International Journal of Earth Science and Geophysics
(ISSN: 2631-5033)
Volume 9, Issue 2
Research Article
DOI: 10.35840/2631-5033/1870
Article Formats
Geochemical Modelling for Determination of the Origin of High Acidity and Iron Content in the Deep and Shallow Aquifers in Nsukka Area, South Eastern Nigeria
Author Details
Ifeoma M Ugwu1* and Smart C Obiora2
1Department of Geology and Mining, Enugu State University of Science and Technology, Agbani, Enugu state, Nigeria
2Department of Geology, University of Nigeria, Nsukka
Corresponding author
Ifeoma M Ugwu, Department of Geology and Mining, Enugu State University of Science and Technology, Agbani, Enugu state, Nigeria.
Accepted: August 12, 2023 | Published Online: August 14, 2023
Citation: Ugwu IM, Obiora SC (2023) Geochemical Modelling for Determination of the Origin of High Acidity and Iron Content in the Deep and Shallow Aquifers in Nsukka Area, South Eastern Nigeria. Int J Earth Sci Geophys 9:070.
Copyright: © 2023 Ugwu IM, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Abstract
Geochemical modelling of groundwater in the deep and shallow aquifers in Nsukka area was carried out in order to provide insight on the mineral-water interactions which could have generated high acidity and iron content in the water chemistry. The study involved analyses of the major and minor ions in water samples from springs, deep and shallow wells, assessment of their suitability for domestic purposes and geochemical modelling to determine their evolution. The order of abundance of the major cations was Mg > Ca > K > Na and that of the major anions was NO 3 > Cl > SO 4 > HCO 3 . The average concentration of Fe ranges from 0.15 to 3.9 mg/L, concentration of Al ranges from 1.8 to 3.9 mg/L while that of Mn ranges from 0.1 to 2.8 mg/L. The pH values of the water samples are mostly acidic. The major ion concentrations are within the WHO guidelines for drinking water. However, high values of Fe and Mn observed in the samples exceed the WHO limits suggesting that these waters are not good for drinking. The geostatistical models indicate that the water type is of calcium chloride with same origin. The results of the inverse modelling of water with PHREEQC reveal that these waters originated from dissolution of halite, alunite, pyrite, pyrolusite, goethite and dolomite, and precipitation of sylvite and anhydrite. However, Saturation Indexes (SI) calculation indicate that the water samples are supersaturated with alunite, gibbsite, goethite, sideriteand undersaturated with anhydrite, dolomite, halite, pyrolusite and sylvite. High iron content of the waters is considered to have originated from dissolution of goethite, which is the major Fe-bearing mineral in the laterite in the overlying layer that dominated the whole area. Alunite is a mineral found to be dissolving in all the models, and this leads to precipitation of gibbsite and decrease in pH. Modelling revealed that these waters can be remediated through precipitation of goethite and pyrolusite by increasing the pH through addition of a definite amount of slaked lime (Ca (OH) 2 ) to prevent calcite and gypsum precipitation as well as very high pH.
Keywords
Inverse modelling, High acidity, High iron, Goethite, Alunite
Introduction
The final composition of water depends on rock types, land use (agricultural or industrial), presence of diffuse or conduit flow path, infiltration rate, climatic conditions, residence time and mechanism of recharge. These factors often determine the presence or absence of contaminants in groundwater. For example, water with high infiltration rate through a porous media usually has low number of dissolved ions. Reconstruction of the processes that gave rise to any water quality requires inverse modelling with known solution and mineral phases. This involves combination of mineral phases, initial and final solution to explain chemical changes such precipitation and dissolution.
Inverse modelling with PHREEQC (PH Redox Equilibrium by C language) have been carried out by many researchers such as Uliana and Sharp [1], Eary, et al. [2] Eraifej [3], Mahlknecht, et al. [4], Machado, et al. [5] Sharif [6], Peikam and Jalali [7] have used inverse modelling for assessment of evolution of waters. Both inverse and forward reactions are used to predict the composition of water that evolved as a result of dissolution and weathering of rock as water flows through the path [8]. However, chemical compositional changes that occurred as water moved along the flow path are determined using inverse modelling [9]. This involves determining different moles of phases that accounts for changes in composition between initial and final water chemistry [10].
Nsukka area does not have recent or historical mining activities, however, the chemical composition of water from both deep and shallow aquifer indicate that high iron concentration and acidity [11-13]. Low pH can be attributed to the influence of acid mine drainage [14], organic matter in soil [15], geology or soil type, landfill leachate, in-stream redox processes, dairy runoff, acidic precipitation, industrial effluents, filamentous algae [16]. The high iron concentration in the groundwater in Nsukka is due to the presence of organic matter and subordinately reduction of ferric hydroxides resulted to low pH [13]. Uzoije, et al. [12] suggested the oxidation of ferrous to ferric iron as a remediation method to remove the high iron content. However, geochemical modelling is required for accurate determination of the sources of low acidity and high iron contents of the waters from these aquifers and suggestion of the remediation.
To this end, this paper aims to investigate the sources of high acidity and high iron content of the waters from springs, shallow and deep wells in Nsukka area through inverse modelling with PHREEQC and suggest the remediation method.
Materials and Methods
Site description
This investigation integrates both field mapping, water sampling, geostatistical methods using (Stiff and Piper diagram) and inverse geochemical modelling with PHREEQC for windows version 1.5.08 [10]. The investigated area is located between latitude 7°18'0"E and 7°22'0"E and latitudes 6°50'0"N to 6°55'0"N. The area is located at north-eastern part of Enugu town (Figure 1) and is situated in Nsukka in Enugu State, Nigeria. In addition, water sample results used in PHREEQC modelling obtained from the works of Onunkwo, et al. [11] and Ozoko [13] extends up to latitude 6°45’ and 7°00’ N and longitude 7°15’ and 7°30’E.
Hydrogeological description and water sampling
The spring water from perched aquifers serves as source of drinking water for rural areas. The study area is located near Nsukka town. The aquifer type in Nsukka area varies from unconfined, semiconfined to confined with average transmissivity of 3.25 × 10 -2 m 2 /s, hydraulic conductivity of 2.3 × 10 -3 m/hr and Specific discharge is 2.24 × 10 -4 m/yr, and average linear velocity is 4.98 × 10 -4 m/yr [12]. Ten water samples were collected from the springs in the area. The water samples were collected with clean dry 500 ml plastic bottles which have been initial rinsed with clean water. Each container was rinsed with the water to be collected. The water bottles were appropriately labelled with the locations and coordinates. Some physical properties of water including pH, temperature and electrical conductivity were measured in the field features whereas cation and anion identification were determined in the laboratory.
Geochemical modelling
The inverse modelling of the thirty three (33) water samples from the springs, deep and shallow wells were performed using PHREEQC for windows version 1.5.08 [10] in order to determine the mineral phases that led to the evolution of the chemical composition. Ten water chemistry results were obtained from this study, additional eight (8) results were obtained from the work of Ozoko, [13] and fifteen (15) results of deep and shallow wells from Onunkwo, et al. [11]. Water chemistry of any water involves the interaction of rocks and recharge water during infiltration, percolation and storage in aquifer. This interaction can be in form of dissolution, ion exchange, adsorption and precipitation. The solubility of minerals largely governs the process of dissolution and precipitation. The tendency of a mineral to dissolve or precipitate in a water is determined by its saturation indices (SI).
Inverse geochemical modelling involves determining different moles of mineral phases that accounts for changes in composition between initial and final water chemistry [10]. To achieve this, the code calculates mass balance equations that can predict the final water composition starting with initial water composition. The input data are composition of pure water with ambient temperature, pH of 7, major ions that are obtained from chemical composition of the final solution and the software’s solution keywords. Predicated mineral phases based on the existing geology in the area were added. The minerals selected were halite, anhydrite, calcite, alunite, sylvite, pyrite, quartz, dolomite, pyrolusite, goethite as well as O 2 (g) and CO 2 (g) as the gas phase. In addition, forward and mixing modelling were performed to validate the results obtained from inverse modelling.
Results and Discussions
Geology
The geology consists of two Formations: Ajali and Nsukka (Figure 1). The Ajali Formation consists of yellowish-creamy fine to medium poorly to moderately sorted moderately consolidated sandstone with lenses of silt and clay. The Nsukka Formation consists of moderately consolidated mudstone with intercalations of siltstone, sandstone, clay, shale and capped by lateritic ironstone overburden. A conspicuous feature of Nsukka Formation is the pronounced degree of lateritization of sediments, which resulted in the development of rock which is vesicular and vulgar often showing mineralogical banding between limonite, goethite and hematite [17]. The lithologic units have general trend of NE-SW and an average dip of 4 in a south-westerly dip direction. The perched aquifers consist of laterites and conglomerates with thickness range of 2 m to 10 m as shown in the generalized lithosection (Figure 2) of the area represented in Figure 1. These aquifers result to contact springs at the intersection between laterites and clays, laterites and indurated ironstone, laterites and carbonanueous sandy shales.
Water quality of the spring waters
The results of the hydrochemical analyses of the spring water are given in Table 1. Water temperature ranged from 28 to 38 °C. The pH of nine out of ten water samples are below the WHO standard for drinking water and has pH values ranging from 5.3 to 6.9 indication mostly acidic. This is consistent with the pH range of 3.6 to 5.8 of the springs [13] and average pH of 5.8 and 6.3 for shallow and deep well, respectively [11]. The order of abundance of the major cations was Mg > Ca > K > Na. The abundance of the major anions was NO 3 > CL > SO 4 > HCO 3 , with average concentrations of 12.2, 1.63, 1.9, 0.05 mg/L, respectively. The average concentration of Fe ranges between 0.15 and 3.9, concentration of Al ranges from 1.8 to 3.9 while that of Mn ranges from 0.1 to 2.8 mg/L. Nine of the ten samples contain amount of Mn and Fe higher than the Maximum Permissible Limit by WHO. The Al content of the waters ranged from 1.8 to 7.1 mg/L Although, there is no health based standard for aluminium, concentration between 0.1-0.2 mg/l leads to precipitation of aluminium hydroxide floc and excessive staining of water by iron [18]. In human, the development of Alzheimer disease can be triggered by exposure of Al [18].
Positive R-values implies simultaneous increase or decrease in values of both variables; 0.50 ≥ r ≤ 1.0 indicates strong positive correlation, 0.10 ≥ r < 0.49 indicates weak correlation and r = 0 implies no linear relationship between variables. Weak relationship may indicate different sources of introduction of the variables while strong relationship indicates that the source of the ions may have originated from one source. The results show a positive correlation between Electric conductivity (EC) and Cl, as well as (Ca + Mg) content in the water samples. The plot of Electric conductivity (EC) vs. Cl - shows that there are two groups; one end member and mixed (Figure 3).
Ten water samples from different locations belong to one end member while five are mixed samples. The regression coefficient R is 0.835 which indicates a strong positive correlation between the two variables.
The graph of HCO 3 - vs. Cl - has regression coefficient of 0.008 which indicates a weak correlation between the two variables. Hence this shows that the source of HCO 3 - ion is not the source Cl - ion (Figure 4). The plot HCO 3 - vs. Ca 2+ + Mg 2+ has its regression coefficient R as 0.0392 which indicates a weak correlation between the two variables implying that the source of HCO 3 - is not the source of Ca 2+ + Mg 2+ (Figure 5).
Statistical analysis of water quality
One of the geostatistical methods that can be used to confirm the presence of dominant ions are piper trilinear diagram [19]. This is used to determine water type, ion exchange and hydrochemical facies [20]. The trilinear diagrams illustrate the relative concentrations of cations (left diagram) and anions (right diagram) in each sample. For the purposes of a Piper diagram, the cations are grouped into three major divisions: Sodium (Na + ) plus potassium (K + ), calcium (Ca 2+ ), and magnesium (Mg 2+ ). The anions are similarly grouped into three major categories: bicarbonate (HCO 3 - ) plus carbonate (CO 3 2- ), sulfate (SO 4 2- ), and chloride (Cl - ). Piper trilinear diagram (Figure 6) for the study area showed just one type of water, the calcium chlorite water type with Ca and Cl as the major dominant ion.
A Stiff diagram is a graphical representation of chemical analyses [21] and can be used to deduce the source rock. The result revealed plots of similar shapes (Figure 7 to Figure 8), which indicates that all the water samples are from the same origin.
Inverse and forward modeling with PHREEQC
The result of the inverse modelling of water from the springs in the study (Table 2) and those from previous authors (Table 3) reveal that these waters are under saturated with respect to halite, alunite, pyrite, pyrolusite, goethite, dolomite, and super saturated with sylvite and anhydrite. More than 20 models were found in some of the simulations; however, we selected maximum of three representative models. In all the models both sylvite and anhydrite precipitated whereas halite, alunite, pyrite, pyrolusite, goethite, dolomite dissolved in all models. However, one of the models indicate that the water is super saturated with goethite and resulted to its precipitation. Goethite is the major mineral in the laterite that dominated the whole area. In addition, the amount of dissolved goethite in the water from the study ranges from 2.37 × 10 -2 to 2.74 × 10 -7 mmol/kg water whereas that from the previous authors ranges from 1.34 × 10 -5 in Iyi Alumu to 4.64 × 10 -9 mmol/Kg in Ogurugu Shallow well. These amounts are higher than the permissible limit of 0.3 mg/L by World Health Organisation (WHO). Although pyrite occurs as one of the dissolved Fe minerals, the amount of dissolved pyrite is significantly lower than that of goethite (Table 2). Therefore, dissolution of goethite is the reason why there is high amount of iron in springs, shallow and deep wells in the Nsukka area.
Alunite is a mineral found to be dissolving in all the models. Both the springs and wells contain alunite ranging from 2.47 × 10 -3 to 1.30 × 10 -5 mmol/Kg. Alunite (KAl 3 (SO 4 ) 2 (OH) 6 ) is one of the main minerals controlling the mobility of aluminium in natural systems in both non mining and mining environments as well as soils under oxic and acidic conditions [22]. The presence of alunite has been described in mining environment [23] and other acid sulphate environments such as soils [24] and surface of Mars [25]. The dissolution of alunite is slower at pH below 4.6 but increases above pH 4.8 [22]. Acero, et al. [22] noted the reactions are involved during the dissolution of alunite at pH between 5.5 to 8 as follows:
KAl 3 (SO 4 ) 2 (OH) 6 + 3OH - → 3Al(OH) 3 + K + + 2SO 4 2- (1)
4KAl 3 (SO 4 ) 2 (OH) 6 + 15H 2 O + 6OH - → 3Al 4 (SO 4 )(OH) 10 5H 2 O + 4K + + 5SO 4 2- (2)
These reactions lead to the precipitation of gibbsite or diaspore or amorphous aluminum hydroxide (eqn. 1) and felsobanyaite (eqn. 2). The precipitation of these aluminium phases would lead to the consumption of OH - , which may lead to a decrease in pH. This is consistent with the observed pH 3 of Iyi Alumu and precipitation of gibbsite and AlOH 3 as indicated by their saturated indices in Table 4. Saturation indexes indicate oversaturation of alunite, gibbsite, goethite, siderite, hematite and under saturation of anhydrite, calcite, dolomite, gypsum, halite, manganite, pyrolusite, sylvite. In addition, least alunite dissolution rates are observed at pH > 5 - 5.5 [26]. This is consistent with small amount of dissolved alunite at pH 5.3 to 5.78 in water from the study area.
PHREEQC forward model at 90 fold concentration of the water chemistry reveal that all the minerals that precipitated or dissolved in the inverse model also precipitates or dissolved in the forward model (Figure 9). The forward model also indicated that most of the iron in solution is contributed by the dissolution of goethite.
Determination of process that contribute to ions in groundwater
The processes that lead to hydrochemical evolution of groundwater depends on various factors such as origin of the recharge water, geology, resident time for water-mineral interaction. The area under investigation is underlain by sedimentary rocks and has undergone weathering. Sedimentary and secondary minerals such as dolomite, calcite, anhydrite, gypsum, pyrite, alunite, pyrolusite, goethite, hematite etc. were selected as minerals that interact with surface water. PHREEQC model was used to predict the minerals that are saturated, undersaturated and oversaturated in the water as indicated by saturation index (SI) is a factor that indicate. If SI > 0, it means oversaturation of the water by the mineral, SI < 0 indicates undersaturation while SI = 0 means equilibrium conditions. Table 3 Shows saturation indices of some of water from the representative springs, shallow and deep wells.
The Ca/Mg ratio of water samples reveal values less than one indicating the dissolution of dolomite rather than calcite, which is consistent with the result of the inverse modelling. The Al concentrations of the water samples were plotted on stability field diagrams (Figure 10). The Al stability diagram indicate they are essentially in the Alunite stability field, showing that equilibrium with this mineral phase is one of the main processes controlling the water chemistry. Similarly, the Fe falls within the stability field goethite indicating the mineral contributed to the presence of Fe in the water composition.
Prediction of ways to removal of acidity, high Fe, Al and Mn content of waters from Nsukka area through modelling
Different ways of removing metallic contaminants from aqueous solutions methods could be classified into photocatalytic, sorption, electric, membrane and chemical methods [27]. We adopted addition of slaked lime (Ca(OH) 2 ) and modelled the simulation using PHREEQC at different concentrations. This was achieved by using the initial pH of the water, the amount of cation and anions and adding specific amount of slake lime during the modelling are measures taken to avoid calcite and gypsum precipitation. The use of lime precipitation is the most effective conventional ways of treating inorganic effluent with high concentration of metals [28]. For example, 0.173 moles of Ca(OH) of slaked lime is added to neutralise the acidity of Ayaka spring, and allow the water to be in equilibrium with goethite, gypsum and calcite as well as undersaturated with pyrolusite. Different amount of slaked lime, final pH for the springs and Saturation Indexes (SI) of calcite, gypsum, goethite and pyrolusite are shown in Table 5. The modelling results indicate that amount of slake lime does not only depend on the pH but also on the amount of dissolved HCO 3 .
Conclusion
Water samples from springs, deep and shallow wells were collected and analyzed to assess the suitability of water for drinking purposes and determine the origin of high acidity and high iron content. Chemical parameters of groundwater such as pH, electrical conductivity (EC), sodium (Na + ), potassium (K + ), calcium (Ca 2+ ), magnesium (Mg 2+ ), bicarbonate (HCO 3 - ), sulphate (SO 4 2- ), Chlorine (Cl - ), manganese (Mn) and iron (Fe) were determined. The results revealed that the concentration of the major cations are within stipulated limits of World Standard Organization (WHO) whereas iron (Fe), and manganese (Mn) are high when compared with the WHO limits. The Piper trilinear and Stiff diagram plot indicate that the water type is the same origin and has Ca and Cl as the major dominant ion. The results of the inverse modelling of water samples with PHREEQC reveal that these waters originated from dissolution of halite, alunite, pyrite, pyrolusite, goethite and dolomite, and precipitation of sylvite and anhydrite. Saturation Indexes (SI) calculation indicate that the water samples are supersaturated with alunite, gibbsite, goethite and undersaturated with anhydrite, dolomite, halite, pyrolusite and sylvite. High iron content of the waters is considered to have originated from dissolution of goethite, which is the major Fe-bearing mineral in the laterite in the overlying layer that dominated the whole area. Alunite is a mineral found to be dissolving in all the models, and this leads to precipitation of gibbsite and decrease in pH. Modelling revealed that these waters can be remediated through precipitation of goethite and pyrolusite by increasing the pH through addition of a definite amount of slaked lime (Ca (OH) 2 ) to prevent calcite and gypsum precipitation and very high pH.