Determination of uranium in tree bark samples by epithermal neutron activation analysis

In this study uranium (U) concentrations were determined in certified reference materials (CRMs) and in tree bark samples collected in “Cidade Universitária Armando de Salles Oliveira” (CUASO) USP, São Paulo. CRMs were analyzed using the experimental conditions used for barks for analytical quality control of U determination. The barks were collected from different species namely Poincianella pluviosa and Tipuana tipu. These bark samples were cleaned, dried, grated and milled for the analyses by epithermal neutron activation analysis method (ENAA). This method consists on irradiating CRMs, samples and U standard in IEA-R1 nuclear reactor with thermal neutron flux of 1.9 x 10 12 n cm -2 s -1 during 40 to 60 seconds depending on their matrices. The U was identified, using gamma-ray spectrometry, by the peak of 74.66 keV of 239 U with half-life of 23.47 minutes. Concentration of U was calculated by comparative method. Results obtained for CRMs presented good precision and accuracy, with |Z score| ≤ 0.39. Uranium concentrations in tree barks varied from 83.1 to 627.6 ng g -1 and the relative standard deviations of these results ranged from 1.8 to 10 %.


INTRODUCTION
The U is a metallic and radioactive chemical element that is naturally distributed in the ground with concentrations between 0.1 and 20 mg kg -1 [1].This element is found in different concentrations in water, air and foods.The increasing application of U in nuclear industry, agriculture and in nuclear weapons fabrication indicates that this element has been systematically extracted from nature and then deposited in the environment, which makes transference to water, plants and food possible.
It is well known that U is a toxic element for human beings and consequently it represents a significant hazard to human health [2].The damages caused by U on humans make essential its study and determination of concentration on environment.Usually U is present on biological and environmental samples at very low concentrations.Therefore, it is necessary to evaluate the quality of U results in order to get reliable data.
Therefore, this research consisted on U determination in tree barks, which are used as environmental pollution biomonitor and certified reference materials by using short irradiation of ENAA method.

TREE BARK SAMPLE COLLECTION
Tree bark samples of two different arboreal species, namely Tipuana tipu (Tipuana) and Poincianella pluviosa (Sibipiruna), were collected at the "Cidade Universitária Armando de Salles Oliveira" (CUASO) of the University of São Paulo, located in São Paulo city at Butantã district.There were eight samples, which were then obtained in different sites of CUASO during the period January and February 2017.Such collection spots were chosen randomly only considering existence of mentioned tree species.The process of getting samples is based on removing barks from trees at about 1.5 m from the topsoil by using a stainless steel knife and then placing them in paper bags.

PREPARATION OF TREE BARK SAMPLES FOR THE ANAL-YSES
Firstly, tree bark samples were dried in an oven (FANEM mod.320-SE) of forced air circulation at 40 o C for about 48 hours.Then they were cleaned using a nylon bristles toothbrush in order to remove dust or any type of extraneous materials.After cleaning them, an external layer of 3 mm of each bark is removed by using a titanium grater.So a Fritsch "Analysette 3 Pulverisette 0" micro mill was used to turn these barks into a powder.After this process, this powder was placed in plastic vials and stored in a desiccator.

CERTIFIED REFERENCE MATERIALS
Certified reference materials (CRMs) were used for ensuring metrological traceability and for eval- To calculate U concentrations on dry weight basis, the humidity of CRMs were determined.This determination consisted on drying an aliquot of each material at temperature of 85 o C and 110 o C for biological and geological materials, respectively for 48 hours or until to obtain constant mass.The percentage of humidity obtained is presented in Table 1.

PREPARATION OF SYNTHETIC STANDARD OF URANIUM
The diluted U standard solution was prepared by using a certified standard solution from Spex Certiprep Chemical (USA) with a concentration of 1003 ± 3 mg L -1 and using MILLI-Q water.The prepared U solution was stored in a refrigerator.A volume of 50 µL of the diluted U standard solution were pipetted onto sheets of Whatman n o 40 filter paper with the dimensions of 1.5 cm x 6 cm.
This synthetic standard was kept in a desiccator in order to dry the pipetted aliquot.After drying, these sheets were folded and placed into polyethylene involucres.The concentration of U diluted solution was of 200.51 µg mL -1 and U mass irradiated was 10.02 µg.

PROCEDURE FOR EPITHERMAL NEUTRON ANALYSIS
Aliquots varying from 100 mg to 200 mg of each CRM or bark samples were weighed in polyethylene involucres.For irradiating, a CRM, sample and U synthetic standard were placed in a cadmium (Cd) capsule and that was placed inside a polyethylene device (rabbit).
The irradiation time range varied from 15 to 40 seconds depending on the type of matrix.The irradiation was carried out using a pneumatic station of the IEA-R1 reactor of IPEN -CNEN/SP.The thermal neutron flux used was of 1.9 x 10 12 n cm -2 s -1 .After irradiation, the CRM, bark samples and the U standard were mounted in a stainless steel base for counting.The induced gamma activities were measured using a high-purity Ge detector (CANBERRA), model GC2018, connected to a DSA 1000 digital spectrum analyzer (CANBERRA).The resolution of the system (FWHM) was 1.05 keV for 121.97 keV peak of the 57 Co and of 1.89 keV for 1332.47 keV gamma ray of 60 Co.A chronometer was used to record the decay time.Spectral data acquisition and its processing were carried out by using Genie 2000 software, version 3.1 (CANBERRA).Counting times of 500 seconds were used for the samples and standard.Uranium was identified by gamma rays of 74.66 keV of 239 U with a half-life of 23.47 minutes [14].The U concentration was calculated by comparative method according to De Soete et al. [15].

TREATMENT OF DATA
Uranium results obtained in this study were evaluated using basic statistic calculating arithmetic mean (M), standard deviation (SD), relative standard deviation (RSD) and relative errors (RE).
Moreover, the standardized difference or Z score value was calculated to evaluate accuracy of the results.According to Konieska and Namièsnik [16] Z score is given by equation 1.
where xlab is the obtained value xref is the certified value, ref u is the combined uncertainty of the certified value and DP is the standard deviation of the results obtained.The criteria of Z score values are [16]: The detection and quantification limits were calculated using equations 2 and 3, respectively, according to Currie [17].
where LDT and LQT are count rates for the minimum detectable and quantifiable concentrations, respectively; BG is counting rate of background, which corresponds to area under the peak and LT is the counting time.Once the values LDT and LQT have been calculated, the limits in con- centration units were obtained using comparative method.

URANIUM RESULTS IN CERTIFIED REFERENCE MATERIALS
In Table 2, are presented results of U concentrations obtained in CRMs, also limits detection, and certified values.These results of U concentrations in CRMs indicate good agreement with certified values.The relative errors are between 0.1 and 3.1 % and |Z score| values are lower than 0.39.In 1575 Pine Needles and M3 Herring Tissue CRMs, U was not detected using short irradiation of ENAA due to its low concentration in these two materials.In Table 2, it can be seen that U concentrations for these both materials are lower than their respective detection limits values.

URANIUM DETERMINATION IN TREE BARK SAMPLES
The U concentrations obtained of tree bark samples are shown in Table 3.In this   Their U concentration found in tree barks is probably due to suspension of soil dust, followed by deposition on the bark.

CONCLUSION
Results obtained in this study demonstrated that procedure of ENAA using short irradiation could be applied satisfactorily in U determination in environmental samples of tree barks.The results obtained in the analyses of certified reference materials presented good precision and accuracy.Unfortunately, there are no environmental regulations or norms that present an acceptable limit value for U in tree barks.This fact indicates that the ENAA procedure established in this study may contribute to the establishment on this acceptable limit value.The detection limits depend on the composition of the samples and in the case of tree barks, these were very low (lower than 122.5 µg g −1 ).
a. Arithmetic mean and standard deviation.b.Number of determinations; c.Relative standard deviation.d.Samples are coded with letters T and S for Tipuana and Sibipiruna tree bark samples, respectively.Uranium concentrations obtained in tree bark samples show precise results with relative standard deviations varying from 1.8 to 10 %.The reproducibility of these results also indicates the homogeneity of prepared bark sample in relation to U concentration.The detection limits presented in Table 3 indicate a high sensitivity of the ENAA procedure for U determined in tree barks.Samples collected at CUASO of São Paulo University presented U concentrations varying from 83.1 to 628 ng g −1 .

Table 1 :
Percentage of humidity determined in the certified reference materials.

Table 2 :
Uranium Table, the mean values of concentrations with standard deviation, relative standard deviation and values of detection limits and quantification limits are presented.

Table 3 :
Uranium concentrations and values of detection and quantification limits obtained in tree bark samples analyses.