https://orcid.org/0000-0001-9416-4070
ABSTRACT
During the dry processing of coffee, beans are classified by size, shape, weight, and color, resulting in the commercial preparation requested by the client, who can be of American, European, or other descent. The preparations consist of a combination of bean sizes and shapes, with a specified maximum number of defects. The objective of this work was to evaluate the effect of the shape and size of the bean of the three main Catimors grown in Mexico, i.e., Colombia, Costa Rica, and Oro Azteca, on the cupping quality attributes of the beverage. The results showed that there was no significant difference between the cupping quality parameters of the Colombia and Costa Rica varieties, whereas Oro Azteca exhibited a final score significantly lower compared to those varieties. For beans of size 13/64 (Z13), significant differences were found between bean forms (flat, peaberry, and mixed) of the Colombia variety. Flat (FL) beans scored almost seven points higher in the final score compared to peaberries (PE). In coffee beans of size Z14 or Z15, mixed (MI) samples scored equal or significantly higher than FL samples in the three varieties. Small beans (Z13 and Z14) of samples not separated by shape (PE and FL together) obtained equal or significantly higher scores in the cup quality attributes of the beverage compared to bigger beans. It was demonstrated that the shape and size of the coffee bean significantly affect the cup quality parameters depending upon the variety.
Introduction
The quality of coffee depends on factors such as the genetic origin, cultivation conditions (type of soil, altitude, and climate), sanitary care, agronomic practices and care of the harvest (degree of maturity), and postharvest handling, including type and control of the production process, storage, roasting, and preparation of the beverage.[Citation1,Citation2] The attributes evaluated in the beverage quality, such as fragrance/aroma, flavor, residual flavor/aftertaste, acidity, body, balance, preference, and sanitary quality of the bean, are important for the acceptance and definition of the bean quality.[Citation3] Which factor or attribute has the major influence on coffee quality definition has not been well elucidated yet, but varietal material has played a relevant role in that definition. Catimors are a group of introgressed Arabica varieties (the cross of Timor Hybrid x Caturra) whose cup quality has been controversial[Citation4], but nowadays, they are essential for coffee farmers for whom coffee leaf rust (Hemileia vastratix Berk. & Broome) is of major concern. This study does not intend to compare the cup quality of Catimors with other varieties, but to bring some light into the effect of the bean size and shape on the cup quality attributes of these materials.
Currently, there is a tendency to improve the quality of coffee, as the final consumer is increasingly aware of the origin and quality of the coffee consumed. For this reason, it is very important to use bean selection methods[Citation5] and to remove defects before roasting. Selection of beans is performed during dry processing, where beans are classified by weight, size, shape, density, and color, such that the physical characteristics of the bean are assessed without establishing a relationship with the cup quality.[Citation6,Citation7] The physical variability of the coffee bean is considerable.[Citation8,Citation9] For example, the size of the bean is closely related to the variety, and for beans of the same variety, the bean size depends on the availability of water during the growing season.[Citation10] Studies[Citation11,Citation12] performed to determine the cup quality with respect to bean size revealed that there were no significant differences between the bean sizes of different varieties of the C. arabica species. It has been shown that no sensory differences among coffee varieties from the same origin are noticed.[Citation13] However, there has been no research suggesting the importance of the selection of bean size and whether this practice can be applied to improve quality attributes in the beverage. The objective of this research was to determine if there are significant differences between the bean sizes of the same variety, between different varieties with the same origin and processing, and between bean shapes, with respect to the quality attributes of the beverage, which can then be highlighted through the selection of green beans.
Materials and methods
This work was carried out in the facilities of the Graduate School (Colegio de Postgraduados, COLPOS) at Córdoba campus.
Coffee samples
Three samples of husked coffee with 12% humidity were obtained from Coffea arabica L. harvested in December 2017 in the experimental coffee plantation of COLPOS, located at 18°88ʹ39” N and 96°92ʹ38” W at 640 m above sea level. Quantities of 50, 11, and 21 kg of husked coffee were obtained from the three main Catimors used in Mexico: Colombia, Costa Rica, and Oro Azteca, respectively. The three evaluated varieties were obtained from the same property. They were cultivated, harvested, and processed under identical conditions, and the same roasting profile and degree of roasting were used[Citation14,Citation15] with a medium grind of coffee.[Citation16]
Dehusking and sample preparation
A balance (Velab granatary scale, model VE-500, d = 100 g) was used to weight samples, which were placed into green coffee containers (SCAA, California, USA). The humidity of beans was measured using a halogen thermobalance (Mettler Toledo, model HG63, Columbus OH, USA). For the coffee dehusking process, a mortar (PROMOR, Veracruz, México) was used at a rate of seven quintals per hour. Quantities of 44, 9, and 17 kg of green coffee beans were obtained from the Colombia, Costa Rica, and Oro Azteca varieties, respectively. One kilogram of green coffee from each variety was obtained with the help of a Boerner Divider (Seedburo Equipment Co., IL, USA) and used for comparison among varieties. Green beans from each variety were first sorted by size with a set of Tyler trays for coffee with circular perforations of 13/64, 14/64, 15/64, 16/64, 17/64, and 18/64 inches in diameter (Mercator Agricola, México), here named as Z13, Z14, and so on. From each of these size groups, a 1-kg sample was taken and labeled as mixed (MI) sample (peaberries and flat beans together) for that size group. MI samples were used to compare the effect of the coffee bean size on the cup quality, within and inter-varieties. MI samples of size 16/64, 17/64, and 18/64, from the three varieties, were not compared (cup quality) with their similar of flat beans (FL) because they contained less than 1% of peaberries (PE). Green coffee beans separated by variety and graded by size were sorted, by hand, into two bean shapes, FL and PE (see . One kilo of PE sample was only available for size 13/64 in varieties Colombia and Oro Azteca, respectively. FL samples (1 kg) were available for the three varieties in sizes from 14/64 (Z14) to 18/64 (Z18), and in varieties Colombia and Oro Azteca for size Z13 or 13/64 (). The amount of sample available of husked coffee for the Costa Rica variety was not sufficient to obtain 1 kg of PE and FL of size 13/64.
Figure 1. Coffee bean shapes: flat beans (FL), peaberries (PE), and a mixture of them (MI)
Table 1. Sample (1 kg) availability by variety, shape, and sieve size of the three main Catimors cultivated in Mexico
Roasting
For coffee roasting, a laboratory roaster (PROMOR, México) with a capacity of 350 g per cycle was used, where 200 g of green coffee per sample was roasted, 24 h before evaluation, for about 10 min with an initial temperature in the roaster drum of 180°C. The roasted coffee was stored in containers designed for this purpose. During the roasting and preparation of the samples, the procedure described in the standards established in the Protocol for Tasting in Cup, published by the Technical Standards Committee of the Specialty Coffee Association of America (SCAA), was applied.[Citation17] To make sure the color of the roasted coffee, according to the International Commission on Illumination (CIE) Standards, was in the range of L* = 20 ± 3, a* = 10 ± 2, and b* = 12 ± 2, a Hunter Lab Mini Scan EZ colorimeter was used, and the results were assessed based on the document “Color Measurement of Roasted Coffee Using HunterLab Spectrophotometers”.[Citation18]
Grinding
For coffee grinding, a grinder (Hamilton Beach) with an adjustable grinding grade and a capacity of 80 g was used. A quantity of 13.75 g of roasted coffee, corresponding to a specific variety, shape, and size, was ground to a medium grind (70% of the grinds pass through the mesh of 0.841 mm or US Standard Sieve number 20) for every five cups per tasting, within 5 min before infusion.
Cupping
Samples of a specific variety, shape, and size were randomly presented to the cuppers. The cup quality evaluation was performed by a group of eight specialized Coffee Tasters with several years of experience in the coffee industry, under the parameters and procedures established by the SCAA. They used trays for coffee samples, round tasting spoons with a capacity of 14 mL, a round tasting turntable with spittoons, white porcelain 250 mL cups, jugs for hot water, electric pots to heat water (97°C), thermometers, jugs of purified (reverse osmosis, carbon filtered) water, samples of roasted coffee, and SCAA standardized cupping formats with nine attributes to be assessed in the following order: fragrance/aroma, flavor, residual flavor/aftertaste, acidity, body, balance, sweetness, uniformity, and clean cup. According to the cupper’s flavor experience with these nine attributes, they gave an overall impression score for the sample (taster’s score). Each attribute, including the taster´s score, was rated on a scale ranging from 0 to 10 with quarter-point increments. Scores below 6.0 are considered bad, from 6 to 6.75 good, from 7 to 7.75 very good, from 8 to 8.75 excellent, and from 9 to 10 as outstanding. Finally, a total score was computed by adding the 10 individual scores. The experiment was designed to assess five cups per sample with five repetitions, and the number of results varied according to the number of cuppers present each day. In total, 700 cups were tasted in the three weeks of the experiment. Each cup was assigned a key so that cuppers were not aware of the variety, shape, or bean size they tasted.
Statistical analysis
Using the R language and environment for statistical computing[Citation19], we applied one-way and two-way analysis of variance for unbalanced data, by fitting generalized linear models, with the use of type III sums of squares for unweighted means. Multiple comparisons of means were carried out with simultaneous tests for general linear hypotheses by using Tukey contrasts (α= 0.05).
Results
Before evaluating the effect of the bean size and shape on the cup quality of coffee, we investigated whether there were significant differences between the three varieties (Catimors) evaluated, and we found that most of the cup quality attributes of the beverage depended on the variety of the coffee bean. The Oro Azteca variety exhibited significantly lower scores compared to the Costa Rica variety, as assessed by aroma, flavor, sweetness, and final score (), and scored significantly lower than Colombia in the attributes of aroma, flavor, aftertaste, acidity, sweetness, taster’s score, and final score. No significant differences were found between the three varieties in the body, uniformity, balance, and clean cup attributes. Next, we evaluated the effect of the size and shape of the green beans, intra- and inter-varieties, obtaining the following results.
Table 2. Cup quality attributes in the Catimors evaluated
Comparison intra-varieties
Depending on the availability of samples (), there were three types of comparisons intra-variety: (a) same size and different shape, (b) same shape and different size, and (c) comparison of shape (MI and FL) and size (Z14 and Z15) together.
Colombia
For small beans (Z13), the cup quality of the Colombia variety depended on the shape of the bean (). FL scored almost seven points higher than PE in the final score, whereas MI and FL were only statistically different in Aroma. For Z14 and Z15, no significant differences were found in the cup quality between MI and FL beans of the Colombia variety. When comparing shape (MI and FL) and size (Z14 and Z15) together, neither main effects nor the interaction was statistically significant. Within shapes (MI and FL), no significant differences were found between the green bean sizes (from Z13 to Z15 for MI and from Z13 to Z18 for FL). Samples not separated by shape (only by size) had similar cup quality attributes as well.
Table 3. Cup quality attributes of the Colombia variety, according to the shape of green beans of size Z13
Costa Rica
In the Costa Rica variety (), the cup quality of beans of size Z14 was different depending on the shape of the bean. In general, MI scored higher than FL beans, with a difference in favor of MI of more than seven points in the final score (). These differences, between MI and FL, were not significant for beans of size Z15, except for aroma. When comparing shape (MI and FL) and size (Z14 and Z15) together, only factor shape presented significant differences between its levels in favor of MI beans, with significant (P < 0.05) interactions between factors, as shown in . Within shapes, the group of MI samples did not present any significant difference between sizes, but the group of FL samples did, as shown in . Beans of size Z15 obtained better scores in flavor, aftertaste, and taster’s score compared to the large beans (Z18), with a significant difference in the final score of more than three points. For beans not separated by shape (only by size), the cup quality of the Costa Rica variety depended on the size of the bean, as shown in . Beans of size Z13 obtained significantly higher scores compared to beans of size Z18, with a difference in the final score of more than five points.
Table 4. Cup quality of the Costa Rica variety according to the size and shape of the coffee bean
Table 5. Observed significance levels for two-way analyses of variance for the cup quality attributes of the Costa Rica variety
Table 6. Effect of size of flat beans (FL) of the Costa Rica variety on the cup quality attributes
Table 7. Cup quality attributes of coffee beans unsorted by shape in the Costa Rica variety
Oro Azteca
The cup quality of the Oro Azteca variety for beans of size Z13 did not depend on the shape of the bean (PE, MI, or FL). When comparing shapes (MI vs. FL) for beans of size Z14 or Z15, there were statistical differences in favor of MI, as shown in . When comparing shape (MI and FL) and size (Z14 and Z15) together, there were significant differences between their levels, in favor of levels MI and Z14, respectively (). Within shapes, the group of MI samples (Z13, Z14, & Z15) did not present any significant difference between sizes, but the group of FL samples (from Z13 to Z18) did, as shown in . Beans of size Z15 obtained lower scores in acidity, sweetness, and taster’s score compared to any other size, with a significant difference in the final score of more than four points compared to the smaller size (Z13) or the larger sizes (Z17 or Z18). For beans not separated by shape (only by size), the cup quality of the Oro Azteca variety depended on the size of the bean, as shown in .
Table 8. Cup quality of the Oro Azteca variety according to the size and shape of the coffee bean
Table 9. Observed significance levels for two-way analyses of variance for the cup quality attributes of the Oro Azteca variety
Table 10. Effect of size of flat beans (FL) of the Oro Azteca variety on the cup quality attributes
Table 11. Cup quality attributes of coffee beans unsorted by shape of the Oro Azteca variety, according to the size of the beans
Comparison inter-varieties
The comparison of MI samples (from Z13 to Z15) showed no significant differences neither between varieties nor between sizes. The interaction MI * Variety was insignificant as well. For FL samples, the comparison of factors variety and size (from Z14 to Z18, as Costa Rica did not have Z13) evaluated together, showed a major influence of variety over the cup quality attributes, as it can be seen in . In flat beans (FL), there was a difference of about three points in the final score in favor of variety Colombia. The size of FL was not determinant in the cup quality attributes of coffee, but there were interesting interaction effects between variety and size for flavor, sweetness, and final score of the coffee brew.
Table 12. Influence of variety and size of FL beans over the cup quality attributes of the beverage
Discussion
Our findings partially coincide with what was reported by Escamilla, Hernández, Rosado, & Zamarripa[Citation20] for the three varieties examined. Their study determined that the Oro Azteca variety was inferior to the other two varieties in terms of aroma, without presenting significant differences between them in the final score. Here, it was found that Oro Azteca scored lower in the main cup quality attributes of coffee: aroma, flavor, aftertaste, acidity, and sweetness. Similarly, Puerta[Citation21] reported that the Colombia variety was better in aroma, flavor, aftertaste, acidity, and sweetness compared to the Oro Azteca variety, which is in agreement with our results.
Almost the whole quantity of PE obtained was of size Z15 or less, with a major proportion on sieve Z13. One-kilogram samples for the three studied shapes (PE, MI, and FL) were only available () for size Z13 in varieties Colombia and Oro Azteca, respectively. When comparing shapes within this size, it turned out that for the Colombia variety, FL scored significantly higher than PE for each compared attribute; meanwhile, in Oro Azteca, there was not any significant difference between shapes. These results do not sustain the popular claim that PE yields a beverage of better cup quality or higher acidity compared to FL. In the Costa Rica and Oro Azteca varieties, beans of small size (Z13 and Z14) unsorted by shape, scored significantly better in acidity compared with bigger beans. Unfortunately, there are no other comparable studies on this subject.
For each variety, MI samples were compared versus FL samples of size Z14 and Z15, respectively. For beans of size Z14, shape significantly modified the overall cup quality attributes of coffee in the Costa Rica variety, flavor and aftertaste in the Oro Azteca variety, and none in the Colombia variety. Every significant difference was in favor of MI samples. In the case of beans of size Z15, shape significantly modulated most of the cup quality attributes of coffee in the Oro Azteca variety, only aroma in the Costa Rican, and none in the Colombia variety. Again, higher scores were attained by MI samples.
It has been reported[Citation11,Citation22] that bean size does not have any correlation with the cup quality attributes of coffee, which is in line with our findings when comparing different bean sizes of shape MI (from Z13 to Z15, ) in the three Catimors evaluated. However, flat beans (FL) of different sizes affected the cup quality attributes of coffee depending upon the variety. In the Colombia variety, there was not any statistical effect of size at all. In the Costa Rica variety, higher scores in flavor, aftertaste, sweetness, taster’s score, and final score were attained by size Z15; meanwhile, in the Oro Azteca variety, Z15 obtained the lower scores in acidity, sweetness, taster’s score, and final score. Coffee beans of different sizes, unsorted by shape, affected the cup quality attributes of the beverage depending upon the variety. In the Colombia variety, there was not any effect of size at all. In the Costa Rica variety, small beans (Z13) obtained the higher score in each cup quality attribute, and in the Oro Azteca variety, only acidity was modulated by the bean size. Small beans (Z13 and Z14) yielded higher acidity compared with the rest of the sizes.
Conclusion
In the Catimors evaluated, the cup quality attributes of the Colombia and Costa Rica varieties are quite similar to each other and significantly higher compared with the Oro Azteca variety. These quality attributes of the beverage could be modulated by selecting the appropriate shape and/or size of the coffee bean. In order to brew a cup of coffee with better quality attributes, it can be recommended the use of flat beans instead of peaberries, the use of coffee beans unsorted by shape (flat beans and peaberries together), and the preference of smaller beans (Z13 and Z14) over bigger ones.
Acknowledgments
We acknowledge the COLPOS Panel of Coffee Tasters, formed with the financial support of the Sectorial Fund SAGARPA-CONACYT, through the project 2016-01-277838, for performing the cupping in this research. We also thank the Postgraduate School, Córdoba Campus, for the use of their materials, equipment, and facilities.
Additional information
Funding
References
- Ameyu, M. A.;. Physical Quality Analysis of Roasted Arabica Coffee Beans Subjected to Different Harvesting and Postharvest Processing Methods in Eastern Ethiopia. Food Sci Qual Manag. 2016a, 57(1), 1–9.
- Bote, A. D.; Vos, J. Tree Management and Environmental Conditions Affect Coffee (Coffea arabica L.) Bean Quality. NJAS - Wageningen J Life Sci. 2017b, 83(1), 39–46. DOI: 10.1016/j.njas.2017.09.002.
- Feria-Morales, A. M.;. Examining the Case of Green Coffee to Illustrate the Limitations of Grading Systems/Expert Tasters in Sensory Evaluation for Quality Control. Food Qual. Prefer. 2002c, 13(6), 355–367. DOI: 10.1016/S0950-3293(02)00028-9.
- Moreira Sobreira, F.; Carlos Baião de Oliveira, A.; Alves Pereira, A.; Sussumu Sakyiama, N. Potential of Híbrido De Timor Germplasm and Its Derived Progenies for Coffee Quality Improvement. Ajcs. 2015d, 9(4), 289–295.
- Gamboa, R. P. Y.; Mosquera, S. S. A.; Paz, N. I. E. Caracterización Física de Café Especial (Coffea arabica) En El Municipio de Chachagüí (Nariño, Colombia). Rev Lasallista Investig. 2015e, 12(1), 90–98. DOI: 10.22507/rli.v12n1a9.
- Oliveros, C. E. T.; Pabón, J. P. U.; Montoya, E. C. R.; Ramírez, C. A. G.; Sanz, J. R. U. Separación de Frutos de Café Verdes Por Medios Mecánicos. Cenicafé. 2010f, 61(3), 260–269.
- Ruge, I.; Pinzon, A.; Moreno, D. Sistema de Selección Electrónico de Café Excelso Basado En El Color Mediante Procesamiento de Imágenes. Rev Tecnura. 2012g, 16(34), 84–93. DOI: 10.14483/udistrital.jour.tecnura.2012.4.a06.
- Severa, L.; Buchar, J.; Nedomová, Š. Shape and Size Variability of Roasted Arabica Coffee Beans. Int. J. Food Prop. 2012h, 15(2), 426–437. DOI: 10.1080/10942912.2010.487967.
- Yuwana,; Silvia, E.; Sidebang, B. Engineering Properties of Coffee Beans from Various Colors of Coffee Cherries. Agric. Agric. Sci. Procedia. 2015i, 3, 274–277.
- Cannell, M.;. Factors Affecting Arabica Coffee Bean Size in Kenya. J Hortic Sci. 1974j, 49(1), 65–76. DOI: 10.1080/00221589.1974.11514552.
- Kathurima, C.; Gichimu, B.; Kenji, G.; Muhoho, S.; Boulanger, R. Evaluation of Beverage Quality and Green Bean Physical Characteristics of Selected Arabica Coffee Genotypes in Kenya. African J. Food Sci. 2009k, 3(11), 365–371.
- Reyes, G. F.; Escamilla, P. E.; Pérez, P. E.; Almaguer, V. G.; Curiel, R. A.; Hernández, G. J. A. Evaluación de Productividad, Calidad Física y Sensorial Del Grano Del Café (Coffea arabica L.), En Cafetos Injertados En El CRUO, Huatusco, Veracruz. Rev Geogr Agrícola. 2016l, 56, 45–53. DOI: 10.5154/r.rga.2016.56.006.
- Puerta, Q. G. I.;. Calidad En Taza de Las Variedades de Coffea arabica L. Cultivadas En Colombia Cenicafé. 1998m, 49(4), 265–278.
- Pizarro, C.; Esteban, D. I.; González, S. J. M.; Forina, M. Use of Near-Infrared Spectroscopy and Feature Selection Techniques for Predicting the Caffeine Content and Roasting Color in Roasted Coffees. J. Agric. Food Chem. 2007n, 55(18), 7477–7488. DOI: 10.1021/jf071139x.
- Baggenstoss, J.; Poisson, L.; Kaegi, R.; Perren, R.; Escher, F. Coffee Roasting and Aroma Formation: Application of Different Time-Temperature Conditions. J. Agric. Food Chem. 2008o, 56(14), 5836–5846. DOI: 10.1021/jf800327j.
- Andueza, S.; Paz de Peña, M.; Cid, C. Chemical and Sensorial Characteristics of Espresso Coffee as Affected by Grinding and Torrefacto Roast. J. Agric. Food Chem. 2003p, 51(24), 7034–7039. DOI: 10.1021/jf034628f.
- SCAA. Specialty Coffee Association of America. Protocols-Cupping Specialty Coffee, 2015q. Published by the Specialty Coffee Association of America, USA.
- Hunter Associates Laboratory Inc. Color Measurement of Roasted Coffee Using HunterLab Spectrophotometers, 2014r. Published by Hunter Associates Laboratory, Inc., USA.
- R Core Team. R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing: Vienna, Austria. URL http://www.R-project.org/. 2013s.
- Escamilla, P. E.; Hernández, A. G.; Rosado, R. O.; Zamarripa, C. A. Calidad En Variedades de Café Orgánico En Tres Regiones de México. Rev Geogr Agrícola. 2015t, 55, 45–55. DOI: 10.5154/r.rga.2015.55.004.
- Puerta, Q. G. I.;. Influencia de Los Granos de Café Cosechados Verdes, En La Calidad Física y Organoléptica de La Bebida. Rev Cenicafé. 2000u, 51(2), 136–150.
- Dessalegn, Y.; Labuschagne, M. T.; Osthoff, G.; Herselman, L. Genetic Diversity and Correlation of Bean Caffeine Content with Cup Quality and Green Bean Physical Characteristics in Coffee (Coffea arabica L.). J. Sci. Food Agric. 2008v. DOI: 10.1002/jsfa.3271.