Citation
Utilization of University of Florida flux meter for estimating arsenic contamination in groundwater

Material Information

Title:
Utilization of University of Florida flux meter for estimating arsenic contamination in groundwater
Creator:
Gupta, Prachee ( Dissertant )
Clark, Clayton J. ( Thesis advisor )
Hatfield, Kirk ( Thesis advisor )
Annable, Michael ( Reviewer )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Copyright Date:
2004
Language:
English
Physical Description:
ix, 43 p. ; ill. , tables

Subjects

Subjects / Keywords:
Civil and Coastal Engineering thesis, M.S.
Dissertations, Academic -- UF -- Civil and Coastal Engineering
Arsenic ( jstor )
Adsorption ( jstor )
Groundwater ( jstor )

Notes

Abstract:
Groundwater contamination is a major problem in today's environment. In Florida, the ground water standards are equivalent to the drinking water standards according to the Florida Department of Environmental Protection (FDEP). Among the various contaminants, organic, inorganic, microbial pathogens and radioactive contaminant, inorganic contaminants are of high interest because most of these contaminants are readily soluble in water, and have a high potential to contaminate groundwater. The present research focused on the groundwater contamination due to arsenic. The maximum permissible concentration of arsenic in drinking water in United States is 10microg l⁻± or 10 ppb as recommended by United States Environmental Protection Agency. The purpose of the research was to evaluate the use of the University of Florida flux meter for estimating arsenic contamination in the groundwater. The flux meters were placed in three wells located perpendicular to the flow lines in the center of model aquifer. The internal adsorbent chosen was activated alumina due to its high efficiency in removal of arsenic. Adsorption of arsenic on activated alumina showed a linear trend with a K sub d value of 1.844 L kg⁻±. The procedure for extracting arsenic adsorbed by activated alumina was evaluated and it was found that extraction efficiency of acid (88%) is comparatively more than that of base (84%). Flux meter tests were conducted for arsenic concentration of 12 ppm at a flow rate of 0.84 cm/hr. The flux meter in the central well estimated 99% of the actual arsenic concentration in aquifer Therefore this provides an efficient and cost effective way to estimate arsenic flux in the field.
General Note:
Title from title page of source document.
General Note:
Document formatted into pages; contains 52 pages.
General Note:
Includes vita.
Thesis:
Thesis (M.S.)--University of Florida, 2004.
Bibliography:
Includes bibliographical references.
Original Version:
Text (Electronic thesis) in PDF format.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Gupta, Prachee. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
8/7/2004
Resource Identifier:
56884647 ( OCLC )

Downloads

This item has the following downloads:


Full Text












UTILIZATION OF UNIVERSITY OF FLORIDA FLUX METER FOR ESTIMATING
ARSENIC CONTAMINATION IN GROUNDWATER















By

PRACHEE GUPTA


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Prachee Gupta
















ACKNOWLEDGMENTS

I would like to express my sincere gratitude to Dr. Clayton J. Clark II, chairperson

of my supervisory committee, for all his guidance and support throughout the project. I

am very grateful to my committee, members, Dr. Kirk Hatfield and Dr. Michael Annable,

for all their valuable guidance, suggestions and encouragement. I would also like to thank

Dr. Jaehyun Cho, Dr. Mark Newman and Mr. Harald Klammler for constant guidance in

performing laboratory experiments and Dr. Tait Cherenji for providing basic data for

computer model simulation.

I greatly appreciate the help of Mr. Thomas Luongo for sample analysis. Thanks go

to the students of Water Resource Research Center for the support I received throughout.

Finally, I want to thank my parents and friends who encouraged me throughout my

academic career.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES ....................................................... ............ ....... ....... vi

L IST O F F IG U R E S .... ...... ................................................ .. .. ..... .............. vii

A B STR A C T ..................... ................................... ........... ................. viii

CHAPTER

1 IN TRODU CTION ............................................... ..... ..... .............. ..

G e n e ra l .......................................................................................................1
Purpose and O bjectives............................................................... ....................... 3

2 THEORETICAL BACKGROUND ............................... ................................. 4

Advective-Dispersive Transport of Reactive Solutes...........................................4.
Linear Isotherm ....................... .................... ... ...... .........4
N on-linear (Freundlich Isotherm ) ........................................... .....................5
N on-linear (Langm uir Isotherm ).................................... ........................ 6
U university of Florida Flux M eter................................................... ... ................... 6
Estimating Fluid Flux ............................. .. .......... ....... ...... ...............7
Estimating Solute Mass Flux........................................ ...............9

3 LITER A TU RE REV IEW ...................... .... ................................. ............... 12

O occurrence of A rsenic ............... .......... .............. .............. .. ..................... ..... 12
Fate and Transport of A rsenic .............. ..................................... ............... 13
Adsorption Characteristics of Arsenic ....... .......... ................... ..............15
L laboratory A naly sis........... ...... ...................................................... .......... .. 18
A d sorption Isoth erm s ........................................ ....................................... 18
Extraction and A analysis: .............. ................................... .... ........ 19
F lux M eter A naly sis ........................................................ .... ...... .... .. ... 19
Com puter M odel Sim ulation.......................................................... ............... 20










4 MATERIALS AND METHODS ........................................ ........................ 22

A dsorption Isotherm s........................................................................... ............... 22
E extraction M ethod .......................................... .. .. ......... .... .. ... 23
A analysis M ethod ............... ................. ....................... ............ 24
F lu x M ete r ........................................................................................................... 2 4
F lux M eter Set-up ..................... .. ...................... .. .. ...... ........... 25
R running E xperim ent ............................................... ............................ 26

5 RESULTS AND DISCUSSION ........................................ ......................... 29

A dsorption E xperim ents ............................................... ......................................29
Extraction Experim ents....................................................................................... 31
Flux M eter Bench Scale Experim ents.................................... ....................... 33
Computer Simulation Results ...........................................................................35

6 CONCLUSIONS .........................................................................37

L IST O F R E F E R E N C E S ....................................................................... ... ................... 39

B IO G R A PH IC A L SK E TCH ..................................................................... ..................43

































v
















LIST OF TABLES

Table p

3-1 Behavior of arsenic with different m inerals ........................................ .................17

5-1 Adsorption of arsenic on activated alum ina ........................................ .................30

5-2 Arsenic extracted using nitric acid ...............................................31

5-3 Arsenic extracted using sodium hydroxide ................................... .................32

5-4 Concentrations measured from flux meter tests............... ......... ............... 34

5-5 Soil properties ........................................................................35






















LIST OF FIGURES


Figure p

2-1 Contaminant sorbed by the sorbing matrix ................................... .................10

3-1 Cumulative probability curve for "As" in Florida surface soils............................. 14

3-2 Eh-pH diagram for inorganic arsenic compounds........................................15

4-1 Bench Scale set-up of flux meter............. ...... ..................................... ....... 27

4-2 Cross sectional dimensions of flux meter ...................... ...................... 27

5-1 Adsorption isotherm of arsenic on activated alumina.................... ..................30

5-2 Amount extracted using acid vs. initial concentration ..........................................32

5-3 Amount extracted using base vs. initial concentration........................... ..........32

5-4 Concentration of arsenic vs. time for varying soil type. .........................................35














Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
UTILIZATION OF UNIVERSITY OF FLORIDA FLUX METER FOR ESTIMATING
ARSENIC CONTAMINATION IN GROUNDWATER

By

Prachee Gutpa

August 2004

Chair: Clayton J. Clark II
Major Department: Civil and Coastal Engineering

Groundwater contamination is a major problem in today's environment. In Florida,

the ground water standards are equivalent to the drinking water standards according to

the Florida Department of Environmental Protection (FDEP). Among the various

contaminants, organic, inorganic, microbial pathogens and radioactive contaminant,

inorganic contaminants are of high interest because most of these contaminants are

readily soluble in water, and have a high potential to contaminate groundwater. The

present research focused on the groundwater contamination due to arsenic. The maximum

permissible concentration of arsenic in drinking water in United States is 10g 1-1 or 10

ppb as recommended by United States Environmental Protection Agency. The purpose of

the research was to evaluate the use of the University of Florida flux meter for estimating

arsenic contamination in the groundwater. The flux meters were placed in three wells

located perpendicular to the flow lines in the center of model aquifer. The internal

adsorbent chosen was activated alumina due to its high efficiency in removal of arsenic.

Adsorption of arsenic on activated alumina showed a linear trend with a Kd value of









1.844 L kg1. The procedure for extracting arsenic adsorbed by activated alumina was

evaluated and it was found that extraction efficiency of acid (88%) is comparatively more

than that of base (84%). Flux meter tests were conducted for arsenic concentration of 12

ppm at a flow rate of 0.84 cm/hr. The flux meter in the central well estimated 99% of the

actual arsenic concentration in aquifer Therefore this provides an efficient and cost

effective way to estimate arsenic flux in the field.














CHAPTER 1
INTRODUCTION

General

Groundwater contamination is a major problem in today's environment since

groundwater constitutes major portion of potable water in Florida. In Florida, the ground

water standards are equivalent to the drinking water standards, as stated by Florida

Department of Environmental Protection (FDEP). Groundwater can be contaminated by

disease-producing pathogens, leachate from landfills and septic systems, careless disposal

of hazardous household products, agricultural chemicals, and leaking underground

storage tanks (Ginn et al., 2002; Day et al., 2001; Leeuwen, 2000). There are four basic

types of contaminants that include organic, inorganic, radioactive elements, and

microbial pathogens (Salas and Ayora 2004; Jones and Huang, 2003; Barry et al., 2002;

Ginn et al., 2002). Among the various contaminants, inorganic contaminants are of high

interest because most of these contaminants are readily soluble in water, and have a high

potential to contaminate groundwater.

Relative to other oxyanion forming minerals, arsenic is problematic because of its

mobility in most of its oxidations states and at almost all pH values. During 2001,

roughly 96% of the arsenic imported into the United States was in the form of arsenic tri-

oxide (Solo-Gabriele et al., 2002). Most of this arsenic trioxide was used for the

production of arsenic acid used for wood preservation. Arsenic is brought into the State

of Florida for the production of chromated copper arsenate (CCA) treated wood (Chirenje

et al., 2003 and Solo-Gabriele and Townsend, 1999). CCA is composed of three major









elemental constituents, copper (Cu), chromium (Cr), and arsenic (As), among which

arsenic is the most hazardous. The present research focused on the soil and groundwater

contamination due to arsenic. Arsenic is a semi-metal element, found in environment

combined with other elements such as oxygen, chlorine, and sulfur to form inorganic

arsenic compounds. It also combines with carbon and hydrogen to form organic

compounds. It most often exists in organic form as monomethyl arsonic acid (MMA) and

dimethyl arsonic acid (DMA). Arsenic exists in the environment most often in two

oxidation states: arsenite, As (III), and arsenate, As (V), respectively (Singh and Pant

2003; Meng et al., 2002). Under oxidation conditions (and aerobic environments)

arsenates are stable species (as oxyanions H3AsO4, H2AsO4-, HAsO4-2 or AsO4-3) and are

strongly sorbed onto clays, iron, and manganese oxides and hydroxides, and organic

matters. However, under most reducing conditions (and anaerobic environment) arsenites

(H3AsO3, H2AsO3-1, HAsO3-2) are the predominant arsenic compounds (Chakravarty et

al., 2002).

According to Kayajanian (2003), studies on Utah cancer mortality as a function of

lifetime arsenic exposure indicated that for concentration range of 1-25 ppb, there are

2.682 cancers per 100 people and in the 5 years following initial arsenic medication,

around 6.45 deaths are expected. The maximum permissible concentration of arsenic in

drinking water in United States is 10g L-1 or 10 ppb as recommended by United States

Environmental Protection Agency, USEPA (2001). In addition to the arsenic

contamination in water sources, dietary intake of arsenic through the food chain via

uptake from contaminated soils may adversely affect human health (Alam et al., 2003).

The current soils clean up goals set by the Florida department of environmental









protection for arsenic in residential and industrial soils are 0.8 mg/kg and 3.6 mg/kg

respectively (Solo-Gabriele et al., 2002).

Various scientists and organizations have made an attempt to identify the source

of contaminant and degree of contamination in groundwater using different field

monitoring techniques (Montas et al., 2000; Brusseau and Srivastava, 1997). This

research discusses a new devise named Flux meter that allows to simultaneously

measuring cumulative dissolved solute fluxes when placed within a flow field. The Flux

meters have been tested for most of the organic contaminant analysis both in lab as well

as field for contaminant flux, groundwater flow and flow direction (Hatfield et al., 2003;

Klammler 2001).

Purpose and Objectives

The purpose of this research was to evaluate the use of flux meter for estimating

arsenic contamination in the groundwater. Before bench scale flux meter tests are

performed, information regarding the type of adsorbents and its interaction with arsenic

should be understood very well. Hence, the objectives of this research were 1) review of

adsorption and extraction characteristics of arsenic to different materials and 2) to

evaluate the applicability of these materials as flux meter sorbent media for arsenic flux

measurement, 3) to simulate the arsenic contamination for various Florida soil types. This

thesis seeks to give the basic understanding of the solute transport and theoretical

contemplation of flux meter for estimating fluid and contaminant flux, arsenic

contamination and need of flux meter for estimating fluid flow and contaminant flux in

the groundwater.














CHAPTER 2
THEORETICAL BACKGROUND

Advective-Dispersive Transport of Reactive Solutes

In non-ideal condition found in the environment, a reactive solute interacts with the

solid matrix during flow. The solute is, therefore distributed between the fluid and the

sorbed phases. If a primary anisotropic system is assumed, for the steady water flow,

under saturated or uniformly unsaturated conditions, the solute transport equation for a

homogeneous media can be represented by (Domenico and Schwartz, 1990)

Sac aS 82C ac (2.
Ow +Pb a =wD, -q a (2.1)
at at ax Qt

where
Ow= volumetric water content
Cw= solute concentration in water (kg L-1)
pb= bulk density of sorbing media (g/cm3)
S= sorbed solute concentration (mg/kg)
Dw = dispersion coeffecient (cm2/hr)
qw = advective mass flux (cm/hr)


Three most commonly known isotherms that define the adsorption characteristics of

any adsorbent are linear isotherm, Freundlich isotherm and Langmuir isotherm (Zheng

and Bennett 2002, Charbeneau 2000). Freundlich and Langmuir isotherms can be used to

represent both linear and non-linear isotherms.

Linear Isotherm

When the sorbed phase solute concentration is linearly proportional to the solute

concentration in the solution phase, the sorption isotherm is linear isotherm, and is

represented by S = Kd.Cw. Therefore, the equation 2.1 is transformed to









aC + pbK 2C Oc
a [1+ ] = D (2.2)
at a ax2 at

The transport velocity (cm/hr) is represented by v. The adsorption coefficient (Kd)

is a measure of how tightly the solute binds or sticks to soil particles. The greater the Kd

value, the less likely a chemical will leach or contribute to runoff.

The retardation factor, R = (1+ pKd/Ow), expresses how much slower a

contaminant moves than does the water itself. Equation 2.2 is therefore transformed to

ac 0 2C ac
a = D*. d2 v (2.3)
at a2 at

D v
where D* and v =
R R

When Kd = 0, it implies that R = 1 and hence D* = D and v* = v, which reverts

the equation (equation 2.3) back to the non-reactive ideal solutes. Hence the equations

below show the effect of retardation factor on the solute transport. As Kd and R increases,

the distance to solute peak decreases. The distance to solute peak (x*) is given by

vt
x* = [v*t] (2.4)
R

The time of arrival of solute peak (t*) is delayed and is given by

LR
t* = [L/v*] = LR (2.5)
V

Non-linear (Freundlich isotherm)

In this type of isotherm the solute mass is proportional to the concentration raise

to N.


S K C N #


(2.6)









pKNCN
Hence, retardation factor R = [1 + P w ]


It is evident from the above equation that R is not a constant but is a function of Cw.

N<1, R (Cw) decreases as Cw increases for Cw >1

N>1, R (Cw) increases as Cw increases for Cw >1

N = 1, R is independent of Cw and hence leads to the linear isotherm.

Non-linear (Langmuir Isotherm)

Langmuir isotherm is the most common type non-linear isotherm and in this

isotherm


S = maxkw (2.7)
1+ kC

pS k
Hence, retardation factor for Langmuir isotherm is R(C,) = [1+ PSmax
O, (1 + kC)2

In non-ideal environmental conditions, it is very difficult to obtain the accurate

values of various parameters and hence accurate flux measurement is difficult. There is a

need for accurate and easy to use equipment for the calculation of groundwater flux and

contaminant flux. A new method utilizing flux meter has been developed at the

University of Florida and laboratory tested for measuring both contaminant and

groundwater fluxes at hazardous waste site.

University of Florida Flux Meter

The University of Florida flux meter measures cumulative dissolved solute fluxes

with directions simultaneously when placed within a flow field (Hatfield et al., 2003).

The device consists of a self-contained permeable unit that intercepts the groundwater

flow without retaining it. A matrix of hydrophobic and hydrophilic sorbents in the device









sorbs dissolved organic and inorganic solutes present in the water intercepting the unit,

thus indicating the amount of contaminant carried by the groundwater. The sorbent

matrix is also impregnated with known amount of one or more fluid soluble "resident

tracers" that leaches from the sorbent at the rate proportional to the fluid flux (Hatfield et

al. 2003) and provides an estimate for the fluid flow. The flux meter has been validated

and used for estimating fluid flow and contaminant flux for various organic

contaminants. This research seeks to explore the use of flux meter for measuring arsenic

contamination. .

Estimating Fluid Flux

Fluid flow can be estimated by placing flux meter device in a monitoring well

perpendicularly intercepting the groundwater flow. The amount of tracer leaching out of

the device gives the measure of fluid flux. It is assumed that a) the various tracers used do

not mutually affect their partitioning properties; b) tracers partition isotherm is linear and,

c) tracer transport due to dispersion is negligible (Hatfield et al., 2003). Therefore the

distance that the tracer is pushed away by the water can be calculated as

vAt
Ax v,,tAt (2.8)


where vtr = tracer transport velocity.


If the water flows through the meter at the rate vtr = v/R, the remaining amount of

the tracer with respect to the initial tracer amount (represented by mR) gives the measure

of how much water has passed the unit. The rate of vtr is measured by vtr = v/R. Given the

cross sectional area of the unit, one can relate this concentration to the area still occupied

by the tracer as










A = 2r2 sinl(a)-aAx (2.9)
r


R = 2 (2.10)
)Tr


A = q 2r [v q= (2.11)
RO 0

where

a= [r2 -( )2 (2.12)

r = radius of the device

If 4, is the dimensionless cumulative volume of fluid conveyed through the

device, then equation 2.10 can also be stated as


mR [sin 1 2)- 2] (2.13)


qAt
= (2.14)
20Rr

However, due to the circular cross section only at the very beginning is the tracer

present over the whole width of the device (2r). As the tracer is desorbed, some water

will pass the device without leaching out any of the tracer. The distance "a" from the

center, this is equal to the radius of the section at the beginning. This also decreases with

growing Ax and the water passing the device at distances greater than "a" from the center

does not leach out tracer any more. Therefore, equation 2.13 must be used to describe the

relation between the relative remaining tracer mass mR and the dimensionless cumulative

volume 4. To simplify equation 2.13, Klammler (2001) performed regression analysis for

the variation of with mR and observed that regression equation is almost linear for less









than equal to 0.6, which corresponds to the mR equal to 0.3. The regression equation

obtained was

mR = -1.2 +1.0 (2.15)

After keeping the flux meter in the groundwater for a determined amount of time,

the amount of tracer left in the sorbent media (mR) is extracted. The dimensionless

cumulative volume of water, intercepted by the flux meter at a specified depth can be

obtained using equation 2.15. The specific discharge, q, can then be calculated using the

equation 2.14. The specific discharge calculated using equation 2.14 is the discharge

through the flux meter. Because the material used in the aquifer can be different from that

used in flux meter, the permeability (or hydraulic conductivity) of flux meter can be

different from permeability of the surrounding aquifer. The specific discharge in the well

and surrounding aquifer can be related as (Hatfield et al., 2003)

q 2k
q k (2.16)
q, k+k0

Where, q and qo are the specific discharge through the flux meter and surrounding

aquifer respectively. k is the hydraulic conductivity of flux meter and ko is the hydraulic

conductivity of aquifer.

Estimating Solute Mass Flux

The dotted area shown in Figure 2-1 is where the contaminant is sorbed in to the

device. The respective retardation factor of the contaminant (Ront) quantifies the fraction

of the cross sectional area of the device that is actually occupied by the sorbed

contaminant (Klammler 2001). Hence the actual concentration of contaminant in aqueous

phase is given by










C = Mco (2.17)
r 2L(1 mR,cont Rcont

where
Moont = mass of contaminant sorbed (mg)
r = radius of the flux meter cylinder (cm)
L = length of sorbent matrix (cm)
mR,ont = relative mass of hypothetical resident tracer retained after time period t.
Reont = retardation of sorbent (contaminant) on matrix (L g')
0 = water content in the device.
To calculate the amount of tracer retained in the flux meter (mR,cont) that is required

for contaminant flux calculations (equation 2.17), an imaginary tracer is assumed in the

flux meter that has the same leaching properties as that of contaminant (i.e. the tracer and

contaminant have the same retardation factor). Equation 2.13 and 2.14 can then be used

for calculating mR,cont using the parameters of sorbent matrix.




Service
Area of sorbed
contaminant



~~ ---------~-- --t------------ ----------







Figure 2-1 Contaminant sorbed by the sorbing matrix

Since the calculated concentration is same in the aqueous phase of the aquifer and

aqueous phase of the device, it can be directly used to calculate the contaminant flux

using


J = qo. C


(2.20)









Like any other process, flux meter also has a disadvantage in it. This method only

works for plume constituents that are actually retained by the sorbing matrix of the

device. For plume constituents that are not retained by the sorbing matrix, the

contaminant flux is not measured. Hence the prior knowledge of type of contaminant in

the area of interest is very important. Evaluating flux meter for monitoring arsenic

concentration and the various parameters needed to be considered while simulating field

conditions in lab are discussed in next chapter. Also, this method has not been validated

for wide scale use for measuring inorganic solutes.













CHAPTER 3
LITERATURE REVIEW

Arsenic predominantly exists in two oxidation states, As (III) and As (V). USEPA

2001 and FDEP 2004 have revised their maximum contamination level limits from 50ig

1-1 to 10g 1- starting from 2006 and 2005, respectively. In this study, arsenic

contamination due to chromated copper arsenate (CCA) leaching from construction and

demolition landfills is considered. Various factors that can affect the sorbed concentration

of arsenic are discussed and affect of varying soil types is also initiated.

Occurrence of Arsenic

Occurrence of arsenic can be both anthropogenic including mining and industrial

waste, and geogenic such as volcanic eruption and weathering, in nature. High

concentration of arsenic in ground water can occur in some areas as a result of inputs

from geothermal sources. Nimick et al. (1998) found up to 370kg 1- arsenic in Madison

River water as a result of geothermal inputs from the yellow-stone geothermal system.

Smedley and Kinniburgh (1996) noted high arsenic concentrations (around 200-300kg 1-

1) in surface waters affected by Sn and Au mining activities. Azcue and Nriagu (1995)

conducted experiments for arsenic concentrations in Moira lake of Canada throughout the

year and showed the significant seasonal difference with an average concentration of 62

.g 1- during summers and 22 .g 1- during winters, they attributed this difference to the

depleted oxygen levels in the bottom lake waters as a result of biological productivity

during summers.









Arsenic is widely used in making wood preservatives like CCA, insecticides and

pesticides and for various other agricultural and industrial purposes. Due to its

widespread use in agriculture, industry, and medicine, it has gained attention of lot of

scientists and researchers for its health effects and required remediation procedures.

Arsenic is brought into the State of Florida primarily for the production of CCA-treated

wood (Chirenje et al., 2003 and Solo-Gabriele and Townsend, 1999). Upon disposal, the

majority of CCA-treated wood is may be disposed in unlined construction and demolition

(C&D) landfills, or recycled as mulch or wood fuel. As a result, a considerable amount of

arsenic can be potentially released into the Florida environment. Townsend et al. (2000)

estimated that, in Florida, CCA-treated decks impact 10 000 ha or 108 m2 of soil, and

the amount of CCA-treated utility poles in use is 7x105 m3. Arsenic is predominant in +5

oxidation state in CCA, though as the pH changes the stable oxidation state for arsenic

varies (Solo Gabriele et al., 2002 Bull, 2000).

Fate and Transport of Arsenic

Relative to other oxyanion forming minerals, arsenic is problematic because it is

mobile in most of the oxidations states and at almost all pH values, whereas other

oxyanion like selenium are mobile as selenate (Se04-2) under oxidizing conditions but are

immobile under reducing conditions either due to the stronger adsorption of its reduced

form, selenite (SeO3-2), or due to its reduction to metal (Smedley and Kinniburgh, 1996).

Chen et al. (2002) collected over 448 samples to determine arsenic distribution in soils of

Florida. The soil order described by Chen et al. (1999) is used to classify different types

of soils including Alfisols (14%), Entisols (22%), Histosols (10%), Mollisols (4%),

Inseptisols (3%), Spodisols (28%) and Ultisols (19%). Clark et al. (2004) grouped these

soil types into three classes, class 1, 2, and 3 of arsenic retention indices. Class 1 includes










Marls and Histosols; these soils have the greatest retention and minimal leaching

capability of arsenic. Class 2 includes Entisols, Alfisols and Ultisols. These soils found to

have moderate retention with moderate potential for leaching. Class 3 have the least

retention and greatest risk for leaching and it include Spodosols. Chen et al. (2002)

observed the decrease in arsenic concentration with following trend in soil order:

Histosols> Inceptisols> Mollisols> Ultisols> Entisols, Alfisols> Spodosols. A cumulative

probability curve for Arsenic concentration in Florida surface soils was developed as

shown in Figure 3-1 (Chen et al., 2002), where x-axis plots a histogram of the data

categories and y-axis plots the frequency of the data in percentage. Numbers in

parentheses are sample sizes of individual soil orders.

100 .. .. .
-o-- Spodosols (122) .
---Entisols (107) I *
80 -o- Alfsols (60) pp .
Upper biislhn
S -- tisos (88) 4" concentration= 702 nag
kg (hen et al., 199)
OM- Moisols (15)
60 -- nceptisoh (10) Florida DEP soil cleanup
I o 3.70 n9l kg9
m Histosols (39) industrial land use)
~- : Florida DEP soil cleanup
40 poal = (A SO rLim
:. (Residential land usw)

a USEPA soil screening
20 level =0.40 Mg kg0'
& :(USEPA, 1996)



0.0 0.1 1.0 10.0 100.0
Arsenic Concentration (rmgkg)

Figure 3-1: Cumulative probability curve for "As" in Florida surface soils
(Chen et al. 2002)










Adsorption Characteristics of Arsenic

Redox potential (Eh), and pH are most often the controlling factors in mobility of

arsenic in the environment. Under oxidizing conditions, H2AsO4- is dominant, at low pH

(pH of 2 to 6); while at high pH (6 to 12), HAsO4-2 becomes dominant (Smedley and

Kinniburgh, 2002). The Eh-pH diagram for arsenic is given in Figure 3-2. Solubility of

arsenic in natural systems is strongly influenced by its capacity to be adsorbed by

different minerals present in soils. The interaction of arsenic with different minerals is

represented in the Table 3-1.


1200 0





400 i ,
E
uJI O .






I IO I I I I -- -]
-400




2 4 6 B 10 12 14
pH
Figure 3-2: Eh-pH diagram for inorganic arsenic compounds
(Smedley and Kinniburgh, 2002)

Among the treatment processes appropriate for removal of arsenic, activated

alumina adsorption is considered to be in-expensive and more versatile (Wang et al.,

2000). Activated alumina (AA) has high efficiency to adsorb arsenic in both of its

oxidation states. However, ionic strength, pH, competing ions, and temperature can

significantly change adsorption characteristics of activated alumina for arsenic (Deliyanni









et al., 2003; Meng et al., 2002; Lin and Wu, 2000; Wang et al., 2000; Halter and Pfiefer

2001). Lin and Wu (2000) found that arsenite uptake on activated alumina is much less

than arsenate uptake at almost all pH conditions. The uptake of arsenite was found to

increase with the increasing pH until pH 7, and then decrease as pH increases. This was

attributed to surface charge of AA, which is positive until pH < pHzpc (point of zero

charge). Deliyanni et al. (2003) observed the effect of ionic strength of solution on the

sorption process of arsenate ions on alkaganeite by varying the concentrations of (0-0.1M

KNO3). Observations indicated that as the concentration of KNO3 is increased, the

removal of arsenic is improved until saturation, when varying the amount of sorbent. This

improvement was attributed to certain depression of negative solid surfaces charges in the

alkaline region, caused by the presence of inorganic electrolyte, which enhanced the

interaction between surface sites and arsenic oxyanions. These adsorption characteristics

of arsenic highly differ in presence of intersecting minerals like phosphorous. British

Geological Survey (BGS 2001) calculated sorption of As (V) and As (III) by Hydrous

Ferric Oxide (HFO) as a function of arsenic concentrations and pH in 0.01M NaCl

background electrolytes. Observations were made in the absence and presence of an

equilibrium dissolved phosphate-P concentration of 1 mg 1-1. Results indicated that over

the pH range of 6 to 8, As (III) sorbs more strongly than As (V), the reverse was seen in

the absence of phosphate. Meng et al. (2002) investigated the combined effect of

phosphate, silicate and bicarbonate on the adsorption of arsenic by iron hydroxides. The

apparent adsorption constants indicated that the affinity of the anions for iron hydroxide

sites decreased in the order as arsenate > phosphate > arsenite > silicate > bicarbonate.









Table 3-1: Behavior of arsenic with different minerals
Mineral Comment Reference
Akaganeit The maximum load capacity was found to be about Deliyanni et. al.
(P-FeO(OH)) about 100-120 mg As(V) per g of Akaganeite (2003)
when .5g 1-1 used. The amount of arsenate
adsorption increases by lowering the pH, increasing
the amount of sorbent and ionic strength of the system.
Clay mineral An increased Arsenic content has been noted with the Galba and
increasing clay content of the contaminated soils. Polacek, (1973)
Hametite & The effect of solute concentration, detention time, pH Singh et. al.
Feldspar temperature, and agitation rate of the adsorbent like (1996)
Hematite and Felspar. The maximum removal was found
to be 100% and 97% with hematite and felspar
respectively at optimum conditions.
Orange Juice Found that for iron-loaded phosphorylated OJR, maximum Ghimire et. al.
Residue (OJR) adsorption capacity for As(V) and As(III) was (2002)
0.94 and 0.91 mol/kg at their optimum pH values 3.1
And 10.0, respectively
Activated Uptake of arsenite is much less than that of arsenate Lin and Wu
Alumina for AA in most pH conditions. Pore diffusion coefficients (2000)
(AA) And tortuosity factors of arsenate and arsenite were
interpreted for AA grains with different grain size and
different pH.
HFO Discussed kinetics and pH dependence of As(V) and As(III) Raven et al.
adsorption on HFO Found very high (1998)
As(V) and As(III) loadings (up to 4-5 mol As kg-1)
at the highest concentrations.
HFO Sorption of As(V) and As(III) on HFO at As Wilkie and
concentration of environmental significance (low Hering (1996)
micromolar range) and pH 4-9. SO4 decreased adsorption
Of As(V) and As(III), especially at low pH, while Ca
Increased As(V) adsorption at high pH. ImM bicarbonate did
Snot act on either As(V) or As(III) adsorption greatly









Phosphate, silicate and bicarbonate decrease the removal of As (III) even at

relatively low concentrations and low surface site coverage (Meng et al., 2002). Effect of

temperature as discussed by Pattanayak et al. (2000) showed that higher temperatures

could lead to lower metal uptakes.

Another easily available adsorbent considered by researchers for arsenic adsorption

is activated carbon. Lorenzen et al., (1995) and Pattanayak et al., (2000) used activated

carbon and its derivatives with different ash contents for arsenic adsorption. Lorenzen et

al., (1995) used peat-based carbon with 5-6% ash content and coconut shell carbon with

3% ash content for arsenic adsorption. Peat based carbon was found to perform better that

coconut shell carbon, implying that carbon with higher ash content can perform better.

Activated Carbon was found to be less effective and efficient for removal of Arsenic than

other adsorbents like activated alumina and iron hydroxide.

Laboratory Analysis

The two adsorbents considered in this research for use of flux meter were activated

alumina and activated carbon. These two adsorbents were chosen because activated

alumina has a very high efficiency for arsenic adsorption and activated carbon is one of

the most commonly used adsorbent for most of the adsorption based remediation

procedures.

Adsorption Isotherms

Adsorption isotherms define the effectiveness of adsorbents (activated alumina and

activated carbon) for the adsorption of arsenic. In this research batch experiments were

performed for adsorption isotherm. Most of these adsorption experiments were performed

at constant temperature and controlled pH.









Extraction and Analysis:

Arsenic extraction can be achieved by using strong acidic or alkaline solutions.

Lorenzen et al., (1995) used copper pretreated peat based carbon with a 100-ppm arsenic

solution for 24 hrs at a pH of 6 and checked the elution with distilled water, acids and

alkalis of different pH at different temperatures and flow rates. The acidic solutions were

found to be more effective than the alkaline solutions based on results seen in this

research. Singh et al., 1996 observed the effect of effluent flow rate on elution and found

that the faster flow rate resulted in lower peak and lower arsenic concentrations at low

number of bed volumes and there was a beneficial effect of a higher flow rate towards the

end of an elution run. Once arsenic is extracted, it can be analyzed using Graphite

Furnace Atomic Adsorption Spectrometer (GFAAS) techniques for concentration range

of 5-100 tg 1- using EPA method 7060 A, Inductively Coupled Argon Plasma

Spectroscopy (ICP) for concentrations above 100 tg 1- using EPA method 3010A and

6010 B, and Spectrophotometer (Bran+Luebbe GmbH) using industrial method 26-71E

for concentration range 0.4-20 ppm. GFAAS can also analyze the solid samples directly,

though often digestion is performed on solid samples for analysis (Sahuquillo et al.

2003). The advantages of this technique are its low detection limits, the minimum sample

manipulation, its relative simplicity and the short time required to obtain the results.

Flux Meter Analysis

Various scientists and organizations have made an attempt to identify the source of

contaminant and degree of contamination in groundwater using different field monitoring

techniques (Brusseau and Srivastava 1997, Montas et al., 2000). Brusseau and Srivastava

(1997) collected a dense, 3 dimensional array of sampling points to obtain time-series

data and spatial distribution which was used to determine the zeroth, first and second









spatial moments of the plume. Montas et al. (2000) presented a methodology for

designing groundwater quality monitoring well network in space and time, and to

evaluate the performance of the resulting network. However, the two techniques are often

time consuming and include various assumptions in the contaminant flux calculations and

large number of wells are drilled for correct measurement of contaminant (Montas et al.,

2000). Zhang et al. (2002) performed experimental investigation of contaminant transport

in coastal groundwater. The experiments were performed in a flow tank and glass beads

were used as the homogeneous porous media. The salt water and fresh water intrusion

was taken into consideration by incorporating the vertical/horizontal ratio of the sloping

seaward boundary keeping freshwater and saltwater at the high and low elevation

respectively. In this thesis, University of Florida flux meter are discussed for their use for

measuring contaminant flux in groundwater. Flux meter has been validated for most of

the organic contaminant analysis both in lab as well as field for contaminant flux,

groundwater flow and flow direction. In this research, use of flux meter for inorganic

contaminant like arsenic was analyzed.

Computer Model Simulation

A numerical model simulation is performed using MODFLOW MT3DMS for

arsenic contamination in various soil types. Soil classification described in earlier section

is used because various scientists working on arsenic contamination and its

concentrations in soils of Florida prefer this classification for research (Chen et al., 1999;

Clark et al., 2004). The advantages of using MODFLOW include numerous facilities for

data preparation, easy exchange of data in standard form, extended worldwide

experience, continuous development, availability of source code, and relatively low price

(McDonald and Harbaugh, 1988).






21


Partition coefficients were calculated for different soil types based on arsenic

concentration of 48 ppm. The Kd values were calculated using solution to solid ratio of

1:5. They range from 0.2 to 0.8, highest in marls and histosols (soil and water

department, University of Florida). Bulk density values were based on the ranges from

USDA county soil classification.

Porosity calculations are done on the basis of linear relation between porosity and

bulk density developed by NASA staff science 2003.














CHAPTER 4
MATERIALS AND METHODS

Various adsorption, extraction and analysis methods were performed for arsenic

adsorption onto activated carbon and activated alumina. The adsorption experiments were

performed as batch experiments for different time periods with respect to different

adsorbents because equilibrium time for each adsorbent is different. Extraction was

performed using strong acidic as well as alkaline solution of nitric acid and sodium

hydroxide respectively. Simple digestion process explained by EPA method 3010A for

extraction of arsenic from soils was used for extracting arsenic from activated alumina

using acid. For extraction using strong bases, standard procedure described by water

chemicals codex (Glaze et al., 1982) was used. Analysis was performed using GFAAS (Graphite

Furnace Atomic Adsorption Spectrometer) for all concentrations since it can measure

concentrations as low as 5 ppb (t-g 1-1). For higher concentrations, samples were diluted

to bring them within the detection range of 5 100 ppb. Typical flux meter set up is then

discussed and various boundary conditions and assumptions involved with the method

are described.

Adsorption Isotherms

Experimental quantities vary significantly for adsorption on activated carbon and

activated alumina because of their different efficiencies to adsorb arsenic. Activated

Alumina acts as a strong adsorbent whereas activated carbon is a weak adsorbent. In this

research, trace metal grade arsenic (Fisher Scientific) was used for analysis. Granular

activated alumina and activated carbon used in this research is commercially available









(Fisher Scientific). Adsorption experiments of arsenic on activated alumina were

conducted using 40 ml glass vials fitted with a Teflon-lined septa screw-top cap. In each

experiment .02-.20 g of activated alumina grains were placed into the vial (Lin and Puls,

2000) and 30 ml solution prepared at a predetermined arsenic concentration using de-ion

water. The temperature of the system was kept constant, and the reciprocating speed was

kept at 70 rpm for all experiments. System was kept on rotator for 48 hrs, for adsorption

to come into equilibrium (Lin and Puls, 2000). In adsorption experiment for activated

carbon, 40 ml vials and varying amount of activated carbon (1 5 grams) were used

(Pattanayak et al., 2000). The reciprocating speed was kept at 70 rpm and system was

kept for 72 hrs for adsorption to come into equilibrium (Pattanayak et al., 2000). After

rotating at their respective times, activated carbon/aluminum was allowed to settle for 12

hrs. Solution was then filtered for analysis using GFAAS. Since the initial concentration

of arsenic in the solution was known and arsenic left in the solution were analyzed,

respective calculations for sorbed concentration was performed.

Extraction Method

Arsenic extraction from activated alumina was performed using strong acid and

alkaline solutions using techniques described by EPA method 3010A and water chemical

codex respectively.

For extraction using acid, a known volume of well-mixed sample was transferred to

a 250 ml Griffin beaker, 2 ml of 30% H202 and sufficient concentrated HNO3 was added

to get an acid concentration of 1% (v/v). It was allowed to digest at 1050C for 2 hrs. The

solution was then filtered and injected into the furnace for analysis.

For extraction using base, first the extraction solution was prepared by adding 100

ml of 0.1 M potassium hydrogen phthalate to 47 ml of 0.1 N sodium hydroxide. Solution









was then diluted to 200 ml using de-ionized water. 0.1 g of sample was taken into a 250

ml beaker and 50 ml of extraction solution was added to completely digest activated

alumina samples. A drop of non-ionic surfactant (Brij 97) was then added and solution

was agitated with a magnetic stirrer for 1 hour. This solution was then filtered and

analyzed using GFAAS.

Analysis Method

As discussed above, arsenic analysis was performed using EPA method 7060 A.

the detection limit for this method is 1 gg/1 (Ippb). Samples were analyzed by GFAAS on

a Perkin Elmer SIMAA 6000 THGA. The details of parameters used are:

Wavelength= 193.7nm
Injection size = 20 tl
Matrix Modifier = 7 kl of 700 ppm Pd(N03)2

These parameters were used on the basis of the guidelines provided by the

manufacturer. Because temperature-sensing mechanisms and temperature controllers can

vary between instruments or with time, the validity of the furnace parameters were

periodically confirmed by systematically altering the furnace parameters while analyzing

a standard. In this manner, losses of analyte due to overly high temperature settings or

losses in sensitivity due to less than optimum settings can be minimized. 20 micro liter

aliquot of sample was injected into the furnace and atomized. If the concentration was

found to be greater than the highest standard (100 ppb), the sample were diluted in the

acid matrix (suggested by EPA method 7060A) and reanalyzed.

Flux Meter

Before conducting bench scale flux meter experiments for estimating contaminant

flux, the basic instrument was set up for simulating true field conditions. Hence this









section is divided into two, one containing information on set-up of the flux meter and

second on running the experiment.

Flux Meter Set-up

The test aquifer was set up in a box of approximately 46 cm length and 30 cm

width. Well casing and screen are constituted by a 3.2 cm inner diameter PVC tube that is

slotted over the entire height of the aquifer and that was glued to the bottom of the box to

keep it in position. A metal grid was placed 3.8 centimeters from the box walls where the

water enters and leaves to avoid intrusion of the aquifer material into glass beads. Glass

beads possesses a very high permeability compared to aquifer, which serves to establish a

vertical plane of constant head over the entire cross sections where the water enters and

leaves the aquifer. The aquifer consists of 20-30 mesh Ottawa sand (hydraulic

conductivity, 4.45 cm/min and porosity 0.3) that reaches up to a height of 11.5 cm in the

box.

The filling of box was performed by adding water simultaneously with the sand in

order to obtain a saturated aquifer from the beginning. The boundary conditions of

uniform flow field have to be approximated by the bench-scale aquifer. These conditions

are constant head at infinity on two opposite sides of the well and straight streamlines

going parallel to the other sides of the infinity. To achieve these conditions, the distance

from the well screen to the box walls on one side and to the gravel pack on the other side

were considered big enough in relation to the well diameter to assume the boundary

conditions for a uniform flow field as valid.

The water supply of the aquifer was established using a Kimax brand aspirator

bottle with tubulation, which was used as a water reservoir that keeps the head at a

constant level. As illustrated in Figure 4-1, this was achieved by placing a tube into the









closed container that maintains atmospheric pressure at its bottom. In order to guaranty

constant head conditions in the aquifer, the bottom of this tube was elevated to the level

of the top part of the aquifer. A constant flow pump on the other side of the aquifer was

used to convey a known constant flow through the aquifer, which was then collected in

another container to allow control of the actual flow over a certain time. A typical setup

of the flux meter is shown in Figure 4.1 (Klammler 2001).

Running Experiment

Before the experiment can be started, the device has to be prepared. As stated in the

previous section, it has to consist of a self-contained permeable unit that intercepts the

groundwater flow without retaining it. In this case, 8-14 mesh activated alumina

(hydraulic conductivity, 4.82 cm/min, porosity 0.382, particle density of 2.45 kg/l, and

bulk density is1.5 kg/1) was used for the permeable media, which took the role of the

sorbent for arsenic contamination. The sorbents (activated alumina) in the device sorbs

dissolved inorganic solutes (arsenic) present in the water intercepting the unit, thus

indicating the amount of contaminant carried by the groundwater. The device was placed

in the well, and pump was set to the required flow rate and started. In this case, flow rate

was approximately 4.83 mL/min, which result in specific discharges of 0.84 cm/hr for the

uniform flow field (cross sectional area A = 420 cm2).

The running times was aimed to be 7 hrs hours, thus conveying a total volume of

about 1500 mL to 2000 mL of water through the aquifer per run. A control volume at the

end of the process is used to quantify the total volume conveyed through the aquifer in

order to compute the actual average flow rate over the running time.













Sorbent media


Aquifer surface


Constant head


Constant flow
pump


'Glass Beads


= <= Qo


Figure4-1 Bench Scale set-up of flux meter (Klammler, 2001)


3.2cm


I I
I I
-------- Well


---------- Well C

------- Well R


15 in


1.5 in


12 in


1.5 in


Figure 4-2 Cross sectional dimensions of flux meter


p

Control
volume


Qo


i:.n

^:i^in

V^^- -'/'-^lS^-: -' A-






28


Since the hydraulic head at the beginning of the aquifer was set to a certain value

by the constant head reservoir and the flow rate was determined by the setting of the

pump, the hydraulic head at the end of the aquifer changes with changing flow rates.














CHAPTER 5
RESULTS AND DISCUSSION

Batch experiments for measuring adsorption coefficient for arsenic adsorption onto

activated carbon and activated aluminum were conducted at predetermined arsenic

concentrations. Strong acid and base extracts were analyzed for extraction efficiency, and

GFAAS was used for analysis. Flux meter tests were performed at a flow rate of 0.69

cm/hr to simulate horizontal flow lines and steady state in the box.

Adsorption Experiments

Batch experiments measuring adsorption coefficient for arsenic adsorption on

activated carbon and activated alumina were conducted for constant concentrations of 36

ppb and 17.4 ppm respectively (Pattanayak et al., 2000; Lin and Wu, 2000). These values

were used based on experiments conducted at different concentrations used for

adsorption of arsenic (Lin and Wu, 2000; Lorenzen et al., 1995). It was observed that

activated carbon is a weak adsorbent for arsenic contamination whereas activated

aluminum is a strong adsorbent for arsenic contaminant. Hence higher concentrations

were used for activated aluminum over activated carbon. Adsorption experiments for

arsenic adsorption on activated carbon indicated that activated carbon is not an adequate

sorbent for arsenic, therefore, it was not examined for extraction and flux meter

experiments. The results obtained from adsorption experiments for arsenic on activated

aluminum are shown in Table 5-1. Activated alumina was allowed to adsorb 30 ml of

17.4-ppm arsenic solution for 48 hrs and final concentration of arsenic left in the solution

was measured, which is indicated as solution concentration (C) in Table 5-1. The total










mass sorbed by activated alumina (total mass mass lest) was calculated. Mass sorbed

was then normalized with respect to mass of activated alumina to give sorbed

concentration (S).

Table 5-1 Adsorption of arsenic on activated alumina
Solution Total Mass Mass Sorbed Concentration
Concentration(C) Mass Left Sorbed (S)
ppm(mg/l) (mg) (mg) (mg) per unit AA mass (mg/g)
8.09 0.52 0.24 0.28 13.89
2.91 0.52 0.09 0.43 7.18
1.58 0.52 0.05 0.47 4.68
1.76 0.52 0.05 0.47 3.41
1.36 0.52 0.04 0.48 2.68
1.33 0.52 0.04 0.48 2.21

The adsorption isotherm (the curve of sorbed concentration Vs concentration in the

solution) was analyzed and for the lower range of activated alumina as in this case (0.02g

to .2g), a linear isotherm is obtained (Figure 5-1).


Figure 5-1: Adsorption isotherm of arsenic on activated alumina

The adsorption isotherm (Figure 5-1) is linear with Kd value of 1.844 L/g. The

adsorption isotherm obtained by (Lin and Wu 2000; Singh and Pant 2003) was found to


l16
14
12
= 1.8444C
- 10
SR2 == 0.9216
8





0 2 4 6 8 10

Solution Concentration, C (mgll)









follow non-linear trend. The difference can be attributed to lower range of solution

concentrations used in the analysis which lie on the lower end of the range used by Lin

and Wu 2000, and Singh and Pant 2003. However from the data it is evident that

activated alumina has a very high efficiency for arsenic and at lower concentrations of

arsenic, it can adsorb almost all of arsenic present in the solution.

Extraction Experiments

Since activated carbon was not found to be a promising adsorbent for arsenic, only

extraction of arsenic from activated alumina was considered for the analysis. Extraction

using acid and base was analyzed using EPA method 3010A and water chemical codex

(Glaze et al., 1982) prescribed method (from materials and methods section). Extraction

efficiency was calculated at the ratio of amount of arsenic extracted per unit amount of

arsenic sorbed. Extractions performed using strong base, sodium hydroxide and strong

acid, nitric acid were compared and it was observed that acidic extraction is more

efficient for extracting arsenic from activated alumina.

Nitric acid was used for extraction and the results obtained are shown in Table 5-2.

A graph of actual amount of arsenic on activated alumina with amount of arsenic

extracted using acid is shown in Figure 5-2.

Table 5-2 Arsenic extracted using nitric acid
Conc. of As in Mass of AA Volume of As per unit mass of AA
extracted solution extraction solution Extracted Actual
(mg/1) (g) (I) (mg/g) (mg/g)
6.20 0.1016 0.05 3.0512 3.4148
6.19 0.1378 0.05 2.2460 2.6838
6.60 0.1622 0.05 2.0345 2.2125












3.50
? 3.00 y = 0.8817x
S. R2 2=0.9598
E 2.50
t 2.00
5 1.50
C 1.00
E 0.50
0.00
0.00 1.00 2.00 3.00 4.00
Amount sorbed (mglg)

Figure 5-2 Amount extracted using acid vs. initial concentration

Graph of amount extracted with respect to actual amount on activated alumina

(Figure 5-2) indicated that extraction efficiency of nitric acid is around 88%. Extractions

were also performed using strong base, sodium hydroxide.

Table 5-3 Arsenic extracted using sodium hydroxide

Conc. Of As in Mass of AA Volume of As per unit mass of AA
extracted solution extraction solution Extracted Actual
(mg/I) (g) (I) (mg/g) (mg/g)
3.54 0.1055 0.0500 1.6777 1.9301
1.96 0.1072 0.0500 0.9142 1.0924
0.96 0.1059 0.0500 0.4514 0.7043


2.00

1.50

1.00

0.50

0.00
0.0


0


0.50 1.00 1.50 2.00
Amount Adsorbed (mglg)


Figure 5-3 Amount extracted using base vs. initial concentration


y 0.8412x
R2 = 0.9702









The results obtained from extraction using base are shown in Table 5-3. The graph

of actual amount of arsenic on activated alumina with amount of arsenic extracted using

base (sodium hydroxide) is shown in Figure 5-3.

Figure 5-3 indicates that extraction efficiency of sodium hydroxide to extract

arsenic is around 84%. However, the results obtained for both types of extraction were

similar i.e. efficiency of 85-90%.

Flux Meter Bench Scale Experiments

The UF flux meter was tested in the lab for its ability to determine arsenic

contaminant flux. Dimensionless cumulative mass () for activated alumina was

calculated (equation 2.15) from given parameters (section of materials and methods) to

be 0.59. Since this is less than 0.6, the linear relation developed by Klammler (2001) was

used to calculate mass of imaginary tracer retained in the flux meter (equation 2.15) as


1-m t = 1.2 qAt (5.1)
20Rkontr

Since the hydraulic conductivity of activated alumina (4.82 cm/min) is more than

the hydraulic conductivity of Ottawa sand (4.45 cm/min), the specific discharge through

flux meter was different from that through the surrounding aquifer. Therefore, to

calculate specific discharge in flux meter, qo was multiplied with a factor of 1.04

(equation 2.16). Also, from the extraction experiment the amount of arsenic sorbed per

unit mass of activated alumina was obtained. If this sorbed mass is denoted by moont, then

the concentration of arsenic in the solution as indicated by flux meter can be calculated as

(using equation 2.14, 2.15 and 2.17)


C = cntpr (5.2)
0.6qAt










where mnt cnt (5.3)
pLzrr
Arsenic was extracted using acid based on EPA method 6010A. Mean and standard

deviation of calculated concentration using actual and corrected mass (88% extraction

efficiency) was calculated. Results obtained from the flux meter analysis are shown in

t s
Table 5-4. From statistical t-test, 95% confidence intervals (/u+ 025 where is the


sample mean, to25 is the t-test parameter, s is the sample standard deviation and n is the

number of samples takes) for actual and corrected concentrations are 9.79 1.79 and

10.97 2.24 respectively. Arsenic concentration used in the influent solution in flux

meter set up was 12ppm, which lies within the 95% confidence interval obtained from the

concentration calculations using corrected mass. However, because the degree of freedom

for the analysis was low (df = 2, which results in high value of parameter to.025)

confidence interval provides a wide range for concentrations.

Table 5-4 Concentrations measured from flux meter tests
Extracted mass Concentration Calculated Using

m
Well cont 0.69 A
Location Actual Corrected Actual mass Corrected mass
(pg/g) (pg/g) (mg/1) (mg/1)
Well L 14.25 15.96 9.36 10.48
Well C 16.17 18.11 10.62 11.89
Well R 14.29 16.01 9.73 10.89
Mean 9.79 10.97
Std Dev 0.72 0.81

Since well L and well R are the peripheral wells (Figure 4-2), and due to possible

convergence of flow lines; flow through well L and well R might have been lower when

compared to the center well. Similarly, the total mass of arsenic passing through the well

on sides might have been less than expected, which in turn would result in lower

estimated flux concentration. From Table 5-4, calculated concentrations in well L and










well R (Figure 4-2) are less than the concentration calculated in well C. Well C provides

a good estimate (estimated value = 11.89ppm) of the actual concentration (12ppm) of

arsenic in the solution. This implies that flux meter is a promising tool to assess arsenic

contamination in groundwater.

Computer Simulation Results

Arsenic concentration is plotted with respects to time for varying soil types. The

parameters used for the analysis were taken from research through various journals and

documents as well as from data collected by University of Florida, soil and water

department.

Table 5-5: Soil properties
Soil Type Silt+Clay Content Bulk Density Kd Porosity
Marls 95 1000 0.8 0.62
Histosols 90 600 0.7 0.76
Spodosols 13.7 1300 0.3 0.51
Entisols 3.85 1300 0.2 0.51


Concentration Vs Time

9
8 -
C. 7
_6 Marls
2 5 --Histosols
S4 Spodosols
3 -x Entisols
c2
0

0 1000 2000 3000 4000
Time (days)

Figure 5-4: Concentration of arsenic vs. time for varying soil type.

The data used is described in the literature review section of the thesis. Most of this

data is obtained from findings of University of Florida, soil and water science department

or from the research conducted by Clark et al. (2004). As can be seen from the data,









marls and histosols have comparatively high silt and clay content. There high clay

content values reflect that higher values of bulk density should be expected. Conversely,

marls and histosols were found to have low bulk densities, and this is due to their higher

particle fraction. These fractions are mostly organic or limestone particles and are not

clay.

Marls are clayey soils and therefore, have a higher efficiency to retain arsenic,

which results in minimal leaching. From Marls to Entisols, the soil characteristics change

from higher composition of clay to lower composition of clay (and higher composition of

sand). From Figure 5-4, one can see that time taken by arsenic to reach maximum

concentration decreases with increasing sand content (or decreasing clay content) and

decreasing Kd values, hence decrease in dispersion from clayey soil to sandy soil (Marls

to Entisols) can be observed. Therefore, break through for entisols was achieved the first,

spodosols the second, histosols the third and marls the fourth.














CHAPTER 6
CONCLUSIONS

The purpose of this research was to evaluate the use of flux meter for estimating

arsenic contamination in groundwater. The primary objective was to evaluate the

capability of activated alumina as a flux meter sorption media for arsenic. Activated

alumina was chosen as the potential sorbents because it has a high efficiency for arsenic

(Lin and Wu, 2000).

From the batch experiments, it was observed that for constant initial arsenic

concentration of 17.4 ppm, with varying amount of activated alumina, a linearly

increasing isotherm with Kd value of 1.84 L/g is obtained. This showed that with higher

concentrations of arsenic in the aqueous phase, higher concentrations were sorbed onto

the solid phase.

Evaluation for determining a suitable technique for extracting arsenic from

activated alumina was explored. Both acid as well as base solutions were used. It was

observed that acid has relatively higher efficiency to extract arsenic from activated

alumina than a base. It was determined that extraction methods, acid and base, extract

around 85-90% of arsenic from activated alumina. The acid and base considered in the

research were nitric acid and sodium hydroxide respectively. The acid was used for

extracting arsenic from flux meter samples.

Bench scale experiments were conducted for evaluating the use of flux meter for

estimating arsenic concentration. A continuous flow of 12 ppm aqueous arsenic

concentration was maintained through the model aquifer at specific discharge of 0.84









cm/hr. The flux meters were placed in three wells located perpendicular to the flow lines

in the center of model aquifer, and arsenic adsorbed by activated alumina in each flux

meter was measured. Concentrations calculated based on the mass sorbed by activated

alumina were compared with the aqueous concentration used in the aquifer (12ppm). The

statistically estimated range of aqueous concentration, calculated on the based of arsenic

extracted from flux meter, provided the 95% confidence interval of 10.97+ 2.24 ppm.

The flux meter in central well estimated 99% of the actual concentration in aquifer.

The results obtained from flux meter proved it to be a promising device for

assessing arsenic contamination in groundwater. Potential applications for this method

are measuring arsenic fluxes for quantifying risk to human from arsenic leaching through

hazardous waste sites, or from source zones before and after remedial efforts.

Future research can be conducted to evaluate compatibility of various available

tracers with activated alumina for fluid flow determination.
















LIST OF REFERENCES


Alam M. G. M., Snow E. T. and Tanaka A. 2003. Arsenic and heavy metal contamination
of vegetables grown in Samta village, Bangladesh. The Science of The Total
Environment. 308: 83-96.

Azcue, J. M. and Nriagu, J. O. 1995. Impact of abandoned mine tailing on the arsenic
concentrations in Moira Lake, Ontario. Journal of Geochemical Exploration. 52:
81-89.

Barry D. A., Prommer H., Miller C. T., Engesgaard P., Brun A. and Zheng C. 2002.
Modeling the fate of oxidisable organic contaminants in groundwater. Advances in
Water Resources. 25: 945-983.

British Geological Survey (BGS) 2001. Groundwater studies for arsenic contamination.
Report WC/00/19. Volumes 1-10.

Brusseau, M., Srivastava R. 1997. Nonideal transport of reactive solutes in heterogeneous
porous media: 2. Quantitative analysis of the borden natural-gradient field
experiment. Journal of Contaminant Hydrology. 28(1-2): 115-155.

Bull D. C. 2000. The chemistry of chromated copper arsenate, characteristics of timber
treatment plant chemical sludge. Wood Science and Technology. 34:367-376.

Chakravarty S., Dureja V., Bhattacharyya G., Maity S. and Bhattacharjee S. 2002.
Removal of arsenic from groundwater using low cost ferruginous manganese ore.
Water Research. 36:625-632.

Charbeneau R. J. 2000. Groundwater hydraulics and pollutant transport. Prentice Hall,
Inc. Upper Saddle River. NJ.

Chen M., Ma L. Q. and Harris W. G. 2002. Arsenic concentrations in Florida surface
soils: influence of soil type and properties. Soil Science Society of American
Journals. 66:632-640.

Chen, M., Ma, L.Q. and Harris, W.G. 1999. Baseline concentrations of 15 trace elements
in Florida surface soils. Journal of Environmental Quality. 28:1173-1181.

Chirenje T., Ma L. Q., Clark C. and Reeves M. 2003. Cu, Cr and As distribution in soils
adjacent to pressure-treated decks, fences and poles. Environmental Pollution.
124:407-417.









Clark J. C., Chirenje T., Annable M. D., Hatfield K. 2004. Developing retention indices
and modeling transport of CCA in Florida soils at unlined landfills. Florida center
of solid and hazardous waste management. Final Project Report.

Day, M. J., Reinke, R. F. and Thomson, J. A. M. 2001. Fate and transport of fuel
components below slightly leaking underground storage tanks. Environmental
Forensics.2: 21-28

Deliyanni E. A., Bakoyannakis, D. N., Zouboulis A. I. and Matis K. A. 2003. Sorption of
As(V) ions by akaganeite-type nanocrystals. Chemosphere. 50:155-163.

Domenico P.A. and Schwartz F. W. 1990. Physical and Chemical Hydrogeology. John
Wiley and Sons, Inc. New York City. New York

Florida Department of Environmental Protection (FDEP) 2004. New arsenic standards
applicable to both domestic and industrial wastewater facilities. Chapter 62-550,
Florida Administrative Code (F.A.C.)

Galba J and Polacek S. 1973. Sorption of arsenates under kinetic conditions in selected
soil types. Acta Fytotech. 28: 187 -197.

Ghimire K. N., Zhul Y., Yano M., Makino K. and Miyajima T. 2002. Effective use of
orange juice residue for removing heavy and radioactive metals from environment.
Geosystem Engineering. 5(2): 31-37.

Ginn T. R., Wood B. D., Nelson K. E., Scheibe T. D., Murphy E. M. and Clement T. P.,
2002. Processes in microbial transport in the natural subsurface. Advances in Water
Resources. 25:1017-1042

Halter W. E. and Pfeifer H. R. 2001. Arsenic (V) adsorption onto a-A1203 between 25
and 700C. Applied Geochemistry. 16(7-8): 793-802.

Hatfield K., Annable M., Cho J, Rao P. S. C. and Klammler H. 2003. A direct passive
method for measuring water and contaminant fluxes in porous media. Journal of
Contaminant Hydrology. In press.

Jones K. D. and Huang W, 2003. Evaluation of toxicity of the pesticides, chlorpyrifos
and arsenic, in the presence of compost humic substances in aqueous systems.
Journal of Hazardous Materials. 103: 93-105.

Kayajanian G. 2003. Arsenic, cancer, and thoughtless policy. Ecotoxicology and
Environmental Safety. 55: 139-142.

Klammler H. 2001. Contaminant hydrology A new method for measurement of
groundwater and contaminant flux. PhD Dissertation. Technical University Graz,
Austria









Leeuwen, F. X. R. 2000. Safe drinking water: the toxicologist's approach. Food and
Chemical Toxicology. 38:S51-S58

Lin T. F. and Wu, J. K. 2000. Adsorption of arsenite and arsenate within activated
alumina grains: equilibrium and kinetics. Water Resource (Oxford). 35: 2049-2057.

Lin Z and Puls R. W. 2000. Adsorption, desorption and oxidation of arsenic affected by
clay minerals and aging process. Environmental Geology. 39: 753-759.

Lorenzen L., Deventer J. S. J. van and Landi W. M. 1995. Factors affecting the
mechanism of the adsorption of arsenic species on activated carbon. Minerals
Engineering. 8:557-569.

McDonald, M.G. and Harbaugh A.W. 1988. A modular three-dimensional finite-
difference ground-water flow model. USGS TWRI. 6-Al: 586

Meng X., Korfiatis G. P., Bang S., Bang K. W. 2002. Combined effects of anions on
arsenic removal by iron hydroxide. Toxological Letters. 133: 103-111.

Montas H. J., Mohtar R. H., Hassan A. E. and Alkhal F. A. 2000. Heuristic space-time
design of monitoring wells for contaminant plume characterization in stochastic
flow fields. Journal of Contaminant Hydrology. 43(3-4): 271-301.

National Aeronautics Space Administration (NASA) Staff Science 2003. (http://www-
eosdis.oml.gov /FIFE /Datasets /Soil Properties /Soil_Survey Ref.html). Last seen
on September 2003.

Nimick, D. A., Moore J. N. Dalby C. E. and Savka M. W. 1998. The fate of geothermal
arsenic in Madison and Missouri rivers, Montana and Wyoming. Water Resource
Research. 34: 3051-3067.

Pattanayak J., Mondal K., Mathew S., Lalvani S. B. 2000. A parametric evaluation of the
removal of As (V) and As(III) by carbon based adsorbents. Carbon. 38: 589-596.

Raven K. P., Jain A. and Loeppert, R. H. 1998. Arsenite and arsenate adsorption on
ferrihydrite: kinetics, equilibrium and adsorption envelopes. Environmental Science
and Technology. 32: 344-349.

Sahuquillo A., Rauret G., Rehnert A. and Muntau H. 2003. Solid sample graphite furnace
atomic adsorption spectroscopy for supporting arsenic determination in sediments
following a sequential extraction procedure. Analytica Chemica Acta. 476:15-24.

Salas J. and Ayora C. 2004. Groundwater chemistry of the Okelobondo uraninite deposit
area (Oklo, Gabon): two-dimensional reactive transport modeling. Journal of
Contaminant Hydrology. 69:115-137.

Singh D. B., Prasad G. and Rupainwar D. C. 1996. Adsorption technique for the
treatment of As(V) rich effluents. Colloids and Surfaces. 111: 49-56.









Singh T. S. and Pant K. K. 2003. Equilibrium, kinetics and thermodynamic studies for
adsorption of As (III) on activated alumina. Separation Purification Technology.
36:139:147.

Smedley P. L. and Kinniburg D. G. 2002. A review of the source, behavior and
distribution of arsenic in natural waters. Applied Geochemistry. 17: 517-168.

Smedley P. L and Kinniburgh D. G. 1996. Arsenic contamination of groundwater in
bangladesh, Sorption and Transport. British Geological Survey Report WC/00/19.
Volume 1.

Solo-Gabriele H., Townsend T.G. 1999. Disposal practices and management alternatives
for CCA-treated wood waste. Waste Manage. Residential. 17:378-389.

Solo-Gabriele H. M., Townsend T. G. and Messick B. 2002. Characteristics of chromated
copper arsenate-treated wood ash. Journal of Hazardous Materials. B89: 213-232

Townsend T., Stook K., Tolaymat T., Song J.K., Solo-Gabriele H., Hosein N., and Khan
B. 2000. New lines of CCA-treated wood research: in-service and disposal issues.
Florida Center for Hazardous Waste. Technical Report.

United States Environmental Protection Agency (USEPA) 2001. National primary
drinking water regulations; arsenic and clarification to contaminants monitoring.
DOCID:fr22ja01-29. 66:6975-7066.

Wang L., Chen A. and Fields K. 2000. Arsenic removal from drinking water by ion
exchange and activated alumina plants. EPA Publication EPA/600/R-00/088.

Glaze W. H., Buescher C. A. and Mahon J. H. 1982. Water chemical codex. Activated
alumina. National Academy press. Washington D. C. pp 57-58.

Wilkie, J. A. and Hering J. G. 1996. Adsorption of arsenic onto hydrous ferric oxide:
effect of adsorbate/adsorbent ratios and co-occuring solutes. Colloidal Surfaces.
107:97-110.

Zhang, W., Cai Y., Tu C., and Ma. L.Q. 2002. Arsenic speciation and distribution in an
arsenic hyper accumulating plant. Science of Total Environment. 300:167-177.

Zheng, C. and Bennett, G.D. 2002. Applied contaminant transport modeling; John Wiley
and Sons, Inc., New York City. New York.















BIOGRAPHICAL SKETCH

Prachee Gupta was born in 1980 in Rajasthan, India. She received a Bachelor of

Technology (B.Tech) from the Indian Institute of Technology (IIT), Mumbai, in 2002. In

the fall of 2002, Prachee joined the University of Florida for a Master of Science in the

Department of Civil and Coastal Engineering under the tutelage of her major professor,

Dr. Clayton J. Clark, II.




Full Text

PAGE 1

UTILIZATION OF UNIVERSITY OF FL ORIDA FLUX METER FOR ESTIMATING ARSENIC CONTAMINATI ON IN GROUNDWATER By PRACHEE GUPTA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Prachee Gupta

PAGE 3

ACKNOWLEDGMENTS I would like to express my sincere gratitude to Dr. Clayton J. Clark II, chairperson of my supervisory committee, for all his guidance and support throughout the project. I am very grateful to my committee, members, Dr. Kirk Hatfield and Dr. Michael Annable, for all their valuable guidance, suggestions and encouragement. I would also like to thank Dr. Jaehyun Cho, Dr. Mark Newman and Mr. Harald Klammler for constant guidance in performing laboratory experiments and Dr. Tait Cherenji for providing basic data for computer model simulation. I greatly appreciate the help of Mr. Thomas Luongo for sample analysis. Thanks go to the students of Water Resource Research Center for the support I received throughout. Finally, I want to thank my parents and friends who encouraged me throughout my academic career. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT.....................................................................................................................viii CHAPTER 1 INTRODUCTION......................................................................................................1 General........................................................................................................................1 Purpose and Objectives...............................................................................................3 2 THEORETICAL BACKGROUND...........................................................................4 Advective-Dispersive Transport of Reactive Solutes.................................................4 Linear Isotherm..................................................................................................4 Non-linear (Freundlich Isotherm) ......................................................................5 Non-linear (Langmuir Isotherm)........................................................................6 University of Florida Flux Meter................................................................................6 Estimating Fluid Flux........................................................................................7 Estimating Solute Mass Flux.............................................................................9 3 LITERATURE REVIEW.........................................................................................12 Occurrence of Arsenic..............................................................................................12 Fate and Transport of Arsenic...................................................................................13 Adsorption Characteristics of Arsenic......................................................................15 Laboratory Analysis..................................................................................................18 Adsorption Isotherms......................................................................................18 Extraction and Analysis:..................................................................................19 Flux Meter Analysis..........................................................................................19 Computer Model Simulation.....................................................................................20 iv

PAGE 5

4 MATERIALS AND METHODS.............................................................................22 Adsorption Isotherms................................................................................................22 Extraction Method....................................................................................................23 Analysis Method.......................................................................................................24 Flux Meter.................................................................................................................24 Flux Meter Set-up...........................................................................................25 Running Experiment.......................................................................................26 5 RESULTS AND DISCUSSION..............................................................................29 Adsorption Experiments...........................................................................................29 Extraction Experiments.............................................................................................31 Flux Meter Bench Scale Experiments.......................................................................33 Computer Simulation Results...................................................................................35 6 CONCLUSIONS.......................................................................................................37 LIST OF REFERENCES...................................................................................................39 BIOGRAPHICAL SKETCH.............................................................................................43 v

PAGE 6

LIST OF TABLES Table page 3-1 Behavior of arsenic with different minerals.............................................................17 5-1 Adsorption of arsenic on activated alumina.............................................................30 5-2 Arsenic extracted using nitric acid...........................................................................31 5-3 Arsenic extracted using sodium hydroxide..............................................................32 5-4 Concentrations measured from flux meter tests.......................................................34 5-5 Soil properties..........................................................................................................35 vi

PAGE 7

LIST OF FIGURES Figure page 2-1 Contaminant sorbed by the sorbing matrix..............................................................10 3-1 Cumulative probability curve for As in Florida surface soils...............................14 3-2 Eh-pH diagram for inorganic arsenic compounds....................................................15 4-1 Bench Scale set-up of flux meter..............................................................................27 4-2 Cross sectional dimensions of flux meter................................................................27 5-1 Adsorption isotherm of arsenic on activated alumina..............................................30 5-2 Amount extracted using acid vs. initial concentration.............................................32 5-3 Amount extracted using base vs. initial concentration.............................................32 5-4 Concentration of arsenic vs time for varying soil type. ...........................................35 vii

PAGE 8

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science UTILIZATION OF UNIVERSITY OF FLORIDA FLUX METER FOR ESTIMATING ARSENIC CONTAMINATION IN GROUNDWATER By Prachee Gutpa August 2004 Chair: Clayton J. Clark II Major Department: Civil and Coastal Engineering Groundwater contamination is a major problem in todays environment. In Florida, the ground water standards are equivalent to the drinking water standards according to the Florida Department of Environmental Protection (FDEP). Among the various contaminants, organic, inorganic, microbial pathogens and radioactive contaminant, inorganic contaminants are of high interest because most of these contaminants are readily soluble in water, and have a high potential to contaminate groundwater. The present research focused on the groundwater contamination due to arsenic. The maximum permissible concentration of arsenic in drinking water in United States is 10g l-1 or 10 ppb as recommended by United States Environmental Protection Agency. The purpose of the research was to evaluate the use of the University of Florida flux meter for estimating arsenic contamination in the groundwater. The flux meters were placed in three wells located perpendicular to the flow lines in the center of model aquifer. The internal adsorbent chosen was activated alumina due to its high efficiency in removal of arsenic. Adsorption of arsenic on activated alumina showed a linear trend with a Kd value of viii

PAGE 9

1.844 L kg-1. The procedure for extracting arsenic adsorbed by activated alumina was evaluated and it was found that extraction efficiency of acid (88%) is comparatively more than that of base (84%). Flux meter tests were conducted for arsenic concentration of 12 ppm at a flow rate of 0.84 cm/hr. The flux meter in the central well estimated 99% of the actual arsenic concentration in aquifer Therefore this provides an efficient and cost effective way to estimate arsenic flux in the field. ix

PAGE 10

CHAPTER 1 INTRODUCTION General Groundwater contamination is a major problem in todays environment since groundwater constitutes major portion of potable water in Florida. In Florida, the ground water standards are equivalent to the drinking water standards, as stated by Florida Department of Environmental Protection (FDEP). Groundwater can be contaminated by disease-producing pathogens, leachate from landfills and septic systems, careless disposal of hazardous household products, agricultural chemicals, and leaking underground storage tanks (Ginn et al., 2002; Day et al., 2001; Leeuwen, 2000). There are four basic types of contaminants that include organic, inorganic, radioactive elements, and microbial pathogens (Salas and Ayora 2004; Jones and Huang, 2003; Barry et al., 2002; Ginn et al., 2002). Among the various contaminants, inorganic contaminants are of high interest because most of these contaminants are readily soluble in water, and have a high potential to contaminate groundwater. Relative to other oxyanion forming minerals, arsenic is problematic because of its mobility in most of its oxidations states and at almost all pH values. During 2001, roughly 96% of the arsenic imported into the United States was in the form of arsenic tri-oxide (Solo-Gabriele et al., 2002). Most of this arsenic trioxide was used for the production of arsenic acid used for wood preservation. Arsenic is brought into the State of Florida for the production of chromated copper arsenate (CCA) treated wood (Chirenje et al., 2003 and Solo-Gabriele and Townsend, 1999). CCA is composed of three major 1

PAGE 11

2 elemental constituents, copper (Cu), chromium (Cr), and arsenic (As), among which arsenic is the most hazardous. The present research focused on the soil and groundwater contamination due to arsenic. Arsenic is a semi-metal element, found in environment combined with other elements such as oxygen, chlorine, and sulfur to form inorganic arsenic compounds. It also combines with carbon and hydrogen to form organic compounds. It most often exists in organic form as monomethyl arsonic acid (MMA) and dimethyl arsonic acid (DMA). Arsenic exists in the environment most often in two oxidation states: arsenite, As (III), and arsenate, As (V), respectively (Singh and Pant 2003; Meng et al., 2002). Under oxidation conditions (and aerobic environments) arsenates are stable species (as oxyanions H3AsO4, H2AsO4-, HAsO4-2 or AsO4-3) and are strongly sorbed onto clays, iron, and manganese oxides and hydroxides, and organic matters. However, under most reducing conditions (and anaerobic environment) arsenites (H3AsO3, H2AsO3-1, HAsO3-2) are the predominant arsenic compounds (Chakravarty et al., 2002). According to Kayajanian (2003), studies on Utah cancer mortality as a function of lifetime arsenic exposure indicated that for concentration range of 1-25 ppb, there are 2.682 cancers per 100 people and in the 5 years following initial arsenic medication, around 6.45 deaths are expected. The maximum permissible concentration of arsenic in drinking water in United States is 10g L-1 or 10 ppb as recommended by United States Environmental Protection Agency, USEPA (2001). In addition to the arsenic contamination in water sources, dietary intake of arsenic through the food chain via uptake from contaminated soils may adversely affect human health (Alam et al., 2003). The current soils clean up goals set by the Florida department of environmental

PAGE 12

3 protection for arsenic in residential and industrial soils are 0.8 mg/kg and 3.6 mg/kg respectively (Solo-Gabriele et al., 2002). Various scientists and organizations have made an attempt to identify the source of contaminant and degree of contamination in groundwater using different field monitoring techniques (Montas et al., 2000; Brusseau and Srivastava, 1997). This research discusses a new devise named Flux meter that allows to simultaneously measuring cumulative dissolved solute fluxes when placed within a flow field. The Flux meters have been tested for most of the organic contaminant analysis both in lab as well as field for contaminant flux, groundwater flow and flow direction (Hatfield et al., 2003; Klammler 2001). Purpose and Objectives The purpose of this research was to evaluate the use of flux meter for estimating arsenic contamination in the groundwater. Before bench scale flux meter tests are performed, information regarding the type of adsorbents and its interaction with arsenic should be understood very well. Hence, the objectives of this research were 1) review of adsorption and extraction characteristics of arsenic to different materials and 2) to evaluate the applicability of these materials as flux meter sorbent media for arsenic flux measurement, 3) to simulate the arsenic contamination for various Florida soil types. This thesis seeks to give the basic understanding of the solute transport and theoretical contemplation of flux meter for estimating fluid and contaminant flux, arsenic contamination and need of flux meter for estimating fluid flow and contaminant flux in the groundwater.

PAGE 13

CHAPTER 2 THEORETICAL BACKGROUND Advective-Dispersive Transport of Reactive Solutes In non-ideal condition found in the environment, a reactive solute interacts with the solid matrix during flow. The solute is, therefore distributed between the fluid and the sorbed phases. If a primary anisotropic system is assumed, for the steady water flow, under saturated or uniformly unsaturated conditions, the solute transport equation for a homogeneous media can be represented by (Domenico and Schwartz, 1990) 22wwwbwwwCCSDqttx wCt (2.1) where w= volumetric water content Cw= solute concentration in water (g L-1) b= bulk density of sorbing media (g/cm3) S= sorbed solute concentration (mg/kg) Dw = dispersion coeffecient (cm2/hr) qw = advective mass flux (cm/hr) Three most commonly known isotherms that define the adsorption characteristics of any adsorbent are linear isotherm, Freundlich isotherm and Langmuir isotherm (Zheng and Bennett 2002, Charbeneau 2000). Freundlich and Langmuir isotherms can be used to represent both linear and non-linear isotherms. Linear Isotherm When the sorbed phase solute concentration is linearly proportional to the solute concentration in the solution phase, the sorption isotherm is linear isotherm, and is represented by S = Kd.Cw. Therefore, the equation 2.1 is transformed to 4

PAGE 14

5 22[1]wbdwwwCKCDtx wCt (2.2) The transport velocity (cm/hr) is represented by v. The adsorption coefficient (Kd) is a measure of how tightly the solute binds or sticks to soil particles. The greater the Kd value, the less likely a chemical will leach or contribute to runoff. The retardation factor, R = (1+ Kd/w), expresses how much slower a contaminant moves than does the water itself. Equation 2.2 is therefore transformed to 22.wwCCDtx wCt (2.3) where = D D R and = R When Kd = 0, it implies that R = 1 and hence D* = D and v* = v, which reverts the equation (equation 2.3) back to the non-reactive ideal solutes. Hence the equations below show the effect of retardation factor on the solute transport. As Kd and R increases, the distance to solute peak decreases. The distance to solute peak (x*) is given by x* = [v*t] = t R (2.4) The time of arrival of solute peak (t*) is delayed and is given by t* = [L/v*] = LR (2.5) Non-linear (Freundlich isotherm) In this type of isotherm the solute mass is proportional to the concentration raise to N. S = K C, N 1 (2.6) Nw

PAGE 15

6 Hence, retardation factor 1[1]NwwKNCR It is evident from the above equation that R is not a constant but is a function of Cw. N<1, R (Cw) decreases as Cw increases for Cw >1 N>1, R (Cw) increases as Cw increases for Cw >1 N = 1, R is independent of Cw and hence leads to the linear isotherm. Non-linear (Langmuir Isotherm) Langmuir isotherm is the most common type non-linear isotherm and in this isotherm max1wwSkCSkC (2.7) Hence, retardation factor for Langmuir isotherm is max2()[1(1)wwwSkRCkC ] In non-ideal environmental conditions, it is very difficult to obtain the accurate values of various parameters and hence accurate flux measurement is difficult. There is a need for accurate and easy to use equipment for the calculation of groundwater flux and contaminant flux. A new method utilizing flux meter has been developed at the University of Florida and laboratory tested for measuring both contaminant and groundwater fluxes at hazardous waste site. University of Florida Flux Meter The University of Florida flux meter measures cumulative dissolved solute fluxes with directions simultaneously when placed within a flow field (Hatfield et al., 2003). The device consists of a self-contained permeable unit that intercepts the groundwater flow without retaining it. A matrix of hydrophobic and hydrophilic sorbents in the device

PAGE 16

7 sorbs dissolved organic and inorganic solutes present in the water intercepting the unit, thus indicating the amount of contaminant carried by the groundwater. The sorbent matrix is also impregnated with known amount of one or more fluid soluble resident tracers that leaches from the sorbent at the rate proportional to the fluid flux (Hatfield et al. 2003) and provides an estimate for the fluid flow. The flux meter has been validated and used for estimating fluid flow and contaminant flux for various organic contaminants. This research seeks to explore the use of flux meter for measuring arsenic contamination. Estimating Fluid Flux Fluid flow can be estimated by placing flux meter device in a monitoring well perpendicularly intercepting the groundwater flow. The amount of tracer leaching out of the device gives the measure of fluid flux. It is assumed that a) the various tracers used do not mutually affect their partitioning properties; b) tracers partition isotherm is linear and, c) tracer transport due to dispersion is negligible (Hatfield et al., 2003). Therefore the distance that the tracer is pushed away by the water can be calculated as x = trt = t R (2.8) where vtr = tracer transport velocity. If the water flows through the meter at the rate vtr = v/R, the remaining amount of the tracer with respect to the initial tracer amount (represented by mR) gives the measure of how much water has passed the unit. The rate of vtr is measured by vtr = v/R. Given the cross sectional area of the unit, one can relate this concentration to the area still occupied by the tracer as

PAGE 17

8 212sin()aArar x (2.9) 2RAmr (2.10) 2qt x rR [v = q ] (2.11) where 2[()2 2 x ar (2.12) r = radius of the device If is the dimensionless cumulative volume of fluid conveyed through the device, then equation 2.10 can also be stated as 122[sin(1)1]Rm 2 (2.13) 2qt R r (2.14) However, due to the circular cross section only at the very beginning is the tracer present over the whole width of the device (2r). As the tracer is desorbed, some water will pass the device without leaching out any of the tracer. The distance a from the center, this is equal to the radius of the section at the beginning. This also decreases with growing x and the water passing the device at distances greater than a from the center does not leach out tracer any more. Therefore, equation 2.13 must be used to describe the relation between the relative remaining tracer mass mR and the dimensionless cumulative volume To simplify equation 2.13, Klammler (2001) performed regression analysis for the variation of with mR and observed that regression equation is almost linear for less

PAGE 18

9 than equal to 0.6, which corresponds to the mR equal to 0.3. The regression equation obtained was mR = -1.2 +1.0 (2.15) After keeping the flux meter in the groundwater for a determined amount of time, the amount of tracer left in the sorbent media (mR) is extracted. The dimensionless cumulative volume of water, intercepted by the flux meter at a specified depth can be obtained using equation 2.15. The specific discharge, q, can then be calculated using the equation 2.14. The specific discharge calculated using equation 2.14 is the discharge through the flux meter. Because the material used in the aquifer can be different from that used in flux meter, the permeability (or hydraulic conductivity) of flux meter can be different from permeability of the surrounding aquifer. The specific discharge in the well and surrounding aquifer can be related as (Hatfield et al., 2003) 2ooqkqkk (2.16) Where, q and qo are the specific discharge through the flux meter and surrounding aquifer respectively. k is the hydraulic conductivity of flux meter and ko is the hydraulic conductivity of aquifer. Estimating Solute Mass Flux The dotted area shown in Figure 2-1 is where the contaminant is sorbed in to the device. The respective retardation factor of the contaminant (Rcont) quantifies the fraction of the cross sectional area of the device that is actually occupied by the sorbed contaminant (Klammler 2001). Hence the actual concentration of contaminant in aqueous phase is given by

PAGE 19

10 2,(1)contRcontcontMCrLmR (2.17) where Mcont = mass of contaminant sorbed (mg) r = radius of the flux meter cylinder (cm) L = length of sorbent matrix (cm) mR,cont = relative mass of hypothetical resident tracer retained after time period t. Rcont = retardation of sorbent (contaminant) on matrix (L g-1) = water content in the device. To calculate the amount of tracer retained in the flux meter (mR,cont) that is required for contaminant flux calculations (equation 2.17), an imaginary tracer is assumed in the flux meter that has the same leaching properties as that of contaminant (i.e. the tracer and contaminant have the same retardation factor). Equation 2.13 and 2.14 can then be used for calculating mR,cont using the parameters of sorbent matrix. Figure 2-1 Contaminant sorbed by the sorbing matrix Device Area of sorbed contaminant v0 v0 Since the calculated concentration is same in the aqueous phase of the aquifer and aqueous phase of the device, it can be directly used to calculate the contaminant flux using J = qo. C (2.20)

PAGE 20

11 Like any other process, flux meter also has a disadvantage in it. This method only works for plume constituents that are actually retained by the sorbing matrix of the device. For plume constituents that are not retained by the sorbing matrix, the contaminant flux is not measured. Hence the prior knowledge of type of contaminant in the area of interest is very important. Evaluating flux meter for monitoring arsenic concentration and the various parameters needed to be considered while simulating field conditions in lab are discussed in next chapter. Also, this method has not been validated for wide scale use for measuring inorganic solutes.

PAGE 21

CHAPTER 3 LITERATURE REVIEW Arsenic predominantly exists in two oxidation states, As (III) and As (V). USEPA 2001 and FDEP 2004 have revised their maximum contamination level limits from 50g l-1 to 10g l-1 starting from 2006 and 2005, respectively. In this study, arsenic contamination due to chromated copper arsenate (CCA) leaching from construction and demolition landfills is considered. Various factors that can affect the sorbed concentration of arsenic are discussed and affect of varying soil types is also initiated. Occurrence of Arsenic Occurrence of arsenic can be both anthropogenic including mining and industrial waste, and geogenic such as volcanic eruption and weathering, in nature. High concentration of arsenic in ground water can occur in some areas as a result of inputs from geothermal sources. Nimick et al. (1998) found up to 370g l-1 arsenic in Madison River water as a result of geothermal inputs from the yellow-stone geothermal system. Smedley and Kinniburgh (1996) noted high arsenic concentrations (around 200-300g l-1) in surface waters affected by Sn and Au mining activities. Azcue and Nriagu (1995) conducted experiments for arsenic concentrations in Moira lake of Canada throughout the year and showed the significant seasonal difference with an average concentration of 62 g l-1 during summers and 22 g l-1 during winters, they attributed this difference to the depleted oxygen levels in the bottom lake waters as a result of biological productivity during summers. 12

PAGE 22

13 Arsenic is widely used in making wood preservatives like CCA, insecticides and pesticides and for various other agricultural and industrial purposes. Due to its widespread use in agriculture, industry, and medicine, it has gained attention of lot of scientists and researchers for its health effects and required remediation procedures. Arsenic is brought into the State of Florida primarily for the production of CCA-treated wood (Chirenje et al., 2003 and Solo-Gabriele and Townsend, 1999). Upon disposal, the majority of CCA-treated wood is may be disposed in unlined construction and demolition (C&D) landfills, or recycled as mulch or wood fuel. As a result, a considerable amount of arsenic can be potentially released into the Florida environment. Townsend et al. (2000) estimated that, in Florida, CCA-treated decks impact ~ 10 000 ha or 108 m2 of soil, and the amount of CCA-treated utility poles in use is 7x105 m3. Arsenic is predominant in +5 oxidation state in CCA, though as the pH changes the stable oxidation state for arsenic varies (Solo Gabriele et al., 2002 Bull, 2000). Fate and Transport of Arsenic Relative to other oxyanion forming minerals, arsenic is problematic because it is mobile in most of the oxidations states and at almost all pH values, whereas other oxyanion like selenium are mobile as selenate (SeO4-2) under oxidizing conditions but are immobile under reducing conditions either due to the stronger adsorption of its reduced form, selenite (SeO3-2), or due to its reduction to metal (Smedley and Kinniburgh, 1996). Chen et al. (2002) collected over 448 samples to determine arsenic distribution in soils of Florida. The soil order described by Chen et al. (1999) is used to classify different types of soils including Alfisols (14%), Entisols (22%), Histosols (10%), Mollisols (4%), Inseptisols (3%), Spodisols (28%) and Ultisols (19%). Clark et al. (2004) grouped these soil types into three classes, class 1, 2, and 3 of arsenic retention indices. Class 1 includes

PAGE 23

14 Marls and Histosols; these soils have the greatest retention and minimal leaching capability of arsenic. Class 2 includes Entisols, Alfisols and Ultisols. These soils found to have moderate retention with moderate potential for leaching. Class 3 have the least retention and greatest risk for leaching and it include Spodosols. Chen et al. (2002) observed the decrease in arsenic concentration with following trend in soil order: Histosols> Inceptisols> Mollisols> Ultisols> Entisols, Alfisols> Spodosols. A cumulative probability curve for Arsenic concentration in Florida surface soils was developed as shown in Figure 3-1 (Chen et al., 2002), where x-axis plots a histogram of the data categories and y-axis plots the frequency of the data in percentage. Numbers in parentheses are sample sizes of individual soil orders. Figure 3-1: Cumulative probability curve for As in Florida surface soils (Chen et al. 2002)

PAGE 24

15 Adsorption Characteristics of Arsenic Redox potential (Eh), and pH are most often the controlling factors in mobility of arsenic in the environment. Under oxidizing conditions, H2AsO4is dominant, at low pH (pH of 2 to 6); while at high pH (6 to 12), HAsO4-2 becomes dominant (Smedley and Kinniburgh, 2002). The Eh-pH diagram for arsenic is given in Figure 3-2. Solubility of arsenic in natural systems is strongly influenced by its capacity to be adsorbed by different minerals present in soils. The interaction of arsenic with different minerals is represented in the Table 3-1. Figure 3-2: Eh-pH diagram for inorganic arsenic compounds (Smedley and Kinniburgh, 2002) Among the treatment processes appropriate for removal of arsenic, activated alumina adsorption is considered to be in-expensive and more versatile (Wang et al., 2000). Activated alumina (AA) has high efficiency to adsorb arsenic in both of its oxidation states. However, ionic strength, pH, competing ions, and temperature can significantly change adsorption characteristics of activated alumina for arsenic (Deliyanni

PAGE 25

16 et al., 2003; Meng et al., 2002; Lin and Wu, 2000; Wang et al., 2000; Halter and Pfiefer 2001). Lin and Wu (2000) found that arsenite uptake on activated alumina is much less than arsenate uptake at almost all pH conditions. The uptake of arsenite was found to increase with the increasing pH until pH ~ 7, and then decrease as pH increases. This was attributed to surface charge of AA, which is positive until pH < pHzpc (point of zero charge). Deliyanni et al. (2003) observed the effect of ionic strength of solution on the sorption process of arsenate ions on alkaganeite by varying the concentrations of (0-0.1M KNO3). Observations indicated that as the concentration of KNO3 is increased, the removal of arsenic is improved until saturation, when varying the amount of sorbent. This improvement was attributed to certain depression of negative solid surfaces charges in the alkaline region, caused by the presence of inorganic electrolyte, which enhanced the interaction between surface sites and arsenic oxyanions. These adsorption characteristics of arsenic highly differ in presence of intersecting minerals like phosphorous. British Geological Survey (BGS 2001) calculated sorption of As (V) and As (III) by Hydrous Ferric Oxide (HFO) as a function of arsenic concentrations and pH in 0.01M NaCl background electrolytes. Observations were made in the absence and presence of an equilibrium dissolved phosphate-P concentration of 1 mg l-1. Results indicated that over the pH range of 6 to 8, As (III) sorbs more strongly than As (V), the reverse was seen in the absence of phosphate. Meng et al. (2002) investigated the combined effect of phosphate, silicate and bicarbonate on the adsorption of arsenic by iron hydroxides. The apparent adsorption constants indicated that the affinity of the anions for iron hydroxide sites decreased in the order as arsenate > phosphate > arsenite > silicate > bicarbonate.

PAGE 26

17 Table 3-1: Behavior of arsenic with different minerals Mineral Comment Reference Akaganeit The maximum load capacity was found to be about Deliyanni et. al. (-FeO(OH)) about 100-120 mg As(V) per g of Akaganeite (2003) when .5g l-1 used. The amount of arsenate adsorption increases by lowering the pH, increasing the amount of sorbent and ionic strength of the system. Clay mineral An increased Arsenic content has been noted with the Galba and increasing clay content of the contaminated soils. Polacek, (1973) Hametite & The effect of solute concentration, detention time, pH Singh et. al. Feldspar temperature, and agitation rate of the adsorbent like (1996) Hematite and Felspar. The maximum removal was found to be 100% and 97% with hematite and felspar respectively at optimum conditions. Orange Juice Found that for iron-loaded phosphorylated OJR, maximum Ghimire et. al. Residue (OJR) adsorption capacity for As(V) and As(III) was (2002) 0.94 and 0.91 mol/kg at their optimum pH values 3.1 And 10.0, respectively Activated Uptake of arsenite is much less than that of arsenate Lin and Wu Alumina for AA in most pH conditions. Pore diffusion coefficients (2000) (AA) And tortuosity factors of arsenate and arsenite were interpreted for AA grains with different grain size and different pH. HFO Discussed kinetics and pH dependence of As(V) and As(III) Raven et al. adsorption on HFO Found very high (1998) As(V) and As(III) loadings (up to 4 mol As kg-1) at the highest concentrations. HFO Sorption of As(V) and As(III) on HFO at As Wilkie and concentration of environmental significance (low Hering (1996) micromolar range) and pH 4. SO4 decreased adsorption Of As(V) and As(III), especially at low pH, while Ca Increased As(V) adsorption at high pH. 1mM bicarbonate did not act on either As(V) or As(III) adsorption greatly

PAGE 27

18 Phosphate, silicate and bicarbonate decrease the removal of As (III) even at relatively low concentrations and low surface site coverage (Meng et al., 2002). Effect of temperature as discussed by Pattanayak et al. (2000) showed that higher temperatures could lead to lower metal uptakes. Another easily available adsorbent considered by researchers for arsenic adsorption is activated carbon. Lorenzen et al., (1995) and Pattanayak et al., (2000) used activated carbon and its derivatives with different ash contents for arsenic adsorption. Lorenzen et al., (1995) used peat-based carbon with 5-6% ash content and coconut shell carbon with 3% ash content for arsenic adsorption. Peat based carbon was found to perform better that coconut shell carbon, implying that carbon with higher ash content can perform better. Activated Carbon was found to be less effective and efficient for removal of Arsenic than other adsorbents like activated alumina and iron hydroxide. Laboratory Analysis The two adsorbents considered in this research for use of flux meter were activated alumina and activated carbon. These two adsorbents were chosen because activated alumina has a very high efficiency for arsenic adsorption and activated carbon is one of the most commonly used adsorbent for most of the adsorption based remediation procedures. Adsorption Isotherms Adsorption isotherms define the effectiveness of adsorbents (activated alumina and activated carbon) for the adsorption of arsenic. In this research batch experiments were performed for adsorption isotherm. Most of these adsorption experiments were performed at constant temperature and controlled pH.

PAGE 28

19 Extraction and Analysis: Arsenic extraction can be achieved by using strong acidic or alkaline solutions. Lorenzen et al., (1995) used copper pretreated peat based carbon with a 100-ppm arsenic solution for 24 hrs at a pH of 6 and checked the elution with distilled water, acids and alkalis of different pH at different temperatures and flow rates. The acidic solutions were found to be more effective than the alkaline solutions based on results seen in this research. Singh et al., 1996 observed the effect of effluent flow rate on elution and found that the faster flow rate resulted in lower peak and lower arsenic concentrations at low number of bed volumes and there was a beneficial effect of a higher flow rate towards the end of an elution run. Once arsenic is extracted, it can be analyzed using Graphite Furnace Atomic Adsorption Spectrometer (GFAAS) techniques for concentration range of 5-100 g l-1using EPA method 7060 A, Inductively Coupled Argon Plasma Spectroscopy (ICP) for concentrations above 100 g l-1 using EPA method 3010A and 6010 B, and Spectrophotometer (Bran+Luebbe GmbH) using industrial method 26-71E for concentration range 0.4-20 ppm. GFAAS can also analyze the solid samples directly, though often digestion is performed on solid samples for analysis (Sahuquillo et al. 2003). The advantages of this technique are its low detection limits, the minimum sample manipulation, its relative simplicity and the short time required to obtain the results. Flux Meter Analysis Various scientists and organizations have made an attempt to identify the source of contaminant and degree of contamination in groundwater using different field monitoring techniques (Brusseau and Srivastava 1997, Montas et al., 2000). Brusseau and Srivastava (1997) collected a dense, 3 dimensional array of sampling points to obtain time-series data and spatial distribution which was used to determine the zeroth, first and second

PAGE 29

20 spatial moments of the plume. Montas et al. (2000) presented a methodology for designing groundwater quality monitoring well network in space and time, and to evaluate the performance of the resulting network. However, the two techniques are often time consuming and include various assumptions in the contaminant flux calculations and large number of wells are drilled for correct measurement of contaminant (Montas et al., 2000). Zhang et al. (2002) performed experimental investigation of contaminant transport in coastal groundwater. The experiments were performed in a flow tank and glass beads were used as the homogeneous porous media. The salt water and fresh water intrusion was taken into consideration by incorporating the vertical/horizontal ratio of the sloping seaward boundary keeping freshwater and saltwater at the high and low elevation respectively. In this thesis, University of Florida flux meter are discussed for their use for measuring contaminant flux in groundwater. Flux meter has been validated for most of the organic contaminant analysis both in lab as well as field for contaminant flux, groundwater flow and flow direction. In this research, use of flux meter for inorganic contaminant like arsenic was analyzed. Computer Model Simulation A numerical model simulation is performed using MODFLOW MT3DMS for arsenic contamination in various soil types. Soil classification described in earlier section is used because various scientists working on arsenic contamination and its concentrations in soils of Florida prefer this classification for research (Chen et al., 1999; Clark et al., 2004). The advantages of using MODFLOW include numerous facilities for data preparation, easy exchange of data in standard form, extended worldwide experience, continuous development, availability of source code, and relatively low price (McDonald and Harbaugh, 1988).

PAGE 30

21 Partition coefficients were calculated for different soil types based on arsenic concentration of 48 ppm. The Kd values were calculated using solution to solid ratio of 1:5. They range from 0.2 to 0.8, highest in marls and histosols (soil and water department, University of Florida). Bulk density values were based on the ranges from USDA county soil classification. Porosity calculations are done on the basis of linear relation between porosity and bulk density developed by NASA staff science 2003.

PAGE 31

CHAPTER 4 MATERIALS AND METHODS Various adsorption, extraction and analysis methods were performed for arsenic adsorption onto activated carbon and activated alumina. The adsorption experiments were performed as batch experiments for different time periods with respect to different adsorbents because equilibrium time for each adsorbent is different. Extraction was performed using strong acidic as well as alkaline solution of nitric acid and sodium hydroxide respectively. Simple digestion process explained by EPA method 3010A for extraction of arsenic from soils was used for extracting arsenic from activated alumina using acid. For extraction using strong bases, standard procedure described by water chemicals codex (Glaze et al., 1982) was used. Analysis was performed using GFAAS (Graphite Furnace Atomic Adsorption Spectrometer) for all concentrations since it can measure concentrations as low as 5 ppb (g l-1). For higher concentrations, samples were diluted to bring them within the detection range of 5 100 ppb. Typical flux meter set up is then discussed and various boundary conditions and assumptions involved with the method are described. Adsorption Isotherms Experimental quantities vary significantly for adsorption on activated carbon and activated alumina because of their different efficiencies to adsorb arsenic. Activated Alumina acts as a strong adsorbent whereas activated carbon is a weak adsorbent. In this research, trace metal grade arsenic (Fisher Scientific) was used for analysis. Granular activated alumina and activated carbon used in this research is commercially available 22

PAGE 32

23 (Fisher Scientific). Adsorption experiments of arsenic on activated alumina were conducted using 40 ml glass vials fitted with a Teflon-lined septa screw-top cap. In each experiment .02-.20 g of activated alumina grains were placed into the vial (Lin and Puls, 2000) and 30 ml solution prepared at a predetermined arsenic concentration using de-ion water. The temperature of the system was kept constant, and the reciprocating speed was kept at 70 rpm for all experiments. System was kept on rotator for 48 hrs, for adsorption to come into equilibrium (Lin and Puls, 2000). In adsorption experiment for activated carbon, 40 ml vials and varying amount of activated carbon (1 5 grams) were used (Pattanayak et al., 2000). The reciprocating speed was kept at 70 rpm and system was kept for 72 hrs for adsorption to come into equilibrium (Pattanayak et al., 2000). After rotating at their respective times, activated carbon/aluminum was allowed to settle for 12 hrs. Solution was then filtered for analysis using GFAAS. Since the initial concentration of arsenic in the solution was known and arsenic left in the solution were analyzed, respective calculations for sorbed concentration was performed. Extraction Method Arsenic extraction from activated alumina was performed using strong acid and alkaline solutions using techniques described by EPA method 3010A and water chemical codex respectively. For extraction using acid, a known volume of well-mixed sample was transferred to a 250 ml Griffin beaker, 2 ml of 30% H2O2 and sufficient concentrated HNO3 was added to get an acid concentration of 1% (v/v). It was allowed to digest at 105oC for 2 hrs. The solution was then filtered and injected into the furnace for analysis. For extraction using base, first the extraction solution was prepared by adding 100 ml of 0.1 M potassium hydrogen phthalate to 47 ml of 0.1 N sodium hydroxide. Solution

PAGE 33

24 was then diluted to 200 ml using de-ionized water. 0.1 g of sample was taken into a 250 ml beaker and 50 ml of extraction solution was added to completely digest activated alumina samples. A drop of non-ionic surfactant (Brij 97) was then added and solution was agitated with a magnetic stirrer for 1 hour. This solution was then filtered and analyzed using GFAAS. Analysis Method As discussed above, arsenic analysis was performed using EPA method 7060 A. the detection limit for this method is 1 g/l (1ppb). Samples were analyzed by GFAAS on a Perkin Elmer SIMAA 6000 THGA. The details of parameters used are: Wavelength= 193.7nm Injection size = 20 l Matrix Modifier = 7 l of 700 ppm Pd(NO3)2 These parameters were used on the basis of the guidelines provided by the manufacturer. Because temperature-sensing mechanisms and temperature controllers can vary between instruments or with time, the validity of the furnace parameters were periodically confirmed by systematically altering the furnace parameters while analyzing a standard. In this manner, losses of analyte due to overly high temperature settings or losses in sensitivity due to less than optimum settings can be minimized. 20 micro liter aliquot of sample was injected into the furnace and atomized. If the concentration was found to be greater than the highest standard (100 ppb), the sample were diluted in the acid matrix (suggested by EPA method 7060A) and reanalyzed. Flux Meter Before conducting bench scale flux meter experiments for estimating contaminant flux, the basic instrument was set up for simulating true field conditions. Hence this

PAGE 34

25 section is divided into two, one containing information on set-up of the flux meter and second on running the experiment. Flux Meter Set-up The test aquifer was set up in a box of approximately 46 cm length and 30 cm width. Well casing and screen are constituted by a 3.2 cm inner diameter PVC tube that is slotted over the entire height of the aquifer and that was glued to the bottom of the box to keep it in position. A metal grid was placed 3.8 centimeters from the box walls where the water enters and leaves to avoid intrusion of the aquifer material into glass beads. Glass beads possesses a very high permeability compared to aquifer, which serves to establish a vertical plane of constant head over the entire cross sections where the water enters and leaves the aquifer. The aquifer consists of 20-30 mesh Ottawa sand (hydraulic conductivity, 4.45 cm/min and porosity 0.3) that reaches up to a height of 11.5 cm in the box. The filling of box was performed by adding water simultaneously with the sand in order to obtain a saturated aquifer from the beginning. The boundary conditions of uniform flow field have to be approximated by the bench-scale aquifer. These conditions are constant head at infinity on two opposite sides of the well and straight streamlines going parallel to the other sides of the infinity. To achieve these conditions, the distance from the well screen to the box walls on one side and to the gravel pack on the other side were considered big enough in relation to the well diameter to assume the boundary conditions for a uniform flow field as valid. The water supply of the aquifer was established using a Kimax brand aspirator bottle with tubulation, which was used as a water reservoir that keeps the head at a constant level. As illustrated in Figure 4-1, this was achieved by placing a tube into the

PAGE 35

26 closed container that maintains atmospheric pressure at its bottom. In order to guaranty constant head conditions in the aquifer, the bottom of this tube was elevated to the level of the top part of the aquifer. A constant flow pump on the other side of the aquifer was used to convey a known constant flow through the aquifer, which was then collected in another container to allow control of the actual flow over a certain time. A typical setup of the flux meter is shown in Figure 4.1 (Klammler 2001). Running Experiment Before the experiment can be started, the device has to be prepared. As stated in the previous section, it has to consist of a self-contained permeable unit that intercepts the groundwater flow without retaining it. In this case, 8-14 mesh activated alumina (hydraulic conductivity, 4.82 cm/min, porosity 0.382, particle density of 2.45 kg/l, and bulk density is1.5 kg/l) was used for the permeable media, which took the role of the sorbent for arsenic contamination. The sorbents (activated alumina) in the device sorbs dissolved inorganic solutes (arsenic) present in the water intercepting the unit, thus indicating the amount of contaminant carried by the groundwater. The device was placed in the well, and pump was set to the required flow rate and started. In this case, flow rate was approximately 4.83 mL/min, which result in specific discharges of 0.84 cm/hr for the uniform flow field (cross sectional area A = 420 cm2). The running times was aimed to be 7 hrs hours, thus conveying a total volume of about 1500 mL to 2000 mL of water through the aquifer per run. A control volume at the end of the process is used to quantify the total volume conveyed through the aquifer in order to compute the actual average flow rate over the running time.

PAGE 36

27 q q0 q0 Q 0 Q 0 p o Q 0 Valve Control volume Valve Sorbent media Well Screen Aquifer surface Glass Beads Ottawa Sand Constant flow pump Constant head p po p = p0 p = p 0 Q 0 Figure4-1 Bench Scale set-up of flux meter (Klammler, 2001) Well R Well C Well L 3 in 3 in 3 in 3 in 3.2c m 12 in 1.5 in 15 in 1.5 in Figure 4-2 Cross sectional dimensions of flux meter

PAGE 37

28 Since the hydraulic head at the beginning of the aquifer was set to a certain value by the constant head reservoir and the flow rate was determined by the setting of the pump, the hydraulic head at the end of the aquifer changes with changing flow rates.

PAGE 38

CHAPTER 5 RESULTS AND DISCUSSION Batch experiments for measuring adsorption coefficient for arsenic adsorption onto activated carbon and activated aluminum were conducted at predetermined arsenic concentrations. Strong acid and base extracts were analyzed for extraction efficiency, and GFAAS was used for analysis. Flux meter tests were performed at a flow rate of 0.69 cm/hr to simulate horizontal flow lines and steady state in the box. Adsorption Experiments Batch experiments measuring adsorption coefficient for arsenic adsorption on activated carbon and activated alumina were conducted for constant concentrations of 36 ppb and 17.4 ppm respectively (Pattanayak et al., 2000; Lin and Wu, 2000). These values were used based on experiments conducted at different concentrations used for adsorption of arsenic (Lin and Wu, 2000; Lorenzen et al., 1995). It was observed that activated carbon is a weak adsorbent for arsenic contamination whereas activated aluminum is a strong adsorbent for arsenic contaminant. Hence higher concentrations were used for activated aluminum over activated carbon. Adsorption experiments for arsenic adsorption on activated carbon indicated that activated carbon is not an adequate sorbent for arsenic, therefore, it was not examined for extraction and flux meter experiments. The results obtained from adsorption experiments for arsenic on activated aluminum are shown in Table 5-1. Activated alumina was allowed to adsorb 30 ml of 17.4-ppm arsenic solution for 48 hrs and final concentration of arsenic left in the solution was measured, which is indicated as solution concentration (C) in Table 5-1. The total 29

PAGE 39

30 mass sorbed by activated alumina (total mass mass lest) was calculated. Mass sorbed was then normalized with respect to mass of activated alumina to give sorbed concentration (S). Table 5-1 Adsorption of arsenic on activated alumina Solution Concentration(C) Total Mass Mass Left Mass Sorbed Sorbed Concentration (S) ppm(mg/l) (mg) (mg) (mg) per unit AA mass (mg/g) 8.09 0.52 0.24 0.28 13.89 2.91 0.52 0.09 0.43 7.18 1.58 0.52 0.05 0.47 4.68 1.76 0.52 0.05 0.47 3.41 1.36 0.52 0.04 0.48 2.68 1.33 0.52 0.04 0.48 2.21 The adsorption isotherm (the curve of sorbed concentration Vs concentration in the solution) was analyzed and for the lower range of activated alumina as in this case (0.02g to .2g), a linear isotherm is obtained (Figure 5-1). S = 1.8444CR2 = 0.921602468101214160246810Solution Concentration, C (mg/l)Sorbed Concentration, S (mg/g) Figure 5-1: Adsorption isotherm of arsenic on activated alumina The adsorption isotherm (Figure 5-1) is linear with Kd value of 1.844 L/g. The adsorption isotherm obtained by (Lin and Wu 2000; Singh and Pant 2003) was found to

PAGE 40

31 follow non-linear trend. The difference can be attributed to lower range of solution concentrations used in the analysis which lie on the lower end of the range used by Lin and Wu 2000, and Singh and Pant 2003. However from the data it is evident that activated alumina has a very high efficiency for arsenic and at lower concentrations of arsenic, it can adsorb almost all of arsenic present in the solution. Extraction Experiments Since activated carbon was not found to be a promising adsorbent for arsenic, only extraction of arsenic from activated alumina was considered for the analysis. Extraction using acid and base was analyzed using EPA method 3010A and water chemical codex (Glaze et al., 1982) prescribed method (from materials and methods section). Extraction efficiency was calculated at the ratio of amount of arsenic extracted per unit amount of arsenic sorbed. Extractions performed using strong base, sodium hydroxide and strong acid, nitric acid were compared and it was observed that acidic extraction is more efficient for extracting arsenic from activated alumina. Nitric acid was used for extraction and the results obtained are shown in Table 5-2. A graph of actual amount of arsenic on activated alumina with amount of arsenic extracted using acid is shown in Figure 5-2. Table 5-2 Arsenic extracted using nitric acid Conc. of As in Mass of AA Volume of As per unit mass of AA extracted solution extraction solution Extracted Actual (mg/l) (g) (l) (mg/g) (mg/g) 6.20 0.1016 0.05 3.0512 3.4148 6.19 0.1378 0.05 2.2460 2.6838 6.60 0.1622 0.05 2.0345 2.2125

PAGE 41

32 y = 0.8817xR2 = 0.95980.000.501.001.502.002.503.003.500.001.002.003.004.00Amount sorbed (mg/g)Amount extracted (mg/g) Figure 5-2 Amount extracted using acid vs. initial concentration Graph of amount extracted with respect to actual amount on activated alumina (Figure 5-2) indicated that extraction efficiency of nitric acid is around 88%. Extractions were also performed using strong base, sodium hydroxide. Table 5-3 Arsenic extracted using sodium hydroxide Conc. Of As in Mass of AA Volume of As per unit mass of AA extracted solution extraction solution Extracted Actual (mg/l) (g) (l) (mg/g) (mg/g) 3.54 0.1055 0.0500 1.6777 1.9301 1.96 0.1072 0.0500 0.9142 1.0924 0.96 0.1059 0.0500 0.4514 0.7043 y = 0.8412xR2 = 0.97020.000.501.001.502.000.000.501.001.502.002.50Amount Adsorbed (mg/g)Extracted Amount (mg/g) Figure 5-3 Amount extracted using base vs. initial concentration

PAGE 42

33 The results obtained from extraction using base are shown in Table 5-3. The graph of actual amount of arsenic on activated alumina with amount of arsenic extracted using base (sodium hydroxide) is shown in Figure 5-3. Figure 5-3 indicates that extraction efficiency of sodium hydroxide to extract arsenic is around 84%. However, the results obtained for both types of extraction were similar i.e. efficiency of 85-90%. Flux Meter Bench Scale Experiments The UF flux meter was tested in the lab for its ability to determine arsenic contaminant flux. Dimensionless cumulative mass () for activated alumina was calculated (equation 2.15) from given parameters (section of materials and methods) to be 0.59. Since this is less than 0.6, the linear relation developed by Klammler (2001) was used to calculate mass of imaginary tracer retained in the flux meter (equation 2.15) as ,1.211.22Rcontcontqtm R r (5.1) Since the hydraulic conductivity of activated alumina (4.82 cm/min) is more than the hydraulic conductivity of Ottawa sand (4.45 cm/min), the specific discharge through flux meter was different from that through the surrounding aquifer. Therefore, to calculate specific discharge in flux meter, qo was multiplied with a factor of 1.04 (equation 2.16). Also, from the extraction experiment the amount of arsenic sorbed per unit mass of activated alumina was obtained. If this sorbed mass is denoted by mcont, then the concentration of arsenic in the solution as indicated by flux meter can be calculated as (using equation 2.14, 2.15 and 2.17) 0.6contmCqt r (5.2)

PAGE 43

34 where 2contcontMmLr (5.3) Arsenic was extracted using acid based on EPA method 6010A. Mean and standard deviation of calculated concentration using actual and corrected mass (88% extraction efficiency) was calculated. Results obtained from the flux meter analysis are shown in Table 5-4. From statistical t-test, 95% confidence intervals ( 0.025*tn s where is the sample mean, t is the t-test parameter, s is the sample standard deviation and n is the number of samples takes) for actual and corrected concentrations are 9.79 1.79 and 10.97 2.24 respectively. Arsenic concentration used in the influent solution in flux meter set up was 12ppm, which lies within the 95% confidence interval obtained from the concentration calculations using corrected mass. However, because the degree of freedom for the analysis was low (df = 2, which results in high value of parameter t0.025) confidence interval provides a wide range for concentrations. 0.025 Table 5-4 Concentrations measured from flux meter tests Extracted mass Concentration Calculated Using contm Actual Corrected Actual mass Corrected mass Well Location (g/g) (g/g) (mg/l) (mg/l) Well L 14.25 15.96 9.36 10.48 Well C 16.17 18.11 10.62 11.89 Well R 14.29 16.01 9.73 10.89 Mean 9.79 10.97 Std Dev 0.72 0.81 Since well L and well R are the peripheral wells (Figure 4-2), and due to possible convergence of flow lines; flow through well L and well R might have been lower when compared to the center well. Similarly, the total mass of arsenic passing through the well on sides might have been less than expected, which in turn would result in lower estimated flux concentration. From Table 5-4, calculated concentrations in well L and

PAGE 44

35 well R (Figure 4-2) are less than the concentration calculated in well C. Well C provides a good estimate (estimated value = 11.89ppm) of the actual concentration (12ppm) of arsenic in the solution. This implies that flux meter is a promising tool to assess arsenic contamination in groundwater. Computer Simulation Results Arsenic concentration is plotted with respects to time for varying soil types. The parameters used for the analysis were taken from research through various journals and documents as well as from data collected by University of Florida, soil and water department. Table 5-5: Soil properties Soil Type Silt+Clay Content Bulk Density Kd Porosity Marls 95 1000 0.8 0.62 Histosols 90 600 0.7 0.76 Spodosols 13.7 1300 0.3 0.51 Entisols 3.85 1300 0.2 0.51 Concentration Vs Time012345678901000200030004000Time (days)Concentration (ppm) Marls Histosols Spodosols Entisols Figure 5-4: Concentration of arsenic vs. time for varying soil type. The data used is described in the literature review section of the thesis. Most of this data is obtained from findings of University of Florida, soil and water science department or from the research conducted by Clark et al. (2004). As can be seen from the data,

PAGE 45

36 marls and histosols have comparatively high silt and clay content. There high clay content values reflect that higher values of bulk density should be expected. Conversely, marls and histosols were found to have low bulk densities, and this is due to their higher particle fraction. These fractions are mostly organics or limestone particles and are not clay. Marls are clayey soils and therefore, have a higher efficiency to retain arsenic, which results in minimal leaching. From Marls to Entisols, the soil characteristics change from higher composition of clay to lower composition of clay (and higher composition of sand). From Figure 5-4, one can see that time taken by arsenic to reach maximum concentration decreases with increasing sand content (or decreasing clay content) and decreasing Kd values, hence decrease in dispersion from clayey soil to sandy soil (Marls to Entisols) can be observed. Therefore, break through for entisols was achieved the first, spodosols the second, histosols the third and marls the fourth.

PAGE 46

CHAPTER 6 CONCLUSIONS The purpose of this research was to evaluate the use of flux meter for estimating arsenic contamination in groundwater. The primary objective was to evaluate the capability of activated alumina as a flux meter sorption media for arsenic. Activated alumina was chosen as the potential sorbents because it has a high efficiency for arsenic (Lin and Wu, 2000). From the batch experiments, it was observed that for constant initial arsenic concentration of 17.4 ppm, with varying amount of activated alumina, a linearly increasing isotherm with Kd value of 1.84 L/g is obtained. This showed that with higher concentrations of arsenic in the aqueous phase, higher concentrations were sorbed onto the solid phase. Evaluation for determining a suitable technique for extracting arsenic from activated alumina was explored. Both acid as well as base solutions were used. It was observed that acid has relatively higher efficiency to extract arsenic from activated alumina than a base. It was determined that extraction methods, acid and base, extract around 85-90% of arsenic from activated alumina. The acid and base considered in the research were nitric acid and sodium hydroxide respectively. The acid was used for extracting arsenic from flux meter samples. Bench scale experiments were conducted for evaluating the use of flux meter for estimating arsenic concentration. A continuous flow of 12 ppm aqueous arsenic concentration was maintained through the model aquifer at specific discharge of 0.84 37

PAGE 47

38 cm/hr. The flux meters were placed in three wells located perpendicular to the flow lines in the center of model aquifer, and arsenic adsorbed by activated alumina in each flux meter was measured. Concentrations calculated based on the mass sorbed by activated alumina were compared with the aqueous concentration used in the aquifer (12ppm). The statistically estimated range of aqueous concentration, calculated on the based of arsenic extracted from flux meter, provided the 95% confidence interval of 10.97 2.24 ppm. The flux meter in central well estimated 99% of the actual concentration in aquifer. The results obtained from flux meter proved it to be a promising device for assessing arsenic contamination in groundwater. Potential applications for this method are measuring arsenic fluxes for quantifying risk to human from arsenic leaching through hazardous waste sites, or from source zones before and after remedial efforts. Future research can be conducted to evaluate compatibility of various available tracers with activated alumina for fluid flow determination.

PAGE 48

LIST OF REFERENCES Alam M. G. M., Snow E. T. and Tanaka A. 2003. Arsenic and heavy metal contamination of vegetables grown in Samta village, Bangladesh. The Science of The Total Environment. 308: 83-96. Azcue, J. M. and Nriagu, J. O. 1995. Impact of abandoned mine tailing on the arsenic concentrations in Moira Lake, Ontario. Journal of Geochemical Exploration. 52: 81-89. Barry D. A., Prommer H., Miller C. T., Engesgaard P., Brun A. and Zheng C. 2002. Modeling the fate of oxidisable organic contaminants in groundwater. Advances in Water Resources. 25: 945-983. British Geological Survey (BGS) 2001. Groundwater studies for arsenic contamination. Report WC/00/19. Volumes 1-10. Brusseau, M., Srivastava R. 1997. Nonideal transport of reactive solutes in heterogeneous porous media: 2. Quantitative analysis of the borden natural-gradient field experiment. Journal of Contaminant Hydrology. 28(1-2): 115-155. Bull D. C. 2000. The chemistry of chromated copper arsenate, characteristics of timber treatment plant chemical sludge. Wood Science and Technology. 34:367-376. Chakravarty S., Dureja V., Bhattacharyya G., Maity S. and Bhattacharjee S. 2002. Removal of arsenic from groundwater using low cost ferruginous manganese ore. Water Research. 36:625-632. Charbeneau R. J. 2000. Groundwater hydraulics and pollutant transport. Prentice Hall, Inc. Upper Saddle River. NJ. Chen M., Ma L. Q. and Harris W. G. 2002. Arsenic concentrations in Florida surface soils: influence of soil type and properties. Soil Science Society of American Journals. 66:632. Chen, M., Ma, L.Q. and Harris, W.G. 1999. Baseline concentrations of 15 trace elements in Florida surface soils. Journal of Environmental Quality. 28:1173-1181. Chirenje T., Ma L. Q., Clark C. and Reeves M. 2003. Cu, Cr and As distribution in soils adjacent to pressure-treated decks, fences and poles. Environmental Pollution. 124:407-417. 39

PAGE 49

40 Clark J. C., Chirenje T., Annable M. D., Hatfield K. 2004. Developing retention indices and modeling transport of CCA in Florida soils at unlined landfills. Florida center of solid and hazardous waste management. Final Project Report. Day, M. J., Reinke, R. F. and Thomson, J. A. M. 2001. Fate and transport of fuel components below slightly leaking underground storage tanks. Environmental Forensics.2: 21-28 Deliyanni E. A., Bakoyannakis, D. N., Zouboulis A. I. and Matis K. A. 2003. Sorption of As(V) ions by akaganeite-type nanocrystals. Chemosphere. 50:155-163. Domenico P.A. and Schwartz F. W. 1990. Physical and Chemical Hydrogeology. John Wiley and Sons, Inc. New York City. New York Florida Department of Environmental Protection (FDEP) 2004. New arsenic standards applicable to both domestic and industrial wastewater facilities. Chapter 62-550, Florida Administrative Code (F.A.C.) Galba J and Polacek S. 1973. Sorption of arsenates under kinetic conditions in selected soil types. Acta Fytotech. 28: 187 Ghimire K. N., Zhu1 Y., Yano M., Makino K. and Miyajima T. 2002. Effective use of orange juice residue for removing heavy and radioactive metals from environment. Geosystem Engineering. 5(2): 31-37. Ginn T. R., Wood B. D., Nelson K. E., Scheibe T. D., Murphy E. M. and Clement T. P., 2002. Processes in microbial transport in the natural subsurface. Advances in Water Resources. 25:1017-1042 Halter W. E. and Pfeifer H. R. 2001. Arsenic (V) adsorption onto -Al2O3 between 25 and 70C. Applied Geochemistry. 16(7-8): 793-802. Hatfield K., Annable M., Cho J, Rao P. S. C. and Klammler H. 2003. A direct passive method for measuring water and contaminant fluxes in porous media. Journal of Contaminant Hydrology. In press. Jones K. D. and Huang W, 2003. Evaluation of toxicity of the pesticides, chlorpyrifos and arsenic, in the presence of compost humic substances in aqueous systems. Journal of Hazardous Materials. 103: 93-105. Kayajanian G. 2003. Arsenic, cancer, and thoughtless policy. Ecotoxicology and Environmental Safety. 55: 139-142. Klammler H. 2001. Contaminant hydrology A new method for measurement of groundwater and contaminant flux. PhD Dissertation. Technical University Graz, Austria

PAGE 50

41 Leeuwen, F. X. R. 2000. Safe drinking wate r: the toxicologist's approach. Food and Chemical Toxicology. 38:S51-S58 Lin T. F. and Wu, J. K. 2000. Adsorption of arsenite and arsenate within activated alumina grains: equilibrium and kinetic s. Water Resource (Oxford). 35: 2049-2057. Lin Z and Puls R. W. 2000. Adsorption, deso rption and oxidation of arsenic affected by clay minerals and aging proce ss. Environmental Geology. 39: 753-759. Lorenzen L., Deventer J. S. J. van a nd Landi W. M. 1995. Factors affecting the mechanism of the adsorption of arsenic species on activated carbon. Minerals Engineering. 8:557-569. McDonald, M.G. and Harbaugh A.W. 1988. A modular three-dimensional finitedifference ground-water flow m odel. USGS TWRI. 6-A1: 586 Meng X., Korfiatis G. P., Bang S., Bang K. W. 2002. Combined effects of anions on arsenic removal by iron hydroxide Toxological Letters. 133: 103-111. Montas H. J., Mohtar R. H., Hassan A. E. and Alkhal F. A. 2000. Heuristic spacetime design of monitoring wells for contaminan t plume characterization in stochastic flow fields. Journal of Contaminant Hydrology. 43(3-4): 271-301. National Aeronautics Space Administration (NASA) Staff Science 2003. (http://wwweosdis.ornl.gov /FIFE /Datasets /Soil_Propert ies /Soil_Survey_Ref.html). Last seen on September 2003. Nimick, D. A., Moore J. N. Dalby C. E. and Savka M. W. 1998. The fate of geothermal arsenic in Madison and Missouri rivers Montana and Wyoming. Water Resource Research. 34: 3051-3067. Pattanayak J., Mondal K., Mathew S., Lalvani S. B. 2000. A parametric evaluation of the removal of As (V) and As(III) by car bon based adsorbents. Carbon. 38: 589-596. Raven K. P., Jain A. and Loeppert, R. H. 1998. Arsenite and arsenate adsorption on ferrihydrite: kinetics, equilibrium and adso rption envelopes. Environmental Science and Technology. 32: 344-349. Sahuquillo A., Rauret G., Rehnert A. and Munt au H. 2003. Solid sample graphite furnace atomic adsorption spectroscopy for supporti ng arsenic determinat ion in sediments following a sequential extraction proce dure. Analytica Chemica Acta. 476:15-24. Salas J. and Ayora C. 2004. Groundwater chem istry of the Oklobondo uraninite deposit area (Oklo, Gabon): two-dimensional reac tive transport modeling. Journal of Contaminant Hydrology. 69:115-137. Singh D. B., Prasad G. and Rupainwar D. C. 1996. Adsorption technique for the treatment of As(V) rich effluent s. Colloids and Surfaces. 111: 49-56.

PAGE 51

42 Singh T. S. and Pant K. K. 2003. Equilibrium, kinetics and thermodynamic studies for adsorption of As (III) on activated alumin a. Separation Purification Technology. 36:139:147. Smedley P. L. and Kinniburg D. G. 2002. A review of the source, behavior and distribution of arsenic in natural waters. Applied Geochemistry. 17: 517-168. Smedley P. L and Kinniburgh D. G. 1996. Arse nic contamination of groundwater in bangladesh, Sorption and Transport. Br itish Geological Survey Report WC/00/19. Volume 1. Solo-Gabriele H., Townsend T.G. 1999. Disposal practices and management alternatives for CCA-treated wood waste. Wast e Manage. Residential. 17:378-389. Solo-Gabriele H. M., Townsend T. G. and Me ssick B. 2002. Characteristics of chromated copper arsenate-treated wood ash. Journa l of Hazardous Materials. B89: 213 Townsend T., Stook K., Tolaymat T., Song J.K., Solo-Gabriele H., Hosein N., and Khan B. 2000. New lines of CCA-treated wood res earch: in-service and disposal issues. Florida Center for Hazardous Waste. Technical Report. United States Environmental Protection Agency (USEPA) 2001. National primary drinking water regulations; arsenic and clarif ication to contaminants monitoring. DOCID:fr22ja01-29. 66:6975-7066. Wang L., Chen A. and Fields K. 2000. Arsenic removal from drinking water by ion exchange and activated alumina plan ts. EPA Publication EPA/600/R-00/088. Glaze W. H., Buescher C. A. and Mahon J. H. 1982. Water chemical codex. Activated alumina. National Academy press. Washington D. C. pp 57-58. Wilkie, J. A. and Hering J. G. 1996. Adsorp tion of arsenic onto hydrous ferric oxide: effect of adsorbate/adsorbent ratios a nd co-occuring solutes. Colloidal Surfaces. 107:97-110. Zhang, W., Cai Y., Tu C., and Ma. L.Q. 2002. Arsenic speciation and distribution in an arsenic hyper accumulating plant. Scie nce of Total Environment. 300:167-177. Zheng, C. and Bennett, G.D. 2002. Applied cont aminant transport modeling; John Wiley and Sons, Inc., New York City. New York.

PAGE 52

BIOGRAPHICAL SKETCH Prachee Gupta was born in 1980 in Rajasthan, India. She received a Bachelor of Technology (B.Tech) from the Indian Institute of Technology (IIT), Mumbai, in 2002. In the fall of 2002, Prachee joined the University of Florida for a Master of Science in the Department of Civil and Coastal Engineering under the tutelage of her major professor, Dr. Clayton J. Clark, II. 43