Vol. 14(14), pp.

171-176, October, 2019

DOI: 10.5897/IJPS2019.4814
Article Number: C53DE1E62206
ISSN: 1992-1950
Copyright ©2019 International Journal of Physical
Author(s) retain the copyright of this article
http://www.academicjournals.org/IJPS Sciences

Full Length Research Paper

Comparative study of caffeine content in beans and

leaves of Coffea arabica using UV/Vis
spectrophotometer
Adane Tadesse Dado1, Yoseph Alresawum Asresahegn2 and Kusse Gudishe Goroya1*
1
Department of Physics, Wolaita Sodo University, P. O. Box 138, Ethiopia.
2
Department of Physics, Kotebe Metropolitan University, P. O. Box 31248, Ethiopia.
Received 29 March, 2019; Accepted 8 August, 2019

Caffeine is one of the compounds found in coffee beans and other parts of coffee plant. However, the
relative caffeine content in different coffee plant parts is least researched. Therefore, this study was
designed to determine the content of caffeine in green coffee leaves and beans. In this work, we report
content of caffeine in green coffee beans and leaves of Coffea arabica using UV/Vis spectrophotometer
with liquid-liquid extraction method. Obtained results indicated that the content of caffeine in green
coffee beans and leaves are in the range of 9.94±0.20 to 11.46±0.10 mg/g and 2.53±0.07 to 2.73±0.16
mg/g, respectively. On top of that, the percentage of caffeine content in green coffee beans is more than
in leaves in the range of 73.44 to 77.31%. These results show that there is statistically significant
different content of caffeine in green coffee beans than its green leaves in a given coffee plant.

Key words: Caffeine, green coffee beans, green coffee leaves, UV/Vis spectrometer.

INTRODUCTION

Due to its pleasant taste, aroma, stimulant effect and different parts of the coffee seedlings (Zheng et al.,
health benefits (Gebeyehu and Bikila., 2015), coffee is 2004). Caffeine content of green coffee beans varies
one of the most widely consumed beverages throughout widely with species and as well as within a given species
the world. Coffee plant is categorized among medicinal due to other factors such as on type of coffee bean,
plants (Eva et al., 2016). Beans and leaves of the coffee degree of maturation, roasting (Ky et al., 2001; Silvarolla
plant are found to have generous amounts of secondary et al., 2004). Coffea canephora coffee beans have higher
metabolites such as phenolic compounds, esters of caffeine content
hydroxycinamic acids and mangiferin which have high with an overall mean value of 2.2%, where as that of C.
level of antioxidant properties and antiinflammatory arabica is about 1.2% (Belay, 2010; Belay et al., 2008;
effects on humans (Campa et al., 2012; Delaroza et al., Franca et al., 2005). Alpdogan et al. (2002) have shown
2017). that percentage of caffeine in coffee beans is 1.36 ±
Literature has revealed that caffeine is distributed in 0.03%. Study of Farah and Donangelo (2006) has

*Corresponding author. E-mail: [email protected].

Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution
License 4.0 International License
172 Int. J. Phys. Sci.

reported that percentage of caffeine content in C. arabica MATERIALS AND METHODS

beans is in the range of 0.96 ± 0.01% to1.23 ± 0.06 %.
Data collection technique
Different researches have been conducted to determine
the concentration content of caffeine in coffee beans. Ripen green coffee beans and matured coffee leaves were taken
Belay et al. (2008) have reported that the percentage of from the same branches of a coffee tree. Simultaneously one coffee
caffeine content in C. arabica is 1.01 ± 0.04% to 1.19 ± bean and coffee leave samples were collected from each woreda.
0.02% while Meareg and Shimelis (2012) have All samples were collected from Southern part of Ethiopia, Hadiya
developed polymer modified glassy carbon electrode for Zone, of five different Woredas, specifically, Gibe, Soro, Gombora,
Misrak Bedawachew and Mirab Bedawachew. Figure 1 displays the
the electrochemical for the determination of caffeine in administrative map of the study area.
coffee. Xiu-Min et al. (2018) showed that the processing The sampling Woredas were selected purposively based on their
and age of the leave of coffee plant have effect on productivity. In total, ten samples (five for leaves and five for beans)
phytochemical profilesand bioactivity of coffee leaves. were collected and analyzed.
Wondimkun et al. (2016) have measured that the content The coffee samples were collected from the model farmers of the
selected Woredas by considering their productivity without
of caffeine in C. arabica beans beverage is determined to considering their varieties. Figure 2 presents green coffee beans
be 501.97 to 564.07 mg/L in different parts of Ethiopia. and leaves collected for sample preparations. The samples were
Gebeyehu and Bikila (2015) have also revealed that the collected in the month August within three day across all sample
percentage of caffeine content of roasted coffee arabica areas.
beans is 0.97 ± 0.049 to 1.53 ± 0.0 03%. According to
Demissie et al. (2016), the caffeine content of green
Standard solution preparations
coffee beans ranges from 0.601 to 0.903 %. Another
research in Ethiopia has revealed that the caffeine Standard stock solution of 1000 ppm (parts per million) of caffeine
content in roasted coffee beans is 1.10% (Shiferaw et al., (Evan, England) was prepared for all sample measurements. The
2018). solution was stirred and heated gently for an hour using magnetic
Zheng et al. (2004) have studied that although caffeine stirrer for more homogeneity. Series of standard solutions were
is distributed mainly in leaves and cotyledons, essentially prepared from the stock solution (2.5, 5, 7.5, 10 and 12.5) ppm for
caffeine from which calibration curve was drafted. All
no caffeine was detected in roots and older brown parts measurements were carried out in short period of time after
of shoots. However, study done by Ashihara et al. (1996) preparation.
has shown that the caffeine content in coffee leaves
(fresh weight) depends on their age’s level. This research
has revealed that the caffeine content in buds, young Sample preparations
leaves, matured leaves and aged leaves is 5.7, 7.1, 2.1 Samples collected from the model farmers were dried at open air
and 2.4 mg/g, respectively. Another study has said that and shaded room temperature for 15 days. Each room temperature
the caffeine content is 3.2 mg/g in fresh young coffee dried sample of green coffee beans and leaves was ground and
leaves and 1.8 mg/g in matured coffee leaves sieved through 500 µm sieve to get a uniform texture granulometry.
(Ratanamarno and Surbkaral, 2017). Gustavo et al. Accurately weighed 120 mg amount of sieved samples was
(2019) have studied the effect of mixture of solvent on dissolved in de-ionized water in a volumetric flask up to mark of 25
mL. Solutions were stirred for half an hour using magnetic stirrer
pigment extraction and antioxidant activity from C. arabica and heated gently (85°C). In addition, the solutions were filtered
leaves. through glass filter to get rid of particles from solution.
Despite of the fact that caffeine exists either in coffee
beans or coffee leaves, consumption of coffee beans is
the common one in most part of the world (Grosso et al., Liquid-Liquid extraction of caffeine
2016). In Ethiopia, besides of consumption of coffee
beans, coffee leaves have been consumed for different Dichloromethane liquid-liquid extraction was deployed in order to
avoid caffeine and CGA spectral overlapping in 200 to 500 nm
reasons since the early times. However, the indigenous wavelength range. The same procedure was followed for both
knowledge of the community about importance of coffee beans and leaves samples. The procedure developed by Belay et
leaves is not the same as for coffee beans. Still today, al. (2008) was deployed to extract caffeine from the solution.
people cook coffee leaves to prepare local beverages Sample solutions prepared above (25 mL solution) were mixed with
and drink it with some other spices. In contrast to the 25 mL dichloromethane (Aldrich-Sigma, Germany) giving total of 50
considerable amount of study on green beans for mL solution of samples.
The solution was stirred for 10 min where a layer was formed
determination of caffeine content, there are relatively few with caffeine making upper layer and CGA making the lower layer.
studies concerned with the metabolite content in leaves. Caffeine was extracted from coffee solution and measured to get
The leaves of the coffee plants which are believed to measured volume (22.5 mL) of the sample. This process was
have high medical effects are overlooked as attention is repeated three times to exhaustively extract caffeine. Either of the
given to beans. extraction for beans and leaves, samples at each round was stored
Therefore, in this work, attempt is made to conduct in separate volumetric flasks. All glass wares and curette used were
thoroughly cleaned, rinsed with de-ionized water and dried before
comprehensive study of determination of caffeine content use. From the extracted and measured caffeine sample solution,
in green coffee leaves and beans for their comparative absorption was measured using double beam UV/Vis
study. spectrophotometer (spectral 50 analytic Jenna, Germany) with
 Dado et al. 173

Figure 1. Map of study area (Hadiya Zone) as indicated by an arrow (Retrieved from
http://www.ethiodemographyandhealth.org/SNNPRAdmMap.jpg)

Figure 2. Collected coffee samples (a) beans and (b) matured leaves.

Wave length ranges of 190 to 1100 nm from which caffeine Lambert’s law at maximum wavelength, the caffeine contents in
concentration were calculated against the standard solution by Beer coffee beans as well as coffee leaves were calculated using
Lambert’s Law at the maximum wavelength. The same extraction equation (1) (Zewdu et al., 2016).
procedure was repeated for all the five areas samples for both
beans and leaves of the coffee.

Determination of caffeine content Statistical analysis

Once the caffeine concentration was calculated from the Data entry management and preliminary summaries were done on
absorbance of the measured sample solution through Beer Microsoft Office Excel spread sheet. All analyses were carried out
174 Int. J. Phys. Sci.

Figure 3. Calibration curve of standard caffeine solution.

Table 1. Concentration of caffeine content in green coffee beans in this work (n = 3).

Total sample volume Measured sample Caffeine content Caffeine content

Study area
(mL) volume(ml) (mg) (mg/g)
Gombora 50 22.5 1.248 10.41B±0.21
Gibe 50 22.5 1.372 11.44A±0.04
Misrak Bedawacho 50 22.5 1.259 10.50BB±0.11
A
Mirab Bedawacho 50 22.5 1.375 11.46 ±0.12
Soro 50 22.5 1.193 9.94C±0.20
Critical value - - - 2.23
LSD - - - 0.27
Means with different letters are statistically different (p < 0.05) in the fifth column.

in triplicates and data were presented as means ± standard and x is concentration in mgL-1. Results obtained in this
deviation. Means of data obtained from quantitative measurement work for green coffee beans are presented in Table 1.
of spectrophotometer were determined. One-way analysis of
variance (ANOVA) at p < 0.05 and SAS 9.1.3 statistical software
Maximum absorbance was obtained at 276 nm
were used to determine statistically significant differences in the wavelength.
mean concentrations of caffeine in leaves and beans as well as The percentage of caffeine content in coffee beans
across study areas. For comparison of the means, the Fisher’s least calculated for five independent measurements ranges
significant difference (LSD) test was used to check the significance from 0.99±0.01% to 1.15±0.01%. As can be seen from
level. Table 1, high concentration value of caffeine is recorded
in Mirab Bedawacho (11.46 mg/g) study area followed by
RESULTS Gibe (11.44 mg/g).
On the other hand, Soro showed low concentration
Validation of the method was carried out in the linearity recording (9.94 mg/g). One-way and SAS analysis of
property of Beer-Lambert’s law from calibration graph variance (ANOVA) at p < 0.05 show that all the mean
correlating the absorption intensity with the concentrations are statistically significant different across
corresponding concentration which was constructed for the study areas, except in case of Gibe nad Mirab
standard caffeine at the highest peak of intensity. The Bedawacho which are statically insignificant figures
calibration curve facilitated measurement of the content (Table 1).
of caffeine samples and validation of the method is In the same way, results obtained in this work for green
displayed in Figure 3. The calibration equation is (Y coffee leaves are presented in Table 2. Maximum
=0.016x + 0.001, R = 0.998, SD = 0.0144%, N = 5) where absorbance was obtained at 272 nm wavelength. The
Y, represents the peak height at maximum wavelength percentage of caffeine content in green coffee leaves
 Dado et al. 175

Table 2. Concentration content of caffeine in green coffee leaves in this work (n = 3).

Total sample volume Measured sample Caffeine content

Study area Caffeine content (mg)
(mL) volume (mL) (mg/g)
Gombora 50 22.5 0.304 2.53B±0.07
Gibe 50 22.5 0.328 2.73A±0.16
AA
Misrak Bedawacho 50 22.5 0.314 2.62 ±0.07
Mirab Bedawacho 50 22.5 0.312 2.60AA±0.15
Soro 50 22.5 0.317 2.64AA±0.08
Critical value - - - 2.23
LSD - - - 0.20
Means with different letters are statistically different (p < 0.05) in the fifth column.

Table 3. Percentage difference of caffeine content between green coffee beans and leaves in this work (n = 3).

Study areas Caffeine in coffee beans (% w/w) Caffeine in coffee leaves (% w/w) Caffeine difference (%)
Gombora 1.04±0.01 0.25±0.01 75.67±1.05
Gibe 1.14±0.01 0.27±0.01 76.12±1.21
Misrak Bedawacho 1.05±0.02 0.26±0.01 74.93±1.68
Mirab Bedawacho 1.15±0.01 0.26±0.00 77.31±1.54
Soro 0.99±0.01 0.26±0.01 73.44±1.34

(calculated for five independent measurements) in this is 1.36 ± 0.03%. The caffeine content of C. arabica beans
work ranges from 0.25±0.01% to 0.27±0.01%. The mean reported by Farah and Donangelo (2006) is 0.96 ± 0.01
concentration recorded in Gibe study area (2.73±0.16 to 1.23 ± 0.06%. Moreover, Belay et al. (2008) showed
mg/g) is observed to be higher while in Gombora area that the caffeine content in C. arabica from different parts
(2.53±0.07 mg/g) study area it is recorded to be less of Ethiopia is 1.01 ± 0.04% to 1.19 ± 0.02%. Results
(Table 2). One-way variance and SAS analysis show that found in this work fall in the concentration range reported
the concentrations of Misrak Bedawacho, Mirab by Shiferaw et al. (2018) (0.46 to 2.82%) which were
Bedawachoand Soro samples are statistically done in the same country. Similarly this work’s result is in
insignificantly different while the others two samples are line with results reported by Demissei et al. (2016) (0.5 to
statistically significantly different. 0.9%) and Belay (2010) (0.95 to 1.27%). Caffeine content
The content of caffeine in green coffee beans and determined in this work is in agreement with the caffeine
leaves are significantly different from each other and the content of C. arabica beans reported by most researchers’
measured caffeine content in each coffee bean samples work in different methods. However, results in the current
is greater than its respective counterpart coffee leave work are found to be less as compared with results
samples (Table 3). reported in (Gebeyehu and Bikila., 2015).
The percentage difference in each study area signifies The caffeine content of green coffee leaves at matured
that the caffeine content in green coffee beans is much growing stage in the current work ranges from 0.25±0.01
higher than that of caffeine content in green coffee leaves (2.53±0.07 mg/g) to 0.27±0.01% (2.73±0.16 mg/g).
in study areas. Caffeine content in green coffee beans is Zheng et al. (2004) showed that caffeine contents for
at least 73.44% higher than its counterpart of caffeine upper leaves or young, middle or matured and lower
content in green coffee leaves. Table 3 displays leaves or an aged leaves is 55.8 μmol/g (10.84 mg/g),
percentage difference of caffeine content between each 47.2 μmol/g (9.17 mg/g) and 42.6 μmol/g (8.27 mg/g),
green coffee beans and coffee leaves samples. respectively. Caffeine content in fresh coffee leaves as
determined by Ratanamarno and Surbkaral (2017) is 3.2
mg/g for young and 1.8 mg/g for matured leaves. The
DISCUSSION current result is less than the results obtained by Zheng
et al. (2004) and Ratanamarno and Surbkaral (2017) but
The caffeine content calculated in this work for green more than the results reported by Ashihara et al. (1996)
coffee beans ranges from 0.99±0.01 (9.94±0.20 mg/g) to (2.1±0.43 mg/g) for matured leaves of C. arabica.
1.15±0.01% (11.46±0.12 mg/g). Alpdogan et al. (2002) Result obtained in current work shows that green
reported that the percentage of caffeine content in coffee coffee leaves are found to have less concentration of
176 Int. J. Phys. Sci.

caffeine content than its respective green coffee beans. Eva BP, Tímea B, N´ora P (2016). Phytochemical over view and
medicinal importance of coffea species from the past until now. Asian
Researchers have revealed that intake of caffeine lowers
Pacific Journal of Tropical Medicine 9:1127-1135.
the level of depression (Grosso et al., 2016) and enhance Farah A, Donangelo CM (2006). Phenolic compounds in coffee.
endurance and sprint of runners and concentration level Brazilian Journal of Plant Physiology 18:23-36.
during tiredness feelings (Conger et al., 2011). The intake Franca AS, Oliveira LS, Mendonca CF, Silva XA (2005). Physical and
chemical attributes of defective crude and roasted coffee beans.
of caffeine also contributes to the completeness of the
Food Chemistry 90:89-94.
desired therapeutic effect for patients (Madhusudhan, Gebeyehu BT, Bikila SL (2015). Determination of caffeine content and
2013; Straube et al., 2011). antioxidant activity of coffee. American Journal of Applied Chemistry
3:69-76. 10.11648/j.ajac.20150302.16.
Grosso G, Micek A, Castellano S, Pajak A, Galvano F (2016). Coffee,
tea, caffeine and risk of depression: A systematic review and dose-
Conclusion response meta‐analysis of observational studies. Molecular Nutrition
and Food Research 60:223-234. 10.1002/mnfr.201500620.
We report comparative study of caffeine content in green Gustavo GM, Claudia DT, Elis DP, Miroslava R (2019) Experimental
coffee beans and leaves of the same plant using double mixture design solvent effects on pigment extraction and antioxidant
activity from C. arabica leaves. Microchemical Journal 146:713-721.
beam UV/Vis spectrophotometer. Results indicate that 10.1016/j.microc.2019.01.073
the concentration content of caffeine in green coffee Ky CL, Louarn J, Guyot B, Dussert S, Hamon S, Noirot M (2001).
beans is 0.99 to 1.15% while that of green coffee leaves Caffeine, trigonelline, chlorogenic acids and sucrose diversity in wild
C. arabica and C. canephore P. accessions. Food Chemistry 75:223-
is 0.25 to 0.27%. These results showed that the caffeine
230.
content in green coffee beans is at least 73. 44% more Madhusudhan SK (2013). Novel analgesic combination of tramadol,
than the one found in matured growing stage green paracetamol, caffeine and taurine in the management of moderate
coffee leaves. to moderately severeacute low back pain. Journal of Orthopaedics
10:144-148. 10.1016/j.jor.2013.07.001.
Meareg A, Shimelis A (2012). Polymer modified glassy carbon electrode
for the electrochemical determination of caffeine in coffee. Talanta
CONFLICT OF INTERESTS 93:122-128. 10.1016/j.talanta.2012.01.058.
Ratanamarno S, Surbkar S (2017). Caffeine and catechins in fresh
The authors have not declared any conflict of interest. coffee leaf (Coffea Arabica) and coffee leaf tea by HPLC. Maejo
International Journal of Science and Technology 11:211-218.
Ratanamarno S, Surbkar S. (2017). Caffeine and catechins in fresh
coffee leaf (Coffea Arabica) and coffee leaf tea by HPLC. Maejo
REFERENCES International Journal of Science and Technology 11:211-218.
Shiferaw M, Adane L, Alemayehu I (2018). Determination of caffeine
Ashihara H, Monteiro A, Gillies FM, Crozier A (1996). Biosynthesis of content of Bale coffee using HPLC analysis. Food Science and
caffeine in leaves of coffee. Plant Physiology 111:747-753. Quality Management 73:23-32.
Alpdogan G, Karbina K, Sungur S (2002). Derivative spectrophotometer Silvarolla MB, Mazzafera P, Fazuoli LC (2004). A naturally
for determination of caffeine in some beverages. Turkish Journal of decaffeinated Arabica coffee. Nature 429: 826.
Chemistry 26:295-302. Straube A, Aicher B, Fiebich BL, Haag G (2011). Combined analgesics
Belay A, Ture K, Redi M, Asfaw A (2008). Caffeine measurement in in (headache) pain therapy: Shotgun approach or precise multi-target
coffee beans with UV/Vis spectrophotometer. Food Chemistry therapeutics. BMC Neurology 11:43. 10.11648/j.ajac.20160402.14.
108:310-315. 10.1016/j.foodchem.2007.10.024. Xiu-Min C, Zhili M, David DK (2018). Effects of processing method and
Belay A (2010). Measurement of integrated absorption cross-section, age of leaves on phytochemical profilesand bioactivity of coffee
oscillator strength and number density of caffeine in coffee beans by leaves. Food Chemistry 249:143-153.
integrated absorption coefficient technique. Food Chemistry 121:585- 10.1016/j.foodchem.2017.12.073
590. 10.1016/j.foodchem.2009.12.052. Wondimkun ZT, Jebessa AG, Molloro LH, Haile T (2016). The
Campa C, Mondolot L, Rakotondravao A, Bidel LPR, Gargadennec A, determination of caffeine level of Wolaita zone, Ethiopia, coffee
Couturon E, Fisca PL, Rakotomalala J, Jay-Allemand C, Davis AP using UV-visible spectrophotometer. American Journal of Applied
(2012). A survey of mangiferin and hydroxycinnamic acid ester Chemistry 4:59-63. 10.11648/j.ajac.20160402.14.
accumulation in coffee (Coffea) leaves: biological implications and Zewdu TW, Jebessa AG, Molloro LH, Haile T (2016). The determination
uses. Annals of Botany 110:1-19. 10.1093/aob/mcs119. of caffeine level of Wolaita zone, Ethiopia, coffee using UV-visible
Conger SA, Warren GL, Hardy MA, Millard-Stafford ML (2011). Does spectrophotometer. American Journal of Applied Chemistry 4:59-63.
caffeine added to carbohydrate provide additional ergogenic benefit 10.11648/j.ajac.20160402.14.
for endurance? International Journal of Sport Nutrition and Exercise Zheng X, Ashihara H (2004). Distribution, biosynthesis and function of
Metabolism 21:71-84. purine and pyridine alkaloids in Coffea Arabica seedlings. Plant
Demissie EG, Woyessai GW, Abebe AS (2016). UV/VIS Science 166:807-813. 10.1016/j.plantsci.2003.11.024.
spectrophotometer determination of caffeine in green coffee beans
from Hararghe, Ethiopia, Using Beer-Lambert’s law and integrated
absorption coefficient techniques. Scientific Study and Research
Chemistry and Chemical Engineering, Biotechnology. Food Industry
17:109-123.
Delaroza F, Rakocevic M, Malta GB, Sanchez PM, Bruns RE,
Scarminio IS (2017). Factorial design effects of plant density, pattern
and light availability on the caffeine, chlorogenic acids, lipids,
reducing sugars and ash contents of C. arabica L. beans and leaves.
Analytical Methods 9:3612-3618. 10.1039/C7AY00721C