Assessment of toxic elements in sediments of Linggi River using NAA and ICP-MS techniques

Graphical abstract

weighed using the analytical balance into polyethylene vials and sealed with heating solder prior to the irradiation process [14].

Irradiation process
The duplicate samples, mix standard solution and SRM were irradiated together in the 750 kW PUSPATI TRIGA Mark II reactor at Malaysian Nuclear Agency, with a thermal flux of 4.0 Â 10 12 n cm -2 s -1 [15]. The isotope of radionuclied and other information were shown in Table 1. The irradiation process of up to 6 h at the rotating rack for the long-life radionuclides (As, Sb, Cr, Zn and Fe) were performed. Cooling  time for a decay process were ranged from 2 to 4 days before performing the first counting, and 3 to 4 weeks for the second counting using gamma spectrometry.

Counting process and concentration measurement by using NAA technique
The counting process of the irradiated samples, mix standard solution and SRM were performed for one hour each, by using gamma spectrometry. The detector was calibrated from low to high energy of gamma ray by using 241 Am (59.5 keV), 109 Cd (88.1 keV), 57 Co (122.1, 136.5 keV), 133 Ba (81.0, 303.0, 356.0, 384.0 keV), 137 Cs (661.7 keV), 60 Co (1173.2, 1332.5 keV) and 88 Y (898.0, 1836.1 keV) [16,17]. The efficiency curve of gamma ray calibration of gamma spectrometer was shown in Fig. 1. The counting geometry of samples, mix standard solution and SRM were performed at 16 cm and 4 cm for the fist and second counting, respectively. The signal of g-ray from the respective energy of elements were detected by a coaxial hyperpure germanium detector supply by EG&G ORTEC. The signals were then amplified and connected to multichannel analyser (MCA) and then the signal was converted to the photopeak (as net count area) by Gamma Vision software. Computations of elemental concentrations were based on comparative method and data were reported in dry weight (d.w.). The concentration of the sample and SRM was measured by using Eq. (1) [15,18] (Fig. 2).
Where: A smp = net count of the selected peak area of an interested element in a sample A std = net count of the selected peak area of an interested element in a standard W smp = Weight of sample used W std = Weight of standard used C std = Concentration of interested element in standard (e.g.: mg/g, mg/kg) C El = Concentration of interested element in sample (e.g.: mg/g, mg/kg)

Digestion of sediment sample
Approximately 0.1-0.2 g of homogenised sediment samples of Linggi River were digested in containing of 5 mL nitric acid (67% HNO 3 -TraceMetal Fisher brand) and 2 mL concentrated hydrofluoric acid (49% HFanalytical grade, QRëC 1 brand) by microwave oven (Mars 5 brand). Each digestion batch was included at least two reagents blank acid, SRM (e.g., IAEA Soil-7) and duplicate samples. The microwave oven was programmed as follows: The power of the microwave setup at 1200 W, temperature setup for 200 C, ramped for 20 min, held for 15 min, pressure setup at 0.6 MPa [7]. Sediment samples were digested for 20 min. After cooling for at least 30 min and microwave temperature below 50 C, samples were removed from the microwave oven. If the solutions contained some residue, 1 mL HNO 3 was added and the digestion process was repeated until clear solutions were obtained. After that, the solution was transferred into Teflon beaker and rinsed with 3 mL Milli-Q water. The Teflon beaker containing the digested sample was heated at 60 C to 70 C on a hot plate until dry and the Teflon beaker was then rinsed with 20 mL Milli-Q water. The solution was then filtered with filter paper (Ø=diameter 125 mm, whatmann brand) into a polyethylene bottle and lastly brought up to a volume of 50 ml with Milli-Q water for ICP-MS analysis.

Analysis of samples using inductively coupled plasmamass spectrometry (ICP-MS)
The commercial mix standard solution (Standard 1 X) was purchased from Perkin Elmer. Instrumental operating conditions and data acquisition settings of ICP-MS was shown in Table 2. The mix standard solution of 10, 50, 100 and 150 mg/L were prepared for standard calibration curve. The calibrations of Pb, Cd, Cu and Ni showed good linearity with a correlation coefficient (R 2 ) >0.9998 ( Table 2). The concentrations of Pb, Cd, Cu and Ni in the samples were analysed by ICP-MS (Perkin Elmer-Elan 6000). The reagent blank acid used in digestion process was monitored throughout the analysis and used to correct the analytical results.

Quality control and quality assurance of analytical method
The SRM (IAEA Soil-7) was used as quality control and quality assurance in the analytical method analysis. The SRM measurement followed the same procedure as a sample analysis. The certified and measured value, percentage of recovery, coefficient of variation (CV) and other information are tabulated in Table 2. The recovery and coefficient of variation percentage of the analysed SRM ranged  Table 3). The calculation of relative bias (%) and U test score are described in Eqs. (2) and (3), respectively. The calculated U test value is compared with critical value listed in the t-statistic table to determine if the analysed result differs significantly from the certified value at a given level of probability (Table 4). The U test score are acceptable with U test value is 1.95 [19].
Where:C analysed = concentration of analysed value C certified = concentration of certified value s analysed = standard deviation of analysed value s certified = standard deviation of certified value

Distributions and concentrations of samples
The concentrations of toxic elements in the surface sediments from Linggi River are listed in Table 5. The mean concentrations of As, Cd, Sb, Pb and Zn were found to be 21.2, 2.9, 6.1, 2.1 and 1.6 times greater, respectively, than the continental crust (CC) values, whilst other elements such as Cu, Cr, Fe, and Ni showed lower concentrations as compared to CC values. All sampling locations of the study area showed higher concentration values of As compared to the CC value. The contamination of As, Cd, Sb, and Zn, were suspected originate from industrial activities. The concentrations of Cu, Pb and Zn concentration values were recorded higher at the downstream area of the Linggi River and were originated from the industrial effluents [12]. Most of the high concentrations of As, Cd, Sb, and Zn were observed at sampling locations of SL 09 -SL15, which are significantly derived from elementals pollution from Linggi River (main river) and its tributary (Simin river) ( Fig. 1). High concentrations of Pb were observed at SL02 and SL03 (Seremban area), SL07 and SL09 (Rantau area), where most of the industrial activities are located. The contamination of Pb was suspected from industrial activities with effluents from metal smelting, electroplating and factories. However, the possible sources of As and Pb elements are originated from phosphate fertilizer, lead-arsenate insecticides, and pesticides used in the agriculture activities [20]. Table 3 Certified and analysed values of standard reference material (SRM) IAEA-Soil-7 and other information.

Enrichment factor (EF)
In order to evaluate possible anthropogenic sources of toxic elements, the enrichment factor (EF) was calculated based on the Eq. (4) below: Where M is the element of interest, R is the reference element, (M/R) measure is the elemental ratio found in the sample, and (M/R) CC is the elemental ratio in the continental crust. Iron (Fe) was used for normalisation purpose to determine the metal and heavy metal pollution of Linggi River. The selection of Fe as a normalisation element and used to be in the EF calculation was due to Fe distribution being not related to other heavy metals [21]. Fe usually has a relatively high natural concentration [22], and therefore not expected to be substantially enriched from anthropogenic source in estuarine sediment [23]. Most of the researchers suggested that EF values as the following: EF < 2 indicates no enrichment, EF = 2 to 3 is minor enrichment, EF = 3 to 5 is moderate enrichment, EF = 5 to 10 is moderately severe enrichment, EF = 10 to 25 is severe enrichment, EF = 25 to 50 is very severe enrichment and EF > 50 is extremely severe enrichment [24]. EF values of less than 2.0 indicate that the element in the sediment originated predominantly from lithogenous materials, whereas EFs are much greater than 2.0 indicating that the element is of anthropogenic origin [25]. The calculated EF values for selected elements of Linggi River are shown in Table 6. The EFs of As show enrichment in all sampling locations (EF values 17.7-75.0). Arsenic pollution can be categorised The analysed result is probably significantly different from the certified value. U-test > 3. 29 The analysed result is significantly different from the certified value. as severe to extreme enrichment. Cd, Pb, Sb and Zn can be categorised as minor to severe enrichment at most of the other sampling locations. Other elements such as Cr, Cu and Ni showed no enrichment at most of the sampling locations. EF of metals and heavy metals can be valuable and have been used as an indirect indicator for evaluation of sediment contamination or toxicity. However, it is not sufficient to use enrichment factor only for the evaluation of sediment toxicity at a particular site. Consideration for the degree of contamination in sediment and comparison with sediment guidelines are useful to evaluate the toxicity of the sediment for the particular site.

Degree of contamination (C d )
To describe the contamination of toxic elements in Linggi River, the following Eqs. 5 and 6 below are used to define as a contamination factor (C f ) and degree of contamination (C d ), respectively; where C d is the degree of contamination, C f is a contamination factor, C n is the metal content in the sediments and C 0 is a background value (reference value of metals). The following terminology was used to describe the contamination factor: C f < 1 low contamination factor (indicating low sediment contamination); 1 C f < 3 moderate contamination factor; 3 C f < 6 considerable contamination factor; C f ! 6 very high contamination factor [26]. The contamination factor (C f ) values were shown in Table 7. At all sampling stations As showed very high contamination factor, except at stations SL01 and SL04. Contamination factors of Cr, Cu and Ni can be categorised as low contamination. The elements of Cd, Pb, Sb and Zn can be categorised as low to very high contamination factor. The results indicated that contamination of sediments of Linggi River were mainly contributed by As, Cd, Pb, Sb and Zn.
The degrees of contamination (C d ) values of Linggi River are shown in Table 7. The degree of contamination (C d ) is defined as the sum of all contamination factors (C f ) of As, Cd, Cr, Cu, Ni, Pb, Sb, and Zn. Degree of contamination can be categorised into four categories according to the Hakanson (1980) classification. For the description of degree of contamination values, the following terminologies have been used: C d < 8 low degree of contamination; 8 C d < 16 moderate degree of Notes: Bold type indicates the enrichment of elemental pollution in sediment (EF value > 2.0). Table 7 Contamination factor (C f ) and degrees of contamination (C d ) values of toxic elements from Linggi River.     [12], 1990; b [27]; c [28] ; d [29]; e [30]; f [31]; g [32]; h [33]; i [34]; j [11]; value in parentheses = mean concentration; TEL - contamination; 16 C d < 32 considerable degree of contamination; C d ! 32 very high degree of contamination. SL01 and SL04 can be categorised as low degree of contamination with C d values 5.3 and 4.9, respectively. Sampling location of SL03 can be categorised as having a moderate degree of contamination. SL02, SL06 and SL11 stations can be categorised into considerable degree of contamination. Most of the sampling stations (eight stations) showed very high degrees of contamination with C d values ranging from 32.6 to 73.6. This indicated very high loading of anthropogenic pollution at these eight sampling locations (SL07 to SL10 and SL12 to SL15).

Comparison of toxic elements with FSQGs
In this paper, we adopted the Canadian Freshwater Sediment Quality Guidelines (Canadian-FSQGs) and the Consensus Freshwater Sediment Quality Guidelines (Consensus-FSQGs) published by MacDonald (2000) for the purpose of comparison between Malaysian rivers sediment and FSQGs. If a trace element concentration in sediment was less than the TEC or TEL values, the sediments were considered to be clean to marginally pollute. No effects on the majority of sediment-dwelling organisms were expected below the TEC or TEL concentration values. If the concentration of toxic element in sediment was greater than the PEC or PEL values, the sediments were to be considered heavily polluted. Adverse effects on the majority of sediment-dwelling organisms were expected when the concentrations exceeded PEC or PEL of FSQGs values.
Comparisons of toxic elements in sediments of the Linggi River and Malaysian rivers to Canadian FSQGs and Consensus FSQGs are shown in Table 8. Mean As concentrations of Linggi River in this study were higher than those of the Canadian-FSQGs -PEL value and Consensus-FSQGs -PEC value. These indicated that the Linggi River sediments were polluted with As and this may cause adverse effects to the majority of sediment-dwelling organisms. However, mean concentrations of Cd, Cr, Cu, Ni, Pb and Zn are less than the Canadian-FSQGs -TEL value and Consensus-FSQGs -TEC value. The concentration of As in sediments of Pelepah Kanan River, Kota Tinggi showed 4.5 times higher than the Consensus-FSQGs -PEC value and mean concentration of Cd in sediments of Langat River (12.1 mg/kg) showed higher concentration as compared to the Consensus-FSQGs -PEC value (4.98 mg/kg), as shown in Table 8.

Conclusions
The enrichment factors and degree of contaminations showed that the sediments collected from the Linggi River were polluted with toxic elements As, Cd, Pb, Sb, and Zn. The source of As, Cd, Pb, Sb, and Zn pollution were originated from industries. Amongst the elements analysed, As showed high EF and C f values in most of the sampling stations. The mean As concentrations of Linggi River sediments showed higher concentration than values from the guidelines of Canadian-FSQGs -PEL (17.0 mg/kg) and Consensus-FSQGs -PEC (33.0 mg/kg). The high concentrations of toxic elements such as As and Cd than those of the PEC-FSQGs could result in an adverse effect on the benthic organisms and marine ecology. The results of the assessment of Linggi River sediments obtained from this study will provide vital information that can be used for future comparison. Information from the present study will be useful to the relevant government agencies and authorities in preparing preventive actions to control direct discharge of toxic elements and other pollutants from industries, agro-based activities and domestic wastes into the rivers.