Struvite Precipitation for Ammonia Nitrogen Removal in 7-Aminocephalosporanic Acid Wastewater

7-Aminocephalosporanic acid wastewater usually contains high concentrations of ammonium (NH4+-N), which is known to inhibit nitrification during biological treatment processes. Chemical precipitation is a useful technology to remove ammonium from wastewater. In this paper, the removal of ammonium from 7-aminocephalosporanic acid wastewater was studied. The optimum pH, molar ratio, and various chemical compositions of magnesium ammonium phosphate (MAP) precipitation were investigated. The results indicated that ammonium in 7-aminocephalosporanic acid wastewater could be removed at an optimum pH of 9. The Mg2+:NH4+-N:PO43−-P molar ratio was readily controlled at a ratio of 1:1:1.1 to both effectively remove ammonium and avoid creating a higher concentration of PO43−-P in the effluent. MgCl2·6H2O + 85% H3PO4 was the most efficient combination for NH4+-N removal. Furthermore, the lowest concentration of the residual PO43−-P was obtained with the same combination. Struvite precipitation could be considered an effective technology for the NH4+-N removal from the 7-aminocephalosporanic acid wastewater.


Introduction
7-Aminocephalosporanic acid (7-ACA) is one of the key intermediates in the production of medically important semisynthetic cephalosporins, such as cephalaglycin and cephalothin. Currently, an enzyme-mediated process for the synthesis of 7-ACA from cephalosporin C has been recommended as an environmentally friendly technology compared to the conventional chemical synthetic process. During the enzyme-mediated process for the synthesis of 7-ACA, high levels of ammonium nitrogen (NH 4 + -N) and high chemical oxygen demand (COD) were found in the wastewater. NH 4 + -N present in wastewater at excess levels may deteriorate the receiving water quality [1]. In addition, NH 4 + -N is harmful to the local ecology [2]. Therefore, these compounds should be removed from the wastewater before entering into aquatic systems. However, the 7-ACA wastewater it is hard to bioremediate, because of the high concentrations of NH 4 + -N, a small quantity of cephalosporin and 7-ACA that can inhibit the growth of, and even destroy, harmful microorganisms. To overcome this difficulty, the precipitation of NH 4 + -N by forming magnesium ammonium phosphate (struvite, MgNH 4 PO 4 ·6H 2 O) is an attractive means of 7-ACA wastewater treatment. NH 4 + -N recovered by sturvite may be reused as slow release fertilizer. Struvite crystallizes is a white orthorhombic crystalline structure consisting of magnesium, ammonium, and phosphate in equal molar concentrations [3]. The basic chemical reaction to form struvite is expressed in Equation (1) The method of chemical precipitation of NH 4 + -N in the form of struvite has been studied widely from various types of wastewaters such as landfill leachate [5], industrial wastewater [6], source-separated human urine [7], anaerobic swine lagoon liquid [8] and semiconductor wastewater [9]. Münch and Barr [10] have reported that the success of struvite precipitation depended on two main factors: Mg 2+ :NH 4 + -N:PO 4 3− -P ratio and the pH of the solution. Li and Zhao [11] found that under an equal molar ratio of Mg 2+ :NH 4 + -N:PO 4 3− -P, the NH 4 + -N concentration could quickly be reduced from 5,618 mg/L to 112 mg/L by pretreating the chemical precipitation. Uludag-Demirer [12] and co-workers treated dairy manure by struvite precipitation and demonstrated that over 92% of NH 4 + -N removal was possible by adding Mg 2+ ions at a concentration higher than 0.06 M. Ryu [13] studied the struvite precipitation process in semiconductor wastewater at the field-scale and found that the optimum reaction for ammonium nitrogen removal occurred at a pH of 9.2. Marti [14]  The objective of this study is to investigate the removal of NH 4 + -N by struvite precipitation from 7-ACA wastewater using different magnesium and phosphate sources. In the experiments, the evaluations were focused on the following objectives: (1) optimizing the effects of operating parameters, such as the pH, Mg 2+ :NH 4 + -N:PO 4 3− -P molar ratio and mixing time for the precipitate; (2) appraising the performance of struvite precipitation on residual PO 4 3− -P and COD removal; and (3) examining the physical properties of the struvite products.

Batch Testing with Nine Combinations of Chemicals
In the first step of the struvite precipitation tests, nine combinations of chemicals denoted A1-A9 were tested with an initial NH 4   The addition of Na 3 PO 4 ·12H 2 O + MgSO 4 (A8), NaH 2 PO 4 ·12H 2 O + MgCl 2 ·6H 2 O (A5), 85%H 3 PO 4 + MgCl 2 ·6H 2 O (A4) or 85% H 3 PO 4 + MgSO 4 (A7) each achieved highly efficient removal of NH 4 + -N, with 70.92%, 67.83%, 74.28% and 70.02% of the total removed, respectively. To assess the quality of the struvite created through precipitation, the four combinations were analyzed by XRD and SEM analysis ( Figure 2 and Figure 3). The XRD pattern generated from these samples matched the database model for struvite. The combination of 85% H 3 PO 4 + MgCl 2 ·6H 2 O showed the strongest match, indicating that a relatively pure precipitate of struvite could be created using 85% H 3 PO 4 + MgCl 2 ·6H 2 O. The results obtained from SEM morphological analysis were compared with the XRD analysis. As shown in Figure 3, the needle-shaped spherical crystal precipitate of 85% H 3 PO 4 + MgCl 2 ·6H 2 O was more distinct than the others, and its size was regular (radius 25-28 nm). Therefore, 85% H 3 PO 4 + MgCl 2 ·6H 2 O is proposed as the best condition to achieve maximum removal of NH 4 + -N from the 7-ACA wastewater.
The COD reduction was lower when compared with the corresponding NH 4 + -N removal fractions in the experiment (Figure 1), which implies that the struvite precipitation technique is highly selective for NH 4 + -N. This result is in good agreement with those reported by Li [11] and indicates that a subsequent biological treatment process may be needed to remove the residual COD.
The change in the pH of the solutions during the experiments was similar regardless of the choice of chemicals used. A decrease in pH value was observed in the course of the struvite reactions ( Figure 1). Stratful [19] demonstrated that, in terms of thermodynamic equilibrium, hydrogen was released into the solution when struvite was formed, resulting in a decrease in pH.

Effect of pH
pH is an important factor for struvite precipitation because it has a direct influence on the solubility of struvite and its thermodynamic properties [7]. The optimum pH for struvite precipitation has been widely investigated. In previous literature concerning struvite precipitation, optimum pH values of 8.5 [20,21], 9 [22], 8.9-9.25 [8], and 9.5-10.5 [23] were reported. In this study, to determine the ( -P was at a stoichiometric ratio of 1:1:1. Figure 4 showed the obtained results.  Under otherwise constant precipitation conditions, changes in pH lead to a direct change in the degree of supersaturation during the precipitation process. At pH 7, no struvite was produced at detectable levels, while at pH 8, only a minute amount of very small crystals were produced. The growth of struvite crystals improved above pH 8, and the amount of precipitate at the bottom of beaker increased when the pH of the solution was gradually raised to 9. The struvite product was formed rapidly and settled quickly at the bottom of the beaker after stirring ceased at pH 9. However, the amount and the speed of formation of struvite precipitate decreased substantially at pH values of 10 and 11. Therefore, the best experimental ammonia removal was obtained at pH 9. At higher pH, the ammonia volatilization is serious. Air flow also plays an important role in ammonia-nitrogen volatilization. However, on the basis of the present experimental procedure (without stripping and only 15 min of stirring time) and also other findings in the literature [16,18], it can be concluded that ammonia volatilization is negligible on the removal of NH 4 + -N from the 7-ACA wastewater, as compared to struvite precipitation. It was likely that when the pH was excessively high, Mg 3 (PO 4 ) 2 was formed instead of struvite, which led to a decrease in the NH 4 + -N removal efficiency. H + in the reaction solution should inhibit struvite precipitation when the pH is lower than the optimum point, which agrees with the reduced precipitation observed at lower pH. The optimum pH for the removal of ammonia observed in this experiment was consistent with other studies. Booker [24] reported that pH 9.2 was optimum, whereas Tünay [25] found pH 8.5-9.3 to be the optimal range. The morphology of struvite precipitation was observed both above and below the optimum pH of 9 ( Figure 5).  -P in the 7-ACA wastewater was higher at pH < 8 than that at pH > 8 conditions. This may be because at low pH, further crystallization and precipitation of struvite was inhibited, and the residual concentration of PO 4 3− -P was maintained. The results indicate that the optimum pH values for the removal of ammonium and phosphate are different. This finding was consistent with the study of Booker [24] who reported that the maximum ammonium removal was found at pH 9.2, whereas the maximum phosphate removal was observed at pH 9.8.
Based on previous results, subsequent experiments were conducted at pH 9.0 with MgCl 2 ·6H 2 O and 85% H 3 PO 4 to investigate the effects of different molar ratios on the NH 4 + -N removal efficiency as well as on the residual PO 4 3− -P and COD.  Figure 6a). This may be due to the formation of other precipitates at higher molar ratios. For example, when an excess concentration of Mg 2+ is in highly alkaline conditions, solid phase Mg(OH) 2 may precipitate. The precipitation of Mg 3 (PO 4 ) 2 may also occur because the precipitation potential of this compound is enhanced by the addition of additional Mg substrate. These results agree with the findings of several previous studies [25,26]. However, some scientists [3,27] have shown that NH 4 + -N removal was generally affected by the amount of magnesium available to the struvite precipitation reaction. In particular, Stratful [19] reported that magnesium ions were a limiting factor for struvite precipitation. The difference between these two contrary results may be due to the properties of the applied water.

Effect of the Mg 2+ :NH 4 + -N:PO 4 3− -P Molar Ratio
The removal fraction of COD increased with an increasing concentration of Mg 2+ species. COD removal reached 20.   -P molar ratio was fixed at a ratio of 1:1:1.1, and the initial pH was 9.0. Overall removal of NH 4 + -N was observed to be similar at different mixing times. At short mixing times, the removal efficiency of NH 4 + -N was not significantly reduced. As the mixing time increased, the removal efficiencies of NH 4 + -N did not significantly increase. The mixing time between 5 and 60 min had a negligible effect on the production of struvite, suggesting that struvite crystals form homogeneously under these conditions and that precipitation is rapid. Examination of the precipitate by SEM microscopy revealed that the maximum crystal size increased with time ( Figure 8). Crystals up to 20 μm were precipitated at 10 min. At a mixing time of 60 min, the maximum crystal size had increased, with some crystals reaching lengths of 75 μm. Stratful [19] also investigated the effect of reaction time on the precipitation of struvite and obtained the same conclusion. Some of the crystals were broken with the time increasing, because of the low strength of the crystal, which were shown in Figure 8. We also found that the precipitation system was impeded. The residual phosphate was lowest at mixing time 20 min (Figure 7). A little amount of phosphate may be released from the broking struvite when the mixing time more than 20 min. Kim [9] investigated the effect of mixing intensity and mixing duration on struvite precipitation and reported that mixing enhanced the transfer of mass from the solute to the crystals, resulting in improved struvite crystallization and growth.

7-ACA Wastewater
The 7-ACA wastewater used in this study was taken from an enzymatic transformation-based production line for antibiotics at a pharmaceutical plant in Hebei, China. The wastewater was generated from the oxidative deamination and hydrolysis catalyzed processes of 7-ACA manufacturing. The characteristics of the 7-ACA wastewater are summarized in Table 2. The analysis techniques used for the 7-ACA wastewater were in accordance with the Standard Method for the Examination of Water and Wastewater [28].

Experimental Procedures
The experiments were performed at 298.15 K with a ZRS-6 variable-speed jar test apparatus (Tangshan Dachang Chemical Ltd., Tangshan, China). The jars were made of polytetrafluoroethene with dimensions of Φ 9.5 cm × 15 cm and held 1.0 L liquid. A two-blade propeller (polytetrafluoroethylene) with diameter of 2.5 cm and height of 7.6 cm was used for stirring. -P, and the mixing time. The detailed precipitation parameters are listed in Table 3. All batch experiments were performed in duplicate. The effectiveness of pH was investigated first. The test jar was filled with ammonia/phosphate solutions, and the pH was adjusted to the given values (from 7 to 11) in different jars using 1 mol·L −1 NaOH. The solutions were then stirred at 100 rpm for 15 min, followed by 30 min of quiescent settling. When the reaction time had elapsed, the pH was measured, and the precipitate that had formed was collected by double filtration through a 0.

Analytical Methods
COD, total suspended solids, NH 4 3+ -N, PO 4 3− -P, turbidity and pH analyses were performed at the Water Quality Lab, as described in the Standard Method for the Examination of Water and Wastewater [29]. Crystal phases of the struvites were obtained by XRD (D/max 2500PC, Rigaku, Tokyo, Japan) with Cu Kα radiation of wavelength 0.154 nm in the range of 2θ = 10-80° with a scan speed of 1.2 °/min. The morphologies of the struvites were analyzed by SEM (S-4800I, Hitachi, Tokyo, Japan) at 3.0 keV, which was equipped with an energy dispersive analysis system of X-ray (EDS).

Observation and Identification of Crystals
The struvites were washed with distilled water through the membrane filter and dried at 303.17 K for 72 h. The crystal size was examined using an Olympus BH-2 light microscope with a camera attachment. X-ray diffraction using a Siemens D5000 diffractometer and monochrome CoKa radiation (40 kV, 30 mA) was used to determine the identity of the precipitate. Scans from 2 to 75° 2θ were recorded with a scan speed of 0.08° 2θ per min. The scan length was 0.02°, and the time constant was 15 s by reference to Card Socabin from Diffract AT.

Conclusions
Struvite precipitation was applied for the removal of NH 4 + -N from 7-ACA wastewater. The effects of the operational parameters on struvite precipitation were also investigated. Based on the results of the experimental tests, the following conclusions could be drawn: (1) MgCl 2 ·6H 2 O + 85% H 3 PO 4 was the most efficient combination for NH 4 + -N removal compared with the other chemical combinations studied. Furthermore, the lowest concentration of the residual PO 4 3− -P was obtained with the same combination.
(2) pH was an important parameter in the removal of NH 4 + -N from 7-ACA wastewater. The optimum pH for NH 4 + -N removal was clearly observed at pH 9, and a slightly higher pH would be required for efficient residual PO