Bio-Adsorbent Derived from Sewage Sludge Blended with Waste Coal for Nitrate and Methyl Red Removal: Synthesis, Oxidation, Performance and Environmental Consideration

Low-cost bio-adsorbents were synthesized using two types of sewage sludge: D, which was obtained during the dissolved air flotation stage, and S, which was a mixture of 10 primary and secondary sludge from the digestion and dewatering stages. The sewage sludge was mixed with waste coal before being activated with Potassium Hydroxide (KOH) and oxidized with ammonium persulfate (APS). The nitrate and methyl red removal capacities of the synthesized bio-adsorbents were evaluated and compared to those of industrial activated charcoal. The oxidation surface area of bio-adsorbents 15 derived from sludge S shrank by six fold after modification, while those derived from D only varied narrowly from 312,72 m 2 /g to 282,22 m 2 /g, but surface modification had no effect on inorganic composition in either case. The adsorption of nitrate and methyl red (MR) was performed in batch mode, and the removal processes followed the pseudo second order kinetic model and the Langmuir isotherm fairly well. The adsorption 20 capacities of nitrate and MR were higher at pH=2 and pH=4, respectively. The total nitrate Langmuir adsorption potential was DC-5-750 (26,735 mg/g) > commercial activated carbon (Com-AC) (20,61 mg/g) > DC-55-750M1 (17,06 mg/g), and for MR, Com-AC (196,07 mg/g) > DC-5-750M2 (175 mg/g). surface without significantly altering the porous structure and increase surface acidity by increasing carboxylic group presence. There are no studies that we are aware of that use ammonium peroxydisulfate to oxidize activated carbon from sewage sludge blended with waste coal


Introduction
Sewage sludge is an unavoidable by-product of wastewater treatment plants, with treatment costs accounting for 25 % to 65 % of the overall operating costs [1]. South 45 Africa's population has gradually increased over the last two decades, rising from 48.8 million in 2005 to 51.58 million in 2010 to 57.7 million in 2018, with more than half of the country's population now residing in densely populated urban areas [2]. Rapid population growth, combined with rapid urbanization, has put additional pressure on existing wastewater treatment plant (WWTP) facilities, resulting in increased sewage 50 sludge production. Several lines of evidence demonstrate that dumping waste sludge in landfills is not a long-term sewage sludge management solution. Despite the risk of soil and subsoil contamination, nearly 80% of wastewater treatment plants continue to discharge sewage sludge at designated sites [3].
The South African Department of Water Affairs and Forestry has identified five 55 potential applications for sewage sludge: agriculture (fertilizer), on-site and off-site land application (within and outside WWTP boundaries), thermal management activities (e.g. full or partial combustion of organic solids by incineration), and benign land application [3]. The use of sewage sludge fertilizers is restricted due to the difficulties in meeting quality requirements (faecal coliform cap of 10000 CFU/g dry, 60 Helminth Ova maximum of one viable ova/g dry, pollution) [3]. As a result of these restrictions, large quantities of untreated sludge have accumulated on open fields. South Africa produces 673 360 metric tons of sewage sludge annually, of which only about 19 % is recycled and the rest is landfilled, necessitating the development of innovative approaches to reduce sewage sludge emissions [2]. Landfilling as a wastewater 65 treatment residue management alternative poses significant environmental concerns about bacteria, heavy metals, and trace organic pollutants; the same conclusion applies to traditional sea dumping and forestry, both of which are deemed unsustainable [4].
Due to the high cost of industrial activated carbons, research on activated carbon materials extracted from bio-waste, such as olive stones, lemon grass, wall nutshells 70 and pine cones has taken center stage [5,6,7,8] and among these, activated carbon extracted from sewage sludge (SBAC) has shown promise in the adsorption of inorganic elements such as Ni, Cu, Pb, Cd and Hg [1,9,10,11,12,13].
However, due to the inorganic composition of raw sewage sludge, SBAC has a low surface area, and increasing porosity by adding richer carbonaceous such as coal, 75 bagasse, and coconut shell may be preferable [ 7,9,11,14,15]. On the other hand, The use of HCl and NaOH, HNO3 and N,N-Dimethylformamide for the oxidation of SBAC in liquid phase has been reported to increase their adsorption potential by adding surface functionalities groups as shown in Table 1 [9,16,17,18]. An in-depth analysis of the existing body of works, on the other hand, exposes knowledge gaps, such as the fact 80 that very few studies take into account the specific chemical composition of the source materials used. The synthesis of activated carbon from sewage sludge has traditionally been oriented toward particular applications and thus has been more focused on results than on understanding the intrinsic structure within the materials (Table 1). In particular, understanding the adsorption of certain inorganic compounds in aqueous 85 solution by activated carbon requires knowledge of the chemical structure of the activated carbon surface, i.e. determining how the chemical structure of the activated carbon surface is oxidized during the chemical activation process in order to achieve the optimal porous surface. When the source material has been chemically treated, the surface of the activated carbons is normally filled with oxygenated sites and probably 90 amine sites which result in three types of oxides on the surface: acidic, basic, and neutral. Acidic sites increase the hydrophilicity of activated carbon, lower the pH in aqueous suspension, and increase the negative charge density at the surface, while simple sites are primarily of the Lewis form and are associated with -rich regions located at the basal planes [19]. Additional benefits can be obtained by oxidizing 95 activated carbon after the activation phase to induce the formation of oxygen complexes. The oxidation process raises the oxygen content by lowering the electronic density of the basal planes, which lowers the basicity at the surface [19,20,21]. The oxidation of activated carbon extracted from sewage sludge blended with coal using ammonium peroxydisulfate [(NH4)2S2O8] is investigated in this paper. In general, the 100 oxidation process is intended to result in the formation of carboxylic sites or the transformation of oxygen sites to carboxylic sites. The choice of ammonium peroxydisulfate is based on its ability to oxidizes the surface without modifying drastically the porous structure and enhance surface acidity by increasing carboxylic group presence [22,23,24]. To our knowledge, there are no reports on the oxidation of 105 activated carbon from sewage sludge blended with waste coal using ammonium peroxydisulfate. Finally, the oxidized activated carbons are tested for nitrate and MR adsorptive remediation. All reagents, including NaOH, NaNO3, KOH, HCl (32%), (NH4)2S2O8 (APS), H2SO4, and NaCl, were obtained from Associate Chemical Enterprise (ACE) and were analytical grade. Sigma-Aldrich (SA) supplied activated charcoal (C3014-500G), which was designated Com-AC.
Two forms of sewage sludge namely D and S were collected at the East Rand Water 120 Care Company (ERWAT) wastewater treatment plant facilities. D was collected during the dissolved air flotation stage, while S was a combination of the primary and secondary sludge from the digestion and dewatering stages, respectively.

Methods
Raw sewage sludges were sun dried for 24 hours before being oven-dried. The dried 125 sludge was crushed to 100 % passing through 125 µm sieves, and pyrolyzed at different temperatures (DC-5-750, DC-7-900, SC-3-600, and SC-5-900). The products from the pyrolysis were mixed with discard coal in a proportion of 1:1 and activated with KOH under N2 atmosphere at a flow rate of 5 l/min. To achieve surface modification via oxidation, 1 g of pyrolyzed sewage sludge was incubated in 25 ml of 1M or 2M APS 130 dissolved in 1M H2SO4 at 60°C for 240 minutes before being washed with distilled water until the pH neutral value was reached. Batch adsorption experiments were carried out with an adsorbent dose of 0.5% (5 mg of adsorbent in 10 ml of solution) and the performance of bio-adsorbents was compared to commercial activated carbon. The best oxidized and/or unoxidized SBACs were chosen based on preliminary evaluation 135 test results conducted at 303 K in an incubator shaker. The initial concentration of nitrate was 50 mg/l, with contact times of 180 and 360 minutes, respectively, while the initial concentration of MR was 75 ppm, with contact times of 90 and 180 minutes, and initial pH of 2 and 4 for nitrate and methyl red, respectively. The initial pH value of 2 in the case of nitrate was chosen because adsorbent surfaces are prone to being 140 positively charged as pH decreases, and nitrate ions cannot compete with hydroxyl anions [25] and pH 4 was chosen in the case of MR to prevent competition with H + cations while attempting to preserve the adsorbent surface deprotonated [26].

Chemical analysis
SEM and EDS analysis were done using a ZEISS SIGMA FESEM 03-39 apparatus 145 on coated samples on a carbon tape support using Gold-Palladium techniques. TGA was performed with the Pioneer (SDT-Q500) equipment. Proximate analysis of sewage was performed using a Perkin-Elmer STA 6000 simultaneous thermal analyser. FT-IR analyses were done using the Perkin-Elmer FT-IR Spectrometer Spectrum 2. Ultimate analysis was conducted with the thermoscientific flash 2000. XRD study was done 150 using a Bruker 2D phaser instrument with CuKα1 radiation at a wavelength of 1,54060 cm -1 and a 2θ angle. The NH4OAC method [27] was used to determine cations exchange capacity. The Boehm titration method was used to determine the acidity of SBAC [28].
The pH point of zero charge was determined using the procedure described by Leng et al. [29]. Surface functionalities and graphitization or carbon disorder structure of the 155 adsorbent were assessed using FT-IR analysis and Raman spectroscopy, respectively.

Pretreatment and characterisation of precursors
The precursors were sun dried, and the moisture content of the sample was 165 determined by mass losses after oven drying at 105 °C for 24 hours as shown in Table   2. The elemental analysis (Table 3) reveals that there is only a minor difference between the two samples, indicating that digestion has no effect on the organic content of sewage sludge, despite the volatile solid content being slightly different. However, although the surface area of sludge from D was approximately 1.5 m 2 /g, the surface area of sludge S could not be determined, possibly due to a lack of porosity in the precursor materials. In Figure 1, the TGA curve of dried sludge shows that between 30 °C and 190 °C, there is a slight mass loss, and the mass of both sludge was roughly 5%, possibly due to adsorbed water. The most mass loss occurred between 200 and 600 degrees Celsius, with mass loss of D and S increasing from about 5% at 200 degrees Celsius to about 185 two-thirds (64%) and half (46%) at 600 degrees Celsius, respectively. The mass loss from 600 °C to 1000 °C is less pronounced for both sludge with 17 and 13 % for D and S respectively. As a result, 600 °C was chosen as the minimum temperature for pyrolysis. S appears to be slightly more stable than D based on the TGA curves.  Figure 5 shows the FT-IR spectra of the raw sewage sludge with wavenumbers 225 ranging from 450 cm -1 to 4000 cm -1 . Reading the peaks spectra wavelengths from left to right, the peak observed at 3688-3619 cm -1 is attributed to OH-kaolinite and gibbsite lattice stretching [31,32], 2988-2901 cm -1 to -C-H group vibration [7,33], 1631 cm -1 and 1538 cm -1 to sulphur and nitrogen functional groups, respectively [33]. The shape of the shoulder peaks at 1050-1090 cm -1 was attributed to Si-C or Si-O-Si bands (Liang 230 et al., 2020), C-O-C vibration [33], and finally, under 1000 cm -1 , the peaks at 749 cm -1 , 535 cm -1 , and 467 cm -1 were attributed to silica or calcium carbonate stretching [32,34]. Discard coal has a lower transmittance from 1498 cm -1 peaks than FT-IR spectra sludge.1007 cm -1 , 1030 cm -1 , implying that waste coal contains more mineral elements whereas sludge S has lower transmittance than D, implying that the former 235 contains more functional classes.   Figure 6.a shows the microscopic examination of the sorbent SC-3-600 after activation. Pyrolysis and washing stages may have facilitated cavity formation of highly formed cavities due to the remarkable depletion of inorganic and organic components [35]. Furthermore, some particles ( Figure 6a and 6b) lack or have insignificant cavities, which may be attributable to a lack of volatile and decomposed matter escaping to 250 facilitate porosity [9]. The EDS qualitative analysis ( Figure 6.c) of the local particle In comparison to feedstock spectra, the disappearance of the wavelength peak at 470 cm -1 and 500 cm -1 in SBAC could be due to inorganic matter solubilisation during the acid washing phase [36,37,38], while 798 cm -1 could be due to dehydrogenation reactions [24], 1631 cm -1 and 1538 cm -1 could be due to thermal degradation of protein 270 for nitrogen related compound or sulphur [33]. The peaks associated with C-H group stretching not only shifted slightly from2859-280 2922 cm -1 in feedstock to a higher value (2901-2988 cm -1 ) in SBAC, which can be related to the presence of saturated group [11] but also transmittance increased after activation for all samples, which is in contrast to some literature [7,33,39] in which it was argued that the disappearance of the peaks was due to the the decomposition of fatty organic matter and dehydration [7]. Organometallic formation may be a possible 285 explanation for the increased transmittance of SBAC pyrolyzed at lower temperatures followed by depletion as temperature rises; for example, the abundance of functional groups (hydroxyl, carboxyl) on the biochar surface synthesized from sewage sludge pyrolyzed at 300 °C reduced extractable cations due to the formation of organometallic compounds [40,41]. The functional groups present in SBAC can be summarized as O- Com-AC has broad peaks (2 = 25,3° and 2 = 44,6°) linked to its amorphous phases, while the XRD pattern (Figure 9a and b) demonstrated mineral phases transformation from broad peak in the precursors to sharp in the manufactured absorbent, which 295 clarified transition from amorphous to crystalline phase due to pyrolysis. XRD verified the existence of minerals such as wustite, quarts, illite, and feldspars in SBAC. It is worth noting that the presence of alkaline earth elements in the minerals (feldspars and illite) led to magnesium, calcium, and iron being classified as exchangeable cations.

Surface modification via oxidation
The oxidation of SBAC resulted in surface functionalities modifications with formation of peaks in the oxidized adsorbent around 1550-1575 cm -1 (Fig. 10),which can be linked to symmetric COOand nanoaromatic C=O stretching entailed in formation of As shown in Table 5, graphitic structure is more predominant than carbon disorder structure and augmented with oxidation, partial graphitization of activated carbon was reported with acidic oxidation treatment with HNO3 [42]. The surface of oxidized adsorbent (Figure 12.b) exhibited less irregularities and soften surfaces than the unoxidized adsorbent (Figure 12.a) and Com-AC (Figure 12.c) probably as results of corrosive H2SO4-adsorbent surface interaction [18] and also disintegration of pore structure situated at the carbon edge [17]. XRD spectra of adsorbent are represented in Figure 13 after oxidation treatment the unoxidized sorbent variation was less than 2%, alike tendency was reported by Ang et al. [22] Furthermore, textural properties in Table 6 receded severely after modification, in case 350 of bio-adsorbent derived from sludge S, for instance prior oxidation SC-3-600 had surface area of 281,72 m 2 /g that shrunk to 68,22 m 2 /g and 46,673 m 2 /g when treated with a solution of 1M and 2M APS respectively, probably due to thinness of walls pores, which are prone to collapse [43] and/or micropore occlusion [44].

Adsorption experiments
To better understand kinetics order and intra particle diffusion, the effect of time was investigated using the following relationships: (2) Where qe (mg/g), C0 (mg/l), Ce (mg/l), m (mg) and V (ml) represent the adsorption capacity, initial concentration, equilibrium concentration, adsorbent mass and volume of liquid in contact with adsorbent respectively. The removal %age is calculated via equation (2). 375 The first and second order kinetics models, as well as interparticle diffusion, are commonly used to understand the adsorption mechanism of pollutants with activated carbon [45].
Adsorption kinetics can be expressed in terms of the hypothesis that adsorbate removals obey a first-order kinetics: where qt and qe are the amount of pollutants adsorbed per mass of adsorbent (mg/g) at the targeted time and equilibrium, respectively, and k1 is the constant rate (min -1 ).
After integration with conditions that qe =0 if t = 0, equation (3) can be written: Or alternatively The adsorption capacity at equilibrium (qe) and the first-order sorption rate constant (k1) can be evaluated from the slope and the intercept respectively from plot of ln (1qt/qe) vs t. 390 The second pseudo order kinetics is defined by equation: where K2 is the second order rate constant, integration of equation (6) with initial conditions when t = 0 and qe =0, lead to: Where Kint is the intraparticle diffusion rate constant (mg. g -1 min -1/2 ) and C is the boundary layer effect intercept; the larger C, the greater the contribution of surface sorption to the rate-controlling step.   (Table 6) or a lack of acidic functional groups due to their depletion during temperature augmentation [14]. This finding emphasizes the critical nature of functional group presence.
As depicted in Figure 15, the synthesised sorbents with the highest adsorption potential were DC-5-750 M1 (127,6 mg/g) and DC-7-900 M1 (124,4 mg/g), and they were 450 therefore chosen for future experiments, along with the unmodified sorbents (DC-5-surface reaction which are deemed to present less component resistance than external diffusion [46], while stagnant trend after 120 min may be ascribed to occupation of available adsorption site by nitrate ions as the process progress [25,47,48], which 455 weaken the interaction between sorbate and adsorbent surface [48]. Preliminary tests shown in Figure 6 revealed that the adsorption capacity increased significantly with time with a little fluctuation in Com-AC from 123, 8 mg/g at 90 minutes to 121, 172 mg/g at 180 minutes, implying that equilibrium had already been reached. 460 Although having a higher surface area, the adsorption capacity of SC-5-900 (422,1 m2/g) and SC-5-900M2 (313,1 m 2 /g) was lower than other SBAC. This could be due to lower carbon content (Table 5) and/or depletion of acidic functional groups as they vanished with temperature rise (28). This observation further corroborated the importance of functional group presence. 465 Based on results in Figure 6, DC-5-750 M1 and DC-7-900 M1 with adsorption capacities of 127, 6 mg/g and 124, 4 mg/g, respectively, were chosen for additional investigations in comparison to Com-Ac (121,1 mg/g) performances. The rapid increase in adsorption capability at the start of the process could be attributed to the abundance of adsorption sites [26,49].
Nitrate pseudo first order (PFO) and Pseudo second order (PSO) plots are recorded in Figure 17a-b and the corresponding parameters are recorded in Table 6. From the results in Figure 17a Table 8.  The effect of pH was measured by varying the pH solution from 2 to 10, as shown in Figure 20. The process was pH dependable, as evidenced by adsorption decrease with pH increasing, in the case of DC-5-750 (pHpzc=6,6) from 20,56 mg/g at pH 2, to 6,24 525 mg/g at pH 6 and 4,2 mg/g at pH 10, Com-AC (pHpzc=10,3) and DC-5-750M2 (pHpzc=3,1). At pH 2, adsorption potential was 16,32 mg/g and 12,24 mg/g, respectively, at pH 6, 11,12 mg/g and 9,6 mg/g, respectively, and at pH 10, 6mg/g and 1,8mg/g. This pattern was more likely caused by: (i) favorable conditions of nitrate removal accentuated by electrostatic attraction as adsorbent surface bears positive 530 charge at lower pH. (ii) the presence of rivalry between nitrate ions and hydroxyl ions in basic solution, as also stated in other works using carbon-based activated carbon content [47,48,51]. In addition to the above-mentioned justifications, it may be further hypothesized that the underperformance of oxidized SBAC is due to the introduction of an acidic surface functional group; in the case of deprotonation, if pH> pHpzc, more 535 binding sites for cationic sorbate are created on the surface than if it was an unoxidized adsorbent [44,52].
Taking into account that MR is negative if pH>pKa and positive if pHpKa (Khan et al., 2018), the introduction of acidic functional groups caused a shift in pHpzc, from neutral 6,6 (DC-5-750) to acidic after oxidation 3,1 (DC-5-750-M1) and Com-AC was basic 540 10,2, surface functional group deprotonated when pHpzc < pH and adsorbents surface becomes negatively charged [53]. In contrast to SBAC, Com-AC (pHpzc=10,2) has a wider range where the surface's net charge is positive. Adsorption of MR with changed SBAC, which had a lower pHpzc than unoxidized, was more pH dependable due to electrostatic attraction between the adsorbent negatively charge and positive MR below 545 pH4 [26]; at pH4, the adsorption potential of DC5-750M1 and DC-7-900M1 was 127.634mg/g and 117.176 mg/g, respectively, from pH 6, pH 8, and pH 10. Adsorption capacity decreased at pH6 from 97,488mg/g for DC-5-750M1 and 109.278 for DC-7-900M1 to 81.854 mg/g for DC-5-750M1 and 71,028 mg/g for DC-7-900M1, the results show that for the oxidized adsorbent at pH4 the adsorption mechanism was related to 550 electrostatic attraction and hydrophobic associated to/or hydrogen bond [54]. The outperformance in basic solution could be attributed to adsorption site rivalry between hydroxyl ion and MR negatively charged ions [49]. Similarly, in an acidic solution with a pH of 2, the rivalry may have been between H + and positive MR [55] 560 or/and repulsion force between protonated adsorbent surface and MR [26].
However, pH solution variation affected slightly the Com-AC adsorption capacity of MR, from 101.894 mg/g at pH 2 to 113,04 mg/g at pH 10, possibly because the electrostatic attraction mechanism was not very pronounced in the process because below pH 5,1 dye was charged positively and the protonated adsorbent had positive net 565 charge, similar results were recorded on adsorption of cationic dye (Methyl blue) [56].
The adsorption capacity increased as initial concentration increased because driving force of concentration gradient prevailed and had propensity to subjugate mass transfer resistance barrier between solid and liquid interface. Conversely, the proportion of nitrate extracted decreased due to adsorbent site saturation, so a fraction of sorbate 570 remained in solution [14]. As the initial pollutant concentration increased, the adsorbent's dye adsorption capacity increased, but the proportion of dye removed     Nitrate Langmuir and Freundlich isotherms were used to investigate the nitrate removal process. The findings are presented in Table 11.
As predicted, the Langmuir isotherm described the process better than the Freundlich model, with a greater R2 for all adsorbents involved on a monolayer surface. The value of RL between 0 and 1 indicated that both sorbate adsorption was favourable on all 620 sorbent surfaces of the Langmuir isotherm model, it is regarded as unfavourable if RL>1 [36], for Freundlich model adsorption intensity (1/n) value was less than 0,5 in all cases, indicating that sorbate was easily adsorbed, it is hardly adsorbed if 1/n>2 [42].
It is worth noting that MR adsorption became more irreversible with increasing 625 concentration since the RL value approached zero at higher concentration [25,26]. The published data are also compared to the present study's nitrate adsorption capacities in Table 12.To assess the effect of ionic strength, MR was diluted in NaCl solution with a concentration of 0.01 M and 0,05M and the pH was adjusted at 4. As shown in Figure   23, presumably Na + cations competition and screening effect of Clanions at the 630 external surface of adsorbent [54] caused adsorption capacity of oxidized SABC to dwindle narrowly with NaCl concentration change from 127,63 mg/g (without NaCl) to 117, 08 mg/g (0,05M NaCl) for DC-5-750M1 and 122,29 mg/g (without NaCl) to 119, 61 mg/g(0,05M NaCl) for DC-7-900M1, contrariwise Com-AC and unoxidized adsorbents (DC-5-750 and DC-7-900) increased with NaCl concentration augmentation 635 from 121,08 mg/g to 129,42 mg/g, 114,43 mg/g without NaCl to 125,31 mg/g and 110,29 mg/g to 122,85 mg/g in presence of 0,05M NaCl respectively, adsorption capacity upward trend could have been imputed arguably to dye aggregation drove by salt ions force [61].

Environmental consideration: toxicity characteristic leaching procedure
The toxicity contaminant leaching procedure was carried out as described elsewhere 645 [39,62], and the element concentrations were determined using a Perkin-Elmer AA spectrometer. Table 13 shows the results of the TCLP test; in general, the concentration of leachable heavy metal in the pyrolyzed adsorbent was lower than the precursor due to the higher thermal stability of heavy metal acquired through pyrolysis [62]. 650 However, after pyrolysis, SC-5-900 released more heavy metals (Fe, Cr, Co, and Ni) than its precursors. This may be due to the disintegration of some stable inorganic minerals (primarily silicate and carbonate) from sludge during pyrolysis with temperature augmentation, which caused the liberation of the fixed metals from the lattice [17]. 655

660
In this paper, two types of sewage sludge were used to make low-cost bioadsorbents: D, which was collected during the dissolved air flotation stage, and S, which was a mixture of primary and secondary sludge from the digestion and dewatering stages. The sewage sludge was mixed with waste coal before being activated with KOH and oxidized with APS. The ability of the synthesized bio-adsorbents to remove nitrate 665 and MR was assessed and compared to that of industrial activated charcoal. The oxidation with APS influenced (i) the textural properties of bio-adsorbent derived from sludge S more negatively than those derived from D, (ii) influenced organic composition only marginally as revealed by ultimate analysis, and (iii) induced the introduction of acidic functional groups as revealed by FT-IR and Raman spectroscopy 670 analysis, respectively. Adsorbents' adsorption capability increased with time and initial concentrations of contaminants. The removal processes of nitrate and MR followed the pseudo second order kinetic model and the Langmuir isotherm reasonably well. At pH=2 and pH=4, nitrate and MR adsorption capacities were higher, respectively.