12
1. Introduction
Recently livestock wastewater treatment and recycling in South
Korea have become a big issue as animal wastewater generation
has reached to about 46,000 Mm
3
/y. Also, the wastewater from
the pig farming industry poses serious social and economic prob-
lems due to the negative effect on the environment with respect
to treatment and recycling of swine wastewater [1-2]. The treatment
and recovery of nutrient from the waste stream is important, since
swine wastewater contains high concentrations of nitrogen and
phosphorus that cause eutrophication in water bodies. Recently,
the depletion of phosphorus resources has been a big issue, which
has been discussed on a global basis [3], although it is estimated
that 7,000 billion kg of phosphate rock still exists [2].
Phosphorus as well as nitrogen in swine wastewater after anaero-
bic digestion has been widely used as a liquid fertilizer in Korea.
However, the quick-release fertilizer application is a principal
source of nitrogen and phosphorus pollution in the farming field,
causing water pollution. Alternative methods, however, such as
crystallization and adsorption processes have been developed to
recover nutrients to high quality [4-6]. Among these techniques,
the crystallization process of the magnesium ammonium phosphate
(MAP, also known as struvite) is considered to be one of the better
techniques, as struvite crystallization is cost-effective and yields
high-quality nutrients, used as valuable slow-release fertilizers [7-9].
Struvite (MgNH
4
PO
4
･6H
2
O) is a crystalline substance consisting
of magnesium, ammonium, and phosphate ions in equal molar
concentrations. Struvite crystal is commonly formed as scale after
anaerobic digestion on pipe walls and reactor vessels. The chemical
equation for struvite crystal formation is as follows [10]:
→
･
(1)
Environ. Eng. Res. 2017; 22(1): 12-18
pISSN 1226-1025
https://doi.org/10.4491/eer.2016.037
eISSN 2005-968X
Effects of pH, molar ratios and pre-treatment on phosphorus
recovery through struvite crystallization from effluent of
anaerobically digested swine wastewater
Daegi Kim
1
, Kyung Jin Min
2,3
, Kwanyong Lee
4
, Min Sung Yu
2
, Ki Young Park
2
†
1
Institute for Advanced Engineering, Gyeonggi-do 17180, Republic of Korea
2
Department of Civil and Environmental System Engineering, Konkuk University, Seoul 05029, Republic of Korea
3
AinChemTech Co., Ltd and EPS Solution Co., Ltd, Gyunggi-do 05029, Republic of Korea
4
Korea Institute of Civil Engineering and Building Technology, Gyeonggi-do 10223, Republic of Korea
ABSTRACT
Struvite precipitation has been proven to be an effective method in removing and recovering ammonia nitrogen (N) and phosphate phosphorus
(P) from wastewater. In this study, effects of pH, molar ratios and pre-treatment of effluent of anaerobically digested swine wastewater were
investigated to improve struvite crystallization. The magnesium : ammonium : phosphate ratio of 1.2 : 1.0 : 1.0 was found to be optimal, yet
the molar ratio in the wastewater was 1 : 74.9 : 1.8. From the analysis, the optimum pH was between 8.0 and 9.0 for maximal phosphate
P release and from 8.0 to 10.0 for maximal ammonia N and phosphate P removal from real wastewater. Analysis from Visual MINTEQ predicted
the pH range of 7-11 for ammonia N and phosphate P removal and recovery as struvite. For pre-treatment, microwave pre-treatment was
ineffective for phosphate P release but ultrasound pre-treatment showed up to 77.4% phosphate P release at 1,000 kJ/L of energy dose. Precipitates
analysis showed that phosphorus and magnesium in the collected precipitate had almost same values as theoretical values, but the ammonia
content was less than the theoretical value.
Keywords: Ammonia and phosphate recovery, Magnesium ammonium phosphate, Pre-treatment, Struvite, Ultrasonic treatment
This is an Open Access article distributed under the terms
of the Creative Commons Attribution Non-Commercial License
(http://creativecommons.org/licenses/by-nc/3.0/) which per-
mits unrestricted non-commercial use, distribution, and reproduction in any
medium, provided the original work is properly cited.
Copyright © 2017 Korean Society of Environmental Engineers
Received February 29, 2016 Accepted September 6, 2016
†
Corresponding author
Email: kypark@konkuk.ac.kr
Tel: +82-2-450-3736 Fax: +82-2-450-3726
Environmental Engineering Research 22(1) 12-18
13
Struvite precipitation from wastewater is influenced by a large
number of parameters such as pH of the reaction, molar ratio,
interfering ions in the feed, reaction time, types of chemicals added,
types of the reactor used and temperature. From these, the reaction
pH and molar ratios of reactants, namely magnesium : ammonium
: phosphate molar ratios, are the main factors for struvite
precipitation.
The pH of the reaction plays a significant role during struvite
precipitation process, and not only affects the amount of struvite
precipitation, but also its purity. Increasing the pH and the reactant
concentration can reach solution saturation but increasing the
pH of the solution is more feasible and allows for more varied
applications [11]. Struvite can be precipitated at a wide pH range
(from 7.0 to 11.5), but the suitable pH range is from 8.0 to 9.5.
Struvite precipitation is a physico-chemical process that can occur
over a range of pH values bounded by the speciation of struvite
components so that the concentrations of magnesium, ammonium
and phosphate ions can be affected by the pH of the solution
[12]. A variety of magnesium and phosphate complex ions patterns
in the reactor solution, including MgOH
+
, Mg(OH)
3
-
, MgH
2
PO
4
+
,
MgHPO
4
, H
3
PO
4
, H
2
PO
4
-
, HPO
4
-2
, MgPO
4
-
can be formed when
the pH of solution is varied [13].
The pH of the solution in the struvite precipitation reactor
influences struvite solubility. With increasing the pH, the struvite
solubility decreases, but the solubility begins to increase when
the pH rises above pH 9; this is because the ammonium ion concen-
tration decreases and the phosphate ion concentration increases
[12, 14]. As various factors such as reaction pH, ionic strength
and temperature affect struvite solubility, which in turn determines
the supersaturation ratio [15]. It is the excess supersaturation in
the liquid that is the major parameter in predicting struvite precip-
itation potential [16]; therefore, it is important to use chemical
equilibrium-based models to calculate and predict the practical
conditions for struvite formation. There is a geochemical equili-
brium speciation model MINTEQ that could be used to model
struvite formation [17]. For calculating metal speciation, solubility
equilibria, sorption, etc., for natural waters, visual MINTEQ is
available as a freeware chemical equilibrium model.
For the anaerobically digested effluent of swine wastewater
from livestock, there is generally less magnesium and phosphate
ions compared with ammonium ion. It is then necessary to add
a source of magnesium and phosphate ions to enhance the struvite
crystallization process. The concentration of phosphate ion is ex-
pected to increase through solubilization of total phosphorus. The
first step, thus, is to enable phosphate P release from solid phases
to increase the recovery of phosphorus by the struvite crystal-
lization process. The general methods to facilitate phosphate P
release are physical and chemical techniques [18-20]. Among these
techniques, alkaline-ultrasonic pre-treatment is preferred as it also
disintegrates the solid from swine wastewater and enhances the
anaerobic digestion process [21-22].
The objectives of this study were to investigate the effect of
pH and molar ratios for magnesium, ammonium, and phosphate
ions on ammonia N and phosphate P removal and recovery. Also
alkaline-ultrasonic pre-treatment was applied to the struvite crys-
tallization process to enhance nutrient recovery.
2. Materials and Methods
2.1. Materials
The anaerobically digested effluent of swine wastewater used
in the study was from the P-city swine wastewater treatment
plant in Korea. The effluent of swine wastewater was concentrated
at 4°C for 24 h and its main characteristics are shown in Table
1. Based on the initial composition of the effluent solution, magne-
sium and phosphate ions concentrations were very low, and
they needed to be increased to reach the desired molar ratios
for magnesium, ammonium, and phosphate ions. For both syn-
thetic and real wastewater, the concentrations of magnesium,
ammonium and phosphate ions were adjusted to the required
molar concentration using MgCl
2
･6H
2
O, NH
4
Cl and KH
2
PO
4
sol-
utions, respectively. All reagents were of analytical grade. To
investigate the effect of reaction pH, 2 N HCl and 2 N NaOH
were used to adjust the pH, and the pH was monitored with
a pH meter. In addition, the pH influenced the phosphorus frac-
tions [22-23].
Table 1. Characteristics of the Effluent from the Anaerobically Digested
Swine Wastewater
Concentration
pH 8.17
T-N (mg/L) 2,350
NH
3
-N (mg/L) 1,775
T-P (mg/L) 612
PO
4
-P (mg/L) 221
2.2. Crystallizations Experiments
A lab-scale airlift reactor with a working volume of 5 L was used
for struvite crystallization. The schematic diagram of the ex-
perimental apparatus for struvite crystallization is shown in Fig.
1. The reactor operated with a 10 min hydraulic retention time
for the mixing zone and 3 h for the whole reactor. The obtained
struvite cake from the process was dried at room temperature
to form a powder.
Fig. 1. Experimental equipment for struvite crystallization.
D. Kim et al.
14
2.3. Pre-treatment for Phosphate P Release
Ultrasonic pre-treatment was performed for phosphate P release
with STH-750S ultrasound (Sonitopia, Korea) with operating fre-
quency of 20 kH and maximal power of 750 watt. Microwave
pre-treatment device had a microwave frequency of 2,450 MHz
with maximal power of 600 watt. During pre-treatment, the sup-
plied energy density ranged from 100 to 20,000 kJ/L. The energy
density of pre-treatment device can be defined with following
Eq. (2) (the energy density conditions of pre-treatment are shown
in Table 2):
･
×
(2)
where Power is in watt, t (time) in s, and V (sample volume)
in L.
Table 2. Energy Density Used in Pre-treatment
Energy density (kJ/L) Electricity (watt) Contact time (s)
100 33 300
200 67 300
500 167 300
1,000 333 300
2,000 667 300
5,000 750 667
10,000 750 1,334
20,000 750 2,667
2.4. Analysis and MINTEQ Model
The concentrations for total nitrogen (T-N), ammonia N, pre-treat-
ment for organic phosphate measurement were determined by
the following standard methods [24]. To study the release of phos-
phate ions concentration at different ultrasonic doses with ultra-
sonic disintegration, total phosphorus (T-P) and ortho-phosphate
P (PO
4
3-
-P) levels in the effluent of the swine wastewater were
established by the Persulfate Digestion Method in HACH methods
10072. The phosphorous concentrations were determined by the
ascorbic acid method, using a UV-V is spectrophotometer at 800
nm (Smart Plus SP-1900PC, Woongki Science, Seoul, Korea). The
pH meter (Orionstar, Thermo Scientific, Waltham, MA, USA) was
calibrated after each experiment. The potential for struvite for-
mation as a function of pH was predicted by using chemical equili-
brium freeware Visual MINTEQ 3.0 developed by the U.S.
Environmental Protection Agency. Using the composition of the
anaerobically digested effluent of swine wastewater from P-city
as input, the model’s.
3. Results and Discussion
3.1. Effect of MAP Molar Ratios on Struvite Formation
In this research, the effects of Mg
2+
: NH
4
+
: PO
4
3-
molar ratios
on struvite crystallization using synthetic swine anaerobic digester
wastewater were analyzed based on indications from previous
work [25-27]. At the start of the experiment, using a 0.1 NaOH
solution, the initial pH of the digester effluent sample was adjusted
to 9.0. Table 2 shows the effect of molar ratios on ammonia and
phosphate ions removal. Molar ratio of Mg
2+
: NH
4
+
: PO
4
3-
for
the effective removal seemed to be 1.2 : 1.0 : 1.1.
Nelson et al. reported that adding magnesium ions did not
play an important role in phosphorus removal [11]. Therefore,
external addition of magnesium and phosphate should be con-
trolled to ensure the feasibility of struvite precipitation from
wastewater. Rahman et al. had a wide range of PO
4
3-
and Mg
2+
ratios tested for struvite precipitation, but in most cases, the effec-
tive ratio was 1 : 1 or 1 : 1.2 [28]. Most research to date has
reported that the optimum molar ratio of Mg
2+
: NH
4
+
: PO
4
3-
for struvite precipitation is between 1.0 : 1.0 : 1.0 and 1.6 : 1.0 :
1.0 [28], although phosphate removal is not affected when Mg
2+
: NH
4
+
: PO
4
3-
molar ratio is more than 1.3 : 1.0 : 1.0 at pH 9.0
in a full-scale plant [26].
There was a significant difference between ion removals in
real and synthetic effluents of swine wastewater anaerobic digester.
Ammonia N removal efficiency from synthetic wastewater was
over 90%, while real wastewater had lower than 50% ammonia
N removal. Addition of magnesium ion facilitated ammonia N
removal up to ratio of 1.2 and then it negatively influenced it.
However, more ammonia N was removed than would be predicted
based on the magnesium removal and the chemical formula for
struvite as shown in Fig. 2.
Table 3. Molar Ratios of Mg
2+
: NH
4
+
: PO
4
3-
for Ammonia and Phosphate
Removal
NH
4
+
Mg
2+
PO
4
3-
NH
4
+
removal (%) PO
4
3-
Removal (%)
1.0
1
1 86.9 97.1
1.1 88.7 88.2
1.2 88.4 83.4
1.3 90.4 76.4
1.5 91.2 70.3
2 83.8 61.1
1.1
1 88.8 99.2
1.1 92.7 96.9
1.2 93.6 89.4
1.2
1 87.5 99.8
1.1 94.5 98.9
1.2 95.8 95.9
1.3
1 87.4 99.4
1.3 98.6 98.2
1.5 97.6 41.8
2 95.4 78.3
1.5
1 92.3 99.9
1.3 98.4 99.3
1.5 98.3 95.3
2 97.7 93.9
2
1 89.9 99.5
1.3 98.1 99.4
1.5 98.7 99.4
2 99.0 95.0
Environmental Engineering Research 22(1) 12-18
15
Fig. 2. Nitrogen and phosphorus removal according to PO
4
3-
: Mg
2+
ratio for the effluent of swine wastewater anaerobic digester at pH 9.
Initial levels of magnesium, ammonium and phosphate ions
in the effluent of swine wastewater anaerobic digester were 32,
1,775, and 221 mg/L (molar ratio 1 : 74.9 : 1.8), respectively. The
ammonium ion concentration was much higher than magnesium
and phosphate ions concentrations. Therefore, magnesium and
phosphate ions sources had to be adjusted to completely remove
the ammonium ion. The experimental design allowed observing
the effects of magnesium and phosphate ions source dosage on
ammonia N and phosphate P removal as struvite. Experiments
were carried out with 10 min in the mixing zone and 3 h for
the whole reactor retention time and a pH of 9.0 according to
previous results.
The addition of magnesium and phosphate ions was required
to maximize ammonia recovery from the effluent of swine waste-
water anaerobic digester. The removal efficiency reached over
95%, and it was almost the same for the synthetic effluent.
Moreover, increasing added magnesium likely attributed to both
improvement of struvite precipitation and reduction of phosphate
P dose. Over dosing of magnesium could also contribute to de-
creased residual phosphorous concentration in the effluent and
phosphate recovery. However, if the concentration of magnesium
was increased up to a certain value, phosphorus removal would
not change [29].
Fig. 3. Addition of PO
4
3-
and Mg
2+
to real wastewater for enhanced
recovery.
3.2. Effect of pH on Ammonia N and Phosphate P Removal
The ideal pH range for struvite precipitation could occur at a
wide pH range of 7.0 to 11.5. However, the suitable pH range
for struvite formation is 8 to 9.5 [30]; this is consistent with many
other reports [17, 29]. Interfering ions in solution also affect the
pH range for struvite precipitation and nutrient removal. To inves-
tigate the effect of pH on ammonia N and phosphate P removal
and recovery from the effluent of swine wastewater anaerobic
digester, the residual concentrations of ammonia N and phosphate
P were examined after each experiment. Removal of ammonia
N and phosphate P was calculated based on the change between
the initial concentration and the residual concentration.
Experiments were carried out under the same reactor conditions
at the pH range of 6.0 to 12.0 and an equal ratio (Mg
2+
: NH
4
+
:
PO
4
3-
= 1.0 : 1.0 : 1.0).
Based on the batch experiment, the optimum pH for struvite
precipitation was investigated. Fig. 4 and Fig. 5 display the removal
efficiencies, depending on the pH, for ammonia N and phosphate
P in synthetic and real wastewaters, respectively. For synthetic
wastewater, as shown in Fig. 4, both ammonia N and phosphate
P removal efficiencies depended on the reaction pH, and the max-
imum ammonia N and phosphate P removal occurred at pH 9.0
and 11.0, respectively. For real wastewater, from Fig. 5, the optimal
Fig. 4. Ammonia N and P phosphate removal according to pH for
synthetic wastewater.
Fig. 5. Ammonia N and phosphate P removal according to pH for
real wastewater.
D. Kim et al.
16
range of phosphate P was reduced to between pH 8 and pH 10.
The maximum removal efficiency of phosphate P achieved were
over 95% in both wastewater types, while the maximum ammonia
N removal efficiency was very low in real wastewater due to the
high initial concentration of ammonia N. Thus, the pH of 8.0-10.0
can be considered as the optimum pH range for both ammonia
N and phosphate P removal from the effluent of swine wastewater
anaerobic digester.
Fig. 6 shows the optical microscope images of the struvite crystals
at various reaction pH values. This indicates larger struvite crystals
seen in higher pH values. Moreover, the increased size affected
struvite formation, and led to more precipitates forming at high
pH values. This could be explained in terms of more ammonia-based
precipitates forming compared to phosphate-based precipitates
at these conditions.
3.3. Pre-treatments Affecting Phosphate P Release
Acid-alkaline pre-treatments were applied for phosphate P release
from the effluent of swine wastewater anaerobic digester of P-city.
In this experiment, the initial pH was 7.2 and the pH was changed
with HCl and NaOH for the test pH range of 2.0-12.0. Acid-alkaline
pre-treatments were carried out under the same conditions so
that the ammonia N and phosphate P removals could be tested
in terms of the pH conditions of the wastewater. As shown in
Fig. 7, both T-P and phosphate P concentrations increased with
Fig. 7. Change of phosphorus concentration according to pH.
the increasing reaction pH, while poly-P concentration was slightly
decreased with the increasing pH. Fig. 8 shows the change of
phosphorus fraction according to pH. Maximum phosphate P re-
lease was observed at pH between 8.0-9.0.
In this work, the effect of ultrasonic and microwave pre-treat-
ment was also studied for changes in phosphate P release from
the effluent of swine wastewater anaerobic digester. The range
of the supplied energy density was from 100 kW/L to 20,000 kW/L.
From this analysis, phosphate P levels increased by increasing
ultrasonic energy density (up until 1,000 kJ/L); however, microwave
Fig. 8. Change of phosphorus fraction according to pH.
Fig. 9. Phosphate P release from swine wastewater according to energy
density.
a b
c
Fig. 6. Struvite crystals (1000X); a) pH 8, b) pH 8.5, c) pH 9.