Removal of Polycyclic Aromatic Hydrocarbons and Heavy Metals from Stormwater using Granular Activated Carbon

This study examines the removal of polycyclic aromatic hydrocarbons and heavy metals from stormwater using granular activated carbon. The experiment found that effective removal of heavy metals occurs at filtration velocity of 5, 10 and 11.5 m/h when a 100 cm height of granular activated carbon was used as the filter column. The filtration was enhanced if a pre-treatment was added to the 100 cm height anthracite filter column at higher velocities of filtration of 10 and 11.5 m/h. The removal efficiency order for the solution at pH of 6.5-7.2 of the batch and column experiments for the single and mixed metals was Pb, Cu>Zn. The study also explores the kinetics of PAHs on granular activated carbon.

ABSTRACT
The discharge of pollutants if storm water has the potential to damage the aquatic and terrestrial
environments if the discharge is not treated before being discharged. The removal of polycyclic
aromatic hydrocarbons and heavy metals from stormwater were examined using batch and fixed-
bed experiments. Studies from the filed illustrated that effective removal of heavy metals occurs
at filtration velocity of 5, 10 and 11.5 m/h when a 100 cm height of granular activated carbon
was used as the filter column. The filtration was enhanced if a pre-treatment was added to the
100 cm height anthracite filter column at higher velocities of filtration of 10 and 11.5 m/h. The
removal efficiency order for the solution at pH of 6.5-7.2 of the batch and column experiments
for the single and mixed metals was Pb, Cu>Zn. The order has a strong correlation with the
solubility product and the constants of the first hydrolysis of the hydroxides of the metals.
Homogenous particle diffusion and shell progressive models allow calculation of the average or
the mean coefficient of the intraparticle diffusion for the case of high levels of naphthalene and
the coefficient of mass transfer in cases of low ranges concentration of the organic molecules.
The shell progressive mechanism and Fick’s law are among the excellent approaches for the
extraction of kinetics of PAHs on granular activated carbon. The experiment illustrated that
granular activated carbon filter is effective and reliable in the removal of heavy metals and
polycyclic aromatic hydrocarbons from stormwater.

ACKNOWLEDGEMENTS
The preparation of this thesis is a demanding task that calls for maximum dedication and
cooperation from various significant people and institutions. Listing all of them might be
impossible. First and foremost I thank the Almighty God for having granted me the life and
energy to enable me get to this far. May power and glory go back to Him who dwells above. I
would like to also thank my supervisor(s) who dedicated a lot of the quality time and
perseverance towards the preparation of this thesis proposal. I am equally humbled to thank them
sincerely for every positive immediate responses, guidance and encouragement that has seen me
get to this level of this task this time. May I also thank the University as well for their faithful
commitment and moral support.

TABLE OF CONTENTS
ABSTRACT................................................................................................................................................................. 1
ACKNOWLEDGEMENTS....................................................................................................................................... 2
TABLE OF CONTENTS........................................................................................................................................... 3
LIST OF FIGURES.................................................................................................................................................... 4
LIST OF TABLES...................................................................................................................................................... 5
ABBREVIATIONS..................................................................................................................................................... 6
SYMBOLS.................................................................................................................................................................... 7
CHAPTER 1: INTRODUCTION............................................................................................................................ 8
CHAPTER 2: LITERATURE REVIEW................................................................................................................ 9
CHAPTER 3: EXPERIMENTAL METHODOLOGY...................................................................................... 14
Granular activated carbon anthracite..................................................................................................... 14
Heavy metals..................................................................................................................................................... 14
Stormwater........................................................................................................................................................ 15
Field fixed-bed column experiments....................................................................................................... 16
Laboratory batch experiments................................................................................................................... 17
Laboratory fixed-bed column experiments.......................................................................................... 18
Naphthalene....................................................................................................................................................... 20
Reagents and solutions................................................................................................................................. 20
Naphthalene analysis..................................................................................................................................... 21
Batch kinetics experiments......................................................................................................................... 21
SEM analysis...................................................................................................................................................... 22
Adsorption Studies.......................................................................................................................................... 22
Theoretical models of kinetics................................................................................................................... 23
CHAPTER 4: RESULTS AND DISCUSSION................................................................................................... 24
Laboratory batch experiments................................................................................................................... 24
Adsorption of heavy metal by granular activated charcoal............................................................24
Individual metals......................................................................................................................................... 24

Mixed metals................................................................................................................................................. 25
Laboratory fixed bed column experiments........................................................................................... 26
Removal of heavy metals......................................................................................................................... 26
Feasibility study and optimum experiment conditions...................................................................28
Analysis in the intraparticle diffusivity mechanisms........................................................................ 30
Kinetics sorption models for non-polar organic micro pollutants..............................................31
Results for PAH analysis using UV-vis Spectrophotometry...........................................................33
The shell progressive model....................................................................................................................... 37
The homogenous particle diffusion model............................................................................................ 38
Factors affecting the adsorption process for naphthalene.............................................................39
Temperature................................................................................................................................................. 39
Solubility......................................................................................................................................................... 40
pH...................................................................................................................................................................... 40
Surface area ad size of the particle...................................................................................................... 40
CHAPTER 5: CONCLUSION............................................................................................................................... 42
References............................................................................................................................................................... 44

LIST OF FIGURES
Figure 1: GAC filtration set-up schematic diagram………………………………………..…. 20
Figure 2: Concentrations of metals used in the experiment ……………………….….….……. 26
Figure 3: Efficiencies of the removal of mixed metals at different concentrations ….…..…..… 27
Figure 4: Curves for the column breakthrough for the three metals tested …………..…………28
Figure 5: Evolution of the concentration of naphthalene with respect to time ………….………29
Figure 6: Scanning examinations of the F400 …………………………………..……….…….. 31
Figure 7: Structure of Naphthalene …………………………………………………….………. 32
Figure 8: PAH analysis using UV-vis Spectrophotometry ………………….…………….…… 35
Figure 9: PAH analysis using UV-vis Spectrophotometry at 0 GAC ……………………...……36
Figure 10: PAH analysis using UV-vis Spectrophotometry at 0.1gm GAC ……….……………36
Figure 11: PAH analysis using UV-vis Spectrophotometry at 0.25gm GAC ……………….…..37
Figure 12: PAH analysis using UV-vis Spectrophotometry 0.75gm GAC ……………...…….. 37
Figure 13: PAH analysis using UV-vis Spectrophotometry 1gm GAC …………………..…… 38

LIST OF TABLES
Table 1: Physical properties of granular activated carbon and anthracite ……………….…….. 15
Table 2: Summary of the characteristics of stormwater ……………………………………….. 17
Table 3: Levels of concentration of the metals used in the experiment …………….……..…… 25
Table 4: Results for PAH analysis using UV-vis Spectrophotometry ………………………..... 33

ABBREVIATIONS
GAC Granular Activated Carbon
PAH Polycyclic Aromatic Hydrocarbons
HDM Homogenous Particle Diffusion Model
SPM Shell Progressive Model
SEM-EDS Scanning Electron Microscopy with Energy Dispersive System

SYMBOLS
∆ H Change in Enthalpy
Keq Coefficient of the equilibrium partition at a specific temperature
∆ S Change in the entropy
R Gas constant
T Absolute temperature
De Effective diffusion coefficient
r radius of sorbent particles
Pb Lead
Zn Zinc
Cu Copper

CHAPTER 1: INTRODUCTION
Water bodies such as lakes and rover are mainly polluted by urban stormwater which usually
carries with it large quantities of both organic and inorganic pollutants in dissolved and solids
states such as heavy metals dissolved organic carbon as well as suspended solids. The effects on
these solids on the environment when discharged if untreated are such devastating that they
cannot be ignored (Alley, 2007, p.158). Suspended solids have a significant impact on making
the water turbid, lessening the penetration of sunlight as well as creating temperature variations
all of which can lead to reducing the growth and activity of photosynthetic organisms. The result
is higher water treatment costs, decline in the resources of fish, aesthetic issues, reduced
longevity of reservoirs and dams, devastating degradation of the ecology of aquatic life besides
reduced navigability of channels.
Suspended solids on the other hand serve as carried to pollution since pollutants including
pesticides, nutrient, heavy metals and organic matter are easily transported by the solids in
particulate form which then are later introduced into the surrounding. Dissolved organic carbon
results in deficiency of oxygen in water and hence causing death of aquatic organisms. Still,
dissolved organic carbons have been found to cause unpleasant taste and smell in water besides
acting as a substrate for the growth of microbes (Gupta, 2012, p.122). Dissolved organic carbons
have the capability of reducing the effectiveness of the processes of water treatment and increase
the demands of disinfections and coagulant and foul membrane filters.

CHAPTER 2: LITERATURE REVIEW
Studies and research have established the existence of high levels of contaminations of heavy
metals including copper; zinc and lead in the urban stormwater which have a significant impact
on the pollution of the water bodies that receive the water. Such heavy metals more specifically
lead and mercury have been listed among the top seven most hazardous chemicals by the by the
US Agency for Toxic Substances and Disease Registry (Bahadori, 2013, p.201). Studies have
reported the adsorptive removal of organics and inorganics contaminants from stormwater. The
studies are however limited to static batch studies as opposed to the dynamic fixed-fed column
studies which have found more relevant in the actual operations of systems when it comes to
stormwater treatment.
The discharge of stormwater pollutants into the aquatic environment without being treated poses
a high level of thereat to the aquatic life. This thus raises the need to eliminate suspended solids,
dissolved organic carbon as well as heavy metals from the storm water as one of the treatment
stages to ensure that the level of harm is greatly reduced if not fully eliminated at the point of
distribution of the treated wastewater. Numerous approaches and techniques have been proposed
and/or are currently under applications in the removal of harmful chemicals and substances from
contaminated water. Oxidation with the help if ozone, carbon absorption, high-energy electron
radiation as well as advanced oxidation process have been considered as some of the highly
effective and promising alternatives for use in the elimination of pollutants from wastewater
(Bingham, 2012, p.214).
Polycyclic aromatic hydrocarbons (PAHs) have been found to cause both carcinogenic and
mutagenic effects in the human body and biota. The compounds are very common in the

atmosphere and the environment due to their stable chemical structure as well as low
bioavailable fraction. Based on their structures, Polycyclic aromatic hydrocarbons are classified
as either low molecular weight or high molecular weight where the low molecular weight
Polycyclic aromatic hydrocarbons have two and three rings structures. High molecular weight
structure of Polycyclic aromatic hydrocarbons have four and above rings. The solubility of
polycyclic aromatic hydrocarbons is influenced by their molecular weight in which an increase
in the molecular weight reduces the level of solubility (Chowdhury, 2013, p.233).
The carcinogenetic of Polycyclic aromatic hydrocarbons increases with an increase in the
molecular mass. As a result of their toxic, mutagenic and carcinogenic nature, Polycyclic
aromatic hydrocarbons have gained the attention and the interest of researchers and governments
in the establishment of the most appropriate removal processes. The removal process is one that
requires an elaborate understanding of the mechanisms involving remediation. Numerous
methods have been established and adopted in the treatment of polycyclic aromatic hydrocarbons
from polluted waters and soil in a bid to mitigate the potential risks of PAHs to the environment
as well as on the human health (Franklin L. Burton, 2013, p.187).
Thermal, biological, chemical, physical and even phytoremediation process have been used as
the major treatment methods of waters and souls contaminated with PAHs. Unfortunately, most
of these methods have been associated with drawbacks and disadvantages among them high
investment and maintenance costs and sophisticated operation procedures rendering their use less
reliable and inefficient (Figueiredo, 2009, p.211).
Still, some of these processes have been found to yield the secondary by-products some of which
are mutagenic and carcinogenic for example trihaomethanes. These by-products further top up

on to the list of adverse effects of PAHs on public health. Sorption using various media such as
granular activated charcoal as alternative methods of treating waters polluted with PAHs has
been found to be reliable and promising techniques. Such techniques are not only economical but
also environmental friendly (Crini, 2010, p.175).
Granular activated carbon filters have their applications in the final step of polishing in the
treatment of drinking water in the elimination of compounds which in most cases are not present
in high levels of concentration such as algae toxins, industrial micro pollutants, odors, tastes and
pesticides. Still, the use of granular activated carbon has been acknowledged as an effective way
of eliminating organic materials that occur naturally as it breaks down animal and plant mater in
the environment (Dotro, 2017, p.415). Granular activated carbon treatment process can result
into bacterial colonization of GAC1-10 which is assumed to be resulting in part from the
following:
 Neutralization of the compounds of stressor
 Adsorptive properties which enhance the enrichment of nutrients and the concentration of
oxygen. The properties also help in the elimination of disinfectant compounds.
 The presence of a range of functional groups on the surface of carbon which facilitate the
attachment of microbes on its surface (Erickson, 2013, p.218)
 The porous nature of the surface of carbon which offers a protective environment from
shear forces of the fluid.
In regard of the above mentioned possible reasons for bacterial colonization, the bacteria that
stick to the surface of the particles of carbon are mostly resistant to infections. Nevertheless, the
existence of such microorganisms in the beds of carbon during the process of water treatment has

effects that could be of beneficial impacts. A possibility of prolonging the carbon-bed life in the
process of biological activated carbon through conversion of a portion of the recalcitrant
organics to biodegradable organics using preozonation is one such benefits of establishment of
microorganisms in the carbon beds (Dotro, 2017, p.278).
The biodegradable portion is then converted to biomass, waste products and carbon dioxide by
the microorganisms that are attached to the granular activated carbon. The material then occupies
the sites of adsorption on the granular activated carbon. The density of the surface charge can be
changed by the biofilm that forms on the surface of the activated carbon and thus an increase in
its negative value. This in turn facilitates its capacity of adsorption against some of the positively
charged pollutant species. These positively charged species include among them most of the
heavy metals (Hahn, 2012, p.322).
All the above consideration is indicative of the how important bacterial adsorption process on
activated carbons is in the process of water treatment. This study aims at exploration of the
process of adsorption of an enteropathegonic bacteria on activated carbons under varied
experimental conditions in a bid to determine which are the parameters used in determining the
rate of the process. The study as well aims at investigating the modifications of carbons
following the adsorption by the bacteria and the effect of the bacteria on the capacity of
adsorption of the carbon as far as the elimination of heavy metals such as zinc; copper and lead
from aqueous solutions are concerned (Hlavinek, 2011, p.310).
The purpose of this study was to evaluate how effective granular activated carbon is in the
removal of a range of pollutants from stormwater by laboratory column filtration and fixed bed
field and hence helping in the prevention of the contaminant from being carried to the waterways

and enhancing the likelihood of recycle and recuse of stormwater (Horwatich, 2008, p.420). The
categorical contaminants of interest and focus in this study include suspended solids, dissolved
organic carbons, heavy metals among them lead, zinc and copper as well as polycyclic aromatic
hydrocarbons (PAHs) in stormwater most specifically naphthalene.
Naphthalene was chosen as the preferred polycyclic aromatic hydrocarbon for the study due to
its common presence in stormwater and its challenging proportions in the stormwater on
comparison to other polycyclic aromatic hydrocarbons. Natural stormwater was used in the
analysis of the removal of dissolved organic carbons and suspended solids in the field
experiment that used anthracite and granular activated carbon columns on series each having
different concentrations (Scholz, 2006, p.425). The concentration of heavy metals in stormwater
was very low and hence the study of their removal was not studied in the field. The experiment
on heavy metals was instead done in the laboratory in which tap water was spiked with heavy
metals so as to test the ability of heavy metals, dissolved organic carbons removal and turbidity
through the use of granular activated carbon.

CHAPTER 3: EXPERIMENTAL METHODOLOGY
Granular activated carbon anthracite
Granular activated carbon anthracite that was used in this experiment and study were extracted
from James Cummins P/L, Australia. The table below provides a summary of the selected
properties of Granular activated carbon anthracite (Riffat, 2012, p.188). A zetasizer nano-
instrument was used to measure the zeta potential that is correlated to the surface charge of the
adsorbent in the granular activated carbon.
Table 1: Physical properties of granular activated carbon and anthracite
Properties Granular activated carbon Anthracite
Maximum moisture content (%) 5 -
Nominal size (mm) 0.3-2.38 1.0-1.1
BET surface area (m2/g) 750 -
Uniformity coefficient - 1.30
Speed gravity - 1.45
Bulk density (kg/m3) 748 660-720
Iodine number (mg/g.min) 800 -
Acid solubility (%) - 1
The instrument generated replicate measurements for each of the samples after which a mean
value was generated. The zeta potential as measured at pH 3 to 10 upon adjusting the pH of the
100 ml suspension of the adsorbent in deionized water and having it agitated at 120rpm for a
period of 6 hours.
Heavy metals
Analar grase nitrate salts of heavy metals among them zinc lead and copper were used in the
study.

Stormwater
The stormwater that was used in the experiments of the field studies was mainly extracted from
the base flow in a plant that harvests stormwater that is located in Sydney. The plant had a
continuous flow of stormwater in the stormwater canal on every rainfall events. Measurements of
the amount of rainfall on the first, second and third days were taken and found to be 19, 13 and
84 mm respectively from the nearby weather station that was proximate to the site (Nobutada
Nakamoto, 2014, p.313). Through a sump pit, the stormwater was drained to the stormwater
canal floor by gravity to an adjacent wet wall.
The water was then pumped into a collection ta through a control valve pit. From this point, the
stormwater was continuously fed to the fixed-bed adsorbent columns at a controlled velocity.
The velocity was controlled through the use of valves that were placed before and after the filter
columns (Negulescu, 2011, p.214). The characteristics of the stormwater are summarized in table
2 below. Foe the case of laboratory experiments and studies, synthetic stormwater was prepared
using tap water to generate a mix of heavy metals that bore the specific required turbidity and the
concentrations of the metals.
Table 2: Summary of the characteristics of stormwater
Parameter Unit Value
Physical and chemical properties
pH - 6.68-7.28
TOC mg/L 4.25-8.96
True color PtCo 18-270
Water Hardness mg/L CaCO3 equivalent 30-95
Turbidity NTU 1.5-370
Bicarbonate mg/L CaCO3 equivalent 22-145
Metals
Cu mg/L 0.008-0.049
Pb mg/L 0.001-0.022
Zn mg/L 0.001-0.004

Field fixed-bed column experiments
Granular activated carbon was the main media that was used in the experiments that were carried
out in dynamic adsorption conditions. The figure below illustrates a schematic diagram of the
various media filters that were setup in the field. PVC columns of internal diameter 10 cm were
used in carrying out the field experiments with one of the columns packed with anthracite and
the other two packed with granular activated carbon to a height of about one meter (Kazner,
2012, p.177). Two setups for filtrated were used in the experiment as shown in the figure: a
single granular activated carbon column and anthracite filter column that was preceded by a
granular activated carbon column.
The need to test the necessity of pre-treatment at different loadings of the hydraulic necessitated
the use of anthracite filter column before the use of granular activated carbon. High loadings of
suspended solids on the granular activated carbon filters may interferes with the ability of the
granular activated carbon filters to effectively eliminate other pollutants such as organics that
may be present in the stormwater through clogging its active pores. Cost considerations and its
capability to be able to sufficiently remove suspended solids are among the reasons why
anthracite was chosen as the filter material for pre-treatment (Mannina, 2017, p.222).
The rates of flow of stormwater through the columns were 5 m/h, 10m/h and 11.5 m/h for 1-2 d,
3-4 d and 5-6 d respectively. The empty bed contact times were 12, 6 and 5.2 minutes at these
flow rates. The columns were backwashed using tap water for about one minute at the end of
each of the day of operation that lasted between 4 to 6 hours of filtration. The columns were then
kept in a moist state for 18 hours until the commencement of the filtration experiments the next
day.

A collection of the influent and effluent samples from every column was done at intervals of 5,
10, 15 and 30 minutes and later after every hour. A Hach Model 2100P Turbidimeter was used in
measured the turbidity of a portion of each of the samples. The samples were then filtered
through the use of 0.45μm filter disks (Microfilms, 2008, p.315). A Multi N/C 2000 analyzer
was used in taking the measurements of the dissolved organic carbon present in the samples of
the stormwater.
Laboratory batch experiments
In order to obtain the data and information on the adsorptive properties of every heavy metal on
granular activated carbon, batch adsorption experiments were carried out under statistic
conditions in closed systems. In comparison to granular activated carbon, anthracite has
properties that are more inert toward heavy metals hence it as not tested (Jakeman, 2016, p.166).
The initial concentration of metals that was used in the experiment was 5 mg/L and a range of
doses of adsorbent from 0.1 to 7.5 g/L were used in offering a range of equilibrium concentration
as well as adsorption capacities of the metals. The range was from low to high values of the
adsorbent. The initial pH of the samples was kept at 6.5 in a bid to simulate a pH that is close to
the pH of natural stormwater.
100 ml of metal solutions were thoroughly mixed with the specified adsorbent dose and agitated
at 120 rpm I a flat shaker in the adoption experiments. This was done for 24 hours at room
temperature and the ionic strength of the solution was maintained at 10-3 M NaNO3. Filter disks
having 0.45 um openings of the pore were then used to filter the suspensions and the
concentrations of the heavy metals in the filtrate analyzed (Shafer, 2012, p.318). An atomic
absorption spectrophotomer was used in measuring heavy metals in the suspension. The
calculation of the amount of absorption of the heavy metals at equilibrium was done by

subtracting the quantity of metals in the solution at equilibrium from the quantity of the metals
added with the aims of the formula below:
qe=(Co−Ce)V
M where Co is the initial concentration of the heavy metals in mg/L, Ce, the
equilibrium concentration of the heavy metals in mg/L, V=the volume of the solution in litres
and M the mass of the granular activated carbon. The formula
Removal Efficieny ( % ) = Co −Ce
Co
× 100 % is used in estimating the efficiency of the removing the
heavy metals using granular activated carbon. Repeats of the experiment were done for the
multi-metals so as to determine the effect of co-existence of ions of heavy metals on each of the
metals being eliminated (Yung-tse, 2012, p.289). The concentration of zinc, lead and copper
were 10.0, 5.0 and 3.0 mg/L respectively. The concentrations of the heavy metals that were used
in the experiment were found to be ten times the maximum achievable concentrations of the
Australian stormwater. The doses of the adsorbent that were used ranged from 0.5 to 10 g/L.
Laboratory fixed-bed column experiments
Illustrated in figure 2 below is a schematic diagram of the filtration units of granular activated
carbon. The granular activated carbon is initially packed into a transparent acrylic column of
diameter 2 cm to a height of 90 cm. deionized waster was then passed through the column for 5
minutes upwards so as to remove air present within the pores of the particles. Using synthetic
stormwater, filtration experiments were carried out with filtration velocity maintained at 5 m/h
with the use of two peristaltic pumps in the gravity flow mode (Shi, 2011, p.188). One of the
filtration experiments was conducted just before the water entered the column while the other
after the water leaves the column. The empty bed contact time at the 5 m/h filtration velocity was
10.8 minutes.

Figure 1: GAC filtration set-up schematic diagram
The concentration of heavy metals in the influent natural stormwater was found to be very low in
comparison with the average concentrations that were established in the stormwater across the
city of Sydney (Sperling, 2005, p.314). It is for this reason that synthetic stormwater prepared by
spiking tap water with heavy metals was used to generate the required concentration of the
various heavy metals used. The concentrations of the heavy metals were 2.0, 1.0 and 0.6 mg/L
for zinc, lead and copper respectively. These concentrations were about double the maximum
concentrations established in the Australian stormwater through research.
Kaolinite of 7 mg/L concentration was used to simulate the turbidity in water. This simulation
generated the required turbidity which was the average f the turbidity values of the stormwater
across the city. Since the water was already contaminated with dissolved organic carbon, it was
not spiked with any organics. The level of dissolved organic carbon in the tap water was 5.1
mg/L, a value that was proximate to the levels of dissolved organic carbon in the field.

Collection of the samples was done at 10 and 30 minutes and thereafter after every hour for 8
hours and then with a low frequency up to 120 hours (Shifrin, 2014, p.458).
Analysis of the effluent samples was then done for dissolved organic carbons, heavy metals, pH
and turbidity. The cumulative column adsorption capacity for ant feed concentration and
filtration velocity was determined by determining the area under the plot of the adsorbed
concentration of metal, Cad (Cad=Co-Ce) with time (Strande, 2014, p.256). The breakthrough
curves were used achieving the adsorption capacity as per the equation below where Q is the
flow rate of the solution in L/min
q total= Q
1000 ∫
t =0
t =120h
Cad dt
Naphthalene
Reagents and solutions
The sample of GAC for use in this study was provided by Aguas de Lavante (Spain) which was a
simple that was specifically and specially made to the treatment of organic pollutants, odor,
color, tastes as well as micro molecules. The BET surface area for the granular activated carbon
sample was 1000m2/g while the macro porous volume was 1.5 cm3/g. the hash content was at
maximum of 5%with the coarse grade being supplied in the range of 0.5-8 mmm. Throughput
the experiment, a GAC that falls in the 33-6 mm size range was used which was subsequently
washed using deionized waster numerous times (Viessman, 2014, p.245). Washing with
deionized water was done to eliminate fines from the sorbent. The sample was then dried at a
temperature of about 110⁰C for 48 hours. The arithmetic mean values of the respective sizes of
the mesh were used in estimating the diameters of the carbon particles which were assumed to be
spherical in shape.

The PAH that was used in the experiment was naphthalene which was purchases from Johnson
and Witner Co. naphthalene solution as prepared from a stock solution of the product as was
packed from the supplier. Proper dilution of the solution was done in deionized water to prepare
synthetic solution.
Naphthalene analysis
The nature of the experiment determined the composition of the aqueous solution. UV-vis
Spectrophotometry was used in determined the content of PAH in the aqueous phase and the
level of sorption determined using the residual concentration of the naphthalene in the solution
that was equilibrated. The absorbance value of naphthalene was determined at a wavelength of
266 nm (Viessman, 2014, p.263).
Batch kinetics experiments
An experiment was set up of a standard agitated rector that was meant to find the kinetic data of
F400. A wet-sieved F400 fractions having particles of narrow sizes ranging 3-6 mm was used an
mechanical shaker used for conducting the dynamic contact between the naphthalene solution
and the granular activated carbon. Shaking on the mechanical shaker was conducted at varied
speeds to establish the minimum speed beyond which the kinetics would not depend on the level
of agitation. This was the speed at which kinetics was not affected by film diffusion. It was
above this minimum speed that measurements of the sorption rates were being taken. Analysis
of the kinetics was dome at a room temperature of 294F. 200ml of naphthalene solution was put
in contact with 0.3 g of the sorbent material in 500 ml glass rector (Vigneswaran, 2009, p.562).
The solution was left in the glass reactor until an equilibrium was attained. Samples of the
solution were the filtered using a 0.45 μmfilter to remove fines that were produced during the
process of shaking the mixtures. Filtering was also done to eliminate any possible chances of

interference in the quantification step that would follow (Wong, 2012, p.313). The quantity of
sorbent in the liquid was measured and the values used in determining the level of sorption that
occurred. The results were highly reproducible following the high levels of accuracy that was
maintained throughout the session.
SEM analysis
The morphology of the surface of granular activated carbon was monitored using a JOEL 3400
Scanning Electron Microscopy with Energy Dispersive System (SEM-EDS).
Adsorption Studies
The batch experiments were used in the exploration of the kinetics measurements of naphthalene
at temperature of 21⁰C using water, and heptane or cyclohexane as the solvents. Approximately
50g of granular activated carbon were correctly and precisely weight and added into a dark flasks
made of glass and were containing naphthalene solution at a constant volume of the initial
concentration that was used. Vigorous stirring was done to the suspensions (100rpm) in a bath
that was regulated with a thermostat. Volumes of the solution were removed from the dark glass
flasks at predetermined time internals to determine adsorbent concentration evolution in the
supernatant liquid. UV spectrometer was used in this process (Zimmerman, 2009, p.291).
To avoid alterations in the volume of the solution, the samples that were extracted for use in
determining the evolution of the adsorbent concentration were reintroduced into the flasks. Blank
experimented were conducted to ensure confirm that there were no losses resulting from
adsorption or violation on the walls of the flask. The equation qt= (C¿¿ 0−Ct )V
m ¿ where
C0 is the initial concentration, Ct is the remaining concentration of the solution after adsorption, V

the volume of the solution and m the mass of the adsorbent material was used in determining the
quantity of the adsorbed materials (Wang, 2009, p.251).
Theoretical models of kinetics
Numerous simplistic mathematic expressions were used in the modeling of the kinetics of
naphthalene adsorption on activated carbon from the solutions that were diluted. Among those
mathematical expression include the pseudo-first order, Elovich model as well as pseudo-second
order. Elovich model describes the process of sorption as one in which is a collection of
mechanisms including diffusion in the bulk solution activated catalytic surfaces and surface
diffusion (Yung-tse, 2012, p.333). A simplified version of the Elovich model is mathematically
expressed as
qt= 1
β ln ( αβ )+ ¿ 1
β ln t ¿ where β and α are the Elovich equation parameters derived from linear
regression analysis of the equation of qt=F (t).

CHAPTER 4: RESULTS AND DISCUSSION
Laboratory batch experiments
Adsorption of heavy metal by granular activated charcoal
Individual metals
The figure below illustrated the efficiencies of removal of all the metals that were studied in this
experiment (lead, zinc and copper) which followed the order of Pb>Cu>Zn at a pH of 6.5.the
same order was maintained for the solubility products constant of the metal hydroxide
precipitate. The metal hydroxide precipitate for the metal hydroxides of Pb, Cu and Zn were
19.9, 19.3 and 16.5 respectively (Bingham, 2012, p.202). This order was also used as the reverse
order for the first hydrolysis constants of the metal the pk1 of the complexes of the metal
hydroxides for Pb, Cu and Zn were 7.7, 7.9 and 9.0 respectively. From these obtained results it
can be observed that higher pks values of a metal meant greater tendency for such a metal to
precipitate into a metal hydroxide. On the other hand, a lower pk1 value meant that the metal
produced a hydroxyl complex of a soluble metal more readily and easily. Metals with high pks
values revealed higher adsorption capacities since the metals were able to form surface
precipitation in the adsorbent substance. It is possible for surface precipitation to take place at
pHs that is lower than those with each precipitation takes place in solution. This is because the
adsorbent substances offer a nucleus that induces the process of precipitation (Crini, 2010,
p.102). The results indicated that the level of affinity for adsorption in metal hydroxide
complexes was than that of ions of divalent metals. It is for this reason that metals readily
forming hydroxyl complexes i.e. metal with lower pk1 had higher capacities of adsorption.

Table 3: Levels of concentration of the metals used in the experiment
1 2 4 8 12
0.0258 0.0614 0.1233 0.2537 0.3657
0.0261 0.0615 0.1228 0.2542 0.3653
0.0257 0.0611 0.1242 0.2543 0.3658
0.0258666
7
0.0613333
3
0.1234333
3
0.2540666
7
0.3656
Figure 2: Concentrations of metals used in the experiment
Mixed metals
The efficiencies of the removal of mixed metals at different concentrations are shown in the
figure below. This study used heavy metals including lead, zinc and copper and their
concentrations were 5.0, 10.0 and 3.0 mg/L. the initial concentrations of the metals was as per
the order Zn>Pb>Cu while the elimination efficiencies of the metals was as per the order
Pb>Cu>Zn. The order for the concentration of the metals is the same as a single metal adsorption
even though there is a huge difference in the initial concentrations (Gupta, 2012, p.212). The

difference may be attributed to the adsorption of metals being controlled and guided by the
mechanism of adsorption than the initial concentration that were used in the mixture.
Figure 3: Efficiencies of the removal of mixed metals at different concentrations
Laboratory fixed bed column experiments
Removal of heavy metals
The figure below represents the curves for the column breakthrough for the three metals tested.
While the column breakthrough for copper was achieved after 24 hours (133 BV), the exhaustion
point did not take place at the same time but instead occurred after 120 hours (666 BV). There
was no clear and concise breakthrough that took place for lead in the 120 hours of the filter
operation. For the case of zinc, a breakthrough took place as soon as 8 hours and within 24 hours
of operation the exhaustion was nearly completed (Erickson, 2013, p.312). The zinc
breakthrough was almost completed after 24 hours. The obtained results are consistent with the
finding document in literature for the other adsorbents.

Figure 4: Curves for the column breakthrough for the three metals tested
An example is the time for breakthrough of metals when the pH is at 5 using dried green algae as
the adsorbent in fixed-bed columns. It this adsorbent, the longest breakthrough is that of lead
followed by copper and then lastly zinc (Mannina, 2017, p.172). These findings also established
that the capacity of adsorption as calculated from the breakthrough curves was in line with the
order Pb>Cu>Zn. It is reported that the Cu had the longest time taken in reach the influent
concentration for the effluent concentration of metals at a pH of 3 for the metal solutions. This
was when the metals were passed through columns of iron oxide coated zeolite and zeolite.
During such an experiment, lead was not detected in the effluent of ant of the columns used.
The order Pb, Cu>Zn was followed for the breakthrough curves in the granular activated carbon
columns which was the same as the results of one of a single metal as well as mixed metals. The
same order was followed for cumulative elimination of heavy metals and cumulative percentage
of the heavy metals eliminated at the 120 hours of filter operation (Ryan, 2010, p.201). Previous
studies and resect on literature for the adsorption of heavy metals using numerous organic and
inorganic adsorbents in both column and batch studies yielded results of riders that are similar to
this order. Still, the arguments and reasons given for this trend earlier in this study specifically on

the discussion about batch columns, among them solubility products of the metal hydroxides and
the constants of hydrolysis of the metals, the concentration difference of metals in the influent
solution can as well be used in offering explanation to the differences in the adsorptions between
the metals. Metals that were in low concentrations in the influent water showed relatively low
adsorption.
Feasibility study and optimum experiment conditions
The studies of the batch and the column illustrated that contaminant from urban stormwater
among them heavy metal and polycyclic aromatic hydrocarbons can be removed effectively by
the use of a single 100 cm height granular activated carbon column when the velocity of the
filtration is set at 5m/h (Scholz, 2006, p.260). When the velocity is set higher at 10 and 11.5 m/h,
the efficient in the removal of other pollutants such as dissolved organic carbon and turbidity
reduce. The performance of the granular activated carbon column systems was improved hen
pretreated with anthracite. The operational conditions may be altered depending on the levels or
concentrations of pollutants at a specific site. This calls for conducting the same experiment in
sites that have different concentration of pollutants, column height and filtration velocities so as
to establish the optimum conditions for the different scenarios. There is no defined optimum
condition for this operation and experiment as it will vary relative to the characteristics of the
pollutants of the water and the column system used.
It should be remembered that this experiment used synthetic wastewater as an alternative for
stormwater since storm water contains insignificant concentration of the heavy metals that were
under study. The results obtained on the removal of heavy metals may therefore not be an exact
replicate of real stormwater even though a lot of information and analysis can be drawn from the
results and may be used to make substantial conclusions (Shifrin, 2014, p.140). To obtain better

results, stormwater having higher concentrations of the heavy metals should be used. This should
also be used in confirming the finding and thus forming the basis of grounding the arguments.
Previous studies have reported the mechanisms and kinetics of liquid phase adsorption of PAH
on various porous adsorbents. This experiment aimed at study the removal of naphthalene from
storm after using granular activated carbon (Crini, 2010, p.179). This study envisages the
perspective of competitive adsorption that is derived from the interactions of adsorbent solvent
and adsorb ate solvent by deploying solvent increasing polarity. A graph of the evolution of the
concentration of naphthalene with respect to time was plotted as shown in the figure.
Figure 5: Evolution of the concentration of naphthalene with respect to time
As illustrated by the figure it is observable that the uptake of naphthalene is very fast at the initial
stages followed closely by the second stage in which the uptake is observed to be increasing
steadily to the point of the equilibrium conditions. Adsorption of naphthalene requires a
relatively longer time from organic solvents and this illustrated the high affinity of naphthalene
molecules to the organic phase of granular activated carbon (Wong, 2012, p.233). The
observation can be attributed to the fact that when adsorption takes place from organic solvents,
there is large suppressing and this appears to be a reasonable point in relation to the observed
high solubility of naphthalene in granular activated carbon.

Analysis in the intraparticle diffusivity mechanisms
The figure below shows the scanning examinations of the F400, illustrating a material of a
heterogeneous phase.
Figure 6: Scanning examinations of the F400
The diameter of the particle size is about 600-750 nm for the granular activated carbon hence
leading to the creating of large pores that will permit fast diffusion of the solutes.as can be seen
from the figure above, thee are micropores an macropores which have irregular surfaces relative
to the hypercrossslinked polymeric sorbents with the same surface area. The diameter of the
micropores contained in the F400 range between 0.6 and 1.4nm with the pores having an average
diameter of between 100-200 nm (Negulescu, 2011, p.112). The volumes of the pores account
for about 20% of the meso and the macropores an illustration that the sample of GAC that has
been analyzed has a relatively significant number of macropores and mesopores as compared to
the number of the fine pores that are available.

Three stages are involving in the sorption of naphthalene onto a porous sorbent for example
granular activated carbon. The initial and the first stage is presumed to be taking place very
rapidly and does not generate a stage that limits the rate in the sorption process of organic
compounds on granular activated carbon. the major resistance to mass flow is proposed to be
mainly occurring in the second stage. In this stage, naphthalene moves or diffused in the internal
structure of the solvent (Scholz, 2006, p.263).
Recommendations are made that the structures of the pores that are used in sorbent particles in
this process be made up of transitional pores, macropores as well as micropores. The transitional
pores, also called mesopores are important in the transportation of the molecules to the
micropores. The pores, pore-surface diffusion control or the surface are the main control
mechanisms onto which most of the molecular sorption pores incorporate diffusive mass
transfer.
Kinetics sorption models for non-polar organic micro pollutants
The sorption process of organic pollutants onto either natural or synthetics sorbents has been
defined as a sophisticated and complicated process. In this process, the properties of the solvent
as well as the sorbet play a crucial role. The process of sorption takes place within the layer of
boundary around the sorbent and continues in the liquefied pores or even along the pore walls of
the sorbent (Shafer, 2012, p.184). These two processes are referred to as internal and external
mass transfer steps respectively. The solubility of naphthalene at 25⁰C is 30.8 mg/dm3, has a log
Kow value of 3.4, molar volume of 148 cm3/mol and a concentration range of 0-5. The structure
of naphthalene is as shown below

Figure 7: Structure of Naphthalene
Numerous processes are used in an attempt to explain the sorption of organic micro pollutants
like naphthalene using granular activated carbon just like is the case with other numerous
heterogeneous process that takes place between solids and fluids. These sequential processes are
used in the determination of the rate of reactions for the sorption process. Among the processes
include:
 Diffusion of the solute through the films of the liquid that are surrounding the particle.
This process is also called control of liquid film diffusion (Erickson, 2013, p.158)
 Diffusion of the solute vi the sorbent matrix of the granular activated carbon also known
as particle diffusion control
 The chemical reaction involving the functional groups that are linked with the matrix
One of the steps is normally associated with a lot of resistance in comparison to the others and
hence may be treated as a step that limits the rate of the process. Despite the chemical reaction of
naphthalene on the surfaces of sorbents being explained as a chemisorption process, the process
in normally presumed to be too rapid to have an effect on the overall sorption rate. This is
however changed should modification on the chemical composition of the reactants during the

process of sorption (Wong, 2012, p.218). The conditions needed for the process of liquid
diffusion control to occur are obvious and are mostly composed of a low degree of agitation,
small size of the particle besides low concentration levels of the solution.
Results for PAH analysis using UV-vis Spectrophotometry
Table 3: Results for PAH analysis using UV-vis Spectrophotometry
Column1 Column2 Column3 Column4 Column5 Column6
0mg of GAC
0.1mg of
Gac
0.25mg of
Gac
0.75mg of
Gac 1mg of Gac
0 Minute 0.2018
0.202
0.2025
0.2021
1 Minute 0.1636
0.1637
0.1635
0.1636
2 Minutes 0.1585
0.1619
0.1556
0.158666667
3 Minutes 0.1169
0.1168
0.117
0.1169
4 Minutes 0.1024
0.1021
0.1017
0.10206666
7

15
Minutes 0.2034 0.0757 0.0374 0.0188 0.0161
0.2035 0.0761 0.0371 0.0179 0.0108
0.2038 0.0763 0.0373 0.0178 0.0109
0.20356666
7
0.07603333
3 0.037266667 0.018166667 0.0126
30
Minutes 0.2027 0.0239 0.0073 0.005 0.0038
0.2021 0.0242 0.0071 0.005 0.003
0.2022 0.0238 0.0062 0.0056 0.003
0.20233333
3
0.02396666
7 0.006866667 0.0052
0.00326666
7
60
minutes 0.2047 0.0111 0.0063 0.0045 0.0017
0.2046 0.0112 0.0066 0.0046 0.0014
0.2047 0.0113 0.006 0.0046 0.0014
0.20466666
7 0.0112 0.0063 0.004566667 0.0015
120
minutes 0.2016 0.0012 0.001 0.0001 0.0002
0.2032 0.0011 0.0009 0.0003 0
0.203 0.0011 0.0009 0.0002 0.0001
0.2026
0.00113333
3 0.000933333 0.0002 0.0001
Figure 8: PAH analysis using UV-vis Spectrophotometry

Figure 9: PAH analysis using UV-vis Spectrophotometry at 0 GAC
Figure 10: PAH analysis using UV-vis Spectrophotometry at 0.1gm GAC

Figure 11: PAH analysis using UV-vis Spectrophotometry at 0.25gm GAC
Figure 12: PAH analysis using UV-vis Spectrophotometry 0.75gm GAC

Figure 13: PAH analysis using UV-vis Spectrophotometry 1gm GAC
Two models; the shell progressive model, also called the shrinking core model and the
homogenous particle diffusion model are the main kinetic models that have been chosen to
explain solution extraction data.
The shell progressive model
Also called the unreacted shrinking core, the shell progressive model is a model of mass transfer
and involves the reaction beginning at the surface of the particle. The particle surface forms a
reacted zone which then moves inwards at a specific velocity. The expression below is used in
determining the relationship between the level of sorption and the sorption time:
The reaction is controlled by the fluid film:
X ( t ) = 3C A 0 KmA
as Cso
t

When the model is under the control of a chemical reaction:
[ 1−(1− X (t 1/3 )) ] = K s CA 0
as
2 Cso
t
When the reaction is controlled by diffusion that occurs through the sorption layer
[ 3−3(1− X (t1/ 3))−2 X (t ) ]=6 De CA 0
as
2 Cso
t
The homogenous particle diffusion model
Numerous possible resistances are used in the extraction mechanism which incorporates the
diffusion of naphthalene molecules from the aqueous solution into the phase of the sorbent.
Fick’s equation is used in explaining the rigorously explain the sorption of the molecules of
naphthalene in which the diffusion of the molecules of naphthalene are applied in a quasi-
homogenous media (Horwatich, 2008, p.280). An explanation on the diffusion of the molecules
of naphthalene in the sorbent phase is from a solution of infinite volume into the sorbent particle
was made. The equation below is derived as the sorption on the spherical particles is controlled
by the rate of diffusion which yield a simultaneous ends of algebraic and differential equations
X ( t ) =1 6
π ∑
z=1
∞ 1
z2 exp [ −z2 π2 De t
r2 ] where X ( t ) refers to the fractional attainment of equilibrium at
time, t, De is effective diffusion coefficient of the sorbent that are found in the sorbent phase, z is
an integer, and r the radius of the particles of sorbent which are assumed to be spherical in shape
(Franklin L. Burton, 2013, p.132). The following equation can be used in estimating the values
of X ( t )

X ( t )= qt
qe where qt and qe are the loadings of the solute on the solid phase at tine t and upon the
attainment of equilibrium respectively. The whole range of the above equation can be fitted
using the Vermeulen’s approximation for sorption on the spherical particles as shown
X ( t )= [1−exp [ −z2 π2 De t
r 2 ] ]
Further simplification of the equation can be done for most of the data points so as to find the
estimations of particle diffusivity. The following expression can be used as the further simplified
form of the equation
−ln ( 1−X2 ( t ) ) =2 Kt , in which K= π 2 De
r2
The analogous expression below is usable in cases where the rate of sorption is controlled by
liquid film diffusion
X ( t )=1−exp [ 3 De C
rCr ]
−ln ( 1−X (t) ) =Kii t where Kii= 3 De C
rCr
Factors affecting the adsorption process for naphthalene
Temperature
The Van’t equation is used in explaining the effect of temperature on the rate of adsorption of
naphthalene and any other polycyclic aromatic hydrocarbons (Bahadori, 2013, p.132)
ln Keq= ∆ H
R × 1
T + ∆ S
TR

Where ∆ H the change in enthalpy, T the absolute temperature, is ∆ S is the change in the
entropy, R is the gas constant and Keq the coefficient of the equilibrium partition at a specific
temperature. Going by the equation, the adsorption process is endothermic if the slope is
negative and exothermic if the slope is positive.
Solubility
The solubility of the adsorbate is highly influenced by the capacity of the adsorbate to be
adsorbed. The rate of adsorption is inversely proportional to the solubility of an adsorbate in the
solvent in which adsorption is taking place (Figueiredo, 2009, p.190). The solubility is also
inverse to the molecular weight of a PAH.
pH
pH affects the distribution of the surface charge of the adsorbent and thus affects the rate and
level of adsorption. A lowered pH increases the positive charges on the surface of the adsorbent
thereby resulting in increased interaction between the molecules of the PAH and the surface of
the adsorbent. On the other hand, when the pH is increased, the hydroxyl ions on the surface of
the adsorbent are increased while the positive ions are reduced (Dotro, 2017, p.188). The
hydroxyl ions interact with the molecules of naphthalene that are on the active sites of adsorption
leading to a decrease in the efficiency of adsorption of the adsorbent. Naphthalene being a basic
organic compound has a pH of about 9. This means there are more hydroxyl ions that positive
charges on the molecules which in turn interact with the active sites for adsorption resulting into
decreased efficiency of adsorption.
Surface area ad size of the particle
The relationship between the sizes of a particle and the adsorption capacity of the adsorbent is
inverse proportionality. For this reason, the adsorption of naphthalene is found to be higher in
estuarine colloids than in sediment and soil. This is because adsorption is a factor of the surface

area and the level of adsorption is proportional to the specified surface area which in turn is a
factor of the total surface area which is availed for the adsorption (Riffat, 2012, p.123). This
argument has been supported by finding and investigations that were conducted using difference
sizes of particle dn the conclusion was such that particles that have smaller sizes have the largest
surface area for mass transfer and hence higher capacity of adsorption.

CHAPTER 5: CONCLUSION
A GAC fixed-bed filter column and a number of anthracite filter which was followed by GAC
filter were successful in the removal of heavy metals and naphthalene from stormwater. A unit
GAC filter column that works at a low velocity of filtration of 5 m/h was able to eliminate most
of the heavy metals. The effectiveness of the performance of the GAC filter was increased even
farther when the velocities was increased to 10 m/h and 11.5 m/h with the inclusion of an
additional pre-treatment offered to the anthracite filter to the columns. High absorptive removals
were recorded for the metals Pb and Cu at natural pHs of the stormwater which range between
6.9 and 7.2 as recorded by the GAC filter column. The GAC adsorption rates for Zn over 120
hours of continuous operation of the filter at a velocity of 5 m/h were observed to be moderate.
The same removal orders were observed in batch adsorption of both single and mixed metals
which was similar to column studies. The process of removal of the metals followed the order
Pb, Cu>Zn, a scenario that can be explained by the solubility product as well as the constants of
the first hydrolysis of these metal hydroxides.
The parameters of kinematics that were obtained for the F400 followed the same order of
magnitude as that obtained by previous authors for F400 when applied in the sorption of organic
pollutants among them phenolic compounds, dyes and hydrocarbons. The shell progressive
model, also called the shrinking core model and the homogenous particle diffusion model are
usable in studying the removal of naphthalene by granular activated carbon. the obtained results
in this study and research illustrate that the step for determining the rate of extraction of
naphthalene is the sorbent phase diffusion. The two models allow calculation of the average or
the mean coefficient of the intraparticle diffusion for the case of high levels of naphthalene and

the coefficient of mass transfer in cases of low ranges concentration of the organic molecules.
The shell progressive mechanism and Fick’s law are among the excellent approaches for the
extraction of kinetics of PAHs on granular activated carbon.

References
Alley, E.R., 2007. Water Quality Control Handbook, Second Edition. 2nd ed. New York:
McGraw Hill Professional.
Bahadori, A., 2013. Waste Management in the Chemical and Petroleum Industries. 4th ed.
London: Wiley.
Bingham, E., 2012. Patty's Toxicology, 6 Volume Set. 5th ed. New York: John Wiley & Sons.
Chowdhury, Z.K., 2013. Activated Carbon: Solutions for Improving Water Quality. 4th ed. New
York: American Water Works Association.
Crini, G., 2010. Sorption Processes and Pollution: Conventional and Non-conventional Sorbents
for Pollutant Removal from Wastewaters. 3rd ed. Berlin: Presses Univ. Franche-Comté.
Dotro, G., 2017. Treatment Wetlands. 4th ed. London: IWA Publishing.
Erickson, A.J., 2013. Optimizing Stormwater Treatment Practices: A Handbook of Assessment
and Maintenance. 9th ed. New York: Springer Science & Business Media.
Figueiredo, J.L., 2009. Carbon Materials for Catalysis. 3rd ed. Sydney: John Wiley & Sons.
Franklin L. Burton, G.T.R.T.H.D.S.M.&.E.I., 2013. Wastewater Engineering: Treatment and
Resource Recovery. 3rd ed. Paris: McGraw-Hill Education.
Gupta, V.K., 2012. Environmental Water: Advances in Treatment, Remediation and Recycling.
4th ed. Chicago: Newnes.

Hahn, H.H., 2012. Pretreatment in Chemical Water and Wastewater Treatment: Proceedings of
the 3rd Gothenburg Symposium 1988, 1.–3. Juni 1988, Gothenburg. 6th ed. London: Springer
Science & Business Media.
Hlavinek, P., 2011. Advanced Water Supply and Wastewater Treatment: A Road to Safer Society
and Environment. 7th ed. London: Springer Science & Business Media.
Horwatich, J.A., 2008. Use of a Stormwater Filtration Device for Reducing Contaminants in
Runoff from a Parking Lot in Madison, Wisconsin, 2005-07. 4th ed. Kansas: Wisconsin DOT.
Jakeman, A.J., 2016. Integrated Groundwater Management: Concepts, Approaches and
Challenges. 5th ed. Oxford: Springer.
Kazner, C., 2012. Water Reclamation Technologies for Safe Managed Aquifer Recharge. 4th ed.
London: IWA Publishing.
Lin, S.D., 2014. Water and Wastewater Calculations Manual, Third Edition. 3rd ed. Oxford:
McGraw Hill Professional.
Mannina, G., 2017. Frontiers in Wastewater Treatment and Modelling: FICWTM 2017. 4th ed.
London: Springer.
Microfilms, U., 2008. Dissertation Abstracts International: The sciences and engineering. 3rd
ed. London: University Microfilms.
Negulescu, M., 2011. Municipal Waste Water Treatment. 6th ed. Cambrige: Elsevier.
Nobutada Nakamoto, N.G.R.G., 2014. Progress in Slow Sand and Alternative Biofiltration
Processes. 4th ed. Vatican: IWA Publishing.

Riffat, R., 2012. Fundamentals of Wastewater Treatment and Engineering. 5th ed. Oxford: CRC
Press.
Ryan, Z., 2010. Building with Water: Concepts Typology Design. 5th ed. London: Walter de
Gruyter.
Scholz, M., 2006. Wetland Systems to Control Urban Runoff. 5th ed. New York: Elsevier.
Shafer, W.H., 2012. Masters Theses in the Pure and Applied Sciences: Accepted by Colleges and
Universities of the United States and Canada, Volume 39. 5th ed. New York: Springer Science &
Business Media.
Shi, C.Y., 2011. Mass Flow and Energy Efficiency of Municipal Wastewater Treatment Plants.
5th ed. Manchester: IWA Publishing.
Shifrin, N., 2014. Environmental Perspectives: A Brief Overview of Selected Topics. 2nd ed.
Paris: Springer.
Sperling, M.V., 2005. Biological Wastewater Treatment in Warm Climate Regions, Volume 1.
4th ed. New York: IWA Publishing.
Strande, L., 2014. Faecal Sludge Management: Systems Approach for Implementation and
Operation. 12th ed. Manchester: IWA Publishing.
Viessman, W., 2014. Water Supply and Pollution Control. 4th ed. London: Pearson Education.
Vigneswaran, S., 2009. Waste Water Treatment Technologies - Volume I. 7th ed. Mancchester:
EOLSS Publications.
Wang, L.K., 2009. Hazardous Industrial Waste Treatment. 6th ed. Beijing: CRC Press.

Wong, M.H., 2012. Environmental Contamination: Health Risks and Ecological Restoration. 3rd
ed. Beijing: CRC Press.
Yung-tse, H., 2012. Handbook Of Environment And Waste Management: Air And Water
Pollution Control. 2nd ed. Beijing: World Scientific.
Zimmerman, J.B., 2009. Environmental Engineering: Fundamentals, Sustainability, Design. 7th
ed. Chicago: Wiley.