Article pubs.acs.org/JAFC
Effect of Household Coffee Processing on Pesticide Residues as a Means of Ensuring Consumers’ Safety Seblework Mekonen,*,†,‡ Argaw Ambelu,‡ and Pieter Spanoghe† †
Department of Crop Protection, Faculty of Bioscience Engineering, Ghent University, Ghent, Belgium Department of Environmental Health Sciences and Technology, College of Health Sciences, Jimma University, Jimma, Ethiopia
‡
ABSTRACT: Coffee is a highly consumed and popular beverage all over the world; however, coffee beans used for daily consumption may contain pesticide residues that may cause adverse health effects to consumers. In this monitoring study, the effect of household coffee processing on pesticide residues in coffee beans was investigated. Twelve pesticides, including metabolites and isomers (endosulfan α, endosulfan β, cypermethrin, permethrin, deltamethrin, chlorpyrifos ethyl, heptachlor epoxide, hexachlorobenzene, p′p-DDE, p′p-DDD, o′p-DDT, and p′p-DDT) were spiked in coffee beans collected from a local market in southwestern Ethiopia. The subsequent household coffee processing conditions (washing, roasting, and brewing) were established as closely as possible to the traditional household coffee processing in Ethiopia. Washing of coffee beans showed 14.63−57.69 percent reduction, while the roasting process reduced up to 99.8 percent. Chlorpyrifos ethyl, permethrin, cypermethrin, endosulfan α and β in roasting and all of the 12 pesticides in the coffee brewing processes were not detected. Kruskal−Wallis analysis indicated that the reduction of pesticide residues by washing is significantly different from roasting and brewing (P < 0.0001). However, there was no significant difference between coffee roasting and brewing (P > 0.05). The processing factor (PF) was less than one (PF < 1), which indicates reduction of pesticides under study during processing of the coffee beans. The cumulative effect of the three processing methods has a paramount importance in evaluating the risks associated with ingestion of pesticide residues, particularly in coffee beans. KEYWORDS: coffee ceremony, processing factor, roasting, washing, coffee brewing
1. INTRODUCTION In recent years, public health issues affecting the food chain have triggered health authorities to increase their concern on the presence of pesticide residues in staple food items. Pesticides are widely used for the control of pests, disease of crops, and vectors of human and animal diseases. Despite their wide range of application, pesticides may have adverse health effects for consumers because their residues remain on food items. As a result, food safety is a growing concern worldwide because of its direct relation to human health. It is important for consumers to know the possibilitiy of intake of pesticides along with their food. As reported in the literature, food processing may reduce pesticide residues in food and the concurrent scenario of unsafe food for consumers.1 Pesticide residue can be detected in any food items, such as vegetables and fruits, cereals, animal products, or tea and coffee as a result of the application of the pesticides in the field and during storage of crops for the control of pests. Usually, pesticide residue analysis is undertaken for raw agricultural commodities to meet the purpose of marketing to consumers, import/export certification, regulatory monitoring, and others. However, to estimate the level of exposure to pesticide residues in food, it is desirable to investigate the level of exposure at the point of consumption, mainly after food processing, which may lead to a large reduction of pesticide residues.2 Food processing is the action of transforming the food to more edible form before the food is consumed and the processing can influence the pesticide residue present after the raw agricultural commodity is harvested.3 Different household and industrial food processing, such as washing with water or with various © XXXX American Chemical Society
chemicals, peeling for fruits and vegetables, frying, boiling, cooking, and baking, have been able to reduce the pesticide residue in food to below the risk level.4 Coffee is a highly consumed and popular beverage all over the world. Coffee is the second most important commodity next to oil as a source of foreign exchange for most producing countries. In addition, it is considered a primary food due to its contents of compounds with an antioxidant effect and other beneficial biological properties. Its characteristic flavor and aroma make it a unique beverage, and thousands of volatile compounds are found in roasted coffee.5 Currently, Ethiopia is the fifth largest world coffee producer and the top producer and exporter in Africa. The country is believed to be the origin of Coffee arabica. On the other hand, half of the coffee produced in Ethiopia is used for domestic consumption. This makes the nation to be the leading African country in coffee consumption with a yearly per capita consumption of 2.4 kg. Additionally, coffee is the most significant agricultural product to the country’s economy.6 Specifically, the beautiful traditional coffee ceremony makes the country very unique in consumption of coffee. The coffee ceremony and drinking coffee is an important part of Ethiopian cultures. Coffee is offered during holidays, visiting friends, and on a daily basis as a staple food item. The Ethiopian coffee ceremony starts with washing the coffee beans with water and Received: July 10, 2015 Revised: September 1, 2015 Accepted: September 7, 2015
A
DOI: 10.1021/acs.jafc.5b03327 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
raw coffee beans and after different processing steps. Thermo Fisher Scientific supplied MgSO4 (98% purity), NaAc (99% purity), the 50 mL centrifuge tube, the 15 mL dispersive solid phase extraction cleanup tube packed with primary secondary amine (PSA: 99% purity), magnesium sulfate (MgSO4, 98% purity), and octadecyl (C18, 99% purity). The pesticides under study (p′p-DDE (99.9%), p′p-DDD (99.3%), o′p-DDT (100%), p′p-DDT (99%), endosulfan α (98.5%), endosulfan β (98%), Permethrin (98%), cypermethrin (98%), deltamethrin (99%), Chlorpyrifos ethyl (99.5%), Heptachlor epoxide (99.5%) and hexachlorobenzene (98%)) with their highest analytical purity, were obtained from Supelco and delivered by Sigma-Aldrich Logistic Analytical. The physicochemical properties of the pesticides studied in the present study are presented in Table 1.
roasting until it turns to dark brown. After roasting, sometimes the roasted coffee beans are brought to the family members or visitors to allow them to experience the aroma. Afterwards, it will be ground to a fine powder using a traditional wooden mortar and pestle. The powder is poured into boiled water in a special local coffee clay pot called a “jebena”. Coffee will be ready to serve when the steam with an attractive flavor starts to come out from the nozzle. After that, the jebena is placed on ground for about 3 min to let the coffee sludge settle to the bottom. The brewed coffee is then poured into small cups and served (personal observation). In general, coffee is an important crop for domestic consumption as well as a source of hard currency for the nation. However, studies have indicated that coffee beans have different pesticide residues. Even though coffee is the most important agricultural product for human consumption as well as for the economic growth of the producing countries, the beans might have different pesticide residues that can expose consumers to hazardous agrochemicals.7 Because of the high consumption of coffee and its economic importance, people from the producer, exporter, and consumer countries give more attention to its safety.8 Even though pesticide contamination in green coffee beans is limited during agricultural treatments, there may be contamination during transportation and storage.9 Different studies have stated that coffee is contaminated by different classes of pesticides applied in the field. According to the U.S. Food and Drug Administration, from a survey of 60 imported green coffee bean samples, some of them contained different pesticide residues, such as chlorpyrifos at a concentration ranging from 0.01 to 0.04 mg/kg and pirimiphos methyl at 0.01 mg/kg.10 From a study done in Brazil, the residues of different pesticides, such as endosulfan, chlorpyrifos, cypermethrin, and captafol, have been detected in coffee beans from either registered or illegal use.11 Additionally, from the previous study done in Ethiopia, DDT, cypermethrin, permethrin, deltamethrin, chlorpyrifos ethyl, and endosulfan (α and β) were detected in coffee beans at concentrations ranging from 0.011 to 1.115 mg/kg7. The pesticide residues, to a variable extent, left in the food materials after harvesting, are beyond the control of consumers and may have deleterious effects on human health. Because Ethiopia is the largest producer, exporter, and consumer of coffee in Africa, a pragmatic solution should be developed to tackle the problem of coffee safety. Different studies have revealed that food processing can be one important solution that has a significant role for the reduction of pesticide residues in different food items.2−4,12 Nowadays, there is an increasing need for information on the effect of various food processes on pesticide residues. Few studies have indicated the effect of coffee roasting on the reduction of pesticide residues.13,14 However, no studies are available on the effect of household coffee processing (washing, roasting, and brewing) methods on pesticide residues. Therefore, the main aim of the present study is to evaluate the effect of household processing on pesticide residues in coffee and to determine the processing factor (PF) for each of the processing steps.
Table 1. Physicochemical Properties of the Pesticides under Studya
a
pesticides
water solubility (mg/L)
log kow
p′p-DDE p′p-DDD o′p-DDT p′p-DDT endosulfan α endosulfan β permethrin cypermethrin deltamethrin heptachlor epoxide hexachlorobenzene chlorpyrifos ethyl
0.025 0.120 0.090 0.085 0.530 0.280 0.006 0.009 0.002 0.350 0.006 1.400
6.91 6.51 6.02 6.79 3.83 3.62 6.10 5.40 6.10 5.40 5.73 4.70
vapor pressure (mmHg) 1.6 1.4 6.0 1.1 1 1 2.15 2.3 1.5 3.0 1.09 1.9
× × × × × × × × × × × ×
10−7 10−6 10−6 10−7 10−5 10−5 10−8 10−7 10−8 10−4 10−5 10−5
mode of action contact contact contact contact contact contact contact contact contact contact systemic contact
log kow = logarithm of octanol-water partion coefficient.
2.2. Sampling. A 3 kg portion of green coffee beans was bought from a local market in Jimma zone, southwestern Ethiopia in August, 2014 and served as a blank and spiked sample. Samples were packed in polyethylene plastic bags, after which they were sealed and labeled properly. The samples were transported to the laboratory and stored at −20 °C until extraction was done. 2.3. Quality Control. 2.3.1. Linearity. The quantitative determination of the pesticide residues in processed and unprocessed coffee beans was done on the basis of an external standard method. The calibration curves were obtained by injecting five different concentrations of the pesticide standards in the range of 0.005−1 mg/L. The regression coefficient (r2) was >0.995 for all the pesticides under study. Identification and quantification of the pesticides were done on the basis of the retention time and peak area, respectively. 2.3.2. Recovery Study. To evaluate the performance of the analytical procedures recovery (% recovery), studies were undertaken for each pesticide. A 10 g sample of pesticide-free coffee beans was weighed on the analytical balance and spiked with 40 μL of 100 mg/L of each pesticide in three replicates. The extraction was performed in a similar way with the samples. The percent recovery was derived from the equation: (amount of pesticide observed)/(amount of pesticides spiked) × 100. The percent relative standard deviation (% RSD) was obtained by dividing the standard deviations by the average concentration and multiplying the result by 100.
2. MATERIALS AND METHODS 2.1. Chemicals and Reagents. Analytical grade acetonitrile (99.9% purity) was supplied by VWZ-prolabo. High performance liquid chromatography grade n-hexane (98% purity) and acetone (98.9% purity) that were obtained from ALLthec were used to extract the pesticide residues from the B
DOI: 10.1021/acs.jafc.5b03327 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry 2.4. Treatment of Raw Coffee Bean. A 10 g portion of raw coffee beans was weighed on an analytical balance and spiked with 40 μL of a 100 mg/L solution of each pesticides under study (DDT metabolites, cypermethrin, deltamethrin, permethrin, endosulfan [α, β] and chlorpyrifos ethyl, heptachlor epoxide, and hexachlorobenzene) in three replicates to increase the reliability of the results. Spiking was done for each household coffee processing method (washing, roasting, and brewing). The pesticides selected for this study were detected in the previous study done on Ethiopian coffee beans,7 except for heptachlorepoxide and hexachlorobenzene, in which the investigators expected them from previous application in the environment. The spraying of the coffee beans was done in a Petri dish to get equal distribution of the pesticides for each bean. After spraying, the Petri dishes were covered with aluminum foil and placed in refrigerator at 4 °C for 24 h to increase the contact time between the pesticides and the matrix. After 1 day, the processing methods (washing, roasting, brewing) and the extraction, clean up, and analysis were followed as stated below. 2.5. Extraction and Cleanup of the Samples. To assess the effect of washing, roasting, and brewing of coffee on pesticide residues, the extraction and cleanup of the raw and processed coffee beans were done using the modified Quick, Easy, Cheap, Effective, Rugged and Safe (QuEChERS) method combined with a dispersive solid phase extraction cleanup (dSPE) method. The analytical procedure was based on the method AOAC 2007.01.15 The subsequent processing steps of the coffee beans were done to correspond, as closely as possible, to the actual traditional household coffee processing in Ethiopia. Blank coffee bean (nonspiked) was also analyzed together with the raw and processed coffee beans. The extraction and cleanup procedures for raw coffee beans and processed coffee beans were as follows: 2.5.1. Raw coffee beans. After holding the spiked samples for 1 day, the treated raw coffee bean was ground and homogenized using a coffee grinder with a knife in it. A 15 mL portion of acetonitrile was added, and the sample was shaken by hand for 30 s. Then 6 g of MgSO4/1.5 g of NaCl/1.5 g of Na3citrate·2H2O/0.750 g of Na2Hcitrate was added, and the mixture was again shaken for 5 min to prevent formation of agglomerates between the water and MgSO4. Then the samples were centrifuged at 5000 rpm for 5 min. The upper organic layer was taken for cleanup using a 15 mL dispersive solid phase extraction tube (d-SPE) containing 300 mg of PSA, 900 mg of MgSO4, and 150 mg of C18 and then shaken for 30 s. The cleaned extracts were centrifuged to get a clear solution. After centrifugation, 5 mL of the upper layer was taken into a 100 mL flat bottom flask and evaporated to dryness using a rotary evaporator (N18673 Rotavapor; Buchi) at a temperature of 40 °C. A 2 mL portion of n-hexane/acetone (9:1 V/V) was added for solvent exchange to make the sample amenable for GC/ ECD analysis. The blank coffee bean was also extracted in a similar way, as were with the spiked samples. Then the extract was placed into a vial for GC/ECD analysis. 2.5.2. Washed Coffee Beans. The spiked coffee beans were washed thoroughly for 5 min under normal tap water (25−30 °C), which resembles coffee bean washing at the household level. Then all other procedures for the unprocessed coffee beans were applied for the extraction, cleanup, and analysis for the presence of pesticides. 2.5.3. Roasted Coffee Beans. The spiked coffee beans were roasted on a stove at a temperature range of 230−240 °C (light
to medium) roasting and an average time of 12−14 min until the characteristics aroma or flavor of coffee appears. The roasting temperature and duration of roasting was based on a study done by Moon and Shibamoto.16 The roasting processes were done in a way similar to the traditional Ethiopian household coffee roasting procedures. Then the other procedures for extraction and cleanup used for raw coffee beans were applied. 2.5.4. Brewed Coffee Beans. Some heat-resistant pesticides may be detected in a cup of coffee after the coffee beans were roasted. To determine the effect of brewing on the pesticides under study, the brewing process was also undertaken. The roasted coffee beans were ground to a fine powder using a coffee grinder (type 4041, 220-230 V/150w). The fine coffee powder was added to a coffee pot traditionally called a “Jebena” containing 100 mL boiled water and brewed for 10−12 min in a way similar to that of the tradition Ethiopian coffee brewing processes (personal observation). After brewing, the coffee pot was picked up from the stove and put on the ground until the infusion of the coffee was cooled and the coffee powder or sludge had settled to the bottom of the coffee pot. Then the upper liquid layer was removed carefully to the cups, and then extraction, cleanup and analysis of the brewed coffee solution was done in a way similar to that of the raw coffee beans. 2.6. Determination of Processing Factor (PF). The effect of household processing on the level of pesticides often correlates with the physicochemical properties of the pesticides under study, so it is important to adequately monitor the processing factor. The processing factor (PF) for all transformation steps was calculated as the ratio between the pesticide concentrations in the processed commodity (mg/kg) to the pesticide concentration in unprocessed (raw) commodity (mg/kg). According to Bonnechere and his colleagues,17 a PF of 1 indicated no reduction in weight or volume, (PF > 1 = a concentration factor). The processing factor is the proportional amount by which residues change when food is processed. For this study, it is calculated by the formula below: processing factor (PF) =
concentration of pesticides in processed coffee beans (mg/kg) concentration of pesticides in raw coffee beans (mg/kg)
After obtaining the processing factor, the percent reduction (% reduction) for each processing method was calculated as follows: % reduction = (1 − PF) × 100. 2.7. Wash Water and Coffee Sludge Remaining at the Bottom of Pot after Coffee Brewing. Before the coffee was roasted, it was washed with water (which may remove some residues on the surface of the coffee beans). Additionally, after the coffee was brewed, there is sludge (powder or coffee residue) remaining in the bottom of the coffee pot, which may contain some pesticide residue that can evade the roasting process. Traditionally, this wash water and coffee sludge is often disposed of in the environment, which may contaminate food grown there. To check whether the wash water from the samples and the sludge of the coffee contain pesticide residues, these two samples were analyzed. The procedure was as follows: A 10 mL portion of the wash water (assuming 10 g of water) and 6−8 g of coffee sludge were weighed on an analytical balance and placed into a 50 mL centrifuge tube. C
DOI: 10.1021/acs.jafc.5b03327 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry Then the extraction, cleanup, and analysis were done in a way similar to that of the raw coffee bean. 2.8. Analytical equipment. Quantitative estimations and chromatographic separation of each pesticide were done by gas chromatography with an electron capture detector (GC−ECD, Agilent Technologies 6890N) with an auto sampler. The coffee bean processing was undertaken to see their effect on the pesticides previously detected in Ethiopian green coffee beans,7 so the chromatographic conditions for this study were the same as the previous paper, as explained below. A HP-5 capillary column of 30 m × 0.25 mm i.d. × 0.25 μm film thickness coated with 5% phenyl methyl siloxane (model no. Agilent 19091J-433) was used in combination with the following oven temperature program: initial temperature, 80 °C; ramp at 30 °C min−1 to 180 °C; ramp at 3 °C min−1 to 205 °C; held for 4 min; ramp at 20 °C min−1 to 290 °C; held for 8 min; ramp at 50 °C min−1 to 325 °C. For deltamethrin, the oven temperature was maintained initially at 130 °C, held for 1 min, ramped at 30 °C min−1 to 280 °C, held for 16 min, ramped at 50 °C min−1 to 325 °C, and held for 3 min. The total GC run time was 27.92 min. Helium (99.9999% purity) was used as a carrier gas at a flow rate of 20 mL min−1, and nitrogen was the makeup gas at a flow rate of 60 mL min−1. An aliquot of 1 μL was injected in split mode at a split ratio of 50:1 and injection temperature of 280 °C. The pesticide residues were detected with an electron capture detector (μ-ECD) operated at a temperature of 300 °C. For the reliability of the results, each sample was analyzed in triplicate, and the mean concentration was computed accordingly. 2.9. Statistical Analysis. All the treatments for the three processing methods were done in three replicates, and values are shown as mean ± SD. Statistical significance was checked using the Kruskal−Wallis test to see if the processing methods (washing, roasting, and brewing) are different in the percent reduction of the pesticides under study. Differences at P < 0.05 were considered as significant.
Table 2. Results of Regression Coefficient, Percent Recovery and Percent RSDa
a
pesticides
r2
% recovery
% RSD
endosulfan α endosulfan β chlorpyrfos ethyl permethrin cypermethrin deltamethrin p′p-DDE p′p-DDD o′p-DDT p′p-DDT heptachlorepoxide hexachlorobenzene
0.998 0.997 0.995 0.998 0.999 0.999 0.999 0.999 0.998 0.988 0.999 0.999
80 130 122.5 112.5 127.5 100 102.5 102.5 125 125 117.5 120
6.3 3.8 2.0 8.9 19.6 15.0 2.4 4.9 4.0 2.0 10.6 0.20
RSD = relative standard deviation.
Figure 1. Mean pesticide residues (mg/kg) at different coffee processing steps. The error bars indicate standard deviation. Pesticides from brewed coffee were not detected and are not shown in this graph.
3. RESULTS AND DISCUSSION 3.1. Quality Control Studies. The mean percent recovery for most of the pesticides was 80−120% and the percent relative standard deviation (% RSD) was below 10% for all the pesticides under study, except for cypermethrin and deltamethrin. All pesticides were in the acceptable analytical range (70− 120% recovery and RSD < 20%). The regression coefficients (r2) were >0.995 for all the pesticides. This method is accurate and precise for the analysis of the pesticides under study and fulfilled the requirement of European Document no. SANCO/ 12495/2011.18 The results are indicated in Table 2. 3.2. Effect of Household Processing of Coffee Beans on Pesticide Residues. This study investigated the effect of household processing of coffee (washing, roasting, and brewing) on the stability of 12 pesticides, including metabolites and isomers. The concentration of the pesticides under study in raw and processed coffee beans are presented in Figure 1. Among the household processes, washing decreased the pesticide residues by 14.63−57.69%. This effect may be due to the fact that most of the pesticides have contact action and are found on the surface of the coffee beans. This washing process may help for the removal of the pesticides with the silverskin of the coffee (very thin layer of a coffee bean located just above the main part of the coffee beans), dust, or soil. Reports indicated that surface residues are easily amenable to removal by simple washing processes with water.4 The majority
of the pesticides applied to agricultural crops are confined to the surface and a much smaller amount undergoes penetration to the plant system.19 As a result, they can easily be acquiescent to being washed, trimmed, or peeled from the surface of the crop. The maximum reduction was observed for endosulfan α and β, for which the residue decreased by 53.13−57.69%, respectively. A similar result was observed from a study done in India: washing of the vegetable brinjal with tap water decreased endosulfan residue up to 55%; however, a lesser removal of heptachlor epoxide in cucumber was observed.20 The rinsability of some pesticides by washing with water may not always correlate with its water solubility.21 As indicated in Table 1, most of the pesticides under study have low water solubility and high octanol-water partition coefficient (log kow > 3). Washing was found comparatively less effective in reducing the residues of p′p-DDD (14.63%), hexachlorobenzene (20.83%), heptachlor epoxide (23.40%), chlorpyrifos ethyl (26.53%), o′p-DDT ((28%), and p′p-DDT (28%). In this study, roasting was found to be effective in affecting the stability of pesticide residues in coffee beans. By this process, reductions of the residues of the pesticides were in the range of 72.5−99.8%. The maximum reductions were observed in hexachlorobenzene (99.8%), heptachlor epoxide (97.9%), p′p-DDE, p′p-DDD (97.6%), and o′p-DDT (96%). Five of the D
DOI: 10.1021/acs.jafc.5b03327 J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Article
Journal of Agricultural and Food Chemistry
roasting (p = 0.0001) and brewing processes (P = 0.00001); however, there is no significant difference between coffee roasting and coffee brewing (P > 0.05). According to Hasmukh et al.,25 the washing process may depend on different factors, such as location, age of residue, water solubility, and temperature; this may be the reason for the variation. From the household processing methods, roasting and brewing are the most effective in decreasing the pesticides in coffee beans. 3.4. Pesticide Residue in Wash Water and Coffee Sludge. To determine the level of pesticide liquid waste (wash water) and the coffee sludge that settled to the bottom of the coffee pot after the coffee was brewed, these two samples were analyzed. The result of the analyses revealed that all DDT metabolites (p′p-DDE, p′p-DDD, o′p-DDT, and p′p-DDT) and deltamethrin were detected in the wash water (Figure 2). This
spiked pesticides (chlorpyrifos ethyl, permethrin, cypermethrin, and endosulfan α and β) were not detected after the roasting process. This may be due to the instability of these pesticides to the heat applied during the roasting process. Thermal processing treatments such as cooking, backing, blanching, steaming, and boiling have been found effective for the breakdown of various pesticides, depending on the type of pesticide.22 Additionally, there may be a loss of the pesticides during the application of heat depending on their physicochemical properties, such as evaporation or thermal degradations.23 From a study done in Japan on the behavior of pesticides in coffee beans, the pesticide residues decreased a significant amount after the coffee beans were roasted.14 After the coffee brewing process, none of the pesticides under study were detected. This process is found to be the most effective for the breakdown of the pesticides under study. As it has been explained by Abou-Arab and Abou Donia,24 the elimination of the pesticide residue from the boiled extract (brewed coffee) may be due to the decomposition of the pesticide residue by the application of heat during brewing. The pesticide residues were significantly influenced by the roasting and brewing processes. In general, the three household coffee bean processing methods have a cumulative effect in the reduction of pesticides and help in solving health problems that can occur from exposure to pesticides in coffee. 3.3. Determination of the Processing Factor. The processing factor for each pesticide from each processing step was determined. The PFs for coffee samples after each processing step was
