investigations on the chemistry of coffee processing
TRANSCRIPT
1
Investigations on the Chemistry of Coffee Processing
by
Seung-Hun Lee
a Thesis submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy in Chemistry
Approved Dissertation Committee Prof. Dr. Nikolai Kuhnert Name and title of Chair Prof. Dr. Thomas Nugent Name and title of Committee Member Prof. Dr. Adam Le Gresley Name and title of Committee Member
Date of Defense: 03. 09. 2018
2
Statutory Declaration
Family Name, Given/First Name Lee, Seung-Hun
Matriculation number 20330669
What kind of thesis are you submitting:
Bachelor-, Master- or PhD-Thesis PhD Thesis
English: Declaration of Authorship
I hereby declare that the thesis submitted was created and written solely by myself without
any external support. Any sources, direct or indirect, are marked as such. I am aware of the
fact that the contents of the thesis in digital form may be revised with regard to usage of
unauthorized aid as well as whether the whole or parts of it may be identified as plagiarism. I
do agree my work to be entered into a database for it to be compared with existing sources,
where it will remain in order to enable further comparisons with future theses. This does not
grant any rights of reproduction and usage, however.
This document was neither presented to any other examination board nor has it been
published.
German: Erklärung der Autorenschaft (Urheberschaft)
Ich erkläre hiermit, dass die vorliegende Arbeit ohne fremde Hilfe ausschließlich von mir
erstellt und geschrieben worden ist. Jedwede verwendeten Quellen, direkter oder indirekter
Art, sind als solche kenntlich gemacht worden. Mir ist die Tatsache bewusst, dass der Inhalt
der Thesis in digitaler Form geprüft werden kann im Hinblick darauf, ob es sich ganz oder in
Teilen um ein Plagiat handelt. Ich bin damit einverstanden, dass meine Arbeit in einer
Datenbank eingegeben werden kann, um mit bereits bestehenden Quellen verglichen zu
werden und dort auch verbleibt, um mit zukünftigen Arbeiten verglichen werden zu können.
Dies berechtigt jedoch nicht zur Verwendung oder Vervielfältigung.
Diese Arbeit wurde noch keiner anderen Prüfungsbehörde vorgelegt noch wurde sie bisher
veröffentlicht.
……………………………………………………………………………………………………….
Date, Signature
3
g{x g{xá|á
Wxw|vtàxw àÉ
lÉâÇzlÉâÇzlÉâÇzlÉâÇz@@@@[xx ctÜ~[xx ctÜ~[xx ctÜ~[xx ctÜ~
TÇw
]tx]tx]tx]tx@@@@[çâv~ _xx[çâv~ _xx[çâv~ _xx[çâv~ _xx
4
Table of Contents
Table of Contents ............................................................................................................................. 4
Abstract ............................................................................................................................................ 6
Acknowledgement ............................................................................................................................ 8
Abbreviations ................................................................................................................................. 11
Achievements ................................................................................................................................. 12
Curriculum Vitae ............................................................................................................................ 14
Preface ............................................................................................................................................ 20
1. Introduction ............................................................................................................................. 21
Coffee Chemistry ........................................................................................................... 21 1.1.
Coffee Compounds ............................................................................................. 21 1.1.1.
Chlorogenic acids ............................................................................................... 22 1.1.2.
1.1.2.1. Etymology and classification ................................................................................. 22
1.1.2.2. Polyphenols ........................................................................................................... 23
1.1.2.3. Chlorogenic acids in coffee .................................................................................... 24
1.1.2.4. Analysis of chlorogenic acid .................................................................................. 36
1.1.2.5. Properties of chlorogenic acid ............................................................................... 38
Caffeine .............................................................................................................. 39 1.1.3.
1.1.3.1. Etymology and classification ................................................................................. 39
1.1.3.2. The role of caffeine ................................................................................................ 40
1.1.3.3. Coffee decaffeination ............................................................................................ 41
Coffee Technology ........................................................................................................ 42 1.2.
Coffee Cherry ..................................................................................................... 42 1.2.1.
1.2.1.1. Classifications ........................................................................................................ 42
1.2.1.2. Processes ............................................................................................................... 43
Green Coffee ...................................................................................................... 44 1.2.2.
1.2.2.1. Classifications ........................................................................................................ 44
1.2.2.2. Processes ............................................................................................................... 48
Roasted Coffee ................................................................................................... 50 1.2.3.
1.2.3.1. Classifications ........................................................................................................ 51
1.2.3.2. Processes ............................................................................................................... 53
Ground Coffee .................................................................................................... 54 1.2.4.
1.2.4.1. Classifications ........................................................................................................ 54
1.2.4.2. Processes (Brewing / Extraction) ........................................................................... 55
Extracted Coffee ................................................................................................. 55 1.2.5.
Coffee Sensory Perception ............................................................................................ 56 1.3.
5
Coffee Aroma ..................................................................................................... 56 1.3.1.
Coffee Taste ....................................................................................................... 58 1.3.2.
Tools for Coffee Sensory Analysis .................................................................... 59 1.3.3.
1.3.3.1. World Coffee Research Sensory Lexicon ............................................................... 59
1.3.3.2. The Coffee Taster’s Flavor Wheel .......................................................................... 60
2. Aim of the Research ................................................................................................................ 70
3. Results ..................................................................................................................................... 72
Research on Green Coffee Step ..................................................................................... 72 3.1.
Research on Roasting Step ............................................................................................ 89 3.2.
Research on Brewing Step ........................................................................................... 125 3.3.
Research on Brewing Step with Different Solvents ......................................... 125 3.3.1.
Research on Brewing Step with Different Methods ......................................... 139 3.3.2.
Research on Application to Other Food ...................................................................... 167 3.4.
4. Conclusion ............................................................................................................................. 194
6
Abstract
The International Coffee Organization announced that coffee consumption has increased for 4
years consecutively by 1.9 % every year from 2013/14 to 2016/17. The economic importance
of coffee is continuously increasing, so that research on coffee becomes more and more
important. Chlorogenic acids (CGAs) and caffeine are the best investigated components
among all the coffee compounds after Clifford group developed easy identification method by
using novel LC-MSn at 2003. They have been identified as the most health-affecting
compounds in coffee.
A lot of coffee research has investigated the chemical composition of green or roasted coffee
beans and the final cups of coffee by academics and industries, respectively. However, still
there is little research studying the role of composition changes caused by each individual
coffee processes. Furthermore, increasing demands of specialty coffee promotes new
developments of coffee processes but the inventors need scientific input for analyzing the
component with bioactivities to support and market their innovations.
In this thesis, the different stages of coffee processing have been investigated using suitable
experiments in those stages; cultivating, grading, roasting and brewing steps were investigated.
Additionally, those methods were applied to analyze Korean traditional green tea samples.
The different grades of Jamaican Blue Mountain coffee, a product of high price and rarity,
were analyzed by LC-MSn together with different originated common coffees. The relative
ratios of regioisomers of caffeoylquinic acids (CQAs) were determined as the possible
specialty flavor for the first time from this source.
A new coffee product marketed as fermented coffee, obtained by a consecutive steaming and
drying process was also investigated. This product, which is marketed as a roasting substitute
for coffee, showed the highest antioxidant property if compared to the green coffee bean and
two roasted coffee beans with different degree of roasting. This was the first time reported
that the heating processed coffee has higher antioxidant property than the green coffee bean
has. The differences of chemical composition of this new process were also identified by
using LC-MSn method.
7
For the brewing step of coffee process experiment, the different composition ratios of
chlorogenic acids by changing extracting solvent, which is easily predicted but not yet
scientifically established and published, were identified. Moreover, cold-brew coffee, which
is a recent coffee brewing method, is analyzed for the first time in this study.
By combining all the coffee experiments, Korean green tea samples were analyzed by the
same LC-MSn method as used in coffee analysis. The high-priced special tea samples were
selected as Jamaica Blue Mountain coffee is in coffee and the producing method includes
consecutive heating and drying process as fermented coffee in coffee analysis. Additionally,
different extracting solvents were used as brewing step experiment in coffee analysis. To sum
up, all the key-points in the coffee experiments in this thesis were applied to other food
analysis and could find the applicability of the methods as components controller factors.
Through these studies, regioisomeric ratio of chlorogenic acids were suggested as a quality
influencing factor, a coffee processing method with higher antioxidant property was found,
the extraction influences on the kind and quantity of solvents were identified and the
applicability of all the methods from coffee to other food techniques was confirmed.
8
Acknowledgement
First of all, I would like to begin by expressing my gratitude to Prof. Dr. Nikolai Kuhnert, not
only as my supervisor of my Ph.D. research, but also my advisor of my life in Germany for
work/life balance, linguistic development and being a father in a family. Through his
guidance, encouragement and motivation, I could develop myself as an analytic scientist,
although it is totally different field I had done in Korea as a synthetic chemist. Moreover, I
could settle in Germany as a leader of 3-membered family.
I would also like to acknowledge Prof. Dr. Thomas Nugent and Prof. Dr. Adam Le Gresley
for taking time to review my thesis. Prof. Nugent made me maintain as a scientist when I lean
toward coffee field during entire of my Ph.D. period. And Prof. Adam Le Gresley gave me
the insight, being as an independent scientist from Prof. Kuhnert’s guidance.
I would like to thank to sending me the coffee and tea samples from all over the world for my
research. The people working in the Embassy of the Republic of Korea in Jamaica and
Young-gi Moon helped me Jamaica Coffee Research by sending regional samples. Cafe
Romeo helped me Cold-brew Coffee Research by sending their brewing machines. Coffeebio
helped me Fermented Coffee Research by sending their Fermented coffee samples and also
helped me to compare with different roasted coffee samples. Geumro helped me Korean
Traditional Green Tea Research by sending their consecutive parched green tea samples. All
the supports did not charged me anything, genuinely supported for the improvement of
scientific food understanding.
Furthermore, Genesis and Lloyd coffee, though it was not free, provided me a special
discount and fruitful informations of the products for my research. Genesis provided their 100
~ 250 g coffee roaster, which is perfect for the lab-scale research. Lloyd coffee provided a
variety of green coffee bean samples in different origins. Additionally, the people working in
the Embassy of the Republic of Korea in Yemen, Eunsang Kim, Coffee and Tea, Seung-Kuk
Park, Nahee Kim, Joon-Kwan Moon, Juho Choi, Stronghold, Jung-Im You, Byunggun Park
and Jikwang Han did not work together with me, but gave fruitful advices for my coffee
research.
And there were a variety of associations which I would like to express my gratitude to.
Starting from Jacobs University Bremen for providing me a nice environment for my research,
DAAD for scholarship, GDCh for scholarship to conferences, VeKNI for making me feeling
9
at home in Germany and scholarship, Cocotea conference for providing continuous
opportunities to keep me up-to-date on coffee research and Kids at Jacobs for letting me
concentrate on my research out of the babycare.
I would like to individually thank all of my colleagues. Starting with our lab’s excellent
technicians, Anja Mueller for her total-care for our lab core facilities, should be firstly
referred because all our lab’s research could not have done without her contribution. Britta
Behrends established our lab’s systematic management system, so directly or indirectly, all
our lab’s results are supported by her contribution. And I thank to those people worked
together, Dr. Rakesh Jaiswal for the instructions of MS, Dr. Roy Dsouza for physical -
calculational help, Dr. Sagar Deshpande for his Energy to our lab, Dr. Hande Karaköse for her
lab tea-time management, Dr. Agnieszka Golon for her happiness to the lab, Dr. Mohamed
Elsadig for his warm-invitation and interesting episodes, Dr. Matei, Marius-Febi for his
interesting stories for German football system, Dr. Maria Alexandra Patras for her exemplary
attitudes and a lot of practical advices of baby care, Dr. Abhinandan Shrestha for taking the
most time in the office without drinking coffee, Dr. Rohan Shah for his encouraging me to
brew better coffee and pleassure talks about coffee, Dr. Inamullah Hakeem Said for his
contributions for managing lab members, Dr. Diana Sirbu for showing diligent attitudes to
make me motivated, Dr. Roberto Megias for his friendly discussions and the honey-tea gift
from Spain, Anastasiia Shevchuk for her contributions for lab-teatime and helps for visiting
Asian Cocotea, Pawel Andruszkiewicz for being the center for connecting before and after
students, Maria Bikaki for her contributions of members’ birthdays, Sabur Badmos for being
my direct theme-successor and made a lot of fruitful results to fulfill my curiosity on coffee,
Fariba Sabzi for interesting informations of nuts and showing improvement of her lifestyle,
Tahira Musarrat for balancing our lab’s theme with her background of synthesis,
Hossameldeen Sharaf for making me comfortable even though we did not have enough time
together. Also, I would like to thank to Euisuk Han and Suhyun Moon, for being my students
to work together with low acidic coffee and human Luwak coffee projects. Additionally, these
people made my Ph.D. period happier even though we did not work together; Beatrix Tetzlaff,
Prof. Hartwig, Dr. Mubo, Prof. Lalith, Borislav Milev, Dr. Ghada Yassin, Nickolet Ncube, Dr.
Jonathan Cave and Dr. Maria Isbel Alarcon.
I would like to thank the Korean communities in Bremen to make me feel at home with
Korean language environment. I thank to Jacobs Korean community to make regular
gathering twice per year, to make Korean foods and good companies. Starting from the big
10
brother Jonghak who has drunk coffee the most times together with me, Aemin, Seo-bros,
Hyeyoung, Eunhye, Hanbit, Hojun who kept me as an early adapter, Minseok, Sojung who
was a nice consultant with drinking coffee, Haeun, Ikyung, Suhyun, Euisuk, Myunggi, Dain,
Jangsup, Jaeho, Youngjun, Hyunseok and all the Koreans I could not mention here made my
Ph.D. period more comfortable. Also, it was a great consolation that the two Korean churches
and Bremen Korean society were always in the backup whenever the Koreans needed any
types of helps. Especially the times with the coincidence-inevitability friend Hyoseup and
Hyejin, Jung’s family who helped a lot for my starting period in Bremen, Sujin and Wang’s
family, Sanga and Hyunggu are my indelible memories. Also, the three Korean restaurants in
Bremen were very helpful whenever I wanted to taste foods from Korea.
The friends, always be there and make me happy from Korea, are also appreciated. My formal
colleges of Samyang Group, Kwangsu, Kyungdo, Wonjun, Sanghun and the friends of my
childhood, Sangjin, Myunggyu, Kyunghwa and the friends of my University periods, Suyeon,
Hyejin and the friends of my English studies were always there, made me feel at home.
Especially, Kyunghwa and Hyejin visited Germany during my Ph.D. periods, so I want to
express my thanks again here. And I would like to thank to Kyung-ju Hwang for his advices
for the health of myself and my families.
I would like to thank to Prof. Jung-Il Jin, who was the advisor of my master degree, since he
always gives me advices for my future, and supports me as much as he can. Together with
him, the members of Gogohoi would also be appreciated with their warm words and cares
from Korea.
Last but not the least, I thank to my family (and family-in-law) in Korea and the family living
in Berlin and Essen in Germany. I thank to especially my parents, sister, Susie, Woosung, and
Junghun for physically coming from Korea to Germany to see us. Although the other family
members could not visit us, I know that they always care for us, and welcome me whenever I
visit Korea from time to time. I thank to Minjee and Subin to be always with me, be my side
and will always be in the future.
11
Abbreviations
BPC Base Peak Chromatrogram
CA Caffeic Acid
CCoQA Caffeoylcoumaroylquinic Acid
CFQA Caffeoylferuloylquinic Acid
CGA Chlorogenic Acid
CGAs Chlorogenic Acids
CoQA Coumaroylquinic Acid
CQA Caffeoylquinic Acid
CQL Caffeoylquinic Acid Lactone
CSA Caffeoylshikimic Acid
diCQA Dicaffeoylquinic Acid
DFQA Dimethoxycinnamoylferuloylquinic Acid
DQA Dimethoxycinnamoylquinic Acid
EIC Extracted Ion Chromatrogram
ESI Electrospray Ionisation
FQA Feruloylquinic Acid
FSQA Feruloylsinapoylquinic Acid
FTMQA Feruloyltrimethoxycinnamoylquinic Acid
HPLC High Performance Liquid Chromatography
M-CQA Caffeoylquinic Acid Methyl Ester
M-diCQA DiCaffeoylquinic Acid Methyl Ester
MeOH Methanol
MS Mass Spectrometry
NMR Nuclear Magnetic Resonance
QA Quinic Acid
SiQA Sinapoylquinic Acid
TIC Total Ion Chromatogram
TOF Time of Flight
12
Achievements
Publications
• Seung-Hun Lee, Rakesh Jaiswal, Nikolai Kuhnert, Analysis on different grades of
highly-rated Jamaica Blue Mountain coffees compared to easily available originated
coffee beans, SCIREA Journal of Food. Vol. 1 , No. 2 , 2016 , pp. 15 – 27
• Seung-Hun Lee, In Jae Kim, Sabur Badmos and Nikolai Kuhnert, Analysis of Steamed
(“Fermented”) Coffee Beans as a Roasted Coffee Substitute, 2018, in preparation
• Seung-Hun Lee, Nikolai Kuhnert, Comparision of the chlorogenic acids content of
laboratorial and practical solvent extracted coffees, 2018, in preparation
• Seung-Hun Lee, Nikolai Kuhnert, Brewing Less- and Enhanced-caffeinated Coffee by
Using Cold Brew Method, 2018, in preparation
• Seung-Hun Lee, Inamullah Hakeem Said, Un-Jae Lee, Nikolai Kuhnert, Analysis on
kinetic parching samples of Korean traditional green tea process with different
brewing methods, 2018, in preparation
Poster Presentations
• Characterization of Three Different Grades of Jamaica Blue Mountain Coffee by LC-
MSn
Second International Congress on Cocoa Coffee and Tea, 2013, Naples, Italy
• Analysis of Fermented Coffee Beans as a Roasted Coffee Substitute
Bremen Life Sciences Meeting 2015, 2015, Bremen, Germany
• Identification of Composition Kinetics from Cold-brewed Coffee by LC-MSn
Third International Congress on Cocoa Coffee and Tea, 2015, Aveiro, Portugal
• Identification of Composition Kinetics of Traditional Korean Green Tea Producing
Method by LC-MS
The Korean Society of Food and Nutrition Summer Conference, 2016, Seoul, Korea
• Analysis of Less- and Enhanced- Caffeinated Coffee: LeeN-Caffeination
Korea Society of Food and Cookery Science, 2017, Seoul, Korea
• Identification of composition kinetics on applications of the LeeN caffeination coffee
13
Fourth International Congress on Cocoa Coffee and Tea, 2017, Turin, Italy
Articles in Popular Press
• Compounds to diabetes in coffee and tea, Coffee & Tea, 2017. 02.
• Polyphenols and antioxidant effect, Coffee & Tea, 2017. 04.
• Flavor of coffee? Minimization of Furans?, Coffee & Tea, 2017. 06.
• Is really Dutch coffee decaffeinated? Are more than 10 % coffee companies
unscrupulous?, Coffee & Tea, 2017. 08.
• Is coffee good for health or not? Why does the debate still go on?, Coffee & Tea,
2017. 10.
• The great scholars on coffee research, Coffee & Tea, 2017. 12.
• Coffee addiction? Caffeine addiction? The story of addiction and dopamine, Coffee &
Tea, 2018. 02.
• Trigonelline and Vitamin B3 in Coffee, Coffee & Tea, 2018. 04., submitted.
Books
• Chlorogenic acids in Coffee, Book chapter invited, The Royal Society of Chemistry
(RSC)
Awards
• 1st Prize, Home Brewers Cup Bremen, Bremen, 2014
• Excellent Poster Prize, The Korean Society of Food and Nutrition, Seoul, 2016
14
Curriculum Vitae
Seung-Hun Lee
Personal detail
Name in Full: Seung-Hun Lee
Date of Birth: 16 December 1982
Gender: Male
Nationality: Republic of Korea
Address: Metroheights 101 Dong 602 Ho, Jeonpo 2-dong, Busanjingu, Busan, Korea
Schllighoern 6, Bremen, Germany
Telephone Number: +49 176 8464 3086
E-mail: [email protected], [email protected]
Education Background
Sep. 2012 ~ Ph.D. Analytical Chemistry
Advisor: Professor Nikolai Kuhnert
Jacobs University Bremen, Germany
Mar. 2005 ~ Feb. 2007 M.Sc. in Organic Chemistry
Advisor: Professor Jung-Il Jin
Korea University, Korea
Mar. 2001 ~ Feb. 2005 B.Sc. Department of Chemistry
Korea University, Korea
Employment
Feb. 2007 ~ Apr. 2010 Researcher
Information & Electronic Materials Research
Advanced Polymeric Materials R&D Center
Samyang Corporation
15
Certification
2008 Balloon Artist , Party Planner Education Association
2009 Barista, Korea Coffee Education Society
2010 Chinese Character, Korea Test Association
2011 Magic Instructor , The Korean Magic Society
2012 Korean History Proficiency, National Institute of Korean History
Scholarship
2005 1st Semester, BK Scholarship
2005 2nd Semester, General Scholarship
2006 1st Semester, Research Assistant Scholarship
2006 2nd Semester, Administration Assistant Scholarship
& Samyang Education Sponsorship
2017 DAAD Merit-based Scholarship (by Wolfgang-Ritter Stiftung)
Activity
2003 Student Leader, Department of chemistry, Korea University
2008 Band Leader, R-band, Samyang Research Center
2009 Mensa Korea Life Member (IQ: over 156 [SD=24])
2011 Master, Coffle SIG, Mensa Korea
Award
2000 Prize for Encouragement, Math Olympiad, Daejeon, Korea
Hannam University & The Korean Mathematical Society
2014 1st Prize, Home Brewers Cup Bremen, Bremen, Germany
2016 Excellent Poster Prize, The Korean Society of Food and Nutrition
16
Achievement
Poster
1. Photopatternable Electroluminescent Polymers: MEH-PPV and PVK
98th General Meeting of Korean Chemical Society, 2006, Gwangju, Korea
2. Characterization of Three Different Grades of Jamaica Blue Mountain Coffee by
LC-MSn
Second International Congress on Cocoa Coffee and Tea, 2013, Naples, Italy
3. Analysis of Fermented Coffee Beans as a Roasted Coffee Substitute
Bremen Life Sciences Meeting 2015, 2015, Bremen, Germany
4. Identification of Composition Kinetics from Cold-brewed Coffee by LC-MSn
Third International Congress on Cocoa Coffee and Tea, 2015, Aveiro, Portugal
5. Identification of Composition Kinetics of Traditional Korean Green Tea Producing
Method by LC-MS
The Korean Society of Food and Nutrition Summer Conference, 2016, Seoul, Korea
6. Analysis of Less- and Enhanced- Caffeinated Coffee: LeeN-Caffeination
Korea Society of Food and Cookery Science, 2017, Seoul, Korea
7. Identification of composition kinetics on applications of the LeeN caffeination
coffee
Fourth International Congress on Cocoa Coffee and Tea, 2017, Turin, Italy
Publication
1. “Poly(fluorenevinylene) Derivatives by Heck Coupling: Synthesis, Photophysics,
and Electroluminescence”
J. Polym. Sci. Part A: Polym. Chem. 44, 4494-4507 (2006).
2. “Photopatternability of poly(vinylcarbazole) bearing cinnamate pendants and its
blends with a soluble poly(p-phenylene vinylene) derivative”
MACROMOLECULAR RESERCH, 15, 142 (2007).
3. “Enhanced Lifetime of Organic Light-Emitting Diodes Using an Anthracene
Derivative with High Glass Transition Temperature”
Journal of Nanoscience and Nanotechnology 13, 4216-4222 (2013).
17
4. “Analysis on different grades of highly-rated Jamaica Blue Mountain coffees
compared to easily available originated coffee beans”
SCIREA Journal of Food. 1(2), 15-27 (2016) .
Patent
1. Polysilsesquioxane-based organic-inorganic hybrid graft copolymer, organo-silane
comprising porogen used for preparation of the same and Method for preparation of
insulating film comprising the same, AN=[10-2007-0140502]
2. Method of preparing insulating film with using norbornene-based silane derivative
and Insulating film prepared by the same, AN=[10-2007-0140517]
3. Norbornene-based silsesquioxane copolymers, Norbornene-based silane derivative
used for preparation of the same and Method of preparing low dielectric insulating
film comprising the same, AN=[10-2007-0134577]
4. Photoactive resin composition for insulation layer, AN=[10-2009-0131139]
5. Positive Photoresist Composition for insulation layer, AN=[10-2009-0131033]
6. A photosensitive composition for insulation layer, AN=[10-2010-0130912]
Article in Popular Press
1. Compounds to diabetes in coffee and tea, Coffee & Tea, 2017. 02.
2. Polyphenols and antioxidant effect, Coffee & Tea, 2017. 04.
3. Flavor of coffee? Minimization of Furans?, Coffee & Tea, 2017. 06.
4. Is really Dutch coffee decaffeinated? Are more than 10 % coffee companies
unscrupulous?, Coffee & Tea, 2017. 08.
5. Is coffee good for health or not? Why does the debate still go on?, Coffee & Tea,
2017. 10.
6. The leading scientists; who study coffee, Coffee & Tea, 2017. 12.
7. Coffee addiction? Caffeine addiction? Addiction and Dopamine, Coffee & Tea,
2018. 02.
8. Trigonelline and Vitamin B3 in Coffee, Coffee & Tea, 2018. 04., submitted.
18
Books
1. Chlorogenic acids in Coffee, Book chapter invited,
The Royal Society of Chemistry (RSC)
Research Skills
1. Structure Design, Synthesizing, and Characterization of new materials
2. Analysis on unknown samples (Tools available: NMR, IR, UV, PL, TGA, DSC,
GPC, EA, GC-MS, ICP, etc)
3. Interpretation of polyphenols and caffeine with HPLC-TOFMS and HPLC-
tandem MS
Interests
1. Food chemistry, analysis of coffee and tea 2. Mass Spectrometry
3. Optoelectronic devices based on organic materials (OLED, OPV, OTFT) 4. Chemical Education
References
Professor Jung-Il Jin
Past President, IUPAC
Department of Chemistry, Korea University, Korea
Professor Nikolai Kuhnert
School of Engineering and Science, Jacobs University Bremen, Germany
Professor Jeonghun Kwak
School of Electrical and Computer Engineering, University of Seoul, Korea
19
20
Preface
Following water and black tea, coffee is the third most consumed beverage globally with an
average daily consumption of 400 mL per capita. Not only in the beverage category, coffee is
the third most traded commodity in the category of food and 15th the entirely of the most
valuable commodity in the world.
The world production of green coffee was 3,522,508 tons in 1976 and became 8,790,005 tons
in 2014. Especially in China, people started to produce green coffee around 2000, joined the
group of coffee producing top 20 countries in the world in 2009, became 12th coffee
producing country in 2013 and still highly increasing their coffee production. Annual coffee
trade is worth an annual 120 billion US$ in raw material and coffee provide a livelihood for
an estimated 25 million farmers.
Also in the research field, coffee is a very popular and still growing topic. In November, 2012,
there were already 38,589 results by SciFinder with the topic of coffee. In August, 2017, the
result increased to 52,275, which means an average 3000 papers were published annually
related to coffee in the recent five years.
In this respect, to know the constituent of coffee, the processes from farm to table of coffee
and coffee analysis methods are needed to be considered together. This project reviews the
composition of coffee, processes and analysis methods and shows results of analysis of
compositional changes for novel methods of coffee processing through the whole coffee
processes.
21
1. Introduction
Coffee Chemistry 1.1.
A substantial body of research has supported the notion that drinking coffee is beneficial for
human health especially affecting the reduction of cancer, type II diabetes1, Parkinson’s
Disease2, liver disease3 and improving athletic performance4. Hence coffee research has
become more and more important. With this flow, the most emphasized compounds in coffee
are chlorogenic acids, especially since Clifford et al. developed a new easy and fast analytical
method for profiling different regioisomers of chlorogenic acids using HPLC-tandem MS in
2003.5 After this research, coffee polyphenols, which is represented by chlorogenic acids,
entered the mainstream of food research using the innovative method of characterization.
In addition, although there are still controversial arguments addressing its health influence,
the compound caffeine is an attractive compound for analysis.2,4,6,7 Particularly many energy-
drinks started to be produced in 21th century, caffeine became a special issue with respect to
daily intake recommendations. Coffee is the main source of caffeine in normal human diet in
many countries, so the caffeine research in coffee is one of the two major research fields,
together with chlorogenic acids research in coffee. Current figures recommend a daily intake
of 400 mg, with consumption of in excess of 1 g / day leading to caffeinism and an LD50 dose
of 150 mg / kg.
In this thesis, the composition of coffee of varying types, chemical compositions and
properties are investigated and discussed with a focus on new applications. LC-MS methods
are predominantly used for coffee composition analysis.
Coffee Compounds 1.1.1.
Coffee is mainly composed of the three energy providing substances; carbohydrate, protein
and lipids.8 The three substances account for about 80 % of the total dry green coffee beans.
Table 1 shows component of coffee in two representative species, Coffea arabica (arabica
coffee) and Coffea canephora (robusta coffee) which are the most commonly used coffee
products. However, those principal ingredients do not attract people’s interest because we
only consume two cups of coffee per day in Germany, which is one of the most coffee
consuming countries in the world, and two cups of coffee has only 5 g of solid coffee content.
22
Table 1. Component of coffee in Arabica and Robusta coffee beans.
Components % on dry basis Arabica Robusta
Caffeine 1.2 2.2 Trigonelline 1.0 0.7 Minerals (41% = K) 4.2 4.4 Acids Chlorogenic acids – total 6.5 10.0 Aliphatic acids 1.0 1.0 Quinic acid 0.4 0.4 Sugars Sucrose 8.0 4.0 Reducing sugars 0.1 0.4 Polysaccharides 44.0 48.0 Lignin 3.0 3.0 Pectin 2.0 2.0 Proteins 11.0 11.0 Free Amino acids 0.5 0.8 Lipids 16.0 10.0
Although energy providing substances are the main ingredients, the total amount of intake is
even less than 1 % of what people usually take from other food. On the other hand,
chlorogenic acids and caffeine take up 5 ~ 10 % and 1 ~ 2 %, respectively in the coffee
component, but their main source of supply to human is coffee. Therefore, chlorogenic acids
and caffeine analysis becomes the main interest on coffee research.
Chlorogenic acids 1.1.2.
1.1.2.1. Etymology and classification
Chlorogenic acids are natural chemical compounds that are in general terms defined as
hydroxycinnamoyl esters of quinic acids.9-11 Typical substituents are caffeic acid, ferulic acid,
p-coumaric acid and others. The most common chlorogenic acids in coffee is 5-caffeoylquinic
acid of which chemical structure is shown in Figure 1 and it is commercially available in the
market with the trivial name of chlorogenic acid.12
HO
HO O
HOOH
O
O
OH
OH
Figure 1. Chemical structure of 5-caffeoylquinic acid
23
The name chlorogenic acid comes from the Greek χλωρός (light green) and -γένος (a suffix
meaning ‘give rise to’), because of the green color produced when chlorogenic acids are
oxidized.13 The green color obtaining reaction has been found by a group of Japanese
scientists, which is by chlorogenic acid oxidizing with protein of sunflower seeds to make
green pigments. 14
Chlorogenic acids are classified as one of the polyphenols from the fruit, and these
polyphenols are regarded as antioxidant compounds.15 After the name polyphenol started to
be used, research on natural polyphenols became one of the mainstreams in order to explain
the benefits of fruits for our health. The ‘French Paradox’16 is one of the most famous
examples illustrating the potential health benefits of phenolics in cardiovascular diseases. 17
1.1.2.2. Polyphenols
Etymology of polyphenol
First of all, it should be highlighted that polyphenols are not polymers. Most of the
polyphenols do not have repeating units in their structures, and their molecular weights are
below 5000 Da. The word polyphenols has coexisted with the terms plant phenolics and
tannins until the term polyphenols dominate the literature around 2000.
The most employed definition of polyphenols is WBSSH definition.18 The White-Bate-Smith-
Swain-Haslam (WBSSH) describes the polyphenol class as water soluble, 500 – 3000 ~ 4000
Da, possessing 12 to 16 phenolic hydroxyl groups on five to seven aromatic rings per 1000 Da.
However, according to this definition, only a few chlorogenic acids might be included in the
polyphenol group. 5-Caffeoylquinic acid, which is the most abundant chlorogenic acid in
coffee, has only 354.31 g/mol, so that it cannot reach the low limit of 500 Da with the
WBSSH definition. Hence CGAs fall under the definition of plant phenolics. It is different
from general cognition that coffee has a lot of polyphenolic compounds including
caffeoylquinic acids.
Stéphane Quideau suggested that the definition need to be changed in 2011.15 He proposed a
new definition of polyphenol to include the small compounds which could be easily identified
after the analytical development of Clifford et al.5 The new definition is shown below.
24
The term “polyphenol” should be used to define plant secondary metabolites derived
exclusively from the shikimate-derived phenylpropanoid and/or the polyketide pathway(s),
featuring more than one phenolic ring and being devoid of any nitrogen-based functional
group in their most basic structural expression.
In summary, the term polyphenol is now widely used as a common designation of compounds,
which contains multiples of phenolic rings.
The role of polyphenols
Polyphenol as UV-absorbers
It is well-known that the UV-B light cause direct DNA-damage to produce thymine-thymine
cyclobutane dimers.19 Since the wavelength of UV-B light is between 290 nm and 320 nm,
polyphenols could easily absorb the light. From benzene to phenol, the absorption goes
through red-shift from 254 nm to 270 nm. Additional hydroxyl groups produce additional red
shift of absorption to the UV-B range. The protection against DNA damaging solar radiation
is one of the reasons of mounting research on polyphenols. It is believed that plants when
leaving the aqueous habitat, moving on land, have evolved polyphenols as secondary
metabolics to serve as UV protecting agents.
Polyphenol as antioxidants
Antioxidant property plays a more important role in recent research on polyphenols than DNA
protection. In view of several neurological researches such as Alzheimer’s disease or
Parkinson’s disease and diabetes, antioxidant study provides important scientific concept in
development of drugs for managing the disorders. Polyphenols discovery as antioxidant
compounds inspired increment of antioxidant investigation,20 and seemed to explain well with
the health beneficial effect by antioxidant theory. However, the low absorption of the
polyphenols into our body weakened the adaption of polyphenol as health beneficial
antioxidants.21 The biological activity of chlorogenic acids would follow to go into details.
1.1.2.3. Chlorogenic acids in coffee
Chorogenic acids content in green coffee bean
Among the polyphenols, chlorogenic acids have over the last two decades attracted scientific
interest. Chlorogenic acids contribute largely to daily dietary burden of polyphenols with an
25
estimated daily intake of at least 1 g. Coffee is the major CGA source with 280 mg / cup of
coffee. A cup is in coffee research defined as a 200 mL volume.
Occurrence
Coffee is known as one of the most abundant natural plant sources of chlorogenic acids with
bamboo and maté. Although chlorogenic acids are also found in most vegetables and fruits
such as peach, prunes and berries22, chlorogenic acids research focuses on coffee because
coffee is most abundant source of obtaining chlorogenic acids in contemporary human diet.
The main compound of chlorogenic acids in coffee is 5-caffeoyl quinic acid. As shown in
Figure 2. a) and b), the three different isomers of caffeoylquinic acids (3-CQA, 5-CQA and
4-CQA) are the main compounds among the coffee composition. Figure 2 shows a typical
chromatogram of a coffee infusion. It can be seen that 5-CQA is generally used as the
representative standard polyphenolic compound in coffee with the highest abundance in the
chromatogram.
Figure 2. LC-MS chromatogram in negative ion mode of a Brazil green coffee extract; a)
total ion chromatogram, b) extracted ion chromatogram of m/z 353 ± 0.5
Structures
As already mentioned the term chlorogenic acid in chapter 1-1-2. Chlorogenic acids,
chlorogenic acids are composed hydroxycinnamic acids esterified with quinic acid. The
chemical structure of chlorogenic acids and their nomenclature are shown in Figure 3 and
Table 2 below. The priority of carbons in quinic acid is followed by IUPAC nomenclature;
Ra, Rc, Rd and Re are the position of 1, 3, 4 and 5, respectively.
26
a) ORa ORc
ORdOReO
R'O
R’ = b1)
O
OH
OH
b2)
O
OH
O
b3)
O
OH
Figure 3. Chemical structure of quinic acid (a)) and typical hydroxycinnamoyl substituents;
b1) Caffeoyl-, b2) Feruloyl-, b3) Hydroxycinnamoyl-
Table 2. List of chlorogenic acid derivatives
No Name Abb Q Rx R’ 1 Caffeoylshikimic acid CSA 1db2 C H 2 Caffeoyl-1,5-quinide CQL 1la5 C la 3 p-Coumaroylquinic acid CoQA - Co H 4 Caffeoylquinic acid CQA - C H 5 Feruloylquinic acid FQA - F H 6 Caffeoylquinic acid methyl ester M-CQ - C CH3 7 diCaffeoylquinic acid diCQA - diC H 8 Caffeoylferuloylquinic acid CFQA - C, F H 9 Feruloyltrimethoxycinnamoylquinic acid FTMQA - TM, F H 10 Feruloylsinapoylquinic acid FSQA - S, F H
In Table 3, there are 69 chlorogenic acid compounds structures published by Jaiswal et al
pursuant to attached groups in Figure 4.23 These chlorogenic acids in table 3 are all found in
natural green coffee beans, and there are also more chlorogenic acids found in roasted coffee
formed by a set of well investigated reactions during roasting such as acyl migration,
lactonisation, epimerization or dehydration.24,25
27
Table 3. Reported chlorogenic acids in coffee bean
No. Name Abbreviation Rc Rd Re
1 3-O-caffeoylquinic acid 3-CQA C H H
2 4-O-caffeoylquinic acid 4-CQA H C H
3 5-O-caffeoylquinic acid 5-CQA H H C
4 3-O-feruloylquinic acid 3-FQA F H H
5 4-O-feruloylquinic acid 4-FQA H F H
6 5-O-feruloylquinic acid 5-FQA H H F
7 3-O-p-cormaroylquinic acid 3-pCoQA pCo H H
8 4-O-p-cormaroylquinic acid 4-pCoQA H pCo H
9 5-O-p-cormaroylquinic acid 5-pCoQA H H pCo
10 3-O-dimethoylcinnamoylquinic acid 3-DQA D H H
11 4-O-dimethoylcinnamoylquinic acid 4-DQA H D H
12 5-O-dimethoylcinnamoylquinic acid 5-DQA H H D
13 3-O-sinapoylquinic acid 3-SiQA Si H H
14 4-O-sinapoylquinic acid 4-SiQA H Si H
15 5-O-sinapoylquinic acid 5-SiQA H H Si
16 3,4-di-O-caffeoylquinic acid 3,4-diCQA C C H
17 3,5-di-O-caffeoylquinic acid 3,5-diCQA C H C
18 4,5-di-O-caffeoylquinic acid 4,5-diCQA H C C
19 3,4-di-O-feruloylquinic acid 3,4-diFQA F F H
20 3,5-di-O-feruloylquinic acid 3,5-diFQA F H F
21 4,5-di-O-feruloylquinic acid 4,5-diFQA H F F
22 3,4-di-O-p-cormaroylquinic acid 3,4-dipCoQA pCo pCo H
23 3,5-di-O-p-cormaroylquinic acid 3,5-dipCoQA pCo H pCo
24 4,5-di-O-p-cormaroylquinic acid 4,5-dipCoQA H pCo pCo
25 3-O-feruloyl-4-o-caffeoylquinic acid 3F-4CQA F C H
26 3-O-caffeoyl-4-o-feruloylquinic acid 3C-4FQA C F H
27 3-O-feruloyl-5-o-caffeoylquinic acid 3F-5CQA F H C
28 3-O-caffeoyl-5-o-feruloylquinic acid 3C-5FQA C H F
29 4-O-feruloyl-5-o-caffeoylquinic acid 4F-5CQA H F C
30 4-O-caffeoyl-5-o-feruloylquinic acid 4C-5FQA H C F
31 3-O-dimethoxycinnamoyl-4-o-caffeoylquinic acid 3D-4CQA D C H
32 3-O-dimethoxycinnamoyl-5-o-caffeoylquinic acid 3D-5CQA D H C
33 4-O-dimethoxycinnamoyl-5-o-caffeoylquinic acid 4D-5CQA H D C
34 3-O-dimethoxycinnamoyl-4-o-feruloylquinic acid 3D-4FQA D F H
35 3-O-dimethoxycinnamoyl-5-o-feruloylquinic acid 3D-5FQA D H F
36 4-O-dimethoxycinnamoyl-5-o-feruloylquinic acid 4D-5FQA H D F
37 3-O-p-coumaroyl-4-o-caffeoylquinic acid 3pCo-4CQA pCo C H
38 3-O-caffeoyl-4-o-p-coumaroylquinic acid 3C-4pCoQA C pCo H
39 3-O-p-coumaroyl-5-o-caffeoylquinic acid 3pCo-5CQA pCo H C
40 3-O-caffeoyl-5-o-p-coumaroylquinic acid 3C-5pCoQA C H pCo
41 4-O-caffeoyl-5-o-p-coumaroylquinic acid 4C-5pCoQA H C pCo
42 4-O-p-coumaroyl-5-o-caffeoylquinic acid 4pCo-5CQA H pCo C
43 3-O-p-coumaroyl-4-o-feruloylquinic acid 3pCo-4FQA pCo F H
44 3-O-p-coumaroyl-5-o-feruloylquinic acid 3pCo-5FQA pCo H F
45 4-O-p-coumaroyl-5-o-feruloylquinic acid 4pCo-5FQA H pCo F
28
46 4-O-dimethoxycinnamoyl-5-o-p-coumaroylquinic acid 4D-5pCoQA H D pCo
47 3-O-p-coumaroyl-4-o-dimethoxycinnamoylquinic acid 3pCo-4DQA pCo D H
48 3-O-p-coumaroyl-5-o-dimethoxycinnamoylquinic acid 3pCo-5DQA pCo H D
49 3-O-sinapoyl-5-o-caffeoylquinic acid 3Si-5CQA Si H C
50 3-O-sinapoyl-4-o-caffeoylquinic acid 3Si-4CQA Si C H
51 3-O-(3,5-dihydroxy-4-methoxy)cinnamoyl-4-o-feruloylquinic acid 3DM-4FQA DM F H
52 4-O-sinapoyl-3-o-caffeoylquinic acid 4Si-3CQA C Si H
53 3-O-sinapoyl-5-o-feruloylquinic acid 3Si-5FQA Si H F
54 4-O-sinapoyl-5-o-feruloylquinic acid 4Si-5FQA H Si F
55 4-O-sinapoyl-3-o-feruloylquinic acid 4Si-3FQA F Si H
56 4-O-trimethoxycinnamoyl-5-o-caffeoylquinic acid 4T-5CQA H T C
57 3-O-trimethoxycinnamoyl-5-o-caffeoylquinic acid 3T-5CQA T H C
58 3-O-trimethoxycinnamoyl-5-o-feruloylquinic acid 3T-5FQA T H F
59 3-O-trimethoxycinnamoyl-4-o-feruloylquinic acid 3T-4FQA T F H
60 4-O-trimethoxycinnamoyl-5-o-feruloylquinic acid 4T-5FQA H T F
61 3-O-dimethoxycinnamoyl-4-o-feruloyl-5-o-caffeoylquinic acid 3D-4F-5CQA D F C
62 3,4,5-tri-O-caffeoylquinic acid 3,4,5-triCQA C C C
63 3,5-di-O-caffeoyl-4-o-feruloylquinic acid 3,5-diC-4FQA C F C
64 3-O-feruloyl-4,5-di-o-caffeoylquinic acid 3F-4,5-diCQA F C C
65 3,4-di-O-caffeoyl-5-o-feruloylquinic acid 3,4-diC-5FQA C C F
66 3-O-caffeoyl-4,5-di-o-feruloylquinic acid 3C-4,5-diFQA C F F
67 3,4-di-O-feruloyl-5-o-caffeoylquinic acid 3,4-diF-5CQA F F C
68 3,4-di-O-caffeoyl-5-o-sinapoylquinic acid 3,4-diC-5SiQA C C Si
69 3-O-sinapoyl-4,5-di-o-caffeoylquinic acid 3Si-4,5-diCQA Si C C
29
OHOH
O
HO OHO
O
HO
HOOH
OHO
HO OHO
OOH
HO
HO OHO
O
OH
OH
O
OHOH
OHOH
O
HO OHO
O
HO
OOH
OHO
HO OHO
OOH
HO
HO OHO
O
OH
O
O
OHO
1 2 3
4 5 6
OHOH
O
HO OHO
O
HO
OHO
HO
HO OHO
OOH
HO
HO OHO
O
OHO
OH
7 8 9
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 1/8)
30
OHOH
O
HO OHO
O
O
OOH
OHO
HO OHO
OOH
HO
HO OHO
O
O
O
O
OO
OHOH
O
HO OHO
O
HO
OOH
OHO
HO OHO
OOH
HO
HO OHO
O
OH
O
O
OHO
10 11 12
13 14 15
O
O
O
OHO
O
HO OHO
O
HOO
OHO
HO OHO
O
OH
16 17
O
HO
OHOH
O
HO
HO OH
OO
HO
HO OHO
O
OH
O
OH
OH
OH OHO
O
HO OHO
O
HOO
O
OHO
18 19
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 2/8)
31
OOH
O
HO OHO
O
OH
O
HO
O O OO
HO
HO OHO
O
OH
O
O
OH
O
20 21
OHO
O
HO OHO
O
HOO
OHO
HO OHO
O
OH
22 23
O
OH
O
HO
OO
HO
HO OHO
O
OH
O
OH
OHO
O
HO OHO
O
HOO
O
OHOH
24 25
OHO
O
HO OHO
O
HOO
HO
OHO
26
OOH
O
HO OHO
O
OH
O
HO
O OH
27
OOH
O
HO OHO
O
OH
O
HO
HO O
28
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 3/8)
32
OO
HO
HO OHO
O
OH
O
OH
29
OHO
O
HO OHO
O
OO
O
OHOH
31
OO
HO
HO OHO
O
OH
O
OH
30
O
OH
OH
O
OOH
O
HO OHO
O
OH
O
O
O OH
32
OO
HO
HO OHO
O
O
O
OH
33
O
OH
OHO
O
HO OHO
O
OO
OHO
HO OHO
O
OH
34 35
O
OH
O
O
O
O
O O
OO
HO
HO OHO
O
O
O
OH
36
OHO
O
HO OHO
O
HOO
HO
OH
38
O
OOH
OO
HO OHO
O
HOO
OHOH
37
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 4/8)
33
OOH
O
HO OHO
O
OH
O
HO
OH
39
OOH
O
HO OHO
O
OH
O
HO
HO
40
OO
HO
HO OHO
O
OH
O
OH
41
OHO
O
HO OHO
O
HOO
OH
43
OH
42
OO
HO
HO OHO
O
OH
O
OH
OH
O
OOH
O
HO OHO
O
OH
O
HO
O
44 45
OO
HO
HO OHO
O
OH
O
OH
O
OHO
O
HO OHO
O
HOO
O
4746
OO
HO
HO OHO
O
O
O
OH
OO
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 5/8)
34
OOH
O
HO OHO
O
O
O
HO
O
48
OOH
O
HO OHO
O
OH
O
HO
O
49
OH
O
OHO
O
HO OHO
O
HOO
OH
50
OH
OHO
O
HO OHO
O
OO
OH
51
O
OHO
O
HO OHO
O
HOO
OH
52
OH
O
O
HO
OH
O
O
OOH
O
HO OHO
O
OH
O
HO
O
53 54
OO
HO
HO OHO
O
OH
O
OH
O
O
O
O O
OHO
O
HO OHO
O
HOO
OH
55 56
OO
HO
HO OHO
O
O
O
OH
OH
O
O
O O O
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 6/8)
35
OOH
O
HO OHO
O
OH
O
O
OH
57
OOH
O
HO OHO
O
OH
O
O
O
58
O
O
O
O
OHO
O
HO OHO
O
OO
OH
59 60
OO
HO
HO OHO
O
O
O
OH
O
O
O
O O
O
OO
O
HO OHO
O
OH
O
O
OH
61
OO
O
HO OHO
O
OH
O
HO
HO
62
OH
O
O O
OH OHO OH
OO
O
HO OHO
O
OH
O
HO
OH
63
OO
O
HO OHO
O
OH
O
HO
O
64
OH
HO
O O
OH OHO OH
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continues 7/8)
36
OO
O
HO OHO
O
OH
O
HO
O
65
OO
O
HO OHO
O
OH
O
HO
HO
66
OHO
O O
OH OHOH O
OO
O
HO OHO
O
OH
O
HO
OH
67
OO
O
HO OHO
O
OH
O
HO
HO
68
OO
O O
OH OHO OH
O
OO
O
HO OHO
O
OH
O
HO
OH
69
O
O
OHOH
O
Figure 4. Chemical structures of chlorogenic acids in Table 3. (Continued 8/8)
1.1.2.4. Analysis of chlorogenic acid
Analytical methods of chlorogenic acids
Although it is possible to analyze chlorogenic acids with a variety of different analytic tools
such as 13C-NMR, 1D26 and 2D-1H-NMR27, FT-IR28, FT-Raman29 and so on, the major
method to analyze chlorogenic acids is HPLC-MS. In recent studies, the comparison of those
spectroscopic techniques for analyzing chlorogenic acids in coffee30 is described.
37
The most common method for analyzing coffee chlorogenic acids was to compare synthesized
standard compounds using HPLC-UV/Vis spectra. Concentrations of the CQA isomers and
caffeine in weight percent from different originated coffee beans were reported by Michael P.
Purdon and David A. McCamey in 1987 as shown in Table 4.a) Concentrations and 4.b)
Information of the coffee beans.
Table 4.a) Concentrations of the CQA isomers and caffeine in weight percent
Table 4.b) Information of the coffee beans used in Table 4(a)
In 2003, different fragmentation patterns of three different caffeoyl quinic acids were
reported.5 CGA of regioisomeric MS fragmentations did not attract scientists’ interest because
regioisomers usually give the same fragments. However, this report showed clearly different
MS2 spectra, allowing unambiguous assignment of regiochemistry, so that HPLC-tandem MS
became the most important and popular tool for analyzing chlorogenic acids in coffee extracts.
38
Underestimated solubility differences for the real coffee consumption
Meanwhile, analyzing coffee started from the view of analyzing food, coffee beans were
extracted with 70 % methanol solutions using common established methods for the
extractions of food, vegetables and fruits. Although we consume all the coffee beans in some
brewing methods like Turkish coffee, we generally extract only 10 ~ 15 %, 15 ~ 25 % and 45
~ 60% of water soluble part of coffee for drinking with different methods of hand-brewing,
espresso and instant coffee, respectively. Therefore, there were always some differences
between research results and real consuming compounds. The differences were studied, 5-
CQA equivalent peak areas were measured and would be discussed in this thesis.
1.1.2.5. Properties of chlorogenic acid
Organoleptic properties
Sweet, acidic, bitter, body, aroma, after taste and green flavor properties are usually discussed
with regards to coffee taste. The chlorogenic acids compounds in Table 3 provide complex
taste perceived in coffee, together with sweetness in sugars, bitterness in caffeine, body in
lipid and the aroma providing volatile compounds that could be identified by GC-MS.31
Generally, the well-known compound suspected to be responsible for the bitterness is caffeine
and diketopiperazines (DKPs).32,33 Caffeine and DKPs have been studied for taste research as
a representative of bitter compounds in coffee, but a variety of chlorogenic acids plays key
role in bitter taste of coffee as well.
Including caffeine and DKPs, a variety of bitter compounds in coffee were identified.34 Not
only different regioisomers of caffeoylquinic acids, but also caffeic acid, di-caffeoylquinic
acids and caffeoylquinic acid lactones27,35 play substantial roles in the bitterness of coffee.
Although there are a lot of bitter-tasting molecules in coffee, those 10 chlorogenic acids; 3-O-
caffeoyl-γ-quinide (1), 4-O-caffeoyl-γ-quinide (2), 4-O-caffeoyl-muco-γ-quinide (3), 5-O-
caffeoyl-muco-γ-quinide (4), 5-O-caffeoyl-epi-δ-quinide (5), 3-O-feruloyl-γ-quinide (6), 4-O-
feruloyl-γ-quinide (7) 3,4-O-dicaffeoyl-γ-quinide (8), 4,5-O-dicaffeoyl-muco-γ-quinide (9)
and 3,5-O-dicaffeoyl-epi-δ-quinide (10); take important parts in including bitterness
perception as reported by the Hofmann group.36
As already mentioned in chapter 1-1-1. Coffee compounds, Robusta coffee has a higher
caffeine and chlorogenic acid content if compared to Arabica coffee. This is the reason why
39
sensory panels assign a higher degree of bitterness to Robusta beans. Since bitterness is a less
desired attribute in coffee, Arabica beans achieve higher prices in the market.
Biological activity
As discussed in chapter previously, chlorogenic acids may play an important role in biological
activity as an antioxidant. For a chemist, an antioxidant is a reducing agent. Since reactive
oxygen species concept based antioxidant theory was simple and well-explainable both
chemically and practically37,38, it has become so successful to be accepted by the public as a
general truth that antioxidant containing food are beneficial to health.
From the starting of coffee research, the biological activity research of chlorogenic acids only
focused on their antioxidantial properties. This flow was due to the view of chlorogenic acids
as an extension of polyphenols. Most polyphenol researches were focused on the usage of
polyphenols as an antioxidant, and chlorogenic acids share the common antioxidant reactivity
of polyphenols.
However, over the last decade, doubts of the antioxidant concept have emerged from various
concerns.21 Most dietary antioxidants show an even lower plasma concentration than ascorbic
acid or tocopherols and most polyphenols undergo extensive, mostly reductive gut floral
metabolism so that produced less efficient antioxidants.39 Although these evidences might
weaken the strong position of coffee as a healthy food, these also give the driving force of
new research not relying on only the simple explanation of antioxidant as a panacea.
As coffee market grows, several works on the bio-activity of coffee have been documented.
The coffee research which includes not only antioxidant properties, but also the antibacterial,
antimutagenic and antiviral activities could be correlated with chlorogenic acids.40,41 These
studies on how coffee components alleviate the effects of diseases are related to caffeine42 and
chlorogenic acids studies43.
Caffeine 1.1.3.
1.1.3.1. Etymology and classification
Caffeine was discovered by a German chemist, Friedlieb Ferdinand Runge in 1819 at the
request of Johann Wolfgang von Goethe.44 At this time, Runge suggested the name
‘Kaffebase’ that means a base that exist in coffee. However, a French chemist Pierre Jean
40
Robiquet made the first oral announcement about it at the meeting of the Pharmacy Society in
Paris with the name of ‘Caféine’, meaning amine (-ine) in coffee in French ‘Café’ and it
became generally known as caffeine.
Caffeine is a natural methylxanthine alkaloid, chemically related to adenine and guanine. The
chemical structure of caffeine is shown in Figure 5. And for the toxicity, it is classified as
‘generally recognized as safe’ by the US Food and Drug Administration.
N
N N
N
O
O
Figure 5. Chemical structure of caffeine
It is known that the individual sensitivity is genetically determined and leads to varying
individual caffeine sensitivity.45 The LD50 value of caffeine in human is an estimated value of
150 ~ 200 mg per kg, and it means almost two cups of coffee per kg. With continuous
overdose, more than 1 g per day per person can cause a physiological effect, so called
‘caffeinism’, including nervousness, irritability, restlessness, insomnia, headaches,
palpitations.46 Therefore, the daily recommended amount of caffeine in most countries is set
400 mg per day for normal adults and 300 mg per day for pregnant woman. An average cup of
coffee contains around 100 mg caffeine. Overdosage of coffee powder can, however, lead to
commercial cups as high as 323 mg.47
Coffee is the main source for caffeine in 21st century’s European and some Asian countries’
diet. It has both positive and negative health effects. By this reason, caffeine becomes the
most attractive compound in coffee and scientists produces a lot of results related to caffeine
in coffee.
1.1.3.2. The role of caffeine
It is speculated that plants are making caffeine to protect themselves against predator insects
and to prevent germination of nearby seeds. To humans, it is the most remarkable compound
in coffee that causes both positive and negative health effects. People started to use coffee as
stimulant drug and this stimulant effect is originated from the caffeine. From twentieth
41
century, people started to make energy drinks containing doses of caffeine of 6 ~ 242 mg / L
to use this stimulant effect of caffeine.48 However, by increasing coffee, chocolate and
caffeine-containing energy drink consumption, negative health effect of the caffeine came to
the fore.
Since caffeine is chemically one of the methylxanthine families, it interferes adenosine
receptors in our brain. By this effect, caffeine intake plays a stimulant role in our body,
together with inducing insomnia. It is reported that the caffeine intake reduces arteriosclerosis,
hepatic cirrhosis, depression, increasing concentration and agility, diuresis and preventing
cancer and obesity.49 On the other hand, it is also reported that the caffeine intake prevents
absorption of iron and calcium, triggers anxiety and hyperacidity, relaxes bronchial muscle
and makes addiction.50 These physiological effect and outburst of intake on caffeine-
containing food made caffeine as an attractive target compound for research.
The European Food Safety Authority reported ‘Scientific Opinion on the safety of caffeine’ in
2015.51 In this report, for adults in the general population, caffeine intakes up to 400 mg per
day do not raise safety concerns. For pregnant women, 200 mg per day do not raise safety
concern for the fetus. Before this report, most countries recommended 400 mg per day for
adults and 300 mg per day for pregnant or breast feeding women followed by the report from
Health Canada in 2012. The recommended 400 mg of caffeine is about 5 cups of coffee. The
LD50 value of caffeine is between 150 ~ 200 mg / kg, it means almost 100 cups of coffee for
an adult.
1.1.3.3. Coffee decaffeination
In 1903, Ludwig Roselius invented a coffee decaffeination method in Bremen, Germany.52 It
was the first chemical application used in coffee processing. Decaffeination was the first and
the last caffeine control method, which could reduce caffeine by more than 97%. At the
beginning, Roselius used benzene for caffeine extraction as a solvent, which is now banned as
carcinogenic chemical compound. Therefore, there have been a lot of different challenges for
extracting caffeine from green coffee beans using methylene chloride, chloroform, ethyl
acetate, supercritical carbon dioxide and water. These decaffeination methods will be
discussed in following chapter coffee technology, green coffee, decaffeination.
42
Coffee Technology 1.2.
From farm to table, coffee changes from the coffee cherry to green coffee to roasted coffee to
ground coffee to a final cup of coffee. In each step, many different methods are used and all
processes influence the coffee flavor to yield different grades of coffee. In this chapter, each
step would be described and the property changes by the processes would be discussed to use
for upgrading processes in health beneficial composition based control.
Coffee Cherry 1.2.1.
1.2.1.1. Classifications
Commercial coffee is composed of three species in the genus Coffea. Coffea Arabica (Arabica
coffee) accounts for about 60 % of the total production while Coffea canephora (Robusta
coffee) accounts for about 40 %. The third species is Coffea Liberica, which is now less than
1 % of production. (ICO Trade statistics tables, 2015) With the demand of low cost from the
global coffee companies, robusta coffee production has increased, but recently, with the
demand of specialty coffee, arabica coffee production is increasing again. (ICO Trade statics
tables, 2015 – 2016)
After the species, origin and grades are the main price-determining factors. Coffee is
cultivated only in the tropical region, referred to as beanbelt or coffeebelt, shown in Figure 6.
There are three main different criteria for determining coffee grades. Height of the cultivation,
size of the coffee bean and the number of defected beans are the three criteria. The application
of the standards depends on the country; Tanzania and Kenya by size, Ethiopia and Indonesia
by the number of defected beans, Guatemala by height of the cultivation, Hawaii by Size and
defected beans and Jamaica by size and height. Together with the main three criteria, there are
some special standards such as animal processed (Indonesia Luwak), wind fermented (India
Monsoon) or Peaberry (Kenya PB).
43
Figure 6. Coffee Belt / Bean Belt (Figure from Wikipedia)
1.2.1.2. Processes
After harvesting the coffee cherry, the next step includes production of green coffee bean by
so called ‘Coffee Process’. The aim of these processes is to remove the flesh of the cherry to
yield the seed referred to as coffee bean. In general, there are three different processes in this
step; wet, dry and semi-dry processes.
Dry process is a predominant method because it was the oldest method and do not need high
cost of equipment. Two third of world coffee production is conducted by the dry process.
Furthermore, after 2010, coffee with a variety of fermentation methods become popular. The
dry process was preferred and could be easily adapted for fermentation. Almost all the robusta
coffee processes and a major portion of arabica coffee in Brazil, Ethiopia, Haiti, Paraguay,
India and Ecuador are dry process. In the dry process, coffee cherries after harvested are
placed in the sun on tables or thin layers on the ground or patios. It takes some days to max. 4
weeks to dry the cherries and they are hulled to make green coffee beans.
Wet process have used in the countries which are not suitable for dry process because of
frequent rainfalls; Vietnam, India and Indonesia. This process needs cost of equipment, but
gives a cleaner product. Therefore, washed-bean or full-washed bean had started to become
popular especially in the highly-evaluated origin coffee regions. However, with the fame of
the fermentation coffee, dry process became again preferred. Although wet process is evolved
44
into ferment-and wash method, still it is used in limitative origins. In the wet process, the fruit
covering the beans is removed before they are dried. The harvested cherries are immersed into
water to sort out the unripe fruit by floating. The sunk cherries are flown by the water flow to
be hulled with mechanical demucilaging and dried to obtain green coffee beans.
Semi-dry process, also called wet-hulled, semi-washed or pulped natural, is a hybrid process
used in several farms of Indonesia and Brazil. This method removes the outer skins by dry-
process and removes mucilage by wet-process.
There are research results investigating different methods, but no published article directly
compares coffee beans of identical origin with different processes. This is because all the
processes are selected by its facilities, but no coffee farm equipped different process facilities
together. Therefore, there are still under investigated research topics, which suffer from
difficulties in sample acquisition.
Green Coffee 1.2.2.
1.2.2.1. Classifications
There are some special coffee beans processed prior to yielding the green coffee bean. These
special treatments are mostly used for improving medicinal properties or are based on
traditional believes, all associated with chemical changes in coffee composition.
Decaffeination, fermentation and animal processed coffees are the results of such special
treatments.
Normal coffee
This is the standard of comparison. Normal coffee is processed by one of the coffee processes
mentioned in 1-2-1-2. Processes without further treatment. By using decaffeination or
fermentation, the normal coffee can change into special coffee.
Special coffee
Special coffee could be obtained by applying a special method both before and after the
normal coffee is obtained. By using the process-completed normal green coffee bean,
decaffeination or fermentation could be applied to change the chemical compositions for
special use. Besides, animal processed coffees could be obtained simply by giving the coffee
45
cherry to the animals. Presently, the animal processed coffee without going through the
normal green coffee process is the most expensive coffee in the world.
Decaffeination
The first chemical application and the first caffeine control method in coffee is decaffeination.
It had been more than 80 years from the first isolation of caffeine from coffee to the first trial
of removing caffeine from coffee. Ludwig Roselius obtained decaffeinated coffee by
steaming or using benzene to extract caffeine from coffee in 1903. This method is patented in
1906 and he founded Kaffee HAG in Bremen to produce the decaffeinated coffee marketing it
with a slogan of stomach-friendly coffee. However, the method is stopped now because it was
proven that the benzene is carcinogen. Therefore, a variety of methods are developed.
Methylene chloride was the first substitute for benzene because of its good extracting
character and low boiling point for removing after extracting caffeine without known harmful
effect on health. However, methylene chloride is now on the list of International Agency for
Research on Cancer (IARC) in Group 2A, which means it is probable carcinogen. For this
reason, many coffee companies have banned this method, but there are still some global
companies using methylene chloride method.53
Also, ethyl acetate was used as the substitute for benzene and it is still accepted because this
solvent is relatively safe (LD50 for rat is 5620 mg / kg) and present as a natural constituent in
fruits, so this method is called as natural process.54 However, this method also needs chemical
solvent and has no other advantages compared to other methods but using the name ‘natural’.
Therefore, the method using ethyl acetate could not occupy the major market of
decaffeination methods.
Supercritical fluid extraction of caffeine is one of the most widely used processes, which use
supercritical CO2 as the extraction solvent.55 This method was invented by Kurt Zosel from
the Max Planck Institute for Coal Research in Mülheim an der Ruhr in 1967 and patented
three years after the development. The only disadvantage of this method is the high cost of
equipment and maintenance along with co-extraction of bioactives in particular CQA-lactones.
The last decaffeination is so called Swiss Water Process.56 This method uses water as
extracting solvent and this is why this method could become popular in relatively short time.
46
First, coffees extracted with water, filtered with charcoal to remove caffeine, and the green
coffee extract without caffeine are used to extract only caffeine from other batch of coffees
and recycled the method from charcoal filtering.
Coffee fermentation
Coffee fermentation is an old terminology, which is used for the short storage period during
the coffee process. No matter which process among the wet, semi-dry or dry process was
selected, they all have a small period for the coffee beans to be exposed with the residues of
mucilage. During this period coffee beans are exposed to microorganisms. After 2000,
specialty coffee becomes a trend, people started to improve the fermentation method for
ameliorating its flavor. A variety of controls have been done with temperature, humidity, sun
dry or additive controls, not only for the traditional coffee fermentation periods but also after
green bean or roasted bean steps. Furthermore, there were natural fermentation processes by
using seasonal wind or animal organs. Especially animal process would be dealt with
following chapter because it has the character not only with the fermentation, but also animal
selection.
Fermentation on coffee process
As specialty coffee becomes more common, the coffee industry tends towards developing
new methods for making coffee beans to be more special. Fermentation control is one of the
efforts conducted from the farm in this regard. Coffee fermentation is already an established
step during coffee process, though there are differences in their detailed methods. From wet to
dry process, the fermentation period is increased so that the dry process is easily adapted by
the fermentation method. Coffee producers started to control temperature, humidity, sunlight
exposure and the use of some bacteria on the coffee beans. The controlled fermentation
method has been improved and finally in 2015, these fermentation controlled coffee bean
became the main source of 13th Korea Barista Championship champion. These enhanced
fermentation coffee is known to improve their fruit flavor in coffee beans.
Fermentation after green coffee bean
In 2011, a research on the use of steaming and drying of coffee was published as a patent
which indicates that this special treatment of green coffee bean changed green flavor to fruit
flavor, as well as the noble processed coffee bean extracted made it aroma richer than the
reference-roasted coffee bean extracted by sensory evaluation.57 Although this method is hard
47
to include as fermented bean because it did not intended to use microorganisms, it did not
prevent from the activities of their inherent microorganisms. However, there is no in-depth
investigation for this novel method processed coffee beans because it is a very recent try and
succeeded in a limited scale of production.
Fermentation by seasonal wind
There is a special yellow green bean in India. The coffee is fermented green coffee beans by
the seasonal wind, ‘Monsoon’. During the times of British Raj, Indian coffee beans were
transported to Europe by sea for three to four months. The humidity and monsoon wind
caused the green coffee to producing a pale yellow color. However, the transportation ways
have been improved; it reduced the exposed time for the green coffee beans to the monsoon
wind. So the beans do not ripen enough as they were, and people started some artificial
fermentation for the coffee beans to replicate the old method, which can give a special soft
coffee flavor. However, since it is only produced in the Malabar coast region, it did not
receive attention from scientists. Therefore, there is very few published research results about
monsooned coffee.58
Animal processing
The most expensive coffee in the world was Kopi Luwak from Indonesia and changed to
Black Ivory from Thailand. The two different coffees have one thing in common. They are
animal processed coffees. Kopi Luwak is the coffee eaten by civets, and Black Ivory is by
elephants. Animals feed on coffee cherries and undigested coffee beans are gathered from the
animals’ excrement, washed to become the commercial green coffee bean product.
Kopi Luwak has two advantages over normal coffee with fermentation and animal selection.
By passing through the civet’s organs, coffee bean is exposed to protease enzymes of the civet,
to make shorter peptides and more free amino acids. In 2013, research on short chain fatty
acids compared Kopi Luwak to normal coffee and showed that Kopi Luwak has higher malic
acid and citric acid than normal coffee.59 Also there is so-called animal selection, which
means that animals have sixth-sense to figure out which fruit is well-ripened to eat. Therefore,
the coffee beans in the excrement of animals are qualified coffee beans by the animal
selection.
48
Black Ivory Coffee claims its advantage over Kopi Luwak that elephants are herbivores in
contrast with civets, which are omnivores. They both use their digestive enzymes to break
down the coffee’s protein, but herbivores utilize fermentation to help break down cellulose.
So the Black Ivory can get smoother taste.
Until now, Black Ivory Coffee was not analyzed and published by scientists because of its
high price and low production. However, to compare different animal processed coffees such
as civet from Indonesia and Philippines, elephant from Thailand and weasel from Vietnam
would be an interesting further research.
1.2.2.2. Processes
Excepting only very few special treatment on green coffee bean, the process on green coffee
is roasting. Including 2-furfuryl thiol which has high odor activity value in coffee, the
compounds, which contribute to ‘coffee flavor’ in technical term, are made by roasting.60,61
That is to say that roasting is an essential process for coffee flavor. Concentration and
extraction yield of major coffee aroma compounds and their chemical structures are shown in
Table 5 and Figure 7, respectively. The detailed coffee volatiles with threshold concentration
would be discussed later in 1.3. Coffee Sensory Perception.
Table 5. Concentrations of the odorants in coffee powder and coffee brew, and extraction
yields of the compounds60
Nr. Compound Concentration Extraction
yield Powder (µg/kg) Brew (µg/l)
1 Acetaldehyde 120,000 4,700 73
2 Methylpropanal 24,000 760 59
3 2-Methylbutanal 26,000 870 62
4 3-Methylbutanal 17,000 570 62
5 2,3-Buranedione 49,000 2,100 79
6 2,3-Pentanedione 35,000 1,600 85
7 2-Ethyl-3,5-dimethylpyrazine 400 17 79
8 2-Ethenyl-3,5-dimethylpyrazine 53 1.0 35
9 2,3-Diethyl-5-methylpyrazine 100 3.6 67
10 2-Ethenyl-3-ethyl-5-methylpyrazine 15 0.2 25
49
11 3-Isobutyl-2-methoxypyrazine 120 1.5 23
12 4-Hydroxy-2,5-dimethyl-3(2H)-furanone 140,000 7,200 95
13 2-Ethyl-4-hydroxy-5-methyl-3(2H)-furanone 16,000 800 93
14 3-Hydroxy-4,5-dimethyl-2(5H)-furanone 1,900 80 78
15 (E)-β-Damascenone 260 1.6 11
16 Guaiacol 2,400 120 65
17 4-Ethylguaiacol 1,800 48 49
18 4-Vinylguaiacol 45,000 740 30
19 Vanillin 4,100 210 95
20 2-Furfurylthiol 1,700 17 19
21 Methional 250 10 74
22 3-Mercapto-3-methylbutyl formate 130 5.7 81
23 2-Methyl-3-furanthiol 60 1.1 34
24 3-Methyl-2-buten-1-thiol 13 0.6 85
25 Methanethiol 4,400 170 72
O OO
O OO
1 2 3 4 5 O
ON
N
N
N
N
N
6 7 8 9
N
N
10
N
N
O
11
OO
HO
12
OO
OH
O
O
HO
13 14 15
O
O
HO
16
O
HO
17
O
HO
18
HO
OO
19
HSO
20
50
O S
21
O O SH
22
O
HS
23
SH
24 25
SH
Figure 7. Chemical structures of major coffee aroma compounds in Table 5.
After reporting the significant decrement of chlorogenic acids content during roasting stage,62
people started to find a better way to consume more chlorogenic acids in the source of coffee.
There was a green coffee extract study in 2011 which showed that regular drinking of green
coffee extract can reduce weight and it was suggested that the reason is possibly related to the
richness of the chlorogenic acids in green coffee extract.63 However, participants in the
studies were instructed to restrict their diet and increase their exercise in addition to taking the
supplement. In 2012, Dr. Oz Show broadcasted this and green coffee extract became famous
as a weight loss supplement until Federal Trade Commission fined the sponsor company in
2014.64
Although the magical weight loss by the green coffee extract was revealed as a happening, the
fact that chlorogenic acids undergo decomposition during roasting coffee became a common
knowledge despite of its short publication time. And the fact that chlorogenic acids can help
losing weight was proven as true through blocking sugar transfer by chlorogenic acids.65 So
the green coffee extract market formed by this happening is still growing.
Not only just removing the roasting step from the coffee process, but also trying to substitute
the roasting step was conducted. It is adapting Korean traditional red-ginseng process into
green coffee bean to make fermented coffee bean. The process and chemical analysis would
be described in this thesis.
Roasted Coffee 1.2.3.
In roasting process, direct, indirect and semi-direct fire roastings were used. Recently,
electricity only, without fire roaster appears with the advantage of reproducibility, represented
by Stronghold. Although the type of roasting process plays an important role in coffee flavor,
it is hard to examine because the cost of equipment is high for roasters. Likewise, the different
classifications of roasting degree and processes after roasting would be discussed in this
research.
51
1.2.3.1. Classifications
Excepting some special methods such as steamented coffee and green coffee extract, coffee
roasting is an essential process prior to forming basis of coffee brew. Coffee needs different
roasting conditions depending on their species, origins and methods of brewing. Recently, the
relations of various different roasting conditions and tastes are started to be researched.
However, traditional classifications of roasting degree were determined by both roasting time
and temperature, changed by 16 level systems by Specialty Coffee Association America
(SCAA) standard, and finally expressed by the results of colorimetric measurements.
Traditional classifications
The basic categories of traditional roasting degree are green, light, medium and dark roast.
Among them, cinnamon, city, full city, French and Italian roast are used, without precise
categorization. Generally, the coffee industry used more common names and scientists used
more degree names. These traditional classifications are still used both industry and academia
as shown in Figure 8, the research of coffee roasting in 2009,62 but step by step Colorimeter
classification substitutes for the traditional classifications. According to the traditional
classifications, cinnamon roast is the period for 1st crack and full city roast is the period for
2nd crack. Cinnamon roast to city roast are used for filter coffee brewing and higher than full
city roasting are used for espresso brewing.
52
Figure 8. Total chlorogenic acids found in various green coffee beans and in various roasted
coffees treated under different conditions62
SCAA 16 step classification
SCAA tried to change the inconsistent traditional roasting degrees because all roasters have
their own roasting process in their separate ways of roasting time and temperatures. SCAA
selected not to take account of the process, but to focus only for the resulting color. They
divided coffee color into 16 levels from green coffee beans to fully-roasted coffee beans and
published as the color-standard for comparing.
53
Colorimeter classification
There is limitation in print-out color grading in human perception. The textile of coffee bean,
the type of papers for standard color printout and even the feeling of the roaster can affect for
the color perception. To correct this, people started to use colorimetric instruments. The first
accepted was Agtron analytic instrument, which has selected by SCAA, for that reason, its
analytical value symbolizes the coffee roasting degree. After coffee industry accepts the
machine value, HunterLab Coffee Color Index (HCCI) value started to show the degree of
roasted coffee. This HCCI value measure the absorbance of 640 nm wavelength, which is
orange-red, and calculate the value with the slope of the UV-Visible spectra. Therefore, this
HCCI value can present not only the absolute darkness of the coffee bean, but also color
changes. However, the relation between the two different machine values and roasting
condition was not studied. In contrast most other foods are subjected to colorimetric
measurements using the L, a and b scale, coffee colorimetric measurements are using only
limited range of wavelengths because the color of coffee beans are limited in yellow green to
dark brown during the roasting period. In general, roasted coffee beans have L, a and b values
from 30, 14, 23 on light roasting to 12, 4 and 3 on dark roasting, respectively.
1.2.3.2. Processes
Storage
Coffee is scientifically known as a shelf-stable product after roasting that it does not spoil by
enzymatic or microbial processes. However, in sensory tests, a drastic taste change in
rancidity is detected, which is triggered from the roasting. These taste changes are originated
from either the chemical reactions by the conditions of temperature, humidity and oxygen or
the physical diffusion.
The flavor perception is the combination of taste and smell. And in specialty coffee, smell
plays more roles. Therefore, though the major compounds of roasted coffee do not change,
only small part of the composition changes can make the flavor totally different especially if
the small part is volatile. During the roasting process, Strecker degradation makes CO2, and
this gives pressure to break the bean shell for the other volatile compounds to be released
from the roasted coffee beans. And lipid oxidation produces odor active aldehydes,
accelerated by temperature, humidity and oxygen atmosphere.
54
For preventing those rancidity sources, coffee roasters use vacuum, nitrogen-charged or one-
way valve packages for coffee storage.
Grinding
After roasting, grinding should be followed for extraction yield before brewing. It is always
recommended that the grinding should be right before coffee brewing because it increases
surface area to accelerate staling.
For grinding coffee, two types of grinders are used; burr grinder and blade grinder. Home
brewing people and most academic scientists usually use blade grinder for its low-cost, but
this blade grinder cannot guarantee the evenness of the coffee powder and gives too much
heat to the coffee to induce the loss of a lot of volatile compounds. However, there is no study
for the composition changes by the grinders because it is known as not the critical point for
the flavor changes than roasting, brewing or even the storages.
Most of the particle size research is conducted by the grinder sellers for showing their quality
of evenness as a way of public relation. The particle sizes of the ground coffee beans are
decided by the way of brewing. Typical particle sizes for their brewing methods are roughly
from 250 µm, 500 µm, 1,000 µm and 1,500 µm for Espresso, Hand-brewing, Cold-brew and
Frenchpress, respectively.
The main issues for the grinding step are size equalization and heat control. For both reasons,
burr grinder is preferred over the blade grinder and the burr grinder is divided into two
categories; flat burr and conical burr grinders. Flat burr has advantage in size equalization and
conical burr has advantage in heat control. Therefore, for most café or companies, conical
burr grinder is recommended because they grind plenty of coffee and it makes much heat, so
heat control is more important for those. However, for home users, they can give enough time
for cooling to the grinder, so they can accept temperature increasing for more even size.
Ground Coffee 1.2.4.
1.2.4.1. Classifications
The ground coffee should have the shortest period among the whole coffee process. By sharp
growing of the surface area, the ground coffee becomes rancid quickly. Therefore, this step
55
does not have subordinate classifications, but to pass through the final cup of coffee by
brewing.
1.2.4.2. Processes (Brewing / Extraction)
The extracting process applied to ground coffee is called brewing. There are a lot of different
coffee brewing techniques; hot and cold brewing, espresso brewing, machine-brewing, mocha
pot, syphon, Turkish brewing and hand-brewing.
Machine-brewing and espresso brewing are now the most popular techniques for consuming
coffee. Hand-brewing is one of the oldest techniques, but started to be re-illuminated by the
increasing market of specialty coffee. Cold brew coffee is the most recent style and the unique
coffee brewing style because it does not boil water to brew coffee. And it becomes popular
with its unique flavor. However, since it is too new to be identified by scientist, there was no
scientific publication for this remarkably growing brewing style.
Because changing brewing method is the only allowed variation for the general coffee
consumers, a lot of different brewing methods were developed for specialty coffee society to
control the flavor of coffee. On the other hand, because brewing cannot give enough energy
for the components to have chemical reactions, coffee brewing methods were not attractive
enough for the scientific researchers. However, with the coffee market grows, results started
to come out that can fulfill both the curiosity for the specialty coffee society and scientific
interpretations for the academia.66
Extracted Coffee 1.2.5.
After brewing coffee, there seems no other process but to drink. However, there could be one
more step to make the extracted coffee into solid powder. The product is called ‘instant coffee’
that the consumers can make a cup of coffee in a short time simply by pouring hot water
because the solid is already extracted once. This method was first developed and patented in
1881, by Alphonse Allais, France, first started to sell with patent in 1890, by David Strang,
New Zealand, first introduced to public in 1901, by Satori Kato, Japan, and first successfully
commercialized in 1910, by George Washington, Belgium. Using instant coffee process or not,
we can only categorize the coffee as hot or cold and pure or blended. Since there are too many
factors for composition change exist in blending with milk, it is hard to discuss in
scientifically. Therefore, most of the research on extracted coffee differences focused on taste
56
panel survey. Until now, it is accepted as truth that more bitter coffee harmonizes with milk
better than less bitter coffee. Thus, acidic and sweet coffee is used for specialty coffee
consumers and bitter and rich-body coffee is used for blended coffee consumers.
There is an emerging field of research about furan moiety. It is known that all the cooking
food makes furan moiety and those compounds have potentially harmful for our health. By
increasing coffee consumption, coffee is also regarded as an important source of furan family.
Therefore, to reduce the furans would be an interesting issue for the further research.
Coffee Sensory Perception 1.3.
Coffee flavor perception is the combination of taste by the solid contents in the coffee liquid
and smell by the volatile compounds. In general, when coffee consumers drink coffee, they
can perceive the aroma, bitterness, body, sour, sweet and after taste properties. These
components of coffee taste are provided through either by taste, smell or both. In this chapter,
compounds comprising coffee taste and smell are reviewed and noble development of coffee
taste classifications are embraced with Coffee Taster’s Flavor Wheel and World Coffee
Research Sensory Lexicon.
Coffee Aroma 1.3.1.
The more coffee market advances, the more the coffee drinkers search for a better coffee taste.
However, our taste perceptive system can only detect 5 different tastes with taste buds; sweet,
sour, bitter, salty and umami. To achieve descriptions of various coffee tastes, the smell of the
coffee is the basic essential, called coffee aroma.
For coffee aroma research, GC-MS is usually used for analyzing volatile compounds in coffee.
In 2018, Nicola Caporaso et al. published the analysis of 50 aroma active volatile compounds
in coffee samples by SPME (Solid phase microextraction)-GC-MS which is shown in Table
6.67
57
Table 6. Identification of volatile compounds in roasted arabica and robusta coffee samples
analyzed by SPME-GC-MS at a single bean level.
The ultimate goal of coffee aroma research is to know what kind of coffee aroma is made by
which coffee volatile compounds. This research is challenging because perception of flavor,
the odor activity values and the perception threshold concentrations of the compounds are all
different individually. Therefore, almost non-detectable composition change by scientific
instruments can play a significant role in coffee taste. The differences of threshold
concentrations of key food odorants are shown in Table 7.68
58
Table 7. Odorant qualities and threshold concentrations of selected key food odorants
Key Food Odorants Odor quality Threshold Conc. (µg/kg water) ethanol alcoholic 990,000
2-methyl-1-propanol malty 19,000 Acetic acid vinegar-like 5,600 1-hexanol green, grassy 590
(E)-2-hexenal green, apple-like 110 2-phenylethanol flowery, wine-like 18
(R)-limonene citrus-like 13 2-methoxy-4-vinylphenol smoky 5 3-hydroxy-4,5-dimethyl-
2(5H)-furanone seasoning-like 2
butan-2,3-dione butter-like 1 3-methylbutanal malty 0.5
3-(methylthio)propanal cooked potato-like 0.4 (E)-2-hexenal green, grassy 0.1
2-acetyl-1-pyrroline popcorn-like 0.05 (E,E)-2,4-decadienal fatty, French fries-like 0.03
wine lactone coconut-like 0.02 (E)-β-damascenone cooked apple-like 0.01
(E,Z)-2,6-nonadienal cucumber-like 0.005 (Z)-1,5-octadien-3-one geranium-like 0.0003 1-p-menthene-8-thiol grapefruit-like 0.0002 2-methyl-3-furanthiol meaty, bouillon-like 0.00003
Unlike other food analytical studies, coffee aroma specifically depends most importantly on
its process; roasting. The concentrations of the odorants in coffee have already been shown in
Table 5, in the roasting chapter. Since the aroma of coffee changes significantly pursuant to
the roasting degree, customized analytical equipment is needed. Therefore, the kinetic studies
of volatile compounds produced during coffee roasting have been conducted using the
roaster-connected Mass spectrometry customized only for the coffee research.69
Coffee Taste 1.3.2.
Although coffee aroma plays more and more roles in specialty coffee trend than coffee taste,
non-volatile coffee compounds take charge of the key perception in coffee – bitterness. The
perception of bitterness is assumed to have developed for protecting human body from toxic
compounds. Therefore, it was hypothesized that human perception has evolved regarding
bitterness as an unpleasant taste. However, this hypothesis is not sufficiently supported
59
because only a little more than half of the bitter compounds are known as toxic to human.70
Furthermore, coffee itself represents bitter beverage, and is known as a health beneficial
beverage. The relationship between bitterness and toxicity needs to be further investigated.
In coffee, caffeine, chlorogenic acids and special sequences of amino acids with
diketopiperazines are the main bitter taste compounds. Among them, the most popular bitter
compound to coffee consumers is caffeine. However, only 1 % of the extracted coffee solid
content is caffeine and its actual role on bitterness in a cup of coffee is only about 10 %.33
This is easily identified that decaffeinated coffee, which has caffeine less than 3 % of normal
coffee, is still bitter.
When the roasting degree is too high, less than 2 % of chlorogenic acids would be remained,
and the bitterness would stem from the final products of Maillard reaction.71 However, normal
consuming coffee still conserve the main bitter taste compounds in coffee. Among the
different bitterness raising compounds, chlorogenic acids play the largest part in coffee
bitterness. About 80 % of bitterness in a cup of decaffeinated coffee originates from 10 major
chlorogenic acids among the total coffee compounds.36
Special sequences of peptides and diketopiperazines are responsible for the remained
bitterness of coffee. The bitterness of Leucine-containing peptides was selected to compare
the differences of threshold concentration of bitterness by the sequences of amino acids.72
Proline-based diketopiperazines are suspected to be important sources of coffee bitterness.33
Tools for Coffee Sensory Analysis 1.3.3.
For the sensory analysis, panels are trained to perceive the coffee taste according to an official
method. The most widely used training materials in coffee analysis are World Coffee
Research Sensory Lexicon by Kansas University Coffee Research Center for the terminology
and The Coffee Taster’s Flavor Wheel by SCA for actual training.
1.3.3.1. World Coffee Research Sensory Lexicon
World Coffee Research started in 2012. Their research mainly focused on applying advanced
agricultural science for coffee and now expended to all the area of coffee research including
sensory analysis.
60
While working with coffee farms all around the world, they found the need for the unification
on coffee taste expression, resulting in the first Sensory Lexicon published in 2016. The
organization is a nonprofit organization. For the efficient terminology unification, the Sensory
Lexicon is available free to all for downloading and printing of personal-use copies. The most
recent publication version is 2.0, published in October, 2017.73
The Sensory Lexicon includes attribute name, definition, references, intensity score and
preparation instructions. By taste the samples prepared with the instruction, training panels
can have reproducible taste terminology and unified intensity scores for different tastes in
coffee.
1.3.3.2. The Coffee Taster’s Flavor Wheel
The Coffee Taster’s Flavor Wheel was originally published in 1995 for training sensory
panels by specialty coffee association of America (SCAA). It has been the industry standard
for over two decades.
While World Coffee Research started to make its first Sensory Lexicon, SCAA was also one
of the advisory group members. After WCR finished publishing the Sensory Lexicon
successfully, it was essential to revise the SCAA Flavor Wheel to be compatible with the
Lexicon as following the rationale behind the Lexicon publication. Therefore, the new Coffee
Taster’s Flavor Wheel was revised in 2016.
The panel training can start from the center to outside in order to obtain easy relation between
the attribute and their taste. It is also recommended to use only the core attributes (roasted,
spices, nutty/cocoa, sweet, floral, fruity, sour/fermented, green/vegetable and others) when the
panels are not well-trained. Panels can use higher attributes depending on their perception of
the relation between the attribute and their taste. The colors of the attributes were selected
with a careful consideration in order to match up the senses of sight with the senses of taste
and smell. There are three different sized gaps between the attributes, and if there is a gap,
which means the tasters thought of them as slightly less closely related.
61
References
1. Zhao Z, Moghadasian MH. Chemistry, natural sources, dietary intake and pharmacokinetic
properties of ferulic acid: A review. Food Chem. 2008;109(4):691-702.
2. Ascherio A, Zhang SM, Hernán MA, et al. Prospective study of caffeine consumption and
risk of parkinson's disease in men and women. Ann Neurol. 2001;50(1):56-63.
3. Cadden ISH, Partovi N, Yoshida EM. Review article: Possible beneficial effects of coffee
on liver disease and function. Aliment Pharmacol Ther. 2007;26(1):1-7.
4. Fisher SM, McMurray RG, Berry M, Mar MH, Forsythe WA. Influence of caffeine on
exercise performance in habitual caffeine users. Int J Sports Med. 1986;7(5):276-280.
5. Clifford MN, Johnston KL, Knight S, Kuhnert N. Hierarchical scheme for LC-MSn
identification of chlorogenic acids. J Agric Food Chem. 2003;51(10):2900-2911.
6. Dulloo AG, Duret C, Rohrer D, et al. Efficacy of a green tea extract rich in catechin
polyphenols and caffeine in increasing 24-h energy expenditure and fat oxidation in humans.
Am J Clin Nutr. 1999;70(6):1040-1045.
7. Reissig CJ, Strain EC, Griffiths RR. Caffeinated energy drinks-A growing problem. Drug
Alcohol Depend. 2009;99(1-3):1-10.
8. Macrae R, Robinson RK, Sadler MJ. Encyclopedia of food science, technology and
nutrition. Academic Press; 1993.
9. Kuhnert N, Jaiswal R, Matei MF, Sovdat T, Deshpande S. How to distinguish between
feruloyl quinic acids and isoferuloyl quinic acids by liquid chromatography/tandem mass
spectrometry. Rapid Communications in Mass Spectrometry. 2010;24(11):1575-1582.
62
10. Kuhnert N, Jaiswal R, Eravuchira P, El-Abassy RM, von der Kammer B, Materny A.
Scope and limitations of principal component analysis of high resolution LC-TOF-MS data:
The analysis of the chlorogenic acid fraction in green coffee beans as a case study. Analytical
Methods. 2011;3(1):144-155.
11. Matei MF, Jaiswal R, Kuhnert N. Investigating the chemical changes of chlorogenic acids
during coffee brewing: Conjugate addition of water to the olefinic moiety of chlorogenic acids
and their quinides. J Agric Food Chem. 2012;60(49):12105-12115.
12. Sigma-Aldrich. Chlorogenic acid.
http://www.sigmaaldrich.com/catalog/product/fluka/00500590. Updated 2015.
13. Wikipedia TFE. Chlorogenic acid. https://en.wikipedia.org/wiki/Chlorogenic_acid.
Updated 2015.
14. Yabuta G, Koizumi Y, Namiki K, Hida M, Namiki M. Structure of green pigment formed
by the reaction of caffeic acid esters (or chlorogenic acid) with a primary amino compound.
Biosci Biotechnol Biochem. 2001;65(10):2121-2130.
15. Quideau S, Deffieux D, Douat-Casassus C, Pouysegu L. Plant polyphenols: Chemical
properties, biological activities, and synthesis. Angewandte Chemie-International Edition.
2011;50(3):586-621.
16. Renaud S, Delorgeril M. Wine, alcohol, platelets, and the french paradox for coronary
heart-disease. Lancet. 1992;339(8808):1523-1526.
17. Cho ES, Jang YJ, Hwang MK, Kang NJ, Lee KW, Lee HJ. Attenuation of oxidative
neuronal cell death by coffee phenolic phytochemicals. Mutat Res -Fundam Mol Mech Mutag.
2009;661(1-2):18-24.
63
18. Haslam E. Practical polyphenolics - from structure to molecular recognition and
physiological action. Cambridge University Press; 2005.
19. Ichihashi M, Ueda M, Budiyanto A, et al. UV-induced skin damage. Toxicology.
2003;189(1-2):21-39.
20. Valko M, Rhodes CJ, Moncol J, Izakovic M, Mazur M. Free radicals, metals and
antioxidants in oxidative stress-induced cancer. Chem -Biol Interact. 2006;160(1):1-40.
21. Kuhnert N. One size does not fit all-bacterial cell death by antibiotics cannot be explained
by the action of reactive oxygen species. Angewandte Chemie-International Edition.
2013;52(42):10946-10948.
22. Dawidowicz AL, Typek R. Transformation of 5-O-caffeoylquinic acid in blueberries
during high-temperature processing. J Agric Food Chem. 2014;62(45):10889-10895.
23. Jaiswal R, Patras MA, Eravuchira PJ, Kuhnert N. Profile and characterization of the
chlorogenic acids in green robusta coffee beans by LC-MSn: Identification of seven new
classes of compounds. J Agric Food Chem. 2010;58(15):8722-8737.
24. Deshpande S, Jaiswal R, Matei MF, Kuhnert N. Investigation of acyl migration in mono-
and dicaffeoylquinic acids under aqueous basic, aqueous acidic, and dry roasting conditions. J
Agric Food Chem. 2014;62(37):9160-9170.
25. Jaiswal R, Matei MF, Golon A, Witt M, Kuhnert N. Understanding the fate of chlorogenic
acids in coffee roasting using mass spectrometry based targeted and non-targeted analytical
strategies. Food & Function. 2012;3(9):976-984.
64
26. Iwai K, Kishimoto N, Kakino Y, Mochida K, Fujita T. In vitro antioxidative effects and
tyrosinase inhibitory activities of seven hydroxycinnamoyl derivatives in green coffee beans.
J Agric Food Chem. 2004;52(15):4893-4898.
27. Kaiser N, Birkholz D, Colomban S, Navarini L, Engelhardt UH. A new method for the
preparative isolation of chlorogenic acid lactones from coffee and model roasts of 5-
caffeoylquinic acid. J Agric Food Chem. 2013;61(28):6937-6941.
28. Pascoa RNMJ, Magalhaes LM, Lopes JA. FT-NIR spectroscopy as a tool for valorization
of spent coffee grounds: Application to assessment of antioxidant properties. Food Res Int.
2013;51(2):579-586.
29. Rubayiza A, Meurens M. Chemical discrimination of arabica and robusta coffees by
fourier transform raman spectroscopy. J Agric Food Chem. 2005;53(12):4654-4659.
30. Deshpande S, El-Abassy RM, Jaiswal R, et al. Which spectroscopic technique allows the
best differentiation of coffee varieties: Comparing principal component analysis using data
derived from CD-, NMR- and IR-spectroscopies and LC-MS in the analysis of the
chlorogenic acid fraction in green coffee beans. Analytical Methods. 2014;6(10):3268-3276.
31. Yeretzian C, Jordan A, Badoud R, Lindinger W. From the green bean to the cup of coffee:
Investigating coffee roasting by on-line monitoring of volatiles. Eur Food Res Technol.
2002;214(2):92-104.
32. Ginz M, Engelhardt UH. Identification of new diketopiperazines in roasted coffee. Eur
Food Res Technol. 2001;213(1):8-11.
33. Ginz M, Engelhardt UH. Identification of proline-based diketopiperazines in roasted
coffee. J Agric Food Chem. 2000;48(8):3528-3532.
65
34. Frank O, Blumberg S, Kunert C, Zehentbauer G, Hofmann T. Structure determination and
sensory analysis of bitter-tasting 4-vinylcatechol oligomers and their identification in roasted
coffee by means of LC-MS/MS. J Agric Food Chem. 2007;55(5):1945-1954.
35. Frank O, Blumberg S, Kruempel G, Hofmann T. Structure determination of 3-O-caffeoyl-
epi-gamma-quinide, an orphan bitter lactone in roasted coffee. J Agric Food Chem.
2008;56(20):9581-9585.
36. Frank O, Zehentbauer G, Hofmann T. Screening and identification of bitter compounds in
roasted coffee brew by taste dilution analysis. Developments in Food Science. 2006;43:165-
168. doi: https://doi.org/10.1016/S0167-4501(06)80039-7.
37. Halliwell B, Gutteridge JMC. Oxygen free-radicals and iron in relation to biology and
medicine - some problems and concepts. Arch Biochem Biophys. 1986;246(2):501-514.
38. Halliwell B, Clement MV, Long LH. Hydrogen peroxide in the human body. FEBS Lett.
2000;486(1):10-13.
39. Crozier A, Jaganath IB, Clifford MN. Dietary phenolics: Chemistry, bioavailability and
effects on health. Nat Prod Rep. 2009;26(8):1001-1043.
40. Daglia M, Cuzzoni MT, Dacarro C. Antibacterial activity of coffee - relationship between
biological-activity and chemical markers. J Agric Food Chem. 1994;42(10):2273-2277.
41. Kishimoto N, Kakino Y, Iwai K, Mochida K, Fujita T. In vitro antibacterial,
antimutagenic and anti-influenza virus activity of caffeic acid phenethyl esters. Biocontrol
Sci. 2005;10(4):155-161.
66
42. Nakaso K, Ito S, Nakashima K. Caffeine activates the PI3K/akt pathway and prevents
apoptotic cell death in a parkinson's disease model of SH-SY5Y cells. Neurosci Lett.
2008;432(2):146-150.
43. Hall S, Desbrow B, Anoopkumar-Dukie S, et al. A review of the bioactivity of coffee,
caffeine and key coffee constituents on inflammatory responses linked to depression. Food
Res Int.
44. Runge FF. Kaffe. In: Phytochemical discoveries. ; 1820:144-159.
45. Yang A, Palmer AA, De Wit H. Genetics of caffeine consumption and responses to
caffeine. Psychopharmacology. 2010;211(3):245-257.
46. Winston AP, Hardwick E, Jaberi N. Neuropsychiatric effects of caffeine. Adv Psychiatr
Treat. 2005;11(6):432-439.
47. Ludwig IA, Mena P, Calani L, et al. Variations in caffeine and chlorogenic acid contents
of coffees: What are we drinking? Food Funct. 2014;5(8):1718-1726.
48. Consumer Reports magazine. The buzz on energy-drink caffeine. Consumer Reports
magazine. December 2012.
49. Higdon JV, Frei B. Coffee and health: A review of recent human research. Crit Rev Food
Sci Nutr. 2006;46(2):101-123.
50. Johansson B, Halldner L, Dunwiddie TV, et al. Hyperalgesia, anxiety, and decreased
hypoxic neuroprotection in mice lacking the adenosine A1receptor. Proc Natl Acad Sci U S A.
2001;98(16):9407-9412.
67
51. European Food Safety Authority (EFSA). Scientific opinion on the safety of caffeine.
EFSA Journal. 2015;13(5):4102.
52. Meyer JF, Jr., Roselius L, Wimmer K, inventors; Kaffee Handels Ag, assignee. Removing
caffeine from coffee. 1908.
53. Saleeb FZ, Zeller BL, inventors; General Foods Corporation, assignee. Roasted coffee
extract decaffeination method. 1985.
54. Noyes RM, Voeller RW, inventors; The Procter & Gamble Company, assignee. Coffee
extraction process. 1982.
55. Zosel K, inventor; Studiengesellschaft Kohle m.b.H., assignee. Decaffeinizing coffee.
1971.
56. Fischer A, Kummer P, inventors; Coffex A.-G. S, assignee. Decaffeinization of raw
coffee. 1980.
57. Kim IJ, inventorA Method For Processing Coffee Beans And The Coffee Beans
Therefrom. 2011.
58. Murthy PS, Manonmani HK. Physico-chemical, antioxidant and antimicrobial properties
of indian monsooned coffee. Eur Food Res Technol. 2009;229(4):645-650.
59. Jumhawan U, Putri SP, Yusianto, Marwani E, Bamba T, Fukusaki E. Selection of
discriminant markers for authentication of asian palm civet coffee (kopi luwak): A
metabolomics approach. J Agric Food Chem. 2013;61(33):7994-8001.
60. Mayer F, Czerny M, Grosch W. Sensory study of the character impact aroma compounds
of a coffee beverage. Eur Food Res Technol. 2000;211(4):272-276.
68
61. Semmelroch P, Grosch W. Studies on character impact odorants of coffee brews. J Agric
Food Chem. 1996;44(2):537-543.
62. Moon J, Yoo HS, Shibamoto T. Role of roasting conditions in the level of chlorogenic
acid content in coffee beans: Correlation with coffee acidity. J Agric Food Chem.
2009;57(12):5365-5369.
63. Onakpoya I, Terry R, Ernst E. The use of green coffee extract as a weight loss
supplement: A systematic review and meta-analysis of randomised clinical trials.
Gastroenterology research and practice. 2011;2011.
64. Federal Trade Commission. Green coffee bean manufacturer settles FTC charges of
pushing its product based on results of “Seriously flawed” weight-loss study. . 2014.
65. Hemmerle H, Burger H-, Below P, et al. Chlorogenic acid and synthetic chlorogenic acid
derivatives: Novel inhibitors of hepatic glucose-6-phosphate translocase. J Med Chem.
1997;40(2):137-145.
66. Gloess AN, Schönbächler B, Klopprogge B, et al. Comparison of nine common coffee
extraction methods: Instrumental and sensory analysis. Eur Food Res Technol.
2013;236(4):607-627.
67. Caporaso N, Whitworth MB, Cui C, Fisk ID. Variability of single bean coffee volatile
compounds of arabica and robusta roasted coffees analysed by SPME-GC-MS. Food Res Int.
2018;108:628-640.
68. Dunkel A, Steinhaus M, Kotthoff M, et al. Nature's chemical signatures in human
olfaction: A foodborne perspective for future biotechnology. Angewandte Chemie-
International Edition. 2014;53(28):7124-7143.
69
69. Gloess AN, Yeretzian C, Knochenmuss R, Groessl M. On-line analysis of coffee roasting
with ion mobility spectrometry–mass spectrometry (IMS–MS). Int J Mass Spectrom.
2018;424:49-57.
70. Nissim I, Dagan-Wiener A, Niv MY. The taste of toxicity: A quantitative analysis of bitter
and toxic molecules. IUBMB Life. 2017;69(12):938-946.
71. Lund MN, Ray CA. Control of maillard reactions in foods: Strategies and chemical
mechanisms. J Agric Food Chem. 2017;65(23):4537-4552.
72. Ishibashi N, Kanehisa H, Okai H, Fukui S. Bitterness of leucine-containing peptides.
Agric Biol Chem. 1987;51(9):2389-2394.
73. World Coffee Research. World coffee research sensory lexicon 2.0. . 2017.
70
2. Aim of the Research
The aim of this thesis is to investigate and ultimately control each steps of the coffee process
to find how the chemical composition of the coffee bean changes using different process. The
results are expected to tailor coffee for people with problems on caffeine or acid sensitivity.
Alternatively consumes that demands on coffee with increased levels of caffeine or
chlorogenic acids can be satisfied.
The various aspects of this work include:
1) To find the effect of coffee grading and specialty coffee: Harvesting and grading step
Unlike coffee consuming trend changes to drink specialty coffee, high priced coffee
could not attract scientific interest because its rarity and high cost prevent from
changing the major coffee consuming style. Jamaica Blue Mountain coffee is one of
the most expensive and rare coffees in the world so did not attract scientific interest. In
this research, finding scientific reason why the coffee is special was conducted by
comparing different grades of Jamaica Blue Mountain coffees and other originated
normal price coffees.
2) To find the effect of roasting process and its alternative: Roasting step
Roasting has been an essential process for drinking coffee except consuming coffee
for medicinal use. However, the roasting process has a critical disadvantage that
degrades chlorogenic acids in coffee, which are regarded as the key compounds
responsible for health benefits in coffee. The noble method, which can substitute
roasting process using steaming and drying green coffee bean was developed in 2011.
However, the method was not analyzed in scientific methods but analyzed only with
sensory panels. This research is to obtain scientific data of the noble method and to
71
search pros and cons of the method compared to the normal roasting method.
3) To find the effect of extracting method: Brewing step
a. Extracting solvent
i. Comparison of water / methanol
ii. Comparison of different acidic water
b. Extracting method
Brewing step is the only adjustable stage for the coffee consumers among the whole
coffee processes. On the other hand, this step is the least interesting stage for chemists
because most of the brewing steps do not trigger chemical reactions. In this research,
the effect of different extracting solvent was compared with selected chlorogenic acids
in order to concatenate the academic coffee analysis result extracted by organic
solvents with practical water extracted coffee composition.
The kinetic study of coffee brewing was not conducted because it takes less than 20
seconds to brew an espresso coffee. Cold brew method has become popular since 2010
and it takes more than 12 hours so that it allows to study brewing kinetics. In this
research, the brewing kinetics using cold brew was studied focusing on the quantities
of chlorogenic acids and caffeine in coffee.
4) To find the applicability of coffee analyzing method to other beverages
All the analytical methods of different processes on coffee are not only useful for the
coffee analysis but also applicable to other food research. In this research, high priced,
specially parched, extracted with different solvents and kinetically obtained Korean
green tea samples were selected in order to prove the applicability of coffee analyzing
method to other food analysis.
72
3. Results
Research on Green Coffee Step 3.1.
Analysis on different grades of highly-rated Jamaica Blue Mountain coffees
compared to easily available originated coffee beans
Seung-Hun Lee, Rakesh Jaiswal, Nikolai Kuhnert*
Department of Life Sciences and Chemistry, Jacobs University Bremen, 28759 Bremen,
Germany
*Correspondence to:
Prof. Dr. Nikolai Kuhnert,
Department of Life Sciences and Chemistry
Jacobs University Bremen gGmbH
Campus Ring 1
28759 Bremen, Germany
Tel: +49 421 200 3120
Fax: +49 421 200 3102
E-mail: [email protected]
73
Abstract
Origin and coffee grades are major factors determining the price of coffee beans.
Jamaica Blue Mountain coffee achieves among the highest prices in coffee trading due
to its assumed high quality. In this contribution we show the first chemical composition
analysis of Jamaica Blue Mountain using LC-MS methods. In particular differences in
chlorogenic acid regioisomer distribution were observed for Jamaica Blue Mountain
coffee if compared to reference coffees sheding light on the chemical basis of grading
systems in coffee.
Keywords: Specialty coffee; Jamaica Blue Mountain coffee; Chlorogenic acid; tandem
MS
74
Abbreviations Used
CG, caffeoyl glucose; (de), derivatives; QA, quinic acid; CQA, caffeoylquinic acid;
Hyd., hydrated; GCQA, glucose caffeoylquinic acid; CQAG, caffeoylquinic acid
glucose; FQAG, feruloylquinic acid glucose; FQA, feruloylquinic acid; CQAL,
caffeoylquinic acid lactone; CoQA, coumaroylquinic acid; CFQA, caffeoyl-
feruloylquinic acid.
75
1. Introduction
Coffee is the third most consumed beverage after water and black tea in 21st century.
According to the increasing economic importance of the coffee industry, numerous
investigations with respect to the influence of the cup quality of coffee; species1, origin2,
coffee production3, roasting4,5 and brewing6 methods have been published.
With fast growing of coffee industry, decision standards of the price of coffee beans
were set. Indonesian Luwak coffee became the most desired and expensive coffee in the
world followed by the Black Ivory coffee, which appeared from Thailand around 2010.
However, these coffee products are additionally-artificially processed so that direct
comparison to other normally processed coffees does not seem to be fair. The main
decision standards of normally processed coffee price are species and origins. In general,
Arabica coffee is more expensive than Robusta coffee, and the price of same Arabica
coffee differs by a factor of up to ten according to its origins. One of the most expensive
coffees in the world is Jamaica Blue Mountain coffee which has a current retail price of
30 to 40 Euros per 250 g roasted, in Bremen, Germany, a factor of fifteen higher
compared to standard supermarket products. Jamaica Blue Mountain coffee is praised
for its rich aroma and long lasting after-taste.
The name Jamaica Blue Mountain comes from the grading system in Jamaica. Three
different grading systems are commonly used in each country. One is grading by
altitude of the coffee tree, another is by size of coffee bean and the other is by defected
beans per bag. Jamaican Blue Mountain (JBM) grade is the grade name by altitude.
76
Jamaican coffee beans are divided Blue Mountain – High Mountain – Prime Washed –
Prime Berry by altitude of the coffee tree. Also, Jamaica Blue Mountain coffee is
graded by size again, Blue Mountain Number 1 (Screen Size 17 ~ 18), Blue Mountain
Number 2 (Screen Size 16) and Blue Mountain Number 3 (Screen Size 15).
Furthermore, there is one more standard that if the beans are small and have some
defected beans, the grade becomes Jamaica Blue Mountain Triage.
Although Jamaican coffee beans receive top evaluations from coffee societies following
panel tasting, only limited scientific information rationalizing its high quality including
environmental7, isotope8 and antioxidant assay9 studies due to its rarity and high price is
currently available. Therefore, this research is the first component analysis of Jamaican
coffee beans by MS measurements including analysis of different JBM grades.
After the chlorogenic acids isomer study of Clifford et al.10, LC-MSn became one of the
most important tools for analyzing different polyphenols in coffee so that a lot of coffee
analysis researches with LC-MSn measurements are published. An extension of that,
HPLC-MSn is used for interpreting chemical structures in Jamaican coffee beans in this
study.
2. Materials and Methods
The experimental methods of HPLC-LC/MSn are followed by previous publications of
our group research2,11,12.
2.1. Chemicals and Coffee Beans
77
All the chemicals (Analytical grade) were purchased from Sigma-Aldrich (Germany).
Jamaica Blue Mountain Number 1, Jamaica Blue Mountain Number 2 and Jamaica Blue
Mountain Triage green coffee beans were provided by Youngki Moon from Jamaica,
Brazil, Colombia and Guatemala coffee beans were provided by Lloyd Caffee
(Fabrikenufer 115, Bremen, Germany).
2.2. Extract of Coffee Beans
All the green coffee beans were ground to fine powder, methanolic extracts were
prepared by Soxhlet extraction using aqueous methanol (70 %, 170 mL) for 3 hours,
using 5 g per each coffee beans. The extract was treated with Carrez reagents (1 mL of
reagent I plus 1 mL of reagent II) to precipitate colloidal material and filtered through a
Whatman no. 1 filter paper and stored at -20 ℃ until required, thawed at room
temperature, diluted to 60 mg / 10 mL, filtered through a membrane filter, and used for
LC/MS.
2.3. LC/MSn
The LC equipment (Agilent 1100 series, Karlsruhe, Germany) comprised a binary pump,
an autosampler with a 100 µL loop, and a diode-array detector (DAD) with a light-pipe
flow cell (recording at 320 and 254 nm and scanning from 200 to 600 nm). This was
interfaced with an ion-trap mass spectrometer fitted with an electrospray ionization (ESI)
source (Bruker Daltonics HCT ultra, Bremen, Germany) operating in full scan, auto
MSn mode (smart fragmentation) using a ramping of the collision energy. Maximum
fragmentation amplitude was set to 1 V, starting at 20 % and ending at 200 %. MS
operating conditions (negative mode) had been optimized using 5-caffeoylquinic acid
78
with a capillary temperature of 365 ℃, a drying gas flow rate of 10 L / min, and a
nebulizer pressure of 10 psi.
2.4. HPLC
Separation was achieved on a 150 × 3 mm i.d. column containing diphenyl modified
silica gel 5 µm, with a 5 mm × 3 mm i.d. guard column (Varian, Darmstadt, Germany).
Solvent A was water / formic acid (1000:0.005 v / v) and solvent B was methanol.
Solvents were delivered at a total flow rate of 500 µL / min. The gradient profile was
from 10 to 80 % B linearly in 70 min followed by 10 min isocratic and a return to 10 %
B at 90 and 10 min isocratic to re-equilibrate.
3. Results and Discussion
3.1. Interpretations of chlorogenic acids
A total of three JBM green bean samples and three reference samples from Brazil,
Colombia and Guatemala were analyzed in this study. Extraction of green coffee beans
and LC-MS analysis using reversed phase chromatography combined to negative ion
mode MS monitoring using a quadrupole ion trap ESI-MS instrument were carried out
as described previously.
As shown in Figure 1, all six different coffees have qualitatively almost the same
chlorogenic acid (CGA) composition. Three isomers of mono caffeoyl quinic acid and
four isomers of dicaffeoyl quinic acids could be readily identified based on retention
time and tandem MS provided in the literature10,13. Additional 40 further minor CGAs
could be identified including feruloyl quinic acids (FQA), p-coumaroyl quinic acids (p-
79
CoQA) and caffeoyl quinic acid lactones (CQAL). Total ion chromatograms (TICs) for
six representative green bean coffee extracts are shown in Figure 1. This similarity is
not unexpected since all samples originated from the same botanical species; Coffea
Arabica. The common chlorogenic acids’ mass spectral data are shown in Table 1 with
their chemical structures in Figure 2.
Figure 1. Total Ion Chromatogram of green coffee bean extracts in negative ion mode
from six different coffees
Jamaica Blue Mountain Number 1
Jamaica Blue Mountain Number 2
Jamaica Blue Mountain Triage
Brazil Santos
Colombia Supremo
Guatemala Antigua
0.0
0.2
0.4
0.6
0.8
8x10Intens.
0.0
0.2
0.4
0.6
0.8
1.0
8x10Intens.
0.0
0.2
0.4
0.6
0.8
1.08x10
Intens.
0.0
0.2
0.4
0.6
0.88x10
Intens.
0.0
0.2
0.4
0.6
0.8
1.08x10
Intens.
0.0
0.2
0.4
0.6
0.8
1.0
8x10Intens.
0 10 20 30 40 50 60 Time [min]
80
Table 1. Tandem MS data of chlorogenic acids identified in coffee extracts of
compounds in Jamaica Blue Mountain Number 1
No. Compd. RT
MS1 MS2 MS3
Parent ion
Base peak Secondary peak Base peak Secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z Int
1 CG (de) 3,7 341 178,6 160,6 34,6 112,8 12,91 160,6 142,6 65,16 89 62,38 112,8 47,43
2 CG (de) 3,7 473 340,8 130,8 21,81 452,8 3,43 178,6 160,6 74,78 281,6 19,68 118,8 17,57
3 CG 4 340,8 178,6 142,6 19,55 160,8 13,37 89 142,6 84,06 160,6 74,55 118,8 58,53
4 CG (de) 4,4 386,8 340,8 178,6 0,85 283 0,37 178,6 160,6 18,94 112,8 15,71 118,8 13,16
5 CG (de) 4,5 478 340,8 178,6 2,61 430 0,76 178,6 160,6 16,04 112,8 10,87 142,8 10,62
6 QA 5,9 190,6 85 93 66,84 126,8 65,86 59,2 -
7 QA (de) 6 405 190,6 386,8 10,38 274,6 6,58 126,8 172,6 100 85 57,26 110,8 46,01
8 CG (de) 6,4 341 178,6 322,8 74,34 142,6 24,76 160,6 89 95,71 124,6 63,19 100,8 58,9
9 CG (de) 6,9 457 340,8 178,6 5,72 438,8 4,14 178,6 142,8 18,5 160,6 17,03 118,8 9,83
10 CG (de) 11,5 487 383 441 33,78 341 8,94 340,8 322,8 18,71 178,4 2,72 112,8 2,11
11 Hyd. CQA 12,3 371 352,8 134,8 44,62 190,6 30,89 190,6 178,6 32,64 172,6 15,54 134,6 8,26
12 Hyd. CQA 12,8 371 352,8 134,8 43,79 190,6 41,44 190,6 178,6 46,51 134,8 11,51 172,6 10,66
13 Hyd. CQA 14,1 371 190,6 352,8 62,62 178,8 2,59 172,6 126,8 84,48 144,6 84,48 110,8 79,24
14 Hyd. CQA 15,2 371 352,8 172,6 64,8 190,6 28,45 172,6 190,6 63,82 178,6 54,62 134,6 7,8
15 Hyd. CQA 16,7 371 352,8 172,6 92,68 190,6 23,44 172,6 178,6 46,8 190,6 29,11 134,6 6,12
16 GCQA 17,9 515 178,6 340,8 29,78 352,8 28,56 134,8 -
17 CQAG 21,2 515 352,8 190,6 88,86 178,6 3,71 190,6 178,6 5,42 134,8 1,4 172,6 0,56
18 3-CQA 22,9 352,8 190,6 178,6 37,61 134,8 10,51 172,6 85 88,11 126,8 78,39 100,8 49,71
19 GFQA 25,6 529 190,6 367 83,05 172,6 58,72 93 175,8 32,47 85 19,56 143,4 4,06
20 CQAG 26,5 515 322,8 352,8 28,26 190,8 25,3 160,6 276,8 6,14 132,8 3,95 178,6 3,31
21 GCQA 26,7 515 352,8 322,8 98,86 340,8 93,59 190,6 172,6 93,73 178,6 43,99 134,8 8,13
22 MeOH-CQA 26,9 385 352,8 190,6 70,58 178,8 1,62 190,6 178,6 5,27 134,8 1,31 214,6 0,84
23 MeOH-CQA 27,4 385 352,8 190,6 78,68 351 32,4 190,6 214,6 21,29 178,6 5,1 172,6 3,14
24 5-CQA 27,8 352,8 190,6 214,6 7,42 178,6 2,86 172,6 85 85,57 110,8 59,36 126,6 59,27
25 3-CoQA 29,1 337 162,6 118,8 8,71 190,6 6,76 118,8 -
26 MeOH-CQA 31,3 385 352,8 172,6 17,28 190,6 9,16 172,6 178,6 61 190,6 12,9 214,6 7,79
27 3-CQAL 31,4 334,8 178,6 290,8 24,79 316,8 12,85 134,6 152,6 3,05 -
28 MeOH-CQA 31,5 385 352,8 172,6 23,21 178,6 8,15 172,6 178,6 62,52 190,6 18,08 134,8 8,75
29 3-FQA 32,2 366,8 192,6 193,6 8,84 133,8 7,78 133,6 148,6 26,47 190,6 3,12 116,8 2,58
30 4-CQA 33,5 352,8 172,6 178,6 54,46 190,6 21,98 93 110,8 88,73 154,6 32,97 71,2 26,15
31 4-CQAL 34,2 334,8 178,6 134,8 20,58 160,6 2,78 134,6 -
32 5-CoQA 35,3 336,8 190,6 162,6 4,36 172,6 1,95 126,6 85 100 172,6 66,2 93 56,48
33 5-FQA 37,1 366,8 190,6 192,6 3,16 172,8 2,36 126,8 172,6 54,36 85 52,09 108,8 29,29
34 4-CoQA 39,9 337 172,6 162,8 5,39 190,6 1,51 93 110,8 69,49 154,6 33,22 71,2 33,18
35 4-FQA 42,3 367 172,6 192,6 10,59 154,8 1,52 100,8 93 90,88 154,6 64,19 71,2 47,96
36 3,4,5-triCQA 43,6 677 515 334,8 15,6 353 13 352,8 178,6 20,66 340,8 16,86 334,8 12
37 3,5-diCQA 46,2 515 352,8 351 19,44 190,6 6,93 190,6 178,6 38,48 176,8 7,44 134,8 7,24
81
38 3,4-diCQA 48,5 515 352,8 334,8 24,61 172,6 19,04 172,6 178,6 71,95 190,6 53,6 134,8 9,15
39 4,5-diCQA 50 515 352,8 172,6 13,63 202,6 10,41 172,6 178,6 61,09 190,6 34,05 134,8 6,81
40 CFQA 51,4 529 366,8 352,8 81,75 351 17,16 192,6 190,6 46,8 172,6 16,43 133,6 9,79
41 CFQA 52,3 529 367 352,8 28,67 178,6 11,02 178,6 192,6 90,15 160,6 84,45 134,8 48,52
42 CQA (de) 53,5 499 352,8 318,8 74,15 336,8 36,67 172,6 178,6 45,97 190,6 17,94 134,8 7,37
43 p-CoQA (de) 54,7 499 336,8 360,8 59,9 318,8 32,49 172,6 335 83,63 162,8 25,32 190,8 5,85
44 CFQA 54,7 529 366,8 172,6 49,19 348,8 35,3 172,6 192,6 28,93 133,8 4,58 190,6 1,23
45 CFQA 56 529 367 352,8 90,57 350,8 40,31 172,6 192,6 52,04 190,6 35,43 133,8 5,95
46 p-CoQA (de) 61 513 336,8 348,8 30,8 172,8 26,62 172,6 162,6 17,14 190,6 1,79 294,8 1,52
47 FQA (de) 61,6 543 367 349 31,46 172,6 17,76 172,6 192,6 38,52 133,6 4 154,6 1,73
82
OH
OH
O
GlcO
OH
O
HOOH
OHOH
CG QA
O
HOOH
OH
OH
OHHO
OO
3-CQA
OH
O
HOOH
OH
OHO
OO
4-FQA
O
HOOH
OH
OH
OHHO
OO
Hyd. CQA
OH
O
HOOH
OH
OH
OHHO
OO
MeOH-CQA
O
OH
O
HOOH
OHO OGlc
OH
O
GCQA
OH OH
OHHO
OO
O O
4-CQAL
OH
O
HOOH
OHO
OH
O
5-CoQA
OH
O
HOOGlc
OHO
OH
OH
O
CQAG
OH
O
HOOGlc
OHO
O
OH
O
FQAG
O
HO
OH
OH
OHHO
OO
O OH
OH
O
3,4-diCQA
O
HO
O
OH
OHHO
OO
OOH
OH
O
HO OH
O
3,4,5-triCQA
O
HO
OH
OH
OHHO
OO
O O
OH
O
3-C-4-FQA
Figure 2. Chemical structures of chlorogenic acids identified in coffee extracts of
compounds in Jamaica Blue Mountain Number 1
83
3.2. Similarity among the different grades in Jamaica Blue Mountain coffee beans.
As shown in Figure 1, the three different graded Jamaica Blue Mountain coffee beans
show qualitatively almost the same CGA composition. Therefore, without using any
more sophisticated statistical tools and a larger number of samples, a comparison of
among the Jamaica Blue Mountain coffee beans is meaningless and was not further
pursued.
3.3. Different components between Jamaican and other originated beans
Although qualitatively the variety of chlorogenic acids are the same in all different
originated coffee beans, the individual relative amounts of CGA regioisomers are
different in Jamaican beans if compared to other reference beans.
In Figure 3, extracted ion chromatograms (EICs) with m/z 353 and 515 corresponding
to monocaffeoyl and dicaffeoyl quinic acid respectively are shown and the integrations
values given in Table 2, integration ratios of 3-CQA and 4-CQA against 5 CQA and
3,4-diCQA against the average of 5-CQA including diCQAs. The relative amount of 5-
CQA for mono caffeoyl derivatives and 3,4-diCQA for dicaffeoyl derivatives have been
set to 100 % for ease of comparison.
84
Figure 3. EIC of 353 (CQA) and 515 (diCQA) in six different coffees; 3-CQA, 5-CQA,
4-CQA, 3,5-diCQA, 3,4-diCQA and 4,5-diCQA peaks in order of retention time
0 2 4 6 7x10
Intens.
0.0
0.2
0.4
0.6
0.88x10
Intens.
0.0
0.2
0.4
0.6
0.88x10
Intens.
0 2 4 6 7x10
Intens.
0.0
0.2
0.4
0.6
0.88x10
Intens.
0.0
0.2
0.4
0.6
0.88x10
Intens.
0 10 20 30 40 50 60 Time [min]
Jamaica Blue Mountain Number 1
Jamaica Blue Mountain Number 2
Jamaica Blue Mountain Triage
Brazil Santos
Colombia Supremo
Guatemala Antigua
3-CQA
4-CQA 4,5-diCQA
3,4-diCQA
3,5-diCQA
5-CQA
85
Table 2. Relative amounts of monocaffeoyl quinic acids and dicaffeoyl quinic acids
against 5-CQA and 3,4-diCQA set at 100 % for comparison obtained by integration of
EICs at m/z 353 and 515 respectively
5-CQA 3-CQA 4-CQA AVG 3,4-diCQA 3,5-diCQA 4,5-diCQA AVG
JBM Number 1 100 24,65 57,96 41,31 100 63,02 125,85 94,43
JBM Number 2 100 29,98 62,50 46,24 100 57,77 117,89 87,83
JBM Triage 100 29,32 66,88 48,10 100 63,61 129,23 96,42
Brazil Santos 100 18,15 42,31 30,23 100 158,92 124,74 141,83
Colombia Supremo 100 18,42 41,98 30,20 100 112,89 181,24 147,06
Guatemala Antigua 100 15,92 38,50 27,21 100 154,79 151,08 152,93
As shown in Table 2, the three Jamaican coffee beans contain considerably more 3- and
4-CQA with respect to total 5-CQA if compared to the reference coffee beans. All three
averages of Jamaican 3- and 4-CQA contents are over 40 % of their 5-CQA contents
whereas other originated beans are only around 30 %. Likewise, three Jamaican coffee
beans have an increased 3,4-diCQA contents if compared to the relative amount of 5-
substituted diCQAs. All three Jamaican average contents of 5-CQA containing diCQAs
(average of 3,5- and 4,5-diCQA) are less than their non-5-CQA containing diCQA (3,4-
diCQA) while other originated coffee beans have around 150 %.
4. Conclusions
We analyzed Jamaica Blue Mountain coffee beans for the first time with HPLC-MS
analysis and interpreted 47 compounds including chlorogenic acids. This interpretation
informs us that the highly evaluated cup quality of coffee does not effect on the contents
of coffee beans in the m/z range of 50 ~ 1500.
86
We found that different grades of Jamaican coffee beans are virtually identical with
respect to their chemical composition.
There are differences between Jamaica Blue Mountain and other reference coffee beans
especially with respect to their quantities of 3-CQA and 4-CQA. Levels of 3-CQA and
4-CQA are considerably higher in JBM coffee beans. Similarly 3,4-diCQA levels are
considerably higher compared to their 5-substituted analogues. For the first time we
could show that a coffee of a defined origin has a significantly altered profile of
regioisomeric CGAs. We would like to suggest that this difference in regioisomer
composition, observed here might be responsible for sensory differences of JBM if
compared to lower valued coffee beans. 5-substituted CGAs are unable to form
chlorogenic acid lactones in coffee roasting, a class of compounds that has been
frequently linked to the sensory attributes of coffee14,15. Those could be connected to
rich aroma of Jamaica Blue Mountain coffee to make this coffee as one of the cup
quality coffee in the world.
Acknowledgements
Excellent technical support by Anja Müller, providing Jamaica coffee beans from
Youngki Moon and providing other originated coffee beans from Lloyd Caffee are
acknowledged.
References
87
1. Lucon Wagemaker, T.A.; Limonta Carvalho, C.R.; Maia, N.B.; Baggio, S.R.; Guerreiro Filho, O. Sun protection factor, content and composition of lipid fraction of green coffee beans. Industrial Crops and Products 2011, 33, 469-473.
2. Kuhnert, N.; Jaiswal, R.; Eravuchira, P.; El-Abassy, R.M.; von der Kammer, B.; Materny, A. Scope and limitations of principal component analysis of high resolution LC-TOF-MS data: the analysis of the chlorogenic acid fraction in green coffee beans as a case study. Analytical Methods 2011, 3, 144-155.
3. Balyaya, K.J.; Clifford, M.N. Chlorogenic acids and caffeine contents of monosooned Indian Arabica and Robusta coffees compared with wet and dry processed coffees from the same geographic area. Colloq. Sci. Int. Cafe, [C. R. ] 1995, 16th, 316-325.
4. Moon, J.; Yoo, H.S.; Shibamoto, T. Role of Roasting Conditions in the Level of Chlorogenic Acid Content in Coffee Beans: Correlation with Coffee Acidity. J. Agric. Food Chem. 2009, 57, 5365-5369.
5. Moon, J.; Shibamoto, T. Role of Roasting Conditions in the Profile of Volatile Flavor Chemicals Formed from Coffee Beans. J. Agric. Food Chem. 2009, 57, 5823-5831.
6. Gross, G.; Jaccaud, E.; Huggett, A.C. Analysis of the content of the diterpenes cafestol and kahweol in coffee brews. Food and Chemical Toxicology 1997, 35, 547-554.
7. Robinson, D.E.; Mansingh, A. Insecticide contamination of Jamaican environment. IV. Transport of residues from coffee plantations in the blue mountains to coastal waters in eastern Jamaica. Environ. Monit. Assess. 1999, 54, 125-141.
8. Weckerle, B.; Richling, E.; Heinrich, S.; Schreier, P. Origin assessment of green coffee (Coffea arabica) by multi-element stable isotope analysis of caffeine. Analytical and Bioanalytical Chemistry 2002, 374, 886-890.
9. Parras, P.; Martinez-Tome, M.; Jimenez, A.M.; Murcia, M.A. Antioxidant capacity of coffees of several origins brewed following three different procedures. Food Chem. 2007, 102, 582-592.
10. Clifford, M.N.; Johnston, K.L.; Knight, S.; Kuhnert, N. Hierarchical scheme for LC-MSn identification of chlorogenic acids. J. Agric. Food Chem. 2003, 51, 2900-2911.
11. Jaiswal, R.; Patras, M.A.; Eravuchira, P.J.; Kuhnert, N. Profile and Characterization of the Chlorogenic Acids in Green Robusta Coffee Beans by LC-MSn: Identification of Seven New Classes of Compounds. J. Agric. Food Chem. 2010, 58, 8722-8737.
12. Jaiswal, R.; Matei, M.F.; Golon, A.; Witt, M.; Kuhnert, N. Understanding the fate of chlorogenic acids in coffee roasting using mass spectrometry based targeted and non-targeted analytical strategies. Food & Function 2012, 3, 976-984.
88
13. Clifford, M.N.; Knight, S.; Kuhnert, N. Discriminating between the six isomers of dicaffeoylquinic acid by LC-MSn. J. Agric. Food Chem. 2005, 53, 3821-3832.
14. Frank, O.; Zehentbauer, G.; Hofmann, T. Bioresponse-guided decomposition of roast coffee beverage and identification of key bitter taste compounds. European Food Research and Technology 2006, 222, 492-508.
15. Dunkel, A.; Steinhaus, M.; Kotthoff, M.; Nowak, B.; Krautwurst, D.; Schieberle, P.; Hofmann, T. Nature's Chemical Signatures in Human Olfaction: A Foodborne Perspective for Future Biotechnology. Angewandte Chemie-International Edition 2014, 53, 7124-7143.
89
Research on Roasting Step 3.2.
Analysis of Steamed (“Fermented”) Coffee Beans as a Roasted Coffee Substitute
Seung-Hun Lee1, In Jae Kim2, Sabur Badmos1 and Nikolai Kuhnert*1
Department of Life Sciences and Chemistry, Jacobs University Bremen, 28759 Bremen,
Germany
Coffeebio, A-B203, 144-3, Sangdaewon-dong, Jungwon-gu, Seongnam-si, Gyeonggi-do,
Korea
*Correspondence to:
Prof. Dr. Nikolai Kuhnert,
Department of Life Sciences and Chemistry
Jacobs University Bremen gGmbH
Campus Ring 1
28759 Bremen, Germany
Tel: +49 421 200 3120
Fax: +49 421 200 3102
E-mail: [email protected]
90
Abstract
In this contribution, steamed coffee beans (also referred to as fermented coffee) that
have been introduced commercially as roasted coffee substitute were chemically
profiled. Total polyphenol content and antioxidant capacities have been determined and
compared to green and roasted coffee beans. Additionally, the chlorogenic acid profile
has been determined using LC-MS methods and compared to green and roasted coffee
beans. Steamed coffee shows a higher antioxidant capacity if compared to green and
roasted coffee as well as an altered chlorogenic acid profile.
91
1. Introduction
From farm to table, diverse processes are used in each step of coffee production. As a
first processing step the flesh is removed from the coffee cherry by dry, wet and semi-
dry processes to obtain green coffee bean. 1,2 The term fermentation in coffee has
already been widely used in the process of transforming coffee cherry to green coffee
bean, and this refers normally to the action of microorganisms on green coffee beans for
a period of three to seven days. 3 As a next step green coffee beans are roasted at
temperatures between 160 and 250 °C for about 5 to 20 min. During roasting, the
chemical composition of coffee changes dramatically with most of the desirable aroma
compounds being produced. Following roasting, grinding of intact roasted beans is
carried out using two different types of grinder; burr and blade grinders. After grinding
coffee bean, Espresso machine, mocha pot, electric coffee maker or hand-brewing paper
drip could be selected in personal preferences to achieve extraction of soluble coffee
constituents under varying conditions. 4,5
To obtain the desirable organoleptic properties of coffee, roasting is considered as
inevitable. However, at elevated temperatures chlorogenic acids (CGAs) are
unavoidably decomposed to produce a series of chemically altered derivatives. This is
accompanied by a reduction in total polyphenol content as measured by reduced
antioxidant capacity assays. 6 On average, 50 % of the total CGAs are decomposed
during roasting with extreme cases such as 98 % decomposition is already reported. 7
Since CGAs have been shown to be the main contributor to the beneficial health effects
of the coffee against some human disorders such as Type-II diabetes, cardiovascular and
92
liver associated diseases, Parkinson syndrome etc., 8,9 coffee processing methods, which
would retain or increase CGAs quantities, are highly desirable. To address this issue, a
multinational company has introduced the Green Blend product containing 35 % of
green coffee beans and hence delivering a beverage with high CGAs content. Increasing
the quantity of green coffee beans to the product negatively affects the palatability and
results in negative sensory reviews of the coffee beverage. Hence, alternative
approaches towards producing a high CGA coffee that is acceptable to coffee
consumers is of interest.
In 2011, a research on the use of steaming and drying of coffee was published as a
patent which indicates that this special treatment of green coffee bean changed green
flavor to fruit flavor, as well as the noble processed coffee bean extracted made it aroma
richer than the reference-roasted coffee bean extracted by sensory evaluation.10 This
type of coffee is as well referred as “fermented coffee”. However, there were limitations
in producing bitterness and burnt flavor which is usually regarded as unique properties
of coffee taste.
Among the many methods on the coffee fermentation patent, steaming without active
use of microorganisms gave a good condition with final fermented bean. The
fermentation procedure used in this paper is this steaming and drying method only with
the allowance of natural containing microorganisms to process the fermentation.
Therefore, the new word ‘steamentation’ is derived from this novel method which is in
between steaming and fermentation. This steamentation process is a newly developed
method of processing coffee which can possibly be a good substitute to the
93
monopolistic roasting method that has been employed for ages. This process involves
starting from the green coffee bean which has already been fermented from the
traditional system of coffee fermentation. It should as well be noted that “stomach
friendly coffee” is frequently steam treated prior to roasting. 11,12
The aim of this study is to analyze antioxidants and to profile chlorogenic acids in
steamented coffee bean using LC-MS techniques, purposely to compare it with the
green and roasted coffee beans.
2. Experimental
The experimental methods of HPLC and LC-MSn were followed from previous
publications. 13,14,15
2.1. Chemicals, Coffee Beans and Brewing Materials
All the chemicals (analytical grade) were purchased from Sigma-Aldrich (Germany).
Green, fermented and two differently roasted Brazil Arabica coffee beans were provided
by Coffeebio, Gyeonggi-do, Korea.
2.2. Steamentation and Roasting Conditions of Coffee Beans
2.2.1. Steamentation
120 kg Brazil green coffee beans were steamed at 90 ℃ for 9 hr, dried at 50 ℃ for 15 hr
and set at room temperature for 3 days. Those beans were steamed again at 72 ℃ for 7
hr and dried at 38 ℃ for 12 hr. In total, the steaming and drying processes were repeated
9 times which resulted into the steamented coffee bean as shown in Figure 1.10
94
2.2.2. Roasting
The two differently roasted coffee beans are roasted using the same roaster (TOPER,
Turkey). Coffee beans were roasted in medium degree of roasting for handdrip roasted
coffee bean and in between medium and city degree of roasting for espresso roasted
coffee bean. Figure 1 depicts pictures of the respective coffee beans.
Figure 1. Pictures of green, steamented and two differently roasted coffee beans
2.3. Extract of Coffee Beans
All the green coffee beans were ground to fine powder, methanolic extracts were
prepared by Soxhlet extraction using aqueous methanol (70 %, 170 mL) for 3 hr, using
5 g per different coffee beans. The extract was stored at -20 ℃ until required, thawed at
room temperature, diluted to 5 % of the concentrated extract liquid, filtered through a
membrane filter, and analyzed by LC/MS. The concentration of the extracted liquid
were measured by evaporating solvent at 100 ℃, 30 min using aluminum boat, by
measuring boat with solution before and after.
2.4. High Resolution LC-ESI-MS
The LC equipment (Agilent 1100 series, Karlsruhe, Germany) comprised a binary pump,
an autosampler with a 100 µL loop, and a diode-array detector (DAD) with a light-pipe
95
flow cell (recording at 320 and 254 nm and scanning from 200 to 600 nm). This was
interfaced with a MicroTOF Focus mass spectrometer (Bruker Daltonics, Bremen,
Germany) fitted with an electrospray ionization (ESI) source and internal calibration
was achieved with 10 mL of 0.1 M sodium formate solution injected through a six port
valve prior into each chromatographic run. Calibration was carried out using the
enhanced quadratic mode. All experiments were carried out in negative ion mode.
2.5. LC-ESI-MSn
The LC equipment (Agilent 1100 series, Karlsruhe, Germany) comprised a binary pump,
an autosampler with a 100 µL loop, and a diode-array detector (DAD) with a light-pipe
flow cell (recording at 320 and 254 nm and scanning from 200 to 600 nm). This was
interfaced with an ion-trap mass spectrometer fitted with an ESI source (Bruker
Daltonics HCT ultra, Bremen, Germany) operating in full scan, auto MSn mode (smart
fragmentation) using a ramping of the collision energy. Maximum fragmentation
amplitude was set to 1 V, starting at 20 % and ending at 200 %. MS operating
conditions (negative mode) had been optimized using 5-caffeoylquinic acid with a
capillary temperature of 365 ℃, a drying gas flow rate of 10 L / min, and a nebulizer
pressure of 10 psi.
2.6. HPLC
Separation was achieved on a 150 × 3 mm i.d. column containing diphenyl 5 µm, with a
5 mm × 3 mm i.d. guard column (Varian, Darmstadt, Germany). Solvent A was water /
formic acid (1000:0.005 v / v) and solvent B was methanol. Solvents were delivered at a
96
total flow rate of 500 µL / min. The gradient profile was from 10 to 80 % B linearly in
70 min followed by 10 min a return to 10 % and 10 min isocratic to re-equilibrate.
2.7. Antioxidant Assay
DPPH (2,2-diphenyl-1-picrylhydrazyl) and FRAP (Ferric ion reducing antioxidant
power) assays were used to determine the antioxidant capacity of the coffee samples
using 0.5 mg/mL garlic acid and 0.5 mg/mL Trolox (both dissolved in 70 % methanol)
as standards. For the DPPH assay, 10 µL of the extracted coffee samples and standards
were pipetted into 96-well plate with 200 µL of 0.5 mM DPPH solution added to each
sample, mixed gently, incubated in the dark for 30 min and measured absorbance at 517
nm. For FRAP assay, freshly prepared FRAP reagent solution was used (10 mL of
sodium acetate buffer mixed with 1 mL each of 10 mM solution of TPTZ (2,4,6-Tris(2-
pyridyl)-s-triazine) in 40 mM HCl and 20 mM solution of Iron (III) chloride in water).
10 µL of each sample and standards were pipetted into 96-well plate, 200 µL of FRAP
solution was added to each, mixed gently, incubated in the dark for 10 min at 37 ℃ and
absorbance measured at 593 nm. 16-18
3. Results and Discussion
3.1. Changes in the Amount of Total Antioxidants as per Processes
The aim of this study is to compare the polyphenol content of steamented coffee with
green and two differently roasted coffee beans. As roasted coffee beans, Brazilian
Arabica coffee beans roasted in medium degree of roasting (Hand drip roast) and in
between medium and city degree of roasting (Espresso roast) were used as a comparison.
97
In analyzing the CGAs changes among the different processed coffees, 5-CQA was
selected as an indicator compound because it is the most abundant chlorogenic acid in
coffee. The amount of 5-CQA is quantified with HPLC-TOF-MS spectra with 280 nm
UV detector. The result is shown in Figure 2.
Figure 2. Amount of 5-CQA in the four different processed coffee samples (mg / L)
Aqueous methanolic extracts of all the four samples were obtained and total antioxidant
capacity was measured by using both DPPH and FRAP assays. All values are
referenced to gallic acid and Trolox equivalents.
0
500
1000
1500
2000
2500
3000
EspressoRoasted
Hand dripRoasted
Steamented
5-C
QA
mg
/ L
Green
98
Previous studies have shown that the amount of antioxidant in coffee decreases by the
increment of roasting degree.7 In the same vein, it is expected and proved with indicator
quantification that the amount of total antioxidant would decrease in this order of green
– steamented – hand drip roasted – espresso roasted because as mentioned in 2.2.1., the
steamentation process in this paper has maximum heating temperature at 90 ℃ which is
between green (room temperature) and roasted coffee processes.
However, the antioxidant values shown in Figure 3 are different from initial
expectation and indicator quantification data. The amount of antioxidant in steamented
coffee beans was the highest among the four different processed coffee beans. This
phenomenon could be explained with enhancing reactivity of degraded antioxidant
compounds. In essence, the changes of antioxidant reactivity are higher than that of the
reducing amount of antioxidant while antioxidant reducing is predominant factor in
roasting process, even though the amount of total antioxidant is reduced by
steamentation process.
99
0
10
20
30
40
50
60
70
EspressoRoasted
Hand dripRoasted
SteamentedGreen
Am
ount
in S
ampl
e
g GAE / kg g TE / kg
3(a) DPPH
100
Figure 3. Antioxidant assay by DPPH (a) and FRAP (b) for the green, steamented,
handdrip roasted and espresso roasted coffee samples
3.2. Chlorogenic Acid Profiling
3.2.1. Analysis of high resolution MS data
To obtain information on the polyphenolics at a molecular level, all the four coffee
samples were profiled by LC-MS techniques. High resolution MS data analysis, with a
LC-ESI-TOF-MS method, was used in obtaining high resolution m/z values and from
this, the molecular formulae were determined. Table 1 shows all CGAs found in the
Brazil coffee beans listed by their retention times with their exact and theoretical m/z
0
10
20
30
40
50
60
70
80
90
EspressoRoasted
Hand dripRoasted
SteamentedGreen
Am
ount
in S
ampl
e
g GAE / kg g TE / kg
3(b) FRAP
101
values. Details of all the chemical structures and numberings in this paper are shown in
the supporting information and follow our previous shorthand nomenclature on CGAs.19
Table 1. High resolution mass (TOF-MS) data of chlorogenic acids
No. Name Mol. Formula Exp. m/z (M-H) Theor. m/z (M-H) Error (ppm)
1d 4-Caffeoylshikimic acid C16H16O8 335.0765 335.0772 2.2
2c 3-Caffeoylquinic acid lactone C16H16O8 335.0782 335.0772 -2.8
1e 5-Caffeoylshikimic acid C16H16O8 335.0770 335.0772 0.7
2d 4-Caffeoylquinic acid lactone C16H16O8 335.0777 335.0772 -1.3
2a 1-Caffeoylquinic acid lactone C16H16O8 335.0775 335.0772 -0.9
3c 3-p-Coumaroylquinic acid C16H18O8 337.0929 337.0929 0
3e 5-p-Coumaroylquinic acid C16H18O8 337.0922 337.0929 2.1
3d 4-p-Coumaroylquinic acid C16H18O8 337.0931 337.0929 -0.7
4c 3-Caffeoylquinic acid C16H18O9 353.0888 353.0878 -2.7
4e 5-Caffeoylquinic acid C16H18O9 353.0901 353.0878 -6.4
4d 4-Caffeoylquinic acid C16H18O9 353.0885 353.0878 -2
5c 3-Feruloylquinic acid C17H20O9 367.1033 367.1035 0.5
5e 5-Feruloylquinic acid C17H20O9 367.1037 367.1035 -0.5
5d 4-Feruloylquinic acid C17H20O9 367.1041 367.1035 -1.8
6c Methyl-3-caffeoylquinate C17H20O9 367.1052 367.1035 -4.8
6d Methyl-4-caffeoylquinate C17H20O9 367.1051 367.1035 -4.3
7u Dimethoxycinnamoylquinic acid C18H22O9 381.1201 381.1191 -2.5
7u Dimethoxycinnamoylquinic acid C18H22O9 381.1205 381.1191 -3.5
8u Caffeoyl-coumaroylquinic acid C25H24O11 499.1245 499.1246 0.3
8u Caffeoyl-coumaroylquinic acid C25H24O11 499.1239 499.1246 1.5
9u Coumaroyl-feruloylquinic acid C26H26O11 513.1388 513.1402 2.9
10ce 3,5-diCaffeoylquinic acid C25H24O12 515.1197 515.1195 -0.3
10cd 3,4-diCaffeoylquinic acid C25H24O12 515.1199 515.1195 -0.8
10de 4,5-diCaffeoylquinic acid C25H24O12 515.1201 515.1195 -1.1
11u Caffeoyl-feruloylquinic acid C26H26O12 529.1379 529.1352 -5.1
11u Caffeoyl-feruloylquinic acid C26H26O12 529.1337 529.1352 2.7
11u Caffeoyl-feruloylquinic acid C26H26O12 529.1346 529.1352 1
11u Caffeoyl-feruloylquinic acid C26H26O12 529.1344 529.1352 1.5
13u Dimethoxycinnamoyl-
feruloylquinic acid C27H28O12 543.1532 543.1508 -4.4
3.2.2. Assignment of chlorogenic acids by tandem MS
Numerous tandem MS data allowing unambiguous assignment of CGAs to their
respective regioisomeric level especially in the p-coumaroyl quinic acid derivatives;
102
caffeoyl quinic acids 20, feruloylquinic acids 21, and others 19 have been reported and
were applied here.
Structures of CGAs in the four different coffee beans were assigned and are listed
according to their retention time as shown in Table 2 along with MS data shown in
Figure 4. Table 2 shows the tandem MS data followed by compound numbering in
Supplementary table 1 and Supplementary figure 1 ~ 3. And all of the four different
processed coffee bean interpretations by fragmentation pattern are provided in
Supplementary table 2 ~ 5. For compound abbreviations, we use the shorthand
nomenclature introduced by Clifford MN.
103
Figure 4. Total ion chromatogram in negative ion mode of four different coffee beans
Green Bean
Handdrip Roasted Bean
Espresso Roasted Bean
0
1
2
3
7 x10
Intens.
0.00
0.25
0.50
0.75
1.00
1.25
7 x10
0
1
2
3
4
5
6 x10
0
2
4
6
6 x10
0 5 10 15 20 25 30 35 40 45
Steamented Bean
Time [min]
4c
4e
5c 4d
3e
5e
5d 10ce 10de
4ei 4c
4e
4ci 3ci
5c
4ei 4d
2c
3e 5e
2d
5d
10ce 10de
4ci 4c
4ci 4c
4ei
4ei
4ci
4e
4e 6c
4ei
4ei 6c
5c
5c
4d
4d
2c
2c
6d
6d
1e
1e
5e
2d
5e
2d
5d
5d
12u
12u
104
Table 2. Tandem MS data in negative ion mode of Chlorogenic acids from Brazil
espresso roasted coffee bean extract
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1c 3-CSA 335.01 178.7 160.8 70 134.9 41 134.8 133.8 6 135.8 6 132.8 4
2c 3-CQAL 334.98 160.7 134.9 93 178.7 12 132.8 - - - - - -
1e 5-CSA 334.98 172.8 160.8 27 178.8 24 110.9 93 73 154.7 31 108.9 12
2d 4-CQAL 334.97 160.8 314.9 17 134.9 15 132.9 - - - - - -
2a 1-CQAL 334.97 160.8 172.7 83 315.9 12 132.8 - - - - - -
3e 5-pCoQA 336.96 190.8 134.9 7 160.7 6 84.9 172.8 57 87.2 53 - -
3d 4-pCoQA 336.97 172.7 162.8 6 173.7 6 110.9 93 48 98.9 24 - -
4c 3-CQA 352.46 190.8 178.8 44 176.8 34 85.1 172.7 71 126.9 66 108.9 34
4e 5-CQA 352.25 190.7 214.8 45 194.7 7 126.9 85.1 66 172.7 55 93.1 46
4d 4-CQA 352.47 172.8 178.8 51 190.8 24 93.1 110.9 69 154.7 16 59.6 13
5c 3-FQA 366.99 192.7 133.9 14 193.7 10 133.8 148.8 24 134.7 3 116.9 1
5e 5-FQA 367 190.8 191.7 7 192.8 6 126.9 172.8 74 85.2 71 110.9 57
5d 4-FQA 367.01 172.7 190.8 13 192.7 12 93 110.9 58 154.7 33 71.3 25
6c M-3-CQ 367 160.7 192.8 16 132.9 11 132.8 - - - - - -
6d M-4-CQ 366.99 160.8 134.9 36 192.7 10 132.9 - - - - - -
10ce 3,5-diCQA 515.02 353 351.1 33 353.9 9 190.7 178.7 53 176.8 15 134.9 8
10cd 3,4-diCQA 514.52 353 335 77 338.9 57 190.7 172.7 100 178.7 69 191.9 35
10de 4,5-diCQA 514.46 353 351 70 335 18 172.7 178.8 59 194.7 34 190.8 28
11ec 5-C-3-FQA 529 367 353 28 351 19 192.7 190.8 24 133.8 12 172.8 11
11u C-4-FQA 529.05 367 349 54 172.8 42 172.8 192.8 35 190.7 5 133.7 5
12u M-diCQA 529.03 367 160.8 20 368 19 160.8 134.8 26 132.9 14 192.7 11
3.3. Comparison of CGA profile in four coffee samples
When comparing the chromatogram of the four samples investigated, it becomes
apparent that the number of peaks observed increases from green beans to steamented
beans and to two differently roasted beans. This observation reflects that the degree of
chemical transformation of CGAs undergo thermal treatment. Typical thermal reaction
products such as CGA lactones and shikimic acid derivatives, both obtained through
dehydration, transesterification products and epimers of quinic acid can be identified in
105
roasted bean samples. The most significant changes were observed when compared to
steamented with roasted coffee beans for series of isobaric ions at m/z 367 and 335.
3.3.1. Differences in extracted ion chromatograms (EICs) at m/z 367
In EICs at m/z 367 ± 0.5, feruloylquinic acids and caffeoylquinic acid methyl esters
corresponding to the molecular formula C17H20O9 can be located. In all the four coffee
bean samples three feruloylquinic acids were observed while signals corresponding to
caffeoylquinic acid methyl ester were only observed in roasted coffee beans. With these
EICs, we observed that steamentation process decreases the relative intensity of FQAs,
but does not yield methyl esters, which are only formed at elevated roasting
temperatures. Figure 5 summarizes EICs at m/z 367 with their compound structures in
Supplementary figure 4 and their MS2 spectra provided in Supplementary figure 5.
106
Figure 5. EICs at m/z 367 ± 0.5 (C17H20O9, M-H) in negative ion mode of the four
different processed coffee beans
3.3.2. Differences in EICs at m/z 335
In EICs at m/z 335 ± 0.5, caffeoylquinic acid lactones and caffeoylshikimic acids
(C16H16O8, M-H) were observed while in the green bean spectrum, those signals at m/z
335 were absent.
In steamented, handdrip and espresso roasted coffee beans, 3- and 4-caffeoylquinic acid
lactones were found. 1-CQL was observed only in roasted samples suggesting that acyl
05022015000002.D: EIC 367.00 -All MS
05022015000003.D: EIC 367.00 -All MS
05022015000005.D: EIC 367.00 -All MS
05022015000006.D: EIC 367.00 -All MS
0.0
0.2
0.4
0.6
0.8
1.0
7x10Intens.
0
1
2
3
6x10Intens.
0.0
0.5
1.0
1.5
2.0
6x10Intens.
0.0
0.5
1.0
1.5
2.0
2.5
6x10Intens.
0 10 20 30 40 50 60 Time [min]
6d
6d
6c
6c
5c
5c
5c
5c
6e
6e
5e
5e
5e
5e
5d
5d
5d
5d
Green Bean
Steamented Bean
Handdrip Roasted Bean
Espresso Roasted Bean
107
migration does not take place during steamentation. Caffeoylshikimic acids were only
observed in roasted coffee beans. With these observations, it could be concluded that
steamentation process lead to elimination of water from CQAs to yield quinic acid
lactones, while elimination of water from the cyclohexanol moiety to produce shikimic
acid derivatives occurs only at higher temperatures as employed in roasting process.
Figure 6 shows the EICs of all the four samples with the chemical structures in
Supplementary figure 6; 4-CSA (1d), 3-CQAL (2c), 5-CSA (1e), 4-CQAL (2d) and 1-
CQAL (2a) respectively identified as species with a pseudo-molecular ion at m/z 335 by
using MS2 spectra shown in Supplementary figure 7. 22
108
Figure 6. EICs at m/z 335 ± 0.5 (C16H16O8, M-H) in negative ion mode of the four
different processed coffee beans
4. Conclusion
In conclusion, we analyzed Brazil Arabica coffee beans with LC-MS in four different
treatment conditions; green coffee, steamented coffee, medium roasted coffee and high
roasted coffee. We analyzed chlorogenic acids compounds with high resolution MS and
interpreted polyphenolic compounds with tandem-MS spectra. In addition, antioxidant
assays were processed with different standards and methods. Most importantly, we
verified for the first time, the differences between steamentation and roasting processes.
Steamentation as a roasting substitute in coffee can make coffee drinkable as long as
05022015000002.D: EIC 335.00 -All MS
05022015000003.D: EIC 335.00 -All MS
05022015000005.D: EIC 335.00 -All MS
05022015000006.D: EIC 335.00 -All MS
0.0
0.2
0.4
0.6
0.8
1.05x10
Intens.
0
1
2
3
4
5x10Intens.
0
1
2
3
46x10
Intens.
0.0
0.5
1.0
1.5
2.0
2.5
6x10Intens.
0 10 20 30 40 50 60 Time [min]
1e 1d
2c
2d
2a
1e
1d
2c
2d
2a
2c
2d
Green Bean
Steamented Bean
Handdrip Roasted Bean
Espresso Roasted Bean
109
chemical reactions are limited when compared with the settled roasting process. We
suggest this research outcome would be useful in improving the trend on well-being
style of coffee consumers for higher antioxidant containing coffee. Further research on
various steamentation methods and combining it with roasting and steamenting
processes would ultimately leads to production of higher antioxidant content and tastier
coffee.
5. Acknowledgements
Excellent technical support by Anja Müller is acknowledged.
This Research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
Conflicts of interest: none
References
1. Balyaya KJ, Clifford MN. Chlorogenic acids and caffeine contents of monosooned indian arabica and
robusta coffees compared with wet and dry processed coffees from the same geographic area.
Colloq Sci Int Cafe, [C R ]. 1995;16th(Vol. 1, Seizieme Colloque Scientifique International sur le Cafe,
1995, Vol. 1):316-325.
2. Balyaya KJ, Clifford MN. Individual chlorogenic acids and caffeine contents in commercial grades of
wet and dry processed indian green robusta coffee. Journal of Food Science and Technology-Mysore.
1995;32(2):104-108.
110
3. Amorim ACL, Hovell AMC, Pinto AC, et al. Green and roasted arabica coffees differentiated by
ripeness, process and cup quality via electrospray ionization mass spectrometry fingerprinting. J Braz
Chem Soc. 2009;20(2):313-321.
4. Ludwig IA, Sanchez L, Caemmerer B, Kroh LW, Paz De Pena M, Cid C. Extraction of coffee
antioxidants: Impact of brewing time and method. Food Res Int. 2012;48(1):57-64.
5. Sanchez-Gonzalez I, Jimenez-Escrig A, Saura-Calixto F. In vitro antioxidant activity of coffees brewed
using different procedures (italian, espresso and filter). Food Chem. 2005;90(1-2):133-139.
6. Chandrasekara A, Naczk M, Shahidi F. Effect of processing on the antioxidant activity of millet grains.
Food Chem. 2012;133(1):1-9.
7. Moon J, Yoo HS, Shibamoto T. Role of roasting conditions in the level of chlorogenic acid content in
coffee beans: Correlation with coffee acidity. J Agric Food Chem. 2009;57(12):5365-5369.
8. Cano-Marquina A, Tarín JJ, Cano A. The impact of coffee on health. Maturitas. 2013;75(1):7-21.
9. Shahidi F, Chandrasekara A. Hydroxycinnamates and their in vitro and in vivo antioxidant activities.
Phytochem Rev. 2010;9(1):147-170.
10. Kim IJ, inventorA Method For Processing Coffee Beans And The Coffee Beans Therefrom. 2011.
11. Rubach M, Lang R, Skupin C, Hofmann T, Somoza V. Activity-guided fractionation to characterize a
coffee beverage that effectively down-regulates mechanisms of gastric acid secretion as compared to
regular coffee. J Agric Food Chem. 2010;58(7):4153-4161.
12. Weiss C, Rubach M, Lang R, et al. Measurement of the intracellular ph in human stomach cells: A
novel approach to evaluate the gastric acid secretory potential of coffee beverages. J Agric Food Chem.
2010;58(3):1976-1985.
111
13. Jaiswal R, Patras MA, Eravuchira PJ, Kuhnert N. Profile and characterization of the chlorogenic acids
in green robusta coffee beans by LC-MSn: Identification of seven new classes of compounds. J Agric
Food Chem. 2010;58(15):8722-8737.
14. Jaiswal R, Matei MF, Golon A, Witt M, Kuhnert N. Understanding the fate of chlorogenic acids in
coffee roasting using mass spectrometry based targeted and non-targeted analytical strategies. Food &
Function. 2012;3(9):976-984.
15. Kuhnert N, Jaiswal R, Eravuchira P, El-Abassy RM, von der Kammer B, Materny A. Scope and
limitations of principal component analysis of high resolution LC-TOF-MS data: The analysis of the
chlorogenic acid fraction in green coffee beans as a case study. Analytical Methods. 2011;3(1):144-155.
16. Huang DJ, Ou BX, Prior RL. The chemistry behind antioxidant capacity assays. J Agric Food Chem.
2005;53(6):1841-1856.
17. Zou Y, Chang SKC, Gu Y, Qian SY. Antioxidant activity and phenolic compositions of lentil (lens
culinaris var. morton) extract and its fractions. J Agric Food Chem. 2011;59(6):2268-2276.
18. Borrelli RC, Visconti A, Mennella C, Anese M, Fogliano V. Chemical characterization and
antioxidant properties of coffee melanoidins. J Agric Food Chem. 2002;50(22):6527-6533.
19. Jaiswal R, Kuhnert N. Hierarchical scheme for liquid chromatography/multi-stage spectrometric
identification of 3,4,5-triacyl chlorogenic acids in green robusta coffee beans. Rapid Communications in
Mass Spectrometry. 2010;24(15):2283-2294.
20. Clifford MN, Johnston KL, Knight S, Kuhnert N. Hierarchical scheme for LC-MSn identification of
chlorogenic acids. J Agric Food Chem. 2003;51(10):2900-2911.
21. Kuhnert N, Jaiswal R, Matei MF, Sovdat T, Deshpande S. How to distinguish between feruloyl quinic
acids and isoferuloyl quinic acids by liquid chromatography/tandem mass spectrometry. Rapid
Communications in Mass Spectrometry. 2010;24(11):1575-1582.
112
22. Jaiswal R, Matei MF, Subedi P, Kuhnert N. Does roasted coffee contain chlorogenic acid lactones
or/and cinnamoylshikimate esters? Food Res Int. 2014;61:214-227.
113
Supplementary table 1. List of chlorogenic acids
No Name Abb R1 R3 R4 R5 R1’ 1c 3-Caffeoylshikimic acid 3-CSA 2db C H H H 1d 4-Caffeoylshikimic acid 4-CSA 2db H C H H 1e 5-Caffeoylshikimic acid 5-CSA 2db H H C H 2a 1-Caffeoyl-1,5-quinide 1-CQL 5C H H 1 H 2c 3-Caffeoyl-1,5-quinide 3-CQL 5 C H 1 H 2d 4-Caffeoyl-1,5-quinide 4-CQL 5 H C 1 H 3c 3-p-Coumaroylquinic acid 3-CoQA H Co H H H 3d 4-p-Coumaroylquinic acid 4-CoQA H H Co H H 3e 5-p-Coumaroylquinic acid 5-CoQA H H H Co H 4c 3-Caffeoylquinic acid 3-CQA H C H H H 4d 4-Caffeoylquinic acid 4-CQA H H C H H 4e 5-Caffeoylquinic acid 5-CQA H H H C H 5c 3-Feruloylquinic acid 3-FQA H F H H H 5d 4-Feruloylquinic acid 4-FQA H H F H H 5e 5-Feruloylquinic acid 5-FQA H H H F H 6c 3-Caffeoylquinic acid methyl ester M-3-QA H C H H CH3 6d 4-Caffeoylquinic acid methyl ester M-4-QA H H C H CH3 6e 5-Caffeoylquinic acid methyl ester M-5-QA H H H C CH3 7c 3-Dimethoxycinnamoylquinic acid 3-DQA H D H H H 7d 4-Dimethoxycinnamoylquinic acid 4-DQA H H D H H 7e 5-Dimethoxycinnamoylquinic acid 5-DQA H H H D H 8u Caffeoyl-coumaroylquinic acid CCoQA [C Co H H]* H 9u Coumaroyl-feruloylquinic acid CoFQA [Co F H H]* H
10cd 3,4-diCaffeoylquinic acid 3,4-diCQA H C C H H 10ce 3,5-diCaffeoylquinic acid 3,5-diCQA H C H C H 10de 4,5-diCaffeoylquinic acid 4,5-diCQA H H C C H 11u Caffeoyl-feruloylquinic acid CFQA [C F H H]* H 12u diCaffeoylquinic acid methyl ester M-diCQA [C C H H]* CH3 13u Dimethoxycinnamoyl-feruloylquinic acid DFQA [D F H H]* H * ‘u’ is used for not confirmed isomeric positions in this paper, unconfirmed and ‘i’ is
used for stereoisomers. The stereoisomeric position numbering is shown in
Supplementary figure 1.
114
R3OOR4
R1O
OR5
O
OR1'
Supplementary figure1. Stereoisomeric positions of quinic acid
115
HOOH
O
O
OH
OH
OHO
1e
OOH
OH
O
HO
HO
OHO
1c
OH
O
OH O
OH
OHHO
O
1d
HO
OOH
O
HO
HOO
O
2c
HO
OH
O
O
OH
OH
OO
2d
O
OHHO
OO
O
OH
OH
2a
HO
HOOH
O
OHO
O
OH
HO
OOH
OH
OHO
O
HO
HO
OH
O
OHOH
O
O
OH
3d 3e3c
HO
HOOH
O
OHO
O
OH
OH
4e
HO
OOH
OH
OHO
O
HO
HO
4c
HO
OH
O
OHOH
O
O
OH
OH
4d
HO
HOOH
O
OHO
O
OH
O
HO
OOH
OH
OHO
O
HO
O HO
OH
O
OHOH
O
O
OH
O
5d 5e5c
HO
HOOH
O
OO
O
OH
OH
6e
HO
OOH
OH
OO
O
HO
HO
6c
HO
OH
O
OHO
O
O
OH
OH
6d
HO
OOH
OH
OHO
O
O
O HO
OH
O
OHOH
O
O
O
O
7d7c
HO
HOOH
O
OHO
O
O
O
7e
Supplementary figure2. Chemical structures of mono-chlorogenic acids
116
HO
OOH
O
OHO
O
OH
OH
10ce
O
HO
HO
HO
OO
OH
OHO
O
HO
HO
O
OHHO
10cd
HO
HOO
O
OHO
O
OH
OH
O
OHHO
10de
Supplementary figure3. Chemical structures of dichlorogenic acids
117
Supplementary table 2. Chlorogenic acids in Brazil green bean coffee interpretations
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
3e 5-pCoQA 337.03 190.8 191.7 8 162.8 5 126.8 93.1 34 172.8 29 85.2 23
4c 3-CQA 352.46 190.8 178.8 44 176.8 34 85.1 172.7 71 126.9 66 108.9 34
4d 4-CQA 352.98 172.8 178.8 46 190.8 17 93.1 110.9 93 154.8 47 71.3 26
4e 5-CQA 352.36 190.8 214.8 30 176.8 5 126.9 85.2 100 172.8 67 93.1 49
5c 3-FQA 366.99 192.7 133.9 14 193.7 10 133.8 148.8 24 134.7 3 116.9 1
5d 4-FQA 367.01 172.7 190.8 13 192.7 12 93 110.9 58 154.7 33 71.3 25
5e 5-FQA 367.02 190.8 191.7 7 192.8 6 126.8 85.2 61 172.8 60 93.1 49
10ce 3,5-diCQA 515.02 353 351.1 33 353.9 9 190.7 178.7 53 176.8 15 134.9 8
10de 4,5-diCQA 514.46 353 351 70 335 18 172.7 178.8 59 194.7 34 190.8 28
11 5-C-3-FQA 529 367 353 28 351 19 192.7 190.8 24 133.8 12 172.8 11
11 C-4-FQA 529.05 367 349 54 172.8 42 172.8 192.8 35 190.7 5 133.7 5
118
Supplementary table 3. Chlorogenic acids in Brazil fermented bean coffee interpretations
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
2c 3-CQAL 334.99 160.7 134.9 82 178.7 14 132.8 - - - - - -
2d 4-CQAL 334.97 160.8 314.9 17 134.9 15 132.9 - - - - - -
3c 3-pCoQA 337.03 162.8 163.8 11 190.8 9 118.9 - - - - - -
3d 4-pCoQA 337.03 172.8 173.7 9 162.8 7 93 110.9 62 - - - -
3e 5-pCoQA 337 190.8 162.8 7 191.7 7 93.1 126.9 50 87.1 30 85.1 29
4c 3-CQA 352.97 190.8 178.8 50 176.8 32 126.9 85.2 70 93.1 58 172.8 54
4ci 3-CQA de 352.96 190.8 178.8 59 176.8 17 - - - - - - -
4ci 3-CQA de 352.97 190.8 178.8 57 176.8 21 93.1 110.9 65 126.8 51 172.7 49
4d 4-CQA 353 172.8 178.8 43 190.8 15 93.1 110.9 47 71.4 14 154.8 12
4e 5-CQA 352.25 190.7 214.8 45 194.7 7 126.9 85.1 66 172.7 55 93.1 46
4ei 5-CQA de 352.96 190.8 214.8 51 176.8 15 86.2 - - - - - -
4ei 5-CQA de 352.98 190.8 178.8 10 306.9 8 85.1 126.8 40 110.9 30 - -
5c 3-FQA 367.02 192.8 133.9 16 193.7 10 133.8 148.8 22 134.8 3 149.8 3
5d 4-FQA 367.03 172.7 192.8 14 173.6 9 93 110.9 30 71.3 18 154.7 11
5di 4-FQA de 366.99 172.8 347.9 17 192.8 14 71.3 136.8 36 154.8 30 93.1 23
5e 5-FQA 367 190.8 191.7 7 192.8 6 126.9 172.8 74 85.2 71 110.9 57
10ce 3,5-diCQA 515.04 352.9 351.1 35 190.8 11 190.7 178.8 53 134.9 14 132.9 12
10de 4,5-diCQA 514.46 353 351 77 172.8 19 172.7 178.7 67 176.8 31 190.8 24
119
Supplementary table 4. Chlorogenic acids in Brazil handdrip roasted bean coffee interpretation
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1c 3-CSA 335.03 178.8 160.8 66 134.9 41 134.8 132.8 6 - - - -
1e 5-CSA 334.99 172.7 160.8 29 178.7 21 93.1 110.9 93 154.8 62 71.3 34
2a 1-CQAL 334.97 160.8 172.7 83 315.9 12 132.8 - - - - - -
2c 3-CQAL 334.98 160.7 134.9 93 178.7 12 132.8 - - - - - -
2d 4-CQAL 334.99 160.8 134.8 18 178.7 17 132.8 - - - - - -
3e 5-pCoQA 337.02 190.8 160.8 9 134.8 9 126.8 172.8 84 85.2 66 110.9 49
4c 3-CQA 352.98 190.8 178.8 44 176.7 35 126.8 93.1 60 85.1 52 170.7 43
4ci 3-CQA de 352.99 190.8 178.8 59 134.8 20 71.4 - - - - - -
4ci 3-CQA de 353 178.8 172.8 30 134.9 23 134.8 132.8 83 133.8 6 135.8 4
4ci 3-CQA de 353.02 190.8 178.8 73 176.8 22 93.1 - - - - - -
4d 4-CQA 352.98 172.8 178.8 46 190.8 18 93.1 110.9 35 154.7 20 71.3 16
4di 4-CQA de 352.95 172.8 178.8 39 190.8 36 154.8 93.1 97 - - - -
4e 5-CQA 352.42 190.8 214.8 38 176.8 7 85.2 126.9 98 93.1 73 172.7 52
4ei 5-CQA de 352.46 190.7 214.8 20 172.8 19 126.9 172.7 80 85.2 50 170.8 46
4ei 5-CQA de 353 190.8 214.8 34 176.7 6 172.8 85.1 60 93.1 58 170.8 57
5c 3-FQA 367.01 192.7 133.9 12 193.7 8 133.8 148.8 21 116.9 6 134.8 4
5d 4-FQA 367.04 172.7 192.8 12 173.7 7 93.1 111 39 154.8 39 71.4 25
5e 5-FQA 367.01 190.8 191.7 8 192.8 6 126.9 85.1 94 93 52 172.7 46
6c M-3-CQ 367 160.7 192.8 16 132.9 11 132.8 - - - - - -
6d M-4-CQ 366.99 160.8 134.9 36 192.7 10 132.9 - - - - - -
12 M-diCQA 529.03 367 160.8 20 368 19 160.8 134.8 26 132.9 14 192.7 11
120
Supplementary table 5. Chlorogenic acids in Brazil espresso roasted bean coffee interpretation
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1c 3-CSA 335.01 178.7 160.8 70 134.9 41 134.8 133.8 6 135.8 6 132.8 4
1e 5-CSA 334.98 172.8 160.8 27 178.8 24 110.9 93 73 154.7 31 108.9 12
2c 3-CQAL 334.99 160.8 134.9 78 161.8 10 132.8 - - - - - -
2d 4-CQAL 334.98 160.7 178.8 29 134.9 24 132.8 - - - - - -
3d 4-pCoQA 336.97 172.7 162.8 6 173.7 6 110.9 93 48 98.9 24 - -
3e 5-pCoQA 336.96 190.8 134.9 7 160.7 6 84.9 172.8 57 87.2 53 - -
4c 3-CQA 352.97 190.8 178.8 46 176.8 35 85.2 172.8 69 110.9 52 126.9 47
4ci 3-CQA de 352.99 190.8 178.7 75 134.8 23 170.8 - - - - - -
4ci 3-CQA de 352.98 178.7 172.8 28 176.8 22 134.8 132.8 94 133.8 9 135.8 7
4d 4-CQA 352.47 172.8 178.8 51 190.8 24 93.1 110.9 69 154.7 16 59.6 13
4di 4-CQA de 352.97 172.8 178.8 44 190.8 42 110.9 93 89 154.7 36 - -
4e 5-CQA 352.47 190.7 214.8 40 176.8 7 85.1 172.8 69 126.8 62 93 59
4ei 5-CQA de 352.41 190.7 214.8 20 172.8 19 126.8 172.8 42 108.9 32 85.1 28
4ei 5-CQA de 352.98 190.8 214.8 30 172.8 7 110.9 172.8 93 126.9 86 93 79
5c 3-FQA 367.04 192.8 133.9 13 193.7 9 133.8 148.8 16 134.7 5 116.9 2
5d 4-FQA 367.02 172.8 192.7 10 173.7 7 93.1 154.8 40 110.9 33 71.3 24
5e 5-FQA 367.01 190.7 191.7 6 192.8 6 126.8 172.8 90 85.2 78 93.1 67
6c M-3-CQ 367.02 160.7 192.7 16 134.9 12 132.9 - - - - - -
6d M-4-CQ 367.03 160.8 134.9 38 132.9 10 132.8 - - - - - -
10cd 3,4-diCQA 514.52 353 335 77 338.9 57 190.7 172.7 100 178.7 69 191.9 35
10ce 3,5-diCQA 515.04 353 351 30 353.9 11 190.8 178.7 42 176.7 17 134.8 10
12 M-diCQA 529.04 367 367.9 18 160.8 17 160.7 134.8 15 192.8 9 132.8 8
121
OOH
HO
OH
OOH
HOOH
HO
O
OOHHO
OOH
HO
OOH
O
HO
OO
O
OHO
HO
O
OOH
HO
OH
OO
HOOH
HO
O
OOHO
OOH
HO
OO
O
HO
HOO
OH
OHO
HO
HO
5c 5d 5e
6c 6d 6e
Supplementary figure 4. Chemical structures of the peaks in EICs at m/z 367
122
Supplementary figure 5. MS2 spectra of espresso roasted coffee bean from EIC at m/z
367
6d
6c
5c
6e
5e
5d
123
HOOH
O
O
OH
OH
OHO
1e
OOH
OH
O
HO
HO
OHO
1c
OH
O
OH O
OH
OHHO
O
1d
HO
OOH
O
HO
HOO
O
2c
HO
OH
O
O
OH
OH
OO
2d
O
OHHO
OO
O
OH
OH
2a
Supplementary figure 6. Chemical structures of the peaks in EICs at m/z 335
124
Supplementary figure 7. MS2 spectra of handdrip roasted coffee bean from EIC at m/z
335
1e
1d
2c
1a
2d
125
Research on Brewing Step 3.3.
Research on Brewing Step with Different Solvents 3.3.1.
Characterization of the chlorogenic acids composition with different extracting solvents
Seung-Hun Lee, Nikolai Kuhnert*
Department of Life Sciences and Chemistry, Jacobs University Bremen, 28759 Bremen,
Germany
*Correspondence to:
Prof. Dr. Nikolai Kuhnert,
Department of Life Sciences and Chemistry
Jacobs University Bremen gGmbH
Campus Ring 1
28759 Bremen, Germany
Tel: +49 421 200 3120
Fax: +49 421 200 3102
E-mail: [email protected]
126
Abstract
Methanol was used as an extracting solvent for analyzing coffee due to its low viscosity,
price and customs of food analysis. The real consuming compounds are analyzed in
addition to a large amount of scientific coffee results on account of knowing the
different solubility between water and methanol extracted coffee. Nevertheless,
scientific analysis of coffee is still processing methanol extraction. In this report, we
interpreted various chlorogenic acids and confirmed chemical reactions during roasting
time with known methods by analyzing tandem MS spectra of both water and methanol
extracts of green and roasted Arabica Colombia Supremo. Furthermore, we identified
the differences of relative composition between water and methanol extracts, focused on
seven major compounds’ relative intensities against 5-CQA in extracted ion
chromatograms.
Keywords: Tandem MS; Chlorogenic acid; Coffee extracting solvent
127
Abbreviations used
QA, quinic acid; CA, caffeic acid; CQA, caffeoylquinic acid; FQA, feruloylquinic acid;
CQAL, caffeoylquinic acid lactone; CoQA, coumaroylquinic acid; M-CQ, methyl-
caffeoylquinate; CFQA, caffeoyl-feruloylquinic acid.
128
1. Introduction
Coffee is one of the most consumed beverages in the world. In order to get the best way
for the taste of coffee, lots of analysis on all the coffee processes have been studied;
species1, coffee production2, roasting3-5 and brewing6. Furthermore, not only analyzing,
but also in-depth chemical studies into the chemical reactions during roasting process
are proceeding briskly7.
A lot of analysis tools are imported to analyze coffee components; HPLC8,9, GC4,10, LC-
MS11,12, FT-IR13,14, UV-Vis Spectrometry15, NMR16, and so on. After the study of
chlorogenic acids by tandem-MS technique17, LC-MSn became the mainstream for
interpreting isomeric p-coumaric acid derivatives. This research also proceeded with
tandem-MS for verifying peaks of isomeric polyphenols.
Chemical analysis of food has been processed with methanolic extracts. Most of the
published coffee researches also used methanol as an extracting solvent in order to
extract as much components as possible. However, especially for the coffee and tea, not
like other food, we rarely eat them, but drink only water extracted part. Although
practical research are done with water extract by governmental organizations, such as
Food and Drug Administration, scientific views are still focused on methanolic extracts,
with regarding coffee and tea as other food.
This trend caused unnecessary double research. Although there already are a lot of well-
interpreted scientific results done in laboratories, additional experiments were needed
when people wanted to know the data of real-consumed coffee. Thus, this research
129
ascertain the effect of extracting solvent, identify the difference between methanol and
water extracted compounds for opening the possibility to apply existing methanolic
scientific data to watery real-consuming data, and help people plan their further research
to be easily applicable to the real consumption.
2. Experimental
The experimental methods of HPLC and LC/MSn are followed by previous publications
of our group research1,7,11.
2.1. Chemicals and Coffee Beans
All the chemicals (Analytical grade) were purchased from Sigma-Aldrich (Germany).
Green Colombia Supremo Arabica coffee bean was purchased from Lloyd Caffee in
Bremen, Germany.
2.2. Roasting
Colombia Supremo Arabica green coffee bean (100 g) was roasted with a Gene Café
Coffee Roaster 101A (Genesis Co. Kyungki, Korea), set temperature 240 ℃ for 15 min.
2.3. Extract of Coffee Beans
Colombia Supremo Arabica coffee beans (5 g per each sample, Green bean and 240 ℃,
15 min roasting) were ground to fine powder; methanolic extracts were prepared by
soxhlet extraction using aqueous methanol (70 %, 170 mL) for 3 hours and water
extracts were prepared by stirring in 100 ℃ water (200 mL) for 1 hour in 250 mL
beaker. The extract was treated with Carrez reagents (1 mL of reagent A plus 1 mL of
130
reagent B) to precipitate colloidal material and filtered through a Whatman no. 1 filter
paper and stored at -20 ℃ until required, thawed at room temperature, diluted to 60 mg /
10 mL, filtered through a membrane filter, and used for LC/MS.
2.4. LC/MSn
The LC equipment (Agilent 1100 series, Karlsruhe, Germany) comprised a binary pump,
an autosampler with a 100 µL loop, and a diode-array detector (DAD) with a light-pipe
flow cell (recording at 320 and 254 nm and scanning from 200 to 600 nm). This was
interfaced with an ion-trap mass spectrometer fitted with an electrospray ionization (ESI)
source (Bruker Daltonics HCT ultra, Bremen, Germany) operating in full scan, auto
MSn mode (smart fragmentation) using a ramping of the collision energy. Maximum
fragmentation amplitude was set to 1 V, starting at 20 % and ending at 200 %. MS
operating conditions (negative mode) had been optimized using 5-caffeoylquinic acid
with a capillary temperature of 365 ℃, a drying gas flow rate of 10 L / min, and a
nebulizer pressure of 10 psi.
2.5. HPLC
Separation was achieved on a 150 × 3 mm i.d. column containing diphenyl 5 µm, with a
5 mm × 3 mm i.d. guard column (Varian, Darmstadt, Germany). Solvent A was water /
formic acid (1000:0.005 v / v) and solvent B was methanol. Solvents were delivered at a
total flow rate of 500 µL / min. The gradient profile was from 10 to 70 % B linearly in
60 min followed by 10 min isocratic and a return to 10 % B at 90 and 10 min isocratic
to re-equilibrate.
131
3. Results and Discussion
3.1. Interpretations of chlorogenic acids
Interpretation methods of chlorogenic acids by LC/MSn have been already studied
especially in p-coumaroyl quinic acid derivatives; caffeoyl quinic acids17,
Feruloylquinic acids18, and others19.
Compounds were interpreted in four different spectra as shown in Table 1 ~ 4 with their
Total MS data in Figure 1.
Figure 1. Total MS spectra of coffee beans in four different conditions
132
Table 1. Colombia Supremo Green Bean Water Extraction Interpretations
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1 Sucrose 341.0 178.6 160.6 22 112.8 15 89 160.6 85 142.6 77 118.8 62
2 QA 190.6 126.8 85.0 99 172.6 74 108.8 98.8 16
3 QA ep. 190.6 110.6 172.6 37 67.4 -
4 3-CQA 352.8 190.6 178.6 41 134.8 9 126.6 172.6 69 85.0 65 110.8 31
5 5-CQA 352.8 190.6 - 126.8 172.6 100 85.0 62 93.0 38
6 3-FQA 366.8 192.6 133.8 11 133.6 148.6 34
7 4-CQA 352.8 172.6 178.6 57 190.6 15 92.8 110.8 87 154.6 40 71.2 24
8 5-CoQA 336.8 190.6 - 126.6 85.0 78 172.6 58 110.8 42
9 5-FQA 366.8 190.6 - 126.6 172.6 86 85.0 78 93.0 49
10 4-CoQA 336.8 172.6 162.6 7 190.6 4 110.8 71.2 27 154.6 26 81.0 21
11 4-FQA 366.8 172.6 192.6 13 92.8 110.8 89 71.2 34 154.6 28
12 3,5-diCQA 514.8 352.8 190.8 8 190.4 178.8 40 134.8 7
13 3,4-diCQA 514.8 352.8 172.6 23 334.8 19 172.6 178.6 65 190.6 63 134.8 8
14 4,5-diCQA 514.8 352.8 172.8 16 202.6 16 172.6 178.6 62 190.6 28 134.8 10
Table 2. Colombia Supremo Green Bean Methanol Extraction Interpretations
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1 Sucrose 376.8 340.8 214.6 15 178.6 160.6 19 112.8 16 142.6 16
2 QA 190.6 126.6 172.6 100 85.0 99 108.8 98.8 16 83.0 14
3 QA ep. 190.6 110.8 172.6 30 67.2 -
4 3-CQA 352.8 190.6 178.6 45 134.8 9 172.6 126.6 92 85.0 78 110.8 55
5 5-CQA 352.8 190.6 - 126.6 172.6 71 85.0 64 93.0 43
6 3-FQA 366.8 192.6 133.8 11 133.6 148.6 28
7 4-CQA 352.8 172.6 178.6 59 190.6 17 93.0 110.8 94 154.6 56 71.2 45
8 5-CoQA 336.8 190.6 162.8 5 126.6 172.6 81 85.0 63 93.0 61
9 5-FQA 366.8 190.6 - 126.8 172.6 100 85.0 73 93.0 50
10 4-CoQA 336.8 172.6 162.6 6 93.0 71.2 71 110.8 71 154.6 24
11 4-FQA 366.8 172.6 192.6 12 93.0 110.8 75 71.2 52 154.6 14
12 3,5-diCQA 514.8 352.8 190.6 11 190.6 178.6 44 134.8 9
13 3,4-diCQA 514.8 352.8 172.8 25 334.8 19 172.6 178.6 83 190.6 65 134.8 9
14 4,5-diCQA 514.8 352.8 202.6 17 172.6 17 172.6 178.6 70 190.6 31 134.8 6
15 CFQA 528.8 366.8 352.8 22 192.8 13 192.6 172.6 13 133.6 11
16 CFQA 528.8 366.8 172.6 44 352.8 38 172.6 192.6 65
17 CFQA 528.8 352.8 266.8 95 172.6 27 172.6 178.6 61 190.6 30 134.6 8
133
Table 3. Colombia Supremo Roasted Bean Water Extraction Interpretations
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1 Sucrose 377.0 340.8 - 178.6 160.8 28 142.8 23 112.8 20
2 QA 190.8 126.8 172.6 52 110.8 42 108.6 85.0 94
3 Pentose-QA 323.0 190.8 172.8 26 93.0 6 126.8 85.0 65 172.8 47 111.0 27
4 QA de. 364.8 190.6 172.6 5 -
5 QA ep. 190.8 110.8 172.6 24 67.4 -
6 QA de. 364.8 172.6 110.8 97 323 50 110.8 -
7 3-CQA ep. 353.0 178.6 134.8 33 172.8 16 134.8 -
8 3-CQA 352.8 190.6 178.8 46 134.8 13 126.8 172.8 61 110.8 48 85.0 37
9 5-CQA ep. 352.8 190.6 - 126.8 85.2 53 172.6 45 109.0 29
10 3-CQA ep. 353.0 190.6 178.6 75 134.8 27 172.6 93.0 54 126.8 44 110.8 25
11 5-CQA 352.8 190.6 - 85.0 172.6 87 126.8 83 93.0 78
12 3-FQA ep. 367.0 192.6 172.8 39 133.8 18 133.8 148.8 26 177.6 8
13 CQAL 335.0 178.6 160.8 51 134.8 37 134.8 -
14 3-FQA 367.0 192.8 133.8 19 133.8 148.8 21
15 4-CQA 352.8 172.8 178.8 63 190.8 19 93.0 111.0 78 154.6 35 71.4 31
16 5-FQA ep. 367.0 190.6 172.8 6 108.8 126.8 91 172.6 70 85.0 69
17 5-CoQA 337.0 190.6 172.8 5 172.6 110.8 88 126.8 47 85.2 33
18 3-CQAL 335.0 160.6 134.8 87 178.6 10 132.8 -
19 CQAL 335.0 160.6 178.8 100 210.8 78 132.8 -
20 5-FQA 367.0 190.6 - 85.2 172.8 66 126.8 58 93.0 42
21 CQAL 335.0 172.6 178.6 13 134.8 8 110.8 136.8 46 154.8 45 93.0 40
22 4-CQAL 335.0 160.6 134.8 26 290.8 17 132.8 -
23 CQAL ep. 335.0 160.6 172.8 51 132.8 11 93.0 110.8 49 71.2 47 154.6 13
24 4-FQA 367.0 172.6 192.6 12 93.0 110.8 54 154.8 27 81.2 16
25 diCQA ep. 515.0 340.8 280.8 86 190.8 77 280.8 296.8 53 254.8 19 322.8 7
26 diCQA ep. 515.0 340.8 322.8 62 280.8 59 296.8 280.8 60 254.8 10 324.0 9
27 3,5-diCQA 515.0 353.0 340.8 16 190.8 9 190.8 178.8 41 134.8 15
28 3,4-diCQA 515.0 353.0 172.8 26 178.8 21 172.8 178.6 77 190.6 53 134.8 31
29 4,5-diCQA 515.0 352.8 172.8 20 202.8 11 172.8 178.6 58 190.8 26 134.8 9
30 CFQA 529.0 367.0 172.6 39 349.0 26 172.6 192.6 25 133.8 9
31 CFQA 529.0 352.8 366.8 85 172.6 20 172.6 190.6 34 178.8 30 135.0 12
134
Table 4. Colombia Supremo Green Bean Water Extraction Interpretations
No Compd.
MS1 MS2 MS3
Parent ion base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1 Sucrose 341.0 178.6 160.8 38 142.8 30 142.8 89.0 97 118.8 63 160.8 58
2 diCA 322.8 304.8 142.8 24 160.6 15 124.8 214.6 25 142.6 12 101.0 12
3 QA 190.6 126.8 172.8 49 110.8 32 108.8 99.0 28 71.2 7
4 Pentose-QA 322.8 190.6 172.6 66 110.8 8 172.4 93.0 96 110.8 61 126.8 45
5 QA de. 365.0 190.6 172.8 3 172.6 87.0 68 126.8 41 108.4 21
6 QA ep. 190.6 110.8 172.6 28 -
7 3-CQA ep. 352.8 190.8 178.6 86 134.8 24 172.8 93.0 81 136.8 24 126.8 13
8 3-CQA ep. 353.0 178.6 134.8 27 190.8 19 134.8 -
9 3-CQA 353.0 190.6 178.8 38 134.8 15 126.8 85.0 76 172.6 61 93.0 55
10 3-FQA ep. 367.0 334.8 160.6 52 178.6 49 172.8 178.8 94 160.6 82 134.8 74
11 5-CQA ep. 353.0 190.6 - 126.8 172.8 55 85.0 31 144.8 24
12 3-CQA ep. 352.8 190.6 178.6 88 134.8 23 93.0 172.6 64 111.0 38 126.8 19
13 5-CQA 352.8 190.6 178.8 4 126.8 172.6 66 93.0 64 85.0 44
14 M-3-CQ 367.0 160.6 192.6 11 132.8 11 132.8 105.0 35 116.6 4
15 3-CQA ep. 353.0 190.6 178.8 83 134.8 20 127.0 172.8 89 162.8 72 93.6 48
16 3-FQA ep. 367.0 192.6 172.6 42 133.8 15 133.8 148.8 27 177.6 16 116.8 5
17 CQAL 334.8 178.6 160.6 60 134.8 41 134.8 -
18 3-FQA 367.0 192.6 133.8 11 133.8 148.8 19
19 3-FQA ep. 367.0 334.8 192.8 31 160.6 26.6 178.6 160.6 44 134.8 31 254.8 26
20 4-CQA 353.0 172.6 178.6 58 134.8 13 93.0 110.8 84 154.8 30 71.4 29
21 5-FQA ep. 367.0 190.6 172.8 6 172.6 126.8 100 108.8 46 85.0 28
22 5-CoQA 337.0 190.6 162.8 9 178.8 6 85.0 126.8 63 172.8 44 108.8 44
23 3-CQAL 335.0 160.6 134.8 81 178.8 10 132.8 105.0 5
24 M-4-CQ 367.0 160.6 134.8 40 192.8 10 132.8 -
25 CQAL ep. 335.0 160.8 178.8 78 254.8 52 158.6 132.8 53
26 CQAL ep. 335.0 172.6 178.8 16 160.8 9 93.0 110.8 77 59.4 25 136.8 24
27 5-FQA 366.8 190.6 192.8 6 126.8 172.8 90 85.0 83 93.0 74
28 4-CQAL 334.8 160.8 134.8 30 178.8 23.8 132.8 -
29 CQAL ep. 335.0 160.6 172.8 42 178.6 7 132.8 -
30 4-FQA 367.0 172.6 192.6 8 93.0 110.8 46 71.2 24 154.6 17
31 diCQA ep. 515.0 340.8 280.8 96 190.6 69 296.8 280.8 59 254.8 12 178.8 7
32 diCQA ep. 515.0 352.8 190.8 52 341.0 32 190.6 178.6 12 134.8 10 172.8 4
33 3,5-diCQA 515.0 352.8 190.8 6 340.8 6 190.6 178.8 51 134.8 13 172.6 4
34 3,4-diCQA 515.0 352.8 172.6 25 178.6 20 172.6 178.6 53 190.8 37 134.8 15
35 CFQA 529.0 367.0 160.8 10 420.0 4 160.6 134.8 30 192.6 14 335.0 6
36 4,5-diCQA 515.0 352.8 172.8 16 202.6 14 172.8 178.8 63 190.8 29 134.8 13
37 CFQA 529.0 367.0 353.0 42 192.8 13 192.6 172.6 6 133.8 5
38 CFQA 529.0 367.0 172.8 56 352.8 42 172.8 192.6 69 110.8 6 119.8 6
39 CFQA 529.0 352.8 367.0 83 172.6 24 172.6 178.6 56 190.6 42 308.8 7
135
3.2. Interpretations of roasted bean data
As already published, a lot of reactions are occurring during the roasting period7. We
can easily find peak forming or increasing after roasting by comparing Figure 1-A and
1-C. Figure 2 shows three different roasting compounds which showed significant peak
increments; epimerization (Fig. 1-C-9), lactonisation (Fig. 1-C-18 and 1-C-22), and
acyl-migration (Fig. 1-C-24).
HOOH
OHO
O
OH
OHOH
O
HOOH
OHO
O
OH
OHOH
O
HOOH
OHO
O
OHOO
OHHO
O
OHOH
OOH
O
HO
Figure 2. Chemical transformations occur during roasting process: (a) epimerization, (b) lactonisation, (c) acyl-migration
3.3. Differences between water and aqueous methanol (70 %) extracts
As shown in Figure 1, meaningful composition differences are found between water
and aqueous methanol (70 %) extracts data; 5-FQA and three di-CQAs give significant
differences, together with other compounds.
a
b c
136
Table 5. Relative ratios against 5-CQA (100 %) of major compounds
Compd Green / Water Green / MeOH Roasted /
Water Roasted / MeOH
5-FQA 12.78 27.09 22.75 36.08 3,5-diCQA 5.47 10.18 2.10 4.42 3,4-diCQA 2.85 8.29 2.71 6.61 4,5-diCQA 4.83 17.05 2.49 8.86 M-3-CQ - - 0.77 10.84 3-CQAL - - 27.34 39.28 4-CQAL - - 24.23 26.92
Table 5 showed seven different compounds which show decided intensity differences.
The intensities were measured in the extracted ion chromatograms (EICs) of 353 (5-
CQA, standard, 100 %), 367 (5-FQA and M-3-CQ), 515 (di-CQAs), and 335 (CQALs).
In all seven shown compounds, relative intensities against 5-CQA show meaningful
increase in methanol extracts than water extracts.
4. Conclusion
In conclusion, we analyzed Arabica Colombia Supremo coffee beans with liquid
chromatography / tandem mass spectrometry in four different conditions; water
extracted green bean, methanol extracted green bean, water extracted 240 ℃ 15 min
roasted bean, and methanol extracted 240 ℃ 15 min roasted bean. We interpreted
polyphenolic compounds in Total MS spectra, and reconfirmed some published
chemical reactions during coffee roasting steps. Most importantly, we verified for the
first time, the effect of coffee extracting solvents between water and methanol by
comparing major compounds’ detected intensities of relative quantities against 5-CQA.
137
We expect this result would help further research directly reflect the real coffee
consumption in order to fill up a gap between academic and practical researches.
Acknowledgements
Excellent technical support by Anja Müller and providing coffee beans from Lloyd
Caffee are acknowledged.
References
1. Kuhnert, N.; Jaiswal, R.; Eravuchira, P.; El-Abassy, R.M.; von der Kammer, B.; Materny, A. Scope and limitations of principal component analysis of high resolution LC-TOF-MS data: the analysis of the chlorogenic acid fraction in green coffee beans as a case study. Analytical Methods 2011, 3, 144-155.
2. Balyaya, K.J.; Clifford, M.N. Chlorogenic acids and caffeine contents of monosooned Indian Arabica and Robusta coffees compared with wet and dry processed coffees from the same geographic area. Colloq. Sci. Int. Cafe, [C. R. ] 1995, 16th, 316-325.
3. Moon, J.; Yoo, H.S.; Shibamoto, T. Role of Roasting Conditions in the Level of Chlorogenic Acid Content in Coffee Beans: Correlation with Coffee Acidity. J. Agric. Food Chem. 2009, 57, 5365-5369.
4. Moon, J.; Shibamoto, T. Role of Roasting Conditions in the Profile of Volatile Flavor Chemicals Formed from Coffee Beans. J. Agric. Food Chem. 2009, 57, 5823-5831.
5. Moon, J.; Shibamoto, T. Formation of Volatile Chemicals from Thermal Degradation of Less Volatile Coffee Components: Quinic Acid, Caffeic Acid, and Chlorogenic Acid. J. Agric. Food Chem. 2010, 58, 5465-5470.
6. Gross, G.; Jaccaud, E.; Huggett, A.C. Analysis of the content of the diterpenes cafestol and kahweol in coffee brews. Food and Chemical Toxicology 1997, 35, 547-554.
7. Jaiswal, R.; Matei, M.F.; Golon, A.; Witt, M.; Kuhnert, N. Understanding the fate of chlorogenic acids in coffee roasting using mass spectrometry based targeted and non-targeted analytical strategies. Food & Function 2012, 3, 976-984.
8. Ky, C.L.; Louarn, J.; Dussert, S.; Guyot, B.; Hamon, S.; Noirot, M. Caffeine, trigonelline, chlorogenic acids and sucrose diversity in wild Coffea arabica L. and C-canephora P. accessions. Food Chem. 2001, 75, 223-230.
138
9. Farah, A.; De Paulis, T.; Trugo, L.C.; Martin, P.R. Effect of roasting on the formation of chlorogenic acid lactones in coffee. J. Agric. Food Chem. 2005, 53, 1505-1513.
10. Semmelroch, P.; Grosch, W. Analysis of roasted coffee powders and brews by gas chromatography-olfactometry of headspace samples. Lebensmittel-Wissenschaft and Technologie 1995, 28, 310-313.
11. Jaiswal, R.; Patras, M.A.; Eravuchira, P.J.; Kuhnert, N. Profile and Characterization of the Chlorogenic Acids in Green Robusta Coffee Beans by LC-MSn: Identification of Seven New Classes of Compounds. J. Agric. Food Chem. 2010, 58, 8722-8737.
12. Matei, M.F.; Jaiswal, R.; Kuhnert, N. Investigating the Chemical Changes of Chlorogenic Acids during Coffee Brewing: Conjugate Addition of Water to the Olefinic Moiety of Chlorogenic Acids and Their Quinides. J. Agric. Food Chem. 2012, 60, 12105-12115.
13. Kemsley, E.K.; Ruault, S.; Wilson, R.H. Discrimination between Coffea-Arabica and Coffea-Canephora Variant Robusta Beans using Infrared-Spectroscopy. Food Chem. 1995, 54, 321-326.
14. Briandet, R.; Kemsley, E.K.; Wilson, R.H. Discrimination of Arabica and Robusta in instant coffee by Fourier transform infrared spectroscopy and chemometrics. J. Agric. Food Chem. 1996, 44, 170-174.
15. Fujioka, K.; Shibamoto, T. Chlorogenic acid and caffeine contents in various commercial brewed coffees. Food Chem. 2008, 106, 217-221.
16. Iwai, K.; Kishimoto, N.; Kakino, Y.; Mochida, K.; Fujita, T. In vitro antioxidative effects and tyrosinase inhibitory activities of seven hydroxycinnamoyl derivatives in green coffee beans. J. Agric. Food Chem. 2004, 52, 4893-4898.
17. Clifford, M.N.; Johnston, K.L.; Knight, S.; Kuhnert, N. Hierarchical scheme for LC-MSn identification of chlorogenic acids. J. Agric. Food Chem. 2003, 51, 2900-2911.
18. Kuhnert, N.; Jaiswal, R.; Matei, M.F.; Sovdat, T.; Deshpande, S. How to distinguish between feruloyl quinic acids and isoferuloyl quinic acids by liquid chromatography/tandem mass spectrometry. Rapid Communications in Mass Spectrometry 2010, 24, 1575-1582.
19. Jaiswal, R.; Kuhnert, N. Hierarchical scheme for liquid chromatography/multi-stage spectrometric identification of 3,4,5-triacyl chlorogenic acids in green Robusta coffee beans. Rapid Communications in Mass Spectrometry 2010, 24, 2283-2294.
139
Research on Brewing Step with Different Methods 3.3.2.
Brewing Less- and Enhanced-caffeinated Coffee by Using Cold Brew
Method
Seung-Hun Lee, Nikolai Kuhnert*
Department of Life Sciences and Chemistry, Jacobs University Bremen, 28759 Bremen,
Germany
*Correspondence to:
Prof. Dr. Nikolai Kuhnert,
Department of Life Sciences and Chemistry
Jacobs University Bremen gGmbH
Campus Ring 1
28759 Bremen, Germany
Tel: +49 421 200 3120
Fax: +49 421 200 3102
E-mail: [email protected]
140
Abstract
Background: Caffeine is the most attractive compound in coffee on account of its
remarkable pros and cons for human health. In contemporary human diet, coffee is the
main source for caffeine intake, so that controlling daily caffeine intake became an issue.
The only caffeine control method existing is decaffeination and it needs a lot of starting
setup expense so that it is allowed only for a major companies.
Methods: A modern coffee brewing method, cold brew method, was used and the
coffee was analyzed in chronological order to interpret sequential changes of extracted
compounds. Roasted coffee powder 50 g was used for extracting cold brew coffee and
samples were collected every 50 mL until 24 samples. HPLC-MS were used for
interpreting compounds and UV integration was used for measuring the quantitative
data.
Results: Total 26 compounds composed of chlorogenic acids of mono- and di-acylated
p-coumaric acid derivatives and caffeine in cold brewed coffee were interpreted. The
most important content change was caffeine in normalized coffee concentration.
Continuous increasing of caffeine content was detected while commercially used cold
brewing method; until 10 times of water to the coffee powder used.
141
Conclusions: Without any additional expense, max. ± 25 % caffeine content control
method is developed with cold-brew method. The LeeN-Caffeination method would be
expected as a standard content ratio control method in many different ways; different
compound control other than caffeine, different brewing method research other than
dripping cold-brew or different extracting materials other than coffee.
Keywords: Coffee; Cold brew coffee; Caffeine control; LeeN Caffeination; LC-MS
142
1. Introduction
It has been continuously emphasized the importance of coffee in 21st century. Although
the old myth started to spread from the book [Uncommon Grounds: The History of
Coffee and How it Transformed our World] by Mark Pendergrast, “Coffee is the second
most valuable exported legal commodity on earth after oil.”, 1 is now known as false, 2
at least rapid growth of coffee market became an important economic issue.
In spite of short history, both words, polyphenol and chlorogenic acids became well-
known to public by advertisements of global coffee companies to make coffee as a
healthy beverage. 3 In 2003, HPLC-tandem MS allowed us to distinguish the
regioisomers of caffeoylquinic acids. 4 After the research, interpretation of chlorogenic
acids became an important part of analyzing the differences in all the coffee processes
such as different origins, species, roasting degrees and brewing methods. On the other
hand, although there is still an animated controversy, caffeine is known to public as a
restriction-needed compound for consistent coffee consumption. 5 Most restrictions of
coffee per day have grounds on the calculation of caffeine intake per day. So the
decaffeinated coffee market is growing rapidly.
The movement of caffeine control in coffee started from 1903 by Ludwig Roselius in
Bremen, Germany. 6 Since then, a lot of decaffeination methods have been used for the
people who do not want to consume large amount of caffeine. Starting of benzene
extracting method by Ludwig Roselius, dichloromethane extraction 7 is used on account
of its low boiling point, and ethyl acetate extraction 8 is used with the name of natural
decaffeination because it exists in natural aromatic fruits. Supercritical carbon dioxide 9
143
is also used as an extracting solvent, not widely used owing to its high-cost, and the
most widely used method is the so called Swiss Water Process 10 for using water as an
extracting solvent. However, all these decaffeination methods have fundamental
limitation of high facility cost for the customers to control caffeine in their own
preference.
Actually, there’re not enough selections for customers on coffee from farm to table.
Coffee production is restricted in so called “Coffee Belt” by the weather and soil. Wet,
semi-dry and dry coffee processes to make green coffee bean from coffee cherry could
only be done near the farm by distance and size of facilities. Uniform roasting needs at
least more than 1 kg roaster which costs more than 10,000 Euro. The only choice
allowed for coffee consumers is to select the brewing method. Espresso, electric coffee
maker and instant coffee are the major part of brewing methods. However, as coffee
gourmets appear, more sincere brewing processes have been growing started from
Europe and USA, and spread from Korea and Japan; a variety of hand-brewing methods
and cold-brewing methods. 11,12
Cold brewing method has two different ways to brew coffee; steeping cold brew and
dripping cold brew. As the name cold brewing method says, people use 0 oC ~ RT water
to brew coffee. In the steeping cold brew method, water is poured into ground coffee,
the water/ground coffee mixture is stored for more than 12 hours and it is filtered to
make concentrated coffee. However, dripping cold brew method, as known as cold drip
or dutch brewing coffee, is a continuous method. It needs three parts, water reservoir,
ground coffee column and coffee bottle. From coffee reservoir water drops with the
144
speed of 1 drop per 1 - 5 seconds to pass through ground coffee column filled with
ground coffee to extract coffee and arrives at the bottom coffee bottle only with the
power of gravity. The commercial ratio of ground coffee versus used water is 1:5 to
1:10. The structure of dripping cold brew is shown in Figure 1.
Figure 1. Structure of dripping cold brew
In this research, dripping cold brew method is used for brewing coffee samples, and the
samples are divided in sequential order. Cold brewed kinetic samples were analyzed
using HPLC/tandem MS to interpret chlorogenic acids and caffeine content changes.
Quantification of caffeine content in kinetic samples was processed in order to separate
Less-caffeinated and Enhanced-caffeinated coffee samples to achieve Less- and
145
Enhanced-caffeine control method by Seung-Hun Lee and Nikolai Kuhnert (LeeN-
caffeination).
2. Experimental
The experimental methods of HPLC and LC/MSn are followed by previous publications
of our group research. 13-15
2.1. Chemicals, Coffee Beans and Brewing Materials
All the chemicals (Analytical grade) were purchased from Sigma-Aldrich (Germany).
Roasted Brazil Arabica coffee bean was purchased from Lloyd Caffee in Bremen,
Germany. Product name ‘Juliet’s Tears’, 500 mL cold brewing set, was provided by
Caferomeo (Seoul, Korea). Water for brewing coffee, product name ‘Mineral Wasser’,
was purchased from EDEKA in Bremen.
2.2. Brewing of Coffee Beans
50 g of roasted Brazil Arabica coffee bean was ground to fine powder and put in the
coffee column on the Kalita 56 mm coffee filter. After flatting with hands patting,
another coffee filter is put on the coffee powder for the even distribution of dropping
water. After the coffee column is filled, water drops with the speed of one drop per three
seconds from water reservoir to the coffee column. The water extracts coffee and passes
through the coffee column to the bottom sampling bottle. The sampling bottle is
changed in every 50 mL. The sampling is continued until 24th sample, 1200 mL in total
extraction has been collected. The 24 consecutive samples are shown in Figure 2.
146
Figure 2. Cold brew coffee samples 1 ~ 24
The extracts were stored at 4 ℃ until required, aged at room temperature, filtered
through a membrane filter, diluted to 5 % with 70 % aqueous methanol and used for
LC/MS. The concentrations of the extracted liquid were measured by evaporating
solvent at 100 ℃ for 30 min using aluminum boat to measure boat, boat with solution
before evaporation and boat with solid after evaporation.
2.3. High Resolution MS
The LC equipment (Agilent 1100 series, Karlsruhe, Germany) comprised a binary pump,
an autosampler with a 100 µL loop, and a diode-array detector (DAD) with a light-pipe
flow cell (recording at 320 and 254 nm and scanning from 200 to 600 nm). This was
interfaced with a MicroTOF Focus mass spectrometer (Bruker Daltonics, Bremen,
Germany) fitted with an electrospray ionization (ESI) source and internal calibration
was achieved with 10 mL of 0.1 M sodium formate solution injected through a six port
valve prior into each chromatographic run. Calibration was carried out using the
enhanced quadratic mode.
147
2.4. LC/MSn
The LC equipment comprised a binary pump, an autosampler with a 100 µL loop, and a
diode-array detector (DAD) with a light-pipe flow cell (recording at 320 and 254 nm
and scanning from 200 to 600 nm). This was interfaced with an ion-trap mass
spectrometer fitted with an ESI source (Bruker Daltonics HCT ultra, Bremen, Germany)
operating in full scan, auto MSn mode (smart fragmentation) using a ramping of the
collision energy. Maximum fragmentation amplitude was set to 1 V, starting at 20 %
and ending at 200 %. MS operating conditions (negative mode) had been optimized
using 5-caffeoylquinic acid with a capillary temperature of 365 ℃, a drying gas flow
rate of 10 L / min, and a nebulizer pressure of 10 psi.
2.5. HPLC
Separation was achieved on a 150 × 3 mm i.d. column containing diphenyl 5 µm, with a
5 mm × 3 mm i.d. guard column (Varian, Darmstadt, Germany). Solvent A was water /
formic acid (1000:0.005 v/v) and solvent B was methanol. Solvents were delivered at a
total flow rate of 500 µL / min. The gradient profile was from 10 to 80 % B linearly in
70 min followed by 10 min a return to 10 % B and 10 min isocratic to re-equilibrate.
3. Results and Discussion
3.1. Interpretations of chlorogenic acids
3.1.1. Molecular formula analysis of high resolution MS data
MicroTOF was used to obtain exact molecular weight to determine chemical formula of
chlorogenic acids in cold brew coffee. Figure 3 is the base peak chromatogram of the
first 50 mL samples (sample 1) of cold brew coffee by LC-TOFMS.
148
Figure 3. Base peak chromatogram of cold brew coffee by LC-TOFMS
The numbering of the compounds is shown in supporting information (supplementary
table 1, supplementary figure 1 and supplementary figure 2). Following Table 1
shows average molecular weight of different chlorogenic acids and caffeine in positive
mode found in the three repeated cold brew sample 1.
149
Table 1. High resolution mass data in the first 50 mL samples of cold brew coffee
No. RT Name Mol. Formula Exp. m/z (M+H) Theor. m/z (M+H) Error (ppm)
4ui 20.0 Caffeoylquinic acid (ep) C16H18O9 355.1017 355.1024 1.9
4ui 21.5 Caffeoylquinic acid (ep) C16H18O9 355.1009 355.1024 4.1
4c 23.9 3-Caffeoylquinic acid C16H18O9 355.1016 355.1024 2.1
4ui 26.6 Caffeoylquinic acid (ep) C16H18O9 355.1014 355.1024 2.7
4ui 27.5 Caffeoylquinic acid (ep) C16H18O9 355.1017 355.1024 1.9
4e 29.3 5-Caffeoylquinic acid C16H18O9 355.1021 355.1024 0.7
Caf 30.3 Caffeine C8H10N4O2 195.0883 195.0877 -3.3
3c 30.3 3-p-Coumaroylquinic acid C16H18O8 339.1050 337.0918 7.2
1d 30.6 4-Caffeoylshikimic acid C16H16O8 337.0899 369.1180 5.6
6c 31.4 3-Caffeoylquinic acid methyl ester C17H20O9 369.1173 369.1180 1.9
5c 33.4 3-Feruloylquinic acid C17H20O9 369.1169 355.1024 3
4d 34.8 4-Caffeoylquinic acid C16H18O9 355.1015 339.1074 2.4
2c 35.9 3-Caffeoyl-1,5-quinide C16H16O8 337.0896 337.0918 6.5
6d 36.3 4-Caffeoylquinic acid methyl ester C17H20O9 369.1163 339.1074 4.6
3e 36.7 5-p-Coumaroylquinic acid C16H18O8 339.1059 339.1074 4.6
1e 37.4 5-Caffeoylshikimic acid C16H16O8 337.0903 369.1180 4.4
5e 38.6 5-Feruloylquinic acid C17H20O9 369.1165 337.0918 4.1
2d 39.4 4-Caffeoyl-1,5-quinide C16H16O8 337.0903 369.1180 4.4
3d 41.1 4-p-Coumaroylquinic acid C16H18O8 339.1064 337.0918 3.1
2a 41.3 1-Caffeoyl-1,5-quinide C16H16O8 337.0902 337.0918 4.7
9u 43.6 Feruloyltrimethoxycinnamoylquinic acid C29H33O13 589.1918 589.1916 -0.4
5d 43.6 4-Feruloylquinic acid C17H20O9 369.1165 369.1180 4.1
7ce 47.4 3,5-Dicaffeoylquinic acid C25H24O12 517.1305 517.1341 6.9
7cd 49.7 3,4-Dicaffeoylquinic acid C25H24O12 517.1323 517.1341 3.4
7de 51.3 4,5-Dicaffeoylquinic acid C25H24O12 517.1331 517.1341 1.8
10u 55.9 Feruloylsinapoylquinic acid C28H31O13 575.1767 575.1759 -1.4
3.1.2. Interpretation methods of chlorogenic acids by their fragmentation pattern
Interpretation methods of chlorogenic acids by LC/MSn have been already studied
especially in hydroxycinnamic acid derivatives; caffeoyl quinic acids, 4 feruloylquinic
acids, 16 and others. 17
150
Figure 4. Total MS spectra of cold brew coffee samples
Compounds were interpreted in four different spectra as shown in Supplementary
table 2 ~ 5 with their Total MS data in Figure 4. We can find that the number of
compounds and integrations decrease as sample changes. We will discuss it in 3.2.
Results on brewing kinetics.
3.2. Results on brewing kinetics
3.2.1. Brewing kinetics on concentration
151
It is easy to predict that the concentration would decrease while brewing is progressing
because the compounds are extracting out, and the real concentration showed as
expected. The average solid content in triplicated extraction samples with standard
deviation are shown in Figure 5.
Figure 5. Brewing kinetics on the total solid content
Common cold-brewing method utilizes 50 g of coffee and 10 times of water (500 mL)
to get about 400 mL of concentrated coffee (around 50 to 100 mL of water are
conserved in the coffee column with coffee). Therefore, the first 8 samples are
commercially used, and others are discarded and even if they are used, the actual
152
content would not influence the total compound ratio because of its low concentration.
In our experimental data, more than 85 % of solid content was until 8th sample in total
24 samples.
3.2.2. Brewing kinetics on caffeine and 5-CQA
Positive mode extracted ion chromatogram (EIC) at m/z 195 ± 0.5 Da and negative
mode EIC at m/z 353 ± 0.5 Da were used for interpreting caffeine and 5-CQA,
respectively and UV spectra at 280 nm was used for quantifying both caffeine and 5-
CQA. As following the concentration data, it was observed that both compounds
showed decreasing intensity by kinetics. The average caffeine and 5-CQA integrations
in triplicated extractions with standard deviations are shown in Figure 6.
153
Figure 6. Brewing kinetics on caffeine and 5-CQA
3.2.3. Brewing kinetics on caffeine and CQAs in normalized concentration
By using kinetic peak integrations, conclusive data could not be obtained as they only
showed continuous decreasing graph influenced by kinetic concentration changes as
shown in Figure 5. In Figure 7, concentration normalized caffeine graph show
increasing contents in the same concentration until sample 7, which is almost the limit
of commercially used extracting volume.
154
Figure 7. Brewing kinetics on normalized caffeine content
The main achievement of this research is to obtain the ability to control caffeine content
by brewing. For achieving this, the quantification data of caffeine is required.
Quantification has been done comparing with reference caffeine purchased from Sigma-
Aldrich with integration data of UV spectra. If we regard commercially consumed
coffee as 1 % solution, the caffeine content in Figure 7 gives 62.1 mg (Sample 1) to
97.2 mg (Sample 7) per one cup of coffee we drink. The average caffeine content in 1%
concentration of 100 mL coffee samples is 78.7 mg (commercial average, sample 1 ~ 8)
which is 26.7 % higher than the same concentration coffee of Sample 1 and 23.5 %
lower than that of Sample 7.
155
There has been a controversy in this brewing method that cold brewing extracts less
caffeine because of its temperature versus cold brewing extracts more caffeine because
of its brewing time. Both claims have validity with real data and are still in controversy.
We can find the reason in Figure 6, this method has high standard deviations so that the
exact amount cannot be guaranteed in the range of two to three folds. It can be
speculated that the high error in this method may be due to the uneven water
distribution for the ground coffee. On the top of coffee column, the water drops only in
the center of the column, so fair distribution of water cannot be guaranteed especially
for the early stages. Furthermore, there is also the possibility for the formation of water
channels that may affect the even distribution of water. In this research, we tried to get
certain data by triplicating and averaging all the processes.
To sum up, in Figure 7, we can get less- and enhanced caffeinated undiluted coffee
with maximum caffeine content difference of 56.5 % from sample 1 to sample 7.
4. Conclusion
In conclusion, 24 kinetic samples of Brazil roasted coffee extracts by using cold
brewing method were analyzed in this study. LC-TOFMS and LC-IontrapMS were used
for analyzing the extracts to interpret chlorogenic acids and caffeine in each sample.
Most importantly, it was found that caffeine content in 100 mL coffee could be
controlled from 62.1 mg (Sample 1) to 97.2 mg (Sample 7) when we drink coffee in 1 %
concentration.
156
This is the first caffeine controlling method after the decaffeination methods were
discovered. And it is the only method that caffeine can be controlled in the final stage of
coffee process. All the decaffeination methods have had to be done in industrial scale
because of its huge cost, but this LeeN-caffeination could be done by not only in café
but also in ordinary houses. Although LeeN-caffeination has the limitation that Less-
caffeination and Enhanced-caffeination are only around 25 % of normal caffeine
content, this difference is enough for some groups. For example, the recommended
caffeine intake per day is 400 mg for normal adult and 300 mg for pregnant women in
most of the countries. The 25 % caffeine reduced coffee allows the pregnant group to
drink coffee the same as what they consumed before the pregnancy which might reduce
some part of their pregnancy stress. And the residues can be used for the people who
need to work actively for the daytime.
A lot of further research in this matter can be expected starting from this research;
kinetic research of espresso and filter coffee machine by changing samples in seconds,
higher- or lower-temperature brewing with cold brewing method and another cold
brewing method; steeping cold brew studies.
5. Acknowledgements
Excellent technical support by Anja Müller, Abhinandan Shrestha for his proofreading
and providing cold brewing machine from Café Romeo (Seoul, Korea) are
acknowledged.
157
6. Author Disclosure Statement
No competing financial interests exist.
References
1. Pendergrast M. Uncommon grounds: The history of coffee and how it transformed
our world. Basic Books; 1999.
2. FAO. Food and agriculture organization of the united nations. www.fao.org.
3. Quideau S, Deffieux D, Douat-Casassus C, Pouysegu L. Plant polyphenols: Chemical
properties, biological activities, and synthesis. Angewandte Chemie-International
Edition. 2011;50(3):586-621.
4. Clifford MN, Johnston KL, Knight S, Kuhnert N. Hierarchical scheme for LC-MSn
identification of chlorogenic acids. J Agric Food Chem. 2003;51(10):2900-2911.
5. Nawrot P, Jordan S, Eastwood J, Rotstein J, Hugenholtz A, Feeley M. Effects of
caffeine on human health. Food Additives and Contaminants Part A-Chemistry Analysis
Control Exposure & Risk Assessment. 2003;20(1):1-30.
6. Meyer JF, Jr., Roselius L, Wimmer K, inventors; Germany, assignee. Removing
caffeine from coffee. 1908.
7. Zacek Z, inventorDecaffeinated green coffee. 1960.
158
8. Morrison LR, Jr., Elder MN, Phillips JH, inventors; Procter and Gamble Co.,
assignee. Recovery of noncaffeine solubles in an extract decaffeination process. 1982.
9. Zosel K, inventor; Studiengesellschaft Kohle m.b.H., assignee. Decaffeinizing coffee.
1979.
10. Fischer A, Kummer P, inventors; Coffex A.-G., assignee. Decaffeinization of raw
coffee. 1980.
11. Bentz HW, inventor; Melitta-Werke A.-G., assignee. Filters. 1938.
12. Tarasch LC, inventorExtracting and preserving macerated coffee. 1957.
13. Jaiswal R, Patras MA, Eravuchira PJ, Kuhnert N. Profile and characterization of the
chlorogenic acids in green robusta coffee beans by LC-MSn: Identification of seven
new classes of compounds. J Agric Food Chem. 2010;58(15):8722-8737.
14. Jaiswal R, Matei MF, Golon A, Witt M, Kuhnert N. Understanding the fate of
chlorogenic acids in coffee roasting using mass spectrometry based targeted and non-
targeted analytical strategies. Food & Function. 2012;3(9):976-984.
15. Kuhnert N, Jaiswal R, Eravuchira P, El-Abassy RM, von der Kammer B, Materny
A. Scope and limitations of principal component analysis of high resolution LC-TOF-
MS data: The analysis of the chlorogenic acid fraction in green coffee beans as a case
study. Analytical Methods. 2011;3(1):144-155.
16. Kuhnert N, Jaiswal R, Matei MF, Sovdat T, Deshpande S. How to distinguish
between feruloyl quinic acids and isoferuloyl quinic acids by liquid
159
chromatography/tandem mass spectrometry. Rapid Communications in Mass
Spectrometry. 2010;24(11):1575-1582.
17. Jaiswal R, Kuhnert N. Hierarchical scheme for liquid chromatography/multi-stage
spectrometric identification of 3,4,5-triacyl chlorogenic acids in green robusta coffee
beans. Rapid Communications in Mass Spectrometry. 2010;24(15):2283-2294.
160
Supplementary table 1. List of chlorogenic acids
No Name Abb Q Rx R’
1 Caffeoylshikimic acid CSA 1db2 C H
2 Caffeoyl-1,5-quinide CQL 1la5 C la
3 p-Coumaroylquinic acid CoQA - Co H
4 Caffeoylquinic acid CQA - C H
5 Feruloylquinic acid FQA - F H
6 Caffeoylquinic acid methyl ester M-CQ - C CH3
7 diCaffeoylquinic acid diCQA - diC H
8 Caffeoylferuloylquinic acid CFQA - C, F H
9 Feruloyltrimethoxycinnamoylquinic acid FTMQA - TM, F H
10 Feruloylsinapoylquinic acid FSQA - S, F H
* ‘a’, ‘c’, ‘d’ and ‘e’ stand for stereochemical positions of quinic acid; 1, 3, 4 and 5,
respectively. ‘u’ is used for unconfirmed isomeric positions and ‘i’ is used for
stereoisomers. The stereoisomeric position numbering is shown in Supplementary
figure 1. Some examples of each numberings are shown in Supplementary figure 2.
161
RaO
RcOORd
ORe
OOR'
12
34
5
6
Supplementary figure1. Stereoisomeric positions of quinic acid
162
OOH
OH
O
HO
HO
OHO
1c
O
OHHO
OO
O
OH
OH
2a
HO
OOH
OH
OHO
O
HO3c
HO
HOOH
O
OH
O
OH
OH
4e
HO
OH
O
OHOH
O
O
OH
O
5d
HO
OOH
OH
OO
O
HO
HO
6c
HO
OOH
O
OHO
O
OH
OH
7ce
O
HO
HO
HO
OOH
O
OHO
O
OH
OH
8ec
O
HO
O
HO
HOO
O
OHO
O
OH
O
9ed
O
OO
O
HO
OO
OH
OHO
10cd
O
OHO
O
O
HO
O
Supplementary figure2. Chemical structures of chlorogenic acids with numbering
163
Supplementary table 2. Chlorogenic acids in Brazil cold brew coffee sample 1
No Compd.
MS1 MS2 MS3
Parent ion
base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
4ci 3-CQA de 352.95 190.8 178.9 85 135 25 93.2 170.8 39 111.1 31 172.8 31
4ci 3-CQA de 352.92 178.8 135 30 190.8 21 134.9 135.9 4 133.9 2 137 1
4c 3-CQA 352.93 190.8 178.9 45 134.9 11 127 85.3 75 93.2 71 172.9 58
4ei 5-CQA de 352.93 190.9 191.8 8 265.9 2 126.9 172.9 81 109.1 51 87.3 39
4ci 3-CQA de 352.95 190.9 178.8 73 134.9 17 93.2 111 57 172.9 51 170.9 27
4e 5-CQA 352.96 190.8 191.8 6 178.9 5 127 172.8 79 85.4 75 93.2 47
5ci 3-FQA de 367.01 192.8 172.9 28 331 13 133.9 148.9 28 117.1 4 134.9 3
5ci 3-CQA de 352.95 190.8 265.9 53 178.9 38 85.3 126.9 79 172.9 49 170.9 39
1c 3-CSA 334.89 178.8 160.9 63 134.9 29 134.9 135.9 5 134 2 107.1 1
4ui CQA de 352.94 190.8 265.9 62 172.9 26 127 93.2 98 85.2 71 172.9 22
5c 3-FQA 366.97 192.8 134 11 193.8 10 133.9 148.9 22 134.9 4 117 4
3ei 5-pCoQA de 336.92 190.9 191.8 8 334.9 4 126.9 172.8 96 85.3 42 87.3 38
4d 4-CQA 352.93 172.9 178.9 57 190.8 12 93.2 111.1 52 154.9 24 109.1 15
4di 4-CQA de 352.94 172.9 190.9 68 178.9 58 93.2 111.1 55 154.9 48 71.5 32
5ei 5-FQA de 366.97 190.8 191.8 5 172.9 4 127 172.9 66 144.9 31 85.3 23
3e 5-pCoQA 336.91 190.8 191.8 9 334.9 7 127 172.9 68 85.3 47 93.2 47
2c 3-CQAL 334.91 160.8 134.9 75 190.9 13 132.9 117 3 134 1 - -
4di 4-CQA de 352.97 190.9 172.9 97 533.1 37 93.2 111.1 23 - - - -
1e 5-CSA 334.92 172.8 178.9 25 160.9 19 111 93.3 45 154.9 38 71.6 37
5e 5-FQA 366.96 190.8 191.8 8 192.8 5 127 172.9 97 85.3 84 93.2 61
2d 4-CQAL 334.92 160.9 178.8 23 135 22 132.9 119 10 158.8 8 134 2
2a 1-CQAL 334.94 160.9 172.9 56 178.8 34 132.9 - - - - - -
4d 4-FQA 366.98 172.9 192.8 9 173.8 7 93.2 111 72 71.5 28 154.9 16
4di 4-FQA de 366.98 172.9 190.8 38 347.9 37 111 93.3 75 97.2 58 99.1 22
1ei 5-CSA de 334.91 172.8 314.8 20 332.9 13 93.2 59.5 95 111.2 94 128.1 84
7ce 3,5-diCQA 515.11 352.9 353.9 21 190.9 9 190.8 178.8 52 135 9 191.8 8
7cd 3,4-diCQA 515.11 352.9 172.9 27 178.9 16 172.9 178.8 64 190.8 58 135 16
7de 4,5-diCQA 515.09 352.9 340.9 28 172.9 19 280.9 296.9 93 297.9 26 254.9 13
8u C-4-FQA 529.12 367 172.9 46 352.9 37 172.8 192.8 42 190.9 10 133.9 6
164
Supplementary table 3. Chlorogenic acids in Brazil cold brew coffee sample 5
No Compd.
MS1 MS2 MS3
Parent ion
base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
4ci 3-CQA de 352.94 190.9 178.9 80 134.9 19 93.2 170.9 33 172.9 28 111.1 23
4ci 3-CQA de 352.93 178.9 135 23 172.9 12 134.9 133 11 135.9 10 134 5
4c 3-CQA 352.97 190.8 178.8 45 191.8 10 127 172.8 88 85.3 81 111 68
4ei 5-CQA de 352.93 190.8 191.8 6 172.9 3 127 172.9 83 144.8 35 85.3 31
4ei 5-CQA de 352.93 190.8 134.9 6 265.9 9 127 93.1 91 85.3 86 111 78
4e 5-CQA 352.94 190.9 191.8 7 214.9 3 172.8 127 88 85.3 85 93.2 69
6c M-3-CQ 366.98 160.9 192.8 16 134.9 12 132.9 - - - - - -
4ci 3-CQA de 352.93 190.8 178.8 21 265.9 7 127 172.8 72 111 62 93.2 58
1c 3-CSA 334.9 178.8 160.9 60 134.9 31 135 135.8 15 134 2 - -
4di 4-CQA de 352.92 190.9 265.9 29 172.9 27 172.8 127 84 108.9 62 170.9 37
5c 3-FQA 366.97 192.8 193.8 13 133.9 10 133.9 148.9 27 117 4 190.8 3
4d 4-CQA 352.93 172.9 178.8 48 190.8 13 93.2 111.1 62 154.9 21 59.7 17
4di 4-CQA de 352.93 172.9 190.8 56 178.9 44 93.3 170.8 24 115.1 23 111.1 18
5ei 5-FQA de 366.93 190.8 191.7 5 172.9 5 126.9 172.9 95 144.9 33 87.2 30
2c 3-CQAL 334.9 160.9 135 87 190.8 26 132.9 93.3 37 - - - -
6d M-4-CQ 367.02 160.9 134.9 24 161.8 13 132.9 - - - - - -
1e 5-CSA 334.92 172.8 178.8 29 160.9 23 111.1 93.2 61 154.8 38 71.5 36
5e 5-FQA 367 190.8 191.8 9 192.9 4 172.8 127 78 85.3 52 93.2 41
2d 4-CQAL 334.89 160.9 178.9 29 135 22 132.9 134 5 116.9 4 158.6 3
3d 4-pCoQA 336.96 172.8 334.9 24 173.8 12 111.1 93.2 93 154.9 71 71.5 49
2a 1-CQAL 334.93 160.9 172.9 46 317.8 24 132.9 - - - - - -
5d 4-FQA 367 172.9 192.9 12 173.8 7 93.2 111.1 59 71.5 33 154.9 17
7ce 3,5-diCQA 515.07 352.9 353.9 15 190.9 14 190.8 178.8 45 134.9 9 191.6 3
7cd 3,4-diCQA 515.1 353 172.9 25 334.9 18 178.9 172.8 83 190.8 54 135 21
7de 4,5-diCQA 515.09 352.9 202.8 17 172.9 15 172.9 178.9 69 190.8 27 134.9 13
165
Supplementary table 4. Chlorogenic acids in Brazil cold brew coffee sample 10
No Compd.
MS1 MS2 MS3
Parent ion
base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
4ci 3-CQA de 352.92 178.8 172.9 16 134.9 15 134.9 132.9 15 135.9 4 - -
4c 3-CQA 352.92 190.8 178.9 40 191.8 8 127 172.9 57 85.3 46 109.1 46
4ei 5-CQA de 352.91 190.8 191.8 7 350.9 4 126.9 172.9 94 85.3 46 109.1 27
4e 5-CQA 352.93 190.8 191.8 8 214.8 3 127 172.8 85 85.3 59 93.2 38
6c M-3-CQ 366.97 160.9 192.8 21 134.9 11 260 132.9 12 117.2 2 - -
4ci 3-CQA de 352.93 190.9 178.8 17 350.9 10 172.8 127 64 111.1 54 109.1 53
1c 3-CSA 334.98 178.9 160.9 52 135 30 134.9 108 4 135.9 4 117 3
5ci 3-FQA de 366.87 192.8 298.7 91 347.8 65 230.7 229.8 73 252.8 17 239.8 17
4di 4-CQA de 352.92 172.9 190.9 88 178.9 42 93.2 154.9 49 59.6 30 111.1 18
5c 3-FQA 366.98 192.8 334.9 17 134 12 133.9 148.9 22 190.8 5 117 3
4d 4-CQA 352.93 172.9 178.8 45 190.8 15 93.2 111.1 81 154.8 36 71.5 30
2ci 3-CQAL de 334.88 160.9 134.9 68 178.9 18 132.8 - - - - - -
2c 3-CQAL 334.91 160.9 134.9 70 178.9 11 132.9 117.1 2 - - - -
6d M-4-CQ 366.99 160.9 135 28 161.9 12 132.9 134 23 131.9 7 - -
1e 5-CSA 334.93 172.9 178.9 25 160.9 21 111 71.5 33 93.2 31 154.9 27
5e 5-FQA 366.97 190.8 191.8 8 192.8 5 172.9 85.3 96 93.2 77 127 74
2d 4-CQAL 334.93 160.9 178.9 42 135 28 132.9 158.9 10 119.1 6 - -
2a 1-CQAL 334.9 160.9 172.9 98 288.8 25 133 - - - - - -
5d 4-FQA 366.94 172.8 190.8 20 192.8 11 93.2 111.1 75 136.9 27 71.5 26
166
Supplementary table 5. Chlorogenic acids in Brazil cold brew coffee sample 15
No Compd.
MS1 MS2 MS3
Parent ion
base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
4c 3-CQA 352.93 190.8 178.9 41 176.9 17 126.9 85.3 58 172.8 49 93.3 31
6c M-3-CQ 366.94 160.8 192.8 16 134.9 9 132.9 133.9 6 117.1 4 - -
4e 5-CQA 352.9 190.8 214.8 13 191.8 8 394.7 126.9 40 93.2 31 172.9 24
5c 3-FQA 366.99 192.9 134 15 193.8 14 133.9 148.9 16 136.4 3 117 2
4d 4-CQA 352.93 172.8 178.8 43 190.9 21 93.2 111.1 60 154.9 46 94.2 26
2c 3-CQAL 334.93 160.9 135 84 161.9 13 132.9 - - - - - -
6d M-4-CQ 366.97 160.9 134.9 32 161.8 12 132.9 - - - - - -
1e 5-CSA 334.91 172.9 178.8 16 317.8 9 111.1 93.2 71 155 20 59.7 14
5e 5-FQA 366.98 190.8 172.9 5 348.9 5 85.3 172.8 74 111.1 54 126.9 41
2d 4-CQAL 334.89 160.9 178.8 59 210.8 37 132.9 143.9 25 - - - -
5d 4-FQA 366.96 172.9 347.9 18 192.8 11 93.2 110.9 52 71.5 48 154.9 23
167
Research on Application to Other Food 3.4.
Identification of Composition Kinetics of Traditional Korean Green Tea
Producing Method by LC-MS
Seung-Hun Lee1, Inamullah Hakeem Said1, Anastasiia Shevchuk1, Un-Jae Lee2, Nikolai
Kuhnert1*
Department of Life Sciences and Chemistry, Jacobs University Bremen, 28759 Bremen,
Germany1
Keumro Handmade Tea, Jeollabuk-do, Korea2
*Correspondence to:
Prof. Dr. Nikolai Kuhnert,
Department of Life Sciences and Chemistry
Jacobs University Bremen gGmbH
Campus Ring 1
28759 Bremen, Germany
Tel: +49 421 200 3120
Fax: +49 421 200 3102
E-mail: [email protected]
168
Abstract
Nine times steaming and drying process is originated by making Korean Red Ginseng
and is adapted to green tea parching process. The samples are prepared in a traditional
way Korean tea making process separating by each step. All the time series samples are
analyzed by HPLC-TOFMS and HPLC-IontrapMS for interpreting and quantifying
polyphenols and caffeine. Those samples were analyzed both all the components in the
leaves and the content of tea infusions consumed. It could be shown that repeated
steaming increases the extractability of catechins.
Keywords: Korean traditional green tea; Nine times steaming and drying; Polyphenol;
Korean tea ceremony
169
1. Introduction
The Korean traditional green tea method did not attract scientific interest since its
production scale is much lower than machine producing methods. However, in March
2016, Korean Cultural Heritage Administration reported that Korean traditional green
tea producing method would become Korean national intangible cultural heritage in
2016.1 Therefore, to identify the extracts of the tea produced by the Korean traditional
method merits scientific interest, based on its importance in cultural heritage.
Nine repetitive parching and drying process, which is one of the Korean traditional
green tea production methods, has been adapted from nine repetitive steaming and
drying process used in Korean red ginseng production. In the Korean red ginseng
method, an increase of health promoting saponins during repetition of the parching
process has already been observed.2 This nine times repeating method is known to the
public to increase the value of the product.
Starting from the explanation for French Paradox,3 polyphenols attracted scientific
interest. Scientists found polyphenols from wine, fruits, coffee and tea and concatenated
them to the scientific grounds of healthy food. Green tea constitute one of the richest
source of flavanols within the human diet with abundant amount of (epi)catechin (EC),
gallocatechin (GC), (epi)gallocatechin gallate (EGCG), (epi)catechin gallate (ECG) and
so on.4
In 2003, three different regioisomers of caffeoylquinic acids were separated and
analyzed with HPLC-tandem MS.5 Influenced by this research, not only for the
170
chlorogenic acids, but also the whole polyphenol compounds could be easily analyzed
by checking their fragmentation. After this research, HPLC-tandem MS became the
most power tool for analyzing all the different families of polyphenols.6-8 Especially for
thearubigins, HPLC-iontrap MS and UPLC-ionmobility MS were selected and
developed to analyze compounds in black tea.9,10
In this research, we analyzed sequential samples by 9 times repetitive parching and
drying of green tea leaves with two different sampling methods; extracting with organic
solvent by sonication of finely ground tea leaves, and aqueous brewing following the
traditional Korean tea ceremony. The two different extracted teas were analyzed by LC-
high resolution ESI-MS and LC-tandem ESI-MS to identify and quantify polyphenols
and caffeine. EIC spectra of specific polyphenols were used to analyze the content
changes by sequential parching samples.
2. Experimental
The experimental methods of HPLC and LC/MSn are followed by previous publications
of our group research. 11-13
2.1. Chemicals and Green Tea Samples
All the chemicals (Analytical grade) were purchased from Sigma-Aldrich (Germany).
Green tea samples in different parching processes were provided from Geumro
Handmade Tea in Jeollabuk-do, Korea.
2.2. Kinetic parching samples or green tea
171
Parching for green tea leaves were conducted as to make commercial samples. Starting
from sample 0 as without parching, green tea leaves were processed with continuous
heating and drying. In the standard nine-times heating and drying process, the first
parching temperature starts from higher than 250 ℃, gradually decreases to 50 ℃ for
the ninth process. Every sample was numbered and took out from the pot to make
samples from number 1 to 9. All the samples were sealed separately and kept in room
temperature until use.
2.3. High Resolution MS
The LC equipment (Agilent 1100 series, Karlsruhe, Germany) comprised a binary pump,
an autosampler with a 100 µL loop, and a diode-array detector (DAD) with a light-pipe
flow cell (recording at 320 and 254 nm and scanning from 200 to 600 nm). This was
interfaced with a MicroTOF Focus mass spectrometer (Bruker Daltonics, Bremen,
Germany) fitted with an electrospray ionization (ESI) source and internal calibration
was achieved with 10 mL of 0.1 M sodium formate solution injected through a six port
valve prior into each chromatographic run. Calibration was carried out using the
enhanced quadratic mode.
2.4. LC/MSn
The LC equipment comprised a binary pump, an autosampler with a 100 µL loop, and a
diode-array detector (DAD) with a light-pipe flow cell (recording at 320 and 254 nm
and scanning from 200 to 600 nm). This was interfaced with an ion-trap mass
spectrometer fitted with an ESI source (Bruker Daltonics HCT ultra, Bremen, Germany)
operating in full scan, auto MSn mode (smart fragmentation) using a ramping of the
172
collision energy. Maximum fragmentation amplitude was set to 1 V, starting at 20 %
and ending at 200 %. MS operating conditions (negative mode) had been optimized
using 5-caffeoylquinic acid with a capillary temperature of 365 ℃, a drying gas flow
rate of 10 L / min, and a nebulizer pressure of 10 psi.
2.5. HPLC
Separation was achieved on a 150 × 3 mm i.d. column containing diphenyl 5 µm, with a
5 mm × 3 mm i.d. guard column (Varian, Darmstadt, Germany). Solvent A was water /
formic acid (1000:0.005 v/v) and solvent B was methanol. Solvents were delivered at a
total flow rate of 500 µL / min. The gradient profile was from 10 to 80 % B linearly in
70 min followed by 10 min a return to 10 % B and 10 min isocratic to re-equilibrate.
2.6. Green tea extracting
2.6.1. Total polyphenol extracting
The sample preparation method for total polyphenol extracting used ultra-sonication for
obtaining sufficient extracting yield following the previous published methods for
analyzing polyphenols of plant leaves.14,15 In this research, 2 g of dried green tea leaves
for each sample and put them in a mortar to grind into powder with the temperature of
liquid nitrogen. Falcon tubes of 15 mL were used to mix 1 g of the powder into 10 mL
of 70 % methanol solution. The tubes were ultra-sonicated for 30 min and filtered
through a membrane filter. The solutions were used directly for LC-MS.
2.6.2. Practical brewing polyphenol extracting
173
The sample preparation method for practical polyphenol extracting was tried to follow
to the Korean traditional green tea ceremony. 1 g of green tea sample was put into a 250
mL beaker, followed by 100 mL deionized water filling. After 2 min of waiting without
stirring, the extracted tea solution was carefully poured out and filtered through a
membrane filter. The filling water, waiting and pouring out extracted solution were
repeated twice to make three different sequential brewing samples. Those filtered
solutions were directly used for LC-MS.
3. Results and Discussion
3.1. Interpretations of polyphenol
MicroTOF was used to obtain exact molecular weight to determine chemical formula
and Iontrap MS was used to get the fragmentation patterns to determine isomeric
structures of polyphenol compounds in Korean green tea samples.
3.1.1. Molecular formula analysis of high resolution MS data
Polyphenolic compounds and caffeine peaks are detected and listed with their Sodium
adduct in positive mode by TOF MS in Table 1. The experimental value is based on the
1st water extracted non-parching sample. After obtaining the chemical formula by high
resolution MS data, tandem MS data were used together to determine the isomeric
composition of each compound.
174
Table 1. High resolution mass data in the 1st water extracted samples of non-parching
Korean green tea sample in positive ion mode
No. RT Name Mol. Formula Exp. m/z (M+H)
Theor. m/z (M+H)
Error (ppm)
1 6.1 Quinic acid C7H13O6 193.0696 193.0707 5.4
2 6.1 Na+ adduct of Quinic acid C7H12NaO6 215.0523 215.0526 1.5
3 25.2 Gallocatechin C15H15O7 307.0816 307.0812 -1.2
4 25.2 Na+ adduct of Gallocatechin C15H14NaO7 329.0631 329.0632 0.3
5 25.3 3-CQA C16H19O9 355.1054 355.1024 -8.6
6 25.3 Na+ adduct of 3-CQA C16H18NaO9 377.0813 377.0843 7.9
7 27.5 procyanidin B2 C30H27O12 579.1464 579.1497 5.7
8 27.5 Na+ adduct of procyanidin
B2 C30H26NaO12 601.1272 601.1316 7.3
9 29.4 Na+ adduct of 5-CQA C16H18NaO9 377.0845 377.0843 -0.5
10 30.4 Caffeine C8H11N4O2 195.087 195.0877 3.6
11 30.4 Na+ adduct of Caffeine C8H10N4NaO2 217.0689 217.0696 3.3
12 30.4 3-CoQA C16H19O8 339.1083 339.1074 -2.5
13 30.4 Na+ adduct of 3-CoQA C16H18NaO8 361.0885 361.0894 2.4
14 32 (epi)gallocatechin gallate C22H19O11 459.0914 459.0922 1.7
15 32 Na+ adduct of
(epi)gallocatechin gallate C22H18NaO11 481.0724 481.0741 3.5
16 32.1 cis-3-CoQA C16H19O8 339.1071 339.1074 1
17 32.1 Na+ adduct of cis-3-CoQA C16H18NaO8 361.089 361.0894 1.2
18 33.6 (epi)catechin C15H15O6 291.0879 291.0863 -5.5
19 33.6 Na+ adduct of (epi)catechin C15H14NaO6 313.0693 313.0683 -3.3
20 36.8 5-CoQA C16H19O8 339.1042 339.1074 9.6
21 36.8 Na+ adduct of 5-CoQA C16H18NaO8 361.0879 361.0894 4.1
22 39.2 (epi)catechin gallate C22H19O10 443.0958 443.0973 3.3
23 39.2 Na+ adduct of (epi)catechin
gallate C22H18NaO10 465.078 465.0792 2.7
24 39.6 4-CoQA C16H19O8 339.107 339.1074 1.3
25 39.6 Na+ adduct of 4-CoQA C16H18NaO8 361.0883 361.0894 2.9
26 41.1 cis-4-CoQA C16H19O8 339.1072 339.1074 0.7
27 41.1 Na+ adduct of cis-4-CoQA C16H18NaO8 361.0892 361.0894 0.4
28 41.2 p-Coumaroylquinic acid C19H19O8 339.1075 339.1074 -0.1
29 41.2 Na+ adduct of p-
Coumaroylquinic acid C19H18NaO8 361.0889 361.0894 1.5
30 56.1 Theaflavin-3-O-gallate C36H29O16 717.1378 717.145 10.1
31 56.1 Na+ adduct of Theaflavin-3-
O-gallate C38H28NaO16 739.1256 739.127 1.8
175
3.2.2. Compound assignment
Compounds were identified based on fragmentation data from tandem-MS spectra. The
methanolic soxhlet extract of non-parching Korean green tea sample spectrum was used
to assign isomeric structures of polyphenols. Table 2 shows assignment of polyphenols
with their fragmentation patterns from the base peak chromatogram shown in Figure 1.
Table 2. Tandem MS data in negative ion mode of polyphenols from non-parching
Korean green tea methanolic soxhlet extract
No RT Compd.
MS1 MS2 MS3
Parent ion
base peak Secondary peak base peak secondary peak
m/z m/z int m/z int m/z m/z int m/z int m/z int
1 4.1 Caffeoylhexose dimer 683.27 341 341.9 10 340.2 1 178.9 160.9 38 113.1 23 143 20
2 5.9 QA dimer 383.06 190.8 191.8 8 347 1 126.9 93.2 61 172.9 52 85.3 34
3 6 QA 190.88 127 85.4 54 93.3 45 109.1 0 0 0 0 0 0
4 11.2 Gallic acid 168.87 125 123 83 0 0 95.2 0 0 0 0 0 0
5 22.7 3-CQA 352.98 190.8 178.8 41 176.8 13 85.3 97.2 49 87.2 36 109 32
6 24.3 Gallocatechin 304.91 178.9 220.9 80 218.9 65 163.9 135 51 150.9 39 137 24
7 25.4 Catechin 288.92 244.8 204.8 36 178.8 11 203 161 30 226.8 24 186.8 22
8 27.1 Procyanidin B2 577.17 407 425 79 289 27 284.8 280.8 87 389 49 242.8 30
9 28.3 5-CQA 352.98 190.8 214.8 13 178.8 5 127 111 31 85.4 30 172.8 29
10 28.7 3-CoQA 336.96 162.9 119.1 10 163.8 9 119 120 4 117 0 0 0
11 30.5 cis-3-CoQA 336.96 162.8 190.8 8 119 7 119 0 0 0 0 0 0
12 31.5 (epi)Gallocatechin gallate 457.05 168.9 330.9 29 304.9 24 125 123.1 4 97.4 3 107 0
13 32.6 (epi)Catechin 288.91 244.9 204.8 42 202.9 15 202.9 186.9 17 160.9 17 226.9 16
14 35.3 5-CoQA 336.96 190.8 163 4 0 0 127 85.4 67 172.8 53 93.2 37
15 37.9 4-CoQA (de) 336.93 172.8 163 6 0 0 93.2 111 61 154.8 20 71.6 19
16 38.1 (epi)Catechin gallate 441.05 288.9 330.9 25 289.9 21 244.9 204.8 40 202.9 14 178.9 13
17 39.6 4-CoQA 336.94 172.9 173.8 8 190.9 6 93.2 111.1 49 71.5 22 154.9 17
18 55.4 Theaflavin-3-O-gallate 715.15 527 563.2 66 545 63 371 399 69 483.2 46 509 23
176
Figure 1. Base peak chromatogram in negative ion mode of non-parching Korean
Green Tea methanolic soxhlet extract
3.2. Results on different number of times on parching
3.2.1. Comparison on linear fit of different extracting methods of the sequential
parching samples
The UV Integrations of selected compounds’ peaks in HPLC spectra of Korean Green
Tea samples were compared in Figure 2. Composition changes by number of parching
of methanolic extracted samples are shown in Figure 2(a) and their linear fits are shown
in Figure 2(b). Three times averaged composition changes by number of parching of
water extracted samples are shown in Figure 2(c) with their linear fits in Figure 2(d).
177
Figure 2. Composition changes of the sequential parching samples and their linear fits;
(a) composition of methanolic extracts (b) linear fit of methanolic extracts (c)
composition of water extracts (d) linear fit of water extracts
Notable differences between the two different extracting solvent data are the different
slopes of linear fits shown in Figure 2(b) and Figure 2(d). In methanolic extracted
samples, the parching times does not affect to the extraction yields of each compound as
presented in Figure 2(b). On the other hand, the more parching times, the more yields
of polyphenolic compounds are obtained on water extracted samples, presented in
Figure 2(d). This phenomenon could be explained by the following assumption that the
9 times parching and dry process does not decompose the flavonoids in tea, so the total
3000
6000
9000
12000
15000
18000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
)
GC Caffeine EC ECG EGCG
3000
6000
9000
12000
15000
18000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
) GC Caffeine EC ECG EGCG
0
10000
20000
30000
40000
50000
0 1 2 3 4 5 6 7 8 9
UV
Int
ens
ity (
mA
U)
GC Caffeine EC ECG EGCG
5000
10000
15000
20000
25000
30000
35000
40000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
)
GC Caffeine EC ECG EGCG
(a) (b)
(c) (d)
178
phenolics are not changing, but this process can soften the green tea leaves to obtain
better extraction yields. The detailed variables of linear fits on each compound in two
different solvent extracted data are shown in Supporting Information Figure 1 to 10.
3.2.1. Comparison on multiple brewing samples following Korean traditional tea
ceremony.
The traditional Korean tea ceremony recommends three times repeating tea brewing for
one batch of tea leaves. As it is indicated in 2.6.2. Practical brewing polyphenol
extracting, the experiment followed traditional method of tea brewing. The entire
sequential composition spectra are shown in Figure 3 classified by the compounds.
179
Figure 3. Composition of the sequential parching samples extracted by practical
brewing method; a) Gallocatechin b) Caffeine c) (Epi)Gallocatechin gallate d)
(Epi)Catechin e) (Epi)Catechin Gallate
15000
30000
45000
60000
75000
b)
9876543210
UV
Inte
nsity
(m
AU
)
First Second Third Average
0
4000
8000
12000
16000
d)
9876543210
UV
Inte
nsity
(m
AU
)
First Second Third Average
0
6000
12000
18000
24000
30000a)
987643210
UV
Inte
nsity
(m
AU
)
5
First Second Third Average
0
15000
30000
45000
60000
75000c)
9876543210
UV
Inte
nsity
(m
AU
)
First Second Third Average
0
5000
10000
15000
20000
25000e)
9876543210
UV
Inte
nsity
(m
AU
)
First Second Third Average
180
As shown in Figure 3, tea flavonoids’ concentrations start to decrease in the third
brewing. In traditional Korean tea ceremony, dried tea leaves start to be unrolled from
the first brewing. Therefore, the first brewing does not have enough surface area to be
extracted. This is the reason why the first and the second brewings do not have
meaningful flavonoid concentration disparity. The third brewing seems the starting
point of diminution. After considering these phenomena, the Korean tea ceremony
could be regarded as a scientific consuming method of green tea flavonoids.
4. Conclusion
In conclusion, nine times parching and drying green tea samples were analyzed and
interpreted with two different extracting methods; soxhlet extraction with aqueous
methanol for total polyphenol analysis and practical brewing method with water for
sequential sample analysis. LC-TOFMS and LC-IontrapMS were used for analyzing the
extracts to interpret polyphenols and caffeine in each sample.
The most important discovery in this study is the different polyphenol contents between
total polyphenol content and actual consuming polyphenol contents. Although the total
polyphenol contents were not changed by its different parching steps, actual consuming
polyphenol contents increased in accordance with the parching step proceeds. As a
result, the nine times parching method is a soft method which does not decompose the
polyphenols but powerful method which influences the extracting yield by practical tea
brewing. On top of that, by analyzing the sequential brewing concentration, Korean
181
green tea ceremony is verified as a scientifically sound tea consuming method through
three times brewing repetition for efficient flavonoids extraction.
The nine times heating and drying method was mainly investigated only for the Korean
red ginseng process and started to spread throughout other foods. The discovery in this
study is important by itself for the green tea development and is expected for adapting
into the process development of the other food innovation.
Acknowledgements
Excellent technical support by Anja Müller is acknowledged.
References
1. Korean Intangible Heritage Digital Archive Korean Traditional Tea Process. 2017,
2. Kim, D.; Kim, Y.; Lee, Y.; Min, J.; Kim, S.; Yang, D. Conversion of ginsenosides by
9 repetitive steamings and dryings process of Korean ginseng root and its inhibition of
BACE-1 activity. Journal of Physiology & Pathology in Korean Medicine 2008, 22,
3. Renaud, S.; Delorgeril, M. Wine, Alcohol, Platelets, and the French Paradox for
Coronary Heart-Disease. Lancet 1992, 339, 1523-1526.
4. Graham, H.N. Green tea composition, consumption, and polyphenol chemistry.
Preventive Medicine 1992, 21, 334-350.
5. Clifford, M.N.; Johnston, K.L.; Knight, S.; Kuhnert, N. Hierarchical scheme for LC-
MSn identification of chlorogenic acids. J. Agric. Food Chem. 2003, 51, 2900-2911.
182
6. Jaiswal, R.; Kuhnert, N. Hierarchical scheme for liquid chromatography/multi-stage
spectrometric identification of 3,4,5-triacyl chlorogenic acids in green Robusta coffee
beans. Rapid Communications in Mass Spectrometry 2010, 24, 2283-2294.
7. Jaiswal, R.; Kuhnert, N. Identification and characterization of five new classes of
chlorogenic acids in burdock (Arctium lappa L.) roots by liquid
chromatography/tandem mass spectrometry. Food & Function 2011, 2, 63-71.
8. Kuhnert, N.; Jaiswal, R.; Matei, M.F.; Sovdat, T.; Deshpande, S. How to distinguish
between feruloyl quinic acids and isoferuloyl quinic acids by liquid
chromatography/tandem mass spectrometry. Rapid Communications in Mass
Spectrometry 2010, 24, 1575-1582.
9. Yassin GH, Grun C, Koek JH, Assaf KI, Kuhnert N. Investigation of isomeric
flavanol structures in black tea thearubigins using ultraperformance liquid
chromatography coupled to hybrid quadrupole/ion mobility/time of flight mass
spectrometry. J Mass Spectrom. 2014;49(11):1086-1095.
10. Yassin GH, Koek JH, Kuhnert N. Model system-based mechanistic studies of black
tea thearubigin formation. Food Chem. 2015;180:272-279.
11. Jaiswal, R.; Patras, M.A.; Eravuchira, P.J.; Kuhnert, N. Profile and Characterization
of the Chlorogenic Acids in Green Robusta Coffee Beans by LC-MSn: Identification of
Seven New Classes of Compounds. J. Agric. Food Chem. 2010, 58, 8722-8737.
12. Jaiswal, R.; Matei, M.F.; Golon, A.; Witt, M.; Kuhnert, N. Understanding the fate of
chlorogenic acids in coffee roasting using mass spectrometry based targeted and non-
targeted analytical strategies. Food & Function 2012, 3, 976-984.
13. Kuhnert, N.; Jaiswal, R.; Eravuchira, P.; El-Abassy, R.M.; von der Kammer, B.;
Materny, A. Scope and limitations of principal component analysis of high resolution
LC-TOF-MS data: the analysis of the chlorogenic acid fraction in green coffee beans as
a case study. Analytical Methods 2011, 3, 144-155.
183
14. Saito, S.T.; Gosmann, G.; Saffi, J.; Presser, M.; Richter, M.F.; Bergold, A.M.
Characterization of the constituents and antioxidant activity of Brazilian green tea
(Camellia sinensis var. assamica IAC-259 cultivar) extracts. J. Agric. Food Chem. 2007,
55, 9409-9414.
15. Sereshti, H.; Samadi, S.; Jalali-Heravi, M. Determination of volatile components of
green, black, oolong and white tea by optimized ultrasound-assisted extraction-
dispersive liquid-liquid microextraction coupled with gas chromatography. J.
Chromatogr. A 2013, 1280, 1-8.
184
Y = 4342.72512 + 71.59505 * X R = 0.22088 P = 0.53972 Supporting Information Figure 1. Linear fit of Gallocatechin extracted by 70 % methanol (aq)
3000
6000
9000
12000
15000
18000
U
V In
tens
ity (
mA
U)
0 1 2 3 4 5 6 7 8 9
Gallocatechin, MeOH (aq) Linear fit of Gallocatechin
185
Y = 4069.66854 – 57.85839 * X R = -0.43275 P = 0.21161 Supporting Information Figure 2. Linear fit of Caffeine extracted by 70 % methanol (aq)
3000
6000
9000
12000
15000
18000
U
V In
tens
ity (
mA
U)
0 1 2 3 4 5 6 7 8 9
Caffeine, MeOH (aq) Linear fit of Caffeine
186
Y = 2196.08093 + 14.03555 * X R = 0.09681 P = 0.79019 Supporting Information Figure 3. Linear fit of Epicatechin extracted by 70 % methanol (aq)
3000
6000
9000
12000
15000
18000
U
V In
tens
ity (
mA
U)
0 1 2 3 4 5 6 7 8 9
Epicatechin, MeOH (aq) Linear fit of Epicatechin
187
Y = 4416.79979 + 25.24202 * X R = 0.10403 P = 0.77487 Supporting Information Figure 4. Linear fit of Epicatechin gallate extracted by 70 % methanol (aq)
3000
6000
9000
12000
15000
18000
U
V In
tens
ity (
mA
U)
0 1 2 3 4 5 6 7 8 9
Epicatechin Gallate, MeOH (aq) Linear fit of Epicatechin Gallate
188
Y = 13067.91881 + 167.27783 * X R = 0.17709 P = 0.62454 Supporting Information Figure 5. Linear fit of Epigallocatechin gallate extracted by 70 % methanol (aq)
3000
6000
9000
12000
15000
18000
U
V In
tens
ity (
mA
U)
0 1 2 3 4 5 6 7 8 9
Epigallocatechin Gallate, MeOH (aq) Linear fit of Epigallocatechin Gallate
189
Y = 3277.95293 + 1184.37219 * X R = 0.83607 P = 0.00258 Supporting Information Figure 6. Linear fit of Gallocatechin extracted by Water
0
10000
20000
30000
40000
50000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
)
Gallocatechin, Water Linear fit of Gallocatechin
190
Y = 37981.13732 + 213.50734 * X R = 0.14388 P = 0.6917 Supporting Information Figure 7. Linear fit of Caffeine extracted by Water
0
10000
20000
30000
40000
50000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
)
Caffeine, Water Linear fit of Caffeine
191
Y = 3143.54382 + 658.77461 * X R = 0.93223 P < 0.0001 Supporting Information Figure 8. Linear fit of Epicatechin extracted by Water
0
10000
20000
30000
40000
50000
U
V In
tens
ity (
mA
U)
0 1 2 3 4 5 6 7 8 9
(Epi)Catechin, Water Linear fit of (Epi)Catechin
192
Y = 3920.76593 + 1165.60438 * X R = 0.93467 P < 0.0001 Supporting Information Figure 9. Linear fit of Epicatechin gallate extracted by Water
0
10000
20000
30000
40000
50000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
)
(Epi)Catechin Gallate, Water Linear fit of (Epi)Catechin Gallate
193
Y = 7692.49752 + 3242.9969 * X R = 0.84168 P = 0.00226 Supporting Information Figure 10. Linear fit of Epigallocatechin gallate extracted by Water
0
10000
20000
30000
40000
50000
0 1 2 3 4 5 6 7 8 9
UV
Inte
nsity
(m
AU
)
(Epi)Gallocatechin Gallate, Water Linear fit of (Epi)Gallocatechin Gallate
194
4. Conclusion
In this study, the hierarchical key LC-MSn method was mainly used for identifying CGAs and
caffeine of different experimental samples from different stages of the coffee processes;
harvesting and grading, roasting, brewing and applications.
By these researches, below are the first results ever reported with scientific methodology.
1. The chlorogenic acids of Jamaica Blue Mountain coffee were investigated.
2. The different grade green coffee beans in a single origin were compared.
3. The possibility as quality influencing factor of ratios of caffeoylquinic acids’
regioisomers were suggested.
4. Higher antioxidant property processed coffee bean than green coffee bean was found.
5. The differences of aqueous and organic solvent extracted coffee were proved.
6. Cold brew coffee, which is the recent brewing method, was analyzed.
7. The composition changes by kinetic samples of coffee were identified.
8. A new caffeine control method other than decaffeination was developed.
9. The applicability of coffee analyzing methods to other food techniques was confirmed.
In the first harvesting and grading step, three different grades of Jamaica Blue Mountain
coffees were compared each other and with other originated coffees. By using LC-MSn
method, 47 different CGAs and their derivatives were identified as other published coffee
papers reported, but first time investigated for the Jamaica Blue Mountain coffee beans and
different grade coffees in the same origins. Additionally, the regioisomer ratio of 3-, 4- and 5-
CQA was suggested as the possible aroma factor for the first time as the specialty Jamaica
beans and other normal beans had different values.
195
In the roasting process, novel processed coffee beans were investigated to search the
possibility as roasting process substitute. The novel process was nine times repeating of
parching and drying method which is derived from Korean traditional red-ginseng process so
this special method is named steamentation from steaming and fermentation. Most
importantly, the antioxidant property of the novel method was higher than any other
processed coffee bean samples including green coffee bean and two different degree roasted
coffee beans. Especially, higher antioxidant property processed coffee bean than the green
coffee bean was first time reported in this study. EIC at m/z 367 and 335 were concentrically
interpreted by LC-MSn method and showed the differences of chemical reactivity of
steamented coffee to green and roasted coffees.
In the brewing stage, maximum 14 times different 5-CQA equivalent was found with 3-CQA
methyl ester integration in 70 % aqueous methanol and water extracted samples. This result
demonstrated that a discrepancy between polyphenol levels determined from optimized
extraction protocols and levels found in actual coffee brews prepared by consumers.
Moreover, cold-brew coffee, the recently became popular coffee brewing method coffee, was
analyzed separated in sequential order separation. The most important discovery in this
research is that the caffeine in the same solid content increases in time dependent manners of
the samples. It means that only separating the coffee while brewing can be a way of caffeine
control method controlling ± 25 % difference if compared to normal coffee consumption. This
is the only caffeine control method other than the decaffeination, and it costs nothing from
normal coffee brewing method unlike the decaffeination method. Therefore, we named this
innovative method LeeN-Caffeination from less- and enhanced- caffeine coffee making
method developed by Seung-Hun Lee and Nikolai Kuhnert.
196
Traditional Korean green tea samples were selected to expand the applicability of the
techniques obtained from the steps of coffee process. The selected samples were regarded as
specialty tea like Jamaica Blue Mountain in coffee, employed the same Korean red-ginseng
process like steamented coffee, used different extraction method for comparing lab research
and practical consuming like different extracting solvent and the brewing was separated three
times like LeeN-Caffeination. The most important discovery in this study was that the
polyphenols are increasing by repeating the parching method in practical brewing, though the
total polyphenols in the green tea leaves were not changing. This means that the special
parching method might be soft not to destroy the polyphenols in the green tea, but it is strong
enough to weaken the tea leaves to extract polyphenols easily in practical brewing. Also it is
proven that all the methods used in coffee in this research might be adaptable to other food
process or analysis.
By conducting all the projects in different coffee processes and application, it is found that the
LC-MSn method could be useful for comparing different developments on different coffee
processes. The continuous searching of new developments of coffee products and adapting the
identification methods are important for the coffee consumer’s health.
The coffee market has been increasing and a lot of new processes have been invented in all
different processes of the coffee processing steps. A new fermentation control methods,
sprout coffee, roasting machines only using electricity, transparent coffee, Nitrogen brewing
and all sort of new coffee products are developed and consumed without careful identification.
There could be special composition changes by controlling the coffee processes and some
composition changes can have critical effect for specific diseases; caffeine for high blood
pressure, chlorogenic acids for type II diabetes or acidity for dental health. The result of this
research is expected to be used for the new developments of coffee in order to give correct
197
information so that all coffee consumers can select coffee according to their individual health
conditions.