Adsorption/photodegradation of methylene blue from synthetic wastewater on titanate nanotubes surfaces

Sweet water is one of the primary requirements for the Earth to survive but it is polluted in a number of ways. Directly or indirectly, inorganic and organic pollutants/contaminants from the manufacturing industries or generated as by products from the different industries/factories are discharged and are drained away (Vimalkumar et al. 2018). These anthropogenic by products cause water pollution for both surface water and groundwater, showing adverse effects to human health. Among the organic pollutants, the water soluble organic dyes are of concern in the present context. They may be natural or synthetic (Anitha 2018). Due to the lack of rules and regulations, several thousand mg/L dyes are used for different purposes and discharged; not only that, but they are also directly discharged to the environment from the commercial synthetic industries too. More than 10,000 dyes have been widely used in the textile, paper, rubber, plastics, leather, cosmetics, pharmaceutical and food industries (CPCB 1990). The discharge of coloured wastes to the water bodies affects the aesthetic properties of water (Mungondori et al. 2018). The dye effluent is highly toxic in nature as it contains high suspended solids, chemical oxygen demand (COD), dyes and chemicals along with high concentrations of heavy metals like Cu, Cd, Zn, Ni and Pb. The dye effluents contaminate the surface and ground water, which makes it unfit for irrigation and domestic uses (Rastogi et al. 2008).

Different organics bearing synthetic wastewater have been tried to remove them from the aquatic environments. There have been attempts to degrade tannery wastewater (Zhao et al. 2017), methylene blue (Mungondori et al. 2018; Santos et al. 2018), Orange II (Demircivi & Simsek 2018), methyl orange (Wang et al. 2018) and many more using only photo-illumination studies and no adsorption studies have been reported. Hence, the organics removal efficiency (%) of the used materials was understood but not the adsorptive capacity of the used precursors (mg/g). Again, the adopted technologies used composite/doped materials, which add an extra cost. Hence, there is a chance to leach the fabricated materials into the effluents, which again demands the expensive post treatment of the effluents.

Among different organic dyes, methylene blue (MB) has been used to treat cyanide poisoning and urinary tract infections, sulphide analysis and water testing, as a biological stain, an indicator, and as a medication used in Methemoglobinemia. It causes some common side effects include headache, vomiting, confusion, shortness of breath and high blood pressure (Rastogi et al. 2008).

Many further efforts have been made with different natural and synthetic materials with different technologies to remove these mentioned toxic organic contaminants from the aquatic environment (Demircivi & Simsek 2018). Different technologies along with adsorption (Demircivi & Simsek 2018), photodegradation (Wang et al. 2018; Ge et al. 2019), coagulation (Zhao et al. 2017), chemical oxidation (Baban et al. 2003), biological process (Ledakowicz et al. 2001) and many more have been implemented to remove/decolour dyes from aqueous media. The developed technologies were found to be good towards dyes, however having a number of limitations, especially the produced sludge. Recently, photocatalytic processes have shown a great potential, being cost effective, environmentally friendly and in the field of sustainable treatment technology with zero waste discharge (Santos et al. 2018). Numerous semiconductor photocatalysts have been used to degrade the solute dye(s) from water medium using UV rays. The common semiconductors photocatalysts used are TiO₂ (Anatase, P-25), SnO_2, ZnO and doped or composite photocatalyst are also used because of their low band gaps (for TiO_2, SnO₂ and ZnO these are 3.2 eV, 3.6 eV and 3.37 eV, respectively). Among these catalysts, titanium dioxide (TiO₂) is found to be used more frequently due to its low band gap. Apart from this, TiO₂ has been studied most commonly due to its high stability, non-toxicity, high catalytic activity and high conductivity. TiO₂ is eco-friendly and shows high photocatalytic activity due to its larger surface area (Santos et al. 2018). It has been found that the TiO₂-derived materials like TiO₂ nanotubes also have the tendency to degrade dyes and thus can be used as a photocatalyst. Many studies have been done by taking titanium nanotubes, but an extensive study has not been done yet (Liu et al. 2014). Hence an attempt has been made to remove MB from water medium using TNT as having a low band gap of 3.2 eV. It is true that biosorbents are very efficient not only towards dye(s) but also for heavy metals. At the same time it is also true for activated carbon, which is a superior adsorbent for any kind of contaminants, but does not have any band gap for the conduction of electron(s) from the valence band (VB) to the conduction band (CB) to degrade the organics. Again, the activated carbon is costly compared to of TiO₂. In this study we try to use both the properties of TNT, that is, adsorptive and photodegradation, which is due to the electronic transition from VB to CB to degrade the organic dye, MB, from the aquatic system in the presence of an adequate amount of oxygen in the suspension.

However, the objectives of the study are as follows:

• (i)

  Synthesis of titanate nanotubes and hydrogen titanate nanotubes by hydrothermal method in the laboratory

• (ii)

  Characterization of both synthesised precursors by XRD, SEM/EDX and TEM (for TNT only)

• (iii)

  Using the synthesized precursors, an attempt is made to remove MB from the aquatic environment

• (iv)

  To understand the optimal conditions for uptaking of the solute on the precursors' surfaces

• (v)

  An attempt was made to fit the obtained data of the adsorption process in isotherm and kinetics studies with a diffusion model

• (vi)

  Turnover study of the spent exhausted materials was carried out for the reusability to make it economic

• (vii)

  An attempt was made to remove spiked MB from the aqueous media applying the developed technology

• (viii)

  The probable photodegradation mechanism of organics is proposed

Analytical grade reagents were used as received. Aqueous solutions were prepared in double distilled water. MB stock solution (500 mg/L) was prepared using C₁₆N₃H₁₈ClS (λ_max = 663 nm) (Himedia) and diluted as required. Titanium dioxide (TiO₂) was purchased from Loba Chemicals and concentrated HCl, NaOH, NaNO₃, Na₂HPO₄, Na₂SiO₃.9H₂O, Na₂EDTA.2H₂O, CaCl₂, NaCl, FeSO₄, Na₂SO₄ were procured from Merck (India). Appropriately diluted solutions were prepared as and when required.

The as-purchased TiO₂ powder (anatase, P-25) was blended with 10 M aqueous solution of NaOH (25 g in 100 mL) and stirred for 1 h at 520–540 rpm, in a beaker placed on an electrically operated magnetic stirrer. The suspension was then taken in a Teflon lined sealed stainless steel hydrothermal unit and then put in a hot air oven at 150 °C for 14 h. The desired temperature raised within 1 h. Hydrothermal treatment causes the formation of TNT (Liu et al. 2014). After the completion of aging time, the cooled suspension was then repeatedly washed in double distilled water (pH ∼7) and centrifuged at 9,000 rpm (Remi R-24) to separate the solids from the suspension. The obtained white slurry material dried at 90 °C for 6 h and kept in a dry black container and stored in a dark place, characterised by XRD, SEM/EDX and TEM. As-prepared TNT was treated with 10⁻² M HCl (3 g in 10 mL, 1 h interval, 6 times), which may help to replace Na⁺ by H⁺. The acid-treated the suspension was repeatedly washed in double distilled water after each time of acid treatment. After achieving pH ∼ 7, suspension was centrifuged and dried as mentioned, characterized, and called HTNT (Thorne et al. 2005; Liu et al. 2014). Efficiency of precursors was understood by the adsorption/photodegradation of MB from synthetic and from spiked pond water. A close agreement on % dye removal efficiency of both synthesized precursors was evaluated. Hence, the entire studies were conducted using TNT only.

SEM/EDX study was conducted using FEI-SEM (Apreo LoVac) operating at 20 kV for both TNT and HTNT adhering 5–10 mg of each sample separately in the supplied sample bar (without sputtering) and introduced into the instrument through the provided sample holder and operated. The produced electron beams from the cathode strike on the surface of the precursors (2 μ in depth), and produce images with the components present at a particular position of the materials (EDX), and are snapped.

TEM analysis was conducted using JEM-1400, JEOL, Japan, operating at 120 kV. The sample was prepared by dropping 10 μL of aqueous TNT solution (using a micropipette) on a carbon coated copper grid and dried for seven days on a Petri dish at room temperature. Finally, the sample coated copper grid was dehydrated with 100% acetone (20 μL, two times with two days interval) and dried at room temperature for another seven days. Dried sample coated (carbon coated) copper grid was inserted into the instrument with the provided sample holder. The electron beams produced in the cathode strike on the sample, passing through the vacuum, showed the image, and snapped.

In a 500 mL beaker (55.44 cm², bed height from light source to solution is 13.2 cm), 3 g/L dose of as-prepared TNT/HTNT was dispersed in 50 mL synthetic MB solution with a concentration of 10 mg/L and at a pH of 6.8 ± 0.2. Placing the beaker with the suspension on a magnetic stirrer, agitated at 520–540 rpm with a magnetic needle/bar ensuring the system had enough oxygen. Dose and dye concentration were in optimum conditions. Above the beaker, a set up of UV lamp was placed as a source of light (photon, Figure 1). The study was conducted in the absence/presence of UV light (NARVA-Germany UV-C, 18 W, 360 nm). The experiment was conducted at 25 ± 2 °C covering the set up with a black cloth to prevent hazardous UV scattering (Figure 1) to the premises. After the completion of aging (optimum), the suspension was centrifuged at 9,000 rpm (Remi, R-24) for 8 min, the supernatant aliquot was collected (not filtered) carefully using a micropipette for absorbance measurement to determine the effluent dye concentration, preparing a standard calibration curve of MB (range 0–8 mg/L), using a spectrophotometer (UV-1800, Shimazdu, Japan) equipped with 1-cm quartz cell.

It is important to mention here that during adsorption of dye on TNT/HTNT in dark (absence of UV light), might have a little chance to degrade in the presence of IR, but at so very slow a rate that we only consider the adsorption process to be prominent.

Spectrophotometric method/technique was adopted to determine the remaining dye concentration in the effluents after treatment. The spectrums of the dye were recorded at 663 nm (λ_max). The standard calibration curve was drawn for the organic dye MB in the concentrations ranges of 0.0–8.0 mg/L, absorbance = 0.202× (mg/L) +0.0695, (R² = 0.997) and was used for further studies.

Treatment of TiO₂ with strong aqueous solution of NaOH, the positive charge Na⁺ is encapsulated between the edge-shared of TiO₆ and octahedral layers of Na₂Ti₃O₇. In this critical situation, a positive strong generated static attractive force holds Na⁺ and TiO₆ units tightly together, prohibiting the rolling of these layers into the nanotubes. In hydrothermal treatment over a period of time, Na⁺ is gradually un-encapsulated with intercalated H₂O molecules into the interlayer space of the TiO₆ sheets. The larger size of H₂O molecule compared to Na⁺ helps to enlarge the interlayer distance during Na⁺ un-encapsulation, weakening the attractive static interaction force existing between TiO₆ and the octahedral layers. Finally, layered Na₂Ti₃O₇ particles were then gradually exfoliated to form numerous sheet-shaped products and rolled to nanotubes. The nanotube formation mechanism is shown in Figure 2 (Wei et al. 2005; Liu et al. 2014).

The crystallinity of the synthesized nanomaterials; that is, TNT and HTNT, by alkaline hydrothermal treatment was understood from the analysis of the obtained XRD pattern, although there are some controversial issues regarding this (Thorne et al. 2005; Liu et al. 2014). The study was carried out using a powder X-ray diffractometer (Shimazdu XRD 6100) with Cu K_α radiation of wavelength λ = 0.15406 nm in the scan range of 2θ= 20°–80° with a data acquisition rate of 0.033° per step,

A lot of crystal modifications are reported between the forms of polytitanic acids, H_{2 m}Ti_nO_{2n +m} and pure TiO₂. Again, it is really difficult to assign the Miller indices of the formed nanomaterials in a particular crystallographic axis as during the formation of nanomaterials by alkaline hydrothermal heat treatment, it changes its axis during wrapping and winding in a certain direction. In addition, the titanate nanotubes are relatively unstable and can easily undergo phase transformation during heat, acid or any other chemical treatments, during or after preparation of nanotubes (Thorne et al. 2005; Liu et al. 2014). Figure 3 shows the XRD pattern of TNT (sample-1) and HTNT (sample-2). Analysis of the obtained XRD pattern by JCPDF-00-044-0131, the well known software, suggested that both the samples do not contain any rutile or anatase phase of TiO₂. Similar types of diffraction peaks are also observed in both the samples except for a little change in intensity. The diffraction peaks of both samples can be correlated with A₂Ti_nO_2n+1 crystallographic series with n= 5 and A = H, Na (Thorne et al. 2005). No additional peaks in the XRD pattern appeared after acid treatment (HCl/H₂O) of sample-1, except a little change in intensity compared to that of sample-1. In the present study, it has been observed that the XRD pattern of sample-2 matches fairly well with the monoclinic (C2/m) H₂Ti₅O₁₁·H₂O structure (JCPDF-00-044-0131, well-known software). These results suggest that sample-2 originated from sodium titanate phase. A comparison of the observed 2θ values (in degree) and interplanar spacing (d_hkl) along with the standard values are summarized in Table 1.

Surface morphology and atomic percentage compositions of both TNT and HTNT were understood by SEM/EDX. With some impurities, the main composition of the synthesized precursors was found to be the titanium, as P-25 is ∼99% pure. The EDX spectrums are shown in Figure 4(a) and 4(b). TEM analysis of the precursor TNT only (Figure 5) was performed to understand the size and shape of the precursor and was determined to be a diameter of 50 nm and tubular in shape.

Contact time between the solute and sorbent is one the important prime parameters that enables us to inform the boundary condition between the two components to achieve the maximum saturation point for the uptake of solute onto the sorbent surfaces, although it will depend on both the nature of solute and sorbent. At the optimum conditions, initially the suspension was allowed to agitate in the absence of UV light (adsorption) to evaluate the adsorption point of saturation. Results of the average of three replicate experiments revealed within 1.5 h of agitation, saturated ∼85% dye adsorbed (in dark), and after 4.5 h of photo-irradiation, with continuous agitation of the same suspension with the same speed of 520–540 rpm, the final dye removal was achieved to be ∼98% (Figure 6, total spent time is 1.5 h + 4.5 h = 6 h). It can be explained by saying that adsorption efficiency (% removal) of the used precursor (TNT) is increased until the numbers of active adsorbent sites (S_BET = 186.8 m²/g, Sorptomatic) are available, remaining minimum numbers of semiconductor photocatalyst TNT sites, which may not be exposed in the adsorption process, may have a chance to degrade the solute in the presence of UV light and in the presence of an adequate amount of oxygen due to transfer of electron(s) from the valence band (VB) to the conduction band (CB). The photo-illumination study, in the presence of UV light for the second case, in the beginning, the removal efficiency (%) rate of MB was increased and then suppressed. The obtained results may be due to the presence of fresh and active TNT photocatalyst surfaces and may be due to the simultaneous adsorption and photodegradation of the solute molecules onto the precursor's surfaces. As the fresh and active sites are slowly exhausted/exposed, photodegradation of the solute is also suppressed.

The solute (chemical moieties, here dye) removal process depends on both the nature of the solute and sorbent and each and every solute has the maximum affinity towards a particular sorbent and vice versa at a particular temperature. Hence, to understand the optimal solute concentration onto the semiconducting sorbent TNT surfaces, a complete photo-illumination study was conducted. Five levels of MB concentration (6, 9, 10, 11 and 13 mg/L) were tried at room temperature. The adsorbent dose was fixed at 3 g/L. All other experimental conditions were kept constant. The removal efficiency, which depends on MB concentration (the number of molecules of chemical moieties) varied from ∼99% to ∼80%. The number of adsorbent sites is assumed to be the same/constant in each case, exhibiting decreasing removal efficiency with increasing solute molecules (Vimalkumar et al. 2018). However, the obtained result suggested that ∼98% MB was removed with a concentration of 10 mg/L and was used for further studies.

Active precursor sites are also important to adsorb or degrade the chemical moieties. With a constant number of chemical moieties (dye molecules), the uptake capacity of any sorbent increases with the increase in its fresh and active sites. This helps to conduct the study from which the required optimal dose of the precursor for the particular number of molecules of any chemical moieties is to be understood. Hence, a complete photo-illumination study was conducted, varying the doses from 0.5–3 g/L at the optimal conditions with a fixed MB concentration of 10 mg/L. The removal efficiency increased from ∼60% to ∼98%, as it depends on the precursor sites. It was observed that 3 g/L dose of TNT was able to remove ∼98% MB from the aqueous media and was optimized for the entire studies.

The effects of dose as well as the MB concentration on the removal of MB can be determined by expressing % removal of MB as a function of the ratio between the MB concentration and dose of TNT, and is presented in Figure 7, to understand the synergistic effect between the solute and sorbent during the adsorption process in dark. The equation (%) removal of MB = −2.3452× {[MB]/Dose of TNT} +102.3848; R² = 0.9289, correlating these are also shown in the figure. It can be noticed from the graph that the % removal of MB decreases steadily with increase in the ratio between solute and sorbent.

In the suspension, during adsorption, where the solute and sorbent come close to each other, due to the surface properties of the sorbent, it attracts the solute moieties towards itself at a particular temperature and, at the same time, the solute molecules also reverse back from the sorbent surfaces to the solution; that is, a dynamic equilibrium is established. In this critical condition, each and every sorbent has the capacity to adsorb the maximum amount of solute from the suspension although it will depend on the agitation speed of the suspension. Hence, to understand the sportive behavior of TNT towards MB (dose variation study, 0.5–3 g/L, 10 mg/L, other conditions fixed), isotherms namely Langmuir and Freundlich were carried out (Vimalkumar et al. 2018).

To interpret the mechanism of solute sorption onto the sorbent, four kinetic models; viz., the first order reaction model based on the solution concentration, the pseudo-first order model based on the solid capacity, the second order reaction model based on the solution concentration and the pseudo-second order reaction model based on the solid phase sorption were understood and a comparison of the best fit sorption mechanism was made (Al-Kadhi 2019). Four integrated linear equations were represented as:

• 1.
  First order reaction model presented as:

  

  (5)
• 2.
  Pseudo-first order reaction model presented as:

  

  (6)
• 3.
  Second order reaction model depicted as:

  

  (7)
• 4.
  The pseudo-second order reaction model written as:

  

  (8)

  h (initial sorption rate constant) = kq²_e

• Terms and symbols are as usual

Based on the specific co-relation co-efficient (R²) values, Figure 9(a)–(d), we can say that the adsorption of MB on TNT surfaces followed the pseudo-second order reaction model better than the three others. The corresponding linear equations with specific co-relation co-efficient values (R²) are shown in Table 3.

The calculated D_f and D_p values were found to be 3.85 × 10⁻¹³ and 2.30 × 10⁻¹⁷. Film diffusion is the rate limiting step if D_f value is in the range of 10⁻⁶ to 10⁻⁸ and for the D_p it should be in the range of 10⁻¹¹ to 10⁻¹³ (Al-Kadhi 2019). So, the obtained result in the present work suggested the adsorption of MB on semiconducting material TNT surfaces followed neither film diffusion nor pore diffusion like heavy metals or metalloids (Table 4) (Maji et al. 2011) and may behave differently with the sorbent surfaces.

In synthetic solution, a complete photodegradation (6 h) study was conducted in order to understand the effect on MB removal efficiency (%) in the presence of different foreign ions in optimum conditions looking into the real dye-bearing sample on TNT surfaces. The obtained results are summarized in Table 5. It was observed that removal efficiency (%) of MB was suppressed with the increasing concentrations of each ion. A similar result was also obtained in the case of Fe (II) in the range of 0.0–0.5 mg/L (∼98–67%). This attributed result may be explained by saying that salts of each ion may have the chance to coat onto the TNT surfaces, causing the prohibition of electronic transfer from VB to CB to mineralize the solute dye MB (Sahoo et al. 2005).

Again, due to the low band gap of TNT, transits the electron(s) transit from VB to CB in the presence of photon(s) produced from UV light (in the presence of an adequate amount of oxygen in the suspension) and during this process electron-rich free radicals and are formed as mentioned above. Finally, an oxidation or reduction, or both, may have taken place due to the transference of electron(s) from HOMO to LUMO of the used chemical moieties and degradations. The probable mechanism of adsorption and photodegradation of MB is shown in Figure 10. In this process there is no need to use any H₂O₂, which is unstable at room temperature, or Fenton reagents, which are also additional chemicals with additional cost (Ge et al. 2019). Again, in the present technique, the use of UV light kills the pathogens in the effluent, if any, which is one of the biggest prospects of concern for environmental pollution.

A complete photodegradation study was conducted at optimal conditions, varying the pH of the aqueous media, adjusting from 2.0–10.0 with diluted aqueous HCl or NaOH. The obtained result revealed the slight suppression of degradation within ±2.0% with pH. The obtained results may be due to the change in the structural property of MB (Sahoo et al. 2005).

The as-prepared TNT was applied to remove MB from the spiked tap and pond water. The pre-and post treated water quality parameters are shown in Table 6. It was found that at optimum conditions, MB removed ∼97% and ∼92%, respectively from pond water and tap water and may be useful for practical purposes.

Regeneration and re-use of the spent material with suitable solvent is meaningful, which reduces the synthesis cost. Hence, to make the economy of the exhausted materials, a turnover study was performed. Two times turnover study of the exhausted TNT was conducted (at optimum conditions, complete photo-illumination study) after successive washing with 100% EtOH (and drying) six times, agitating at 520–540 rpm at 25 ± 2°C until ∼85% of adsorbed MB was procured each and every time separately. The % MB up-take capacity of the regenerated TNT was found to reduce gradually (∼72% and ∼54%); this may be due to unavailability of active and fresh surface sites of precursor.

Researchers used different materials to elucidate their efficiency towards MB at room temperature from the aquatic system (adsorption process only). Hence, a comparative study was made to compare the MB removal efficiency of the present developed material, TNT, with others. Table 7 shows the corresponding results.

TNT/HTNT was synthesized hydrothermally and was characterized by SEM/EDX, XRD and TEM analysis. The as-prepared precursors were applied to remove MB from the aquatic system. The technique followed both adsorption (in dark) and photodegradation (in the presence of UV light). The adsorption saturation point came after 1.5 h (∼85% removal) of stirring whereas ∼98% (final) MB was removed within another 4.5 h of photo-irradiation at an initial concentration of 10 mg/L, at a dose of 3 g/L at room temperature of 25 ± 2 °C and at pH of 6.8 ± 0.2. Adsorption followed the Langmuir isotherm better than Freundlich. The efficiency of TNT was understood from the Langmuir isotherm and was found to be 151.51 mg MB/g TNT. The adsorption process followed the pseudo-second order reaction model. The adsorption process was found to be favorable but thermodynamically non-spontaneous and did not follow either film diffusion or pore diffusion. The developed precursor was applied to remove MB from spiked tap and pond water with water quality measures. A turnover study was also conducted to find the efficiency (%) of the spent and exhausted materials, if any, for another two cycles. A probable dye removal mechanism on TNT surfaces was proposed. The technique was applied to remove MB from spiked pond water in the same experimental conditions with water quality measures.

Authors are thankful to the DST, SERB, Government of India, for the financial support for this work under the contract number of ECR/2016/001315. For SEM/EDX and XRD, we are thankful to Dr Rupam Goswami, BITS, Pilani and Dr Arup Kole, Durgapur Women's College, WB.

All relevant data are included in the paper or its Supplementary Information.