IN THE NAME OF ALLAH, THE MOST MERCIFUL, THE BENEFICENT
Removal of selected metal ions from aqueous media by agricultural wastes: Kinetic and
thermodynamic studies
BY
Abida Kausar
M.Phil. (UOS)
A Thesis Submitted in Partial Fulfillment of the Requirement for the Degree of
DOCTOR OF PHILOSOPHY In
CHEMISTRY
DEPARTMENT OF CHEMISTRY FACULTY OF SCIENCES
UNIVERSITY OF AGRICULTURE FAISALABAD
2014
To
The Controller of Examinations, University of Agriculture, Faisalabad.
“We, the Supervisory Committee, certify that the contents and form of
thesis submitted by Miss. Abida Kausar, Regd. 2010-ag-608 have been found
satisfactory and recommend that it be processed for evaluation, by the External
Examiner (s) for the award of degree”.
Supervisory Committee
Supervisor
Prof. Dr. Haq Nawaz Bhatti
Member
Dr. Raja Adil Sarfraz
Member
Dr. Muhammad Shahid
DECLARATION
I hereby declare that the contents of the thesis “Removal of selected metal ions from
aqueous media by agricultural wastes: Kinetic and thermodynamic studies” are product
of my own research and no part has been copied from any published source (except the
references, standard mathematical or genetic models/equations/formulate/protocols etc). I
further declare that this work has not been submitted for award of any other diploma/ degree.
The University may take action if the information provided is found inaccurate at any stage.
(In case of any default the scholar will be proceeded against as per HEC plagiarism policy).
Abida Kausar 2010-ag-608
Acknowledgement First and foremost I will thank Almighty GOD, Merciful, who bestowed me with the potential and ability to complete this work. Praise for His last Prophet MUHAMMAD (PBUH) who advised all of us to continue getting education from cradle to death.
Well, the list of the people I need to thank will not fit to a single Acknowledgement section. I just mention some people whose contribution is obvious. The best and worst moments of my doctoral journey have been shared with many people. It has been a great privilege to spend several years in the Department of Chemistry & Biochemistry at University of Agriculture, Faisalabad, and its members will always remain dear to me.
My first debt of gratitude must go to my supervisor Prof. Dr. Haq Nawaz Bhatti, Department of Chemistry, University of Agriculture, Faisalabad. He patiently provided the vision, encouragement and advice necessary for me through the doctoral program and complete my dissertation. He has been a strong and supportive adviser to me throughout my career, and he has always given me great freedom to pursue independent work.
I feel pleasure, in expressing my humble gratitude to Prof. Dr. Muhammad Asghar, Chairman, Department of Biochemistry, University of Agriculture, Faisalabad, and my committee members, Dr. Raja Adil Sarfraz, Assistant Professor, Department of Chemistry, University of Agriculture, Faisalabad and Dr. Muhammad Shahid, Associate Professor, Department of Biochemistry, University of Agriculture, Faisalabad for their support, guidance and helpful suggestions.
My friends in UAF, SUERC and other parts of the world were sources of laughter, joy, and support during my work. All staff members of Environmental Chemistry Laboratory particularly Shazia Noreen and Faiza Amin, University of Agriculture, Faisalabad, and Caroline Donnelly, Scottish Universities Environmental Research centre, Scotland, UK also deserve my sincerest thanks, their friendship and assistance has meant more to me than I could ever express.
I acknowledge with great pleasure to Dr. Gillian Mackinnon, Scottish Universities Environmental Research Centre, Scotland, UK for allowing me to be part of a great professional community. I am very grateful for her valuable guidance in experimental work, data analysis and manuscript preparation. I would like to thank Dr. Justin Hargreaves and Abdul, School of Chemistry, University of Glasgow, for the help in analysis of my samples.
I wish to thank my parents; their love provided my inspiration and was my driving force. I owe them everything and wish I could show them just how much I love and appreciate them. I will give a heartfelt “Thanks” to my husband for love, encouragement allowed me to finish this journey. I also want to thank to lovely brothers for their unconditional support.
I am thankful to Higher Education Commission (HEC) of Pakistan for providing funds to accomplish this work.
May God bless all of us
Abida Kausar
LIST OF CONTENTS
Page No.
List of Figures I
List of Tables IV
Abstract VI
CHAPTER-1
1. INTRODUCTION 1
CHAPTER-2
2. REVIEW OF LITERATURE 5
2.1. Batch biosorption 7
2.2. Linear and non-linear regression analysis 13
2.3. Response surface methodology 16
2.4. Column biosorption 20
CHAPTER-3
3. MATERIALS & METHODS 23
3.1. Collection and preparation of biosorbent 23
3.2. Chemicals 23
3.3. Analytical determination of metal ions 23
3.4. Initial screening of biosorbents 24
3.5. Pre-treatments of biomasses 24
3.6. Immobilization of biosorbents 25
3.7. Batch biosorption 25
3.7.1Effect of pH 25
3.7.2 Effect of biosorbent amount 26
3.7.3. Effect of contact time 26
3.7.4 Effect of initial metal ion concentration 26
3.7.5 Effect of temperature 26
3.8. Sorption kinetics 26
3.8.1. Pseudo-first order kinetic model 26
3.8.2. Pseudo-second order kinetic model 27
3.9. Equilibrium study 27
3.9.1. Freundlich isotherm 28
3.9.2. Langmuir isotherm 28
3.9.3. Redlich-Peterson isotherm 29
3.10. Error analysis for kinetic and equilibrium models optimization 30
3.11. Thermodynamic study 32
3.12. Effect of interfering ions 32
3.13. Response surface methodology 32
3.14. Desorption studies 34
3.15. Biosorbent characterization 35
3.15.1. Determination of elemental composition 35
3.15.2.Determination of chemical composition 35
3.15.3.Determination of surface area 35
3.15.4.Determination of surface morphology 36
3.15.7. Determination of thermal stability 36
3.15.6. Determination of functional groups 36
3.16. Column biosorption 36
3.16.1. Thomas model 37
3.16.2. Bed-depth service time (BDST) model 37
3.17. Statistical analysis 38
CHAPTER-4
4. RESULTS AND DISCUSSION 39
4.1. Screening of biosorbent 39
4.2. Effect of pre-treatments 41
4.3. Effect of initial pH 43
4.4. Effect of biosorbent amount 46
4.5. Effect of contact time 48
4.6. Biosorption kinetics 51
4.6.1 Pseudo-first order kinetic model 51
4.6.2. Pseudo-second order kinetic model 52
4.7. Error analysis for optimization of kinetic model 59
4.8. Effect of initial metal ion concentration 62
4.9. Equilibrium modeling 64
4.9.1. Freundlich isotherm 65
4.9.2. Langmuir isotherm 66
4.9.3. Redlic-Peterson isotherm 67
4.10. Error analysis for optimization of sorption isotherms 75
4.11. Effect of temperature 81
4.12. Thermodynamics studies 83
4.13. Effect of interfering ions 86
4.14. Desorption studies 89
4.15. Response surface methodology 92
4.15.1. Fitness of model 92
4.16. Biosorbent characterization 104
4.16.1. Surface studies. 104
4.16.2. Elemental analysis 104
4.16.3. Thermogravimetric analysis 105
4.16.4. X-Ray diffraction (XRD) studies 108
4.16.5. Scanning electron microscope and Energy dispersive
X- Rays
110
4.16.6. FT-IR Studies 113
4.17. Column biosorption 119
4.17.1. Effect of bed height 119
4.17.2. Effect of flow rate 121
4.17.3. Effect of initial metal ion concentration 124
4.17.4. Application of Thomas model. 126
4.17.5. Application of Bed Depth Service Time (BDST) model 127
CHAPTER-5
5. Summary 128
LITERATURE CITED 132
i
List of Figures
Figure
No.
Title Page
No.
4.1 Screening of biosorbent for U(VI) removal. 39
4.2 Screening of biosorbent for Zr (IV) removal. 40
4.3 Screening of biosorbent for Sr (II) removal. 40
4.4 Effect of pre-treatments on biosorption of U (VI) onto rice husk 41
4.5 Effect of pre-treatments on biosorption of Zr (IV) onto bagasse. 42
4.6 Effect of pretreatments on biosorption of Sr (II) onto peanut husk. 42
4.7 Effect of initial pH on U(VI) biosorption onto rice husk 44
4.8 Effect of initial pH on Zr(IV) biosorption onto bagasse. 45
4.9 Effect of initial pH on Sr(II) biosorption onto peanut husk 46
4.10 Effect of sorbent amount on biosorption of U (VI) onto rice husk. 47
4.11 Effect of sorbent amount on biosorption of Zr (IV) onto bagasse. 47
4.12 Effect of sorbent amount on biosorption of Sr (II) onto peanut
husk.
48
4.13 Effect of time on biosorption of U (VI) onto rice husk. 49
4.14 Effect of time on biosorption of Zr(IV) onto bagasse. 50
4.15 Effect of time on biosorption of Sr (II) onto peanut husk. 50
4.16 Comparison of kinetic models for U(VI) sorption onto rice husk 54
4.17 Comparison of kinetic models for Zr(IV) sorption onto bagasse. 56
4.18 Comparison of kinetic models for Sr(II) sorption onto peanut
husk.
58
4.19 Effect of initial metal ion concentration on U(VI) biosorption onto
rice husk.
62
4.20 Effect of initial metal ion concentration on biosorption of Zr(IV)
onto bagasse.
63
4.21 Effect of initial metal ion concentration on Sr(II) biosorption onto
peanut husk.
64
4.22 Comparison of equilibrium isotherms for U(VI) sorption onto rice
husk
70
4.23 Comparison of equilibrium isotherms for Zr(IV) sorption onto bagasse
72
ii
4.24 Comparison of equilibrium models for Sr(II) sorption onto peanut
husk.
74
4.25 Effect of temperature on U(VI) biosorption onto rice husk 81
4.26 Effect of temperature on Zr(IV) biosorption onto bagasse 82
4.27 Effect of temperature on biosorption of Sr(II) onto peanut husk. 83
4.28 Comparison of different desorbing agents on U(VI) biosorption
onto rice husk
90
4.29 Comparison of different desorbing agents on Zr(IV) biosorption
onto bagasse
91
4.30 Comparison of different desorbing agents on Sr(II) biosorption
onto peanut husk.
91
4.31 (a) The plot of predicted sorption capacity q (mg/g) versus actual
for U(VI) sorption onto native rice husk. The studentized residual
and normal % probability plot for U(VI) sorption onto native rice
husk.
96
4.32 (a) The plot of predicted sorption capacity q (mg/g) versus actual
for Zr(IV) sorption onto native bagasse. The studentized residual
and normal % probability plot for Zr(IV) sorption onto native
bagasse.
97
4.33 (a)The plot of predicted sorption capacity q (mg/g) versus actual
for Sr(II) sorption onto NaOH-treated peanut husk. The
studentized residual and normal % probability plot of removal
Sr(II) onto NaOH-treated peanut husk.
98
4.34 Contour plot showing effect of pH, sorbent dose and initial U(VI)
concentration on U(VI) sorption onto rice husk.
100
4.35 Contour plot showing effect of pH, sorbent dose and initial Zr(IV)
concentration on Zr(IV) sorption onto bagasse. .
102
4.36 Contour plot showing effect of pH, sorbent dose and initial Sr(II)
concentration on Sr(II) sorption onto peanut husk.
103
iii
4.37 TGA of rice husk 105
4.38 TGA of bagasse 106
4.39 TGA of peanut husk 107
4.40 XRD pattern of rice husk 108
4.41 XRD pattern of bagasse. 109
4.42 XRD pattern of peanut husk 109
4.43 SEM-EDX spectra of rice husk. 110
4.44 SEM-EDX spectra of bagasse 111
4.45 SEM-EDX spectra of peanut husk 112
4.46 FT-IR spectra of rice husk. 114
4.47 FT-IR spectra of bagasse. 116
4.48 FT-IR spectra of peanut husk 118
4.49 Breakthrough curves at different bed heights for U(VI) and Zr(IV)
biosorption onto rice husk and bagasse.
120
4.50 Breakthrough curves at different flow rates for U(VI) and Zr(IV)
biosorption onto rice husk and bagasse.
122
4.51 Breakthrough curves at different initial inlet metal ion
concentration for U(VI) and Zr(IV) biosorption onto rice husk and
bagasse.
125
iv
LIST OF TABLES
Table
No.
Title Page
No.
3.1. Experimental ranges and levels of independent variables. 34
4.1 Comparison of parameters of kinetic models for uranium sorption
onto rice husk by linear and non-linear regression methods.
53
4.2 Comparison of parameters of kinetic models for zirconium
sorption onto bagasse by linear and non-linear regression methods.
55
4.3 Comparison of parameters of kinetic models for strontium sorption onto peanut husk by linear and non-linear regression methods.
57
4.4 Kinetic model optimization for U(VI) ions sorption onto rice husk
by error functions.
59
4.5 Kinetic model optimization for Z(IV) ions sorption onto peanut
husk by error functions.
60
4.6 Kinetic model optimization for Sr(II) ions sorption onto peanut
husk by error functions.
61
4.7 Equilibrium models parameters for U(VI) sorption onto rice husk
by linear and non-linear regression methods.
69
4.8 Equilibrium models parameters for Zr(IV) sorption onto bagasse
by linear and non-linear regression methods.
71
4.9 Equilibrium models parameters for Sr(II) sorption onto peanut
husk by linear and non-linear regression methods.
73
4.10 Optimization of equilibrium isotherm for U(VI) sorption onto rice
husk by error functions.
78
4.11 Optimization of equilibrium isotherm for Zr(IV) sorption onto
bagasse by error functions.
79
4.12 Optimization of equilibrium isotherm for S(II) sorption onto
peanut husk by error functions.
80
4.13 Thermodynamic parameters for U (VI) biosorption onto rice husk
as a function of temperature
84
v
4.14 Thermodynamic parameters for Zr (IV) biosorption onto bagasse
as a function of temperature.
85
4.15 Thermodynamic parameters for Sr (II) biosorption onto peanut
husk as a function of temperature.
86
4.16 Comparison of the effect of different interfering cations and
anions on U(VI) ions (50 mg L-1) biosorption onto rice husk.
87
4.17 Comparison of the effect of different interfering cations and anions on Zr(VI) ions (50 mg L-1) biosorption onto bagasse
88
4.18 Comparison of the effect of different interfering cations and anions on Sr(II) ions (10 mg L-1) biosorption onto peanut husk.
89
4.19 Analysis of variance (ANOVA) for response surface quadratic
model for U(VI) sorption onto native rice husk
93
4.20 Analysis of variance (ANOVA) for response surface quadratic model for Zr(IV) sorption onto native bagasse.
94
4.21 Analysis of variance (ANOVA) for response surface quadratic model for Sr(II) sorption onto NaOH-treated peanut husk.
95
4.22 Brunauer-Emmett-Teller (BET) surface area analysis and Barrett-Joyner-Halenda (BJH) pore size and volume analysis.
104
4.23 Elemental (C, H and N) analysis of native rice husk, bagasse and
peanut husk
104
4.24 Functional groups in rice husk by FTIR by spectra 113
4.25 Functional groups in bagasse by FT-IR by spectra 115
4.26 Functional groups in peanut husk by FTIR spectra. 117
4.27 Column sorption capacity and breakthrough time with different
bed heights, flow rates and inlet concentrations.
123
4.28 Thomas Model parameters for the removal of U(VI) and Zr (IV)
by rice husk and bagasse
126
4.29 Bed Depth Service Time model parameters for the removal of
U(VI) and Zr (IV) by rice husk and bagasse.
127
vi
Abstract
In the present research study, biosorption efficacy of agro-wastes (rice husk, bagasse,
peanut husk, cotton sticks and wheat bran) for U, Zr and Sr removal from aqueous media
was investigated. Rice husk, bagasse and peanut husk were selected as most efficient
biosorbent for the removal of U, Zr and Sr ions respectively. These selected biomasses
were subjected to different pre-treatments (Physical and chemical) and modifications
(immobilization). Batch biosorption affecting parameters like pH, sorbent dose, initial
metal ion concentration and temperature were optimized for native, pre-treated and
immobilized biomasses to get maximum removal. Maximum biosorption capacity values
were found at pH (4-5), (3-4) and (7-9) for U, Zr and Sr ions respectively for native, pre-
treated and immobilized biomasses. The amount of metal ions sorbed (mg/g) decreased
with increasing biosorbent dose and increased at higher initial metal ion concentration.
Linear and non-linear regression forms of pseudo-first and second-order were studied and
value of R2 and six non-linear regression error functions namely hybrid fractional error
function (HYBRID), Marquardt’s percent standard deviation (MPSD), average relative
error (ARE), sum of the errors squared (ERRSQ/SSE), sum of the absolute errors (EABS)
and Chi-square test (χ2) were used to predict the most optimum kinetic model. Sorbent-
sorbate reaction nature was estimated by fitting equilibrium data by non-linear and
transformed linear forms of the Langmuir, Freundlich and Redlich-Peterson isotherms
and most optimum isothermal model was optimized by comparing linear and non-linear
R2 value and non-linear regression error functions. Calculated values of thermodynamic
parameters i.e. ΔG˚, ΔH˚ and ΔS˚ showed that studied processes are feasible and
spontaneous. Response surface methodology using face-cantered central composite
design was used to design experiments for biosorption of U(VI), Zr(IV) and Sr(II) ions
onto biomasses. Significance of main, interaction and square effects of quadratic model
was determined by ANOVA, F-test and p value. Adsorption/desorption studies showed
that biosorbents can be reused successfully. Effect of interfering ions (cations & anions)
on the removal efficiencies was studied. The column biosorption was also done and effect
of bed height, flow rate and initial metal ion concentration was also studied by
breakthrough curves and applying Bed Depth Service (BDST) and Thomas model. BET,
SEM-EDX, TGA, XRD and FTIR analysis were carried out to characterize the
biomasses. The whole study proved that selected agro-wastes have good removal
potential for U(VI), Zr(IV) and Sr(II) ions containing wastewater.
1
CHAPTER-1
_____________________________________INTRODUCTION
Maintenance and development in the quality of environment is one of the crucial
concerns of this century. Synthetic and natural pollutants; especially toxic heavy metal
cations, as most of these are persistent, non-biodegradable and carcinogenic, are
important factors accountable for strengthen degradation of the biosphere (Gupta et al.,
2011). Continuous increase in ecological pollution particularly water pollution by toxic
heavy metal ions leads to corresponding increase in the demand for precise and
responsive quantitative metal investigation in different environmental samples (Akhter et
al., 2009; Erikson and Donner, 2009).
Toxic heavy metal ions ground for physical distress and sometimes life-
threatening diseases and irreparable harms to vital body systems. To minimize the heavy
metal pollution; numerous process like adsorption, precipitation, coagulation, ion
exchange, reverse osmosis, electro-dialysis, cementation, and electro-coagulation have
been in use (Arshad et al., 2008; Kausar and Bhatti, 2013). Adsorption is nowadays
documented as an effectual and fiscal method for heavy metal wastewater management.
The adsorption technology presents flexibility in design and operation. In addition,
adsorption is sometimes reversible and adsorbents can be regenerated for recycling by
suitable desorption method (Akhter et al., 2009; Nurchi et al., 2010; Siege and Zuo,
2000).
Biosorption of heavy metals can be an effective method for the uptake and
recovery of heavy metal ions from aqueous systems. The bioadsorption of metal ions to
the biomass surface involves mechanism of either physical binding involving London–
vander waals forces or electrostatic attraction, or by chemical linkage such as ionic or
covalent binding between the adsorbent and the adsorbate. This technology is gaining
attention of many researchers as it is more cost-effective and poses less health hazards
than many of the current techniques (Akhter et al., 2009; Arshad et al., 2008).
Radionuclides are released into the environment through contaminated wastes
produced from a variety of industrial activities including mining, oil production and
electricity generation by nuclear power and also through accidental release. The presence
of radionuclides, even at low concentrations, is of major concern as they pose serious
radiological toxicity to living organisms. Conventional treatment techniques for the
2
cleanup of contaminated wastes are often expensive, inefficient and produce large
volumes of waste resulting in further disposal problems therefore it is essential to find
suitable alternatives which are inexpensive, efficient and can complement or replace
existing technologies. Biosorption, the accumulation of metal ions by biological
materials, is one of the possible innovative techniques (Ngwenya and Chirwa, 2010).
The severe accident of Fukushima Daiichi Nuclear Power Station has caused
radioactive contamination of environment including drinking water. Radioactive iodine
(I), uranium (U), caesium (Cs), strontium (Sr), barium (Ba) and zirconium (Zr) are
hazardous fission products of the high yield and/or relatively long half-life. Many
industrial activities dealing with radioactive materials create low, intermediate and high
level radioactive wastes, causing serious threats to our environment. The elimination of
radionuclides such as uranium, strontium, zirconium from wastewater is very important
issue in ecological controls (Kutahyali and Eral, 2010; Akhtar et al., 2008; Chegrouche et
al., 2009).
Considerable amounts of uranium (U) have found their way into the environment
through various nuclear and industrial activities, posing a threat not only to surface and
groundwater but also public health (Abdel Rahman et al., 2011). The United States
Environment Protection Agency (USEPA) set a maximum acceptable level of 30 μg L-1
and the World Health Organisation (WHO) strictly recommends a maximum level of 2 μg
L-1 for U (Saifuddin and Dinara, 2012). Hence, the removal of U from wastewater has
considerable importance.
During the last decades, the increased industrial use of zirconium (Zr) has
generated a potential risk of Zr contamination in the environment. Zirconium compounds
are used in the ceramic industry, glazes, refractories, enamels, and for electrical ceramics.
In nuclear industry zirconium (Zr) is used for cladding uranium fuel elements for nuclear
power plants (Akhtar et al., 2008). According to the available literature, stable isotopes
zirconium could have a low order of harmful for the living organisms. However, fission
reactions produces, the long half-life isotope 93
Zr (t1/2 = 106
years) in radioactive wastes.
Therefore, the understanding of Zr fate in the environment is required. Due to its low
solubility and strong affinity for polymerization, Zr is generally considered as immobile
and is used as a reference element in weathering processes studies (Monji et al., 2008).
Strontium (Sr) has two important isotopes i.e. 90Sr which emits β radiation with a
half-life of 28 years and 85Sr which is a ϒ emitter with a half-life of 64.8 days. Strontium
3
naturally occurs at an average amount of 0.04% and is 15th in abundance in the earth’s
crust (Chegrouche et al., 2009). The behaviour of strontium (Sr) isotopes in the soil,
which may be discharged to the ecosystem as a result of nuclear weapons testing nuclear
accidents comes into soil and plants, is of great interest. Beyond the four stable isotopes
which are naturally present in soil, Sr90 is also present in the surface soil almost
everywhere in the world as a result of fallout from past atmospheric nuclear weapons tests
(Bascetin and Atun, 2010). Strontium carbonate (SrCO3) is mostly used in making
electroceramics and X-ray absorbing glass for cathode ray tubes, oxide superconductors
(Guan et al., 2011). Strontium also has many commercial applications in optics, and it
produces the red flame colour of pyrotechnic devices such as fireworks and signal flares,
as oxygen eliminator in electron tubes and to produce glass for colour television tubes.
90Sr, with its long half-life is considered to be the more dangerous strontium isotope,
having a tendency to be retained within the living bodies, mostly in the bones, a source of
long term radiation of bone marrow (Bascetin and Atun, 2010; Kocherginsky et al.,
2002).
Reverse osmosis, precipitation, electrochemical treatment, solvent extraction,
flocculation, sorption on activated carbon (AC) and membrane processes are often
expensive, inefficient and produce toxic chemical sludge resulting in disposal problems
(Zhang et al., 2012). Conventional and most frequently used technique for the
remediation of heavy metals including U, Zr and Sr such as ion-exchange, are expensive
and less efficient. It is therefore necessary to find suitable alternative methods which are
affordable, efficient and can be complement or replace the existing methods. Biosorption
is one of the possible novel techniques involved in the remediation of heavy metals and
radionuclides from wastewaters (Saleem and Bhatti, 2011; Aytas et al., 2011).
Biosorption involves the accumulation of metals ions by biological materials either by
metabolically mediated methods or by purely physico-chemical means. Compared with
conventional treatment methods, biosorption is seen as a low cost, energy-saving
alternative, which has high efficiency and selectivity for absorbing metals in low
concentrations and operates over broad ranges of pH and temperature (Arshad et al.,
2008).
In recent years, number of agricultural wastes such as bagasse, sawdust, pine bark,
tree fern, spent grain, corn cobs, apple residue, hazelnut shells, coconut husk, rice husk,
coconut coir husk, coir pith carbon, potato peels, peat, tea leaves, orange peel, cocoa
shell, olive stone, walnut, hazel nuts, almond shells, barley straw and grape stalk have
4
been employed for the removal of metal ions (Rehman et al., 2008; Demibras, 2008) from
aqueous media, but high volumes of wastewater, still, demands exploration of newer
adsorbents. In many developing countries, the low-cost, high sorption capacity and easy
regeneration of agricultural biowastes has focused attention on their use for the
remediation of heavy metals from wastewater.
No attempts have been yet made to consume and understand the binding
mechanism of agro-wastes based on indigenous sources as sorbent for the removal of
uranium, zirconium and strontium ions, so this work will be a novel and cost effective
method for treatment of water loaded with these metal ions.
This research work was planned to search inexpensive and easily available
biosorbent with following objectives:
Exploration of agricultural waste biomasses such as rice husk, peanut husk,
bagasse, cotton sticks and rice bran for the removal of selected metal ions from
aqueous media.
Pre-treatments of biosorbents to boost up their adsorption capacity.
Equilibrium, kinetic and thermodynamic studies of sorption process.
Desorption of the sorbed metal ions for the recovery of biosorbents.
Characterization of selected biosorbents to understand the binding mechanisms of
sorbent-sorbate.
5
Chapter-2
____________________________REVIEW OF LITERATURE
The need for a cleaner world, more satisfying for both ourselves and the next age group,
has led to the advancement of approaches based on the “cleaner production” idea, which
refers to the constant application of integrated efficient defensive environmental methods
in order to reduce both the quantity and the toxicity of emissions and wastes (Fu and
Wang, 2011; Kikuchi and Sanaka, 2012).
Bioremediation is exploitation of plants or microorganisms to get rid of or
immobilize contaminants in soils or water and to reinstate the normal function of tainted
environment (Zuo et al., 2001). Among the existing remediation approaches for heavy
metal and radionuclide tainted environment, bioremediation is one of the most promising
procedures for developing countries and grounds for least disturbance to the ecological
unit. Adsorption is currently used method for radionuclides and other heavy metals ions
removal when low concentration of metals ions has to be removed or recovered. The
fundamental principle of adsorption is the transport of metal ions from the solution phase
to the active surface of adsorbent. The movement is controlled by suitable optimal
experimental conditions in the system for target metal ions, sorbent and desorption. The
main advantages of the adsorption are its flexibility, low cost, environment friendly
nature and regeneration (Veglio and Biolechni; 1997; Demirbas, 2008). Efficient removal
of uranium, zirconium and strontium by adsorption using different types of adsorbents,
viz. activated carbon, organic, inorganic, microorganisms, agricultural wastes, synthetic
materials etc. have been reported (Dushenkov, 2003; Miguel et al., 2006; Groudev et al.,
2008).
A huge amount of literature is being published and reported which mostly deals with
various adsorbents and particularly biosorbents previously being used for heavy metals
ions remediation of aqueous systems. Adsorption is a complex chemical/physical
phenomenon due to involvement of different factors on which it depends. Adsorption of
heavy metal ions by adsorbents has been reviewed by different reviewer in the past
(Alluri et al., 2007; Misra, 2009; Wang and Chen, 2009; Ngah and Hanafiah, 2009;
Opeolu et al., 2010; Nurchi et al., 2010; Malamis and Katsou, 2013; Kausar and Bhatti,
2013) and all literature shows high potential of various adsorbents for heavy metal ions
removal in a very economical way.
6
The biosorbents can be modified by physical pre-treatments using heat treatment,
autoclaving and vacuum drying, treating the biosorbent with chemicals like acids, alkali,
detergents, organic solvents or by mechanical disruption. These types of pre-treaments
modify the cell surface which is essential for biosorption either by removing or masking
the groups or exposing more metal binding sites (Gupta et al., 2000; Yan and
Viraraghavan, 2000; Chen and Wu, 2004). Preetreatment of biomasses can increase or
decrease sorption capacity of biomaterials (Cabuk et al., 2005; Bhatti et al., 2007; Zafar
et al., 2007; Nadeem et al., 2008; Bhatti et al., 2009; Nadeem et al., 2009).
One more very important and useful modification of biosorbents is immobilization
(Zhang and Banks 2006; Kiran et al., 2007; Vijayaraghavan and Yun, 2007;
Vijayaraghavan et al., 2008; Hanif et al., 2009; Asgher and Bhatti, 2010; Kumar et al.,
2011; Ullah et al., 2013) which possess high potential of regeneration and removal of
pollutants from aqueous media.
In the attempt to explore novel adsorbents with characteristic adsorption properties it is
vital to establish the best fitting adsorption equilibrium correlation, which is essential for
consistent estimation of adsorption factors and computable evaluation of adsorbent
behaviour for diverse adsorbent processes or for different experimental situations. For
proper investigation of adsorption, study of equilibrium of reaction provides vital
information. In equilibrium study, it is assumed that a relationship exists between amount
of adsorbate in solution and adsorbent surface. Equilibrium concentrations are dependent
on temperature so equilibrium process is studied at a specific temperature. Various
isotherms are used like Freundlich, Langmuir, Dubinin-Radushkevich, Flory-Huggins,
Halsey, Temkin, Redlich-Paterson and Sips; each describes different characteristic of the
adsorption process, but most important are Freundlich and Langmuir (Foo and Hameed,
2010). Search for the best fit adsorption isotherm and kinetic model using the method of
linear regression is the most widely used technique by researchers to predict the optimum
isotherm and kinetic models. Currently, non-linear regression method is found to be the
best way in selecting the optimum isotherm (Ho, 2004).
Adsorption equilibrium study reveals the efficacy of the adsorbent but adsorption
mechanism study needs information regarding kinetics of the process also. Adsorption
mechanism and rate-controlling step are explored by different types of kinetic models
have been exploited. Kinetic study of adsorption procedure is also helpful to select time
scale optimized situations for full batch study removal process. Various kinetic models
have been developed like pseudo-first-order, pseudo-second-order and Weber & Morris
7
sorption kinetic models. Inspite of the well-established and easy operational adsorption
process, still there is a great deal of confusion concerning the evaluation of experimental
conditions and data for proper description of adsorption process. Nature and feasibility of
reaction is well established after thermodynamic studies of the process. Various
thermodynamic parameters including standard Gibbs free energy (ΔGo), enthalpy change
(ΔHo) and entropy change (ΔSo) of adsorption are calculated from the temperature data
obtained from the adsorption process. Desorption study is also important to explore the
efficiency of the adsorbent in terms of regeneration of the adsorbent and recovery of the
sorbate for further applications (Zuo et al., 2001). High desorption efficiency with
appropriate desorbing agent is determined by trial and error method mostly using
different eluents at variable concentrations. Desorption efficiency is function of adsorbent
nature, forces of attraction between adsorbate and adsorbent, pH and temperature of
solution.
Changing the one variable while keeping other constant is classical and frequently used
method for optimization of experimental conditions. The main disadvantage of this
method is large number of experiments (Can et al., 2006) It is worthy to quote that the
response surface methodology does not elucidate the mechanism of the processes studied
but only ascertains the effects of factors upon response and the interactions between the
factors (Kalavathy et al., 2009). Adsorption is studied both in batch and column modes.
Basic parameters of adsorption are first optimized in batch mode and then large scale
application of the process is studied in column mode (Sadaf and Bhatti, 2013; Noreen et
al., 2013). A brief review of biosorbents used for sorption of metals focusing uranium,
zirconium and strontium is discussed taking into account pre-treatments, immobilization
of biosorbents and influence of various physicochemical factors influencing and
equilibrium, kinetic and thermodynamic modeling and finally the regeneration of
biosorbent in batch mode. The use of linear and nonlinear regression method in
optimization of kinetic and equilibrium data and response surface methodology is also
reviewed. Recent trends in column biosorption affecting parameters and modeling are
also reviewed briefly.
2.1. Batch biosorption
Akhtar et al. (2008) carried out sorption-desorption studies from diluted solutions
of Zr by using Candida tropicalis. Initial pH and metal ion concentration highly affect the
biosorption process. The maximum Zr ions biosorption capacity of C. tropicalis was
8
179 mg/g of biosorbent at optimized conditions with distribution coefficient value of
3968 mL/g. Biosorption process was best explained by Langmuir and pseudo first order
kinetic model at low concentration and by pseudo-second order at high concentration. To
study adsorption process different theoretical thermodynamic models were also
explained. Sorption-desorption studies were carried out by Na2CO3.
El-Kamash (2008) studied the synthesized hydrous titanium oxide for the removal
of Cs, Co and Sr ions from Cl waste solutions in batch mode sorption. The uptake of both
Sr and cobalt ions was found to be greater than that of Cs and biosorption capacity of
each ion was enhanced at higher temperature. The value of free energy (∆G°) was
decreased as increased the temperature, indicating that the sorption reaction for each
metal was favorable at higher temperature. The positive values of enthalpy (∆H°)
suggested that endothermic and chemisorption was the main mechanism involved in the
reaction.
Monji et al. (2008) examined the biosorption Zr (IV) and Hf (IV) onto RB, WB
and leaves of Platanus orientalis tree. Sorption affecting conditions like pH, contact time,
T, and metal ions concentration were optimized. The results indicated that sorption
equilibriums were attained in short time 1, 5 and 40 min for RB, WB, and leaves
respectively. Metal biosorption onto leaves showed pronounced by pH while RB and WB
showed no significant change with pH change. Both Freundlich and Langmuir were
employed to understand the data but Langmuir isotherm showed better results.
Thermodynamic studies showed the spontaneous nature of sorption process. In the
optimum conditions, the other metal ions such as Cu2+, Fe3+, Pb2+, Hg2+, La3+, Ce3+ were
not sorbed considerably as Zr(IV) and Hf(IV) ions, so these biomasses are first-rate
biosorbents for the acceptance of Zr and Hf ions from aqueous media.
Chegrouche et al. (2009) reported sorption of Sr(II) from aqueous solutions onto
activated carbon (AC) at optimum of conditions obtained as at : pH of 4.0, contact time =
8 h, initial 100 mg/L concentration of Sr(II), particle size = 270 μm and temperature of
293.15 K. Kinetic and equilibrium data followed pseudo-first order and Langmuir
isotherm. A dimensionless separation factor (RL) was used to judge the favourable
adsorption. Mass transfer coefficient βL (cm/s) at different temperatures indicated that the
velocity of βL of Sr(II) ions onto AC was slow. The Gibbs free energy (∆G°) showed the
physiosorption and feasibility of the process.
Ahmadpour et al. (2010) investigational almond green hull (AGH), eggplant hull
(EPH), and moss were used as sorbents for the sorption of Sr(II) from aqueous media,
9
showing AGH as most efficient sorbent (116.3 mg/g). The optimum doses of GH were
found to be 0.2 and 0.3 g for 45 and 102 mg/L for the maximum Sr(II) removal
respectively. Rapid Sr(II) removal was achieved in short time (3 min). The kinetics
followed the pseudo second-order. Both Langmuir and Freundlich models were well
fitted with the experimental isothermal data.
Başçetin and Atun (2010) studied the adsorption features of montmorillonite and
zeolite minerals and mixtures of both for 90Sr(II) removal by using a isotopic radiotracer
technique. Sr(II) adsorption was endothermic and spontaneous process. The calculation of
site distribution function by using the Freundlich isothermal parameters, provided
valuable evidences about mechanism of the reaction.
Ghaemi et al. (2011) reported adsorptive removal of Sr(II) by dolomite powder
with maximum adsorption capacity was found to be 1.172 and 3.958 mg/g for Strontium
and Ba(II) with greatest fitness of data with Langmuir isotherm respectively. The kinetic
was fitted well with the pseudo-second order. The thermodynamic studies indicated that
the sorption for both Sr and Ba ions was feasible and exothermic.
Sato et al. (2011) done work to remove I-1, IO3−, Cs and Ba by water purifiers
with efficiencies about 85, 40 and 75-90% and higher than 85 %, respectively using
adsorbents as purifiers in pot-type water purifiers. Sr was removed with initial
efficiencies from 70-100%, but was slightly reduced after each cycle of use. Synthetic
zeolite A4 efficiently removed Cs, Sr and Ba, but had no effect on I and Zr ions.
Mao et al. (2011) evaluated the efficacy of Pseudomonas alcaligenes for the
elimination of Sr(II) ions from aqueous system. Batch biosorption experimental data were
analyzed well by Langmuir isotherm models. The maximum removal capacity of 67.35
mg/g was obtained by at optimized pH value of 6.0 and 5 g/L. Kinetic data was fitted to
second-order rate expression. The FTIR analysis of Pseudomonas alcaligenes confirmed
different possible functional groups responsible for sorption.
Torab-Mostaedi et al. (2011) studied the removal of Sr and Ba ions from aqueous
system onto expanded perlite (EP). Effect of pH, interaction time, EP amount, and
temperature (T). Equilibrium and kinetic data was satisfactory analyzed by Langmuir,
pseudo-second order respectively. Thermodynamic studies endorse the exothermic and
spontaneous nature of the studied process.
Aydin et al. (2012) reported Posidonia oceanica (L.) Delile as an important sea
plant in the Mediterranean Sea (dead leaves) as adsorbent for uranium. Kinetic data
obeyed the pseudo-second order. Freundlich and D-R models were maximum uptake
10
capacities obtained as 5.67 and 9.81 mg g-1, respectively and (-∆G°) showed the
adsorption process was spontaneous in nature.
Ding et al. (2012) characterized tea waste by SEM-EDX before and after the
adsorption treatment. The uranium removal up to 86.80 % at optimum pH 6 in 12 h at 308
K described by the pseudo-second-order equation and Langmuir model. The amount of
adsorption increases from 22.92 - 142.21 mg g-1 with the decrease of tea waste dosage
from 100 to 10 mg for solution with an initial uranium concentration of 50 mg/L.
Desorption for the four strippants is higher than 80 %.
Kubota et al. (2012) reported removal of Cs-134, Sr-85, and I-131 were produced
by neutron irradiation of CsCl, SrCl2, and K2TeO3 using bentonite, zeolite, and AC using
real samples. Cs-134 and Sr-85 were effectively removed using bentonite and zeolite, and
I-131 was removed using activated carbon.
Kumar and Jain (2012) evaluated the removal efficacy of functionalized carbon
nanotubes through the experimental removal of Sr(II) from aqueous system. Sorption
affecting conditions were optimized like initial concentration of Sr (II), contact time and
pH. Adsorbent was characterized by SEM and FTIR. Equilibrium and kinetic data was
explained well with Langmuir and pseudo second order kinetic e.
Kumar et al. (2012) studied the removal of strontium (II) using silver (Ag) nano
particle saturated with Al2O3 prepared by reduction process and characterized by using
UV-Vis spectroscopy, XRD and SEM. All batch biosorption experimental conditions
were optimized as pH, initial Sr ion concentration. Freundlich model was fitted well to
data of equilibrium studies.
Park et al. (2012) reported that sorption competencey of montmorillonite, MnO2 -
coated montmorillonite (MOCM) and Fe coated montmorillonite (IOCM) to investigate
the single-and bi-solute competitive sorptions of Co(II), Sr(II)and Cs ions. Data was
fitted to Freundlich, Langmuir and D-R models in single-solute sorption system data well
(R2 > 0.95). In the bi-solute sorptions, the sorbed amount of one solute was decreased in
presence of other solute.
Yu at al. (2012) reported the Sr(II) removal using activated Na trititanate whisker
(STW) in batch system. The optimum conditions for 20 mg/L of Sr (II) were as pH (5.0),
STW amount ( 0.2 g), shaking time ( 5.0 min), and reaction time (3.0 h) with qmax 8.37
mg/g. Data followed Langmuir and pseudo-second order kinetic model. Thermodynamic
studies showed exothermic, spontaneous, and a physical nature of the studied reaction.
11
Yi and Li (2012) investigated the likelihood of chitosan powder as a novel type of
adsorbent for U(VI) removal from wastewater. Batch biosorption system with an initial
pH of 5.0 was most appropriate in the studied the temperature range from 20°C-70°C.
The % removal increased with increasing chitosan amount, while the adsorption capacity
decreased with qmax 175 mg/g.
Bhatti and Amin (2013) explored the potential of white rot fungus, Coriolus
versicolor to remove Zr ions from aqueous system. Optimal experimental conditions for
the removal of Zr using C. versicolor was examined for effect of medium pH, C.
versicolor concentration, and concentration of Zr ions, interaction time and
temperature(T). The isothermal studies showed that the ongoing biosorption reaction
obeyed the Langmuir equation. The values of (∆G◦) , (∆S◦) show that biosorption of Zr
onto C. versicolor was practicable, and spontaneous at ordinary room temperature. the
kinetic data indicated operation followed pseudo-second order process. Maximum
removal (71.0 mgZr/g) of C. versicolor s was seen under optimized conditions.
Hanif et al. (2013) studied the zirconium removal by live and dead mycelia of
Ganoderma lucidum is reported by demonstrate that at pH 3.5 biosorption capacity
value of 142.5 mg/g was obtained in 240 minutes. Langmuir and pseudo second kinetic
expression order models were fitted to equilibrium and kinetic studies respectively.
H2SO4 was proved good desorbing agent and characterization of biomaterial was done by
FTIR.
Hussein and Taha (2013) investigated uranium removal from a nitric acid
raffinate (waste) solution using prepared solvent (tri-butyl phosphate, TBP) immobilizing
PVC cement (SIC) as a suitable adsorbent. The studied relevant factors affecting uranium
adsorption onto SIC adsorbent involved; contact time, solution molarity, initial uranium
concentration and temperature. The obtained adsorption isotherm of uranium onto the SIC
adsorbent was fitted to Langmuir, Freundlich and Dubinin-Radushkviech (D-R)
adsorption models. The results showed that the obtained equilibrium data fitted well the
Langmuir isotherm. Additionally, it was found that the adsorption process obeys the
pseudo second-order kinetic model. On the other hand, the calculated theoretical capacity
of our prepared SIC adsorbent reached about 17 g U/kg SIC. Uranium adsorption from
the studied raffinate solution was carried out applying the attained optimum conditions.
The obtained data showed that 58.4 mg U/5 g SIC were adsorbed. However, using of 2 M
HNO3 solution as an eluent, 93 (54.3 mg U) from the adsorbed amount were eluted.
12
Keshtkar et al. (2013) reported adsorption capacity of polyvinyl alcohol/tetraethyl
orthosilicate/aminopropyltriethoxysilane (PVA/TEOS/APTES) nanofiber membrane
prepared by the sol-gel/electrospinning method and its use for adsorption of uranium from
solutions. The prepared membranes were characterized by FTIR, SEM and BET analysis.
Experimental results indicated that the pH (4.5) and high temperature (45°C at studied
condition) proceeded earlier than the adsorption of uranium ions onto the both of
prepared membranes. Detailed equilibrium and kinetic studies were done.
Thermodynamic studies showed the feasible, spontaneous and endothermic. Five
sorption-desorption cycles and the results showed that these membranes can be utilized
extensively in industrial activities.
Lian et al. (2013) evaluated the sorption capacity of sunflower straw for Sr2+
ionsfrom aqueous system and morphological studies of sorbent were done by FTIR and
SEM. Maximum removal capacity (17.48 mg/g) happened at about pH 3-7 and
equilibrium was achieved in 5 min and data followed pseudo second order kinetic model
and Langmuir for equilibrium data.
Park et al. (2013) explored the fishbone to remediate groundwater tainted with Co
and Sr through single- and bi-solute competitive sorptions. Freundlich, Langmuir and D-
R models were fitted to single-solute sorption experimental data with (R2 > 0.91). The
coefficients of determination indicated that all models fitted well for both Co and Sr.
Xia et al. (2013) reported the banyan leaves (BLs) as efficient biosorbent for
uranyl ions. The experimental study shows that the optimal removal effect was at seen pH
3.0, the initial U concentration was 100 mg/L, BLs amount 5 g L -1, and T was 293 K.
The biosorption kinetic data could be explained well by a pseudo-second-order model
preceded very quickly in 30 min, and got equilibrium in 50 min. The biosorption could be
described better by Freundlich isotherm. This design of U uptake BLs achieved by
multiple-molecule form, rather by single adsorption form. The thermodynamic studies of
∆H0, ∆S0, and ∆G0 parameters suggested that the processes was endothermic and
spontaneous. The sorption responsible sites of BLs for U, are hydroxyl(-OH), carbonyl
(C=O), P-O, and Si=O which played an important role in biosorption.
Yi et al. (2013) tested the adsorption of uranyl cations (UO22+) by apricot shell
activated carbon (ASAC) in a batch mode. The U(VI) reached an equilibrium state at 120
min in solution of pH( 6.0). Temperature increase slightly effected the U(VI) sorption.
The U(VI) removal efficacy was improved with increasing ASAC dose, whereas
adsorption capacity decreased with dose. Equilibrium data obeyed both Langmuir and
13
Freundlich isotherms and maximum removal was established 59.17 mg/g by Langmuir
isotherm. The reaction kinetics can be very well defined by the pseudo-first-order rate
expression.
Zhang et al. (2013) reported that synthetically prepared heat-treated carboxyl-rich
hydrothermal carbon spheres characterized using Boehm titrations, SEM, FT-IR and
elemental-analysis for good U removal and recovery from solution. The U(VI) sorption
capacity of HCSs after heat treatment was increased from 55.0 to 179.95 mg/g at 300 °C
for 5 h. Selective removal of studies U(VI) was good after heat-treatment in the presence
of other co-existing ions, Na1+, Ni2+ , Sr2+, Mn2+, Mg2+ and Zn2+. Regeneration by 0.05
mol/L hydrochloric acid for the recovery of U(VI). Excellent removal (99.0 %) of U(VI)
from 1.0 L industrial wastewater containing 15.0 mg U(VI) ions was done with 5.0 g.
2.2. Linear and non-linear regression analysis
Jumasiah et al. (2005) prepared activated carbon from agrowaste i.e. from palm
kernel shell, (PKS), were employed to remove a Basic Blue 9 from aqueous system.
Batch mode experiments were done at a fixed temperature (28°C). The sorption kinetics
and equilibrium of Blue 9 onto PKSAC were studied in detail. The isotherm data were
well described by the Re-P isotherm, with constants parameters calculated from non-
linear regression. The sorption kinetics of blue onto PKSAC was well defined by the
pseudo-second-order kinetic model.
Ho and Ofomaja, (2006) reported a new agrowaste t sorbent i.e. palm kernel fibre
(PKF) in West Africa and for the uptake of Cu ions from aqueous media. A comparison
was made between linear least-squares method and a trial-and-error non-linear regression
method of the pseudo-second-order kinetic equation for the sorption of Cu onto PKF.
Nouri et al. (2007) studied the Cd batch sorption process using wheat bran (WB)
as a sorbent from aqueous system. The effect of sorption operational conditions such as
contact time, cadmium initial concentration, WB mass, temperature, pH, agitation speed
and ionic strength on the sorption process of Cd was studied. Pseudo-second-order model
was estimated using the six linear forms and the non-linear curve fitting analysis. Kinetic
results show that sorption data was best fitted to non-linear pseudo-second-order kinetic
model. Isotherms data at different temperatures was determined and was explained with
equations such as Langmuir and Freundlich models with better fitness to tha Langmuir.
The detail study of five Langmuir linear equations as well as the non-linear curve fitting
14
analysis method was discussed. Results indicated that the non-linear method may be a
better way to obtain the Langmuir constant parameters.
Ncibi, et al. (2009) reported that Posidonia oceanica (L.) fibres has potential to
treat contaminated Cr(IV) aqueous system. Several adsorption kinetic models were
applied to fit the experimental data, as first-order, first-order (reversible and
irreversible), pseudo-second-order Brouers-Sotolongo, and Elovich equations using
both linear and nonlinear regression analyses.
Ncibi et al. (2009) used dried raw and modified Mediterranean green alga
Enteromorpha spp in batch biosorption experiments for the uptake of basic dye i.e.
Methylene blue, from aqueous system. Equilibrium data were fitted to five isotherms. The
results revealed that the experimental data were very well explained by the Langmuir for
the linear regression and both the Langmuir and R-P isothermal model for the non-
linear analysis.
Svilovic et al., (2009) studied the uptake of Cu ions from aqueous solutions
using zeolite 13X inbatch technique. Pseudo first and the second order models were
investigated for data fitness by using nonlinear regression while Weber-Morris model
by linear least squares method.
Rao et al. (2010) explored that Syzygium cumini L. leaf powder as sorbent for the
removal of Cd(II) from aqueous system. The Cu loaded Syzygium cumini L was
characterized using both FTIR and SEM. The biosorption of Cd(II) ions was studied
in batch sorption technique as a function of pH, contact time, adsorbate concentration,
Syzygium cumini L amount, anion and cation concentrations. The biosorption capacities
and kinetic rates transfer of Cd ions onto S. cumini L. were evaluated. The kinetics could
be best described by both linear and non-linear pseudo-second order models. The
isothermic data fitted in the order Freundlich>R-P>Langmuir>Temkin.
Zolgharnein, and Shahmoradi (2010) used a statistical experimental technique to
adjust the conditions for maximum uptake of Hg(II) by Fraxinus tree leaves through
a batch biosorption process. Sorbent-sorbate reaction nature was estimated by fitting
equilibrium data by nonlinear and transformed linear forms of the Langmuir, Freundlich,
and Redlich-Peterson isotherms. The study exposed that nonlinear regression is a more
reliable method for equilibrium study. The biosorption process was fast and was
monitored by pseudo-second order kinetic equation. Biosorption reaction mechanism was
evaluated by Fourier transform infrared (FT-IR) and X-ray diffraction (XRD) techniques.
15
Chowdhury and Saha (2011) carried out batch mode biosorption experiments
for elimination of malachite green from aqueous system using pre-treated rice husk (RH).
Four pseudo-second-order kinetic linear equations both linear and nonlinear forms were
discussed. The value of R2 and chi-square as error analysis to decide possible the best-
fitting equation to kinetic experimental data. The results proved that non-linear method
and chi square test are the best way to explain kinetic data.
Chowdhury et al. (2011) reported alkali treated rice husk (RH), by-product of the
agro-industry, for the uptake of safranin from aqueous batch adsorption system. The
equilibrium study was done using the two-parameter isotherms i.e. Freundlich, Langmuir,
and Temkin by linear and nonlinear regression methods to select appropriate sorption
equation for the experimental data. Four linearized Langmuir models were discussed. To
get best-fit isotherm predicted by each method, seven error functions namely,r2, SSE,
SAE, ARE, HYBRID, MPSD and the chi-square test were used. Nonlinear method is a
better way to find the isotherm parameters Langmuir isotherm model was best fitted to
experimental data.
Milosavljevic et al. (2011) used a new hydrogels derived from chi pH-sensitive
based on chitosan, itaconic and methacrylic acid as adsorbents for the uptake of Zn2+ ions
from aqueous system. The sorption procedure was well fitted to the pseudo-second order
kinetic equation. The hydrogels adsorbent was characterized by spectral (FTIR,
SEM/EDX and AFM analyses. The negative of ∆G° and ∆H° indicated that
the process was spontaneous and exothermic. The best appropriate isotherms suggested
by both linear and nonlinear methods were Langmuir and R-P isotherm.
Salman et al. (2011) used date seed (DS), an abundant and low-priced ordinary
sorbent in Iraq. DSAC was prepared by activated carbon by activation with KOH and
CO2 at 850°C for 3h and 37min.. The adsorption kinetic data were analysed by non-
linear fitting using adsorption of bentazon and carbofuran was better described by the
pseudo-second-order equation. Langmuir and Freundlich isotherm models both linear and
non-linear forms were fitted to equilibrium data. Equilibrium data fitted better with the
Freundlich model for both bentazon and carbofuran. DSAC showed higher adsorption for
carbofuran and bentazon.
Krusic et al. (2012) examined the potential of poly acrylamide-co-sodium
methacrylate (AAm/SMA) hydrogel for the removal of Pb2+. FTIR spectra showed that -
NH2 and hydroxyl groups are accountable for Pb2+ ion adsorption. It was found that the
Pb2+ ion followed pseudo-first-order kinetics. Nonlinear regression analysis of six
16
isotherms, Langmuir, Freundlich, R-P , Toth, D-R, and Sips, have been applied to the
sorption data, while the best fitness by R-P isotherm. Separation factor (RL), shows that
Pb2+ ion sorption is favourable, while -∆°G indicate that the lead ion adsorption process
is spontaneous.
2.3. Response surface methodology
Can et al. (2006) reported the 23 factorial central composite design to optimize the
pH, initial nickel ion concentration and Pinus sylvestris amount and removal as 100 %
was achieved just in 20 experiments. The optimum removal efficiency of Ni(II) was
calculated as 100 %. The experimental conditions at this possible optimum point were pH
= 6.17, Pinus sylvestris = 18.8 g/l, C0 = 11.175 mg/L and removal efficiency of nickel
was 99.91 % and model was highly fitted as R2 = 0.985 and adjusted R2 = 0.968) showing
a high significance of the model.
Kiran et al. (2007) examined capacity of alginate immobilized algal beads for the
removal of Cr from aqueous system using a novel cyanobacterium, Lyngbya putealis
isolated from metal containing soil under optimized conditions. Batch mode experiments
were done to study the adsorption equilibrium and kinetic performance of Cr in solution
allowing the calculation of kinetic constant parameters and maximum Cr sorption
capacity. Other parameters like initial Cr ion concentration (10–100 mgL-1), pH (2–6) and
temperature (25–45 ◦C) on Cr adsorption, applying Box–Behnken design. Very good
regression coefficient between the studied sorption variables and the response (R2 =
0.9984) shows superb estimation of experimental data by second-order polynomial
regression equation. The response surface method(RSM) showed that mgL-1 initial Cr
concentration, 2–3 pH and a temperature of 45 ◦C were best for biosorption of Cr by
immobilized L. putealis, giving 82% of the Cr elimination from the solution.
Tan et al. (2008) prepared AC from coconut husk (CH) using KOH +CO2
gasification method. The effects of three preparation affecting variables (CO2 activation
temperature, CO2 activation t and KOH: char) on the (2,4,6-TCP) uptake and AC yield
were explored. Using the CCD, two quadratic models were designed to see effect of the
preparation variables to the two responses. From the ANOVA, the most significant factor
on each experimental design response was recognized. The AC preparation affecting
conditions were adjusted by maximizing the 2,4,6-TCP uptake and AC yield. The
predicted 2,4,6-TCP uptake and AC yield agreed satisfactory with experimental values.
The optimum conditions for preparing AC from CH for adsorption of 2,4,6-TCP were as :
17
CO2 activation T of 750 ◦C, CO2 activation time of 2.29 h and KOH:char impregnation
ratio of 2.91, which lead to 191.73 mg/g of 2,4,6-TCP removal and 20.16 % of AC yield.
ANOVA for RS quadratic model for 2,4,6-TCP rmoval, the model F-value of 8.25
suggested that the model was significant. Values of Prob > F less than 0.05 showed that
the model was significant. From the ANOVA for response surface quadratic model for
AC yield, the model F-value of 24.75 implied that the model was significant.
Garg et al. (2008) observed the effect of sugarcane dose, pH and shaking speed on
Ni removal from aqueous solution. Batch mode experiments were carried out to measure
the adsorption equilibrium. The influence of three sorption affecting parameters on the
removal of Ni(II) was also examined using a RSM approach. The central composite face
centerd-CCD in RSM for designing the experiments and for full response surface
estimation. The optimum conditions for maximum removal of nickel from an aqueous
solution of 50 ppm were as follows: adsorbent dose (1500 mgL-1), pH (7.52) and shaking
speed (150 rpm). This was proved by the higher value of r2 = 0.9873. The value of R2and
adjusted R2 is close to 1.0 that is very high and advocating a high closeness between the
observed and the predicted responses. This shows that regression model providing superb
description of the connection between the independent factors (variables) and the %
adsorption (response).
Garg et al. (2009) studied the effect of succinic acid treated sugarcane bagasse
dose, pH and shaking speed for the uptake of Cr from aqueous system. The CC Face-
Centered Experimental Design in RSM by Design Expert Version 6.0.10 (Stat Ease,
USA) was used for designing the experiments as 20 trials suggested by model were
performed as well as for surface estimation of response. The optimum conditions for
maximum uptake of Cr from an aqueous solution of 50 mgL-1 were as: biomass dose (20
gL-1), pH (2.0) and agitation speed (250 rpm). This was shown by the higher value of
r2=0.9873.
Kalavathy et al. (2009) performed experiments designed by CC Rotary Design
using RSM by Design Expert Version 5.0.7 and 50 experimental trials were done for
optimization of 5 factors. The maximum uptake of Cu (II) i.e. 5.6 mgg-1 was calculated
under optimized concentrations of 35 mgL−1, temperature of 26 ◦C, C loading of 0.45 g
(100 mL)−1, adsorption time 208 min and pH ( 6.5) in batch mode. The value of
determination coefficient R2 0.9859 and adjusted R2 0.9798 suggested high significance
of the model.
18
Sahu et al. (2009) discussed the RSM as an well-organized method for predictive
model building and optimization of Cr sorption on prepared AC. In this study, the use of
RSM is presented for optimizing the uptake of Cr(VI) ions from aqua solutions using AC
as sorbent. A 24 full factorial CCD experimental design was used. ANOVA showed a
high R2 (0.928) and satisfactory prediction second-order regression model. The optimum
AC dose, T, initial Cr(VI) concentration and initial pH of the Cr(VI) solution were
established to be 4.3 g/L, 32 ◦C, 20.15 mg/L and 5.41 respectively. Under optimized
value of process parameters, high removal (>89%) was obtained for Cr(VI) removal. The
values of R2 and R2adj were found to be 0.888 and 0.785, respectively.
Singh et al. (2010) studied the removal of Rhodamine B dye using four-factor
CCD in RSM using synthetic nanocomposite. Quadratic model predicted the responses
of statisticalally designed experiments well. The ANOVA and t-test were used to test the
significance of the factors and their interactions. Suitability of the model was tested by
the closeness between experimental and predicted response and enumeration of prediction
errors. A high R2 = 0.97 among the predicted and the experimental values of the response
suggested for the suitability of the selected quadratic model in predicting the response
variable for the validation data set comprised of different combinations of the process
variables.
Sert and Eral (2010) synthesized NH2–MCM-41 sorbent and was characterized by
using XRD, SEM, BET, and FT-IR. This well characterized NH2–MCM-41 was
examined for U sorption using the batch experiments. The CCD design of RSM was
designated to know the effects of independent parameters and their interactions for the
uptake of UO2+2 ions. The optimum levels of the parameters calculated were 4.2 for the
initial pH, 600C for the T, 90 mgL-1 for the initial U ion concentration and 173 min for the
agitation time using the RSM. ΔH0 and ΔS0 were determined from the slope and the
intercept of plots of ln Kd versus 1/T. Langmuir, Freundlich, D–R isotherm have been
considered to explain the adsorption performances. The experiments were carried out by
the four independent process variables, initial pH, temperature, initial UO2+2
concentration and reaction time according to the central CCD, employing a total of 31
experiments. ANOVA of data was calculated at 95% confidence level. The F-test gave P
< 0.05 with a model F value of 18.20 which shows that this regression model is
statistically significant. The R2 of 94% showed that there was a high closeness between
the measured values and the predicted.
19
Anupam et al. (2011) investigated the adsorptive removal of chromium from
aqueous solution onto commercial PAC by using RSM to optimize best optimum
experimental conditions like pH, initial Cr6+ concentration, PAC dose, reaction time and
temperature (T) on adsorption efficacy. This results showed that with initial concentration
was 50 ppm, 100% Cr6+ removal was possible with pH 2 and 2 g L−1 PAC amount. The
experiments were performed according to central composite rotatable design (CCRD).
The optimum pH, PAC dose and time were found to be 2.32, 1.79 g L−1 and 25.76 min.
Here ANOVA of the regression model demonstrates that the model is highly important as
proof from the calculated F value (31.52) and a very low P = 0.000 value. The predicted
R2 of 0.7409 is in considerable agreement with the adjusted R2 of 0.9353.
Chatterjee et al. (2012) investigated the removal of dye i.e. Methylene Blue (MB)
using Design Expert software to get the optimum condition for removal of dye using CP,
four input parameters viz., initial concentration of MB (25–50 mg/L), amount of CP (0.2–
0.5 g), pH (5to9) and temperature (T) in range of 30–40 ◦C, performing the statistically
designed experiments with removal upto 93.4%. The values of R2 (0.9477) and R2adj
(0.9236) showed good fitness of the model.
Auta and Hameed (2011) prepared and used waste tea activated carbon (WTAC)
under optimum conditions for adsorption of both anionic and cationic dyes. The WTAC
was prepared through chemical activation with potassium acetate for sorption of MB and
AB29 dyes. RSM statistical technique was used to get optimum preparation situations
which were activation temperature, activation time and chemical impregnation ratio (IR);
with % yield and removal as the required responses. The R2 supporting the closeness
between the selected variables and the responses in respect to the predicted and measured
data were graphically represented and R2 values were 0.91 for MB and 0.92 for AB29.
Jain et al. (2011) studied the effect of three parameters like pH (2.0–7.0), initial Cr
concentration (10–70 mg/L) and treated Helianthus annuus amount (0.05–0.5 g/ 100 mL)
was studied for the removal of Cr(VI). Box–Behnken model experimental design
suggested only 17 experiments. The model is considered to be statistically significant
because the associated Prob > F value for the model is lower than 0.05.
Im et al. (2012) used O3/UV/H2O2 system to remove carbamazepine (CBZ) from
aqueous system. Predictions of responses calculated by statistical models were in close
agreement with the experimental findings, demonstrating the suitability of the procedure.
Fatima et al. (2013) proposed the U (VI) removal from aqueous solutions on
synthetic zeolite NaY using a 23 full factorial design to study the effect of the main effects
20
and interaction parameters for optimization of procedure. The pH is the most significant
parameter affecting U (VI) ions deposition onto zeolite NaY. Langmuir and pseudo-
second order followed the experimental data. Thermodynamic studies suggested the
exothermic and spontaneity of the reaction.
Han et al. (2013) used the Box–Behnken design of the RSM to optimize four most
significant adsorption parameters (initial As concentration, pH, temperature and time) and
to examine the interactive effects of these variables on As adsorption capacity of
mesoporous alumina (MA). According to ANOVA the interactive influence of initial As
concentration and pH on As(V) adsorption capacity was highly significant. The predicted
maximum removal capacity was about 39.06 mg/g, and the corresponding optimal
parameters of adsorption process: time 720 min, temperature 52.8 °C, initial pH 3.9 and
initial concentration 130 mgL-1 with the value of adjusted multiple R2 = 0.9697.
Muhamad et al. (2013) reported the potential of pilot-scale granular AC
sequencing batch biofilm reactor (GACSBBR) for removing (COD, ammoniacal nitrogen
NH3-N and (2,4-DCP) from recycled paper wastewater using a central composite face-
centred design (CCFD). Quadratic model with highly significant with value of R2 (>0.9)
obtained from the ANOVA.
2.4. Column biosorption
Steudel et al. (2007) worked on immobilized Bacillus sphaericus sorption in
column experiments with waters from a U remediation site in East Germany. In
experiments with U using real drainage waters, a specific U sorption capacity of 2.34
mg/g was determined.
Gurbuz (2009) explored the removal of the Cr(VI) ions from the aqueous phase
employing batch and column experiments using algae immobilised on silica gel. Results
showed that at pH 2 Scenedesmus obliquus and Arthospira maxima were employed as
adsorbents. The maximum uptake of Cr(VI) ions from the aqueous phase was 18.98 ±
0.32 mg/mg free S. obliquus and 18.37 ± 0.28 mg/mg immobilised S. obliquus. HCl was
proved as very effective for Cr(VI) ions.
Zou et al. (2009) ecplored the adsorption of U(VI) on the MnO2 coated zeolite
(MOCZ) in a fixed-bed column (pH 6) that of increase in bed height, decrease in flow
rate flow rate, small particle size showed more sorption capacity in presence of other
competing ions and also the breakthrough time was reduced. The Thomas model
explained all the experimental observations very well and BDST was used to see the
21
effect of bed height. For four adsorption-desorption cycles using 0.1 molL-1 NaHCO3
solution so MOCZ could be reused to adsorb U (VI) with good amazing sorption
capacity, compared to raw zeolite.
Kalavathy et al. (2010) studied the adsorption of Ni and Zn onto AC of Hevea
brasiliensis sawdust via batch and column mode under various operating conditions. The
qmax of Ni and Zn were 17.21 and 22.03mgg-1, respectively, at 30°C as by Langmuir
model. Kinetic experimental data followed pseudo second-order equation. Column
breakthrough curves were best described by Adam-Boharts model and Thomas model.
The desorbing agent used for the regeneration of the Ni and Zn was 0.1M H2SO4.
Das et al. (2012) carried out Zn biosorption onto yeast species viz. Candida
rugosa and Candida laurentii in aqueous environment in column. Significant
enhancement in Zn(II) uptake wasobserved using dead yeast biomass treated with anionic
surfactant SDS, analyzing using 2, 3 and 4 parameter isothermal models. Freundlich
model showed best fitness to the data. FT-IR analysis showed –NH, –C=O and –COOH
functional groups are responsible for binding of Zn(II) by yeast.
Zou and Zhao (2012) studied the U(VI) by citric acid modified pine sawdust
(CAMPS) in both batch and fixed-bed column modes sorption. The equilibrium data was
well explained by Langmuir and Koble-Corrigan models. In fixed-bed column, the effects
of bed height, flow rate, and inlet U(VI) concentration were studied by breakthrough
curve. The Thomas, the Yan and the BDST models were applied to the column data to
determine the characteristic parameters of the column adsorption.
Roy et al. (2013) explored the jute fiber for the removal of azo dye in both batch
and fixed-bed column mode. The batch sorption shows that sorption process was highly
dependent on different sorption affecting variables, namely, the pH, initial azo dye
concentration of, jute fiber dosage, reaction time, ionic strength, and temperature. Kinetic,
equilibrium and thermodynamic studies showed that pseudo-second order, Langmuir
isotherm and exothermic and spontaneous nature of the process. The column
performances were predicted by the application of the BDST model and Thomas model to
the experimental data. The adsorbent characterization was performed by FTIR and SEM
analyses.
Tofan et al. (2013) explored the sorbent for Co(II) ions uptake from aqueous
solutions in batch and column mode. Batch studies showed the maximum value of 7.5-7.8
mg/g when the initial pH of solution was 5 in the concentration range of (25-200 mgL-1).
Langmuir and pseudo-second order kinetic models were best fitted to experimental
22
equilibrium and kinetic data respectively. The fixed bed column removal (15.44 mg/g)
was better than batch and data was well analyzed by Thomas model.
Zhu et al. (2013) reported the removal of Sr(II) ions from solution by expanding
rice husk (ERH) in a fixed-bed column. The effects of different column design showed
that the equilibrium uptake (qeq ) of the ERH enhanced with the increase in initial Sr
concentration but decreased with the increase in flow rate and bed height respectively.
Adsorption capacities of 2.32 mgg-1 were obtained under the optimized conditions at a
flow rate of 10 mL/min and bed height of 6 cm and explained well by BDST model. XPS
analysis confirmed that the Sr(II) ion was absorbed.
23
Chapter-3
__________________________MATERIALS AND METHODS
The research work reported in this dissertation was carried out in the Environmental
Chemistry Laboratory, Department of Chemistry, University of Agriculture, Faisalabad
and Scottish Universities Environmental Research Centre, Scotland, United Kingdom.
3.1. Collection and preparation of biosorbent
Selected biomasses i.e., rice husk, cotton sticks, peanut husk, sugarcane bagasse, rice bran
and wheat bran were collected from different agro-fields and industries of Faisalabad,
Pakistan. Firstly biomasses were extensively washed with tap water and then three times
with double distilled deionized water (DDW) to remove water soluble surface
contaminants. After washing, biomasses were air dried at ambient temperature then cut,
ground and sieved to obtain a homogenous material of uniform size. The prepared
biomass material was then stored in desiccators until use. The sieve shaker (Octagon
Siever (OCT-Digital 4527-01)) was used to obtain the desired uniform size of biomass.
3.2. Chemicals
All chemicals such as UO2(NO3)2.6H2O, ZrOCl2.8H2O, N2O6Sr, Arsenazo III Dye,
Xylenol orange dye, DTPA, H2SO4, HNO3, HCl, EDTA, NaOH, MgSO4.7H2O, SDS,
CTAB, NH4OH, sodium alginate etc. were of analytical grade, purchased from Sigma-
Aldrich Chemical Co, USA. Stock solution of U, Zr and Sr ions were prepared by
dissolving the salt in double distilled water (pH 7, conductance (4 µS/cm) and working
standards of desired concentration were prepared by diluting the stock solution.
3.3. Analytical determination of metal ions
Quantification of uranium and zirconium concentration in sample solution was
determined using CECIL CE-7200 spectrophotometer. For uranium determination, 0.5
mL of sample solution was mixed with 1 mL of complexing solution of 2.5% DTPA and
0.5 mL Arsenazo-III in 25 mL volumetric flask. Finally, the volume was made up to the
mark by DDW of pH 2 and allowed to stand for 3-4 minutes, after which a pink-violet
coloration developed and the reading at 655 nm was noted against the corresponding
blank (Bhatti et al., 1991).
A solution of xylenol orange (0.05 %) was prepared by dissolving the dry powder in 0.6
N HCl. This reagent was added in the ratio of 2:23 (v/v) to sample solution containing up
24
to 50 µg of zirconium/mL of final volume. The solutions were mixed and allowed to
stand for approximately 10 min and absorbance was measured at 535 nm (Akhtar et al.,
2008).
The concentration of Sr(II) before and after sorption was determined by Optical Emission
Spectroscopy (Guan et al., 2011) on a Perkin Elmer OES-Optima 5300 DV.
3.4. Initial screening of biosorbents
Initially screening was carried out by adding 0.1 g of each biomass (rice husk, cotton
sticks, peanut shell, bagasse, rice bran and wheat bran) in 250 mL Erlenmeyer flasks
containing 50 mL of 100 mg L-1 U(VI) solution of pH 4 (most optimum in previous
literature). Solutions were shaken for 2 h at 125 rpm and then filtered (Whatman No 42
filter paper).
Screening experiment for Zr was done by adding 0.1 g of each biosorbent (rice husk,
cotton sticks, peanut shell, bagasse, rice bran and wheat bran) in 250 mL Erlenmeyer
flasks containing 50 mL of 50 mgL-1 Zr(IV) solution of pH 3.5 (most optimum in
previous literature). Solutions were shaken for 2 h at 125 rpm and then filtered (Whatman
No 42 filter paper). U and Zr containing filtrate were analyzed by spectrophotometric
method.
Screening experiment for Sr(II) was done by mixing 0.1 g of peanut husk and bagasse in
SARSTED (50 mL) tubes containing 25 mL of 10 mg/L of Sr(II) solution of pH(3-9) and
shaked for 2 h at125 rpm. After centrifugation and filtration the filtrate was analyzed for
Sr(II) quantification.
The biosorption equilibrium capacity of each metal ion per unit biomass (mgg-1) dry
weight of the biomass was calculated using formula
q C C VW (3.1)
Where Co and Ce are the initial and equilibrium concentrations of metal ions in solution,
V is volume of metal solution of desired concentration in litres and W is the amount of
biosorbent in grams. The pH each of each solution was adjusted with dilute NaOH and
HCl. Particle size of 300 µm of all biosorbent was in all experiments and shaking speed
of 125 rpm was kept constant for specified period in each experimental trial.
3.5. Pre-treatments of biomasses
After screening for each metal, selected biomasses were treated chemically by shaking
1.0 g of biomass with 100 mL of either 5 % HCl, HNO3, EDTA, NaOH, SDS, CTAB,
25
NH4OH, PEI, CaCl2, EDTA, glutaraldehyde, CTAB and Triton solutions for 2 h. Then
each treated biomass was extensively washed with deionized water. Biomasses were
physically modified by autoclaving (1.0 g of biosorbent/100 mL of water for 15 min.) and
boiling (1.0 g of biosorbent /100 mL of water for 10 min.). Finally, all chemically and
physically treated biomass samples were oven dried at 30 0C, ground with mortar and
pestle and were kept in air tight jars for further use.
3.6. Immobilization of biosorbents
Immobilization of the optimized biomasses was carried out by sodium alginate. Sodium
alginate (1.0 g) was dissolved in 100 mL (1 % w/v) of water by heating and then the
solution was cooled down to 40oC. 2g of each selected biomass was then added to each
100 mL mixture and stirred until a homogeneous mixture was formed. Then the mixture
was added drop wise into a solution of 1% CaCl2 (w/v) to form uniform beads of Ca-
alginate. After an hour, the beads were washed and were stored at 4oC in deionized
distilled water (Kiran et al., 2007; Hanif et al., 2009).
3.7. Batch biosorption
Metal containing effluents have a variety of chemical composition and their binding
interactions with biosorbents depend on the chemical structure of the particular metal ion,
the specific chemistry and morphology of the biosorbent surface and properties of the
metal ions in solution or wastewater. Therefore, it is necessary to see effects of
parameters like pH, biosorbent dose, contact time, initial metal ion concentration and
temperature on reaction to investigate true mechanism of reaction as well experimental
conditions optimization. The effect of different experimental parameters upon the
biosorption efficiency of native, sodium alginate immobilized and chemically treated
biosorbents was studied.
3.7.1 Effect of pH
To determine the optimum pH for biosorption, of U(VI) and Zr(IV) ions, experiments
were performed using 0.1g/50 mL of biosorbent at pH 2-9 and 1-4 respectively at
temperature of 30C and shaking speed of 125 rpm for 2 hours. For Sr(II) the experiment
was carried out by mixing 0.1 g/25mL of biosorbent at pH 3-9 at temperature of 30C
and shaking speed of 125 rpm for 2 hours. The experiments were performed using 50
mg/L of initial metal concentration for U(VI), Zr(VI) and 10 mg/L for Sr(II) ions.
26
3.7.2 Effect of biosorbent amount
The effect of biosorbent amount on biosorption of the selected metal ions was studied by
using different amounts (0.05-0.3g/50 mL for U(VI) and Zr(IV) and 0.05-0.3 g/25 mL for
Sr(II) solutions) of biosorbents. The experiments were performed using 50 mg/L of
U(VI), Zr(VI) and 10 mg/L for Sr(II) initial concentration at optimized pH for all metal
ions at 30C, 125 rpm shaking speed for 2 h.
3.7.3. Effect of contact time
The equilibrium time required by the biosorbent to bind to metal cations was determined
by adding biosorbent (0.05 g/50 mL) in solution of 50 mg/L of U(VI) and Zr(IV) ions
and shaking at 125 rpm and 30C for time periods until equilibrium was reached at
optimum pH. The same procedure was repeated for 25 mL of Sr(II) ions having initial
concentration of 10 mg/L.
3.7.4. Effect of initial metal ion concentration
Initial metal ion concentration is an important driving force to overcome all mass transfer
resistances of the metal ions between the aqueous and solid phases and affects the
efficiency of biosorption. Higher initial concentration of metal ions results in higher
driving forces for biosorption (Aksu, 2005). The experiments were carried out at different
initial metal ion concentrations by adding 0.05 g of biosorbent to metal ions at previously
optimized conditions.
3.7.5 Effect of temperature
The biosorption experiments were performed at different temperatures (30-60°C) under
previously optimized conditions.
3.8. Sorption kinetics
To understand the mechanism, controlling the biosorption, the most commonly used
pseudo-first and pseudo-second order kinetic models were used to interpret the
experimental data assuming that measured concentrations are equal to cell surface
concentrations. Linear regression analysis of kinetic models were performed using
Microsoft excel 2007 and non- linear by statistical software i.e. R -Version 2.15.1.
3.8.1. Pseudo-first order kinetic model
The pseudo-first order kinetic model (Lagergren, 1898) based on solid capacity, expresses
the mechanism of removal as a sorption preceded by diffusion through a boundary. It
considers that the sorption is partial first ordered depending on the concentration of free
sites. Pseudo-first order kinetic model is based on the fact that the change in metal ions
27
concentration with respect to time is proportional to the power one. The non-linear and
linear forms of model are given below
1 (3.2)
log log.
t (3.3)
Where qe and qt are the amount of metal ions adsorbed (mg/g) at equilibrium and at time t
(min), respectively, and k1 (min-1) is the pseudo- first-order rate constant. Values of k1 are
calculated from the plots of log(qe - qt) versus t.
3.8.2. Pseudo-second order kinetic model
Pseudo-second order kinetic (Ho and Mckay, 1999 Ho, 2006) model is based on the
assumption that biosorption follows a second rate kinetic mechanism. So, the rate of
occupation of sorption sites is proportional to the square of the number of unoccupied
sites. Linear and nonlinear forms of pseudo-first and second-order expressions used in
this study are presented below.
q (3.4)
t (3.5)
Where qe and qt in pseudo-second order equations are the amount of metal ions adsorbed
on adsorbent (mg/g) at equilibrium and at time t (min), respectively, and k2 is the pseudo-
second order rate constant (g/mg min). Based on the experimental data of qt and t, the
equilibrium sorption capacity (qe) and the pseudo-second-order rate constant (K2) can be
determined from the slope and intercept of a plot of t/qe versus t.
3.9. Equilibrium study
Adsorption isotherms are used to characterize the biosorption process and for evaluating
biosorption capacity. An isotherm describes the relationship between the amount of
sorbate sorbed and the metal ion concentration remaining in solution. Linear regression
analysis of equilibrium models were performed using Microsoft excel 2007 and non-
linear by statistical software i.e. R -Version 2.15.1.
28
3.9.1. Freundlich isotherm
The Freundlich (1906) equation is an empirical equation employed to describe
heterogeneous systems, in which it is characterised by the heterogeneity factor 1/n.
Hence, the empirical equation can be written:
q K C (3.6)
1/n is the heterogeneity factor. When n = 1, the Freundlich equation reduces to Henry’s
Law. A linear form of the Freundlich expression can be obtained by taking logarithms of
Eq (3.6)
log q log K 1 log C (3.7)
The values of KF and 1/n are calculated from the intercept and slope respectively in linear
regression method.
3.9.2. Langmuir isotherm
Langmuir (1916) proposed a theory to describe the adsorption of gas molecules onto
metal surfaces. The Langmuir adsorption isotherm has found successful application to
many real sorption processes of monolayer adsorption. The Langmuir equation is based
on the assumption of a structurally homogeneous adsorbent where all sorption sites are
identical and energetically equivalent. Theoretically, the sorbent has a finite capacity for
the sorbate. Therefore, a saturation value is reached beyond which no further sorption can
take place. The saturated or monolayer (as Ce →∞) capacity can be represented by the
expression
q
(3.8)
The Langmuir equation degenerates to Henry’s Law at low concentration
A linear expression of the Langmuir equation is:
Ce (3.9)
29
Therefore, a plot of Ce/qe versus Ce gives a straight line. Where qe is the amount of metal
ions sorbed on the biomass (mg/g) at equilibrium, Ce is the equilibrium concentration of
metal ions (mg/L), qm is the maximum biosorption capacity describing a complete
monolayer adsorption (mg/g) and Ka is adsorption equilibrium constant (L/mg) that is
related to the free energy of biosorption.
3.9.3. Redlich-Peterson isotherm
Redlich-Peterson (1959) incorporated three parameters into an empirical isotherm. The
Redlich-Peterson isotherm model combines elements from both the Langmuir and
Freundlich equation and the mechanism of adsorption is a hybrid one and does not follow
ideal monolayer adsorption. The Redlich-Peterson equation is widely used as a
compromise between Langmuir and Freundlich systems.
q (3.10)
Where A, B, and g are Redlich-Peterson parameters. When g=1 it becomes Langmuir
equation
q (3.11)
When g = 0 it reads like the Henry’s Law equation:
q (3.12)
Further non-linear form (3.9) can be converted into linear form by taking logarithms:
ln A 1 gln C ln B (3.13)
Redlich-peterson constant B and g can be calculated from linear plot of ln A 1
Vsln C . the value of redlich-peterson constant A can be calculated by by maximizing
R2 using trial and error method in Microsoft excel solver adds in function (Chan et al.,
2012).
30
3.10. Error analysis for kinetic and equilibrium models optimization
The optimization procedure requires the error functions in order to evaluate the best fit
isotherm to explain the experimental kinetic and equilibrium data (Kumar and Sivanesan,
2006; Foo and Hameed, 2010; Hadi et al., 2010; El Hamidi et al., 2012 ) In this study,
six non-linear error functions were examined using statistical software i.e. R-Version
2.15.1, by minimizing the respective error function across the time and concentration
range studied. The error functions employed are as follows:
The sum of the squares of the errors (SSE) (Boulinguiez et al., 2008)
, , (3.14)
Although this is the most common error function in use, it has one major drawback that
isotherm parameters derived using this error function will provide a better fit as the
magnitude of the errors and thus the squares of the errors increase-biasing the fit towards
the data obtained at the high end of the concentration range
A composite fractional error function (HYBRD) (Porter et al., 1999) an attempt to
improve the fit of the sum of the squares of the errors at low concentrations by dividing it
by the measured value. It also includes the number of degrees of freedom of the system.
The number of data points, n, minus the number of parameters, p, of the isotherm
equation as a divisor.
, ,
, (3.15)
Average relative error (ARE) (Kapoor and Yang (1989)
, ,
, (3.16)
This error function attempts to minimise the fractional error distribution across the entire
concentration range. Sum of absolute error (EABS) (Ng et al., 2003)
q , q , (3.17)
31
The approach is similar to the sum of square error function, with an increase in the errors
will provide a better fit, leading to the bias towards the high concentration data.
Marquardt (1963) developed the error function similar in some respects to a geometric
mean error distribution modified according to the number of degrees of freedom of the
system.
100 , ,
, (3.18)
Nonlinear chi-square test (Boulinguiez et al., 2008)
,
, (3.19)
Non-linear chi-square test is a statistical tool necessary for the best fit of sorption system,
obtained by judging the sum squares differences between the experimental and the
calculated data, with each squared difference is divided by its corresponding value
(calculated from the models). Small χ2
value indicates its similarities while a larger
number represents the variation of the experimental data (Boulinguiez et al. 2008).
If data from the model are similar to the experimental data, errors will be a small number
and if they are different, the error will be a large number. The subscripts “exp” and “calc”
show the experimental and calculated values and n is the number of observations in the
experimental data. Small error values suggest the better the curve fitting.
The general procedure to find an adequate model by means of the error functions is to
calculate the error function for all isotherms and make a comparison between values
obtained by different error functions for each isotherm. Overall, optimum parameters are
difficult to identify directly, hence, ordering results to try to make a comparison between
values of error functions can lead to meaningful results.
The value of coefficient of determination (R2) for non-linear regression was evaluated by
following formula (Boulinguiez et al., 2008)
R , ,
, , , _ ,
(3.20)
32
3.11. Thermodynamic study
The thermodynamic parameters for the adsorption process, namely Gibbs energy (Go),
enthalpy of adsorption (Ho) and entropy of adsorption (So) were determined by
carrying out the adsorption experiments at different temperatures and using the following
equations (Khan et al., 1995):
G° = H° – T S° (3.21)
Log (qe/Ce) = -ΔHo/2.303RT + ΔSo/2.303R (3.22)
Thermodynamic parameters ΔHo and ΔSo were computed from linear plot of Log (qe/Ce)
and 1/T from slope and intercept respectively using Microsoft Excel 2007 and ΔGo from
equation 3.21.
3.12. Effect of interfering ions
The effect of different cations like Ni2+, Pb2+, Co2+ , Mn2+ , Cd2+ , Cu2+ , Zn2+ (50, 75 and
100 ppm of each ion) and anions such as NO3-1, Cl-1, SO4
-2, PO43- (0.1 M of each anion)
onto U (50 ppm) biosorption onto native, SDS-treated and immobilized rice husk.
The effect of metal ions such as Ni2+ , Pb2+ , Co2+ , Mn2+ , Cd2+ , Cu2+ , Zn2+ and anions
such as NO3-1, Cl-1, SO4
-2, PO43-(0.1 M of each anion) onto Zr(IV) ions sorption onto
native, SDS-treated and immobilized bagasse.
The effect of metal ions such as Co2+ , Cu2+ , Ni2+ , Cd2+ , Zn2+ , Mn2+ , Pb2+ 5, 10 and 15
ppm of each anion Cl-1 , CH3COO-1 , SO43 , I-1 , PO4
3 (0.1 M each anion) onto Sr (II) ions
biosorption onto native, NaOH-treated and immobilized peanut husk.
3.13. Response surface methodology
Response surface methodology is a group of mathematical and statistical techniques that
are helpful for evaluating the effects of several independent variables and their
interactions on the response (Box and Draper, 1987). These techniques are based on the
fit of experimental data to the empirical models in relation to the experimental design.
This model provides relatively few combinations of variables to determine the complex
response function (Jain et al., 2011; Bezerra at al., 2008; Tavares et al., 2009). The most
popular response surface method is the central composite design due to its suitability to fit
quadratic surface which usually works well for process optimization. This design consists
of two level factorial design points, axial or star points and center points.
33
In this study, central composite design was used to study the effect of three variables; pH,
biosorbent dose and initial metal ion concentration with two levels on the sorption
capacity of biosorbents. A total of 20 experiments were performed in duplicate. Design
expert software (Stat Ease, 7.0.0 trial Version) was used for regression and graphical
analysis of sorption data. The chosen independent variables used in this study were coded
according to following equation.
∆ (3.23)
Where xi is the dimensionless coded value of the ith independent variable, X0 is the value
of Xi at the center point and ΔX is the step change value. The behavior of system is
explained by the following empirical second-order polynomial model Eq.
∑ ∑ ∑ ∑ ɛ, (3.24)
where Y is the predicted response, xi, xj, . . ., xk are the input variables, which affect the
response Y, x2i , x2
j , . . ., x2k are the square effects, xixj, xixk and xjxk are the interaction
effects, β0 is the intercept term, βi (i=1, 2, . . ., k) is the linear effect, βii (i=1, 2, . . ., k) is
the squared effect, βij (i=1, 2, . . ., k; j=1, 2, . . ., k) is the interaction effect and ɛ is a
random error The goodness of fit of the model was calculated using coefficient of
determination (R2) and the analysis of variance (Amini et al., 2008). Experimental ranges
and levels of these variables suggested by face-centered central composite design in
response surface methodology (RSM) using Design Expert 7.0.0 are given in Table 3.1.
34
Table 3.1.
Experimental ranges and levels of independent variables.
Factor range and Level (Coded) -1 0 +1
Uranium
pH(2-9) 2 5.5 9
Sorbent amount (0.0-0.3g/50 mL) 0.05 0.175 0.3
U(VI) concentration ( 10-100 mg/L) 10 55 100
Zirconium
pH(1-4) 1 2.5 4
Sorbent amount (0.05-0.3g/50 mL) 0.05 0.175 0.3
Zr(IV) concentration ( 25-200 mg/L) 25 112.5 200
Strontium
pH(3-9) 3 6 9
Sorbent amount (0.08-0.3g/25mL) 0.08 6 9
Sr(II) concentration ( 20-70 mg/L) 20 45 70
3.14. Desorption studies
Desorption studies help to elucidate the mechanism of adsorption and regeneration of
biosorbent making the treatment process more economical. Desorption studies to
regenerate the adsorbent were done using eluting agents such as EDTA, H2SO4, HCl,
NaOH and MgSO4, to compare their capacity to elute sorbed metal ions. To regenerate
the adsorbent, first U(VI) and Zr(IV) (50 mg/L) and Sr(II) (10 mg/L) were desorbed
under optimized conditions and the metal loaded residues dried in oven at 40 0C for 24 h.
The loaded biosorbent was then desorbed in 100 mL of 0.1 M solution of each selected
35
eluting agent, by shaking for one hour at speed of 125 rpm. Percent desorption was
calculated by formula
% 100 (3.25)
And
q C VW (3.26)
qdes is eluted metal content (mg g-1) and Cdes (mg L-1) is metal concentration in eluent
solution of volume V (L) and biomass weight (W) in gram.
3.15. Biosorbent characterization
Biosorbents were characterized physically and chemically by scanning electron
microscope equipped with carbon, hydrogen and nitrogen analysis, X-Ray Diffraction
(XRD), surface analysis (BET, BJH), energy dispersive X-Ray (SEM-EDX), Fourier
transform infra-red spectroscopy (FTIR) and thermogravimetric analysis (TGA).
3.15.1. Determination of elemental composition
The percentage of C, H and N was determined using an Exeter CE-440 Elemental
Analyzer. About 2.0 mg of biomass was weighed into a tin capsule and then transferred
into a combustion reactor for analysis. Acetanilide was run as standard, oxygen was used
as oxidant and helium served as carrier gas.
3.15.2. Determination of chemical composition
X-ray diffraction was used to determine the chemical composition of the sorbent. X-ray
diffraction analyses of native rice husk, bagasse and peanut husk were performed using a
Siemens D5000 X-ray Diffractometer operated at 40 kV and 40mA with CuKα radiation
(λ = 1.54056A˚ ). X-ray diffractograms were collected in the 2 θ range from 5◦ to 90◦,
using a step size of 0.02◦ and a counting rate of 1 s per step.
3.15.3. Determination of surface area
Specific studies of native biosorbents (rice husk, bagasse and peanut husk) determined by
BET (Brunauer, Emmett and Teller) and Barrett-Joyner-Halenda (BJH) methods were
performed on a surface area analyzer (NOVA 2200, Quanta Chrome, USA) using
nitrogen as a standard.
36
3.15.4. Determination of surface morphology
Surface composition of the sorbent was examined using a JEOL model 2300 Scanning
Electron Microscope equipped with an electron dispersive spectrometer (SEM-EDX)
before and after activation. Surface elemental composition of biosorbent was examined
by EDX. These analyses were conducted on each sample under optimized conditions
using Pt coating to avoid charge indulgence during SEM scanning in an Ar atmosphere at
current of 6 mA.
3.15.7. Determination of thermal stability
Thermal analysis was done using a Perkin Elmer Diamond Series unit (USA) which
consisted of a microbalance and a furnace through which inert Nitrogen, N2 flowed at 100
cc (STP) min-1. The heating rate was set at 10⁰C/ min-1 and temperature started from 30 to
1000⁰C.
3.15.6. Determination of functional groups
FT-IR analysis (IR Perkin Elmer 1600 spectrometer) of uranium and zirconium unloaded
and loaded rice husk and bagasse were carried out to identify the chemical functional
groups, responsible for sorption of metal ions. FTIR data were observed over 400-4000
cm-1 by preparing KBr disks of sorbent material and spectra were recorded on software
Bio-Rad Merlin.All Sr (II) unloaded and loaded peanut husk samples were recorded in a
Fourier Transform Infrared Spectrometer (FTIR-8400S, Shimadzu) using a silver gate
apparatus by measuring percentage of transmittance against wavenumber in the range of
600-4000 cm-1 and number of scans were 20-30 and resolution 2 cm-1.
3.16. Column biosorption
The adsorption performance of adsorbents in a continuous system is important factor in
accessing the feasibility of adsorbent in real applications. Continuous adsorption
experiments in a fixed-bed column were conducted in a glass column (20 mm ID and 43
cm height), packed with a known quantity of biomass. At the bottom of the column, a
stainless steel sieve was attached followed by a layer of glass wool. A known quantity of
the rice husk and bagasse was packed in the column to yield the desired bed height of the
sorbent (1, 2 and 3 cm). Metal ion solutions of known concentrations at pH 4 and 3.5 for
U and Zr respectively was pumped upward through the column at a desired flow rate (1.8,
3.6 and 5.4 mL/min) controlled by a peristaltic pump (Prominent, Heidelberg, Germany).
The metal ion solutions at the outlet of the column were collected at regular time intervals
37
and the concentration was measured using a double beam UV-visible spectrophotometer.
All the experiments were carried out at room temperature (30 ± 1°C).
Effluent volume (Veff) can be calculated as
Veff = F.t total (3.27)
Where ttotal and F are the total flow time (min) and volumetric flow rate (mL/min).
Breakthrough capacity Q0.5 (at 50% or Ct/Co = 0.5) expressed in mg of metal ion
adsorbed per gram of biosorbent was calculated by the following equation.
(Q50 (mg/g))= %
(3.28)
3.16.1. Thomas model
The linearized form of Thomas (1944) model can be expressed as follows
ln 1 K C t (3.29)
Where KTh (mL/min.mg) is the Thomas rate constant; qo (mg/g) is the metal ions uptake
per g of the biosorbent; Co (mg/L) is the metal ion concentration; Ct (mg/L) is the outlet
concentration at time t; W (g) the mass of biosorbent, and flow rate Q (mL/min) and t
(min) stands for flow time. A linear plot of ln[(Co/Ct) -1] against time (t) was employed
to determine values of kTh and qo from the slope and intercept of the plot respectively.
3.16.2. Bed-depth service time (BDST) model
The bed depth service model based on Bohart and Adams equation (1920) gives the
information about the linear relationship among the service time (t) and the bed depth (Z).
The following expression shows the BDST model.
t ln 1 (3.30)
where Co is the initial metal ion concentration (mg/L), Cb is the breakthrough metal ion
concentration (mg/L), U is the linear velocity (cm/min), No is the biosorption capacity of
bed (mg/L), Ka is the rate constant in BDST model (L/mg/min), t is the time (min) and Z
is the bed height (cm) of the column.
Above equation can be re written in the form of a straight line.
t= az-b
38
a = slope= No/CoU
b = intercept = 1 /KaCo ln (Co/Cb - 1)
The value of R2 showed the fitness of BDST model on the column data obtained at
different Cb/Cin ratios. By keeping linear velocity and inlet concentration constant, the
value of biosorption capacity N0 (mg/L) and rate constant Ka(L/mg.min) for respective
Cb/Cin ratio was estimated by using the slope and intercept value respectively.
3.17. Statistical analysis
All results were discussed by reporting means along with standard deviations. The
coefficients of equilibrium, kinetic and thermodynamic models were determined by using
regression techniques (Steel et al., 1997).
39
Chapter-4
RESULTS AND DISCUSSION
4.1. Screening of biosorbent
Initially experiments were performed to select potential and optimal agro-waste biomass
for U(VI), Zr(IV) and Sr(II) removal from synthetic aqueous solutions. The initial
screening experiments were carried out to select the biosorbent showing the best potential
for U(VI) uptake. Biosorption capacity of rice husk (RH), cotton sticks (CS), peanut husk
(PH), bagasse, rice bran (RB) and wheat bran (WB), were 26.84, 23.73, 23.71, 22.52,
21.78 and 21.70 mg/g respectively. It is clear from the obtained results (Fig.4.1) that all
biosorbents tested possessed good biosorption capacity for U(VI) but RH showed the
highest.
Fig. 4.1. Screening of biosorbent for U(VI) removal. (Biosorbent dosage= 0.1g/50mL,
C0 =100 mg/L, reaction time 2 h, T = 30 0C and initial pH= 4).
Screening experiment was performed to select potential and optimal agro-waste
biosorbent by adding 0.1 g of each biomass is separate flask along with 50 mL of 50
mg/L zirconium solution having initial pH 3.5. Adsorption capacity of each biosorbent
was calculated and Fig.4.2 shows the obtained results. The sorption capacity values for
bagasse, WB, CS, RB, RH, PH are 10.48, 8.83, 7.64, 7.83, 6.35 and 7.46 mg/g
respectively. Due to high uptake capacity; bagasse was selected for further studies.
0
5
10
15
20
25
30
Rice Husk CottonSticks
Peanut husk Bagasse Rice Bran Wheat Bran
Sorption Cap
acity (m
g/g)
Biosorbents
40
Fig.4.2. Screening of biosorbent for Zr (IV) removal. ( Biosorbent dosage 0.1 g/50 mL,
C0 = 50 mg/L, reaction time 2 h, T = 300C and initial pH =3.5).
The removal of Sr (II) was done using agro-wastes i.e. peanut husk and bagasse at initial
pH range 3-9, of mixture containing 10 mg/L of Sr(II) solution and 0.1 g of each biomass.
The mixture was shaked for 2 h, at 125 rpm and 30°C. The results obtained (Fig.4.3)
shows that peanut husk (1.4 mg/g at an initial pH of 9) has more Sr(II) removal potential
as compared to bagasse and was used in further experiments. To avoid metal precipitation
experiment was conducted upto pH 9.
Fig.4.3. Screening of biosorbent for Sr (II) removal. (Biosorbent dosage =0.1g/25 mL, C0 = 10 mg/L, reaction time 2 h, T = 300C and initial pH 3-9).
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
3 4 5 6 7 8 9
Sorption cap
acity (m
g/g)
pH
Peanut husk
Bagasse
0
2
4
6
8
10
12
Bagasse Wheat bran Cotton sticks Rice bran Rice husk Peanut husk
Sorption cap
acity (m
g/g)
Biosorbents
41
4.2. Effect of pre-treatments
Metal affinity to biomass can be modified by pre-treating the biomass with any base, acid,
salts or surfactant. The biosorption capacity (qe) values of untreated or native (control),
physically and chemically modified RH were in the following order: SDS (26.74 mg g-1)
> PEI (25.70 mg g-1) > MgSO4.7H20 (24.81 mg g-1) > boiling (24.72 mg g-1) > NaOH
(24.70 mg g-1) > benzene (24.05 mg g-1) > CaCl2 (21.63 mg g-1) > NH4OH (20.78 mg g-1)
> HNO3 (20.62 mg g-1) = autoclave (20.62 mg g-1) > NaNO3 (19.73 mg g-1), HCl (18.89
mg g-1), H2SO4 (17.91 mg g-1), Triton (17.64 mg g-1), EDTA (16.13 mg g-1),
glutaraldehyde (14.83 mg g-1), CTAB (14.56 mg g-1) and native (13.64 mg g-1).
Fig. 4.4. Effect of pre-treatments on U(VI) biosorption onto rice husk.
(Biosorbent dosage =0.1g/50 mL of each biomass, C0 = 50 mg/L, reaction time 2 h, T =
30 0C and initial pH= 4).
The results regarding the effect of pre-treatments for Zr (IV) removal are shown in Fig.
4.5. The biosorption capacity (qe) values of native biosorbent (No-treatment), physically
and chemically modified bagasse were in the following order: No-treatment (Native),
glutaraldehyde, CTAB, SDS, Triton, HCl, H2SO4, HNO3, PEI, EDTA, NH4OH, NaOH,
CaCl2, NaNO3, MgSO4.7H2O, autoclave, boiling 11.89, 9.11, 10.72, 13.49, 6.62, 5.33,
0
5
10
15
20
25
30
No treatm
ent
Gluterald
ehyde
CTA
B
SDS
Triton
HCl
H2SO
4
HNO3
PEI
EDTA
NH4OH
NaO
H
CaCl2
NaN
O3
MgSO
4.7H2O
Autoclave
Boilin
g
Benzen
e
Sorption cap
acity (m
g/g)
42
6.19, 7.939, 4.85, 6.31, 10.88329, 0.8169, 4.0795, 1.839, 2.6657, 9.5, 8.56 mgg-1
respectively.
Fig.4.5. Effect of pre-treatments on biosorption of Zr (IV) onto bagasse. (Biosorbent
dosage 0.1 g/50 mL, C0 = 50 mg/L, reaction time 2 h, T = 30 0C and initial pH =3.5)
Sr (II) affinity to peanut husk was modified using 5 % HNO3, H2SO4, HCl and 1 %
NaOH and removal efficiency was studied in the pH range 4-9 for all treated peanut husk
forms. The results showed that acids have no pronounced effect on uptake capacity and
decrease the sorption capacity as compared to untreated peanut husk but 1% NaOH has
tremendous increase in uptake capacity of peanut husk.
Fig.4.6. Effect of pretreatments on biosorption of Sr (II) onto peanut husk.
(Biosorbent dosage 0.1 g/25 mL, C0 = 10 mg/L, reaction time 2 h, T = 30 0C and initial
pH 4-9)
0
2
4
6
8
10
12
14
16
Sorption cap
acity (m
g/g)
0
0.5
1
1.5
2
2.5
3
4 5 6 7 8 9
Sorption cap
acity (m
g/g)
pH
1% NaOH
HCl
H2SO4
HNO3
43
An increase in the biosorption capacity of modified biosorbents can be attributed to
increased exposure of active metal binding sites caused by chemical modifications of the
cell wall components or removal of surface impurities. For example, basic pre-treatment
causes increase in biosorption capacity by removing lipids and proteins that mask binding
sites. Pre-treatment of biomass with acids may remove some mineral matter which will
increase access to metal binding sites. Of greater significance however, is the introduction
of oxygen surface complexes that change the surface chemistry by increasing the porosity
and surface area of the original sample (Safa et al., 2011). Surfactant pre-treatment
introduces lyophobic and lyophilic groups capable of adsorbing at the biosorbent surface:
solution interface. The adsorption of heavy metals onto biomass from aqueous solution
can be enhanced in the presence of surfactants due to reduced surface tension and
increased wetting power (Yesi et al., 2010). From all the modified treatments, SDS-
treated RH and bagasse showed maximum removal of U(VI) and Zr IV) ions respectively
and were selected for further biosorption optimization studies. This finding is
complimentary to the work of Chen et al. (2011), and Yesi et al. (2010) who reported an
increase in sorption capacity of surfactant modified silkworm exuviae and Bentonite
respectively. Das et al. (2012) also observed an increase in sorption capacity of two yeast
species for zinc (II) removal by SDS treatment. In case of Sr (II), NaOH treated treatment
was best for peanut husk biomass and was selected for further studies. The increase in
sorption capacity with the sodium hydroxide treatment has been reported by Jian et al.
(2013), Afkhami et al. (2007) for the removal of dyes and cations respectively.
4.3. Effect of initial pH
The initial pH of the solution is critical in controlling the equilibrium loading capacity of
the adsorption process. It affects the surface of the adsorbent and the chemistry of metal
ion in solution which, in turn, depends upon the concentration of metal ions. Ionization
state of the functional groups like carboxylate, phosphate, imidazole, and amino groups of
the cell wall is also affected by pH of the solution (Ozdemir et al., 2003; Elmaci et al.,
2007). The effect of pH on U(VI) sorption onto RH (Native, immobilized and SDS-
treated) was studied in the pH range 2-9. Fig.4.7 clearly illustrates that biosorption
capacity of native, SDS-treated and immobilized RH first increases with increasing pH
and then decreases. Maximum biosorption capacity was observed at pH 4 for native
(29.56 mg g-1) and immobilized (17.59 mg g-1), and pH 5 for SDS-treated (28.09 mg g-1)
biosorbent which is consistent with the optimum pH range for RH previously reported in
44
the literature. A further increase in pH does not favor increase in biosorption capacity.
This change in sorption capacity with pH can be explained by the change in uranyl ion
chemistry in solution at different pHs, which also depends on U ion concentration. In
acidic conditions UO22+ is the dominant species whereas at pH 4-5, monovalent uranyl
species UO2OH+, (UO2)2(OH)22+ [(UO2)3(OH)5
+] are commonly found. At very low pH,
the net charge on the biosorbent surface is positive which inhibits the approach of
positively charged species (Saleem and Bhatti, 2011). As pH is increased, functional
groups on the biosorbent surface such as carbonyl, phosphate and amino would be
available for adsorption hence maximum removal of U(VI) occurs at pH 4. U(VI)
biosorption onto RH is followed by ion-exchange processes between U(VI) ions and
protons introduced to the biosorbent surface of RH by acids. At very high pH, insoluble
precipitates of uranium such as schoepite (4UO3.9H2O) form in solution, decreasing the
uranium concentration in solution which subsequently leads to a lower biosorption
capacity of RH (Aytas et al., 2011; Zhou et al., 2012).
Fig.4.7. Effect of initial pH on U(VI) biosorption onto rice husk. (Biosorbent dosage
0.1g/50 mL, C0 = 50 mg/L, reaction time 2 h, T = 30 0C and pH 2-9).
The effect of pH on zirconium sorption onto bagasse (native, immobilized and SDS-
treated) was studied in the range of 1-4 by adding 0.1 g of bagasse in each flask
containing 50 mL of 50 mg/L zirconium solution. The influence of the initial pH on
biosorption of zirconium ions was evaluated in the pH range of 1-4. Experiments could
not be conducted at pH values above 4 because of visual precipitation of Zr(OH)4 at these
pH values which make estimation of the true sorption studies impossible (Guo et al.,
0
5
10
15
20
25
30
35
0 2 3 4 5 6 7 8 9
Sorption cap
acity (m
g/g)
pH
Native
SDS‐treated
Immobilized
45
2004). It is clear from Fig.4.8 that extremely acidic conditions did not favor zirconium
sorption and increase with increasing pH and maximum loading was observed at pH 4 for
native (14.098 mg/g) and immobilized (10.65 mg/g) while 3 for SDS-treated (14.87
mg/g) biomass. Further increase in pH does not favor increase in biosorption capacity.
The results obtained allowed establishing the optimum pH 3.5 for untreated, immobilized
biomass and 3 for SDS-treated. Akhtar et al. (2008) studied the biosorption of zirconium
by Candida tropicalis and observed maximum biosorption at pH 3.5. Monji et al. (2008)
reported that maximum biosorption of zirconium by using Platanus orientalis leaves
occurred in the pH 3.
Fig.4.8. Effect of initial pH on Zr(IV) biosorption onto bagasse. ( Biosorbent dosage
0.1 g/50 mL, C0 = 50 mg/L, reaction time 2 h, T = 30 0C and pH (1-4).
The effect of pH on strontium sorption onto peanut husk (native, NaOH-treated and
immobilized) was studied in the range of 3-9, adding 0.1 g of peanut husk in each tube
containing 25 mL of 10 mg/L strontium solution. It is clear from Fig.4.9 that acidic
conditions did not favor Sr (II) sorption and increase with increasing pH and maximum
loading was observed at pH 9 for native (1.45 mg/g) and 7 for immobilized (2.35 mg/g)
and NaOH-treated (2.76 mg/g) biomass. The results obtained allowed establishing the
optimum pH 9 for native and 7 for immobilized and NaOH-treated peanut husk. At low
pH values, competitive sorption of H3O+ ions and Sr2+ ions for the same positively
charged sites on the sorbents surface lowers the sorption capacity. With the increase of
pH values, the sorbents surface became more negative and electrostatic attraction between
0
2
4
6
8
10
12
14
16
18
1 1.5 2 2.5 3 3.5 4
Sorption cap
acity (m
g/g)
pH
Native
SDS‐treated
Immobilized
46
the Sr2+ and sorbent surface (Guan, et al., 2011) was likely to be increased. Similar results
have been found by several researchers for Sr2+ sorption on different adsorbents (Lu et al,
2008; Li et al., 2010). Ozeroglu and Keçeli (2006) studied the removal behavior of
strontium ions on a cross-linked copolymer containing meth acrylic acid. It is found that a
maximum adsorption of Sr(II) ions can be obtained after 30 minutes and at pH 8.
Fig.4.9. Effect of initial pH on Sr(II) biosorption onto peanut husk. ( Biosorbent
dosage 0.1 g/25mL, C0 = 10 mg/L, reaction time 2 h, T = 30 0C and pH (3-9).
4.4. Effect of biosorbent amount
Removal efficiency of any biomass is highly dependent upon sorbent amount as it
controls the sorbate-sorbent equilibrium of the sorption system. This is due to fact that the
number of available binding functional groups on the adsorbent surface is a function of
adsorbent amount. The effect of biosorbent amount on U(VI) biosorption was studied in
range 0.05-0.3 g/50 mL of 50 mg L-1 U(VI) solution and the results are illustrated in
Fig.4.10. Results indicated that a maximum biosorption capacity of 29.6, 31.6 and 27.8
mg g-1 was obtained for native, SDS-treated and immobilized RH respectively with 0.05
g. Further increase in biosorbent amount decreased the biosorption capacity which could
be due to the fact that the increase in biomass amount caused aggregation of the biomass
particles and subsequently decreased the available surface area for biosorption of U(VI)
ions (Saleem and Bhatti, 2011; Ofomaja and Ho, 2007).
0
0.5
1
1.5
2
2.5
3
3 4 5 6 7 8 9
Sorption cap
aacity (mg/g)
pH
Native
NaOH‐treated
Immobilized
47
Fig.4.10. Effect of sorbent amount on biosorption of U (VI) onto rice husk. Biosorbent dosage 0.05-0.3/50mL g, C0 = 50 mg/L, reaction time 2 h, T = 30 0C, pH 4 for
native and immobilized and pH 5 =SDS-treated.
The effect of dose on Zr (IV) sorption was studied in dose range of 0.05-0.3 g/50 mL and
results are depicted in Fig.4.11. Results indicate that maximum biosorption capacity of
30.95 mg/g, 35.89 mg/g and 21.94 mg/g was observed for native, SDS-treated and
immobilized bagasse respectively with dose concentration of 0.05 g. Further increase in
biosorbent dose decreased the biosorption capacity.
Fig.4.11. Effect of sorbent amount on biosorption of Zr (IV) onto bagasse.
(Biosorbent dosage 0.05-0.3 g/50mL, C0 = 50 mg/L, reaction time 2 h, T = 30 0C, pH=3.5
(native and immobilized) and pH 3 (SDS-treated)
0
5
10
15
20
25
30
35
0 0.05 0.1 0.15 0.2 0.25 0.3
Sor
pti
on c
apac
ity
(mg/
g))
Sorbent amount (g)
Native
SDS‐treated
Immobilized
0
5
10
15
20
25
30
35
40
0.05 0.1 0.15 0.2 0.25 0.3
Sorption cap
acity (m
g/g)
Biosorbent Dose (g)
Native
SDS‐treated
Immobilized
48
The effect of biosorbent dose on Sr (II) sorption was studied in dose range of 0.05-0.3
g/25 mL and results are depicted in Fig.4.12. Results indicated that maximum biosorption
capacity of 2.99 mgg-1, 5.24 mgg-1 and 4.32 mgg-1 was observed for native, NaOH-treated
and immobilized peanut husk respectively with dose amount of 0.05 g. Further increase in
biosorbent dose decreased the biosorption capacity.
Fig.4.12. Effect of sorbent amount on biosorption of Sr (II) onto peanut husk. (Biosorbent dosage 0.05-0.3 g/25mL, C0 = 10 mg/L, reaction time 2 h, T = 30 0C and
pH= 9 (native), pH 7 (NaOH-treated and immobilized).
This decrease in biosorption capacity with increased sorbent dose is explained
hypothetically by different scientists as increase in concentration may lead to aggregation
of biomass and subsequently decrease the available effective exposed area for biosorption
of metal ions. Decrease in uptake capacity of adsorbent surface may be attributed to lower
concentration of metal ions as depicted by decrease in removal efficiency with increased
sorbent amount (Tangaromsuk et al., 2002; Ahalya et al. 2005).
4.5. Effect of contact time
The effect of contact time on the biosorption of U(VI) by native, SDS-treated and
immobilized RH was investigated over the time intervals of 5 to 740 min as shown in
Fig.4.13. A maximum biosorption capacity value of 39.9, 41.0 and 31.9 U mgg-1 was
obtained for native, SDS-treated and immobilized RH respectively. During the initial
stages of the sorption process, adsorption rate was rapid, after which, uptake rate slowly
declined and tended to attain equilibrium at 320 min. It can be hypothesized that during
the initial stages of the adsorption process, the higher concentration of U(VI)) ions
0
1
2
3
4
5
6
0.05 0.1 0.15 0.2 0.25 0.3
Sorption cap
acity (m
g/g)
Sorbent amount (g)
Native
NaOH Treated
Immobilized
49
provide the driving force to facilitate ion diffusion from solution to the active sites of the
biosorbent. As the process continues, occupation of the active sites and the decrease of
the U(VI) ion concentration, leads to a decrease in uptake rate until equilibrium is
achieved. The equilibrium time for U(VI) biosorption by RH is in accordance with the
previously reported U biosorption studies on other biosorbents (Pang et al., 2010; Saleem
an Bhatti, 2011).
Fig.4.13. Effect of time on biosorption of U (VI) onto rice husk. (Biosorbent dosage
0.05g, C0 = 50 mg/L, reaction time 5-740 min T = 30 0C and pH 4 for native and
immobilized, pH 5 (SDS-treated)
The effect of contact time on the biosorption of Zr(IV) by native, SDS-treated and
immobilized bagasse was investigated over the time intervals of 5 to 740 min as shown in
Fig.4.14. A maximum biosorption capacity value of 35.2, 45.2 and 31.25 mg g-1 was
obtained for native, SDS-treated and immobilized bagasse respectively. During the initial
stages of the sorption process, adsorption rate was rapid, after which, uptake rate slowly
declined and tended to attain equilibrium at 160 min for native and SDS-treated and 320
for immobilized bagasse.
0
5
10
15
20
25
30
35
40
45
50
0 5 10 20 40 80 160 320 740
Sorption cap
acity (m
g/g)
Time (min)
Native
SDS‐treated
Immobilized
50
Fig. 4.14. Effect of time on biosorption of Zr(IV) onto bagasse. (Biosorbent dosage
0.05g, C0 = 50 mg/L, reaction time 5-740 min T = 30 0C and pH 3.5 (native and
immobilized, pH 3 (SDS-treated).
The effect of contact time on the biosorption of Sr(II) by native, NaOH-treated and
immobilized peanut husk was investigated over the time intervals of 5 to 320 min as
shown in Fig.4.15. A maximum biosorption capacity value of 3.81, 5.18 and 4.44 mg g-1
was obtained for native, NaOH-treated and immobilized peanut husk respectively. During
the initial stages of the sorption process, adsorption rate was rapid, after which, uptake
rate slowly declined and tended to attain equilibrium at 80 min for native and NaOH-
treated and 160 for immobilized peanut husk.
Fig.4.15. Effect of time on biosorption of Sr (II) onto peanut husk. (Biosorbent dosage
0.05g, C0 = 10 mg/L, reaction time 5-320 min T= 30 0C and pH 9 (native) and
7(Immobilized and NaOH-treated).
05
101520253035404550
0 5 10 20 40 80 160 320 740
Sorption cap
acity (m
g/g)
Time (min)
Native
SDS‐Treated
Immobilized
0
1
2
3
4
5
6
0 5 10 20 40 80 160 320
Sorption cap
acity (m
g/g)
Time (min)
Native
NaOHtreated
Immobilized
51
The results of the study revealed that adsorption took place in two phases where the metal
ion were physically/chemically up taken onto the surface of the biosorbent before being
taken up into the inner adsorption sites of dead cells (Nadeem et al., 2008; Riaz et al.,
2009). The first phase, known as a passive surface transport, took place quite rapidly,
while the second passive diffusion step transport, could take much longer time to
complete (Pavasant et al., 2006; Sekhar et al., 2003).
4.6. Biosorption kinetics
Kinetic study has an important role in technology transfer from laboratory to industrial
scale. Appropriate models can be helpful in understanding the process mechanisms,
analyzing experimental data, predicting plan for process optimization of future
operational conditions (Limousin et al., 2007). The rate of biosorption process depends
on the physical and chemical properties of the biosorbent material and the mass transfer
mechanism (Boulinguiez et al., 2008). The kinetic study also determines how a reaction
proceeded between sorbed metal ions i.e. U(VI), Zr(IV) and Sr(II) and solution by
following an appropriate pathway with the passage of time. A number of models have
been proposed in order to estimate the removal rate and the kinetic parameters to evaluate
the mechanism of the process.
To understand the controlling mechanism of biosorption, most commonly used pseudo-
first and pseudo-second order kinetic models were used to interpret the experimental data
assuming that measured concentrations are equal to cell surface concentrations. Linear
regression is frequently used to determine the best-fitting kinetic model as appealing
because of simplicity of equations in linear forms. In certain cases, it has been illustrated
that a different axis setting (during transformation of equation in linear forms) would alter
the regression results, influencing its consistency and accuracy However, during the last
few years, a increased interest in the utilization of nonlinear optimization modeling has
been noted. On the contrary, the nonlinear isotherm models are conducted on the same
abscissa and ordinate, thus avoiding such drawbacks of linearization. Few researchers
also report that it would be more rational and reliable to interpret adsorption data through
a process of linear and nonlinear regression (Rivas, et al., 2006; Ayoob and Gupta, 2008;
Han, et al., 2009; Foo and Hameed, 2010).
4.6.1 Pseudo-first order kinetic model
Pseudo-first order kinetic model is based on the fact that the change in metal ion
concentration with respect to time is proportional to the power one. Linear and non-linear
52
forms of pseudo-first order model (Section 3.8.1) were used for kinetic study of the U
(VI), Zr(IV) and Sr(II) onto rice husk, bagasse and peanut husk respectively. The criteria
of comparing R2 and also the maximum experimental qe with calculated qe values by
linear and non-linear method were employed to predict good fitness of the model.
The values of R2 calculated by linear and non-linear methods for uranium, zirconium and
strontium show that pseudo-first order kinetic model does not show good fit with the
kinetic sorption data of native, treated and immobilized forms of biosorbents used;
similarly the calculated and experimental maximum sorption capacity values are not in
good agreement with the experimental values as shown in Table 4.1, 4.2 and 4.3 So, the
first order kinetic model is not fitted well for whole data range of contact time. However,
immobilized peanut husk showing more fitness of data by pseudo-first order kinetic
model for Sr(II) removal.
4.6.2. Pseudo-second order kinetic model
The pseudo-second order model is based on the assumption that biosorption follows a
second-order rate mechanism. So, the rate of occupation of adsorption sites is
proportional to the square of the number of unoccupied sites. It expresses the sorption as
being partial second ordered with respect to free sites. The linear and non-linear forms of
pseudo-second order (section 3.8.2) equations were used to evaluate the behavior of U, Zr
and Sr ions sorption behavior onto selected biosorbents.
The biosorption mechanism over a complete range of the contact time is explained by the
pseudo-second order kinetic model. It was found that the pseudo-second order model is
best fit for all three metal ions sorption. Both Linear and non-linear regression analysis
showed the good fitness of the pseudo-second order kinetic model to the experimental
kinetic data of U(VI), Zr (IV) and Sr (II) biosorption onto all native and modified forms
of biosorbents. Immobilized peanut husk for Sr(II) removal showed better fitness of data
for pseudo-first-order in non-linear method as shown in Table 4.1-4.3. Small value of R2
for pseudo-second order by non-linear method as compared to linear method for Zr(IV)
sorption onto immobilized bagasse shows poor explanation of data by non-linear method
of regression. The R2 values obtained by non-linear method are comparatively small as
compared to linear method but agreement between calculated and experimental sorption
capacity values (qe) is fantastic as shown in Table 4.1, 2 and 3. The comparison between
experimental and predicted by pseudo-first and second-order model values is shown in
Fig.4.16-18.
53
Table 4.1.
Comparison of parameters of kinetic models for uranium sorption onto rice husk by linear
and non-linear regression methods.
Kinetic model
Pseudo-first order kinetic model Linear regression method
Non-linear regression method
K1(L min-1) qe calculated(mg/g) qe experimental(mg/g)
R2
Native SDS- treated
Immobilized Native SDS-treated Immobilized
0.000069 33.9
4.036 0.582
0.00138 42.25 10.44 0.562
0.0011 30.9 5.24 0.669
0.296 31.4 33.9 0.507
0.252 38.037 42.25 0.607
0.034 28.883
30.9 0.606
Kinetic model
Pseudo-second order kinetic model Linear regression method
Non-linear regression method
K2(g/mg min) qe calculated (mg/g)
qe experimental (mg/g) R2
Native SDS-treated
Immobilized Native SDS-treated Immobilized
0.0069 33.9
34.129 0.999
0.0064 42.25 40.48 0.999
0.000924 30.9 32.8 0.997
0.108 32.60 33.9 0.828
0.012 39.526 42.25 0.888
0.002 30.80 30.9 0.775
54
Fig. 4.16. Comparison of kinetic models for U(VI) sorption onto rice husk. a (Native),
b(SDS-treated), (c) immobilized
0
5
10
15
20
25
30
35
40
0 200 400 600 800
Sorption cap
acity qe
Time (min)
qe
Pseudo‐first order
Pseudo‐secondorder
(a)
0
5
10
15
20
25
30
35
0 200 400 600 800
Sorption cap
acity (qe)
Time (min)
qe
Pseudo first order
pseudo second order
(b)
0
5
10
15
20
25
30
35
40
45
0 200 400 600 800
sorption cap
acity (qe)
Time (min)
qe
Pseudo‐first order
pseudo‐second order
(c)
55
Table 4.2. Comparison of parameters of kinetic models for zirconium sorption onto bagasse by
linear and non-linear regression methods.
Kinetic model Pseudo-first order kinetic model
Linear regression method
Non-linear regression method
K1(L min-1) qe calculated(mg/g) qe experimental(mg/g)
R2
Native SDS-treated Immobilized Native SDS-treated
Immobilized
5.07 × 10-3
35.2 3.584 0.673
6.64 × 10-3
45.2 2.49 0.694
2.303× 10-3
31.25 5.50 0.747
0.354 33.168 35.2 0.426
0.457 43.627 45.25 0.442
0.350 27.911 31.25 0.353
Kinetic model Pseudo-second order kinetic model Linear regression method
Non-linear regression method
K2(g/mg min)
qe calculated (mg/g) qe experimental(mg/g)
R2
Native SDS-treated Immobilized Native SDS-treated
Immobilized
0.011 35.2 35.21
1
0.014 45.2 45.45
1
3.45 × 10-3 31.54 31.25 0.998
0.022 34.243 35.21 0.783
0.030 44.515 45.25 0.816
0.029 28.685 31.25 0.88
56
Fig.4.17. Comparison of kinetic models for Zr(IV) sorption onto bagasse. a (Native),
b(SDS-treated), (c) immobilized
15
20
25
30
35
40
0 200 400 600 800
Sorption cap
acity (m
g/g)
Time (min)
Experimental qe Pseudo‐first order Pseudo second order
(a)
38
39
40
41
42
43
44
45
46
0 200 400 600 800
Sorption cap
acity (m
g/g)
Time (min)
Experimental qe Pseudo‐ first order Pseudo‐ second order
(b)
20
22
24
26
28
30
32
0 200 400 600 800
Sorption cap
acity (m
g/g)
Time (min)
Experimental qe Pseudo‐first order Pseudo‐second order
(c)
57
Table: 4.3. Comparison of parameters of kinetic models for strontium sorption onto peanut husk by
linear and non-linear regression methods.
Kinetic model
Pseudo-first order kinetic model Linear regression method
Non-linear regression method
K1(L min-1) qe calculated(mg/g)
qe experimental (mg/g)
R2
Native NaOH-treated Immobilized Native NaOH-treated
Immobilized
0.0123 2.63 3.8
0.629
0.0241 0.153 5.18 0.515
0.0161 1.4662
4.44 0.883
0.4197 3.7157
3.8 0.715
0.5957 5.1545 5.18 0.854
0.1901 4.3023 4.44 0.918
Kinetic model Pseudo-second order kinetic model Linear regression method
Non-linear regression method
K2(g/mg min)
qe calculated (mg/g) qe experimental
(mg/g) R2
Native NaOH-treated Immobilized Native NaOH treated
Immobilized
0.319 3.8 3.8 1
0.894 5.19 5.18
1
0.084 4.47 4.44 0.999
0.328 3.799 3.8
0.971
0.665 5.197 5.18 0.957
0.073 4.524 4.44 0.880
58
Fig.4.18. Comparison of kinetic models for Sr(II) sorption onto peanut husk. (a)Native, b(NaOH-treated), (c), immobilized
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
0 50 100 150 200 250 300 350
Sorption cap
acity (m
g/g)
Time (min)Experimental qe Pseudo‐ first order Pseudo‐ second order
(a)
4.85
4.9
4.95
5
5.05
5.1
5.15
5.2
5.25
0 50 100 150 200 250 300 350
Sorption cap
acity (m
g/g)
Time (min)
Experimental qe Pseudo‐ first order Pseudo ‐second order
(b)
2
2.5
3
3.5
4
4.5
5
0 50 100 150 200 250 300 350
Sorption cap
acity (m
g/g)
Time (min)Experimental qe Pseudo first order immobilized
immobilized second order
(c)
59
4.7. Error analysis for optimization of kinetic model
Six non-linear error functions were used for kinetic equation optimization for biosorption
process of U(VI), Zr(IV) and Sr (II) ions onto RH, bagasse and peanut husk for pseudo-
first and pseudo-second order are given in Tables 4.4-4.6.
The error factions can be arranged in for U(VI) biosorption kinetic models onto native
rice husk from Table 4.4.
ERRSQ/SSE pseudo-first order>pseudo-second order EABS pseudo-first order>pseudo-second order ARE pseudo-first order>pseudo-second order HYBRID pseudo-first order>pseudo-second order MPSD pseudo-first order>pseudo-second order χ2 pseudo-first order>pseudo-second order
The same trend was observed for SDS-treated and immobilized rice husk. Due to very
small values of error functions as shown in Table 4.4, it is concluded that the second-
order-kinetic model is best fitted for the U(VI) sorption onto rice husk (native, SDS-
treated and immobilized). Wang et al. (2013) showed that pseudo-second order equation
is better fitted to describe uranium adsorption onto SBA-15. Adsorption process of U(VI)
on CMK-3 and PANI–CMK-3 was expressed by pseudo-second order kinetic model (Liu,
et al., 2013).
Table: 4.4.
Kinetic model optimization for U(VI) ions sorption onto rice husk by error functions.
Error Function
Native
SDS-treated
Immobilized
Pseudo-first order
Pseudo-second order
Pseudo-first order
Pseudo-second order
Pseudo-first order
Pseudo-second order
ERRSQ/SSE 32.293 11.215 47.830 13.628 186.306 106.444
EABS 12.650 8.608 16.754 9.169 33.263 25.611
ARE 5.355 3.613 6.020 3.292 24.673 18.373
HYBRID 18.265 6.296 23.315 6.666 199.022 109.250
MPSD 7.916 4.632 8.325 4.463 36.671 26.742
Chi-Sq/ χ2 1.096 0.378 1.399 0.400 11.941 6.555
60
The error factions can be arranged in from Table 4.5 for Zr (IV) biosorption kinetic
models onto native, treated and immobilized bagasse.
ERRSQ/SSE pseudo-first order>pseudo-second order EABS pseudo-first order>pseudo-second order ARE pseudo-first order>pseudo-second order HYBRID pseudo-first order>pseudo-second order MPSD pseudo-first order>pseudo-second order χ2 pseudo-first order>pseudo-second order It is concluded that the second-order kinetic model is best fitted for the Zr(IV) sorption
onto bagasse native, SDS-treated and immobilized. Thus, zirconium biosorption by C.
versicolor biomass followed the pseudo-second-order kinetics (Bhatti and Amin, 2013).
The pseudo-second-order kinetic model provided excellent kinetic data fitting for removal
of zirconium from aqueous solution by modified clinoptilolite (Faghihian and Kabiri-
Tadi, 2010).
Table: 4.5.
Kinetic model optimization for Z(IV) ions sorption onto bagasse by error functions.
Error Function
Native
SDS-treated
Immobilized
Pseudo-first order
Pseudo-second order
Pseudo-first order
Pseudo-second order
Pseudo-first order
Pseudo-second order
ERRSQ/SSE 29.36 11.118
18.532 6.116 32.963 25.596
EABS 14.118 8.686
10.678 6.768 14.188 11.808
ARE 5.506 3.416
3.108 1.978 6.2679 5.240
HYBRID 15.360
5.875
7.228 2.392 18.680 14.853
MPSD 6.963 4.327
4.118 2.372 7.999 7.220
Chi-Sq/ χ2 0.897
0.349
0.434
0.142
1.119
0.889
61
In case of Sr(II) removal the values of error functions obtained are very small suggesting
good agreement between models and experimental data. For Sr (II) removal onto native
and NaOH-treated peanut husk the error functions decrease in order given below
suggesting best fitness of pseudo-second order kinetic model to sorption data as shown in
Table 4.6.
ERRSQ/SSE pseudo-first order>pseudo-second order EABS pseudo-first order>pseudo-second order ARE pseudo-first order>pseudo-second order HYBRID pseudo-first order>pseudo-second order MPSD pseudo-first order>pseudo-second order χ2 pseudo-first order>pseudo-second order There is very minor difference in values of first and second-order kinetic models for Sr
(II) removal by immobilized peanut husk. However, the error functions suggesting best
fitness of pseudo-first order kinetic model to sorption data as shown Table 4.6.
Ahmadpour et al. (2010) reported that the kinetics of Sr(II) adsorption on almond green
hull was also examined and it was observed that it follows the pseudo second-order
behavior. Previous reports showed that adsorption of Sr(II) on activated carbon follows
pseudo-first order kinetics (Chegrouche et al., 2009).
Table. 4.6.
Kinetic model optimization for Sr(II) ions sorption onto peanut husk by error functions.
Error Function
Native
NaOH-treated
Immobilized
Pseudo-first order
Pseudo-second order
Pseudo-first order
Pseudo-second order
Pseudo-first order
Pseudo-second order
ERRSQ 0.062 0.006 0.009 0.003
0.227
0.332
EABS 0.58 0.189 0.192 0.102
1.109 0.953
ARE 2.292 0.745 0.536 0.2851
4.215
4.10
HYBRID 0.342 0.035 0.0375 0.011 1.204
2.003
MPSD 3.080 0.981 0.859 0.459
5.728
7.981
Chi-Sq 0.017
0.002
0.002
0.0005
0.060
0.100
62
4.8. Effect of initial metal ion concentration
The effect of changing U(VI) ion concentration was studied in the range of 10-100 mg L-1
by keeping the other parameters like pH, biosorbent dose 0.05 g, temperature 30C,
shaking speed (125 rpm) and contact time constant. The effect of U(VI) concentration is
shown in Fig.4.19 and illustrates the uptake capacity of native, SDS-treated and
immobilized RH increases rapidly before reaching constant value after a certain
concentration. The initial rapid increase is due to the availability of more active sites
which then become saturated. Tian et al. (2011) observed the same trend during uranium
sorption using oxime-grafted ordered mesoporous carbon CMK-5 for concentrations in
the range of 25-250 mg L-1.
Fig.4.19. Effect of initial metal ion concentration on U(VI) biosorption onto rice husk. (Biosorbent dosage 0.05g/50 mL, C0 = 10-100 mg/L, reaction time 320 min T = 30
0C and pH 4 (native and immobilized, pH 5 (SDS-treated). The effect of changing initial metal ion concentration on zirconium removal was studied
in the range of 25-200 mg/L by keeping the other parameters constant. The effect of metal
ion concentration is shown in Fig.4.20 and metal uptake capacity is very high at higher
concentration. Maximum biosorption capacity of 107.4, 111.4, 71.5 mg/g were obtained
for native, SDS-treated and immobilized bagasse respectively. The same trend of initial
Zr(IV) concentration on sorption removal efficiency of biomaterials has been reported by
Bhatti and Amin, 2013 and Hanif et al., 2013.
0
5
10
15
20
25
30
35
40
45
50
10 20 30 40 50 60 70 80 90 100
Sor
pti
on c
apac
ity
(mgg
-1)
Concentration (mg/L)
Native
SDS‐treated
Immobilized
63
Fig .4.20. Effect of initial metal ion concentration on biosorption of Zr(IV) onto
bagasse. (Biosorbent dosage 0.05g, C0 = 25-200 mg/L, reaction time 160 min (Native,
SDS bagasse), 320 min (immobilized) T = 30 0C and (pH 3.5 native and Immobilized, pH
3 = SDS-treated).
The effect of changing initial metal ion concentration on strontium removal was studied
in the range of 10-100 mg/L by keeping the other parameters constant. The effect of metal
ion concentration is shown in Fig.4.21 and metal uptake capacity of is very high at higher
concentration for immobilized peanut husk as compared to native and NaOH-treated.
Maximum biosorption capacity values of 9.4, 17.6, 38.04 mg/g were obtained for native,
NaOH- treated and immobilized peanut husk respectively. Guan et al., 2011 has reported
the same trend of initial Sr(II) ion concentration effect for Sr(II) ions sorption onto
potassium tetratitanate whisker and sodium trititanate whisker. Gok et al., 2013 studied in
detail the biosorption of radiostrontium by alginate beads and proved an efficient and
inexpensive method of Sr(II) ions removal.
0
20
40
60
80
100
120
25 50 75 100 125 150 200
Sorption cap
acity (m
g/g)
Concentration (mg/L)
Native
SDS‐treated
immobilized
64
Fig.4.21. Effect of initial metal ion concentration on Sr(II) biosorption onto peanut
husk. (Biosorbent dosage 0.05g, C0 = 10-100 mg/L, reaction time 80 min (Native,
NaOH-treated) and 160 (immobilized peanut husk) T = 30 0C and (pH 9 (native) and
7(Immobilized and NaOH-treated).
Initial metal ion concentration seems to have significant impact on biosorption, with a
higher concentration resulting in a high solute uptake. This is because at lower initial
metal ion concentrations, the ratio of the initial moles of solute to the available surface
area is low; subsequently, the fractional sorption becomes independent of the initial metal
concentration. However, at higher concentrations, the sites available for sorption become
fewer compared to moles of metal ions present; hence, the removal of solute is strongly
dependent upon the initial metal ion concentration (Ho et al., 2002; Binupriya et al.,
2007).
4.9. Equilibrium modeling
The design and operation of sorption processes require equilibrium sorption data for use
in mass transfer models which can then be used to predict the performance of the sorption
contact processes under a range of operating conditions.
Graphs from equilibrium data correlating the variation of solid phase concentration or the
amount of solute adsorbed per unit mass of solid (qe), to the variation of equilibrium
solution phase concentration (Ce) are termed sorption isotherms.
The most widely used two parameter sorption isotherms i.e. Langmuir, Freundlich and
three parameter equation i.e. Redlich-Peterson were used to determine the metal ions
biosorption mechanism from linearized and non-linearized forms.
0
5
10
15
20
25
30
35
40
45
10 20 20 40 50 60 70 80 90 100
Sorption cap
acity (m
g/g)
Concentration (mg/L)
Immobilized
Native
NaOHTreated
65
The detailed U, Zr and Sr removal equilibrium studies are discussed below by employing
both linear and non-linear regression methods.
4.9.1. Freundlich isotherm
The Freundlich isotherm was developed to describe heterogeneous systems and
exponential decay of energy distribution on sorption sites but it lakes fundamental
thermodynamics and not reduce to Henry’s law at low concentration (Ho et al., 2002).
The values of parameters calculated by linear and non-linear method for U(VI) are given
in Table 4.7. The magnitude of the Freundlich sorption capacity is an indication of
favorability of adsorption. The value of ranges from 2-10 indicate good adsorption
capacity, 1-2 moderate adsorption capacity and less than 1 indicate poor adsorption
capacity. As the values of n obtained from linear method are 2.25, 2.579 and 3.32 for
untreated, SDS-treated and immobilized RH respectively. The values obtained suggests
the favorability of the studied sorption process for U(VI) wastewater treatment.
The R2 values obtained from linear regression method 0.943, 0.982 and 0.969 suggest that
Freundlich model also present good fitness for the uranium sorption onto rice husk but the
Langmuir is more suitable according to linear method. The maximum sorption capacity
calculated from Freundlich model by linear regression method are 36.22, 40.92 and 35.31
mg/g for native, SDS-treated and immobilized rice husk. The criteria of closeness
between calculated and experimental qe values favor suitability of Freundlich model more
as compared to Langmuir by linear method.
According to non-linear method the R2 values for native, SDS-treated and immobilized
rice husk are 0.964, 0.975 and 0.949 respectively. The results also present the favorability
of sorption process explanation by Freundlich model but the Fig.4.22 obtained from non-
linear regression shows that experimental data is not very close to the Freundlich model
data.
In case of Zr(IV) removal (See Table 4.8) the value of R2 and agreement between
closeness of calculated and experimental values does not favor fitness of Freundlich
model to data. The values of n calculated for all forms indicate good moderate adsorption
process.
The Sr (II) removal data by peanut husk in native and modified forms is well fitted to
Freundlich as shown in Table 4.9. The criteria of closeness between calculated and
experimental qe values favor suitability of Freundlich model and the obtained values of
the n also show studied adsorption as favored process.
66
4.9.2. Langmuir isotherm
Langmuir developed a theoretical equilibrium isotherm model that is widely used in
research; both linear and non-linear forms are presented in Section 3.9.2. The values of
Langmuir sorption isotherm parameters calculated for uranium, zirconium and strontium
sorption process by linear and non-linear method are given in Table 4.7, 8 and 9
respectively.
The data used was from batch study performed to evaluate the sorption of U(VI) onto
native, SDS-treated and immobilized rice husk. The results presented in Table 4.7 shows
that R2 value of the linear regression were 0.997, 0.991 and 0.994 respectively for native,
SDS-treated and immobilized forms of rice husk. The experimental values of maximum
sorption capacity obtained for native, SDS-treated and immobilized forms are 38.9, 42.4
and 38 mg/g respectively. The maximum sorption capacity values calculated by Langmuir
for native, SDS-treated and immobilized forms of rice husk are 45.24, 47.16 and 40 mg/g
and are close to the experimental values. The criteria of R2 and closeness of experimental
and calculated values of model suggest good fitness of Langmuir adsorption isotherm to
the sorption equilibrium data of U(VI) for native and modified forms of rice husk. The
value of RL helps in estimating the nature of the sorption process.
RL value Nature of biosorption mechanism
RL > 1 Unfavourable
RL = 1 Linear
0< RL<1 Favourable
RL = 0 Irreversible
The values of RL obtained in the present study are in the range of 0-1(see Table 4.7),
describing that the biosorption process is favourable for uranium removal from
wastewater using rice husk. Both linear and non-linear regression shows high suitability
of Langmuir model for all forms of rice husk in the following order
Native> Immobilized > SDS-Treated
The R2 value obtained by non-linear regression is in following order 0.992, 0.975, and
0.947 for native, SDS-treated and immobilized forms respectively. Equilibrium data of all
three forms of bagasse is fitted to Langmuir model in following order Native>
immobilized>SDS-Treated by non-linear regression methods and on the basis of R2 and
closeness of experimental and model sorption capacity values as shown in Fig.4.22.
67
The results presented in Table 4.8 shows that R2 value of the linear regression were 0.967,
0.983 and 0.975 respectively for native, SDS-treated and immobilized forms of bagasse
for Zr(IV) removal. The maximum sorption capacity values calculated by Langmuir for
native, SDS-treated and immobilized forms of bagasse are not close to the experimental
values. The criteria of R2 shows good fitness of Langmuir adsorption isotherm to the
sorption equilibrium data of Zr(IV) for native and modified forms of bagasse. Closeness
of experimental and calculated maximum sorption capacity values of model is not good.
The values of RL obtained in the present study are in the range of 0-1(see Table 4.8),
describing that the biosorption process is favourable for zirconium removal from
wastewater using bagasse.
The results presented in Table 4.9 shows that R2 value of the linear regression were
0.9729, 0.966 and 0.819 respectively for native, NaOH-treated and immobilized forms of
peanut husk for Sr(II) removal by linear method. Closeness of experimental and
calculated maximum sorption capacity values of model suggest best fit of the results. The
values of RL obtained in the present study are in the range of 0-1(see Table 4.9),
describing that the biosorption process is favourable for strontium removal from
wastewater using peanut husk.
4.9.3. Redlic-Peterson isotherm
Redlich and Peterson proposed is an empirical three parameter equation,” which may be
used to represent adsorption equilibria over a wide concentration range. This equation
reduces to a linear isotherm at low surface coverage, and to the Langmuir isotherm when
g = 1. The exponent, g lies between 0 and 1. Thus, when g=1, the Redlich-Peterson equation
becomes the Langmuir equation, and, when g = 0, equation presents the Henry’s law. Linear
and non-linear forms of the Redlich and Peterson sorption isotherm are given in section
3.9.3. The experimental data used to evaluate this model is identical to that used for the
Freundlich and Langmuir isotherm evaluation.
The results given in Table 4.7 and Fig.4.22 show that Redlich -Peterson equation is more
suitable as compared to Freundlich and comparable with Langmuir isotherm for U(VI)
removal by RH as suggested by high R2 calculated by both linear and non-linear
regression analysis.
The equilibrium study of results Zr(IV) removal by bagasse are given in Table 4.8 and
Fig.4.23 for the comparative study of equilibrium model. The results shows that R2 values of
Redlich-Peterson 0.84, 0.914, 0.858 calculated by linear regression and 0.992, 0.895 and
0.918 by non- linear regression for native, SDS-treated and immobilized bagasse respectively.
68
The high values of correlation coefficient suggest that Redlich-Peterson comparatively much
betterthan Freundlich and comparable with Langmuir isotherm.
The Table 4.9 presents the equilibrium modeling results of Sr(II) removal by peanut husk.
The visual inspection of the R2 values showing high correlation coefficient values for native,
NaOH-treated and immobilized peanut husk for Redlich-Peterson isotherm by linear and non-
linear isothermal results.
The comparative values of experimental sorption capacity qe and predicted by Freundlich,
Langmuir and Redlich-Peterson for uranium, zirconium and strontium are presented in Fig.
4.22,23 and 24 respectively.
69
Table: 4.7. Equilibrium models parameters for U(VI) sorption onto rice husk by linear and non-linear regression methods.
Equilibrium model
Freundlich isotherm Linear regression method
Non-linear regression method
KF(mg/g)(L/mg)n
n
R2
Native SDS-treated Immobilized Native SDS-treated
Immobilized
7.186
2.25
0.9435
9.594
2.579
0.982
1.183
3.32
0.969
9.259
2.76
0.964
11.025
2.915
0.985
13.149
3.816
0.947
Isothermal model Langmuir isotherm Linear regression method
Non-linear regression method
qm(mg/g)
Ka(L/mg)
RL
R2
Native SDS-treated Immobilized Native SDS-treated Immobilized
45.24
0.099
0.101
0.997
47.16
0.129
0.079
0.991
40
0.212
0.049
0.994
38.66
0.213
0.0494
0.992
46.68
0.104
0.097
0.975
48.519
0.129
0.079
0.947
Isothermal model Redlich-Peterson isotherm
Linear regression method
Non-linear regression method
A (L/g)
B (dm3/mg)g
g
R2
Native SDS-treated Immobilized Native SDS-treated
Immobilized
4.4
0.071
1
0.994
12.5
0.552
0.834
0.998
19
1.435
0.708
0.998
5.811
0.198
0.900
0.995
15.031
0.834
0.772
0.996
13.936
1.853
0.797
0.961
70
Fig.4.22. Comparison of equilibrium isotherms for U(VI) sorption onto rice husk. (a)
Native, (b)SDS-treated and (c) immobilized
5
10
15
20
25
30
35
40
45
0 20 40 60
Sorption cap
acity (q
e)
Ce
qe
Freundlich
Langmuir
Redlich Peterson
0
10
20
30
40
50
0 20 40 60
Sorption Cap
acity qe
Ce
qe
Freundlich
Langmuir
Redlich Peterson
0
5
10
15
20
25
30
35
40
0 10 20 30 40 50 60 70
Sorption Cap
acity qe
Ce
qe
Freundlich
langmuir
Redlich Peterson
(a)
(b)
(c)
71
Table: 4.8. Equilibrium models parameters for Zr(IV) sorption onto bagasse by linear and non-linear regression methods.
Equilibrium model
Freundlich isotherm Linear regression method
Non-linear regression method
KF(mg/g)(L/mg)n
n
R2
Native SDS-treated Immobilized Native SDS-treated
Immobilized
12.43 1.817 0.840
18.845 2.206 0.838
9.727 2.410 0.807
23.534 0.364 0.794
29.246 0.321 0.785
17.422 0.318 0.745
Isothermal model Langmuir isotherm Linear regression method
Non-linear regression method
qm(mg/g)
Ka(L/mg)
RL
R2
Native SDS-treated Immobilized Native SDS-treated Immobilized
129.87 0.064 0.096 0.967
125
0.970 0.007 0.983
82.644 0.063 0.113 0.976
136.697 0.062 0.097 0.937
128.363 0.098 0.0642 0.888
88.163 0.056 0.125 0.901
Isothermal model Redlich-Peterson isotherm
Linear regression method
Non-linear regression method
A (L/g)
B (dm3/mg)g
g
R2
Native SDS-treated Immobilized Native SDS-treated
Immobilized
6.1
0.045 0.914 0.843
10.8 0.076
1 0.914
12.2 0.839 0.604 0.858
5.291 0.003 1.549 0.992
9.856 0.045 1.143 0.895
2.903 0.003 1.531 0.972
72
Fig.4.23. Comparison of equilibrium isotherms for Zr(IV) sorption onto bagasse. ( a)
Native, (b)SDS-treated and (c) immobilized.
0
20
40
60
80
100
120
0 20 40 60 80 100Sorption cap
acity (m
g/g)
Ce (mg/L)
Experimental qe Freundlich
Langmuir Redlich peterson
0
20
40
60
80
100
120
140
0 20 40 60 80 100
Sorption cap
acity (m
g/g)
Ce (mg/L)
qe experimental Freundlich
Redlich Peterson Langmuir
0
10
20
30
40
50
60
70
80
90
0 50 100 150
Sorption cap
acity (m
g/g)
Ce (mg/L)
experimental qe Freundlich
Redlich‐ Peterson Langmuir
(b)
(c)
73
Table: 4.9. Equilibrium models parameters for Sr(II) sorption onto peanut husk by linear and non-linear regression methods.
Equilibrium model
Freundlich Isotherm Linear regression method
Non-linear regression method
KF(mg/g)(L/mg)n
n
R2
Native NaOH-treated Immobilized Native NaOH-treated
Immobilized
3.005
3.344
0.990
8.752
6.527
0.957
10.26
2.474
0.892
3.854
0.2067
0.807
8.943
0.1469
0.896
7.112
0.594
0.961
Isothermal model Langmuir isotherm Linear regression method
Non-linear regression method
qm(mg/g)
Ka(L/mg)
RL
R2
Native NaOH-treated Immobilized Native NaOH-treated
Immobilized
8.889
0.429
0.032
0.973
16.502
0.522
0.027
0.966
49.75
0.142
0.070
0.819
9.383
0.242
0.012
0.882
15.268
1.162
0.012
0.714
34.24
0.770
0.014
0.782
Isothermal model Redlich-Peterson isotherm
Linear regression method
Non-linear regression method
A (L/g)
B (dm3/mg)g
g
R2
Native NaOH-treated Immobilized Native NaOH-treated
Immobilized
2.691
0.286
1.00
0.99
2840
272.162
0.8595
0.999
28
3.093
0.413
0.981
2.518
0.296
0.976
0.883
2497.62
272.8
0.86
0.897
4.399
0.021
1.379
0.967
74
Fig.4.24. Comparison of equilibrium models for Sr(II) sorption onto peanut husk. (a) Native (b) NaOH-treated (c) immobilized
0
2
4
6
8
10
12
14
0 20 40 60 80
Sorption cap
acity (m
g/g)
Ce(mg/L)
Experimental qe Freundlich
Langmuir Redlich peterson
0
2
4
6
8
10
12
14
16
18
20
0 10 20 30 40 50 60 70
Sorption cap
acity(mg/g)
Ce (mg/L)qe experimental Freundlich
Redlich Peterson Langmuir
0
10
20
30
40
50
0 5 10 15 20
Sorption cap
acity (m
g/g)
Ce (mg/L)qe experimental Freundlich
Redlich peterson Langmuir
(a)
(b
(c)
75
4.10. Error analysis for optimization of sorption isotherms
Due to the inherent bias resulting from the transformation which riding towards a diverse
form of parameters estimation errors and fits distortion, several mathematically rigorous
error functions are required for optimization procedure in order to evaluate the best fit
isotherm to explain the experimental equilibrium data (Foo and Hameed, 2010). In this
study, six non-linear error functions were examined and in each case a set of isotherm
parameters were determined by minimizing the respective error function across the
concentration range studied.
The general procedure to find an adequate model by means of the error functions is to
calculate error function for all isotherms and made a comparison for between values
obtained by different error functions for each isotherm. Overall optimum parameter set is
difficult to identify directly, hence, order to try to make a comparison between values of
error functions can lead to meaningful results.
The error functions for the native rice husk are in the following order
ERRSQ/SSE Frendlich>Langmuir>R-P EABS Frendlich>Langmuir>R-P ARE Freundlich> R-P> Langmuir HYBRID Freundlich> R-P> Langmuir MPSD Freundlich> R-P> Langmuir χ2 Freundlich> R-P> Langmuir The results obtained from error function for equilibrium biosorption studies of U(VI)
removal from wastewater showed that both Langmuir and Redlich-Peterson have good
co-relation with the experimental values as shown in Fig.4.22(a). There is a minor
difference between error function values of these two isotherms but overall trend of
results favors Langmuir isotherm sorption process for U(VI) onto native rice husk.
The error functions for the SDS-treated rice husk are in following order for each isotherm
ERRSQ/SSE Langmuir> Frendlich>R-P EABS Langmuir> Frendlich>R-P ARE Langmuir> Frendlich>R-P HYBRID Langmuir> Frendlich>R-P MPSD Langmuir> Frendlich>R-P χ2 Langmuir> Frendlich>R-P
For SDS-treated biomass the trend of error functions is in strong favor of Redlich and
Peterson sorption isotherm. This same trend is also clear from the equilibrium curve
Fig.4.22(b) obtained from the non- linear regression for all three isotherms.
For immobilized rice husk
76
ERRSQ/SSE Langmuir> Frendlich>R-P EABS Freundlich> R-P>Langmuir ARE Langmuir> Frendlich>R-P HYBRID Langmuir> R-P >Frendlich MPSD Langmuir> R-P >Frendlich χ2 Langmuir> Frendlich>R-P
The trend of error functions for sorption isotherm for immobilized rice husk is
complicated and visual estimation of above shown trend is not sufficient. If we see the
values of error in Table 4.10, very small values of the errors for Redlich-Peterson as
compared to Freundlich and Langmuir isotherm. So we concluded that equilibrium data
of the U(VI) sorption onto immobilized rice husk best fitted to Redlich-Peterson sorption
isotherm.
The error functions for the native bagasse are in the following order
ERRSQ/SSE Frendlich>Langmuir>R-P
EABS Frendlich>Langmuir>R-P ARE Freundlich> Langmuir >R-P HYBRID Freundlich> Langmuir> R-P MPSD Freundlich> Langmuir >R-P χ2 Freundlich> Langmuir> R-P The error functions for the treated bagasse are in the following order
ERRSQ/SSE Frendlich>Langmuir>R-P
EABS Frendlich>Langmuir>R-P ARE Freundlich> Langmuir >R-P HYBRID Freundlich> Langmuir> R-P MPSD Freundlich> Langmuir >R-P χ2 Freundlich> Langmuir> R-P The error functions for the immobilized bagasse are in the following order
ERRSQ/SSE Frendlich>Langmuir>R-P EABS Frendlich>Langmuir>R-P ARE Freundlich> Langmuir >R-P HYBRID Freundlich> Langmuir> R-P MPSD Freundlich> Langmuir >R-P χ2 Freundlich> Langmuir> R-P
The trend of error functions for sorption isotherm for bagasse is simple and visual
estimation of above shown trend concluded that equilibrium data of the Zr(IV) sorption
onto native, SDS-treated and immobilized bagasse is best fitted to Redlich-Peterson
sorption isotherm.
77
The error functions for the native peanut husk are in the following order
ERRSQ/SSE Frendlich>Langmuir =R-P EABS Frendlich>Langmuir>R-P ARE Freundlich> Langmuir >R-P HYBRID Freundlich> Langmuir> R-P MPSD Freundlich> Langmuir >R-P χ2 Freundlich> Langmuir> R-P However there is very minor difference between values of all error function for Langmuir
and Redlich-Peterson sorption isotherm (Table 4.12).
The error functions for the NaOH-treated peanut husk are in the following order
ERRSQ/SSE Langmuir > Frendlich> R-P
EABS Langmuir > Frendlich> R-P ARE Langmuir > Frendlich> R-P HYBRID Langmuir > Frendlich> R-P MPSD Langmuir > R-P > Frendlich χ2 Langmuir > Frendlich> R-P
The error functions for the immobilized peanut husk are in the following order
ERRSQ/SSE Langmuir > Frendlich> R-P
EABS Freundlich>R-P>Langmuir ARE R-P>Freundlich>Langmuir HYBRID R-P>Langmuir>Freundlich MPSD R-P>Langmuir>Freundlich χ2 R-P>Langmuir>Freundlich
The trends of error functions for sorption isotherm for peanut husk are simple and visual
estimation of above shown trend can conclude that equilibrium data of the Sr(II) sorption
onto native and NaOH-treated is best fitted to Redlich-Peterson sorption isotherm and
immobilized peanut husk is well fitted to Freundlich sorption isotherm. Ahmadpour et
al., 2010 reported previously that biosorption of Sr(II) onto selected biomaterials was
well explained by both Freundlich and Langmuir isothermal model. The previous
research reports of Sr(II) removal onto montmorillonite and zeolite and mixtures of both
adsorbents was well explained by Freundlich isotherm (Bascetin and Atum., 2010). The
same trend was shown for Sr(II) removal by nano-particle impregnated by alumina
(Kumar et al., 2012).
78
Table: 4.10. Optimization of equilibrium isotherm for U(VI) sorption onto rice husk by error functions.
Error
Function
Native
SDS-Treated
Immobilized
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
ERRSQ 35.969 8.162 5.233 16.969 28.524 4.282 43.142 44.948 33.49
EABS 14.66 7.268 6.010 11.462 14.721 5.281 17.811 13.426 16.22
ARE 9.060 2.917 3.062 5.671 7.378 1.762 7.427 9.100 7.377
HYBRID 35.96 4.051 3.910 12.785 21.316 1.726 21.738 53.038 22.99
MPSD 19.66 4.185 5.351 10.557 13.429 2.356 10.435 23.187 11.74
Chi-Sq/ χ2 2.877 0.324 0.274 1.023 1.705 0.121 1.739 4.243 0.274
79
Table: 4.11 Optimization of equilibrium isotherm for Zr(IV) sorption onto bagasse by error functions.
Error
Function
Native
SDS-treated
Immobilized
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
ERRSQ
1397.26
425.59 54.943
1513.37 787.17 736.64
691.14 267.60
76.640
EABS
91.448 52.97 14.869
92.17 63.34 56.294
58.38 36.728
18.419
ARE
26.62 14.42 4.1859
24.24 13.28 11.092
21.54 12.464
5.055
HYBRID
626.12 165.58 24.187
543.26 209.05 239.6
375.88 120.26
31.857
MPSD
47.33 22.90 7.8106
38.78 17.80 18.32
38.75 19.968
7.812
Chi-Sq/ χ2
21.781
6.900
0.954
14.662
5.608
1.321
23.837
11.23
10.281
80
Table: 4.12. Optimization of equilibrium isotherm for S(II) sorption onto peanut husk by error functions.
Error
Function
Native
NaOH-Treated
Immobilized
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
Freundlich isotherm
Langmuir isotherm
Redlich-Paterson isotherm
ERRSQ
5.985 3.646
3.621 14.074
38.78
13.914
46.488
41.122
39.77
EABS
6.070 5.027 4.866 9.342
15.035
9.159
16.841
15.331
15.406
ARE 8.625 7.105 6.7147 6.729
17.321
6.311
13.735
13.459
13.845
HYBRID
10.528 6.232 6.824 11.461
69.129
12.855
46.851
58.558
67.743
MPSD
12.948 9.658 9.807 8.869 35.195 9.245
26.455
33.047
35.792
Chi-Sq/ χ2 0.7528
0.486
0.4575
0.948 65.029
0.934
8.124
48.647
58.825
81
4.11. Effect of temperature
Temperature of the solution is important factor during the process of biosorption. It
affects the interaction between the biomass and the metal ions, usually by influencing the
stability of the metal–sorbent complex, and the ionization of the cell wall moieties (Sag
and Kutsal, 1995). The effect of temperature on biosorption of U(VI) ions onto native,
SDS-treated and immobilized RH is shown in Fig.4.25. The effect of temperature on the
biosorption process was small and the maximum biosorption capacity was obtained at
30C. Decrease in the biosorption capacity was observed at high temperature and the
effect was more pronounced in SDS-treated RH as compared to native and immobilized
forms.
Fig.4.25. Effect of temperature on U(VI) biosorption onto rice husk. (Initial pH 4 for native and immobilized, pH 5 for SDS-treated). (Biosorbent dosage 0.05g/50mL, C0 = 50
mg/L, reaction time 320 min T = 30 0C.
The effect of temperature on biosorption of Zr(IV) ions onto native, SDS-treated and
immobilized bagasse is shown in Fig.4.26. The effect of temperature on the biosorption
process was prominent and the maximum biosorption capacity was obtained at 30C.
Decrease in the biosorption capacity was observed at high temperature and the effect was
more pronounced in native as compared to treated and immobilized forms of bagasse.
0
5
10
15
20
25
30
35
40
45
50
30 35 40 45 50 55 60
Sorption cap
aciyy (m
g g‐
1)
Temperature (0C)
Native
SDS‐treated
Immobilized
82
Fig.4.26. Effect of temperature on Zr(IV) biosorption onto bagasse. (Biosorbent
dosage 0.05g/50mL, C0 = 50 mg/L, reaction time 160 min (Native, SDS-treated bagasse),
320 min (immobilized) T = 30 0C and (pH 3.5 native and Immobilized, pH 3 SDS-
treated).
The effect of temperature on biosorption of Sr(II) ions onto native, NaOH-treated and
immobilized peanut husk is shown in Fig.4.27. The effect of temperature on the
biosorption process was small during initial increase 7of temperature and rapid at high
temperature. Decrease in the biosorption capacity was observed at high temperature. It
may also be attributed to deactivation of adsorbent surface at higher temperatures as
described by Aksu and Isoglu (2006).
0
5
10
15
20
25
30
35
40
45
30 35 40 45 50 55 60
Sorption cap
acity (m
g/g)
Temperature (0C)
Native
Treated
Immobilized
83
Fig.4.27. Effect of temperature on biosorption of Sr(II) onto peanut husk.
(Biosorbent dosage 0.05g/25mL, C0 = 10 mg/L, reaction time 80 min(Native, NaOH
treated) and 160 (immobilized peanut husk) T = 30 0C and (pH 9 (native) and 7(
Immobilized and NaOH-treated)
4.12. Thermodynamics studies
Thermodynamic parameters are calculated for the biosoption process of uranium,
zirconium and strontium are presented in Tales 4.13, 14 and 15. Thermodynamic
parameters at various temperatures for native, SDS-treated and immobilized RH are
presented in Table 4.13. The negative value of ΔHo suggests that the process is
exothermic with ΔH values less than 40 kJ mol-1, suggesting the reaction is physical in
nature. The negative values of ΔG for all three forms of RH provide evidence of the
spontaneity of the reaction. The positive values of entropy change ΔS suggest that
randomness increases as the reaction proceeds and biosorption of U(VI) ions onto native,
SDS-treated and immobilized RH is a favorable process.
0
1
2
3
4
5
6
30 35 40 50 60
Sorption cap
acity (m
g/g)
Temperature oC
Native
NaOH‐Treated
Immobilized
84
Table 4.13.
Thermodynamic parameters for U (VI) biosorption onto rice husk as a function of
temperature.
Temperature (Co)
Native SDS-treated
Immobilized
30 35 40 45 50 55 60
ΔG
-37.57 -38.19 -38.81 -39.43 -40.05 -40.66 -41.28
ΔH
-113.3
ΔS
123.6
ΔG
-30.89 -31.39 -31.91 -32.42 -32.92 -33.43 -33.94
ΔH
-86.95
ΔS
101.7
ΔG
-21.94 -22.29 -22.66 -23.02 -23.38 -23.74 -24.10
ΔH
-70.18
ΔS
72.16
* ΔGo= kJ mol-1; ΔHo= kJ mol-1; ΔSo= J mol-1 K-1
Thermodynamic parameters at various temperatures for native, SDS-treated and
immobilized bagasse are presented in Table 4.14. The positive value of ΔHo suggests that
the process is endothermic with ΔH values greater than 40 kJ mol-1, suggesting the
reaction might be chemical in nature. The negative values of ΔG for all three forms of
bagasse provide evidence of the spontaneity of the reaction at low temperature. The
negative values of entropy change ΔS suggest that randomness decreases as the reaction
proceeds. All the thermodynamic studies shows that and biosorption of Zr (IV) ions onto
native, SDS-treated and immobilized bagasse is a favorable process.
85
Table 4.14
Thermodynamic parameters for Zr (IV) biosorption onto bagasse as a function of
temperature.
Temperature (Co)
Native SDS-treated
Immobilized
30 35 40 45 50 55 60
ΔGo
-0.004 -0.003 -0.002 0.001 0.002 0.003 0.005
ΔHo 99.89
ΔSo -0.317
ΔGo -0.0026 -0.00179 -0.00094 -9.99E05 0.00074 0.00159 0.00242
ΔHo 53.61
ΔSo -0.168
ΔGo -0.00114 -0.00036 0.000427 0.00121 0.001993 0.002776 0.003559
ΔHo 48.592
ΔSo -0.157
* ΔGo= kJ mol-1; ΔHo= kJ mol-1; ΔSo= J mol-1 K-1
Thermodynamic parameters at various temperatures for native, NaOH-treated and
immobilized peanut husk are presented in Table 4.15. The positive value of ΔHo suggests
that the process is endothermic with ΔH values lesser than 40 kJ mol-1, suggesting the
reaction is physical in nature. The negative values of ΔG for all three forms of peanut
husk provide evidence of the spontaneity of the reaction. The negative values of entropy
change ΔS suggest that randomness decreases as the reaction proceeds. All the
thermodynamic studies shows that and biosorption of Sr (II) ions onto native, NaOH-
treated and immobilized peanut husk is a favorable process.
86
Table.4.15
Thermodynamic parameters for Sr (II) biosorption onto peanut husk as a function of
temperature.
Temperature (Co)
Native NaOH-treated
Immobilized
30 35 40 50 60
ΔGo
-0.00384 -0.00377 -0.00369 -0.00355 -0.00341
ΔHo 8.1898
ΔSo -0.015
ΔGo -0.01221 -0.01219 -0.01217 -0.01215 -0.01213
ΔHo 13.48
ΔSo -0.0042
ΔGo -0.00336 -0.00327 -0.00319 -0.0031 -0.0030
ΔHo 8.41
ΔSo -0.017
* ΔGo= kJ mol-1; ΔHo= kJ mol-1; ΔSo= J mol-1 K-1
4.13. Effect of interfering ions
Biosorption of selected metal ions (uranium, zirconium and strontium) was studied in the
presence of other cations and anions. Industrial wastewater contains many other
background electrolytes which may interfere with the biosorption process so the
biosorption process must study in the presence of these competing ions. Solutions of
competing ions were prepared and the influence on the biosorption capacity of RH
biosorbents was studied. The effect of ionic interaction on the sorption process may be
represented by the ratio of sorption capacity in the presence of interfering ion (qmix) and
without interfering ion (qo), such that for:
>1 sorption is promoted in presence of other interfering ions
=1 sorption is not influenced in presence of other interfering ions
< 1 sorption is suppressed in presence of other interfering ions [Pereira, et al., 2010]
The effect of cations and anions on the biosorption capacity of RH is reported in Table
4.16. Among the cations studied, no significant effect on adsorption capacity of native
and SDS-treated RH was observed at low concentration (50 ppm) but at higher
concentrations, these competing cations showed an inhibiting effect. In the case of the
anions selected, nitrate caused the maximum interference on native and SDS-treated RH
forms while sulphate and phosphate also had suppressing effect. The immobilized RH
appeared not to be strongly influenced by the presence of these anions. Chloride did not
seem to compete with the U(VI) ions for adsorption sites on native and SDS-treated RH
but greatly suppressed adsorption on the immobilized RH.
87
Table 4.16.
Comparison of the effect of different interfering cations and anions on U(VI) ions (50 mg
L-1) biosorption onto rice husk.
Cations
Native SDS-Treated
Immobilized
50 ppm
75 ppm
100 ppm
50 ppm
75 ppm
100 ppm
50 ppm
75 ppm
100 ppm
Ni+2 Pb+2 Co+2 Mn+2 Cd+2 Cu+2
Zn+2
0.98 0.97 0.97 0.96 0.96 0.96 0.96
0.62 0.84 0.32 0.89 0.64 0.69 0.97
0.66 0.66 0.13 0.79 0.61 0.58 0.27
0.85 0.84 0.84 0.85 0.75 0.83 0.79
0.62 0.68 0.64 0.75 0.62 0.75 0.75
0.66 0.43 0.43 0.52 0.43 0.53 0.49
0.72 0.88 1.11 1.54 0.82 0.39 0.98
0.28 0.24 0.50 0.78 0.63 0.24 0.66
0.06 0.02 0.27 0.77 0.53 0.17 0.34
Anions
Native
0.1M
SDS-Treated
0.1M
Immobilized
0.1M
NO3
-1 Cl-1
SO42-
PO43-
0.68 0.91 0.79 0.88
0.89 0.94 1.00 0.94
0.99 0.04 1.02 0.94
The effect of cations and anions on the biosorption capacity of bagasse is reported in
Table 4.17. Among the cations studied, significant effect on adsorption capacity of native
and SDS-treated and immobilized bagasse was observed at even low concentration (50
ppm) but at higher concentrations, these competing cations showed an inhibiting effect.
However, the selected anions had less suppressing effects than cations.
88
Table 4.17. Comparison of the effect of different interfering cations and anions on Zr(VI) ions (50 mg L-1) biosorption onto bagasse.
Cations
Native SDS-Treated
Immobilized
Ni+2
Pb+2
Co+2
Mn+2
Cd+2
Cu+2
Zn+2
50pp
m
75ppm 100 ppm 50 ppm 75ppm 100 ppm 50 ppm 75ppm 100 ppm
0.123 0.422 0.213 0.653 0.376 0.401 0.977
0.067 0.254 0.127 0.427 0.2543 0.264 0.425
0.069 0.143 0.085 0.294 0.173 0.153 0.296
0.165 0.211 0.155 0.215 0.143 0.119 0.525
0.065 0.089 0.118 0.149 0.069 0.109 0.376
0.628 0.123 0.628 0.670 0.579 0.548 0.255
0.152 0.152 0.213 0.366 0.306 0.216 0.609
0.079
0.109 0.152 0.268 0.134 0.104 0.216
0.033
0.057 0.062 0.062 0.069 0.051 0.108
Anions
(0.1M)
Native SDS-Treated
Immobilized
NO3
-1 Cl-1 SO4
2- PO4
3-
0.959 0.832 0.834 0.816
0.692 0.734 0.788 0.863
0.891 0.945 0.868 0.848
The effect of cations and anions on the biosorption capacity of peanut husk is reported in
Table 4.18. Among the cations studied, significant effect on adsorption capacity of native
and NaOH-treated was observed at low concentration (5 ppm) but at higher
concentrations, these competing cations showed an inhibiting effect. In the case of the
anions selected, maximum interference on native and NaOH-treated peanut husk. The
immobilized peanut husk appeared not to be strongly influenced by the presence of these
anions.
89
Table 4.18. Comparison of the effect of different interfering cations and anions on Sr(II) ions (10 mg L-1) biosorption onto peanut husk.
Cations
Native NaOH-Treated
Immobilized
Co2+ Cu2+ Ni2+ Cd2+ Zn2+ Mn2+ Pb2+
5ppm 10ppm 15 ppm 5ppm 10ppm 15ppm 5ppm 10 ppm 15
ppm
0.016 0.016 0.016 0.02 0.016 0.016 0.016
0.016 0.016 0.016 0.01 0.016 0.016 0.016
0.016 0.016 0.016 0.01 0.016 0.016 0.016
0.018 0.035 0.09 0.05 0.03 0.08 0.05
0.018 0.006 0.04 0.003 0.008 0.007 0.05
0.018 0.006 0.04 0.003 0.02 0.05 0.05
0.85 0.84 0.85 0.88 0.87 0.86 0.87
0.68 0.72 0.85 0.75 0.75 0.73 0.72
0.53 0.56 0.45 0.61 0.60 0.75 0.56
Anions
(0.1M)
Native NaOH-Treated
Immobilized
Cl-1 CH3COO-1 SO4
3 I-1 PO4
3-
0.199 0.172 0.051 0.064 0.126
0.372 0.362 0.209 0.304 1.032
0.704 0.7067 0.552 0.866 0.727
4.14. Desorption studies
Desorption of the adsorbed U(VI) ions as a function of fixed U(VI) concentration by
different desorbing agents was studied in a batch system. Desorption efficiency of the
selected chemicals was found to be at a maximum with H2SO4 for native and SDS-treated
RH (86 %) and with EDTA for immobilized RH (92%). The selected desorbing agents
efficiency decrease in following order for native, SDS-treated and immobilized RH
respectively
H2SO4 > HCl > EDTA >NaOH >MgSO4 (native)
H2SO4 > HCl > EDTA >NaOH >MgSO4 (SDS-treated)
EDTA > HCl > H2SO4>NaOH >MgSO4 (immobilized)
90
A desorption experiment to study the effect of changing concentrations of H2SO4 was
conducted for native and SDS-treated RH. The results indicate that the elution capacity
of native and SDS-treated RH by H2SO4 increased from 79 to 92% and 87 to 94%
respectively when the H2SO4 concentration was increased from 0.1M to 0.5M. The
elution capacity of the immobilized RH was increased from 92% to 98% by increasing
the EDTA concentration from 0.1M to 0.5M. Previous studies on U(VI) biosorption by
different adsorbents also reported good desorption efficiency of the EDTA and acids
(Gonzilez-Muiuoz, et al., 1997; Saleem and Bhatti, 2011; Zhang et al., 2013)
Fig.4.28.Comparison of different desorbing agents on U(VI) biosorption onto rice
husk . (Concentration of each desorbing agent =0.1M).
The selected desorbing agents efficiency decrease in following order for native, SDS-
treated and immobilized bagasse respectively for desorption of Zr(IV) ions. .
H2SO4> HCl> EDTA> MgSO4> NaOH (Native)
H2SO4> HCl> MgSO4> NaOH> EDTA (SDS-treated)
H2SO4> HCl> EDTA> MgSO4> NaOH (immobilized bagasse)
The desorption efficiency of the H2SO4 is on accordance with the previous reports by
(Bhatti and 2013 and Hanif et al.; 2013) for Zr(IV) biosorption.
0
10
20
30
40
50
60
70
80
90
100
NaOH EDTA MgSO4 HCl H2SO4
% Desorption
Eluating agents
Native
Treated
Immobilized
91
Fig.4.29. Comparison of different desorbing agents on Zr(IV) biosorption onto
bagasse. (Concentration of each desorbing agent =0.1 M.) The selected desorbing agents efficiency decrease in following order for native, NaOH-
treated and immobilized peanut husk respectively for desorption of sorbet Sr(II) ions.
HCl>EDTA>>H2SO4>NaOH (Native)
EDTA>HCl>H2SO4>NaOH (NaOHTreated)
HCl>H2SO4>EDTA>NaOH (Immobilized)
Fig.4.30. Comparison of different desorbing agents on Sr(II) biosorption onto
peanut husk. (Concentration of each desorbing agent=0.1 M).
0
10
20
30
40
50
60
70
80
90
100
NaOH EDTA MgSO4 HCl H2SO4
% Desorption
Desorbing agents
Native
Treated
Immobilized
0
10
20
30
40
50
60
70
80
90
100
0.1 M HCl 0.1 M NaOH 0.1 M EDTA 0.1 M H2SO4
% Desorption
Desorbing agents
Native
NaOH‐Treated
Immobilized
92
4.15. Response surface methodology
The experimental design matrix derived from central composite design for three coded
independent variables (pH, sorbent dose and initial metal ion concentration) along with
observed responses i.e. sorption capacity q(mg/g), for uranium, zirconium and strontium.
The experimental results were evaluated and polynomial equations fitted to sorption data
after base log10 transformation for uranium, zirconium and strontium respectively in
terms of final equation coded factors are given as follows
Log10(R1)= +1.27 +0.093* A -0.29* B +0.38* C -0.078* A * B +0.11* A * C +0.17* B *
C -0.26* A2 -0.10 * B2 -0.37 * C2
Log10(R1) = +1.83 +0.15* A -0.14 * B +0.35 * C +0.085 * A * B +0.028 *A * C-0.064
* B * C -0.48* A2 +0.023*B2-0.23*C2
Log10(R1) =+0.75 +0.11 * A-0.31* B +0.068 * C -0.054 * A * B +0.15 * A * C-0.024 * B
* C -0.15 * A2-0.037 * B2-0.035 * C2
Here A, B and C are three independent variables representing pH, sorbent dose and initial
metal ion concentration respectively and R1 representing the response i.e. sorption
capacity calculated for each designed experiment by using Design Expert software. The
experiments were done at 30οC and saking at 125 rpm for already optimized equilibrium
time in batch biosorption. The obtained results are transformed to base log 10 as
suggested by Box Cox plot of the design expert for better fitness of model and more
reliable results.
4.15.1. Fitness of model
Checking the adequacy of model is an important step of data analysis; otherwise the
model may give poor or misleading results (Sharma et al., 2009). The plot of normal %
probability versus studentized residuals shown in Figs. 4.31-4.33 for uranium, zirconium
and strontium respectively, indicate that the model satisfies the assumptions of the
analysis of variance (ANOVA) where the studentized residuals measure the number of
standard deviations separating the actual and predicted values.
The statistical significance of the fitted quadratic model was determined by the analysis
of variance (ANOVA), F and p values. ANOVA is a statistical technique that subdivides
the total variation in a set of data into component parts associated with specific sources of
variation for the purpose of testing hypotheses on the parameter of the model (Kim et al.,
2003). Results of Analysis of variance (ANOVA) for biosorption of U, Zr, Sr are reported
in Tables 4.19-21 respectively. According to the ANOVA, F values are very large
93
showing that model terms are significant and most of the variation in the response can be
explained by the regression equation. Values of “p value less than 0.0500 also indicate
high significant regression at 95% confidence level. (Sharma et al., 2009; Kim et al.,
2003).
The fitness of the model was checked by the coefficient of determination (R2). The values
of R2 and adjusted R2 are very high and show a high correlation between the observed and
predicted values. This reveals that the regression model explains the relationship between
the independent variables and the response q(mg/g) very well. The coefficient of
determination for were R2 0. 897, Adj R2 0.8076 for U; R2 0.9788, Adj R2 0.9597 for Zr
and R2 0.9302 Adj R2 0.8675 for Sr indicating high significance of the model. These
results indicate that the regression model explains the relationship between the
independent variables and the response very well.
Table 4.19.
Analysis of variance (ANOVA) for response surface quadratic model for U(VI) sorption
onto native rice husk.
Source Sum of squares
df Mean squares
F value P value Prob>F
Model 4.79 9 0.53 9.86 0.0007
A-pH 0.087 1 0.087 1.61 0.2334
B-sorbent dose 0.84 1 0.84 15.53 0.0028
C-Concentration 1.47 1 1.47 27.15 0.0004
AB 0.049 1 0.049 0.90 0.3648
AC 0.10 1 0.10 1.91 0.1967
BC 0.23 1 0.23 4.35 0.0636
A2 3.32 1 3.32 3.32 0.0983
B2 0.030 1 1.377E-003 0.55 0.4763
C2 0.37 1 0.37 6.92 0.0251
94
The Model F-value of 9.86 implies the model is significant. There is only a 0.07%
chance that a "Model F-Value" this large could occur due to noise. Values of "Prob > F"
less than 0.0500 indicate model terms are significant. In this case B, C, C2 are significant
model terms.
Table 4.20.
Analysis of variance (ANOVA) for response surface quadratic model for Zr(IV) sorption onto native bagasse.
Source Sum of Squares df Mean
Square F Value Prob p-
value > F
Model 3.748 9 0.416 51.23976 0.0001
A-PH 0.238 1 0.238 29.33225 0.0003
B-Dose 0.197 1 0.197 24.22724 0.0006
C-Concentration 1.252 1 1.252 154.0122 0.0001
AB 0.058 1 0.058 7.099345 0.0237
AC 0.006 1 0.006 0.786705 0.3959
BC 0.033 1 0.033 4.062231 0.0715
A2 0.628 1 0.628 77.29767 0.0001
B2 0.001 1 0.001458 0.179441 0.6808
C2 0.144 1 0.143654 17.67491
0.0018
The Model F-value of 51.24 implies the model is significant. There is only a 0.01%
chance that a "Model F-Value" this large could occur due to noise. Values of "Prob > F"
less than 0.0500 indicate model terms are significant. In this case A, B, C, AB, A2, C2 are
significant model terms.
95
Table. 4.21.
Analysis of variance (ANOVA) for response surface quadratic model for Sr(II) sorption onto NaOH-treated peanut husk.
Source Sum of squares df Mean squares
F value P value Prob>F
Model
1.54 9 0.17 14.82 0.0001
A-pH
0.11 1 0.11 9.81 0.0107
B-sorbent dose
0.98 1 0.98 84.95 0.0001
C-Concentration
0.046 1 0.046 3.97 0.0743
AB
0.023 1 0.023 2.01 0.1872
AC
0.17 1 0.17 14.83 0.0032
BC
4.658E-003 1 4.658E-003 0.40 0.5402
A2
0.064 1 0.064 5.49 0.0411
B2
3.664E-003 1 3.664E-003 0.32 0.5862
C2
3.360E-003 1 3.360E-003 0.29 0.6019
The Model F-value of 14.82 implies the model is significant. There is only a 0.01%
chance that a "Model F-Value" this large could occur due to noise. Values of "Prob > F"
less than 0.0500 indicate model terms are significant. In this case A, B, AC, A2 are
significant model terms.
96
Fig. 4.31. (a) The plot of predicted sorption capacity q (mg/g) versus actual for U(VI)
sorption onto native rice husk. The studentized residual and normal % probability
plot for U(VI) sorption onto native rice husk.
Design-Expert® SoftwareLog10(Sorption capacity)
Color points by value ofLog10(Sorption capacity):
1.66894
-0.200659
Actual
Pre
dic
ted
Predicted vs. Actual
-0.40
0.13
0.65
1.18
1.70
-0.40 0.12 0.64 1.15 1.67
Design-Expert® SoftwareLog10(Sorption capacity)
Color points by value ofLog10(Sorption capacity):
1.66894
-0.200659
Internally Studentized Residuals
No
rma
l % P
rob
ab
ility
Normal Plot of Residuals
-2.63 -1.40 -0.16 1.07 2.31
1
5
10
20
30
50
70
80
90
95
99
(a)
(b)
97
Fig. 4.32. (a) The plot of predicted sorption capacity q (mg/g) versus actual for
Zr(IV) sorption onto native bagasse. The studentized residual and normal %
probability plot for Zr(IV) sorption onto native bagasse.
Design-Expert® SoftwareLog10(Sorption cpacity)
Color points by value ofLog10(Sorption cpacity):
2.01452
0.521138
Internally Studentized Residuals
Norm
al %
Pro
babili
ty
Normal Plot of Residuals
-2.86 -1.54 -0.21 1.12 2.44
1
5
10
20
30
50
70
80
90
95
99
Design-Expert® SoftwareLog10(Sorption cpacity)
Color points by value ofLog10(Sorption cpacity):
2.01452
0.521138
Actual
Pre
dic
ted
Predicted vs. Actual
0.50
0.90
1.30
1.70
2.10
0.51 0.88 1.26 1.64 2.01
(a)
(b)
98
Fig.4.33. (a) The plot of predicted sorption capacity q (mg/g) versus actual for Sr(II) sorption onto NaOH-treated peanut husk. The studentized residual and normal %
probability plot of removal Sr(II) onto NaOH-treated peanut husk.
Design-Expert® SoftwareLog10(sorption capacity)
Color points by value ofLog10(sorption capacity):
1.30103
0.0293838
Internally Studentized ResidualsN
orm
al %
Pro
ba
bili
ty
Normal Plot of Residuals
-2.38 -1.16 0.07 1.29 2.52
1
5
10
20
30
50
70
80
90
95
99
Design-Expert® SoftwareLog10(sorption capacity)
Color points by value ofLog10(sorption capacity):
1.30103
0.0293838
22
Actual
Pre
dic
ted
Predicted vs. Actual
0.00
0.35
0.70
1.05
1.40
0.03 0.35 0.67 0.98 1.30
(a)
(b)
99
The surface responses of the quadratic model, with one variable maintained at a midrange
value and the other two varying within the given experimental ranges, are shown in Fig.
4.34-4.36 for U(VI), Zr(IV) and Sr(II) onto native rice husk, bagasse and NaOH- treated
peanut husk respectively.
Experiments were carried out as per selected model within range of pH and sorbent dose
to investigate the combined effect of pH and sorbent dose on the sorption capacity of rice
husk for U removal. RSM technology was used and results are shown in the form of
contours plots. Fig. 4.34 (a) and shows that if sorbent dose is increased from 0.05 to 0.30
g/50 mL keeping U (VI) concentration (55 mg/L) constant, the maximum response was
found at low biosorbent dose 0.05 and pH range 5-6. Similarly designed experiments in
the selected range of U(IV) concentration and pH were done to see the combined effect
on the sorption capacity. Results are shown in the form of contours plots shown in Fig.
4.34 (b) and showing that in the range of 10 to 100 mg/L U ion concentration and pH (2-
9) keeping sorbent dose constant (0.17 g). Higher concentration and pH 5-6 are giving
maximum response.
The results of sorbent dose and concentration change keeping pH (5.5) constant are given
in Fig. 4.34 (c) and its clear that low biosorbent dose and higher concentration gives
maximum response. The trends in maximum response are very similar to classical batch
sorption study. The results are also supporting the previous one factor batch experiments.
100
Fig.4.34. Contour plot showing effect of pH, sorbent dose and initial U(VI)
concentration on U(VI) sorption onto rice husk. (a) Effect of pH and biosorbent dose
for U(VI) sorption onto native rice husk (b)Effect of pH and initial metal ion
concentration for U(VI) sorption onto native rice husk (c) Effect of sorbent dose and
initial metal ion concentration for U(VI) sorption onto native rice husk.
Design-Expert® SoftwareTransformed ScaleLog10(Sorption capacity)
Design Points1.66894
-0.200659
X1 = A: pHX2 = B: sorbent dose
Actual FactorC: Concentration = 55.00
2.00 3.75 5.50 7.25 9.00
0.05
0.11
0.17
0.24
0.30Log10(Sorption capacity)
A: pH
B: sorbent d
ose
0.7546330.7546330.901015
1.0474
1.19378
1.34016
666666
Design-Expert® SoftwareTransformed ScaleLog10(Sorption capacity)
Design Points1.66894
-0.200659
X1 = A: pHX2 = C: Concentration
Actual FactorB: sorbent dose = 0.17
2.00 3.75 5.50 7.25 9.00
10.00
32.50
55.00
77.50
100.00Log10(Sorption capacity)
A: pH
C: C
once
ntratio
n
0.436425 0.4364250.628111
0.819798
1.01149
1.20317666666
Design-Expert® SoftwareTransformed ScaleLog10(Sorption capacity)
Design Points1.66894
-0.200659
X1 = B: sorbent doseX2 = C: Concentration
Actual FactorA: pH = 5.50
0.05 0.11 0.17 0.24 0.30
10.00
32.50
55.00
77.50
100.00Log10(Sorption capacity)
B: sorbent dose
C: C
once
ntratio
n
0.211383
0.466746
0.72211
0.977473
1.23284666666
(a)
(b)
(c)
101
Fig.4.35 (a, b, c) represents the effect of changing pH, sorbent dose and zirconium ion
concentration on the adsorption capacity of bagasse for zirconium removal under the
predefined experimental conditions planned by face-cantered response central composite
design by surface methodology. The contour graph Fig.4.35 (a) showing that pH 2-3 and
low sorbent dose 0.05 g/50mL, giving maximum response while keeping initial zirconium
ion concentration (112.5 mg/L) constant. Fig.4.35 (b) showing maximum response in the
pH range (2-3) and at higher zirconium ion concentration keeping sorbent dose (0.17 g/50
mL) constant. The results of experiments at constant pH (2.50) to see the combined effect
of zirconium ion concentration and sorbent dose suggest that higher zirconium
concentration and small biosorbent dose favour the response (Fig.4.35(c)).
Combined effects of pH and Sr(II) ion concentration and sorbent dose has been analyzed
from the face centred central composite design (Fig.4.36 a,b,c). It has been estimated
from Fig. 4.36 (a) that at constant Sr(II) ion concentration (45 mg/L) the response is
maximum at pH 7-9 and at very low sorbent amount (0.08 g/25mL). The Fig.4.36 (b)
represents that maximum response was obtained about pH 7 and Sr(II) ion concentration
of 40-50 mg/L at constant sorbent amount of 0.19 g/25 mL. The results of experiments
conducted to see the combined effect of sorbent dose and concentration of Sr(II) ions at
constant pH 6.0 given in Fig.4.36(c) showing that results obtained at concentration above
45 mg/L and low biosorbent dose giving best maximum response.
102
Fig.4.35. Contour plot showing effect of pH, sorbent dose and initial Zr(IV)
concentration on Zr(IV) sorption onto bagasse (a) Effect of interaction of pH and
biosorbent dose on Zr(IV) sorption onto native bagasse. (b)Effect of pH and initial metal
ion concentration on Zr(IV) sorption onto native bagasse. (c) Effect of sorbent dose and
initial metal ion concentration for Zr(IV) sorption onto native bagasse.
Design-Expert® SoftwareTransformed ScaleLog10(Sorption cpacity)
Design Points2.01452
0.521138
X1 = A: pHX2 = B: Sorbent dose
Actual FactorC: Concentration = 112.50
1.00 1.75 2.50 3.25 4.00
0.05
0.11
0.17
0.24
0.30Log10(Sorption cpacity)
A: pH
B: S
orb
ent d
ose
1.1648
1.33151
1.498211.66491 1.66491
1.83162
666666
Design-Expert® SoftwareTransformed ScaleLog10(Sorption cpacity)
Design Points2.01452
0.521138
X1 = A: pHX2 = C: Concentration
Actual FactorB: Sorbent dose = 0.17
1.00 1.75 2.50 3.25 4.00
25.00
68.75
112.50
156.25
200.00Log10(Sorption cpacity)
A: pH
C: C
once
ntratio
n
0.869485
1.09274
1.09274
1.31599
1.53924
1.76249
666666
Design-Expert® SoftwareTransformed ScaleLog10(Sorption cpacity)
Design Points2.01452
0.521138
X1 = B: Sorbent doseX2 = C: Concentration
Actual FactorA: pH = 2.50
0.05 0.11 0.17 0.24 0.30
25.00
68.75
112.50
156.25
200.00Log10(Sorption cpacity)
B: Sorbent dose
C: C
once
ntratio
n
1.36226
1.52724
1.69221
1.85719
2.02217
666666
(a)
(b)
(c)
103
Fig.4.36. Contour plot showing effect of pH, sorbent dose and initial Sr(II)
concentration on Sr(II) sorption onto peanut husk. (a) Effect of pH and biosorbent
dose for Sr(II) sorption onto NaOH -treated peanut husk(b) Effect of pH and Sr(II) ion
concentration for Sr(II) sorption onto NaOH- treated peanut hudk(c) Effect of of sorbent
dose and Sr(II) ion concentration for Sr(II) sorption onto NaOH-treated peanut husk.
Design-Expert® SoftwareTransformed ScaleLog10(sorption capacity)
Design Points1.30103
0.0293838
X1 = A: pHX2 = B: Sorbent dose
Actual FactorC: Concentration = 45.00
3.00 4.50 6.00 7.50 9.00
0.08
0.14
0.19
0.25
0.30Log10(sorption capacity)
A: pH
B: S
orb
ent d
ose
0.3373050.483036
0.628768
0.774499
0.920231
666666
Design-Expert® SoftwareTransformed ScaleLog10(sorption capacity)
Design Points1.30103
0.0293838
X1 = A: pHX2 = C: Concentration
Actual FactorB: Sorbent dose = 0.19
3.00 4.50 6.00 7.50 9.00
20.00
32.50
45.00
57.50
70.00Log10(sorption capacity)
A: pH
C: C
once
ntratio
n
0.459273
0.544349
0.544349
0.629426
0.629426
0.714503
0.79958
666666
Design-Expert® SoftwareTransformed ScaleLog10(sorption capacity)
Design Points1.30103
0.0293838
X1 = B: Sorbent doseX2 = C: Concentration
Actual FactorA: pH = 6.00
0.08 0.14 0.19 0.25 0.30
20.00
32.50
45.00
57.50
70.00Log10(sorption capacity)
B: Sorbent dose
C: C
once
ntratio
n
0.4449110.5720730.6992350.826398
0.95356
666666
104
4.16. Biosorbent characterization
4.16.1. Surface studies
Physiochemical properties of the native rice husk, bagasse and peanut husk are given in
Table 4.22. The results obtained highlight the predominance of meso-pores (IUPAC
Classification 20Ǻ < d < 500 Ǻ) in biosorbents which is desirable for the adsorption of
metal ions from the aqueous phase.
Table 4.22.
Brunauer-Emmett-Teller (BET) surface area analysis and Barrett-Joyner-Halenda (BJH)
pore size and volume analysis.
Biosorbents
Rice husk Bagasse Peanut husk
Method BJH adsorption
Multi-point BET
BJH adsorption
Multi-point BET
BJH adsorption
Multi-point BET
Average particle size (μm)
300 300 300 300 300 300
Pore Volume
(cc g-1)
0.32 - 0.18 - 0.11
Pore diameter (A0)
129.14 - 108.69 - 80.13
Surface area
(m2/g)
58.48 50.76 108.89
4.16.2. Elemental analysis
The results of C, H and N percentage obtained present in native rice husk, bagasse and
peanut husk are given in Table 4.23.
Table: 4.23.
Elemental (C, H and N) analysis of native rice husk, bagasse and peanut husk
Biosorbents % C %H %N
Rice husk 33.72 – 33.96
4.41 – 4.45
0.52 – 0.45
Bagasse 42.34 – 42.13
5.61 – 5.51
0.68 – 0.69
Peanut husk 45.03 - 45.18
5.60 – 5.47
1.08- 1.17
105
4.16.3. Thermogravimetric analysis
Thermogravimetric analysis (TGA) is the most common technique for investigating the
volatilization behaviour of biomass. TGA involves heating a sample mass at specific
heating rate and measuring the change in mass as a function of temperature and time. A
number of researchers have used this method to investigate the thermo-chemical
characteristics of biomasses. In TGA, the lignocellulosic structure of biosorbents can be
qualitatively identified from the change in weight of a sample which is recorded as a
function of time or temperature. As illustrated in Fig.4.37, 38 and 39 for native rice husk,
bagasse and peanut husk respectively. The first stage (below 200C) corresponded to the
drying period where light volatiles, mainly water were liberated causing minimal
reduction in sample weight, The second stage of decomposition, occurring between 200
and 500C, corresponds to a significant percentage weight loss of sample due to liberation
of volatile hydrocarbons from rapid thermal decomposition of hemicelluloses, cellulose
and some parts of lignin. During stage 3, a continuous weight loss was observed until the
highest temperature was reached (1000C), primarily due to the steady decomposition of
the remaining heavy components mainly from lignin (Ghorbani and Eisazadeh, 2012 ).
Fig . 4.31. TGA of rice husk Native
Fig. 4.37. TGA of rice husk
0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
1 0 0
0 1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 1 0 0 0
Te m p e r at u r e (o C )
weigh
t loss (%)
S t a g e 3
37 7 ‐ 1 0 00 C = 1 3% l o s s
S t a g e 1
< 20 0 C = 7% l o s s
S t a g e 2
2 0 0 ‐ 3 7 7 C = 5 0 % l o s s
106
Four distinct parts can be observed in Fig.4.38 for the thermal decomposition of
lignocellulosic bagasse materials, including moisture content removal at the beginning of
the curves (approximately upto 200°C). Hemicellulose (Approximately 300°C) and
cellulose decomposition (between 300 and 450oC) are the dominant events for the rate of
mass loss. After their decomposition, the lignin content (450-600°C) is more difficult to
decompose due to its structural chemical complexity (Kassia et al., 2010).
Fig. 4.38. TGA of bagasse
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600
% weight loss
Temperature (oC)
Stage 1200 °C> 5 %
Stage 2200‐380°C> 65 %
Stage 3380‐600 °C> 35 %
107
TGA was used to determine the moisture content and stability of peanut husk as shown in
Fig. 4.39. Three stages of decomposition occur: The first stage in the temperature range
<200 °C weight loss due to the moisture released from the sample during heating. TG
analysis of PH revealed that the major decomposition occurred in the second stage 200–
400 °C, which may be due to the decomposition of cellulose, hemicelluloses, and lignin
to carbon. Further heating above 500°C resulted very low weight loss due to the
formation of volatile products like CO, CO2, etc (Reddy et al., 2013)
Fig. 4.39. TGA of peanut husk
0
10
20
30
40
50
60
70
80
90
100
0 100 200 300 400 500 600
% weight loss
Temperature (0C)
Stage 1200°C >7 %
Stage 2200‐400°C> 71 %
Stage 3
400‐600°C> 15 %
108
4.16.4. X-Ray diffraction (XRD) studies
X-Ray diffraction (XRD) studies pattern of the native rice husk, bagasse and peanut husk
are shown below in Fig. 4.40, 41 and 42. The sharp peaks support crystalline nature of the
rice husk, bagasse and peanut husk. XRD studies shows that silica is the main component
of the all three biosorbents.
Fig. 4.40. XRD pattern of rice husk
109
Fig.4.41. XRD pattern of bagasse.
Fig.4.42. XRD pattern of peanut husk
110
4.16.5. Scanning electron microscope and Energy dispersive X-Rays
Scanning electron microscope and Energy dispersive X-Rays (SEM-EDX) images were
used to study surface morphology of native RH, before and after loading with U(VI) ions
are illustrated in Fig. 4.43. The images show morphological changes after uranium
sorption. The EDX spectra show the adsorbed ions on loaded rice husk.
The adsorption of Zr (IV) and Sr(II) was seen on EDX spectra of loaded bagasse and
peanut husk respectively as shown in Fig. 4.43. 4.45.
Fig.4.43. SEM-EDX spectra of rice husk. (a,b) SEM images of unloaded native rice husk at low and high resolution (c,d) SEM images of loaded rice husk at low and high
resolution (e) EDX spectra of U(VI) loaded rice husk.
(a) (b)
(c) (d)
(e)
111
Fig.4.44. SEM-EDX spectra of bagasse. (a,b) SEM images of unloaded native bagasse at low and high resolution (c,d) SEM images of loaded bagasse at low and high resolution
(e) EDX spectra of Zr(IV) loaded bagasse.
(a) (b)
(c) (d)
112
Fig.4.45. SEM-EDX spectra of peanut husk (a) SEM spectra of loaded peanut husk (b)
EDX spectra of loaded peanut husk.
(a)
(b)
113
4.16.6. FT-IR Studies
The presence of active functional groups responsible for sorption of selected metal ions
(U(VI), Zr(IV) and Sr(II) sorption onto RH, bagasse and peanut husk is confirmed by
FTIR (see Fig.4.46- 4.48).
The organic part of the RH is composed of cellulose, hemicelluloses and lignin, which
contain mostly alkenes, esters, aromatics, ketones and aldehydes. The presence of OH
groups on the RH is confirmed by presence of a band between 3000 and 3750 cm-1. OH
groups bound to methyl radicals, which are common in lignin, show a signal between
2940-2820 cm-1. The peak at 1053 cm-1 represents the Si-O-Si linkage as part of the
inorganic portion of the RH. Comparative analysis of vibrational frequencies of the
functional groups of biosorbents (native RH, SDS -treated and immobilized RH shown in
Table. 4.24) shows the involvement of cellulose, lignin and silica functional moieties in
adsorption.
Table 4.24.
Functional groups in rice husk by FT-IR.
Native rice husk U Loaded rice husk (cm-1)
SDS-treated loaded rice husk(cm-1)
Immobilized loaded rice husk(cm-1)
3826.77 cm-1 (Si-OH) 3294.42 (O-H stretching of hydroxyl cellulose) 2924.09 (OH bound to –CH3 of lignin) 1529.55 (Aromatic C=C stretching Lignin/phenolic backbone) 1053.13 (Si-O-Si)
Absent
3290.56
2918.30
1531.4
1037.70
3761.19
3442.94
2929.87
1585.49
1072.42
3761.19
Absent
2924.09
1587.42
1064.71
114
Fig.4.46. FT-IR spectra of rice husk. (a) Rice husk native (b) Rice husk loded native(c) SDS treated loaded Rice husk (d) Immobilized loaded rice husk.
115
The FT-IR spectra of native, SDS-treated and immobilized bagasse were plotted to
determine the vibration frequency changes in the functional groups. Table 4.25 presents
the fundamental peaks of all possible functional groups in native, modified and loaded
bagasse with zirconium ions. Spectra in Fig.4.47 showing a number of absorption peaks,
indicating the complex nature of the biosorbents. According to the literature, the
absorption band wave number of the ester and carboxyl acid groups in the organic
compounds is approximately 1740 cm−1. Therefore, it can be concluded that the
absorption band at 1730 cm−1 is attributed to the absorption of ester and carboxyl acid
groups (Martin-Lara et al., 2010). The strong C–O band at about 1051.2 cm−1 due to –
OCH3 group, also confirms the presence of lignin structure in sugarcane baggase and can
be assigned to 1055.06 cm−1(Garg et al., 2008). The intense and broad bands at around
3350 cm−1 and 2900 cm−1 are assigned to OH group and C–H stretching respectively
(Zhang et al., 3013).
Table 4.25. Functional groups in bagasse by FT-IR spectra.
Possible function groups
Native bagasse(cm-1)
SDS-treated bagasse (cm-1)
Immobilized bagasse (cm-1)
Zr loaded bagasse(cm-1)
N-H stretching(amides, amines)
3534.0783
O-H functional groups (carboxylic acids, phenols and alcohols)
3329.14 3371.6628 3300.0214 3214.20
C-H(stretching of alkanes)
2910.58 2918.1811 - -
C=O (Carboxylic acid)
1730.15 - - -
C-H stretching (aldehydes)
1595.13 1513.6058 1595.2032 1552.70
C-N (amines) 1055.06 1035.3994 1027.7229 1051.20
C-H stretching 834.68, 910.12 833.2177 819.0531 862.1
116
Fig.4.47. FTIR spectra of bagasse. (a) Native bagasse (b) Zr(IV) loaded native bagasse
(c) Immobilized bagasse(d) SDS treated bagasse.
117
The FTIR spectra and functional groups in peanut husk e native, modified and loaded
with Sr (II) are described in Table 4.26 and Fig. 4.48. The O-H vibrations found at about
3344 cm−1 lignocellulosic and cellulosic material. The intense band at about 2930 cm−1
for peanut husk was attributed to the C H stretching vibration. The intense band at about
2930 cm−1 for peanut husk attributed to the C-H stretching vibration. The bands located at
1570 cm−1 and 1215 cm−1 in the spectra of the peanut husk also demonstrates that peanut
husk contained large amounts of lignocellulosic material, as lignin was a complex (Zhong
et al., 2012; Noreen et al., 2013).
Table 4.26. Functional groups in peanut husk by FTIR spectra.
Possible function groups
Native peanut husk(cm-1)
NaOH-treated Peanut husk (cm-1)
Immobilized peanut husk (cm-1)
Sr loaded peanut husk(cm-1)
O-H functional groups (carboxylic acids, phenols and alcohols)
3344.14 3373.5527 3300.1383 3352.2960
C-H(stretching of alkanes)
2969.41 2919.9152 - 2909.0043
C=O (carboxylic acis)
1726.35 - - -
N=O (R-NO2)
1501.85 1506.6696 1592.8847 1508.5205
C-O(Alcoho, ether, ester carboxylic acid)
1217.12, 1038.09,
1030.3921
1030.3921 1026.3785 1027.8398
C=C stretching
628.81
621.0019
609.5066
653.7539
118
Fig. 4.48. FT-IR spectra of peanut husk (a) Peanut husk native(b) Peanut husk Sr(II) loaded (c) Immobilized peanut husk (d) NaOH treated peanuthusk.
119
4.17. Column biosorption
The results of biosorption of uranium and zirconium onto native rice husk and bagasse
biomasses in a continuous system have been presented in the form of breakthrough curves
which showed the loading behavior of these metal ions to be adsorbed from the solution
expressed in terms of normalized concentration defined as the ratio of the outlet metal ion
concentration to the inlet metal ion concentration as a function of time (Ct/Co vs. t).
4.17.1. Effect of bed height
Fixed bed column uranium biosorption studies were conducted using column filled with
rice husk for three different bed heights of 1, 2, and 3 cm at a constant flow rate of 1.8
mL/min and inlet concentration of 50 mg/L uranium solution having pH 4. Similarly,
zirconium biosorption studies were conducted in column using bagasse at different
heights and 1, 2 and 3 cm at constant flow rate of 1.8 mL/min and inlet concentration of
50 mg/L zirconium solution having pH 3.5. The breakthrough results are summarized in
Fig.4.50.
In all curves, when adsorption was continued beyond the breakthrough point, the Cout/Cin
would rise rapidly to about 0.5 and then more slowly approach 1 and make S-shape
curves. The results show that at higher bed height slope decreases rapidly. Biosorption
capacity increases by increasing the bed height of the column as shown in Table 4.27.
The Breakthrough time and treated volume also increased as bed height increased. The
increase in metal uptake capacity with the increase of bed height in the fixed bed column
may be due to increased surface area of the adsorbent, which provided more binding sites
for the adsorption (Zulfadhly et al., 2001; Ghasemi et al, 2011; Zou et al., 2009).
120
Fig.4.49. Breakthrough curves at different bed heights for U(VI) and Zr(IV)
biosorption onto rice husk and bagasse. (a) Effect of bed height on U(VI) biosorption
at 1, 2 and 3 cm bed height, initial metal ion concentration of 50 mg/L having pH 4 at
constant flow rate of 1.8 mL/min.(b) Effect of bed height on Zr(IV) biosorption at 1, 2
and 3 cm bed height, initial metal ion concentration of 50 mg/L having pH3.5 at constant
flow rate of 1.8 mL/min.
0
0.2
0.4
0.6
0.8
1
1.2
0 200 400 600 800
Cout/Cin
Time (min)
1 Cm
2 cm
3 cm
(a)
0
0.2
0.4
0.6
0.8
1
1.2
0 200 400 600 800 1000 1200
C out/Cin
Time (min)
1 Cm
2 cm
3 cm
(b)
121
4.17.2. Effect of flow rate
The breakthrough curves for U(VI) and Zr(IV) at various flow rates of 1.4, 3.6 and 5.4
mL/min through a 3 cm bed height column and inlet concentration of 50 mg/L of uranium
and zirconium are shown in Fig.4.49 and the breakthrough parameters of column
calculated are presented in Table 4.27. The metal adsorption per unit mass decreases at
high flow rate. This indicates that among three steps of the adsorption process (film
diffusion, pore diffusion, surface diffusion), the surface diffusion by occupation of active
sites on the surface of biosorbents is the dominant step at the overall metal removal
process. The results show that with increasing the flow rate from 1.8 to 5.4 mL/min the
breakthrough curves shift towards a lower time scale and the breakthrough time decrease
because if the flow rate increases, not all the metal ions will have enough time to
penetrate from the solution to the biomasses (Shahbazi et al., 2011; ).
122
Fig.4.50. Breakthrough curves at different flow rates for U(VI) and Zr(IV)
biosorption onto rice husk and bagasse. (a) Effect of flow rate on U(VI) biosorption
onto rice husk at1.8, 3.6 and 5.4 mL/min cm bed height at initial metal ion concentration
of 50 mg/L having pH 4 at constant bed height of 3 cm. (b) Effect of flow rate on Zr(IV)
biosorption onto bagasse at using 1.8, 3.6 and 5.4 mL/min cm bed height at initial metal
ion concentration of 50 mg/L having pH 3.5 at constant bed height of 3 cm packed with
bagasse.
0
0.2
0.4
0.6
0.8
1
1.2
0 200 400 600 800
Cout/Cin
Time (min)
1.8 L/min
3.6mL/min
5.4 mL/min
(a)
0
0.2
0.4
0.6
0.8
1
1.2
0 200 400 600 800 1000 1200
Cout/Cin
Time (min)
1.8 mL/min
3.6mL/min
5.4 mL/min
(b)
123
Table 4.27. Column sorption capacity and breakthrough time with different bed heights,
flow rates and inlet concentrations.
Inlet
concentration
(mg/L)
Bed
height
(cm)
Flow rate
(mL/min)
Treated
volume
(mL)
Breakthrough
Point (50%)
(min)
Biosorption
capacity
(mg/g)
50 %
Uranium
50 1 1.8 540 100 8.64
50 2 1.8 900 220 9.504
50 3 1.8 1260 480 13.824
50 3 3.6 1584 140 9.216
50 3 5.4 1836 80 6.912
25 3 1.8 1440 480 10.528
75 3 1.8 1044 320 14.1696
Zirconium
50 1 1.8 1116 180 15.228
50 2 1.8 1440 420 17.766
50 3 1.8 1764 680 19.176
50 3 3.6 1404 300 16.92
50 3 5.4 828 160 13.536
25 3 1.8 1656 740 10.716
75 3 1.8 1764 580 25.6824
124
4.17.3. Effect of initial metal ion concentration
The effect of the inlet metal ions concentration on the adsorption of U(IV) and Zr(IV)
onto rice husk and bagasse was investigated using various concentrations of 25, 50 and 75
mg/L at a constant bed height of 3 cm and flow rate of 1.8 mL/min and results are
presented in Fig.4.51 and Table 4.27. Adsorption capacity increases at higher
concentration while breakthrough time decreases. This is due to the high driving force for
the adsorption process at higher concentration.
This can be explained by the fact that more adsorption sites are being covered with the
increase in metal ion concentration. The larger the influent concentration, the steeper the
slope of the breakthrough curve and smaller the breakthrough time. These results
demonstrated that the change of concentration gradient affected the saturation rate and
breakthrough time, or in other words, the diffusion process was concentration dependent
(Goel et al., 2005; Vijayaraghavan et al., 2004)
125
Fig.4.51. Breakthrough curves at different initial inlet metal ion concentration for
U(VI) and Zr(IV) biosorption onto rice husk and bagasse. (a) Effect of initial metal
ion concentration on U(VI) biosorption 25, 50 and 75 mg/L having pH 4 at constant bed
height of 3 cm packed with rice husk. (b) Effect of initial metal ion concentration onto
Zr(IV) biosorption. 25, 50 and 75 mg/L of Zr (IV) solution having pH 3.5 at constant bed
height of 3 cm packed with bagasse.
0
0.2
0.4
0.6
0.8
1
1.2
0 200 400 600 800 1000
Cout/Cin
Time (min)
50 ppm
25 ppm
75 ppm
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 200 400 600 800 1000 1200
Cout/Cin
Time (min)
50 ppm
25 ppm
75 ppm
126
4.17.4. Application of Thomas model.
Data of different column depth, flow rate and initial metal ion concentration were also
tested for Thomas model and the parameters are given in Table 4.28. The results represent
that the value of qo (sorption capacity) was increased, KTh reduced by increasing the bed
heights from 1 to 3 cm. Reduction in Kth value by increasing the initial concentration of
metal ions from 25 to 75 mg/L was observed. The opposite results were obtained for
column data of flow rates. The qo value was decreased while the KTh value was increased
by increasing the flow rate from 1.8 mL/min to 5.4 mL/min.
Table 4.28.
Thomas Model parameters for the removal of U(VI) and Zr (IV) by rice husk and
bagasse.
Concentration (mg/L)
Bed height (cm)
Flow rate (mL/min)
KTH
(mL/min mg)
qo (mg/g) R2
Uranium 50 1 1.8 0.5 8.129 0.925
50 2 1.8 0.2 8.146 0.972
50 3 1.8 0.2 9.674 0.751
50 3 3.6 0.3 7.786 0.893
50 3 5.4 0.6 7.155 0.894
25 3 1.8 0.2 6.099 0.940
75 3 1.8 0.1 11.856 0.952
Zirconium
50 1 1.8 0.2 18.40 0.984
50 2 1.8 0. 1 16.06 0.983
50 3 1.8 0.1 15.61 0.929
50 3 3.6 0.5 16.56 0.986
50 3 5.4 0.259 14.92 0.965
25 3 1.8 0.178 8.304 0.806
75 3 1.8 0.215 20.32 0.915
127
4.17.5. Application of Bed Depth Service Time (BDST) model
The BDST model is based on physically measuring the capacity of the bed at different
breakthrough values. This simplified design model ignores the intraparticle mass transfer
resistance and external film resistance such that the sorbate is adsorbed onto the adsorbent
surface directly (Sadaf and Bhatti, 2013). With these assumptions, the BDST model
works well and provides useful modeling equations for the changes of the system
parameters.
The values of slope and intercept for respective Ct/Cin ratio are listed in 4.29. The rate
constant Ka(L/mg min) represents the rate of rate of transfer of metal ions from its
solution to biomass surface. From Table 4.29, it can be seen that high correlation
coefficient values suggesting that data fixing on BDST model. With the values of Ct/C0
increasing, the values of N0 increased while Ka decreased. The BDST model parameters
can be helpful to scale up the process for other flow rates without further experimental
run (Han et al., 2008).
Table 4.29.
Bed Depth Service Time model parameters for the removal of U(VI) and Zr (IV) by rice husk and bagasse.
Ct/C0 a b Ka(L/mg min)
Nο (mg/L) R2
Uranium
0.2 100 -100 0.000253 2280 0.893
0.4 155 -106.67 0.000654 3534 0.945
0.6 200 -10 -0.0000799 4560 0.971
Zirconium
0.2 70 26.667 -0.00099 1562.53 0.993
0.4 190 -46.667 0.0000282 4241.71 0.999
0.6 240 26.667 -0.000413 5358 0.991
128
Chapter-5
___________________________________________SUMMARY
This study was aimed to explore the potential of selected indigenous abundantly available
agro-wastes as sorbents for the removal of U, Zr and Sr from synthetic aqueous solutions.
Results obtained from the present investigation indicate that agro-wastes are very
efficient and promising biosorbents for the removal of selected radioactive metals (U, Zr
& Sr) from aqueous media. Initially experiments were conducted to select most efficient
biosorbent among selected i.e rice husk, peanut husk, wheat bran, bagasse and cotton
sticks for removal of U, Zr and Sr. The screening results showed that rice husk, bagasse
and peanut husk were most optimal sorbents for U, Zr and Sr respectively. The selected
biosorbents were than subjected to different physical and chemical treatment to see the
effect of these pretreatments on removal efficiency and most efficient biomass was
further used for batch mode sorption. The immobilization of the selected biomasses was
done using sodium alginate and the results obtained describe that pre-treatments (physical
& chemical) and modification (immobilization) of the biomasses showed great effect on
biosorption capacity. The effect of batch mode sorption affecting parameters like pH,
sorbent amount, time, initial metal ion concentration and temperature on metal ions
removal was studied over the wide range of these parameters for native, pretreated and
immobilized biomasses for each metal ion. The results showed that pH of the medium,
sorbent amount and initial metal ion concentration strongly affected the removal
efficiency of biosorbents as compared to time of contact and temperature of the solution.
Kinetic (Pseudo-first order and Pseudo-second order) model and equilibrium (Freundlich,
Langmuir and Redlich-Peterson) models were satisfactory optimized by comparing R2
value of both linear and non-linear regression and six different non-linear error functions
(hybrid fractional error function (HYBRID), Marquardt’s percent standard deviation
(MPSD), average relative error (ARE), sum of the errors squared (ERRSQ), sum of the
absolute errors (EABS) and Chi-square test (χ2 )). Acids showed good desorption capacity
for sorbed U, Zr and Sr ions. Biosorption efficiency was mostly decreased in presence of
other competing metal ions however, anions showed less suppressing effect than
competing metal ions. The central composite face-centered experimental design in
response surface methodology (RSM) by Design Expert Version 7.0.0 (Stat Ease, USA)
129
was used for designing the experiments as well as for full response surface estimation to
see the combined effect of independent variables i.e. pH(A), sorbent amount (B) and
initial metal ion concentration (C) on response (sorption capacity). Characterization of
biosorbent materials by XRD, BET, SEM-EDX, TGA and FTIR helped in studying
composition of biosorbents and sorption mechanism. The functional groups of cellulose,
hemicellulose and lignin in biomaterials were found to be involved in adsorption of metal
ions. The effect of bed height, flow rate and initial metal ion concentration was also
studied in fixed bed column for removal of U and Zr ion in continuous sorption mode.
Breakthrough curves shows that bed height and initial metal ion concentration increased
the sorption capacity of the column while increase in flow rate decreased the column
efficiency for metal removal. Column biosorption data was satisfactory explained by
Thomas and BDST model. Column biosorption studies showed that metal uptake
efficiency is decreased as compared to batch biosorption mode experiment for U and Zr
removal; however the removal in continuous system has considerable efficacy for metal
ions removal at large scale.
Results showed that rice husk has the potential for U(VI) uptake in wastewater. The pre-
treatment with SDS showed highest increases the removal efficiency as compared to
native rice husk. pH of the medium strongly affected the removal of U(VI) and 4 was
optimized pH for native and immobilized and pH 5 for SDS-treated. Equilibrium was
achieved in 320 minutes and kinetic pseudo-second order was fitted well to native, SDS-
treated and immobilized rice husk. Maximum biosorption capacity value of 38.9, 42.4 and
38 mg/g were obtained for native, SDS-treated and immobilized rice husk. Langmuir
model provided the best correlation to the experimental data of native and Redlich-
Peterson to SDS-treated and immobilized rice husk. Thermodynamics studies showed that
removal of U is spontaneous and favorable at studied temperature. H2SO4 and EDTA
proved most successful eluting agents. The quadratic model regression coefficient R2 =
0.897 and AdjR2= 0.807 means that provides an excellent explanation of the relationship
between the independent variables and response (sorption capacity). The significant terms
in model as suggested by ANOVA are B, C and C2. The bisorption capacity in column
was increased with increasing bed height and initial inlet metal ion uranium concentration
and decreased with increasing flow rate. Column data was satisfactory analyzed by
Thomas and BDST model.
130
Screening results showed that bagasse has the highest potential to treat Zr(IV) containing
wastewater among selected agro-wastes. Pre-treatment with SDS enhanced the removal
efficiency of the bagasse. pH of the medium strongly affected the removal of Z(IV) and
maximum loading was observed at pH 3.5 for native and immobilized while 3 for SDS-
treated bagasse. Equilibrium was achieved in 160 minute for native and SDS-treated and
and 320 for immobilized bagasse and kinetic data was fitted well to pseudo-second order
model. Redlich-Peterson model provided the best correlation to the experimental data of
native, SDS-treated and immobilized bagasse for Zr(IV) removal. Maximum biosorption
capacity values of 107.4, 111.4, 71.5 mg/g were obtained for native, SDS-treated and
immobilized bagasse respectively. Thermodynamic studies showed that removal of Zr
was spontaneous at low temperature. H2SO4 was proved as most successful eluting
agents. A, B, C, AB, A2, C2 are significant model terms suggested by response surface
methodology and good fitness of quadratic model with R2= 0.979 and Adj R2= 0.9597.
Column parameters satisfactory explained by Thomas and BDST model.
Peanut husk has potential to remove the Sr(II) ions from wastewater even in low
concentration. Sorbent amount strongly affected the sorption capacity of Sr(II) onto
peanut husk. The pH of the medium affected the sorption capacity and most optimal value
were pH 9 for native and 7 for immobilized and NaOH-treated peanut husk. Equilibrium
was achieved in 80 minutes for Sr sorption onto native and NaOH-treated while in 160
min for immobilized peanut husk. Native and NaOH-treated kinetic data was fitted to
pseudo-second order and immobilized peanut husk data by pseudo-first order model.
Redlich-Peterson isothermal model had the best correlation to the experimental data of
native and NaOH while Freundlich by immobilized. Maximum biosorption capacity 9.4,
17.6, 38.04 mg/g for native, NaOH-treated and immobilized peanut husk.
Thermodynamics showed that removal of Sr(II) was spontaneous and favorable at all
studied temperatures. The R2 = 0.9302 and AdjR2=0.867 means that regression model
showed good correlation between among the predicted and the experimental values of the
response suggested for the suitability of the selected quadratic model in predicting the
response variable for the validation data set comprised of different combinations of the
process variables. ANOVA shows that in this case A, B, AC and A2 are the significant
model terms. HCl and EDTA proved most successful eluting agents for sorbed Sr(II)
ions.
131
From results of the present research work for biosorption of selected metal ions (U, Zr
and Sr), we can conclude that wastewater containing radioactive metal ions can be treated
in a very efficient and economical way in developing and agricultural country like
Pakistan by using this method. So, we can suggest that sorption technology using
agrowaste at large scale would be effective, ecofriendly and inexpensive method for
wastewater treatment of these metal ions.
132
LITERATURE CITED
Ahalya, N., R. D. Kanamadi and T. V. Ramachandra. 2005. Biosorption of
chromium(VI) from aqueous solutions by the husk of Bengal gram (Cicer
arientinum). European Journal of Biotechnology, 8: 258–264.
Aksu, Z., 2005. Application of biosorption for the removal of organic pollutants: a
review. Process Biochemistry, 40: 997-1026.
Aksu, Z., I. and A. Isoglu. 2006. Use of agricultural waste sugar beet pulp for the removal
of Gemazol turquoise blue-G reactive dye from aqueous solution. Journal of
Hazardous Material, 137: 418–430.
Afkhami, A., T. Madrakian, Z. Karimi and A. Amini. 2007. Effect of treatment of carbon
cloth with sodium hydroxide solution on its adsorption capacity for the adsorption
of some cations. Colloids and Surfaces A: Physicochemical Engineering Aspects,
304:36–40.
Alluri, H. K., S. R. Ronda, V. S. Settalluri, B. J. Singh, V. Suryanarayana, and P.
Venkateshwar. 2007. Biosorption: An eco-friendly alternative for heavy metal
removal. African Journal of Biotechnology, 6:2924-2931.
Arshad, M., M. N. Zafar, S. Younis and R. Nadeem. 2008. The use of Neem biomass for
the biosorption of zinc from aqueous solutions. Journal of Hazardous Materials,
157: 534–540.
Amini, M. H. Younesi, N. Bahramifar, A. A. Z. Lorestani, F. Ghorbani, A. Daneshi and
M. Sharifzadeh. 2008. Application of response surface methodology for
optimization of lead biosorption in an aqueous solution by Aspergillus niger.
Journal of Hazardous Materials, 154: 694–702.
Akhtar, K., M. W. Akhtar and A. M. Khalid. 2008. Removal and recovery of zirconium
from its aqueous solution by Candida tropicalis. Journal of Hazardous Materials,
156: 108-117.
Ayoob, S. and A. K. Gupta. 2008. Insights into isotherm making in the sorptive removal
of fluoride from drinking water. Journal of Hazardous Material, 152:976–985.
Akhter, M., M. I. Bhanger and M. Moazzam. 2009. Utilization of organic by-products for
the removal of organophosphorous pesticide from aqueous media. Journal of
Hazardous Materials, 69: 63-70.
133
Asgher, M. and H. N. Bhatti. 2010. Mechanistic and kinetic evaluation of biosorption of
reactive azo dyes by free, immobilized and chemically treated Citrus sinensis
waste biomass. Ecological Engineering, 36:1660-1665.
Ahmadpour, A., M. Zabihi, M. Tahmasbi and T. R. Bastami. 2010. Effect of adsorbents
and chemical treatments on the removal of strontium from aqueous solutions.
Journal of Hazardous Materials, 182:552–556.
Abdel Rahman. R. O., H. A. Ibrahium and Y-T. Hung. 2011. Liquid Radioactive Wastes
Treatment: A Review. Water, 3: 551-565.
Anupam, K., S. Dutta, C. Bhattacharjee and S. Datta. 2011. Adsorptive removal of
chromium (VI) from aqueous solution over powdered activated carbon:
Optimisation through response surface methodology. Chemical Engineering
Journal, 173:135– 143.
Auta, M., and B. H. Hameed. 2011. Optimized waste tea activated carbon for adsorption
of Methylene Blue and Acid Blue 29 dyes using response surface methodology.
Chemical Engineering Journal, 175: 233– 243.
Aytas, S., D. A. Turkozu and C. Gok. 2011. Biosorption of uranium(VI) by bi-
functionalized low cost biocomposite adsorbent. Desalination, 280:354–362.
Aydin, M., L. Cavas and M. Merdivan. 2012. An alternative evaluation method for
accumulated dead leaves of Posidonia oceanica (L.) Delile on the beaches:
Removal of uranium from aqueous solutions. Journal of Radioanalytical and
Nuclear Chemistry, 293: 489-496.
Bohart, G.S, and Adams, E.Q. 1920. Some aspects of the behaviour of charcoal with
respect to chlorine. Journal of American Chemical Society, 42:523–44.
Box, G. E. P. and N. R. Draper. 1987. Empirical Model-building and Response Surfaces,
John Wiley & Sons, New York.
Bhatti, M., A. M. Amin, K. A. Malik and A. M. Khalid. 1991. Spectrophotometric
determination of Uranium(VI) in Bacterial Leach Liquores Using Arsenazo III.
Journal of Chemical Technology and Biotechnology, 52: 331-341.
Bhatti, H. N., B. Mumtaz, M. A. Hanif and R. Nadeem. 2007. Removal of Zn(II) ions
from aqueous solution using Moringa oleifera Lam. (horseradish tree) biomass.
Process Biochemistry, 42: 547–553.
Binupriya, A. R., M. Sathishkumar, D. Kavitha, K. Swaminathan, S. E. Yun and S. P.
Mun. 2007. Experimental and isothermal studies on sorption of Congo red by
modified mycelial biomass of wood rotting fungus. Clean, 35: 143–150.
134
Boulinguiez, B., P. L. Cloirec and D. Wolbert. 2008. Revisiting the determination of
Langmuiparameters application to tetrahydrothiophene adsorption onto activated
carbon. Langmuir, 24: 6420–6424.
Bezerra, M. A., R. E. Santelli, E. P. Oliveira, L. S. Villar and L. A. Escaleira, 2008.
Response surface methodology (RSM) as a tool for optimization in analytical
chemistry. Talanta, 76: 965-977.
Bhatti, H. N., R. Khalid and M. A. Hanif. 2009. Dynamic biosorption of Zn(II) and Cu(II)
using pretreated Rosa gruss an teplitz (red rose) distillation sludge. Chemical
Engineering Journal, 148:434–443.
Başçetin, E. and G. Atun. 2010. Adsorptive removal of strontium by binary mineral
mixtures of montmorillonite and zeolite. Journal of Chemical and Engineering
Data, 55:783-788.
Bhatti, H. N. and M. Amin. 2013. Removal of zirconium(IV) from aqueous solution by
Coriolus versicolor and thermodynamic study. Ecological Engineering, 51:178–
180.
Chen, J. P and S. Wu. 2004. Acid/base-treated activated carbons: characterization of
functional groups and metal adsorptive properties. Langmuir, 20: 2233–2242.
Cabuk, A., S. Ilhan, C. Filik and F. Caliskan. 2005. Pb2+ biosorption by pre-treated fungal
biomass. Turkish Journal of Biology, 29:23–28.
Can, M. Y. and Y. Kaya and F. Algur. 2006. Response surface optimization of the
removal of nickel from aqueous solution by cone biomass of Pinus sylvestris.
Bioresource Technology, 97:1761–1765.
Chegrouche, S., A. Mellah and M. Barkat. 2009. Removal of strontium from aqueous
solutions by adsorption onto activated carbon: kinetic and thermodynamic studies.
Desalination, 235: 306–318.
Chen, H., J. Zhao, J. Wu and G. Dai. 2011. Isotherm, thermodynamic, kinetics and
adsorption mechanism studies of methyl orange by surfactant modified silkworm
exuviae. Journal of Hazardous Materials, 192: 246–254.
Chowdhury, S. and P. D. Saha. 2011. Comparative analysis of linear and nonlinear
methods of estimating the pseudo-second-order kinetic parameters for sorption of
malachite green onto pretreated rice husk. Bioremediation Journal, 15: 181-188.
Chowdhury, S., R. Misra, P. Kushwaha and P. Das. 2011. Optimum Sorption Isotherm by
Linear and Nonlinear Methods for Safranin onto Alkali-Treated Rice Husk.
Bioremediation Journal, 15:77-89.
135
Chan, L. S., W. H. Cheung, S. J. Allen and G. McKay. 2012. Error Analysis of
Adsorption Isotherm Models for Acid Dyes onto Bamboo Derived Activated
Carbon. Chinese Journal of Chemical Engineering, 20: 535-542.
Chatterjee, S., A. Kumara, S. Basu and S. Dutta. 2012. Application of Response Surface
Methodology for Methylene Blue dye removal from aqueous solution using low
cost adsorbent. Chemical Engineering Journal, 181– 182:289– 299.
Dushenkov, S. 2003.Trends in phytoremediation of radionuclides. Plant and Soil,
249:167–175.
Demirbas, A. 2008. Heavy metal adsorption onto agro-based waste materials, A review:
Journal of Hazardous Materials, 15: 220-229
Das, D., G. Basak, V. Lakshmi and N. Das. 2012. Kinetics and equilibrium studies on
removal of zinc(II) by untreated and anionic surfactant treated dead biomass of
yeast: Batch and column mode. Biochemical Engineering Journal, 64:30–47.
Ding, D.-X., X.-T. Liu, N. Hu, G.-Y. Li and Y.-D. Wang. 2012. Removal and recovery of
uranium from aqueous solution by tea waste. Journal of Radioanalytical and
Nuclear Chemistry, 293:735-741.
Elmaci, A., T. Yonar and N. Ozengin. 2007. Biosorption characteristics of copper(II),
chromium(III), nickel(II), and lead(II) from aqueous solutions by Chara sp. and
Cladophora sp. Water Environment Research, 79: 1000–1005.
El-Kamash, A. M. 2008. Evaluation of zeolite A for the sorptive removal of Cs+ and Sr2+
ions from aqueous solutions using batch and fixed bed column operations. Journal
of Hazardous Materials, 151:432-445
Eriksson, E. and E. Donner. 2009. Metals in greywater: Sources, presence and removal
efficiencies. Desalination, 248: 271-278.
El Hamidi, A., S. Arsalane and M. Halim. 2012. Kinetics and Isotherm Studies of Copper
Removal by Brushite Calcium Phosphate: Linear and Non-Linear Regression
Comparison. E-Journal of Chemistry, 9: 1532-1542
Freundlich, H. M. F. 1906. Over the adsorption in solution. Journal of Physical
Chemistry, 57:385–471.
Faghihian, H. and M. Kabiri-Tadi. 2010. Removal of zirconium from aqueous solution by
modified clinoptilolite. Journal of Hazardous Materials, 178:66–73.
Foo, K. Y. and B. H. Hameed. 2010. Insights into the modeling of adsorption isotherm
systems. Chemical Engineering Journal, 156: 2–10.
136
Fu, F. and Q. Wang. 2011. Removal of heavy metal ions from wastewaters: A review.
Journal of Environmental Management, 92:407-418.
Fatima, H., N. Djamel, A. Samira and B. Mahfoud. 2013. Modelling and adsorption
studies of removal uranium (VI) ions on synthesised zeolite NaY. Desalination
and Water Treatment 51: 5583-5591
Gonzilez-Muiioz, M. T., M. L. Merroun, N. B.Omar and J. M. Arias. 1997. Biosorption
of Uranium by Myxococcus Xanthus. International Biodeterioration &
Biodegradation, 40: 107-114.
Gupta R., P. Ahuja, S. Khan, R. K. Saxena and H. Mohapatra. 2000. Microbial
biosorbents: meeting challenges of heavy metal pollution in aqueous solutions.
Current Science, 78: 967-973.
Guo, G. Y., Y. L. Chen and W. J. Ying. 2004. Thermal spectroscopic and X-ray
diffractional analyses of zirconium hydroxides precipitated at low pH values.
Materials Chemistry and Physics, 84: 308-314.
Goel, J., K. Kadirvelu, C. Rajagopal and V. K. Garg. 2005. Removal of lead (II) by
adsorption using treated granular activated carbon: batch and column studies.
Journal of Hazardous Material, 125: 211-220.
Garg, U. K., M. P. Kaur, V. K. Garg and D. Sud. 2008. Removal of Nickel(II) from
aqueous solution by adsorption on agricultural waste biomass using a response
surface methodological approach. Bioresource Technology, 99:1325–1331.
Garg, U., M. P. Kaur, G. K. Jawa, D, Suda and V. K. Garg. 2008. Removal of cadmium
(II) from aqueous solutions by adsorption on agricultural waste biomass. Journal
of Hazardous Materials, 154 :1149–1157.
Groudev, S., P. Georgiev, I. Spasova and M. Nicolova. 2008. Bioremediation of acid
mine drainage in a uranium deposit. Hydrometallurgy, 94: 93–99.
Garg, U. K., M. P. Kaur, D. Sud and V. K. Garg. 2009. Removal of hexavalent chromium
from aqueous solution by adsorption on treated sugarcane bagasse using response
surface methodological approach. Desalination, 249: 475–479.
Gurbuz, F. 2009. Removal of toxic hexavalent chromium ions from aqueous solution by a
natural biomaterial: Batch and column adsorption. Adsorption Science and
Technology, 27:745-759
Ghaemi, A., T-M, Meisam M. and G-M. Maragheh. 2011. Characterizations of strontium
(II) and barium(II) adsorption from aqueous using dolomite powder. Journal of
Hazardous Materials, 190: 916–921.
137
Gupta, C., C. Umesh and S. Gupta. 2011. Metal toxicity in humans and its preventive
and control measures. Current Nutrition and Food Science, 7: 221-231.
Guan, W., J. Pan, H. Ou, X. Wang, X. Zou, W. Hu, C. Li , X. Wu. 2011. Removal of
strontium(II) ions by potassium tetratitanate whisker and sodium trititanate
whisker from aqueous solution: Equilibrium, kinetics and Thermodynamics.
Chemical Engineering Journal, 167:215–222.
Ghasemi, M., A. R. Keshtkar, R. Dabbagh, S. and J. Safdari. 2011. Biosorption of
uranium(VI) from aqueous solutions by Ca-pretreated Cystoseira indica alga:
Breakthrough curves studies and modelling. Journal of Hazardous Materials, 189 :
141–149.
Ghorbani, M. and H. Eisazadeh. 2012. Fixed bed column study for Zn, Cu, Fe and Mn
removal from wastewater using nanometer size polypyrrole coated on rice husk
ash. Synthetic Metals, 162:1429– 1433.
Gok, C. U. Gerstmann and S. Aytas. 2013. Biosorption of radiostrontium by alginate
beads: application of isotherm models and thermodynamic studies. Journal of
Radioanalytical and nuclear chemistry, 295:777–788.
Ho, Y.S. and G. Mckay. 1999. Pseudo-Second Order Model for Sorption Processes.
Process Biochemistry, 34: 451-465.
Ho, Y. S., J. F. Porter and G. Mckay. 2002. Equilibrium isotherm studies for the sorption
of divalent metal ions onto peat: copper, nickel and lead single component
systems. Water, Air, and Soil Pollution, 141: 1–33.
Ho.Y. S. 2004. Selection of optimum sorption isotherm. Carbon, 42: 2113–2130
Ho, Y. S., 2006b. Second-order kinetic model for the sorption of cadmium onto tree fern:
a comparison of linear and non-linear methods. Water Research, 40:119–125.
Ho, Y-S and A.E. Ofomaja. 2006. Kinetic studies of copper ion adsorption on palm
kernel fibre. Journal of Hazardous Materials, 137: 1796-1802.
Han, R. P., J. J. Zhang, P. Han, Y. F. Wang, Z. H. Zhao and M. S. Tang. 2009. Study of
equilibrium, kinetic and thermodynamic parameters about methylene blue
adsorption onto natural zeolite. Chemical Engineering Journal, 145:496–504.
Hanif, A., H. N. Bhatti and M. A. Hanif. 2009. Removal and recovery of Cu(II) and
Zn(II) using immobilized Mentha arvensis distillation waste biomass. Ecological
Engineering, 35: 1427–1434
138
Hadi, M., M. R. Samarghandi and G. McKay. 2010. Equilibrium two-parameter
isotherms of acid dyes sorption by activated carbons: Study of residual errors.
Chemical Engineering Journal, 160 :408–416.
Han, C., H. Pua, H. Li, L. Deng, S.Huang, S. He and Y. Luo. 2013. The optimization of
As(V) removal over mesoporous alumina by using response surface methodology
and adsorption mechanism. Journal of Hazardous Materials, 254– 255: 301– 309.
Hanif, A., H. N. Bhatti and M. A. Hanif. 2013. Removal of zirconium from aqueous
solution by Ganoderma lucidum: biosorption and bioremediation studies.
Desalination and Water Treatment, DOI:10.1080/19443994. 837005.
Hussein, A. E. M. and M. H. Taha. 2013. Uranium removal from nitric acid raffinate
solution by solvent immobilized PVC cement. Journal of Radioanalytical and
Nuclear Chemistry, 295:709-715.
Im, J-K, I-H. Cho, S-K. Kim and K-D. Zoh. 2012. Optimization of carbamazepine
removal in O3/UV/H2O2 system using a response surface methodology with
central composite design. Desalination, 285: 306–314.
Jumasiah, A., T. G. Chuah, J. Gimbon, T. S. Y. Choong and I. Azni. 2005. Adsorption of
basic dye onto palm kernel shell activated carbon: Sorption equilibrium and
kinetics studies. Desalination, 186: 57-64.
Kalavathy, H. M., I. Regupathib, M. G.Pillai and L.R. Miranda. 2009. Modelling,
analysis and optimization of adsorption parameters for H3PO4 activated rubber
wood sawdust using response surface methodology (RSM). Colloids and Surfaces
B: Biointerfaces, 70:35–45.
Kalavathy, H., B. Karthik and L. R. Miranda. 2010. Removal and recovery of Ni and Zn
from aqueous solution using activated carbon from Hevea brasiliensis: Batch and
column studies. Colloids and Surfaces B: Biointerfaces, 78: 291-302.
Jain, M., V. K. Garg and K. Kadirvelu. 2011. Investigation of Cr(VI) adsorption onto
chemically treated Helianthus annuus: Optimization using Response Surface
Methodology. Bioresource Technology, 102: 600–605.
Jian, Z., P. Qingwei, N. Meihong, S. Haiqiang and L. Na. 2013. Kinetics and equilibrium
studies from the methylene blue adsorption on diatomite treated with sodium
hydroxide. Applied Clay Science, 83–84:12–16.
Kapoor, A. and R. T. Yang. 1989. Correlation of equilibrium adsorption data of
condensable vapours on porous adsorbents. Gas and Separation. Purification,
3:187–192.
139
Khan, S. A., R. Rehman and M. A. Khan. 1995. Adsorption of Cr (III), Cr (VI) and Ag (I)
on Bentonite. Waste Management, 15: 271–282.
Kim, H. M., J. G. Kim, J. D. Cho and J. W. Hong. 2003. Optimization and
characterization of UV curable adhesives for optical communication by response
surface methodology. Polymer Testing, 22: 899–906.
Kocherginsky, N. M., Y. K. Zhang and J. W. Stucki. 2002. D2EHPA based strontium
removal from strongly alkaline nuclear waste. Desalination, 144: 267– 272.
Kumar, K.V., S. Sivanesan. 2006. Pseudo second order kinetics and pseudo isotherms for
malachite green onto activated carbon: comparison of linear and nonlinear
regression methods, Journal of Hazardous Material B, 136: 721–726
Kiran, B., A. Kaushik and C. P. Kaushik. 2007. Response surface methodological
approach for optimizing removal of Cr(VI) from aqueous solution using
immobilized cyanobacterium. Chemical Engineering Journal, 126: 147-153.
Kutahyali, C. and M. Eral. 2010. Sorption studies of uranium and thorium on activated
carbon prepared from olive stones: Kinetic and thermodynamic aspects. Journal of
Nuclear Materials, 396: 251–256.
Kassia, G. S., S. L. Taisa, V. M. Valeria, M. G. Marcos and A. S. Barrozo. 2010.
Pyrolysis of Sugarcane Bagasse: A Consecutive Reactions Kinetic Model from
TGA Experiment. Materials Science Forum, 660-661: 593.
Kumar, R., D. Bhatia, R. Singh, S. Rani and N.R. Bishnoi. 2011. Sorption of heavy
metals from electroplating effluent using immobilized biomass Trichoderma
viride in a continuous packed-bed column. International Biodeterioration &
Biodegradation, 65: 1133-1139.
Krusic, M. K., N. Milosavljevic, A. Debeljkovic, O. B. Uzum and E. Karadag. 2012.
Removal of Pb2+ ions from water by poly(acrylamide-co-sodium methacrylate)
hydrogels. Water, Air, and Soil Pollution, 223:4355-4368.
Kubota, T., S. Fukutani, T. Ohta and Y. Mahara. 2012. Removal of radioactive cesium,
strontium, and iodine from natural waters using bentonite, zeolite, and activated
carbon. Journal of Radioanalytical and Nuclear Chemistry, 296: 981-984.
Kikuchi, T and T. Sanaka. 2012. Biological Removal and Recovery of Toxic Heavy
Metals in Water Environment. Critical Reviews in Environmental Science and
Technology, 42:1007-1057.
Kumar, R. and S. K. Jain. 2012. Removal of strontium (II) from aqueous solution using
functionalized carbon nanotubes. International Journal of Nanoscience, 11:6.
140
Kumar, R., T.N. Abraham, and S.K. Jain. 2012. Silver nano particles impregnated
alumina for the removal of strontium(II) from aqueous solution. Advanced
Materials Letters, 3: 507-510.
Kausar, A. and H. N. Bhatti. 2013. Adsorptive removal of uranium from wastewater: A
review. Journal of Chemical Society of Pakistan, 35:1041-1052
Keshtkar, A. R., M. Irani and M. A. Moosavian. 2013. Removal of uranium (VI) from
aqueous solutions by adsorption using a novel electrospun PVA/TEOS/APTES
hybrid nanofiber membrane: Comparison with casting PVA/TEOS/APTES hybrid
membrane. Journal of Radioanalytical and Nuclear Chemistry, 295: 563-571.
Lagergren, S. 1898. Handlingar, 24: 1-39.
Langmuir, I. 1916. The constitution and fundamental properties of solids and liquids.
Journal of American chemical society, 38: 2221–2295.
Limousin, G., J. P. Gaudet, L. Charlet, S. Szenknect, V. Barthes and M. Krimissa. 2007.
Isotherms: a review on physical bases, modeling and measurement. Applied
Geochemistry, 22: 249–275.
Lu, N. P., A. G. Carles and A. M. El-Kamash. 2008. Evaluation of zeolite A for the
sorptive removal of Cs+ and Sr2+ ions from aqueous solutions using batch and
fixed bed column operations. Journal of Hazardous Material, 151: 432–445.
Li, Q., H. N. Liu and T. Y. Liu. 2010. Strontium and calcium ion adsorption by
molecularly imprinted hybrid gel. Chemical Engineering Journal, 157: 401–407.
Lian, A., X. G. Luo, X. Y. Lin and S. Z. Zhang. 2013. Removal of strontium ions from
aqueous solution by sunflower straw. Advanced Materials Research, 726-
731:1922-1925
Liu, Y., Q. Lia, X. Cao, Y. Wang, X. Jiang, M. Li, M. Hua and Z. Zhang. 2013. Removal
of uranium(VI) from aqueous solutions by CMK-3 and its polymer composite.
Applied Surface Science, 285:258– 266.
Marquardt, D. W. 1963. Analgorithm for least-squares estimation of nonlinear
parameters. Journal of the society for industrial and Applied Mathematics,
11:431–441
Miguel, G. S., S. D. Lambert and N. J. D. Graham. 2006. A practical review of the
performance of organic and inorganic adsorbents for the treatment of
contaminated waters. Journal of chemical technology and biotechnology, 81:1685-
1696
141
Monji, A. B., S. J. Ahmadi and E. Zolfonoum. 2008. Selective biosorption of zirconium
and hafnium from acidic aqueous solutions by rice bran, wheat bran and Platanus
orientalis tree leaves. Separation Science and Technology, 43: 597-608.
Misra, V. A. 2009. Review on the use of biopolymers for the removal of toxic metals
from liquid industrial effluents. International Journal of Environment and Waste
Management, 3:393-410.
Martin-Lara, M. A., I. L. R. Rico, I.d. l. C. A. Vicente, G.B. Garcia and M. C. D. Hoces.
2010. Modification of the sorptive characteristics of sugarcane bagasse for
removing lead from aqueous solutions. Desalination, 256: 58–63
Milosavljevic, N. B., M. T. Ristic, A. A. Peric-Grujic, J. M. Filipovic, S. B. Strbac, Z. L.
J. Rakocevic and M.T. K. Krusic. 2011. Sorption of zinc by novel pH-sensitive
hydrogels based on chitosan, itaconic acid and methacrylic acid. Journal of
Hazardous Materials, 192 : 846-854.
Mao, Y.-L., X-T. Wang, S-T. Luo and W-F. Liu. 2011. Adsorptive removal of strontium
from aqueous solution by utilizing Pseudomonas alcaligenes biomass as
biosorbent. Proceedings-3rd International Conference on Measuring Technology
and Mechatronics Automation, 351-354
Malamisa, S. and E. Katsoua. 2013. A review on zinc and nickel adsorption on natural
and modified zeolite, bentonite and vermiculite: Examination of process
parameters, kinetics and isotherms, Journal of Hazardous Materials, 252–
253:428– 461
Muhamad, M. H., S. Rozaimah, S. Abdullah, A. B. Mohamad, R. A. Rahman and A. A.
H. Kadhum. 2013. Application of response surface methodology (RSM) for
optimisation of COD, NH3-N and 2,4-DCP removal from recycled paper
wastewater in a pilot-scale granular activated carbon sequencing batch biofilm
reactor (GAC-SBBR). Journal of Environmental Management, 121:179-190.
Ng, J. C. Y., W. H. Cheung and G. McKay. 2003. Equilibrium studies for the sorption of
lead from effluents using chitosan. Chemosphere, 52: 1021–1030.
Nouri, L., I. Ghodbane, O. Hamdaoui and M. Chiha. 2007. Batch sorption dynamics and
equilibrium for the removal of cadmiumions from aqueous phase using wheat
bran, Journal of Hazardous Material, 149 : 115–125.
Nakbanpote, W., B. A. Goodmanb and P. Thiravetyanc. 2007. Colloids and Surfaces A:
Physicochemical and Engineering Aspects. 304:7–13.
142
Nadeem, R., T. M. Ansari and A.M. Khalid. 2008. Fourier Transform Infrared
Spectroscopic characterization and optimization of Pb(II) biosorption by fish
(Labeo rohita) scales. Journal of Hazardous Materials, 156: 64-73.
Nadeem, R., M. H. Nasir and M. A. Hanif. 2009. Pb (II) sorption by acidically modified
Cicer arientinum biomass Chemical Engineering Journal, 150: 40–48.
Ncibi, M. C., B. Mahjou, M. Seffen, F. Brouers and S. Gaspard. 2009. Sorption dynamic
investigation of chromium(VI) onto Posidonia oceanica fibres: Kinetic modelling
using new generalized fractal equation. Biochemical Engineering Journal, 46:141-
146.
Ncibi, M. C., M. B. Hamissa A. AFathallah, M. H. Kortas, T. Baklouti, B. Mahjoub and
M. Seffen. 2009. Biosorptive uptake of methylene blue using Mediterranean green
alga Enteromorpha spp. Journal of Hazardous Materials, 170:1050-1055.
Ngah, W. S. W. and M. A. K. M. Hanafiah. 2009. Removal of heavy metal ions from
wastewater by chemically modified plant wastes as adsorbents: A review.
Biotechnology Advances, 27:195–226.
Ngwenya, N. and E. M. N. Chirwa. 2010. Single and binary component sorption of the
fission products Sr2+, Cs+ and Co2+ from aqueous solutions onto sulphate reducing
bacteria Minerals Engineering, 23: 463–470.
Nurchi, V. M., G, Crisponi and I. Villaescusa. 2010. Chemical equilibria in wastewaters
during toxic metal ion removal by agricultural biomass. Coordination Chemistry
Reviews, 254:2181-2192.
Noreen, S. H. N.Bhatti, S. Nausheen, S. Sadaf and M. Ashfaq. 2013. Batch and fixed
bed adsorption study for the removal of DrimarineBlack CL-B dye from aqueous
solution using a lignocellulosic waste:A cost affective adsorbent. Industrial Crops
and Products, 50: 568– 579.
Ozdemir, G., T. Ozturk, N. Ceyhan, R. Isler and T. Cosar. 2003. Heavy metal biosorption
by biomass of Ochrobactrum anthropi producing exopolysaccharide in activated
sludge. Bioresource Technology, 90: 71–74.
Ozeroglu, C. and G. Keceli. 2006. Removal of strontium ions by a crosslinked copolymer
containing methacrylic acid functional groups. Journal of Radioanalytical and
Nuclear Chemistry. 268:.211-219.
Ofomaja, A. E., Y. S. Ho. 2007. Equilibrium sorption of anionic dye from aqueous
solution by palm kernel fiber as sorbent. Dyes Pigment, 74: 60-66.
143
Opeolu, B. O., O. Bamgbose, T. A. Arowolo, and M. T. Adetunji. 2010. Utilization of
biomaterials as adsorbents for heavy metals removal from aqueous matrices.
Scientific Research and Essays, 5:1780-1787.
Porter, J. F., McKay, G. and Choy, K. H. 1999. The prediction of sorption from a binary
mixture of acidic dyes using single- and mixed-isotherm variants of the ideal
adsorbed solute theory. Chemical Engineering Science, 54: 5863-5885.
Pavasant, P., R. Apiratikul, V. Sungkhum, P. Suthiparinyanont, S. Wattanachira and T. F.
Marhaba. 2006. Biosorption of Cu2+, Cd2+, Pb2+ and Zn2+ using dried marine green
macroalga Caulerpa lentillifera. Bioresource Technology, 97: 2321-2329.
Pang, C., Y. Liu, X.Cao, R. Hua, C. Wang and C. Li. 2010. Adsorptive removal of
uranium from aqueous solution using chitosan-coated attapulgite. Journal of
Radioanalytical and Nuclear Chemistry, 286 : 185-193.
Park, Y., W. S. Shin and S-J. Choi. 2012. Sorptive removal of cobalt, strontium and
cesium onto manganese and iron oxide-coated montmorillonite from groundwater.
Journal of Radioanalytical and Nuclear Chemistry, 291: 837-852
Pereira, F. V., L.V. A. Gurgel and L. F. Gil. 2010. Removal of Zn2+ from aqueous single
metal solutions and electroplating wastewater with wood sawdust and sugarcane
bagasse modified with EDTA dianhydride (EDTAD).Journal of Hazardous
Material, 176:856-863.
Park Y., W. S. Shin and S.-J. Choi. 2013. Removal of cobalt and strontium from
groundwater by sorption onto fishbone. Journal of Radioanalytical and Nuclear
Chemistry, 295:789-799.
Redlich, O. and D. L. Peterson. 1959. A useful adsorption isotherm, Journal of Physical
Chemistry. 63: 1024–1026.
Rivas, F. J., F. J. Beltran, O. Gimeno, J. Frades and F. Carvalho. 2006. Adsorption of
landfill leachates onto activated carbon equilibrium and kinetics. Journal of
Hazardous Material B, 131:170–178.
Rehman, A., F. R. Shakoori and A. R. Shakoori. 2008. Heavy metal resistant freshwater
ciliate, Euplotes mutabilis, isolated from industrial effluents has potential to
decontaminate wastewater of toxic metals. Bioresource Technology, 99: 3890-
3895.
Riaz, M. R. Nadeem, M. A. Hanif, T. M. Ansari and K. U. Rehman. 2009. Pb(II)
biosorption from hazardous aqueous streams using Gossypium hirsutum (Cotton)
waste biomass. Journal of Hazardous Materials, 161: 88-94.
144
Rao, K. S., S. Anand and P. Venkateswarlu. 2010. Cadmium removal from aqueous
solutions using biosorbent Syzygium cumini leaf powder: Kinetic and equilibrium
studies. Korean Journal of Chemical Engineering, 27:1547-1554.
Reddy, P. M. K., S. Mahammadunnis, B. Ramaraju, B. Sreedhar and C. Subrahmanyam.
2013. Low-cost adsorbents from bio-waste for the removal of dyes from aqueous
solution. Environmental Science and Pollution Research, 20:4111-24.
Roy, A., S. Chakraborty, S. P. Kundu, B. Adhikari and S. B. Majumder. 2013.
Lignocellulosic jute fiber as a bioadsorbent for the removal of azo dye from its
aqueous solution: Batch and column studies. Journal of Applied Polymer Science,
129: 15-27.
Sag, Y. and T. Kutsal. 1995. Biosorption of heavy metals by Zoogloea ramigera: use of
adsorption isotherms and a comparison of biosorption characteristics. Chemical
Engineering Journal, 60: 181–188.
Steel, R. G. D., J. H. Torrie and D. Dickery. 1997. Principles and procedures of statistics.
A biomaterial Approach, 3rd Ed. McGraw Hill Book Co. Inc., New York, USA.
Siege, l. J and Y. Zuo. 2000. Using seafood processing waste to clean up wastewater.
Biocycle, 41:34-34.
Sekhar, K.C., C.T. Kamala, N.S. Chary and Y. Anjaneyulu. 2003. Removal of heavy
metals using a plant biomass with reference to environmental control.
International Journal of Mineral. Processing, 68: 37–45.
Steudel, K., G. Horak, S. Willscher, W. Pompe and P. Werner. 2007. Removal of copper
and uranium from contaminated waters in biosorption columns. Advanced
Materials Research, 20-21:627-630.
Sahu, J., N. J. Acharya and B. C. Meikap. 2009. Response surface modeling and
optimization of chromium(VI) removal from aqueous solution using Tamarind
wood activated carbon in batch process. Journal of Hazardous Materials, 172:818–
825.
Sharma, P., l. Singh and N. Dilbaghi, 2009. Response surface methodological approach
for the decolorization of simulated dye effluent using Aspergillus fumigatus
fresenius. Journal of Hazardous Material, 161: 1081-1086.
Svilovic, S., D. Rusic and R. Stipisic. 2009. Modeling batch kinetics of copper ions
sorption using synthetic zeolite NaX. Journal of Hazardous Materials, 170:941-
947.
145
Sert, S. and M. Eral. 2010. Uranium adsorption studies on aminopropyl modified
mesoporous sorbent (NH2–MCM-41) using statistical design method. Journal of
Nuclear Materials, 406:285–292.
Singh, K. P., S. Gupta, A. K. Singha and S. Sinha. 2010. Experimental design and
response surface modeling for optimization of Rhodamine B removal from water
by magnetic nanocomposite. Chemical Engineering Journal, 165: 151–160.
Safa, Y., H. N. Bhatti, I. A. Bhatti and M. Asgher. 2011. Removal of direct Red-31 and
direct Orange-26 by low cost rice husk: Influence of immobilization and
pretreatments. Journal of Chemical Engineering, 89:1565.
Salman, J. M., V. O. Njoku and B. H. Hameed. 2011. Bentazon and carbofuran
adsorption onto date seed activated carbon: Kinetics and equilibrium. Chemical
Engineering Journal, 173:361-368
Sato, I., H. Kudo and S. Tsuda. 2011. Removal efficiency of water purifier and adsorbent
for iodine, cesium, strontium, barium and zirconium in drinking water. Journal of
Toxicological Sciences, 36:829-834.
Shahbazi, A., H. Younesia and A. Badiei. 2011. Functionalized SBA-15 mesoporous
silica by melamine-based dendrimer amines for adsorptive characteristics of
Pb(II), Cu(II) and Cd(II) heavy metal ions in batch and fixed bed column.
Chemical Engineering Journal, 168 : 505–518.
Saleem, N. and H. N. Bhatti., 2011, Adsorptive removal and recovery of U(VI) by citrus
waste biomass., Bioresources, 6: 2522-2538.
Saifuddin, N. and S. Dinara. 2012. Immobilization of Saccharomyces Cerevisiae onto
cross-linked Chitosan coated with magnetic nanoparticles for adsorption of
Uranium (VI) ions. Advances in Natural and Applied Sciences, 6: 249-267.
Sadaf, S. and H. N. Bhatti. 2014. Batch and fixed bed column studies for the removal of
Indosol Yellow? BG dye by peanut husk. Journal of Taiwan Institute of Chemical
Engineering, 45:541-553
Thomas, H. C. 1944. Heterogeneous ion exchange in a flowing system. Journal of
American Chemical Society, 66:1466–664
Tangaromsuk, J., P. Pokethitiyook, M. Kruatrachue and E. S. Upatham. 2002. Cadmium
biosorption by Sphingomonas paucimobilis biomass. Bioresource Technology, 85:
103-105.
146
Tan, I. A. W., A.L. Ahmad and B. H. Hameed. 2008. Preparation of activated carbon
from coconut husk: Optimization study on removal of 2,4,6-trichlorophenol using
response surface methodology. Journal of Hazardous Materials, 153: 709–717.
Tavares, P. M. A., R. O. Cristovao, J. M. Loureiro, R. A. R. Boaventura and E. A.
Macedo. 2009. Application of statistical experimental methodology to optimize
reactive dye decolorization by commercial laccase. J. Hazardous Material, 162:
1255–1260.
Tian, G., J. Geng, Y. Jin, C. Wang, S. Li, Z. Chen, H. Wang, Y. Zhao and S. Li. 2011.
Sorption of uranium(VI) using oxime-grafted ordered mesoporous carbon CMK-
5. Journal of Hazardous Material, 190:442.
Torab-Mostaedi, M., A. Ghaemi, H. Ghassabzadeh and M.Ghannadi-Maragheh. 2011.
Removal of strontium and barium from aqueous solutions by adsorption onto
expanded perlite. Canadian Journal of Chemical Engineering, 89: 1247-1254.
Tofan, L., C. Teodosiu, C. Paduraru and R. Wenkert. 2013. Cobalt (II) removal from
aqueous solutions by natural hemp fibers: Batch and fixed-bed column studies.
Applied Surface Science, 285:33-39.
Ullah, I., R. Nadeem, M. Iqbal and Q. Manzoor. 2013. Biosorption of chromium onto
native and immobilized sugarcane bagasse waste biomass. Ecological
Engineering, 60:99– 107.
Veglio, F. and F. Beolchini. 1997. Removal of metals by biosorption: a review.
Hydrometallurgy, 44:301-316.
Vijayaraghavan, K., J. Jegan, K. Palanivelu and M. Velan. 2004. Removal of nickel (II)
ions from aqueous solution using crab shell particles in a packed bed up flow
column”, Journal of Hazardous Material, 113: 223-230.
Vijayaraghavan, K. and Y-S. Yun. 2007. Chemical modification and immobilization of4
Corynebacterium glutamicum for biosorption of Reactive Black 5 from Aqueous
Solution. Industrial & Engineering Chemistry Research, 46: 608-617
Vijayaraghavan, K., J. Mao and Y-S. Yun. 2008. Biosorption of methylene blue from
aqueous solution using raw and polysulfone-immobilized Corynebacterium
glutamicum: Batch and column studies. Bioresource Technology, 99: 2864–2871.
Wang, J. and C. Chen. 2009. Biosorbents for heavy metals removal and their future.
Biotechnology Advances, 27:195-226.
Wang, X., G. Zhu and F. Guo. 2013. Removal of uranium (VI) ion from aqueous solution
by SBA-15. Annals of Nuclear Energy, 56 : 151–157.
147
Xia, L., K. Tan, X. Wang, W. Zheng, W. Liu and C. Deng. 2013. Uranium removal from
aqueous solution by banyan leaves: Equilibrium, thermodynamic, kinetic, and
mechanism studies. Journal of Environmental Engineering, 139: 887-895.
Yan, G. and T. Viraraghavan. 2000. Effect of pretreatment on the bioadsorption of heavy
metals on Mucor rouxi, Water SA, 26: 119–123.
Yesi, F., P. Sisnandy, Y. H. Ju, F. E. Soetaredjo and S. Ismadji. 2010. Adsorption of Acid
Blue 129 from Aqueous Solutions onto Raw and Surfactant-Modified Bentonite:
The Application of Temperature Dependence Form of Adsorption Isotherms.
Adsorption Science & Technology, 28: 847-868.
Yi, Z. and J. Li. 2012. Removal of uranium(VI) from aqueous solution by dry chitosan
powder. Advanced Materials Research, 366: 434-437.
Yu, Z., J. Dai, R. Zhao, L. Xu and Y. Yan. 2012. Removal of strontium ions from
aqueous solution by adsorption onto sodium trititanate whisker. Fresenius
Environmental Bulletin, 21:19-25.
Yi, Z.-J., J. Yao, F. Wang, H-L. Chen, H-J. Liu and C. Yu. 2013. Removal of
uranium(VI) from aqueous solution by apricot shell activated carbon. Journal of
Radioanalytical and Nuclear Chemistry, 295: 2029-2034.
Zulfadhly, Z., M. D. Mashitah, S. Bhatia. 2001. Heavy metals removal in fixed-bed
column by the macro fungus Pycnoporus sanguineus, Environmental Pollution,
112: 463-470.
Zuo, Y., J. Zhan and Y. Deng. 2001. Environmental application of chitin and chitosan
extracted from seafood processing waste. In Advance in Environmental
Materials,Environmentally Preferred Materials, 2: 249-260,
Zuo, Y., J. Zhan, N. Costa. 2001. Use of shell chitin extracted from seafood processing
waste in recycling of industrial wastewater. Proceedings of SPIE:
Environmentally Conscious Manufacturing, 4193:403-413.
Zhang, Y. and C. Banks. 2006. A comparison of the properties of polyurethane
immobilized Sphagnum moss, seaweed, sunflower waste and maize for the
biosorption of Cu, Pb, Zn and Ni in continuous flow packed columns. Water
Research, 40: 788–798.
Zafar, M. N., R. Nadeem and M. A. Hanif. 2007. Biosorption of nickel from protonated
rice bran. Journal of Hazardous Materials, 143: 478–485.
148
Zou, W., L. Zhao and R. Han, 2009. Removal of Uranium (VI) by Fixed Bed Ion-
exchange Column Using Natural Zeolite Coated with Manganese Oxide. Chinese
Journal of Chemical Engineering, 17: 585-593.
Zolgharnein, J. and A. Shahmoradi. 2010. Characterization of sorption isotherms, kinetic
models, and multivariate approach for optimization of Hg(II) adsorption onto
Fraxinus tree leaves. Journal of Chemical and Engineering Data, 55:5040-504.
Zou, W. and L. Zhao. 2012. Removal of uranium(VI) from aqueous solution using citric
acid modified pine sawdust: Batch and column studies. Journal of Radioanalytical
and Nuclear Chemistry, 291:585-59.
Zhang, X. F., L. Y. Ji, J. Wang, R. M. Li, Q. Liu, M. L. Zhangand, L.H. Liu. 2012.
Removal of uranium(VI) from aqueous solutions by magnetic Mg-Al layered
double hydroxide intercalated with citrate: Kinetic and thermodynamic
investigation. Colloids and Surfaces A-Physicochemical and Engineering Aspects,
414: 220-227.
Zhong, Z-Y., Q. Yanga, X-M. Li, K. Luo, Y. Liua and Guang-Ming. 2012. Preparation
of peanut hull-based activated carbon by microwave-induced phosphoric acid
activation and its application in Remazol Brilliant Blue R adsorption. Industrial
Crops and Products, 37: 178– 185.
Zhou, L., C. Shang, Z. Liu, G. Huang and A. A Adesina. 2012. Selective adsorption of
uranium(VI) from aqueous solutions using the ion-imprinted magnetic chitosan
resins. Journal of Colloid and Interface Science, 366:165-72.
Zhang, Z. I. M. O. Hara, G. A. Kent and W. O. S. Doherty. 2013. Comparative study on
adsorption of two cationic dyes by milled sugarcane bagasse. Industrial Crops and
Products, 42 :41– 49.
Zhang, Z-B., W-B. Nie, Q. Li, G-X. Xiong, X.-H., Cao and Y.-H. Liu. 2013. Removal of
uranium(VI) from aqueous solutions by carboxyl-rich hydrothermal carbon
spheres through low-temperature heat treatment in air. Journal of Radioanalytical
and Nuclear Chemistry, 298: 361-368.
Zhu, G. L., X. Y. Lin and X .G. Luo. 2013. Fixed-bed column study for the removal of
strontium (II) ions from aqueous solutions using expanding rice husk. Advanced
Materials Research, 726-731:654-657.