Using pH as a Scientific Indication of Coffee Cup Quality

Full text

21
Chemical Technology • March 2014
Chemical Technology • March 2014
20
Petrochemicals
According to Global Exchange, the coffee bean is
the most traded commodity (after petroleum) with
an annual consumption of approximately 5 billion
kilograms. Globally the production of coffee involves an
estimated 23 million farmers and labourers, as well as
11 million hectares of available farm land dedicated
solely to the cultivation of the coffee bean. The largest
producer and exporter of coffee is Brazil, followed by
Columbia, Vietnam and Indonesia, which are responsible
for the 500 000 x 60 kg bags imported to South Africa.
Currently the quality of coffee is determined by sensory
analyses conducted by a trained panel of experts. This evalu-
DWLRQV\VWHPKDVPDQ\ÁDZVDQGDVDUHVXOWWRJHWKHUZLWK
the importance of coffee as a commodity; it is the objective
of this article to explore the methods of harvesting and
URDVWLQJLQRUGHUWRVXJJHVWDSRVVLEOHVFLHQWLÀFTXDQWLWDWLYH
method which indicates coffee cup quality.
Background
The coffee bean belongs to the botanical family of Rubia-
ceae which consists of more than 90 different species. How-
ever, only a few species, including Coffea Arabica (Arabica),
CoffeaCanephora (Robusta) and CoffeaLiberica(Liberian)
are of commercial value. Arabica and Robusta account for
approximately 64% and 35% of the worldwide production
respectively and therefore are the main focus of this article.
Arabica plants are characterized by having large bushes
with dark green oval leaves. They genetically differ from
other species as they have four sets of chromosomes rather
than two. The fruit are oval and require a maturation period
RIDSSUR[LPDWHO\HLJKWPRQWKV7KLVUHVXOWVLQWZRÁDWVHHGV
which become the green coffee beans after drying.
Robusta coffee is a name given to a wide variety of spe-
cies. These plants are characterized as robust shrubs or
small trees which grow up to a height of 10 m. They yield
round fruit which undergo an 11-month maturation period,
resulting in comparatively smaller oval seeds. Robusta is
grown in West and Central Africa, Asia and Brazil while
Arabica can be found in South America, Central and East
Africa and Indonesia. The chemical composition of green
coffee beans of Arabica and Robusta are given in Table 1.
Harvesting and processing
Green coffee beans are obtained through harvesting
coffee cherries when they become bright, red, glossy and
ÀUP+DUYHVWLQJRFFXUVWKURXJKQRQVHOHFWLYHVWULSSLQJRI
whole branches or the more time-consuming, but superior,
product quality method of hand picking. The coffee fruit is
then separated from the pulp through wet or dry process-
ing. Dry processing is simple and inexpensive as it involves
drying of the whole cherry in the sun, followed by mechani-
cal separation of the pulp and hull to yield the green bean.
The wet method requires the utilization of specialized
equipment with substantial amounts of water. This method
Using pH as a
scientic indication
of coffee cup quality
by S Naidoo, L Motingwa, V Mphaga, M Talenga, K G Harding, all of the School of Chemi-
cal and Metallurgical Engineering, University of the Witwatersrand, Johannesburg, South
Africa and L Jewell, University of South Africa
An experiment shows that pH can
EHXVHGDVDVFLHQWLÀFDQDO\WLFDO
WRROWRGHWHUPLQHWKHFRIIHHFXS
TXDOLW\RI.HQ\DQDQG(WKLRSLDQ
EHDQVVLQFHUHODWLRQVKLSVH[LVW
EHWZHHQLPSRUWDQWIDFWRUVVXFKDV
sucrose content, mass percentage
ORVVDEVRUEDQFHDQGS+
ensures that sensory qualities of the bean are better pre-
served resulting in higher quality green beans with fewer
defects. The key difference between the wet and dry meth-
ods is that the wet processing involves the separation of
the beans from the pulp before the drying stage. Once all
the pulp has been removed, the beans are placed in large
fermentation tanks where enzymes are used to break down
the mucilage layer surrounding the seed. The seeds are
then washed and dried either by the sun or in mechani-
cal dryers, in order to reduce the moisture content to the
required level of 12,5%.
Roasting
Green coffee beans only provide the characteristic aroma
DQG ÁDYRXU RI SUHSDUHG FRIIHH RQFH URDVWLQJKDV WDNHQ
place. Roasting of coffee beans typically occurs at 200-
240°C for various periods of time depending on the desired
product. Roasting encompasses a multitude of complex
chemical reactions which include oxidation, reduction,
hydrolysis, polymerization and decarboxylation. These reac-
tions result in physical and chemical changes which lead
to the formation of substances essential to the sensory
quality of the coffee.
The most important reactions which occur are caramel-
ization, the Maillard reactions and the Strecker degradation.
Caramelization is the degradation of sugars within the
green beans and begins to occur at 155°C. Lowering of
the sugar content results in changes in taste and browning
of the beans. Both the Maillard and Strecker degradation
reactions are complex non-enzymatic browning reactions
which are responsible for the degradation of proteins,
polysaccharides, trigonelline and chlorogenic acids. These
reactions begin to occur at 190°C and require rapid cooling
through the use of water or air once roasting is complete
in order to stop the reactions and prevent the formation of
any undesirable characteristics.
Component Arabica coffee Robusta coffee
Polysaccharides 49,8 54,4
Sucrose 8,0 4,0
Reducing sugars 0,1 0,4
Other sugars 1,0 2,0
Lipids 16,2 10,0
Proteins 9,8 9,5
Amino acids 0,5 0,8
Aliphatic acids 1,1 1,2
Quinic acids 0,4 0,4
Chlorgenic acids 6,5 10,0
Caffeine 1,2 2,2
Trigonelline 1,0 0,7
Minerals (as oxide ash) 4,2 4,4
Volatile aroma traces traces
Water 8-12 8-12
Table 1: Chemical
composition of
green Arabica and
Robusta coffee
beans (g/100g

23
Chemical Technology • March 2014
Chemical Technology • March 2014
22
Dependent on the extent of heating, roasting can be
categorized as light, medium or dark. Light roasts result in
QRQXQLIRUPEHDQFRORXUZLWKVRXUXQGHUGHYHORSHGÁDYRXU
medium roasts produce a balanced taste and aroma with a
FLWUXVÁDYRXUZKLOHGDUNURDVWVIRUPFRIIHHVRIORZDFLGLW\
Parameters such as temperature, colour and mass loss are
used as indicators for the degree of roast. However, these
FKDUDFWHULVWLFVRQO\DVVLVWZLWKÁDYRXUSURÀOHVXQGHUOLPLWHG
conditions and are vague as well as subject to individual
perception.
Studies carried out by Wang (2012) indicate that the
quality of roasted coffee cannot be accurately described
solely by physical parameters. It was found that reaching
D VSHFLÀF ÁDYRXU SURÀOHZDV GHSHQGHQW RQ WKH FDUHIXO
control and monitoring of the path taken during roasting,
UDWKHUWKDQLQLWLDODQGÀQDODVVHVVPHQWVRIWKHVXJJHVWHG
parameters. Analytical methods for quantifying the degree
of roast include monitoring the ratio of free amino acids,
alkylpyrazines and chlorogenic acid content. Fobe (1968)
conducted a study on chemical composition of Arabica
roasted at 230°C for various periods of time. It was found
that as roasting time increased the sugar and protein con-
tent decreased and fatty acids increased with negligible
effects on caffeine content.
Chemical compounds produced during the roasting pro-
cess can be broadly divided into volatile substances which
are responsible for aroma, and non-volatile substances
which result in taste sensations such as sourness, bitter-
ness and astringency. Of the non-volatile components, the
main constituents responsible for taste include proteins,
caffeine, acids and sucrose. Protein content is an important
attribute as it increases foaming ability of the brew.
Compositions of amino acids vary depending on tem-
perature and time as proteins tend to deplete rapidly during
the Maillard reactions. Natural acids such as carboxylic
and acetic acids are responsible for the acidity, aroma and
bitterness in sensory impact of brewed coffees. The higher
quality Arabica brews are more acidic with a pH existing
in the range of 4,8-5,1 while lower quality Robusta brews
have a pH of 5,2-5,4.
Sucrose is the principal sugar in coffee and begins to
degrade at 75°C resulting in condensation and the forma-
tion of water, while caramelization begins to occur at 155°C,
UHVXOWLQJ LQ WKH ÀUVW ¶FUDFN· DV ZDWHU DQG FDUERQ GLR[LGH
are released. The sucrose content within roasted beans
contributes to colour, aroma and taste. Caffeine is a major
contributor to strength, body and bitterness of coffee. The
caffeine content of green coffee beans varies according to
species; Robusta and Arabica have a caffeine content of
2,2% and 1,2% by weight respectively.
&XUUHQWPHWKRGVRIGHWHUPLQLQJFRIIHH
FXSTXDOLW\
Coffees are assessed by taste at every stage of process-
ing; especially during grading in the country of origin and
when it is sold and imported to other countries. In addition,
samples from a variety of batches and different beans are
tasted daily to ensure accurate sensory analyses. The cof-
fee beans are not only analysed in this manner for inherent
VHQVRU\ FKDUDFWHULVWLFV DQG ÁDZV EXW DOVR IRU EOHQGLQJ
abilities with other coffees as well as for determining the
correct degree of roast.
The traditional method of grading coffee has several
ÁDZV7KHPDLQ FODVVLÀFDWLRQ FULWHULRQ UHODWHV WR DOWLWXGH
ZLWKVHQVRU\GHVFULSWLRQVXVHGDVFRQÀUPDWLRQIRUWKHGLI-
ferentiation, as beans may have identical sensory evalua-
tions and only differ by geographical aspects. Placing such
a large emphasis on place of origin and altitude of the bean
LVLQVXIÀFLHQWIRUFXSTXDOLW\VLQFHPLVKDQGOLQJRIWKHEHDQV
during harvesting and processing, as well as the presence
of unknown amounts of defective beans, affects cup quality.
In addition, there is a variable nature in the description
of different coffees at the place of origin which prevents a
ZRUOGZLGH VWDQGDUGL]HG FODVVLÀFDWLRQ RI VHQVRU\ TXDOLW\
This occurs as panels are trained differently across the
world, making cross-comparisons of sensory analyses
impossible. Attributes used as evaluation criteria also dif-
fer between regions and companies with only acidity, body
and cleanliness being used uniformly. This method for
HYDOXDWLQJFRIIHHFXS TXDOLW\LV IXUWKHU ÁDZHGDVGHVSLWH
existing ISO standards for the preparation and brewing of
samples, panels seldom follow these conditions and assess
coffee quality based on a unique manner of preparation
also preventing any cross-comparisons. In addition, blind
taste-testing is not a norm, as expert tasters know certain
information such as place of origin and altitude prior to the
coffee quality evaluation, resulting in biased judgement.
Objective
Based on the reviewed literature, it is proposed that once
VXIÀFLHQW DQG VDWLVIDFWRU\ LQIRUPDWLRQ SHUWDLQLQJ WR WKH
green coffee beans is obtained and, based on existing ana-
lytical methods, coffee quality can be measured impartially
DQGVFLHQWLÀFDOO\WKURXJKS+7KLVFDQEHDFKLHYHGWKURXJK
ÀQGLQJFRUUHODWLRQVH[LVWLQJEHWZHHQVXFURVHFRQWHQWPDVV
loss, absorbance and pH of various degrees of roasted
coffee beans.
([SHULPHQWDOSURFHGXUH
A thermogravimetric analysis (TGA) was carried out on
initial samples of Kenyan and Ethiopian coffee beans
obtained from Bean There Coffee Trading (44 on Stanley,
Braamfontein Werf). This was done in order to determine
suitable temperatures for the roasting process. Weighed
samples of 25 g of each bean were crushed to within a
range of 75- 150 μm, and placed on a crucible within the
machine. The thermogravimetric machine was set so that
temperature increase by 10°C/min and operated in the
presence of oxygen. The mass of the sample was then
measured throughout the operating temperature range of
the machine (30-600°C).
Once suitable temperatures for roasting were estab-
lished, 25 g samples of each green bean type were weighed
out for each run. An electronically controlled hot air oven
with a maximum set point temperature of 250°C was
utilized for the roasting process. The weighed samples
of the coffee types were then placed into the oven within
separate crucibles and roasted for roasting periods of 8,
12 and 16 min at each of the chosen temperatures which
were 180°C, 200°C and 220°C.
Once each run was complete, the samples were removed
simultaneously and cooled. The roasted samples were then
ground and percolated using 300 ml of boiling distilled water
within a Sunbeam designer electric coffee grinder (model
SCG – 2012) and Logik 10 Cup electric coffee percolator. The
percolated samples which were roasted at various tempera-
tures and times were refrigerated and stored before further
processing and analysis. Refrigeration of the bean samples
was necessary in order to stop roasting from continuing and
prevent the formation of any undesirable characteristics
which may have compromised further testing and results.
The percolates of each bean sample were then prepared
IRUIXUWKHUDQDO\VLVWKURXJKÀOWUDWLRQDQGWKHQWHVWHGIRUS+
WKURXJKWKHXVHRIDS+PHWHU7KHÀOWHUHGSHUFRODWHGVDP-
ples were then passed through cuvettes for an absorbance
analysis using a UV/Vis photo-spectrometer and then tested
for sucrose content through the use of a High Performance
Liquid Chromatographer (HPLC).
5HVXOWV
TGA
Based on the TGA results, it was found that both the Kenyan
and Ethiopian samples experienced the greatest mass loss
between 190°C and 350°C. The green beans both started
WRORVH VLJQLÀFDQW PDVV IURP &*UHHQ.HQ\DQEHDQV
showed a greater change in mass at a temperature range
of between 220°C and 340°C, whereas green Ethiopian
beans experienced a greater mass loss within a tempera-
ture range of 220°C to 480°C. The percentage of mass
loss increased with increase in roasting temperature. The
rapid mass loss can be attributed to various non-enzymatic
reactions occurring during the drying and pyrolysis stages of
roasting. Based on the TGA results of the two green beans,
together with relevant literature, the roasting conditions for
the beans were determined as 180°C, 200°C and 220°C
for roast times of 8, 12 and 16 min respectively
Composition and pH
Figures 1 – 3 indicate the results obtained for sucrose
content, pH, absorbance and percentage of mass loss for
green Kenyan beans roasted at 180°C, 200°C and 220°C
respectively for roasting times of 8, 12 and 16 min. Similar
ÀJXUHVZHUHREWDLQHGIRUVXFURVHFRQWHQWS+DEVRUEDQFH
and percentage of mass loss for green Ethiopian beans
roasted at the same temperatures and times respectively
(not presented here).
For both bean types it was found that as roast-
ing temperature or time increased there was a
corresponding decrease in sucrose content and
percentage mass loss while the absorbencies of the
beans increased. It was also found that an increased
roasting temperature resulted in a greater decrease
in these characteristics than when roasting time
was increased. In addition it was found that pH ex-
perienced a consistent decrease within increasing
roasting temperature and time which corresponded
to the decrease in sucrose content, mass percent-
age loss and the increase in absorbencies.
Figure 1: Composition and pH of Kenyan Coffee beans
roasted at 180°C
Figure 2: Composition and pH of Kenyan beans
roasted at200°C
Figure 3: Composition and pH of Kenyan beans
roasted at 220°C
Discussion
Compositional changes over varied roasting
temperature and time
Sucrose
The sucrose content is found to decrease in both bean type
percolates as roasting time and temperature increase. This
is consistent with similar studies carried out by Farah (2006)
which attributed the decrease in sucrose concentration to
caramelization. Caramelization is the degradation of sugars
within the green beans which occurs during roasting. The
lowering of the sugar content results in changes in taste
and browning of the beans. The highest concentrations
of sucrose were 3,053 g/l and 2,432 g/l for Kenyan and
Ethiopian beans respectively, which were observed at the
ORZHVWWHPSHUDWXUHRI&7KLVZDVVLJQLÀFDQWO\KLJKHU
relative to the sucrose contents found at 200°C and 220°C
which were obtained as 0 g/l, 0 g/l and 0,032 g/l, 0.011 g/l
for Kenyan and Ethiopian roasted beans respectively. This
is an expected occurrence as caramelization reactions are
favoured during pyrolysis which occurs typically at tem-
SHUDWXUHV H[FHHGLQJ &WKXV D VLJQLÀFDQWO\ JUHDWHU
degradation of sugars would occur at roasting temperatures
of 200°C and 220°C.
Absorbance
The absorbencies of the roasted bean percolates were found
to increase with increasing roast time and temperature
for both Kenyan and Ethiopian bean samples. The initial
absorbencies of green unroasted Kenyan and Ethiopian
beans were obtained as 0,013 and 0,014 respectively. The
bean percolates at various increasing temperatures and
times have shown a progressively increasing absorbance,
indicating that the percolates have an increased amount
of substances which absorb light in the UV or visible light
region of the electromagnetic spectrum. The chemistry of the
roasting process is very complex and not fully understood as
hundreds of reactions occur. However, based on the results
of the absorbencies, it implies that as the roasting process
LQWHQVLÀHGGXHWRHLWKHULQFUHDVHLQWHPSHUDWXUHRUWLPHWKH
concentrations of various compounds increased resulting in
a progressively greater absorbance. In addition this suggests
that a relationship between caramelization and the Maillard
reactions and absorbance, as these reactions are known to
RFFXUGXULQJWKHURDVWLQJSURFHVVDQGDUHFKLHÁ\UHVSRQVLEOH
IRUWKHÀQDOSURGXFWVLQEUHZHGFRIIHH
Mass loss
The percentage of mass loss was found to increase with
increase in roast time and increasing roast temperature for
both bean types. This relationship is further supported by
the results obtained from the thermogravimetric analyses
Petrochemicals