The effect of coffee beans roasting on its chemical composition

Abstract

Drinking coffee has become part of our everyday culture. Coffee cultivation is devoted to over 50 countries in the world, located between latitudes 25 degrees North and 30 degrees South. Almost all of the world's coffee production is provided by two varieties, called 'Arabica' and 'Robusta' whereas the share of Arabica is 70% of the world's coffee harvest. Green (raw) coffee can not be used to prepare coffee beverages, coffee beans must first be roasted. Roasting coffee and reaching a certain degree of coffee roasting determine its flavor and aroma characteristics. In the present study the fate of sucrose, chlorogenic acid, acetic acid, formic acid, lactic acid, caffeic acid, total phenolic compounds and 5-hydroxymethylfurfural was studied in coffee (Brazil Cerrado Dulce, 100% Arabica) roasted in two ways (Medium roast and Full city roast). It has been found that almost all sucrose has been degraded (96-98%) in both roasting ways. During Medium roast 65% of chlorogenic acid contained in green coffee was degraded while during Full city roast it was 85%. During both Medium and Full city roasting, the formation of acetic acid but especially formic and lactic acid was recorded. The highest concentration of organic acids was recorded at Full City roasting at medium roasting times (3.3 mg.g-1 d.w. acetic acid, 1.79 mg.g-1 d.w. formic acid, 0.65 mg.g-1 d.w. lactic acid). The amount of phenolic substances also increased during roasting up to 16.7 mg.g-1 d.w. of gallic acid equivalent. Highest concentrations of 5-hydroxymethylfurfural were measured at medium roasting times at both Medium (0.357 mg.g-1 d.w.) and French city (0.597 mg.g-1 d.w.) roasting temperatures. At the end of roasting, the 5-hydroxymethylfurfural concentration in coffee were 0.237 mg.g-1 d.w. (Medium roast) and 0.095 mg.g-1 d.w. (Full city roast).

Full text

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 344 No. 1/2019
Potravinarstvo Slovak Journal of Food Sciences
vol. 13, 2019, no. 1, p. 344-350
https://doi.org/10.5219/1062
Received: 10 February 2019. Accepted: 11 March 2019.
Available online: 28 May 2019 at www.potravinarstvo.com
© 2019 Potravinarstvo Slovak Journal of Food Sciences, License: CC BY 3.0
ISSN 1337-0960 (online)
THE EFFECT OF COFFEE BEANS ROASTING ON ITS CHEMICAL
COMPOSITION
Pavel Diviš, Jaromír Pořízka, Jakub Kříkala
ABSTRACT
Drinking coffee has become part of our everyday culture. Coffee cultivation is devoted to over 50 countries in the world,
located between latitudes 25 degrees North and 30 degrees South. Almost all of the world's coffee production is provided
by two varieties, called ‘Arabica’ and ‘Robusta’ whereas the share of Arabica is 70% of the world's coffee harvest. Green
(raw) coffee can not be used to prepare coffee beverages, coffee beans must first be roasted. Roasting coffee and reaching a
certain degree of coffee roasting determine its flavor and aroma characteristics. In the present study the fate of sucrose,
chlorogenic acid, acetic acid, formic acid, lactic acid, caffeic acid, total phenolic compounds and 5-hydroxymethylfurfural
was studied in coffee (Brazil Cerrado Dulce, 100% Arabica) roasted in two ways (Medium roast and Full city roast). It has
been found that almost all sucrose has been degraded (96 – 98%) in both roasting ways. During Medium roast 65% of
chlorogenic acid contained in green coffee was degraded while during Full city roast it was 85%. During both Medium and
Full city roasting, the formation of acetic acid but especially formic and lactic acid was recorded. The highest concentration
of organic acids was recorded at Full City roasting at medium roasting times (3.3 mg.g-1 d.w. acetic acid, 1.79 mg.g-1 d.w.
formic acid, 0.65 mg.g-1d.w. lactic acid). The amount of phenolic substances also increased during roasting up to
16.7 mg.g-1 d.w. of gallic acid equivalent. Highest concentrations of 5-hydroxymethylfurfural were measured at medium
roasting times at both Medium (0.357 mg.g-1 d.w.) and French city (0.597 mg.g-1 d.w.) roasting temperatures. At the end of
roasting, the 5-hydroxymethylfurfural concentration in coffee were 0.237 mg.g-1 d.w. (Medium roast) and 0.095 mg.g-1
d.w. (Full city roast).
Keywords: coffee; roasting; hydroxymethylfurfural; sucrose; organic acids
INTRODUCTION
Coffee is made up of modified seed of the fruit of various
tropical to subtropical trees or coffee shrubs. Coffee has
a large number of varieties, only a few of them have
economic significance. Almost all of the world's coffee
production is provided by two varieties, called ‘Arabica’
and ‘Robusta’ (Butt and Tauseef Sultan, 2011). Arabica
(Coffea arabica), is the most important botanical species,
especially for the high quality of its fruits. It comes from
about 70% of the world's green coffee production. Robusta
(Coffea canephora), is the second most important variety
of coffee, and its share of world production is steadily
growing mainly due to its greater adaptability to habitats
and disease resistance. Other reasons to increase demand
of Robusta coffee in the market are the growing demand
for instant coffee, which is preferentially made from
Robusta coffee and last but not least, lower price of
Robusta coffee compared to Arabica coffee (Kemsley et
al., 1995). World coffee production in 2017 was about
9.5 million tonnes, which makes coffee the second most
important commodity of world trade (FAOSTAT, 2018).
After processing the coffee beans by wet or dry drying
technologies (Arya and Jagan Mohan Rao, 2007;
Guimar, Berbert and Silva, 1998), coffee beans are
roasted. During coffee roasting, coffee beans get brown
colour and their characteristic flavour and aroma
(Yeretzian et al., 2002). In different countries, different
roasting styles have been created according to population
preferences. These roasting styles differ from each other at
the roasting temperature used and the total coffee roasting
time (Moon, Yoo and Shibamoto, 2009; Dórea and
DaCosta, 2005).
Due to the popularity of coffee in the world many
researchers were involved in coffee research in the last
quarter of a century. Most studies on coffee are health-
related studies (Ciaramelli, Palmioli and Airoldi, 2019;
Poole et al., 2017; Ludwig et al., 2014; Butt and Tauseef
Sultan, 2011; Dórea et al., 2005). Other studies are
focused on the role of roasting conditions of the coffee in
the level of selected compounds. Information on what is
happening in roasted coffee beans is quite sufficient in the
available literature, but there are only few studies that deal
with the complex monitoring of changes in the chemical
composition of coffee beans during roasting (Wei et al.,
2012).

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 345 No. 1/2019
In the present study the fate of sucrose, chlorogenic acid,
acetic acid, formic acid, lactic acid, caffeic acid, total
phenolic compounds and 5-hydroxymethylfurfural was
studied in coffee roasted in two ways (Medium roast and
Full city roast).
Scientific hypothesis
Higher temperature and higher time of coffee beans
roasting cause higher amounts of 5-hydroxymethylfurfural,
organic acids and phenolic compounds while reducing the
carbohydrate content in the coffee beans.
MATERIAL AND METHODOLOGY
Chemicals and reagents
All water used in this study was ultrapure water (Elga
pure lab classic, Veolia water systems, UK). All chemicals
used in this study were analytical grade chemicals
purchased from Sigma-Aldrich (Germany) company
except of Karl-Fisher titration reagent and water standard
which have been purchased from Labicom (Czech
Republic).
Sample preparation
The 100% Arabica coffee (Brazil Cerrado Dulce,
4coffee, Czech Republic) was used in this study. Total
amount of 25g of green coffee beans was roasted in a
home coffee roaster (Gene café CBR101, 4coffee, Czech
Republic). Coffee was roasted with two roasting degrees
as Medium roast and Full city roast. Roasting on Medium
roast degree was done at 210 °C and the total roasting time
was 14 min. Roasting on Full city roast degree was done at
225 °C and the total roasting time was 19 min. Extraction
of organic acids, sucrose, 5-hydroxymethylfurfural and
total phenolic compounds was performed in 25 mL
Erlenmeyer flasks. One gram of sample weighted on
analytical balancer and 10 mL of solvent (80 °C water
mixed with ethanol in 60:40 volume ratio) were used for
extraction. Extraction was carried out on a magnetic stirrer
for 30 minutes. After the extraction, the samples were
centrifuged at 5000 rpm in centrifuge and the supernatant
was filtered using nylon syringe filters (0.45 m, Labicom,
Czech Republic) and used for analysis. All samples were
prepared in two replicates.
Chemical analysis
Acetic, formic and lactic acid were determined using ion
chromatography (Metroohm 850 professional IC,
Metroohm, Switzerland) with conductivity detector. An
Agilent Infinity 1260 liquid chromatograph (Agilent
Technologies, USA) equipped with ELSD detector was
used for determination of sucrose. Both methods are
described in detail at work published by Diviš et al.
(2018). Total phenolic compounds were determined using
Helios gamma spectrophotometer (Spectronic Unicam,
Great Britain) through the Folin-Ciocalteus method
(Singleton et al. 1999) and expressed as gallic acid
equivalent. Concentration of 5-hydroxymethylfurfural was
determined on Agilent Infinity 1260 liquid chromatograph
with DAD detector using Kinetex EVO-C18 column and
acetonitrile mixed with water in 15:85 volume ratio.
Chlorogenic and caffeic acid were determined on Agilent
Infinity 1260 liquid chromatograph with DAD detector
using Kinetex EVO-C18 column and mixture of 2.5%
formic acid and acetonitrile in 90:10 volume ratio as
mobile phase. Water content in all samples was
determined by Karl-Fisher titration (Verhoef and
Barendrecht, 1977) using KF Titrino 701 titrator
(Metroohm, Switzerland). The pH value was measured
using pH meter with combined electrodes (WTW,
Germany). All parameters for a single sample were
measured in three replicates. All measured concentrations
were recalculated to dry weight of coffee.
Statistic analysis
All experimental data were statistically processed using
software XLstat (Addinsoft, USA). Obtained data were
pre-treated by using Analysis of Variance (ANOVA) to
find statistical significant differences between groups.
Tukey’s comparative test on the significance level 0.05 has
been performed for individual parameters observed during
coffee roasting. The pre-treated data were used as input
parameters in Principal Component Analysis (PCA) to find
correlation between the chemical composition changes
during the roasting process.
RESULTS AND DISCUSSION
Coffee beans contain, in addition to water and minerals,
a large number of organic substances. The main
component of coffee beans are carbohydrates. The coffee
beans contain various hemicelluloses, starch,
oligosaccharides and mainly sucrose. The amount of
monosaccharides is relatively small. Other substances
contained in coffee are proteins, non-protein nitrogenous
substances, phenolic substances, non-volatile organic
acids, volatile substances and oils (Arya and Jagan
Mohan Rao, 2007; Farah and Marino Donagelo, 2006;
Redgwell and Fisher, 2006).
Concentration of sucrose in green coffee beans and in
roasted coffee beans is presented in Table 1 and Table 2.
The results show that sucrose degradation occurs during
the roasting process. Sucrose degradation is explained by
sucrose hydrolysis to glucose and fructose, which may be
further fragmented to form aliphatic acids, or which may
participate in Maillard reactions with proteins or amino
acids (Ginz et al. 2000). The sucrose concentration
decreased in the middle of the roasting process by 47% in
the case of Medium roast and by 59% in the case of Full
city roast. At the end of the roasting process, almost all
sucrose has already been degraded (96 – 98%).
Another substance that has been observed to reduce the
concentration during the roasting process was chlorogenic
acid. Concentration of chlorogenic acid in green coffee
beans and in roasted coffee beans is presented in Table 3
and Table 4. Chlorogenic acid concentration decreased in
the middle of the roasting process by 45% in the case of
Medium roast and by 42% in the case of Full city roast. At
the end of the roasting process, 67% of chlorogenic acid
contained in green coffee was degraded during Medium
roast and 85% during Full city roast. Chlorogenic acid is
involved in colour, flavour and aroma formation of coffee
Farah and Marino Donagelo, 2006; Farah et al., 2005).
Major degradation products of chlorogenic acid are
melanoids and low molecular weight compounds.

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 346 No. 1/2019
Strong significant correlation was found between sucrose
concentration and concentration of organic acid during
coffee roasting (r >0.9).
Concentration of organic acid in coffee during the roasting
process is shown in Table 3 and Table 4. While the sucrose
concentration in coffee decreases during roasting, the
concentration of organic acids significantly increases. The
most significant change was found in the lactic and formic
acid content. The content of these acids in coffee beans
rose almost 100 times after coffee roasting. Trend of
changes in the concentration of organic acids was similar
for Medium roasting and French roasting, however in the
case of French roasting decrease in organic acid content
was observed in the later stage of roasting. Ginz et al.
(2000) lists the content of organic acids in Robusta coffee
after roasting to be approximately 2 mg.g-1 in the case of
formic acid and acetic acid and 0.2 mg.g-1 in the case of
lactic acid. Formation of organic acids in coffee is
described by Lobry-deBruyn-vanEckenstein
rearrangement reaction in which fructose or glucose
produced by sucrose hydrolysis is involved, and by
formation of 1,2-endiole or 2,3-endiole as acid precursors
(Ginz et al., 2000). Formation of organic acids in coffee
during roasting process did not significantly affect the pH
of the coffee (Table 1 and Table 2.). This finding can be
caused due to highly complex buffering effects and the
wide distributions of salts and acids present in coffee.
Jeszka- Jeszka-Skowron et al. (2016) measured pH value
Table 1 Content of sucrose, 5-hydroxymethylfurfural, total phenolic compounds and pH value of coffee roasted to
Medium roast degree.
Time
pH
(mg.g-1 ±SD)
sucrose
(mg.g-1 ±SD)
HMF
(mg.g-1 ±SD)
TPC
(mg.g-1 ±SD)
0
6.09 ±0.05a
70.1 ±4.9a
<0.010
8.5 ±0.8e
4
5.93 ±0.05ab
58.5 ±2.1b
<0.010
8.9 ±0.8e
5
5.89 ±0.05ab
57.0 ±6.4bc
<0.010
9.6 ±0.9de
6
5.78 ±0.05ab
50.1 ±3.5c
0.013 ±0.004d
10.8 ±0.6cd
7
5.89 ±0.05ab
38.8 ±3.8d
0.044±0.016d
12.8 ±0.6ab
8
5.68 ±0.05ab
23.8 ±2.5e
0.136 ±0.013c
12.0 ±0.7bc
9
5.73 ±0.05ab
14.3 ±2.5f
0.281 ±0.031ab
12.7 ±0.3ab
10
5.68 ±0.05ab
12.2 ±1.9f
0.232±0.014b
13.0 ±0.2ab
11
5.72 ±0.05ab
6.3 ±1.8gh
0.139 ±0.021c
13.7 ±0.3a
12
5.65 ±0.05ab
4.4 ±0.5gh
0.264 ±0.046b
14.4 ±0.4a
13
5.68 ±0.05ab
4.1 ±0.7gh
0.346 ±0.016a
14.0 ±0.5a
14
5.65 ±0.05ab
3.0 ±0.4h
0.224 ±0.018b
13.4 ±0.4ab
Note: Values in the same column with different letters are significantly different at p <0.05.
Table 2 Content of sucrose, 5-hydroxymethylfurfural, total phenolic compounds and pH value of coffee roasted to Full
city roast degree.
Time
pH
(mg.g-1 ±SD)
sucrose
(mg.g-1 ±SD)
HMF
(mg.g-1 ±SD)
TPC
(mg.g-1 ±SD)
0
6.09 ±0.05a
70.1 ±4.9a
<0.010
8.5 ±0.8e
4
5.95 ±0.05a
62.8 ±3.9ab
<0.010
8.9 ±0.3e
5
5.95 ±0.05a
59.3 ±3.4b
<0.010
11.5 ±0.7d
6
5.76 ±0.05ab
59.5 ±5.1ab
0.017 ±0.003h
13.7 ±0.9c
7
5.73 ±0.05ab
41.6 ±4.2c
0.091 ±0.006gh
13.9 ±0.5c
8
5.69 ±0.05ab
30.6 ±2.8d
0.326 ±0.018bcd
13.1 ±0.7c
9
5.78 ±0.05ab
16.3 ±2.5e
0.341 ±0.025bc
13.2 ±0.9c
10
5.77 ±0.05ab
12.1 ±2.2f
0.259 ±0.021de
13.8 ±0.8c
11
5.75 ±0.05ab
5.7 ±1.3g
0.207 ±0.013ef
14.9 ±0.9bc
12
5.69 ±0.05ab
3.4 ±0.4gh
0.549 ±0.029a
15.8 ±1.2ab
13
5.62 ±0.05ab
2.5 ±0.3h
0.510 ±0.017a
14.0 ±0.3bc
14
5.59 ±0.05ab
2.3 ±0.2h
0.408 ±0.022b
14.1 ±0.5bc
15
5.55 ±0.05b
2.0 ±0.2hi
0.406 ±0.019b
14.8 ±0.3bc
16
5.62 ±0.05ab
1.9 ±0.3hi
0.303 ±0.023cd
14.6 ±0.7bc
17
5.72 ±0.05ab
1.7 ±0.3hi
0.236 ±0.015de
16.7 ±0.8a
18
5.66 ±0.05ab
1.5 ±0.2i
0.121 ±0.014fg
15.6 ±0.6ab
19
5.70 ±0.05ab
<1.0
0.108 ±0.017g
14.4±0.9bc
Note: Values in the same column with different letters are significantly different at p <0.05.

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 347 No. 1/2019
of water treated Brazil Arabica green coffee to be 4.92.
Ginz et al. 2000 monitored pH changes in Robusta coffee
during the roasting process and recorded pH change from
6.1 to 5.7 similar to this study (Table 1 and Table 2).
Another strong correlation was found between
chlorogenic acid concentration and caffeic acid
concentration in the case of Medium roast (r = 0.8025).
However, in the case of Full city roast, this correlation was
not significant and week (r = 0.2403). Chlorogenic acid is
an ester of quinic acid and phenolic acid, mostly caffeic,
ferulic or 3-hydroxycinnamic acid (Farah and Marino
Donagelo, 2006). During shorter roasting at lower
temperatures chlorogenic acid can be hydrolysed and
concentration of caffeic acid in coffee beans may
temporarily increase. On the other side, with longer
roasting times and higher temperatures, caffeic acid
released from chlorogenic acid can be further degraded.
From the results summarized in Table 4 it can be seen a
significant increase of caffeic acid concentration in roasted
coffee beans and subsequent reduction of caffeic acid
concentration over longer periods of roasting. The
formation of phenolic substances during roasting of coffee
is also evident from the total concentration of phenolic
compounds presented in Table 1 and Table 2. Measured
concentrations of chlorogenic acid, caffeic acid or total
phenolic compounds in this study are comparable with
data published in literature. Chlorogenic acid
concentration in green coffee beans is reported within the
Table 3 Content of organic acids in coffee roasted to Medium roast degree.
Time
Organic acids
Acetic
(mg.g-1 ±SD)
Formic
(mg.g-1 ±SD)
Lactic
(mg.g-1 ±SD)
Chlorogenic
(mg.g-1 ±SD)
Caffeic
(mg.g-1 ±SD)
0
0.345 ±0.036gh
<0.005
<0.005
22.7 ±1.8a
0.741 ±0.082c
4
0.262 ±0.055h
0.032 ±0.005f
0.019 ±0.009e
17.8 ±0.5b
0.873 ±0.033c
5
0.345 ±0.013gh
0.055 ±0.016f
0.037 ±0.006e
15.7 ±0.9bc
0.971 ±0.021bc
6
0.378 ±0.033g
0.069 ±0.004f
0.036 ±0.004e
14.5 ±0.6cd
0.932 ±0.029bc
7
0.726 ±0.067f
0.133 ±0.027ef
0.039 ±0.007e
13.6 ±0.4de
1.03 ±0.08abc
8
1.07 ±0.08e
0.291 ±0.028de
0.114 ±0.011d
13.7 ±0.3de
1.05 ±0.11abc
9
1.16 ±0.06e
0.383 ±0.014d
0.099 ±0.021d
12.2 ±0.2def
1.17 ±0.12ab
10
1.21 ±0.03de
0.432 ±0.029cd
0.144 ±0.013d
11.7 ±0.3ef
1.03 ±0.07abc
11
1.39 ±0.04cd
0.571 ±0.025bc
0.198 ±0.009c
10.4 ±1.1fg
1.02 ±0.14abc
12
1.58 ±0.07c
0.633 ±0.063b
0.239 ±0.017c
8.9 ±0.9gh
1.18 ±0.05ab
13
1.81 ±0.04b
0.717 ±0.113ab
0.317 ±0.018b
7.9 ±0.8gh
1.17 ±0.08a
14
2.17 ±0.07a
0.875 ±0.057a
0.392 ±0.022a
7.2 ±0.4h
1.05 ±0.13abc
Note: Values in the same column with different letters are significantly different at p <0.05.
Table 4 Content of organic acids in coffee roasted to Full city roast degree.
Time
Organic acids
Acetic
(mg.g-1±SD)
Formic
(mg.g-1±SD)
Lactic
(mg.g-1±SD)
Chlorogenic
(mg.g-1±SD)
Caffeic
(mg.g-1±SD)
0
0.345 ±0.036j
<0.005
<0.005
22.7 ±1.8a
0.741 ±0.082abc
4
0.214 ±0.015k
0.011 ±0.005i
0.013 ±0.004i
17.1 ±0.3b
0.852 ±0.130 abc
5
0.484 ±0.016i
0.061 ±0.008h
0.046 ±0.006h
16.2 ±0.7bc
0.97 ±0.112 abc
6
0.492 ±0.012i
0.104 ±0.011gh
0.053 ±0.006h
15.3 ±0.6bcd
0.932 ±0.091 abc
7
0.856 ±0.025h
0.293 ±0.041fg
0.051 ±0.004h
14.1 ±0.5cde
0.974 ±0.133 abc
8
1.26 ±0.08g
0.421 ±0.013ef
0.180 ±0.025g
12.9 ±0.8def
1.25 ±0.18ab
9
1.46 ±0.09fg
0.519 ±0.038e
0.208 ±0.028g
12.5 ±0.9def
1.42 ±0.19a
10
1.41 ±0.03fg
0.559 ±0.029de
0.217 ±0.025fg
12.2 ±0.6efg
1.03 ±0.09 abc
11
1.72 ±0.08ef
0.732 ±0.047cd
0.315 ±0.023ef
11.1 ±0.5fgh
0.95 ±0.07 abc
12
1.85 ±0.09de
0.786 ±0.040bc
0.343 ±0.013de
9.9 ±0.6ghi
1.45 ±0.15a
13
2.17 ±0.17cd
0.906 ±0.045abc
0.394 ±0.029cde
9.5 ±0.4ghi
1.22 ±0.13ab
14
3.22 ±0.11a
1.13 ±0.21a
0.628 ±0.043a
9.0 ±0.7hij
1.05 ±0.09 abc
15
2.61±0.12b
1.07 ±0.09a
0.442 ±0.041bcd
8.6 ±0.5ij
0.951 ±0.141 abc
16
2.51 ±0.14b
0.917 ±0.048abc
0.497 ±0.021bc
7.5 ±0.4jk
0.873 ±0.062bc
17
2.43 ±0.05bc
0.921 ±0.050abc
0.410 ±0.016bcde
6.8 ±0.8jk
0.852 ±0.015bc
18
2.61 ±0.06b
0.984 ±0.033ab
0.454 ±0.061bc
5.8 ±0.5k
0.755 ±0.073bc
19
2.51 ±0.05b
1.02 ±0.05a
0.513 ±0.032b
3.2 ±0.9l
0.613 ±0.082c
Note: Values in the same column with different letters are significantly different at p <0.05.

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 348 No. 1/2019
range of 34 – 57 mg.g-1 while in roasted coffee beans in
the range of 2 – 19 mg.g-1 (Ludwig et al., 2014; Narita
and Inouye, 2015; Farah and Marino Donagelo, 2006;
Moon, Yoo and Shibamoto, 2009).
Total phenolic compounds content in coffee is reported to
be 14 – 30 mg.g-1 (gallic acid equivalent) while caffeic
acid content in coffee is reported to be 1.4 – 3 mg.g-1
(Bauer et al., 2018; Hall, Yuen and Grant, 2018).
Almost all foods that are heat-treated are monitored for
content of 5-hydroxymethylfurfural. This compound is
generated in food by the Maillard reaction (Antal et al.,
1990). Increased interest in 5-hydroxymethylfurfural stems
from a partially verified suspicion that this compound is
a health hazard compound that may be mutagenic,
carcinogenic and cytotoxic (Abraham et al., 2011). The
content of 5-hydroxymethylfurfural in roasted coffee is
reported to be 0.3 – 1.9 mg.g-1 (Murkovic and Pichler,
2006). In this study, maximum concentration of
5-hydroxymethylfurfural was measured to be 0.549 mg.g-1
in coffee during Full city roast. During Medium roast
Figure 1 PCA score of the monitored analytes in coffee beans roasted on Medium roast degree. Note: S = short time of
roasting (0 – 6 min), M = medium time of roasting (7 – 10 min), L=long time of roasting (8 – 14min). CGA =
chlorogenic acid, SUC = sucrose, LAC = lactic acid, FOR = formic acid, AAC = acetic acid, HMF =
hydroxymethylfurfural,
TPC = total phenolic compounds, CFA = caffeic acid.
Figure 2 PCA score of the monitored analytes in coffee beans roasted on Full city roast degree. Note: S = short time of
roasting (0 – 7 min), M = medium time of roasting (8 – 14 min), L = long time of roasting (15 – 19 min).
CGA = chlorogenic acid, SUC = sucrose, LAC = lactic acid, FOR = formic acid, AAC = acetic acid,
HMF = hydroxymethylfurfural, TPC = total phenolic compounds, CFA = caffeic acid.
S
S
S
S S
S
S
S
M
M
M M
M M
M M
L L
L L
L L
L
L
AAC
LAC
FOR
pH
HMF
CGA
CFA
SUC
TPC
-8
-6
-4
-2
0
2
4
6
8
-8 -6 -4 -2 02468
F2 (5.93 %)
F1 (85.56 %)
S
S
S
S S
S
S
S
S
S
M
M M
M
M
M
M
M
M
M M
M M
M L
L
L
L
L
L L
L
L
L
AAC
LAC
FOR
pH
HMF
CGA
CFA
SUC
TPC
-6
-4
-2
0
2
4
6
8
10
-8 -6 -4 -2 0 2 4 6 8 10
F2 (15.99 %)
F1 (68.47 %)

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 349 No. 1/2019
maximum content of 5-hydroxymethylfurfural was found
to be 0.346 mg.g-1. Relatively interesting is the course of
5-hydroxymethylfurfural concentration during roasting.
During both Medium and Full city roast two sharp maxima
in 5-hydroxymethylfurfural concentration were recorded,
which could correspond to the first and second crack in
coffee beans. After reaching the second maximum,
concentration of 5-hydroxymethylfurfural in coffee
decreases, because of its degradation to organic acids
(Murkovic and Bornik, 2007).
To investigate the overall composition changes during
the roasting process, PCA was performed on the data for
all coffee bean extracts at different time of roast. The
roasting process was divided into three categories
according to the total roasting time (short, medium and
long time). The results are shown in Figure 1 and Figure 2.
The PCA plots confirmed that sucrose and chlorogenic
acid degraded during the roasting process and also that
with longer roasting times the content of organic acids in
coffee increases. Both Figures 1 and Figures 2 also show
that content of 5-hydroxymethylfurfural is the highest in
medium roasting times.
CONCLUSION
This study proved that coffee roasting is a complex
chemical process. The basic processes detectable during
coffee roasting are the decomposition of sucrose and
chlorogenic acid. Almost all sucrose is degraded during
roasting independently of the roasting method.
Degradation of chlorogenic acid is higher with longer
roasting at higher temperatures. During the Full city roast
(225 °C, 19 min) up to 85% of chlorogenic acid was
degraded. Degradation products of sucrose and
chlorogenic acid are low molecular organic acids and
phenolic acids. Formic or lactic acid concentrations in
coffee beans increased up to 100-fold during roasting. The
increase in the concentration of phenolic compounds was
not so steep, but it was observable. From the measured
results, it cannot be clearly stated that with higher
temperature and with higher roasting time concentration of
5-hydroxymethylfurfural is increasing. Highest
concentrations of 5-hydroxymethylfurfural were measured
at medium roasting times at both Medium and French city
roasting temperatures. Conversely, at shorter roasting time
and lower temperature higher concentrations of
5-hydroxymethylfurfural were found at the end of roasting.
REFERENCES
Abraham, K., Gürtler, R., Berg, K., Heinemeyer, G.,
Lampen, A., Appel, K. E. 2011. Toxicology and risk
assessment of 5-Hydroxymethylfurfural in food. Mol. Nutr.
Food Res., vol. 55, no. 5, p. 667-678.
https://doi.org/10.1002/mnfr.201000564
Antal, M. J., Mok, W. S. L., Richards, G. N. 1990.
Mechanism of formation of 5-(hydroxymethyl)-2-furaldehyde
from d-fructose and sucrose, Carbohydr. Res., vol. 199, no. 1,
p.91-109. https://doi.org/10.1016/0008-6215(90)84096-D
Arya, M., Jagan Mohan Rao, L. 2007. An impression on
coffee carbohydrates. Crit. Rev. Food Sci. Nutr., vol. 47, no.
1, p. 51-67. https://doi.org/10.1080/10408390600550315
Bauer, D., Abreu, J., Jordao, N., Santos daRosa, J., Freitas-
Silva, O., Teodoro, A. 2018. Effect of roasting levels and
drying process of Coffea canephora on the quality of
bioactive compounds and cytotoxicity. Int. J. Mol. Sci., vol.
19, no. 11, p. 3407. https://doi.org/10.3390/ijms19113407
Butt, M. S., Tauseef Sultan, M. 2011. Coffee and its
Consumption: Benefits and Risks, Crit. Rev. Food Sci. Nutr.,
vol. 51, no. 4, p. 363-373.
https://doi.org/10.1080/10408390903586412
Ciaramelli, C., Palmioli, A., Airoldi, C. 2019. Coffee
variety, origin and extraction procedure: Implications for
coffee beneficial effects on human health. Food Chem., vol.
278, p. 47-55.
https://doi.org/10.1016/j.foodchem.2018.11.063
Diviš, P., Smilek, J., Pořízka, J., Štursa, V. 2018. The
quality of ketchups from the Czech Republic market in terms
of their physico-chemical properties. Potravinarstvo Slovak
Journal of Food Sciences, vol. 12, no. 1, p. 233-240.
https://doi.org/10.5219/898
Dórea, J. G., daCosta, T. H. M. 2005. Is coffee a functional
food? The British Journal of Nutrition, vol. 93, no. 6, p. 773-
82. https://doi.org/10.1079/BJN20051370
FAOSTAT. 2018. Trade and markets. Available at:
http://www.fao.org/statistics/en/
Farah, A., DePaulis, T., Trugo, L. C., Martin, P. R. 2005.
Effect of roasting on the formation of chlorogenic acid
lactones in coffee. J. Agric. Food Chem., vol. 53, no. 5, p.
1505-1513. https://doi.org/10.1021/jf048701t
Farah, A., Marino Donangelo, C. 2006. Phenolic
compounds in coffee. Braz. J. Plant Physiol., vol. 18, no. 1, p.
23-36. https://doi.org/10.1590/S1677-04202006000100003
Ginz, M., Balzer, H. H., Bradbury, A. G. W., Maier, H. G.
2000. Formation of aliphatic acids by carbohydrate
degradation during roasting of coffee. Eur. Food. Res.
Technol., vol. 211, no. 6, p. 404-410.
https://doi.org/10.1007/s002170000215
Guimarã, A. C., Berbert, P. A., Silva J. S. 1998. Ambient-
Air Drying of Pre-Treated Coffee (Coffea Arabica L.). J.
Agric. Eng. Res., vol. 69, no. 1, p. 53-62.
https://doi.org/10.1006/jaer.1997.0222
Hall, S., Yuen, J. W., Grant G. D. 2018. Bioactive
constituents in caffeinated and decaffeinated coffee and their
effect on the risk of depression-A comparative constituent
analysis. Beverages, vol. 4, no. 4, p. 79.
https://doi.org/10.3390/beverages4040079
Jeszka-Skowron, M., Sentkowska, A., Pyrzynska, K., Paz
dePena, M. 2016. Chlorogenic acids, caffeine content and
antioxidant properties of green coffee extracts: influence of
green coffee bean preparation. Eur. Food Res. Technol., vol.
242, no. 8, p. 1403-1409. https://doi.org/10.1007/s00217-016-
2641-y
Kemsley, E. K., Ruault, S., Wilson, R. H. 1995.
Discrimination between Coffea arabica and Coffea canephora
variant robusta beans using infrared spectroscopy. Food
Chem., vol. 54, no. 3, p. 321-326.
https://doi.org/10.1016/0308-8146(95)00030-M
Ludwig, I. A., Clifford, M. N., Lean, M. E. J., Ashihara, H.,
Crozier, A. 2014. Coffee: Biochemistry and potential impact
on health. Food Funct. vol. 5, no. 8, p. 1695-1717.
https://doi.org/10.1039/c4fo00042k
Ludwig, I. A., Mena, P., Calani, L., Cid, C., DelRio, D.,
Lean, M. E. J., Crozier, A. 2014. Variations in caffeine and
chlorogenic acid contents: what are we drinking? Food
Funct., vol. 5, no. 8, p. 1718-1726.
https://doi.org/10.1039/c4fo00290c
Moon, J. K., Yoo, H. S., Shibamoto, T. 2009. Role of
roasting conditions in the level of chlorogenic acid content in
coffee beans: correlation with coffee acidity. J. Agric. Food

Potravinarstvo Slovak Journal of Food Sciences
Volume 13 350 No. 1/2019
Chem., vol. 57, no. 12, p. 5365-5369.
https://doi.org/10.1021/jf900012b
Murkovic, M., Bornik, M. A. 2007. Formation of 5-
hydroxymethyl-2-furfural (HMF) and 5-hydroxymethyl-2-
furoic acid during roasting of coffee. Mol. Nutr. Food Res.,
vol. 51, no. 4, p. 390-394.
https://doi.org/10.1002/mnfr.200600251
Murkovic, M., Pichler, N. 2006. Analysis of 5-
hydroxymethylfurfual in coffee, dried fruits and urine. Mol.
Nutr. Food Res., vol. 50, no. 9, p. 842-846.
https://doi.org/10.1002/mnfr.200500262
Narita, Y., Inouye, K. 2015. Chlorogenic acids from coffee.
In Preedy, V. R. Coffee in health and disease prevention.
Amsterdam, Netherlands : Elsevier, p. 189-199. ISBN 978-0-
12-409517-5. https://doi.org/10.1016/B978-0-12-409517-
5.00021-8
Poole, R., Kennedy, O. J., Roderick, P., Fallowfield, J. A.,
Hayes, P. C., Parkes, J. 2017. Coffee consumption and health:
Umbrella review of meta-analyses of multiple health
outcomes. BMJ : British Medical Journal, vol. 359, p. 1-18.
https://doi.org/10.1136/bmj.j5024
Redgwell, R., Fisher, M. 2006. Coffee carbohydrates. Braz.
J. Plant Physiol., vol. 18, no. 1, p. 165-174.
https://doi.org/10.1590/S1677-04202006000100012
Singleton, V. L., Orthofer, R., Lamuela-Raventós, R. M.
1999. Analysis of total phenols and other oxidation substrates
and antioxidants by means of folin-ciocalteu reagent. Methods
in Enzymology, vol. 299, p. 152-178.
https://doi.org/10.1016/s0076-6879(99)99017-1
Verhoef, J. C., Barendrecht, E. 1977. Mechanism and
reaction rate of the Karl Fischer titration reaction. Analytical
implications. Anal. Chim. Acta, vol. 94, no. 2, p. 395-403.
https://doi.org/10.1016/S0003-2670(01)84541-4
Wei, F., Furihata, K., Koda, M., Hu, F., Miyakawa, T.,
Tanokura, M. 2012. Roasting process of coffee beans as
studied by nuclear magnetic resonance: Time course of
changes in composition. J. Agric. Food Chem, vol. 60, no. 4,
p. 1005-1012. https://doi.org/10.1021/jf205315r
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. Eur. Food Res.
Technol., vol. 214, no. 2, p. 92-104.
https://doi.org/10.1007/s00217-001-0424-7
Acknowledgments:
This work was financially supported by project FCH-S-18-
5334 (The Ministry of Education, Youth and Sports of the
Czech Republic).
Contact address:
*Pavel Diviš, Brno University of Technology, Faculty of
Chemistry, Department of Food chemistry and
Biotechnology, Purkyňova 118, 612 00 Brno, Czech
Republic, Tel.: +420541149454,
E-mail: divis@fch.vut.cz
ORCID: https://orcid.org/0000-0001-6809-0506
Jaromír Pořízka, Brno University of Technology, Faculty
of Chemistry, Department of Food chemistry and
Biotechnology, Purkyňova 118, 612 00 Brno, Czech
Republic, Tel.: +420 54114 9320,
E-mail: porizka@fch.vut.cz
ORCID: https://orcid.org/0000-0002-2742-8053
Jakub Kříkala, Brno University of Technology, Faculty
of Chemistry, Department of Food chemistry and
Biotechnology, Purkyňova 118, 612 00 Brno, Czech
Republic, Tel.: +420541149393,
E-mail: xckrikala@fch.vut.cz
ORCID: https://orcid.org/0000-0002-4776-9517
Corresponding author: *