Radiological hazard of coffee to humans : a comparative study of Arabian and Turkish coffees

Thirty-nine Arabian and Turkish coffee powder samples purchased from various markets in Saudi Arabia were analyzed by High Pure Germanium (HPGe) gamma spectrometry to determine the activity concentrations of the natural and artificial radionuclides 238 U, 226 Ra, 232 Th, 137 Cs, and 40 K. All samples, (except for two samples of Turkish coffee) were found to contain a high mean content of 40 K, ranging from 839.83 to 1197.11 Bq/kg and from 161.312 to 2411.215 Bq/kg for Arabian and Turkish coffee powders, respectively. The concentrations of 226 Ra and 232 Th were found to be 2.57 to 10.63 Bq/kg and nondetectable to 8.01 Bq/kg for Arabian coffee and nondetectable to 10.09 Bq/kg and nondetectable to 9.75 Bq/kg for Turkish coffee, respectively. Based on these values, we estimated the potential radiological hazards to consumer health from coffee powder. We determined the radium equivalent, annual effective dose rate, and external and internal hazard for each element, and all were found to be below the limit recommended by the United Nations Scientific Committee on the Effects of Atomic Radiation. However, absorbed dose rate values for some samples were higher than the permissible limit. In all samples, the 137 Cs concentration was below the detection limit. The average annual committed effective dose values reported in this study were far below the world average value of 0.30 mSv/yr for an individual. In addition, the limit for the threshold consumption rate was calculated. The statistical methods were applied to study the relationship between all the calculated natural radionuclides and their hazard parameters. Results indicated that the use of these types of coffee had no significant radiological health risks. This study may contribute data on coffee powder for formulating regulations related to radiological health care.


INTRODUCTION
Coffee is one of the most popular and widely consumed beverages in the world, and its consumption is increasing (Roselli et al., 2013).However, variation exists in the annual consumption between countries.In addition, coffee is grown in many countries, where the coffee trade has played a crucial role in their economic development (Roselli et al., 2013).Coffee comes from the plant genus Coffea, which has two primary species, C. arabica (Arabica) and C. canephora (also known as C. robusta, or robusta).Some naturally occurring radioisotopes and other elements present in soil are drawn into the roots of plants via ion channels or specific transporters (Sugiyama et al., 2009;Jibiri et al., 2016).Their distribution throughout the plant tissues depends on their chemical characteristics and several parameters of soil and the plants themselves (Awudu, 2012).Several studies have been performed to determine the concentrations of elements in various types of coffee and to estimate whether these concentrations contribute to toxicity (Szymczycha-Madeja et al., 2014;Welna et al., 2014;Jarošová et al., 2014;Zaidi et al., 2005).Jarošová et al. (2014) determined the toxic elements and mineral nutrients, including Pb, Ni, Mn, Cr, Cd, Zn, Mg, Fe, Cu, and Ca, in five types of coffee by using inductively coupled plasma mass spectrometry (ICP-OES) and atomic absorption spectrometry (AAS).No significant differences were found between the results obtained through the two methods, but a multivariate analysis helped to identify variations among samples collected from different locations.Zaidi et al. (2005) used instrumental neutron activation analysis to measure 20 trace elements (toxic, essential, and nonessential) in samples of coffee beans from four different origins and two instant coffee brands consumed in Pakistan.They next estimated the daily intake of trace elements and compared those levels with tolerance limits.The intake of toxic elements was well below the recommended safety limits, but the cumulative intake of Mn was four times higher than the acceptable amount.
Knowledge of the concentrations and distributions of natural radionuclides, such as 40 K, 238 U, and 232 Th and their decay products, in soils, plants, sediments, and so forth is useful for monitoring environmental radioactive contamination (El-Reefy et al., 2006).Use of chemical fertilizers, especially phosphorus fertilizer, may lead to increased contamination of agricultural crops by enriching the soil with radioactive elements as well as chemical toxins (Alharbi, 2013).Roselli et al. (2013)  137 Cs concentrations less than the detection limit (2.0 Bq/kg).
The aim of this study was to compare the concentration of naturally occurring radioactive materials (NORMs) in Arabian and Turkish coffee powder using gamma spectrometry.The imported samples were bought from selected local markets.Additional goals were to determine the radiological hazard associated with drinking coffee made from the various coffee powders and to estimate the average annual committed effective dose via ingestion of the radionuclides in the coffee and the threshold consumption rate.
The study focused on coffee powder to enrich the radiological information in Saudi Arabia in particular and the world in general because information on the concentration of NORMs and their presence in coffee is scarce.

MATERIALS AND METHODS
Thirty-nine coffee powder samples were purchased at different markets in Saudi Arabia.Each powdered was grinded and dissolved as homogenous solution and weighing about 200 to 250 g.The samples were then transferred to polyethylene 650-mL Marinelli type beakers (of known weight) were hermetically sealed with an insulating tape to impede contact with air moisture say labeled, and packed into radon-impermeable plastic containers to prevent radon gas escape as much as possible.Containers of the same size and geometry were used for reference materials to calibrate the system for measuring radioactivity.The samples and reference materials were stored and kept for a period of 1 month to attain secular radioactive equilibrium among 226 Ra, 232 Th, and their respective short-lived decay products 226 Ra and its decay products in the uranium series and 228 Ra and its decay products in the thorium series (Kurnaz et al., 2007;Samad et al., 2012).Finally, each Marinelli container was analyzed using a HPGe detector.

The radiometric measurement
Detection of the amount of natural radioactivity for 226 Ra, 238 U, 232 Th, and 40 K in coffee powder samples was carried out with gamma spectrometry using a high-purity germanium (HPGe) detector (P-type vertical coaxial, Canberra model GC4018).The detector had 25% relative efficiency and a resolution of 1.85 at 1332 keV of 60 Co gamma ray.The gamma acquisition and analysis were determined by using a multichannel analyzer (16,000 channels spectral memory) coupled to a computer using Genie-2000 Basic Spectroscopy Software.Quality-assured standard samples in QCY4 solution (obtained from the International Atomic Energy Agency) were used for the calibration and the absolute efficiency of the detector.The mixture of radionuclides in the solution (with corresponding energies) included 60 Co (1173 and 1333 keV), 88 Yt (898 and 1836 keV), 137 Cs (662 keV), 85 Sr (514 keV), 113 Sn (391.69 keV), 203 Hg (279 keV), 139 Ce (1656 keV), 57 Co (122 keV), 109 Cd (88 keV), and 241 Am (60 keV).Detailed information concerning calibration of the HPGe detector and the procedures followed can be found in previous study (Tufail et al., 2006;Fawzia, 2010).To measure the environmental gamma background, an empty identical Marinelli beaker was used.All samples and the background were counted for 36,000 s.After measurement, the activity concentrations were calculated by subtraction of the background values.The 137 Cs and 40 K concentrations were measured by determining their characteristic gamma lines of energies, 661.6 and 1460.8 keV, respectively.The 226 Ra activity concentration was determined by averaging the calculated activity concentrations of three photopeaks of 214 Bi (1764.5, 1120.3, and609.3 keV) and two of 214 Pb (295.1 and 352 keV).The activity of 232 Th was inferred from the weighted mean activities of the gamma peaks of 208 Tl (583.0 keV), 212 Pb (238.6 keV), and 228 Ac (911.2 and 338.4 keV).In addition, the activity of 238 U was given by the line of gamma of its product decay: 234 Th (63.22 and 92.78 keV) (Darko, 2015).The minimum detectable activity concentrations of 40 K, 238 U, 232 Th, 226 Ra, and 137 Cs using HPGe detector were 2.51, 1.0, 0.80, 0.75, and 0.85 Bq/kg, respectively.

Absorbed dose rate and annual effective dose rate equivalent from ingestion of coffee powder
According to United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR, 2010), the external outdoor absorbed gamma dose rate (D) from natural radionuclides in the samples under study (at 1 m above the ground level) can be calculated from the following equation: Where, ARa, ATh, and AK are the activity concentrations of radium, thorium, and potassium, respectively.The absorbed dose can be converted to the annual effective dose equivalent by using the conversion factor 0.7 Sv/Gy (UNSCEAR,2000) and the indoor occupancy factor of 0.8, based on people spending an average of 80% of their time indoors and 20% outdoors (Jibiri, 2016).The total annual effective dose (AE, the external dose rates) from the 226 Ra, 232 Th, and 40 K is obtained by using the following equation (Giri, 2013): 6 (mSv/yr) 24 365.25 0.8 0.7 10 The absorbed dose rate (D) and the annual effective dose (AE) for the coffee powder samples are presented in Table 4.

The radium equivalent
The radium equivalent activity for coffee powder is given by the following equation (Matiullah et al., 2004): Equation 3 compares the activity concentrations of coffee powder containing different amounts of 40 K, 232 Th, and 226 Ra.The permissible limit of this index is 370 Bq/kg as reported by UNSCEAR (2000).

The internal and external hazard indices
The internal and external exposures to gamma radiation in coffee powder were determined using the Equations 4 and 5, respectively (Fawzia, 2010):

Alharbi and Alamoudi 329
To safely consume the coffee powder, the upper limit for both the internal index (Hin) and the external index (Hex) should be less than unity as reported by ICRP (2007).

Average annual committed effective dose or dose ingested by the consumer
Estimation of the average annual committed effective dose (AACED) due to ingestion of NORMs in coffee can be calculated using the following equation (Njinga et al., 2015;Chandrashekara and Somashekarappa, 2016): where, Eave (Sv/yr) is the AACED, Ai is the activity concentration of each radionuclide i , Cr is the consumption rate of radionuclide, and DCFi is the standard dose conversion factor (2.8 × 10 −4 , 2.3 × 10 −4 , and 6.2 × 10 −6 mSv/Bq for 226 Ra, 232 Th, and 40 K, respectively, for an adult) (UNSCEAR, 2000).The annual threshold consumption rate (kg/yr) for each coffee sample was obtained using the following equation: Where, A1, A2, and A3 are the activity concentrations of 226 Ra, 232 Th, and 40 K, respectively DCF1, DCF2, and DCF3 are the dose coefficients for ingestion for the radionuclides (Sv/Bq) and Eave = 0.3 (mSv/yr) is the threshold AACED due of NORMs in the studied samples.

RESULTS AND DISCUSSION
The activity concentrations of Ra, and 232 Th in Arabian and Turkish coffee powders imported for Saudi markets from different countries were estimated, and the results are summarized in Tables 1 and 2  Turkish samples CT20 and CT21, all Arabian and Turkish coffee powders had 40 K concentrations that were higher than the acceptable value (412 Bq/kg) (UNSCEAR, 2010).Based on Figure 3, the mean activity concentrations of 40 K in Arabian and Turkish coffee powders are 1.53 and 2.69 times higher than the recommended limit, respectively.In addition, the 40 K concentration was 46.28% higher in the Turkish coffee than in the Arabian coffee.Moreover, the current results from Arabian and Turkish coffee powders showed a mean concentration of 40 K that was higher than that obtained by Roselli et al. (2013), although the mean concentration of 226 Ra was within the range found in the same study.This finding may be linked to contaminated soils in the farming areas (Fatima et al., 2008;Faanu et al., 2016).Uranium 238 was detectible in all study samples except samples CA10, CA11, CT2, CT15, CT20, CT21, and CT22.The highest 238 U concentrations were found in sample CA5 (135.11Bq/kg) and sample CT10 (57.16 Bq/kg), representing Arabian and Turkish coffee powders, respectively.
Hence, the mean activity concentration of 238 U in Arabian coffee was approximately 98.09% higher than the Turkish coffee.The activity concentration of 232 U ranged from ND to 66.33 Bq/kg with a mean value of 44.61 Bq/kg for Arabian coffee powder and from ND to 57.16 Bq/kg with a mean value of 22.52 Bq/kg.Figure 4 shows a comparison of 238 U in Arabian and Turkish coffee powders, in which the mean activity in Arabian samples is less than the permissible limit, but the mean activity in Turkish samples was 35.18% higher than the acceptable limit.The measurements show that all studied samples had concentrations of the artificially produced radionuclide 137 Cs below the detectable limit except in samples CA5 and CT19, which had the values of 3.44 and 1.26 Bq/kg, respectively.Tables 3 and 4 list the calculated values of the radium equivalent, absorbed dose rate, annual effective dose rate, and gamma hazard indices H ex and H in used in Equations 1 to 5, respectively.The results showed that the radium equivalent lay in the ranges of 77.382 to 114.254 Bq/kg (CA9 and CA5, respectively) for Arabian coffee powder with a mean value 90.595 Bq/kg and from 12.421 to 202.159 Bq/kg for Turkish coffee powder with a mean value of 124.586Bq/kg.These values are less than the permitted limit of 370 Bq/kg.The estimated values for the absorbed dose rate in Arabian and Turkish coffee powders varied from 40.703 to 59.665 nGy/h (CA9 and CA5, respectively) with a mean value of 48.655 nGy/h and from 6.727 to 107.791 nGy/h (CT20 and CT24, respectively) with a mean value of 66.709 nGy/h.According to UNSCEAR (2010), all results fell below the recommended limit for the average exposure rate (84 nGy/h).External and internal hazards were also determined based on Equations 4 and 5. Tables 5 and 6 show that the external and internal hazard indexed are less than unity for Arabian and Turkish coffee powders.With a unit consumption rate of 1 kg per annum being used, the AACED due to the ingestion of radionuclides 232 Th, 226 Ra, and 40 K from Arabian and Turkish coffee was estimated using Equation 6, and the results are presented in Tables 5 and 6.The AACED varied from 0.00232 to 0.00396 mSv/yr and from 0.00032 to 0.00598 mSv/yr for Arabian and Turkish coffee powders, respectively.The mean values of AACED in the study samples were far below the world AACED of 0.3 mSv/yr for ingestion of natural radionuclides detailed in UNSCEAR (2010).The threshold consumption rate (Table 5 and 6) is the amount at which the AACED value exceeds 0.3 mSv for any of the coffee powders.Consequently, Tables 5 and 6 provide baseline data indicating that an individual with a consumption rate below the threshold values would experience an insignificant radiological health risk, but someone with a consumption rate that is slightly higher than the threshold values faces significant radiological health risk.

Descriptive statistics
Basic descriptive statistics were used to characterize the central tendency and variation of the data.The distribution of radionuclides measured in Arabian and Turkish coffees and the radiological hazards using SPSS, version16.0for Windows (Sivakumar et al., 2014).7 and  8 summarize the statistical parameters such as minimum, maximum, arithmetic mean, standard deviation, skewness, and kurtosis.The asymmetry of the probability distribution for real values of random variables can be characterized by measuring the degree of skewness.A normal distribution has a skewness of zero.Therefore, positive (or negative) skewness indicates a distribution with an asymmetric tail extending towards values that are more positive (or negative) (Adam and Eltayeb, 2012).The radionuclides in Arabian and Turkish coffees (Tables 7 and 8 Kurtosisis determined if the distributions for data are peaked or flat relative to a normal distribution.Positive kurtosis indicates a relatively peaked distribution, while negative kurtosis indicates a relatively flat distribution.Higher kurtosis means that more of the variance is explained by infrequent extreme deviations, as opposed to frequent slight deviations (Raghu et al., 2015).According to Tables 7 and 8, the distributions associated with

Frequency distribution
To estimate the probability distribution of continuous variables the histograms are represented in Figure 8A to D) for Arabian coffee and in Figure 9A to D for Turkish coffee.The frequency distributions (according to the normality test for Shapiro-Wilk) for all activity concentrations in Arabian and Turkish coffees were analyzed (Tanaskovi et al., 2012).The graphs in Figure 8 show that the 40 K was distributed in non-normal distribution, while the 238 U, 232 Th, and 226 Ra were distributed in a normal (bell-shaped) distribution in Arabian coffee.All radionuclides in Turkish coffee were distributed in non-normal distribution except 238 U as shown in Figure 9.

Pearson's correlation analyses
Pearson's correlation analyses were performed to determine the interrelation between the natural radionuclides and the calculated radiological hazard parameters (Liu et al., 2003).The obtained correlation coefficients are presented in Tables 9 and 10 U in Turkish coffee.Moreover, strong correlations among most of the radiological hazard parameters were noted and each other.

Conclusion
In this study, measurement of radioactivity in Arabian and Turkish coffee powders sold in Saudi Arabia markets was  workers may face a greater health hazard because they experience external exposure to potentially large amounts of coffee powder and may thus accumulate large doses.
The annual effective ingestion dose from coffee should be added to the total global dose from food consumption.The data obtained from this research provides information on the activity concentration of natural radionuclides and increases the knowledge about natural radioactivity.The statistical analysis also confirms that these samples under the study of Arabian and Turkish coffees do not possess significant gamma radiation effects.Coffee quality must be strictly controlled, and the study of radionuclide concentration in this matrix has great significance.In terms of the coffee market, determining the radioactivity values of the commodity is important for customers.

Figure 1 .
Figure 1.A comparison between the mean activity concentration of 226 Ra in Arabian and Turkish coffee with the permissible limit.

Figure 2 .
Figure 2. A comparison between the mean activity concentration of 232 Th in Arabian and Turkish coffee with the permissible limit.

Figure 3 .
Figure 3.A comparison between the mean activity concentration of 40 K in Arabian and Turkish coffee with the permissible limit.

Figure 4 .
Figure 4.A comparison between the mean activity concentration of 238 U in Arabian and Turkish coffee with the permissible limit.

Figure 5 .
Figure 5.The AAECD as a function of various consumption rates within the range of 0 to 100 kg/year.

Figures
respect to sample code numbers in the Arabian and Turkish coffee.Tables ) have positive skewness values, indicating that the distributions are asymmetric, with the exceptions of 40 coffee have positive kurtosis values, indicating peaked distributions, and the others radionuclides have negative kurtosis values, indicating flat distributions.

Figure 6 .
Figure 6.The distributions of 226 Ra, 232 Th, 40 K, and 238 U with respect of sample code numbers in Arabian coffee.

Figure 7 .
Figure 7.The distributions of 226 Ra, 232 Th, 40 K, and 238 U with respect of sample code numbers in Turkish coffee.

igure 9
Frequency distributions of activity concentrations of 226 Ra, 232 Th, 40 K, And 238 U in 9 Frequency distributions of activity concentrations of 226 Ra, 232 Th, 40 K, And 238 U in igure 9 Frequency distributions of activity concentrations of 226 Ra, 232 Th, 40 K, And 238 U in igure 9 Frequency distributions of activity concentrations of 226 Ra, 232 Th, 40 K, And 238 U in

Figure 9 .
Figure 9. Frequency distributions of activity concentrations of 226 Ra, 232 Th, 40 K, And 238 U in Turkish coffee.

Table 1 .
, respectively.As shown in Table1, the activity concentrations of with a mean of 1507.18Bq/kg.The lowest activity was recorded for sample CT20, while the highest activity was recorded for sample CT24.With the exception of two Activity concentration (Bq/kg) in different types of Arabian coffee powder.

Table 2 .
Activity concentration (Bq/kg) in different types of Turkish coffee powder.

Table 3 .
Radium equivalent, absorbed dose rate, annual effective dose rate, external hazard index, and internal hazard index in differ ent types of Arabian coffee powder.

Table 4 .
Radium equivalent, absorbed dose rate, annual effective dose rate, external hazard index, and internal hazard index in differ ent types of Turkish coffee powder.

Table 5 .
AACED due to the ingestion of radionuclides 232 Th,226Ra, and 40 K from Arabian coffee and the threshold consumption rate.

Table 6 .
AACED due to the ingestion of radionuclides 232 Th,226Ra, and 40 K from Turkish coffee and the threshold consumption rate.

Table 7 .
Descriptive statistics of radionuclides and radiological hazard of Arabian coffee.

Table 8 .
Descriptive statistics of radionuclides and radiological hazard of Turkish coffee.

Table 9 .
Pearson correlation coefficients of radionuclides and radiological hazard of Arabian coffee.

Table 10 .
The Pearson correlation coefficients of radionuclides and radiological hazard of Turkish coffee.Correlation is significant at the 0.05 level (2-tailed) .**Correlation is significant at the 0.01 level (2-tailed). *