Measurement of Natural and Artificial Radioactivity in Infant Powdered Milk and Estimation of the Corresponding Annual Effective Dose

Abstract

Infant powdered milk (i.e., infant formula) is a rich and convenient source of nutrients, substituting for human breast milk in many countries. Moreover, it is the basic foodstuff for the infants because of its mineral and protein content, which is essential for growth and development. However, there are still limited data on radioactivity levels in infant powdered milk around the world, including Malaysia, for radiological health risk assessment. Thus, it is important to assess the radioactivity levels and the associated dose in the widely consumed powered infant milk. As a result, activity concentrations of ²²⁶Ra, ²³²Th, ⁴⁰K, and ¹³⁷Cs were determined in 14 brands of powdered milk consumed by Malaysian infants, which are imported from various regions around the world. Mean activities of ²²⁶Ra, ²³²Th, ⁴⁰K, and ¹³⁷Cs were 3.05 ± 1.84, 2.55 ± 2.48, 99.1 ± 69.5, and 0.27 ± 0.19 Bq/kg, respectively. Among the analyzed milk samples, the brand from Philippines (Lactogen) showed low level of radioactivity, while a Singaporean brand (S26 SMA Gold) showed the highest. The artificial radionuclide, ¹³⁷Cs, is virtually not detected in most of the brands investigated. Estimated mean annual effective doses due to consumption of powdered milk were 635.13 and 111.45 μSv/year for infants ≤1 year and infants 1–2 years old, respectively. The obtained dose value is significantly higher (in case of infants ≤1 year old) and lower (in case of infants 1–2 years old) compared to the UNSCEAR reported value (290 μSv/year) for the general population. In general, values are lower than the FAO/WHO and ICRP recommended limit of 1.0 mSv/year for all ages.

Introduction

Human exposures to radiation are mainly due to natural sources (up to 87%), whereas anthropogenic contribution is generally limited or lower (UNSCEAR, 2008). Specifically, naturally occurring radioelements such as uranium, thorium, and their progenies and potassium form the largest contributions to internal radiation dose received by humans due to their widespread distribution in the environment, whereas contributions from anthropogenic radionuclides are minimal. In most cases, anthropogenic radionuclides are localized to a particular area through routine and/or accidental releases and unplanned disposal of industrial and/or radioactive wastes etc., but finally find their way into the environment and foodstuffs (Al-Mastri et al., 2004; Ademola and Ehiedu, 2010). Besides this, radiation dose to different populations could vary in some order of magnitudes due to the geological and geographical variability, differential bioaccumulation of radionuclides by organisms, contamination of food through processing and preparation techniques, and feeding habits by a particular population (both man and animals).

Irrespective of natural or anthropogenic origin, internal exposure to humans is majorly through ingestion and, to some extent, through inhalation of radionuclides (UNSCEAR, 2008). Once ingested, radionuclides, such as ²²⁶Ra, ²³²Th, and their progenies, are easily assimilated into the soft body tissues, resulting into complex health-threatening issues, which could be for the entire lifetime. Moreover, long-term exposure to low-intensity radiation, particularly due to ²²²Rn and its short-lived alpha-emitting decay products, ²¹⁸Po and ²¹⁴Po, may cause lung cancer (William et al., 2000). Thyroid cancer is a major consequence of radionuclide ingestion (mostly by the artificial ¹³¹I) and, in acute cases, it can cause some changes in genes or disorders in reproductive function, heart problems, etc. Note that due to the short lifetime ¹³¹I (T_1/2 = 8.0252 days), radionuclide decay before detailed contamination surveys could be performed. As a result, ¹³⁷Cs deposition density is the primary measure to determine the accidental contamination of a specific area (Stepanenko et al., 2004). Thus, the assessment of radioactivity in foodstuffs is essential for adequate monitoring of the radiation levels to which humans are exposed.

Milk is an important ingredient used to make cookies, ice cream, yogurt, chocolate, powdered chocolate, and many other foodstuffs. As an important component of the food chain and a major source of nutrients in the daily human diet, milk is consumed in a significant amount, particularly by infants <1 year old and children, since its minerals and proteins are essential for their growth and development. Apart from this, infant formula is mostly produced from purified cow's milk and other rich protein sources. Infant formulas are liquids or reconstituted powders fed to infants and young children and serve as substitutes for breast milk and play a significant role in baby diets. Since the first year of life is a very sensitive period in the development of human nervous, reproductive, digestive, respiratory, and immune systems, detail study on the composition of baby foods (especially concentration of radionuclides) is crucial to ensure the safety and suitability of these foods (Pandelova et al., 2012). Although breast feeding is the best feeding choice during infancy (WHO, 2008), but due to many reasons and the possibility of certain body toxins being transferred to the infants through human milk, an increasing number of mothers are limiting the breast feeding to their infants and replacing it with industrially processed infant formula (Cattaneo, 2004). As a result, some infant populations may be more at risk when consuming contaminated milk since industrially processed infant formula may represent a source of intake of certain contaminants, including radionuclides, leading to prolonged health effects. It has been reported that children are more susceptible to exposure (Tripathi et al., 2001) because of their greater intestinal absorption than adults and a lower threshold for adverse effects (Cambra and Alonso, 1995). These pollutants may arise from the raw materials, poor quality of production process, subsequent pooling and the storage of milk, adulteration of infant foods (Fein and Falci, 1999), and through the soil–grass–cow milk pathway, especially when cows have fed on hay and silage in higher natural radioactivity areas. In the recent time, several studies have been conducted on milk and milk products consumed in different countries for adequate radiological assessment to forestall unnecessary radiation exposure to humans, particularly the infant population above the limiting index (Osibote et al., 1999; Navarrete et al., 2007; Afshari et al., 2009; Ababneh et al., 2010; Shanthi et al., 2010; Giri et al., 2011; Al-Zahrani, 2012; Alamoudi, 2013). In Malaysia, no surveys of radiation dose through the consumption of infant formula have been carried out and no baseline data of naturally occurring and anthropogenic radionuclides have been reported, considering the fact that over 90% of powdered milk consumed by Malaysian infants (<2 years old) is just being packed locally, but their origins are either Australia or New Zealand. As a result, the influence of environmental factors of the country of origin and the raw materials involved in the processing of the infant formula, as well as the conditions under which it is produced, cannot be overemphasized in radioactivity contents of the infant formula. It is therefore important to have an accurate knowledge of radioactivity content in the varieties of powdered infant milk consumed in the country since distribution of natural radioactivity is completely heterogeneous with respect to various regions of the world and the human activities in respective regions. In addition, milk is also considered to be a sensitive indicator of fallout contamination, which is a result of deposition of radioactive particles on the grass, soil, and earth surface following nuclear weapon detonation or nuclear accidents (such as Chernobyl and Fukushima nuclear reactor accidents). This is because ¹³⁷Cs (or ⁹⁰Sr) is easily transferred to the milk through the grass–cow–milk pathway due to its relatively high mobility in the soil–plant ecosystem (Nisbet and Woodman, 2000). Therefore, the activity concentrations of some long-lived natural radionuclides (²²⁶Ra, ²³²Th, and ⁴⁰K) and artificial radionuclide, ¹³⁷Cs, have been measured and the annual effective dose due to the intake of radionuclides is estimated for radiological risk assessment to the Malaysian infant population. Note that other studies such as Giri et al. (2011) have successfully applied such data to generate risk estimates. The values reported in this study will serve as reference data for natural and artificial radioactivity in the widely consumed powdered infant milk in Malaysia.

Materials and Methods

In this study, 14 brands of powered infant milk imported from different countries and also produced in Malaysia were collected from major supermarkets within Selangor, Malaysia (Table 1). Sample selections were made based on the information gathered from the sales departments of those supermarkets (i.e., higher selling rate) and assumed that these are the widely consumed powdered infant milk among the available brands in Malaysia. It should be noted that this work represents a cross-sectional survey, but not a generalizable study due to the involvement of limited samples. The collected samples were weighed and packed (without previous treatment) in their natural state, or in the form they are usually imported, into Marinelli beakers of 500 mL sizes (after washing and rinsing the beakers with diluted H₂SO₄ to avoid contaminations) with masses varying between 23.5 and 160 g. A total of 28 samples were prepared from the collected brands, two samples per brand. The samples were then stored for at least 6 weeks in a shielded area before analysis to ensure attainment of secular equilibrium between the parent nuclides and their respective short-lived progenies. The radioactivities of the sampled milk were determined using γ-ray spectrometry by means of a p-type coaxial HPGe γ-ray detector having 28.2% relative efficiency and an energy resolution of 1.67 keV−Full width at half maximum at the 1332.5 keV peak of ⁶⁰Co (Khandaker et al., 2015). The γ-rays emitted from the samples were analyzed by Gamma Vision 5.0 software (EG&G Ortec). Cylindrical multinuclide γ-ray sources being traceable to the Isotope Products Laboratories, containing ²⁴¹Am, ¹⁰⁹Cd, ⁵⁷Co, ²⁰³Hg, ¹¹³Sn, ⁸⁵Sr, ¹³⁷Cs, ⁸⁸Y, and ⁶⁰Co emitting γ-rays with energy ranging between 59.541 and 1836.063 keV (having homogeneously distributed activity with the same volume and shape as the Marinelli beakers containing the investigated samples so that the detection geometry remained the same), were used for detector energy calibration and the absolute photopeak efficiency evaluation. The calibration of the detector was done to take into account self-absorption. A detailed description of the measurement system and calibrations of the detector have been explained in Amin et al. (2013). Meanwhile, the detector system has a cylindrical lead shielding arrangement (consisting of 10-cm-thick lead internally lined with cadmium–copper coating), having a fixed bottom and a movable cover surrounding the detector to reduce background radiation. The energy calibration of the detector system was regularly repeated at an interval where a significant shift in the energy is noticed. Since ²²⁶Ra and ²³²Th emit very low-intensity gamma rays, which are not suitable to be detected through the conventional γ-ray spectrometry, an indirect approach was adopted to determine these nuclides. This approach involved the detection of their short-lived progenies through their characteristic γ-lines that are intense enough to be acquired by the conventional gamma-ray spectrometry. Hence, activity (Bq/kg) of ²²⁶Ra was determined using the characteristic γ-lines released by its progenies ²¹⁴Pb (E_γ = 351.93 keV, I_γ = 35.60%) and ²¹⁴Bi (E_γ = 609.32 keV, I_γ = 45.49%); ²³²Th by its progenies ²¹²Pb (E_γ = 238.632 keV, I_γ = 43.6%), ²⁰⁸Tl (E_γ = 583.187 keV, I_γ = 85.0%), and ²²⁸Ac (E_γ = 911.204 keV, I_γ = 25.8%), while ⁴⁰K and ¹³⁷Cs were determined using their respective single γ-lines of 1460.822 keV (I_γ = 10.66%) and 661.657 keV (I_γ = 85.10%), respectively. Measurement of gamma emission as the analytical technique to determine the concentrations of radionuclides can be justified as follows: ¹³⁷Cs decays to ¹³⁷Ba by emitting beta particle (E_β− = 187.1 keV; I_β− = 100%), followed by the emission of intense gamma line (E_γ = 661.657 keV, I_γ = 85.10%). Since the gamma ray is an energetic form of radiation (i.e., massless), its penetration power is much higher than the counterpart radiation of alpha and beta particles. Due to this property, most of the gamma rays can easily come out from the sample matrix, which is not the same case for all alpha and beta particles. Thus, accurate determination of samples' radioactivity is highly probable through gamma-ray spectrometry compared with its counterpart of alpha or beta spectrometry. Consequently, each sample was placed on the detector and counted for a period of 24 h ( = 86,400 s) to ensure good counting statistics. The background spectra for the same counting period were collected for each week of the counting and the background counts were subtracted from the region of interest. The activity concentration of the radionuclides, expressed in Bq/kg, in milk samples was calculated using the Equation (1) taken from Khandaker et al. (2013). \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}A = \frac { N \times 1000 } { \varepsilon _ { \gamma } \times I_ { \gamma } \times W_S } \tag { 1 } \end{align*} \end{document}

Table 1.

Identity of Analyzed Powdered Infant Milk, Its Characteristics, and Assessed Radioactivity Concentrations

Origin	Brand name	Source	Age group	Production date (dd/mm/yy)	²²⁶Ra (Bq/kg)	²³²Th (Bq/kg)	⁴⁰K (Bq/kg)	¹³⁷Cs (Bq/kg)
New Zealand	Anmum	Cow milk	0–6 months	27/05/2013	2.49 ± 0.23	0.31 ± 0.10	51.6 ± 2.77	ND
	Bebelac	Cow milk	0–12 months	20/01/2014	2.47 ± 0.25	1.33 ± 0.30	62.6 ± 3.38	0.73 ± 0.47
	Mamex	Cow milk	0–12 months	28/05/2013	1.89 ± 0.24	1.37 ± 0.47	66.9 ± 3.56	0.63 ± 0.46
	Karihome	Goat milk	0–6 months	29/10/2013	2.47 ± 0.24	1.16 ± 0.30	78.2 ± 4.05	0.41 ± 0.38
Mean ± SD		—	—	—	2.33 ± 0.29	1.04 ± 0.50	65 ± 11	0.59 ± 0.16
Malaysia	Dulac	Cow milk	0–12 months	11/12/2013	3.12 ± 0.36	1.83 ± 0.85	60.5 ± 4.56	ND
	Dutch Baby	Cow milk	0–12 months	02/02/2014	1.94 ± 0.22	2.23 ± 0.78	53.7 ± 2.84	ND
Mean ± SD		—	—	—	2.53 ± 0.83	2.03 ± 0.28	57.1 ± 4.81	ND
Thailand	Enfalac A⁺	Cow milk	0–12 months	15/04/2014	2.37 ± 0.38	1.83 ± 0.87	94.3 ± 4.70	0.16 ± 0.11
	Sustagen	Cow milk	1–3 years	31/02/2014	4.95 ± 0.69	2.42 ± 0.86	254 ± 12.8	0.23 ± 0.18
Mean ± SD		—	—	—	3.66 ± 1.82	2.13 ± 0.42	174 ± 113	0.20 ± 0.05
The Netherlands	Frisolac	Cow milk	0–12 months	01/09/2013	1.68 ± 0.22	2.10 ± 0.75	66.5 ± 3.51	0.28 ± 0.20
Philippines	Lactogen	Cow milk	0–12 months	16/11/2013	1.36 ± 0.14	0.55 ± 0.27	40.3 ± 2.10	ND
	Nan Pro	Cow milk	0–12 months	19/09/2013	2.17 ± 0.22	2.00 ± 0.81	48.7 ± 2.62	ND
Mean ± SD		—	—	—	1.77 ± 0.57	1.23 ± 1.03	44.5 ± 5.94	ND
Australia	Neo Baby	Cow milk	0–9 months	07/09/2013	1.51 ± 0.28	1.04 ± 0.54	47 ± 2.9	0.11 ± 0.09
Spain	Similac	Cow milk	0–12 months	10/11/2012	3.86 ± 0.44	2.22 ± 0.38	104 ± 5.3	0.17 ± 0.08
Singapore	S26 SMA Gold	Cow milk (milk and soy)	0–12 months	15/06/2013	7.06 ± 0.84	8.57 ± 2.11	235 ± 12.3	ND
Mean ± SD	All brands	—	—	—	3.05 ± 1.84	2.55 ± 2.48	99.1 ± 69.5	0.27 ± 0.19

ND, not detected; SD, standard deviation.

where N is the net counts per second under corresponding photopeak, ɛ_γ is the detection efficiency, I_γ represents the intensity of γ-ray, and W_s is the sample weight in gram. Meanwhile, the detection efficiency used in activity determination was fitted by the polynomial fitting function previously described in Khandaker et al. (2013). The final activity concentration was obtained using a weighted mean approach by combining the different values obtained for the individual γ-ray energy. By this, the uncertainty in the derived values is considered to be reduced compared with the use of a single transition.

Statistical analyses of the obtained results were performed using the TIBCO Spotfire S+ software package to find the existing relationship between the investigated radionuclides. The data of activity concentrations of the investigated radionuclides in the powdered infant milk are presented as mean ± standard deviation.

The estimation of uncertainties was similar to our previous report (Amin et al., 2013). However, it estimates the uncertainties in this measurement and we considered four independent uncertainties, which were added in quadrature. These are statistical uncertainties of the γ-ray counting having the relative standard deviation (RSD) range of 0.5%–10%, uncertainties in the efficiency calibration of the detector (RSD = ∼4%), uncertainties in sample weight (RSD = ∼1.5%), and uncertainty in gamma-ray intensity (RSD = ∼1%).

Determination of minimum detectable activity (MDA) of the measurement system was done following the procedure laid out elsewhere (Xhixha et al., 2013; Olatunji et al., 2014) as shown below: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}MDA = \frac { 2.71 + 4.65 \sqrt { B } } { \varepsilon _ { \gamma } \times I _ { \gamma } \times T \times W_S } \ ( { \rm Bq } / { \rm kg } ) \tag { 2 } \end{align*} \end{document}

where B is the background count for the region of interest of a certain radionuclide, and ɛ_γ, I_γ, and W_s retain their usual meanings as defined in Equation (1). Any activity concentration below MDA (0.103, 0.363, 0.033, 0.064, 0.097, 1.43, and 0.024 Bq/kg for ²¹⁴Pb, ²¹⁴Bi, ²¹²Pb, ²⁰⁸Tl, ²²⁸Ac, ⁴⁰K, and ¹³⁷Cs, respectively) is considered below detection limit.

Results and Discussion

Measured activity concentrations of ²²⁶Ra, ²³²Th, ⁴⁰K, and ¹³⁷Cs found in the powdered infant milk are presented in Table 1. In all the sampled brands, the mean activity concentration of ²²⁶Ra in the infant milk with respect to the country of origin shows a range of values with the lowest being from Philippines (Lactogen) (1.36 ± 0.14 Bq/kg) and the highest from Singapore (SMA Gold) (7.06 ± 0.84 Bq/kg) and a mean of 3.05 ± 1.84 Bq/kg (all brands). Since the passage of radionuclides to organisms is through the food chain (Licata et al., 2004), the relatively high concentration of ²²⁶Ra obtained in the Singaporean milk as well as in Thailand, Spain, and Malaysia could be attributed, probably in part, due to the radium content of the environments, which is being transferred to the milk through the grass–cow–milk pathway. The radium-226 content in soils of these areas has been reported in several environmental studies: Saat et al. (2010) reported a range of 61.92–141.68 Bq/kg in Malaysia and similar values are found in the review work of Ahmad et al. (2015) in Malaysian soils, 8–310 Bq/kg in Spanish soils (Quîndós et al., 1994), and 55.3–65.2 Bq/kg in Thai soils (Santawamaitre et al., 2011), but no such data are available in the literature about Singaporean soils based on our knowledge. In any case, it is worth mentioning that one of the sources of Singaporean milk/formula is soy; hence, the production of soy and processing to infant formula may enhance the activity level. Note that in the global marketplace, many industrial facilities import materials from different supply chains around the world. It is very possible that the final manufacturing step in making these products includes raw materials from other regions, and thus, the detected activity of the investigated infant formula might be the result of raw materials imported from elsewhere in the world. However, the variation in the radium content in the sampled powdered infant formula is not significant from one brand to another. The obtained mean activity of radium-226 has been compared with the available literature data (Table 2). It shows that the mean activity of ²²⁶Ra in powdered infant milk available to the Malaysian infant population is higher than in Slovenia (Štrok and Smodiš, 2011), Iran (Hosseini et al., 2006), Poland (Pietrzak-Flis et al., 1997), India (Shanthi et al., 2010), Jordan (Ababneh et al., 2010), and Saudi Arabia (Al-Zahrani, 2012) with the exception of 9.64 Bq/kg reported by Alamoudi (2013), but significantly lower than the reported data from Nigeria (Osibote et al., 1999).

Table 2.

Comparison of Mean/Range of Activity Concentration (Bq/kg) of ²²⁶ Ra, ²³² Th, ⁴⁰ K, and ¹³⁷ Cs with Literature Values

Country	²²⁶Ra	²³²Th	⁴⁰K	¹³⁷Cs	References
Malaysia	3.05 ± 1.84 (1.36–7.06)	2.55 (0.31–8.57)	99.1 (40.3–254)	—	Present study
Saudi Arabia	9.64	6.77	74.5	—	Alamoudi (2013)
Saudi Arabia	0.25–0.85	0.09–0.76	210–257	—	Al-Zahrani (2012)
Jordan	BDL, 2.14	BDL, 1.28	296.8–392.9	ND, 1.5	Ababneh et al. (2010)
Israel	—	ND, 0.8	54.2–472.3	ND, 0.08	Lavi et al. (2006)
Iran	0.05–0.186	0.094–0.166	434.1–610	—	Hosseini et al. (2006)
Indian	2.5 ± 1.2	1.48 ± 0.4	34.35 ± 5.2	—	Shanthi et al. (2010)
Iran	—	—	17.1, 31	—	Afshari et al. (2009)
Brazil	—	1.6–3.6, 1.7–3.7	475–489	5.1–7.3, 7–11.2	Melquiades and Appoloni (2001, 2002)
Slovenia	0.041–0.110	—	—	—	Štrok and Smodiš (2011)
Nigeria	23.07 ± 7.75	4.35 ± 2.06	831.66–54.83	ND	Osibote et al. (1999)
Syria	—	—	129–435	—	Al-Mastri et al. (2004)
IAEA-152 milk powder	—	—	540 ± 40	2.21E + 03	Altzitzoglou and Bohnstedt (2008)
New Zealand	—	—	—	0.45–0.76	Hermanspahn (2011)

BDL, below detection limit; ND, not detected; —, not reported.

With regard to the ²³²Th (an indicator for ²²⁸Ra) activity in the sampled infant milk brands, the activity ranged between 0.31 ± 0.10 Bq/kg (Anmum) and 8.57 ± 2.11 Bq/kg (SMA Gold) with an average value of 2.55 ± 2.48 Bq/kg (all brands). Again, the imported infant milk brand from Singapore shows the highest value of ²³²Th activity, which supports the earlier fact that the raw materials involved in processing the infant formula are normally obtained from different supply chains around the world. However, this is limited, in that the global economy necessitates the use of materials imported from all around the world. Meanwhile, the nature of the soil–plant ecosystem and particular food (e.g., silage, hay) intake by different organisms/animals (cow in this case) may influence the radioactivity level in the infant formula/milk (Pietrzak-Flis et al., 1997; Giri et al., 2011; Štrok and Smodiš, 2011). Apart from this, the method of food processing and preparation to some extent contributes to the radionuclide redistributions (Jibiri et al., 2007); this could be another reason for the level of radioactivity in this brand of infant milk under consideration. However, the present study does not investigate the method of milk processing and preparation, but could be considered for the future research. Meanwhile, several reports in the past have attributed infant deaths and illnesses to some unhealthy manufacturing processes of infant formula (Solomon, 1981; Mount, 1985; Lonnerdal and Hernell, 1998; WHO, 2003; Abbott Press Release, 2010). Thus, it can be assumed that the reported incident was not due to radioactive compounds, but due to chemical contamination. With an exception to SMA Gold brand from Singapore, the obtained data from other brands show similar low values of ²³²Th activity, which could be attributed to the poor mobility of this radionuclide within the environmental compartments and consequently low transfer through the soil–plant/grass–cow–milk pathway. On the other hand, the obtained Th-232 activity is compared with the data from other countries. As shown in the Table 2, the mean activity of all milk brands is similar to those reported from Brazil (Melquiades and Appoloni, 2001, 2002), but relatively higher than the data from Jordan (Ababneh et al., 2010), India (Shanthi et al., 2010), Saudi Arabia (0.09–0.76 Bq/kg; Al-Zahrani, 2012, exception is 6.77 Bq/kg reported by Alamoudi, 2013), Israel (Lavi et al., 2006), and Iran (Hosseini et al., 2006), and considerably lower compared with the data from Nigeria (Osibote et al., 1999).

In respect to the activity of ⁴⁰K in the powdered infant milk under investigation, it shows a range of relative high values. This is expected as potassium is an essential element for the body metabolism and widely dispersed in the environment, thus it can easily be transferred through the food chain. Hence, the activity ranged between 40.3 ± 2.10 Bq/kg (Lactogen from Philippines) and 254 ± 12.8 Bq/kg (Sustagen from Thailand) with a mean activity being 99.1 ± 69.5 Bq/kg (all brands). With respect to countries, the infant milk sample from Singapore gives the highest mean activity, followed by those from Thailand, but a sample from Philippines shows the least mean value (Table 2). When compared with the literature, the activity of ⁴⁰K obtained in the present study show a mixed behavior: lower than the data reported from Brazil (Melquiades and Appoloni, 2001, 2002), Nigeria (Osibote et al., 1999), Hong Kong (Yu and Mao, 1994), Jordan (Ababneh et al., 2010), and IAEA milk powder reference material (Altzitzoglou and Bohnstedt, 2008); higher than the data reported from India (Shanthi et al., 2010), Saudi Arabia (Alamoudi, 2013), and Iran (Afshari et al., 2009); similar to Al-Zahrani (2012), and lie within the values reported from Syria (Al-Mastri et al., 2004) and Israel (Lavi et al., 2006).

To understand the additional radiation exposure to humans from environmental pollution such as the nuclear accidents (Chernobyl and Fukushima accidents in particular) resulting in the deposition of radionuclides on soil/grass and transfer to infant milk through the soil–grass–cow–milk pathway, the activity of artificial radionuclide (¹³⁷Cs) has been measured in the present study. It shows that although the sampled infant milk brands are mostly produced in the countries within the vicinity of Japan where the recent Fukushima nuclear accident occurred, the obtained results (mean activity 0.27 ± 0.19 Bq/kg) indicate that ¹³⁷Cs is virtually not present in the powdered infant milk, which also suggests less impact of the recent nuclear accident on the neighboring environments or country of origin of the investigated milk brands. This observation may not be unconnected to the proactive programs of Japan in reducing the environmental contamination from Fukushima. However, the present results are higher than the obtained result in milk imported from various European countries and consumed in Nigeria, 10 years after the Chernobyl accident (Osibote et al., 1999), but less than the reported data from Jordan (Ababneh et al., 2010), in dairy milk powder from New Zealand (Hermanspahn, 2011), dairy milk products from Brazil (Melquiades and Appoloni, 2001, 2002), and IAEA-152 milk powder reference material (Altzitzoglou and Bohnstedt, 2008). It is, however, important to stress that the relative ranks may not be considered as representative if there is likely high variability in the measures as, for instance, a small sample size (two samples only for each brand) is involved in the present study, which makes it difficult to generalize the findings.

Annual effective dose to different infant age groups through ingestion

The annual effective dose, D (μSv/year), to individual infant population due to the ingestion of radionuclides through the consumption of powdered milk is estimated using the Equation (3) taken from the UNSCEAR (2000) report: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}D = AIE \tag{3}\end{align*} \end{document}

where A is the activity concentration of radionuclides in the studied milk (Bq/kg), I is the annual intake of radionuclide representing the average milk consumption by the infants, which depends on the age group (for infants ≤1 year old is 22.4 kg/year and infants 1–2 years old is 15 kg/year according to ICRP reports) (International Commission on Radiological Protection, 1996), and E is the dose conversion factor of the radionuclides of interest (Table 3) taken from the UNSCEAR (1993). The estimated annual committed effective dose due to the ingestion of ²²⁶Ra, ²³²Th, ⁴⁰K, and ¹³⁷Cs shows that ²²⁶Ra is the major contributor to internal radiation dose received by the infant population through consumption of powdered milk (Table 4), but this can be justified by the high-dose conversion factor of this radionuclide compared with the others. The percentage contribution of dose by country shows that the Singaporean milk brand contributed the highest (34% for infants ≤1 year old and 31% for infants 1–2 years old) to the annual effective dose and the least contributors are the milk brands from Australia and Philippine (6%, respectively) (Fig. 1). Meanwhile, the obtained ⁴⁰K dose in this study is used only for comparison purposes. This is because potassium is an essential element strictly under homeostatic regulation in the body and not influenced by variation of environmental levels, thus the dose from ⁴⁰K within the body is constant. In case of ¹³⁷Cs, the dose due to this artificial radionuclide is of no significant value (Table 4) and hence shows less ¹³⁷Cs contamination in the investigated milk. The results here agreed with the fact that naturally occurring radionuclides are the major contributors to internal radiation dose received by humans as earlier stated, while artificial radionuclides are less and, in some cases, they make only an insignificant contribution. However, the obtained total mean of annual effective dose based on the age group is estimated to be 635.13 μSv/year and 111.45 μSv/year for infants ≤1 year and infants 1–2 years old, respectively. Again, the results obtained here are due to the cross-sectional survey (due to the involvement of limited samples), which may not be considered as the representative of the commonly consumed infant formula throughout the country. However, the estimated dose value is higher (for infants ≤1 year old) compared with the UNSCEAR (2008) reported value of 290 μSv/year through ingestion of radionuclides. The guidelines of the Codex Alimentarius Commission (FAO/WHO, 1995) stipulated the intervention level of 1 mSv per food group, above which it is considered a public health risk for such food. Note that this limit is similar to the new recommendation of ICRP (2007).

FIG. 1.

Contribution by country of origin of investigated infant milk brands to annual effective dose.

Table 3.

Dose Conversion Factors of ⁴⁰ K, ²²⁶ Ra, ²²⁸ Ra, and ¹³⁷ Cs for Age Groups, Infants <1 Year Old and Infants 1–2 Years Old (UNSCEAR, 1993 )

	Dose conversion factors (nSv/Bq)
Category	²²⁶Ra	²³²Th	⁴⁰K	¹³⁷Cs
Infants ≤1 year old	4,700	4,600	62	21
Infants 1–2 years old	960	450	42	12

Table 4.

Estimated Annual Effective Dose (μSv) Through Consumption of Infant Formula

		Individual effective dose (μSv) due to
Age group	Range	²²⁶Ra	²³²Th	⁴⁰K	¹³⁷Cs	Annual effective dose (μSv)
Infants ≤1 year old	Minimum	143	31.9	56.0	0.052
	Maximum	743	883	326	0.28
	Mean	296	214	125	0.13	635.13
Infants 1–2 years old	Minimum	19.6	2.09	25.4	0.02
	Maximum	102	57.8	160	0.11
	Mean	40.4	14.0	57	0.049	111.45

Statistical analysis

Detailed analysis of the results from the individual measurement was further performed using the S+ software package to find the correlation among the radionuclides and their significant differences at 95% confidence level. Table 5 shows the degree of association existing among the measured radionuclide parameters for the investigated powdered infant milk. Figures 2 and 3 exhibit that ⁴⁰K makes a strong positive correlation with ²²⁶Ra (r = 0.898) and ²³²Th (r = 0.710). Similarly, ²³²Th and ²²⁶Ra are positively correlated (0.842) as seen in Fig. 4, which indicates that both thorium and radium are part of the same decay chain. Hence, the individual results of one nuclide can be used to predict the individual results of the two others.

FIG. 2.

²²⁶Ra versus ⁴⁰K concentrations in the brands of infant formula.

FIG. 3.

²³²Th versus ⁴⁰K concentrations in the brands of infant formula.

FIG. 4.

²³²Th versus ²²⁶Ra concentrations in the brands of infant formula.

Table 5.

Correlation Coefficients Among Radioactive Parameters for the Investigated Powdered Infant Milk

					AED
Variables	²²⁶Ra	²³²Th	⁴⁰K	¹³⁷Cs	Infants ≤1 year old	Infants 1–2 years old
²²⁶Ra	1
²³²Th	0.842	1
⁴⁰K	0.898	0.710	1
¹³⁷Cs	−0.165	−0.222	−0.049	1
AED
Infants ≤1 year old	0.968	0.941	0.889	−0.177	1
Infants 1–2 years old	0.968	0.841	0.972	−0.118	0.971	1

AED, annual effective dose.

Radiological risk analysis

Risk assessment due to the ingestion of radionuclides through infant milk was estimated for the age groups (infants ≤1 year old and infants 1–2 years old) using the calculated effective dose. The essence is to have an idea of total risks from all of the investigated radionuclides compared with the lifetime risk from overall exposure to radionuclides coming through different pathways to the whole body for the typical duration of human life (70 years). The following Equation (4) (Giri et al., 2011) was used to estimate the risk: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} \begin{split}{\rm Risk} &= {\rm Average \ daily \ dose} \times {\rm SF} \\ & \quad \times {\rm Duration \ of \ infant \ exposure} \end{split} \tag{4}\end{align*} \end{document}

where SF = slope factor (morbidity risk coefficient) obtained from the US EPA (1993) expressed in activity/risk (pCi/risk; 1 Bq = 27 pCi) and it is specific for each radionuclide (Table 6). Duration of infant exposure was considered as 365 days for each age group. Table 6 shows the estimated risk for the age groups to be 3.22 × 10⁻⁶ and 2.16 × 10⁻⁶ for infants ≤1 year old and infants 1–2 years old, respectively. These values are within the recommendation of the excess lifetime cancer risk of 1 × 10⁻⁶ and 1 × 10⁻⁴ by US EPA (1991).

Table 6.

Risk Estimate Due to Ingestion of Radionuclide Through Infant Formula

Parameter	Infant age	²²⁶Ra	²³²Th	⁴⁰K	¹³⁷Cs	Total risk
Daily intake (pCi)	≤1 year	5.05	4.23	164	0.45
	1–2 years	3.38	2.83	110	0.30
Slope factor (risk/pCi)		5.14 × 10⁻¹⁰	1.33 × 10⁻¹⁰	3.43 × 10⁻¹¹	3.74 × 10⁻¹¹
Risk	≤ 1 year	9.48 × 10⁻⁷	2.05 × 10⁻⁷	2.06 × 10⁻⁶	6.11 × 10⁻⁹	3.22 × 10⁻⁶
	1–2 years	6.35 × 10⁻⁷	1.37 × 10⁻⁷	1.38 × 10⁻⁶	4.09 × 10⁻⁹	2.16 × 10⁻⁶

Conclusion

Activity concentrations of naturally occurring radionuclides (²²⁶Ra, ²³²Th, and ⁴⁰K) and artificial ¹³⁷Cs in widely consumed powdered milk (i.e., infant formula) in Malaysia have been determined by HPGe γ-ray spectrometry. Due to the involvement of small sample numbers, this study represents a cross-sectional survey, but not a generalized one. The activity of ²²⁶Ra, ²³²Th, ⁴⁰K, and ¹³⁷Cs in all of the brands of infant milk ranged from 1.36 ± 0.14–7.06 ± 0.84 Bq/kg, 0.31 ± 0.10–8.57 ± 2.11 Bq/kg, 40.3 ± 2.10–254 ± 12.8 Bq/kg, and ND−0.73 ± 0.47 Bq/kg, respectively. With regard to the country of origin of the studied milk brands, the brand (Lactogen) from Philippines shows low levels of radioactivity, while the brand from Singapore (S26 SMA Gold) depicts relatively higher values. It is possible that the detected activity might be the result of raw materials imported from elsewhere in the world. The present results lie within the literature data measured from different brands of milk around the world. Meanwhile, the mean annual effective dose due to the ingestion of radioactivity through powdered infant milk was estimated and found to be above the UNSCEAR reference value. As far as our concern, this is the first study (i.e., it can also be considered a pilot study) for the radioactive content of powdered infant milk consumed by Malaysian infants. The data reported here might be useful to establish a baseline for natural and artificial radioactivity in milk and help to develop future guidelines in the country for radiological protection for the relevant population.

Footnotes

Acknowledgment

This work was supported by the University of Malaya Research Grant nos. RP006D-13AFR and IPPP PG065-2014A.

Author Disclosure Statement

No competing financial interests exist.

References

Ababneh

Z.Q.

, Alyassin

A.M.

, Aljarrah

K.M.

, and Ababneh

A.M.

(2010). Measurement of natural and artificial radioactivity in powdered milk consumed in Jordan and estimates of the corresponding annual effective dose. Radiat. Prot. Dosimetry, 138, 278.

Abbott Voluntarily Recalls Certain Similac^® Brand Powder Infant Formulas, Abbott Press Release, issued September 22, 2010. Available at: http://www.abbott.com/global/url/pressrelease/en_us/60.5:5/press_release_0900.htm (accessed May 5, 2015).

Ademola

J.A.

, and Ehiedu

S.I.

(2010). Radiological analysis of ⁴⁰K, ²²⁶Ra and ²³²Th in fish, crustaceans and sediment samples from fresh and marine water in oil exploration area of Ondo state, Nigeria. Afr. J. Biomed. Res., 13, 99.

Afshari

N.S.

, Abbasisiar

, Abdolmaleki

, and Nejad

M.G.

(2009). Determination of ⁴⁰K concentration in milk samples consumed in Tehran–Iran and estimation of its annual effective dose. Iran. J. Radiat. Res., 7, 159.

Ahmad

, Jaafar

M.S.

, Bakhash

, and Rahim

(2015). An overview on measurements of natural radioactivity in Malaysia. J. Radiat. Res. Appl. Sci., 8, 136.

Alamoudi

Z.M.

(2013). Assessment of natural radionuclides in powdered milk consumed in Saudi Arabia and estimates of the corresponding annual effective dose. J. Am. Sci., 9, 267.

Al-Mastri

M.S.

, Mukallati

, Al-Hamwi

, Khalili

, Hassan

, Assaf

, Amin

, and Nashawati

(2004). Natural radionuclides in Syrian diet and their daily intake. J. Radioanal. Nucl. Chem., 260, 405.

Altzitzoglou

, and Bohnstedt

(2008). Characterization of the IAEA-152 milk powder reference material for radioactivity with assigned values traceable to the SI units. Appl. Radiat. Isot., 66, 1702.

Al-Zahrani

J.H.

(2012). Natural radioactivity and heavy metals in milk consumed in Saudi Arabia and population dose rate estimates. Life Sci. J., 9, 651.

10.

Amin

Y.M.

, Mahat

R.H.

, Nor

R.M.

, Khandaker

M.U.

, Takleef

G.H.

, and Bradley

D.A.

(2013). The presence of natural radioactivity and ¹³⁷Cs in the South China Sea bothering Peninsular Malaysia. Radiat. Prot. Dosim., 156, 475.

11.

Cambra

, and Alonso

(1995). Blood lead levels in 2- to 3-year-old children in the greater Bilboa area (Basque county, Spain): Relation to dust and water levels. Arch. Environ. Health, 50, 362.

12.

Cattaneo

(2004). Protection, promotion and support of breastfeeding in Europe: A blueprint for action. EU Project contract N. SPC 2002359.

13.

FAO/WHO. (1995). Codex Alimentarius, General Requirements, Section 6.2, Guideline Levels for Radionuclides in foods Following Accidental Nuclear Contamination for Use in International Trade. Joint FAO/WHO Food Standards Programme, Rome.

14.

Fein

S.B.

, and Falci

C.D.

(1999). Infant formula preparation, handling and related practices in the United States. J. Am. Diet. Assoc., 99, 1234.

15.

Giri

, Singh

, Jha

V.N.

, and Tripathi

R.M.

(2011). Risk assessment due to ingestion of natural radionuclides and heavy metals in the milk samples: A case study from a proposed uranium mining area, Jharkhand. Environ. Monit. Assess., 175, 157.

16.

Hermanspahn

(2011). Environmental Radioactivity in New Zealand and Rarotonga-Annual Report 2010. National Radiation Laboratory, Ministry of Health.

17.

Hosseini

, Fathivand

A.A.

, Barati

, and Karimi

(2006). Assessment of radionuclides in imported foodstuffs in Iran. Iran. J. Radiat. Res., 4, 149.

18.

ICRP. (2007). Recommendations of the international commission on radiological protection ICRP publication 103. Ann. ICRP, 37, 2.

19.

International Commission on Radiological Protection. (1996). Age-Dependent Doses to Members of the Public from Intake of Radionuclides. Part 5: Compilation of Ingestion and Inhalation Coefficients. ICRP publication 72. Oxford, United Kingdom: Pergamon Press.

20.

Jibiri

N.N.

, Farai

I.P.

, and Alausa

S.K.

(2007). Estimation of annual effective dose due to natural radioactive elements in ingestion of foodstuffs in tin mining area of Jos-Plateau, Nigeria. J. Environ. Radioact., 94, 31.

21.

Khandaker

M.U.

, Olatunji

M.A.

, Shuib

K.S.K.

, Hakimi

N.A.

, Nasir

N.L.M.

, Asaduzzaman

K.H.

, Amin

Y.M.

, and Kassim

H.A.

(2015). Natural radioactivity and effective dose due to the bottom sea and estuaries marine animals in the coastal waters around peninsular Malaysia. Radiat. Prot. Dosimetry [Epub ahead of print]; DOI:10.1093/rpd/ncv243.

22.

Khandaker

M.U.

, Wahib

N.B.

, Amin

Y.M.

, and Bradley

D.A.

(2013). Committed effective dose from naturally occurring radionuclides in shellfish. Radiat. Phys. Chem., 88, 1.

23.

Lavi

, Golob

, and Alfassi

(2006). Monitoring and surveillance of radio-cesium in cultivated soils and foodstuff samples in Israel 18 years after the chernobyl disaster. Radiat. Meas., 41, 78.

24.

Licata

, Trombetta

, Cristiani

, Giofre

, Martino

, Calo

, and Naccari

(2004). Levels of “toxic” and “essential” metals in samples of bovine milk from various dairy farms in Calabria, Italy. Environ. Int., 30, 1.

25.

Lonnerdal

, and Hernell

(1998). Effects of feeding ultrahigh-temperature (UHT)-treated infant formula with different protein concentrations or powdered formula, as compared with breast-feeding, on plasma amino acids, hematology, and trace element status. Am. J. Clin. Nutr., 68, 350.

26.

Melquiades

F.L.

, and Appoloni

C.R.

(2001). Radiation of powdered milk produced at Londrina, PR, Brazil. Radiat. Phys. Chem., 61, 691.

27.

Melquiades

F.L.

, and Appoloni

C.R.

(2002). ⁴⁰K, ¹³⁷Cs and ²³²Th activities Brazilian milk samples measured by gamma ray spectrometry. Ind. J. Pure Appl. Phys., 40, 5.

28.

Mount

(1985). $27 Million for Victims of Baby Formula. Chicago Tribune, March 1, 1985. Available at: http://pqasb.pqarchiver.com/chicagotribune/doc/290828620.html?FMT=ABS&FMTS=ABS:FT&type=current&date=Mar%2001,%201985&author=Charles%20Mount&pub=Chicago%20Tribune%20(pre-1997%20Fulltext)&edition=&startpage=&desc=$27%20MILLION%20FOR%202%20VICTIMS%20OF%20BABY%20FORMULA (accessed May 5, 1981).

29.

Navarrete

J.M.

, Martinez

, and Cabrera

(2007). Comparative study between radioactive contamination in powder milk by Chernobyl accident (¹³⁷Cs) and natural radioactivity (⁴⁰K). J. ‘Radioanal. Nucl. Chem., 272, 277.

30.

Nisbet

A.F.

, and Woodman

R.F.M.

(2000). Soil-to-plant transfer factor for radiocaesium and radiostrontium in agricultural systems. Health Phys., 78, 279.

31.

Olatunji

M.A.

, Uwatse

O.B.

, Khandaker

M.U.

, Amin

Y.M.

, and Faruq

(2014). Radiological study on newly developed composite corn advance lines in Malaysia. Phys. Scr., 89, 125002.

32.

Osibote

, Olomo

, Tchokossa

, and Balogun

(1999). Radioactivity in milk consumed in Nigeria 10 years after Chernobyl reactor accident. Nucl. Instrum. Methods Phys. Res. A, 422, 778.

33.

Pandelova

, Lopez

W.L.

, Michalke

, and Schramm

K.W.

(2012). Ca, Cd, Cu, Fe, Hg, Mn, Ni, Pb, Se, and Zn contents in baby foods from the EU market: Comparison of assessed infant intakes with the present safety limits for minerals and trace elements. J. Food Comp. Anal., 27, 120.

34.

Pietrzak-Flis

, Chrzanowski

, and Dembinska

(1997). Intake of ²²⁶Ra, ²¹⁰Pb and ²¹⁰Po with food in Poland. Sci. Total Environ., 203, 157.

35.

Quîndós

L.S.

, Fernández

P.L.

, Soto

, Ródenas

, and Gómez

(1994). Natural radioactivity in Spanish soils. Health Phys., 66, 194.

36.

Saat

, Kassim

, Hamzah

, and Farisz

(2010). Determination of surface radiation dose and concentrations of uranium and thorium in soil at UITM Perhilitan research station Kuala Keniam, Taman Negara, Pahang. J. Nucl. Relat. Technol., 2, 49.

37.

Santawamaitre

, Malain

, Al-Sulaiti

H.A.

, Matthews

, Bradley

D.A.

, and Regan

P.H.

(2011). Study of natural radioactivity in riverbank soils along the Chao Phraya river basin in Thailand. Nucl. Instrum. Methods Phys. Res. A, 652, 920.

38.

Shanthi

, Kumaran

J.T.T.

, Raj

G.A.G.

, and Maniyan

C.G.

(2010). Natural radionuclides in the South Indian foods and their annual dose. Nucl. Instrum. Methods Phys. Res., 619, 436.

39.

Solomon

(1981). The Controversy Over Infant Formula Available at: http://www.nytimes.com/1981/12/06/magazine/the-controversy-over-infant-formula.html (accessed May 5, 2015).

40.

Stepanenko

V.F.

, Voillequé

P.G.

, Gavrilin

Y.I.

, Khrouch

V.T.

, Shinkarev

S.M.

, Orlov

M.Y.

, Kondrashov

A.E.

, Petin

D.V.

, Iaskova

E.K.

, and Tsyb

A.F.

(2004). Estimating individual thyroid doses for a case-control study of childhood thyroid cancer in Bryansk Oblast, Russia. Radiat. Prot. Dosimetry, 108, 143.

41.

Štrok

, and Smodiš

(2011). Natural radionuclides in milk from the vicinity of a former uranium mine. Nucl. Eng. Des., 241, 1277.

42.

Tripathi

R.M.

, Raghunath

, Mahapatra

, and Sadasivan

(2001). Blood lead and its effect on Cd, Cu, Zn, Fe and haemoglobin levels of children. Sci. Total Environ., 277, 161.

43.

UNSCEAR. (1993). United Nations Scientific Committee on the Effects of Atomic Radiation. Reports to the General Assembly. New York: United Nations.

44.

UNSCEAR. (2000). Sources and Effects of Ionizing Radiation. Report to the General Assembly with Scientific Annexes. New York: United Nations.

45.

UNSCEAR. (2008). Sources and Effects of Ionizing Radiation. New York: United Nations Publication, pp. 223–233.

46.

US-EPA. (1991). Role of the Baseline Risk Assessment in Superfund Remedy Selection Decisions (Memorandum from D. R. Clay, OSWER 9355.0–30, April 1991). Washington DC: US Environmental Protection Agency. Available at: www.epa.gov/oswer/riskassessment/baseline.htm (accessed May 7, 2015).

47.

US-EPA. (1993). Carcinogenicity Assessment. IRIS (Integrated Risk Information System), 2003. Washington DC: US Environmental Protection Agency. Available at: www.epa.gov/iris (accessed May 7, 2015).

48.

WHO. (2003). Global Strategy for Infant and Young Child Feeding. Geneva. Available at: http://whqlibdoc.who.int/publications/2003/9241562218.pdf (accessed May 5, 2015).

49.

WHO. (2008). Strengthening action to improve feeding of infants and young children 6–23 months of age in nutrition and child health programmes. In Report of Proceedings. Geneva, October 6–9 (ISBN: 978 92 4 159789 0).

50.

William

F.R.

, Steck

D.J.

, Smith

, Brus

C.P.

, Fisher

, Neuberser

J.S.

, Platz

C.P.

, Robinson

R.A.

, Woolson

R.F.

, and Lynch

C.F.

(2000). Residential radon gas exposure and lung cancer. Am. J. Epidemiol., 151, 1091.

51.

Xhixha

, Bezzon

G.P.

, Broggini

, Buso

G.P.

, Caciolli

, Callegari

, De Bianchi

, Fiorentini

, Guastaldi

, Xhixha

M.K.

, Mantovani

, Massa

, Menegazzo

, Mou

, Pasquinin

, Alvarez

C.R.

, and Shyti

(2013). The worldwide NORM production and a fully automated gamma-ray spectrometer for their characterization. J. Radioanal. Nucl. Chem., 295, 445.

52.

K.N.

, and Mao

S.Y.

(1994). Application of high resolution gamma-ray spectrometry in measuring radioactivities in drinks in Hong Kong. Appl. Radiat. Isotopes, 45, 1031.