Heavy Metals and Polycyclic Aromatic Hydrocarbons in Soil from E-waste Dumpsites in Lagos and Ibadan, Nigeria

Background. Soil contamination from heavy metals and polycyclic aromatic hydrocarbons (PAHs) released during informal e-waste processing and disposal poses human and ecological health risks in Nigeria. Objectives. This study assesses the levels of heavy metals and PAHs in soils of e-waste dumpsites in Lagos and Ibadan, Nigeria. Methods. Composite soil samples were collected at depths of 0–15 cm, 15–30 cm and 30–45 cm from major e-waste dumpsites in Lagos and Ibadan and analyzed for lead (Pb), cadmium (Cd), copper (Cu), nickel (Ni), zinc (Zn), chromium (Cr) and PAHs to evaluate the potential contaminant contribution from e-waste activities. Control samples were collected at the Botanical Garden, University of Ibadan. Samples were analyzed for heavy metals after acid digestion using atomic absorption spectrophotometry, while PAHs were extracted using cold solvent extraction and quantified by gas chromatography-mass spectrometry. Blank determination and recovery studies were carried out for each metal. Contamination and ecological risks were assessed using soil contamination indices such as contamination factor, geo-accumulation and pollution load indices, and potential ecological risk index to categorize contaminant concentrations and associated impacts. Soil physico-chemical characteristics such as pH and total organic matter were also determined. Results. Metals concentrations in the dumpsite soils ranged from 114–2,840 mg/kg and not detectable - 6.50 mg/kg for Pb and Cd, and 42.8–5,390 mg/kg, 27.5–3,420 mg/kg, 11.0–128 mg/kg and 94.0–325 mg/kg for Cu, Zn, Ni and Cr, respectively. Serious metals accumulation was observed at every e-waste dumpsite, as shown by the pollution load index. The potential ecological risk values were between 584 and 10,402 at all of the dumpsites, signifying very high ecological risk. The total PAHs ranged from 1,756–2,224 μg/kg at the 0–15 cm level, 1,664–2,152 μg/kg at 15–30 cm and 278 μg/kg in the top- and sub-soil of the control site. Discussion. The total PAHs in the soil of e-waste dumpsites was significantly higher than in the control soil. Conclusions. The results of this study indicate that indiscriminate dumping and open burning of e-waste are potential sources of PAH and toxic metal emissions, which can pose serious human health and ecological risks.


Introduction
Soil contamination resulting from uncontrolled dumping of municipal, industrial, and agricultural solid waste, as well as hazardous waste such as e-waste, has become a public health concern in Nigeria. 1-6 Of particular concern is soil contamination at informal electronic waste recycling and disposal sites. In Nigeria, domestic and imported e-waste streams are growing steadily due to the increased availability of secondhand computers used in computer training centers, printing houses, cyber cafes, business centres and homes. Researchers have estimated that, on average, 500 shipping containers, with 400,000 computer monitors or 175,000 large TV sets enter the port of Lagos, Nigeria per year. As much as 75% of this waste is unserviceable and unable to be refurbished, and thus becomes e-waste. 7-10 In addition to precious metals such as gold, silver, and platinum, e-waste contains toxic metals such as lead (Pb) and cadmium (Cd), arsenic (As), and mercury (Hg). 11 Informal Polycyclic aromatic hydrocarbons, another class of toxic chemicals, are released by low-temperature combustion of e-waste. 19,29,30 Although limited data exist on the distribution and transport of polycyclic aromatic hydrocarbons (PAHs) from e-waste dumpsites in Nigeria, PAHs are known to be lipophilic and accumulate in the food chain near contaminated sites. 20,31-37 Their lipophilicity also makes dermal absorption possible. Epidemiological studies on occupational exposure to PAHs indicate that they can contribute to induction of skin and lung cancers. It has been reported that certain PAH metabolites interact with deoxyribonucleic acid (DNA) and are genotoxic, causing malignancies and heritable genetic damage in humans. 38 The lower molecular weight PAHs (e.g., 2-3 rings) such as naphthalene, fluorene, phenanthrene and anthracene have significant acute toxicity to aquatic organisms, while higher molecular weight PAHs (4-7 rings) such as chrysene and coronene do not, but are carcinogenic.
This study assesses the distribution and levels of toxic metals and PAHs in the soil of selected e-waste dumpsites in Lagos and Ibadan, Nigeria, where open burning is prevalent. Data of this nature are currently lacking for Nigeria, and understanding local contaminant levels is important for effective health risk assessment. We also estimated human and ecological health risks using the pollution load index and ecological risk index, using our soil concentration data as inputs. 39, 40 A secondary objective was to determine contaminant origin (lithogenic versus anthropogenic) using the index of geo-accumulation and contamination factors. 39,41

Study Area
Lagos and Ibadan are located in southwestern Nigeria. Alaba international market, Ojo (LLS 1 ) and Chinatown, Ojota (LLS 2 ) are the locations of the two e-waste dumpsites selected for the present study in Lagos. The Alaba market sampling site is a large expanse of land adjacent to the market shopping complex. The major wastes observed on this site were e-waste, followed by polythene bags, cartons, cardboards and cans. The Chinatown dumpsite is located on a small plot of land adjacent to the Chinese building at Ojota, a suburb in Lagos. Wastes observed there included broken monitor glass, plastics, cans, polythene bags and paper. The three Ibadan dumpsites include along Iwo road/Ile-pupa, located behind an electronics shopping complex (ISS 1 ), the Ogunpa dumpsite, adjacent to the Ogunpa River channel (ISS 2 ), and the Dugbe dumpsite, adjacent to residential buildings (ISS 3 ). At every site except for Dugbe (ISS 3 ), open burning to recover copper and other valuable materials is commonly practiced. At Dugbe, no traces of burning were apparent among the e-waste piles. Control samples were also collected at the Botanical Garden, University of Ibadan, Ibadan.

Soil Sample Collection
Samples for PAH determination were collected with a stainless steel hand trowel, while plastic was used for collection of samples for heavy metal determination. The stainless hand trowel and plastic were cleaned thoroughly to prevent cross contamination. Samples were collected randomly at almost 5 m distance from five different points and combined to form a composite sample, with this process repeated at three different depths (0-15 cm, 15-30 cm and 30-45 cm) for heavy metal determination and two depths (0-15 cm and 15-30 cm) for PAH determination. Samples for PAHs were packed in pre-cleaned aluminum foil, which was previously solvent rinsed and dried at 80 0 C. Polyethylene bags were Soil pH and total organic carbon (TOC) were determined by standard methods using a Jenway 3310 pH meter in ratio 1:2 (wt/vol) and the Walkey-Black method, respectively. 44 Approximately 0.5 g of each of the sieved samples were weighed, 10 mL of standard potassium dichromate used for packing soils for heavy metal determination. Samples for metals and soil characteristics determination were air-dried in the laboratory after manual removal of stones, twigs and other large materials then ground in a porcelain mortar and passed through a 2-mm sieve. PAH samples were preserved on ice and kept in the refrigerator prior to extraction and analyses.

Analytical Procedures
Samples were analyzed for PAHs, heavy metals and soil characteristics. For the metals analysis, approximately 1 g each of the sieved samples were weighed into digestion tubes and 10 ml aqua regia (concentrated hydrogen chloride and nitric acid, ratio 3:1 vol/vol) added (United States Environmental Protection Agency (USEPA) method 3050b). 42 The tubes were covered, heated in a water bath to 100 0 C for 2 hours with intermittent shaking, cooled to room temperature, and then filtered using filter papers (pore size 110 mm). The filtrate was diluted with distilled water to 25 mL and analyzed for total Pb, chromium (Cr), Cd, nickel (Ni), zinc (Zn) and copper (Cu) using atomic absorption spectrophotometry (Buck Scientific Model 205A). Metal recovery was carried out by spiking 1 g of the soil sample with known concentrations of each metal. The concentrations of the metals were determined after taking the spiked sample through the entire procedure. The concentrations of each metal in the unspiked sample was deducted from that of the spiked sample and divided by the concentrations of the metals used for spiking, then multiplied by 100. The recovery was between 93.2 -100.4% for all the metals.
Sixteen target PAHs were analyzed using gas chromatography-mass spectrometry (GS/MS) following modified USEPA methods (method 827 0 C). 43 Approximately 5 g of each sample and 5 g of anhydrous sodium sulphate were weighed and homogenized to a complete mixture. The mixtures were transferred to pre-cleaned extraction tubes, and 25 mL dichloromethane added. The tubes were tightly capped, allowed to stand for 30 minutes, and then shaken vigorously for 30 minutes. The solids were allowed to settle and solvent layers were filtered using filter papers. The procedure was repeated with 25 mL dichloromethane. The two extracts were combined, concentrated on a rotary evaporator (Büchi Rotavapor R-114), exchanged with 5 mL of n-hexane and re-concentrated to 1 mL for clean-up. The extracts were then eluted with 25 mL dichloromethane/ hexane (20:80 v/v) on a silica gel column. The extracts were evaporated and re-dissolved in 1 mL n-hexane. The cleaned extracts were analyzed for the 16 representative PAHs using a Shimadzu GS/MS QP 2010 model. Helium gas was used as the carrier gas with a constant flow rate of 1 mL/ min, HP-1 ms column (30 m x 0.25 µm 0.25 mm ID), injection mode was pulsed splitless, volume of extract injected was 1 µL, injection port temperature was 290 0 C, pulse pressure and flow were 35 psi (0.5 min) and 20 mL/min (2 min), respectively; solvent delay was 5 min, initial oven temperature and hold time was 50 0 C (1 min), ramped at 30 0 C/min to 280 0 C and 15 0 C/min to 310 0 C with final hold time of 4 min. External calibration using PAHs standard was used for analytes quantification, while identification was based on retention time. The quantification limit of the PAHs in the standard and the samples was 0.001 ppm. The average response factor for the weight ranges were calculated and used for sample quantification. The concentration of each analyte was determined by calculating the amount of analyte

Research
Where; C f i is the contamination factor of each metal. 40

Potential Ecological Risk Index
In this study, a simplified approach to risk assessment based on comparison of the measured level of contamination in the soil of the studied sites with the background value from the control sample was adopted. 49 Although the ecological risk index (RI) is primarily intended by Hakanson to express the ecotoxic potential of increased concentrations of toxic metals such as arsenic, Cu, Ni, cobalt, Pb, Cd, and mercury in consumable fish, it can also be applied for the assessment of the potential risk from toxic substances to biota and non-human biota in other similar media such as contaminated soils. 39, 52 We used the RI introduced by Hakanson to characterize the metal contamination of each sample in terms of their potential ecotoxicity using the Equations 4 to 6.

RI = ∑ n i=1 Eir
where; Fi is the single metal pollution index; Cs is the concentration of metal in the samples; Cr is the reference value for the metal; Eir is the monomial potential ecological risk factor; Tir is the metal toxic response factor according to Hakanson, and Zn = 1< Cr = 2 < Cu = Ni = Pb = 5 < As = 10 < Cd 30. 39, 53 The ecological risk index is the potential ecological risk caused by the overall contamination categorized in the four classes as shown in Table   solution added, and swirled to mix, 15 mL of concentrated sulphuric acid was added gently and mixed. The flasks were allowed to stand for 30 minutes. Five drops of ferroin indicator was added and the resulting mixtures were titrated against ferrous ammonium sulphate until color change from blue green to violet red was observed. Total organic carbon was determined using an appropriate mathematical expression and multiplied by a factor to obtain the total organic matter (TOM). 22,24

Soil Contamination
The degree of contamination of the dumpsite and the control site soils was evaluated using four indices. where,

Geo-accumulation Index
Cn is the measured concentration of a particular metal in a particular soil sample; Bn is the geochemical background value in average shale of element n and 1.5 is a background matrix correction factor, accounting for lithogenic effects. 46 We then classified each I geo using Forstner et al. descriptive categories: <0, unpolluted; 0-1, unpolluted to moderately polluted; 1-2, moderately polluted; 2-3, moderately to highly polluted; 3-4, highly polluted; 4-5, highly to very highly polluted, and >5, very highly polluted. 47

Contamination Factor
The contamination factor (C f i ) was used by Hakanson to assess soil contamination by comparing the contaminant concentration in the surface layer to a background value. 39, 48 We used a modified C f i formula, using metals concentrations in the control samples instead of background values, which are currently lacking for Nigeria. 49 It is expressed using Equation 2.

Equation 2
Contamination factor, where, C f i = contamination factor; C i 0-1 = mean concentration of each metal in the soil; Cn i = baseline or background value (concentration of each metal in the control sample was used); n = number of analyzed elements; i = ith element (or pollutants). We then classified the C f i using descriptive categories: C f i < 1, low contamination; 1 ≤ C f i < 3, moderate contamination; 3 ≤ C f i < 6, considerable contamination; and 6 ≤ C f i , very high contamination.

Pollution Load Index
Pollution load index (PLI) was also used to assess the metal accumulation and multi-element contamination resulting in increased overall metal toxicity. 50 Heavy metal contamination is associated with a mixture of contaminants rather than one metal contaminant. 51 The higher the pollution load index, the more serious the heavy metal accumulation in the soil. 50 We used the PLI to characterize the aggregate contamination of the six target metals using Equation 3.

Equation 3
PLI Adeyi, Oyeleke Research 1. The potential ecological risk caused by Cu, Zn, Cd, Pb, Ni and Cr on the e-waste dumpsite soils in Lagos and Ibadan were calculated based on the potential ecological risk factor (Eir). The ecological RI value characterizes the sensitivity of the local ecosystem to the pollutants i.e., metals, and represents the ecological risks resulting from the overall contamination. The overall RI was calculated as the sum of all the four risk factors.

Statistical Analysis
Obtained data (i.e., soil properties, metals concentrations and total concentrations of PAHs) were subjected to descriptive statistics and Pearson's correlation coefficient to determine whether there were significant relationships between total PAHs, metals concentrations and soil properties. The statistical analysis was performed using Statistical Package for Social Sciences (SPSS) version 16.0.

Soil Characteristics and Total Metals Concentrations
The pH of topsoil (0-15 cm) ranged from 5.77-5.80 and 5.84 -6.30, respectively, in samples collected in Lagos (LSS) and Ibadan (ISS), while total organic matter ranged from 8.32-8.85% and 3.27-8.65%, respectively ( Table 2). In general, e-waste dumpsite soils were more acidic than the control soil. This might be attributed to the parent material and burning of wastes on the dumpsites. Among dumpsite soils, the Lagos samples were more acidic, with high TOM compared to Ibadan samples. Metals concentrations across dumpsites varied widely. Topsoil Pb ranged from 193-2,240 mg/kg in Lagos and 246-2,090 mg/ kg in Ibadan, while Cu ranged from 50.5-5,390 mg/kg and 79.3-1,150 mg/kg, Zn ranged from 220-1930 mg/kg and 27.5-3420 mg/kg, Cd ranged from 0.43-5.85 mg/kg and not detectable-6.50 mg/kg, Ni ranged from 11.0-51.5 mg/kg and 27.7-128 mg/ kg, and Cr ranged from 108-118 mg/ kg and 94.0-325 mg/kg, in Lagos and Ibadan, respectively ( Table 2). Metals concentrations in the control sample were generally lower than what was detected in e-waste dumpsite soils by over one hundred orders of magnitude in some metals, which may be at least partly explained by e-waste burning activity at the dumpsites. Most heavy metals determined in soils collected from Wenling, an emerging e-waste recycling city in Taizhou Table 2

Metals Contamination Indices
The I geo analysis showed that the soil of LLS 1 was very highly polluted with Pb and Cu, moderately to highly polluted with Zn and Cd, and unpolluted with Cr and Ni (      Abbreviations: Eir, ecological risk factor; RI, risk index  Abbreviations: Eir, ecological risk factor; RI, risk index

Research
The ratio of PAH profiles maybe used to track their origin as petrogenic, biogenic and pyrogenic sources. Poland) classification showed that soils with total PAH < 1,000 μg/kg dry weight (dw) can be considered to be unpolluted. 59 The total PAHs concentrations of all of the samples in the dumpsite soils and the control exceeded the typical concentration of arable topsoil (around 200 μg/kg) in Sweden. 60 The target established by the Dutch government for PAHs in uncontaminated soil is 20-50 μg/kg (dw). 61 The total PAHs concentrations at depths of 0-15 cm and 15-30 cm in all the e-waste dumpsites in Lagos and Ibadan exceeded the 50 μg/kg limit. Thus, all of the study sites were considered to be highly polluted by PAHs.
anthracene to benzo(a)anthracene plus chrysene and indeno(1,2,3-cd) pyrene to indeno(1,2,3-cd)pyrene plus benzo(g,h,i)perylene were also used for source identification. 19,30,65, 66 The ranges obtained were 0.43-0.50, 0.44-0.50; 0.46-0.55, 0.52-0.55; 0-1.0, 0-0.37, respectively, at the 0-15 cm and 15-30 cm levels, respectively, for these PAHs. These values indicated that PAHs had both pyrolytic and petrolytic origins. The results obtained in this study were compared with those in the literature and are presented in Table 10. It was reported that total PAHs in soil collected from Wenling, an emerging e-waste recycling area in Taizhou, China ranged from 371.8 to 1231.2 μg/kg, and relatively higher PAHs concentrations were found in soils taken from simple household workshops. 19 Adeyi, Oyeleke

Discussion
Migration of Cd from topsoil to the subsurface soil was observed in both the Lagos and Ibadan dumpsites. In most cases, Cu, Zn and Pb concentrations were highest in topsoil, which was evidence of recent/ anthropogenic contamination, but with limited evidence of migration to the subsoil. 72 This indicates that there is little risk of groundwater contamination at these sites. All of the e-waste dumpsites in Lagos and Ibadan exhibited multi-element contamination from anthropogenic inputs, most likely from e-waste burning activity. The indices of potential ecological risk were found in the following order at the different sites:

Statistical Analysis
Statistical analyses of the results obtained in the e-waste dumpsites in Lagos and Ibadan using Pearson's correlation coefficient (Tables 11 and  12) showed very strong and negatively significant correlations between total PAHs versus Cd (r = -0.955, p< 0.05), Ni (r = -0.973, p< 0.05) and TOC (r = -0.899, p< 0.05) in Ibadan, suggesting that these contaminants might have originated from similar sources, such as burning of e-waste at dumpsites. There was no significant correlation between total PAHs and TOC (r = -0.395, p< 0.05), and no significant correlation with most of the metals except for Zn (r = 0.648, p< 0.05) in soils of e-waste dumpsites in Lagos, suggesting different emission sources. The PAHs profile pattern in soils of four e-waste dumpsites in Lagos and Adeyi, Oyeleke       Our work shows that improper e-waste handling at these sites may contribute additional metals and PAH contamination and highlights the need for regular soil monitoring at major dumpsites in Nigeria.

Copyright Policy
This is an Open Access article distributed in accordance with Creative Commons Attribution License (http://creativecommons.org/ licenses/by/3.0/).