulsions (Pickering emulsions) and particle containing nano-carrier
systems can cause abrasion in the disruption unit [13,14].
.
Coalescence and agglomeration can occur at higher volume fractions of the disperse phase (e.g.,
in dairy homogenization) [10,15]. In general, coalescence is more likely to occur in high-pressure
homogenizers compared to rotor-stator-systems because of the high energy input at extremely
short residence times [16].
.
Emulsifiers need to meet specific criteria such as fast adsorption kinetics in order to fulfill their
purpose in high-pressure homogenizers [17]. Some of the emulsifiers typically used in the food
industry (biopolymers or proteins), for example, are heat-or pressure sensitive [1,7] which
complicates the production of emulsions at higher temperatures.
One potential approach for overcoming these limitations and boundaries is the ‘Simultaneous
Emulsification and Mixing’ (SEM) process. This process is based on a modified disruption unit which
combines the unit operations mixing and emulsification by inserting a micromixer shortly after the
disruption unit [4,18,19].
This article will focus on the properties and possible applications of the SEM process, but will
also introduce other approaches made in order to improve the conventional homogenization process.
It will begin with providing some theoretical background of the topic (Chapter 2). Chapter 3 will
focus on the improvements of the high-pressure process that have been developed in the last years
including the SEM process. Some applications of the SEM process will be discussed in Chapter 4.
Finally, Chapter 5 will provide a short summary and outlook.
2. Theoretical Background
This chapter will give an introduction into the process of high-pressure homogenization (HPH).
Furthermore, it will summarize the most important characteristics of mixing, since those are
important for the operating principle of the SEM process discussed later on.
2.1. High-Pressure Homogenization
Usually, HPH is conducted in two steps. First, a coarse emulsion is produced. Then the droplet
sizes are reduced in a high-pressure homogenizer [15]. The idea of combining a high-pressure pump
with a disruption unit in order to produce small-scale emulsions dates back to 1899 [20]. Piston pumps
with pressure ranges between 20 and 4000 bar are most commonly used [3]. The high-pressure
pump builds up energy that is then relaxed after the disruption unit and therefore leads to droplet
disruption [16]. The energy input during droplet size reduction can be expressed as the energy
density Ev that describes the average energy input per emulsion volume [21,22]. In case of high
pressure homogenization processes, the energy density equals the pressure drop in the disruption
unit. The obtained droplet diameter decreases with increasing pressure difference [21].or increasing
energy density.unless coalescence occurs [23].
The achievable droplet sizes also depend on the disruption unit as they influence the flow pattern
which in turn determines droplet breakup. Disruption units can be divided by their flow guidance
into radial diffusers, counterjet dispergators, and axial nozzle aggregates [2]. Typical representatives
of these types are shown in Figure 1.
Processes 2016, 4,46 3of15
Processes 2016, 4, 46 3 of 15
Figure 1. Disruption units in high-pressure homogenization. (1) radial-diffusers (reproduced with the
permission from [3]; Behrs Verlag, 2012); (2) examples for counter-jet-dispergator: jet-dispergator
adapted from [24], Microfluidizer® adapted from [25]; (3) axial flow nozzle-system.
Radial diffusers contain an axially mobile valve seat [21,26] which enables the variation of the
flow rate by varying the slit width. Counterjet dispergators include a collision area of two or more
opposed jets of the emulsion [16,21]. Both the counterjet dispergators and the axial flow-nozzle-
systems contain no movable parts which makes them suitable for very high pressures [2]. Nozzle
aggregates can be distinguished by their axial flow direction. Simple orifices usually consist of round
holes of 0.1.2 mm diameter [10]. It has been reported that different disruption units lead to different
droplet sizes when the same energy density is applied. For example, flat valves are less energy
efficient than orifices or the Microfluidizer geometries, esp. for oil in water (o/w)-emulsions
containing high viscosity oil [23,26].
Different mechanisms can cause droplet breakup: laminar shear or elongation stresses, turbulent
stresses, and cavitation. According to recent reports, droplet breakup occurs after passing the
disruption valve. In orifices, for example, the droplets are first elongated in the inlet area [10] and
then disrupted in the turbulent and cavitating flow in the discharge area [26,27]. On top of that, the
components of the emulsions influence the flow regimes and thus the mechanisms predominantly
responsible for droplet breakup [28.31]. Further information on droplet breakup can be found in our
second paper in this journal.
2.2. Mixing
Mixing is a unit operation of process engineering in which several substances, that differ in at
least one property, are distributed in a defined volume. Its goal is to achieve homogeneity in order to
improve product quality, chemical or biological conversions, or heat- and mass-transfer [32]. In
theory, ideal mixing occurs when all starting materials are equally distributed instantly. In reality,
however, this rarely occurs, so that either the time needed for complete mixing (mixing time) or the
degree of mixing after a certain time (mixing quality) is used for characterization of the mixing quality
[18].
Macromixing is the rate-determining step in most mixing processes and is caused by the largest
scales of motion in the fluid. On the other hand, mixing on the smallest scale of motion and the final
scales of molecular diffusivity is called micromixing [33].
Continuous mixers are often used in continuous processes and can be classified by their
residence time, by their residence-time behavior, and by the way in which the mixing energy is
Figure 1. Disruption units in high-pressure homogenization. (1) radial-diffusers (reproduced with
the permission from [3]; Behrs Verlag, 2012); (2) examples for counter-jet-dispergator: jet-dispergator
adapted from [24], Microfluidizer® adapted from [25]; (3) axial flow nozzle-system.
Radial diffusers contain an axially mobile valve seat [21,26] which enables the variation of the flow
rate by varying the slit width. Counterjet dispergators include a collision area of two or more opposed
jets of the emulsion [16,21]. Both the counterjet dispergators and the axial flow-nozzle-systems contain
no movable parts which makes them suitable for very high pressures [2]. Nozzle aggregates can be
distinguished by their axial flow direction. Simple orifices usually consist of round holes of 0.1.2 mm
diameter [10]. It has been reported that different disruption units lead to different droplet sizes when
the same energy density is applied. For example, flat valves are less energy efficient than orifices or the
Microfluidizer geometries, esp. for oil in water (o/w)-emulsions containing high viscosity oil [23,26].
Different mechanisms can cause droplet breakup: laminar shear or elongation stresses, turbulent
stresses, and cavitation. According to recent reports, droplet breakup occurs after passing the
disruption valve. In orifices, for example, the droplets are first elongated in the inlet area [10] and
then disrupted in the turbulent and cavitating flow in the discharge area [26,27]. On top of that, the
components of the emulsions influence the flow regimes and thus the mechanisms predominantly
responsible for droplet breakup [28.31]. Further information on droplet breakup can be found in our
second paper in this journal.
2.2. Mixing
Mixing is a unit operation of process engineering in which several substances, that differ in at
least one property, are distributed in a defined volume. Its goal is to achieve homogeneity in order to
improve product quality, chemical or biological conversions, or heat-and mass-transfer [32]. In theory,
ideal mixing occurs when all starting materials are equally distributed instantly. In reality, however,
this rarely occurs, so that either the time needed for complete mixing (mixing time) or the degree of
mixing after a certain time (mixing quality) is used for characterization of the mixing quality [18].
Macromixing is the rate-determining step in most mixing processes and is caused by the largest
scales of motion in the fluid. On the other hand, mixing on the smallest scale of motion and the final
scales of molecular diffusivity is called micromixing [33].
Continuous mixers are often used in continuous processes and can be classified by their residence
time, by their residence-time behavior, and by the way in which the mixing energy is introduced.
Processes 2016, 4,46 4of15
The required residence time and residence time behavior are influenced by process conditions, reaction
kinetics (in case of biological or chemical reactions), and other factors [33].
Furthermore, the mixing energy can either be introduced by moving parts such as stirrers or it
can be withdrawn directly from the flowing medium like in the jet mixer, for example [33,34]. The jet
mixer can achieve rapid mixing in a short length of pipe mostly in turbulent flow regimes [33]. At least
one component has to be added to the main stream with a considerably higher velocity than the main
stream. The same working principle is used in the SEM process.
3. Developments in High-Pressure Homogenization
Several ideas to improve the conventional HPH process have been investigated in the last years.
These can be classified as follows: geometrical modifications of the disruption valve, inserting a second
homogenization step, and including a micromixer in the disruption unit. It should be mentioned
that in this paper, only systems based on the conventional HPH process containing a high-pressure
pump and a disruption unit are discussed in detail. However, desired droplet size distributions could
also be produced by alternative concepts such as jet homogenizer [35,36] or microfluidic systems [37].
Furthermore, this article will not go into detail on the possibilities of post-homogenization treatment,
since these are already discussed in [38].
3.1. Geometrical Modifications of the Disruption Valve
Flat valves can be divided into three generations. While the first generation had a flat valve seat
and stamp, the valve seat in the second generation was designed in a conical shape which induced
extended elongation and reduced total pressure loss. The third generation flat valve is characterized by
a wider valve diameter at a smaller homogenization gap [10]. Droplet disruption could be improved
in the second and third flat valve generation while applying the same energy density or pressure loss
compared to a first generation flat valve [15].
Stansted Fluid Power Ltd. (Stansted, UK) developed a modified flat valve made of ceramic
material that enables much higher levels of pressure (3500 bar) than a conventional flat valve [6].
Both the conventional and the Stansted valve consist of a valve piston and a valve seat, but the flow
directions through the valve are reversed. First, the fluid passes the mobile valve piston and is then
accelerated by the narrow gap between the piston valve and the piston seat [39]. Although very high
pressures could be realized, the achieved droplet sizes were limited due to coalescence.
Modifications of orifices have been constructed in order to influence flow conditions and thus
also to influence the mechanisms causing droplet breakup. Conical and circular inlets influence the
elongation of the droplets, while conical outlets enhance the stabilization of the droplets while at the
same time decreasing the pressure loss [27]. A modified orifice with two diagonal bores affects the
intensity and distribution of turbulence after the homogenization unit. It has been shown that these
orifices, also called two beam jet valves, result in smaller droplet sizes of an (o/w)-emulsion compared
to conventional orifices [40]. The smallest droplet sizes were found at angles of 60.
in relation to
the direction of flow. A similar principle is used in the Y-chamber of the Microfluidics® (Westwood,
MA, USA) disruption units [25]. Turbulence in the discharge area of the orifice can also be influenced
by inserting an impact bead [41] or by using a redirecting valve [42,43]. These modified orifices are
displayed in Figure 2.
However, according to Aguilar et al. [28], the geometry of the orifice should be adapted to the
material characteristics of the emulsion. The authors report that at low viscosity ratio between the
disperse and the continuous phase, droplets are easily elongated but also tend to relax faster after
elongation. Therefore, emulsions of low viscosity ratio require a fast build-up of turbulent flow, with
the intensity of turbulence being more important for efficient droplet breakup than the time scales
over which droplets are subjected to turbulent stresses. The opposite is true for emulsions of high
viscosity ratios.
Processes 2016, 4,46 5of15
Processes 2016, 4, 46 5 of 15
Figure 2. Modifications of orifices. (a) Concial orifice adapted from [27]; (b) Two beam jet valve
adapted from [44]; (c) Valve with impact bead adapted from [41]; (d) Redirecting valve adapted from
[43].
3.2. Inserting a Second Homogenization Step
In industrial homogenizers, a double stage process is often installed in order to increase the
emulsification efficiency of the HPH. Here, a second homogenization unit is inserted in line after the
first one. With regard to flow conditions, the second unit applies a back-pressure and thus influences
the pressure drop over the first device. This has implications particularly for the occurrence of
cavitation [45,46]. The ratio of back-pressure and inlet pressure is defined as a Thoma number (Th)
[46,47]. It is reported that cavitation decreases with increasing Th and disappears at Thoma numbers
of 0.3 < Th < 0.5 [48]. The Thoma numbers that also influence droplet breakup and minimum droplet
sizes were found at 0.15 < Th < 0.35 [48.50], depending on the high pressure unit and the emulsion
composition. Nevertheless, it was pointed out that only the first homogenization unit is responsible
for droplet breakup. Therefore, the second homogenization unit can also be replaced by other process
units inducing back-pressure, such as a needle valve or a pressure vessel [46]. In dairy
homogenization, the second homogenization unit is also responsible for deagglomerating fat globule
aggregates that result from casein bridging in the first homogenizing step [10,51].
The two homogenization units can be realized as two simple orifices or as modified valves.
Displacing the second homogenization unit axially can intensify the turbulent flow in the discharge
area of the first orifice [52]. Kolb et al. [5] stated that this new type of homogenization valve can
reduce droplet sizes compared to a Microfluidizer or a flat valve due to reducing coalescence in the
turbulent area between the two orifices. Karasch and Kulozik [53] proposed a double valve consisting
of two beam jet valves for dairy homogenization.
3.3. Simultaneous Emulsifying and Mixing
The basic idea of the ‘Simultaneous Emulsification and Mixing’ process is to combine the unit
operations mixing and emulsification in order to create synergies between them [18]. It should be
noted that the SEM process is also named ‘High Pressure Post Feeding’ (HPPF) process in some
publications [13,54,55]. Just as in the conventional process, a high-pressure pump is combined with
a disruption unit. The SEM disruption unit however combines a simple homogenization orifice and
a micromixer [4]. Two streams enter the SEM disruption unit (Figure 3): one stream which creates a
turbulent jet and an additional mixing stream [18]. The first one, passing the valve or orifice, is often
called the ‘homogenization stream’ as it is responsible for setting up flow conditions responsible for
droplet or filament breakup. It is also often called ‘main stream’ as it is placed centrally even if its
Figure 2. Modifications of orifices. (a) Concial orifice adapted from [27]; (b) Two beam jet valve adapted
from [44]; (c) Valve with impact bead adapted from [41]; (d) Redirecting valve adapted from [43].
3.2. Inserting a Second Homogenization Step
In industrial homogenizers, a double stage process is often installed in order to increase the
emulsification efficiency of the HPH. Here, a second homogenization unit is inserted in line after
the first one. With regard to flow conditions, the second unit applies a back-pressure and thus
influences the pressure drop over the first device. This has implications particularly for the occurrence
of cavitation [45,46]. The ratio of back-pressure and inlet pressure is defined as a Thoma number
(Th) [46,47]. It is reported that cavitation decreases with increasing Th and disappears at Thoma
numbers of 0.3 < Th < 0.5 [48]. The Thoma numbers that also influence droplet breakup and minimum
droplet sizes were found at 0.15 < Th < 0.35 [48.50], depending on the high pressure unit and the
emulsion composition. Nevertheless, it was pointed out that only the first homogenization unit is
responsible for droplet breakup. Therefore, the second homogenization unit can also be replaced by
other process units inducing back-pressure, such as a needle valve or a pressure vessel [46]. In dairy
homogenization, the second homogenization unit is also responsible for deagglomerating fat globule
aggregates that result from casein bridging in the first homogenizing step [10,51].
The two homogenization units can be realized as two simple orifices or as modified valves.
Displacing the second homogenization unit axially can intensify the turbulent flow in the discharge
area of the first orifice [52]. Kolb et al. [5] stated that this new type of homogenization valve can reduce
droplet sizes compared to a Microfluidizer or a flat valve due to reducing coalescence in the turbulent
area between the two orifices. Karasch and Kulozik [53] proposed a double valve consisting of two
beam jet valves for dairy homogenization.
3.3. Simultaneous Emulsifying and Mixing
The basic idea of the ‘Simultaneous Emulsification and Mixing’ process is to combine the unit
operations mixing and emulsification in order to create synergies between them [18]. It should be
noted that the SEM process is also named ‘High Pressure Post Feeding’ (HPPF) process in some
publications [13,54,55]. Just as in the conventional process, a high-pressure pump is combined with
a disruption unit. The SEM disruption unit however combines a simple homogenization orifice and
a micromixer [4]. Two streams enter the SEM disruption unit (Figure 3): one stream which creates
a turbulent jet and an additional mixing stream [18]. The first one, passing the valve or orifice, is often
called the ‘homogenization stream’ as it is responsible for setting up flow conditions responsible for
Processes 2016, 4,46 6of15
droplet or filament breakup. It is also often called ‘main stream’ as it is placed centrally even if its
throughput may be lower than that of the ‘mixing stream’ or ‘side stream’. The desired synergies
between emulsification and mixing can only be achieved when the mixing stream is induced shortly
after the disruptive unit where the disruptive flow conditions are fully built up [4,17,56,57].
Processes 2016, 4, 46 6 of 15
throughput may be lower than that of the ‘mixing stream’ or ‘side stream’. The desired synergies
between emulsification and mixing can only be achieved when the mixing stream is induced shortly
after the disruptive unit where the disruptive flow conditions are fully built up [4,17,56,57].
Figure 3. Function principle of SEM process.
The emulsion morphology produced by SEM can be influenced by geometrical parameters as
well as process and material parameters [4,18,56,57]. According to Kohler [18], the SEM process can
be operated in seven different operational modes (Figure 4). The type of operational mode depends
on whether pure phases or premixes are used and whether these phases are applied in the
homogenization or mixing stream [17,18].
Figure 4. Operational modes of SEM adapted from [19]. The different phases (continuous phase,
premix, disperse phase) are displayed in different colors according to the axes.
The operational modes of SEM can be classified as follows [18]:
Figure 3. Function principle of SEM process.
The emulsion morphology produced by SEM can be influenced by geometrical parameters as
well as process and material parameters [4,18,56,57]. According to Kohler [18], the SEM process can be
operated in seven different operational modes (Figure 4). The type of operational mode depends on
whether pure phases or premixes are used and whether these phases are applied in the homogenization
or mixing stream [17,18].
Processes 2016, 4, 46 6 of 15
throughput may be lower than that of the ‘mixing stream’ or ‘side stream’. The desired synergies
between emulsification and mixing can only be achieved when the mixing stream is induced shortly
after the disruptive unit where the disruptive flow conditions are fully built up [4,17,56,57].
Figure 3. Function principle of SEM process.
The emulsion morphology produced by SEM can be influenced by geometrical parameters as
well as process and material parameters [4,18,56,57]. According to Kohler [18], the SEM process can
be operated in seven different operational modes (Figure 4). The type of operational mode depends
on whether pure phases or premixes are used and whether these phases are applied in the
homogenization or mixing stream [17,18].
Figure 4. Operational modes of SEM adapted from [19]. The different phases (continuous phase,
premix, disperse phase) are displayed in different colors according to the axes.
The operational modes of SEM can be classified as follows [18]:
Figure 4. Operational modes of SEM adapted from [19]. The different phases (continuous phase,
premix, disperse phase) are displayed in different colors according to the axes.
The operational modes of SEM can be classified as follows [18]:
Processes 2016, 4,46
7of15
.
In the operational modes 2 and 6, pure phases are mixed in the micromixer. Primary droplets
are produced due to the turbulent current after the orifice. These operational modes can also be
referred to as simultaneous primary emulsification and mixing (SpEM).
.
In the operational modes 1, 3, and 4, the already existing droplets of the premixes are disrupted
into smaller droplets. These operational modes can also be referred to as simultaneous
homogenization and mixing (SHM).
.
In the operational modes 5 and 7, primary and secondary droplet breakup occurs since both pure
disperse phase and emulsion premix are inserted into the homogenizer.
The operational modes in which the disperse phase does not pass the orifice itself are based on
the idea that the homogenization stream delivers the local flow conditions required for emulsifying
the disperse phase and mixing both streams [58].
The operating principle of the SEM process can be realized using different disruption units.
Figure 5 illustrates some of the SEM valves that have been presented in the literature. They can
either be simple orifices, modified orifices, or even double orifices. SEM flat valves have also been
designed [18,59], but most of the recent research was conducted with orifices.
Processes 2016, 4, 46 7 of 15
. In the operational modes 2 and 6, pure phases are mixed in the micromixer. Primary droplets
are produced due to the turbulent current after the orifice. These operational modes can also be
referred to as simultaneous primary emulsification and mixing (SpEM).
. In the operational modes 1, 3, and 4, the already existing droplets of the premixes are disrupted
into smaller droplets. These operational modes can also be referred to as simultaneous
homogenization and mixing (SHM).
. In the operational modes 5 and 7, primary and secondary droplet breakup occurs since both
pure disperse phase and emulsion premix are inserted into the homogenizer.
The operational modes in which the disperse phase does not pass the orifice itself are based on
the idea that the homogenization stream delivers the local flow conditions required for emulsifying
the disperse phase and mixing both streams [58].
The operating principle of the SEM process can be realized using different disruption units.
Figure 5 illustrates some of the SEM valves that have been presented in the literature. They can either
be simple orifices, modified orifices, or even double orifices. SEM flat valves have also been designed
[18,59], but most of the recent research was conducted with orifices.
Figure 5. SEM valves. (a) orifice SEM valve adapted from [4]; (b) double SEM valve adapted from
[58]; (c) SEM valve with modified inlet geometry adapted from [55].
4. Applications of SEM Homogenization
The SEM process has been tested in various application fields. This chapter will give an overview
over the conducted research.
4.1. Preparation of Hybrid Nanoparticles
Hybrid nanoparticles are of scientific and industrial interest as they can be used for several
applications.e.g., paints of high color intensity [60.62], electronic devices [63,64], and medical
applications [65.67]. Hybrid nanoparticles can be prepared by miniemulsion polymerization in a
two-stage process [68.70]: first, a nanoparticle-in-monomer suspension is emulsified in a continuous
phase and then the polymerization of the filled submicron-sized monomer droplets is conducted [55].
Hecht et al. consider high-pressure homogenizers to be the most suitable device to produce small
sizes for particle filled monomer droplets at a high throughput [55]. Figure 6 displays some
nanostructured particles produced in dynamic high-pressure processes via miniemulsions [11].
Figure 5. SEM valves. (a) orifice SEM valve adapted from [4]; (b) double SEM valve adapted from [58];
(c) SEM valve with modified inlet geometry adapted from [55].
4. Applications of SEM Homogenization
The SEM process has been tested in various application fields. This chapter will give an overview
over the conducted research.
4.1. Preparation of Hybrid Nanoparticles
Hybrid nanoparticles are of scientific and industrial interest as they can be used for several
applications.e.g., paints of high color intensity [60.62], electronic devices [63,64], and medical
applications [65.67]. Hybrid nanoparticles can be prepared by miniemulsion polymerization in
a two-stage process [68.70]: first, a nanoparticle-in-monomer suspension is emulsified in a continuous
phase and then the polymerization of the filled submicron-sized monomer droplets is conducted [55].
Hecht et al. consider high-pressure homogenizers to be the most suitable device to produce small sizes
for particle filled monomer droplets at a high throughput [55]. Figure 6 displays some nanostructured
particles produced in dynamic high-pressure processes via miniemulsions [11].
Processes 2016, 4,46 8of15
In the past, ultrasonic systems have been used to produce particle loaded droplets [71.73].
Since nanoparticles can cause abrasion in homogenization valves [12], Hecht et al. used the SEM
process for the miniemulsion polymerization especially with regard to a high degree of nanoparticle
filling (up to 60 wt %) [55]. As disruption unit, a simple orifice (Figure 5a) was used. To avoid abrasion
of the orifice, operational mode 1 (see Figure 4) was chosen: pure continuous phase was pumped
through the orifice, while the premix emulsion containing the nanoparticle-in-monomer-droplets
was inserted as mixing stream. Droplets of the desired size range (<500 nm) were produced, which
demonstrates that the particle-loaded droplets did not require an elongation prior to their break-up [55].
Winkelmann et al. [44] used a similar SEM process to produce zinc oxide nanoparticles by miniemulsion
precipitation. Experimental investigations and computational fluid dynamics (CFD) simulations
showed that different factors influence the mixing quality: the homogenization pressure, the disruption
unit geometry, and the distance between the outlet of the disruption unit and the inlet of the second
feed stream. Mixing quality was shown to be responsible for the size of precipitated nanoparticles
especially when the process was run in the single-emulsion-mode. Here, precursor 1 of the particles
to be precipitated is dissolved in the miniemulsion droplets, while precursor 2 is mixed into the
continuous phase. Its transport into the miniemulsion droplet starts precipitation. On top of that, SEM
homogenizers can also be efficiently used to disperse nanoparticles in a liquid [13].
Processes 2016, 4, 46 8 of 15
In the past, ultrasonic systems have been used to produce particle loaded droplets [71.73]. Since
nanoparticles can cause abrasion in homogenization valves [12], Hecht et al. used the SEM process
for the miniemulsion polymerization especially with regard to a high degree of nanoparticle filling
(up to 60 wt %) [55]. As disruption unit, a simple orifice (Figure 5a) was used. To avoid abrasion of
the orifice, operational mode 1 (see Figure 4) was chosen: pure continuous phase was pumped
through the orifice, while the premix emulsion containing the nanoparticle-in-monomer-droplets
was inserted as mixing stream. Droplets of the desired size range (<500 nm) were produced, which
demonstrates that the particle-loaded droplets did not require an elongation prior to their break-up
[55]. Winkelmann et al. [44] used a similar SEM process to produce zinc oxide nanoparticles by
miniemulsion precipitation. Experimental investigations and computational fluid dynamics (CFD)
simulations showed that different factors influence the mixing quality: the homogenization pressure,
the disruption unit geometry, and the distance between the outlet of the disruption unit and the inlet
of the second feed stream. Mixing quality was shown to be responsible for the size of precipitated
nanoparticles especially when the process was run in the single-emulsion-mode. Here, precursor 1 of
the particles to be precipitated is dissolved in the miniemulsion droplets, while precursor 2 is mixed
into the continuous phase. Its transport into the miniemulsion droplet starts precipitation. On top of
that, SEM homogenizers can also be efficiently used to disperse nanoparticles in a liquid [13].
Figure 6. Nanostructured particles produced in dynamic high-pressure processes via miniemulsions.
Reproduced with the permission from [11]; Woodhead Publishing, 2016.
4.2. Particle Stabilized Emulsions
Particle stabilized emulsions (PSE), also called Pickering emulsions, use small particles to
stabilize the interface of emulsions [74,75]. They have regained interest in scientific literature [76] as
the availability of suitable particles has increased [58]. Since the energy supplied during
emulsification determines the droplet break-up for emulsions including PSE [77], HPH of PSE could
be of great interest [58], although the particles may cause abrasion as already discussed.
By using an SEM process in which the stabilizing particles are added in the mixing stream, this
problem can be reduced. Kohler et al. investigated the influence of process parameters, composition
and operational mode on the homogenization results of (o/w)- emulsions prepared in the SEM
process which were stabilized by Stober silica particles [58]. The investigated operational modes were
numbers 1, 3, and 6 (see Figure 4). In all operational modes, the particles were added in the mixing
stream. First experiments in the operational mode 3 revealed that small particles in the range of 12
nm were needed to achieve fast stabilization kinetics. Larger particles in the range of 200 nm resulted
Figure 6. Nanostructured particles produced in dynamic high-pressure processes via miniemulsions.
Reproduced with the permission from [11]; Woodhead Publishing, 2016.
4.2. Particle Stabilized Emulsions
Particle stabilized emulsions (PSE), also called Pickering emulsions, use small particles to stabilize
the interface of emulsions [74,75]. They have regained interest in scientific literature [76] as the
availability of suitable particles has increased [58]. Since the energy supplied during emulsification
determines the droplet break-up for emulsions including PSE [77], HPH of PSE could be of great
interest [58], although the particles may cause abrasion as already discussed.
By using an SEM process in which the stabilizing particles are added in the mixing stream, this
problem can be reduced. Kohler et al. investigated the influence of process parameters, composition
and operational mode on the homogenization results of (o/w)-emulsions prepared in the SEM process
which were stabilized by Stober silica particles [58]. The investigated operational modes were numbers
Processes 2016, 4,46 9of15
1, 3, and 6 (see Figure 4). In all operational modes, the particles were added in the mixing stream.
First experiments in the operational mode 3 revealed that small particles in the range of 12 nm were
needed to achieve fast stabilization kinetics. Larger particles in the range of 200 nm resulted in
significantly larger droplet sizes. At lower homogenizing pressures (100.500 bar), the obtained droplet
sizes in all operational modes were comparable to emulsions stabilized by a conventional homogenizer.
At higher pressures (800.1000 bar) however, droplet sizes could not be further reduced since the
stabilization kinetics of the particles was apparently not fast enough and droplets recoalesced.
4.3. Dairy Homogenization
For decades, drinking milk and dairy products have been homogenized either in full-stream or
partial-stream processes [51]. Homogenization reduces the milk fat globule diameter from around 4 m
to 0.6.0.7 m [78]. The lower droplet size is crucial to prevent e.g., creaming of the fat droplets within
the shelf life of milk. In the partial-stream process, the milk is first separated into cream and skim milk,
before both are mixed again to yield a fat content of maximum 17 vol % [10,51] and homogenized.
In full-stream homogenization, the fat content is adjusted to the final fat content (e.g., 3.5 vol %)
upstream the homogenization step and the whole volume is homogenized.
The partial-stream homogenization process allows a fat content up to 17 vol % upstream of the
homogenization step, which results in saved energy due to the reduced over-all processed volume.
In a downstream standardization step, the fat content is then adjusted to the final fat content.
Raising the fat content during milk homogenization above 17 vol % is the key to save energy
during homogenization [79]. Nevertheless, in conventional partial stream milk homogenization the
fat content cannot exceed 17 vol % due to coalescence and aggregation of the fat droplets [4] after the
homogenization step.
Kohler et al. [4] used SEM homogenization in order to enable the homogenization of cream with
up to 42 vol % fat content which corresponds to the concentration at which cream exits the separation
process in conventional dairy processing lines. In this case, the operational mode 3 and therefore
a SpEM process was used. The homogenized cream was diluted with skim milk, also coming from
the separation process, in the micromixer unit instantly after droplet breakup. Using this setup, fat
globule aggregation could be prevented while still allowing for the homogenization of increased fat
contents. Since the product volume to be pressurized was reduced, energy and investment costs could
be cut. Furthermore, it was possible to simplify the process line because two mixing units could be
eliminated. From the application point of view, the SpEM process allows an increase of the throughput
of a dairy process line by a factor of up to 8 without investment in new high-pressure pumps [10,79].
In order to further improve the SpEM process of milk homogenization, Kohler et al. [4] investigated
the influence of the distance between the exit of the orifice and the inlet of the mixing stream (distance
d in Figure 3) on the milk fat globule size and mixing quality. Both experiments and CFD simulations
of the process indicated that there is an optimal distance for the injection of the skim milk as mixing
stream. Short distances improve the mixing quality while long distances enable an undisturbed
disruption process. As compromise, the mixing stream was inserte