Journal of Industrial and Engineering Chemistry 58 (2018) 148–154
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry
journal homepage: www.elsevier.com/locate/jiec
Ammonium removal from digested effluent of swine wastewater by
using solid residue from magnesium-hydroxide flue gas
desulfurization process
Huanjia Lina,1, Yunqin Lina,1, Dehan Wanga,* , Yuwan Panga,b , Fabao Zhangb ,
Shuanhu Tanb
a
College of Natural Resources and Environment, South China Agricultural University, Guangzhou 510642, China
b
Institute of Agricultural Resource and Environment, Guangdong Academy of Agricultural Sciences, Guangzhou 510642, China
A R T I C L E I N F O A B S T R A C T
Article history:
Received 25 May 2016 This paper focused on the ammonium removal from the digested effluent of swine wastewater (DESW)
Received in revised form 1 September 2017 via MAP precipitation using the residue from magnesium-hydroxide flue gas desulfurization (MFGD)
Accepted 12 September 2017 process as a Mg source. The MAP precipitation was proceeded in a 500-mL beaker with a magnetic stirrer.
Available online 20 September 2017 It was found that around 60% of ammonium in the DESW was removed at pH 9.5, HRT 10 min and the P:
Mg:N molar ratio of 0.6:0.7:1. The NH4+ concentration (183.2 mg/L) in the DESW after MAP precipitation
Keywords: was lower than 200 mg L1, being alternative for the downstream biological process. A low content of
Digested effluent of swine wastewater PO43 (3.64 mg L1) and a high concentration of Mg2+ (67.17 mg L1) were also observed in the solution.
Ammonium removal
The main composition of the precipitate was proved to be MAP via XRD and SEM-EDX analysis. Besides,
Solid residue from MFGD process
the total chemical cost was saved by 29.17% using the MFGD residue as a Mg source instead of the
MAP precipitation
commercial MgCl26H2O. Hence, reusing MFGD residue as a Mg source to remove ammonium from the
DESW was feasible with cost saving, resource recovery and environment protection.
© 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights
reserved.
Introduction oxygen, eutrophication and red tidal phenomena in water courses
if the effluent is drained directly, maybe even resulting in some
Swine raising has been developed rapidly in China in recent worse consequences to the ecosystem [4,5]. Thus, appropriate
years. It was reported that around 65–70 * 103 swine farms had techniques are further required to remove pollutants in the DESW.
already been built in China, and simultaneously about 21 * 108 tons On the other hand, the biological conversion technology is known
of swine wastewater was produced annually [1]. How to deal with as an efficient and low-cost method for wastewater treatment, in
the huge amount of swine wastewater is a tough task for swine which pollutants are removed via biodegradation or bioconversion
farmers. Therefore, many related approaches to cleansing the [6,7]. For instance, ammonium is usually removed by autotrophic
swine wastewater are investigated, among which anaerobic nitrification (the conversion of NH4+ to NO3) and heterotrophic
digestion is well known as an essential basic process to remove denitrification (the conversion of NO3 to gaseous nitrogen). So far,
the organics in swine wastewater, along with the benefits of biogas biological processes like constructed wetlands and stabilization
generation [2,3]. However, the concentrations of chemical oxygen ponds have become good options for farmers adopted to treat the
demand (COD), NH4+–N and PO43 in the digested effluent of swine large amount of DESW [8,9]. However, the high ammonium
wastewater (DESW) are still much higher than the discharge concentration in the DESW always inhibits the microbe’s activities
standard of pollutants for livestock and poultry breeding in China in the biological process and reduces the removal efficiencies of
(GB 18596-2001). It might facilitate the depletion of dissolved pollutants including ammonium itself, even causing the failure of
the process [10–12]. For the success of the downstream biological
process, ammonium is usually the most critical element that needs
a prior removal from the DESW due to its relatively high
* Corresponding author. Fax: +86 020 85287672.
E-mail address:
[email protected] (D. Wang). concentration, treatability limitations and the high removal cost
1
These authors contributed equally. [13].
http://dx.doi.org/10.1016/j.jiec.2017.09.019
1226-086X/© 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
H. Lin et al. / Journal of Industrial and Engineering Chemistry 58 (2018) 148–154 149
So far, many physico-chemical processes which have widely contamination in power plants due to its higher desulfurization
been used can be applied in the removal of ammonium from performance, lower cost, less space occupation and pipe blockages.
wastewater, such as air stripping [14], membrane separation Simultaneously, the amount of the solid residue has increased.
processes [15], chemical precipitation [16]. Compared to air Therefore, various methods have been investigated to reuse the
stripping and membrane separation, chemical precipitation solid residue in the world. In Poland, the solid residue was made
showed obvious advantages in resource recovery and environment into MgSO4 fertilizer via aeration and crystallization, by which
protection by converting ammonium into a component of the solid more benefits in economy and environment were achieved
byproduct (MAP precipitate), instead of producing NH3 or N2 and compared to recycling the residue from limestone flue gas
releasing them to the air. Recently, chemical precipitation of desulfurization system [41]. In America, the solid residue is
ammonium by forming magnesium ammonium phosphate (MAP) calcined into SO2 and MgO, and then SO2 is reused to produce
precipitate has been studied widely [17–19]. Huang et al. [20] H2SO4, and MgO is reused as a basic refractory material to
investigated the recovery and removal of nutrients from swine manufacture crucibles [42]. While in China, the yield of the solid
wastewater by employing a struvite recycling process, which used residue from MFGD process is less because the MFGD process is
magnesium phosphate (MP) as the active component. Liu et al. [16] only applied in some small- or medium-scale power plants. No
studied the effects of the aeration rate and magnesium: phosphate significant economic benefits can be expected to reuse the small
(Mg: P) molar ratio on struvite crystallization in swine wastewater amount of MFGD residue in the same ways applied in some
using a newly designed unit process. Zhang et al. [19] used the developed countries, like USA and EU countries. Generally, it is
“Ultrasound/Fenton oxidation — MAP chemical precipitation” landfilled or dumped as a solid waste in China. Since the main
method to remove COD and ammonia nitrogen from the old compositions of the MFGD residue are MgSO3 and MgSO4, there
landfill leachate. After the oxidation treatment, Na2HPO412H2O should be a strategy to address this issue by reusing it in China.
and MgSO47H2O were used as precipitation reagents to remove Then a new idea about applying the MFGD residue as a novel Mg
ammonia nitrogen, and the concentration degraded from 910 to source to remove ammonium from the DESW by MAP precipitation
11 mg L1. was suggested by co-authors of this work. It was assumed to
Generally, MAP is a white crystalline compound reduce the ammonium removal cost, and solve the secondary
(MgNH4PO46H2O), also known as struvite, consisting of Mg, pollution caused by the MFGD residue.
ammonium (NH4), and P in equal-molar concentrations. Struvite The objectives of this research are (a) to evaluate the possibility
forms in an alkaline condition, following the reaction shown below of ammonium removal from the DESW by MAP precipitation with
[21,22]. the MFGD residue application as a Mg source; (b) to investigate the
ammonium removal efficiencies with different processing param-
Mg2+ + NH4+ + H2PO4 + 6H2O ! MgNH4PO46H2O # + 2H+
eters, like pH regulators, hydraulic retention time (HRT), P:Mg:N
Besides, struvite precipitation depends on two main factors: the molar ratios, so as to establish an optimal working condition to
molar ratio of Mg:N:P and the pH value of the wastewater [23]. treat DESW for the downstream biological process without
Since chemical precipitation has become a key method to produce ammonia inhibition; (c) to analyze the ammonium removal cost
novel needed materials [24–27], it is ideal to remove ammonium along with the removal efficiency by applying the MFGD residue,
nitrogen from the MESW by MAP precipitation because simulta- compared to using the traditional Mg source (MgCl26H2O).
neously it recovered P from the DESW to form struvite which is
famous as a high-quality fertilizer characterized by slow-release of Materials and methods
nutrients and low contents of heavy metals, compared to P rocks
which helps make it environmentally friendly [28,29]. Moreover, Experimental materials preparation
an efficient removal of organic carbon could also be successfully
achieved by MAP precipitation in treating ammonium nitrogen in The DESW was collected from a swine farm in Boluo
semiconductor wastewater [12]. (Guangdong, China). The fresh raw effluent was contained in a
As regards to MAP formation, magnesium chloride (MgCl2) is 10-L air-tight bucket and settled down for 24 h to remove the
generally used as a Mg source because it is easy to be dissociative sediment. The supernatant was filtered to another 10-L open
[30,31], but the treatment cost is high since the market price of bucket by using a 2 mm screen, and then aerated at the flow rate of
magnesium chloride has reached 0.12 $/kg [32,33]. In order to be 0.4 L/Lsupernatant min for 2 h to remove CO2 by using an aerator (AE2,
cost-efficient, researchers have been looking for other Mg sources JiYin, China). It is helpful for the following MAP precipitation which
as substitutes, like magnesite mineral [34], bittern [35], seawater occurs in an alkaline environment [21]. The basic properties of the
[36], seawater NF concentrate [37,38], MgO [39] and etc. However, supernatant after aeration was determined, showing in Table 1.
some problems occurred when these substitute Mg sources were The MFGD residue was provided by a power plant in Dongguan,
applied to form MAP, such as more solid residue, lower Mg2+ Guangdong, China. The collected residue was ground using a
concentrations, new ion pollution, and location restriction.
On the other hand, magnesium-hydroxide flue gas desulfuriza-
tion (MFGD) is a magnesia scrubbing method to remove SO2 from Table 1
the flue gas. During the process of SO2 absorption into magnesia Properties of the DESW supernatant and the liquor made from MFGD residue.
slurry, the following chemical reactions would take place: Unit Concentration
MgO + H2O = Mg(OH)2 DESW supernatant
pH — 8.32 0.05
PO43–P mg L1 18.73 0.52
NH4+–N mg L1 440.39 4.34
Mg(OH)2 + SO2 + 5H2O = MgSO3 6H2O COD g L1 1.54 0.02
Mg2+ mg L1 35.16 1.79
Also, it is easy to oxidize MgSO36H2O into MgSO47H2O in the
air. Thus, the solid residue produced in the MFGD process mainly Liquor made from MFGD residue
contains MgSO3 and MgSO4 [40]. So far, the MFGD process has pH — 8.74 0.03
Mg2+ g L1 31.42 0.07
become one of the most widely used techniques to eliminate SO2
150 H. Lin et al. / Journal of Industrial and Engineering Chemistry 58 (2018) 148–154
0.15 mm screen, and then kept in zip lock bags. The main Statistical analysis
characteristics of the MFGD residue were determined, in which
the contents of MgSO3 and MgSO4 were 35.7 2.58 and Statistical significance was examined by analysis of variance
8.54 0.46 mg/L, respectively. Meanwhile, the ground residue (ANOVA) using SPSS (Version 16.0 for Windows). The probability
was throughly mixed with deionization (DI) water at the ratio of 3: level of p < 0.05 was considered significant.
10(w/v)using a blender (JB40-C, Shanghai, China) at 100 rpm for
24 h, along with aeration at 6.67 L/Lsupernatant min using the same Results and discussion
aerator to remove CO2. The well-mixed liquor was used as the
magnesium source for MAP precipitation, and the basic character- Effects of pH regulators
istics of the liquor were measured (See Table 1).
Sodium dihydrogen phosphate dihydrate (NaH2PO42H2O, pH is the main driving force behind struvite precipitation and
analytical grade, >98% pure) was purchased from Tianjin Chemical variation of pH effects the solubility or formation of struvite
Sci-Tech Co., Ltd (Tianjin, China) and applied as a phosphate source crystals. To authors’ knowledge, the optimal pH value for MAP
in MAP precipitation. Magnesium chloride (MgCl26H2O, analytical precipitation was not constant. Different optimal pH values were
grade, >98% pure) was purchased from Guangzhou Chemical Sci- observed in different wastewater treatment systems. Li et al. [44]
Tech Co., Ltd (Guangzhou, China) and applied as a traditional reported that the optimal pH value was 9 when treating raw
commercial Mg source. Sodium hydroxide (NaOH, analytical grade, landfill leachate via struvite precipitation. Zhang et al. [45] found
>98% pure) was also bought from the same company in Guangzhou the highest NH4+–N removal efficiency was noted at pH 10 when
to adjust pH values in the experiments. treating swine wastewater by MAP precipitation. Gunay et al. [34]
reported that all struvite crystals were dissolved in acidic pH
MAP precipitation values. In contrast, decrease in solubility of struvite was observed
as a result of pH increase. The minimum solubility of struvite was
Three factors affecting MAP formation were evaluated, includ- observed in the range of pH 8.8–9.4 and solubility began to
ing pH, HRT which was divided into stirring time and settling time, increase above pH of 9.5.
and molar ratios of PO43:Mg2+:NH4+. The ammonium removal In this study, the ammonium removed from the DESW was
efficiencies and the concentrations of NH4+–N, PO43–P, Mg2+ in the considered to be throughly converted into MAP. Thus the
solution were used as the main indicators to determine the optimal ammonium removal efficiency was highly relevant to the MAP
MAP forming conditions. precipitation reaction. As shown in Fig. 1, a significant distinction
To investigate the effects of pH regulators on MAP precipitation, (p < 0.05) of NH4+–N removal efficiencies was noted when pH
the experiments were performed at a P:Mg:N molar ratio of 1:1:1, regulators varied in the range of 8.5–9.5 during MAP precipitation
varying pH values from 8.5 to 10.5. Subsequently, to evaluate the at the P:Mg:N molar ratio of 1:1:1. The NH4+–N removal efficiency
effects of HRT on ammounium removal efficiencies, HRT from increased with an enhanced pH value till the highest removal
10 min to 35 min was designed under the optimal pH value and a P: efficiency of 89.06% was determined at pH 9.5. Then it went down
Mg:N molar ratio of 1:1:1. Finally, the effect of P:Mg:N molar ratios with an increased pH value until 64.52% of ammounium removal
on MAP precipitation were investigated at the optimal pH value efficiency was observed at pH 10.5. This result was similar to that
and HRT. MAP precipitation efficiencies increased from 40% to 80% when pH
Regarding the experimental procedure, briefly, the MAP values increased from 8.2 to 9.5 [46]. It was caused by the fact that
precipitation reaction was proceeded in a 500-mL beaker. The ion varieties and contents changed in the solution under different
liquor (MFGD residue completely mixed with DI water) and pH values [45,47]. As is known, more H+ is noted in a solution with
NaH2PO42H2O powder were added to the beaker which has a lower pH value, so it is easier to form Mg(H2PO4)2 instead of MAP
contained 300 mL DESW supernatant to obtain the P:Mg:N molar in an acidic solution. On the other hand, more OH is available in a
ratio at a desired value in the solution. Then, the H2SO4 solution solution with a higher pH value which will combine Mg2+ to form
(3 mol/L) and the NaOH solution (6 mol/L) were used to adjust the Mg(OH)2 rather than MAP. Besides, based on the solubility, it is
pH value. The mixed solution was stirred using a magnetic stirrer easier to occur Mg3(PO4)2 when the pH value was higher than 9.5.
(MYP11-2A, Chijiu, China) for 5 min at the desired pH value. Then, Thus, Mg(OH)2 and Mg3(PO4)2 could appeare under different pH
the mixed solution was kept still for 5 min, and 20 mL supernatant values during MAP precipitation in this work. Moreover, a part of
was sampled and filtered through a 0.45 mm membrane filter for
NH4+–N, PO43–P and Mg2+ analysis.
Analytical methods
The characteristics of the DESW supernatant and the liquor
from the MFGD residue were measured according to the Standard
Methods [43]. The ammonium concentration was determined by
the Nessler’s reagent colorimetric method using a spectropho-
tometer (7200, UNIC, China). The PO43 concentration was
measured by the molybdate–ascorbic acid colorimetric method
using the same spectrophotometer as that used for the ammonium
measurement. The concentration of Mg2+ was determined using an
atomic absorption spectrophotometer (ZA3300, HITACHI, Japan).
The pH value of the solution was measured using a pH meter (pHs-
3D, China). The collected precipitate was dried in a vacuum drying
oven at 40 C for 48 h, and then analyzed by X-ray diffraction (XRD,
D8 ADVANCE, Bruker, Germany) and scanning electron microscopy
with energy dispersive X-ray analysis (SEM-EDX, merlin, zeiss,
Germany). Fig. 1. NH4+–N removal efficiencies under various pH regulators.
H. Lin et al. / Journal of Industrial and Engineering Chemistry 58 (2018) 148–154 151
Fig. 2. NH4+–N removal efficiencies via MAP precipitation (a) and PO43–P concentrations in the solution after MAP precipitation (b) under different hydraulic retention time.
NH4+-N could also be converted into NH3-N under a high pH value standard of pollutants for livestock and poultry breeding in China
which would further reduce the yield of MAP precipitate [48]. (GB 18596-2001). It might cause eutrophication to the surface
water system, and the cost to remove ammonium in the DESW was
Effects of hydraulic retention time increased too, since the excess commercial NaH2PO4 was applied.
For the sake of environment and ecomomy, finding an optimal P:
HRT, the sum of stirring time and settling time, is a parameter Mg:N molar ratio to keep a lower PO43–P concentration in the
related to the processing efficiency which is important to a DESW after MAP precipitation was vital.
wastewater treatment plant. Therefore, effects of HRT varying from
10 min to 35 min at the P:Mg:N molar ratio of 1:1:1 with pH 9.5 on Effects of P:Mg:N molar ratios
MAP precipitation to remove ammonium from the DESW was
investigated. The highest NH4+-N removal efficiency was 89.44%, Different loads of PO43
and no significant difference (p < 0.05) was noted under various PO43, Mg2+, and NH4+ are three basic ions to form MAP
HRT (Fig. 2a). It meant around 90% of MAP could be yielded in the precipitate and affect the struvite crystallization. An appropriate P:
first 10 min. Prolonging HRT (over 10 min) was not contributed to Mg:N molar ratio is important for pollutants removal and resource
the MAP formation. It is because MAP precipitation belongs to a recovery. In this work, the P:Mg:N molar ratios from 0.4:1:1 to
fast chemical reaction, which is mainly affected by stirring 0.8:1:1 were investigated. It was found that the NH4+–N removal
methods and speeds, reactor sizes and wastewater properties. efficiency improved from around 40%–80% when the NaH2PO4
Based on the above results, it could be concluded that 10 min was adding ratio increased from 0.4 to 0.8 (Fig. 3a). Because excess Mg2
+
the optimal HRT to remove ammonium from the DESW through and NH4+ were provided in this MAP precipitation reaction, and
MAP precipitation by using the MFGD residue as the Mg source. PO43 became the rate-limiting ion, then the NH4+–N removal
Besides, the concentration of PO43–P in the solution after MAP efficiency was enhanced with the increased PO43 load. This result
precipitation was also determined. No significant variation was consistent with that reported by Zhang et al. [47].
(p < 0.05) of PO43–P concentrations was noted with different Consequently, the NH4+–N concentrations in the solution during
HRT (Fig. 2b). This result was in line with that expressed in Fig. 2a. MAP precipitation decreased from 268.14 to 120.39 mg L1. And the
However, the PO43 concentration in the DESW after MAP leftover NH4+–N concentration in the DESW after MAP precipita-
precipitation was around 35–37 mg L1, which was about 3.5-fold tion was lower than 200 mg L1. It meant that the DESW after MAP
higher than the concentration (8 mg L1) required in the discharge precipitation was alternative to the downstream biological
Fig. 3. NH4+–N removal efficiencies via MAP precipitation (a) and the concentrations of PO43, Mg2+ and NH4+ in the solution after MAP precipitation (b) at different P:Mg:N
molar ratios by changing PO43 loads.
152 H. Lin et al. / Journal of Industrial and Engineering Chemistry 58 (2018) 148–154
Fig. 4. NH4+–N removal efficiencies via MAP precipitation (a) and the concentrations of PO43, Mg2+ and NH4+ in the solution after MAP precipitation (b) at different P:Mg:N
molar ratios by changing Mg2+ loads.
treatment without ammonia inhibition, which was also confirmed Mg2+ concentration rose significantly (p > 0.05). It implied that
by some other researchers [12,49]. Mg2+ was excess in the MAP precipitation reaction. In order to
On the other hand, the concentrations of PO43–P and Mg2+ in reduce the Mg2+ concentration in the DESW after MAP precipita-
the solution during MAP precipitation were also measured tion and save the cost, the optimal P:Mg:N molar ratio loading to
(Fig. 3b). The concentrations of PO43–P were in the average of the MAP precipitation reactor to remove ammonium in the DESW
3.00 mg L1. No significant difference (p < 0.05) of PO43–P was considered to be 0.6:0.7:1.
concentrations was observed when the P:Mg:N molar ratios were
less than or equal to 0.8:1:1. It implied that the concentrations of Characteristics of MAP precipitate
NH4+–N and Mg2+ in the reactor were abundant to form MAP, and
the PO43–P concentration became the rate-limiting factor in the The main components of the precipitate could be concluded by
MAP precipitation reaction at the P:Mg:N molar ratios from 0.4:1:1 X-ray diffraction (XRD) analysis, via comparing the patterns of
to 0.8:1:1. Taking the NH4+–N removal efficiencies, the leftover samples to those of the standard substance [50,51]. As is shown in
concentrations of NH4+–N and PO43–P in the DESW after MAP Fig. 5, the characteristic peaks, which refer to the main
precipitation, and the cost effectiveness into consideration, the components, of the precipitate were similar to those of the
optimal loading ratio of PO43 should be 0.6. standard MAP (MgNH4PO46H2O) (JCPDS 15-0762) [52]. It indicat-
The Mg2+ concentrations in the solution decreased with the ed that the precipitate formed in the DESW precipitation process
increased PO43 molar ratios due to the formation of MAP was about MAP. Otherwise, the characteristic peak intensity of the
precipitate. When the optimal PO43 molar ratio (0.6) was loaded, precipitate was not exactly the same as the standard MAP. It was
252.04 mg L1 of Mg2+ in the solution was determined. And the caused by some flaws in the crystals which originated from the
lowest Mg2+ concentration of 118.50 mg L1 was noted when the P: impurities in the DESW.
Mg:N molar ratio was equal to 08:1:1. However, these Mg2+ To further identify the precipitate, SEM-EDX analysis was
concentrations were too high for the subsequent biological process performed to determine the surface characteristics of the
for further pollutants removal in the DESW, in which salt inhibition precipitate (Fig. 6a and b). It was found that the precipitate
might occur [47]. Additionally, it was also a waste to drain surface was full of complete and incomplete X-shaped white
wastewater containing a high concentration of Mg2+. Hence, crystals. The incomplete X-shaped crystals were coarse, and the
finding an appropriate Mg2+ load was necessary in this DESW size was irregular (15–40 mm length) by SEM analysis. According to
treating process. the XRD results, MAP precipitate belongs to the complete X-shaped
crystal. These incomplete crystals further proved the exist of the
Different loads of Mg2+ impurities in the DESW during precipitation. The complete X-
Based on the optimal PO43–P molar ratio (0.6), the Mg2+ molar
ratio from 0.7–0.9 was studied. The results indicated that no
significant difference (p < 0.05) of NH4+–N removal was noted at
different Mg2+ molar rations. Around 60% of NH4+–N removal
efficiency was achieved at the P:Mg:N molar ratios from 0.6:0.7:1
to 0.6:0.9:1. Also, no markedly difference (p < 0.05) of NH4+–N
concentrations in the solution during MAP precipitation was
observed. But the NH4+–N concentrations in the solution were all
lower than 200 mg L1 (Fig. 4a). The reason lied in the fact that the
PO43–P concentration was the rate-limiting factor in the MAP
precipitation reaction, and Mg2+ was always excess in the reaction
even at its lowest load ratio of 0.7.
This explanation was further proved by the data shown in
Fig. 4b. The leftover concentration of PO43–P in the solution
during MAP precipitation was almost keeping the same, while the
Fig. 5. X-ray diffraction analysis of the precipitate.
H. Lin et al. / Journal of Industrial and Engineering Chemistry 58 (2018) 148–154 153
L1. However, the leftover NH4+–N concentration with the MFGD
residue application was 7.43% lower than that with the commercial
MgCl26H2O. The leftover PO43–P concentration with the MFGD
residue application was also 1.92% lower than that with the
MgCl26H2O. Contrarily, the leftover Mg2+ concentration with the
MFGD residue application was 17.82% higher than that with the
MgCl26H2O. It indicated that the MAP precipitation reaction
responded better with the MFGD residue application even though
the initial Mg2+ load was the same with the commercial
magnesium source (MgCl26H2O). It might be attributed to some
cooperatives caused by other compounds in the MFGD. Further
study is needed in future.
According to the above results, the economic evaluation of
ammonium removal from 1 m3 DESW was performed. Regarding
the economic assessment, the manpower costs and the commer-
cial value of MAP precipitate were not considered; only the costs of
the chemicals used in the process were taken into account. The
MFGD residue used in the process was solid waste and free of
charge. The market prices of the chemicals and the total chemical
costs were listed in Table 2. The total chemical cost to remove
ammonium in the 1 m3 DESW by using MgCl26H2O as the
magnesium source was 3.60 USD, whereas the cost of using the
Fig. 6. Scanning electron microscopy analysis (a) and energy dispersive X-ray
MFGD residue as the magnesium source was 2.55 USD. The
analysis (b) of the precipitate.
chemical cost was saved by 29.17% using the MFGD residue as the
magnesium source, which could cover a part of expenses of other
shaped white crystals were sampled and followed with EDX chemicals, like NaH2PO4 and NaOH. Moreover, extra income would
analysis. The results showed that the surface of the crystal be achieved if the MAP precipitate was sold as a fertilizer in market.
contained a great deal of C (22.26 wt%), O (44.72 wt%), P (15.10 wt And reusing the MFGD residue in the MAP precipitation process
%), Mg (13.45 wt%), a moderate content of N (2.76 wt%) and K would reduce the solid waste treating fee as well.
(1.55 wt%), and a little of Fe (0.16 wt%). Since O, Mg, P and N are the However, the saving cost was changed when the different
main components of MAP, it could be infered that the main magnesium source was applied. Huang et al. [53] reported that the
composition of the precipitate was MAP (MgNH4PO46H2O). This cost to form struvite by using MgO-saponification wastewater as
result was consistent with the one obtained through XRD analysis the magnesium source was reduced by 12.4%, compared to using
(Fig. 5). Furthermore, it was noticed that the precipitate was rich of the pure magnesium chemical. Gunay et al. [34] reported that
C, which meant a part of COD in the DESW was also removed in the 90.9% of ammonium in the leachate was removed using magnesite
MAP precipitation process. It was helpful to reduce the COD instead of MgCl2, with a resultant drop of 18% in the operation cost.
content in the DESW, and the similar result was reported in While Huang et al. [54] found that approximately 90% ammonium
another literature [12]. Additionally, the high contents of N, Mg and could be removed using the pyrolyzate of magnesite as the
P were observed in the precipitate, which suggested that being magnesium source, saving 34% of the operation cost compared
reused as a fertilizer was an alternative to the precipitate. with that by using the pure magnesium chemical. According to the
previous research, variations of the saving cost were also
Economical evaluation connected to the types of wastewater.
Overall, the MFGD residue reused as a magnesium source to
The MAP precipitation experiments were performed to remove remove ammonium in the DESW was feasible with obvious
ammonium in the DESW by using the MFGD residue and the economic benefits.
commercial magnesium chloride (MgCl26H2O), respectively, at
the P:Mg:N molar ratio of 0.6:0.7:1 and pH 9.5. The NH4+–N Conclusions
removal efficiencies and the concentrations of NH4+–N, PO43–P
and Mg2+ in the DESW after MAP precipitation were all analyzed. Using the solid residue from the MFGD process as a magnesium
As is shown in Table 2, The NH4+–N removal efficiencies were both source to remove ammonium in the DESW via MAP precipitation to
higher than 55% by adding the MFGD residue and the commercial avoid ammonia inhibition in the downstream biological process
magnesium chloride. And the leftover NH4+–N concentrations in was investigated in this study. The ammonium removal efficien-
the DESW after MAP precipitation were both lower than 200 mg cies, pH regulators and HRT, as well as the concentrations of NH4+,
Table 2
The ammonium removal performaces and total chemical costs in the DESW treatment processes via MAP precipitation by using different Mg sources.
NH4+–N removal efficiency (%) Concentrations in the DESW after MAP precipitation(mg/L) Market price ($/kg) Total chemical cost($/m3)
+ 3 2+
NH4 –N PO4 –P Mg
MFGD residue 58.45 2.63 183.20 11.59 3.64 0.11 67.17 3.91 0.00 2.55a
MgCl26H2O 55.36 0.28 196.82 1.23 3.71 0.09 55.20 0.64 0.12 3.60b
a
Total chemical cost = CostMFGDresidue + CostNaH2 PO4 2H2 O + CostNaOH = 0 + 0.56 ($/kg, the market price for NaH2PO42H2O) 2.86 (kg/m3, the NaH2PO42H2O dosage used to
treat 1 m3 DESW) + 0.35 ($/kg, the market price for NaOH) 2.71 (kg/m3, the NaOH dosage used to treat 1 m3 DESW) = 2.55 ($/m3).
b
Total chemical cost = CostMgCl2 6H2 O + CostNaH2 PO4 2H2 O + CostNaOH = 0.12 ($/kg, the market price for MgCl26H2O) 8.36 (kg/m3, the MgCl26H2O dosage used to treat 1 m3
DESW) + 0.56 ($/kg, the market price for NaH2PO42H2O) 2.86 (kg/m3, the NaH2PO42H2O dosage used to treat 1 m3 DESW) + 0.35 ($/kg, the market price for NaOH) 2.83
(kg/m3, the NaOH dosage used to treat 1 m3 DESW) = 3.60 ($/m3).
154 H. Lin et al. / Journal of Industrial and Engineering Chemistry 58 (2018) 148–154
PO43, and Mg2+ in the solution during MAP precipitation were [21] N.C. Bouropoulos, P.G. Koutsoukos, Cryst. Growth Des. 213 (2000) 381.
considered, in order to establish an optimal process to treat DESW. [22] G.E. Diwani, S.E. Rafie, N.N.E. Ibiari, H.I. El-Aila, Desalination 214 (2007) 200.
[23] E.V. Munch, K. Barr, Water Res. 35 (2001) 151.
Analysis of the precipitate was also performed. It was concluded [24] L. Nejati-Moghadam, A. Esmaeili Bafghi-Karimabad, M. Salavati-Niasari, H.
that the MFGD residue played an important role in the MAP Safardoust, JNS 5 (2015) 47.
precipitation reaction, which worked in the way of cost saving [25] G. Nabiyouni, S. Sharifi, D. Ghanbari, M. Salavati-Niasari, JNS 4 (2014) 317.
[26] T. Mahmoudi, M. Salavati-Niasari, Cluster Sci. 23 (2012) 1003.
compared to using a commercial Mg source and environment [27] M. Panahi-Kalamuei, M. Salavati-Niasari, J. Mater. Sci.: Mater. Electron. 26
protection by reusing the solid waste. (2015) 2851.
[28] K. Suzuki, Y. Tanaka, T. Osada, M. Waki, Water Res. 36 (2002) 2991.
[29] J.E. Lee, M.M. Rahman, C.S. Ra, J. Hazard. Mater. 169 (2009) 801.
Acknowledgements [30] K. Suzuki, Y. Tanaka, K. Kuroda, D. Hanajima, Y. Fukumoto, Bioresour. Technol.
96 (2005) 1544.
The authors would like to thank the National Key Research and [31] Z.G. Liu, Q.L. Zhao, D.J. Lee, N. Yang, Bioresour. Technol. 99 (2008) 6488.
[32] L. Pastor, D. Mangin, R. Barat, A. Secoc, Bioresour Technol. 99 (2008) 6285.
Development Program of China (No: 2016YFD0800607) and the
[33] M.M. Rahman, Y.H. Liu, J.H. Kwag, C.S. Ra, J. Hazard. Mater. 186 (2011) 2026.
Foundation of Academy of Agricultural science (No: YNK2015-227) [34] A. Gunay, D. Karadag, I. Tosun, M. Ozturk, J. Hazard. Mater. 156 (2008) 619.
for financially supporting this research. [35] Z.L. Ye, S.H. Chen, M. Lu, J.W. Shi, L.F. Lin, S.M. Wang, Water Sci. Technol. 64
(2011) 334.
[36] D. Crutchik, J.M. Garrido, Water Sci. Technol. 64 (2011) 2460.
References [37] M. Telzhensky, L. Birnhack, O. Lehmann, E. Windler, O. Lahav, Chem. Eng. J. 175
(2011) 136.
[1] C. Lu, Pigs Today 6 (2013) 26. [38] O. Lahav, M. Telzhensky, A. Zewuhn, Y. Gendel, J. Gerth, W. Calmano, L.
[2] C.N. Mulligan, An evaluation of the sustainability of the anaerobic digestion of Birnhack, Sep. Purif. Technol. 108 (2013) 103.
manure [G], Proceedings 10th World Congress A D, Montreal, Canada, 2004, [39] H. Huang, Y. Chen, Y. Jiang, L. Ding, Chem. Eng. J. 254 (2014) 418.
pp. 11796. [40] R.T. Guo, W.G. Pan, X.B. Zhang, H.J. Xu, J.X. Ren, Fuel 90 (2011) 7.
[3] Y. Lin, D. Wang, Q. Li, M. Xiao, Bioresour. Technol. 102 (2011) 3673. [41] H. Kikkawa, A. Nakamoto, M. Morishita, J. Ind. Eng. Chem. Res. 41 (2002) 3028.
[4] K.R. Lopata, L. Auerswald, P. Cook, Aquaculture 261 (2006) 678. [42] J.H. Beeghly, Product development of FGD recovered magnesium hydroxide, A.
[5] P.K.S. Shin, N.W.Y. Lam, R.S.S. Wu, P.Y. Qian, S.G. Cheung, Mar Pollut. Bull. 56 16th Proceedings Annual International Pittsburgh Coal Conference (1999) 462.
(2008) 282. [43] APHA, Standard Methods for the Examination of Water and Wastewater, 21st
[6] R.H. Zhang, P. Yang, Z. Pan, T.D. Wolf, J.H. Turnbull, Trans. ASAE 47 (2004) 243. ed., American Public Health Association, Washington, D.C, 2005.
[7] T. Hao, P. Xiang, H.R. Mackey, Water Res. 65 (2014) 1. [44] X. Li, B. Dolores, J. Chen, Environ. Chem. Lett. 9 (2011) 71.
[8] G.J. Liu, D. Zheng, L.W. Deng, Q. Wen, Y. Liu, Environ. Technol. 35 (2014) 2660. [45] D. Zhang, Y. Chen, G. Jilani, W. Wu, W. Liu, Z. Han, Bioresour. Technol. 116 (2012)
[9] C. Plaza De Los Reyes, G. Pozo, G. Vidal, J. Environ. Sci. Health A: Tox. Hazard. 386.
Subst. Environ. Eng. 49 (2014) 218. [46] N.A. Booker, A.J. Priestley, I.H. Fraser, Environ. Technol. 20 (1999) 777.
[10] J. Chung, W. Bae, Y. Lee, B. Rittmann, J. Process Biochem. 42 (2007) 320. [47] T. Zhang, L. Ding, H. Ren, J. Hazard. Mater. 166 (2009) 911.
[11] H.S. Kim, Y. Choung, S. Ahn, H.S. Oh, Desalination 223 (2008) 194. [48] Y. Song, P. Yuan, B. Zheng, J. Peng, F. Yuan, Y. Gao, Chemosphere 69 (2007) 319.
[12] H. Ryu, S. Lee, J. Process Biochem. 45 (2010) 563. [49] S. Park, W. Bae, Process Biochem. 44 (2009) 631.
[13] M. Vanotti, A. Szogi, J. Environ Qual. 37 (2008) S86. [50] M. Ghiyasiyan-Arani, M. Masjedi-Arani, D. Ghanbari, S. Bagheri, M. Salavati-
[14] L. Zhang, Y.W. Lee, D. Jahng, J. Hazard. Mater. 199–200 (2012) 36. Niasari, Sci. Rep. 6 (2016) 25231.
[15] Y.L. Ye, P.E. Saikaly, B.E. Logan, Bioresour. Technol. 244 (2017) 456. [51] M. Salavati-Niasari, M. Hosseinpour-Mashkani, F. Mohandes, S. Gholamrezaei,
[16] Y.H. Liu, J.H. Kwag, J.H. Kim, C.S. Ra, Desalination 277 (2011) 364. J. Mater. Sci.: Mater. Electron. 26 (2015) 2810.
[17] S. Sugiyama, M. Yokoyama, H. Ishizuka, K. Sotowa, T. Tomida, N. Shigemoto, J. [52] International Centre for Diffraction Data, Ammonium magnesium phosphate
Colloid Interface Sci. 292 (2005) 133. hydrate (Standard #15-0762), A computer database. 1996.
[18] M.H. Bhuiyan, D.S. Mavinic, R.D. Beckie, J. Cryst. Growth 310 (2008) 1187. [53] H. Huang, D. Xiao, Q. Zhang, L. Ding, J. Environ. Manage. 145 (2014) 191.
[19] J. Zhang, T. Yang, H. Wang, K. Yang, F. Cheng, L. Bin, X. Yang, Chem. Speciation [54] H.M. Huang, X.M. Xiao, L. Yang, B. Yan, Water Pract. Technol. 5 (2010) 7.
Bioavailability 27 (2015) 175.
[20] H.M. Huang, J.H. Liu, S.F. Wang, Y. Jiang, D. Xiao, L. Ding, Ecol. Eng. 92 (2016) 111.
