2008:058
M A S T E R ' S T H E S I S
Iron Removal at the
Drinking Water Treatment Plant of Luleå
Ahmed Fahim Faisal
Luleå University of Technology Master Thesis, Continuation Courses
Environmental Engineering
Department of Civil and Environmental Engineering
PREFACE
This work was carried out from October 2007 to June 2008. The work is targeted to find out a suitable solution to activate the drinking water well 21, at the drinking water plant in Luleå which is severely contaminated by iron. The well had never been connected to the main distribution plant due to this serious contamination by iron. There are a number of investigations to add the well in distribution system, mentioned in the report.
Unfortunately those were not successful. Hence the recent study is performed to find a suitable way to get rid of iron from drinking water. In this regard several samples have been collected in different times and analyzed in the laboratory. The results are analyzed and discussed in this report.
The 8 months study is mandatory to achieve Masters Degree. The work was arranged by Luleå University of Technology and Technical office of Luleå Kommun. The work was performed at the Water Treatment Plant in Luleå. Several travels have been made to the plant to examine the samples collected of different times.
I place special thanks to my supervisor Professor Jörgen Hanaeus, Division of Sanitary Engineering for his continuous guidance and instruction for the work. I am undoubtedly grateful to him for solving my thousands of doubts. I am also thankful to all the operators at the plant Lars-Olof Eriksson, Börje Lantto and Patrik Fahlen for their time-to-time help and all necessary background information I need for composing the report. I specially thank Patrik Fahlen for his willful support in establishing the pilot filter and providing all the necessary information about the well as well as the water quality without which the work is incomplete. On this occasion I would also like to thank my wife Ashfia for her encouragement and support to perform the work.
ABSTRACT
The study is aimed to find a suitable way to treat the iron contaminated well 21 in the Luleå water Treatment Plant in Gäddvik, Luleå. Rapid sand filtration was established to treat water pumped from the well. A pilot filter was constructed in October, 2007. The filter was run continuously and the filtrate was examined in the laboratory in different interval. The 8 months long experiment reveals the efficiency of the filter in removing iron. The period runs from the beginning of spring, winter and the starting of summer.
There is a large difference in surface water temperature during this period.
There are several wells in the plant contributing to the supply system. Most of them are free from iron problem. Unfortunately the well 21 is contaminated with iron and has never been contributing to the main plant due to severe iron contamination. The probable reason for this contamination is also discussed in this study depending on existing data from previous surveys.
The performance of the filter in removing iron is satisfactory throughout the run.
However sometimes even the pumped raw water had low concentration of iron. But even though the iron concentration is very high in the raw water the filter was capable of removing 95% of it. The filter was run at different flow rates facilitating different retention times. In all cases the iron removal was quite successful. The filtrate had iron concentration lower than the standard value fixed by EPA allowable in drinking water.
The plant has been suffering from turbidity problem since long time. The turbidity values from some wells are higher than the standard value (0.5 NTU). In this experiment the same problem occurred with turbidity. The filtrate water turbidity is well higher than the allowable limit. Different options like pH adjustment or aeration or the combination of both was applied to reduce turbidity although those were not quite successful. The relation of iron concentration with turbidity is also discussed in this study depending on the test results. Furthermore some recommendations have been made to solve this problem.
Table of content
1 INTRODUCTION………..………1
2 BACKGROUND....……….2
2.1 Iron chemistry…………...………...………2
2.1.1 Iron in water…………...………...………...2
2.1.2 Source of iron in water…………...………..………....3
2.1.3 Effect of iron in water…………...………..……….4
2.1.4 Standard for iron in drinking water…………...………...………5
2.2 Description of the plant………...………….5
2.2.1 Treatment method ……….…7
2.2.2 Process ………..7
3 WATER TREATMENT TECHNIQUE………...8
3.1 Unit processes ……….………....………...8
4 IRON REMOVAL BY FILTRATION...…...………..10
4.1 Filter medium………...………...………10
4.2 Properties of filter media……….……….10
4.3 Rapid sand filtration ……….……..………..……11
4.3.1 Filter media……….………..12
4.3.2 Head loss development………….………12
4.3.3 Flow control during filtration…………..………..12
4.3.4 Backwashing………...………..12
5 OBJECTIVES………...……...……...13
6 MATERIAL AND METHOD………..………13
6.1 Pilot filter………..………...……….14
6.2 Setting up the pilot filter ……...………...………..14
6.3 Sampling the pilot filter….………...16
6.4 Laboratory equipment….………..16
6.5 Laboratory analysis………...………16
6.6 Laboratory experiments………18
7 RESULTS………..…....18
7.1 The pilot filter………...………...……….18
7.1.1 Determination of the parameters………..19
7.1.2 Determination of pressure drops………..23
7.2 Results of the laboratory experiments………...………..23
8 DISCUSSION………26
8.1 Iron removal……….………...……….26
8.2 Turbidity reduction ………..………...………..…………..27
8.3 Optional experiments……….………….………...28
8.4 Summary discussion……….………..29
9 CONCLUSIONS..………..……….………30
9.1 Findings……….………...……….30
9.2 Suggestions for future work…..……….………...……….31
10 List of references………31
Appendices Appendix I Treatment plant layout……….………..33
A) Treatment process….……….………..33
B) Location of the wells………….……….………..33
Appendix II Previous investigation in well 21, 2000………..………..34
Appendix III WHO guideline for aesthetic quality……….………..34
Appendix IV River water quality parameters……….………...35
Appendix V Size distribution of particles in water and of filter materials………….…...36
Appendix VI Equipment for laboratory analysis……….………..36
A) Thermometer……….………36
B) pH meter…….……….………..………37
C) HACH 2100 Turbidity meter……….………...37
D) Color kit……….……….………..38
E) Colorimeter……..….……….………...38
F) Yellow Springs 58 Oxygen meter………….………39
Appendix VII Method of iron detection by colorimeter.….………..40
Appendix VIII Test results. ……….………..41
A) Results for different parameters at pilot filter runs….………..41
B) Results if pressure drops…...….……….………..43
Tables and Figures
Tables
1 Pressure drop for different flow rates in different time duration ………23
2 Turbidity and iron content after stepwise addition of acid to raw water from well 21 and filtration……….24
3 Turbidity and iron content after addition of alkali and filtration ………25
4 Turbidity and iron content after aeration of different duration followed by filtration .. ………..………….26
Figures 1 Location of the drinking water plant in Luleå ……..………..………….1
2 pE-pH diagram of iron system ………..………..………….4
3 Water treatment processes in Luleå water treatment plant .……….6
4 Process of rapid sand filtration ………..…………11
5 The pilot filter………..………...15
6 Temperature recording of water samples from the pilot filter ………..………….19
7 pH values of water samples from the pilot filter ………...……….20
8 Turbidity of water samples from the pilot filter ….…….………..………….20
9 Color of water samples from the pilot filter …………...………...………….21
10 Iron content of water samples from the pilot filter…..………..………….22
11 Oxygen content of water samples from the pilot filter ………..………22
12 Co-relation between turbidity and iron content in treated samples from well 21..27
1. INTRODUCTION
Water is considered as another name of life. About 66% of the human body is considered to be water. It is the main transporter of all nutrient and air to cells. If there is shortage of water occurs dehydration follows. On the other hand drinking of polluted water will cause different diseases like diarrhea or cholera. Hence pure drinking water is a fundamental right of all on the earth.
Only 3% of the water in the world is fresh. About 2/3rd of this is frozen in polar ice caps and glaciers. Major portion of the rest is ground water and only 0.3% appears as surface water. About 7/8th of this fresh surface water is contained by freshwater lakes. The other sources are swamps and rivers. Hence the portion of drinkable water is negligible on earth. In this condition to fulfill the huge demand of drinking water proper treatment is required (Bolonkin, 2007).
The Luleå drinking water plant is situated in the northern part of Gäddvik (Fig 1) along the side of the Lule ålv . The process applied in the plant is artificial recharge of ground water and afterwards treatment in the main plant (Appendix I [A]). There are two artificial infiltration ponds in the plant. There are 28 wells constructed (Appendix I [B]) for pumping artificially recharged ground water. Not all of the wells are contributing to the main supply network for treatment. There exists a long term problem with well no. 21 contaminated with iron. There have been a number of attempts to activate the well solving the problem of iron contamination. However they were not successful due to unsatisfactory results. Hence this study was run to find the current situation and find a suitable way to treat the well water free from iron.
Fig 1: Location of the drinking water plant in Luleå
There have been a number of investigations (Appendix II) (Official Documents, 2000) in the well 21 since it was constructed in 1974 to find out the iron content of the well water
The Plant
and an ideal way to treat it. This study is targeted to investigate further the way of treating the well water. Rapid sand filtration is applied for treating the water. In this regard a pilot filter was established at the well. Along with iron the other parameters related to water quality were also measured in this investigation.
2. BACKGROUND
The recommended daily intake of iron varies from 7-27 mg/d depending on age, sex or state of pregnancy (National Academy of Sciences, 2004). However higher consumption of iron may cause digestion problems. Hence excess of iron in drinking water must be treated prior to drinking.
2.1 Iron chemistry
Iron is measured to be the 2nd most abundant metal in the earth's crust. Of which about 5% appears in the outer crust. Iron in elemental form is rarely found in nature. Iron ions Fe2+ and Fe3+ generally combine with oxygen- and sulfur-containing compounds forming oxides, hydroxides, carbonates, and sulfides. Most common forms found in nature are iron-oxides (WHO, 1996).
Groundwater directly pumped from anaerobic conditions may contain several mg iron (II) /l without discoloration or turbidity. However turbidity and colour may develop in pipes at even low iron concentration levels of 0.05–0.1 mg/litre (WHO, 1996).
2.1.1 Iron in water
Iron is mainly present as Fe(OH)2+
(aq) in water at oxygen-rich conditions. On the other hand iron appears as divalent iron under poor oxygen conditions. Iron also appears in many organic and inorganic complexes generally soluble in water
Iron corrodes in the presence of water and oxygen. The color changes from silvery to reddish-brown due to the formation of hydrated oxides. The reaction is as follows:
4 Fe + 3 O2 + 6 H2O -> 4 Fe3+ + 12 OH- -> 4 Fe(OH)3 or 4 FeO(OH) + 4 H2O Iron is mainly released in water through weathering of iron containing minerals. Under normal conditions elementary iron dissolves in water. However naturally occurring iron oxide, iron hydroxide, iron carbide and iron penta carbonyl are water insoluble. pH values sometimes govern the solubility of iron compounds. Solubility of some iron compounds increases as pH decreases. FeCO3 has solubility of 60 mg/L in water while FeS is soluble upto 6 mg/L. Furthermore iron chelation complexes are water soluble.
Generally Fe2+ compounds are water-soluble while Fe3+ compounds are water insoluble.
Fe3+ compounds are only soluble in water under strongly acidic condition. But water solubility increases when these are reduced to Fe2+ (Lenntech, 2008).
Fe solubility largely depends on pH and redox potential. Generally ground water exists in deoxidized zone. Hence dissolved oxygen content is rather low. Iron is only soluble in water in pH<6.5 and in absence of oxygen. Normal solubility of iron in ground water is
<1mg/l. However even this solubility will increase significantly if there is only slightest change in pH. Sometimes peaty soils contain humic acid and deoxygenating occurs due to contamination by organic matter releasing CO2. Consequently pH reduces and iron solubility increases (Gray, 2005).
2.1.2 Source of iron in water
The solubility of iron in water is primarily governed by its oxidation state. In higher oxidation state it is highly soluble within the pH range of 6-8. In addition to Redox potential and pH iron concentration of ground water also depends on the following factors-
- Geology
- Characteristics of the aquifer.
- Soil and bed rock type.
- Ground water flow pattern in the aquifer.
- Existence of iron processing microorganisms eg. Bacteria.
Iron is released in solution by primary weathering of minerals. In one process solid iron (Fe0(S)) is oxidized by O2 to Fe+2:
Fe0(S)+ O2+2H+ Fe+2+ H2O
This chemical reaction occurs when the mineral comes in contact with water with dissolved oxygen. Fe+2 may be further oxidized by more dissolved oxygen which will generate insoluble ferric hydroxide:
O2(g) +2Fe+2 + 2H+
2Fe+3 +2H2O Fe3++ 3OH-
Fe(OH)3(s)
At natural ground water conditions both Fe+2 and Fe+3 are present. However Fe2+ is dominating under normal pH values (6.6-8.3). The proportion of Fe+2 and Fe+3 depends on the redox potential of the system. Furthermore iron solubility is also dependent on pH as can be seen from the following figure 2.
Fig 2: pE-pH diagram of iron system (Montesinos, 2005)
In deep soil CO2 is generated from oxidation of organic matter. Hence ground water remains in equilibrium with partial pressure of CO2. This lowers the pH and increases the solubility of the minerals. When ground water is aerated CO2 is released and amount of O2 increases oxidizing the soluble Fe+2 to Fe+3. Hence aeration of ground water helps removal of iron (Hedlund et al, 2000).
2.1.3 Effect of iron in water
Iron is not very harmful for human health. Still it influences a lot on the taste and color of drinking water. It poses poor tasting as well as unattractive color. Iron taste threshold is about 0.3 mg/l however it varies among individuals (Gray, 2005). Iron rich water stains both plumbing fixtures, clothing and shows unattractive color when mixed with tea, coffee or alcoholic beverages. Foods cooked in iron rich water gives dark appearance and unpleasant taste. Even a low concentration, as 0.3 mg/l of iron, can create reddish-brown stains on plumbing fixtures, dishes and clothing. These stains are difficult to remove. Iron is responsible for red water staining. Sometimes, Fe (III) iron deposits in water supply pipes cause cracks and supply brown tap water. Some bacteria generates in iron deposits which results in slimy coating inside water supply pipes, which pose unpleasant taste and impose yellow stains on cloths during wash. Iron deposits sometimes clog water pipes, water filter, pump screens which may be difficult to fix and expensive as well.
Iron most commonly found in drinking water is not considered as serious health concern.
On the contrary small concentrations of iron in water is even beneficial for human health.
Iron is necessary for the red blood cells to transport oxygen in the blood stream (Colter et al, 2006).
2.1.4 Standard for iron in drinking water
The development of setting the standard of drinking water has a long history. In 1913 the first formal and comprehensive review of drinking water was launched. After long research and experiments interim standards were set in 1975. In addition to health related enforceable standards. The non enforceable guidelines were set in 1979 for contaminants that affects aesthetic quality of drinking water (Appendix III) .The standards were set as secondary maximum contaminant level (SMCL) (AWWA, 1990).
Iron standard in drinking water is decided on two criteria set by the US Environmental Protection Agency (EPA, 2002). Primary standard is fixed upon health risk (classes of pollutants e.g. toxicity, hygiene, radioactivity etc.) and secondary standard is decided on aesthetic issues e.g. turbidity, color, taste, odor, corrosion, foaming and staining effect of water on properties. Standard of iron is controlled by the Secondary Maximum Contaminant Level (SMCL) .The value of SMCL for iron in drinking water is 0.3 mg/l (milligrams per liter), or 0.3 ppm (parts per million) (Colter et al, 2006).
The World Health Organization sets iron standard in drinking water as 0.3 mg/l. However the European Commission directive has fixed iron standard in drinking water to 0.2 mg/1.
Guideline value for iron in drinking water is set ≤0.05 mg/1 in the Netherlands. To minimize the cost of maintenance in distribution network many Dutch companies prefer to keep iron standard below 0.03 mg/l (Sharma, 2001).
2.2 Description of the plant
The Luleå water treatment plant is an artificial ground water recharge plant. The average daily treatment volume of water is about 20,000 m3. The treated water is supplied to about 65,000 people residing in Luleå and it’s neighbourhood. The source of water for artificial recharge is the river Lule älv flowing beside the plant. Water from the river is directly pumped into the plant where it undergoes rapid sand filtration. After this the filtrate is sent to two artificially constructed infiltration ponds by gravity. The water infiltrates through the permeable bottom of the ponds and reaches ground water. Hence the aquifer is artificially recharged. The flow diagram of the treatment processes is illustrated in fig 3.
Fig 3: Water treatment processes in Luleå water treatment plant (Info från Luleå Kommun, 2008)
1- Water is pumped from Luleå river 2- Pump house
3- Intake water goes through rapid sand filtration 4- Filtered water is sent to the artificial recharge ponds
5- Water infiltrates through the permeable bed of the ponds and recharge the aquifer 6- Recharged ground water is pumped up through a number of wells and slaked lime, carbon dioxide and hypochlorite are added
7- Water is supplied to the overhead reservoir for distribution after disinfection
The river as a source is a nice choice for artificial recharge as it is more or less clean being undisturbed by any human activities upstream (Boulanger, 2004). The basic parameters of the river water as surveyed by a consulting company (Alcontrol Laboratories, 2007) are given in Appendix IV.
1
2 3
4
5
7
6
2.2.1 Treatment method
Artificial recharge of groundwater has been in practice in Luleå for a long time. At the starting of the plant in 1906 ground water was pumped directly from the wells. But after a period of pumping the water quality deteriorated due to lowering of ground water level, even below the river bottom, and intrusion of saline water from the sea took place . The pumped water became seriously saline and undrinkable. Investigation during the period of 1915-1928 showed that the salt concentration varied from 182 mg/l-575 mg/l. One sample even showed 1200 mg/l of salt (Official Documents, 1934). Therefore a decision was made to recharge ground water artificially in 1935. The infiltration pond actually acts as a slow sand filter media. Water infiltrated in the ground remain there for several weeks.
There are 28 wells constructed to pump ground water. Pumped water is sent to the main plant for further treatment. Disinfection by chlorination and lime and carbon dioxoide addition for corrosion control is done here. Sodium hypochlorite (NaOCl) is added as the disinfecting agent. Lime and carbon dioxide (Ca(OH)2+ CO2) are added which increases the pH and hence reduces the corrosive action. Finally the treated water is stored in two underground reservoirs and afterwards pumped to two overhead reservoirs in the town.
Fresh water is distributed to the consumers from these high reservoirs.
2.2.2 Process Disinfection
Chlorination is used for disinfection. The amount of chlorine added is about 0.3 mg Cl/L.
NaOCl is easily mixed with water due to it’s liquid phase.
Corrosion control
Lime and carbon dioxide are added for corrosion control. Dissolution is required for addition of lime as it is delivered as a powder. A small stream of water is diverted from the plant inlet for dissolution of lime. The solution passes through two successive dissolution chambers having a volume of 15 m3 and 80 m3 respectively. After dissolution the stream is added to the main stream serving the purpose of lime addition. Addition of NaOCl, Ca(OH)2 and CO2 is done at the same point. The following chemical reactions occur during mixing
Ca(OH)2 Ca+2 +2OH- CO2+ OH- HCO3-
(pH increases)
Due to occurrence of Ca+2 hardness increases while pH increases due to HCO3-
formation. In this way corrosion control can be mentioned as pH correction as well as alkalinity and hardness adjustment (Sotty, 2000).
3. Water treatment technique
Drinking water should be free from organisms that may cause diseases from harmful minerals and organic substances. It should also have acceptable levels of turbidity, color, odour and be free from objectionable taste. It should have reasonable temperature. Water having that quality is called potable water meaning it can be consumed without any health risk.
Selection of water treatment processes is determined by the following facts- - Water source quality
- Desired quality of supply water - Reliability of process
- Skill of personnel
- Space requirement for construction of treatment facilities.
- Options for waste disposal
- Establishment and operation cost.
3.1 Unit processes
The following unit processes are generally used for water treatment in any water treatment plant.
Air stripping and aeration
Air stripping is done mainly for the following purposes:
- To remove CO2, H2S and other taste or odour causing compounds and VOC.
- Add O2 to water to oxidize Fe & Mn and prevent fomentation of reduced environment that may cause taste & odour problems.
Coagulation process
Alum and Iron (III) salts are mainly applied as coagulants used for water treatment. It is dependent on particular suspension, quantity, chemical dose, pH, temperature, ionic strength and reaction time. Coagulation is mainly applied for:
- Mixing, destabilization and flocculation
- For promoting aggregation of small particles into larger particles and consequently remove by sedimentation or filtration.
- To remove clay and silt based turbidity; natural organic matter such as microbial contaminants, toxic metals, synthetic organic matter such as microbial contaminants, toxic metals, synthetic organic chemicals and iron & manganese.
Sedimentation and flotation
These processes are mainly dependent on gravity. Heavier particles fall to the bottom of a water column and can be removed after accumulation on the bottom. In flotation air bubbles are introduced that attach the solid particles and bring them up to the water surface. Afterwards the accumulated particles are removed from the top.
Filtration
It is applied mainly to remove suspended particles like clay, silt, colloidal and precipitated natural organic matter, metal salt precipitates from coagulation, lime softening precipitates, microorganisms. Granular media is commonly used as filter for potable water treatment. Water is passed through the filter either by gravity flow or by pressurized passage through the filter media. The type of granular media, size, shape and depth determine pore volume, pore size and pore distribution. This affects solid holding capacity, head loss characteristics, filtrate quality and backwash flow requirement.
Sand, crushed anthracite coal, garnet, Ilmenite and granular activated carbon (GAC) are commonly used as filter media. The performance of the filter depends on the rate of filtration. Normally a rate of 5 m/ h is applied in rapid sand filtration.
Backwashing at regular intervals is important to maintain optimum effluent quality as well as to prevent gradual deterioration of the media and eventual replacement. Sufficient pretreatment and screening of larger particles are necessary for better filter performance.
Ion exchange and Inorganic adsorption
Ion exchange with synthetic resins or adsorption onto activated alumina, is applied when the source water is rich in mineral content. It is basically applied for water softening.
Both the processes are applied for removal of contaminants like ions of toxic or radio- active substances such as Barium, Arsenic, Chromium, Fluorine, Nitrate, Radium and Uranium.
Chemical precipitation
Precopitation is applied for softening and Fe and Mn removal. It also helps removing heavy metals, radionuclide as well as viruses and bacteria.
Membrane processes
There are a number of membrane processes applied in water treatment. These are
• Reverse osmosis (RO)
• Electrodialysis (ED)
• Electrodialysis Reversal (EDR)
• Ultra-filtration (UF)
• Nano-filtration (NF)
• Micro-filtration (MF)
Ultra-filtration and Nano-filtration are applied for removing particulates, color, tri- halomethane (THM) precursors and hardness. RO, ED and EDR are applied basically to treat brackish water.
Chemical Oxidation
This is basically applied for controlling growth of biological components in pipelines and basins, control of color, taste and odor, oxidation of Fe & Mn, and disinfection.
Adsorption of organic compounds
Powdered or Granulated activated carbon (PAC or GAC) adsorption is generally used for the removal of dissolved organic compounds, color and taste and odor causing compounds. These adsorbents basically adsorb high molecular weight and non polar compounds.
Disinfection
It is done as the last stage of water treatment to assure microbiologically safe finished water. It is also required to control odor developing biologically growth algal growth in basin and channels. Common disinfectants used are free or combined chlorine, chlorine dioxide, ozone, ultraviolet irradiation and other miscellaneous disinfectants (Frederick, 1990).
4. Iron removal by filtration
Filtration is the best option for removing iron. However filtration depends of various factors and properties of filter.
4.1 Filter medium
Common types of medium used for filters are silica sand, anthracite coal and garnet or ilmenite. A filter can contain one or more of those filter medias in different layers.
Sometimes some adsorbents are also used in filters like GAC. It is useful in reducing taste and odor problems.
4.2 Properties of filter media
Filtration performance is dependent on different properties of the filter media.
• Grain size and size distribution
• Grain shape and roundness
• Grain density and specific gravity
• Grain hardness
• Fixed bed porosity
Two types of filtration are applied for water filtration. Slow sand filtration and Rapid sand filtration. Rapid sand filtration is mostly applied in large water treatment plants. In this study Rapid sand filtration is applied for iron removal.
4.3 Rapid sand filtration
Rapid sand filtration is the most common filter operation for treating water. Water pass through the filter by gravity at a rate of 5-25 m/h. During the flow the solids are screened from the water and mostly accumulated at the top of the filter. The voids and the top of the filter become clogged with those solids which create gradual head loss. This leads to a backwash of the filter after a certain interval. Operating time of the filter between two consecutive backwashes is called filter run and total head loss measured before backwashing is called terminal head loss. Generally filter runs vary from 12-96 hours depending on the rate of filtration. Increased rate of filtration shortens filter run roughly inverse with the rate. A typical mode of rapid sand filtration is depicted in fig 4.
Fig 4: Process of rapid sand filtration (Cheremisinoff, 1995)
4.3.1 Filter media
Rapid filters usually consist of sand, crushed anthracite coal, Granulated activated carbon (GAC), garnet or ilmenite. Grain sizes of common filter media are presented in Appendix V. GAC is used sometimes in filter as an adsorbent of organic compounds and consequently reducing taste and odor. Generally lifetime of a GAC bed is assumed to be 1-5 years unless high loading of organic rich water is used.
The quality of the filtrate varies during the filter run. Generally at the beginning and at the end the quality is poor. Water quality improves after some time of run and reaches the best value and finally it begins to deteriorate (filter break through).
4.3.2 Head loss development
Head loss is another phenomenon that develops during the filter run. It develops at a rate roughly proportional to the solids captured by the media. If it can be assumed that total volume of incoming solids is captured, the development of head loss can be declared proportional to the rate of filtration. Head loss also depends on the size of grains of the filter. If it is a fine grained filter the solids are mostly accumulated at the top, clogging the filter and hence increasing the head loss. On the other hand in case of coarse grained filter the captured solids penetrate deep inside the filter allowing sufficient time to develop head loss. The most common head loss pattern seen in rapid sand filtration is linear with respect to volume of filtrate.
4.3.3 Flow control during filtration
Flow through the filter should be carefully controlled to avoid any sudden fluctuations in flow rate, which may deteriorate the filtrate quality. Mechanical flow control or flow control by inherent hydraulics is a nice way of handling flow rate.
4.3.4 Backwashing
Backwashing is necessary for the following reasons.
• For long term effective filtration
• To keep the filter acceptably clean
• To protect generation of any foreign media such as mud balls
• To protect the filter from cracks, tunnels.
Typical backwash rate should be 37-49 m/h according to US practice (AWWA, 1990).
The bed expansion should be 15-30%. Backwash is done until the water over the filter is sufficiently clear. A turbidity of 10 NTU is considered sufficient to stop backwashing.
Water amount required for backwashing depends on the depth of filter media and the depth up to which the particles are penetrated in the filter media. More water is required
for deep filters as well as in case of deep penetration. According to US practice the typical water volume required for backwashing is 4-6 m3/m2.
The expansion of the filter media during backwashing depends on size and size gradation and grain shape and density. 15-30% expansion of the filter bed is considered satisfactory during backwashing.
Other phenomena like stratification and intermixing are important at backwashing. The tendency of stratification depends on the bulk density of different sizes of sand grains.
Fine grains have lower bulk density and hence expand more during backwashing. As a result the finer layer remains at the upper portion of a filter after backwashing. This phenomenon is helpful in removing any foreign particle from the filter (AWWA, 1990).
5. Objectives
The work has the following objectives:
• Measure the degree of contamination by iron in well 21 water
• Find an ideal way to treat iron from water
• Monitor the performance of the filter media in removing iron
• Check other parameters of both raw and filtered water (e.g temperature, turbidity)
• Recommend the plant about how to activate the well on the basis of the findings in the research
6. Material and method
The main objective set for this research work is to find a suitable way to reduce the iron contamination in well 21 and to consequently add this well to the main water treatment plant. Filtration through sand bed in a pilot filter is checked as one of the recommended options of iron removal.
An experiment was performed throughout the winter until summer. A pilot filter was installed in December 07 and sampling was done until end of May 08. Hence the surface water temperature differs in a large scale during this period. However the temperature of the ground water remained stable throughout the period. Artificial heating device was set inside the pumping station, which affects the temperature of the treated samples. As the temperature remains stable there should not be any significant changes in physical as well as chemical properties of water like solubility of iron or biological actions due to iron bacteria. Therefore this stability of temperature indicates smooth filtration of water through the ground and uniform travelling time. It also indicates there exists no mixing of water or oxygen from any anthropogenic sources.
Experiments were run in two stages. First the well water was run through the pilot filter and afterwards the filtrate and the raw water were analyzed in the Laboratory.
6.1 Pilot filter
The materials required for setting up the pilot filter were as follows, a) Transparent plastic tube
(Height 2m, Inner diameter 12.5 cm, Outer diameter 13.2 cm)
The tube has 8 numbers of connections for the pressure pipes. The cumulative height from the base are 15, 30, 45, 60,70, 75, 80, 85 cm (fig 5) respectively. There are three connections for flow pipes. Two are at the top as inlet and outlet of overflow and one at the bottom for filtrate outlet (Fig 5). There is a screen at the bottom to protect the filter material from washing away.
b) Steel frame
It is provided to support the filter to remain in vertical position. It has flared (L- shape) legs and a distribution chamber at the top.
c) Filter medium
German sand was used as the filter medium having following properties.
Name: German sand Grain size: 0.8-1.2 mm d) A suction pump
e) Flow meter (Traditional for households) f) Valves (4 no)
g) Plastic pipes, measuring band, insulating material, heater, Plastic bottles for sampling.
6.2 Setting up the pilot filter
The outline of the installed filter is given in the following figure 5.
Fig 5: The pilot filter
At first the plastic pipe was filled up with sand to the depth of 103 cm (≈1m). Care was taken to avoid segregation of sand grains. The sand was poured in a slow and smooth way. The filter was erected vertical and bound to the steel stand for support. Transparent rubber pipes were connected to the joints for pressure pipes. The inlet was joined with pipe extension from the pump having valves in two ends. The outlet pipe had two branches. One was sent to the filtrate outlet and other branch was connected to the pump outlet for backwashing purpose. There existed a valve at the end. A flow meter was set after the pump to measure the inflow. A measurement band is attached to the column starting zero at the surface of the water. The height of the water column above the sand was measured as 83 cm. Hence the total height of the water column and filter medium became 2 m approximately.
The initial flow rate started at the beginning was 1 l/m. The flow pattern was monitored for some time so that there should not be any occurrence of channeling. The inflow was measured by the flow meter attached in front of the plant. Pressures at different levels were detected by the pressure pipes. Before the filter run the filter media was backwashed
Valve Piezometer Inlet
Filter media Water column Overflow outlet
Backwashing pipe
Treated water Flow meter
85 80 75
70
60
45 30
15
Height from bottom of the filter in cm
to clean the sand grains. The total working time of the filter was about six months including time for backwashing. The highest flow rate applied at the end was about 20 l/m.
Backwashing was done in suitable intervals depending on the pressure drops measured from the pressure pipes. Generally when the pressure drop became significant backwashing followed. The flow rate was set to a minimum of 1 l/min in the to maximum of 20 l/min at the end of the experimented period which corresponds to 5 m/h and 100 m/h respectively. The higher the flow rate the more frequent the backwashing was done.
Flow rate for backwashing was about 100 m/h and backwashing time was about 10-15 minutes. During backwashing the filter bed expansion was about 38-40%. The pressure pipes were inspected every time to avoid any wrong readings due to clogging or air bubbles.
Some parameters were measured in the field such as temperature, pH, and oxygen content. Those were detected in different parts of the well and pilot filter. For example oxygen content was measured in the water inside the well, water column above the filter media both top and bottom. These parameters were checked to ensure that there existed no physical or chemical changes that might affect iron removal process.
6.3 Sampling the pilot filter
Samples were collected from the well at different times .At each occasion 4 samples were collected: 2 raw water samples and 2 treated water samples. Before collecting the samples pressure drops were measured. If the drop was significant backwashing followed. During sampling sincere care was taken so that there should not be any mixing between raw and treated water. Samples were rushed to the laboratory for analysis as soon as possible so that no changes appeared in the samples like precipitation of iron. The sampling period was from December 07 to May 08.
6.4 Laboratory equipment
For laboratory analysis the following equipment have been used 1. Temperature meter
2. pH meter 3. Colorimeter 4. Oxygen meter 5. Ferozine pillows 6. Color meter 7. Turbidity meter
8. Sampling bottles (1 l capacity) 9. Glass tubes
6.5 Laboratory analyses
Samples were tested for detection of the following parameters.
Temperature
The electrode of the thermometer (Appendix VI A) was placed deep in the sample
retained in a beaker. Temperature reading was taken after a few seconds when the reading became stable.
pH
The pH meter (Appendix VI B) was appointed to measure the pH. The electrode was placed in the sample is such a way that it remained drowned in the sample .The meter starts stabilizing itself for pH detection. A beep is given when the pH meter is stabilized.
Then the reading of pH is taken.
Turbidity
The HACH 2100 turbidity meter (Appendix VI C) was appointed to measure the turbidity in the laboratory. This device detected the turbidity depending on light refraction resulted from particles remaining in water. A small bottle of 25 ml was inserted inside the hole in the equipment. It took about 3-4 minutes to detect the reading of the turbidity.
Color
A simple color kit (Appendix VI D) was used for comparison of color of the samples to the normal potable water produced by the plant. Color disks were used to compare the color. There are two glass cylinders in this kit. One was filled up with the sample and another with distilled water as the reference for comparison. The kit was located in a place with sufficient light so that the color comparison could be done accurately.
However the detection was done depending on eye inspection. Hence the detected values can have standard deviation of ± 5 mg Pt/l.
Iron Content
The amount of total iron was detected by DR/890 Colorimeter (Appendix VI E) manufactured by HACH. It is a microprocessor controlled, led sourced filter photometer suitable for calorimetric test in lab or field. It is pre-calibrated for common colorimetric measurements. It displays results of concentration, absorbance or percentage transmittance. Total iron Ferozine method was applied to detect the total iron concentration in the samples. The step by step procedure is given below.
Procedure
1. Enter the program no. for Fe .(37) 2.Fill a sample cell with 25 ml of sample.
3.Press zero.
4. Add the contents of one Ferrozine iron reagent solution pillow to the cell. Shake well for mixing.
5. Timer (5min). Violet color develops which indicates presence of Fe.
6. After 5 minutes the machine beeps:press read and get the volume from display.
-Standard deviation ± 0.004 mg/l
-Estimated detection limit for iron is 0.011 mg/l
The reagent forms a purple color complex with trace amounts of Fe in the samples that are buffered to a pH of 3.5. This method is suitable for detecting trace levels of iron (HACH Manual, 2008). Detailed method for iron detection is shown in Appendix VII.
Oxygen Content
A Yellow Springs 58 oxygen meter (Appendix VI F) was appointed to measure the volume of dissolved oxygen in water samples. The saturation value of oxygen for the water temperature was checked according to the standard table printed on the oxygen meter. Reading was taken when the value became stable for few seconds.
6.6 Laboratory experiments
Besides detecting the above parameters some other options for reducing turbidity in laboratory scale were adopted such as aeration, alkalization or acidification followed by filtration. These experiments needed the following extra tools and chemicals:
• Aerator
• Sodium hydroxide solution
• Sulfuric acid solution (0.1 N)
• Filter paper (Diameter- 15cm, OOH-SI-80-40)
• Funnel and support stand for funnel
The aerator was used to aerate the sample for different times. After the aeration the raw samples were filtrated through the filter paper and afterwards the concentration of iron and turbidity was determined.
Sodium hydroxide solution was added to the samples in different amount resulting in different pH increase. The pH was checked before and after addition of this solution.
Finally the samples were filtrated through filter paper and iron concentration and turbidity were determined.
7. Results
7.1 The pilot filter
A total of 39 samples (Raw & Treated) have been analyzed at the laboratory during different periods. Two raw and two treated samples were treated at a time. During filtration operation the pressure drop was monitored before every backwash operation.
7.1.1 Determination of parameters Temperature
The temperature of different samples at different time is shown in fig 6.
Temperature Vs Time Curve
0 5 10 15 20 25
0 5 10 15 20 25 30 35 40
Sample no. & Time
Temperature C
Raw water temperature C Treated water temperature C
Fig 6: Temperature recording of water samples from the pilot filter
The temperature record shows that the temperature of both raw and treated samples remained almost the same. There was no significant fluctuation in temperature for different samples. The temperature varies from 15 C to 7 C until February 08. Afterwards the temperature became almost stable and varied between 7.5-8.5 C.
pH
The pH of the different samples is shown in fig 7.
Dec 07 May 08
pH
0 2 4 6 8 10
0 5 10 15 20 25 30 35 40
Sample no.
pH
Raw water pH Treated water pH
Fig 7: pH values of water samples from the pilot filter
The pH values of the samples both raw and treated was between 7.5 to 9 reflecting a certain alkalinity of the water. There exists no effect on pH values due to filtration.
Turbidity
The turbidity of the different samples, both raw and treated, is shown in fig 8.
Turbidity
0 2 4 6 8 10 12 14 16 18 20
0 5 10 15 20 25 30 35 40
Sample no.
Turbidity, NTU
Raw w ater turbidity (NTU) Treated w ater turbidity (NTU)
Fig 8: Turbidity of water samples from the pilot filter
The turbidity results show that there are significant fluctuations in raw water turbidity while the treated water turbidity remains almost same for all samples. Although the
Dec 07 May 08
Dec 07 May 08
reduction of turbidity by the filter was high the residual turbidity is too high to be accepted as potable water.
Color
The color detection of different samples both raw and treated is shown in fig 9.
Color
0 5 10 15 20 25 30 35 40
0 5 10 15 20 25 30 35 40
Sample no.
Color, mg Pt/l
Raw w ater color (mg Pt/l) Treated w ater color (mg Pt/l)
Fig 9: Color of water samples from the pilot filter
Color values of different raw samples are fluctuating in different periods. It is related to iron content of the water samples as iron gives yellowish color. However the color values for treated samples shows quite stable values which indicates an active color reduction by filter operation.
Iron content
Removal of iron is the main objective of this research work. Raw water iron concentration is quite higher than the guideline values and unacceptable for consumption.
The iron content of different samples both raw and treated is shown in fig 10.
Dec 07 May 08
Iron content
0 0.5 1 1.5 2 2.5 3 3.5
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39
Sample no.
Iron content, mg/l
Raw w ater iron content (mg/l) Treated w ater iron content (mg/l)
Fig 10: Iron content of water samples from the pilot filter
The figure shows that the filter is quite successful in removing iron from the raw water well below the recommended values of 0.3 mg/l. However sometimes the raw water iron concentration is low and hence the treated sample iron concentration is almost zero.
Oxygen content
About 18 samples were investigated for oxygen content, both raw and treated. The result is shown in fig 11.
Oxygen content
0 2 4 6 8 10
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Sample no.
Oxygen content, mg/l
Raw water oxygen content (mg/l) Treated water oxygen content (mg/l)
Fig 11: Oxygen content of water samples from the pilot filter
Dec 07 May 08
Dec 07 March 08
Results of oxygen content shows that both raw and treated samples have almost similar concentrations of oxygen. It indicates there is no significant mixing of air during filtration operation.
Detailed results of all the parameters are given in Appendix VIII A.
7.1.2 Determination of pressure drops
Pressure drops were checked before every backwash governed by the flow rate. The higher the flow rate the higher is the backwashing frequency. Maximum pressure drops observed for different flow rates at different time are illustrated in the following Table 1.
Table 1: Pressure drop for different flow rates in different time duration Date Between Consequent
Backwashings
Maximum Pressure Drop
(cm)
Flow rate (m/h)
Days Hours
07.12.17 7 168 14 7.5
07.12.29 12 288 75 12.5
08.02.09 4 96 52 15
08.02.19 10 240 66.5 22.5
08.02.29 10 240 79 30
08.04.11 11 264 82 40
08.05.05 21 504 100 50
08.05.12 4 96 93 75
08.05.16 2 48 97 100
The results illustrates simple co-relation between flow rate and pressure drops. In higher flow rate the development of pressure drop is quite frequent. Detailed results of the pressure drops are given in Appendix VIII B.
7.2 Results of laboratory experiments
The raw water samples were also tested in the laboratory to reduce the turbidity by pH adjustment. The results for addition of acid (H2SO4, 0.1 N) are shown in table 2.
