Available via license: CC BY
Content may be subject to copyright.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
*Corresponding author:
Phone: +903562521616 (2888)
E-mail: hilal.isleroglu@gop.edu.tr
https://doi.org/10.17113/ftb.58.01.20.6544 original scientific paper
Evaluation of Process Conditions for Ultrasonic Spray-Freeze Drying of
Transglutaminase
Running title: Ultrasonic Spray-Freeze Drying of Transglutaminase
Hilal Isleroglu* and Izzet Turker
Tokat Gaziosmanpasa University, Faculty of Natural Sciences and Engineering,
Food Engineering Dept., 60150 Tokat, Turkey
Received: 16 October 2019
Acepted: 11 March 2020
SUMMARY
In this study, a commercial transglutaminase enzyme was dried using an ultrasonic spray-freeze
drying method and the effects of the process conditions (ultrasonic spraying and freeze-drying) were
optimized to maximize the final transglutaminase activity. Accordingly, process parameters affecting
enzyme activity were selected, such as nozzle frequency (48 and 120 kHz), flow rate (2, 5 and 8
mL/min) and plate temperature for secondary drying (25, 35 and 45 °C). Moreover, the effects of
different pH levels (pH=2.0 and pH=9.0) and high temperature (80 °C) on enzyme activity, physical
properties and particle morphology of transglutaminase were discussed. According to the results,
transglutaminase preserved its activity despite ultrasonic spray-freeze drying. Activity was even
enhanced regarding the sonication process. The optimum process conditions were determined using
the desirability function method to be a 3.10 mL/min flow rate, 45 °C plate temperature and 120 kHz
nozzle frequency. The predicted activity ratio was 1.17, and this was validated experimentally as
having an activity ratio of 1.14±0.02. Furthermore, enzyme produced by ultrasonic spray-freeze drying
had low moisture values (2.92-4.36 %) at 8 hours of drying time. When the morphological structure
of the transglutaminase particles produced by ultrasonic spray-freeze drying at the optimum
conditions was examined, spherical particles with pores on their surfaces were observed. In addition,
flowing properties of the transglutaminase powders were considered as fair under most conditions
according to the Carr index.
Key words: spray-freeze drying, ultrasonication, transglutaminase, optimization, physical properties,
particle morphology
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
INTRODUCTION
Through recent developments in the food industry, usage of food additives obtained from natural
sources to enhance food properties has been becoming popular. Microbial enzymes produced from
different strains can be used for numerous purposes for production of more desirable food products
or development of the novel and unique products. In this respect, protein modification has been
becoming an important issue especially in foods having high protein content. Because of the proteins
are one of the most important components of food products, modification of the food proteins via
different methods (chemical, physical or enzymatic) can provide producing novel products and can
improve the functional properties (1). In this regard, transglutaminase (TG) is the most effective
enzyme because of its unique properties. Transglutaminase (EC 2.3.2.13) is a transferase catalyzing
the acyl transfer reactions between γ-carboxyamide group of protein-bound glutamine residues and
primary amines, deamidation of protein-bound glutamines and cross-linking glutamine and lysine
peptide residuals (2-4). TG, is an extracellular enzyme and generally produced by fermentation of
Streptomyces strains. TG acts in the pH range of 5.0-8.0, it has ability being active at 40-70 °C. Also,
metal ions (Ca2+) or cofactors are not needed for activation of the enzyme. Because of these unique
characteristics, TG is noted as a reliable additive for foods in terms of human digestion (1,5,6). In
consequence of its cross-linking properties, a variety of proteins (soy proteins, whey proteins,
albumins, myofibrillar proteins) can be suitable substrates for TG (1,2). Thus, TG can alter the
mechanical and textural properties of meat, dairy and bakery products. Many studies have been
reported about the successful usage of TG to develop the functional properties of different food
products (7-9).
In the food industry, sonication techniques are generally used for the inactivation of enzymes. This
inactivation effect comes from extreme conditions of sonication which can cause breakdown of
hydrogen bonds and Van der Waals interactions in enzyme molecules (10). However, ultrasonic
applications can not inactivate all kind of enzymes. When sonication and ambient conditions are
suitable, enzymes can be still stable in terms of their biological activity. There are several studies in
literature showing that sonication applications have not had negative effects on some enzymes such
as γ-glutamyltranspeptidase, lactoperoxidase and chymotrypsin (11,12). It is also known that some
proteins are resisted to sonication and some conformational changes of these proteins may even
enhance the enzyme activity (13,14).
Drying of biological materials such as enzymes is an important topic due to their sensitiveness to the
process conditions. Lately, spray-freeze drying (SFD) has drawn attention for obtaining the powder
forms of enzymes both in literature and food industry (15). SFD is a process consisting of mainly three
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
steps: (1) atomization of the bulk solution/liquid, (2) freezing of the atomized droplets by cryogenic
fluids and (3) drying of frozen particles by sublimation at low temperature and pressure (15,16). It is
asserted that enzymes can be easily dried by SFD because of the low temperature during the
operation and shorter drying times achieved via atomization step which the fine droplets can be
generated (17). Different techniques can be preferred for the atomization of bulk solution and
ultrasonic spraying is the most effective one to generate the fine droplets (15,18).
When enzymes or protein containing solutions are subjected to drying operations, stress factors such
as freezing and dehydration may affect the biological activity of such products adversely (19). Sonner
et al. (20) studied the stability of tryipsinogen activity during the ultrasonic spray freeze drying (USFD)
process. They revealed that the activity loss of trypsinogen mainly occurred at the freeze-drying step.
Likewise, Yu et al. (21) investigated the influence of freeze-drying and spray-freezing processes on
the biological activity of lysozyme and they reported that drying step decreased the enzyme activity.
On the other hand, the adverse effects of drying step might be eliminated using the ultrasonic nozzles
due to positive effects of ultrasonic applications on the enzymes in terms of activity. Parallel to this
concept, Isleroglu et al. (22) revealed that the activity of microbial TG was enhanced by ultrasonic
atomizing. Furthermore, Ishwarya et al. (15) stated that usage of ultrasonication in SFD can enhance
the controlling of particle size distribution and can help production of the porous particles. Hence,
D’addio et al. (18) carried out USFD process successfully to obtain powders of lysozyme having
controlled particle size distribution. Despite USFD is considered as a suitable method for drying of
protein-containing solutions, there are only a small number of studies in literature revealing the
success of it (20,23). For this reason, the number of studies in which USFD and enzymes are jointly
used needs to be increased in order to explain the interactions between them.
The scope of the current study is the optimization of ultrasonic spraying and drying conditions of a
commercial TG during USFD to get the highest final enzyme activity. Moreover, physical properties
of TG such as bulk and tapped densities, moisture content, water activity and wettability under
different conditions were determined. The effect of high temperature and different pH levels for each
drying condition were also discussed and particle morphology of the TG powder achieved by the
optimization of the USFD was investigated.
MATERIALS AND METHODS
Materials
A commercial form of TG (Benosen, Tegen 20X, China), which was produced by fermentation of
Streptomyces strains and powdered through spray-drying using only maltodextrin as a bulking agent,
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
was used to prepare enzyme solutions. The moisture content and the enzyme activity of the
commercial TG powder were 5.41±0.18% and 4298.75±483.94 U/(g protein), respectively. 15 g of
powdered enzyme were dissolved in 60 mL of distilled water, and this solution was prepared freshly
before every drying operation. For determination of enzyme activity, the γ-glutamyl donor substrate
of TG (Sigma, Z-Gln-Gly, Germany) was used.
Determination of specific enzyme activity
The specific activity of TG was determined using the hydroxamate method and Bradford protein assay
(24) and defined as U/g protein. The hydroxamate method was carried out to determine the enzyme
activity in U/mL as previously described by Isleroglu et al. (22). To calculate the specific enzyme
activity in U/g protein, a modified Bradford protein assay was used as follows: First, Bradford solution
was prepared. For the preparation of Bradford dye solution, 50 mg of brilliant blue G-250 was
dissolved in 50 mL of pure methanol, then 100 mL of 85 % (m/V) H3PO4 was added and the mixture
was diluted to 1 L with distilled water. The resulting dye solution was stored at +4 °C until use for
analysis, and the solution was filtered by Whattman No.1 filter paper before assays. For the analysis,
300 µL of enzyme solution was mixed with 700 µL of dye solution and the absorbance of samples
was measured at 595 nm. The protein content of the samples was calculated using bovine serum
albumin as a standard and the enzyme activity of samples was determined using Eq. 1.
Specific enzyme activity (U g protein
⁄ )=Enzyme activity (U mL
⁄)
Protein amount ( g protein mL
⁄) /1/
Ultrasonic spray-freeze drying of TG and experimental design
The commercial TG solutions were dried by USFD using different nozzles (SonoTek Inc., USA), an
ultrasonic generator (ECHO, SonoTek Inc., USA) and a syringe pump (Syringe Pump TI, SonoTek
Inc., USA). Nozzle frequencies were 48 and 120 kHz, flow rates were 2, 5 and 8 mL/min and plate
temperatures of secondary drying phase were set at 25, 35 and 45 °C. The USFD rig was set up
according to Isleroglu et al. (22). 50 mL of enzyme solution (20 %, m/m) was atomized and droplets
began to be frozen by the nitrogen vapor. Then, freezing of the droplets was finished when they totally
sank into the liquid nitrogen. At each condition, the frozen particles were dried for 6 hours at 1 mbar
for main drying and 2 hours at 0.01 mbar for secondary drying (CHRIST, Alpha 1-4 LSC Plus,
Germany). Plate temperatures were set at the beginning of the secondary drying phase. Flow rate
(X1), plate temperature (X2) and nozzle frequency (X3) were determined as independent variables.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
Three-level factorial design was used for the experimental setup (25). Flow rate and plate
temperatures were selected as numeric factors, nozzle frequency was a categoric factor having two
treatments (48 and 120 kHz), and the center point for each treatment was set as 5. The experimental
design of 26 runs is given in Table 1.
To determine the total effect of the USFD process on TG activity, activity ratio values were used. Initial
enzyme activity of the enzyme solutions was determined by measuring the activity of fresh solutions
prepared prior to the drying operation. After drying operations were carried out, final activity of powder
samples was measured by rehydration of the resulting products to their initial dry matter percentage
(20 %), matching the fresh solutions used before drying. Activity ratios were calculated using Eq. 2.
Activity ratio = Final activity of sample (U g protein
⁄)
Initial enzyme activity of solution ( (U g protein
⁄) /2/
Table 1.
Investigation of the effect of different pH and temperature on TG
The effect of high temperature on the enzyme activity was investigated at 80 °C by using an incubation
temperature of 80 °C instead of 37 °C, as in the standard enzyme activity assay (Hydroxamate
method). To determine the effect of different pH levels on TG activity, buffer solutions (pH=2.0 and
pH=9.0) were used instead of pH=6.0 buffer solution as in standard enzyme activity assay. The results
were also defined as activity ratios.
Physical properties of TG
Moisture content
All samples were analyzed by an infrared moisture analyzer (Shimadzu, MOC 63U, Japan) to
determine the moisture content. The analyses were carried out at 90 °C using the device’s automatic
mode. All analyses were duplicated.
Water activity
The water activity measurements of the powder samples were carried out with a water activity
measurement device (Aqualab, 3TE, USA) and all measurements were done in parallel.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
Bulk and tapped densities and flowing behaviors
The masses of the powder samples and the volume of the samples at the same mass were measured
to determine the bulk density (ρb) values. First, approximately 2 g of sample was weighed and the
exact mass was recorded. Then the sample was poured into a measuring cylinder (25 mL) and the
volume was determined. ρb values were calculated by dividing the mass of powders by their volumes.
After the volume values were determined for ρb, the measuring cylinder was tapped 100-150 times
until a steady volume was reached to calculate tapped density (ρt). The ρt of the samples were
calculated by dividing the mass of the samples by their tapped volume (22).
Carr index (CI) and Hausner ratio (HR) values were also calculated for bulk and tapped densities to
determine the flow behavior of the powder samples (26,27). Eq. 3 and Eq. 4 were used to calculate
CI and HR, respectively.
CI= ρt-ρb
ρt
×100 /3/
HR= ρt
ρb
/4/
Wettability
Wettability assay of the powder samples was carried out according to the method described by
Isleroglu et al. (22). 0.1 g of samples were analyzed for every run and the wettability values were
identified as the time (s) needed for the powders to get totally wet. All analyses were duplicated.
Particle morphology
Particle morphology analysis was carried out by scanning electron microscopy for the sample
obtained by the determined optimum process condition (Zeiss, Evo LS 10, Germany). Aluminum
stages with adhesive carbon-conducting tape on their surfaces were covered with samples, and the
samples were coated using gold (200 seconds). Images were taken at different magnifications. The
analyses were done under a 9.75 x 10-5 torr vacuum and accelerating voltage of 20 kV.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
Statistical analysis
Experimental data obtained from the design was fitted to a second-order polynomial model and the
desirability function method was used to determine the optimum process conditions to ensure the
highest activity ratio. Analysis of variance was used to determine the significant variables at a
confidence level of 95 %, and the p-values of lack of fit were expected to be over 0.05. Verification of
the optimum point was carried out by one sample t-test. After investigating the effects of different pH
and temperature on TG, the data were analyzed by Duncan difference test (28).
RESULTS AND DISCUSSION
All activity ratio values obtained are shown in Table 1. According to the results, the activity ratio values
ranged between 0.76 and 1.31 (Table 1). One can observe that in the most conditions the final activity
level was higher than the initial activity level of the enzyme solution. Considering that the protein
content of the samples could not be changed by the USFD process, it can be assumed that the activity
(U/mL) of the enzyme is increased under most of the process conditions. It is known that the stability
of enzymes under sonication conditions is unique and specific, as the conformational structures of all
kinds of enzymes are different (14). Isleroglu et al. (22) properly exhibited the positive effect of the
ultrasound treatment on TG activity. It was found that the activity after sonication and before drying
was higher than initial activity of the enzyme (activity ratio between 1.02 and 1.31). Furthermore,
Isleroglu et al. (29) observed the same results for microencapsulated TG. According to their results,
the enzyme activity increased during the atomization and freezing stage, which was not related to the
coating type or other process conditions. These studies showed that atomization and freezing did not
harm the TG; conversely, its activity was slightly enhanced.
Many studies have shown the inhibitory effect of the sonication process on different kinds of enzymes.
Raviyan et al. (30) studied the thermal and thermo-sonication inactivation kinetics of pectin
methylesterase in tomatoes, revealing that the sonication process positively affected the inhibition of
the enzyme. Likewise, Tian et al. (31) revealed that activity of trypsin in an aqueous medium was
decreased with an increase in ultrasound power (20 kHz, 100-500 W). Although sonication processes
are mainly used for inactivation of enzymes in literature, there are several studies showing positive
effects or at least not inducing effects of ultrasound treatments on enzyme activity. Manas et al. (32)
investigated the effect of ultrasound on egg white lysozyme and found that a 15-minute-long
application of ultrasound at 20 kHz frequency did not affect enzyme activity. Apar et al. (33) also
demonstrated that alpha-amylases produced from Bacillus species were not inhibited by 20 kHz of
sonication. Moreover, Gebicka and Gebicki (34) detected that catalase was not inactivated under the
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
conditions of 22 kHz frequency and 5 °C of ambient temperature. Similar to our study, Froment et al.
(35) determined a slight transient enzyme activation for human butyryl-cholinesterase at varying
density input, using 20 kHz of sonication frequency. They concluded that the small changes in the
catalytic activity might be caused by ultrasound-induced slight conformational changes affecting the
active site reactivity. In our study, it is thought that the active TG site was rearranged by ultrasonic
forces, which might lead to a positive effect on enzyme activity when suitable process conditions are
applied. As stated by some researchers, ultrasound treatments at proper frequencies and intensity
may cause an increase in enzyme activity through physical and biochemical effects (10).
For USFD processes, flow rate and nozzle frequency are the most important parameters affecting the
success of the operation (15). Moreover, drying time is an important factor that can lead to critical
decreases in enzyme activity (20). In our previous study, in which microbial TG was used, it was
determined that atomization did not affect the TG activity negatively; however, the activity decreased
at the drying step drastically (22). In this study, TG activity was thought to be enhanced by ultrasonic
atomization. This increase in the enzyme activity at the atomization step might have defeated the
negative effects of drying, such as dehydration stress on enzyme activity. Under these circumstances,
the effect of the sonication process was prominently seen.
The mathematical model that expresses the relationship between the process variables was formed
by multiple linear regression analysis. For this purpose, linear effect terms for each variable were first
used. Then, quadratic and interaction effect terms were added to the model, and the increase in sum
of squares and lack of fit values were analyzed. The variance analysis table is given in Table 2. The
model was found to be statistically significant at the confidence level of 95 % (p<0.05). According to
data from Table 2, the linear effects of different parameters did not affect the activity ratio significantly
(p>0.05). Interaction of flow rate and nozzle frequency affected the activity ratio significantly (p<0.05);
however, the enzyme activity was not affected by the plate temperature (p>0.05) according to the
model used.
Table 2.
3D response surface graph and contour lines are shown in Fig. 1a for 120 kHz nozzle frequency when
the activity ratio was chosen as a response. The circular form of the lines demonstrates that the
interaction between flow rate and plate temperature is not significant as previously shown in the
ANOVA table (Table 2). Furthermore, the linear trend of the plate temperature (x2) axis showed that
plate temperature did not have a significant impact on activity ratio (p>0.05). The second-order
polynomial model (obtained from regression analysis for the activity ratio used for the optimization
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
study) is given in Eq. 5 using coded variables. The relationship between the values estimated from
Eq. 5 and the experimental values for the activity ratio is shown in Fig. 1b. It can be seen that the
model’s activity ratio estimates were consistent with experimental data.
Activity ratio = +0.21101+0.10973X1+0.032279X2-8.70804×10-4X1X2-0.0114X1
2-2.96269×10-4X2
2 /5/
Fig. 1.
TG is an enzyme that can be used in various food systems of different pH values. At different pH
values, the activity changes of TG are an important topic affecting its usage in the process. To
determine the process conditions on TG activity at different pH levels, buffer solutions of pH=2.0 and
pH 9.0 were used instead of pH=6.0 Tris buffer solution as described in the activity assay (22). At
pH=2.0, activity ratios ranged between 0.25 and 0.53 (Fig. 2a). Likewise, TG activity was also inhibited
at pH=9.0 and activity ratios were in the range of 0.34-0.60 (Fig. 2b). TG activity was decreased
dramatically under all conditions; however, at least 25 % of the activity was preserved. Isleroglu et al.
(22) demonstrated that minimum and maximum relative activity values ranged between 27 and 67 %
at pH=2.0 and pH=9.0. However, Cui et al. (36) demonstrated that microbial TG lost all activity at
lower pH values. In this study, none of the process conditions statistically affected the activity ratios
of different pH levels (p>0.05) (Table 2). Nevertheless, the effect of the sonication process is thought
to be an important factor because of leading conformational changes on the surface of enzyme
molecules, which render TG more stable at different pH levels.
Temperature is another substantial parameter for the TG activity. The effect of high temperature on
sonicated TG was investigated to reveal the resistance of the enzyme activity. Similar to the changes
at different pH levels, the activity ratios decreased more sharply at high temperature for both nozzle
frequencies. Nevertheless, at 80 °C with 1 hour of incubation, the enzyme still had an activity ratio of
at least 0.23 (Fig. 2c). When ANOVA was run for model parameter effects on activity ratios of high
temperature, only nozzle frequency had a significant statistical effect (p<0.05) (Table 2). The greater
resistance of TG produced with a 120 kHz nozzle to 80 °C than that of a 48 kHz nozzle can be
explained by a stronger structural conformational change due to higher frequency.
Fig. 2.
For the optimization study, activity ratio values were used to determine the effects of independent
variables on TG activity. The desirability function method was chosen to obtain the highest activity
ratio at different conditions-namely, flow rate, plate temperature and nozzle frequency. The optimum
point was determined as 3.10 mL/min of flow rate, 45 °C of plate temperature and 120 kHz nozzle
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
frequency with a desirability of 0.76. Estimated activity ratio was calculated as 1.17, and validation of
estimated optimum point prediction was carried out under the selected conditions. The experimental
result (activity ratio) at the optimum point was 1.14±0.02. Experimental results did not differ
statistically when compared with the values predicted by the model according to the paired t-test
results (p>0.05).
The physical properties of TG powders under each USFD condition were also determined (Table 3).
Moisture content, which is one of the most important parameters for powdered products, can
especially affect the storage stability of biological materials. Water activity is considered as a key
parameter for storage, like moisture content, and these parameters should be controlled wisely (37).
The moisture values ranged between 2.92 and 4.36 %, and all samples showed lower water activity
than the value of 0.035, which is under the limit of our water activity measurement device. The results
obtained were very promising for an enzyme powder, which means that if this powder form can be
stored carefully, the enzyme can remain active for very long times. Because of their very low water
activity, samples produced under any condition in our study might have longer shelf-life in terms of
less possibility of oxidation and microbial growth (38). It is also assumed that the generation of fine
crystal formations at the rapid-freezing step of USFD may lead to the production of porous particles,
resulting in lower moisture content and water activity (17). When individual particles have a porous
surface upon atomization and freezing, drying times can get shorter and moisture values can be lower
(15). Table 4 represents the effects of the process parameters and their interactions on the physical
properties of TG powder samples. According to the results, plate temperatures had the most
statistically significant effect on moisture values (p<0.05), and when the plate temperature was
increased, the moisture values of powdered TG was decreased. In this respect, the highest plate
temperature selected in the study (45 °C) did not have an impact on the activity ratio (p>0.05) and
also provides the production of powder products with low moisture content. Bulk and tapped densities
of the powder products are important parameters in terms of packaging, transportation and storage
(39,40). In this study, bulk and tapped densities were calculated in the range of 195-257 kg/m3 and
268-334 kg/m3, respectively (Table 3). However, it was determined that none of the process
parameters or their interactions statistically affected bulk or tapped density (p>0.05) (Table 4).
Wettability is considered an important reconstitution property, and wettability of powder products is
mainly affected by their drying method (41). Our results showed that wettability values obtained at a
120 kHz nozzle frequency were lower, and this phenomenon might be related to the smaller particle
size of the samples regarding the higher sonication frequency (Table 3). The inverse relation between
particle size and atomization frequency has been previously stated in different studies (22,42).
ANOVA results showed that only nozzle frequency statistically affected wettability (Table 4). The
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
flowability of powders and their flow behaviors at different conditions are very important for processing
and transportation operations (43). In this study, flowability of the powders was defined by Carr index
and Hausner ratio. The critical compressibility value between free-flowing (granular) and non-free-
flowing (powder) is about 20-25 %. Most of the samples showed ‘fair’ flowability while some of them
exhibited ‘good’ flowability in terms of Carr’s classification (26). The linear effect of flow rate and plate
temperature significantly impacted (p<0.05) on Carr index and Hausner ratio values, whereas nozzle
frequency and interactions of the independent variables were not (p>0.05) (Table 4). The SEM images
of TG produced at the optimum point (Fig. 3) show the spherical shape of the sample and the fair
flowability of samples can be explained by the generation of the spherical particles. These tend to
decrease the cohesive forces, resulting in an increase in flowing ability. Similar results were obtained
by other researchers’ studies on spray-freeze drying (44).
Table 3
Table 4
The observation of the microstructure of spray-freeze dried products is important to determine
whether the collapse phenomenon is absent or not, which can be a signature of successful drying
operation (17). In our study, SEM images of TG produced at the optimum point were taken at
magnifications of 100, 1000, 3000 and 5000X (Fig. 3). At 100X magnification, tiny spherical particles
having moderate particle size distribution were observed (Fig. 3a). When a 1000X image was
investigated, spherical particles having pores on their surface were clearly seen (Fig. 3b). The same
spherical particles were observed 3 times closer at 3000X magnification (see Fig. 3c), and it was
observed that the spherical particles had small fine pores on their surfaces depending on the
sublimation of ice crystals. In Fig. 3d, 5000X magnification was used and the fine pores were observed
in much better detail, indicating that no collapse occurred during the drying operation. The fine and
small pores are thought to have formed because of fast freezing rates and pore sizes much smaller
than freeze-dried products (17). Our findings were consistent with spray-freeze studies found in
literature (17,45). The low moisture values of the USFD samples in our study might be strongly related
with particle morphology. The porous structure of TG particles produced by USFD might also had an
impact on the samples’ low wettability values (Table 3). It is thought that TG particles rapidly dissolve
due to the fast infusion of water through micropores. Rogers et al. (46) reported similar findings in
their characterization study of milk powders produced by SFD. Moreover, the TG activity might be
enhanced by configuration of the surface proteins of the enzyme, followed by generation of the
particles having micropores on their surface due to fast freezing and sublimation.
Fig. 3.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
CONCLUSION
In this study, the USFD process conditions for obtaining TG powder were optimized to attain the
highest possible activity ratio. Moreover, physical properties of the powders produced using USFD
were determined, and the process parameters affecting their physical properties were considered.
The effects of the process on the activity ratios at different pH levels and at high temperature were
also investigated. Additionally, SEM images of the powdered TG were taken in order to observe the
microstructure of the particles. The results showed that the USFD process did not negatively affect
TG in terms of its biological activity. Furthermore, TG activity increased after the whole process. Low
moisture content and fair flowability of TG samples were also obtained by USFD. Moreover, the
spherical and porous surface characteristics of USFD powders were shown by SEM images. The
findings of this study indicate that TG powder can be produced by the USFD method while increasing
its activity. Even though USFD is an expensive technique for drying, it can be a feasible way to
produce TG in powder form for large-scale productions in the long term because of its low moisture
content and the resulting higher activity of the enzyme powders.
FUNDING
This research was supported by the Scientific and Technological Research Council of Turkey
(TUBITAK, Project Number: 115O216) financially.
CONFLICT OF INTEREST
The authors have no conflict of interest to declare.
ORCID ID
H. Isleroglu https://orcid.org/0000-0002-4338-9242
I. Turker https://orcid.org/0000-0003-0107-1962
REFERENCES
1. Gaspar ALC, de Góes-Favoni SP. Action of microbial transglutaminase (MTGase) in the
modification of food proteins: A review. Food Chem. 2015;171:315-22.
https://doi.org/10.1016/j.foodchem.2014.09.019
2. Lorenzen PC. Effects of varying time/temperature-conditions of pre-heating and enzymatic cross-
linking on techno-functional properties of reconstituted dairy ingredients. Food Res
Int. 2007;40(6):700-8.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
https://doi.org/10.1016/j.foodres.2006.12.001
3. Zhang D, Zhu Y, Chen J. Microbial transglutaminase production: understanding the
mechanism. Biotechnol Genet Eng. 2009;26(1):205-22.
https://doi.org/10.5661/bger-26-205
4. Queirós RP, Gouveia S, Saraiva JA, Lopes-da-Silva JA. Impact of pH on the high-pressure
inactivation of microbial transglutaminase. Food Res Int. 2019;115:73-82.
https://doi.org/10.1016/j.foodres.2018.08.002
5. Kuraishi C, Yamazaki K, Susa Y. Transglutaminase: its utilization in the food industry. Food Rev
Int. 2001;17(2):221-46.
https://doi.org/10.1081/FRI-100001258
6. Aalami M, Leelavathi K. Effect of microbial transglutaminase on spaghetti quality. J Food
Sci. 2008;73(5):C306-12.
7. Romeih E, Walker G. Recent advances on microbial transglutaminase and dairy application.
Trends Food Sci Tech. 2017;62:133-40.
https://doi.org/10.1111/j.1750-3841.2008.00741.x
8. Aaron L, Torsten M. Microbial transglutaminase: a new potential player in celiac disease. Clin
Immunol. 2019;199:37-43.
https://doi.org/10.1016/j.clim.2018.12.008
9. Liu Y, Liu Y, Xu Z, Shan M, Ge X, Zhang, Y, et al. Effects of Bacillus subtilis transglutaminase
treatment on the functional properties of whey protein. LWT. 2019;116:108559.
https://doi.org/10.1016/j.lwt.2019.108559
10. Mawson R, Gamage M, Terefe NS, Knoerzer K. Ultrasound in enzyme activation and inactivation.
In: Feng H, Bárbosa-Canovas GV, Weiss J, editors. Ultrasound technologies for food and
bioprocessing. New York, NY, USA: Springer; 2011. pp. 369-404.
https://doi.org/10.1007/978-1-4419-7472-3_14
11. Villamiel M, de Jong P. Influence of high-intensity ultrasound and heat treatment in continuous
flow on fat, proteins, and native enzymes of milk. J Agric Food Chem. 2000;48(2):472-78.
https://doi.org/10.1021/jf990181s
12. Ovsianko SL, Chernyavsky EA, Minchenya VT, Adzerikho IE, Shkumatov VM. Effect of
ultrasound on activation of serine proteases precursors. Ultrason Sonochem. 2005;12(3):219-23.
https://doi.org/10.1016/j.ultsonch.2003.10.012
13. Guzey D, Gulseren I, Bruce B, Weiss J. Interfacial properties and structural conformation of
thermosonicated bovine serum albumin. Food Hydrocolloid. 2006;20(5):669-77.
https://doi.org/10.1016/j.foodhyd.2005.06.008
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
14. Feng H, Barbosa-Cánovas, GV, Weiss J. editors. Ultrasound technologies for food and
bioprocessing. New York, USA: Springer; 2011.
https://doi.org/10.1007/978-1-4419-7472-3
15. Ishwarya SP, Anandharamakrishnan C, Stapley AG. Spray-freeze-drying: a novel process for the
drying of foods and bioproducts. Trends Food Sci Tech. 2015;41(2):161-81.
https://doi.org/10.1016/j.tifs.2014.10.008
16. Wanning S, Sueverkruep R, Lamprecht A. Pharmaceutical spray freeze drying. Int J
Pharm. 2015;488(1-2):136-53.
https://doi.org/10.1016/j.ijpharm.2015.04.053
17. Anandharamakrishnan C, Rielly CD, Stapley AG. Spray-freeze-drying of whey proteins at sub-
atmospheric pressures. Dairy Sci Technol. 2010;90(2-3):321-34.
https://doi.org/10.1051/dst/2010013
18. D’Addio SM, Chan JGY, Kwok PCL, Prud’homme RK, Chan HK. Constant size, variable density
aerosol particles by ultrasonic spray freeze drying. Int J Pharm. 2012;427(2):185-91.
https://doi.org/10.1016/j.ijpharm.2012.01.048
19. Mensink MA, Frijlink HW, van der Voort Maarschalk K, Hinrichs WL. How sugars protect proteins
in the solid state and during drying (review): mechanisms of stabilization in relation to stress
conditions. Eur J Pharm Biopharm. 2017;114:288-95.
https://doi.org/10.1016/j.ejpb.2017.01.024
20. Sonner C, Maa YF, Lee G. Spray‐freeze‐drying for protein powder preparation: Particle
characterization and a case study with trypsinogen stability. J Pharm Sci. 2002;91(10):2122-39.
https://doi.org/10.1002/jps.10204
21. Yu Z, Johnston K P, Williams III RO. Spray freezing into liquid versus spray-freeze drying:
influence of atomization on protein aggregation and biological activity. Eur J Pharm Sci.
2006;27(1):9-18.
https://doi.org/10.1016/j.ejps.2005.08.010
22. Isleroglu H, Turker I, Tokatli M, Koc B. Ultrasonic spray-freeze drying of partially purified microbial
transglutaminase. Food Bioprod Process. 2018;111:153-64.
https://doi.org/10.1016/j.fbp.2018.08.003
23. Nguyen XC, Herberger JD, Burke PA. Protein powders for encapsulation: a comparison of spray-
freeze drying and spray drying of darbepoetin alfa. Pharm Res. 2004;21(3):507-14.
https://doi.org/10.1016/j.fbp.2018.08.003
24. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein
utilizing the principle of protein-dye binding. Anal Biochem. 1976;72(1-2):248-54.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
https://doi.org/10.1016/0003-2697(76)90527-3
25. Design-Expert®, v. 7.0.0, Stat-Ease, Inc, MN, USA; 2005. Available from:
https://www.statease.com/software/design-expert/
26. Carr RL. Evaluating flow properties of solids. Chem Eng. 1965;72:163-68.
27. Hausner HH. Friction conditions in a mass of metal powder. Powder Metall. 1967;13:7-13.
28. SPSS Statistics, v. 21.0, IBM Corporation, NY, USA; 2012. Available from:
https://www.ibm.com/support/pages/spss-statistics-210-available-download
29. Isleroglu H, Turker I, Koc B, Tokatli M. Microencapsulation of microbial transglutaminase by
ultrasonic spray-freeze drying. Food Bioproc Tech. 2019;12(12):2004-17.
https://doi.org/10.1007/s11947-019-02353-4
30. Raviyan P, Zhang Z, Feng H. Ultrasonication for tomato pectinmethylesterase inactivation: effect
of cavitation intensity and temperature on inactivation. J Food Eng. 2005;70(2):189-96.
https://doi.org/10.1016/j.jfoodeng.2004.09.028
31. Tian ZM, Wan MX, Wang SP, Kang JQ. Effects of ultrasound and additives on the function and
structure of trypsin. Ultrason Sonochem. 2004;11(6):399-404.
https://doi.org/10.1016/j.ultsonch.2003.09.004
32. Manas P, Munoz B, Sanz D, Condon S. Inactivation of lysozyme by ultrasonic waves under
pressure at different temperatures. Enzyme Microb Tech. 2006;39(6):1177-82.
https://doi.org/10.1016/j.enzmictec.2005.11.053
33. Apar DK, Turhan M, Özbek B. Enzymatic hydrolysis of starch by using a sonifier. Chem Eng
Comm. 2006;193(9):1117-26.
https://doi.org/10.1080/00986440500354424
34. Gebicka L, Gebicki JL. The effect of ultrasound on heme enzymes in aqueous solution. J Enzym
Inhib. 1997;12(2):133-41.
https://doi.org/10.3109/14756369709035814
35. Froment MT, Lockridge O, Masson P. Resistance of butyrylcholinesterase to inactivation by
ultrasound: effects of ultrasound on catalytic activity and subunit association. BBA-Protein Struct
M. 1998;1387(1-2):53-64.
36. Cui L, Du G, Zhang D, Liu H, Chen J. Purification and characterization of transglutaminase from
a newly isolated Streptomyces hygroscopicus. Food Chem. 2007;105(2):612-18.
https://doi.org/10.1016/S0167-4838(98)00105-8
37. Syamaladevi RM, Tang J, Villa‐Rojas R, Sablani S, Carter B, Campbell G. Influence of water
activity on thermal resistance of microorganisms in low‐moisture foods: a review. Compr Rev
Food Sci F. 2016;15(2):353-70.
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
https://doi.org/10.1111/1541-4337.12190
38. Rao Q, Klaassen Kamdar A, Labuza TP. Storage stability of food protein hydrolysates—a review.
Cr Rev Food Sci. 2016;56(7):1169-92.
https://doi.org/10.1080/10408398.2012.758085
39. Walton DE. The morphology of spray-dried particles a qualitative view. Drying
Technol. 2000;18(9):1943-86.
https://doi.org/10.1080/07373930008917822
40. Fernandes RVDB, Borges SV, Botrel DA. Gum arabic/starch/maltodextrin/inulin as wall materials
on the microencapsulation of rosemary essential oil. Carbohyd Polym. 2014;101:524-32.
https://doi.org/10.1016/j.carbpol.2013.09.083
41. Hogekamp S, Schubert H. Rehydration of food powders. Food Sci Technol Int. 2003:9(3), 223-
35.
https://doi.org/10.1177/10820132030349387
42. Febriyenti MN, Mohamed N, Hamdan MR, Salleh SNM, Baie SBB. Comparison of freeze drying
and spray drying methods of Haruan extract. Int J Drug Deliv. 2014;6, 286-91.
43. Teunou E, Fitzpatrick JJ, Synnott EC. Characterisation of food powder flowability. J Food
Eng. 1999;39(1):1-37.
https://doi.org/10.1016/S0260-8774(98)00140-X
44. Rajam R, Anandharamakrishnan C. Spray freeze drying method for microencapsulation of
Lactobacillus plantarum. J Food Eng. 2015;166:95-103.
https://doi.org/10.1016/j.jfoodeng.2015.05.029
45. Hundre SY, Karthik P, Anandharamakrishnan C. Effect of whey protein isolate and β-cyclodextrin
wall systems on stability of microencapsulated vanillin by spray–freeze drying method. Food
Chem. 2015;174:16-24.
https://doi.org/10.1016/j.foodchem.2014.11.016
46. Rogers S, Wu WD, Saunders J, Chen XD. Characteristics of milk powders produced by spray
freeze drying. Drying Technol. 2008;26(4):404-12.
https://doi.org/10.1080/07373930801929003
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
Table 1. Experimental design and the obtained activity ratio values
Flow Rate (X1)
/(mL/min)
Plate Temperature (X2)
/°C
Nozzle Frequency (X3)
/kHz
Activity Ratio
5
35
48
1.10±0.00
8
25
48
1.04±0.04
8
25
120
0.76±0.01
5
35
120
1.31±0.06
2
25
120
0.88±0.01
5
35
120
1.06±0.01
5
25
48
1.01±0.01
5
35
48
1.13±0.00
8
45
48
1.06±0.00
8
35
120
0.87±0.01
5
35
48
1.02±0.01
5
35
48
1.01±0.01
5
35
120
1.14±0.01
2
45
48
0.91±0.00
5
35
48
1.02±0.02
5
35
120
1.08±0.01
5
35
120
1.05±0.01
8
35
48
1.17±0.02
2
25
48
0.96±0.03
5
25
120
1.09±0.00
5
45
48
1.12±0.00
2
35
120
0.97±0.01
5
45
120
1.12±0.01
2
45
120
1.22±0.03
8
45
120
0.82±0.00
2
35
48
1.03±0.00
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
Table 2. Statistical results for the effects of process parameters on activity ratio
Source
DF
Activity ratio
pH=6.0
(main condition)
pH=2.0
pH=9.0
80 °C
Sum of
squares
p-value
Sum of
squares
p-value
Sum of
squares
p-value
Sum of
squares
p-value
Model
8
0.22
0.0325
0.016
0.9496
0.042
0.1727
0.046
0.1126
x1
1
5.54x10-3
0.4541
2.35x10-3
0.5481
1.67x10-3
0.4753
0.011
0.0724
x2
1
0.022
0.1487
3.26x10-4
0.8220
4.53x10-5
0.9056
1.61x10-3
0.4669
x3
1
1.84x10-3
0.6648
3.27x10-4
0.8217
0.013
0.0598
0.016
0.0300
x1 x2
1
5.51x10-3
0.4553
2.39x10-3
0.5445
4.71x10-3
0.2360
3.67x10-4
0.7273
x1 x3
1
0.082
0.0091
6.74x10-3
0.3139
9.56x10-3
0.0982
7.56x10-3
0.1258
x2 x3
1
0.010
0.3067
2.00x10-3
0.5790
8.35x10-4
0.6117
1.99x10-3
0.4205
x12
1
0.059
0.0233
2.79x10-4
0.8353
0.012
0.0713
2.88x10-3
0.3341
x22
1
4.90x10-3
0.4810
1.61x10-3
0.6188
5.23x10-3
0.2131
7.07x10-3
0.1379
Residual
17
0.16
0.11
0.053
0.050
Lack of fit
Uygunsuzluğu
9
0.10
0.2876
0.072
0.1950
0.035
0.2436
0.039
0.0531
Pure error
8
0.060
0.034
0.019
0.010
Total
25
0.38
0.12
0.095
0.096
DF: Degree of freedom, X1: Flow rate/ (mL/min), X2: Plate temperature/°C, X3: Nozzle Frequency/kHz
(a)
(b)
Fig. 1. Effects of process conditions on activity ratio (a) 3D response surface graph for 120kHz, (b) the relation
between experimental and predicted values
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
(a)
(b)
(c)
Fig. 2. Effects of process conditions on activity ratio at (a) pH=2.0, (b) pH=9.0, (c) 80 °C.
Table 3. Physical properties of TG at different drying conditions
Nozzle
Frequency
/kHz
Plate
temperature
/°C
Flow rate
/(mL/min)
Moisture
Content/
%
Bulk density
/(kg/m3)
Tapped density
/(kg/m3)
Wettability
/s
CI
/%
HR
48
25
2
(3.31±0.20)a
(205.47±1.37)a
(298.56±3.75)ab
(4.21±0.04)ab
(31.17±0.40)abc
(1.45±0.01)bc
5
(3.90±0.45)a
(219.31±1.69)a
(313.97±4.27)ab
(6.71±0.30)ab
(30.14±0.41)bc
(1.43±0.01)bc
8
(3.63±0.24)a
(204.46±0.26)a
(249.19±0.31)b
(7.78±0.34)a
(17.95±0.00)e
(1.22±0.00)e
35
2
(3.45±0.17)a
(195.85±1.22)a
(267.67±1.67)ab
(3.78±0.25)b
(26.83±0.00)bcd
(1.37±0.00)bcd
5
(3.11±0.21)a
(216.47±25.43)a
(301.33±46.19)ab
(6.20±2.02)ab
(27.78±3.59)bcd
(1.39±0.07)bcd
8
(3.64±0.18)a
(205.86±3.86)a
(278.37±9.04)ab
(6.72±0.30)ab
(26.01±1.01)cd
(1.35±0.02)cd
45
2
(3.70±0.10) a
(200.82±0.72)a
(317.28±2.88)ab
(4.55±0.20)ab
(36.70±0.80)a
(1.58±0.02)a
5
(2.92±0.04)a
(209.49±2.09)a
(272.43±4.81)ab
(5.62±0.12)ab
(23.09±0.59)de
(1.30±0.01)de
8
(3.22±0.10)a
(225.55±1.88)a
(333.58±1.92)a
(5.78±0.28)ab
(32.38±0.95)ab
(1.48±0.02)ab
120
25
2
(3.70±0.18)zyx
(211.88±1.04)z
(307.85±0.30)z
(4.13±0.12)zy
(31.17±0.40)z
(1.45±0.01)zy
5
(3.65±0.17)zyx
(256.67±3.33)z
(320.77±0.77)z
(4.93±0.15)zy
(19.98±1.23)x
(1.25±0.02)x
8
(4.36±0.24)z
(202.40±1.60)z
(258.10±7.10)z
(3.43±0.19)y
(21.54±1.54)yx
(1.28±0.02)yx
35
2
(3.54±0.25)zyx
(202.44±0.44)z
(290.80±2.23)z
(4.05±0.12)zy
(30.38±0.38)zy
(1.44±0.01)zyx
5
(3.15±0.17)x
(228.23±56.71)z
(313.57±61.00)z
(5.53±1.04)z
(27.99±5.84)zyx
(1.40±0.01)zyx
8
(4.02±0.10)zy
(220.21±0.95)z
(298.86±1.29)z
(3.73±0.25)y
(26.32±0.00)zyx
(1.36±0.00)zyx
45
2
(3.49±0.24)yx
(215.99±5.18)z
(326.91±14.91)z
(3.37±0.24)y
(33.86±1.43)z
(1.51±0.03)z
5
(3.09±0.07)x
(200.10±0.00)z
(294.72±1.72)z
(4.59±0.15)zy
(32.10±0.40)z
(1.47±0.01)z
8
(3.37±0.07)yx
(231.37±0.74)z
(323.92±1.04)z
(3.70±0.22)y
(28.57±0.00)zyx
(1.40±0.00)zyx
(a–e) Means with uncommon superscripts within a column and flow rate are significantly different (p<0.05) for 48 kHz nozzle.
(z–x) Means with uncommon superscripts within a column and flow rate are significantly different (p<0.05) for 120 kHz nozzle.
None of the letters indicate comparison between the rows.
*Water activity values were not given in the table due to having smaller values than <0.035
Food Technology and Biotechnology 58 (1) 2020 www.ftb.com.hr
Please note that this is an unedited version of the manuscript that has been accepted for publication. This
version will undergo copyediting and typesetting before its final form for publication. We are providing this
version as a service to our readers. The published version will differ from this one as a result of linguistic and
technical corrections and layout editing.
Table 4. Statistical results for the effects of model parameters on physical properties
Source
DF
Moisture
Content/
%
Bulk density
/(kg/m3)
Tapped
density
/(kg/m3)
Wettability
/s
CI
/%
HR
p-value
Model
8
0.0010
0.9629
0.9348
0.0393
0.2167
0.2668
X1
1
0.1969
0.6126
0.6488
0.1066
0.0296
0.0337
X2
1
0.0026
0.8812
0.4166
0.3982
0.0411
0.0481
X3
1
0.1764
0.4278
0.4847
0.0143
0.9746
0.9639
X1x2
1
0.0251
0.5848
0.3543
0.7036
0.3205
0.4457
X1x3
1
0.2299
0.9290
0.8791
0.0569
0.9656
0.9491
X2x3
1
0.3461
0.7900
0.9851
0.6480
0.5770
0.6333
x12
1
0.0012
0.3611
0.5271
0.0483
0.5730
0.5986
x22
1
0.2997
0.8585
0.7784
0.7172
0.9792
0.8736
Residual
17
Lack of fit
9
0.2007
0.9977
0.9774
0.9943
0.6493
0.7002
Pure error
8
DF: Degree of freedom, X1: Flow rate/ (mL/min), X2: Plate temperature/°C, X3: Nozzle Frequency/kHz
(a)
(b)
(c)
(d)
Fig. 3. Particle morphology of obtained TG by scanning electron microscopy at different magnification (a) 100x,
(b) 1000x, (c) 3000x and (d) 5000x
