Accepted Manuscript

What kind of coffee do you drink? An investigation on effects of

eight different extraction methods

Giulia Angeloni, Lorenzo Guerrini, Piernicola Masella, Maria

Bellumori, Selvaggia Daluiso, Alessandro Parenti, Marzia
Innocenti

PII: S0963-9969(18)30809-3
DOI: doi:10.1016/j.foodres.2018.10.022
Reference: FRIN 7992
To appear in: Food Research International
Received date: 8 June 2018
Revised date: 1 October 2018
Accepted date: 7 October 2018

Please cite this article as: Giulia Angeloni, Lorenzo Guerrini, Piernicola Masella, Maria
Bellumori, Selvaggia Daluiso, Alessandro Parenti, Marzia Innocenti , What kind of coffee
do you drink? An investigation on effects of eight different extraction methods. Frin
(2018), doi:10.1016/j.foodres.2018.10.022

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

What kind of coffee do you drink? An investigation on effects of eight different

extraction methods

Giulia Angelonia,* [email protected] , Lorenzo Guerrini a, Piernicola Masella a, Maria Bellumori b,

Selvaggia Daluiso b, Alessandro Parenti a, Marzia Innocenti b.

a
Department of Management of Agricultural, Food and Forestry System, University of Florence, Italy

b
Department of NEUROFARBA, Division of Pharmaceutical and Nutraceutical Sciences, via U. Schiff 6,

PT
Sesto F.no, Florence, Italy

RI
*Corresponding author at: Department of Management of Agricultural, Food and Forestry System, University

SC
of Florence, Italy. Piazzale delle Cascine 16, 50144, Firenze, Italy.
NU
Abstract
The chemical composition of brewed coffee depends on numerous factors: the beans, post-harvest
MA

processing and, finally, the extraction method. In recent decades, numerous coffee-based
beverages, obtained using different extraction techniques have entered the market. This study
characterizes and compares eight extraction coffee methods from a chemical-physical point of
ED

view, starting from the same raw material. Specifically, three types of Espresso, Moka, French
Press, and 3 filter coffee that for the first time are reported in the scientific literature Cold Brew,
T

V60, and Aeropress are compared.

EP

Physical measurements included the quantification of total dissolved solids, density, pH,
conductivity, and viscosity. Chemical analyses identified 15 chlorogenic ac ids (CGAs): six
caffeoylquinic acids, one p-Coumaroylquinic acid, one Feruloylquinic Acid, four Caffeoylquinic
C

lactones, and three Dicaffeoylquinic acids. Maximum caffeine and CGA concentrations were found
AC

in Espresso coffees, while Moka and filtered coffees were three to six times less concentrated. The
classic Espresso method was most efficient for caffeine and CGA recovery, with a yield almost
double that of other methods. Per-cup caffeine and CGAs were higher in Cold Brew than Espresso
coffees, as a function of the volume of beverage, which ranged from 30 mL (for espresso) to 120
mL (for filtered coffees). In light of these results, it is not possible to establish how many cups of
coffee can be consumed per day without exceeding the recommended doses, since according to
the applied brewing method, the content of the bioactive substances varies considerably.
Keywords: brewing methods, coffee extraction, bioactive compounds, caffeine, chlorogenic acids

1
 ACCEPTED MANUSCRIPT

1. Introduction

Coffee is one of the most widely-consumed beverages worldwide (ICO, 2016), and numerous
brewing and extraction methods are used depending on the geographic, c ultural and social
context, not to mention personal preferences. Typically, its preparation involves three main stages.
First, the green beans are roasted. Following this, the roasted beans are ground to facilitate
extraction during the final, brewing, stage. In beverage form, quality characteristics such as smell,
taste, color, and body are relevant, and highly appreciated attributes (Nunes, 1997). The flavor of a
freshly-prepared cup of coffee is the final expression, and perceptible result of a long chain of

PT
transformations (Yeretzian et al., 2002).
This complex beverage contains over 1000 compounds that are responsible for its pleasant flavor

RI
and aroma (Nijssen et al.,1996). Of these, caffeine (1,3,7-trimethylxanthine) is the most widely
studied. Caffeine exerts most of its biological effects through the antagonism of the adenosine

SC
receptor inducing generally stimulatory effect in the central nervous system (Cano-Marquina et al.,
2013; Bae et al 2014). Infact, its positive effects are well-known; in particular, improvements
NU
related to cognitive abilities such as better perception, reduced tiredness, and shorter duration of
sleep (Borota et al., 2014). Recently, it was demonstrated that the risk of Alzheimer disease was
lower in those who regularly consume caffeine-containing coffee than those who did not drink it. In
MA

addition, the physiolocal effects of caffeine intake include acute elevation of blood pressure,
increasing metabolic rate and diuresis (Bae et al 2014).
The alkaloid is heat stable, and the amount present in raw coffee can vary significantly depending
ED

on many factors, among which the most important are origin and cultivar. Its concentration and
biological activity depend on a blend of factors, such as raw materials (Arabica or Canephora)
T

(Severini et al., 2017), agricultural practices (traditional or organic), post-harvest techniques (wet or
EP

dry), duration and conditions of storage, roasting degree (light, medium, or dark), roasting process
(standard or torrefacto), type of commercial coffee (ground roasted or instant), and grinding and
C

brewing method (boiled, filtered, or expresso). Altogether, this means that we never drink two cups
of coffee with the same chemical composition, even when they come from the same outlet (De
AC

Mejia et al., 2014).

Many studies have demonstrated that coffee is one of the most important sources of polyphenols
and caffeoylquinic acids (CQAs) (Kamiyama et al., 2015).
The major polyphenol in coffee in chlorogenic acid and it is one of the major strong antioxidant
compounds in coffee (Bae et al 2014).
It is known that, on average, about one third of the ingested amount of chlorogenic acids through
coffee can be absorbed in the human gastrointestinal tract, metabolized in the stomach, intestine,
liver, and kidney and can probably exert a series of beneficial biological properties in the body,

2
 ACCEPTED MANUSCRIPT

explaining at least partially why coffee consumption has been associated with higher longevity and
lower incidence of various degenerative and nondegenerative diseases in epidemiological studies
(Farah A., and Duarte G. 2015).
These water-soluble acids are abundant in coffee, and they are formed by the coffee plant through
esterification of trans-cinnamic acids (most notably caffeic, ferulic, and p-coumaric) with quinic acid
(Higdon, 2006). CGAs and their derivatives are known to contribute to the acidity, astringency, and
bitterness of the final coffee beverage (Scholz and Maier, 1990; Trugo and Macrae, 1984). The
main CGAs are 5-O-caffeoylquinic acid (5-CQA), and its isomers 3-O-caffeoylquinic acid (3-CQA)

PT
and 4-O-caffeoylquinic acid (4-CQA), which together account for 80% of total CGAs (Farah and
Donangelo, 2006; Moeenfard et al., 2014)

RI
Coffee preparation is a solid–liquid extraction process, involving: (1) water absorption by ground
coffee; (2) mass transfer of soluble solids from ground coffee into hot water; and (3) separation of

SC
the resulting extract from spent solids. Several variables can modify in-cup coffee quality, including
the contact time between the water and ground coffee, extraction time, the ground coffee/water
ratio, water temperature and pressure (for espresso coffee), type of filter, and the boiling process.
NU
All of these factors play important roles in modifying caffeine content and other compounds
(Gloess et al., 2013; Niseteo et al., 2012; Andueza et al., 2003; Andueza et al., 2007).
MA

There are many ways to prepare coffee and consumer preferences for a particular mode are
influenced by various factors such as lifestyle, culture, and flavor preferences (Illy, 2005). Of the
various brewing methods that use pressure, the most famous is the espresso machine. Espresso
ED

coffee (EC) is one of the most appreciated brews; the term espresso is derived from the Italian
word for ‘express’ since expresso is made for, and served immediately to, the customer. EC is
T

prepared on request from roasted and ground coffee beans. A limited amount of pressurized hot
water quickly percolates through a ground coffee cake to yield a small cup of concentrated foamy
EP

beverage (Petracco, 2001). The original EC formulation used 7 gr of coffee powder to obtain
around 30 gr of expresso beverage. Nowadays, there are many different recipes, of which the most
C

popular is specialty coffee. This preparation uses 7 gr of coffee powder to produce 14 gr of

AC

expresso beverage. As every gram of ground coffee turns into 2 grams of liquid the final beverage
is a strong expresso with an extraction formula of 50% (SCAA 2016). Recently, a new expresso
brewing method, namely Caffè Firenze (EU Patent 06 023 798.9; US 2010/0034942 A1) has been
developed, which uses a sealed chamber and pressurized air (Masella et al., 2015). Another
pressurized method is the Moka pot. Traditionally, this is the most popular method in Italian homes
as the machine is cheap, and it is quick to brew. However, quality is often compromised as the risk
of over or under extraction is high (depending on the grind).
Lungo is an alternative to EC. This less-intense beverage is characterized by a different water/
ground coffee ratio and a larger cup size (100–250 ml), depending on cultural habits. Numerous
3
 ACCEPTED MANUSCRIPT

brewing methods may be used to prepare lungo coffee: steeping using a French press, filtration or
dripping in the V60, Aeropress and cold drip technique, and boiling.
Standard preparation methods have been developed for different types of extraction. These
methods differ in terms of the process, grams of coffee, amount of water, and grain size of ground
coffee. Several studies have compared these different techniques, and described the
physicochemical attributes and sensory profile of the coffees that are produced (Andueza et al.,
2003; Gloess et al., 2013; Caporaso et al., 2014; Parenti et al., 2014; Masella et al., 2015). These
studies reveal that there is no ‘best’ extraction method, but that each technique has its own

PT
characteristics. This study extends the literature and examines several new brewing techniques
that are already well-known by baristas and consumers, but for which there are, as yet, no data.

RI
The aim was to describe and compare eight extraction methods: three espresso systems , classic
(EC), specialty espresso (ECS), and Caffè Firenze (ECF); one cold brew system (Cold Brew); and

SC
four filter methods (V60, Aeropress, French Press, and Moka) that use different pressures and filter
techniques. These methods were characterized by the analysis of physicochemical parameters.
This was supplemented by an in-depth investigation of caffeine and CGA content based on high-
NU
performance liquid chromatography with diode-array detector (HPLC-DAD) analyses. Quantitative
data related to bioactive substances were expressed as concentration (mg/mL of beverage),
MA

extractive capacity (mg/g of ground coffee) and per-cup dosage (mg/cup).

This study provides a comprehensive scientific overview of the most common coffee extraction
methods currently used worldwide. It compares eight different extraction methods in term of it
ED

provides the concentration (mg/mL), extraction capacity (mg/g), and per-cup content of caffeine
and CGA. To the best of our knowledge, this is the first time that data for Cold Brew, V60, and
T

Aeropress techniques are reported in the literature.

EP

2. Materials and methods

C

2.1. Experimental design

AC

The experiment was designed to highlight differences between extraction methods in terms of the
physicochemical characteristics of brewed coffee, and its sensory aspects. A specific recipe was
followed for each of the eight methods. Standardized procedures were developed that differed in
terms of the grind, the amount of coffee used, water temperature and, last but not least, the
equipment. The extraction parameters were summarized in Table 1. Six replicates were performed
for each brewing method. The order of beverage preparation was completely randomized.

2.2. Coffee samples and extraction methods

4
 ACCEPTED MANUSCRIPT

The same batch of 100% Arabica coffee (Ethiopian Gera Estate) was used for all extractions. Each
pack of beans (250 g) was opened immediately before brewing to avoid oxidative damage. Beans
were ground using a professional grinder (EK43 Mahlkönig AG, Switzerland). Coarse-ground
coffee was used for all lungo and filter methods (Clarke, 2008), while a fine grind was used for
expresso and Moka methods. Size distribution was analyzed using laser diffractometry, which is
suitable for ground coffee particles ranging from 5–2000 µm. As water quality plays an important
role in coffee beverage quality (Navarini and Rivetti, 2010) all samples were prepared using the
same commercial brand of mineral water.

PT
EC Espresso classical method

RI
A conventional bar machine (GS3, LaMarzocco, Italy) was used. Two cups of EC were prepared
(14.5±0.2 g). Physicochemical analyses were only performed on one of the two ECs. Extraction

SC
parameters were: water temperature 92 °C, water pressure 9 bar, and 30 s of percolation time,
assuming an optimal flow rate of about 1 ml s −1 (Illy and Viani, 2005).
NU
ECS Espresso Specialty method
MA

ECS was produced with the bar machine described above. This preparation follows the Specialty
Coffee Association of America (SCAA) standard procedure (SCAA, 2016), and differs from the
classic method in two respects: more coffee powder (18 g), and slower percolation (25 s).
ED

ECF Espresso Caffè Firenze

T

Caffè Firenze (ECF) samples (Patent 06 023 798.9; US 2010/0034942 A1) were produced
EP

following the procedure given in Masella et al., 2015. The method uses a sealed extraction
chamber in which water and air are at higher pressures than other extraction methods, resulting in
C

a pronounced difference in foam characteristics.

AC

Cold Brew

Samples were prepared using cold drip equipment with 25 g coffee powder and 250 mL mineral
water at room temperature (22 °C). Equipment comprised three parts. An upper (glass) section,
containing water, was equipped with a tap. The tap was used to control the flow rate and extraction
time. The coffee/water mixture was placed in a central container. Water entering from above
passed through a filter and into a lower carafe, where the final brew was collected. Spent coffee
grounds were retained in the filter. The average extraction time was approximately 5.5–6 h.

5
 ACCEPTED MANUSCRIPT

Moka

A three-cup expresso maker was used (Bialetti Industrie SpA, Italy). Moka is the most popular
technique in Italian households. Samples were produced following the procedure given in Navarini
et al., 2009.

French Press

Coarse-ground coffee (25 g) and hot water (250 g at 95 °C) were mixed in a brewer fitted with a

PT
mesh plunger. The mixture was brewed for 5 min, then the plunger was pressed to trap coffee
grounds at the bottom of the container, following the SCAA standard procedure (SCAA 2016).

RI
V60

SC
This coffee maker consists of three parts: a cone-shaped upper dripper with ridges along the inner
edges and a single, large hole at the bottom, a paper filter, and a glass ves sel (Hario server, 300
NU
mL). Water was poured into the V60 to create a small crater in the middle of the ground coffee.
Next, 70 mL of water at 98 °C, was poured over the coffee, which was left to pre-infuse for 30 s.
MA

Finally, 180 mL of water was added in concentric circles and left to drawdown for three minutes.
The brew ratio was 60 g/L.

Aeropress
ED

The Aeropress was invented in 2005 by Aerobie; the device consists of two nested cylinders. One
T

has a flexible airtight seal, and fits inside the larger cylinder, similar to a syringe. The procedure
EP

was as follows: first, 16.5 g of ground coffee was put into the cylinder, and then 250 mL of water at
93 °C was added. Coffee was steeped for one minute and then forced through a filter by pressing
the plunger through the tube. Paper filters were used. The average quantity of beverage obtained
C

was 215 mL.

AC

2.3. Physicochemical analyses

2.3.1. Physical analyses

All samples were brought to 20 °C before selected parameters were analyzed and evaluated. A
digital pH meter (GLP 21, Crison Instruments, Spain) was used to determine pH. Viscosity was
measured with a capillary viscometer (Ostwald-type) fitted with an automatic optical reader
(ViscoClock, Schott Instruments, Germany) and expressed as mN s m −2. Relative density was

6
 ACCEPTED MANUSCRIPT

measured with a 25 mL pycnometer. Total dissolved solids (TDS) was measured using a
refractometer (VST LAB Coffee III Refractometer, USA) to calculate extraction yields. TDS was
converted into the total percentage of ground coffee dissolved in the brewed coffee: Total Coffee
Brewed (g) * TDS % / powder used (g).

2.3.2. Analyses of caffeine and CGAs

Coffee samples were centrifuged at 12000 rpm for 5 min and diluted 1:10 with water before HPLC-
DAD analysis.

PT
HPLC was carried out using an Agilent HP 1100 system equipped with an autosampler, column
heater module and quaternary pump, coupled to a diode array detector (DAD) all from Agilent

RI
Technologies (Palo Alto, CA, USA). A 150 mm × 3 mm i.d., 2.7 μm Poroshell 120, EC-C18 column
(Agilent Technologies) was used, equipped with a pre-column of the same phase, and maintained

SC
at room temperature. Injection volume was 5 μL. The elution method was performed at a flow rate
of 0.4 mL/min using water at pH 3.2 by formic acid (solvent A) and acetonitrile (solvent B). All
NU
solvents used were Chromasolv for HPLC grade (Sigma Aldrich S.R.L). The multistep linear
solvent gradient technique is described in detail in Angeloni et al. (2018). Starting from 95% A, up
to 10% A, over 24 min (the total analysis time) UV–vis spectra were recorded in the range 220–600
MA

nm. Chromatograms were registered at 330 nm for CGAs, and 278 nm for caffeine. Caffeine and
CGAs were identified by comparing their retention times, UV–vis spectra to those of the respective
standard, when it was possible, or with published data (Angeloni et al 2018). CGAs were evaluated
ED

by HPLC- DAD using a five-point calibration curve of chlorogenic acid (purity 99 %)

(Extrasynthèse, Genay, France) at 330 nm (0-1.776 g; r2=0.9991) and caffeine content was
T

determined by HPLC-DAD using a six-point calibration curve from Extrasynthèse (purity 95%) at
EP

278 nm (0-0.632 g; r2=0.9994).

Quantitative data related to bioactive substances were expressed as concentrations (mg/mL of
C

beverage), extractive capacity (mg/g of coffee powder) and per-cup dosage (mg/cup).
AC

2.4. Cluster analysis

Cluster analysis is an exploratory, multivariate technique used to explore the data structure and
overall characteristics when little (or even no) information about group structure is available (Ares
2014). It is a convenient method for identifying homogenous groups of objects. Objects (in our
case, brewing methods) in a specific cluster share many characteristics and are dissimilar to
objects not belonging to that cluster (Sarstedt and Mooi, 2014). It is a hierarchical approac h, based
on the determination of the distance between objects (degree of similarity/dissimilarity), and the
application of an agglomerative (amalgamation) method to establish clusters of n-objects.
7
 ACCEPTED MANUSCRIPT

Variables included in the analysis were physical measurements, and concentrations (mg/mL) of
caffeine and CGAs for each brewing method.

2.5. Statistical analyses

Conventional analysis of variance (ANOVA) was used to compare means and standard deviation
determined for the different extraction methods. The tested factors were considered significantly
different at p<0.05. All statistical analyses were performed using R software (version 3.4.0 for
Windows).

PT
3. Results and discussion

RI
Extraction parameters were optimized for each brewing method in order to follow, as closely as

SC
possible, the settings used by baristas, while guaranteeing the best possible comparability.

3.1. Cluster analysis

NU
Homogenous groups of brewing techniques were identified by a cluster analysis. As shown in
Figure 1, cluster analysis made it possible to divide the eight methods into two main groups, with
MA

four subclasses in each group: the first group comprised Cold Brew, Aeropress, French Press, and
V60 and a second included Moka, ECF, ECS, and EC.
Similar concentrations were frequently found for these two groups of extraction methods. Within
ED

the filter group the French Press method could be distinguished from the other methods, probably
due to a different time of extraction and temperature, as reported in Table 1.
T

Within the expresso group, another differentiation was found between ECS–EC and ECF–Moka,
EP

confirmed by the results of physicochemical analyses.

As expected, EC and ECS resulted similar because the extraction method was the same and the
C

only difference it was in the ratio of powder/water.

AC

3.2. Physical analyses

The physical characterization of the coffee beverage produced using the different preparation
methods is shown in Table 2. This analysis highlighted significant differences between the eight
brewing methods for TDS %, extraction %, and viscosity. Concerning TDS %, the highest values
were found for ECS followed by EC, Moka and ECF methods. No difference was found among the
remaining extractive methods, where values were lower. TDS % directly correlates with coffee
strength: high TDS % is consistent with a strong brew. It reflects the level of extraction of the
coffee. High temperature and pressure increase extraction yield and rate, seen in the difference
8
 ACCEPTED MANUSCRIPT

between expresso and Moka coffees, and filtered brews (López-Galilea et al., 2007). It is well-
known that TDS % affects the sensory property described as ‘body’ (Gloess et al., 2013), and
seems to be related to the coffee/water ratio (Andueza et al., 2007), and the brewing procedure
(López-Galilea et al., 2007). Although the literature contains no data related to TDS, this factor is
employed by baristas, and is recommended by SCAA to assess the correct degree of extraction.
Concerning extraction %, the highest value was found for Moka (28.6±1 %) and the lowest value
for ECF. Intermediate values were recorded for the other two expresso preparations, EC and ECS.
Percentages were similar for Cold Brew and Aeropress, although different quantities of ground

PT
coffee were used. The value for the V60 method was similar to the EC method, and the value for
the French Press method was similar to the ECS method. SCAA guidelines state that extraction %

RI
should be in the range 18–23%. Our data is generally consistent with this range, except for ECF
(which appears to be under-extracted), and Moka (which appears to be over-extracted).

SC
Relating viscosity, Moka and ECF were similar to each other but different from other expresso
coffees. No significant differences were found among the remaining methods (V60, Aeropress,
Cold Brew, and French Press).
NU
No significant differences were found for densities, which were around 1.05 g/mL, and for pH
values, which were around 5.16.
MA

3.3. Chemical analyses

ED

The qualitative profile of bioactive substances detected by HPLC-DAD was almost the same for all
samples. A total of 15 CGAs were detected. Figure 2 presents chromatographic profiles at 278 and
T

330 nm. Peaks were identified based on UV spectra and elution/retention sequences reported in
the literature, and confirmed by their mass spectrometric behavior, as reported in our earlier work
EP

(Angeloni et al., 2018).

Fujioka and Shibamoto (2008) report that the most abundant CGAs in coffee are caffeoylquinic
C

acids (CQAs), notably 5-O-caffeoylquinic (5-CQA) followed by its isomers 3-and 4-CQA.
AC

Dicaffeoylquinic acid (3,4-, 3,5- and 4,5-diCQA), feruloylquinic acid (3-, 4-and 5-FQA),
diferuloylquinic acid (dFQA) and p-coumaroylquinic acid (3-, 4- and 5-p-CoQA) isomers were also
found in our samples, although less abundant.
Any comparison of caffeine and CGAs must take into consideration the fact that every operational
condition (e.g. particle size and dose of ground coffee, tamping, water temperature and pressure,
coffee/ water ratio, and the final volume of the drink) create considerable differences in bioactive
compound extraction kinetics. Of these, one of the most important factors is the ratio of ground
coffee to the final volume of water (Andueza et al., 2007). For this reason, the results of chemical
analyses are presented in three ways: concentration (mg/mL), extraction efficiency (mg/g of ground
9
 ACCEPTED MANUSCRIPT

coffee), and total bioactive content per cup (mg/cup), (Tables 3, 4, and 5 respectively).
Furthermore, Figure 3 (a, b, and c) reports mean values for caffeine and total CGAs

3.3.1. Concentration of bioactive compounds (mg/mL)

Table 3 shows that there was a significant difference in caffeine concentration for the methods
tested (p ≤ 0.05). Values were highest for ECS and EC, on the contrary lowest concentrations were
observed for Aeropress, V60 and French Press methods. Significant differences were found
between these groups and other extraction methods (Cold Brew, ECF, and Moka).

PT
These data agree with Severini (2017), who assessed the main variables that affect caffeine
concentrations in coffee-based beverages. Several studies have indicated that caffeine content

RI
ranges from 2.4 to 4.5 mg/mL for expresso (25 mL), from 0.4 to 1.4 mg/mL for American or filtered
(200 mL), from 0.2 to 0.5 for French or Plunger (100 mL), and from 0.7 to 5.4 mg/mL for Moka

SC
(30 mL) (López-Galilea et al., 2007; Caporaso et al., 2014). Caffeine is moderately soluble in water
at room temperature 20 °C (1.46 mg/mL), it increases at 80 °C (180 mg/mL), but becomes very
NU
soluble at 100 °C (670 mg/mL) (Prankerd 2007). Despite the lower solubility of caffeine in water at
room temperature, data for the Cold Brew method shows that concentrations are similar to Moka
and ECF. This fact could be explained by the extensive contact time between water and the
MA

ground coffee (around six hours). Regarding ECF, the lower caffeine concentration could be due to
the fact that the chamber in which the coffee panel was placed in direct contact with water at 75 °C
(Masella et al., 2015). Consequently, water that is in contact with the coffee panel is at a lower
ED

temperature than classic espresso.

Concerning CGAs, CQAs dominated for all preparations ranging about 75% of the total, followed
T

by CQLs (about 12%) then di-CQAs (about 7%), 5-FQAs (about 4.5%) and finally 5-pCoQAs
EP

(about 1.5%) according to previous literature data (Ludwig et al., 2012). Moreover, 5-CQA was
always the most abundant compound, ranging from 35–39% of total CGAs (for ECF and Moka,
C

respectively), followed by 4-CQA and 3-CQA. CGA concentrations followed the trend observed for
caffeine. For all 15 CGAs, values were highest for EC and ECS preparations. An interesting finding
AC

is that ECF, Cold Brew, and Moka methods have a mean total CGA concentration that is
significantly different from the other two expresso methods, and from Aeropress, French Press and
V60 preparations (p ≤ 0.05). Intermediate values were found for the latter (Table 3 and Figure 3a).
Several studies have assessed the influence of contact time and brew ratio on bioactive compound
extraction (Andueza et al., 2007; Crozier et al., 2009; Caprioli et al., 2015). The results show that
most extractable compounds are brought into solution in the first few seconds of the extraction
process under higher pressure, as previously reported by Ludwig at al 2012, that evidenced the

10
 ACCEPTED MANUSCRIPT

technological differences between espresso and filter coffeemaker. This could explain the highest
CGA concentrations in EC and ECS coffees compared to the other preparation methods.
These trends agree with the results reported by Gloess et al. (2013), in which the highest
concentration of CGAs was reported for espresso, followed by Moka and, finally, filter coffee. In
this earlier work, concentrations ranged from 17.0 mg/mL for expresso, to 2.43 mg/mL for French
Press. The present study evaluated five other methods that are not widely known in the scientific
literature; of these, concentrations in at least three methods (Aeropress, French Press, and V60),
were comparable to those of the filter coffees reported by Gloess et al. (2013).

PT
3.3.2. Extraction efficiency (mg/g ground coffee)

RI
Extraction efficiency can be defined as the ratio of the mass of ground coffee powder that passes
into the cup, and the total amount of ground coffee used (Clarke, 2008). Table 4 shows that there

SC
was a significant difference in extraction efficiency among all 15 CGAs, for the tested methods
(p ≤ 0.05, letters indicate statistically significant differences between groups). The analysis showed
NU
that extraction efficiency was highest for the EC method, both for caffeine and all CGAs.
Specifically, for EC caffeine extraction efficiency was about double that of the ECS method
(17.4±0.62 mg/g compared to 8.5±0.12 mg/g for ECS). Given that the extraction time was similar
MA

(25±5 s), this observation could be explained by the different ground coffee/mL beverage ratio
(7g/30mL for EC and 9g/18mL for ECS). For Moka, although the concentration was similar to that
of ECF, extraction efficiency was similar to V60, Cold Brew, and Aeropress. This could be
ED

explained by the contact time, which was much longer than that used for expresso preparation
(25±5 s). Finally, extraction efficiency was lowest for ECF (5.76 ±0.33 mg/g).
T

Concerning CGA concentrations, trends were similar to those for caffeine for all 15 detected
EP

compounds. Figure 3b show that EC was able to extract 52.09±4.81 mg/g of total CGAs, with an
extraction capacity about twice that of ECS, Moka and ECF. French Press and ECF they were
C

been least efficient and significantly different to V60, Cold Brew, and Aeropress methods. These
trends agree with earlier data (Gloess 2013), which found highest concentrations of the most
AC

abundant CGAs for expresso, followed by Moka and filter coffee.

3.3.3. Bioactive content per cup

In the context of caffeine and CGA content in a coffee brew, some factors must be taken into
consideration. First, the usual amount of coffee in a cup varies enormously in different cultures and
traditions, ranging from 18-30 mL for expresso, to over 200 mL for filtered coffee. Therefore, we
adopted a ‘typical’ volume for each type of beverage: 30 mL for expresso; 18 mL for ECS; 40 mL
for Moka; and 120 mL for the other types. Romani et al. (2004) argues that the ratio between the
11
 ACCEPTED MANUSCRIPT

dose of ground coffee, and volume of coffee is a variable that strongly affects the final caffeine
content in the Espresso cup. Similarly, it is reasonable to affirmative that this could explain the high
caffeine content in a cup of Cold Brew coffee (149.52±13.80 mg/cup).
As reported in Table 5, EC contained much more caffeine than ECS. However, these two expresso
were prepared with different cup volumes the ECS cup being almost half the size of the EC cup.
Caffeine content for a cup of Moka and ECF was lower than for the other expresso methods,
although the ANOVA analysis found that these two methods were not significantly different from
each other, they showed different to other extraction methods. High per-cup levels of caffeine were

PT
found for V60 and Aeropress methods, these values were lower than the Cold Brew method, and
different to the other methods.

RI
Concerning per-cup CGA content, the same trend was observed for all individual compounds. The
highest level was observed for Cold Brew followed by EC. As reported in Table 5 and Figure 3c,

SC
highest concentrations of all 15 compounds were detected for the Cold Brew method (sum of
CGAs 433.25±52.50 mg/cup). This result was expected as extraction is cold, limiting the
degradation of compounds.
NU
This information is relevant in the context of the maximum recommended daily dose of caffeine. In
2012, the FDA (2012) stated that, for healthy adults, a dose of caffeine up to 400 mg/day was not
MA

associated with adverse effects. This work highlights that the intake of bioactive components is
highest for lungo coffee, although the consumer often considers that a long coffee is more diluted
and therefore contains less bioactive substances.
ED

4. Conclusions
T

This study provides important information on concentrations (mg/mL), extraction capacity (mg/g),
EP

and per-cup caffeine and CGA content for eight types of beverage preparation. Some of these
methods, which are very popular among consumers and industry experts, have not previously
C

been investigated in the scientific literature. Here, they are assessed and compared for the first
time.
AC

Technical differences in these extraction methods led to quantitative differences in extraction

efficiencies, and produce coffees with different profiles. In general, the concentration of bioactive
compounds was higher for the expresso group than the filter group. However, when content per
cup was compared, filter coffees were found to have a higher content. The cluster analysis
identified clear differences between and among these two groups. Clusters can be distinguished
based on caffeine and CGA concentrations.

12
 ACCEPTED MANUSCRIPT

This study reviewed extraction methods for coffee production. The aim was not to establish “the
best method” but to highlights that different extraction methods produce coffee beverages with
different qualitative and quantitative characteristics, starting from the same raw material.
In light of these results it is not possible to establish how many cups of coffee can be consumed
per day without exceeding the recommended doses, since according to the applied brewing
method, the content of the bioactive substances varies considerably.

PT
References

RI
Angeloni, G., Guerrini, L., Masella, P., Innocenti, M., Bellumori, M., Parenti, A. (In press).
Characterization and comparison of cold brew and cold drip coffee extraction methods. Journal of

SC
the Science of Food and Agriculture

Andueza, S., Vila, A.M, De Peña, M.P, Cid, C. (2007). Influence of coffee/water ratio on the final
NU
quality of espresso coffee. Journal of the Science of Food and Agriculture, 87, 586-592.
MA

Andueza, S., Maeztu, L., Pascual, L., Ibáñez, C., De Peña, M.P., Cid,C. (2003). Influence of
extraction temperature on the final quality of espresso coffee. Journal of the Science of Food and
Agriculture, 83, 240-248.
ED

Ares, G. (2014). Cluster analysis: application in food science and technology. Mathematical and
Statistical Approaches in Food Science and Technology, 103-119. Oxford: Wiley Blackwell.
T

Bae J.H., Park J.H., Im S.S., Song D.K. (2014). Coffee and health. Integr Med Res, 3,189–191.
EP

Borota, D., Murray, E., Keceli, G., Chang, A., Watabe, J. M., Ly, M., & Yassa, M. A. (2014). Post-
C

study caffeine administration enhances memory consolidation in humans. Nature

AC

neuroscience,17(2), 201-203.

Cano-Marquinaa A., Tarínb J.J., Canoc A. (2013) . The impact of coffee on health. Maturitas, 75,
7–21.

Caporaso, N., Genovese, A., Canela, M. D., Civitella, A., & Sacchi, R. (2014). Neapolitan coffee
brew chemical analysis in comparison to espresso, moka and American brews. Food Research
International, 61, 152-160.

13
 ACCEPTED MANUSCRIPT

Caprioli, G., Cortese, M., Sagratini, G., & Vittori, S. (2015). The influence of different types of
preparation (espresso and brew) on coffee aroma and main bioactive constituents." International
journal of food sciences and nutrition, 66(5), 505-513.

Clarke, R., & Vitzthum, O. G. (2008). Coffee: recent developments. John Wiley & Sons.

Clifford, MN. (2000).Chlorogenic acids and other cinnamates – nature, occurrence, dietary burden,
absorption and metabolism. Journal of the Scince of Food Agriculture, 80,1033-1043.

PT
Crozier, A., Jaganath, I., Clifford, M. (2009). Dietary phenolics: chemistry, bioavailability and effects
on health. Natural Product Reports , 26,1001–1043.

RI
De Mejia, E. G., & Ramirez-Mares, M. V. (2014). Impact of caffeine and coffee on our health.

SC
Trends in Endocrinology & Metabolism, 25(10), 489-492.

Farah, A., Donangelo, C. M. (2006). Phenolic compounds in coffee. Brazilian Journal of Plant
NU
Physiology, 18(1),23–36.

Farah, A., & Duarte, G. (2015). Bioavailability and metabolism of chlorogenic acids from coffee.
MA

In Coffee in Health and Disease Prevention (pp. 789-801).

FDA. Food and Drug Amministration. (2012). http://www.fda.gov.

ED

Fujioka, K., & Shibamoto, T. (2008). Chlorogenic acid and caffeine contents in various commercial
T

brewed coffees. Food Chemistry, 106(1), 217-221.

EP

Gloess, A. N., Schönbächler, B., Klopprogge, B., Lucio, D., Chatelain, K., Bongartz, A. & Yeretzian,
C. (2013). Comparison of nine common coffee extraction methods: instrumental and sensory
C

analysis. European Food Research and Technology, 236(4), 607-627.

AC

Higdon, J. V., & Frei, B. (2006). Coffee and health: a review of recent human research.Critical
reviews in food science and nutrition, 46(2), 101-123.

ICO. International Coffee Organization. (2016). http://www.ico.org/monthly_coffee_trade_stats.asp.

Accessed: 11-11-2016.

Illy, A., & Viani, R. (2005) Espresso coffee: the science of quality. Academic Press.

14
 ACCEPTED MANUSCRIPT

Kamiyama, M., Moon, J. K., Jang, H. W., & Shibamoto, T. (2015). Role of degradation products of
chlorogenic acid in the antioxidant activity of roasted coffee. Journal of agricultural and food
chemistry, 63(7), 1996-2005.

López-Galilea I, de Peña MP, Cid C. (2007). Correlation of selected constituents with the total
antioxidant capacity of coffee beverages: Influence of the brewing procedure. Journal of
Agricultural and Food Chemistry, 55,6110-6117.

Ludwig, I. A., Sanchez, L., Caemmerer, B., Kroh, L. W., De Peña, M. P., & Cid, C. (2012).

PT
Extraction of coffee antioxidants: Impact of brewing time and method. Food Research
International, 48(1), 57-64.

RI
Masella, P., Guerrini, L., Spinelli, S., Calamai, L., Spugnoli, P., Illy, F., & Parenti, A. (2015). A new

SC
espresso brewing method. Journal of Food Engineering,146, 204-208.

Moeenfard, M., Rocha, L., & Alves, A. (2014). Quantification of caffeoylquinic acids in coffee brews
NU
by HPLC-DAD. Journal of analytical methods in chemistry.
MA

Navarini, L. & Rivetti, D. (2010). Water quality for espresso coffee. Food Chemestry, 122(2) 424-
428.

Navarini, L., Nobile, E., Pinto, F., Scheri, A., & Suggi-Liverani, F. (2009). Experimental
ED

investigation of steam pressure coffee extraction in a stove-top coffee maker. Applied Thermal
Engineering, 29(5), 998-1004.
T
EP

Niseteo, T., Komes, D., Belščak-Cvitanović, A., Horžić, D., & Budeč, M. (2012). Bioactive
C

composition and antioxidant potential of different commonly consumed coffee brews affected by
AC

their preparation technique and milk addiction. Food Chemistry., 134,1870-1877.

Nijssen, L.M. , Visscher, C.A. , Maarse, H. , Willemsense, L.C. , Boelens, M.H. (1996). Volatile
compounds in food. TNO Nutrition and Food Research Institute, Zeist, The Netherlands

Nunes, F. M., Coimbra, M. A., Duarte, A. C., & Delgadillo, I. (1997). Foamability, foam stability, and
chemical composition of espresso coffee as affected by the degree of roast. Journal of Agricultural
and Food Chemistry, 45(8), 3238-3243.

15
 ACCEPTED MANUSCRIPT

Parenti, A., Guerrini, L., Masella, P., Spinelli, S., Calamai, L., & Spugnoli, P. (2014). Comparison of
espresso coffee brewing techniques. Journal of Food Engineering, 121, 112-117.

Petracco, M. (2001). Technology IV: beverage preparation: brewing trends for the new millennium.
Coffee: recent developments, 140-164.

Prankerd, R. J. (2007). Critical compilation of pKa values for pharmaceutical substances. Profiles
of drug substances, excipients and related methodology, 33, 1-33.

PT
Romani S., Severini C., Fiore A.G., Pinnavaia G.G. (2004, October). Quality of espresso coffee: A
studyperformed through Italian coffee shops. International Scientific Colloquium onCoffee-

RI
Proceedings (20th ASIC) Bangalore, India.

SC
Sarstedt, M., & Mooi, E. (2014). Cluster analysis. In A concise guide to market research (pp. 273-
324). Springer, Berlin, Heidelberg. NU
SCAA. Specialty Coffee Association of America. (2016). http://www.scaa.org/PDF/resources/best-
practices.
MA

Scholz, B. M., & Maier, H. G. (1990). Isomers of quinic acid and quinide in roasted coffeeIsomere
Chinasäuren und-lactone in Röstkaffee. Zeitschrift für Lebensmittel-Untersuchung und
Forschung, 190(2), 132-134.
ED

Severini, C., Derossi, A., Ricci, I., Fiore, A. G., & Caporizzi, R. (2017). How Much Caffeine in
T

Coffee Cup? Effects of Processing Operations, Extraction Methods and Variables. In The Question
EP

of Caffeine. InTech.

Trugo, L. C., & Macrae, R. (1984). Chlorogenic acid composition of instant coffees. Analyst,
C

109(3), 263-266.
AC

Yeretzian, C., Jordan, A., Badoud, R., & Lindinger, W. (2002). From the green bean to the cup of
coffee: investigating coffee roasting by on-line monitoring of volatiles. European Food Research
and Technology, 214(2), 92-104

16
 ACCEPTED MANUSCRIPT

1
Table 1: Extraction parameters: extraction method , grind, amount of ground coffee in grams, volume of
water per cup or jug in milliliters, temperature in degrees centigrade, pressure in bar, time in seconds, total
amount of beverage in milliliters, and extraction % .

Extraction Grinding Powder Water Temperature Pressure Beverage Extraction

Time
method level (g) (mL) (°C) (bar) (g) %

EC fine 14 - 93 9 27± 1.7(s) 29.6±1.7 22.8±1.3

ECF fine 15 - 92 20 70±10(s) 30±5 13.1±1.6

PT
ECS fine 18 - 93 9 26.50±1.8(s) 17.4±1.6 17.5±0.9

RI
Moka fine 15 150 100 1.5 2.13±0.13 (min) 134±1.8 28.4±1.1

V60 coarse 15 250 93 1 2.3±0.1(min) 206±5 22.1±0.7

SC
Cold Brew coarse 25 250 20 1 4.7±0.1(h) 199±10 23.3±0.9

Aeropress coarse 16.5 250 93 1 1.35±0.08(min) 212±4 20.4±1.2

NU
French Press coarse 15 250 93 1 5(min) 199±4 18.7±1.1
1. EC, espresso coffee; ECS, specialty espresso, ECF, Caffè Firenze;
MA
T ED
C EP
AC

17
 ACCEPTED MANUSCRIPT

1, 2
Table 2: Physical characterization of coffee beverages .

Density Viscosity
pH TDS % Extraction % -2
20°(g/mL) (mN s m )

ECF 5.16 ± 0.10 a 3.32 ± 0.40 a 13.46 ± 1.56 a 1.02 ± 0.03 a 115.15 ± 3.29 a
ECS 5.30 ± 0.25 a 8.44 ± 0.38 b 17.54 ± 0.86 b 1.01 ± 0.01 a 151.59 ± 7.01 b
EC 5.17 ± 0.07 a 5.20 ± 0.35 c 22.59 ± 1.51 c 1.04 ± 0.03 a 123.13 ± 2.70 c

PT
V60 5.15 ± 0.12 a 1.55 ± 0.04 d 22.14 ± 0.65 c 1.07 ± 0.09 a 99.76 ± 3.44 d
Cold Brew 5.12 ± 0.10 a 1.54 ± 0.06 d 20.89 ± 0.82 d 1.05 ± 0.05 a 100.83 ± 2.40 d
Aeropress 5.16 ± 0.11 a 1.52 ± 0.06 d 20.56 ± 0.67 d 1.06 ± 0.05 a 101.74 ± 2.62 d

RI
French Press 5.16 ± 0.13 a 1.35 ± 0.03 d 18.61 ± 1.20 b 1.07 ± 0.07 a 98.25 ± 3.97 d
Moka 5.10 ± 0.24 a 3.40 ± 0.15 a 28.60 ± 1.03 e 1.06 ± 0.02 a 111.61 ± 2.56 a

SC
1.Data are expressed as mean ± standard deviation. Letters (a,b,c,d,e) indicate statistically significant differences betw een extraction
methods.
2.EC, espresso coffee; ECS, specialty espresso, ECF, Caffè Firenze;
NU
MA
T ED
C EP
AC

18
 ACCEPTED MANUSCRIPT
1,2
Table 3: Chemical characterization beverages. Concentrations (mg/mL) of Caffeine, CQAs, CeQAs, 5-FQA, 5-pCoQA, CQLs and diCQAs are reported .

ECF ECS EC V60 COLD BREW AEROPRESS FRENCH PRESS MOKA

1.43 ± 0.07 4.20 ± 0.09 4.10 ± 0.16 0.74 ± 0.09 1.25 ± 0.12 0.78 ± 0.09 0.52 ± 0.06 1.28 ± 0.04
Caffeine
†
b
0.07 ± 0.02
a
0.20 ± 0.02
a
0.18 ± 0.03
c
0.03 ± 0.00
b
0.04 ± 0.00
c
0.02 ± 0.01
c
0.02 ± 0.00
b
0.04 ±

P T 0.01

I
CQA
b a a c c a c c
0.60 ± 0.06 1.86 ± 0.01 1.80 ± 0.30 0.31 ± 0.05 0.50 ± 0.06 0.27 ± 0.04 0.21 ± 0.03 0.45 ± 0.07
3-CQA†

CeQA†
b
0.08 ±
b
0.01
a
0.23 ±
a
0.02
a
0.24 ±
a
0.04
b
0.03 ±
c
0.00
b
0.06 ±
b
0.01
b
0.03 ±
bc
0.01
b
0.02 ±

C
c
0.00
R b
0.05 ±
bc
0.01

CeQA

5-CQA
† 0.08 ±
b
1.56 ±
b
0.02

0.17
0.17 ±
a
4.80 ±
a
0.02

0.30
0.17 ±
a
4.46 ±
a
0.02

0.10
0.03 ±
c
0.80 ±
c
0.00

0.08
0.05 ±
b
1.39 ±
b
0.01

0.15
0.03 ±
c
0.72 ±
c
0.01

U
0.11 S
0.02 ±
c
0.53 ±
c
0.00

0.07
0.04 ±
bc
1.22 ±
b
0.01

0.18

4-CQA

5-pCoQA
0.85 ±
b
0.09 ±
b
0.11

0.02
2.50 ±
a
0.27 ±
a
0.30

0.07
2.59 ±
a
0.23 ±
a
0.14

0.05
0.44 ±
c
0.03 ±
b
0.04

0.00
0.76 ±
b
0.06 ±
b
0.08

0.02
A
0.31 ±
c
0.04 ±
b
N0.16

0.02
0.31 ±
c
0.02 ±
b
0.04

0.00
0.50 ±
bc
0.05 ±
b
0.20

0.01

5-FQA
0.22 ±
b
0.04 0.71 ±
a
0.08 0.50 ±
a
0.20 0.09 ±
cb
0.01 0.18 ±
b
M
0.03 0.09 ±
c
0.01 0.07 ±
c
0.01 0.15 ±
b
0.03

CQL

4-CQL
† 0.04 ±
b
0.11 ±
0.01

0.02
0.12 ±
a
0.31 ±
0.04

0.07
0.17 ±
a
0.31 ±
0.01

0.06
0.01 ±
c
0.04 ±
0.00

0.01
E D
0.02 ±
b
0.07 ±
0.01

0.02
0.02 ±
b
0.05 ±
0.00

0.02
0.01 ±
c
0.03 ±
0.00

0.00
0.01 ±
bc
0.06 ±
0.00

0.02

T
b a a c bc c c bc
† 0.21 ± 0.04 0.61 ± 0.07 0.43 ± 0.19 0.09 ± 0.02 0.16 ± 0.02 0.11 ± 0.02 0.07 ± 0.01 0.16 ± 0.03

P
CQL
b a a bc c bc c b
0.19 ± 0.03 0.52 ± 0.09 0.41 ± 0.09 0.08 ± 0.02 0.12 ± 0.02 0.07 ± 0.02 0.05 ± 0.00 0.13 ± 0.02
CQL†

1,4-diCQA
b
0.10 ±
b
0.03
a
0.28 ±
a
0.08
a
0.33 ±
a
C E
0.09
c
0.03 ±
b
0.00
bc
0.06 ±
b
0.02
c
0.05 ±
b
0.02
c
0.02 ±
b
0.00
bc
0.05 ±
b
0.02

C
0.08 ± 0.02 0.21 ± 0.07 0.26 ± 0.11 0.02 ± 0.00 0.04 ± 0.01 0.03 ± 0.00 0.02 ± 0.00 0.04 ± 0.01
3,5-diCQA
b a a b b b b b
4,5-diCQA

1.
0.15 ±
b
0.03 0.41 ±
a
0.11

A
0.38 ±
a
0.04 0.05 ±
b
0.01 0.09 ±
b
0.02 0.07 ±
b
0.03 0.03 ±
b
0.01 0.09 ±

Data are expressed as mean ± standard deviation. Letters indicate statistically significant differences between extraction methods.
b
0.02

† indicates that the acylation position w as uncertain

2. CGA, chlorogenic acid; 5-CQA, 5-O-caffeoylquinic acid; 3-CQA , isomers 3-O-caffeoylquinic acid; 4-CQA, 4-O-caffeoylquinic acid; CeQA. caffeoyl epi-quinic acid;
p-CoQA, p-Coumaroyl Quinic Acid; FQA, Feruloyl Quinic Acid; CQL, Caffeoyl Quinic Lactone Acid; diCQA:, di-Caffeoyl Quinic Acid. EC, espresso coffee;
ECS, specialty espresso, ECF, Caffè Firenze;

19
 ACCEPTED MANUSCRIPT
1, 2
Table 4: Chemical characterization of beverages. Extraction efficiency (mg/g) of caffeine, CQAs, CeQAs, 5-FQA, 5-pCoQA, CQLs and diCQAs are reported .

ECF ECS EC V60 COLD BREW AEROPRESS FRENCH PRESS MOKA

5.76 ± 0.33 8.50 ± 0.12 17.40 ± 0.62 10.19 ± 0.97 9.67 ± 0.64 10.14 ± 1.21 6.89 ± 1.00 10.17 ± 0.33
Caffeine
d c a b b b c b
CQA†
0.30 ±
b
0.08 0.42 ±
b
0.05 0.77 ±
a
0.12 0.35 ±
b
0.05 0.33 ±
b
0.06 0.29 ±
b
0.08 0.26 ±
b
0.03 0.29 ±
b
0.15

P T
3-CQA†

CeQA †
2.42 ±
b
0.34 ±
b
0.27 3.79 ±
b
0.06 0.48 ±
b
0.21 6.82 ±

0.05 1.00 ±
a

a
0.32 4.29 ±

0.17 0.43 ±
b

b
0.57

0.03
3.90 ±
b
0.50 ±
b
0.63 3.63 ±

0.13 0.42 ±
b

b
0.56

0.06
2.76 ±
b
0.30 ±
b
0.41 3.06 ±

0.03 0.35 ±
b

R
b
I 1.55

0.19

CeQA†

5-CQA
0.31 ±
b
6.32 ±
0.07 0.34 ±
b
0.70 9.75 ±
0.05 0.72 ±
a
0.66 18.91 ±
0.11 0.37 ±
b
0.18 11.02 ±
0.04

0.95
0.39 ±
b
10.39 ±
0.10 0.34 ±

1.73 9.52 ±
b
0.06

1.49
0.23 ±
b
7.06 ±
S C
0.04 0.30 ±

1.10 8.17 ±
b
0.16

4.12

4-CQA
c
3.44 ±
c
0.37 ±
b
0.45 5.20 ±
b
0.10 0.55 ± 0.16 0.98 ±
a
0.53 11.00 ±
a
0.47 6.04 ±

0.21 0.35 ±
b

b
0.47

0.06
b
5.70 ±
b
0.44 ±
0.95 4.16 ±

0.14 0.54 ±
b

b
1.21

0.33
N U c
3.99 ±
c
0.28 ±
0.54 3.22 ±

0.03 0.32 ±
b

bc
2.48

0.18
5-pCoQA

5-FQA

CQL†
b
0.91 ±
b
0.15 ±
b
0.14 1.44 ±
b
0.03 0.24 ±
0.17 2.11 ±

0.08 0.71 ±
a

a
0.93 1.27 ±

0.49 0.09 ±
b

b
0.11

0.01
b
1.38 ±
b
0.14 ±
M A
0.26 1.22 ±

0.07 0.30 ±
b

b
0.19

0.43
b
0.91 ±
b
0.09 ±
0.14 0.99 ±

0.01 0.09 ±
b

b
0.53

0.06

4-CQL
b
0.45 ±
b
b
0.07 0.64 ±
b
0.15 1.33 ±
a

a
0.28 0.51 ±
b

b
0.09

E D b
0.55 ±
b
0.16 0.68 ±
b

b
0.28
b
0.35 ±
b
0.04 0.43 ±
b

b
0.23

T
† 0.84 ± 0.18 1.23 ± 0.15 1.82 ± 0.14 1.31 ± 0.25 1.17 ± 0.22 1.39 ± 0.20 0.87 ± 0.16 1.10 ± 0.56
CQL
b b a b b b b b
CQL†

1,4-diCQA
0.79 ±
b
0.41 ±
0.13 1.06 ±
b
0.10 0.58 ±
0.19 1.73 ±

0.17 1.40 ±
a
0.32 1.04 ±

E
0.41 0.45 ±
b
P 0.25

0.03
0.95 ±
b
0.46 ±
0.19 0.92 ±

0.13 0.69 ±
b
0.32

0.38
0.70 ±
b
0.30 ±
0.08 0.88 ±

0.03 0.36 ±
b
0.45

0.21

C
b b a b b b b b
0.32 ± 0.11 0.45 ± 0.14 1.20 ± 0.48 0.32 ± 0.04 0.28 ± 0.07 0.43 ± 0.03 0.22 ± 0.03 0.26 ± 0.14
3,5-diCQA

C
b b a b b b b b
0.60 ± 0.13 0.84 ± 0.21 1.63 ± 0.18 0.71 ± 0.13 0.59 ± 0.13 1.03 ± 0.44 0.46 ± 0.06 0.62 ± 0.32

A
4,5-diCQA
bc bc a bc bc b c bc

1. Data are expressed as mean ± standard deviation. Letters indicate statistically significant differences between extraction methods.
† indicates that the acylation position w as uncertain
2. CGA, chlorogenic acid; 5-CQA, 5-O-caffeoylquinic acid; 3-CQA , isomers 3-O-caffeoylquinic acid; 4-CQA, 4-O-caffeoylquinic acid; CeQA. caffeoyl epi-quinic acid;
p-CoQA, p-Coumaroyl Quinic Acid; FQA, Feruloyl Quinic Acid; CQL, Caffeoyl Quinic Lactone Acid; diCQA:, di-Caffeoyl Quinic Acid. EC, espresso coffee;
ECS, specialty espresso, ECF, Caffè Firenze;

20
 ACCEPTED MANUSCRIPT
Table 5: Chemical characterization of beverages. Bioactive content per cup (mg/cup) of caffeine, CQAs, CeQAs, 5-FQA, 5-pCoQA, CQLs and diCQAs are
1,2
reported .

ECF ECS EC V60 COLD BREW AEROPRESS FRENCH PRESS MOKA

T
42.78 ± 2.15 75.51 ± 1.54 122.40 ± 4.95 89.04 ± 11.25 149.52 ± 13.80 93.36 ± 10.32 62.16 ± 6.92 51.14 ± 1.43
Caffeine
e d b c a c d e
CQA†
2.12 ± 0.54 3.67 ± 0.37
c b
5.46 ± 0.88 3.05 ±
a b
0.47 5.28 ±
a
0.55 2.64 ±
b
0.76
c
P
2.28 ± 0.17 1.69 ± 0.29

I c

R
17.97 ± 1.89 33.52 ± 1.86 54.02 ± 10.08 37.50 ± 6.37 61.20 ± 6.69 32.42 ± 5.05 24.96 ± 3.17 18.12 ± 2.85
3-CQA†
c b a b a b b c

C
† 2.52 ± 0.42 4.20 ± 0.41 7.08 ± 1.18 3.78 ± 0.39 7.68 ± 1.65 3.80 ± 0.60 2.69 ± 0.34 2.05 ± 0.47
CeQA
c b a b a b c c
CeQA†

5-CQA
2.27 ± 0.49 3.00 ± 0.39
c b
5.05 ± 0.69 3.24 ±
a
46.92 ± 4.91 86.03 ± 5.97 133.86 ± 2.91 96.04 ±
d c b
b

c
0.43 6.17 ±
a

a
1.26 3.05 ±

U S
b

c
0.60 2.08 ± 0.36 1.73 ± 0.48
c
9.21 167.29 ± 13.26 86.08 ± 13.26 63.81 ± 8.72 48.63 ± 7.23
d
c

d
4-CQA

5-pCoQA
25.54 ± 3.22 45.73 ± 4.51 77.76 ± 4.08 52.66 ±
c
2.70 ± 0.75 4.81 ± 1.29
b a
6.98 ± 1.55 3.04 ±
b
5.30 90.96 ±

0.56
A
6.61 ±
a
N
8.58 37.71 ± 19.17 36.78 ± 4.30 19.78 ± 9.75

2.15 4.91 ±
b
2.95
bc c
2.54 ± 0.40 1.80 ± 0.57

M
b a a b a a b b
6.73 ± 1.08 12.73 ± 1.73 15.14 ± 6.81 11.02 ± 0.87 21.77 ± 3.28 10.93 ± 1.73 8.24 ± 3.53 5.85 ± 1.23
5-FQA
c b a b a b bc c
CQL†

4-CQL
1.12 ± 0.22 2.09 ± 0.63
bc
3.41 ± 0.56 5.63 ± 1.33
b
4.99 ± 3.36 0.81 ±
a
9.39 ± 1.83 4.42 ±
c

E D0.09

0.69
2.44 ±

8.48 ±
a
1.25 2.80 ±

2.62 6.21 ±
a
3.92

2.55
0.79 ± 0.12 0.57 ± 0.16
c c
3.18 ± 0.21 2.49 ± 0.59

CQL †
dc

c
bc
6.28 ± 1.33 10.99 ± 1.33 13.03 ± 3.78 11.30 ±
b
a

ab
P T c

b
1.84 18.78 ±
ab

a
2.73 12.59 ±
b

b
1.88
d d
7.83 ± 1.31 6.53 ± 1.10
c c
CQL†

1.4-diCQA
c
3.06 ± 0.79 5.05 ± 1.57
c
b

b
C E
5.73 ± 0.92 9.34 ± 1.74 12.33 ± 2.52 9.01 ±
ab
9.95 ± 2.75 3.92 ±
a
b

cb
1.88 14.91 ±

0.35 7.01 ±
a

a
2.87 8.30 ±

2.32 6.34 ±
b

ab
2.81

3.63
6.35 ± 0.39 5.26 ± 0.90
bc c
2.68 ± 0.14 2.15 ± 0.67
c c
3.5-diCQA

4.5-diCQA
2.41 ± 0.77 3.79 ± 1.24
c b

A C 7.83 ± 3.21 2.77 ±

4.41 ± 1.10 7.45 ± 1.91 10.53 ± 1.32 6.17 ±

c b
a

a
bc

bc
0.35 4.44 ±

0.96 10.22 ±
a

ab
1.04 3.97 ±

2.18 8.88 ±
b

ab
0.32

3.96
1.90 ± 0.23 1.56 ± 0.36
c

c
c
3.92 ± 0.45 3.68 ± 0.70
c
1. Data are expressed as mean ± standard deviation. Letters indicate statistically significant differences between extraction methods.
† indicates that the acylation position w as uncertain
2. CGA, chlorogenic acid; 5-CQA, 5-O-caffeoylquinic acid; 3-CQA , isomers 3-O-caffeoylquinic acid; 4-CQA, 4-O-caffeoylquinic acid; CeQA. caffeoyl epi-quinic acid;
p-CoQA, p-Coumaroyl Quinic Acid; FQA, Feruloyl Quinic Acid; CQL, Caffeoyl Quinic Lactone Acid; diCQA:, di-Caffeoyl Quinic Acid. EC, espresso coffee;
ECS, specialty espresso, ECF, Caffè Firenze;

21
 ACCEPTED MANUSCRIPT

Figure captions

Figure 1: Cluster analysis of extraction methods. List of acronyms: EC, espresso coffee; ECS, specialty espresso, ECF, Caffè Firenze;

sample.

P T
Figure 2: Overlapping of HPLC/DAD chromatograms at 278 nm (whole line) and 330 nm (dotted line) for CGAs and caffeine monitoring of a representative coffee

R I
1: CQA*; 2: 3-CQA; 3: CeQA*; 4: CeQA*; 5: 5-CQA (chlorogenic acid); 6: 4-CQA ; 7: 5-p-CoQA; 8: 5-FQA; 9:CQL*; 10:4-CQL*; 11: CQL*; 12:CQL*; 13:1,4-diCQA; 14: 3,5-diCQA
;15: 4,5-diCQA. *acylation position in uncertain . List of acronyms : CQA: Caffeoyl Quinic Acid; CeQA: caffeoyl epi-quinic acid; p-CoQA:p-Coumaroyl Quinic Acid; FQA: Feruloyl
Quinic Acid; CQL: Caffeoyl Quinic Lactone Acid; diCQA: di-Caffeoyl Quinic Acid.

S C
N U
Figure 3: Content per mL of extract (a), per gram of coffee powder (b), and per cup of coffee brew (c) of caffeine and of sum of CGAs. Letters indicate statistically

correspond to the standard deviation (95 %).

M A
significant differences between extraction methods. Capital letters indicate difference in CGAs while lowercase letters indicate differences in caffeine. Error bars

E D
P T
C E
A C

22
 ACCEPTED MANUSCRIPT
Highlights

 Numerous coffee beverages obtained using different extraction have entered the market

 This study characterizes and compares eight brewing methods

P T
 The qualitative profile of bioactive substances was detected by HPLC-DAD
R I

C
The study informs on concentrations and on per-cup caffeine and chlorogenic acids content

S
N U
M A
E D
P T
C E
A C

23
Figure 1
Figure 2
Figure 3