Microplastics pollution indices of bottled water from South Eastern Nigeria

ABSTRACT The pollution indices of microplastic content in bottled water from South Eastern Nigeria were evaluated. Microplastic contamination factors, pollution load and risk index, polymer risk indices and estimated daily intake for adults and children were determined. Microplastic concentrations, types and shapes were assessed using scanning electron microscopy (SEM) – energy-dispersive X-ray spectroscopy (EDS). From documented literature, this is the first study on microplastics in bottled water from Africa. The microplastics were found in 92% of the samples. The four types of microplastics were polyethylene (PE), polyvinyl chloride (PVC), polyethylene terephthalate (PET) and polydimethyl siloxane (PDMS) while the dominant shapes were fragment, film and pellet/granule. Principal component analysis shows strong association of the concentration of microplastics with total solids and total suspended solids. Pollution risk indices show medium pollution risks of microplastics in bottled water. The estimated daily intake was generally low and therefore shows no risk from daily consumption but reveals a higher intake of microplastics for children than adults. The source of the MPs was attributed to leaching from the packaging material, which should be monitored by appropriate regulatory bodies.


Introduction
Microplastics are tiny plastic particles less than 5 mm long.These particles have various forms, sizes and levels in soil, water, air, food and other distant areas.Around 8 million tons of plastic is discharged into the sea each year.From this number, 1% consists of little plastic trash alone which will likely double by 2030 if no action is taken [1].Microplastics have been found in different conditions, yet additionally in staples like fish [2], salt [3] and drinking water, both from the tap and packaged [4,5].
Microplastics are classified as essential and auxiliary.Essential microplastics are delivered and implied for outside human use.Auxiliary microplastics come from the breakdown of bigger plastic items through regular enduring cycles after entering the climate.Some microplastics come in as crude materials from different items; some happen because of the breakdown of enormous sizes of plastics, while others are found in tyres, materials, paint, ropes and waste treatment.Wastewater and sewage treatment plants are the wellsprings of essential microplastics, while the wellsprings of auxiliary microplastics are huge plastics, clothing, fabricated products, the travel industry, delivery and cataclysmic events like flooding [6].Microplastics are carried away by the wind from far away and easily dispersed within the environment [1].
Adsorption to the outside of plastics by synthetic compounds, and in certain blends of substances and sorts of plastics lead to higher fixations inside the plastics when contrasted with the general climate.The essential dangers of microplastics (MPs) to environments are their omnipresence and bioavailability for ingestion, snare or inward breath [5].There are numerous reports of their ingestion by marine, soil creatures and plants from dirtied soils [7].Human exposure to microplastics could prompt oxidative pressure, DNA harm and irritation, among other medical issues.Especially, when aggravation becomes constant, this can lead to intense medical conditions.Some potential health impacts that might be connected to groupings of ingested microplastics include metabolic interruption, invulnerable brokenness, neurodegenerative illnesses, and constant irritation, which can prompt malignant growth.Microplastics can act as carriers of other molecules that stick to them.Microplastics in drinking water pose human health risk.The perils related with microplastics come from particles that present an actual risk, synthetic substances and microorganisms known as biofilms that may connect and colonise on them [8].
People are unaware of the presence of microplastics and their effects on the environment over a long period of time.Microplastics enter the ecosystem through anthropogenic processes.They also enter freshwater environments from surface run-off and wastewater effluent as well as from consolidated sewer floods, debased plastic wastes and climatic processes.Sources adding to the presence of perilous substances in microplastics include those substances added to the plastics or utilised as crude materials for the creation of plastics, and those in the environment adsorbing to the outer part of the plastic particles which, over the long haul, might be ingested into the plastic network.Plasticisers, for example, phthalates and chlorinated paraffins just as added biocides are likewise present in microplastics found in the environment.It is undeniably true that hazardous substances are spread by means of plastics.
Various studies have been done on microplastics in different matrices [9].The vast majority of these investigations were outside Nigeria.In drinking water from groundwater sources, Mintenig et al. [4] tracked down a limit of 7 MP for every m 3 while Weber et al. [10] did not identify any MP (≥10 um).Mason et al. [11] in the US obtained 325.33 MPs per g/L/ m 3 of microplastics in bottled water.Microplastic contamination was observed for 93% of the bottles.Different investigations have been completed in Lebanon, Mexico, Thailand, the USA, Nairobi, India, Brazil, Indonesia, Germany [7,11] and China [12].However, more data is needed on microplastic content in drinking water for a better understanding of the potential exposure and health risk assessment.However, based on reviewed literature, there is currently no documented study of microplastics in bottled water in Africa including Nigeria.Therefore, this study aimed to investigate (i) types, composition and levels of microplastics in five brands of bottled water in South eastern Nigeria, ii) effects of the physico-chemical properties on the levels of microplastics (iii) pollution risk indices of microplastics.

Sample collection and physicochemical analysis
Five brands of plastic bottled (750 mL) water samples were purchased from major stores in Anambra (A-E), Imo (F-J), Enugu (K-O), Abia (P-T), and Ebonyi (U-Y) states in South Eastern, Nigeria.Four batches of each brand were purchased from each state.A total of 100 samples were obtained and analysed immediately after purchase.
pH and electrical conductivity were determined with a Hanna pH meter and Consort digital conductometer respectively while total solids and total suspended solids were determined by gravimetry.

Quality assurance/quality control to prevent particle contamination
A long sleeve laboratory coat of 100% cotton and particle free gloves were worn during the filtration and handling of samples in the laboratory.Hands were washed with detergent and rinsed thoroughly with deionised water.All glass vessels were checked for cracks, cleaned with detergent and rinsed with deionised water to ensure they do not also contribute to particle contamination.
Before filtration of water samples, the plastic bottles were also cleaned in order to prevent the risk of contamination from outside.Deionised water was also filtered and used as a blank sample to account for background or laboratory contamination.

Filtration
The bottled water samples were prepared using Nile red dye (Loba Chemie).Nile red are highly photostable and fluorescent organic dyes from the benzo[a]phenoxazine family.It was used as a result of its absorption affinity to plastics but not naturally occurring materials and making smaller particles to be detected by fluoresces under specific wavelengths of light [13].Methanol was used to prepare the Nile red to 1 mg/ml to yield a working solution of 10 ug/ml and used for each sample.
0.75 L of each of the samples were vacuum filtered sequentially through a cellulose filter paper with a nominal pore size of 11um (Whatman No. 1, Catalogue No. 1001110, UK) with the aid of a suction pump.After the filtration, 10 ml of Nile red solution was added to stain the filter paper and allowed to dry.It was then kept in sterile petri dishes of 90 mm diameter below room temperature away from light and later viewed.The filtration set-up was always covered with a watch glass to avoid external contamination of samples.After filtration, the filter papers were kept inside the sterile petri dish, capped and left in a dark cupboard before proceeding to particle counting and identification.

Counting and identification of microplastics with optical microscope and scanning electron microscopy (SEM) -energy dispersive X-ray spectroscopy (EDS)
Preliminary analysis of the samples via visual inspection was initially done to identify any dubious microplastics using an Olympus optical microscope Model 0706962 with a camera.Through a red-light shielding plate, the filter papers were each placed in an uncovered petri dish on the stage of the microscope with an objective lens of 40-fold magnification.Microplastics were visually identified and characterised by handling the fluorescence particles (enhanced by the Nile red) for resiliency with a sharp pointed tweezer [14].All particles were counted, sorted and photographed.All sorted particles and plastic bottles were capped in sterile petri dishes and then analysed with SEM-EDS.The samples were mounted on stubs with adhesive carbon and coated in 20 nm gold with a Quorum SC 7620 mini sputter coater, and then analysed with a Zeiss EVO LS10 SEM equipped with an Oxford XMax 50 Silicon Drift Energy Dispersive X-ray detector at 10KV under high vacuum.
Pictures of the isolated microplastic particles were taken and classified into three dominant shapes: fragments, film and granules/pellets, and four types: polyethylene (PE), polyvinyl chloride (PVC), polyethylene terephthalate (PET) and polydimethyl siloxane (PDMS), based on their physical properties.

Microplastics contamination factors and pollution load index
The microplastics contamination factors (MPCfs) and pollution load index (MPPLI) in the bottled water were estimated as described in previous studies [15,16].The MPCf refers to the contamination of MPs in the studied bottled water compared to the background values.The MPCf is known as a standardised monitoring and assessment approach for determining the level of contamination between different samples.The MPCfs were categorised according to Verla et al. [17].Values with MPCf<1 are low contamination, 1≤ MPCf<3 are moderately contaminated, 3≤ MPCf ≤6 are considerably contaminated and MPCf ≥6 very highly contaminated.The MPCf and MPPLI were mathematically computed using equations ( 1) and (2).Where MP i is the quantity of MPs in sample i while MP b is the minimum baseline concentration taken from the lowest MPs abundance recorded in the study of Mason et al. [11] as it shares similar environments and analytical context as this study.

Microplastics polymer risk indices and pollution risk index
The MP polymeric risk indices and pollution risk for the different bottled branded water and also for the entire study area were computed following the description presented previously [15].The equations for computing the polymer risk indices (H i ) for the different samples and index (MPR area ) for the entire area (Southeast, Nigeria) are presented in equations ( 3) and ( 4).The hazard scores assigned based on their toxicity levels to ecosystems presented by Lithner et al. [18] were used in the computation of polymeric risks such as the chemical toxicity coefficient or risk score (S j ) for the identified MP polymers in the samples.The hazard scores for the polymers identified in the samples were PE = 11; PET = 4; and PVC = 10,551, while PDMS was not available and was excluded in the calculation.However, P ji is the number of each single MPs polymer identified in sample i and the MPR area is computed as the n th root of the polymer risk indices products.

Estimated daily intake
An individual risk pathway as a result of human exposure to microplastic contamination of drinking water could be through oral ingestion.Therefore, the estimated daily intake (EDI) due to exposure to overall MPs resulting from ingestion of contaminated water is determined using equation (7) where, EDI q : estimated daily intake of MPs based on quantity (EDI q ) through ingestion of the bottled water (particle/0.75/BWday);MP i : average quantity of the MPs in bottled water (MP particle/0.75L); RI: ingestion rate (2.2 L/day for adults; 1.8 L/day for children); B W : average body weight (70 kg for adults; 15 kg for children) as described in earlier reports [19][20][21].

Statistical analysis
Descriptive statistics was done using Excel 2016.The data were analysed using SPSS 24 software package.A one-way ANOVA was done to detect any significant differences (P < 0.05) between the brands of the plastic bottled water and the states where they were purchased.Nearest Neighbour Analysis (NNA) was conducted to evaluate the spread of MP data and the degree to which a collection of samples are clustered or evenly spaced.The city block metric was applied for the distance computation and a predictor factor (k) was set at 1. Principal component analysis was done on the data for MPs and physico-chemical properties to find any association.As recommended by the Kaiser criteria, varimax rotation and retention of main components with eigenvalues greater than one were utilised.

Shape of MPs in bottled water
The quantities of microplastics (MPs) shaped in bottled water from South eastern Nigeria are presented in Table 1 while the distribution for the different samples is presented in Figure 1.Three dominant shapes (fragment, film and pellet/granule) of MPs with the following total quantities 42.83 MP particles/0.75L, 1.16 MP particles/0.75L and 10.82 particles/0.75L, respectively, were generally identified and presented in Figure 1.The distribution of particles based on shapes were fragments (77.94%) > granules/pellets (19.92%) > film (2.14%).High quantity of fragments was also reported for bottled water by Mason et al. [11].By states, film was not found in samples from Imo, Abia and Ebonyi states.Enugu showed the highest distribution for all shapes followed by samples from Abia (for fragments: 9.33 MP particles/0.75L) and the lowest generally recorded in Imo state samples (Figure 2).

Quantification of MPs
The quantities of MPs identified in the bottled water samples after removing any possible background contamination are summarised in Table 1.Of the 100 bottles of water analysed, eight (i.e 8%) showed no contamination by MP indicating that the     remaining 92 (92%) bottled water samples tested showed some level of contamination by MPs.This observation is in agreement with previous reports of Mason et al. [11], who reported that 7% of 259 branded bottled water from many world countries analysed were not contaminated by microplastics.The sources of MPs in the bottled water could occur during bottle cleaning, filling and capping processes.Weisser et al. [22] showed that MPs were present in all steps of mineral water bottling, which increased more during bottle capping.Additionally, carbonisation and the bottle age seem to play a role in MP load in the bottled water [22].The distribution of MPs in different bottled water samples is presented in Figure 3.The sizes of MPs were generally between 20 and 100 µm (Figure 4).The highest and lowest MPs were found in samples N (6.67 ± 5.51 MP/0.75 L) and G (0. Similarly, in re-used PET and glass bottles, Oßmann et al. [5] observed MPs of 4889 ± 5432 MP/L and 3074 ± 2531 MP/L, respectively.The high load of MP in these studies is mainly due to the cleaning process of re-usable bottles and the analytical methods adopted.Schymanski et al. [7] applied the micro-Raman spectroscopy, which is very efficient in analysing much smaller particles.However, Mason et al. [11] who applied Nile Red staining also used in this study, obtained some results that were comparable with this study for particles >100 µm. In order to determine the MP data spread and evaluate the degree of association between samples, the Nearest Neighbour Analysis (NAA) was computed.The nearest neighbour index quantifies the degree to which characteristics are spatially dispersed over a given distance.It is a numeric number between 0 and 2.15 that indicates whether the samples are clustered (0 to less than 1), random (1), or regular (greater than 1 to 2.15).
The nearest neighbour quadrant mapping for the MPs in different bottled water samples is presented in Figure 4 while the distance between nearest neighbours is presented in Table 2. Based on these distances, all samples showed clustered mapping (0 to 0.222) indicating that the presence of MPs in the bottled samples may come from similar source (s), potentially due to leaching from the packaging material.

Identification of MP types in bottled water
For the identification of MP types in the bottled water samples, the EDS analysis was conducted on the particle identified by the SEM.The EDS has been adopted in studies of MPs [23][24][25] and have been adjudged to be a very powerful method of analysing the composition of microplastics [26,27].The method enables the differentiation of MPs [rich in carbon (C), chlorine (Cl), silicon (Si), sulphur (S) and titanium (Ti)] from natural materials due to simultaneously collected images and elemental mapping.The results of the EDS analysis for all samples are presented in Figure 5(I-IV).The EDS was compared with EDS spectral data of pure PET (plastic bottle of water samples) and those reported in other studies (Figure S1), to be able to identify the type of polymer.
In sample A, all MPs identified were assigned as PVC due to the high intensities of chlorine (Cl).The EDS spectra further corresponded with the EDS spectra for pure PVC reported by Hadi et al. [28] (Figure S1).Similarly, sample B also showed a correlation with PVC while some MPs also in sample B showed EDS spectra, which suggest that they were made up of PET, which also corresponded with the EDS spectra for PET collected from the   environment reported by Fotopoulou and Karapanagioti [24], Wang et al. [25] and Kohutiar et al. [29].In Sample C and E (Figure 5I), MPs showed EDS spectra corresponding to a silicon-based polymer (PDMS: polydimethyl siloxane) due to the high intensities of silicon (Si).PDMS are used in the production of functional additives such as plasticisers and lubricants used in the production of plastics (e.g.PVC) to improve the flexibility, durability and stretchability of polymeric films, reducing, at the same time, melt flow [30].Therefore, the presence of the polymer in the bottled water could be from the plastic bottle container through leaching processes.This has been demonstrated by Dopico-Garcııa et al. [31].The MP particles in sample D showed EDS spectra, which were assigned as PE due to the high carbon content (~93%) and also corresponded to the EDS spectra of pure industrial PE reported by Liço et al. [32].Overall, bottled water samples from Anambra state showed contamination by PVC, PET, PE and PDMS types of MPs.Similar results were also obtained for bottled water samples (P to T) collected in Abia state and samples (U to Y) collected in Ebonyi state.Bottled water samples (F to J) from Imo State showed the two types of MP composition, viz., PDMS and PE.Samples H and J do not contain MPs, and the particles observed (Figure 4II) were considered as sand particles due to EDS spectra showing richness in aluminium (Al) and neodymium (Nb) (Table 1) and do not correspond to any of the reference EDS spectra for the polymers.In bottled water samples (K to O) from Enugu state, PE and PET MPs were recorded.The presence of inorganic elements such as aluminium (Al), calcium (Ca), magnesium (Mg), sodium (Na), oxygen (O) and zinc (Zn) in all samples is most likely from additives added during plastic production.Since MP can potentially adsorb elements on its surface, the EDS analysis was conducted from inside the particle to be certain it is from additives.For example, compounds of Ca and Zn present in samples A and B are commonly used as heat stabilisers in the production of PVC [30].The strong nitrogen peaks appearing on the surface of PE (samples B and L) and PET (sample U) fragments can be used as a proxy for biomass, indicating potential interactions between MPs and organic materials [27,33].Organic materials such as nitrates have been detected in bottled water sold in Nigeria [34].
In all 92% of samples, four MP types were generally identified, i.e.PVC, PE, PDMS and PET.The distributions of these types are presented in Figure 6.The distribution followed the order PET (45.08%) > PE (24.79%) > PDMS (17.90%) > PVC (12.23%).The high abundance of PET could be due to the packaging bottles as they are made of PET.This is in agreement with the results of Schymanski et al. [7], but in contrast to the study of Mason et al. [11] that reported PP to be the most abundant with 54%.

Microplastics contamination factors (MPCf) and pollution load index (MPPLI)
The results for the computed microplastics contamination factors and pollution load index are presented in Figure 7.Following the classification of MPCf, forty percent (40%) of all samples showed low contamination, 32% showed moderate contamination, 20% showed considerable contamination while the remaining 8% were high.The overall pollution load was 1.71 which is >1, indicating pollution by microplastics for all bottled water samples analysed in the South east area.However, by states, MPCf followed the decreasing order of contamination as Enugu > Anambra > Abia > Ebonyi.Contamination of drinking water by microplastics is undesirable as they are capable of accumulating in the body, causing deleterious effects [35].

Microplastics polymer risk indices and pollution risk index
Results for the MP polymer risk indices (MPR) and pollution risk index are presented in Table 3. MP polymeric risk indices quantify the risks the MP pose to health and ecosystems based on its composition.According to Kabir et al. [16], the classification for the polymeric risk indices (H i ) is given as low when <150, medium when 150-300, considerable when 300-600, high 600-1200 and very high when >1200.Following the classification, only samples from Anambra state showed high risks with MPR of 926.93.This is due to the presence and high hazard scores of PVC [17].However, for the entire Southeast, Nigeria, the MPR showed low polymeric risks (22.38).
For MPs pollution risk index (MPRI), it is classified as follows: Class I when <10, class II when 10-100, Class III when 101-1,000, Class IV when 1001-10,000 and Class V when >10,000 [16].Only samples from Imo and Ebonyi states with MPRI of 1.62 and 5.70, respectively, belong to Class I (low pollution risk category).Medium pollution risk (Class II) was shown by samples from Enugu (23.55) and Abia (73.36), while high pollution risk (Class IV) was shown by samples from Anambra state (1997.00).For the entire study area, the MPRI was 46.13, which indicates medium pollution risks.So far, none of the published studies applied chemometric models in evaluating MP pollution in drinking water.The absence of MP pollution assessment using models in previous studies poses difficulties in comparability of the results with this study.

Estimated daily intake
The results for computed estimated daily intake of MPs through consumption of the studied bottled water for adults and children are presented in Table 3.All EDIs are generally less than 1, indicating a low daily intake of MPs and therefore may pose no risk from consumption daily.However, results generally reveal a higher intake of MPs for children than adults.Studies have generally suggested a higher intake of MPs by children compared to adults [36].Information on the risks of MPs to children as well as adult health is still very unclear.Other than exposure, the fate and transport of ingested MPs in the human body, which incorporate intestinal assimilation and biliary discharge, have not been tended to in the previous investigation and remained generally obscure.Amassing of MPs in the body tissue could cause actual pressure and harm, irritation, oxidative pressure, and insusceptible reactions.Up to this point, impact studies looking at MPs on human cells discovered little proof of effect on cell practicality  [37][38][39].Nonetheless, it is questionable if the scope of exposure concentrations utilised in such investigations is really illustrative of the MP gathered in body tissues.Therefore, an urgent action is needed to evaluate the health implications of these MPs when they enter the human body.However, it is assumed that with other toxic chemicals, MPs may result in low intelligence quotient (IQ), stunted growth and organ underdevelopment in children.

Physicochemical properties and relationship with MPs
The levels of pH, electrical conductivity (EC), total solids (TS), and total suspended solids (TSS) in the bottled water are presented in Table 4.The pH ranged from 5.98 ± 0.05 in sample E to 9.35 ± 0.1 in sample U. Thirty-two (32%) of the samples showed pH outside the normal range (6.5 to 8.5) for drinking water, according to the World Health Organization.Only samples Q, R, U and V (16%) showed higher EC values above the stipulated limits of 100 µS/cm for drinking water.All bottled samples analysed showed high TSS values (>50 mg/L), while TS was low (<500 mg/L).Table 5 contains the results of the principal component analysis for physicochemical characteristics and MPs, whereas Figure 8 has the component plot in rotated space.
PCA was carried out to detect the relationship between the data for physico-chemical properties and MPs.The component loadings, as well as the eigenvalues, are shown in Table 4. Two components, however, have initial eigenvalues higher than one (>1) as indicated.The first component is dominated by MPs, TS and TDS (with a high factor loading value of 0.714, 0.947 and 0.940) accounting for 46.22% of the total variance, while the second component is dominated by pH and EC (with a high factor loading value of 0 .868and 0.849) accounting for 30.06% of the total variance.All components represent strong association or relationship.MPs showed a strong association with TS and TSS, indicating that if the bottled water sample contains high TS or TSS, there are likely to be high MPs present in the water.

Conclusion and recommendation
This study evaluated for the first time the quantity and types of microplastics in bottled water brands sold in Southeast, Nigeria.Ninety-two per cent of the bottled water samples showed some level of contamination with microplastics.Fragments were the dominant shape and polyethylene terephthalate was the main type of microplastic.There was a strong association  of total solids and total dissolved solids with the microplastic levels in the water.This is due to the high hazard scores of polyvinyl chloride.However, for the entire samples, the microplastics pollution risks showed low polymeric risks.Generally, EDIs revealed a higher intake of microplastics for children than adults.There are currently no maximum permissible limits for microplastics in bottled water.Therefore, further studies on evaluation of the health effects of various concentrations of microplastics in the human body are necessary.

Figure 2 .
Figure 2. Distribution of MPs shapes in all bottled water samples.Error bars represent 5 % standard error.

Figure 4 .
Figure 4. Nearest neighbour quadrant mapping for the different bottled water samples.The numbers represent different samples i.e A to Y represented with 1-25.

Figure 6 .
Figure 6.Distribution of MP polymer types in bottled water.

Figure 8 .
Figure 8. Principal component plot in rotated space.

Table 1 .
Characteristics of microplastics in bottled water.

Table 2 .
Nearest neighbours and distances between samples.

Table 3 .
Microplastics polymer risk indices, pollution risk index and estimated daily intake.

Table 4 .
Physico-chemical properties of bottled water.

Table 5 .
Total variance of physico-chemical properties and MPs concentration in bottled water.