of Surrev CAFFEINE CONTENTS IN WET-, DRY- AND ...epubs.surrey.ac.uk/851634/1/13803878.pdfUniversity...

University of Surrev

THE CHLOROGENIC ACID AND

CAFFEINE CONTENTS IN WET-, DRY-

AND MONSOON- PROCESSED

COMMERCIAL GRADES OF GREEN

AND ROASTED INDIAN COFFEE

BEANS

A thesis submitted in accordance with the requirements o f the University o f

Surrey, fo r the degree o f CdCte&Sof Philosophy 1998

Food Safety Research

Group

The Nutrition and Food

Safety Research Centre,

School of Biological

Sciences

University of Surrey,

Guildford,

Surrey , United

by

Kondalkana Jayarama Balyaya, BSc, MSc.

ProQuest Number: 13803878

All rights reserved

INFORMATION TO ALL USERS The qua lity of this reproduction is d e p e n d e n t upon the qua lity of the copy subm itted.

In the unlikely e ve n t that the au tho r did not send a co m p le te m anuscrip t and there are missing pages, these will be no ted . Also, if m ateria l had to be rem oved,

a no te will ind ica te the de le tion .

uestProQuest 13803878

Published by ProQuest LLC(2018). C opyrigh t of the Dissertation is held by the Author.

All rights reserved.This work is protected aga inst unauthorized copying under Title 17, United States C o de

M icroform Edition © ProQuest LLC.

ProQuest LLC.789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 4 81 06 - 1346

ABSTRACT

This research study reports the content of individual chlorogenic acids and caffeine in

commercial grades of Indian wet-, dry- and monsoon- processed Arabica and Robusta

coffees. It also reports on the composition of roasted coffee beans as obtained by

reversed-phase high-performance liquid chromatography (HPLC). A partial study

involved roast and beverage quality of a series of roasted coffees of the above,

covering the lightest and darkest roasts up to the burnt, especially in the case of dry-

and wet- processed Arabica and Robusta coffee. Significant differences with reference

to individual chlorogenic acids (CGA), its isomers, sub groups, and the total CGA

contents were observed between wet-, dry- and monsoon- processed coffee beans of

different main grades, immature and mature dry processed Robusta AB (flat beans)

grade green coffee.

Monsooned Arabica and Robusta coffees were characterised by significantly higher

contents of caffeic acids and had higher total CGA contents than wet- and dry-

processed coffees.

In the case of wet- and dry- processed Arabica coffee, CGA content steadily decreased

during the initial phase, followed by a rapid decrease at extreme degrees of roasting.

An apparent increase in 3-CQA and 4-CQA isomers as 5-CQA and di-CQA contentis Hit'

decreases/due tc/pyrolytic effect on 5-CQA and di-CQA. There is a rapid decrease in

all CGA isomers at the darker degree of roast.

There is a proportionate increase in feruloylquinic acid (FQA) content from light-

medium and medium-dark roasted Arabica and Robusta coffees. From this

investigation, it is evident that about 50 percent of CQA and di-CQA contentam

disappeared at a certain degree of roast and/apparent increase in FQA content which

may correlate with the roast and beverage quality in both these coffees.

There is a considerably higher content of chlorogenic acids and caffeic acid in both

monsooned coffees than in wet- and dry- processed. This may be correlated with fully

matured coffee beans of dry and monsooning processes. It has also been observed that

CGA content rapidly disappears at shorter degree of roasting times in both monsooned

Arabica and Robusta coffees. FQA content increases at 5 to 6 minutes roasting time

and there is a rapid moisture and roasting loss. This is required for medium to standard

roasting of monsooned coffees.

The kinetics of chlorogenic acids degradation in different processed Indian Arabica and

Robusta coffees during roasting have been studied. The rate of degradation of CQA,

di-CQA and FQA sub groups and total CGA content have been reported during

roasting.

Dedicated

to

my mother Mrs. Lakshmi Gopalakrishna

and

loving memory of my (late) wife Mrs. Vathsala Jayarama

Declaration

I declare that this thesis has not been previously submitted and is not

currently being submitted, for any other degree than that of jplosfeiv of

Philosophy of the University of Surrey.

All work reported in this thesis is my own, except where acknowledged in

the text.

Sicj wxttOUA) fi

le

Candidate

Supervisor

1998

iv

CONTENTSAbstract.........................................................................................................................i

Acknowledgments...................................................................................................... ix

Abbreviations.............................................................................................................. x

List of figures.............................................................................................................. xi

List of tables............................................................................................................. xiii

Chapter 1 Introduction............................................................................1.1 Brief history of coffee...........................................................................................1

1.2 Processing of green coffee.................................................................................... 2

1.3 Grades of Indian coffee.........................................................................................4

1.4 Composition of green coffee................................................................................. 6

1.4.1 Carbohydrates..........................................................................................7

1.4.2 Lipids...................................................................................................... 8

1.4.2.1 Crude lipid content....................................................................8

1.4.2.2 Triglycerides............................................................................. 8

1.4.2.3 Diterpenes................................................................................. 9

1.4.2.4 Sterols........................................................................................9

1.4.3 Proteins and free amino acids.............................................................. 10

1.4.3.1 Crude proteins..........................................................................11

1.4.3.2 Crude proteins corrected for alkaloid nitrogen........................11

1.4.3.3 Free amino acids...................................................................... 11

1.4.4 Minerals............................................................................................... 12

1.4.5 Chlorogenic acids.................................................................................12

1.4.6 Alkaloids.............................................................................................. 15

1.4.6.1 Caffeine....................................................................................15

1.4.6.2 Trigonelline..............................................................................16

1.5 Roasting and other processes.............................................................................. 17

1.6 Indian coffee quality............................................................................................18

1.6.1 Visual quality.................................................................................... . 19

1.6.2 Cupping evaluation............................................................ 20

1.7 Analytical methods................................................................................................20

1.8 Aim and Objectives...............................................................................................21

Chapter 2 Materials and Methods...................................................... 232.1 Materials............................................................................................................... 24

2.1.1 Arabica coffee.........................................................................................24

2.1.1 1 Wet-processed.................................................................................... 24

2.1.1.2 Dry- processed.................................................................................... 25

2.1.1.3 Monsooned Malabar AA..................................................................... 26

2.1.2 Robusta coffee.....................................................................................27

2.1.2.1 Wet processed (washed) coffee............................................... 27

2.1.2.1.1 Robusta parchment PB coffee..................................27

2.1.2.1.2 Robusta parchment AB coffee..................................28

2.1.2.2 Dry- processed (unwashed) coffee..........................................29

2.1.2.2.1 Robusta (unwashed) cherry PB coffee..................... 29

2.1.2.2.2 Robusta (unwashed) cherry AB coffee.................... 30

2.1.2.3 Monsooned Robusta A A .........................................................31

2.1.3 Chemicals............................................................................................ 32

2.2 Methods............................................................................................................... 32

2.2.1 Extraction...............................................................................................32

2.2.2 Analytical HPLC ........................................................................... 33

2.2.2.1 Standard calibration for standard 5-CQA................... 34

2.2.2.2 Standard calibration for standard caffeine.................. 35

2.2.3 Moisture content.................................................................................... 35

2.2.4 Roasting..................................................................................................36

2.2.4.1 Moda coffee bean sample roaster............. 36

2.2.4.2 Grinding of roasted coffee samples............................ 36

2.2.5 Brewing................................................................................................36

2.2.6 Evaluation of coffee beverage.............................................................37

2.2.7 Storage.................................................................................................37

Chapter 3 Results and Discussion : Green coffee..............................383.1 Results.......................................................................................... 39

3.1.1 Moisture content of green coffee beans................................................. 39

3.1.2 Identification of peaks on chromatogram.............................................. 39

3.1.3 Individual chlorogenic acids and caffeine content in Indian Arabica

green coffee..................................................................................................... 45

3.1.4 The chlorogenic acids and caffeine contents in Indian Robusta green

coffee.............................................................................................................. 46

3.1.5 The chlorogenic acids and caffeine contents in immature and fully

mature dry processed Indian Robusta green coffee.........................................47

3.2 Discussion............................................................................................................. 47

3.2.1 Moisture content of green coffee............................................................47

3.2.2 The chlorogenic acids and caffeine content of Indian Arabica green

coffee.............................................................................................................. 48

3.2.3 Chlorogenic acid and caffeine contents in Indian Robusta green coffee49

3.2.4 Chlorogenic acid and caffeine contents in immature and fully mature dry

processed AB grade Indian Robusta green coffee........................................... 50

3.3 Conclusion.............................................................................................................50

Chapter 4 Results and Discussion :

Roasted coffee and Beverage quality......................................... 524.1 Roasted coffee analysis....................................................................................... 53

4.1.1 Indian Arabica coffee............................................................................. 53

4.1.1.1 Washed Indian Arabica A grade coffee...................................53

4.1.1.2 Unwashed Indian Arabica AB grade coffee............................58

4.1.1.3 Monsooned Arabica AA grade coffee..................................... 59

4.1.2 Indian Robusta coffees........................................................................... 61

4.1.2.1 Unwashed Robusta AB grade coffee....................................... 61

4.1.2.2 Unwashed Robusta PB grade coffee....................................... 63

4.1.2.3 Washed Robusta AB grade coffee........................................... 64

4.1.2.4 Washed Robusta PB grade coffee........................................... 68

4.1.2.5 Monsooned Robusta AA grade coffee.....................................70

4.2 Beverage Quality:.................................................................................................. 72

4.2.1 Arabica coffee.........................................................................................72

4.2.1.1 Washed Arabica A grade coffee.............................................. 73

4.2.1.2 Unwashed Arabica AB grade coffee.......................................74

4.2.1.3 Monsooned Malabar (Arabica) AA grade coffee....................75

4.2.2 Robusta coffee........................................................................................76

4.2.2.1 Unwashed Robusta AB grade coffee.......................................76

4.2.2.2 Unwashed Robusta PB grade coffee.......................................77

4.2.2.3 Washed Robusta AB grade coffee...........................................78

4.2.2.4 Washed Robusta PB grade coffee...........................................79

4.2.2.5 Monsooned Robusta AA coffee.............................................. 80

4.3 Conclusion.............................................................................................................81

Chapter 5 Kinetics of Chlorogenic Acids : Degradation in

Indian Arabica and Robusta coffees during roasting.............. 825.1 Introduction......................................... 83

5.2 Materials and Methods..........................................................................................83

5.3 Results and Discussion..........................................................................................84

5.3.1 Kinetics of Chlorogenic Acids degradation in Indian washed, unwashed

and monsoon- processed Arabica coffees....................................... 88

5.3.2 Kinetics of Chlorogenic Acids degradation in Indian washed, unwashed

and monsoon- processed Robusta coffees.......................................................88

5.4 Conclusion........................................................ 92

Chapter 6 General Discussion and Conclusion...................................936.1 Proposal for future work.................................................................................... ..97

References............................................................................................. 98

Appendices............................................................................................107

ACKNOWLEDGEMENTS

The author is greatly indebted to Dr. Michael Newton Clifford, B.Sc., Ph.D.,

MIFST, Leader—Food Safety Research Group and Reader in Food Science, a world

coffee and tea scientific expert, for his keen continuous supervision, encouragement,

guidance, incredible advice and critical suggestions for the successful completion of

this document of research.

Thanks are due to M/s Consolidated Coffee Limited, Pollibetta, Kodagu,

Karnataka, India, for the supply of green coffee beans used during this research

study. The author is particularly grateful to Mr. T Kazi of the Excelsior Coffee and

Tea Company for providing roasting facilities.

The Government of India for the award of National Overseas Scholarship for

this study will be specially acknowledged.

I am also highly obliged to Mr. C. Rajendra Prasad, Managing Director,

Continental Coffee Ltd, India (formerly managing Director of Asian Coffee Ltd,

India), for his support and encouragement. Sincere thanks also go out to Dr. Ronald

Clarke, formerly of General Foods Ltd, UK, Mr. Celsius A. Lodder, Executive

Director of International Coffee Organisation UK, Ms. Gill Hooper, Argus Business

Media Limited, Surrey, Mr. Ted Lingle, Executive Director, Speciality Coffee

Association of America, and others who have involved directly or indirectly helping

me pursue this research.

My sincere thanks are also due to Professor Fred Hoskins, Washington State

University, USA Professor Harry E. Nursten, Professor of Food Science, University

of Reading, UK and Professor Ron Walker, Professor of Food Science, University of

Surrey, UK, for their critical evaluation and valuable suggestion in making and

improving this thesis of research.

The co-operation of all staff, technicians and all my research colleagues in the

Food Safety Research Group and in the School of Biological Sciences are specially

acknowledged.

Finally, I am thankful to all my family members for their loving and prayerful

support throughout this part of my academic and research career and especially my

mother, without whom I would never have completed this thesis.

ABBREVIATIONS

AO AC - Association of Official Analytical Chemists

ACN - Acetonitrile

CA - Caffeic acid

CGA - Chlorogenic acid

CQA - Caffeoylquinic acid

CFQA - Caffeoylferuloylquinic acid

CoA - Coumaric acid

CoQA - Coumaroylquinic acid

FA - Ferulic acid

FAQ - Fair average quality

FQA - Feruloylquinic acid

d.b. - dry basis

DiCQA -Dicaffeoylquinic acid

FCQA - Feruloylcaffeoylquinic acid

GC - Gas chromatography

GLC - Gas-liquid chromatography

HPLC - High performance liquid chromatography

ISO - International standard organisation

IUPAC- International Union of Pure and Applied Chemistry

MS - Mass Spectroscopy

S.d. - Standard deviation

TMAH- Tetramethylammonium hydroxide

TFA - Trifluoroacetic acid

QA - Quinic acid

x

LIST OF FIGURES

1.1 Fluxogram method of processing Indian coffee................................................ 3

1,2a Structure of a quinic acid in preferred chair conformation............................... 13

1,2b Structure of esterifying residues found in coffee bean chlorogenic acids.........13

1.3 Structure of caffeine..........................................................................................16

1.4 Mechanical principles in roasting methods.......................................................17

2.1a Plantation (washed Arabica) A grade coffee................................................... 24

2. lb Arabica cherry (unwashed) AB grade coffee................................................... 25

2.1c Monsooned Arabica (Malabar) AA grade coffee............................................. 26

2.2a Robusta parchment (washed) PB grade coffee................................................ 27

2.2b Robusta parchment (washed) AB grade coffee................................................ 28

2.2c Robusta (unwashed) cherry PB grade coffee................................................... 29

2.2d Robusta (unwashed) cherry AB grade coffee.................................................. 30

2.2e Monsooned Robusta AA grade coffee..............................................................31

2.3 Schematic diagram of the HPLC system.......................................................... 33

2.5 Standard calibration curves o f:...........................................................................

a) 5-CQA...............................................................................................................34

b) Caffeine.............................................................................................................35

3.1. Chromatogram of a70 % methanolic extract of an Indian washed green

Arabica A coffee.............................................................................................41

3.2 Chromatogram of a70 % methanolic extract of an Indian washed green

Robusta AB coffee .......................................................... ..............................42

4.1a CQA content vs Degree of Roast of Indian Washed Arabica A grade

Coffee............................................................................................................ 56

4. lb FQA content vs Degree of Roast of Indian Washed Arabica A grade

Coffee............................................................................................................ 56

4.1c Di-CQA content vs Degree of Roast of Indian Washed Arabica A grade

Coffee............................................................................................................ 56

4.1 d Chlorogenic acid and caffeine content vs Degree of Roast of Indian Washed

Arabica A grade Coffee...................................................................................57

4.2a CQA content vs Degree of Roast in Indian Washed Robusta AB Grade Coffee

........................................................................................................................ 67

4.2b FQA content vs Degree of Roast in Indian Washed Robusta AB Grade Coffee

4.2c

4.2d

4.2e

5.1a

5.1b

......................................................................................................................... 67

Di-CQA content vs Degree of Roast in Indian Washed Robusta AB Grade

Coffee........................................................ 67

CFQA + FCQA and caffeoyl tryptopahan vs Degree of Roast in Indian

Washed Robusta AB Grade Coffee.................................................................. 68

CGA Sub-group and caffeine contents vs Degree of Roast in Indian Washed

Robusta AB Grade Coffee.............................................................................. 68

PMoP C qb Cln\bda6(\ld<

c&YXc ft Q&j$ap, > ■ * - * * - * * * - - - - 86

serd&nl' C

■fdv ppg&ix £0fjju, 86

xii

LIST OF TABLES

1.1

1.2

1.3

1.4

1.5

3.1

3.2

3.3

3.4

3.5

3.6

4.1a

4.1b

4.1c

4.2a

4.2b

Grade designations of Indian green coffee....................................................... 5

Green coffee composition................................................................................. 6

Composition of the lipid fraction of green coffee beans..................................10

Roast coffee composition.................................................................................18

Analytical methods for determination of coffee composition......................... 21

Moisture content of Indian green coffee..........................................................39

Retention time (minutes), relative retention times and Xmax. for the individual

chlorogenic acids and caffeine......................................................... ; ............. 43lh-

The content of individual chlorogenic acids content( % dry basis)/nine sets of

Indian Robusta( washed) AB grade green coffee......................................... 44

Content of individual chlorogenic acids and caffeine (% dry basis) in Indian

Arabica green coffee beans............................................................................. 45

The chlorogenic acids and caffeine content in Indian Robusta green coffee

beans............................................................................................................... 46

The content of individual chlorogenic acids and caffeine contents (% dry

basis) of immature and fully mature dry processed Indian Robusta green

coffee beans..................................................................................................... 47

The data for chlorogenic acid and caffeine contents (% dry basis) for green

coffee of washed Arabica A grade (mean of triplicate determinations) and the

meansof duplicate determinationsfor roasted coffee........................................ 54


coffee of unwashed Arabica AB grade (mean of triplicate determinations) and

the meansof duplicate determinationsfor roasted coffee...................................58


coffee of Monsooned Arabica AA grade(mean of triplicate determinations)

and the meansof duplicate determinationsfor roasted coffee............................60


coffee unwashed Robusta AB grade(mean of triplicate determinations) and the

meansof duplicate determinatiomfor roasted coffee........................................ 62


coffee unwashed Robusta PB grade(mean of triplicate determinations) and the

meansof duplicate determinationsfor roasted coffee........................................64xiii

4.2c The data for chlorogenic acid and caffeine contents (% dry basis) for green

coffee washed Robusta AB grade(mean of triplicate determinations) and the

meansof duplicate determinationsfor roasted coffee........................................66

4.2d The data for chlorogenic acid and caffeine contents (% dry basis) for green

coffee washed Robusta PB grade(mean of triplicate determinations) and the

meansof duplicate determinationsfor roasted coffee........................................69

4.2e The data for chlorogenic acid and caffeine contents (% dry basis) for green

coffee monsooned Robusta AA grade(mean of triplicate determinations) and

the meansof duplicate determinationsfor roasted coffee..................................71

4.3 Beverage quality evaluation reports o f :............................................................

4.3.1a Washed Arabica A grade coffee......................................................................72

4.3.1b Unwashed Arabica AB grade coffee...............................................................74

4.3.lc Monsooned Malabar ( Arabica ) AA grade coffee..........................................75

4.3.2a Unwashed Robusta AB grade coffee...............................................................76

4.3.2b Unwashed Robusta PB grade coffee...............................................................77

4.3.2c Washed Robusta AB grade coffee...................................................................78

4.3.2d Washed Robusta PB grade coffee.............. 79

4.3.2e Monsooned Robusta AA grade coffee.............................................................80

5.1

" \6 f 9U )aU ^nf . -3

f i r d i fo o fo ' 87

52 The C6? Pi e& nJjLm t C 7? ■ (a

^YiclXGm Pjdiust<^A> : ' 1.................................................................. 89

xiv

CHAPTER 1

INTRODUCTION

1.1 Brief History of Coffee

Coffee was discovered about 1000 years ago in a province called Kaffa in

Ethiopia. It is an indisputable fact that coffee is the second largest commodity in the

world in international trade after oil. The Latin name Coffea is derived from one of the

Arabic words Caova cova or kahwah, qahwah. The history of coffee cultivation and its

use as a non-alcoholic beverage is centred in Arabia and then spread to Aden, Cairo,

Persia, Turkey and Venice. Arabica coffee, Coffea arabica, was introduced to India in

1600 AD by a Muslim pilgrim, Baba Budan. He is reported to have had seven seeds

from Yemen, presumably Mocha coffee, and raised seedlings on his hermitage on the

hills near Chikmagalore in the Karnataka State of India (Anon., 1985).

Coffee has been the most sought after refreshing drink. In the olden days, it

was believed that coffee could comfort the brain, help relieve pains in the head,

lethargy, cough, indigestion and prevent sleepiness. It was also considered as a useful

remedy for rheumatism, gout and fever. Today, as a stimulating non-alcoholic

beverage, it is very popular, with various brands in the market jostling each other for

shelf space and consumer preferences.

Coffea belongs to the family Rubiaceae and comprises evergreen shrubs and

small trees. The four commercially important species are: Coffea arabica L., Coffea

canephora, Coffea liberica and Coffea excelsa. The former two are commercially

known as Arabica and Robusta coffees, respectively. The latter two species, namely C.

liberica and C. excelsa, are less commercially important and are small tree coffees.

There are many botanical varieties and cultivars in both Arabicas and Robustas found

and developed by the coffee research units world wide.

1

1.2 Processing of green coffee

In India the two major species of coffee, i.e., Coffea arabica L. and Coffea

canephora var. robusta (Robusta), are grown in southern states, especially in

Karnataka, Kerala and Tamilnadu. Worlds' 1995/96 crop production will be

82,577,000 bags (60 kg/bag), comprising 26,603,000 bags of Robusta and 55,974,000

bags of Arabica green coffees { ^nr,: « 1996). India expected

to produce a record crop in 1995/96 of 1.32 million bags. Arabica production in

1995/96 is expected to reach 1.72 million bags while Robusta coffee production would

be around 2.00 million bags which is 18 percent more than was produced in 1994/95.

The Indian Coffee Board has stated that the quality of Indian Robusta coffee is

one of the best in the world as far as green coffee and cup quality are concerned

(Menon, 1984). Indian coffee is famous for its use in the best blends in the world

coffee market.

Indian coffee is processed mainly by two methods, i.e., the dry process and the

wet process. Dry processing of coffee yields a product described as “unwashed” or as

‘cherry coffee’. Wet processing yields ‘parchment’ or ‘washed coffee’. In addition to

the above, there is an additional method of processing of coffee found exclusively in

India, known as the monsooning process, yielding ‘monsooned coffee’.

2

The three distinct processing methods are shown in the following Fluxograms 1

Harvesting

Sorting

Monsoon process

Selected whole crop A-grade cherry clean green coffee monsooning with the moisture content being in the range o:' 13 to 14.5 % (prepared on the west coast . at curing work* during monsoon period June- September months)

Grading as per the size, such as Monsoon Malabar AA, Monsoon Basanally, Monsoon Arabica triage, MonsoonRobusta AA, MonsoonRobusta triage

PB

Garbling

AB PB AB BBB

Grading

Cherry or unwashed clean green coffee

GarblingParchment or washed clean green coffee

Wet process Dry process

Washing

Pulping

Demucilaging or washing

Fermentation/treatment wilh alkali/removal of mucilage by friction____________

Parchment coffee

Sun drying

Dried parchment coffee

Sun drying

Cherry coffee

Hulling

Polishing

3

1.3 Grades of Indian coffee

Indian coffee processed by the wet method is called parchment coffee, whereas

the dry method produces cherry coffee. Indian washed (parchment) Arabica coffee, is

also called Plantation coffee and dry-processed Arabica and Robusta coffees are

known as Arabica cherry coffee and Robusta cherry coffee, respectively. Indian wet-

processed (washed) Robusta coffee is known as Robusta parchment coffee in the

trade. Indian coffee is graded based on the size of the beans and percentage of

imperfections. For example, the Indian plantation coffee grades are ‘Mysore nuggets

EB’ (extra bold), Plantation PB (Pea berry - rounded beans), A, B, C, Blacks and Bits

and Arabica unwashed (cherry) coffee grades are PB, AB, C, B/B/B (Blacks, Browns

and Bits). Robusta unwashed (cherry) and Robusta washed (parchment) coffee grades

are ‘Robusta Kaapi Royale’ (extra bold beans), PB, AB, C, and B/B/B, and finally the

grades of Monsooned Arabica coffee grades are Monsooned Malabar AA, Monsooned

Basanally and Monsooned Robusta as AA. The standard moisture content of the

washed coffees is in the range of 10-11 %, with allowed permissible variation at +

0.5% for washed and unwashed coffees, respectively. The moisture standards for the

monsooned coffees were 13 to 14 percent. Mysore Nuggets EB (extra bold) grades of

Indian Arabica and Robusta coffees are the speciality of Indian coffees

Grade designations of Indian green coffees and standards as referred in this

research study are shown in Table 1.1.

4

Tabl

e 1.1

: G

rade

desig

natio

ns o

f Ind

ian

green

co

ffee.

CA

12h bo-5■o •- c" *0Vi

oNJJ

w> .’Sfl V3*p« wn

I ar h « O a.

-ai-asnsfiCd4-»CA

00s*>.2*

jo<u00cd"S<DOJ-Ha>

P h

T3<U.s"34->CD

J3op'53£

a)N

*CA

0)

&.B.

so13 * 2 s o .2P

OQ

a>a.►»H

<d00cd

<D00cd

«4-(om cd

PQO h<X*

O<N

«+iO

CN

<41O

<D 0 0 • 2

<d CD0 0 0 0 00cd t—i +-> cd cdu,

Cm

<41o

* £4-> ’C4-><+4

O i n O o<N <N m

o oON ON

oO n

O O O O n O n O n

o m O O m oo NO 1 ° . . o ( N oNO NO NO NO r - ‘

o

aCDa

J poc3CXWho

T 3<D

^ 'm<D

^ § O >ao

T 3<D

J pCAcd

IP

S 'cd

o

T3CD-aCAcd

£

Ib IJ-* o <d u

cdo ex

cd

£cd 5 * -P otn cd

+_> cd -pfl-S “ c d cd P£ Id "§■s s <§

•*

cdexcd cd cdCAP•§

CA CA

P P JO X> O Op4

<d <uc c cd o Oca O O~ CA CAR e a o oi2 S S

1 Tria

ge

: Con

sists

of br

oken

be

ans

of no

t les

s tha

n on

e-th

ird

size

of the

who

le fla

t be

an,

with

ered

, sh

rivell

ed,

spot

ted,

elep

hant

- siz

ed

bean

s, sm

all,

disc

olou

red,

malf

orm

ed

bean

s, pa

les,

and

bean

s da

mag

ed

durin

g the

pu

lping

pr

oces

s( p

ulpe

r cu

ts).

1.4 Composition of green coffee

tf>Coffee has been much researched, and/one of the most popular beverages

throughout the world. Green coffee as such is devoid of any characteristic coffee

colour and flavour. The green coffee bean has a complex chemical composition and

varies according to the species, variety, climatic conditions and location, production

and other factors such as method of cultivation and processing. Analytical results also

depend on the methods of test, equipment used and whether modem or long-

established laboratory techniques have been employed.

Clarke (1995) tabulated the average chemical composition of the two main

species of coffee, and these are shown in Table 1.2

Table 1.2 : Green coffee composition

Component Typical average content (% dry basis)

Arabica RobustaCaffeine 1.2 2.2Trigonelline 1.0 0.7Minerals (as oxide ash; 41% K and 4% P) 4.2 4.4AcidsTotal chlorogenic 6.5 10.0Aliphatic 1.0 1.0Quinic 0.4 0.4SugarsSucrose 8.0 4.0Reducing 0.1 0.4Arabinogalactan, mannan and glucan 44.0 48.0Others 1.0 2.0Lignin 3.0 3.0Pectin 2.0 2.0ProteinaceousProtein 11.0 11.0Free amino acids 0.5 0.8LipidsCoffee oil (Triglyceride with unsaponifiables) 16.0 10.0

TOTAL 100.0 100.0

6

The main non-volatile components of green coffee are carbohydrates, lipids,

proteins, chlorogenic acids, minerals, alkaloids (caffeine) and small quantities of

trigonelline, free aliphatic and quinic acids.

1.4.1 Carbohydrates:

Green coffee contains both low and high molecular weight carbohydrates.

Sucrose, the major sugar, is present in both Arabica and Robusta green coffees. It has

been reported that Arabica green coffee contains more free sugar than Robusta. In

duplicate samples of Arabica and Robusta green coffee, Tressl et al. (1982) determined

free sugars by gas chromatography and reported that Arabica coffee contained 8.2-8.3

% on a dry basis and Robusta coffee 3.3 - 4.1 % d.b. Trugo et al. (1982) determined

free sugars by high-performance liquid chromatography (HPLC) and reported that

Arabica and Robusta green coffee contain 6.1% d.b. and 3.4 % d.b., respectively. The

presence of a very small quantity of other simple sugars such as glucose and fructose

were reported by Tressl et al (1982) in two samples of Arabica and Robusta green

coffees. According to Trugo (1985) the nature and content of these simple sugars

appears to be important for the flavour development and caramelisation products in

roasted coffee.

Polysaccharides, such as arabinogalactan, mannan and glucan, are important

constituents of green coffee and present in the range of 44 - 48 % d.b., as reported by

Bradbury et al. (1987). The importance of various polysaccharides fractions on flavour

retention and their affect on coffee quality has yet to be established (Trugo et al.,

1982) and their contribution to the yield of soluble coffee is not known in Arabica

coffee. Robusta contains more arabinogalactan.

7

1.4.2 Lipids

Substantial amounts of lipids, in the form of coffee oil are present in green

coffee. Coffee oil consists of the following:

• Crude lipid content

• Triglycerides

• Diterpenes

• Sterols

1.4.2.1 Crude lipid contents

The lipid content of green coffee beans is described as crude lipid, which

includes any substance extracted by a specified solvent and may thus include non

lipids.

Average figures of lipid contents in green coffee are shown in Table 1.2, mainly

based on the compilations of Maier (1981), Clifford (1985b), and others. Streuli (1973)

reported that Arabica green coffee contains more lipid, i.e. 15.0 ± 0.78 % d.b., than

that of Robusta green coffee, which contains 10.0 ± 1.41 % d.b. The lipid fraction of

coffee is mainly composed of triglycerides. The composition of the lipid fraction of

green coffee beans was summarised by Viani (1986) and is reproduced in Table 1.3.

1.4.2.2 Triglycerides

About 70-80 % of total lipid content in green coffee is trigyceride molecules. It

has been reported that in the coffee bean oil triglyceride molecules, linoleic acid and

palmitic acid are predominate (Kaufmann & Hamsager, 1962)

Free fatty acid content is lower in Arabica (1.0-1.5%) than in Robusta green

coffee (1.0-2.7%) (Speer et al., 1993). Triglycerides are little affected by roasting

except for slight hydrolysis and decomposition with liberation of free fatty acids and

formation of volatile components, possibly through oxidation.

1.4.2.3 Diterpenes

The diterpene alcohols namely, cafestol which occurs in a larger percentage,

and a lower level of kahweol, are present in unsaponifiable matter of coffee oil and

their structures have been elucidated by Djerassi et al. (1958b) and Haworth et

al. (1957). The diterpenes are partially degraded with loss of water, with exception of

16-O-methylcafestol for Robusta which is thermally quite stable and characteristic of

Robusta (Speer et al., 1991). The ratio cafestol: kahweol can vary in Arabica between

40 : 60 and 70 : 30 ( Viani, 1988). The diterpenes can be determined by capillary gas

chromatography and a very distinct peak of 16-O-methylcafestol is an indication of

Robusta coffee.

1.4.2.4 Sterols

Sterols are found in free and esterified form in coffee. The sterol fraction in

Arabica and Robusta shows some variation between the species( Mariani and Fedeli,

1991). In Robusta these are higher levels of 24-methylenecholesterol and V5 ’

avenasterol and in Arabica sitostanol is higher. The determination of sterols in green

and roasted regular and decaffeinated coffee is by using capillary gas chromatography

(Mariani and Fedeli, 1991)

In addition, traces of di-amines, putrescine, spermine and spermidine, have also

been identified in green coffee beans and these readily decompose while roasting.

9

Table 1.3 : Composition of the lipid fraction of green coffee beans

Lipid contents Proportion wt %_ _ ComponentsTriglycerides mainly esters of linoleic and

palmitic acidsFree fatty acids (as % oleic) 0.5-2.0 mainly esters of palmitic andDiterpene esters 15-0-18.5 linoleic acids

16-O-methylcafestol inTriterpene, sterol, and 1.4-3.2 Robusta only and inmethylsterol esters Arabicas kahweol andFree diterpenes 0.1-1.2 atrotligenin mainly

sitosterol, stigmasterol andFree triterpenes and sterols 1.3-2.2 campesterol

Phospholipids 0.1

Hydrocarbons traces mainly squalene andnonacosane

5 -Hydroxytryptamides 0.3-1.0 amides of arachidic,behenic, and lignoceric acids

Tocopherols 0.3-0.7 a-, P-, and y- isomers

Speer (1989) quantified the cafestol and kahweol content in a variety of

Arabica, Robusta and hybrid (of Arabica and Robusta) coffee samples. An interesting

16-O-methyl derivative of cafestol has also been reported in coffees by Speer et al.

(1989; 1991) by using an HPLC. Subsequently in 1993, Speer et al. reported that

Arabica and Robusta coffees contain about 15 % and 10 % lipid, respectively, and also

determined free fatty acid contents in green Arabica and Robusta coffees by using the

permeation chromatography method.

1.4.3 Protein and free amino acids

In green coffee, the proteins are predominantly present in an unbound form in

the cytoplasm, but some are bound to polysaccharides in the cell walls. Apart from the

amino acids bound in protein, some free amino acids are present in green coffee

(Macrae, 1985)v

10

1.4.3.1 Crude protein

Based on the determination of crude nitrogen and multiplication by the factor

6.25, various values for the protein content in green coffee have been reported. Crude

protein values found in green coffees have been reported in the range of 13-16 % by

Barbiroli(1965). These values include non-protein nitrogen from caffeine, trigonelline,

etc.

1.4.3.2 Crude protein corrected fo r alkaloid nitrogen

Real protein values are reported as 10-12.5 %, after correction is made for

caffeine-nitrogen, by Roffi et a l, 1971 and Thaler et a l, 1963. In green coffee,

proteins are present in the form of water-soluble (albumin) and water- insoluble

proteins, 50 % of the total being water-soluble proteins (Macrae, 1985).

Characterisation of water-soluble protein fractions by many techniques has been

reported by Amorim et al.,(1977). Thaler and Gaigl (1963) quantified the percentage

of total protein and albumin in Arabica and Robusta green coffees.

1.4.3.3 Free amino acids

Tressl (1980 ) stated that free amino acids found in the green coffee bean may

have an influence on the aroma of coffee and thus beverage quality. The proteins in

green coffee consist of a water-soluble and a water-insoluble fraction. The free amino

acids and protein contents of green and roasted coffee beans have not been thoroughly

studied (Maier, 1993). It has been reported that most of the proteins and carbohydrates

in green coffee react during roasting and thus form melanoidin (Steinhart et a l, 1993)

1.4.4 Minerals

The ashed mineral content of green coffees mainly consists of potassium and

other small quantities of calcium and magnesium, and the non-metallic elements

phosphorus and sulphur.

11

Kroplein (1961) has made a detailed investigation of ash contents of different

Arabica and Robusta green coffees with consistent results. Consistent values of ash

contents of Arabica and Robusta green coffee beans have been reported using

standard ashing temperature of 580 °C (Clarke and Walker, 1974 and 1975). The ash

content of green coffee is about 4% d.b., of which about 40 % is potassium with a

further 33 % of other trace elements having been quantified. Clarke and Walker (1974,

1975) reported that there was a real difference in average content of potassium

between dry- and wet-processed green coffee, - •.•••-u the former

having a slightly higher level of potassium. They suggested the lower content of

potassium in wet-processed coffees to be due to leaching out of potassium and other

minerals during wet processing.

1.4.5 Chlorogenic acids

Chlorogenic acids (CGA) are the most studied and major acids of coffee and

contribute a series of esters of quinic acid and a very small quantity of free quinic acid

also occurs in green coffee beans.

An extensive study on chlorogenic acids in coffee beans has been carried out

mainly by Clifford(1985), Macrae(1985) and Trugo(1984), and Ohiokpehai(1982).

Robiquet et a l,(1837) probably first reported these acids and later Payen (1846)

designated this green-tinged crystalline substance extracted from green coffee beans as

‘chlorogenic acid’. This particular compound was established as 3-CQA (Freudenberg,

1920), and later on redefined as 5-CQA as per IUPAC (1976).

Chlorogenic acids(CGA)were defined by Clifford (1985) as quinic acids

esterified with/>-coumaric acid, caffeic acid and or ferulic acid and are now subdivided

into the following groups:

CQA caffeoylquinic acids

di-CQA dicaffeoylquinic acids

FQA feruloylquinic acids

p -CoQA />-coumaroylquinic acids

12

CFQA caffeoylferuloylquinic acids

FCQA feruloylcaffeoylquinic acids.

The structure of quinic acid in its preferred conformation (carboxyl equatorial)

with the ring carbons numbered according to the IUPAC recommendation (IUPAC,

1976) is shown in Fig. 1.2a. The CGA found in coffee are mono- and di-esters at

positions 3, 4 and 5. The structures of the esterifying residues (is-cafifeic acid, fs-ferulic

acid and £-p-coumaric acid) associated with coffee are shown in Fig. 1.2b,

respectively.

OH 4

O

\/

OH

OH

OH

OH

Fig.l.2a : Structure of a quinic acid in preferred chair conformation

COOH

R = H = p- coumaric Acid R = OH = caffeic Acid R = OCH3 = ferulic Acid

Fig. 1.2b : Structure of esterifying residues found in coffee bean chlorogenic

acids

13

The six CGA sub-groups are known respectively as :

• caffeoylquinic acids (CQA) which are esters of caffeic acid with quinic acids;

• /?-coumaroylquinic acids (p-CoQA) which are esters of p-coumaric acid with

quinic acid;

• . feruloylquinic acids (FQA) which are esters of ferulic acid with quinic acid;

and the recently characterised (Clifford, 1986).

• dicaffeoylquinic acids ( di-CQA) which are esters of two molecules of caffeic

acid with one molecule of quinic acid;

• caffeoylferuloylquinic acids (CFQA) which are esters with one residue of

caffeic acid plus one residue of ferulic acid attached to the same residue of

quinic acid and

• feruloylcaffeoylquinic acids (FCQA ) - esters with one residue of ferulic acid

plus one residue of caffeic acid attached to the same residue of quinic acid.

In addition to the above, 7V-/?-caffeoyl-L-tryptophan and A-/?-caffeoyl-L-

tyrosine and traces of unesterifed ^-caffeic acids, ^-ferulic acid and E-p-coumaric acid

have also been reported (Clifford, 1985a, b.). The mixed di-esters and the amino acid

derivative, namely caffeoyl-L-tryptophan, are found only in Robusta coffees (Morishita

et al, 1987)., Clifford et a l, 1989b, c and Correia et al, 1994 identified a new

compound, namely ‘caffeoyl tyrosine’, and confirmed it as a marker compound

exclusive to Angolan Robusta coffees.

The mixed di-esters have previously been partially resolved into three peaks,

containing possibly two isomers each, namely, 3C,4-FQA +3F,4-CQA, 3C,5-

FQA+3F,5-CQA and 4C,5-FQA + 4F,5-CQA (Clifford et a l, 1989 a). One of the

most hydrophobic pair co-eluted with the leading edge of the peak of caffeoyl-

tryptophan isolated and was identified as a novel compound /?-coumaroyl-L-tryptc^ian

in green Robusta coffee beans by Murata et a l, 1995. Recently, De Maria et a l, 1995

proposed an adequate method for the simultaneous determination of total CGA,

trigonelline and caffeine in green coffee by high performance gel filtration

chromatography.

14

Various data for total chlorogenic acids in green coffee beans were summarised

by Clifford (1985b), who reported that Arabica and Robusta green coffee beans

contained about 5-7.5% and 7-10 % dry basis, respectively. A comparison between

photometric and HPLC methods of determination of chlorogenic acid has been studied

by Schunemann et al. (1986 and 1987)^howed that values for chlorogenic acids in

green coffee obtained by both methods were equal. Clifford and Jarvis (1988) reported

that chlorogenic acids vary in green Robusta coffees grown in different geographic

areas and thus might be indicative of geographic origin. DIN-Norm (1992) adopted

HPLC as the standard method for determination of chlorogenic acids.

Clifford and Kazi (1987), who reported increases of chlorogenic acids, caffeine

and trigonelline during the maturation of the fruit, found the contents of chlorogenic

acids, i.e., isomers of caffeoylquinic acids, had considerable increase in the ratio of

CQA and di-CQA. In 1994, De Menezes also reported that there were significant

differences in the ratio of CQA and di-CQA content in dry- and wet- processed

Arabica green coffees. Notable changes in the chlorogenic acids content during the

process of storage (Schunemann and Maier, 1986) and roasting (Scholz-Bottcher,

1991) are reported. Studies on the taste of pure 5-CQA and on its astringency were

carried out by Nagel et al (1987) and Naish et al (1993), respectively.

1.4.6 Alkaloids

1.4.6.1 Caffeine

Caffeine, 1,3,7-trimethylxanthine (Figure 1.3) is a much researched component

in coffee mainly due to its physiological properties of stimulation which improves

alertness and concentration. Caffeine is a white compound with a melting point of

236°C and has a marked bitter taste. Caffeine is also present in tea, cola drinks, cocoa

and chocolate, as well as some medicines, such as painkillers. In addition to this major

alkaloid, traces of theophylline and theobromine have also been reported in green

coffees.

15

Fig. 1.3. Structure of caffeine.

The wide range of values of caffeine content in green coffees was summarised

by Macrae (1985). The average caffeine content in Arabica and Robusta green coffees

is 1.2-1.5% (d.b.) and 2.2-2.5% (d.b.). For 25 years caffeine has been frequently

determined using different methods based on spectrophotometry and chromatography.

Gennaro and Abrigo (1992) determined caffeine and theobromine simultaneously

without clean-up of column by HPLC on RP-18, and Kazi (1987) with clean-up

method. Din-Norm (1992) also proposed a rapid method of determination of caffeine

without clean-up, but it yields higher values.

1.4.6.2 Trigonelline

Trigonelline is thermally unstable and hygroscopic, consisting of colourless

crystals. It is present in green coffees (Arabica 0.6-1.2%, Robusta 0.3-0.9%) (Viani,

1986). This compound has received much attention as its thermal decomposition

during roasting yields nicotinic acid, pyridine, and other volatile aroma constituents,

which are important both from sensory and nutritional points of view. However,

variability in the level of trigonelline between the species of coffee and differences in

analytical methods of determination of trigonelline may also be a factor (Macrae,

1985). Trugo and Macrae (1989) reported that the degradation of trigonelline and

sucrose in Arabica and Robusta coffee yielded nicotinic acid. They used reversed-

phase HPLC with a mass detector for determination of trigonelline and sucrose,

reversed phase HPLC for caffeine and ion-pair reversed-phase HPLC for nicotinic

acid. Following this, Mazzafera (1991) determined trigonelline in various species of

coffee by using another HPLC method (ODS-Hypersil; NaOAc 5%, pH 5) and found

higher values for Arabica, Robusta and Liberica coffee beans than all other preceding

investigators.

1.5 Roasting and other processes

The desirable taste and aroma of coffee beverage are mainly formed during

roasting. The pleasant flavour of roasted coffee depends on several factors, such as the

variety of coffee, the processing method, type of blend, shape and size of coffee bean,

origin, climate, location of coffee production, and, most importantly, the roasting

conditions, i.e., type of roaster used, temperature, time, colour and cooling, etc.

Roasting is a process dependent upon time and temperature. Green coffees are subject

to many chemical changes. Roasting of the green beans develops the aromatic coffee

oil which produces the unique characteristic flavour and aroma associated with coffee.

During this process, changes occur, such as loss of moisture content, gaseous carbon

dioxide and other volatile products. The time and temperature chosen for roasting of

coffee beans are responsible for obtaining a certain degree of roasting from very light

to very dark and this depends on consumer preferences. The ideal temperature and

time for a medium roast are 190-240 °C and 8-10 minutes, respectively. Roasting is

normally carried out under atmospheric conditions with hot combustion gas, or with

other heating agents.

Mechanical principles in roasting methods are shown in Fig. 1.4

Fig. 1.4 Mechanical principles in roasting methods.

Roasters

Horizontal rotating Vertical static drum Vertical rotating Fluidised Pressure bowl drum with blades bowl bed roasting

Perforated wall Solid wall

The chemical changes taking place in roasting for each of the individual

compounds in green coffee give rise to the characteristic flavour and aroma of roasted

coffee and its brews.

The typical average of the final chemical composition for medium/standard

roasted coffee, summarised by Clarke (1995), is reproduced in Table 1.4 :

Table 1. 4 : Roast coffee composition

Component Typical average percentage content (dry basis figure for a medium roast)

Arabica Robusta

Alkaloids (caffeine) 1.3 2.4Trigonelline(including roasted 1.0 0.7by- products)Minerals (oxide ash) 4.5 4.7Acids

residual chlorogenic acid 2.5 3.8quinic 0.8 1.0aliphatic 1.6 1.6

Sugarssucrose 0.0 0.0reducing 0.3 0.3

polysaccharides 33.0 37.0

Lignin 2.00 2.0Pectin 3.00 3.0Proteinaceous

protein 10.0 10.0free amino acids 0.00 0.0

Lipids (coffee oil) 17.0 11.0Caramelised/condensationproducts (melanoidins etc.)by difference 23.00 22.5Total 100.0 100.0

1.6 Indian coffee quality

Coffee quality is the summative index of many characteristics, such as

appearance in the raw green, roast and ground form, as well as liquor or beverage

18

qualities, comprising parameters, like aroma, taste, acidity and body. Indian coffee has

been maintained through strict quality control measures enforced at all levels by

planters, curers and traders. The quality control department of the Indian Coffee Board

plays a vital role in improving and maintaining the quality of Indian coffee. These

schemes include selecting and preparing coffee of uniform quality for export, making

payment to coffee growers to produce good quality coffee based on their visual and

cupping assessment, encouraging farmers to produce good quality coffees, screening

of coffee for export, quality assessment of research trials, and assessment of quality of

coffee grown in non-traditional areas. The quality control department carries out the

quality assessment with visual and organoleptic methods (Coffee & Cocoa

International, 1995)

1.6.1 Visual quality

The quality of coffee is evaluated as soon as coffee is cured and out turned.

Characteristics, such as moisture content, colour, size, evenness of shape, boldness,

density of beans, presence or absence of defects, unripe beans, stones and other

extraneous matter and odour of the beans, are evaluated by experienced, trained,

expert assessors. These are the important criteria for assessing raw or green coffee

quality and are known as visual assessment of coffee. These assessments have been

done against standard coffee samples prepared for a particular crop year or season

and known as ‘fair average quality’ (FAQ) samples. FAQ samples are prepared for

each of the principal types and grades of coffee, such as Plantation A (for washed

Arabicas), Arabica cherry AB for unwashed Arabica coffee, Robusta parchment AB

coffee for washed Robusta, and finally Robusta cherry AB grade for unwashed

Robusta coffee by senior assessors of the quality control division of the Coffee Board

of India. These samples should possess moisture contents within the prescribed limits,

a healthy colour and be very representative of the quality for the particular season.

This sample is treated as a standard for assessing visual quality of clean and

processed green/raw for premium payment for planters, for assessing samples for

domestic, export and all other sales. Coffee processed at the estate level from both

19

small and large growers is then further processed as per moisture, grading and garbling

standards laid down by the Indian Coffee Board.

1.6.2 Cupping evaluation

Coffee cupping is a method of systematically evaluating the aroma and taste

characteristics of a sample of coffee beans. Coffee tasting is an art by which the

intrinsic qualities of coffee, such as aroma, flavour, body, and acidity, are evaluated

based on a standardised procedure. It is a predominant and well known method of

evaluation throughout the world . The taster or tasters are always the same people who

have long experience in the art and have well acquired memory of various flavours.

The steps in evaluating the quality of a procedure are to examine the physical

appearance of the sample, the roast and grind, aroma of grind, and finally to taste the

brew for assessing overall quality.

A standard amount of clean green coffee is roasted in a sample roaster to a

standard or medium degree of roast. After roasting, the beans are allowed to cool and

ground to medium or coarse grind size. A measured amount of roast and ground

coffee is put into a porcelain or glass cup and a standard amount of freshly just boiling

water poured over the coffee. The brewing time is approximately six minutes. At the

end of brewing, the floats are removed and allow to cool to a palatable temperature.

The taster smells the brew, then sucks a sample from a tasting spoon into his mouth,

swirls it around, then spits it out and passes on to the next sample. During the tasting,

within the short time the brew is in the mouth, the strength (body), acidity (not sour)

and flavour are assessed. The combination of body, acidity and flavour are used to

rank samples. The overall cup quality is designated as excellent, fine, good, average,

falling off and poor.

1.7 Analytical methods:

The determination of acidity, pH, caffeine, amino acids, chlorogenic acids,

carbohydrates, oil content, proteins, trigonelline and volatile components are quite

20

important to define total beverage quality. Various analytical methods employed by

several researchers for different coffee components are summarised in Table 5 (Clarke

e ta l , 1985)

Table 1.5 : Analytical methods for determination of coffee composition.

Component_________________________________Analytical methods______________pH and titratable acidity glass electrode and pH meter, liquid chromatography,organic acids in coffee column chromatography, high performance liquid

chromatography, gas chromatography, enzymatic methods

caffeine Kjeldahl method: macro Bailey-Andrew method andmicro-Bailey-Andrew method, spectrophotometric method, HPLC methodpaper chromatography/ninhydrin combination of paper chromatography and electrophoresis thin layer chromatography, gas-liquid chromatography, HPLC methodpetroleum light extraction method (AOAC method) Kjeldahl method, gel filtration paper chromatography, HPLC method gas chromatography, column chromatography HPLC, gas chromatography-mass spectrometry method spectrophotometric method chromatographic methods: gas chromatography, liquid chromatography, HPLC

1.8 Aim and objectives

It is clear from the literature survey, that only few data have been reported on

the CGA and caffeine contents of Indian Arabica and Robusta coffees of commercial

grades. The purpose of this research is therefore to determine the chlorogenic acid and

caffeine contents of different processed and commercial grades of Indian Arabica and

Robusta green and roasted coffee by using sophisticated and precise analytical HPLC

methods to accomplish the following objectives.

1 To determine the chlorogenic acids and caffeine contents in most of the important

commercial varieties and main grades of dry-, wet- and monsoon- processed Indian

Arabica and Robusta green coffees.

21

ammo acids

carbohydrates

oil content proteins trigonelline volatile component

Chlorogenic acids

2 To determine the chlorogenic acid, and caffeine contents at various degrees of

roasting ( from the lightest to darkest roasts) of the Indian Arabica and Robusta

coffees and to evaluate the corresponding beverage quality.

3 To compare chlorogenic acid and caffeine contents of differently processed Indian

Arabica and Robusta coffees.

4 To carry out kinetic study on chlorogenic acids degradation in Indian Arabica and

Robusta coffees during the roasting process.

22

CHAPTER 2

MATERIALS AND METHODS

23

2.1 M aterials

The samples of wet-, dry- and monsoon- processed commercial green coffee

beans were kindly supplied by Consolidated Coffee Ltd, Pollibetta, Kodagu,

Karnataka, India, and were used for this entire research study unless otherwise stated.

The green coffee samples were :

2.1.1 Arabica coffee

2.1.1.1 Wet processed (washed) (Figure 2.1a )

Figure 2.1a: Washed Arabica A grade

Washed Arabica coffee -Plantation A : Plantation A grade is one of the grade

designations of Indian washed Arabica coffee. Its sieving requirements are 90 %

weight shall stand on a sieve with round holes of 6.65 mm. Not more than 1.5 % by

weight shall pass through a sieve with round holes of 6.0 mm. This grade shall be

clean garbled. It shall not contain PB (rounded coffee beans) subject to a tolerance of

2% by weight. It can contain 3 % triage by weight.

24

2.1.1.2 Dry processed (unwashed) Arabica coffee : Arabica cherry coffee A B

(Figure 2.1b)

Figure 2.1b Arabica cherry (unwashed ) AB grade coffee

It is one of the grade designations of Indian unwashed Arabica green coffee. Its

sieving requirements are that 90 % by weight shall stand on a sieve with round holes of

6.0 mm and not more than 1.5 % by weight shall pass through a sieve with round holes

of 5.50 mm. This grade shall be clean garbled and shall not contain PB beans subject to

a tolerance of 2% by weight. It can contain 3 % triage by weight.

25

2.1.1.3 Monsooned Malabar AA (Figure 2. lc)

Figure 2.1c Monsooned Arabica (Malabar) AA grade coffee

This grade refers to one of the grade designations of Monsooned Arabica

cherry coffee superior quality green coffee beans from unwashed or dry processed or

cherry coffee which have been subjected to moisture absorption during the monsoon

period (from June to September), as a result of which they swell and change colour to

golden/ light brown). The sieving requirement of this coffee is that 90 % shall stand on

a sieve with round holes of 7.25 mm and garbling standard is 2% triage by weight.

26

2.1.2 Robusta coffee :

2.1.2.1 Wet processed (washed) coffee:

2.1.2.1.1 Robusta parchment coffee PR (peaherry) (Fig. 2.2a)

Figure 2.2a Robusta (washed) parchment PB grade coffee

This grade shall be clean garbled. It shall not contain flat (AB) beans subject to

a tolerance of 2 % by weight, but it can contain 3 % of Pea berry triage by weight (Pea

berry is a coffee bean of nearly ovaloid form, resulting from the development o f a

single seed in the fruit).

27

2.1.2.1.2 Robusta parchment coffee AB (flat beans) (Figure 2.2b)

Figure 2.2b Robusta parchment (washed) AB grade coffee

It is one of the grade designations of Robusta Parchment (washed) coffee. Its

sieving requirements are that 90 % by weight shall stand on a sieve with round holes of

6.0 mm and not more than 1.5 % by weight shall pass through a sieve round holes of

5.50 mm. fh is grade shall be clean garbled. It shall not contain PB beans subject to a

tolerance of 2 % by weight and contain not more than 3 % triage by weight.

28

2.1.2.2 Dry; processed (unwashed) coffee : Robusta cherry coffee

Robusta cherry coffee is obtained by harvesting and drying coffee fruits. This

type is also called unwashed cotfee.

2.1.2.2.1 Robusta (unwashed) cherry coffee PB ( Figure 2.2c)

Figure 2.2c Robusta (unwashed) cherry PB grade coffee

The grading and garbling standards are the same as the Robusta parchment PB

coffee.

29

2.1.2.2.2 Robusta (unwashed) cherry coffee A B (flat beans) (Figure 2.2d)

Figure 2.2d Robusta (unwashed) cherry AB grade coffee

The grading and garbling standards of this coffee are similar to th©*.of Robusta

Parchment .AB coffee.

Moisture Standards : Moisture content of 10 -10.5 % for Indian washed Arabica

and Robusta coffees and 10.5 - 11.0 % for Indian unwashed Arabica and Robusta

coffees are prescribed.

30

2.1.2.3 Monsooned Robusta AA ( Figure 2.2e)

Figure 2.2e Monsooned Robusta AA grade coffee

It is one of grade designations of Indian Monsooned Robusta cherry coffee and

its sieving requirement is that 90 % shall stand on a sieve with round holes of 7.0 mm.

with a 2 % maximum permissible limit of triage. Prescribed moisture standard for

monsooned coffee is in the range of 13 -14 %.

31

2.1.3 Chemicals

5-CQA (caffeoylquinic acid), caffeine, caffeic acid and trifluoroacetic acid

(TFA), were obtained from Sigma Chemical Company Ltd, Poole, Dorset, UK.

Methanol and acetonitrile (ACN) of HPLC grade were obtained from Fisons Ltd,

Loughborough, UK.

Water/ aqueous means distilled water unless otherwise specified. All other

reagents were standard items from reputable commercial sources.

Carrez Reagent A was prepared by dissolving 21.9 g zinc acetate dihydrate in

water containing 3.0 g of glacial acetic acid and diluting to 100 ml with water.

Carrez Reagent B was prepared by dissolving 10.6 g of potassium ferrocyanide

trihydrate in 100 ml water. These reagents were stored at -4 °C (Egan et a l ., 1981).

Prior to grinding and extraction, all abnormal and defective beans were

removed from each sample by manual sorting. The immature (green coloured) flat or

‘AB’ grade beans were removed from dry-processed Robusta AB grade green coffee

sample, and these were analysed separately from fully mature (golden brown coloured)

beans.

2.2 Methods

2.2.1 Extraction

The green coffee bean samples were frozen, ground in a hammer mill to pass

0.7 mm and 500 mg extracted (4 x 25 ml, 25 minutes each) with 70 % ( v/v) boiling

aqueous methanol, using a Tecator HT-1043 (Tecator, Sweden) continuous extraction

system with the oil bath set to 140 °C. The bulked extracts were treated with Carrez

Reagents (1 ml A plus 1 ml B) to precipitate colloidal material, diluted to 100 ml with

32

aqueous 70 % methanol and filtered through a Whatman No. 1 filter paper. These

extracts were stored at -4°C and used directly for HPLC analysis. The roast and

ground coffee samples were extracted in an identical manner.

2.2.2 Analytical HPLC

A Spectra Physics P-4000 gradient pump, coupled to a Spectra physics AS-

3000 auto-sampler and a Spectra physics forward optical scanning detector were used

in conjunction with a 250 mm x 4.6 mm internal diameter column packed with

Spherisorb 5 ODS 2 (a fully end-capped 5 |Lim reverse-phase packing material, having a

very low level of residual silanol groups) prepared by Hichrom, Theale, Berkshire, UK.

Chromatographic and spectral data collection as well as integration were performed

using Spectra focus software on an IBM PS/2 computer.

A schematic diagram of the HPLC system used for this study is shown in Figure 2.3

Solvent A Solvent B

HPLC waste solvent

HPLC column Spectra Physicsscanningdetector Spectra focus

software + IBM PS/2 computer

Spectra Physics P- 40000 gradient pump + Auto sampler

Figure 2.3 : Schematic diagram of the HPLC

The chromatographic conditions, developed for analysis of standard 5-CQA,

caffeine and coffee extracts (in 70% v/v aqueous methanol), were as follows :

Injection volume : 20 pi

Mobile phase : Solvent A : 5 ml trifluo . ^cetic acid (TFA) per litre of water

Solvent B : 5 ml TFA added to 450 ml acetonitrile and 550 ml water

Gradient: A linear gradient from 100 % solvent A to 100 % Solvent B

over 56 minutes at flow rate 1 ml per minute

33

The above gradient provides good resolution of the peaks. Peaks were detected

at 280 nm for caffeine and 315 nm for chlorogenic acids. As required, UV spectra

were recorded between 200 and 360 nm. Peak identification was achieved for

chlorogenic acids where possible by spiking with authentic material, or by relative

retention time and the effect of treatment with tetramethylammonium hydroxide

(TMAH) (Clifford et al., 1989 ).

2.2.2.1 Standard calibration curve for standard 5-CQA

Concentrations of 0, 1, 5, 10, 20, 50, and 100 pg/ml of standard 5-CQA were

prepared in 70 % (v/v) aqueous methanol. Calibration curves were prepared using

duplicate injections of each concentration of standard 5-CQA and (20 pi) analysed as

described above. The peak was detected at 315 nm. The peak area and the retention

times in correspondence with the concentration were noted and was calculated. The

linear regression coefficient and a calibration curve drawn ( Figure 2.5a).

9.0E+6

8.0E+6 --

7.0E+6 --

6.0E+6 --

5.0E+6 --re<Di-

4.0E+6 --

3.0E+6 --

2.0E+6 --

1.0E+6 --

000.0E+0400 20 60 80 100

Concentation of 5-CQA(in ug/ml)

Figure 2.5a : Standard calibration curve for 5-CQA.

34

2.2.2.2 Calibration standard o f Caffeine

The following concentrations of standard caffeine were prepared in 70% (v/v)

aqueous methanol for construction of a calibration curve and calculation of linear

regression coefficient: 0, 15.6, 31.25, 62.5, 125 and 250. pg/ml.

The samples were injected (20pl in duplicate) as described earlier and the caffeine peak

has detected at 280 nm.

The calibration curve for caffeine is shown in Figure 2.5b.

20.0E+6

18.0E+6 --

16.0E+6 --

14.0E+6 --CNE 12.0E+6 -- _c (0 10.0E+6 --£mraa>

£L

8.0E+6 --

i.OE+6 --

4.0E+6 --

2.0E+6 --

000.0E+00 50 100 150 200 250

- Peak area

Concentation of Caffeine(ug/ml)

Figure 2.5b : Standard calibration curve for Caffeine.

2.2.3 Moisture content

Duplicate samples of each ground green and roast and ground coffee sample

were dried conventionally to constant weight at 105 °C in a vacuum oven for about 6

hours.

35

2.2.4 Roasting

2.2.4.1 Moda coffee bean roaster

Samples were roasted in a Moda F-134 laboratory sample electric roaster

(Moda, made in Hong Kong, Model F-134, 240 volts with metallic mesh with a lid

supplied by FCF Ltd, Poole, Dorset, England, UK). The Moda coffee bean sample

roaster was pre-set to a temperature of 190 - 200 °C for about 3 minutes and, in order

to preheat the roaster, approximately 50 g of green coffee were inserted and roasted

for about five minutes, the power was switched off and the beans were removed and

discarded. After this trial run, the samples of Arabica, Robusta, and Monsooned green

coffees of 50 g each in weight were roasted in the preheated electric Moda coffee bean

sample roaster. The roasting time for Arabica and Robusta coffee bean samples were

of 3, 4, 5, 6, 8, 10, 14 and 16 minutes. For Monsooned Arabica and Robusta coffees,

light, medium and dark roasts of 3, 5 and 6 minutes, respectively were used. When all

the samples had been roasted as specified, the power was switched off and the beans

were transferred onto white porcelain plates to cool down with vigorous stirring. After

cooling, the roasted beans were weighed immediately in order to calculate the roasting

loss.

2.2.4.2 Grinding o f roasted coffee sample

An electrical coffee bean sample grinder, namely Krups - 75, KM - 75, Type

203 (240 volts), was used for grinding roasted coffee samples throughout this study.

Each sample was ground for 10 seconds to produce medium-filter ground size. These

roast and grounded samples were used for domestic style aqueous brewing and

aqueous methanolic extraction.

2.2.5 Brewing

Freshly prepared roast and ground coffee was extracted in duplicate by

weighing approximately 7 g into a white porcelain bowl. Freshly boiled potable water

36

(150 ml) was added with stirring. The temperature of the mixture should be not less

than 90 °C at the time of pouring water over the coffee. The brew was allowed to cool

to about 60 °C or similar palatable temperature and the beverage quality evaluated.

Evaluation was completed before the brew became cold. The beverage was slurped

with a suction of air into the mouth (with accompanying noise). This slurping action

cools the coffee beverage, while allowing aroma and flavour to be thoroughly wafted

into the nostrils. The pH value of the aqueous brew was measured using pH meter

calibrated with appropriate standards. About 50 ml of each aqueous brew were

collected and stored in refrigeration at -4°C and used for HPLC analysis for

quantifying chlorogenic acid and caffeine contents after treatment with Carrez reagents

as described in Sub-section 2.1.3.

2.2.6 Evaluation o f coffee beverage

The. evaluation of coffee beverage was done by myself having, about nine years

of experience in quality control and beverage evaluation in the Directorate of Quality

Control of Coffee Board of India and being Chief Quality Controller and Cup Taster,

Head of Quality Control Department, Asian Coffee Limited ( 100% pure soluble coffee

manufacturing and 100% export oriented unit), India. In addition, I have undergone a

short term course in “Sensory Evaluation of Foods “ at Central Food Technological

Research Institute, India.

2.2.7 Storage.

All the samples and standards extracts, etc. were stored in glass containers in a

refrigerator at -4 °C.

This concludes the chapter on materials and methods utilised during the course

of research study. The next chapter concentrates on the results obtained and discusses

the analysis performed on green coffee.

37

CHAPTER 3

GREEN COFFEE ANALYSIS

RESULTS AND DISCUSSION

38

3.1 Results

3.1.1 Moisture content o f green coffee beans

The mean moisture content of duplicate determination of each green coffee

sample was determined. The values are as shown in Table 3.1

Table 3.1 : Moisture content of green coffee

Type and grade of coffee Moisture content

(n=2, % dry wt.)

Washed Arabica (Plantation) A 7.8

Unwashed Arabica (Arabica cherry) AB 6.5

Unwashed Robusta (Robusta cherry) AB 7.5

Unwashed Robusta (Robusta cherry) PB 7.6

Washed Robusta (Robusta parchment) AB 7.8

Washed Robusta (Robusta parchment) PB 9.0

Monsooned Arabica (Malabar) cherry AA 7.3

Monsooned Robusta cherry AA 7.4

These values of moisture content of green coffees are commensurate with those

observed in commercial practice (Clarke, 1985).

3.1.2 Identification o f peaks on chromatogram

The three CQA isomers (3-CQA, 4-CQA and 5-CQA), caffeic acid, 5-p-

CoQA, two FQA isomers (4-FQA and 5-FQA), three isomers of di-CQA (3,4-di-CQA,

3,5-di-CQA and 4,5-di-CQA) and caffeoyl tryptophan were located easily on the

chromatograms by a combination of spiking with authentic material and behaviour,39

when treated with tetramethylammonium hydroxide (TMAH), as reported previously

(Clifford et a l, 1989 b, d). Further evidence of identity and purity was obtained from

the calculation of relative retention times and spectral matching. Caffeic acid identity

also was confirmed by running a standard sample of caffeic acid under the same

gradient conditions. These data \ ± were consistent with those obtained previously

(Clifford et a l, 1989 b, d). It was not possible to locate 3-/?-CoQA or 4-/>-CoQA,

which would be very minor components, or 3-FQA, which would be expected to co

elute with 4-CQA, as reported previously (Trugo and Macrae, 1994; Cifford et al,

1989b, d). Confirmation of the behaviour of 3-FQA, which would co-elute with 4-

CQA, was obtained by analysing the mixture generated by treating with TMAH 5-

FQA, isolated from green coffee extract.

The mixed di-esters have previously (Clifford et a l, 1989 b) been partially

resolved into three peaks, containing two isomers each (3C,4-FQA + 3F,4-CQA; 3C,

5-FQA + 3F,5-CQA; and 4C,5-FQA + 4F,5-CQA). However, in the present case ,the

most hydrophobic pair co-eluted with caffeoyl tryptophan, where their presence on the

leading edge of that peak was detectable spectrally. Specimen chromatograms of

Indian Arabica and Robusta green coffee are shown in Figs. 3.1 & 3.2, respectively.

Only minute traces (<0.01 %) of/?-coumaric acid and ferulic acid were observed, and

these have not been included in the tabulated results.

40

Fig.

3.1

: Chr

omat

ogra

m

of a

70 %

met

hano

lic

extra

ct o

f an

Indi

an

wet

-pro

cess

ed

gree

n A

rabi

ca

coffe

e (P

lant

atio

n A

). C

hrom

atog

raph

ic

cond

ition

s: S

olve

nt

A: 0

.5 %

TFA

; So

lven

t B

: 0.5

% TF

A in

45%

aque

ous

acet

onitr

ile;

grad

ient

100

%

, A

to 10

0%

B lin

early

in

56 m

inut

es;

Colu

mn

25 cm

x 4.6

mm

pa

cked

wi

th Sp

heris

orb

OD

S2,

5 pm

; 20

pi in

ject

ion

volu

me,

det

ectio

n at

280

and

3 15

nm.

Key

peak

1=

3-C

QA

; 2=

4-C

QA

; 3=

5-C

QA

; pe

ak

4 =

caffe

ic

acid

5 =

caffe

ine

(at 2

80

nm)

mas

king

5-

/;-Co

QA

(at

3 15

nm);

peak

6-

4-

FQA

; pe

ak

7 =

5-FQ

A;

peak

s 8

and

9 =

Unk

now

n 1;

peak

10

= 3,

4-di

-CQ

A;

peak

1 I

= 3,

5-di

-CQ

A;

peak

12

= 4,

5-di

-CQ

A;

peak

13=

un

know

n 2

E Ec c8 in<N CO(TJ (U

8

soueqjosqv

Ret

entio

n tim

e (m

inut

es)

ino '

<a £ <!> P

VOor- x

x

£o

c

"O

ooo'JSC lcOu.b cot o£o

in(NC£

_ 3oo

c /o ’o3c

■p

-CU ^

54i

, cC O « 00

(N

t oc_o-4-4o <u-4—><u

~ a

c

o <u

vOPj3

COCUp .

<aP-4

i n

c0<uCL

m

to

<a0

U1CL

s oc

p 2C/0cOP

£coooCN

<Uc?scO

< J

xcOu.bO

 P - o n

H D‘ £ O

< a

. o

§ 4

PPcOCUCL

<au iinP-Tm+<aP-

UmII

<N

p*: cO (U CL

< a u

iXp-T

<N+< . a gu . J2

i ~ p

^ 3<-> II^ 4ii —

p*s ’ 1 cO

i o ^CU • -

bJ)< "SR &dU> c

4 ^X cO I CU

ir , —

T §

c£o

< 2 aJ 4 U

<5 ^CL ^

4-

C J <.1 aX PL

O

£ m _ ocO O

b b • •O CQ>4—'03 -*-•£ o

oc /o’ C Cl) r “

< f ’§- a 2

.JL ^-O +H

O f >4

11 % 00 o

•S 11O CO C l —

E £ c c

8 SCM CO

ro ro

as

£oo

CQ<5(A3

PSO

a :ca>a>i_O)

■oasszwTOScTO

,J5ccTO

TOJC■+Hto£

oh -TO

CTO05o

•+Hre£ow

SZocv|co

05iZ

CO

o

03

CO

CD

in

aoueqjosqv

Ret

entio

n tim

e (m

inut

es)

Table 3.2 : Retention time (minutes), relative retention times and Xmax for the individual chlorogenic acids and caffeine.

Compound analysed Retention time (min)

Relative retention times * ^ m a x

3-CQA 15.5 0.74 1.00 326

4-CQA 20.6 0.99

op

328

5-CQA 21.0 1.00

X>OO

328

Caffeic acid 22.6 1.08 325

5-/?-CoQA 24.7 1.18 313

3-FQA

4-FQA 25.1 1.20 0.96a 1.22 326

5-FQA 26.2 1.25 1.00a 1.25 327

Unknown 1 27.3 1.30 325

3,4-di-CQA 31.6 1.51 1.00b 1.00 327

3,5-di-CQA 32.7 1.56 1.03b 1.00 329

4,5-di-CQA 34.9 1.66 1 . 1 0 b 329

3C,4-FQA + 3F,4- CQA

36.0 1.71 1.00° 1.14° 328

3C,5-FQA +3F,5- CQA

37.5 1.80 1.04 1.15d 329

Caffeoyl tryptophan 39.3 1.88 222, 292, and 324

Unknown 2 43.1 2.05 222, 292 and 309

Caffeine 23.7 1.18 274* Relative retention times having the same superscript are directly comparable.

As a preliminary study, the commercial washed Robusta AB green coffee was

sampled extensively and ground material was extracted using each of the six sample

positions on the Tecator HT-1043 apparatus, in order to assess properly whether any

43

significant variation was introduced due to this operational factor. The variation

observed was no greater than that between replicate determinations associated with a

single position, and the nine sets of data so obtained were subsequently pooled and

shown in Table 3.3.

Table 3.3 The content of individual chlorogenic acids content( % dry basis) nine sets

of Indian Robusta ( j washed) AB grade green coffee.

Compoundanalysed

1 2 3 4Sam5

3le num6

ber7 8 9 Mean Sd.(±)

3-CQA4-CQA5-CQA

0.440.604.61

0.480.655.01

0420.524.31

0.430.544.30

0.410.514.48

0.420.584.36

0.450.604.46

0.470.614.52

0.460.564.39

0.440.574.49

0.030.060.22

CQASub-total 5.65 6.14 5.24 5.27 5.40 5.36 5.51 5.60 5.41 5.51 0.28Caffeic acid 0.07 0.07 0.11 0.05 0.05 0.05 0.05 0.10 0.05 0.07 0.025-p-CoQA 0.05 0.05 0.15 0.04 0.05 0.01 0.04 0.05 0.04 0.05 0.004-FQA5-FQA

0.090.77

0.130.92

0.131.01

0.100.92

0.090.92

0.080.80

0.080.91

0.100.91

0.090.88

0.090.89

0.010.07

FQA Subtotal

0.86 1.05 1.14 1.02 1.01 0.88 0.99 1.01 0.97 0.99 0.08

Unknown 1 0.09 0.07 0.10 0.05 0.06 0.05 0.04 0.06 0.05 0.06 0.023.4-di-CQA3.5-di-CQA4.5-di-CQA

0.520.690.52

0.540.800.49

0500.820.50

0.470.750.41

0.430.500.82

0.460.760.41

0.580.930.49

0.600.950.51

0.560.830.49

0.520.820.47

0.050.080.05

di-CQASub-total

1.73 1.83 1.82 1.63 1.75 1.63 2.00 2.06 1.88 1.81 0.15

3C,4-FQA+ 3F,4-CQA 3C,5-FQA+ 3F,5-CQA

0.12

0.11

0.13

0.13

0.15

0.16

0.12

0.14

0.13

0.12

0.11

0.11

0.13

0.14

0.13

0.18

0.14

0.13

0.13

0.14

0.01

0.02CFQA+FCQASub-total

0.23 0.26 0.31 0.26 0.25 0.22 0.27 0.32 0.27 0.27 0.03

Caffeoyltryptophan

0.54 0.57 0.55 0.56 0.57 0.51 0.60 0.70 0.61 0.58 0.06

Unknown 2 0.17 0.18 0.19 0.14 0.16 0.17 0.16 0.17 0.16 0.17 0.01CGA total 9.36 10.19 9.51 9.01 9.29 8.90 9.64 10.06 9.43 9.49 0.43

The contents of caffeine and individual chlorogenic acid (as CGA equivalents,

without correction for response factors) were calculated using the following

regression equations :

Caffeine (fig/ml) = -3.348 + 1.37 x 10'5 (peak area), r > 0.999 and

5-CQA (fig/ml )=1.281 + 1.10x 10 '5 (peak area), r > 0.999.

44

and b, respectively.

3.1.3 Individual chlorogenic acid and caffeine content in Indian Arabica green

coffee

The data so obtained are presented (% dry matter basis, mean ± standard

deviation) in Table 3.4.

Table 3.4 : Content of individual chlorogenic acids and caffeine (% dry basis) inIndian Arabica coffee

Compoundanalysed

Dry-processed

n=3 sd (±)

Wet-processed

n=3 sd (±)

Monsoon-processed

n=3 sd (±)3-CQA 0.37 0.01 0.34 0.01 0.63 0.004-CQA 0.50 0.01 0.42 0.01 0.71 0.015-CQA 4.23 0.01 3.64 0.01 4.19 0.01CQA Sub - total

5.10 0.01 4.40 0.03 5.54 0.02

Caffeicacid

0.08 0.02 0.07 0.01 0.19 0.01

5-p-CoQA 0.07 0.01 0.05 0.01 0.05 0.004-FQA 0.05 0.01 0.06 0.01 0.05 0.005-FQA 0.35 0.01 0.33 0.02 0.26 0.00FQA Subtotal

0.40 0.02 0.39 0.03 0.31 0.00

Unknown 1 0.04 0.02 0.04 0.02 0.02 0.013,4-di-CQA

0.18 0.01 0.15 0.02 0.24 0.01

3,5-di-CQA

0.37 0.02 0.41 0.05 0.50 0.00

4,5-di-CQA

0.27 0.04 0.21 0.03 0.34 0.00

di-CQASub-total

0.82 0.08 0.77 0.01 1.08 0.01

Unknown 2 0.07 0.02 0.08 0.02 0.07 0.00Total CGA 6.60 0.20 5.80 0.02 7.05 0.04Caffeine 1.50 0.10 1.27 0.06 1.52 0.05

45

3.1.4 The chlorogenic acid and caffeine content in Indian Robusta green coffees

Table3.5 : The chlorogenic acid and caffeine content (% dry mass basis) in Indian ____________ Robusta green coffee.__________________

Compoundanalysed

Wet-processed Dry-processed Monsoon-processed

‘AB’n=9

sd(±) ‘PB’n=3

sd(±) ‘AB’n=6

sd(±) ‘PB’n=3

sd(±) ‘AA’n=3

sd(±)

3-CQA 0.44 0.03 0.43 0.01 0.43 0.11 0.43 0.01 0.64 0.004-CQA 0.57 0.02 0.59 0.03 0.48 0.10 0.53 0.02 0.74 0.015-CQA 4.49 0.02 4.90 0.01 3.49 0.46 3.76 0.06 4.50 0.01CQA Subtotal

5.50 0.07 5.92 0.04 4.39 0.67 4.73 0.09 5.88 0.02

Caffeic acid 0.07 0.02 0.08 0.01 0.07 0.02 0.07 0.01 0.12 0.025-p-CoQA 0.05 0.00 0.05 0.01 0.05 0.01 0.05 0.03 0.02 0.00

4-FQA 0.09 0.01 0.09 0.01 0.11 0.01 0.10 0.01 0.09 0.005-FQA 0.89 0.07 0.85 0.05 0.83 0.05 0.83 0.01 0.69 0.00FQA Subtotal

0.98 0.08 0.94 0.05 0.94 0.05 0.93 0.01 0.78 0.00

Unknown 1 0.06 0.02 0.08 0.01 0.09 0.01 0.08 0.01 0.02 0.013,4-di-CQA 0.52 0.05 0.51 0.03 0.44 0.05 0.46 0.07' 0.71 0.013,5-di-CQA 0.81 0.08 0.77 0.06 0.61 0.08 0.67 0.13 0.71 0.014,5-di-CQA 0.47 0.05 0.52 0.08 0.47 0.04 0.52 0.03 0.84 0.01di-CQASub-total

1.80 0.18 1.80 0.17 1.52 0.17 1.65 0.23 2.26 0.01

3C,4-FQA + 3F,4- CQA

0.13 0.01 0.12 0.01 0.15 0.01 0.14 0.02 0.13 0.01

3C,5-FQA

+3F,5-CQA

0.14 0.02 0.12 0.01 0.13 0.01 0.12 0.01 0.08 0.00

CFQA +FCQASub-total

0.27 0.03 0.24 0.01 0.28 0.01 0.26 0.03 0.21 0.01

Caffeoyltryptophan

0.58 0.06 0.57 0.03 0.45 0.10 0.48 0.09 0.57 0.00

Unknown 0.17 0.01 0.17 0.01 0.14 0.02 0.15 0.02 0.12 0.00

CGA total 9.49 0.04 9.84 0.03 7.94 0.08 8.38 0.02 9.99 0.06

Caffeine 3.15 0.02 3.00 0.02 2.90 0.04 2.70 0.03 2.52 0.06

46

3.1.5 The chlorogenic acids and caffeine content in immature and mature dry-

processed Indian Robusta green coffee beans

Table 3.6 : The content of individual chlorogenic acids and of caffeine (% dry mass

basis) in immature and fully mature dry-processed Indian Robusta green coffee beans

Compound analysed Immature beans Mature beans‘AB’- grade ‘AB’- grade

n=3 sd(±) n=3 sd(±)3-CQA 0.45 0.04 0.43 0.044-CQA 0.60 0.03 0.56 0.025-CQA 4.51 0.51 3.89 0.30CQA Sub-total 5.56 0.58 4.88 0.33Caffeic acid 0.07 0.01 0.07 0.015-p-CoQA 0.07 0.01 0.06 0.004-FQA 0.08 0.01 0.11 0.005-FQA 0.66 0.02 0.81 0.04FQA Sub-total 0.74 0.01 0.92 0.04Unknown 1 0.07 0.00 0.08 0.013,4-di-CQA 0.62 0.04 0.52 0.033,5-di-CQA 0.80 0.11 0.65 0.094,5-di-CQA 0.61 0.01 0.58 0.02di-CQA Sub-total 2.23 0.16 1.75 0.143C,4-FQA + 3F,4-CQA 0.16 0.01 0.16 0.023C,5-FQA +3F,5-CQA 0.11 0.01 0.11 0.01CFQA + FCQA Sub-total 0.27 0.01 0.27 0.02Caffeoyl tryptophan 0.67 0.02 0.56 0.02Unknown 2 0.17 0.01 0.15 0.01CGA total 9.65 0.70 8.73 0.53Caffeine 2.60 0.02 2.40 0.08

3.2 Discussion

3.2.1 Moisture content o f green coffee

The mean moisture content (duplicate determination, % dry weight) of

commercial grades of wet-, dry- and monsoon-processed Indian Arabica and Robusta

green coffee are shown in Table 3.1. Several methods have been proposed for the

determination of water content (moisture) in green, roasted coffee beans (Clarke,

1985). Recently, some of the methods have been developed by ISO and developed

47

green coffee and Karl Fischer method for green, roasted and soluble coffee (Illy &

Viani, 1995). The values of moisture content of Indian green coffee obtained by loss in

mass at 105°C method are commensurate with trade only.

3.2.2 The chlorogenic acids and caffeine content o f Indian Arabica green coffee

The contents of individual chlorogenic acids content and caffeine of Indian dry-

, wet- and monsoon-processed Arabica green coffees were determined by HPLC, on -

% dry basis, and are shown in table 3.4. The wet-processed Arabica coffee showed

lower levels of total chlorogenic acids than dry and monsoon processed Arabica

coffee. The level of 5-CQA, total CQA, di-CQA, as well as caffeine in wet processed

Arabica green coffees were also lower than dry- and monsoon- processed Arabica

green coffees. The monsoon-processed Arabica coffee contained a much higher level

of caffeic acid than that of dry- and wet-processed Arabicas. In wet- and dry-

processed Arabica coffees, the level of 3,5-di-CQA was significantly higher than other

isomers of di-CQA.

Monsoon-processed Arabica coffee showed higher level of CQA and di-CQA

and thus total CGA, as compared with dry- and wet- processed Arabica coffees.

However, from the Table 3.4 it is clear that there was a lower content of FQA in

monsooned coffee than in wet- and dry- processed Arabica coffees. From Table 3.4 it

is also clear that there was more caffeic acid than in wet- and dry- processed Arabica

green coffee. This seems to be the most significant compositional difference

characterising monsoon- processed Arabica coffee.

Previous studies of Clifford and Kazi (1987) have shown that there is an

increase in the CQA : di-CQA ratio in the final 5-6 weeks of fruit maturation. De

Menezes (1994) reported that there was a statistically significant association between

the CQA : di-CQA ratio and maturity in samples of Coffea arabica cv vermelho

prepared by both dry- and wet-processing methods. It is generally accepted that the

quality of Indian Arabica coffee is rated as Monsoon- > Wet- > Dry- processed.

48

It is possible that the higher levels of CQA, di-CQA and thus total CGA, and

the greater content of caffeic acid and slightly lower level of FQA content in

monsooned Arabica coffee may be due in some way to the method of preparation and

contribute directly or indirectly to the unique quality of the beverage, in particular the

neutral mellow2 flavour.

There were no significant differences in the level of caffeine between dry-,

wet- and monsoon- processed Arabica coffees.

3.2.3 Chlorogenic acid and caffeine content in Indian Rohusta green coffee

The data obtained for wet-, dry- and monsoon- processed Indian Robusta

green coffee beans are shown in Table 3.5. The data indicate that the wet-processed

AB grade of Robusta coffee had a higher level of 5-CQA and 3,4-di-CQA (both p <

0.5) total CQA, 3,5-di-CQA and total CGA (p < 0.01) than dry-processed Robusta AB

green coffee beans. There were no significant differences in total FQA and caffeine

contents between these two samples of green coffee. As was seen for monsoon-

processed cherry coffee, there was a lower content of total FQA found in Monsooned

Robusta cherry coffees than wet- and dry- processed Robusta AB grade of coffee (p<

0.05). There were also significant differences in total CQA, caffeic acid, total di-CQA

and total CGA (p < 0.01) in monsooned Robusta AA grades of coffee compared with

cherry AB coffee. The monsooned Robusta cherry AA coffees had greater content of

3-CQA, and caffeic acid, (p > 0.01) and a lower level of total FQA (p < 0.05) than

Robusta wet- processed AB coffee. There were no significant differences in the

caffeine, total CFQA or caffeoyl tryptophan contents between these three processed

Indian Robusta green coffee samples.

2 Mellov/a balanced coffee whose basic organoleptic characteristics are just at the right level ,with none particularly apparent, giving an impression of roundness, and reflecting a harmonious balance in the strength/body not too acid, not too bitter but dense and rich flavour as defined by Lingle(1988)

49

- 3.2.4 Chlorogenic acid and caffeine contents in immature and fu lly matured dry

processed AB grade o f Indian Rohusta coffee

The quantitative data for chlorogenic acids and caffeine contents on % dry

basis, obtained by HPLC analysis, are shown in Table 3.6. The data indicate that the

immature (green coloured) dry- processed AB grade coffee beans, analysed as an

additional study, had a significantly lower content of 5-FQA (and hence total FQA)

and significantly higher 3,4-di-CQA (and hence total di-CQA), when compared with

the fully mature (golden brown coloured) beans from the same commercial grades of

Robusta cherry AB samples. The mature beans had a slightly lower (p < 0.05) caffeine

content than the immature coffee beans.

Clifford and Kazi (1987) previously reported on the progressive changes in the

CGA content of two clones of Ivory Coast Robusta as the fruits matured. Firstly, the

beans used in the 1987 study were obtained from fruit harvested a definite number of

weeks after flowering and thus each sample was fairly uniform in maturity, where as

the immature beans used in this study were obtained by manual sorting of a commercial

batch, and thus were of a less precisely defined stage of maturity. Secondly, there is

some evidence that seasonal and clonal factors influence the changes observed

(Clifford and Kazi, 1987). However, an increase in the CQA : di-CQA ratio seems to

be a general feature of the final five to six weeks of maturation (Clifford and Kazi,

1987). The corresponding data for the two samples compared in this study are : mean

ratios of 2.5 : 1 and X' 8 : 1, which suggests a similar trend, although the magnitude of

this change is not statistically significant.

3.3 Conclusion

In conclusion we found the chlorogenic acid and caffeine contents in various

processed Indian Arabica and Robusta green coffees varied from process to process.

The order of total CGA content for Robusta green coffee increased, starting with dry-

(approx. 7.6^d.b.), significantly increased for wet- (approx. 9.^d.b.) and concluded

with monsoon-processed coffee(approx. lft/d.b.). The variation found in the case of

Arabica green coffee was different, starting with wet- (approx. 5.8 d.b.), increasing for

50

dry- (approx. 6.6 d.b.) and concluded with monsoon-processed coffee (approx. 7.0

d.b.). There was also a small variation in the quantity of total CGA content between

the flat and rounded bean grades of the different processed coffees.

Confirmation was also obtained that fully matured Robusta dry-processed

Indian coffee beans (flat bean) had a lower CGA content than the immature variety.

The next chapter discusses the analysis and beverage evaluation of roasted

Indian coffee.

51

CHAPTER 4

ROASTED COFFEE

RESULTS AND DISCUSSION

52

The level of individual chlorogenic acids and of caffeine content of Indian

Arabica and Robusta coffees subjected to different degrees of roast were determined

by HPLC. The compounds were quantified in comparison with standard 5-CQA and

caffeine, respectively, without correction for individual response factors.

4.1 Roasted coffee analysis

4.1.1 Indian Arabica coffee

The quantitative data for individual chlorogenic acids, total chlorogenic acids

and caffeine of the samples of Indian Arabica coffee roasted to different degrees are

given in the following Tables.

4.1.1.1 Washed Indian Arabica A grade coffee

The chlorogenic acid content and caffeine content (% dry mass basis) for green

coffee (triplicate determination) and a series of roasts that cover from the lightest to

the darkest (almost burnt) including the sub commercial medium to standard roasting

times. The sequential roasting times were 3, 4, 5, 6, 10, 14 and 16 minutes. The data

(mean of duplicate determinations in percentage d. b., with standard deviation ( ±) for

roasted coffee) were given in the Table 4. la.

53

Tabl

e 4.

1a:

The

data

for c

hlor

ogen

ic ac

ids a

nd ca

ffeine

co

nten

ts (%

dry

basis

) for

gree

n co

ffee

of wa

shed

Ar

abica

A

grad

e (m

ean

of tri

plic

ate

dete

rmin

atio

ns)

and

the

mean

of

dupl

icate

deter

min

ation

for

roa

sted

coffe

e

\D

096 25.8

| | 900 18

.6 1

o o o’ o’

oo’ 0.0

6 | |

100 0.02

1900 0.0

6 |

0.15

|

00

6.36

840

16.8 00o

o' 9.80

0.33

0.33

0.87

1.53

900Ol'O 0.

090.

400.

48

600

600900800 0.

190.

022.

46 1.45

5.46

o

009 15.4

1.55

9.20

0.53

0.60 1.39

2.52

0.06

OI'O 0.10

0.50

0.59

900

no

80011*0

©©

o© 3.

69 1.43

5.14

00 480

14.0

1.75 o©

00

890190 1.

492.

780.

05 00oo' 0.

090.

400.

49

900 0.12

0.09

0.13

0.34 00o

o’ 3.87 1.43

5.00

VO 360

12.6

2.75

7.90

0.67

0.80 1.70

3.17 o

o't-oo' 0.

090.

360.

450.

050.

120.

090.

130.

35 l>oo' 4.

19 1.40

4.93

300 V

ll 3.66

7.60

0.78

0.93 1.94

3.65

0.03

0.28

0.08

0.28

0.35

£00 0.12

0.10

0.14

0.37

900 4.56 1.35

4.96

240

9.2

4.29

5.90

901080 2.

013.

87 ©o’ 0.

240.

070.

23 oCO© 0.

290.

130.

130.

150.

410.

055.

19 1.34

5.22

CO o 00 r- <

VL

6.22

5.80

|

901880 2.

434.

380.

03

£00 0.06

0.22

0.28

£00 0.15

0.13

0.21

0.48

900 5.28 1.29

5.44

o © o00t"'

ooo’ 0.

340.

443.

644.

410.

07 moo' 0.

060.

330.

39 ©o' 0.15

0.41

0.21

0.78 00o

o' 5.81 r-cs

| DEG

REE

OF

ROAS

T (M

IN.)

DEGR

EE

OF

RO

AST

ING

SE

CO

ND

S| R

OAST

ING

LOSS

%

(wt/w

t)| M

OIST

URE

CONT

ENT

%I %

DRY

MAS

S LO

SS3-

CQ

A4-

CQ

A| 5

-CQ

ATO

TAL

CQ

A|

CAFF

EIC

AC

IDI 5

-p-C

oQA

4-FQ

A5-

FQA

| TOT

AL

FQA

* 3.4-

di-C

QA

3.5-

di-C

QA

4.5-

di-C

QA

I TOT

AL

di-C

QA

**

| TO

TAL

CGA

| CA

FFEI

NE

| pH

OF

AQUE

OUS

BREW

Tj-

Roasting washed Arabica A grade coffee for progressively longer times has

provided interest in data on the fate of chlorogenic acids and caffeine and also on

roasting loss, dry mass loss and loss of moisture content. From Table 4.1a, the

moisture content ( %) of the green beans dropped from 7.8, to 0.1. Total CQA slowly

fell initially from 3 to 6 minutes but after 6 minutes, the degradation rate of total CQA

increased, degradation becoming very rapid after 10 minutes. However, initially there

were increases in 3-CQA and 4-CQA, seemingly due to the thermally induced

isomerisation of 5-CQA. this proportional increase was due to thermally induced

isomerisation of 5-CQA. There was an increase in the total FQA content from the 6 to

10 minutes roast falling sharply after 14 minutes. Roasts of 8 and 10 minutes are

categorised as medium and standard. At such degrees of roast, maximum flavour

development as well as good roasting quality were observed. During such roasting, the

total CQA and di-CQA contents were reduced to about 50 % of that of green coffee.

The caffeine content remained almost constant during the roasting process.

From the data reported above, it is clear that the individual chlorogenic acid in Indian

Arabica coffees degrade during roasting at different rates. It was reported that 5-CQA

is rapidly destroyed and 3-CQA produced in Arabica coffee, and di-CQA subgroup is

destroyed more rapidly than the CQA subgroup.

These data showed that CGA degrade during the roasting process but not all

CGA Sub-groups such as CQA, di-CQA and FQA behaved in an identical ■'

: * ■ " I manner, as seen in Fig .4.1 for Arabica washed coffee.

5-CQA decreases early in roasting ( 3-6 minutes). At this stage 5-CQA is more

sensitive to thermal treatment and partially converted in to 3-CQA and 4-CQA. (Fig

4.1). During this stage there is no destruction of 3-CQA and 4-CQA. Early in roasting

(3-6 minutes) 5-CQA degrades progressively. In this period, 3-CQA and 4-CQA

increase, although total CQA declines. It has previously been suggested that 5-CQA is

isomerised to 3-CQA and 4-CQA and this is consistent with the data reported here.

55

—♦ — 3-CQA

—■— 4-CQA

5-CQA

—X— TOTAL CQA

Fig. 4.1a :CQA content vs Degree of Roast of Indian Washed Arabica -A GradeCoffee

-d 3.5T3£ 31 25 C8 2

$ 1.5 o1

400 600 800

Degree of Roast (in seconds)

1000 1200

Fig. 4.1 b : FQA content vs Degree of Roast in Indian Washed Arabica -A GradeCoffee

—* — 4-FQA

5-FQA

TOTAL FQA

0 200 400 600 800 1000 1200


— 3,4-Di-CQA

— 3,5-Di-CQA

4,5-Di-CQA

—X— T OT ALDi-CQA

Fig. 4.1c : Di-CQA content vs Degree of Roast in Indian Washed Arabica AGrade Coffee

200 400 600 800 1000 1200

Dearee of Roast fin seconds)

—# — TOTAL CQA

—■ — TOTAL FQA

TOTAL Di-CQA

—X— TOTAL CGA

—* — CAFFEINE

< 7TO -Q O -d

Fig-

7.000

6.000

5.000

4.000

3.000

2.000

1.000

0.000

4.1 d : Chlorogenic acid and caffeine contents vs Degree of Roast in Indian Washed Arabica A Grade Coffee

200 400 600 800 1000 1200


A similar effect is seen among the di-CQA, but here 3,5-di-CQA degrades

more rapidly than the other isomers namely 3,4-di-CQA and 4,5-di-CQA, suggesting

inter-conversion. Total di-CQA declines, and, if there is some loss by hydrolysis, the

mono-ester generated may also at later stage of roasting, ranges from 8 to 10 minutes

duration. Degradation rate of all isomers of CGA is slower in the early stage of

roasting at extreme degree of roasting, but all CGA sub-groups are degrading very

rapidly , as chemical reaction taking place endothermically.

One might expect FQA to behave similarly to CQA, although possibly with a

difference in rate. At first, the total FQA content appears to be increasing. Since

Arabica coffees do not contain CFQA, production of FQA by partial hydrolysis of

CFQA is not an option. There was little likelihood of FQA being synthesised de novo

during roasting. A possible explanation would be the formation of a co-eluting

contaminant, but spectral analysis of the peaks associated to 4-FQA and 5-FQA

revealed typical CGA spectra. Therefore any contaminant must have similar

characteristics.

57

4.1.1.2 Unwashed Arabica AB grade coffee

The data compiled for CGA and caffeine content (triplicate determination for

green coffee analysis) and duplicate determination for roasted unwashed Arabica AB

grade coffee are shown in Table 4. lb

Table 4.1b : The data for chlorogenic acids and caffeine contents (% dry basis) for green coffee of unwashed Arabica AB grade (mean of triplicate determinations) and

the mean o f duplicate determination for roasted coffee

DEGREE OF ROAST (IN MIN..) 0 3 4 5 6 8 10ROASTING LOSS % (wt/wt) 0.00 9.60f 12.60 15.00 15.40 20.40 22.00MOISTURE CONTENT % 6.50 5.44 3.42 2.23 1.71 1.59 1.09% DRY MASS LOSS 0.00.. 8.50l 9.50 10.00 10.75 15.50 16.603-CQA 0.40 0.96 0.97 0.73 0.62 0.08 0.054-CQA 0.50 1.20 1.18 0.90 0.77 0.09 0.055-CQA 4.23 2.93 2.39 1.80 1.53 0.22 0.12TOTAL CQA 5.12 5.10 4.54 3.43 2.91 0.39 0.21CAFFEIC ACID 0.08 0.03 0.04 0.06 0.03 0.015-p-Co QA 0.07 0.07 0.08 0.10 0.07 0.044-FQA 0.05 0.09 0.10 0.10 0.12 0.07 0.055-FQA 0.35 0.28 0.36 0.38 0.45 0.15 0.09TOTAL FQA 0.40 0.36 0.45 0.48 0.57 0.22 0.14* 0.04 0.04 0.05 0.08 0.07 0.043,4-di-CQA 0.18 0.17 0.15 0.14 0.12 0.05 0.033,5-di-CQA 0.37 0.15 0.14 0.13 0.12 0.03 0.034,5-di-CQA 0.27 0.24 0.16 0.14 0.12 0.05 0.03TOTAL di-CQA 0.82 0.56 0.45 0.40 0.36 0.12 0.10** 0.07 0.04 0.05 0.03 0.02 0.01TOTAL CGA 6.60 6.21 5.65 4.57 4.04 0.84 0.44CAFFEINE 1.50 1.62 1.69 1.80 1.78 1.68 1.56pH OF AQUEOUS BREW - 5.20 5.00 4.96 5.00 5.92 5.96

There was an apparent, slow i increase in caffeine content with increasing

mass loss during roasting, reflecting caffeine’s relative stability to heat. In contrast the

total CGA content deteriorated progressively.

This was true also for 5-CQA, total CQA and total di-CQA. However, there

were short-term increase in 3-CQA and 4-CQA during the early stages of roasting,

reflecting the isomerisation of 5-CQA and possibly the hydrolysis of di-CQA, while the

beans contained sufficient water.

58

As 4-CQA, 4-FQA also increased between 3 to 6 minutes, but, in contrast to 5-

CQA, 5-FQA showed apparent ' increase between 5 and 6 minutes. Since there

is no obvious source for additional 5-FQA, it is suggested that this is due to an artefact

co-eluting with 5-FQA.

As this particular grade AB coffee of unwashed Arabica contained a mixture of

mature and immature beans, the roasting times of 14 and 16 minutes were deleted and

only six roasts were performed. The rates of degradation of CQA content at the initial

roasts, 3 and 4 minutes, and rapid decrease in total CQA content appeared from 5 to 6

minutes with drastic degradation of CQA content appearing in 8 and 10 minutes. There

was a proportional increase of FQA content at 5 and 6 minutes roasts. There were

drastic degradation of CQA, FQA and di-CQA contents, and hence total CGA content

from 8 and 10 minutes were observed as these roasts are considered extreme in this

particular study. There was also a slight increase in caffeine content during the

roasting process.

4.1.1.3 Monsooned Arabica -AA grade coffee

In this special study, only three degrees of roasting were performed, 3 5 and

6 minutes, as light, medium and dark roast, respectively. The data obtained for

chlorogenic acid and caffeine contents (triplicate determination in the case of green

coffee) and duplicate determinations for roasted Monsooned Arabica AA grade coffee

are given in the Table 4.1c.

59

Table 4.1c : The data for chlorogenic acids and caffeine contents (% dry basis) for green coffee of Monsooned Arabica AA grade (mean of triplicate determinations) and

the mean of duplicate determination for roasted coffee

DEGREE OF ROAST 0 Light Medium DarkDEGREE OF ROAST ( Min.) 0 3 5 6ROASTING LOSS % (wt/wt) 0 12 16 22MOISTURE CONTENT % 7.27 3.54 1.82 1.383-CQA 0.63 0.82 0.43 0.164-CQA 0.71 0.94 0.39 0.185-CQA 4.19 2.00 0.70 0.33TOTAL CQA 5.54 3.76 1.53 0.67CAFFEIC ACID 0.19 0.05 0.06 0.025-p-CoQA 0.05 0.06 0.06 0.054-FQA 0.05 0.08 0.10 0.075-FQA 0.26 0.43 0.47 0.26TOTAL FQA 0.31 0.51 0.56 0.32* 0.02 0.05 0.04 0.033,4-di-CQA 0.24 0.16 0.08 0.053,5-di-CQA 0.30 0.14 0.10 0.034,5-di-CQA 0.34 0.20 0.05 0.02TOTAL di-CQA 0.87 0.51 0.22 0.10** 0.07 0.07 0.04 0.00TOTAL CGA 7.05 5.01 2.51 1.19CAFFEINE 1.52 1.59 1.64 1.55pH OF AQUEOUS BREW 5.11 5.45 5.76

As described in chapter 2, this monsoon- processed Indian Arabica cherry AB

grade is a fully matured coffee (whole crop cherry) and A grade in raw and cupping

quality. Monsoon-processed coffee is a speciality exclusive to India Monsooned coffee

is characterised by having a bean size one and half times or twice that of dry- and wet-

processed green coffee and it is lighter pale white in colour. The data from the Table

show that roasting loss and loss of moisture content during the roasting process are

very rapid, and percentage of roasting loss at 3, 5 and 6 minutes roasting were 12, 16

and 22, compared with, say, unwashed Arabica AB of 8.5,10 and 10.75 %,

respectively. Sub-totals of CQA, di-CQA and caffeic acid contents decreased during

the roasting process, but total FQA was higher at light and medium roast. Total

CGA content decreased during the roasting process, but caffeine content remained

virtually constant. The rate of degradation of CQA, di-CQA and total CGA content

were almost similar. Very rapid disappearance of caffeic acid was observed in this

60

roasting study. The pre-existing caffeic acid is lost rapidly. Presumably caffeic acid is

released by hydrolysis of di-CQA and CQA hydrolysis but does not accumulate.

4.1.2 Indian Robusta coffees

Quantitative data in the form of means of duplicate determination ( % d. b.)

for chlorogenic acids and caffeine for the different degrees of roasted commercial

grades of wet- and dry- processed Indian Robusta coffees are presented. The degrees

of roast were 3, 4, 5, 6, 8, 10,14 and 16 minutes respectively for Robusta unwashed

and washed AB and PB grades. However, in the case of speciality coffees, i.e.,

Monsooned Robusta AA grade coffee, only three degrees of roasting were carried out,

3 minutes for light, 5 minutes for medium to standard and 6 minutes for dark roast.

They were analysed for CGA and caffeine content ( % d. b., in duplicate).

4.1.2.1 Unwashed Robusta AB grade coffee

The data obtained for chlorogenic acids and caffeine contents (% dry basis) for

green coffee of unwashed Robusta AB grade (mean of triplicate determinations) and

the mean of duplicate determination for roasted coffee are given in Table 4.2a.

61

Table 4.2a : The data for chlorogenic acids and caffeine contents (% dry basis) for green coffee of unwashed Robusta AB grade (mean of triplicate determinations) and

the mean of duplicate determination for roasted coffee

DEGREE OF ROAST(MIN.) 0 3 4 5 6 8 10 14 16IN SECONDS 180 240 300 360 480 600 840 960ROASTING LOSS %(wt/wt) 7.2 8.8 9.6 12.8 15.2 16.4 16.8 18MOISTURE CONTENT % 7.50 5.08 4.06 3.16 1.83 1.70 1.69 1.60 1.44% DRY MASS LOSS 0.00 4.80 5.30 7.11 7.11 8.41 8.90 9.30 11.903-CQA4-CQA5-CQA

0.350.473.29

0.751.012.36

0.761.032.21

0.821.141.94

0.590.831.56

0.240.360.65

0.170.320.51

0.120.160.38

0.080.080.17

TOTAL CQA 4.11 4.12 4.00 3.90 2.99 1.24 1.00 0.66 0.34CAFFEIC ACID 0.06 0.04 0.03 0.03 0.05 0.06 0.04 0.04 0.035-p-Co QA 0.04 0.04 0.04 0.05 0.10 0.10 0.08 0.07 0.054-FQA 0.10 0.17 0.20 0.22 0.22 0.16 0.17 0.14 0.095-FQA 0.85 0.52 0.55 0.59 0.65 0.37 0.29 0.27 0.17TOTAL FQA 0.95 0.69 0.76 0.81 0.87 0.53 0.46 0.41 0.26* 0.09 0.04 0.04 0.03 0.06 0.07 0.06 0.06 0.04CAFFEOYL TYROSINEANGOLA II3,4-di-CQA 0.41 0.34 0.32 0.28 0.23 0.17 0.07 0.07 0.053,5-di-CQA 0.60 0.24 0.24 0.21 0.23 0.12 0.08 0.05 0.034,5-di-CQA 0.45 0.41 0.39 0.33 0.24 0.10 0.09 0.04 0.03TOTAL di-CQA 1.45 0.99 0.95 0.81 0.70 0.39 0.24 0.15 0.103C,4-FQA + 3F,5-CQA 0.14 0.13 0.11 0.10 0.09 0.08 0.07 0.05 0.023C, 5-FQA + 3F,5-CQA 0.12 0.06 0.05 0.03 0.23 0.19 0.03 0.02 0.02TOTAL CFQA + FCQA 0.26 0.19 0.16 0.13 0.32 0.27 0.10 0.08 0.04CAFFEOYL TRYPTOPHAN 0.45 0.44 0.41 0.38 0.37 0.34 0.18 0.16 0.12** 0.14 0.09 0.09 0.11 0.11 0.11 0.06 0.06 0.05TOTAL CGA 7.55 6.63 6.48 6.24 5.55 2.94 2.23 1.69 1.03CAFFEINE 2.90 2.94 2.97 2.99 3.10 3.13 3.40 3.43 3.45pH OF AQUEOUS BREW 5.80 5.60 5.34 5.22 5.57 5.70 5.91 6.10

The total CGA content steadily decreased during the early roasting time

between 3 to 6 minutes. It is then steadily reduced at 8 to 10 minutes and then

drastically reduced at 14 to 16 minutes. A steady decrease in caffeoyl tryptophan and

total FQA content during early stage of roasting from 3 to 8 minute and a rapid

decrease in quantity after 10 minutes up to 16 minutes. There were rapid increase in

3-CQA and 4-CQA content at the initial stages of roasting at 3 to 5 minutes and at 6

and 8 minutes a steady increase was observed, corresponding to rapid decreases in 5-

CQA and diCQA, as these isomers are sensitive to heat treatment. However, after 10

t j iu

minutes roasting a rapid disappearanceysub-totals of CQA, FQA and di-CQA, CFQA +

FCQA and hence the total CGA content occurred. A slight increase in caffeine content

was also observed during the roasting process from about 2.7 % to 3.1 %. From above

data, the thermal instability of the 5-CQA, FQA, di-CQA and CFQA + FCQA

during the roasting is evident. From the above observation, over a series of roasting

levels from lightest to darkest, about 11% to 13 % roasting loss occurred,

corresponding to light to medium and standard roast levels and the pH of aqueous

brews were optimum 5.2 to 5.3, and providing well balanced liquor quality.

4.1.2.2 : Unwashed Robusta PB grade coffee

The quantitative data obtained for the chlorogenic acid and caffeine content of

green coffee of unwashed Robusta PB grade (triplicate determination) and roasted

coffees (duplicate determination) are given in Table 4.2b.

From the data shown in Table 4.2b, it can be seen that rounded or PB beans

naturally provides even roasts and heat transfers to the bean uniformly as this roll

readily while roasting. As compared to the flat beans, the quantitative data as stated

above reveals that the rapid effect of heat on chlorogenic acids and degradation rates

of total CQA, di-CQA and hence total CGA content faster than that of AB grades of

unwashed Robusta. Caffeoyl tryptophan content decreases steadily up to 8 minutes

roasting then degrading sharply, the total FQA content was relatively stable up to 8

minutes roasting time, but then FQA decreased more rapidly. At 16 minutes CQA and

di-CQA, retained only about 5-10 % of the original contents of the green coffee, and

about 15 % of total CGA.

63

Tabl

e 4.2

b : T

he

data

for c

hlor

ogen

ic ac

ids a

nd ca

ffeine

co

nten

ts (%

dry

basis

) for

gree

n co

ffee

of un

wash

ed

Robu

sta

PB gr

ade

(mea

n of

tripl

icate

dete

rmin

atio

ns)

and

the

mean

of

dupl

icate

deter

min

ation

for

roa

sted

coffe

e___

____

____

____

____

o'OON

00 O 00l-H i-Ho ’ o ’ o ’

ON ON © 1-Ho ’ o ’

co co o o o ’ o ’

oo

o00

NO 00 '3-1—I <—I COo ’ o ’ o ’

CS Oi— I C O

O NO V )i-H O ©o ’ o ’ o ’

'3- COo o 00

ooNO

oCO

*—i cs csCO CO NOo ’ o ’ o ’

ot Tt1-H -O"o o ’

CO ON I-H

o ’ o ’ o ’Tt Tl-o o

<N

o00Tt'n o n in r- o

I—I NOcs in o ’ o '

NO ON CO CS I-H CSo ’ o ’ o ’

on <n o o o ' o '

<o©•o'

cscs’

t " l-H ©NO 00 NOo ’ o ' hH

cs oCS NO

NO ON h CS i—i c so ’ o ’ o '

on >n o o >

>n

oC S

ooooocs

NO i-Ht" oO >—i

CO cs CS NOo ' o '

00 00 00 cs i-H cs o <nhH O <nI-H cs

•O"NO

oTtCS

rt i—i Tt t" O © CS CO CS NO

00 h oCS I—I COo ' o ’ o ’

00 rt O O rtO

'O*>-l

o00 om i n n o oot"- O hHo’ I—i CS

cs ^tCS NOo ’ o '

00 00 I—I C S I-H CO

o ' o ' o ’ON T to o o ’ o ’

oNO

CO CO NO in t-o ’ o ’ CO

NO t cs oc no i no ’ o ’ o ’

r t CS

o ’ o 'O n<o00

tZ3<§PhOm%owQ

o'er*inOJO£i—iHm<S

< <3 <a c / a0 u u1 I Icn m

< < o oPh Ph rt in

c c c0 0 0 u o u1 I ITt T3 "O■ i iTt in inCO*' cn r t

5XPLhoH §mmXO

oXa

4.1.2.3 Washed Indian Robusta AB grade coffee

The contents of individual chlorogenic acids and caffeine, obtained for green

coffee duplicate determination and triplicate determination roasted coffees, are

presented in Table 4.2c on the next page.

The data for the CGA and caffeine content of washed Robusta AB grade coffee

indicate that the rate of degradation of total CGA content was slow from 3 to 4

minutes roasting, more rapid thereafter ' up to 16 minutes. As

expected from the previous experiments, the FQA content was almost stable to heat

from 3 to 14 minutes roasting. Degradation of total CGA content is slower during the

roasting of washed Robustas than unwashed Robustas. This slower degradation may

be due to the washing effect and the maturity of the coffee beans. As expected, there

were no changes of caffeine content during the roasting process, even at extreme roast

the roasting loss was 14.4 % when only about 30 % CGA had been retained.

Quantitative data for individual chlorogenic acid in green and roasted coffee

shows that CGA degrades during the roasting process but not all CGA subgroups such

as CQA, di-CQA and FQA behave in an identical manner as shown in Fig 4.2 for

Robusta coffee.

65

Tabl

e 4.2

c : T

he

data

for c

hlor

ogen

ic ac

ids a

nd ca

ffeine

co

nten

ts (%

dry

basis

) for

gree

n co

ffee

of wa

shed

Ro

busta

AB

gr

ade

(mea

n of

tripl

icat

e

dete

rmin

atio

ns)

and

the

mean

of

dupl

icate

deter

min

ation

for

roa

sted

coffe

e __

____

_

so oso tt CSSO i-HCS <Nco OnCO Oso ©vs Tfo 00o 00 00Tf soso Tfo SOo o CS HH COCO cso © Tfo t-hH r-© 00Os CSTf r-SOos hH o r-~ o O o .-H o o o o o o o o o o o o © o o © cs! CO vs

Tf oTf 00 COso oso ON oo oOS soo o 00 vsvs cst" vso r-o cs vs co ©Tf Tfo o vso hHcs r-o 5 00 COV00 hH o so o o i-H o o o o o o o o o o o o o o o © CO co V

O oo © i-H oCO OsCS rHV soo £ CS 00SO OS00 o Tfvs cscs SO cocs 1-Hso r»o Tf© H1-H TfCO OS© t\CO cs oTfso i-H so o o <N o o o o o o © o o o o © © o o © "o CO V

00 ©00 © soVS e00 <NO 00 CS soo vsco Os ot- OS00 CCSo 00o r-CS t" cs cst" 00© Tfo cshH 00CO © oiH cscsTf i-H vs o o 1—1co o © o o o o © o o o © © © o o © Vo* co V

so oso 00to CSH ©1-H r00 00ON <No Tf00 Tfo f-o OS ol> Os00 coo CSco cs <Nco vs00 o vso VtH oTf cs CO 00o sohHco CS V o o cs co o o o o o o o o o o o © o o © <5 CO V

VS oo © Oos oCS coo oCS oTf coso Tfo Tfo oCS or- oOS coo vsTf co 00Tf Tfcs Tf so© ocs COCO cs •o Tfo ecsco us CS Tf CS Tf o o o o o o o o o 1-H © © o o © K CO vs

Tf oTf cs SOt" o1-H cs (Scs Os Tfco co© Tfo f- vsSO cs00 Tfo f"SO Tf csvs oso Tf SO© ocs TfV cs o© Vcs Tf TT CN vs o o o o o o o o o v-1 © © o o © «o co vs

co e00 00 CS OhH rcs SO CSo CSTf Tfo COo f- so 00t" vso cs so■O" COso 00 vs r-~o COcs V V Vo asOS t"Tfhh vs co 1—1co vs o o o o © o o o o hH © o o o © Os cs vs

o © o o00 oo r Osvs TfSO 00so 00o uoo o oOs oo 00o csvs SOC" ovs OS co co t"-cs VV oo Vo osost" o o o Tf IT) o © o o hH o o o o o o o o © OS cs

/■—v1 £ £

Hcn5«toOwto0too

Vi

OUto5

COCOoto05gOto M

OIS

TUR

E C

ON

TE

NT

% DR

Y M

ASS

LO

SS3-

CQ

A4-

CQ

A5-

CQ

ATO

TAL

CQ

AI C

AFFE

IC

AC

ID

§0o1*vs 14

-FQ

A|5

-FQ

ATO

TAL

FQA

* ICAFF

EOY

L TY

RO

SIN

E

1=158

3.4-

di-C

QA

3.5-

di-C

QA

4.5-

di-C

QA

ItOT

AL

di-C

QA

c*ovstoCO+

&pH1o'CO

§O■vstoco+

sto■vsUCO TO

TAL

CFQA

+

FCQ

A too$eto><otototo<u ** TO

TAL

CGA

1 CA

FFE

INE

siVi

§to<toowto

—♦ —3-CQA - ■ —4-CQA

5-CQA —X — TOTAL CQA

Fig. 4.2a : CQA content vs Degree of Roastin indian Washed RobustaAB Grade Coffee

6.000

5.000

4.000■o

3.000oo§ 2.000 o

1.000

200 300 400 500 600 700Degree of Roast 9in seconds)

0.0000

i t t i i s

■

SSifflS____ .

—♦ — 4-FQA—■ —5-FQA

TOTAL FQA

1.200

Fig.4.2b :FQA content vs Degree of Roast in Indian Washed RobustaAB Grade Coffee

1.000.Q■o 0.800

1 0.600co< 0.400

0.200

200 300 400 500 600 700


0.0000

Fig.4.2c Di-CQA content vs Degree of Roast in indian Washed RobustaAB Grade Coffee

2.0001.800

2 1.6002 1.400 0s■£ 1.200 0)•£ 1.000 8 0.800 g 0.600 5 0.400

0.200 0.000

0 200 400 600 800 1000 1200Degree of Roast (in seconds)

3.4-Di-CQA3.5-Di-CQA4.5-Di-CQA TOTAL Di-CQA

—♦— 3C,4-FQA +3F.5-CQA —■— 3C,5-FQA + 3F.5-CQA

TOTAL CFQA + FCQA —X— CAFFEOYL TRYPTOPHAN

Fig. 4.2d : CFQA + FCQA and Caffeoyl tryptophan vs Degree of Roast in Indian washed Robusta AB grade Coffee

> * 2 0.50<12 T3 3- soS 5r o.4o

T J Cc <v re ■£§ 8 0 :30O c u. re+ •§. 0.20 S IO £ 0.10

0 200 400 600 800 1000 1200

Degree of Roast( in seconds)

Fig. 4.2e : CGA Sub-group and caffeine contents vs Degree of Roast in Indian Washed Robusta AB Grade Coffee

12.00

10.00

■TOTALCQA -TOTALFQA TOTAL Di-CQA TOTAL CFQA + FCQA

-CAFFEOYL TRYPTOPHAN -TOTALCGA CAFFEINE

200 400 600 800

Degree of Roast (in seconds)1200

4.1.2.4 Washed Robusta PB grade coffee

The levehof individual chlorogenic acidr-and caffeine of washed Robusta PB

grade green coffee (triplicate determination) and of coffees with different degrees of

roast (duplicate determination) are given in Table 4.2d orynext page.

68

Tabl

e 4.2

d : T

he

data

for c

hlor

ogen

ic ac

ids a

nd ca

ffeine

co

nten

ts (%

dry

basis

) for

gree

n co

ffee

of wa

shed

Ro

busta

PB

grad

e (m

ean

of tri

plic

ate

dete

rmin

atio

ns)

and

the

mean

of

dupl

icate

deter

min

ation

for

roa

sted

coffe

e

NO

| 096

| I 15

.2 | |

oro | I

6.23

||

0.20

||

0.24

||

0.47

| 1 160

| | 0.0

5 |

| 0.0

9 |

0.14

0.42

| 0.5

6 |

| 0.0

33

|

| 0.0

6 |

0.11

0.09

0.03

0.23

|

100ZO'O

rooo 0.1

1 |

0.06

|2.1

2 |

3.47

|5.8

6 |

Tft-H

| 84

0|

14.6

1 1-

24|

6.23

| 0.

25|

0.29 oNO

o ’1

1.14

|

Tfoo

ooo ’ 0.

160.

49|

0.65

|

oo

1 0.1

2 |

0.12

0.10

0.07

| 0.2

9 |

0.07

0.03

| oro 0.1

7 |

0.06

|2.6

7 |

3.45

|5.

75

©l-H

009 | I

13.4

| 1.

66|

6.20

| 0.

45|

0.53

1 1.

09I

2.07

| 0.

05 oo ’ 0.

160.

63I

0.79

|

oo'

1 zvo

|

0.16

0.12

0.14

0.42

|0.

080.

030.1

1 |

0.25

|0.0

9 | |

OO

P 3.40

|5.

46

00

| 48

0I

12.4

| 1.

70|

5.20

1 0.

55

990 | 1

1-43

I 2.

64|

0.07

|

oooo 0.

180.

64|

0.82

|r"oo ’

1 no

|

0.22

0.15

0.23

0.59

|0.

110.

040.1

5 |

0.32

| | 600 4.9

4 |

3.35

|5.

30

NO

| 36

0I

11.4

981 | 1

4.70

| 0.

67|

0.79

| 1.6

8 |

1 3.1

4 |

r-oo

| 0.0

6 1

0.19

0.66

I 0.8

5 |

I 0.0

6 1 1

iro I

0.27

0.22

0.29 00

o’ 0.11

0.06

0.17

|0.3

9 |

0.12

|5.7

4 |

3.30

|5.

11

in

| 30

0

| 2.

37I

4.50

1 0.

74 00o

o00

I 3.3

9 |

00oo ’

NOoo ’ 0.

190.

70 as00©

| 0.0

6 | 1

iro |

0.32

0.26

0.32

| 060 0.

120.

050.1

7 |

0.40

|0.1

2 | |

919

3.20

|5.

13

Tf

| 24

0 801 |

LL'Z | |

4.40

| 0.

76 00o

1 1.

94I

3.55 00p

o

| 0.0

6 I

0.20

0.71

1 I6'0

I

t"-oo'

NTlo© 0.

340.

290.

36

| 66

0 0.13

0.06

| 610 0.4

5 |

0.14

|6.4

7 | |

0l’£ 5.

18

ro

| 18

0I

8.2

| 3.

57|

2.90

| 0.

97|

1.05

| 2.

271

4.29

oro I

Tfoo 0.

210.

72I

0.93

|

00oo ’

I 0.0

5 1

T f T f NO Tf CT> T fo ’ o ’ o ’ 1.2

4 |

0.13

0.07

0.20

|0.5

2 |

0.15

|7.6

0 |

2.99

|5.

22

oo

| 8.

98 oo©

| 0.

431

0.59

| 4.

901

5.92 00o

o

| 0.

050.

090.

85|

0.94

|

00oo 0.

500.

770.

521.7

9 |

0.12

0.12

0.24

|0.5

7 |

0.17

|9.8

4 | |

OOC1

DEG

REE

OF

ROAS

T (M

in)

1 N

SEC

ON

DS

1 R

OA

STIN

G

LOSS

%

(wt/w

t)I

MO

ISTU

RE

CO

NTEN

T %

I %

DRY

MAS

S L

OSS

1 3-

CQ

A|

4-C

QA

1 5-C

QA

11

TOTA

L CQ

A 1

I CA

FFEI

C A

CID

| 5-

p-Co

QA

|4-

FQA

5-FQ

AI

TOTA

L FQ

A I

*

I CA

FFEO

YL

TYRO

SIN

E I

| AN

GOLA

H

|3.

4-di

-CQ

A3.

5-di

-CQ

A4.

5-di

-CQ

AI

TOTA

L di

-CQA

I

3C, 4

-FQA

+3

F, 4

-CQ

A

3C,5

-FQ

A+3

F,5-

CQ

A|

TOTA

L CF

QA

+ FC

QA

I|

CA

FFEO

YL

TRY

PTO

PHA

N

I

*

| TO

TAL

CGA

\|

CA

FFEI

NE

I|

pH OF

A

QU

EOU

S B

RE

W

ONNO

The data shown in the above table indicate that heat transfer to the beans

during the roasting process woh , even in PB grades of washed Robusta coffee and that

the degradation rate of total CQA and di-CQA and hence total CGA content is higher

than that for flat beans. At 8 minutes roast, the total CGA content reduced to about 50

% of that of green coffee.Slow steady degradation of total FQA content was observed.

Roasting loss at various degrees of roasting/almost similar values in AB and PB grades

of washed Robusta coffees. Unlike AB grades of washed Robusta, degradation of

CGA content was higher in this case because of uniform heat transfer to the body of

rounded coffee beans while roasting and the CGA compounds behave -

4.1.2.5 Indian Monsooned Robusta AA grade coffee

The levels of CGA and caffeine determined for Monsooned Robusta AA grade

green coffee (mean of triplicate determination) and duplicate analysis for light, medium

to standard and dark roasted coffees are summarised in the following Table 4.2e.

70

Table 4.2e : The data for chlorogenic acids and caffeine contents (% dry basis) for green

coffee grade Indian Monsooned Robusta AA(mean of triplicate determinations) and the

mean of duplicate determination for roasted coffee

Degree of Roast (Min.)

0 Light(3)

Medium(5)

Dark(6)

In Seconds 0 180 300 360ROASTING LOSS %(wt/wt) 0 12 16 20MOISTURE CONTENT % 7.38 4.75 1.5 1.35% DRY MASS LOSS 9.37 10.12 13.973-CQA 0.64 0.93 0.42 0.234-CQA 0.74 1.13 0.56 0.265-CQA 4.50 2.88 0.93 0.48TOTAL CQA 5.88 4.95 1.91 0.97CAFFEIC ACID 0.12 0.04 0.05 0.035-p-Co QA 0.02 0.17 0.08 0.054-FQA 0.09 0.17 0.17 0.155-FQA 0.69 0.52 0.61 0.42TOTAL FQA 0.78 0.68 0.78 0.57* 0.02 0.05 0.04 0.04CAFFEOYL TYROSINEANGOLA II 0.05 0.14 0.143,4-di-CQA 0.71 0.56 0.17 0.123,5-di-CQA 0.71 0.45 0.15 0.104,5-di-CQA 0.84 0.64 0.17 0.11TOTAL di-CQA 2.26 1.65 0.49 0.333C,4-FQA + 3F,5-CQA 0.13 0.12 0.06 0.043C,5-FQA + 3F,5-CQA 0.08 0.06 0.04 0.02TOTAL CFQA + FCQA 0.21 0.18 0.11 0.06CAFFEOYL TRYPTOPHAN 0.57 0.48 0.24 0.08* 0.12 0.10 0.10 0.05TOTAL CGA 9.98 8.34 3.93 2.32CAFFEINE 2.52 2.59 2.70 2.78pH OF AQUEOUS BREW 5.48 5.46 5.65

The above data showed ’ ' that slow degradation of CGA content at a

lighter roast corresponding to 3 minutes, ushUtat medium roast of 5 minutes roasting

time, the degradation rates of total CGA content is very high. At 6 minutes roast,

corresponding to dark roast, the CGA content was reduced to 25% of the original

content in raw coffee beans.

The following section summarises the Beverage Quality of Indian Arabica and

Robusta coffees.

71

4.2

Beve

rage

Q

ualit

y

<D

$8oo<D

T 3cSi-i60

T 3

t £§

»Cs

g,**«i<N

CO

203H

0>bx2 s> 1 t) 3

« a

a ft 2 3 3

O U O'

esEoj-

u3o>03

b £p2 •'**«s ^3

i . , .o S*>

&■§ 3 «

woc§

« 5 C

33U % 0) cs > o O #

IIO'<ft

^ J2cSo ops 43

ba>> ftBj~ CSCfl•£o

T3 £K. ftCSft

£ s

O'£

1 i£ 3

. ! ? & f t >

ftbX

•C §

t l

ft ft C3

CJ <L>S ftu - |o s?

ft ®^!a §o > 13

*2 ° s£S 1=1 b 43 £ oft pw ft

c sON

O'cft

a<ft

(Uft33

bX

J s

b

IcS cSft ft

+a ft ft

o o < o o ft O O

o +1r\j ft f t O O

< o oft o aftoobx

f t

ftoobx

S3

1ftftOobx

ftoobxo

ft 3 * ° - 5 f to > oc« O cs O5 G f tO

O '

ta<ft

vocs’

a<ft

o'S-

ftoo

O

ftoo

O

ftoo

O

ftoobx

•ftft

boftftoobx

T t•o’

+ftooa

+O '<ft

+O 'cft

ftftbx

bft

43

b;>tA | f t o o >■° ftbx ® a +S ■*t/3 43

00VO

O 'cft

O 'cft

acft

3ftbx3o

b(L)i>

b ^O ft • ° obx 03 C 52 o ft f t a tS es

</sc s

ft

bx 3

bx “P

m b bx'g° ^ -c °I> ft

bX cs

c sr -

4.2.1.1: Washed Arabica A grade coffee

Table 4.3.1a indicate that 5,6, 8 and 10 minutes roasting which corresponds to

11.4 %, 12.6 %, 14 % and 15.4 % roasting loss, and results in fair average quality (

FAQ ) to good to fine in both roast and beverage quality. At the early stage of roastingIAUl J.cntfc

no proper roast and beverage properties have developed an^the extreme roast with

drastic degradation of CGA content. ^ a/nd. burnt and darker colour of roast

and liquor -Jfo beverage quality declined, yielding a strong bitter taste. Flavour was

found to be better when the degree of roast was 6 to 10 minutes.

73

•8Ia;5Spj

g

5S£!§<N*■4<N

<D

s§Oo<DT3cdUh&J)

kJo3cd

T3<D

cnaI£

fOrf—2€8H

T3

ft 2s ^ U O'bJQ ttjPh ^ C nT3

T3 bJO

. tn+2 o ra

£ | “ “o ^2x> w ^ W « H (TO O .

t >

cy oa « 2 QW>

oVO

csVOO v cs

T3

>• TS

Q o

oVO 00 In

the

roas

t and

be

vera

ge

quali

ty re

port

of thi

s pa

rticu

lar

unwa

shed

Ar

abica

AB

gr

ade

coffe

e, on

ly 7

roas

ts we

re pe

rform

ed,

as th

ese

parti

cular

sa

mpl

es

cont

ained

m

ixtu

res

of im

matu

re

and

matu

re co

ffee

bean

s. H

owev

er,

in thi

s pa

rticu

lar

case

, at

10 to

14 m

inut

es

roas

ting,

dar

ker

and

dark

est

roas

t co

rresp

onde

d to

22 and

31

% ro

astin

g lo

ss.

At 5

to 6

min

utes

roa

sting

, pr

oper

med

ium

to st

anda

rd

roas

t wa

s ob

taine

d and

thu

s yie

lded

fair

to go

od

cup

quali

ty.

At 3

, 4,

10 and

14

min

utes

roa

sting

, cup

qu

ality

tend

to FA

Q to

poor

.

4.2.1

.3 M

onso

oned

M

alab

ar

(Ara

bica

) AA

gr

ade

coff

ee:

1§

X )£ooc«£O

o

cd

g<ubQcd<3s03

COtT3cdH

T3

CQ O'

cr

so ~ AS so^ i*3 3 3Q» u bO K) «h C M).1=j

2 . £ - tf *2.in § o •§,cd o o r2 o ;£

[iH ^ Cd U O EA

<N

60 cd

T3

Ml

■a _o S Si’S Q S

C O

oobQo

T 3

2a>sI-h

a>

£oocdo

T 3CD£ooonaoa

<4-1o"O£

cd

§•obQcdgST3£cd

cdOl-H

<D.£

cdg-O h£O<ubQcd<3§

fcdM-4<D>O

Is<u§bO

.■£ 'O

bOflt 3£Ocuw<UWhl-Hoo

4 - >C/3cdOUc

cd-o

od

g5

T 3£cd

cd

§ •<L)bOcdV-4<D5

X>T 3£cd

cdOVhOJ£

£3

bT3T3£cd

TDOt-HcdO haoo£s

bO#£

C/3cdOl-H

Vh<L>tSo,£

C/3

.£0

1cd3o4-i

§•<DbQcd<3S

X )T3£cd

C/3cdOWhl-H<d4-J4-><L>

-O<D£bQ<D

OoT£<D£OO0 3£Osod£3T3<U£<D0 3

o proc

esse

d . T

he

faster

ro

ast

may

be du

e to

bloate

d and

lig

hter

whit

e co

lour

ed

raw

bean

s wh

ich

faci

litate

, qu

icker

tra

nsfe

r of

heat

to the

be

ans,

quick

ly de

velo

ping

pr

oper

col

our

and

swell

ing

and

rich

aroma

and

fla

vour

.

dl

CJScoS»C3*<NciTf

d>

CJ■82&J0

Sso S<i©

d>*SiCO«£S§ririrf

<D£ooCD

T3cdWh00

cd -♦-» cn P

X>

&TP

CDX5

1 /3cd

IP

Cl—,O

*cdPcrCD00cd<3§

03

cd<NCOrf3cdH

T3

a o a<3 o <3 X x

T3

S J3 O' fe* r .£ r*> T3

T3

CT1 cr

WO

r fvdcs 00oo'

00

o o<1 o IX

T3

T3

« "-> « '2 •£ '2 P<w D Q o

00

The

roas

t and

be

vera

ge

quali

ty an

alysis

da

ta sh

owing

tha

t in

unwa

shed

Ro

busta

AB

gr

ade

coffe

e at

5 to

10 m

inut

es

degr

ee

of ro

ast

lead

s

to fai

r av

erag

e to

good

ro

ast

and

liquo

r qu

ality

. 8

and

10 m

inut

es r

oasti

ng

gives

goo

d ro

ast

and

beve

rage

qu

ality

yieldi

ng

good

bo

dy,

neut

ral

and

bette

r ta

ste.

s *

■s1 Sa5A3©

* 3

* 3$2

g«N<N<N

fa£ofafat fcdt-HGO

PQOha+->wfaX0

p4t ffa- a

c a

cd

1 fa

cd

§■faGOcdUtfag

PQ

£rsrnr r

3cd

H

P2 £2 « 5> =n a

ooPh

S ' S(2 2 to

^ .ts2 f ! g O o o '(2

SS '

x S b

+acfin

+a

£

++ 1 11 1-HPo o o O P< < c < ^fan fan X fan h-h

++ 1 11 P IH PO o o

^ oc < <fan fan fan 2

3so'«

<

.H* s'S ®3 3

O ' s

£ £ § X

.2" S ‘3^ <

£"3OCO

<D-asg3

tf(Gcd

POCT1

XbO

-3 °1 ° £ eX*3 X« .55ed cp fecd

Xbo

£b£c

O 5« VOt f S C K

52g -M X>H 5« t f ^> o s §o tf 0(2

8 ^ I? I£ GG 2 &

oo

X

& 1 X3 w •2 £.<2 x"2 -iG2 X fG <+-<JiP4 >-*' £ § . S ’Sw u

i-o o O GG

x Sb

§1^ 2 p ^ GO o cd id

fa 2 fa o o > > cd cd cd

cfaT* x.fa o o

boo

fa fa«2 fa **■'

8 &

faI

Gh

tfOObJO

I

Xbo

bo o "P3 faCA .faed _ O cd

-rt 3 fa *

t-i tft-» o fa o X> bO

cd a) H k t* w fa cdP h-»fa cd

h-hb. CAtf ,0 o o x tfo 5 § .EP

O *3

X3CA’BT3p1-1

<D& "S'fa edO i-*-fa Cd"2 1=5 O ^5 O Oa tf

p

!fafa

r f abO

T3facd

| ^£ ° cu otf bo p

»-T Jg p f a■W

2 fa

’fa ZTp sfa tfCA T3

-2 fa4 tg fa fa

^ fa* p

^ *p T3 a° cd-O .2

if §2 fato J3

^’’S^ tf

aCAcdOt-HUlfa-t->-t->fa

XCA

t f’fa>%

C+H’fa

CA

GOfa

OOo-*->

m"cdCAfacdfa

Xt ffat ffafaOUh

PQPincd-»->CA

faX0

p4t ffaX

CAcd

1

§X

facdfaXfa

£ofafa

t fcdV*HGOPQPh_fa

t-H.oM-HCAfacdt-H

cdfaX

2t-H

S2fa<pl-Ho2

t f§fa

xo

'! •

CAt fcd^ f a

GO#C*H-»

CAcdOt-H

JDX

CAfacdfa

Xtf<OGO

.2"o

t-H

t-HfaQhot-HOhfaCA

SOfaXt ffa_fa"24-»xo

tfHoCAfa

t fcdt-HGO

PQPLh

t ffacd

OXfa

X

CA0

1XCAcd0

P4CA

t ft-Hcd

1O

oofa

XH->

2egen-Hfa#fa

’fafat f

cd&faGOcdfag

Xt ffacd+->CAcdO

facdfa

Xfa

t fcdt-HGO

S3cd

qp

r-r-

fafa

cn <D«8ofacdH->CA

faX0

fant ffaX

CAcd

1f a

CD*8oo0)

TDcdLito

0

*81a>5s»p©

*si

i<N*>*

cd-mMpX>

(2at<D

XIMcdb

CmO

cd

g<<vtocd<3s«

CDfSCOrr«2ed

H

qWQ ^2 .-a*^ S V fa« O'

ooCL,

B Xs §■! gO U O '(2

c

o £ g<1 o <1[in ^ Oh

+o O'< cOh Oh

T3OOa o<! o Oh *-

edaou<

•• S-►» 5•M Oa >ed 3 fa s O'® £ &i§ 3 <

£aoCQ

61) _ •5s wB «i O cnPS 5

a60

Vh13Oa60 §■

a .M^ Oh

m d§ ^ •E bi - i q

•SP fa—H 03b-3*^ O > X

O q ca0 tfl *._ fa fa fa ^ g•H H-> O CA I—H r- Cd OH S "1 a -so ’2 xf a f a CA

^ E ’£

•22 fa 2£ £ ^ -2 S ^73 -SP a * *3 £b? ^ fa^ fa O H O #£ f ig60 J2 ' i?

a a m J

K*~>a60

T3faedq■M ,CA S ied 3

to cd cd c22 faS3 «2

a -q£ f=3 ed

3 3

1 * O JDfa ° ■° PS

^ .£> a 12a j i1 o 3 o o 60 > edO X

a •fa 0 c£ 60

00o

aOoO

q

■§ E.9 o O ■,_‘

b ’gs §^ 60

a fa fa £ faO !_.§? o

§ a 2 8 •° I.q 3

a a•q q2 £•> &

£ 35 -a__, 'b 3o U 2O g ô 3 a

ocs

++ +o O oc «! <Oh Oh Oh

+ao o Oo c <O Oh Oh

§aT38 ,ed 60 fa!

fa h q«2 psa a^ £

*a »i•ps 2^ E, a cr ca afaT3 O

a 'o^ o’S co ^60 O

^ 5o “> 2

fa(Dfa

£Oa

« Is ^- an“ .fa§ «S0 ’S° g•S S3

1 ^2 2^ I•a Ied fa a ed

oocn

a

<D>vfafao

<D

•aeda >i>r> fa a p5 §fa + 2 .faa £

toa <d« SB ' S 5

2 § s ’S)O t f O j

o73

d)7da

BfaoBa

a60

fa£oM

aa

CA

o73

PS fa d ) . f a

| ' sa â60

f a CA

a o 60 >

a60

a60

fa<Da060 60

„ fa +h a ca a ed d>

2 £T CA

1 2^ .3fa a^ <D’S a aCAq g

Vh

a60

• If73£

ao•H—*B.2

a(D

S

fafaCA*'d>

7 dO h

cs

| | |

^ ^ ’S O ' a C / a a O a a 2 g I < : s 3 < ; S 3 < : o S 3 < :c B t S f c a f e a h ^ a t i H

a oq _ P

.5 P 60

^ % o .fa

a "a m (D .fa >i aqa fa

CA facd

2 1 £ 2

=§ g3 J§O §

a §•s? aa ed

Ia

a60

faa

1•§C/3Ctf

1C/2

f r ’g I

| 'fa o fa o 20 -s• a t

1 2o £ m a

a qo. 3

faoM

a

fa fa ^ §^ 8 a "rt 60 b a fa

■s?■° Iw' oC/3 ±5§.fa*-• .cd S i ..q q

a q 1 •§ a 3CAb >rJq >a -b fa s ,cd o o

% !§ Cd r-Q fa

b5 £ o cS c£q -Mb j£ ca a

42 B q oê sq «*_ O °

The

resu

lts

of ro

ast

and

liquo

; r

quali

ty of

Was

hed

Robu

sta

AB

grad

e co

ffee

led

to FA

Q +

fine

roas

t, cup

qu

ality

at 8

to 10

min

utes

roas

ting

and,

even

at

14 m

inut

es r

oasts

qua

lity

still

good

. Th

is ma

y be

due

to the

eff

ect

of the

was

hing

proc

ess

and

the

matu

rity

of co

ffee

bean

s.

oo

CD

CJ•8I

§ssoS©*

*st

I><NN

<D

&oo<D

*T3cdVh00

PQPhcd-»-ncoP

-OoP4T3<DCOcd£t+-iO

.S'

§•ooocd<3

PQ

T3CSroTf

—2cdH

T3O

g o c /p

& 2 3 SU O'

cr1bJO

W)

T3TJ

W>H ibfl + T3

w 2 w x>

00o

csincsoo'

T3< Ph

T3

23 0

<+H

o vo00

<D

«Soo<D

TDcdi- h00

PQP h

cd-*-»C/3

p0 KT 3<Ucocd

1<D-T3

cd

at-H

£‘33t d

l-HcoPX>£T3<D

-C Scocd£C+-HOfl>COcdOD.£3

min

utes

ro

astin

g yie

lds

FAQ

to Go

od

roas

t qu

ality

and

beve

rage

qu

ality

. As

far

as

roas

ting

loss

is co

ncer

ned,

the

re wa

s no

t mu

ch

diffe

renc

e

betw

een

AB

and

PB gr

ades

of

wash

ed

Robu

sta

coffe

es,

but

were

lower

after

5 m

inut

es

roas

ting

loss

when

co

mpa

red

to the

dr

y-pr

oces

sed

R obu

sta

coffe

es.

< £>

s *o

a5*3a

* C )

" a5JSt

<55*5a

I• n

N

r s

<L)

£oo

P+->x nP

X>O

p4td

<DPooc/3poo

s<H-Ho.ITI d

§ •

p(30Pt-HCD>P

PQP

c s

r o

TT2s

p

H

■gaJ £ § *

O O

T3

O 0h

5 3 O ejVh O W)J! « H *rs_, OJO K > 2

wo

c s

T3g 8 < o

T3

T3

’D ^

Q

tSoo

0 )• pcd

t-Hto

c dH->C/3

pX >

O

P4*P

Paooxnao2tfHo

cdPcrO h

3to.5’- p

x nc dOt-H

C/3P-t->gg

V O

edap

tSP h

Po

pX3

w r nP -P4-i3 P

3 ^

g 2

£ o15aacd x n cd P

V O

T3§

i n

c c T

O-»->

to# p

- 3.poO hx nP

t lOo

c dOt-H

04c3*p

- ppcd

a.3* 3<13a

2to2o

cd

§ •

P h

PO

"PPP

PO

P d

T3OOto

T D

9toP

POt-H

t-HP

H->H-H>CD

X>x n

21 3

H->1/3poUh

C/3P

H->

gi n

3X jH->

<DH->PO

* 3p

pM

O

tP

X 3H

p

§ ■

t-Ho§ •

•p9

p-*->H->P

P

x n

2’ p

* £ »

T D

§> >

2o

¥pot-H

P

Sop

pH->C/3

PPop 4

T3PPOox nPO

2x n<D

s§oo

pp

T 3PPOox nPO

2

<T D

OOto

*ppp

£o

p

£op

15 ^ap

X S

T3 P P

C/3 C/3P P Ol-HO h

to . 5’po o

x n P O

3p

xH->

X >

T3 PP P xn O O

Pt-HP

C/3P P P

X >

P X

9p

XP

*8op

* PP

3otpp

t-HP

2to

p

p*P

pX

a *

pX

0

P h

’ PPP

PP

1T 3

PPOOx nPO

OJ O

P C/3Poa

a- - H Q

x n P

SS O hO x n

^ h m

to ^. S

poP 4

top

’poo

top

’ CpT3

3p

3op

p

3t+-Ho

o

!gP

PX 3H->

Px o

aCD

x nctf

4 -»

2

ap "H

* 9a . g

to.a'h—*x nPOt - i

P~ o

P>

Pot-H

0 4

c3* p

T3PP

. P

* 3pa

2to

to.a•3poP hx nP

o x

( NC S

OCS• ppp

VO

cs

T3PP

t -P

o0 0

4.3 Conclusion

This chapter discussed CGA and caffeine contents of variously processed

Indian Arabica and Robusta roasted coffees. Degradation of CGA content was shown

to be slow initially, culminating in drastic degradation at higher roasting degrees.

Beverage quality was found to be good when the roasting time was 6-10 minutes. At

lower roasting times the beverage quality was poor and for times between 10 to 14

minutes the quality was FAQ. For darker roasts the beverage quality declined.

Chapter 5 briefly discusses the kinetics of CGA in Indian Arabica and Robusta

coffee during roasting.

81

CHAPTER 5

KINETICS OF CHLOROGENIC

ACIDS DEGRADATION IN

INDIAN ARABICA AND

ROBUSTA COFFEES DURING

ROASTING

5.1 Introduction

The contents of individual CGA were determined by HPLC, as previously

described. The samples analysed in the foregoing chapter 3 and 4 of this thesis were

washed, dry- and monsoon- processed Indian Arabica and Robusta commercial coffees

both before and after roasting. From the data, it has been observed that CGA contents

are found in the range of 5.8 to 10.0 percentage dry basis. The CQA and di-CQA

isomers predominate over the FQA isomers. As a whole, CGA are degraded depending

upon the degree of roasting. Chemical changes and reactions can take place within the

coffee beans at widely differing rates. They may be exothermic, partially endothermic

and endothermic. Recently a degradation mechanism for chlorogenic acids during

roasting has been proposed (Leloup et al, 1995).

From the data reported in Chapter 4, it is clear that the individual chlorogenic

acids in Indian Arabica and Robusta coffees degrade during roasting at different rates.

It was reported that 5-CQA is rapidly destroyed and 3-CQA produced in Arabica

coffee, and di-CQA subgroup is destroyed more rapidly than the CQA subgroup. This

Chapter explains rate of reaction and behaviour of individual isomers of CGA

pyrolysis. Despite many studies on the chlorogenic acids in green coffee beans

worldwide only a few studies have considered the kinetics of CGA degradation during

roasting Clifford (1985a). Hence, the detailed kinetic behaviour of CGA at various

degree roasting in both Indian Arabica and Robusta coffees,- wet-, dry- and monsoon-

processed has been studied.

5.2 Materials and Methods

Wet-, dry- and monsoon- processed Indian Arabica and Robusta green coffee

samples were roasted at 240°C for various times. Samples were analysed by HPLC and

quantitative results for individual chlorogenic acid during roasting process were

reported and discussed in the Chapter 4. Mean values (triplicate determinations in the

case of green coffees and duplicate determinations in the case of roasted coffee beans)

are taken for this kinetic study of chlorogenic acid degradation.

83

5.3 Results and Discussion

Quantitative data for individual chlorogenic acid in green and roasted coffee

were obtained arid discussed in the Chapter 3 and 4, respectively. These data showed

that CGA degrade during the roasting process but not all CGA subgroups such as

CQA, di-CQA and FQA behaved in an identical manner, as shown in Fig 4.1 and 4.2

for Arabica and Robusta coffees, respectively.

5-CQA decreases early in roasting (3 compared with 6 minutes). At this stage

5-CQA is more sensitive to thermal treatment and partially converted in to 3-CQA and

4-CQA. (Fig 5.1). During this stage there is no destruction of 3-CQA and 4-CQA.

Early in roasting (3-6 minutes) 5-CQA degrades progressively. In this period, 3-CQA

and 4-CQA increase, although total CQA declines. It has previously been suggested

that 5-CQA is isomerised to 3-CQA and 4-CQA and this is consistent with the data

reported here.

A similar effect is seen among the .di-CQA, but here 3,5-di-CQA degrades

more rapidly than the other isomers namely 3,4-di-CQA and 4,5-di-CQA, suggesting

inter-conversion. Total di-CQA declines, and, if there is some loss by hydrolysis, the

mono-ester generated may also at later stage of roasting, ranges from 8 to 10 minutes

duration. Degradation rate of all isomers of CGA is slower in the early stage of

roasting and faster at extreme degrees of roasting, but all CGA sub-groups are

degrading very rapidly, as chemical reaction taking place endothermically.

One might expect FQA to behave similarly to CQA, although possibly with a

difference in rate. At first, the total FQA content appears to be increasing. Since

Arabica coffees do not contain CFQA, production of FQA by partial hydrolysis of

CFQA is not an option. There was little likelihood of FQA being synthesised de novo

during roasting. A possible explanation would be the formation of a co-eluting

contaminant, but spectral analysis of the peaks associated to 4-FQA and 5-FQA

revealed typical CGA spectra. Therefore any contaminant must have similar

84

characteristics.

Rate of reaction for the kinetics of First Order has been calculated for

individual chlorogenic acids and its sub groups by using the equation as follows:

The rate of reaction at time t (degree of roast) is proportional to the content of

CGA at time t, i.e. -dCa/dt = kj Ca. The proportionality factor ki is called the specific

rate constant of the first order reaction. On setting Ca = 1, the significance of this

constant may be seen to be the speed of reaction when CGA content is constant and

equal to unity. The units of ki follow from kj = (-1 /Ca)(dCa/dt) as a reciprocal time.

For first order reaction constant k} should be a constant characteristic of the reaction,

independent of the content, and a function of the temperature only.

Let a be the initial content of CGA and x the decrease in percent of CGA due

to reaction up to time t. Then Ca = (a-x) at time t - dCa/dt = -d(a-x)/dt = djdt, and

now equation -dCa/dt becomes ki(a-x), where ki is constant at time t.

This equation gives the rate for a first order reaction in terms of the initial percentage

of CGA content and the amount of the CGA reacted. Then equation converts at the

start of reaction t = 0, and x = 0, and x = x at time t, obtain

I0X dy/a-x = y ki dt

[-ln(a-*)]0x= M o1

In a/(a-x) = kit

First knowing the initial percentage of CGA and the percentage of reactant at various

elapsed times, a, (a-x), and t may be substituted into the equation and k\ calculated

for. If the reaction is first order, a series of Ary’s is thus obtained which are constant

within the accuracy of the experiment. The above equation was tested graphically in

this experiment. Now the equation becomes

In (a-x) = -kit + In a or log i0 (a-x) = (-/q/2.303)/ + logio a

In this experiment, a is constant, a plot of log (a-x) vs t yielding a straight line in which

the y intercept will be logiofl and the slope (-&//2.303). If the plot constructed from the

experimental data is found to be linear, the reaction is considered as of first order, and

by taking the slope line, results in ki = -2.303(slope). The unit of the rate of reaction is

per second in this experiment. Fig. 5.1a and 5.1b are graphic representations of the rate

of degradation of CQA sub-groups and CGA subgroups vs degree of roast in washed

Arabica A grade coffee, respectively.

Figure 5.1a: Plot of CQA content (ln(a-x)) against time of roasting (s) for Washed

Indian Arabica A grade Coffee.

2.00

y = -0.0013x +1.0

0.50-

= -0.0016x +0.00

3^8 400 450 500 550 600 650 700 750 8<

-0.50

y = -0.0005x -

♦ 3-CQA H 4-CQA A 5-CQA X TOTALCQA

Linear (TOTAL CoA)"““ Linear (5-CQA) “ “ Linear (4-CQA) “ “ Linear (3-CQA) Linear (TOTAL CqA)

‘ Linear (3-CQA)

Figure 5.1b: Plot of CGA content (ln(a-x)) against time of roasting (s) for Washed

Indian Arabica A grade Coffee

8

♦ TOTAL CQA E3 TOTAL FQA

A TOTAL DICQA

X TOTAL CGA Linear (TOTAL CQA)

“ “ Linear (TOTAL CGA)

Linear (TOTAL DICQA)

— Linear (TOTAL FQA)

Linear (TOTAL CGA)

— Linear (TOTAL CQA)“ " Linear (TOTAL FQA)

Linear (TOTAL DICQA)

y = ;-PQ013x +

y y0.0015x - 0.4420

2.00

1.50

1.00

0.50

0.00

-0.50

- 1.00

- 1.50

- 2.00

I rL frU- CS

86

CJ£o-CCOoVhCCSc nCD

*8ooRjo

4

c.2c

l - H

T3<DC/lin<DOoUia.

C!<Du,

£e

TDCOo<dubOa>

c<uVh

iS

rJ

OR j

o■au00o

V-iO3o<+Ho

C/3C/3

<DX5H m

o3H

(ARjCD

4

c.2*3c»—i

V-i

<2

■ —h-»C/3R jOVhO-ioi n<U£

3O•e>V-

cg

CJ<u"2oo<ou

<L>-CH

V)<u£osH

oVO CN00 or~1—< —H OsCO OvVO voCO COF—1 CNO voin voOv TfCO OVCO COl-H O’CO r-rH

co 1—H I ■ 1—1 ( Oi CNI i—H1 l-H1 CO1 CO1 CO1 CN1 o1 l-H1 CN1 ©

oo ■o-00 as •nCO CN•O' ■O'CO VOr~ 00•n os•n OvCN f"o O•n CN 00in ©in CNOSco o1 ©■ o1 o CNi oi o■ CNi CNi CO1 l-H1 O o■ 1-H1 ©

o00 ocn

VOoOsVO COCO •O' in00 VO N-00 •nOV Osin 00vo COCO t-VO 00vo

rHVO

o1 9 O 1—1 CNi o1 9 1—H 1 1—H 1 ■ o1 i-H o1 ©1 rH

oo ol-H 00o Tfl-H r-in VOo 00CO Ov r tr t Tfin 00CO inCO t"in r-'Ov •nCOrH00

VL> co■ CO1 CNi r—I 1 CO1 CN1 i-H1 COi COi CO1 CN1 1“H1 l-H1 CN1 ©■

o00r—1<n VOCO CN•n COOV r-VO 0000

l-H•n Ol-H ■O'■O' oi-H oi-H COos l-H

in ©l-H 00l-H■*t cni CN1 f-H1 9 CNi l-H1 l-H1 CO1 CO1 CO1 CN1 o1 l-H1 CN1 ©1

oVO Ov•*t CN CO•O' r-~o O?—i o00 VO•n OVo

inl-H ol-H1—HO r-o soin

i-H© ©■*t

CO o1 Oi o rH o1 o■ CN1 CNi CNI 1—H 1 l-H ©i 1—H 1 l-H

oo CNco—H 1—1 Os

•n COCN t"CN 00os COf-OvOs 00o

00ov CNOv COCN COr- CNOv CN•nCO o■ o1 o —4 ©1 o1 l-H1 CNi H1 Oi ©■ ©1 l-H

oH"coo r-i* r-~00

i—iin cn COo OVr~ 0000 VOOv VO00 o00

i-Hin OV

t" ©00 COr-CN o■ o o —— 9 l-H1 o■ l-H1 l-H1 l-H1 o1 l-H ©■ 9 1—1O00 o 00 l—H 00o COVO t"rh osCN CNO vor- oOv CN C"in COvo CN© in CO00i-H o1 o rH 1—1 CNi 1—1 1 i-H1 I-H1 l-H1 l-H1 o1 l-H l-H1 ©1 rH

oTfr1—Hi-H r-H rH CN■O' CNOs CO • tin vo00 ■o- •nVO CN•O' COr- VO ©Ov

00 rH1 l—H 1 9 o 9 ©i 9 CN■ CN1 l-H1 o ©■ rH1 ©

oocoVO

f—1<n COCO CNOv •nCO of-

cn<o r-l-H in•n oCNOvl-H CNOV CO•n ONrH ©CO

vu o• o■ o O 9 o1 o1 CN■ CN■ 9 i 1 o ©i rH1 rH

o00asTf OvCO o CNO oO'

l-HOs Hr-

l-Hl-H l-HN- 00o

OsO CNo

l-Hr- ON

oinCO

't o1 o1 o <—i 9 Oi oi CNi CNi 9 l-H1 l-H ©■ rH1 rH

oVO oTf CNCN COin inrH COO' s

H00 CNl-H OsCO CNo VOO ml-H i-H

00 voO COH"co o1 o1 o rH 9 H1 o1 CNi CN1 CN1 l-H1 l-H ©1 rH1 l-H

ooinCN I"*o VOVO

OvCN 00•n 00CN o Oi-H 00CN os oo

OvCN O'© oo CNinCO o■ o1 o rH CNi rH1 rH1 CNi CNi rH1 i-H1 i—H i-H1 rH1 l-H

o CNcn

VOo ot-inCO r-VO ONrH CNo t"O <NC\ oOv inCO Ov ©Ov inVOCN o1 o o ——I CNi i rH1 CNi CNi rH1 O1 i—H l-H1 ©i 1—H

o00 CO ——1 VOoOs00 00o- inr- m*n r*CN COOv VOO 00iTi Tfr- 00O' t"CN f- VOvorH o1 o o l-H CNi rH1 rH1 l-H1 CNI rH1 oi l-H l-H1 ©i i-H

tflT3«oo<ucn<-/Hcn<§ow

own 13

-CQ

A <au1■o-

<0u1m IT

OTAL

C

QA

14-F

QA <

ain IT

OTAL

FQ

A

<ao■■

co~

ca01

co"

<OCJ■■in

|T0T

AL

di-C

QA

|T0T

AL

CQ

AjT

OTAL

FQ

AIT

OTAL

di

-CQ

A|T

0TAL

C

GA

t-oo

5.3.1 Kinetics o f chlorogenic acid degradation in Indian washed, unwashed and

monsoon- processed Arabica coffees

The relative chlorogenic acid percent loss per second for washed, unwashed

and monsooned Arabica shows that there were significant differences in the

coefficients between monsooned and washed or unwashed coffee during roasting.

However, there are no significant differences in degradation rate of chlorogenic acids

between washed and unwashed Arabica when compared at 180, 300 and 360 seconds

roasting periods where as the monsooned Arabica coffees have significantly higher rate

of degradation of all isomers of CGA and hence of total CGA. The degradation rate of

total chlorogenic acids in unwashed Arabica coffee is much more rapid than washed

Arabica coffee when compared at later stage of roasting (480 and 600 seconds). The

differential rate of degradation loss of chlorogenic acids may implicate the method of

processing of coffee. The most striking feature of these coefficients is that CGA

content very rapidly degraded in the shorter roasting periods in the case of monsooned

when compared with the dry and wet processed Arabica coffees. The variation in

coefficients is greater during the early roasting period.

5.3.2 Kinetics o f chlorogenic acid degradation in Indian unwashed, monsooned

and wet- processed Robusta coffees

The relative coefficients for chlorogenic acids in series of roasted Indian

Robusta coffees are shown in Table 5.2. The series of roasted Robusta coffee beans

prepared by different processes reveals that, as Arabica coffee, here also the

degradation rate of chlorogenic acids are much higher than in washed or unwashed

Robusta flat coffee beans. The degradation rates of chlorogenic acids in monsooned

Arabica coffees were significantly higher than in monsooned Robustas. However from

the data, it is also indicated that there are no significant differences in the coefficient

values between flat (AB grade) and rounded (PB grade) within the dry- and wet-

processed Robusta coffees during roasting process. The relative coefficient values in

washed Robusta coffee are slightly higher than in dry- processed Robusta coffees.

However, as far as FQA content in both Robusta and Arabicas is concerned, the

relative coefficients are similar, as discussed in earlier paragraphs of this chapter.

88

vo00

mc n ts Ov 00

tS ts

0000

o00

fS O00ts*

fS <Soot- t sts*

intsts"

om <nt s

Ov«noo o

Ov 00 00c n

00c n

Ovin©’

00 c n c n t sts •no

vo00 ots ts tso

ts00

ts OVo00

c n

t s

00ov 00 c nots

tstststs

r-00

voint s ’

o OV00

Ov00

c noo c nooc n00- 00Ovtsts00

ts ts o vt sts"

tsts00 00voovt" c n

vqts*

o00o*

vqts’

ints* ts

00Ov 0000OV

c nvoo*

c nvoc n

00 00Ov00 ints*

c n

ts*t s00 tsts tso

00"3- CS 00 vo o0000tsin OVoVO tsoo

00Ovo*

00ov

00ts tsvoc n ts ts

o*tsrsts .00c/a ts tso

00 vo00

00ts00tsOVi n

inOvO*

c n ooOvc n

Ovc n Ovts

ts

CS00

oOv

ts00o*

0000r-

00c n

in00ts*

n-00 o

inin

c n

c n

in

Ov00

Tabl

e 5.2

C

ontin

ued

ROBU

STA

CHER

RY

(UN

WA

SHED

) PB

| 096

00T}-c4■ -2.

35

|ot"S -1.

02

|-2.

39

|-1.

64

|

in<N1 -2

.81

|-3.

19

|-3.

44

|-2.

02

|-3.

47

|

1—HVO

cni -2.85

|

-2.12

|

-1.02

|

-1.25

|

-2.02

|

-2.85

| |

ZYZ-

r-t-Ho'

840

-1.8

2-1

.71

-1.0

7 00cn9

I -2

.15

-1.2

1 0000o'1 -2

.31

-2.8

3

ON

CN1 -1.5

6-3.

35

|-3

.51

|-2

.73

-1.8

3-0

.38 0000

o*1 -1.5

6-2

.73

-1.8

30.

70

009

001 -1

.14

-0.4

80.

22-1

.95

-0.8

2 •no'1 -2

.01

-2.4

7-2

.23

-1.1

2 CNcni -3

.18

-2.5

3-1

.69

0.22 <n

o'1 -1.1

2-2

.53 691-

660

480

-0.59

•

cno’■

oo' 1.

02-1

.58

-0.5

9-0

.27

-1.3

5-1

.66

-1.4

7-0

.39

-2.3

9-3

.08

-1.9

8-1

.09

1.02

-0.2

7-0

.39

-1.9

8-1

.09

1.62

360 o

o1

| -0

.22

0.47

ZVl -1.5

2-0

.52

-0.2

1-1

.34

-1.6

4-1

.31

-0.3

2-2

.40

-3.04

|

-1.9

7 oo1 1.

12-0

.21

-0.3

2-1

.97

-1.0

01.

70

300

-0.2

8 100990 1.

31-1

.48 r-

oa -0

.16 t"CN

t -1.7

3-1

.26

-0.3

0-2

.33

-2.98

|

-1.9

1-0

.98

1.31

91 0- -0

.30

-1.9

1-0

.98 00 ?—H

240

-0.3

1 TOO 0.71 1.33

-1.5

1

Oit-Oi -1

.28

-1.7

5 l-Hr-H 1 , -0

.29

-2.5

8 r~-cncn■ -2

.18 060- 1.

33

t"r—Ho’1 -0

.29

-2.1

8 060- 1.82

o00

-0.2

9 900

00t-o’ 1.

38-1

.51

o’1 -0.1

5-1

.27

-1.7

0-1

.18

-0.2

6-2

.38

-3.3

2-2

.05 CN00o'1 1.

38 o ’1 -0.2

6-2

.05

-0.8

2 981

' 1

i i ■M

i .4 - . IDEG

REE

OF

ROAS

T (s

econ

ds)

3-C

QA

14-C

QA

5-C

QA

|TOT

AL

CQA

|4-F

QA

|5-F

QA

|TOT

AL

FQA

3,4-

di-C

QA <

aVa• H

9«ncn"

<<ao

«n

|T0T

AL

di-C

QA

<ao■incn+<o

ucn 3 C,5

-FQA

+ 3F

,5-C

QA

|T0T

AL

CFQA

+

FCQ

ACA

FFEO

YL

TRY

PTO

PHA

N|T

OTAL

C

QA

ITOT

AL

FQA

TOTA

L di

-CQ

ATO

TAL

CFQA

+

FCQ

A| C

AFFE

OYL

TRY

PTO

PHA

N|T

0TA

L CG

A

ROBU

STA

PARC

HMEN

T (W

ASH

ED)

PB

| 096

-1.6

21-1.

44

|-0

.75| oro-

<nr H1 -0.

86

|-0.

57

|-2

.18

-2.39

|

-3.51

|

-1.45

|

-4.05

|

-4.71

|

-3.63

|

-2.23

|

orHo« -0.

57

|-1

.451

-3.63

|

-2.23

|

0.7

5

840

-1.3

8-1

.24

-0.5

2 £10

-1.8

6-0

.71

©1 -2.1

5-2

.26

-2.6

3-1

.23

-2.7

2 r H

coi -2.3

1 00r-rH1

£10 o’1 -1.2

3-2

.31

-1.7

80

.98

009

o00o’1 -0

.64 00o

o’ 0.73

-1.8

2-0

.47

-0.2

4-1

.85

-2.1

6-1

.97

-0.8

9-2

.53

-3.5

1-2

.21

-1.3

70.

73-0

.24

-0.8

9-2

.21

-1.3

71

.40

480

090- -0.4

20.

360.

97-1

.72

o1 -0.2

0-1

.53

-1.9

2-1

.47

-0.5

2-2

.23

-3.1

5 061- -1.1

3| 0

.97 ots

o’1 -0.5

2 oONrH1 -1

.13 0

9'I

096

rHo1 -0

.23

0.52

1.14

-1.6

9 rH©1 -0

.17

-1.3

1-1

.52

-1.2

4-0

.25

-2.1

8-2

.88

-1.7

8-0

.93

1.14 t- 1—<

o’1 -0.2

5 00r-■41 -0

.93

1.7

5

300

-0.2

9 t"rH©1 0.

59 1.22

-1.6

5-0

.36

-0.1

2-1

.15

-1.3

4-1

.14 rHrH

o’1 -2.1

5-3

.04

-1.8

0 160- 1.22

-0.1

2 r HrHo1

o00rH1 -0.9

11

.82

I RO

BUST

A PA

RCHM

ENT

(WA

SHED

) A

B

240

-0.2

8-0

.16 990 1.

27-1

.62

-0.3

5 oH

o100©rHI -1

.24

-1.0

3 rH©o’■ -2

.07

-2.8

8 o1

o00o’1

| 1.

27

010-100- -1

.70 o00

o’1 1.8

7

o00H -0.0

30.

05 CS00o’ 1.

46-1

.54

-0.3

3-0

.07

-0.8

2 00©41 -0.7

90.

21-2

.03

-2.6

0-1

.58

-0.6

51.

46-0

.07

0.21

-1.5

8-0

.65

2.0

3

096

-1.1

4-0

.95

-0.2

3 o'Sto’ -1

.74 'tr-

o’■ -0.4

2-2

.27

-2.1

1-2

.21 oH

r H1 -3.7

3-4

.27

-3.2

7-1

.80 o

o’ -0.4

2 on-

t"fSm*■

ooorH1

60

'I

840 0000

o’1 -0.7

2 ooo’ 0.

64-1

.74 190- -0

.33

-2.1

1-1

.89

-2.0

6-0

.92

-3.3

0-4

.27

-2.9

8-1

.58

1 0.

64-0

.33

-0.9

2-2

.98

-1.5

81

.28

009

-0.6

8-0

.34

0.26

0.92

I -0.3

8-0

.12

-1.5

0-1

.83 00

rH1 -0.4

9-2

.64

-3.2

4-2

.20 00o

rH11

0.92

-0.1

2-0

.49

-2.2

0 00orH1 1.6

8

480

-0.3

3-0

.17

0.54 1.19

991- -0.3

6 zvo- -1.3

0-1

.75

-1.3

1-0

.33

-2.5

9-3

.22

-2.1

6-0

.97

I 1.

19-0

.12

-0.3

3-2

.16

-0.9

71

.79

360 rH

o1 -0.0

2 oo’ 1.

35-1

.68

-0.3

6-0

.12

-1.1

4-1

.55

-1.1

4-0

.16

-2.3

5-2

.98

-1.9

2 160- 1.35

-0.1

2-0

.16

-1.9

2 160- 1.8

5

300

0.03

810

0000o’ 1.

53-1

.62

-0.3

6 rHrH

©1

rH00o’1 -1

.17

-0.7

40.

21-2

.00

-2.7

5-1

.61 orH

rH1 1.53

no- 0.21

-1.6

1 on-

2.0

2

240 610 0.

20 1.07

891

-1.7

5-0

.44

-0.2

0-0

.41

-0.8

9-0

.65

0.47

-1.9

8-2

.75

-1.6

0-0

.62

1 1.

68-0

.20

0.47

oVO■ -0

.62 LVZ

o00r H 0.

220.

15

III691

-1.7

9-0

.49

-0.2

5-0

.33

-0.7

8-0

.46

0.59

0000rH1 -2

.62

-1.4

9-0

.57 691

|

-0.2

50.

59-1

.49

-0.5

72

.20

‘ ■ it .'A '■ *' ./

i DEGR

EE

OF

RO

AST

(sec

onds

)3-

CQ

A4-

CQ

A|5

-CQ

A|T

OTAL

C

QA

|4-F

QA

5-FQ

A|T

0TA

L FQ

A

<ao■■

c n |3,5

-di-C

QA Caui*3i*n

TOTA

L di

-CQ

A

<au■<oc n

+<abiuc n

<cx0 ■c n

+<atLl1Uc n TO

TAL

CFQA

+

FCQ

ACA

FFEO

YL

TRY

PTO

PHA

N|T

0TA

L CQ

ATO

TAL

FQA

TOTA

L di

-CQ

A|T

0TAL

CF

QA

+ FC

QA

CAFF

EOYL

TR

YPT

OPH

AN

TOTA

L CG

A

The rates of degradation of chlorogenic acids in dry processed Robusta coffee

AB and PB are almost identical. The relative coefficients for total CGA content in wet-

processed did not differ significantly either. Table 5.2 shows that the relative

coefficients for CGA content in wet- processed are higher than dry-processed

Robustas. In particular, CGA content is more rapidly destroyed in both monsooned

Arabica and Robusta coffees at early degree of roasting than in dry- or wet- processed

Robustas.

5.4 Conclusion

This chapter attempted a brief study of the kinetics of CGA degradation in

Indian dry-, wet- and monsoon-processed Arabica and Robusta coffees at various

degrees of roasting. The kinetics showed that CGA content very rapidly degraded in

the shorter roasting periods in the case of monsooned when compared with the dry-

and wet- processed Arabica coffees.

The next and final chapter provides a general discussion and overall conclusion

to the research carried out on CGA and caffeine contents in commercial grades of

green and roasted Indian coffee beans.

92

CHAPTER 6

GENERAL DISCUSSION

AND

CONCLUSION

93

The main objective of this study was to determine quantitatively the contents

of chlorogenic acid and caffeine on percentage dry basis in main commercial grades

of wet- dry-, monsoon- processed Indian Arabica and Robusta green and roasted

coffees by using reversed phase HPLC analysis. The study involved beverage quality

of a series of roasted coffees from the lightest to darkest roasts up to the burnt in the

case of both wet- and dry- processed, and light, medium to standard and dark roasts

with respect to monsooned Arabica and Robusta coffees.

The term total chlorogenic acids (total CGA) comprises mainly totals of CQA,

FQA and di-CQA in case of Arabicas. The totals of CFQA + FCQA, and caffeoyl-

tryptophan added in the case of Robusta. Significant differences between the total

CGA content were noticed in wet-, dry-, monsoon- processed Arabica and Robusta

green coffees. There were no significant differences in total CGA and caffeine content

between AB (flat beans) and PB (pea berry) grades within wet- and dry- processed

coffees. Significantly different CGA content were also observed in immature and

fully matured dry- processed Robusta flat green coffee beans.

The mean chlorogenic acid contents (% dry basis) of wet-, dry- and monsoon-

processed Indian Arabica and Robusta green coffees by HPLC are in the range of 5.8

to 7.1 and 7.6 -10.0, respectively. Higher content of chlorogenic acids are found in

both monsooned Arabicas and Robustas when compared to dry- and wet- processed.

Monsooned Arabica and Robusta green coffees are characterised by significantly

higher caffeic acid and total chlorogenic acid contents when compared ' / wet- and

dry- processed coffees.

The means of duplicate determination of caffeine content (% dry basis) were

obtained by HPLC and the values are in the range of 1.3 to 1.5 and 2.7 to 3.0 in Indian

Arabica and Robusta green coffees, respectively. There were no significant

differences in caffeine content associated with method of processing of green coffees

of Arabica and Robusta.

94

Monsooned coffees were also characterised by shorter roasting periods and

more rapid destruction of CGA when compared with wet- and dry- processed Arabica

and Robustas.

Chlorogenic acids steadily degraded during initial degree of roasting and

rapidly decomposed during darker and extreme degree* of roasting. The loss of total

CGA may reach 80-90 % in darkest roasted Arabica and Robusta coffees. Higher

levels of 5-FQA and hence of total FQA than those of green coffee beans have been

observed at medium to standard degrees of roasts in Arabica coffee, irrespective of

processing methods. The increase in the levels of 3-FQA and 4-FQA in slightly

roasted has been reported (van der Stegan & van Duijn, 1980). The relative stability

of FQA has been reported by Pictet and Rehacek (1983) and Clifford (1972). The

increase in 5-FQA and total FQA in this roasting study of Indian Arabica coffee can

be seen from the CGA data analysis of different degree of roasting. Spectral matching

was obtained for those peaks assigned as 5-FQA in both green coffee beans and

roasted coffee beans, signifying the increase is due only to 5-FQA. In all three

processed Robusta coffees, during the roasting process, the degradation rate of FQA is

relatively slow in comparison with caffeoyl tryptophan. This is in marked contrast to

thos*? of CQA and di-CQA and CFQA + FCQA.

The roasting and beverage study carried out reveals that both washed and

unwashed Arabica coffee at medium to standard degree of roasts (viz., 6-10 minutes)

yield better overall roast and liquor quality with optimum beverage pH between 4.9 -

5.0. The roasting losses are about 12-16 % and CGA losses at this stage of roasting

are in the range of 60-50 %.

The optimum beverage pH is obtained at medium to standard degree of roasts

corresponding to 6-10 minutes. In unwashed and washed Robusta this pH value was

observed to be 5.1 - 5.2. In the case of monsooned Arabica and Robusta at medium

roasts, the pH ranges from 5.4- 5.5. However, in all the three processed Arabica and

Robusta coffees the pH ranges from 5.7 to 6.1 at darker to extreme degree of roasts.

With reference to monsooned coffees, it has been observed that the CGA falls

95

rapidly already at a roasting time of 5 minutes/corresponding roasting loss is about 16

%. This contrasts with wet- and dry- processed coffee, where degradation is slower.

This rapid loss of CGA, totalling around 65 %, and a corresponding slight increase

in FQA content yields a medium to standard roast and beverage quality. The

beverage pH at this stage of roasting is 5.4 and yields a well balanced neutral

‘mellow’ cup quality. Even at darker roasts of around 6 minutes, with a loss of about

85% CGA, a good cup quality is still obtained. Overall, monsooned coffee yields a

better roast, a unique neutral, fully developed, balanced ‘mellow” flavour and

smoother cup. This is in divergence from the higher CGA and caffeic acid content in

un-roasted monsooned coffees. This implies that maturity of coffee beans, higher

content of CGA and the method of processing of coffee impact on roast and beverage

quality.

The acidity of coffee brew is an important sensory attribute in determining

desirable coffee quality since it contributes the characteristic coffee flavour. The

optimum beverage acidity corresponding to pH 4.9 to 5.1 is ideal for fine cup quality.

In green coffee beans body/strength, flavour retention and acidity for the beverage are

contributed by chlorogenic acid content, besides other major acids such as citric,

malic, quinic, non aromatic and aromatic acids. This can be seen in the example of the

comparison of degradation rate of CGA content between Robusta coffee and Arabica.

The higher CGA content retention in the case of roasted Robusta tends to yield a

stronger body/strength to the brew.

The present study of chlorogenic acids and caffeine on various processed

green and roasted Arabica and Robusta shows particularly that synergistic effects of

CQA, FQA and caffeic acids may contribute to the roast and beverage quality, in

respect of beverage colour, acidity, flavour and body /strength to the cup. This finding

supports the views of Clifford (1985 b), Ohiokephai et al. ,(1982), Trugo (1984), Illy

and Viani (1995) and also confirms that degradation of chlorogenic. acids during

roasting can be used as an index of degree of roasts. The kinetic study on degradation

chlorogenic acid in Indian Arabica and Robusta coffees of various processed coffees

during roasting process that all sub groups of CGAs are not degrading at the

same rate.

96

6.1 Proposal for future work

1 Fractionation and isolation of different isomers of chlorogenic acids with special

reference to 5-FQA, 3,4-di-CQA, 3,5-di-CQA, 4,5-di-CQA, CFQA + FCQA, caffeoyl

tryptophan and to study the individual behaviour of the said isomers at various degree

of temperature by using HPLC method and further kinetic study.

2. Fractionation, isolation and identification of Angola II peak compound which is

eluted in between 3,4- and 3,5-di-CQA in Angolan Robusta, and also another

unidentified peak compound which appears after caffeoyl tryptophan in Robusta

green coffee.

3. Quantification of actual chlorogenic acids and caffeine contents in the aqueous

brew at different degree of roasted Indian Arabica and Robusta coffees.

97

REFERENCES

98

Amorim HV, (1970) : Nutritional status of the coffee plant and beverage, Indian

Coffee, (12), 331-335

Amorim HV and Amorim V I, (1977) : In, Enzymes in food and beverage processing,

ACS Symposium series, Ed. RL Ory and A. J. St Angelo, No. 4, 27 - 56.

Anon, (1983) : Indian Coffee, 47 : 9-30

Anon, (1985) Coffee in India. In : Coffee Guide, 1-6

Anon, (1995) : Coffee & Cocoa International issue five pp 34-5, Argus Business

Media Ltd, Redhill, Surrey, UK.

Anon, (1996) : Coffee International Coffee Year Book 1996 F. O Licht, Razeburg

GMBH.

Anon, (1996) : Quality Control - Indian expertise. In : Coffee & Cocoa International

issue five, 1996, 34-5, Argus Business Media Ltd, Redhill, Surrey, UK.

Balyaya KJ and Clifford MN, (1995a) : Individual chlorogenic acids and caffeine

contents in commercial grades of wet and dry processed Indian coffee.

Journal Food Science and Technology 32 : 2, 104-108.

Balyaya KJ and Clifford MN, (1995b) : Chlorogenic acids and caffeine contents of

monsooned Indian Arabica and Robusta coffees compared with wet and

dry processed coffees from the same geographic area. Proceedings 16th

International conference on coffee. Science, Kyoto, Japan, April 9 -14,

1995, ASIC 1995,1, 316-325.

Barbaroli G, (1965) : Rassegna Chimica. 1965, 17, 261.

Bradbury AGW and Halliday DJ, (1987) : ~ - - of polysaccharides

in green coffee beans. Proc. 12th ASIC Coll, 1987,265-269.

Clarke RJ and Walker LJ, (1974) : J. Sci. Food Agric., 1974. 25 1389-1404.

Clarke RJ and Walker LJ, (1975) : Proc. 7th Coll. ASIC, 1975, 159-63.

99

Clarke RJ and Macrae R(eds)1988 : Coffee, Volume 3 Technology, Else' . er Applied

Sciences, London, 1-31.

Clarke RJ, (1995): Coffee In : Encyclopaedia of Food Science, Food Technology and

Nutrition (eds.), Academic Press, London Vol. II, 1114 - 1146.

Clifford MN, (1972) : The phenolic compounds of green and roasted coffee beans.

PhD thesis, University of Strathclyde.

Clifford MN and Ohiokpehai O, (1983) : Coffee astringency. Analytical Proceedings

29: 83-86.

Clifford MN, (1985a) : Chlorogenic acids. In : Clarke RJ and Macrae R (eds.) Coffee,

1 Chemistry, Elsevier Applied Science, London, 153-202.

Clifford MN, (1985b) : Chemical and physical aspects of green coffee and coffee

products. In: Clifford MN, Willson KC (eds.), Coffee: Botany,

Biochemistry and Production of Beans and Beverage. Croom Helm,

London, 305-374." I

Clifford MN, (1986) : Coffee bean dicaffeoy^uinic acids. Phytochemicstry, 1986, 25,

1767-1769.

Clifford MN, Kazi T, and Crawford S, (1987) : The content and washout kinetics of

chlorogenic acids in normal and abnormal green coffee beans. 12th Coll. A ^ c

Montreux, 221-228.

Clifford MN, Kellard B and Birch GG (1987) : The chlorogenic acids-physiological

and organoleptic properties. Douzieme Colloque International sur la

Chimie des Cafes Verts, torrefiees et Leurs Derivees, Montreux.

Association Internationale du Cafe, Paris. 254-259.

Clifford MN and Kazi T (1987a) : The influence of coffee bean maturity on the

content of chlorogenic acids, caffeine and trigonelline. Food Chemistry

100

Clifford MN and Jarvis T, (1988) : The chlorogenic acids content of green robusta

coffee beans as a possible index of geographic origin. Food Chemistry

26:291-298.

Clifford MN, Kellard B and Ah-Singh E, (1989a) : Caffeoyl tyrosine from green

robusta coffee beans. Phytochemistry 28: 1989-1990.

Clifford MN, Kellard B and Birch GG, (1989b) : Characterisation

acids by simultaneous isomerisation and trans-esterification with

tetramethylammonium hydroxide. Food Chemistry, 34: 81-88.

Clifford MN, Williams T and Bridson D, (1989c) : The seed content of chlorogenic

acids and caffeine as possible taxonomic criteria in Coffea and Psilanthus.

Phytochemistry, 29: 829-838.

Clifford MN, Kellard B and Birch GG, (1989d) : Characterisation of chlorogenic

acids by simultaneous isomerisation and trans-esterification with

tetramethylammonium hydroxide. Food Chemistry 33 : 115-123.

Correia AMNG, Leeitao MCA and Clifford MN, (1994) : Caffeoyl tyrosine and

Angola II as characteristic markers for Angolan robusta coffees. Food

Chemistry 53 : 309-313.

Dejarassi C and Bemdas H, (1955) : Chem. Ind. 1481-82.

Dejarassi C, Cais M, Mitscher LA, (1958a) : J. Amer .Chem. Soc., 80: 247-248.

Dejarassi C, Cais M and Mitscher LA, (1958b): J. Amer. Chem. Soc., 81: 2386-2398. v

De Maria CAB, Trugo LC and Moriera RFA, (1995) : Simultaneous determination of

of caffeoyl^uinic

total chlorogenic acid, trigonelline and caffeine in green coffee samples

by high performance gel filtration chromatography. Food Chemistry 52,

447- 449.

De Menezes HC, (1994) : The relationship between the state of maturity of raw coffee

beans and the isomers of caffeoylquinic acids. Food Chemistry, 50. 293-

296.

DIN-Norm, (1992) Untersuchung von Kaffee und Kaffee-Erzeugnissen. Bestimmung

des Gehaltes an Chlorogensauren. HPLC Verfahren : DIN 10 766.

Egan H, Kirk RS and Sawyer R, (1981) : Pearson’s Chemical analysis of Foods

Churchill Livingston, London, UK, 234.

Freudenberg K, (1920) : Ber., 1920, 53, 232-239.

Gennaro MC and Abrigo C, (1992) : Fresenius J Anal Chem. 343 : 523- 525.

Haworth RD and Johnstone RAW, (1956) : Chem. Ind., 168.

Haworth RD and Johnstone RAW, (1957) J. Chem. Soc., 1492-1496.

Haworth RD, Jubb AH and McKenna J, (1955) : J. Chem. Soc., 1983-1989.

Humphrey CJ and Macrae R, (1987) : Determination of chlorogenic acids in instant

coffee using derivative spectrophotometry and its application to the

characterisation of instant coffee chicory mixtures. Douziemie Colloque

International sur la Chimie des Cafes Verts, Torreficees et Leurs derivees,

Montreux. Association Internationale du Cafe, Paris, 179- 186.

Illy A and Viani R, (1995) : Espresso coffee (eds.). Academic press Ltd, London

IUPAC, (1976) : Nomenclature of cyclitols. Biochem. J, 153:23-31.

Kaufmann HP and Hamsager RS, (1962) : Zur Kentniss der Lipoide der Kaffeebohne

I : Ueber Fettsaure ester des Cafestols; Fette Seiffen Anstrichm, 64, 206-

213.

102

Kazi T, (1987) : Determination of caffeine and other purine alkaloids in coffee and tea

products by high performance liquid chromatography. In : Proc. 12th Coll.

ASIC, (1987), 227-244.

Kroplien U, (1961) : Green and Roasted coffee tests, Gordian, Hamburg, 1961.

Leloup V, Louvrier A and Liardon, (1995) : Degradation mechanisms of Chlorogenic

acids during roasting. Proceedings 16th International conference on

coffees Science, Kyoto, Japan April 9 -14, 1995, ASIC 1995, I pp 192-

198.

Lingle TR, (1980) : The basics of cupping coffee. Coffee Development Group,

Washington, D. C. 1-32.

Macrae R, (1985) : Nitrogenous Compound,^Coffee, Vol. 1 Chemistry (RJ Clarke &

R. Macrae, eds.), Elseiver Applied Sciences Publishers, London, UK, 1 lb-

152.

Maier HG, (1981) : Kaffee, Paul Parey. Berlin, 1981 35-6.

Maier HG, (1993) : Status of research in the field of non-volatile coffee components

Proc. 15th ASIC, Coll. Montpellier, 567- 576.

Mariani C, Fedeli E, (1991) : Gli steroli delle specie Arabica e Robusta del Caffe' Riv

Sostanze Grasse, 68,111 -115.

Mazzafera P, (1991) : Trigonelline in coffee., Phytochemistry 30, 2309-10.

Menon SN, (1984) : Indian coffee in Europe - An evaluation of quality standards and

cup quality parameters of world coffees, 1-30.

Morishita H, Iwahashi H and Kido R, (1986) : 3-o-caffeoyl-4-o-feruloylquinic acid

from green Robusta coffee beans. Phytochemistry, 25, 2679- 2680.

Morishita H, Takai Y, Yamada H, Fukada F, Sawada M, Iwahashi H and Kido R,

103

(1987) : Caffeoyl tryptophan from green Robusta coffee beans.

Phytochemistry, 26, 1195-6.

Murata M, Okada H and Homma S, (1995) : A novel hydroxy cinnamic acid

derivative of coffee bean, Proceedings 16th International conference on

coffees Science, Kyoto, Japan April 9 -14, 1995, ASIC 1995, I, 199-

207.

Nagel CW, Herrick IW and Clifford MN, (1987) : 1 y

Is chlorogenic acid is bitter? J. Food Sci. 1987, 52 (1): 213.

Naish M, Clifford MN and Birch GG (1993) : Sensory astringency of 5-o-

caffeoyquinicacid, tannic acid and grape-seed tannin by a time intensity

procedure. J. Sci. Food Agric., 1993, 61, 57-64.

Ohiokephai O, Brumen G and Clifford MN, (1982) : The chlorogenic acids content of

some peculiar green coffee beans and their implications for beverage

quality. Dixieme Colloque International sur la Chimie des Cafes Verts,

Torrefiees et Leurs Derivees, Salvador do Bahia, Association

Internationale du Cafe, Paris, 177-185.

Ohiokeehai O, (1982) : Chlorogenic acid content of green coffee beans. PhD thesis,

1982, University of Surrey.

Payen A, (1846) : Compt. rend., 60, 286 -294.

Pictet G, Rehacek J, (1983) : 10th international scientific Colloquium on coffee,

ASIC, Paris, 219.

Robiquet and Bourton, (1837) : Annalen der Pharmacie, f837, 23, 93-95. .

Roffi J, (1971) : Proc. 5th Coll. ASIC, 1973, 61-72.

Scfr'iunemann W and Maier HG. (1986) : Dtsch Lebenm Rdschau, 82 (3); 73-76.

Schunemannn W and Maier HG, (1987) : Dtsch Lebenn Rdschau.

Scholz-Bottcher B, (1991) : Bildung von Sauren und Lactonen, insbesondere aus

Chlorogenesaure, beim Rosten von Kaffee. Dissertation TU

Braunschweig.

Speer K, (1989) : Z. Lebensm Unters Forsch, 189 : 326-330.

Speer K, Tewis R and Montag A, (1991): ASIC, 14th Coll. San Francisco, 1991, 615-

621.

Speer K, Sehat N and Montag A, (1993) : Fatty acids in coffee; Proc. 15th ASIC Coll.,

1993, 583-592.

Steinhart H and Packert A, (1993) : Proc. 15th ASIC Coll, 1993, 593-600.

Streuli H, (1973) : Der heutige Stand der kaffee-Chemie.,Proc. 6th Coll. ASIC, 1973,

61-72.

Thaler H and Gaigl R, (1963) : Z. Lebensm. Unters. Forsch., 1963, 120, 357-63.

Tressl R, (1980) : Proc. 9th Coll. ASIC, 1980, 55-56.

Tressl R , Holzer M. and Kamperschroer, (1982) : Proc. 10th Coll. ASIC, 1982, 272-

292.

Trugo LC and Macrae R, (1982) : The determination of carbohydrates in coffee

products using high performance liquid chromatography. Proc. 10th Coll.

ASIC, 1982, 187-92.

Trugo LC and Macrae R, (1984) : A study of the effect of roasting on the chlorogenic

acid composition of coffee using HPLC. Food Chemistry, 15, 219-227.

Trugo LC, (1984) : HPLC in cc/fee analysis, PhD thesis, University of Reading.

Trugo LC, (1985) : Carbohydrates. In : Coffee I Chemistry eds.(R. J. Clarke & R.

Macrae) Elsevier Applied Science Publishers, London, UK, 83-114.

Trugo LC and Macrae R, (1989) : Archivos Latino Americanos de Nutricion XXXIX

105

(1): 96-107.

van der Stegen GHD and van Duijin J, (1980) : Analysis of chlorogenic acids in

coffee. In : 9th International Scientific Colloquium on coffee, London,

1980. ASIC, Paris, 107-12.

Viani R, (1986) : Coffee ,In : Ullmann’s encyclopaedia of industrial chemistry, Vol.

A7, 315-339. Weinheim :VCH.

Viani R, (1988) : Physiologically active substances in coffee. In : Coffee Vol. 39/

Physiology, Clarke RJ and Macrae R, (eds.), 1-31, Els&ryer Applied

Sciences, London.

106

MATERIAL REDACTED AT REQUEST OF UNIVERSITY

of Surrev CAFFEINE CONTENTS IN WET-, DRY- AND ...epubs.surrey.ac.uk/851634/1/13803878.pdfUniversity...

Documents

Transcript of of Surrev CAFFEINE CONTENTS IN WET-, DRY- AND ...epubs.surrey.ac.uk/851634/1/13803878.pdfUniversity...