**2. Phosphorus recovery as struvite**

The shift into circular use of P passes through the implementation of multidimensional innovations focused on the creation of P recovery/recycling technologies [14]. The main well-recognized "waste" streams containing nutrients and causing negative *Phosphorus Recovery through Waste Transformation: Implication for an Alternative Fertilizer DOI: http://dx.doi.org/10.5772/intechopen.111856*

#### **Figure 2.**

*Phosphorus production, consumption, and recycling pathways [13]. The color code indicates the impact in terms of environmental health and sustainability: Blue – Neutral (currently applied practices); red – Negative practices to be avoided or minimized; green – Processes with positive impact.*

environmental effects include animal manure, urban wastewater and sewage sludge, and food processing wastewater [15]. In the following sections, this chapter discusses two specific potential sources of P, namely the sewage and manure waste streams.

Domestic wastewater and sludge could be regarded as an important secondary source of P. The sludge water from municipal wastewater treatment plants (WWTP) is a fluid containing a reasonable level of P (120–160 mg/l PO4 3−), which allows the extraction of this nutrient with good economic effect, especially taking into account the global volumes of this waste stream. According to many authors, its potential counts for 15–20% of the global phosphorus demand [8, 16, 17]. For this reason, phosphate recovery has high priority in sewage treatment. European Union estimations show that the sludge amount produced annually is over 7 Mt. dry solid (DS), while globally, 1.3 Mt. P/year is treated in WWTPs worldwide [8, 17]. Animal manure production is another significant waste stream containing P. The amount of P in manure compared to municipal wastewater depends strongly on the manure source and type. Common *o-*phosphate concentrations of manure liquid fraction varied in a wide range of 50–1200 mg/l. However, the high suspended and dissolved organic matter should be taken into consideration in choosing the P recovery option. The available data show that in evaluating the by-products and waste sources of P in Mt. per year (**Table 1**) P in animal manure is almost one order of magnitude higher in sewage sludge (20–30 Mt. vs. 3–5 Mt., respectively) [19].


#### **Table 1.**

*Series of experiments using supernatants produced after centrifugation of sludge from WWTP-Burgas [18].*

#### *Phosphorus in Soils and Plants*

Interestingly, several technologies for P recovery from wastewater and sludge are operating at either full or demonstration scale. However, P-recovery technologies are focused mainly on the aqueous phase of sludge (the so-called sludge water). In the solid phase, the technologies are directed either to recover P from the dewatered sewage sludge or from mono-incinerated sewage sludge ashes [20]. The accepted means differ in respect of technology choice, costs, efficiency, and product purity.

Chemical precipitation of P into salts of low solubility is a common method for removing dissolved phosphorus from wastewater: in the form of magnesium ammonium phosphate (MAP) hexahydrate (MgNH4PO4·6H2O), also called struvite, or calcium phosphates [21]. Struvite has the advantage of being a slow-dissolving salt, while calcium phosphates are characterized with extremely low water solubility. P-recovery is usually based on the chemical precipitation of struvite from concentrated wastewater or the liquid fraction remaining after anaerobic digestion of sludge in WWTP. The chemical precipitation of struvite removes up to 98% of the soluble phosphates and 20–30% of the soluble ammonia contained in the targeted liquor [22]. Notably, the main challenge in relation to struvite precipitation is the P-recovery from wastewater characterized by phosphorus concentration of less than 50 mg/L and suspended solids concentration (TSS) above 1000 mg/L [23].

The struvite formation process is a function of pH and the molar ratio between magnesium, ammonium, and phosphate ions. The precipitation occurs in alkaline conditions and optimal crystal formation is observed at pH above 9 and equimolarity of the constituent ions [24]. However, several studies show that struvite can be synthesized in a wide range of pH values. Even experiments applying a pH slightly above 8 show positive results, but the formation and precipitation rates are lower compared to the optimal pH range. This is confirmed by studies with real dewatered sludge liquor (DSL) taken from the MWWTP of Burgas, Bulgaria (initial PO4 3−concentration of 86.7 mg/l) [18]. In the same study, MgCl2 was used as a precipitation agent while different pH and molar ratio Mg:PO4 were applied (**Table 1**). Obviously, both pH and mole ratio Mg:PO4 are crucial factors affecting precipitation efficiency. However, the highest effect is observed at pH near and above 9.0. At extremely high pH values (above 11), the struvite yield decreased due to the formation of Mg(OH)2 and transformation of ammonia ions into free NH3 (reducing the general availability of Mg2+ and NH4 + in the medium). Experiments carried out by Saidou et al. [25] showed that at initial solution pH of 10, another phosphate mineral, namely Mg3(PO4)2.2H2O starts precipitating. Such formation results in other dominant species formation. In fact, varying the pH levels results in different species of phosphate ions. The precipitation of struvite requires equal molar ratios of its components: magnesium, ammonia, and phosphate ions. Different molar ratios have been shown to influence several characteristics of the struvite. It could be argued that as more Mg is available, more crystal units are generated, hence larger crystals could be formed. As reported by Merino-Jimenez et al. [26], several studies have applied the ratios of phosphate and magnesium ions around 1:1.2 for optimum struvite yield. The latter result confirms the effects shown in **Table 1**.

As it was shown above, one of the main factors for struvite formation is the pH. In the targeted P-containing waste streams, approximately 90% of P is trapped in sewage sludge following primary and secondary sedimentation [27]. The pH of these fluids (usually centrate) is in the range of 7.3 and 7.5; far from the optimal pH values providing technologically acceptable conditions for MAP crystals formation. In this case, as a general option struvite crystallization is achieved through alkalization by reagents such as NaOH.

*Phosphorus Recovery through Waste Transformation: Implication for an Alternative Fertilizer DOI: http://dx.doi.org/10.5772/intechopen.111856*

**Figure 3.** *Air stripping unit design [28].*

Besides the direct reagent alkalization there, pH can be elevated by carbon dioxide (CO2) stripping through aeration of the reagent mixture. **Figure 3** shows the principal scheme of the striping system [28]. The mechanism of this treatment is based on the manipulation of the dissolved carbon dioxide mass balance in the liquid. The aeration of the solution leads to the decomposition of the carbonic acid (H2CO3) to H2O and CO2 which results in pH elevation. The described mechanism seems to be appropriate in the case of sludge water treatment as it contains high amounts of organic matter which potentially enriches the liquid with CO2 due to the microbial activity occurring in the suspension. It was found that the rate of pH elevation depends strongly on the spangling area of the air distribution system while the air flow rate does not influence considerably the dissolved oxygen level which governs the CO2 stripping process [28]. The theoretically calculated values of the volumetric mass transfer coefficient have been compared with those obtained experimentally. Based on the data obtained, relationships of pH/kLa (mass transfer coefficient) were developed. These correlations serve as a tool for the prediction of pH during the struvite precipitation process. The pH dependence of air rate is given in **Figure 4**.

The results showed evidently that within the volumetric air rates applied pH elevation from 7.5 to 8.4 was achieved by CO2 stripping in less than 25 minutes. Such a retention time in the reactor is technologically applicable. Actually, a pH value around 8.5 is high enough for effective struvite precipitation even in several studies the optimal pH for the process is shown to be in the range of 9 to 9.5. The targeted pH 9 could be also achieved in an acceptable retention time of less than 1 hour. On the other side, the slow pH change during the aeration could be considered an advantage of the CO2 stripping process because it restricts the rapid increase of solution saturation. At such conditions, the struvite crystallization process predominates, and in addition, Ronteltap et al. [29] reported that the crystal sizes at pH values of 7–11, and found that the largest crystals occurred at pH 8.

The concentration of magnesium is the other limiting factor for struvite crystal formation, the choice of Mg source is important as it forms nearly 75% of the struvite production costs [21]. Several commercial magnesium salts such as MgSO4, MgCl2, MgO, or Mg(OH)2 have been used to precipitate struvite from different liquid wastes [30]. Cheaper options based on seawater as a source of Mg ions are thoroughly studied as well demonstrating a comparatively high phosphorus recovery rate (80–90%) and precipitation of several different products with the domination

#### **Figure 4.**

*Elevation in pH by air stripping through applying different volumetric air rates [28].*

of struvite (but also containing magnesium calcite and calcite due to the complex ion composition of the seawater) [31].

Similar results were obtained in the research group of the authors, where seawater brine was used to precipitate P from WWTP sludge centrate [32]. Also, the seawater brine used was found to contain significant amounts of calcium and potassium (**Table 2**). However, the magnesium concentration is higher (59.5 g/l) than the calcium concentration (3.5 g/dm3) which resulted in effective struvite precipitation with minimal calcium phosphate co-production during the process.

The MAP crystals obtained by sweater brine used during this study were compared with struvite produced via the application of a conventional magnesium source (MgCl2.6H2O). Starting from a model solution of (NH4)2HPO4 the precipitation was carried out at a mole ratio Mg:P = 2:1 and pH of 9.5, within 15 minutes of continuous slow mixing (50 rpm). The crystals obtained following a period of 30 minutes for struvite agglomeration, were studied under a scanning electron microscope (SEM) (**Figures 5** and **6**).

The SEM images showed that in both cases the crystals obtained are of typical struvite morphology. Brine-induced crystals have an average size (length) of 280–300 μm, while the crystals precipitated by MgCl2 are of size close to 200 μm. The difference can be explained by the calcites produced during the precipitation of seawater brine.


#### **Table 2.**

*The concentration of main elements present in the seawater brine [32].*

*Phosphorus Recovery through Waste Transformation: Implication for an Alternative Fertilizer DOI: http://dx.doi.org/10.5772/intechopen.111856*

**Figure 5.** *Struvite crystals precipitated by seawater brine.*

**Figure 6.** *Struvite crystals precipitated by MgCl2.6H2O.*
