Adsorption of dimethylarsinate (DMA) monomethylarsonate (MMA), and arsenate on goethite (alpha FeOOH

ADSORPTION OF DIMETHYLARSINATE (DMA),
MONOMETHYLARSONATE (MMA), AND ARSENATE ON
GOETHITE (α-FeOOH)










ZHANG JUNSHE
( M.Eng., Tianjin University )







A THESIS SUBMITED
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF CHEMICAL & BIOMOLECULAR
ENGINEERING
NATIONAL UNIVERSITY OF SINGAPORE
2005
DEDICATION

To my mother and father



i
ACKNOWLEDGEMENTS

First, I am particular grateful to my supervisors, A/P Robert Stanforth and A/P Simo
Olavi Pehkonen, for their continual encouragement, support and inspiration throughout the
entire investigation. Their kind understandings and characters make the study pleasing.

I would also like to thank my colleagues, Mr. Jiao Lei, Mr. Zhong Bin, Mr. Tian Kun,
Ms. Thet Su Hliang, Dr. Xu Ran, Mr. Xu Tongjiang and all other friends who shared with
me their knowledge and experience as well as their support.

Special thanks go to Ms. Li Fengmei, Ms. Chia Yuit Ching, Mr. Chia Phai Ann and Ms.
Li Xiang for their instructions with the equipment.

I wish to thank my parents, my wife Li Meiling, and two younger sisters for their
support and understanding.

Finally, I would like to express my sincere gratitude to the National University of
Singapore for providing a research scholarship to make this research project possible.


ii
TABLE OF CONTENTS

1.1 Arsenic in natural systems 1

1.2 Arsenic adsorption on goethite 3
1.3 Objectives and scope 4
2.1 Aqueous speciation of arsenic 7
2.2 The physical properties of goethite 8
2.3 Surface complexation models (SCMs) 10
2.4 Charge development on the goethite surface 12
2.5 Interaction between arsenic and the goethite surface 16
2.5.1 Surface species of adsorbed arsenic 17
2.5.2 Proton-anion adsorption ratios 19
2.5.3 Adsorption kinetics 20
2.5.4 Competitive adsorption between arsenic and phosphate 23
3.1 Synthesis and characterization 25
ACKNOWLEDGEMENTS i
TABLE OF CONTENTS ii
SUMMARY v
LIST OF FIGURES vii
LIST OF TABLES xii
LIST OF SYMBOLS xiv
NOMENCLATURE xvii
Chapter 1
Introduction 1
Chapter 2
Literature Review 7
Chapter 3
Materials and Methods 25


iii
3.2 Adsorption isotherms 25

3.3 Adsorption edges 26
3.4 Methods to determine surface coverages 26
3.5 CO
2
-free system 27
3.6 Acid-base titration 28
3.7 Arsenic and phosphate analysis and error analysis 28
3.8 Zeta potential 29
3.9 Effective particle sizes 29
3.10 Adsorption and desorption kinetics 30
3.11 Back titration to determine proton-anion adsorption ratios 30
3.12 Competitive adsorption between arsenic and phosphate 31
4.1 Characterization of goethite 32
4.1.1 Specific surface area and porosity of goethite 32
4.1.2 Morphology of goethite 34
4.1.3 Acid-base titration 37
4.1.4 Estimation of the density of reactive sites 39
4.2 Solubility of goethite and ferric arsenate 41
4.3 Adsorption edges 43
4.3.1 Error analysis 44
4.3.2 Effect of pH and background electrolyte on the uptake of DMA 44
4.3.3 Effect of pH and background electrolyte on the uptake of MMA 47
4.3.4 Effect of pH and background electrolyte on the uptake of arsenate 49
4.3.5 Adsorption edges of three arsenic species 51
4.4 Adsorption isotherms 53
4.4.1. Methods to determine surface coverages and error analysis 53
4.4.2 Adsorption isotherms at pH 4.00 and 7.20 56
4.4.3 Effect of the solids concentration 60
4.5 Adsorption and desorption kinetics 64
4.5.1 Error analysis 64

4.5.2 Zeta potential and effective particle size 65
Chapter 4 Results and Discussion 32


iv
4.5.3 Effect of background electrolyte concentrations on adsorption kinetics 72

4.5.4 Effect of pH on absorption kinetics 77
4.5.5 Rate-determining step during adsorption 79
4.6 Proton-anion adsorption ratios 86
4.6.1 Charge development on the goethite surface 87
4.6.2 Method to calculate proton-anion adsorption ratios 96
4.6.3 Error analysis 97
4.6.3 Proton-anion adsorption ratios 98
4.7 Competitive adsorption between arsenic and phosphate 111
4.7.1 Error analysis 112
4.7.2 Effect of the order of addition 112
4.7.3 Replacing the adsorbed anion with another anion 120
4.8 Acid-base properties of surface groups and adsorption mechanism 131
4.9 Environmental implications and industrial applications 133
5.1 Summary 136
5.2 Recommendation future research 141
A Desorbility of arsenate at pH 4.00 using 0.1 M NaNO
3
solution 160
B Fraction of various charged hydroxyls on the goethite surface 162
C Publications 164

Chapter 5
Conclusions and Recommendations 136
References 144
Appendices 160


v
SUMMARY

Arsenic (As) is highly toxic to humans and is widespread in aquatic environments, soils,
and sediments due to natural and anthropogenic sources. The bioavailability of arsenic in
oxidized systems is mainly influenced by its interactions with the mineral surface.
Adsorption is one of these important interactions. Adsorption of inorganic arsenic on
goethite (α-FeOOH) has been studied for many years; however, the adsorption mechanism
of arsenic is still in debate and the reactivity of the goethite surface is not well understood.

The purpose of this study was to investigate the adsorption mechanism of three arsenic
species – arsenate, monomethylarsonate (MMA), and dimethylarsinate (DMA)-on goethite
and chemical properties of the goethite surface by comparing the individual adsorption
edges, isotherms, and effect on zeta potential of the three arsenic species, and the
competitive adsorption behaviors between the three arsenic species and phosphate.
Adsorption kinetics and proton-anion adsorption ratios provide very useful insight into the
chemical properties of the goethite surface.

DMA, MMA, and arsenate form inner-sphere complexes by ligand exchange on the
goethite surface. At low surface coverages, MMA and arsenate predominately adsorb as
bidentate complexes; however the contribution of monodentate complexes is important at
high surface coverages. As pH decreases and the surface coverage increases, some adsorbed
arsenate is protonated. The low affinity of DMA for goethite compared with MMA or
arsenate and the influence of electrolyte anion and carbonate on the adsorption of DMA
suggests that DMA only forms monodentate complexes. Adsorbed DMA can be completely


vi
displaced by phosphate, but only a portion of adsorbed MMA or arsenate can be displaced
by phosphate, indicating that adsorption of MMA or arsenate is not totally reversible with
respect to phosphate. The non-exchangeable part of MMA or arsenate is attributed to
bidentate complexes, although some bidentate complexes are also exchangeable. DMA only
exchanges with phosphate adsorbed as monodentate complexes.

There are two types of reactive sites on the goethite surface for MMA or arsenate based
on the affinity: high-affinity sites and low-affinity sites. On the high affinity sites MMA or
arsenate adsorbs as bidentate complexes, and a part of the adsorbed MMA or arsenate on
these sites cannot be displaced by phosphate. Monodentate complex formation is the
dominant mechanism on the low affinity sites, and adsorption isotherms of MMA or
arsenate on these sites follow Freundlichian behavior. It seems that only one type of reactive
sites exists for DMA, since logarithmic plots of all adsorption data of DMA fall on one
straight line. The heterogeneity also gives rise the Elovichian kinetics of the three arsenic
species adsorption. Neither a 2-pK model, nor a two-site 1-pK or two-site 2-pK model can
interpret both the proton-arsenic adsorption ratio and the effect of adsorption of the three
arsenic species on the zeta potential. Singly coordinated hydroxyls are the reactive sites on
the goethite surface, but the contribution of other hydroxyls to the total surface charge and
particle charge is significant at certain pH values. Thus a 1-pK MUSIC (Multisite
Complexation) model should be used in describing the adsorption of anions on the goethite

surface.





vii
LIST OF FIGURES

Figure 2.1 Crystal morphology of large goethite twin (a) and the goethite particle (b)
reported by Cornell and Schwertmann (2003) 10

Figure 2.2 Schematic representation of the surface structure of the (101) face 16
Figure 2.3 Proposed AsO
4
surface complexes on the goethite surface. 18
Figure 4.1 N
2
-adsorption and desorption isotherms on goethite at 77.4K 33
Figure 4.2 Pores sizes distribution of goethite based on BJH method 34
Figure 4.3 TEM image of goethite crystals 35
Figure 4.4 FESEM image of goethite crystals 35
Figure 4.5 AFM image of goethite crystals. 36
Figure 4.6 Optical microscopy of goethite aggregates in suspension 37
Figure 4.7 A schematic presentation of goethite crystals 37
Figure 4.8 Acid-base titration curves of goethite at three NaNO
3
concentrations. Goethite
concentration is 10 g L
-1

. 39
Figure 4.9 Activity of single ion species and total Fe
III
ion concentrations in equilibrium
with goethite as a function of pH 42

Figure 4.10 Total As concentrations without scorodite precipitation in equilibrium with
goethite as a function of pH 43

Figure 4.11 The uptake of DMA as a function of pH and concentrations of NaNO
3
. Goethite
concentration is 1.98 g L
-1
, and the initial DMA concentration is 68 µM. 45
Figure 4.12 The uptake of DMA as a function of pH in 1.0 M NaCl, NaNO
3
and NaClO
3
.
Goethite concentration is 1.98 g L
-1
, and the initial DMA concentration is 60 µM.
46

Figure 4.13 The uptake of DMA as a function of pH in an open and a close system in 0.01
M NaNO
3
. Goethite concentration is 1.2 g L
-1

, and the initial DMA concentration
is 59 µM. 48

Figure 4.14 The uptake of MMA as a function of pH and concentrations of NaNO
3
.
Goethite concentration is 1.98 g L
-1
, and initial the MMA concentration is 213
µM 48



viii
Figure 4.15 The uptake of MMA as a function of pH in an open and a close system in 0.01
M NaNO
3
. Goethite concentration is 1.2 g L
-1
, and the initial MMA concentration
is 93 µM. 49

Figure 4.16 The uptake of arsenate as a function of pH and NaNO
3
concentrations. Goethite
concentration is 1.98 g L
-1
, and the initial arsenate concentration is 180 µM 50
Figure 4.17 The uptake of arsenate as a function of pH in an open and a close system in 0.01
M NaNO

3
. Goethite concentration is 1.2 g L
-1
, and the initial arsenate
concentration is 72 µM. 51

Figure 4.18 The uptake of the three arsenic species on goethite as function of pH in 0.01 M
NaNO3. Goethite concentration is 1.2 g L
-1
and the initial arsenic concentration is
70 µM 52

Figure 4.19 Adsorption isotherms of the three arsenic species on goethite at pH 4.00 in 0.01
M NaNO
3
. Goethite concentration is 1.2 g L
-1
57
Figure 4.20 Adsorption isotherms of the three arsenic species on goethite at pH 7.20 in 0.01
M NaNO
3
. Goethite concentration is 1.2 g L
-1
57
Figure 4.21 Logarithmic plots of adsorption of MMA and arsenate at two pH values in 0.01
M NaNO
3
. Filled symbols for adsorption at pH 4.00 and open symbols for
adsorption at pH 7.20. Goethite concentration is 1.2 g L
-1

60
Figure 4.22 Adsorption isotherms of DMA at pH 4.00 in 0.1 M NaNO
3
with four solids
concentrations 61

Figure 4.23 Adsorption isotherms of MMA at pH 4.00 in 0.1 M NaNO
3
with four solids
concentrations 62

Figure 4.24 Adsorption isotherms of arsenate at pH 4.00 in 0.1 M NaNO
3
with four solids
concentrations 63

Figure 4.25 Surface coverages (a) and zeta potential (b) of DMA adsorbed on goethite as a
function of pH in 0.001 M NaNO
3
with three initial DMA concentrations.
Goethite concentration is 0.25 g L
-1
67
Figure 4.26 Surface coverages (a) and zeta potential (b) of MMA adsorbed on goethite as a
function of pH in 0.001 M NaNO
3
with three initial MMA concentrations.
Goethite concentration is 0.25 g L
-1
68

Figure 4.27 Surface coverages (a) and zeta potential (b) of arsenate adsorbed on goethite as
a function of pH in 0.001 M NaNO
3
with three initial arsenate concentrations.
Goethite concentration is 0.25 g L
-1
69
Figure 4.28 The effective particle sizes of goethite at different initial DMA concentrations
in 0.001 M NaNO
3
. Goethite concentration is 0.25 g L
-1
. 70


ix
Figure 4.29 The effective particle sizes of goethite at different initial MMA concentrations
in 0.001 M NaNO
3
. Goethite concentration is 0.25 g L
-1
. 71
Figure 4.30 The effective particle sizes of goethite at different initial arsenate concentrations
in 0.001 M NaNO
3
. Goethite concentration is 0.25 g L
-1
. 71
Figure 4.31 Kinetics of DMA adsorption at pH 4.00 at three NaNO
3

concentrations.
Goethite concentration is 1.98 g L
-1
, and the initial DMA concentration is 130µM.
Lines are Elovich fitting. 73

Figure 4.32 Kinetics of MMA adsorption at pH 4.00 at three NaNO
3
concentrations.
Goethite concentration is 1.98 g L
-1
, and the initial MMA concentration is 217
µM. Lines are Elovich fitting 73

Figure 4.33 Kinetics of arsenate adsorption at pH 4.00 at three NaNO
3
concentrations.
Goethite concentration is 1.98 g L
-1
, and the initial arsenate concentration is 120
µM 74

Figure 4.34 Adsorption of DMA on goethite at pH 4.00 at three NaNO3 concentrations as
described by a plot of Γ vs. ln t. Goethite concentration is 1.98 g L-1, and the
initial DMA concentration is 130 µM 75
Figure 4.35 Adsorption of MMA on goethite at pH 4.00 at three NaNO
3
concentrations as
described by a plot of Γ vs. ln t. Goethite concentration is 1.98 g L
-1

, and the
initial MMA concentration is 217 µM 76

Figure 4.36 Adsorption of arsenate on goethite at pH 4.00 at three NaNO
3
concentrations as
described by a plot of Γ vs. ln t. Goethite concentration is 1.98 g L
-1
, and the
initial arsenate concentration is 120 µM 76

Figure 4.37 Adsorption of DMA on goethite at two pH values in 0.1 M NaNO
3
as described
by a plot of Γ vs. ln t. Goethite concentration is 1.98 g L
-1
, and the initial arsenate
concentration is 130 µM 78

Figure 4.38 Adsorption of MMA on goethite at three pH values in 0.1 M NaNO
3
as
described by a plot of Γ vs. ln t. Goethite concentration is 1.98 g L
-1
, and the
initial arsenate concentration is 217 µM 78

Figure 4.39 Adsorption of arsenate on goethite at three pH values in 0.1 M NaNO
3
as

described by a plot of Γ vs. ln t. Goethite concentration is 1.98 g L
-1
, and the
initial arsenate concentration is 130 µM 79

Figure 4.40 Effect of mixing methods (ultrasonication vs. magnetic stirring) on the
adsorption kinetics of arsenate at pH 4.20 in 0.001 M NaNO
3.
Goethite
concentration is 1.98 g L
-1
and the initial arsenate concentration is 120 µM 82


x
Figure 4.41 Adsorption (at pH 4.00) and desorption (at pH 10.0 and 12.0) kinetics of DMA
in 0.1 M NaNO
3.
Goethite concentration is 1.98 g L
-1
, and the initial DMA
concentration is 130 µM. 84

Figure 4.42 Adsorption (at pH 4.00) and desorption (at pH12.0) kinetics of MMA and
arsenate in 0.1 M NaNO
3
.Goethite concentrations is 1.98 g L
-1
. The initial
concentration of MMA is 217 µM and arsenate is 120 µM 85

Figure 4.43 Two proton affinity constants calculated from acid-base titration by
extrapolation to zero charge 89

Figure 4.44 Acid-base titration of goethite in 0.001 M NaNO
3
and surface charge predicted
by two models. Goethite concentration is 10 g L
-1
93
Figure 4.45 Distribution of various charged hydroxyls on the goethite surface as a function
of pH predicted by the 2-pK TLM model. 93

Figure 4.46 Distribution of various singly coordinated hydroxyls on the goethite surface as a
function of pH predicted by the two-site 1-pK BSM model 96

Figure 4.47 Proton-DMA adsorption ratios vs. surface coverages of DMA at pH 4.25 and
6.75. Goethite concentration is 2.0 g L
-1
and NaNO
3
concentration is 0.001M 99
Figure 4.48 Zeta potential of goethite vs. surface coverages of DMA at pH 4.25 and 6.75.
Goethite concentration is 0.25 g L
-1
and NaNO
3
concentration is 0.001M. 100
Figure 4.49 Relationship between the proton-MMA adsorption ratio and the surface
coverage of MMA at two pH values. Goethite concentration is 2.0 g L

-1
and
NaNO
3
concentration is 0.001M 104
Figure 4.50 Zeta potential as a function of MMA surface coverages at pH 4.00 and 6.77.
Goethite concentration is 0.25 g L
-1
and NaNO
3
concentration is 0.001M. 105
Figure 4.51 Relationship between proton-arsenate adsorption ratios and the surface
coverage of arsenate at two pH values. Goethite concentration is 2.0 g L
-1
and
NaNO
3
concentration is 0.001M 108
Figure 4.52 Zeta potential of goethite as a function of arsenate surface coverages at two pH
values. Goethite concentration is 0.25 g L
-1
and NaNO
3
concentration is 0.001M.
109

Figure 4.53 Total surface coverages of phosphate and DMA as a function of total
equilibrium concentrations in three systems and adsorption isotherms of DMA
and phosphate at pH 4.00 in 0.01 M NaNO
3

. Goethite concentration is 1.2 g L
-1
.
114

Figure 4.54 Total surface coverages of phosphate and MMA as a function of total
equilibrium concentrations in three systems and adsorption isotherms of MMA


xi
and phosphate at pH 4.00 in 0.01 M NaNO
3
. Goethite concentration is 1.2 g L
-1
.
118

Figure 4.55 Surface coverages of phosphate (filled symbols) and DMA (open symbols) as a
function of initial DMA/P molar ratios with three initial phosphate concentrations
at pH 4.00 in 0.01 M NaNO
3
. The values in bracket are the surface coverage of
phosphate without DMA. Goethite concentration is 1.2 g L
-1
121
Figure 4.56 Surface coverages of phosphate (filled symbols) and DMA (open symbols) as a
function of initial P/DMA molar ratios with an initial DMA concentration of 149
µM at pH 4.00 in 0.01 M NaNO
3
. The values in bracket are the surface coverage

of DMA without phosphate. Goethite concentration is 1.2 g L
-1
123
Figure 4.57 The relationship between the surface coverage of phosphate to DMA on
reversible sites and the equilibrium concentration of phosphate to DMA at pH
4.00 in 0.01 M NaNO
3
. Goethite concentration is 1.2 g L
-1
. 125
Figure 4.58 Surface coverages of phosphate (open symbols) and MMA (filled symbols) as
a function of initial MMA/P molar ratios with three initial phosphate
concentrations at pH 4.00 in 0.01 M NaNO
3
. The values in bracket are the
surface coverage of phosphate without MMA. Goethite concentration is 1.2 g L
-1
.
126

Figure 4.59 Surface coverages of phosphate (open symbols) and MMA (filled symbols) as a
function of initial P/MMA molar ratios with three initial MMA concentrations
at pH 4.00 in 0.01 M NaNO
3
. The values in bracket are the surface coverage of
MMA without phosphate. Goethite concentration is 1.2 g L
-1
. 126
Figure 4.60 The relationship between the surface coverage of phosphate to MMA on
reversible sites and the equilibrium concentration of phosphate to MMA at pH

4.00 in 0.01 M NaNO
3
. Goethite concentration is 1.2 g L
-1
. 127
Figure 4.61 Surface coverages of phosphate (open symbols) and arsenate (filled symbols) as
a function of initial As/P molar ratios with three initial phosphate concentrations
at pH 4.00 in 0.01 M NaNO
3
. The values in bracket are the surface coverage of
phosphate without arsenate. Goethite concentration is 1.2 g L
-1
. 129
Figure 4.62 Surface coverages of phosphate (open symbols) and arsenate (filled symbols) as
a function of initial P/As molar ratios with three initial arsenate concentrations at
pH 4.00 in 0.01 M NaNO
3
. The values in bracket are the surface coverage of
phosphate without arsenate. Goethite concentration is 1.2 g L
-1
. 129
Figure 4.63 The relationship between the surface coverage of phosphate to arsenate on
reversible sites and the equilibrium concentration of phosphate to arsenate at pH
4.00 in 0.01 M NaNO
3
. Goethite concentration is 1.2 g L
-1
. 130

xii
LIST OF TABLES

Table 2.1 The density of singly, doubly, and triply coordinated oxygens on three crystal
faces 10

Table 3.1 The effect of background electrolyte concentrations on the analysis of arsenic by
ICP-OES 29

Table 4.1 The mean percent deviation and standard deviation of the uptake for Arsenate,
MMA and DMA 44

Table 4.2 Desorbility of three arsenic species using different NaOH concentrations
*
55
Table 4.3 Comparison the surface coverages at pH 4.00 determined by three methods
*
55
Table 4.4 The mean percent deviation and standard deviation of surface coverages for
arsenate, MMA and DMA 56

Table 4.5 Parameters of Langmuir and Freundlich equation determined from linear plots 58
Table 4.6 The mean percent deviation and standard deviation of surface coverages of
arsenate, MMA and DMA, zeta potential and effective particle sizes 65

Table 4.7 The effective particle sizes of pure goethite at different pH values and NaNO
3

concentrations. Goethite concentration is 0.25 g L
-1

. 69
Table 4.8 Parameters of the Elovich equation for DMA, MMA, and arsenate at pH 4.00 at
three background electrolyte concentrations 72

Table 4.9 Linear fitting results of the three arsenic species adsorption on goethite at pH
4.00 at three NaNO
3
concentrations 77
Table 4.10 Linear fitting results of three arsenic species adsorption on goethite at different
pH values in 0.1 M NaNO
3
79
Table 4.11 The effective particle size (nm) of pure goethite using two mixing methods in
0.001 M NaNO
3
. Goethite concentration is 0.25 g L
-1
. 82
Table 4.12 A comparison of four surface complexation models 88
Table 4.13 Chosen modeling parameters 92
Table 4.14 Mean percent deviation of the replicates and standard deviation of proton-anion
adsorption ratios. 97

Table 4.15 Proton-DMA adsorption data at pH 4.25
a
98


xiii
Table 4.16 Proton-DMA adsorption data at pH 6.75

a
99
Table 4.17 Proton-MMA adsorption data at pH 4.00
a
103
Table 4.18 Proton-MMA adsorption data at pH 6.77
a
104
Table 4.19 Proton-arsenate adsorption data at pH 4.00
a
107
Table 4.20 Proton-arsenate adsorption data at pH 6.75
a
108
Table 4.21 Mean percent deviation of the replicates and standard deviation for surface
coverages of the three arsenic species and phosphate. 112

Table 4.22 The individual surface coverage of DMA and phosphate in three systems at
pH4.00
*
114
Table 4.23 Equilibrium concentrations and surface coverage ratios of phosphate to DMA in
DMA+P system at pH4.00
*
115
Table 4.24 Equilibrium concentrations and surface coverage ratios of phosphate to DMA in
P/DMA system at pH4.00
*
115
Table 4.25 Equilibrium concentrations and surface coverage ratios of phosphate to DMA in

DMA/P system at pH4.00
*
115
Table 4.26 The individual surface coverage of MMA and phosphate in three systems at
pH4.00
*
117
Table 4.27 Equilibrium concentrations and surface coverage ratios of MMA to phosphate in
MMA+P system at pH4.00
*
119
Table 4.28 Equilibrium concentrations and surface coverage ratios of MMA to phosphate in
P/MMA system at pH4.00
*
119
Table 4.29 Equilibrium concentrations and surface coverage ratios of MMA to phosphate in
MMA/P system at pH4.00
*
119



xiv
LIST OF SYMBOLS

C
Concentration of acid or base used in acid-base titration (µM) or constant
capacity in equation 4.45 (F m
-2
)

C
A,s

Concentration of stock arsenic solution (µM)
C
A,e


Equilibrium concentration of arsenic

(µM)

C
0

Initial concentration (µM)
C
d

Concentration of arsenic in desorption solution (µM)
C
e

Equilibrium concentration (µM)
C
H

Concentration of acid used in back titration (µM)
C
s

Solids concentration of goethite suspension (g L
-1
)
C
1

Constant capacitance (F m
-2
)
C
2

Constant capacitance (F m
-2
)
f


Activity coefficient
F
Faraday constant (96493 C mol
-1
)
I
Ionic strength (M)
k
1/2

Partition coefficient

K
Intrinsic acidity constants in 1-pK model or parameter of Langmuir or
Freundlich equation
K
1

Intrinsic acidity constants in 2-pK model
K
2

Intrinsic acidity constants in 2-pK model
K
A

Stability constants for anion pair formation
K
C

Stability constants for cation pair formation


xv
K
DMA,int

Equilibrium constant
K
DMA

Acid constant

n
Parameter in the Freundlich equation
N
x

Total surface area of goethite per unit of solution volume (m
2
L
-1
)
N
s

Density of surface sites (C m
-2
)
m
Mass of goethite (g)
R
Gas constant (8.314 J mol K
-1
)
s
Specific surface area (m
2
g
-1
)
t
Time (hours)

T
Temperature (K)
V
Volume of goethite suspension (L)
V
1

Volume of acid used in back titration (L)
V
2

Volume of arsenic stock solution added to goethite suspension (L)
z
Charge of ions
α
Parameter of Elovich equation or the protonation degree of arsenic
β
Parameter of Elovich equation or the protonation degree of arsenic
γ
Protonation degree of singly coordinated groups in 2-pK model
δ
Protonation degree of adsorbed arsenate
λ
Protonation degree of singly coordinated groups in 1-pK model
ζ
Zeta potential (mV)
σ
0
Surface charge (C m
-2

)
σ
β
Charge at IHP (C m
-2
)
σ
d
Charge at OHP (C m
-2
)


xvi
ψ
0

Surface potential (V)
ψ
β

Potential at IHP (V)
ψ
d

Potential at OHP (V)
Γ
Surface coverage (µmol m
-2
)

Γ
max

Maximum surface coverage (µmol m
-2
)



xvii
NOMENCLATURE

AFM Atom Force Microscopy
ATR-FTIR Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy
BSM Basic Stern Model
CCM Constant Capacity Model
CD Charge Distribution
CIP Common Intercept Point
DLM Diffusion Layer Model
DMA Dimethylarsinate
EXAFS Extended X-Ray Absorption Fine Structure
FESEM Field Emission Scanning Electron Microscopy
FLM Four Layer Model
FTIR Fourier Transform Infrared Spectroscopy
ICP-OES Inductively Coupled Plasma - Optical Emission Spectroscopy
ICP-MS Inductively Coupled Plasma - Mass Spectrometry
IEP Isoelectric Point
IHP Inner Helmholtz Plane
IUPAC International Union of Pure and Applied Chemistry
MMA Monomethylarsonate

MUSIC Multi Site Complexation
OHP Outer Helmholtz Plane
PZC Point of Zero Charge


xviii
PSD Pore Size Distribution
SCM Surface Complexation Model
SEM Scanning Electron Microscopy
TEM Transmission Electron Microscopy
TLM Triple Layer Model
XRD X-ray Diffraction
Chapter 1 introduction

1
Chapter 1 Introduction

Adsorption of arsenic on the mineral surface is an important process that influences
the potential bioavailability of arsenic in natural systems. A wide variety of minerals,
especially aluminum oxides and ferric (hydr)oxides, have a high adsorption capacity for
arsenic (Jain et al., 1999; Ladeira et al., 2001; Dixit and Hering, 2003; Arai et al., 2004).
Goethite (α-FeOOH) is one of the most abundant ferric (hydr)oxides in natural systems
(Cornell and Schwertmann, 2003) and has been widely used in adsorption studies. Goethite
has a well-characterized crystal structure and can be easily synthesized in the laboratory.
Unfortunately, many aspects of arsenic adsorption are not clearly understood, which limits
our ability to model arsenic movement in the environment.

1.1 Arsenic in natural systems
Arsenic is widespread in aquatic environments, soils, and sediments due to natural and
anthropogenic sources, generally at relatively low concentrations. Natural processes such as

erosion and weathering of crustal rocks lead to breakdown and translocation of arsenic from
the primary sulfide minerals. An extensive range of anthropogenic sources, such as mining
and metallurgy, wood preservation, urban and industrial waste disposal, and application of
sewage sludge and fertilizer, may also increase arsenic concentrations in natural systems.

Arsenic exists in several different species in natural water systems, depending on the
pH and the redoxida potential. Most common are the arsenic oxyanions, notably arsenite
(As(III)) and arsenate (As(V)), and the methylated arsenic forms of monomethylarsonate
Chapter 1 introduction

2
(MMA, As(V)), and dimethylarsinate (DMA As(V)), these two species are being found
with increasing frequency in natural water systems (Del Razo et al., 1990; Anderson and
Bruland, 1991; Chen et al., 1995; Banerjee et al., 1999). Arsenic (As) has been known to be
highly toxic to humans for a long time. Arsenic interferes with enzyme action, DNA
transcription, and metabolism, so its effects on the human body are variable. These include
dermatological, cardiovascular, neurological and cancinogenic effects (Mandal and Suzuki,
2002). Arsenic enters human bodies predominantly through drinking water. Drinking water
is derived from a variety of sources such as surface water (river, lakes, reservoirs and ponds),
ground water (aquifer) and rainwater. Alongside point sources of arsenic contamination,
high concentrations are mainly found in groundwaters. The range of arsenic concentrations
found in natural waters is unusually large, ranging from less than 0.5 µg
L
-1
to more than
5000 µg L
-1
. Based on the accumulation of evidence for the chronic toxicological effects of
As in drinking water, the WHO guideline value for As in drinking water is l0 µg
L

-1
. At
present, Several million people worldwide are drinking water with arsenic concentrations
above drinking water standards, most notably in the Bengal basin of India and Bangladesh
(Karim, 2000; Watanabe et al., 2004).

In an oxidized system, the solubility of arsenic in aquatic environments is mainly
controlled by the interactions between arsenic and mineral surfaces. Among these
interactions, adsorption of arsenic on the surface of minerals is one of the most important
processes. Adsorption of inorganic arsenic on aluminum oxides and ferric (hydr)oxides
has been studied for many years (Malotky and Anderson, 1976; Gupta and Chen, 1978;
Hingston, 1981; Xu et al, 1991; Bowell, 1994; Cox and Ghosh, 1994; Jaskson and Miller,
2000; Goldberg and Johnston, 2001; Arai et al., 2004). At present there are scare studies
Chapter 1 introduction

3
on the adsorption of DMA and MMA on goethite (16. Bowell, 1994; Xu et al., 1991;
Lafferty and Loppert, 2005). Iron oxides and oxyhydroxides such as hematite and goethite
show a strong affinity for arsenic and numerous studies have quantified and modeled
arsenic adsorption on goethite (Cornell and Schwertmann, 2003).

1.2 Arsenic adsorption on goethite
Adsorption data is frequently presented as a plot of the amount of adsorbate per gram
or unit surface area of adsorbent vs. the equilibrium concentration in solution at a constant
pH, solids concentration, and background electrolyte concentration. The adsorption data is
usually fitted to an adsorption isotherm such as the Langmuir and Freundlich equation
(Stumm, 1992; Sparks, 2003; Matis et al., 1999; Liu et al., 2001). These equations are useful
for summarizing and comparing equilibrium data. Although both equations can be used to
predict adsorption behavior under conditions other than those of actual experiment, they
provide little information about either the adsorption mechanism or the molecular structure

of the adsorbed species and such predictions are tenuous. More information is available
from a variety of surface complexation models developed in recent years (Dzombak and
Morel, 1990; Lutzenkirchen, 2002).

Adsorption kinetics of arsenic or other strongly bond anions (e.g. phosphate) on
goethite have been less studied than the “equilibrium” conditions. However, adsorption
kinetics can provide insight into the mechanism of arsenic adsorption on goethite and the
nature of the goethite surface. A two-phase kinetics pattern, with an rapid initial stage
followed by a slow stage, has been observed for arsenate adsorption on ferric (hydr)oxides
(Barrow, 1983; Anderson et al., 1985; Strauss et al., 1997; Raven et al., 1998; O'Reilly et al.,
Chapter 1 introduction

4
2001; Zhao and Stanforth, 2001; Waltham and Eick, 2002). Two factors have been
suggested to control the slow stage. One is diffusion, such as inter-particle or intra-particle
diffusion, and the other is heterogeneity of surface reactions, such as surface precipitation,
rearrangement of the surface complex, or heterogeneity of the surface sites. The diffusion
process is dependent on the effective (or aggregate) particle sizes, porosity, or crystal
structure of ferric (hydr)oxides, whereas surface reactions are related to the surface energy
or the transition from one type of surface complex to another. Both reaction-control and
diffusion-control have been reported for the slow stage of phosphate adsorption on goethite
(Strauss et al., 1997; Zhao and Stanforth, 2001). However, the slow stage of arsenic
adsorption on goethite is still not clearly understood. Few studies have focused on the effect
of the effective particle sizes on the slow stage of adsorption kinetics and desorption for
arsenic interaction with goethite.

1.3 Objectives and scope
The main goals of this study were to investigate the acid-base properties of surface
groups on the goethite surface and the adsorption mechanism of two methylated arsenicals
on goethite by comparing the adsorption equilibrium and kinetics of arsenate, MMA, and

DMA on goethite at the same experimental conditions. The objectives of this study are
given below:
1. Effect of methyl groups in arsenic compounds on its affinity for goethite. DMA
(pK=6.27) (CRC Handbook of Chemistry and Physics, 2004) and MMA (pK
1
=3.6,
pK
2
=8.2) (Gettar et al., 2000) have a similar molecule structure and size with arsenate
(pK
1
=2.20, pK
2
=6.97, pK
3
=11.53) (Burrow and Turner, 1920) except for two or one
Chapter 1 introduction

5
hydroxyl groups replaced by methyl groups. There is only one hydroxyl group in
DMA and two in MMA, thus the adsorption behavior of DMA and MMA may
different from that of arsenate which bonds to the goethite surface by inner-sphere
complexes (Fuller et al., 1993; Waychunas et al., 1993; Hiemstra and Van Riemsdijk,
1996; Manning and Goldberg, 1996; Sun and Doner, 1996; Fendorf et al., 1997;
Grossl et al., 1997)
2. The types of heterogeneity exist on the goethite surface. There are potentially several
different types of heterogeneity on the goethite surface, e.g. (i) the presence of various
reactive crystallographic faces, (ii) defects, surface relaxation, and different
coordinated hydroxyls on individual faces, (iii) different bonding sites on individual
faces, and (iv) adsorption of one ion altering the bonding energy for the next ion.

3. The factor which is responsible for the slow stage of arsenic adsorption on goethite.
Effective particle sizes, crystal sizes and structure influence the diffusion process.
Studying the effect of these factors on the slow stage will differentiate diffusion-
controlled and reaction-controlled process. Desorption of arsenic after increasing the
pH of system will provide supplementary information on the slow stage.
4. The model that can interpret both proton-anion adsorption ratios and the effect of
adsorption of arsenic on zeta potential. Either a 2-pK or a two-site 1-pK model can
describe the charging behavior of the goethite surface well, but the predicted degree of
protonation of singly coordinated hydroxyls is different in the two models. The singly
coordinated hydroxyls are assumed to be reactive sites for anions, thus these two
models may give different proton-anion adsorption ratios.