Red mud (RM) is the industrial waste of alumina production and causes serious environmental risks. In this paper, a novel activation procedure for RM (mechano-chemical processing) is proposed in order to improve the nitrate adsorption from water. High-energy milling and acidification were selected as mechanical and chemical activation methods, respectively. Synthesized samples of adsorbent were produced considering two parameters of activation: acid concentrations and acidification time in two selected milling times. Optimization of the activation process was based on nitrate removal from a stock solution. Experimental data were analyzed with two-way analysis of variance and Kruskal–Wallis methods to verify and discover the accuracy and probable errors. Best conditions (acceptable removal percentage > 75) were 17.6% w/w for acid concentrate and 19.9 minutes for acidification time in 8 hours for milling time. A direct relationship between increase in nitrate removal and increasing the acid concentration and acidification time was observed. The adsorption isotherms were studied and compared with other nitrate adsorbents. Characterization tests (X-ray fluorescence, X-ray diffraction, Fourier transform infrared spectrophotometry, dynamic light scattering, surface area analysis and scanning electron microscopy) were conducted for both raw and activated adsorbents. Results showed noticeable superiority in characteristics after activation: higher specific area and porosity, lower particle size and lower agglomeration in structure.

INTRODUCTION

Red mud (RM) is a reddish-brown colored solid waste generated during the production of alumina (Wang & Liu 2012). Based on present technologies, 1–2.5 tons of RM is generated for each ton of alumina produced, depending on the quality of the raw material processed. Globally, the total amount of RM produced every year is about 60 million tons (Liu & Wu 2012). RM is a highly alkaline waste material with pH 10–12.5. Due to the alkaline nature and the chemical and mineralogical species present in RM, this solid waste causes a significant impact on the environment and proper disposal of waste RM presents a huge challenge where alumina industries are installed (Wang et al. 2008).

RM has a wide range of composition: Fe2O3, Al2O3, SiO2, Na2O, CaO, TiO2 and trace elements. The particle diameter of RM is 0.8–50 μm (average: 14.8 μm) and 98% of particles are less than 1 μm. The measured value of the density of RM is 2.70 g/cm3. The specific surface area of RM is around 10–25 m2/g (Wang et al. 2008; Wang & Liu 2012; Nath et al. 2015).

RM can be used in the ceramic and cement industries, in agriculture to improve soil quality, in building materials, e.g. blocks and bricks, and in recovering valuable elements, e.g. titanium (Paramguru et al. 2005). In recent years, development in using RM as an adsorbent for water and wastewater treatment has been widely reported, e.g. in removal of organic materials (Gupta et al. 2004; Tor et al. 2006; Zhang et al. 2012), heavy metals (Ma et al. 2009; Chandra Sahu et al. 2011; Smiljanić et al. 2011), inorganic anions (Cengeloglu et al. 2006; Tor et al. 2009; Zhao et al. 2011), dyes (Crini 2006; Ratnamala et al. 2012) and pathogens (Ho et al. 1990). Utilization of industrial wastes for another waste treatment produces many benefits in terms of economy and the environment and could help environmental pollution abatement, in solving both solid waste disposal as well as liquid waste problems (Wang et al. 2008; Ratnamala et al. 2012).

Nitrate, due to its high water solubility, is possibly the most widespread groundwater contaminant in the world, imposing a serious threat to human health and contributing to eutrophication. Among several treatment technologies applied for nitrate removal, adsorption has been explored widely and offers satisfactory results especially with carbon-based, mineral-based and surface-modified adsorbents (Bhatnagara & Sillanpaab 2011; Hashemi et al. 2011; Hong et al. 2012). RM (in raw or activated forms) is reported as an efficient and low-cost adsorbent to remove nitrate from aqueous environments (Cengeloglu et al. 2006).

Nowadays, development of low-cost adsorbents from industrial and agricultural wastes has been extensively researched. RM as a low-cost adsorbent exhibits promising adsorption capacity (Wang et al. 2008). It has been shown that physical and chemical activation can significantly change the adsorption capacity of RM (Crini 2006). Three methods, acidification (chemical method), thermal treatment and high-energy milling (mechanical method), are usually employed to activate RM. Acid and heat treatments will improve the surface physicochemical properties, resulting in higher adsorption capacity. High-energy milling will increase the surface energy of adsorbent and produce a finer particle size, leading to higher adsorption capacity too (Wang et al. 2008).

Development of novel materials and improvement of the properties of existing materials have been the preoccupation of materials scientists for several decades. Mechano-chemical processing (MCP) is a novel method applied to the powder process, in which chemical reactions, structural changes and phase transformations are activated by the application of mechanical energy. MCP is normally a dry, high-energy ball milling technique and has been employed to produce a variety of commercially useful and scientifically interesting materials. Some attributes of MCP are listed below (Suryanarayana & Ivanov 2013):

  • Production of fine dispersion of second phase (usually oxide) particles.

  • Refinement of grain sizes down to nanometre range.

  • Synthesis of novel crystalline and quasicrystalline phases.

  • Inducement of chemical reactions at low temperatures for mineral and waste processing.

In this paper, a novel mechano-chemical activation process for RM adsorbent was intended to be optimized based on maximizing nitrate removal from a constant stock nitrate solution. Experimental parameters selected were acid concentration and acidification time in two different milling times.

MATERIALS AND METHODS

Chemicals

Raw red mud (RRM) was obtained from Iran Alumina Co. (Jājarm, Iran). For representative results, each sample was a mixture of RRM taken from three different points of the waste dam. HCl, NaOH and NaNO3 was obtained from Merck Co. (Darmstadt, Germany). Ultra-pure water (UPW) was used to prepare nitrate solution.

Instruments and standards

Adsorbent preparation

The pH values were determined by a BEL W3B pH meter. Milling of adsorbent was carried out with an Asia San'at Rakhsh PM2400 planetary ball mill with hardened steel cups and hardened steel balls (HRC (Rockwell C Hardness) of 58). Concentration of nitrate ion was determined by a Metrohm 850ProfIC ion chromatography (IC) system.

For determination of residual concentration of nitrate ion by IC, ASTM ISO 10304 standard method was used.

Adsorbent characterization

X-ray fluorescence (XRF) mineralogical composition of the samples was determined using a Philips PW2404 analyzer. X-ray diffraction (XRD) analysis was carried out on a Philips PW3040/60 diffractometer using Cu K–alpha radiation, voltage of 40 kV, current of 30 mA, and scanning angle of 5° to 80°. Qualitative phase analysis was conducted using the X'Pert Highscore plus search match software. Fourier transform infrared spectrophotometry (FT-IR) was carried out on a Bruker Tensor27 spectrophotometer. Spectra were recorded and analyzed with the Opus software. A dynamic light scattering (DLS) test was carried out using a Malvern ZS zetasizer instrument. Scanning electron microscopy (SEM) observation was performed by a TESCAN VEGA II scanning electron microscope. The Brunauer–Emmett–Teller (BET) test was performed by a BEL Belsorp Mini II with N2 gas and 150 mA current sensor.

Preparation of adsorbents

RRM preparation

RM lumps were first pounded and sieved with mesh #100 (0.149 mm). Then, it was washed thoroughly with UPW three times (liquid to solid weight ratio was 2:1), and the mixture was let to sediment for about 1 hour. After filtration (using Whatman grade 42 filter paper with pore size 2.5 μm) the powder was kept overnight in an oven (70 °C) to get completely dry, and then sieved again with mesh #100 (0.149 mm) to obtain a uniform sized powder of adsorbent.

Mechano-chemical activated red mud preparation

To prepare mechanical activated red mud (MRM), RRM powder was milled in a planetary ball mill with ball to powder weight ratio of 25:1 for 4 and 8 hours with rotating speed of 200 rpm. Then, it was pounded and sieved with mesh #100 (0.149 mm). For further chemical activation, 10 g of MRM was boiled in 200 mL of 5 to 20% wt. HCl for 5 to 20 minutes. The acid slurry was then filtered and the residue washed with distilled water to remove residual acid and soluble Fe and Al compounds. After that, the sample was washed thoroughly with UPW five times (liquid to solid weight ratio was 2:1). After filtration (using Whatman grade 42 filter paper with pore size 2.5 μm), the mechano-chemical activated red mud (MCRM) adsorbent was kept overnight in an oven (70 °C) to get completely dry, and was then sieved again with mesh #100 (0.149 mm) to obtain a uniform sized powder.

Nitrate stock solution preparation

Nitrate solution was prepared from 1,000 mL sodium nitrate (NaNO3) stock solution (150 mg/L NO3). The salt was weighed carefully and dried at 105 °C for 1 hour to lose excessive water. Then, it was dissolved in UPW, and then water was added to reach the volume of 1,000 mL.

Nitrate removal tests

The experimental parameters pH, adsorbent dosage, contact time and initial nitrate concentration were set at constant values for all adsorption experiments as mentioned in Table 1.

Table 1

Constant conditions for nitrate removal tests

pH Contact time (min) Initial nitrate concentration (mg/L) Adsorbent dosage (g/L) Stirrer speed (rpm) Temperature (°C) 
60 150 0.1 500 25 
pH Contact time (min) Initial nitrate concentration (mg/L) Adsorbent dosage (g/L) Stirrer speed (rpm) Temperature (°C) 
60 150 0.1 500 25 

The pH values were controlled by addition of adequate volume of 1 M NaOH or HCl solutions. At room temperature conditions (25 °C), adsorbent was added to solution and mixed with a magnetic stirrer with 500 rpm, then filtered, centrifuged and separated from solution. Residual concentration of nitrate was determined with the IC analyzer.

Statistical analysis

In this research, two-way analysis of variance (ANOVA) and Kruskal–Wallis methods were applied on experimental data. Two-way (or multi-way) ANOVA is an appropriate method for a study with a quantitative outcome and two (or more) categorical explanatory variables. The usual assumptions of normality, equal variance, and independent errors apply. One of the possible means models for two-way ANOVA is the interaction model. In the interaction model, the effects of a particular level change for one explanatory variable does depend on the level of the other explanatory variable (Seltman 2014).

The Kruskal–Wallis method is a non-parametric test and its validity does not depend on the data being drawn from any particular distribution. The test is applied to, say, k independent random samples of sizes ni, i = 1, 2,…k, with a total of n observations. The advantage of the non-parametric method is that it can be used with any set of data with their distribution being unknown (Kottegoda & Rosso 2008).

The data were statistically analyzed using Minitab® Version 16.2.4 software.

RESULTS AND DISCUSSION

Nitrate removal tests

The results of nitrate removal tests are given in Tables 2 and 3. Selection of milling time was done based on the authors' previous experiments in laboratory. Because of the nature of RM (combination of several metal oxides), optimal milling time for RM is a laboratorial parameter that can be determined by experiment. In the optimal milling time, a lower particle size of sample with lower agglomeration is obtained. It should be noted that milling time is an important parameter which determines the amount of accumulated energy in the particles due to the milling process. The stored energy can affect the dissolution process during chemical activation process.

Table 2

Results of nitrate removal tests for MCRM-4

Sample no Acid concentration (% w/w) Acidification time (min) Nitrate removal (%) 
2.5 59.30 
60.01 
10 60.75 
20 66.14 
10 2.5 61.33 
10 62.49 
10 10 61.27 
10 20 65.11 
20 2.5 66.75 
10 20 67.48 
11 20 10 67.96 
12 20 20 70.64 
Sample no Acid concentration (% w/w) Acidification time (min) Nitrate removal (%) 
2.5 59.30 
60.01 
10 60.75 
20 66.14 
10 2.5 61.33 
10 62.49 
10 10 61.27 
10 20 65.11 
20 2.5 66.75 
10 20 67.48 
11 20 10 67.96 
12 20 20 70.64 
Table 3

Results of nitrate removal tests for MCRM-8

Sample no Acid concentration (% w/w) Acidification time (min) Nitrate removal (%) 
13 2.5 68.03 
14 68.41 
15 10 69.66 
16 20 70.50 
17 10 2.5 69.83 
18 10 71.33 
19 10 10 71.89 
20 10 20 74.33 
21 20 2.5 72.16 
22 20 74.86 
23 20 10 74.80 
24 20 20 75.11 
Sample no Acid concentration (% w/w) Acidification time (min) Nitrate removal (%) 
13 2.5 68.03 
14 68.41 
15 10 69.66 
16 20 70.50 
17 10 2.5 69.83 
18 10 71.33 
19 10 10 71.89 
20 10 20 74.33 
21 20 2.5 72.16 
22 20 74.86 
23 20 10 74.80 
24 20 20 75.11 

The results of two-way ANOVA analysis are given in Table 4 and Figure 1, which show the accuracy, correlation of parameters and probable errors of laboratorial results. It shows the appropriate set of data with the normal distribution. The standardized residual of the linear regression model was used and the error term ‘ɛ’ was actually normally distributed.
Table 4

Results of two-way ANOVA analysis

Sample p-value
 
Selected confidence level (%) R2 (%) Accuracy 
Acidification time Acid concentration 
MCRM-4 0.003 0.000 99.50 95.58 ok 
MCRM-8 0.008 0.000 99.50 91.90 ok 
Sample p-value
 
Selected confidence level (%) R2 (%) Accuracy 
Acidification time Acid concentration 
MCRM-4 0.003 0.000 99.50 95.58 ok 
MCRM-8 0.008 0.000 99.50 91.90 ok 
Figure 1

Normal probability plot of residues for (a) MCRM-4 and (b) MCRM-8.

Figure 1

Normal probability plot of residues for (a) MCRM-4 and (b) MCRM-8.

The linear coefficient of determination (R2) was employed for the error analysis (0 < R2 < 1). R2 = 1 shows that 100% of the variation of experimental data is explained by the regression equation. The R2 parameter was also accurate (above 90%) for both MCRM-4 and MCRM-8 results.

To determine whether there was a significant difference between the removal data groups (measured for MCRM-4 and MCRM-8) and for verification of ANOVA analysis data, Kruskal–Wallis statistical analysis was conducted and intergroup comparisons were done. Assuming α > 0.05 for the level of significance, the results showed that there was a significant difference between the data groups. The MCRM-8 (p-value < 0.003) showed much better removal efficiency of nitrate than MCRM-4 (p-value = 0.08). Considering the mechanical activation, nitrate removal values are relatively higher for 8 hours milling time than for 4 hours milling time (Figure 2). Consequently, 8 hours milling time was selected for the mechanical activation process for RM adsorbent.
Figure 2

Comparative box plot for MCRM-4 and MCRM-8 removal data groups.

Figure 2

Comparative box plot for MCRM-4 and MCRM-8 removal data groups.

Optimizing the chemical activation parameters

For 4 hours and 8 hours milling times, a surface diagram of nitrate removal versus acid concentration and acidification time is illustrated in Figure 3(a) and 3(b), respectively. It can be noticed that nitrate removal values generally increase with increasing the acid concentration and acidification time. Also, the maximum removal of nitrate (75.11%) occurs in acid concentrate of 20% w/w and acidification time of 20 minutes.
Figure 3

Nitrate removal versus acid concentration and acidification time for constant milling time of (a) 4 hours and (b) 8 hours.

Figure 3

Nitrate removal versus acid concentration and acidification time for constant milling time of (a) 4 hours and (b) 8 hours.

The parameters are minimized in order to obtain at least 75% removal. Results of the optimization process indicated that removal percentage of 75% can be obtained in acid concentrate of 17.6% w/w and acidification time of 19.9 minutes (Table 5).

Table 5

Conditions for nitrate removal tests

Optimal acid concentration (% w/w) Optimal acidification time (min) Milling time (hour) 
17.6 19.9 
Optimal acid concentration (% w/w) Optimal acidification time (min) Milling time (hour) 
17.6 19.9 

Mechano-chemical activation can improve the characterization of RM adsorbent including increase in the percentage of iron and aluminum content, reduction in diversity of minerals and elimination of ineffective phases, finer particle size and unified grading, modification to the shape of particles and reduction in agglomeration, higher specific surface area, decrease in mean pore size and modification to pore size distribution. It also increases the capacity of adsorbent to remove nitrate from solution.

Adsorption isotherm study

The equilibrium adsorption isotherm is of fundamental importance in the design of adsorption systems and indicates how the molecules subjected to adsorption distribute between the liquid phase and the solid phase when the adsorption process reaches an equilibrium state (Demiral & Gündüzoğlu 2010). The Langmuir and Freundlich isotherm models (linear shape) are represented by Equations (1) and (2) (Menkouchi Sahli et al. 2008): 
formula
1
 
formula
2
where qe (mg/g) is the nitrate concentration in the adsorbent, Ce (mg/L) is the equilibrium nitrate concentration in solution, qmax (mg/g) is monolayer capacity of the adsorbent, b is the Langmuir adsorption constant, Kf (L/g) and n are Freundlich constants which indicate the relative sorption capacity and sorption intensity, respectively.

The isotherm data were correlated with Freundlich and related parameters are listed in Table 6 for further investigation. Values of Kf, n, qmax and b were calculated from the intercept and slope of the linear plots. For nitrate adsorption with MCRM, the Langmuir adsorption isotherm model showed more significant correlation than the Freundlich isotherm model.

Table 6

Isotherm parameters for removal of nitrate

  Abb. Freundlich isotherm
 
Langmuir isotherm
 
Reference 
Kf n R2 qmax b R2 
Raw red mud RRM 0.874 1.599 0.962  N.A.a Cengeloglu et al. (2006)  
Chemical activated red mud CRM  N.A. 5.858 65.654 0.999 Cengeloglu et al. (2006)  
Activated carbon AC  N.A. 27.550 2.112 0.995 Demiral & Gündüzoğlu (2010)  
Chemical activated sepiolite CS 2.490 1.133 0.901  N.A. Öztürk & Ennil Bektaş (2004)  
Chitosan – 99.480 0.616 0.940 8.030 0.007 0.940 Menkouchi Sahli et al. (2008)  
Mechano-chemical activated red mud MCRM  N.A. 1.007 0.912 0.975 This research 
  Abb. Freundlich isotherm
 
Langmuir isotherm
 
Reference 
Kf n R2 qmax b R2 
Raw red mud RRM 0.874 1.599 0.962  N.A.a Cengeloglu et al. (2006)  
Chemical activated red mud CRM  N.A. 5.858 65.654 0.999 Cengeloglu et al. (2006)  
Activated carbon AC  N.A. 27.550 2.112 0.995 Demiral & Gündüzoğlu (2010)  
Chemical activated sepiolite CS 2.490 1.133 0.901  N.A. Öztürk & Ennil Bektaş (2004)  
Chitosan – 99.480 0.616 0.940 8.030 0.007 0.940 Menkouchi Sahli et al. (2008)  
Mechano-chemical activated red mud MCRM  N.A. 1.007 0.912 0.975 This research 

aNot applicable.

It is well known that Langmuir-type adsorption is resulted with monolayer type adsorption Negative values for the Langmuir isotherm constants indicate the inadequacy of the isotherm model to explain the adsorption process (Cengeloglu et al. 2006).

The Freundlich isotherm model describes adsorption on a heterogeneous surface and is not restricted to monolayer formation. In general, as the Kf value increases, the adsorption capacity of the adsorbent increases. Usually, 1 < n < 10 provides a good adsorbent. A smaller value of n indicates better adsorption and formation of relatively strong bonds between adsorbate and adsorbent (Zabihi et al. 2009).

The nitrate sorption with RRM obeys the Freundlich isotherm model while CRM and MCRM obey the Langmuir isotherm model. The nitrate adsorption process with activated forms could be described as a monolayer adsorption and limited by surface site saturation due to homogeneous and negligible interaction between adsorbed molecules. The RRM surface is composed of a heterogeneous mixture of several minerals. A less heterogeneous mineral assemblage on the surface of RM happens by activation and results in a homogeneous surface. The activation process improves the adsorption capacity by increasing binding sites.

The results of isotherm studies of nitrate adsorption by different adsorbents (Table 6) showed that the adsorption capability of the prepared MCRM was more than that of the other adsorbents. It may due to the fact that MCRM adsorbent provided active sites to contact with nitrate ions directly. Diffusion resistance may be smaller in this sample; therefore, the amount of adsorption is greater.

Characterization of adsorbents

Chemical compounds

The chemical composition of two types of RM adsorbent (RRM and MCRM) determined by XRF analysis is given in Table 7.

Table 7

Major oxide composition of two types of RM adsorbent (%)

  CaO Fe2O3 Al2O3 SiO2 TiO2 Na2MgO K2SO3 P2O5 
RRM 25.37 18.48 14.71 12.36 6.91 3.17 1.45 0.38 0.34 0.15 
MCRM 7.25 52.66 5.74 12.84 13.86 0.92 0.53 0.35 0.21 0.17 
  CaO Fe2O3 Al2O3 SiO2 TiO2 Na2MgO K2SO3 P2O5 
RRM 25.37 18.48 14.71 12.36 6.91 3.17 1.45 0.38 0.34 0.15 
MCRM 7.25 52.66 5.74 12.84 13.86 0.92 0.53 0.35 0.21 0.17 

The Fe2O3 and TiO2 content of MCRM is considerably higher than in the RRM sample, while the CaO and Al2O3 content is considerably reduced. The reason is the dissolution of Al- and Ca-containing materials during chemical activation process. The changes in the content of other oxides are negligible. In addition, MCRM contains relatively less rare earths than RRM.

Mineral phases

The mineral phases of two types of RM adsorbent (RRM and MCRM) determined by XRD analysis is given in Figure 4.
Figure 4

XRD pattern of (a) RRM adsorbent and (b) MCRM adsorbent.

Figure 4

XRD pattern of (a) RRM adsorbent and (b) MCRM adsorbent.

The mineral composition of RRM was: calcite (CaCO3), hematite (Fe2O3), magnesium–aluminum–oxide (MgAl2O4), perovskite (CaTiO3), calcium–aluminum–iron–oxide (CaAl4Fe8O19), anatase (TiO2), magnesium–silicon (Mg2Si), larnite (Ca2SiO4), cancrinite (Na6Ca2Al6Si6O24(CO3)2), katoite (Ca3Al2(SiO4)(OH)8) and calcium–silicon (CaSi2), while the mineral compositions of MCRM was: hematite (Fe2O3), perovskite (CaTiO3), aluminum–silicon–titanium–oxide (Al4Ti2SiO12) and calcium–iron–oxide (Ca2Fe7O11). This indicates that they have obviously different mineral compositions, and diversity of phases is reduced in MCRM adsorbent. Also, most of the iron and aluminum content of MCRM is in the form of Fe3+ (ferric) and Al3+.

Chemical bonds

The nature of the chemical bonds in the two types of RM adsorbent was characterized by FT-IR spectrophotometry. The FT-IR spectra of RRM and MCRM are shown in Figure 5.
Figure 5

FT-IR pattern of two types of RM adsorbent: (a) RRM and (b) MCRM.

Figure 5

FT-IR pattern of two types of RM adsorbent: (a) RRM and (b) MCRM.

The spectrum pattern of RRM shows peaks at 3,430 and 1,798 cm−1, related to O–H and H–O–H bonds. It indicates existence of water in RM. Also the peak at 1,431 cm−1 is related to C=O bonds, because of carbonate compounds in RM phase structure. Silicate phases (e.g. katoite) show their effect at 991 cm−1. Peaks at 461, 561 and 682 cm−1 are related to hematite, aluminum silicate and titanium phases, respectively. The spectrum pattern of MCRM also shows a peak at 3,391 cm−1 related to O–H bonds (water content), a peak at 1,628 cm−1 related to C=O bonds, a peak at 1,077 cm−1 related to C–O bonds (calcite mineral phase) and peaks at 552 and 446 cm−1 related to hematite and silicates mineral phases.

Size of particles

The particle size distribution of the two types of RM adsorbent (from DLS analysis) is shown in Figure 6.
Figure 6

The particle diameter distribution plots and sieve curves of two types of RM adsorbent.

Figure 6

The particle diameter distribution plots and sieve curves of two types of RM adsorbent.

The mean size of RRM particles is 983.6 nm, while the mean size of MCRM particles is 196.7 nm (about five times finer). Compared with RRM, the MCRM had small particle diameter that corresponds to higher specific surface area of the adsorbent and its higher adsorption capacity. Comparing the shape of grading curves also showed more uniformity in particle size of MCRM adsorbent

Specific surface area

The BET analysis data for specific surface area and BJH (Barrett, Joyner and Halenda) method diagram for pore size distribution of the two types of RM adsorbent are given in Table 8 and Figure 7, respectively.
Table 8

BET analysis data for two types of RM adsorbent

  as BET (m2/g) Mean pore size diameter (nm) 
RRM 11.18 22.25 
MCRM 68.95 7.93 
  as BET (m2/g) Mean pore size diameter (nm) 
RRM 11.18 22.25 
MCRM 68.95 7.93 
Figure 7

Pore size distribution of two types of RM adsorbent: (a) RRM and (b) MCRM.

Figure 7

Pore size distribution of two types of RM adsorbent: (a) RRM and (b) MCRM.

The results indicated that specific area of MCRM is about six times larger than that of RRM adsorbent due to further mechano-chemical activation. MCRM also showed a noticeable decrease in mean pore size diameter. Principally, a considerable increase in adsorption properties of MCRM is expected.

The BJH diagram of the two types of RM adsorbent showed uniformity in pore size distribution for MCRM adsorbent. It can be observed that for the MCRM adsorbent, more than 85% of pores have diameter size under 10 nm.

Morphology of microstructure

SEM observations of RRM and MCRM samples are shown in Figure 8.
Figure 8

SEM micrograph of two types of RM adsorbent: (a) RRM and (b) MCRM.

Figure 8

SEM micrograph of two types of RM adsorbent: (a) RRM and (b) MCRM.

A comparative analysis on the two images showed less agglomeration, more dispersion, and finer particle size MCRM compared with RRM, which confirms results of DLS and BET analysis. SEM analysis also showed more angular shape of the RRM particles and more spherical shape of the MCRM particles. Accordingly, improvement of qualitative characteristics of RM adsorbent in the form of MCRM is predictable.

Mechanism of nitrate adsorption on RM

RM contains insoluble metal oxides in its structure. Other researchers demonstrated that metal oxide was an effective positively charged adsorbent with a high maximum adsorption capacity for anionic ions (e.g. nitrate). The charge of the anion is an important factor for the adsorption due to a kind of ion-exchange mechanism (Crini 2006).

Ligand exchange is proposed as the main reaction between the nitrate ion and metal oxide in the RM/activated RM based on studying the effect of pH on the adsorption rate (Liu et al. 2011). The extent of adsorption of anions is strongly governed by the pH of the solution. Since anion adsorption is coupled with OH ions, the adsorption is favored in low or neutral pH values. RM is an adsorbent containing different metal oxides in its structure. In a humid environment, hydroxylated surfaces of these oxides developed charge on the surface (Cengeloglu et al. 2006).

The mechanism for nitrate removal was explained by surface chemical interaction of RM with nitrate ions as shown in Equations (3) to (5) (Wang et al. 2008): 
formula
3
 
formula
4
 
formula
5
where M represents metal ions (Al, Fe or Si).
Among anionic ligands, sulfate is believed to form both inner- and outer-sphere complexes with surface active sites, Cl, F and NO3 forming outer-sphere complexes. Therefore, sulfate ions can block more available adsorption sites than nitrate and chloride ions. The affinity sequence for anion adsorption on RM is sulfate > nitrate ≫ chloride (Tor et al. 2006). The mechanism of nitrate anion adsorption on RM is illustrated in Figure 9.
Figure 9

Mechanism of nitrate anion adsorption on RM (M is the metal oxide surface.

Figure 9

Mechanism of nitrate anion adsorption on RM (M is the metal oxide surface.

CONCLUSION

Novel MCRM adsorbent was successfully synthesized. A mechanical and subsequent chemical activation method was selected for this purpose. It was shown that 8 hours milling and subsequent chemical activation (with 17.6% w/w for acid concentrate and 19.9 minutes for acidification time) were the optimal conditions for MCP. A direct relationship existed between increase in nitrate removal percentage while increasing the acid concentration and acidification time. An equilibrium study showed the monolayer nitrate adsorption process on MCRM. The raw and synthesized adsorbents were subjected to characterization tests. Research results indicated that novel MCP has considerable effect on the general characteristics of RRM adsorbent and can result in significant improvement of sorption ability.

ACKNOWLEDGEMENTS

The authors are grateful for financial support for this research work provided by Zanjan Province's Urban Water and Wastewater Co. (Zanjan, Iran). Nitrate detection tests were done in Water Research Institute (WRI) central laboratory.

REFERENCES

REFERENCES
Bhatnagar
A.
Sillanpaa
M.
2011
A review of emerging adsorbents for nitrate removal from water
.
Chemical Engineering Journal
168
,
493
504
.
Cengeloglu
Y.
Tor
A.
Ersoz
M.
Arslan
G.
2006
Removal of nitrate from aqueous solution by using red mud
.
Separation and Purification Technology
51
,
374
378
.
Chandra Sahu
R.
Patel
R.
Chandra Ray
B.
2011
Adsorption of Zn(II) on activated red mud: Neutralized by CO2
.
Desalination
266
,
93
97
.
Gupta
V. K.
Ali
I.
Saini
V. K.
2004
Removal of chlorophenols from wastewater using red mud: an aluminum industry waste
.
Environmental Science and Technology
38
,
4012
4018
.
Hashemi
S. E.
Heidarpour
M.
Mostafazadeh-Fard
B.
2011
Nitrate removal using different carbon substrates in a laboratory model
.
Water Science & Technology
63
(
11
),
2700
2706
.
Ho
G. E.
Gibbs
R. A.
Mathew
K.
1990
Bacteria and virus removal from secondary effluent in sand and red mud columns
.
Water Science and Technology
23
(
1–3
),
261
270
.
Hong
S.
Zhang
J.
Feng
C.
Zhang
B.
Ma
P.
2012
Enhancement of nitrate removal in synthetic groundwater using wheat rice stone
.
Water Science & Technology
66
(
9
),
1900
1907
.
Kottegoda
N. T.
Rosso
R.
2008
Applied Statistics for Civil and Environmental Engineers
. 2nd edn.
Wiley-Blackwell, Chichester
,
UK
.
Ma
Y.
Lin
C.
Jiang
Y.
Lu
W.
Si
C.
Liu
Y.
2009
Competitive removal of water-borne copper, zinc and cadmium by a CaCO3-dominated red mud
.
Journal of Hazardous Materials
172
,
1288
1296
.
Menkouchi Sahli
M. A.
Annouar
S.
Mountadar
M.
Soufiane
A.
Elmidaoui
A.
2008
Nitrate removal of brackish underground water by chemical adsorption and by electrodialysis
.
Desalination
227
,
327
333
.
Öztürk
N.
Ennil Bektaş
T.
2004
Nitrate removal from aqueous solution by adsorption onto various materials
.
Journal of Hazardous Materials
B112
,
155
162
.
Paramguru
R. K.
Rath
P. C.
Misra
V. N.
2005
Trends in red mud utilization: a review
.
Mineral Processing & Extractive Metallurgy Review
26
,
1
29
.
Seltman
H. J.
2014
Experimental Design and Analysis
.
Carnegie Mellon University Press
. .
Smiljanić
S.
Smičiklas
I.
Perić-Grujić
A.
Šljivić
M.
Đukić
B.
Lončar
B.
2011
Study of factors affecting Ni2+ immobilization efficiency by temperature activated red mud
.
Chemical Engineering Journal
168
,
610
619
.
Suryanarayana
C.
Ivanov
E.
2013
Mechanochemical synthesis of nanocrystalline metal powders
. In:
Advances in Powder Metallurgy: Properties, Processing and Applications
(
Chang
I.
Zhao
Y.
, eds).
Woodhead Publishing
,
Sawston, UK
, pp.
42
68
.
Tor
A.
Cengeloglu
Y.
Aydin
M. E.
Ersoz
M.
2006
Removal of phenol from aqueous phase by using neutralized red mud
.
Journal of Colloid and Interface Science
300
,
498
503
.
Tor
A.
Danaoglu
N.
Arslan
G.
Cengeloglu
Y.
2009
Removal of fluoride from water by using granular red mud: batch and column studies
.
Journal of Hazardous Materials
164
,
271
278
.
Zabihi
M.
Ahmadpour
A.
Haghighi Asl
A.
2009
Removal of mercury from water by carbonaceous sorbents derived from walnut shell
.
Journal of Hazardous Materials
167
,
230
236
.
Zhang
Y.
Chen
W.
Lv
G.
Lv
F.
Chu
P. K.
Guo
W.
Cui
B.
Zhang
R.
Wang
H.
2012
Adsorption of polyvinyl alcohol from wastewater by sintered porous red mud
.
Water Science and Technology
65
(
11
),
2055
2060
.
Zhao
Y.
Zhang
L.
Ni
F.
Xi
B.
Xia
X.
Peng
X.
Luan
Z.
2011
Evaluation of a novel composite inorganic coagulant prepared by red mud for phosphate removal
.
Desalination
273
,
414
420
.