Accepted Manuscript

Optimization of microwave-assisted extraction of natural antioxidants from spent

espresso coffee grounds by response surface methodology
Marija Ranic , Marina Nikolic , MSc Marija Pavlovic , MSc Aneta Buntic , MSc Slavica
Siler-Marinkovic , PhD Suzana Dimitrijevic-Brankovic , PhD
PII:

S0959-6526(14)00537-X

DOI:

10.1016/j.jclepro.2014.05.060

Reference:

JCLP 4353

To appear in:

Journal of Cleaner Production

Received Date: 6 July 2013

Revised Date:

5 May 2014

Accepted Date: 20 May 2014

Please cite this article as: Ranic M, Nikolic M, Pavlovic M, Buntic A, Siler-Marinkovic S, DimitrijevicBrankovic S, Optimization of microwave-assisted extraction of natural antioxidants from spent espresso
coffee grounds by response surface methodology, Journal of Cleaner Production (2014), doi: 10.1016/
j.jclepro.2014.05.060.
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.

ACCEPTED MANUSCRIPT

Reference: JCLP 4353

Editorial reference: JCLP_JCLEPRO-D-13-00940
To be published in: Journal of Cleaner Production

RI
PT

Abstract: Optimization of microwave-assisted extraction of natural antioxidants from spent espresso

coffee grounds by response surface methodology

AC
C

EP

TE
D

M
AN
U

SC

Espresso spent coffee grounds (SCG) that is a waste material abundantly produced by restaurants,
cafeterias and in domestic environment could be used as a low-cost and rich source of valuable
polyphenol compounds. The benefit would be twofold: extraction of health beneficial natural
polyphenol antioxidants and reducing the cost to facilitate SCG waste management. The overall
objective of this study was to examine an optimal range of extraction conditions for extraction of
antioxidants from spent espresso coffee. Optimization of the extraction process from SCG was
carried out using response surface methodology (RSM). Microwave-assisted extraction (MAE) has
been used as a potential alternative to conventional solvent extraction for the isolation of
polyphenol compounds from SCG. A complete central composite 2 factorial experimental design
has been used to monitor the extraction characteristics, as affected by different variables, extraction
time (ET), liquid-to-solid ratio (LSR), and microwave power (MWP). Low concentration ethanol in
aqueous solutions was employed as non-toxic extracting media. With the reduced time of
extraction, low power and medium liquid to solid ratio while using minimal concentration of
ethanol, the polyphenols extract with high antioxidant activity can be achieved. The obtained
experimental values were in solid agreement with predicted values. The FRAP and the DPPH
antioxidative activity showed good correlation with the total polyphenol content (TPC), with high
correlation factors. The presented data could be a reliable guidelines for establishing full-scale,
sustainable cost-effective and resource-effective industrial process.

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

7328 words in total

Optimization of microwave-assisted extraction of natural antioxidants from

spent espresso coffee grounds by response surface methodology

RI
PT

Marija Ranica*, Marina Nikolica, MarijaPavlovicb, AnetaBunticb, Slavica Siler-Marinkovicb, Suzana

Dimitrijevic-Brankovicb
a

Institute for Medical Research, Centre of Research Excellence in Nutrition and Metabolism, University of
Belgrade, TadeusaKoscuska 1, PAK 104 201, 11 158 Belgrade, Serbia
b

Marina Nikolic, MSc, marina.nikolic12@yahoo.com

M
AN
U

Marija Pavlovic, MSc, marija.pavlovic87.tmf@hotmail.com

SC

Faculty of Technology and Metallurgy, Division: Biochemical Engineering and Biotechnology, University
of Belgrade, Karnegijeva 4, 11120 Belgrade, Serbia

Aneta Buntic, MSc, anetabuntic@gmail.com

Slavica Siler-Marinkovic, PhD, siler@tmf.bg.ac.rs

Suzana Dimitrijevic-Brankovic, PhD, suzana@tmf.bg.ac.rs

TE
D

Keywords: Spent espresso coffee, Microwave-assisted extraction, Response surface methodology, Total
polyphenol content, DPPH, FRAP

EP

Abbreviations: SCG, Spent coffee grounds; SCGE Spent coffee grounds extract; MAE, Microwave-assisted
extraction; RSM, Response surface methodology; TEY, Total extraction yield; TPC, Total polyphenol
content; ET, Extraction time; LSR, Liquid-to-solid ratio; MWP, Microwave power; d.w., dry weight; DPPH,
2,2-diphenyl-1-picrylhydrazyl; FRAP, ferric reducing/antioxidant power.

AC
C

Corresponding author: Tel: +381 113281564; fax: +381 112030169.

E-mail address: marija.ranic@gmail.com (Marija Ranic).

ACCEPTED MANUSCRIPT

1. Introduction

M
AN
U

SC

RI
PT

Antioxidant activity in foods and beverages is one of the properties that have generated much
interest within the scientific community (Gulcin, 2012). It has to be mentioned that food commodities
are usually fortified with additional quantities of antioxidants, either natural, or synthetic due to
depletion of antioxidants present in raw food when subjected to processing, preservation and storage.
The synthetic antioxidants have proven to be cheaper and more effective in the retardation process of
oxidative degradation, comparing to the natural ones. The synthetic antioxidants, introduced a few
decades ago, are mostly butylated phenols and polyphenols, such as butylated hydroxyl toluene (BHT),
an effective, widely used and low cost antioxidant (Mukhopadhay, 2006). However, due to safety
concerns, interest in natural antioxidants has intensified (Shahidi, 2000) and synthetic antioxidants are
subjected to evaluating the safety. Many studies were investigated whether BHT possess
mutagenic/genotoxic risk to man (Bomhard et al., 1992). Recently, the EFSA Panel (European Food
Safety Authority) gave the Scientific opinion on the re-evaluation of BHT as a food additive (EFSA, 2012).
Despite contradictory recommendations, BHT it is still in use for food products preservation specially in
developing countries, thanks to cost advantages and availability and it is still very popular antioxidant in
many different fields; 80.000 100.000 tons per year is been producing in USA, Western Europe and
Japan (Mukhopadhay, 2006). The superiority of natural antioxidants has to be emphasized vs. synthetic
in terms of tolerance, safety and non-toxicity. In addition to their safety advantages, food phenolics
may augment bodys source of natural antioxidants.

TE
D

Polyphenol compounds have received substantial attention due to their beneficial effects on
human health, such as a protective action against chronic degenerative diseases (cataracts, macular
degeneration, neurodegenerative diseases, and diabetes mellitus), cancer and cardiovascular diseases,
which is attributed to their antioxidant activity (Mussatto et al., 2011a). Recent studies report
considerable evidence of health benefits for healthy adults as a result of moderate coffee consumption
(Butt and Sultan, 2011). The benecial effect of coffee consumption is partially attributed to the
antioxidant activity of the polyphenols present in coffee (Vignoli et al., 2011; Cammerer and Kroh, 2006).
Tannins are consider to be the main phenolic compounds in coffee pulp, while in the seed, polyphenol
compounds are present predominantly as a family of esters formed between certain hydroxycinnamic
acids and quinic acid, known as chlorogenic acids (CGA) (Farah and Donangelo, 2006).

AC
C

EP

Coffee, as the one of the most popular beverages, is consumed by millions of people all over the
world, several times a day. In the coffee industry, during the manufacture of soluble and concentrate
coffee, huge amount of solid residue known as spent coffee grounds (SCG) is produced. It was estimated
that only from the Province of Rome over 10,000t of SCG are available per year for the production of
polyphenols rich extracts and bioenergy, which means that around 1400t/year of polyphenol extract
could be produced (Zuorro at al., 2012). With regard to spent coffee, two types should be
differentiated. The soluble coffee industry bringing together almost 50% of the world coffee production,
with a proportional amount of spent coffee residues. Another 50% is used for preparation of beverages.
Depending on the intended reuse, espresso spent coffee bring out a similar or even greater reuse
potential than the one expected from spent grounds obtained from the soluble coffee industry,
exhausted of most of its soluble components (Cruz at al., 2012). Regardless of the brewing process
applied, during the preparation of coffee beverages, extraction efficiency is clearly lower comparing to
industrial level. Therefore, the residues, being richer in unextracted coffee components comparing to
coffee brew, opening the series of potential applications. That fact leads us to the assumption that spent
coffee produced in tons by restaurants, cafeterias and in domestic environment, could be important
source of natural antioxidants, also from an economical point of view (Bravo et al., 2012). Adams and

ACCEPTED MANUSCRIPT

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

Ghaly, 2006 suggested an alternative use of coffee industry by-products, among others, for maximizing
the sustainability of the coffee industry, through identification of new opportunities and cost and risk
reductions. Waste minimization can also be achieved by turning wastes into value added products of
economic value. Several previous studies have dealt with residues generated during the entire coffee
processing (Yen et al., 2005; Adams and Ghaly, 2006, Kondamudi et al., 2008; Murthy et al., 2010; Bravo
et al., 2013; Mussatto et al., 2011a; Esquivel et al., 2012; Zuorro et al., 2012; Cruz et al., 2012). In
addition, the more recent studies continue to deal with different aspects of possible SCG utilization: oil
extracted from SCG as a substrate for the production of poly (3-hydroxybutyrate) (Obruca et al., 2014),
production of polyhydroxyalkanoates from SCG oil obtained by supercritical fluid extraction technology
(Cruz et al., 2014),) potential feedstock for mannooligosaccharides (Chiyanzu et al., 2014) and mannans
from SCG (Passos et al., 2014), etc.
Extraction is the first step in the isolation of polyphenol compounds from SCG. Different
techniques have been applied to recover antioxidant polyphenol compounds from natural sources
including solidliquid extraction with organic solvents, ultrasound-assisted extraction (UAE), microwaveassisted extraction (MAE), supercritical fluid extraction, and high pressure processes (Mussato et al.,
2011). Different extraction techniques have certain limitations in terms of extraction time, energy,
solvent consumption and cost and also influenced by several factors such as the type and concentration
of solvent, the solvent/solid ratio, pH, time, temperature, etc. Comparing to the SCG extracts (SCGE)
obtained with continuous (Soxlet, 1h and 3h) and discontinuous methods (solid-liquid extraction) (Bravo
et al., 2011), MAE has been considered as a potential alternative to conventional solvent extraction for
the isolation of polyphenol compounds from SCG (Upadhyay et al., 2012). So far, it has been used for
extracting a number of components from different sample matrices (Chen and Spiro, 1994; Young, 1995;
Lopez-Avila et al., 1996; Bureau et al. 1996; Kwon al., 2003; Rostagno et al., 2007; Zhanga et al., 2008;
Ballard et al., 2009; Hyun-Ku et al., 2012; Li et al., 2012; Singh et al., 2011). Optimization of the
extraction procedure of plant constituents through the traditional method, i.e., one variable at-a-time,
is time consuming and not cost effective, while statistical optimization methods provide insight about
the interactions of the variables when generating the response. In the context of optimization of the
whole system, response surface methodology (RSM) is the most widely used (Bezerra et al., 2008). It
derives a model equation that fits the experimental data and it calculates the optimal response of the
system with a limited number of experiments. The RSM was used so far for optimization of antioxidant
activity of different extracts (Karacabey et al., 2010; Saha et al., 2011; Gupta et al., 2011; Ghafoor et al.,
2011; Singh et al., 2011). In the work of Zuorro et al. 2012, a two-level factorial design was used to
study the effects of temperature, extraction time and liquid-to-solid ratio, and ethanol concentration on
the recovery of phenolic compounds from SCG. They suggest future improvement by subsequent
optimization based, like on RSM.
Over the last few decades, an espresso coffee started to be dominant in the restaurants and
cafeterias. Increase consumption of espresso coffee generates a considerable amount of SCG, as a
waste, that could be reused for industrial scale-up. The overall objective of this study was to identifying
optimal range of extraction conditions and to develop an effective and eco-friendly MAE strategy to
optimize the extraction of polyphenol compounds from samples of spent espresso coffee, using RSM.
The correlation between the antioxidant activities and polyphenol content (Dudonn et al., 2009) was
studied by employing two well-established assays for antioxidant activity, 2, 2-diphenyl-1-picrylhydrazyl
(DPPH.) decolorization assay and ferric reducing antioxidant power (FRAP) assay. To the best of our
knowledge, such an approach has not been reported elsewhere, so far.

1. Materials and methods

ACCEPTED MANUSCRIPT

2.1 Chemicals and reagents

RI
PT

We used Doncafe - Espresso Aromatico-cialde corresponding to Arabica/Robusta blends in

undisclosed exact proportion. All reagents that were used in this work were of analytical grade and were
purchased from Sigma Chemical Co. (St. Luis, USA), Aldrich Chemical Co. (Steineheim, Germany), Merck
(Darmstadt, Germany) and Sineks laboratory (Serbia). The ethanol (20%) and distilled water were used
as solvents. Total polyphenol content was measured using Folin-Ciocalteau reagent and then calculated
using Gallic Acid (GA) as a standard. The antioxidant capacity of spent coffee extracts was measured
using 2, 2-diphenyl-1-picrylhydrazyl (DPPH.) decolorization assay. The FRAP method, was performed
using complex ferric ion-TPTZ (2, 4, 6-tri (2-pyridyl) - 1, 3, 5-triazine).

SC

2.2 Spent coffee preparation

2.3 Experimental design

TE
D

M
AN
U

A total number of samples Doncafe - Espresso Aromatico - cialde were donated by a local
company Strauss-Adriatic d.o.o. Doncafe Espresso Aromatico is mix obtained from Arabica (below 50%)
and Robusta (above 50%) (supplied from Santos, Guatemala, Columbia, India, Honduras and Vietnam
depending on the current offer on the market, while taking into account similarity in composition).
Espresso coffee was prepared on Italian Frog Espresso Coffee Pod Machine (Power: 450W; Voltage:
230/120V) that use cialda pre-packaged espresso ground coffee in coffee pods. 30 ml of espresso
coffee brew was obtained from 7g of ground coffee in cialda, within 25s of extraction process at about
90oC. After brewing, spent coffee grounds from coffee pods, was widespread in a very thin layer, with a
large contact area, thus shortening the duration of the next phase (drying stage). Samples of collected
spent coffee were preserved by oven-drying at 40oC (Memmert). The samples were stored at 3-4C
during the experiment.

AC
C

EP

The RSM was employed to optimize the MAE of phenolic compounds from SCG samples, and to
determine the optimum extraction characteristic by employing the central composite design (CCD)
(Cochran et al., 1992). In this case, the RSM was used to study the effects of extraction time (ET), liquid to-solid ratio, (LSR) and microwave power (MWP) to total polyphenol content (TPC) in the extract. Our
previous work showed that 20% ethanol solution in water is optimal for polyphenol extraction from
spent coffe so this parameter was not considered in this work (Pavlovi et al., 2013).
The experimental domain was defined taking into account the results obtained in preliminary
tests, as well as the operational limits of the instrument and all significant parameters in a typical MAE
process were chosen as independent variables: ET (A=11.0-209.0s), LSR (B=4.76-13.24ml/g) and MWP
(C= 240, 400 and 560W). The range and levels of 3 independent variables and code values are
presented in Table 1. The variables A and B have been coded at 5 levels -, -1, 0, 1, + while variable C
was coded at 3 levels (-1, 0, 1). The response variable were: total extraction yield (TEY; in mg/g d.w.
SCG), total polyphenol content (TPC; % (w/w) in extract), FRAP (mmolFe2+/l) and DPPH (% of DPPH
inhibition). The complete design consists of 30 experiments that were conducted in triplicate (Table 2).
Experimental runs were randomised in order to minimize the effects of unexpected variability in the
observed responses. The data from CCD were analysed by multiple regression to fit to a second-order
polynomial regression model containing the coefficient of linear, quadratic, and two factor interaction

ACCEPTED MANUSCRIPT

effects. The model equation of response (Y) of three independent variables (A, B and C) is given in the
following equation:
Y=  3i=1   3i=1        

(1)

SC

RI
PT

where Y is the dependent variable, 0 is the constant coefficient,  is the linear coefficient (main effect),
 is the quadratic coefficient, and  is the two factor interaction coefficients. The model adequacy
was determined by evaluating the lack of fit, coefficient of determination (R2), and the Fisher test value
(F-value) obtained by analysis of variance (ANOVA) that was generated by the Design Expert 8 software
(Stat-Ease, Inc., Minneapolis, MN, USA). Differences were considered significant if p<0.05. Optimal
independent variables (extraction conditions) for maximised responses were pre-established by
superimposing the corresponding contour plots, by which two subsequent confirmatory experiments
were carried out to validate their equations.

M
AN
U

2.4 Analysis of responses

TE
D

Total extraction yield (TEY): MAE was performed using a domestic microwave oven (SAMSUNG
CE104VD,1250W). In brief, 3g d.w. SCG was macerated with appropriate volume of 20% (v/v) ethanol in
250ml Erlenmeyer flasks. Extraction with different ET, LSR, and MWP was performed according to the
experimental design (Table 2) in a total number of 30 samples. After extraction, solid phase was
separated using vacuum-pump (BCHI Vacuum Pump V-700), and extracts are amended to the initial
volume for further analysis. The TEY of each sample was assessed by plotting the dry matter calibration
curve made by optical density measurement (green filter, 570nm) of different concentration of dry
samples, obtained using the Bchi Mini Spray Dryer B-290 (BchiLabortechnik AG, Flawil, Switzerland)
after dissolving in 20% (v/v) ethanol concentration, which is the working condition of this unit, using
20% (v/v) ethanol concentration as a blind test.

AC
C

EP

Total polyphenol content (TPC): TPC in spent espresso coffee extracts was measured using the
Folin-Ciocalteau reagent according to the modified method of Dordjevic et al. 2009. Briefly, water
solution (100l) of the investigated extracts (adjusted to 1mg d.w. /ml) was shaken for 60s on the Vortex
mixer with 500 L of FolinCiocalteu reagent and 6 mL of methanol. After that, 2 mL of 15% Na2CO3 was
added and the mixturewas shaken once again for 30s. Finally, the solution was brought up to 10 mL by
adding distilled water. After 2 h, the absorbance was read on the UV/visible spectrophotometer
(Ultrospec 3300 pro, Amersham Bioscience, Uppsala, Sweden) at 750 nm (25oC) using glass cuvettes
against a blank (100L of distilled water instead test samples). The TPC was assessed by plotting the
gallic acid calibration curve (from 1 to 1500 g/ mL) and the results expressed as gallic acid equivalent
are recalculated as the percent of dry extract.
The antioxidant capacity of DPPH assay: The ability to neutralize DPPH (2, 2-diphenyl-1picrylhydrazyl), stable free radical is a method commonly used to determine the antioxidant activity
(Pisoschi et al., 2011). The free radical scavenging capacities of spent espresso coffee extracts were
determined according to the previously described method (Cuendet et al., 1997). Briefly, the aqueous
solutions of the extracts (50l) with concentrations 1mg d.w. /mL were further dissolved in methanol up
to 4ml. After that 1 mL of 0.2mmol DPPH in methanol was added. The obtained solutions were
vigorously shaken and left at room temperature in the dark. After 30 min of incubation the absorbance
was measured against a blank (methanol) at 517 nm using the UV/visible spectrophotometer (Ultrospec
3300 pro, Amersham Bioscience, Uppsala, Sweden). Inhibition of DPPH radical was calculated as a
percentage (%) using the equation (2):

ACCEPTED MANUSCRIPT

Scavenging activity (%) = [(A control-A sample) /A control] *100

(2)

where, A control is the absorbance of the control reaction (containing all reagents except tested
extract), and A sample is the absorbance of the tested extract. Ascorbic acid was used as a positive
control and all tests were carried out in triplicates.

M
AN
U

SC

RI
PT

Ferric reducing antioxidant power (FRAP): FRAP method is based on the ability of phenolic
substances, soluble in water, to reduce Fe3+ to Fe2+. FRAP assay uses antioxidants as reductants in a
redox-linked colorimetric method. The FRAP method relies on the reduction by the antioxidants, of the
complex ferric ion-TPTZ (2,4,6-tri(2-pyridyl)- 1,3,5-triazine) (Pisoschi et al., 2011). FRAP method
described previously (Szllsi and Szllsi, 2002) was used to assess overall antioxidant activity. Water
solution (150 L) of espresso SCGE was mixed with 4.5 mL of FRAP reagent. In the FRAP method, the
yellow Fe3+TPTZ complex is reduced to the blue Fe2+TPTZ complex by electron- donating substances
under acidic conditions. Any electron donating substance with a half reaction of lower redox potential
than Fe3+/Fe2+TPTZ will drive the reaction and the formation of the blue complex forward. The
absorbance readings started after 5 min at 593 nm using a spectrophotometer (Ultrospec 3300 pro,
Amersham Bioscience, Sweden). The blank probe consisted of FRAP reagent. The final absorbance of
each sample was compared with those obtained from the standard curve made from ferric sulphate
(FeSO4 7H2O) (200, 400, 600, 800, and 1000 mol/L). The results were expressed in mmol Fe2+/g of
extract.

2. Results and discussion

TE
D

3.1 Selection of the solvent

AC
C

EP

A constant solvent concentration of 20% (v/v) ethanol was used based on our preliminary
results. The influence of solvent to polyphenols extraction was investigated extensively (Bilek S. E., 2010;
Zhang et al., 2011; Wu et al., 2011). Ethanol as a solvent was chosen based on comparing the phenolic
content obtained from spent espresso using different solvent systems (water < methanol < ethanol
extract) (Milutinovic et al., 2013), confirmed by DPPH and FRAP assays. Further analyses included
phenolics extraction from espresso SCG under MAE conditions, optimised using one factor design
under RSM. The investigated factors, ET and percent of ethanol in water solution ranged from 40s
to 360s and 20% to 80% (v/v), respectively (unpublished results).The yield of TPC reached a
maximum using 20% (v/v) ethanol concentration as it is shown on Fig. 2 (I). Our results are in
agreement with results of Joki et al., 2010 that are also found that lower percent of ethanol in water
positively affected polyphenols recovery from soybean. Aqueous ethanol is a safe and efficient solvent,
and also bio-solvent that can be generated through fermentation of different raw materials, with
possibilities for recycling, as a basic element for sustainable industrial process (Kiassos et al., 2009).

3.2 Optimization of the total extract yield

The yield of total dry extracts was optimized by using RSM approach. In Table 2 actual and
predicted values for total extract yield (TEY) were given. The yield ranged from 7,694 to 31.216 mg/g
d.w SCG. The maximum yield was recorded under following experimental conditions: 180s ET, 12 ml/g
LSR, and 550W MWP (run 24). The lowest result is recorded at following conditions: 40s extraction

ACCEPTED MANUSCRIPT

time, 6 ml/g LSR, and 400W MWP (run 1). According to analysis from present study an increase of ET (A),
LSR (B) and MWP (C) result in an increase of TEY.

M
AN
U

SC

RI
PT

By applying a multiple regression analysis of the experimental data, the response variable
related to the second-order polynomial equation is given in Table 3. In Table 4 analysis of variance
(ANOVA) for the regression equation is shown. The linear term and quadratic term were very significant
(p<0.01). The most significant interaction between independent variables is between A and C. The value
of lack of fit (0,0790) verifies the adequacy of the proposed model indicating that the model fits the
experimental data adequately (p>0.05). The relationship between the experimental and predicted
values showed that the plotted points cluster around the diagonal line in Fig. 1 (A), indicating good
fitness with a low dispersion, because the value of predicted R-squared of 0.9586 is in satisfactory
agreement with adjusted R-squared of 0.9402. Rather low value of coefficient of variance C.V.% of 7,304
indicated a high degree of precision and reliability of the experimental values. Fig. 2 (A) shows the threedimensional response surface plots of the TEY as affected by A (ET) and C (MWP), indicated as significant
with a p value of 0.0002. An increase of ET (160-180s) with higher values of MWP (480-550W) results in
an increase of TEY to a maximum at a certain level. Verification experiments (A and B in Table 2), carried
out in the predicted conditions, showed values reasonably close to predicted ones and further
confirmed the adequacy of predicted models.

3.3 Optimization of the total polyphenol content

AC
C

EP

TE
D

Present experimental and predicted data for TPC are given in Table 2 expressed as percent (%,
w/w) in dry SCGE. The yield ranged from 18.83 up to 79.83%w/w. The maximum yield was recorded in
sample run 16 under following experimental conditions: 40s extraction time, 240 W MWP and 6 fold
solvent to SCG ratio. The response variable related to the second-order polynomial equation is given in
Table 3. In Table 4, analysis of variance (ANOVA) for the regression equation is shown. The linear and
quadratic term were very significant (p<0.01). The most significant interaction between independent
variables is between A and C, and B and C. The value of lack of fit (0,0679) indicating that the model fits
the experimental data adequately (p>0.05). The relationship between the experimental and predicted
values showed that the plotted points group around the diagonal line as it is shown at Fig. 1 (B),
indicating good fitness with a low dispersion, the value of predicted R-squared of 0.845 and adjusted Rsquared of 0.892. Coefficient of variance C.V.% of 13.667 indicated a satisfactory degree of precision and
reliability of the experimental values. Fig. 2 (B) shows the three-dimensional response surface plots of
the TPC as affected by A and C indicated as significant with a p value of 0.0007, and B and C, Fig. 2 (C),
with a p value of 0.0013. The results suggest that shorter ET (40-110s) and lower range of MWP lead to
higher TPC (plot B). The LSR lower than 6mg/g with the use of 240-400W MWP also results in higher TPC
(plot C).
Similar results regarding to ET and MWP can be found in a work of Zhang et al., 2012, where 35s
of ET were enough to reach the maximal yield of polyphenols from Lotus seeds with MWP of 400W.
Further increasing of these two factors leads to decreasing of polyphenols yields in Lotus seeds extracts.
Somporn et al., 2011, showed that temperature applied to roasting process can influence the TPC in
coffee extracts and that higher temperature and longer time of roasting decrease the TPC in extracts.
The influence of LSR is a little more difficult to compare with other authors due to the fact that we
investigated the lower level of LSR in a narrow range, from 4.76 to 13.24 ml/g SCG. Some investigations
showed that the TPC total yield is higher with the higher LSR. For example, Mussatto et al., 2011a
showed that 25 LSR of methanol solution (in range of 10-50) was optimal for polyphenols recovery from
spent coffee grounds. Zuorro and Lavecchia, 2012 are concluded that the increasing of LSR (in range of

ACCEPTED MANUSCRIPT

10-50, ml/g) lead to higher TPC in spent coffee grounds extracts. However, Franco et al, 2008 showed
that phenolic recovery from grape pomace was higher with the lowest LSR (in range of 1-5, ml/g of
ethanol solution). In addition, Mukhopadhyay et al., 2006 showed that a higher portion of solids in the
black cohosh extraction process (ranged from 0-200 S/L, mg/ml) positively influence the TPC in extract.

M
AN
U

3.4 Optimum extraction conditions based on the DPPH method

SC

RI
PT

Comparing our results of TPC with the results of total extraction yield - TEY (total solids of
extracts) it is clear that the TPC is in inverse proportion to TEY, as it is shown in Fig. 3 (A). The value of
correlation factor is -0.747, negative slope. Experimental conditions that are optimal for the maximum
of total extraction yield (longer ET, higher LSR and higher MWP) are not optimal for the maximum
content of TPC (shorter ET, lower LSR and lower MWP). A possible explanation for this outcome could be
that the extracts with higher values of dry matter also content an increase level of other (ballast)
substances such as polysaccharides, proteins and lipids, thus reducing the proportion of TPC (Mussatto
et al., 2011b).The similar results can be noticed at work of Wu et al., 2011 when recalculations of TPC
contribution in total yield were carried out.

TE
D

Experimental and predicted data for antioxidative activity of SCGE determined by the DPPH
method are given in Table 2. The antioxidative activity ranged from 36,560%, in run 6 up to 98,240% in
run 28, under experimental conditions of 11s ET, 9 fold solvent to SCG ratio and 240 W, MWP. The
response variable related to the second-order polynomial equation is given in Table 3. In Table 4,
analysis of variance (ANOVA) for the regression equation is shown. The used model is significant
(p<0.01). The most significant interaction between independent variables is between A and C. The value
of lack of fit (0,1377) indicating that the model fits the experimental data adequately (p>0.05). The
relationship between the experimental and predicted values showed that the plotted points cluster
around the diagonal line (Fig. 1 (C), indicating good fitness with a low dispersion, with the value of
predicted R-squared of 0.9734 and adjusted R-squared of 0.9805, R2 is 0.9845. Coefficient of variance
C.V.% of 3.711 indicated a very high degree of precision and reliability of the experimental values.

AC
C

EP

Fig. 2 (D) shows the three-dimensional response surface plots of the DPPH as affected by A and
C indicated as significant with a p value <0.001. Shorter MAE time (40-120s) with the use of lower range
of MWP (240-400W) results in higher % in neutralization of DPPH radicals.These results are in direct
correlation with TPC (correlation factor 0,841) and in inverse correlation to TEY (correlation factor 0.786), as it is shown in Fig. 3 (A) and (B),resp. It should be noticed that the samples in our investigations
were adjusted to a concentration of 1 mg/ml of extracts, so these findings also confirm the influence of
investigated factors on TPC proportions in total solids of SCGE. The other researcher also noticed
positive correlation of TPC and DPPH inhibition activity (Franco et al., 2008, Wang et al., 2011).

3.5 Optimum extraction conditions based on the FRAP activity

Experimental and predicted data for antioxidative capacity of SCGE determined by FRAP method
are given in Table 2. The antioxidative activity ranged from 2.620mmol Fe2+/l , run 12, up to 6.660mmol
Fe2+/l in run 28, under experimental conditions of 11s ET, 9 fold solvent to SCG ratio and 240W MWP.
The response variable related to the second-order polynomial equation is given in Table 3. In Table 4,
analysis of variance (ANOVA) for the regression equation is shown. The used model is significant
(p<0.01), with all significant parameters. The most significant interaction between independent
variables is between A and C. The value of lack of fit (0,4664) indicating that the model fits the

ACCEPTED MANUSCRIPT

RI
PT

experimental data adequately (p>0.05). The relationship between the experimental and predicted
values showed that the plotted points cluster around the diagonal line (Fig. 1 (D), indicating good fitness
with a low dispersion. The squared correlation coefficient R2 of the model was 0.9433, which confirms
that the model can adequately represent the true relationship between the parameter chosen.
Coefficient of variance C.V.% of 8.324 indicated a high degree of precision and reliability of the
experimental values. FRAP value as affected by relation of A and C indicated as significant with a p value
<0.0001. As evident in Fig. 2 (E) shorter MAE time (40-110s) with the use of the lower range of MWP
(240-300W) are more effective in extracting antioxidative polyphenols from espresso SCG using MAE.
This result is favorable since using of higher MWP with extended ET may lead to thermal degradation of
the polyphenol compounds (Chen et al., 2007).

M
AN
U

SC

It can be noticed that FRAP values are also in direct correlation with TCP (correlation factor
0,859), Fig. 3 (D) and in inverse proportion to TEY (correlation factor -0.770), as it is shown in Fig. 3 (E).In
general, the relationship of FRAP was the similar to what was observed in the DPPH assay. Correlation
between DPPH and FRAP values (Fig. 3 (F) shows that samples with high antioxidative activity
determined by DPPH method, also have high FRAP values. The highest FRAP value in the range from 6.0
to 7.0 mmol Fe2+/l, and the highest antioxidant capacity determined by DPPH (70% to 100%) correspond
to TPC range from 65% and 80%of dry SCGE.9 fold solvent to SCG ratio, shorter MAE time (40-110s), and
a lower range of (240-400W) gives the optimal DPPH and FRAP values. The minor deviations could be
the result of different solubility of polyphenols showing antioxidative properties in different solvents
(methanol for DPPH and water for FRAP method).

TE
D

The results presented in this study show that used models are efficient to predict the outcome
variables (TEY, TPC, DPPH and FRAP) of espresso SCG using the independent variables (ET, LSR, MWP).
The results obtained from all models are highly correlated, which means that each model can be used to
predict the outcome based on listed independent variables. Validation of the models showed no
statistically significant differences between the results obtained by the experiment and those predicted
by the models.

AC
C

EP

Overall, espresso SCG could be an inexpensive source of natural antioxidants. Extracts obtain
form espresso SCG in this study had high TPC contents (up to 79.83%) of dry matter obtained under
optimal extraction conditions. In this regards the SCG can be compared to other sources that were used
for the same purpose, such as kinema fermented soybean food with 135 mg GAE/g d.w. (Saha et al.,
2011); potato peels, 3.94 mg/g d.w. (Singh et al., 2011); grape (Vitis vinifera) seeds, 5.44mg GAE/100mL
(Ghafoor et al., 2009); tomato, 4.89-9.97mg GAE/g d.w. (Li et al., 2012); or onion (Allium cepa) solid
waste, which is theoretically predicted to have 93.4214.35mg GAE/g d.w., under optimal conditions
(Kiassos et al., 2009).
Many companies from food and cosmetic industry used MAE for plant nutrient extraction. It is
worth to mention CRODAROM and CODIF companies that use microwaves for extraction of antioxidant
and colours of plants and marine algae, respectively (Ying Li et al., 2013). The industrial scale microwave
equipment is still developing especially for continuous processes (Radoiu, 2013 ). Mandal et al., 2007.,
gives a detailed review of MAE for medicinal plant extraction and point out the significance of MAE
process optimisation for the maximum yield. Kumar Das and coworker, 2013, compared various method
for extraction of lupeol from Ficus racemosa leaves. They concluded that the MAE is more rapid,
convenient and appropriate extraction method, with a higher yield and lower solvent consumption
when compared with conventional extraction techniques. They also concluded that the response surface
methodology can be applied to all natural products and if explored properly, can prove to be efficient
for large-scale industrial application.

ACCEPTED MANUSCRIPT

M
AN
U

SC

RI
PT

Our results also suggested that MAE can be an efficient technique for waste coffee polyphenolic
extraction. With the short time of extraction, low power and medium liquid to solid ratio as well as with
minimal concentration of ethanol the polyphenol extract with high antioxidant activity can be obtained.
The SCG is widely available from restaurants with no-costs. It gives great importance to the economic
feasibility of its use. In addition, an environmental standard promotes proper, responsible and safe
management of waste and successful waste prevention that not only benefits the environment, but also
has the potential to eliminate certain amount in annual disposal costs. (Nelson and Pound, 2009).
Starbucks Corporation, an American global coffee company and coffee house chain, possess 20,891
Starbucks stores, as of March 2013 (Loxcel, 2013) in 61 countries and territories. Starbucks used to make
spent espresso/coffee grounds available for free. Costa Coffee, a British multinational coffee house
company, internationally operates to over 1,700 stores across 29 countries (Costa Coffee, 2013). In
2009, 1.6 billion K-cups were sold, mostly fulfilled with different kind of espresso coffee (Keurig, 2013).
It is worth to mention an ecological device for quick removal of the spent espresso coffee grounds (EP
1803380 A2). This device allows the user to remove, with little effort, in just one quick stage the used
coffee grounds from the metallic filter, housed in the portafilter of an espresso machine, after the
preparation of a coffee beverage (Campos, 2012). The foregoing facts contribute to Waste-to-Product
(WtP) novel concept (Bosmans, 2013), in a sustainable context. Turning waste into value added
products, ie. proposed cyclic reuse of coffee waste is shown in Fig. 4.

4. Conclusions

AC
C

EP

TE
D

The sustainability of the coffee processing system can be substantially improved through the use
of by-products, by adoption of new technologies that maximize process profitability. The espresso
coffee grounds are non-cost by product of coffee consumption. It contains a great amount of valuable
phenolic and other components with high antioxidative properties. In our study, MAE along with RSM
proved to be effective in estimating the effects of three independent variables (ET, LSR and MWP) on
the phenolics extraction from espresso SCG. We were able to extract high level of phenolics from SCG,
using less solvent with considerably shorter extraction time than previously reported. Although a
detailed cost-benefit analysis is needed to assess the economic practicability of the proposed approach,
the analysis and the results reported in this study could be of great significance for future exploitation of
espresso SCG as a valuable source of natural antioxidant in industrial scale-up. All presented facts could
be a reliable guidelines for development of a full-scale project and good business opportunities for
SMEs, producing fortified foods or dietary supplements with pharmacological doses and safety
assessments of the applied doses performed (Mennen et al., 2005).

Acknowledgments

This work is funded by the Ministry of Education, Science and Technological Development, Republic of
Serbia, Projects TR31035 and III 41030. Language help and proof reading the article provided by Romana
Novakovic, PhD is greatly appreciated. The authors would also like to thank the Strauss Adriatic d.o.o,
imanovci, Serbia for providing materials for investigation.

ACCEPTED MANUSCRIPT

References
Adams, M., Ghaly, , A.E., 2007. Maximizing sustainability of the Costa Rican coffee industry. Journal of
Cleaner Production, 15, 1716-1729.

RI
PT

Ballard, T.S.,Mallikarjunan, P., Zhou, K.,OKeefe, S.F., 2009. Optimizing the extraction of phenolic
antioxidants from peanut skins using response surface methodology. Journal of Agriculture and
Food Science, 57, 3064-3072.
Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S., Escaleira, L. A., 2008. Response surface
methodology (RSM) as a tool for optimization in analytical chemistry. Talanta, 76 (5), 965-977.

SC

Bilek, S.E., 2010. The effects of time, temperature, solvent: solid ratio and solvent composition on
extraction of total phenolic compound from dried olive (Oleaeuropaea L.) leaves. GIDA, 35, 6,
411-416.

M
AN
U

Bomhard, E.M,, Bremmer, J.N., Herbold, B.A., 1992. Review of the mutagenicity/genotoxicity of
butylated hydroxytoluene. Mutation Research, 277, 187-200.
Bosmans , A., Vanderreydt , I., Geysen, D., Helsen, L. , 2013. The crucial role of Waste-to-Energy
technologies in enhanced landfill mining: a technology review. Journal of Cleaner Production,
55, 10-23.

TE
D

Bravo, J., Juniz, I., Monente, C., Caemmerer, B., Kroh, L. W., De Pea, M. P., Cid, C. , 2012. Evaluation of
Spent Coffee Obtained from the Most Common Coffeemakers as a Source of Hydrophilic
Bioactive Compounds. Journal of Agricultural and Food Chemistry, 60 (51), 12565-12573.
Bravo, J., Monente, C., Juniz, I., De Pea, M. P., Cid, C., 2013. Influence of extraction process on
antioxidant capacity of spent coffee. Food Research International, 50 (2), 610-616.

EP

Bureau, S., Razungles, A., Baumes, R., Bayonove, C. , 1996. Glycosylated flavour precursor extraction by
microwave from grape juice and grapes. Journal of Food Science, 61, 557561.
Butt, M.S., Sultan, M.T., 2011. Coffee and its consumption: benefits and risks. Crit Rev Food Sci Nutr, 51,
4,363-73.

AC
C

Cammerer, B., Kroh, L.W. , 2006. Antioxidant activity of coffee brews. European Food Research and
Technology, 223, 469474.
Campos, D.A., 2012. Ecological device for quick removal and without knocking impact of the spent
espresso coffee grounds, EP 1803380 A3
(http://www.google.com/patents/EP1803380A2?hl=sr&cl=en).
Chen, S.S., Spiro, M., 1994.Study of microwave process of essential oil constituents from plant materials.
The Journal of Microwave Power and Electromagnetic Energy, 29, 231241.
Chen, Y., Xie, M.-Y., Gong, X.-F., 2007. Microwave-assisted extraction used for the isolation of total
triterpenoid saponins from Ganoderma atrum. Journal of Food Engineering, 81 (1), 162170.

ACCEPTED MANUSCRIPT

Chiyanzu, I., Brienzo, M., Garca-Aparicio, M.P., Grgens, J.F., 2014. Application of Endo--1,4,Dmannanase and Cellulase for the Release of Mannooligosaccharides from Steam-Pretreated
Spent Coffee Ground. Appl Biochem Biotechnol, 172, 7, 3538-57.

Costa Coffee, 2013. (http://www.costacoffee.com/)

RI
PT

Cochran, W. G., Cox, G. M., 1992. Some Methods for the Study of Response Surfaces. In Experimental
Designs; John Wiley and Sons, Inc., the United States of America, 335-375.

Cruz, M.V., Paiva, A., Lisboa, P., Freitas, F., Alves, V.D., Simes, P., Barreiros, S., Reis, M.A., 2014.
Production of polyhydroxyalkanoates from spent coffee grounds oil obtained by supercritical
fluid extraction technology. Bioresour Technol, 157, 360-3.

SC

Cuendet, M., Hostettmann, K., Potterat, O., 1997. Iridoid glucosides with free radical scavenging
properties from Fagraea blume. Helvetica Chimica Acta, 80, 11441152.

M
AN
U

Das, A.K., Mandal, V., Mandal,S.C., 2013. Design of Experiment Approach for the Process Optimisation of
Microwave Assisted Extraction of Lupeol from Ficus racemosa Leaves Using Response Surface
Methodology. Phytochem. Anal., 24, 230247.
Dordevic,T.M., Siler-Marinkovic, S.S., Dimitrijevic-Brankovic, S.I. , 2009. Effect of fermentation on
antioxidant properties of some cereals and pseudo cereals, Food Chemistry, 119, 957963.

TE
D

Dudonn, S., Vitrac, X., Coutire, P., Woillez, M., Mrillon, JM., 2009. Comparative Study of Antioxidant
Properties and Total Phenolic Content of 30 Plant Extracts of Industrial Interest Using DPPH,
ABTS, FRAP, SOD, and ORAC Assays. Journal of Agricultural and Food Chemistry, 57 (5), 17681774.
EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS), 2012. Scientific Opinion on the
re evaluation of butylated hydroxytoluene BHT (E 321) as a food additive. EFSA Journal, 10, 3,
2588.

EP

Esquivel, P., Jimnez, V.M. , 2012. Functional properties of coffee and coffee by-products. Food
Research International, 46, 488-495.
Farah, A., Donangelo, C.M., 2006. Phenolic compounds in coffee. Braz. J. Plant Physiol. 18, 1.

AC
C

Franco, D., Sineiro, J., Rubilar, M., Snchez, M., Jerez, M., Pinelo, M., Costoya, N., Nez,M.J.,
2008.Polyphenols from plant materials: extraction and antioxidant power. EJEAFChe, 7, 8, 32103216.
Ghafoor, K., Choi, Y. H., Jeon, J. Y., Jo, I. H., 2009. Optimization of Ultrasound-Assisted Extraction of
Phenolic Compounds, Antioxidants, and Anthocyanins from Grape (Vitisvinifera) Seeds. Journal
of Agricultural and Food Chemistry, 57 (11), 4988-4994.
Gulcin, I., 2012. Antioxidant activity of food constituents: an overview. Archives of Toxicology, 86,345
391.
Gupta, S., Banerjee, R., 2011. Radical scavenging potential of phenolics from Bryophyllumpinnatum
(LAM.) OKEN. Prep BiochemBiotechnol, 41 (3), 305-19.

ACCEPTED MANUSCRIPT

Joki, S., Veli, D., Bili, M., Buci-Koji,A., Planini, M.,Tomas, S., 2010. Modelling of the Process of
Solid-Liquid Extraction of Total Polyphenols from Soybeans, Czech J. Food Sci. 28, 3, 206212.
Keurig, 2013. (http://www.keurig.com/)

RI
PT

Kiassos, E., Mylonaki, S., Makris, D. P., Kefalas, P., 2009. Implementation of response surface
methodology to optimise extraction of onion (Allium cepa) solid waste phenolics. Innovative
Food Science and Emerging Technologies, 10 (2), 246-252.
Kwon, J.H., Belanger, J.M.R. Pare, J.R.J. , 2003. Optimization of microwave assisted extraction (MAP) for
ginseng components by response surface methodology. Journal of Agriculture and Food Science,
51, 1807-1810.

SC

Li, H., Deng, Z., Wu, T., Liu, R., Loewen, S., Tsao, R., 2012. Microwave-assisted extraction of phenolics
with maximal antioxidant activities in tomatoes. Food Chemistry, 130 (4), 928-936.

M
AN
U

Li, Y., Fabiano-Tixier, AS., Abert-Vian, M., Chemat, F., 2013. Microwave-assisted Extraction of
Antioxidants and Food Colors, in Microwave-assisted Extraction for Bioactive Compounds: Theory
and Practice, edited by Farid Chmat, Giancarlo Cravotto, Springer, pp. 103-126.
Lopez-Avila, V., Young ,R., Teplitsky, N.Y.R. , 1996. Microwave-assisted process as an alternative to
Soxhlet, sonication and supercritical fluid process. J AOAC Int, 79, 142156.
Loxcel, 2013. (http://www.loxcel.com/sbux-faq.html)

TE
D

Lu,J., Zhou, C., Rong, O., Xu,Y., Zhou, B., and Zhonghai, L., 2013.Optimization of Microwave-assisted
Extraction of Flavonoids from Cryptotaenia japonica Hassk using Response Surface
Methodology. Advance Journal of Food Science and Technology 5, 3, 310-317.
Mandal, V., Mohan, Y., Hemalatha, S., 2007.Microwave Assisted Extraction An Innovative and
Promising Extraction Tool for Medicinal Plant Research. Pharmacognosy Reviews, 1, 1, 7-18.

EP

Mennen, L.I., Walker, R., Bennetau-Pelissero, C., Scalbert, A., 2005. Risks and safety of polyphenol
consumption. Am J Clin Nutr, 81(suppl), 326S9S.

AC
C

Milutinovi,M.D., iler-Marinkovi, S.S., Antonovi, D.G., Mihajlovska, K.R., Pavlovi, M.D., DimitrijeviBrankovi, S.I., 2013. The antioxidant properties of dried extracts from the spent espresso
coffee. Chemical Industry, 67, 2, 261-267.
Mukhopadhyay, A.K., 2006. Antioxidants Natural and Synhtetics, Armani International Publishers, Kiel.
Mukhopadhyay, S.,Luthri, D.L., Robbins, R.J., 2006. Optimization of extraction process for phenolic acids
from black cohosh (Cimicifugaracemosa) by pressurized liquid extraction. J Sci Food Agric, 86,
156162.
Mussatto, S. I., Ballesteros, L.F., Silvia, M., Teixeira, J. A ., 2011a. Extraction of antioxidant phenolic
compounds from spent coffee grounds. Separation and Purification Technology, 83, 173-179.

ACCEPTED MANUSCRIPT

Mussatto, S., Carneiro, L.M., Silva, J.P.A., Roberto, I.C., Teixeira, J.A., 2011b. A study on chemical
constituents and sugars extraction from spent coffee grounds. Carbohydrate Polymers, 83,2,
368-374.

RI
PT

Nelson, V. Pound,B., 2009. The Last Ten Years: A Comprehensive Review of the Literature on the
Impact of Fairtrade Natural Resources Institute (NRI), University of Greenwich
(http://www.fairtrade.org.uk/resources/natural_resources_institute.aspx)
Obruca, S., Petrik, S., Benesova, P., Svoboda, Z., Eremka, L., Marova, I., 2014. Utilization of oil extracted
from spent coffee grounds for sustainable production of polyhydroxyalkanoates. Appl Microbiol
Biotechnol, [Epub ahead of print].

SC

Passos, C.P., Moreira, A.S., Domingues, M.R., Evtuguin, D.V., Coimbra, M.A., 2014. Sequential microwave
superheated water extraction of mannans from spent coffee grounds. Carbohydr Polym, 103,
333-8.

M
AN
U

Pavlovi MD, Bunti AV, iler-Marinkovi SS, Dimitrijevi-Brankovi SI, 2013. Ethanol influenced fast
microwave-assisted extraction for natural antioxidants obtaining from spent filter coffee.
Separation and Purification Technology 118, 503-510.
Pisoschi, A.M.,and Negulescu, G.P., 2011. Methods for Total Antioxidant Activity Determination: A
Review. Biochemistry and Analytical Biochemistry, 1,106.
Radoiu, M. The scale-up of microwave-assisted extraction, International Congress on Green Extraction of
Natural Products - GENP Avignon, France, April 16 17, 2013, book of abstract, pp. 119.

TE
D

Rostagno, M. A., Palma, M., Barroso, C. G., 2007. Microwave assisted extraction of soyisoflavones.
AnalyticaChimicaActa, 588, 274282.
Saha, J., Biswas, A., Chhetri, A., Sarkar, P.K., 2011. Response surface optimization of antioxidant
extraction from kinema a Bacillus-fermented soybean food. Food Chemistry, 129, 507-513.

EP

Shahidi,F., 2000. Antioxidants in food and food antioxidants. Nahrung, 44, 3, 158-63.

AC
C

Singh, A., Sabally, K., Kubow, S., Donnelly, D. J., Gariepy, Y., Orsat, V., Raghavan, G. S. V., 2011.
Microwave-Assisted Extraction of Phenolic Antioxidants from Potato Peels. Molecules, 16 (3),
2218-2232.
Somporn, C., Kamtuo, A., Theerakulpisut, P., Siriamornpun,S., 2011. Effects of roasting degree on radical
scavenging activity, phenolics and volatile compounds of Arabica coffee beans (Coffeaarabica L.
cv. Catimor), International Journal of Food Science and Technology, 46, 22872296.
Szllsi, R., Szllosi Varga, I., 2002. Total antioxidant power in some species of Labiate (adaptation of
FRAP method). Acta Biologica Szegediensis, 46, 125127.
Upadhyay, R., K. R., Mohan Rao, L.J., 2012. Microwave-assisted extraction of chlorogenic acids from
green coffee beans.Food Chemistry, 130, 188.
Vignoli, J. A., Bassoli, D. G., Benassi, M. T., 2011. Antioxidant activity, polyphenols, caffeine and
melanoidins in soluble coffee: The influence of processing conditions and raw material. Food
Chemistry, 124, 863868.

ACCEPTED MANUSCRIPT

Wang, Z., Pan, Z., Ma, H., Atungulu, G.G., 2011. Extract of Phenolics From Pomegranate Peels. The Open
Food Science Journal, 5, 17-25.

RI
PT

Wu,X., Yu, X., Jing, H., 2011. Optimization of Phenolic Antioxidant Extraction from Wuweizi
(Schisandrachinensis) Pulp Using Random-Centroid Optimazation Methodology.Int. J. Mol. Sci.
12, 6255-6266.
Young, J.C., 1995. Microwave-assisted process of the fungal metabolite ergosterol and total fatty acids.
Journal of Agricultural Food Chemistry, 43, 29042910.

SC

Zhang, Y., Zheng, B., Tian, Y., Huang, S., 2012. Microwave-assisted Extraction and Anti-oxidation Activity
of Polyphenols from Lotus (NelumbonuciferaGaertn.)Seeds.Food Sci. Biotechnol. 21,6, 15771584.
Zhang,L., Wang, Y., Wu, D., Xu, M., Chen, J., 2011. Microwave-Assisted Extraction of Polyphenols from
Camellia oleifera Fruit Hull. Molecules, 16, 4428-4437.

M
AN
U

Zhanga, B., Yang, R., Liua, C. Z., 2008. Microwave-assisted extraction of chlorogenic acid from flower
buds of Lonicera japonica Thunb. Separation and Purification Technology, 62, 480483.

AC
C

EP

TE
D

Zuorro, A., Lavecchia, R., 2012. Spent coffee grounds as a valuable source of phenolic compounds and
bioenergy. Journal of Cleaner Production, 34, 49-56.

ACCEPTED MANUSCRIPT

Figure captions

Fig. 1. The relationship between the experimental and predicted values for TEY (A), TPC (B), DPPH
(C) and FRAP (D)

RI
PT

Fig. 2. Response surface plots showing the combined effects of ethanol and ET (I), ET and MWP (B)
and LSR and MWP (C) to TPC; ET and MWP to TEY (A); ET and MWP to DPPH (D); ET and MWP to
FRAP (E)
Fig. 3. Different bivariate plots: TEY vs. TPC (A), DPPH vs. TPC (B), DPPH vs. TEY (C), FRAP vs. TPC
(D), FRAP vs. TEY (E) and FRAP vs. DPPH (F)

AC
C

EP

TE
D

M
AN
U

SC

Fig. 4. Proposed model for cyclic reuse of coffee waste

ACCEPTED MANUSCRIPT

Table 1. Independent variables and their levels employed in a Central Composite Rotatable Design for
optimization of SCG extraction
Unit

Label

Time of extraction

-
11.0

Coded levels
-1
0
1
40.0
110 180.0

Liquid-to-solid ratio

ml/g

4.76

6.00

12.0

Microwave power

240

400

560

SC
M
AN
U
TE
D
EP
AC
C

+
209.0

RI
PT

Independent variable

13.24

ACCEPTED MANUSCRIPT

Actual
7,694b
17,245
13,840
22,064
15,269
27,387
16,031
17,541
12,713
15,870
22,195
28,053
12,433
26,768
30,620
8,408
25,741
21,534
26,678
23,109
14,729
28,270
13,823
31,216

Predicted
9,357
16,970
15,396
20,344
15,429
29,221
17,917
18,115
12,755
16,279
20,494
28,736
11,684
27,222
28,904
6,836
23,699
21,165
24,860
22,992
16,279
29,076
15,139
31,552

Actual
55.918
29.423
66.100
19.055
69.372
31.100
46.286
54.782
30.464
36.327
37.555
29.100
48.372
19.055
18.827
79.827
41.100
39.010
26.464
33.691
41.945
36.918
39.827
27.055

Predicted
49.631
29.763
64.563
24.547
59.380
30.082
42.368
50.051
34.522
37.528
40.067
33.373
55.266
19.791
17.283
82.833
44.584
46.054
24.250
27.407
37.528
34.085
41.528
30.463

SC

400 (0)
550 (1)
240 (-1)
400 (0)
240 (-1)
550 (1)
400 (0)
550 (1)
400 (0)
400 (0)
550 (1)
240(-1)
400 (0)
400 (0)
400 (0)
240 (-1)
550 (1)
240 (-1)
400 (0)
550 (1)
400 (0)
550 (1)
550 (1)
550 (1)

Response variables
TPC (% in dry extract)
DPPH (%)

M
AN
U

6,00 (-1)
4,76 (-)
12,00 (1)
6,00 (-1)
9,00 (0)
9,00 (0)
12,00 (1)
9,00 (0)
4,76 (-)
9,00 (0)
9,00 (0)
9,00 (0)
9,00 (0)
9,00 (0)
12,00 (1)
6,00 (-1)
12,00 (1)
6,00 (-1)
13,24 (+)
6,00 (-1)
9,00 (0)
13,24 (+)
6,00 (-1)
12,00 (1)

TE
D

40,00 (-1)a
110,00 (0)
40,00 (-1)
180,00 (1)
110,00 (0)
208,99 (+)
40,00 (-1)
11,01 (-)
110,00 (0)
110,00 (0)
110,00 (0)
208,99 (+)
11,01 (-)
208,99 (+)
180,00 (1)
40,00 (-1)
40,00 (-1)
180,00 (1)
110,00 (0)
180,00 (1)
110,00 (0)
110,00 (0)
40,00 (-1)
180,00 (1)

EP

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

TEY (mg/g d.w. SCG)

AC
C

Exp.
No.

Independent variables
ET (s)
LSR (ml/g) MWP (W)

RI
PT

Table 2. Experimental data on total extract yield (TEY), total polyphenol content, (TPC), DPPH, and FRAP of spent espresso coffee extracts under different
conditions of extraction time (ET), liquid-to-solid-ratio (LSR), and microwave power (MWP) based on central composite design for response surface
analysis

Actual
81,520
51,870
91,350
45,250
80,600
36,560
77,870
70,760
64,050
64,520
54,240
51,380
85,860
39,070
45,320
91,630
61,290
54,320
59,140
42,110
67,390
54,090
63,470
41,770

Predicted
77,860
51,784
92,584
47,786
78,097
35,088
77,899
69,322
62,618
66,541
55,707
48,904
84,305
41,773
47,824
92,545
64,131
56,213
62,673
39,885
66,541
51,839
64,093
39,924

FRAP (mmol Fe2+/l)

Actual
4,440
3,940
6,260
3,540
5,910
2,750
3,860
3,670
4,870
3,560
4,170
2,620
4,240
2,630
2,680
6,660
3,430
3,850
2,990
3,440
3,460
3,130
4,040
2,630

Predicted
4,723
4,282
6,142
3,173
5,498
2,751
3,947
3,370
4,445
3,896
3,733
2,890
4,320
2,127
2,396
6,919
3,227
4,181
3,347
3,566
3,896
3,184
4,004
2,790

ACCEPTED MANUSCRIPT

110,00 (0)
9,00 (0)
550 (1)
21,228
20,494
36.645
40.067
110,00 (0)
240
(-1)
22,578
24,011
13,24 (+)
31.736
38.318
110,00 (0)
9,00 (0)
240 (-1)
15,112
15,429
68.645
59.380
9,00 (0)
240 (-1)
10,451
8,471
76.464
85.386
11,01 (-)
180,00 (1)
12,00 (1)
240 (-1)
30,581
29,725
29.418
27.784
110,00 (0)
240 (-1)
11,277
11,905
4,76 (-)
71.555
64.156
Data for validation
A
40
6,78
240
7,876*
7,694
76.941*
79.641
B
40
8
240
8.259*
8.565
82.453*
80.362
a
Numbers in parentheses are coded symbols for levels of independent parameters as per text
b
Min and max values are given in bold

54,330
78,100
79,610
98,240
56,710
72,260

55,707
74,229
78,097
100,286
56,251
74,174

3,440
5,290
5,860
6,660
2,940
6,240

3,733
4,949
5,498
6,761
3,405
6,047

92,963*
93.912*

92,593
94.289

6,647*
6,782*

6,775
6.660

M
AN
U

SC

RI
PT

25
26
27
28
29
30

AC
C

EP

TE
D

*Not significant

ACCEPTED MANUSCRIPT

Table 3. Second-order polynomial equations calculated for extraction conditions of spent espresso
coffee

DPPH
FRAP

EP

TE
D

M
AN
U

A, extraction time (s), B, Liquid-to-solid-ratio (ml/g), C, microwave power (W)

AC
C

SC

TPC

Second-order polynomial equation1

YTEY = 16,195 + 5,546*A + 4,280*B + 2,533*C 1,619*A*C + 1,587*A2 +
1,264*B2 + 1,766*C2
YTPC=37.83 12.72*A 3.80*B 9.66*C + 5.66*A*C + 5.33*B*C 4.07*B2
11.90*C2
YDPPH=66,902 15,135*A + 0,019*B 11,195*C + 3,031*A*C - 1,751A2
1,948*B2
YFRAP=66,902 15,135*A + 0,019*B 11,195*C + 3,031*A*C - 1,751A2
1,948*B2

RI
PT

Response variable
Total extract yield

ACCEPTED MANUSCRIPT

Regression coefficients and ANOVA results

Model
A
B
C
AC
BC
A2
B2
C2
Lack of Fit
R-Squared
Adj R-Squared
Pred R-Squared
C.V. %

< 0.0001
< 0.0001
< 0.0001
< 0.0001
0.0002

DPPH

FRAP

P value Prob>F
<0.0001
< .0001
0.0040
<0.0001
0.0007
0.0013

< 0.0001
< 0.0001
0.9684
< 0.0001
< 0.0001

< 0.0001
< 0.0001
< 0.0001
< 0.0001
< 0.0001

0.0087
<0.0001
0.0679
0.918165
0.892127
0.844539
13.667171

AC
C

EP

TE
D

0.0005
0.0036
0.0044
0.0790
0.968599
0.958607
0.940209
7.304383

TPC

0.0117
0.0057

RI
PT

Total extract
yield

SC

Source

0.1377
0.984501
0.980458
0.97341
3.710808

M
AN
U

Table 4.

0.0005

< 0.0001
0.4664
0.943311
0.928522
0.90661
8.324233

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

ACCEPTED MANUSCRIPT

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

ACCEPTED MANUSCRIPT

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

ACCEPTED MANUSCRIPT

AC
C

EP

TE
D

M
AN
U

SC

RI
PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

HIGHLIGHTS

EP

TE
D

M
AN
U

SC

RI
PT

RSM was used to monitor the extraction characteristics of espresso SCG

The effect of extraction time, liquid-to-solid ratio, and microwave power were investigated
The results obtained from all models were highly correlated
The FRAP and the DPPH antioxidative activity showed substantial correlation with TPC
The model can be useful tool for optimization at industrial scale

AC
C