/Rapport+Olegsander+Panasiuk

Oleksandr Panasiuk

PHOSPHORUS REMOVAL AND RECOVERY FROM WASTEWATER USING MAGNETITE

Supervisors: Per Olof Persson & Bengt Hultman Examiner: Per Olof Persson,

Master of Science Thesis STOCHOLM 2010

PRESENTED AT

INDUSTRIAL ECOLOGY ROYAL INSTITUTE OF TECHNOLOGY www.ima.kth.se

TRITA-IM 2010:25 ISSN 1402-7615 Industrial Ecology, Royal Institute of Technology www.ima.kth.se

Abstract The aim of this work was to study the possibilities of using magnetite for phosphorus removal and recovery from wastewater. It was also aimed to investigate how the structure of magnetite influences the efficiency of adsorption and desorption of phosphorus.

Methodology used in this study is literature review and laboratory experiments. The study is mainly focused on the influence of Fe(II)/Fe(III) ratio in magnetite (coefficient K) on the P removal and recovery rate. Several sets of experiments were also done to study the influence of some factors (e.g. contact time, starting concentrations, amount of base needed, etc.) on the efficiency of the processes.

Study results showed that magnetite has a great potential for phosphorus removal because of its high efficiency, especially at low concentrations of input phosphorus. It was also found that the contact time and sedimentation time of the method is relatively small.

Recovering of magnetite is also possible, but for its reuse additional renovation stage is needed.

It was concluded that magnetite purification could be preferably used as the polishing method. It can be introduced in already existing wastewater treatment facilities and substitute some older technologies. The method seems to be easy in starting and operation; it has relatively low operational and investment costs.

Key words: phosphorus, removal, recycling, wastewater treatment, magnetite.

III

Acknowledgements The work for this thesis was done at the Department of Industrial Ecology at KTH and the experiments were conducted at the research facility Hammarby Sjรถstadsverk.

First of all, I would like to express gratitude to my supervisors Per Olof Persson (KTH) and Bengt Hultman (KTH) for their valuable comments and ideas that helped me a lot during performing of presented work.

I would also like to thank all people from Hammarby Sjรถstadsverk for the nice working atmosphere during my experimental part there. Special thanks to Christian Baresel and Lars Bengtsson for providing all necessary conditions for me to perform the experiments.

I am very thankful to the Royal Institute of Technology (KTH) and all my teachers from the Master Program of Sustainable Technology.

Thanks to my parents and sister who have always believed in me and to my friends who have always been there when needed.

Last but not least I would like to thank to my wife for her love, patience and constant support during my studies at the Master program.

IV

Table of content 1.

Introduction ........................................................................................................................................................... 1 1.1. Aims...................................................................................................................................................................... 1 1.2. Objectives .......................................................................................................................................................... 1

2.

Theoretical background ................................................................................................................................... 3 2.1. Origins of phosphate in water bodies .................................................................................................. 3 2.2. Eutrophication ................................................................................................................................................ 4 2.3. Methods used to decrease phosphate concentration................................................................... 5 2.3.1. Precipitation, coagulation, and flocculation....................................................................... 6 2.3.2. Enhanced biological phosphorus removal ......................................................................... 7 2.3.3. Other methods for phosphorus removal ............................................................................. 9 2.4. Drawback of current methods.............................................................................................................. 10 2.4.1. Disadvantages of chemical precipitation ......................................................................... 10 2.4.2. Disadvantages of enhanced biological phosphorus removal ................................. 10 2.4.3. Disadvantages of membrane technologies ...................................................................... 11 2.5. Phosphorus recycling ............................................................................................................................... 11 2.5.1. Methods of phosphorus recycling........................................................................................ 12 2.6. Adsorption of phosphorus using magnetite .................................................................................. 13 2.6.1. Magnetite ......................................................................................................................................... 13 2.6.2. Magnetite in wastewater treatment ................................................................................... 14 2.6.3. Methods of magnetite synthesis ........................................................................................... 15

3.

Experiments description............................................................................................................................... 17 3.1. Materials ......................................................................................................................................................... 17 3.2. Experiments .................................................................................................................................................. 17 3.2.1. Phosphorus stock solution preparation ........................................................................... 17 3.2.2. Measuring of phosphorus concentration. Dilution ...................................................... 17 3.2.3. Magnetite synthesis .................................................................................................................... 19 3.2.4. Magnetite concentration .......................................................................................................... 20 3.2.5. Effect of different Fe(II)/Fe(III) ratio in magnetite on P removal efficiency .. 21 3.2.6. Adsorption kinetics..................................................................................................................... 21 3.2.7. Adsorption isotherm of phosphorus on magnetite ..................................................... 21 3.2.8. Amount of base needed for phosphorus recovery from magnetite .................... 21 3.2.9. Recovery of phosphorus from magnetite......................................................................... 21 V

3.2.10. Using recovered magnetite for phosphorus removal ................................................ 22 4.

Results.................................................................................................................................................................... 23 4.1. Magnetite concentration ......................................................................................................................... 23 4.2. Effect of different Fe(II)/Fe(III) ratio in magnetite on P removal efficiency................. 23 4.3. Adsorption kinetics.................................................................................................................................... 24 4.4. Adsorption isotherm of phosphorus on magnetite .................................................................... 25 4.5. Amount of base needed for phosphorus recovery from magnetite ................................... 26 4.6. Recovery of phosphorus from magnetite ....................................................................................... 26 4.7. Using recovered magnetite for phosphorus removal ............................................................... 28

5.

Discussion ............................................................................................................................................................ 29 5.1. Results’ discussion ..................................................................................................................................... 29 5.1.1. Results’ accuracy .......................................................................................................................... 29 5.2. Future study possibilities ....................................................................................................................... 30

6.

Conclusion............................................................................................................................................................ 31

References..................................................................................................................................................................... 33

VI

1. Introduction Phosphorus is a vital element for every plant and animal. Lack of phosphorus in ground can result in limited crop production. Phosphorus is mainly used in agriculture as a fertilizer or as food additive in the animal feed. The other applications of phosphorus include ingredients for human food, pharmaceuticals, detergents and some especial chemicals (DEFRA, 2008).

The phosphorus content in natural waters is usually regulated by microorganisms, so there is a balance in the available phosphorus and ecosystem requirements. However, if P input exceeds the ecosystem consumption rate, the problem of increased phosphorus concentration appears.

Excess of phosphorus stimulates algae growth in water bodies, which in its turn decreases oxygen concentration and leads to eutrophication of the surface water bodies.

Increased phosphorus concentration also results in higher water treatment costs, decreased recreational value, and livestock losses. There is also a high probability of the sub-lethal effect because of toxins from algae in the drinking water (Defra, 2008). The probability and effect of these problems are as higher as smaller the water body is. The new EU Directive 2000/60/EC, 2000 requires controlling of phosphorus discharge to maintain or improve the water ecology (EC, 2000).

As phosphorus is non-renewable element, its recycling is also of great interest. Huge amount of phosphorus is lost annually for lack of P recovery. Methods of phosphorus extraction from wastewater for its recycling should be developed.

Current methods of phosphorus removal and recovery used on the wastewater treatment facilities have several strong drawbacks. This causes different environmental problems and also limits recycling of phosphorus in present treatment systems. In this case application of magnetite for phosphorus adsorption seems to be good alternative.

Presented work studies different aspects of magnetite application both for removing and recycling of phosphorus.

1.1.

Aims

The aims of this work are: — to study the possibilities of using magnetite to remove P from waste water; — to investigate the influence of magnetite structure on the P removal efficiency; — to study the possibility of phosphorus recovery, its efficiency depending on magnetite properties.

1.2.

Objectives

Objectives of the study are: — to make a literature review about present technologies of phosphorus removal, available techniques; and current usage of magnetite in wastewater treatment; — to carry out laboratory experiments showing how magnetite affects phosphorus concentration in wastewater; 1

— to analyze the possible application of the method; — to define the future perspective of magnetite application for wastewater treatment in general and for P removal in particular.

2

2. Theoretical background 2.1.

Origins of phosphate in water bodies

Phosphorus cycle in the nature could be described with the following scheme (Spellman, 2008) (Figure 1).

Organic P plants, animals, and bacteria

Phosphate rocks, guano deposits, and fossil bone deposits Protoplasm synthesis

Erosion Dissolved phosphates

Marine birds and fish

Volcanic apatite Phosphatizing bacteria

Excretion

Shallow marine sediments

Bones, teeth

Loss to deep sediments

Figure 1. Phosphorus cycle (from Spellman, 2008)

Unlike the nitrogen and carbon cycles that are global, the phosphorus cycle is local. This is because the main abiotic reservoirs of phosphates are rocks. Erosion from the rocks transforms phosphorus compounds into soluble phosphates. Then dissolved phosphates either leach to the water bodies or sink into the soil. In the latter case they are consumed by growing plants, which in their turn are eaten by animals. Animals and plants are returning dissolved phosphates to the cycle in form of excretion and when they die, detritivores transform organic phosphorus compounds back into inorganic one.

There are three main ways of how phosphates enter wastewater and ground water: 1. leaching from the natural mineral deposits; 2. agriculture; 3. sewage disposal and liquid urban waste (domestic and industrial).

Sewage disposal is considered as the main source of phosphorus entering the water bodies.

For example, about 75 percents of total phosphorus discharge in the Netherlands comes from the domestic wastewater, compared to the total municipal wastewater volume (van Starkenburg & Rijs, 1988). Looking at some agricultural areas with intensive water usage from lakes, the agricultural P input could be dominant in this case (Steen, 1998).

Phosphorus in wastewater exists in three main forms: orthophosphate ion, polyphosphates and organic phosphorus compounds (Mahmut & Ayhan, 2003).

3

The primary material that is used for phosphorus production is apatite [Ca5(PO4)3OH]. The process of P production is highly energy-consuming and causes a lot of environmental problems connected mainly to the stage of raw material extraction (Better Crops, 1999). In this case, closing of material loops of phosphorus by its removal and recycling from wastewater is of increasing interest during last years.

2.2.

Eutrophication

As was written above, phosphorus is the vital element for the plants and animals.

Phosphorus is presented in natural waters usually in form of phosphates. They are forming insoluble compounds with calcium and other salts settling as the sediments on the bottom.

Thus, available concentration of P in natural water bodies is usually very low and it is the limiting factor for the algae and other organisms to grow. An excessive amount of phosphorus breaks this limit and rapid microorganisms’ growth is initiated. This widespread today problem is known as eutrophication, Eutrophication in general is the process during which the formerly deficient nutrients are coming to the surface water. Looking backwards, eutrophication was a natural process and took centuries to transform water body into a dry land. Nowadays, the process is greatly accelerated by human activities (Figure 2).

Figure 2. Lake aging: natural process and accelerated by land use (Artwork by W. Feeny. Source: Carpenter et al., 1998).

According to research (UNEP, 1994) more than half of all lakes in Europe and Asia are eutrophic (53% and 54% correspondingly), about the half (48%) — in North America, 41% in South America and 28% in Africa. In the USA the water bodies that cannot be used for 4

drinking, fishing, recreation, irrigation or industry purposes are impaired in the 60% of cases because of eutrophication (Carpenter et al., 1998). This shows the scale of the problem and its importance.

Eutrophication has several negative impacts that result one from another and are interconnected: — the proliferation of bloom-forming species that could be toxic or inedible; — decreasing of water quality (e.g. color, transparency, odor, taste and water treatment problems); — water plants die causing propagation of bacterial decomposers; — bacteria consume more oxygen causing its shortage in water body, and thus, fish kills; — loss of biodiversity; — decrease of esthetic and recreational value of the water body.

During the period of 1950 to 1995, approximately 600 million tonnes of phosphorus were used as a fertilizer globally. In the same period, about 250 million tonnes of P were collected in form of different crops. Fifty million tonnes of P came back to the croplands with the manure form the livestock eating part of harvested crops. It appears from this that 400 million tonnes of phosphorus were added to the soil during that period. Taking into account that 3 to 20% of that amount was washed to the water bodies it results in 12 to 80 million tonnes of phosphorus discharged in the surface water (Carpenter et al., 1998).

In this case the phosphorus removal from wastewater is extremely important as it both saves enormous amount of non-renewable phosphorus and also prevent water eutrophication and contamination.

2.3.

Methods used to decrease phosphate concentration

There are several ways to reduce the amount of phosphate entering into environment from inland human activity: — to reduce emission from the primary sources: point (e.g. wastewater treatment plant) and diffuse (e.g. agriculture); — to decrease phosphate input to the material flow and, as a result, to the sewage (e.g. replacement of phosphates in detergents); — to remove phosphates at the wastewater treatment plant (e.g. by introducing additional removal capacity).

This study is focused on the last option, i.e. technical measures to reduce phosphorus concentration in the effluent of the wastewater plant. In order to prevent negative impact of phosphorus on the environment the limits on total P discharges have to be set. In many countries this limit is 1 mg/L or 2 mg/L of total phosphorus. Such a low limit can be explained also by the fact that concentration of phosphorus less than 0.5 mg/L limits (inhibits or even blocks) the growth of algae (Dryden & Stern, 1968).

There are two fundamental ways to remove phosphorus from wastewater: (1) physicalchemical precipitations and (2) enhanced biological removal. Combinations of these two methods and some other specific technologies are also used for phosphorus removal.

5

2.3.1. Precipitation, coagulation, and flocculation Chemical precipitation is one of the most common ways of phosphorus removal used for a long time. It is a method that causes dissolved phosphorus to settle out of solution (Figure 3). This insoluble phosphorus-containing sediment can then be settled, centrifuged, filtered, or separated from the liquid by other methods and is called precipitate (EPA, 2000). The liquid part after precipitation is called supernate. During formation and settling, the precipitate can catch ions and particles from solution, increasing efficiency of the method (Tchobanoglous & Burton, 1991).

Figure 3. Chemical precipitation diagram

In order to get precipitation to take place, the agent called coagulant need to be added. It causes small suspended matter to group into bigger aggregates. Compounds of iron, aluminum and calcium are chemicals primarily used in P precipitation: ferric chloride, ferrous sulfate, aluminum sulfate (alum), and lime.

If lime is used, special conditions should be fulfilled to ensure the reaction between excess calcium ions and phosphate. This could be done if pH of the solution is not less than 10, so it is important to add sufficient amount of lime. (Tchobanoglous et al., 2002). In spite of the fact that lime is an effective agent for P removal, the application of this reagent is slightly diminished because of high volume of produced sludge.

In case of iron or aluminum salts are used in phosphorus removal, insoluble metal phosphates are produced. The formation of these compounds is pH-dependent: pH level strongly affects the degree of insolubility of metal phosphates. Another issue that should be taken into account is different completing reactions that occur in the system in addition to the main one. As a result the amount of metal salts should be determined practically during the experiments for each case and cannot be calculated simply based on the chemical reaction with phosphorus (Tchobanoglous & Burton, 1991).

Study on precipitation (Song et al., 2002) based on thermodynamics showed that the theoretical removal rate of phosphorus depends on such factors as P concentration, temperature, pH, and ionic strength. From the other hand, it has also been shown that the real picture is much more complicated than the theoretical calculations and there are more 6

factors which influence the P removal. Prediction of the best performance of P chemical precipitation also varies in different studies from 0.005-0.04 mg/L (TakĂĄcs, 2006) to 0.05 mg/L (Neethling & Gu, 2006).

Coagulation and flocculation allow removing of the suspended compounds or compounds in colloidal form from wastewater. In the first case polyvalent ions like Fe3+ or Al3+ are used, while in the second the long-chain polymers are added as the flocculation agents (Figure 4a and 4b). a)

b)

Figure 4. Two different mechanisms for chemical flocculation of colloidal particles: a) addition of longchain polymers and b) addition of aluminum ions (redrawn from Persson & Nilson, 2005) .

Polymers’ long-chain molecules can be either positively (cationic) or negatively (anionic) charged or be neutral (nonionic). These electrical qualities of polymers are very useful since in wastewater treatment the interactions usually take place between ions or other charged particles. In this case, molecules of polymers help to connect particles in the solution or to neutralize them (Amirtharajah & O’Mella, 1990; Jacangelo, 1987) (Figure 4a).

The coagulation process is based on another mechanism: charged ions can attach to the surface of colloidal particles making neutral particles; Van der Waals forces between particles can ensure formation of the bigger particles that will settle more easy (Figure 4b).

The efficiency of the process depends a lot on the charge of salt ions added. The higher charge the ions have, the better coagulation is taking place (Persson & Nilson, 2005).

There are different strategies on how, when and where to add the precipitation, flocculation and coagulation agents.

2.3.2. Enhanced biological phosphorus removal Enhanced biological phosphorus removal (EBPR) is a widely used method to decrease P concentration in a full-scale wastewater treatment plants (Sedlak, 1991). It is of great interest because of possibility to reach low or even very low P concentrations (less than 0.1 7

mg/L). The other strong sides of the method are minimal sludge production and moderate operational cost (Strom, 2006). Compared to chemical precipitation, the main advantage of EBPR is the absence of metal ions from coagulant in the sludge.

EBPR is a process held in the environment that force biomass to consume more phosphorus than generally is needed for growth in the normal conditions (Metcalf and Eddy, 1991). During EBPR process activated sludge should adapt to the changing anaerobic-aerobic conditions (Figure 5). Biodegradable carbon sources are available in the anaerobic stage only (Loosdrecht et al., 1997). The excessive biological sludge after aerobic stage is partly returned to the anaerobic reactor.

Figure 5. Process flow diagram for an EBPR process (adapted from Woods et al., 1999).

Phosphorus removal is performed mainly by a group of microorganisms known as the polyphosphate accumulating organisms (PAO). These organisms can consume and store P in form of intracellular polyphosphate. This fact leads to the decreasing phosphorus content in the liquid phase and a concentration of P in the activated sludge. PAOs are able to consume carbon sources, e.g. volatile fatty acids (VFAs) in the anaerobic conditions, and store them in form of carbon polymers (poly-β-hydroxyalkanoates, PHAs) inside the cell.

Ability to uptake carbon sources is the distinguishing feature of PAOs in comparison to most other microorganisms (Mino et al., 1998).

For phosphorus consumption and polyphosphate storage, as well as for biomass growth, stored PHAs can be used by PAOs as the energy source under aerobic conditions. Overall phosphorus reduction is reached by removing activated sludge waste with high polyphosphate concentration. Usually the P removal is achieved in anaerobic-aerobic process in EBPR. Still, some POAs can use nitrite or nitrate instead of oxygen as acceptor for electron, and anaerobic-anoxic process will also result in P uptake. Moreover, the anaerobic-anoxic process allows simultaneous removal both phosphorus and nitrate. In this case both savings in aeration and smaller need in amounts of additional carbon source result in decreasing of process operational costs. Nowadays there are a lot of process configurations that combine both phosphorus and nitrogen uptake (Henze et al., 1997; Tchobanoglous et al., 2002).

During successful operation the EBPR is quite moderately priced and is also environmentally sustainable process for phosphorus removal. From the other hand, reliability and stability of this method could be a problematic issue. Process breakdown, performance worsening, and even failure can happen during EBPR process (Oehmen et al., 2007). In the work done by Neethling et al. (2005) factors affecting the EBPR reliability 8

were studied. BOD/P ratio more than 25:1 and recycle streams control are some of the factors that help to keep reliable and high removal level. In the same work it was also found that the lower concentration of phosphorus one can achieve the smaller period of time it can be maintained on the given level: (i) P concentration less than 0.1 mg/L — during quite long period (more than a month); (ii) 0.03 mg/L — during a week; (iii) less than 0.02 mg/L — during a couple of days.

2.3.3. Other methods for phosphorus removal Physical methods include (a) particulate phosphorus removal and (b) membrane technologies. (a) Particulate phosphorus removal is a method based on the removing of all phosphorus compounds, both organic and inorganic, that can be hold by the filter (Rigler, 1973). The method could be used as the primary stage in highly contaminated water with high content of particulate P. It could be done using e.g. sand filter or wetland filtration (Lowe et al., 1992). (b) Membrane purification is based on the semipermeable properties of the selective separation wall: certain components can go through membrane, while other are caught (Figure 6).

Figure 6. Example of membrane technology: in membrane bioreactor (MBR) microorganisms consuming dissolved materials unable to cross the membrane, so only purified water is going out.

Membrane technologies have become recently of great interest for wastewater treatment.

Later their application for phosphorus removal has been studied. It has been shown that membrane methods are able to remove not only phosphorus in total suspended solids but also the dissolved P. According to report (Reardon, 2006), methods involving membrane technology showed good results in full-scale plants: less than 1 mg/L of total phosphorus in their effluent. These methods include membrane bioreactors (MBRs), reverse osmosis and tertiary membrane filtration. In the same report current reliable limits were suggested: 0.04 mg/L for tertiary membrane filtration and MBRs, and 0.008 mg/L for reverse osmosis.

9

There are other chemical methods of P removal (based mainly on the physical-chemical adsorption). For example, magnetically enhanced coagulation method can achieve low concentrations of phosphorus less than 0.03 mg /L of total phosphorus (Woodard, 2006).

System based on iron reactive filtration can achieve up to 0.01 mg/L of total P at the average flow plant (Möller 2006).

One more method that is becoming more popular recently is removal of phosphorus from wastewater in form of struvite (Magnesium ammonium phosphate hexahydrate, MgNH4PO4∙6H2O). The simplified reaction of struvite formation can be shown as follows: Mg2++NH4++PO43-+6H2O → MgNH4PO4∙6H2O The benefit of struvite formation is that the method is used not just for P removal but also as a phosphorus recovery process. It was found that about 80–90% P recovery from anaerobic digester supernatant could be obtained using, e.g. fluidized bed MAP (Magnesium Ammonium Phosphate) crystallizer (Adnan et al., 2003).

2.4.

Drawback of current methods

As was described above, present treatment methods have some advantages but also a lot of disadvantages.

2.4.1. Disadvantages of chemical precipitation The major drawback of the chemical precipitation is that during phosphorus removal an additional amount of sludge is produced. Situation is becoming worse if lime is used for the primary treatment (Tchobanoglous et al., 2002). In this case the volume of sludge can increase up to 50%. Application of alum instead of iron salts and, especially, lime results in much smaller amount of sludge but doesn’t solve the problem completely (Strom, 2006).

Other disadvantages of chemical precipitation are (from EPA, 2000): — Usually, calculations of accurate dosages of chemicals are impossible because of changing levels of alkalinity, different competitive reactions and other factors. And so, it is necessary to make batch tests often to prove the optimal treatment conditions. One should also keep in mind that overdosing can result in reduction of treatment efficiency.

— It is more difficult to recover P from the sludge in comparison to the EBPR. — There are increased concerns about operator safety as chemical precipitation can involve contact with corrosive chemicals.

— Necessity to transport large amounts of chemicals to the wastewater treatment facilities usually needed.

— High prices for polymers that can be used.

— And, finally, the huge amounts of phosphorus in sludge are mainly lost.

2.4.2. Disadvantages of enhanced biological phosphorus removal As it was stated above, the EBPR process has low reliability and stability. The decreasing in the efficiency of the method or even failures can happen during EBPR process because of different reasons. For example, high rainfall, nutrient limitation, excessive nitrate input in the anaerobic reactor and other external disturbances can cause the process upsets.

10

The other reason for decreased efficiency of P removal could be microbial competition between polyphosphate accumulating organisms (PAOs) and glycogen (nonpolyphosphate) accumulating organisms (GAOs). GAOs are also able to breed under changing anaerobic-aerobic conditions like PAOs. The difference is that GAOs do not uptake P therefore don’t contribute to the P removal process (Mino et al., 1995). GAOs are highly unwanted in EBPR systems as they consume VTA (volatile fatty acids) without phosphorus removing; nevertheless GAOs have been found in EBPR many plants. One of the solutions in this case is to increase amount of VTA during anaerobic stage. Solving GAOs’ problem is recently becoming of great interest because of the increasing cost-effectiveness opportunities of current process (Oehmen et al., 2007).

2.4.3. Disadvantages of membrane technologies There are several significant drawbacks of utilization of membrane separation for phosphorus removal: — quite high initial investments; — expensive membrane replacement, shorter economic life of membrane; — higher energy consumption in comparison to other methods; — larger amounts of sludge produced.

All the above-listed result in much higher operational and maintenance costs of membrane technology than for other treatment methods (Jiang et al., 2004).

2.5.

Phosphorus recycling

Aside from removal of phosphorus from wastewater, this paper addresses also methods of its recycling and reusing. It is very important as phosphorus is a deficient vital resource and unlike fossil fuels cannot be substituted by other resources (Steen 1998). Moreover, large industrial unit CEEP (Centre Européen d’Etudes sur les Polyphosphates) that coordinates some programmes in phosphorus recycling announced the goal to reach 25% of phosphorus recovery during the next decade (CEEP, 1998). Sweden has even more ambitious plans to recover at least 75% of phosphorus from wastewater (Hultman et al., 2001).

Historically some amount of phosphorus is recycled in agriculture: animal manure, human waste, and burned wood ash are returned to the land (Hislop, 2007). Modern society with its rapid growth and increased volumes of phosphorus-containing sewage water is discharging enormous amount of phosphorus to rivers, lakes and seas or losing it in form of sludge. Without recycling of phosphorus to the ground, modern agriculture depends a lot on the mineral phosphorus fertilizers. These fertilizers are produced from phosphorus rock and form about 87% of total phosphate rock consumption. Remaining 13% are represented by industrial use: detergents — about 8%; food — about 1%; and other purposes — around 4% (Wind, 2007).

As the reserves of phosphorus rock are limited, it is extremely important to close the material loops and recycle phosphorus for sustainable development. In addition to that, phosphates recycling can save energy if one compares it to the common manufacturing and transportation of fertilizers from the phosphate rock.

11

2.5.1. Methods of phosphorus recycling There are different approaches in recovering of phosphorus after its removal from wastewater (Rybicki, 1997): — wastewater streams separation to recover phosphorus from the most concentrated ones; — recovery of phosphates using ion exchange; — phosphates and coagulants recovery using sulfate oxidizing and sulfite reducing bacteria; — recovery of magnesium or calcium phosphates from biological treatment plants.

There are three main methods of P removal from human wastes: a) As mentioned above, separation of black and grey wastewater. This can result in more than 80% nitrogen and phosphorus removal (Jenssen & Etnier 1996). b) Utilization of old agricultural method of using wastewater treatment sludge as manure. Some questions about environmental safe and social acceptance may accompany this method (Kuile et al., 1983). c) Solubilization of sludge parts using chemical, biological and/or thermal methods.

This results in organics, metals, nitrogen and phosphorus compounds to be released in a concentrated liquid stream. Different products can then be recovered using selective methods. Phosphorus obtained in such a way can be of high purity (Dirkzwager et al., 1994) In general, phosphorus recovery from the sludge of wastewater treatment plants can be described as follows (Stark, 2002): 1. by using EBPR or algae transfer soluble compounds of phosphorus into biomass or by using chemical precipitation transfer it into chemically bound compounds; 2. by using different physical, chemical, mechanical or biological methods in order to dissolve phosphorus compound in a smaller stream, thus concentrating it; 3. as the concentrated stream can also contain other components, separation technologies (e.g. ion exchange, membrane technologies, crystallization, or chemical precipitation) should be used to obtain clean enough phosphorus product.

Not so long ago just the first stage of phosphorus removal was being developed and now technologies for two last stages need to be improved.

In principle, release of phosphorus (and other compounds) can be done in two ways: using acid or base. The positive feature of using acids is high recovery grade. From the other hand, a lot of other metals including heavy metals are also released causing separation problems and increasing complexity of process technology. Leaching with base has up to 50% lower recovery efficiency however metal leaching is also reduced. Another point that should be taken into account while using base leaching is that calcium presented in sludge will more likely bind phosphorus into calcium phosphate (Levlin & Hultman, 2005).

It is also important to mention that during the process of phosphorus adsorption, regeneration of adsorbent and phosphorus precipitation from regeneration liquid, base

12

(e.g. NaOH) that is produced after precipitation can be recycled for regeneration stage. It can be illustrated with the following chemical reactions (Jiang et al., 2004): Phosphorus adsorption: (Al2O3)n∙2Al(OH)3 + 3H+ +PO43- → (Al2O3)n∙Al(OH)3∙AlPO4 + 3H2O Adsorbent regeneration: 3Al(OH)3∙AlPO4 + 15NaOH → 6NaAlO2 + 3Na3PO4 + 12H2O NaOH recycling

Phosphorus precipitation: 6NaAlO2+3Na3PO4+8Ca(OH)2 → Ca5(PO4)3OH+3Ca(AlO2)2+15NaOH

The opportunity to recycle the base minimizes its irreversible consumption in the process to almost negligible level.

In some cases (e.g. when phosphorus is strongly bound to the metal ions) combination of different approaches is needed for P recovery. For example, acids together with high temperature and pressure are used in the KREPRO commercial system. This system is already operating almost in full-scale plant in the city of Helsingborg. Another system, BioCon, is based on drying, incineration and acid leaching of produced ash. The recovery of the product is realized by using the system of ion exchangers. It was planned to be built in two Swedish cities: Falun and Mora, but then the project was cancelled (Stark, 2002; Levlin et al., 2004).

In the papers written by Kuroda et al. (2002) and Takiguchi et al. (2004) heating was studied as a method for phosphorus recovery. It was found that at pilot plant running EBPR process heating of activated sludge is a promising method for P recovery. Calcium chloride (CaCl2) was added in order to precipitate phosphorus.

Recently, recovery of phosphorus through struvite (MgNH4PO4∙6H2O) crystallization become of great interest (Jaffer et al., 2002). This method was found to be very effective in phosphorus recovery: about 80% to 90% efficiency (Adnan et al., 2003).

The study (Lesjean et al., 2003) was held in order to examine possible potential of P recovery by existing methods after system improvements and upgrades. Comparisons between different phosphorus recovery methods are also of great interest (Hasselström, 2001). This is an evidence for increased attention to phosphorus recovery and recycling nowadays.

2.6.

Adsorption of phosphorus using magnetite

As it was mentioned above, current methods of phosphorus removal have significant drawbacks. An alternative solution for P adsorption suggested in this work is utilization magnetite.

2.6.1. Magnetite Magnetite is the oldest known magnetic material (Wilke et al., 2009) with the highest magnetic properties of all natural minerals on Earth (Harrison et al., 2002). Its chemical formula is Fe3O4, but can be also written as FeO·Fe2O3. The last variant shows that iron in magnetite is present in two different oxidation states, FeO (wüstite) and Fe2O3 (hematite).

13

The actual composite of natural magnetite varies over the regions and formation conditions.

Magnetite is widespread in the nature despite the fact that it is thermodynamically unstable in comparison to hematite. Magnetite can be found in rocks, soils and clays, both natural and anthropogenic aerosols in the atmosphere, bacteria, insects, and even on the Mars surface (Cornell & Schwertmann, 1996). Sweden is rich on magnetite that could be found all over the country. 22 out of 25 provinces of Sweden have magnetite deposits (except Bohuslän, Halland and Öland) (Mindat, 2010).

In the temperatures around 25°C magnetite is oxidizing very slowly to maghemite (c-Fe2O3). If temperature increases the oxidation goes till hematite (Tang et al., 2002).

Magnetite is used in different fields of science: biology, medicine, physics, and material science (Shourong et al., 2005). Recently, also some applications of magnetite for wastewater treatment have been studied.

2.6.2. Magnetite in wastewater treatment Numerous studies have been done in order to investigate the possible magnetite application in wastewater treatment to remove different pollutants.

Magnetite is often used for heavy metals removal. Paper written by Navratil (2009) describes the application of magnetite and ferrites for mine water treatment and remediation. The results show that magnetite has some advantages for wastewater treatment in general and, in particular, for mine water treatment. It was found to be “an improved, simple, and an inexpensive method for mine water treatment and remediation” (Navratil, 2009).

In the study made by Oskay (2003) removing of some metals (mainly zinc and nickel) from wastewater of metal finishing industries and synthetic wastewater using magnetite has been investigated. It was shown that such factors as dosage of magnetite, its particle size, contact time with wastewater and pH of the solution have effect on the method efficiency.

In general, application of magnetite was found to be promising for Zn and Ni removal from wastewater of metal finishing industries. Studies about utilization of magnetite for removal of hexavalent chromium [Cr(VI)] (Yuan et al., 2010), mercury (Girginova et al., 2010) and other heavy metals also showed good results.

Magnetite can also be used for arsenic removal from wastewater (Mayo et al., 2007). In his work Mayo is mainly focused on the impact of particles size of magnetite on the efficiency of As(III) and As(V) removal. It was found that efficiency of the method increases dramatically at particle decrease to nanosize. The paper states that magnetite and magnetic separation has big potential for arsenic removal.

Several researchers like Lorenc and Hyde (1973) and more recently Moniwa et al. (2010) got the US patents for the development and introduction of the methods of oil removal from wastewater. In general, the methods of oil removal are based on the following quite simple method: the magnetite particles are distributed in the solution, adsorb the oil substances and then removed from the wastewater together with adsorbed oil.

14

Magnetite can also be used for color and turbidity removal (Anderson et al., 1982), as well as bacteria cells removal (Mac Rae & Evans, 1984). It can also be used as the carrier for microorganisms to remove, e.g. pesticides (Mac Rae, 1985) or other pollutants from wastewater (Wong & Fung, 1997).

Particles of magnetite are also used for phosphorus removal as an additional to coagulants agent to facilitate separation of solid and liquid phases (Huang et al., 2004).

This method could also be used for other non-magnetic substances like suspended solids, organic substances, algae and viruses (Ying et al., 2000). In case of phosphorus removal the ordinary coagulants for chemical precipitation are used with subsequent or simultaneous adding of magnetite (Booker et al., 1991). This method has improvement compared to common precipitation: less time needed for coagulant settlement, better extraction of settled particles and therefore smaller volume of tanks needed. On the other hand it has all other drawbacks that are peculiar to the chemical precipitation.

Some decades ago magnetite usage for phosphorus removal was already studied (Hultman, 1975). At that time the method didn’t find its application in wastewater treatment.

Magnetite for wastewater treatment has several advantages. Of course, its main advantage is possibility of using magnetic field in order to separate magnetite from the water. This results in shorter sedimentation time, and thus, in size reduction and space saving. There is no sludge production during magnetite usage for wastewater purification, it could be easily renovated and recycled back into the process. The other positive thing is that magnetite can be synthesized directly at the WWT facility, so no additional transportation is needed.

2.6.3. Methods of magnetite synthesis Magnetite particles can be synthesized in different ways: (a) by co-precipitation of iron (II) and iron (III) aqueous salts solution in alkaline medium (Booker et al., 1991; Lee at al., 1996; and Kang et al., 1996); (b) by thermal decomposition of Fe-complex (Woo et al., 2004); and (c) using sonochemical approach (Suslick et al., 1996).

Because of strong magnetic dipole, magnetite particles tend to aggregate. Polymeric compounds with special functional groups or surfactants can be added into solution in order to stabilize the solution (Wormuth, 2001; and Harris et al., 2003). Synthesis of sizecontrolled magnetite particles are also of great interest today due to changing physical and chemical parameters with size decreasing and wide application perspectives of magnetite nanoparticles (Harris et al., 2003; Liu et al., 2004; Rabelo et al., 2001; and Mayo et al., 2007).

Another issue that was taken into account during magnetite synthesis is that magnetic properties of magnetite depend on the ratio of iron ions concentrations. This ratio is called coefficient K. =

(

(

)

)

15

where C (Fe II) and C (Fe III) are concentrations of correspondingly bivalent and trivalent iron ions in magnetite.

Magnetic properties of magnetite are shown in the diapason of K=0.2–3 (i.e. from 1:5 till 3:1 ratio of Fe2+:Fe3+) (Booker et al., 1991). The maximum is observed at K=0.5 (Fe(II)/Fe(III) ratio is 1:2, as in natural magnetite). Along with coefficient K, pH and ionic strength also affect the chemical composition of the magnetite particle surface (Li et al., 2005).

16

3. Experiments description As the aims of this work were to study the opportunities of removing P using magnetite, phosphorus recovering, as well as examine the influence of magnetite structure on the removal and recovery efficiency, some experiments were done.

All of the experiments were made at the research facility Hammarby Sjöstadsverk that is located on top of Henriksdal WWTP that belongs to the Stockholm Water company. The facility itself is owned and operated by IVL Swedish Environmental research Institute and the Royal Institute of Technology (KTH) (Hammarby Sjöstadsverk, 2010).

3.1.

Materials

Chemicals that were used in experimental part of this work are presented in Table 1.

Table 1. Chemicals used in experiments.

1.

2.

3.

4.

Name Sodium hydroxide pellets Iron(II) sulfate heptahydrate Iron(III) chloride hexahydrate di-Sodium hydrogen phosphate dihydrate

Chemical formula NaOH FeSO4∙7H2O FeCl3∙6H2O Na2HPO4∙2H2O

Assay ≥ 99 % 99.5– 102.0 % 99.0–102.0 % ≥ 99.5 %

All chemicals mentioned in Table 1 were ordered from VWR International.

All apparatus, glassware, additional materials (e.g. filters etc.) were either present at the Sjöstadsverket laboratory or were provided by Industrial Ecology department, KTH. If nothing else mentioned, filters Munktell #5 (Ø55 mm, typical pore size >20 μm) have been used.

3.2.

Experiments

Several sets of experiments were done in order to see how magnetite reacts with phosphorus. Factors affecting the phosphorus removal and recovery were studied.

3.2.1. Phosphorus stock solution preparation Stock solution is the solution that is the basis for preparation of all working solutions used in the experiments. Its concentration is chosen for easy preparation of working solutions and couldn’t be lower than maximum concentration needed.

During the study, stock solution with the phosphorus concentration of 400 mg/L was prepared as following. Weighed sample of phosphate salt (Na2HPO4∙2H2O) with mass of 2.2968 g was dissolved in small amount of distilled water (about 100 ml). Then solution was transferred quantitatively into volumetric flask of 1000 ml and diluted with distilled water to specified volume.

3.2.2. Measuring of phosphorus concentration. Dilution Concentrations of phosphorus in all experiments were checked using HACH LANGE cuvette tests (Figure 7). Two types of cuvettes that have different measuring ranges were used. For high concentration of phosphorus tests with 0.5-5 mg/L range of total phosphorus were 17

used. For trace concentration the 0.05-1.5 mg/L range of total P was used. Measurements were performed using Dr. Lange spectrophotometer. The accuracy of the method is three significant digits (i.e. result are shown e.g 1.23 mg/L or 0.123 mg/L).

Figure 7. HACH LANGE cuvette tests for phosphorus trace concentration (range of 0.05-1.5 mg/L of total P).

Before measuring, sample was diluted if needed. Dilution rate was calculated based on approximate expected concentration (or results from previous similar experiments) and chosen tests range. For experiments with starting concentration 100 mgP/L the dilution rate was usually 62.5. For 0.05-1.5 mg/L range cuvettes it results in 3.125–93.75 mg/L of final range that covers almost all expected values. In case of high probability that final result will be in a specific narrow diapason, the smaller dilution rate was taken. This strategy allowed minimizing the number of out-of-range measurements and thus the total number of tests used.

Two methods were used for dilution: one-step or two-step dilution. The one-step dilution was used for smaller dilution rates (less than 100 times). The necessary volume of sample (Vsample) was taken using automatic pipettes and transferred into a volumetric flask of certain volume (Vflask) and then diluted with distilled water to specified volume. The dilution rate (DR) was calculated as =

.

Two-step dilution was used for high dilution rate (100 times and more). It consists of dividing the dilution process in two stages that are, in fact, two sequential one-step dilutions. For example, dilution in 10 times and then dilution in 15 times results in 150 times dilution.

After measuring all samples the result was multiplied on the dilution rate to obtain the actual concentration of phosphorus in the probe: Cactual = Cmeas.∙ DR 18

3.2.3. Magnetite synthesis Magnetite was synthesized using standard method of chemical co-precipitation of Fe (II) salt (iron sulfate heptahydrate, FeSO4∙7H2O) and Fe (III) salt (iron chloride hexahydrate, FeCl3∙6H2O) in the presence of base (sodium hydroxide, NaOH).

In order to prepare magnetite with different Fe(II)/Fe(III) ratio, amounts of Fe2+ and Fe3+ presented in Table 2 were used (amount corresponds to 100 ml of obtained magnetite suspension).

Table 2: mass of Fe(II) and Fe(III) and mass of FeSO4∙7H2O and FeCl3∙6H2O salts for magnetite synthesis with different Fe(II)/Fe(III) ratio (coefficient K).

Mass, g

Mass, g

Fe (II)/Fe(III) ratio

K

1:5

0.2

0.333

1.667

1.655

8.036

1:2

0.5

0.667

1.333

3.310

6.429

1:1

1.0

1.000

1.000

4.964

4.821

2:1

2.0

1.333

0.667

6.619

3.214

3:1

3.0

1.500

0.500

7.446

2.411

Fe2+

Fe3+

FeSO4∙7H2O FeCl3∙6H2O

Fe(II) and Fe(III) salts in amounts as stated in Table 2 were dissolved separately in small amount of distilled water. After dilution solutions of iron salts were quantitatively transferred to one 100 ml volumetric flask, mixed and diluted with distilled water to fixed volume (100 ml) (Figure 8). Then the mixture is transferred to chemical beaker. During constant stirring on the magnetic stirrer at 600 rpm pH level 9.5–10.0 was achieved by sharp adding of 20%-solution of sodium hydroxide (NaOH). Solution color was immediately changed from transparent deep orange to opaque dark black indicating the formation of magnetite particles. After stirring for 10 min magnetite solution was kept settling for 15 min and then washed three-four times with the distilled water to neutral pH. Separation of magnetite from liquid was done by decantation. Finally, magnetite solution was settled to the volume of 100 ml.

Figure 8. Fe3+ (on the left) and Fe2+ salts dissolved in water.

19

In case of small experiments that need less amount of magnetite, the same methodology with correspondingly fewer amounts of salts and final volume was used. Just newly synthesized magnetite of no more than 3 days age was used in experiments. If nothing else stated, magnetite with K=0.5 (i.e. Fe(II)/Fe(III) ratio is 1:2) is used.

3.2.4. Magnetite concentration Because of complex chemical reactions and also conditions of synthesis of magnetite (e.g. losing some magnetite during washing stage) it is impossible to calculate accurate concentration of magnetite in the solution. Therefore method of weighing of solid residue was used.

Filters were dried until constant mass at the temperature 105oC and weighed on the electronic wages (mfilter). Measured volumes (5 ml or 10 ml) of stirred up magnetite solution (Vm) were put in the coned filter (Figure 9) and dried at 105oC. Next day, after stabilizing of the mass, filters with dried magnetite were weighed (Figure 10). The difference between mass of filter with magnetite (mf+m) and mass of filter only (mfilter) shows the actual mass of magnetite in measured sample. Ratio between this mass and corresponding volume of sample gives the concentration of magnetite in solution (Cm): =

−

Figure 9. Magnetite with different K ratio (from right to left: 3.0, 1.0, 2.0, 0.5, and 0.2) in the coned filters before drying.

K=0.2

K=0.5

K=1.0

K=2.0

K=3.0

Figure 10. Magnetite with different K ratio after drying.

20

3.2.5. Effect of different Fe(II)/Fe(III) ratio in magnetite on P removal efficiency Magnetite solutions with coefficients K 0.2, 0.5, 1.0, 2.0, and 3.0.were prepared; 5 ml of each of them were added to separate beakers with 50 ml of 110 mg/L solution of P stirring the magnetic stirrer at 600 rpm. After 20 min of mixing, beakers were put on the magnets.

After clarification of the upper layer samples were taken, diluted and P concentration was measured.

3.2.6. Adsorption kinetics Changes of P concentrations with time were studied. In 500 ml sample of 110 mg/L solution of phosphorus, 50 ml of magnetite was added during stirring on the magnetic stirrer at 600 rpm. 2 ml samples were taken after the following time intervals: 2.5 min, 5 min, 10 min, 15 min, 20 min, 30 min, and 60 min. Samples were filtered to remove magnetite particles and then diluted and measured using cuvette tests.

3.2.7. Adsorption isotherm of phosphorus on magnetite Equilibrium isotherms were obtained as follows. A number of solutions with different known P concentrations (10 mg/L, 50 mg/L, 100 mg/L, 200 mg/L, 300 mg/L, and 400 mg/L) and same volume (100 ml) was made and transferred to the separate chemical beakers. During constant mixing on the magnetic stirrers at 500 rpm, 50 ml of magnetite was added to each sample. After stirring for 20 min the beakers were placed for 5 min on the magnets in order to settle magnetite particles. Samples were taken from the upper clear part of the solution then diluted and equilibrium concentrations were measured using cuvette tests.

3.2.8. Amount of base needed for phosphorus recovery from magnetite Phosphorus recovery was made using base leaching method. 20%-solution of NaOH was used as a base.

In the beakers containing 25 ml of 110 mg/L phosphorus solution was added 2.5 ml of magnetite. Mixture was stirred at 600 rpm for 20 min and then filtered using a vacuum filter. In order to find the residual concentration of phosphorus after water treatment, samples from the filtrate were taken. Magnetite that stayed on the filter was washed three times with distilled water in order to remove not-absorbed phosphorus. 2.5 ml, 5 ml, 10 ml, 15 ml, 20 ml, and 30 ml of 20%-solution of NaOH were added to magnetite and mixed together for 10 min. After magnetite settling, samples were taken from the upper clear part of the solution. After dilution of samples, concentrations of phosphorus that was desorbed from magnetite to the solution were measured.

3.2.9. Recovery of phosphorus from magnetite Phosphorus recovery was made in two ways: using one-step base leaching and using twostep base leaching. 20%-solution of NaOH was used as a base.

5 ml of magnetites with different K (0.2, 0.5, 1.0, 2.0, and 3.0) were added to 25 ml of 120 mg/L phosphorus solution during stirring at 600 rpm. After 20 min the mixtures were filtrated using vacuum filter. Samples from the filtrate were taken to find the residual P concentration. Kept on filter magnetite was washed three times with distilled water.

21

In one-step base leaching 10 ml of 20%-solution of NaOH was added to washed magnetite, stirred for 10 min and then kept settling on magnets. Samples from liquid part were taken to measure the phosphorus amount that was discharged from magnetite. One-step process finished at this moment and two-step process run further as follows: magnetite was separated from NaOH by decantation and washed three times with distilled water; after that the same phosphorus desorption procedures as described in one-step process was done.

Results for recovery efficiency were calculated based on the amount of phosphorus adsorbed by magnetite (taken as 100%). This means that for two processes with the same recovery rate but different removal rate, actual amount of phosphorus recovered will be different. For example, 1.0 g of phosphorus entered the treatment and was adsorbed with 40% removal efficiency by magnetite. Magnetite now contains 0.4 g of P that is taken as 100%. After leaching, 0.3 g of P was desorbed from magnetite to liquid phase using base.

Efficiency of recovering in this case is 0.3 g/0.4 g = 0.75 or 75%. Actual amount of phosphorus recovered is 0.3 g or 0.3 g/1.0 g = 0.3 or 30%.

3.2.10. Using recovered magnetite for phosphorus removal Experiments on how the recovered magnetite adsorbs phosphorus were done as follows.

After phosphorus recovery from magnetite as described in previous section, magnetite was added to 25 ml of 120 mg/L phosphorus solution, stirred for 20 min at 600 rpm and filtrated on vacuum filter. Samples from filtrate were taken to measure the residual P concentration. Magnetite was washed and then two-step base leaching using 10 ml of 20%solution of NaOH on each step as described above was done. Recovered in the second time magnetite was again added to a new phosphorus solution (120 mgP/L, 25 ml), stirred (20 min, 600 rpm) and filtrated. Samples from filtrate were taken to control P content.

Results of experiments include three residue P concentrations: one after using fresh magnetite, and two after using recovered first and second time magnetite.

22

4. Results 4.1.

Magnetite concentration

As was mentioned above the concentration of magnetite couldn’t be calculated through the chemical reactions and thus has to be measured. It was found that for chosen method of magnetite synthesis the concentration of different form of magnetite is varying in a range between 20 mg/ml and 25 mg/ml. Just magnetite ratio of 0.2, 0.5, 1.0, 2.0, and 3.0 that are used in following experiments were studied. Results of measuring magnetite concentration are shown in the Figure 11.

Magnetite concentration 30 25

mg/ml

20 15 10 5 0 0,0

0,5

1,0

1,5

2,0

2,5

3,0

Fe(II) / Fe(III) ratio (coeficient K)

Figure 11. Magnetite concentration for magnetite with different Fe(II)/Fe(III) ratio

For simplification of calculations for other experiments’ results the average concentration of 22.5 mg/ml was assumed to be the same for all types of magnetite.

4.2.

Effect of different Fe(II)/Fe(III) ratio in magnetite on P removal efficiency

One of the main fields of interest of current work was how the different Fe(II)/Fe(III) ratio in magnetite effects the phosphorus removal rate.

It was found that indeed different magnetite structure does have an impact on the phosphorus adsorption. Three independent sets of experiments (A, B and C) with the same input characteristics showed the following correlation: efficiency of P removal decreases with increasing of coefficient K. So, the maximum was observed in Fe(II):Fe(III) as 1:5 and the minimum in the 3:1 ratio (see Figure 12).

23

Effect of different K on P removal

Removal efficiency, %

60 50 40

Average

30

Exp. A

20

Exp. B

10

Exp. C

0 0,0

0,5

1,0

1,5 2,0 Fe(II) / Fe(III) ratio

2,5

3,0

3,5

Figure 12. Effect of magnetite composition (Fe(II)/Fe(III) ratio) on the phosphorus removal rate

It is also worth to mention that the magnetites with K=0.2 and K=0.5 have the minimum time of magnetite sedimentation and purification of upper liquid layer. Magnetites with K=1.0 and, especially, K=2.0 and K=3.0 have very long time of upper layer purification.

Taking into account the sedimentation time and also the fact that the natural magnetite has K=0.5 it was decided to choose this Fe(II)/Fe(III) ratio as the base for all further experiments.

4.3.

Adsorption kinetics

Removal efficiency, %

Experiments on the adsorption curve have shown that adsorption time has small impact on the P removal efficiency. The efficiency increases rapidly before the moment of time 10 min. After that just small variations were observed. After mixing more than 30 min small decreasing in P removal efficiency was observed. It could be partly explained by desorption of small amounts of phosphorus from magnetite. Adsorption curve is presented on the Figure 13.

Adsorption curve

42 41 40 39 38 37 36 35 34 33 32 31 30 0

5

10

15

20

25 30 35 40 Adsorption time, min

45

50

55

60

Figure 13. Adsorption of phosphorus curve by magnetite over time

24

4.4.

Adsorption isotherm of phosphorus on magnetite

It was important to see how magnetite adsorbs phosphorus in general and evaluate its capacity. Figure 14 illustrates the dependence of P removal efficiency on the initial concentration of phosphorus solution. For magnetite capacity adsorption isotherm was built (Figure 15).

Efficiency of P removal

100 90

70 60 50 40 30 20 10 0 0

50

100

150

200

250

300

250

300

Input P concentration, mg/L

Figure 14. Efficiency of P removal

Adsorption isotherm 9 8 7 Loading, mgP/gM

Removal efficiency, %

80

6 5 4 3 2 1 0 0

50

100

150

200

Phosphorus equilibrium concentration

Figure 15. Adsorption isotherm

25

It is obvious that the efficiency of phosphorus removal will decrease with increased concentration because the same amount of magnetite added has to adsorb larger amount of phosphorus in the solution. The efficiency of magnetite at low concentrations are especially interesting. For concentration about 6.7 mg/L phosphorus removal efficiency reaches almost 99%; for concentration of 33.3 — up to 77% of phosphorus removal. Taking into consideration that for many WWTP the input P concentration is rarely exceed 20 mg/L the proposed method can be efficiently used for phosphorus removal at those WWTP.

4.5.

Amount of base needed for phosphorus recovery from magnetite

Solution of 20%-NaOH was used for phosphorus recovery from magnetite. It was found that recovery efficiency comes on plateau after 10–15 ml of NaOH added at the level of about 43%. In spite of the fact that small increasing in efficiency observed with more amount of NaOH added, addition of more than 15 ml of base is inexpedient and non-valueadded on the real plant recovery process.

The results of experiments are shown of the Figure 16.

Amount of NaOH needed for P recovery from magnetite 60

Recovery efficiency %

50 40 30 20 10 0 2,5

5

10

15

20

30

V (NaOH), ml

Figure 16. Phosphorus removal efficiency from magnetite using different amounts of base.

4.6.

Recovery of phosphorus from magnetite

The study showed that phosphorus recovery from magnetite has the same dependence from Fe(II)/Fe(III) ratio as P removal had. Both one-step and two-step base leaching decreased efficiency during Fe(II)/Fe(III) ratio increasing (see Figure 17).

26

P removal and recovery efficiency 70 60 P removal rate

Efficiency, %

50 P recovery (one-step leaching or step 1 of two-step leaching)

40 30

P recovery (step 2 of two-step leaching)

20 Total recovery of two-step leaching

10 0 0,0

0,5

1,0

1,5

2,0

2,5

3,0

Fe(II) / Fe(III) ratio

Figure 17. Phosphorus removal and recovery in one-step and two-step base leaching.

Removal rate is shown for evaluating the actual amount of phosphorus adsorbed by magnetite. As was described above this amount is taken as 100% for measuring recovery efficiency. Results had shown maximum of about 50% during one-step base leaching (i.e.

50% of phosphorus adsorbed by magnetite was recovered) for magnetite with Fe(II)/Fe(III) ratio equal to 0.2. About 23% desorption rate was obtained during second step of two-step leaching resulting in more than 60% of total recovery efficiency using twostep base leaching. Detailed information about actual amounts of phosphorus that were adsorbed by magnetite and then desorbed in one-step or two-step leaching process is presented in Table 3.

Table 3. Amounts of P removed and recovered from wastewater

K

0,2 0,5 1 2 3

Input P amount, mg 3000 3000 3000 3000 3000

P removed, mg 1269,0 1146,0 903,0 792,0 660,0

Phosphorus recovering, mg One-step Total recovery leaching or step Step 2 of twoof two-step 1 of two-step step leaching leaching leaching 633,3 146,5 779,8 531,5 96,3 627,8 257,5 58,8 316,3 164,0 26,1 190,1 126,1 14,7 140,8

27

4.7.

Using recovered magnetite for phosphorus removal

Experiments with recovered magnetite showed that efficiency of removing of phosphorus from solution decreased dramatically after first recovery with two-step base leaching (see Figure 18). There are almost no changes in removal efficiency after second recovery of magnetite.

Removal efficiency of recovered magnetite 45 40

Removal efficiency, %

35 30 25 P removal using fresh magnetite

20

P removal after first recovery

15

P removal after second recovery

10 5 0 K=0.2

K=0.5

K=1.0

K=2.0

K=3.0

Fe(II) / Fe(III) ratio

Figure 18. Phosphorus removal efficiency of fresh, once and twice recovered magnetite.

The possible explanation for such decreasing in removal process could be changes that happen in upper layer of magnetite under NaOH exposure during base leaching process. An additional stage to refresh magnetite is probably needed.

28

5. Discussion 5.1.

Results’ discussion

The first thing that was found after conducting the experiments is that magnetite can be in fact used for phosphorus removal and recovery from wastewater. Efficiency of removal process is high enough (65% and higher) at low concentrations (lower than 30 mgP/L).

This makes application of magnetite a good option for P removal from low-concentrated fluid or as the polishing method.

Synthesis of magnetite is relatively easy process and could be automated on the WWTP if needed. Still, specific requirements should be met in order to synthesize magnetite with required properties.

It was also found that contact time of magnetite with phosphorus and settling time is relatively small compared to other methods. This allows making smaller reaction and sedimentation tanks thus saving space at the facility.

Experiments with different Fe(II)/Fe(III) ratio (coefficient K) showed that the best removal as well as recovery rate is reached at the K=0.2. Nevertheless, the natural composition of magnetite with K=0.5 showed also good results and could be used for P removal and recovery from wastewater. It is very strong advantage of the method as natural magnetite is widespread mineral and it is relatively cheap.

After conducting the experiments connecting to the phosphorus recovery is was found that P desorption is taking place with the efficiency of about 40%. For recycling of magnetite additional renovation stage is needed. Amount of base (NaOH) is moderate low and it is also possible to be recycled.

Comparing one- or two-step base leaching process, the two-step leaching has better P-yield but is more complicated and requires more chemicals. So in case the cost of the method is more significant than P recovery efficiency, the one-step base leaching could be advised.

It is also important to keep in mind current specific properties of wastewater and its phosphorus content in order to decide the actual parameters of P removal and recycling.

Results of this work can be used for finding the general correlation between removal and recovery of phosphorus and different factors that are affecting these processes.

5.1.1. Results’ accuracy There are several sources of errors, observed during the experiments: — Method error.

— Measuring equipment errors (e.g. balance error during sample weighting).

— Dilution error, e.g. mistake in volume added.

— Miscalculation.

— Other types of errors (e.g. human error).

29

Measuring and dilution errors took place every time during weighting or dilution of the probe, but their influence on the results is less than 0.05% for weighting, no more than 0.5% for phosphorus concentration measurements using HACH Lange tests and less than 0.5% for dilution of solutions. Miscalculation and other types of errors have quite small probability and could be neglected.

In general, the reliability of the results is rather high. This can also be explained by the fact that almost every experiment has 2-3 parallel measurements for each point.

5.2.

Future study possibilities

There are several proposals for the future investigation.

Firstly, the method of magnetite renovation for its recycling should be found. One of the suggestions that could help to renovate magnetite is to add iron salts during or after the base leaching process. It will be also interesting what amount of magnetite could not be recycled and thus should be substituted with the new one.

Secondly, magnetite seems to remove not just phosphorus but also other compounds from wastewater, both inorganic and organic. Experiments studying of sorption affinity of magnetite to a range of contaminants, i.e. selectivity of adsorption of different pollutants from the mixture, are also of great interest. Real wastewater could be used for this kind of study.

Thirdly, that partly appears from previous point, is that the pilot-scale plant study should be done to examine magnetite in real conditions.

During presented work it was found that several factors have influence the magnetite structure and thus P removal by magnetite and recovery efficiency, but no further study had been done. Factors like temperature, pH, stirring and settling time and some other seemed to have effect on the magnetite properties. Detailed investigation studying what factors and how exactly affect magnetite is necessary in order to develop the method.

It is also interesting to study the other possibility for P recovery aside base leaching that was used in this work.

30

6. Conclusion After performing the study, it could be concluded that magnetite has a great potential for phosphorus removal from wastewater. It has the highest efficiency as a polishing method for P removal, as the removal rate of the method is quite high at the low phosphorus concentrations. Taking into account the efficiency of the method at higher concentrations, several steps process may be needed for higher phosphorus removal rate.

Purification stage using magnetite could be introduced in already existing WWTP, replacing older technologies (e.g. sand filter). This makes upgrading of old WWTP easier, cheaper and more efficient.

As was mentioned above, magnetite purification reduces reaction and settling time, and thus saves space on the facility. The method has relatively low energy consumption; the operational and investment costs are also low. This makes described method economically feasible and profitable.

As magnetite loses its adsorption properties after phosphorus removal, renovation of magnetite is necessary for its recycling. The method of renovation is still need to be investigated.

31

32

References Adnan A., Mavinic D.S., Koch F.A. (2003) Pilot-scale study of phosphorus recovery through struvite crystallization – examining the process feasibility, J. Environ. Eng. Sci., 2:315-324.

Amirtharajah A., O’Mella C.R. (1990) Coagulation Processes: Destabilization, Mixing, and Flocculation, In: Pontius F.W. (ed.) Water Quality and Treatment, A Handbook of Community Water Supplies, AWWA 4th Ed. McGraw-Hill, Inc. NY, USA. Anderson N. J., Bolto B. A., Blesing N. V., Kolarik L. O., Priestley A. J., Raper W. G. C. (1982) Colour and turbidity removal with reusable magnetite particles—VI pilot plant operation, Water Research, 17, 10:1235-1243.

Better Crops (1999) World Production of Phosphate Rock, Better Crops, 83, 1.

Booker N.A., Keir D., Priestley A.J., Ritchie C.B., Sudarmana D.L., Woods M.A. (1991) Sewage clarification with magnetite particles, Water Sci. Tech., 23, 7-9:1703-1712.

Carpenter S.R., Caraco N.F., Smith V.H. (1998) Nonpoint pollution of surface waters with phosphorus and nitrogen, Ecological Applications, 8:559-568.

CEEP (1998) Phosphates, a sustainable future in recycling, Centre Européen d’Etudes sur les Polyphosphates, Dépôt légal D/1998/3158/15.

Cornell R. M., Schwertmann U. (1996) Iron Oxides Structure, Properties, Reactions, Occurrence and Uses, VCH, Weinheim, Germany.

DEFRA (2008) Consultation on options for controls on phosphates in domestic laundry cleaning products in England, The Department for Environment, Food and Rural Affairs, London, UK. Dirkzwager A.H., Eggers E., Ferdinandy-van Vierklen M.M.A. (1994) Developments in waste water technologies and municipal waste water treatment systems in the future, European Water Pollution Control, 4, 1:9-19.

Dryden F.O., Stern G. (1968) Full-scale cyclic activated sludge system phosphorus removal, Wat. Sci. Tech., 26 (9-11):2253-2256.

EC (2000) Directive 2000/60/EC of the European Parliament and of the Council establishing a framework for the Community action, Official Journal of the European Communities, L 327:1-72.

EPA (2000) Wastewater Technology Fact Sheet: Chemical Precipitation, United States Environmental Protection Agency, Office of Water, Washington, USA. Girginova P.I., Daniel-da-Silva A.L., Lopes C.B., Figueira P., Otero M., Amaral V.S., Pereira E., Trindade T. (2010) Silica coated magnetite particles for magnetic removal of Hg2+ from water, J Colloid and Interface Science, 345, 2:234-240.

Hammarby Sjöstadsverk assessed 2010-04-15.

(2010)

http://www.sjostadsverket.se/Welcome_en.html

33

Harris L. A., Goff J. D., Carmichael A. Y., Riffle J. S., Harburn J. J., Pierre T. G. St., Saunders M. (2003) Magnetite Nanoparticle Dispersions Stabilized with Triblock Copolymers, Chem.

Mater., 15:1367-1377.

Harrison R.J., Dunin-Borkowski R.E., Putnis A. (2002) Direct imaging of nanoscale magnetic interactions in minerals, PNAS, 99, 26: 16556–16561.

Hasselström J. (2001) Fosforatervinning ur slam: En jamforelse av tre olika processer, KTH, Stockholm, Sweden (in Swedish).

Henze M., Harremoes P., la Cour Jansen J., Arvin, E. (1997) Wastewater Treatment: Biological and Chemical Processes, 2nd ed., Springer, Berlin, Germany.

Hislop H. (ed.) (2007) The nutrient cycle: closing the loop, Seacourt, Oxford, UK. Huang Z., Hu Y., Xu J., Zheng C. (2004) Removal of phosphate from municipal sewage by high gradient magnetic separation, J Centr. South Univ. of Techn., 11, 4: 391-394.

Hultman B., Levlin E., Stark K. (2001) Phosphorus recovery from sewage sludges: Research and experiences in Nordic countries, Scope Newsletter, Scientific committee on Phosphates in Europe, 41.

Jaffer Y., Clark T. A., Pearce P., Parsons S. A. (2002) Potential phosphorus recovery by struvite formation, Water Research, 36:1834-1842.

Jancangelo J.G., Demarco J., Owen D.M., Randtke S.J. (1987) Selected Processes for Removing NOM: An Overview, Journal of the AWWA, 87(1):64 - 77.

Jenssen P.D., Etnier C. (1996) Ecological engineering for wastewater and organic waste treatment in urban areas: An overview, Material of conference "Water saving strategies in urban renewal", European Academy of the Urban Environment, Wien, Austria.

Jiang F., Beck M.B., Cummings R.G., Rowles K., Russell D. (2004) Estimation of Costs of Phosphorus Removal in Wastewater Treatment Facilities: Construction De Novo, Water Policy Working Paper #2004-010.

Kang Y. S., Risbud S., Rabolt J. F., Stroeve P. (1996) Synthesis and Characterization of Nanometer-Size Fe3O4 and γ-Fe2O3 Particles, Chem. Mater., 8:2209–2211.

Kuile M., White R.E., Beckett P.H.T. (1983) The availability to plants of phosphate in sludges recipitated from the effluents from sewage treatment, Water Pollut. Control, 82:582-589.

Kuroda, A., Takiguchi, N., Gotanda, T., Nomura, K., Kato, J., Ikeda, T., Ohtake, H. (2002) A simple method to release polyphosphate from activated sludge for phosphorus reuse and recycling, Biotechnol. Bioeng., 78:333–338.

Lee J., Isobe T., Senna M. (1996) Preparation of Ultrafine Fe3O4 Particles by Precipitation in the Presence of PVA at High pH, J Colloid and Interface Science, 177:490–494.

Lesjean B., Gnirss R., Adam C., Kraume M., Luck F. (2003) Enhanced biological phosphorus removal process implemented in membrane bioreactors to improve phosphorous recovery and recycling, Water Sci. and Techn., 48, 1:87–94.

34

Levlin E., Hultman B. (2005) Phosphorus recovery from sewage sludge — ideas for further studies to improve leaching, Joint Polish - Swedish Reports, Stockholm, Sweden.

Levlin E., Löwén M., Stark K. (2004) Phosphorus recovery from sludge incineration ash and supercritical water oxidation residues with use of acids and bases, In: Plaza E., Levlin E., Hultman B. (eds.), Integrated and optimisation of urban sanitation systems, Polish–Swedish seminar, Wisla, Poland, pp. 19–28.

Li Z., Sun Q., Gao M. (2005) Preparation of Water-Soluble Magnetite Nanocrystals from Hydrated Ferric Salts in 2-Pyrrolidone: Mechanism Leading to Fe3O4, Angew. Chem.

Int. Ed., 44:123 –126.

Liu Z.L., Wang X., Yao K.L., Du G.H., Lu Q.H., Ding Z.H., Tao J., Ning Q., Luo X.P., Tian D.Y., Xi D. (2004) Synthesis of magnetite nanoparticles in W/O microemulsion, J Materials Science, 39:2633 – 2636.

Loosdrecht M.C.M. van, Hooijmans C.M., Brdjanovic D., Heijnen J.J. (1997) Biological phosphate-removal processes, Appl. Microbiol. Biotechnol., 48:289–296.

Lorenc W.F., Hyde J.A. (1973) Oil removal from waste waters, United States Patent, #3,767,571.

Lowe E.F., Battoe L.E., Stites D.L., Coveney M.F. (1992) Particulate Phosphorus Removal via Wetland Filtration: An Examination of Potential for Hypertrophic Lake Restoration, Env. Manag., 16, 1:67-74.

Mac Rae I. C. (1985) Removal of pesticides in water by microbial cells adsorbed to magnetite, Water Research, 19, 7: 825-830.

Mac Rae I. C., Evans S.K. (1984) Removal of bacteria from water by adsorption to magnetite, Water Research, 18, 11: 1377-1380.

Mahmut O., Ayhan S. (2003) Effect of tannins on phosphate removal using alum, Turkish J. Eng. Environ. Sci., 27:227-236.

Mayo J.T., Yavuz C., Yean S., Cong L., Shipley H., Yu W., Falkner J., Kan A., Tomson M., Colvin V.L. (2007) The effect of nanocrystalline magnetite size on arsenic removal, Science and Technology of Advanced Materials, 8:71–75.

Metcalf and Eddy (1991) Wastewater Engineering: Treatment, Disposal, and Reuse, 3rd ed., McGraw-Hill, Inc., Montreal, USA. Mindat (2010) Mindat.org — the mineral and locality database http://www.mindat.org/show.php?id=2538&ld=1#themap assessed 2010-04-09.

Mino T., Liu W.T., Kurisu F., Matsuo T. (1995) Modeling glycogenstorage and denitrification capability of microorganisms in enhanced biological phosphate removal processes, Water Sci. Technol., 31 (2):25–34.

Mino T., Loosdrecht M.C.M., Heijnen J.J. (1998) Microbiology and biochemistry of the enhanced biological phosphate removal process, Water Res., 32 (11):3193–3207.

35

Möller G. (2006) Absolute (1000 Fold) Phosphorus removal: performance, mechanisms and engineering analysis of iron-based reactive filtration and coupled CEPT at the Hayden, ID WWTP, Session P2 in WERF 2006.

Moniwa S., Shiire H., Ebihara S., Ashikaga N., Kiuchi T. (2010) Water Treatment System, United States Patent Application, 20100059444.

Navratil J.D. (2009) Mine Water Treatment Using Iron Ferrites and Magnetite, International Mine Water Conference, Pretoria, South Africa.

Neethling J.B., Bakke B., Benisch M., Gu A., Stephens H., Stensel H.D., Moore R. (2005) Factors Influencing the Reliability of Enhanced Biological Phosphorus Removal, Water Environment Research Foundation, Alexandria, USA. Neethling J.B., Gu A. (2006) Chemical phosphorus removal constraints – Introduction.

Session P2 in WERF. Oehmen A., Lemos P.C., Carvalho G., Yuan Z., Keller J., Blackall L.L., Reis M.A.M. (2007) Advances in enhanced biological phosphorus removal: From micro to macro scale, Water Research, 41:2271 – 2300.

Oskay E. (2003) Treatment of wastewater using magnetite, Izmir, Turkey.

Rabelo D., Lima E. C. D., Reis A. C., Nunes W. C., Novak M. A., Garg V. K., Oliveira A. C., Morais P. C. (2001) Preparation of Magnetite Nanoparticles in Mesoporous Copolymer Template, Nano Letters, 1, 2:105-108.

Reardon, R. (2006) Technical introduction of membrane separation processes for low TP limits, Session P3 in WERF, The Water Environment Research Foundation, Alexandria, USA. Rigler F.H. (1973) A Dynamic View of The Phosphorus Cycle in Lakes, In: Griffith E.J., Beeton A., Spencer J.M., Mitchell D.T. (eds), Environmental Phosphorus Handbook, John Wiley & Sons, Hoboken, USA. Rybicki S. (1997) Phosphorus Removal from Wastewater, Joint Polish - Swedish Reports, Stockholm, Sweden.

Sedlak R. I. (1991) Phosphorus and nitrogen removal from municipal wastewater, 2nd ed, Lewis Publishers, New York, USA. Shourong W., Junsheng H., Husheng Y., Keliang L. (2005) Size-controlled preparation of magnetite nanoparticles in the presence of graft copolymers, J. Mater. Chem., 16:298–303.

Song Y., Hahn H.H., Hoffmann E. (2002) Effects of solution conditions on the precipitation of phosphate for recovery: A thermodynamic evaluation, Chemosphere, 48:1029–1034.

Spellman F. R. (2008) Handbook of Water and Wastewater Treatment Plant Operations, 2nd ed., CRC Press, Boca Raton, USA. Stark K. (2002) Phosphorus release from sewage sludge by use of acids and bases, Licentiate thesis, Water Resources Engineering, KTH, TRITA, Stockholm, Sweden.

36

Starkenburg W. van, Rijs G.B.J. (1988) Phosphate in sewage and sewage treatment. Proc. of SCOPE Phosphorus Cycles Workshop, Poznan.

Steen I. (1998) Phosphorus Availability in the 21st Century: Management of a NonRenewable Resource, Phosphorus and Potassium, 217:25-31.

Strom P.F. (2006) Technologies to Remove Phosphorus from Wastewater, Rutgers University, USA. Suslick K.S., Fang M., Hyeon T. (1996) Sonochemical Synthesis of Iron Colloids, J. Am. Chem. Soc., 118 (47):11960–11961.

Takács, I. (2006) Modeling chemical phosphorus removal processes, Session P2 in WERF. Takiguchi N., Kishino M., Kuroda J.K., Ohtake H. (2004) A Laboratory-Scale Test of Anaerobic Digestion and Methane Production after Phosphorus Recovery from Waste Activated Sludge, Biosc. and Bioeng., 97, 6:365–368.

Tang J., Myers M., Bosnick K.A., Brus L.E. (2002) Magnetite Fe3O4 Nanocrystals: Spectroscopic Observation of Aqueous Oxidation Kinetics, J. Phys. Chem., 107:7501-7506.

Tchobanoglous G., Burto F.L. (1991) Wastewater Engineering, Treatment, Disposal and Reuse, Metcalf & Eddy, McGraw Hill International Editions, Civil Engineering Series, 3rd Edition, pp. 733.

Tchobanoglous G., Burton F.L., Stensel H.D. (2002) Wastewater Engineering: Treatment and Reuse, 4th ed., Metcalf & Eddy Inc., McGraw-Hill Science Engineering, NY, USA. UNEP (1994) The Pollution of Lakes and Reservoirs: UNEP Environment Library No. 12, United Nations Environment Programme: Nairobi.

Wilke M., Caliebe W.A., Machek P. (2009) Magnetite at low temperature: Resonant Inelastic X-ray Scattering (RIXS) at the Fe K-edge, J. of Physics: Conference Series 190:012090.

Wind T. (2007) The Role of Detergents in the Phosphate-Balance of European Surface Waters, Official Publication of the European Water Association (EWA).

Wong P.K., Fung K.Y. (1997) Removal and recovery of nickel ion (Ni2+) from aqueous solution by magnetite-immobilized cells of Enterobacter sp. 4-2, Enzyme and Microbial Technology, 20, 2:116-121.

Woo K., Hong J., Choi S., Lee H.W., Ahn J.P., Kim C.S., Lee S.W. (2004) Easy Synthesis and Magnetic Properties of Iron Oxide Nanoparticles, Chem. Mater.,16:2814-2818.

Woodard S. (2006) Magnetically enhanced coagulation for phosphorus removal, Session B2 in WERF. Woods N.C., Sock S.M., Daigger G.T. (1999) Phosphorus recovery technology modeling and feasibility evaluation for municipal wastewater treatment plants, Environmental Technology, 20 (7):663-679.

Wormuth K. (2001) Superparamagnetic Latex via Inverse Emulsion Polymerization, J Colloid and Interface Science, 241:366–377.

37

Ying T. Y., Yiacoumi S., Tsouris C. (2000) High-gradient magnetically seeded filtration, J Chemical Engineering Science, 55:1101-1113.

Yuan P., Liu D., Fan M., Yang D., Zhu R., Ge F., Zhu J., He H. (2010) Removal of hexavalent chromium [Cr(VI)] from aqueous solutions by the diatomite-supported/unsupported magnetite nanoparticles, J Hazard Mater, 173(1-3):614-21.

38