Pyrethroid and metal residues in different coffee bean preparing processes and their human health risk assessments via consumption

The study was conducted on 50 samples of coffee beans from various origins. The samples included green coffee beans, roasted beans, brew coffee drinks and coffee sludge. Three processes were used to prepare these samples: dried, semi-washed, and washed. Three synthetic pyrethroid insecticides and nine heavy metals were subsequently analyzed using modified Quick, Easy, Cheap, Effective, Rugged and Safe (QuEChERS) and acid digestion methods, respectively. The quantification of pyrethroids was performed by GC-μECD whereas those of metals were determined using flame atomic absorption spectrophotometer. According to the results, concentrations of both pyrethroids and heavy metals were predominantly found in green coffee beans except for Cr. Pyrethroid insecticides were not detectable in brew coffee drink and heavy metal concentrations were below the acceptable daily intake (ADI) level. Risk estimations for daily coffee intake using the health risk indices (HRIs) and target hazard quotients (THQs) of normal and the 97.5 percentile Thai consumers were less than 1. This indicated that the coffee drinks from studied samples could not cause potential health risk.


31
Coffee is a worldwide favorite beverage. It has several beneficial antioxidants and is 32 one among the richest chlorogenic acid sources (1). However, coffee drinking can potentially 33 lead human to expose to multiple toxicants. Those of which include pesticides, heavy metals, 34 organic solvents and pharmaceutical agents. These hazardous chemicals have abundantly been 35 used to protect coffee trees from pests and diseases during cultivation and manufacturing 36 processes (2). Even though coffee beans are roasted, pesticide residues are still present. Recent 37 studies revealed that coffee beans and coffee wastes were contaminated by numerous pesticides 38 and metals (3)(4)(5). 39 Applications of pesticides and insecticides to coffee crops are either directly to soil or The elements present in the roasted and ground coffee samples and their infusions can differ 77 (9, 13). In addition, brewing types and roasting conditions also affected the concentration of 78 elements in resulting infusions (14,15). de Queiroz, Azevedo (9) determined heavy metals 79 (Cd, Cr, Cu, Mn, Ni, Pb, Zn) in the roasted and ground coffee beans and brew. They reported 80 that for all the infusions, the metals evaluated were found in lower concentrations with respect 81 to the maximum permissible daily intake, except for Pb. Hence, it is important to determine 82 heavy metal concentrations in various forms of coffee (raw, roasted, brew). 83 To the best of our knowledge, no research study has been conducted for analyzing the 84 heavy metals and pyrethroid insecticides in green coffee beans, roasted coffee, and brew in 85 Thailand. Therefore, the main goals of this study were 1) to determine the residues of 86 pyrethroid insecticides (flumethrin, cypermethrin, and cyfluthrin) and heavy metals (Cd,Co,87 Cr, Cu, Fe, Mn, Ni, Pb, and Zn) in various forms of coffee (green, roasted, brew, sludge) 88 originated from green coffee beans prepared by 3 processes (dry, semi-washed, washed); 2) to 89 evaluate the human health risk from insecticide and metal contaminated coffee drink using the 90 estimate daily intake/EDI, health risk index/HRI, and target hazard quotients/THQs. The 91 results of this study would provide useful baseline data on the levels of contaminants and 92 human health risks through coffee consumption.
6 126 drink were added with 10 mL acetonitrile, and hand shaken for 1 min. The samples were then added 127 with 4 g MgSO 4 and 1.5 g C 2 H 3 NaO 2 , and shaken for 1 min. They were subsequently centrifuged 128 at 5000 rpm for 5 min. A 2-mL upper organic layer was taken and translocated to a d-SPE tube 129 containing 600 mg MgSO 4 , 300 mg PSA and 100 mg alumina. The extract was shaken for 1 min and 130 centrifuged. After centrifugation, 200 µL of upper layer was taken and placed into a vial for the 131 analysis of pyrethroid residues using gas chromatography equipped with a micro electron capture 132 detector (GC-µECD). 133 Roasted coffee powder and coffee sludge: 2.5 g fine roasted coffee powder and 2.5 g 134 coffee sludge were added with 10 ml acetonitrile and shaken for 30 sec. Then 4 g of MgSO 4 , 135 1 g of NaCl, 1 g of C 6 H 9 Na 3 O 9 and 0.5 g C 6 H 5 Na 3 O 7 were added and shaken for 30 sec.

136
Samples were later centrifuged at 5000 rpm for 5 min. Two mL upper organic layer was taken 137 and placed into d-SPE tube containing 600 mg magnesium sulfate anhydrous, 100 mg PSA and 138 100 mg alumina then shaken for 1 min and centrifuged thereafter at 5000 rpm for 5 min. After  The acid digestion method was used to extract 9 metals: i.e., Cd, Co, Cr, Cu, Fe, Mn, 7 151 FAAS analysis.

152
Brew coffee drink: 25 mL of each sample was digested with 5 mL 70% HNO 3 at 180°C for 153 at least 3 h. The digested samples were cooled to room temperature (25°C) and then filtered through 154 No. 4 Whatman filter paper. The solution volume was adjusted to 25 mL using ultra-high purity 155 water and kept at 4°C until FAAS analysis.

Sample analysis
157 Gas chromatographic analysis 158 Pyrethroid residues analyses were carried out using an Agilent gas chromatography 159 (GC 7890B, Agilent Technologies, Germany) with a micro-electron capture detector (µ-160 ECD). The system was equipped with a 30 m x 0.32 mm x 0.25 µm fused silica capillary 161 column (HP-5, Agilent Technologies, Germany). High purity (99.999%) helium and nitrogen 162 were used as carrier gas and make up gas at a constant flow rate of 20 mL/min and 60 163 mL/min, respectively. The GC oven was operated as follows: initial temperature at 150°C 164 held for 1 min, followed by the ramp of 15°C/min to 255°C and then 20°C/min to 300°C and 165 held for 5 min. The injector and detector temperatures were set at 260°C and 315°C, 166 respectively. The total run time was 28 min per sample. One microliter of each sample was 167 injected to the GC system under splitless mode. Pyrethroid concentrations were quantified 168 using standard calibration curves. Each sample was analyzed in triplicates to ensure reliable 169 results.

170
The linearity, percentage recovery, precision, limit of detection (LOD) and limit of concentration. The precision was obtained by spiking blank sample with each pyrethroid at 6 177 concentrations across the work range and express the variation in terms of relative standard 178 deviation (%RSD). The LOD and LOQ were determined by comparing the target signal to 179 noise ratio in blank sample spiked with known pyrethroid concentrations. For the LOD and 180 LOQ the minimum signal to noise ratio was 3 and 10, respectively. The analytical 181 performances were provided in Table 2.  The linearity (r 2 > 0.999) of calibration curve was evaluated using 5 replicates of 6 190 metal concentrations ranging from 0.05 -2.00 g/g for Cd, 0.05 -5.00 g/g for Co and Cu, 191 0.10 -10.00 g/g for Cr, 0.05 -4.50 g/g for Fe, 0.25 -4.00 g/g for Mn, 0.50 -10.00 g/g 192 for Ni, 0.50 -20.00 g/g for Pb, and 0.10 -2.00 g/g for Zn. The accuracy of metal analysis 193 was assessed using known spiked blank samples and the precision was also assessed by 194 analyzing %RSD. The LOD and LOQ of each metal were calculated using signal-to-noise 195 ratio (S/N) of 3 and 10, respectively. The obtained values were presented in Table 2. 196 Determination of processing factor 197 The processing factor (PF) for each transformation step (roasting, brewing and sludge) 198 was calculated as the ratio of pyrethroid insecticide or metal in processed samples (roasted 199 coffee powder, brew coffee drink, coffee sludge) (µg/kg) to those in non-processed samples 200 (green coffee bean) (µg/kg) using the following equation: Processing factor (PF) = Level of pyrethroid insecticide or metal in processed sample (μg/kg) Level of pyrethroid insecticide or metal in unprocessed sample (μg/kg)

202
The PF of < 1 (reduction factor) indicated that there was a sample reduction of 203 pyrethroid insecticide or metal by the processing step, whereas PF > 1 (concentration factor) 204 indicated that there was no reduction in those toxic residues (16). In addition, the percentage 205 reduction (% reduction) for individual processing step was calculated using the following The estimate daily intake/EDI, health risk index/HRI (the ratio of calculated EDI to 210 ADI or RfD) and target hazard quotients (THQs) were used to evaluate human health risk from 211 insecticide-and metal-contaminated coffee drink. Risk determination was also calculated for 212 the average consumption of brew coffee drink and at the 97.5 th percentiles of coffee drinker 213 (extreme consumer). These allowed more comprehensive evaluation of human health risk 214 associated with the consumption of pyrethroid and metal residues in coffee drink.

240
Pyrethroid residues in coffee samples 241 Residues of pyrethroid insecticides in green coffee beans, roasted coffee powder, brew 242 coffee drink and coffee sludge are shown in Table 3. All studied pyrethorids were 243 predominantly found in green coffee beans from various processes (dried, semi-washed, and 244 washed) but not in brew coffee drink. The residues in green coffee beans differed significantly 245 (p < 0.05) among processing steps (Table 3). Cyfluthrin 2, 3, cypermethrin CisA, and 246 cyfluthrin concentrations in dried processed coffee beans were the highest (p < 0.05). In 247 addition, concentrations of flumethrin and cypermethrin TransD found in semi-washed 248 processed coffee beans were significantly the highest (p < 0.05). Simple washing process was proven to easily remove these residues (19). Pesticides and insecticides were removed from 250 the outer and silver skins of the beans following washed and semi-washed processes (3).

251
However, this residue removal may not always correlate to the water solubility of each residue 252 compound. The peeling and refrigeration storage may also affect residue reduction (20).

253
Concentrations of all studied pyrethroids in brew coffee drink (Table 3) were lower 254 than their corresponding limits of detection ( Table 2). The roasting and brew coffee steps among coffee bean processes in roasted coffee beans and brew coffee drinks.

260
Most pyrethroid residue levels and their detection percentages in coffee sludge samples 261 were lower than those found in roasted coffee beans (Table 3). Cyfluthrin 1 concentration 262 found in coffee sludge sample from semi-washed beans was the highest (0.16 ± 0.02 µg/g).

263
However, all pyrethroid residue concentrations in coffee sludge did not significantly differ 264 among coffee bean processes (p > 0.05).

265
Heavy metal residues in coffee samples 266 Most of the studied 9 metals were found in green coffee beans, roasted coffee powder, 267 brew coffee drink, and coffee sludge except for Cr in green coffee beans and Cd in brew coffee 268 drink (only in semi-washed process samples) (Table 3). In addition, the lowest and the highest 269 metal concentrations were in brew coffee drink and green coffee beans, respectively. The 270 significant differences (p > 0.05) of Cd and Fe were found among the brew coffee drink.

271
The highest trace and toxic metal concentrations in green coffee beans were Cu ( Nędzarek, Tórz (30) but slightly higher than those found in others. These differences were 305 caused from a wide range of geographical areas in coffee culture (12,24,27 for such purposes should be of concern. The regulation of metal residues in coffee sludge 313 should then be established.

314
Determination of processing factors 315 The processing factor (PF) and percentage reduction (%) for individual pesticide and 316 metal from each process were determined (Table 4). No significant difference (p > 0.05) of PF 317 was found among the bean processes. All PFs of pyrethroids from all coffee bean processes 318 were less than 1. This indicated that coffee roasting and brewing decreased pyrethroid residues.

319
The results are in agreement with other studies whose PFs were decreased following food

30.
Nędzarek The rip beans were removed after sundrying coffee fruits. The hybrid method combining washed and dried processes. The rip beans were removed after coffee fruit fermentation.  Table 3 Average ± SE (µg/g or µg/mL) and percentage detection (