Journal of Food Engineering 139 (2014) 31–42

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Spray drying of orange peel extracts: Yield, total phenolic content,

and economic evaluation
Dian Shofinita ⇑, T.A.G. Langrish
Drying and Process Technology Group, School of Chemical and Biomolecular Engineering, The University of Sydney, Chemical Engineering Building J01, NSW 2006, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Orange peel extract powders were produced using two steps: microwave-assisted extraction and spray
Received 10 December 2013 drying. The extraction solvent-to-solid ratio has been found to significantly affect the level of total phe-
Received in revised form 24 February 2014 nolic compounds in the extract and powder. Under the outlet air temperatures between 43 and 79 °C,
Accepted 30 March 2014
spray drying of orange peel extracts resulted in a peak yield trend (between 75% and 92%) and high
Available online 30 April 2014
TPC (total phenolic content) recoveries. An evaluation of economics has also been conducted, suggesting
that this process is economically feasible, with profits of 6.1 USD/kg and 8.8 USD/kg for solvent to solid
Keywords:
ratios of 2 and 14, respectively. The uncertainty analysis of the economics showed that the selling price,
Phenolic compounds
Folin–Ciocalteu assay (FCR)
the labor cost and the orange peel cost are the three most important parameters compared with the costs
Antioxidant of electricity, natural gas, and water.
Microwave assisted extraction (MAE) Ó 2014 Elsevier Ltd. All rights reserved.
Manufacturing cost

1. Introduction
The consumption of antioxidants, which are abundant in fruits
and vegetables, has been suggested because of their benefits for
health (Liu, 2003). Antioxidants, such as vitamin C, vitamin E,
carotenoids, and phenolic compounds, have the ability to scavenge
reactive oxygen and nitrogen species and prevent oxidative dam-
age to important biological macromolecules, such as DNA, lipids,
and proteins (Carr and Frei, 1999). In the food industry, antioxi-
dants are used to maintain nutritional quality and avoid undesir-
able changes in the color, flavor, and texture (Finley and Given,
1986). Several synthetic antioxidants, such as butylated hydroxy-
toluene (BHT) and butylated hydroxyanisole (BHA), are commonly
used as food additives (Ajila et al., 2007). However, these antioxi-
dants have been reported to be potentially toxic and carcinogenic
(Ito et al., 1997). On the other hand, natural antioxidants have been
found to be safer than many synthetic ones and to provide addi-
tional nutrition value (Ajila et al., 2007). This has led to increasing
development of natural antioxidants, especially from fruits and
vegetables.
Citrus fruits have been explored as one of the sources for natu-
ral antioxidants (Bocco et al., 1998). Orange, as one of the main cit-
rus fruits grown in Australia, had a total production amount of
507,232 tonnes in 2008, which was 78% of the total produced cit-
rus fruits in Australia (Keogh et al., 2010). Peel and seed are the
major by-products of orange fruit processing, which are sometimes
⇑ Corresponding author. Tel.: +61 02 9351 5660; fax: +61 02 9351 2854.
E-mail address: dsho6535@uni.sydney.edu.au (D. Shofinita).
used as animal feed and often simply sent to waste. This agricul-
tural waste may cause major environmental problems due to high
associated chemical and biological oxygen demand (COD and BOD)
(Lin et al., 2013). There is also a potential problem with citrus
extract in waste water treatment, which can kill composting
microorganisms. Therefore, the utilization of orange peel to pro-
duce valuable products is expected to lower its environmental
impacts. Bocco et al. (1998) have reported that orange peel extract
contains phenolic compounds, such as phenolic acids and flavo-
noids, so the extract has significant antioxidant activity.
Drying is a method that can be used to reduce the moisture con-
tent of food products and prolong their shelf-life. Compared with
liquid extracts, the production of dried extracts could result in
lower storage and transportation costs, and also better storage sta-
bility (Fang and Bhandari, 2011). Spray drying is a commonly-used
method to convert feed from a liquid state into a powder form
(Masters, 1979). In the spray-drying process, the moisture content
of the feed is removed very quickly. This process usually occurs
with the temperature of the product being much lower than
100 °C, so it may be suitable for the drying of heat-sensitive mate-
rials, such as phenolic compounds in orange peel extracts.
Several studies have been conducted in the area of spray drying
for natural extracts. Ersus and Yurdagel (2007), for example, found
that higher air inlet temperatures resulted in higher anthocyanin
losses, due to degradation, during the spray drying of black carrot
extracts. Another study was carried out by Fang and Bhandari
(2011) on the spray drying of bayberry juices. They found good
retention of total phenolics and anthocyanins content after spray
drying, which were 96% and 94%, respectively. Kha et al. (2010)

http://dx.doi.org/10.1016/j.jfoodeng.2014.03.028
0260-8774/Ó 2014 Elsevier Ltd. All rights reserved.
32 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42
studied spray drying of Gac fruit aril extract and found a significant
decrease in the total carotenoid content as the spray-drying tem-
perature increased from 120 °C to 200 °C. This behavior may occur
due to thermal degradation and oxidation. However, the economic
evaluation of the production of phenolic powder, by considering its
properties, has not been explored yet.
The objective of this study was to evaluate the effect of different
drying conditions on the total phenolic content recovery and the
yield from spray drying the orange peel extract. Moreover, the
effect of extraction solvent to solid ratio on the total phenolic con-
tent of spray-dried powder was assessed. An evaluation of econom-
ics was also performed to analyse the feasibility of these processes.
2. Materials and methods
2.1. Chemicals
Gallic acid (G7384), sodium carbonate (S2127), and Folin Cio-
calteu’s reagent (FCR) (F9252) used in this study were obtained
from Sigma Aldrich.
2.2. Extraction of phenolics from orange peels
Navel oranges were obtained from the local supermarket (Coles
Broadway, Sydney). The sample fruits were washed with tap water
and peeled using gloves. The peels were then cut into small sizes
(1–3 mm) and stored in a refrigerator for the experiments.
The prepared orange peel samples, together with deionized
water, were added to the microwave vessel. The solvent to solid
ratios were adjusted to 2 and 14, with the total amount of solvent
and peels used for each solvent to solid ratio being kept constant
(15 g) to provide the same heat load per unit mass during extrac-
tion. The temperature of the system was increased to 135 °C for
7 min. The extraction was then carried out using a microwave sys-
tem (ETHOS SEL Microwave Solvent Extraction Labstation) at con-
stant temperature (135 °C) for 3 min. The extraction temperature
was chosen to be high enough to yield an extract with high antiox-
idant activity, but not too high to reduce the possibility of Maillard
reactions occurring during the extraction (Ahmad and Langrish,
2012). After the extraction was completed, the microwave vessel
was cooled down to the ambient temperature. Solid residue was
then separated from the liquid extract using filter paper, and the
liquid extract was stored in a cool place for the determination of
the total phenolics content.
2.3. Determination of total phenolics content (TPC)
The TPC in the extracts was assessed according to the method
described by Singleton et al. (1999). The sample was prepared by
mixing 0.1 mL of orange peel extract with 7.9 mL of deionized
water and 0.5 mL of FCR. The mixture was left in the dark for 5–
8 min, before 1.5 mL of 20% sodium carbonate solution was added
to the sample. The mixture was then left again in the dark for 2 h.
The absorbance value of the sample was then read using a UV/VIS
Spectrophotometer (Cary 50, Varian, USA) at a wavelength of
765 nm. Gallic acid was used to produce a standard calibration
curve. The TPC was expressed as mg of gallic acid equivalents per
g of dry matter (mg GAE/g DM).
2.4. Spray drying
The orange peel extract was dried using a Buchi B-290 mini
spray dryer with a high performance cyclone. The operating condi-
tions were: drying air inlet temperature varying from 100 to
200 °C, atomization air flow rate of 50 mm (1052 L/h), liquid feed
pump rate of 4 mL/min (10%), and aspirator rate of 38 m3/h
(100%). All the inlet and outlet air temperatures for each run are
shown in Table 1. The yield (or recovery) from spray drying has
been calculated using Eq. (1).
mass of solid in the collecting v essel
Yield ð%Þ ¼  100% ð1Þ
mass of solid in the feed
2.5. Statistics
Data in this study were obtained from two replicates for each
experiment and are presented as means ± standard deviation. For
statistical analysis, differences were tested for significance by
using the ANOVA method, using a significance level of P 6 0.05.
2.6. Evaluation of the economics
The evaluation of the economics for the process was carried out
using data from the extraction and spray-drying experiments, as
explained in the previous sections. The required experimental data
include the yield from spray drying, the total phenolic content and
moisture content of the powder, the optimum extraction time, the
mass fraction of solvent (water) remaining in the residue, and the
humidity of the ambient air. The manufacturing cost of orange peel
extract powder was calculated using the approach proposed by
Turton et al. (2009), as shown in Eq. (2). The cost of manufacturing
(COM) is a function of the fixed capital cost for investment (FCI),
the cost of operational labor (COL), the cost of utilities (CUT), the
cost of water treatment (CWT), and the cost of raw material (CRW).
COM ðUSD=yearÞ ¼ 0:280  FCI þ 2:37  COL þ 1:23
 ðCUT þ CWT þ CRMÞ ð2Þ
The FCI was calculated by predicting the purchased equipment
costs using the cost correlation method (Towler and Sinnott,
2013; Turton et al., 2009). The flowsheet used to calculate the FCI
in this study consists of three 100 L microwave-assisted extractors
with integrated mesh filters (Li et al., 2012), one solvent reservoir
tank, two centrifugal pumps, one compressor, one indiret heater,
and one spray dryer. The process flow diagram is shown in Fig. 1.
It was assumed that the industrial-scale unit would work with
the same performance as the lab-scale unit at the same solvent to
solid ratio, temperature, pressure, and extraction time (Rosa and
Meireles, 2005). The plant was assumed to run 24 h per day,
330 days per year, resulting in 7920 h per year. The efficiency of
the pumps, the fan, and the indirect heater were assumed to be
75%, 70%, and 85%, respectively (Baker and McKenzie, 2005;
Turton et al., 2009).
Turton et al. (2009) proposed a method to calculate the number
of operators required per shift (NOL), which is shown in Eq. (3):
Table 1
The inlet and outlet air temperatures from the spray drying experiments at
atomization air flow rate of 1052 L/h, liquid feed pump rate of 4 mL/min, and
aspirator rate of 38 m3/h.
Run Inlet temperature (°C) Outlet temperature (°C)
1 100 43
2 100 43
3 125 52
4 125 50
5 150 66
6 150 64
7 175 75
8 175 73
9 200 79
10 200 79
 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42
Inlet air
Fan
Natural gas
Indirect Heater
Orange peel
Water
Pump 1
Reservoir Tank
Fig. 1. Process flow diagram illustrating the production of antioxidant from orange peel.
0:5
NOL ¼ ð6:29 þ 31:7P2 þ 0:23Nnp Þ
where P is the number of processing steps involving the handling of
particulate solids, and Nnp is the number of nonparticulate process-
ing steps. An operator has been considered to work 49 weeks per
year, 5 shifts per week, and 8 h per shift.
3. Results and discussion
3.1. Effect of drying conditions on yield from spray drying
Figs. 2 and 3 show the effect of different inlet and outlet air
temperatures on the yield from spray drying. All of the resulting
yields, at outlet air temperatures ranging from 43 to 79 °C (corre-
sponding to the inlet temperatures of between 100 and 200 °C),
were above 75%. In comparison with the yield from industrial-scale
spray drying of milk, which should be more than 99% to be accept-
able, the yield in this study might appear to be low. However, the
yields of more than 60% in a small-scale spray dryer should be con-
sidered to be good due to the high deposition of wet particles on
the spray dryer walls (Bhandari et al., 1997; Hanus and Langrish,
2007; Fang and Bhandari, 2011).
Figs. 2 and 3 both show an increase and then decrease of the
yield from spray drying as the inlet and outlet temperatures
increase. ANOVA tests suggested that both temperatures affect
the yield from spray drying significantly. The outlet air tempera-
ture of the spray drying can be used to approach the particle tem-
perature during the process (Bhandari et al., 1997; Langrish and
Premarajah, 2013). At first, the yield of spray drying increased as
Fig. 2. The effect of different inlet air temperatures on the yield from the spray
drying of orange peel extract (error bars indicate ± standard deviation). The black
line represents a regression of polynomial order 2 with equation
y = 0.005x2 + 1.465x  12.77 and r2 of 0.86.
33
Flue gas
Outlet air
Orange peel extract
Pump 2 Spray Dryer powder
Orange peel residue
Microwave Extractor
ð3Þ
Fig. 3. The effect of different outlet air temperatures on the yield from the spray
drying of orange peel extract (error bars indicate ± standard deviation). The black
line represents a regression of polynomial order 2 with equation
y = 0.041x2 + 4.870x  50.58 and r2 of 0.72.
the outlet air temperature increased. This was caused by the lower
moisture content of the powder at higher outlet temperatures,
which reduced its stickiness. However, at outlet temperatures
above 50 °C, higher outlet temperatures resulted in a lower yield.
This peak yield trend in the spray drying of natural extract was also
reported by Bhandari et al. (1997). As explained by these authors,
lower moisture content also leads to an increase in the glass-tran-
sition temperature, and the yield of spray drying is actually a
function of the overall difference between the particle and glass-
transition temperatures. In the beginning, the glass-transition tem-
perature increases faster than the particle temperature. As the
overall difference between the particle and glass-transition tem-
perature decreases, the yield increases. Then, the particle temper-
ature starts to increase more rapidly than the glass-transition
temperature, resulting in an increasing overall difference between
the particle and glass-transition temperatures, and hence the yield
decreases.
3.2. Effect of drying conditions on total phenolic compounds of the
powder
The total phenolics content (TPC) of the orange peel extracts
(liquids) were between 20.0 and 23.3 mg/g DM. After spray drying,
the TPCs of the powders were between 18.3 and 21.2 mg/g DM. The
TPC recovery can be calculated using Eq. (4) and plotted as a func-
tion of different outlet air temperatures (Fig. 4).
TPC in powder ðmg=g DMÞ
TPC recov eryð%Þ ¼  100% ð4Þ
TPC in extract ðmg=g DMÞ
34 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42
Fig. 4. The effect of different outlet air temperatures on the TPC recovery (error bars
indicate ± standard deviation). The black line represents a linear regression with
equation y = 0.044x + 88.09 and r2 of 0.89.
Statistical analysis performed has showed that there was no signif-
icant effect of the outlet air temperatures on the TPC recovery with
95% confidence. The TPC recoveries ranged between 89.4% and
91.9% for inlet air temperatures between 100 °C and 200 °C and out-
let temperatures below 80 °C. As reported by Liazid et al. (2007),
most of the phenolic compounds do not degrade below temperature
of 100 °C. High recoveries of TPC have also been reported by Fang
and Bhandari (2011) and Akkaya et al. (2012) during spray drying
of bayberry juice and carob molasses, respectively. This results sug-
gests that spray drying is a suitable process for the drying of some
heat-sensitive materials, such as phenolic compounds.
3.3. Effect of extraction solvent to solid ratio on the TPC of the powder
Fig. 5 shows the effect of extraction solvent to solid ratio on the
TPC of the extracts and the spray-dried powders. In this experi-
ment, the total amounts of solvent and peels used for each solvent
to solid ratio were kept constant (15 g) to provide the same heat
load per unit mass during extraction. Based on the statistical anal-
ysis, the extraction solvent to solid ratio had a significant effect on
the TPC of extract and spray-dried powder. Increasing the solvent
to solid ratio resulted in a greater TPC of the extract, and a greater
TPC of the spray-dried powder.
A higher solvent to solid ratio provides a higher concentration
gradient between the phenolic concentrations inside and at the
surface of the orange peels, resulting in a higher extraction rate
and a higher extraction yield. However, large amounts of solvent
produce a more dilute extract, which increases the energy require-
ment for the spray-drying process. A tradeoff between the amount
of water used and the yield of phenolics can be done by doing an
evaluation of the economics.
Fig. 5. The effect of the extraction solvent to solid ratio on the TPC of the extracts
and the spray-dried powders at outlet air temperature of 50 °C (error bars
indicate ± standard deviation).
3.4. Evaluation of the economics
3.4.1. Overall results and uncertainty analysis of the economics
The evaluation of economics has been carried out using the pro-
posed methodology. From the preliminary experiment, it was
found that three minutes of extraction, with seven minutes of pre-
heat time, was sufficient to yield the maximum extractable pheno-
lic compounds from orange peels. This optimum extraction time
has also been reported in a study conducted by Ahmad and
Langrish (2012). Moreover, the filtration experiment result sug-
gested that 20% of the solvent remained in the wet orange peel res-
idue. It was also measured, in the spray drying experiment, that the
humidity of the ambient air and the moisture content of the pow-
der were 0.0126 kg moisture/kg dry air and 0.04 kg moisture/kg
powder, respectively. All these data, together with the results of
the yield and the total phenolic compounds of extracts and pow-
ders for outlet air temperature of 50 °C in the previous sections,
have been used for the evaluation of the economics. It has been cal-
culated from the uncertainty analysis that the experiment condi-
tions, such as the extraction time, the moisture content of the
powder, and the humidity of the air, resulted in relatively low
impact on the economic evaluation. The systematic uncertainties
for these variables were 0.05 and 0.39 for solvent to solid ratios
of 2 and 14, respectively. The highest impact of uncertainty on
the results was achieved by the extraction time, which has been
known as an important parameter regarding extraction
(Mezzomo et al., 2011; Rosa and Meireles, 2005; Vieira et al.,
2013).
In this section, the analysis of the economics for antioxidant
production from orange peels using the average values (Table 2)
is discussed. The average values have been calculated as the mean
of minimum and maximum values for each input variable. As men-
tioned previously, the COM consists of several costs: the fixed cap-
ital cost of investment (FCI), the cost of operational labor (COL), the
cost of utilities (CUT), the cost of water treatment (CWT), and the
cost of raw material (CRW). The CWT was assumed to be zero,
since there are several potential uses of the solid materials after
extraction, such as for animal feed or as a fiber source (Mezzomo
et al., 2011).
In order to determine the maximum and minimum values for
each input variable, several assumptions have been made:
3.4.1.1. Basic rate of labor. It has been reported by the United States
Bureau of Labor Statistics (2013) that Australia had high hourly
labor costs in manufacturing. Several countries in Europe (Norway,
Switzerland, and Denmark) have been reported to have hourly
labor costs in manufacturing that are higher than Australia. How-
ever, there is no defined minimum wages in these countries. In
Norway, for example, the wages are usually negotiated by the
labor, employers, and local governments. Thus, in this study, the
basic labor rate in Australia has been used as a reference for the
maximum value, which is 14.73 USD per hour (2013), as set
nation-wide by Fair Work Commission Australia (2013). The max-
imum value has been calculated by adding 5% to that value to
allow for the inflation rate, resulting in 15.47 USD per h.
The minimum basic rate of labor has been set based on the min-
imum wages in Bangladesh. Chowdhury (2006) reported that, for
medium size industries (with 11–49 workers), the minimum
wages in Bangladesh were 1250 taka per month, or about 0.10
USD per hour.
3.4.1.2. Cost of orange peel. The maximum price of orange peel used
in this study has been specified to be 2.17 USD per kg including a
5% inflation rate). This value has been set based on the market
price of whole Navel orange fruit in Australia (Coles Broadway,
Sydney, July 2013). On the other hand, the minimum price of
 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42 35

Table 2
The range of economic parameters used for evaluating the production of orange peel extract powder.

Cost Minimum Average 1.1  Average Maximum

Basic rate of operational labor/OL (USD/h) 0.10 7.79 8.56 15.47
Cost of orange peel/OP (USD/kg) 0 1.09 1.20 2.17
Cost of water/W (USD/kL) 1.00 3.29 3.62 5.59
Cost of electricity/E (USD/kW h) 0.05 0.20 0.21 0.34
Cost of natural gas/N (USD/MJ) 0.50 6.00 6.59 11.49
Selling price/SP (USD/kg) (1) For solvent to solid ratio of 2: 8.11 (1) 25.06 (1) 27.56 42.00
(2) For solvent to solid ratio of 14: 17.49 (2) 29.75 (2) 32.72

orange peel has been set to zero due to the possibility of utilizing
orange peel as an agroindustrial residue (Mezzomo et al., 2011).
3.4.1.3. Cost of water. The maximum price of water used in the
uncertainty analysis has been specified based on the study con-
ducted by Leal et al. (2008). In their evaluation of economics for
the production of natural antioxidants in Brazil, the authors
defined the cost of water as 5.32 USD per kL (2008). The maximum
value has been calculated by adding 5% to that value to allow for
the inflation rate, resulting in 5.59 USD per kL. Turton et al.
(2009) suggested that the price of deionized water was 1.00 USD
per kL (2009), which has been used as the minimum water price
here. In addition, Hoang et al. (2009) reported that the price of
desalinated water in Australia was mostly between 1.25 and
2.00 USD/kL (2009), which is between the determined maximum
and minimum values.
3.4.1.4. Electricity price. The maximum electricity price has been
taken as the electricity price for industry in Australia, which is
0.324 USD per kW h (Origin Energy Australia, 2013). By consider-
ing an additional 5% for inflation, the maximum electricity price
used in this study was 0.34 USD per kW h. This electricity price
was higher than the energy prices for industrial consumers in
European countries. Cyprus has the highest price for electricity
industry in Europe, which was 0.19 € per kW h, or about
0.26 USD per kW h (Eurostate, 2013).
The value of the minimum electricity price has been taken from
the price in Korea. The International Energy Agency (2012)
reported that the Republic of Korea has a lower electricity price
compared with other countries, such as European countries, US,
New Zealand, Japan, and Turkey. The electricity price in Korea
was 57.90 won per kW h (Korea Electric Power Corporation, data
for spring and fall 2013), or about 0.05 USD/kW h.
3.4.1.5. Cost of natural gas. It has been reported that South Korea
has a very high price for natural gas compared with other countries
(Reed, 2013). Thus, the natural gas price in South Korea, which was
11.49 USD/GJ (including a 5% inflation rate), has been used as the
maximum value for the price of natural gas in this study. On the
other hand, Reed (2013) showed that Iran has the lowest natural
gas price (0.5 USD/GJ), and so this value has been used as the min-
imum price of natural gas here. Meanwhile, the price of natural gas
in Australia (4.6 USD/GJ) was in between the prices in those coun-
tries (Reed, 2013).
3.4.1.6. Selling price. The antioxidant power of the orange peel
extract powder in this study has been reported as mg GAE/g DM.
Its selling price has been determined based on the market price
of gallic acid, which was 427.6 USD/kg (Sigma Aldrich, September
2013). The minimum selling price of the product in this study
has been initially calculated based on the gallic acid price and
the specific antioxidant power of the powder (detail calculation
could be found in Appendix A). The minimum selling prices have
been determined to be 8.11 USD/kg for a solvent to solid ratio of
2 and 17.49 USD/kg for a solvent to solid ratio of 14. On the other
hand, the maximum selling price (42 USD/kg) has been specified
based on the highest market price of orange peel extract powder
produced in China.
Table 3 shows the results of the economic evaluations for the
average values of the profits at solvent to solid ratios of 2 and
14. However, there is uncertainty in the results, and this uncer-
tainty can be quantified. Uncertainty analysis is usually conducted
to consider the impact of uncertainty from several input variables
on the results of any study. Regarding an economic evaluation, this
analysis can give an insight about the levels of possible risk and
profit as a result of changes in several economic parameters, and
give guidance for the management of risk. Uncertainty analysis is
also considered to be one way to increase the generalisability of
an economic evaluation (Briggs et al., 1994). The economic param-
eters used to evaluate the economic viability may vary from loca-
tion to location. One of the benefits of doing uncertainty analysis is
to assess whether or not the results of a study conducted in one
location are also valid in another location.
In this study, the output variable that has been evaluated is the
profit (P) from the production of orange peel extract powders
(without considering taxes, storage, and transportation costs),
while the basic rate of operational labor (OL), the cost of orange
peel (OP), the cost of water (W), the cost of electricity (E), the cost
of natural gas (NG), and the selling price (SP) have been varied. The
range of variation for each variable used in this work is shown in
Table 2 as the minimum and maximum values.
The change of profit (P) as result of a change in the value of one
input variable can be determined by using the partial derivative of
P for that variable. For example, Eq. (5) has been used to determine
the amount of profit (P) changes due to a change in operational
labor (OL) (symbolized by DPOL). In that equation, oP/oOL is the
partial derivative of P relative to the OL, and DOL is the variation
in the variable OL (difference between the maximum and mini-
mum possible values of OL).
@P
DPOL ¼  DOL ð5Þ
@OL
The partial derivatives of P for each input variable (DOL, DOP, DW,
DE, DNG, and DSP) have been evaluated using 10% variations
around the average values, as shown in Table 2. The values of partial
derivatives for all input variables are shown in Table 4. It can be
seen that there is considerable variation in the partial derivatives
at different solvent to solid ratios. In order to determine the contri-
bution of each input variable to the total uncertainty, the values of
Table 3
The results of the economic evaluation for the average values.
Solvent to solid ratio (g Specific COM (USD/ Selling price Profita
solvent/g peel) kg powder) (USD/kg) (USD/kg)
2 1.26 25.06 23.79
14 6.36 29.75 23.38
a
Profit without considering taxes, storage, and transportation costs.
36 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42

Table 4 orange peels in one location, without the complications of sourcing

Values of the partial derivatives with respect to each input variable. small amount of orange peels from several locations. Table 5 also
Partial derivatives Solvent to solid ratio 2 Solvent to solid ratio 14 shows that the electricity, natural gas, and water costs are the
oP/oOL 0.02 0.19 three least important variables in the uncertainty analysis. This
oP/oOP 0.55 0.84 result suggests that variations in these prices have much less
oP/oW 0.00 0.01 impact on the overall profit and so deserve less management atten-
oP/oE 0.28 2.57 tion than the selling price, labor, and orange peel costs.
oP/oNG 0.01 0.07
oP/oSP 1.00 1.00
3.4.2. Economic evaluation results for the most likely case
This section will discuss the results of economic analysis for the
most likely case, which has been evaluated using the current eco-
systematic uncertainty and random uncertainty need to be calcu-
nomic parameters in Australia. These economic parameters are
lated. The equations used to determine the value of both uncertain-
shown in Table 6.
ties are provided in Eqs. (6) and (7). The values of both uncertainties
The effect of solvent to solid ratio on the specific COM (specific
are provided in Table 5.
COM (USD/kg) = COM (USD/year)/production rate (kg/year)) is
dP systematic ¼ DPOL þ DPOP þ DPW þ DPE þ DPNG þ DPSP ð6Þ shown in Table 7. Increasing the solvent to solid ratio caused an
increase in the specific COM. From the calculation, it was found that
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi increasing the solvent to solid ratio resulted in a decrease in the
u" #
u ð DP 2 þ DP 2 þ DP 2 þ DP 2 þ DP 2 þ DP 2 Þ COM (USD/year). However, in this case, the extract would be more
dP random ¼ t OL OP W E NG SP
ð7Þ
6 diluted, so the productivity would decrease, resulting in a higher
specific COM. In addition, the selling price of the peel extract pow-
Systematic uncertainty represents the uncertainty where all the der at higher solvent to solid ratios was considered to be higher,
variations happen at the same time, while random uncertainty since the powder contains more phenolic compounds, as shown
shows the uncertainty where the parameters vary randomly. in Fig. 5. If the powder profitability (profit = selling price  specific
Systematic uncertainty has been considered to be the upper bound COM; without considering taxes, storage, and transportation costs)
for the overall uncertainty. It may appear that the values of system- is evaluated, the profit of the process with an extraction solvent to
atic uncertainties are higher than the average case profits for both solid ratio of 14 (8.8 USD/kg) is higher than that for a solvent to
solvent to solid ratios. The high values of uncertainty suggest that solid ratio of 2 (6.1 USD/kg), if the selling price, the basic rate of
the production of orange peel extract may not be viable, unless operational labor, and the cost of orange peel are the same.
the overall uncertainty is reduced because of the possibility of zero From the specific COM and the selling price of the orange peel
or negative profits. The priority in the reduction of the uncertainties extract powder, the production process (microwave-assisted
in the input variables can be based on the relative contributions of extraction and spray drying) appears to be a feasible process.
the individual uncertainties to the total uncertainty. Pereira et al. (2013) summarized the economic evaluation for the
The results in Table 5 show that the selling price has the largest production of extracts from several natural sources using super-
impact on the profits from producing orange peel extract powders critical technology. The specific COMs were between 5.22 and
for both solvent to solid ratios, thus its uncertainty should be 48,000 US$/kg of crude extract, for peach almond extract and jato-
reduced first by, for example, a negotiation with the buyers. A sim-
ilar further reduction of the overall uncertainty can also be done Table 6
for other costs. Overall, at both solvent to solid ratios, the labor cost Economic parameters used for the most likely case.
and the orange peel cost are the second or third most important Cost Value
parameters. This issue suggests that the location of manufacturing,
Basic rate of operational labor 14.73a
and the associated labor cost, needs careful consideration. The (USD/h)
same review should also be done for the cost of orange peel by, Cost of orange peel (USD/kg) 2.07b
for example, integration with an orange juice factory so the orange Cost of water (USD/kL) 1.65c
peel is considered to be unutilized potential feedstock for further Cost of electricity (USD/kW h) 0.324d
Cost of natural gas (USD/MJ) 4.6e
processing, and its cost can be minimized. This co-location would
Selling price (USD/kg) (1) For solvent to solid ratio 2: 8.11f
also ensure that there would be a sufficiently large amount of (2) For solvent to solid ratio 14: 17.49f
a
Fair Work Commision Australia (2013).
b
Table 5 Market price of whole Navel orange in Australia (Coles Broadway Sydney, July
Results of the uncertainty analysis. 2013).
c
Boxall et al. (2012).
Uncertainty and Solvent to solid ratio 2 Solvent to solid ratio 14 d
Origin Energy Australia (2013).
average case e
Systematic Random Systematic Random Reed (2013).
profit f
uncertaintya uncertainty uncertaintya uncertainty Based on gallic acid selling price of 427 USD/kg (Sigma Aldrich, September
2013).
DPOL 0.38 (3) 0.15 2.94 (2) 8.62
DPOP 1.19 (2) 1.41 1.83 (3) 3.34
DP W 0.01 (6) 0.00 0.06 (6) 0.00
DP E 0.08 (4) 0.01 0.74 (4) 0.55 Table 7
DPNG 0.06 (5) 0.00 0.73 (5) 0.53 Specific cost of manufacturing (COM) and COM composition of fixed capital cost of
DPSP 33.89 (1) 1148.53 24.51 (1) 600.74 investment (FCI), cost of operational labor (COL), cost of raw material (CRW), and cost
Value of overall ±35.61 ±13.84 ±30.77 ±10.11 of utilities (CUT) for the most likely case.
uncertainty
Solvent to solid Specific Selling COM composition (%)
Average case 23.79 23.38
ratio (g solvent/ COM (USD/ price
profit (USD/kg FCI COL CRM CUT
g peel) kg powder) (USD/kg)
powder)
a
2 2.0 8.1 19.27 18.29 56.67 5.78
Numbers in the brackets represent the ranking of each variable’s contribution
14 8.7 17.5 34.60 32.22 20.17 13.00
to the total uncertainty.
 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42 37

ba bark extract, respectively (Mezzomo et al., 2011; Veggi et al., A.1. Fixed Capital of Investment (FCI)
2011). The values of specific COM here (2.0–8.7 USD/kg) may
appear to be low, since this study evaluated the specific COM for A.1.1. 3  100 L microwave extractor (batch operation)
extract powder, which is more complicated than the specific
COM for crude liquid extracts but also adds value due to the longer
storage life of a powder product (the extra COM reflects the added Design capacity 100 L
value). This difference may be due to the application of supercrit- Operating Conditions Temperature 135 °C
ical technology in the previous works. The supercritical extraction Solvent to solid ratio 2 L/kg
usually requires long extraction times, for instance 30–360 min Volume solvent per cycle 50 L
and 40–77 min for peach almond extract and jatoba bark extract, Mass orange peel per cycle 25 kg
respectively (Mezzomo et al., 2011; Veggi et al., 2011). On the Extraction time 3 min
other hand, this study evaluated the use of microwave-assisted Data from experiments 20% Solvent remaining in the wet
extraction with an extraction time of 3 min. The longer extraction orange peel residue
time results in a lower productivity of the extract for the same
capacity of equipment, increasing the specific COM.
Overall, the COL and CRM make a large contribution to COM.
This trend has also been reported by Vieira et al. (2013) in their
80
Solvent volume in extract ¼  50 L=batch ¼ 40 L=batch:
study of ultrasound-assisted and agitated bed extraction from jus- 100
sara pulp. The value of COL was constant since it was only affected Calculation of cost
by the number of processing steps, as described in Eq. (3), yet its Cost of microwave extractor 1.5 L 2004: USD 60,000 (ETHOS SEL
importance in the COM increased at a higher solvent to solid ratio Microwave Solvent Extraction Labstation)
due to the lower productivity. Table 7 also shows that, at a lower
solvent to solid ratio, the CRM (cost of raw material) has a more CEPCI 2004 = 444.2.
significant influence on the COM. For a constant total amount of CEPCI 2011 = 585.7.
solvent and orange peels, a decrease in the solvent to solid ratio
caused an increase in the orange peel mass and a decrease in the
Cost of microwave extractor 100 L 2011
solvent volume fed to the extractor. However, the cost of orange
 0:6
peel (2.30 USD/kg) is much higher than the cost of water 100 L 585:7
¼ USD60; 000   ¼ USD983; 090
(0.00165 USD/kg), hence the effect of the increasing cost for the 1:5 L 444:2
orange peel is more significant than the decreasing cost of water
at lower orange peel masses. Cost of 3  100 L microwave extractors = 3  USD 983,090 = USD
2,949,269.
4. Conclusions Time allocation per cycle

The spray drying experiments to produce orange peel extract at - Loading: 5 min.
outlet temperatures between 43 and 79 °C resulted in yields of - Preheat Time: 7 min.
between 75% and 92%. However, the high TPC recoveries of - Extraction Time: 3 min.
spray-dried powder (89–92%) showed that the spray-drying tem- - Cooling Time: 10 min.
perature did not appear to significantly affect the TPC of the pow- - Unloading Time: 5 min.
der. This result suggested that spray drying is a suitable process for Total per cycle: 30 min.
drying of heat-sensitive materials, such as phenolic compounds.
Moreover, the extraction solvent to solid ratio was found to signif- A.1.2. Spray dryer
icantly affect the TPC of both extract and powder. Operating Conditions: Inlet temperature 125 °C.
The economic evaluation for using microwave-assisted extrac- Outlet temperature 50 °C.
tion and spray drying to produce antioxidant powder from orange Data from experiments:
peel has also been studied. The uncertainty analysis suggested that
the powder selling price has the largest impact on the profits, fol- - Spray drying yield: 90%.
lowed by the labor cost and the orange peel cost. On the other - Total solid concentration in the feed (Xi): 0.6 kg solid/kg extract.
hand, the electricity, natural gas, and water costs are the three least - Moisture content of the antioxidant powder: 0.04 g/g powder.
important variables. Using the current economic parameters in - Humidity of air at ambient temperature: 0.0126 kg/kg air.
Australia, this process appears to be economically feasible with
profits of 6.1 USD/kg and 8.8 USD/kg for solvent to solid ratios of Feed mass per batch ¼ Mass of solvent þ Mass of solid
2 and 14, respectively. Xi
¼ Mass of solvent þ
ð1  X i Þ
Acknowledgements  Mass of solvent
0:6
The authors would like to acknowledge the Commonwealth ¼ 40 L=batch  1 kg=L þ
ð1  0:6Þ
Department of Innovation, Industry, Science and Research (DIISR)
Australia for the scholarship funding through International Post-  40 L=batch  1 kg=L
graduate Research Scholarship (IPRS). ¼ 100 kg=batch

Appendix A. Example of calculation 100 kg=batch kg

Feed pump flowrate ¼  3 extractor ¼ 10
30 min=batch min
Example of calculation for solvent to solid ratio of 2 (most likely ¼ 600 kg=h
case)
38 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42

Evaporation rate Purchased tank cost in 2011 ¼ Purchased cost in 2010

CEPCI2011
¼ feed flow rateð1  total solid contentÞ  feed flow rate 
CEPCI2007
moisture content
total solid content  1moisture content
: 585:7
  kg moisture ¼ 24; 300 USD 
0:04 550:8
¼ 600 kg 1  0:6 kgkgextract
solid
 600 kg  0:6 kgkgextract
solid
 kg wet powder
h h 10:04 kgkgwet
moisture
powder ¼ 25; 840 USD
¼ 225 kg=h
A.1.5. Centrifugal pump for solvent extraction
Yield
¼ Yield fraction  ðfeed flow rate  evaporation rateÞ Water needs per cycle  no: of extractor
  Feed flow rate ¼
¼ 0:9  600 kg  225 kg time allocated per cycle
h h
3 extractor  50 L=extractor=cycle
¼ 337:5 kg=h ¼ 2673 ton=year ¼ ¼ 5 L=min
30 min=cycle
Calculation of cost ¼ 0:083 L=s
Data from Chemical Engineering Design (Towler and Sinnott,
2013): Calculation of cost
Data from Chemical Engineering Design (Towler and Sinnott,
Purchased spray drying cost Jan2010 2013):
 0:7
kg Purchased pump cost Jan 2010
¼ 410; 000 þ 2; 200 evaporation rate
h ¼ 8000 þ 240 ðflowrate LsÞ
0:9
¼ 8025
0:7
kg Purchased spray drying cost in 2011
¼ 410; 000 þ 2; 200 ð337:5 Þ ¼ 507; 488 USD
h ¼ Purchased cost in2010  CEPCI 2011
CEPCI2010
¼ 8025 USD  585:7
550:8
¼ 8534 USD
Purchased spray drying cost in 2011
CEPCI 2011
¼ Purchased cost in 2010  A.1.6. Centrifugal pump for spray dryer feed
CEPCI 2010
Density feed = 1018 kg/m3
585:7
¼ 507; 488 USD  ¼ 539; 643 USD
550:8 600 kg=h L 1h
Feed flow rate ¼  1000 3  ¼ 0:164 L=s
1018 kg=m3 m 3600 s
A.1.3. Fan Calculation of cost
Air flowrate = 1.815 m3/s (detail calculation in the cost of utility Data from Chemical Engineering Design (Towler and Sinnott,
section). 2013):
Based on the cost-capacity equation from Turton et al. (2009),
cost of fan in 2001 = 3003 USD. Purchased pump cost Jan 2010
0:9
¼ 8000 þ 240ðflowrate LsÞ ¼ 8047
Purchased blower cost in 2011 ¼ Purchased cost in 2001 Purchased spray drying cost in 2011
CEPCI2011 ¼ Purchased cost in 2010  CEPCI 2011
 CEPCI 2010
CEPCI2001 ¼ 8047 USD  585:7
¼ 8557 USD
550:8
585:7
¼ 3; 003 USD 
397
¼ 4; 431 USD A.1.7. Indirect heater
Assumption: Heater efficiency = 85%.
Energy absorbed by air for spray dryer = 349.7 kW (detail calcu-
lation in the cost of utility section).
A.1.4. Water reservoir tank
Design requirement for 6 days. 349:7 kW
Energy supplied by natural gas ¼ ¼ 411 kW
No. of extractors: 3 0:85
¼ 0:411 MW
Water needs per cycle  no: of extractor
Volume of tank ¼ Calculation of cost
time allocated per cycle
Data from Chemical Engineering Design (Towler and Sinnott,
 6 days 2013):
3 extractor  50 L=extractor=cycle
¼  60 min Purchased cylindrical furnace cost Jan 2010
30 min=cycle
0:8
 24 h  6 days ¼ 80; 000 þ 109; 000 ðduty in MWÞ ¼ 133; 563 USD
3
¼ 43; 200 L ¼ 43:2 m Purchased spray drying cost in 2011
CEPCI 2011
Calculation of cost ¼ Purchased cost in 2010 
Estimated cost in 2010 for cone roof tank (Towler and Sinnott, CEPCI 2010
2013): USD 24,300 585:7
¼ 133; 563 USD  ¼ 142; 026 USD
550:8
 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42 39

A.2. Cost of operational labor (COL) energy requirement to increase the solvent temperature from room
temperature, 25 °C, to saturated liquid at 100 °C. QMAE,2 is the
The number of operators required to run the process unit per energy requirement to change the solvent phase from saturated
shift (NOL) can be determined using Eq. (3) (as shown in the mate- liquid at 100 °C to vapor–liquid equilibrium at 135 °C. QMAE,3 is
rial and methods section). In this plant, the value of P is 0, while the the energy requirement to increase citrus peels temperature from
value of Nnp is 8, resulting in the NOL value of 2.85. 25 °C to 135 °C.
A single operator works on the average 49 weeks a year QMAE,1
(3 weeks time off for vacation and sick leave), five 8-h shifts a
Q MAE;1 ¼ mwater  Cpwater  DT
week. This amounts to (49 weeks/year  5 shifts/week) 245 shifts
per operator per year. A chemical plant normally operates 24 h/ kJ
¼ 50 kg  4:187  ð100  25Þ  C ¼ 15; 701 kJ
day. This requires (355 days/year  3 shifts/day) 1095 operating kg C
shifts per year. The number of operators needed to provide this
QMAE,2
number of shifts is [(1095 shifts/year)/(245 shifts/operator/year)]
QMAE,2 can be calculated by determining the vapor–liquid equi-
or approximately 4.5 operators. Four and one-half operators are
librium condition at 135 °C. The mass of vaporized water is deter-
hired for each operator needed in the plant at any time.
mined using the vapor pressure at 135 °C (3.1 atm) and vessel
The number of operating labor required = 4.5  2.77 = 13.5  14
volume of 100 L, resulting in 83.3 g of water in the vapor state.
operators.
The enthalpy of saturated liquid and vapor from Smith et al., 2005:
From Fair Work Commision Australia, the basic rate of opera-
tional labor is 14.73 USD/h. This corresponds to 29,466 USD/year.
- At 100 °C Hsaturated liquid = 419.1 kJ/kg.
The cost of operational labor can then be calculated:
- At 135 °C (from interpolation) Hsaturated liquid = 567.7 kJ/kg.
USD
COL ¼ 29; 466  14 operators ¼ 412; 524 USD=year
year Hsaturated vapor ¼ 2726:6 kJ=kg

A.3. Cost of raw material (CRM) Q MAE;2 ¼ mwater  ðHequilibrium at135C  Hsaturated liquid at100C Þ
 
83 g kJ 49; 917 g kJ kJ
¼ 50 kg   2726:6 þ  567:7  419:1
A.3.1. Orange peel 50; 000 g kg 50; 000 g kg kg
¼ 7610 kJ
No. of extractor: 3.
QMAE,3
Orange peel requirement per batch: 25 kg.
Time per cycle: 30 min. Q MAE;3 ¼ mcitruspeel  Cpcitruspeel  DT
kJ
¼ 25 kg  3:77  ð135  25Þ  C ¼ 10; 368 kJ
Total orange peel requirement kg C
3 extractor  25 kg=extractor=batch QMAE,i
¼ ¼ 2:50 kg=min
30min=batch
Q MAE;i ¼ Q MAE;1 þ Q MAE;2 þ Q MAE;3 ¼ 33; 679 kJ
¼ 1; 188; 000 kg=year:
By assuming that the energy requirement to maintain constant
Price of orange peel = 2.07 USD/kg (assumed same as the price temperature at 135 °C for 3 min is 10% of the energy requirement
of whole navel orange). to increase temperature, then Q MAE ¼ 1:1  Q MAE;i ¼ 37; 046 kJ. Since
Cost of orange peel = 2.07 USD/kg  1,188,000 kg/ it takes 30 min for each batch, so Q MAE ¼ 74; 093 kJ=h
year = 2,459,160 USD/year.
A.4.2. Spray dryer
A.3.2. Water Operating conditions: Inlet temperature 125 °C.
Outlet temperature 50 °C.
No. of extractor: 3. Data from experiments:
Water requirement per cycle: 50 L.
Time per cycle: 30 min. - Spray drying yield: 90%.
- Total solid concentration in the feed (Xi): 0.6 kg solid/kg extract.
3 extractor  50 L=extractor=cycle - Moisture content of the antioxidant powder: 4%.
Total water requirement ¼ - Humidity of air at ambient temperature: 0.0126 kg/kg air.
30 min=cycle
¼ 5 L=min ¼ 2376 kL=year:
Fuel consumption
In order to calculate the energy consumption on spray drying,
Price of water = 1.65 USD/kL.
dew point temperature (Tdp) and outlet air humidity (Yo) should
Cost of water = 1.65 USD/kL  2376 kL/year = 3913 USD/year.
be determined by the following steps (Baker and McKenzie,
2005):
A.4. Cost of utilities (CUT)
(a) Guess the value of DTo (the difference between the outlet air
A.4.1. Microwave extractor (batch operation)
temperature and its dew point)
The energy requirement in microwave extractor can be calcu-
lated using the following formula:
DT o ¼ 11:9  C:
Q MAE;i ¼ Q MAE;1 þ Q MAE;2 þ Q MAE;3

where QMAE,i is the total energy requirement to increase the solvent (b) Calculate Tdp = To  DTo, where Tdp is the dew point temper-
and citrus peel temperature from 25 °C to 135 °C. QMAE,1 is the ature and To is the temperature of the outlet air
40 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42

T dp ¼ T o  DT o ¼ 50  C  11:9  C ¼ 38:1  C: of single-stage spray dryers found by Baker and McKenzie, 2005).
The formula for calculating Es is shown in Eq. (A.6).
To
! Yo
!
(c) Calculate Pdp using Eq. (A.1), where Pdp is the vapor pressure 1g
 Ta 1g
 Ya
Es ¼ Cpg þk ðA:6Þ
of water at Tdp (dew point temperature) Yo  Ya Yo  Ya
 
3816:44  50 
Pdp ðkPaÞ ¼ 0:13132 exp 18:3036  ðA:1Þ kJ 10:3
 25
T dp ðKÞ  46:13 Es ¼ 1:006
kg C 0:0423  0:0126


   0:0423 
3816:44 kJ 10:3
 0:0126
Pdp ¼ 0:13132 exp 18:3036  ¼ 6:58 kPa: þ 2501
ð38:1 þ 273Þ  46:13 kg 0:0423  0:0126
¼ 5596 kJ=kg evaporated water

(d) Calculate Yo using Eq. (A.2), where Yo is the humidity of out- In the previous calculation, it has been determined that the evapo-
let air and Pa is the atmospheric pressure. ration rate (Ev) is 225 kg/h, so:Es = 5596 kJ/kg evaporated
18:02 water  225 kg/h = 1,259,004 kJ/h = 350 kW.
29:97
 Pa Electric power consumption
Yo ¼ ðA:2Þ The dry air flowrate (G) can be determined using the water bal-
Pa  Pdp
ance over the dryer:
18:02
29:97
 100 kPa Ev 225 kg=h
Yo ¼ ¼ 0:0423 kg moisture=kg dry air: G¼ ¼ ¼ 7558 kg=h
100 kPa  6:58 kPa ðY o  Y a Þ ð0:0423  0:0126Þ
The air flow rate (Ga) can be determined by considering the humid-
(e) Calculate Ho using Eq. (A.3), where Ho is the enthalpy of out- ity of ambient air
let air, Cpg is the specific heat of dry air (1.006 kJ/kg °C), Cpv
is the specific heat of water vapor (1.84 kJ/kg °C), and k is the Gm ¼ ð1 þ Y i ÞG ¼ ð1 þ 0:0126Þ  7558 kg=h ¼ 7654 kg=h
latent heat of vaporization at 0 °C (2501 kJ/kg). Baker and McKenzie (2005) described a simplified equation to cal-
culate the electric power consumption for the fan (Ef), as shown in
Ho ¼ Cpg  T o þ Y o ðk þ Cpv  T o Þ ðA:3Þ
Eq. (A.7). Gm represents the mass flow rate of the air including its
moisture (ton/h), DP is the outlet pressure in mbar, qa is the density
kJ of the humid air (kg/m3), and ef is the fan efficiency. In this study,
Ho ¼ 1:006  47  C
kg  C fan efficiency is assumed to be 70% and the average pressure drop
 
kg moisture kJ kJ  across the drying system is 57 mbar, as found by Baker and
þ 0:0436 2501 þ 1:84  47 C
kg dry air kg kg  C McKenzie (2005).
kJ Gm DP
¼ 160:16 Ef ¼ ðt=hÞ ðA:7Þ
kg 35:5qa ef
Using Eq. (A.7), it is found that the Ef in this case equals to 14.99 kW
(f) Calculate Es,a using Eq. (A.4), where Es,a is the specific energy or 53,965 kJ/h. This results in a heat-to electricity consumption ratio
consumption of an adiabatic dryer, Ta is the ambient temper- of 23.
ature of air, and Ya is the humidity of ambient air.
  A.4.3. Pump
To  Ta
Es;a ¼ Cpg þk ðA:4Þ Boyce (1999) described an equation to calculate the work of
Yo  Ya
pump, as shown in Eq. (A.8). In this equation, H is the total dynamic
  head in Pa, Q is the capacity in m3/h, and g is the pump efficiency.
kJ 50  25 kJ
Es;a ¼ 1:006 þ 2501 HQ
kg  C 0:0423  0:0126 kg Work input ðkWÞ ¼ ðA:8Þ
¼ 3346 kJ=kg evaporated water gx3:599x106
Centrifugal pump for solvent extraction
(g) Calculate Ti using Eq. (A.5), where Ti is the inlet air 3 extractorx50 L=extractor=cycle
Feed flowrate ¼¼ ¼ 5 L= min
temperature. 30 minutes=cycle
ðCpg þ Y o Cpv ÞT o þ kðY o  Y i Þ ¼ 0:3 m3 =h
Ti ¼ ðA:5Þ
ðCpg þ Y i Cpv Þ
100; 000 Pa  0:3 m3 =h
Work ¼ ¼ 0:011 kW ¼ 40:01 kJ=h
  0:75  3:599  106
1:006 kgkJ C þ 0:0423  1:84 kgkJ C 50  C þ 2501 kg
kJ
ð0:0423  0:0126Þ
Ti ¼   Centrifugal pump for spray dryer feed
kJ kJ
1:006 kg  C þ 0:0126  1:84 kg  C
600 kg
h m3
¼ 125  C Feed flowrate ¼¼ ¼ 0:589
1018 kg=m3 h
The calculated value of Ti is the same with the operating inlet air
100; 000 Pa  0:589 m3 =h
temperature, so the guess of DTo is right. Work ¼ ¼ 0:022 kW ¼ 78:6 kJ=h
The specific energy consumption of non-adiabatic dryer (Es) can 0:75  3:599  106
then be calculated using the value of Yo from the iteration and by Total cost utility
assuming a thermal heat loss factor (g) of 0.3 (the average g value Electricity consumption
 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42 41

Basic cost of electricity = 0.324 USD/kWh (Origin energy, Akkaya, Z., Schröder, J., Tavman, S., Kumcuoglu, S., Schuchmann, H.P., Gaukel, V.,
2012. Effects of spray drying on physical properties, total phenolic content and
Australia)
antioxidant activity of carob molasses. Int. J. Food Eng. 8 (4), 20.
Baker, C.G.J., McKenzie, K.A., 2005. Energy consumption of industrial spray dryers.
Electricity consumption ¼ 3  Q MAE þ Ef þ Work pump 1 Drying Technol. 23 (1–2), 365–386. http://dx.doi.org/10.1081/drt-200047665.
Bhandari, B.R., Datta, N., Crooks, R., Howes, T., Rigby, S., 1997. A semi-empirical
þ Work pump 2
approach to optimise the quantity of drying aids required to spray dry sugar-
¼ 3  74; 093 kJ=h þ 53; 965 kJ=h rich foods. Drying Technol. 15 (10), 2509–2525. http://dx.doi.org/10.1080/
07373939708917373.
þ 40:01 kJ=h þ 78:6 kJ=h Bocco, A., Cuvelier, M.-E., Richard, H., Berset, C., 1998. Antioxidant activity and
phenolic composition of citrus peel and seed extracts. J. Agri. Food Chem. 46 (6),
¼ 276; 362 kJ=h ¼ 76:77 kW 2123–2129. http://dx.doi.org/10.1021/jf9709562.
Boxall, P., Cox, M.J., Krieger, M.S., 2012. Review of Prices for Sydney Water
Electricity consumption per year = 76.77 kW  330 days  24 h = Corporation’s Water, Sewerage, Stormwater Drainage and Other Services. New
607,997 kWh/year. South Wales: Independent Pricing and Regulatory Tribunal of New South Wales.
Boyce, M.P., 1999. Transport and Storage of Fluids. In: Perry’s Chemical Engineers’
Cost of electricity = 0.324 USD/kW h  607,997 kW h/year = Handbook, 7th edition (eds: R. H. Perry and D. W. Green). McGraw-Hill, New
196,991 USD/year. York.
Fuel consumption Briggs, A., Sculpher, M., Buxton, M., 1994. Uncertainty in the economic evaluation of
health care technologies: the role of sensitivity analysis. Health Econ. 3 (2), 95–
104. http://dx.doi.org/10.1002/hec.4730030206.
Natural gas price = 4.608 USD/GJ (Reed, 2013). Carr, A., Frei, B., 1999. Does vitamin C act as a pro-oxidant under physiological
Fuel consumption for heater = 1.5 GJ/h = 11,268 GJ/year. conditions? FASEB J. 13 (9), 1007–1024.
Chowdhury, M.S., 2006. Minimum wages in Bangladesh – issues and challenges.
Cost of natural gas = 4.608 USD/GJ x 11,731 GJ/year = 54,056 - Chittagong Univ. J. Law 11 (1), 1–32.
USD/year. Ersus, S., Yurdagel, U., 2007. Microencapsulation of anthocyanin pigments of black
carrot (Daucus carota L.) by spray drier. J. Food Eng. 80 (3), 805–812. http://
dx.doi.org/10.1016/j.jfoodeng.2006.07.009.
Total cost of utility Eurostate, 2013. Electricity and Natural Gas Price Statistics.
Fair Work Commission Australia, 2013. National Minimum Wage Order 2013.
Total Cost of Utility ¼ Cost of electricity þ Cost of natural gas Melbourne: Fair Work Commission Australia.
Fang, Z., Bhandari, B., 2011. Effect of spray drying and storage on the stability of
¼ 196; 991 USD=year þ 54; 056 USD=year
bayberry polyphenols. Food Chem. 129 (3), 1139–1147, http://dx.doi.org/
¼ 251; 047 USD=year 10.1016/j.foodchem.2011.05.093.
Finley, J.W., Given Jr., P., 1986. Technological necessity of antioxidants in the food
industry. Food Chem. Toxicol. 24 (10–11), 999–1006, http://dx.doi.org/10.1016/
0278-6915(86)90280-2.
A.5. Cost of manufacturing Hanus, M.J., Langrish, T.A.G., 2007. Re-entrainment of wall deposits from a
laboratory-scale spray dryer. Asia-Pacific J. Chem. Eng. 2 (2), 90–107. http://
dx.doi.org/10.1002/apj.47.
The manufacturing cost of orange peel extract powder was cal- Hoang, M., Bolto, B., Haskard, C., Barron, O., Gray, S., Leslie, G., 2009. Desalination in
culated using the equation proposed by Turton et al. (2009), as Australia: CSIRO: Water for a Healthy Country National Research Flagship.
shown in Eq. (A.9). Cost of manufacturing (COM) is a function of International Energy Agency, 2012. Electricity Information 2012. IEA, France.
Ito, N., Hirose, M., Imaida, K., 1997. Antioxidants: Carcinogenic and
the fixed capital cost of investment (FCI), the cost of operational
Chemopreventive Properties Encyclopedia of Cancer, vol. 1. Elsevier, USA, pp.
labor (COL), the cost of utilities (CUT), the cost of water treatment 89–101.
(CWT), and the cost of raw material (CRW). Keogh, R., Robinson, A., Mullins, I., 2010. Pollination aware case study: citrus a real
value of pollination in Australia. In: Rural Industries Research and Development
COMðUSD=yearÞ ¼ 0:280  FCI þ 2:37  COL þ 1:23 Corporation (Ed.), vol. Pub. No. 10/081, pp. 1–2.
Kha, T.C., Nguyen, M.H., Roach, P.D., 2010. Effects of spray drying conditions on the
 ðCUT þ CWT þ CRMÞ ðA:9Þ physicochemical and antioxidant properties of the Gac (Momordica
cochinchinensis) fruit aril powder. J. Food Eng. 98 (3), 385–392, http://
dx.doi.org/10.1016/j.jfoodeng.2010.01.016.
COMðUSD=yearÞ ¼ 0:280  3; 678; 302 þ 2:37  412; 524 Korea Electric Power Corporation. <http://cyber.kepco.co.kr> (accessed: September/
2013).
þ 1:23  ð251; 047 þ 0 þ 2; 463; 073Þ Langrish, T.A.G., Premarajah, R., 2013. Antioxidant capacity of spray-dried plant
¼ 5; 340; 934 USD=year extracts: experiments and simulations. Adv. Powder Technol. 24 (4), 771–779,
http://dx.doi.org/10.1016/j.apt.2013.03.020.
Leal, P., Maia, N., Carmello, Q.C., Catharino, R., Eberlin, M., Meireles, M.A., 2008.
5; 345; 975 USD=year Sweet basil (Ocimum basilicum) Extracts Obtained by supercritical fluid
COMðUSD=kgÞ ¼ ¼ 2:00 USD=kg extraction (SFE): global yields, chemical composition, antioxidant activity,
2; 673; 000 kg=year
and estimation of the cost of manufacturing. Food Bioprocess Technol. 1 (4),
326–338. http://dx.doi.org/10.1007/s11947-007-0030-1.
Li, Y., Radoiu, M., Fabiano-Tixier, A.-S., Chemat, F., 2012. From laboratory to
A.6. Selling price industry: scale-up, quality, and safety consideration for microwave-assisted
extraction. In: Chemat, F., Cravotto, G. (Eds.), Microwave-Assisted Extraction for
Bioactive Compounds. Springer, New York, pp. 207–230.
Price of gallic acid = 426.6 USD/kg gallic acid (data from Sigma Liazid, A., Palma, M., Brigui, J., Barroso, C.G., 2007. Investigation on phenolic
Aldrich). compounds stability during microwave-assisted extraction. J. Chromatogr. A
TPC of powder = 19 mg GAE/g DM = 0.019 kg GAE/kg DM. 1140 (1–2), 29–34, http://dx.doi.org/10.1016/j.chroma.2006.11.040.
Lin, C.S.K., Pfaltzgraff, L.A., Herrero-Davila, L., Mubofu, E.B., Abderrahim, S., Clark,
Selling price = 426.6 USD/kg gallic acid  0.019 kg GAE/kg J.H., Koutinas, A.A., Kopsahelis, N., Stamatelatou, K., Dickson, F., Thankappan, S.,
DM = 8.11 USD/kg. Mohamed, Z., Brocklesby, R., Luque, R., 2013. Food waste as a valuable resource
for the production of chemicals, materials and fuels. Current situation and
global perspective. Energy Environ. Sci. 6 (2), 426–464. http://dx.doi.org/
10.1039/c2ee23440h.
References Liu, R.H., 2003. Health benefits of fruit and vegetables are from additive and
synergistic combinations of phytochemicals. Am. J. Clin. Nutr. 78 (3), 517S–
Ahmad, J., Langrish, T.A.G., 2012. Optimisation of total phenolic acids extraction 520S.
from mandarin peels using microwave energy: the importance of the Maillard Masters, K., 1979. Spray Drying Handbook. Halsted Press, New York.
reaction. J. Food Eng. 109 (1), 162–174, http://dx.doi.org/10.1016/ Mezzomo, N., Martínez, J., Ferreira, S.R.S., 2011. Economical viability of SFE from
j.jfoodeng.2011.09.017. peach almond, spearmint and marigold. J. Food Eng. 103 (4), 473–479, http://
Ajila, C.M., Naidu, K.A., Bhat, S.G., Rao, U.J.S.P., 2007. Bioactive compounds and dx.doi.org/10.1016/j.jfoodeng.2010.10.032.
antioxidant potential of mango peel extract. Food Chem. 105 (3), 982–988, Origin Energy Australia. <http://www.originenergy.com.au/> (accessed: September/
http://dx.doi.org/10.1016/j.foodchem.2007.04.052. 2013).
42 D. Shofinita, T.A.G. Langrish / Journal of Food Engineering 139 (2014) 31–42
Pereira, C.G., Prado, J.M., Meireless, M.A.A., 2013. Economic evaluation of natural
product extraction processes. In: Rostagno, M.A., Prado, J.M. (Eds.), Natural
Product Extraction: Principles and Applications. Royal Society of Chemistry,
Cambridge, pp. 442–471.
Reed, T., 2013. Energy Shock: The Gas Cruch is Here. The Australian Industry Group,
North Sydney.
Rosa, P.T.V., Meireles, M.A.A., 2005. Rapid estimation of the manufacturing cost of
extracts obtained by supercritical fluid extraction. J. Food Eng. 67 (1–2), 235–
240, http://dx.doi.org/10.1016/j.jfoodeng.2004.05.064.
Sigma Aldrich Australia. <http://www.sigmaaldrich.com> (accessed: September/
2013).
Singleton, V.L., Orthofer, R., Lamuela-Raventos, R.M., 1999. Analysis of total phenols
and other oxidation substrates and antioxidants by means of folin–ciocalteu
reagent. Oxid. Antioxid. A 299 (4), 152–178.
Smith, J.M., Van Ness, H.C., Abbott, M.M., 2005. Introduction to Chemical
Engineering Thermodynamics. McGraw-Hill, Boston.
Towler, G., Sinnott, R.K., 2013. Chemical Engineering Design: Principles, Practice,
and Economics of Plant and Process Design, second ed. Butterworth-
Heinemann, Boston.
Turton, R., Bailie, R.C., Whiting, W.B., Shaeiwitz, J.A., 2009. Analysis, Synthesis, and
Design of Chemical Processes, third ed. Prentice Hall, New Jersey.
United States Bureau of Labor Statistics, 2013. International Labor Comparisons.
Veggi, P.C., Cavalcanti, R.N., Meireles, M.A.A., 2011. Modifier effects on Supercritical
Fluid Extraction (SFE) of some Brazilian plants: antioxidant activity and
economical evaluation. Proc. Food Sci. 1, 1717–1724, http://dx.doi.org/
10.1016/j.profoo.2011.09.253.
Vieira, G.S., Cavalcanti, R.N., Meireles, M.A.A., Hubinger, M.D., 2013. Chemical and
economic evaluation of natural antioxidant extracts obtained by ultrasound-
assisted and agitated bed extraction from jussara pulp (Euterpe edulis). J. Food
Eng. 119 (2), 196–204, http://dx.doi.org/10.1016/j.jfoodeng.2013.05.030.