Journal of Food Engineering 186 (2016) 27e33

Contents lists available at ScienceDirect

Journal of Food Engineering

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

Evolution of sulforaphane content in sulforaphane-enriched broccoli

during tray drying
Andrea Mahn*, Constanza Martin, Alejandro Reyes, Aldo Saavedra
n Central, Santiago 9170019,
Department of Chemical Engineering, University of Santiago of Chile, Avenida Libertador Bernardo O'Higgins 3363, Estacio
Chile

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

Article history: Sulforaphane is a natural anticancer compound found in broccoli that comes from hydrolysis of glu-
Received 14 January 2016 coraphanin. Conversion of glucoraphanin to sulforaphane has been optimized, however, its use as
Received in revised form functional ingredient is limited because sulforaphane is thermo-labile. We investigated the effect of
17 March 2016
drying air temperature (60, 70, 80
C) in tray drying of sulforaphane-enriched broccoli on the evolution
Accepted 4 April 2016
Available online 6 April 2016
of sulforaphane content. Broccoli temperature and sulforaphane content were registered in time. Sul-
foraphane content profiles were adjusted to a first-order kinetic model, showing acceptable agreement
(r > 0.90). Sulforaphane formation occurred below 40
C; formation and degradation occurred at broccoli
Keywords:
Sulforaphane degradation
temperature above 40
C, until the low content of glucoraphanin and moisture, prevents reaction. After
Broccoli that, only sulforaphane degradation was detected. The highest sulforaphane content at X/X0 ¼ 0.1 was
Tray drying 67.6 mg/100 g DW, obtained with drying air temperature of 70
C, being 4-fold higher than that found in
Kinetic model fresh broccoli, and the highest reported so far in any dehydrated food.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction disfavor sulforaphane formation. As an option to deliver sulfo-

Sulforaphane is an isothiocyanate that comes from brassicaceae
vegetables, and it is considered to be a powerful natural anticancer
compound (Matusheski et al., 2004; Elbarbry and Elrody, 2011). Its
precursor is glucoraphanin, which is the most abundant glucosi-
nolate in some broccoli cultivars. Among brassicaceae, broccoli has
by far the highest content of glucoraphanin. The hydrolysis of
glucoraphanin to yield sulforaphane (Fig. 1A) proceeds through the
action of myrosinase (EC 3.2.1.147) to give an unstable intermediate,
which can be subsequently converted spontaneously into nitriles,
thiones, thiocyanates or isothiocyanates, depending on the chem-
ical conditions i.e. pH, temperature, presence of Feþ2, presence and
activity of epithiospecifier protein (ESP) (Gu et al., 2012). Sulfo-
raphane is the main product of the reaction when it proceeds at
neutral pH and when ESP is inactive (Shen et al., 2010). Although
glucoraphanin can be hydrolyzed by myrosinase, which is released
from the myrosin cells of the vegetable during mastication and
digestion in the intestine, the bioavailability of sulforaphane in this
case is rather low, since the chemical conditions in intestine
* Corresponding author.
E-mail address: andrea.mahn@usach.cl (A. Mahn).
rez et al. (2014)
raphane instead of its precursor in the vegetable, Pe
proposed an optimized process that consists of blanching at 57
C
for 3 min, followed by an incubation step at 40
C. This resulted in
94% conversion of glucoraphanin to sulforaphane, reaching a con-
centration of 142 mg sulforaphane per 100 g dry weight, the
highest sulforaphane content reported so far (Pe rez et al., 2016). In
order to exploit these results in the field of functional foods, the
pre-processed broccoli, rich in sulforaphane, could be dehydrated
so that it can be incorporated as a functional ingredient in further
elaborated foods. The main hindrance of this approach relies on the
thermal instability of sulforaphane, which can result in consider-
able loss of the compound during hot air drying.
Research about the behavior of bioactive compounds and sul-
foraphane content in brassicaseae during drying is scarce. Jin et al.
(2014) reported a dynamic optimization strategy to determine
the best moisture-temperature trajectories that maximize reten-
tion of glucosinolates and vitamin C in broccoli stalks with mini-
mum energy consumption in convective drying. They developed a
model that considers degradation kinetics of glucosinolates and
vitamin C, both following first order reaction models. The authors
showed that this strategy allowed keeping 55% vitamin C and
improved energy efficiency in 65%. However, this is a theoretical
study that lacks experimental verification. Tanongkankit et al.

http://dx.doi.org/10.1016/j.jfoodeng.2016.04.007
0260-8774/© 2016 Elsevier Ltd. All rights reserved.
28 A. Mahn et al. / Journal of Food Engineering 186 (2016) 27e33
Fig. 1. Representation of (A) the enzymatic hydrolysis of glucoraphanin through the action of myrosinase (adapted from Tanongkankit et al. (2011)), and (B) pathway for degra-
dation of sulforaphane to thiourea proposed by Jin et al. (1999).
(2011) studied the evolution of sulforaphane content in cabbage
leaves during hot air drying. They described this behavior through a
semi-empirical model that considers heat transfer and synthesis-
degradation kinetics. The authors solved the model by adjusting
it to experimental data obtained from drying experiences per-
formed in a tunnel dryer at 40, 50, 60 and 70
C, obtaining good
agreement between experimental data and the model (r > 0.90).
Besides, the authors reported that sulforaphane degradation takes
place when the substrate temperature exceeds 40
C. Based on the
results of Tanongkankit et al. (2011), Lekcharoenkul et al. (2014)
proposed a hybrid drying technique with temperature stepwise
changes to enhance sulforaphane content in cabbage leaves. They
found a similar behavior to that reported by Tanongkankit et al.
(2011), but with a higher maximum sulforaphane content.
Currently, no reports about sulforaphane evolution during drying of
broccoli are available.
In this work we investigate the effect of temperature in
convective drying of sulforaphane-enriched broccoli on the evo-
lution of sulforaphane content. The results obtained here could be
used for designing an industrial process in order to produce a
dehydrated functional ingredient naturally enriched in
sulforaphane.
2. Materials and methods
2.1. Plant material
Broccoli (Brassica oleracea var italica cv. Avenger) heads (three
days from harvesting) were purchased at the local market (San-
tiago, Chile) from a single supplier. Broccoli florets were subjected
to blanching followed by incubation in previously optimized con-
ditions to maximize the sulforaphane content (Pe rez et al., 2014).
This optimized process ensures that myrosin cells are broken so
that myrosinase can enter in contact with glucoraphanin, the epi-
thiospecifier protein is inactive and myrosinase remains active.
Broccoli heads were washed and cut into 5-cm length and
0.7e0.9 cm width (stem). Broccoli pieces were immersed in
deionized water in a thermostatic bath (Stuart, United Kingdom,
Great Britain) at 57
C for 13 min. After blanching, broccoli pieces
were immediately put in an ice-water bath. After that, broccoli was
 A. Mahn et al. / Journal of Food Engineering 186 (2016) 27e33
crushed and then subjected to incubation by putting the broccoli
pulp in empty, hermetically closed flasks and immersed in a ther-
mostatic water bath (Trilab, Mexico City, Mexico) at 40
C during
rez et al., 2016).
1 h (Pe
2.2. Drying
Drying experiments were performed in a laboratory tray dryer
(20 cm  20 cm cross-section). Pre-processed sample (750 g) was
spread on two squared trays (10 cm wide  10 cm length and 0.5 cm
thickness); one tray was used to take broccoli samples for analysis
purposes and the other tray was used to build the drying curve.
Samples (3e5 g) were taken out every 15 min. Drying experiments
were conducted at 60, 70 and 80
C, with constant air flow rate
equal to 2 m/s and relative air humidity in the range of 40e60%,
until the sample reached constant moisture content. The temper-
ature of the sample was measured using a type-K thermocouple
inserted in the broccoli bed. The moisture content of the sample
was determined according to AOAC 920.151 (AOAC, 1990) by drying
at 80
C in a vacuum oven at 0.5 atm (Cole & Palmer, model 60061
Holdpack, Illinois) until constant mass was attained, and the
average of three measurements was considered for calculation.
Moisture ratio (X/X0) was used to describe the drying behavior of
the substrate, and was defined as the ratio between moisture
content of the sample at any stage of the drying process (X) and
moisture content of the substrate at t ¼ 0 (X0) in dry base. Equi-
librium moisture ratio (Xeq/X0) was defined as equilibrium mois-
ture content (Xeq) divided by moisture content at t ¼ 0 in dry base.
2.3. Sulforaphane content
Samples were homogenized in a mortar until obtaining a ho-
mogeneous pulp. One gram of the sample was extracted two times
with 10 mL methylene chloride (J.T. Baker, Center Valley, PA, USA)
combined with 0.5 g anhydrous sodium sulfate (SigmaeAldrich,
Schnelldorf, Germany). The methylene chloride fraction was dried
at 30
C under vacuum. The residue was dissolved in 1 mL aceto-
nitrile (Merck, Darmstadt, Germany) and filtered through a 0.22 mm
membrane filter. Sulforaphane content was assessed by in a HPLC-
DAD (Shimadzu, Tokyo, Japan) using a C18 column (5 mm particle
size, 250  4.6 mm) (Agilent Technologies, Santa Clara, CA, USA)
(Liang et al., 2006). The solvent consisted of 20% acetonitrile in
water; this solution was then changed linearly over 10 min to 60%
acetonitrile and maintained at 100% acetonitrile for 5 min to purge
the column. The temperature of the column oven was set at 30
C,
the flow rate was 1 mL min1, and 20-mL aliquots were injected into
the column. Sulforaphane was detected by absorbance at 254 nm
and expressed in mg/100 g dry weight (DW). All determinations
were made in triplicate.
2.4. Glucoraphanin content
The glucoraphanin standard was analyzed by LC-MS (Agilent
Technologies, Waldbronn, Germany) aiming to determine the
retention time. Ionization was performed by ESI-ion trap electro-
spray- IT Esquire 4000 (Bruker Dalronik GmbH, Germany) assisted
with ultra-pure nitrogen at a temperature of 325
C, pressure and
flow rate were 30.0 psi with a flow of 7 L min1. For quantitative
analysis, samples were analyzed using the same column and con-
ditions in a HPLC-DAD equipment (Shimadzu, Tokyo, Japan) pro-
vided with a C18 column (5-mm particle size, 250 mm  4.6 mm)
(Agilent Technologies, Santa Clara, CA, U.S.A.). Glucoraphanin peak
was detected at a retention time of 6.2 min and the following
fragmentation pattern 371.9; 258.8; 420.8; 193.6 with a precursor
of 436.3 m/z (Cataldi et al., 2010).
29
2.5. Statistical analyses
The experiments were conducted in complete random. Data was
analyzed by ANOVA at 95% confidence. Significant differences be-
tween samples were identified by the Least Significant Difference
(LSD) multiple range test. Statistical analyses were performed with
Statgraphics Centurion XVI.II (Statistical Graphics Corp., Princeton,
NJ, USA).
3. Theory and model
3.1. Theory
Sulforaphane comes from the hydrolysis of glucoraphanin
catalyzed by the enzyme myrosinase, and the mechanism com-
prises an intermediate step to form glucose and an unstable in-
termediate that may be converted to different compounds, such as
thiocyanates, nitriles and the isothiocyanate sulforaphane (Bones
and Rossiter, 2006). Fig. 1A shows a representation of the enzy-
matic reaction. The resulting end compound depends on the
chemical conditions in the reaction medium. In this work, broccoli
was processed in order to reach the conditions that favor sulfo-
raphane formation in spite of the other compounds. The pre-
processing conditions were established to almost complete con-
version (94%) of glucoraphanin into sulforaphane, so that broccoli
to be dried contained a very high sulforaphane content (Pe rez
et al., 2014, 2016). However, some glucoraphanin remains avail-
able for reaction in the vegetable tissue. Then, sulforaphane for-
mation can occur in the first stage of the drying process.
Sulforaphane is thermo labile, then it is expected that when the
broccoli temperature during drying exceeds 40
C (Tanongkankit
et al., 2011), degradation of the compound will occur, which will
be reflected as a reduction of the sulforaphane content in broccoli.
Accordingly, the model to represent sulforaphane evolution during
drying must consider the formation and degradation of the
compound.
The thermal degradation of sulforaphane has been studied by Jin
et al. (1999). The authors determined that in aqueous solution the
main non-volatile product of sulforaphane degradation was N,N-di-
(methylsulfinyl)butyl thiourea. Also, the authors proposed a
mechanism for the formation of this compound from sulforaphane.
This mechanism is given in Fig. 1B.
3.2. Model
The model proposed here to describe the evolution of sulfo-
raphane content in a drying process considers that the formation
and degradation of sulforaphane can be represented by two irre-
versible consecutive reactions, as shown in Eq. (1). The model
considers simplifications of the real mechanisms that underlie both
formation and degradation of sulforaphane. We reduced the steps
for sulforaphane formation to one step, and sulforaphane degra-
dation was also represented by only one step. According to litera-
ture (Jin et al., 2014; Wu et al., 2013, 2014), we considered first
order kinetics in both cases. The simplified reaction is given in Eq.
(1). Here, k1 and k2 are the rate constants (s1). Eqs. (2) and (3)
describe the kinetic model, where CA, CB and CC are the concen-
tration of glucoraphanin, sulforaphane and thiourea, respectively.
k1
glucoraphanin þ H2 Oƒƒƒƒ!sulforaphane þ D
k2
 glucose ƒƒƒƒ!N; N  di  ðmethylsulfinylÞbutyl thiourea
(1)
30 A. Mahn et al. / Journal of Food Engineering 186 (2016) 27e33

80
C, the equilibrium moisture ratio (Xeq/X0 ¼ 0.01) was reached

after 4.50 and 4.25 h, respectively. At 60
C, the equilibrium
moisture content was somewhat higher (Xeq/X0 ¼ 0.02), and it was
reached after 6.50 h. This difference between the behavior at 60
C
and the other runs may be due to a higher humidity content in the
drying air, in addition to the lower drying air temperature.

4.2. Effect of drying air temperature on sulforaphane content

Fig. 3 shows the evolution of sulforaphane content and the

temperature profile inside the broccoli bed during the runs per-
formed at different drying air temperatures. The initial temperature
of broccoli was approximately 24
C and it raised as the solid was
heated by the drying air during the process. Temperature profiles
were somewhat irregular, they did not follow a clear asymptotic
shape, probably because of the characteristics of the substrate. The
substrate was a heterogeneous bed of chopped broccoli pieces, with
Fig. 2. Drying curves of pre-processed broccoli at different drying air temperatures.
size ranging from 0.5 to 1.0 cm length and 0.3e0.7 cm width. The
oscillations seen in the temperature profiles may obey to shrinkage
of the substrate. Since the thermocouple was inserted in the
dC broccoli bed and was kept in the same place during the whole run,
 A ¼ k1 ,CA (2) as drying proceeds some pores were formed, and it is possible that
dt
the thermocouple was located inside one of them. Table 2 shows
dCB the maximum sulforaphane content achieved in each drying run,
¼ k1 ,CA  k2 ,CB (3) and the corresponding drying time and moisture ratio.
dt
In drying at 60
C (Fig. 3A) the maximum sulforaphane content
dCC corresponds to the initial content (at t ¼ 0 h). At the beginning of
¼ k2 ,CB (4) drying sulforaphane content diminished (at t ¼ 15 min), but after
dt
that there was an increase until 1 h of drying, when temperature in
Eqs. (2)e(4) were solved resulting in the solution given by Eq. the broccoli bed was 36
C. This can be interpreted as sulforaphane
(5), where t is the drying time, CA0 and CB0 are the initial (at t ¼ 0) formation in this period, since there was glucoraphanin available
concentration of glucoraphanin and sulforaphane, respectively. for reaction (glucoraphanin content at t ¼ 0 h was 22.8 ± 2.6 mg/
100 g DW) and myrosinase remains active at temperatures below
CB C k1  k1 t 
70
C (Howard et al., 1997). Between 1 and 2 h of drying there was a
¼ ek2 t þ A0 e  ek2 t (5)
CB0 CB0 k2  k1 decrease in sulforaphane content, probably because sulforaphane
Based on results reported by Tanongkankit et al. (2011), mois- degradation was faster than the hydrolysis of glucoraphanin. At
ture content does not affect sulforaphane content during drying; t ¼ 2.5 h, a new increase in sulforaphane content was observed. At
only the substrate temperature is the dominant variable that de- this point, glucoraphanin content was 27.7 ± 3.5 mg/100 g DW,
termines the evolution of sulforaphane content during the process. higher than glucoraphanin content at t ¼ 0 h. Since there is no
Hence, the rate constants were expressed through the Arrhenius glucoraphanin synthesis, the only possible interpretation is that the
equation (Eq. (6)), where ki (s1) is the rate constant, k0i is the availability of glucoraphanin was higher, probably due to shrinkage
frequency factor (s1), Eai is the activation energy (KJ/mol) and T is which in turn produces important tissue damage during drying,
the absolute temperature (K). thus facilitating the release of glucoraphanin. Then, sulforaphane
formation was faster than sulforaphane degradation. Despite
Eai
broccoli temperature was below 40
C, the increase of sulforaphane
ki ¼ ki0 ,e R,T (6)
content apparently stopped at t ¼ 3.5 h (broccoli temperature of
38
C); probably because of the low glucoraphanin content
4. Results and discussion (8.7 ± 1.3 mg/100 g DW) and low moisture content. When broccoli
temperature reached 42
C (4 h of drying), a drastic reduction of
4.1. Drying of pre-processed broccoli sulforaphane content could be observed, from 86.8 ± 9.1 mg/100 g
DW to 18.6 ± 3.0 mg/100 g DW by the end of drying, representing a
The drying curves obtained from convective drying of broccoli at 78% reduction in the last two hours of drying. This agrees with the
60, 70 and 80
C are given in Fig. 2. As expected, the drying rate was results reported by Tanongkankit et al. (2011), who reported that
higher in the runs executed at higher temperature. Table 1 shows thermal degradation of sulforaphane in hot air drying begins at
the equilibrium moisture ratio (Xeq/X0) and the time needed to 40
C. In drying at 70
C (Fig. 3B) there was an increase of 18% of
reach it for the different runs. In the runs performed at 70 and sulforaphane content until 0.25 h of drying, coinciding with a

Table 1
Drying time to achieve moisture ratio (X/X0) of 0.1 and equilibrium moisture ratio of broccoli (as average ± standard deviation) at different
temperatures. Different superscripts indicate that values are significantly different (p < 0.05).

Drying temperature (
C) Drying time (h) Equilibrium moisture ratio (Xeq/X0)

60 6.50 0.02 ± 0.002a

70 4.50 0.01 ± 0.006a
80 4.25 0.01 ± 0.005a
 A. Mahn et al. / Journal of Food Engineering 186 (2016) 27e33 31

broccoli temperature of 39
C. After that, sulforaphane diminished

Fig. 3. Evolution of sulforaphane content and temperature profile in the broccoli bed
during drying at (A) 60
C, (b) 70
C and (C) 80
C. C0 is the sulforaphane content at
t ¼ 0.
from 148.4 ± 9.3 mg/100 g DW to 118.9 ± 13.5 mg/100 g DW until
broccoli temperature was 40
C, and kept relatively constant until
broccoli temperature was 48
C. At this temperature sulforaphane
content decreased drastically to reach 6.5 ± 0.2 mg/100 g DW until
the end of drying (t ¼ 5 h). Then it can be presumed that sulfo-
raphane degradation is apparently accelerated when temperature
exceeds 48
C, giving some insight about the mechanism of thermal
degradation of sulforaphane. In the period of constant sulforaphane
content, glucoraphanin content diminished slightly, from
22.3 ± 0.7 mg/100 g DW to 20.1 ± 1.0 mg/100 g DW, then there was
a slow glucoraphanin consumption. Since sulforaphane content
remained constant in this period, it can be supposed that glucor-
aphanin consumption rate equals the rate of sulforaphane degra-
dation. In drying at 80
C (Fig. 3C) sulforaphane content increased
very slightly (3%) in the first 15 min, when broccoli reached 48
C.
This increase was not significant, then it cannot be supposed that
sulforaphane formation took place, especially considering that
sulforaphane degradation proceeds above approximately 40
C.
After that sulforaphane content (106.0 ± 6.2 mg/100 g DW at 0.25 h
of drying) decreased slowly until broccoli temperature was 55
C,
reaching 97.3 ± 4.6 mg/100 g DW after 2 h of drying. From that
moment onwards sulforaphane decrease was considerably more
drastic, reaching 0.88 ± 0.06 mg/100 g DW after 5 h of drying.
In summary, sulforaphane formation apparently occurred in
drying with an air temperature of 60 and 70
C, until broccoli
temperature reached approximately 40
C. At this temperature
sulforaphane degradation began. At higher drying air temperature
(80
C) no significant formation of sulforaphane could be observed,
probably because of a very low myrosinase activity. A decrease of
sulforaphane content could be seen along drying. In all the drying
runs there was a change in the slope of the profile of sulforaphane
degradation when broccoli temperature reached approximately
50
C. This can be interpreted as a degradation reaction that differs
from which occurs at temperature between 40 and 50
C. Jin et al.
(1999) studied thermal degradation of sulforaphane in aqueous
solutions, and found that higher temperatures accelerate the
degradation rate of sulforaphane. The authors proposed a mecha-
nism of sulforaphane degradation that considers the formation of
different volatile compounds such as S-methyl methylthiosulfinate,
methyl (methylthio)methyl disulfide and 4-isothiocyanato-1-
(methylthio)-1-butene. They also reported that the main non-
volatile degradation compound was N,N-di-(methylsulfinyl)butyl
thiourea. Then, the different degradation rates of sulforaphane at
40 and 50
C can also be attributed to a different mechanism for the
thermal degradation of sulforaphane, or to degradation to different
products. On the other hand, if there was still glucoraphanin
available, the decrease in sulforaphane content can be attributed to
the fact that the rate of enzymatic reaction was lower than the
degradation rate at temperatures below 48
C. The change in the
decreasing slope at that temperature could then be attributed to
low glucoraphanin concentration and low moisture content, thus
preventing sulforaphane formation. The change in the degradation

Table 2
Maximum content of sulforaphane (average ± standard deviation) achieved during drying at different temperatures and the corresponding moisture ratio and drying time. DW
means dry weight. Different superscripts indicate that values are significantly different (p < 0.05).

Drying temperature (
C) Drying time (h) Moisture ratio (X/X0) Sulforaphane content (mg/100 g DW) Increment regarding pre-processed (%)

Pre-processed Maximum in drying

60 0.00 1.00 90.6 ± 2.8 90.6 ± 2.8 0.0 ± 0.0a

70 0.25 0.93 125.5 ± 6.2 148.9 ± 9.1 18.9 ± 9.5b
80 0.25 0.62 103.7 ± 0.8 106.3 ± 5.3 2.25 ± 5.3a
32 A. Mahn et al. / Journal of Food Engineering 186 (2016) 27e33

Table 3
Sulforaphane content in broccoli at X/X0 ¼ 0.1 and retention percentage regarding pre-processed broccoli. DW means dry weight. Different superscripts indicate that values are
significantly different (p < 0.05).

Drying temperature (
C) Drying time (h) Sulforaphane content (mg/100 g DW) Retention
(%)

60 5.5 18.1 ± 1.6 19.9 ± 1.2a

70 4.0 67.6 ± 0.6 54.0 ± 3.7c
80 4.0 43.0 ± 0.8 41.5 ± 5.3b

rate of sulforaphane at 50
C was not detected in the runs at drying

air temperature of 60
C because the temperature inside the
broccoli bed did not exceed 45
C.
Sulforaphane content increased during drying, reaching a
maximum that varied depending on drying air temperature. The
maximum sulforaphane content was 148.9 mg per 100 g dry mat-
ter, obtained after 15 min drying at air temperature of 70
C. This
represents an 18.9% increase with respect to the initial sulforaph-
ane content (at t ¼ 0).
Table 3 shows the sulforaphane content when moisture ratio (X/
X0) was 0.1, for the different drying air temperatures. Drying at
70
C gave the highest value (67.6 mg per 100 g dry matter), rep-
resenting a 54% retention with respect to the initial sulforaphane
content. This value is 4-fold higher than that found in fresh broccoli.
Additionally, this value is much higher than those reported by
Lekcharoenkul et al. (2014) (12 mg per 100 g dry matter) and by
Tanongkankit et al. (2011) (2.4 mg per 100 g dry matter), probably
because the sulforaphane content in fresh cabbage is much lower
than in fresh broccoli. Besides, the authors did not perform any
procedure to maximize the initial sulforaphane content before
drying, so that most glucoraphanin was converted into sulforaph-
ane. Our results seem promising in the field of functional foods,
since we obtained dehydrated broccoli that, at a moisture ratio that
ensures microbial and chemical stability (X/X0 ¼ 0.1), presents the
highest sulforaphane content that has been attained so far in a
dehydrated product.

4.3. Modeling of sulforaphane content evolution during drying

The kinetic constants are function of broccoli temperature, and

broccoli temperature is in turn function of time, as shown by Eq.
(7). Then, the model shown in Eq. (5) was solved by estimating the
rate constants using Eq. (6) from the temperature data for each
experimental point, thus obtaining one value of rate constant for
each point. Besides, the frequency factor and the activation energy
were calculated. Fig. 4 shows the model adjustment to the exper-
imental data obtained in the different drying runs. Table 4 presents
the frequency factors, activation energies and correlation coeffi-
cient of the model for each drying air temperature.

Fig. 4. Comparison between predicted and experimental evolutions of sulforaphane
during drying at drying air temperature of (A) 60
C, (B) 70
C and (C) 80
C. C0 is the
sulforaphane content at t ¼ 0.
ki ¼ f ðTðtÞÞ (7)
The model gave a good agreement with experimental data,
resulting in correlation coefficients equal or higher than 0.9. In
drying at 80
C (Fig. 4C) the model gave the best fit, with r ¼ 0.98,
representing the sulforaphane evolution during the complete
process in an acceptable way, probably because the increase of
sulforaphane content at the beginning of the process was very
slight and occurred in a short period of time. In drying at 70
C
(Fig. 4B) the correlation coefficient was also 0.98, but despite the
model adjusted very well in the intermediate and final periods of
drying, it was not able to represent adequately the initial behavior
of sulforaphane content. This can be attributed to the fact that the
model cannot explain a sharp increase of sulforaphane content, as it
happened in this run. The poorest fit was observed for drying at
 A. Mahn et al. / Journal of Food Engineering 186 (2016) 27e33
Table 4
Arrhenius parameters for estimating the kinetic constants of eqn. (5) as a function of temperature.
Drying temperature (
C) k01 (s1)
60 7.27  103
70 7.84  103
80 9.67  103
60
C (Fig. 4A), probably because in this condition the evolution of
sulforaphane showed two periods where sulforaphane content
increased, and the model considers only one.
The activation energy was higher for drying at 60
C. In drying at
70 and 80
C, activation energy had similar values. The frequency
factors showed the same behavior. The frequency factors obtained
in the different drying conditions have the same order of magni-
tude than those reported by Tanongkankit et al. (2011) in the drying
of cabbage leaves at 60
C. The activation energies obtained in this
work are two orders of magnitude highest than those reported by
Tanongkankit et al. (2011).
5. Conclusion
In all the drying runs there was a period where sulforaphane
was simultaneously synthesized and degraded, resulting in two
decreasing stages with different slopes in the sulforaphane content
profile. This period started at broccoli temperature of 40
C, when
sulforaphane degradation began, but simultaneously there was
sulforaphane formation until glucoraphanin and moisture content
allowed it. This happened approximately at broccoli temperature of
50
C, depending on the drying air temperature, and was reflected
as a change in the slope of sulforaphane content profile from this
point onwards.
Sulforaphane content increased at the beginning of drying due
to higher availability of glucoraphanin and temperature close to the
optimum of myrosinase. Afterwards it decreased until the end of
the process. The maximum sulforaphane content was obtained
after 15 min drying at drying air temperature of 70
C (148.9 mg per
100 g dry matter). This represents an 18.9% increase with respect to
the initial sulforaphane content, and an 8.5-fold increase with
respect to the fresh vegetable. The highest sulforaphane retention
at X/X0 ¼ 0.1 was 54%, obtained with drying air temperature of
70
C. This value is 4-fold higher than that found in fresh broccoli,
and the highest reported so far in any dehydrated food.
The model proposed here gave a relatively good agreement with
the experimental data, resulting in correlation coefficients equal or
higher than 0.9. The best fit was obtained with drying air temper-
ature of 80
C (r ¼ 0.98), representing the sulforaphane evolution
during the complete process in an acceptable way. The model
should be improved in order to attain a better fit.
33
Ea1 (KJ/mol) k02 (s1) Ea2 (KJ/mol) r
112.3 1.94  107 70.4 0.90
48.6 2.83  105 57.8 0.98
49.2 2.23  105 58 0.98
Acknowledgement
Fondecyt Grant 1130384 supported this research.
References
Association of Official Analytical Chemists (AOAC), 1990. Official Methods of Anal-
ysis. The Association of Official Analytical Chemists, Arlington, VA.
Bones, A.M., Rossiter, J.T., 2006. The enzymic and chemically induced decomposi-
tion of glucosinolates. Phytochemistry 67, 1053e1067.
Cataldi, T.R., Lelario, F., Orlando, D., Bufo, S.A., 2010. Collision-induced dissociation
of the aþ 2 isotope ion facilitates glucosinolates structure elucidation by elec-
trospray ionization-tandem mass spectrometry with a linear quadrupole ion
trap. Anal. Chem. 82, 5686e5696.
Elbarbry, F., Elrody, N., 2011. Potential health benefits of sulforaphane: a review of
the experimental, clinical and epidemiological evidences and underlying
mechanisms. J. Med. Plants Res. 5, 473e484.
Gu, Z.X., Guo, Q.H., Gu, Y.J., 2012. Factors influencing glucoraphanin and sulfo-
raphane formation in Brassica plants: a review. J. Integr. Agric. 11, 1804e1816.
Howard, L.A., Jeffery, E.H., Wallig, M.A., Klein, B.P., 1997. Retention of phytochemicals
in fresh and processed broccoli. J. Food Sci. 62, 1098e1104.
Jin, Y., Wang, M., Rosen, R.T., Ho, C., 1999. Thermal degradation of sulforaphane in
aqueous solution. J. Agric. Food Chem. 47, 3121e3123.
Jin, X., van der Smar, R.G.M., van Straten, G., Boom, R.M., van Boxtel, A.J.B., 2014.
Energy efficient drying strategies to retain nutritional components in broccoli
(Brassica oleracea var. italica). J. Food Eng. 123, 172e178.
Lekcharoenkul, P., Tanongkankit, Y., Chiwchan, N., Devahastin, S., 2014. Enhance-
ment of sulforaphane content in cabbage outer leaves using hybrid drying
technique and stepwise change of drying temperature. J. Food Eng. 122, 56e61.
Liang, H., Yuan, Q.P., Dong, H.R., Liu, Y.M., 2006. Determination of sulforaphane in
broccoli and cabbage by high-performance liquid chromatography. J. Food
Compos. Analysis 19, 473e476.
Matusheski, N.V., Juvik, J.A., Jeffery, E.H., 2004. Heating decreases epithiospecifier
protein activity and increases sulforaphane formation in broccoli. Phytochem-
istry 65, 1273e1281.
Perez, C., Barrientos, H., Mahn, A., 2016. Optimization of an Incubation Step to
Maximize Sulforaphane Synthesis in Pre-Processed Broccoli. Unpublished
results.
Perez, C., Barrientos, H., Roman, J., Mahn, A., 2014. Optimization of a blanching step
to maximize sulforaphane synthesis in broccoli florets. Food Chem. 145,
264e271.
Shen, L., Su, G., Wang, X., Du, Q., Wang, K., 2010. Endogenous and exogenous
enzymolysis of vegetable-sourced glucosinolates and influencing factors. Food
Chem. 119, 987e994.
Tanongkankit, Y., Chiewchan, N., Devahastin, S., 2011. Evolution of anticarcinogenic
substance in dietary fibre powder from cabbage outer leaves during drying.
Food Chem. 127, 67e73.
Wu, Y., Mao, J., Mei, L., Liu, S., 2013. Kinetic studies of the thermal degradation of
sulforaphane and its hydroxypropyl-b-cyclodextrin inclusion complex. Food
Res. Int. 53, 529e533.
Wu, Y., Mao, J., You, Y., Liu, S., 2014. Study on degradation kinetics of sulforaphane in
broccoli extract. Food Chem. 155, 235e239.