Received: 18 June 2021 | Revised: 7 August 2021 | Accepted: 9 August 2021

DOI: 10.1002/fsn3.2549

ORIGINAL RESEARCH

Optimization extraction of rosemary essential oils using

hydrodistillation with extraction kinetics analysis

Hanyue Chen | Zhanying Gu | Ling Yang | Ruonan Yang | Yaxin Ji |

Qingyang Zeng | Fangmeng Xiao | Peng Huang

Central South University of Forestry and

Technology, Changsha, China Abstract
Rosemary (Rosmarinus officinales L. (Labiatae)) is one of the major economic crops in
Correspondence
Zhanying Gu, Central South University of the world, and rosemary essential oil (REO) is one of the top products derived from
Forestry and Technology, Changsha, China. rosemary and has excellent commercial prospects. Many factors affect the yield of
Email: [email protected]
REO extracted by hydrodistillation (HY). This study was proposed to identify and
Funding information analyze these factors to maximize the yield of essential oils and reduce the cost. First,
2020 Hunan Province Graduate Education
Innovation Project; Professional Ability two different single-­factor extraction experiments were conducted, (1) adding NaCl
Promotion Project, Grant/Award Number: and (2) using various organs of the plant, to determine the influence of each factor
2020-­41; School-­level Science and
Technology Innovation Project of University on the oil yield. Based on single-­factor experiments, the orthogonal experiments (L9,
Students in Central South University of 33) were designed to determine the optimal conditions for the extraction of rosemary
Forestry and Technology in 2020, Grant/
Award Number: 2020-­18 and 2020-­73; oil. Meanwhile, the kinetic extraction analysis of the test data was carried out. The
Education Department of Hunan results revealed that the highest oil yield was achieved when rosemary leaves were
crushed to 2 cm, the ratio of water to the material was 1:3, and NaCl concentra-
tion was 5%. A simple first-­order kinetic model has also proved to be an acceptable
general choice and allows to predict the output of extraction operations overtime
accurately and robustly in practice. This study provides a reference scheme for using
hydrodistillation to extract rosemary essential oil.

KEYWORDS

essential oil, extraction kinetics, first-­order kinetic model, hydrodistillation, rosemary

1 | I NTRO D U C TI O N characteristic fragrance (de Oliveira et al., 2019). Its health benefits
have been recognized since ancient times, and it is the raw material
Rosemary (Rosmarinus officinalis L.) is an essential perennial ar- of many culinary traditions (Bellumori et al., 2015; Spadi et al., 2021).
omatic, evergreen shrub belonging to the family Lamiaceae and Moore et al. (2016) mentioned that Rosemary extract has antioxi-
indigenous to the Mediterranean region (Naimi et al., 2017; de dant, anti-­inflammatory, antidiabetic, and anticancer properties.
Oliveira et al., 2019). However, it could be found worldwide and Widespread use of rosemary is to extract the essential oil.
grown widely in northern central Europe, America, and East Asia Rosemary essential oil (REO) is a colorless or pale-­yellow volatile
(Ahmed et al., 2016; Tigrine-­Kordjani et al., 2012). This plant has a liquid extracted from the branches and leaves with a characteris-
height of up to 2 m, with branches full of green leaves that exude a tic odor. It consists mainly of monoterpenes such as 1,8-­cineole,

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,
provided the original work is properly cited.
© 2021 The Authors. Food Science & Nutrition published by Wiley Periodicals LLC.

Food Sci Nutr. 2021;9:6069–6077.  www.foodscience-nutrition.com | 6069

6070 | CHEN et al.

camphor, and α-­pinene (Rašković et al., 2014). These compounds in improve the yield. Therefore, it is necessary to study the joint in-
REO make REO can be used in many aspects. REO is considered a fluence of these operating factors to maximize the recovery rate
natural antioxidant, and it has been reported as an inhibitor of lipid of REO.
oxidation in meat products (Sebranek et al., 2005). Furthermore, In addition, distillation time (DT) is known as also very important
REO can be used for treating dyspepsia and mild spasmodic dis- (Spadi et al., 2021). Many studies have modeled the extraction pro-
orders of the gastrointestinal tract (Raskovic et al., 2015), and ex- cess of volatile essential oils and simulated the kinetics by making
tensively used in industry, especially in the flavor and cosmetics many assumptions (Ait Amer Meziane et al., 2019). When dealing
industries, as additives in active packaging and in agriculture as with the relationship between the yield of essential oil and distil-
additives to repel insects and sometimes acting as pheromones lation time by kinetic data, two types of curves were revealed: ex-
(Silvestre et al., 2019). ponential curve and S curve, while HY belonged to the exponential
These characteristics, as well as consumers' increasing demand curve (Benyoussef et al., 2005).
for healthy natural products, make REO a commercial product with In this context, we aimed to optimize the method for extracting
wide application and high economic value (Spadi et al., 2021). REO rosemary essential oil based on the HY method under different de-
accounts for about 1%–­2.5% of the total plants. Like other essential signed experiments. The DT is an independent variable to carry out
oils, their quantity and quality are affected by various internal and ex- the first-­order kinetic modeling. The main objectives of this study
ternal factors. In particular, the content of its chemical composition were (1) to investigate the effect of adding sodium chloride (NaCl) on
varies with the geographical region where plants grow, climate, plant the extraction of rosemary essential oil, (2) to examine the difference
parts used, and extraction methods (Borges et al., 2019). Current in the amount of essential oil extracted from rosemary branches and
methods for extraction of rosemary essential oil include hydrodis- leaves, (3) according to the data of the previous two single-­factor
tillation, steam distillation, ultrasound-­
assisted extraction (Heck experiments, further to determine the optimal extraction of essen-
et al., 2018), molecular distillation, adsorption, supercritical water tial oil using orthogonal experiments, and (4) Lastly, considering the
extraction, CO2 supercritical extraction (Allawzi et al., 2019), and distillation time, different parts of the rosemary, the ratio of water-­
enzyme-­assisted extraction (Nadar et al., 2018). Hydrodistillation to-­raw material, and concentration of NaCl as influencing factors,
(HY) is the simplest method and commonly used to achieve indus- the kinetic extraction model describing the extraction process was
trial production. In HY, plant materials containing essential oils are successfully constructed.
placed in a distiller mixed with water (Spadi et al., 2021). A heat
source heats the mixture, make them undergo physical alterations,
and oil vapors along with water vapors come out due to thermal dif- 2 | M ATE R I A L S A N D M E TH O DS
fusion (Solanki et al., 2018), and the steam that is created passes
through a condenser that allows the recovery of essential oil (Spadi 2.1 | Study site
et al., 2021). This procedure achieves component isolation according
to their degree of hydro-­solubility rather than their boiling points Materials for this study were collected from the Production and
(Presti et al., 2005). Since the cell wall provides the maximum re- Research Base of Spice Plant Cultivation and Utilization (113.08°E,
sistance, this process takes long extraction, relatively high solvent 27.78°N) of Central South University of Forestry and Technology
consumption, and usually irreproducibility (Mohamad et al., 2019; (CSUFT), Zhuzhou, Hunan Province, China. The site is plain,
Solanki et al., 2018). with an average elevation of 60 m. The local climate is a typical
Although this method has been used for a long time, few scien- humid, subtropical monsoon climate, with four distinct seasons,
tific research and operating conditions for REO extraction, some average annual precipitation of 1400–­1700 mm, and an average
ways, mention the use of pressurized heating or the addition of yearly temperature of 16–­18°C. Soil type is mainly acidic red loam
organic solvents to promote extraction, but this does not meet (Marburg, 1994).
the premise of saving energy and protecting the environment.
Therefore, the extraction conditions are significant for maximizing
the yield of REO and optimizing parameters such as energy, time, 2.2 | Plant material collection
raw materials, and solvents (de AR Oliveira et al., 2016). Many
factors affect the extraction process of HY. Adding NaCl to the Rosemary was the 4-­year-­old rooted cuttings in the study site.
water from which essential oil is extracted is an influencing fac- The total sampling area is about 0.67 ha, and the whole rosemary
tor, improving the REO yield during extraction. The oil content in was conducted every 4 m along the diagonal line of the field. The
organs of different parts of plants is different, so it is also a key entire rosemary plants were divided into leaves, 1-­year branches,
factor. The solid/liquid ratio is another critical factor. Optimizing and perennial branches to accommodate different experimental
this ratio should maximize REO production and reduce solvent designs. Plant materials were collected randomly in March 2019,
consumption, thus improving economic and environmental effi- placed in a cooler, and sent immediately to the laboratory. The
ciency (Spadi et al., 2021). Finally, a study (Smallfield et al., 2001) plant materials were placed in a refrigerator at 4°C to ensure
reported that pretreatment, such as crushing raw materials, can freshness.
CHEN et al. | 6071

2.3 | Single-­factor experiments any two factors. In this experiment, variables were determined
based on the results of two single-­factor experiments. Factor A
The two single-­f actor experiments are independent of each other crushed rosemary leaves to verify the influence of particle size on
and have no interaction. Experiment 1 (E1) was an exploratory ex- rosemary essential oil extraction (unbroken, 2-­cm pieces, and 1-­cm
periment; it examined the differences in oil yield related to the pieces, respectively). Furthermore, factors B and C verified the ratio
amount of NaCl added. The experiment used a combination of of water-­to-­raw material (this factor was divided into 1:1; 1:2, and
fresh rosemary leaves and branches. The rosemary material was 1:3) and NaCl concentration (1.5%, 3.0%, and 5.0%) on the yield of
mixed with NaCl and distilled water in a distillation flask, and hy- essential oil extracted, respectively. Table 1 shows the experimental
drodistillation was continued for 3 h to ensure no more essential design for extracting essential oil from rosemary.
oil was obtained. Rosemary contained volatile oil experiences The whole experiment was performed twice with 18 extractions.
thermal stress due to thermal energy supplied by the water during Distillation was performed for 2 h each time, and the amount of es-
hydrodistillation. This phenomenon leads to oil diffusion, which is sential oil was recorded every 30 min to facilitate later calculation.
then carried by water vapor generated during the process toward SPSS20.0 obtained the orthogonal table.
the condenser (Solanki et al., 2018). For extracting the oil–­water
mixture distilled from each condition, the extraction process
started at room temperature (25°C) (Akhbari et al., 2018). The use 2.5 | Kinetic model
of a separating funnel for direct extraction is considered in terms
of environmental protection and economy, and no organic solvent This study is based on the first-­order kinetic model mentioned in
is needed to assist. The ratio of water to the material was 1:3. The Zhang et al. (2020):
following concentrations of NaCl were assessed: 0 (control), 0.5,
1.5, 3.0, 5.0, and 5.5 (unit: %). dYt
(1)
( )
= K1 Ye − Yt
Experiment 2 (E2) examined the variance in oil yield from the dt
different parts of the rosemary. In this experiment, the ratio of water
to the material was 1:1, both of which are 400 (g/ml). Because the where Ye (%) and Yt (%) represent the yield of EO at equilibrium or at
size and essential oil content of different parts of rosemary are dif- any time, respectively, t (min) represents the distillation time, K1 rep-
ferent, the volume of perennial branches is larger than that of 1-­year resents the rate constant.
branches and fresh leaves. However, the maximum bearing capacity According to the initial and boundary conditions: Yt = 0 at t = 0
of the hydrodistillation device used in the laboratory is 800–­1000ml, and Yt = Yt at t = t, Equation (1) is rearranged to a linear equation:
so the solid/liquid ratio of 1:1 can meet the needs of extracting dif-
ferent parts of rosemary. In addition, after the exploration of experi- (2)
( )
ln Ye − Yt = lnYe − K1 t
ment 1, the total distillation time of E2 was determined to be 2 h, and
the amount of essential oil was recorded every 30 min. The distilled A nonlinearized form can be obtained from Equation (2), as
essential oil was bottled individually and labeled with the extraction follows:
date and experiment number. All subsequent experiments are dis-
tilled for 2 h. Both experiments were repeated three times to exam- Yt = Ye 1 − e−K1 t (3)
( )

ine the experimental error. All the experiments use 220 V voltage
in the first 15 min of distillation and change it to 100 V after the According to the actual data obtained in experiments 2.3 and 2.4,
temperature of the distillation flask stabilizes. after processing and analyzing with the first-­order kinetic model, it
is known that the result of experiment 2.3 accords with the Equation
(2) of linear regression in the first-­order kinetic model. In contrast,
2.4 | Multi-­factor experiment the result of the orthogonal experiment is more suitable for fitting
with the nonlinear regression of Equation (3).
The orthogonal L9 (33) design was used to optimize the extraction Therefore, we made the scatter plot of the yield of REO and
conditions. This design assumes that there is no interaction between distillation time (DT) of different rosemary parts in experiment 2.3.

TA B L E 1 Factors and levels for the

Level Level Level
orthogonal experiments
Factors 1 2 3

A. Leaf and branch integrity of the current Unbroken 2-­cm pieces 1-­cm pieces
year
B. Ratio of water-­to-­raw material 1:1 1:2 1:3
C. Concentration of NaCl 1.5% 3.0% 5.0%
6072 | CHEN et al.

With time t as abscissa and REO yield (%) as ordinate, fitted the at 5.5%, the yield was only 0.36%. The extraction yield without
yield increment of essential oil extracted from fresh leaves, annual NaCl was the lowest, at only 0.02%. Overall, with an increase in
branches, and perennial branches by linear regression, and obtained NaCl concentration, the oil yield first increased remarkably, then
three fitting equations describing the extraction increment and time decreased.
(Wang et al., 2015). Furthermore, He and Tian (2003), Li et al. (2014) In extracting essential oil, adding NaCl is a method thought to
proposed a simple method describing the extraction process of improve the oil yield, mainly because the NaCl can effectively tar-
volatile oil (Laplace transform method). The process of extracting get the analyte on a molecular level and enter the rapid extraction
volatile oil from plant cells by hydrodistillation is usually as follows: stage to significantly improve its analysis signal (Bibi et al., 2016).
(i) It diffuses from the inside of cells to the inner wall of cells, that The results show that a certain amount of NaCl can promote the
is, the gas–­solid interface and (ii) volatile oil enters the gas phase yield of rosemary essential oil, but it will inhibit the yield of es-
through the gas–­solid interface. Because the solubility of volatile oil sential oil beyond that amount. One reason for this inhibition is
in water is minimal, the transmission process of a small number of that, with the increase in salinity, the water content of rosemary
volatile oil molecules in the liquid phase can be ignored; (iii) volatile leaves decreases, while the content of phenols, alkaloids, and
oil molecules entering the gas phase are taken out of the system by brass compounds increases (Móricz et al., 2016). These substances
continuously generated water vapor and transferred into contain- are usually the main components of rosemary essential oil, and
ers for collection and preservation. Then, used the best extraction the phenols have good antioxidant properties, which is one of
conditions obtained by orthogonal test, and the data of essential oil the medicinal uses for rosemary essential oil (Moore et al., 2016).
yield and DT were fitted by nonlinear regression. Equation (3) is the However, excessively high salinity will increase the oxidative
BoxLucas1 model in the exponential function, and the kinetic ex- stress of organic components in essential oil, thereby inhibiting
traction equation under this treatment is obtained by exponential their extraction (Móricz et al., 2016).
fitting with Origin. This study demonstrates that, with a water-­to-­raw material
ratio of 1:3, the addition of NaCl influences the oil yield. The ex-
traction yields under the conditions of 3% and 5% NaCl are very
2.6 | Statistical analysis high, reaching 0.61% and 0.60%. When NaCl concentration is
1.5%, the extraction yield is 0.55%; according to the results ob-
Descriptive statistical analysis and one-­way analysis of variance tained in E1, therefore, it is recommended to use a concentration
were used to analyze the data for the two single-­factor experi- of NaCl between 1.5% and 5.0% for distillation extraction, which
ments. For the orthogonal experiments, Univariate Analysis of can save costs and ensure a high extraction yield and applied to
Variance in the general linear model was used to find the relevant the subsequent orthogonal experiment of three factors and three
values such as the extreme value (R) and F-­value of each factor. levels.
Oil yield for all experiments was calculated using the following
equation:
3.2 | Variation in oil yield from different
Meo (g)
(4)
parts of the rosemary
Reo(%) = × 100
Md (g)
The branches and leaves of rosemary contain aromatic oil (Rafael
Where Reo = REO yield (% w/w), Meo = REO mass extracted (g), and et al., 2011). Due to the essential oil glands located in the leaves
Md = dry mass of rosemary leaves (g) (Spadi et al., 2021). and the flowers (Presti et al., 2005), rosemary essential oil is usu-
The figure and all tables were created using Excel 2016 and ally conducted by hydrodistillation of fresh leaves of rosemary
Origin 9.65. annual flowering buds (Zheljazkov et al., 2015). Based on previ-
ous studies, REO is mainly located in leaves and flowers. It seems
that few studies have mentioned whether rosemary branches can
3 | R E S U LT S A N D D I S CU S S I O N be used as the main raw materials for essential oil extraction.
Therefore, in this study, we set up experiments to verify whether
3.1 | Effect of NaCl on REO extraction yield the 1-­year branches and perennial branches except leaves can
extract essential oils. The oil yield of fresh leaves and 1-­year
E1 is an exploratory experiment completed under the conditions branches was sharply higher than perennial branches (Figure 1).
of a material-­to-­water ratio of 1:3 and distillation time of 3 h. It is In comparison, the average oil yield of 1-­
year branches was
confirmed that NaCl concentration directly influences the oil pro- slightly higher than that of fresh leaves. The mean difference is
duction rate (Table 2). When the concentration of NaCl reached significant at the 0.05 level. The direct yield of 1-­year branches
3%, the extraction yield reached a maximum of 0.61%. When the is slightly higher than that of fresh leaves, but the difference be-
concentration of NaCl reached 5%, the extraction yield reached tween the two factors is not significant (p > .05), so there is no
0.60%. This trend does not hold for increased NaCl concentrations; comparative value. However, the results alone are also intriguing,
CHEN et al. | 6073

TA B L E 2 Oil yield related to concentration of NaCla In the long run, as a perennial economic plant, rosemary can survive
for decades under suitable environmental conditions. Seasonal har-
Concentration of NaCl (%) Yield (%)
vest of the rosemary leaves and branches is suggested to maximize
0 0.02 ± 0.003e
product benefits.
0.5 0.20 ± 0.003d
1.5 0.55 ± 0.028b
3.0 0.61 ± 0.020a 3.3 | Orthogonal experimental design generates
5.0 0.60 ± 0.010a optimum results
5.5 0.36 ± 0.040 c

Note: Based on one-­way analysis of variance. Values represent In E1 and E2, we found that adding NaCl to the solution can improve
means ± SD. the yield in the process of hydrodistillation of rosemary, while the
b,c, d, e
Yields are significantly different froma and from each other. extraction yield of 1-­year branches and fresh leaves of rosemary is
a
Yields at 3.0% and 5.0% are not significantly different. higher. Therefore, the rosemary raw material used in the orthogo-
nal test is the leafy rosemary (including 1-­year branches and fresh
leaves), and the influence of the pretreatment of crushing rose-
mary raw material on the yield is further discussed. Based on the
single-­f actor tests above, the concentration of NaCl and different
parts of rosemary appear to be the significant factors that affect
the yield of essential oil from rosemary. In the present study, these
factors were examined using an orthogonal L9 (3 3) test design. The
test results are shown in Table 3, which indicates that the maxi-
mum yield of essential oil was 1.52%. However, we cannot select
the best extraction conditions based only on the outcomes, and
therefore, a further orthogonal analysis was warranted. Thus, the
values of K, k, and R values were calculated and listed in Table 3.
As seen from Table 3, the influence on the extraction yield of es-
sential oil decreased in the order: B > C>A. The ratio of water-­
to-­
r aw material was the most important determinant of yield
according to the R values. The maximum yield of essential oil was
obtained when the plant material crushed 2-­cm pieces, the water
F I G U R E 1 Oil yields from three different parts of the rosemary.
*Error bars represent the standard deviation of the mean. aYields in ratio to raw material was 1:3, and the NaCl concentration was at
fresh leaves and one-­year branches are not significantly different. 5.0% (combination A2B3C3), respectively.
b
Yields are significantly different from a We next evaluated the F-­value and the significance level of each
factor, as shown in Table 4 (p = .05). The results show that, although
the water-­to-­material ratio has the most significant influence on the
proving that rosemary branches also contain many essential oils, oil yield among all factors, the influence of the water-­to-­material
especially the branches of the 1-­year. Our guess is that because ratio is not statistically different from the influence of the other
rosemary has a unique plant morphology, many tiny leaves and factors.
unopened buds gather at the top of short branches to form a We examined different water-­to-­material ratios, blade integrity,
race. These microscopic parts containing essential oils are dis- and NaCl concentration in the orthogonal experiments of this study.
tilled together with the branches of the 1-­year, which makes the The best processing conditions in the traditional sense are obtained
yield of branches of the 1-­year increase somewhat. This result by evaluating the range R, and the mean K. Borhan et al. (2013) men-
can also inspire the follow-­u p researchers and further explore tioned the size of the particles that can interfere in the extraction
the comparison between the 1-­year branches' yield results and process. The smaller the particle size, the higher the interaction
fresh leaves. between the plant sample and the solvent to obtain (de Oliveira
However, as far as this experiment is concerned, from the point et al., 2019). By comparing the k value of factor A, it can be seen that
of view of saving cost and increasing yield, it can be advocated to ex- the extraction rate is the highest when the materials are crushed to
tract essential oil from 1 year's leafy branches instead of just taking 2 cm. Using the results of significance tests (Table 4), the F-­values
fresh leaves. In the subsequent orthogonal experiment, the 1-­year of the three factors are more significant than Sig. (p > .05) indicates
leafy branches of rosemary were used as experimental materials. that factor A, leaf integrity, factor B, water-­to-­material ratio, and fac-
The influence of crushing the one-­year leafy branches on the yield tor C, the NaCl concentration, have no significant effect on the oil
was further verified from pretreatment of experimental materials. yield.
 6074 | CHEN et al.

TA B L E 3 Analysis of orthogonal
A. Leaf and branch integrity B. Ratio of water-­ C. Concentration Extraction
experimental design
No. of the current year to-­raw material of NaCl yield (%)

1 1 1 1 0.67
2 1 2 2 0.47
3 1 3 3 1.52
4 2 1 2 0.89
5 2 2 3 0.74
6 2 3 1 1.05
7 3 1 3 0.62
8 3 2 1 0.57
9 3 3 2 0.92
K1 2.66 2.18 2.29
K2 2.68 1.78 2.28
K3 2.11 3.49 2.88
k1 0.89 0.73 0.76
k2 0.89 0.59 0.76
k3 0.70 1.16 0.96
R 0.19 0.57 0.20

Note: K1 represents the average value of extraction yield from three experimental replicates under
Level 1, and the same is true for K 2 and K3 under Levels 2 and 3, respectively.
k1 represents the value obtained by dividing K1 by the test times under Level 1, and the same is
true for k 2 and k3 under Levels 2 and 3, respectively.
R1 refers to the results of the extreme analysis.

TA B L E 4 Test of significance of the

Deviation sum Mean
orthogonal experiment
Factors of squares df Square F Sig.

A. Leaf and branch integrity 0.205 2 0.103 1.558 .391

B. Ratio of water-­to-­raw material 0.183 2 0.091 1.388 .419
C. Concentration of NaCl (%) 0.293 2 0.147 2.227 .310

Note: df, degrees of freedom; Sig., significant differences in the influence of factors on the
extraction rate of essential oil. p = .05.

0.011
Fresh leaves

0.009 One-year branches

Perennial branches

0.007 First-order kinec model

Yirld of REO (%)

0.005

0.003

0.001 F I G U R E 2 Kinetic curve for the yield

of essential oil obtained by distilling
different parts of rosemary. Points
-0.001
represent actual experimental data, and
-10 10 30 50 70 90 110 130
lines represent fitting behaviors predicted
Time (min) by first-­order kinetic model

In the comprehensive comparison, the water-­to-­rosemary ratio considering both the addition of NaCl and the physical breakdown
is the most significant factor affecting the oil yield, and the best of the plant parts to achieve the optimal extraction conditions and
level is 1:3. Therefore, in actual industrial production, we suggest the maximum oil yield.
CHEN et al. | 6075

TA B L E 5 The first-­order kinetic model of experiment 2.3, DT is

the distillation time, and the number is the fitting parameter

Factors Fitting equation R2

Fresh leaves Y = 0.00008DT−0.0002 .9995

One-­year branches Y = 0.00007DT+0.00005 .9937
perennial branches Y = 0.00003DT−0.0003 .9883

2.67% theoretically. The results show that the regression coefficient

R 2 is more significant than .95, which indicates that the equation
is reliable, and the obtained results can be used for optimization.
Under the support of this theory, the follow-­up research group used
rosemary raw materials from the same origin to carry out the same
experimental treatment, which was repeated three times. The final
F I G U R E 3 The first-­order kinetic curve of essential oil yield actual yield was between 2.52% and 2.73%, which was equivalent
was obtained under A2B3C3 treatment. Points represent data, and to the theoretical value, proved the treatment's reliability (Table 5).
lines represent fitting behaviors predicted by the first-­order kinetic
model

4 | CO N C LU S I O N

3.4 | Extraction kinetics Extraction of rosemary essential oil was studied in single-­factor ex-
periments and orthogonal experiments and used the first-­order ki-
The kinetic data of the single-­factor experiment and orthogonal ex- netics model to verify the data. The single-­factor experiments used
periment on different parts of rosemary and the kinetic curves fitted different concentrations of NaCl and different parts of the rosemary
by the first-­order kinetic model are shown in Figures 2 and 3. The plant to examine extraction conditions. Orthogonal tests examined
extraction yield of essential oil from different parts of rosemary has water-­to-­material ratios, concentrations of NaCl, and integrity of
an excellent linear relationship with DT (R 2 > 0.98), which accords leafy branches of rosemary. When extracting essential oil by hydro-
with the first-­order kinetic model. It can be observed from Figure 2 distillation, the results show that it is preferred to add NaCl concen-
that the yield of REO gradually increases with the extension of distil- tration at 5%, crushed the branches and leaves from the current year
lation time. Under the same distillation time, the REO yield of fresh about 2 cm, and use a water-­to-­material ratio of 1:3 to obtain the
branches and existing shoots showed apparent advantages, which high extraction yield. This is the result of economic and environmen-
indicated that different parts of rosemary had a particular influence tal considerations. In addition, according to the kinetic model, the
on the extraction yield of volatile oil and verified the conclusion of yield under the optimum extraction conditions is estimated to be
3.2. With the extension of distillation time, the extraction content 2.67%. However, essential oil content in rosemary may be affected
of fresh leaves was slightly higher than that of annual branches. by many factors, such as plant growth conditions (including soil, cli-
According to the results of content determination, the final selection mate, precipitation, etc.), so this study cannot wholly solve essen-
conditions were as follows: fresh rosemary leaves, distillation for 2 h, tial oil extraction. In conclusion, rosemary is a volatile oil product
applied to the orthogonal test. used widely in many fields and has significant economic value in the
According to the optimization results of the orthogonal test, the market. Our experiment revealed the best conditions for extracting
distillation time (DT = 2 h) is taken as the abscissa, and the increase rosemary essential oil by hydrodistillation.
of REO yield is taken as the ordinate. The curve is drawn by nonlin-
ear fitting (Figure 3). The kinetic model equation of REO yield is as AC K N OW L E D G M E N T
follows: The authors would like to acknowledge the financial support of the
2020 Hunan Province Graduate Education Innovation Project and
Yt = 0.0267(1 − e−0.0074DT ) (5)
Professional Ability Promotion Project (2020-­41) and School-­level
Science and Technology Innovation Project of University Students in
Yt is the REO yield (%) of the device under the processing con- Central South University of Forestry and Technology in 2020 (2020-­
ditions, and DT is the distillation time. We can get the best treat- 18, 2020-­73), and the Education Department of Hunan.
ment result through the orthogonal test, get a more considerable
predicted value of rosemary essential oil yield after balance, and es- C O N FL I C T O F I N T E R E S T
tablish a characteristic kinetic model to describe the process better. It is declared that there is no conflict of interest in the publication
Under this model, the rate of return after REO equilibrium can reach of this work.
6076 | CHEN et al.

AU T H O R C O N T R I B U T I O N involved. Journal of Ethnopharmacology, 229, 29–­ 45. https://doi.

org/10.1016/j.jep.2018.09.038
Hanyue Chen: Data curation (lead); Formal analysis (lead);
Borhan, M. Z., Ahmad, R., Mohd, R. M., & Abdullah, S. (2013). Impact of
Methodology (equal); Visualization (lead); Writing-­original draft (lead); nano powders on extraction yield of Centella asiatica. Advances in
Writing-­review & editing (equal). Zhanying Gu: Conceptualization Materials Research, 667, 246–­250.
(lead); Funding acquisition (lead); Investigation (equal); Methodology de Oliveira, J. R., Camargo, S., & de Oliveira, L. D. (2019). Rosmarinus
officinalis L. (rosemary) as therapeutic and prophylactic agent.
(equal); Project administration (lead); Resources (lead); Supervision
Journal of Biomedical Science, 26(1), 5. https://doi.org/10.1186/s1292​
(lead); Validation (equal); Writing-­
review & editing (equal). Ling 9-­019-­0 499-­8
Yang: Investigation (equal); Methodology (equal). Ruonan Yang: He, B., & Tian, J. (2003). Dynamic model of volatile oil extraction pro-
Investigation (equal). Yaxin Ji: Investigation (equal). Qingyang Zeng: cess and determination of its parameters. Journal of Luzhou Medical
Investigation (equal). Peng Huang: Investigation (equal). Fangmeng College, 26, 478–­481. https://kns.cnki.net/kcms/detai​l/detail.aspx?-
FileN​ame=LXYB2​0 0306​0 02&DbNam​e=CJFQ2003
Xiao: Investigation (equal).
Heck, R. T., Lucas, B. N., Santos, D., Pinton, M. B., Fagundes, M. B., de
Araújo Etchepare, M., Cichoski, A. J., de Menezes, C. R., Barin, J. S.,
E T H I C A L A P P R OVA L Wagner, R., & Campagnol, P. (2018). Oxidative stability of burgers
Our research did not contain any animal experiments and human containing chia oil microparticles enriched with rosemary by green-­
extraction techniques. Meat Science, 146, 147–­ 153. https://doi.
subjects.
org/10.1016/j.meats​ci.2018.08.009
Li, J. E., Nie, S. P., Shao, D. Y., Li, C., & Xie, M. Y. (2014). Dynamic rela-
DATA AVA I L A B I L I T Y S TAT E M E N T tionship between extraction yield and extraction time of essential oil
The data that support the findings of this study are available on re- from Mosla Chinensis Maxim cv. Jiangxiangru. Journal of Food Science
and Biotechnology, 33, 308–­314.
quest from the corresponding author. The data are not publicly avail-
Marburg, M. W. (1994). Herbal drugs and phytopharm aceuticals.
able due to privacy or ethical restrictions. Medpharm Scientific Publishers.
Mohamad, N., Ramli, N., Abd-­ A ziz, S., & Ibrahim, M. F. (2019).
ORCID Comparison of hydrodistillation, hydrodistillation with enzyme-­
assisted and supercritical fluid for the extraction of essential oil from
Zhanying Gu https://orcid.org/0000-0002-3233-2291
pineapple peels. 3 Biotech, 9(6), 234. https://doi.org/10.1007/s1320​
5-­019-­1767-­8
REFERENCES Moore, J., Yousef, M., & Tsiani, E. (2016). Anticancer effects of rosemary
Ahmed, Z., Ouassila, L., Abdeslam, H. M., Christelle, C., & Elisabeth, B. (Rosmarinus officinalis L.) extract and rosemary extract polyphenols.
(2016). Optimization of Algerian rosemary essential oil extraction Nutrients, 8(11), 731. https://doi.org/10.3390/nu811​0731
yield by supercritical CO2 using response surface methodology. Móricz, Á. M., Horváth, G., Böszörményi, A., & Ott, P. G. (2016). Detection
Comptes Rendus Chimie, 19, 538–­ 543. https://doi.org/10.1016/j. and identification of antibacterial and antioxidant components
crci.2015.08.011 of essential oils by TLC-­biodetection and GC-­MS. Natural Product
Ait Amer Meziane, I., Bali, N., Belblidia, N., Abatzoglou, N., & Communications, 11(11), 1705–­1708. https://doi.org/10.1177/19345​
Benyoussef, E. (2019). The first-­ o rder model in the simulation 78X16​01101120
of essential oil extraction kinetics. Journal of Applied Research Nadar, S. S., Rao, P., & Rathod, V. K. (2018). Enzyme assisted extraction
on Medicinal and Aromatic Plants, 15, 100226. https://doi. of biomolecules as an approach to novel extraction technology: A
org/10.1016/j.jarmap.2019.100226 review. Food Research International (Ottawa, Ont.), 108, 309–­330.
Akhbari, M., Masoum, S., Aghababaei, F., & Hamedi, S. (2018). https://doi.org/10.1016/j.foodr​es.2018.03.006
Optimization of microwave assisted extraction of essential oils from Naimi, M., Vlavcheski, F., Shamshoum, H., & Tsiani, E. (2017). Rosemary
Iranian Rosmarinus officinalis L. using RSM. Journal of Food Science extract as a potential anti-­hyperglycemic agent: Current evidence and
and Technology, 55(6), 2197–­ 2207. https://doi.org/10.1007/s1319​ future perspectives. Nutrients, 9(9), 968. https://doi.org/10.3390/
7-­018-­3137-­7 nu909​0968
Allawzi, M., Allaboun, H., & Almasri, A. (2019). CO₂ supercritical ex- Oliveira, G. de A. R., de Oliveira, A. E., da Conceiç˜ao, E. C., & Leles,
traction of essential oil of Jordanian rosemary. Journal of AOAC M. I. (2016). Multiresponse optimization of an extraction proce-
International, 102(2), 662–­ 665. https://doi.org/10.5740/jaoac​ dure of carnosol and rosmarinic and carnosic acids from rosemary.
int.18-­0221 Food Chemistry, 211, 465–­ 473. https://doi.org/10.1016/j.foodc​
Bellumori, M., Michelozzi, M., Innocenti, M., Congiu, F., Cencetti, G., hem.2016.05.042
& Mulinacci, N. (2015). An innovative approach to the recovery of Presti, M. L., Ragusa, S., Trozzi, A., Dugo, P., Visinoni, F., Fazio, A., Dugo,
phenolic compounds and volatile terpenes from the same fresh foliar G., & Mondello, L. (2005). A comparison between different tech-
sample of Rosmarinus officinalis L. Talanta, 131, 81–­87. https://doi. niques for the isolation of rosemary essential oil. Journal of Separation
org/10.1016/j.talan​t a.2014.07.073 Science, 28(3), 273–­280. https://doi.org/10.1002/jssc.20040​0 037
Benyoussef, E.-­H., Yahiaoui, N., Khelfaoui, A., & Aid, F. (2005). Water Rafael, B. S., Argimiro, R. S., Rafael, D. P. S., & Eduardo, C. (2011). Dynamic
distillation kinetic study of spearmint essential oil and of its major simulation of rosemary essential oil extraction in an industrial steam
components. Flavour and Fragrance Journal, 20(1), 30–­33. https://doi. distillation unit. I&EC Research, 50, 3955–­3959.
org/10.1002/ffj.1406 Rašković, A., Milanović, I., Pavlović, N., Ćebović, T., Vukmirović, S., &
Bibi, S., Shazia, I., & Zahoor, A. B. (2016). Chemical composition of rose- Mikov, M. (2014). Antioxidant activity of rosemary (Rosmarinus of-
mary (Rosmarinus officinalis L.) leaves under salt stress. Pure and ficinalis L.) essential oil and its hepatoprotective potential. BMC
Applied Biology, 2, 355–­360. Complementary and Alternative Medicine, 14, 225. https://doi.
Borges, R. S., Ortiz, B. L. S., Pereira, A. C. M., Keita, H., & Carvalho, J. org/10.1186/1472-­6882-­14-­225
C. T. (2019). Rosmarinus officinalis essential oil: A review of its phy- Raskovic, A., Milanovic, I., Pavlovic, N., Milijasevic, B., Ubavic, M., &
tochemistry, anti-­inflammatory activity, and mechanisms of action Mikov, M. (2015). Analgesic effects of rosemary essential oil and its
CHEN et al. | 6077

interactions with codeine and paracetamol in mice. European Review Analytical Methods, 5(3), 596–­ 603. https://doi.org/10.1007/s1216​
for Medical and Pharmacological Sciences, 19(1), 165–­172. 1-­011-­9283-­4
Sebranek, J. G., Sewalt, V. J., Robbins, K. L., & Houser, T. A. (2005). Wang, H., Shi, Y. J., Deng, H., Song, R., & Wang, M. (2015). Extraction
Comparison of a natural rosemary extract and BHA/BHT for relative kinetics of hydroxysafflor yellow A in carthamus tinctorious. Anhui
antioxidant effectiveness in pork sausage. Meat Science, 69(2), 289–­ Medical and Pharmaceutical Journal, 10, 1853–­1856. https://kns.cnki.
296. https://doi.org/10.1016/j.meats​ci.2004.07.010 net/kcms/detai​l/detail.aspx?FileN​ame=AHYY2​01510 ​0 05&DbNam​
Silvestre, W. P., Medeiros, F. R., Agostini, F., Toss, D., & Pauletti, G. F. e=CJFQ2015
(2019). Fractionation of rosemary (Rosmarinus officinalis L.) essential Zhang, Q., Huo, R. Y., Ma, Y. B., Yan, S. M., Yang, L., & Chen, F. L. (2020).
oil using vacuum fractional distillation. Journal of Food Science and A novel microwave-­assisted steam distillation approach for separa-
Technology, 56(12), 5422–­5434. https://doi.org/10.1007/s1319​7-­ tion of essential oil from tree peony (Paeonia suffruticosa Andrews)
019-­0 4013​-­z petals: Optimization, kinetic, chemical composition and antioxi-
Smallfield, B. M., van Klink, J. W., Perry, N. B., & Dodds, K. G. (2001). dant activity. Industrial Crops & Products, 154, 112669. https://doi.
Coriander spice oil: effects of fruit crushing and distillation time on org/10.1016/j.indcr​op.2020.112669
yield and composition. Journal of Agricultural and Food Chemistry, Zheljazkov, V. D., Astatkie, T., Zhalnov, I., & Georgieva, T. D. (2015).
49(1), 118–­123. https://doi.org/10.1021/jf001​024s Method for attaining rosemary essential oil with differential com-
Solanki, K. P., Desai, M. A., & Parikh, J. K. (2018). Sono hydrodistilla- position from dried or fresh material. Journal of Oleo Science, 64(5),
tion for isolation of citronella oil: A symbiotic effect of sonication 485–­496. https://doi.org/10.5650/jos.ess14258
and hydrodistillation towards energy efficiency and environment
friendliness. Ultrasonics Sonochemistry, 49, 145–­ 153. https://doi.
org/10.1016/j.ultso​nch.2018.07.038
How to cite this article: Chen, H., Gu, Z., Yang, L., Yang, R., Ji,
Spadi, A., Angeloni, G., Guerrini, L., Corti, F., Michelozzi, M., Cencetti,
G., Parenti, A., & Masella, P. (2021). Using a Plackett-­Burman design
Y., Zeng, Q., Xiao, F., & Huang, P. (2021). Optimization
to maximise yield of rosemary essential oil by distillation. Industrial extraction of rosemary essential oils using hydrodistillation
Crops and Products, 166, 113488. https://doi-­org.ezpro​x y.bangor. with extraction kinetics analysis. Food Science & Nutrition, 9,
ac.uk/10.1016/j.indcr​op.2021.113488 6069–­6077. https://doi.org/10.1002/fsn3.2549
Tigrine-­Kordjani, N., Meklati, B. Y., Chemat, F., & Guezil, F. Z. (2012).
Kinetic investigation of rosemary essential oil by two methods:
Solvent-­free microwave extraction and hydrodistillation. Food