Introduction

Arsenic (As) is a Class I carcinogen due to its high toxicity to organisms1,2. In the past few decades, millions of people around the world suffered from various chronic diseases related to regular consumption of As contaminated water3,4. It has become a prerequisite to remove As from wastewater to prevent As contamination to our ecosystems. To date, many advanced technologies have been developed for this purpose, such as membrane separation, ion exchange, electrokinetics, coagulation and adsorption5,6,7. Coagulation and adsorption are well-studied strategies in water treatment due to their simplicity and cost-effectiveness8,9. A lot of coagulants/adsorbents have been reported with good As removal performance, such as oxides or hydroxides of aluminum (Al), iron (Fe), manganese (Mn), magnesium (Mg), zinc (Zn), titanium (Ti), zirconium (Zr), lanthanum (La), cerium (Ce), and yttrium (Y)8,9,10,11,12,13. However, existing adsorbents usually have low adsorption capacities at low As concentration, which leads to large amount of As-containing hazardous sludge that needs further disposal8,9. Therefore, there is still a great need to develop more efficient As-removal reagents and related technologies.

MgO is an eco-friendly material with low cost14. At present, MgO-based materials have been widely studied for water As removal, but most of them focus on MgO-based composites, such as Mg–Al layered double hydroxides15,16, zinc-magnesium oxides (ZnO-MgO)-based materials17, Mg-Fe-based hydrotalcite and oxides18,19,20. Although these MgO-based hybrid materials exhibited good performance on As removal, their complicated synthesis process will lead to increasing production cost and hinder their large-scale practical application. In fact, pure MgO or Mg(OH)2 itself already possesses high adsorption capacity to As21. MgO nanoparticles were reported with a maximum As adsorption capacity up to 115.27 mg/g10. In another study, nest-like micro-/nanostructured N-MgO was developed with a maximum adsorption capacity of 378.79 mg-As(V)/g-MgO, which was much higher than other micro-/nanostructured metal oxides22. Mechanism analyses indicated that As species were all removed by Mg(OH)2 because MgO hydrolyzed quickly forming Mg(OH)2 after it was added into water22,23. Because the in-situ formed Mg(OH)2 from MgO hydrolysis had more high affinity surface hydroxyl groups than pre-formed Mg(OH)2. Thus, generally, MgO performed better than Mg(OH)2 on As removal. Nevertheless, the cost of MgO, especially nanoparticles, is still too high for scale application in water treatment. Considering low material cost, MgCl2, a byproduct of salt industry and raw chemical for MgO production, is the cheapest Mg-related reagents. It might be much more cost-effective to use MgCl2 instead of MgO for water As removal. However, researches on using MgCl2 as As-removal reagents are still rare to date.

In-situ formed metal hydroxides were found more effective to adsorb pollutants than there pre-formed counterparts24,25,26. Therefore, we hypothesize that in-situ formed Mg(OH)2, using MgCl2 as Mg2+ source, might perform better than nano-MgO on water As removal. To verify this hypothesis, a simple comparison was conducted between MgO nanoparticles and in-situ formed Mg(OH)2 (by adding MgCl2 and NaOH) to reveal the advantages of As removal efficiency and kinetics of in-situ formed Mg(OH)2. Arsenate (As(V)) was used to simulate As-contaminaed wastewater because it is usually the most abundant As specie in surface water and arsenite (As(III)) is ease to be transformed into As(V) through an oxidation pre-treatment5,6,7. Detailed mechanism analysis was performed based on multiple analysis of the formed Mg-As sludge.

Experimental

Synthesis of nano-MgO

All reagents in this study were analytical grade and used directly without any further purification. Nano-MgO was synthesized following our previous report27. Briefly, 100 mL of Mg2+ solution (1 mol/L) was prepared by dissolving magnesium chloride in distilled water. Then, 15 mL of ammonium water (28 wt.%) was added quickly into Mg2+ solution with vigorous stirring to form Mg(OH)2. After boiling to let water evaporate, the obtained Mg(OH)2/NH4Cl mixture was then annealed at 450 °C for 2 h. Finally, loose white powder, nano-MgO, was obtained after simple grinding. The specific surface area and mesopore volume of synthesized nano-MgO were 66.83 m2/g and 0.135 cm3/g, respectively. They were acquired from the N2 adsorption/desorption isotherm which was measured using an automated gas sorption analyzer (Autosorb, Quantachrome, USA)27.

Batch experiments

All batch experiments were conducted in a glass reactor contained 200 mL of 10 mg/L inorganic As(V) (Na2HAsO4·7H2O, Sigma-Aldrich) solution which was stirred by a magnetic stirrer. As(V) was chosen as the model As species because As(V) is the major component and the most stable form of As in aquatic environments28. For nano-MgO treatment, nano-MgO powder was added to each reactor to desired concentrations under stirring (~ 150 rpm). For in-situ formed Mg(OH)2 treatment, 0.1 mol/L MgCl2 and NaOH solutions were added successively drop by drop to each reactor to desired concentration. Specifically, in As(V) removal experiment, there were 5 application dosages (0.5, 1.0, 1.5, 2.0, and 2.5 mmol/L) for both nano-MgO and in-situ formed Mg(OH)2. All glass reactors were sealed and stirred (~ 150 rpm) at 25 °C for 24 h. For adsorption kinetics, 2 mmol/L of material dosage was chosen based on the As(V) removal experiment, and samples were collected at desired time intervals.

The effect of pH on As(V) removal by in-situ formed Mg(OH)2 was explored with pH values ranged from 10.0 to 11.5. The pH range started at 10.0 because the pH of solution is ~ 10 when Mg2+ and OH react at molar ratio of 1:2 to form Mg(OH)2. Initial pH was adjusted by 0.1 mol/L NaOH, and the pH at the end of experiment was also tested. The effect of co-existed ions and humic acid (HA) were conducted with co-existed ions (Na+, SO42−, Ca2+, PO43−, and CO32−) of 10 mg/L or HA of 100 mg/L. All experiments were conducted at 25 ± 1 °C, except for the reaction temperature experiment, which was conducted at temperatures ranging from 25 to 40 °C.

There were three replicates for each experiment. At the end or desired time intervals, approximately 4 mL aqueous sample was collected and filtered through a 0.22 μm membrane filter. The concentration of residual As(V) was analyzed using a liquid chromatography-atomic fluorescence spectrometry (LC-AFS) (ELSPE-2, Guangzhou Pulin Sheng Technology Co., Ltd, China).

Characterization

To further explore the underlying mechanism, the precipitates were collected via centrifugation after coagulation, washed with distilled water and dried at 80 °C. Fourier transform infrared (FTIR) spectra of samples were recorded with KBr pellets in the range of 4000–400 cm−1 using a Thermo Scientific Nicolet 6700 spectrometer. The morphologies of adsorbents were observed using a field emission scanning electron microscope (SEM) (S-4800, Hitachi, Japan). The As elemental mapping images were acquired using energy-dispersive X-ray spectroscopy (EDX) module coupled with the SEM under an electron accelerating voltage of 20 kV. The x-ray diffraction patterns (XRD) were acquired using a powder diffractometer with Cu Kα radiation (λ = 1.5418 Å) (D2 PHASER, AXS, Germany). The elemental valence and abundance analysis was conducted by using an X-ray photoelectron Spectrometry (XPS).

Results and discussion

Performance for arsenic removal

The adsorption performance of nano-MgO and in-situ formed Mg(OH)2 were compared by adding them into a simulated As(V) contaminated wastewater and analyzing the As(V) removal efficiencies. The results indicate that in-situ formed Mg(OH)2 and nano-MgO realized similar removal efficiencies while in-situ formed Mg(OH)2 realized slightly higher As(V) removal efficiency (Fig. 1). As the concentration of nano-MgO increased from 0.5 to 1.5 mmol/L, As(V) removal efficiency increased from 29.7% to 91.0%, while that of using in-situ formed Mg(OH)2 increased from 34.2% to 95.7% at the same application levels. In other words, to obtain an As concentration less than 0.5 ppm, which is the municipal wastewater treatment standard in China (GB 8978–1996), more than 2.0 mmol/L of nano-MgO will be needed, but only 1.5 mmol/L of in-situ formed Mg(OH)2 are needed. Nano-MgO has been proved to possess high adsorption capacity for both organic and inorganic As10,23. Thus, these results indicate that in-situ formed Mg(OH)2 from cheap MgCl2 and NaOH has similar or even better As(V)-removal capability than pre-formed nano-MgO with a much higher price.

Fig. 1
figure 1

Comparison of the performance for As(V) removal by nano-MgO and Mg(OH)2-in-situ (initial As(V) = 10 mg/L, time = 24 h).

The adsorption kinetics of As(V) on nano-MgO and in-situ formed Mg(OH)2 were studied, as illustrated in Fig. 2. A removal efficiency growth and equilibrium stages were observed for both chemicals, especially for in-situ formed Mg(OH)2. Specifically, for in-situ formed Mg(OH)2, the growth stage was extremely fast and finished within only 10 min, and nearly 99% of As(V) in water was adsorbed. This may be attributed to the abundant active binding sites of in-situ formed Mg(OH)2. Because As(V) could be efficiently incorporated into both the inner and outer surfaces of in-situ formed Mg(OH)2 flocs during the progress of flocs growth. Similar phenomenon was observed in adsorption of As(III) by in-situ formed Ti(OH)424. On the other hand, the As(V) removal efficiency of nano-MgO, in contrast to that of in-situ formed Mg(OH)2, increased very slowly in the growth stage. It took as long as ~ 12 h to reach As(V) adsorption equilibrium. A possible reason might be that As(V) is removed during the phase transition of MgO into Mg(OH)2, which is a heterogeneous reaction between MgO and water, and will take longer time than a homogenous reaction between ions in water23. In addition, for the in-situ formed Mg(OH)2 treatment, the formed flocs settled down to the bottom very quickly within 30 min. This will facilitate efficient slurry separation via traditional gravity settling. Based on the above results, 10 min was chosen as the reaction time for the following experiments considering operational convenience and As(V)-removal efficiency. It should be noted that in-situ formed Mg(OH)2 didn’t realized higher As(V) removal efficiency but only accelerated the removal process. Thus, this process can’t be explained by a co-precipitation mechanism, which is process-controlled.

Fig. 2
figure 2

Comparison of As(V) removal kinetics between nano-MgO and Mg(OH)2-in-situ (initial As(V) = 10 mg/L, chemical dosage = 2 mmol/L).

Influence factors on arsenic removal by in-situ formed Mg(OH)2

Usually, pH was a key factor during adsorption of As(V). On the one hand, pH value affects As species in water according to their dissociation constants, which will influence the adsorption capacities of adsorbents29. On the other hand, pH value affects the formation of Mg(OH)2 which needs a slight basic pH condition24. In this study, As(V) removal by in-situ formed Mg(OH)2 was highly pH dependent and basic pH conditions could facilitate the As removal process (Figure S1). Differently, temperature had little influence on the As removal efficiency from 25 to 40 ℃, possibly due to the high reactivity of Mg2+ with OH (Figure S2).

Many ions co-exist with As in wastewater, such as Na+, Ca2+, and PO43−. Moreover, HA, a common natural organic matter, was frequently reported to affect As removal29,30,31,32. Therefore, it is imperative to consider the presence of coexisting cations, anions, and HA while evaluating As(V) sorption capacity of in-situ formed Mg(OH)2. Here, two kinds of cation, Na+ and Ca2+, three kinds of anion, SO42−, CO32−, and PO43−, and HA were chosen to investigate the effect of co-existed ions and HA on As(V) removal by in-situ formed Mg(OH)2. As illustrated in Fig. 3, Na+, Ca2+, CO32−, and SO42− had slight or negligible influence on As(V) removal even at a concentration up to 100 mg/L.

Fig. 3
figure 3

Effect of co-existed ions and HA on As(V) removal by in-situ formed Mg(OH)2 (initial As(V) = 10 mg/L, Mg(OH)2 = 1 mmol/L, time = 10 min). Here error bars mean standard deviation. A bar with an asterisk (*) represents significant difference from the control (p < 0.05).

Humic acid showed a slight negative effect on As(V) removal at high concentration of 100 mg/L, but no negative effect was observed at low concentration of 10 mg/L. The resistance to HA is an advantage of using in-situ formed Mg(OH)2. It was reported that the As(V) removal efficiency of an iron-cerium bimetal oxide material significantly decreased from 96.19% to 56.12% at the presence of only 10 mg/L HA29. Preformed sorbents are ease to encounter HA contamination on the material-water interface and lose their advanced adsorption property33. In this study, in-situ formed Mg(OH)2 has renewed material-water interface along with the in-situ formation of Mg(OH)2 so that the adsorption property could be retained25,27.

In contrast, PO43− was found to inhibit As(V) removal significantly with decreased efficiencies by 28.3% and 58.4% when encountering the interference of 10 and 100 mg/L of PO43−, respectively. This is consistent with existing reports1,29. It is known than P and As are congeners in the periodic table and have similar chemical properties. PO43− acts as a competitive ion of AsO43− during As(V) removal using in-situ formed Mg(OH)234.

Properties of the formed Mg-As sludge by in-situ formed Mg(OH)2

To explore the mechanism of the superior As(V) removal performance of in-situ formed Mg(OH)2, a series of characterizations were conducted on in-situ formed Mg(OH)2 flocs after As(V) adsorption. Firstly, the crystal structure of the in-situ formed Mg(OH)2 was characterized using XRD (Fig. 4). Obviously, there was no crystalline MgO or Mg(OH)2 in the in-situ formed Mg(OH)2 flocs compared to the standard XRD patterns of crystalline MgO (periclase) and Mg(OH)2 (brucite). Actually, no apparent diffraction peaks was observed in the flocs, indicating that the in-situ formed Mg(OH)2 existed as an amorphous form. Amorphous metal hydroxides were frequently reported with bigger specific surface areas, more surface hydroxy groups, and thus better adsorption performance35. Moreover, no apparent diffraction peaks were found for As minerals, indicating that As possibly existed as adsorbed ions. These results are consistent with existing reports on As coagulation by in-situ formed Ti(OH)4 and Fe(OH)324,25.

Fig. 4
figure 4

XRD image of MgO, Mg(OH)2 and the As(V)-loaded Mg(OH)2-in situ-As(V).

Micro-morphological results showed that the in-situ formed Mg(OH)2-As(V) sludge had a smooth bulk outlook (Fig. 5a). However, large numbers of wrinkles was seen on the surface and side at higher magnification (Fig. 5, b-d), indicating that in-situ formed Mg(OH)2 possessed a loose structure, which is consistent with its XRD pattern. Notably, the loose amorphous structure of in-situ formed Mg(OH)2 is completely different from the nanosheet-like Mg(OH)2 formed from nano-MgO hydrolysis23. This may be one of the reasons why in-situ formed Mg(OH)2 has superior As(V) removal performance, as an amorphous structure usually indicates a bigger specific surface area and more As binding sites.

Fig. 5
figure 5

SEM image of the Mg(OH)2-in situ-As(V) sludge at different magnifications.

The elemental mapping picture indicates that the removed As(V) element distributed uniformly within the Mg-As(V) sludge (Fig. 6). Thus, As(V) was highly possible encapsulated within the precipitate, rather than adsorbed on the particle surface. It is a common sense that adsorption occurs on sorbent-water interface and is partially or completely reversible. The encapsulation might be the main reason about the high and fast As removal efficiency, which could transform reversible superficial adsorption into irreversible encapsulated co-precipitation and thus alleviate the desorption or second release of adsorbed As(V) back into water phase.

Fig. 6
figure 6

Elemental mapping of As in the Mg(OH)2-in situ-As(V) sludge. (a) SEM picture, (b) As distribution.

FTIR analysis was used to identify the main functional groups of in-situ formed Mg(OH)2 after As(V) adsorption. As shown in Figure S3, compared to MgO and Mg(OH)2 without As(V) exposure, there was an absorption peak appeared at 841 cm−1 in in-situ formed Mg(OH)2 after reacting with As(V). This can be attributed to the stretching vibration of As-O bonds, indicating that As (V) was transferred from the simulated wastewater to the newly generated precipitate. XPS was used to analyze the elemental composition and chemical valence of Mg and As in the flocs. According to the full spectrum, the flocs mainly consisted of Mg, O, and As (Fig. 7a). The refined Mg 2p and As 3d electron spectra indicate that Mg existed as Mg(OH)2 (49.89 eV) and As existed as AsO43− (45.01 eV) (Fig. 7b). The comparable peak intensity indicates high As content in the formed Mg-As sludge.

Fig. 7
figure 7

XPS analysis of the as-formed Mg(OH)2-in situ-As(V) sludge. (a) Full spectrum, (b) refined spectrum of Mg and As.

EDX reveals a very low Mg/As molar ratio of ~ 6.3 (Fig. 8). These result indicates a very high atomic utilization efficiency of Mg2+ during the As(V) removal process, which could hardly be achieved by using MgO nanoparticles. As(V) was quite possibly encapsulated uniformly within amorphous Mg(OH)2 via coordination rather than forming magnesium arsenate crystal. Similarly, no magnesium arsenate was observed in As(V) removal by preformed commercial Mg(OH)221. The high Mg atomic utilization efficiency implies a low cost and highly reduced mass amount of Mg-As sludge. Therefore, the post-treatment of the hazardous As-sludge will be easier.

Fig. 8
figure 8

EDX spectrum image of the Mg(OH)2-in situ-As(V) sludge.

Mechanism of As removal using in-situ formed Mg(OH)2

Based on the above results, a simplified mechanism is proposed and shown in Fig. 9. During the successive addition of MgCl2 and NaOH solutions to As(V) contaminated wastewater, amorphous Mg(OH)2 tiny nuclei (Mg(OH)2(in-situ)) are formed with high affinity surface hydroxyl groups and big surface area. Simultaneously, As(V) is adsorbed onto the Mg(OH)2 surface. Then, the nuclei collide with each other and form bigger flocs with As(V) encapsulated within the flocs. Ultimately, Mg-As(V) composite flocs settle down to bottom and As(V) is removed from water. As the Mg(OH)2 sorbent is in-situ formed within the As(V) contaminated water, tiny Mg(OH)2 nuclei are formed with renewed Mg(OH)2-water interface from basic Mg2+ and OH ions with very abundant adsorption active sites. Meanwhile, the adsorbed As(V) hinders the further growth of Mg(OH)2 nuclei. Thus, amorphous Mg-As sludge is formed instead of Mg(OH)2 nanocrystals.

Fig. 9
figure 9

The possible mechanism for the enhanced As(V) removal from water by Mg(OH)2-in situ.

Conclusions

Efficient As(V) removal from water is realized by using in-situ formed Mg(OH)2, which is obtained by simply adding cheap MgCl2 and NaOH reagents. Notably, the removal process is very quick and can be performed within several minutes while traditional methods using MgO nanoparticles need several hours. This process is not only timely efficient but also cost-effective than using MgO nanoparticles as sorbent. Further, the amount of the resulted Mg-As sludge can be reduced significantly and thus adapt better to the demand of mass reduction on hazardous waste disposal. The mechanism still needs further study and a possible explanation is that in-situ formed Mg(OH)2 has more accessible and high affinity surface hydroxyl groups because of the precipitation reaction starting from basic Mg2+ and OH ions. As(V) is incorporated into the inner surface of in-situ formed Mg(OH)2 agglomerates, hinders its crystallization process, and finally forms an amorphous Mg-As precipitate. Although the feasibility in treating real wastewater in large scale is still unclear, considering the very fast, efficient As(V) removal process and low cost of MgCl2 and NaOH reagents, this As removal technology based on in-situ formed Mg(OH)2 has great potential for scale application and will imply a bunch of other in-situ formed nanomaterials for water treatment.