Waste and Biomass Valorization

Valorization of colombian agroindustrial wastes: Evaluation of conventional and non-

convectional extraction techniques to obtain antioxidant extracts from cocoa bean shell
(Theobroma cacao L.).
--Manuscript Draft--

Manuscript Number: WAVE-D-18-00858R1

Full Title: Valorization of colombian agroindustrial wastes: Evaluation of conventional and non-
convectional extraction techniques to obtain antioxidant extracts from cocoa bean shell
(Theobroma cacao L.).

Article Type: Original research article

Keywords: accelerated lipid oxidation assay; Antioxidant activity; Phenolic compounds;

pressurized fluid extraction; Theobroma cacao L.

Corresponding Author: Fabián Parada-Alfonso, Ph.D.

Universidad Nacional de Colombia
Bogota, COLOMBIA

Corresponding Author Secondary

Information:

Corresponding Author's Institution: Universidad Nacional de Colombia

Corresponding Author's Secondary

Institution:

First Author: Luis Miguel Buelvas Puello, Chemical Engineer

First Author Secondary Information:

Order of Authors: Luis Miguel Buelvas Puello, Chemical Engineer

Gabriela Franco Arnedo, Chemical Engineer

Isabel C. Paz Astudillo, Chemical Engineer

José Angel Colina Marquez, Chemical Engineer

Fabián Parada-Alfonso, Ph.D.

Blanca L Ortiz-Quintero

Order of Authors Secondary Information:

Funding Information: Dirección de Investigación, Universidad Not applicable

Nacional de Colombia

Abstract: The aim of this study was to employ cocoa bean shells as a source of antioxidant
compounds, obtaining extracts by two methods (i) using high-pressure extraction with
CO2 and ethanol mixture and (ii) using Soxhlet extraction. Regarding high-pressure
extraction method the effect of the pressure and the cosolvent fraction on the
antioxidant activity of obtained extracts was evaluated using Surface Response
Methodology (SRM). The Soxhlet extraction was made with solvents of different
polarity. The antioxidant activity of extracts was studied by reduction of the radical
DPPH and the best results were: for pressurized extraction (250 bar, 12,1% ethanol),
18.8% of DPPH inhibition, and for Soxhlet extraction with ethanol with biomass
previously extracted with ethyl acetate (EtAc-EtOH), 30.5% of DPPH inhibition. The
optimized response by SRM, was 248 bar and 12.8% ethanol. The most bioactive
extracts from both methods were tested in a lipid oxidation assay by quantifying lipid
hydroperoxides and TBARS produced in edible vegetable oil. The best result was
obtained by Soxhlet extraction using EtAc-EtOH with a comparable behaviour with
synthetic antioxidant TBHQ.

Response to Reviewers: According to the revisions made by the reviewer, in relation to the statement of novelty,
we have modified and added this section within the manuscript trying not to repeat the
same information contained in the summary.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
 Previously, we had noted that there was a section in the editorialmanager to upload the
novelty statement independently, however, this section no longer appears. However,
according to the author's guide, this section should be included mainly in the
manuscript.

We want to thank for the opportunity provided to be able to share this manuscript with
you and we hope to have resolved the initial revisions. Likewise, we hope that the
contribution of our research work will open the possibility of continuing with this
research topic.

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript Click here to access/download;Manuscript;Blinded Manuscript-
2.docx

1 1
2
3
4
51 Valorization of colombian agroindustrial wastes: Evaluation of conventional and non-convectional extraction techniques to
6
72 obtain antioxidant extracts from cocoa bean shell (Theobroma cacao L.).
8
93 Abstract
10
11
12 4 The aim of this study was to employ cocoa bean shells as a source of antioxidant compounds, obtaining extracts by two
13
14 5 methods (i) using high-pressure extraction with CO2 and ethanol mixture and (ii) using Soxhlet extraction. Regarding high-
15
16 6 pressure extraction method the effect of the pressure and the cosolvent fraction on the antioxidant activity of obtained extracts
17
18 7 was evaluated using Surface Response Methodology (SRM). The Soxhlet extraction was made with solvents of different
19
20 8 polarity. The antioxidant activity of extracts was studied by reduction of the radical DPPH and the best results were: for
21
22 9 pressurized extraction (250 bar, 12,1% ethanol), 18.8% of DPPH inhibition, and for Soxhlet extraction with ethanol with
23
2410 biomass previously extracted with ethyl acetate (EtAc-EtOH), 30.5% of DPPH inhibition. The optimized response by SRM,
25
2611 was 248 bar and 12.8% ethanol. The most bioactive extracts from both methods were tested in a lipid oxidation assay by
27
28
12 quantifying lipid hydroperoxides and TBARS produced in edible vegetable oil. The best result was obtained by Soxhlet
29
30
13 extraction using EtAc-EtOH with a comparable behaviour with synthetic antioxidant TBHQ.
31
32
14
33
34
3515 Statement of Novelty
36
3716 The aim of this research work was to propose an alternative of valorisation for cocoa bean shell, an unexplored by-product
38
3917 from Colombian cocoa industrial processing. This study assesses conventional (Soxhlet) and non-convectional (Supercritical
40
4118 CO2 and ethanol as entrainer as solvents) extraction techniques for obtaining enriched- phenolic compounds extracts, which
42
4319 can be able to inhibit the oxidation of edible oils. From our best of knowledge, there is scarce information about food
44
4520 applications of cocoa bean shell extracts, resulting of interest for the natural food additive industry. The results obtained
46
4721 herein showed, for the first time, that the extracted obtained can be used to replace for synthetic antioxidants such as TBHQ,
48
4922 which is widely employed as food additive.
50
5123 Keywords: accelerated lipid oxidation assay, antioxidant activity, phenolic compounds, pressurized fluid extraction,
52
5324 Theobroma cacao L.
54
5525 Highlights:
56
5726  Antioxidant activity extracts from cocoa bean shell were obtained using green solvents.
58
5927  Cocoa bean shell extracts were using as protective agent against the lipid oxidation of an edible oil.
60
61
62
63
64
65
 1 2
2
3
4
528  The extract obtained by pressurized CO2 and ethanol with the most lipo-protective effect was obtained at the higher
6
729 level of the experimental design of pressure and ethanol percentage, these conditions allowed to obtain the best yield
8
930 of the fat fraction.
10
1131  There was a higher correlation, of nearly 90%, between total phenolic content and antioxidant activity.
12
13
14
1532 1. INTRODUCTION
16
1733 Cocoa bean shell (CBS) is an agroindustrial residue obtained from the bean in the toasting stage, or from pre-treatment with
18
1934 thermal shock [1]. It represents a problem for the industries as a large volume that holds little economic value is generated
20
2135 [2]. It has been anticipated that the generation of these residues in Colombia will increase, due to a cocoa production
22
2336 incentive, as described in the Cocoa Growers National Development Plan (2012-2021) [3]. It is therefore necessary to find a
24
2537 technique and a product with greater value than the current applications for this residue.
26
2738 The supercritical fluids extraction has been widely studied in recent years [4–6], given that it is a green technique that allows
28
2939 high purity extraction without leaving residues of organic solvents that are toxic to humans. CO2 is the most commonly used
30
3140 supercritical fluid due to its known benefits: it can be recirculated during the process, is cheap, inert when combined with
32
3341 many substances, easy to eliminate and not flammable [7]. Recently, studies that combine CO2 extraction with cosolvents,
34
3542 such as ethanol, have been carried out with the aim to improve the extraction of high-polarity compounds assuring the
36
3743 solvents used are FDA (U. S. Food and Drug Administration) approved as GRAS (Generally Recognised As Safe)
38
3944 classification [8], making the process suitable to obtain extracts which can be used as potential additives for the food
40
4145 industry. However, as mentioned by Adil et al. [9], the critical point temperature of the CO2-ethanol mixture increases as the
42
4346 proportion of ethanol does, presenting a risk of analyte degradation in some samples if one tries to work in the supercritical
44
4547 state.
46
4748 Directed to cytotoxic and antioxidant activity [2] previous studies of antibacterial activity using colombian CBS have been
48
4949 carried out [10] or also, supercritical CO2 extraction using cocoa pod husk to obtain phenolic extracts [6]. However, there are
50
51
50 still few studies about CBS as a food additive, especially as a natural antioxidant for edible oils. Therefore, this research is
52
53
51 relevant as it contributes to the development of the cacao biorefinery concept using agroindustrial residues from a biomass
54
55
52 that has great market potential at a global level to obtaining antioxidant extracts and fat fractions.
56
57
53
58
59
60
61
62
63
64
65
 1 3
2
3
4
554 2. MATERIAL AND METHODS
6
7
8
955 Chemicals and reagents
10
1156 The Folin-Ciocalteu reagent and pure grade Gallic acid were obtained from Merck®; the anhydrous sodium carbonate from
12
1357 JT Baker®; the radical 2,2 diphenyl-1- picrylhydrazyl (DPPH) and the Trolox from Aldrich®; the methanol and pure
14
1558 isooctane from Panreac®; and the tertiary butylhydroquinone TBHQ from TCI-Tokyo Kasei®.
16
17
18
59 Sample treatment
19
20
2160 The shell of Theobroma cacao L. was obtained from fermented colombian cocoa beans, that had been pre-treated with
22
2361 infrared and dehydrated, from one batch supplied by the company Casa Luker. The sample was first homogenized, ground in
24
2562 an electric grinder and then sieved to a particle size between 1,18-0,30 mm.
26
27
2863 Vacuum Soxhlet extraction
29
30
3164 Extraction was carried out over 8 hours with solvents of different polarities (benzene-Ben, ethylacetate-EtAc, ethanol-EtOH),
32
3365 using a vacuum system to reduce the temperature of the extraction to avoid degradation of possible thermosensitive analytes.
34
3566 Figure 1 shows the conditions and sequence in which the 6 extracts were obtained. Two formation phases, soluble and
36
3767 insoluble, were observed, separated through centrifugation using a Hettich Universal 320R centrifuge. The yield obtained in
38
3968 each phase was quantified and the extracts were stored in darkness at -20°C until the moment of their use.
40
41
4269 Extraction with pressurized mixture of CO2 - ethanol
43
44
70 The equipment (Figure 2) consisted of a 16 mL 316 stainless steel extraction cell, covered with an electric heat jacket, which
45
46
4771 connected to a PID temperature controller. The CO2 was pressurized through a diaphragm pump, connected to a frequency
48
4972 regulator. The pipes were made of 316 stainless steel with a diameter of 1/8”. To avoid the formation of carbonic snow at the
50
5173 moment of batch discharge, the pipes are heated using coated resistance wire. This resistance is connected to a PID
52
5374 temperature controller. In the discharge zone, a BPR valve (pressure regulator) is installed. Given that the discharge rate of
54
5575 the CO2 can carry the extract with it, two collection recipients were installed in series. The ethanol was pumped using a 140A
56
5776 model HPLC Beckman pump.
58
59
60
61
62
63
64
65
 1 4
2
3
4
577 The extraction cell was filled with approximately 20 g of the sample and test conditions were set for the extraction, which
6
778 was carried out in batch mode. A factorial central composite experimental design 22 was implemented to observe the effect of
8
979 adding ethanol as a cosolvent at 96%, whilst maintaining a constant temperature of 60°C. The cosolvent percentage (6,4%-
10
1180 12%) and the pressure of the extraction vessel (150-250 bar) were varied to obtain a total of 13 experimental points, using 5
12
1381 replicates in the central point of the experimental design.
14
1582 The yield of each extract was determined (mg extract/g dry biomass), considering the two obtained phases: ethanolic phase
16
1783 (assumed polar) and insoluble phase (assumed non-polar), which were separated following the methodology implemented in
18
1984 the soxhlet extraction. A batch type extraction was performed on 21 batches at 10 minutes each, for a total extraction time of
20
2185 210 minutes, which, according to previous tests was enough time to exhaust the sample.
22
23
24
2586 Total phenolic content
26
2787 The total phenolic content (TPC) was quantified using the Folin–Ciocalteu method (F-C) according to the methodology
28
2988 shown by(Carrillo et al., [11] with modifications. Gallic acid (G.A.) was used as a standard, drawing from a stock solution.
30
3189 100 μL of 2 mg/mL extract was mixed with 750 μL Folin–Ciocalteu reagent (10%), shaken, and left to rest for 5 minutes in
32
3390 darkness. Then 750 μL of anhydrous Na2CO3 solution (6%) was added, shaken and left for 90 minutes in darkness while it
34
3591 reacted. The absorbance of the blue colour was measured as 765 nm in a UV-Vis Thermo-Scientific Genesys 10
36
3792 spectrophotometer. The total phenolic content is expressed in milligrams equivalent to gallic acid per gram of dry sample
38
3993 material (mg-eq GA/g dry matter). All measurements were made in triplicate. The test was carried out for both the Soxhlet
40
4194 extracts and the pressurized solvent extracts.
42
43
44
4595 Antioxidant activity tests (AA)
46
4796 Two tests were carried out. DPPH (2,2 diphenyl-1- picrylhydrazyl) reduction assay, which measures the capacity of the
48
4997 sample to trap free radicals, was conducted as a screening test to carry out the second test. Then, an evaluation of the most
50
5198 active extracts identified, from both the Soxhlet and pressurized solvent extraction, was carried out to measure the level of
52
5399 protection against lipid oxidation in an edible vegetable oil.
54
55
56
57
58
59
60
61
62
63
64
65
 1 5
2
3
4
5
100 2.6.1. DPPH (2,2 diphenyl-1- picrylhydrazyl) radical scavenging assay
6
7
101 For this test the methodology from(Carrillo et al.. was used with modifications [11]. The test was carried out with 975 μL of a
8
9
102 0.1 mM DPPH solution, which was prepared in methanol, mixed with 25 μL of 2 mg/mL of either the extract, or the standard,
10
11
103 in methanol. As a reference, the same amount of DPPH and 25 μL of solvent (methanol) was used. The mixtures were shaken
12
13
104 and left to react at room temperature in darkness for 30 minutes. The absorbance was registered as 517 nm in the
14
15
105 spectrophotometer. The results were expressed as a percentage inhibition (as shown in equation 1), and TEAC values (μmol
16
17
106 of trolox/g of dry matter) were expressed through the construction of a calibration curve using several concentrations of the
18
19
107 antioxidant Trolox.
20
21 𝐴𝑏𝑠𝑡=0 − 𝐴𝑏𝑠𝑡=30
22
108 %𝐼 = ∗ 100 (1)
23 𝐴𝑏𝑠𝑡=0
24
25
26
109 2.6.2. Accelerated oxidation test for commercial vegetable oil
27
28
110 Edible mixed vegetable oil was used, which according to Team Foods, the company who provided the sample, is highly
29
30
111 consumed in Colombian households. It contains a mixture of soy oil (principally linoleic and oleic acid [12]; and a smaller
31
32
112 proportion of palm olein, principally palmitic and oleic acid) [13].
33
34
113 The procedure used by(Castro-Vargas et al. was followed [14]. The most active extracts, obtained from both extraction
35
36
114 methods were reconstituted at 8 mg/mL in ethanol as well as the standard solution, synthetic antioxidant, tert-
37
38
115 Butylhydroquinone (TBHQ). An oil with 200 mg extract/kg oil was prepared, in accordance with the concentration of
39
40
116 antioxidants recommended in the codex alimentarius for fats and oils [15]. Finally, the oxidation accelerator, FeCl2. 4H2O,
41
42
117 was added to obtain a 3.5 ppm concentration of iron in the oil. Bubbling air was applied for 5 minutes every 3 hours, four
43
44
118
45 times a day for six days at a constant temperature of 70°C. A sample of the oil was taken on day zero and then every 48
46
119
47 hours. The oxidation kinetics were measured by quantifying the total hydroperoxides (HPL) and a test of thiobarbituric acid
48
120
49 reactive species (TBARS).
50
121
51 Quantification of hydroperoxide lipids (HPL)
52
122
53 A sample of 50 mg of oil with no antioxidants added was taken and 500 µL of isooctane was added to it. Then a 20 µL
54
123
55 aliquot of this solution was taken and 2 mL of isooctane were added. The calibration curve was made with linoleic acid. The
56
124
57 absorbance was recorded as 234 nm, where the maximum absorption peak of linoleic acid (cis-trans) is registered, using
58
125
59 isooctane as blank. The results were expressed in mmol of linoleic acid/kg oil.
60
61
62
63
64
65
 1 6
2
3
4
5
126 Species reactive to thiobarbituric acid (TBARS)
6
7
127 A 50 mg sample of oil was taken and the procedure shown in (Castro-Vargas et al. was followed [14]. The calibration curve
8
9
128 was made with 1,1,3,3-tetraethoxypropane (TEP), with an absorbance reading of 532 nm. The results were expressed in
10
11
129 mmol of malondialdehyde (MDA) / kg oil.
12
13
14
130
15 Statistical Analysis
16
131
17 The soxhlet yield, percentage inhibition of DPPH, lipid oxidation assay and TPC results are expressed as averages ± SD
18
132
19 (standard deviation) for n=3. Averages with different letters were significantly different in the level of p<0,05 according to
20
133
21 the Tukey test carried out using R Project software. For the pressurized fluid extraction, a multifactorial ANOVA was carried
22
23
134 out. A surface response analysis was used to choose the most active extracts of the experimental design for pressurized
24
25
135 solvent extraction. This data was then adjusted to a statistical model using Statgraphics Centurion XVI software, to maximise
26
27
136 TPC.
28
29
30
137
31 3. RESULTS AND ANALYSIS
32
33
34
138 Yield extraction
35
36
139 In Figure 3 the results from the Soxhlet extraction are shown. The most representative extracts in the soluble phase were
37
38
140 EtOH and benzine, while the benzine extract was found to be among the lowest in the insoluble phase yield (Figure 3a y 3b).
39
40
141 This owes to the fact that benzine only solubilizes substances with low polarity as it is assumed that the insoluble phase is
41
42
142 mostly fat, which prevents the formation of the two phases. On the contrary, the extract with ethanol allows compounds of
43
44
143 low and medium polarity as its dipolar moment promotes formation in both phases. A progressive decrease in total yield of
45
46
144 Ben, Ben-EtAc and Ben-EtAc-EtOH was observed due to sample exhaustion in the successive extractions.
47
48
145 The Figure 4 shows the yields from extraction using the pressurized mixture CO2/ethanol (CO2-EtOH). Extract 9 (200 bar
49
50
146
51 and 13%) represents the extract with the highest total yield (5.8%), and at the same time the highest yield in the ethanolic
52
147
53 phase (3.6%). The highest insoluble phase (assumed fat) yield (2.5%) was obtained at 270 bar and 9% of ethanol (extract 4),
54
148
55 which indicates that the fat extraction is improved at high pressures. On the other hand, pressurized extraction allows for
56
149
57 higher yields of fat extraction compared with Soxhlet, however, it is only possible to recuperate a maximum of 55% of the
58
59
60
61
62
63
64
65
 1 7
2
3
4
5
150 yield in the polar phase and 74% of the total yield, compared to polar and total phase yield of the ethanol Soxhlet,
6
7
151 respectively.
8
9
10
152
11 Total Phenolic Content
12
153
13 The soxhlet extract with the largest quantity of phenols was EtAc-EtOH, followed by EtOH y Ben-EtAc-EtOH, mainly the
14
154
15 ethanolic extracts, which indicates that this high polarity solvent is more selective with the analytes of interest (Table 1). It is
16
155
17 also believed that the cleaning of the sample with a medium polarity solvent, such as ethyl acetate, for the fat extraction could
18
156
19 have favoured ethanol access to the polyphenols, compared to the biomass extracted with only ethanol.
20
157
21 In the pressurized extracts, extract 6 stood out (250 bar, 12% of ethanol) with 44.433 mg-eq AG/g extract. In comparison, the
22
23
158 results obtained through the soxhlet extraction were lower, because the latter technique uses a larger quantity of solvent,
24
25
159 which, constantly renewed, increases its capacity to transfer the mass matrix to the solvent. In addition to this, the contact
26
27
160 time used was longer which also increases extraction capacity. However, for this extract recovery rates reached 88% in half
28
29
161 the time of extraction, compared to EtAc-EtOH extract, which had the greatest TPC of all the extracts.
30
31
32
162
33 DPPH radical scavenging assay
34
163
35 The percentage inhibition and oxidation of the radical DPPH of the soxhlet extracts were compared with the standard (Table
36
164
37 1) at the same concentration as the extracts. The standard almost totally reduced the radical, followed by the extracts EtAc-
38
165
39 EtOH, EtOH and Ben-EtAc-EtOH. These results coincided with the extracts with greatest TPC and indicates that there is a
40
41
166 correlation between TPC and antioxidant activity. The results obtained for the DPPH of the pressurized extracts showed
42
43
167 extract 6 as the most bioactive one, with a value of 14.8%.
44
45
46
168 3.3.1. TPC and AA Correlation
47
48
169
49 The results in Figure 5 show a lineal correlation between AA and the TPC in the pressurized extracts. This indicates that,
50
170
51 preliminarily, the polyphenols quantified through the Folin-Ciocalteu method were responsible for inhibiting the oxygen.
52
171
53 These results agreed with those obtained in other reports [16], which indicates that a larger the number of phenols can inhibit
54
172
55 in a powerful way the oxidation. This is might be due to their impact in breaking the chain reaction of the free radicals by
56
173
57 donating an atom to the free hydrogen. From this, it can be deduced that extracted compounds can have a reductive capacity.
58
59
60
61
62
63
64
65
 1 8
2
3
4
5
174 94.3% of the experimental results fitted with the linear model, with a coefficient correlation of 88%, which confirms that
6
7
175 there is a correlation between the polyphenols identified through this technique and antioxidant activity.
8
9
10
176
11 Statistical analysis of the pressurized extraction
12
177
13 According to the analysis of variance (Table 2), in the experimental design variable responses, like total phenol content and
14
178
15 antioxidant activity, have a p-value less than 0.05, which indicates that all the results obtained are different to one another for
16
179
17 each of the effects studied by experimental design. However, the results of the p-value for ethanolic yield and fat yield are
18
180
19 slightly higher than 0.05. This is probably because there was a small sample extraction (20 g), and as a result the extracts
20
181
21 obtained were also small, meaning that at the moment of separation of the phases, small quantities of fat were not eliminated
22
23
182 due to ethanol saturation, and as such, contributed to the yield in the solvent phase meaning the results had a statistical
24
25
183 similarity.
26
27
28
184 3.4.1. Non-linear regression for experimental results
29
30
185
31 For the statistical model the effect of every considered variable and the interaction between them was observed. Figure 6
32
186
33 indicates the standardized effect of the independent variables on every one of the responses. The variable with the biggest
34
187
35 effect was the cosolvent percentage, which suggests that polyphenol extraction is mostly favoured by the addition of a polar
36
188
37 solvent, probable due to its affinity with analytes of interest within the matrix, a phenomenon previously observed in soxhlet
38
189
39 extraction.
40
190
41 However, the pressure has a significant effect on the total polyphenol content (Figure 6a), which suggests that minor changes
42
191
43 in pressure influence the power of solvation of the pressurized CO2 -ethanol mix. In other words, the increase in pressure
44
192
45 favoured the entry of the extractor agents in the matrix of the shell. It has been well demonstrated that the increase in
46
193
47 cosolvent percentage increases fat extraction [17, 18], however, the modification made by the use of ethanol on the constant
48
194
49 dielectric of the mixture promote extraction of compounds with higher polarity [19], like phenols, which explains why at
50
195
51 higher pressure and cosolvent percentage higher total yields and TPC yield are achieved.
52
53
54
196 3.4.2. Surface response methodology
55
56
197
57 For the optimization process, the TPC variable was chosen. This is principally justified by the existence of a proportional
58
198
59 correlation between the TPC and AA. The optimum point was estimated at 247.6 bar and 12.8%, obtaining likely TPC value
60
61
62
63
64
65
 1 9
2
3
4
5
199 of 1.271 mg-eq AG/g dry biomass with a polar phase yield of 3.1%. This is consistent with the effects of the variables
6
7
200 throughout the study: more pressure, better matrix penetration and more cosolvent, higher polarity and affinity with the
8
9
201 desired compounds.
10
11
202 TPC = 3.37882 − 0.02450(P) − 5.79584 (X) + 0.00006(P 2 ) + 10.95740 (𝑋 2 ) + 0.00821 (PX)
12
13
203 𝐸𝑐. (2)
14
15
204
16
17
205 Figure 7 shows the surface response obtained from the experimental TPC values, adjusted to the statistic model of the
18
19
206 equation (2). The model explains 77.66% of the effects generated by the manipulated variables. This model did not
20
21
207 correspond more than 90% for various possible factors, specifically because (i) the TPC values per gram of CBS were
22
23
208 calculated from the ethanolic yield, which deviates from the real figures because the fat could not completely separate in the
24
25
209 soluble phase, generating an increase in the weight of all the extracts and causing statistical similarities between them, as
26
27
210 indicated in the p-value in Table 2 for ethanolic yield; (ii) during the pressurized extraction phase analytes were lost during
28
29
211 decompression due to the pull from the flow of CO2.
30
31
32
33
212 Lipid Oxidation
34
35
213 The best extracts of two methodology of extraction were selected according with the most DPPH inhibition. In evaluating the
36
37
214 protection of the extracts in edible oil against lipid oxidation, hydroperoxide formation(Figure 8) and TBARS Figure 9 were
38
39
215 measured. A decrease in the formation of these compounds was noted, which could indicate that the antioxidants interrupted
40
41
216 the propagation phase of self-oxidation. In the final two days a decline in hydroperoxide formation was noticed, even in the
42
43
217 oil without antioxidants, indicating a state of primary oxidation on completion [20].
44
45
218 All the extracts, Soxhlet and pressurized, presented reductions in hydroperoxide formation (Table 3), which shows that, to
46
47
219 some extent, that they inhibit the generation of hydroperoxide. However, on the second day EtAc-EtOH presented greater
48
49
220 inhibition in relation to all test agents. It is probable that its good performance was reduced due to the depletion of
50
51
221 antioxidants in the sample.
52
53
222
54 About to the TBARS assay (Table 4), which measures the formation of the final products of oxidation, during the experiment
55
223
56 all extracts presented lower values than the oil sample without antioxidants, confirming, to some degree, that the extracts
57
224
58 deter oxidation, avoiding the formation of these final compounds. This affirms that the extracts have the capacity to inhibit
59
60
61
62
63
64
65
 1 10
2
3
4
5
225 oxidation. Again, the EtAc-EtOH extract showed most activity out of all the extracts on day 2 and a relative inhibition of
6
7
226 75% at day 6. The pressurized extract also had a relative inhibition of 71% in relation to TBHQ.
8
9
227 In addition to this, the test for accelerated lipid oxidation gives a closer approximation to reality when compared to the test
10
11
228 for the capture of radical DPPH, as in the radical DPPH test the reference antioxidants presented an almost total radical
12
13
229 reduction behaviour and in the oil the behaviour was similar compared to the extracts. This is probably because the test with
14
15
230 oil allows the evaluation of possible synergic phenomena occurring between the extract and the oil, which is not the case with
16
17
231 the radical DPPH, a synthetic species not found naturally in live organisms or foods.
18
19
20
232
21 4. CONCLUSION
22
233
23 It can be concluded that with the techniques used in this study, analytes which have similar behaviour in terms of oxidation
24
234
25 inhibition were obtained. It also reconfirms that the addition of ethanol is the most influential variable in the study in terms of
26
235
27 its capacity to extract analytes with antioxidant activity. The extracts with highest TPC and antioxidant activity were the
28
236
29 Soxhlet extracts EtAc-EtOH and EtOH and the pressurized extracts 4 (270 bar and 9% EtOH) and 6 (250 bar and 12.3%
30
237
31 EtOH), ethanol being the variable which had most influence. Furthermore, a lineal correlation was discovered between the
32
238
33 CTF and AA of the extracts obtained. Through the analysis of the surface response it was possible to determine optimal
34
35
239 extraction conditions: 60°C, 247.6 bar and 12.8%, obtaining a value of TPC 1.271 mg-eq AG/g dry biomass and a yield of
36
37
240 5.9%.
38
39
241 Finally, it was accomplished that it is possible to obtain extracts from Colombian cocoa bean shells with antioxidant activity
40
41
242 comparable to those of commercial antioxidants, which have been widely used on a commercial scale (TBHQ).
42
43
243 It is recommended that other experiments using non-conventional methods of extraction be carried out with an apolar
44
45
244 dissolvent to withdraw larger quantities of fat from the sample to improve extraction of polar compounds later. Furthermore,
46
47
245 a sensorial analysis is recommended to verify if there are changes to the organoleptic properties of the oil on applying these
48
49
246 kinds of extracts. For future research, a synergic study into the extracts and commercial antioxidants could be carried out in
50
51
247 order to evaluate the effect that the combination of these two substances have on the oxidation protection index. In addition
52
53
248 to this, is needed to identify a potential use for the fat obtained during the pressurized extraction.
54
55
56
57
58
59
60
61
62
63
64
65
 1 11
2
3
4
5
249 5. ACKNOWLEDGEMENTS
6
7
250 The authors would like to thank the companies Casa Luker for providing the cocoa bean shells and the Team Foods Company
8
9
251 for providing oil without antioxidants. As well as the Research Division of the Bogotá headquarters of the National
10
11
252 University of Colombia for the approval of project 37479 through "NATIONAL PROJECT CALL FOR STRENGTHENING
12
13
253 OF THE RESEARCH, CREATION AND INNOVATION OF THE UNIVERSITY NATIONAL OF COLOMBIA 2016-
14
15
254 2018 "
16
17
18
255
19 6. REFERENCES
20
256
21 1. Wollgast, J., Anklam, E.: Review on polyphenols in Theobroma cacao : changes in composition during the
22
23
257 manufacture of chocolate and methodology for identification and quantification. 33, (2000)
24
25
258 2. Arlorio, M., Coisson, J.D., Travaglia, F., Varsaldi, F., Miglio, G., Lombardi, G., Martelli, A.: Antioxidant and
26
27
259 biological activity of phenolic pigments from Theobroma cacao hulls extracted with supercritical CO 2. Food Res.
28
29
260 Int. 38, 1009–1014 (2005). doi:10.1016/j.foodres.2005.03.012
30
31
261 3. Ministerio de Agricultura y Desarrollo Rural: Plan de Desarrollo Cacaotero 2012-2021. (2012)
32
33
262 4. Herrero, M., Castro-Puyana, M., Mendiola, J.A., Ibañez, E.: Compressed fluids for the extraction of bioactive
34
35
263 compounds. TrAC - Trends Anal. Chem. 43, 67–83 (2013). doi:10.1016/j.trac.2012.12.008
36
37
264 5. Leng, Y., Gan, S., Morris, A., Kiat, H.: Ethyl lactate as a potential green solvent to extract hydrophilic ( polar ) and
38
39
265 lipophilic ( non-polar ) phytonutrients simultaneously from fruit and vegetable by-products. Sustain. Chem. Pharm. 4,
40
41
266 21–31 (2016). doi:10.1016/j.scp.2016.07.003
42
43
267 6. Valadez-carmona, L., Ortiz-moreno, A., Ceballos-reyes, G., Mendiola, J.A., Ibáñez, E.: The Journal of Supercritical
44
45
268 Fluids Valorization of cacao pod husk through supercritical fl uid extraction of phenolic compounds. J. Supercrit.
46
47
269 Fluids. 131, 99–105 (2018). doi:10.1016/j.supflu.2017.09.011
48
49
270 7. Lee, J.W., Fukusaki, E.: Application of supercritical fluid carbon dioxide to the extraction and analysis of lipids.
50
51
271 Bioanalysis. 4, 2413–2422 (2012). doi:10.4155/bio.12.198
52
53
272 8. Food and Drug Administration: Direct food substances affirmed as Generally Recognized As Safe. (2007)
54
55
273 9. Adil, I.H., Çetin, H.I., Yener, M.E., Bayindirli, A.: Subcritical (carbon dioxide + ethanol) extraction of polyphenols
56
57
274
58 from apple and peach pomaces, and determination of the antioxidant activities of the extracts. J. Supercrit. Fluids. 43,
59
275
60 55–63 (2007). doi:10.1016/j.supflu.2007.04.012
61
62
63
64
65
 1 12
2
3
4
5
276 10. Cuéllar, O., Guerrero, G.: Actividad antibacteriana de la cáscara de cacao, Theobroma cacao L. Rev. MVZ Córdoba.
6
7
277 17, 3176–3183 (2012)
8
9
278 11. Carrillo, L.C., Londoño-Londoño, J., Gil, A.: Comparison of polyphenol, methylxanthines and antioxidant activity in
10
11
279 Theobroma cacao beans from different cocoa-growing areas in Colombia. Food Res. Int. 60, 273–280 (2013).
12
13
280 doi:10.1016/j.foodres.2013.06.019
14
15
281 12. Jaselskis, B., Stemm, N.L., Johnston, W.D.: Determination of the fatty-acid composition of soybean oil by high-
16
17
282 pressure liquid chromatrography. Talanta. 29, 54–56 (1982)
18
19
283 13. Martínez Ávila, O.M., Sánchez Castellanos, J., Suárez Palacios, O.Y.: Ethyl ester production from (RBD) palm oil.
20
21
284 Rev. Ing. e Investig. 27, 34–43 (2007)
22
23
285 14. Castro-Vargas, H.I., Baumann, W., Parada-Alfonso, F.: Valorization of agroindustrial wastes: Identification by LC-
24
25
286 MS and NMR of benzylglucosinolate from papaya (Carica papaya L.) seeds, a protective agent against lipid
26
27
287 oxidation in edible oils. Electrophoresis. 1–8 (2016). doi:10.1002/elps.201500499
28
29
288 15. Food and Agricultural Organization: SECTION 1. Codex General Standard for Fats and Oils
30
31
289 16. Azizah, A.H., Ruslawati, N.M.N., Tee, T.S.: Extraction and characterization of antioxidant from cocoa by-products.
32
33
290 Food Chem. 64, 199–202 (1999)
34
35
291 17. Salajegheh, D., Vaziri, A., Bastani, D.: Supercritical extraction of cocoa butter from cocoa seed, using pure carbon
36
37
292 dioxide, carbon dioxide with ethanol as co-solvent and ethane. Middle East J. Sci. Res. 13, 1010–1015 (2013).
38
39
293
40 doi:10.5829/idosi.mejsr.2013.13.8.3763
41
294
42 18. Tan, T.-J., Jinap, S., Kusnadi, A.E.D.I., Hamid, N.: Extraction of Cocoa Butter By Supercritical Carbon Dioxide :
43
295
44 Optimization of Operating Conditions and Effect of Particle Size. J. Food Lipids. 15, 263–276 (2008)
45
296
46 19. Wesch, A., Dahmen, N., Ebert, K..: Measuring the static dielectric constants of pure carbon dioxide and carbon
47
297
48 dioxide mixed with ethanol and toluene at elevated pressures. Berichte der Bunsengesellschaft für Phys. Chemie.
49
298
50 100, 1368–1371 (1996). doi:10.1073/pnas.0703993104
51
299
52 20. Wasowicz, E., Gramza, A., Hes, M., Jelen, H.H., Korczak, J., Malecka, M., Mildner-Szkudlarz, S., Rudzinska, M.,
53
300
54 Samotyja, U., Zawirska-Wojtasiak, R.: Oxidation of lipids in food. J. Food Nutr. Sci. 13, 87–100 (2004)
55
301
56
57
58
59
60
61
62
63
64
65
 1 13
2
3
4
5
6
7. TABLES AND FIGURES
7
8
9 Table 1. Summary of the yield results, total phenol content and antioxidant activity of the extracts obtained with CO 2-ethanol and
10 Soxhlet extraction. Different letters indicate a different significance (p<0.05) for each extraction technique according to Tukey test.
11
12
13
14 Point Pressure (bar) Cosolvent Solvent Fat Total TPC DPPH DPPH (µmol
15 (%p/p) Yield Yield Yield (mg eq-G. (% de Trolox/g
16
(%) (%) (%) A/g extract) inhibition) extract)
17
1 150.0 12.1 2.3 2.4 4.7 36.801 ± 13.3 ± 0.6cd 88.165 ± 4.053cd
18
3.788d
19
20 2 200.0 5.2 1.4 1.0 2.4 24.190 ± 7.0 ± 0.4a 42.647 ± 3.092a
21 2.859a
22 3 150.0 6.4 1.3 1.5 2.9 25.770 ± 9.2 ± 0.2ab 58.403 ± 1.715ab
23 2.514ab
24 4 270.7 9.3 1.9 2.5 4.4 36.860 ± 14.8 ± 0.2d 99.370 ± 1.715d
25 2.756d
26 5 200.0 9.3 2.6 1.7 4.2 31.881 ± 12.8 ± 0.2bcd 85.014 ± 1.310bcd
27 0.304bd
28 250.0 2.7 2.4 5.1 44.433 ± 18.9 ± 0.5e 128.782 ± 3.930e
6 12.1
29 0.637e
30
7 200.0 9.3 1.1 2.2 3.3 31.911± 10.5 ± 0.8ac 68.207 ± 5.447ac
31
0.263bd
32
33 8 129.3 9.3 2.5 1.0 3.5 36.890 ± 12.4 ± 1.5bcd 81.863 ±
34 0.360d 10.856bcd
35 9 200.0 13.2 3.6 2.2 5.8 33.939 ± 11.2 ± 0.6bcd 73.109 ± 4.538bcd
36 0.446cd
37 10 200.0 9.3 1.2 2.4 3.6 30.689 ± 10.7 ± 0.4ac 69.258 ± 3.247ac
38 1.257bd
39 11 200.0 9.3 1.1 2.0 3.1 29.258 ± 10.9 ± 0.1bcd 71.359 ± 0.495bcd
40 0.112abc
41 12 250.0 6.4 1.8 2.0 3.7 27.916 ± 9.6 ± 0.4ac 61.555 ± 2.573ac
42 0.345abc
43
13 200.0 9.3 2.2 2.0 4.2 30.212 ± 10.7 ± 1.9bcd 69.608 ± 3.768bcd
44
0.292abc
45
46 Benzine 5.1  0.6 0.3  0.1 5.4  0.7 3.411  2.2  0.5a 7.983  3.431 a
47 1.610a
48 Ben-EtAc 4.2  1.0 0.4  0.2 4.5  1.1 11.281  7.3  0.1b 44.748  0.858 c
49 0.788b
50 Ben-EtAc-EtOH 2.4  0.4 0.1  0.0 2.5  0.4 35.996  21.0  1.5c 143.838  10.481
51 2.791c d

52 EtAc 4.0  0.2 0.5  0.3 4.6  0.5 6.809  4.8  0.6ab 27.241  4.401 b
53 1.160ab
54 EtAc-EtOH 4.1  0.4 0.2  0.1 4.3  0.6 50.276  30.8  1.3d 214.566  9.090 e
55 3.627d
56 EtOH 6.6  0.7 1.2  0.3 7.8  1.1 37.307  22.6  0.2c 155.742  1.785 d
57
1.795c
58
59
TBHQ - - - - 96.7  0.1
60
61
62
63
64
65
 1 14
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23 Table 2 Analysis of Variances for results obtained through pressurized extraction. Taken from data produced by Statgraphics
24
25 software.
26
27
28 Polar Phase Non-Polar TPC (mg-eq Antioxidant Activity
Model Total Yield
29 Yield Phase Yield GA/g extract) (µmol Trolox/g extract)
30 P-valor 0.1676 0.1026 0.0060 0.0038 0.0080
31 Error d.f. 7 7 7 7 7
32 Stand.
33 0.006 0.004 0.005 2.450 10.645
error
34 R2 61.01 67.04 86.47 88.20 85.23
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 15
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23 Table 3. Kinetics of total hydroperoxide production expressed in mmol malondialdehyde/ kg oil. Averages with different numbers
24
25 were significantly different at level of p-value according to Tukey test. The test was realized for three replicates.
26
27 Day 0 Day 2 Day 4 Day 6 %inhibition
28 Day 6
29
30 EtAc_EtOH 37.987 37.479±3.358a 120.367±0.424b 163.248±5.888c 52%
31 EtOH 37.987 42.868±3.764 a
141.646±8.457 c
165.446±9.593cd 48%
32
33 270.7 bar, 9.3% 37.987 58.450±1.321b 152.496±5.634c 181.670±8.448de 20%
34 250.0 bar, 12.1% 37.987 56.795±2.304b 144.767±5.949c 172.284±2.791cd 36%
35 TBHQ 37.987 39.216±0.200a 124.733±0.143b 134.766±5.901b
36 100%
37 Oil CONTROL 37.987 85.873±6.176c 175.126±5.270d 193.748±7.006e 0%
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 16
2
3
4
5
6
7
8
9
10
11
12
13 Table 4. Kinetics of production of species reactive to thiobarbituric acid expressed in mg MDA/ kg oil. Averages with different
14
15 letters were significantly different to the level of the p-value according the Tukey test. The test was realized for three replicates.
16
17 Day 0 Day 2 Day 4 Day 6 Relative
18 inhibition
19 Day 6
20 EtAc_EtOH 178.419 218.780±21.433a 687.544±29.343b 1190.217±109.985bc 75%
21 EtOH 178.419 292.940±26.834a 822.109±60.999bd 1201.930±115.885c
22 73%
23 270.7 bar, 9.3% 178.419 447.076±9.929b 964.280±46.387d 1442.681±99.689d 46%
24 250.0 bar, 12.1% 178.419 447.572±36.131b 881.800±10.259cd 1225.523±55.281cd 71%
25 178.419 323.600±30.727ab 754.991±77.101bc 967.097±62.149b
TBHQ 100%
26
27 Oil CONTROL 178.419 1158.041±105.149c 1479.649±63.235e 1846.733±48.656e 0%
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 17
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25 Figure 1 Diagram showing collection of Soxhlet extracts.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 18
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21 Figure 2 Flow chart of pressurized extraction equipment. The standard setup is: 1. CO2 reservoir, 2a. CO2 inlet manometer, 2b.
22
23 CO2 outlet manometer, 3a. Cell Jacket temperature controller, 3b. Pipeline heater temperature controller, 4. Compressor, 5. CO 2
24
25 inlet flow valve, 6. Ethanol inlet flow valve, 7. HPLC type pump, 8. Ethanol reservoir, 9. Extraction cell, 10. Flow valve, 11. Back
26
27 Pressure Regulator (BPR) valve, 12. Extract collectors
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 19
2
3
4
5
8.0 2.0
6 c

Yield [%- g insoluble/100 g

Yield [%- g soluble/100 g

7.0 bc 1.8 b
7 b 1.6
6.0 1.4
8 5.0 ab ab
dry matter]

1.2 ab

dry matter]
9 4.0 1.0
10 3.0
a 0.8
0.6 a a a
11 2.0
0.4
12 1.0 0.2 a
13 0.0 0.0
Benzine Ben - Ben - EtAc EtAc - EtOH Benzine Ben - Ben - EtAc EtAc - EtOH
14 EtAc EtAc - EtOH EtAc EtAc - EtOH
15 EtOH EtOH
16 Extracts (a) Extracts (b)
17
18
19 Figure 3 Yield from soxhlet extraction in (a) soluble y (b) insoluble phase with error bars. Different letters indicate significant
20
21 differences with p<0.05 according to a Tukey analysis. The test was realized for three replicates.
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 20
2
3
4
5
6 Ethanolic Yield
6 Insoluble Phase Yield
7 Total Yield
8
5
9
10
4
11
Yield %

12
3
13
14
15 2
16
17 1
18
19 0
200 150 250 129,3 200 270,7 150 250 200 P [Bar]
20
5,2 6,4 9,3 12,1 13,2 Co [%]
21
22
23
24 Figure 4 Fat and ethanolic phases yields and Total yields obtained in the extraction with CO2-EtOH.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 21
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 Figure 5 Correlation between Total Phenolic Content (TPC) and Antioxidant Activity of pressurized extracts. Taken from
23
24 data produced by Statgraphics software.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 22
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27 Figure 6 Standardized Pareto chart of results obtained from the experimental design. The grey bar indicates increase or an
28
increase above the response variable. Taken from data produced by Statgraphics software. A: Pressure B: Cosolvent
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 23
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44 Figure 7 Estimated surface response with Pressure and Cosolvent vs Total Phenolic Content (TPC). Taken from data produced by
45
46 Statgraphics software.
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 24
2
3
4
5
6
7
8
9
250
10
11
12
mmol Linoleic Acid/ kg oil

200
13
14 EtAc_EtOH
15
150 EtOH
16
17 270.7 bar, 9.3%
18 250.0 bar, 12.1%
19 100
20 TBHQ
21 Oil CONTROL
22 50
23
24
25 0
26 Day 0 Day 2 Day 4 Day 6
27
28 Figure 8. Kinetics of oil oxidation through quantification of total hydroperoxides.
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 25
2
3
4
5
6
7
8
9
2,000.000
10
11 1,800.000
12
13 1,600.000
14 1,400.000 EtAc_EtOH
mg MDA/ kg oil

15
16 1,200.000 EtOH
17 270.7 bar, 9.3%
18 1,000.000
19 250.0 bar, 12.1%
800.000
20 TBHQ
21 600.000
22 Oil CONTROL
400.000
23
24 200.000
25
26 0.000
27 Day 0 Day 2 Day 4 Day 6
28
29 Figure 9 Kinetics of oil oxidation through the quantification of species reactive to thiobarbituric acid.
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
 1 26
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31 Figure 10 Graphical Abstracts
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
Fig 1 Soxhlet Extraction.eps Click here to access/download;Figure;Fig 1 Soxhlet Extraction.eps

Solvent Polarity
- +
Petroleum Ethyl
Ethanol Extract 3
Benzine Acetate
Solvent Polarity

Extract 1 Extract 2
Biomass

Ethyl Ethanol
Acetate
Extract 4 Extract 5

Ethanol Extract 6

+
Fig 2 Extraction System.eps Click here to access/download;Figure;Fig 2 Extraction System.eps

Ethanol
Convention
1. CO2 Reservoir
Preheat 2. Pressure indicators
7 8
2.a CO2 in to pump
5 6 2.b CO2 out of pump
PI

1 PI PI
2.c 2.c CO2 In Extraction Cell
3.Temperature Controller
TC
Temperature Controller
3.a Cell Jacket
2.a 2.b 3.a 3.b Preheat and output line.
Cell 3.b 4. Compressor Pressure Control
Extraction
9
CO2 5. CO2 valve access
Outlet CO2 6. Inlet valve ethanol
7. Type piston pump
12 8. Ethanol reservoir
4
BPR 9. Cell Extraction heating jacket
11 10. Exhaust outlet valve
10 11. Control valve pressure drop
Extracts
12. Trap receiving statements
.
Fig 4 Yield.eps Click here to access/download;Figure;Fig 4 Yield.eps

Ethanolic Yield
6 Insoluble Phase Yield

Total Yield

4
Yield %

0
200 150 250 129,3 200 270,7 150 250 200 P [Bar]

5,2 6,4 9,3 12,1 13,2

Co [%]
 Fig 5 CorrelationTPC vs AA.eps Click here to access/download;Figure;Fig 5 CorrelationTPC vs AA.eps
Antioxidant Activity (μmol Trolox /g extract)

142

122

102

82

62

42
24 28 32 36 40 44 48

TPC (mg eq GA/g extract)

Fig 6 Pareto Chart.eps Click here to access/download;Figure;Fig 6 Pareto Chart.eps

Plot of Fitted Model

AA = -41,7304 + 3,66473*TPC
Presión (Bar) vs Cosolvente (%Etanol)
142
0,57
122

0,47
102

cosolvente
AA
0,37
82

0,27
62

42 0,17
120 160 200 240 280
24 28 32 36 40 44 48
Presión
TPC

Correlation Coefficient = 0,94365

a) TPC (mg GA/g extract) b) Antioxidant Activity (μmol Trolox/g extract)

B B

AA AA

A A

AB AB

BB BB
0 1 2 3 4 5 6 0 1 2 3 4 5
Standardized effect Standardized effect +
-
c) Polar Fase Yield (g/g dry matter) d) Non Polar Fase Yield (g/g dry matter)

B B
BB A
AA BB
AB AB
A AA
0 0,5 1 1,5 2 2,5 3 0 0,5 1 1,5 2 2,5 3

Standardized effect Standardized effect

Fig 7 Response Surface.eps Click here to access/download;Figure;Fig 7 Response Surface.eps

TPC
0,0
0,25
2,5 0,5
0,75
TPC (mg GA/g dry matter)

2 1,0
1,25
1,5
1,5
1,75
1
2,0
0,5
2,25
2,5
0,57
0 0,47
2,75
0,37
120 0,27 Cosolvent (fraction)
160
200
240 0,17
280
Pressure (bar)
Fig 8 Yield.eps Click here to access/download;Figure;Fig 8 Yield.eps

Ethanolic Yield
6 Insoluble Phase Yield

Total Yield

4
Yield %

0
200 150 250 129,3 200 270,7 150 250 200 P [Bar]

5,2 6,4 9,3 12,1 13,2

Co [%]
Title Page Click here to access/download;Attachment to manuscript;Title
Page.docx

1 Valorization of colombian agroindustrial wastes: Evaluation of conventional and non-

2 convectional extraction techniques to obtain antioxidant extracts from cocoa bean shell

3 (Theobroma cacao L.).

4 Luis M. Buelvas-Puello1 (ORCID: 0000-0002-3787-0670), Gabriela Franco-Arnedo1,

5 Isabel C. Paz-Astudillo2, Jose A. Colina-Márquez3, Blanca L. Ortiz-Quintero3, Fabián

6 Parada-Alfonso4* (ORCID ID: 0000-0003-0896-4114)

1
7 Department of Chemical Engineering. National University of Colombia. Av. Carrera 30 #

8 45-03. Postal code 111321. Bogotá, Colombia.

2
9 Department of Plant Production and Health. Faculty of Agricultural Engineering. Tolima

10 University. Barrio Santa Helena High Section. Postal code 730006299. Ibagué, Colombia.

3
11 Department of Chemical Engineering. University of Cartagena. Campus Piedra de

12 Bolívar. Calle 30, Avenida Consulado. Postal code 130015. Cartagena, Colombia.

4
13 Department of Chemistry. National university of Colombia. Av. Carrera 30 # 45-03. Postal

14 code 111321. Bogotá, Colombia. [email protected] Cell phone: +573114750329.