Accepted Manuscript Thermal degradation of grape marc polyphenols Katalin Sólyom, Ruth Solá, María José Cocero, Rafael B. Mato PII: DOI: Reference:
S0308-8146(14)00406-3 http://dx.doi.org/10.1016/j.foodchem.2014.03.021 FOCH 15550
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
18 November 2013 30 January 2014 5 March 2014
Please cite this article as: Sólyom, K., Solá, R., Cocero, M.J., Mato, R.B., Thermal degradation of grape marc polyphenols, Food Chemistry (2014), doi: http://dx.doi.org/10.1016/j.foodchem.2014.03.021
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
2
Thermal degradation of grape marc polyphenols
3
Katalin SÓLYOM, Ruth SOLÁ, María José COCERO, Rafael B. MATO*
4
5
University of Valladolid, Industrial Engineering School, Department of Chemical
6
Engineering and Environmental Technology; High Pressure Processes Group; C/Dr.
7
Mergelina, s/n, 47011-Valladolid, Spain
8 9
*Corresponding author:
10
Rafael B. Mato
11
email: [email protected]
12
telephone: +34 983 423 177
13
1
14
ABSTRACT
15
Bioactive compounds of wine making by-products are of interest in food, cosmetics and
16
pharmaceutical industries. Extraction of antioxidants under mild conditions is time-
17
consuming, giving ground to the development of intensification processes where the
18
operation at high temperature may deteriorate extract quality. This study examined thermal
19
degradation of grape marc and its filtered extract (80, 100 and 150 °C). The decrease in
20
anthocyanin content was modelled under non-isothermal conditions by first order kinetics,
21
using the Arrhenius equation. Simulated degradation under isothermal heating showed that
22
the grape marc is more sensitive by one order of magnitude to heat than the filtered extract.
23
This tendency was also confirmed by analyses of the total phenolic content and antioxidant
24
activity. It is suggested that an optimal combination of temperature, treatment time and
25
also raw material environment could be found in process intensification.
26
Keywords:
27
Grape pomace, pH differential method, ORAC, degradation kinetics, non-isothermal
28
heating, anthocyanin
29 30
2
31
32
1. Introduction
33
Worldwide wine production generates a large amount of semi-solid residue (grape marc),
34
with high antioxidants content. Flavonoids, anthocyanins and stilbene derivatives are
35
bioactive compounds present in this residue, and they are of interest for food, cosmetics
36
and pharmaceutical industries due to their antioxidant, antimicrobial, antiviral or
37
anticarcinogenic features (Nassiri Asl & Hosseinzadech, 2009). The elimination of these
38
compounds is also desirable before using the grape pomace as fertilizer, in order to avoid
39
inhibition problems (Negro, Tommasi, & Miceli, 2003, Pinelo, Arnus, & Meyer, 2006).
40
Mild extraction techniques (maceration and stirred extraction) are time consuming and
41
usually require an excessive amount of solvent, leading to an inefficient and costly
42
recovery process (Spigno, & De Faveri, 2007, Spigno, Tramelli, & De Faveri, 2007, Lafka,
43
Sinanoglu, & Lazos, 2007, Amendola, Spigno, & De Faveri, 2010).
3
44
Novel techniques and pre-treatments in order to enhance the extraction kinetics of valuable
45
components include high temperature during processing. Extraction at high pressure and
46
temperature (110 °C) for 60 minutes was applied and compared to microwave assisted
47
extraction for the recovery of bioactive compounds from grape waste (Casazza,
48
Aliakbarian, Mantegna, Cravotto, & Perego, 2010). Peralbo-Molina and co-workers (2012)
49
compared superheated extraction at 180 °C for 60 min with microwave and ultrasound
50
assisted extraction in order to enhance the extract quality and process efficiency, when
51
compared to conventional maceration. The effect of temperature in microwave processing
52
can also alter the interaction between the material and the electromagnetic field (Solyom et
53
al, 2013). Drying pre-treatment was also applied for grape stalks at temperatures between
54
40 and 115 °C, and improved extraction kinetics were described by a diffusion model
55
(Garcia-Perez, García-Alvarado, Carcel, & Mulet, 2010).
56
Although these process intensification methods increased the quantity or the quality of the
57
extract, their destructive effect on the antioxidant content may be masked by a larger
58
beneficial effect on the extraction stage efficiency. The physical complexity of the natural
59
raw material and the variety of experimental conditions may modify interactions between
60
active compounds, thus changing the final product quality. The effect of thermal
61
treatments on liquid extracts has been previously reported (Wang, & Xu, 2007, Cruz,
62
COnde, Domínguez, & Parajo, 2007, Cisse, Vaillant, Acosta, Dhuique-Mayer, & Dornier,
63
2009), and first order kinetic degradation was found for anthocyanins, while for other
64
antioxidants this tendency was not so clear. Drying of raw material deteriorated the
65
antioxidant quality of grape peels when treated above 60°C (Larrauri, Ruperez, & Saura-
66
Calixto, 1997).
4
67
Heating of grape seeds prior to extraction released more polyphenols, verified by
68
antioxidant activity and HPLC analysis (Kim et al, 2006). The anthocyanin content of dry
69
bilberry extracts was reduced after 30 min, 30% in dry bilberry extracts at 100 °C, while
70
antioxidant activity increased with increasing temperature (Yue, & Xu, 2008). It is clear
71
that the raw materials variability and the different heating conditions are hardly
72
comparable in terms of temperature and time, solid content, or the presence of extraction
73
solvent. Consequently, the aim of this study was to compare the effects of heating on the
74
semi-solid grape marc waste and on its liquid extract. To avoid excessively high operating
75
pressures (above 10 bar), the temperature range was restricted to between 80 and 150ºC,
76
which corresponds to the range in boiling temperatures of 50%vol mixtures of ethanol-
77
water mixtures between 1 and 10 bar. To this end heating was performed at 80, 100 and
78
150 °C for up to 4 hours in order to study degradation kinetics in terms of total phenol
79
content, antioxidant activity and anthocyanin content. Kinetic parameters of anthocyanin
80
degradation were obtained by regression of non-isothermal heating experimental values to
81
a first order kinetic model. These parameters were later used to estimate and compare
82
degradation in isothermal heating of semi-solid grape marc and its liquid extract.
83
2. Materials and methods
84
2.1. Grape marc
85
A mixture of red grape seeds and skins (grape marc) of the Tempranillo grape variety from
86
the region of Toro (Spain) was used in the experiments. This residue was generated in the
87
wine making process, after long maceration with a post-fermentative phase. The grape
88
marc was distributed in 100 g packages and stored at -18 °C until defrosting was
89
performed at 4 °C overnight.
90
2.2. Heat treatment of grape marc
5
91
In these experiments, the raw material was first exposed to heat and then further extracted
92
with solvent. Grape marc (10 0.001 g) was filled in three identical stainless steel tubes
93
(length: 220 mm, inner diameter: 12.3 mm) and closed with stainless steel stoppers to hold
94
up the arising pressure during the experiment. The tubes were placed in an oven, set at 80,
95
100 and 150 °C, and were removed consecutively after 60, 120 and 240 minutes for
96
analysis. The series at 150 °C was repeated with sampling times at 20, 40 and 60 minutes
97
to better follow the fast degradation kinetics. Tubes were immediately cooled down in an
98
ice bath and further stored at 4 °C until extraction, which was performed within 4 hours.
99
Also, one fraction of original grape marc was kept for extraction, without previous heat
100
treatment, to use as a control sample in comparison to the heated material.
101
After the heat treatment, grape marc was removed from the tubes and placed in glass
102
vessels for extraction with 50 ml of 50% volumetric mixture of ethanol (96%) and
103
acidified water (pH=1; H2SO4). Extraction was performed for 1 hour at 40 °C, under
104
continuous stirring on an orbital shaker. Extraction was stopped by separating liquid and
105
solid phases in centrifuge (4000 rpm, 5 min). The supernatants were filtered (PET
106
membrane, 0.2 µm) for further analyses.
107
2.3. Heat treatment of filtered extract
6
108
In the second series of experiments, extraction was first performed from the grape marc,
109
and the obtained extract was further exposed to heat treatment. A 40 g sample of raw
110
material was extracted in 200 ml 50% volumetric mixture of ethanol (96%) and acidified
111
water (pH=1; H2SO4) at 40 °C for 24 hours in the dark. The aim of this experiment was to
112
study the effect of temperature on the stability of polyphenols in the absence of a solid
113
phase. A longer extraction time (24 h) was chosen to obtain a high extraction yield, and so
114
that most of the polyphenols originally present in the solid sample could be extracted to a
115
liquid before heat treatment was performed. In the case of the extraction of the preheated
116
grape marc, a 1 hour extraction time was sufficient to evaluate the differences after the
117
heat treatments. After the extraction, a centrifuge was used for phase separation (4000
118
rpm, 5 min) and the supernatant was filtered through a 0.45 µm borosilicate filter
119
(Millipore). The filtered extract obtained was divided into five stainless steel tubes (length:
120
160 mm, inner diameter: 3.9 mm). In this instance thinner tubes were used to achieve
121
faster heating of the filtered extract. In the case of the grape marc, a larger inner tube
122
diameter was necessary to fit the skin and seed particles into the tube. Closed tubes were
123
placed in the oven at 80, 100 and 150 °C, and removed one by one after 20, 40, 60, 120
124
and 240 minutes. After immediate cooling in an ice bath, the extracts were filtered (PET
125
membrane, 0.2 µm) and kept refrigerated for further analysis. A fraction of the original
126
unheated filtered extract was also kept as control sample for comparison.
127
2.4. Temperature evolution in heat treatments
128
The thermal history of the experimental setup, was followed by temperature measurement
129
inside separate tubes with a K-type thermocouple. Temperature was recorded until the set
130
value was reached.
131
2.5. Analysis
7
132
The moisture content of the grape marc (≈ 85% w/w) was determined in triplicate for
133
every grape marc batch used by recording the weight difference after drying at 105ºC for
134
24 hours, until a constant weight was observed.
135
Total phenol content (TPC) was quantified using the Folin-Ciocalteu method, as Gallic
136
acid equivalents per gram of dry material [mg GA/g DM] (Singleton, Orthofer, &
137
Lamuela-Raventos 1999). A 40 μl volume of filtered sample was diluted with distilled
138
water (3 ml) and mixed with 200 μl of Folin-Ciocalteu reagent. After 10 minutes, saturated
139
Na2CO3 was added to the solutions and the samples were incubated at 40ºC for 30 min.
140
Absorbance was measured at 765 nm (UV 2550 Shimadzu UV/VIS spectrometer) and
141
standard dilutions (50-900 ppm) of Gallic acid were used for calibration.
142
Total monomeric anthocyanin pigment content was determined by the pH differential
143
method (AOAC 2005.02.). Results are expressed on the basis of cyanidin-3-glucoside [mg
144
cyanidin-3-glucoside/ g DM]
145
Antioxidant activity was measured in terms of Oxygen Radical Absorbance Capacity
146
(ORAC) based on the Application note 148 (BMG-Labtech). A 96 well-plate was used in a
147
Fluostar Optima (BMG-Labtech) to determine the signal quench from fluorescent probe
148
(100 nM Fluorescein in PBS) due to the reactive oxygen species, generated from the
149
thermal decomposition of 240 mM AAPH [2,2`-azobis(2-methylpropionamidine)
150
dihydrochloride] at 37°C. Trolox (12.5 – 200 µM) standard (3 fold), or diluted (1:100)
151
grape marc extract (6 fold) were added in the wells. Depending on the antioxidant activity,
152
the stability of the fluorescence signal was affected. Fluorescence was recorded over time,
153
and from integrating the area under the kinetic curves for each well, the antioxidant
154
capacity was estimated (Optima-MARS Data Analysis). ORAC values of the extracts were
155
compared to Trolox standards and expressed as Trolox equivalents (TE) [µmol Trolox/ g
156
dry material].
8
157
2.6. Experimental design
158
Heat treatment on grape marc was performed from different packages of 100 g. Every day
159
a new package was opened, from which the corresponding moisture content determination
160
was performed and control sample was extracted. Repeatability between the different
161
batches was studied by control samples and a heat treatment in triplicate at 150°C with 3
162
sample tubes withdrawn after 60, 120 and 240 minutes, respectively. Variations between
163
the different tubes was determined from one grape marc batch by exposing the 3 tubes to
164
150°C for 60 minutes heat treatment.
165
Heat treatment on filtered extract was performed from one extraction batch. Repeatability
166
was studied by a heat treatment at 150°C for 30 minutes in triplicate. All analytical
167
procedures were used on the repeated treatments. The estimated errors from the repeated
168
experiments were considered for the other experimental conditions as well. An overview
169
of the performed experiments is presented in Table 1.
170 171 172
3. Results and discussion 3.1. Repeatability
9
173
The analyses of the different grape marc batches after 1 hour of extraction without heat
174
treatment (Table 1, samples 1-6.) resulted in 4.01 ± 0.57 mg cyanidin-3-glucoside/ g DM,
175
44.81 ± 10.48 mg GA/g DM, and 426.9 ± 115.1 µmol Trolox/ g DM for anthocyanin, total
176
phenol and ORAC values, respectively. Errors determined from the standard deviation of
177
triplicated analyses from the same batch with the same treatment (Table 1, sample 1)
178
resulted in 9.5%, 1.6% and 13.0% for anthocyanin, total phenol and antioxidant values,
179
respectively. These errors allow the comparison between results of the same batch to study
180
the behaviour of the kinetics during the treatment, however, because of sample variability,
181
separated batches would not provide exact comparable results. For this reason,
182
experimental values are compared to their corresponding control value from the same
183
grape marc package.
184
In the case of the filtered extract (Table 1, samples 7-9) untreated extracts were compared
185
from the same batch on three days in a row. Anthocyanin content was 3.81 ± 0.10 mg
186
cyanidin-3-glucoside/ g DM, total phenol content was 82.79 ± 2.67 mg GA/g DM and
187
ORAC value was 470.0 ± 62.5 µmol Trolox/ g DM. The triplicated experiment on filtered
188
extract with the same heat treatment (Table 1, sample 10) resulted in 16%, 1.1% and
189
13.3% of error for anthocyanin, total phenol and antioxidant values, respectively. As in the
190
case of the thermal treatment on grape marc, kinetic studies of anthocyanin degradation in
191
filtered extract were presented as relative concentration values (Section 3.2.).
192 193
3.2. Anthocyanin content
194
From the experimental values obtained for anthocyanin degradation (Figure 1 a, b), first
195
order kinetics can be proposed to describe the concentration evolution (dC/dt) of
196
monomeric anthocyanins over the heating process (Eq 1).
197
dC BC dt
(1) 10
198
The rate parameter (B) depends on temperature according to the Arrhenius model (Eq 2),
199
as previously suggested by others (Wang, & Xu, 2007, Cisse, Vaillant, Acosta, Dhuique-
200
Mayer, & Dornier, 2009, Patras, Brunton, O’Donell, & Tiwari, 2010):
B B0 e
201
Ea RT
(2)
202
The heating of the sample inside of the tubes was slow and a period between 30 and
203
60 min was required to reach the temperature value set in the oven. Since the rate
204
parameter (B) is temperature dependent, it was necessary to correlate temperature
205
evolution in the heating process as a function of time to integrate Eq. 1. In previous papers
206
(Patras, Brunton, O’Donell, & Tiwari, 2010, Dolan 2003, Mishra, Dolan, & Yang, 2008),
207
this effect has been modelled using the “time-temperature history (β)”, a scaling factor of
208
the rate constant obtained by integration of the rate parameter over time.
209
In this paper, the heating temperature evolution of the samples was found to be
210
successfully correlated by first order kinetics (Eq 3):
211
T (t ) Tinitial (T final Tinitial )(1 e kT t )
(3)
212
The value of the heating rate constant (kT) was obtained for each experiment by
213
minimizing the sum of the squares of the differences between experimental and calculated
214
temperatures. Correlation parameters (Table 2) were calculated by the regression function
215
from experimental values.
216
By combining Eqs 1 to 3, anythocyanin degradation can be described (Eq. 4) by the values
217
of the pre-exponential constant of the degradation rate parameter (B0), the degradation
218
energy of activation (Ea) and the heating rate constant of the thermal treatment(kT) :
219
Ea dC C B0 exp k t T R (T dt ( T T )( 1 e ) initial final initial
(4)
11
220
This differential equation was integrated by the Euler’s method, with a time step Δt =
221
0.1 min.
222
A single pair of values (B0, Ea) is used to describe all degradation experiments with the
223
same material (grape marc or liquid extract), regardless of the heating process or
224
temperature. These values were obtained by minimizing the following objective function
225
(Eq. 5):
2 Cijexp Cijcal s
226
nj
2
(5)
j 1 i 1
227
exp cal where, Cij and Cij are the experimental and calculated values of anthocyanin
228
concentration, nj the number of data points in the degradation experiment j, and s the total
229
number of experiments with the same material (grape marc or liquid extract).
230
To locate initial estimates for B0 and Ea, fixed values of B0 were established in a range of
231
104 – 107 min-1, and corresponding Ea values were calculated by minimizing the objective
232
function (χ2). The pair (B0, Ea) providing the minimum value of χ2 was used as an initial
233
estimate for the simultaneous minimization of both parameters.
234
The standard error of the correlation parameters (B0, Ea) was calculated (Eq. 6):
235
2 2 N Ci 2 a N p i 1 ak
1
(6)
k
s
236
where, N is the total number of data points N n j , p the number of correlation j 1
237
parameters and ak the corresponding correlation parameter (B0 or Ea).
12
238
Considering the wide sample variability and that a single set of parameters is used at each
239
temperature for all experiments, the correlation results show an acceptable agreement
240
between experimental and correlated values (Figure 1 a, b): R2= 0.9261 in the case of the
241
grape marc and R2= 0.9731 in the case of the filtered extract. The graphs initially show a
242
short flat region, corresponding to the heating stage, where temperature was not high
243
enough to produce significant anthocyanin degradation. After 10 or 20 min, degradation
244
begins and increases rapidly due to the combination of rising temperature and first order
245
degradation kinetics. The activation energy for the grape marc (Table 2) is slightly higher
246
than the one for the filtered extract, which suggest that temperature rise will degrade
247
anthocyanins in grape marc samples more rapidly. The pre-exponential term is higher by
248
one order of magnitude for the degradation of anthocyanins in the presence of grape marc
249
than in the isolated filtered extract. This demonstrates a higher tendency for reactions
250
converting monomeric anthocyanin in other molecules when the solid is present. Precise
251
reaction mechanisms are difficult to describe in complex, natural raw materials, where
252
numerous possibilities and molecules may play important roles. However, the activation
253
energy values are in agreement with those from other studies (Cisse, Vaillant, Acosta,
254
Dhuique-Mayer, & Dornier, 2009, Yue, & Xu 2008, Patras, Brunton, O’Donell, & Tiwari,
255
2010).
256
13
257
Isothermal heating degradation can be predicted at any temperature from the obtained
258
correlation parameters. For an efficient extraction process, more than 10% degradation of
259
valuable compounds would not be convenient. Using the degradation kinetic parameters in
260
isothermal simulation, 10% decrease in liquid and solid phase can be compared (Figure 2)
261
at different temperatures. It is clearly seen that anthocyanin molecules suffer higher
262
degradation in the presence of the solid phase, by one order of magnitude. These
263
degradation rates should be taken into account in the development of new grape marc
264
extraction processes, in order to avoid major deterioration of anthocyanins in the material.
265 266
3.3. Total phenol content
267
In the case of thermal treatment on grape marc at 80 ºC (Figure 3a), significant decrease
268
can be observed in TPC values. The slow heating up of the experimental setup could give
269
rise to polyphenol oxidase enzyme activity in the material, lowering the phenol content in
270
the first hour (Kalt, 2005). Further degradation of TPC did not occur at 80 °C.
271
When heating up to 100 °C, TPC value was maintained for the first hour. However, a
272
slight increment was achieved during the remaining time. An increase was detected at 150
273
°C for the first two hours, but a reduction took place later due to the thermal degradation of
274
phenolic compounds. When grape seeds were heated before extraction, liberation of
275
polyphenols was achieved at 100 °C and 150 °C after 60 and 40 minutes, respectively
276
(Kim et al 2006). On the other hand, total extractable polyphenols of grape pomace peels
277
decreased after drying at 100 °C and 140 °C, while no change was observed at 60 °C
278
(Larrauri, 1997).
14
279
The raw material studied contains both grape seeds and peels, and thus a combination can
280
occur: on one hand initial thermal degradation of free polyphenols and later degradation of
281
bounded polyphenols as well. On the other hand, the release of bound polyphenol
282
compounds and phenolic acid derivatives, because of partial lignin degradation in the
283
seeds (Maillard, & Berset, 1995).
284
Filtered extracts did not behave the same way as solid grape marc. At 80 °C and 100 °C
285
there was no TPC content change (Figure 3b). However, at 150 °C a significant increase
286
was observed from 81.3 2.0 (mg gallic acid/ g dry material) to 117.1 2.9 after 4 hours.
287
Enzyme activity did not affect the material, since the extraction previous to the thermal
288
treatment was performed at low temperature (40 °C) for 24 hours. Extracts from grape
289
pomace after fermentation did not change in polyphenol concentration at 100 and 150 ºC,
290
while at 200 °C, 72% degradation was observed after 120 min (Cruz, Conde, Domínguez,
291
& Parajo, 2007). Grape marc contains a wide range of phenolic components, including
292
anthocyanins, flavonols, flavan-3-ols and flavanonols. These compounds are responsible
293
for the nutritional quality and antioxidant activity of foodstuffs and afford the positive
294
health effects associated with this and other similar natural materials (Larrauri, 1997). Due
295
to their diverse structure, different polyphenols are found in the distinct parts of the grape
296
skin and seed. Some of them are free and can be found in vacuoles, while others are
297
associated to cell wall compounds or to polysaccharide structures in the skin cells (Pinelo,
298
Arnous, & Meyer, 2006). The structural and bonding state of the different present
299
polyphenols would explain the distinct overall behaviour of the total phenol concentration
300
due to thermal treatments (Kim et al, 2006).
301
3.4. Antioxidant activity (ORAC)
15
302
Antioxidant power was analysed by the capacity to absorb oxygen radicals and compared
303
to the power of known trolox concentration. Antioxidant activity of solid grape marc
304
(Figure 4 a) showed similar behaviour to the one observed for TPC values. Treatment at 80
305
°C caused a slight decrease at the beginning of the process, but no further degradation
306
occurred.
307
The initial increment of ORAC values turned to a decomposition of active compounds in
308
the second hour of the thermal treatment. Similar behaviour was found when drying
309
methods were studied on grape skin with the ferric thiocyanate method (Larrauri, 1997).
310
Radical scavenging and reducing power of pre-heated grape seed was positively modified
311
at 100 °C and 150 °C, while at 200 °C both values decreased (Kim et al, 2006). In liquid
312
extracts (Figure 4 b) the antioxidant activity did not change at 80 °C, while at 100 °C an
313
increment was observed without final degradation. However degradation was once more
314
observed at 150 °C. Radical scavenging power of wine making by-products and of bilberry
315
extracts showed a similar tendency after heat treatments (Cruz, Conde, Domínguez, &
316
Parajo, 2007, Yue, & Xu, 2008). Either the elevated reactivity of the degradation products
317
or the activation of previously bound phenolics could explain the results obtained here
318
(Maillard, & Berset 1995, Guillot, Malnoë, & Stadler, 1996). However, the exact
319
composition and in vivo health benefits of the altered compounds need to be further
320
studied.
321
4. Conclusions
16
322
This study explored the effects of heat treatments on a by-product of the wine making
323
industry and its extract, in terms of anthocyanin degradation, total phenol content (TPC)
324
and antioxidant activity (ORAC). In the case of anthocyanins, a fast degradation behavior
325
was observed. A non-isothermal procedure was developed and used to model degradation
326
kinetics, where isothermal behaviour was simulated by first order kinetics and the
327
Arrhenius model. The presence of solid grape marc showed a 10-fold increase in the
328
thermal degradation rate of anthocyanins, when compared to the heating of the filtered
329
extract. TPC and ORAC values were less sensitive to heating, where even an initial
330
increase of concentration was observed over 100 °C. As observed with the anthocyanins,
331
the presence of semi-solid grape marc made TPC and ORAC more vulnerable to heat
332
treatment than the isolated liquid extract. From these results, it can be concluded that
333
process intensification methods for antioxidant extraction from grape marc may preserve at
334
least 90% of the active compounds up to 150°C, if the processing time does not exceed 1
335
minute at this temperature.
336
17
337
5. Abbreviations and symbols
338
Abbreviations:
339
TPC
total phenol content
340
ORAC
oxygen radical absorbance capacity [µmol trolox/ g dry material]
341
DM
dry material
342
GA
gallic acid
343
PBS
phosphate buffer solution
344
TE
trolox equivalent
345
Symbols:
346
C
anthocyanin concentration
[mg gallic acid/ g dry material]
347
t
time
[min]
348
B
degradation rate parameter
[min-1]
349
B0
pre-exponential constant of B
[min-1]
350
T
temperature
[K]
351
T initial
temperature before heating treatment
[K]
352
T final
final temperature after heating treatment
[K]
353
kT
heating rate constant at corresponding T°C
[min-1]
354
Ea
energy of activation
[J/mol]
355
R
gas constant
8.314 [J/molK]
356
χ2
sum of the squares of errors
357
s
total number of experiments
[mg gallic acid/ g dry material]
358
experimental value of anthocyanin concentration
359
calculated value of anthocyanin concentration
360
nj
number of experimental data points of experiment j
18
361
N
Total number of experimental data points with the same material (grape
362
marc or liquid extract)
363
p
number of correlation parameters
364
ak
parameter in standard error calculation (Ea or B0)
365 366
Acknowledgments
367
The authors wish to thank the Junta de Castilla y León for the financial support of the
368
project VA330U13, and the Spanish Ministry of Economy and Competitiveness for the
369
Projects 356 CTQ2010-15475 and ENE2012-33613. The research group is ‘‘Grupo de
370
Excelencia GREX-11 de la Junta de Castilla y León”.
371
References
372
Amendola, D., De Faveri, D. M., & Spigno, G. (2010). Grape marc phenolics:
373
Extraction kinetics, quality and stability of extracts. Journal of Food Engineering, 97,
374
384 – 392.
375
AOAC Official Method 2005.02. (2006). Total Monomeric Anthocyanin Pigment
376
Content of Fruit Juices, Beverages, Natural Colorants, and Wines. pH Differential
377
Method. AOAC International, 37.1.68.
378
Casazza, A. A., Aliakbarian, B., Mantegna, S., Cravotto, G., & Perego, P. (2010).
379
Extraction of phenolics from Vitis vinifera wastes using non-conventional techniques.
380
Journal of Food Engineering, 100, 50 –55.
381
Cisse, M., Vaillant, F., Acosta, O., Dhuique-Mayer, C., & Dornier, M. (2009). Thermal
382
Degradation Kinetics of Anthocyanins from Blood Orange, Blackberry, and Roselle
383
Using the Arrhenius, Eyring, and Ball Models. Journal of Agricultural and Food
384
Chemistry, 57, 6285–6291. 19
385
Cruz, J.M., Conde, E., Domínguez, H., & Parajo, J.C. (2007). Thermal stability of
386
antioxidants obtained from wood and industrial wastes. Food Chemistry, 100, 1059–
387
1064.
388
Dolan, K.D. (2003). Estimation of Kinetic Parameters for Nonisothermal Food
389
Processes. Journal of Food Science, 68(3), 728 – 741.
390
Garcia-Perez, J.V., García-Alvarado, M.A., Carcel, J.A., & Mulet, A. (2010).
391
Extraction kinetics modeling of antioxidants from grape stalk (Vitis vinifera var.
392
Bobal): Influence of drying conditions. Journal of Food Engineering, 101, 49–58.
393
Guillot, F.L., Malnoë, A., & Stadler, R.H. (1996). Antioxidant Properties of Novel
394
Tetraoxygenated Phenylindan Isomers Formed during Thermal Decomposition of
395
Caffeic Acid. Journal of Agricultural and Food Chemistry, 44 (9), 2503–2510.
396
Kalt, W. (2005). Effects of Production and Processing Factors on Major Fruit and
397
Vegetable Antioxidants. Journal of Food Science, 70(1), R11 – R19.
398
Kim, S.Y., Jeong, S.M., Park, W.P., Nam, K.C., Ahn, D.U., & Lee, S.C. (2006). Effect
399
of heating conditions of grape seeds on the antioxidant activity of grape seed extracts.
400
Food Chemistry, 97, 472–479.
401
Lafka T.I., Sinanoglou, V., & Lazos, E. S. (2007). On the extraction and antioxidant
402
activity of phenolic compounds from winery wastes. Food Chemistry, 104, 1206–1214.
403
Larrauri, J.A., Rupérez, P., & Saura-Calixto, F. (1997). Effect of Drying Temperature
404
on the Stability of Polyphenols and Antioxidant Activity of Red Grape Pomace Peels.
405
Journal of Agricultural and Food Chemistry, 45, 1390-1393.
20
406
Maillard, M.N., & Berset, C. (1995). Evolution of Antioxidant Activity during Kilning:
407
Role of Insoluble Bound Phenolic Acids of Barley and Malt. Journal of Agricultural
408
and Food Chemistry, 43, 1789-1793.
409
Mishra, D.K., Dolan, K.D., & Yang L. (2008). Confidence Intervals for Modeling
410
Anthocyanin Retention in Grape Pomace during Nonisothermal Heating. Journal of
411
Food Science. 73(1), E9 - E15.
412
Nassiri-Asl, M., & Hosseinzadeh, H. (2009). Review of the Pharmacological Effects of
413
Vitis vinifera (Grape) and its Bioactive Compounds. Phytotheraphy Research, 23, 1197
414
– 1204.
415
Negro, C., Tommasi, L., & Miceli, A. (2003). Phenolic compounds and antioxidant
416
activity from red grape marc extracts. Bioresource Technology, 87, 41 – 44.
417
Patras, A., Brunton, N.P., O’Donnell, C., & Tiwari, B.K. (2010). Effect of thermal
418
processing on anthocyanin stability in foods; mechanisms and kinetics of degradation.
419
Trends Food Science and Technology, 21, 3 – 11.
420
Peralbo-Molina, A., Priego-Capote, F., & Luque de Castro, M. D. (2012). Comparison
421
of extraction methods for exploitation of grape skin residues from ethanol distillation.
422
Talanta, 101, 292–298.
423
Pinelo, M., Arnous, A., & Meyer. A.S. (2006). Upgrading of grape skins: Significance
424
of plant cell-wall structural components and extraction techniques for phenol release.
425
Trends Food Science and Technology, 17, 579 – 590.
21
426
Singleton, V.L., Orthofer, R.., & Lamuela-Raventos, R.M. (1999). Analysis of total
427
phenols and other oxidation substrates and antioxidants by means of Folin–Ciocalteu
428
reagent. Methods in Enzymology, 299, 152–178.
429
Solyom, K., Kraus, S., Mato, R.B., Gaukel, V., Schuchmann, H.P., & Cocero, M.J.
430
(2013). Dielectric properties of grape marc: Effect of temperature, moisture content
431
and sample preparation method. Journal of Food Engineering, 119, 33-39.
432
Spigno, G., & De Faveri, D.M. (2007). Antioxidants from grape stalks and marc:
433
Influence of extraction procedure on yield, purity and antioxidant power of the
434
extracts. Journal of Food Engineering 78, 793 – 801.
435
Spigno, G., Tramelli, L., & De Faveri, D.M. (2007). Effects of extraction time,
436
temperature and solvent on concentration and antioxidant activity of grape marc
437
phenolics. Journal of Food Engineering, 81, 200 – 208.
438
Wang, W.D., & Xu, S.Y. (2007). Degradation kinetics of anthocyanins in blackberry
439
juice and concentrate. Journal of Food Engineering, 82, 271–275.
440
Yue, X. & Xu, Z. (2008). Changes of Anthocyanins, Anthocyanidins, and Antioxidant
441
Activity in Bilberry Extract during Dry Heating. Journal of Food Science, 73(6), C494
442
– C499.
443
22
444
FIGURE CAPTIONS
445
Figure 1. Experimental (Exp.) and model fit (Regr.) values of the non-isothermal
446
degradation kinetics of a) grape marc and b) filtered extract anthocyanins at 80, 100 and
447
150 °C.
448
Figure 2. Required combination of time and temperature for 10 % anthocyanin decrease in
449
semi-solid grape marc and liquid extract exposed to isothermal heating.
450
Figure 3. Total phenol content of a) grape marc and b) filtered extract at 80, 100 and 150
451
°C.
452
Figure 4. Antioxidant activity of a) grape marc and b) filtered extract at 80, 100 and 150
453
°C (ORAC values).
454 455
23
456
457
TABLES
458 459
Table 1 Experimental design
460
461
Table 2. Correlation parameters of the heating process and the anthocyanin degradation
462
kinetics.
463 464
24
495
496 497
28
498 499
29
500
501 502
30
503
504 505
31
Table 1 Experimental design
Filtered extract (Same batch)
Grape marc (different batches)
Sample
Temperature (°C) 1 2 3 4 5 6
150 150 150 80 100 150
7
80
8
100
9
150
10
150
Time (min) 0; 60; 60; 60 0; 60;120;240 0; 60;120;240 0; 60;120;240 0; 60;120;240 0; 20;40 0; 20; 40; 60; 120; 240 0; 20; 40; 60; 120; 240 0; 20; 40; 60; 120; 240 0; 30; 30; 30
Table 2. Correlation parameters of the heating process and the anthocyanin degradation kinetics.
Grape marc
Parameter
Filtered extract
k80°C
min -1
0.0486
0.0006 R2 = 0.9995
0.0688
0.0014 R2 = 0.9988
k100°C
min -1
0.0615
0.0010 R2 = 0.9996
0.0851
0.0015 R2 = 0.9979
k150°C
min -1
0.0688
0.0009 R2 = 0.9995
0.1156
0.0030 R2 = 0.9952
B0
min -1
Ea
J/mol
1.67*106 55980
0.34*106 620
1.32*105 53196
1.62*104 399
514
Highlights:
515 516 517 518 519 520
• • • • •
Thermal treatments on enhanced polyphenol extraction degrades active compounds. Antioxidant degradation was studied at 80-150°C of grape marc and filtered extract Anthocyanins (AC) follow first order degradation; total phenol content and ORAC not. Degradation was 10 times faster in grape marc than in filtered extract. 90% of AC can be preserved in a pretreatment at 150°C, if no longer than 1 minute.
521 522
33
S0308-8146(14)00406-3 http://dx.doi.org/10.1016/j.foodchem.2014.03.021 FOCH 15550
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
18 November 2013 30 January 2014 5 March 2014
Please cite this article as: Sólyom, K., Solá, R., Cocero, M.J., Mato, R.B., Thermal degradation of grape marc polyphenols, Food Chemistry (2014), doi: http://dx.doi.org/10.1016/j.foodchem.2014.03.021
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
2
Thermal degradation of grape marc polyphenols
3
Katalin SÓLYOM, Ruth SOLÁ, María José COCERO, Rafael B. MATO*
4
5
University of Valladolid, Industrial Engineering School, Department of Chemical
6
Engineering and Environmental Technology; High Pressure Processes Group; C/Dr.
7
Mergelina, s/n, 47011-Valladolid, Spain
8 9
*Corresponding author:
10
Rafael B. Mato
11
email: [email protected]
12
telephone: +34 983 423 177
13
1
14
ABSTRACT
15
Bioactive compounds of wine making by-products are of interest in food, cosmetics and
16
pharmaceutical industries. Extraction of antioxidants under mild conditions is time-
17
consuming, giving ground to the development of intensification processes where the
18
operation at high temperature may deteriorate extract quality. This study examined thermal
19
degradation of grape marc and its filtered extract (80, 100 and 150 °C). The decrease in
20
anthocyanin content was modelled under non-isothermal conditions by first order kinetics,
21
using the Arrhenius equation. Simulated degradation under isothermal heating showed that
22
the grape marc is more sensitive by one order of magnitude to heat than the filtered extract.
23
This tendency was also confirmed by analyses of the total phenolic content and antioxidant
24
activity. It is suggested that an optimal combination of temperature, treatment time and
25
also raw material environment could be found in process intensification.
26
Keywords:
27
Grape pomace, pH differential method, ORAC, degradation kinetics, non-isothermal
28
heating, anthocyanin
29 30
2
31
32
1. Introduction
33
Worldwide wine production generates a large amount of semi-solid residue (grape marc),
34
with high antioxidants content. Flavonoids, anthocyanins and stilbene derivatives are
35
bioactive compounds present in this residue, and they are of interest for food, cosmetics
36
and pharmaceutical industries due to their antioxidant, antimicrobial, antiviral or
37
anticarcinogenic features (Nassiri Asl & Hosseinzadech, 2009). The elimination of these
38
compounds is also desirable before using the grape pomace as fertilizer, in order to avoid
39
inhibition problems (Negro, Tommasi, & Miceli, 2003, Pinelo, Arnus, & Meyer, 2006).
40
Mild extraction techniques (maceration and stirred extraction) are time consuming and
41
usually require an excessive amount of solvent, leading to an inefficient and costly
42
recovery process (Spigno, & De Faveri, 2007, Spigno, Tramelli, & De Faveri, 2007, Lafka,
43
Sinanoglu, & Lazos, 2007, Amendola, Spigno, & De Faveri, 2010).
3
44
Novel techniques and pre-treatments in order to enhance the extraction kinetics of valuable
45
components include high temperature during processing. Extraction at high pressure and
46
temperature (110 °C) for 60 minutes was applied and compared to microwave assisted
47
extraction for the recovery of bioactive compounds from grape waste (Casazza,
48
Aliakbarian, Mantegna, Cravotto, & Perego, 2010). Peralbo-Molina and co-workers (2012)
49
compared superheated extraction at 180 °C for 60 min with microwave and ultrasound
50
assisted extraction in order to enhance the extract quality and process efficiency, when
51
compared to conventional maceration. The effect of temperature in microwave processing
52
can also alter the interaction between the material and the electromagnetic field (Solyom et
53
al, 2013). Drying pre-treatment was also applied for grape stalks at temperatures between
54
40 and 115 °C, and improved extraction kinetics were described by a diffusion model
55
(Garcia-Perez, García-Alvarado, Carcel, & Mulet, 2010).
56
Although these process intensification methods increased the quantity or the quality of the
57
extract, their destructive effect on the antioxidant content may be masked by a larger
58
beneficial effect on the extraction stage efficiency. The physical complexity of the natural
59
raw material and the variety of experimental conditions may modify interactions between
60
active compounds, thus changing the final product quality. The effect of thermal
61
treatments on liquid extracts has been previously reported (Wang, & Xu, 2007, Cruz,
62
COnde, Domínguez, & Parajo, 2007, Cisse, Vaillant, Acosta, Dhuique-Mayer, & Dornier,
63
2009), and first order kinetic degradation was found for anthocyanins, while for other
64
antioxidants this tendency was not so clear. Drying of raw material deteriorated the
65
antioxidant quality of grape peels when treated above 60°C (Larrauri, Ruperez, & Saura-
66
Calixto, 1997).
4
67
Heating of grape seeds prior to extraction released more polyphenols, verified by
68
antioxidant activity and HPLC analysis (Kim et al, 2006). The anthocyanin content of dry
69
bilberry extracts was reduced after 30 min, 30% in dry bilberry extracts at 100 °C, while
70
antioxidant activity increased with increasing temperature (Yue, & Xu, 2008). It is clear
71
that the raw materials variability and the different heating conditions are hardly
72
comparable in terms of temperature and time, solid content, or the presence of extraction
73
solvent. Consequently, the aim of this study was to compare the effects of heating on the
74
semi-solid grape marc waste and on its liquid extract. To avoid excessively high operating
75
pressures (above 10 bar), the temperature range was restricted to between 80 and 150ºC,
76
which corresponds to the range in boiling temperatures of 50%vol mixtures of ethanol-
77
water mixtures between 1 and 10 bar. To this end heating was performed at 80, 100 and
78
150 °C for up to 4 hours in order to study degradation kinetics in terms of total phenol
79
content, antioxidant activity and anthocyanin content. Kinetic parameters of anthocyanin
80
degradation were obtained by regression of non-isothermal heating experimental values to
81
a first order kinetic model. These parameters were later used to estimate and compare
82
degradation in isothermal heating of semi-solid grape marc and its liquid extract.
83
2. Materials and methods
84
2.1. Grape marc
85
A mixture of red grape seeds and skins (grape marc) of the Tempranillo grape variety from
86
the region of Toro (Spain) was used in the experiments. This residue was generated in the
87
wine making process, after long maceration with a post-fermentative phase. The grape
88
marc was distributed in 100 g packages and stored at -18 °C until defrosting was
89
performed at 4 °C overnight.
90
2.2. Heat treatment of grape marc
5
91
In these experiments, the raw material was first exposed to heat and then further extracted
92
with solvent. Grape marc (10 0.001 g) was filled in three identical stainless steel tubes
93
(length: 220 mm, inner diameter: 12.3 mm) and closed with stainless steel stoppers to hold
94
up the arising pressure during the experiment. The tubes were placed in an oven, set at 80,
95
100 and 150 °C, and were removed consecutively after 60, 120 and 240 minutes for
96
analysis. The series at 150 °C was repeated with sampling times at 20, 40 and 60 minutes
97
to better follow the fast degradation kinetics. Tubes were immediately cooled down in an
98
ice bath and further stored at 4 °C until extraction, which was performed within 4 hours.
99
Also, one fraction of original grape marc was kept for extraction, without previous heat
100
treatment, to use as a control sample in comparison to the heated material.
101
After the heat treatment, grape marc was removed from the tubes and placed in glass
102
vessels for extraction with 50 ml of 50% volumetric mixture of ethanol (96%) and
103
acidified water (pH=1; H2SO4). Extraction was performed for 1 hour at 40 °C, under
104
continuous stirring on an orbital shaker. Extraction was stopped by separating liquid and
105
solid phases in centrifuge (4000 rpm, 5 min). The supernatants were filtered (PET
106
membrane, 0.2 µm) for further analyses.
107
2.3. Heat treatment of filtered extract
6
108
In the second series of experiments, extraction was first performed from the grape marc,
109
and the obtained extract was further exposed to heat treatment. A 40 g sample of raw
110
material was extracted in 200 ml 50% volumetric mixture of ethanol (96%) and acidified
111
water (pH=1; H2SO4) at 40 °C for 24 hours in the dark. The aim of this experiment was to
112
study the effect of temperature on the stability of polyphenols in the absence of a solid
113
phase. A longer extraction time (24 h) was chosen to obtain a high extraction yield, and so
114
that most of the polyphenols originally present in the solid sample could be extracted to a
115
liquid before heat treatment was performed. In the case of the extraction of the preheated
116
grape marc, a 1 hour extraction time was sufficient to evaluate the differences after the
117
heat treatments. After the extraction, a centrifuge was used for phase separation (4000
118
rpm, 5 min) and the supernatant was filtered through a 0.45 µm borosilicate filter
119
(Millipore). The filtered extract obtained was divided into five stainless steel tubes (length:
120
160 mm, inner diameter: 3.9 mm). In this instance thinner tubes were used to achieve
121
faster heating of the filtered extract. In the case of the grape marc, a larger inner tube
122
diameter was necessary to fit the skin and seed particles into the tube. Closed tubes were
123
placed in the oven at 80, 100 and 150 °C, and removed one by one after 20, 40, 60, 120
124
and 240 minutes. After immediate cooling in an ice bath, the extracts were filtered (PET
125
membrane, 0.2 µm) and kept refrigerated for further analysis. A fraction of the original
126
unheated filtered extract was also kept as control sample for comparison.
127
2.4. Temperature evolution in heat treatments
128
The thermal history of the experimental setup, was followed by temperature measurement
129
inside separate tubes with a K-type thermocouple. Temperature was recorded until the set
130
value was reached.
131
2.5. Analysis
7
132
The moisture content of the grape marc (≈ 85% w/w) was determined in triplicate for
133
every grape marc batch used by recording the weight difference after drying at 105ºC for
134
24 hours, until a constant weight was observed.
135
Total phenol content (TPC) was quantified using the Folin-Ciocalteu method, as Gallic
136
acid equivalents per gram of dry material [mg GA/g DM] (Singleton, Orthofer, &
137
Lamuela-Raventos 1999). A 40 μl volume of filtered sample was diluted with distilled
138
water (3 ml) and mixed with 200 μl of Folin-Ciocalteu reagent. After 10 minutes, saturated
139
Na2CO3 was added to the solutions and the samples were incubated at 40ºC for 30 min.
140
Absorbance was measured at 765 nm (UV 2550 Shimadzu UV/VIS spectrometer) and
141
standard dilutions (50-900 ppm) of Gallic acid were used for calibration.
142
Total monomeric anthocyanin pigment content was determined by the pH differential
143
method (AOAC 2005.02.). Results are expressed on the basis of cyanidin-3-glucoside [mg
144
cyanidin-3-glucoside/ g DM]
145
Antioxidant activity was measured in terms of Oxygen Radical Absorbance Capacity
146
(ORAC) based on the Application note 148 (BMG-Labtech). A 96 well-plate was used in a
147
Fluostar Optima (BMG-Labtech) to determine the signal quench from fluorescent probe
148
(100 nM Fluorescein in PBS) due to the reactive oxygen species, generated from the
149
thermal decomposition of 240 mM AAPH [2,2`-azobis(2-methylpropionamidine)
150
dihydrochloride] at 37°C. Trolox (12.5 – 200 µM) standard (3 fold), or diluted (1:100)
151
grape marc extract (6 fold) were added in the wells. Depending on the antioxidant activity,
152
the stability of the fluorescence signal was affected. Fluorescence was recorded over time,
153
and from integrating the area under the kinetic curves for each well, the antioxidant
154
capacity was estimated (Optima-MARS Data Analysis). ORAC values of the extracts were
155
compared to Trolox standards and expressed as Trolox equivalents (TE) [µmol Trolox/ g
156
dry material].
8
157
2.6. Experimental design
158
Heat treatment on grape marc was performed from different packages of 100 g. Every day
159
a new package was opened, from which the corresponding moisture content determination
160
was performed and control sample was extracted. Repeatability between the different
161
batches was studied by control samples and a heat treatment in triplicate at 150°C with 3
162
sample tubes withdrawn after 60, 120 and 240 minutes, respectively. Variations between
163
the different tubes was determined from one grape marc batch by exposing the 3 tubes to
164
150°C for 60 minutes heat treatment.
165
Heat treatment on filtered extract was performed from one extraction batch. Repeatability
166
was studied by a heat treatment at 150°C for 30 minutes in triplicate. All analytical
167
procedures were used on the repeated treatments. The estimated errors from the repeated
168
experiments were considered for the other experimental conditions as well. An overview
169
of the performed experiments is presented in Table 1.
170 171 172
3. Results and discussion 3.1. Repeatability
9
173
The analyses of the different grape marc batches after 1 hour of extraction without heat
174
treatment (Table 1, samples 1-6.) resulted in 4.01 ± 0.57 mg cyanidin-3-glucoside/ g DM,
175
44.81 ± 10.48 mg GA/g DM, and 426.9 ± 115.1 µmol Trolox/ g DM for anthocyanin, total
176
phenol and ORAC values, respectively. Errors determined from the standard deviation of
177
triplicated analyses from the same batch with the same treatment (Table 1, sample 1)
178
resulted in 9.5%, 1.6% and 13.0% for anthocyanin, total phenol and antioxidant values,
179
respectively. These errors allow the comparison between results of the same batch to study
180
the behaviour of the kinetics during the treatment, however, because of sample variability,
181
separated batches would not provide exact comparable results. For this reason,
182
experimental values are compared to their corresponding control value from the same
183
grape marc package.
184
In the case of the filtered extract (Table 1, samples 7-9) untreated extracts were compared
185
from the same batch on three days in a row. Anthocyanin content was 3.81 ± 0.10 mg
186
cyanidin-3-glucoside/ g DM, total phenol content was 82.79 ± 2.67 mg GA/g DM and
187
ORAC value was 470.0 ± 62.5 µmol Trolox/ g DM. The triplicated experiment on filtered
188
extract with the same heat treatment (Table 1, sample 10) resulted in 16%, 1.1% and
189
13.3% of error for anthocyanin, total phenol and antioxidant values, respectively. As in the
190
case of the thermal treatment on grape marc, kinetic studies of anthocyanin degradation in
191
filtered extract were presented as relative concentration values (Section 3.2.).
192 193
3.2. Anthocyanin content
194
From the experimental values obtained for anthocyanin degradation (Figure 1 a, b), first
195
order kinetics can be proposed to describe the concentration evolution (dC/dt) of
196
monomeric anthocyanins over the heating process (Eq 1).
197
dC BC dt
(1) 10
198
The rate parameter (B) depends on temperature according to the Arrhenius model (Eq 2),
199
as previously suggested by others (Wang, & Xu, 2007, Cisse, Vaillant, Acosta, Dhuique-
200
Mayer, & Dornier, 2009, Patras, Brunton, O’Donell, & Tiwari, 2010):
B B0 e
201
Ea RT
(2)
202
The heating of the sample inside of the tubes was slow and a period between 30 and
203
60 min was required to reach the temperature value set in the oven. Since the rate
204
parameter (B) is temperature dependent, it was necessary to correlate temperature
205
evolution in the heating process as a function of time to integrate Eq. 1. In previous papers
206
(Patras, Brunton, O’Donell, & Tiwari, 2010, Dolan 2003, Mishra, Dolan, & Yang, 2008),
207
this effect has been modelled using the “time-temperature history (β)”, a scaling factor of
208
the rate constant obtained by integration of the rate parameter over time.
209
In this paper, the heating temperature evolution of the samples was found to be
210
successfully correlated by first order kinetics (Eq 3):
211
T (t ) Tinitial (T final Tinitial )(1 e kT t )
(3)
212
The value of the heating rate constant (kT) was obtained for each experiment by
213
minimizing the sum of the squares of the differences between experimental and calculated
214
temperatures. Correlation parameters (Table 2) were calculated by the regression function
215
from experimental values.
216
By combining Eqs 1 to 3, anythocyanin degradation can be described (Eq. 4) by the values
217
of the pre-exponential constant of the degradation rate parameter (B0), the degradation
218
energy of activation (Ea) and the heating rate constant of the thermal treatment(kT) :
219
Ea dC C B0 exp k t T R (T dt ( T T )( 1 e ) initial final initial
(4)
11
220
This differential equation was integrated by the Euler’s method, with a time step Δt =
221
0.1 min.
222
A single pair of values (B0, Ea) is used to describe all degradation experiments with the
223
same material (grape marc or liquid extract), regardless of the heating process or
224
temperature. These values were obtained by minimizing the following objective function
225
(Eq. 5):
2 Cijexp Cijcal s
226
nj
2
(5)
j 1 i 1
227
exp cal where, Cij and Cij are the experimental and calculated values of anthocyanin
228
concentration, nj the number of data points in the degradation experiment j, and s the total
229
number of experiments with the same material (grape marc or liquid extract).
230
To locate initial estimates for B0 and Ea, fixed values of B0 were established in a range of
231
104 – 107 min-1, and corresponding Ea values were calculated by minimizing the objective
232
function (χ2). The pair (B0, Ea) providing the minimum value of χ2 was used as an initial
233
estimate for the simultaneous minimization of both parameters.
234
The standard error of the correlation parameters (B0, Ea) was calculated (Eq. 6):
235
2 2 N Ci 2 a N p i 1 ak
1
(6)
k
s
236
where, N is the total number of data points N n j , p the number of correlation j 1
237
parameters and ak the corresponding correlation parameter (B0 or Ea).
12
238
Considering the wide sample variability and that a single set of parameters is used at each
239
temperature for all experiments, the correlation results show an acceptable agreement
240
between experimental and correlated values (Figure 1 a, b): R2= 0.9261 in the case of the
241
grape marc and R2= 0.9731 in the case of the filtered extract. The graphs initially show a
242
short flat region, corresponding to the heating stage, where temperature was not high
243
enough to produce significant anthocyanin degradation. After 10 or 20 min, degradation
244
begins and increases rapidly due to the combination of rising temperature and first order
245
degradation kinetics. The activation energy for the grape marc (Table 2) is slightly higher
246
than the one for the filtered extract, which suggest that temperature rise will degrade
247
anthocyanins in grape marc samples more rapidly. The pre-exponential term is higher by
248
one order of magnitude for the degradation of anthocyanins in the presence of grape marc
249
than in the isolated filtered extract. This demonstrates a higher tendency for reactions
250
converting monomeric anthocyanin in other molecules when the solid is present. Precise
251
reaction mechanisms are difficult to describe in complex, natural raw materials, where
252
numerous possibilities and molecules may play important roles. However, the activation
253
energy values are in agreement with those from other studies (Cisse, Vaillant, Acosta,
254
Dhuique-Mayer, & Dornier, 2009, Yue, & Xu 2008, Patras, Brunton, O’Donell, & Tiwari,
255
2010).
256
13
257
Isothermal heating degradation can be predicted at any temperature from the obtained
258
correlation parameters. For an efficient extraction process, more than 10% degradation of
259
valuable compounds would not be convenient. Using the degradation kinetic parameters in
260
isothermal simulation, 10% decrease in liquid and solid phase can be compared (Figure 2)
261
at different temperatures. It is clearly seen that anthocyanin molecules suffer higher
262
degradation in the presence of the solid phase, by one order of magnitude. These
263
degradation rates should be taken into account in the development of new grape marc
264
extraction processes, in order to avoid major deterioration of anthocyanins in the material.
265 266
3.3. Total phenol content
267
In the case of thermal treatment on grape marc at 80 ºC (Figure 3a), significant decrease
268
can be observed in TPC values. The slow heating up of the experimental setup could give
269
rise to polyphenol oxidase enzyme activity in the material, lowering the phenol content in
270
the first hour (Kalt, 2005). Further degradation of TPC did not occur at 80 °C.
271
When heating up to 100 °C, TPC value was maintained for the first hour. However, a
272
slight increment was achieved during the remaining time. An increase was detected at 150
273
°C for the first two hours, but a reduction took place later due to the thermal degradation of
274
phenolic compounds. When grape seeds were heated before extraction, liberation of
275
polyphenols was achieved at 100 °C and 150 °C after 60 and 40 minutes, respectively
276
(Kim et al 2006). On the other hand, total extractable polyphenols of grape pomace peels
277
decreased after drying at 100 °C and 140 °C, while no change was observed at 60 °C
278
(Larrauri, 1997).
14
279
The raw material studied contains both grape seeds and peels, and thus a combination can
280
occur: on one hand initial thermal degradation of free polyphenols and later degradation of
281
bounded polyphenols as well. On the other hand, the release of bound polyphenol
282
compounds and phenolic acid derivatives, because of partial lignin degradation in the
283
seeds (Maillard, & Berset, 1995).
284
Filtered extracts did not behave the same way as solid grape marc. At 80 °C and 100 °C
285
there was no TPC content change (Figure 3b). However, at 150 °C a significant increase
286
was observed from 81.3 2.0 (mg gallic acid/ g dry material) to 117.1 2.9 after 4 hours.
287
Enzyme activity did not affect the material, since the extraction previous to the thermal
288
treatment was performed at low temperature (40 °C) for 24 hours. Extracts from grape
289
pomace after fermentation did not change in polyphenol concentration at 100 and 150 ºC,
290
while at 200 °C, 72% degradation was observed after 120 min (Cruz, Conde, Domínguez,
291
& Parajo, 2007). Grape marc contains a wide range of phenolic components, including
292
anthocyanins, flavonols, flavan-3-ols and flavanonols. These compounds are responsible
293
for the nutritional quality and antioxidant activity of foodstuffs and afford the positive
294
health effects associated with this and other similar natural materials (Larrauri, 1997). Due
295
to their diverse structure, different polyphenols are found in the distinct parts of the grape
296
skin and seed. Some of them are free and can be found in vacuoles, while others are
297
associated to cell wall compounds or to polysaccharide structures in the skin cells (Pinelo,
298
Arnous, & Meyer, 2006). The structural and bonding state of the different present
299
polyphenols would explain the distinct overall behaviour of the total phenol concentration
300
due to thermal treatments (Kim et al, 2006).
301
3.4. Antioxidant activity (ORAC)
15
302
Antioxidant power was analysed by the capacity to absorb oxygen radicals and compared
303
to the power of known trolox concentration. Antioxidant activity of solid grape marc
304
(Figure 4 a) showed similar behaviour to the one observed for TPC values. Treatment at 80
305
°C caused a slight decrease at the beginning of the process, but no further degradation
306
occurred.
307
The initial increment of ORAC values turned to a decomposition of active compounds in
308
the second hour of the thermal treatment. Similar behaviour was found when drying
309
methods were studied on grape skin with the ferric thiocyanate method (Larrauri, 1997).
310
Radical scavenging and reducing power of pre-heated grape seed was positively modified
311
at 100 °C and 150 °C, while at 200 °C both values decreased (Kim et al, 2006). In liquid
312
extracts (Figure 4 b) the antioxidant activity did not change at 80 °C, while at 100 °C an
313
increment was observed without final degradation. However degradation was once more
314
observed at 150 °C. Radical scavenging power of wine making by-products and of bilberry
315
extracts showed a similar tendency after heat treatments (Cruz, Conde, Domínguez, &
316
Parajo, 2007, Yue, & Xu, 2008). Either the elevated reactivity of the degradation products
317
or the activation of previously bound phenolics could explain the results obtained here
318
(Maillard, & Berset 1995, Guillot, Malnoë, & Stadler, 1996). However, the exact
319
composition and in vivo health benefits of the altered compounds need to be further
320
studied.
321
4. Conclusions
16
322
This study explored the effects of heat treatments on a by-product of the wine making
323
industry and its extract, in terms of anthocyanin degradation, total phenol content (TPC)
324
and antioxidant activity (ORAC). In the case of anthocyanins, a fast degradation behavior
325
was observed. A non-isothermal procedure was developed and used to model degradation
326
kinetics, where isothermal behaviour was simulated by first order kinetics and the
327
Arrhenius model. The presence of solid grape marc showed a 10-fold increase in the
328
thermal degradation rate of anthocyanins, when compared to the heating of the filtered
329
extract. TPC and ORAC values were less sensitive to heating, where even an initial
330
increase of concentration was observed over 100 °C. As observed with the anthocyanins,
331
the presence of semi-solid grape marc made TPC and ORAC more vulnerable to heat
332
treatment than the isolated liquid extract. From these results, it can be concluded that
333
process intensification methods for antioxidant extraction from grape marc may preserve at
334
least 90% of the active compounds up to 150°C, if the processing time does not exceed 1
335
minute at this temperature.
336
17
337
5. Abbreviations and symbols
338
Abbreviations:
339
TPC
total phenol content
340
ORAC
oxygen radical absorbance capacity [µmol trolox/ g dry material]
341
DM
dry material
342
GA
gallic acid
343
PBS
phosphate buffer solution
344
TE
trolox equivalent
345
Symbols:
346
C
anthocyanin concentration
[mg gallic acid/ g dry material]
347
t
time
[min]
348
B
degradation rate parameter
[min-1]
349
B0
pre-exponential constant of B
[min-1]
350
T
temperature
[K]
351
T initial
temperature before heating treatment
[K]
352
T final
final temperature after heating treatment
[K]
353
kT
heating rate constant at corresponding T°C
[min-1]
354
Ea
energy of activation
[J/mol]
355
R
gas constant
8.314 [J/molK]
356
χ2
sum of the squares of errors
357
s
total number of experiments
[mg gallic acid/ g dry material]
358
experimental value of anthocyanin concentration
359
calculated value of anthocyanin concentration
360
nj
number of experimental data points of experiment j
18
361
N
Total number of experimental data points with the same material (grape
362
marc or liquid extract)
363
p
number of correlation parameters
364
ak
parameter in standard error calculation (Ea or B0)
365 366
Acknowledgments
367
The authors wish to thank the Junta de Castilla y León for the financial support of the
368
project VA330U13, and the Spanish Ministry of Economy and Competitiveness for the
369
Projects 356 CTQ2010-15475 and ENE2012-33613. The research group is ‘‘Grupo de
370
Excelencia GREX-11 de la Junta de Castilla y León”.
371
References
372
Amendola, D., De Faveri, D. M., & Spigno, G. (2010). Grape marc phenolics:
373
Extraction kinetics, quality and stability of extracts. Journal of Food Engineering, 97,
374
384 – 392.
375
AOAC Official Method 2005.02. (2006). Total Monomeric Anthocyanin Pigment
376
Content of Fruit Juices, Beverages, Natural Colorants, and Wines. pH Differential
377
Method. AOAC International, 37.1.68.
378
Casazza, A. A., Aliakbarian, B., Mantegna, S., Cravotto, G., & Perego, P. (2010).
379
Extraction of phenolics from Vitis vinifera wastes using non-conventional techniques.
380
Journal of Food Engineering, 100, 50 –55.
381
Cisse, M., Vaillant, F., Acosta, O., Dhuique-Mayer, C., & Dornier, M. (2009). Thermal
382
Degradation Kinetics of Anthocyanins from Blood Orange, Blackberry, and Roselle
383
Using the Arrhenius, Eyring, and Ball Models. Journal of Agricultural and Food
384
Chemistry, 57, 6285–6291. 19
385
Cruz, J.M., Conde, E., Domínguez, H., & Parajo, J.C. (2007). Thermal stability of
386
antioxidants obtained from wood and industrial wastes. Food Chemistry, 100, 1059–
387
1064.
388
Dolan, K.D. (2003). Estimation of Kinetic Parameters for Nonisothermal Food
389
Processes. Journal of Food Science, 68(3), 728 – 741.
390
Garcia-Perez, J.V., García-Alvarado, M.A., Carcel, J.A., & Mulet, A. (2010).
391
Extraction kinetics modeling of antioxidants from grape stalk (Vitis vinifera var.
392
Bobal): Influence of drying conditions. Journal of Food Engineering, 101, 49–58.
393
Guillot, F.L., Malnoë, A., & Stadler, R.H. (1996). Antioxidant Properties of Novel
394
Tetraoxygenated Phenylindan Isomers Formed during Thermal Decomposition of
395
Caffeic Acid. Journal of Agricultural and Food Chemistry, 44 (9), 2503–2510.
396
Kalt, W. (2005). Effects of Production and Processing Factors on Major Fruit and
397
Vegetable Antioxidants. Journal of Food Science, 70(1), R11 – R19.
398
Kim, S.Y., Jeong, S.M., Park, W.P., Nam, K.C., Ahn, D.U., & Lee, S.C. (2006). Effect
399
of heating conditions of grape seeds on the antioxidant activity of grape seed extracts.
400
Food Chemistry, 97, 472–479.
401
Lafka T.I., Sinanoglou, V., & Lazos, E. S. (2007). On the extraction and antioxidant
402
activity of phenolic compounds from winery wastes. Food Chemistry, 104, 1206–1214.
403
Larrauri, J.A., Rupérez, P., & Saura-Calixto, F. (1997). Effect of Drying Temperature
404
on the Stability of Polyphenols and Antioxidant Activity of Red Grape Pomace Peels.
405
Journal of Agricultural and Food Chemistry, 45, 1390-1393.
20
406
Maillard, M.N., & Berset, C. (1995). Evolution of Antioxidant Activity during Kilning:
407
Role of Insoluble Bound Phenolic Acids of Barley and Malt. Journal of Agricultural
408
and Food Chemistry, 43, 1789-1793.
409
Mishra, D.K., Dolan, K.D., & Yang L. (2008). Confidence Intervals for Modeling
410
Anthocyanin Retention in Grape Pomace during Nonisothermal Heating. Journal of
411
Food Science. 73(1), E9 - E15.
412
Nassiri-Asl, M., & Hosseinzadeh, H. (2009). Review of the Pharmacological Effects of
413
Vitis vinifera (Grape) and its Bioactive Compounds. Phytotheraphy Research, 23, 1197
414
– 1204.
415
Negro, C., Tommasi, L., & Miceli, A. (2003). Phenolic compounds and antioxidant
416
activity from red grape marc extracts. Bioresource Technology, 87, 41 – 44.
417
Patras, A., Brunton, N.P., O’Donnell, C., & Tiwari, B.K. (2010). Effect of thermal
418
processing on anthocyanin stability in foods; mechanisms and kinetics of degradation.
419
Trends Food Science and Technology, 21, 3 – 11.
420
Peralbo-Molina, A., Priego-Capote, F., & Luque de Castro, M. D. (2012). Comparison
421
of extraction methods for exploitation of grape skin residues from ethanol distillation.
422
Talanta, 101, 292–298.
423
Pinelo, M., Arnous, A., & Meyer. A.S. (2006). Upgrading of grape skins: Significance
424
of plant cell-wall structural components and extraction techniques for phenol release.
425
Trends Food Science and Technology, 17, 579 – 590.
21
426
Singleton, V.L., Orthofer, R.., & Lamuela-Raventos, R.M. (1999). Analysis of total
427
phenols and other oxidation substrates and antioxidants by means of Folin–Ciocalteu
428
reagent. Methods in Enzymology, 299, 152–178.
429
Solyom, K., Kraus, S., Mato, R.B., Gaukel, V., Schuchmann, H.P., & Cocero, M.J.
430
(2013). Dielectric properties of grape marc: Effect of temperature, moisture content
431
and sample preparation method. Journal of Food Engineering, 119, 33-39.
432
Spigno, G., & De Faveri, D.M. (2007). Antioxidants from grape stalks and marc:
433
Influence of extraction procedure on yield, purity and antioxidant power of the
434
extracts. Journal of Food Engineering 78, 793 – 801.
435
Spigno, G., Tramelli, L., & De Faveri, D.M. (2007). Effects of extraction time,
436
temperature and solvent on concentration and antioxidant activity of grape marc
437
phenolics. Journal of Food Engineering, 81, 200 – 208.
438
Wang, W.D., & Xu, S.Y. (2007). Degradation kinetics of anthocyanins in blackberry
439
juice and concentrate. Journal of Food Engineering, 82, 271–275.
440
Yue, X. & Xu, Z. (2008). Changes of Anthocyanins, Anthocyanidins, and Antioxidant
441
Activity in Bilberry Extract during Dry Heating. Journal of Food Science, 73(6), C494
442
– C499.
443
22
444
FIGURE CAPTIONS
445
Figure 1. Experimental (Exp.) and model fit (Regr.) values of the non-isothermal
446
degradation kinetics of a) grape marc and b) filtered extract anthocyanins at 80, 100 and
447
150 °C.
448
Figure 2. Required combination of time and temperature for 10 % anthocyanin decrease in
449
semi-solid grape marc and liquid extract exposed to isothermal heating.
450
Figure 3. Total phenol content of a) grape marc and b) filtered extract at 80, 100 and 150
451
°C.
452
Figure 4. Antioxidant activity of a) grape marc and b) filtered extract at 80, 100 and 150
453
°C (ORAC values).
454 455
23
456
457
TABLES
458 459
Table 1 Experimental design
460
461
Table 2. Correlation parameters of the heating process and the anthocyanin degradation
462
kinetics.
463 464
24
495
496 497
28
498 499
29
500
501 502
30
503
504 505
31
Table 1 Experimental design
Filtered extract (Same batch)
Grape marc (different batches)
Sample
Temperature (°C) 1 2 3 4 5 6
150 150 150 80 100 150
7
80
8
100
9
150
10
150
Time (min) 0; 60; 60; 60 0; 60;120;240 0; 60;120;240 0; 60;120;240 0; 60;120;240 0; 20;40 0; 20; 40; 60; 120; 240 0; 20; 40; 60; 120; 240 0; 20; 40; 60; 120; 240 0; 30; 30; 30
Table 2. Correlation parameters of the heating process and the anthocyanin degradation kinetics.
Grape marc
Parameter
Filtered extract
k80°C
min -1
0.0486
0.0006 R2 = 0.9995
0.0688
0.0014 R2 = 0.9988
k100°C
min -1
0.0615
0.0010 R2 = 0.9996
0.0851
0.0015 R2 = 0.9979
k150°C
min -1
0.0688
0.0009 R2 = 0.9995
0.1156
0.0030 R2 = 0.9952
B0
min -1
Ea
J/mol
1.67*106 55980
0.34*106 620
1.32*105 53196
1.62*104 399
514
Highlights:
515 516 517 518 519 520
• • • • •
Thermal treatments on enhanced polyphenol extraction degrades active compounds. Antioxidant degradation was studied at 80-150°C of grape marc and filtered extract Anthocyanins (AC) follow first order degradation; total phenol content and ORAC not. Degradation was 10 times faster in grape marc than in filtered extract. 90% of AC can be preserved in a pretreatment at 150°C, if no longer than 1 minute.
521 522
33
