;, 1

'rr
, J
.1 •
.

,,

REMOVAL OF MANGANESE FROM GROUNDWATER BY

OXIDATION AND ADSORPTION PROCESSES

SNIGDHA AFSANA

DEPARTMENT OF CIVIL ENGINEERING

BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY
DHAKA
AUGUST 2004
1111111111111111111111111111111111'
1199604#
.~
- --

.r.~:
\..~ .~,
 "

REMOVAL OF MANGANESE FROM GROUNDWATER BY

OXIDATION AND ADSORPTION PROCESSES

A Thesis Submitted
by
SNIGDHA AFSANA

In partial fulfillment of the requirements for the degree of

Master of Science in Engineering (Civil & Environmental)

DEPARTMENT OF CIVIL ENGINEERING

BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY
DHAKA
AUGUST 2004
 BANGLADESH UNIVERSITY OF ENGINEERING AND TECHNOLOGY
DEPARTMENT OF CIVIL ENGINEERING

CERTIFICATE OF APPROVAL

We hereby recommended that the thesis titled "Removal of Manganese from

Groundwater by Oxidation and Adsorption Processes" submitted by Snigdha Afsana,
Student No. 100104156 (P) be accepted as fulfilling this part of the requirements for the
degree of Master of Science in Engineering (Civil & Environmental).

BOARD OF EXAMINERS

Dr. M. Ashraf Ali

:Chairman (Supervisor)
Associate Professor.
Dept. of Civil Engineering, BUET

Dr. S(
Le.
Sekender Ali :Member
Professor and Head
Department of Civil Engineering, BUET

Dr. Feroze Ahmed :Member

Professor
Department of Civil Engineering, BUET

'~~
D~i1 :Member
Professor
Department of Civil Engineering, BUET

Dr. M. S ah Alam Khan :Member (External)

Assistant Professor
Institute of Water and Flood Management, BUET

11
 DEDICATIONS

To my Family Members

III

•
 DECLARATION
~.

It is hereby declared that except for the contents where specific references have been
made to the work of others, the studies contained in this thesis titled "Removal of
Manganese from Groundwater by Oxidation and Adsorption Processes", is the result of
investigation carried out by the author. No part of this thesis has been submitted
elsewhere for any Degree, Diploma or other qualification.

(Dr. M. Ashraf Ali) (Snigdha Afsana)

Counter Signed by Supervisor Signature of the Candidate

IV
 •

ACKNOWLEDGMENT

First of all, I want to express my deepest gratitude to Almighty Allah for His
graciousness, unlimited kindness and with the blessings of Whom the good deeds are
fulfilled.

I am delighted to express my heartiest gratitude and sincerest indebtedness to my

supervisor Dr. M. Ashraf Ali, Associate Professor, Department of Civil Engineering,
BUET, for his continuous guidance, invaluable suggestions, affectionate encouragement
and generous help at every stage of this study. His active interest in this topic and
valuable advice was my source of inspiration to carry out the study.

Many thanks to the technical staff of the Environmental Engineering Laboratory, BUET,
for their assistance and cooperation during the study.

A very special debt to my mother, husband, mother-in-law, father-in-law and other

relatives and friends, who are always a constant source of inspiration throughout my life.

v
 ABSTRACT

Although significant research works have been carried out on removal of arsenic and iron
from groundwater, relatively little work has been done on the removal of manganese from
groundwater in Bangladesh, in spite of its existence at relatively high concentration in
many areas of Bangladesh. The primary concern regarding presence of manganese in
water stems from its capability to stain sanitary-ware and laundry at concentration> 0.1
mg/L and its deposition in water distribution system, even at concentration as low as 0.02
mg/L. Chemical oxidation followed by filtration, is by far the most widely used
manganese removal technique. Although coagulation by iron and aluminum salts has
been widely used in Bangladesh for arsenic removal, little information is available on its
capability in removing manganese from water. The present study is focused on assessing
removal of manganese from groundwater using two chemical oxidants (potassium
permanganate and bleaching_powder) commonly available in Bangladesh. Besides,
manganese removal by ~imple aeration, a method widely used for iron removal, has also
been assessed. In addition possIble -manganese removal by coagulation with iron salts has
also been evaluated.

In this study, it has been found that potassium permanganate is capable to remove
manganese from groundwater very effectively over a wide range of initial manganese
concentration. In the near-neutral natural pH range of groundwater, maximum manganese
removal by permanganate oxidation was achieved at a permanganate dose equal to that
required from stoichiometric consideration. However, for an initial manganese
concentration of 2.0 mg/L, greater than 95% removal was achieved for a dose fraction
equal to 0.8 times that required from stoichiometric consideration. A slightly (20%)
higher permanganate dose did not have any significant effect on manganese removal.

Removal of manganese has been found to be highly dependent on pH. In general, removal
increased as pH increased. Beyond pH of about 8, removal of manganese with
permanganate is almost independent of pH. One drawback of permanganate oxidation is
the development of color. It has been found that sand filters, with depths of 10 to 20 cm,
are capable to reduce the color below Bangladesh drinking water standard. The sand
filters were also found to be very effective in removing oxidized manganese solids.

In case of oxidation by chlorine (bleaching powder), better manganese removal could be

achieved only at pH values 10 and above. In the neutral pH range, removal of manganese
was relatively poor (less than 50%) for initial manganese concentration ranging from 1.0
to 10.0 mg/L. However, removal increased significantly as pH increased and almost
complete removal was achieved at pH 10. Significant manganese removal (over 95%) can
also achieved by simple aeration at high pH value (>10). Thus, at higher pH values,
oxidation of manganese by air may have contributed to the higher manganese removal by
bleaching powder.

In this study, experiments were carried out to evaluate the influence of dissolved iron on
manganese removal by oxidation. It was found that when high amount of dissolved iron
was present in water, manganese removal by chemical oxidation (with potassium
permanganate) was very poor. This is partly due to the fact that a significant part of the.
oxidizing agent added for oxidation of manganese may have been utilized for oxidation of
VI
dissolved ferrous iron. It was also found that when both manganese and iron are present
in water, pre-oxidation of iron (e.g., by aeration) prior to the addition of the oxidizing
agent would improve manganese removal.

In this study, effectiveness of coagulation by iron salts in removal of manganese was

evaluated. Coagulation experiments were carried out with and without the addition of an
oxidizing agent (potassium permanganate) in order to ascertain the dominant removal
mechanism (either oxidation or adsorption onto iron floes). From results obtained in this
study, it appears that very little manganese could be removed by adsorption onto the
coagulated floes of iron solids. The increased removal of manganese by coagulation with
ferric chloride at higher pH values appears to be due the oxidation of manganese by
atmospheric oxygen rather than adsorption of manganese on iron solids and their
subsequent precipitation.

va
 TABLE OF CONTENTS

.Page
ACKNOWLEDGEMENT v
ABSTRACT VI

CONTENTS Vlll

LIST OF TABLES Xli

LIST OF FIGURES Xlll

LIST OF ABBREVIATIONS XVlll

CHAPTER 1 INTRODUCTION
1.1 General I
1.2 Rationale of the Study 2
1.3 Objectives of the Study 4

1.4 Organization of the Thesis 4

CHAPTER 2 OCCURRENCE OF MANGANESE: LITERATURE REVIEW

2.1 futroduction 6
2.2 Occurrence of Manganese 6
2.2.1 Physical and Chemical states 6
2.2.2 Sources of Manganese in the Environment 10
2.2.3 Environmental Fate 13
2.2.4 Manganese futake 14
2.3 Undesirable Effects of Manganese 14
2.4 Health Effects of Manganese 15
2.4.1 Short-term Exposure Studies 16
2.4.2 Long-term Exposure Studies 17
2.4.3 Standards and Guideline Values for Manganese 18
2.5 Occurrence of Manganese in Groundwater of Bangladesh 18

CHAPTER 3 CHEMISTRY OF MANGANESE AND MANGANESE

REMOVAL TECHNIQUES: LITERATURE REVIEW

3.1 futroduction 33

viii
 3.2 Chemistry of Manganese in Water 33
3.2.1 Process Kinetics for Oxidation of Manganese 34
3.2.2 pE - pH and Eh- pH Diagrams for Manganese in Solution 36
3.3 Factors Affecting Manganese Oxidation 40
3.3.1 Effect of pH 43
3.3.2 Presence of Organic Matter 43
3.3.3 Effect of Temperature 43
3.3.4 Presence of Dissolved Oxygen 44
3.3.5 Mn(H) Concentration in Solution 44
3.3.6 Effect of Alkalinity 46
3.3.7 Presence of Oxide Surfaces 46
3.3.8 Presence of Reductive Substances 49
3.3.9 Effect of presence of Metal Ions 49
3.4 Manganese Removal Techniques 49
3.4.1 Factors That Must Be Known When Choosing a Treatment Process 49
3.4.2 Overview of All Categories of Treatment 51
3.4.3 Removal of Dissolved Manganese Using Water Softeners 51
3.4.4 Aeration Followed by Filtration 54
3.4.5 Oxidizing Filter 55
3.4.6 Chemical Oxidation Followed by Filtration 57
3.4.7 Biological Oxidation 62
3.4.8 Filtration 63
3.5 Treatment Types Not Recommended 65
3.5.1 Magnetic manganese Removal Devices 65
3.5.2 Electrodialysis 65
3.5.3 Reverse Osmosis 67
3.5.4 Bag Filtration 67
3.6 Sequestering Process: Phosphate Treatment 67

CHAPTER 4 MANGANESE REMOVAL FROM GROUNDWATER BY

OXIDATION
4.1 Introduction 69
. 4.2 Materials And Methods 70
4.2.1 Manganese Removal by Oxidation with KMn04 71

ix
 4.2.2 Manganese Removal by Oxidation with Bleaching Powder 74
4.2.3 Manganese Removal by Aeration 75

4.2.4 Chemicals, Preparation of Stock Solutions 76

4.3 Results And Discussions 76
4.3.1 Removal of Manganese by Oxidation with KMnO. 76
4.3.2 Removal of Manganese by Oxidation with Bleaching Powder 89
4.3.3 Manganese Removal by Aeration 94
4.4 Summary 96

CHAPTER 5 MANGANESE OXIDATION IN PRESENCE OF IRON

5.1 Introduction 98
5.2 Materials and Methods 100
5.2.1 Manganese Oxidation with KMnO. in Presence of Dissolved Iron 101
5.2.2 Manganese Removal by Aeration in Presence ofIron 102
5.2.3 Chemicals and Measurement of Parameters 102
5.3 Results and Discussions 103
5.3.1 Removal of Manganese by KMnO. Oxidation in Presence of Iron 103
5.3.2 Removal of Manganese and Iron in Water by Aeration 113

CHAPTER 6 MANGANESE REMOVAL BY COAGULA TION-ADSORPTION-

COPRECIPITATION
6.1 Introduction 114
6.2 Materials and Methods 116
6.2.1 Manganese removal by Coagulation-Adsorption-Coprecipitation 119
6.2.2 Effect of pH on Coagulation of Manganese with Fe (III) Coagulation 119
6.3 Results And Discussions 119
6.3.1 Manganese Removal by Iron Coagulation 119

CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS

7.1 Conclusions 126

7.2 Recommendations for Future Studies 128

REFERENCES 129

x
APPENDIX A 139
APPENDIXB 142
APPENDIXC 156
APPENDIXD 164

Xl
 LIST OF TABLES

Table Description Page

No No

2.1 Physical and Chemical Properties of Manganese and that of Some

Common Manganese Compounds 8

2.2 Uses of Some Common Manganese Compounds 12

2.3 Adequate Manganese Intakes for Men, Women and Children 14

2.4 Mn Concentration Expressed as Percentile 20

2.5 Manganese Occurrence Summaries for Different Divisions 23

2.6 Manganese Statistics for the 17 Most Contaminated Districts 24

2.7 Percentages of Wells in Shallow and Deep Wells Above and Below 25
Bangladesh Standard and WHO Health-based Guideline Value

2.8 Manganese Occurrence Summaries for Different Depth Range of Wells 26

2.9 Manganese Concentration in Shallow Wells of Different Years of 26

Construction

2.10 Comparative occurrence of Arsenic and Manganese in different 28

Divisions
2.11 Comparative Occurrence ofIron and Manganese in Different Divisions 31

3.1 Stoichiometry for Manganese Oxidation 62

3.2 Treatment of Manganese and Iron in Drinking Water 66

xu
 LIST OF FIGURES

Figure Descriptiou Page

No No

2.1 Dissolved manganese concentration with variation in depth of well 9

2.2 Manganese distributions with number of wells 21

2.3 Distribution of manganese in groundwater of Bangladesh 21

2.4 Smoothed map of manganese concentration in groundwater 22

2.5 Distribution of wells with respect to year of construction 27

2.6 Comparative Distributions of Arsenic and Manganese III 29

Groundwater of Bangladesh.

2.7 Comparative Distributions of Iron and Manganese in Groundwater 32

of Bangladesh.
3.1 Forms of Manganese in aqueous solution 41

3.2 pE-pH diagram for the simple ions and hydroxides of manganese 41

3.3 Eh-pH diagram (Pourbaix diagram) for manganese at 25° C 42

3.4 Effect of pH on Manganese Rate of Oxidation 45

3.5 Effect of Dissolved Oxygen on Manganese Oxidation 45

3.6 Effect of Manganese Concentration in Solution 47

3.7 Manganese oxidation in Bicarbonate Solution 47

3.8 Commonly Used Manganese removal Teclmiques 51

4.1 Mixing ofKMn04 with groundwater containing Manganese 73

4.2 Flow diagram of experimental procedure followed for evaluating 73

manganese removal by oxidation process
4.3 Removal of manganese by oxidation with KMn04 for different 78
initial manganese concentrations (stoichiometry ratio = 1.2; Initial
pH =7.7)
4.4 Residual manganese remaining in solution after filtration for 78
different initial manganese concentration after oxidation with
KMn04.(Stoichiometry ratio = 1.2)
4.5 (a) Color of precipitated manganese solids for groundwater with initial 80
manganese concentration 5 mg/L (pH '= 7.4; Eh = 230 mY)

X1l1
Figure Descriptiou Page
No No
4.5(b) Color of precipitated manganese solids for groundwater with initial 80
manganese concentration 10 mg/L (pH = 7.9; Eh = 560 mY)

4.6 Removal of manganese by oxidation with KMn04 for different pH 82

value (Initial Mn concentration = 2.0 mg/L; KMn04 stoichiometry
ratio = 1.2)
4.7 Removal of manganese by oxidation with KMn04 for different doses 83
of KMn04 expressed as multiple of stoichiometric ratio (Initial Mn
concentration = 2.0 mglL)

4.8 Removal of manganese by oxidation with KMn04 as a function of 85

initial manganese concentration for different settling times. (KMn04
stoichiometric ratio =1.2; Initial pH = 7.7)

4.9 Residual manganese concentrations as a function of settling time 85

after oxidation with KMn04 for different initial Mn concentrations.
(stoichiometric ratio = 1.2; Initial pH 7.7).

4.10 Color remaining in solution as a function KMn04 doses for different 86

settling times. (Stoichiometry ratio= 1.2; Initial pH 7.7)

4.11 Color remained in solution as a function of settlement time for different 87

amount of initial manganese concentration (Stoichiometry ratio = 1.2)

4.12 Color remammg in solution after passage through sand filter 88

column, as a function of pore volume of liquid passed. (Initial Mn
concentration = 2.0 mg/L; Stoichiometry ratio for KMn04 = 1.2)

4.13 Residual manganese remaining in solution after passage through 88

sand filter column, as a function of pore volume of liquid passed.
(Initial Mn concentration 2.0mglL; Stoichiometry ratio for KMn04 =
1.2)

4.14 Removal of manganese by oxidation with bleaching powder, as a 90

function of initial manganese concentration at different pH values
(Chlorine stoichiometry ratio = 1.2)

4.15 Residual chlorine remained in solution after oxidation with chlorine 90

of groundwater containing different amount of initial manganese for
amount of chlorine dose applied (Initial Mn Concentration = 2.0
mglL; Chlorine stoichiometry ratio = 1.2)

XIV
Figure Description Page
No No

4.16 (a) Color of precipitated manganese solids with initial manganese 91

concentration of 1.0 mglL (pH =7.4; Eh = -70 mV)

4.16 (b) Color of precipitated manganese solids with initial manganese 91

concentration of 5.0 mg/L (pH S.4;Eh -176mV)

4.16 (c) Color precipitated manganese solids with initial manganese 91

concentration of 10.0 mglL (pH 9.7; Eh -SOmV)

4.17 Removal of manganese by oxidation with chlorine at different pH 92

values (Initial Mn concentration =2.0 mg/L; Chlorine dose added =
2.5S mglL)
4.IS Removal of manganese by oxidation with chlorine as a function of 93
initial manganese concentration for different contact times (Initial
Mn conc. = 2.0 mglL; Initial pH = 10; chlorine stoichiometry
ratio=l.2)

4.19 Removal of manganese by aeration for different initial manganese 95

concentration at different pH values ( Initial Mn = 2.0 mglL)

4.20 Comparison of removal of manganese (after filtration) by aeration 95

and by bleaching powder oxidation at different pH values ( Initial
Mn = 2.0 mglL; for aeration contact time= 60 mins; for bleaching
powder oxidation contact time = 30 mins)

5.1(a) Percentage removal of manganese as a function of different initial 104

iron concentration present in groundwater (KMn04 stoichiometry
ratio ~ 1.2)

5.1(b) Percentage removal of iron as a function of initial iron concentration 104

present in groundwater (KMn04 stoichiometry ratio for Mn = 1.2)

5.2 Residual manganese in solution after filtration for different initial 105
iron concentration in water

5.3 Manganese removal as a function of initial iron concentration by 106

oxidation with KMn04 at two different stoichiometry ratio (S. R.) of
KMn04 (initial manganese concentration = 2.0 mglL)

5.4(a) Comparison of manganese removal with KMn04 as a function of 107

different initial iron concentration in low pH -low alkalinity water
and in normal pH - normal alkalinity condition.

xv
Figure Description Page
No No

5.4(b) Comparison of % iron removal as a function of different initial iron 108

concentration in low pH -low alkalinity water and that in normal pH
-normal alkalinity condition.(Initial Mn concentration = 2.0 mglL;
Stoichiometry ratio for KMn04 = 1.0)

5.5 Residual manganese remain in solution after oxidizing with KMn04 109
at low pH low-alkaline water for presence of dissolved iron at
different concentrations. (Initial Mn concentration = 2.0 mglL;
KMn04 stoichiometric ratio = 1.0)

5.6 Simultaneous removal of manganese and iron for different KMn04 110
dose expressed as a fraction of that required for complete oxidation
of both iron and manganese. (Initial Mn concentration = 2.0 mglL;
initial iron concentration = 5.0 mglL)

5.7 Manganese concentration remaining in solution after simultaneous III

removal of manganese and iron at different KMn04 dose expressed
as a fraction of that required for complete oxidation of both iron
and manganese. (Initial Mn concentration = 2.0 mg/L; initial iron
concentration = 5.0 mglL)

5.8 Removal of manganese with KMn04 oxidation in presence of iron 112

with pre-oxidation of iron by aeration (KMn04 stoichiometry ratio
for Mn= 1.0).

5.9 Simultaneous removal of manganese and iron by simple aeration for 113
different initial iron concentration (Initial Mn concentration = 2.0
mglL).

6.1 Experimental set-up for manganese removal by Coagulation 118

6.2 Removal of manganese by iron coagulation-adsorption process by 120

using dissolved FeS04 as coagulant with and without addition of
KMn04 (Initial Mn concentration =2.0 mglL; KMn04 stoichiometry
ratio =1; pH =7.6)

6.3 Residual manganese concentration in solution after coagulation with 122

FeS04, with and without pre-oxidation of dissolved manganese,
usmg KMn04. (Initial Mn concentration =2.0 mglL; KMn04
stoichiometry ratio =1; pH =7.6)

XVI
Figure Description Page
No No

6.4 Removal of manganese by iron coagulation-adsorption process 123

using Fe(III) as coagulant, with and without pre-oxidation of
dissolved manganese using KMn04. (Initial Mn concentration = 2.0
mglL; KMn04 stoichiometry ratio =1)
6.5 Comparison of Fe(III) and Fe(II) as coagulant in removing 124
manganese from water. (Initial Mn concentration = 2.0 mglL;
KMn04 stoichiometry ratio =1)

6.6 Effect of pH on removal of manganese by Fe (III) coagulation- 125

adsorption process using Fe(III) as coagulant without pre-oxidation
of disso.lved manganese using KMn04.( Initial Mn concentration =
2.0 mglL; KMn04 stoichiometry ratio =1)

XVIl
 LIST OF ABBREVIATIONS

AAS Atomic Absorption Spectrophotometer

AWWA American Water Works Association

BGS British Geological Survey

BAMWSP Bangladesh Arsenic Mitigation Water Supply Project

DFID Department for International Development

DPHE Department of Public Health Engineering

MW Molecular Weight

NBS National Hydrochemical Survey

rpm Revolution per minute

WHO World Health Organization

XV111
 CHAPTER!

INTRODUCTION

1.1 GENERAL

Groundwater extracted through hand pump tubewells is the primary source of safe
drinking water for the majority of the rural population of Bangladesh. Urban water supply
is also heavily dependent on groundwater extracted through deep tubewells.
Groundwater, being free from pathogenic microorganisms, receives wide public
acceptance. However, groundwater contains a wide range of dissolved minerals and
presence of certain dissolved constituents in excess of quantities recommended for
drinking purpose may make it unsuitable or less acceptable as a source of water supply.
The most widely reported groundwater-quality problems in Bangladesh include excessive
concentration of arsenic, iron and salinity (in coastal areas). However, a close inspection
of the available groundwater quality data (e.g., DPHE/ BGS, 2001) suggests that
excessive concentration of manganese is also a significant problem in many areas of the
country. The DPHE-BGS groundwater survey indicates that in Bangladesh manganese is
present in groundwater at relatively high concentration. In the nationwide groundwater-
quality survey (DPHE/ BGS, 2001), concentration of manganese up to 10 mg/l has been
found, with an average value of 0.5 mg/l.

Presence of excessive manganese in potable water may cause significant adverse health
impacts. It may also cause problems related to aesthetics and may cause precipitation in
the water distribution system. Evidence of manganese neuro-toxicity has been found in
people following long-term exposure to dusts containing manganese (such as people
working in mines). The World Health Organization (WHO) has a provisional health-
based guideline value of 0.5 mg/l for manganese in drinking water (WHO, 1993). The
WHO guideline value fromconsumer acceptability consideration is 0.01 mg/l. Bangladesh
Standard for manganese in drinking water is also 0.10 mg/l. At levels exceeding 0.1 mg/l,
manganese in water supplies stains sanitary ware and laundry and causes undesirable taste
in beverages. The presence of manganese in drinking water may lead to accumulation of
deposits in the distribution system. Even at a concentration of 0.02 mg/l, manganese may
form coating on distribution pipes, which may slough off as a black precipitate. In
 addition, certain nuisance organisms concentrate manganese and give rise to taste, odor
and turbidity problems in distribution system. Therefore, the acceptability threshold value
of manganese (0. I 0 mg/I) is five times less than the provisional health-based guideline
value (0.5 mg/I)

A large percentage of wells in Bangladesh fail to satisfy the drinking water standard for
manganese. Results of DPHE-BGS survey showed that about 39% of shallow tubewells
and 2% of deep tubewells exceeded the WHO health-based guideline value (0.01 mg/I),
whereas 79% of shallow tubewells and 22% of deep tubewells exceeded Bangladesh
Standard value. High concentrations of manganese are found in many areas of the
country, but particular high-manganese areas are seen in the current Brahmaputra and
Ganges floodplains. The distribution generally does not correspond with that of arsenic.
This means that groundwater with acceptable concentration of arsenic may not have
acceptable concentration of manganese (BGSlWaterAid, 2001). Again similar
comparison between iron and manganese shows that water with high manganese
concentration may not have high iron content. Distribution of manganese shows a
distinctly different pattern from those of arsenic and iron. People of areas with high iron
or manganese contents are more inclined to use the unprotected surface water sources,
many of which are dangerously contaminated and completely unsuitable for domestic use
without any treatment.

1.2 RATIONALE OF THE STUDY

High concentrations of manganese are found in many areas of the country, but particular
high-manganese areas are seen in the current Brahmaputra and Ganges floodplains. The
distribution generally does not correspond with that of arsenic. This means that
groundwater with acceptable concentration of arsenic may not have acceptable
concentration of manganese (BGS/WaterAid, 2001). Again similar comparison between
iron and manganese shows that water with high manganese concentration may not have
high iron content. Distribution of manganese shows a distinctly different pattern from
those of arsenic and iron. People of areas with high iron or manganese contents are more
inclined to use the unprotected surface water sources, many of which are dangerously
contaminated and completely unsuitable for domestic use without any treatment (Shahid,
1998).

2
 Although significant research works have been carried out on removal of arsenic and iron
from groundwater, relatively little studies (e.g., Ali et ai, 2001; BAMWSPI DFIDI Water
Aid, 2001; Tahura et ai. 2001) has been done on the removal of manganese from
groundwater of Bangladesh. Manganese removal processes include ion exchange,
aeration, greensand filtration, chemical oxidation and subsequent filtration, and biological
manganese removal (Varner, 1992). Chemical oxidation and subsequent filtration is by
far the most widely used technique for manganese removal from water (Raveendran et ai,
2001). A number of water quality parameters, such as pH, Eh, iron, organic matter, etc.
can affect the efficiency of,manganese removal from water (Seeling et aI., 1992). Choice
of oxidizing agent is also an important consideration. Potassium permanganate and
bleaching powder is the most commonly used oxidizing agent in Bangladesh. Although
potassium permanganate is widely used, color development due to its presence is a
concern. Effectiveness of commonly available oxidizing agents in removing manganese
in groundwater and their impact on the overall removal process needs to be studied in
more details (Tahura et ai, 2001). Moreover, color produced from permanganate
oxidation of manganese may be removed by using commonly used color removal
techniques (e.g., sand filtration).

It should be noted that processes typically used for arsemc and iron removal, e.g.,
coagulation-adsorption- coprecipitation (for arsenic), adsorptive-filtration (for arsenic),
and oxidation-precipitation-filtration (for iron), can possibly be used to remove
manganese from water. In the project conducted by BAMWSPI DFIDI Water Aid (2001),
it has been found that some household treatment methods for arsenic removal (e.g., Aclan
Activated Alumina Method, Sono 3-Kolshi Method) were capable to remove manganese
in addition to arsenic. But however, in the study areas, manganese concentration in the
feed water to this treatment plants were much below I mglL. Where as mentioned earlier
in the most of the areas in Bangladesh, manganese concentration is above I mglL.
Therefore, applicability of these methods of arsenic and iron removal needed to be
assessed further for higher manganese concentrations.

Presence of iron in water can also considerably influence manganese removal from water
by the oxidation process. Excess iron present in water can act as a coagulant and thereby
may assist the settling of manganese solids in water. Moreover, iron hydroxide flocs

3
present in water may also act as an adsorbent for dissolved manganese. Thes

'" be ,rudie"" more '''ml, I

1

1.3 OBJECTIVES OF THE STUDY:

The overall objective of this study was to evaluate the effectiveness of oxidation and
adsorption processes in removing manganese from groundwater. The specific objectives
of this research work include:
(1) Evaluation of the effectiveness of the commonly available oxidizing agents
(potassium permanganate and bleaching power) in Bangladesh in removing
manganese from groundwater;
(2) Better understanding of the processes and parameter (e.g., pH, oxidant dose, Eh,
settling time etc) affecting removal of manganese from groundwater
(3) Evaluation of the effectiveness of aeration in removal of manganese present III

groundwater.
(4) Assessment of the effectiveness of sand filtration in removing color produced due to
the presence of permanganate;
(5) Evaluation of effectiveness the of potassium permanganate III removlllg both
dissolved manganese and iron in groundwater; and
(6) Assessment of the effectiveness of coagulation-adsorption-coprecipitation processes
(using both Fe(I!) and Fe(III) salts) in removing manganese from groundwater.

1.4 ORGANIZATION OF THE THESIS

The study is presented in seven chapters, the first of which is general introduction.
Chapter two and three contain a brief and selective review of relevant literature. In these
chapters occurrence of manganese in groundwater, problem associated with manganese in
groundwater of Bangladesh, manganese chemistry, and commonly used manganese
removal techniques are discussed.

In Chapter Four, laboratory studies carried out on oxidation of manganese in groundwater

using chemical oxidants (potassium permanganate and bleaching powder) and aeration
have been described. Effect of pH, oxidant dose, manganese concentration, and settling
and contact times on oxidation processes are explained and discussed.

4
In Chapter Five, experiments on oxidation of natural groundwater containing both
dissolved iron and manganese are described with a view to find out the effect of the
presence of iron on manganese removal by chemical oxidation as well as by aeration.

In Chapter Six, laboratory studies conducted to evaluate the effectiveness of iron

coagulation (using both Fe(II) and Fe(III) salts) in removing manganese in groundwater
are described. Effect of pH, coagulant dose, and manganese concentration in coagulation
process are also discussed.

Conclusions and recommendations for future studies are given III Chapter Seven.
Attempts are made to draw conclusions from various findings of the study.
Recommendations presented in this chapter provide a basis for further study.

'5
 CHAPTER 2
OCCURRENCE OF MANGANESE

2.1 INTRODUCTION

Manganese in objectionable concentration has been detected in many water supply

sources. Alone or in combination with other elements, manganese is capable to cause
serious impairment of water quality. The problem is severe in context of Bangladesh, as
groundwater of Bangladesh contains significantly high amount of dissolved manganese
and at the same time groundwater is the vital source for safe water supply for drinking as
well as other domestic and agricultural purposes. The number of tubewells in Bangladesh
is not known but estimates put the number at around 6- 11 million. The vast majority of
these are private tubewells, which penetrate the shallow alluvial aquifers to depths
typically of 10-60 m. Irrigation boreholes typically tap shallow as well as deeper aquifers
in the region of 70- 100 m depth. In some areas, notably the south and the Sylhet Basin of
North-East Bangladesh, deep tubewells abstract groundwater from depths of 150 m or
more. In the south, the deep tubewells have been installed to avoid high salinity at
shallower levels (DPHE/ BGS, 200 I). Shallow hand-dug wells occur in some areas,
though they are much less common than tubewells. In many areas in Bangladesh,
concentration of manganese is at much higher level than the limit acceptable to the rural
people. People of such areas generally refuse to use tubewells water and inclined to use
unsafe and impure pond and river water (DPHE/ BGS, 2001).

In this chapter, relevant literatures on occurrence and distribution of manganese in

groundwater, and some common removal techniques have been reviewed. Possible health
and other effects of manganese have aJso been focused.

2.2 OCCURRENCE OF MANGANESE

2.2.1 Physical and Chemical states

Manganese is a naturally-occurring element that can be found ubiquitously in the air, soil,
and water. Manganese is the third most abundant metals on the earth's surface (9.5 x 102
ppm), making up approximately 0.1% of the earth's crust. Manganese is not found
 naturally in its pure (elemental) form, but is a component of over 100 mine
2004). Manganese is naturally occurring in many surface and ground waf
in soils that may erode into these waters. However, human activities are ~>J>~. .. ..

for much ofthe manganese contamination in water in some areas. ,. '-- .

Manganese is typically released from the rock strata and enters surface and ground water
during mining. Manganese is dissolved in anoxic and acid water. Homogeneous
precipitation ofMn(H) as an oxide phase does not occur:below pH 8 but Mn(H) oxidation
does occur in the presence of different mineral surfaces and/or via bacterial processes
between pH 6 and 8. It is also known that bacterially mediated oxidation of dissolved
manganese (Davidson et ai, 1989).

Manganese is an important component of metals redox chemistry in soils, and marine and
fresh water environments (Martin, 2003). Among the microorganisms capable of
oxidizing Mn (H) are bacteria, algae, yeast, and fungi (Ehrlich et ai, 1996). Because
aqueous environments are dynamic, and water chemistry, aquatic plants, and algal
communities can change with time, even when manganese (H) precipitates as stable oxide
phases, these precipitates can be resolubilized if new reducing Environments form. The
chemical and physical properties of elemental manganese and common manganese
compounds are presented in Table 2.1 (USEPA,2001).

~anganese problems are most likely to develop in water from wells with high carbonate
and low oxygen as shown in the middle well in Figure 1. Problems occur when this type
of water is pumped to the surface. The chemical equilibrium is changed upon exposure to
the atmosphere. The end result is precipitation of manganese compounds in plumbing, on
fixtures, and on clothing, dishes, and utensils. However, the Amount of manganese
dissolved in water often follows a trend of low to high back to low again as depth of the
well increases (Seelig et ai, 1992). A typical representation of dissolved manganese
concentration with variation in well depth is shown in Figure 2.1.

7
Table 2.1 Physical and Chemical Properties of Manganese and that of Some Common
Manganese Compounds

Manganese Manganese Potassium

Manganous Manganese
Name Manganese (II, III) Dioxide Pennanganate
Chloride Sulfate
Oxide
Manganese
Manganese Trimanga-
Elemental peroxide;
chloride; Manganese nese Permanganic
manganese; manganese
manganese sulfate; Tetroxide; acid;
Colloidal binoxide;
dichloride; Sulfuric Mangano potassium salt;
Synonyms manganese; manganese
manganese acid, Manganic chameleon
Cutaval; black;
bichloride; manganese oxide; material
Magnacat; battery
manganese (II) salt Manganese
Tronamang manganese;
(II) chloride Tetroxide
oVTolusite

Valence 0 +2 +2 +2 and +3 +4 +7

Chemical MnCI,
Mn MnSO, Mn,O, MnO, KMnO,
Formula

Molec.
54.9 125.84 151.00 228.81 86.94 158.04
Weight

Pink Pale rose- Purple

Color Black Black
red

Physical
Solid Solid Solid Solid Solid Solid
State
Melting
1244 650 700 1564 535 240
Point(oC) .

Boiling Decomposes
PointCC)
1962 1190
at 850°C
- - -

Solubility
In water Decomposes 723 (25°C) 120 (25°C) Insoluble Insoluble 63.8 (20°C)
(g/100 mI)

Taste
Threshold
- - - - - -
.

Odor
Threshold - - - - - -
(air)
- No date available. (Source: USEPA, 2004)

8
 Low Mn High Mn Low Mn
Well Well Well
•.••
p

Non-water beal11ng
su rfac'8 mal.s rial

~.
. '.' ... .:,'.'
~: ;-.
. : i~ : :

Aquifer f.. ::; :~I:f ,'U,Ti~n2+

..':"~ -. ~.
-,.,.
,: Mn (s)

Impermeable ~.;
matenalls ~ f. •
tzzzI Oxygenated zone
[]]] Ae.duced zone {low oxygen)

Figure 2.1 Dissolved manganese concentration with variation in depth of well

(Seelig et ai, 1992)

9
2.2.2 Sources of Manganese in the Environment

Manganese compounds are widely distributed in air, soil, and water. Sources of
atmospheric manganese include industrial emissions, fossil fuel combustion, and erosion
of manganese-containing soils.

Water

Manganese is naturally occurring in many surface and ground water sources and in soils
that may erode into these waters. However, human activities are also responsible for
much of the manganese contamination in water in some areas. Ambient manganese
concentrations in seawater have been reported to range from 0.4 to 10 /-!g1L(ATSDR,
2000), with an average of about 2 /-!g/L (US EPA, 2004). Levels in freshwater typically
range from 1 to 200 /-!g1L (USEPA, 2004). The United States Geological Survey's
National Ambient Water Quality Assessment (NA WQA) has gathered limited data since
1991 on representative study basins around the U.S. This report indicates a median
manganese level of 16 /-!g1Lin surface waters, with 99th percentile concentrations of 400
'to 800 /-!g1L(USEPA, 2004). Higher levels in aerobic waters are usually associated with
industrial pollution. Manganese can be released to water by discharge from industrial
facilities or as leachate from landfills and soil (Gregory et aI, 1996). In the USA, reported
industrial discharges in 1991 ranged from 0 to 17.2 t for surface water, from 0 to 57.3 t
for transfers to public sewage, and from 0 to 0.114 t for underground injection
(Barceloux, 1999).

An estimated total of 58.6 t, or 1% of the total environmental release of manganese in the

USA, was discharged to water in 1991 (Casale et ai, 2001). Overall, the detection
frequency of manganese in U.S. groundwater is high (approximately 70% of sites assayed
have measurable manganese levels) due to the ubiquity of manganese in soil and rock. In
another worldwide survey, manganese is detected in about 97% of Surface water sites (at
levels far below those likely to cause health effects) and universally in sediments and
aquatic biota tissues (at levels which suggest that it does not bioaccumulate (USEPA
1984).

10
Soil

Manganese constitutes approximately 0.1% of the earth's crust, and is a naturally

occurring component of nearly all soils (ATSDR, 2000). Natural levels of manganese
range from less than 2 to 7,000 ppm, with a geometric mean concentration of 330 ppm
(Shacklette and Boemgen, 1984). Accumulation of manganese occurs in the subsoil
rather than on the soil surface (ATSDR, 2000). Land disposal of manganese-containing
wastes is the principal source of manganese releases to soil (Barceloux, 1999). An
estimated 60-90% of soil manganese is associated with the sand fraction (WHO, 1981, as
cited in ATSDR, 2000).

Air

Air levels of manganese compounds vary widely depending on the proximity of point
sources such as ferro alloy production facilities, coke ovens, or power plants. The main
sources of manganese releases to the air are industrial emissions, combustion of fossil
fuels, and re-entrainment of manganese-containing soils (Lioy, 1983; USEPA, 1983,
1984, 1985a, 1985b). Average ambient levels near industrial sources have been reported
to range from 220 to 300 ng MnJm3, while levels in urban and rural areas without point
sources have been reported to range from 10 to 70 ng MnJm3 (Barceloux, 1999). Air
erosion of dusts and soils is also an important atmospheric source of manganes~, but no
quantitative estimates of manganese release to air from this source were identified
(USEPA; 1984). Volcanic eruptions can also release manganese to the atmosphere
(Hasan, 2004).

Food

Manganese is found in a variety of foods including many nuts, grains, fruits, legumes, tea,
leafy vegetables, infant formulas, and some meat and fish. Food is the most important
source of manganese exposure in the general population (ATSDR, 2000; 10M, 2002;
USEPA, 2003a). Particularly green vegetables (2 mglkg), nuts (14.9 mglkg), bread (8
mglkg) and other cereals (6.81 mglkg) have considerable amount of manganese. Infant
formulas contain 50 to 300 !J.g/L manganese (Collipp et aI., 1983), compared to human

11
milk which contains approximately 3.5 to IS Ilg/Lmanganese (ATSDR, 2000; U.S. EPA,
1997). Tea is a rich source of manganese, containing 2.71 mglkg and is the largest
contributor to manganese intake (Kondakis, 1989). Therefore heavy tea drinkers may
have a higher manganese intake than the general population. An average cup of tea may
contain 0.4 to 1.3 mg manganese (ATSDR, 2000).

Industrial Source of Manganese

Manganese compounds are produced from manganese ores or from manganese metal.
Metallic manganese (ferromanganese) is used principally in steel production along with
cast iron and superalloys to improve hardness, stiffuess, and strength (NTP, 1993;
USEPA, 1984). Manganese compounds have a variety of uses. The most common uses of
manganese compounds are presented in Table 2.2 (Hasan, 2004).

Table 2.2 Uses of Some Common Manganese Compounds

Compound Use
Methylcyclopentadienyl manganese
tricarbonvl cMMT) Fuel additive

Manganous carbonate Ferrites; animal feeds; ceramics etc

Catalyst in organic compound chlorination; trace
Manganese chloride mineral supply for animal feed; brick_colorant;dye;
dry-cell batteries; linseed oil drier; disinfecting;
Manganous acetate Mordant in dyeing; drying agent for paint and varnish;

Manganese ethylenebisdithiocarbamate Agricultural fungicide

Manganese oxide Ferrites; ceramics; fertilizer; livestock feed additive

Manganese phosphate Ingredient of proprietary solutions for phosphating iron

and steel
Livestock feed additive; fertilizer; glazes; varnishes;
Manganese sulfate
ceramics; fungicides
Manganous trifluoride Fluorinating agent in organic chemistry
Drying agent for varnish and oil; linseed oil drier;
Manganese borate
leather industry
Porcelain colorants; manufacture of reagent grade
Manganous nitrate
manganese dioxide
Manganese dioxide (electrolytic Dry-cell batteries; matches; fireworks; porcelain; glass
manganese, pyrolusite) bonding materials; amethyst glass; oxidizer
Oxidizing agent; water and air disinfectant; antialgal
Potassium perrnanganate agent; metal cleaning, tanning, and bleaching agent;
fresh flower and fruit preservative
Sources: u.S. EPA (1994a); ATSDR (2000); Merck (1983).

12
2.2.3 Environmental Fate

Manganese compounds may be present In the atmosphere as suspended particulates

resulting from industrial emissions, soil erosion, volcanic emissions, application of
manganese-containing pesticides, and the burning of MMT -containing gasoline (USEP A,
2004). Early analysis of emissions suggested that manganese from combustion of MMT
is emitted primarily as manganese tetroxide (Mn304) (Ter Haar et a!., 1975, as cited in
ATSDR, 2000). However, more recent testing suggests that when very low levels of
MMT are combusted (i.e., concentrations comparable to the currently allowed levels),
manganese is emitted primarily as manganese phosphate and sulfate. The reported formal
charge of the emitted manganese is +2.2, with a mass median aerodynamic diameter of 1
to 2 microns (Ethyl Corporation, 1997, as cited in Lynam et a!., 1999). Uncombusted .
MMT rapidly decomposes to manganese oxide, carbon dioxide, and organic compounds
in the atmosphere and has a half-life of only a few seconds in the presence of sunlight
(Lynam et a!., 1999; Zayed et a!., 1999). Because particle size is small, atmospheric
manganese distribution can be widespread. These particles will eventually settle out into
surface waters or onto soils via the process of dry deposition. Little information is
available on the chemical reactions of atmospheric manganese, but it is expected to react
with sulfur and nitrogen dioxide. The half-life of manganese in air is only a few days
(ATSDR,2000).

Transport and partitioning of manganese in water is dependent on the solubility of the

manganese form. In surface waters, manganese occurs in both dissolved and suspended
forms, depending on such factors as pH, anions present, and oxidation-reduction potential
(ATSDR, 2000). Often, manganese in water will settle into suspended sediments.
Anaerobic groundwater often contains elevated levels of dissolved manganese. The
2
divalent form (Mn +) predominates in most water at pH 4-7, but more highly oxidized
forms may occur at higher pH values or result from microbial oxidation (ATSDR, 2000).
Manganese in water can be significantly can bioaccumulate in lower organisms (e.g.,
phytoplankton, algae, mollusks, and some fish), but not in higher organisms, and
biomagnification in food-chains is not expected to be significant (ATSDR, 2000).
Bioconcentration factors (BCFs) of 10000-20000 for marine and freshwater plants, 2500-
6300 for phytoplankton, 300-5500 for marine algae, 800-830 for inter-tidal mussels, and
35-930 for fish have been estimated (Folsom et a!., 1963; Thompson et a!., 1972). The

13
high reported BCFs probably reflect the essentiality of manganese for a wide variety of
organisms; specific uptake mechanisms exist for essential elements. Little information is
available on the biodegradation of manganese-containing compounds in water, but factors
such as pH and temperature are important for microbial activities (Hurley, 1984).

2.2.4 Manganese Intake

Adequate Intake (AI) values have been determined for manganese by the Food and
Nutrition Board of the Institute of Medicine has fixed the amount of manganese intake
per day for different gourp of people. The tabular form is shown in Table 2.3 (US EPA,
2004)

Table 2.3 Adequate Manganese Intakes for Men, Women and Children

Age Group Males Females

Infants, 0-6 months 3 Ilg/day 3 Ilg/day
Infants, 7-12 months 0.6 mg/day 0.6 mg/day
Children, 1-3 years 1.2 mg/day 1.2 mg/day
Children, 4-8 years 1.5 mg/day 1.5 mg/day
Boys, 9-13 years 1.9mg/day --

Boys, 14-18 years 2.2 mg/day --

Girls, 9-18 years -- 1.6 mg/day
Adults, 19 years 2.3 mg/day 1.8 mg/day
Women- pregnant (lactating) -- 2 mg/day (2.6 mg/day)
Source: (USEP A, 2004)

2.3 UNDESIRABLE EFFECTS OF MANGANESE

Manganese is one of the most difficult elements to remove from surface waters. Presence
of manganese primarily interferes with water uses. In higher concentrations manganese
causes the following problems (Seelig et ai, 1992):
1. Staining: At levels exceeding 0.1 mg/I, manganese in water supplies stains sanitary
ware and laundry. Where the concentration of manganese is high, the color of the
staining tends toward more black or gray.
2. Taste: Manganese causes a metallic or vinyl type taste in the water. .

;
14
3. Appearance: Manganese will often give an oily appearing, "crusty" sheen to the water
surface. (Oil does not appear "crusty" when disturbed, but "feathers out" like a
rainbow).
4. Sulfur Taste: Hydrogen Sulfide, which causes a characteristic "rotten egg" odor, can
also be liberated by the same conditions (i.e. low dissolved oxygen and low pH) that
cause manganese to dissolve in water. Hydrogen sulfide is frequently encountered in
water with excessive manganese. Some of the treatment methods used to remove iron
and manganese will also "remove" hydrogen sulfide gas. Presence in high
concentrations, manganese may cause an unpleasant metallic taste to the water
(Raveendran et ai, 2001).
5. Deposits accumulation: The presence of manganese in drinking water may lead to
accumulation of deposits in the distribution system. Even at a concentration of 0.02
mg/I, manganese may form coating on distribution pipes, which may slough off as a
black precipitate.
6. Clogging: Manganese supports the growth of manganese bacteria. This non-health
related bacteria can clog strainers, pumps, and valves. Periodic or continuous
chlorination is the best means to control manganese bacteria. Once present,
manganese bacteria are difficult to purge from a well.

In agricultural fields, where groundwater is used for irrigation purpose, presence of

manganese in water causes irrigation difficulties. Dissolved manganese in water becomes
insoluble either by bacteria mediated oxidation or by atmospheric oxidation. Manganese
solids the deposits and results in clogging in different parts of the pump. Oxidized
manganese produced by iron bacteria often results in the formation of an ochre sludge or
slime mass, which is capable of plugging the entire irrigation system (Boman et aft
,1999).

2.4 HEALTH EFFECTS OF MANGANESE

Manganese is an essential element for many living organisms, including humans. It is

necessary for proper functioning of some enzymes (manganese superoxide dismutase)
and for activation of others (kinases, decarboxylases, etc) (USEP A, 2004). The National
Academy of Science set an adequate intake for manganese at 2.3 mg/day (for men) to 1.8
mg/day (for women), with an upper limit of 11 mg/day (NRW A, 2004).

,.'
15
Adverse health effects can be caused by inadequate intake or over exposure. Manganese
deficiency in humans appears to be rare because manganese is present in many common
foods. Animals experimentally maintained on manganese-deficient diets exhibit impaired
growth, skeletal abnormalities, reproductive deficits, ataxia of the newborn, and defects in
lipid and carbohydrate metabolism (Keen et aI., 1999; Hurley and Keen, 1987; U.S. EPA,
1984). The health effects from over-exposure of manganese are dependent on the route of
exposure, the chemical form, the age at exposure, and an individual's nutritional status.
Irrespective of the exposure route, the nervous system has been determined to be the
primary target with neurological effects generally observed (USEPA, 2004). Exposure to
toxic levels of manganese affects the nervous system, and may cause neurological and
behavioral symptoms, including dementia, anxiety, and a "mask-like" face (NRW A,
2004). These symptoms are generally the result of very high exposures via inhalation, as
might occur in an industrial setting.

2.4.1 Short-term Exposure Studies

Neurological

Kawamura et al. (1941) reported health effects resulting from the ingestion of
manganese-contaminated well water for an estimated 2-3 months by 25 individuals. The
source of contamination was identified as leachate from approximately 400 dry cell
batteries buried near the drinking water well. The concentration of manganese in the well
water was in average 28 mg Mn/L or higher. Assuming a daily water intake of 2L, with a
minimum of 2 mg Mn from food, a dose of at least 58 mg Mn/day is estimated. This
exposure level is quite uncertain and it is estimated that it is around 25-30 times the level
considered to be safe and adequate by the Food and Nutrition Board of the Institute of
Medicine (10M, 2002).

Health effects reported by Kawamura et al. (1941) included lethargy, increased muscle
tonus, tremor and mental disturbances. Out of 25 people examined, 15 had symptoms.
Five cases were considered severe, 2 cases were categorized as moderate, and 8 cases
were described as mild. The most severe symptoms were observed in the elderly.
Younger people were less affected, and symptoms of intoxication were completely absent
in young children (age I to 6 years) (USEPA, 2004).

16
"
2.4.2 Long-term Exposure Studies
Neurological

The neurological effects of inhaled manganese have. been well documented in humans
chronically exposed to elevated levels in the workplace (ATSDR, 2000; Canavan et aI.,
1934; Cook et aI., 1974; Roels et aI., 1999). The syndrome known as "manganism" is
caused by exposure to very high levels of manganese dusts or fumes and is characterized
by a "Parkinson-like syndrome" including weakness, anorexia, muscle pain, apathy, slow
speech, monotonous tone of voice, emotionless "mask-like" facial expression, and slow
clumsy movement of the limbs. In general, these effects are irreversible (USEPA, 2004).

By the oral route, manganese is often regarded as one of the least toxic elements, .
although there is some controversy as to whether the neurological effects observed with
inhalation exposure also occur with oral exposure .. Several case reports of oral exposure
to high doses of manganese have described neurological impairment as an effect, but the
quantitative qualitative details of exposure necessary to establish direct causation are
lacking. An individual who took large mineral supplements over several years displayed
symptoms ofmanganism (Banta and Markesbery, 1977).

Results from studies of an Aboriginal population in Groote Eylandt have been cited as
additional evidence for a relationship between elevated manganese exposure, violent
behavior, and adverse health effects (COMA, 1998).

Cancer and Mutagenicity Studies

Mutagenicity

The genotoxic potential of high manganese exposure in humans is not known (IPCS,
1999). Elias et al. (1989) found an increase in the incidence of chromosomal aberration in
metal active gas welding workers who had been welding for 10-24 years. Occupational
exposure to nickel, as well as manganese, was reported. Since nickel is known to cause
chromosomal aberration via inhalation, the results could not be attributed solely to the
influence of manganese (USEP A, 2004)

17
Carcinogenicity
No studies are available on the potential carcinogenicity of high exposure to manganese
in humans (ATSDR, 2000). USEP A believes that the available data on occurrence,
exposure, and other risk considerations suggest that regulating manganese does not
present a meaningful opportunity to reduce health risk. Only the non-enforceable

.
secondary MCL for manganese of 0.05 mg/L will remain in effect (NRW A, 2004) .

2.4.5 Standards and Guideline Values for Manganese

The Environmental Protection Agency (EPA) standards for drinking water fall into two
categories: Primary Standards and Secondary Standards. Primary Standards are based on
health considerations and are designed to protect people from three classes of pollutants:
pathogens, radioactive elements and toxic chemicals. Secondary Standards are based on
taste, odor, color, corrosivity, foaming and staining properties of water. Iron and
manganese are both classified under the Secondary Maximum Contaminant Level
(SMCL) standards. The SMCL for manganese in drinking water is 0.05 mg/I (ppm)
(Lemley, 1999). The World Health Organization (WHO) has a provisional health-based
guideline value of 0.5 mg/l for manganese in drinking water (10M, 2002). The WHO
guideline from aesthetic perspective is 0.10 mg/L. Bangladesh Standard for manganese in
drinking water is also 0.10 mg/l.

2.5 OCCURRENCE OF MANGANESE IN GROUNDWATER OF BANGLADESH

The most widely reported groundwater-quality problems in Bangladesh include excessive

concentration of arsenic, iron and salinity (in coastal areas). Although Considerable
research works have been carried out on occurrence and removal of arsenic as well as of
iron, relatively little work has been done on the manganese problem in groundwater of
Bangladesh. The National Hydrochemical Survey (NHS) conducted in 2000 by British
Geological Survey (BGS) in collaboration with DPHE and DFID (DPHEIBGS, 2001)
provides the most comprehensive information on manganese in groundwater of
Bangladesh. Using this survey data various aspects of manganese distribution and
occurrence in groundwater of Bangladesh has been prepared by Hasan (2004).
Groundwater survey indicates that iron and manganese are present in high concentrations.
Concentration of manganese in groundwater has been found as high as 10 mg/l, with an

18
average value of 0.5 mg/L. The high values are related to the anaerobic conditions
dominant in the aquifers (BGSI WaterAid, 2001). A large percentage of wells in
Bangladesh fail the health guideline value specified by WHO (0.5 mglL). A total of 938
samples (26,5% of all samples) had manganese concentration below O. I mglL.

Percentile distribution of manganese in groundwater of Bangladesh is summarized in

Table 2.4. The table reports the number of well with specific manganese concentration
range. Table 2.4 indicates the median concentration of manganese is about 0.3 mgl!.
Distribution cif manganese in number of wells is shown in Figure 2.2. From Fig. 2.2 it can
be noted that the most frequent concentration occurrence of manganese lay within the
concentration range of o. I to 0.5 mg/L, accounting for about 38.6% of all sample wells.
About 35% of sample wells exceed WHO. health based guideline value and about 73%
exceeds Bangladesh guideline value.

The distribution of manganese in groundwater of Bangladesh is shown in Figure 2.3.

From Figure 2.3 it appears that high concentrations of manganese are found all over the
country and the distribution are very scattered. It also indicates that groundwater iil .
coastal region of south part of the country has lower manganese concentration « 0.1
mglL). However, the regional trends, more precisely shown in Figure2.4, indicate that the
central, north and south-east regions of Bangladesh have higher concentrations of
manganese.

19
 Table 2.4 Mn Concentration Expressed as Percentile
. (n=3534)

Percentile Mn Concentration less or

eQual to (ml!!L)
10 0.03
20 0.068
30 0.123
40 0.185
50 0.287
60 0.428
70 0.613
80 0.892
90 1.365
95 1.9
99 3.697
(Source: Hasan, 2004)

00.001-0.1
C10.1-0.5
00.5-1.0
01.0-1.5
IJ 1.5-2.0
02.0-3.0
03.0-4.0
04.0-5.0
IJ >5.0

Figure 2.2: Manganese distributions in groundwater

20
 Manganese
(mg L-1)
• <0.1
• 0.1-0.3
• 0.3-0.75
• >1

India

~ 20
Bay of Bengal
10
~- 200 km ----<
0- -
'Groundwater Studies of Arsenio Contamination in Bangladeshi
DPHElBGSlDFID (2000)

Figure 2.3 Distribution of manganese in groundwater of Bangladesh

(8ouree: BGSIDPHE. 2001)

21
 Manganese (mg L-1)
• <0.18 • 0.5-0.6
• 0.18-0.28 • 0.6-0.7
• 0.28-0.36 • 0.7-0.9
• 0.36-0.44 • 0.9-1.2
• 0.44-0.5 • >1.2

Bay of Bengal

1-- __ 200 km ---<

'Groundwater Studies of ASlnic Contamination in Bangladesh'

I)PHEIB<7S1DFII) (2000)

89°
Figure 2.4 Smoothed map of manganese concentmtion in groundwater
(Source: BGSIDPHE, 2001)

22
The average manganese concentration distribution for each individual division is
presented in Table 2.5. It shows that Rajshahi division has the highest average manganese
concentration (0.731 mg/L), where 521 wells among the 1072 wells exceed the WHO
health-based guideline value (0.5 mg/L) and 84% of the test wells (about 900 wells)
exceed Bangladesh guideline value of manganese. The lowest average manganese
concentration is found in Barisal division. The average concentrations for each division
above the WHO guideline value lie between 0.738-1.03 mg/L (Hasan, 2004).

Table 2.5 Manganese Occurrence Summaries for Different Divisions

Concentration below WHO Concentration above WHO health- ,

health- based guideline value based guideline value
Total

Division Mean Max. Mean Mean

No of Percent- No of Percent- No of
cone. cone. cone. Cone.
wells age (%) wells age (%) Wells
(mg/L) (mg/L) (mg/L) (mg/L)
Barisal 277/ 7.8 0.055 18 0.5 2.07 1.03 295 0.115
Chillagong 311 J 8.8 0.19 134 3.8 4.67 1.10 445 0.645
Dhaka 600/ 17.0 0.175 388 11.0 8.39 1.37 988 0.646
Khulna 330. 9.3 0.181 144 4.1 3.24 1.10 474 0.460
Rajshahi 551 15.6, 0.202 521 14.7 9.98 1.29 1072 0.731
Sylhet 216 6.1 0.202 44 1.2 2.06 0.74 260 0.292
Total 2285 64.7 0.172 1249 35.3 3.98 1.25 3534 0.554
(Source; Hasan, 2004)
~

The most contaminated district with respect to manganese Kurigram, with average
manganese concentration of .1.336 mg/L, of which about 80% wells have manganese
concentration above of WHO health-based guideline value. Manganese concentration
distribution in the most contaminated 17 districts is shown in Table2.6. About 80% or
more. of the tubewells in. these 17 districts contain manganese above the Bangladesh
standard with average concentrations exceeding the WHO health-based guideline value.
The highest manganese concentration (about 10 mg/L) is found in Jaipurhat. But this
concentration was found in only one testwell~

23
 Table 2.6 Manganese Statistics for the 17 Most Contaminated Districts

Percentage of wells in
MinimumMn Percentage of wells
Avg Max specific Mn cone. range
No. Cone. (mg/L) exceeding
Mn Mn (mg/L)
District of
cone. cone. Bangladesh
WHO
Wells guideline
(mg/L) Shallow Deep (mg/L) <0.1 0.1.0.5 0.5.2.0 >2.0 Standard (0.1
value
mglL)
0.5 moiL'

iKurigram 77 1.336 0.Dl5 <.001 5.23 4 19 56 21 96 77

~arayanganj 30 1.276 0.003 0.133 8.39 7 27 43 23 93 67

Sirajganj 89 1.249 0.016 <.001 3.77 4 15 63 18 . 96 81

iRajbari 34 1.195 0.085 0.943 3.87 3 9 76 12 97 88

Pabna 78 1.083 0.111 <.001 5.54 0 25 65 10 100 74

Narsingdi 56 0.979 0.009 <.001 4.03 18 23 50 9 82 59

Magura 32 0.971 0.087 <.001 3.14 3 25 63 9 97 72
Tangai1 91 0.922 0.006 <.001 3.8 12 23 56 9 88 64
IT'alpur h at
40 0.907 0.079 <.001 9.98 3 47 42 8 98 48
Manikganj 47 0.868 0.001 <.001 6.03 22 37 28 13 79 40
Rajshahi 78 0.859 0.017 <.001 3.82 13 23 55 9 87 64
Natore 51 0.841 0.145 <.001 2.13 0 24 73 4 100 76
Sherpur 51 0.814 0.032 0.404 . 7.83 2 53 41 4 98 45
caridpur 63 0.806 0.008 0.133 3.83 13 35 43 10 87 52

Gaibandha 71 0.787 0.038 <.001 4.59 8 28 61 3 92 63

ama1pur 63 0.771 0.014 <.001 4.6 10 38 48 5 90 52
Meherpur 15 0.743 0.01 <.001 2.35 7 20 67 7 93 73
(Source: Hasan, 2004)

24
Distribution and concentration of manganese in groundwater is significantly influenced
by the depth of well. From the National Hydrochemical Survey (BGSIDPHE 2001),39%
of shallow tubewells and 2% of deep tubewells exceeded the WHO health-based
guideline value. The distribution of manganese concentration with type of well is shown
in Table2.7 & Table 2.8.

Table 2.7 Percentages of Wells in Shallow and Deep Wells Above and Below
'Bangladesh Standard and WHO Health-based Guideline Yalue

Bangladesh guideline value WHO health-hased guideline

Well type (0.10 mg/L) value (0.50 mg/L)
Below Above Below Above
Shallow 21 79 61 39
Deep 78 22 98 2
(Source: Hasan, 2004)

Table 2.8 presents a summary of Manganese concentration in different depth intervals.

The highest average concentration (0.717 mg/L) occurs for wells of less than 15m depth.
There is little difference in average concentration of manganese for well with depth up to
90m; but for wells above 90m depth, average concentration falls sharply. From Table 2.8,
it is evident that shallow wells with depth up to 60m, are the most vulnerable for
manganese contamination (Hasan, 2004).

Table 2.9 shows a relationship between year of construction of shallow wells and
different manganese ranges. Deep wells have been excluded from this table because most
of their manganese concentrations are below the national/WHO standards. From Table
2.9, recent wells appear to be more contaminated.

The trend appears to be alarming because more wells are being constructed recently in
areas which are prone to be manganese contaminated. Figure-2.5 shows the distribution
of wells with respect of year of construction over the country.

25
 Table 2.8 Manganese Occurrence Summaries for Different Depth Range of Wells

Percentage of Wells
Depth AvgMn MaximumMn Percentage of
No. of Percentage 0 > Bangladesh
Range Cone. Concentration Wells> WHO
Wells Wells guideline value
(m) (mg/L) (mg/L) guideline value

<15 287 8 0.717 7.83 79 43

15-30 1180 33 0.653 9.98 80 42
30-60 1258 36 0.588 5.54 79 40
60-90 317 9 0.519 8.39 78 29
90-150 165 5 0.293 3.50 76 15
150-200 32 I 0.196 1.04 56 6
>200 295 8 0.073 1.59 18 2
Total 3534 100 0.554 9.98 74 35
(Source: Hasan, 2004)

Table 2.9 Manganese Concentration in Shallow Wells of Different Years of Construction

Percentage of Total Wells Having Specific Manganese Concentration Range

Year of (mg/L)
Construction .
0.001-0.1 0.1-0.5 0.5-1.0 1.0-1.5 1.5-2.0 2.0-3.0 >3.0

Before 1970 25 31 25 15 4 0 0
1970-74 16 41 24 II 5 2 I
1975-79 23 41 16 11 5 2 2
1980-84 24 44 21 4 4 I I
1985-89 21 43 18 10 5 3 0
1990-95 22 40 20 9 4 3 2
1995 and later 20 39 20 10 5 4 2

All years 21 40 20 9 4 3 2
(Source: Hasan, 2004)

26
 India

Bay of Bengal

~- 200 km ----4
'Groundwater Studies of Arsenic Contamin.tion in Bangladesh'
DPHElBGSlDFID (2ODO)

Figure 2.5 Distribution of wells with respect to year of construction

(Source: DPHFlBGS, 2001)

27
 Manganese and Arsenic Ratio

Arsenic is considered to be the most significant groundwater quality problem in

Bangladesh. However, available data do not show any correlation between occurrence
and distribution of arsenic and manganese in groundwater. This means that groundwater
with acceptable concentrations of arsenic may not have acceptable concentrations of
manganese Arsenic and manganese showed distinctly different regional pattern that is
shown in Figure 2.6. Division wise comparison of the occurrence of manganese and
arsenic is shown in Table2.10. Although about 43.5 % of wells in Rajshahi safe for
drinking with respect to arsenic concentration, but about 84% of these tubewells fails to
satisfy Bangladesh guideline value for manganese. It is notable that groundwater from the
deep aquifer contain relatively low concentrations of both arsenic and manganese. Only
33% of shallow tubewells had both arsenic and manganese concentration below the
respective WHO guideline value. Whereas about 93% of deep tubewells satisfied
respective WHO guideline value (DPHE/BGS, 2001).

Table 2.10 Comparative occurrence of Arsenic and Mangauese in different Divisions

As vsMn

As < 50 flg/L As < 50 flg/L As> 50 flg/L As> 50 flg/L

Total
Mn<0.5 mg/L Mn>0.5 mg/L Mn< 0.5 mg/L Mn> 0.5 mg/L
Division
Percen Percen Perce Percen Percen
No of No of Noof No of No of
tage tage ntage tage tage
wells wells wells wells wells
(%) (%) (%) (%) (%)
Barisal 249 84.4 6 2.0 28 9.5 12 4.1 295 100.0
Chillagong 145 32.6 76 17.1 166 37.3 58 13.0 445 100.0
Dhaka 426 43.1 259 26.2 174 17.6 129 13.1 988 100.0
Khulna 174 36.7 105 22.2 156 32.9 39 8.2 474 100.0
Rajshahi 540 50.4 466 43.5 11 1.0 55 5.1 1072 100.0
Sylhet 162 62.3 42 16.2 54 20.8 2 .8 260 100.0
Total 1696 48.0 954 27.0 589 16.7 295 8.3 3534 100.0

28
 At'6SiiitiC 'VB MangaiiiK.1!
M {Jg IL-~!Mill {WIg L~t)
• <fiO ,o(/)ll
• ",50 :loOll
• :>50 ,0(/);5
• :>00 >Oll

-
23
.,.

8lly 01 Dmijjltl'l
I ~a.fw--"

Figure 1.6 Comparative Disttibutions of Atsenie and Manganese in Groundwater

ofBaogladesh. (Source: DPHElBGS, 2001)

29
 Manganese vs Iron

Occurrence of high levels of iron is the most common water quality problems in
groundwater of Bangladesh. In many areas iron concentration much higher than WHO
standard (0.30 mglL) as well as Bangladesh guideline value (1.0 mglL) for iron.
Especially the standard limit set by WHO can hardly be maintained in rural water supply
in Bangladesh (Shahid, 1998). From National Hydrochemical Survey (NHS), it is found
that more than 65% of tubewells had iron concentration greater than WHO guideline
value. Therefore assessment of presence of iron in groundwater of Bangladesh has been
done with respect to Bangladesh standard value for iron.

Although both iron and manganese have physical as well as chemical similarities,
distribution of these elements in Bangladesh groundwater show very scattered pattern.
Figure 2.7 shows relative molar distribution of iron and manganese in Bangladesh
groundwater. From figure it is found that in about 25% areas of Bangladesh, Mn/Fe molar
ratio exceeds I and in about 22% area, the ratio is below 0.05.

Tab.le 2.11 represents a comparative occurrence of manganese and Fe in groundwater of

different divisions of Bangladesh. Iron concentration has been evaluated with respect to
Bangladesh guideline value and for manganese WHO health-based standard is used. In
Chittagong division only 37% of wells had iron concentration below Bangladesh standard
value but about 70% wells satisfied the guideline value for manganese. Again in Rajshahi
division about 59% and 49% of wells failed to satisfy respective guideline values for iron
and manganese. In Dhaka division more than 24% of wells had found to exceed
respective standard values of iron and manganese. Barishal division is found to be less
affected with respect to these two elements.

From National Hydrochemical Survey (NHS), it is also found that shallow tubewells are
more associated with high iron and manganese concentration. More than 65% of shallow
tubewells had exceeded Bangladesh guideline value for iron whereas about 61% of that
exceeded WHO guideline for manganese. In case of deep tubewells, about 26% had iron
concentration and 11% had manganese concentration greater than respective guideline
values. Only 27% of shallow tubewells had both iron and manganese concentrations

30
 below the standard limits, whereas about 95% of deep tubewells satisfied the respective
standard values (DPHEIBGS, 2001).

Table 2.11 Comparative Occurrence oflron and Manganese in Different Divisions

Mn vsFe
Fe < I mg/L Fe> 1 mg/L
Fe < 1 mg/L Fe> 1 mg/L
Mn<0.5 mg/L Mn<0.5 Total
Division Mn>0.5 mg/L Mn> 0.5 mg/L
m /L
Percen Percen Perce Percen
No of No of No of No of Percent No of
tage tage ntage tage
wells wells wells wells age (%) wells
(%) (%) (%) (%)
Barisal 211 71.5 4 1.4 65 22 15 5.1 295 100.0
Chittagong 123 27.6 42 9.4 188 42.3 92 20.7 445 100.0
Dhaka 300 30.4 148 14.9 301 30.5 239 24.2 988 100.0
Khulna 78 16.5 90 19.0 251 52.9 55 11.6 474 100.0
Rajshahi 381 35.5 279 26.0 170 15.9 242 22.6 1072 100.0
Sylhet 52 20 7 2.7 164 63.1 37 14.2 260 100.0
Total 1145 32.4 570 16.2 1139 32.2 680 19.2 3534 100.0

31
 Mn/Fe (molar) .
• <0.05
0.05-0.2
0.2-1
>1

India

~ 20
Bay of Bengal
10
~-200km-~
o
'Groundwater Studies 01 Arsenio Contamination in Bangladesh'
DPHElBGSJl>FID (2000)

Figure 2.7 Comparative Disuibutions of Iron and Manganese in Groundwater of

Bangladesh. (Source: DPHElBGS, 20eH)

32
 CHAPTER 3
CHEMISTRY OF MANGANESE AND MANGANESE REMOVAL
TECHNIQUES

3.1 INTRODUCTION

. Manganese is the third most abundant transition metal in the earth's crust (9.5 x 102
ppm). The redox chemistries of manganese (I/III/IV) have important roles and impacts in
the environment. An important principle to remember about chemical reactions is that, if
allowed enough time, it will reach equilibrium with the surrounding environment. When
the conditions of that environment are changed, such as pumping water from an
underground aquifer, the chemical equilibrium is upset. This will lead to either solution of
manganese or its precipitation.

In this chapter relevant literature on chemistry of manganese that influence manganese

oxidation and dissolution have been reviewed. Different environmental factors which
influence manganese oxidation has also been reviewed briefly. This chapter also provides
an overview of manganese removal techniques from water.

3.2 CHEMISTRY OF MANGANESE IN WATER

A general rule of thumb is that oxygenated water will have only low levels of manganese.
The reason is that manganese reacts with oxygen to form compounds that do not stay
dissolved in water (Varner, 1994). Surface water and shallow groundwater usually have
enough dissolved oxygen to maintain manganese in an undissolved state. In surface
water, manganese is most likely to be trapped within suspended organic matter particles.
Oxidized manganese precipitates are so small sized that complete removal is not possible
by settling (Seelig, 1992). Regardless of the removal the basic approach utilized to
remove manganese from water involves oxidation of soluble manganese to one of more
insoluble forms. Thus the removal chemistry of manganese predominantly includes
oxidation chemistry of manganese.

<0' ..•...t"
. ';
"
.•.. ', .
~, .
 3.2.1 Process Kinetics for Oxidation of Manganese

Although iron and manganese are chemically similar, the rate of manganese oxidation
does not follow the same rate law as for Fe (II) oxygenation (Stumm and Morgan, 1981).
The chemistry of manganese is substantially more complex than that of iron and only a
limited understanding of manganese oxidation exists (Montgomery, 1985 ).The oxidation
and control of manganese is complicated by factors that ranges from misunderstanding of
the reaction chemistry to the relatively slow kinetics and the numerous oxidation states
that result from this oxidation (Stumm and Morgan, 1981). Various forms of manganese
in aqueous solution is shown in Figure 3.1. The oxidation of manganese (II) with
molecular with oxygen is an autocatalytic process, that is, the spontaneous oxidation by
free oxygen at room temperature. According to Stumm and Morgan (1981), the reaction
might be visualized as proceeding to the following pattern:

slow
___ ._ Mn02 (s) ----------------------(a)

fast
• Mn2+. Mn02 (s) ---------(b)
slow
--.... 2Mn02 (s) ----------( c)

Although other interpretations of autocatalytic nature of the reaction are possible, the
following experimental findings are in accord with such reaction scheme:

1. The extent of manganese (II) removal during oxygenation reaction is not accounted
for by the stoichiometry of the oxidation reaction alone; that is not all the manganese
(II) removed from the solution is oxidized. Therefore this mechanism offers a
reasonable explanation why a less than stoichiometric oxidant dose may be required
in the "oxidation" of manganese (II).
ll. The products of non-stoichiometric oxygenation of manganese (II) show vanous
average degree of oxidation ranging from MnOu to Mn01.9 (30 to 90% oxidation to
Mn02; commonly expressed as Mn0x) under varying alkaline conditions.
lll. The higher valent manganese oxide suspensions show large sorption capacities for
Mn2+ in slightly alkaline solutions.

34
Therefore, this removal mechanism indicates that presence of manganic dioxide generally
increases the apparent rate of oxidation of both iron (H) and manganese (H).The
integrated form of autocatalytic reaction rate of manganese (H) can be expressed as
follows:

- d [Mn (II)]
dt
= ko [Mn (H) ] + k [Mn (H)] [MnOz] (1)

where,
- d [Mn (II)] = Rate of manganese (H) oxidation [mol/Umin]
dt z
ko = Reaction rate constant [I /molz.atm.min]

k 3
= Reaction rate constant [1 /moe .atm.min]

[Mn (H)], [MnOz] = Ionic concentration [mol/L]

Both manganese (H) oxidation and removal rates follow the rates law of equation (1). The
rate dependence on the oxygen concentration can expressed as :

- d [Mn (II)]
(2)
dt

Therefore, k in equation (1) can be formulated as

k = k' [OH' ]z Poz
Where,
OH' = Hydroxide ion concentration [mol/L]
Poz = Partial pressure of oxygen [atm]

Graveland (1975) suggested another reaction rate equation for manganese. According to
Graveland overall manganese removal is also dependent on temperature, alkalinity as
well as rate of filtration and media diameter. His equation is expressed as,

- d [Mn (II)]
dt = k.[Mn(H)].[Oz]. {[OH' ]_10'70 }.{[4.32xl0'3 +[HC03']}.
exp(-7000/T).Vo035dm,1I1 (3)

Where,
[Mn(H)] = concentration ofMn(H) (mg/I)
t = time (sec)

35
 k = reaction rate constant (sec-! )
[02] = oxygen concentration (mg/I)
[OK] = concentration of hydroxyl ion (g/l)
[HC03-] = bicarbonate concentration (mg/l)
T = temperature ( 0 Kelvin)
Vo = Filtration rate (cm/sec)
dm = mean particle diameter (cm)

Aqueous Mn(H) is oxidized by reaction with dissolved oxygen. The reaction proceeds
through the aqueous Mn(OH)2 species, although the bimolecular rate constant of
Mn(OH)2 with O2 is 1052 lower than that of Fe(OH)2. The reaction product (Mn(III)), in
the absence of strongly complexing ligands, rapidly polymerizes to form Mn oxide solids,
which catalyze further Mn(H) oxidation. Hence, separating homogeneous from
heterogeneous pathways in Mn(H) oxidation is difficult because they occur
simultaneously under most experimental conditions (Martin, 2003).

3.2.2 pE - pH and Eh- pH Diagrams for Manganese in Solution

In order to describe the stability relationships of distribution of various soluble and

insoluble forms of aqueous elements, two types of graphical analysis are essential been
used: first, equilibria between chemical species in a particular oxidation state as a
function of pH and solution composition; second, equlibria between chemical species at a
particular pH as a function of pE (or EH). These diagrams can be combined into pE-pH
diagrams. Such pE-pH stability field diagrams show in a comprehensive way how proton
and electrons simultaneously shift the equilibria under various conditions and can indicate
which species predominate under any given condition ofpE and pH (Stumm and Morgan,
1981). Natural waters are in highly dynamic state with regard to oxidation -reduction
rather than in or near equilibrium. Most oxidation-reduction reactions have a tendency to
be much slower than acid-base reactions, especially in the absence of suitable
biochemical catalysis. Nonetheless equilibrium diagrams can greatly aid attempts to
understand the possible redox patterns of water (Stumm and Morgan, 1981).

There is an analogy between acid-base and reduction-oxidation reactions. Acid-base

reactions exchange protons. Acids are proton donors and bases proton acceptors. Redox

36
 processes reactions exchange electrons. Reductants are electron donors and oxidants are
electron acceptors. As there are no free hydrogen ions (protons), there are no free
electrons. It implies that every oxidation is accompanied by a reduction and vice versa
(Jimbo and Goto, 2001). Like pH has been introduced as the proton activity, we may
introduce an electron activity defined as:

pE=-log{e' }

For as aqueous solution at a given pH each pE value is associated with a partial pressure
ofH2 and O2.
The equilibrium redox equations for these elements can be expressed in logarithmic form:
log PH2 = 0 - 2pH - 2pE
log P02 = -83.1 + 4pH + 4pE

These equations can be rewritten as

pE = 0 - pH - Yz log PH2
pE = 20.78 - pH - Y.log P02

These equations can be plotted in a pE-pH diagram with a slope d pEl d pH of -I for both
lines. Similarly, pE-pH diagram for aqueous manganese can be developed using
equilibrium equation (Considering solid concentration in pE equations equals I f.lM.):

Yz Mn02 (s) + 2H+ + e = Yz Mn2+ + H20

Then at equilibrium,
pE= pEa + log [Red 1] [OX2]
[OX1] [Red2]

pE = pEa + log [M n 2+]% ]

[M 2(S)]' [H]
I
= 20.58 - Yz log[Mn2+ ] + 2log[H +]

pEa is the electron activity standard condition of temperature (2S°C) and pressure (I atm)
= 20.58.

37
Rewriting the above equation,

pE = 20.58 - Y2log[Mnz+] - 2pH (4)

Thus pE-pH diagram derived according to equation (4) is shown in Figure 3.2 for a total
manganese concentration 10-6 M at 25°C. The dashed lines show the water stability region
for I atm of gases. Above the top line, water should thermodynamically form Oz if
oxygen partial pressure is below I atm. Similarly, below the bottom line, water should
thermodynamically form Hz if hydrogen partial pressure is below I atm. Natural
environments have a range of pE/pH values: for oceans pE = O2 saturated, pH = 8),
surface waters oflakes and rivers (pE = Oz saturated, pH = 4 to 6), mine waters (pE = 02
saturated, pH = I to 3), groundwater and sediments (pE = 0 to 3, pH = 7 to 9), and
swamps (pE = 0 to -3, pH = 5 to 7) (Martin, 2003).

In the aqueous phase in the manganese system, [Mn(HzO)6f+ is the unique dominant at
the common pE and pH values of natural waters. Hydrolysis begins only for pH > 10,
while the stabilization of aqueous Mn(III) requires strong ligands. The white pE-pH
regions show where MnT is thermodynamically speciated entirely in the aqueous phase.
For instance, if the prescription of 10-6M MnT initially includes solid species in the white
region, then it is predicted that these solids will dissolve. However, the dissolution rate is
slow compared to the timescales of days, weeks, and months usually relevant to the
environment. The gray pE-pH regions conversely show where some (viz. near the
boundary lines) or most (viz. moving inw~rd in the gray region) of 10-6 M MnT should
thermodynamically include solid phases, with a small amount of aqueous manganese
species in equilibrium (e.g., 10-9 to 10-15 aqueous Mn). However, the precipitation rate
may be slow. For instance, in the absence of catalysis, aqueous Mn(II) at pE = 10 and
pH= 8 persists for years in solution, even though oxidation and precipitation ofMn oxide
solid phases is thermodynamically favored. By increasing pH, super saturation increases
and Mn(H) oxidizes and precipitates (Martin, 2003).

However, the first solid to form is generally the least favored thermodynamically, a
phenomenon described as Ostwald's rule of stages (Martin, 2003). For instance, Mn(III)
solids such as MnOOH form initially even when the free energy of formation for Mn(IV)
O2 is greater. The Mn(IV) oxides form only after long aging times of Mn(III)OOH

38
(Martin, 2003).Stability relationship of aqueous elements can also be expressed by redox
potential (Eh) -pH diagram. Redox potential is defined as the oxidation or reduction
potential of a particular environment (Douglas, 1994).

The redox potential, a measure of the oxidizing power of a system, is a variable of major
importance in characterizing systems containing elements that exhibit more than one
oxidation state (Minear et ai, 1982). There are some close relationship between Eh and
pH. Eh essentially measures the environments ability to supply electrons in an aqueous
solution (limbo and Goto, 2001). The relationship between pE and the redox potential, Eh
can be expressed as follows:

pE= F x Eh (5)
2.3 RT

Where,
F = Faraday's number = 96485 Clmol = the charge of 1 mol of electrons
R = Gas constant = 8.314 J/mol K = 0.082057 liter atrn I mol K.

Nernst Law provides the basis for measurement of redox potential. According to Nernst
Law (limbo and Goto, 2001):

2.3 RT ~
Eh =Eho + log (6)
nF [ red]

Where,
n = Number of electrons gained or lost
Eh° = Standard redox potential corresponding to the {ox} = {red} = 1.

From standard redox potential table it is found that for reduction reaction of Mn(Il) at
room temperature (25°C) is ~ 0.615 V (for half reaction; as in reduction reaction of
manganese two electrons are gained) (Jimbo and Goto, 2001).

Combining equation (5) and equation (6) and putting values of constants a Eh-pH
diagram can be developed. Figure 3.3 represents the Eh-pH diagram for manganese at
standard condition. In the figure, vertical lines separate species that are in acid-base
equilibrium. Non vertical lines separate species related by redox equilibria. Horizontal

39
lines separate specIes In redox equilibria not involving hydrogen or hydroxide ions.
Diagonal boundaries separate species in redox equilibria in which hydroxide or hydrogen
ions are involved. Dashed lines enclose the practical region of stability of the water
solvent to oxidation or reduction (Douglas, 1994). This diagram is called Pourbaix
Diagram.

From Pourbaix diagram for manganese it can be observed that (Douglas, 1994):
o Strong oxidizing agents and oxidizing conditions are found only at the top of
Pourbaix diagrams.
o Strong oxidizing agents have lower boundaries that are also high on the diagram.
In the diagram it is shown that permanganate is an oxidizing agent over all pH
ranges. It is very strongly oxidizing even at low pH.
o Reducing agents and reducing conditions are found at the bottom of a diagram and
not elsewhere. Strong reducing agents have low upper boundaries on the diagram
and it can be noticed that manganese metal is a reducing agent over all pH ranges
and is strongest in basic conditions.
o When the predominance area for a given oxidation state disappears completely
above or below a given pH and the element is in an intermediate oxidation state,
the element will undergo disproportionation, therefore, MnO/- tends to become
disproportionate.

3.3 FACTORS AFFECTING MANGANESE OXIDATION

The oxidation and control of manganese reactions is complicated by factors that ranges
from misunderstanding of reaction chemistry to the relatively slow kinetics and the
numerous oxidation states (Montgomery, 1985). Further adding to the complexity of
manganese chemistry is the fact that there are difficulties with analytical techniques used
to remove manganese. The gross analytical techniques do not allow for differentiation of
the speciation of the manganese. Therefore, presumptions are made with respect to the
oxidative states of manganese without through knowledge of its speciation (Montgomery,
1985). However, in general, the removal of manganese is greatly influenced by some
environmental parameters, such as pH, dissolve oxygen, temperature etc. Some of these
factors are discussed below.

40
 u-MnU2iS)
Mn(IV) ~-MnO:z{s)
bimessite (1I1,IV)

Mn(lI) Mn(lII)
Mn2;(aq) MnOOH{S)
Mn(ll)l(aq) M~O.(s) m,lII)
MnC03(s) Mn{I1I}L{aq)
Mn(OH),{S) (L = strong ligand)

Figure 3.1 Forms of Manganese in aqueous solution

(source: Martin, 2003)

30
.25°C
25 10-8M Mn(IIXaq)

20
<I-Mn02(S)

15

10
w
0-
5
Mn(II)(aq)

0
-- H2O
H2 ................
-5 ,

I
-10
- ...._-,J
Mn(OHl2(sjl

-15
o 2 6 10 12
pH

Figure 3.2 pE.pH diagram for the simple ions and hydroxides of manganese
at 2S°C. (Source: Martin, 2003)

41
 ~ T r
.:!.ll

I.ll.
Mn0..a- lrurplel
1.6

I.'
l.~
1.0
O.tl
0.•
n.4

t
;;
0 ..2
-- ---- --
~
-0.2
0
--- -------
-0,4
------ --
-0 .•
-0.8
-1.0
-- --- --
cHMnO~-)

-I.:! MntOH1~
-l.-l

-I.t'
-1.~ 10 II 1.2 13 14 l~ 16
_.2 -I (l

pH
Figure 3.3 Eh-pH diagram (Pourbaix diagram) for manganese at 25° C
(Source: Douglas, 1994)

42
3.3.1 Effect of pH

Oxygenation kinetics equation for manganese (Equation 2) clearly shows that rate of
reaction ofMn (II) has a second order relationship with hydroxyl ion concentration which
indicate that an increase in one pH unit, cause 100 fold increase in rate of reaction
(Stumm and Morgan, 1984). The pH dependence of Mn(II) oxidation is shown in Figure
3.4. Reaction ofMn (II) with 02 is at least 106 times slower than that occurs for iron (II)
oxidation at circumneutral pH (Martin, 2003). Only for pH > 8 does the reaction rate
become appreciable.

According to Marble et al (1998), overall mass transfer of Mn(II) from solution to active
sites at the surface decreases as pH decreases because of competition with ft. As
reported by Ben~~hoten and Lin (1992) sorption of Mn (II) in excess of 0.5 mol of Mn(II)
adsorbed per mole of MnOz (s), where as the capacity of oxide surface at pH 9 is about 2
mole of Mn(II) adsorbed per mole of Mn02 (s). However, strong oxidizing agent like
permanganate or chlorine dioxide can effectively oxidize manganese at a pH range from 5
to 10 (Samblebe, 2003). But for slow oxidizing agents like chlorine, it is necessary to
raise the pH above 8.5 for effective oxidation reaction of manganese (Samblebe, 2003).

3.3.2 Presence of Organic Matter

Mn (II) is capable of forming complexes with organic matter and as such, is resistant to
oxidation. The relative strength of such complexes has a stability constant of
approximately 104 (Shahid, 1998). Again, presence of oxidizable organics or inorganics
in the water reduces the oxidation effectiveness of the oxidizing agent (e.g., chlorine,
permanganate, etc.) used to remove manganese from water because some of the applied
dose will be consumed in the oxidation of organics and inorganics. Permanganate
oxidizes a wide variety of inorganic and organic substances in the pH range of 4 to 9
(Hazen and Sawyer, 1992).

3.3.3 Effect of Temperature

Change in temperature can affect the oxidation reaction rate of manganese. As ionization
constant of water is dependent on temperature variation, which in tum effects hydroxyl

43
ion concentration of water. On this basis, from oxidation reaction kinetics it can be found
that at a given pH value, the rate increases about 10 fold for a 15° C increase in
temperature (Shahid, 1998).

A number of research review indicates that oxidation rate gets slower with decrease in
temperature (Benschoten and Lin, 1992). As reported by Montgomery (1985), oxidation
of manganese by permanganate solution needs a contact time of 5 mins at 20° C and a
°
contact time of 10 mins at 1 C.

3.3.4 Presence of Dissolved Oxygen

•
As stated in equation (2), the rate of manganese oxidation is of the first order with respect
to the partial pressure of oxygen, P02 (Stumm & Morgan, 1981). It is also observed that
above about 30% of saturation value of dissolve oxygen, there exhibits no significant
dependence on concentration of dissolve oxygen. Below this value the net rate of Mn(II)
removal was approximately first-order with respect to DO concentration (Marble et ai,
1999) The DO dependence of the observed net rate of removal is presented in Figure 3.5.

Many other researchers [e.g., Graveland, 1975; Tebo and Emerson, 1985; etc] reported
no dependence of the rate of Mn(II) removal on DO above concentrations of about 1
mglL(i.e.- 12% air saturation, or 0.03 mM) and an approximate linear dependence at
lower DO values (Marble et ai, 1998).

3.3.5 Mn (II) Concentration in Solution

As described in Marble et al (1999), the net rate of Mn (II) removal is directly

proportional to Mn (II) and that a simple first-order dependence on Mn (II) is a
reasonable assumption. The dependence of the net rate of removal of Mn (II) on Mn(II)
concentration is shown in Figure 3.6. However, it is apparent from the Figure that the net
rate of Mn(II) removal does not extrapolate to a zero intercept at zero Mn(II)
concentration.

44
 25'C
(
Po, = 0.2 aim
~ Rate = dlogtMn (Ill]
>, dl
oj
.,,;
~
0) -1
"'id
e<:
OJ;
..s
-2

-3

8.0 8.5 9.0 9.5

pH

Figure 3.4 Effect of pH on Manganese Rate of Oxidation

(Source: Marble et aI, 1998)

10
I
I
I
I

.
I

••
I
I

I.
+. ..--.
. I
~'
~,
4,!
f
/
I
I
I
I

:::s,"" ,,.1*+
I Ail"
I
I :BatlUra'lon
.,l I

e/" !'ir&1.Qrder
I

I
.

, l[)e;penden •••• I
I
f I
f I
I
1
OJBt 0l:10 -1I.DD

Dissolve Oxygen (mM)

Figure 3.5 Effect of Dissolved Oxygen on Manganese Oxidation

(Source: Marble et aI, 1998)

45
In other studies [e.g., Tebo and Emerson, 1985; Tebo et aI, 1997] of Mn (II) removal
(oxidation) in microbially active media, it has been demonstrated that the rate of Mn (II)
removal is consistent with a Michaelis-Menten-type of rate expression (Marble et aI,
1998).

- d [Mn (II)] _ [Mn (II)]

(7)
dt - Vm [Mn (II)] + Km

Where
Vm = maximum rate (nM/sec)
Km = Michaelis constant

The study by Marble et al (1998) found that it was more difficult to remove Mn when the
initial concentration was low, regardless of the oxidant used.

3.3.6 Effect of Alkalinity

As stated in Stumm and Morgan (1985), sorption capacities of Mn (II) increases at

slightly alkaline solutions as low alkalinity waters tend to dissolve minerals and metals.
Figure 3.7 shows removal of manganese in bicarbonate solution (Stumm and Morgan,
1985). As indicated by Graveland (1975) in his manganese oxidation rate equation that
manganese oxidation rate is directly dependent on bicarbonate alkalinity. A review of
research involving reactions of permanganate and Mn(II) indicates that experiment often
carried out in acid media to avoid complications that arise from precipitation of Mn (II)
oxides and heterogeneous oxidation reaction (Benschoten and Lin, 1992). There is lack
of clear information about the mechanism of direct effect of alkalinity on manganese
removal.

3.3.7 Presence of Oxide Surfaces

The rate of Mn(II) oxidation by O2 is catalyzed by metal oxide surfaces (>S). These
surfaces are terminated by hydroxyl groups (>SOH), which bind Mn(II) as (>SO)2Mn.
The inner-sphere surface complexes promote rapid oxidation. The catalysis occurs both
on foreign surfaces, e.g., Mn(II) on FeOOH and also for the special case of autocatalysis,

46
 ,

20 -
-
.

-
<>
<>

,
o
o l'
[Ulnfll)) - ",M
Figure 3.6 Effect of Manganese Concentration in Solution
(Source: Marble et aI, 1998)

J
~. 0
.--c p!H9.0

:E

2:5" C
. pili ~O~"'" .1
!

.2:
a 40 so 120 160 200
MinUtes

Figure 3.7 Manganese oxidation in Bicarbonate Solution

(Source: Stumm and Morgan, 1981)

47
e.g., Mn(II) on MnOOH producing additional MnOOH (Martin, 2003). As reported by
\
Bensschoten et al (199,2), the kinetics of the reaction rate can be expressed as:

- d [Mn (II)) (>SOH) [Mn (II)] (8)

dt =k x A x P02 [H+f

Where,
(>SOH) = surface concentration of hydroxyl sites
A = mass solids concentration
P02 = Partial pressure of oxygen"

The surface hydroxyl site concentration is hypothesized to be proportional to the

concentration of surface Mn(II) species (Bensschoten and Lin, 1992). Reaction rates at
surfaces are given either as the conversion rate per unit surface area of the foreign surface
(mol/m2-s) or as the conversion rate per liter of a particulate suspension (MIs). The later is
the basic observable in experiments employing particulates, whereas the former is a more
intrinsic measure, which can be estimated for a suspension of known loading (g/ L ) and
specific surface area (m2/g ) (Martin, 2003).

Again in autocatalytic reaction, heterogeneous oxidation occurs when the product of the
oxidation further accelerates the reaction rate. For example, the oxidation of Mn(1I)
produces MnOOH(s), as follows:

homogeneous
---~~- 4MnOOH (s) + 8 H + (9)
or heterogeneous

As the reaction proceeds, the MnOOH(s) surface area and hence the heterogeneous
reaction rate increase. The rate laws of autocatalysis are less precise than those of
heterogeneous reactions on foreign mineral surfaces. Detailed descriptions for the
autocatalysis pathways are hindered both by the complexities of separating homogeneous
from heterogeneous pathways and by limitations in characterizing the increasing mineral
surface area and the altering mineral phases during reaction (Martin, 2003).

48
3.3.8 Presence of Reductive Substances

Reductants rapidly accelerate manganese oxide dissolution. Examples of reductants are

ascorbic acid, hydrogen sulfide, and phenols. A reductant typically forms an inner-sphere
complex at the surface, though not always so. When an electron is transferred to a
Mn(III/IV) oxide, a surficial Mn(II) ion locked inside an oxide lattice is formed. Because
Mn(II) oxides are much more soluble than the corresponding higher oxides (Figure 3.2),
rapid Mn(II) depolymerization occurs, which is followed by release to the aqueous phase
ofMn(II).

3.3.9 Effect of presence of Metal Ions

Unlike iron (II), metal ions like Cu + and complex formers do not appear to have any
marked effect upon reaction rate of manganese (Stumm and Morgan, 1985).

3.4 MANGANESE REMOVAL TECHNIQUES

If manganese is present in a water source, there would be two basic options - obtain an
alternate water supply or use some type of treatment to remove the impurity. Selection of
appropriate option for manganese contaminated water depends on various factors, such as
concentration of manganese, availability of alternate sources etc. Use of alternate source
instead of groundwater is less preferable from cost as well as bacteriological
contamination view point. Therefore using groundwater with suitable treatment for
manganese removal would be more appropriate.

3.4.1 Factors that must be known when choosing a Treatment Process

Type of Manganese Present

Manganese may be present in any of three different forms ranging from clear to
discolored as described below. Not all treatment methods work on all forms of manganese
(Varner et aI, 1994).

49
1. Water is totally clear when drawn from the tap: Manganese is present in the dissolved
form. The term "Clearwater manganese" is often used to describe this form. The
scientific name for clearwater Manganese is called "manganous".
2. Water is rusty colored when drawn from the tap: When exposed to oxygen or other
oxygen like chemicals, clearwater manganese will precipitate to form fine blackish
(manganic) "rust" particles. The tendency to precipitate is also influenced by changes
of water temperature, pressure, pH and other factors. It is this precipitated form which
stains wate; use fixtures and discolors laundry. These rust particles will settle if the
water is not disturbed.
3. Water has a yellow tint, but is totally transparent and the color does not settle out with
time: In this case the manganese is probably combined with dissolved organic matter
in the water. This is commonly called colloidal iron. It is more commonly found in
surface water than in groundwater; therefore, you should also have the bacterial
quality of your well checked if this form is present. Testing for the organic
components of tannins is also suggested.

This form of manganese will not settle out when the water is undisturbed, is too small
to be removed by filtration, and cannot be effectively treated by ion exchange (i.e.,
water softener). Colloidal manganese is very difficult to remove.

Water Quality Tests

In order to determine which treatment process will work for a particular water quality; it
is essential to know certain water quality factors. Typically important factors for
mang~nese removal include (Samblebe, 2003):

• The concentrations of manganese.

• pH and hardness.
• Dissolved oxygen for some treatment types; this must be field measured.
• The presence of manganese bacteria.

50
 •

3.4.2 Overview of All Categories of Treatment

Generally speaking, there are two basic methods for treating water containing manganese
either by exchanging manganese with any other cation or by oxidizing soluble manganese
to precipitate as insoluble form(s). The diagram below shows the treatment options for
manganese removal.

Oxidation processes, both physical-chemical and biological basically involves oxidation-

raction and surface adsorption followed by suitable filtration option. Surface adsorption is
influenced by autocatalytic behavior of manganese. It is also dependent on the type of
oxidation procedure used to remove manganese. A separate explanation for each type of
manganese removal processes is summarized in Figure 3.8.

Manganese Removal

Oxidation Ion Exchange

(e.g. ion exchange
olecular oxidation followed using water softener)
by filtration (Aeration)
hemical oxidation followed by
filtration ( KMn04, Cb etc.)
xidation filter (manganese
green sand filter, zeolite filter)
iological oxidation

Figure 3.8 : Commonly Used Manganese removal Techniques

3.4.3 Removal of Dissolved Manganese Using Water Softeners

A water softener removes manganese, which is in the dissolved "clearwater" form.

Softening also removes calcium (Ca++) and magnesium (Mg++) ions which are the
primary minerals responsible for "hard" water. The treatment process consists of passing

51
the water through an ion exchange resin media bed. The manganese ions and also calcium
and magnesium ions in the water are "exchanged" for sodium (Na+) ions which have
been temporarily stored in the resin material (Kassim, 1994).

As the hardness and manganese are removed from the water, sodium is added
proportionally. For every 10 mg!1 of hardness and manganese removed, approximately 5
mg/l of sodium will be added to the treated water. For those concerned with elevated
sodium levels in their drinking water, potassium chloride (KCI) can be used in place of
sodium chloride (NaC\). The cost of potassium chloride is higher than sodium chloride
(Ficek, 1985).

Eventually the removal capacity of the ion exchange resm material will become
exhausted and the media will need to be rejuvenated. The rejuvenation process begins
with a physical backwash of the media. The resin is then immersed in a strong salt brine
solution. The sodium from the salt enters the resin and displaces the previously removed
manganese and hardness. After a period of time (approximately 20 minutes), the
remaining brine, along with the displaced manganese and hardness are flushed out of the
device and disposed of into a dry well, septic tank or sewer. Studies by the Water Quality
Association (a trade organization of the home water conditioning industry) indicate that
waste brine does not injure leach fields or septic tanks (Kassim, 1994).

Ion exchange softening is described as effective for water containing less than 2- 5 mg/L
of dissolved (I.e., colorless) manganese. The system will not work at all where the
manganese has turned to a rusty color. Other aspects of water quality are not overly
important. Where both manganese and hardness are high, softening is an appropriate
treatment technique (Seelig, 1992).

Where manganese is bound to organic matter, or concentrations of manganese is very

high, manganese bacteria is present, a strong oxidizing substance must be applied before
filtration. The most commonly used chemical in these systems is household bleach
(hypochlorite) injected ahead of the pressure tank. This procedure disinfects the water and
at the same time oxidizes iron (if present), manganese, and organic matter, which will
then precipitate. Sedimentation and/or filtration are then needed to remove the
precipitants. Activated carbon units or reverse osmosis units should then be used to

52
remove the remammg chlorine and possible halogenated hydrocarbons created from
organics. Final choice of the method will depend on manganese concentration, pH of
water, and th~ presence of the bacteria (Sharma et ai, 2001).

Advantages of Water Softening

-Softener resin can be rejuvenated and re-used.

- Ion exchange can consistently remove dissolved Fe/Mn from water to extremely
low levels.
- There are lower backwash water requirements than oxidizing filters.
- The manganese removal is not appreciably affected by pH.

Disadvantages of Water Softening

- Softening will not operate satisfactorily if manganese bacteria or rusty colored water
exists, even if occasionally. If particles are present, a sediment filter is often placed
before the resin tank (Samblebe, 2003).
- The softener and iron filter are effective only if the manganese is not bound to organic
matter and there are no iron or manganese bacteria in the water. The oxidizing media of
the iron filters are not strong enough to break these materials down. In such case it is
necessary to pre-oxidize water with appropriate oxidizing agent e.g., permanganate,
bleaching powder etc (Sharma et ai, 2001).
- A water softener will not remove hydrogen sulfide odor.
- Water softeners produce waste brine that must be disposed off. In absence of a sewer or
other suitable drainage system, disposal of the waste brine will likely be into the
ground. This creates the potential of polluting the groundwater. This disposal
consequence can be lessened by minimizing the number of backwash rejuvenation
cycles (Samblebe, 2003).

New water softeners allow the frequency of softener rejuvenation cycles to be reduced.
The controls on these devices include: those that measure the water's electrical
conductivity or those that measure the volume of water treated. In each case, rejuvenation
is triggered based on actual need rather than the passage of time. Excessive backwash
needlessly increases salt use and the generation of waste brine (Sharma et aI, 2001).

53
3.4.4 AERATION FOLLOWED BY FILTRATION

Aeration can be used to oxidize manganese ions to mangamc dioxide. However, the
kinetics of oxidation by oxygen is slow in typical water treatment conditions and so a
long detention time is required especially at pH less than 8.5 (Raveendran et aI, 2001).
Aeration is useful as an option to oxidize manganese in reservoirs. The reaction between
manganese and molecular oxygen is:

2Mn02 (s) + 4H+ (10)

The amount of molecular oxygen required to oxidize manganese can be obtained from
stoichiometric relation of manganese oxidation as shown in Table 3.1, which shows that
0.29 mg 02 is required to oxidize per mg of manganese.

Operation of the aeration process requires careful control of the flow through the process.
If the flow becomes too great, not enough air is applied to oxidize the iron and
manganese. If the flow is too small and the aeration is not cut back, the water can become
saturated with dissolved oxygen and, consequently, become corrosive to the distribution
system. During aeration, slime growths may be created on the aeration equipment. If
these growths are not controlled, they could produce taste and odor problems in the water.
The growth of slime can be controlled by the addition of chlorine at the head of the
treatment plant. The process should be inspected regularly to catch the problems in their
early development (Seelig, 1992).

Manganese removal by simple aeration reqUires longer contact time depending on

manganese concentration present in water. Basically for lower manganese concentration
higher reaction time is required, even as high as contact time of 1 to several hours may be
needed (Montgomery, 1985). This may due to the autocatalytic behavior of manganese
and presence of higher concentration accelerates removal. Aeration is ineffective in
oxidizing organically bound manganese. Due to involvement of high cost, complexity in
pH adjustment and being time consuming, aeration can only be used as a preliminary
treatment to oxidize manganese. Where further oxidation is necessary an oxidizing agent
must be introduced to reduce the manganese levels (Raveendran et aI, 2001).

54
After reaction and precipitation of insoluble manganese, the water is allowed to flow
through a filter where various filter media are used to screen out oxidized particles of
manganese and some elements co precipitated with manganese. The selection of media is
important. The media should have a large effective size (>1.5 mm) to reduce head loss
and should not have a low uniformity coefficient (Montgomery, 1985). The most
important maintenance step involved in operation is periodic backwashing of the filter.
As manganese oxidation is slower than for iron, it requires greater quantities of oxygen
(Seelig, 1992).

Advantages of aeration

- No foreign chemicals are added to the water.

- Low labor cost.
- Can handle a wide range of manganese concentration in water.
- Can often reduce some odor.

Disadvantages of aeration

- Aeration is not recommended for water containing organic complexes of manganese or

manganese bacteria that will clog the aspirator and filter.
- Generally aeration involves high capital costs and high running costs.
- Aeration alone cannot completely oxidize all manganese at pH less than 10 (Seelig,
1992) and without longer detention time.

3.4.5 OXIDIZING FILTER

An oxidizing filter treatment system is an option for moderate levels of dissolved iron and
manganese at combined concentrations up to 15 mgll. The filter material is usually
natural manganese greensand or manufactured zeolite coated with manganese oxide,
which adsorbs dissolved iron and manganese. In this adsorption process Mn304 acts as a
catalyst on which Mn2+ is adsorbed Mn2+gets oxidized to. Mn304 while older Mn304 gets
oxidized to Mn02 (Sharma et ai, 2001).

55
As water is passed through the filter, soluble iron and manganese are pulled from solution
and later react to form insoluble iron and manganese. Insoluble iron and manganese will
build up in the greensand filter and must be removed by backwashing. Backwashing
should be done regularly twice a week or as recommended by the designer (Seelig. 1992)

Synthetic zeolite requires less backwash water and softens the water as it removes iron
and manganese. The system must be selected and operated based on the amount of
dissolved oxygen. Dissolved oxygen content can be determined by field test kits, some
water treatment companies or in a laboratory (Kassim, 1994).

Manganese greensand is a specially processed medium for iron, manganese, and

hydrogen sulphide removal. Manganese greensand is a premium non-proprietary filter
medium which is processed from glauconitic greensand on which a shiny, hard finite
thickness manganese oxide coating is formed and is firmly attached on every grain by a
controlled process. This process utilizes the ion exchange properties of greensand to form
a manganese base material which is converted to manganese oxides by oxidation with
potassium permanganate. Manganese greensand contains 0.3% manganese or 0.45%
manganese dioxide.

This material has a high buffering or oxidation- reduction capacity due to the well defined
manganese oxide coating. Actually, the manganese greensand can oxidize over 300 grains
of manganese per cubic foot or reduce over I oz. of potassium permanganate per cubic
foot, by far the most of any iron and manganese removal filter media. The grains of
manganese greensand are of both the size and shape to capture the fine precipitates of
iron and manganese which pass through the upper coarse anthracite layer during normal
service conditions. No expensive polymer or other filter aid is needed to prevent leakage
of these oxidation products (Gregory, 1996).

Acidity or pH of the water will influence the ability of the filter to remove both iron and
manganese. If the pH of the water is lower than 6.8, the greensand will probably not
adequately filter out the iron and manganese. The pH can be raised above 7.0 by running
the water through a calcite filter (Seelig, 1992).

56
Advantages of oxidation filter

- The manganese greensand process requires no detention time, no secret expensive filter
media, no high concentration of chlorine, and no sulphur dioxide (Gregory, 1996)
- Optimum grain size and shape to retain oxidation precipitation products of iron and
manganese (Gregory, 1996)
\

- Unequalled oxidation-reduction buffer capacity. Can tolerate slight over or underfeed of

continuously fed oxidants (Gregory, 1996).
- Manganese oxide coating is not removed during backwashing (Gregory, 1996).

Disadvantages of oxidation filter

- High level of dissolved oxygen must be present for effective removal.

- Oxidizing filter size should be suitable to source water output. Where well output is low
and the required filter size is large, two smaller filters might be substituted so that each
can be backwashed separately. This isolation would be expensive however.
- Backwash requirement is greater than that for softening process (Kassim, 1994).

3.4.6 Chemical Oxidation Followed by Filtration

High levels of dissolved or oxidized manganese can be treated by chemical oxidation,

using an oxidizing chemical such as chlorine, permanganate, or sodium hypochlorite,
chlorine dioxide or ozone, followed by a sand trap filter to remove the precipitated
material. This treatment is particularly valuable when manganese is combined with
organic matter or when manganese bacteria are present (Varner et ai, 1994).

Chlorine

Chlorine is a stronger oxidizing agent than oxygen. Chlorine forms hypochlorous acid
when dissolved in water. For manganese oxidation chlorine needs to be added at the head
works or just before filtration. After a retention time of at least 20 minutes to allow for
oxidation of soluble manganese into the insoluble manganic form, the solid particles are
filtered out (Seelig, 1992). The chemical reaction can be written as:

57
 Mn2+ + Ch + 2HzO = MnOz (s) + 2cr + 4H+ (11)

Stoichiometric relation indicates that (Table 3.1) about 1.29 mg ofCh dose is needed for
per mg of manganese removal. Even though, the stoichiometric requirement of chlorine is
less than potassium permanganate, in practice the chlorine requirement has been found to
be much higher due to the chlorine demand by organic carbon (Raveebdran, et aI, 2001).

As chlorine is a weak oxidant, manganese removal by chlorination would not be very

effective until pH is raised above 8.5 and for high level of manganese it is often needed to
raise pH above 9.5 (Benschoten et ai, 1990). Moreover, it required alkaline conditions to
oxidize manganese. Soda ash injected with the chlorine will increase the pH to optimum
levels. Adjusting the pH to alkaline levels also reduces the corrosivity of the water to
pipes and plumbing (Seelig, 1992)

Pre-chlorination has a higher potential to react with organic compounds and to produce
trihalomethane (THM) which is carcinogenic. Chlorinators and appropriate safety
equipment are required to dose chlorine (Samblebe, 2003). When chlorine is used, the
treated water can have an unpleasant taste if a particle filter of calcite, sand, anthracite, or
aluminum silicate is used. To overcome this problem, use of an activated carbon filter can
remove both excess chlorine and solid manganese particles. The insoluble materials
produced by chlorination may be highly dispersed and therefore coagulation and filtration
is required (Montgomery, 1986). Frequent backwash is needed for effective removal
(Singer, 1988).

Sodium Hypochlorite

Sodium hypochlorite also forms hypochlorous acid when dissolved in water. The sodium
hypochlorite reaction slightly increases the pH whereas the reaction of chlorine gas
slightly reduces the pH. Commercially available sodium hypochlorite has a concentration
of 12.5 %. Large quantities of sodium hypochlorite required to achieve adequate Mn
removal. Even though sodium hypochlorite is about twice the cost as equivalent chlorine

58
gas, sodium hypochlorite is used only in small systems due its ease of handling and safety
(Singer, 1988).

Potassium Permanganate

Although oxidation of manganese by KMn04 is slower than CI02 and ozone, it is most
extensively used in oxidation purposes as it is easily available in almost everywhere
(Minear et aI, 1982). Potassium permanganate is a stronger oxidant than chlorine and
sodium hypochlorite. Whilst it is effective in oxidizing manganese, it has also been used
for the treatment of taste and odor problems in water supplies (Boman et ai, 1999). Unlike
chlorine, the reaction of potassium permanganate with organic compounds will not
produce trihalomethanes but will actual1y reduce them (Singer, 1988). The stoichiometric
equation for manganese ion oxidation by potassium permanganate is given as below.

(12)

This reaction shows that alkalinity is consumed through acid production at the rate of
1.21 mg/L as CaC03 per mglL of Mn +2 oxidized. This consumption of alkalinity should
be considered when permanganate treatment is used along with alum coagulation, which
also requires alkalinity to form precipitates. According to the stoichiometric equation
(Table 3.1), it would require 1.92 mg of potassium permanganate to oxidize I mg of
manganese ion. In practice, the actual amount of potassium permanganate used has been
found to be less than that indicated by stoichiometry. It is thought that this is because of
the catalytic influence ofMn02 on the reactions (O'Connell, 1978).

Permanganate being highly reactive oxidant, adsorption of Mn(II) to the oxide surface is
the rate-limiting step that is rapid surface oxidation reaction is less effective for low
manganese concentration (Benschoten et aI, 1992). Where as in contrast for less reactive
oxidants like chlorine, surface adsorption is rapid relative to solution and surface
oxidation reaction. General1y a detention time of 5 to 15 minutes is recommended for
manganese removal (O'Connell, 1978).

According to a study by Desjardins, oxidation of manganese by potassium permanganate

occurred in less than 5 minutes where the manganese was not in a complexed form

59
 (Gravel and and Heertjes, 1975). As permanganate is very strong oxidizing agent, it is
capable to remove manganese over a wide pH range of 5-10 (Samblebe, 2003). But for
rapid oxidation it is preferable to raise pH above 7.0 (Benschoten et aI, 1992). Slight
overdosing of permanganate (up to 0.1 mglL) has been found not to cause any adverse
effects (Raveendran et aI, 200 I), .but presence of excess permanganate produce pink color
to the water. Although the cost of permanganate is more than that for chlorine, it has been
reported to be as efficient and may require considerably less equipments and capital
investment (Montgomery, 1986).

In order to re~ove manganese, potassium permanganate is usually added to solution

ahead of a fi.lter. After reaction, oxidized water is delivered to this filter media to remove
oxidized substances and as well as color (if produced). Usually manganese green sand, or
silica or even anthracite can be used as filter media (Montgomery, 1985). Greensand
media requires periodic regeneration with potassium permanganate solution.

Chlorine Dioxide

Chlorine dioxide or ozone is extremely rapid under most solution conditions (Benschoten,
and Lin 1992). Cl02 has relatively higher oxidation potential than ozone and potassium
permanganate. CI02 is capable to reduce residual manganese concentration to a level less
than 10 ugiL within 60-120 sec when manganese concentration is within 1,000 IlgiL
(Gregory and Carlson, 1996). The stoichiometric equation for manganese ion oxidation
by Cl02 is given as below (Benschoten and Lin, 1992).

From this equation it can be found that about 2.45 mg of CI02 is needed to oxidize I mg
of soluble manganese. However, as Cl02 is a strong oxidizing agent, during oxidation,
relatively less amount of dose is required (Gregory and Carlson, 1996). Best results are
obtained when the pH is greater than 7 (Stevens, 1982). Chlorine dioxide has also been
reported to oxidize organically bound manganese (Masschelein, 1979). One important
advantage of chlorine dioxide oxidation is that it does not produce THMs during
oxidation like chlorine (Stevens, 1982).

60
' .. ' :
However, excessive dosing may lead to increase in cr concentration especially at pH less
than 7 (Benschoten and Lin, 1992). The chlorine dioxide is a so strong oxidant that reacts
with organic material to produce a variety of oxidized by-products. Practicing indicated
that CI02- produced hemolytic anemia at lower ~xposure levels than those required to
produce significant increases in methemoglobin. Additional studies extended these
findings to chlorine dioxide and CI03-. Chlorite remained the most potent of the three
chemical species for causing signs of hemolytic oxidative stress in animals (Cao rui yu et
aI, 2001). These factors often limit the use of Chlorine dioxide as oxidant in removal of
manganese and iron (Cao rui yu et aI, 2001).

Ozone

Ozone is 12.5 times more soluble in water than is oxygen, leading to better mixing in
water treatment (Evans, 1972). Also, the products of its reaction with organics are
oxygen, carbon dioxide, and water. This prevents the incomplete disinfection products
that could lead to trihalomethanes (THMs) in drinking water. Ozone is effective against
odor-producers because it can easily oxidize these unsaturated compounds. Although
ozone is extremely rapid under most solution conditions, it is rarely practiced for
oxidation of manganese (Benschoten and Lin, 1992). Ozone may not be effective for
oxidation in presence of humic or fulvic materials. When ozone is applied to water,
excess air or oxygen is applied in sufficient quantities to supersaturate the dissolved
oxygen content of the water. The excess transferred oxygen is of concern due to its effect
on accelerating corrosion rates and out gassing via effervescence. The stoichiometric
equation for manganese ion oxidation by CI02 is given as below (Benschoten and Lin,.
1992).

2Mn2+ + 03 (g) + 2H20 2Mn02 (s) + 4H+ (14)

From stoichiometric relation it can be found that about 0.44 mg ozone is reqired for 1 mg
manganese oxidation. Oxidation with ozone often results in Mn2+ greater than 20 ug/L
and increasing TOC concentrations causes increasing Mn2+ residuals (Gregory and
Carlson, 1996). It is necessary to control appropriate dosing of ozone as overdosing due
to ozone's ability to oxidize Mn2+ to Mn7+ or permanganate. The formation of

61
permanganate and the resulting pink water are drawbacks to the use of ozone for
manganese removal (Montgomery, 1985).

Table 3.1 Stoichiometry of Manganese Oxidation

Hydrogen ion produced Alkalinity destroyed as
Dose required
Oxidizing agent as H+per mg ofMn CaCO, per mg ofMn
per mg ofMn
removal removal
Oxygen 0.29 (as 0,) 0.04 1.80
Permanganate, 1.92 (as
0.02 1.21
KMn04 KMn04)
Chlorine, CI, 1.29 (as Cl,) 0.07 3.64
Chlorine
2.45 (as CIO,) 0.06 0.73
Dioxide, CIO,

Ozone, 0, 0.44 as (0,) 0.05 1.82

- Not given(Source:Raveendran,2001; Benschoten and Lin, 1992; Gregoryand Carlson,1996)

3.4.8 Biological Oxidation

When manganese is in complexed with substances like humic acids, polyphosphates,

silica etc., removal of manganese by simple physical-chemical processes may become
ineffective. In such condition biological oxidation can be utilize to remove manganese
from water. In biological oxidation process manganese is oxidized by several types of
bacteria, such as Lepothrix, Crenothrix, Siderocapsa, Mettallogenium Pseudomonas etc
(Sharma et ai, 2001). These bacteria which can remove iron or manganese are referred to
as 'Iron Bacteria'. In general, these bacteria are found wherever there is a detectable level
of iron or manganese in water.

Similar to chemical oxidation, biological oxidation of manganese is highly dependent on

pH and redox potential of water. Manganese oxidation by 'iron bacteria' requires higher
oxidation - reduction potential (ORP) values (Eh > 300 mv) than that for iron and pH of
water should be greater that 7.5 for effective removal. Oxidation occurs according to
some variation of the following three step reaction (Gage et ai, 2001):
Mn2+ + 02 = Mn02 + Energy
Mn2++ Mn02 = Mn02oMn2+
Mn02oMn2+ + O2 = 2Mn02

62
Once a biological iron or manganese removal plant is constructed, the system must be
given time to 'seed' with bacteria naturally present in the water source. This seeded
biomass is naturally and continuously regenerated during the life of the plant and is
periodically partially removed through backwashing. For manganese removal plants the
seeding time can be considerably longer, anywhere from 3 weeks to 3 months (Gage et ai,
2001).

Advantages of biological oxidation process

- Smaller sized plants can be used because of higher applied filtration rates, (sometimes
in excess of 50 mlhr versus 10 - 15 mIhr) or because aeration and filtration can take
place simultaneously in the same vessel.
- Longer filter runs because manganese retention in the filter due to the formation of
more dense precipitates and the use of a more course media
- Produce denser backwash sludge that is easier to thicken and de-water
- Higher net productions due to less water being required for backwashing and being able
to use raw water for the backwash; less frequent backwashing is needed
- Require no chemical addition; and no deterioration of water quality over time;
- Lower capital and operating costs through the elimination of chemicals.

Disadvantages of .biological oxidation process

,
-Difficult to control as the process occurs naturally
-Responsible bacteria require more stringent conditions for effective oxidation
- Not suitable for small scale removal

3.4.8 Filtration

The filtration step involves the final removal of manganese from the water. It therefore is
a critical link in the process. There are two basic types of filters that are used; gravity high
rate filters and pressure filters. Basically, they include a means of introducing the water,
the filter media and a collection system for the filtered water. The collection system also
serves as a distribution system for the backwash water used to clean the filters. The

63
 selection of filtration media and operational cycle of a gravity filter is somewhat similar
to that of a pressure filter.

The media for the filters can include anthracite filter material, sands and manganese
greensand together with the support sands and gravels. If manganese removal is not
required then the filter can be anthracite and sand, sand only or anthracite only. On the
other hand, if manganese removal is required, then normally manganese greensand is
used. If there are any significant iron levels present, it is beneficial to have an anthracite
cap on top of the manganese greensand to protect it from a lot of iron sediment.

There are two basic operations associated with filtration. They are described below:

(a) Rate of Flow

The filtration step includes application of water uniformly to the top of the filter. Often
the rate of water application is described in relation to the area of the filter surface. The
application rate can be expressed in cubic meters per hour of water per square meter of
area (meters per hour) or gallons per minute per square foot. The rates that can be
appropriately used will depend on the raw water quality, the pretreatment provided and
the media used in the filter.

A typical rate for filtration would be 6 meters per hour (m/hr) or 2 gallons per minute per
square foot (2 gpm/ft') although some filters may have been designed for higher rates.
Normally, the lower rate will permit a better operation and subsequently less treatment
and filter problems. The under drain system should be so designed that the water can be
collected evenly from the filter. This type of system can be either a hub or lateral system
which has a spoke type configuration or it could be a system of evenly spaced nozzles or
collectors on a false bottom. The use of fine porous plates is normally not recommended
for manganese removal since the small pore size is susceptible to clogging.

(b) Backwashing
The cleaning or backwashing of a filter is one of the most important aspects of filter
operation. The process is to reverse the flow upwards from the under drain or distribution
system up through the filter and waste that water. The cleaning action arises from the
expansion of the bed and the rubbing of the filter particles so that all of the deposits

64
 become free and pass out in the wastewater. For effective backwashing, it is important
that the rate of water applied be sufficiently large to permit a good expansion of the bed.
This rate is also expressed as a water flow per unit area of filter. Typical backwash rates
would be in the order of 60 to 70 meters per hour or 10 to 12 gallons per minute per
square foot.

It is important that the distribution water be uniformly applied so that the entire filter bed
is expanded evenly. Backwashing is normally carried out until the wastewater turns clear.
The backwash water is then stopped and the filter is rinsed to waste before placing it into
service. For manganese greensand, an air scour or air wash system is also useful to ensure
the media becomes clean. This air wash is normally used during the backwash cycle after
the filter has been initially flushed. It is also important that treated water is used to
backwash and clean the filter media, particularly in the case of manganese greensand.

A brief description of common manganese treatment options and their suitability is shown
in tabular form in table 3.2.

3.5 TREATMENT TYPES NOT RECOMMENDED

3.5.1 Magnetic manganese Removal Devices

The Water Quality Association (the professional association representing the home water
treatment industry) has indicated that there is no proof that magnetic manganese removal
devices are effective (Cameron an Bourgin, 1995).

3.5.2 Electrodialysis

This process will become clogged by any rust particles, manganese bacteria, silt etc. The
treatment membranes cannot be rejuvenated and new membranes will be necessary. This
equipment is very expensive to purchase and operate (Cameron and Bourgin, 1995).

65
Table 3.2 Treatment of Manganese and Iron in Drinking Water

Indication Cause Treatment

Water clear when drawn but Dissolved iron or • Water softener «5 mg/I combined
red-brown or black particles manganese concentrations of iron and manganese)
appear as water stands; red-
brown or black stains on • Oxidizing filter (manganese greensand
fixtures or laundry or zeolite) « 15 mg/I combined
concentrations of iron and manganese)

• Aeration (pressure) «25mg/l combined

concentrations of iron and manganese)

• Chemical oxidation with potassium

permanganate or chlorine; followed
with filtration (> I0 mg/I ironl
manganese)

Water contains black Oxidized manganese Particle filter (if quantity of oxidized
particles when drawn; due to exposure of material is high, use larger filter than
particles settle out as water water to air prior to tap inline; e.g., sand filter)
stands

Black slime appears in toilet manganese bacteria Kill bacteria masses by shock treatment
tanks or from clogs in with chlorine or potassium
faucets permanganate, then filter; bacteria may
originate in well, so it may require
continuous feed of chlorine or potassium
permanganate, then filter

Black color that remains Colloidal manganese; Chemical oxidation with chlorine or
longer than 24 hours organically complexed potassium permanganate; followed with
manganese filtration
(Source: Varner et ai, 1994)

66
; ".;;'
3.5.3 Reverse Osmosis

This process will become clogged by rust particles, manganese bacteria, silt, etc. and
cannot be rejuvenated. New membranes would be required. Like eletrodialysis process,
this method is also expensive to operate (Cameron and Bourgin, 1995).

3.5.4 Bag Filtration

This method uses chemicals to cause formation of rust particles. The particles are
removed from the water by passage through bag filters. The cost of this system is
comparatively low. However, the bags must be manually cleaned which creates higher
operational costs (Cameron and Bourgin, 1995).

3.6 SEQUESTERING PROCESS: PHOSPHATE TREATMENT

Sequestering of soluble manganese is the opposite of oxidation. It means to bind up or

complex forming so as to prevent natural chemical reaction (Montgomery, 1986).
Chemical used for sequestering is sodium hexametaphosphate, commonly known as
polyphosphate. Low levels of up to 2 mg/I can be remedied using phosphate compound
treatment (Singer, 1988). Phosphate compounds are a family of chemicals that can
surround minerals and keep them in solution. Phosphate compounds injected into the
water system can stabilize and disperse dissolved manganese at this level. As a result,
manganese compounds are not available to react with oxygen and separate from solution.
The phosphate compounds must be introduced into the water at a point where the iron is
still dissolved in order to maintain water clarity and prevent possible iron staining. This
should be before the pressure tank and as close to the well discharge point as possible
(Varner et aI, 1994).

Phosphate compound treatment is a relatively inexpensive way to treat water for low
levels of iron and manganese. Phosphate treatment is effective in the pH range of 5.0 to
8.0 (NRWA, 2004). Since phosphate compounds do not actually remove manganese,
water treated with these chemicals will retain a metallic taste. In addition, too great a
concentration of phosphate compounds will make water feel slippery (Varner et aI, 1994).

67
Phosphate compounds are not stable at high temperatures. If phosphate compound-treated
water is heated (for example, in a water heater or boiled water), the phosphates will break
down and release iron and manganese. The released iron and manganese will then react
with oxygen and precipitate.

Adding phosphate compounds is not recommended where the use of phosphate in most
cleaning products is banned. Phosphate, from any source, contributes to excess nutrient
content in surface water (Seelig, 1992).

68
 CHAPTER 4

MANGANESE REMOVAL FROM GROUNDWATER BY

OXIDATION

4.1 INTRODUCTION

Besides arsenic, iron and salinity (in coastal areas), excessive concentration of manganese
is also a significant groundwater quality problem in many areas of Bangladesh (DPHEI
BGS, 2001). Presence of manganese in potable water may cause problems related to
aesthetics (e.g., development of color in clothes and plumbing fixtures) and may cause
precipitation in the water distribution system. Evidence of manganese neuro-toxicity has
been found in people following long-term exposure. The World Health Organization
(WHO) has a provisional health-based guideline value of 0.5 mg/l for manganese in
drinking water (WHO, 1993). The WHO guideline value from aesthetics consideration is
0.1 mg/I. Bangladesh Standard for manganese in drinking water is also 0.10 mg/I. In the
nationwide groundwater-quality survey (DPHEI BGS, 2001), it has been found that about
35% of sample wells exceeds WHO health-based guideline value (0.5 mg/l) and about
73% of samples exceed Bangladesh guideline value (0.1 mg/l). Manganese concentration
as high as 10 mg/l has been detected in this survey. Therefore, development of
appropriate manganese removal technology in the context of Bangladesh is of prime
importance.

Broadly manganese removal technologies can be categorized in two groups: (i) ion
exchange and (ii) oxidation followed by precipitation of manganese in insoluble form(s).
As described in Chapter 2, chemical oxidation followed by filtration, is by far the most
widely used manganese removal technique. Common chemical oxidants include
potassium chlorine, ozone, and permanganate. Besides, simple aeration has also been
used for oxidation and subsequent removal of manganese. Efficiency and effectiveness of
manganese removal depend on a range of factors including type of oxidant used, initial
manganese concentration, pH, alkalinity, and a range of other water quality parameters.
Though considerable work has been done elsewhere in the world, very limited works
(e.g., BAMWSPI DFIDI Water Aid, 2001; Tahura et aI, 2001) have so far been done on
the effectiveness of different oxidizing agents in removing manganese from groundwater
of Bangladesh.

In Bangladesh, bleaching powder and potassium permanganate are easily available and
these chemicals are also widely used for oxidation of As(II) to As(V) in many arsenic
removal systems (Tahura et ai, 2001). It may be mentioned that aeration followed by
filtration is often used for removal of dissolved iron from groundwater in Bangladesh.
However, vary limited data are available on effectiveness of these chemicals and
processes in removing manganese from water (e.g., Tahura et ai, 2001).

In this study, removal of manganese from natural groundwater by the oxidation process
has been assessed using two commonly available oxidizing agents, potassium
permanganate and chlorine. Effectiveness of manganese 'removal by simple aeration has
also been evaluated. Besides, removal of color produced during permanganate oxidation
of manganese has been assessed by using sand filtration. This chapter presents results of
manganese removal from natural groundwater by aeration and chemical oxidation using
potassium permanganate and chlorine (bleaching powder).

4.2 MATERIALS AND METHODS

In this study, the effectiveness of two oxidizing agents, bleaching powder and potassium
permanganate, in removing manganese from groundwater by the oxidation process has
been evaluated. Laboratory batch experiments were carried out to assess removal of
manganese from groundwater under various conditions. In this study groundwater
collected from a deep tubewell pumping stations at BUET was used. The advantage of
using this water is that the concentrations of both manganese and iron in this water are
very low, and therefore concentrations of these two parameters could be varied by
appropriate spiking. Table 4.1 shows the characteristics of groundwater used in this study.

All batch experiments were carried out in I-L glass beakers. In a typical batch
experimental set up, 500-mL groundwater sample was taken in each of a series of 1-L
glass beakers. Initial manganese concentrations in the groundwater samples were varied
by spiking with a stock solution of manganese (containing 500 mg/I of Mn), prepared by
dissolving anhydrous manganous sulfate salt (MnS04.HzO) in deionized water.

70
Required doses of the oxidizing agents (either potassium permanganate or bleaching
power) were added to the beakers by from stock solutions of potassium permanganate and
bleaching powder, respectively.

After addition of the oxidizing agent, the content of each beaker was mixed in a jar test
apparatus at 100 rpm for 10 minutes (when KMn04 was used as oxidant; Fig. 4.1) or 30
minutes (when bleaching powder was used as oxidizing agent). The samples were then
kept at rest for 30 minutes to allow the manganese solids, formed as a result of oxidation,
to precipitate. The pH and Eh of the water samples in each beaker were then measured.
About 50 ml of water sample was then drawn from each beaker from a depth of about 1-
cm below the surface of liquid. These samples were tested for total manganese, color
(during chemical oxidation) and residual chlorine (when bleaching powder was used as an
oxidizing agent). Part of the sample withdrawn from each beaker was filtered through a
0.45 J.!m filter, and the filtrate was analyzed for total manganese. Each experiment was
carried out in duplicate. Similar experiments were carried out to assess the effect of pH on
manganese removal by both the oxidizing agents. The pH of water sample was adjusted
by addition of solution of either NaOH or Hel as required.

Figure 4.2 shows a flow diagram of the experimental procedure followed in this study.
Besides, additional experiments were carried out to evaluate the removal of manganese
from groundwater by aeration. Experimental evaluation was also done to determine the
effectiveness sand filter to remove color produced from oxidation of manganese with
potassium permanganate. Additional details of each type of experimental set up are
briefly described below.

4.2.1 Manganese Removal by Oxidation with KMn04

Available information (e.g., Raveendran et aI, 2001) suggest that oxidation of manganese
by potassium permanganate occurs in less than 5 minutes where the manganese was not
present in complexed form. In this study, a mixing time of 10 minutes was allowed for
complete oxidation of manganese.

71
Effect of initial manganese concentration on it removal efficiency was evaluated by
varying the initial concentration of manganese from about 1.0 to about 10.0 mg/L. Initial
pH of the samples were measured to be within the range of 7.7 to 7.8. Required dose of
permanganate was calculated from the stoichiometry of Eq. 4.1. According to Eq. 4.1, for
each mg/l of dissolved manganese, the required dose of potassium permanganate for
complete oxidation of manganese is 1.917 mg/I. For these experiments, concentration of
permanganate was fixed at 1.2 times that required from stoichiometric consideration.

2+. + +
3 Mn + 2 KMn04 + 2 H20 = 5 Mn02 (s) + 2K + 4H (4.1)

To evaluate effect of pH on manganese removal, experiments were carried out with an

initial manganese concentration of about 2.0 mg/I. For these experiments, permanganate
dose was set at exactly that required from stoichiometric consideration (i.e., 2.84 mg/I of
KMn04 for 2.0 mg/I of manganese). The pH of the water samples were varied from about
4.8 to 10.0, by either NaOH (N/44) or Hel (concentrated) solutions.

Experiments were carried out to evaluate the effect of the dose of potassium
permanganate (which also contains manganese) on the removal of manganese and the
presence of residual manganese. These experiments were carried out with and initial
manganese concentration of 2.0 mg/I. The dose of potassium permanganate was fixed at
six different factors of that required from stoichiometric ratio. The factors were 0.25, 0.5,
0.8, 1.0, 1.2, and 1.5. That is the permanganate dose was varied from 0.25 to 1.5 times
that required from stoichiometric consideration.

The effect of settling time on manganese removal was determined, in similar

experimental set up, in order to assess the settling of manganese solids by gravity. Initial
manganese concentration was set at 1.0 mg/I, 2.0 mg/I, 5.0 mg/I, and 10.0 mg/I.
Permanganate dose was fixed at 1.2 times that required from stoichiometric
consideration). In these experiments, after the initial mixing of 10 minutes (at 100 rpm),
the contents of the beaker were allowed to settle and water samples were collected (from
a depth of about 1 cm below the water surface) at time intervals of 30, 90 and 180
minutes.

72
 -
~{rrr
,~.~- ,~b
.~
•

- .,,;;,;,r

-:=:~

Figure 4.1: Mixing of KMn04 with groundwater containing Manganese

Water sample

Procedures Measurement

Add required Mn stock solution

Mn conc., pH, Eh
Adjust pH (ifrequired) [NaOHlHCI]
pH
Add oxidant (KMnOJ Cl,)

Stir for specific detention time

Allow settling for 30 mins or
more ( if re uired
pH,Eh
Collect supernatant
Residual Mn, Color I residual Cl,
Filter sample (0.45 urn)

ResidualMn

Figure 4.1: Flow diagram of experimental procedure followed for evaluating manganese
removal by oxidation process

73
In manganese removal processes involving oxidation with potassium permanganate, it is
customary to use some kind of filtration device to remove the manganese solids as well as
the color produced due to addition of potassium permanganate. In Bangladesh sand
filtration is common in many water treatment processes (e.g., arsenic and water treatment
processes). Hence efficiency of sand filtration in removing manganese solids (formed as a
result of oxidation with potassium permanganate) and color was evaluated in this study.
The experimental set up is similar to that used by Ali et al. (2001). Sand filters were
prepared in glass burettes having a cross-sectional area of 1 sq. cm. A pre-washed locally
available sand sample was oven-dried at 105° C for 24 hrs.

The oven-dried sand was sieved and the portion of sand passing sieve # 30 and retained
on sieve #40 was selected as filter media. This portion of the sand sample was then
poured in to the burettes. Care was taken to ensure that no void space existed between the
sand particles and the burette. Experiments on color removal were carried out for two
different depths of sand filter, 10 cm and 20 cm.

For these experiments, water samples (3-litres) having initial manganese concentration of
about 2.0 mg/I were treated with potassium permanganate at a dose 1.2 times that
required from stoichiometric consideration. After the initial mixing for 10 minutes (at 100
rpm), the water sample was passed though the sand column. The flow rate was controlled
to maintain a minimum contact time of I minute. The filtrate coming out of the bottom of
the burette was collected at 30-minute interval and was analyzed for residual manganese
and color. The experiments were carried out for a period of about 150 minutes.

4.2.2 Manganese Removal by Oxidation with Bleaching Powder

Available information (e.g., Seelig, 1992) suggest that oxidation of manganese by

chlorine requires a contact time of about 20 minutes. In this study, a mixing time of 30
minutes was allowed for complete oxidation of manganese by chlorine (added in the form
of bleaching powder).

Effect of initial manganese concentration on it removal efficiency was evaluated by

varying the initial concentration of manganese from about 1.0 to about 10.0 mg/L.
Required dose of chlorine (in the form of bleaching powder) was calculated from the

74
stoichiometry ofEq. 4.2. According to Eq. 4.2, for each mg/l of dissolved manganese, the
required dose of chlorine for complete oxidation of manganese is 1.29 mg/l. For these
experiments, concentration of chlorine was fixed at 1.2 times that required from
stoichiometric consideration. As oxidation by chlorine is known to be strongly dependent
on pH (Benschoten et ai, 1990), these experiments were carried out at three different pH
values - 7.5, 8.5 and 10.0.

Mnz+ +Ch +2HzO = MnOz(s) +2cr+4H+ (4.2)

To evaluate effect of pH on manganese removal, experiments were carried out with initial
manganese concentration varying from about 1.0 mg/l. For these experiments, chlorine
dose was set at exactly that required from stoichiometric consideration (i.e., 2.58 mg/l of
Chlorine for 2.0 mg/l of manganese). The pH of the water samples were varied from
about 5.2 to 10.0, by either NaOH (N/44) or HCl (concentrated) solutions.

A set of experiments was carried out to assess the effect of initial mixing time on
manganese removal. For this purpose, initial contact time was fixed at 15, 25 and 35
minutes. For this set of experiments, initial manganese concentration varied from about
1.0 mg/L to 10.0 mg/l and pH was fixed at a value of about 10.0.

4.2.3 Manganese Removal by Aeration

Manganese removal by simple aeration was also evaluated in this study. The
experimental set up for this purpose was similar to that used for manganese removal by
an oxidizing agent, except that instead of adding an oxidizing agent, the water samples in
the experimental beakers were aerated by vigorous mixing.

Available literature (e.g., Montgomery, 1986) suggest that for removal of manganese by
simple oxidation, a contact time of several minutes to an hour may required for oxidation
of manganese, depending on manganese concentration as well as some other parameter,
e.g., pH. In this study, aeration (by vigorous mixing) was carried out for a period varying
from 20 minutes to an hour. The mixing was done in a jar test apparatus, where mixing
was carried out at 100 rpm. After completing the mixing, samples were filtered using 0.45
/-Lmfilter and analyzed for residual manganese.

75
For these experiments, initial manganese concentration was fixed at 2 mg/I and pH was
varied from 7 to 11.

4.2.4 Chemicals, Preparation of Stock Solutions

All chemicals used in this study were of reagent grade. Stock solutions were prepared by
dissolved appropriate salts to deionized water (Barnstead Fistreem III).

Manganese stock solution (containing about 500 mg/I of Mn) was prepared by dissolving
anhydrous manganous sulfate salt (MnS04.H20; MW = 169.04) in deionized water. The
stock solution was kept at a pH below 2.0 by acidifying with concentrated HCI. Stock
solution of potassium permanganate (containing about 500 mg/I KMn04) was prepared
by dissolving KMn04 crystals (MW = 158.07) in deionized water. The stock solution was
kept in dark. A chlorine stock solution (having about 400 mg/I of Chlorine) was prepared
by dissolving bleaching powder (with chlorine content of about 32%) in deionized water.

In this study, manganese concentration in water samples was measured by flame atomic
absorption spectrophotometry, using an AAS (Shimadzu, 6800). A sample standard curve
for manganese is shown in Fig. Al in Appendix A. Measurement of pH was carried out
with a digital pH meter (HACH, Sension I) and Eh was measured with an Eh meter
(WTW, Multiline P4). Color was measured with a spectrophotometer (HACH, DR2010).
Residual chlorine in water samples was also measured with the spectrophotometer (by
chlorine total DPD method). Other parameter was measured following standard methods
(AWW A, 2002).

4.3 RESULTS AND DISCUSSIONS

4.3.1 Removal of Manganese by Oxidation with KMn04

This section describes the results of batch experiments designed to remove manganese
from groundwater by oxidation with potassium permanganate. Batch experiments were
conducted primarily to evaluate: (i) the effect of initial manganese concentration on
manganese removal, (ii) effect of pH on manganese removal, (iii) effect of permanganate
dose on manganese removal, and (iv) effect of settling time on manganese removal.
Results from each set of experiment are described below.

76
 Effect of Initial Manganese Concentration on Manganese Removal
Mechanisms involved in the removal of dissolved manganese from groundwater as a
result of oxidation include: (i) oxidation (in solution) of dissolved Mn(H) into insoluble
manganese solids [e.g., Mn02(S), MnZ03(s), Mn304(S), and Mn(OH)z(s), MnOOH(s)]; (ii)
surface mediated oxidation of Mn(H); and (iii) adsorption of dissolved Mn(H) onto oxide
surfaces (e.g., on the insoluble manganese oxide surfaces). Permanganate being a highly
reactive oxidant, oxidation (in solution) of Mn(H) into insoluble manganese solids is the
primary removal mechanism (Benschoten et aI., 1992).

Figure 4.3 shows percentage of removal of manganese for different initial manganese
concentrations, varying from about 1.0 mgll (0.0182 mM) to about 10.0 mgll (0.182
mM). The permanganate dose for these experiments was fixed at 1.2 times that required
from stoichiometric consideration. Figure 4.3 shows very good removal of manganese
(expressed as "% removed") for all different initial manganese concentrations. In fact, for
initial manganese concentration of up to about 2.0 mgll, almost complete removal was
achieved (i.e., residual manganese concentration below the MDL of 0.001 mg/l). For
initial concentration ranging from about 4 to 10 mgll, % removal approached about 98%.

Figure 4.4 shows residual manganese concentration (after filtration) for different initial
manganese concentration. It shows that although manganese removal, when expressed "%
Mn removed", does not appear to depend .significantly on initial manganese concentration
(Fig. 4.3), in terms of residual manganese the effect is not insignificant. For example, for
an initial manganese concentration of about 10.0 mgll, 98% removal means a residual
concentration of 0.20 mgll, which does not satisfy the Bangladesh drinking water
standard as well as the WHO guideline value (from aesthetics consideration). However,
the maximum residual concentration of manganese (0.22 mgll for initial manganese
concentration of about 10.0 mgll) was much below the WHO health-based guideline
value.

77
 105 -,------------------------------

o 2 4 6 8 10 12
Initial Mn Concentration (mg/L)

Figure 4.3: Removal of manganese by oxidation with KMn04 for different initial
manganese concentrations (stoichiometry ratio = 1.2; Initial pH =7.7)

0.25

/+
bn
E
~ 0.2
.~

/.
1;;
•...
Bangladesh
""•... 0.15
i-'< drinking water
standard
"
q:::
-<
0.1
~
/
'"
~
0.05
"'"
P::

a
0 2 4 6 8 10 12

Initial Mn Concentration (mg/L)

Figure 4.4: Residual manganese remaining in solution after filtration for different initial
manganese concentration after oxidation with KMn04.(Stoichiometry ratio
= 1.2)

78
In previous studies (e.g., Marble et aI., 1999), it has been found that it is more difficult to
remove Mn when the initial concentration is low, regardless of the oxidant used.
However, for the lowest initial manganese concentration (about 1.0 mg/I) used in this
study, this effect was not apparent. In comparIson with the study conducted by
Raveendran et al (2001), better removal is achieved in this study. In the study of
Ravvendran et al (2001), for an initial manganese concentration of about 0.3 mg/L only
50% removal of manganese was achieved for KMn04 dose of 1.2 times of that obtained
from stoichiometric consideration. Since pH of water sample for the different experiments
varied over a relatively narrow range of 7.44 to 7.90, pH does not appear to have a
significant effect on removal of manganese among the different experiments.

Results presented in Figs. 4.3 and 4.4 suggest that although oxidation by potassium
permanganate (followed by filtration) could remove significant manganese from
groundwater, for higher initial concentrations exceeding about 5.0 mg/I manganese, the
residual manganese achievable may not satisfy the Bangladesh drinking water standard of
0.10 mg/I.

Measured Eh values for these experiments varied from a low of 230 mY for a
permanganate dose of 2.3 mg/I (for I mg/I of manganese) to a high of 556 mY for a
permanganate dose of about 23 mg/I (for 10 mg/I of manganese). As noted earlier pH for
these experiments varied from 7.44 to 7.90. For these ranges of pH and Eh, the Eh-pH
diagram of manganese (Fig. 3.3 of Chapter 3) suggest that the precipitated solids would
be either Mn20](s) (dark brown to black) or Mn]04(s) (reddish brown). The precipitated
foim would change from Mn]04(s) to Mn20](S) as Eh value increases.

The observed color of the precipitates for different experimental conditions matched well
with the color of the expected precipitate for that particular condition. For example,
figures 4.5(a) shows photographs of precipitates of manganese solids (reddish brown) for
an initial manganese concentration of 4.98 mg/I, for which measured pH and Eh were
7.74 and 225 mY, respectively. According to Fig. 3.3 of Chapter 3, for this condition the
most likely precipitate is Mn]04(s), having a reddish brown color, which matched well
with the observed color. Similarly, Fig. 4.5(b) shows dark brown colored precipitates of
manganese solids for an initial manganese concentration of 10.02 mg/I, for which
measured pH and Eh were 7.44 and 556 mY, respectively. According to Fig 3.3 of

79
ttl
" -~ --
 77

Figure 4.5 (a) Color of precipitation for water

with initial manganese concentration 5 mgIL
(pH = 7.4; Eh = 230 mY)

Figure 4.5 (b) Color of precipitation for water

with initial manganese concentration 10 mgIL
(pH = 7.9; Eh = 560 mV)

80
Chapter 3, for this condition the most likely precipitate is MnzO](s), having a dark brown
to black color.

Effect of pH on Manganese Removal

In this study, effect of pH manganese removal was evaluated in batch experiments with
an initial manganese concentration of about 2.0 mg/l, with a permanganate dose set at
exactly that required from stoichiometric consideration (i.e., 2.84 mg/l of KMn04 for 2.0
mg/l of manganese). The pH of the water samples were varied from about 4.8 to 10.0.
Figure 4.6 shows removal of manganese as a function of pH. Thus pH appears to have a
major influence on the removal of manganese. In general, removal increased as pH
increased. Beyond pH of about 8, removal is almost independent of pH.

It should be noted that oxygenation kinetics equation for manganese clearly shows that
rate of reaction of Mn (II) has a second order relationship with hydroxyl ion
concentration, which indicate that an increase in one pH unit, cause 100 fold increase in
rate of reaction (Stumm and Morgan, 1984).

The pH of water also affects manganese removal by influencing the adsorption of

dissolved Mn(II) on the MnOz(s) formed as a result of oxidation. According to Marble et
al (1998) overall mass transfer of Mn(II) from solution to active sites at the surface
decreases as pH decreases because of competition with H+. The capacity of oxide surface
at pH 9 is about 2 mole of Mn(II) adsorbed per mole of MnOz (s). However,
permanganate being a very strong oxidizing agent, it is capable to oxidize manganese
over a wide pH range of 5-10 (Samblebe, 2003), though for rapid oxidation it is
preferable to raise pH above 7.0 (Benschoten et aI., 1992).

Thus, results from this study appear to agree well with those observed by other
researchers (e.g., Raveendran,et aI, 2001). Since pH of natural groundwater in
Bangladesh usually falls around the neutral range, manganese from such groundwater
could be effectively removed by potassium permanganate without any pH adjustment.

81
 110

~
"'" 100
c
0
• •
~
u:: 90
~(])

'«"
80
~
E
(])
a::
c 70
:;;

60
4 5 6 7 8 9 10 11
pH

Figure 4.6: Removal of manganese by oxidation with KMn04 for different pH value
(Initial Mn concentration= 2.0 mg/L; KMn04 stoichiometry ratio = 1.2)

Effect of Permanganate Dose on Manganese Removal

According to the stoichiometric equation for manganese oxidation by potassium
permanganate (Eq. 4.1), it would require 1.92 mg of potassium permanganate to oxidize 1
mg of manganese ion. In practice, the actual amount of potassium permanganate used has
been found to be less than that indicated by stoichiometry. It is thought that this is
because of the catalytic influence of MnOz on the reactions (O'Connell, 1978). Unused
permanganate, if any, would contribute to be increased residual manganese concentration.

In this study, effect of permanganate dose on manganese removal was evaluated in batch
experiments with an initial manganese concentration of about 2.0 mg/l, where
permanganate dose was varied from 0.25 times to 1.5 times that required from
stoichiometric consideration. For 12 sets of experiments carried out for this purpose, pH
varied from 7.56 to 8.10.

Figure 4.7 shows removal of manganese for different doses of potassium permanganate
(expressed as multiple of stoichiometric ratio). It shows that manganese removal was low
(about 50%) for a stoichiometric ratio of 0.25, and has increased as permanganate dose
increases from 0.25 up to a stoichiometric ratio of 1.0. At a stoichiometric ratio of 1.0,
!

82
manganese removal approaches 100%. At a stoichiometric ratio of 1.2, manganese
removal decreased slightly and was about 97%, with residual manganese concentration of
about 0.05 mg/I, below the Bangladesh drinking water standard of 0.1 0 mg/1. At even
higher stoichiometric ratio (1.5), manganese removal decreased significantly. This was
because of the presence of unused potassium permanganate that contributes to the
residual manganese. At a stoichiometric ratio of 1.5, manganese removal drops to about
85% with residual manganese concentration of about 0.30 mg/1.

120

~ 100
"--"
0

~ 80
'"iI
:;;
'"
«
"0
60

91
0
E 40
"
0::
"
:;; 20 -&--SetA
-G-SetB
a
a 0.25 0.5 0.75 1 , 1.25 1.5 1.75

Pennanganate Dose (as Multiple Fraction of Stoichiometric Ratio)

Figure 4.7: Removal of manganese by oxidation with KMn04 for different doses of
KMn04 expressed as multiple of stoichiometric ratio
(Initial Mn concentration= 2.0 mg/L)

Thus, results from this study show maximum manganese removal at a permanganate dose
equal to that required from stoichiometric consideration. However, a slight overdose of
permanganate (e.g., for ensuring a factor of safety) would not affect residual manganese
concentration significantly when initial manganese concentration is below 5.0 mg/L.
However, for higher values of initial Mn concentration (> 5.0 mg/L) is a concern.

As noted earlier, pH value of these experiments varied from 7.56 to 8.1, which is a
favorable range for manganese oxidation. Permanganate dose may have a higher
influence on manganese removal below neutral pH range.

83
Eh values measured for these experiments varied from 174 mY to 230 mY; generally
increasing with increasing permanganate dose. For this range of Eh and pH values
(varying from 7.56 to 8.1), the precipitated solid would most likely be Mn]04(s),
according to the Eh-pH diagram shown in Fig 3.3 in Chapter 3. As before, the precipitates
were found to be reddish to dark brown.

Effect of Settling Time on Manganese Removal and Color

In order to evaluate effect of settling time (detention time) on removal of manganese (by
gravity settling of manganese solids) and color, experiments were carried out with initial
manganese concentration varying from about 1.0 to about 10.0 mg/1. Potassium
permanganate dose was set at 1.2 times that required from stoichiometric consideration.
These experiments are the same as those carried out to assess the effect of initial
manganese concentration on manganese removal.

Figure 4.8 shows removal of manganese from solution as a function of time for different
initial manganese concentrations. It shows that for a particular initial manganese
concentration, manganese removal (by gravity settling of manganese flocs) increases with
increasing settling/ detention time. For example, for an initial manganese concentration of
about 2.0 mg/I, manganese removal (for a sample taken from l-cm below the surface of
water) after 30 minutes was about 86%, after 90 minutes about 96%, and after 3 hours
about 100%. Figure 4.8 shows that after about 3 hours of detention time, manganese
removal by gravity settling approaches that achievable by filtration (with a 0.45 m
filter).

Figure 4.9 shows residual manganese concentration (for water sample collected form 1-
cm below water Burface) in water as a function of settlement! detention time for different
initial manganese concentrations. It shows the with prolonged detention time (3 hours),
residual manganese concentration could be brought down to levels satisfying the WHO
standard (health-based). But however, for initial manganese concentration equal to or
greater than 5.0 mg/L, residual manganese concentration was above Bangladesh standard
(WHO aesthetic-based standard) even after allowing 3 hrs settlement/detention time.

84
 110

100

C 90
<::
:2:
'5 80
!ii
0
.,
E
oc 70

60
-.-
___ after 30
after 90 nins
nins
-/i.- after 3 hrs
50 -x- after filtration
0 2 468 10 12
Initial Mn Concentration (mg/L)

Figure 4.8: Removal of manganese by oxidation with KMn04 as a function of initial

manganese concentration for different settling times. (KMn04
stoichiometric ratio =1.2; Initial pH = 7.7)

3
M1 Concentration
-..- lrrg/L
-<>-2 rrg/L
2.5 -.- 5 rrg/L
--G- 7.5rrg/L

:::; -.-101TQ!l
0,
.s 2
<::
.2
'5
o
(J)
1.5
WHO standard
.!:
<:: (health-based)
:2:
ro:J
-0
~
~ 0.5
'---.-----::::=::::
o •
o 30 60 90 120 150 180 210
Settlement (Detention )lime (mins)

Figure 4.9: Residual manganese concentrations as a function of settling time after

oxidation with KMn04 for different initial manganese concentrations.
(stoichiometric ratio = 1.2; Initial pH 7.7 ).

Figure 4.10 shows concentration of color (in Pt.-Co. Unit) in water at different times after
addition of potassium permanganate. It shows that for a particular time interval, color
increases as initial manganese concentration (and hence corresponding potassium

85
permanganate dose) increases. For example, for an initial manganese concentration of
about 1.0 mg/I, the potassium permanganate dose was 2.3 mg/I and the corresponding
color, after 30 minutes, varied from 78 to 82. For initial manganese concentration of
about 10.0 mg!l, the potassium permanganate dose was 23.0 mg/I and the corresponding
color, after 30 minutes, varied from 549 to 556. Thus, color concentration appears to be
proportional to the dose of potassium permanganate added.

500

•
~ 400
'c
y"
[ 300
c
:so
<5
(/)
200
.S
~
o
o
U 100
-+- after 30
___ after 90 mns
mns
-Is- after3 hrs
o
o 5 10 15 20 25
Potassium Pennanganater Dose (mg/L)

Figure 4.10: Color remaining in solution as a function KMnO. doses for different settling times.
(Stoichiometry ratio= 1.2; Initial pH 7.7)

Figure 4.10 also shows that for a particular initial manganese concentration (and hence
for a particular permanganate dose), color decreases with time as a result of settling. For
example, for an initial manganese concentration of about 2.0 mg/I, the potassium
permanganate dose was 4.6 mg/I and the corresponding color (average value) after 30
minutes was 140 Pt.-Co. unit, after 90 minutes 54 Pt.-Co. unit, and after 3 hours 24 Pt.-
Co. unit.

Figure 4.11 shows color of water samples as a function of settling/ detention time for
different initial manganese concentration. It shows that except for the lowest initial
manganese concentration (i.e., 1.0 mg/I), all color concentrations were above the
Bangladesh drinking water standard, even after 3 hours of settling/ detention time. Thus,
it is clear that settling! detention alone would not remove color from water treated with

86
 potassium permanganate. Hence, appropriate filtration would be required to remove.
color, although it appears that manganese solids (produced as a result of oxidation wit?
potassium permanganate) could be removal by prolonged gravity settling.

500
__ -M1=1.0
-0- fv1n=2.0
-/1- fv1n=5.0
-+-rv'r1=7.5
400 -e-M1=10

~
~
~ 300
e;.
0>
.1"
'ro 200
E
i"
o
8 100

o
o 30 60 90 120 150 180 210

Settlement /Detention lime ( mins)

Figure 4.11: Color remained in solution as a function of settlement time for different amount of
initial manganese concentration (Stoichiometry ratio= 1.2)

Removal of Color and Manganese by Sand Filtration after Oxidation with Potassium
Permanganate
Figures 4.12 shows removal of color as a functionofbed volume of liquid passed through

.
the 10-cm and 20-cm deep sand filter column. Initial concentration of color of the water
samples, having an initial manganese concentration of about 2.0 mgll and treated with.
potassium permanganate, were 244 and 225 Pt.-Co. unit, respectively. Significant color
removal was achieved with both the filters. However, for the lO-cm filter, measured color
slightly exceeded the Bangladesh drinking water standard (15 Pt.-Co.) for bed volume of
up to about 80.

Figures 4.13 shows removal of manganese as a function of pore volume of liquid passed
through the 10-cm and 20-cm deep sand filter column. Manganese concentration in the
filtrate varied from 0.01 to 0.04 mg/l for both the sand filters. In similar experiments
performed earlier in this study, residual manganese concentration. after filtration with a
0.45 flm filter paper was < 0.001 mgll. Though such low levels of manganese could not

87
be achieved by sand filtration, all residual manganese concentrations (after sand filtration)
were well below the Bangladesh drinking water standard of 0.1 0 mg/l.

40
-+- 1O-em sand filter
-13- 20-cm sand filter
35

-
'E
::> 30
0
,
U
25 Bangladesh
[ drinking
c
0 20 water "'"
~ standard ~
(5
(f)
15
c
~
0
(5 10
U 8_________ +-
G +
5
~[j 0
0
0 20 40 60 80 100 120 140
Bed Volume Passed

Figure 4.12: Color remaining in solution after passage through sand filter column as a
function of pore volume of liquid passed. (Initial Mn concentration = 2.0
mg/L; Stoichiometry ratio for KMn04 = 1.2)

0.1 -,------------ ,---:--:-;:- __ -:;-;:;,....,

-+-10-cmsand filter
1
-G- 20-cm sand filter

Bangladesh
Standard

0+----.----,-------,--_--.- .-__ ~--__1

o 20 40 60 80 100 120 140
Bed Volume Passed

Figure 4.13: Residual manganese remaining in solution after passage through sand filter
column. as a function of pore volume of liquid passed. (Initial Mn
concentration 2.0mg/L; Stoichiometry ratio for KMn04 = 1.2)

88
 4.3.2 Removal of Manganese by Oxidation with Bleaching Powder
Effect of Initial Manganese Concentration and pH on Manganese Removal

Figure 4.14 shows manganese removal by oxidation with chlorine (added in the form of
bleaching powder) at different pH values for different initial manganese concentrations
(varying from about 1.0 mg/I to about 10.0 mg/I). The chlorine dose was fixed at 1.2
times that required from stoichiometric consideration. As explained earlier, experiments
were carried out at three different pH values: 7.5, 8.5 and 10.0. pH was adjusted with
NaOH (N/44) solution. After the equilibration period, the final pH was also recorded. For
the 10 sets of experiments run at pH '" 7.5, the final measured pH varied from 7.24 to
7.42; for experiments run at pH '" 8.5, the final measured pH varied from 8.21 to 8.42;
and for experiments at pH '" I 0.0, the measured pH varied from 9.68 to 9.75.

Figure 4.14 shows for all three pH values, manganese removal (expressed as % removal)
decreased as initial manganese concentration increased. At pH '" 7.5, manganese removal
is relatively poor, varying from about 50% for an initial manganese concentration of I
mg/l to about 23% for an initial concentration of 10 mg/l. At pH '" 8.5, manganese
removal improved significantly, varying from about 83% for an initial manganese
concentration of I mg/I to about 73% for an initial concentration of 10 mg/l. At pH close
to 10, complete manganese removal was achieved (i.e., residual manganese below the
MDL of 0.001 mg/I).

Figure 4.15 shows chlorine dose added and residual chlorine remaining in solution at
three different pH values. It is clear that utilization of added chlorine becomes much
better as pH increases. At pH '" 7.5, measured Eh values varied from - 77 to - 54 mY; for
pH", 8.5, Eh varied from - 176 to - 70 mY; and for pH", 10, measured Eh varied from-
88 to - 70 mY. For these ranges of pH and Eh, the Eh-pH diagram of manganese (Fig.3.3
of Chapter 3) suggest that the precipitated solids would be either Mn(OHMs) (cream
colored) or Mn304(S) (reddish brown). The precipitated form would change from
Mn(OHMs) to Mn304(S) as Eh value increases. Observed color of precipitates is shown
in Fig. 4.16(a), (b), and (c). It indicates that the color of the precipitate was off-white,
when pH was relatively low and cream-brown to light reddish-brown, when pH was
relatively high.

89
 120

'c-"-
0
100

80
.--.-------.------ ..------.
m_
llil
_

~
u:
Q;
'----
60
'"«
"0
!!I
0
E 40
w
':;;"c
20
+- pH - 10.(
.........f!I-pH=8.5

0 -lJ.-pH= 7.5

0 2 4 6 8 10 12
Initial Mn Concentration (mg/L)

Figure 4.14: Removal of manganese by oxidation with bleaching powder, as a function

of initial manganese concentration at different pH values (Chlorine
stoichiometry ratio = 1.2)

1.8

1.6
....•... ••••
::;
rn 1.4 .,"',.'.
.s .--E
Q)
c
1.2 .. ••• •
...
'

.<= •
0 .....
_
....
III... ...
:<: ••• iii'" ...........- •.. '.'
u
rn
:J
0.8
••••e ••••••
-"

:5"
0.6
cr"'
.. -,."
Q)
iii'
0.4

0.2
./
I!i"
.. ,

...... ••
.-+ .. pH = 10.0
•• pH = 8.5
- •••. pH=7.5
0
0 2 4 6 8 10 12 14 16 18
Chlorine Dose Applied(mg/L)

Figure 4.15: Residual chlorine remained in solution after oxidation with chlorine of
groundwater containing different amount of initial manganese for amount of
chlorine dose applied (Initial Mn Concentration = 2.0 mg/L; Chlorine
stoichiometry ratio = 1.2)

90
Figure 4.16(a) Color of precipitation Figure 4.16(b) Color of precipitation
for water with initial manganese for water with initial manganese
concentration 1.0 mgIL (pH ""'7.4; Eh concentration 5.0 mglL (pH ""'8.4; Eh
"'" -70 mY) "'" -176 mY)

Figure 4.16(c) Color of precipitation

for water with initial manganese
concentration 10.0 mglL (pH ""'9.7; Eh
"'" -80 mY)

91
Figure 4.17 shows results of another set of experiments carried out to evaluate the effect
of pH on manganese removal by bleaching powder. This experiment was carried out with
an initial manganese concentration of about 2.0 mgll; chlorine dose was set at exactly that
required from stoichiometric consideration and pH was varied from 5.2 to 10. As shown
in Fig. 4.15, manganese removal was found to be strongly dependent on pH. Removal of
manganese varied from a very low of about 8.5% at pH 5.2 to about 100% at pH 10.

These results are not surprising as chlorine is a weak oxidant, and manganese removal by
chlorination is usually not very effective until pH is raised above 8.5; and for high level
of manganese it is often needed to raise pH above 9.5 (Benschoten et aI, 1990). Samblebe
(2003) also reported that for slow oxidizing agents like molecular oxygen or chlorine it is
necessary to raise the pH above 8.5 for effective oxidation reaction of manganese.
Besides higher pH also promotes rate of oxidation of manganese by oxygen. In treatment
systems, soda ash is often injected with the chlorine to raise pH to optimum levels.
Adjusting the pH to alkaline levels also reduces the corrosivity of the water to pipes and
plumbing (Seelig, 1992).

120

~ 100
;,'!.
~
c:
0
80
~
u:
~
<ll 60
'"
<{
'0
9!
0 40
E
&!
c:
::;; 20

0
5 6 7 8 9 10 11

pH

Figure 4.17: Removal of manganese by oxidation with chlorine at different pH values

(Initial Mn concentration =2.0 mglL; Chlorine dose added = 2.58 mglL)

a
92
Effect of Contact Time on Manganese Removal by Bleaching Powder
Since chlorine (bleaching powder) is a slow oxidizing agent, effect of contact time on
manganese removal was evaluated in this study. For this purpose, experiments were
carried out with initial manganese concentrations varying from about 1.0 to 10.0 mgll;
chlorine dose was set at 1.2 times that required from stoichiometric consideration and pH
was fixed at close to 10.0. The contact time of the oxidant (bleaching powder) was fixed
at 15 minutes, 25 minutes and 35 minutes. Figure 4.18 shows manganese removal as a
function of initial manganese concentration for the three different contact times. It shows
that for any particular initial manganese concentration, removal increases as contact time
increases. For example, for an initial manganese concentration of about 2.0 mgll, average
mangatiese removal for 15, 25 and 35 minute contact times are about 65, 89 and 99
percent, respectively. Thus, it appears that efficiency of manganese removal by bleaching
powder could be improved by increasing the contact time with the oxidant.

110

100

~
c
0 90
~
II

2 80
:;;0
"0
W
>
g 70 El
&' o
:!E o
60 [iJ 15 nins rrixing
o 25 nins nixing
. l!J. 35 nins rrixing
50
0 2 4 6 8 10 12
Initial fvtl Concentration (rrg/L)

Figure 4.18: Removal of manganese by oxidation with chlorine as a function of initial

manganese concentration for different contact times (Initial Mn cone. =
2.0 mglL; Initial pH = 10; chlorine stoichiometry ratio=1.2)

93
4.3.3Manganese Removal by Aeration

Figure 4.19 shows removal of manganese (present at an initial concentration of about 2.0
mg/I) by simple aeration as a function of pH for different period of mixing (aeration). It
shows that for a particular pH manganese removal increased as aeration period increased;
and for a particular aeration period removal increased as pH increased. However, these
results show that pH has a much more pronounced effect on manganese removal than
aeration time. For example for a 30 minute contact time, removal of manganese increased
from about 10% to about 93% as pH increased from 7.04 to about 10.

These results are not surprising because the rate of oxidation of Mn (II) has a second
order relationship with hydroxyl ion concentration, which indicate that an increase in one
pH unit would cause about 100 fold increase in rate of the oxidation reaction (Stumm and
Morgan, 1984). On the. other hand, manganese oxidation is not strongly dependent on the
concentration of dissolved oxygen. It has been observed that above about 30% saturation
value of dissolve oxygen, there is no significant dependence of the manganese oxidation
reaction on the concentration of dissolve oxygen. Many other researchers [e.g., Graveland
and Heertjes, 1975; Tebo and Emerson, 1985; Tebo and others, 1991] reported no
dependence of the rate of Mn(II) removal on DO above concentrations of about I mg/I
(i.e.about 12% air saturation, or 0.03 mM) and an approximate linear dependence at lower
DO values (Marble et aI, 1999).

A comparison of manganese removal (after filtration) by bleaching powder and simple

aeration at different pH values (pH 7 to 10) is shown in Figure 4.20. The figure shows
that at pH 7 removal of manganese by both methods was almost same. But at pH values
in between 7 to 9, manganese removal by bleaching powder oxidation was slightly higher
than that from aeration. At higher pH (at 10), removal was almost same. It should be
noted that for bleaching powder a contact time of 30 mins was maintained and in case of
aeration, removal (after filtration) for a contact time of 60 mins was plotted. Therefore, it
appears that oxidation by mixing with bleaching powder is slightly more effective within
the pH range of 7 to 9 compared to oxidation by aeration (by mixing). However, it should
be noted that, during bleaching powder oxidation, aeration also occurred during mixing.
Therefore, manganese removal by bleaching powder oxidation may also include effect of
oxidation by aeration during mixing.

94
 120

100

80
~
c
:;;
'0 60
~
0
E
& 40

20
-+- 20 rrin nixing
___ 30 rrin nixing
-,d- 60 rrin nixing
~- after filtration
0
6 7 8 9 10 11 12

pH

Figure 4.19: Removal of manganese by aeration for different initial manganese

concentration at different pH values ( Initial Mn = 2.0 mg/L)

120 ___ Aeration

-e- Bleaching powde

~ 100
~
c:
0
80
~
u::~
Q)
60
'"'-0:
-0
~
0
40
E
Q)
0::
c:
:2 20

0
6 7 8 9 10 11
pH

Figure 4.20: Comparison of removal of manganese (after filtration) by aeration and by

bleaching powder oxidation at different pH values ( Initial Mn = 2.0 mg/L;
for aeration contact time= 60 mins; for bleaching powder oxidation contact
time = 30 mins)

95
4.4 SUMMARY

Potassium permanganate is found to be very effective oxidizing manganese. Very good

removal of manganese for different initial manganese concentrations varying from 1.0 to
10.0 mg/L was achieved. For initial manganese concentration of up to about 2.0 mg/I,
almost 100% removal was achieved. Although oxidation by potassium permanganate
(followed by filtration) could remove significant manganese from groundwater, for higher
initial concentrations exceeding about 5.0 mg/I manganese, the residual manganese
achievable may not satisfy the Bangladesh drinking water standard of 0.1 0 mg/l.

Manganese oxidation with KMn04 increases rapidly above pH 7.5 (greater than 95%).
Complete removal can be achieved around pH 9. Since pH of natural groundwater in
Bangladesh usually falls around the neutral range, it appears, manganese from such
groundwater could be effectively removed by potassium permanganate without any pH
adjustment.

Optimum manganese removal (100%) has been found at a permanganate dose equal to
that required from stoichiometric consideration. However, a slight overdose of
permanganate (1.2 times of stoichiometry ratio) did not affect residual manganese
concentration significantly. A dose about 0.8 times of stoichiometry ratio removed about
95% manganese for an initial manganese concentration of2.0 mg/L.

Sand filter has been proved very effective (for both lO-cm and 20-cm depth) in removal
of the color developed in water due to addition of KMn04 Sand filtration has also been
found to be very effective in removing solid manganese oxides efficiently (>99%, for
initial Mn concentration of2.0 mg/L).

For an initial manganese concentration of 2.0 mg/L, complete removal was possible for a
settlement time of 3hrs. For initial Mn concentration varying from 5.0 to 10.0 mg/L,
removal by a 3-hrs settlement was greater than 95%.

Manganese oxidation using Bleaching powder has been found to be less effective than
permanganate in the natural.pH range of groundwater. At pH 7.5, removal by chlorine

96
oxidation varied from about 23% to 49% for initial Mn concentration of 10.0 to 1.0 mglL,
respectively. At pH 8.5, corresponding removal was 83% and 73%, respectively. At pH
10, complete removal occurred regardless of initial concentration.

Manganese oxidation by aeration is possible at high pH value. At pH 7, removal after

aeration, followed by filtration has been found to be only 15% for an initial Mn
concentration of 2.0 mglL. At pH 9, removal increased to 89% (for a contact time of 60
mins) Complete removal was achieved at pH 11, regardless of contact time.

97
 CHAPTERS

MANGANESE OXIDATION IN PRESENCE OF IRON

5.1 INTRODUCTION

Occurrence of iron in groundwater is one of the most wide spread and significant water
quality problems in Bangladesh. In most of the areas of Bangladesh significantly high
amount of iron is present in groundwater. As described in chapter two, more than 65%
tubewells surveyed during National Hydrochemical Survey (NBS), exceeded WHO
guideline value (0.3 mg/L) for iron (DPHEIBGS, 2001). Presence of iron in water is not a
health concern but cause the water to be unsightly, taste bad, sticky hair and stain
plumbing fixtures and laundry (Cameron and Bourgin, 1995). In literature reviews in
chapter two, it is described that manganese can cause similar problems even existing at
very low concentration (less than 0.02). From NHS it is found that only about 47% of
tubewells of Bangladesh have both iron and manganese concentration below the
acceptable limit from aesthetic view point (0.3 mg/L and 0.10 mg/L respectively).
Therefore, in many areas of Bangladesh it may be necessary to remove both of these
elements from groundwater to increase the effective use of the groundwater.

Considerable works have been carried out in Bangladesh over the last three decades for
removal of iron from groundwater. In recent years substantial amount of work has been
done on removal of arsenic and iron-arsenic. Consideration of possible presence and
removal of manganese has given less emphasis. Only a few studies (e.g., Tahura et aI,
2001; Ali et aI, 2001; BAMWSPIDFID/Water Aid, 2001) on iron and arsenic removal
have provided data on manganese removal by those treatment options of iron and arsenic
removal. Moreover in those studies, manganese concentration was below 1.0 mg/L. Only
two tubewells had concentration above 1.0 mg/L (BAMWSPIDFID/Water Aid, 2001) but
it was found that removal of manganese was poor for that tubewells. As both iron and
manganese are present at considerable level in many groundwater sources, it is necessary
to develop any treatment option for simultaneous removal of these two elements.

Processes in which oxidation is followed by filtration are found to effectively remove

soluble iron from water. Usually oxidation of iron is accomplished by simple aeration or
chlorination or potassium permanganate application (Shahid, 1998). These oxidizing
agents are also used for manganese oxidation. Iron is oxidized at relatively lower pH
value than manganese. Soluble ferrous iron is oxidized very rapidly to insoluble ferric at
pH range 7 to 11(Shahid, 1998). Whereas, manganese oxidation is more effective at
relatively higher pH range (above 7.5 in case of permanganate oxidation). So in natural
pH range, for a water source where both iron and manganese are present at elevated level,
a portion of oxidation dose provided for manganese removal may be consumed by
dissolved iron. These may lead to inefficient removal of both iron and manganese by
oxidation.

Again during oxidation processes insoluble iron floes are formed. Manganese solids
produced from oxidation gradually settle down with time. As evaluated in chapter four,
considerable amount of settlement time is required for removal of oxidized manganese
solids from solution. Presence of iron may influence settlement of solid manganese
particles.

In chapter four, it is found that during permanganate oxidation and as well as chlorine
oxidation of high manganese concentration, significant amount of color is produced.
Therefore it was necessary to provide father treatment option for this color removal.
Presence of soluble iron in water may have some effect on color removal produced during
oxidation.

In many iron treatment plants, iron is removed by simple aeration at elevated pH (7.5 to
8) followed by suitable filtration option (Shahid, 1998). From assessment of manganese
removal by simple aeration (chapter four), it is found that at pH below 10 manganese
removal by aeration is not very effective. However, if soluble iron is present in such
water, that can be precipitated easily at this pH value. This precipitation of insoluble iron
may enhance the removal of manganese by simple aeration.

In this study, oxidation of manganese present in groundwater has been evaluated in

presence of dissolved iron using potassium permanganate as oxidizing agent.
Permanganate oxidation has been done on both natural groundwater and artificially
prepared samples. Besides, effectiveness of simple aeration in manganese removal in
presence of iron has also been assessed. This chapter presents results of laboratory

99
,
investigations on manganese removal from water by permanganate oxidation and aeration
in presence of dissolved iron.

5.2 MATERIALS AND METHODS

In this study, laboratory batch experiments were carried out in a way similar to that
described in chapter four. Natural groundwater collected from BUET pumping station
was used in these experiments. In order to assess effect of alkalinity on manganese
removal, additional experiments were carried out in synthetic water samples prepared
with deionized water, having low alkalinity.

All batch experiments were carried out in l-L glass beakers using in 500-ml samples.
Initial manganese concentration of the water samples were varied by spiking with
manganese stock solution (500mg/L) prepared according to the description given in
chapter four.

Required concentration of iron in groundwater was added by spiking the samples in

beakers with iron stock solution (having 500 mg/L Fe), prepared by dissolving anhydrous
ferrous sulfate salt (FeS04. 7H20).

All batch studies were carried out according to the procedure described in chapter four.
Additional details of different experimental set up are briefly described in the next
section.

5.2.1 Manganese Oxidation with KMn04 in Presence of Dissolved Iron

This study was carried following the procedure described in article 4.2.1 for manganese
removal by oxidation with KMn04.

Effect of presence of dissolved iron on manganese removal from natural groundwater

(alkalinity = 240 mg/L as CaCO]) by oxidation with KMn04 was studied by varying the
initial iron concentration from about 1.0 to 10.0 mg/L. Initial manganese concentration
was fixed at 2.0 and 5.0 mg/L. Dose of KMn04 was set at 1.0 and1.2 times to that
required from stoichiometric relation for manganese oxidation (Eq. 4.1). Groundwater

100
 ...
was first spiked with manganese, and then iron stock solution was added to achieve a
range of iron concentrations (e.g., 1,3,5 and 10 mglL). After adding the oxidant, mixing
was done for 10 mins and samples were allowed to settle for 30 mins before filtration in
similar way described in chapter four.

Experiments were carried out to assess the effect dissolved iron on manganese oxidation
at low alkalinity and low pH. Instead of using natural groundwater, in this study,
deionized water was used. Alkalinity of the artificial water was varied from 89 to 102
mglL as CaC03 using a NaHC03 solution (20 giL). pH was varied between 5.98 to 6.1 by
adding concentrated HCI solution. The experiments were carried out with initial
manganese concentration of about 2mglL. Dissolved iron concentration was varied from
1.0 mglL to 10.0 mglL. Permanganate dose was set at exactly that required from
stoichiometric requirement for manganese oxidation (Eq. 4.1).

Effect of dose of potassium permanganate on simultaneous removal of iron and

manganese in natural groundwater was determined. Permanganate dose was calculated
from stoichiometric relation for manganese and as well as for iron. According to Eq. 5.1,
for complete oxidation of iron byKMn04, about 0.94 mg of KMn04 dose is required for
1 ml of dissolved iron.

3Fe 2+ + KMn04 + 7H20 = 3Fe(OH)3 «)s) + Mn02 s + K + + 5H + (5.1)

Therefore for simultaneous oxidation of 1.0 mglL iron and 1.0 mglL manganese in
groundwater, the total amount of KMn04 dose required is 2.86 mglL. For these
experiments, KMn04 dose was varied by different factors this requirement. The factors
were 0.25, 0.5, 0.75 and 1.0. Manganese concentration was fixed at 2.0 mglL and iron at
5.0 mglL. Therefore, permanganate dose was varied from 0.25 to 1.0 times of that
required to oxidize 2 mglL manganese and 5.0 mglL iron simultaneously according to
stoichiometry ofEq. 4.1 and Eq. 5.1. For example, stoichiometry fraction 0.25 means that
of 25% of KMn04 dose required for 2.0 mglL manganese and 5.0 mglL iron has been
used. Alkalinity of natural groundwater was found to be 238 mglL as CaC03

Dissolved iron is known to get oxidized in contact with air under suitable alkalinity and
pH condition. Hence, removal of manganese from water (having dissolved iron) that has

101
""
,
\
~~
" .'
'.f!
been aerated (for iron oxidation) was investigated. In this experiment pH of water
samples were raised to about 8.0 (In between 8 to 8.1). pH was increased using NaOH
solution (N/44). Water was spiked with an initial manganese concentration of about 2.0
mg/L Iron concentration was varied from 1.0 to 10. 0 mg/L. After spiking water with
iron, samples were stirred for 15 mins in order to oxidize iron in water by aeration before
adding manganese to water. After that manganese stock solution was added to water
samples to provide an initial concentration of 2.0 mg/L. Perrnanganate dose required to
oxidize manganese only was added to water. Dose was set to exactly that required from
stoichiometric consideration for manganese oxidation.

5.2.2 Manganese Removal by Aeration in Presence of Iron

Manganese oxidation by simple aeration in presence of dissolved iron was also evaluated
in this study. Experiments similar to that described in article 4.2.4 were carried out. Initial
manganese concentration of manganese was fixed to about 2.0 mg/L. Initial iron
concentration was varied from 1.0 to 10.0 mg/L. After spiking water with required dose
of manganese and iron, water samples were aerated by vigorous mixing at 100 rpm.
Mixing was carried out for about 30 mins. Samples were then allowed to settle for 30
mins and then filtered and analyzed for iron and manganese. Initial pH varied from 7.65
to 7.72.

5.2.3 Chemicals and Measurement of Parameters

As described in chapter four, all chemicals used in this study were of reagent grade. Iron
stock solution (containing about 500 mg of iron/L of solution) was prepared by dissolving
anhydrous ferrous sulfate (FeS04. 7HzO; MW = 278.0) in deionized water. The stock
solution was kept at a pH below 2.0 by acidifying with concentrated Hel solution.
Manganese and perrnanganate stock solutions were prepared according to the procedure
described in chapter four in article 4.2.6.

In this study both iron and manganese concentration in water was measured by flame
atomic absorption spectrophotometry, using an AAS (Shimadzu, 6800). A sample
standard curve for iron is shown in Fig. A2 in Appendix A. Measurement of other
parameters (e.g., pH) was carried out following procedures described in article 4.2.6.

102
5.3 RESULTS AND DISCUSSIONS
5.3.1 Removal of Manganese by KMn04 Oxidation in Presence ofIron

This section describes the results of batch experiments carried out to assess manganese
removal from groundwater when iron is also present in water. Oxidation was performed
either by using potassium permanganate or by aeration. Batch experiments were
conducted to assess: (i) the effect of iron present in varying concentration in natural
groundwater on removal of manganese by permanganate oxidation, (ii) Effect of iron on
manganese removal from synthetic water having low alkalinity and pH, (iii) removal of
both iron and manganese present in groundwater by oxidation with KMn04, and (iv)
removal of manganese by chemical oxidation from groundwater which has been aerated
at elevated pH for removal of iron.

Effect of Dissolved Iron Concentration on Manganese Removal

Figure 5.I(a) and 5.1(b) show the percentage of manganese and iron removal respectively
as a function of initial iron concentration, varying from 1.0 mg/L (0.0178 mM) to about
10.0 mg/L (0.178 mM). The figure is plotted for initial manganese concentrations of 2.0
mg/L (0.036 mM) and 5.0 mg/L (0.091 mM). The permanganate dose for these
experiments was fixed at 1.2 times that required from stoichiometric consideration for
manganese oxidation. pH of water samples for these experiments varied from 7.77 to
7.62.

The figure shows that for higher initial manganese concentration (5.0 mg/L) removal of
both iron and manganese was good. About 90% of manganese removal was achieved
even when initial iron concentration was as high as 10.0 mg/L, while removal was over
98% for an initial iron concentration of 1.0 mg/L. But for lower initial manganese
concentration (2.0 mg/L), percentage of manganese removal decreased significantly with
increasing iron concentration. For initial iron content of 10.0 mg/I, manganese removal
was about 78%. For both cases iron removal was very good. For initial Mn concentration
of 2 mg/L, removal of iron was more than 99% even when iron concentration was about
10.0 mg/L. This is clear from the figure that as iron is capable to be oxidized to its
undissolved form under the experimental conditions. Part of the oxidizing agent added for
manganese oxidation may have been utilized for oxidation of iron, resulting in lower
removalofMn.

103
•,
 105

~
~ 100
c
g~ij,
o
'"~ 95
;.<. -.._,,-,-,~:
u:: '.
~
~" 90 ~ -8- <>
""'"

~
E 85
&!
c
":fi 80

'"
Ol
~ 75
::;;
-0- Mn = 5 rrglL
70 [] Mn = 21T\JIL
o 2 4 6 8 10 12
InitialIronConcentration (mg/L)

Figure 5.1 (a): Percentage removal of manganese as a function of different initial iron
concentration present in groundwater (KMn04 stoichiometry ratio
= 1.2)

105 ,--------------- ~

100 III III 0 0

~ & &
c
.Q
~ 95
u::
Q;
""«
"0
90
9Z
0
E
Q)
n::
Q) 85
LL

I Mn = 5 IT\J/L
.•.

o Mn= 2 IT\JIL
80
0 2 4 6 8 10 12
InitialIronConcentration (mg/L)

Figure 5.1 (b): Percentage removal of iron as a function of initial iron concentration
present in groundwater (KMn04 stoichiometry ratio for Mn = 1.2)

104
Figure 5.2 shows the amount ofresidual manganese remaining in solution after filtration
for different initial iron concentration in water. The permanganate dose was 1.2 times of
that required from stoichiometric consideration. It shows that although the removal was
good when expressed as percentage (fig. 5.1), but the residual concentrations of
manganese were high, especially in presence of higher iron contents. For an initial
manganese concentration of 5.0 mg/L, residual manganese after filtration exceeded
Bangladesh drinking water standard value for manganese (0.10 mg/L) as well as WHO
Health-based guideline value (0.50 mg/L), for initial iron concentrations of 3 mg/L or
higher. And in case of manganese concentration 2mg/L, residual concentration was above
allowable limit for initial iron content of 5.0 mg/L or higher.

0.6

:::J 0.5 B
0,
S- o
c
.2 0.4 o
~
""OJ 0.3
'"'"c
:s
0
0.2 Bangladesh
0 drinking water
U>
standard
.£
c 0.1
::;;
B o M1 = 2 rrg/L
<> M1 = 5 rrg/L
0
0 2 4 6 8 10 12
InitialIronConcentration (mg/L)

Figure 5.2: Residual manganese in solution after fil1ration for different initial iron
concentration in water

If these results are compared with experiments described in chapter four article 4..3.1,
where manganese was oxidized with KMn04 in the absence of iron, residual
concentration was much less than allowable limit for initial manganese concentrations of
2 and 5mg/L. This indicates that the presence of iron, decreased manganese removal
efficiency significantly and increased residual concentration of manganese in solution.

105
 \
•
Figure 5.3 shows comparison of manganese removal (initial Mn concentration = 2.0
mg/L) in presence of iron (varied from 1.0 to 10.0 mg/L) for two different dose of
KMn04 (1.0 and 1.2 times that required for Mn oxidation). Quantitatively, for an initial
manganese concentration of 2 mg/L, amount of KMn04 dose added was 3.84 and 4.61
mg/L for stoichiometric fraction of 1.0 and 1.2 respectively. The figure shows that in case
of stoichiometric ratio 1.2, manganese removal was relatively good (as % removal) even
when iron was as high as 10.0 mg/L. But in case of stoichiometric ratio 1.0 for KMn04,
removal gradually decreased with increase in iron concentration. This probably indicates
that when less oxidant is available, iron consumes a part of the oxidant before manganese
could get oxidized.

120
-<)-
__
SR. ~ 1.2
S.R.~1.0
~
~ 100
c
0
:;:;
~ 80
'"u:~
Q)
60
~
"0
92 40
0
E
Q)
Il::
c 20
:2
0
0 2 468 10 12
InitialFe Concetration (mg/L)

Figure 5.3: Manganese removal as a function of initial iron concentra1ion by

oxidation with KMn04 at two differen1 stoichiometry ratio (S. R.) of
KMn04 (initial manganese concentration = 2.0 mg/L)

Results presented in Fig. 5.1 and 5.2 suggests that complete manganese oxidation in
presence of high iron content in water is not possible with an oxidant dose required just
for manganese oxidation. In pH range (7.5 to 7.8) at which the experiments were
conducted was more favorable for iron oxidation than for manganese.

106
Manganese Removal from Low Alkalinity-Low pH Water in Presence of Iron
Previous studies on iron removal (e.g. Shahid, 1998) suggest that at low pH (less than 7)
and at low alkalinity, iron removal efficiency is also reduced. Therefore an attempt was
made to assess manganese oxidation in presence of iron in water at low pH and low
alkalinity, where iron may not get oxidized before manganese.

Figure 5.4(a) and 5.4(b) show the comparison between manganese and iron removal
respectively under normal pH (7.6 to 7.8) and alkalinity (240 mglL as CaC03) conditions
with that at low pH ( 6 to 6.1) and low alkalinity «100 mg/L as CaC03) value. From Fig.
5.4 (a) and (b), it is clear that manganese removal follows the same trend under both
experimental conditions. That is in both cases, manganese removal decreased with
increase in initial iron concentrations. However the removal was a little less for low pH
and low alkaline condition than that in case of normal pH range. For example, at initial Fe
concentration of 3 mglL, % of manganese removal after filtration was about 83% and
80% at normal pH and at low pH condition, respectively. Figure 5.4 also indicates that
iron removal was also high even at low pH and low alkalinity, even for initial iron
concentration as high as 10.0 mglL.

120
-. -- normal pH
ID low pH
~ 100
e....
c
1--- --
.........
0
80
~
u::
:;; 60
'«"
"0
- --.
"'" mJ -'.
91
0 40
E ..- -.
'"
0::
c
:;;;; 20

0
0 2 3 4 5 6 7 8 9 10 11
Initial Iron Concentration (mg/L)

Figure 5.4 (a): Comparison of manganese removal with KMn04 as a function of

different initial iron concentration in low pH -low alkalinity water and in
normal pH - normal alkalinity condition.

107
 105
" norrral pH
~ I
a low pH
~

•
c 100 II.
0
;;J
co
~
==
i.L 8 o
~ 95 o
<ll
~
"0
<ll 90
>
0
E
<ll
0:: 85
<ll
LL

80
0 2 3 4 5 6 7 8 9 10 11
InitialIron Concentration (mg/L)
Figure 5.4(b): Comparison of% iron removal as a function of different initial iron
concentration in low pH -low alkalinity water and that in normal pH -
normal alkalinity condition.(Initial Mn concentration = 2.0 mg/L;
Stoichiometry ratio for KMn04 = 1.0)

This is probably due to the fact that permanganate being a very strong oxidizing agent
capable to oxidize both iron and manganese even at low pH (for pH range of 5 to 10;
Samblebe,2003).Though alkaline environment enhance oxidizing power ofpermanganate,
its effect was not significant for the experimental condition used in the study.

Figure 5.4 suggests that even at low pH and low alkaline environment, manganese
removal by KMn04 is poor in presence of iron. As KMn04 is a strong oxidant agent, it is
capable to oxidize iron at low pH and low alkaline condition.

Figure 5.5 shows the residual manganese remaining in solution after KMn04 oxidation in
presence of iron at low pH and low alkalinity condition. It shows that residual manganese
was always above the Bangladesh standard (0.10 mg/L) for J'!langanese. However for
initial iron concentration less or equal to 3.0 mg/L, concentration of residual manganese
was below WHO health-based guideline value (0.50 mg/L) for manganese. Fig. 5.5 shows
that although the removal of iron (expressed as "% removed") was high, for higher initial
iron concentration residual iron was greater than who guideline value for iron (0.30
mg/L).

108
 1.8

1.6

;;;r 1.4
E
c: 1.2
.Q

~
2 WHO
:< 0.8
guideline
~
ro~ 0.6 ,/
",2 0.4 Bangladesh
/ standard
0.2

o
o 2 468 10 12
Initial Fe Concentration (rrg/L)

Figure 5.5: Residual manganese remain in solution after oxidizing with KMn04 at low
pH low-alkaline water for presence of dissolved iron at different
concentrations. (Initial Mn concentration = 2.0 mglL; KMn04 stoichiometric
ratio = 1.0)

Removal of Both Iron and Manganese by Oxidation with KMn04 Dose

Results presented earlier shows that manganese removal in the presence of iron,
decreased with increase in iron concentration. This shows that, when both iron and
manganese are present in water, it is necessary to provide sufficient oxidant dose to
oxidize and remove both of these two elements. For oxidation KMn04 dose required may
be less than that indicated by stoichiometry because of possible oxidation of iron by air.
Therefore experiments were conducted to assess the amount of KMn04 dose required to
remove iron and manganese simultaneously.

Figure 5.6 shows % of manganese removal (after filtration) for different combined
stoichiometry fraction. The figure indicates that regardless of the amount of KMn04 dose
applied, complete removal of iron occurred after filtration. But for manganese, removal
increased with increase of dose fraction up to the fraction 0.75, then again it decreased.
Removal of manganese was greater than 93% in case of stoichiometric fraction of 0.75;
whereas it was 84% when the dose was exactly that indicated by stoichiometry of iron
and manganese oxidation.

109
Quantitatively, to oxidize 5.0 mg/L of Fe, about 4.7 mglL of KMn04 dose is needed
according to Eq. 5.1. For removal of both 2.0 mglL manganese and 5.0 mglL of iron,
about 8.54 mglL of KMn04 is required. Fig. 5.6 indicates that almost 100% Fe removal
was achieved after filtration irrespective of KMn04 dose. Therefore, if the oxidation of
iron is carried out by KMn04, then for combined KMn04 fraction of 0.25 (i.e., 2.14 mglL
KMn04) and 0.5 (i.e., 4.27 mglL KMn04), no KMn04 was available to oxidize
manganese. However, for these KMn04 fractions, removal of manganese about 8% and
60 %, respectively was achieved. For KMn04 fraction 0.75, (i.e., 6.41 mglL KMn04),
about 1.71 mg/L KMn04 in excess of that required for complete Fe oxidation was
available to oxidize Mn, which is capable to oxidize 44.5% Mn. Whereas at this dose,
removal of Mn was about 94%. Therefore, oxidation of Fe did not occurred by KMn04
alone, aeration (during mixing) and autocatalytic oxidation of manganese may play
significant role in removing both Mn and Fe.

120

100 -----~D.\------ID.~-----D.

:oR
~ 80
+----.
>-
"
c:
'"w'"' 60

~
0
E 40
&!
20

-+- % M1 rerroved
-A- % Fe rermved
0
0 0.25 0.5 0.75 1.25
KMn04 Dose Stochimetric Fraction

Figure 5.6: Simultaneous removal of manganese and iron for different KMn04 dose
expressed as a fraction of that required for complete oxidation of both iron
and manganese. (Initial Mn concentration = 2.0 mglL; initial iron
concentration = 5.0 mglL)

110
 2.00
-+- Mnafter 30 nins -6- Mnafter filter

:::J
en 1.50
S
c
o
:g
o
(/J 1.00
.":;;: WHO

<ii
/'Iue
guideline

"
:Q
(f)

&!
0.50
•
0.00
o 0.25 0.5 0.75 1.25
KMn04 stochiometric ratio

Figure 5.7: Manganese concentration remaining in solution after simultaneous removal

of manganese and iron at different KMn04 dose expressed as a fraction of
that required for complete oxidation of both iron and manganese. (Initial Mn
concentration = 2.0 mg/L; initial iron concentration = 5.0 mg/L)

Figure 5.7 shows residual manganese concentration for different dose ofKMn04 It shows
that although the removal of manganese was high in case of stoichiometric fraction 0.75
(as % removed), residual concentrations were well above Bangladesh drinking water
standard and WHO aesthetics based guideline, though lower than WHO health based
guideline.

Manganese Oxidation by KMn04 with Pre-Oxidation of Iron by Aeration

Previous results from this study indicate that when iron is present in groundwater, it is
difficult to remove manganese completely, even with addition of required amount of
oxidant dose for removal of both iron and manga~ese. On the other hand, almost
complete removal of manganese was achieved in the absence of iron, under similar
experimental conditions. Therefore, removal of iron by aeration prior to addition of
KMn04 may improve manganese removal in the presence of iron.

Experiments were carried out to evaluate the removal efficiency of manganese with
KMn04 oxidation with pre-oxidation of iron by aeration. For these experiments, initial
manganese concentration was fixed at 2.0 mg/L and initial pH was adjusted to about 8.0

III
(to enhance iron removal by aeration). Iron concentration was varied from 1.0 to 10.0
mg/L. KMn04 dose was set at exactly that required for complete oxidation of manganese
only.

Figure 5.8 shows removal of both manganese and iron from these experiments. It shows
that pre-oxidation removes about 30% to 47% of initial iron in water varied from 1.0 to
10.0 mg/L respectively. After addition of permanganate dose, complete removal of iron
was achieved regardless of initial concentration. The figure also indicates that with pre-
oxidation of iron by aeration, greatly increase the removal of manganese, and complete
removal of manganese was also achieved after oxidation with KMn04 This indicates that
prior aeration for removal of iron can enhance simultaneous removal of both manganese
and iron.

120

100
:oR
~
Q)
LL 80
"0
C

'"c
:;; 60
a
g! 40
a
EQ)
0::
20
I- ,Ii
Fe after filtration
6. Mn after Filtration
0
0 2 4 6 8 10 12

Initial Fe Concentration (mg/L)

Figure 5.8: Removal of manganese with KMn04 oxidation in presence of iron

with pre-oxidation of iron by aeration (KMn04 stoichiometry ratio for
Mn= 1.0).

112
5.3.2 Removal of Manganese and Iron from Water by Aeration

From previous experimental result it has been found that aeration enhances the removal
of iron from water. An attempt was made to evaluate the removal of both manganese and
iron from water by simple aeration.

100
I. ~
90 I m Fe
80
~
~ 70
c
0 iii
60
~
u:: 50
~
Q)
40
-<
""
30
~
E
Q)
0::
20
10
• • • •
0
0 2 4 6 8 10 12
InitialIron(mg/L)

Figure 5.9: Simultaneous removal of manganese and iron by simple aeration for different
initial iron concentration (Initial Mn concentration= 2.0 mg/L).

Figure 5.9 shows the % of removal of manganese and iron by simple aeration. The figure
shows that both iron and manganese removal decreased with increase of initial iron
concentration in water. However ifFig.S.8 is compared with the Fig. 4.19 in chapter four
showing manganese removal by aeration, it would indicate that in presence of iron
removal of was slightly less than manganese removed by aeration in absence of iron. This
implies that presence of iron may not enhance manganese removal by simple aeration.

113
 CHAPTER 6
MANGANESE REMOVAL BY COAGULATION-ADSORPTION-
COPRECIPITATION

6.1 INTRODUCTION

In many water treatment facilities chemical coagulation process is used to enhance the
removal of colloidal and dissolved substances from water. Especially in Bangladesh
chemical coagulation (e.g. iron coagulation, alum coagulation etc.) processes are
extensively used to remove arsenic as well iron from groundwater. When both iron and
arsenic is present in water, iron coagulation enhances removal of both of these
substances. In addition to iron and arsenic, manganese is another very common element
found in water in almost allover the world. Many researches suggest (Benschoten and
Lin, 1992; Raveendran et aI, 2001; Samblebe, 2003; Seeling, 1992; Sharma et aI, 2001
etc.) chemical oxidation, especially with KMn04, CIOz, as the most effective treatment
option for manganese removal. One of the common difficulties with chemical oxidation
is the removal of color, for which it is necessary to filter water through a suitable filter
media after chemical oxidation. Where as color produced from water treatment can
effectively reduced through flocculation and coagulation.

The subject of controlled coagulation for the removal of the objectionable manganese
dioxide resulting from manganese oxidation has received little attention (Passelt et aI,
1967). According to Morgan and Stumm (1981) oxygenation of manganese (II) show
various average degree of oxidation (MnOx). Variations in x often can be attribute to
sorption of manganese (II) by the oxide, thus increasing the apparent ratio Mn:O and
decreasing the magnitude of x. Despite the variable nature of x, the formula for
manganese dioxide is commonly given, as the name implies, as MnOz (Gabono et aI,
1965).

According to Passelt et al (1967), for the pH range of practical interest for water
treatment (5 to 11), colloidal MnOz is characterized by a net negative particle charge.
Hence, for normal conditions, colloidal hydrous Mn02 is should behave in a fashion
similar to many other negatively charged colloids (such as Si02), respond accordingly to
treatment with common coagulants such as ferric sulfate and alum.

An important characteristic of hydrous manganese dioxide in aqueous solution is its

ability to exchange surface-bound H+ and OR ions in response to changes in the relative
concentrations or activities of these ions in solution phase; i.e., in response to changes in
pH. If the negatively charged OR ion is present in excess at the hydrated surface, the
colloid will exhibit a net negative charge, while a net positive surface charge will be
exhibited in the presence of an excess ofH+ ions (Passelt et ai, 1967).

The hydrolyzing metal ions Fe (Ill) and Al (Ill) are the most common coagulants for
water treatment. The behavior of these metals with respect to coagulation of colloidal
hydrous Mn02 produced from dissolved manganese oxidation is thus direct interest.
According to Passelt (1967), both Fe(Ill) and Al(Ill) undergo relatively extensive
hydrolysis and polymerization in aqueous solution to yield a broad spectrum of charged
species of different "molecular" size and charge density. The nature of predominant
hydrolysis and polymerization species formed is strongly dependent upon pH, other
factor being constant. At low pH these species are positively charged, but increases in pH
lead eventually to formation of negatively charged metal-hydroxo polymeric forms.

Coagulant aids often promote settling, when the sedimentation characteristics of flocs
formed by normal coagulation procedures are poor. There are two categories of coagulant
aids. The first includes those for which a substantial increase in floc size occurs as a
result of sorption and enmeshment of particles by long-chain polymeric molecules. The
second includes clay-type minerals that affect an increase in the density of floc particles.
Mn02, because of its high specific gravity, might well act as a coagulant aid in water
treatment operations and thus be included in the second category (Black, 1985).
Therefore, removal of manganese by coagulation, insoluble manganese itself may act as
coagulant aid and thus may enhance further removal.

115
All the studies discussed above highlights the possibility of removal of oxidized
manganese (Mn02) by coagulation. Therefore, for this option it is necessary to oxidize
manganese first and then Mn solids would be removed by coagulation.

In context of Bangladesh, coagulation using FeCI] and alum is widely used to remove
arsemc from groundwater (e.g., Ali et aI, 2001; Tahura et aI, 2001;
BAPWSP/DFID/Water Aid, 2001). In these studies it has been found that some of the
methods also removed manganese, but in those cases manganese concentration was
below 1.0 mglL. Moreover in many methods of arsenic removal by coagulation utilized
some permanganate to oxidize As (III) to As(V) (e.g. Ali et aI, 2001). And coagulation
was followed by filtration to enhance surface adsorption of arsenic.

No reliable data has been found on effective removal of dissolved manganese in water by
coagulation-adsorption process alone. By addition of coagulant under favorable
environment it may be possible that dissolved manganese get adsorbed onto coagulant
flocs and thus get settled by gravity. Therefore, it is necessary to assess the possibility of
removing dissolved manganese by coagulation adsorption and coprecitation.

Iron coagulation is being widely used in removal of arsenic from groundwater of

Bangladesh. In this study, experiments have been carried out to assess manganese
removal by Fe(III) coagulation, adsorption and coprecipitation processes. Moreover,
study was also made to evaluate the possibility of manganese removal by utilizing
dissolved ferrous iron as adsorbent in coagulation process.

6.2 MATERIALS AND METHODS

In this study, effectiveness of manganese removal by coagulation with two common iron
salts [ferric chloride and ferrous sulfate) was evaluated. Moreover, effect of pH on
manganese removal. by Fe(III) coagulation was also assessed. As in previous studies
described in chapter four and five, groundwater collected from BUET pumping station
was used for this study. Alkalinity of groundwater was 242 mglL as CaCO]. Stock

116
solutions for manganese, potassium permanganate, and iron (II) were prepared following
the procedures in chapter four and five. Stock solution of Fe(III) salt (i.e., ferric chloride)
was prepared by dissolving anhydrous ferric chloride salt (FeCh.6HzO; MW = 270.30) in
deionized water. A stock solution having 500 mg Fe(III) per liter of solution was
prepared. The pH ofthe stock solution was kept below 2.0 by adding concentrated HCl in
order to keep ferric iron in dissolved form.

For a particular initial manganese concentration and coagulant dose, one set of
experiments was conducted without the addition of any oxidizing agent, while another set
was conducted with the addition of oxidizing agent (potassium permanganate). This was
be done for differentiating the mechanisms (oxidation versus adsorption) of manganese
removal from water. In, each set of experiment, water samples (500 ml) in I-L glass
beakers were coagulated in a digital paddle-type coagulation apparatus. The coagulant
dose was varied from 2 mg/I (as iron) to 25 mg/I (as iron).

After addition of potassium permanganate (if needed), coagulant dose was added to the
beaker from stock solution of either Fe(II) or Fe(III). Then the contents of the beakers
were mixed rapidly for I minute at a speed 100 rpm. Then a slow mixing was done for 15
minutes at a speed of 45 rpm. Samples were then allowed to settle for 30 minutes. Then
clear supernatant was colh;cted from the beaker from about I-em below the water surface
and analyzed for manganese, color, and iron concentrations. Collected samples were
then filtered through OA5-flm filter in order to determine total dissolved manganese
present in water. In order to determine the effect of pH on manganese removal by
coagulation with ferric solution as coagulant, pH was varied from 7 to II by adding
NaOH (N/44) solution. Figure 6.1 shows the experimental set up for manganese removal
by coagulation with the addition of KMn04 as oxidizing agent. Additional details of
each type of experimental set up are briefly described below.

117
Figure 6.1: Experimental Set-up for manganese removal by Coagulation

118

,•
6.2.1 Manganese removal by Coagulation-Adsorption-Coprecipitation

Experiments were conducted for initial manganese concentration of 2 mg/L. Ferrous

sulfate concentration was varied from 2 to 25 mg/L (as Fe). Two sets of experiments
were carried out. In one set KMn04 was added at a concentration (2.84 mg/I) exactly
equal to that required from stoichiometric consideration for manganese oxidation. After I
minute of rapid mixing (100 rpm), IS minutes of slow mixing (45 rpm) followed by 30
minutes of settlement, supernatant samples were collected. The pH values of the samples
were within the range of 7.5 to 7.7. Similar experiments were conducted with ferric
chloride added as coagulant. Coagulant dose in this case was varied from 10.0 mg/I to
25.0 mg/L (as Fe). The pH values for these experiments varied from 7.59 to 7.72.

6.2.2 Effect of pH on Coagulation of Manganese with Fe (III) Coagulation

In order to evaluate the influence of pH ort removal of manganese from groundwater by

coagulation with FeCi}, experiments similar to those described above were carried out
where pH of the samples was varied from 7 to II by adding NaOH (N/44). Experiments
were carried out without addition of the chemical oxidant (e.g., KMn04.) Manganese
concentration was fixed at about 2 mg/L. Coagulant dose (ferric chloride) was fixed to 15
mg/L (as Fe).

6.3 RESULTS AND DISCUSSIONS

6.3.1 Manganese removal by Iron Coagulation

In this study, batch experiments were carried out to evaluate the effectiveness of
coagulation-adsorption process in removing dissolved manganese from groundwater.
Experiments were carried to determine: (i) the effectiveness of Fe(IT) as coagulant in
removing manganese with or without pre-oxidation of manganese by KMn04 dose, (ii)
the effectiveness of Fe (III) as coagulant in removing manganese with or without pre-

119
oxidation of manganese by KMn04 dose, and (iii) effect of pH on removal of manganese
with iron (III) coagulation. Experimental results are briefly described below.

Manganese Removal by Iron (II) (Ferrous Sulfate) Coagulation

The primary mechanism involved in the removal of metal ions by coagulation with iron
(II) (ferrous sulfate) involves adsorption of dissolved metal ions onto the surface of iron
floes and subsequent precipitation of the floes. In case of manganese, in addition to
adsorption, some manganese could be removed by oxidation of dissolved Mn(II) to
insoluble manganese solids, either by oxygen (from air) or potassium permanganate (if
added).

Figure 6.2 -shows the % removal of manganese after coagulation. The figure gives a
comparison between removal efficiency with and without KMn04 oxidation. Fig. 6.2
shows that manganese removal by coagulation was relatively good (as % removed) when
manganese was oxidized with KMn04.

90

80

~ 70
c:
a
~ 60
'5
OJ

8'" 50
>.
.0
40
9ia iii
E 30 III
•
Q)
c::
c: 20 a
:2
10
lowith KM104
0 I_ w ithoU!KM1~
0 5 10 15 20 25 30
FeSO, Dose (mg/L as Fe)

Figure 6.2: Removal of manganese by iron coagulation-adsorption process by using dissolved

FeSO, as coagulant with and without addition of KMnO, (Initial Mn
concentration =2.0 mglL; KMnO, stoichiometry ratio =1; pH =7.6)

120
Poor removal (about 20% to 34%) was achieved in case of coagulation without the
chemical oxidation. Part of this removal may be due to oxidation of dissolved manganese
into insoluble manganese by oxygen (from air) and its subsequent precipitation.

In case of coagulation without chemical oxidation, % removal decreased with increase in

coagulant dose, though the reason for this apparent trend is not clear. In case of
coagulation with chemical oxidation of manganese, removal decreased with increase in
coagulant dose up to 10 mglL of iron (II), and then increased with increase in coagulant
dose.

For coagulant FeS04dose of 3, 5 and 10 mglL, manganese removal has been found to be
about 78%, 66% and 60%, respectively. In chapter 5 it has been found that for similar Mn
and Fe (II) contents, removal of Mn was 82%, 54% and 29%, respectively for dissolved
Fe (II) concentration of 3,5 and 10.0 mglL respectively (Fig. 5.3). This probably implies
that controlled mixing for the purpose of coagulation enhanced Mn removal in presence
ofFe(II).

From results obtained in this study, it appears that when the oxidizing agent is added,
oxidation of dissolved manganese to insoluble forms and their subsequent precipitation is
the primary mechanism for manganese removal. Very little dissolved manganese could
be removed in the coagulation experiments through adsorption of dissolved manganese
onto iron flocs ..

Figure 6.3 illustrates the amount of residual manganese remammg m solution after
coagulation. It shows that although the removal of manganese was good (when expressed
"as % removed"), in case of coagulation with chemical oxidation, the residual
concentrations were well above the allowable national standard (0.10 mgll) and the limit
set by WHO (aesthetics). For coagulant dose of 5.0 and 10.0 mglL, residual manganese
was greater than WHO health-based limit (0.5 mgll).

121
 •

2.0 ,---------------------------r~~~=="9
Q -- With KMn04

I --.- Without KMl04

:::J
d>
.s -.
.-'
.--_.,.
c:
o
1.5 ------'-
•__.............•............
-1-.' ."
~
"S
g>
8
~ 1.0
«
c;
::; ~. WHO
.0..... ....
~
"C
""""""'_ IUideline
.~ 0.5 '$'/':/' . >.,<.S':: .
0:
Bangladesh -
.... ~
standard \

0.0 +-----~----~----~----~----~---_1
o 5 10 15 20 25 30
FeS04 Dose (mglL as Fe)

Figure 6.3: Residual manganese concentration in solution after coagulation with FeS04, with
and without pre.oxidation of dissolved manganese using KMn04. (Initial Mn
concentration =2.0 mglL; KMn04 stoichiometry ratio =1; pH =7.6)

Manganese Removal by Iron (III) (Ferric Chloride) Coagulation

Iron (III) salts, such as ferric chloride, are widely used as coagulant for removal of
arsenic from groundwater. In this study, effectiveness offerric chloride coagulation in
removing dissolved manganese from groundwater was evaluated.

Figure 6.4 illustrates the removal of manganese by coagulation with ferric chloride, with
and without chemical oxidation of manganese. It shows that regardless of amount of
coagulant, removal of manganese in case of coagulation with chemical oxidation was
greater than 90%. In contrast, removal of manganese by coagulation without chemical
oxidation was much less, varying from about 25% (at a coagulant dose of 10 mg/I) to
about 37% (at a coagulant dose of 25 mg/I). Results presented in Fig. 6.4 also suggest
that chemical oxidation of manganese (and not adsorption of dissolved manganese) is the
primary mechanism of manganese removal in the coagulation experiments where KMn04
was used.

122
 120

100
~
--" ~-- ---<>
c
.Q
1ij 80
"S
Ol
ro
0
0 60
'"
.0

~
0
E 40
i£ ---~
c
:;;
20

-~ .. - With KMl04
I '" Wtl:hout KMl04
0
5 10 15 20 25 30
Coagulant Dose [FeC13l (mg/L)

Figure 6.4 : Removal of manganese by iron coagulation-adsorption process using Fe(III) as

coagulant, with and without pre-oxidation of dissolved manganese using KMn04'

(Initial Mn concentration = 2.0 mg/L; KMn04 stoichiometry ratio =1)

However, if these results are compared with results obtained from coagulation with
ferrous sulfate (Fig 6.5), it becomes clear that ferric chloride coagulation was more
effective in removing manganese from groundwater that ferrous sulfate. The difference
was more significant for coagulation with chemical oxidation by potassium
permanganate. This may be due to the fact that in case of coagulation by ferrous sulfate,
part of the oxidant added (KMn04) was used up for conversion of ferrous iron to ferric
iron; and this portion of the oxidant was no longer available for oxidation of dissolved
manganese. It may be noted that as described by Black (1985), solid oxide form of
manganese (Mn02), because of its high specific gravity, might well act as a coagulant aid
in water treatment operations.

123
 40
~
:><. 35
~
c
0 30
:;:;
co
:; 25
Ol
co
0
0 20
>-
.0
"0 15
!I
0
E 10
'"
0::
c 5 --.-FeS04
2 -IlI-Fe03
0
5 10 15 20 25 30
Coagulant Dose (mg/L)

Figure 6.5: Comparison of Fe(II!) and Fe(II) as coagulant in removing manganese from
water. (Initial Mn concentration ~ 2.0 mglL; KMn04 stoichiometry ratio ~1)

Effect of pH on Manganese Removal by Iron (III) Coagulation

Since adsorption of dissolved manganese on to metal hydroxides is likely to be favored at
higher pH values, experiments were carried out to assess if greater manganese removal
could be achieved by coagulation at higher pH values. Results from this study presented
earlier suggest that Fe(III) is better coagulant than Fe(II) for manganese removal. Effect
of pH on manganese removal by coagulation was therefore carried out with ferric
chloride.

Figure 6.6 shows the removal of manganese by Fe(III) coagulation as a function of pH. It
shows that removal of manganese was low (less than 40%) for pH less than 9. However,
at higher pH values manganese removal increased sharply and almost complete removal
was achieved at pH II.

124
 120

~ 100
c
o
'iii
"S
80
OJ

<3 60
E
"0

~ 40
E
&!
c
::;;
20 0-°---- "'--
0------'-'
o

o
6 7 8 9 10 11 . 12
pH

Figure 6.6: Effect of pH on removal of manganese by Fe (III) coagulation-adsorption

process using Fe(III) as coagulant without pre-oxidation of dissolved
manganese using KMn04.( Initial Mn concentration = 2.0 mg/L; KMnO,
stoichiometry ratio =1)

In previous experiment on manganese removal by simple aeration (Chapter Four), it has

been found that high manganese removal (greater than 90%) could be achieved by simple
aeration pH value 10 and 11 (regardless of contact time). Thus, the apparent higher
removal of manganese in coagulation experiments (shown in Fig. 6.5) appears to be the
result of oxidation of manganese by air. The coagulant (ferric chloride) does not appear
to be responsible for this high removal of manganese at higher pH values. These results
once again demonstrated that very little manganese removal could be achieved by
coagulation alone. Thus, very little manganese removal could be expected in traditional
arsenic removal systems based on coagulation.

125

. f';/
 CHAPTER 7
CONCLUSIONS AND RECOMMENDATIONS

7.1 CONCLUSIONS
From the experimental results obtained in this study, following conclusions can be drawn:

• Potassium permanganate is very effective oxidant for removal of manganese from

groundwater. Manganese removal by permanganate oxidation was maximum at
permanganate dose equal to that required from stoichiometric consideration. However
for very high initial manganese concentration, although the removal was significantly
high with slight over dose of permanganate, residual manganese concentration was well
above the allowable limit set by Bangladesh drinking water standard (0.1 mg/L)

• Potassium permanganate, being a very strong oxidizing agent, is capable to oxidize

manganese over a wide pH range of 5-10. However, removal rate above pH 7.0 was
significantly high and above pH 8.0, manganese removal with KMn04 oxidation
becomes independent of pH. This implies that natural groundwater containing
dissolved manganese can be effectively removal with permanganate without any
adjustment to natural pH.

• One drawback of permanganate oxidation is the formation of color due to the oxidation
of dissolved manganese. It has been found that even for the lowest concentration of
manganese used in this study (1.0 mg!L), all color concentrations were above the
Bangladesh drinking water standard, even after 3 hours of settling! detention time.
Thus, it is clear that settling! detention alone would not remove color from water treated
with potassium permanganate. Hence, appropriate filtration option would be required to
remove color.

• Significant color (produced from potassium permanganate oxidation of manganese)

removal would be possible using of sand filters having 10 to 20 cm depth. However,
sand filter with greater depth found to be more effective in removing color. Sand filter
is also capable to remove solid manganese formed by oxidatiorl. Although residual
concentrations after sand filtration were not be as low as that obtained by 0.4511m filter
 \

paper «0.001 mglL), all residual manganese concentrations (after sand filtration) were
well below the Bangladesh drinking water standard of 0.10 mg/I.

• Manganese removal by chlorination in natural pH range (7. to 8.5) would be very poor.
Manganese oxidation with chlorine would not be very effective until pH is raised above
8.5; and for complete oxidation of manganese, it is necessary to raise pH near 10.

• Manganese oxidation with chlorine is also dependent on contact time. Insufficient

contact time may lead to poor removal of manganese by chlorine oxidation even at an
elevated pH.

• Manganese removal by simple aeration is also possible. Similar to chlorine oxidation,

manganese removal by aeration requires sufficient contact time (often may need 30 to
60 minutes contact). pH has a very pronounced effect on manganese removal by
aeration. For effective removal of manganese by aeration, pH is needs to be raised
above 10.

• Within pH range of 7 to 9, manganese oxidation with bleaching powder is more

effective than aeration. However, at higher pH values (= 10), both of these options are
capable to remove manganese completely.

• Presence of dissolved iron in groundwater may significantly influence manganese

oxidation with potassium permanganate. In natural pH range a part of the oxidant added
for manganese oxidation could be consumed by iron and could result in ineffective
removal of manganese.

• When both iron and manganese is present in water, providing oxidation dose for
removal of both iron and manganese may lead to sufficient removal of manganese.
However dose requirement would be less than that required from stoichiometric
consideration for iron and manganese. This is due to the fact that some manganese as
well as iron would be removed from water by the aeration during mixing.

127
• When dissolved iron is present in water, removal of manganese by chemical oxidation
may become effective, if water is aerated for a suitable time before adding the oxidant.
By aeration a significant portion of iron would get oxidized and therefore would not
influence manganese oxidation and thus complete removal of both iron and manganese
could be possible.

• Ferric chloride coagulation is found to be more effective in removing manganese from

groundwater that ferrous sulfate. However, without pre-oxidation of manganese (e.g.,
using KMn04) coagulation would not be capable to remove dissolved manganese
significantly.

• Coagulation (with FeCI)) at higher ph values increased manganese removal. But at

higher pH values (> 10) manganese removal is dominant by oxidation with air rather
than by adsorption onto iron floes. These results demonstrated that very little
manganese removal could be achieved by coagulation alone.

7.2 RECOMMENDATIONS FOR FUTURE STUDIES

• To conduct study on manganese removal using groundwater having naturally occurring

high manganese concentrations

• To investigate removal options for reducing residual manganese levels below the
Bangladesh standard (0.1 mglL) for higher initial manganese concentration.
• To investigate effect of parameters such as organic matter on manganese removal
efficiency.
• To investigate the effectiveness of sand filtration alone or coagulation followed by
conventional sand filtration as used in iron and arsenic removal plants in removing
manganese from groundwater.

128
 REFERENCES

Ali, M. A, Badruzzaman, AB.M., Jalil,M.A, Hossain, M. D., Hussainuzzaman, M. M.,

Badruzzaman, M., Mohammad, OJ. and Alder, N., (2001), "Development of Low-cost
Technologies for Arsenic Removal from Groundwater", BUET-UNU International
Workshop on Technologies for arsenic Removal from Drinking Water, May 2001

ANSVNSF,(2000), "American National Standards Institute, NSF International. Drinking

water treatment chemicals-Health effects", ANSVNSF 60-2000. Ann Arbor, MI

ASTDR, (2000), "Toxicological Profile for Manganese", US Department of Health ond

Human Services, Public Health Service, Agency for Toxic Substances and Desiese
Registry, Atlanta, GA.

BAMWSP/ DFID/ Water Aid, (2001), "Rapid Assessment of Household Level Arsenic
Removal Technologies", Phase II report, March 2001.

Banta,R.G. and W.R.mARKESBERY,(1977), "Elevated manganese levels associated

with dementia and extrapyramidal signs". Neurology 27:213-216.

Barceloux, D. G., (1999), "Manganese Isuues", Clin. Toxicol. 37:293-307

Benschoten, J. B., and Lin, W., (1992), "Kinetic Modeling of Manganese (II) Oxidation
by Chlorine Dioxide and Potassium Permanganate", Environmental Science and
Technology, Vol. 26 NO.7. Journal of the American Water Works Association 83[6], 80-
87. 1992.

BGS/WaterAid (2001), "Groundwater Quality: Bangladesh", Department of Public

Health Engineering, British Geological Survey, January 2001.

Black, A. P., (1985), "Determination of Mobility of Colloidal Particles", Jour. AWW A

54:926, August

129
Boman, B. J., Wilson, P. C, and Ontermaa E. A, (1999), "Understanding Water Quality
Parameters for Citrus Irrigation and Drainage Systems", Env. Sci. & Tech., 18(5),781

Cameron, LR (1995) "Biological Iron and Manganese Removal: An Untapped Potential",

Water 23,2 pp 25-28, AWWA Regional Conference, Coolangatta.

Cameron, L R, and Bourgin, F. (1995), "New Frontier - Biological Iron and Manganese
Removal from Drinking Water", AWW A Queensland Regional Conference.

Canavan, M,M., Cobb, S. and CSrinker, (1934), "Chronic manganese poisoning". Arch.
Neurol. Psych. 32:501-12

Cao, R. Y., Gu G. W., Huang, Z. M. and Ye H., (2001), "The Generation, Oxidation and
Disinfection of Chlorine Dioxide", Tongji Quality Supervision and Examination Center
of Environment Protective product NEPA

Casale, R J., LeChevallier, M.W., and Pontius, F.W., (2001), "Review of Manganese
Control and Related Manganese Issues", Denver: American Water Works Association
(AWW A) Research Foundation and AWW A

Collipp, P. J., S. Y. Chen, and S.Maitinsky, (1983), "Manganese in infant formulas and
learning disability". Annals Nutr. Metabol. 27:488-494.

COMA (1998).Nutrition and Bone Health:with particular reference to calcium and

vitamin D.Report of the Subgroup on Bone Health,Working Group on the Nutritional
Status ofthe Population,Committee on Medical Aspects of Food and Nutrition Policy.The
Stationery Office,London

Cook,D.G.,S.Fahn, and K.A.Brait,(1974),Chronic manganese intoxication". Archives of

Neurology 30:59-64.

Davidson, L.,A., Cederblad, B. Lonnerdal., and B. Sandstorm., (1989), "Manganese

absorption from human milk, cow's milk, and infant formulas in humans". Amer.J.Dis.
Children 43(7):823-827

.....•,
130
••
 Devenyi, A. G., T. F. Barron, and Mamourian, A. c., (1994), "Dystonia, hyperintense
basal ganglia, and high whole blood manganese levels in Algille's syndrome".
Gastroenterol. 106(4): 1068-1071.

Douglas, B., McDaniel, D., and Alexander, J., (1994), "Concepts and Models of
Inorganic Chemistry", 3rd ed.; Wiley & Sons:New York

DPHE/BGS (2001), "Arsenic Contamination of Groundwater in Bangladesh",

Department of Public Health Engineering, DFID, British Geological Survey, 2001.

Ehrlich, P.R. Dobkin, D.S., (1996), "Manganese clinical Nutrition III Health and
Disease", (Handbook of Nutrtionally Essential Mineral elments.

Elias, Z.,J.M.Mur, F.Pierre, et al,(I989), "Chromosome aberrations in peripheral blood

lynphocytes of welders and characterisation of their exposure by biological samples
analysis". J.Occup. Med.31 :477-483.

Evans, F.L., (1972), "Ozone in water and wastewater treatment", Ann Arber Science
Publishers, London.

Fechter, L.D, (1999), Distribution of manganese". Neuro Toxicol. 20:197-201.

Ficek, K. J. (1985), "Manganese Removal Using Potassium Permanganate in Low pH,

Low Hardness Waters", AWWA Annual Conference, Washington D.C.

Gabono, J. P., Etienne, P., and Lavrent, J. F., (1965), "Study of the Surface Properties of
Manganese Dioxide", Electrochim. Acta, 10:947.

Gage, B., O'Dowd, Dr. D. H., Williams P., (2001), "Biological Iron and Manganese
Removal: Pilot and Full Scale Applications", Ontario Water Works Association
conference, May 3rd, 2001

131
Graveland, A., Heertjes, P.M., (1975), "Removal of manganese from ground water by
heterogeneous autocatalytic oxidation", Transactions of the Institution of Chemical
Engineers, 53,154-164.

Gregory, D., Carlson, K. (1996), "Oxidation of Dissolved Manganese in Natural Waters,

1996 Annual Conference Proceedings"; American Water Works Association

Hauser, R.A., T.A.Zesiewicz, C.Martinez, A.S.Rosemurgy, and C.W.Olanow,

(1996),Blood manganese correlates with brain magnetic resonance imaging changes in
patients with liver disease". Can.I. Neurol.Sci. 23:95-98.

HSDB,(2001), "Manganese compounds".Hazardous Sabstances Data Bank. Last

revised:May 23, 2001.

Hurley,L.S., C.L.Keen and D.L. Baly,(1984), "Manganese deficiency and toxicity: effects
on carbohydrate metabolism in the rat". Neuro Toxicol. 5:97-104.

IOM,(2002), Institute of Medicine. Directary Reference intakes: Vitamin A, Vitamin K,

Arsenic, Boron, Chromium, Copper, Iodine, Iron, Manganese, Molybdenum, Nikel,
Silicon, Vanadium and Zinc". Food and Nutrition Board, Institute of Medicine.
Washington DC: National Academy Press.pp.10-1-1O-22.

IPCS,(1999), "Manganese and its Compounds. Concise International Chemical

Assessment Document 12, International Programme on Chemical Safety, World Health
Organisation.

limbo, Y. and Goto, K., (2001), "Iron and manganese removal by a membrane filtration
system ", IW A Publishing, Water Supply Vol 1 No 5-6 pp 357-364

Kassim, Dr. M. A.(1994), "Biological Iron and Manganese Removal", 20th WEDC
Conference Colombo, Srilanka, p-281-283.

132
Kwamura, c.L., H. Ikuta, S. Fukuzimi, R. Yamada, S, Tsubaki, T. Kodama, and S.
kurata,(1941), "Intoxication by manganese in well water". Kitasato Arch. Exp. Med.
18:145-169.

Keen, C.L., J.G. Bell and B.Lonnerdal,(1986), " The effect of age on manganese uptake
and retention from milk and infant formulas in rats". J.Nutr. 116:395-402.
Kondakis,X.G., N.Makris, M. Prinou and T.Papapetropoulos,(1989), "Possible Health
effects of high manganese concentration in drinking water". Arch. Environ. Health
44:175-178.

Lemley, A., Schwartz, J. J., and Wagenet, L. (1999), "Iron and Manganese in Household
Drinking Water", Fact Sheet 6, Cornell Cooperative Extension, New York State College
of Human Ecology.

Lynam, D.R., J.W. Roos, G.D. Pfeifer, B.F.Fort and T.G. Pullin,(1999), "Environmental
effects and exposures to manganesefrom use metyllcyclopentadienyl manganese
,
tricarbonyl(MMT) in gasoline". Neuro Toxico!. 20: 145-150".

Marble, J. C., Cotley, T. L., Conklin, M. H., and Fuller, C. C., (1999), "Environmental
Factors Affecting Oxidation of Manganese in Pinal Creek, Arizona", U.S. Geological
Survey Toxic Substances Hydrology Program--Proceedings of the Technical Meeting
Charleston South Carolina March 8-12,1999--Volume I of 3--Contamination From Hard-
Rock Mining, Water-Resources Investigation Report 99-4018A

Martin, S. T., (2003), "Precipitation and Dissolution of Iron and Manganese Oxides",
Washington, D.C.: American Chemical Society, 1986, pp 487-502.

Masschelein, W.J., (1979), "Chlorine Dioxide: Chemistry and Environmental Impact of

Oxychlorine Compounds", Ann Arbor Sci. Pub!., Ann Arbor, MI.

Minear, R. A.; Keith, and Lawrence H., (1982), " Water analysis", Academic Press,
Otlando

133
Montgomery, J. M., (1985), "Water Treatment Principles and Design", Wiley-Intuscical
Publication, New York.

Moss, AJ., A.S.Levy, LKim, YK.Park,OI989), " Use of vitamin and mineral
supplements in the United States: current uses, types of products, and nutrients. Advance
Data, Vital and Health Statistics of National Center for Health Statistics. Hyattsville, MD:
National Center for Health Statistics. No174. Hyattsville, MD:National Center for Health
Statistics(as cited in 10M, 2002)

NGO FORUM (1997), "Annual Report on Drinking Water and Sanitation"

NTP,(1993), " Toxicology and carcinogenesis studies of manganese(II)Sulfate

Monohydrate (CAS No. 10034-96-5)in F344/N Rats and B6C3F] Mice (Feed Studies).
NTP Tech. Rep.Ser.428. National Toxicology Program, Research Triangle Park, NC.

National Rural Water Association (NRWA), (2004) Regulatory Policy Development

Proj ect: Manganese.

O'Connell, R.T., (1978), "Suspended Solids Removal." Water Treatment Plant Design,
Ann Arbor Science Publishers, Inc, Ann Arbor, ML

Posselt, H.S., F. J. Anderson, and WJ. Webber., (1967), "The Surface Chemistry of
Hydrous Manganese Dioxide," Presented at meeting of Water, Air, and Waste Chemistry
Division, American Chemical Society, Bar Harbor, FL, April.

Raveebdran, R., Asgworth, B., Chatelier, B., and (2001) "Manganese Removal in
th
Drinking Water Systems", 64 Annual Water Industries Engineers and Operators
Conference, Bendigo, Bendigo, 5-6 September 2001.

Rayner-Canham, G.(1996), "Descriptive Inorganic Chemistry", Freeman:New York,

USA

134
 Roles, H.A, M.l. Ortega Eslava, E. Ceulemans, ARobert, and D.Lison,(1999), "
Prospective study on the reversibility of neurobehavioral effects in workers exposed to
manganese dioxide"" NeuroToxicol. 20:255-271.

Royal Academy of Engineering International Travel Grant Report, (2003), 8 th

International Congress on Mine Water and the Environment,19 th -22 nd October 2003.

Hasan, S., "Manganese removal from Groundwater of Bangladesh", Unpublished

undergraduate thesis work.

Samblebe, M. (2003), "Manganese Removal Using Chlorine Oxidation and Powdered

Activated Carbon", 66 th Arumal Water Industry Engineers and Operators' Conference,
Eastbank Centre - Shepparton, 3- 4 September 2003, p 86-95

Seelig, B., Derickson, R., and Bergsrud, F., (1992) "Iron and Manganese Removal:
Treatment Systems for Household Water Supplies", NDSU Extension Service, North
Dakota State University, USA

Singer, P.c. (1991), "Research Needs for Alternative Oxidants and Disinfectants",
Annual AWW A Conference, Philadelphia, June 23-27.

Shacklette, H.T., and J.G. Boerngen, (1984), " Element concentrations in soils and other
surficial materials of the coterminous United States". USGS Paper 1270. Washington,
DC: US Govern. Printing Office,p.6.

Shahid, S., (1998)., "Removal of Iron from Groundwater by Coagulation and

Sedimentation", M.sc. Thesis, Department of Civil Engineering, Bangladesh University
of Engineering & Technology, Dhaka, Bangladesh.

Sharma, S.K., Petrusevski, B., J.C. Schippers, (2001), "Advance Groundwater Treatment
Iron, Manganese, Fluoride and Boron removal" published by IDE-Delft, International
Institute for Infrastructure, Hydraulic and Environmental Engineering, Netherlands.

135
 Standard Methods (2002), "Standard Methods for the Examination of Water and
Wastewater", American Public Health Association, AWW A, and Water Pollution Control
Fed., Washington, D.C.

Stevens, AA, (1982), "Reaction of Chlorine Dioxide", Environ. Health Perspec!. 46; 101

Stumm, W. and Morgan, J.J., (1981), "Aquatic Chemistry", Willey & Sons Inc. New
York.

Tahura, S., Shahidullah,S. M., Rahman, T. and Hasnat, A, (2001), Evaluation of an

Arsenic Removal Household Device: Bucket Treatment Unit", BUET-UNU International
Workshop on Technologies for arsenic Removal from Drinking Water, May 2001.

Tebo, B.M., Emerson, S., (1985), "Effect of oxygen tension, Mn(II) concentration, and
temperature on the microbially catalyzed Mn(II) oxidation rate in a marine fjord",
Applied and Environmental Microbiology, 50(5), 1268-1273.

Tebo, B.M., Ghiorse, W.C., Waasbergen V., L.G., Siering, P.L., Caspi, R., (1997),
"Bacterially-mediated mineral formation: Insights. into manganese (II) oxidation from
molecular genetic and biochemical studies", Reviews in Mineralogy,. 35,225-266.

Ter, H, G.M.Griffing, M.Brandt, D.Oberding and M.Kapron, (1975),

"Methylcyclopentadienyl manganese trycarbonyl as an antiknock: Composition and fate
of manganese exhaust products. J Air Poilu!. Control assoc. 25:585-860(as cited in
ATSDR,2000).

Thomson,A.B., D.Olatunbosun, L.S.Valverg,(l971), "Interrelation of intestinal transport

system for manganese and iron ". J.Lab. Clin. Med. 78:642-655.

US Department of Health and Human Services, Agency for Toxic Substances and
Disease Registry (1997), ".Draft toxicological profile for manganese update".

136
US EPA, (1984) , "Health Assessment Document for manganese. Environmental Critaria
and Assessment Office, U.S. Environmental Protection Agency, Cincinati, OH (EPA-
600//83113F)

USEP A. (1986a), "Reference Values for Risk Assessment", Environmental Criteria and
Assessment Office. Cincinati, OH.

USEP A (1986b), "Guidance for Carcinogenic risk Assessment", US Environmental

Protection Agency, Fed. reg. 51(185):33992-34003.

USEPA (1990), "Comments on the Use of Methylcyclopentadienyl Manganese

Tricarbonyl in Unleaded Gasoline", U.S. Environmental Protection Agency, Office of
Research and Department, Research Triangle Park, NC.

USEPA (1996a), "Exposure factors handbook", Vol-I-General factors, U.S.

Environmental Protection Agency, Washington, DC.

USEP A (1996b), "Proposed Guidelines Carcinogenic Risk Assessment", Environmental

Protection Agency, Office of Research and Department, Washington, DC.

USEPA, (2001), "Contaminant Candidate List Preliminary Regulatory Determination

Support Document for Manganese", EPA #815- R-01-013. (November) Washington DC

USEP A (2002), "Announcement of Preliminary Regulatory Determinations for Priority

Contaminants on the Drinking Water Contaminant Candidate List", Federal Register
67: 106:38222-38244.

USEP A, (2002), "Health Effects Support Document for Manganese" (External Review
Draft --April) EPA #R -02-029. Washington DC

US EPA, (2004), "Drinking Water Health Advisory for Manganese" U.S. Environmental
Protection Agency, Washington, DC.

~ "1""
"
137
Varner, D., Skipton, S., Hay, D., and Jasa P. (1994), "Drinking Water: Iron and
Manganese", NHDES Technical Bulletin, WD-WS-3-7.

Vieregge,P .,Heinzow,B.Korf,G., Teichert,H.F .,Schleifenbaum,P. ,Mosinger,H. U.( 1995),.

"Long term exposure to manganese in rural well water has no neurological
effects",Canadian Journal ofNeurological Sciences 22 ,286-289

Wang, J.D., C.C.Chuang, Y.H.Hwang, J.M.Lin and J.S.Chen,(1989), "Mnganese induced

parkinsonism: An outbreak due to unrepaired ventilation control system in a
ferromanganese smelter". Br.J.Ind. Med.46:856-859.

Woolf, A.,R.Wright, C.Amarasiriwardena. and D.bellinger,(2002), "Achild with chronic

manganese exposure from drinking water. Environ. Health Persp. 110:613-616.

Zayed, J., C.Thibault, L. Gareau, et ai, (1999), "Airborne manganese particulates and
methylcyclopentadienyl manganese tricarbonyl (MMT) at selected outdoor sites in
Montreal". Neuro Toxico!. 20(2-3):151-158

138
APPENDIX- "A"
 A
b 0.450 -----i-----~------l------r-----i-----~------,------
I Itt I I I

s •
I
I
I
I
I
I
I
I
I
I
I
I
I ,
0.400 • t
-----T-----'------r-----r-----T-----'
I I Itt ,
l_ ----,..---
•
I I
I I
I
I
I
r
I
I
I
, ,
0.350 -----A--
•
~
t I
I
I
L ~
I
~_
, , ,
- __ .1. L _
I
I
I
r I
r
I
I I
I
,, ,,
-----l-----J
I I I
L
I
L l.
I
, ,, - ,,
0.300 I t I I I
--~------~-----r---
,, ,, ,,
I I I I
I I I I
I I I I
0.250 -----T-----'------,------r-- I
--T-----'-----_, I I I
,- _
, ,, , ,
,, ,,
I I t I
I I I I

0.200 --L-----~I I
~ I
~
I
L _
I

, ,
,,
-----~-----~---- __ I__ -

,,, ,,
I I r I I
I I I I I
, r I I 1 I
0.150 -----1-----,---- r I I I I I
,
, , -,------r-----i-----,------,------r---
I I I I I t
, I I I I I I

0.100 -----~----
,,
I

~------~-----~-----~-----~------I------~---
t I
I

I
I

I
I

I
I

I
I

, I
I
I
I
I
I
I
I
I
I
I
I
I
I
0.050 - - - - - J 1 L ..1 J , L _
• 1 1 1 r I I
I 1 r I I I I
I I I I I 1 I

0.000
0.000 0.500 1.000 1.500. 2.000 2.500 3.000 3.500,:4.'000. '.
::~,"'.-
Cone (ppm] .'.. .. "

".,

Abs = 0.1116 Conc. + 0

Regression, r = 0.9993

Figure AI: A sample standard curve for manganese standards plotted in AAS (Shimadzu,
6800) for determination of manganese in groundwater

140
'"
 A I

b I
I
s I
I I I I
0.125 ---------,---------r---------,---------r-----
t I I I I
I I I I I
I I I I I
I I I I
I I I I
I I I I I
0.100 ---------I---------,---------r------
I I I
--,---------,---
I I
I I I I I
I I I I I
I I I I I
I I I I
I I I I I
0.075 ---------r---------r------ -r---------r---------r---
I I I I I
I I I I I
I I I I I
I I I I I
I I I I I
I I I I
0.050 ---------,-------
I
-r---------r---------r---------r---
I I I I
I I I I I
I I I I I
I I I I I
I I I I I
t I I I.
0.025 ,---------r---------r---------r---------r---
I I I I ",
I I I r I
I I I I I
I I I r I
I I I I I

0000
0.000 lOOO 2.000 3.000 4000. ,:
.5.000
'~- :\-..~;.'
Cone [ppm) • ','
.; . -
,\
<t: •• "', ~

Abs = 0..0.271Cone. + 0.
Regression, r = 0..9998

Figure A2: A sample standard eurve for iron standards plotted in AAS (Shimadzu, 680.0.)
for detemlination of manganese in groundwater

141
 APPENDIX- "R"
Study on Manganese Removal from Groundwater by Oxidation
Table Bl: Water quality parameters for groundwater collected from BUET pumping
station.

Bangladesh
Concentration Standard for
Sl. no. Water Quality Parameters Unit
Present Drinking Water
(ECR'97)

I pH -- 7.78 6.5-8.5
PI. Co.
2 Color 2.00 15
unit
3 Turbidity NTU 0.35 10

4 Carbon-dioxide, CO2 mg/L 41.0 -.

5 Total Alkalinity as CaCO, mg/L 238.0 -
6 Total Hardness as CaCO, mg/L 236.0 200-500

7 Iron, Fe mg/L 0.02 0.3-1.0

8 Manganese, Mn mg/L 0.004 0.1

9 Arsenic, As giL <1.0 50

10 Chloride, Cl mg/L 130.0 150-600

11 Fluoride, F mg/L 0.29 1

Nitrate-Nitrogen, NO,-N as
12 mg/L 0.2 10 as N
N
Total Dissolved Solids
13 mg/L 504.0 1000
(TDS)

143
Table B-2: Effect of settling time on removal of manganese present in different
concentration in groundwater with KMn04 oxidation (Stoichiometry ratio = 1.2)

Unit lA 1B 2A 2B 3A 3B
Initial Mn 2.02 4.95 4.98
mg/L 1.03 1.01 2.03
Concentration
KMn04Dose 4.608 11.52 11.52
mg/L 2.304 2.304 4.608
Added
Residual Mn After 1.02 1.08
mg/L 0.15 0.17 0.27 0.3
30 mins Settlement
Residual Mn After 0.41
mg/L 0.05 0.04 0.09 0.11 0.35
90 mins Settlement
Residual Mn After
3 hrs settlement mg/L 0 0 0 0 0.11 0.13
(mg/I)
Residual Mn After 0.06 0.08
mg/L 0 0 0 0
Filtration
Manganese
Removal After 30 % 85.4 83.2 86.7 85.1 79.4 78.3
Mins Settlement
Manganese
Removal After 90 % 95.1 96.0 95.6 94.6 92.9 91.8
Mins Settlement
Manganese
Removal After 3 hrs % 100.0 100.0 100.0 100.0 97.8 97.4
Settlement
Manganese
Removal After % 100.0 100.0 100.0 100.0 98.8 98.4
Filtration
pH After Removal 7.88 7.9 7.84 7.86 7.74 7.7

Eh After Removal mY 230 228 236 248 338 325

Color after 30 mins Pt-Co

78 82 145 150 197 208
settlement unit

Color after 90 mins Pt-Co

32 40 52 55 121 130
settlement unit

Color after 3 hrs Pt-Co

12 18 24 22 71 70
settlement unit

Color after Pt-Co

0 0 0 0 4 0
Filtration unit

144
Table B-2 (continued) : Effect of settling time on removal of manganese present in
different concentration in groundwater with KMn04 oxidation (Stoichiometry ratio = 1.2)

Unit 4A 4B 5A 5B
Initial Mn
mg/L 7.48 7.52 9.98 10.02
Concentration
KMn04Dose ,
mg/L 16.128 16.128 23.04 23.04
Added
Residual Mn After .
mg/L 1.86 1.79 2.54 2.66
30 mins Settlement
Mn Concentration
After 90 mitis mg/L 0.7 0.77 1.14 1.18
Settlement
Residual Mn After
3 hrs settlement mg/L 0.24 0.23 0.35 0.38
(mg/I)
Residual Mn After
mg/L 0.16 0.15 0.22 0.21
Filtration
Manganese
Removal After 30 % 75.1 76.2 74.5 73.5
mins Settlement
Manganese
Removal After 90 % 90.6 89.8 88.6 88.2
mins Settlement
Manganese
Removal After 3 hrs % 96.8 96.9 96.5 96.2
Settlement
Manganese
Removal After % 97.9 98.0 97.8 97.9
Filtration
pH After Removal 7.6 7.58 7.5 7.44

Eh After Removal mY 469 480 549 556

Color after 30 mins Pt-Co
settlement 298 315 456 448
unit
Color after 90 mins Pt-Co
settlement 179 188 225 220
unit
Color after 3 hrs Pt-Co
120 128 162 151
settlement unit
Color after Pt-Co
Filtration 8 8 12 15
unit

145
Table B-3: Effect of pH on manganese removal from groundwater by oxidation
with KMn04 (Stoichiometry ratio =1.0)

Unit I 2 3 4 5 6 7
Initial Mn mg!
2.01 2.03 1.99 2.02 1.99 2.01 1.98
Concentration L
Adjusted pH 4.82 6 7.1 7.5 8.07 9.1 10
Mn conc after
mg!
30 mins 0.96 0.78 0.33 0.28 0.16 0.18 0.21
L
settling
Mn
mg!
Concentration 0.45 0.3 0.14 0.07 0.02 0 0
L
After Filtration
MnRemoval
After 30 mins % 52.24 61.58 83.42 86.14 91.96 91.04 89.39
Settling
MnRemoval
% 77.61 85.22 92.96 96.53 98.99 100.00 100.00
After Filtration
pH After
7.54 7.68 7.78 7.86 8.20 8.34 8.78
Removal
Eh After
mV 182 205 228 230 239 235 232
Removal
Pt-
Color After
Co 92 85 81 75 68 75 71
Removal
unit

146

•
Table 8-4: Effect ofKMn04 dose on oxidation of manganese in groundwater by varying the stoichiometric fraction

Unit 1a 1b 2a 2b 3a 3b 4a 4b 5a 5b 6a 6b
Initial Mn
mg/L 2.01 2.02 1.99 2.02 1.98 1.98 2.01 2.03 1.98 2.01 2.01 1.98
Concentration
KMn04 Fraction
of Stoichiometric 0.25 0.25 0.5 0.5 0.8 0.8 1 1 1.2 1.2 1.5 1.5
Ratio
Mn Concentration
After 30 mins mg/L 1.46 1.51 0.78 0.69 0.36 0.33 0.28 0.29 0.42 0.39 0.95 0.93
Settlement
Mn Concentration
mg/L 1.03 1.07 0.56 0.53 0.08 0.09 0 0.01 0.06 0.05 0.31 0.28
After Filtration
Color After 30 Pt-Co
224 236 110 119 78 74 85 91 130 141 359 385
mins Settlement unit

pH After Removal 7.57 7.56 7.84 7.84 7.95 7.98 8.1 8.05 8.01 8.02 8.02 8.04

Eh After Removal mY 174 177 196 192 210 215 224 230 218 .222 212 202
% MnRemoval
% 48.76 47.03 71.86 73.76 95.96 95.45 100.00 99.51 96.97 97.51 84.58 85.86
After Filtration
% MnRemoval
After 30 Mins % 27.36 25.25 60.80 65.84 81.82 83.33 86.07 85.71 78.79 80.60 52.74 53.03
Settlement

147
Table B-5: Removal of color produced during oxidation of manganese present in
groundwater by potassium permanganate using 10 cm depth sand filter.
(KMn04 Stoichiometry ratio = 1.2)

Unit 1 2 3 4 5
Initial Mn
mglL 2.01 2.01 2.01 2.01 2.01
Concentration
Initial Color Pt-Co
244 244 244 244 244
After Mixing unit
Time Interval for
Mins 30 30 30 30 30
Measurement
Volume of
ml 285 268 230 192 178
Filtrate Water
Rate of Filtration m1/min 9.50 8.93 7.67 6.40 5.93

Contact Time Mins 1.05 1.12 1.30 1.56 1.69

Color After Pt-Co
18 18 15 8 7
Filtration unit
Residual Mn
mg/L 0.03 0.04 0.03 0.02 0.02
after Filtration
Color Removed
% 92.62 92.62 93.85 96.72 97.13
by Filtration
Mn Removal by
% 98.51 98.01 98.51 99.00 99.00
Filtration ,

148
Table B-6: Removal of color produced during oxidation of manganese present in
groundwater by potassium permanganate using 20 cm depth sand filter.
(KMn04 Stoichiometry ratio = 1.2)

Unit 1 2 3 4 5
Initial Mn
mglL 2.03 2.03 2.03 2.03 2.03
Concentration
Initial Color Pt-Co
225 225 225 225 225
After Mixing unit
Time Interval for
Mins 30 30 30 30 30
Measurement
Volume of
ml 495 475 435 420 395
Filtrate Water
Rate of Filtration ml/min 16.50 15.83 14.50 14.00 13.17

Contact Time Mins 1.21 1.26 1.38 1.43 1.52

Color After Pt-Co
Filtration 6 7 4 2 2
unit
Residual Mn
mg/L 0.04 0.03 0.02 0.01 0.02
after Filtration
Color Removed
% 97.33 96.89 98.22 99.11 99.11
by Filtration
Mn Removal by
Filtration % 98.03 98.52 99.01 99.51 99.01

149
 Table B-7: Oxidation of groundwater containing different amount of manganese by chlorination at pH 7.5. (Stoichiometry ratio = 1.2)

Unit 1a 1b 2a 2b 3a 3b 4a 4b Sa Sb
Initial Mn
mg/L 1.01 1.02 2.01 2.02 4.98 5.02 7.5 7.48 9.97 10.01
Concentration
Chlorine Dose
mg/L 1.548 1.548 3.096 3.096 7.74 7.74 11.61 11.61 15.48 15.48
Added
Mn Concentration
after 30 min mg/L 0.56 0.61 1.29 1.35 3.21 3.31 5.62 5.64 7.79 7.81
settling
Mn Concentration
mg/L 0.51 0.54 1.18 1.21 3.16 3.21 5.49 5.4 7.58 7.7
After Filtration
MnRemoval
After 30 mins % 44.55 40.20 35.82 33.17 35.54 34.06 25.07 24.60 21.87 21.98
Settlement
Mn Removal After
% 49.50 47.06 41.29 40.10 36.55 36.06 26.80 27.81 23.97 23.08
Filtration

pH after removal 7.24 7.28 7.37 7.35 7.36 7.37 7.38 7.41 7.42 7.4

Eh After Removal mY -58 -65 -77 -74 -68 -70 -60 -58 -54 -60

Residual Clz mg/L 1.06 1.04 1.21 1.24 1.56 1.51 1.72 1.79 1.82 1.88

Color After 30
Pt-Co unit 0 4 0 0 5 4 0 4 8 9
mins Settlement

150
.0
 Table B-8: Oxidation of groundwater containing different amount of manganese by chlorination at pH 8.5. (Stoichiometry ratio = 1.2)

Unit 1a 1b 2a 2b 3a 3b 4a 4b 5a 5b
Initial Mn
mglL 1.01 1.03 2.01 2.01 4.98 4.99 7.49 7.51 9.98 10.02
Concentration "

Chlorine Dose
mglL 1.548 1.548 4.64 4.64 7.74 7.74 11.61 11.61 15.48 15.48
Added
Mn Concentration
after 30 min mglL 0.25 0.27 0.61 0.62 1.69 1.6 3.01 2.95 4.39 4.45
settling ,

Mn Concentration
mglL 0.17 0.18 0.4 0.39 1.12 1.15 1.97 1.92 2.72 2.66
After Filtration
MnRemoval
After 30 mins % 75.25 73.79 69.65 69.15 66.06 67.94 59.81 60.72 56.01 55.59
Settlement
MnRemoval'
% 83.17 82.52 80.10 80.60 77.51 76.95 73.70 74.43 72.75 73.45 '
After Filtration

pH after removal 8.25 8.21 8.29 8.31 8.42 8.4 8.37 8.35 "8.39 8.41

Eh After Removal mV -72 -70 -80 -82 -105 -104 -155 -154 -174 -176

Residual Cb mglL 0.41 0.38 0.52 0.55 0.99 0.87 1.04 1.09 1.3 1.35

Color After 30 Pt-Co

0 4 0 0 5 8 12 17 18 22
mins Settlement unit -

151
..
r,
'
 Table B-9 Oxidation of groundwater containing different amount of manganese by chlorination at pH 10.0 with variation in contact time.
(Stoichiometry ratio = 1.2)

Unit la Ib 2a 2b 3a 3b 4a 4b 5a 5b
Initial Mn
mg/L 1.02 0.99 2.01 2.02 5.01 5.02 7.51 7.48 10.04 10.01
Concentration
Chlorine Dose
mg/L 1.548 1.548 4.64 4.64 7.74 7.74 11.61 11.61 15.48 15.48
Added
Mn Concentration
mg/L 0.41 0.39 0.73 0.68 1.48 1.52 2.02 2.14 2.67 2.55
After 15 min contact
Mn Concentration
mg/L 0.15 0.16 0.21 0.24 0.35 0.39 0.43 0.41 0.51 0.59
After 25 min contact
Mn Concentration
mg/L 0.01 0 0.02 0.01 0.05 0.03 0.08 0.07 0.08 0.08
After 35 min contact
Mn Concentration
mg/L 0 0 0 0 0.03 0.01 0.04 0.02 0.03 0.02
After 30 min settling
Mn Concentration
mg/L 0 0 0 0 0 0 0 0 0 0
After Filtration
Mn Removal After
% 59.80 60.61 63.68 66.34 70.46 69.72 73.10 71.39 73.41 74.53
15 min contact

Mn Removal After
% 85.29 83.84 89.55 88.12 93.01 92.23 94.27 94.52 94.92 94.11
25 min contact

~ 152
 Table 8-9 (Continued.): Oxidation of groundwater containing different amount of manganese by chlorination at pH 10.0 with variation in
contact time. (Stoichiometry ratio = 1.2)

Unit 1a Ib 2a 2b 3a 3b 4a 4b Sa Sb
Mn Removal After
% 99.02 100.00 99.00 99.50 99.00 99.40 98.93 99.06 99.20 99.20
35 min Contact
Mn Removal After
% 100.00 100.00 100.00 100.00 99.40 99.80 99.47 99.73 99.70 99.80
30 min Settlement
Mn Removal After
% 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Filtration

pH After Removal 9.72 9.7 9.75 9.69 9.74 9.76 9.68 9.76 9.68 9.74

Eh After Removal mY -88 -86 -76 -70 -74 -75 -70 -89 -72 -73

Residual Clz mglL 0.28 0.26 0.42 0.39 0.81 0.82 0.95 0.99 1.19 1.12

Color After 30 min Pt-Co

4 6 2 5 0 4 2 0 0 0
Settlement unit

153
.
Table B-10: Effect of pH on oxidation of manganese in groundwater by chlorine
(Stoichiometry ratio = 1.0)

Unit I 2 3 4 5 6 7
Initial Mn
mg/L 2.01 2.02 2.01 2.0 2.03 2.02 2.04
Content

Adjusted pH 5.2 6.02 7.04 7.5 7.98 9 10.01

Chlorine dose
mglL 2.58 2.58 2.58 2.58 2.58 2.58 2.58
added
Residual Mn
After 30 mins mglL 1.89 1.84 1.73 1.42 0.86 0.39 0.03
Settling
Mnin
Solution After mglL 1.84 1.76 1.64 1.15 0.62 0.16 0
Filtration
MnRemoval
After 30 mins % 5.97 8.91 13.93 29.00 57.64 80.69 98.53
Settling
MnRemoval
After % 8.46 12.87 18.41 42.50 69.46 92.08 100.00
Filtration
pH After
5.47 6.34 7.21 7.46 7.90 8.55 8.73
removal (

Eh After rem
mY 91 36 -23 -42 -71 -88 -100
(mY)
Residual Cb mglL 2.08 2.01 1.78 1.52 1.26 0.97 0.51
Color after Pt-Co
4 8 12 17 22 8 5
removal unit

154
Table B-11: Removal of manganese present in natural groundwater with simple
aeration with variation in contact time

Unit 1 2 3 4 5
Initial Mn
mglL 2.03 2.04 2 .2.02 2.02
Concentration

Initial pH 7.04 8.08 9.1 10.02 11

Mn Concentration
mglL 1.86 1.54 0.82 0.26 0.1
After 20 min Mixing
Mn Concentration
mglL 1.82 1.41 0.63 0.15 0.05
After 30 min Mixing
Mn Concentration
mglL 1.71 0.98 0.22 0.04 0.02
After 60 min Mixing .

Mn Concentration
mglL 1.68 0.92 0.2 0.02 0.01
After Filtration
Mn Removal After
20 min Mixing % 8.37 24.51 59.00 87.13 95.05

Mn Removal After
30 min Mixing % 10.34 30.88 68.50 92.57 97.52

Mn Removal After
60 min Mixing % 15.76 51.96 89.00 98.02 99.01

Mn Removal After
Filtration % 17.24 54.90 90.00 99.01 99.50

pH After Removal 6.94 7.91 8.86 9.82 10.74

Eh After Removal mV 5 -72 .110 -170 .228

Pt. Co
Color After Removal 0 4 2 4 4
unit

155

'.
 APPENDIX "C"
Study on Manganese Oxidation in Presence of Iron
Table C-l: Manganese removal from groundwater by oxidation with KMn04 in presence of dissolved iron in water
(Initial Mn concentration = 5.0 mglL; Stoichiometry ratio =1.2)

Unit la Ib 2a 2b 3a 3b 4a 4b
Initial pH 7.65 7.62 7.64 7.62 7.7 7.65 7.61 7.67
Initial Mn
mg/L 5.02 5.01 4.98 5.01 5.03 5.02 4.99 5.02
Concentration
Mn Concentration
After 30 mins mg/L 1.76 1.88 1.54 1.48 1.32 1.26 1.21 1.2
Settlement
Mn Concentration
mg/L 0.08 0.06 0.11 0.12 0.18 0.21 0.49 0.51
After Filtration
Mn Removal After
% 64.94 62.48 69.08 70.46 73.76 74.90 75.75 76.10
30 mins settlement
Mn Removal After
% 98.41 98.80 97.79 97.60 96.42 95.82 90.18 89.84
Filtration
Initial Fe
mg/L I I 3 3 5 5 10 10
Concentration
Fe Concentration
After 30 mins mg/L 0.33 0.42 1.21 1.29 1.72 1.82 2.87 3.25
Settlement
Fe Cone after
mg/L 0 0 0 0 0 0 0 0
Filtration
Fe Removal After
% 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Filtration

Final pH 7.77 7.72 7.65 7.68 7.84 7.82 7.88 7.85

157
 Table C-2: Manganese removal from groundwater by oxidation with KMn04 in presence of dissolved iron in water
(Initial Mn concentration = 2.0 mg/L; Stoichiometry ratio =1.2)

Unit la Ib 2a 2b 3a 3b 4a 4b
Initial pH 7.61 7.64 7.68 7.57 7.59 7.7 7.66 7.58
Initial Mn
mg/L
Concentration 2.02 2.01 1.99 2.02 2 2.03 2.03 2.02
Mn Concentration
After 30 mins mglL
Settlement 0.14 0.15 0.25 0.22 0.37 0.35 0.51 0.57
Mn Concentration
mglL
After Filtration 0 0.01 0.04 0.03 0.22 0.2 0.44 0.41
Mn Removal After
%
30 mins settlement 93.07 92.54 87.44 89.11 81.50 82.76 74.88 71.78
Mn Removal After
%
Filtration 100.00 99.50 97.99 98.51 89.00 90.15 78.33 79.70
Initial Fe
mg/L
Concentration 1 1 3 3 5 5 10 10
Fe Concentration
After 30 mins mglL
Settlement 0.46 0.48 0.71 0.68 1.02 1.08 1.48 1.56
Fe Cone after
mglL
Filtration 0 0 0 0 0.03 0.02 0.06 0.07
Fe Removal After
%
Filtration 100.00 100.00 100.00 100.00 99.40 99.60 99.40 99.30
Final pH
7.83 7.8 7.77 7.78 7.84 7.86 7.9 7.92

158
1"
 Table C-3: Manganese removal from groundwater by oxidation with KMn04 in presence of dissolved iron in water
(Initial Mn concentration = 2.0 mg/L; Stoichiometry ratio =1.0)

Unit 1a 1b 2a 2b 3a 3b 4a 4b
Initial pH 7.55 7.69 7.58 7.71 7.56 7.55 7.7 7.68

Initial Mn
mg/L 1.98 2.01 1.99 1.98 2.02 1.97 2.02 2.01
Concentration
Mn Concentration
After 30 mins mg/L 0.65 0.69 0.8 0.82 1.31 1.28 1.56 1.55
Settlement
Mn Concentration 1.44
mg/L 0.08 0.05 0.3 0.35 0.91 0.89 1.42
After Filtration
Mn Removal After
% 67.17 65.67 59.80 58.59 35.15 35.03 22.77 22.89
30 mins settlement
Mn Removal After
% 95.96 97.51 84.92 82.32 54.95 54.82 29.70 28.36
Filtration
Initial Fe
mg/L 1 1 3 3 5 5 10 10
Concentration
Fe Concentration
After 30 mins mg/L 0.28 0.3 0.53 0.54 0.82 0.92 1.7 1.76
.
Settlement
Fe Conc after
mgIL 0 0 0 0 0.05 0.04 0.09 0.08
Filtration
Fe Removal After
% 100.00 100.00 100.00 100.00 99.00 99.20 99.10 99.20
Filtration

Final pH 7.91 7.92 7.87 7.85 7.84 7.83 7.9 7.88

159
)
'.
f1lo
Table C-4: Manganese removal from artificially prepared, low alkalinity and low pH water by oxidation with KMn04 in presence of
dissolved iron at condition. (Stoichiometry ratio =1.0)

Unit la Ib 2a 2b 3a 3b 4a 4b
Adjusted pH 7.71 7.68 7.62 7.6 7.74 7.70 7.61 7.64
mg/L as
Adjusted Alkalinity 89 91 101 98 100 98 95 102
CaC03
Initial Mn
mglL 2.01 1.99 1.99 2.01 2.03 2.02 2 2.02
Concentration
Mn After Filtration mglL 0.15 0.18 0.37 0.41 1.1 1.08 1.59 1.62

Mn Removal After
% 92.54 90.95 81.41 79.60 45.81 46.53 20.50 19.80
Filtration
Initial Fe
mg/L I 1 3 3 5 5 10 10
Concentration
Fe after 15 mins
mg/L 0.88 0.85 2.69 2.8 4.49 4.34 8.04 7.92
mlxmg
Fe Conc after
mg/L 0 0 0.03 0.04 0.16 0.18 0.3 0.37
Filtration
Fe Removal After
% 100.0 100.0 98.88 98.57 96.44 95.85 96.27 95.33 .
Filtration
Final pH 7.81 7.77 7.6 7.72 7.61 7.54 7.38 7.41
Color After 30 Pt-Co
24 15 84 79 115 118 178 160
mins settlement unit

160
 Table C-5: Simultaneous removal of manganese and iron from groundwater by oxidation with KMn04 dose at varying stoichiometric
fraction with respect to initial iron and manganese concentrations

Unit 1a 1b 2a 2b 3a 3b 4a 4b
combined KMn04
025 0.25 0.5 0.5 0.75 0.75 1 1
ratio

Initial pH 694 6.93 6.91 7 6.91 6.96 6.94 7.04

Initial Mn
mglL 1.99 2.01 1.98 2 2.01 2.03 2.02 1.97
Concentration
Mn Concentration
After 30 min mglL 1.83 1.82 1.14 1.19 0.31 0.35 0.55 0:58
settlement
Mn Concentration
mglL 1.72 1.75 0.8 0.78 0.14 0.12 0.33 .0.3
After Filtration
Mn Removal After
% 13.57 12.94 42.42 40.50 84.58 82.76 72.77 70.56
30 mins settlement
Mn Removal After
% 8.04 9.45 59.60 61.00 93.03 94.09 83.66 84.77
Filtration
Initial Fe
mg/L 5 5 5 5 5 5 5 5
Concentration
Fe Concentration
After 30 mins mg/L 0.88 0.87 0.59 0.62 0.41 0.42 0.25 0.3
Settlement
Fe Concentration
mg/L 0 0 0 0 0 0 0 0
After Filtration
Fe Removal After
% 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
Filtration
pH After Removal 7.45 7.42 7.75 7.71 8.1 8.04 7.9 7.79
Color After 30 mins Pt-Co
22 31 61 52 88 93 131 140
settlement unit

~ 161
 Table C-6: Potassium permanganate oxidation of manganese present in groundwater (containing dissolved iron) with pre- oxidation of
iron by aeration at elevated pH (Stoichiometric ratio ofKMn04 = 1.0)

Unit la Ib 2a 2b 3a 3b 4a 4b

Adjusted pH
8.1 8.06 8.01 8.09 8.08 8.1 8.11 8.08
Initial Mn
mglL
Concentration 1.99 2.01 1.97 1.99 2.01 1.98 1.98 2.02
Mn Concentration
mg/L
After Filtration 0 0 0 0 0 0 0 0
Mn Removal After
% 100.00 100.00
Filtration 100.00 100.00 100.00 100.00 100.00 100.00
Initial Fe
mg/L
Concentration. 1 1 3 3 5 5 10 10
Fe Concentration
after 15 mins mg/L
stirrina 0.68 0.71 2.02 1.89 3.14 3.18 5.11 5.26
Fe Concentration
mg/L
after Filtration 0 0 0 0 0 0 0 0
Fe Removal after
%
Filtration 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00

pH after removal
7.79 7.78 7.74 7.71 7.69 7.7 7.62 7.65
Pt-Co
Calor after removal
unit 62 58 49 55 70 58 66 72

..•
,
.~~.
...• <./
~ 162
Table C-7: Oxidation of manganese by aeration in presence of iron in groundwater in varying concentration

Unit la Ib 2a 2b 3a 3b 4a 4b
Initial pH 7.68 7.65 7.71 7.7 7.69 7.74 7.7 7.72

Initial Mn
mg/L 2.03 2.02 2.03 2.01 2.01 2.03 2.02 2.01
Concentration
Mn Concentration
After 30 min mg/L 1.78 1.75 1.74 1.71 1.44 1.39 .1.74 1.78
settlement
Mn Concentration
mg/L 1.56 1.58 1.59 1.58 1.62 1.65 1.66 1.65
After Filtration
Mn Removal After
% 12.32 13.37 14.29 14.93 28.36 31.53 13.86 11.44
30 mins settlement
Mn Removal After
% 23.15 21.78 21.67 21.39 19.40 18.72 17.82 17.91
Filtration
Initial Fe
mg/L 1 1 3 3 5 5 10 10
Concentration
Fe Concentration
After 30 mins mgIL 0.17 0.14 0.53 0.54 1.42 1.39 3.99 4.28
Settlement
Fe Concentration
mgIL 0.08 0.09 0.45 0.41 1.26 1.24 3.56 3.5
After Filtration
Fe Removal After
% 83.00 86.00 82.33 82.00 71.60 72.20 60.10 57.20
30 mins Settlement
Fe Removal After
% 92.00 91.00 85.00 86.33 74.80 75.20 64.40 65.00
Filtration

pH After Removal 7.84 7.81 7.78 7.8 7.74 7.75 7.8 7.82

163
 APPENDIX "D"
Study OilMallgallese Removal by Coagulatioll-Adsorptioll- Coprecipitatioll
Table D-l: Removal of manganese from groundwater by coagulation-adsorption-coprecipitation using ferrous sulfate [Fe(II)] as coagulant
{ with addition of potassium permanganate as oxidant (Stoichiometry ratio for KMn04 =1.0)

Unit la Ib 2a 2b 3a 3b 4a 4b Sa 5b
Initial pH 7.55 7.52 7.6 7.62 7.61 7.58 7.49 7.54 7.64 7.6

Initial Mn 2.02
mg/L 2.01 2.01 1.99 1.98 1.99 2.01 1.98 2.02 2.03
Concentration

Fe (II) Dose mg/L 25 25 15 15 10 10 5 5 2 2

Mn Concentration 0.46 0.49

mg/L 0.35 0.33 0.47 0.44 0.77 0.79 0.68 0.67
After Coagulation

Mn Concentration 0.4 0.43

mg/L 0.27 0.28 0.42 0.38 0.73 0.75 0.63 0.61
After Filtration
.
Mn Removal After 77.34 75.74
% 82.59 83.58 76.38 77.78 61.31 60.70 65.66 66.83
Coagulation

Mn Removal After
% 86.57 86.07 78.89 80.81 63.32 62.69 68.18 69.80 80.30 78.71
Filtration

Fe Concentration 0.39 0.33 0.05 0.08

mg/L 4.79 4.96 2.87 2.71 0.68 0.74
After Coagulation .

Fe Concentration
mg/L 0.04 0.03 0.04 0.05 0.03 0.04 0.01 0 0 0
After Filtration

pH After Removal 7.77 7.69 7.78 7.71 7.75 7.7 7.7 7.76 7.72 7.73

Color After Pt-Co

90 88 84 80 90 79 89 90 85 88
Coagulation unit

165
Table D-2: Removal of manganese from groundwater by coagulation-adsorption-coprecipitation using ferrous sulfate [Fe(IT)] as
coagulant without addition of potassium permanganate

Unit la Ib 2a 2b 3a 3b 4a 4b Sa Sb
Initial pH 7.61 7.65 7.62 7.63 7.69 7.64 7.7 7.68 7.66 7.59
Initial Mn
mg/L 2.01 2.01 1.99 1.98 1.99 2.01 1.98 2.02 2.03 2.02
Concentration

Fe (II) Dose mg/L 25 25 15 15 10 10 5 5 2 2

Mn Concentration
mg/L 1.59 1.58 1.48 1.47 1.41 1.4 1.36 1.35 1.31 1.32
After Coagulation
Mn Concentration
mg/L 1.35 1.36 1.31 1.33 1.29 1.31 1.24 1.21 1.26 1.22
After Filtration
MnRemoval
% 20.90 21.39 25.63 25.76 29.15 30.35 31.31 33.17 35.47 34.65
After Coagulation .

MnRemoval
% 32.84 32.34 34.17 32.83 35.18 34.83 37.37 40.10 37.93 39.60
After Filtration

Fe Concentration
mg/L 5.29 5.16. 3.32 3.44 1.68 1.61 0.77 0.73 0.32 0.29
After Coagulation

Fe Concentration
mg/L 2.38 2.51 1.36 1.49 0.84 0.79 0.26 0.28 0.15 0.13
After Filtration

Fe Removal After
% 78.84 79.36 77.87 77.07 83.20 83.90 84.60 85.40 84.00 85.50
Coagulation

pH After Removal 6.68 6.69 6.78 6.72 6.9 6.94 7.02 7.1 7.4 7.42

166
 Table D-3: Removal of manganese from groundwater by coagulation-adsorption-coprecipitation using ferric chloride [Fe(III)] as
coagulant with addition of potassium permanganate as oxidant (Stoichiometry ratio for KMn04 =1.0)

Unit 1a 1b 2a 2b 3a 3b
Initial pH 7.71 7.67 7.72 7.7 7.64 7.61

Fe (III) Dose mg/L 10 10 15 15 25 25

Initial Mn
mg/L 1.99 1.98 2.01 2 1.98 1.97
Concentration
f
Mn Concentration ~
mg/L 0.14 0.15 0.1 0.09 0.07 0.08 1
After Coagulation
Mn Concentration
mg/L 0.02 0.03 0 0.01 0 0
After Filtration
MnRemoval
% 92.96 92.42 95.02 95.50 96.46 95.94
After Coagulation
MnRemoval
% 98.99 98.48 100.00 99.50 100.00 100.00 ~
After Filtration

pH after removal 7.83 7.88 7.89 7.9 7.94 7.91

Color after
Pt-Co unit 92 109 126 131 166 155
removal

~ 167
Table 0-4: Removal of manganese from groundwater by coagulation-adsorption-co precipitation using ferric chloride [Fe(III)]
as coagulant without addition of potassium permanganate as oxidant

Unit la Ib 2a 2b 3a 3b
Initial pH 7.64 7.6 7.59 7.6 7.68 7.64

Fe (III) Dose mg/L 10 10 15 15 25 25

Initial Mn
mg/L 2 2.02 2.03 2.02 2.03 2.03
Concentration
Mn Concentration
mg/L 1.51 1.49 1.38 1.37 1.26 1.29
After Coagulation

Mn Concentration
mg/L 1.42 1.41 1.33 1.32 1.22 1.19
After Filtration

MnRemoval
% 24.50 26.24 32.02 32.18 37.93 36.45
After Coagulation

MnRemoval
% 29.00 30.20 34.48 34.65 39.90 41.38
After Filtration

pH after removal 6.86 6.91 6.94 6.92 6.8 6.85

Color after
Pt-Co unit 23 18 .27 30 35 42
removal

168
 Table D-5: Effect of pH on Mn removal by coagulation using ferric chloride [Fe(III)] as coagulant

Unit 1 2 3 4 5 6 7

Initial pH 7 7.25 7.7 8.3 9 10 11

Initial Mn
mg/L 15 15 15 15 15 15 15
Concentration

Fe (III) Dose mg/L 1.97 2.01 1.98 1.95 2.01. 2.03 1.97
..
Mn Concentration
mg/L' 1.58 1.56 1.41 1.35 1.28 0.46 0.02
After Coagulation

Mn Concentration
mg/L 1.45 1.40 1.26 1.25 1.15 0.31 0.01
After Filtration
Mn Removal After
% 19.80 22.39 28.79 30.77 36.32 77.34 98.98
Coagulation
Mn Removal After
% 26.40 30.35 36.36 35.90 42.79 84.73 99.49
Filtration
Fe Concentration
mg/L 0.02 0.04 0 0 0 0 0.04
After Filtration
Fe Removal After
% 99.87 99.73 100.00 100.00 100.00 100.00 99.73
Filtration

pH after removal 6.78 6.91 7.02 7.67 7.29,' 7.89 9.36

.'
()
n..
169